HUPO 2024 Abstracts

Table of contents Plenary lectures (PL-05) .............................................................................................................................................. 3 Keynote lectures (KN-02 to KN-48) ............................................................................................................................. 4 Oral presentations (OP-01 to OP-88) ........................................................................................................................ 28 Poster presentations of HUPO & EuPA initiatives (iP-01 to iP-13) ......................................................................... 132 Defining signaling networks - Funcyional PTMs (P-I-0001 to P-I-0053) ................................................................. 146 Microbiology and microbiome analysis (P-I-0054 to P-I-0096) .............................................................................. 200 New technology: MS-based proteomics (P-I-0097 to P-I-0204) ............................................................................ 243 Population proteomics in health and disease (P-I-0205 to P-I-0232) .................................................................... 353 Spatial and imaging proteomics (P-I-0233 to P-I-0273) ......................................................................................... 382 Clinical proteomics I (P-I-0275 to P-I-0383) ............................................................................................................ 425 New technology: Sample preparation (P-II-0384 to P-II-0434) .............................................................................. 539 New technology: Chromatography (P-II-0435 to P-II-0444) .................................................................................. 589 New technology: AI and bioinformatics in mass spectrometry (P-II-0445 to P-II-0504) ....................................... 599 Multiomics approaches (P-II-0505 to P-II-0558) .................................................................................................... 659 Structural proteomics (P-II-0559 to P-II-0594) ....................................................................................................... 711 One health approaches (plant, food, nutrition, animal, environment) (P-II-0595 to P-II-0617) ........................... 744 Clinical proteomics II (P-II-0618 to P-II-0723) ......................................................................................................... 767 Infectious biology insights (P-II-0724 to P-II-0746) ................................................................................................ 884 New technology: Non MS-based proteomics (P-III-0747 to P-III-0762) ................................................................. 908 Data integration: With bioinformatics to biological knowledge (P-III-0763 to P-III-0852) .................................... 923 Chemical biology insights (P-III-0853 to P-III-0875).............................................................................................. 1012 Organisation of the proteome (PPI) (P-III-0876 to P-III-0890) ............................................................................. 1035 Glycobiology insights (P-III-0891 to P-III-0916) .................................................................................................... 1051 Cell biology insights (P-III-0917 to P-III-0997) ...................................................................................................... 1076 Human health insights (neurobiology, cardiovascular, liver, kidney etc.) (P-III-0999 to P-III-1083) ................... 1159 Immunopeptidomics (P-III-1084 to P-III-1108) ..................................................................................................... 1248 Late breaking poster (LP-01 to LP-10) .................................................................................................................. 1275 2

Keynote lectures (KN-02 to KN-48) KN-02 (Phospho)proteomics for personalized precision oncology Connie Jimenez1 1Amsterdam University Medical Center, Medical Oncology, Amsterdam, Netherlands Cancer is a heterogeneous disease characterized by aberrant cellular signaling. Hyperactive kinase activity is one of the drivers of tumorigenesis and cancer progression and results in altered phosphorylation abundance and activity of downstream substrates. A prime prerequisite for personalized treatment is that kinase activity inference can be based on a single (patient) sample. This is pivotal in a clinical setting with the aim to select the best kinase inihibitor treatment for individual patients. To this end, phosphoproteomics coupled to INKA analysis has emerged as an advantageous approach that achieves an optimized ranking of inferred kinase activities in single samples (www.inkascore.org) (1-5). INKA integrates kinase phosphorylation and substrate phosphorylation to infer kinase activity. The output is a ranked kinase activity bar graph and an individualized kinase-substrate network that can be used by biologists and clinicians to extract actionable information from modern deep comprehensive (patient) phosphoproteomic data. In my lecture, I will present our progress in (phospho)proteomics profiling and INKA of xenograft tumor models and tumor needle biopsies collected as part of side-studies in precision medicine clinical trials (IMPACT, colorectal cancer patients treated with anti-EGFR therapy; REPOSIT, melanoma patients treated with anti-BRAF- MEK combination therapy; DRUP-Lenvatinib cohort, pan-cancer). Our findings show that needle biopsy protein yields are compatible with phosphoproteomics (~200 mg protein) and a large fraction of the samples is also amendable to pTyr-phosphoproteomics that we recently down-scaled to 500 mg protein input. The analyses offer explanation for (lack of) drug response and uncover actionable resistance mechanisms. Ultimately, prospective clinical trials that include drug selection based on patient phosphoproteomic data coupled to inference, will show the added value of phosphoproteomics (combined with genomics) for precision medicine. individualized kinase and pathway activity 1. Beekhof R, Henneman AA, Jimenez CR et al.,. INKA, an integrative data analysis pipeline for phosphoproteomic inference of active kinases. Mol Syst Biol. 2019 May 24;15(5):e8981. 2. Cordo' V, et al., Phosphoproteomic profiling of T cell acute lymphoblastic leukemia reveals targetable kinases and combination treatment strategies. Nat Commun. 2022 Feb 25;13(1):1048. 3. Vallés-Martí A, et al. Phosphoproteomics guides effective low-dose drug combinations against pancreatic ductal adenocarcinoma. Cell Rep. 2023 Jun 1;42(6):112581. 4. Beekhof R, et al., Phosphoproteomics of patient-derived xenografts identifies targets and markers associated with sensitivity and resistance to EGFR blockade in colorectal cancer. Sci Transl Med. 2023 Aug 16;15(709):eabm3687. 5. Vallés-Martí A, et al. Kinase activities in pancreatic ductal adenocarcinoma with prognostic and therapeutic avenues. Mol Oncol. 2024 Aug;18(8):2020-2041. 4

KN-04 Multi-omics analysis in CTC-based liquid biopsy Hyun Woo Park1 1Yonsei University, Biochemistry, Seoul, South Korea A specialized mechanism that reprograms anchorage dependency of solid tumor cells into circulating tumor cells (CTCs) during the metastatic cascade remains elusive. Using multi-omics approaches in liquid biopsy, we discovered a biological phenomenon referred to as Adherent-to-Suspension Transition (AST) that reprograms adherent cells into suspension cells via aberrant induction of hematopoietic transcriptional regulators, IKZF1, NFE2, BTG2, and IRF8, which are hijacked by solid tumor cells to disseminate into CTCs. During dissemination, tumor microenvironment triggers the epigenetic alteration of AST factors in the invasive front of solid tumors to evoke spontaneous cell-matrix dissociation, acquire anoikis resistance, and bypass immune surveillance of CTCs, however, in the absence of lineage differentiation and EMT. By performing scRNAseq in a cohort of de novo metastatic breast cancer patients, which consists tissue- and liquid biopsy paired specimens of primary tumor, CTCs, and metastatic lesions, we uncovered how Adherent-Suspension Plasticity (ASP) regulated anchorage plasticity during the dissemination and colonization process within the metastatic cascade. Finally, we demonstrate therapeutic strategies that target AST factors to specifically abrogate CTC formation and suppress distant metastases.Together, our finding underscores the mechanism that reprograms anchorage dependency of cancer cells and highlights undescribed hematopoietic targets for the development of anti-metastatic therapy. Furthermore, ASP opens unexplored scenarios of how reprogramming anchorage dependency in diverse physiological events orchestrates development, organ morphogenesis, and immune cell activation. Fig. 1 (optional) 5

KN-05 Deep visual proteomics for precision oncology Fabian Coscia1 1Max Delbrück Center Berlin, Spatial Proteomics, Berlin, Germany Through innovative single-cell and spatial omics, cell-to-cell heterogeneity can be mapped today at unprecedented molecular detail, profiling thousands of analytes per cell, offering a promising avenue towards personalized medicine. Albeit still at an early stage, unbiased (i.e. untargeted) liquid chromatography mass spectrometry (LC-MS) based spatial proteomics approaches promise to revolutionize our understanding of cell heterogeneity from a global, functional and phenotype-centric perspective. Here, we recently co-developed deep visual proteomics (DVP) that combines whole-slide tissue imaging, machine-learning based image analysis and ultrasensitive LC-MS based proteomics to analyze cell type and spatially resolved proteomes at unprecedented biological resolution. In my presentation, I will give an overview of state-of-the-art spatial proteomics approaches, highlight our previous and ongoing work to integrate targeted and untargeted methods and show several examples how we apply such approach for translational cancer research. Examples include the deep proteomic profiling of cancer precursor lesions, the spatially resolved analysis of melanoma progression obtained from a 17-years old archival specimen, and how the integration with spatial transcriptomics can prioritize actionable and niche-specific drug targets in an aggressive lung adenocarcinoma. Finally, I will give an outlook on promising new directions to improve throughput, sensitivity and accessibility of multimodal spatial proteomics in general, and the DVP pipeline in particular. 6

KN-06 A spatio-temporal single-cell type map of the human proteome based on transcriptomics, high-resolution antibody-based imaging and artificial intelligence Cecilia Lindskog1, Feria Hikmet Noraddin1, Loren Méar1,2, Filippa Bertilsson1, Borbala Katona1, Rutger Schutten1, Kalle von Feilitzen3, Mattias Forsberg3, Mathias Uhlen2,3 1Uppsala University, Immunology Genetics and Pathology, Uppsala, Sweden 2Karolinska Institute, Stockholm, Sweden 3KTH Royal Institute of Technology, Stockholm, Sweden For a fundamental understanding of human health, molecular medicine and targeted treatment, it is necessary to map processes unique to each tissue and cell type. The major collection of bioimages in the Human Protein Atlas (HPA) database, comprising 10 million high-resolution images of stained tissues and cells, constitutes an important resource for studying protein expression in relation to tissue and cell morphology. While the staining of the most common cell types and structures have been manually annotated and is presented as protein expression levels, the immunohistochemical data has limitations in terms of relative protein abundance, subjectivity and challenges distinguishing specific subsets of cells or rare cell types. In an effort to increase the resolution of the tissue-based protein expression data, we here aimed to set up a stringent workflow for mapping human tissues at the single-cell type level, and utilized this workflow to create a high-resolution spatio- temporal maps of >1,000 tissue or cell type-specific proteins in eight different tissues. Based on single-cell RNA sequencing data, we identified tissue-specific cell types and subsets thereof, some of which cannot be distinguishable by regular immunohistochemistry. Using a multiplex immunofluorescence pipeline, we then built antibody panels specifically outlining each of these cell types. The fixed antibody panels were stained together with the candidate protein of interest, one at a time, to pinpoint the exact protein localization at a cellular and subcellular level. Using artificial intelligence, the high-resolution images were quantified using an automated image analysis workflow. The integrated data allowed us to study temporal mRNA and protein expression gradients along with maturation processes and identify which mRNAs that are consistently translated into proteins from those that vary from a spatio-temporal aspect. We were also able to assign presumed functions to numerous uncharacterized proteins. The established pipeline presented will be further used for other human tissues, decoding the complexity of human cells, tissue heterogeneity and homeostasis, and link quantitative data with tissue morphology. The data is freely available on www.proteinatlas.org and the detailed data will likely contribute to valuable insights into protein function and molecular processes linked to disease. 7

KN-13 Challenges and opportunities of genome-wide association studies with proteomics Karsten Suhre1 1Weill Cornell Medicine - Qatar, Research, Doha, Qatar Protein quantitative trait loci (pQTLs) are important tools in the drug discovery process as they provide supporting evidence for drug target identification and hypothesis generation regarding their modes of action. Most genome-wide association study studies conducted so far have used affinity proteomics platforms, which have limited discovery to protein panels covered by the available affinity binders. Furthermore, it is not clear to which extent protein epitope changing variants interfere with the detection of pQTLs. Mass spectrometry-based (MS) proteomics can overcome some of these limitations and provide insights regarding the role that specific protein isoforms may play. However, MS proteomics also have their own challenges, such protein quantification in the presence of peptides containing protein altering variants. Here we report insights from a recent GWAS conducted using MS-based proteomics and how the results relate to pQTLs discovered using affinity platforms. 8

KN-14 Molecular health perspectives from self-sampled circulating proteomes Jochen Schwenk1 1KTH Royal Institute of Technology, Science for Life Laboratory, Stockholm, Sweden Recent advancements in proteomics technologies have made it possible to analyze thousands of blood proteins across thousands of samples, significantly enhancing expectations for breakthroughs in precision medicine. However, despite the influx of new molecular health data, distinguishing observational associations from actionable links remains crucial for translating this new knowledge into clinically relevant information about human biology. This presentation details our research approaches to the dynamic architecture of circulating proteins as they can be associated with various human phenotypes, health status, and disease risks. The talk will also present our work studying the impact of medication use on these data. Beyond the established biobanks of liquids, we will discuss the innovative approach of self-sampling blood, enabling the exploration of population health beyond traditional hospital environments. This "new" sample type can simplify continuous blood collection at home, offering new immune and health monitoring opportunities. For proteomics-informed precision medicine to be realized, future efforts must prioritize improved management of pre-analytical factors, a deeper understanding of technology-centric insights, and greater emphasis on the conditions of donors at the time of sampling. As multi-modal data from novel study designs and diverse sample types emerge, there is a pressing need for data-driven approaches to learn from the community data and effectively address the heterogeneity present at baseline and throughout the longitudinal monitoring of phenotypes. 9

KN-15 An expanded RAS proteoform landscape provided by top-down analysis of malignant cell lines Grace Scheidemantle1, Robert D'Ippolito1, Kanika Sharma1, Nicole Fer1, Brian Smith1, Mackenzie Meyer1, Scott Eury1, Abigail Neish1, Katie Powell1, Vanessa Wall1, William Burgan1, Dominic Esposito1, Anna Maciag1, Frank McCormick1,2, Dwight Nissley1, Caroline DeHart1 1Frederick National Laboratory for Cancer Research, NCI RAS Initiative, Frederick, United States 2Helen Diller Family Comprehensive Cancer Center, University of California, San Francisco, San Francisco, United States Top-down mass spectrometry (TDMS), which analyzes intact and modified protein forms (proteoforms), can confidently identify and precisely localize mutations or post-translational modifications (PTMs) while maintaining the primary sequence. This provides unparalleled molecular detail, particularly in the case of the RAS GTPases (HRAS, KRAS4A, KRAS4B, NRAS), which are the most frequently mutated proteins in human cancer. Their high sequence identity, labile PTMs, and low endogenous abundance have all provided formidable challenges to standard peptide-based proteomic methods. However, by directly analyzing immunoprecipitated RAS proteoforms via targeted TDMS (IP-TDMS), the NCI RAS Initiative has identified novel RAS PTMs and context- specific RAS proteoforms which provide new insight into RAS-dependent signaling mechanisms. Previously, we developed an optimized RAS IP-TDMS assay on an Orbitrap Fusion Lumos mass spectrometer. By analyzing recombinant, FLAG-tagged, and endogenous versions of each RAS identified chromatographic gradients, proteoform charge states, and isoform-specific targeted MS2 parameters to facilitate optimal proteoform characterization. However, after an expanded analysis of malignant cell lines employing these optimized parameters, many endogenous RAS proteoforms still could not be successfully characterized due to their low relative abundance and certain expected RAS proteoforms (HRAS, KRAS4A) were still not observed. To address this, we further optimized our RAS IP-TDMS assay on an Orbitrap Exploris 480 mass spectrometer. Here, we have substantially expanded the RAS proteoform landscape by performing IP-TDMS analysis on fourteen malignant cell lines (NCI-60 and NCI Ras Initiative validated) using previously optimized parameters and both instrument platforms. isoform, we In doing so, we identified an unanticipated wealth of novel HRAS, KRAS4A, KRAS4B, and NRAS proteoforms, the majority of which differed markedly from the established literature. We also observed further evidence of context-dependent RAS proteoforms, including apparent tissue-, isoform- and mutation-specific modifications. Our results underscore the ability of proteoform-resolved measurements to improve our understanding of RAS- dependent signaling while emphasizing that much work remains to be done to elucidate the role of RAS PTMs and proteoforms in oncogenesis. 10

KN-17 Single-cell and spatial proteomics workflows for in-depth tissue mapping Claudia Ctortecka1, Brian W. Boyle1, Natalie M. Clark1, Anjali Seth1, DR Mani1, Michael Gillette1, Shankha Satpathy1, Namrata D. Udeshi1, Steven A. Carr1 1Broad Institute, Cambridge, United States The composition of heterogenous cell types and their tightly controlled spatial localization within tissues dictates health or disease. To characterize and map these cellular identities, recently developed workflows for single-cell or spatial proteomics have demonstrated unprecedented sensitivity. The key improvements relate to automated nanoliter sample preparation and use of latest generation mass spectrometry (MS)-instrumentation for reproducible proteome analysis of single mammalian cells or small tissue sections. Despite these advances, sample acquisition throughput at biologically relevant proteome depth to provide sufficient tissue representation remains challenging. To address these limitations, we have combined a dedicated nano-96-well chip to in-line sample clean-up and chromatographic separation. In conjunction with highly sensitive and robust MS acquisition, the workflow enables robust and highly sensitive low-input proteomics, including single cell and spatial tissue analysis. Using 40SPD on the Evosep with optimized low-input dia-PASEF methods on the timsTOF Ultra2 we identify and quantify more than 4,000 proteins per single cell in solution or 25 fresh-frozen cells. To preserve the spatial context of cells within tissues we additionally combine our proteoCHIP workflow with laser capture microdissection and excise regions of interest down to 25 FFPE or fresh-frozen cells. The combination of acquisition throughput (40SPD) with this surprising proteome depth enables us to build a spatially resolved tissue catalogue including diverse tumor types and normal adjacent tissue. Based on this integrative single cell and spatial workflow we will demonstrate the analysis of clinically relevant samples. 11

KN-19 Deep learning for peptide property prediction and beyond Mathias Wilhelm1 1Technical University of Munich, Computational Mass Spectrometry, Freising, Germany Various deep learning-assisted data analysis pipelines developed recently are able to boost the performance of mass spectrometry-based proteomics. One such pipeline is Oktoberfest that can assist the analysis of DDA, DIA, and PRM data. Oktoberfest is based on Prosit, a deep neural network that predicts multiple peptide properties. Since its release in 2019, it was used to generated >20 billion predicted spectra. Here, we summarize the available Prosit models for peptide property predictions and specifically highlight recent developments to generate a new single model for fragment ion intensity prediction covering tryptic and non-tryptic, modified and unmodified, and labeled and unlabeled peptides for different mass analyzers and fragmentation methods. The application of deep learning in proteomics is still in its early stages. While it holds the potential to assist upstream and downstream analysis and interpretation of proteomics data even further, the lack of reproducibility and FAIRness of models are widely acknowledged concerns. To limit such issues, we have developed a number of re-usable open-source packages relevant to the development and application of deep learning models in proteomics. Beyond the development of novel peptide property predictors, we will showcase the potential of AI on use cases such as interpretable AI and drug response prediction. 12

KN-20 Signalling architecture underlying breast cancer cell motility Emily Goodall1, Joseph Parsons1, Harriet Ferguson1, Joanne Watson1, Jennifer Ferguson1, Paul Fullwood1, Jean- Marc Schwartz1, Patrick Caswell1, Robert Clarke1, Michael P. Smith1, Chiara Francavilla1,2 1University of Manchester, Manchester, United Kingdom 2Technical University of Denamrk, Lyndby, Denmark Cell migration is vital to a variety of biological processes including wound healing, the immune response, embryonic development, and cancer cell metastasis. Growth factors and proteins in the extracellular matrix (ECM) provide cues to guide migrating cells. However, the specific contribution of each of these signals and potentially their integration via intracellular signalling pathways to promote cell migration are still poorly understood. Here, combining quantitative proteomics and phosphoproteomics with bioinformatics, and functional validation in both breast cancer cell models and patient samples, we investigated how: 1) Receptor Tyrosine Kinases affect signalling networks underlying cell motility, mitogenicity, and survival 2) non canonical signalling pathways regulate cell migration in response to both ECM proteins and growth factors 3) ECM proteins and kinases promote breast cancer bone metastasis Our findings show that a more intricated signalling architecture than anticipated fine tunes the motility of breast cancer cells with profound implications for our understanding of breast cancer metastasis and for improving the therapeutic options available for patients 13

KN-21 Combining phospho- (and other) proteomics with biological knowledge into computational models to understand signal transduction Julio Saez-Rodriguez1 1EMBL-EBI & heidelberg University, Heidelberg, Germany Proteomics in its various forms – from phosphor-proteomics to thermal proteome profiling - provides rich information about cellular processes and their deregulation in disease. Given the complexity of the data, we need computational methods to effectively extract mechanistic information from it. We believe that including biological knowledge in the methods can be instrumental to move from pure correlation to causation in large data sets, and thereby identify the molecular processes that underlie signaling events, that can be ultimately used as therapeutic targets and biomarkers. In this talk I will describe various approaches that follow this strategy, from kinase-activity estimation to signaling network reconstruction, and illustrate them in examples from various contexts. 14

KN-23 Neuroproteomics to decipher how the COVID-19 affects the human brain Daniel Martins-de-Souza1 1University of Campinas (Unicamp), Dept of Biochemistry, Campinas, Brazil COVID-19 is among the most deadly pandemic of humankind. The disease caused by the SARS-CoV-2 is rather versatile, since it may affect different physiological systems, associated to a range of symptoms, which may pass or may become sequelae. Neurological symptoms are among the most prevalent of the extrapulmonary complications of COVID-19, affecting more than 30% of patients. Here, based on data obtained in living COVID- 19 patients, postmortem brain tissue, and cell cultures, we found that the SARS-CoV-2 is able to reach the human brain, where it infects mostly astrocytes. Moreover, astrocytes are susceptible to SARS-CoV-2 infection through a noncanonical mechanism and respond to the infection by remodeling energy metabolism, which in turn, alters the levels of metabolites used to fuel neurons and support neurotransmitter synthesis, according to proteomic and metabolomic analysis. The altered secretory phenotype of infected astrocytes, in turn, impairs neuronal viability. Morphological and molecular data we obtained explains the damage and structural changes observed in the brains of COVID-19 patients, which might be associated with the long-lasting symptoms (and eventually sequelae) known as "long COVID". 15

KN-26 Global organelle profiling reveals subcellular localization and remodeling at proteome scale Marco Y. Hein1,2, Duo Peng1, Verina Todorova1, Frank McCarthy1, Kibeom Kim1, Chad Liu1, Laura Savy1, Camille Januel1, Rodrigo Baltazar-Nunez1, Sophie Bax1, Shivanshi Vaid1, Madhuri Vangipuram1, Ivan Ivanov1, Janie R. Byrum1, Soorya Pradeep1, Carlos G. Gonzalez1, Yttria Aniseia1, Eileen Wang1, Joseph S. Creery1, Aidan H. McMorrow1, James Burgess3, Sara Sunshine4, Serena Yeung-Levy1, Brian C. DeFelice1, Shalin B. Mehta1, Daniel Itzhak1, Joshua E. Elias1, Manuel D. Leonetti1 1Chan Zuckerberg Biohub, San Francisco, United States 2Medical University of Vienna, Max Perutz Labs, Vienna, Austria 3Stanford University, Institute for Computational & Mathematical Engineering, Palo Alto, United States 4University of California San Francisco, San Francisco, United States Defining the subcellular distribution of all human proteins and its remodeling across cellular states remains a central goal in cell biology. Here, we present a high-resolution strategy to map subcellular organization using organelle immuno-capture coupled to mass spectrometry. We apply this workflow to a cell-wide collection of membranous and membrane-less compartments. A graph-based analysis reveals the subcellular localization of over 7,600 proteins, defines spatial networks, and uncovers interconnections between cellular compartments. Our approach can be deployed to comprehensively profile proteome remodeling during cellular perturbation. By characterizing the cellular landscape following hCoV-OC43 viral infection, we discover that many proteins are regulated by changes in their spatial distribution rather than by changes in abundance. Our results establish that proteome-wide analysis of subcellular remodeling provides unique insights for the elucidation of cellular responses, uncovering an essential role for ferroptosis in OC43 infection. Our dataset can be explored at organelles.czbiohub.org. 16

KN-28 π-DDK-Linker: Towards an automated intelligent knowledge mining system for proteomics data Dong Li1 1Beijing Proteome Research Center, Beijing, China The development of high-throughput proteomics technology is expected to accelerate the process of drug target identification and drug discovery. Data analysis has become a significant challenge in the proteomic study. The manual interpretation process has become a bottleneck in the knowledge mining of omics data. To address this, I have conducted a series of studies around the drug-related knowledge graph and related knowledge discovery algorithms, successively constructing multiple knowledge bases including cancer drug-sensitive gene signals, drug response gene signals, TCM small molecule-drug target interactions, and ubiquitination modification networks. Besides the BATMAN-TCM database search algorithm, I proposed a knowledge graph-based data analysis algorithm, π-DDK-Linker, which explores disease signals hidden in omics data through the association (Linker) between omics data (D) and known disease-drug knowledge (K), providing clues for research on drug targets and other areas. In the future, π-DDK-Linker will continue to enhance its performance in all aspects and will continuously integrate a series of knowledge bases developed in the π-Hub project, such as biological networks, precision medicine, and disease genes, using the knowledge graph as a carrier to develop a new omics data drug knowledge mining system based on artificial intelligence algorithms. 17

KN-31 Disease-state and progression biomarker discovery in multiple myeloma plasma DR Mani1, Elizabeth Lightbody2, Hasmik Keshishian1, Michael Gillette1, Irene Ghobrial2, Steven A. Carr1 1Broad Institute of MIT and Harvard, Proteomics, Cambridge, MA, United States 2Dana-Farber Cancer Institute, Boston, MA, United States Multiple Myeloma (MM) develops from well-defined precursors—Monoclonal Gammopathy of Undetermined Significance (MGUS) and Smoldering Multiple Myeloma (SMM)—where patients remain stable or may rapidly progress. Bone marrow (BM) biopsies are used for staging and identifying high-risk genomic abnormalities associated with progression. However, BM biopsies are invasive and cannot be repeated often for monitoring tumor burden in the precursor setting, and these patients are limited to monitoring a few markers within peripheral blood (PB) for signs of progressive disease. Deep proteome profiling of PB plasma may advance non- invasive precursor disease staging, monitoring, and characterization. Precursor patients are currently monitored via a few peripheral blood biomarkers for signs of progressive disease. Plasma proteome profiling to identify effective biomarkers has been hindered by the wide dynamic range of plasma and often low abundance of disease-specific markers. Here we performed the first comprehensive analysis of the plasma proteome of MM and its precursor conditions using the Olink proximity extension assay (PEA) that targets ca. 3000 proteins. We profiled 529 PB plasma samples from 485 individuals, including MGUS (n=100), SMM (n=203), MM (n=100), and age-matched healthy donors (HDs) (n=82). Samples from patients with progressive disease (n=32) and stable precursor disease (n=32) with matched clinical follow-up time were also profiled. Analysis of the data enabled us to develop a classifier that utilizes plasma proteins alone to accurately classify disease stages and identified a prognostic protein signature associated with progressive disease. Advancements are underway to validate our high-risk disease candidates, test the performance of an improved risk stratification model in external cohorts, and determine how best to implement proteome profiling into clinical practice. 18

KN-32 Alcohol shapes multiple microenvironments in oral cancer initiation Ariane Busso-Lopes1, Daniella de Figueiredo1, Tatiane De Rossi1, Ana Karina de Oliveira1, Nayane Galdino2, Rodrigo Cruz1, Larissa Carvalho-Silva1, Jamile Sá1, Sami Yokoo1, Reynaldo Melo1, Jackson Gabriel Miyamoto1, Maurício Camacho1, Bianca Pauletti1, Romênia Domingues1, Daniela Campos Granato1, Elaine C Cardoso1, Heloísa MA Prado1, Wagner FO da Silva1, Fabio Patroni1, Gabriela Raposo3, Mariana Boroni3, Mariela Piccin2, Tiago Medina2, Guilherme Tamarindo1, Sandra Dias1, João Ormonde1, Marcos Alborghetti1, Mauricio Sforca1, Ana Carolina Prado Ribeiro4, Thais Brandão4, Mariane Amano5, Miyuki Uno4, Alan Santos-Silva6, Marcio Lopes6, Kenneth Gollob7, Luiz Paulo Kowalski8, Adriana Paes Leme1 1Brazilian Center for Research in Energy and Materials, Campinas, Brazil 2A.C. Camargo Cancer Center, São Paulo, Brazil 3National Cancer Institute, Rio de Janeiro, Brazil 4Sao Paulo Cancer Institute, São Paulo, Germany 5Sirio-Libanes Hospital, São Paulo, Brazil 6Faculty of Dentistry of Piracicaba, UNICAMP, Piracicaba, Brazil 7Israelita Albert Einstein Hospital, São Paulo, Brazil 8USP Medical School, São Paulo, Brazil Head and neck cancer (HNC) is ranked the seventh leading cause of cancer worldwide, with a 5-year survival rate of less than 50% in advanced cases. Together with tobacco smoking, alcohol drinking is one of the major risk factors for HNC. Despite the advances in understanding HNC progression, the mechanism involved in HNC initiation is largely unknown. Through spatially resolved and bulk proteomics using holistic and reductionist approaches, we demonstrate in a 94 patient-cohort that alcohol modifies the proteome pattern of stromal and epithelial cells, circulating immune cells, primary fibroblasts and their extracellular vesicles (EVs). Besides tissue biopsy-derived EVs, alcohol-treated fibroblasts and their EVs shape the immune response. In addition, a selected protein, identified in both biopsy-derived EVs and those from alcohol-treated fibroblasts, modulates the phenotype of specific immune cells. These results improve our understanding of the molecular changes induced by alcohol in multiple microenvironments, and highlight the role of EVs in expanding its toxicity. 19

KN-34 The biology of antigen processing revealed through the lens of immunopeptidomics Anthony Purcell1 1Monash University, Biochemistry, Clayton, Australia Human Leukocyte antigen (HLA) molecules play a key role in health and disease by presenting antigen to T- lymphocytes for immunosurveillance. Immunopeptidomics involves the study of the collection of peptides presented within the antigen binding groove of HLA molecules. Identifying their nature and diversity is crucial to understanding immunosurveillance especially during infection or for the recognition and potential eradication of tumours. I will discuss new insights into the fidelity of antigen processing and presentation and how novel informatics approaches and databases have shed light on the diversity of peptide antigens derived from unconventional sources including peptides derived from transcripts associated with frame shifts, long noncoding RNA, incorrectly annotated untranslated regions, post translational modifications and proteasomal splicing. Several challenges remain in successful analysis of immunopeptides, yet these developments point to unexplored biology waiting to be unravelled. 20

KN-37 Beyond a bird's eye view towards One Health: Demystifying proteogenomic landscape and organeller control of health to disease linking multi-host response Subhra Chakraborty1 1National Institute of Plant Genome Research, New Delhi, India Complex nature of the interactions between plant, animal, human and ecosystem renders multisectoral transdisciplinary approaches critical to address future food and health risks and challenges. Proteogenomics and organeller proteomics uncover new strategies for one health problems by facilitating discovery of molecular signature of disease and protein biomarkers to multi-host resistance. Fusarium sp., an ascomycetes fungus and multi-host pathogen is a causal agent of vascular wilt in plants, neuronal stress in worm and fusariosis in humans. Although precisely controlled innate immune response is governed by conserved cellular events in phylogenetically diverse hosts, underlying molecular mechanisms by which this process is regulated against multi-host pathogen remain unknown. We have established the proteo-genomics landscape of a food legume, chickpea and reanalysed the data for worm in healthy and disease state based on DDA library and SWATH-MS with a high-quality spectral library for rapid and deep profiling. Further, with an iTRAQ based workflow combining integrated transcriptomics and metabolomics, we also demonstrated multilayered analyses at the resolution of organellar proteomics and phosphor-proteomic and their application to reveal new insight on the immune signature and pathogenesis during fusariosis. Most excitingly, network analysis identified significant functional modules pointing toward fusarium responsive common mechanism of remodelling, homeostasis and cell death response through divergent regulatory pathways across kingdom. 21

KN-41 Microproteins and the non-canonical proteome in childhood cancer Sebastiaan van Heesch1 1Princess Máxima Center for Pediatric Oncology, Utrecht, Netherlands Ribosome profiling (Ribo-Seq) has proven transformative for our understanding of the human genome and proteome by illuminating thousands of noncanonical sites of ribosome translation outside the currently annotated coding sequences (CDSs). A conservative estimate suggests that at least 7000 noncanonical ORFs are translated, which, at first glance, has the potential to expand the number of human protein CDSs by 30%, from ~19,500 annotated CDSs to over 26,000 annotated CDSs. Yet, additional scrutiny of these ORFs has raised numerous questions about what fraction of them truly produce a protein product and what fraction of those can be understood as proteins according to the conventional understanding of the term. I will discuss the current state of noncanonical ORF research, databases, and interpretation, focusing on how to assess whether a given ORF can be said to be "protein coding." I will do so in the context of childhood cancer, where strategies targeting the small microproteins produced from these ORFs hold great promise for the development of new (immuno)therapies against these deadly early-onset cancers. 22

KN-42 Profiling metalloproteomes by chemical and computational proteomics Chu Wang1 1Peking University, Beijing, China Metal binding proteins (MBPs) are widely distributed in living systems. They play key roles in diverse biological processes, including but not limited to catalysis, stabilization of proteins" structure, and signal transduction. So far, there are few methods to identify MBPs on the proteome level. In this talk, I will describe our recent and ongoing efforts to profile MBPs in a global scale by chemical and computational proteomics methods1. Our approach identified a large number of previously unannotated MBPs, some of which can be experimentally and structurally validated. Our chemoproteomic and computational methods will provide an enabling tool for discovering more MBPs systematically and the data will generate more hypothesis for the community of studying metal biology. 23

KN-43 Proteomics tools facilitate the discovery of the new function of PTMs Mingliang Ye1 1Dalian Institute of Chemical Physics, Chinese Academy of Science, Dalian, China Post-translational modification (PTM) regulates protein functions and influence almost all aspects of normal cell biology and pathogenesis. Therefore, identifying and understanding PTMs is crucial in the study of cell biology and disease treatment and prevention. Recently, we have developed a few proteomics tools to facilitate the discovery of the new function of PTMs. A highly robust N-glycoproteomics (HRN) platform comprising an automated enrichment method, a stable microflow LC-MS/MS system, and a sensitive glycopeptide-spectra- deciphering software tool (i.e. Glyco-Decipher) is developed for large-scale quantitative N-glycoproteome analysis, which enabled the identification of over 20,000 unique site-specific glycans and discovered one with good performance in diagnosing early-stage gastric cancer. A methyl-specific metabolic labeling approach was developed for the identification of methylations on 8 types of residues. Interestingly, of the identified methylation events, His methylation is found to be preferably occurred in C3H1 zinc fingers (ZFs). The His methylation is found to stabilize the structure of ZFs. U2AF1 is used as a proof-of-concept to highlight the functional importance of His methylation in ZFs in RNA binding and RNA metabolism. A special limited proteolysis approach, "peptide-centric local stability assay"(PELSA), was developed to enable proteome-wide target identification and binding-region determination of diverse ligands. Beyond offering users unprecedented sensitivity for characterizing diverse target-ligand interactions, it can also be applied to identify the reader proteins of PTMs. This was demonstrated by target proteins of a synthetic phosphotyrosine peptide pYEEI via using PELSA. It was founs that 28 proteins were significantly stabilized, among which 9 proteins contain SH2 domains, the reading domains for tyrosine phosphorylation. Our results demonstrated that PELSA can be readily applied to investigate the recognition domains of PTMs, which is crucial for understanding the biological functions of the PTMs. identifying the 24

KN-44 Inferring protein function from proteomics data Kathryn Lilley1 1University of Cambridge, Department of Biochemistry, Cambridge, United Kingdom There are between 2-10 billion separate protein molecules per average human cell depending on the cell type, which are encoded by approx. 20000 genes. There is an expansion in the number of different protein species per cell through a number of mechanisms including alternative splicing (potentially 70000 splice variants per cell), millions of potential single nucleotide polymorphisms, plus thousands of possible RNA base modifications within coding regions. Taken together, even before translation has taken place, the potential size of the proteome is vast. Coupled with the post translational processing of proteins through truncation and chemical modifications (enzyme catalysed and otherwise), the potential number of proteoforms in a cell could be in the range of the total number of proteins per cell! Although it isunlikely that every potential proteoform persists in a cell, the functional expansion of the proteome through the above mechanisms cannot be underestimated. The concept of one gene - one protein - one function is obsolete. Instead we are faced with the unenviable task of identifying the repertoire of proteoforms that are present in cells and ascribing a function to each. To date only a very small proportion of the proteome is experimentally verified, with a mere 5% of proteins being characterised with greater than 50% coverage, with approx. 10% of the predicted proteome typically yet to be detected by current methodologies. Moreover, only half of the human proteome carries accurate functional predictions, and these rarely take into consideration how different functions may be impacted by subtle changes in sequence, post transcriptional and post translational processing. Currently commonly used proteomics approaches are limited in capturing the true size of the functioning proteome, and importantly in assigning functions to proteoforms. Shotgun methods have resulted in disconnected information among intact protein species, and the lack of suitable tools means that many rarer PTMs are under sampled. Moreover, smaller proteins are often neglected and generally our ability to identify proteins is only as good as the gene models we have at our disposal In this presentation, I will present the problems that exist in defining the functional proteome and offer some insights in how we can use existing data and resources to infer function. Finally, I will look to the future to the large scale community efforts required to conquer the functional proteome. 25

KN-45 Bioorthogonal photocatalytic proximity labeling for spatiotemporal proteomics Ziqi Liu1, Xinyuan Fan1 1Peking University, Beijing, China Proteins are the direct executors of life processes, involved in metabolism, signal transduction, gene expression, protein synthesis, transport, and degradation, among others. Their functions within living cells often require precise spatiotemporal control, and any disruption can lead to severe diseases. The main techniques used for proteome profiling are enzyme-catalyzed labeling methods, such as APEX and BioID, which have limitations such as low time resolution and complex operations. Moreover, these methods are only limited to cells that are easy to genetic transfection, while hard-to-transfect cells, primary and clinical samples are hard to analyze. To address these issues, we focused on the development and exploration of chemical biological approaches, and developed a bioorthogonal photocatalytic labeling technology. This technique combines the unique advantages of photocatalytic chemistry and bioorthogonal reactions, providing precise spatiotemporal resolution in hard-to- transfect samples. It uses photocatalysts and external light to control the labeling of neighboring proteins in living cells (spatial control by targeting of the photocatalyst, and temporal control by the external light), allowing for in situ analysis and mapping of the proteome. The new generation of the chemical technology for proteomic research is expected to overcome the limitations of current techniques, offering a promising approach to understanding the complex roles of proteins in cellular processes. 26

KN-48 Neuro-symbolic AI reveals the functions of proteins in complex systems Robert Hoehndorf1 1King Abdullah University of Science and Technology, Computational Bioscience Research Center, Thuwal, Saudi Arabia The prediction of functions of proteins is crucial for understanding biological mechanisms and has significant implications in medical diagnostics. My talk will trace our decade-long exploration in designing novel machine learning and deep learning techniques to predict protein functions, described by the Gene Ontology (GO). I will discuss the evolution of these methods, highlight significant breakthroughs that have enhanced our ability to predict protein functions accurately and efficiently. A particular focus will be on the integration of neuro- symbolic AI methods, which combine machine learning with logical reasoning. These methods not only enable the incorporation of biological background knowledge into our models but also allow for zero-shot predictions -- - predicting functions for proteins without prior observed examples. I will demonstrate how these advancements have provided interpretable insights into protein function and have been used in identifying candidate genes for Mendelian diseases and enhancing variant prioritization for rare disease diagnosis. This approach not only broadens our understanding of protein functions in biotechnology, health, and disease, but also opens new avenues for protein design. 27

Oral presentations (OP-01 to OP-88) OP-01 Federated learning enables cancer subtyping by proteomics Zhaoxiang Cai1, Emma Boys1, Zainab Noor1, Adel Aref1, Dylan Xavier1, Natasha Lucas1, Steven Williams1, Jennifer Koh1, Rebecca Poulos1, Yangxiu Wu1, Michael Dausmann1, Karen MacKenzie1, Adriana Aguilar2, Carolina Niell3,4, Maria Barranco3,4, Mark Basik2, Elise Bowman5, Rory Clifton-Bligh6,7,8, Elizabeth Connolly1, Wendy Cooper9, Bhavik Dalal5, Anna de Fazio10,11,12, Martin Filipits13, Peter Flynn14, J Dinny Graham10,15, Jacob George6,16, Anthony Gill6,7,16,17, Michael Gnant18,19, Rosemary Habib10,20,21, Curtis Harris5, Kate Harvey22,23, Lisa Horvath6,24,25, Christopher Jackson26, Maija Kohonen-Corish27,28, Elgene Lim25, Georgina Long29, Reginald Lord30, Graham Mann31, Geoff McCaughan32,33, Lucy Morgan6, Leigh Murphy34, Adnan Nagrial6,35, Ben Panizza26, Jas Samra6,36, Richard Scolyer29, Ioannis Souglakos37, Alexander Swarbrick22,23,37, David Thomas38, Peter G. Hains1, Rosemary Balleine1,10, Phillip J. Robinson1, Qing Zhong1, Roger R. Reddel1 1The University of Sydney, ProCan, Children’s Medical Research Institute, Faculty of Medicine and Health, Westmead, Australia 2Jewish General Hospital, Lady Davis Institute at the Jewish General Hospital, Montreal, Canada 3Germans Trias i Pujol Research Institute (IGTP), Childhood Liver Oncology Group, Badalona, Spain 4Networking Biomedical Research Centre (CIBER) in Hepatic and Digestive Diseases, Barcelona, Spain 5National Institutes of Health, National Cancer Institute, Bethesda, MD, United States 6The University of Sydney, Faculty of Medicine and Health, Sydney, Australia 7Royal North Shore Hospital, Kolling Institute of Medical Research, St Leonards, Australia 8Royal North Shore Hospital, Department of Endocrinology, St Leonards, Australia 9Royal Prince Alfred Hospital, NSW Health Pathology, Camperdown, Australia 10The University of Sydney, The Westmead Institute for Medical Research, Westmead, Australia 11The University of Sydney, a joint venture with Cancer Council, The Daffodil Centre, Woolloomooloo, Australia 12Westmead Hospital, Department of Gynaecological Oncology, Westmead, Australia 13Medical University of Vienna, Center for Cancer Research, Vienna, Austria 14Nepean Hospital, Kingswood, Australia 15Westmead Hospital, Westmead Breast Cancer Institute, Westmead, Australia 16Westmead Hospital, Storr Liver Centre, The Westmead Institute for Medical Research, Westmead, Australia 17Royal North Shore Hospital, NSW Health Pathology, Department of Anatomical Pathology, St Leonards, Australia 18Medical University of Vienna, Comprehensive Cancer Center, Vienna, Austria 19Austrian Breast and Colorectal Cancer Study Group, Vienna, Austria 20Westmead Hospital, Crown Princess Mary Cancer Centre, Westmead, Australia 21Blacktown Hospital, Blacktown Cancer and Haematology Centre, Blacktown, Australia 22Garvan Institute of Medical Research, Cancer Ecosystems Program, Darlinghurst, Australia 23University of New South Wales, School of Clinical Medicine, St Vincent’s Clinical Campus, Sydney, Australia 24Chris O’Brien Lifehouse, Camperdown, Australia 25Garvan Institute of Medical Research, Darlinghurst, Australia 26Princess Alexandra Hospital, Department of Otolaryngology, Head and Neck Surgery, Brisbane, Australia 27Macquarie University, Woolcock Institute of Medical Research, Sydney, Australia 28Macquarie University, Macquarie Medical School, Faculty of Medicine, Sydney, Australia 29The University of Sydney, Melanoma Institute Australia, Sydney, Australia 30St. Vincent’s Centre for Applied Medical Research, Darlinghurst, Australia 31Australian National University, John Curtin School of Medical Research, Canberra, Australia 32Royal Prince Alfred Hospital, AW Morrow Gastroenterology and Liver Centre, Camperdown, Australia 33Centenary Institute, Camperdown, Australia 34University of Manitoba, Cancer Care Manitoba Research Institute, Manitoba Tumour Bank, Winnipeg, Canada 35Western Sydney Local Health District, Westmead, Australia 36Royal North Shore Hospital, Department of Upper GIT Surgery, St Leonards, Australia 37University Hospital of Heraklion, University of Crete, Department of Medical Oncology, Faculty of Medicine, Crete, Greece 38University of New South Wales, Centre for Molecular Oncology, Sydney, Australia Introduction: The availability of large quantities of online data is underpinning the rapid development of many artificial intelligence (AI) applications. However, in biological and healthcare settings AI applications face challenges, including data privacy and security. Federated learning (FL), a distributed machine learning method, enables analysis of large biomedical datasets while ensuring data privacy, confidentiality and regulatory compliance. In this approach, local models are trained using data held inside several participating sites, with the data remaining private behind local firewalls. Local model updates are then aggregated in the server to improve the global model. Methods: We obtained 30 distinct tissue sample cohorts, including a pan-cancer cohort (N = 1,260) from biobanks in Australia, with the remainder (N = 6,265) being largely single-cancer cohorts from centres in North America, Europe and Australia. We quantified 9,051 proteins across 7,525 tissue samples encompassing 57 28

cancer subtypes, using data-independent acquisition mass spectrometry (DIA-MS). From this we developed ProCanFL, a deep learning-based FL model that is trained on multiple simulated local sites using samples from the pan-cancer cohort and 90% of the 29 other cohorts. We simulated four local sites. The first site was fixed and housed the proteomic data and histopathology labels of the pan-cancer cohort, while sites 2-4 contained data from random combinations of the 29 cohorts. This randomization process was repeated 10 times to create 10 cohort allocations. A local model was trained at each site and the global model incorporated parameter updates from the local models. Results: We used a hold-out test set (10% of the 29 cohorts) to evaluate the performance of individual local models in predicting 15 cancer subtypes for which a minimum number of samples were represented in the proteomic dataset. The average accuracy across the four sites, evaluated on the unseen samples in the test set was 0.68±0.14. After application of ProCanFL, the global model had a significantly enhanced predictive performance, achieving an accuracy of 0.97±0.01. Conclusion: We demonstrated that FL enables a global model to be trained using local model updates from multiple decentralized sites containing private data, achieving high performance in cancer subtype prediction. Our study underscores the potential role of FL in the biomedical domain, where data sharing among global consortium members is often limited by several barriers, including country-specific data regulations. Further, the prediction accuracy from the FL model can be improved as more local sites and data are added. By harnessing the full wealth of large global datasets while keeping local data behind firewalls, FL offers a promising solution for developing omic foundation models and advancing the future of digital health. 29

OP-02 Proteogenomic analysis revealed prognosis biomarker and therapeutic targets in sarcomas Shuhei Iwata1, Shungo Adachi2, Rei Noguchi1, Yuki Adachi1, Julia Osaki1, Koichi Ogura3, Shintaro Iwata3, Seiji Ohtori4, Akira Kawai3, Tadashi Kondo1 1National Cancer Center Research Institute, Division of Rare Cancer Research, Tokyo, Japan 2National Cancer Center Research Institute, Department of Proteomics, Tokyo, Japan 3National Cancer Center Hospital, Department of Musculoskeletal Oncology, Tokyo, Japan 4Chiba University, Department of Orthopedic Surgery, Chiba, Japan Backgrounds: Sarcomas are rare mesenchymal malignant tumors. More than 100 histological subtypes have been characterized, and most of them have limited treatment options. Genomic analyses for sarcomas were reported, but proteomic analyses leading to diagnosis and treatment have not yet been conducted. Objective: The purpose of this study was to develop diagnostic or prognostic biomarkers and discover therapeutic targets in sarcomas. Methods: Proteogenomic analyses were performed on frozen tissues and cell lines derived from approximately 1,000 patients with 12 histologically different sarcomas. All materials including clinical and pathological information were obtained from the National Cancer Center Biobank, Japan. Proteins were extracted, purified with single-pot solid-phase-enhanced sample preparation (SP3), digested by Trypsin/Lys-C Mix, and subjected to the Orbitrap Astral mass spectrometer for quantitative proteomics using Data-Independent Acquisition (DIA). We integrated proteomic data and previously obtained genomic and transcriptomic data. Results: Sarcomas were classified into various proteomic clusters based on the expression pattern of 8000-9000 proteins. Pathway enrichment analysis revealed the specific molecular pathways unique to individual clusters. The integrated analysis showed correlations between genomic and proteomic data. Comparative analysis of proteomic and transcriptomic profiles highlighted the proteomic-specific features for optimal risk stratification. We identified proteins common to the poor prognosis group as prognostic biomarkers, and molecular pathways that could be therapeutic targets for molecular target drugs. Conclusion: This proteogenomic study revealed the utility of proteogenomics for identifying molecular subgroups with implications for risk stratification and therapy selection. We confirmed that this study provided a rich resource for future sarcoma research. We will conduct further analyses for the development of precision medicine for sarcomas. 30

OP-03 Integrating AI-based Digital Pathology and Proteome Profiling to Uncover Melanoma Progression and Heterogeneity Jéssica Guedes1, Nicole Woldmar1, Leticia Szadai2, András Kriston3, Ede Migh3, Henriett Oskolás1, Ferenc Kovács3, Roger Appelqvist1, Kun Hsing Yu4, Eugene R. Semenov4, Johan Malm1, Peter Horvath3, István Balázs Németh2, György Marko-Varga1, Jeovanis Gil Valdés1 1Lund University, Department of Translational Medicine, Lund, Sweden 2University of Szeged, Szeged, Hungary 3Biological Research Centre, Szeged, Hungary 4Harvard Medical School, Boston, MA, United States Melanoma, the deadliest form of skin cancer, is one of the fastest-growing cancers in the world. The marked molecular and morphological heterogeneity of this disease is directly connected to its aggressive behavior, progression, and treatment resistance. In this study, we integrate AI-based digital pathology and proteome profiling to deeply characterize histopathological features and melanoma subclones, and determine their molecular fingerprint in different melanoma cases to uncover the mechanism linked with progression. images, tumor and stroma areas, Using artificial intelligence approaches and high-resolution HE histopathological components, and melanoma subclones were laser microdissection, and analyzed by a proteomics approach. Over 100 FFPE samples were investigated, corresponding to primary tumors (n=84) and paired metastases (n=20). The samples were selected from patients who experienced disease recurrence within 5 years, with metastases at the moment of diagnosis, and those without any progression even after 5 years of the initial diagnosis. By applying this approach, we investigated the whole progression of melanoma from early-stage to advanced metastatic cases, decoding the pathways involved in the disease progression. isolated by identified, further The AI-based digital pathology results unveiled histopathological signatures in tumor cells and their surroundings, encompassing melanoma subclones and stromal components. Determining the proteomic signature of these areas revealed dysregulation of mitochondrial function and immune response-related pathways as the most significant drivers of progression. Proliferative pathways and histidine metabolism were up- and down-regulated, respectively, in line with the advancement of the disease. Additionally, suppression of immune system response in the tumor microenvironment of recurrent and metastatic melanomas underscores the importance of immunotherapies to improve patient outcomes in these groups of patients. The analyses of different melanoma subclones revealed molecular heterogeneity linked to metastatic melanoma and evidenced the critical role of mitochondria and histidine metabolism in melanoma progression, opening an opportunity to identify new druggable targets for therapy. Additionally, our study identified morphological markers associated with a high risk of melanoma recurrence and progression. Specifically, the presence of microerosion, budding formations, and melanocytic atypia, along with the absence of regression areas, were found to be significant risk factors for melanoma progression. Altogether, our findings highlighted that integrating AI-based digital pathology and proteome profiling is a powerful strategy to map melanoma progression, identify new druggable targets for treatment, and determine markers of recurrence and progression, which may illuminate new early-stage melanoma prognostic factors and guide a new era of personalized therapy. 31

OP-04 An integrated landscape of mRNA and protein isoforms Henrik Zauber1, Amir Kedan1, Mengran Wang1, Wei Chen1, Matthias Selbach1 1MDC Berlin, Proteome Dynamics, Berlin, Germany Cellular processes such as alternative splicing (AS) and proteolytic processing generate various protein isoforms from a single protein-coding gene. Studies using high-throughput sequencing indicate that > 95% of multi-exon genes in humans undergo AS [1,2]. In addition, proteolytic processing contributes to proteome complexity and is involved in processes like cell cycle regulation, cell signaling, and apoptosis. Current proteomic methods are not well suited to detect protein isoforms. Standard shotgun (bottom-up) proteomics involves digesting proteins into peptides. Although this approach identifies many proteins, it results in a loss of isoform information [3]. In contrast, mass spectrometric analysis of intact proteins (top-down proteomics) can distinguish protein isoforms but only covers a small subset of the proteome [4]. Here, we developed peptide correlation profiling (PepCP) as a method to obtain protein-level information from peptide-centric (bottom-up proteomic) data: First, proteins are fractionated by SDS-PAGE into polypeptides of different lengths. Second, individual protein fractions are digested into peptides. Third, peptides are identified and quantified across all fractions using quantitative mass spectrometry-based proteomics. Finally, peptide abundance profiles across fractions are analyzed to obtain protein-level information. We combined PepCP with long-read RNA sequencing to provide an integrated landscape of mRNA and protein isoforms in human RPE-1 cells. We established a reference set of 45,223 full- length transcripts and developed a computational framework to automatically detect different proteoforms at both the mRNA level (32612 protein isoforms from 12,400 unique genes) and protein level (16,983 protein isoforms from 8,168 unique genes), providing the largest available integrated dataset. Our data captures many well-known protein isoforms created by AS, alternative translation, and protein processing. Most of the detectable alternatively spliced transcripts are indeed detected at the protein level. In addition, we provide evidence for numerous new protein isoforms presumably generated by protein processing that cannot be observed at the mRNA level. Our data provides a unique resource of integrated mRNA and protein isoforms and paves the way for a deeper understanding of proteome complexity. 1. Pan, Q. et al. Deep surveying of alternative splicing complexity in the human transcriptome by high- throughput sequencing. Nat Genet 40, 1413–1415 (2008). 2. Wang, E. T. et al. Alternative isoform regulation in human tissue transcriptomes. Nature 456, 470–476 (2008). 3. Sinitcyn, P. et al. Global detection of human variants and isoforms by deep proteome sequencing. Nat Biotechnol 41, 1776–1786 (2023). 4. Tran, J. C. et al. Mapping intact protein isoforms in discovery mode using top-down proteomics. Nature 480, 254-258 (2011). Fig. 1 (optional) 33

OP-05 Multiomics characterization of in vivo resistance to anti-VEGF therapy in ovarian carcinoma Ilaria Piga1, Giulia Franciosa1, Oana Palasca1, Daniele Boso2, Sonia Minuzzo2, Lars Juhl Jensen1, Jesper Olsen1, Stefano Indraccolo1 1University of Copenhagen, Novo Nordisk Foundation Center for Protein Research (CPR), Copenhagen, Denmark 2University of Padua, Padova, Italy Epithelial ovarian cancer (OC) is the most lethal gynecologic malignancy in developed countries. Due to the lack of a screening strategy, the disease is usually diagnosed at an advanced stage and has a poor prognosis. In the advanced stage, antiangiogenic therapy combined with platinum-based chemotherapy represents a first-line clinical treatment option, but despite initial clinical benefit, resistance eventually occurs. The aim of this study is to investigate the mechanisms behind anti-VEGF therapy resistance in OC, by generating and analyzing multiomics data obtained from patient-derived xenografts (PDXs). We used a panel of seven PDXs, originally derived from the ascitic fluid of high-grade serous OC (HGSOC) patients and propagated in immune deficient mice. To a subset of the mice in each PDX group, we administered bevacizumab, a monoclonal antibody targeting VEGFA. To comprehensively profile the treated and wild-type mice within each group, we performed RNA-Seq, mass spectrometry (MS) proteomics, and whole exome sequencing (WES). In line with clinical data, the in vivo experiment reveals two mechanisms of resistance: three PDXs were intrinsically resistant to anti-VEGF therapy and the other four PDXs responded in the beginning but eventually developed resistance. To understand the reasons of these two mechanisms, we explore and integrate the three different layers of omics data. We analyze the mutational profiles of the different samples and identify genes differentially mutated in treated vs wild-type samples, and employ a proteogenomics approach to enhance the accuracy of peptide identification from MS data. We also examine the transcriptomics and proteomics profiles across tumors and replicates and identify genes potentially involved in the mechanisms of resistance. Genomic data reveals high mutational rates in all PDX models, as described in other studies on HGSOC. Mutations are particularly enriched in MUC, HLA and T53 genes. No significant mutations are enriched or selected during antiangiogenic therapy, suggesting that methylation might play an important role. Transcriptomics and proteomics data show that non-responders" models have a poor modulation at the transcriptome and proteome levels. Conversely, acquired resistant tumors display a wide transcriptome and proteome rewiring. Our data indicates that acquired resistance models have a strong upregulation of interferon signatures at both transcriptomics and proteomics levels. In addition, acquired resistance models downregulate cell cycle and DNA packaging signatures as a mechanism of drug adaptation. RNA splicing analysis uncovers that one model is producing a splicing VEGFA variant as a mechanism to escape VEGFA inhibition, while other models adapt to antiangiogenic therapy by modulating their metabolism. After disclosing the mechanisms of resistance, we will validate our findings in vivo through combination treatment and survival analysis. 34

OP-06 Multi-omics biomarker approaches in precision medicine Alain van Gool1, Hans Wessels1, Jolein Gloerich1, Purva Kulkarni1, Charissa Wijnands2, Anastasia Tzasta2, Gad Armony1, Dirk Lefeber3, Joannes Jacobs2, Peter-Bram 't Hoen4 1Radboud University Medical Center, Human Genetics, Nijmegen, Netherlands 2Radboud University Medical Center, Department of Laboratory Medicine, Nijmegen, Netherlands 3Radboud University Medical Center, Department of Neurology, Nijmegen, Netherlands 4Radboud University Medical Center, Department of Medical Biosciences, Nijmegen, Netherlands Introduction: Exponential developments in molecular omics technologies such as mass spectrometry and next generation sequencing have enabled us to obtain increasing insights in the molecular components of human biology and their interactions. Novel personalized diagnostics and high precision therapies that modulate selected disease mechanisms are now driving the new paradigm of precision medicine. However, we are rediscovering that human biology is a highly complex system and that we need multiple viewing angles to begin to understand disease mechanisms, identify its key nodes and define optimal personalized therapies. Measuring health using multiple biomarkers in a multi-omics approach can be a powerful way to achieve this. Proteomics takes up a unique part in multi-omics approaches, as it provides insights in enzymatic and structural components of biological pathways, yielding candidate biomarkers for diagnosis, prognosis, therapy prediction and monitoring. Optimally, multi-omics biomarker discovery, validation and development to clinical applications should be done in an integrated manner. Methods: In the selected case studies that will be presented, we applied an LC-MS proteomics workflow using the Evosep One LC (Evosep) and timsTOF Pro MS (Bruker Daltonics), in combination with the PaSER (Parallel Search Engine in Real-time; Bruker Daltonics) platform that enables real-time processing of PASEF DDA and dia- PASEF spectral data streamed from the timsTOF Pro to the PaSER box. Proteomics analysis was aimed at either expression proteomics or glycoproteomics in untargeted or targeted (glyco)peptide analysis. Other omics analysis (genomics by WES, methylomics by ME-Seq, transcriptomics by RNA-Seq, metabolomics by (un)targeted LC-MS, lipidomics by (un)targeted LC-MS, glycoproteomics by (un)targeted LC-MS) were performed at laboratories that are part of the Netherlands X-omics Consortium or as part of the EATRIS Biomarker Platform. Omics data analyses using various workflows including Multi-Omics Factor Analysis and Similarity Network Fusion were done at the X-omics data team at Radboudumc. Results: Lessons learned from two multi-omics case studies will be presented. A cohort of 120 healthy individuals was analysed on 7 omics levels, yielding information on natural variation and inter-individual differences of analytes in a healthy population. Induced pluripotent stem cells derived from patients with three types of inherited diseases were analysed by 8 omics platforms, yielding information on differences between patients and between iPSC clones. Two examples will be discussed where targeted proteomics analysis of (glyco)peptides is being translated to certified clinical diagnostics. 35

OP-07 Filter-aided expansion proteomics Zhen Dong1,2,3, Wenhao Jiang1,2,3, Jiayi Chen1,2,3, Chunlong Wu1,2,3, Pingping Hu1,2,3, Ting Chen4, Xuan Ding1,2,3, Shu Zheng4, Kiryl Piatkevich1,2,3, Yi Zhu1,2,3, Tiannan Guo1,2,3 1Westlake University, School of Medicine, Hangzhou, China 2Westlake Laboratory of Life Sciences and Biomedicine, Westlake Center for Intelligent Proteomics, Hangzhou, China 3Westlake University, Research Center for Industries of the Future, School of Life Sciences, Hangzhou, China 4Second Affiliated Hospital of Zhejiang University School of Medicine, Cancer Institute, Key Laboratory of Cancer Prevention and Intervention, Ministry of Education, Hangzhou, China Introduction: Understanding tissue heterogeneity is crucial for studying disease evolution and drug targets. High- throughput spatial proteomics with subcellular resolution can significantly advance this field. Current antibody- based techniques like CODEX offer subcellular resolution but are limited in throughput and antibody availability, hindering new protein biomarker discovery. Mass spectrometry (MS) methods like MALDI-MS do not require antibodies but can only characterize a limited number of peptides and proteins. Improvements in MS instruments have made it feasible to analyze minute tissue samples via laser microdissection (LCM), as demonstrated by methods such as LCM-nanoPOTs and LCM-DVP. However, LCM-based methods may compromise sample quality, rely heavily on expertise, and face technical bottlenecks for samples with resolutions below 10-20 µm. Methods: Here we present a spatial proteomics technology, termed filter-aided expansion proteomics (FAXP), using hydrogel-based tissue expansion and data-independent acquisition (DIA) MS for proteomics of formalin- fixed paraffin-embedded (FFPE) specimens. The FFPE tissue sections are embedded in customized hydrogel and can be expanded by a linear scale of 4-5 folds. After expansion, a minute tissue area, down to near single cell level, could be manually dissected and subject to DIA-based MS analysis. To achieve subcellular resolution, we further enhance FAXP by combining iterative tissue expansion with super- resolution imaging, LCM, robotics, and advanced MS techniques. This combination overcomes the resolution limits of LCM, enabling spatial subcellular proteomics analysis. Results: Compared to our original ProteomEx method1, FAXP, now partially automated via robotics, achieves a 14.5-fold increase in volumetric resolution of sample preparation. It generates a higher peptide yield, resulting in a 250% rise in protein identifications, while reducing sample preparation time by 50%, leading to identification of over 6700 proteins from FFPE specimens containing different stages of malignant changes of colorectal adenomas at specific positions. Building on FAXP, we iteratively expand cells and tissue samples from 512-fold to over 8000-fold in volume. This new strategy enables sub-100 nm super-resolution imaging using conventional confocal microscopes (e.g., Zeiss 980). Assisted by super-resolution imaging, we have delineated intricate subcellular organelle structures and identified nearly 2000 proteins from a single bona fide nucleus. Next, we plan to map smaller organelles like mitochondria and analyze them in AD mouse models for key factors. Conclusions: A novel spatial proteomics technology, FAXP, has been developed and implemented on clinical FFPE samples, enabling subcellular analysis guided by imaging. In this presentation, we will discuss its technical details, utility, pros, and cons. 1. Li, L. et al. Spatially resolved proteomics via tissue expansion. Nat. Commun. 13, 7242 (2022). 36

OP-08 Backside laser irradiation on histological slide sections for multi-omics applications: spot-by-spot LC-MS imaging Léa Ledoux1, Yanis Zirem1, Laurine Lagache1, Michel Salzet1, Isabelle Fournier1 1PRISM Laboratory, Villeneuve d'Ascq, France Introduction: Biological tissues exhibit complex spatial heterogeneity that directs the functions of multicellular organisms. Quantifying the expression of molecules is essential for elucidating processes within these complex biological assemblies. Indeed, spatial omics can provide deep insights into cellular compositions and interactions in tissues. Mass spectrometry imaging is a powerful emerging tool for mapping the spatial distribution of molecules across tissue surfaces. However, it does not permit the identification of detected molecules without post-analysis. Past efforts to achieve spatially resolved proteome across tissues have had limited proteome coverage and have relied on multi-step workflows. Here, we demonstrate an automated approach to imaging that uses label-free quantification omics (proteins, lipids, and metabolites) to analyze rat brain tissue sections with a spatial resolution of 220 µm. Methods: The SpiderMass technique is used here to microsample a region of tissue the size of the laser beam at the focal point (220um). The advantage, given its wavelength and energy, is that glass is transparent at the wavelength used, allowing irradiation to be performed from the back, with tissue ablation towards the front of the slide. This facilitates the collection of ablated tissue, which can then be processed for all types of omic analysis. Rat brains were irradiated, extracted tissue fragments are then collected in a tube containing 200 µl of ACN and subjected to a Folch extraction to recover three samples, which are used for simultaneous lipidomic, metabolomic, and proteomic analysis by LC-MS. Results: Various parameters were tested and refined to achieve optimal proteomic analysis via backside irradiation. Key parameters included tissue section thickness, extraction method, sample collection technique, protein digestion type, and slide type. These parameters were optimized using LC-MS analysis of 220 µm irradiation zone tenplicates in RB cortex. Excellent proteomic results (more than 5000 proteins) were obtained with 12 µm-thick tissue sections and Trypsin liquid digestion for both fresh and FFPE tissue. This methodology is compatible with white or ITO slides, demonstrating its feasibility for post-MSI analysis or post-HE staining. Additionally, this methodology was used to create 2D protein images of rat brain tissue sections, including various areas of interest. Using the LFQ, it was possible to map the expression levels of all detected proteins back onto the tissue image. This highlights the specific localization of certain proteins, which is of immense interest for future analyses of cancerous tissues, biomarker localization, and direct identification. The optimized parameters were similarly used to achieve high-quality results in additional omics analyses (lipids, metbolites). This comprehensive approach allows for a robust, multi-dimensional understanding of the molecular composition and spatial distribution within biological tissues. 37

OP-09 Deep visual proteomics advances human colon organoid models by revealing a switch to an in vivo-like phenotype upon xenotransplantation Frederik Post1, Annika Hausmann2, Sonja Kabatnik1, Sophia Steigerwald3, Alexandra Brand2, Ditte Lynge Clement2, Jonathan Skov2, Theresa Louise Boye4, Toshiro Sato5, Casper Steenholdt4, Andreas Mund1, Ole Haagen Jensen4, Kim Bak Jensen2, Matthias Mann1,3 1University of Copenhagen, Novo Nordisk Foundation Center for Protein Research, Copenhagen, Denmark 2University of Copenhagen, Novo Nordisk Foundation Center for Stem Cell Research, Copenhagen, Denmark 3Max Planck Institute of Biochemistry, Department of Proteomics and Signal Transduction, Martinsried, Germany 4Herlev Hospital, Department of Gastroenterology, Herlev, Denmark 5Keio University School of Medicine, Department of Integrated Medicine and Biochemistry, Tokyo, Japan Introduction: Organoid models are essential tools in disease modeling and regenerative medicine, providing a unique platform to closely mimic human tissue architecture and function. However, it is important to assess to which degree these models accurately replicate physiological features. Likewise, addressing safety concerns in organoid transplantation is critical, as cellular misbehavior after transplantation can prevent integration and functionality in clinical settings. This study introduces drastic improvements to our Deep Visual Proteomics (DVP) technology, overcoming limitations in sensitivity and depth, thus providing a colon mucosa atlas and profound insights into organoid cellular behaviors. Methods: Organoids were cultured under in vitro conditions or transplanted into murine colons to simulate a more dynamic biological environment. Simultaneously, native human colon samples were analyzed to serve as the in vivo benchmark of intestinal epithelial cells (IECs). To characterize 136 samples consisting of ~100 to 500 cell contours, we improved our DVP technology, which integrates high-resolution fluorescence imaging with AI- guided cell segmentation and classification, laser microdissection, and mass spectrometry-based proteomics. We dramatically increased the sensitivity of the proteome acquisition by coupling the Evosep One running the Whisper40 nanoflow gradient on a 15 cm ionopticks Aurora column to the Orbitrap Astral mass spectrometer. This setup enabled us to detect 8,865 unique proteins across samples and ~5000 unique proteins from only ~20 intestinal stem cells isolated from transplanted human colon organoids. Results: Our improved DVP technology revealed remarkable insights into the proteomic landscape of IECs. We created an in-depth spatial proteome atlas of the human colon mucosa that serves as a critical reference for assessing organoid models and future research on human intestinal biology. Comparing proteomes of IECs in vivo and those grown as organoids in vitro revealed a robust correlation. However, IECs grown in vitro had more proliferative and lower functional signatures compared to their in vivo counterparts. Xenotransplantation of human colon organoids into the murine colon as well as adjusting the culture conditions led to a reversion of the proliferative phenotype to a more in vivo-like state. Outlook: By increasing the sensitivity of DVP, we are now able to substantially cover single cell-type proteomes in vivo. This enhanced our understanding of IEC behavior in different biological systems and provided a valuable reference for researchers looking to design more accurate in vitro models that better mimic in vivo conditions. The ability of xenotransplanted organoids to revert to an in vivo-like phenotype highlights their potential for use in regenerative medicine. This includes applications in treating diseases such as inflammatory bowel disease by replenishing impaired IECs. 38

OP-10 Characterizing the target space of molecular glues and bifunctional degraders in cells and animal models by bottom-up proteomics Patrick Ruether1, Niklas Gombert1, Sebastian Krebs1, Beate Scholz1, David Banczyk1, Philipp Cromm1, Ulrich Bothe1, Anja Giese1, Heike Schaecke1, Laura Luh1, Christian Karl Frese1 1Bayer AG, Drug Discovery Sciences, Wuppertal, Germany Targeted protein degradation facilitates the development of selective therapies against protein targets that were previously deemed "undruggable" by conventional inhibitors, for instance because they lack a distinctive active site. Characterizing the biological activity of degrader compounds, including proteolysis-targeting chimeras (PROTACs®) and molecular glues, is both easy and hard: While degrader targets can be readily identified by global quantitative proteomics, the effects can be highly tissue-specific, as they depend not only on the main target but also on the simultaneous expression, activity, and localization of a ubiquitin ligase complex. Therefore, it is desirable to validate target degradation and search for potential off-targets across many different biological models ranging from cell culture and primary cells to tissues from animal models. Such experiments with large cohorts and often limited sample amounts require high throughput analytics with good sensitivity. Here, we present a streamlined and largely automated bottom-up proteomics workflow quantifying up to 9,000 proteins at a throughput of 30 samples per day. As a proof-of-concept, we confirmed the (off-) targets of the clinical compounds Indisulam, Lenalidomide, and Pomalidomide in a human T-cell model. To narrow down the main targets and rule out secondary effects, we performed further experiments in dose-response, with a time course, and with pharmacological co-treatments to selectively inhibit proteasomal degradation or Cullin-RING ligases. To demonstrate applicability for more complex biological models, we studied a bifunctional degrader in murine liver, spleen, and adipose tissue and in human peripheral blood mononuclear cells (PBMCs). Our data successfully confirmed target engagement in all four matrices including the sample-limited PBMCs and was key in the discovery of a previously unknown off-target. Furthermore, we characterized a PROTAC® in a patient- derived xenograft (PDX) model at four time points, each at four different doses. After 48 h of MS analysis time, we obtained global proteome data with time- and dose-resolution that helped us to validate target degradation and identify potential off-targets. Taken together, we established a global degrader target identification method that can be universally applied to a range of relevant disease models and offers relatively high throughput und sensitivity. We anticipate that our method will accelerate drug discovery by providing a detailed picture of the target landscapes of degrader modalities, regardless whether they come from phenotypic screens or from rational drug design. 40

OP-11 Combining chemoproteomics and phenotypic screening for identifying STING modulators Shelly Kalaora1, Shane Dawson1, Joao Paulo1, Steven Gygi1 1Harvard Medical School, Cell Biology, Boston, MA, United States The cGAS–STING pathway is a crucial component of the innate immune system that functions to detect the presence of cytosolic DNA and, in response, trigger the expression of inflammatory genes, therefore playing a critical role in autoimmune diseases, inflammation, and cancer. Yet, there are no approved STING inhibitors in the clinic. By high-throughput phenotypic screening of thousands of small covalent molecule fragments, we were able to identify >30 compounds that inhibit STING pathway activation. To understand the compound"s mechanism we investigated the cysteines to which the compounds bind using activity-based protein profiling (ABPP), followed by an in-depth quantitative whole proteome and phospho-proteome analyses to identify the proteome-wide changes by the compounds. Although different compounds demonstrated varying levels of STING modulation, by the profiling of the cysteines to which the compounds bind, we identified that many of them target STING cysteine 91 (C91), resulting in inhibition of STING palmitoylation. The phospho-proteome and whole proteome analyses complemented this observation and validated the changes resulting from STING inhibition. In addition to identifying the binding site of the compounds, the cysteine profiling approach revealed changes in the reactivity of other cysteines. We observed changes in the reactivity of STING C148 and C206, suggesting higher availability of these positions to the desthiobiotin iodoacetamide probe due to STING conformational changes. As well as changes in the availability of cysteines in STING interactors, for example, IRF3 C371 which resides in the proximity of STING-IRF3 interaction. These examples demonstrate the utility of this approach for the understanding of the global changes in protein interactions and conformation by the compounds. Identifying structurally variable compounds in the screen directs us to improve the specificity of the compounds to STING. In addition, ABPP results enable us to identify off target proteins to which the compounds bind. Together, this information led us to design a library of elaborated compounds and tested their binding specificity by ABPP and their activity using the functional assay, and identifying compounds with improved specificity and binding to STING. The combination of high-throughput phenotypic screening and proteomic approaches facilitates the identification of functionally relevant inhibitors and their mechanisms. 41

OP-12 Rationale prediction of drug combinations based on large-scale perturbation proteomics Liujia Qian1, Rui Sun1, Zhangzhi Xue1, Xuedong Zhang1, Guangmei Zhang1, Kunpeng Ma1, Yi Zhu1, Tiannan Guo1 1Westlake University, Hangzhou, China identifying successful drug repositioning combinations. We propose The pharmaceutical industry is hindered by high failure rates, steep expenses, and sluggish drug development. Drug repositioning can alleviate these challenges by slashing costs, expediting development, and unveiling new targets for existing drugs, but standalone efficacy is often limited by low plasma concentrations. Drug combinations may enhance repositioning success, but the vast number of possible pairings is a significant obstacle. Current computational methods, which rely heavily on genomic and transcriptomic data, fall short in systematically integrating a drug repurposing library and post-perturbation proteomic data to enhance the success rate of drug combination repositioning. In our study, we focused on three cell lines, namely NCI-H1299, HCT116, and 786-O. To identify effective drug combinations, we implemented a 2x7 concentration matrix approach, using 25 cancer-specific anticancer drugs as 'anchor' compounds, each at two optimized concentrations. Leveraging the molecular representation learning model, Uni-Mol, we carried out unsupervised clustering on 3000 repurposed drugs, and picked 192 library drugs, each set at seven concentration points across a 1000-fold range. We conducted perturbed proteomic analysis of 3000 drugs, and 300 each of effective and ineffective drug combinations on these cell lines at two time points (6 and 24 hrs), using data-independent acquisition on Orbitrap Exploris™ 480 Mass Spectrometers. We generated over 40,000 viability measurements for over 4000 pairwise drug combinations across three cell lines—the largest dataset of its kind focusing on repurposed drugs for anticancer regimens. Combinations demonstrating shifts beyond Bliss in potency (ΔIC50 > 3) or efficacy (ΔEmax > 0.2) were classified as synergistic. Notably, less than 10% of the combination-cell line tuples exhibited synergy, underlining the rarity of effective combinations. Furthermore, synergistic drug combinations showed a strong correlation with tissue origin. Next, we produced over 20,000 perturbed proteomic data, quantifying over 10,000 proteins with high confidence and technical reproducibility. Unsupervised clustering showed cell line-specific proteome responses to perturbations, highlighting the cell context-dependence of the perturbation proteome. We compared perturbed proteomic data between synergistic and ineffective drug combinations to uncover potential molecular mechanisms of synergy effects. Utilizing baseline genomic data, perturbed proteomic data, and SMILE as training inputs, we developed a neural Ordinary Differential Equations model to predict the efficacy of independent drug combinations from a pool of repurposing drugs and anticancer drugs. The model achieved an AUC above 0.80. We further validated a few representative drug synergies in cell line models and explored the targets of the selected drug combinations by thermal proteome profiling experiments. Fig. 1 (optional) 42

OP-13 Absolute quantification of the synaptic blood marker β-synuclein by targeted mass spectrometry ranks synaptic degeneration after amyloid deposition in autosomal dominant Alzheimer's disease Patrick Oeckl1,2, Benjamin Mayer3, Randall J. Bateman4, Gregory S. Day5, Nick C. Fox6, Edward D. Huey7, Laura Ibanez8, Takeshi Ikeuchi9, Mathias Jucker10,11, Jae-Hong Lee12, Johannes Levin13,14,15, Jorge J. Llibre-Guerra4, Francisco Lopera16, Eric McDade4, John C. Morris8, Yoshiki Niimi17, Jee Hoon Roh18, Raquel Sánchez-Valle19, Peter R. Schofield20,21, Markus Otto22 1German Center for Neurodegenerative Diseases (DZNE) Ulm, Ulm, Germany 2Ulm University Hospital, Neurology, Ulm, Germany 3Ulm University, Ulm, Germany 4Washington University School of Medicine, Saint Louis, WA, United States 5Mayo Clinic in Florida, Jacksonville, FL, United States 6UCL Queen Square Institute of Neurology, London, United Kingdom 7Alpert Medical School of Brown University, Providence, RI, United States 8Washington University, Saint Louis, WA, United States 9Niigata University, Niigata, Japan 10German Center for Neurodegenerative Diseases (DZNE) Tübingen, Tuebingen, Germany 11University of Tübingen, Tuebingen, Germany 12University of Ulsan College of Medicine, Asan Medical Center, Seoul, South Korea 13LMU University Hospital, LMU Munich, Munich, Germany 14German Center for Neurodegenerative Diseases (DZNE) Munich, Munich, Germany 15Munich Cluster for Systems Neurology (SyNergy), Munich, Germany 16Antioquia University, Medicine Faculty, Medellin, Colombia 17The University of Tokyo, Tokyo, Japan 18Korea University College of Medicine, Seoul, South Korea 19University of Barcelona, Barcelona, Spain 20Neuroscience Research Australia, Sydney, Australia 21University of New South Wales, Sydney, Australia 22Department of Neurology, Martin-Luther-University Halle-Wittenberg, Halle (Saale), Germany Introduction: Synaptic degeneration is a major hallmark of Alzheimer´s disease (AD) and the pathological correlate of memory impairment. Monitoring of synaptic changes in AD patients is therefore of importance for diagnosis, prognosis and assessing disease progression but also to track beneficial effects on synaptic loss in clinical trials. The protein β-Synuclein is a promising and easily accessible blood marker to track synaptic degeneration in AD but changes in preclinical AD are unclear. The Dominantly Inherited Alzheimer Network (DIAN) observational study is a worldwide multicenter initiative to longitudinally study dominantly inherited AD and provides the unique opportunity to study blood β-synuclein levels in very early (asymptomatic) AD and assess the timing and relation of β-synuclein changes to other pathological alterations. Methods: We used immunoprecipitation and multiple reaction monitoring mass spectrometry (IP-MS) with an Eksigent MicroLC200 and a QTRAP6500 for the absolute quantification of β-synuclein in serum samples of 178 participants of the DIAN study including 69 cognitively unimpaired mutation non-carriers (NC), 78 cognitively unimpaired AD mutation carriers (asymptomatic AD, aMC) and 31 symptomatic MC (sMC). Changes of β- synuclein serum levels were compared and correlated with cerebrospinal fluid (CSF) biomarkers (Aβ42/40, tTau, pTau variants), amyloid deposition ([11C]-Pittsburgh Compound B - positron emission tomography (PiB-PET)), brain atrophy (magnetic resonance imaging (MRI)) and metabolism ([18F]-fluorodeoxyglucose - PET (FDG-PET)), axonal degeneration (serum NfL) and cognitve impairment (MMSE, Clinical Dementia Rating sum-of-boxes (CDR- SB), DIAN cognitive composite). Longitudinal trajectories were estimated using the expected years to symptom onset (EYO). Results: Blood β-synuclein levels were already higher in asymptomatic autosomal dominant AD (ADAD) mutation carriers (median 7.29pg/mL, interquartile range (IQR) 5.89-8.59) compared with non-carriers (5.25pg/mL, IQR 4.63-6.66) and highest in sMC (14.4pg/mL, IQR 11.7-18.4). Longitudinal trajectories based on the EYO and correlation analyses indicated that β-synuclein levels start to rise after amyloid deposition preceding axonal degeneration, brain atrophy and hypometabolism and cognitive decline. β-synuclein levels were associated with cognitive increased with declining cognition. Thus, our data rank synaptic degeneration as one of the earliest events in the ADAD pathophysiological cascade as it is close to the start of amyloid deposition and it supports a retrograde degeneration mechanism starting in the synapse. impairment and gradually 43

OP-14 Identification of novel diagnostic protein biomarkers in human plasma for breast cancer by targeted protein assay Kristrun Yr Holm1,2, Finnur Eiriksson1,3, Yassene Mohammed4,5,6, Christoph H. Borchers4,6,7,8, Sigridur Klara Bodvarsdottir2, Margret Thorsteinsdottir1,3,2 1University of Iceland, Faculty of Pharmaceutical Sciences, Reykjavik, Iceland 2University of Iceland, Biomedical Center, Reykjavik, Iceland 3ArcticMass, Reykjavik, Iceland 4McGill University, Segal Cancer Proteomics Centre, Montreal, Canada 5Leiden University Medical Center, Center for Proteomics and Metabolomics, Leiden, Netherlands 6Jewish General Hospital, Lady Davis Institute for Medical Research, Gerald Bronfman Department of Oncology, Montreal, Canada 7McGill University, Division of Experimental Medicine, Montreal, Canada 8McGill University, Department of Pathology, Montreal, Canada Introduction: Breast cancer (BC) is the most common cancer among women worldwide and ranks as the second leading cause of cancer-related deaths. Fortunately, the prognosis of BC is good when detected at an early stage, however, sensitive diagnostic tools for early detection of BC are vital for improving survival rates. This study aims to identify protein biomarkers with diagnostic and prognostic value in human plasma to improve early BC diagnosis. Methods: An absolute quantification of 131 proteins was performed on 279 well-defined Icelandic biobank- based plasma samples from 135 BC patients and 144 healthy controls by UPLC-MRM-MS assay. Among the BC patients, 45 were BRCA2 mutation carriers. Absolute quantification of the proteins were conducted using a PeptiQuantTM protein human kit containing synthetic light peptide and matching heavy peptide as internal standard for each protein. Prior to analysis the plasma samples were proteolytically cleaved with trypsin, internal standards were added to the peptide mixture and concentrated by solid-phase extraction using liquid handling robot. Data analyses were conducted using Skyline Quantitative Analysis software for pre-processing of the data. SIMCA Pro-17, R studio, and Python were used for statistical analysis, multivariate data analysis, and machine learning. Predictive models including random forest, support vector machine, and logistic regression were developed and validated through cross-validation using the SciKit-Learn library. Results: The targeted proteomic assay was successfully implemented for absolute quantification of 131 proteins in human plasma samples with precision and accuracy for calibration standards and quality controls within 20% relative standard deviation. Out of the 131 proteins, 99 proteins were quantifiable in the Icelandic bio-bank plasma samples, surpassing the lower limit of quantification. The samples were analyzed in eight batches, each containing 25 cases and 25 controls, showing minimal batch-to-batch variability. The results reveal that protein signatures of the patients with a BRCA2 germline mutation are different from patients with sporadic BC. This difference was found to be significant for few proteins in comparison to control groups as well as between the patient groups. Notably, predictive models could distinguish between patients with sporadic BC from BC patients with BRCA2 mutation with AUC > 0.8, offering valuable insights for targeted screening for BC. Furthermore, seven proteins showed the potential to discriminate BC patients with the luminal B subtype from controls. Conclusions: Targeted proteomics using UPLC-MRM-MS shows potential for identifying and quantifying protein biomarkers in human plasma for diagnosis of BC in BRCA2 mutation carriers and for the luminal B subtype. 45

OP-15 Preoperative diagnosis of thyroid nodules using a multi-omics test Xue Cai1, Yaoting Sun1, Lu Li1, Zhangzhi Xue1, Yi Zhu1, Yu Wang2, Tiannan Guo1 1Westlake University, School of Medicine, School of Life Sciences, Hangzhou, China 2Fudan University, Oncology, Shanghai, China Background: Approximately 15-30% of thyroid nodules display indeterminate cytology on fine needle aspiration (FNA) biopsy, creating a clinical diagnostic dilemma in determining their benign or malignant nature. Multiple gene-panel tests have been developed; however, these tests also possess certain limitations, such as heterogeneity. Protein samples are less susceptible to spontaneous degradation in clinical specimens. We hypothesized that the introduction of proteomics molecular testing may help tackle this clinical challenge. Methods: We conducted a prospective, noninterventional, multicenter study. All patient's FNA samples were classified using the Bethesda System for Reporting Thyroid Cytopathology. Results of a histopathological review served as the reference standard. Proteins are extracted and digested from FNA samples and peptide samples mixed with synthesized stable isotope labeled peptides (called AQUA peptides) were analyzed using multiple reaction monitoring (MRM) mass spectrometry (MS) on a Sciex 4500 MD instrument over an 8-min LC gradient. MS data were analyzed using OpenMS. BRAF V600E mutation status was determined by ARMS-PCR Kit. A machine learning approach was employed, utilizing the Sklearn package to construct a random forest model. Results: A total of 1954 FNA samples from 1952 patients collected from 15 medical centers from January 2020 to September 2022 were used for feature selection, model training and validation. Based on previous studies, 211 proteins were selected as candidate diagnostic biomarkers. After multiple rounds of protein biomarker screening, including machine learning model screening, evaluation of AQUA peptides stability and linearity, and MS signal screening, three proteins were ultimately selected as the final features of the classifier model (ThyroProt). The ThyroProt incorporated the absolute quantitative data of three proteins, along with age, sex, and BRAF V600E mutation information. Independent test set is composed of 290 FNA samples collected from 283 patients from five medical centers, spanning from October 2022 to March 2023. Classifier test on independent set 1 showed an area under the curve (AUC) of 0.94, an accuracy of 89.7% (95% confidence interval [CI]: 89.6-89.7), sensitivity of 85.9% (95% CI: 80.7-91.0), specificity of 95.6% (95% CI: 91.8-99.4), positive predictive value (PPV) of 96.8% (95% CI: 94.1-99.6) and negative predictive value (NPV) of 81.2% (95% CI: 74.6-87.8) for thyroid malignancy. Test on 45 FNA samples reported as Bethesda III or IV showed an AUC of 0.87 and an accuracy of 88.9% (95% CI: 88.5-89.3), sensitivity, specificity, PPV, and NPV of 86.7% (95% CI: 69.5-100), 90.0% (95% CI: 79.3-100), 81.2% (95% CI: 62.1-100), and 93.1% (95% CI: 83.9-100), respectively. Conclusions: This study demonstrates the feasibility and practicality of using targeted mass spectrometry to detect multiple proteins and machine learning for clinical diagnostics in real-world scenarios. 46

OP-16 Large scale cross-linking to discover, evaluate, and validate the human cell structural proteome and interactome Tara Bartolec1,2, Xabier Vázquez-Campos2, Alexander Norman3, Clement Luong4, Marcus A. Johnson4, Richard J. Payne3,5, Marc Wilkins2, Joel P. Mackay4, Jason K.K. Low4 1EMBL Heidelberg, Genome Biology, Heidelberg, Germany 2University of New South Wales, Systems Biology Initiative, School of Biotechnology and Biomolecular Sciences, Sydney, Australia 3The University of Sydney, School of Chemistry, Sydney, Australia 4The University of Sydney, School of Life and Environmental Sciences, Sydney, Australia 5The University of Sydney, Australian Research Council Centre of Excellence for Innovations in Peptide and Protein Science, Sydney, Australia structure modellers have enabled access Recent advances in structural biology have expanded our ability to create experimental structures for proteins and complexes, but many proteins remain refractory to these approaches or have not yet been analysed. Machine-learning based to highly accurate protein structure predictions for entire proteomes. These modellers are trained on experimental structures available in the Protein Data Bank (PDB), which constitute a relatively small subset of proteins, with many structures solved using non-native conditions or sequences. Therefore, a critical question is whether predicted (and PDB) structures reflect the bona fide structures and complexes formed by proteins in their native environment. We investigate this question using a large-scale cross-linking mass spectrometry (XL-MS) resource generated for the human cell. To generate a high density and high depth XL-MS dataset for human HEK293 cells, we utilised a multipronged approach. Briefly, we cross-linked four subcellular fractions (nucleus, endoplasmic reticulum, mitochondria, cytosol) using three different cross-linkers with orthogonal chemistries (DHSO, DSSO, DMTMM). Then, we enriched cross-linked peptides using offline size-exclusion chromatography followed by further fractionation by high pH reverse phase HPLC. Mass spectrometry was performed on concatenated fractions using hybrid-MS2- MS3, or MS/MS with EThcD or HCD, fragmentation strategies. Cross-linked peptides were identified using XlinkX 2.3 or pLink2, using stringent search parameters and post-hoc filtering to control the false discovery rate to <2% at the unique residue pair (URP) or PPI levels. We identified 28,910 URPs representing 4,084 unique proteins and 2,110 unique putative PPIs. Subcellular fractionation before cross-linking resulted in significantly improved proteome coverage, whilst orthogonal reactivities (D/E-D/E, K-K and K-D/E) improved the density of cross-linking per protein, especially in intra-protein links. We demonstrate that our resource of URPs confirm and rediscover existing experimental structures, capturing proteoforms and complexes within their approximate subcellular niches and range of conformations. Remarkably, our intra-molecular URPs also largely corroborate thousands of new structures predicted by next- generation modeller AlphaFold2, including those involving proteins (or regions of proteins) without existing resolution, and those lacking any structural precedent. Furthermore, our inter-protein crosslinks recapture the topology of well-described complexes and PPIs, whilst supporting or discovering poorly characterised PPIs. Finally, the inter-protein crosslinks also help localise PPI interfaces, and we use this information to assess quaternary protein structures modelled in AlphaFold-Multimer. 47

OP-17 Advancing stability proteomics for functional characterization of intact proteoforms Kellye Sutton1, Anju Teresa Sunny1, Yanting Guo2, Trishika Chowdhury1, Patrycja Szamweber1, Si Wu1 1University of Alabama, Chemistry and Biochemistry, Tuscaloosa, AL, United States 2University of Oklahoma, Chemistry and Biochemistry, Norman, OK, United States Introduction: High-throughput top-down proteomics has significantly advanced the characterization of intact proteoforms in complex biological systems. However, linking proteoform structure with function remains challenging. Stability proteomics methods assess protein functionality by analyzing stability under a denaturation gradient, but bottom-up proteomics techniques limit their ability to characterize intact proteoforms with modifications. To address this, we developed top-down stability proteomics methods, including Top-Down Thermal Proteome Profiling (TD-TPP) and Methionine Oxidative Footprinting in Intact Proteins (MOFIP), using thermal and chemical denaturant gradients to probe protein stability. These methods were employed to study the effect of protein modification on proteoform stability in complex samples, aiming to bridge the gap between structure and function. Methods: We developed TD-TPP and MOFIP to probe proteoform stability in E. coli, mouse brain, and HeLa lysates. For TD-TPP, aliquots were heated (37-56 °C) and centrifuged to remove aggregates. For MOFIP, lysates were treated with H2O2 and quenched with L-methionine. They were then incubated with increasing concentrations of guanidinium hydrochloride (GdmCl) followed by H2O2 incubation and quenching. Label-free quantitative TD proteomics was performed using our one-dimensional ultra-high-pressure liquid chromatography top-down mass spectrometry (1D UPLC-TDMS) platform for deep characterization of intact proteoforms. Results: Our top-down platform enabled the characterization and profiling of hundreds of proteoforms in complex biological samples. Using TD-TPP, we analyzed the standard proteins β lactoglobulin A&B. βLG B was slightly destabilized as a result of an amino acid substitution (D80G) to a more nonpolar residue on an exposed turn as well as the exchange to a more polar residue (V134A) internal beta-sheet. TD-TPP was further applied to HeLa lysate and determined the average melting point of the intact proteoforms was 59.3 °C for cytosolic proteoforms, consistent with previous reports. The MOFIP was applied to evaluate solvent-accessible Met residues based on their secondary structure. Interestingly, 26 proteoforms were partially oxidized while natively folded indicating multiple populations may co-exist in cell lysate. Furthermore, proteoform stability was examined and one protein, the Ribose import binding protein (RbsB), contains 1 solvent-accessible methionine residue on domain 1 and 3 solvent-inaccessible methionine residues on domain 2. We found that the three residues on domain 2 were simultaneously oxidized upon denaturation demonstrating the potential of MOFIP to obtain domain-specific information. Conclusions: Our novel top-down stability proteomics platforms have shown great potential for deep proteoform characterization and stability evaluation at the proteoform level. 48

OP-18 Maintenance of Mitochondrial Complexes Ensures Bioenergetic Function in Differentiated Cells Parviz Ghezellou1, Juliana Heidler1, Heiko Giese2, Ralf P. Brandes1, Ilka Wittig1 1Goethe University Frankfurt, Institute for Cardiovascular Physiology, Frankfurt, Germany 2Goethe University Frankfurt, Molecular Bioinformatics, Frankfurt, Germany The assembly sequence of mitochondrial complexes has been intensively investigated in proliferating cells, primarily focusing on de novo assembly. However, these studies provide limited insights into the dynamics of protein complexes in differentiated cells and tissues. In postmitotic tissues, the state of protein complexes is likely maintained through a balance of biosynthesis and degradation. This raises an important question: are protein complexes continuously assembled de novo, or do remodeling and repair mechanisms play a crucial role in maintaining mitochondrial function? To address this, we utilized complexome profiling, a technique combining blue native electrophoresis with quantitative mass spectrometry, able to identify scarce sub-complexes, assembly intermediates, and complex remodeling events. Furthermore, we integrated complexome profiling with pulse stable isotope labeling of amino acids in cell culture (Pulsed-SILAC) to investigate the turnover and half-life of individual proteins within protein complexes in differentiated post-mitotic C2C12-myotubes. Our findings provide a comprehensive dataset on the dynamics of stable mitochondrial protein complexes. We determined that the complete replacement of all complexes within the oxidative phosphorylation system (OXPHOS) requires approximately one month in mouse myotubes. Notably, we observed higher turnover rates among the interface subunits of dimeric and oligomeric ATP synthase and supercomplexes, indicating more dynamic interactions between complexes rather than within them. We further explored the maintenance mechanisms of complex I. Newly synthesized proteins are incorporated as spare parts, replacing modules within complex I. Applying this method to patients with mitochondrial disorders and mouse models revealed specific service factors involved in the repair process. In particular, we identified complex I subunits with higher turnover rates in the electron transport modules, suggesting that quality control and repair mechanisms within assembled complexes are essential for maintaining full bioenergetic function in postmitotic tissues. 49

OP-19 Advancing medical diagnosis with fully untargeted proteomics-based virus identification Clément Lozano1, Olivier Pible1, Jean Armengaud1 1CEA, Bagnols-sur-Cèze, France Introduction: The fast and accurate identification of microorganisms is essential for medical diagnosis. Traditional microbiological methods rely on cultivation, which is time-consuming and fails to detect all microorganisms. DNA amplification-based methods, such as PCR, require prior knowledge of the microorganism's genome and allow only the detection of the targeted species. MALDI-TOF only permits the detection of species included in a dedicated spectral library and performs poorly on unknown pathogens or in the presence of host material. In this context, a universal method that does not require prior knowledge of the sample components and is not hindered by host proteins would greatly advance the detection and identification of pathogens. Here, we present a groundbreaking tandem mass spectrometry proteotyping method for rapid and universal virus identification, eliminating the need for predefined targets, and permitting detection from multiple matrices of clinical interest. Methods: We evaluated our proteotyping method on a range of viruses including SARS-CoV-2, Vaccinia virus, and Monkeypox virus, among others. To demonstrate its clinical relevance, we tested the method using various clinically significant matrices, such as positive saliva and feces samples. Proteins were extracted and digested via microplate-adapted SP3 protocol. The resulting peptides were subjected to a 30 min LC-MS/MS analysis using a Vanquish Neo UHPLC coupled to an Exploris 480 mass spectrometer (ThermoFisher). The resulting spectra were analyzed using an in-house tool, based on a cascade search permitting the taxonomical identification of the sample components using a generalist database. Results: Based on the combination of shared and taxa-specific peptides, our method successfully identified all viruses. Additionally, the other Eukaryotic and Prokaryotic organisms present in the sample were identified. Notably, SARS-CoV-2 and Vaccinia virus were detected from saliva, a complex biological matrix, from 106 pfu/ml, a clinically relevant concentration. In addition, rotavirus from gastroenteritis fecal samples was successfully identified, along with the different microorganisms, depicting the full sample microbial pictures in a single query. The entire experimental workflow, including digestion in 96-well plates and short LC-MS/MS analyses is straightforwardly amenable to high throughput. Such detection procedures pave the way for a rapid, accurate, and non-invasive diagnosis. Fig. 1 (optional) 50

OP-20 Mapping the Crosstalk Between Cellular or Viral-mediated Proteolysis and Ubiquitination and Phosphorylation During Norovirus Infection Leandro Neves1, Philip Brownridge1, Edward Emmott1 1University of Liverpool, Centre for Proteome Research, Liverpool, United Kingdom Human noroviruses are the leading cause of viral gastroenteritis accounting for>200,000 deaths per annum and major disruption to healthcare systems worldwide. Despite its medical relevance, our knowledge of human norovirus biology is limited given difficulties in propagating the virus life cycle in vitro. Alternatively, Murine Norovirus (MNV) represents a useful cell culture model for studying infection and replication of noroviruses. It"s known that during MNV replication viral proteins can be post-translationally modified (PTM), with regulated proteolysis being one essential PTM for viral replication. MNV encodes a chymotrypsin-like protease (NS6 or 3CL protease) that mediates the proteolytic processing of the ORF1 polyprotein encoded in the viral genome, thus releasing functional non-structural proteins required for viral replication. Although polyprotein processing is essential and conserved in many other positive-sense single-stranded RNA (+ssRNA) viruses, mounting evidence suggests that other PTMs such as ubiquitination and phosphorylation a regulated during MNV replication playing an important role in the outcome of the infection. Therefore, this work aims to characterise quantitatively proteolytic processing of viral and cellular proteins via N-terminomics and investigate the interplay of proteolysis with protein ubiquitination and phosphorylation. For this, we performed a time course infection of murine BV2 cells with Murine Norovirus, collecting cells from 0-12 hours post-infection (psi) following LC-MS analysis. Our newly developed TMT-HUNTER methodology using multiple proteases was applied for the characterisation of neo-N-termini generated including ubiquitination and phosphorylation, was performed using both open search workflow of Fragpipe v22 software. Over 3,700 neo-N- termini were detected across all infection time points revealing a temporal regulation of proteolysis during MNV infection. Motif analysis showed that a fraction of neo-n-termini identified in murine proteins matched the conserved 3CL proteases cleavage motif, suggesting that cellular proteins are also substrates for viral proteases. On the other hand, neo-n-termini matching caspase cleavage motif has been detected in viral non-structural proteins, indicating a convoluted network in which proteases and their substrates can have either viral or cellular origin. Ubiquitinome and phosphoproteome profiling are being carried out to assess the interplay of these PTMs and proteolysis. Our N-terminomics results provided an expanded repertoire of cellular substrates potentially targeted by MNV protease. The role of the proteolysis substrates and other PTMs in subverting host cells and causing disease will be further investigated. infection. PTM profiling, in response to MNV 51

OP-21 Population-based metaproteomics reveals functional associations between gut microbiota and phenotypes Shuang Liang1, Yingying Sun1, Zelei Miao1, Bang-yan Li2, Ziyuan Xing1, Yuting Xie1, Enci Cai1, Sainan Li1, Pu Liu3, Min Yang1, Menglei Shuai1, Wanglong Gou1, Wenhao Jiang1, Youming Wang4, Huanhuan Gao1, Ke Zhang1, Jing Yu3, Xue Cai1, Xingbing Wang4, Yu-ming Chen2, Ju-Sheng Zheng1, Tiannan Guo1 1Westlake University, Life Sciences, Hangzhou, China 2Sun Yat-sen University, Epidemiology, Guangzhou, China 3Westlake Omics, Hangzhou, China 4University of Science and Technology of China, Hefei, China The association of the gut microbial community with various physiological disorders underscores the significance of understanding its biological functions, which have yet to be fully explored. In this study, we characterized 86,876 microbial proteins and 1,277 human proteins in 1,967 Chinese stool specimens of GNHS cohort. Our findings demonstrate that, compared to metagenomic data, metaproteomic data can effectively capture the microbial biodiversity of an entire population using about 500 individuals. Additionally, it reveals greater inter- individual variability in microbial functions and serves as a more reliable indicator for assessing the temporal stability of the microbiota. We delineated the highly expressed functions of microbes among population and identified the core microbes across individuals that are likely to contribute to SCFA and vitamin B production. Moreover, our analysis discovered 10,714 associations between metaproteomic taxon, microbial function, or human proteins with 39 phenotypes, uncovering 1,381 microbial taxa, functions, and host proteins that related to Type 2 Diabetes (T2D). Among them, five features exhibited consistent and significant associations with T2D in both the GNHS cohort and an independent cohort. Finally, we demonstrated that Megasphaera elsdenii could mitigate elevated blood glucose levels in the host by metabolizing lactate to produce butyric acid, suggesting a protective mechanism against T2D. Our study underscores the power of population-based metaproteomics to offer unique functional insights into the gut microbiota. Fig. 1 (optional) 52

OP-22 The surfaceome architecture of a human B cell Julia Boshart1,2,3, Meret Zimmermann1,2,3, Arend Keller1,2,3, Sandra Götze1,2,3,4, Bernd Wollscheid1,2,3 1ETH Zürich, Department of Health Sciences and Technology, Zurich, Switzerland 2Institute of Translational Medicine (ITM), Zurich, Switzerland 3Swiss Institute of Bioinformatics (SIB), Lausanne, Switzerland 4ETH PHRT Swiss Multi-Omics Center (SMOC), Zurich, Switzerland Cellular function is orchestrated by extracellular signaling cues to elicit intracellular responses to stimuli. Cell surface residing proteins, collectively termed the surfaceome, fulfill their function as signal transducers within protein communities. This cell surface receptor interaction network is complex and dynamic, forming a highly organized surfaceome architecture essential for cellular functionality. Exploring the lateral proximity network of cell surface residing proteins facilitates the understanding of the surfaceome architecture, hence allowing for manipulation by utilizing bivalent therapeutics, such as bispecific antibodies. Targeting cells based on specific combinations of proteins present on their surface holds great potential in the search for selecting disease-driving sub-populations. We set out to identify novel target combinations for such therapies by elucidating the surfaceome architecture of B cells. To probe the spatial surfaceome, we utilized the in-house developed technology LUX-MS, enabling light-mediated proximity detection of dynamic protein interactions on the cell surface. Small molecule Singlet Oxygen Generators (SOGs) facilitate light-mediated photo-oxidation of cell surface proteins in the nanometer vicinity. When delivered through a ligand-of-interest, SOGs enable the capture of receptor proximity networks. LUX-MS of the mature B cell marker IgM revealed proximity candidates, of which 15 were initially chosen for individual LUX-MS experiments to confirm their proximity to IgM in a reciprocal approach. Some of these candidates exhibit similar proximity networks to IgM, such as the co- receptor CD79a, but others display very distinct organizations, as in the case of the costimulatory molecule and CD27 ligand CD70. The extensive proximity network of CD70 suggests a bystander functionality due to its high expression levels but can also hint at distinct organizational hubs on a single cell level. Intersecting and distinct proximity networks of individual target receptors facilitate the establishment of a global view of the B cell surfaceome architecture through reciprocal network walking. Having established initial architectural insight into the B cell surfaceome landscape, we are currently expanding our datasets with a large set of CD-annotated receptor targets in an effort to generate a comprehensive view of the B cell surfaceome architecture. Establishing a model for the B cell surfaceome architecture offers an opportunity to expand the toolbox for improved targeted immunotherapy by revealing drug candidate pairs. 53

OP-23 Mapping of the human solute carrier transporter protein interaction landscape Fabian Frommelt1, Rene Ladurner1, Eva Lineiro-Retes1, Peter Sykacek2, Ulrich Goldmann1, Sabrina Lindinger1, Iciar Serrano1, Eirini Christodoulaki1, Zuzana Gelová1, Ann-Katrin Hopp1, Alvaro Ingles-Prieto1, Anastasiia Pantielieieva1, Tamara A. M. Mocking3, Jasper F. Ooms3, Vojtech Dvorak1, Svenja Onstein1, Tabea Wiedmer1, André C Müller1, Thomas J. Hannich1, Adriaan P. IJzerman4, Laura H. Heitman3,4, Giulio Superti- Furga1,5 1CeMM, Vienna, Austria 2University of Natural Resources and Life Sciences, Department of Biotechnology, Vienna, Austria 3Leiden University, Division of Drug Discovery and Safety, Leiden Academic Centre for Drug Research, Leiden, Netherlands 4Oncode Institute, Leiden, Netherlands 5Medical University of Vienna, Center for Physiology and Pharmacology, Vienna, Austria Solute carrier (SLC) transporters form a superfamily of membrane proteins facilitating transmembrane transport of a wide range of substrates. SLCs play a vital role in cellular homeostasis and metabolic regulation, as they control nutrient, ion and drug transport. They consist of about 450 members which locate in cellular membranes including subcellular compartments. To achieve fine-tuned transport functions necessary for proper metabolic homeostasis, transporters act in concert with numerous other membrane, cytoskeletal and proteostatic proteins. More than half lack a well described interactome. We applied an AP-MS protocol tailored to transmembrane proteins to map the protein interaction network for more than 400 solute carriers. We further devised a ML- based scoring approach integrating scores of commonly used scoring tools (e.g. SAINT, CompPASS) and multiple quantitative and uniqueness features. The generated interactome covers more than 18"000 novel high- confidence protein interactions, therefore providing a basis for understanding the molecular regulation of this protein family. Clustering and functional enrichment analysis revealed interaction partners throughout the life span of the transporters. The extracted subnetworks allowed us to link interactors to proteostatic regulation pathways, including protein translation and posttranslational modification processes. In addition, we identified close to stoichiometric interactions, revealing known and novel stable assemblies of SLCs with chaperone-proteins. We performed a large-scale validation approach focusing on protein stability and localization changes upon genetic perturbation of prey proteins by RNAi. For 135 bait-prey combinations either approach was tested, with 40% showing a functional effect on the target SLC. Our dataset empowers the identification of core regulatory modules in which SLCs are involved and further contributes to the understanding of the biological roles of this understudied family of proteins. 54

OP-24 Dynamic proximal interaction networks of stress granules during assembly and disassembly reveal RNA regulation important for stress granule formation. Ji-Young Youn1,2,3, Yi-Cheng Tsai1,2, Sean Millar1,2,4, Hiroyuki Uechi5,6, Eileigh Kadijk1, Alon Minkovich1,2, Karl Schreiber1, Sarah Zhang7, Jie Qi Huang2, Hannes Rost2,8, Anthony Hyman5 1SickKids Research Institute, Molecular Medicine, Toronto, Canada 2University of Toronto, Molecular Genetics, Toronto, Canada 3SickKids Research Institute, Cell Biology, Toronto, Canada 4Thermo Fisher Scientific, Toronto, Canada 5Max Planck Institute, Cell Biology and Genetics, Dresden, Germany 6Tohoku University, Ecological Developmental Adaptability Life Sciences, Sendai, Japan 7University of Toronto, Physiology, Toronto, Canada 8Terrence Donnelly Centre for Cellular Biomolecular Research, Toronto, Canada Stress granules (SGs) are biomolecular condensates that transiently assemble during cellular stress and assist in the cellular stress response. SG formation is thought to be driven by multivalent interactions between proteins and transcripts released from translation machineries. Efforts to elucidate the mechanisms underlying SG formation has revealed the importance of protein-protein, protein-RNA and RNA-RNA interactions, which involve protein and RNA scaffolds. However, we lack in understanding additional mechanisms required for SG formation. Here, we present quantitative analyses of multi-bait SG proximity labeling (i.e., BioID) experiments performed during stress time course paralleled with a genome-wide chemical genetic interaction screen. First, we implemented pulse biotin labeling of SG proteome during stress and recovery, augmented by data- independent acquisition (DIA) method. Experimentally, we used a group of bait proteins that effectively label over 50% of the known SG proteome and performed BioID labeling during two different stress conditions (oxidative and hyperosmotic) that induce SG via distinct mechanisms. Through SAINT analysis, we identify ~1,300 high-confident proximal interactors (preys) in steady state, stress, and recovery periods, and predict that about a third of these preys represent SG proteome in either condition. To measure quantitative changes in the SG protein interaction network during SG assembly and disassembly, we utilized DIA-NN analyses to obtain protein abundance and exploited ANOVA tests to determine statistically significant changes. About 20% of preys show significant quantitative changes during assembly and disassembly of SGs, representing a pool of dynamic associations and dissociations, while ~80% remain the same. Regardless of the type of stressor, we observe dynamic proximal interaction changes between SG baits and the components of the CCR4-NOT deadenylase, global regulator of polyA tail lengths. In parallel to the proteomics screening, we utilized a small molecule that can inhibit SG formation, called lipoamide, to interrogate mechanisms underlying SG formation. We identified that CCR4-NOT structural components, CNOT1, CNOT2 and CNOT3, are required for lipoamide inhibitory action on SGs. Based on the function of CCR4-NOT complex, we propose that lengthening of poly(A) tails by loss of CCR4-NOT complex lowers the threshold for phase separation through increasing valency and overcomes the inhibitory effect of lipoamide. Together, we provide dynamic map of SG proximal interaction networks during SG formation and dissolution and propose that regulation of polyA lengths of mRNAs by CCR4-NOT complex is important for SG formation. Fig. 1 (optional) 55

OP-25 A plasma protein atlas of the UK biobank Maik Pietzner1, Carl Beuchel1, Summaira Yasmeen1, Burulça Uluvar1, Martijn Zoodsma1, Mine Köprülü2, Julia Carrasco-Zanini3, Claudia Langenberg1,2,3 1Berlin Institute of Health at Charité – Universitätsmedizin Berlin, Computational Medicine, Berlin, Germany 2University of Cambridge, MRC Epidemiology Unit, Cambridge, United Kingdom 3Queen Mary University of London, Precision Healthcare University Research Institute, London, United Kingdom Affinity-based proteomic techniques now measure thousands of proteins circulating in blood at population scale, but only ~10% of which have established roles in blood leaving the sources and relevance of variation for most proteins poorly understood. Here, we used machine learning to systematically identify and quantify major predictors of plasma levels of ~3,000 protein targets among 43,240 participants of the UK Biobank. We identify 427 diverse, modifiable and non-modifiable, factors that cumulatively explained on average 23.7% (range: 0.001%-79.7%) of variation in plasma levels of 2,846 protein targets (97.4% of all protein targets), selected from over 1,800 participant and technical characteristics. We demonstrate that protein targets segregate into 11 clusters, of which all but one, are predominantly explained by a single predictor of plasma protein variances, including platelet count (n=635), germline genetic variation (n=372), renal (n=199) and liver function (n=22), low-grade inflammation (n=208), ancestry (n=150), and sex (n=10). With few exceptions (2.7%), we obtain largely similar explained variances in plasma protein levels across the sexes and ancestries (British Central South Asians and Africans), but with evidence that selected participant characteristics differed by sex or ancestry. Our results highlight new tissue and cell-type disease links, as well as potentially (adverse) effects of medications. For example, intake of warfarin or clopidogrel explains variance in plasma levels of proteins specifically expressed in fibroblasts. In summary, we provide a data-driven approach to better understand the factors that shape the human plasma proteome and help guide blood-based protein biomarker discovery. 56

OP-26 A comprehensive human pan-disease blood atlas based on next-generation protein profiling María Bueno Álvez1, Josefin Kenrick1, Emil Johansson1, Thanadol Sutantiwanichkul1, Martin Zwahlen1, Kalle von Feilitzen1, Fredrik Edfors1, Mathias Uhlen1 1KTH Royal Institute of Technology, Protein science, Stockholm, Sweden In this study, we present the creation of an open-access resource designed to explore the protein profiles in blood across a wide range of diseases. Using a highly sensitive and specific multiplex assay, we conducted parallel analysis of 1,463 proteins in blood from patients with 57 diseases. This approach allowed us to identify several proteins correlating with age and BMI, as well as sex-related proteins showing significant differences in males and females. Notably, we identified proteins significantly altered in various cancers, including acute myeloid leukemia, glioma, and ovarian cancer, whereas fewer proteins were observed for some of the common cancers, such as breast and prostate cancer. Our findings also revealed that proteins elevated in diseases like pancreatic cancer and autoimmune disorders were similarly elevated in patients with infectious diseases, indicating these proteins are markers for inflammation rather than disease-specific markers. Furthermore, proteins elevated in hepatocellular cancer were also elevated in other liver diseases, showcasing the power of our pan-disease approach in identifying proteins with cross-disease patterns. For cardiovascular disease, several proteins were identified, including proteins elevated in stenosis of the carotid artery. In pediatric patients, we observed protein elevation patterns similar to those in corresponding adult diseases, along with unique proteins elevated across all pediatric samples. The analyzed protein levels for each disease are visualized and compared against (i) a cohort of healthy individuals, (ii) all other diseases, and (iii) all other diseases within the same class, providing comprehensive differential expression results. This resource offers valuable insights for biomarker discovery and enhances our understanding of protein behavior across various diseases. Fig. 1 (optional) 57

OP-27 Mass spectrometry-based proteomics at extreme scale reveals biomarkers for adverse pregnancy outcomes Vincent Albrecht1, Vincenth Brennsteiner1, Medini Steger1, Aneeta Hotwani2, Javairia Khalid3,4, Imran Nisar3,4, Margaret Kasaro5,6, Humphrey Mwabe7, Dominique Davis8, Joni Price5, Waqassuddin Khan3,4, Fyezah Jehan3,4, Kristina De Paris8, Jeffrey SAS Stringer5, Johannes Müller-Reif1, Matthias Mann1 1Max Planck Institute of Biochemistry, Department of Proteomics and Signal Transduction, Martinsried, Germany 2Aga Khan University, Medical College, Infectious Diseases Research Lab, Karachi, Pakistan 3Aga Khan University, Medical College, Department of Pediatrics and Child Health, Karachi, Pakistan 4Aga Khan University, Medical College, Biorepository and Omics Research Group, Karachi, Pakistan 5University of North Carolina, School of Medicine, Division of Global Women's Health, Chapel Hill, NC, United States 6University of Zambia, Deptartment of Obstetrics and Gynecology, Lusaka, Zambia 7University of North Carolina, Global Project Zambia, Lusaka, Zambia 8University of North Carolina, Department of Microbiology and Immunology, Chapel Hill, NC, United States Objective: Proteomic technology has progressed tremendously in the past years fueling a revival of mass spectrometry (MS)-based plasma proteomics (Bader, PMID 37209816). We have recently developed a highly robust, deep and economic workflow that should enable the analysis of very large cohorts. We are currently applying this to study adverse pregnancy outcomes (APOs) in low- and middle-income countries and identify biomarkers for improved prediction aiding clinical decision-making. Methods: The Multi-Omics for Mothers and Infants (MOMI) consortium funded by the Bill & Melinda Gates Foundation has collected more than 40,000 longitudinal plasma samples from women in Africa and South Asia, taken at up to three different time points during pregnancy. Using a semi-automated liquid handling robot, we subjected these samples to our plasma profiling workflow, which features an adaption of the recently reported perchloric-acid enrichment technique (Viode, PMID 36989362). Here, high abundant proteins partition more to the precipitate effectively enriching other plasma proteins in the supernatant. Subsequently we heat-denature and digest them using LysC/trypsin. Peptides are measured on an Orbitrap Astral mass spectrometer interfaced with the Evosep One system using IonOpticks columns. Here, we report on results of three sub-cohorts accounting in total for more than 10,000 samples provided by the Pakistani AMANHI (Alliance-for-Maternal-and-Newborn-Health-Improvement) and the Zambian IPOP (Improving Pregnancy Outcomes with Progesterone) and ZAPPS (Zambian Preterm Birth Prevention Study) cohorts of MOMI. Results and Conclusion: The AMANHI cohort with 5,500 plasma samples was measured with a throughput of 60 samples-per-day, yielding on average 2,000 protein groups. Our preliminary data suggest that this in-depth dataset reveals statistically significant changes in more than 1,500 proteins between the first and second trimesters. We identified both known and novel pregnancy-related markers, while corroborating the presence of known markers for APOs such as PAPP-A. Potential novel markers of APOs are currently being validated across the different MOMI subcohort in a "rectangular study approach" (Geyer, PMID 28951502). Initial results from the two Zambian cohorts with more than 5,000 plasma samples are being measured with a throughput of 100 samples-per-day yielding on average 2,400 protein groups. Results appear to support findings from the AMANHI cohort, promising novel biomarkers that would allow for improved early detection of APOs in routine clinical diagnostics. 58

OP-28 Single-cell molecular pixelation of CAR T-cell surface proteome reveals spatial patterns characteristic of chronic stimulation Anthony Cesnik1, Theodore Roth2, Oliver Takacsi-Nagy2, Trang Le1, Ansuman Satpathy2, Emma Lundberg1,2 1Stanford University, Bioengineering, Stanford, CA, United States 2Stanford University, Pathology, Stanford, CA, United States Customized immunotherapies using CAR T-cells are already making their way into the clinic and are showing much promise for the treatment of cancers including leukemia. However, stimulating CAR T-cells against the target antigen, such as that presented on a tumor, has the challenge that chronic stimulation leads to deterioration of the treatment potential of the CAR T-cells. Identifying the effects of chronic stimulation is currently challenging, making it difficult to select CAR T-cells that are effective for treatment. We applied a recently developed antibody-based spatial profiling technique for characterizing the surface proteome, named molecular pixelation, to determine characteristic surface-based signatures of chronic CAR T-cell stimulation. This technique uses DNA-conjugated antibodies against 80 markers, 4 controls and 76 target proteins, and two rounds of gap-fill ligation and rolling circle amplification to determine the relative position of the protein molecules on the surface of a cell. We analyzed 8504 CAR T-cells at a single-cell level, collected from 3 donors and stimulated once for acutely stimulated cells and stimulated repeatedly over 14 days for chronically stimulated cells, followed by selection for live cells and a surface-based marker NGFR indicating the expression of the CAR construct. The sequencing counts that correspond to individual antibodies allow us to distinguish acutely and chronically stimulated cells, and the molecular pixelation protocol allows us to identify polarization and colocalization of the markers on the cell surface. A total of 33 of the 76 proteins (43.4%) exhibited significant changes in polarization, and there was an overall increase in polarization after chronic stimulation. Significant colocalization changes were observed for 561 proteins (19.7% of all 2850 combinations); one particular colocalization event that stood out was CD37/CD82, which increased significantly after chronic stimulation (adjusted p-value < 1e-250). Molecular pixelation is a new spatial proteomics technique for the discovery of proteomic polarization and colocalization on the cell surface that can identify new types of biomarker candidates, such as were identified for chronically stimulated T-cells in this study. 59

OP-29 Sensitivity and dynamic range of massively parallel single-molecule protein quantitation using protein identification by short-epitope mapping (PrISM) Sheri Wilcox1, Taryn Gillies1, Rob Grothe1, Vivek Budamagunta1, Katie Winters1, Maureen Newman1, Maria Villancio-Wolter1, Pengyu Hao1, Sonal Tonapi1, Kathy Lazaruk1, Kara Juneau1, Jarrett Egertson1, Andreas Huhmer1, Parag Mallick1 1Nautilus Biotechnology, Probe Development and Bioinformatics, San Carlos, CA, United States Many biological processes in the cell are driven by protein abundance changes across the full complement of the cellular proteome. Detection of those changes requires a proteomic measurement capable of detecting minute changes in protein abundance simultaneously with high-abundance protein changes. Protein Identification by Short-epitope Mapping (PrISM) is a novel single-molecule based proteome analysis approach that may enable identification and quantification of nearly the entire human proteome in a single experiment. Here we assess the dynamic range associated with the PrISM approach utilizing a simple mixture of standard proteins. Two proteins from a twenty-one protein standard mixture were deposited in separate control lanes on the flow cell and in a sample lane as part of the twenty-one protein mixture. In a second set of lanes, the two proteins were varied over multiple orders of magnitude. PrISM data were acquired on the flow cell using 150 sequential incubate/rinse/image/remove cycles with 50 multi-affinity probes. Frequency of binding of the probes to the standard proteins in the control lanes compounded over multiple cycles of measurement provided the basis for confident identification and quantification of the two proteins in the sample lane. The results indicate that the two standard proteins were reproducibly quantified over multiple orders of magnitude in the presence of the mixture with one standard varying as much as one million fold vs the other standard. We also demonstrate that the protein decoding signal has a wider dynamic range than quantitation from any single multi-affinity probe due to the ability of the decode algorithm to increase the fidelity of quantitation by accumulating information across multiple multi-affinity probes measurement events. Furthermore, we demonstrate assay sensitivity into the yoctomole range. These attributes showcase the platform's robust capability for precise and sensitive quantification of protein variations in complex biological samples. 60

OP-30 Resolving spectral congestion of (any?) intact glycoprotein using DIA-PTCR Luis Schachner1, Christopher Mullen2, Wilson Phung1, Joshua D Hinkle2, Michelle Irwin Beardsley1, Tracy Bentley1, Peter Day1, Christina Tsai1, Siddharth Sukumaran1, Tomasz Baginski1, Danielle DiCara1, Nicholas J. Agard1, Matthieu Masureel1, Joshua Gober1, Adel M. ElSohly1, Rafael Melani2, John E. P. Syka2, Romain Huguet2, Michael T. Marty3, Wendy Sandoval1 1Genentech, Inc., South San Francisco, CA, United States 2Thermo Fisher Scientific, San Jose, CA, United States 3University of Arizona, Tucson, AZ, United States The molecular complexity of proteins with multiple glycosylation sites challenges traditional analytical methods for intact mass analysis. Glycans play a pivotal role in defining the functional and structural properties of proteins; however, their biosynthesis is not template-driven and are thereby (rather unpredictably) structurally diverse. This inherent molecular diversity impedes thorough characterization using only mass spectrometry (MS) or other conventional techniques, especially for the purposes of assigning function to glycoforms subpopulations or individual proteoforms. Current approaches to analyzing heterogeneous glycoprotein samples, such as those involving biotherapeutics or in targeted structural/glycobiological studies, involve fragment measurements or partial digestion of glycans. In our recent Nature Communications publication (4/2024), we present a streamlined workflow for the direct, intact analysis of heterogeneous proteins that have previously posed challenges for intact MS characterization. This innovative method, termed "DIA-PTCR," employs a data-independent approach to analyze intact glycoproteins and non-covalent complexes using proton-transfer charge-reduction tandem MS (PTCR). Owing to the high-mass range quadrupole of the new Ascend Orbitrap platform, which can isolate ions up to 8000 m/z with an exceptionally narrow m/z window, DIA-PTCR provides mass information on spectrally unresolvable proteins without the need for denaturation, upfront digestion, or separation. We tested the approach on various heavily glycosylated molecules, including cytokines, TNF-fusions, and peptide-bound MHC class II complexes to demonstrate efficacy in measuring their proteoform-level diversity while also retaining information about non-covalent structure, co-occurrence of glycans and other PTMs such as phosphorylation and acetylation. We observed subunit truncations, differential glycosylation between samples, and were able confirm subunit assembly in the midst of extreme proteoform diversity. Additionally, we establish a bioinformatic data analysis strategy for DIA-PTCR that may offer insights into glycan composition. Due to the staggering combinatorial diversity of glycosylation, probable annotations for detected intact masses –reflecting the limited set of glycan combinations– were modeled using validated glycan candidates from glycopeptide and glycomic data, providing relative abundances of putative glycoforms for each DIA-PTCR molecular weight. Notably, we inferred the glycoform abundances for hundreds of molecular weights for the eight-times glycosylated cytokine IL22-Fc dimer, enabling correlations between sialylated glycoform sub- populations and their interaction with the IL22 receptor. Further development of the DIA-PTCR workflow may incorporate top-down fragmentation for glycoform characterization as well analysis of immuno-enriched glycoproteins to elucidate the network of readers, writers and erasers of glycosylation. 61

OP-31 Proving the proteoform hypothesis via spatial top-down proteomics of the human cell Jake Kline1, Jingjing Huang2, Christopher Mullen2, Joseph Greer3, David Bergen2, Michael Belford2, Cornelia Boeser2, Kenneth Durbin3, Vlad Zabrouskov2, Rafael Melani2, Luca Fornelli1 1University of Oklahoma, School of Biological Sciences, Norman, OK, United States 2Thermo Scientific, San Jose, CA, United States 3Proteinaceous, Inc, Evanston, IL, United States Proteins exist in their biological context as molecularly defined "protein forms", or proteoforms, carrying specific sets of genetic and chemical modifications. The "proteoform hypothesis" posits that the proteome-to-phenotype correlation is better explained through the analysis of endogenous proteoforms via top-down proteomics rather than by the analysis of protein surrogates (i.e., proteolytic peptides) investigated through bottom-up proteomic approaches. Here we present the results of the study of the intact 0-70 kDa proteome of HEK293 cells performed at the subcellular level. Subcellular fractions were obtained using the dynamic organellar mapping (DOM) protocol. Intact proteoforms extracted from each of the seven subcellular fractions (nucleus, mitochondria, ER, endosomes, Golgi apparatus, larger protein complexes, cytosol) were analyzed using a specific approach, depending on their molecular weight (MW). Proteoforms <30 kDa were fractionated using an in-house cast, high-resolution 15% T GELFrEE, with each MW fraction (~5 kDa mass bins) liquid chromatography-field asymmetric ion mobility spectrometry-tandem mass spectrometry (LC-FAIMS-MS2) on an Orbitrap Eclipse Tribrid (Thermo Scientific). Larger proteoforms 30-70 kDa were fractionated by 10% T GELFrEE and each fraction was analyzed using the targeted proton transfer charge reduction data acquisition method (tPTCR) on an Orbitrap Ascend Tribrid (Thermo Scientific). In tPTCR experiments, 4 partially overlapping m/z windows (1.5 Th width) are surveyed per LC-MS2 run. Cations from each narrow isolation window are first subjected to PTCR, to obtain a new charge state envelop from which the average mass of the proteoform can be derived; and then, they are fragmented using higher-energy collisional dissociation for amino acid sequencing. Five LC-MS2 tPTCR measurements were run for each fraction, sampling a total of ~24 m/z units, to obtain information on "all" proteoforms included in a given GELFrEE fraction. interrogated via Preliminary results from four of the seven subcellular fractions show three to four-fold increase in the number of identified UniProt accessions (>3,000) and proteoforms (>28,000) compared to a similar top-down analysis performed on HEK293 whole cell lysate. Additionally, while each characterized DOM fraction contains a large portion of unique UniProt accessions, the variability at the proteoform level is substantially higher than at the protein family level. For instance, 36 kDa Y-box-binding protein 1 (P67809), known to be present both in the nucleus and in the cytoplasm, was identified as full-length proteoforms in the nuclear fraction and as C- terminally truncated proteoforms (20-29 kDa range) in the cytosolic fraction. Taken together, these results demonstrate the validity of the proteoform hypothesis and represent the most comprehensive catalog of proteoform for a single cell line ever produced. 63

OP-32 Quantitative assessment of Tau proteoform changes in iPSC-derived neurons during differentiation Sanjib Guha1, James Joly1, Vivek Budamagunta1, Rosemary Wang1, Abdul-Rizaq Hamoud1, Kevin Chen1, Gajalakshmi Dakshinamoorthy1, Steven Tan1, Maryam Jouzi1, Andreas Huhmer1, Greg Kapp1, Kara Juneau1, Parag Mallick1 1Nautilus Biotech, San Carlos, CA, United States Alzheimer's disease (AD) and related neurodegenerative diseases, such as primary Tauopathies are a significant burden in healthcare due to growing prevalence. One central element in these diseases, particularly in tauopathies, is the microtubule-associated protein, Tau. There are six major isoforms generated by alternative mRNA splicing of exons 2 and 3 (either 0N, 1N, or 2N isoforms) and exon 10 (3R or 4R isoforms) resulting in the mature human brain. Specific post-translational modifications of tau have been associated with Alzheimer’s disease and have been suggested as clinical biomarkers and/or therapeutic targets. However, it remains unclear which tau proteoform(s) contribute(s) most to neural dysfunction and neurodegeneration, and how. Current methods do not adequately address the molecular diversity of Tau proteoforms in neuronal tissue. We have utilized a single-molecule detection technology, which allows for highly sensitive and precise identification of Tau proteoforms and their ratios in iPSC-derived neurons. Here we apply this single-molecule approach with single-molecule sensitivity and a wide dynamic range to the discovery of Tau proteoforms from iPSC- derived neurons. We utilize the single-molecule detection capability to detect and assess the specific proteoforms and their relative ratios present in the tissue. We demonstrate the detection of the proteoforms using a multi-cycle method adopting commercially available protein-isoform and PTM-specific antibodies and validate that the single-molecule platform exhibits a linear analytic response. First, to determine single-molecule binding, single protein nanoparticle-conjugates are deposited on the flow cell. In a second step, an antibody specific for a Tau epitope of interest is added to analyze all single proteins on the flow cell massively in parallel. In the third step, bound antibody is imaged by fluorescence, localizing each antibody binding event to an individual protein at a specific location. The cycle of binding, imaging, and removal is repeated multiple times to probe for Tau epitopes of interest. Applying a panel of 8 epitope probes (256 potential proteoforms) covering several Tau isoforms and phosphoproteoforms we quantify the relative amounts of Tau proteoform in differentiated iPSC cells confirming the relative amounts of Tau isoforms indicated by Western blot. In addition, the single molecule nature of the platform also allows us to determine relative quantities of phosphorylated proteoforms present in those cells. These results clearly demonstrate the unique capability of the platform to quantify biologically relevant Tau proteoforms accelerating our understanding of the role of Tau proteoforms in Tau aggregation, a hallmark of Alzheimer's disease. 64

OP-33 System-wide functional glycoproteomics approaches Clement Potel1, Mira Lea Burtscher1, Martin Garrido-Rodriguez1, Isabelle Becher1, Amber Brauer-Nikonow1, Athanasios Typas1, Michael Zimmermann1, Mikhail Savitski1 1EMBL Heidelberg, Heidelberg, Germany Introduction: Protein glycosylation is a highly diverse post-translational modification, playing key roles in the regulation of cellular processes such as cell signaling, adhesion and cell-cell interactions. A major obstacle in our understanding of the role of glycosylation in health and disease, along with the molecular mechanisms underlying the modulation of glycoprotein functions, has been the lack of systematic methods for assessing the impact of glycosylation alteration on protein function. Methods: We recently developed a glycoproteomics workflow enabling in-depth quantification of glycosylation. Building on this advancement, we now aim to study glycoforms within a functional context. First, we have devised methods to differentiate surface-exposed, mature glycoforms from intermediate glycosylated proteoforms by treating intact HEK293 cells with either proteinase K or PNGase F. To improve our understanding of how glycosylation alteration affects protein states, we developed techniques to systematically measure the effect of glycosylation alteration on biophysical properties of proteins in vivo, such as thermal stability or solubility. Such approaches have recently been used to assess phosphorylation sites functionality, and rely on the fact that changes in protein states (interaction, localization,...) can be associated with changes in protein biophysical properties. Results: We could successfully differentiate mature, surface-exposed glycoforms from intermediate intracellular species, addressing a longstanding question in the field. We demonstrate that high-mannose glycoforms are in general less surface exposed, although this tendency is site-dependent and related to structural features. Conversely, most of the fucosylated/sialylated glycoforms are, at least partially, surface-exposed. We observed good correlation between the enzymatic treatments, both at the glycopeptide level and glycan compositions level, and could identify compositions predominantly surface-exposed. Next, we measured the solubility and thermal stability of thousands of glycoforms in the mouse brain. Our results illustrate significant variability in biophysical behaviors among different glycoproteoforms, allowing us to establish connections between glycan compositions or structural features of glycoproteins and their thermal stability and solubility. Conclusions: Here we present a method that offers a system-wide, site-specific view of mature and immature glycans in a human cell line. We believe this approach has the potential to enhance understanding of the biological functions of glycosylation and contribute to the development of therapies by informing on which glycosylation events to target. Our biophysical glycoproteomics approaches constitute an initial step towards systematically understanding the impact of glycoform modulation on protein function, providing a framework for future research in this area. 65

OP-34 From discovery to clinic: deep visual proteomics for advanced precision medicine Lisa Schweizer1, Ishani Ummat1, Lukas Oldenburg1, Stefanie Strasser1, Maximilian T. Strauss1, Andreas Mund1 1OmicVision Biosciences ApS, Copenhagen, Denmark The individual architecture of tissue holds information that is invaluable for clinical diagnosis and treatment. Molecular approaches such as genomics consult the pathological assessment of patient specimens, but often fall short owing to nonlinear genomic-phenotypic relationships. Immuno-based imaging of tissue sections provides spatial resolution of a phenotype by targeting specific proteins but lacks information depth and requires a-priori knowledge on the analysis outcome. Correlating the visual and proteomic phenotype of single cells while maintaining their localization in the native tissue landscape is hence key to improving precision medicine. We aim to explore disease-causing proteins and to identify druggable targets in individual patients. To this end, we employed a robust and scalable adaptation of the Deep Visual Proteomics (DVP) technology. DVP integrates high-content imaging, AI-based image analysis, single-cell isolation via high-precision laser microdissection and ultra-sensitive mass spectrometry-based proteomics. By employing cutting-edge artificial intelligence models, our platform enables accurate segmentation and classification of diverse cell types within tissue samples. This advanced phenotyping approach facilitates the automated identification of morphologically distinct regions of interest within each tissue, laying the foundation for targeted analysis and discovery of novel therapeutic targets. Integrating laser microdissection (LMD7, Leica Microsystems) for precise single-cell isolation from minute tissue samples with advanced mass spectrometry (Astral, ThermoFisher) and standardized liquid chromatography (Evosep), our platform achieves a throughput of 80 samples per day, yielding in-depth proteomes of over 6,000 proteins from single-cell isolated phenotypes across diverse diseases and tissues. The generation of proteomic maps based on the original tissue specimens resolves distinct cell types or cell states within AI-annotated regions that represent the phenotypic heterogeneity of a given disease. We demonstrated the versatility and effectiveness of our platform by applying our spatial approach to complex oncology cases across various indications. In summary, our approach enables robust and scalable spatial proteomics to identify cell-type specific, druggable protein targets across diverse indications. This resolves the cellular heterogeneity of a disease in synergy with clinical pathology and presents clinicians with disease-relevant therapeutic options at the individual level. The aim of precision medicine is to match the right drug to the right patient. Using our spatial proteomic approach, we translate recent technological advances from discovery to the clinic, identifying novel and druggable protein targets. 66

OP-35 Global analysis of protein turnover dynamics in single cells during stem cell differentiation Pierre Sabatier1,2, Zilu Ye2,3, Maico Lechner2, Ulises H. Guzman2, Christian Beusch4, Fabiana Izaguirre5, Anjali Seth5, Sergey Rodin1, Karl-Henrik Grinnemo1, Jesper Olsen2 1Uppsala University, Department of Surgical Sciences, Uppsala, Sweden 2University of Copenhagen, NNF-CPR, Copenhagen, Denmark 3Chinese Academy of Medical Sciences, Suzhou Institute of Systems Medicine, Suzhou, China 4Emory University School of Medicine, Department of Pathology and Laboratory Medicine, Atlanta, GA, United States 5Cellenion SASU, Lyon, France Even with recent improvements in sample preparation and instrumentation, single-cell proteomics (SCP) analyses mostly measure protein abundances, making the field unidimensional. In this study, we employ a pulse stable isotope labelling by amino acids in cell culture (pSILAC) approach to evaluate protein abundance and turnover in single cells (SC-pSILAC). We demonstrate that two SILAC labels are detectable from ~4000 proteins in single HeLa cells, using state-of-the-art SCP workflow, providing known biological information about protein turnover in HeLa cells and making pulsed and multiplex SILAC applicable to single-cell analysis. Then, we investigate drug-altering global and specific proteins turnover and highlight those treatments effects in single cells. Lastly, we perform a large-scale time-series SC-pSILAC analysis of human induced pluripotent stem cells undirected differentiation encompassing six sampling times over two months and > 1000 cells. Abundance measurements highlight cell-specific markers of stem cells and cells from various lineages and organs. Turnover dynamics highlights differentiation-specific turnover of proteins and co-regulation of core members of protein complexes over the differentiation time. Finally, we show that histone turnover can highlight dividing and non- dividing cells with potential in stem cell and cancer research. Our study represents the most comprehensive SCP analysis to date, offering new insights into cellular diversity and pioneering functional measurements beyond protein abundance, distinguishing SCP from other single-cell omics methods and enhancing its scientific relevance in biological research in a multidimensional manner. Fig. 1 (optional) 67

OP-36 In-depth and high-throughput single-cell proteomics workflow defines function subtypes Marvin Thielert1, Marc Oeller1, Enes Ugur1, Maximilian Zwiebel1, Sophia Steigerwald1, Georg Wallmann1, Tim Heymann1, Yaara Oren2, Thierry M. Nordmann1, Matthias Mann1 1Max Planck Institute of Biochemistry, Martinsried, Germany 2Tel Aviv University, Department of Human Molecular Genetics & Biochemistry, Tel Aviv, Israel The cell, the fundamental unit of life, is crucial for understanding the molecular mechanisms and functions of organisms in both healthy and diseased states. Mass spectrometry-based single‐cell proteomics aims to characterize biological functions and heterogeneity at the level of proteins in an unbiased manner. Despite its clear potential, single-cell proteomics faces formidable challenges in sensitivity and throughput. Here we develop a single-cell workflow for consistent, high-throughput and in-depth proteomic data acquisition. Our approach integrates an optimized oil-free workflow on a nanoliter dispensing device (cellenONE) with the Orbitrap Astral mass spectrometer (Thermo Scientific), effectively streamlining the proteomic acquisition process. By implementing the Whisper Zoom technology (Evosep), we increased the throughput to 80 or 120 samples per day (SPD). We also optimized a data-independent acquisition method with variable windows to enhance proteomic depth and quantitative consistency. This approach allowed us to achieved a proteomic depth of over 3,000 to 5,000 protein groups depending on cell size. The integration of multiplex-DIA (mDIA) can further multiply throughput by at least two-fold to 160 or 240 cells per day. Compared to single-cell RNA sequencing, single-cell proteomics exhibits lower variability and enables the definition of various cell states. We applied this refined workflow to a range of biological questions. For instance, in the study of melanoma single cells, we successfully characterized the proteomic distinctions between treatment resistant "persister cells" and a control population of cells. We achieved a proteomic depth of about 5,000 protein groups with high data completeness, even covering many important transcription factors. We identified elevated signatures like epithelial- mesenchymal transition, glycolysis metabolism and hypoxia in the persister cells, compared to cell cycle and DNA repair terms in the control melanoma cells. Our study demonstrates that single-cell proteomics can be performed with high throughput and significant proteomic depth, a depth that was state of the art for bulk proteomes not long ago. We also highlight the potential of single-cell proteomics in addressing complex biological questions and advancing our understanding of cellular heterogeneity. Fig. 1 (optional) 68

OP-37 End-to-end transfer learning in DIA proteomics for the characterization of unseen peptide modifications Georg Wallmann1, Marvin Thielert1, Tim Heymann1, Mohamed Kotb1, Xie-Xuan Zhou1, Wen-Feng Zeng1, Matthias Mann1 1Max Planck Institute of Biochemistry, Munich, Germany Data independent acquisition (DIA) based proteomics has established itself as key for fast and deep characterization of complex proteomes. Identification of peptides and proteins from DIA data is commonly performed using deep-learning generated spectral libraries which were trained with a large corpus of known spectral matches. DIA search is therefore challenged when the properties of unseen post translation modifications or instrument setups should be predicted. Based on our recently published modular open-source search framework alphaDIA, we have introduced a new strategy for DIA search termed end-to-end transfer learning. AlphaDIA is an open-source framework for complete DIA search workflows. It builds on the scientific python stack and the alphaX ecosystem allowing flexible search strategies as well as default workflows accessible through a Python API, Jupyter notebooks, a command line interface. AlphaDIA encompasses the entire workflow, from raw file processing to protein quantity reporting, and supports files and proprietary formats from major vendors. Engineered for streamlined processing of large cohorts and diverse data sizes, it operates seamlessly on Windows, Linux, and Mac platforms, or can be deployed in a distributed manner in the cloud using Slurm or Docker. By integrating deep-learning prediction using our deep-learning framework AlphaPeptDeep and with DIA search, custom models are trained for specific experimental setups. Following an initial search, confident identifications are re-quantified and their fragment spectra are collected. Our robust and automated deep-learning pipeline optimizes the neural network for retention time, ion mobility, charge state and fragment spectra prediction while ensuring generalization through a held out test set. In a second search alphaDIA uses this trained model to increase identifications and decrease false discoveries. We show the potential of this technology by applying it to chemical as well as post translational modifications. In our experiments, end-to-end transfer learning can increase the number of identified precursors drastically by close to 50% percent, leading to more than 25% additional protein identifications. Rigorous benchmarking established that the effects aren"t the result of overfitting but reflect generalization of the network to the new data distribution. To unlock the potential, we show the potential application to acetylated peptide discovery, HLA-peptide characterization and non-tryptic digests. Fig. 1 (optional) 69

OP-38 Causal inference enables the estimation of outcomes of interventions from observational mass spectrometry (MS)-based proteomics experiments Devon Kohler1,2, Karen Sachs3,4,5, Jeremy Zucker6, Benjamin M. Gyori1,2, Lindsay Pino7, Olga Vitek1,2 1Northeastern University, Khoury College of Computer Sciences, Boston, MA, United States 2Northeastern University, Barnett Institute for Chemical and Biological Analysis, Boston, MA, United States 3Answer ALS, Washington, DC, United States 4Next Generation Analytics, Palo Alto, CA, United States 5Modulo Bio, San Diego, CA, United States 6Pacific Northwest National Laboratory, Richland, WA, United States 7Talus Bioscience, Seattle, WA, United States Understanding the proteome"s response to perturbations is an important step towards a full understanding of its function. Traditionally this is done experimentally, by perturbing the system using external mechanisms such as drugs and observing the resulting effects on protein abundance. Alternatively, machine learning methods, such as neural networks can be used to estimate the effect of perturbations by considering joint coregulation of many proteins. These approaches can be successful in some circumstances, however, they do not incorporate prior knowledge about mechanisms between proteins. In contrast, causal inference methods make explicit use of prior knowledge and have been shown to perform well in the estimation of perturbations, for example, using transcriptomics data. These methods take as input abundances from an observational experiment (i.e., an experiment that does not implement a perturbation), and a graph of known causal relationships, and estimate the impact of perturbations on the graph. Here we describe a novel approach to translating causal inference methods to mass spectrometry (MS)-based proteomics, moving beyond the characterization of associations. The implementation of causal inference methods in the context of MS-based proteomics has a number of computational challenges that must be overcome in order for the methods to be effective. First, MS proteomics data is inherently noisy, is subject to batch effects and missing values, and quantifies peptide-fragments rather than proteins. Second, creating an accurate graph of causal relationships between proteins is non-trivial. We leverage the INDRA database of biological mechanisms which combines the content of pathway knowledgebases with literature mining to extract causal relations between the proteins quantified in the experiment. We employ a variety of filtering strategies to create a context-specific network with edges that support the data generating process. Finally, we develop a Bayesian model that leverages the network to model the relationships between proteins and is fit to the observational data, to perform model-based inference. We validate this method on synthetic data, then demonstrate it on real biological experiments. We highlight areas where machine learning methods such as neural networks fail to accurately predict the effect of interventions and show how causal inference can overcome this. We use computer simulations to investigate the circumstances in which the causal inference methods are likely to succeed, such as favorable network topology and number of biological replicates. Finally, we show the accuracy of the method using an experiment which investigates the effect of drug compounds on the chromatin-binding activity of transcription factors. In this setting, the causal model is trained on the observational DMSO control data and evaluated by comparing the model"s predictions of the effect of drug interventions against experimental measurements. 70

OP-39 Privacy-preserving multi-center differential protein abundance analysis with FedProt Yuliya Burankova1,2, Miriam Abele2, Mohammad Bakhtiari1, Fabian Gruhn3, Christine von Toerne3, Teresa Barth4, Lisa Schweizer5, Pieter Giesbertz6, Johannes Schmidt7, Stefan Kalkhof7,8, Janina Müller-Deile9, Yassene Mohammed10, Elke Hammer11, Lis Arend1,2, Klaudia Adamowicz1, Tanja Laske1, Anne Hartebrodt12, Tobias Frisch13, Chen Meng2, Julian Matschinske1, Julian Späth1, Richard Röttger13, Veit Schwämmle13, Stefanie Hauck3, Stefan Lichtenthaler6, Axel Imhof4, Matthias Mann5, Christina Ludwig2, Bernhard Küster2, Jan Baumbach1, Olga Zolotareva1,2 1University of Hamburg, Hamburg, Germany 2Technical University of Munich, Freising, Germany 3Helmholtz Center Munich, Munich, Germany 4Ludwig-Maximilians-University Munich, Planegg, Germany 5Max Planck Institute of Biochemistry, Martinsried, Germany 6German Center for Neurodegenerative Diseases, Goettingen, Germany 7Fraunhofer Institute for Cell Therapy and Immunology IZI, Leipzig, Germany 8University of Applied Science Coburg, Coburg, Germany 9University Hospital Erlangen, Erlangen, Germany 10Leiden University Medical Center, Leiden, Netherlands 11University of Greifswald, Greifswald, Germany 12Friedrich-Alexander-Universität Erlangen-Nürnberg, Erlangen, Germany 13University of Southern Denmark, Odense, Denmark Expanding proteomics data availability unlocks potential for large-scale biomedical research, but integrating patient-derived mass spectrometry data can be problematic due to privacy concerns. Protein expression profiles, just like transcriptomics data, can be subject to genotype reconstruction attacks and must be treated as confidential information [1]. To enable privacy-preserving, robust analysis of distributed proteomic data despite data heterogeneity, we designed FedProt — the first tool for federated differential protein abundance analysis. FedProt represents the mathematical equivalent of DEqMS, a state-of-art modification of the limma method [2], and utilizes the hybrid approach of federated learning (FL) [3] and additive secret sharing [4]. In FL, multiple parties holding sensitive data jointly participate in the computation without revealing their data. This involves splitting the workflow into steps performed locally and aggregating results globally by a trusted server. In FedProt, this approach allows us to obtain the same result as centralized pooled data analysis, but without violating privacy. We evaluated FedProt using two multi-center datasets, a TMT human plasma dataset of 60 samples (2 conditions, 3 cohorts) and a LFQ dataset of 118 Escherichia coli samples (2 conditions, 5 cohorts). All MS data were uniformly preprocessed, quantified, and submitted to FedProt, which identified differentially abundant protein groups. We conducted a central DEqMS analysis as a baseline and compared the results of FedProt against it. Additionally, we studied class label and cohort size imbalance using simulated data. In all tests, FedProt effectively handled batch effects and produced results that closely matched the results of DEqMS baseline; fold-changes and negative log-transformed adjusted p-values were almost identical (maximal differences no greater than 1*10-11). It can analyze proteins absent in some cohorts and matches centralized analysis more precisely than typical meta-analyses like Fisher's or Stouffer's methods, RankProd, or the random effects model, representing the only privacy-preserving approach to differential protein abundance analysis to date. Using FL and additive secret sharing, FedProt effectively manages proteomic data complexity, including missing values and batch effects, enhancing sample sizes and statistical power while minimizing privacy risks without compromising accuracy. Accessible as a user-friendly FeatureCloud App (https://featurecloud.ai/app/fedprot), FedProt could catalyze larger collaborations and advance proteomics research, improving the robustness of differential protein abundance analysis results. 1. Geyer, P.E., et al. Mol Cell Proteomics. 20, 2021, 100035. 2. Zhu, Y., et al. Mol Cell Proteomics. 19, 2020, 1047–1057. 3. McMahan, B., et al. PMLR, AI and Statistics, 2017, pp. 1273–1282. 4. Cramer, R., et al., 2015. SMPC. Cambridge University Press. 71

OP-40 Illuminating the human phosphoproteome by dose-resolved proteomics of clinical kinase inhibitors Florian P. Bayer1, Julian Mueller1, Matthew The1, Bernhard Küster1 1Technical University Munich, Proteomics, Freising, Germany Introduction: Pathological kinase signaling is among the key causes of human diseases, as it shifts the phosphoproteome out of its homeostasis and leads to uncontrolled cell growth or inflammation. Rationally, kinase inhibitors have become a popular treatment strategy to restore the phosphoproteome. Although many phosphorylation sites (p-sites) can now be analyzed by mass spectrometry, the biological contexts and functions of these p-sites remain hidden in the dark for most cases. Consequently, the complete sets of mechanisms of action(s) of kinase inhibitors in cells remain similarly vague. The known polypharmacology of most inhibitors must be carefully considered in any perturbation experiment and can only be dissected by exploiting affinity differences from dose-resolved experiments. Only by systematically screening inhibitors in various cell line models in a dose-dependent manner, it becomes now possible to understand the consequences of kinase inhibition on a systems level. Methods: We developed a scalable "decryptM" screening platform in a 96-well format and combined it with a mostly automated phosphoproteomic workflow. MS raw files were searched with MaxQuant and boosted with SIMSI and Percolator. Relevant p-site-specific dose-response curves were identified using the novel tool "CurveCurator". Kinase inhibition estimates were obtained from curated seed p-sites. New kinase substrate relationships were obtained by characteristic potency signatures across the data set. Results: We performed systematic and short dose-resolved perturbations of a total of 134 approved and phase 3 clinical kinase inhibitors across 5 different cell lines. This results in a complete data matrix of 715 decryptM experiments that are based on 7,865 individual perturbations at a depth of around 20-25"000 p-sites per drug and cell line. All 28 million dose-response curves were fitted and statistically evaluated using our new analysis tool, CurveCurator, which combines statistical significance with biological effect size into a single relevance score that outperforms traditional Benjamini-Hochberg procedures in terms of specificity and sensitivity. Overall, the global degree of perturbation is between 1-10%, constraining the response always close to the affected pathways. Harnessing the potency dimension of decryptM experiments, which hints at the initially inhibited kinase(s), we identified many known and unknown p-sites that follow their direct upstream kinase inhibition/activation signature across the entire data matrix. Additionally, we also collected clear evidence for numerous questionable kinase-substrate relationships in public databases. On a higher level, we also obtained an activity-based view of the dependencies of different kinases and cellular backgrounds. Conclusion: This is the largest decryptM study existing which allowed a systematic grouping of signaling relevant p-sites into a kinase regulative context by exploiting the potency dimension of drugs. 72

OP-41 Proteomics of neuroinflammatory signalling: illuminating cellular mechanisms driving neurodegenerative diseases Katrin Stuber1, Florian Mau1, Jesper Olsen1 1NNF Center for Protein Research, University of Copenhagen, Faculty of Health and Medical Sciences, Copenhagen, Denmark Alzheimer’s (AD) and Parkinson"s disease (PD) are prevalent neurodegenerative diseases, but their underlying mechanisms remain poorly understood. Emerging evidence suggests that neuroinflammation, characterized by the release of pro-inflammatory cytokines like IL-1β, TNF-α, and INF-γ, plays a pivotal role in these diseases, challenging the conventional focus on protein aggregates (Aβ, αSyn). In AD, it has been shown that elevated IL- 1β levels induce the translocation of NEDD8 from the nucleus to the cytoplasm, where it co-localizes with the E3-ligase Parkin. Together with the kinase PINK1, Parkin is crucial for the removal of defective mitochondria, with loss-of-function mutations Intriguingly, phosphorylated NEDD8 can activate Parkin, interconnection between neuroinflammation, NEDD8-and Parkin-regulated phosphorylation-dependent signaling in brain cells. Yet, a mechanistic understanding of cellular processes triggered by neuroinflammation remain elusive. in these genes being the primary cause of early-onset PD. implicating a more complex This study aims to investigate spatiotemporal (phospho-)proteome changes in response to neuroinflammatory stress elicited by pro-inflammatory cytokines and αSyn fibrils. To achieve this, spatial-proteomics using subcellular fractionation is combined with high-sensitivity narrow-window data-independent acquisition (nDIA) using the Orbitrap Astral mass spectrometer. This workflow resulted in identification and quantification of ⁓8000 protein groups and between ⁓6000 and 15,000 localized phosphorylation sites per subcellular fraction. IL-1β-dependent subcellular relocation of hundreds of (phospho)proteins was determined in initial analyses of neuronal cell line models. This strategy will be extended to a more physiologically-relevant model system of hPSC derived neurons, astrocytes, microglia, and co-cultures thereof. This system enhances the translational relevance by mimicking neuroinflammation in the complex tissue architecture of the brain. The data obtained reveals the cellular protein abundance changes and phosphorylation site regulation induced under neuroinflammatory stress. This study aspires to unravel the cellular mechanisms underlying neuroinflammation, potentially unlocking new avenues for therapeutic strategies against neurodegenerative diseases. 74

OP-42 Time- and spatially-resolved TRAP analysis identifies drug targets and deciphers signaling network in vitro and in vivo Hanqing Zhang1, Ning Wan1, Yi Li1, Hui Ye1 1China Pharmaceutical University, State Key Laboratory of Natural Medicines, Nanjing, China Target identification reveals how drugs act, and the discovered targets represent a valuable repertoire for therapeutic target discovery. Current approaches mostly investigate the direct binding targets (directTs) for drugs and lack the ability to decipher the downstream effector proteins (defined as "indirectTs"). Previously, we developed a target-responsive accessibility profiling (TRAP) approach, which measures ligand-binding induced changes in accessibility for target identification by global labeling of reactive proteinaceous lysines via reductive demethylation (Nat Chem Biol., 2023, 19, 1480–1491). We proposed that when the directTs bind to its indirectTs, the chemical accessibility also changes and can thus be detected by TRAP. Since the engagement between the drug and its directTs precedes the engagement between the directTs and the indirectTs, we sought to perform time-resolved TRAP to simultaneously identify the direct -indirect target networks. To meet the high throughput requirements of time-resolved TRAP, data-independent acquisition (DIA) (Nature, 2016, 537, 347–355) is incorporated into the TRAP workflow for label-free quantification of TRAP-labelled peptides. Using TEPP-46 as a model compound, we incubated it with cells and then performed TRAP labelling at 6 different time points after administration (0, 5 min, 15 min, 30 min, 1 hr, 2 hr). Subsequent DIA-TRAP analysis allowed us to identify its bona fide target, PKM2, which showed the most significant decrease in accessibility at the earliest time point (5 min). Interestingly, changes in the accessibility of glycolytic enzymes such as ALDOA and GAPDH were observed at 30 min post-administration. These proteins have been annotated as interacting proteins of PKM2. Surprisingly, at a later time point (2 h), we observed that proteins involved in the cellular response to stress and the organic cyclic compounds metabolic process, which have not been linked to PKM2, also showed altered accessibility. AlphaFold was used to assess the possibility of interactions and biochemical experiments are ongoing to elucidate the signaling pathways involved. With the analysis power enabled by DIA-TRAP, we asked whether we can assess ligands-induced accessibility changes in vivo and identify targets in a spatially-resolved manner. As a proof-of-concept, we administered TEPP- 46 to mice inoculated with transplanted tumors and performed TRAP analysis on divided tumor tissues (core and periphery). TRAP analysis showed that the target PKM2 most significantly responded among all quantified TRAP peptides in all assayed regions. Notably, TRAP ratio of PKM2 extracted from tumor periphery most markedly changed when compared to the core, agreeing with the in vivo distribution of TEPP-46. Overall, we demonstrate that DIA-TRAP enables time and spatially-resolved TRAP analysis. It allows us to identify targets for bioactive ligands and profile the downstream effectors, and is applicable to in situ contexts. 75

OP-43 Ultra-deep longitudinal profiling of plasma and serum proteome provides critical insights into the interaction of human organism and pig xenotransplants Alexey Stukalov1, Eloi Schmauch2, Brian Piening3, Simon Williams4,5, Ian Jaffe4,5, Maedeh Mohebnasab6, Shadi Ferdosi1, Alexa Dowdell3, Karen Khalil4, Adam Griesemer4,5, Jacqueline Kim4,5, Jef Boeke7, David Ayares8, Marc Lorber9, Massimo Mangiola4, Harvey Pass4,10, Vasishta Tatapudi4,5, Jeffrey Stern4,5, Robert Montgomery4,5, Serafim Batzoglou1, Asim Siddiqui1, Brendan Keating4,5,7 1Seer, Inc., Redwood City, CA, United States 2Broad Institute of MIT and Harvard, Cambridge, MA, United States 3Providence Cancer Center, Earle A. Chiles Research Institute, Portland, OR, United States 4NYU Langone Health, NYU Langone Transplant Institute, New York, NY, United States 5NYU Grossman School of Medicine, Department of Surgery, New York, NY, United States 6University of Pittsburgh Medical Center, Division of Molecular & Genomic Pathology, Pittsburgh, PA, United States 7NYU Langone Health, Institute for Systems Genetics, New York, NY, United States 8Revivicor Inc., Blacksburg, VA, United States 9United Therapeutics, Silver Spring, MD, United States 10NYU Langone Health, Department of Cardiothoracic Surgery, New York, NY, United States Xenotransplantation of pig organs to humans holds immense potential to address the critical shortage of donor organs. Recent advancements in genetic engineering and immunosuppressive therapies have brought xenotransplantation closer to reality, with successful pig-to-human kidney and heart transplants marking significant milestones. However, understanding the complex molecular interactions between the recipient's organism and the xenograft remains a major challenge. Measuring plasma proteome of these patients provides critical information about immune response and status of many organism subsystems. In a series of studies, we employed an advanced proteomic workflow, combining nanoparticle-based sample preparation (Proteograph™ XT, Seer, Inc.) with the state-of-the-art LC-MS using Ultimate™ 3000 HPLC and Thermo Scientific™ Orbitrap™ Astral™ MS. Both components allowed us to significantly improve the proteome coverage in comparison to the standard workflows and deliver ultra-deep profiling of plasma or serum proteomes in human recipients of a genetically modified pig organ, including the unprecedented daily serum monitoring of the brain-dead patient with xenokidney over 61 days. The samples were analyzed in a DIA mode with 3 m/z isolation window using a 30-min gradient. In these datasets we identified around 100K peptides, half of which could be specifically assigned to human or pig proteins and allowed us to reconstruct around 6,200 human and 1,000 pig protein groups. Our analysis revealed the intricate regulation patterns for thousands of human and pig proteins. We observed significant changes in the complement system, cytokines, immunoglobulins, and various metabolic pathways, reflecting the recipient's immune response and the physiological adaptation to the xenograft. Notably, we could track the abundance dynamics for hundreds of pig-specific proteins in the recipient's blood, highlighting the molecular details of xenograft integration, including its immune system response. By using the whole plasma proteome as a comprehensive indicator of a recipient"s state, we could observe the differential dynamics of individual xenotransplantation patients that correlated with the other clinical and omics data. Such analysis is instrumental in identifying potential human and xenograft biomarkers that would contribute to more efficient post-operational treatment of these patients. Ultimately, this research paves the way for xenotransplantation to become a viable solution for patients in need of life-saving organ transplants. 76

OP-44 Parallel global proteome and glycaproteome of human brain reveal altered glycation of key proteins linked to Alzheimer's disease Qin Fu1, Elizabeth Anderson1, Gary Gibson2, Sarah Flowers3, SHENG ZHANG1 1Cornell University, Institute of Biotechnology, Ithaca, NY, United States 2Weill Cornell Medicine, White Plains, NY, United States 3University of Virginia, Charlottesville, VA, United States Although amyloid-beta (Aβ) deposition and tau neurofibrillary tangles in brain are hypothesized as a central force driving Alzheimer’s disease (AD) pathogenesis, the mechanism underlying the development pathology is unknown. Protein glycation with primary amine groups of proteins reacting with aldehyde or reactive carbonyl group of a sugar affects structure, bioactivity and molecular stability. While increased glycation has been reported in AD brains, omic studies have never been done. Here we analyzed proteome and glycaproteome of brains from AD patients and cognitively normal individuals in parallel to explore roles of glycation in AD pathology, and to identify potential new biomarkers and novel therapeutic targets. Proteins were extracted from human brains of five controls and five AD patients. TMT10-plex labeling was used for global proteomics and boronate affinity enrichment of glycapeptides was performed for glycaproteome. All samples were analyzed by nanoLC-MS/MS on Orbitrap Eclipse and data processed by Proteome Discoverer. In the global proteome, 530 proteins of the 4,848 that were quantified with statistical differences between AD and controls. Confident quantification of 814 glycapeptides from 395 glycaproteins revealed six different glycated forms including Nε-fructosyl, on both Lys and Arg residues were determined. Glycation peptides were consistently increased in the AD samples. While 105 unique glycated peptides from 73 unique glycated proteins were significantly increased (p<0.05) in AD patients, no glycapeptides were significantly decreased. Glycation of several key proteins associated with AD were increased including amyloid-beta precursor protein (APP), Tau,glyceraldehyde-3-phosphate dehydrogenase (GAPDH), glial fibrillary acidic protein (GFAP), fibrinogen and synaptophysin. APP is the precursor protein that when cleaved generates the β-amyloid peptides, and was found with Nε-fructosyl modification on its key K612 residue. This is remarkable as we found previously the same site of APP was modified by succinylation, which is linked to form amyloid plagues. Increased glycation was found on K311, K353, K369 and K375 of Tau protein. K311 was succinylated promoting pathological tau aggregation to tangles and impairing microtubule assembly. The increased glycation fold change over the change of their protein abundance suggests that glycapeptides may represent sensitive biomarkers of AD. Our data also suggest there appears crosstalk between glycation and other post-translation modification such as succinylation in potential regulation of AD pathogenesis. 78

OP-45 Personalized neuroinflammatory status in neurodegenerative diseases by a panel of 20 cerebrospinal fluid markers Ignacy Rzagalinski1, Johanna Knoeferle2,3, Mikael Simons2,3, Andrej Shevchenko1 1Max Planck Institute of Molecular Cell Biology and Genetics, Dresden, Germany 2Technical University Munich, Institute of Neuronal Cell Biology, Munich, Germany 3German Center for Neurodegenerative Diseases (DZNE), Munich, Germany Neuroinflammation plays a critical role in the pathogenesis and progression of Alzheimer’s Disease (AD) and other neurodegenerative disorders. We developed and validated a sensitive and accurate multiprotein mass spectrometry assay for the absolute (molar) quantification of 20 key neuroinflammation markers in human CSF, and employed the method to study the inflammatory status along the AD continuum in the cohorts of subjective cognitive decline (SCD), mild cognitive impairment (MCI) and AD dementia, as well as corticobasal degeneration (CBD), a rare progressive neurodegenerative disorder. The panel comprised apolipoproteins: ApoA1, ApoD, ApoJ as well as ApoE with all of its three isoforms (E2, E3 and E4), the strongest genetic risk factor for AD. It also included the markers of reactive astrogliosis (GFAP, YKL- 40 and complement proteins: C1q, C3, C4) and microglia activation (TREM2). Besides, it contained proinflammatory cytokines (MIF, osteopontin and alpha-1-antichymotrypsin) and the two TAM receptors (Axl and Tyro3). Altogether, the quantified marker proteins spanned 10,000-fold molar concentrations ranging from 20 pM (GFAP) to 200 nM (ApoD) and were detected in a single LC-MS/MS run with excellent precision (median CV <6%) and inter-peptide agreement (typical r >0.98). We applied the developed panel to the cohort of 109 patients (SCD=24; MCI=42; AD=43). We investigated the association of panel proteins with other CSF clinical markers (Aβ42, p-Tau and t-Tau) using the ATN classification system, as well as with the mini-mental state examination (MMSE) scores. Notably, four proteins (GFAP, osteopontin, YKL-40 and MIF) showed a significantly higher CSF molar concentrations in AD (and two proteins in MCI) vs. our controls (SCD). Among them, osteopontin yielded the best diagnostic performance in the discrimination of AD from controls (AUC 0.81). Next, by investigating whether a minimal combination of biomarkers could discriminate AD from controls with higher accuracies, we identified a panel of 5 proteins (GFAP, osteopontin, MIF, C1q and C3) that showed a considerably higher accuracy than any marker alone (AUC 0.91). Finally, we compared the CSF neuroinflammatory markers for AD (n=35) and Aβ-negative CBD patients (n=37). Interestingly, the results showed no discrimination between these two groups, thus suggesting a similar neuroinflammation profile shared by these two tauopathies. In the end, we propose that the developed multiprotein panel will contribute to personalized diagnostics by monitoring inflammatory signatures of CSF in the broad spectrum of neurodegenerative diseases. Fig. 1 (optional) 79

OP-46 High-throughput Glycoproteomics: deciphering age-related Glycosylation changes in the mouse brain Delf-Magnus Kummerfeld1, Johanna Schrader2, Aleksander Nowak3, Joanna Kirkpatrick4, Peter Milland4, Tabiwang Arrey4, Nicolaie Eugen Damoc4, Boris V. Skryabin5, Hans Wandall3, Timofey S. Rozhdestvensky5, Sergey Vakhrushev3 1Fritz Lipmann Institute, Leibniz Institute on Aging, Jena, Germany 2Leibniz University Hannover, L3S Research Center, Hannover, Germany 3University of Copenhagen, Cellular and Molecular Medicine, Copenhagen, Denmark 4Thermo Fisher Scientific, Bremen, Germany 5Muenster University, Transgenic animal and genetic engineering Models, Muenster, Germany Glycoproteomics is essential for studying diseases like cancer, cardiovascular disorders, autoimmune diseases, neurodegenerative conditions, and metabolic syndromes. However, its progress is limited by the lack of sensitive, high-throughput technologies. We introduce a novel glycoproteomics sample preparation pipeline, sLWAC-HTP, capable of processing up to 400 samples within a 3-5 hour timeframe. Utilizing the Orbitrap Astral mass spectrometer, known for its exceptional speed and sensitivity, our comprehensive high-throughput glycoproteomics platform integrates sLWAC-HTP for sample preparation, Orbitrap Astral MS for analysis, and GlycoDIA for data processing. This platform enabled us to investigate age-related glycosylation alterations in distinct compartments of the mouse brain, providing insights into glycosylation dynamics during brain development, maturation, and aging. Brain regions from male C57BL/6J mice at various ages—postnatal days 14 and 21, and at 2, 5, 18, and 20 months—were dissected, snap-frozen, and stored at -80°C. Tryptic O-glycopeptides from different brain compartments across six age points, each with four biological replicates, were enriched using the Jacalin lectin- based sLWAC-HTP platform. Peptides were separated by reversed-phase UHPLC on a PepMap 150 μm x 15 cm column over a 3.1-minute linear gradient using the Thermo Scientific Vanquish Neo system in Trap&Elute mode. Data were acquired on a Thermo Scientific Orbitrap Astral mass spectrometer in high-speed narrow window data-independent acquisition (DIA) mode and processed using Spectronaut software. Our miniaturized lectin-based platform, evolved from the standard sLWAC method, has advanced to a high- throughput variant termed sLWAC-HTP. This innovation enables rapid glycopeptide enrichment from complex proteomic samples with enhanced efficiency, comparable to the traditional LWAC method. The system has been optimized for high-throughput applications using a 96-well plate format and a standard benchtop centrifuge, allowing processing up to 4x96 samples in approximately 3-5 hours. Using the Orbitrap Astral MS with an ultra- short 4-minute LC gradient and the GlycoDIA platform, we achieved a throughput of 220 samples per day. Applying this pipeline for quantitative glycoproteomics analysis of 10 different brain compartments at six age points, we detected, on average, 7000-10000 unique glycopeptide precursors per compartment, per age. These findings are unprecedented in brain O-glycoproteomics. The age-related differences in glycosylation across brain regions provide valuable insights into dynamic changes during brain development and aging. Our high throughput glycoproteomics platform represents a significant advancement in glycoproteomic analysis, marking a major shift in research approaches for disease studies and biomarker discovery, leading to transformative progress in glycobiology and biomedical research. 80

OP-47 Finding the sweet spot – the pursuit of single cell glycomics by CE-MS Wenjun Wang1,2, Jing Zheng1, Karthika Korumadathil Shaji1, Hector Rhault Loponte1, Alienke van Pijkeren1, Peter Horvatovich1, Manfred Wuhrer2, Guinevere Lageveen-Kammeijer1 1Groningen Institute of Pharmacy, University of Groningen, Analytical Biochemistry, Groningen, Netherlands 2Leiden University Medical Center, Center for Proteomics and Metabolomics, Leiden, Netherlands Introduction: There is still an urgent clinical need for new specific molecular targets that can be utilized to establish better diagnostic, prognostic as well as therapeutic treatments for cancer patients. A hallmark of many diseases, including cancer, are alterations in glycosylation (e.g. covalent attachment of sugar moieties to proteins or lipids), providing new opportunities for better patient stratification. This is not surprising as glycosylation is involved in many biological processes, and glycan structures cover our cells in a dense layer and contribute to the malignant phenotype of cancer cells by promoting proliferation, metastasis, and immunosuppression. However, a more thorough investigation is needed on the glycomic intratumoral heterogeneity (e.g. primary versus metastatic tumors), as tumors consist of a network of cell populations that are interconnected in tissue matrix. Therefore, it is of utmost importance that analytical workflows are used that can distinguish between these populations, and reveal the relevant cells by single cell analysis. However, current glycomic methods still require a large amount of cells (20,000 up to 100,000 cells). Here, we present an innovative and miniaturized high-sensitivity glycomics workflow (capillary electrophoresis coupled to mass spectrometry; CE-MS). Methods: Various conditions of the 96-well polyvinylidene difluoride (PVDF) membrane approach were evaluated (e.g. membrane blocking and N-glycan release conditions). By implementing the release on the membrane as well as using a hydrazide labeling procedure, no additional purification steps were required. Analysis was performed on a CE-MS platform and the complete workflow was further evaluated by its application for the characterization of the N-glycome from total plasma, extracellular vesicles and two cell lines (PaTu-S and JY) using various amount of lysed cells (1 – 200,000 cells). Results: A highly diverse N-glycome was observed for all samples and the CE-MS demonstrated its ability to separate isomeric structures, in particular differently linked sialic acids, without the need of an additional derivatization step. A total of 72 N-glycans could be relatively quantified from 5,000 PaTu-S cells using the optimized protocol. The five most abundant N-glycans could still be detected using ~1 cell after cell lysis. Conclusions: In this study we achieved an exceptional sensitivity for glycomic analysis that only requires minimal sample amounts (e.g., up to a single cell or 100 pg of protein weight). Future endeavors will focus on using the workflow for isolated single cells (sorted by the CellenONE system) as well as to apply it on laser-capture microdissected samples. In time, this will allow us to gain in-depth insights into the role of glycosylation in tumor progression and how these alterations can be exploited for personalized treatment strategies through improved diagnosis and prognosis of cancer patients. 81

OP-48 New insights into the lysosomal glycoproteome Sayantani Chatterjee1, Rebeca Kawahara2, Markus Damme3, Thomas Braulke4, Morten Thaysen-Andersen1,2 1Macquarie University, School of Natural Sciences, Sydney, Australia 2Nagoya University, Institute for Glyco-core Research (iGCORE), Aichi, Japan 3Christian-Albrechts-University Kiel, Institute of Biochemistry, Kiel, Germany 4University Medical Center Hamburg-Eppendorf, Department of Osteology and Biomechanics, Hamburg, Germany Lysosomes, essential catabolic cellular organelles packed with diverse lysosomal hydrolases, play critical roles in biomolecular degradation, autophagic processes and extracellular cell-cell communication. While the protein complement of the lysosome is well-studied, the lysosomal glycoproteome remain poorly understood despite the recognised importance of protein glycosylation including mannose-6-phosphorylation (M6P) in lysosome biology. To this end, we performed an in-depth glycoprofiling of both soluble and membrane protein fractions from isolated liver lysosomes (tritosomes) from wild-type (WT, n = 3-5) and acid-phosphatase 2/5 double knockout (ACP2/5-dKO, n = 3) mice using quantitative glycomics and glycoproteomics. ACP2/5-dKO mice were included in the study to also map the unprocessed lysosomal glycoproteome ahead of the final M6P removal from trafficking lysosomal glycoproteins. From WT lysosomes, we identified 1000+ unique N-glycopeptides covering 200+ N-glycosylation sites from 120 source glycoproteins comprising the expected lysosomal hydrolases. Glycomics identified 60 different N-glycan isomers spanning a surprisingly high abundance of paucimannose (Man1-3GlcNAc2Fuc0-1, >50% of the N-glycome) in addition to oligomannosidic and complex- type N-glycans. As expected, no M6P-tagged glycoproteins were found in the WT lysosomes. The WT lysosome displayed different N-glycome profiles across the soluble and membrane protein fractions indicating differential N-glycan processing of those protein subsets. Surprisingly, the glyco(proteo)mics data of ACP2/5-dKO lysosomes revealed very low levels of M6P (<10% of the N-glycoproteome) on only a few soluble lysosomal proteins (prosaposin, cathepsin D/L, dipeptidyl-peptidase-1). Similar to the WT lysosomes, the ACP2/5-dKO lysosomes instead displayed a paucimannose-rich glycophenotype (~50% of the N-glycome) and only relatively little oligomannosylation and complex-type N-glycosylation indicating that only a small proportion of the lysosomal hydrolases traffic to mouse liver lysosomes via the well-established M6P-tagging pathway. This study generates, for the first time, a comprehensive and quantitative N-glycoproteome map of the mouse liver lysosome representing an important resource to further unravel the glycobiology of lysosomes and their trafficking pathways. 83

OP-49 Immunoproteomic analysis of cellular and adaptive responses in naïve and recovered TB patients following intra- lung re-challenge with live BCG Michelle Mullins1, Javan Okendo1, Clemens Hermann1, Keertan Dheda1, Jonathan Blackburn1 1University of Cape Town, Institute of Infectious Disease & Molecular Medicine, Cape Town, South Africa Tuberculosis disease (TB) remains a major global health challenge today, despite the availability of vaccine and an armoury of small molecule drugs. Notably, response rates to the century-old bacillus Calmette-Guerin (BCG) vaccine vary widely, for reasons that remain poorly understood. We have therefore developed a lung-oriented controlled human infection model of tuberculosis disease to better understand TB immunopathogenesis at the site of disease and across a spectrum of disease susceptibilities. We recruited 99 healthy South African participants who had previously either single or multiple bouts of TB disease, a latent TB infection, or who appeared to have sterilising immunity. Live BCG, sterile PPD and saline were bronchoscopically instilled into separate lung segments (n = 65). A control group (n = 34) underwent a single bronchoscopy without challenge. Bronchoalveolar lavage (BAL) samples were collected at baseline and 3 days after re-challenge and analysed using SWATH-based proteomics, as well as using mycobacterial protein- and lipid-microarrays, in order to quantify cellular immune responses and anti-microbial antibody signatures as a function of challenge and disease susceptibility. We quantified 4901 peptide IDs and 949 human protein IDs in the BAL samples through the SWATH-based analysis, with a data completeness of 79% across all samples. Differential expression was analysed in limma with Benjamin-Hochberg multiple testing correction, comparing baseline and 3 days post-rechallenge samples across the four disease susceptibility groups. We also quantified antigen-specific IgG and IgA titres in BAL and serum using custom mycobacterial protein NAPPA microarrays comprising 687 cell wall-associated antigens, as well as on home-made mycobacterial lipid microarrays. We observed that immune responses were highly compartment specific (BAL vs. blood) and localized to the challenged lung segments. We also found that cellular and humoral responses increased in magnitude with increasing disease susceptibility, but that these were highly heterogeneous within a challenge- and susceptibility group. In particular, integration of our multi-omic proteomic and antibody data demonstrates the presence of previously unrecognised sub-groups within the clinically-defined latent TB infection group and points to significant mechanistic differences in the immune responses upon re-exposure to the pathogen between these sub-groups. More specifically, STRING analysis of our data identifies a co-ordinated antigen-specific IgA- and neutrophil-driven response at the site of disease in one LTBI sub-group that is likely to be protective but which is absent in in the other LTBI sub-group. Collectively, our integrated proteomic data therefore sheds important new light on the latent TB infection state, with implications amongst others for control of TB transmission, as well as for WHO modelling of elimination of TB disease by 2050. Fig. 1 (optional) 84

OP-50 Population scale proteomics enables adaptive digital twin modelling in sepsis Johan Malmström1 1Lund University, Department of Clinical Sciences, Lund, Lund, Sweden Sepsis is one of the leading causes of mortality in the world. Currently, the heterogeneity of sepsis makes it challenging to determine the molecular mechanisms that define the syndrome. Here, we leverage population scale proteomics to analyze a well-defined cohort of 1364 blood samples taken at time-of-admission to the emergency department from patients suspected of sepsis. We identified panels of proteins using explainable artificial intelligence that predict clinical outcomes and applied these panels to reduce high-dimensional proteomics data to a low-dimensional interpretable latent space (ILS). Using the ILS, we constructed an adaptive digital twin model that accurately predicted organ dysfunction, mortality, and early-mortality-risk patients using only data available at time-of-admission. In addition to being highly effective for investigating sepsis, this approach supports the flexible incorporation of new data and can generalize to other diseases to aid in translational research and the development of precision medicine. 85

OP-51 Organ atlas proteome mapping of cryptococcosis defines novel putative biomarkers and promising druggable targets for personalized health management Jennifer Geddes-McAlister1, Benjamin Muselius1, Arnaud Droit2, Florence Roux-Dalvai2 1University of Guelph, Molecular and Cellular Biology, Guelph, Canada 2Laval University CHU – Quebec, Research center, Quebec, Canada The ability to profile changes in both host and pathogen proteomes during infection provides a unique opportunity to identify novel mechanisms of interplay driving disease. Such information supports an improved understanding of a diseased state and promotes novel strategies for disease detection and treatment. Here, we profiled the interplay between the human fungal pathogen, Cryptococcus neoformans, and the mammalian host using high-resolution mass spectrometry-based proteomics to demonstrate the dynamic nature of infection and to define novel diagnostic biomarkers of disease. We established an in vivo murine model of cryptococcosis and profiled 13 clinically relevant organs, tissues, and fluids across a temporal continuum of disease. Tandem mass tags combined with high pH fractionation improved protein quantification and increased proteome coverage across 780 samples with the proteomic data curated and publicly available through our development of the "CProteo Atlas". We demonstrate the value of this resource by defining novel putative dual species biomarker signatures for detecting and monitoring progression of infection, as well as provide organ and temporal-specific drivers of disease. We also leverage our findings to reveal a novel putative druggable target with a key role in blood brain barrier crossing through in vitro and in vivo assays. Overall, we demonstrate dynamic proteome remodelling of cryptococcosis in a temporal and spatial-dependent manner to uncover new putative biomarkers and druggable targets of infection for informed detection, monitoring, and treatment of disease. 86

OP-52 Maximizing LC-MS sensitivity and throughput in single-cell proteomics Runsheng Zheng1, Rupert Mayer2, Manuel Matzinger2, Alec Valenta1, Tabiwang Arrey3, Xuefei Sun4, Christopher Pynn1, Ece Aydin1, Wim Decrop1, Dominic G. Hoch5, Martin Samonig1, Karl Mechtler2 1Thermo Fisher Scientific, Germering, Germany 2Research Institute of Molecular Pathology (IMP), Vienna, Austria 3Thermo Fisher Scientific, Bremen, Germany 4Thermo Fisher Scientific, Sunnyvale, CA, United States 5Thermo Fisher Scientific, Reinach, Switzerland Introduction: Proteomics has recently shifted towards profiling smaller sample quantities, including single-cell analysis. Although state-of-the-art liquid chromatography-mass spectrometry (LC-MS) technology has demonstrated unprecedented performance in investigating proteome changes, it lacks sensitivity for low- abundance peptides in single-cell proteomics (SCP) applications. Naturally, LC-MS sensitivity can be enhanced by utilizing ultralow LC flow rates (<20 nL/min) with small inner diameter columns (20-30 µm) and emitters, but this comes at the cost of decreased sample throughput and restricting the possibility to analyze large sample cohorts. Therefore, a balance between sensitivity and sample throughput is the prerequisite for large-scale single-cell sample profiling. To address this dilemma, we successfully developed a low-nanoflow UHPLC separation setup with a wide-window data-independent acquisition (WW-DIA) strategy to achieve deep proteome profiling in a high-throughput fashion. Methods: We established a novel low nano flow LC-MS configuration running gradients at 100 nL/min on a Vanquish Neo UHPLC system with an Acclaim PepMap 100 C18 50 µm ID x 15 cm column (2 um dp) in direct injection and trap-and-elute workflows. Electrospray ionization was carried out via a 10 µm ID glass emitter into a FAIMS Pro interface operated at a single compensation voltage to reduce background ion interference, followed by data acquisition in an Orbitrap Exploris 480 or an Orbitrap Astral mass spectrometer. Preliminary results: Seven chromatography methods were developed to permit fast sample loading, efficient column washing, and equilibration while maintaining a stable 100 nL/min flow during the gradient. This permits the analysis of up to 120 samples/day with a 8-min peptide elution window (i.e., 67% MS utilization), whilst also offering extremely high levels of sensitivity for routine SCP analysis. Using this optimized front-end configuration, we identified ~2,200 protein groups from 250 pg of standard protein digest using the WW-DIA strategy in a library-free search without booster channels. A proof of principle study indicated that >1,700 protein groups were identified and quantified from single-cell samples in a label-free experiment with negligible carryover at 100 cells/day (CPD) throughput. Furthermore, employing the Orbitrap Astral mass spectrometer significantly improved performance by doubling the protein identifications and quantification. In conclusion, we established a high-performance LC-MS configuration for SCP by balancing the sensitivity and throughput, thus enabling the investigation of cellular heterogeneity in clinically and biologically relevant cellular populations. 87

OP-53 Combining Proteome Integral Solubility Alteration (PISA) assay and one-tip workflow for a streamlined and high- throughput drug target deconvolution Maico Lechner1, Pierre Sabatier1,2, Jesper Olsen1 1Novo Nordisk Foundation Center for Protein Research, Copenhagen, Denmark 2Uppsala University, Department of Surgical Sciences, Uppsala, Sweden The Proteome Integral Solubility Alteration (PISA) assay is a well-established mass spectrometry (MS)-based proteomics methodology. It identifies changes in protein-ligand complexes based on their solubility variation under different treatment conditions based on thermal probing at several temperatures and is applicable to living cells and cell lysates. PISA usually necessitates relatively high sample input, but biomedically-relevant material available for analysis is often limited. This is the case for patient derived samples or challenging and costly cell cultures, and thus reducing sample requirements would increase the widespread applicability of PISA. Utilizing the One-Tip workflow, a novel sample preparation technique based on the Evotip, streamlines analysis of low sample input due to direct protein digestion on top of the C18-disk in the tip, followed by desalting, concentration and injection into the Evosep One LC system. The addition of 0.2% n-Dodecyl-Beta-Maltoside (DDM) during the freeze-thaw cell lysis after the PISA assay allows for a swift transition to the One-Tip method. Since DDM is a MS compatible detergent, there is no need to perform a buffer exchange of the protein sample prior to enzymatic Lys-C/trypsin digestion, thus streamlining the method by decreasing sample preparation time as well as minimizing variability thanks to fewer pipetting steps. To test the sensitivity of the integration of PISA with the One-Tip workflow, we assayed the cellular targets of staurosporine, a broadband kinase inhibitor, in different quantities of HeLa cells. The PISA samples were analyzed on the Orbitrap Astral mass spectrometer using narrow-window DIA. As a readout, we investigated the correlation between cell number input and subsequent protein IDs and identified kinase targets. This showed that One-Tip-PISA can be performed on very low cell inputs, as low as 2500 HeLa cells for each temperature point sampled. This yielded reproducible results showing a strong correlation with known staurosporine targets, surpassing the protein depth and sample reproducibility of comparable methods like the Protein Aggregation Capture (PAC) protocol, when using a limited sample amount. Furthermore, the One-Tip-PISA method enabled us to prepare and analyze a full 96-well plate of HeLa cells treated with 11 different kinase inhibitors at 2 different concentrations within a single day. The resulting injection-ready Evotips can be analyzed within a subsequent day by fast LC-gradients. The One-Tip-PISA method is promising for higher throughput and lower sample input compared to the existing PISA analysis, with potential applications in a wide range of research areas. 88

OP-54 Enhanced secretomics workflow reveals organ-dependent secreted protein landscape across multiple murine tissues Amanda Smythers1,2, Joao Paulo1, Melanie Mittenbühler1,2, Katherine Blackmore1,2, Sumeet Khetarpal1,2, Bruce Spiegelman1,2, Steven Gygi1,2 1Harvard Medical School, Cell Biology, Boston, MA, United States 2Dana-Farber Cancer Institute, Cancer Biology, Boston, MA, United States Exercise exerts benefits across multiple organ systems and is associated with prevention and treatment of many diseases, including cardiovascular disease, pre-eclampsia, and type 2 diabetes mellitus, among others. Despite this, the identity, origin, and mechanism(s) of action of molecules facilitating these beneficial outcomes are relatively unexplored. The interrogation of the exercise- and organ-dependent secretome is essential to understanding the origin, destination, and therapeutic potential of exerkine-mediated inter-organ signaling. However, comprehensive organ-dependent secretomics faces several analytical challenges, including large dynamic range, blood contamination, and risk of cell burst during preparation. Further, unconventional protein secretion pathways make bioinformatic prediction of all secreted proteins challenging, resulting in poor proteome annotation and requiring robust validation of secreted candidates. We have developed a robust workflow for the interrogation of organ-dependent secretomes from murine tissue. Tissues are dissected within minutes of animal sacrifice and are rapidly rinsed with PBS buffer to remove residual blood. We implement a low-speed filtered centrifugation to remove the extracellular fluid. Fluid is centrifuged to remove cells followed by albumin/IGG depletion, and then further separated via ultra-centrifugation or magnetic beads to distinguish between extracellular vesicles and released proteins. All samples are analyzed with LC- MS/MS and quantified via TMT. We applied this protocol to iWAT, skeletal muscle, lung, and placenta. Both the lung and placenta produce the highest volume of EF of 10 tested murine tissues; both fluids had protein concentrations of >50 ug/uL, producing >700 ug of protein per sample, on average. While the lung is amenable to our secretomics method without modification, the placenta required a slightly modified procedure with extra perfusion. Similarly, the iWAT tissue required a brief fixation in paraformaldehyde. Western blots show that these fluids are distinct from organ lysates and lack markers that would indicate cell breakage and/or other contaminant infiltration. LC-MS/MS analysis revealed >3000 proteins in the extracellular fluid as well as the extracellular vesicles, with unique proteins present in each fraction that reinforce the need for extracellular fluid fractionation. Comparing sedentary and exercised mice revealed lung specific cytokines present only in the lung EF, including the CXL15, an inflammation-induced neutrophil trafficking protein. Additionally, MCK-PGC1alpha transgenic mice show distinctions in secreted proteomes among their muscle, lungs, and placenta, emphasizing the advantage of our methods for the characterization of physiology-dependent signaling. Overall, this optimized secretome isolation procedure has the ability to reveal dynamic changes in organ-dependent signaling that can further enhance therapeutic discovery efforts. 89

OP-55 A complete and automated end-to-end sample preparation strategy for high-throughput and standardized proteomics with high sensitivity Dorte Bekker-Jensen1, Magnus Huusfeldt1, Hope Cabalo Zisco1, Djordje Vasiljevic1, Nicolai Bache1 1Evosep, Odense, Denmark Proteomics is going through a significant paradigm shift and is rapidly evolving towards higher throughput and robust easier-to-use technologies. The use of short LC gradients combined with fast and sensitive mass spectrometers has been a key aspect in this development. The Evosep One is designed for high-throughput and high-volume applications, where standardized methods with increasing robustness enable routine analysis of hundreds of samples per instrument per day. This has moved the bottleneck towards sample preparation, which requires standardized and fully automated end-to-end workflows. We have developed an automated end-to-end workflow for proteomics sample preparation, designed for fast and robust protein digestion capable of processing up to 500 samples on one AssayMap Bravo in a single workday. The workflow results in ready-to-use samples for LC-MS as tryptic peptides are loaded directly onto Evotips in a complete hands-free manner, starting from protein lysate. It is optimized for speed and is completed within just 90 minutes. To illustrate the robustness and reproducibility, 500 samples were prepared in one day and analyzed with the corresponding high-throughput 500 samples per day method in a randomized order. At this level of throughput, the cost per sample is an important consideration. Therefore, we scaled down the use of consumables and designed the workflow to digest only what is needed for one LC-MS injection to minimize enzyme usage. To examine the sensitivity of our sample preparation workflow, we scaled down to even lower starting input materials and observed significant sensitivity down to just 1 ng protein starting amount. 90

OP-56 Novel data-independent acquisition and targeted quantitation on Orbitrap Tribrid instruments, unleashing the power of Hybrid-DIA technology Michael Hoopmann1, Jesse Canterbury2, David Baxter1, Rafael Melani2, Khatereh Motamedchaboki2, Yue Xuan3, Robert L. Moritz1 1Institute for Systems Biology, Seattle, WA, United States 2Thermo Fisher Scientific, San Jose, CA, United States 3Thermo Fisher Scientific, Bremen, Germany For any proteomic analysis using mass spectrometry, the desired outcome defines the methodology of choice. However, each methodology has tradeoffs, and relying on a single approach may be insufficient to the task. For example, rapid chromatography increases sample throughput, but limits the sensitivity and quantitative capabilities of the sample analysis. Tandem mass tagging of peptides multiplexes sample analysis, but requires high resolution and data-dependent methods. Targeted methods focus on a small set of peptides, leaving much of the sample unexplored. To overcome this challenge, sample analysis requires serially applied methodologies, a costly redundancy if sample or instrument availability is limited. Hybrid-DIA MS technology offers a novel solution by intelligently integrating targeted peptide analysis and discovery-based data independent acquisition (DIA), simultaneously maximizing instrument utilization and increasing sample throughput. Hybrid-DIA MS operates through the Thermo Fisher Scientific Moonshot application, so far only available to the Exploris 480 Orbitrap mass spectrometer via the Instrument Application Programming Interface (IAPI). We present Moonshot Pro, a redevelopment of the Moonshot application for the Tribrid series of Orbitrap mass spectrometers, with greatly improved functionality, visual output, and performance of Hybrid-DIA MS technology. The application analyzes DIA mass spectra in real-time to detect peptide transitions of interest. These precursor peptides are then targeted for quantitation on the fly. Targeted scans are fully customizable for resolution, injection time, and ion accumulation to maximize sensitivity. Moonshot Pro combines various targeted methodologies with DIA, leveraging a transition mass inclusion list for peptides of interest. By scanning DIA windows for candidate peptides, Moonshot Pro enhances the sensitivity and specificity alongside survey scanning based on the high-resolution accurate precursor mass. Quantitative PRM scans only occur after observation of the correct transitions for a peptide, allowing DIA to remain largely uninhibited to maximize scan frequency and sample coverage. Moonshot Pro performs SureQuant analysis with addition of heavy synthetic peptides to trigger the targeting of endogenous peptide counterparts. Customizing targeted scan settings for sensitivity allows endogenous peptide quantification even when abundances are below detection in DIA. MSx mode simultaneously fragments synthetic and endogenous peptides, enhancing quantification. Additionally, samples labeled with TMTpro18 and synthetic peptides labeled with Super Heavy TMTpro provide a means to multiplex the analysis and quantitation. This dramatically increases sample throughput by quantifying target peptides from multiple samples in a single discovery-based DIA analysis, enhancing data completeness for targets of interest, with potential for use in proteogenomic and translational research studies. 91

OP-57 Proteomic translation and degradation screen of HECT E3 ligases reveals functional redundancies and exclusivity Bhavesh Parmar1, Christina Schülein-Völk2, Ursula Eilers2, Nikita Verheyden1, Martin Eilers2, Christian Münch1 1University of Frankfurt, Frankfurt, Germany 2University of Würzburg, Wuerzburg, Germany Homologous to E6AP C-terminus (HECT) E3 ubiquitin ligases are crucial for maintaining cellular proteostasis by regulating ubiquitination and subsequent degradation of substrates. These enzymes are integral to various biological processes, including protein trafficking, subcellular localization, and immune responses. Defects in HECT E3 ligase functionality have been implicated in various cancers, neurodegenerative diseases, and viral infections. With the advent of PROteolysis TArgeting Chimaera (PROTAC) technology, HECT E3 ligases have emerged as promising targets for therapeutic intervention. Despite the wide range of pharmacological possibilities, only a subset of the 28 annotated human HECT E3 ligases have been studied in depth. Consequently, much remains unknown about their substrates and regulatory mechanisms. To address this, we performed a high-throughput translation/degradation proteomic screen coupled with siRNA- induced knockdown of 27/28 HECT E3 ligases. Expanding on our previously established multiplexed enhanced protein dynamic (mePROD) approach, we combined an siRNA screen with pulsed stable isotope labelling in cells (pSILAC) and isobaric tagging (TMT) to temporally quantify protein degradation and synthesis for each HECT E3 ligase upon knockdown. We were able to quantify >8,000 proteins/siRNA in both translation and degradation proteoform states from a single well of a 96-well cell culture plate. Moreover, knockdown was confirmed for the majority of siRNA targets in proteomics data, further increasing the confidence in the efficacy of our optimized high-throughput experimental approach. Seventeen out of 27 siRNAs showed no drastic difference in overall protein translation and degradation or cell viability, suggesting a narrow range of downstream modules and functional synergism, with a few unique set of differentially regulated proteins corresponding to each HECT E3 ligase. Ten out of 27 siRNAs severely impacted the overall proteome turnover, with a broad effect on protein translation and degradation, suggesting exclusive dependence on these specific HECT E3 ligases and their crucial role in the regulation of cellular proteostasis. UBR5 has previously been reported to regulate a large number of transcription factors and nuclear proteins. Our data corroborates this, with evidence of reduced protein degradation and cell viability. Similarly, our data confirm known modalities in HECTD2-4, NEDD4, HUWE1 and provides evidence for novel functional substrates with potentially wide therapeutic relevance ranging from cancer, cardiac hypertrophy, inflammation to antidepressants. In conclusion, we provide here a high-throughput approach to study protein synthesis and degradation that can be combined with siRNA, CRISPR and compound screens. Our findings provide a valuable resource for understanding HECT-mediated protein degradation and paves the way for drug design in various therapeutic contexts. 93

OP-58 Divergent proteome tolerance against gain and loss of chromosome arms Yi Di1, Wenxue Li1, Joan Josep Castellano2, Joanna N. Kopko2, Barbora Salovska1, Alison M. Taylor2, Yansheng Liu1 1Yale University School of Medicine, Department of Pharmacology, West Haven, CT, United States 2Columbia University Irving Medical Center, Department of Pathology and Cell Biology, New York, NY, United States Aneuploidy, defined as the abnormal gain or loss of chromosomes or chromosome arms, causes human genetic syndromes and many cancers. For example, in lung squamous cell carcinoma, chromosome 3q is gained in over 60% of tumors, while chromosome 3p loss occurs in 78% of tumors. However, how the cellular proteome tolerates the gain or loss of chromosomes—a fundamental question in aneuploidy research—remains controversial. We and others have previously reported that aneuploid eukaryotic cells enhance the dynamic degradation of unbound protein complex subunits encoded by excessive chromosomes, maintaining protein complex stoichiometry for cellular biochemical reactions. This mechanism was recently speculated to occur in both gain and loss-type aneuploidies. However, this means that loss-type aneuploid cells would have to degrade all proteins not encoded by the aneuploid (deficient) chromosome but participating in the same protein complexes to a much lower level, which is not economic for the proteomic organization. In the present work, using a CRISPR engineering strategy, we generated an isogenic chromosome 3 isogenic aneuploidy model exhibiting either 3p loss or 3q gain in lung epithelial cells, along with euploid controls. We then performed a systems analysis of protein abundance, turnover, and thermal stability using mass spectrometry (MS) technologies, including data-independent acquisition (DIA)-MS quantification, dynamic SILAC or pulse SILAC (pSILAC) labeling with plex-DIA, and cellular thermal shift assays (CETSA), on a series of isogenic cells. Our plex-DIA data, based on nearly 9,000 proteins, remarkably found no changes in protein degradation rate for complex-in or complex-out 3p proteins in 3p loss cells, demonstrating that deficient gene copies due to chromosomal loss do not entail slower protein degradation. Instead, we discovered that protein dosage compensation is tightly associated with a relatively enhanced synthesis rate of 3p-encoded proteins participating in protein complexes. Additionally, the interaction partners of proteins encoded by duplicated 3q or deleted 3p are both pervasively regulated post-transcriptionally (mRNA-protein correlations are barely 0.047~0.13). CETSA results further show that proteins encoded on deleted 3p exhibit higher thermal stability, implying the presence of stronger protein-protein interactions. In stark contrast, thermal stability is not regulated for 3q proteins in 3q gain cells. Finally, to corroborate the tolerance mechanism of selective protein translation in lost-type aneuploidy, we found that cytosolic ribosomal proteins, although downregulated, are dramatically stabilized at the post-translational level through longer protein lifetimes and higher thermal stability in 3p loss cells, suggesting a fine-tuning of the protein synthesis process. In summary, our results demonstrate that cells utilize distinct proteome response mechanisms to tolerate the gain and loss types of aneuploidies. 95

OP-59 Mapping communication dynamics between physically interacting cells using HySic Alexia Carré1, Sofía Ibáñez-Molero2, Joannes Pruijs2, Alisha Atmopawiro2, Fujia Wang1, Alexandra Terry2, Maarten Altelaar1, Daniel Peeper2, Kelly Stecker1 1Utrecht University, Utrecht, Netherlands 2Netherlands Cancer Institute, Amsterdam, Netherlands Summary: Cytotoxic CD8+ T cells engage their targets and form physical connections that drive the release of cytotoxic compounds, thus playing a crucial role in anti-cancer immunity. Yet, our full understanding of the proteins involved in this process remains elusive. To study the mechanisms underlying tumor resistance to T cell attack,ar events, we have developed HySic (hybrid quantification of stable isotope labeling by amino acids in cell culture [SILAC]-labeled interacting cells), a method to quantify short-term protein and phosphorylation dynamics between and within physically-interacting cells (PMID: 38150364). We benchmarked our method by analysing the communication between activated primary CD8+ T-cells and non-small-cell lung cancer cell lines over a time course of 0-6 hours. Method: First, we incorporated heavy SILAC labels into cultured tumor cells. Then, we co-cultured them with primary activated T cells in media containing intermediate SILAC label. All cells were harvested together and we used label-free quantification (LFQ) to quantify protein and phosphorylation dynamics in T-cells (light SILAC), tumor cells (heavy SILAC) and in newly-synthesized proteins (intermediate SILAC). Next, we sought to increase the throughput and sampling depth of our HySic method by replacing LFQ quantification with multiplexed TMT combined to FAIMS. Here, each co-culture timepoint was tagged with a different TMT channel. After sample pooling and fractionation, we employed real-time-searching (RTS) to trigger MS3-based TMT quantification. Using their respective SILAC label, each target population (tumor, T-cell or newly-synthesized proteins) can be enriched with RTS. Results: In total HySic quantified 2742 proteins and 7005 phosphosites in T-cells, and 5469 proteins and 19428 phosphosites from tumors while HySic-TMT led to the quantification of 2.5-times more proteins in 2 times less runtime. Enrichment analysis of T-cell proteome data yielded leukocyte-specific terms, indicating selective quantification of T-cells in a mixed proteome population. We observed specific downregulation of granulin in tumor cells and CD8 in T-cells, which were both validated by additional methods. In our phosphoproteome data, we observed that following engagement, both T-cells and tumors upregulate Rho-GTPase signaling through PAK1 activation. By applying a PAK1 inhibitor, we were able to increase the migration and the production of cytotoxic granzyme-B by T cells, which resulted in increased tumor killing. Significance: Altogether, our results demonstrate the ability of HySic to identify actionable targets that improve immune-mediated tumor clearance. Moreover, implementing TMT-RTS-MS3 in the HySic workflow opens the possibility to quantify large experimental datasets of mixed-cell signaling dynamics with minimal instrumentation time. We demonstrated that our HySic method is an innovative and efficient approach to probe signaling dynamics between interacting cells. 96

OP-60 Top-down characterization of genetically encoded Bcl-XL phosphorylation and phosphomimetics using electron capture dissociation Rachel Franklin1, Cat Hoang Vesely2, Richard B. Cooley2, Joseph Meeuswen1 1Agilent Technologies, Mass Spectrometry Division, Corvallis, OR, United States 2Oregon State University, GCE4ALL center, Corvallis, OR, United States Genetically encoded phosphoserine and non-hydrolyzable phosphonate-containing phosphorylation mimics play a crucial role in studying phosphorylation-dependent processes. Genetic code expansion (GCE) allows the creation of site-specifically modified proteins with non-hydrolyzable and native phosphoserine. The precise engineering of phosphorylated proteins enables the measurement of phosphorylation-specific effects on protein function. Combining GCE with top-down mass spectrometry, enables an in-depth characterization of phospho- proteoforms which informs downstream binding studies. Here, we employ top-down analysis to characterize Bcl- XL, a protein central to cellular apoptosis. Within minutes of data collection, we performed sequence and impurity analysis on Bcl-XL variants containing phosphoserine and phospho mimics. These results underscore the importance of top-down characterization to inform studies on phosphorylation-dependent processes. Top-down analysis of Bcl-xL was performed using an Agilent 6545XT quadrupole time-of-flight (QTOF) mass spectrometer equipped with an ExD cell for electron fragmentation. Bcl-XL variants were expressed and purified by the GCE4All Center using amber codon reassignment to create pure, site-specifically modified Bcl-XL. Intact masses were determined using Mass Hunter BioConfirm. ExDViewer v4.5.11 was used for top-down sequence analysis and localization of modifications. Downstream binding studies were accomplished using isothermal titration calorimetry (ITC). In contrast to traditional bottom-up techniques that demand extensive sample preparation, our approach minimizes these requirements and reveals residue-level differences among samples within a 1-minute data collection period. We isolated the 19+ precursor of intact Bcl-XL proteins (~23kDa) with encoded phosphoserine, non-hydrolyzable phosphoserine, serine, glutamine, or alanine in position 62. Utilizing electron capture dissociation (ECD), we achieved 85-90% sequence coverage for all Bcl-XL variants including site-specific evidence of mutations and modifications. The spectrum included both ECD and collision-induced dissociation (CID) fragmentation ions which contributed to overall sequence coverage. For the wild-type Bcl-xL sample, the fragmentation results yielded 84% coverage considering ECD ions alone and 87% coverage when considering both ECD and CID-type ions. In addition, we show that the 2 Da difference between non-hydrolyzable phosphoserine with a phosphonate group and genuine phosphoserine can be differentiated using top-down electron capture mass spectrometry. The confirmation of modification identity & site via top-down ECD revealed a contaminant in one phosphomimetic sample, leading to re-analysis of downstream binding studies using ITC. Effective analysis of unknown contaminants and precise mapping of modifications demonstrate the utility of pairing genetic code expansion with top-down analysis for phosphoprotein characterization. 97

OP-61 High-throughput, single-cell top-down proteomics using multisegmented spray-capillary CE-MS Si Wu1 1University of Alabama, Tuscaloosa, AL, United States Introduction: Ultrasensitive top-down proteomics techniques provide valuable insights into PTM-regulated cellular functions in mass-limited samples, including single cells. However, top-down proteomics analysis of mass-limited samples and single cells is challenging due to high sample complexity, wide dynamic range, and wide isotopic/charge state distributions of proteoforms. We recently developed "Spray-Capillary", an ESI-assisted device for quantitative ultralow-volume sampling and online CE-MS analysis. In this study, we optimized the spray-capillary capillary device for improved sensitivity for intact protein analysis via PEI coating and reducing capillary inner diameter. We further coupled this platform droplet-based sample preparation to analyze single mammalian cells. Additionally, to improve throughput, we developed a multisegmented spray-capillary injection method that allows simultaneous separation of 10+ samples in a single analysis. Methods: Spray-capillary devices were fabricated as discussed previously (Huang et al. 2019). An ECE-001 CE autosampler (CMP scientific) was used for automation and a customized 3D stage was used for droplet sample injection. Data was collected on a Thermo Orbitrap Exploris 240 or Orbitrap Ascend Tribrid MS. Single cells were obtained from Hui Zhang at Johns Hopkins. Biopharma Finder and TopPIC Suite were utilized for mass feature deconvolution and identification, respectively. Results: We have developed and successfully applied spray-capillary CE-MS for quantitative single-cell metabolomics analysis in single onion cells (Huang et al. 2021). Here, we adapted this platform for automated, high-throughput single-cell top-down proteomics. First, we introduced a polyethyleneimine (PEI) coating to reduce sample loss due to protein adhesion and modulate EOF. The intensity for the coated capillary was 1.81E8 (RSD=17%), which represents a >500-fold increase compared with the bare capillary. Additionally, 20 and 50 µm I.D. PEI-coated spray-capillary devices were evaluated, and the 20 µm capillary resulted in a ~32-fold higher BPE intensity compared with 50 μm. The optimized spray-capillary device was used to analyze 50 pg intact E. coli lysate and >200 proteoforms were detected, representing one of the most sensitive detection methods for top- down proteomics. Additionally, we coupled a nanodroplet-based sample preparation method with this optimized spray-capillary CE-MS platform. We detected 867 and 711 intact proteoforms in ~45 HeLa and OVCAR-8 cells, respectively. We further applied this nanodroplet-based platform to single PC3 cells, detecting 60-80 proteoforms in single cells with high cell-to-cell reproducibility. Furthermore, to improve throughput, a multisegment sample injection method was coupled to our platform to improve throughput. Up to 17 sample segments could be injected and achieve baseline separation in a single spray-capillary CE-MS top-down proteomics analysis. 98

OP-62 Spatially resolved multi-omics to decipher molecular pathology for idiopathic pulmonary fibrosis U-Ming Lim1, Esther Cheow1, Matthew Choo1, Nikhil Tulsian1, Tze Khee Chan1, Brian Henry1, Aaron Zefrin Fernandis1 1MSD International GmbH (Singapore Branch), Quantitative Biosciences, Singapore, Singapore Introduction: Idiopathic pulmonary fibrosis (IPF) is a chronic, progressive lung disease characterized by progressive scarring of lung tissue, resulting in respiratory failure. Despite extensive research, the underlying molecular mechanisms driving IPF progression remain elusive. Here, we highlight the application of multi-omics spatial mass spectrometry tissue imaging coupled with laser capture microdissection (LCM)-LC-MS/MS technologies to identify key drivers of fibrosis progression, enabling target identification and therapeutic invention strategies for IPF. Methods: Sequential MALDI-MSI analysis was conducted on fresh-frozen IPF lung tissue sections for metabolite/lipid analysis and intact protein analysis using HiPlex-IHC (Ambergen) on a timsTOF fleX MALDI-2 (Bruker) instrument. SCiLS lab software (Bruker) was used for data visualization and analysis, while image segmentation algorithms embedded within QuPath and SCiLS lab were subsequently applied to enable the selection of regions of interest (ROI) for downstream analysis. LCM was performed using the LMD7 system (Leica) to isolate these ROIs for in-depth proteomic characterization via LC-MS/MS (timsTOF fleX MALDI-2 coupled with EvoSep). Preliminary data: Existing spatial omics studies in IPF have analyzed single omics layers, limiting their ability to capture the complex interplay between biomolecules. Our innovative approach utilizes sequential multi-omics MALDI imaging, which enables us to concurrently analyze and integrate information from untargeted metabolites and lipids alongside targeted proteins within a single tissue section. Our multi-omics analysis found dysregulation of inflammatory mediators, including increased expression of inflammatory metabolites like lysophosphatidic acid and arachidonic acid. These elevated levels coincide with the spatial distribution of CD68 macrophages, indicating a possible connection between these mediators and macrophage activity in the progression of IPF. Furthermore, our methodology incorporates laser capture microdissection (LCM) for targeted isolation of fibrotic regions, guided by a multi-modal approach. First, we utilize MALDI-IHC targeting α-SMA, an established protein marker for fibrosis, to identify regions of interest. The reliability of MALDI-IHC with α-SMA probes in determining fibrotic regions is then verified using pentachrome staining, a gold standard for characterizing fibrotic foci, ensuring that the most relevant areas for in-depth proteomic characterization via LC-MS/MS were isolated. Preliminary data from this approach indicates dysregulation in IPF-related pathways, including upregulated TGF-β signaling in fibrotic tissue, aligning with its role in ECM synthesis and remodeling. This localized analysis offers a deeper understanding of the molecular composition in fibrotic regions, overcoming limitations of bulk tissue analysis. Fig. 1 (optional) 99

100

OP-63 MoDPA: investigating diseases through modification-dependent protein associations Enrico Massignani1, Kevin Velghe1, Natalia Tichshenko1, Pathmanaban Ramasamy1, Lennart Martens1 1Ghent University, Center for Medical Biotechnology, Department of Biomolecular Medicine, Ghent, Belgium Both protein-protein associations and Post-Translational Modifications (PTMs) are crucial for regulating various biological processes, but the relationship between these two regulatory mechanisms is not fully understood. To bridge this gap, I am currently developing MoDPA (Modification-dependent Protein Associations), a computational approach that aims to identify functional associations between PTMs. Co-expression analysis is a widely utilized method for exploring the function and regulation of genes. Various techniques exist for assessing the relationship between two proteins, with the simplest being based on co- occurrence. My research project seeks to employ a similar approach to identify consistently co-occurring clusters of PTMs across experiments. To obtain PTM identifications, public datasets from the online PRIDE repository were uniformly reprocessed using ionbot, a machine learning-based peptide search engine. To tackle the high dimensionality and sparsity of PTM data, I employed a Variational Autoencoder (VAE) deep- learning model to compress the dataset into a low-dimensional latent space while retaining its essential features. An adjacency matrix was then constructed by calculating correlations between PTMs in the latent space, resulting in a network of co-occurring PTMs. Each node in the network represents a modification on a specific protein residue, and two modifications are connected by an edge if they were consistently co-identified in the analyzed mass spectrometry data. A clustering analysis revealed that proteins within related pathways tend to bear similar PTMs. For instance, proteins involved in phagocytosis display co-occurring Lysine and Arginine methylation events, whereas proteins related to translation displayed co-occurring Lysine acetylation events. Intriguingly, clusters containing a combination of acetylation, phosphorylation and methylation also emerged from the analysis, offering an opportunity to study the cross-talk between these modifications. The PTM association network, generated by MoDPA and annotated with biological information, will be made public and will serve as a source of novel hypotheses for researchers in life sciences. Fig. 1 (optional) 100

101

OP-64 A next-generation covariation map advances functional annotation of the human proteome Georg Kustatscher1 1University of Edinburgh, Wellcome Centre for Cell Biology, Edinburgh, United Kingdom Most research aiming to understand the molecular foundations of life and disease has focused on a limited set of increasingly well-known proteins. By contrast, thousands of human proteins remain "understudied": their biological function is poorly understood and annotation of their molecular properties is scarce. This annotation inequality hinders biomedical progress because mechanistic investigations of gene–disease associations typically focus on proteins that are already well known, a phenomenon also known as the street-light effect (Kustatscher et al, Nat Methods 2022, PMID 35534633). We have previously shown that protein covariation (coexpression) analysis is a powerful proteomics approach to link uncharacterised proteins to known cellular processes (Messner et al, Cell 2023, PMID 37080200; Kustatscher et al, Nat Biotechnol 2019, PMID 31690884). Here, we present a new covariation map of the human proteome. To assemble it, we developed a proteomics data processing pipeline that is fast, scalable, and includes appropriate FDR control for very large datasets. Based on the Fragpipe platform, this pipeline enabled us to re- process 23,000 previously published MS runs in less than two weeks. This resulted in the ProteomeHD.2 dataset, which covers the abundance changes of 16,000 proteins in response to 2,500 biological perturbations, quantified using SILAC labelling. This includes many microproteins for which proteomic evidence had so far been lacking (Kourtis et al, in preparation). To determine which proteins have similar covariation patterns across ProteomeHD.2, and might thus be functionally related, we developed a machine-learning-based strategy. This strategy not only considers whether two proteins have correlated expression but also incorporates additional information from the proteomics data that reflect the robustness and reliability of the underlying protein quantitations. The resulting proteome covariation map reveals functional associations for about 10,000 human proteins. We show that it captures functional associations as well as more established techniques, such as affinity-purification MS experiments, allowing us to make predictions on the potential biological functions of many previously unidentified and understudied human proteins. 101

102

OP-65 PTMeXchange species specific PTM builds: meta-analysis of datasets and dissemination of high-quality PTM data for community use Kerry Ramsbottom1, Ellen Boswell1, Oscar Martin Camacho1, Shireen Al-Momani1, Ananth Prakash2, Yasset Perez Riverol2, Zhi Sun3, Deepti Jaiswal Kundu2, Emily Bowler-Barnett2, Maria Martin2, Jun Fan2, Eric Deutsch3, Juan Vizcaino2, Andrew Jones1 1University of Liverpool, Liverpool, United Kingdom 2EMBL-European Bioinformatics Institute (EMBL-EBI), Cambridge, United Kingdom 3Institute for Systems Biology, Seattle, WA, United States Post-translational modifications (PTMs), of which phosphorylation is the most studied, play an important role in biological functions. Mass spectrometry (MS) and database searching are commonly used to detect and localise modification sites on proteins, with confidence being governed by peptide-spectrum match (PSM) and PTM localisation statistics. The aim of the PTMeXchange project is to re-analyse public enriched PTM datasets, focusing on accurate PTM localisation, integrating data across studies and disseminating the data to UniProtKB, linking it to the original MS evidence in PRIDE and PeptideAtlas in order to make PTM data FAIR (Findable, Accessible, Interoperable and Reusable). Here we demonstrate the re-analysis workflow using publicly available mass spectrometry proteomics datasets. Publicly available mass spectrometry datasets from ProteomeXchange were selected and curated to identify relevant datasets for each investigated species and modification. An open data re-analysis pipeline using the Trans Proteomic Pipeline (TPP) and statistical methods to control PTM false localisation were used. The "PTM build" data from these are then integrated into public databases, including UniProtKB, PRIDE and PeptideAtlas. This enables detailed exploration of scores and visualisation of source mass spectra, as a full evidence trail. Here we demonstrate the pipeline for species specific PTM builds, including data from our first major build, phosphorylation in Asian rice (Oryza sativa). For our first build, Oryza sativa, we identified eight relevant phosphoproteomics datasets. We preformed a simple meta-analysis combining all datasets, and assigning sites to categories based on their PTM localisation scores and occurrences in datasets: Gold-Silver-Bronze. This resulted in identifying 15,565 high-quality phosphosites on serine, threonine and tyrosine residues on rice proteins. We incorporated the use of pAla decoy identifications, enabling validation of false reporting within these categories. Only two pAla hits were observed in the Gold set, with 475 sites across all categories, indicating the overall FLR is very low. We have now generated several phosphorylation builds for a variety of species including Plasmodium falciparum, Saccharomyces cerevisiae, Homo sapiens and Mus musculus. We are also covering other modifications, requiring more complex workflows, including SUMOylation, lysine acetylation and ubiquitination. The re-analysed data for the completed builds have been loaded into UniProt Knowledge-Base - enabling researchers to visualise sites alongside other data on relevant proteins e.g. structural models from AlphaFold2, PeptideAtlas and the PRIDE database - enabling visualisation of source evidence, including scores. We also demonstrate how Universal Spectrum Identifiers (USIs) can be used to visualise mass spectra supporting modification sites and can be a valuable tool to investigate the evidence supporting identified modifications. 102

103

OP-66 A comparison of peptide- and spectrum-centric search engines beyond bar charts Michelle T. Berger1, Johanna Tüshaus2, Alexander Hogrebe1, Florian Seefried1, Bernhard Küster2,3, Daniel Zolg1, Mathias Wilhelm3,4, Martin Frejno1 1MSAID, Garching, Germany 2Technical University of Munich, Chair for Proteomics and Bioanalytics, Freising, Germany 3Technical University Munich, Munich Data Science Institute (MDSI), Garching, Germany 4Technical University of Munich, Computational Mass Spectrometry, Freising, Germany Background: Data-independent acquisition (DIA) offers reproducibility and proteome coverage but yields complex spectra. Various algorithms were developed for peptide identification from DIA data. Peptide-centric engines rely on fragment ion elution curves, while spectrum-centric methods analyze spectra individually. Algorithms aim to separate true from false identifications (IDs) through diverse scores, with q-values controlling false discovery rates (FDR). We compare DIA-NN, Spectronaut, and Chimerys examining their unique statistics and scores. Methods: Publicly available mixed species and single-cell Orbitrap Astral DIA raw files were downloaded from PRIDE. Datasets were searched library-free with DIA-NN 1.8.1, Spectronaut 18.5 and Chimerys 2.0 in Proteome Discoverer 3.1 against normal or entrapment databases. Chimerys deconvolutes spectra, distributing shared fragment ion intensities among co-isolated precursors. Peptides are quantified by aggregating contributions over retention time with mokapot controlling FDR. Peptide and protein FDR is calculated by Proteome Discoverer 3.1. Empirical FDR measurements were conducted via entrapment experiments. Results: Search engine comparisons typically focus on ID counts, yet FDR control and reporting levels differ among DIA-NN, Spectronaut, and Chimerys. Filtering is done at the peptide group (Chimerys) or precursor level (DIA-NN and Spectronaut), which affects the number of reported IDs. We investigated the effect of mismatching the level at which FDR is estimated and at which IDs are counted. We observe that the number of peptide groups artificially increases by ~10% if FDR is controlled at the precursor level, but IDs are counted at the peptide group level. Applying 1% precursor FDR to all search engines reveals that Chimerys detects the same number of precursors as Spectronaut, while DIA-NN seemingly identifies 17% more. The accuracy of FDR estimates significantly impacts the number of IDs. We empirically assessed FDR control using entrapment experiments. Chimerys and Spectronaut maintain precursor counts, while DIA-NN loses ~16% of IDs at 1% empirical FDR. Accurate FDR control equalizes the performance of all search engines in identifying precursors on this dataset. However, identifying spurious IDs from DIA-NN necessitates entrapment analyses. Subsequently, we investigated IDs unique to each specific tool. Unique IDs from each tool often rely on few data points or fragments, and measures of fragment ion interference can inflate ID numbers. In conclusion, close inspection of search engine results is vital for accurate peptide identification and quantification from DIA data. Conclusion: We demonstrate the importance of scrutinizing search engine scores and FDR for accurate peptide identification and quantification from DIA data. 103

104

OP-67 Achieving quantitative accuracy in large-scale DIA proteomics Lukasz Szyrwiel1, Justus L. Grossmann1, Ludwig Roman Sinn1, Vadim Demichev1 1Charité – University Medicine Berlin, Quantitative Proteomics Lab, Berlin, Germany Data-independent acquisition (DIA) proteomics has recently seen rapid growth in popularity as a robust analytical tool in basic and translational research. New mass spectrometry instrumentation has pushed the boundaries of sensitivity and speed, and in tandem with novel acquisition schemes and advances in analysis software enables innovative applications, including spatial and single-cell proteomics. The high diversity of sample types and DIA methods calls for a reliable framework to evaluate and optimise the accuracy of protein quantification and ensure that the quantitative methods scale well to large experiments. In this work, we designed and recorded a large-scale benchmark dataset to thoroughly examine the performance of quantification algorithms in DIA proteomics. The dataset contains a series of multi-species mixes similar to the LFQbench approach, but with the crucial difference that our design introduces a variable background to mimic natural proteome variation in a controlled fashion. The digest mixes were recorded in a high-throughput setting with fast gradients (200 - 500 samples/day) on an Evosep One system coupled to the latest-generation Bruker timsTOF Ultra mass spectrometer using dia-PASEF and slice-PASEF. Injection amounts ranged from 15 ng down to 0.75 ng for the human background matrix. Our dataset thus permits to investigate the effects of injection amount, chromatography method, acquisition scheme and sample heterogeneity on accuracy, precision and reproducibility of quantification. It further allows to ensure that the data processing algorithms used can cope with the challenges posed by large experiment size and sample diversity. We use the new benchmark dataset to validate our machine learning-driven quantification framework QuantUMS (Quantification using an Uncertainty Minimising Solution), integrated in the DIA-NN software. We find that not only ultra-fast chromatographic gradients but also low sample amounts exasperate the inherent signal ratio compression in DIA proteomics. Next, we observe that QuantUMS uniquely alleviates the ratio compression, thus enabling comprehensive quantitative proteomics of low sample amounts at a throughput of hundreds of samples per day. Our results further show that QuantUMS enables reliable quantification irrespective of the dataset size or the presence of acquisitions with sample amounts spanning an order of magnitude within the same experiment, meeting the demands of the recent challenging applications of DIA proteomics. In summary, we present a large-scale and diverse controlled benchmark data set acquired using the latest- generation mass spectrometry platform. This dataset allows to gain insights into the quantitative performance of DIA at the limits of instrument speed and sensitivity, as well as permits to evaluate, optimise and validate the next-generation computational methods for protein quantification. 104

105

OP-68 PTMs and proteoforms shaping tissue proteomes Tine Claeys1, Juan Vizcaino2, Kris Gevaert1, Lennart Martens1 1Ghent University, Center for Medical Biotechnology (CMB), Ghent, Belgium 2EMBL – EBI, Proteomics, Cambridge, United Kingdom Tissue-specific characteristics are influenced by numerous factors, with proteins playing a pivotal role. Alterations in PTMs can modify the tissue-specific functionality of proteins, but this has been largely explored in a restricted set of proteins and PTMs. New open modification search engines have expanded the possibilities for studying PTMs from a more unbiased perspective. In previous research(1), we trained a machine-learning model on public datasets of healthy human tissue, revealing tissue-specific protein patterns. Applying this model to diseased data demonstrated its effectiveness in biomarker detection, aided by explainable AI highlighting the proteins contributing to classification. To evaluate the tissue-predictive capability of PTMs in a similar manner, we reprocessed data from the human draft map of the proteome (2) with the ionbot open modification search engine. This data was utilized in three ways. First, a model was trained on only the protein information, achieving a high F1-score of 71%. Secondly, we added the PTM information to the protein data, resulting in a more nuanced view and a 4% improvement of the F1-score. Lastly, only the PTM information was used for training which resulted in a surprising F1-score of 60%. To further test the PTM classification capabilities, dataset (3) was used for evaluation. This resulted in an accurate classification of 65% of tissues, confirming the hypothesis that tissues exhibit specific PTM patterns. Explainable AI methods revealed the contribution of PTMs to each tissue, showing strong PTM patterns in metabolically active tissues like the liver, brain, T-cells, and monocytes, driven notably by glycosylation patterns. This observation was further confirmed when studying the tissue-specificity of proteoforms of widespread proteins using the same (2) dataset. We selected proteoforms from proteins identified in 75% of tissues, such as actin, filamin, glyceraldehyde-3-phosphate dehydrogenase, etc. Using this dataset, containing a minimal amount of information with 384 proteoforms from 26 proteins, the model achieved an F1-score of 38% across 16 tissues, effectively classifying highly metabolic tissues but performing poor on other tissues. In conclusion, our research highlights the potential of combining open modification searches with machine learning and explainable AI to study tissue-specific PTMs and proteoforms. This approach not only enhances our understanding of tissue-specific proteomes but also opens new avenues for biomarker discovery and disease characterization. The identification of distinct PTM patterns in metabolically active tissues underscores the importance of PTMs, providing a clear direction for future investigations into their specific roles. 1. Claeys, T., et. al. J Proteome Res (2023). 2. Kim, M.-S. et al. Nature (2014). 3. Wang, D. et al. Mol Syst Biol. (2019) 105

106

OP-69 quantms: a cloud-based pipeline for quantitative proteomics enables the reanalysis of public proteomics data Chengxin Dai1, Julianus Pfeuffer2, Hong Wang3, Ping Zheng3, Lukas Käll4, Timo Sachsenberg5, Vadim Demichev6, Mingze Bai3, Oliver Kohlbacher5, Yasset Perez Riverol7 1Beijing Proteome Research Center, State Key Laboratory of Proteomics, Beijing, China 2Freie Universität Berlin, Algorithmic Bioinformatics, Berlin, Germany 3Chongqing University of Posts and Telecommunications, Chongqing Key Laboratory of Big Data for Bio Intelligence, Chongqing, China 4School of Engineering Sciences in Chemistry, Biotechnology and Health, KTH Royal Institute of Technology, Science for Life Laboratory, Stockholm, Sweden 5University of Tübingen, Institute for Bioinformatics and Medical Informatics, Tuebingen, Germany 6Universitätsmedizin Berlin, Charité, Berlin, Germany 7European Bioinformatics Institute (EMBL-EBI), Proteomics Services, Cambridgeshire, United Kingdom Background: The increasing volume of public proteomics data presents significant computational challenges for large-scale reanalysis. Addressing this, we introduce quantms (https://quantms.org/), an open-source, cloud- based pipeline designed for massively parallel proteomics data analysis. The quantms.org resource includes results for 16,599 proteins, with 16,270 quantified in normal tissues, 11,374 in cell lines, and 4,993 in human plasma. Methods: Our workflow for proteomics data analysis begins with parsing input files and supporting peptide identification through multiple search engines like SAGE, Comet, and MSGF+, integrating tools such as ConsensusID for scoring and protein inference, and supporting post-translational modifications with LuciPHOr2. For label-free quantification (LFQ), we developed proteomicsLFQ, which employs both spectral counting and intensity-based methods. At the same time, for isobaric labelling, quantms uses IsobaricAnalyzer for reporter ion normalisation and benchmarking against gold-standard datasets. Data-independent acquisition (DIA) analysis is facilitated by parallelising the DIA-NN tool across compute nodes, ensuring robust identification and quantification even at varying protein concentrations. The downstream analysis integrates MSstats for differential expression and pmultiqc for quality control, with all tools available as BioConda packages and BioContainers, ensuring compatibility with various computational infrastructures and seamless submission to PRIDE and ProteomeXchange. Results: Using quantms, we reanalyzed 83 public ProteomeXchange datasets (Figure), which included 29,354 instrument files from 13,132 human samples. The pipeline quantified 16,599 proteins based on 1.03 million unique peptides. quantms demonstrated superior performance and scalability compared to traditional tools like MaxQuant. For datasets exceeding 1,000 instrument files, quantms performed up to 40 times faster. Benchmarking against MaxQuant showed that quantms could quantify more proteins with similar accuracy, although it underestimated fold changes at low concentrations. In specific reanalyses, quantms identified more differentially expressed proteins and unique peptides across various tissues and conditions. Additionally, it successfully processed 118 human datasets, highlighting its robustness and efficiency. Conclusions: quantms addresses major bottlenecks in the field of proteomics by enabling automated, large-scale quantitative analysis in cloud and high-performance computing environments. Its modular, open-source nature allows for continuous updates and improves the reproducibility and portability of proteomics data analysis, making it a valuable resource for the proteomics community. Future developments will focus on expanding the repository of reanalyzed datasets and integrating protein expression profiles with other omics data to facilitate advanced biological insights. integration of new tools and workflows. quantms 106

108

OP-70 Integration of genetics and proteomics reveals circulating TIMP2 role in cognitive performance Federica Anastasi1,2,3, Patricia Genius1,2,3, Blanca Rodriguez-Fernandez1,2,3, Chengran Yang4,5, Jigyasha Timsina4,5, Priyanka Gorijala4,5, Felipe Hernández-Villamizar1,3, Luigi Lorenzini6, Marta Del Campo1,3,7, Armand G. Escalante1,3, Gonzalo Sánchez-Benavides1,3,8, Carolina Minguillon1,3,8, Manel Esteller8,9, Arcadi Navarro2,3,10,11, Carlos Cruchaga4,5, Marc Suárez-Calvet1,3,12, Natalia Vilor-Tejedor1,2,3,13 1Hospital del Mar Research Institute, Barcelona, Spain 2Centre for Genomic Regulation (CRG), Barcelona Institute of Science and Technology (BIST), Barcelona, Spain 3BarcelonaBeta Brain Research Center, Fluid Biomarkers and Translational Neurology, Barcelona, Spain 4Washington University, Department of Psychiatry, St. Louis, MO, United States 5Washington University, NeuroGenomics and Informatics. Washington Univer, St. Louis, MO, United States 6Amsterdam University Medical Center, Department of Radiology and Nuclear Medicine, Amsterdam Neuroscience, Amsterdam, Netherlands 7Universidad San Pablo-CEU, CEU Universities, Departamento de Ciencias Farmacéuticas y de la Salud, Facultad de Farmacia, Madrid, Spain 8Instituto de Salud Carlos III, Centro de Investigación Biomédica en Red de Fragilidad y Envejecimiento Saludable (CIBERFES), Madrid, Spain 9Josep Carreras Leukaemia Research Institute (IJC), Badalona, Spain 10Institució Catalana de Recerca i Estudis Avançats (ICREA), Barcelona, Spain 11Institute of Evolutionary Biology (CSIC-UPF), Barcelona, Spain 12Servei de Neurologia, Hospital del Mar, Barcelona, Spain 13Radboud University Medical Center, Department of Genetics, Nijmegen, Netherlands Murine studies have identified blood proteins influencing mice cognitive function, yet translating these findings to humans remains challenging. We report an innovative bioinformatic approach to investigate proteins' potential effects on the human brain. We generated genetic proxies to predict plasma protein levels through a clumping-thresholding approach, and validated their robustness in two independent cohorts. We then assessed their association with cognitive performance in humans. First, we identified circulating proteins in mice affecting brain aging through a systematic review. We retrieved protein quantitative trait loci (pQTLs) summary statistics associated with these proteins in human plasma from the Fenland study (Pietzner et al., 2021). Clumping-thresholding polygenic scores (protPRSs) were computed as proxies for plasma protein levels in 1,380 cognitively unimpaired (CU) individuals from the Knight-ADRC cohort and 410 CU individuals from the ALFA+ study. We validated the predictive ability of the protPRSs by analyzing their association with plasma protein levels measured by Somalogic. As a proof of concept, we assessed associations of genetically predicted protein levels with cognitive performance using linear models adjusted for age, sex, and education in ALFA+ study participants. Stratified models by sex, APOE-ε4 carriership, and Aβ status were also assessed. We annotated significant pQTLs for biological significance through enrichment analysis. Gene-level tissue specificity was evaluated by integrating differential gene expression data from GTEx v8. We identified 12 circulating proteins affecting mice brain aging. Most computed protPRS significantly predicted plasma protein levels across the two independent cohorts: Knight-ADRC (10/13) and ALFA+ (7/13). In ALFA+, a significant association was found between TIMP2-protPRS and better cognitive performance, measured by Preclinical Alzheimer’s Cognitive Composite (PACC) and episodic memory composite (EM). Associations of TIMP2-protPRS with PACC remained significant in stratified models. The pQTLs incorporated into the TIMP2- protPRS were located on chromosome 3, potentially crucial for regulating TIMP2 protein levels. Notably, significant variants were annotated within the TLR9, SMIM4, STAB1, and RTF1 genes. Integrating genetic data with expression profiles across 54 tissues revealed tissue-dependent up-regulation of genes associated with TIMP2 pQTLs, particularly in brain regions. This innovative approach integrating genetic and proteomics data, overcomes translational challenges from animal studies and provides a complementary framework for assessing the association of circulating proteins with brain performance in humans. We demonstrated that genetically predicted plasma TIMP2 levels, known for its rejuvenating effect on mice brains, are linked to better cognitive performance in humans, identifying TIMP2 as a potential therapeutic target for age-related brain diseases. 108

110

OP-71 Cross-platform clinical proteomics using the Charité Open Standard for Plasma Proteomics (OSPP) Ziyue Wang1, Vadim Farztdinov2, Ludwig Roman Sinn1, Johannes Hartl1, Michael Mülleder2, Markus Ralser1 1Charité – Universitätsmedizin Berlin, Institute of Biochemistry, Berlin, Germany 2Charité – Universitätsmedizin Berlin, High Throughput Mass Spectrometry Core Facility (HT-MS), Berlin, Germany Introduction: Human plasma is a preferred body fluid for clinical investigations due to its minimally invasive collection and rich physiological information. LC-MS-based plasma proteomics holds immense promise in characterising human disease, identifying biomarkers, and advancing diagnostic technologies. However, for widespread clinical adoption, there is an ongoing need to improve proteomic workflows in terms of accuracy, reproducibility, platform transferability, and cost-effectiveness. Objective: To develop a peptide panel based on a previously developed targeted panel for COVID-19. The new panel, Charité Open Peptide Standard for Plasma Proteomics (OSPP), consists of 211 stable-isotope labeled peptides from 131 consistently quantified plasma proteins, it is designed for high-throughput targeted and untargeted plasma and serum proteomics studies in large clinical cohorts. Methods: We utilised targeted plasma proteomics MRM (Agilent 6495C), MRM-HR (ZenoTOF 7600) and discovery proteomics techniques on various LC (Waters M Class, Agilent 1290, Thermo Vanquish Neo) and MS platforms (ZenoTOF, TimsTOF, and Exploris 480) to comprehensively test the peptide panel. Results: The targeted peptide panel that was previously established for COVID-19 demonstrated exceptional abilities in categorising the disease, predicting outcomes, and comparing across different platforms. It also has the potential to be applied to comparable types of diseases (Fig 1). By analysing more than 10,000 plasma samples from various disease categories, we have increased the number of peptides in OSPP to 211, derived from 131 plasma proteins that were consistently quantified. These peptides have shown consistent quantification properties in human studies, across different platforms and matrices, and are suitable for chemical synthesis (Fig 2). The OSPP measures proteins that are crucial for human disease, which are derived from proteins involved in many biological processes, including those commonly employed in clinical tests or targeted by FDA-approved medications . In an acute COVID-19 patient cohort (Fig 3), we used the OSPP to i) achieve patient classification and biomarker identification, ii) generate comparable quantitative proteome data with both targeted and untargeted approaches, and iii) estimate absolute peptide quantities for cross-platform alignment across various MS methods. The OSPP is cost-effective, adding only less than 1€ per sample. OSPP also successful implementation in a real-world cohort comprising over 15,000 samples underscores the scalability and practical utility of our peptide panel in addressing critical translational medicine challenges. Conclusion: The OSPP offers high analytical consistency, cost-effectiveness, and versatility, facilitating adaptable and accessible high-throughput plasma proteomics in large clinical cohorts, addressing critical needs in translational medicine. 110

112

OP-72 Proteome profiling with precision on a clinical cohort: Pioneering a large-scale ultra-high throughput validation study employing a novel high-speed hybrid nominal mass platform Qin Fu1, Philip Remes2, Jihyeon Lee1,2, Cristina Jacob2, Koen Raedschelders1,2, Ali Haghani1,2, Sandy Joung1,2, Susan Cheng1,2, Dalin Li3, Scott Peterman2, Dermot McGovern3, Jennifer Van Eyk1,2 1Cedars Sinai Medical Center, Heart Institute, Los Angeles, CA, United States 2Thermo Fisher Scientific, San Jose, CA, United States 3Cedars Sinai Medical Center, Los Angeles, CA, United States Protein and peptide quantification methods, such as Multiple Reaction Monitoring (MRM) and Parallel Reaction Monitoring (PRM), utilize LC-MS for the targeted, bottom-up quantification of protein biomarkers by monitoring signature peptides representative of candidate biomarkers. The challenges are balancing throughput with assay/study variables, peptide parameters that define precision, ease for establishing quantifiability (linearity, sensitivity, reproducibly) and data acquisition across large cohort. The recent development of a novel high-speed hybrid nominal mass platform has enabled us to perform large-scale, high-throughput proteomic profiling of 83 Stable Isotope Labeled (SIL) peptides representing 53 Health Surveillance Panel (HSP) proteins (containing 21 FDA-approved proteins assay) in an inflammatory bowel disease (IBD) cohort (n=498) and age/sex-matched healthy individuals (n=547), totaling 1,005 plasma samples. Using a pooled human plasma sample (100 healthy males and 100 healthy females), we evaluated the targeted peptides' sensitivity, linearity, LLOD, LLOQ, and reproducibility across four different throughput levels: 100 samples per day (SPD), 144 SPD, 180 SPD, and 300 SPD. Excellent reproducibility was observed at all four throughputs, with median CV% ranging from 5% to 10% for all 83 SILs. The median points per peak were 10 (100 SPD), 11 (144 SPD), 8 (180 SPD), and 5 (300 SPD). The observed sensitivity for the 83 SIL peptides across the four throughputs was sub-fmol (LOQ < 0.5 fmol), with 93% of the peptides achieving this for 100 SPD, 87% for 144 SPD, 48% for 180 SPD, and 53% for 300 SPD. For LLOQ (>5 fmol), the percentages were 99% for 100 SPD, 97% for 144 SPD, 49% for 180 SPD, and 72% for 300 SPD. For linearity, more than 95% of all peptides had an R² > 0.9. The total median CV% based on injections each day for 5 days for the 83 SIL peptides was less than 10% at all throughputs. The implementation of the Adaptive Retention Time (ART) algorithm, which allows real-time retention time adjustments, proved advantageous in achieving robustness and enhancing reproducibility when acquiring the IBD cohort of 1,005 samples within 14 days (180 SPD). The study included 2,000 injections when accounting for system suitability tests, low, mid, and high-quality controls, washes, and blanks, performed robustly and reproducibly. In conclusion, the combination of a novel high-speed hybrid nominal mass platform and the ART algorithm provides a powerful solution to the challenges of large-scale, high-throughput, targeted, and scheduled MS-based quantification by increasing the ease to establishing quantifiability (establishing linearity, sensitivity, reproducibly) and ease for consistency data acquisition across large cohort is highlighted. This approach allows us to profile a large IBD cohort (n=1,005) with precision and robustness, offering insights into biologically important plasma proteins, including 21 FDA- approved plasma protein markers. 112

113

OP-73 Breaking barriers to proteomics at scale: the nELISA platform profiles hundreds of circulating proteins in high throughput and at low cost Nathaniel Robichaud1, Grant Ongo1, Kevin Hernandez1, Kiran Edwardson1, JiaMin Huang1, Jeffrey Munzar1, Milad Dagher1 1Nomic Bio, Montreal, Canada Despite the great promises of proteomics, the low abundance of many signalling proteins in circulation, such as cytokines and metabolic mediators, renders mass spectrometry-based analysis challenging. Targeted approaches using affinity binders provide more sensitive quantification of these proteins, but have been challenging to multiplex due to cross-reactivity between reagents (rCR) when scaling content beyond ~25-plex. Recent advances using the Proximity Extension Assay have helped avoid issues with rCR, but high cost and low throughput have limited their widespread adoption for large-scale applications. The nELISA is a high-throughput, highly multiplexed immunoassay platform capable of profiling 1536 samples per instrument per day at a fraction of the cost of other platforms. This is achieved by miniaturizing the sandwich immunoassay, whereby antibody pairs are pre-packaged at the surface of color-coded microparticles that can be readout by high-throughput flow cytometry. We recently demonstrated the nELISA"s ability to profile 10,000 cell culture supernatants in 1 week to reveal immune phenotypes from PBMCs; here, we describe its use in profiling circulating proteins to identify biomarkers of disease, and compare our findings to xMAP (Luminex) and PEA (Olink) platforms. We profiled circulating protein abundance in 44 plasma samples across 4 diseases (type 2 diabetes, colorectal cancer, COVID19, rheumatoid arthritis) and healthy controls using 3 platforms, nELISA (191-plex), PEA (92-plex), and xMAP (48-plex). For targets shared across platforms, nELISA and PEA measurements were generally well- correlated, with a median Spearman correlation of 0.71. In line with this, both platforms distinguished healthy and disease phenotypes, and identified individual protein markers with differing expression patterns between healthy and disease states. In contrast, correlations were significantly lower when comparing PEA (0.53) or nELISA (0.45) to xMAP, purportedly due to well-known cross-reactivity issues with the latter. We then performed a follow up study on an independent set of 96 samples, which we profiled with an expanded nELISA panel (275-plex), as well as a 384-plex PEA panel (Olink Explore). Unsurprisingly, the expanded content of both platforms resulted in the identification of an additional set of biomarkers not available in the initial screen. Among proteins quantified in both sample sets, approximately 40% of biomarkers were validated using the nELISA, but fewer markers were validated by PEA: this may be due to differences in sensitivity between different PEA panels, and/or the use of different reagents in the different panels. In contrast, nELISA reagents and platform sensitivity were maintained regardless of plex level. Of note, the cost of profiling was 15X more expensive using PEA than nELISA. Thus, the nELISA provides a new, high-throughput and cost-effective means of profiling circulating proteins for large-scale studies. Fig. 1 (optional) 113

114

OP-74 Multi-omics analysis of human skeletal muscle in response to acute exercise: initial insights from MoTrPAC Hasmik Keshishian1, Christopher A. Jin2, Patrick Hart1, Gregory R. Smith3, Gayatri Iyer4, James Sanford5, Scott Trappe6, Natalie Clarck1, Paul M. Coen7, Bret H. Goodpaster7, Anna Thalacker-Mercer8, Samuel Montalvo Hernandez2, Daniel H. Katz2, Charles F. Burant4, Josh Adkins5, Steven A. Carr1, MoTrPAC Study Group9 1Broad Institute of MIT and Harvard, Proteomics Platform, Cambridge, MA, United States 2Stanford University School of Medicine, Palo Alto, CA, United States 3Icahn School of Medicine at Mount Sinai, New York, NY, United States 4University of Michigan, Ann Arbor, MI, United States 5Pacific Northwest National Laboratory, Richland, WA, United States 6Ball State University, Muncie, IN, United States 7AdventHealth Translational Research Institute, Orlando, FL, United States 8The University of Alabama at Birmingham, Birmingham, AL, United States 9MoTrPAC Study Group, Cambridge, MA The Molecular Transducers of Physical Exercise Consortium (MoTrPAC) was established to deeply and systematically characterize the molecular basis of exercise through integrative, multi-omics analysis of tissues obtained from rats and human participants1. For the clinical study, sedentary participants were recruited and randomized into endurance exercise (EE), resistance exercise (RE) and control (CON) groups. Vastus lateralis muscle biopsies were collected before a sub maximal acute bout of either EE or RE and at one of several time points after following the acute bout (30min, 3.5h, and 24h). Tissue collection was performed in the CON group at the same time points, but without exercise. Muscle tissue samples were analyzed by ATAC-seq, transcriptomics, proteomics, phosphoproteomics and metabolomics, with the goal of identifying dynamic molecular changes in response to an acute bout of EE or RE including divergent molecular responses between the two modes of exercise. Data QC and differential analysis for all omes were performed using limma and variancePartition packages in R2, respectively. For differential analysis, a linear mixed model was used with participant as the random effect, timepoint-exercise intervention as an interaction term and sex, age, clinical site, and BMI as covariates. Exercise intervention groups (EE and RE) were compared to matched timepoint CON values to identify significantly changed features. Initial multi-omics analysis of an acute bout of exercise showed divergent behavior of different omes throughout the time course. Maximal changes in ATACseq, phosphoproteome and metabolome occur earlier (30min) than changes in the transcriptome (3.5h) and proteome (3.5h and 24h). Furthermore, the proportion of genes with annotated differentially accessible regions (DARs) based on ATACseq that are also differentially regulated genes (DEGs) in the RNAseq data overlap significantly, especially at the later time points (3.5h and 24h), suggesting a time delay between chromatin remodeling and gene expression. Additionally, PTM signature enrichment analysis, using PTM-SEA3 of the phosphoproteome data, shows enrichment of many kinase signatures, more pronounced after RE than EE. These include several MAPK kinase signatures following an acute bout of either exercise mode, consistent with prior findings4. Some overlap of omic dynamics is observed between changes in EE and RE; however, there is a considerable difference in magnitude of changes between modes of exercise, with RE resulting in a greater number of changed features. These MoTrPAC data provide initial insights into the multi-omic exercise responses in human skeletal muscle highlighting a dynamic and varied time course across the omes with common and specific signatures between EE and RE. 114

115

OP-75 The HDL proteome represents a novel, sensitive biomarker of metabolic inflammation and metabolic health in obesity that can track both disease progression and regression Aleksandra Dudzik1,2,3, Rachel Byrne1,2,3, Andrew Hogan4, Donal O’Shea5, James Gibney6, Juan Valderas7, Camilo Boza7, Stephen Pennington1,3, Carel Le Roux1,2,3, Fiona McGillicuddy1,2,3 1UCD Conway Institute, Dublin, Ireland 2UCD Diabetes Complications Research Centre, Dublin, Ireland 3University College Dublin, School of Medicine, Dublin, Ireland 4Kathleen Lonsdale Institute, Maynooth University, Department of Biology, Maynooth, Ireland 5St. Vincent’s University Hospital, Obesity Immunology Group, Education and Research Centre, Dublin, Ireland 6Robert Graves Institute of Endocrinology, TUH, Dublin, Ireland 7Pontificia Universidad Catolica De Chile, Faculty of Medicine Department of Nutrition, Diabetes and Metabolism, Santiago, Chile Introduction: Low-grade metabolic inflammation is causally implicated in the development of multiple obesity- related complications, such as cardiovascular disease, chronic kidney disease (CKD) and liver disease. Our previous preclinical studies in mice have shown that changes in the proteome of high-density lipoprotein (HDL) particles in obesity occur independently of changes in HDL-cholesterol. The HDL proteome in obesity has increased pro-inflammatory acute phase proteins and reduced antioxidant/anti-inflammatory proteins. In the current study we hypothesized that the HDL proteome may serve as a novel biomarker of metabolic inflammation in human obesity, and that a reversal in 'adverse' particle protein composition would be evident after weight-loss. Methods: HDL particles were isolated from serum using a novel two-step extraction and purification method followed by tryptic digestion for LC-MS/MS. Proteins associated with HDL were identified by data-dependent acquisition (DDA) mass spectrometry using a timsTOF Pro (Bruker Daltonics) coupled with an EvoSep One system (EvoSep BioSystems). MS data was analysed using MaxQuant (v2.0.3.0) and Perseus (v2.0.3.1) software. HDL proteome composition was determined in pair-matched individuals ± obesity and in a separate cohort of patients with obesity and CKD at baseline and 1 year post gastric bypass (n=20), sleeve gastrectomy (n=19), or liraglutide±SGLT2 inhibitors treatment (n=18). An algorithm to determine inflammatory profile of the HDL particle (HDL inflammatory index score) was generated, using 20 proteins identified in obesity setting. Results: LC-MS/MS analysis of HDL protein composition identified a total 194 proteins in ≥70% of the HDL samples. There was a significant (p≤0.05) change in 64 proteins with increase in inflammatory-related proteins and loss of antioxidant/anti-inflammatory proteins in HDL isolated from obese individuals. Surgically induced weight-loss reversed the pro-inflammatory phenotype of HDL particles with changes in 55 proteins, including reductions in Complement factor I (CFI), Complement factor B (CFB), Vitronectin (VTN) and increases in Serum paraoxonase 3 (PON3), Apolipoprotein D (APOD). In contrast, intervention with liraglutide had little impact on the HDL proteome. An HDL-inflammatory index score significantly (p≤0.0001) delineated the non-obese and obese groups. Furthermore, significant (p≤0.0001) recovery of the HDL inflammatory index was evident upon surgically induced weight-loss. Conclusions: The HDL proteome exhibits a pro-inflammatory phenotype with deteriorating metabolic health independent of static measurements of HDL-C. The adverse effects of obesity on the HDL proteome are reversible with surgical-induced weight-loss. The data presented here suggest that the HDL proteome composition and an associated inflammatory index score can potentially be used to monitor metabolic health and identify patients at highest risk of cardiometabolic complications 115

116

OP-76 High-throughput and deep plasma proteomics allows the classification of cancer types Arthur Viodé1, Sandra Schär1, Christopher Below1, Lukas Reiter1, Roland Bruderer1 1Biognosys AG, Schlieren, Switzerland Plasma proteomics is challenging with the 22 most abundant proteins constituting more than 99% of the total protein content, limiting the detection of less abundant proteins using discovery mass spectrometry–based proteomics. To address this limitation, workflows based on protein corona formation have emerged as a strong method to dig deeper into the proteome. Plasma proteins upon exposure to particles forms a layer around it, known as protein corona, leading to a reduction in dynamic range and enabling of detection of lesser abundant proteins. Here, we introduce an enrichment workflow that allows for robust, quantitative and deep enrichment of low-abundant plasma proteins, termed P2 Plasma Enrichment System. We applied this workflow to a cohort of 62 plasma samples from 4 cancers and 36 healthy patients to discover classification biomarkers. We developed and optimized the P2 Plasma Enrichment System starting from 100µL of plasma followed by tryptic digestion. All samples were analyzed using a 17min analytical gradient and data independent acquisition on a timsTOF HT (Bruker). Data analysis was performed using Spectronaut 19 (Biognosys) using directDIA. From a reference pool plasma, we achieved a proteome depth of 6324 proteins. The enrichment factor is up to 10x the one of neat plasma and 2x compared to an optimized enrichment based on MagReSyn SAX-beads (Mag-Net protocol). Our experimental conditions allow for robust protein corona formation leading to high workflow reproducibility with coefficient of variation on protein level of 6.7% in median, 5498 proteins with a CV <20%. To evaluate quantitative performance of our workflow we performed a controlled quantitative experiment (CQE) based on human plasma spiked with known amounts of chicken plasma (1:9). The mixture of plasma of different species shows the conservation of relative protein quantities, demonstrates its suitable for quantitative plasma proteomics. We applied the enrichment workflow to 62 plasma samples from 4 cancers type (Breast n= 15; Colon Cancer n=19, Lung n=13 and Prostate n=15) and 36 healthy patients. We identified cumulatively 6979 protein groups in the study and 5498 protein groups on average per sample with a gradient of 17min. We probed the 4 cancer cohort for classification of the plasma specimen into the cancer types and healthy. Therefore, an analysis pipeline was developed consisting of an ANOVA analysis with filtering by adjusted p-value <0.05. Next, correlating features were removed (R2 < 0.7), finally a s-PLSDA model was trained on 70% of the data (with 10-fold cross validation). The model was tested for classification accuracy on the 30% hold out data and was found to be 92% on average for 10 repeats of the analysis pipeline. Taken together, the optimized plasma proteomic workflow showcased a powerful way to profile plasma biomarkers. The ease of automating and scaling up such an approach could enable a broader application to other indications and biofluids. 116

117

OP-77 CAN-IMMUNE: a comprehensive platform for cancer-specific immunopeptidomics-based neoantigen mapping and evaluation Chen Li1,2, Sanjay SG Krishna1,2, Mohammad Shahbazy1,2, Gabriel Goncalves1,2, Pouya Faridi1,2,3, Anthony Purcell1,2 1Monash University, Department of Biochemistry and Molecular Biology, Melbourne, Australia 2Monash University, Biomedicine Discovery Institute, Melbourne, Australia 3Hudson Institute of Medical Research, Centre for Cancer Research, Clayton, Australia Neoantigens are formed due to the somatic mutations that arise during carcinogenesis; they give rise to unique neopeptides bearing cancer-specific mutations that are presented on the surface of the tumour in complex with human leukocyte antigen (HLA) molecules for scrutiny by T cell recognition. A subset of these neopeptides can be recognised by tumoricidal T cells. As such, the neoepitopes derived from these neoantigens can serve as targets for cancer vaccines and antigen-based immunotherapies. Liquid Chromatography-Mass Spectrometry (LC-MS) is the technology of choice for identifying HLA-presented peptides, generally through spectral matching to inferred peptide sequences generated from a reference proteome database. However, the reference proteome database only contains canonical wild-type protein/peptide sequences and lacks cancer-specific mutations, hindering the identification of cancer-specific neoantigens using conventional database searching approaches. To tackle this challenge, we developed a comprehensive platform, CAN-IMMUNE, for MS-based identification of cancer-specific neopeptides. To construct a cancer-specific mutant peptide library, we collected missense mutations identified by whole exome sequencing (WES) from various sources, including the Catalogue of Somatic Mutations in Cancer (COSMIC), the Cancer Cell Line Encyclopedia (CCLE), and published literature. To date, CAN-IMMUNE documents 854,813 mutant peptides across 33 cancer types from 1428 cancer cell lines and 52 tumour tissues. CAN-IMMUNE allows users to browse, search, and download bespoke mutant peptide libraries for their cancer of interest. In addition, CAN-IMMUNE provides customisable source codes for users to generate their mutant peptide libraries using WES data. To assess the utility of CAN-IMMUNE, we searched our raw spectra data of MDA-MB-231 (a triple-negative breast cancer cell line), HCT116 (a colorectal cancer cell line), and various melanoma cell lines using both PEAKS 11 Online and deep learning-rescoring-assisted MSFragger database search engines. The results demonstrate that our constructed databases successfully identified potential cancer-specific neoantigens that cannot be found using the canonical proteome library, providing promising neoantigen candidates for further immunogenicity assessment. We anticipate that CAN- IMMUNE can accelerate the identification of cancer neoantigens via its WES-based mutant peptide libraries, thereby better informing the development of targeted cancer immunotherapies. 117

118

OP-78 The proteomic and immunopeptidomic landscape of non-small cell lung cancer Adrianna Seredynska1,2, Ruth Fiestas Cueto1,3, Lina Goncharenko1, Antonio Eduardo Cabrita Figueiredo1, Konrad Kurowski1, Patrick Westermann4, Peter Bronsert1, Martin Werner1, Christoph Messner4, Katja Bärenfaller4, Oliver Schilling1,2,3 1Institute for Surgical Pathology, Faculty of Medicine, Medical Center, University of Freiburg, Freiburg, Germany 2German Cancer Consortium (DKTK) and Cancer Research Center (DKFZ), Heidelberg, Germany 3ProtPath Research Training Group, Freiburg, Germany 4Swiss Institute of Allergy and Asthma Research (SIAF), University of Zurich, Davos, Switzerland Introduction: Non-small cell lung carcinoma (NSCLC) is the most common type of lung cancer, with adenocarcinoma (LUAD) and squamous cell carcinoma (LUSC) being the most prevailing histological subtypes with distinct biological signatures. Correct classification and accurate molecular profiling are essential for effective therapeutic guidance. In this research, we performed an in-depth mass spectrometry (MS)-based proteomic study of a large patient cohort representing the two aforementioned subtypes of NSCLC. This MS- based profiling of LUAD and LUSC revealed proteomic alterations that might determine their clinical behavior. High-coverage proteome analysis opens the possibility to unravel the mechanisms of immune evasion in NSCLC. Moreover, additional immunopeptidomics study should elucidate distinct tumor-associated antigens, that are pivotal for the development of epitope-specific cancer immunotherapies. Methods: In our work, the patient-derived samples were processed with an optimized, semi-automated single- pot, solid-phase enhanced sample preparation (SP3-beads) workflow utilizing a Bravo pipetting platform. Thereby, extracted proteins were digested and the resulting peptides were analyzed via liquid-chromatography tandem-mass spectrometry (LC-MS/MS). For the immunopeptidome study, we established and optimized a workflow to enrich HLA-I and HLA-II-bound antigens. This workflow involves automated affinity purification with crosslinked monoclonal antibodies, performed on a AssayMAP Bravo platform. This adaptable method enables processing of fresh frozen tissue from biopsies, among a variety of sample types. Results: The MS-analysis of the patient cohort (consisting of LUAD (n = 83), LUSC (n = 70), and tumor-adjacent (n = 138)) resulted in high proteome coverage of a total of over 10000 unique proteins. Initial statistical analysis shows significant differences between LUAD and LUSC and suggests a possible distinction based on patient's gender. In our pilot study of the immunopeptidome, we successfully identified > 4000 peptides after HLA class I enrichment and > 7500 peptides after HLA class II enrichment. As for the patient cohort, the initial analysis of HLA-I enriched samples revealed a total of 40k unique peptides acquired in data independet acquisition. This coverage enables a robust analysis of the immunopeptidome, providing valuable insights into antigen presentation. Furthermore, a de novo sequencing analysis of the HLA-I peptides was performed. Outlook: Deep proteomic coverage of a large patient cohort enables us to gain deeper insight into the fundamental molecular biology of LUAD and LUSC. The further immunopeptidomic investigation gives a promising outlook to reveal a subset of tumor-associated antigens unique to each NSCLC subtype, offering precise targets for epitope-specific cancer immunotherapies. This study should facilitate the development of personalized immunotherapies tailored to each patient's tumor profile. 118

120

OP-79 Immunopeptidomics shows how deleterious knockouts in the HLA class I antigen processing and presentation machinery induce distinct changes in the antigen landscape Ilja Shapiro1, Clelia Maschke1, Antonius de Waard2, Tamara Verkerk2, Robert Spaapen3, Michal Bassani- Sternberg1 1CHUV, Department of Oncology, Lausanne, Switzerland 2University of Amsterdam, Amsterdam, Netherlands 3Sanquin Research, Amsterdam, Netherlands Antigen presentation is an indispensable trigger for T-cell-based immune responses where T-cell receptors recognize HLA-peptide complexes on the cell surface. In the case of HLA class I, cells constantly sample the cytoplasmic proteome for short peptides to present them as antigens. For antigen presentation, an entire network of proteins act in concert to uphold a molecular machinery that delivers a flow of HLA-peptide complexes to the cell surface to potentially bind T-cell receptors. Functionally impairing mutations of the antigen processing and presentation machinery (APPM) are a frequently observed immune evasion mechanism in cancer. Our aim is to understand how deleterious knock out of key players in the APPM affects the antigenic landscape, and, consequentially, potential to stimulate T-cells. HAP1 is a near-haploid leukemia-derived cell line model expressing highly frequent HLA class I alleles among the European population, and, as such, a suitable model to generate and study the immunopeptidome of APPM knock-outs. In total, a series of 11 knock-outs have been previously generated (IRF2, B2M, GANAB, CALR, CALX, TAPBP, PDIA3, ERAP1, TAP1, TAP2, SPPL3), including genes involved in modifying, chaperoning, translocating, and loading the HLA molecules with peptides. In order to study antigen presentation, we investigated each knock-out"s immunopeptidome with our high- throughput 96-well plate-based immunopeptide enrichment workflow, in triplicates and we further measured the respective proteomes. Data was acquired in DDA and DIA on a HF-X Q Exactive. With help of the FragPipe search engine environment and using chimeric spectra to find PSMs with a database search, we accumulated a library of more than 22"000 unique immunopeptides, mapping to more than 8"000 unique proteins, for a subsequent library search of the DIA data (FDR 1%). Our data shows varying levels of impact on the overall ability to present antigens, ranging from complete elimination of antigen presentation capability, like in the case of B2M, to sampling depths resembling those of the wildtype controls as in CALR. We determined knock-out specific effects that are directly linked to the biological function, e.g., knocking-out ERAP shifted the peptide length distribution to longer peptides. In addition, due to the quantitative quality of DIA library searches, we also determined an HLA allele-dependent shift in antigen presentation levels in different knock-outs, directly linking the dependencies of HLA-alleles on APPM components regarding the regulation of their presentation. Proteomics data provided insights on the efficiency of sampling source proteins for presentation. We demonstrate that impairments in the APPM result in distinct features in the antigenic landscape. This research enhances our understanding of the roles played by various APPM actors in shaping the immunopeptidome and could lead to improved HLA binding predictions that account for defects in the machinery. 120

121

OP-80 Proteomic characterisation of rare quiescent therapy-resistant cancer cells Emily Gaizley1, Xiuyuan Chen1, Amandeep Bhamra1, Bowie Lam1, Anni Poysti1, Jason Wray1, Simona Parrinello1, Silvia Surinova1, Tariq Enver1 1UCL Cancer Institute, London, United Kingdom Many cancers are characterised by the presence of a rare subpopulation of quiescent stem-like cells which are a source of heterogeneity, phenotypic therapy resistance, relapse and further disease evolution. These cells are able to evade cytotoxic chemotherapy, which selectively targets proliferating cells, thus representing a major therapeutic challenge. The mechanisms controlling quiescence remain largely unknown, and transcriptomic analysis of quiescent cells has yielded only limited insights. Proteomic analysis of rare cancer stem cells requires the development of a low-input methodology in order to capture global protein changes and differentially expressed post-translational modifications. To identify and isolate rare quiescent cells, we generated and validated a fluorescent label-retaining in vitro model using patient-derived glioblastoma cell lines and used somatic mouse models with the same inducible reporter. Using these model systems, we have demonstrated the presence of a pre-existing quiescent population and shown that these cells exhibit self-renewal capacity and increased therapy resistance. We have developed a robust proteomic workflow for analysing rare FACS-sorted cells, using multiplexed isobaric labelling and phosphopeptide enrichment. This method enables the processing of protein loads down to the low-microgram/sub-microgram level to identify signalling networks that govern quiescence. Our findings indicate that the quiescent proteome is conserved across models of glioblastoma, regardless of driver mutations. Network modelling with causal integration of multi-omic data has led to the identification of potential mechanisms of resistance and targetable vulnerabilities. Functional assays indicate PML as a potential therapeutic target for overcoming quiescence mediated therapy resistance in glioblastoma. We propose extracellular interferon signalling initiates upregulation of PML leading to suppression of mTORC1-mediated translation and ultimately entry to cellular quiescence. By bridging the gap in knowledge between transcriptomic regulation and protein-driven phenotypes, we aim to contribute to the understanding of phenotypic therapy resistance and provide new strategies for modulating quiescence. 121

122

OP-81 A proteomic atlas of the human monocytic compartment Kyra van der Pan1, Indu Khatri1,2, Enrique De La Rosa3,4, Sara Kassem1, Anniek L de Jager1, Brigitta AE Naber1, Inge F de Laat1, Alesha Louis1, Marjolijn Hameetman5, YK Onno Teng6, Meta Roestenberg7, Stijn Crobach8, Wouter BL van den Bossche9, Daniela Damasceno10,11,3, José Manuel Sánchez-Santos12, Alberto Orfao10,11,3, Javier de las Rivas3,4, Jacques JM van Dongen1,10,11,3, Cristina Teodosio1,10,11,3, Paula Díez13,1,10,11,3,14 1Leiden University Medical Center, Department of Immunology, Leiden, Netherlands 2Leiden University Medical Center, Leiden Computational Biology Center, Leiden, Netherlands 3Institute of Biomedical Research of Salamanca (IBSAL), Salamanca, Spain 4Cancer Research Center, Bioinformatics Group, Salamanca, Spain 5Leiden University Medical Center, Flow cytometry Core Facility, Leiden, Netherlands 6Leiden University Medical Center, Department of Nephrology, Leiden, Netherlands 7Leiden University Medical Center, Department of Parasitology, Leiden, Netherlands 8Leiden University Medical Center, Department of Pathology, Leiden, Netherlands 9Erasmus University Medical Center, Department of Neurosurgery, Erasmus, Netherlands 10Cancer Research Center, Translational and Clinical Research Program, Salamanca, Spain 11University of Salamanca, Department of Medicine, Salamanca, Spain 12University of Salamanca, Department of Statistics, Salamanca, Spain 13University of Oviedo, Department of Functional Biology, Oviedo, Spain 14University of Oviedo, Department of Physical and Analytical Chemistry, Oviedo, Spain Monocytes (Mo) and macrophages (MAC) are key players in tissue-cleaning processes during body homeostasis and response to stressors (e.g. infection, cancer). Mo derive from precursor cells in the bone marrow (BM) and differentiate into MAC after leaving the bloodstream and entering tissues. Despite the existing knowledge about these cells, their plasticity, functional heterogeneity, and the discovery of new subsets, currently hamper the definition of their exact maturational relationship. To address this, we combined single cell-based mass cytometry (CyTOF) and bulk mass spectrometry (MS)-based proteomics approaches to define the proteome and maturational relationship of monocytic cells across tissues (BM, peripheral blood (PB), skin, peritoneal dialysate (PD) and colon). Firstly, >100 proteins were screened to design a 35-marker CyTOF panel to evaluate >15 Mo/MAC populations. Based on this data, we purified 5 (pro)monocyte subsets from BM (n=4-5, from 9 donors), 8 monocytic subsets from PB (n=2-5, from 7 donors), and MAC from skin (n=4, from 7 donors), PD (n=3, from 6 donors) and colon (n=5, from 7 donors) samples, for TMT-based MS. Based on population relationships, as defined using single-cell trajectory analysis, 11 distinct protein expression patterns were observed during maturation in BM, reflecting diverse functions (e.g. metabolism, signalling, lysosome assembly, antigen presentation) at different stages of differentiation. Analysis of PB subsets suggested a close relationship between CD62L- FcεRI- classical Mo (cMo), intermediate Mo (iMo) and CD36+ Slan- non- classical Mo (ncMo) subsets, with iMo acting as a bridge between cMo and ncMo. Within ncMo subpopulations, a distinct functional profile distinguished Slan- cells, primarily involved in metabolic and protein production processes, from CD36- subsets, showing a stronger commitment to protein production and vesicle transport. Regarding MAC, a common core functional identity signature encompassing 425 functions (79.6% of all MAC functions) was identified across all tissues. A microenvironment effect was also observed, as colon MAC exhibited distinctive TOR and TLR signalling pathways, skin MAC expressed proteins involved in cellular responses to gamma radiation, and PD MAC reflected processes like immunological synapse formation. Notably, when compared to Mo, CD62L+ FcεRI- cMo from PB displayed the closest resemblance to the general MAC signature, despite differences in their immune profiles. While MAC were prominently involved in antiviral responses, the cMo subset played a main role in T-cell activation, receptor signalling, cytokine production and antigen processing and presentation processes. In conclusion, by integrating CyTOF and MS-based proteomics approaches, we provide a detailed proteome atlas of monocytes and MAC across tissues, shedding light on the complex maturational relationships and functional heterogeneity of these cells. 122

123

OP-82 Developing a miniturboid mass spectrometry-based method for characterizing aggregate proteomics in metabolically distressed cells Allen Teng1,2, Jonathan St. Germain3,4, Uros Kuzmanov1,2, Brian Raught3,4, Anthony Gramolini1,2 1University of Toronto, Physiology, Toronto, Canada 2University of Toronto, Ted Rogers Centre for Heart Research, Toronto, Canada 3University Health Network, Princess Margaret Cancer Centre, Toronto, Canada 4University of Toronto, Medical Biophysics, Toronto, Canada Protein aggresomes are a hallmark of impaired proteostasis, but the assembly and detailed cellular function of this biomolecular condensate remain largely unknown. Until recently, systematic methods for identifying aggresome proteins were either unavailable or incomplete. Past efforts including centrifugation or gel filtration could fractionate aggregate proteins in vitro, but they fail to capture proteins in situ or in vivo. To address this limitation, we introduce AggresomeID as a novel approach for screening aggresome proteins in live cells. Autophagy adaptor protein sequestome1 (SQSTM1) is visibly enriched in aggresomes in response to various pathophysiological stresses. Here, we genetically fused human SQSTM1 with the biotin ligase miniTurboID to create a recombinant protein (an aggresome sensor). With this fusion construct, we engineered a stable HEK293 cell line. Fluorescent microscopy showed that aggresome sensors were cytosolic in DMSO-treated cells but coalesced with the aggresome markers ubiquitin and HDAC6 following 2 µM MG132 (a proteasome inhibitor) treatment for 18 hours. Proximity-based biotinylation of aggresome proteins was enriched using immobilized streptavidin followed by SAINT bioinformatics. SAINT analyses identified over 250 proteins that were significantly enriched in aggresomes (Log2 enrichment > 2 between the groups, MG-132/DMSO; Bayesian false discovery rate cut-off < 1%). Interestingly, several clusters of these enriched proteins appear to be functionally relevant for protein degradation, including autophagy receptors (NDP52, TAX1BP1, TOLLIP, and NBR1), autophagosome biogenesis and assembly (ATG7, ATG101), proteasome 20s subunits (PSMA1, PSMA5, PSMB1, PSMB2), and valosin-containing protein complex subunits (VCP/p97, UFD1, UBXD1). We also confirmed the enrichment of these proteins in aggresomes by flag-tagged fusion constructs expressed in cells, with MG132 treatment followed by confocal microscopy and immunoblots. Ongoing work include building a AggresomeID toolkit in this model system, as well as explore additional cell types. In conclusion, our study is the first report to address the technical limitation in studying aggresome proteomics and our approach enables the continuous monitoring of aggresome proteomics in live cells. Ongoing studies will focus on the characterization of aggresome subtypes in different lines of cells following pathological stresses (e.g., heat shock, oxidative stress, and proteasome inhibition) and establish a timeline (and possibly mechanisms) through which candidate proteins participate in aggresome biogenesis. 123

124

OP-83 Protein restriction reprograms the multi-organ proteomic landscape of mouse aging Tian Lu1,2,3, Yuting Xie1,2,3, Xiling Lin4, Xue Cai1,2,3, Chang Su5, Wanglong Gou1, Pengfei Shan4, Ju-Sheng Zheng1, Huijun Wang5, Yi Zhu1,2,3, Tiannan Guo1,2,3 1Westlake University, School of Medicine, Hangzhou, China 2Westlake Laboratory of Life Sciences and Biomedicine, Westlake Center for Intelligent Proteomics, Hangzhou, China 3Westlake University, Research Center for Industries of the Future, School of Life Sciences,, Hangzhou, China 4The Second Affiliated Hospital of Zhejiang University School of Medicine, Department of Endocrinology, Hangzhou, China 5Chinese Center for Disease Control and Prevention, National Institute for Nutrition and Health, Beijing, China Aging is a major risk for life-threatening diseases. In recent years, protein restriction (PR) has gained recognition as a viable and practical intervention for promoting healthy aging. However, considering the complexity of aging and the importance of assessing the effectiveness and safety of PR prior to implementation, it is essential to delve into the molecular alterations occurring in different organs. While several extensive transcriptomics studies on aging covering five to 23 organs have been conducted, the need for proteomics studies remains crucial due to the limited correlation between transcripts and proteins. However, the existing proteomic investigations have primarily focused on no more than ten organs at just two time points, excluding vital organs including the adrenal glands, lymph nodes, and reproductive organs. Additionally, the two-time-point approach may introduce biases by confounding the effects of aging with those of diseases. Moreover, the impact of PR on multiple organs and the optimal timeframe for intervention remain largely unexplored. To address these gaps, we conducted an extensive study involving the collection of 41 mouse organs at eight distinct time points spanning the mouse lifespan and implemented a dietary intervention with varying protein intake for different age and sex groups. We also obtained plasma samples from both mice and humans to investigate the systemic effects of aging and PR. We applied Pulse Data-Independent Acquisition (PulseDIA) for analyzing all 2,565 tissue samples and Tandem Mass Tag (TMT) for the 592 depleted plasma samples. We observed tissue-specific expression patterns of aging hallmarks and identified 36 proteins that consistently changed across multiple organs, with immunoglobulins and serpins demonstrating particular significance. Notably, many of these changes started prior to maturation. PR mitigate age-related molecular changes in a tissue-specific manner, including epigenomic modifications in the urinary system and protein phosphorylation alterations in brown adipose tissue, brain, and seminal vesicle. Our findings were supported by Reduced- representation bisulfite sequencing, phosphoproteomics, and staining experiments. These molecular changes indicated improved health conditions under PR, such as reduced cancer risks in the bladder and enhanced adipose function through increased thermogenesis. Importantly, initiating PR during middle age yielded the most profound effects. Furthermore, our analysis of human and mouse plasma proteomic data indicated improved cardiovascular health under PR, although extreme low protein intake showed signs of increased inflammation. In summary, our study provides the most comprehensive molecular insights to date regarding aging and PR effects in multiple organs. This work expands our understanding of PR as a geroprotective intervention, highlights the potential benefits and risks, and identifies the optimal time frame for PR interventions. 124

126

OP-84 Reversing antifungal resistance through disruption of ClpX Michael Woods1, Arianne Bermas1, Brianna Ball1, Benjamin Muselius1, Davier Gutierrez-Gongora1, Kerry Woolnough1, Norris Chan1, Jason McAlister1, Stephan Sieber2, Jennifer Geddes-McAlister1 1University of Guelph, Molecular and Cellular Biology, Guelph, Canada 2Technical University Munich School of Natural Sciences, Department Biosciences, Munich, Germany Fungal disease impacts the lives of almost a billion people across the globe. The opportunistic human fungal pathogen, Cryptococcus neoformans, causes cryptococcal meningitis in immunocompromised individuals with high fatality rates in response to limited treatment options. Moreover, the emergence of azole-resistant isolates in the clinic following prolonged treatment regimes, environmental fungicide exposure, and fungal evolution, threatens the outcome of current therapeutic options, endangering the survival of infected individuals. By quantitatively characterizing the proteomes of fluconazole-susceptible and -resistant C. neoformans strains using state-of-the-art tandem mass spectrometry, we defined ClpX, an ATP-dependent unfoldase, as a target to overcome resistance. We discovered that disruption of ClpX through deletion or inhibition re-introduces fluconazole susceptibility into the resistant strains, rendering treatment effective once again. We further explored the mechanism of susceptibility and determined interruption to heme biosynthesis and ergosterol production associated with ClpX. Our results contribute to the understanding of novel mechanisms driving fluconazole resistance and provide support for targeting proteins as a therapeutic strategy to combat resistance. 126

127

OP-85 PECA, proteome integral solubility alteration for chemical safety, a tool for target identification, and exposome impact analysis within the One Health framework Ana Carrasco del Amor1, Rebecca Fonseca-Pilzecker1, Mariana Plakhotniuk1, Susana Cristobal1 1Linköping University, Biomedical and Clinical Sciences, Linköping, Sweden The EU's zero pollution ambition for 2050 is to reduce pollution to levels no longer harmful to human and environmental health. Current research has predominantly focused on human health, leaving a significant gap in understanding how these exposures affect animal health. This knowledge gap has consequences, as animal health is intricately linked to human health through the One Health concept. The lack of high-throughput methods to analyse the effects of the exposome on animal health not only endangers animal welfare but could pose risks to food safety and, consequently, human health. As we present here, the Proteome Integral Solubility Alteration for Chemical Safety (PECA) implements the PISA assay for a fast and sensitive chemical safety assessment within the One Health concept. The modification aims: i) constrain the proteome of analysis on the globular native proteome; ii) increase the completeness of the target analysis by combining different principles that induce alteration in protein stability: temperature, acid or salt concentration; iii) select a temperature, acid, salt concentration that increases target identification while reducing sample requirement, iv) introduce lessons learnt from single-cell proteomics to reduce steps, time, and sample loss in sample preparation. Here, we showed that PECA, as a simplified PISA assay, can identify proteins interacting with environmental chemicals, providing insights into the mechanism of action and predicting potential adverse effects. We first applied the method to determine protein targets of chemicals from human cell lines from the liver, skin, and endothelium, which are relevant to predicting the impact. Finally, we confirmed the method's applicability by analysing farm animal tissues. PECA aligns with the One Health concept by examining the broad impact of chemicals across different organisms, which is crucial for safeguarding the health of all species and ensuring the integrity of our food supply. Funding: This work has been performed with funding from Cristobal Lab from the ERA-NET Marine Biotechnology project CYANOBESITY, which FORMAS, Sweden grant nr, cofounds. 2016-02004. The project GOLIATH has received funding from the European Union's Horizon 2020 research and innovation program under grant agreement No 825489; IKERBASQUE, Basque Foundation for Science; Basque Government Research Grant IT-476-22 and LiU verification grant. 127

128

OP-86 High-quality peptide evidence for translated non-canonical open reading frames presented by HLA-I Leron Kok1,2, Eric Deutsch3, Michal Bassani-Sternberg4,5,6, Jyoti Choudhary7, Ivo Fierro-Monti8, Robert L. Moritz3, Jonathan Mudge8, John Prensner9,10, Jorge Ruiz-Orera11, Nicola Ternette12,13,14, Juan Antonia8, Sebastiaan van Heesch1,2 1Princess Máxima Center, Utrecht, Netherlands 2Oncode Institute, Utrecht, Netherlands 3Institute for Systems Biology, Seattle, WA, United States 4University of Lausanne, Ludwig Institute for Cancer Research, Lausanne, Switzerland 5Centre Hospitalier Universitaire Vaudois (CHUV), Department of Oncology, Lausanne, Switzerland 6Agora Cancer Research Centre, Lausanne, Switzerland 7The Institute of Cancer Research, Functional Proteomics group, London, United Kingdom 8European Bioinformatics Institute (EMBL-EBI), European Molecular Biology Laboratory, Cambridge, United Kingdom 9University of Michigan Medical School, Department of Pediatrics, Ann Arbor, MI, United States 10University of Michigan Medical School, Department of Biological Chemistry, Ann Arbor, MI, United States 11Max Delbruck Center for Molecular Medicine in the Helmholtz Association (MDC), Cardiovascular and Metabolic Sciences, Berlin, Germany 12University of Oxford, The Jenner Institute, Oxford, United Kingdom 13University of Oxford, Nuffield Department of Medicine, Oxford, United Kingdom 14University of Utrecht, Department of Pharmaceutical Sciences, Utrecht, Netherlands We and others have recently demonstrated the widespread translation of thousands of short non-canonical open reading frames (ncORFs). Translation of ncORFs is frequent in cancer and indicative of protein production, which could drastically expand the proteome. This "dark proteome" has gained interest as a source of new tumor-specific antigens for immunotherapeutic targeting. However, most ncORFs are currently not supported by high-confidence protein-level evidence. To address this, we established an international consortium with experts from PeptideAtlas, HUPO-HIPP, GENCODE, and the Ribo-seq ORF Consortium. We reanalyzed 3.8 billion raw spectra of 414 different studies comprising 95,520 MS runs using the PeptideAtlas workflow for the presence of 7,264 ncORFs. Focusing on a subset of 118 immunopeptidomics datasets comprising 240M spectra from 10,095 MS runs, we could confirm the peptide level presence of 1,785 out of 7,264 ncORFs. Of these ncORFs, 39% was supported by at least two unique HLA peptides, and 348 could be qualified as Tier 1B "Presented" protein candidates according to our previously proposed non-canonical protein annotation guidelines (Prensner et al., MCP 2023). We made several observations based on our analyses. First, we demonstrate that the detected ncORF peptides show strong concordance (95%) with in silico HLA binding predictions and are mostly processed from ncORF-derived microproteins with a high isoelectric point. Second, we noticed that certain HLA alleles presented more ncORF peptides than others. We could link this directly to the enrichment of certain amino acids in ncORFs that precisely matched the anchor residues of specific HLA alleles. This could indicate that cancer patients carrying such HLA alleles might be more susceptible to future immunotherapeutic strategies targeting antigens derived from ncORFs. Our analyses provide reference-annotation quality and manually curated evidence for the HLA presentation of ncORFs, providing critical insights that can aid in advancing immunotherapy development against this class of antigens. 128

129

OP-87 The pluto problem in protein annotation Jonathan Mudge1, Robert L. Moritz2, Sandra Orchard1, Elspeth Bruford3 1EMBL – EBI, Cambridge, United Kingdom 2Institute for Systems Biology, Seattle, WA, United States 3University of Cambridge, Cambridge, United Kingdom The human transcriptome undergoes extensive translation beyond the portion annotated as protein-coding. Especially, thousands of "non-canonical" translations have been identified using Ribo-seq, hundreds of which now have conventional proteomics or immunopeptidomics support. Should reference annotation projects such as Ensembl-GENCODE, UniProtKB and HGNC add these to the protein "canon"? Here, we see a fascinating parallel with astronomy. In 2006, following lively debate, it was considered that Pluto is unusual compared to the other eight planets, and one of many similarly small objects now observed in the solar system. Thus, after 75 years as a canonical planet, it was reclassified as a "dwarf planet". The vast majority of these newly detected non-canonical translations are also unorthodox; especially, the vast bulk are very small, and lack obvious signatures of protein evolution. Furthermore, many are thus far detected only in cancer samples or cell lines, suggesting that some might be aberrant molecules; others are transposon-derived. Pertinently, we observe that these characteristics are shared by reasonable numbers of canonical proteins, such as HMHB1 and MYEOV. How should annotation projects proceed into this space? We aim to start a similar discussion in our community, noting that decisions taken will have profound implications. In fact, just as the astronomists were led to reconsider the question of what actually is a planet, it may be necessary to rethink what we mean by a protein. Is it simply a molecule built from amino acids that can be detected in the cell? Or should the term carry a deeper meaning, invoking knowledge of actual physiology? Since most non-canonical translations remain functionally obscure, the question is then how to annotate when such knowledge is unavailable. At present, our projects are reluctant to annotate these non-canonical translations as proteins en masse; this would significantly inflate the catalog of protein-coding genes, while functional uncertainties would pass on as ambiguities to our users. However, it would be a mistake not to annotate them at all. We emphasise that numerous non-canonical translations have now turned out to encode bona fide functional microproteins, and we assume that more will be elucidated once studied in more detail. Meanwhile, a standardised annotation of aberrant translation products could be a boon in the study of disease; they are potential diagnostic markers or even therapeutic targets. Thus, we favour the dwarf planet option, whereby non-canonical translation products are classified as something else, while existing canonical proteins with such characteristics are reclassified. This would demand the creation of a new system of protein annotation and nomenclature, and a profound expansion and reorganisation of our catalogs. We will put proposals forward for discussion. 129

130

OP-88 Customized proteomics workflow enables discovery of a microprotein restoring drug sensitivity in cancer Qian Zhao1 1Hong Kong Polytechnic Univeristy, Hong Kong, Hong Kong Microproteins, encoded by small open reading frames (sORFs) have been historically overlooked and are rapidly reshaping our understanding of the proteome. However, large-scale, and confident identification of these microproteins has been technically challenging, compounded by the largely unknown functions of these entities. To address this issue, we have customized sample preparation and data analysis workflows and identified functionally important microproteins in cancer. First, we compared more than ten methods and reported that size-exclusion chromatography (SEC) could simultaneously enrich and fractionate microproteins from complex proteomes, enabling the discovery of more AltProts with higher overall intensities. Next, after a comprehensive comparison of mass spectrometry methods and software, we designed a Data- Independent Acquisition (DIA) workflow that leverages a fragmentation spectra predictor for the efficient construction of DIA libraries for microproteins. Our method achieved a two-fold increase in the identification of microproteins and a 50% reduction in missing values compared to traditional methods. Using this optimized approach, we detected and quantified over a thousand microproteins from hepatocellular carcinoma (HCC) cells. Among them, a novel 54-amino acid microprotein named SEP6, encoded by a lncRNA, was significantly down-regulated in drug-resistant HCC. In both in-situ cell assays and in-vivo animal tests, upregulation of SEP6 restored the sensitivity of HCC and lung cancer cells to multiple drug treatments, including Lenvatinib and regorafenib. Conversely, the knockdown of SEP6 induced drug resistance. Mechanistically, we found that SEP6 interacted with a phosphate, increasing the degradation of g-pg and subsequently enhancing the intracellular concentration of drug molecules. Our work not only provides a robust workflow from sample preparation to mass spectrometry analysis but also highlights the transformative potential of microprotein research. By uncovering the role of SEP6 in drug sensitivity, we demonstrate the critical impact of microproteins on cancer treatment, paving the way for new therapeutic strategies and enhancing our understanding of the proteome. Reference: [1] Yang Y, Wang HW, Zhang YL, Chen L, Chen GL, Bao ZS, Yang Y, Xie Z, Zhao Q*. Molecular & Cellular Proteomics, 2022, 22, 100480 [2] Zhang YL, Chen L, Wang HW, Xie Z, Zhao Q*. Unpublished [3] Chen L, Yang Y, Zhang YL, Wong C, Lee KW, Zhao Q*. Unpublished 130

132

Poster presentations of HUPO & EuPA initiatives (iP-01 to iP-13) iP-01 π-HuB: the proteomic navigator of the human body Fuchu He1, Ruedi Aebersold2, π-HuB Consortium1,3 1National Center for Protein Sciences (Beijing), Beijing, China 2ETH Zürich, Institute of Molecular Systems Biology, Zurich, Switzerland 3International Academy of Phronesis Medicine, Guangdong, China The human body contains trillions of cells, classified in specific cell types, diverse morphologies and functions. Additionally, cells of the same type can assume different states within an individuals" body during their lifetime e.g. by reversibly or irreversibly adjusting to changing conditions. Understanding the complexities of the proteome in the context of a human organism and its many potential states is an inevitable requirement to understand human biology, but these complexities can neither be predicted from the genome, nor have they been systematically measurable with available technologies. Recent advances in proteomic technology and computational sciences now provide opportunities to investigate the intricate biology of the human body at unprecedented resolution and scale. Here, we introduce an international big-science endeavor called π-HuB (Proteomic Navigator of the Human Body). The π-HuB project is a 30-year, large-scale, international high technology and life science research program supported by a substantial multibillion China Renminbi (RMB) investment. The π-HuB project has three central goals: 1) to build proteomic maps of the human body based on anatomy-based proteomic spaces at single-cell or cell-type resolution; 2) to develop the Meta Homo Sapiens model for tracing lineages of proteomic states in lifetime at the population scale; 3) to build a computation- driven, iteratively designed "navigator" for the human body and its states, with the ultimate goals to increase understanding of human biology, to facilitate disease risk assessment and to devise early disease diagnosis, optimal therapeutic interventions and intelligent healthcare. To achieve these goals, the π-HuB project is structured into six pillars: i) human biospecimen isolation and processing, ii) measurement technology deployment and innovation, iii) computational technology deployment and innovation, iv) establishment of a collaborative Big-Science infrastructure, v) sharing of open resources, and vi) establishment of international research teams. During the initiation and development phase (2024-2033), the π-HuB project will assemble an international consortium focused on achieving three major outcomes, specifically, 1) establish principles of cell type-based tissue organization, 2) establish proteomics-driven health guidelines, and 3) generalize proteomics- driven precision medicine. These initial goals will provide the π-HuB project with significant momentum and establish modes of operation and governance of the consortium. Moving forward, the π-HuB project will continue to catalyze global collaboration to integrate results from of a worldwide community of multidisciplinary scientists into a comprehensive, widely accessible resource. We anticipate that the π-HuB project will make major contributions to the understanding of biological systems, biomedical research and medical care for decades to come. Fig. 1 (optional) 132

133

iP-02 Canadian artificial intelligence and mass spectrometry consortium (CAN-AIMS) Jennifer Geddes-McAlister1, Arnaud Droit2 1University of Guelph, Molecular and Cellular Biology, Guelph, Canada 2Laval University CHU – Quebec, Research center, Quebec, Canada Disease spans diverse demographics and negatively impacts human health, affecting each individual in a specific manner. The ability to diagnose, monitor, and treat individuals in a strategic and personalized manner is limited due to lack of disease knowledge and discrepancies in accessibility of healthcare. To overcome these limitations, an increased understanding of the causes, regulatory mechanisms, and treatment options for diseases are needed. Importantly, the identification of proteins, metabolites, and pathways responsible for disease presents a critical starting point. However, the integration of datasets across biological networks and platforms is challenging to ensure comprehensive and robust analyses. Herein, we introduce the Canadian Artificial Intelligence and Mass Spectrometry Consortium (CAN-AIMS), which brings together researchers from across Canada with diverse expertise in human disease, proteomics, computation, and bioethics. The goals of CAN- AIMS are three-fold, to: i) explore innovative research strategies from discovery to translation, ii) develop a hands-on training platform for the next generation of scientists, and iii) build capacity in leading-edge instrumentation and computational recourses for Canada. Together, CAN-AIMS provides the first cohesive group of researchers working collectively to define and mitigate disease within Canada using a combination of mass spectrometry-driven technologies and computational platforms for integration of disease knowledge to improve diagnostics and treatments. 133

134

iP-03 EUPA Initiatives Franck Vandermoere1, Maria Hernandez–Valladares2, Oxana Trifononva3, Richard Unwin4, Shabaz Mohammed5, Viviana Greco6 1Univ Montpellier, INSERM, CNRS, Montpellier, France 2University of Granada and Instituto de Investigación Biosanitaria ibs, Department of Physical Chemistry, Granada, Spain 3Institute of Biomedical Chemistry, Moscow, Russian Federation 4University of Manchester, Division of Cancer Sciences, Faculty of Biology, Medicine and Health, Manchester, United Kingdom 5Oxford University, Oxford, United Kingdom 6Università Cattolica del Sacro Cuore, Rome, Italy The current mission of the EuPA Initiatives Committee is: • • • support existing European proteomics initiatives by providing funding and infrastructure as well as promoting their scientific and educational activities in a fair and transparent way encourage the creation of new initiatives that value scientific excellence, equality, inclusion and diversity foster links between initiatives and stimulate interdisciplinary exchanges to favor scientific innovation Visit our website https://eupa.org/eupa-initiatives/ to discover the 5 Initiatives currently supported by EuPA: European Bioinformatics Community for Mass Spectrometry (EuBIC-MS) Food and Nutrition Initiative Initiative for Model Organism Proteomics (iMOP) Standardization Initiative • • • • • Metaproteomics Initiatives The EuPA Initiatives Committee is composed of researchers from diverse countries in Europe - you probably already know one of us! Visit our poster during the session dedicated to Initiatives to meet us in person, to discuss what the initiatives do, what EuPA can do for you and your colleagues and maybe to explore how to create your own initiative. Fig. 1 (optional) 134

135

iP-04 Initiative for Model Organism Proteomics (iMOP) Fabrice Bertile1, Sabine Matallana-Surget2, Tristan Cardon3, Andreas Tholey4, Susana Cristobal5, Jean Armengaud6 1CNRS Université de Strasbourg, Hubert Curien Pluridisciplinary Institute - IPHC, Strasbourg, France 2University of Stirling, Stirling, United Kingdom 3Inserm, University of Lille, Lille, France 4Kiel University, Kiel, Germany 5Linköping University, Linköping, Sweden 6Paris-Saclay University, CEA, Marcoule, France The initiative for Model Organism Proteomics (iMOP), active within HUPO and EuPa, aims to advance knowledge of proteins important for human health and disease, and more generally for the environment, as part of the "One-Health" concept. This goal is achieved by promoting the emergence of new model organisms suitable for i) the study of biological mechanisms important to human health and disease, ii) a better understanding of pathogenicity, pathogen reservoirs, and the emergence of resistance, and iii) deciphering the toxic effects of pollutants and exposomes on biological systems and in this context to define sensitive sentinels of our environment, or to be relevant to the One-Health concept. Furthermore, iMOP aims to characterize non- referenced proteins across various models by describing their structural features, determining their activity and function, and establishing interspecies comparisons. The application of these efforts range from Evolutionary Biology to Medicine, as well as Environmental Proteomics and Ecotoxicology. The main innovative approaches include comparative proteomics, proteogenomics, and (eco)toxicoproteomics. By fostering collaborations between biologists and multi-omics specialists and promoting multidisciplinary approaches, new scientific questions have already been tackled, including, e.g., how age and social environment both shape the molecular phenotype of animals, the deciphering of the molecular effects of inorganic compounds on various species, the identification of the initiating molecular events when model cell lines are exposed to toxic substances identifying a circadian clock in cyanobacteria, a key biological mechanism in new prokaryotic models and showing that proteogenomics can identify and characterize the structure of new coding sequences to better define their function. Overall, iMOP aims to bring together expertise to enhance our understanding of proteome at the structural and functional level to answer fundamental biological questions using the most relevant biological models. Researchers working or interested in those questions are invited to join us for discussion at this poster session and through the iMOP initiative homepages at any time (https://hupo.org/Initiative-for-Model- Organisms-Proteomics-(iMOP) or https://eupa.org/eupa-initiatives/imop/). 135

136

iP-05 The Metaproteomics Initiative: an international community by and for metaproteomics researchers Tim Van Den Bossche1, Benoît Kunath2 1Ghent University, CompOmics, Ghent, Belgium 2University of Luxembourg, Luxembourg Centre for Systems Biomedicine, Esch-sur-Alzette, Luxembourg The Metaproteomics Initiative promotes dissemination of metaproteomics fundamentals, advancements, and applications through collaborative networking in microbiome research. We aim to be the central information hub and open meeting place where newcomers and experts interact to communicate, standardize and accelerate experimental and bioinformatic methodologies in this field. Currently, we host over 300 members from 85 labs in 22 countries, and are proudly supported by both the European Proteomics Association (EuPA, https://eupa.org/eupa-initiatives/metaproteomics/) and (HUPO, www.hupo.org/metaproteomics). More information about our vision can be found in Van Den Bossche, T. et al., Microbiome 9, 243 (2021) (doi: 10.1186/s40168-021-01176-w). the Human Proteome Organization Our efforts are structured around three core pillars: communication & collaboration, education, and standardization. Through the first pillar, communication & collaboration, we share advancements, connect researchers, and organize community-driven benchmark studies, the so-called CAMPI studies. Two, currently ongoing projects, are CAMPI-2 and CAMPI-3. CAMPI-2 aims to establish methods for sample handling to preserve metaprotein function prior to microbiota analysis, while CAMPI-3 evaluates critical decision points in the bioinformatics workflow. Through the second pillar, education, we provide the microbiome community with technical insights and easy access to expert knowledge. At this moment, we"re drafting a comprehensive guide, "Metaproteomics for Beginners", to introduce and explain metaproteomics to researchers. Through the third pillar, standardization, our efforts are focused on developing and implementing robust (meta)data standards to ensure the FAIRness (Findable, Accessible, Interoperable, Reusable) of data in the field. Therefore, we are currently working on SDRF-metaproteomics metadata standards, which will integrate attributes from the MIxS metadata standard for metagenomics environments in the SDRF-Proteomics metadata standard. We invite microbiome and proteomics researchers to join our growing community of over 300 members, to stay up-to-date on the latest developments in metaproteomics. For more information (and to become a member), please (www.metaproteomics.org). via e-mail (info@metaproteomics.org) or follow us on X (@MetaP_Init). visit our website contact us You can also Fig. 1 (optional) 136

137

iP-06 Sperm is still a relevant source for detecting missing proteins in the context of the chromosome-centric human proteome project Océane Girard1, Charline Keller2, Régis Lavigne3,4, Lila Gely5, Emmanuelle Com3,4, Blandine Guevel3,4, Pierre- Olivier Schmit6, Quentin Enjalbert7, Thomas Freour1,5, Christine Carapito2, Charles Pineau3,4 1Nantes University, CHU Nantes, Inserm, CR2TI, Nantes, France 2Laboratoire de Spectrométrie de Masse BioOrganique, Strasbourg, France 3Univ Rennes, Inserm, EHESP, Irset - UMR_S 1085, Rennes, France 4Univ Rennes, CNRS, Inserm, Biosit UAR 3480 US_S 018, Protim Core Facility, Rennes, France 5CHU Nantes, Service de Biologie de la Reproduction, Nantes, France 6Bruker Daltonique SA, Wissembourg, France 7PreOmics GmbH, Planegg, Germany The Chromosome-centric Human Proteome Project (C-HPP) aims at identifying the proteins as gene products encoded by the human genome, characterizing their isoforms and functions. The existence of products has now been confirmed for 93.1% of the genes at the protein level. The remaining mostly correspond to proteins of low abundance or difficult to access. Over the past decade, we have significantly contributed to the identification of over 300 missing proteins in the human sperm. Thanks to technical progresses in mass spectrometry and to the development of efficient sample preparation, we have been able to push further our MS analyses of the human spermatozoa, with the aim to identify new missing proteins. Interestingly, we were able to characterize additional missing proteins using the TimsTOF Ultra coupled to a NanoElute but also using the TimsTOF Pro coupled to a NanoElute based on seven different sample preparations with PreOmics kits. Here, we present the characterization of 9 new missing proteins, and can also propose 12 one-hit wonders proteins, thus contributing again to the efforts of the C-HPP. Keywords: Human Proteome Project, Missing Proteins, human Sperm, Bruker, PreOmics 137

138

iP-07 Defining a functional evidence metric for the human proteome project Eric Deutsch1, Charles Pineau2, Cecilia Lindskog3, Christopher M. Overall4, Sandra Orchard5, Nuno Bandeira6, Robert L. Moritz1, Gilbert S. Omenn7 1ISB, Seattle, WA, United States 2University Rennes, Rennes, France 3Uppsala Universitet, Uppsala, Sweden 4University of British Columbia, Vancouver, Canada 5EMBL – EBI, Cambridge, United Kingdom 6University of California, San Diego, La Jolla, CA, United States 7University of Michigan, Ann Arbor, MI, United States As the HPP draws near to completion of its first goal of achieving confident detections of all entries in the human proteome parts list, the HPP has embarked on a more audacious Grand Challenge Project of determining at least one molecular function of each protein. There will, of course, always be more to learn about every protein, but the HPP sets as its goal that we achieve a solid understanding of at least one function of each protein at the molecular level, ideally via at least two orthogonal methods. In order to track the progress in this goal, a metric must be developed since there is currently no suitable metric with which to measure the distance from our goal. The PE score, developed by Swiss-Prot decades ago, has served the HPP well in measuring its progress in its first goal, detection of each protein and information about where it is found. We have therefore developed an analogous FE ("function evidence") score. As PE1 is the ultimate goal for detection, FE1 is the ultimate goal for each protein for function, denoting a good understanding of its molecular function, which, as noted above, is not to imply that nothing more can be learned about an FE1 protein. FE5 represents the lowest category, meaning that essentially nothing is known. In practice, to be a useful metric the FE score must be computable from a repository of protein function information, and UniProtKB is currently the most complete repository of protein functions. All functional information in UniProtKB can be downloaded as an XML file, parsed, and used for calculation. The FE working group has defined a set of heuristics for computing the FE score that are precisely based on information available in UniProtKB. This information includes a free-text functional description and associated publications, Gene Ontology terms, and EC numbers. Furthermore, all annotations have an "ECO code", which is a term from the Evidence and Confidence Ontology describing how each piece of information is known. The best ECO codes describe annotations that are manual assertions based on direct experimental evidence. These stem from Swiss- Prot curators reading papers in the literature that describe experiments that provide functional evidence. At the lowest level are automatic assertions based on computational results not individually validated by curators. Here we present a draft of the FE score, how it is computed, and the progress of the Human Proteome Project in pursuit of achieving a solid understanding of at least one function of each protein in the human proteome at the molecular level. 138

139

iP-08 Young proteomics investigators club: Empowering tomorrow's proteomics talents Maurine Fucito1, . YPIC Team1, Megan Ford2 1We are 31 people representing 14 European countries, ---, United States 2German Cancer Research Center (DKFZ), Heidelberg, Germany YPIC (Young Proteomics Investigators Club) is an inclusive group of early career researchers in proteomics from all over Europe. YPIC was founded at the 10th annual EuPA conference in Istanbul in 2016. The team aims at representing the extra-scientific aspects of being a young researcher and favoring the networking of young European scientists in proteomics. All in all, our aim is to connect and support early career researchers on their path through and beyond academia! Throughout the year, YPIC organizes career, mentoring, "Meet-the-Expert", and proteomics intro sessions. This includes the educational day at the yearly EuPA conference and our own YPIC annual proteomics gathering, where the winner of the Student Proteomics Fund is announced. The team also organizes non-scientific seasonal webinars as well as proteomics challenges. 139

140

iP-09 Second community evaluation of glycoproteomic informatics solutions Nicholas Riley1, Keira Mahoney2, Nara Chung2, Vincent Chang2, Valentina Rangel-Angarita2, Devon Kohler3, Kiyoko Aoki-Kinoshita4, Frédérique Lisacek5, Nichollas Scott6, Stacy Malaker2 1University of Washington, Chemistry, Seattle, United States 2Yale University, Chemistry, New Haven, United States 3Northeastern University, College of computer science, Boston, United States 4Soka University, Glycan and Life Systems Integration Center, Tokyo, Japan 5University of Geneva, Department of Computer Science & Section of Biology, Geneva, Switzerland 6University of Melbourne, Department of Microbiology and Immunology, Melbourne, Australia in The field of glycoproteomics has experienced considerable growth within the last decade. Analytical improvements instrumentation, fragmentation, and sample preparation have allowed for biological discoveries in how glycosylation contributes to protein function and cellular health. However, one critical bottleneck in glycoproteomics is data analysis via search algorithms due to the vast heterogeneity in glycosylation. In the first Human Glycoproteomics Initiative study, researchers were sent the same data file and were asked to report glycopeptides found. Unfortunately, vast differences were reported between the groups. Additionally, since then, many new programs and developments have emerged that warrant consideration. Thus, here we propose the second HGI study, wherein developer teams will be sent data files comprising N- glycopeptides, O-glycopeptides, and/or a mixture of both. The results will allow us to recommend suitable settings, parameters, and programs for various sample types. Ultimately, we believe this work will prove useful to the broad glycoproteomics community. Fig. 1 (optional) 140

141

iP-10 Proteomics Multi-laboratory Experiment 13 (PME13). EuPA Standardization Initiative Fernando Corrales1 1Centro Nacional de Biotecnología, CSIC, Macromolecular Structure, Madrid, Spain This study has been done by all the authors listed below, who share the authorship. The abstract has been submitted by the coordinator of the EuPA Standardization Initiative. Mass Spectrometry & Proteomics Core Facility, IRB Barcelona; Proteomics Platform, CIC bioGUNE, Derio; Proteomics Facility of Toulouse, CNRS; Proteome Pole of Montpellier (PPM),CNRS/UAR BioCampus Montpellier; Proteomics Platform-MSBio, Institut Pasteur, Paris; ETHZ, Zurich; Proteomics Laboratory DBMR - Bern University; Koç University Protemics Facility; Clinical Proteomics Laboratory, Institute for Biomedical Technologies, National Research Council (ITB-CNR); Microbiology, University of Cádiz; Department of Molecular Immunology and Toxicology,National Institute of Oncology, Budapest; Protein Chemistry Facility (PCF),Institute of Biochemistry of the Romanian Academy (IBRA); Servei de Proteomica SCSIE, Universitat de València; Proteomics Laboratory, Instituto de Biomedicina de Sevilla; Laboratory of Pharmaceutical Biotechnology, Ghent University; Proteomic Unit (UCM), Complutense University of Madrid; Cardiovascular Proteomics Laboratory, CNIC, Madrid; Kuttam Omics Laboratory, Koç University; IMIBIC Mass Spectrometry and Molecular Imaging Unit (IMSMI), IMIBIC, Cordoba; Proteomics Core Facility, Centro de Biología Molecular Severo Ochoa, Madrid; Proteomics Core Facility, Instituto de Salud Carlos III, Madrid; Proteomics Unit, UPF-CRG, Barcelona; Proteomics Unit, ISPA-CSIC, Oviedo; Proteomics Core Facility, University of the Basque Country; Servicio de Proteómica y Genómica, CIB-CSIC, Madrid; Proteomics Platform, Navarrabiomed, Pamplona; CNB Proteomics Facility, Centro Nacional de Biotecnología (CNB-CSIC) Madrid; Single cell analysis at the proteome level has become feasible as new technologies for sample preparation and mass spectrometry analysis ensure efficient protein extraction and processing, as well as analytical sensitivity in the attomolar range. However, the implementation of high sensitivity proteomics (HSP) protocols remains a challenge. In this context, the EuPA standardization initiative launched the Proteomics Multicentric Experiment (PME13) with the main objectives of benchmarking HSP analysis protocols, providing tools to facilitate HSP implementation and assessing laboratory performance. The reference sample used in the study consisted of a serial dilution of E. coli protein extract from 100 to 1 ng, provided to participants in intact and trypsin digested formats. A pilot analysis was conducted to optimise the analytical pipeline and to minimize sources of uncertainty. After analysis, DDA and DIA datasets were generated and submitted for centralized processing along with metadata specific to each participant. Data were collected via NextCloud, processed using MaxQuant v2.0.2.0 and integrated, visualized and analyzed using R. Searches were mainly performed using default parameters, agreement between runs within the same concentration level and sample, and 1% FDR at peptide and protein level. Thirty-six laboratories from eight different countries requested PME13 samples, almost half of which (17 out of 36) had not participated in previous PMEs. We received DDA results from 28 out of 36 laboratories (78%) and DIA results from 16 out of 23 laboratories (70%). Overall, our data indicate that the reference sample provides a reliable basis for benchmarking, establishing protocols and assessing performance. The influence of analytical variables and the correlation with the physico-chemical properties of the analyzed peptides are presented and discussed. 141

143

iP-11 EuBIC-MS Marie Locard-Paulet1, Louise Marie Buur1, Ralf Gabriels1 1EuBIC-MS, Toulouse, Germany The European Bioinformatics Community for Mass Spectrometry (EuBIC-MS) is a EuPA initiative that was started in November 2015. Since then, it brings together people from the bioinformatics community who work on mass spectrometry (MS) data to provide an open, collaborative and constructive environment. It is carried enthusiastically by the bioinformatics and MS community and welcomes everyone who is willing to contribute. EuBIC-MS hosts an annual conference focused on computational MS, uniting leading bioinformatics researchers, early-career scientists, and industry professionals from omics fields such as proteomics and metabolomics. This event alternates each year between a Winter School, featuring keynote lectures and hands-on workshops, and a Developers Meeting, centered on hackathons. In addition to the main conference, EuBIC-MS offers workshops at various MS and proteomics conferences and actively contributes to bioinformatics hubs at these events. Along with organizing and being present at events, EuBIC-MS also has several community driven projects. Our poster will describe the main activities of EuBIC including the upcoming Developers Meeting in the beginning of 2025, give an overview of the active community projects, and provide information about how to become a member and contribute to EuBIC-MS. 143

144

iP-12 Critical Assessment of Metaproteome Investigations (CAMPI): a Series of International-Lab Comparisons of Experimental and Bioinformatic Workflows Tim Van Den Bossche1, Benoît Kunath2, Sergio Uzzau3, Pratik Jagtap4, Robert Hettich5 1CompOmics (VIB - Ghent University), Gent, Belgium 2University of Luxembourg, Luxembourg Centre for Systems Biomedicine, Esch-sur-Alzette, Luxembourg 3Università degli Studi di Sassari, Sassari, Italy 4University of Minnesota, Minneapolis, MN, United States 5Oak Ridge National Laboratory, Oak Ridge, TN, United States Metaproteomics is a powerful approach for characterizing the functional metabolic activities of complex microbiome systems. For example, clinical metaproteomics provides details about how the human microbiome affects/responds to clinical conditions, whereas environmental metaproteomics provides information on how the microbiome affects/responds to the ecosystem. Metaproteomics is still under rapid development, with a few standardized methods and approaches even among the experienced practitioners. To better understand and enhance the overall power of metaproteomics, the Metaproteomics Initiative [1] has launched a series of Critical Assessment of Metaproteome Investigations, termed "CAMPI." The first campaign, CAMPI-1, compared experimental and bioinformatic workflows across international laboratories, demonstrating a strong correlation in microbial functionals [2]. Building on this, two new CAMPI projects are currently in progress. CAMPI-2, involving 11 labs, focuses on evaluating a variety of sample preparation protocols for fecal or soil microbiomes to enable sample shipment at ambient temperatures. CAMPI-3, meanwhile, is evaluating bioinformatic tools for protein assignments and inferring taxonomic assignments using metaproteomic raw data from a synthetic community (with a known representation) and a complex fecal sample. Over 20 labs participated, revealing a wide variety of strategies and accuracies in the analysis. These findings are being compiled into a manuscript that will provide critical insights and propose best practices. Future CAMPI projects are being prioritized to address key challenges in metaproteomics, driving the field forward to become an essential tool for tackling global challenges in microbiome research. References [1] Van Den Bossche, T., Arntzen, M.Ø., Becher, D. et al. The Metaproteomics Initiative: a coordinated approach for propelling (2021). https://doi.org/10.1186/s40168-021-01176-w characterization of microbiomes. Microbiome 9, 243 the functional [2] Van Den Bossche, T., Kunath, B.J., Schallert, K. et al. Critical Assessment of MetaProteome Investigation (CAMPI): a multi-laboratory comparison of established workflows. Nat Commun 12, 7305 (2021). https://doi.org/10.1038/s41467-021-27542-8 144

145

iP-13 HUPO ECR Committee: Empowering the next generation proteomics researcher Ruth Huttenhain1 1Stanford University, Molecular and Cellular Physiology, Stanford, United States The mission of the HUPO Early Career Researcher (ECR) Committee is to create a vibrant network for early- career proteomics researchers, enabling them to thrive, share knowledge, and develop the skills necessary to lead in the ever-evolving field of proteomics. The committee aims to: Promote integration and networking opportunities for ECRs within the HUPO community and with members at various career stages. Provide scientific visibility and research opportunities for ECRs. Offer career development activities that enhance the competitiveness of ECRs in the job market. Improve equity, diversity, and inclusion (EDI) within proteomics by integrating EDI principles into all ECR activities. Create inclusive spaces where diverse voices are heard, fostering collaboration and mutual support. Disseminate HUPO"s activities, mission, and vision among ECRs globally. To achieve these goals, the HUPO ECR Committee organizes various activities throughout the year, with a strong focus on the HUPO World Congress. These activities include: ECR Mentoring Sessions: Offering invaluable insights into career development, these sessions feature lively discussions and interactive polls, allowing participants to address critical career-related challenges openly. Proteomics Highlight of the Year: This manuscript competition showcases groundbreaking research by postdocs and early-career principal investigators, with finalists presenting their work in a dedicated plenary session. The competition emphasizes the significant contributions of young scientists to the field of proteomics. 3-Minute Thesis Competition: In a dynamic, fast-paced format, selected PhD students present their research in just three minutes to a lay audience, highlighting the importance of clear and effective science communication. Poster Competition: Graduate students and postdocs present their research in a competitive setting, with evaluators recognizing excellence in innovation, impact, and presentation. Networking Evening: This relaxed event allows early-career researchers and senior scientists to connect, share ideas, and explore potential collaborations. Beyond the annual HUPO Congress, the committee also organizes regular online panel discussions throughout the year on various topics relevant to early-career researchers, such as scientific communication, grant writing, and starting a new lab. Additionally, we have ECR representatives on all HUPO committees to advocate for the needs of early-career researchers within the proteomics community. The HUPO ECR Committee plays a pivotal role in the professional development of young proteomics researchers. Through our activities, we are building a supportive community that empowers ECRs to excel in their scientific endeavors and contribute meaningfully to the global proteomics community. As we look to expand our offerings with new opportunities for skill development, networking, and career advancement, we encourage all ECRs to join our efforts, stay engaged with our activities, and take advantage of the resources and support we offer. Stay tuned for more exciting events and opportunities and don"t miss any HUPO ECR event at HUPO2024! 145

146

Defining signaling networks - Funcyional PTMs (P-I-0001 to P-I-0053) P-I-0001 Phosphoproteomic alterations in Mesenchymal stem cells induced by Dimethyl fumarate treatment Maryam Adelipour1, Jeongkwon Kim1 1Chungnam National University, Daejeon, South Korea Mesenchymal stem cells (MSCs) have garnered attention as promising candidates for treating degenerative diseases; however, concerns persist regarding their stability and viability in cell therapy applications. In response, this study endeavors to augment MSC viability and efficacy by subjecting them to dimethyl fumarate (DMF) preconditioning, a process evaluated through phosphoproteomic analysis. To delve into the effects of DMF preconditioning, the researchers conducted protein extraction and digestion procedures on both untreated and DMF-treated MSCs. Utilizing the s-trap method alongside trypsin/lysin enzymes ensured comprehensive processing of the samples. Following this, phosphorylated peptides were enriched using TiO2 and identified via nanoLC-MS/MS technology. The resulting data underwent thorough analysis utilizing Maxquant and Perseus software tools. Remarkably, the analysis revealed a notable increase in the expression levels of 559 phosphorylated peptides among the 837 commonly detected in DMF-treated MSCs compared to their untreated counterparts. Network analysis conducted through Reactome shed light on RNA Binding Motif Protein 39 (RBM39) as a potentially pivotal mediator of cellular responses, hinting at its involvement in crucial RNA processing events and gene expression regulation mechanisms. Furthermore, the efficacy of TiO2 enrichment in boosting the fraction of phosphorylated peptides by approximately 50% underscores its robustness as a technique. In conclusion, this study serves as a testament to the significant impact of DMF treatment on the phosphoproteomic landscape of MSCs. The proposed avenue for further exploration of identified pathways via high-throughput phosphoproteomics promises a deeper and more comprehensive understanding of the underlying molecular mechanisms driving the observed effects. 146

147

P-I-0002 Insights into phosphorylation post-translational modifications in the proteome of C. albicans during oxidative stress response Víctor Arribas1, Ana Borrajo1, Raquel Martinez1, Maria Luisa Hernáez2, Lucía Monteoliva1, Gloria Molero1, Concha Gil1,2 1Complutense University of Madrid, Microbiology and Parasitology, Madrid, Spain 2Complutense University of Madrid, Proteomics Unit, Madrid, Spain Candida albicans is a common fungus of the human microbiota that can turn into an opportunistic pathogen under specific circumstances, particularly in individuals with a weakened immune system, leading to invasive candidiasis. During infection, C. albicans interacts with phagocytes, initiating the respiratory burst which produces reactive oxidant species (ROS) such as hydrogen peroxide (H2O2) in the intend to destroy the fungus. C. albicans survival depends on its ability to respond to this oxidative stress. Several well-known pathways have been involved in the response to the oxidative stress, such as the high osmolarity glycerol (HOG) MAPK pathway, that stands out as a primary mechanism for signalling the oxidative stress response (1), and Cap1 as the main transcription factor involved (2), finally promoting the expression of antioxidant proteins. To dig in other proteins and routes contributing, our group has done several proteomic studies, that have unveiled important proteins for C. albicans survival such as Prn1 and Oye32 (3, 4). The present work describes a quantitative proteomic assay (DDA-MS) conducted with the wild-type strain SC5314 to characterize the changes in protein abundance and PTMs in response to a 200-min 10 mM H2O2 treatment. The assay identified approximately 1,900 phosphopeptides from 895 proteins that contained 2421 phosphosites. Gene Ontology (GO) term enrichment analysis of the proteins with significantly altered phosphorylation revealed an enrichment in proteins implicated in chromatin remodelling and in nucleotide binding proteins. Several protein kinases also exhibited significant changes in phosphorylation and, in concordance, npr1∆, kis1∆, and kin2∆ mutant strains displayed increased susceptibility to H2O2. However, Hog1 and Mkc1 MAPK signalling pathways, which sense H2O2, were deactivated after 200 min of H2O2 treatment. Among the nucleotide-binding GO Term enriched proteins we identified several transcription factors, such as Frc1 and Gzf3. As in the case of MAPK mutants, frz1∆ and gzf3∆ strains presented increased susceptibility to H2O2. An integrative analysis of significant changes in both phosphopeptide and peptide abundances revealed that autophagy and cell cycle are strongly regulated after H2O2 treatment. Specifically, the decreased phosphorylation of Rim15 between Ser1183-Ser1190 might suggest that autophagy signalling pathway might be activated. In addition, cells undergo a G2 cell cycle arrest probably mediated by decreased phosphorylation of Cdc5 at Thr25. 1. Alonso-Monge R, et al. Eukaryot Cell. 2003 2. Dai B Di, et al. FEBS. 2013 3. Amador-García A, et al. mSystems. 2021 4. Arribas V, et al. Antioxidants. 2024 147

148

P-I-0003 Using phosphoproteomics to explore autophagy relevant TBK1 signal transduction Ibrahim Bayramoglu1, Jörn Dengjel1 1University of Fribourg, Biology, Fribourg, Switzerland TBK1 (TANK-binding kinase 1), a member of the IKK (IκB kinase) kinase family, is a crucial regulator of various cellular processes, including autophagy, immunity, and inflammation. Autophagy is a constitutive cellular mechanism which involves the degradation and recycling of cellular components, such as organelles and proteins. Autophagy is upregulated in response to diverse stimuli like nutrient deprivation, oxidative stress, and infection. TBK1 emerges as a critical player in autophagy regulation by interacting with key autophagy-related proteins, including OPTN (optineurin). Our study aims to investigate the downstream targets of TBK1 in the context of metabolic stress and mitophagy. Mitophagy, the selective autophagy of mitochondria, is induced here by subjecting MCF7 cells to glucose starvation, achieved via shifting the cells to galactose-containing media without any glucose in it. Galactose shifting elevates oxidative phosphorylation due to its less efficient ATP generation compared to glucose glycolysis, resulting in increased mitochondrial damage and subsequent mitophagy upregulation. We have observed a notable shift towards a more respiratory state and significant phosphorylation changes in mitochondrial proteins. However, linking these changes to mitophagy remains challenging. To further assess the impact of oxidative conditions, instead of shifting, we are adapting the cells to galactose media. The more oxidative conditions potentially lead to increased mitochondrial turnover, aiding in the detection of mitophagy- related proteins and TBK1-relevant phosphorylation changes. To overcome the challenges posed by the galactose shifting experiments, we are exploiting mitophagy-inducing pharmaceutical agents such as Antimycin A, Oligomycin A, and CCCP with the addition of a TBK1 inhibitor to investigate the mitophagy-relevant TBK1 phosphoproteome. So far, the TBK1 phosphoproteome remains unexplored. We anticipate that our investigation will unveil novel downstream targets of TBK1, shedding light on its role in autophagy regulation. 148

149

P-I-0004 Active oxidative metabolism and impaired glyoxalase system amid elevated intracellular oxidative stress in non- small cell lung cancer shown by immediate post surgery alkylation and redox proteomics Ruth Birner-Grünberger1,2, Tamara Tomin1, Sophie Honeder1, Laura Liesinger1, Daniela Gremel1, Bernhard Retzl1, Joerg Lindenmann2, Luka Brcic2, Matthias Schittmayer1 1TU Wien, Vienna, Austria 2Medical University of Graz, Graz, Austria Reactive oxygen species can oxidatively modify enzymes to reroute metabolic pathways according to tumor needs but we lack overview of all potential targets. Thiol groups are most susceptible to oxidative modifications but rarely analyzed in clinical settings due to their reactivity. To accurately address the cross-talk between redox signaling and metabolism we collected tumor and healthy tissue from 70 individuals with non-small cell lung cancer right after surgery into a thiol-quenching solution, then carried out redox-proteomics based on relative MS1 quantitation of isotope labeled cysteines discriminating between their reductive and oxidative state and averaging redox ratios of different peptides for the same site. As a result of such an unbiased and meticulous approach, we, for the first time, show evidence of higher oxidation of a number of key metabolic and especially glucose related enzymes and their specific cysteine sites in tumor. Integrating redox and standard label free quantitative proteomics we demonstrate that lung cancer strives to maintain oxidative metabolism amid the rise of intracellular oxidative stress. We also report both redox and protein level deactivation of the glyoxalase system, which might be compensated by higher excretion or lower production of toxic methylglyoxal, aiding cancer progression. 149

150

P-I-0005 Proteome-wide in vitro kinase and phosphatase assays to determine enzyme-substrate interactions Melanie Brunner1, Zehan Hu1, Jörn Dengjel1 1University of Fribourg, Biology, Fribourg, Switzerland i.e. direct events are difficult to determine. (De)phosphorylation reactions are highly dynamic and involved in virtually all cell fate decisions. Protein kinases and phosphatases are hence interesting drug targets. For targeted drug development, knowledge about direct and indirect enzyme-substrate interactions is critical. Whereas in vivo phosphoproteomics supports the routine analysis of (de)phosphorylation events, causal In vitro kinase/phosphatase assays are still the gold-standard to discriminate potential direct from indirect events. However, in their classical setup respective assays are time- and labor-intensive, and often result in false-positive events. Here, we describe in vitro reactions using native proteome as input which, in combination with in vivo phosphoproteomics, support screening approaches to determine kinase/phosphatase-substrate interactions in high throughput. Importantly, cell lysates are generated under native conditions preserving 4D protein structures and respective protein complexes are immobilized on NHS-activated Sepharose beads. Endogenous protein kinases and phosphatases, which are still active due to lysis conditions, are inactivated using covalent inhibitors. Next, purified kinases/phosphatases-of-interest are added and in vitro (de)phosphorylation reactions are performed on proteome scale. Comparison of active and inactive enzymes-of-interest enable the identification of enzyme target sites. Combining in vitro with classical in vivo data using e.g. enzyme inhibitors supports the identification of bona fide target sites. In proof-of-concept studies we used either a kinase of interest, or the serine-threonine phosphatases PP1 and PP2 complexes and performed in vivo as well as in vitro phosphoproteomics using our newly developed On Beads In Vitro Kinase Assay (OBIKA) and On Beads In Vitro Phosphatase Assay (OBIPhA). We identified several tens of thousand phosphosites of which several hundred were characterized as bona fide enzyme target sites, i.e. identified as regulated in vivo and in vitro, highlighting the increased throughput of the on beads assays compared to classical in vitro assays. 150

151

P-I-0006 Proteomic and phosphoproteomic characterization of FFPE lung adenocarcinoma harboring ALK, EGFR or KRAS Mutations, and triple wild-type tissue sections Fanni Bugyi1,2, Mirjam Balbisi2, Simon Sugár2, Lóránd Váncza3, Ilona Kovalszky3, Tünde Harkó4, Ibolya Laczo5, Gabor Kecskemeti6, Zoltán Szabó6, Judit Moldvay4,7, László Drahos8, Lilla Turiak2 1ELTE Eötvös Loránd University, Hevesy György PhD School of Chemistry, Budapest, Hungary 2HUN-REN Research Centre for Natural Sciences, MTA-TTK Lendület (Momentum) Glycan Biomarker Research Group, Budapest, Hungary 3Semmelweis University, Department of Pathology and Experimental Cancer Research, Budapest, Hungary 4National Korányi Institute of Pulmonology, Budapest, Hungary 5Békés County Central Hospital, Gyula, Hungary 6University of Szeged, Department of Medical Chemistry, Interdisciplinary Excellence Centre, Szeged, Hungary 7University of Szeged, Department of Pulmonology, Albert Szent-Györgyi Medical School, Deszk, Hungary 8HUN-REN Research Centre for Natural Sciences, MS Proteomics Research Group, Budapest, Hungary Lung cancer (LC) is the leading cause of cancer-related mortality. Lung adenocarcinoma subtype (LUAC) accounts for approximately 40% of all LC cases. Mutations in key oncogenes have been identified as important genetic alterations in LUAC, including genes encoding epidermal growth factor receptor (EGFR), Kirsten rat sarcoma viral oncogene homolog (KRAS), and anaplastic lymphoma kinase (ALK). Our aim was the characterization of the altered biological pathways and phosphorylation events in LUAC tumors harboring EGFR, KRAS, or ALK oncogenic mutation and triple wild-type (WT) LUAC tumors by MS-based quantitative proteomics and phosphoproteomics analyses of 84 formalin-fixed paraffin-embedded (FFPE) tissue sections. The on-surface tryptic digestion and subsequent phosphopeptide enrichment enabled the identification of differentially expressed proteins and phosphosites, respectively using label-free quantitation. During the proteomic analysis, 1995 proteins were subjected to statistical analysis. To identify proteins differentially expressed between the four sample groups, two-group comparisons were performed. Altogether, the expression of 160 proteins was significantly altered in at least one of the six comparisons. Based on the functional enrichment analysis, most of these proteins were affiliated with intracellular organelle lumen, actin cytoskeleton, Golgi apparatus, vesicle, and focal adhesion terms. Samples with EGFR mutation showed the most differences. Moreover, 31 proteins had been identified as mutation-specific (significantly altered in one mutation compared to the others) serving as potential therapeutic targets. In the phosphoproteomic analysis, 386 phosphosites were considered for statistical analysis. Principal component analysis of these phosphosites revealed considerable differences between the samples with different mutations. The expression of 202 sites was significantly altered between the four sample groups. Additionally, 61 sites were found to be significantly dysregulated specifically in one sample group. These 61 phosphosites derived from 48 phosphorylated proteins. Functional enrichment analysis revealed that most of these proteins were involved in cellular processes, like cytoskeleton organization, and metabolic processes, like regulation of primary metabolic process, regulation of cellular biosynthetic process, gene expression, and RNA processing including RNA splicing. Kinase-substrate enrichment analysis revealed altered activities of 10 kinases. Our results could provide significant insights for developing therapeutic strategies for LUAC. This project was supported by the Lendület (Momentum) Program of the Hungarian Academy of Sciences (HAS, MTA). Funding from the National Research, Development, and Innovation Fund of Hungary (FK 131603) is acknowledged. Samples were run at the ISTA LSF Mass Spectrometry Facility, part of the Scientific Service Units of ISTA. The Authors are grateful to Goran Mituloviĉ for the column. 151

152

P-I-0007 Functional phosphoproteomics and network models to elucidate mechanisms of deregulated signaling Mira Lea Burtscher1, Clement Potel1, Martin Garrido-Rodriguez1,2, Isabelle Becher1, Julio Saez-Rodriguez2, Mikhail Savitski1 1EMBL Heidelberg, Genome Biology unit, Heidelberg, Germany 2Heidelberg University, Institute for Computational Biomedicine, Heidelberg, Germany is a powerful method to study such deregulated signaling processes, but Deregulated cellular signaling is a hallmark of cancers such as melanoma, the majority of which have a implications. Quantitative phosphomimetic mutation of the BRAF kinase with significant phenotypic phosphoproteomics low phosphoproteome coverage and the need for sophisticated computational methods often constrain the obtained biological insights. We applied a phosphoproteomics method which combines high-sensitivity detection and precise multiplexed quantification to profile the response of three BRAF-mutated cell lines to the BRAF inhibitor Dabrafenib based on more than 50k phosphopeptides. Further, multiplexing of six different Dabrafenib treatment time points resulted in a quantitative time-resolved profile for more than 80k phosphopeptides in the melanoma cell line A2058. We used computational methods such as clustering, kinase activity prediction with DecoupleR and the KinaseLibrary as well the causal reasoning method PHONEMEs to extract mechanistic insights from these datasets using network models. To deepen our understanding of the functionality of the observed phosphorylation events, we performed solubility proteome profiling experiments at the phosphoproteome level across three cell lines. This approach enabled us to correlate specific phosphorylation events with protein condensation processes and to elucidate the effects of BRAF mutations on transcription. These findings highlight the potential of deep phosphoproteome profiling on abundance as well as functionality level combined with computational methods to study drug effects on signaling networks in cancer. 152

153

P-I-0008 Dissecting the spatiotemporal regulation of the cell cycle using phosphoproteomics and highlighting functional differences within proteoform families Anthony Cesnik1,2,3, Christian Gnann2,3, Frank McCarthy3, Daniel Itzhak3, Emma Lundberg1,2,3,4 1Stanford University, Bioengineering, Stanford, CA, United States 2KTH Royal Institute of Technology, Science for Life Laboratory, Department of Protein Science, Stockholm, Sweden 3Chan Zuckerberg Biohub, San Francisco, CA, United States 4Stanford University, Pathology, Stanford, CA, United States Protein-coding genes are expressed as dynamic ensembles of unique molecular forms known as proteoforms, and differences between their amino acid sequences or post-translational modifications can imbue important functional differences. These functional differences help to drive fundamental biological processes, such as the cell cycle, a highly conserved process during which a cell grows and divides into two daughter cells. Post- translational modifications in particular can impact the localization of proteins and create important functional differences and compartmentalization of proteoform families. We have conducted a deep phosphoproteomic study of the cell cycle using bottom-up phosphoproteomics to identify over 77,000 phosphosites in U-2 OS cells following fluorescence-aided cell sorting using FUCCI cell cycle markers. These results have expanded the network of cell cycle regulation, with over 4,000 phosphosites displaying abundance changes over the cell cycle, including sites that may be linked to protein translocation events and protein abundance changes independent of RNA levels. These phosphopeptide results provide insights into the origins of cell-to-cell heterogeneity of proteoform expression, including that introduced by the cell cycle, which have relevance in studying human health and diseases such as cancer, and which provide fodder for interrogating the proteoform systems biology of the cell cycle in the future. 153

154

P-I-0009 Simultaneous profiling of sialoglycoproteome and phosphoproteome in TKI-resistant NCSLC cells using metal ion- decorated ZIC-cHILIC StageTip Yi-Ju Chen1, Yan-Lin Chen1, Kun-Hao Chang1, Hsiang-Chun Cheng1, Chiao-Chun Chang1, Hao Fang1, Yu-Ju Chen1 1Academia Sinica, Institute of Chemistry, Taipei, Taiwan The cellular plasma membrane proteins are densely decorated with surface glycosylation and intracellular phosphorylation whose interplay determines the cell-cell communication and signaling cascades. However, their concomitant characterization remains extreme challenges due to complexity of glycan structures, low stoichiometry (particularly sialylation), labile, dynamic nature and low detectability of both phosphopeptides and glycopeptides in mass spectrometry. In this study, we introduce a streamlined metal ion-decorated ZIC-cHILIC strategy for simultaneous, sensitive and large-scale (sialo-)glycoproteomic and phosphoproteomic profiling. We previously reported a stepwise ZIC-cHILIC StageTip with high enrichment efficiency and specificity for large-scale glycoproteomic characterization (Chen, et al., Anal. Chem. 2021). Based on different degree of negative charge on phosphate and glycans, we further implemented metal ions into a tandem ZIC-cHILIC StageTip and simple pH control as a streamlined strategy for simultaneous enrichment and stepwise fractionation of both (sialo)glycopeptides and phosphopeptides profiles. Despite tyrosine kinase inhibitor (TKI) demonstrated precision oncology for EGFR-mutant patient in non-small cell lung cancer (NSCLC), nearly every patient develop resistance. The strategy was demonstrated on patient-derived TKI-resistant NSCLC cells, identifying 13,714 high- confident glycopeptides (including 3,142 sialo-glycopeptides) and 15,164 phosphopeptides in 4 different EGFR mutated NSCLC cells. Compared to naked ZIC-cHILIC, the metal ion under pH control increases 5.6-fold specificity (from 15% to 84%) and 10-fold phosphopeptides (17,036). Without protein immunoprecipitation, the approach identified important glycosylation and phosphorylation on druggable protein, such as 10 of 12 N- glycosylation sites and driver autophosphorylation sites, such as Y1172 and Y1110 of EGFR protein, the target therapy in lung cancer. Quantitative comparison also showcase the sensitivity with high coverage of many oncoproteins and their downstream signaling cascades, such as dysregulated glycosylation and phosphorylation under different EGFR downstream cascades in PI3K and MAPK pathway in these patients. Our approach may offers a high performance approach to reveal the dynamic interplay of multi-PTMomic landscape in cell membrane. 154

155

P-I-0010 A multi-proteomics analysis of FGFR family mutations on cancer signalling networks Iftekhar Mahmud Chowdhury1, Xiaonan Liu1, Salla Keskitalo1, Markku Varjosalo1 1University of Helsinki, Institute of Biotechnology, Helsinki, Finland Protein kinases play critical roles in regulating cellular processes, including growth, survival, differentiation, and apoptosis. Mutations in protein kinase are frequently associated with development and progression of cancer. These mutations, often referred to as cancer driver mutations can lead to the continuous activation or inhibition of kinase signalling pathways, which in turn promotes oncogenic transformations. Our study focuses on the physical and functional interaction networks of 180 established driver mutations (from the COSMIC database) in 58 cancer driver protein kinases (Figure1). We generated Flp-In™ T-REx™ 293 stable cell lines harbouring the mutations including those in the FGFR family (FGFR1/2/3/4). Fibroblast growth factor receptors (FGFRs) are a family of essential transmembrane receptors involved in diverse cellular activities. Mutations of FGFR including amplifications, activating mutations, rearrangements and fusions lead to the imbalance of signalling pathways resulting in oncogenic transformation of healthy cells. Here we present the impact of such driver mutations in the FGFR family on cancer signalling networks using a multi- proteomics strategy. Utilizing affinity purification mass spectrometry (APMS) and proximity labelling (BioID), we mapped the protein- protein interaction networks for each mutated kinase relative to its wild-type form. This allowed us to understand the altered interactor profiles and the signalling pathways affected by the mutations. Further, phosphoproteomics was applied to illustrate the phosphorylation landscape induced by the mutations, including specific phosphosites. To functionally validate the consequences of the mutations, kinase-dead mutants of FGFR family were also expressed and analysed. Our findings indicate a significant upregulation of several critical cancer driver proteins, including ABL2, ARAF, PIK3CA, PIK3CB, PLCG1 and STAT5B by FGFR mutations. These changes lead to distinct signalling cascades as confirmed by pathway-specific reporter assays. In addition, confocal microscopy and MS-microscopy demonstrated notable changes in the subcellular localization of the mutated FGFRs which may account for their altered interaction profiles. Many of the identified proteins are attractive candidates for therapeutic drug targeting in FGFR-induced cancers. Future directions of this research will focus on the validation of the functional relevance of the cancer driver proteins by gene silencing and specific drug targeting experiments as well as immunohistochemical analysis on tissue from the mutation-confirmed cancers of interest followed by whole proteome analysis from the tissue samples to provide a comprehensive understanding of the global effect of the mutation. This multi-proteomics strategy allows the characterization of the complex signalling changes induced by FGFR mutations and identifies new targets for precision cancer medicine. 155

157

P-I-0011 Traceability of post-translational modifications in atherosclerosis from plasma to tissue to single-cell proteomics David del Rio Aledo1, Cristina Amparo Devesa Arbiol1, Estefanía Núñez1,2, Consolación Marín1,2, Paula Nogales1, Laura Carramolino1, Miguel Sánchez1, Inmaculada Jorge1,2, Andrea Laguilllo1,2, Rafael Barrero-Rodríguez1,2, José Manuel Rodríguez1,2, Jose Luis Martin Ventura3, Miguel Ángel del Pozo1, Jacob Fog Bentzon1,4, Jesús Vázquez1,2 1Centro Nacional de Invstigaciones Cardiovasculares Carlos III, Cardiovascular Proteomics Laboratory, Madrid, Spain 2CIBER de Enfermedades Cardiovasculares (CIBERCV), Madrid, Spain 3IIS-Fundación Jiménez Díaz Hospital, Autonoma University of Madrid (UAM), Madrid, Spain 4Aarhus University, Clinical Medicine, Aarhus, Denmark Atherosclerosis is a chronic inflammatory disease characterized by the accumulation of LDL, primarily composed of Apolipoprotein B (ApoB). Various strategies have been proposed for the treatment of atherosclerosis, mainly focused on reducing LDL levels, such as statins. However, these approaches are often insufficient to prevent the risk of thrombosis. It is known that post-translational protein modifications (PTMs) regulate the activity of numerous cellular pathways, and some evidence relate them to atherosclerosis. However, their involvement in the development of atherosclerosis and their potential as therapeutic targets have not yet been explored form an unbiased, high-throughput perspective. To study the role of PTM and its association with plaque activity, we analysed tissue and plasma samples from pig and mouse atherosclerotic regression models, as well as patients with subclinical atherosclerosis, totaling 1080 samples. We conducted a comprehensive and unbiased analysis of PTMs using an "Open-Search" strategy using Comet-PTM and a novel integrative quantitative workflow, both developed in our laboratory. We have detected numerous modifications that are associated with decreased plaque activity. Since Smooth Muscle Cells (SMCs) are the principal producer of extracellular matrix during atherosclerotic plaque development and are responsible for plaque stability, we also performed a PTM-based analysis focused on these cells. Specifically, we analyzed an extracellular matrix preparation from primary cultured SMCs using a mouse regression model (n=10). We have also performed a preliminary study using Single Cell Proteomics of mouse aortic SMCs from a mouse regression model. Preliminary results revealed that several PTMs previously observed at the tissue and plasma level are present in SMCs and their extracellular matrix, reflecting the relevance of PTMs in SMCs during atherosclerosis. Our results suggest the existence of PTMs associated with atherosclerosis progression and regression, supporting the idea that modifications could have unprecedent implications in vascular remodelling, and open the way to the study of their mechanistic implications. Fig. 1 (optional) 157

158

P-I-0012 Mass spectrometry analysis of post translational modification in adeno-associated virus capsid proteins Vrushali Deshpande1, Ramaraj Kannan1, Ravi Kiran1, Ruchita Selot1, Riya Patra1, Ashish Khaparde1, Sharath Babu1, Rohit Shetty2, Arkasubhra Ghosh1 1Narayana Nethralaya Foundation, GROW LAB, Bangalore, India 2Narayana Nethralaya, Cornea and Refractive surgery, Bangalore, India Introduction: Adeno-associated virus (AAV) capsids are currently the most favoured vectors for gene delivery. Recombinant viruses produced in a specific cell line are likely to acquire unique post-translational modifications during the intracellular maturation of their capsid proteins. These post-translational modifications (PTM) on the capsid proteins are known to regulate the stability, packaging and transduction efficiency and shield potentially immunogenic protein epitopes. Understanding these fine tune regulations may aid in gene therapy development. Methodology: AAV vectors were produced by triple transfection in HEK293T cells. The cell lysate and supernatant were extracted and purified using cesium chloride method and affinity chromatography respectively post 72 hours. The purified protein was denatured and alkylated. Before trypsin digestion, EndoH treatment was given to enrich the glyco proteins. TripleTOF6600 coupled with UPLC was used for protein identification. The molecular dynamics simulation approach was used to assess the protein structure and stability upon modification. identified comparatively a greater number of viral capsid proteins Results: Our approach in affinity chromatography purified supernatants than cell lysates. Moreover, we found lesser host cell protein carry over contamination in affinity purified samples. A total of 4, 12 and 3 novel glycosylation modifications and 8, 14 and 3 phosphorylation modifications were identified for AAV6, AAV8 and AAV9 vectors respectively. Interestingly, our MDS results showed that PTMs on loop 4, loop 6 and loop 8 region of the capsid proteins have functional importance in the capsid proteins. Conclusion: The current data reported novel capsid modifications which can be explored further to understand the biological relevance of modified AAV in transduction, stability and immune evasion. This will aid in development of novel bioengineered AAV vectors for gene therapy applications. Keywords: Adeno-associated virus; capsid; proteomics; LC-MS simulation identification; PTM; Molecular Dynamics 158

160

P-I-0013 The regulatory role of CARP-NF-κB in breast cancer metastasis Penchatr Diskul Na Ayudthaya1,2, Seon-Joo Bae1, Yun-Ui Bae3, Ngu Trinh Van1, Wootae Kim1, Daranee Chokchaichamnankit2, Jisnuson Svasti2, Chantragan Srisomsap2, Seongho Ryu1,4 1Soonchunhyang University, Department of Integrated Biomedical Sciences, Chungcheongnam-do, South Korea 2Chulabhorn Research Institute, Laboratory of Biochemistry, Bangkok, Thailand 3Konkuk University, Konkuk University Medical Center, Seoul, South Korea 4Soonchunhyang University, Department of Pathology, Chungcheongnam-do, South Korea In the year 2023, the survival of breast cancer patients is relatively increase and become the leading cause of cancer death worldwide. The early diagnosis of metastatic breast cancer is usually detected by mass of tumors or clinical examination. Once cancer starts to metastasize, the disease carries the worse prognosis and greater decrease of patient survival. In this study, we performed the experiments by using a microarray to compared low (MCF-7) and high (LM-2) metastatic breast cancer cells. we found the increasing of Cardiac Ankyrin Repeat Protein (CARP) level in LM-2 cells. Then we did both overexpression and knockdown of CARP protein in MCF-7 and LM-2 cells, respectively. The results showed that in CARP-overexpressing cells, the ability of cell migration and invasion was increase, while CARP-knockdown cells were reduced. These studies demonstrated that CARP protein is essential for both cancer migration and invasion. For In vivo studies, we did subcutaneous and tail vain injection of CARP-knockdown and control cells in NOD.SCID mice, the results showed less of cancer metastasis from primary to secondary site after injected CARP-knockdown cells. We performed immunohistochemistry using human breast tissue array and found high grade of tumors expressed high levels of CARP compared to low grade. To elucidate the underlying mechanism of CARP protein, we performed Next-Generation Sequencing, mass spectrometer and immunoblotting in CARP-knockdown and control cells. The results indicated that the NF- κB acts as a downstream of CARP protein. The present studies reveal an aspect of the precise molecular mechanism of CARP protein in cancer metastasis. These findings suggesting that CARP protein might be a candidate biomarker as well as targeted therapeutic for metastatic breast cancer. 160

161

P-I-0014 Light changes promote distinct responses of plastid protein acetylation marks in the model plant Arabidopsis thaliana Juergen Eirich1, Iris Finkemeier1 1IBBP, Muenster, Germany Protein acetylation is a key co- and post-translational modification. How different types of acetylation respond to environmental stress is still unknown. GNAT2, a member of the newly discovered family of plastid acetyltransferases in plants, which is featuring both lysine- and N-terminal acetyltransferase activities, was used to obtain a holistic multi-omics acetylation-dependent view of the acclimation of plants to short-term light changes. We utilized mass spectrometry-based proteomic and acetylation analysis to investigate the plants" responses to changes in light intensity. We grew WT and gnat2 plants under the same conditions and subjected them to high light, darkness, or standard growth conditions for two hours. To analyse the protein abundance and acetylation changes, we used quantitative mass spectrometry, which allowed the quantification of 5,620 protein groups in total. The analysis revealed that both WT and gnat2 plants showed a strong accumulation of heat shock family proteins under high light conditions. The proteomic analysis also showed that two proteins involved in oxidative stress tolerance, were found up-regulated in the WT upon high light stress but not in the gnat2 mutant. The acetylome analysis showed that plastid NTA yield did not significantly change under different light conditions. However, the gnat2 mutant displayed downregulation of transcripts involved in translation-related pathways under darkness, suggesting that GNAT2 might be involved in the light-dependent control of translation. The acetylome analysis revealed that the two different types of acetylations, catalyzed by GNAT2 in the chloroplast, distinctively respond to short-term changes in light conditions. The gnat2 mutant showed a more pronounced de-regulation in the lysine acetylome under high light treatment, with 50 acK sites up-regulated and only nine acK sites also significantly up-regulated in the WT. In conclusion, our study highlights the unique strategies of plant acclimation to different light treatments involving specific protein modifications and suggests that plastid K- and N-terminal acetylations may respond differently to environmental or developmental stimuli. The research provides valuable insights into the plant's responses to changes in light intensity and the interplay between genetic and environmental factors, with a particular focus on mass spectrometry-based proteomic and acetylome analysis. 161

162

P-I-0015 Effect of lysine post-translational modifications in the proteome of extracellular vesicles from non-small cell lung cancer cells Orlando Morales-Tarré1, Magdalena Hernández-Ortiz1, Alberto Paradela2, Fernando Corrales2, Sergio Manuel Encarnación-Guevara1 1Universidad Nacional Autónoma de México, Centro de Ciencias Genómicas, Mexico City, Mexico 2Centro Nacional de Biotecnología, CSIC, Functional Proteomics Laboratory, Madrid, Spain Exosomes, small cell-derived vesicles found in a wide range of biological fluids, play a pivotal role in numerous biological processes. They have been implicated in coagulation, protein degradation, immune signaling, antigen presentation, and cancer metastasis, underscoring their diverse and significant functions. Emerging evidence suggests a potential link between ubiquitylation and protein export to exosomes, with acetylation potentially competing for the ε-amino groups of lysine. Given that enzymes tightly control the abundance of these modifications, it is plausible that the balance of these modifications, in concert with the metabolic state and microenvironment of the cells, governs the presence of specific proteins in exosomes. Importantly, dysregulation of acetylation and ubiquitylation, and the enzymes that govern them, has been implicated in the pathogenesis of non-small cell lung cancer. Our hypothesis posits that the equilibrium of acetylation and ubiquitylation levels is a critical player in regulating protein export to EVs. This process is crucial in modifying vesicular cargo in various non-small cell lung cancer models. the inhibition of deubiquitinases, proteasomal degradation, We are using LC-MS/MS and lysine acetyltransferases, and deacetylases to perform protein quantitation to evaluate and compare the changes in the profiles of intracellular proteins and the different populations of extracellular vesicles, characterizing the proteome and post-translational modifications (acetylation and ubiquitylation) in lysine residues (KPTM) in various NSCLC cell lines, and correlating these modifications to the targeting of proteins towards EVs, and comparing the protein composition of EVs in correlation to intracellular protein profiles in different non-small cell lung cancer models and blood-derived vesicles from cancer patients in different stages of the disease, with the possibility to identify tumor-related markers contained in EVs. In the first stage of the analysis, we have identified 1638 proteins, highlighting proteins participating in vesicle formation, such as tetraspanins CD9, CD81 and CD151, the accessory protein ALIX (PDCD6IP), HGS protein (part of the ESCRT-0 complex), CHMB- 4b and IST1 (which is part of the ESCRT-III complex), and the GTPases Rab-7, Rab-35 and Rab-27, related to endosomal differentiation to multivesicular bodies, and their movement towards the plasma membrane to fuse and release exosomes contained in multivesicular bodies. The results of this project will provide information on how the protein cargo of EVs is conformed during lung cancer development. Understanding this biological process can lead to the identification of biomarkers with diagnostic and prognostic value, and new therapeutic targets to stop metastasis formation. 162

163

P-I-0016 Exploring EGFR and MET cross-talk in NSCLC: a pathway to improved therapy efficacy Dario Frey1,2, Barbara Helm1,2, Florian Schwörer1,2, Simone Clas1, Helge Hass3, Siegfried Hänselmann4, Luisa Schwarzmüller1, Dominic Helm1, Franziska Gödtel1, Yannik Dieter1,2, Yu Qiang5, Katerina Sulková1, Yueyang Xie1,6, Cong Quan Ta1,6, Bernhard Steiert3, Clemens Kreutz3, Marcus Rosenblatt3, Katherina Lauk3, Marc Schneider2,7, Felix Braun4, Thomas Muley2,7, Michael Meister2,7, Laura Klotz2,7, Hauke Winter2,7, Michael Thomas2,7, Andreas Raue8, Karl Rohr5, Dirk-Peter Herten4, Junyan Lu6, Marcel Schilling1, Jens Timmer3,9, Ursula Klingmüller1,2 1German Cancer Research Center (DKFZ), Heidelberg, Germany 2German Center for Lung Research (DZL), Translational Lung Research Center (TLRC), Heidelberg, Germany 3University of Freiburg, Institute of Physics, Freiburg, Germany 4Heidelberg University, Cellnetworks Cluster and Institute of Physical Chemistry, Heidelberg, Germany 5Center for Quantitative Analysis of Molecular and Cellular Biosystems (BioQuant), Heidelberg, Germany 6Heidelberg University, Institute of Computational Biomedicine, Heidelberg, Germany 7Thoraxklinik at University Hospital Heidelberg, Heidelberg, Germany 8University of Augsburg, Department of Computer Science, Augsburg, Germany 9University of Freiburg, BIOSS Centre for Biological Signaling Studies, Freiburg, Germany Non-small cell lung cancer (NSCLC) is a major global cause of cancer-related deaths. Targeted therapies like epidermal growth factor receptor (EGFR) tyrosine kinase inhibitors (TKIs) can extend survival in NSCLC patients, but their effects are temporary. The MET receptor, which binds to hepatocyte growth factor (HGF), has been linked to EGFR-TKI resistance, but the mechanisms behind this are not fully known. We studied how EGF and HGF signals are transmitted in different NSCLC cell lines, which have varying levels and mutations of EGFR and MET. We first used immunoblotting to generate data, which showed that upon co-stimulation with EGF and HGF the receptors" phosphorylation was in one of the cell lines prolonged due to synergistic effects. Mass- spectrometry (MS) based phosphoproteomics using data independent acquisition confirmed this prolonged receptor phosphorylation in certain conditions and revealed further downstream communication, such as prolonged phosphorylation of mitogen-activated protein kinase (MAPK) pathway members. Furthermore, co- stimulation with EGF and HGF resulted in a sustained receptor half-life which influences downstream signaling behavior. We confirmed the hypothesis of receptor interaction using fluorescent live-cell microscopy, retroviral transduction, and siRNA knockdown of the receptors. Our dynamic pathway model indicated that a high receptor ratio could enhance the effectiveness of EGFR-TKIs. We are currently analyzing preliminary tissue samples from NSCLC patients treated with EGFR-TKIs to validate the abundance and phosphorylation of the receptors in primary samples. In conclusion, our findings suggest a strategy to potentially enhance the efficacy of EGFR and MET-targeting treatments. 163

164

P-I-0017 Phosphotyrosine-based phosphoproteomics scaled down to sub-mg needle biopsy level for analysis of individual tumor biology and treatment selection Berend Gagestein1, Alex Henneman1, Thang V. Pham1, Jaco Knol1, Connie Jimenez1 1Amsterdam University Medical Center, Medical Oncology, Amsterdam, Netherlands Introduction: Mass spectrometry-based phosphoproteomics of cancer cell and tumor tissue lysates provides insight in aberrantly activated signaling pathways and potential drug targets. For a comprehensive understanding of individual patient's tumor biology that includes phosphotyrosine signaling and to allow selection of tyrosine kinase inhibitors in individual patients, pTyr-based phosphopeptide enrichment coupled to mass spectrometry is required. However, it needs to be feasible and reproducible with the low protein yields obtained from tumor needle biopsies. Previously, we down-scaled the phosphotyrosine immunoprecipitation to 1 mg, yet most needle biopsies have protein yields that are lower. Here we explore whether further down- scaling of the tyrosine immunoprecipitation is possible and yields reproducible profiles. Methods: To this end, phosphopeptide immunoprecipitation using anti-phosphotyrosine beads was performed using 5, 2, 1 and 0.5 mg protein input from lysates of a mix of colorectal cancer (CRC) tissues that yields similar phosphosite numbers as our previously published CRC cell line reference HCT116. All volumina were linearly scaled or left at 1 mg protein input level. In addition, for low protein input samples, smaller vials and glassware was used. Results: The total number of phosphopeptides captured and detected by LC-MS/MS (Timstof-HT) searched using MQ 2.3.1.0 ranged from 2387 at 5 mg input to 457-604 at 0.5 mg CRC mix input, depending on the condition used. These numbers may be further boosted by match-between-run ID propagation. Reproducibility of 6 replicates at the 0.5 mg protein level was high. Detailed results will be reported along with optimization of MS acquisition and exploration of phosphoproteomics in DIA mode. Conclusions: This study demonstrates the feasibility of label-free pTyr-phosphoproteomics at sub-mg protein level, a protein yield that in our experience may be obtained from ~50% of tumor needle biopsies collected in clinical trials. 164

165

P-I-0018 Deciphering the signaling mechanisms of the C-type lectin receptor Dectin-1 in myeloid cells using high throughput proteomics Rim Abderrahim1, Alexandre Stella1, Karima Chaoui1, Marie Locard-Paulet1, Odile Schiltz1, Olivier Neyrolles1, Yoann Rombouts1, Anne Gonzalez de Peredo1 1CNRS, Institute of Pharmacology and Structural Biology (IPBS), Toulouse, France Dectin-1 is a Pathogen Recognition Receptor expressed on myeloid immune cells, known to play a major role in the antimicrobial immune response. It belongs to the C-type lectin Receptors (CLR) family and can bind to carbohydrate structures, such as β-glucans present on pathogens. This recognition triggers the phosphorylation of key tyrosine residues in the ITAM domain of Dectin-1, leading to the activation of SYK-dependent and independent signaling pathways, and finally to the secretion of pro-inflammatory cytokines and reactive oxygen species. Even though Dectin-1 is an important CLR known to be involved in many diseases such as fungal infections, inflammation and autoimmune diseases, its signaling mechanisms remain unclear. In this project, we used global phosphoproteomics to obtain a time-resolved characterization of the early signaling events taking place in the first minutes following Dectin-1 engagement, and identify key intracellular signaling proteins involved in that pathway. In addition, we established a high-throughput approach for secretome analysis, to monitor in a fast and global manner the inflammatory mediators secreted by immune cells 24 h after Dectin-1 stimulation. To that aim, ER-Hoxb8 immortalized murine cell lines from either WT or Dectin-1 KO progenitors were differentiated into dendritic cells and stimulated with depleted Zymosan, a canonical ligand for Dectin-1, during different time length. Phosphorylated peptides were enriched using a two-step protocol based on TiO2 and phospho-Tyrosine antibody affinity purification, and resulting phosphopeptides were analyzed on an Orbitrap- Exploris instrument equipped with a FAIMS interface. To monitor the secretory phenotype of DCs, a high- throughput approach for secretome analysis was developed using the Evosep chromatographic system and DIA- PASEF analysis on a TIMS-TOF SCP mass spectrometer. Using this analytical pipeline, we could globally detect more than 10,000 class I phosphosites, and identify after statistical analysis more than 300 phosphosites significantly regulated upon Dectin-1 stimulation. These included sites from known proteins of the Dectin-1 signaling pathway, as well as novel targets that were selected for further functional validation by generating the corresponding KO cells. We could also provide a detailed characterization of the DC phenotypic response following Dectin-1 activation, with more than 500 proteins significantly over-represented in the secretome of activated cells, including many canonical inflammatory mediators. The high-throughput secretomic analytical workflow based on the Evosep system and fast DIA-MS runs provides a quick and comprehensive read-out to monitor the phenotypic response of immune cells, allowing the systematic screening of different KO cells in on-going validation studies. 165

166

P-I-0019 Hydrogen sulfide signalling as an additional factor to sex-related differences in ALS progression Anna Percio1,2, Alida Spalloni3, Patrizia Longone3, Andrea Urbani1,2, Viviana Greco1,2 1Fondazione Policlinico Universitario Agostino Gemelli IRCCS, Rome, Italy 2Catholic University of the Sacred Heart, Department of Basic Biotechnological Sciences, Intensivological and Perioperative Clinics, Rome, Italy 3Fondazione Santa Lucia, IRCCS, Rome, Italy, Department of Experimental Neuroscience, Molecular Neurobiology Unit, Rome, Italy Amyotrophic lateral sclerosis (ALS) is a neurodegenerative and devastating disorder characterized by the death of motor neurons in spinal cord, brain stem and motor cortex, with still no convincing therapy. Several etiologic factors, including genetic and environmental ones, can contribute to such neurodegeneration. Despite the underlying causes of motor neuron death remain largely unknown, a scenario of strong oxidative stress (OS) and mitochondrial dysfunction (MD) has long been associated with ALS[1]. Recent evidence points out that this dysregulation may act as real driver of neurodegeneration worthy of greater importance than previously recognized [2]. In this perspective, a disruption of hydrogen sulfide (H₂S) metabolism has been proposed as contributor to ALS pathogenesis. Harmful H2S concentrations have been shown in a mouse model of familial ALS (fALS), SOD1G93A, in primary spinal cord cultures prepared from fALS transgene [3,4], as well as in a cohort of ALS patients compared to other neurological disorders [5]. ALS women exhibit higher levels than men do, as do SOD1G93A female mice compared to males, with increased production correlating to disease progression from early to late symptomatic stages [6].The mechanisms driving this alteration is still elusive. Therefore, the overall goal of the current study is to shed light on this still hidden H2S-ALS crosstalk, particularly by examining gender differences in ALS. Label free proteomics analysis was performed on spinal cord tissues (total extracts and mitochondrial enriched fractions) prepared from female and male ALS SOD1G93A and SODWT mice. Dedicated mass spectrometry- based proteomics analysis have been developed to quantify the expression of enzymes involved in both H2S synthesis (CBS, CSE, 3MST) and catabolism (ETHE1 and SQR) to elucidate the origin of such alteration in ALS. In addition, the enrichment and mapping of cysteine-persulfides was performed.Peptide identification and quantification were carried out by Peaks and MaxQuant softwares respectively; bioinformatics analysis was performed by Ingenuity Pathway analysis (Quiagen). An altered expression of H2S producing enzymes was shown in ALS tissues, particularly in female groups. Moreover, site-specific identification and quantification of cysteine persulfides highlighted a different profile in SOD1G93A compared to SODWT especially for those proteins related to mitochondrial quality control complexes. This investigation provides a comprehensive mapping of the redox scenario in ALS, revealing hitherto unexplored molecular features. References: [1] Greco, V. et al. In Mitochondria in Health and in Sickness; Springer, 2019; pp. 71–82. [2] Jagaraj, C. et al Front. Cell. Neurosci., 2021 [3] Davoli, A.; Greco, V. et al. Ann. Neurol. 2015, 77, 697–709. [4] Greco, V. et al. Antioxidants 2018, 7, 87. [5] Greco, V. et al. Metabolites 2021, 11 [6] Spalloni, A.; Greco, V.; et al. Int. J. Mol. Sci. 2019, 20, 2550. 166

167

P-I-0020 A targeted proteomic approach to assess the presence of mouse-specific histone H3 variants in brain and testis Hassan Hijazi1, Delphine Pflieger1, Sabine Brugière1, Marie Courçon1, Anne-Marie Hesse1, Christophe Bruley1 1Exploring the Dynamics of Proteomes (EDyP), University Grenoble Alpes, CEA, Inserm, IRIG-BGE, Grenoble, France Histones are proteins in charge of DNA packaging and compaction in eukaryotic nuclei. Histone post- translational modifications (hPTM) serve as epigenetic knobs by which eukaryotic cells fine-tune DNA-templated processes such as replication, transcription and DNA damage repair. Dysregulation of hPTMs has been linked to disease states such as cancer and neurodegenerative disorders. Classical hPTMs include acetylation and mono-, di- and tri-methylation of lysine residues. Over the past 15 years, structures have been progressively discovered that can also modify lysines: they resemble acetylation but vary in length and hydrophobicity and are collectively called acylations. The big question that has emerged is whether they play specific or only redundant roles compared to acetylation. Besides, in mouse, several sequence variants of histone H3, which differ by a handful of amino acids scattered over the sequence, have been recently described. Some specific functions have just started being attributed to a few of them. Screening for many combinations of biologically relevant lysine acylations and methylations on several H3 variants leads to numerous isomeric peptides that are strictly isobaric. Mass spectrometry (MS) has proven crucial for identification and quantification of histones and histone PTMs. Nonetheless, data processing and interpretation of histone mass spectra is still challenging due to the absence of a decoy strategy for database search and the presence of ample isobaric peptides that require manual inspection to validate their presence. In this study, we focused on identification and quantification of the variably modified K27-R40 stretch in canonical H3, H3.3 and two mouse-specific H3 variants (H3mm7 and H3mm13) by targeted proteomic analysis using Parallel Reaction Monitoring (PRM). To this end, we built and refined a DDA spectral library through: (1) selecting a suitable preprocessing algorithm, (2) validating PTMs by looking at informative ions in the low-mass region and (3) utilizing synthetic peptides and Indexed Retention Time (iRT) peptides to increase identification confidence. We used PRM to exclusively target variant × PTM combinations of the stretch K27-R40 of histone H3 proteins extracted from mouse testis and brain tissues. None of H3mm7 and H3mm13 variably modified H3 K27- R40 peptides were identified except the unmodified form of H3mm7. While tentatively identified in DDA analysis, these variants are easily confused with other variably modified peptidoforms from canonical H3 and H3.3, which indicates the necessity of extra measures to confirm identifications. In all, histone PTM confident characterization requires validating multiple aspects including the retention time, the presence of diagnostic ions coming from modified lysines in the low-mass range and their relative intensities, in addition to the high- mass y fragments classically considered for reading amino acid sequence. 167

169

P-I-0021 Single-shot tyrosine phosphoproteomics through FAIMS-DIA, and FAIMS-PRM mass-spectrometry afterwards, in revealing the signaling networks associated with BTKi-mediated drug mechanism of action Mahmud Hossain1, Gregory Wirak1, Dimitry Ofengeim1, Dhiman Ghosh1, Bailin Zhang1 1Sanofi, Cambridge, MA, United States One of the compelling needs in drug discovery is to connect cellular responses to pharmacological substance with that of drug mechanism of action (MOA). While genomic signatures provide an extensive view of genome- wide alterations, there are genome-independent mechanisms, such as posttranslational modifications (PTMs), that cannot be revealed through genomics. Besides, drugs that target cellular proteins which are interconnected and integrated with various biochemical pathways can be perturbed by those. Bruton"s tyrosine kinase inhibition (BTKi) shows therapeutic promise for the treatment of multiple sclerosis (MS). Potential mode of action that contribute to this efficacy includes the effects on both B lymphocytes and macrophage signaling cascades, as both cell types contribute to MS pathology and employ BTK signaling. Towards better understand the biology of potential therapeutic mechanisms and to identify putative drug targets, it is advantageous to detect protein substrates that are phosphorylated by BTK in these cell-types under immune-responsive conditions. To attain a deeper proteome coverage and to avoid common issues such as low sensitivity, throughput, or reproducibility that are associated with conventional DDA-based mass spectrometry approaches, here we present label-free single-shot FAIMS-DIA mass spectrometry platform in exploring phospho-tyrosine networks of proteins downstream of BTK signaling, and subsequent validation by FAIMS-PRM mass spectrometry. Human B lymphocytes (Ramos RA1 cell line) and THP-1-derived macrophages were immunologically stimulated by goat anti-human IgM F(ab")2 fragment and Fc OxyBURST, respectively, and with or without the presence of the BTK inhibitor tolebrutinib, and then perform tyrosine phosphoproteomics on the cell lysates. Together with ion-mobility and gas-phase ion separation at high resolution, the current phosphoproteomics platform can increase the sensitivity of detection, reducing chemical noises, improving the depth of phosphoprotein coverage and phospho-sites localization accurately. Besides, being label-independent, the FAIMS-DIA platform increases the throughput in sample processing and improves quantification, followed by multiplexed FAIMS-PRM for targeted assay. Using tool compound targeting BTK (BTKi), we observed distinct downstream phospho-tyrosine signaling through FAIMS-DIA mass spectrometry in activated B cell and macrophages that highlights the organization of intrinsic cellular complexities of protein networks in achieving a desired pharmacological effect. A subset of phosphopeptides from these phospho-tyrosine networks has been further validated through targeted FAIMS- PRM and will be presented. Comparison of tyrosine phosphoproteomics with and without BTKi in stimulated human B lymphocytes and macrophages will identify potential downstream targets of BTK in both cell types that will eventually provide strategies for an effective therapeutic development of MS. 169

170

P-I-0022 Redox proteome analysis of rat liver: gender-specific effects to diet and aging Isabel Huber1,2, Theresa Hofmeister3, Michal Rackiewicz1, Sorel Sagu3, Kristine Wolf1,4, Jens Raila1, Matt Fuszard5, Katja Witzel6, Harshadrai Rawel1, Andrea Henze3,1 1University of Potsdam, Institute of Nutritional Science, Potsdam, Germany 2Health and Medical University, Anatomy, Potsdam, Germany 3Martin-Luther-University Halle-Wittenberg, Nutrition Physiology, Halle (Saale), Germany 4Brandenburg Medical School Theodor Fontane, Brandenburg, Germany 5Martin-Luther-University Halle-Wittenberg, Core Facility Proteomic Mass Spectrometry, Halle (Saale), Germany 6Leibniz Institute of Vegetable and Cornamental Crops, Research Group Plant-Microbes, Grossbeeren, Germany Increasing life expectancy combined with an unbalanced diet in Western societies promotes the development of diseases such as obesity, type 2 diabetes mellitus and non-alcoholic fatty liver disease. One potential molecular cause for the progression of age- and diet-related diseases is an imbalance in redox homeostasis favouring oxidative processes. Understanding the redox proteome is crucial for the selective modulation of the redox sensor system for therapeutic approaches. This study systematically analysed the impact of diet and aging on the redox proteome of the liver as a central metabolic organ. Therefore, 84 Wistar ats (Rattus norvegicus) of both sexes were fed a moderate Western diet (WD) mimicking dietary patterns in central Europe (WD, 36 kcal % fat, 13 % sugar) or a standard diet (SD, 10 kcal % fat) for up to 78 weeks. Various in vivo tests (glucose tolerance, activity, blood pressure) were conducted, and liver tissues were collected at defined time points (26, 52 and 78 weeks of age) for further analysis. Changes in cysteine redox status may reflect alterations in redox signalling pathways. In this study, non-oxidized cysteine peptides were labelled using N-ethylmaleimide (NEM), providing an indicator of cysteine oxidation status. A decrease in relative protein levels of NEM-sites indicates an increase in the oxidized status of the proteins. Using sequential alkylation and modified C-STrap sample preparation, the samples were fractionated after tryptic digestion into a non-cysteine fraction containing NEM-labelled non-oxidized cysteine peptides as well as cysteine-free peptides and a N-propylmaleimide (NPM)-labelled oxidised cysteine fraction. Following TMT-labelling the samples were analysed using high-resolution LC-MS/MS (Orbitrap). Analysis of NEM-sites revealed 84 and 78 significantly regulated redox-sites for male and female rats, respectively, differing between WD and SD. In females, NEM-sites involved in pyruvate and lipid metabolism were enriched during pathway analysis. In males, pathway enrichment highlighted proteins involved in purine and pyrimidine metabolism and energy homeostasis due to altered redox profiles as a consequence of WD. Age-related analysis of NEM-sites showed a higher number of regulated proteins in males, with 16 sites compared to 7 in females. For instance, males exhibited among other observations a reduction in NEM-sites of glutathione S-transferases and increased levels of thioredoxin (TXN), suggesting increased oxidation levels with age and potential changes in redox signalling pathways, respectively. TXN's role in regulating transcription factors and protecting against oxidative stress may be affected by changes in its oxidation status. These findings provide insights into early metabolic changes and adaptation mechanisms in rats under a moderate WD, with a notable sex-specific liver proteome profile that may explain differences in prevalence and manifestation of diet- and age-related pathologies. 170

171

P-I-0023 A pharmacological dimension to personalised phosphoproteomics of human skeletal muscle Hannah Huckstep1, Magnus Leandersson2, Johan Onslev2, Nicolai Henriksen2, Jonas Kristensen2, Magnus Zankel2, Jesper Birk2, Rasmus Kjøbsted2, Lina Le1, James Burgess1, Jørgen Wojtaszewski2, Sean Humphrey1 1Murdoch Childrens Research Institute, Functional Phosphoproteomics, Parkville, Australia 2University of Copenhagen, Department of Nutrition, Copenhagen, Denmark Recent studies by our group suggest that the kinase mTORC1 (the mammalian target of rapamycin) may regulate changes to insulin sensitivity in skeletal muscle following exercise. To investigate this directly in humans we extended an approach recently developed by our group ("personalised phosphoproteomics") to include a pharmaco-dimension through the addition of treatment with the mTORC1 inhibitor Rapamycin. We recruited 13 healthy human subjects and administered a single dose of Rapamycin using a double blinded placebo crossover study design. Subjects performed one-legged exercise for 2 hours, and 4 hours post-exercise a 2 hour hyperinsulinemic-euglycemic clamp was performed. Muscle biopsies from m. vastus lateralis were obtained immediately prior to the clamp in both the rested and exercised leg, and at the conclusion of the clamp. Frequent blood samples from catheterised femoral and arterial veins were additionally taken, concomitantly with blood flow measurements, enabling the direct measure of glucose utilization by the muscle bed. Our global phosphoproteomics analysis quantified over 60,000 Class I phosphopeptides in human skeletal muscle tissue, and through comprehensive mass spectrometry-based analyses we identified and quantified phosphosites within the mTOR pathway across eight different conditions in 13 subjects (104 biopsies in total). As we observed previously, skeletal muscle phosphoproteomes exhibited high inter-subject variation, however our paired experimental design enabled us to uncover numerous novel phosphorylation events modulated by mTORC1 and its downstream effectors. Ongoing investigation into these phosphosites and their potential kinases are shedding light on the complex interplay between mTOR, insulin and exercise at an individual and group level. 171

172

P-I-0024 Analysing non-proteinaceous ubiquitination via NoPro-clipping Marco Jochem1,2, Simon Cobbold1, Anthony Cerra1, Tahnee Saunders1, David Komander1,2 1Walter and Eliza Hall Institute (WEHI), Ubiquitin Signaling Division, Parkville, Australia 2University of Melbourne, Department of Medical Biology, Melbourne, Australia The small protein ubiquitin dictates the cellular fate of other proteins via covalent modification and subsequent degradation by proteasomal or lysosomal pathways. But ubiquitin signalling has consequences beyond degradation and is involved in almost every cellular process. Consequently, it plays a role in most human diseases, including various cancers, infections, or neurodegenerative and autoimmune diseases. Recent studies have described the ubiquitination of lipopolysaccharides (LPS), sugars, lipids and nucleic acids, expanding the role of ubiquitin modifications beyond proteins. However, a lack of suitable methods and reagents prevents reliable detection of these species and hinders us from understanding their true cellular significance. Here we present NoPro-clipping, a mass spectrometry (MS)-based method for the unbiased detection of non- proteinaceous ubiquitination (see Figure). A key component of NoPro-clipping is the use of viral and bacterial proteases, referred to as "clippases", which cleave ubiquitin at its C-terminus and leave characteristic GlyGly marks on ubiquitinated substrates. Subsequently, we use molecular weight filtration to enrich GlyGly-modified small molecules, including non-proteins, and analyse them by ammonium acetate/acetonitrile-based microflow liquid chromatography (LC)-MS/MS. We further increase the sensitivity of NoPro-clipping by enzymatically attaching a small peptide onto GlyGly-modified molecules. This transformative step renders non-proteinaceous metabolite samples peptidic, allowing the subsequent use of conventional proteomics workflows for further enrichment and nanoflow LC-MS/MS analyses. NoPro-clipping has been validated with a variety of in- vitro ubiquitinated sugars (e.g., Maltose, Ribose, GlcNAc) and is compatible with human cell lines and tissue samples. It can be applied in both targeted and untargeted studies, facilitating the detection of known and novel ubiquitinated non-proteins. To unravel the underlying mechanisms of various human and horse glycogen storage diseases (e.g., McArdle's disease, Lafora disease, equine polysaccharide storage myopathy), we established liver cell lines in which we can observe and monitor the formation of ubiquitinated glycogen bodies via microscopy. Following cell lysis, α- amylase digestion, and NoPro-clipping, we successfully detected multiple ubiquitinated maltose species (e.g., GG-maltose, GG-maltotriose, GG-maltotetraose), demonstrating that NoPro-clipping is capable of detecting ubiquitinated glycogen, which, to our knowledge, has not been previously reported in vivo. Fig. 1 (optional) 172

173

P-I-0025 Accelerating single-shot phosphoproteomics by identification and quantitation of > 100,000 phosphosites in human cell lines Claire Koenig1, Hayoung Cho1, Pierre Sabatier1, Ana Martinez-Val2, Jesper Olsen1 1Center for Protein Research, University of Copenhagen, Copenhagen, Denmark 2Centro Nacional de Investigaciones Cardiovasculares Carlos III (CNIC), Madrid, Spain Site-specific protein phosphorylation is a highly dynamic post-translational modification that influences protein activity and function. With the advances in automated phosphopeptide enrichment and rapid MS analysis, the current bottleneck in phosphoproteomics workflows is the identification and quantification of phosphopeptides. With the generation of large phosphoproteomics raw files (>10Gb), using a spectral library-free approach for phosphopeptide identification present a significant challenge. In Spectronaut, search times can be reduced up to four-fold using a dedicated spectral library, however, the resulting depth of coverage using a spectral library is constrained by the library's quality and comprehensiveness. Here, we used a wide panel of human cell lines, with and without stimulation, for single-shot phosphoproteomics analysis using an Evosep One LC system coupled to the Orbitrap Astral MS using narrow window data-independent acquisition (nDIA). Cell lysates were processed using PAC-based on-bead LysC/trypsin digestion followed by on-beads (Zr-IMAC HP) phosphopeptide enrichment in the KingFisher platform. From most cell lines including unstimulated HeLa, we quantified over 50,000 localized phosphosites (class I) with a median coefficient of variation below 20% with online LC gradients of 44 minutes (30SPD). The generated data was searched using Spectronaut for the generation of a comprehensive and high quality phosphopeptide library containing over 100,000 localized phosphosites (class I). This library was then benchmarked against a spectral library-free approach on a separate sample set. For biological insights, phosphorylation sites were quantified between the different cell line conditions to elucidate activated signaling pathways. This project aims at addressing the challenge imposed by search time of phosphoproteomics data on the Orbitrap Astral MS by providing the largest experimental human phosphopeptide library. This resource reduces computational analysis time, enabling large-scale phosphoproteomics studies without compromising depth or requiring project-specific spectral libraries. 173

174

P-I-0026 Comparison of phosphopeptide enrichment and LC-MS/MS acquisition methods to study cancer phosphoproteome Petr Lapcik1, Damian Liptak1, David Potěšil2, Zbynek Zdrahal2, Pavel Bouchal1 1Department of Biochemistry, Faculty of Science, Masaryk University, Brno, Czech Republic 2Central European Institute of Technology, Masaryk University, Brno, Czech Republic Phosphorylation, a ubiquitous mechanism of cellular signaling, is one of the most common post-translational modifications involved in regulation of cellular processes. Dysregulation of kinase pathways is associated with various cancers including renal cell carcinoma (RCC). As number of mass spectrometry-based strategies to study phosphoproteome was developed, we aim to compare widely used strategies of phosphopeptide enrichment and data acquisition to gain as deep as possible insight in changes of phosphoproteome in two RCC cell lines, 786-0 and Caki-1, after treatment with sunitinib, a potent tyrosine kinase inhibitor that targets VEGF and PDGF receptors and is commonly used for RCC treatment. In our work we compare i) the use of two phosphopeptide enrichment methods based on Fe-NTA and TiO2 performance, (ii) data-dependent acquisition (DDA) and data- independent acquisition (DIA) methods for phosphoproteome analysis using timsTOF Pro 2 mass spectrometer (Bruker), iii) the outputs from the DIA data processing software Spectronaut 18.5 and DIA-NN 1.8.1, and iv) deregulation of kinase pathways in 786-0 and Caki-1 cell lines in response to the sunitinib treatment. In total, use of DIA strategy with data processing in Spectronaut 18.5 reached higher phosphoproteome coverage with 9,419 identified phosphopeptides compared to 6,007 with the data acquisition in DDA mode. Within the DIA dataset, 10.6% more phosphopeptides were identified in phosphopeptide samples enriched by Fe-NTA than in the samples enriched by TiO2. The number of phosphopeptides identified exclusively using Fe-NTA enrichment was five times higher than those identified exclusively after TiO2 enrichment. The number of identified tyrosine- phosphorylated peptides was comparable between enrichment methods, with only 2.5% more phosphopeptides identified with Fe-NTA than with TiO2 enrichment. Nevertheless, 86% of phosphopeptides was identified using both enrichment methods across both cell lines. Analysis of DIA data indicated decreased phosphorylations of key substrates of kinases involved in the VEGF signaling pathway, such as MAPK14, MAPK1, PTK2, and AKT1 after sunitinib treatment. Simultaneously, increased phosphorylation of PTK2 kinase and an increased abundance of AXL kinase were observed. In a summary, this work compared various methodological approaches for phosphoproteome analysis, highlighted the benefits of DIA acquisition for analysis of phosphopeptides and showed a deeper coverage of RCC phosphoproteome with Fe-NTA enrichment. Supported by Ministry of Health of the Czech Republic (project NU22-08-00230), all rights reserved, by MEYS CR infrastructure project LM2023042 and European Regional Development Fund-Project "UP CIISB" (No. CZ.02.1.01/0.0/0.0/18_046/0015974), and by the project National Institute for Cancer Research (Programme EXCELES, ID Project No. LX22NPO5102)—Funded by the European Union—Next Generation EU. 174

175

P-I-0027 Posttranslational modification of angiotensin II by pyridoxal-5'-phosphate as a new drug approach against high blood pressure Michaela Lellig1, Juan Muñoz-Castañeda2, Juliane Hermann1, Heidi Noels1, Mariano Rodríguez2, Joachim Jankowski1, Vera Jankowski1 1University Hospital RWTH Aachen, Institute for Molecular Cardiovascular Research (IMCAR), Aachen, Germany 2Maimonides Institute for Biomedical Research (IMIBIC), Córdoba, Spain Introduction: Although hypertension is a worldwide health problem, treatment options in low-income countries are still limited. This results in high mortality. Therefore, the development of new therapeutic drugs that are affordable and accessible to everyone is still an urgent need. Since angiotensin II is one of the essential vasoconstrictive peptides in the human organism, we analysed the impact of its posttranslational modification to pyruvamide-angiotensin II by pyridoxal-5'-phosphate on blood pressure. Pyridoxal-5'-phosphate is a less expensive vitamin B6 derivative and therefore could be a novel, cost-effective drug for the treatment of hypertension. Methods: The posttranslational modification of angiotensin II to pyruvamide-angiotensin II by pyridoxal-5'- phosphate was investigated in vitro by mass spectrometry. Calcium ion influx into vascular smooth muscle cells treated with angiotensin II or pyruvamide-angiotensin II, respectively, was also investigated in vitro. In ex vivo experiments, the vasoconstrictive effect of angiotensin II and pyruvamide-angiotensin II, respectively, was analysed by using the bioassay of the isolated perfused rat kidneys. Furthermore, angiotensin II and pyruvamide- angiotensin II were investigated in in vivo experiments. Both, Wistar Kyoto rats and spontaneously hypertensive rats were treated with pyridoxal-5'-phosphate by using mini-osmotic pumps, and the blood pressure was measured time-dependently. Results: Angiotensin II, incubated with pyridoxal-5'-phosphate, was posttranslationally modified to pyruvamide- angiotensin II. In vascular smooth muscle cells, the calcium ion influx after stimulation with pyruvamide- angiotensin II was significantly decreased compared to angiotensin II. The perfusion pressure of isolated perfused rat kidney increased less by pyruvamide-angiotensin II than by unmodified angiotensin II. The systolic and diastolic blood pressure of spontaneously hypertensive rats treated with pyridoxal-5'-phosphate decreased after three days of treatment significantly. The blood pressure of Wistar Kyoto rats treated with angiotensin II increased to hypertensive values, while the blood pressure of Wistar Kyoto rats co-treated with angiotensin II and pyridoxal-5'-phosphate was not increased. Conclusions: Pyridoxal-5'-phosphate significantly decreased blood pressure and might be a cost-effective drug for hypertension treatment. Respectively, an increased pyridoxal-5'-phosphate intake recommendation in the form of vitamin B6 intake might have a decreasing effect on blood pressure in the hypertensive population. 175

176

P-I-0028 Deciphering the role of ADP-ribosylation in CardioLaminopathy Deena Leslie Pedrioli1, Nicolas Vignier2, Antoine Muchir2, Michael Hottiger1 1University of Zürich, Department of Molecular Mechanisms of Disease, Zurich, Switzerland 2Sorbonne Université, Paris, France Protein ADP-ribosylation is a physiologically and pathologically important post-translational modification (PTM) that regulates numerous cellular activities. ADP-ribosylation quickly alters the cellular states under acute stress conditions and disease pathologies. Laminopathies are a group of disorders caused by mutations in the LMNA gene, which encodes A-type lamins, components of the nuclear lamina. Recent studies characterizing striated muscle laminopathies have provided novel perspectives on the role of the nuclear lamina and the pathological consequences of altered nuclear lamina. Interestingly, studies have revealed that nicotinamide adenine dinucleotide (NAD+) salvage pathway was altered in the cardiac tissues carrying LMNA mutations, which drastically reduced cellular NAD+ levels. It is interesting to speculate that these changes in NAD+ availability could induce global ADP-ribosylome changes in striated muscle laminopathies that affect cardiac function. To address this hypothesis, we defined the ADP-ribosylomes of heart tissues isolated from an autosomal-recessive Lmna- H222P/H222P mouse model of Emery Dreifuss Muscular Dystrophy (EDMD), a striated muscle laminopathy, in which altered NAD+ homeostasis is thought to contribute cardiac dysfunction. Compilation of the individual WT and Lmna-H222P/H222P ADP-ribosylomes using our well-established MS-based ADP-ribosylome revealed that the Lmna- H222P/H222P mutation reduced ADP-ribosylation in mouse hearts. STRING analysis of the 291 ADP-ribosylated proteins identified in the WT mouse hearts revealed a specific enrichment in proteins required for mitochondrial and muscle contraction functions. Importantly, we also discovered that the Lmna-H222P/H222P mutation specifically reduced the modification of mitochondrial, cell membrane, and secreted proteins. Collectively, these studies provide important steps forward for the rapidly developing field of ADP-ribosylation research and have provided novel mechanistic insights into the role of ADP-ribosylation in diverse biological samples under both healthy and disease conditions. 176

177

P-I-0029 Secretome and degradome analysis reveal effects of Kallikrein-related peptidase 6 on ECM arrangement and immune recognition in PDAC Mujia Li1,2,3, Miguel Cosenza-Contreras1,2,4, Bettina Wehrle1,2, Oliver Schilling1,2 1Uniklinik Freiburg, Freiburg, Germany 2Uniklinik Freiburg, Institute of Clinical Pathology, Freiburg, Germany 3University of Freiburg, Pharmaceutical Sciences, Freiburg, Germany 4University of Freiburg, Biology, Freiburg, Germany Kallikrein-related peptidase 6 (KLK6) is a secreted serine protease that is involved in inflammatory pathways in skin and central nervous system. Additionally, its overexpression was detected in several tumor entities including pancreatic ductal adenocarcinoma (PDAC), thereby rendering it as a potential cancer biomarker candidate. However, distinct substrates of KLK6 are only sparsely identified and the extent of its pleiotropic biological effects in PDAC remains to be fully understood. This study aims to elucidate biological mechanisms of KLK6 in a MiaPaCa2 knockdown model. For that, cell conditioned medium was generated and a mass spectrometry-based secretome and degradome analysis was performed. By using an isobaric labeling approach (TMT 16-plex), a quantitative analysis and comparison between control and KLK6 knockdown condition was achieved. The secretome analysis revealed over 1400 proteins classified as “extracellular space” or “secreted” by Gene Ontology (GO). An upregulation of the TGF-receptor binding proteins TGFB2 and THBS1 could be detected in the knockdown condition, aligning with previously reported effects of KLKs on the TGFβ-signalling pathway. Effects on the extracellular matrix (ECM) assembly by KLK6 can be suggested due to the upregulation of several ECM structure proteins such as VCAN, FN1, SPON1, COL3A1, COL5A2 and COL6A1. On the opposite, COL1A2 and PCOLCE2 were found downregulated, indicating an effect on ECM structure assembly or a compensatory feedback mechanism. A downregulation of MHC class I (HLA-A, HLA-B, HLA-C, HLA-E) and a higher abundance of secreted MICA and MICB indicate an influence on immune escape and NK-cell and T-cell recognition, caused by a suppression of expression of the immunoproteasome (PSMB8, PSMB9, PSMB10). Interestingly, many interferon- stimulated genes were found downregulated in the KLK6 knockdown condition, suggesting an effect on IFN signaling by KLK6. For the degradome analysis, neo-N-termini were labelled with TMT 16-plex before tryptic digestion and analyzed by LC-MS/MS. The analysis revealed 2871 neo-N-termini classified as secreted by GO, providing a deeper insight into the effects of KLK6 on ECM remodelling: Differential processing of the C-terminal sides of COL18A1, COL3A1 and COL1A2 as well as LOXL2 highlight the biological importance of proteolytic processing as a posttranslational modification and the value of mass spectrometry-based proteomics. Altogether, this study represents a first mass spectrometry-based secretome and degradome analysis to unravel the biological effects of KLK6 in PDAC. Given that KLK6, along with other KLKs, is already considered as potential biomarker or therapeutic target, our findings promise to furnish crucial and valuable insights into the physiological mechanisms that could be influenced by KLK6 inhibition. Acknowledgments The Schilling laboratory acknowledges funding by the Deutsche Forschungsgemeinschaft SFB1479-Oncoescape (project ID: 441891347) 177

178

P-I-0030 SICyLIA-TMT: a multi-step labelling strategy for global redox proteomics Sergio Lilla1, Samuel Atkinson1,2, Jiska van der Reest1,3, Joanna Kirkpatrick4, Sonja Radau5, Atul Shahaji Deshmukh6, Eyal Gottlieb7, Sara Zanivan1 1Beatson institute for cancer research, Glasgow, United Kingdom 2Institute of Medical Sciences, University of Aberdeen, Aberdeen, United Kingdom 3KU Leuven, Center for Cancer Biology, Leuven, Belgium 4Thermo Fisher Scientific, Huntingdon, United Kingdom 5Thermo Fisher Scientific, Dreieich, Germany 6Novo Nordisk Foundation Center for Basic Metabolic Research, University of Copenhagen, Copenhagen, Denmark 7MD Anderson Cancer Center, University of Texas, Houston, TX, United States High levels of reactive oxygen species can cause oxidation of different classes of molecules causing profound consequences on cellular functions. ROS have also been shown to be a signal intermediates that modulate enzyme activity and trigger signalling cascades. Notably, Cysteines are an important stress sensor that swiftly respond to external oxidative stimuli. We have previously developed SICyLIA, a method for accurate measurement of Cysteine oxidation levels in cells and tissues using light or stable-isotope labelled, heavy, iodoacetamide (IAA) as readout for cysteine thiol oxidation. Here we further developed the method, taking advantage of TMT technology (SICyLIA-TMT). SICyLIA-TMT improved the throughput and additionally enabled the measurement of reversibly oxidised cysteines and total protein levels within the same sample. Using SICyLIA- TMT, we also reduce by half both sample preparation and analysis at the mass spectrometry. To swiftly block free thiols, we add light IAA to lysis buffer. Reversibly-oxidised Cysteines are then reduced, and newly generated thiols blocked with heavy IAA. This creates two labelled populations of free and reversibly- oxidised thiols, which are further labelled with TMT, fractionated and analysed by mass spectrometry. In normal conditions, high levels of intracellular Glutathione maintain the redox balance. This keeps the quantity of reversibly oxidised thiol to a minimum level, which results in making their detection very difficult. To overcome this issue, we use one of the TMT channels for a heavy Iodoacetamide-labelled sample, to act as "carrier channel" to boost the signal of the low-level reversibly oxidised peptides. We benchmarked the SICyLIA-TMT method against the same samples analysed with the original SICyLIA method. SICyLIA-TMT outperformed the old method in respect to the number of peptides robustly quantified (>9,000 with SICyLIA; >12,000 light and >1,000 heavy with SICyLIA-TMT). Depth of the redox proteome was also evaluated using a new generation mass spectrometer, Thermo Astral. Finally, SICyLIA-TMT can determine the oxidation state of both free and reversibly-oxidised thiol fraction. . This enables to calculate the oxidation stoichiometry of each Cysteine-, under the assumption that, for each Cysteine residue, the summed intensities of free and reversible oxidised thiol fraction represent 100% of the residue. In conclusion, we have developed SICyLIA-TMT as a new simple method for proteome-wide analysis of oxidative signalling and proteome dynamics. 178

179

P-I-0031 A dynamic interaction landscape of the mammalian Hh signaling pathway Xiaonan Liu1, Kari Salokas2,1, Markku Varjosalo2,1 1University of Helsinki, HiLIFE, Helsinki Institute of Life Science, Helsinki, Finland 2University of Helsinki, Institute of Biotechnology, Helsinki, Finland Hedgehog (Hh) signaling transduction is mediated by consecutive protein-protein interactions (PPIs). To comprehensively map the interaction network of the mammalian Hh pathway, we employed a proximity-based labeling method, BioID. Initially, we mapped the dynamic PPI network surrounding the core components in the absence and presence of the Hh ligand in 3T3 cells. Our comparative analysis of PPIs revealed that the network recapitulates many known ciliary proteins, underscoring the essential role of primary cilia in Hh signaling. We then characterized the interaction network at different stages of cilium progression by monitoring PPIs under conditions of cilia absence, ciliogenesis, and cilia resorption (Figure 1). The significant overlap of interactions among these conditions indicated that proteins could regulate Hh signaling outside the primary cilia. Additionally, we applied quantitative phosphoproteomics to elucidate the phosphorylation dynamics that drive pathway activation in 3T3 cells. Our phosphoproteomic profiling suggests a central role for phosphorylation in the regulation of ciliary assembly, trafficking, and Hh signal transduction. Moreover, distinct and novel phosphorylation events were observed in key Hh signaling components, including SMO, SUFU, PTCH, GLI2, and GLI3. This study provides a comprehensive proteomic and phosphoproteomic landscape of the mammalian Hh signaling pathway, offering new insights into its regulation and function. Fig. 1 (optional) 179

180

P-I-0032 Delineating organ phosphoproteomic landscapes and deregulated metabolic associated signaling under dietary intervention Ziliang Ma1,2, Eduardo Hideo Gilglioni3,4, Fujia Wang1, Esteban N. Gurzov3,4, Wei Wu1,2,5 1Utrecht University, Utrecht, Netherlands 2Singapore Immunology Network (SIgN), Agency for Science, Technology and Research (A*STAR), Singapore, Singapore 3Université libre de Bruxelles, Brussels, Belgium 4WEL Research Institute, Wavre, Belgium 5National University of Singapore, Singapore, Singapore Mammalian organs are biochemically optimized for their unique physiological roles through tightly regulated, phosphorylation-dependent intracellular signaling, impacting human health. Overconsumption of prevalent high- fat, high-fructose, high-cholesterol (HFHFHC) diet often leads to obesity and metabolic syndrome in organs, which typically stems from dysregulated phosphorylation events in signaling pathways. To better understand the influence of HFHFHC diet intake on organs in living organisms, we established a comprehensive mouse organ- specific phosphoproteome atlas quantified across 11 mouse organs and tissues. The atlas harbors 34,551 Class I phosphosites from 6,306 phosphoproteins, inclusive of 5,364 novel phosphosites that were not reported in previous mouse phosphoproteome libraries. Through deep phosphoproteome analysis, we profiled different phosphorylation patterns in metabolic and non-metabolic organs related to their physiological functions and the deregulated signaling pathways under HFHFHC dietary conditions. Tyrosine phosphorylation is a rapid and rare protein modification controlled by kinase and phosphatase. We built up a receptor-type protein tyrosine phosphatase κ (PTPRK) knockout mouse model to investigate the regulatory mechanism of phosphatase in metabolic reprogramming. With these, we found most signaling interference is caused by HFHFHC diet-induced reactive oxygen species (ROS), and mTOR/AMPK signaling was enriched in most organs indicating a pro-survival signal in early adulthood mice. This rich dataset is complementary to the current phosphoproteome online resource. More importantly, we elucidated the influence of diet predispositions in organisms in early adulthood and comprehended molecular mechanisms underlying metabolic disease signaling pathways under dietary intervention. 180

181

P-I-0033 Comparative quantitative proteome and phosphoproteome analysis of castor bean (Ricinus communis L.) seed endosperm at different imbibition stages Yara Martins da Silva1, Wassali V. Sousa2, Francisco A. P. Campos3, Gilberto B. Domont1, Fabio Cesar Sousa Nogueira1 1Federal University of Rio de Janeiro, Biochemistry, Rio de Janeiro, Brazil 2Federal University of Rio de Janeiro, Genetics, Rio de Janeiro, Brazil 3Federal University of Ceará, Molecular Biology and Biochemistry, Fortaleza, Brazil Castor bean (Ricinus communis L.) stands out as one of the world's major oilseed crops, recognized for its role in studying fatty acid metabolism with biotechnological applications. Its seeds not only yield oil rich in triricinoleate, valuable across various industrial sectors and in health, but they also store significant amounts of allergenic proteins and harmful toxins. Given the relevance of its characteristics, investigating seed development and germination is essential, as these processes are dynamic and influenced by post-translational modifications (PTMs). The study aims to characterize the proteome and phosphoproteome of castor oil seed endosperm throughout imbibition and germination processes. Seeds endosperm collected at four imbibition stages (3 hours after the onset of imbibition – HAI – 6 HAI, 36 HAI and 62 HAI) and mature stage were dissected. Proteins were extracted with a solution containing phenol and tris buffer pH 7.9, and digested with trypsin (1:50, enzyme: substrate ratio). Phosphorylated peptides were enriched using titanium dioxide and eluted with 1.5% ammonium hydroxide solution (pH 11). The unretained and enriched fractions were then subjected to a liquid chromatography system coupled to an Orbitrap ExplorisTM 480. In the analysis of the unretained fractions, 9676 peptides and 1743 proteins were identified. The groups M, 3, 6, 36, and 62 have 881, 889, 859, 808 and 1011 proteins, respectively. Quantitative analysis revealed a higher number of down-abundant proteins compared to 97 up-abundant proteins, including ribosomal proteins involved in protein synthesis. The breaking of seed quiescence begins between 6 and 36 HAI, becoming active after 36 HAI with a higher number of positively regulated proteasomal, heat shock, and ribosomal proteins. In the phosphoproteomic analysis, 2613 phosphopeptides and 634 phosphoproteins were identified. The data highlighted 27 differentially abundant phosphoproteins and the quantitative analysis of phosphopeptides identified 55 differentially abundant phosphopeptides, with an overrepresentation of SP-type motif. Among the differentially abundant proteins, seven are potentially regulated by phosphorylation. Phosphorylated motif analysis revealed 13 overrepresented, including 3 exclusives to castor bean seed germination, with distinct abundance trends among stages. In conclusion, the comparative analysis of early castor bean seed germination highlights protein synthesis as a crucial process, along with identifying the phases determining quiescence breaking. Proteins regulated by phosphorylation were identified exhibiting variations in abundance when phosphorylated, including allergenic and toxic proteins. Overrepresented motifs display distinct patterns during imbibition, including those exclusive to castor bean seeds. These results highlight the impact of phosphorylation throughout proteins in seeds germination, as well as a more comprehensive understanding of the seed imbibition process. 181

182

P-I-0034 The phosphoproteome of stress granules Sophie Moggridge1, Alexis Chang1, Yuzhen Liu1, Matthew Berg1, Ricard A. Rodriguez-Mias1, Brian Beliveau1, Judit Villén1 1University of Washington, Genome Sciences, Seattle, WA, United States Under stress conditions, eukaryotic cells rapidly partition mRNAs and proteins into condensates called stress granules (SGs) to protect these molecules from degradation, and to initiate changes in gene expression and signalling pathways. SGs are integral cellular structures involved in proteostasis and aging, and the persistence and dysfunction of SGs contributes to many cancers and neurodegenerative diseases. The protein composition of SGs differs depending on the type of stress, however, the specific protein selection mechanisms are largely unknown. Protein phosphorylation can disrupt or promote interactions required for SGs and has been implicated in SG biogenesis. We hypothesize that phosphorylation plays a more substantial role in regulating SG interactions than previously thought, particularly mediating the inclusion or exclusion of proteins under different stress conditions. To understand this type of phosphoregulation, we measured the phosphoproteome of SGs under diverse stress conditions in human osteosarcoma cells (U-2 OS) with the goal of identifying general and stress-specific phosphorylation responses. To confirm SG formation, we used an engineered cell line with a GFP-fused SG marker, Ras GTPase-activating protein-binding protein 1 (G3BP1). Using mass spectrometry, we quantified stress-specific changes to the total proteome and phosphoproteome in response to oxidative stress (sodium arsenite), osmotic stress (sorbitol), heat stress, and proteasome inhibition (bortezomib) in six biological replicates per condition. We quantified over ten thousand phosphosites, 150 of which were significantly regulated in stress. Comparing the identified phosphoproteins to published SG protein databases, we selected candidate phosphosites that localize to SGs and are likely involved in SG regulation. Localization of candidate phosphosites was performed by BioID proximity labelling using multiple SG baits. Overall, we characterized the phosphorylation-specific changes to proteins in SGs in multiple stress conditions to more fully understand the mechanisms regulating stress responses, SG persistence and SG dysfunction that may contribute to health and disease. 182

183

P-I-0035 Defining intracellular and extracellular role of AGR2: insights of from CRISPR-targeted breast cancer cells and DIA-NN analysis M. Aiman Mohtar1, Syazalina Zahari1, Saiful Effendi Syafruddin1, Teck Yew Low1, Tiannan Guo2 1Universiti Kebangsaan Malaysia, Kuala Lumpur, Malaysia 2Westlake University, School of Life Science, Zhejiang Province, China Anterior Protein Gradient-2 (AGR2) is an endoplasmic reticulum resident protein that catalyzes thiol-disulfide interchange reactions during protein maturation within the secretory pathway. Overexpression of AGR2 is prevalent in various epithelial cancers, particularly in estrogen receptor-positive (ER+) breast cancer, where it correlates with poor prognosis. Emerging evidence suggests AGR2 can also be secreted extracellularly and exert pro-tumorigenic effects similar to its intracellular counterpart, highlighting its potential role in the tumor microenvironment. However, the precise mechanism of AGR2 secretion and how it impacts on the tumor microenvironment remains unclear. This study aims to elucidate the intricate roles of both intracellular and extracellular AGR2 signaling. We employed CRISPR/Cas9 genome editing to knock out AGR2 in two ER+ breast cancer cell lines, MCF7 and T47D, resulting in two AGR2 knockout cell pools per cell line. Using cutting-edge proteomics techniques, specifically Data Independent Acquisition-Neural Network (DIA-NN) mass spectrometry, we analyzed the impact of AGR2 depletion on cellular protein expression. In the whole cell lysate proteome of MCF-7 and T47D cell lines, we identified 9698 and 9984 proteins, respectively. Among these, we discovered 119 and 601 commonly differentially expressed proteins (DEPs) in the AGR2 knockout pools of MCF-7 and T47D, respectively. Functional enrichment analysis revealed shared perturbations in transmembrane transporter activity, estrogen signaling pathways, and oxidoreductase activity. Secretome analysis identified 1041 and 1626 proteins in MCF-7 and T47D cell lines, respectively. We identified 92 and 601 DEPs in the secretomes of MCF-7 and T47D, respectively, primarily involved in protein chaperone activity, metabolic pathways, and cytoskeleton organization. Protein-protein interaction and hub gene analysis further identified potential key binding partners in the AGR2 signaling axis. In summary, our findings provide a comprehensive insight into the intricate signaling pathways involving AGR2, spanning both intracellular and extracellular proteomes. We identified a subset of proteins highly reliant on AGR2-mediated regulation, suggesting that AGR2's oxidoreductase activity may extend beyond cellular boundaries. Moreover, this study emphasizes a potential 'druggable' stage within the AGR2- dependent pathway, proposing avenues for targeted therapeutic interventions. 183

184

P-I-0037 Comprehensive analysis of the redox proteome reveals survival pathways activated upon resistance to Pevonedistat Michaela Myšáková1, Kristýna Pimková1 1First Faculty of Medicine, Biocev, Prague, Czech Republic Pevonedistat (PEVO) is a selective inhibitor of neddylation-dependent protein degradation and has been shown to enhance the efficacy of the hypomethylating drug 5-azacytidine (AZA) in leukemia therapy. However, despite promising results, even combination therapy fails due to the emergence of resistance. Our recent data showed that AZA alters redox homeostasis and oxidative modifications of proteins, which further contributed to the selective pressure of AZA. PEVO activates the main antioxidant defense pathway by inhibiting the degradation of nuclear factor erythroid 2-related factor 2 (NRF2), a substrate of ubiquitin ligase Kelch-like ECH-associated protein 1 (KEAP1), and thus also affects cellular redox homeostasis. Redox changes are sensed by protein thiol groups which undergo various oxidative modifications leading to modulation in protein function. Therefore, in the presented work we investigated the impact of PEVO on redox homeostasis and redox proteome with the aim to reveal the role of redox homeostasis in the response of MDS/AML cells to the combined therapy. We used cell line models of MDS/AML myeloblasts sensitive and resistant to AZA and PEVO developed in our laboratory. We performed quantitative proteomic analysis using TMT10-plex and Orbitrap Fusion Tribrid MS instrument on PEVO-sensitive cells treated with PEVO and PEVO-resistant cells. We identified and quantified 6,886 proteins. PEVO treatment affected the level of 98 proteins, and resistance was associated with changes in levels of 566 proteins. Principal component analysis separated the analyzed groups, with differing levels of NRF2 target proteins contributing most to separating the PEVO-treated sample. Indeed, the activity of NRF2, determined by its accumulation in the nucleus, was significantly increased in the PEVO-treated sample. This was associated with changes in the oxidative state monitored by flow cytometry using fluorescent probes. Interestingly, enhanced activity of NRF2 was detected in untreated PEVO-resistant cells as well. Mass spectrometry-based redox proteomic analysis of reduced and oxidized cysteine thiol groups using iodoTMT isobaric labelling and Orbitrap Fusion Tribrid MS instrument identified 26,265 cysteine-containing peptides from 6,634 proteins and showed that PEVO resistance is accompanied by changes in the redox state of 228 proteins. We identified increased oxidation on cysteines 289 and 290 of Sequestosome-1, previously shown to be important for the autophagic clearance of ubiquitinated proteins. We have identified prominent modifications in proteins with roles in autophagy, indicating a role in autophagy regulation that could contribute to the adaptive mechanisms leading to the emergence of resistance. Our results indicate that the altered redox state has a significant role in PEVO and AZA resistance. 184

185

P-I-0038 Proteomic characterization of the protein S-glutathionylation pattern in stem and leukemic cells Dušan Němec1, Lubomír Minařík2, Tomáš Stopka2, Jenny Hansson3 1First Faculty of Medicine, Charles University, BIOCEV, Laboratory of Stem Cell Proteomics, Vestec, Czech Republic 2First Faculty of Medicine, Charles University, BIOCEV, Laboratory of Haematooncology and Stem Cells, Vestec, Czech Republic 3Lund Stem Cell Center, Department of Experimental Medical Science, Lund, Sweden Proteins are critical mediators of redox signaling. Cysteines within its thiol groups sense the oscillations in oxidative state and undergo different reversible oxidative modifications. One of such modifications is protein S- glutathionylation (PSSG), a conjugation of glutathione (GSH) to protein thiols that not only protects them from oxidative damage, but also regulates a wide variety of cellular functions including signal transduction, cell survival and apoptosis. PSSG has been described to play an important role in cancer initiation, progression, response to anti-leukemic therapy and the emergence of drug resistance in leukemia cells. The process of PSSG is driven by fluctuations in the availability of GSH and oxidized glutathione (GSSG) and by enzymes catalyzing the removal and conjugation of GSH to thiol residues such as glutaredoxins and glutathione-S-transferases. Our recent data showed that response of leukemic cells to hypomethylation therapy is associated with disturbed GSH/GSSG ratio, lower expression of glutathione S-transferase P (GSTP1) and overexpression of glutaredoxin 1. Therefore, our aim was to find a pattern of thiol oxidation upon modulation of PSSG and next to identify the impact of perturbations in GSH metabolism on PSSG in healthy human embryonic stem cells (hESCs) and leukemic cell lines. To do so, we induced oxidative stress in hESCs by treatment with hydrogen peroxide and monitored its impact on protein thiol modifications in the presence of GSSG and both GSSG and GSTP1 inhibitor. To identify and quantify free and oxidized protein cysteine sites we applied IodoTMT-based redox proteomic approach using LC- MS3 method. We identified 2485 cysteine containing proteins from which 773 had significantly changed thiol oxidation status. Inducing oxidative stress by H2O2 and increasing the level of oxidized glutathione by NOV-002 compound profoundly elevated the median protein oxidation from 4 to 6 %, while simultaneous inhibition of GSTP1 with Ezathiostat rescued the oxidation to control levels. Next, we studied the impact of perturbations in GSH metabolism and inhibition of thioredoxin on protein PSSG using selective reduction and blocking of PSSG sites with subsequent TMT labeling in OCI-M2 leukemic cells. We found out that perturbation of GSH metabolism at the node of GSH synthesis (glutamate-cysteine ligase) led to significant alterations in protein S-glutathionylation and significantly inhibited leukemic cells growth. Collectively, these data indicate that protein S-glutathionylation plays a critical role in antioxidant defense of the cell, and modulation of PSSG in leukemia can lead to proliferation arrest and induce cell death that could increase the efficacy of anti-leukemic therapy. 185

186

P-I-0039 PM2.5 engages impairment of mucosal immunity, involving alteration of complement activation and cell junction assembly by regulated IL-17 levels: Pathological and proteomic approaches Warat Palpai1, Thamradee Tansrivorarat1, Aekkacha Moonwiriyakit1, Wararat Chiangjong2 1Mahidol University, Chakri Naruebodindra Medical Institute, Faculty of Medicine Ramathibodi Hospital, Samut Prakan, Thailand 2Mahidol University, Department of Pediatrics, Faculty of Medicine Ramathibodi Hospital, Bangkok, Thailand Fine particulate matter (PM2.5) exacerbates respiratory morbidity, posing significant health and environmental concerns. Although PM2.5 causes various adverse effects on the respiratory tract, the mechanisms of PM2.5- induced lung toxicity have been incompletely characterized. Thus, this study aims to investigate the mechanisms of PM2.5-induced lung inflammation. BALB/c mice were intratracheally administered with concentrations of PM2.5 daily for 7 days. Histopathological analyses of lung tissue were performed. Bronchoalveolar lavage fluid proteomics was analyzed using QTOF LC-MS/MS, and differentially expressed genes were visualized through volcano plots and heatmaps. Functional protein data were further assessed by GO enrichment analysis, KEGG pathways, and STRING protein networking. Moreover, specific proteins of interest were validated using multiplex ELISA. We found that PM2.5 induced marked neutrophilic inflammation in the lungs, as evidenced by the increase of polymorphonuclear cells and macrophages in a dose-dependent manner. Proteomics analysis identified 147 differentially expressed proteins, with 43 down-regulated and 104 up-regulated. Gene ontology analysis revealed that up-regulated proteins are mainly secretory and implicated in the inflammatory response, complement activation, and assembly of various cell junctions. Additionally, a decrease in serum IL-17A levels, supported by STRING network analysis, suggests its involvement in PM2.5-induced mucosal immunity impairment in the respiratory tract and potentially in non-exposed organs. These findings provide mechanistic insight into PM2.5-induced inflammation, facilitating the development of biomarkers of PM2.5 exposure and targeted therapies for PM2.5-related lung diseases. 186

187

P-I-0040 Redox proteomics identifies a novel adaptive mechanism of leukemic cells emerging upon exposure to hypomethylation therapy Kristýna Pimková1, Dušan Němec1, Juraj Kokavec1, Lubomír Minařík1, Michaela Myšáková1, Tomáš Stopka1 1BIOCEV, 1st Faculty of Medicine, Vestec, Czech Republic Hypomethylation therapy by 5-azacytidine (AZA) is indicated upon progression of onco-hematological diseases: myelodysplastic syndrome (MDS) and acute myeloid leukemia (AML). AZA represents a backbone treatment option in MDS/AML that significantly prolongs patient survival, but early development of resistance seriously complicates the therapy. AZA inhibits DNA methylation, induces DNA damage as well as the production of reactive oxygen species (ROS). Our and others data have shown that ROS, through oxidative modifications of thiols in proteins, fine-tune intrinsic stress and survival cell programs of cells. Therefore, we hypothesized that AZA induces protein oxidative modifications that control adaptive processes when MDS/AML cells respond to AZA. In this study, we applied a quantitative mass spectrometry-based proteomic approach to identify protein targets of oxidative modifications in a cellular MDS/AML model of AZA resistance (Minarik, Pimkova et al., 2022) and in primary patient samples, and complemented the results with transcriptomic analysis. Identification of the redox state of 8,686 cysteine sites in 3,380 proteins revealed that AZA preferentially regulated the redox state of proteins that are activated in response to the DNA damage and contain a zinc finger domain essential for DNA/RNA binding, and that oxidation induced cell death is one of the mechanisms of action of AZA. MDS/AML OCI-M2 cell and AZA-resistant primary myeloblasts are characterized by high glutathione (GSH) levels and low S- glutathionylation of proteins. GSH depletion by buthionine sulfoximine increased protein S-glutathionylation and restored response to AZA in AZA resistant cells thus validating our assumptions. This study unveils a role of redox homeostasis in the toxicity and mechanism of adaptation to the hypomethylation therapy. This is the first work that comprehensively described the effect of AZA treatment on oxidative modifications of proteins in MDS/AML cell line responding and not responding to treatment and compared these data with redox changes on proteins in patients relapsing on AZA. Remarkably we found that AZA-induced redox changes of the proteome in MDS/AML cell line forms a substantial part of AZA toxicity and also contribute significantly to the development of AZA resistance. 187

188

P-I-0041 Histone variants and post-translational modifications in the regulation of plant callus formation Pavlína Pírek1, Karolína Kryštofová1,2, Ingrid Kováčová1, Anna Kromerová2, Dagmar Zachová1, Ondřej Helia1,2, Klára Panzarová3, Jiří Fajkus1,2, Zbynek Zdrahal1,2, Gabriela Lochmanová1,2, Miloslava Fojtová1,2 1CEITEC Masaryk University, Mendel Centre for Plant Genomics and Proteomics, Brno, Czech Republic 2Masaryk University, Faculty of Science, National Centre for Biomolecular Research, Brno, Czech Republic 3PSI (Photon Systems Instruments) spol. s.r.o., Drásov, Czech Republic Introduction: The ability of plants to regenerate from dedifferentiated cells opens up many possibilities for molecular bioengineering. By manipulating gene expression in calli, we can derive crops with specific traits. Epigenetically active compounds are frequently used to impact callus formation and subsequent processes of plant organ regeneration. However, little is known about epigenetic changes during callus formation and epigenetic imprint in long-term callus propagation. Objectives: Comparison of histone variant composition and post-translational modification status in calli induced from the shoot and root parts of Arabidopsis thaliana seedlings in the presence / absence of histone deacetylase inhibitors (HDACi), 0.5 mM sodium butyrate (NaB) or 0.5 µM trichostatin A (TSA). Materials and methods: Callus cultures originated from shoot or root parts of 7-day-old seedlings were collected for MS analysis after 20 days of cultivation or 1-year propagation. Nuclei from seedlings and calli were isolated using Percoll gradient followed by histone extraction into sulfuric acid. Chemical derivatization of amine residues was performed prior to LC-MS/MS on Orbitrap Fusion Lumos Tribrid. Peaks of histone peptide precursors were quantified in Skyline software. Results: Increased levels of the histone H3.3 variant were prominent in 20-day calli. The methylation status in root- and shoot-derived calli reached the same level during long-term propagation, while differences in acetylation levels provided a lasting imprint of their origin. Epigenetic markers disappeared after 20 days of callus propagation in the presence of HDACi, with NaB-treated calli resembling root-derived calli and TSA-treated calli resembling shoot-derived calli. Conclusion: Understanding the factors that control epigenetic changes during callus formation is a prerequisite for targeted plant regeneration and the development of crop manipulation technologies. Here we show that histone variant replacement plays a substantial role in chromatin remodeling during callus induction and propagation. Changes in histone acetylations and methylations represent an additional ﬁne-tuning mechanism that is important for activation/repression of speciﬁc genes. Interestingly, signs of calli origins (shoot vs. root) seem to disappear during long-term passaging and in the 20-day cultivation in the presence of HDACi. Funding: This research was supported by the European Regional Development Fund project "SINGING PLANT" (No. CZ.02.1.01/0.0/0.0/16_026/0008446) and Czech Science Foundation (project No. 22-28190S). CIISB: Instruct-CZ Centre of Instruct-ERIC EU consortium, funded by MEYS CR infrastructure project LM2023042, is gratefully acknowledged for the financial support of the measurements at the CEITEC Proteomics Core Facility. Computational resources were provided by the e-INFRA CZ project (ID:90254), supported by MEYS CR. 188

189

P-I-0042 Inhibition of TBK1/IKKε mediated RIPK1 phosphorylation sensitizes tumors to immune cell killing Anastasia Piskopou1, David Vredevoogd2, Kong Xiangjun2, Daniel Peeper2, Maarten Altelaar1, Kelly Stecker1 1Utrecht University, Biomolecular Mass Spectrometry and Proteomics, Bijvoet Center for Biomolecular Research and Utrecht Institute for Pharmaceutical Sciences, Utrecht, Netherlands 2Netherlands Cancer Institute, Division of Molecular Oncology and Immunology, Amsterdam, Netherlands Introduction: Resistance to immune cell-mediated cytotoxicity poses a significant challenge in cancer treatment, compromising the efficacy of promising immunotherapeutic approaches, such as immune checkpoint blockade (ICB) therapy. To enhance treatment outcomes, it is crucial to elucidate interventions that synergize with ICB to overcome tumor resistance. Therefore, we need to identify the cellular mechanisms that sensitize tumors to cytotoxic CD8+ T cells. CD8+ T cells rely on cytokines such as TNFα to carry out their cytotoxicity against cancer cells. TNFα is a major mediator of cell death and immunity, and recent findings link select tumor mutations within the TNF pathway to increased T cell killing, in a manner dependent on the protein RIPK1 (Vredevoogd et al., 2019). Key findings: Through MS-based proteomics and phosphoproteomics methods, we demonstrate that sensitized tumor cells fail to initiate inhibitory RIPK1 phosphorylation at site S25 upon T cell attack, thereby foregoing a pro-survival checkpoint early in TNF signal transduction and driving RIPK1-mediated cell death. Consequently, tumor cells experiencing a loss of TNF-induced RIPK1 S25 phosphorylation exhibit increased RIPK1 activation and fail to recruit non-canonical IKK kinases (TBK1 and IKKε) to the TNFR1 complex. Functional knockouts of TBK1 and IKKε in melanoma cells lack RIPK1 S25 phosphorylation and result in heightened sensitivity not only to CD8+T cell but also to Natural Killer cell attacks. These findings implicate TBK1 & IKKe as the checkpoint kinases responsible for pro-survival RIPK1 S25 phosphorylation and a key regulator of immune-mediated cell death in cancer cells. MS application: To quantify phosphorylation dynamics of RIPK1 within protein signaling complexes, we combine phosphoproteomic analysis with Affinity Purification-Mass Spectrometry (AP-MS). Melanoma cell lines were stimulated with biotinylated TNFα and downstream complexes were isolated via streptavidin-coated magnetic beads. The purified complexes were digested utilizing S-TRAP microfilters, with subsequent enrichment of phosphopeptides carried out in an automated manner using the AssayMAP Bravo Platform, prior LC-MS analysis using an Orbitrap Exploris MS. In addition to our proteomics workflow, a Targeted Mass Spectrometry method was developed to quantify RIPK1 phosphorylation and validate our findings. Impact: Our findings indicate that preventing TBK1 and IKKε recruitment to the TNF signaling complex, thereby blocking RIPK1 pro-survival phosphorylation and promoting direct RIPK1 activation, can be considered a tractable strategy to increase tumor sensitivity to immune cell killing and has the potential to benefit current immunotherapy interventions. 189

190

P-I-0043 Global, site-resolved analysis of ubiquitylation occupancy and turnover rate reveals systems properties Gabriela Prus1, Shankha Satpathy1, Brian Weinert1, Takeo Narita1, Chunaram Choudhary1 1Novo Nordisk Foundation Center for Protein Research, University of Copenhagen, Proteomics Program, Copenhagen, Denmark Ubiquitylation regulates most proteins and biological processes in a eukaryotic cell. However, the site-specific occupancy (stoichiometry) and turnover rate of ubiquitylation have not been quantified. Here we present an integrated picture of the global ubiquitylation site occupancy and half-life. Ubiquitylation site occupancy spans over four orders of magnitude, but the median ubiquitylation site occupancy is three orders of magnitude lower than that of phosphorylation. The occupancy, turnover rate, and regulation of sites by proteasome inhibitors are strongly interrelated, and these attributes distinguish sites involved in proteasomal degradation and cellular signaling. Sites in structured protein regions exhibit longer half-lives and stronger upregulation by proteasome inhibitors than sites in unstructured regions. Importantly, we discovered a surveillance mechanism that rapidly and site-indiscriminately deubiquitylates all ubiquitin-specific E1 and E2 enzymes, protecting them against accumulation of bystander ubiquitylation. The work provides a systems-scale, quantitative view of ubiquitylation properties and reveals general principles of ubiquitylation-dependent governance. Fig. 1 (optional) 190

191

P-I-0044 Proteomic workflows for substrate screening and site mapping of ubiquitin-like protein modification Elena Rudashevskaya1, Zehra Vural1, Nils Brose1, Marilyn Tirard1, Olaf Jahn1 1Max Planck Institute for Multidisciplinary Sciences, Department of Molecular Neurobiology, Goettingen, Germany Ubiquitin-like proteins (UBLs) such as SUMO, Nedd8 or UFM1 are conjugated to target proteins to regulate their activity, stability, subcellular localization, or macromolecular interactions. In recent years, UBLs have become an important subject of study in biomedical research as they play a role in central biological processes like stress response, and as UBL dysfunction is associated with many pathological states, including neurodevelopmental and neurodegenerative disorders. Knowledge of the modified lysine residues within a conjugated substrate is key for a detailed functional analysis, e.g. by replacing acceptor lysine with non-modifiable arginine residues. However, despite advances in mass spectrometry (MS)-based proteomics, mapping of UBL modification sites remains difficult not only because of the low stoichiometry of conjugation. Particular technical challenges include the detection and sequencing of low-abundant branched peptides, in which the size of the remnant from the UBL may vary considerably depending on the type of UBL present and the protease used for digestion during sample preparation. To screen for substrates conjugated with SUMO2, the most abundant SUMO paralogue in neurons, we made use of a novel HA-SUMO2 knock-in mouse model, which expresses HA-tagged SUMO2 from the endogenous genetic locus. Using anti-HA affinity purification from whole brain, followed by label-free protein quantification, we identified several hundred putative SUMO2 candidates, including the transcription factor Zbtb20, of which we have previously characterized the SUMO acceptor sites [1]. We thus selected HEK cells expressing both tagged Zbtb20 and tagged SUMO2 as model for establishing a proteomic workflow for UBL conjugation site mapping. Partially using the strategy introduced by Hendriks et al. [2] as a starting point, we revisited critical steps of the workflow and established optimized elution conditions of the anti-tag affinity purification, C8 stagetip-based enrichment of branched peptides, and MS acquisition parameters with customized mass cutoffs and collision energies on an Orbitrap Exploris 480 instrument. Our workflow will be used in future for the site mapping of synaptic UBL targets with the aim of gaining insights into the role of UBL alterations in synapse biology and pathology. [1] Ripamonti S, Shomroni O, Rhee JS, Chowdhury K, Jahn O, Hellmann KP, Bonn S, Brose N, Tirard M. SUMOylation controls the neurodevelopmental function of the transcription factor Zbtb20. J Neurochem 2020, 154:647. [2] Hendriks IA, Lyon D, Su D, Skotte NH, Daniel JA, Jensen LJ, Nielsen ML. Site-specific characterization of endogenous SUMOylation across species and organs. Nat Commun 2018, 9:2456. 191

192

P-I-0045 Proteomic and phosphoproteomic profiling of a TRIB2-KO glioblastoma-derived cell line reveals new roles in cell signaling Pablo San Segundo Acosta1, Alba Orea Soufi2, Wolfgang Link3, Bibiana I. Ferreira2, Rodrigo Barderas1,4 1Instituto de Salud Carlos III, Functional Proteomics, Madrid, Spain 2Algarve Biomedical Center Research Institute-ABC-RI University of Algarve, Faro, Portugal 3Sols-Morreale Biomedical Research Institute (IIBM), Spanish National Research Council (CSIC), Universidad Autónoma de Madrid (UAM), Madrid, Spain 4CIBERFES, Madrid, Spain Glioblastoma, a type of cancer of the brain or the spinal cord, forms from astrocytes supporting nerve cells. Despite the various treatments available, the prognosis for glioblastoma is poor, with a median survival time of about 14 months. Tribble proteins, or Tribbles, are pseudokinases with very important roles in cell signaling regulation. These proteins present at least a pseudokinase domain and a C-terminal domain, by which they act as scaffold for a diverse range of proteins to regulate their post-translational modifications and their cellular location. They primarily signal via noncatalytic mechanisms, such as protein-protein interactions, contributing to the development of various human diseases. Trib2 has been shown to interact specifically with E3-ubiquitin ligases and MAP kinases through its C-terminal domain, and with their substrates via the C-terminal domain and the pseudokinase domain. Thus, TRIB2 acts as a regulator of important cellular pathways such as PI3K-Akt, NFkB, or Wnt, all of them related to cell survival and proliferation. These pathways are key in the development and progression of many types of cancer, including glioblastoma. Here, we have analyzed the proteome and phosphoproteome of the highly-TRIB2-expressing glioblastoma- derived cell line Lnn229, and of two different Crispr-Cas9-generated TRIB2-KO clones. SILAC and label-free quantitative proteomics analyses have been performed using data-independent acquisition with an Orbitrap Astral Mass Spectrometer to address the interactome and phosphoproteome associated to this pseudokinase. Our data shows that TRIB2 regulates the phosphorylation and expression levels of important cellular proteins, which play a key role in survival, metastasis, and proliferation. Results suggest that the modulation of the pathways where Trib2 is involved could help in the management of glioblastoma and other cancers, where these pathways are dysregulated. 192

193

P-I-0046 Spike-in enhanced phosphoproteomics uncovers synergistic signaling responses to MEK inhibition in colon cancer cells Mirjam van Bentum1,2, Bertram Klinger2,3, Anja Sieber2,3, Nadine Lehmann3, Sheyda Naghiloo1, Mohamed Haji1, Sylvia Niquet1,4, Philipp Mertins1,4, Nils Bluethgen2,3, Matthias Selbach1,3 1Max Delbruck Center for Molecular Medicine, Berlin, Germany 2Humboldt University, Berlin, Germany 3Charité – University Medicine Berlin, Berlin, Germany 4Berlin Institute of Health at Charité – Universitätsmedizin Berlin, Berlin, Germany Targeted kinase inhibitors are a cornerstone of cancer therapy, but their success is often hindered by the complexity of cellular signaling networks that can lead to resistance. Overcoming this challenge necessitates a deep understanding of cellular signaling responses. While standard global phosphoproteomics offers extensive insights, lengthy processing times, the complexity of data interpretation, and frequent omission of crucial phosphorylation sites limit its utility. Here, we combine data-independent acquisition (DIA) with spike-in of synthetic heavy stable isotope-labeled phosphopeptides to facilitate the targeted detection of particularly informative phosphorylation sites. Our spike-in enhanced detection in DIA (SPIED-DIA) approach integrates the improved sensitivity of spike-in-based the discovery potential of global phosphoproteomics into a simple workflow (Fig. 1). We employed this method to investigate synergistic signaling responses in colorectal cancer cell lines following MEK inhibition. Our findings highlight that combining MEK inhibition with growth factor stimulation synergistically activates JNK signaling in HCT116 cells. This synergy emphasizes the therapeutic potential of concurrently targeting MEK and JNK pathways, as evidenced by the significantly impaired growth of HCT116 cells when treated with both inhibitors. Our results demonstrate that SPIED-DIA effectively identifies synergistic signaling responses in colorectal cancer cells, presenting a valuable tool for uncovering new therapeutic targets and strategies in cancer treatment. targeted detection with Fig. 1 (optional) 193

194

P-I-0047 Phosphoproteomic delineation of platelet activation states Eva Smit1, Carmen van der Zwaan1, Arjan Hoogendijk1, Maartje van den Biggelaar1 1Sanquin Research, Amsterdam, Netherlands Background: Primary hemostasis is a tightly orchestrated process initiated by the adhesion, activation and aggregation of blood platelets to stop bleeding. This multi-step process depends on receptor interactions, including VWF-Glycoprotein Ib collagen fibers -Glycoprotein VI/α2β1 integrin, epinephrin to A2α and released of platelet granule interactions of thromboxane A2, (TxA2 [TP]) with TP and ADP with P2Y1 and P2Y12.These interaction in platelet functions are regulated by kinases and phosphatases. In platelet function disorders, each of the aforementioned receptors can be impaired in response to stimuli, resulting in increased bleeding risk. The extent of diverging and converging signaling pathways between the intra-platelet signaling is unresolved. Further resolving these is called for to better understand biological mechanisms in the platelet function disorders that these receptors are involved in. To this end, we aimed to delineate receptor-specific platelet activation states for key receptors on platelets. Method: Activation of platelets was induced with a panel of agonists, including arachidonic acid (AA), TxA2 mimic U46619, ristocetin, collagen, epinephrin, PAR-1 AP and PAR-4 AP, by stimulation for 30sec, 1min, 2min, 5min and 10min. After lysis, denaturation, reduction and alkylation, proteins were digested and phosphopeptides were enriched from the peptides with Fe3+-IMAC and analyzed with data independent acquisition on a TimsTof HT. Results: A large coverage of phosphosites across the range of platelet activation states was obtained with a total of 25680 quantified phosphorylation sites. As previously reported, we observed MAPK1 phosphorylation (T185- MAPK1 and Y187-MAPK1) as well as dephosphorylation of the PAR-1 receptor (S412-F2R, S413-F2R, S418-F2R and Y420-F2R) upon activation with PAR-1 AP. Longitudinal statistical analysis showed limited (de)phosphorylation upon stimulation with epinephrine and ristocetin, whereas strong agonists (PAR-1 AP, PAR- 4 AP and U46619) demonstrated a large impact on the platelet phosphoproteome. For each of the agonists we observed unique differentially expressed phosphorylation sites including dephosphorylation of the receptor for the agonists, including S314-A2α by epinephrine, S418-F2R by PAR-1 AP, S331-TP by U46619 and S366-F2RL3 by PAR-4 AP. Furthermore, the strong agonists shared over 2000 differentially regulated phosphorylation sites. These might be involved in TxA2 synthesis and platelet granule secretion, which are specific to strong agonists. Notably, focusing on the TxA2 synthesis pathway through the thromboxane A2 precursor AA, revealed that the AA phosphorylation signature was greater than the other strong agonists including thromboxane A2 itself. Conclusion: This study dissected the longitudinal changes upon stimulation of receptors that are important in primary hemostasis. It highlights the overlap and unique signatures between the platelet activation states. 194

195

P-I-0048 Multiproteomics approach reveals an axonogenesis and synaptogenesis disruption throughout congenital Zika syndrome progression Patricia Sosa-Acosta1,2,3, Raissa Rilo4, Tailene Rabello4, Jefferson Rodrigues Quintanilha4, Jéssica Guedes1,2,3, Mauricio Quiñones-Vega1,2,3, Jéssica Ferreira4, Patrícia Garcez4, Gilberto B. Domont1,2,3, Fabio Cesar Sousa Nogueira1,2,3 1Federal University of Rio de Janeiro, Biochemistry, Rio de Janeiro, Brazil 2Federal University of Rio de Janeiro, LADETEC (Insitute of Chemistry), Laboratory of Proteomics (LabProt), Rio de Janeiro, Brazil 3Federal University of Rio de Janeiro, Precision Medicine Research Center, Rio de Janeiro, Brazil 4Federal University of Rio de Janeiro, Institute of Biomedical Science, Rio de Janeiro, Brazil Congenital Zika Syndrome (CZS) is caused by vertical transmission of the Zika virus (ZIKV) during pregnancy, leading to various pathologies in affected fetuses. Although global ZIKV outbreaks have diminished, CZS remains a significant concern in endemic countries. Nowadays, CZS progression studies remain unexplored. To address this gap, we conducted a comprehensive study using in-vivo mice models to investigate the molecular mechanisms underlying CZS development, including acute and chronic infection stages. Our approach focused on examining proteome, N-glycoproteome, and phosphoproteome alterations across different neurodevelopmental stages. Through quantitative proteomics, N-glycoproteomics, and phosphoproteomics analyses, we detected 4,870 and 5,062 proteins in acute and chronic stages, respectively. Among them, 802 and 934 N-glycoproteins, 1,131 and 1,030 phosphoproteins, 1,786 and 2,188 N-glycosites, and 2,593 and 2,245 phosphosites, corresponding to acute and chronic infections. During early ZIKV infection, N-glycan biosynthesis was negatively disrupted through alpha-1,3/1,6-mannosyltransferase impairment of N- glycosylation may be related to proteasome activation in the acute stage. In the CZS stage, cell adhesion proteins exhibited a drastic hypoglycosylation affecting complex neurodevelopmental processes such as axonogenesis and synaptogenesis, also detected in the total proteomics approach. In addition, histological examination of postnatal ZIKV infection shows a decreased dendritic arborization of pyramidal neurons which may impact synapse formation. During acute stage, Golgi organization process was enhanced and regulated by phosphorylation, which might be related to N-glycosylation disruption. Conversely, the synaptic vesicle cycle was negatively modulated by phosphorylation the phosphatidylinositol 4-kinase type 2-alpha, nucleoside diphosphate kinase B, cyclin-dependent kinase 2, and dual specificity protein kinase CLK2 remain altered throughout CZS development. Multi-proteomics approach proved to be a valuable tool to monitor CZS progression and allowed to identify specific proteins and pathways for potential therapeutic targeting. the dysregulated kinases, inhibition. The in CZS stage. Among 195

196

P-I-0049 Proteome-wide analysis of PTM-dependent protein conformational changes Damiano Tosolini1, Claudio Bassot1, Tomas Koudelka1, Ilaria Piazza1 1Max-Delbrück-Centrum für Molekulare Medizin, Berlin, Germany Intrinsically disordered regions (IDRs) of proteins are difficult to study, because they lack a defined three- dimensional structure, yet they play important biological roles. Many IDRs can mediate protein-protein interactions through regulation by post translational modifications (PTMs). IDRs also affect protein functions, as 20% of missense pathogenic mutations are present in IDRs. Understanding why disease associated mutations are located in IDRs, why protein interactions interfaces are particularly present in IDRs, and how this is linked with post-translational modifications is essential for unravelling disease mechanisms. Here we investigate how post- translational modifications globally modulate protein conformations using structural proteomics. We find that hundreds of putative structural changes are triggered by PTMs across the human proteome, recapitulating known regulatory events. The co-localization of these protein regions with known disease-associated mutations suggest that specific protein conformational changes and proteoforms pairs might be involved in the pathogenesis of genetic diseases. 196

197

P-I-0050 Pitfalls in phosphoproteomic data analysis Christine von Toerne1, Fabian Giehler2,3, Fabian Gruhn1, Michael Bock1, Julia Mergner4, Ignasi Forné5, Arnd Kieser2,3, Stefanie Hauck1 1Helmholtz Center Munich – German Research Center for Environmental Health, Metabolomics and Proteomics Core, Munich, Germany 2German Center for Infection Research (DZIF), Partner Site Munich, Munich, Germany 3Helmholtz Center Munich – German Research Center for Environmental Health, Research Unit Signaling and Translation (SAT), Munich, Germany 4Technical University of Munich, Bavarian Center for Biomolecular Mass Spectrometry at Klinikum rechts der Isar (BayBioMS@MRI), Munich, Germany 5Ludwig-Maximilians-University Munich, Faculty of Medicine-BioMedical Center, Munich, Germany Phosphorylation of proteins plays a major role in signal transduction and provides insights into the activation status of pathways in response to stimuli. Investigation of global changes in post-translational modifications (PTMs) is best achieved using mass spectrometry (MS)-based proteomics. Various techniques for enriching phosphorylated peptides have advanced "phospho"-proteomics in recent years. Similar to global proteomics, the phospho-proteomics field is transitioning from data-dependent acquisition (DDA) combined with the MaxQuant/Perseus (MPI, Martinsried) analysis pipeline to data-independent acquisition (DIA) methods. Initially, DIA spectra were commonly analyzed using Spectronaut software (Biognosys), with precursor lists exported and converted into a Perseus-readable format for further data analysis. However, recent versions of Spectronaut include a generic phospho data analysis capability. In this study, we compare data analysis in Perseus with Spectronaut using a real-life cell culture experiment. Phosphopeptides were enriched using Zr-IMAC beads (ReSyn) and measured in DIA (QE-HFX). Spectra were searched in directDIA mode in Spectronaut (version 17). Collapsed peptides were first analyzed in Perseus (version 2.0.3.1). Antibody-based validation of statistically significant differential peptides was performed using Western blots (WBs) for seven candidates. Subsequently, data analysis, including statistical evaluation, were tested in Spectronaut. These datasets were systematically compared, focusing on localization probability cut-offs, imputation strategies, and different normalization options. Given that neither the data analysis parameters themselves, nor the order in which steps are performed are independent of each other, we aim to highlight potential pitfalls in data analysis and propose a best practice approach, guided by our positive control candidates. 197

198

P-I-0051 Deep learning-assisted lactylprotein mining (DeepLaM) identifies potential pan-cancer biomarkers Ning Wan1, Chenguang Liu1, Tianze Ling2, Zimeng He1, Rui Han1, Wenjie Yuan1, Xiaoyu Pu1, Jiahao Chen1, Chang Shao1, Haiping Hao1, Cheng Chang2, Hui Ye1 1China Pharmaceutical University, Nanjing, China 2National Center for Protein Sciences, Beijing, China The study of the functionality of lactylation, a recently discovered post-translational modification (PTM), has been emerging lately. However, its role in cancer progression remains to be elucidated. Previously, we found a lactylated lysine-derived cyclic immonium (cycIm) ion and demonstrated that its presence can signify true lysine lactylation (Klac) and identify novel Klac sites from unriched proteomic datasets. Here, we advanced this data mining strategy to DeepLaM, which involves the use of artificial intelligence (AI) algorithms to predict true Klac sites after training with the largest Klac-bearing peptide library to date. With DeepLaM, we efficiently identied several Klac sites that are consistently present in multiple cancer types, including lung adenocarcinoma, colon cancer, pancreatic ductal adenocarcinoma, squamous lung cancer, and small cell lung cancer by searching public data collected by the Clinical Proteomic Tumor Analysis Consortium (CPTAC). This finding implies these Klac sites as potential pan-cancer biomarkers and their association with cancer progression. Interestingly, these Klac- modified protein substrates converge in their involvement in NADPH-dependent reductase catalysis. To further evaluate the potential of these novel Klac sites as cancer biomarkers and even drivers of cancer progression, we performed Klac proteome profiling and detected their upregulation in our collected lung adenocarcinoma samples. Their functionality was elucidated in vitro using genetic codon expansion (GCE). We site-specifically engineered the discovered Klac proteins in lung cancer cell lines, and demonstrated that these proteins modulate the redox homeostasis in cancer cells and contribute to cell growth and migration. Together, by integrating AI-based spectral learning and prediction, Klac proteome profiling and GCE, we are expanding our ability to efficiently discover cancer biomarkers and uncover their contribution to cancer. Such knowledge deepens our understanding in lactylation and cancer biology, and may provide novel target proteins for cancer prognosis and treatment. 198

199

P-I-0053 Quantitative phosphoproteomics identifies filamin C as a signaling hub of MAP kinases in skeletal muscle cells during acute mechanical stress Johannes P. Zimmermann1, Anna L. Fricke1,2,3, Anja Schwäble3,4, Lorena Heil5, Lena Reimann3,4, Wignand W. D. Mühlhäuser3,6, Madeleine Christoph5, Thien-Ly J. Dinh3,7, Peter F. M. van der Ven5, Dieter Fürst5, Bettina Warscheid1,3 1University of Würzburg, Biochemistry II, Wuerzburg, Germany 2Sartorius CellGenix GmbH, Freiburg, Germany 3University of Freiburg, Functional Proteomics, Freiburg, Germany 4Celonic AG, Basel, Switzerland 5University of Bonn, Institute for Cell Biology, Bonn, Germany 6Scitaris Biopharma, Berlin, Germany 7University of Freiburg, Institute for Surgical Pathology, Freiburg, Germany Objective: Skeletal muscle cells are continuously exposed to mechanical forces. Mechanical sensors such as filamin C (FLNc) and integrins mediate intracellular signaling pathways governed by protein kinases to maintain contractile functions even under extensive mechanical strain. Changes in protein folding and posttranslational modifications can directly regulate the function, turnover, and the interactions of key players within the contractile apparatus to maintain proteostasis in mechanically stressed myotubes and thereby enable muscle fibers to adapt and protect themselves. In this work, we designed a functional, quantitative phosphoproteomics approach to identify signaling mechanisms involved in acute mechanical stress protection in skeletal muscle cells. Methods: To analyze the phospho-signaling network in C2 myotubes under mechanical stress, we applied electrical pulsed stimulation (EPS) and subsequently performed global quantitative phosphoproteomics using triple stable isotope dimethyl labeling in combination with high-pH RPLC for peptide fractionation and IMAC for phosphopeptide enrichment before LC-MS. Phosphoproteomics data analysis was performed to reveal stress- activated protein kinases. Functional proteomics and biochemical analyses were performed to identify those kinases specifically targeting the large signaling adaptor FLNc to regulate its protein interactions and functions at the myofibrillar Z-disc of contracting skeletal myotubes. Results: We established a vast myotube phosphoproteome with 33,000 phosphosites localized in 7,226 proteins. 21,428 phosphosites were quantified in ≥ 3/6 biological replicates. Under acute mechanical stress, 2,243 phosphopeptides were significantly upregulated, while 700 were downregulated. Kinase-substrate enrichment analysis showed that mechanical stress markedly increased the activity of JNK1, p38α, ERK1 and PKCα. Most phosphorylation events on FLNc were found in its mechano-sensing domain 20 and its dimerization domain 24. Both domains are known to mediate binding to Z-disc proteins and proteostasis factors. We found that numerous proline-directed motifs in FLNc showed increased phosphorylation under mechanical stress. Inhibition of the MAP kinases in skeletal myotubes during mechanical stress revealed JNK1, p38α, and ERK1 as regulators of PxSP sites, which show different time-dependent activation profiles. Functional analysis of mechanical stress responsive phosphosites in d24 using FLNc mutants mimicking either the non- or the phosphorylated isoform(s) showed regulation of FLNc-HSPB7 binding, which affects FLNc homodimerization for actin-crosslinking at Z-discs. On a physiological level, both FLNc phosphomutants led to higher lesion areas in mechanically stressed skeletal myotubes. We concluded that dynamic phosphorylation/dephosphorylation cycles are crucial for the role of FLNc in stabilizing and maintaining sarcomeres as well as for FLNc homeostasis during mechanical stress. 199

200

Microbiology and microbiome analysis (P-I-0054 to P-I-0096) P-I-0054 Staphylococcus aureus and skin infection: a proteomic journey to decipher adaptation strategies Dina Al Nahhas1,2, Salvatore Pisanu1, Sergio Uzzau2, Daniela Pagnozzi1 1Porto Conte Ricerche, Alghero, Italy 2University of Sassari, Sassari, Italy Skin infections caused by methicillin-resistant Staphylococcus aureus (MRSA) pose significant clinical challenges due to their virulence and antibiotic resistance. Understanding the proteome variations of S. aureus during skin infection can reveal critical insights into pathogenic mechanisms and help identify novel targets for therapeutic interventions. This study aimed to characterize the in vitro proteome of S. aureus from infected skin tissues using a 3D skin model. We focused on two strains of MRSA, ST398 and JE2 (a USA300 isolate), representing different epidemiological contexts. We employed a 3D skin model that mimics human skin, consisting of fibroblasts and keratinocytes, to provide a highly relevant context for human skin infections. These in vitro tissues were infected with 108 CFU/mL of the respective MRSA strains. After 2 days, we washed the tissues with sterile PBS and collected both the bacterial supernatant and the underlying tissues for protein extraction. The extracted proteins were analyzed using shotgun proteomics with a Q-Exactive mass spectrometer. Raw data were processed with Proteome Discoverer using SEQUEST-HT and evaluated by a label-free quantification (LFQ) approach using precursor ion abundances to identify differentially expressed proteins (DEPs). We identified a total of 2,194 proteins from ST398 and 2,074 proteins from JE2. Analysis of the DEPs revealed that proteins associated with amino acid metabolism, such as those involved in arginine biosynthesis, pyruvate metabolism, and betaine biosynthesis, increased during the infection. We also identified numerous virulence proteins shared by both strains, including acid phosphatase (sapS), autolysin (Atl), leucocidin subunits (hlgA, hlgB), MAP protein, Immunoglobulin G-binding protein A (spa), and ATP-dependent molecular chaperones (clpB and clpC). Additionally, we observed strain-specific responses, with certain proteins being exclusively present in either the JE2 or ST398 strains, reflecting their unique adaptation strategies. Our study provides a comprehensive proteomic profile of MRSA during skin infection, highlighting both common and unique adaptive mechanisms employed by different strains. These findings offer valuable insights into the pathogenicity of MRSA and identify potential targets for novel therapeutic interventions for better management of skin infections. 200

201

P-I-0055 Antibiotic resistance detection and concomitant species identification of ESKAPE pathogens by proteomics Christian Blumenscheit1, Yvonne Pfeifer2, Guido Werner2, Charlyn John1, Andy Schneider1, Peter Lasch1, Jörg Döllinger1 1Robert Koch Institute, Proteomics and Spectroscopy, Berlin, Germany 2Robert Koch Institute, Nosocomial Pathogens and Antibiotic Resistance, Wernigerode, Germany Antimicrobial resistance (AMR) of bacteria represents a significant and emerging threat to public health. In 2019, 4.95 M deaths were associated and 1.27 M deaths were attributed to resistant bacteria. Appropriate antibiotic selection for therapy relies heavily on bacterial species identification and antibiotic susceptibility testing (AST) in clinical diagnostic laboratories. Currently, the taxonomic identity of clinical isolates is determined using MALDI- ToF MS-based biotyping, which is known for its speed and cost-effectiveness. Phenotypic AST methods, such as broth dilution or disk diffusion, which assess bacterial growth in the presence of various antibiotics are time- consuming. Recently, genomics has gained attraction as a complementary approach for predicting the AMR phenotype from the genotype. Integrating molecular data for microbial diagnostics offers several advantages over phenotypic methods. It allows for simultaneous species identification and AMR detection, eliminating the need for secondary cultivation, and can predict phenotypes beyond AMR, such as virulence features. The objective of this study was to establish and evaluate whole bacterial cell proteomics for the detection of antibiotic resistance and the identification of species of ESKAPE pathogens. A representative panel of clinical isolates comprised 129 bacterial strains from 16 species, representing the full spectrum of ESKAPE genera and 30 AMR determinants. In a previous small-scale proof-of-concept (POC) proteomics study we showed that simultaneous antibiotic resistance detection and species identification from DIA-MS data is possible with high specificity and sensitivity. The diagnostic workflow was based on universal bacterial sample preparation using SPEED and data-independent acquisition mass spectrometry (DIA-MS), which potentially enables high- throughput proteomics but poses major challenges for data analysis, such as the identification of bacterial species from large peptide sequence data sets and the detection of low abundant AMR determinants. The results of this study showed, that this new workflow (see Figure 1) enables the reporting of AMR phenotypes from primary cultures within two hours with high specificity (99.9%) and sensitivity (92.3%). Furthermore, it allows for species identification from very large sequence databases with high accuracy (91%). The integration of species identification and AMR detection into a single proteomics-DIA-MS-based workflow has the potential to massively improve the current diagnostic landscape, offering faster, more accurate, and comprehensive capabilities. This innovative approach has the potential to improve patient outcomes by enabling timely and appropriate AMR phenotypes and as such guide antibiotic therapy allowing to combat the growing threat of antimicrobial resistance. Fig. 1 (optional) 201

202

P-I-0056 Impact of different chemicals on the gut microbiota: an in vitro approach Vittoria Federica Borrini1, Denise Manfredi2, Elettra Barberis1, Francesco Dondero1, Marcello Manfredi2 1University of Piemonte Orientale, Department of Sciences and Technological Innovation, Alessandria, Italy 2University of Piemonte Orientale, Department of Translational Medicine, Novara, Italy The human gut microbiota is the totality of the microorganisms that colonizes our gut and many studies have demonstrated its involvement in human health and in the development of many diseases. Humans get in contact with hundreds of chemicals every day, thus the gut microbiota is daily exposed to many factors that could alter its status. The field of gut microbiota toxicology is still relatively new, more data are needed to evaluate the impact of chemicals on human health. Given this, an already validated method was used to study the effects that different substances have on the gut microbiota. We employed an in vitro method for the culture of gut microbiota that maintains the functional and compositional profiles of in vivo gut microbiota. Then, stool samples from healthy donors were used to test the impact of chemicals on the gut microbiota using metaproteomic and metabolomic analyses. The in vitro gut microbiota models were treated separately with the several chemicals, including PFOA, six replacement PFAS, Glyphosate and Cypermethrin, at different concentrations. Control samples grown without chemicals were also prepared and analyzed. The analysis was performed on a UHPLC Vanquish System coupled with an Orbitrap Q Exactive Plus. The results of the analysis were elaborated using Metalab MAG software. The metaproteomic analyses of the in vitro experiments were used to perform both phylogenetic and functional analysis. For example, at the level of phyla only the stimulation with Glyphosate led to statistically significant changes. For the other taxonomic levels investigated -orders, genera and species- all the chemicals led to significant alterations. Almost all modulated taxa were previously studied either to be important for the health and stability of the gut (decreased levels after treatments) or involved in some inflammatory processes (increased levels after treatments). The functional analysis of the treated samples treated showed a clear distinct pattern, at the molecular functions and biological processes levels, in respect with the control group. Finally, the stimulation of the gut microbiota with the six replacement PFAS induced significant microbiota changes. The in-vitro approach coupled with metaproteomics is a valuable tool to investigate the impact of chemicals on the gut microbiota. 202

203

P-I-0057 Metaproteomic analysis of gut microbiota responses to acute and prolonged xenobiotic exposure using a Simplified Human Microbial Model (SIHUMIx) Victor Castaneda-Monsalve1, Sven-Bastiaan Haange1, Ulrike Rolle-Kampczyk1, Martin von Bergen1,2,3, Nico Jehmlich1 1Helmholtz Centre for Environmental Research (UFZ), Molecular Toxicology, Leipzig, Germany 2University of Leipzig, Institute of Biochemistry, Faculty of Biosciences, Pharmacy and Psychology, Leipzig, Germany 3German Centre for Integrative Biodiversity Research (iDiv) Halle-Jena-Leipzig, Leipzig, Germany The human gut microbiota, a complex symbiotic community, performs crucial functions such as breaking down complex dietary fibers, regulating the immune system, maintaining intestinal health, and biotransforming xenobiotics. However, xenobiotic intake can disrupt microbial composition, impairing these functions and posing risks to host health. While research on individual bacterial species has highlighted chemical interactions and their effects on bacterial metabolism, the overall impact on the entire gut microbiome remains insufficiently understood. In this study, we explore how a simplified human gut microbial model (SIHUMIx) responds to acute and prolonged exposure to xenobiotics using a metaproteomic approach. Our aim is to capture the structural and functional dynamics of the microbiota, providing a comprehensive understanding beyond traditional genomic profiling. For acute exposure, SIHUMIx was exposed to 90 xenobiotics in 96 deep-well plates for 24 hours. Changes in community structure and function were assessed using metaproteomics. One treatment that significantly altered the SIHUMIx community was selected for prolonged exposure in bioreactors, where it was continuously exposed for six days, followed by a four-day recovery period. Similar analyses of changes in structure and function were conducted. We observed that 44 out of the 90 xenobiotics influenced SIHUMIx's structure or function. Notably, the colorant Brilliant Blue caused a decrease in SCFAs, reduced the abundance of C. butyricum, and altered several pathways. Therefore, it was chosen for a prolonged exposure study. Exposure to this food colorant increased the abundance of Bacteroides thetaiotaomicron while decreasing other species. It also altered the community's metabolic profile by affecting pathways involved in energy metabolism, short- chain fatty acid metabolism, amino acid metabolism, secretion metabolism, and antibiotic resistance. This study presents two efficient in vitro techniques that provide valuable insights into the effects of a broad range of chemicals on a human microbiota model. Environmentally relevant concentrations of these chemicals caused significant changes in bacterial abundance and altered key metabolic pathways. The use of SIHUMIx allows for a detailed description of effects due to its controlled complexity. The use of metaproteomics offers a robust approach, providing broader insights into how bacterial metabolic phenotypes are altered after exposure and identifying proteins involved in these changes. Our approach is essential for the functional characterization of chemical-microbiota interactions, which is highly relevant for risk assessments by regulatory agencies. Fig. 1 (optional) 203

204

P-I-0058 Detection of lipid A and its modifications by MALDI-TOF in Pseudomonas aeruginosa isolates with or without colistin selection pressure Elodie Degout-Charmette1, Nadine Perrot1, Justine Lannes1, Emilie Bisceglia1, Jean-Marc Roche1, Jean-Philippe Charrier1 1bioMerieux, Marcy l'Etoile, France Introduction: Colistin antibiotic (i.e., Polymyxin E) is known since the 1950"s, but its use into the clinic was suspended for years due to side effects. Today, it is once again prescribed in humans as last-resort therapy for otherwise untreatable Gram-negative infections (e.g., Pseudomonas aeruginosa infections). Colistin kills efficiently bacteria by disrupting the outer membrane integrity, but unfortunately resistance and heteroresistance (HR) have been described. HR is a poorly characterized phenomenon where subpopulations of apparently isogenic bacteria exhibit a range of antibiotic susceptibilities. Colistin resistance and HR are mainly due to the covalent modification of Lipid A (LpA) membrane by positively charged groups, such as 4-amino-4- desoxy-L-arabinose (L-Ara-4N) or phosphoethanolamine (pEtN). Mass differences between native and modified LpA peaks are observable by MALDI-TOF and allow the detection of colistin resistance but were not demonstrated on possibly HR isolates. The purpose of this study was to evaluate MALDI-TOF detection of native/modified LpA peaks in clinical isolates of Pseudomonas aeruginosa, a species known for HR prevalence, grown with or without colistin selection pressure. Materials and methods: Two P. aeruginosa controls (susceptible & resistant) and 5 P. aeruginosa clinical isolates showing variable minimum inhibitory concentration (MIC) and consequently suspected of HR, were characterized according to : • • • the percentage of resistant cells in a total bacterial population after culture on non-selective versus selective colistin medium; the detection of LpA, LpA-L-Ara-4N, and LpA-pEtN by MALDI-TOF (VITEK®MS); the MIC, measured by broth microdilution (BMD), of the different bacterial subpopulations, resistant or not. Results: After selection on colistin agar, the possibly HR strains showed a very significant increase in their proportion of resistant cells, accompanied by an increase in MIC and the appearance of modified LpA. However, before selective culture, the low percentage of resistant cells in the initial populations was insufficient to allow the observation of modified LpA characteristic peaks. Conclusion: MALDI-TOF detection of modified LpA does not seem suitable for the characterization of colistin resistance for strains of P. aeruginosa, suspected of HR, cultivated without selection pressure. 204

205

P-I-0060 The HadB-HadD complex, from the fatty acid synthase type II system, is important for biosynthesis, fitness and virulence of Mycobacterium smegmatis and Mycobacterium tuberculosis Manuelle Ducoux-Petit1, Pascaline Bories2, Julie Rima1, Richard Boulon3, Cyril Lefebvre1, Samuel Tranier1, Julien Marcoux1, Hedia Marrakchi1, Cecile Bon1, Odile Burlet-Schiltz1, Fabienne Bardou1, Annaik Quemard1 1CNRS, IPBS, Toulouse, France 2Laval University CHU – Quebec, Research center, Quebec, Canada 3INRS Institut national de la recherche scientifique, Montreal, Canada Worldwide Mycobacterium tuberculosis (Mtb), the causative agent of tuberculosis, is the second leading infectious killer and multidrug resistance severely hampers disease control. Additionally, the number of diseases" cases caused by other nontuberculous mycobacteria is alarmingly increasing. Thanks to the nontuberculous and nonpathogenic effects, mycobacteria such as Mycobacterium smegmatis (Msm), can be used routinely and easily in the laboratory. Mycolic acids (MA) are a unique category of lipids that are essential for viability, virulence, and persistence of mycobacteria. They are synthetized by enzymes of the the fatty acid synthase type II system (FAS-II), which thus represent an important class of drug targets. In order to identify new therapeutic targets, we used a combination of affinity purification experiments and proteomics analyses to explore the composition and organization of the FasII system in mycobacteria. We expressed tagged version of HadA, member of the Had family of (3R)-hydroxyacyl-ACP dehydratase proteins, taking part in the FAS-II system from either Msm or Mtb, purified the associated complexes, and characterized them by mass spectrometry analysis. We discovered a mycobacterial protein, HadD, displaying highly specific interaction with the dehydratases HadA and HadB of FAS-II, and investigated its function in Msm and Mtb. In Msm, a hadD knockout mutant produced only the medium-size alpha"-MAs. Data strongly suggest that HadD is involved in building the third meromycolic segment during the late FAS-II elongation cycles, leading to the synthesis of the full-size alpha- and epoxy-MAs. The change in the envelope composition induced by hadD inactivation strongly alters the bacterial fitness and capacities to aggregate and to assemble into colonies or biofilms. In Mtb, we showed that the HadD protein is dedicated mainly to the production of keto-mycolic acids and plays a determinant role in Mtb biofilm formation and virulence. We discovered that HAD activity requires the formation of a tight heterotetramer between HadD and HadB. Using biochemical, structural, and cell-based analyses, we showed that HadB is the catalytic subunit, whereas HadD is involved in substrate binding. Our study shows that the HadB-HadD, and not HadD alone, is the biologically relevant functional unit. These results suggest that the target to consider is not an isolated subunit, but the whole HadB-HadD complex, either in Mtb or Msm, and have important implications for designing innovative antivirulence molecules to fight tuberculosis as well as other pathologies caused by mycobacteria. 205

206

P-I-0061 Proteomics and metabolomics analyses of Streptococcus agalactiae isolates from human and animal sources Shymaa Enany1, Yasmin Tartour2, Rania Kishk1, Ahmed Gadallah1, Eman Ahmed3, Sameh Magdeldin3 1Suez Canal University, Microbiology, Ismailia, Egypt 2Zagazig University, Faculty of Veterinary Medicine, Zagazig, Egypt 3Children's Cancer Hospital CCHE-57357, Cairo, Egypt Streptococcus agalactiae (S. agalactiae), group B Streptococcus (GBS), a major cause of infection in a wide variety of diseases, have been compared in different human and animal sources. We aimed to compare the bacterial proteome and metabolome profiles of human and animal S. agalactiae strains to delineate biological interactions relevant to infection. With the innovative advancement in mass spectrometry, a comparative result between both strains provided a solid impression of different responses to the host. For instance, stress-related proteins (Asp23/Gls24 family envelope stress response protein and heat shock protein 70), which play a role in the survival of GBS under extreme environmental conditions or during treatment, are highly expressed in human and animal strains. One human strain contains ꞵ-lactamase (serine hydrolase) and biofilm regulatory protein (lytR), which are important virulence regulators and potential targets for the design of novel antimicrobials. Another human strain contains the aminoglycosides-resistance bifunctional AAC/APH (A0A0U2QMQ5) protein, which confers resistance to almost all clinically used aminoglycosides. Fifteen different metabolites were annotated between the two groups. L-aspartic acid, ureidopropionic acid, adenosine monophosphate, L- tryptophan, and guanosine monophosphate were annotated at higher levels in human strains. Butyric acid, fumaric acid, isoleucine, leucine, and hippuric acid have been found in both human and animal strains. Certain metabolites were uniquely expressed in animal strains, with fold changes greater than 2. For example, putrescine modulates biofilm formation. Overall, this study provides biological insights into the substantial possible bacterial response reflected in its macromolecular production, either at the proteomic or metabolomic level. 206

207

P-I-0063 Universal, unbiased DIA-Based metaproteomics sample processing and database search workflow for comprehensive metaproteomic analysis Tariq Ganief1, Juandre Makaka2, Tara Miller2, Nina Radzey2, Andrea Abrahams2, Anna Happel2, Lindi Masson2,3, Heather Jaspan2,4, Jonathan Blackburn2 1University of Cape Town, Integrative Biomedical Sciences, Cape Town, South Africa 2IDM UCT, Integrative biomedical sciences, Cape Town, South Africa 3Burnet Institute, Melbourne, Australia 4Seattle Children’s Research Institute, Seattle, WA, United States Microbiomes play a pivotal role in human health, environmental ecosystems, and agriculture. While genomic approaches dominate microbiome research, metaproteomics offers unique insights into the microbiomes functionality. However, challenges persist, including universal sample preparation, analysis depth, and efficient database search strategies. Here, we introduce a universal metaproteomic sample preparation method. Our approach ensures lysis of both gram-positive and gram-negative bacteria without the use of detergents, chaotropic agents, or any other mass spectrometry-incompatible reagents. The streamlined 96-well format enables rapid protein extraction, purification, and digestion. Coupled with Evosep LC and data-independent acquisition (DIA) mass spectrometry, we achieves highly efficient metaproteomic analysis of up to 60 samples per day while maintaining deep metaproteomic coverage. As shotgun proteomics spectral matching requires some prior knowledge, proteome databases and/or spectral libraries are required. Since its inception, this simple reality has plagued metaproteomics and several strategies have been developed to address this; these are typically not applicable to DIA-based metaproteomics. Although we have previously successfully used 16s analysis coupled with a DIA-NN library free search, these analyses rely on prior genomic analysis. To achieve a truly unbiased, comprehensive means to independently acquire quantitative taxonomic and metaproteomic data directly from a single mass spectrometry run, we employ a multi-tiered search strategy which progressively increases specificity while preserving sensitivity without compromising search the FDR using all Uniprot bacterial proteins as an intital reference proteome database. To validate the sensitivity and specificity of this approach, we analysed a well-characterized commercial microbiome preparation and achieving nearly 100% identification of bacterial biomass of a gut microbiome in a single injection. Thus, our universal, rapid, and unbiased DIA-based metaproteomics requires less than 2 hours of hands-on sample preparation and identifies nearly 100% of microbial biomass along with deep proteome coverage facilitating adaptable and unbiased metaproteomic analysis with applications in clinical, agricultural, and environmental metaproteomics. 207

208

P-I-0064 Exploring the microbiome in long COVID patients: a metaproteomic study Carmen García Durán1, Leticia Gómez Artiguez1, Samuel De la Camara2, Ines Zapico2, Maria Luisa Hernáez2, Lucía Monteoliva1, Concha Gil1,2 1Complutense University of Madrid, Microbiology and Parasitology Department, Madrid, Spain 2Complutense University of Madrid, Proteomics Unit, Madrid, Spain Most people infected with COVID-19 recover over time, but a percentage develop what is known as long COVID. Long COVID is the persistence or development of new symptoms with no other explanation 3 months after the SARS COV-2 infection and lasting at least 4 weeks. Previous research has confirmed a relationship between an altered gut microbiota and COVID-19, suggesting that a similar association may exist with long COVID. In this study, our main objective was to investigate changes in microbiota composition and function in patients with long COVID using metaproteomics. A cohort of 100 patients with long COVID and 34 controls was recruited. Faecal and saliva samples were collected from each patient and control. Faecal samples were processed using differential centrifugation and a combination of bead beating and sonication, whereas saliva samples were processed by sonication only. Peptides were analysed on a timsTOF Pro2 using MSFragger for protein and peptide identification. The Integrated Gene Catalog and the Human Oral Microbiome Database were used for microbial protein identification of gut and oral samples, respectively. Metalab program was used for taxonomic and functional annotation of gut microbiota samples, and for the taxonomic analysis of oral samples. The functional analysis of salivary microbiota was performed using the Unipept Desktop application due to the incompatibility of the database with the functional annotation option of Metalab. In total, we identified 7.437 protein groups, corresponding to 557 different taxa in the gut microbiota and 1.154 protein groups corresponding to 344 taxa in the oral microbiota in at least 50% of the samples. RStudio was used for visualization and statistical analysis. We observed differences in the relative abundance of different taxa between groups (long COVID and control) in both samples. For example, Akkermansia genus, which is thought to have beneficial effects on host health, was significantly more abundant in control samples. A taxon-symptom correlation analysis was performed to evaluate the importance of these taxa in long COVID condition. We also identified 3497 human proteins in saliva samples. Functional analysis of these proteins revealed interesting results. For example, the String analysis showed a significant cluster associated with fibrinogen proteins. High levels of fibrinogen have been associated with a poor prognosis in COVID-19, and coagulation abnormalities have also been observed in long COVID patients. In conclusion, our findings highlight the importance of studying the microbiota to understand and potentially manage the complexities of long COVID. This research provides valuable insights into the disease that may improve patient outcomes. 208

209

P-I-0065 Lysine acetylation and its regulatory enzymes in Cyanobacteria Feng Ge1, Mingkun Yang1 1Institute of Hydrobiology, Chinese Academy of Sciences, Wuhan, China Lysine acetylation is a conserved regulatory posttranslational protein modification that is performed by lysine acetyltransferases (KATs). By catalyzing the transfer of acetyl groups to substrate proteins, KATs play critical regulatory roles in all domains of life; however, no KATs have yet been identified in cyanobacteria. Here, we tested all predicted KATs in the cyanobacterium Synechococcus sp. PCC 7002 (Syn7002) and demonstrated that A1596, which we named cyanobacterial Gcn5-related N-acetyltransferase (cGNAT2), can catalyze lysine acetylation in vivo and in vitro. Eight amino acid residues were identified as the key residues in the putative active site of cGNAT2, as indicated by structural simulation and site-directed mutagenesis. The loss of cGNAT2 altered both growth and photosynthetic electron transport in Syn7002. In addition, quantitative analysis of the lysine acetylome identified 548 endogenous substrates of cGNAT2 in Syn7002. We further demonstrated that cGNAT2 can acetylate NAD(P)H dehydrogenase J (NdhJ) in vivo and in vitro, with the inability to acetylate K89 residues, thus decreasing NdhJ activity and affecting both growth and electron transport in Syn7002. In summary, this study identified a KAT in cyanobacteria and revealed that cGNAT2 regulates growth and photosynthesis in Syn7002 through an acetylation-mediated mechanism. 209

210

P-I-0066 Unravelling the role of Candida albicans Prn1 in the Oxidative stress response through a proteomic approach. Study of Prn1 orthologues Víctor Arribas1, Lucía Monteoliva1, Maria Luisa Hernáez2, Concha Gil1,2, Gloria Molero1 1Complutense University of Madrid, Microbiology and Parasitology, Madrid, Spain 2Complutense University of Madrid, Proteomics Unit, Madrid, Spain Candida albicans Prn1 is a protein that shares similarities with mammalian Pirin but with an unknown function in the yeast. Orthologues of Prn1 have been identified in other pathogenic fungi but not in Saccharomyces cerevisiae, suggesting a relationship with pathogenesis. Prn1 increase in abundance after H2O2 treatment has been previously shown [1]. Quantitative proteomic assay (DDA-MS) in a wild-type (SN250) and a prn1∆ strains treated with hydrogen peroxide identified around 1,800 proteins. These assays indicated a lower increase of proteins with oxidoreductase activity after treatment in the prn1∆ strain compared to the wild-type, as well as an increase in proteasome-activating proteins and a decrease in translation-involved proteins. Accordingly, Prn1 absence, under H2O2 treatment, led to a lower survival rate and a higher percentage of apoptosis, together with higher reactive oxygen species levels and higher proteasome activity. Besides, remarkable differences in the abundance of some transcription factors were observed between prn1∆ mutant and wild-type strains. Among them, Mnl1, involved in Prn1 expression, Nrg1, a Mnl1 antagonist, or Bas1, presented an inverse pattern of expression between both strains. Interestingly, orf19.4850, a protein orthologue to S. cerevisiae Cub1, was only expressed in the wild-type strain under the oxidative stress. Studies with the orf19.4850∆/orf19.4850 mutant strain have shown its involvement in the response to H2O2 and in proteasome function. Under basal conditions, the proteomics results also suggest a possible involvement of Prn1 in mitochondrial oxidative stress detoxication. All the experiments confirm Prn1 as a relevant actor in C. albicans oxidative response. Surprisingly, the Prn1 orthologues, Prn2, Prn3 and Prn4, have not shown a redundant function or expression. The proteomics analyses only detected a slight but significant increase in Prn4 abundance, which is the most similar to Prn1, in the wild-type strain after H2O2 treatment, but not in the prn1∆. Prn2 and Prn3 were not detected in any of the four conditions studied. Thus, we are currently studying the role of Prn2, Prn3 and Prn4 under oxidative stress and other conditions that could give us a clue about their function. 1. Amador-García A, et al. mSystems. 2021 210

211

P-I-0067 Ultra-sensitive metaproteomics (uMetaP) redefines the "dark" metaproteome, enables single-bacterium resolution, and discovers hidden functions in the gut microbiome Feng Xian1, Christoph Krisp2, Malena Brenek1, Ranjith Kumar1, Manuela Schmidt1, David Gomez Varela1 1University of Vienna, Pharmaceutical Sciences, Vienna, Austria 2Bruker Daltonics GmbH & Co. KG, Bremen, Germany The interplay between members of microbial communities (microbiomes) is crucial for planetary health. While genomic methods have greatly expanded the taxonomic characterisation of microbiomes, complementary techniques for directly quantifying biological functions are needed for deeper understanding. Metaproteomics, which analyzes microbial samples using liquid chromatography-tandem mass spectrometry (LC-MS)-based proteomics, has emerged as a powerful tool for studying host-microbiome interactions. It offers a direct view of functional changes within microbes and their host. However, metaproteomics faces significant challenges compared to single-organism LC-MS. Microbial samples are incredibly complex, containing an estimated 100 million unique peptide species with a vast dynamic range. Consequently, over 80% of bacterial species detected by common genomic methods (e.g., 16S rRNA) remain undetected by metaproteomics, limiting functional characterization. A significant leap in sensitivity, depth, and accuracy is needed to tap into this "dark metaproteome". We present a novel, ultra-sensitive metaproteomic solution called uMetaP, which leverages cutting-edge technologies for unprecedented sensitivity. Utilizing mouse feces as a model, uMetaP achieves dramatic improvements in key metrics. This includes quantifying unprecedented amounts of protein groups and associated functional pathways for both the host and its microbiome (47925 protein groups and 223 KEGG pathways). Remarkably, uMetaP quantified 220 microbial species, which is comparable to the average amount reached by full-length 16S rRNA, with exceptional reproducibility and quantitative precision in DIA-PASEF runs using low sample amounts (25 ng of peptides). Additionally, uMetaP revealed (i) previously hidden functions, identifying 3252 proteins of unknown function (PUFs), (ii) taxonomic and functional annotation of 2,181 small proteins (sProt), and (iii) 553 proteins gathering predicted antimicrobial peptide sequences (AMPs). Leveraging uMetaP with a meticulous experimental design utilizing novel SILAC-labelled bacteria, we addressed two longstanding challenges in metaproteomics: 1) the true limit of detection for the so-called "dark metaproteome", and 2) detection at the single-bacterium resolution. We confidently detected 10000 spiked-in bacteria among a theoretical amount of 10 billion, lowering the limit of detection 5000-fold. Moreover, we robustly identified species-specific peptides from a 500 fg two-bacteria peptide mix with remarkable quantification precision and accuracy. From deciphering the interplay of billions of microorganisms with the host to exploring microbial heterogeneity, uMetaP represents a quantum leap in metaproteomics with the potential to open new avenues for our understanding of the microbial world and its connection to health and disease. 211

212

P-I-0068 Unveiling the distant interaction of staphylococcus aureus with pseudomonas aeruginosa: a novel analysis of long-term and short-term responses on the proteome level Jonas Grossmann1,2, Lukas Schwyter3, Selina Niggli3, Witold Wolski1,2, Natalia Zajac1, Lucy Poveda2, Panse Christian1,2, Ralph Schlapbach1, Rolf Kümmerli3 1University of Zürich, Functional Genomics Center, Zurich, Switzerland 2Swiss Institute of Bioinformatics (SIB), Lausanne, Switzerland 3University of Zürich, Quantitative Biomedicine, Zurich, Switzerland Staphylococcus aureus and Pseudomonas aeruginosa frequently co-occur in infections, and there is evidence that their interactions negatively affect disease outcome. P. aeruginosa is known to inhibit S. aureus through the secretion of inhibitory compounds and therefore P. aeruginosa is thought to be the dominant species in this interaction. However, S. aureus may adapt to inhibitory compounds in cases of prolonged co-occurrence in chronic infections. We performed a 30-day evolution experiment with S. aureus exposed to P. aeruginosa supernatant containing growth-inhibitory compounds. While afterwards various phenotypic changes were observed, only a few genetic mutations were identified after whole genome sequencing. Based on these findings we hypothesize that resistance results from a combination of phenotypic responses and genetic adaptations. We tested this hypothesis using label-free quantitative proteomics and targeted metabolomics for one particular evolved S. aureus clone and its ancestor grown in either only normal media or exposed to growth-inhibitory compounds secreted from P. aeruginosa. As predicted, we found that S. aureus shows fundamental phenotypic changes at the proteome level when exposed to P. aeruginosa supernatant regardless of whether S. aureus had evolved in the supernatant or not. Using the prolfqua R package and a linear model with two factors, we identified differentially expressed proteins associated with both genotype and growth-condition effects. Major changes involved the downregulation of virulence factors and a deregulation of multiple membrane transporters. At the genetic level, only a single mutation in tcyA encoding a transmembrane transporter was observed. We further show that this mutation leads to a non-functional protein and therefore a complete loss of function and confers protection against selenocysteine, a toxic compound secreted by P. aeruginosa. Our findings suggest a two-stage response of S. aureus to P. aeruginosa inhibition: A phenotypic response securing instant survival and growth, followed by genetic adaptation reducing the negative effects of toxic compounds in the long-term. This two-stage model could explain how pathogens adapt to one another and how co-existence in polymicrobial infections can arise. Fig. 1 (optional) 212

213

P-I-0069 A metaproteomic landscape of human gut microbiome in controls and colorectal cancer progression stages Alex Henneman1, Roza Opperman1, Thang V. Pham1, Sofie Bosch1, Meike de Wit2, Beatriz Carvalho2, Remond Fijneman2, Gerrit Meijer2, Nanne de Boer1, Connie Jimenez1 1Amsterdam University Medical Center, Medical Oncology, Amsterdam, Netherlands 2Netherlands Cancer Institute, Department of Pathology, Amsterdam, Netherlands Introduction: The human gut microbiome is a complex ecosystem that is increasingly connected to a variety of conditions, such as obesity and depression. To date, the majority of current investigations use genome-wide or targeted DNA sequencing, such as 16S. This type of approaches fall short of the determination and quantification of the actual proteins produced and secreted by both host and microbiome, while these may provide more complete insights. To that end, we used mass spectrometry to examine a large set of stool samples of healthy controls, adenoma and cancer patients to map the extent of the microbiome constitution and the corresponding variation across these progression stages. Importantly, current small-host-number data analysis approaches are inadequate in a metaproteomics setting, requiring the development of novel processing and visualization methods, to which we aim to contribute to foster the emerging field of gut microbiome metaproteomics. Methods: Approximately 380 human stool samples, of which a subset also had corresponding 16S sequencing data, were profiled using LC-MS/MS in data-dependent acquisition mode. Resulting fragmentation spectra were searched using MSFragger in search space composed of the human proteome supplemented with the proteins of approx. 350 carefully selected gut bacterial species from the Human Microbiome Project. Peptide spectrum matches were processed using Trans Proteomic Pipeline elements in conjunction with in-house developed software into FDR-controlled count-based peptide abundance tables. A novel method, specially developed for our metaproteomics applications, was used for aggregating these peptide tables into protein tables and also higher-level taxonomic units. Subsequently, specialized diversity estimation and visualization methods were developed and applied to this dataset. Results and conclusions: This study provides an extensive and complex proteomic human-microbial landscape extending across several orders of magnitude in abundance level. A clear signal reflected by both human and microbial protein components can be discerned across normal controls, adenoma and cancer patient stool samples. Proteomics and 16S signatures are largely in agreement, although the unbiased nature of a proteomics- based approach provides a more informative and higher resolution landscape. 213

214

P-I-0070 Meta-analysis of fecal metaproteomic datasets reveals robust human and microbial signatures for inflammatory bowel disease Robert Heyer1, Maximilian Wolf2, Kay Schallert1, Luca Knipper2 1ISAS Dortmund, MdOA, Dortmund, Germany 2Bielefeld University, Bielefeld, Germany Introduction: Metaproteomics provides valuable insights into the processes within the human gastrointestinal tract, helping us understand the interplay between humans, their gut microbiome, and various diseases, including colon cancer, diabetes, and inflammatory bowel diseases (IBD). However, comparing different studies is challenging due to significant variations between individuals and differences in experimental and bioinformatic workflows. This complicates the integration of studies and the identification of general biomarkers for different diseases. Materials and Methods: To draw general conclusions from metaproteomics toward the development of general biomarkers, we conducted a fecal metaproteomics metastudy using the following inclusion criteria: (i) based on Data-Dependent Acquisition (DDA), (ii) published before November 2023, (iii) no chemical labeling, and (iv) spectral data available on PRIDE. We performed a complete bioinformatic re-analysis of the data from all studies using the MetaProteomeAnalyzer software and developed a novel biostatistical workflow for data analysis. Results: Nine studies met the inclusion criteria, covering a total of 429 samples, While clustering revealed a separation based on different labs, experimental workflows, and mass spectrometers rather than diseases, variance analysis enabled separation based on diseases. For example, focusing on IBD, we identified 28 human biomarker candidates, mostly neutrophil- and immunoglobulin-derived, and 55 microbial biomarker candidates. These biomarkers allowed for a distinction between healthy and diseased states with an accuracy of 95%. Moreover, integrating all the different datasets resulted in smaller p-values than individual studies, providing more robust results. Conclusion: Our study demonstrates a promising strategy for conducting metaproteomics, confirming known biomarker candidates, and identifying novel candidates for fecal diagnosis of diseases such as IBD. This approach enhances the validity of results and supports the development of general biomarkers for disease diagnosis. 214

215

P-I-0071 The Peptonizer2000: probaility-based taxonomic profiling of microbiome samples Tanja Holstein1 1Ghent University, Center for Medical Biotechnology, Department of Biomolecular Medicine, Ghent, Belgium Metaproteomics, the large-scale study of proteins from microbial communities, presents complex challenges in taxonomic inference due to sequence homologies between proteins within and across taxa. Commonly, taxonomic inference relies on heuristics such as peptide-spectrum match counting. Only few more advanced methods are available. We introduce the Peptonizer2000, a novel graphical model-based workflow designed to provide high-resolution taxonomic identifications of metaproteomic samples with associated confidence scores. Our tool integrates peptide scores from any proteomic search engine with peptide-taxon mappings from Unipept. We subsequently represent the joint probability distribution of peptides and taxa through a factor graph and use belief propagation to compute marginal probabilities of a taxon's presence, resulting in taxonomic identifications with associated probability scores. We demonstrate the Peptonizer2000's accuracy and robustness through the analysis of various publicly available metaproteomic samples, including lab-assembled communities and actual microbiomes. We showcase the Peptonizer's ability to deliver reliable probabilistic taxonomic identifications at various taxonomic resolution levels. Our results highlight the Peptonizer2000's potential to improve the specificity and confidence of taxonomic assignments in metaproteomics, providing a valuable resource for the study of complex microbial communities. 215

216

P-I-0072 Investigating key host, variant and microbial peptides for detection of oral cancer using a multi-omics approach Beverly Wuertz1, Ruben Shreshta2, Monica Kruk3, Subina Mehta4, Matthew Willetts5, Frank Ondrey1, Timothy Griffin4, Pratik Jagtap4 1University of Minnesota, Otolaryngology, Minneapolis, MN, United States 2Bruker Scientific LLC, San Jose, CA, United States 3University of Minnesota, Minneapolis, MN, United States 4University of Minnesota, Biochemistry, Molecular Biology and Biophysics, Minneapolis, MN, United States 5Bruker Scientific LLC, Billerica, MA, United States There is an unmet need for risk stratification biomarkers for precancerous oral leukoplakia patients. While host biomarkers are available for detection, there is a clear need to expand the biomarker panel to improve predictive performance. We have used the latest methods for enrichment of low-abundance proteins from non- invasively collected samples, sensitive mass spectrometry (MS) methods, and bioinformatic analysis to delve deeper into microbial proteome along with host and variant proteins from precancerous patients. Oral rinse samples from six oral precancerous lesion patients (pre/post chemoprevention agent treatment) were processed using the PreOmics ENRICH kit to facilitate the detection of low abundance proteins. Digested proteins were analyzed by DIA-PASEF on a hybrid TIMS QTOF mass spectrometer (Bruker). RNASeq analysis was used to generate customized proteogenomics and metaproteomics databases. The MS data was searched against the human proteome along with variant proteins and microbial proteins using Spectronaut 19 (Biognosys). Bioinformatic and statistical analysis was performed to detect differentially expressed host proteins, microbial proteins and novel proteoforms. Enrichment of low abundance proteins coupled with DIA-PASEF MS allowed for deeper quantitative analysis of oral host proteome and taxonomic and functional analysis of oral microbiome. Several human, microbial and variant proteins were detected to be differentially abundant in pretreatment and treated samples. Out of the 5894 human proteins quantified across the six replicates, 81 proteins were differentially abundant (51 in treated samples and 30 in pretreatment samples). Proteins associated with Gamma-carboxylation of protein precursors and complement cascade were upregulated and vesicle-mediated transport, apoptosis and inflammasome pathways were downregulated after treatment. Metaproteomics analysis detected 6830 microbial protein groups, out of which 38 microbial proteins were detected to be differentially abundant amongst treated and untreated samples. Notably, dextransucrase (glycotransferase involved in exopolysaccharide synthesis and promotes biofilm formation) from Streptococcus salivarius was abundant in treated samples. Additionally, D-Ala D-Ala carboxypeptidase (involved in peptidoglycan synthesis) from S. salivarius was also abundant in treated samples. Lastly, Alkyl hydroperoxide reductase subunit C (catalytic subunit responsible for the detoxification of reactive oxygen species) from Veillonella was found to be abundant in pretreated samples. Finally, variant proteins and peptides associated with PYCARD and GDI2 are upregulated after treatment. Taken together we have more than 300 peptides (127 host, 125 microbial and 80 variant peptides) that will prioritized to generate a biomarker panel which can be later assessed for a) risk stratification in oral leukoplakia and b) deciphering mechanisms for host-microorganism interaction in oral carcinogenesis. 216

218

P-I-0073 Clinical metaproteomics workflow to study host-microbiome dynamics Katherine Do1, Subina Mehta2, Surbhi Bihani3, Monica Kruk4, Aryan Gupta3, Kevin Murray2, Andrew Rajczewski2, Reid Wagner2, Dechen Bhuming2, Kristin Boylan4, Amy Skubitz4, Theresa Laguna5, Sanjeeva Srivastava3, Timothy Griffin2, Pratik Jagtap1 1University of Minnesota, Biochemistry, Molecular Biology and Biophysics, Maple Grove, MN, United States 2University of Minnesota, Biochemistry, Molecular Biology and Biophysics, Minneapolis, MN, United States 3Indian Institute of Technology Bombay, Mumbai, India 4University of Minnesota, Minneapolis, MN, United States 5Seattle Children’s Hospital, Department of Pediatrics, Division of Pulmonary and Sleep Medicine, Seattle, WA, United States Clinical metaproteomics has the potential to offer insights into the host-microbiome interactions underlying diseases. However, the field faces challenges in characterizing microbial proteins found in clinical samples, which are usually present at low abundance relative to the host proteins. As a solution, we have developed an integrated workflow coupling mass spectrometry-based analysis with customized bioinformatic identification, quantification, and prioritization of microbial proteins, enabling targeted assay development to investigate host- microbe dynamics in disease. The bioinformatics tools are implemented in the Galaxy ecosystem which offers integrated development and dissemination of complex bioinformatic workflows. The modular workflow integrates MetaNovo (to generate a reduced protein database), SearchGUI/PeptideShaker and MaxQuant (to generate peptide-spectral matches (PSMs) and quantification), PepQuery2 (to verify the quality of PSMs), Unipept (for taxonomic and functional annotation), and MSstatsTMT (for statistical analysis). We have utilized this workflow in ongoing projects to identify microbial peptide panels for cystic fibrosis (CF) disease progression studies, co-infection status during COVID-19 pandemic waves, and ovarian cancer biomarker discovery. In the CF study, broncho-alveolar samples from pediatric CF and disease control patients were characterized for their microbial diversity. In addition to the above described modules, targeted analysis of microbial and host proteins from individual samples led to detection of a peptide-panel that includes known and novel peptide targets that can be used to track host-microbe protein dynamics during CF progression. In another study, nasopharyngeal swabs from SARS-CoV-2 infected patients from a hospital in Mumbai, India, were analyzed to identify microbial peptides corresponding to potential co-infecting microorganisms during two pandemic waves. The study detected several peptides belonging to opportunistic microbial pathogens, such as Streptococcus pneumoniae, Klebsiella pneumoniae, Rhizopus microsporus, and Syncephalastrum racemosum. Microbial proteins with a role in stress response, gene expression, and DNA repair were upregulated in severe COVID-19 patients compared to those who tested negative for SARS-CoV-2. Lastly, we analyzed Papanicolaou (Pap) test protein samples from benign, normal, and ovarian cancer patients for host and microbial proteins that are differentially expressed. The results of the quantitative analysis of these samples, and the potential use of combined host-microbe protein markers for cancer detection, will be discussed during the presentation. The complete workflow, including training data and documentation, is available via the Galaxy Training Network, empowering non-expert studies. https://training.galaxyproject.org/training-material/learning-pathways/clinical-metaproteomics.html these powerful researchers to utilize tools in their clinical 218

220

P-I-0074 Multi-Omics analysis reveals mechanisms of colistin resistance in Acinetobacter nosocomialis Hayoung Lee1, Seung Il Kim1 1KBSI, Digital Omics analysis, Cheongju, South Korea Background: Acinetobacter nosocomialis, an opportunistic pathogen responsible for pneumonia and bacteremia, poses a significant threat in intensive care units. The increasing prevalence of colistin-resistant strains in hospital settings has raised serious concerns about patient safety and healthcare outcomes. Therefore, elucidating the mechanisms driving colistin resistance, effectively tracking the spread of resistant strains, and developing strategies to overcome the resistance are imperative. Methods: We conducted a comparative genomic analysis to uncover the antimicrobial resistance genes or underlying mechanisms by comparing susceptible and resistant bacterial isolates. Additionally, we examined the molecular characteristics of A. nosocomialis strains under colistin-exposed conditions (4 mg/L for 4 hours) using quantitative transcriptomics and proteomics. Label-free proteomics was performed through in-gel digestion, and the eluents were analyzed using LC-MS/MS with Orbitrap Q Exactive Plus. Results: Through comparative genomic analysis, we discovered a distinct subtype of the capsular polysaccharide (K) locus in A. nosocomialis KAN02, a colistin-resistant strain. This K locus is closely associated with capsular polysaccharide (CPS) production, which contributes to antimicrobial resistance and promotes survival through surface modification. Furthermore, we identified differentially expressed genes (DEGs) and differentially expressed proteins (DEPs). Our comprehensive transcriptomic and proteomic analysis revealed a total of 2,289 DEGs and 660 DEPs. Notably, 14 out of the 22 genes within the K locus showed significant upregulation in both omics datasets, indicating increased synthesis of the capsule in the KAN02 isolate. Conclusion: This study represents the first comprehensive multi-omics investigation aimed at enhancing our understanding of the mechanisms driving colistin resistance in A. nosocomialis. Through multi-omics analysis, we identified molecular characteristics associated with the mechanisms of colistin resistance in A. nosocomialis KAN02. Our findings revealed that the upregulation of the K locus, as assessed through multi-omics analysis, plays a significant role in conferring colistin resistance in A. nosocomialis KAN02. 220

221

P-I-0075 Metaproteomics reveals altered gut microbiota and differential responses to diet intervention in Crohn's disease patients according to disease location Stefano Levi Mortera1, Valeria Marzano1, Federica Rapisarda1, Chiara Marangelo1, Ilaria Pirona2, Pamela Vernocchi1, Marta Di Michele1, Federica Del Chierico1, Maria Alejandra Qintero3, Irina Fernandez3, Hajar Hazime3, Rose M Killian3, Luis Garces3, David H Kerman3, Amar R Deshpande3, Maria Teresa Abreu3, Lorenza Putignani1 1Bambino Gesù Children's Hospital, Human Microbiome, Rome, Italy 2Catholic University of the Sacred Heart, Istituto di Patologia Speciale Medica, Rome, Italy 3University of Miami, Division of Gastroenterology, Department of Medicine, Miami, FL, United States Crohn's disease (CD) is a type of inflammatory bowel disease (IBD) that causes chronic mucosal inflammation in different regions of the gastrointestinal tract, due to genetic and environmental factors. Diet is a key modifiable factor influencing the gut microbiome (GM) and a risk factor for CD. However, the impact of diet modulation on GM function in CD patients has not been carefully studied so far. In this work, we evaluated the effect of a high- fiber, low-fat diet (the Mi-IBD diet) on the GM function in CD patients. CD patients were divided in three main groups. The first one (Group 1, n=10) followed a Mi-IBD diet after a one- time diet counseling. The second one (Group 2, n=16) received catered food for 8 weeks while the third one (Group 3, n=22 dyads) received catered food for 8 weeks for a CD patient and a healthy household control (HHC). Group 3 patients and HHCs also participated in Dyadic Psychosocial Support (DPS) sessions. Stool samples were collected at baseline, week 8, and week 36 and analyzed by liquid chromatography-tandem mass spectrometry. Results showed a high heterogeneity between patients, groups of subjects and across the time points, making quite challenging the identification of a modulation of the GM functionality prompted by the diet. Diet-induced changes involving the activation of bacterial metabolic pathways were observed, including those related to carbohydrate metabolism due to an increased fiber uptake. Interestingly, the activity of taxa accomplishing such functions was affected by the inflamed environment and the microbial composition underlying the metaproteomic changes differed between patients with ileal only disease or with colonic involvement. The Mi-IBD diet changed the microbial function of both CD patients and HHC. Our metaproteomic results highlight functional differences in the microbiome according to disease location. Comparing the effect of a diet intervention between diverse CD patients and HHCs provided additional insight into diet optimization to achieve defined salutary functions. Fig. 1 (optional) 221

222

P-I-0076 PhyloFunc: phylogeny-informed functional distance as a new ecological metric for metaproteomic data analysis Luman Wang1, Caitlin M.A. Simopoulos2, Joeselle M. Serrana2, Zhibin Ning2, Boyan Sun3, Jinhui Yuan3, Daniel Figeys2, Leyuan Li3 1Peking University, Beijing, China 2University of Ottawa, Ottawa, Canada 3Beijing Proteome Research Center, Beijing, China Background: Beta-diversity is a fundamental ecological metric for exploring dissimilarities between microbial communities. On the functional dimension, metaproteomics data can be used to quantify beta-diversity to understand how microbial community functional profiles vary under different environmental conditions. Conventional approaches to metaproteomic functional beta diversity often treat protein functions as independent features, ignoring the evolutionary relationships among microbial taxa from which different proteins originate. A more informative functional distance metric that incorporates evolutionary relatedness is needed to better understand microbiome functional dissimilarities. Results: Here, we introduce PhyloFunc, a novel functional beta-diversity metric that incorporates microbiome phylogeny to inform on metaproteomic functional distance. Leveraging the phylogenetic framework of weighted UniFrac distance, PhyloFunc innovatively utilizes branch lengths to weigh between-sample functional distances for each taxon, rather than differences in taxonomic abundance as in weighted UniFrac. Proof-of-concept using a simulated toy dataset and a real dataset from mouse inoculated with a synthetic gut microbiome and fed different diets show that PhyloFunc successfully captured functional compensatory effects between phylogenetically related taxa. We further tested a third dataset of complex human gut microbiomes treated with five different drugs to compare PhyloFunc"s performance with other traditional distance methods. PCoA and machine learning-based classification algorithms revealed higher sensitivity of PhyloFunc in microbiome responses to paracetamol. Conclusions: Unlike traditional approaches that consider metaproteomics features as independent and unrelated, PhyloFunc acknowledges the role of phylogenetic context in shaping the functional landscape in metaproteomes. In particular, we report that PhyloFunc accounts for the functional compensatory effect of taxonomically related species. It is effective, ecologically significant, and has better sensitivity, as evidenced by the particular applications we presented. 222

224

P-I-0077 Peering into the microbial world: exploring proteomic diversity across the bacterial domain of life for enhanced bacterial identification Christina Ludwig1, Miriam Abele1,2, Armin Soleymaniniya3, Florian P. Bayer2, Nina Lomp1, Etienne Doll4, Chen Meng1, Klaus Neuhaus5, Siegfried Scherer4, Mareike Wenning6, Nina Wantia7, Bernhard Küster1,2, Mathias Wilhelm3 1Technical University of Munich, Bavarian Center for Biomolecular Mass Spectrometry (BayBioMS), Freising, Germany 2Technical University of Munich, Chair of Proteomics and Bioanalytics, Freising, Germany 3Technical University of Munich, Computational Mass Spectrometry, Freising, Germany 4Technical University of Munich, Research Department Molecular Life Sciences, Freising, Germany 5Technical University of Munich, Core Facility Microbiome / ZIEL Institute for Food & Health, Freising, Germany 6Bavarian Health and Food Safety Authority, Oberschleissheim, Germany 7Technical University of Munich, Institut für Medizinische Mikrobiologie, Immunologie und Hygiene, Munich, Germany Despite the vast diversity of the prokaryotic domain of life, relatively few bacterial proteomes have been comprehensively characterized with mass spectrometry so far. Here, we present the largest proteomic resource for bacteria to date, encompassing 305 species, 119 genera, and five phyla. This resource provides mass spectrometric evidence for over 635,000 expressed bacterial proteins and confirms the existence of over 35,000 hypothetical proteins. The complete dataset resource ProteomicsDB (https://www.proteomicsdb.org/), enabling quantitative exploration of all detected proteins within a species as well as across different species. To facilitate cross-species analyses, we have grouped bacterial proteins into orthogroups using the tool OrthoFinder, which identifies homology relationships between proteins based on their sequence similarities. is available in the public This comprehensive dataset facilitated the development of a new bacterial identification algorithm called PyTaxon. PyTaxon employs the rapid data analysis tool MSFragger to query in a first iteration the complete species-level proteome space deposited in NCBI (n = 13,855 species). Subsequently, a second iteration analysis can be conducted to refine the results to strain-level by focusing on the identified taxonomical region from the first iteration. Using the dataset of the 305 bacterial species as a benchmark, we demonstrate that PyTaxon achieves identification accuracies of >99% on species-level. When applied to a dataset comprising 94 strains of the genus Pseudomonas spp. and 29 strains of the Bacillus cereus s.l. group, PyTaxon attains unprecedented identification accuracies of >89% on strain-level. Finally, we evaluated PyTaxon's performance as a tool for food safety and clinical diagnostics. When applied to 55 pure bacterial isolates from dairy products, PyTaxon provided more accurate results than MALDI-TOF or FTIR, with identifications that were 94% consistent with 16S rRNA gene sequencing. In 570 clinically derived bacterial isolates, PyTaxon successfully identified 64 different bacterial species from 20 genera and 3 phyla. Additionally, we explored the acquired data for the expression of proteins responsible for antibiotic resistance. Beta- lactamases, including also extended-spectrum beta-lactamases (ESBL), were detected in every fourth clinical sample. This alarming finding underscores the urgency for rapid antimicrobial resistance screenings to enable informed antibiotic treatment decisions and ideally mitigate further acquisition of (multiple-) antibiotic resistance. In conclusion, our work establishes a comprehensive proteomic resource detailing protein expression profiles across the bacterial domain of life. This resource facilitated the development of an advanced bacterial identification algorithm and provided insights into the prevalence of antimicrobial resistance markers in clinical samples. 224

226

P-I-0080 Effect of selenium (Se(IV)) on the protein profile and ethanol production in Saccharomyces cerevisiae cultures – a proteomic study using 15N metabolic labeling Luis Mejia1, Katarzyna Wrobel1, Kazimierz Wrobel1, Alma Rosa Corrales1 1University of Guanajuato, Department of Chemistry, Guanajuato, Mexico The species-dependent benefic effects of selenium have been demonstrated in a variety of organisms, and its role in human health and disease has been studied extensively. S. cerevisiae, biofortified with Se is the most common and effective form of this element [1, 2]. Although this microorganism is not strictly accepted as a probiotic, it is used to make functional foods, specifically to produce Se-enriched dairy products. To understand the mechanisms for the health-promoting effects of selenized yeast and for those involved in the production of selenized food, it is necessary to investigate the impact of Se on the growth and metabolism of yeasts. In this regard, the selenium pathway is well documented [3] and the metabolic response under selenium exposure has been studied [4-6]. However, less effort has been devoted to unraveling the effects of Se at the proteome level. One promising strategy in comparative proteomics relies on the metabolic labeling of target microorganisms with heavy nitrogen during their growth. Thanks to the isotopic capability of mass spectrometry, samples from two different treatments and labeled differentially can be combined and analyzed in a single run, thereby increasing the reliability, and facilitating comparative assessment. In practice, inverse metabolic labeling is often performed by preparing two sets of biological samples; in one of them 15N is applied to the control culture whereas in another set, 15N is introduced into the exposed organisms. In this way, the interpretation of mass spectrometry data becomes less error-prone [7-10]. This work aimed to perform 15N - metabolic labeling of baker´s yeasts to ensure a high-fidelity assessment of Se(IV) impact at the proteome level. For this purpose, nitrogen-containing salt was added to the growth medium as a sole nitrogen source, using 15N salt for yeast exposed to Se and a salt with natural 14N isotope composition for non-exposed controls (and inversely). Protein extracts from differentially labeled Se-exposed and non- exposed control cultures, were combined, and prepared for peptide analysis by capillary liquid chromatography- electrospray ionization - mass spectrometry with quadrupole - time of flight mass analyzer (HPLC-ESI-QTOF-MS). The incorporation of selenium into three proteins participating in glycolytic processes and the increased abundance of these proteins in the presence of Se were found. The results obtained suggested that selenized yeasts may have better fermentation capacity compared to non-exposed organisms. The increase in ethanol production by selenized yeasts was confirmed by analyzing the respective spent media using gas chromatography with flame ionization detection. 226

227

P-I-0081 Metaproteome analysis of exhaled breath particles Susann Meyer1, Vivien Pietsch1, Nicole Hüttig1, Dierk-Christoph Pöther1 1Federal Institute for Occupational Safety and Health, Bioaerosols, Berlin, Germany Indoor air hygiene is a crucial aspect of human health and well-being. Previously, the concentration of carbon dioxide (CO2) is used as a benchmark for assessing the hygienic quality of indoor air and ventilation strategies were based on this value. Nevertheless, the recent pandemic has once again demonstrated the significant impact that individuals can have on indoor air quality through the emission of bioaerosol particles. To date, the particles exhaled by humans that could be emitted into indoor air have not been extensively studied in the context of indoor air hygiene and occupational health and infection control. The objective of this study therfore is to analyze the particles exhaled by healthy workers and to evaluate the influence of human bioaerosol particles on indoor air quality. In order to achieve this, human volunteers were instructed to generate respiratory particles from the lower airways using a specific breathing technique and then to breathe them into the exhaled particle collection device, designated as PExA (particles in exhaled air). Subsequently, the particles with a diameter below 7 µm were collected by impaction on filters. Following protein extraction and digestion, the peptides were analyzed using the Thermo Astral mass spectrometer with a Data Independent Acquisition (DIA) method. The data were subjected to metaproteomic analysis using an in-house protein database comprising predicted microbial and viral representatives of the respiratory tract and the human proteome. The results of this study indicate the presence of a specific profile of human proteins that come from the lower respiratory tract into the exhaled human breath. These proteins are derived from the human immune system, particularly from neutrophil degranulation and haemostasis, as well as from the area of protein metabolism, including surfactant metabolism. Furthermore, a few proteins of the microbiome of the respiratory tract were identified in the exhaled air of the human test subjects. These belong, for example, to bacterial representatives that can be assigned to the phylum Pseudomonadota, as well as fungal representatives that belong to the phylum Ascomycota. The results demonstrate that the exhaled air of healthy test subjects releases bioaerosol particles that are predominantly composed of human proteins, with a portion originating from microbial representatives of the respiratory tract. Further studies will concentrate on the quantification of specific proteins in exhaled air, as well as on the question of whether these proteins can be quantified in indoor air. The data will be employed as a foundation for an examination of the correlation between CO2 concentration and the concentration of bioaerosol particles emitted. This will facilitate a reassessment of the recommendations for indoor ventilation, which were previously based on CO2 concentrations. The objective is to enhance indoor air quality in the context of occupational health and infection control. 227

228

P-I-0082 Deep phosphoproteomics of Klebsiella pneumoniae reveals HipA-mediated tolerance to ciprofloxacin Payal Nashier1, Isabell Samp2, Marvin Adler2, Carsten Jers3, Ivan Mijakovic3,4, Sandra Schwarz2, Boris Macek1 1University of Tübingen, Proteome Center Tuebingen, Tuebingen, Germany 2University of Tübingen, Institute of Medical Microbiology and Hygiene, Tuebingen, Germany 3Technical University of Denmark, The Novo Nordisk Foundation, Center for Biosustainability, Kongens Lyngby, Denmark 4Chalmers University of Technology, Systems and Synthetic Biology Division, Gothenburg, Sweden Klebsiella pneumoniae belongs to the group of bacterial pathogens causing the majority of antibiotic-resistant nosocomial infections worldwide; however, the molecular mechanisms underlying post-translational regulation of its physiology are poorly understood. Antibiotic tolerance plays a significant role in shaping the evolutionary dynamics of bacterial populations subjected to repeated antibiotic treatments. A subset of the population developing tolerance to high concentrations of antibiotics and characterized by slow growth are called persisters. Exposure to various bactericidal antibiotics commonly employed in the treatment of K. pneumoniae infections has revealed the presence of multidrug-tolerant persister cells. Persister cells have also been identified in clinical isolates from individuals experiencing recurring bloodstream infections, demonstrating genomic alterations in relapsed isolates that evolved within the host. However, the mechanism of formation of these multidrug-tolerant persister cells in K. pneumoniae is understudied. One of the best-studied drivers of persistence is the eukaryotic-type Ser/Thr kinase, HipA in E. coli and a recent study has identified homologs of the hipA in many bacterial species. We used a comprehensive analysis of Klebsiella phosphoproteome, focusing on HipA, a Ser/Thr kinase involved in antibiotic tolerance in Escherichia coli. We showed that overproduced K. pneumoniae HipA (HipAkp) is toxic to both E. coli and K. pneumoniae and its toxicity can be rescued by overproduction of the antitoxin HipBkp. Importantly, HipAkp overproduction leads to increased tolerance against ciprofloxacin, a commonly used antibiotic in the treatment of K. pneumoniae infections. Proteome and phosphoproteome analyses in the absence and presence of ciprofloxacin confirmed that HipAkp has Ser/Thr kinase activity, auto-phosphorylates at S150, and shares multiple substrates with HipAec including GltX (glutamate tRNA synthetase) at S239. In E. coli, this phosphorylation leads to the accumulation of uncharged glutamate tRNA and activates the stringent response in cells, ultimately leading to antibiotic tolerance and persistence. Our study provides a valuable resource to clarify the molecular basis of tolerance and the role of Ser/Thr phosphorylation in this human pathogen. 228

229

P-I-0083 Transcriptional and post-translational regulation by the Ser/Thr protein kinase HipH Niels Neumann1, Payal Nashier1, Boris Macek1 1University of Tübingen, Quantitative Proteomics, Tuebingen, Germany Dynamic protein phosphorylation is one of the cornerstones in bacterial signal transduction. In recent years the eukaryotic type Ser/Thr kinase HipA received increasing attention due to its role in conferring antibiotic tolerance. In Escherichia coli HipA is encoded together with HipB on the hipBA operon, forming a toxin antitoxin (TA) module. If not neutralized by its antitoxin HipB, HipA phosphorylates different substrate pools in E. coli, promoting growth stop and inducing persistence. Phylogenetic analysis revealed that homologues of hipBA like TA systems are widespread among various bacterial species, including clinically relevant strains [1]. The roles of these different TA modules are still debated and include genetic chromosome stabilization, phage defense and antibiotic tolerance. In a recent study we investigated the monocistronic hipA homologue HipH (yjjJ), demonstrating its toxic effect on cell division and identified several novel phosphorylation targets [2]. Intriguingly, HipH carries a putative DNA binding helix-turn-helix (HTH) domain at its N-terminus which is usually found in HipB-like antitoxins to regulate expression of its own TA module. In this work we investigate the role of the HTH domain of HipH regarding its ability to bind DNA/RNA and its potential role in transcriptional regulation. We show that the HTH-domain of HipH, in contrast to HipB, is not involved in regulating its own expression but is instead essential for the toxicity of HipH. Thereby, we show via shotgun proteomics that the HTH-domain regulates a variety of different genes, including potential virulence factors. We therefore postulate, that this domain provides a regulatory function on the transcriptional level for HipH in addition to the post-translational regulation by its kinase activity. 1. Gerdes K, Bærentsen R, Brodersen DE. Phylogeny Reveals Novel HipA-Homologous Kinase Families and Toxin-Antitoxin Gene Organizations. mBio. 2021 Jun 29;12(3):e0105821. 2. Gratani FL, Englert T, Nashier P, Sass P, Czech L, Neumann N, Doello S, Mann P, Blobelt R, Alberti S, Forchhammer K, Bange G, Höfer K, Macek B. E. coli Toxin YjjJ (HipH) Is a Ser/Thr Protein Kinase That Impacts Cell Division, Carbon Metabolism, and Ribosome Assembly. mSystems. 2023 229

230

P-I-0084 Mapping the functional proteome of gut microbiome specie Dimitris Papagiannidis1, Martin Garrido-Rodriguez1,2, Frank Stein1, Isabelle Becher1, Nassos Typas1, Mikhail Savitski1 1EMBL – EBI, Heidelberg, Germany 2Heidelberg University, Institute for Computational Biomedicine, Heidelberg, Germany The bacterial proteome is highly diverse and quickly adapts to physiological stimuli, available nutrients and competition. The gut microbiome is a complex integral part of human physiology and has long been known to play an essential role in digestion and host metabolism. Increasing evidence further highlights its impact on human behavior and it has been associated with a variety of diseases, ranging from inflammatory bowel disease to cancer. Despite the intricate interplay between the microbiome and its host, mechanistic insights into their interactions are scarce and to a large extent based either on metagenomic data or animal studies. Although indispensable for the characterization and analysis of the microbiome on a systems level, these approaches provide only indirect mechanistic information on the subcellular level. Protein annotation in these species is incomplete with information about protein existence, let alone function, being mainly based on predictions. A systematic understanding of protein functions in gut bacteria will not only provide insights into their metabolism, but also contribute to better understanding of the formation, interplay and molecular function of the microbiome in relation to the host. Here, we develop a multiplexed Thermal Proteome Profiling-based approach to map the functional proteome of the gut microbiome species P.vulgatus and B.uniformis. Our approach involves combinatorial treatment of lysates with pools of compounds, allowing us to explore a large number of potential protein-ligand pairs. We have probed the binding partners of 56 compounds, ranging from nutrients to host hormones and drugs and validated our approach using highly conserved processes. Our data provide insights into previously uncharacterized aspects of gut bacterial biology such as sugar sensing, antibiotic transport and response to host hormones. 230

231

P-I-0085 Model of protein translation and quantitative assay identify dozens of novel slippery sites causing ribosomal frameshifting Jonas Poehls1,2, Cedric Landerer1,2, Doris Richter1,2, Anna Shevchenko1, Andrej Shevchenko1, Agnes Toth- Petroczy1,2 1Max-Planck Institute of Molecular Cell Biology and Genetics, Dresden, Germany 2Center for Systems Biology Dresden, Dresden, Germany Cellular life critically depends on accurate translation of mRNAs into proteins. However, translation is an imperfect process, and the ribosome is known to make a variety of errors, termed phenotypic mutations. Phenotypic mutations include single amino acid substitutions, Stop codon readthrough, or ribosomal frameshifting, where the ribosome switches from one reading frame to another during translation. Ribosomal frameshifting in particular is implicated in human health, as many viruses, including HIV and Sars-CoV2, use it and depend on it. In addition, recent evidence suggests that frameshifting occurs during translation of the Sars- CoV2 mRNA vaccines. Mechanistically, frameshifting is generally dependent on signals in the mRNA sequence, so-called "slippery sites". While the biochemical mechanisms of frameshifting have been studied in detail, few attempts have been made to identify frameshifting sites independent of known cases and at high throughput. We developed a mechanistic model of translation errors purely based on the mRNA sequence and the cell"s tRNA pool that predicts frameshifting sites. Using a reporter protein construct that allows frameshift quantification by fluorescence and mass spectrometry, we tested the model"s predictions in vitro. Our assay identified dozens of 10-mers that cause frameshifting by more than 10%, both in the -1 and +1 direction. Considering that all of the tested sequences occur in real genomes, this indicates that frameshifting outside of viruses is more common than previously assumed. In addition, our model predicts -1 and +1 frameshifting with an average precision of >80%, both on the experimental library and a larger dataset of known frameshift motifs and non-frameshifting sequences. Using the experimental data and the model"s predictions, we are now scanning genomes for frameshifting sites. This will allow us to assess the baseline levels of frameshifting and thus its impact on an organism"s fitness. In addition, we are analyzing the putative protein products of these frameshift sites in terms of their conservation, structure and family membership to identify novel functional cases of frameshifting. Fig. 1 (optional) 231

232

P-I-0086 Proteome-wide absolute quantification using a generic internal standard Viditha Rao1, Ignacy Rzagalinski1, Isaac Gabriel Justice2, Ha Ngoc Anh Nguyen2, Bharath Kumar Raghuraman3, HongKee Moon1, Andre Gohr1, James P. Sáenz2, Andrej Shevchenko1 1Max Planck Institute of Molecular Cell Biology and Genetics, Dresden, Germany 2B Cube-Centre for Molecular Bioengineering, Dresden, Germany 3Anavo Therapeutics BV, Heidelberg, Germany Absolute (molar) quantification of proteins helps us to determine stoichiometric ratios within metabolic pathways. Neither genome nor transcriptome sheds light on how metabolic networks are organized: genomics tells what could be the components; transcriptomics shows which genes are expressed, but neither of the two translate into the actual protein concentrations. Here, we aimed at demonstrating that proteome-wide absolute quantification supports the direct comparison of proteome composition of phylogenetically distant minimal organisms. These minimal organisms help us to understand the fundamental principles of life and the basic requirements for cellular function, including metabolism. We aim to determine and compare the quantities of functionally related proteins in the whole cells of minimal JCVI-syn3A (2) bacteria and near minimal bacterium Mesoplasma Florium (3). We underscore that protein levels comparison could not be achieved by considering their genomes or (if available) transcriptomes. We have developed a workflow for proteome –wide absolute quantification by DDA with a family of FUGIS (Fully Unlabelled Generic Internal Standard) (1) proteins as the internal standard. The FUGIS chimeric protein is co- digested with a known amount of BSA (reference protein) along with protein test samples. The digest is analyzed using LC-MS/MS workflow and the raw data files are processed with MSFragger (Fig 1A). The .tsv output from the MSFragger with peptide ID"s and their respective raw XIC peak areas is used by in-house built GlobeQuant(1) software for quantification. The software calculates the molar amount of the FUGIS peptides and relates it to the median area of XIC peaks of FUGIS peptides peaks and further uses it as a single-point calibrant to determine the molar abundance of any codigested protein. The peptides of the codigested sample proteins are selected by the principle of BestNAverage, where the "BestN" is selected from the Top peptides based on a criteria of CV% (Fig 1B). These minimal cell organisms allow us to perform absolute quantification in an untargeted way to quantify the whole proteome across organisms hence quantifying and comparing the homologous proteins at the organism level. References J. Proteome Res.2022, 21, 1, 132–141. doi:10.1021/acs.jproteome.1c00596 1. 2. Pelletier, J.F. et al. (2021) "Genetic requirements for cell division in a genomically minimal cell", Cell, 184(9). doi:10.1016/j.cell.2021.03.008. 3. Matteau, D. et al. (2020) "Integrative characterization of the near‐minimal bacterium "mesoplasma florum", Molecular Systems Biology, 16(12). doi:10.15252/msb.20209844. Fig. 1 (optional) 232

233

P-I-0087 A novel cell biological tool to explain mechanics and dynamics in fission yeast Muhammad Ali Rasheed1 1University of karachi, Proteomics, Karachi, Pakistan The Rho guanosine triphosphatase hydrolase enzyme (GTPase) is required for the control of the actin cytoskeleton, but its activation in vivo condition is unknown. The study's goal was to find a new synthetic nanobody VHH (P-36 tagged with mNeonGreen) that interacts strongly with the Rho GTPase. We present the first novel synthetic nanobody, VHH (P-36 tagged with mNeonGreen), tested in fission yeast cells and found to have a particular interaction with Rho1GTPase. Plasmids were constructed by using of certain enzymes to digest the pDUAL-pef1a vector plasmid to produce a protein that was encoded by cloned genes. A varied VHH library was created synthetically, then transformed into yeast cells, and positive clones were chosen using chemical agents. To investigate protein interactions and cellular reactions, several studies were carried out, such as live cell imaging, growth curve analysis, coimmunoprecipitation, structural analysis, and cell therapies. Prism and RStudio were used for the statistical analysis. The presence of VHH (P-36) has no effect on the growth pattern making it an appropriate model for studying cytokinesis in vivo. According to a computational biological study, its affinity to interact with Rho1GTPase with all the complementarity-determining region (CDR) regions found on VHH (P-36) is extremely strong. We were able to track its subcellular target by localization using a fluorescent confocal microscope, ensuring the maintenance of cell polarity and morphology. Spheroplast analysis revealed a circular- shaped cell with an even distribution of Rho1 tagged VHH (P-36), indicating that the interaction occurs near the plasma membrane. The introduction of latrunculin-A (Lat-A) disrupted Rho GTPase localization, demonstrating the control over actin production, and the cell did not show evidence of mitotic phase commencement while Lat-A was present. Finally, this important biological tool can aid in our understanding of the mechanics and dynamics of cytokinesis in relation to Rho1GTPase. Fig. 1 (optional) 233

234

P-I-0088 A new proteomic approach to detect and characterise microorganisms in peritoneal mucinous carcinomatosis Antonio Romero-Ruiz1,2, Rafaella Pezzopane1, María C Vázquez-Borrego2, Ana Martinez-Lopez3, Florina Iulia2, Melissa Granados-Rodríguez2, Ana Moreno-Serrano2, Maria Torres-Martinez2, Lidia Rodriguez-Ortiz4, Francisca Valenzuela Molina4, Blanca Rufian-Andujar4, José Alhama2, Carmén Michán2, Alvaro Arjona-Sanchez2,4 1University of Córdoba, Biochemistry and Molecular Biology, Córdoba, Spain 2Maimonides Biomedical Research Institute of Cordoba (IMIBIC), GE09, Córdoba, Spain 3Reina Sofia University Hospital of Cordoba, Pathology Unit, Córdoba, Spain 4Reina Sofia University Hospital of Cordoba, Surgery Unit, Córdoba, Spain Pseudomyxoma peritonei (PMP) is a rare malignancy and the paradigm of Peritoneal Mucinous Carcinomatosis, which causes the continuous accumulation of mucus-secreting cells and glycoprotein-rich mucus in the abdomen. The only available treatment involves complete cytoreductive surgery combined with hyperthermic intraperitoneal chemotherapy. Despite this aggressive management, relapses are common in PMP, leading to debilitating symptoms and a fatal end. It's therefore important to look for ways to improve treatment and better understand the pathogenesis of the disease. In this context, microorganisms have previously been detected in PMP by genomic approaches and antibacterial treatment has been suggested, but data are limited. Our group aimed to develop the first method based on a "proteomic approach" to test and characterise the presence of microorganisms in mucinous tumour tissues. Samples obtained from patients during CRS-HIPEC surgery were treated with our pioneering protein isolation protocol and then analysed by nanoHPLC-MS/MS. To identify and characterise the presence of microorganisms in the tumour tissue, a new human and microbial protein database was created. Our results demonstrate for the first time the presence of fungal proteins in mucin samples from PMP tumours, both low and high grade (Figure 1), and was validated by histological staining of tumoral tissue at our pathology service. This work presents a novel proteomic and bioinformatic approach to identify the presence of microbiota. This is in line with the literature where different genera of bacteria and fungi have been identified by genomic approaches in different types of cancer. However, further research is needed to determine the role of this advance in the pathogenesis of mucinous tumours, to understand the mechanisms of oncogenesis and to develop better therapeutic strategies for these patients. Funding sources: This work is supported by Instituto de Salud Carlos III (ISCIII), "PI22/01213"; co-funded by the European Union, and by Asociación Española contra el Cáncer, "PRYES223170ARJO". Fig. 1 (optional) 234

235

P-I-0089 In-depth metaproteomics analysis of tongue coating for gastric cancer: a multicentre diagnostic research study Jiahui Chen1, Yingying Sun1, Jie Li2, Mengge Lyu3, Li Yuan4, Jiacheng Sun5, Shangqi Chen6, Can Hu1, Qing Wei4, Zhiyuan Xu1, Tiannan Guo3, Xiangdong Cheng1 1Zhejiang Cancer Hospital, Hangzhou, China 2Tsinghua University, Beijing, China 3Westlake University, Hangzhou, China 4Chinese Academy of Sciences, Hangzhou, China 5The Affiliated Hospital of Wenzhou Medical University, Wenzhou, China 6University of Chinese Academy of Sciences, Ningbo, China Our previous study identified significant differences in tongue images between individuals with and without gastric cancer, but the underlying biological mechanisms remain unclear. In this study, we employed pressure cycling technology and data-independent acquisition (PCT-DIA) mass spectrometry to acquire tongue-coating metaproteomics data from 180 gastric cancer patients and 185 controls across five centers in China. We characterized 1,432 human proteins and 13,780 microbial proteins in these tongue-coating samples and demonstrated the temporal stability of protein expression levels using a time-series cohort. We also compared the functional and taxonomic characteristics of the tongue-coating proteomes between individuals with and without gastric cancer, identifying notable alterations in gastric cancer patients. Notably, keratins KRT2 and KRT9, and the microbial ABC transporter COG1136 were downregulated in gastric cancer patients, indicating reduced lingual mucosa defense. Importantly, we developed a machine learning model using 50 microbial proteins to identify high-risk individuals, achieving an AUC of 0.91 in the independent validation cohort. This study highlights the potential of tongue-coating proteins as indicators for non-invasive gastric cancer screening. Fig. 1 (optional) 235

236

P-I-0090 An indispensable protein catalyzing nitrite reduction in "Candidatus Kuenenia stuttgartiensis" strain CSTR1: activity and physiological properties Emea Okorafor Ude1,2, Pranathi Sure2, Chang Ding2, Lorenz Adrian2 1UFZ, Leipzig, Germany 2Helmholtz Centre for Environmental Research (UFZ), Department of Molecular Environmental Biotechnology, Leipzig, Germany The physiological enzyme that reduce nitrite in anaerobic ammonium oxidizing (anammox) bacteria belonging to "Candidatus Kuenenia stuttgartiensis" is a mystery. The ongoing curiosity in identifying this enzyme has led to the proposal of myriads of proteins as potential candidates. So far, most of the studies on nitrite reductase in "Ca. Kuenenia" were centred only on purified proteins which mostly resulted in inconclusive evidence. However, crude cell extract can be a platform for identification of major protein candidates and their physiological properties. Here, we performed a comprehensive screening of nitrite-reducing activity in the proteome of "Ca. Kuenenia stuttgartiensis" strain CSTR1 which was dominant in a continuous stirred tank reactor (CSTR) containing planktonic cells. We hypothesized that a fast and reliable methods capable of separating the proteins in crude cell extract would enable a broader evaluation of the protein activities. In this context, fast protein liquid chromatography (FPLC) with anion exchange chromatography (AEC) or size exclusion chromatography (SEC) were adopted sequel to proteomics analysis. Separation of the cell extract of strain CSTR1 with SEC yielded active fractions with a molecular size range of 150-200 kDa. Conversely, using AEC resulted in significant loss of activity compared to SEC separation, and activities were majorly distributed in 3 complexes. Combinatorial experiment from AEC fractions show that the activity can be recovered when certain AEC fractions are combined but removal of a certain component resulted to a huge loss of activity up to 74%. This component that accounted for the majority of the nitrite reducing activity was denoted as an indispensable protein fraction. A hierarchical cluster analysis of the mass spectrometric identifications in the indispensable protein fraction revealed that a heme protein and electron transfer protein were linked to nitrite reducing activity. The evaluation of the physiological characteristics of the indispensable protein component revealed its ability to catalyse nitrite reduction in the presence of protein components such as AEC fractions, SEC fractions, reduced and oxidized haemoglobin, and bovine serum albumin (BSA). In contrast, these proteins without the indispensable protein component, as well as the indispensable protein component alone, exhibited no activity or significant loss of activity. Therefore, the discovery of the indispensable protein component in "Ca. Kuenenia stuttgartiensis" is an intriguing finding. These results suggest that protein interaction could be an important requirement for nitrite reduction in "Ca. Kuenenia stuttgartiensis". 236

237

P-I-0091 The role of polybetahydroxybutyrate (PHB) accumulation in the stoichiometry of the L-acetylation and the metabolism of Rhizobium etli Jocelin Rizo1, Carmen Vargas-Lagunas1, Jeovanis Gil Valdés2, Ramiro Alonso-Bastida1, Angel Gabriel Martínez- Batallar1, Magdalena Hernández Ortíz1, Sandra Contreras-Martínez1, Alma Ruth Reyes González1, Noé Arroyo1, Sergio Manuel Encarnación-Guevara1 1Universidad Nacional Autónoma de México, Centro de Ciencias Genómicas, Mexico City, Mexico 2Lund University, Division of Oncology, Department of Clinical Sciences Lund, Lund, Sweden Rhizobium is a type of bacteria found in the rhizosphere that can fix atmospheric nitrogen in association with leguminous plants. Different species have been identified in this genus capable of accumulating poly-beta- hydroxybutyrate (PHB), which is stored in symbiosis and free life. The process of PHB accumulation typically starts with the condensation of two molecules of acetyl-CoA to form acetoacetyl-CoA. Subsequent reduction and polymerization reactions lead to the production of PHB. This compound serves as a reserve material for carbon and/or electron storage when the bacteria face suboptimal growth conditions. However, the exact role of PHB in the bacteria's metabolism is not fully understood. Lysine acetylation is a protein modification that plays a crucial role in various cellular processes, such as central metabolism, transcriptional regulation, chemotaxis, and pathogen virulence. It can alter protein functions by affecting DNA binding, enzymatic activity, protein interactions, stability, and localization. In prokaryotes like Rhizobium etli, lysine acetyltransferases. lysine acetylation occurs both non-enzymatically and through the action of In R. etli, a mutant lacking PHB-synthase (phaC-), and unable to accumulate PHB, shows decreased growth when cultured in a minimal medium with glucose or pyruvate as the sole carbon source. This mutant also exhibits a high excretion of organic acids and an excess of accumulated reducing power. This suggests that PHB synthesis is necessary for R. etli to utilize glycolytic carbon sources for growth. In the absence of PHB synthesis, the accumulation of acetyl-CoA can lead to the chemical acetylation of proteins, which regulates their activity. This proteomic study using LC-MS/MS is being conducted with R. etli strains of varying PHB accumulation capacities, including the PHB synthase gene mutant, to investigate the regulation of non-enzymatic protein acetylation by PHB synthesis in R. etli. The study aims to determine the target proteins, the stoichiometry of post-translational modifications, and their implications in cellular metabolism under free-living and symbiotic conditions with the legume Phaseolus vulgaris (beans) in the presence or absence of PHB synthesis. By integrating lysine acetylation stoichiometric analysis with quantitative proteomics, the study has identified and quantified nearly 2000 proteins. Our initial analysis highlighted proteins that participated in carbon metabolism. The results will be shown and discussed. The research aims to reveal how metabolic processes are affected by the stoichiometry of acetylation under different conditions, providing insights into the regulation of PHB accumulation and its impact on protein acetylation. 237

238

P-I-0092 Unipept next: expanding the scope beyond tryptic peptides for advanced microbiome analysis Pieter Verschaffelt1,2,3, Tibo Vande Moortele2, Bart Mesuere2, Peter Dawyndt2, Bram Devlaminck2 1Ghent University, Center for Medical Biotechnology (CMB), Ghent, Belgium 2Ghent University, Department of Applied Mathematics, Statistics and Computer Science, Ghent, Belgium 3Ghent University, Center for Medical Biotechnology, Department of Biomolecular Medicine, Ghent, Belgium (https://doi.org/10.1021/acs.jproteome.8b00716), a pioneering Unipept in metaproteomics, has significantly advanced the analysis of complex ecosystems by facilitating both taxonomic and functional insights from environmental samples. Initially, Unipept's capabilities focused on tryptic peptides, utilizing the predictability and consistency of trypsin digestion to efficiently construct a protein reference database. However, the evolving landscape of proteomics and emerging fields like immunopeptidomics necessitate a more versatile approach that extends beyond the analysis of tryptic peptides. tool In metaproteomics, samples are typically prepared using trypsin which cleaves proteins at specific cleavage sites (https://doi.org/10.1038/s41467-021-27542-8). Occasionally, trypsin skips a cleavage site, resulting in fragments composed of multiple tryptic peptides, referred to as peptides with missed cleavages. Semi-tryptic peptides, produced when one terminus is generated by trypsin cleavage and the other is not, were previously undetectable by Unipept. Accurately identifying semi-tryptic peptides is crucial for understanding the complex proteolytic activities of the gut microbiome (https://doi.org/10.1186/s40168-020-00967-x). Immunopeptidomics (https://doi.org/10.1016/j.smim.2023.101727), an upcoming research discipline, focuses on characterizing HLA-peptides presented by MHC molecules, which are vital to immune responses. These peptides are not restricted to tryptic forms and exhibit a diverse range of lengths and cleavage patterns. The current limitation of Unipept to match only tryptic peptides constrains its utility in the use-cases described above. Over the past year, we have integrated a range of sophisticated index structures from the field of metagenomics, such as Enhanced Suffix Arrays (ESA), FM-indices, and the R-index, to enhance Unipept's capability in identifying substrings in large text bodies efficiently. Unipept Next is the next-generation of Unipept, capable of analyzing peptides regardless of how their originating proteins were cleaved. Utilizing an Enhanced Suffix Array, Unipept Next can identify the proteins from which a peptide originates in milliseconds, enabling rapid taxonomic and functional analysis of protein samples. We validated the performance and correctness of Unipept Next using six datasets from the SIHUMI study, demonstrating its massively enhanced performance and versatility (see Figure 1). Fig. 1 (optional) 238

239

P-I-0093 Fecal metaproteomics enables functional characterization of remission in patients with inflammatory bowel disease Maximilian Wolf1,2, Julian Lange3, Patrick Hellwig3,4, Kay Schallert5, Dirk Benndorf6, Udo Reichl3,4, Matthias Gunzer5,7, Konrad Aden8, Robert Heyer1,5 1Bielefeld University, Faculty of Technology, Bielefeld, Germany 2Bielefeld Institute for Bioinformatics Infrastructure (BIBI), Graduate School DILS, Bielefeld, Germany 3Otto-von-Guericke University Magdeburg, Bioprocess Engineering, Magdeburg, Germany 4Max Planck Institute for Dynamics of Complex Technical Systems, Magdeburg, Germany 5Leibniz-Institut für Analytische Wissenschaften – ISAS – e.V., Dortmund, Germany 6Anhalt University of Applied Science, Applied Biosciences and Process Engineering, Koethen, Germany 7University Hospital Essen, Institute for Experimental Immunology and Imaging and Imaging Center Essen, Essen, Germany 8Christian-Albrechts-University Kiel, Institute of Clinical Molecular Biology, Kiel, Germany Inflammatory Bowel Disease (IBD) describes a condition characterized by chronic inflammation of the gastrointestinal tract. Its pathogenesis is linked to genetic susceptibility, environmental factors, host immune response, and gut microbiome. There are various therapy approaches, including treatment with biological antibodies. Alleviation of symptoms, which is termed remission, is often assessed by colonoscopy, which means an invasive measure for the patient. Therefore current research is focused on characterizing remission and finding microbial or human biomarkers for therapy success by non-invasive methods, like analysis of fecal samples by metaproteomics. The goal of this work was the functional characterization of differences in the fecal metaproteome of patients who did or did not achieve remission. Therefore, fecal samples of a cohort of IBD patients were collected before and after 14 weeks of treatment with three different biologics. Clinical disease activity scores were used to determine clinical response and remission. The fecal metaproteomes of the remitting patients (n=12) and of non-remitting patients (n=12) were compared at baseline and changes within both groups were assessed over the course of the therapy to identify functional changes and potential human and microbial biomarkers. The proteins of these samples were extracted via bead beating with phenol and tryptically digested. The peptides were separated using reversed-phase liquid chromatography and measured with a timsTOF mass spectrometer. The MetaProteomeAnalyzer was used for protein identification and taxonomical and functional annotation. The results show, that the therapy"s success is clearly depicted in the fecal metaproteome of IBD patients. The abundance of proteins associated with the intestinal barrier, neutrophilic granulocytes, and immunoglobulins significantly decreased in remitting patients, and showed, in contrast, an increase in non-remitting patients. Furthermore, the remission was characterized by an increased abundance of microbial proteins involved in the pentose phosphate pathway, tricarboxylic acid cycle, or butyrate fermentation. These results imply diminished immune system activity and a restored microbial metabolism in remitting patients. Finally, we proposed new potential biomarkers for monitoring of therapy success, e.g. human lysosome-associated membrane glycoprotein 1, a cytotoxicity marker, or microbial anthranilate synthase component 2, a part of the tryptophan metabolism, both increasing in non-remitting patients. Our study suggests, that there are distinct changes in gut inflammation and gut microbiome metabolism dependent on whether therapy with biologics induced remission or not, proving that metaproteomics could be a useful tool for monitoring remission in IBD therapies. 239

240

P-I-0094 Accurate and effecitve protein database construction strategy of metaproteomics Enhui Wu1 1Fudan University, Chemistry, Shanghai, China Metaproteomics can provide a direct means to identify microbial proteins in microbiota for compositional and functional characterization. However, in-depth and accurate metaproteomics is still limited due to the extreme complexity and high diversity of microbiota samples. We proposed an accurate taxonomic annotation pipeline for genes from metagenomic data, namely contigs directed gene annotation (ConDiGA), and used this method to build protein sequence databases for metaproteomic analysis. ConDiGA can tackle the current taxonomic annotation reliability problem in metagenomics-derived protein sequence database and can promote the in- depth metaproteomic analysis of microbiome. Moreover, when metagenomics-derived database is not available, public databases are frequently employed in metaproteomics. Nevertheless, utilizing large public databases for metaproteomic analysis can cause an expansion in the search space, prolong the search duration, and diminishing the sensitivity of identification. To address this issue, we present a high-abundance protein-guided hybrid spectral library strategy for in-depth data independent acquisition (DIA) metaproteomic analysis (HAPs- hyblibDIA). HAPs-hyblibDIA achieves effective filtering of public databases whose protein and peptide identification numbers are comparable to metagenomics-derived databases. 240

241

P-I-0095 Towards establishing proteomic analyses of indoor air bioaerosols – optimizing protein filter extraction Leonie T D Wuerger1, Jennifer Bortlik1, Robert M Leidenfrost1, Dierk-Christoph Pöther1 1Federal Institute for Occupational Safety and Health, Berlin, Germany People spend most of their time indoors, both at work or at home. The quality of indoor air is therefore crucial for the well-being of the population. The composition of bioaerosol as a part of indoor air, has not yet been sufficiently characterized. Previously established, usually culture-based, analysis methods do not allow a full characterization of bioaerosols in indoor spaces. Therefore, modern omics methodologies may provide a more comprehensive characterization of bioaerosols. As part of the presented project, proteomic analysis of bioaerosols will be established and applied to bioaerosol samples from various indoor spaces. A challenge in indoor air analysis is the choice of filter material used to collect the bioaerosols. The filter material plays an important role, as its properties determine sample yield. While quartz filters are commonly used, they generally show lower bioaerosol recovery in comparison to hydrophilic poly-carbonate (PC) or polyvinylidene fluoride (PVDF) filters. The length of the sampling period can also have an effect, as bioaerosols might dry out on the filter over time. Furthermore, bioaerosol size plays a role in recovery as well, as smaller microorganisms can be embedded deeper into the fibers of the filters. Therefore, the choice of filter material plays an important role regarding the optimization of protein extraction. To establish a pipeline for proteome analysis of indoor air, a standard sample consisting of a mixture of microorganisms was applied to different filter materials. Furthermore, we tested different lysis methods and compared the extracted protein amounts to the standard sample to determine the optimal filter material and lysis method for indoor bioaerosol proteomics. For verification of the method bioaerosol samples were collected by filtration from varying indoor spaces for metaproteomic analysis. As bioaerosol concentrations in indoor spaces are generally low, an optimized analysis pipeline for proteomics may improve detectability of low abundance proteins otherwise overlooked. 241

242

P-I-0096 Machine learning based on multi-omics data reveals microbial contribution during cervical pre-cancer lesions development Jianxujie Zheng1 1Fudan university, Chemistry, Shanghai, China closely related to women's is vaginal microbiota vaginal health. Persistent vaginal The microbiota disorder can greatly increase the risk of human papillomavirus (HPV) infection and may cause gynecological diseases including squamous intraepithelial lesion (SIL) and cervical cancer. However, little is known about the microbial contribution during HPV-related cervical cancer development. Herein, we applied machine-learning classifiers to a dataset of clinical and multi-omic molecular features from patients for SIL grading. Among all single-analyte classifiers, the accuracy of vaginal human proteins, species abundance based on 16S sequencing and vaginal metabolites was greater than 0.9. In addition, combining multiple clinical and omic analytes gave better classification accuracy than using single one. Features related to bacterial glycogen and D-amino acid metabolism showed higher importance in multi-omic classifiers, which reveals that the vaginal microbiota of high-grade SIL patients are better able to utilize glycogen secreted by host cells and to synthesize D-amino acids, thus causing an inflammatory response compared to that of low-grade SIL patients. This approach is expected to transfer for prediction of SIL progression and to assist in clinical diagnosis and treatment. 242

243

New technology: MS-based proteomics (P-I-0097 to P-I-0204) P-I-0097 High throughput absolute quantitation of plasma proteins in young and old cohort Bich Nguyen1, Shreya Chand1, Balazs Szoke1, Sara Ahadi1 1Alkahest, San Carlos, CA, United States Mass spectrometry (MS)-based proteomics is a powerful technique for analyzing plasma proteins. It enables comprehensive protein cataloging at an unprecedented scale, offering immense potential for biomarker discovery and understanding disease mechanisms. Combined with stable isotope-labeled internal standards (SIS), targeted MS allows absolute protein quantification. This surpasses antibody-based methods like ELISA, which are often less robust and time-consuming. However, achieving absolute quantification in high-throughput MS analysis remains a challenge, as it becomes more difficult with increased multiplexing. The introduction of SureQuant™ technology addresses this limitation. SureQuant utilizes spiked-in SIS peptides to trigger real-time acquisition of endogenous peptides, eliminating the need for complex retention time scheduling. For plasma analysis, SureQuant can be combined with the PQ500™ Assay Panel, containing over 800 reference peptides representative of over 500 proteins, and SpectroDive TM software for high-throughput analysis of targeted proteomics data. Together, this combined workflow provides a multiplexed, targeted approach that enables rapid and robust quantification of hundreds of plasma targets with easy method setup and streamlined data analysis. Here, we highlight the application of this streamlined workflow for analyzing plasma samples from young and old healthy individuals collected using EDTA and sodium citrate as anticoagulants, and via plasmapheresis. In average 415 proteins are quantified in EDTA, 399 in sodium citrate and 403 in plasmapheresis samples. Shared proteins quantified among all three sample types are highly correlated with each other with spearman rho of 0.97, 0.98 and 0.99 between EDTA and sodium citrate, EDTA and plasmapheresis and sodium citrate and plasmapheresis pairs, respectively. However, Log2 foldchange between pools of young and old analyzed separately for male and female samples are less correlated between the plasma types, with highest correlation observed between plasmapheresis and citrate samples (spearman rho = 0.71). We have then compared the absolute protein values from our SureQuant analysis with estimated and published values from Human Protein Atlas (HPA) database. There are 513 proteins overlapped between the two datasets. In average protein concentrations (pg/mL) in SureQuant dataset are most different from HPA values in citrate and plasmapheresis samples and closest to EDTA samples, average log2FC of -0.57, -0.44 and -0.054, respectively. Overall, our pipeline makes high throughput absolute quantitation easy to implement for more than 500 proteins. Moreover, it is also extendable to other targets with high confidence, paving the way for development of future-high throughput SureQuant based analysis. 243

244

P-I-0098 Proximity labeling and SILAC based proteomic approach identifies proteins at the interface of homotypic and heterotypic cancer cell interactions Nazan Saner1, Ceren Uzun2, Büşra Aytül Akarlar1, Daniel Jon Geiszler1, Nurcan Tunçbağ2, Nurhan Özlü1 1Koç University, Department of Molecular Biology and Genetics, Istanbul, Turkey 2Koç University, Department of Chemical and Biological Engineering, Istanbul, Turkey Cellular communication, particularly between different types of cancer cells, plays a crucial role in tumor progression and invasion. This study employed a combinatory approach of horseradish peroxidase (HRP)- dependent proximity biotinylation and stable isotope labeling by amino acids in cell culture (SILAC) to identify proteins involved in homotypic (same cell type) and heterotypic (different cell types) interactions in breast cancer cell lines. Epithelial (MCF7) and mesenchymal (MDA-MB-231) breast cancer cells were co-cultured to model these interactions. HRP was targeted to the extracellular leaflet of the plasma membrane in 'bait' cells to label proximal proteins in 'prey' cells. Prey cells were metabolically labeled with heavy isotopes, enabling differentiation of self-biotinylated proteins from those of interacting cells. Mass spectrometry analysis identified cell surface and extracellular proteins enriched at the interface of cell interactions. The study revealed distinct and common proteomic profiles for homotypic and heterotypic interactions and demonstrating that the methodology effectively maps the proteome of cancer cell interactions. This approach provides insights into the molecular mechanisms of cancer cell communication, potentially informing therapeutic strategies to inhibit tumor invasion. 244

245

P-I-0099 Single-Cell proteomics using inCapS to dissect cellular states through protein correlation analysis Syed Azmal Ali1, Adela-Eugenie Vrsanova1, Lina-Marie Wagner1, Jeroen Krijgsveld1 1German Cancer Research Center, Division of Proteomics of Stem Cells and Cancer, Heidelberg, Germany Protein and mRNA levels within cells fluctuate to maintain cellular homeostasis, leading to proteome differences among individual cells. However, it is unknown to what extent variation in protein expression at the single cell level is a random or coordinated process, by what principles it is guided, and which proteins are involved. Therefore, here we employed a novel single-cell proteomics approach to investigate protein co-regulation in three breast cancer cell lines. This revealed distinct modules of proteins whose expression show strong co- or anti-regulation, either in all cell lines or in a cell-line specific manner, suggesting that co-regulation of specific proteins define cellular states within heterogeneous populations. To understand protein co-regulation, we developed a robust novel proteomic workflow called inCapS (in- capillary sample preparation) for single-cell proteome analysis. This method integrates miniaturized sample preparation, low flow-rate chromatography, and state-of-the-art mass spectrometry using a timsTOF ultra. Utilizing inCapS, we achieved extensive proteome coverage of approximately 3,000 proteins in a label-free manner across 178 samples, examining proteome differences in three breast cancer cell lines: MB468, T47D, and MCF7. We evaluated interdependencies of protein expression for the pairwise combinations of all proteins in the single-cell proteomic data sets of each cell line (926,841 combinations). This revealed distinct modules of proteins whose expression showed a strong positive correlation, such as CFL1, ATP5F1A, and CCT6A that each correlated with the expression of >140 other proteins operating in the same complex or pathway. In addition, >1000 protein pairs correlated negatively, indicating mutual exclusivity in their expression. For instance, across all three cell lines, FLI1 was consistently the most negatively correlated with 83 other proteins, including the Nop56p-associated pre-rRNA complex. Furthermore, in a GSEA analysis, SETD7 and SNRNP70 were identified as key transcription factors associated with proteins that are anti-correlated across all three cell lines. These results suggest the co-existence of cells with distinct proteome compositions, representing different cellular states. Both positively and negatively regulated protein modules only partially overlapped between the three cell lines, suggesting cell-line specific proteome regulation in individual cells. Altogether, these findings potentially underlie the distinct biological properties of the respective cell lines. Collectively, our single-cell proteome analysis revealed coordinated protein covariation as a novel dimension of proteome regulation afforded by single-cell proteomics, and it showed the existence and properties of cellular states as a defining principle of heterogeneity within cellular populations. 245

246

P-I-0100 High-throughput workflow for in-depth proteome coverage and quantitative analysis of low sample amounts and single-cells Tabiwang Arrey1, Jenny Ho2, Min Huang3, Haoran Huang3, Nicolaie Eugen Damoc1 1Thermo Fisher Scientific, Bremen, Germany 2Thermo Fisher Scientific, Hemel Hempstead, United Kingdom 3Thermo Fisher Scientific, Shanghai, China In MS based proteomics, millions of cells are analyzed at the same time and protein changes are presented as a sum to the total number of cells analyzed. However, it is known that each cell carries out its own biological processes. Therefore, a better understanding of this diversity can be achieved by studying as many individual cells as possible. Improvements in MS instrumentation have enhanced sensitivity, dynamic range, and large-scale profiling with improved throughput. However, maintaining the proteome depth at throughput is still challenging. Thus, we optimized and benchmarked the performance of a low nanoflow LC-MS methods in data-independent acquisition (DIA), on the Orbitrap Astral MS, to achieve deep proteome profiling at highest throughput for single cell. For the dilution, Pierce™ HeLa digest (20 ug) was reconstituted by adding 200 μL of 0.015% DDM (Dodecyl β D maltosid) solution, sonicated for 5 min, then diluted to 5 ng/μL in 0.015% DDM. Individual HeLa cells were isolated using CellenONE® system from Cellenion, followed by reduction, alkylation, and trypsin digestion as per manufacturer"s instructions. The data was acquired on a Thermo Scientific™ OrbitrapTM Astral™ mass spectrometer with a Thermo Scientific™ FAIMS Pro™ interface. Samples were separated using Thermo Scientific™ Vanquish™ Neo UHPLC system on an IonOpticks Aurora 25 cm TS C18 column. Data was acquired in DIA mode. Raw files were processed with Spectronaut 18, with and without library, using a human database with 20k sequences. Though throughput is determined by sample preparation, LC separation and MS acquisition, we focused more on the LC optimization since the MS part has been previously optimized [Santosh et al.]. During the separation, the flow rate is ramped multiple times, starting from 450nl\min at the beginning of the gradient. During peptide elution, the flow rate dropped to 200nl\min and later ramped up to 300nl\min during the washing step. Spectronaut analysis, with no library, from 250pg HeLa digest on column, separated using a 19.5min active gradient, we could identify more than 5600 protein groups from individual files. When triplicated files are combined and processed together, the number of protein groups at 1 % FDR goes up to over 6100, with median CVs below 10%. With a library generated from 10ng the protein groups increased to over 7400. To evaluate the reproducibility and reliability of the optimized method, it was transferred to multiple Thermo Fisher Scientific sites across the globe to be executed on different instruments and users. All results from the different sites were within ± 10%, demonstrating the reproducibility and reliability of this method. Analyzing real HeLa single cells, we identified on average of over 6,500 protein groups without using a library. 246

247

P-I-0101 Unveiling proteomic insights: Orbitrap Astral's performance in multiplexed TMT analyses of brain Organoids in Alzheimer's disease research Ana Montero-Calle1, Patricia Mateos-Martínez1, Raquel Coronel1, Victoria López-Alonso1, Isabel Liste1, Rodrigo Barderas1,2 1Instituto de Salud Carlos III, Chronic Disease Programme, Madrid, Spain 2CIBERFES, Madrid, Spain New mass spectrometry technologies are crucial for driving scientific progress. In the last years, the development of fast, highly sensitive and resolution mass spectrometers allowing deep proteome coverage and accurate protein quantification has significantly expanded the capabilities and applications of this analytical technique. The Orbitrap Astral state-of-the-art mass spectrometer delivers unprecedent performance in proteomics analyses. While the Orbitrap Astral has demonstrated impressive performance in label-free proteomics workflows, its applicability in multiplexed experiments, such as tandem mass tag (TMT) analyses, has not been extensively explored yet. To address this question, we have compared here the performance of the Orbitrap Astral to other leading Orbitrap-based mass spectrometers, the Orbitrap Exploris 480 and Orbitrap Eclipse Tribrid, in TMT-based quantification experiments. To this end, we have investigated the protein content of brain organoids derived from healthy and Alzheimer's disease (AD) patients, which mimic the healthy and pathological brain context, respectively. A 16-plex TMT experiment was performed to comprehensively characterize the proteome of these organoids, representing two healthy brain models and two AD models (overexpressing APP or with a mutation in PSEN1), harvested at two time points (5 weeks and 2 months). AD is the most common type of dementia worldwide, and an in-depth characterization of the pathological changes during disease progression is crucial to elucidate novel proteins and biological processes that may serve as prognostic or diagnostic biomarkers, as well as potential therapeutic targets. The comparison of four different DDA-based TMT LC- MS/MS methods using the previously described three mass spectrometers aimed to identify the best approach for sensitive and accurate analysis of these organoids TMT samples. After protein extraction, reduction, and alkylation, proteins were trypsin digested and TMT labeled. Labeled peptides were pooled together, and off-line fractionated into 12 fractions using a high-pH reverse phase column. Individual fractions were then analyzed by mass spectrometry on the three mass spectrometers in a 2-hours gradient. In addition, a 30-minutes gradient was also investigated with the Orbitrap Astral. Then, raw data were analyzed with the MaxQuant (version 2.4.2) and with the R program for the identification of dysregulated proteins in the disease. The results obtained indicate that the Astral mass analyzer can deliver comparable or even superior depth of proteome coverage and quantification accuracy compared to the established Orbitrap platforms. A higher number of peptide and protein identifications were obtained with the Orbitrap Astral mass spectrometer, with significant improvements observed when using a 2-hour gradient instead of a 30-minute gradient. 247

248

P-I-0102 Novel data-independent acquisition (DIA) strategy for higher throughput quantitative proteomics Ihor Batruch1, Jason Causon1, Katherine Tran1, Patrick Pribil1, Naomi Diaz2 1SCIEX Canada, Concord, Canada 2SCIEX, Dallas, TX, United States Introduction: Obtaining insight into biological systems requires the ability to detect and reproducibly quantitate peptides of interest. ZT Scan DIA is a Zeno trap-enabled scanning DIA strategy currently deployed on a research- modified Zeno trap-enabled QTOF instrument with a sliding quadrupole (Q1) through the precursor m/z to select precursors for CID (collision-induced dissociation) fragmentation in a collision cell. Continuous quadrupole movement across the precursor masses can be directly correlated to the continuously changing intensity of fragment ions arising from those masses, thereby enabling the direct assignment of those precursors to their respective MS/MS spectra in a time-dependent manner. Since each fragment ion must have an identical pattern of increasing and decreasing intensity mapping to the precursor being scanned by the quadrupole, fragments arising from different precursors can be distinguished enabling improved spectral deconvolution. Thereby, we have used varying levels of commercial K562 tryptic digest loads (low to 200 ng) across gradients as short as 1 min up to 30 min with nanoflow and microflow with ZT Scan DIA and Zeno SWATH DIA to assess protein detections and quantitation. Methods: A research modified Zeno trap-enabled QTOF with an OptiFlow Turbo V ion source in line with a Waters M-Class LC system with nano- and microflow columns was used to acquire data with ZT Scan DIA and Zeno SWATH DIA. ZT Scan DIA with nanoflow (150-500 nl/min) active gradient lengths (water to 5-35% acetonitrile with 0.01 % formic acid) from 5-30 min in length and microflow (5 µL/min) gradients of 1-15 min in length were evaluated across varying K562 cell digest (SCIEX) loads with acquisition speeds up to 640 Hz. Data was processed with DIA-NN (v. 1.8.1) and PEAKS Studio (v. 12 beta) using a spectral-library or library-free (FASTA) mode. Results: A novel acquisition strategy, ZT Scan DIA, delivers over 2x gains in quantitative performance at low loads using nanoflow relative to Zeno SWATH DIA using similar acquisition time and identical gradient length. At higher loads, ZT Scan DIA offers similar quantitative performance with an active LC gradient that is 3-5x shorter than with Zeno SWATH DIA. For a 200 ng K562 load with a 3 minute LC gradient at microflow, we can detect nearly 5300 protein groups with over 4600 having a CV under 20%. Conclusions: ZT Scan DIA improves qualitative and quantitative performance at higher throughput relative to Zeno SWATH DIA. 248

250

P-I-0103 Enabling scalable single-cell proteomics by utilizing the unique analytical properties of the Evotip Pure and Whisper Zoom methods Dorte Bekker-Jensen1, Frederik Haugaard Holck1, Marko Tijardović1, Ole Bjeld Hørning1, Nicolai Bache1 1Evosep, Odense, Denmark Single-cell proteomics is a rapidly developing field with recent advancements improving robustness and sensitivity in all parts of the workflow. The new Whisper Zoom methods set a new standard for enabling high- end technology. This ensures that researchers with all levels of expertise will benefit from the robustness and ease-of-use to achieve reproducibility in high-end LC-MS proteomics. While high performance and run-to-run reproducibility is a hard requirement for all types of experiments the additional complexity of setting up and maintaining multiple LC-MS instruments with decent inter-instrument reproducibility is challenging - even for experts. The Evosep One technology has solved this through a standardized offering, that combined with the new Whisper Zoom methods provides cutting-edge performance with unprecedented instrument-to-instrument reproducibility – even for non-experts. Here we benchmark the performance of the new Whisper Zoom methods, 20, 40, 80 and 120SPD and evaluate and reproducibility in internal and external labs. We have evaluated the performance on ten different instruments, five internally and five from external sites. We demonstrate a significant improvement in both inter and intra lab reproducibility between the legacy Whisper methods and the new Whisper Zoom methods From Whisper 80SPD the intra-instrument reproducibility was measured to 4.6 seconds and the inter-instrument reproducibility is 15.9 seconds which is an improvement of 85% compared to the legacy Whisper methods. We also demonstrate the relative performance through HeLa dilution curves representing sub cellular sample amounts to higher loads to cover a broad range of applications. We demonstrate cutting-edge performance which is possible to easily setup and maintain on multiple instruments in the same lab as well as external labs to either share or compare data within or across projects. 250

251

P-I-0104 A systematic optimization of diagonal-PASEF acquisition for low- to medium-input acquisitions on the timsTOF HT mass spectrometer Christopher Below1, Stephanie Kaspar-Schoenefeld2, Oliver Bernhardt1, Sander Willems3, Tejas Gandhi1, Roland Bruderer1, Dennis Trede2, Jonathan Krieger2, Lukas Reiter1 1Biognosys AG, R&D, Schlieren, Switzerland 2Bruker Daltonics GmbH & Co. KG, Bremen, Germany 3Bruker Belgium NV, Kontich, Belgium Liquid chromatography mass spectrometry methods aim to yield high peptide and protein identifications while retaining quantitative accuracy and precision. To increase the analytical performance of the timsTOF mass spectrometers, researchers have recently proposed diagonal dia-PASEF acquisition methods termed as synchro- or midia-PASEF which attempt to acquire the ion cloud more efficiently (Skowronek, 2023; Distler, 2023). These methods operate by seamlessly and continuously following the observed diagonal shape of the precursor ion density. While such methods promise to cover the observed ion-cloud more efficiently, their utility is currently still hampered by a lack of systematic method evaluation. We therefore set out to optimize the diagonal-PASEF acquisitions on a timsTOF-HT mass spectrometer for different applications. We first developed an automated Rshiny-based solution that can seamlessly and efficiently compute any diagonal-PASEF method from dda-PASEF or dia-PASEF acquisitions. We show that this tool can build methods with different specifications depending on user-desires on vastly different biological matrices. Next, we optimized the diagonal-PASEF acquisitions in respect to their ion-mobility settings. We infused a HEK-293 sample at 100 ng on a 17-minute analytical gradient on our timsTOF-HT and analyzed the data in Spectronaut 19. We found that the optimal ion-mobility settings yielded up to 10% more protein groups and up to 56% more precursors compared to settings that could have been chosen if no systematic optimization had been conducted. Next, we optimized the overall width of the diagonal-PASEF methods and the number of slices for each diagonal-PASEF method systematically at a loading amount of 10, 100 and 800 ng. The optimal method was found to strongly differ across the tested loadings indicating that diagonal-PASEF strongly benefits from systematic method optimization efforts. The best performing diagonal-PASEF method achieved on average 9% and 2% more protein group identifications against a set of matching dia-PASEF methods at 10 ng and 100 ng of loading specifically while at 800 ng diagonal-PASEF achieved 2% lower protein group identifications against a set of dia-PASEF methods. The experiments indicated that diagonal-PASEF could be a significant extension of the dia-PASEF group of methods. To test the performance of our optimized method we evaluated the accuracy and candidate recovery in a controlled quantitative experiment using a 4-species mixture in known ratios. Here, we found that the diagonal- PASEF method not only improved the precision by up to 30% but also the accuracy by up to 26% compared to dia-PASEF. This resulted in an increase of the truly differential abundant proteins at 5% error rate by up to 53%. In conclusion, we show that the improved duty cycle of the diagonal window dia-PASEF methods together with the systematic optimization of the methods holds a significant potential for quantitative DIA proteomics. 251

252

P-I-0105 Proteome-wide monitoring of drug action in living cells using a novel label-free solvent-based shift assay Raffaela Berger1, Dominik Steinbrunn1, Hannes Hahne1 1OmicScouts GmbH, Freising, Germany Detection of proteome-wide interactions between a ligand and cellular proteins is essential during drug discovery and development. It allows for drug-target identification and mechanism of action deconvolution, as well as identification of off-targets and potential liabilities. To this end, various label-free and label-based chemical proteomic assays have been developed over the past 25+ years. Among the label-free approaches, Thermal Proteome Profiling (TPP) and Solvent-Induced Proteome Profiling (SIPP) enable monitoring ligand- induced stabilization of proteins against denaturation by heat or exposure to organic solvents [1, 2]. SIPP works analogous to TPP, but without the need for applying heat. To explore whether the SIPP assay can be used to detect drug-target interactions in living cells, we established a solvent-based shift assay to probe the biophysical stability of proteins in cell extracts as well as in cells. The assay has been automated on a liquid handling station, and works with adherent and suspension cells equally well. We evaluated the assay, which we termed SideScout™, with well characterized tool compounds and drugs of different classes: We employed the SideScout™ assay to detect the interaction between methotrexate and its known target DHFR, as well as stabilization of the known targets of two kinase inhibitors (MK-2206, SCIO-469), a histone deacetylase inhibitor (Vorinostat) and a natural product (Cyclosporine A) in cell extracts and in living cells. We also explored the assay"s capabilities to detect target engagement of covalent drugs (Ibrutinib, THZ-1), heterobifunctional degraders and molecular glues (Indisulam, MZ1). In summary, the live cell assay successfully recapitulated the identification of the designated targets of the example tool compounds and drugs in all cases. The results also demonstrate that live cell experiments lead to more complex results than comparable experiments in cell extracts. However, this allows detecting downstream and MoA-related effects, which eventually modulate protein stability (such as changes in PPIs, phosphorylation status or substrate availability) as well as the target profiles of pro-drugs or drug metabolites. Taken together, the SideScout™ solvent shift assay complements the chemical proteomics toolbox and represents a valuable and reliable tool for monitoring drug action in living cells. Literature: [1] Savitski M., Science 2014, 346(6205):1255784 [2] Zhang X., Anal. Chem. 2020, 92, 1, 1363–1371 Patent pending 252

253

P-I-0106 Single-cell proteomics and AP-MS to elucidate auditory mechanoelectrical transduction in mammalian cochlea Ruxandra Bighiu1, Elise Pepermans1, Andrea Lelli2, Christine Petit2, Geert Baggerman3, Vincent Van Rompaey1 1University of Antwerp, Translational Neurosciences, Antwerp, Belgium 2Institut Pasteur, Hearing Institute, Paris, France 3University of Antwerp, AdRem Lab, Antwerp, Belgium The mammalian cochlea houses two distinct mechanosensitive receptor cell types: inner hair cells (IHCs) and outer hair cells (OHCs). Despite their critical role in auditory function, the intricate molecular processes that underlie their unique morphological and functional characteristics remain largely unexplored. Our study aims to fill this gap by constructing an exhaustive proteomic profile of IHCs and OHCs isolated from postnatal day 7 (P7) murine cochlea. This deep proteomic profiling enables the identification of pivotal proteins implicated in precise cellular functionalities or pathological pathways. Given the limited quantity of hair cells within each cochlea—approximately 3,500 IHCs and 12,000 OHCs—and their inherent fragility, conventional high-throughput collection methods are inadequate for single-cell proteomics. To address this challenge, we employed the suction pipette technique to isolate individual OHCs from enzymatically treated and mechanically dissociated murine cochlea at P7. These isolated OHCs were then cryopreserved in liquid nitrogen. Sample preparation involved a streamlined one-pot method that included lysis, sonication, denaturation, and tryptic digestion in a minimal volume, thereby reducing sample handling and potential contamination. For mass spectrometric analysis, we utilized the TimsTOF SCP, coupled with an Evosep One chromatography system. This method has yielded up to 500 protein groups per OHC and 2200 protein groups for 5 OHCs. Our primary focus is on OHCs due to their greater abundance and ease of collection. Ultimately, we aim to conduct a comparative analysis between OHCs and the more challenging-to-isolate IHCs. Additionally, we seek to correlate this dataset with established molecular structures in auditory mechanisms and use it to complement affinity-purification mass spectrometry (AP-MS) experiments targeting known components of the hearing machinery, such as protocadherin 15 (Pcdh15). This approach includes utilizing AP-MS with specialized antibodies against Pcdh15 to identify new components of the lower tip-link complex, crucial for mechano- electrical transduction (MET). Protein-protein interactions between potential new and known MET components will be validated using Alphafold2. These findings promise to provide critical insights into cochlear mechanosensation by elucidating the molecular mechanisms of both IHCs and OHCs. Furthermore, our collection method and subsequent sample preparation approach offer a viable option for single-cell proteomics studies on other challenging-to-collect cell types. This research advances our understanding of auditory biology and lays the groundwork for innovative therapeutic strategies to address hearing impairments, potentially transforming the landscape of auditory health and disease management. Fig. 1 (optional) 253

254

P-I-0107 SysQuan: a novel strategy for the absolute quantitation of human proteins Christoph H. Borchers1,2,3,4, Vincent Richard4, Robert Popp5, Rene Zahedi6,7,8,9, Yassene Mohammed2,4,10 1McGill University, Department of Pathology, Montreal, Canada 2Jewish General Hospital, Gerald Bronfman Department of Oncology, Montreal, Canada 3McGill University, Division of Experimental Medicine, Montreal, Canada 4Jewish General Hospital, Segal Cancer Proteomics Centre, Montreal, Canada 5MRM Proteomics, Montreal, Canada 6University of Manitoba, Manitoba Centre for Proteomics and Systems Biology, Winnipeg, Canada 7University of Manitoba, Department of Biochemistry and Medical Genetics, Winnipeg, Canada 8University of Manitoba, Department of Internal Medicine, Winnipeg, Canada 9CancerCare Manitoba Research Institute, Winnipeg, Canada 10Leiden University Medical Center, Leiden, Netherlands Introduction: The "gold standard" for quantitative proteomics is absolute quantitation using targeted mass spectrometry (MS) with triple quadrupole MS and stable-isotope-labeled (SIL) peptides. The high cost of SIL peptides has, thus far, precluded the use of proteomics in large clinical studies where 1000s of samples often need to be measured. We have developed a new method that allows the absolute quantitation of 1000s of proteins in different human biofluids and tissues at a fraction of the current cost, and which can be used for both classical targeted MS and non-targeted approaches, such as data independent acquisition (DIA). This method is based on the use of body fluids and tissues from stable-isotope (13C- and 15N) labelled (SIL) mice. Methods: This method is based on the use of body fluids and tissues from stable-isotope labelled (SIL) mice. By analyzing the labeled mouse samples, we can identify surrogate peptide sequences that are identical between mouse and human proteins. When mixed with the corresponding human tissue/body fluids, all of the peptides that can be used for absolute quantitation will appear as doublets (light/heavy pairs). Then, using cost-effective "light" (i.e., unlabelled) synthetic peptide standards corresponding to the native peptides (NAT), we can determine the concentrations of these peptide sequences in the SIL mouse reference materials. Results: We have calculated that 13,926 proteins of the human proteome (based on 146097 tryptic peptides 7 to 25 amino acids long) can theoretically be quantified using this approach. This means that, for the first time, the "proteome-wide" absolute quantitation can be performed, without the current limitations of quantitative MS. For those proteins without suitable tryptic peptides for quantitation, we will use alternative enzymatic digestion procedures. We have already shown that it is possible to achieve absolute quantitation of hundreds of proteins in human plasma in a single non-targeted LC-MS run on the timsTOF. In a fractionated run using that number increases to >650 proteins. Using high pH pre-fractionation (with 48 fractions) followed by DDA on an Evosep interfaced to a timsTOF we were able to detect 1159 proteins (covered by 3639 heavy-light peptide pairs) in plasma and 6987 proteins (covered by 48760 heavy-light peptide pairs) in liver. We are in the process of transforming our untargeted LC- MS/MS assay into a targeted LC-MRM/MS approach for absolute quantitation using the Evosep/Agilent 6495D QQQ equipment. Our first project will be the >1000 proteins detected in liver which are involved in metabolic pathways. All of the acquired peptide doublets can later be assigned to SIL mouse concentrations, even if these are determined post-acquisition. This new SysQuan approach will be a "game-changer" for non-targeted and absolute protein quantitation as it theoretically would allow a researcher to quantify the entire human proteome. 254

255

P-I-0108 Mass spectrometry-based single cell proteomic analysis of primary human neutrophils Alejandro Brenes1, Rupert Mayer2, Pranvera Sadiku1, Leila Reyes1, Paul Brennan3, Ailiang Zhang1, Patricia Coelho1, Karl Mechtler2, Sarah Walmsley1 1University of Edinburgh, Centre for Inflammation Research, Edinburgh, United Kingdom 2Research Institute of Molecular Pathology (IMP), Vienna, Austria 3University of Edinburgh, CRUK Edinburgh Centre, Edinburgh, United Kingdom Introduction: Neutrophils are vital immune cells that possess an arsenal of antimicrobial mechanisms, enabling them to kill pathogens but also damage healthy tissue if misused. They have been described as heterogeneous with direct involvement in human diseases ranging from respiratory disease to cancer. They are produced in the bone marrow, where they egress as with little transcriptional capacity, making scRNASeq poor at characterising them. Thus, a single cell proteomic (SCP) analysis would provide a much closer link to their phenotypes and functional capacities. Methods: Single neutrophils were sorted into 384 well plates using flow cytometry. They were frozen and later centrifuged at 300xg for 3 min before placing them in the CellenONE X1 at 50°C and 85% humidity, followed by incubation for 2h at 50°C at 85% relative humidity. Samples were kept hydrated every 15 min by addition of 500 nL water. 30 min post incubation, 500nL of 3ng/μL trypsin are added. After lysis and digestion, 2.5 μL of 0.1% TFA with 5% DMSO were added for quenching and storage. Samples were analysed on an Orbitrap Astral, using the Vanquish Neo UHPLC with an Aurora Ultimate TS25cm nanoflow column with a 50 SPD method. The MS1 spectra were recorded at 240k res. from m/z 400-900 with AGC target of 500% and a max. injection time of 100 ms. MS2 in DIA mode used the Astral analyzer, with non- overlapping isolation windows of m/z 20 and a scan range from m/z 400–800. The precursor accumulation time set to 60ms and the AGC target to 800%. Raw files were searched on Spectronaut18. Results: >1,250 proteins were identified within a single neutrophil, > tripling 10x scRNASeq coverage. The increased depth enabled the mapping of the prime metabolic enzymes, and the study of cell signalling components, including vital pattern recognition receptors like TLR2, important kinases like SYK, adaptor vital adaptor proteins like FCER1G and phosphatases like PTPN6. It has also enabled a comprehensive coverage of the main neutrophil granules. This data has enabled the study of the heterogeneity within cells, validating the differences in important markers like CD177 and showing the consistency of some of the granules proteins like ELANE. It also enabled the discovery of a neutrophil population that displayed significant reductions in the abundance of the vital granules as well as the rate limiting metabolic enzymes. Conclusions: By using SCP, we have identified >1,250 proteins per cell, enabling an expanded coverage of neutrophil functions at a single cell level. The data mapped the main metabolic pathways, important receptors that respond to tissue damage and infections, signalling modulators as well as important granule proteins. This work empowers the study of heterogeneity across neutrophils inter and intra individuals, enabling a better mapping neutrophil functional capacities and phenotypes and help to discover new populations and target those that are associated with disease and disfunction. 255

256

P-I-0109 High-throughput high-resolution data-independent acquisition workflow on an Orbitrap exploris 480 hybrid mass spectrometer for accurate label-free quantitation Kevin Yang1, Julia Kraegenbring2, Julian Saba3, Maciej Bromirski4, Amirmansoor Hakimi1 1Thermo Fisher Scientific, San Jose, CA, United States 2Thermo Fisher Scientific, Bremen, Germany 3Thermo Fisher Scientific, Winnipeg, Canada 4Thermo Fisher Scientific, Warsaw, Poland Traditional data-dependent acquisition (DDA) approaches have been widely employed for LFQ experiments, but they suffer from run-to-run inconsistencies due to intensity-based stochastic triggering of precursors, often leading to undersampling of low-abundant proteins. Missing values become more likely as sample size increases, so data-independent acquisition (DIA) has emerged as a popular technique for large scale quantitative analyses. A suitable analytical workflow addresses the need for reproducible sample preparation, robust separations, high-quality quantitative measurements, and reliable data analysis. In this study, we developed a high-throughput Velocity DIA workflow on the Orbitrap Exploris 480 mass spectrometer with in-depth proteome coverage and highly accurate label-free quantitation. As an evaluation of performance, HeLa digest and three- proteome mixtures were loaded onto a 50 cm μPAC Neo HPLC column in direct injection mode using a Vanquish Neo UHPLC system over 9 min, 30 min, and 60 mins active LC gradients, respectively, before being transferred into the Orbitrap Exploris 480 mass spectrometer. Acquired data has been processed by Spectronaut (Biognosys, v18) using a directDIA approach, DIA-NN (v3.1.0.638) using CHIMERYS™ intelligent search algorithm by MSAID. (v1.8.1) or Proteome Discoverer software With a 30 min active gradient, 6,300+ proteins and 40,000+ peptides were identified, along with a protein group CV of approximately 5%, suggesting that the 30 min active gradient method provides the strongest analytical balance to achieve throughput while maximizing identification and quantitative performance. We extended this workflow to a 60 min active gradient and successfully identified close to 7,200 proteins and >60,000 peptides, highlighting that deeper proteome coverage can be achieved in the Velocity DIA workflow by using a longer gradient. To test the robustness and reproducibility of quantitation accuracy, in this data set, we used two samples with different amounts of spiked microbial proteins to mimic biological samples where proteins might be up- or downregulated under different conditions. The Velocity DIA workflow used for relative quantitation of e.coli and Yeast proteins in a high amount of human peptides as background yields excellent quantitation accuracy across all ratios, with median values exhibit a high degree of proximity to the theoretical ratios, as well as a narrow distribution of all data points around the median values, indicating high quantitative accuracy and precision of the workflow. Additionally, the three-proteome mix experiment further highlights the proteome depth that can be achieved with an Orbitrap Exploris 480 MS. In the 60 min active gradient, >10,000 protein groups were identified. In conclusion, we demonstrate the high-throughput, high-resolution and high quantitative precision, and accuracy of the Velocity DIA workflow for label-free quantitation on an Orbitrap Exploris 480 MS. 256

257

P-I-0110 Generation of ground-truth data for the systematic evaluation of data normalization strategies in single-cell proteomics data Marcel Bühler1,2, Emin Araftpoor1,2, Christophe Vanderaa3, Lieven Clement3,4, An Staes1,2,5, Kris Gevaert1,2 1Ghent University, Center for Medical Biotechnology (CMB), Zwijnaarde, Belgium 2Ghent University, Center for Medical Biotechnology, Department of Biomolecular Medicine, Ghent, Belgium 3Ghent University, Department of Applied Mathematics, Statistics and Computer Science, Ghent, Belgium 4Ghent University, Bioinformatics Institute, Ghent, Belgium 5Ghent University, Proteomics Core Facility, Ghent, Belgium Single-cell technologies have become a fundamental part of biomedical research in the past years, yet these have been mainly limited to the study of transcriptomes. In contrast, proteomic diversity within or between cell populations is still heavily undercharacterised. Mass spectrometry-based single-cell proteomics (SCP) offers novel opportunities to study proteome variation on the single-cell level. Such variation can, however, be attributed to various sources, both biological and technical in nature. To fully unlock the quantitative potential of SCP, novel data normalization approaches must be implemented to separate the biological variation of interest from other confounding factors. This requires the assessment of such confounders, both biological and technical in nature, such as variation in digestion, matrix-associated ionization suppression and injection accuracy, cell size and morphology, and cell cycle. To facilitate the development of different data normalization strategies and assess their performance, we propose the generation of empirical datasets containing several levels of ground truth data. This involves the use of DIGESTIF protein spike-ins to assess and correct for technical variation, as well as a Small Molecule-Assisted Shutoff (SMASh)-tag system to generate proteome-based readouts for cell size across a panel of morphologically distinct cell lines in vitro. In conjunction with a Fluorescent Ubiquitination-based Cell Cycle Indicator (FUCCI) system, this allows to uncouple cell size from cell cycle effects and to individually assess their contributions on proteome complexity and data normalization. Such reference data are expected to help develop new quantitative normalization models for SCP data, in particular towards heterogenous datasets comprising multiple cell types. 257

258

P-I-0111 Regulation of Keratin 8 Phosphorylation during cancer cell division Ceyda Seren Ceyhan1, Ali Yurtseven1, Aybaran Olca Kebabcı1, Büşra Aytül Akarlar1, Nurhan Özlü1 1Koç University, Molecular Biology and Genetics, Istanbul, Turkey Keratins, distributed within epithelial cells, play a crucial role in providing tissue strength and are utilized as diagnostic markers in tumor pathology due to their varied expression patterns. Despite their robust structure, their regulation during dynamic processes such as cell division, particularly cytokinesis, remains poorly understood. Phosphorylation has been identified as a mechanism to regulate Keratin solubility and phenotype. Our study focuses on Keratin 8, aiming to elucidate its plasticity during cancer cell division through an investigation of its phosphorylation profile. Methodologically, cancer cell lines including MCF7, HeLa.S3, MDA-MB-231, and CAKI-2 are arrested at different cell cycle stages (interphase, mitosis, and cytokinesis). Phosphorylation profiles are examined using stable isotope dimethyl labeling-based quantitative phosphoproteomics, with samples from various cell cycle stages combined for each cell line. To simplify analysis, labeled peptides are fractionated via Strong Cation Exchange Chromatography and low-abundance phospho-peptides are enriched using TiO2 beads. LC-MS/MS analysis is then performed on the enriched phospho-peptides, with raw data processed using PD 2.3, MaxQuant and Perseus software. To comprehend Keratin 8 phosphorylation dynamics during cancer cell division, triplex dimethyl labeling is conducted, and cell cycle stage-specific phospho-sites are analyzed for each cell line. Inter-cell type comparisons reveal commonly regulated phospho-sites in the head domain of Keratin 8 across all cell lines. Intra-cellular phosphorylation analysis indicates that MCF7 cells exhibit the highest number of phospho-sites, correlating with elevated Keratin 8 expression levels. The number of phospho-sites decreases as Keratin 8 expression levels decline from epithelial to epithelial-mesenchymal cells. Our findings uncover upregulated phospho-sites of Keratin 8 at specific cell cycle stages, including both commonly regulated and cell-type unique phospho-sites. These findings present potential marker candidates for cell division and tumor diagnosis. 258

259

P-I-0112 Single cell proteomics analysis: a way-to-do proposal Blandine Chazarin1, Soulaimane Aboulouard1, Oumaima Hamhoum1, Maheul Ploton1, Anna Danchielli1, Isabelle Fournier1, Michel Salzet1 1PRISM, U1192 Inserm, Univ. Lille, Villeneuve d'Ascq, France Single Cell Proteomics (SCP) is an emerging field that labs start to implement. It allows to reach a unique level of information using mass spectrometry. The identification and the quantification of the proteome of single cells could lead to the characterization of cell sub-populations with specific proteome signatures and associated regulations. This will participate to develop new strategies for personalized medicine approaches. However, implementation of SCP remains challenging and the "way-to-do" is unclear. Here, we propose a demonstration for the implementation of various technologies. In this study, HeLa digest was used as standard sample with a dilution range allowing the analysis of 20ng down to 0.05ng. This standard was used for the development of a method on a NanoAcquity liquid chromatography (Waters) coupled with Eclipse mass spectrometer (Thermo). The tests included the change of the analytical column from 25cm to 10cm length, the removal of the pre-column, the gradient optimization (total run time of 30min vs 15min), the addition of FAIMS technology and the evaluation of software as MaxQuant and Proteome Discoverer and DIA-NN. Finally, we propose an easy to implement method on the CellenONE (Scienion) to significantly improved the rate of isolation of viable cells, on MCF7 cells for which apoptosis was enhanced. This project aims to provide guidelines for the implementation of SCP analysis using broadly used technologies. Publications demonstrated setups specifically developed for SCP analysis, but none of the presented studies highlighted the gain of each of the technologies as a whole [1], [2]. Furthermore, not all the documented technologies [3], [4] are available broadly, limiting the implementation in labs. In this study, demonstrations will be obtained on the gain in analysis throughput using shorter column, the improved depth of analysis with FAIMS technology, for example. Additionally, a method for the capture of viable cells at a higher rate and with a non- toxic and simple method on the CellenONE platform, was demonstrated on MCF7 cells. [1] Cong et al., Chem. Sci., 2021 [2] Furtwängler et al., Mol. Cell. Proteomics, 2022 [3] Kreimer et al., Anal. Chem., 2023 [4] Mund et al., Nat. Biotechnol., 2022 259

260

P-I-0113 Benchmarking TimsTOF HT and Orbitrap eclipse for deep proteome analysis using DDA and DIA workflows Weiqiang Chen1, Markus Hartl1 1University of Vienna, Mass Spectrometry Facility, Max Perutz Labs, Vienna, Austria Introduction: Bottom-up proteomics is rapidly evolving with new instruments and methods demanding continuous evaluation. This study benchmarks the TimsTOF HT and the Orbitrap Eclipse mass spectrometer using three-proteome mixtures (yeast, E. coli, human) in known ratios analyzed with both data-dependent acquisition (DDA) and data-independent acquisition (DIA) workflows. Methods: Mixtures of the three proteomes (total protein loads: 0-600 ng) were separated on ionopticks Aurora Ultimate columns and analyzed on TimsTOF HT and Obitrap Eclipse in DDA and DIA modes. DDA data was processed with FragPipe, while DIA data utilized Biognosys™ Spectronaut™. Custom Python scripts (MsReport) were employed for computational analysis. Results: Analysis of three replicates per sample mixture in DIA mode on TimsTOF HT identified over 12,000 protein groups (7,800+ human, 2,000+ yeast, 700+ E. coli) and 130,000+ unique peptides, while over 11,000 protein groups were identified with DDA on TimsTOF HT. Notably, >80% of quantified proteins had a coefficient of variation (CV) below 10%, with a median CV of less than 5%. Experimental protein ratios are closely centered around the theoretical ratios. Very few E. coli proteins were quantified in the 0 ng E. coli sample, indicating a low false discovery rate (around 0.4%). DIA identified up to 19% more protein groups than DDA, comprising mostly low-abundant proteins. Conclusions: The established workflows offer deep proteome coverage with high accuracy and low false discovery rates, demonstrating their readiness for application to complex biological samples. 260

261

P-I-0114 Ultra-high throughput workflow for robust and sensitive quantitative single cell analysis Bernard Delanghe1, Julia Kraegenbring1, Jana Richter1, Tabiwang Arrey1, Fernanda Salvato2, David Hartlmayr3, Anjali Seth3, Nicolaie Eugen Damoc1, Tonya Pekar Hart2, Thomas Moehring1 1Thermo Fisher Scientific, Bremen, Germany 2Thermo Fisher Scientific, San Jose, CA, United States 3Cellenion SASU, Lyon, France Introduction: Thousands of proteins can be measured from different cell lines using MS based proteomics. However, to understand and characterize cell heterogeneity, these cells need to be studied individually. In recent years single cell analysis has profited from advances in LC-MS based proteomics approaches. Nevertheless, there are still challenges in this field of application. Besides sample preparation, the key challenges in looking at individual single cell proteomes are sensitivity, coverage, dynamic range, and throughput. To address some of these challenges, new technological developments, as well as improvements on existing LC-MS- based proteomics workflows are a necessity. Here we described a high-throughput workflow of TMTpro multiplexed single cells using an Orbitrap Astral mass spectrometer. Methods: Individual HeLa cells were sorted, followed by reduction, alkylation, trypsin digestion and TMT labeling using CellenONE as per the manufacturer"s protocols. To enhance the MS1 signal of the peptides a booster channel was used. The multiplexed single cell digests were analyzed using an Orbitrap Astral MS with FAIMS Pro Duo interface coupled to a Vanquish Neo UHPLC system. Data was acquired in DDA mode and searched with Proteome Discoverer Software 3.1. Results: Using optimized methods for analyzing 250 Samples Per Day on average more than 3000 proteins and 13.000 peptides were quantified. 80% of those proteins had less than 10% CV. 261

262

P-I-0115 State-of-the-art glycopeptide identification using an Orbitral(TM)-Astral(TM) mass spectrometer Matthew James Garland1, Tabiwang Arrey1, Nicolaie Eugen Damoc1, Eduard Denisov1, Bernd Hagedorn1, Christian Hoch1, Teeradon Phlairaharn1, Hamish Stewart1 1Thermo Fisher Scientific, Life Sciences Mass Spectrometry (LSMS), Bremen, Germany Introduction: Glycoproteomics leverages powerful new tandem mass spectrometry (MS/MS) tools for exploring glycosylation at a granular level. While improvements in instrumentation, computational tools, and sample preparation methodologies have made glycoproteomics attainable by many labs, consistently detecting glycopeptides in clinical samples such as human plasma remains a challenging endeavor due to its complexity. One way to improve the ability to detect glycopeptides in human plasma is to use a redefined electron- transfer/higher-energy collision dissociation (ET/HCD) to fragment peptides and glycan in the same spectrum. Recent reports have suggested that using two different collision energies (CE) improves glycopeptide detection in complex samples such as human plasma. In this work, we sought to demonstrate that an Orbitrap(TM) Astral(TM) mass spectrometer is a suitable platform for glycoproteomics analysis Methods: We configured an Orbitrap Astral mass spectrometer to collect data-independent acquisition (DIA) MS/MS from 500 ng human plasma using two different normalized collision energies known as "stepped collision energies" per precursor,. The raw files from this study were processed using MSFragger Glyco. Preliminary Data: By analyzing enriched plasma samples on an Orbitrap Astral mass spectrometer coupled with the Vanquish(TM) Neo UHPLC system, we can identify 1774 unique glycopeptides from 5988 glycopeptide- spectrum matches with two CEs and 1073 unique glycopeptides from 4016 glycopeptide-spectrum matches with single CE. This study demonstrates that using stepped collision energy in an HCD device at a relatively slow scan rate of 75 Hz, but coupled with the fast and accurate Orbitrap Astral allows highly sensitive, state-of-the-art identification of glycopeptides without a fully optimized routine. These results pave the way for even faster stepped CE scans with more optimized routines to provide ever-improved sensitivity of glycopeptide identification in the field of glycoproteomics. 262

263

P-I-0116 Evaluation of midia-PASEF acquisition mode for the analysis of phosphoproteome samples Ute Distler1, Mateusz Krzysztof Łacki1, Michał Piotr Startek2, David Teschner3, Tanja Ziesmann1, Stephanie Kaspar- Schoenefeld4, Jens Decker4, Florian Krohs4, Jonathan Krieger5, Oliver Raether4, Stefan Tenzer1,6,7 1University Medical Center Mainz, Institute for Immunology and Research Center for Immunotherapy (FZI), Mainz, Germany 2University of Warsaw, Faculty of Mathematics, Informatics, and Mechanics, Warsaw, Poland 3Johannes-Gutenberg University Mainz, Institute for Informatics, Mainz, Germany 4Bruker Daltonics GmbH & Co. KG, Bremen, Germany 5Bruker Ltd, Milton, Canada 6Helmholtz Institute for Translational Oncology (HI-TRON), Mainz, Germany 7Deutsches Krebsforschungszentrum (DKFZ), Heidelberg, Germany Background: The recently introduced DIA acquisition scheme, midia-PASEF, integrates the concepts and benefits of dia-PASEF, scanning quadrupole and overlapping quadrupole window acquisition. Using an overlapping ion mobility-encoded quadrupole window scheme facilitates fragment-precursor ion assignment and increases duty cycle as compared to fixed window schemes. midia-PASEF acquisition improves the precision of precursor- fragment relationships to <2 Th and resulting deconvoluted MIDIA-MS/MS fragment ion spectra display DDA-like quality with fragment ion mass accuracies below 10 ppm. In the present work, we conducted an in-depth evaluation of the midia-PASEF workflow for the analysis of phosphoproteomic samples. Methods: Mouse brains were lysed and tryptically digested. Phosphopeptides were enriched using TiO2 tips (GL Sciences, Japan) according to the manufacturer´s protocol. Phosphopeptides were analysed by LC-MS on mass spectrometers from the Bruker timsTOF series coupled to nanoElute LC systems (Bruker Corporation). Samples were separated on a reversed phase C18 column (IonOpticks, 25 cm x 75 μm 1.7 μm). Data were acquired using parallel accumulation serial fragmentation (PASEF) enhanced data-dependent (DDA) and data-independent (DIA) methods such as dia-PASEF, as well as midia-PASEF applying different isolation schemes for precursor ion selection. Data were analysed applying different software tools including PEAKS, Spectronaut and our in-house developed pipeline for midia-PASEF data processing. Results: We evaluated and compared different midia-PASEF acquisition schemes applying different quadrupole isolation window widths (36 Th, 24 Th, 20 Th). In the original midia-PASEF method, quadrupole isolation window positions are shifted 12 Th as compared to the previous frame during acquisition resulting in an overlap of 24 Th between two neighboring frames. Here, we additionally tested the impact of changing the window overlap to 12 Th and 10 Th. The performance of the different midia-PASEF methods was compared to PASEF and dia-PASEF applying different tools for data processing. We could demonstrate that midia-PASEF provides, upon deconvolution, DDA-like MS/MS spectra that can be analyzed using either algorithms tailored for DDA data processing or our in-house developed pipeline, which has been optimized for the analysis of midia-PASEF data. Best results could be obtained applying a 36x24 Th window scheme achieving higher and similar phoshoproteome coverage as compared to dia-PASEF and dda-PASEF, respectively. Conclusions: Evaluation of midia-PASEF for the analysis of phosphoproteomic samples revealed best results for a 36x24 Th-window scheme and subsequent processing using our in-house developed midia-PASEF processing pipeline, midiaID, combined with MS2 rescore. 263

264

P-I-0117 Enhancing whole-proteome quantification in mass spectrometry-based proteomics with multi-point internal calibration curves Zahra Elhamraoui1,2, Cristina Chiva1,2, Julia Morales-Sanfrutos1,2, Olga Pastor1,2, Eduard Sabidó1,2 1Universitat Pompeu Fabra, Barcelona, Spain 2Centre for Genomic Regulation, Barcelona, Spain Mass spectrometry-based proteomics is well-known for its ability to quantify proteins with high accuracy and precision. External calibration curves and single-point calibration methods are often employed in targeted proteomics for protein quantification. These methods involve using external curves to determine the range of linearity, or adding isotopically labeled internal standards to samples to derive analyte concentration. However, they have limitations because the range of linearity is not directly determined within the sample. This issue is particularly significant when dealing with large patient cohorts. To address this challenge, a few years ago we developed the Isotopologue Multiple-point Calibration (ImCal) strategy. ImCal uses a mixture of isotopically labeled peptides at different concentrations to establish a multi-point internal calibration curve, ensuring precise and accurate quantification of specific peptides of interest in targeted proteomics applications. Unfortunately, the use of multiple isotopologue peptides in ImCal increases the cost of each targeted assay and limits the number of analytes measured due to cycle time constraints. We extended the ImCal approach to the whole proteome by developing an inexpensive TMT-based multi-point internal calibration curve strategy (TMTCal) created from serial dilutions of total protein extracts. These dilutions are used to generate internal calibration curves within a single experiment for all peptides identified in a proteome. For this purpose, human ovarian cancer cells (SK-OV-3) were cultured in triplicate, with and without cis- platinum, and subsequently digested with trypsin. Concurrently, a pooled sample was created to generate several dilutions and create an internal multi-point calibration curve. Tandem mass tags (TMT-11) were used to label the samples, which were then mixed. Following basic pH reversed-phase fractionation, samples were analyzed using a 90-min gradient in an Orbitrap Eclipse mass spectrometer and a Real-Time Search MS3 method (RTS-MS3). Acquired data were processed using Proteome Discoverer (v2.4). Using the TMTCal, we observed that the majority of identified peptides demonstrated high quantitative response, with most exhibiting complete or nearly complete linear regression dilution curves (R²>0.95). Furthermore, we confirmed that the vast majority of peptides identified in the human ovarian cancer cell samples fell within the linear range of quantitation, ensuring accurate and precise relative quantification using either label-free direct comparison or by utilizing abundances derived from the individualized regression lines for each peptide. In conclusion, the TMTCal technique allows for a more comprehensive analysis of peptide quantitative behavior within the human proteome and, by extending the concept of multi-point internal calibration curves to the entire proteome, enables enhanced accuracy and precision in proteome-wide relative quantification. 264

265

P-I-0118 Protein-Centric antibody repertoire profiling on a timsTOF: enhanced sensitivity using minimal sample volume Jan Fiala1,2, Dina Schuster1,2, Simon Ollivier1,2, Stuart Pengelley3, Markus Lubeck3, Florian Busch4, Andris Jankevics1,2, Oliver Raether3, Jean-Francois Greisch4, Albert J. R. Heck1,2 1Utrecht University, Biomolecular Mass Spectrometry & Proteomics, Bijvoet Center for Biomolecular Research & Utrecht Institute for Pharmaceutical Sciences, Utrecht, Netherlands 2Netherlands Proteomics Center, Utrecht, Netherlands 3Bruker Daltonics GmbH & Co. KG, Bremen, Germany 4Bruker Switzerland AG, Faellanden, Switzerland Introduction: Endogenous antibodies, or immunoglobulins (Igs), are pivotal for the immune system's ability to identify and neutralize foreign threats. Despite their abundance in body fluids, the enormous variability in sequence and concentration of these molecules presents significant analytical challenges. Recent advances in protein-centric analysis utilizing Orbitrap-based systems have facilitated the detailed profiling of human antibody repertoires by analyzing fragment antigen-binding (Fab) regions. However, effective Fab profiling requires optimized separation techniques and rapid mass spectrometric analysis to differentiate hundreds of Fabs with nearly identical masses but potentially diverse biochemical properties. Here, we demonstrate for the first time that timsTOF mass spectrometer coupled to nanoLC can be effectively employed to analyze antibody repertoires from minimal amounts of serum. Methods: IgG antibodies were affinity-purified using CaptureSelect FcXL resins and digested with the IdgE hinge- cleaving protease. This process released the Fabs from the remaining fragment crystallizable (Fc) region attached to the bead. Prior to Fab profiling, LC runs with 7 recombinant monoclonal Fab mixture were performed to compare the sensitivity of UHPLC system versus the nanoLC, whereby the nanoLC provided approx. 50-fold increase in intensity. Further, Fab fragments from serum samples were profiled on both MS platforms with identical nanoLC and separation methods. LC-MS data were analyzed by Thermo BioPharma Finder or Bruker DataAnalysis with in-house implemented sliding window algorithm, followed by MaxEnt deconvolution. Results: Advancement in the analysis and profiling of the circulating antibody population in human serum samples are presented, marking the first-time use of a Bruker timsTOF HT mass spectrometer. It was observed that injections of 1-2 µg into the nanoLC-timsTOF system induced a noise-like signal deteriorating phenomenon in the 1000−2000 m/z range. Consequently, sample amounts were reduced (~56-110ng), resulting in a notable sensitivity enhancement, leading to the detection of hundreds of distinct IgG1 clones. Notably, the timsTOF HT data yielded >1.5 times more identified clones compared to the Orbitrap Eclipse. This increase was primarily due to the identification of clones with abundance lower than 1%. On the other hand, the 100 most abundant clones from both platforms demonstrated high identity overlap, reinforcing confidence in our proposed workflow. Our data reveal that the timsTOF mass spectrometer can effectively track antibody repertoires, matching the quality of prior workflows. Its high sensitivity allows for detailed Fab profiling from just 50 nL of serum, which is a substantial decrease in sample consumption from the 1-2 µL typically required in previous studies. 265

267

P-I-0119 Stability DIF-FRAC: a systematic characterization of protein complex stability Samantha Fischer1, Anna Mallam2, Matthew O'Meara3, Kevin Drew1 1University of Illinois at Chicago, Biological Sciences, Chicago, IL, United States 2University of Texas at Austin, Molecular Biosciences, Austin, TX, United States 3University of Michigan at Ann Arbor, Computational Medicine and Bioinformatics, Ann Arbor, MI, United States Macromolecular protein assemblies are responsible for most functions within cells, including critical processes like protein degradation, protein folding, and vesicle trafficking, exemplified by structures such as the 26S proteasome, the prefoldin complex, and the commander complex. To fulfill their functions within the cell, protein complexes need to be stably assembled into three-dimensional architectures. However, the stability of each protein complex, the physicochemical factors contributing to this stability, and the relationship between stability and function are not fully understood. To bridge this knowledge gap, we introduced Stability-DIFFRAC, a high-throughput, label-free proteomic methodology designed to evaluate the stability of human protein complexes. This approach leverages biochemical denaturation (e.g., urea), HPLC chromatographic separation, mass spectrometry, and a modeling workflow to systematically analyze the stabilities of thousands of protein complexes and subcomplexes. This analysis identifies over 7,000 unique proteins and 4,500 potential protein complexes (and sub-complexes), when compared to the 7,000 protein complexes of the Hu.MAP 2.0 resource. We observe destabilization globally and at the level of individual complexes, revealing distinct stability behaviors both between and within complexes, such as the higher stability of the proteasome 20S core compared to the greater 26S proteasome holoenzyme and the commander complex. Illustrating the connection between stability and function, this analysis reveals that several vesicle-trafficking complexes exhibit low stability. In contrast, complexes such as the 20S proteasome core and prefoldin complex display high stability behavior, correlating with their critical roles in protein homeostasis. Assessment of complex stability is essential for deriving general principles of biophysical properties which confer stability and lead to protein complex self-assembly. The insights derived from our work are expected to contribute to the development of more stable and efficient therapeutics and to the understanding of disease mechanisms linked to the disruption or misfolding of stable protein complexes. 267

268

P-I-0120 Expanding TMTpro reagents to 32-plex for high-throughput quantitative proteomics on Orbitrap platforms Dustin Frost1, Joao Paulo2, Steven Gygi2, Karsten Kuhn3, Ian Pike3, Ryan Bomgarden1 1Thermo Fisher Scientific, Rockford, IL, United States 2Harvard Medical School, Boston, MA, United States 3Proteome Sciences, London, United Kingdom TMTpro reagents are popular reagents for highly multiplexed mass spectrometry (MS)-based proteomics, enabling relative quantification of many samples in a single LC-MS/MS acquisition. 18-plex TMTpro reagents have been previously demonstrated to provide accurate and precise quantification of proteins and peptides by high-resolution LC-MS/MS analysis on Orbitrap platforms. Each tag in the multiplex set incorporates nine 13C & 15N stable isotopes at unique positions within the structure to enable generation of reporter ions during MS/MS fragmentation with distinct masses that differ by a minimum of 6 mDa. To further increase multiplexing capacity, we designed 17 structurally identical isotopologues that substitute one 2H stable isotope into the reporter group to impart a 3 mDa mass difference between the new variants and the existing 18-plex tags. We strategically located this deuterium isotope on the isobutyl group proximal to the nitrogen atom on the reporter to minimize chromatographic shifts and for synthetic tractability using the existing synthesis route and isotopic starting materials. The deuterated TMTpro reagents were synthesized using existing manufacturing methods and chemical purity & isotopic incorporation were verified by direct infusion MS. BSA digest and HeLa digest samples labeled with deuterated and non-deuterated TMTpro reagents were subjected to LC-MS/MS analysis using a Vanquish Neo UPLC with a 50cm EASY-Spray HPLC column interfaced with an Orbitrap Eclipse MS system. MS2and RTS SPS-MS3 acquisitions were performed at various eFT and TurboTMT resolving powers. Data was processed with Proteome Discoverer to determine labeling efficiency and numbers of identified & quantified peptides & proteins. BSA digest labeled with a subset of tags was analyzed using a targeted MS/MS acquisition method to evaluate chromatographic shifts between peptides labeled with the deuterated and non-deuterated tags and to determine the impact on quantitative metrics. HeLa digest samples were labeled with 32-plex TMTpro reagents to determine appropriate instrument parameters, evaluate peptide identification rates, and assess quantitative performance. An Orbitrap resolving power (RP) of ≥90K was sufficient to baseline resolve the reporter ions differing in mass by 3 mDa in MS/MS spectra, while TurboTMT acquisition at RP 45K permitted distinguishing reporter ions with greater instrument duty cycle efficiency to realize greater proteomic depth. Our preliminary LC-MS/MS data using 18-plex and 32-plex TMTpro reagents demonstrate comparable labeling efficiency, peptide identification rates, and quantitative performance with fewer missing values for the expanded reagent set. 268

269

P-I-0121 Increasing throughput while maintaining coverage depths in single cell proteomics using the timsTOF Ultra Shourjo Ghose1, Christoph Krisp1, David Hartlmayr1, Verena Tellestroem2 1Bruker Scientific LLC, Proteomics, Billerica, MA, United States 2Bruker Daltonics GmbH & Co. KG, Bremen, Germany Introduction: For single cell proteome analysis, throughput while maintaining coverage depths is one of the bottlenecks in the field. Therefore, fast scanning, ultra-high sensitivity mass spectrometry is a key to reach a proteome coverage necessary for understanding the cellular heterogeneity on a cell-by-cell level across large sample cohorts. The latest enhancements in 4D-proteomicsTM on the timsTOF platform with high ion intake, a higher-pressure segment for more effective ion collection and two orthogonal deflections, to maintain robustness, pushes the limits of detection to the single cell level. Combined with automated single cell isolation and sample preparation on the cellenONE® platform for protein-loss reduced preparation with direct sample pickup from the label-free proteoCHIP® enables deep proteome coverage and high reproducibility. Methods: HeLa cells at different cell counts were isolated into the LF48 proteoCHIP®, directly lysed and digested using the cellenONE platform. Samples were either dried down in the chip or manually transferred into 96-well plates. Peripheral blood mononuclear cells (PBMCs) were FACS sorted into T-Cells (CD4+, CD8+) B-Cells (CD19+) and monocytes (CD14+), isolated with the cellenONE and prepared as the HeLa cells above. The LF48 proteoCHIP was placed into the nanoElute® 2 autosampler. Sample were picked up with the dissolve sample function and injection onto a 5 cm Aurora Rapid 75 or a 25 cm Aurora Ultimate C18 column (IonOpticks) and eluted into a timsTOF Ultra at 80 or 32 SPD. dia-PASEF® data were processed with Spectronaut 18 (Biognosys) using directDIA+. Preliminary Data: A K562 protein extract digests (Promega) dilution series were run on both column setups. Processing of the dia-PASEF data with Spectronaut 18 demonstrated good quantitative reproducibility for individual concentration loads. For loads of les than 500 pg the 80 SPD (10 min active gradient), and 32 SPD (22 min active gradient. Showed same protein coverage depths only for the high loads > 2 ng, the 32 SPD setup outperformed the faster 80SPD setup. Therefore, the 80 SPD setup was used for the following experiments. Sample pickup directly from the label-free proteoCHIP with the dissolve sample function of the nanoElute® 2 was compared to proteome depths obtainable with samples manually transferred to 96 well plates. Both setups performed equally well delivering a protein depth of about 5,000 proteins per single cell and increasing to >7,000 proteins for 20 cells. The PBMC analysis Workflow from FACS sorted T-Cells (CD4+, CD8+) B-Cells (CD19+) and monocytes (CD14+) in the 80 SPD setup processed with Spectronaut 18 in directDIA+ identifying over 2,000 protein. groups across the different cell types with good group identification rate and quantitation reproducibility. Clear differences were observed, distinguishing the two T-Cells classes from each other and also from the B-Cells and monocytes. 269

270

P-I-0122 Measuring proteome of 1,000 neutrophils directly isolated from inflammatory sites Susmita Ghosh1, Ali Ata Tuz Tuz2, Martin Stenzel3, Vikramjeet Singh2, Mathis Richter4, Oliver Soehnlein4, Emanuel Lange3, Robert Heyer3,5, Zülal Cibir2, Alexander Beer2, Marcel Jung2, Dennis Nagel2, Dirk M. Hermann6, Anja Hasenberg2, Anika Grüneboom3, Albert Sickmann3,7,8, Matthias Gunzer2,3 1Leibniz-Institut für Analytische Wissenschaften – ISAS – e.V., Biofluorescence, Dortmund, Germany 2Institute for Experimental Immunology and Imaging, University Hospital, University of Duisburg-Essen, Essen, Germany 3Leibniz-Institut für Analytische Wissenschaften – ISAS – e.V., Dortmund, Germany 4Institute for Experimental Pathology, University of Münster, Muenster, Germany 5Bielefeld University, Multidimensional Omics Analyses Group, Faculty of Technology, Bielefeld, Germany 6University Hopital Essen, University of Duisburg-Essen, Department of Neurology, Essen, Germany 7Ruhr-Universität Bochum, Medizinisches Proteom-Center, Bochum, Germany 8University of Aberdeen, Department of Chemistry, Aberdeen, United Kingdom Neutrophils are typically the earliest and most frequent inflammatory cells in sterile inflammation. However, to achieve a thorough comprehension of their involvement in pathophysiology, neutrophils must be characterized molecularly either as isolated cells or remaining on individual inflammatory site. While these cells have been extensively studied in different inflammatory situations using transcriptome analyses, this does not fully represent their molecular state due to their post-mitotic nature. High-throughput proteomic technologies can fill the gap by providing a comprehensive insight into their structural and functional architecture. The currently available approach for neutrophil proteomics analyzed millions of cells, which limits the analysis of scant population of these cells at individual inflammatory site, especially in case of chronic inflammation or steady- state infiltration. Although recent proteomics approaches are sensitive enough to analyze very low input samples, even at single-cell resolution, their application in understanding neutrophils (ten times lesser cell volume than HeLa cells) remains unexplored, presenting a necessity for further investigation. We here present a pipeline for the proteome analysis of down to 1,000 human (~60ng) and mouse (~40ng) neutrophils isolated directly from inflammatory sites. This approach enabled identification of ~4800 and ~3400 proteins from 1,000 isolated circulating murine and human neutrophils with high reproducibility and 102-108 copies/cell. We also provide a searchable database for human and murine protein copy numbers, a few of which were also validated with quantitative flow cytometry approach. Our analysis demonstrated the activation of circulatory neutrophils upon stroke induction compared to sham controls and naive mice. These proteome changes were also observed in circulating neutrophils isolated from human patients compared to healthy controls. Further we found that post-stroke brain-infiltrated neutrophils exhibited a strongly distinct proteome compared to circulatory neutrophils in the same host. Brain-infiltrated neutrophils showed a metabolic switch towards increased mitochondrial activity and ROS-production to adapt glucose-deprived condition. Our approach was also efficient to demonstrate a functional shift of neutrophils transmigrated to oral cavity upon transient oral inflammation. Overall, this method represents a significant leap in intricate neutrophil proteome characterization, even from minute samples, and holds great promise for expanding our understanding of neutrophil biology in health and disease. We further envisage to investigate neutrophil at single-cell resolution to reveal the heterogeneity and functional diversity of these intriguing immune cells. 270

271

P-I-0123 Ion mobility free high throughput peptide collisional cross-section measurements in fourier transform mass spectrometer on the proteome wide scale Ulises H. Guzman1, Konstantin Aizikov2, Kyle L. Fort2, Pedro Navarro2, Martin Rykaer1, Jeppe Madsen1, Alexander Makarov2, Jesper Olsen1 1Novo Nordisk Foundation Center for Protein Research, Proteomics Department, Copenhagen, Denmark 2Thermo Fisher Scientific, Bremen, Germany The integration of Ion Mobility (IM) devices with mass spectrometers has facilitated the precise determination of peptide collisional cross sections (CCS), introducing an additional dimension to Mass Spectrometry (MS)-based investigations. While the utilization of transient decay rate in Fourier Transform Mass Spectrometry (FTMS) for assessing CCS values has been reported for Orbitrap™ and Ion Cyclotron Resonance (ICR) MS systems, this study presents a novel method that enables real time highly reproducible and accurate measurement of CCS using FTMS in high-throughput LC-MS studies on entire proteome scale. The key innovation of this approach lies in its ability to infer peptide CCS values directly from the FTMS spectra, without the need for a dedicated IM device, by relating the observed spectral peak resolution to the rate of collisions experienced by ions during the detection determined by their CCS. Experiments were conducted on the research grade Thermo Scientific™ Orbitrap Exploris™ 480 mass spectrometer using HeLa cell tryptic and Lys-C lysates, as well as chemically modified peptides (methylation, acetylation, and phosphorylation). The inferred CCS values showed strong correlation with the IM-MS values reported in the literature. The reproducibility of the CCS measurements in the shotgun proteomics experiments was on par with the values obtained using dedicated IM devices. Our findings illustrate the utility of IM-free FTMS based CCS in the analysis of peptide secondary structures and Post Translational Modifications (PTMs), revealing that the conformation of peptides is likely determined by the balance between hydrophobic interactions and charge repulsion. Modifications such as phosphorylation, methylation, and acetylation were found to impact peptide CCS values, likely due to increased hydrophobicity and loss of charge. This novel FTMS-based method for CCS determination offers a high-throughput and widely accessible solution for integrating CCS information into standard proteomics workflows without relying on the dedicated IM devices. 271

272

P-I-0124 Unveiling hidden depths: A high-throughput plasma proteomics workflow on Orbitrap astral mass spectrometer for enhanced biomarker discovery Kevin Yang1, Nicholas Mucci2, Evangelina Bahu2, Andrea Cerda2, Lee Cantrell2, Khatereh Motamedchaboki1, Stephanie Samra1, Amirmansoor Hakimi1 1Thermo Fisher Scientific, San Jose, CA, United States 2Seer, Redwood City, CA, United States Introduction: Cancer is the second leading cause of death, accounting for millions of deaths annually. Late diagnosis hinders prognoses and highlights the urgency for improved early detection tools. Existing diagnostic tests lack sensitivity or require invasive biopsies. While analyzing blood plasma for cancer biomarkers holds promise, complex workflows involved limit progress. We address these challenges by developing a high- throughput plasma proteomics workflow on the cutting-edge Thermo Scientific™ Orbitrap™ Astral™ mass spectrometer, which yields the deepest plasma proteome coverage yet. We have evaluated the performance of this workflow in a mini cohort of diverse cancer and non-cancer samples. This paves the way for rapid cancer cohort and translational research studies, offering the depth needed for early disease detection through comprehensive plasma analysis. Methods: Plasma samples were processed with a standard plasma proteomics workflow using the Proteograph XT Assay from Seer or top 14 High Select™ Depletion Spin Columns from Thermo Fisher Scientific. Resulting peptides were then analyzed using a robust liquid chromatography (LC) method with EASY-Spray™ PepMap™ Neo columns on a Vanquish™ Neo UHPLC system coupled to the new Orbitrap Astral mass spectrometer in Data Independent Acquisition (DIA) mode with few variations of LC-MS methods which provided a variety of high- throughput options to enable translational researchers to choose a workflow optimized for their cohort size and preferred turnaround time. The LC-MS data analysis was done using Thermo Scientific™ Proteome Discoverer software 3.1 and Seer"s Proteograph Analysis Suite. Preliminary Data: Our preliminary results show over 700 protein groups identified in neat plasma with the standard high-throughput proteomics workflows through LFQ-DIA analysis on an Orbitrap Astral mass spectrometer. The top 14 depletion column enables identification of 1500 protein groups, which doubled the plasma proteome coverage without sacrificing the throughput, showing an economical method to capture subtle disease biomarkers. When coupled with the Proteograph XT Assay sample processing, we were able to identify 7,000 protein groups, which represents a 10-fold increase in proteome coverage. We also evaluated the quantitation accuracy by mixing different quantities of human and chicken plasma. The results showed that we were able to obtain a ratio that is close to the expected ratio, highlighting the quantitation accuracy from our plasma proteomics workflow. Together, the novel DIA LC-MS workflow represents a powerful combination of scale and depth of coverage, empowering the development of improved classification models and enhancing biomarker discovery through robust and larger statistically powered early-stage biomarker discovery studies. 272

273

P-I-0125 Identifying drug targets with thermal proteome profiling using IBT-16plex Piliang Hao1, Zhaomei Shi1, Shuwei Li2 1ShanghaiTech University, Shanghai, China 2Nanjing Apollomics Biotech Inc., Nanjing, China Thermal proteome profiling (TPP) has been widely applied in the identification of drug targets for several years, and TMTpro-16plex has recently been evaluated for TPP of vehicle and drug treated samples in a single labeling process to reduce missing values and save instrument time. A novel isobaric labeling reagent, IBT-16plex, was just developed with slightly better performance in protein identification and quantification than the commercially available TMTpro-16plex. In this study, we applied the newly developed IBT-16plex into the target identification of methotrexate and panobinostat using TPP. The known targets of these two drugs were successfully identified with elevated melting temperatures, and some known off-targets and potential new off- targets were also identified. It indicates that IBT-16plex can be a cost-effective replacement of TMTpro-16plex for TPP applications. 273

274

P-I-0126 Enhancing biomarker discovery: Automatable, enrichment-based plasma pipeline allows rapid and deep analysis of large sample cohorts Katrin Hartinger1, Leander Runtsch1, Xaver Wurzenberger1, Godfred Boateng1, Claudia Martelli2, Irina Maerki2, Roland Bruderer3, Nils Kulak1 1PreOmics GmbH, Planegg, Germany 2Bruker Switzerland AG, Faellanden, Switzerland 3Biognosys AG, Zurich, Switzerland Introduction: Blood plasma represents an invaluable source for proteomic biomarker discovery as it is easily accessible and provides comprehensive information about an individual´s health status. However, the high dynamic range as well as sample heterogeneity and complexity pose significant challenges for LC-MS-based proteomics, limiting access to the entire proteome. Moreover, common plasma workflows require tedious manual processing and are often incompatible with automation or the analysis of large cohorts. With ENRICHplus, we present a fully automatable plasma workflow that allows parallel processing of 96 samples in 5 hrs and enhances proteome coverage for human plasma by up to 7-fold compared to neat sample preparation. Combined with state-of-the-art data acquisition and analysis, it provides a high-throughput plasma platform for the analysis of large-scale proteomic studies. Methods: The ENRICHplus workflow utilizes a proprietary single particle solution for dynamic range compression by enriching low-abundance proteins from 20-80 µL of plasma onto paramagnetic beads. Upon enrichment, proteins were further processed according to the iST-BCT protocol. The full sample preparation workflow was done manually or fully automated in 96well format. Peptides were analyzed by a nanoLC coupled to a timsTOF HT instrument using diaPASEF® acquisition mode. Data processing was done with Spectronaut®. Preliminary results: For a human EDTA plasma pool, almost 4,700 protein groups were identified in a short 17- min LC-gradient, which corresponds to a 6.8-fold increase compared to neat samples and provides access to low- abundance proteins by covering a protein concentration range of 10 orders of magnitude. Moreover, superb precision was achieved with a CV of 11%. To further enhance the throughput of ENRICHplus, the sample preparation workflow was successfully automated and combined with 5-mins gradients, allowing rapid analysis of large sample cohorts. Preliminary results revealed over 2,100 protein groups identified and no decrease in precision compared to long gradients of 35-mins. To assess its potential for biomarker discovery, the ENRICHplus kit was applied to a small cohort of colorectal cancer patients (n=6) and matched healthy donors (n=6), resulting in more than 5,800 protein group identifications. Statistical analysis showed a clear stratification between healthy and sick donors, and the 13 proteins that showed substantial regulation are all known to be associated with cancer, including the serum amyloid A1 (SAA1), which is a hallmark of malignant disease progression in colorectal cancer. Conclusion: The presented plasma platform with the ENRICHplus technology enables rapid and deep analysis of large plasma sample cohorts, opening up new possibilities for high-throughput biomarker discovery. 274

275

P-I-0127 Optimized DIA-MS workflow for Host Cell Proteins (HCP) characterization and quantification in bioreactors and top-down mass spectrometry analysis for monoclonal antibody production Peter Hoffmann1, Janik Seidel1, Abdulafeez Akinloye1, Clifford Young1, Manuela Klingler-Hoffmann1, Leigh Donnellan1, Mark Condina1, Sri Ramarathinam2, Lee Xin Chong2, Alok Shah2 1University of South Australia, Clinical Health Sciences, Adelaide, Australia 2CSL Limited, Melbourne, Australia Biopharmaceutical proteins have revolutionized treatment of diseases and are one of the fastest growing segments of the pharmaceutical industry. Close monitoring of Critical Quality attributes (CQAs), - attributes of the product or impurities known to influence efficacy, safety or stability of the final product, - is necessary. Host cell proteins (HCPs) are especially challenging to monitor and to deplete trough purification. The industrial standard to monitor HCP are enzyme-linked immunosorbent assays (ELISA). These assays are only able to detect immunoreactive HCP species and only give a total quantity. In contrast mass spectrometry (MS) is a non- targeted approach and allows unbiased identification and quantification of HCPs population down to single HCPs. Here, we present an optimized workflow to monitor and quantify HCP using a DIA bottom-up MS approach. The developed method is capable of quantifying more than 1000 HCPs from harvest samples including 20 high risk proteins identified as being especially problematic for biopharmaceuticals, while having improved throughput due to reduction in processing and acquisition time through the implementation of a high temperature digest in combination with a micro-flow LC system and untargeted DIA acquisition on a SCIEX ZenoTOF 7600. The full workflow as well as unit operations were evaluated for robustness, sensitivity, repeatability and influences on the quantification results. We will also show top-down mass spectrometry analysis of the monoclonal antibody product and good fragmentation by electron activated dissociation (EAD)in order to look at post translational modifications in a shorter timeframe than with a bottom-up approach. 275

276

P-I-0128 Intensity-based site localization of post-translational modifications utilizing predicted peptide fragmentation spectra Alexander Hogrebe1, Michael Graber1, Daniel Zolg1, Markus Schneider1, Tobias Schmidt1, Vishal Sukumar1, Florian Seefried1, Siegfried Gessulat1, Martin Frejno1 1MSAID, Berlin, Germany Background: Modifications (PTMs), particularly phosphorylation, are commonplace in cell signaling studies, and precise PTM site localization is pivotal for understanding underlying biology. Current localization tools mainly rely on peak m/z matching, neglecting fragment ion intensity differences between modified peptide isomers. Leveraging our deep-learning framework, we accurately predict physicochemical properties of modified peptides, facilitating intensity-based localization of modification sites. We assess various scoring methods and compare our approach with existing tools. Methods: We curated a vast training dataset (>1.6 million spectra) of post-translationally modified peptides to develop a precise model for predicting their properties. Our Chimerys search algorithm employs these predictions to score peptide spectrum matches (PSMs) from experimental MS2 spectra, aiming to explain maximum MS2 intensity with minimal peptides. For localization, we predict spectra for all possible modification isomers simultaneously, calculating normalized scores and localization probabilities for each potential site. Results: Previously, we showed that our deep-learning model accurately predicts fragment ion intensities of phospho-peptides, which can be leveraged for identification and deconvolution of MS2 spectra in the Chimerys search algorithm. Here, we extend its capabilities to the localization of PTM sites in DDA and DIA data, exemplified by but not limited to phosphorylation. First, we designed a score that combines information on matched peaks with fragment ion intensity information. Calculating this score on the most intense b- and y-ions as well as neutral losses was sufficient for the accurate localization of PTM sites. In a DDA ground truth data set of purified phosphorylated peptides, the intensity-based scoring outperforms PTMProphet and MaxQuant at identifying known localizations. Here, the algorithm correctly localized 2% and 6% more overlapping PSMs at 99% precision, respectively. On this data set, ptmRS did not reach 99% precision. We then investigated how intensity-based scoring behaves in more complex samples, such as a publicly available data set of synthetic phospho-peptides, spiked into a complex yeast phospho-peptide background. Here, the algorithm correctly localizes 8% more overlapping PSMs at 99% precision than MaxQuant in DDA and 5% more overlapping precursors than DIA-NN in DIA. When analyzing all PSMs, we identify 32% more correctly localized PSMs than the FragPipe-PTMProphet pipeline. In both DDA and DIA, ptmRS did not reach 99% precision. In conclusion, we integrated intensity-based localization in Chimerys, obsoleting the need to pair Chimerys with ptmRS. The algorithm is compatible with DDA and DIA data analysis, providing a powerful new tool for PTM research. Conclusions: We present a novel algorithm utilizing predicted fragment ion intensities for the intensity-based localization of PTM sites in DIA and DDA data. 276

277

P-I-0129 Automated in-StageTip (iST)-based sample preparation for scaling up large cohort proteomics studies Zehan Hu1, Vincenzo Romaniello1, Torsten Pirch1, Claudia Martelli2, Irina Maerki2, Barbara Steigenberger3, Katrin Hartinger1, Nils Kulak1 1PreOmics GmbH, Application Development, Planegg, Germany 2Bruker Switzerland AG, Application, Faellanden, Switzerland 3Max-Planck-Institute of Biochemistry, Mass Spectrometry Core Facility, Martinsried, Germany Introduction: LC-MS is a powerful tool in proteomics, helping to decode biological processes, discover new biomarkers, and understand the function of new drugs. For comprehensive insights, analysis of large sample numbers is crucial. However, one of the main bottlenecks remains the lack of a robust and rapid workflow for sample preparation. To overcome the above challenges, we present a unique in-tip stationary phase-based proteomics workflow integrated in a fully automated, yet compact liquid handling platform, the Automated ProteinPrep 96 (APP96TM), to process up to 96 samples/day. Notably, our approach does not require a centrifuge, positive pressure, or disposable tips. Methods: The automated workflow is based on the PreOmics iST chemistry and includes all LC-MS sample preparation steps in a 96-well format: lysis, reduction and alkylation, digestion, and peptide clean-up. Plasma samples as well as human, insect and yeast cell pellets corresponding to a protein amount range between 2 and 50 µg were processed according to PreOmics iST protocol. 300 ng of peptides was separated with a 6 (for plasma samples)- or 30 (for others)-min gradient on the nanoElute 2 system and analyzed usingdia-PASEF® schemes on the timsTOF HT (Bruker). DIA search was performed using MaxQuant or Spectronaut 17/18 (Biognosys) using directDIA mode. Results: Automated sample preparation on the APP96 provided equal or higher protein and precursor identifications for all tested sample types (plasma samples, mammalian, insect, and yeast cells) compared to manual sample preparation. In experiments with HeLa cells, more than 6800 protein groups and 110´000 peptides were identified with an impressive 98% overlap of protein groups between manual and automated workflows. Notably the introduction of newly designed POPTip achieved significant reduction of methionine oxidation due to reduced air contact during clean-up. Moreover, excellent intra-day (manual: 8-10 %CV vs. automated: 5-8 %CV) and inter-day (manual: 23 %CV vs. automated: 17 %CV) repeatability in protein quantity could be obtained with the automated workflow, outperforming the manual approach. Conclusions: The APP96 platform combines robust, flexible, and fully automated sample preparation for LC-MS sample analysis with minimal environmental impact. Based on the high-quality and reliable data obtained, the APP96 enables confident upscaling of proteomics sample preparation, ideal suited large cohort studies. 277

278

P-I-0130 Enhancing protein quantification and sample throughput with the new TMTpro 32plex reagent and extended supporting features from thermo scientific Orbitrap ascend MultiOmics tribrid mass spectrometer Jingjing Huang1, David Bergen1, Dustin Frost1, Ryan Bomgarden2, Rosa Viner2, Graeme McAlister1, Rafael Melani1 1Thermo Fisher Scientific, San Jose, CA, United States 2Thermo Fisher Scientific, Rockford, IL, United States Multiplexed quantitative proteomics has gained significant attention in various research domains, including chemoproteomics, drug discovery, single-cell analysis, and plasma profiling. To address the demand for higher multiplexing capabilities and improved sample throughput, a novel TMTpro 32-plex reagent was developed and recently introduced, allowing for extensive protein quantification with enhanced accuracy and precision. To fully leverage the potential of the TMTpro 32plex reagent, it is crucial to employ mass spectrometers capable of resolving 3 mDa spaced reporter ion peaks. Consequently, the Thermo Scientific Orbitrap Ascend MultiOmics Tribrid mass spectrometer now offers new extended resolving power (RP) settings at Turbo 45k, Turbo 60k, eFT 75k, and eFT 90k, ensuring optimal resolution and speed for accurate quantification of proteins and peptides in highly multiplexed experiments. One of the key features that enhances data acquisition efficiency while maintaining quantification accuracy is the Real-Time Search (RTS) feature, implemented and available on Orbitrap Tribrid MS systems. RTS enables on-the- fly peptide spectrum matches (PSMs) for MS2 spectra with online database search, triggering Synchronous Precursor Selection (SPS)-MS3 only upon confident identifications. This feature significantly enhances data acquisition and quantification efficiency in complex sample mixtures. Our experiments utilized the Orbitrap Ascend MultiOmics Tribrid MS and Vanquish Neo systems with a PepMap 50cm column, employing a 140-minute LC gradient length with about 1ug sample loaded on column. Both MS2 and Real-Time Search MS3 experiments were performed under Turbo 45k, Turbo 60k, eFT 75k, and eFT 90k resolving power settings, with optimized maximum injection time for each resolving power and an AGC target of 300%. Data analysis was performed using Proteome Discoverer 3.2 TMTpro 32 plex quan node. The sample set comprised TMTpro 32plex-labeled HeLa digest with 1:1 and 1:4 ratios across channels, enabling assessment of the instrument parameters necessary to resolve reporter ions. Additionally, TMTpro-labeled HeLa digest with a Pierce 6 protein digest mix was used, with HeLa digest labeled 1:1 across all channels and the protein mix spiked in at different ratios across channels. Preliminary results from our TMTpro 32-plex experiment revealed quantification of over 4000 protein groups and more than 22,000 quantified peptides in the MS2 experiments. Evaluation with RTS-SPS-MS3 further demonstrated improved quantification accuracy. Instrument parameters, including resolving power, AGC, and injection time, were optimized. Comparison between 16-plex and 32-plex label experiments indicated equal or greater quantified proteins/peptides in the 32-plex experiment compared to two sets of 16-plex experiments. Quantification using ratio reference within sets exhibited accurate and precise results. 278

279

P-I-0131 Combining a new hybrid nominal mass platform and intelligent data acquisition to enable highly multiplexed targeted proteomics Philip Remes1, Cristina Jacob1, Lilian Heil1, Nicholas Shulman2, Brendan MacLean2, Michael J MacCoss2 1Thermo Fisher Scientific, Life Sciences Mass Spectrometry, San Jose, CA, United States 2University of Washington, Department of Genome Sciences, Seattle, WA, United States Targeted mass spectrometry (tMSn) traditionally is employed at the end of the biomarker discovery pipeline for small numbers of diagnostic compounds. Although tMSn produces the best quantitative data, it has had limited applicability to upstream experiments due to its low throughput.We describe a novel high-speed hybrid nominal mass platform designed for tMSn that has ~10x higher throughput for targeted peptide quantitation than triple quadrupole (QQQ) platforms.The increased performance is due to a combination of hardware/instrument control developments coupled with a real-time AdaptiveRT algorithm that adjusts for retention time shifts, allowing narrower acquisition windows. We also present a Skyline-based software plugin that streamlines instrument method creation.These advances enable large-scale tMSn with applications to translational and even discovery proteomics. Parallel ion management is used to achieve high duty cycles for ion injection and analysis, maximizing sensitivity and acquisition rates. A new long-life detector maintains stable gain while achieving single-ion detection limits. The instrument acquires nominal resolution mass spectrometry data at a rate of 65- 100 Hz for peptide analysis, depending on the choice of acquisition parameters. The Skyline plugin filters transition data according to metrics like area and signal-to-background, and filters precursors by a minimum number of qualifying transitions. Precursors are scheduled for analysis with a load-balancing technique that chooses the best peptides per protein and ensures a minimum points-per-peak. The AdaptiveRT algorithm estimates retention time shifts and updates acquisitions using cross correlation-based comparisons of real-time with reference data. Large scale discovery experiments are now capable of probing the depths of the proteome in unprecedentedly short times. However, the data independent and data dependent acquisition strategies used to perform those experiments carry tradeoffs in sensitivity and selectivity, as well as instrument complexity and cost. On the other hand, targeted experiments on QQQ are more cost effective and have good sensitivity but have not traditionally been easily scalable to more than a few hundreds of peptides. The new MS platform fills the gap between the capabilities of QQQ and HRAM instruments, with a capacity to target more than 5000 peptides per hour. We present a unified workflow for creating large scale tMSn assays on the new instrument from discovery data. With a new Skyline-based software plugin, precursors are selected that have the best chance of being good quantitative targets. We present experimental use cases that include relative and absolute quantitation, as well as MS2 and MS3 acquisition modes. While the platform"s additional throughput can be used to extend the number of targeted compounds, we show that the gains can be used to shorten the gradients used with smaller numbers of targets. 279

280

P-I-0132 Wideband PRM with fully automated dual online reverse-phase liquid chromatography for High-throughput targeted proteomics Chaewon Kang1, Dowoon Nam1, Hokeun Kim1, Jiwon Hong1, Kwon Hee Bok1, Jingi Bae1, Sang-Won Lee1 1Korea university, Center for proteogenome research, chemistry, Seoul, South Korea Background: Targeted proteomics is a technique used to obtain quantitative information about target proteins, making it widely used for verification and validation of candidate cancer biomarker proteins. Parallel-reaction monitoring (PRM), is favored due to the ease of assay development and the capability to detect all fragment ions. However, PRM has the limitation that the stable isotope labeled (SIL) peptides and endogenous (ENDO) peptides are monitored at different retention time. Methods: In this study, we developed a wide-band PRM (WBPRM) method to address the limitations of conventional PRM method. The WBPRM utilizes multiplexed single ion monitoring (MSX) scans to increase sensitivity of low-abundance target precursor ions. Furthermore, WBPRM features a wide isolation window that allows for simultaneous isolation and fragmentation of ENDO and SIL precursor ions. We also developed a fully automated dual online reverse-phase liquid chromatography system for high- throughput analysis. This LC system employs three additional switching valves and two analytical columns that enable mutually independent operation with hidden isocratic elution. As a result, this system eliminates dead time associated with sample injection, column washing and equilibration, thereby increasing experimental throughput by two folds. With this LC system, we optimized 15 minutes gradient with micro flow rate to analyze 96 samples per day (SPD). integrated the WBPRM method with dual online reverse-phase Result and conclusions: We liquid chromatography system to develop pancreatic ductal adenocarcinoma (PDAC) subtype identification technology using 149 PDAC patient samples. Dual LC system allows short gradient without deadtime, while WBPRM enables the targeting a large number of peptides in highly complex proteome samples within a short LC gradient. Consequently, in a 15-minutes gradient experiment, WBPRM targeted 148 PDAC peptides and detected transition profiles of 106 of 148 SIL peptides, and quantified 93 ENDO peptides on average. This combined system is thus ideal for high-throughput targeted proteomics, involving large cohort sample. 280

281

P-I-0133 Technical repeatability and efficiency in deep proteomics: a comparative study of various biological matrices Paraskevi Karousi1,2, Ioannis Kollias3, Panagiota Efstathia Nikolaou4, Foteini Paradeisi1,5, Maria Voumvouraki2,6, Ieronymos Zoidakis1,5, Efstathios Kastritis3, Nikolaos Thomaidis2,6, Guillaume Médard2,6, Julie Courraud2,3,6 1National and Kapodistrian University of Athens, Biochemistry and Molecular Biology, Athens, Greece 2National and Kapodistrian University of Athens, Proteomics Core Facility, Athens, Greece 3National and Kapodistrian University of Athens, Department of Clinical Therapeutics, School of Medicine, Athens, Greece 4National & Kapodistrean University of Athens, Laboratory of Pharmacology, School of Pharmacy, Athens, Greece 5Academy of Athens, Proteomics Laboratory, Biomedical Research Foundation, Athens, Greece 6National and Kapodistrian University of Athens, Laboratory of Analytical Chemistry, Department of Chemistry, Athens, Greece Understanding the technical repeatability of proteomic analyses is crucial for designing robust experimental studies. At the Proteomics Core Facility of the University of Athens, we emphasise the importance of conducting preliminary tests on different matrices to prove the feasibility of the chosen sample preparation method and ensure the consistency and reliability of results. We analysed replicates of protein extracts from undepleted serum, CD138+ cell pellets, protein phenol phase samples after RNA extraction (PPP) of CD138+ cells, murine bronchoalveolar lavage (BAL) fluid, and murine heart tissue. Proteins from BAL and PPP samples were precipitated with solvents and pelleted by centrifugation. Following adaptations from the SPEED protocol (Doellinger et al., 2020), proteins were denatured with trifluoroacetic acid (TFA), reduced and alkylated, then digested overnight with trypsin. Purified peptides were separated onto a C18 reverse-phase column (nanoElute 2) during a 30-min elution gradient. For each matrix, we optimised the injected number of cells or nanograms of protein to maximise coverage and ensure scalability. Data was acquired on a timsTOF fleX (Bruker) using library-free Data-Independent Acquisition (DIA) Parallel Accumulation Serial Fragmentation. Data was analysed using DIA-NN. Across the different sample types, we detected 400-5,000 protein groups, demonstrating high depth of analysis compared to other published studies. Serum, CD138+ cell pellet, and PPP samples showed robust repeatability with 81%, 81% and 81.5% of the 430, 4,800, and 3,600 protein groups detected below 20% coefficient of variations (CV), respectively. BAL samples were precipitated with ethanol, acetonitrile, or acetone, as well as with the SP3 method. Acetone precipitation yielded the best results regarding repeatability and protein identification, with an average of 1400 protein groups detected, 89% of which with CV <20%. Heart tissue analysis was much more challenging with no significant difference between TFA lysis and traditional RIPA buffer protein extraction. While more than 3,200 protein groups could be detected, only a quarter had a CV <20%. These tests allow us to improve our protocols and bring knowledge as to the reliability of proteomics analyses. The use of TFA enhances biological safety and simplifies the preparation process. We have further streamlined our workflows with a 30-min gradient, significantly shorter than traditional proteomics protocols, achieving high throughput and scalable proteomics pipelines. Depth of analysis was achieved thanks to careful optimisation of DIA windows. Along with ensuring the feasibility of the analyses, understanding the unique properties of each matrix often has a significant impact on study design, sample selection or collection, and appropriateness of proteomics to answer research questions. We recommend performing such tests very early in any study planning to include proteomics. 281

282

P-I-0135 Implementation of MS1 based real–time retention time adjustment on a hybrid Orbitrap mass spectrometer for directed and targeted proteomics Markus Kellmann1, Yovany Cordero Hernandez1, Hanno Resemann1, Philip Remes2, Jesse Canterbury2, Will Barshop2, Julia Kraegenbring1, Heiner Koch1 1Thermo Fisher Scientific, Bremen, Germany 2Thermo Fisher Scientific, San Jose, CA, United States Targeted proteomics studies with large cohorts often suffer from retention time instabilities caused by column aging, especially when low flow HPLC is utilized. To compensate for this, the time windows for scheduling the targets are chosen to be larger, which in turn reduces the number of meaningfully measurable targets. Real-time correction of retention times provides the possibility to get reproducible identification and accurate quantification of many peptide targets during a LC run regardless of column aging effects which usually result in instabilities of the retention times in large cohorts. Several solutions have already been developed and implemented to tackle this drawback. For example, Exploris Orbitrap mass spectrometers have implemented a landmark-based method that uses isotopically labelled peptides. Although this method is simple and effective, it has the disadvantage that the retention time is corrected at relatively long intervals (distances between the eluting peptides), measuring time is "lost" due to the monitoring of the standards and the costs for the synthetic peptides. A robust method without the need of spiking synthetic standards has been presented by Remes et al.. This method utilizes cross-correlations at MS1 level between reference run and current runs for calculating an estimated RT shift and applies this correction in real-time to the scheduled target list. Here we show the implementation on a modified Orbitrap Exploris 480 mass spectrometer. For initial assessment we tested different input variables of the algorithm to find the optimal balance between processing speed and correction performance. Additionally, we present results of a targeted long-term study with >150 samples and more than 800 targets as well as a data directed study with more than 5000 targets. Hela cell digests in different concentrations are used to mimic complex samples. In a first set of experiments, slightly different gradient slopes were used to mimic retention time instabilities and to assess the correction performance. For the long cohort study, replicate injections of HELA digest were used in combination with a nLC column. Preliminary experiments with 5-10 replicates and retrospective offline processing show a robust retention time correction on replicates of HELA cell lysate samples with differing gradient slopes. Setting the retention time window to 30s to 45s for a scheduled targeted experiment >95% of the targets (~800) are covered. In data-directed experiments, the window size can be reduced to 8-10 sec, as quantification is performed at MS1 level or via TMT. In this case, up to 95% of a list of 5000 peptides can be reproducibly found across the replicates. Short processing times for cross-correlation calculations and fast data transfer times make real-time correction possible within one scan cycle, thereby delivering superior performance compared to real-time correction based on isotopically labeled standard peptides. 282

283

P-I-0136 Comparative glycosylation analysis of etanercept and its biosimilars using TMT labeling and ZIC-HILIC HPLC chromatography Hyejin Kim1, Heeyoun Hwang2, Myung Jin Oh3, Hyun Joo An3, Jong Shin Yoo2, Jin Young Kim2,4 1Korea Advanced Institute of Science and Technology, Department of Bio and Brain Engineering, Daejeon, South Korea 2Korea Basic Science Institute, Cheongju, South Korea 3Chungnam National University, Daejeon, South Korea 4Korea Resesarch Instiitute of Biology and Biotechnology, Daejeon, South Korea Etanercept is a biopharmaceutical fusion protein composed of tumor necrosis factor receptor 2 (TNFR2) and the Fc region of human immunoglobulin G (IgG1), used to treat autoimmune diseases. Due to its heterogeneous glycosylation at multiple sites, a detailed glycoprotein analysis is essential for characterizing this biopharmaceutical. In this study, we present a workflow that integrates the enrichment and fractionation of TMT-labeled glycopeptides with ZIC-HILIC HPLC chromatography. This method was applied for an in-depth analysis of site-specific N- and O-glycopeptides from Etanercept, comparing the original product with two biosimilars. We identified 87 N-glycopeptides and 67 O-glycopeptides from three N-glycosylation sites and 15 O- glycosylation sites, respectively. For the first time, we report 12 O-glycopeptides from two novel O-glycosylation sites. The N149 site in the TNFR2 region accounted for over 70% of afucosylated complex-type glycans, while the N171 site predominantly featured fucosylated complex-type glycans. In the IgG region, the main glycan types were asialo fucosylated complex and high mannose. At all O-glycosylation sites, over 70% of the glycopeptides were of the core1 type. Glycosylation patterns among the three pharmaceuticals were similar in macro heterogeneity but differed in micro heterogeneity. Our method demonstrates that TMT labeling combined with ZIC-HILIC HPLC fractionation provides precise quantitation and higher throughput for site-specific N- and O- glycopeptide analysis in therapeutic fusion proteins, enabling the identification of novel glycosylation sites and characterization of low abundant glycosylation types. 283

284

P-I-0137 Human transcription factor interactions and environment Matias Kinnunen1, Markku Varjosalo1 1University of Helsinki, Institute of Biotechnology, Helsinki, Finland Our research focuses on human transcription factor (TF) protein interactions and close protein environment. TFs are proteins that control gene expression by recruiting transcription cofactors to regulate not only transcription itself but also the epigenetic landscape and access to the genomic DNA. One of the outstanding challenges, and point of interest, in the study of TFs and their cofactors is that these interactions occur in context of phase separation. Phase separation forms dynamic aggregates of proteins and RNA where both contribute to the interactions within the aggregate. Phase separation is typically initiated by intrinsically disordered regions (IDRs), which are found in many TF effector regions. We use biotin proximity labeling (BioID) to study the human TFs in the HEK293 model system. BioID detects both stable and transient protein interactors by biotinylating nearby protein cysteine residues from the tagged TF of interest. This method can capture dynamic and mobile protein constituents of phase separation aggregates. The resulting complex protein samples are analyzed using mass spectrometry. By thoroughly examining the interactions of human TFs on a large scale, we can identify the mechanisms and roles of individual TFs, the cofactors they interact with and potential interactions with other TFs. By analyzing the TF IDRs and interactors, we can also start to understand how particular IDR sequences govern specific interactions in the cell. 284

285

P-I-0138 Evaluation of high dynamic range MS1 acquisition methods on a hybrid Orbitrap mass spectrometer Hanno Resemann1, Markus Kellmann1, Catharina Crone1, Julia Kraegenbring1, Thoeing Christian1, Claire Dauly1, Heiner Koch1 1Thermo Fisher Scientific, Bremen, Germany Advancements in MS-based proteomics have enabled the analysis of the proteome comprehensively and quickly. However, accurately identifying and quantifying low-abundant human proteins remains difficult due to the wide range of protein abundance found in tissues and body fluids. Gas phase fractionation methods like BoxCar as well as BoxCar Assisted MS Fragmentation (BAMF) have addressed this challenge by adding multiple narrow mass-to-charge segments to fill the Orbitrap more efficiently. In our study, we have further developed these ideas. We made modifications to an OrbitrapTM Exploris 480 mass spectrometer to allow gas phase fractionation and intelligent MS1 multiplexing schemes with optimized injection times for high dynamic range Orbitrap MS1 scans. Pierce HeLa samples were analyzed by LC-MS/MS with 60min-per-run methods. The analysis was performed on a Thermo ScientificTM VanquishTM Neo UHPLC system with a Thermo ScientificTM PepMapTM column (2μm, 150μm x 15cm) connected to a Thermo ScientificTM OrbitrapTM Exploris 480 mass spectrometer. The instrument was operated in DDA mode with Orbitrap-MS1 resolution at 120,000, and ddMS2 scans acquired using 2-Th quadrupole isolation windows and a resolution of 15,000. The raw data was evaluated with Proteome DiscovererTM using the CHIMERYSTM search engine. We have performed data acquisition with proof-of-concept MS acquisition methods to achieve higher dynamic range (HDR) in Orbitrap based MS1 scans. This approach encompasses the on-the-fly decomposition of MS1 spectra into two full mass range sub[CC2] -scans, which are reconstituted as one afterwards. For gas phase fractionation of the ion current, we have implemented two different strategies, an equidistant mode (HDR equidistant) and a dynamic mode (HDR dynamic) to fractionate horizontally along the m/z range. Both methods profit from using low fill times for fractions with high-abundant signals and high fill times for fractions with low- abundant signals, achieving high dynamic range across the fully assembled MS1 spectrum. Gas phase fractionation for HDR leads to a significant increase of intra spectrum MS1 complexity when a tryptic HeLa digest is applied. Both approaches result in a substantial increase of the signal-to-noise ratio of Orbitrap MS1 scans with gain factors >20 for lower abundant ions. Enabling HDR also leads to a significant increase of global peptide isotopic patterns found in a 1h LCMS run of a HeLa proteolytic digest. The HDR equidistant mode results in detection of additional 48% isotopic patterns while additional 59% are detected in dynamic mode when compared to a control. Our findings confirm that the HDR approaches works in principle, leaving room for further improvements on identification and quantification of proteins with a wide range of abundance. 285

286

P-I-0139 Assessment of high-resolution DIA methods and short gradients on high-throughput µPAC columns for maximum proteome coverage and quantitative performance Julia Kraegenbring1, Dominic G. Hoch2, Riccardo Stucchi3, Jeff Op de Beeck4, Ece Aydin5, Robert van Ling4, Heiner Koch1, Kevin Yang6, Amirmansoor Hakimi6 1Thermo Fisher Scientific, Bremen, Germany 2Thermo Fisher Scientific, Dreieich, Germany 3Thermo Fisher Scientific, Reinach, Switzerland 4Thermo Fisher Scientific, Ghent, Belgium 5Thermo Fisher Scientific, Germering, Germany 6Thermo Fisher Scientific, San Jose, CA, United States Understanding biological processes on a global proteome-wide scale is crucial in scientific research. Achieving statistically significant quantitative data often involves studying large cohorts comprising hundreds of biological samples and replicates. An effective workflow must be capable of processing, measuring, and analyzing a high volume of samples while ensuring data reproducibility and consistent performance throughout the study. The primary aim of this study was to introduce a robust workflow for high-throughput label-free quantification using a Vanquish Neo LC system equipped with a 5.5 cm uPAC Neo High Throughput column and an Orbitrap Exploris 480 mass spectrometer. We evaluated three gradient lengths, namely 3.5, 5.5, and 10 minutes. The mass spectrometric method, predominantly the isolation window widths, have been optimized for two different objectives: one method for giving maximum proteome coverage of the sample, another for gaining high-quality quantitative information on the protein abundances. To further showcase the quantitative capabilities of these methods, we employed a three-species proteome mix, achieving accurate quantification results across all three gradient lengths using the MS method for quantitative analysis. Furthermore, we demonstrated the workflow's robustness through an analysis of several 100 consecutive injections, proving this workflow to be suitable for the reliable analysis of large cohort studies. Data processing was carried out using different software tools with library-free analysis approaches. The main objective of this study was to evaluate the suitability of µPAC columns with 5.5cm flow path for maximizing the throughput of MS analysis gaining reliable and robust quantitative information. Each gradient length was analyzed for the number of identified proteins and peptides from different amounts of HeLa digests. Despite the short gradients, the data clearly illustrate that a median of 5-6 points per peak could be measured for isolated HeLa peptide XICs, achieved by meticulous duty cycle adjustments in each of the three DIA methods tested. In experiments focused on label-free quantification (LFQ), precision and accuracy in measuring relative protein and peptide abundances are crucial. To assess the quantitative performance of the described methods, the study analyzed a mix of three species proteomes, i.e. tryptic digests of human, yeast, and E.coli proteins mixed in specific ratios. The quantification precision, evaluated through coefficients of variation (CV) across different technical replicates at both the protein and peptide levels, consistently demonstrated CVs well below 10% and minimal missing values, indicating high technical reproducibility. 286

287

P-I-0140 Label-free DIA-based workflow for single-cell proteomic analysis on an Orbitrap Exploris 480 mass spectrometer Julia Kraegenbring1, Fernanda Salvato2, Bernard Delanghe1, David Hartlmayr3, Anjali Seth3, Heiner Koch1, Amirmansoor Hakimi2 1Thermo Fisher Scientific, Bremen, Germany 2Thermo Fisher Scientific, San Jose, CA, United States 3Cellenion SASU, Lyon, France Recent advances in LC-MS have enabled label-free single-cell proteome analysis, revealing unexpected functional diversity in cells. However, there are still key challenges in this field, such as sensitivity, coverage, dynamic range, and throughput. To address some of these challenges, new method developments as well as optimization of existing LC-MS-based proteomics workflows are necessary. Here, we demonstrate the use of the Orbitrap Exploris 480 mass spectrometer and the Vanquish Neo UHPLC system for high-throughput single cell applications. Individual HeLa cells were sorted, followed by reduction, alkylation, and trypsin digestion using CellenONE as per the manufacturer"s protocols. Pierce HeLa digest was used for dilution series from 50 pg to 10 ng loaded on column. Single cell digests and the diluted standard HeLa digest samples were analyzed using the Orbitrap Exploris MS with the FAIMS Pro interface coupled to the Vanquish Neo UHPLC system. Separation was performed on the Aurora Ultimate TS 25cm column. Data was acquired in a DIA mode and searched with a beta version of Proteome Discoverer Software 3.1 and Spectronaut 18. The performance of this ultra-sensitive LC-MS workflow was optimized using a dilution series of HeLa digest. From 250 pg of HeLa digest load, we could identify on average > 2,000 protein groups by library-free approach whereas using DIA-library generated from 5 ng HeLa digest, we were able to identify >3,000 protein groups using a method with a throughput of 50 samples per day. 287

288

P-I-0141 Brucellin proteomic analysis for a standardized skin tests fostering Brucella infection diagnosis in animals Ivanka Krasteva1, Mirella Luciani2,1, Federica D’Onofrio1, Fabrizio De Massis1, Flavio Sacchini1, Tiziana Di Febo1, Chiara Di Pancrazio1, Fabrizia Perletta1, Marta Maggetti1, Luca Freddi3, Vitomir Djokic3, Acacia Ferreira Vicente3, Claire Ponsart3, Manuela Tittarelli1 1Istituto zooprofilattico dell'abruzzo e del molise, Teramo, Italy 2University of Teramo, Teramo, Italy 3ANSES-EU/WOAH, Reference Lab. for Brucellosis, Paris, France The Brucellin Skin Test (BST) detects brucellosis in animals through a cell-mediated immune response to a protein extract from B. melitensis strain 115, which is almost free of lipopolysaccharide. It is highly specific and used to confirm suspected false positives in small ruminants and swine, but not recommended for screening due to low sensitivity. Despite its diagnostic significance, the protein composition of brucellin has not been fully characterized. This study used nLC-ESI-MS/MS analysis and bioinformatics tools to evaluate brucellin's protein composition and identify immunoreactive proteins. An allergen suspension of purified proteins (free of S-LPS) of EU Standard Brucellin (batch 2011-01), produced by ANSES was used. A total of 248 proteins were identified 206 of which are immunogenic. Key proteins identified include the L7/L12 ribosomal protein, bacterioferritin, outer membrane protein BP26/OMP28, and EipB, which are crucial in inducing DTH and for Brucella pathogenesis. The L7/L12 protein, induces strong delayed-type hypersensitivity (DTH), dependent on post-translational modification. It is also a major component of tuberculin purified protein derivative (PPD) and is a major heat- resistant protein inducing strong delayed-type hypersensitivity Additionally, proteins such as chaperonin, superoxide dismutase, DNA-binding proteins, and ABC transporter substrate-binding proteins are known for their immunogenic roles in cellular functions and stress responses These findings underscore the potential of these proteins as candidates for detecting cellular immunity to Brucella. Developing recombinant Brucella- allergenic proteins could help in standardizing skin tests, providing reliable allergens and reducing exposure risks. Moreover, a serological test using these recombinant proteins could improve sensitivity and specificity in Brucella infections diagnostic, eliminating false positivities related to LPS-based diagnostics. Bachrach G, Banai M, Bardenstein S, Hoida G, Genizi A, Bercovier H. Brucella ribosomal protein L7/L12 is a major component in the antigenicity of brucellin INRA for delayed-type hypersensitivity in brucella-sensitized guinea pigs. Infect Immun. 1994 Dec;62(12):5361-6. doi: 10.1128/iai.62.12.5361-5366.1994. PMID: 7960115; PMCID: PMC303276 Nandini P, Jakka P, Murugan S, Mazumdar V, Kumar D, Prakash R, Barbuddhe SB, Radhakrishnan G. Immuno- profiling of Brucella proteins for developing improved vaccines and DIVA capable serodiagnostic assays for brucellosis. Front Microbiol. 2023 Oct 4;14:1253349. doi: 10.3389/fmicb.2023.1253349. PMID: 37860136; PMCID: PMC10582347 He CY, Yang JH, Ye YB, Zhao HL, Liu MZ, Yang QL, Liu BS, He S, Chen ZL. Proteomic and Antibody Profiles Reveal Antigenic Composition and Signatures of Bacterial Ghost Vaccine of Brucella abortus A19. Front Immunol. 2022 Apr 22;13:874871. doi: 10.3389/fimmu.2022.874871. PMID: 35529865; PMCID: PMC9074784. 288

289

P-I-0142 Scalable single cell analysis towards 1000 samples per day using the Evosep whisper zoom methods on the timsTOF Ultra2 Christoph Krisp1, Verena Tellstroem1, David Hartlmayr2, Anjali Seth2, Guilhem Tourniaire2, Dorte Bekker-Jensen3, Ole Bjeld Hørning3, Nicolai Bache3, Markus Lubeck1 1Bruker Daltonics GmbH & Co. KG, Bremen, Germany 2Cellenion SASU, Lyon, France 3Evosep, Odense, Denmark Scalability is one of the crucial factors for single cell proteomics to allow meaningful data collection with reliable and robust statistical power for elucidating heterogeneity in ultra-high sensitive single cell and near single cell tissue input applications. Multiplexing as well as fast chromatography with low overhead time aids in upscaling proteome analyses. Here, we demonstrate applicability of the new Whisper Zoom methods on the Evosep system for speeding up label free sample analysis up to 120 samples per day (SPD) and exceeding >1000 samples per day in a multiplexing approach with data acquisition on the timsTOF Ultra2. HeLa and HEK293 as well as small HEK 293 spheroids were isolated into the proteoCHIP® EVO 96, directly lysed and digested using the cellenONE platform. Samples were transferred by centrifugation onto Evotips, separated in Whisper Zoom 120, 80 and 40SPD, analyzed on timsTOF Ultra2 in dia-PASEF® mode and processed with Spectronaut 19 using directDIA+. Further, protein digests were labelled with a subset 16-plex TMTpro, combined to 9-plex single cell equivalent samples (250 pg per labeled), analyzed in Whisper Zoom 120SPD on the timsTOF Ultra2 in dda-PASEF and analyzed with Spectromine 4.5. Analyses of HeLa cell digest dilutions from 50 ng to 0.25 ng were performed to assess sensitivity of the different Whisper Zoom methods identifying more than 4000 protein groups from 250 pg at 120SPD, 5,000 proteins at 80SPD and 5500 protein groups at 40SPD speed and reaching 6,500 protein groups at 120SPD, 7000 protein groups at 80SPD and 7700 protein groups at 40SPD from the higher loads. From single Hela cells, protein identification rates correlated with gradient length, identifying on average 2500 protein groups at 120SPD, 3000 protein groups at 80SPD and 4000 protein groups at 40SPD. This setup was then used to investigate cellular responses to LPS treatment on single cell level (HEK 293 cells 40x treated, 40x controls) and small spheroids (3 – 10 cells per spheroid; 12x treated, 12x controls). Results were compared to responses seen in 30 cell isolation bulk samples (4x treated, 4x controls). As expected, sample to sample variability decreased with an increasing cell number. However, comparing protein groups identified in the undisturbed non-treated samples, 65% of all proteins identified in 30 cell bulk were identified at single cell level and increased to 85% on small spheroid level. LPS treatment responses on 30 cell bulk samples, spheroids and single cell level were comparable showing expected activation of inflammatory pathways. The 9-plex multiplexing approach run at Whisper Zoom 120SPD speed resulted across the 1080 samples on average in the identification of 1500 protein group per label/cell type. Excellent tip to tip chromatographic reproducibility as well as quantitative accuracy within and across TMT batches was observed, accurately separating the 360 HEK 293 samples from the 360 HeLa samples and the 360 K562 samples. 289

290

P-I-0143 Low-variation proteome profiling across 11 Labs, identifying 7200+ protein groups in 5 minutes with dia-PASEF Stephanie Kaspar-Schoenefeld1, Tobias Kroniger1, Ann-Christine König2, Stefanie Hauck2, Sebastian Johansson3, Axel Karger4, Uli Ohmayer5, Matteo Pecoraro6, Stefan Tenzer7, Ute Distler7, Torsten Mueller1 1Bruker Daltonics GmbH & Co. KG, Bremen, Germany 2Helmholtz Center Munich, Metabolomics and Proteomics Core, Munich, Germany 3PreOmics GmbH, Planegg, Germany 4Friedrich-Loeffler-Institute, Institute of Molecular Virology and Cell Biology, Greifswald, Germany 5NEOsphere Biotechnologies GmbH, Planegg, Germany 6Università della Svizzera italiana, IRB, Bellinzona, Italy 7University Medical Center of the Johannes Gutenberg-University Mainz, Institute for Immunology, Mainz, Germany High throughput proteomics is gaining traction as it facilitates screening of large sample cohorts or many projects especially important in clinical research and systems biology studies. Recent development in mass spectrometry proteomics resulted in improved hardware and software making high throughput proteomics accessible to a wide range of laboratories. Here we benchmark dia-PASEF, a data-independent acquisition scheme for deep proteome analysis of human cell lines in a short 5-minute gradient. To show intra- and interlaboratory reproducibility, we performed a multi-laboratory study including 11 sites. We demonstrate that on average 7286 (± 32) protein groups were identified on the timsTOF HT instrument with low variation. Our results underline that dia-PASEF data acquisition combined with reproducible chromatography enables high reproducibility and data consistency across instruments and laboratories, which is a prerequisite for translational biomedical insights. 290

291

P-I-0144 Comparative evaluation of proteomics software suites for DDA and DIA acquisition strategies Karl Kristian Krull1,2, Syed Azmal Ali1, Jeroen Krijgsveld1,2 1DKFZ, Proteomics of Stem Cells and Cancer, Heidelberg, Germany 2University of Heidelberg, Medical Faculty, Heidelberg, Germany Mass spectrometry (MS) is a very powerful approach for proteome analysis, offering different instruments that can acquire raw data in different acquisition modes. In addition, an increasing palette of software tools have become available for the analysis of MS data, making the selection of suitable methods a complex decision. Here, we provide a comparative evaluation of six current software suites, namely FragPipe, MaxQuant, PEAKS Studio, DIA-NN, MaxDIA and Spectronaut for data acquired on three different mass spectrometers: a QExactive- HF, Exploris480 and a timsTOF Pro. To assess the software in a fair and systematic manner, we investigated their performances analyzing the same data of a human proteome sample spiked with different amounts of E.coli proteins, which we generated on each instrument by both DDA and DIA. We evaluated the capabilities of each software by the attained proteome coverage, we determined false-identifications and focused specifically on protein quantification to give practical advice to the end-user. Notably, we found that performance of software differed profoundly across instruments, which calls for certain software to be used only in combination with specific instruments, while indicating that tools may be optimized for more universal application. Moreover, in an unconventional approach, we linked the magnitudes of the detected proteome range to their respective quantitative accuracy, allowing filtering based on data quality before making biological inferences. When comparing software tools across acquisition modes, we observed that among the tested DIA suites only results from DIA-NN aligned to DDA-derived data, making DIA-NN particularly powerful for integration of data sets. We further noticed that DDA software handled our entrapment approach much better than the DIA tools, leading to fewer false-identifications, and we therefore explored ways to tighten the existing filters, thus suggesting new standards for the analysis of DIA data. Beyond serving as a comprehensive reference for state-of-the-art software capabilities, we envision this work to constitute a valuable resource to peer researchers, encouraging adaptation and refinement of data analysis in the proteomics community. 291

292

P-I-0145 Development of a high-throughput kinase activity platform using nanoLC-MS/MS with DIA approach for studying the anti-cancer mechanism of Taxol in ovarian cancer Chia-Hsin Lee1, Yu-Ching Liu2, Chao-Jung Chen2,3 1China Medical University, Department of Chinese Pharmaceutical Sciences and Chinese Medicine Resources, Taichung, Taiwan 2China Medical University Hospital, Proteomics Core Laboratory, Department of Medical Research, Taichung, Taiwan 3China Medical University, Graduate Institute of Integrated Medicine, Taichung, Taiwan Protein phosphorylation carried out by protein kinases increases protein diversity and further influence various cellular physiology. Peptide substrates have been used for in vitro phosphorylation with purified kinase or cell lysates. Herein, a quantitative high-throughput platform was developed for assessing multikinase activity using nano-LC-MS/MS with a data-independent acquisition (DIA) approach. This platform was evaluated by studying the kinase activity in Taxol-treated SKOV3 cells. A library containing 38 peptide substrates was designed and analyzed to determine the activities of major kinases involved in cancer development. Our results showed that 23 synthetic peptide substrates had significance phosphorylated changes in triplicate biological experiments. The 23 peptide substrates were further verified by western blotting. Our findings reveal that Taxol suppressed SKOV3 cell survival by activating AMPK and suppressing the PI3K-Akt-dependent pathway, ultimately leading to mTOR inhibition. Furthermore, in combination with ERK, Akt, SGK, CK1 and ErbB2 inhibitors, Taxol enhanced the inhibitory effect on ovarian cancer. This platform can be an attractive approach for large-scale kinase activity studies to comprehensively elucidating the mechanisms of drug-disease treatments, enabling the investigation of more effective therapeutic strategies. 292

293

P-I-0146 High-throughput high-resolution data-independent acquisition workflow on an Orbitrap ascend MultiOmics tribrid mass spectrometer for accurate label-free quantitation Kevin Yang1, Julia Kraegenbring2, Jingjing Huang1, Amanda Lee1, Julian Saba3, Amirmansoor Hakimi1 1Thermo Fisher Scientific, San Jose, CA, United States 2Thermo Fisher Scientific, Bremen, Germany 3Thermo Fisher Scientific, Winnipeg, Canada Quantitative proteomics is an essential tool for understanding global protein expression and the mechanisms of biological processes and disease states. Accurately quantifying the abundances of proteins of interest in complex samples is a prerequisite for developing suitable statistical models to gain biological insights from experimental data sets. Statistical significance is improved by decreasing variability in measurements and/or increasing the sample set. A suitable analytical workflow addresses the need for reproducible sample preparation, robust separations, high-quality quantitative measurements, and reliable data analysis. In the present study, we developed a high-throughput Velocity LFQ-DIA analysis workflow on the Orbitrap Ascend MultiOmics Tribrid mass spectrometer for comprehensive proteome coverage and excellent quantitation accuracy. To evaluate the performance of the Velocity DIA workflow, HeLa digest and three-proteome mixtures were loaded onto a 50 cm μPAC Neo HPLC column and separated using a Vanquish Neo UHPLC system over 9 min, 30 min, and 60 mins active LC gradients, respectively, before being transferred into the Orbitrap Ascend MultiOmics mass spectrometer. Acquired data has been processed by Spectronaut (Biognosys, v18) using a directDIA approach, DIA-NN (v1.8.1) or Proteome Discoverer software (v3.1.0.638) using CHIMERYS™ intelligent search algorithm by MSAID. Various run times of 9, 30, and 60 minutes were evaluated for a Hela digest standard to meet different throughput needs. With a 30 min active gradient, 7,000+ proteins and 47,000+ peptides were identified, along with a protein group CV of approximately 5%, suggesting that the 30 min active gradient method enables relatively high throughput while maximizing identification and quantitative performance. We extended this workflow to a 60 min active gradient and successfully identified close to 7,800 proteins and >76,000 peptides, highlighting that deeper proteome coverage can be achieved in the Velocity DIA workflow by using a longer gradient. To test the quantitative accuracy of the Velocity DIA workflow, we created a 3-proteome mix to mimic biological samples where proteins might be up- or down-regulated under different conditions. The Velocity DIA workflow yielded excellent quantitative accuracy across a wide dynamic range, with median values extremely close to the theoretical ratios, as well as a narrow distribution of all data points around the median values, indicating high quantitative accuracy and precision of the workflow. Additionally, the three-proteome mix experiment further highlights the proteome depth that can be achieved with an Orbitrap mass spectrometer. In the 60 min active gradient experiment, >11,000 protein groups were identified. Together, our data demonstrated deep proteome coverage and excellent quantitation accuracy afforded by the Orbitrap Ascend MultiOmics mass spectrometer. 293

294

P-I-0147 Top-down and bottom-up analysis of proteins and PTMs using an Orbitrap-Omnitrap instrument equipped with ExD, UVPD and IRMPD Nikita Levin1,2, Yana Demyanenko1,2, John Sidda1,2, Kyle L. Fort3, Alexander Makarov3, Athanasios Smyrnakis4, Dimitris Papanastasiou4, Shabaz Mohammed1,5,6 1Rosalind Franklin Institute, Didcot, United Kingdom 2University of Oxford, Department of Pharmacology, Oxford, United Kingdom 3Thermo Fisher Scientific, Bremen, Germany 4Fasmatech Science and Technology, Athens, Greece 5University of Oxford, Department of Biochemistry, Oxford, United Kingdom 6University of Oxford, Department of Chemistry, Oxford, United Kingdom For many years, CID has been the method of choice for peptide sequencing in bottom-up proteomics. However, as the focus of the proteomics community has been shifting in recent years to more challenging tasks such as identification and localization of complex PTMs and characterization of proteoforms by means of (native) top- down MS, a need for more advanced fragmentation techniques has emerged. Here, we report the results of protein and peptide characterization by various electron based (ECD, EID) and laser based (IRMPD, UVPD) fragmentation techniques and combinations thereof integrated in a novel omnitrap instrument as well as the results of the implementation of these techniques on the LCMS scale for the analysis of complex peptide and glycopeptide mixtures. All spectra were obtained in an omnitrap MS platform coupled with an Orbitrap Exploris 480 mass spectrometer. The omnitrap is equipped with an electron source for ECD and EID experiments, a UV 193 nm laser for UVPD, and an IR 10um laser for IRMPD and activated-ion fragmentation. We first performed an initial characterization of UVPD, IRMPD, EID, ECD, and activated-ion ECD (AI-ECD) available in the Omnitrap using several standard polypeptides in direct infusion experiments. The aptitude of each of the dissociation techniques listed above for conventional bottom-up proteomics were probed in a series of large-scale LCMS experiments on cell lysates. Cell lysates were digested using various proteases (trypsin, LysC, GluC, LysN, chymotrypsin) and the digests were offline fractionated prior to LCMS analysis. The identified peptides allowed us to compare the fragmentation techniques in terms of their suitability for peptide sequencing and localisation of PTMs. This experiment yields data which is orthogonal/complementary to conventional CID data. In a separate series of experiments, we examined the utility of Omnitrap UVPD, EID, ECD, and AI-ECD for the analysis of complex glycopeptide mixtures. We focused on the efficiency of each technique for sequencing of peptide and glycan parts of a glycopeptide and its ability to localise glycosites. We demonstrated that the Orbitrap-omnitrap hybrid is capable of generating high-quality spectra with reasonable (up to 20 Hz) scan rate suitable for LCMS experiments. In this work, we compare alternative fragmentation techniques from the perspective of sequence coverage, fragmentation efficiency, and the ability to localise PTMs and discuss their utility for the top-down and bottom- up analysis of (modified) proteins. In addition, we generated a large peptide dataset which provides a great material for training deep learning models for spectra prediction and thus further enhances the quality of deep proteome sequencing and PTM mapping. 294

295

P-I-0148 Proteomic investigation of proteolytic processing dependent lysosomal hydrolase maturation Jiaran Li1, Dominic Winter1 1University of Bonn, Bonn, Germany Lysosomes are the main degradative organelles of almost all eukaryotic cells and crucial for decomposing biological macromolecules. This function is facilitated by ~70 hydrolases which are contained in the lysosomal lumen. In order to form mature lysosomes, luminal proteins, membrane proteins, and lysosome-associated proteins have to be transported from the Golgi apparatus to lysosomes and assembled in the correct way. For a large number of lysosomal luminal proteins, this includes the transition from an immature pre-form to a fully functional mature version by proteolytic processing. While this transition was characterized for some lysosomal hydrolases, for others the exact events of maturation are not known. In order to compare the pre-form and mature form of lysosomal hydrolases, HEK293 were treated with ammonium chloride, which triggered exocytosis and resulted the secretion of lysosomal hydrolases from the Golgi apparatus in the culture medium. Lysosomes were enriched with SPIONS (pulse 24 hours followed by chase 24 hours). Concentrated medium and lysosome-enriched fractions were digested under RapiGest, digested peptides were analyzed using a Dionex Ultimate 3000 nanoUHPLC system coupled to an Orbitrap Fusion Lumos mass spectrometer. Mass spectrometry analyses were performed in the DIA mode, DIA *.raw files were analyzed with the Pulsar search engine available in Spectronaut. By comparing peptides identified for lysosomal hydrolases from fractions derived by hypersecretion from the Golgi apparatus (containing pre-forms) and from lysosome-enriched fractions (containing mature forms), the significant peptides were selected with log2 fold change > 0.58. We analyzed the ratio distribution for individual peptides of 77 lysosomal hydrolases. In total, we were able to identify maturation-specific differentially regulated peptides, indicating cleavage events for 30 hydrolases. This included 10 hydrolases for which the proteolytic activation process was hitherto unknown. We further performed DDA analysis and semi-tryptic searches to further define the exact region/site of proteolytic processing. 295

296

P-I-0149 Evaluation of plasma sample preparation methods for high-throughput mass spectrometry based proteomics and application to colorectal cancer Jianing Liu1, Liva Stalidzane2, Berend Gagestein1, Catarina Almeida-Marques1, Thang V. Pham1, Jaco Knol1, Connie Jimenez1 1Cancer Center Amsterdam, OncoProteomics Laboratory, Amsterdam, Netherlands 2Vrije Universiteit Amsterdam, Amsterdam, Netherlands Introduction and Objectives: The significance of plasma as a source of potential biomarkers is immense. In recent years, methods for high-throughput plasma proteomics based on mass spectrometry have emerged that enable profiling of large clinical cohorts. The aim of our project is to compare and implement a high-throughput plasma proteomics workflow that enables profiling at a depth of >500 proteins per sample in single shot. Methods: Plasma processing methods of interest enable sample processing at a throughput of 30 samples per day and compatible with low sample input (ideally a few ml). Sample preparation methods that are being explored: 1. In-solution digestion; 2. Suspension Traping filter (Strap)-based sample processing (Li et al., 2023); and 3. a precipitation method using perchloric acid (Viode et al., 2023). To this end, aliquots of one donor sample will be used for analysis of workflow triplicates. Moreover, we explored timsTOF high-throughput quantitative performance using different plasma protein input (400ng, 500ng, 600ng). Results: To determine the most optimal workflow, the data are being compared in terms of total number of protein identifications, unique identifications, reproducibility of protein identification and quantification. We attempt several alterations and their combinations to ensure the Viode et al method can be tailored for our requirements, the precipitation was carried out as described in the article, with volume adjustments when using HLB cartridges. So far our results show that the Perchloric acid (perCA) depletion method yields the highest number of proteins (330 proteins) per sample at good reproducibility (CV=13%). Since the aim amount of proteins is over 500, we will try to use DIA instead of DDA to analyze plasma proteins. Conclusion: The best performing method will be applied to a clinical cohort of plasma samples of patients with colorectal cancer and control subjects to explore the detection of previously defined colon cancer secretome biomarker candidates. 296

297

P-I-0150 Evaluation of diagonal-PASEF for data independent acquisition in proteomics Markus Lubeck1, Stephanie Kaspar-Schoenefeld1, Christoph Krisp1, Andreas Schmidt1, Silke Neustifter1 1Bruker Daltonics GmbH & Co. KG, Bremen, Germany Introduction: Data independent acquisition became the most widely used method for proteomics due to its high level of sensitivity, reproducibility, and data completeness. To increase sensitivity and sample throughput, a lower number of broader isolation windows needs be applied, which leads to reduced selectivity. The combination with dispersive ion mobility separations like Trapped Ion Mobility (TIMS) prior to quadrupole isolation increases selectivity, however the quadrupole has to switch its isolation position in multiple steps to follow the mobility separation. A more efficient way to sample the mass-mobility range is to use a scanning quadrupole, that can continuously follow the ion cloud eluting from the TIMS device (diagonal-PASEF). Methods: Diagonal-PASEF is enabled on recent instruments of the timsTOF series (Bruker, Bremen), running the latest instrument control software. Peptide identifications as well as quantitative reproducibility among technical replicates were evaluated with dilution series of human cell line digests (K562, Promega and HeLa, Pierce) in the pg to µg range, separated by nanoHPLC (250nl/min). Depending on gradient length, 5cm or 25cm columns (0.075mm ID, IonOpticks, Melbourne) were used. Data were processed with Spectronaut 19 (Biognosys, Schlieren). Results: Scanning vs. switching quadrupole isolation schemes were designed to cover the same m/z - mobility range. A widely used window scheme consists of 8 isolation windows of 25 m/z width, that change their mass position twice during an ion mobility scan, covering almost all peptide ions in a range of 400-1000 m/z. With a scanning quad, 4 windows, also of 25 m/z width, can be used to cover a very similar set of precursor ions. This doubles acquisition speed as well as ion usage, as individual peptides are fragmented twice as often over the course of a chromatographic peak, so improvements were expected for low sample amounts and/or fast gradients. For 0.25ng of diluted cell digests, diagonal-PASEF outperformed the stepping window scheme in this technical comparison. Diagonal-PASEF also slightly increased the number of identified peptides and proteins from real single HeLa cells when comparing to the current optimal high sensitivity dia-PASEF method. For high sample amounts, identifications were similar between classic and diagonal-PASEF; for very short gradients a slight improvement in the percentage of identified precursors with a coefficient of variation (CV) below 10 % resp. 20% was observed. 297

298

P-I-0151 Proteomic characterization of outbreak-associated Listeria monocytogenes strains in Italy: discovery of virulence markers Mirella Luciani1,2, Ivanka Krasteva1, Federica D’Onofrio1, Luigi Iannetti1, Marina Torresi1, Chiara Di Pancrazio1, Patrizia Centorame1, Fabrizia Perletta1, Silvia Valentinuzzi3, Manuela Tittarelli1, Francesco Pomilio1, Nicola D'Alterio1, Antonello Paparella2, Piero Del Boccio3, Flavio Sacchini1, Maria Schirone2, Francis Butler4 1Istituto zooprofilattico dell'abruzzo e del molise, Teramo, Italy 2University of Teramo, Department of Bioscience and Technology for Food, Agriculture and Environment, Teramo, Italy 3University “G. d’Annunzio” of Chieti-Pescara, Department of Pharmacy, Chieti, Italy 4University College Dublin, Biosystems Engineering, UCD School of Agriculture, Food Science and Veterinary Medicine, Dublin, Ireland Listeria monocytogenes, a persistent foodborne pathogen associated with ready-to-eat (RTE) foods, remains a significant threat to food safety and consumer health, particularly for vulnerable populations. This research delves into the proteomic analysis of hypervirulent L. monocytogenes strains from meat-based foods to better understand their pathogenicity, virulence, and adaptation under stress conditions. We used immunoproteomics and bioinformatics to characterize the immunogenic protein profile of an ST7 strain linked to severe listeriosis outbreaks in central Italy (Chiaverini et al., 2021). Our findings showed that L. monocytogenes produces several immunogenic proteins in response to mild acid, high osmolarity, and low temperature stress conditions (D"Onofrio et al., 2022). Six specific proteins were identified, related to cell motility (FliM, lmo0723), carbohydrate metabolism (PdhB), oxidative stress (TrxB, lmo2679), and DNA repair (lmo0132). To further explore L. monocytogenes' stress response, we analyzed the whole proteomes of two different L. monocytogenes 1/2a strains: 268M "ATCC BAA-679TM" as the reference strain and 3178-CB-2018, a wild type from pork sausage. We identified 265 immunogenic proteins, with gene enrichment analysis confirming the activation of virulence and pathogenesis pathways under stress conditions. A subsequent study focused on a highly virulent strain (ST155) from a 2022 listeriosis outbreak in Italy, using a quantitative proteomic approach. This strain was cultivated under control conditions (C1) and conditions simulating the frankfurter matrix (C2). The analysis identified potential immunogenic targets, including lmo0723 proteins previously associated with the ST7 strain, and additional targets linked to flagellar assembly, surface interactions, host cell entry, and adhesion. Overall, understanding the adaptive mechanisms of L. monocytogenes can provide insights into new biomarkers for identifying strains in food products, thereby enhancing food safety and public health. Chiaverini, A., Guidi, F., Torresi, M., Acciari, V. A., Centorotola, G., Cornacchia, A., Centorame, P., Marfoglia, C., Blasi, G., Di Domenico, M., Migliorati, G., Roussel, S., Pomilio, F., & Sevellec, Y. (2021). Phylogenetic Analysis and Genome-Wide Association Study Applied to an in microbiology, 12, 750065. https://doi.org/10.3389/fmicb.2021.750065 Italian Listeria monocytogenes Outbreak. Frontiers D'Onofrio, F., Visciano, P., Krasteva, I., Torresi, M., Tittarelli, M., Pomilio, F., Iannetti, L., Di Febo, T., Paparella, A., Schirone, M., & Luciani, M. (2022). Immunoproteome profiling of Listeria monocytogenes under mild acid and salt stress conditions. Proteomics, 22(18), e2200082. https://doi.org/10.1002/pmic.202200082 298

299

P-I-0152 Optimizing data-independent acquisition for the Orbitrap exploris 240: enhanced proteome coverage and quantitative accuracy and precision Reynaldo Melo1, Bianca Pauletti1, Jackson Gabriel Miyamoto1, Adriana Paes Leme1 1Brazilian Center for Research in Energy and Materials - CNPEM, Brazillian Biosciences National Laboratory - LNBio, Campinas, Brazil Proteomic analysis through data-independent acquisition (DIA) has advanced significantly, driven by improvements in instrumentation, data analysis, and acquisition strategies. These advancements have enabled deep proteome coverage and enhanced quantitative accuracy and precision. However, achieving such high metrics requires careful optimization of liquid chromatography (LC) and mass spectrometry (MS) methods. In this study, we optimized DIA parameters for the new Orbitrap Exploris 240 and evaluated the performance of various acquisition strategies. We defined the LC elution peak width using a 124-minute gradient, from 7 % to 45 % of acetonitrile concentration, in a 50-cm analytical column and optimized MS parameters, including MS resolution, isolation width, and m/z range, to achieve a median of at least 10 points per peak. Additionally, we tested dynamic and automatic injection times (IT) and various MS2 AGCs in the SWATH-MS acquisition strategy, followed by identification and quantification tests using different HEK cell lysate peptide amounts. Further, library generation with gas phase fractionation (GPF) and staggered window approaches were evaluated, and the optimized methods' quantitative merits were assessed using a complex mixture of proteomes from three species with known fold change (FC) values. Identification and quantification were performed in Spectronaut with default parameters. Initial tests achieved a median LC peak width of 17 seconds, suggesting that an isolation window of 16 m/z with 15k MS2 resolution in the 400-1000 m/z range resulted in a median of 10 points per peak, suitable for quantitation. Dynamic IT did not significantly improve the identification rate compared to auto mode, nor did varying MS2 AGC between 300% and 800%. Injecting peptide amounts from 0.14 µg to 1 µg resulted in identifying 5311 to 6879 protein groups (PGs), with a plateau near 0.7 µg. The spectral library created from GPF identified 2495 more precursors and 697 more PGs compared to a library- free strategy. The staggered window approach for GPF further increased precursor identifications by 2.36% compared to SWATH-MS, although it did not yield significant quantitative differences. Our optimized methods identified over 8000 protein groups from human HEK293 cells, achieving a median coefficient of variation (CV) of 5.0%. The complex proteome mixture analysis showed absolute median deviations of 0.51 and 0.13 from expected FC values of 2 and 0.5, respectively. These results demonstrate a deep proteome coverage, quantitative accuracy, and precision achievable with the optimized methods for the Exploris 240 mass spectrometer. 299

301

P-I-0153 Application of targeted proteomics for quantification of protein targets to support PK/PD-modelling and TMDD de-risking of a bispecific agonistic antibody Sebastian Malchow1, Hannah Joerissen1, Stephan Menz1 1Bayer Aktiengesellschaft, Research & Development, Drug Metabolism & Pharmakokinetics, Wuppertal, Germany A bispecific agonistic antibody has been designed to facilitate tumor-selective activation of protein target 1 (PT1) by specifically binding protein target 2 (PT2) expressed on cancer-associated cells (CAC). The bispecific binding mode is designed to increase intra-tumoral PT1 agonism via PT2-mediated receptor hyperclustering while reducing systemic PT1 activation. Pharmacokinetic (PK) studies in non-tumor bearing mice showed strong non- linear pharmacokinetic behavior as it is often observed for antibodies binding to membrane targets. Prediction of human PK for molecules exhibiting this phenomenon referred to as targeted-mediated drug disposition (TMDD) can be very challenging and relies on quantitative knowledge of target expression across species. In the case of PT2 binding, strong PK non-linearity was not originally expected in non-tumor bearing mice where CACs are absent. However, in addition to specific expression on CACs, PT2 has also been described to be expressed in skin, which is a possible explanation for the observed non-linear pharmacokinetic behavior in mice. Therefore, we developed an LC-MS based method to quantify the amount of PT2 in skin samples from different mouse strains, cynomolgus monkey and human. Absolute quantification of PT2 using stable isotope labeled peptides as surrogate standards gave inconclusive results, so we resorted to relative target quantification in skin samples. Although not quantitative, this approach helped to guide the need and choice of animal models for further pharmacokinetic studies and allowed to de-risk the prediction of TMDD in humans. In addition, we aim to develop an absolute quantitative method for the quantification of PT2 in murine and human tumor samples to support quantitative PK/PD modeling approaches. As alternative approach to using peptides as surrogate standards, we will use stable isotope labeled and unlabeled recombinant proteins for the second attempt. 301

302

P-I-0154 A multi-year longitudinal harmonization of quality controls in mass spectrometry proteomics core facilities Cristina Chiva1, Roger Olivella1, An Staes2, Teresa Mendes Maia2, Francis Impens2, Simon Devos2, Christian Panse3, Karel Stejskal4, Karl Mechtler4, Thibaut Douché5, Bérangère Lombard6, Damarys Loew6, Mandy Rettel7, Dominic Helm8, Andrea Schuhmann9, Anna Shevchenko9, Mariette Matondo5, Paolo Nanni3, Eduard Sabidó1 1Centre de Regulació Genòmica, Barcelona Institute of Science and Technology (BIST), Barcelona, Spain 2Ghent University, Center for Medical Biotechnology, Department of Biomolecular Medicine, Ghent, Belgium 3ETH Zürich, Functional Genomics Center, Zurich, Switzerland 4Vienna Biocenter (VBC), IMBA Institute of Molecular Biotechnology of the Austrian Academy of Sciences, Vienna, Austria 5Institut Pasteur, Mass Spectrometry Proteomics, Paris, France 6Institut Curie, PSL Research University, Centre de Recherche, CurieCoreTech, Paris, France 7European Molecular Biology Laboratory (EMBL), Proteomics Core Facility, Heidelberg, Germany 8German Cancer Research Center (DKFZ), Proteomics Core Facility, Heidelberg, Germany 9Max Planck Institute, Molecular Cell Biology and Genetics, Dresden, Germany Introduction: Common quality control procedures in proteomics core facilities ensure data quality, reproducibility, and comparability, benefiting proteomics infrastructures, the scientific community, and their users. The Proteomics Working Group of Core for Life (https://coreforlife.eu) conducted a four-year harmonization study to enhance quality control comparability, identify community reference values, monitor intra-laboratory performance, and find improvement opportunities. Unlike standardization, harmonization does not impose uniform methods but agrees on practices for data collection. This flexible approach accommodates diverse setups and instrumentation, overcoming the challenges faced by many standardization efforts. Methods: Eight different proteomics core facilities (CRG, EMBL, FGCZ, Institute Curie, IMP, Institute Pasteur, MPI, and VIB) have incorporated 30 LC-MS systems corresponding to 5 types of mass spectrometer (Orbitrap Eclipse, Fusion Lumos, Q Exactive HF, Q Exactive HF-X, and Exploris 480) and 4 liquid chromatography systems (UltiMate 3000 RSLC, EASY-nLC 1200, ACQUITY UPLC M-Class, Evosep One) into an harmonization study that spanned over four years (2019-2022). A commercially available QC4Life standard has been developed and analyzed with each LC-MS setup using a DDA method with HCD fragmentation (and ETD, EThcD and CID when available) and a flexible chromatographic gradient (30-60 min). The QCloud platform was established as a common software platform to evaluate the results and minimize data analysis heterogeneity. Results: To improve quality control consistency, we first established a common standard using a sample of 25 ng K562 cells proteome extract spiked with an isotopologue mix at five different concentrations, now available as the QC4Life Standard Mix (Promega, CS30240). We also agreed on using QCloud for automated, reproducible, and centralized analysis of quality control raw files from various instruments across participating sites. Over four years, 9054 raw files were generated, yielding 826,696 quality control data points from 109 parameters per file. Each laboratory accessed its own and other sites' data through QCloud, promoting discussions to address performance differences. This collaboration instruments or misconfigurations, leading to reduced variation in parameter measurements across all instruments. We established average reference values and ranges for each quality control parameter by instrument type. These values, reflecting diverse configurations across Core for Life laboratories, are available in the QCloud interface to help other laboratories assess their instruments, methods, and standard control samples. like under-performing identified issues 302

303

P-I-0155 Mass spectrometry acquisition methods for subcellular spatial proteomics: a direct comparison of resolution and reproducibility Kieran McCaskie1, Charlotte Hutchings1, Yong-In Kim1, Eneko Villanueva1, Renata Feret1, Lisa Breckels1, Mike Deery1, Kathryn Lilley1 1University of Cambridge, Biochemistry, Cambridge, United Kingdom Compartmentalisation of eukaryotic cells into organelles allows separate biochemical processes to occur simultaneously. The subcellular localisation of a protein influences its function by defining both the surrounding physical conditions and the biomolecules available for interaction. Movement of proteins between organelles is essential for normal cellular activity, allowing multiple roles to be fulfilled in different environments, and mislocalisation is linked with an increasingly wide range of diseases. The optimisation of techniques for profiling the spatial proteome is imperative, to gain new insight into both normal cellular function and responses to perturbation, especially in cases where overall protein abundance stays constant but localisation changes. The LOPIT-DC protocol is able to map the subcellular localisation of thousands of proteins in a single experiment by utilising differential ultracentrifugation fractionation, before isotope tagging, DDA mass spectrometry and correlation profiling based analysis. Within the same experimental pipeline, we compare the established DDA- MS based LOPIT-DC method to a DIA-MS approach, for a direct comparison of resolution and reproducibility. Initial analysis of the resolution suggests the DIA workflow can identify significantly more proteins, with the DDA workflow generally showing tighter organelle marker clustering after PCA analysis. Both methods are effective at distinguishing proteins from nuclear, cytosolic, and mitochondrial compartments, as well as key secretory pathway organelles including the Golgi apparatus and endoplasmic reticulum. The DIA workflow reduces the amount of mass spectrometry instrument time required, since fewer fractions are run. In terms of sample preparation, a lower number of initial cells can be used with the DIA workflow as the amount of protein mass required in each fraction for TMT labelling protocols is no longer a limiting factor. 303

304

P-I-0156 Quantifying 500 proteomics samples per day with ZT Scan DIA Dietrich Merkel1, Anjali Chelur2, Stephen Tate2, Claudia Alvarez2, Gordana Ivosev2, Yves Le Blanc2, Nic Bloomfield2 1AB SCIEX Germany GmbH, Application Support, Darmstadt, Germany 2SCIEX Canada, Concord, Canada High throughput proteomics requires advanced instrumentation coupled with chromatographic improvements to push the boundaries of coverage depth with quantitative accuracy. Shorter gradient MS-based proteomics requires lower cycle times to achieve results that strike a balance between the number of identifications and the precision of measurements. This creates a need for a data-independent acquisition (DIA) mechanism that can run scans at a higher rate than Zeno SWATH DIA without compromising on specificity. We present ZT Scan DIA, a Zeno trap-enabled scanning DIA workflow with the ability to run at faster scan rates to enable both higher identifications and quantitative precision than conventional Zeno SWATH DIA. In ZT Scan DIA, the quadrupole isolation window slides along the mass range of interest at an MS/MS acquisition rate of 640 Hz. Each fragment is visible when the trailing edge of the quadrupole passes the precursor m/z and disappears when the leading edge of the quadrupole passes it. Thus, fragments can be mapped to the precursors bringing in an additional dimension along with the m/z, intensity and RT dimensions. ZT Scan DIA does not require emptying the collision cell, thereby saving on cycle time. We explored a high-throughput combination of 500 SPD with ZT Scan DIA where 3 modes of quad scan settings broadly categorized for different peak widths were available. The 500 SPD setting gives a peak width half height of 0.9 sec, making it a good candidate for the scan speed setting of 750 Da/s (5 Da Q1 window size). A corresponding variable window Zeno SWATH DIA method with equivalent cycle time and MS/MS accumulation time was acquired for comparison Using different loadings of commercially available K562 digest and a research-modified Zeno trap-enabled QTOF system, we demonstrate that higher specificity would yield better gains in terms of protein numbers over Zeno SWATH DIA because shorter gradients require optimization of Zeno SWATH DIA windows, either widening the MS/MS windows or cutting down on the scan range to maintain the cycle time. 304

305

P-I-0157 A mass-spectrometry-based approach for the detection of gene fusion products Frauke Textor1,2, Larissa Meyer1,3, Bettina Wehrle1, Stefan Tholen1, Matthias Fahrner1,4, Cornelius Miething2, Oliver Schilling1,4 1Instituts for Surgical Pathology, University of Freiburg, Freiburg, Germany 2Department of Internal Medicine I, Faculty of Medicine and Medical Center, Freiburg, Germany 3Faculty of Biology, Albert-Ludwigs-Universität Freiburg, Freiburg, Germany 4German Cancer Consortium (DKTK) and Cancer Research Center (DKFZ), Freiburg, Germany Background: Multiple gene fusions have been associated with tumor development and prognosis and in some cases are characteristic for certain malignancies. These gene fusions can result in hybrid proteins that may contain functional domains of the individual fusion partners. Molecular diagnostics of gene fusion events mainly relies on nucleic acid-based sequencing approaches and can be complemented by immunohistochemical (IHC) staining. However, IHC is antibody-dependent and serves only as a surrogate approach by detecting increased levels of either fusion proteins or upregulation of downstream signaling. Additionally, many anticancer drugs target proteins, so a deep insight into these hybrid proteins would enhance patient-specific molecular diagnostics. Therefore, we set out to establish an antibody-independent LC-MS/MS-based approach for directly detecting the fusion proteins. Methods: In a proof-of-concept experiment, three cell lines with stable gene fusion mutations were cultured and the hybrid proteins were detected by western blot. The fusion proteins were purified and digested using the S- Trap protocol. The resulting peptides were analyzed on a TIMS-TOF flex mass spectrometer using data- independent acquisition (DIA) in parallel accumulation-serial fragmentation (PASEF) mode. The results were analyzed using DIA-NN and compared to a human spectral library that included amino acid sequences of the hybrid proteins. Results and Outlook: We detect hybrid proteins in all three cell lines using western blot as well as our LC-MS/MS- based approach. Furthermore, in our LC-MS/MS approach we detect peptides that cover the actual gene fusion site. Hence, our approach provides an antibody-independent alternative for the direct assessment of gene fusion events and paves the way for more detailed molecular characterization of gene fusion-driven malignancies. Future experiments aim to develop a spectral library based on previously described and most common gene fusions. This library will be utilized to investigate hybrid proteins in formalin-fixed paraffin-embedded (FFPE) samples from patients with known gene fusion mutations. Establishing a robust workflow for detecting hybrid proteins in cancer patients could enhance the molecular diagnostics of gene fusion-driven malignancies, thereby improving personalized cancer treatment. 305

306

P-I-0158 Elucidation of Caenorhabditis elegans germline protein-protein interaction networks by in vivo photo- crosslinking Fränze Müller1,2, Sowmya Sivakumar Geetha2, Karl Mechtler1, Verena Jantsch2 1Research Institute of Molecular Pathology (IMP), Protein TechHub, Vienna, Austria 2Max Perutz Labs Vienna, Department of Chromosome Biology, Vienna, Austria Understanding protein interaction networks in their native environment is crucial for elucidating biological roles. Incorporating unnatural amino acids (unAAs) labeled with fluorescent dyes enables direct protein labeling in live cells, facilitating the study of protein context. Meanwhile, chemical crosslinking allows the identification of proximal amino acids in native protein structures, aiding in structural biology. However, system-wide crosslinking poses challenges in peptide identification due to complex sample backgrounds. In this study, we propose a novel approach combining biolabeling and crosslinking mass spectrometry (CLMS) to study protein-protein interactions (PPIs) in live Caenorhabditis elegans. We employ photoactivatable amino acids for crosslinking, enabling precise timing and unaffected translation. This innovative pipeline promises to generate a comprehensive PPI network in a living organism, enhancing in vivo CLMS depth and capturing transient PPIs. By utilizing this strategy, we aim to establish protocols for studying PPI networks and protein complexes in C. elegans, particularly in isolated gonad nuclei, and extend this to the entire germline. The integration of biolabeling, CLMS, and pull-down assays offers a multifaceted approach to unraveling complex protein interactions within a live organism, paving the way for deeper insights into cellular mechanisms. 306

307

P-I-0159 Broadening the horizon of MS optimization for enzymatic digestions in proteomics Kinga Nagy1,2, Péter Sándor1, László Drahos1, Ágnes Révész1 1MS Proteomics Research Group, HUN-REN Research Centre for Natural Sciences, Budapest, Hungary 2Hevesy György PhD School of Chemistry, Eötvös Loránd University, Faculty of Science, Institute of Chemistry, Budapest, Hungary Tandem mass spectrometry is an essential tool for protein analysis. Proteins are most often identified based on MS/MS spectra of their peptides produced by tryptic digestion, which are compared with theoretical spectra of peptides based on a protein database. The reliability of the peptide spectrum match is characterized by a score. The identification depends on the experimental parameters and the software used for the evaluation. Many times, instead of trypsin, enzymes with other specificity are preferable, so their use is more and more widespread. However, the optimization of MS/MS parameters is still focused on tryptic peptides. Even though collision energy (CE) has a great influence on the quality of MS/MS spectra, CE optimization in detail for digestions with alternative enzymes has not yet been carried out for a large number of non-tryptic peptides. The objective of this study was to optimize the MS-based identification of non-tryptic pep-tides by examining the reliability of database search-based identification of peptides digested by alternative enzymes as a function of the collision energy used during MS/MS measurement. We focused on four enzymes typically used in proteomics (Asp-N, Arg-C, chymotrypsin, Glu-C,) and examined trypsin as reference. NanoLC-MS/MS measurements were performed using varied CE settings to determine the optimal value of the CE for a large number of non-tryptic peptides. Human blood plasma and MS compatible human protein extract were used as complex search engines. Results indicate that the enzyme specificity has a significant effect on the optimal collision energy, and the observed differences can be explained by the fragmentation characteristics of the peptides obtained by different enzymatic digestions. While the peptides obtained with Arg-C behave analogously to tryptic peptides, in the case of Asp-N, chymotrypsin and Glu-C the optimal collision energy is 5-15 eV lower. The optimum does not depend significantly on the sample type and the database search engine used. Based on our results, a more efficient MS- based method is proposed for the investigation of non-tryptic peptides by adjusting the collision energy based on the type of enzyme. When determining its performance, in the case of alternative enzymes the number of peptides and the sequence coverage increased significantly compared to the method optimized for tryptic peptides. Financial support from the NKFIH (grant FK-138678) is gratefully acknowledged. samples. The evaluation was carried out with Byonic and Mascot 307

308

P-I-0160 Automated high-throughput proteomic analysis of stored blood cells from a large cohort of non-domestic felids Kevin L. Schauer1, Kevin Yang1, Nicolaie Eugen Damoc2, Sega Ndiaye3, Amirmansoor Hakimi1, Lily A.B. Parkinson4 1Thermo Fisher Scientific, San Jose, CA, United States 2Thermo Fisher Scientific, Bremen, Germany 3Thermo Fisher Scientific, Courtaboeuf, France 4Brookfield Zoo, Chicago, IL, United States Introduction: Blood transfusions occur infrequently in non-domestic cats housed at zoos, but remain a necessary, life-saving treatment. For this reason, long-term blood storage for non-domestic felid species is required. Little is known about how long blood can be stored from these species before degradation limits its clinical efficacy, however. Non-targeted proteomics provides a sensitive, unbiased method to look at molecular changes in blood cells that result from prolonged storage. Here, over 500 blood cell samples were analyzed using an automated, high-throughput LC-MSMS and data analysis workflow. Methods: Blood samples were obtained from 135 non-domestic cats, consisting of 18 different species, housed at US zoos. Fresh, anticoagulated, whole blood samples were aliquoted and stored at 4⁰C in a clinical blood bank refrigerator for 0, 7, 14, or 28 days, after which the red blood cells were pelleted. Pelleted blood cells were then prepared for bottom-up proteomic analysis using the Thermo Scientific™ AccelerOme™ that allowed for hands- off, automated sample preparation of over 500 individual blood cell samples analyzed in triplicate resulting in 1600 injections. A high-throughput (85 samples per day) method on a Thermo Scientific™ Vanquish™ Neo LC at capillary flow rates was used for peptide separation before being analyzed by a Thermo Scientific™ Orbitrap™ Astral™ mass spectrometer in data independent (DIA) acquisition mode. Data analysis was automated using the Thermo Scientific™ Ardia™ Platform via the connection of acquisition using the Thermo Scientific™ Xcalibur™ software to processing by the Thermo Scientific™ Proteome Discoverer™ software. Upon the completion of each injection, result files were automatically uploaded to the Ardia platform by the Xcalibur acquisition software, and analysis by the Proteome Discoverer software was automatically triggered. Proteome Discoverer searches leveraged the CHIMERYS™ algorithm running locally on the Ardia server for both spectral identification and quantification. Preliminary Data: Each injection resulted in the identification of approximately 1200 to 1500 proteins and 8000 to 12,000 peptide groups. The protein and peptide identification rates were approximately 2.5-fold and 5-fold higher, respectively, than a pilot study that was performed on a Thermo Scientific™ Orbitrap Exploris™ 480 mass spectrometer, despite using a method that was nearly twice as long. The variation in the number of identifications observed here can be attributed to differences in the completeness of the proteomes that were available for each species. However, intensities and identification rates across samples derived from the same species were very consistent, highlighting the robustness of the sample preparation and LC-MS setup used. Label-free relative quantification revealed few changes in protein abundance across storage times, which was consistent with limited metabolic activity due to storage at 4⁰C. Interestingly, numerous peptides showed changes in abundance as storage lengths increased, which could indicate protein degradation or chemical changes that occur during storage. 308

309

P-I-0161 Label-Free quantitation with high accuracy and precision on a single cell scale and beyond with Orbitrap Astral mass spectrometer Anna Pashkova1, Jenny Ho2, Sonja Radau3, Roxana Martinez-Pinna3, Shio Watanabe4, Pedro Navarro1, Florian Marty5, Tabiwang Arrey1, Nicolaie Eugen Damoc1 1Thermo Fisher Scientific, Bremen, Germany 2Thermo Fisher Scientific, Hemel Hempstead, United Kingdom 3Thermo Fisher Scientific, Dreieich, Germany 4Thermo Fisher Scientific, Yokohama, Japan 5Thermo Fisher Scientific, Reinach, Switzerland Analysis of protein expression at the single-cell level, allowing to reveal subtle cellular heterogeneity, gained increasing importance over the last 5 years. Nanoflow LC coupled with high-resolution mass spectrometry is the method of choice for proteomics studies, due to its sensitivity, dynamic range, and throughput. In this study, we assessed the accuracy and precision of quantitation over the wide range of sample loads, for low-input samples, from as low as 10 pg up to 500 ng per injection, at a less than 20 min gradient length. For 10 pg-20 ng loads, FAIMS was used to decrease interference from background ions. Starting from 20ng and up MS data acquisition without FAIMS becomes more beneficial for both peptide and protein identifications. A wide (up to 40 Th) isolation window DIA methods with long (up to 100 ms) Astral ion injection times were employed to maximize the proteome coverage for low-input samples. With these settings, more data points per LC peak are acquired at MS1 level with the Orbitrap analyzer, ensuring high quantitation accuracy and precision, and MS2 information is used only for identification purposes. 100 ms maximum injection time with a 500% normalized AGC target, set at the Orbitrap analyzer, allowed delivery of a sufficient number of ions to be transferred downstream to the Astral analyzer, to ensure reliable quantitation. Linear correlation (up to R2=0.99) was observed for analyte peaks measured with the Orbitrap analyzer with FAIMS, at sample loads from 10 pg to 500 ng, demonstrating a wide dynamic range of quantitation at the scale spanning over 4 orders of magnitude. A systematic study of parameters critical for high quantitation accuracy was performed, and the best MS methods were selected. Data was analyzed using SpectronautTM 18 software (Biognosys) with directDIA, library-free processing. As a result, ~3400 protein groups, >13,000 peptides identified at 10 pg/injection, 82 to 90% of them quantified in one of the 3 ratios of proteomes; ~9300 protein groups, ~63,000 peptides identified at 250 pg/injection (single-cell level), with nearly 98% of them quantified. At 500 ng/injection, over 15.500 protein groups and >230,000 peptides were identified and quantified. Quantitation accuracy at the single-cell level loads (10-250pg) was very close to the expected values, with less than 10% relative error. Quantitation accuracy remained high throughout the entire range of sample loads, regardless of the number of data points per peak. Single-cell level LFQ was performed in three different laboratories, showing similar results, and demonstrating that high accuracy and precision of quantitation can be achieved on a routine basis. Fig. 1 (optional) 309

310

P-I-0162 Refining advanced sample preparation and nanoLC-MS/MS analysis techniques for single-cell proteomics Pauline Perdu-Alloy1,2, Charline Keller1,2, Jeewan Babu Rijal1,2, Charline Fuseau3, Delphine Rolland1,2,3, Fornecker Matthieu1,2,3, Laurence Sabatier1,2, Christine Carapito1,2 1Infrastructure Nationale de Protéomique ProFI, FR2048, Strasbourg, France 2Laboratoire de Spectrométrie de Masse BioOrganique, IPHC (UMR 7178), CNRS, Université de Strasbourg, Strasbourg, France 3Institut de Cancérologie Strasbourg Europe, Strasbourg, France Introduction: Recent advancements, such as the integration of TIMS coupled to a high sensitivity mass spectrometer (TimsQTOF) with automated processes, have significantly enhanced the precision and efficiency of single-cell proteome analysis. In addition, the control of ion populations is advancing in tandem with instrumental innovations, enabling the creation of new precursor scan designs and the management of their fragmentation concurrently with their accumulation. Notably, the emergence of DIA-PASEF (Data-Independent Acquisition - Parallel Accumulation Serial Fragmentation) technology allows for the successive isolation and fragmentation of all precursors within selected mass windows, significantly enhancing analytical efficiency and depth. Methods: After five washes with frozen PBS, freshly prepared HeLa cells were diluted to a final concentration of approximately 200 cells/µL. Sorting, lysis, and digestion were performed using the CellenONE instrument (Cellenion, Lyon). To optimize the sample preparation, various parameters were adjusted: the protein-to- protease digestion ratio (1:1, 1:10, 1:20, and 1:50), digestion times (30 minutes, 1 hour, 1.5 hours, and 2 hours), robot humidity levels (75%, 85%, and 90%), and the final volume of peptide recovery (1 µL, 2 µL, 3 µL, and 3.5 µL). Additionally, different sample plates, namely the proteoCHIP and EVO96, were evaluated as targets for the sample preparation workflow. Peptide mixtures were separated on an Aurora C18-RP 100Å (75µm x 250mmx1.5µm) and C18-RP 120Å (75µmx50cmx1.7µm) (IonOpticks) columns using a nanoElute 2 coupled to a timsTOFUltra (Bruker Daltonics) mass spectrometer. Separation was performed on a gradient from 5 to 35% ACN in 0.1% FA over 30min and 14min at 0.25µL/min. Data Independent Acquisition – PASEF (DIA-PASEF) and Slice- PASEF were performed followed by a comparative analysis to assess the performance of the direct-DIA against the spectral library processing using DIA-NN. Results: Following the extensive method optimizations, more than 3500 proteins could be identified in single HeLa cells. We observe a non-significant difference between data processing with and without spectral libraries. Building on these advancements, our focus is now on demonstrating that the optimized workflow can be successfully applied to characterize the cellular heterogeneity in B-cell mantle lymphomas. Promising preliminary results allow us to aim at reaching a proteome depth of 1200 proteins in single B-lymphocytes. Conclusion: While still requiring method refinements at all steps of the workflow, SCP starts delivering biological valuable results and the proteome coverage reached is compatible with its application for a deeper understanding of to diseases characterized by cellular heterogeneity. 310

311

P-I-0163 Multiple Accumulation Precursor Mass Spectrometry (MAP-MS) extends precursor dynamic range and improves peptide detection in data-independent analysis on an OrbitrapTM platform Teeradon Phlairaharn1,2, Brian Searle2 1Technical University of Munich, Freising, Germany 2Ohio State University, Department of Biomedical Informatics, Columbus, OH, United States In this study, we introduced a multiplex acquisition strategy at the precursor level, called "Multiple Accumulation Precursor Mass Spectrometry" (MAP-MS). This technique enhances the dynamic range by accumulating multiple precursor mass ranges while the Orbitrap (OT) analyzes MS/MS spectra without increasing the cycle time. Our results confirm that peptide detection in OT-based data-independent acquisition (DIA) analysis can be improved by extending the precursor dynamic range using MAP-MS. This is achieved through a software package that combines precursor and fragment information, demonstrating that MAP-MS can serve as an alternative procedure for precursor analysis in OT-based mass spectrometers. By analyzing 125 HeLa tryptic peptides, we identified 99,056 peptides and 7,721 protein groups using MAP-MS, compared to 82,336 peptides and 6,736 protein groups identified with the standard DIA method on an Orbitrap ExplorisTM 480 mass spectrometer with a 110-minute gradient. Additionally, we introduced a true positive validation method based on gas-phase fractionation (GPF). This method can be applied to verify post-acquisition method optimization using retention times (RT) from GPF-DIA as a reference. 311

312

P-I-0164 Exploring the influence of the number of Ions on the peptide fragmentation spectra Teeradon Phlairaharn1,2, Denis Chernyshev3, Brian Searle2 1Technical University of Munich, Freising, Germany 2Ohio State University, Department of Biomedical Informatics, Columbus, OH, United States 3Thermo Fisher Scientific, Bremen, Germany Introduction: Single-cell and low-input proteomics leverage powerful new tandem mass spectrometry (MS/MS) tools for exploring cell biology at a granular level. While improvements in instrumentation and sample preparation methodologies have made single-cell proteomics attainable by many labs, consistently detecting peptides in single-cell experiments remains a challenging endeavor. One way to improve our ability to detect peptides in low-input experiments is to use reference spectra acquired from bulk samples with more input material. Recent reports have suggested that the quality of MS/MS spectra may change as input material decreases, and in this work, we sought to determine how many ions are required for an OrbitrapTM (OT) and AstralTM (AST) mass analyzer to produce internally consistent spectra using the same instrument. Preliminary Data: Comparisons of low-input and bulk proteomics spectra rely on using different acquisitions for each sample type. In some cases, these measurements have been performed in different labs on instruments with different configurations and tunings. Even when performed on the same instrument, changes in sample complexity and composition affect what background ions can interact with peptides, potentially changing ion interference. For this work, we designed a DIA method to assess the influence of the number of ions on the quality of MS/MS spectra in an OT within a single injection. Rather than modulate sample concentrations to change the number of ions in the IRM, we adjusted AGC to mimic low-input level measurements by limiting the number of ions. This gave us a unique platform to test the effects of limited ions on the quality of MS/MS, where higher AGC spectra act as controls. Integrating the precursor isolation window from preceding precursor spectra, we calculated the number of ions within each MS/MS measurement, given the ion injection time. Within each packet, we calculated the spectral contrast angle between the optimal MS/MS (1000% AGC) and MS/MS with fewer ions limited by lower AGC targets. We observed that the quality of MS/MS spectra measured in the OT was tied to the ion count of the highest fragment, suggesting that the relationship is governed by ion statistics, rather than total ion count. We found the OT required approximately 150 ions in the OT as the predominant fragment to produce >80% spectral angle. This corresponded to approximately 1400 total precursor ions, equivalent to only 1.4% of the normalized AGC target in MS/MS analysis in most low-input DIA experiments. Consequently, we suggest that optimal acquisition parameters e.g. longer IIT and higher OT resolving power can improve MS/MS spectra quality in single-cell measurements and eliminate differences, observed from the bulk analysis. 312

313

P-I-0165 Advancement in nanoparticle-based proteomic analysis of whole blood obtained from various dried blood spot collection devices Maik Pruess1, Maedeh Zamani2, Xiaoyuan Zhou2, Shao-Yung Chen2, Aaron Steven Gajadhar2 1Seer Bio GmbH, Field Application Scientist, Frankfurt, Germany 2Seer, Inc., Strategic Applications Marketing, Redwood City, CA, United States Microsampling is a methodology for collection and analysis of low-volume biological samples, typically under 100 μL. Dried blood spot (DBS) collected through microsampling affords several advantages compared to conventional venipuncture blood draw: diminished invasiveness, enhanced convenience, and cost-effectiveness. It allows collecting whole blood samples for widespread application in clinical trials, therapeutic drug monitoring, pharmacogenomics and disease diagnosis. While microsampling of DBS has demonstrated significant advantages in proteomics studies, the depth of protein coverage remains limited owing to the high dynamic range of proteins present in whole blood. Also, the biological differences, particularly the differences in hematocrit level, along with low sample volume may cause variations in sample quantification and analysis, which ultimately results in lack of reproducibility of the measurements. Whole blood samples were obtained intravenously, alongside capillary blood collected via finger-prick from the same subjects to saturate microsampling devices. A simple extraction method was utilized to extract whole blood from microsampling devices, including MitraTM devices (30uL), CapitainerTM devices (10uL), and WhatmanTM paper. The extracted proteins underwent Seer's ProteographTM XT workflow, followed by mass spec analysis using Orbitrap ExplorisTM 480 and AstralTM. Incubation of the protein extracts from all three devices with the nanoparticles in ProteographTM XT workflow significantly enhanced the number of protein groups detected, compared to the direct processing of the samples. Introducing Seer"s nanoparticles to the extracts from all DBS devices resulted in detection of around 3,000 protein groups on Exploris 480, with additional protein coverage on Astral. In the absence of nanoparticles, the direct processing of the extract from all three microsampling devices resulted in identifying only around 1,000 proteins. The results also exhibited low coefficients of variation (CVs) below 10% across multiple subjects, with similar protein group profiles. The capacity for unbiased proteomic studies at such depth, provided by ProteographTM XT workflow offers a distinct advantage for early biomarker discovery and health status monitoring. Considering the annual collection of millions of dried blood spot (DBS) samples, Seer"s technology can be employed as a powerful tool for accurate and large-scale proteomic studies using DBS samples. 313

314

P-I-0166 hadexversum: tools for HDX-MS analysis Weronika Puchała1, Michał Burdukiewicz2, Michał Dadlez1 1PAS Institute of Biochemistry and Biophysics, Warsaw, Poland 2Medical University of Białystok, Białystok, Poland Hydrogen deuterium exchange (HDX) is a unique technique of protein structural studies as it provides insight into the dynamic properties of protein chains entangled in more or less stable spatial structures. The HDX-MS data is complex and it requires a dedicated solution. To answer the needs of the community, we developed a family of applications - hadexversum, personalizable workflows for multidimensional challenges. The very first created tool is HaDeX - a versatile software for processing, analyzing, and visualizing output data on the peptide level. The application provides a complete analytic workflow, with precise uncertainty calculations and reporting features recommended in the community guidelines. Various visualization methods (including novel ones) ensure in-depth data exploration. To expand the possible overview of the exchange phenomenon, we created HRaDeX. We utilized a well-known mathematical exchange model and proposed a unique method of uptake curve classification, using color code corresponding to the exchange pattern. Ultimately, the classification results are aggregated to the high- resolution level and presented on the 3D structure, enabling the correspondence of the spatial data with experimental results. As HRaDeX works on one biological state at a time, compaHRaDeX provides comparative analysis to ensure comfort of use. All applications are available as a web-servers, R-packages, and standalone software. The excessive documentation accompanies transparency as an open-source solution (https://github.com/hadexversum/). to ensure maximal them 314

315

P-I-0167 Advanced proteomics for understanding cellular health and disease by unveiling lysosomal protein interactions Ernst Heinz Pulido1, Emma Lundberg1 1Stanford University, Biophysics, Stanford, CA, United States Lysosomes are essential for degrading macromolecules, removing faulty organelles, and regulating cellular homeostasis, with dysfunction linked to severe diseases. This study uses advanced subcellular proteomics, organelle purification techniques, and structural biology to elucidate complex protein interactions within and around lysosomes. We will isolate intact lysosomes using lysosomal immunopurification (LysoIP) and apply membrane-penetrating and impermeable cross-linkers to stabilize protein interactions. This allows us to capture a detailed snapshot of these interactions as they occur naturally within the cell. The cross-linked proteins will be analyzed using high-resolution mass spectrometry to map interactions in the lysosomal lumen and on its membrane. Identified interactions will be verified and functionally investigated under different nutrient conditions to understand their regulatory mechanisms. Bioinformatics tools will be employed for data analysis and structural biology, providing a comprehensive view of lysosomal protein dynamics. Our research will focus on both the internal and external protein interactions of lysosomes, which are crucial for their role in maintaining cellular health. By integrating bioinformatics for structural biology, we will gain detailed insights into the 3D arrangements and interactions of lysosomal proteins. To verify the identified protein interactions and investigate their specific functions, we will use spatial proteomics, particularly fluorescence microscopy, to better understand their regulatory mechanisms. By exploring how these protein interactions change under varying nutrient levels, we aim to gain insights into how lysosomes help regulate the cell"s response to its environment. This interdisciplinary research will significantly enhance our understanding of lysosome functions, potentially leading to new treatments for diseases related to lysosomal malfunction and innovative strategies to maintain cellular health. Through this study, we aim to contribute foundational knowledge that will advance the field of cellular homeostasis and disease treatment. 315

316

P-I-0168 Enabling ultra sensitive, superior-throughput proteomics from data acquisition to data analysis Rui Qiao1, Stephen Tate2, Anjali Chelur2, Ihor Batruch2, Katherine Tran2, Lei Xin1, Haibo Bian1, Zia Rahman1, Zac Anderson1, Baozhen Shan1 1Bioinformatics Solutions Inc., Waterloo, Canada 2SCIEX Canada, Concord, Canada Challenges in quantitative accuracy for high-throughput mass spectrometry remain a concern as demands continue to increase for proteomics-based precision medicine and clinical studies. With a new wave of research shifting towards DIA strategies to increase proteomic coverage, depth, and reproducibility, larger amounts of data are generated as more information is acquired. Furthermore, successful quantitative measurements in complex samples require a great deal of method optimization to ensure critical factors, such as the number of experimentally measured points across an LC peak, are met. Ultimately, the throughput of current quantitative proteomics workflows is negatively affected by these factors to accommodate the longer times needed to collect and process large sets of data, or the quality and sensitivity of the data is sacrificed to maintain high throughput. Here, we present a full workflow solution to overcome these barriers and optimize the throughput of large scale, LC-MS/MS proteomics from data acquisition to analysis. First, we utilize an extremely fast, novel acquisition method (ZT Scan DIA), that requires minimal method development compared to existing DIA workflows. In this work, ZT Scan DIA exploits a continuous sliding quadrupole (Q1), at a rate of 640Hz, in combination with a research-grade Zeno trap enabled QTOF to simultaneously assign precursor masses to the MS/MS fragments. Furthermore, 20x more fragment ions can be detected with Zeno trap activation. Then, with the varying quadrupole scanning rate, a generally accepted range in m/z of 500 amu is used to cover the peptide parent masses, along with having the smallest quadrupole isolation window to maximize selectivity. In the second part of the workflow, we introduce a software platform to automate the analysis of ZT Scan DIA data. The software utilizes a new algorithm integrating DDA and DIA data analysis approaches to increase sensitivity and accuracy of identification and quantitation results. Specifically, a DDA approach is used for associating precursor masses to the MS/MS fragments generated by the Q1. Using data acquired by ZT Scan DIA for 50ng of K562 on a 5-min gradient, the software identified 29886 peptides and 3716 protein groups at 1% FDR with direct database search. Compared with data from the same sample acquired by Zeno SWATH DIA, ZT Scan DIA demonstrates improved identification rates by >15% on both peptide- and protein-levels contributing to the added Q1 dimension. In addition, the software displays the Q1 profile of each identified peptide in an interactive GUI so that results can be easily interpreted and validated. Visualization of the Q1 profile demonstrates the added ability to deconvolve co-eluting fragment ion signals which originate from native and modified peptides. Overall, ZT Scan DIA together with the customized software platform, provide an extremely sensitive, superior- throughput workflow for large-scale quantitative proteomics. 316

317

P-I-0169 Mass spectrometry as a quantitative methodology to assess the biodistribution of monoclonal antibodies Domenico Ravazza1, Sheila Dakhel Plaza1, Andrea Galbiati1, Andrea Ciamarone2,3, Samuele Cazzamalli1, Dario Neri4,5, Ettore Gilardoni1 1Philochem AG, Otelfingen, Switzerland 2Instituto Italiano di Tecnologia, Department of Computational and Chemical Biology, Genova, Italy 3University of Bologna, Department of Pharmacy and Biotechnology, Bologna, Italy 4Philogen S.P.A., Sovicille, Italy 5Swiss Federal Institute of Technology, Department of Chemistry and Applied Biosciences, Zurich, Switzerland Over the past thirty years, various naked and armed monoclonal antibodies (mAbs) have been clinically utilized as targeted therapeutics in the field of oncology. Understanding the quantitative biodistribution of mAbs is crucial to accurately evaluate the accumulation of the drug at the site of disease and in healthy tissues. In this study, we describe a hybrid platform based on affinity enrichment and LC-MS for the ex vivo quantification of tumor-targeting mAbs in cancer lesions and in healthy organs upon systemic administration in tumor-bearing mice. To reduce sample complexity and increase the analyte concentration, different enrichment modalities were evaluated and compared for their efficiency of antibody recovery. A homologous mAb with a different light chain isotype, a stable-isotope-labeled peptide, or a stable isotope-labeled protein were evaluated as putative internal standards. This was done by comparing their reliability in normalising errors introduced during the sample preparation or the LC-MS analysis, with the stable-isotope labelled protein outperforming the others. After enrichment and protein digestion, tryptic peptides were subjected to nanoHPLC-HRMS analysis on an Orbitrap Q-Exactive mass spectrometer coupled to an EASY nanoLC 1000 system via a Nano Flex ion source. Chromatographic separation was carried out at room temperature on an Acclaim PepMAp RSLC column (50 µm x 15 cm, particle size 2 µm, pore size, 100 Å). Different MS acquisition modes (i.e., DDA-top12, a SIM-dda and a PRM) were evaluated comparing the peptides" total signals and signals to noise ratio. As a case study, the ex vivo biodistribution of a fully human monoclonal antibody targeting the extra domain B of fibronectin in 129sv mice bearing teratocarcinoma tumor was carried out 24 h after a single dose administration. Data obtained with the mass spectrometer were compared to biodistributions performed in the same condition using radiolabelled antibody (i.e., 89Zr or 125I). 317

318

P-I-0170 LC-MS/MS-based structural glycoproteomics Joshua Maliepaard1, Weiwei Wang1, Karli Reiding1 1Utrecht University, Biomolecular Mass Spectrometry & Proteomics, Utrecht, Netherlands Mass spectrometry (MS) is a fundamentally important tool for proteomics, providing protein sequences, quantification, as well as detailed information on post-translational modifications (PTMs). However, mass alone is not sufficient to solve the challenging problem of isomerism, the situation in which biologically distinct analytes have different structures but the same mass. An example of this can be found in leucine and isoleucine, which share a mass but are distinct amino acids. Isomerism is a particular issue in the field of glycosylation analysis. Glycans ubiquitously cover our cells and proteins, often being the first barrier of inter-cellular communication and pathogen-host interactions, as well as being the most-accessible biomarker reservoir in a spatial sense. Glycosylation has long been studied by mass spectrometry, e.g., the glycans themselves after enzymatic release (glycomics), or the glycopeptides that result from proteolytic digestion (glycoproteomics). However, while glycans have a multitude of isomeric properties, including fucose position, sialic acid linkage and antennary branching structures, these are not readily determinable by MS. For released glycans, this can be partially overcome by orthogonal means, e.g., ion mobility, liquid chromatography (LC) using porous graphitic carbon, or chemical derivatization, but for glycopeptide analysis these techniques are not directly compatible. At the same time, only at the glycopeptide level can glycans reliably be positioned at given glycoproteins and glycosylation sites. As such, to enable structural glycoproteomics, we have developed a combination of MS/MS-based structural glycopeptide characterization and high-resolution nano-HILIC-LC (Figure 1). As we demonstrate on a growing number structurally-defined glycopeptide standards, oxonium ion ratios follow a distinct path through collision- energy-space depending on the structural characteristics (Maliepaard, et al., 2023, Anal. Chem.). While this itself can already be used to quantify isomer ratios from a mixture, it also allows the structural characterization of glycopeptide isomers that show chromatographic separation, for which we have optimized HILIC-LC. By applying the resulting LC-MS/MS methodology on complex samples such as human plasma and recombinant IgGs, we reveal, for instance, a distinct branch asymmetry present amongst the IgG subclasses: galactosylation of IgG1 dominantly occurs at the 6-branch, opposite to IgG2 for which it is the 3-branch instead. Structural glycoproteomics is expected to yield similarly remarkable observations for other proteins and their interactions, to lead to improved drug design, as well as to significantly deepen the pool of glycosylation-based biomarkers. Fig. 1 (optional) 318

319

P-I-0172 Enrichment of transmembrane protein identification in mass spectrometry with the BeatBox® sample homogenizer Julien Saint-Pol1, Katrin Hartinger1 1Université d'Artois, Blood-Brain Barrier Lab (BBB Lab, UR2465), Lens, France The identification of membrane proteins in mass spectrometry (MS) analyses requires the attention of researchers as pathophysiological regulators of signaling pathways or inter/extracellular exchanges, as pathological markers or as potential mediators of pharmacological vectorisation. However, conventional methods of sample preparation do not provide access to the full pattern of membrane proteins and in particular transmembrane proteins due to their low abundance, hydrophobic nature, and limited enzyme accessibility. Based on the use of the iST kit for the preparation of LC-MS samples, the study aims to investigate the potential of the BeatBoxÒ homogenizer to improve the identification of membrane and transmembrane proteins from whole cell lysates or cellular fractions enriched for cell membranes by ultracentrifugation, using brain pericytes as the study cell type. By using BeatBox for protein extraction and solubilization in conjunction with iST sample preparation, we observed an improvement in the identification of membrane and transmembrane proteins compared to the iST kit alone. Better peptide alignment scores further highlight that the BeatBox improves the detection of such proteins in a state-of-the-art manner. Further investigations are focusing on strategies to further increase the extraction and digestion efficiency of transmembrane proteins, for example through detergent-mediated destabilization of lipid bilayers. 319

320

P-I-0173 Depletion-dependent activity-based protein profiling SWATH/DIA mass spectrometry discriminates aggressive lung adenocarcinomas through molecular abundances and catalytic action of serine hydrolases Tatjana Sajic1,2, Matej Vizovišek2, Stephan Arni3, Aurelien Thomas1, Sven Hillinger3, Ruedi Aebersold2 1Lausanne University Hospital, University Center of Legal Medicine, Lausanne, Switzerland 2ETH Zürich, Department of Biology, Institute of Molecular Systems Biology, Zurich, Switzerland 3University Hospital Zürich (UHZ), Division of Thoracic Surgery, Zurich, Switzerland Background: Aberrant molecular mechanisms at the origin of pathologies depend on the activity and regulation of molecules, primarily proteins and their interactions. Abundance measurements of proteins or transcripts, widely used in molecular medicine, are generally blind to functional protein status. Determining the functional state of proteins, in particular the catalytic state of enzymes and their regulation by genetic variation, compartmentalisation, post-translational modifications, ligands or molecular interactions, is a prerequisite for understanding cancer progression and therapy resistance. Activity-based proteome profiling (ABPP) is the tool of functional MS-based proteomics for the quantification of catalytically active enzymes1. Here, we describe a novel variant of ABPP, depletion-dependent ABPP (dd-ABPP) combined with an automated SWATH/DIA-MS workflow, which simultaneously determines three molecular layers of data: i) the fractions of catalytically active enzymes; ii) the relative abundances of enzymes and background proteins; and iii) context-dependent protein- protein interactions involving enzymes. Methods: To demonstrate the capabilities of the platform, 32 lung biopsies from a cohort of patients with long and short survival type IIIA lung adenocarcinoma (LUAD) and their normal tissue counterparts embedded in OCT were retrospectively analysed. We monitored the activities of the serine hydrolase (SH) family of enzymes known to be involved in cancer1 and abundances of approximately 4000 contextual tissue proteins using the dd- ABPP-SWATH/DIA-MS workflow, which integrates bespoke software tools for automation2,3. Commercial assays and subsequent lipidomic, metallomic and genomic analyses were used for data validation. Results: A discriminative signature of the activity profiles of 23 SHs and 59 tissue proteins associated with the active enzymes identified accelerated de-palmitoylation of lipoproteins, which were enriched sevenfold with dysregulated hydrolases within the discriminative signature compared to standard proteomes (Fig 1). In aggressive tumours, subsequent lipidomics confirmed an increase in monounsaturated metabolites of palmitic acid in line with increased expression of stearoyl-CoA desaturase, a rate-limiting enzyme of palmitate desaturation. Further results showed that dysregulated catalytic fractions of palmitoyl-(protein) hydrolases are not compensated by biological levels of enzymes, but are most likely related to the specificity of enzyme-protein sequences or molecular tissue heterogeneity (e.g. somatic mutations of SH genes, tissue content of "tobacco" metals). Conclusions: Taken together, the ensemble of our proteomic data, integrated via computational modelling tools, has the ability to provide new insights into the functional regulation of enzymes in tumours. 1.Jessani, N., et al., Nat Methods, 2005. 2(9): p. 691-7. 2.Rost, H.L., et al., Nat Biotechnol, 2014. 32(3): p. 219-23. 3.Rost, H.L., et al., Nat Methods, 2016. 13(9): p. 777-83. 320

322

P-I-0174 Revolutionizing translational research: large-scale targeted PRM proteomics assays enabled by a novel hybrid mass spectrometer Stephanie Samra1, Kevin Yang1, Amarjeet Flora1, Bhavin Patel1, Cristina Jacob1, Scott Peterman1, Neloni Wijeratne1, Khatereh Motamedchaboki1, Amirmansoor Hakimi1 1Thermo Fisher Scientific, San Jose, CA, United States Targeted mass spectrometry assay development for biomarker validation is no simple task. Although, the alure of finding a conclusive biomarker to indicate the presence of disease seems simple and straightforward in the discovery phase, the verification step is a painstakingly costly process. While traditional methods of analyzing target proteins defer to immunological or affinity-based assays, they suffer from lack of specificity and sensitivity. This is of critical concern for potential low abundant biomarkers in highly complex biomatrices. Here we present a simplified and streamlined approach moving from discovery to target verification by liquid chromatography coupled with a novel hybrid mass spectrometer instrument enabling targeted PRM assay for > 1000s of peptides associated with healthy and disease states and affected pathways. A quantitative targeted mass spectrometry (MS) method was created for > 1000s of peptides associated with proteins in disease related pathways, in less than 3 days by utilizing a gas-phase fractionation, Data Independent Acquisition (DIA) method to quickly map retention times and select charge state and optimal peptide transitions for targeted pathway protein assay. The targeted assays were then built by including optimal transitions in a PRM assay. The new instrument collects PRM data with 5x the throughput of previous generations of MS instruments, which provides the flexibility to create a high throughput (>60 SPD) quantitative targeted assay on a larger set of target compounds. To evaluate the assay"s quantitative performance, we spiked in the synthetically prepared disease biomarker associated SIL peptides by serial dilution into neat plasma digest processed. The depletion resin removed high abundant proteins like albumin and IgGs which constitute roughly 90 % of the plasma proteome, masking the identification of the low abundant proteins in a complex sample such as plasma. This was done in a 96-well plate format to demonstrate high-throughput capability of the workflow. The depleted samples were analyzed in triplicate to assess precision, accuracy and minimum Datapoints Per Peak (DPP) of the targeted assay. Early evaluation of the peptides demonstrated a % CV of < 20 % and a minimum of 10 DPP for roughly 70 % of the peptides in the quantitative PRM method. To further demonstrate the utility of this targeted assay we analyzed a mini cohort study that contained plasma samples of various high-prevalence disease with age- and ethnicity-matched normal plasma samples for protein/peptide biomarker detection and assessment. Novel Aspect: Novel Hybrid Mass Spectrometer Enables Streamlined Verification of Large-Scale Disease Biomarker Panel of more than 1000s targets. 322

323

P-I-0175 Optimization of fractionation methods for cross-linking mass Spectrometry in studying protein-protein interactions and protein structures in complex mixtures Suparat Scheu1, Mandy Rettel1, Tara Bartolec2, Martin Garrido-Rodriguez2, Nassos Typas2, Mikhail Savitski1,2 1European Molecular Biology Laboratory (EMBL), Proteomics Core Facility, Heidelberg, Germany 2European Molecular Biology Laboratory (EMBL), Genome Biology, Heidelberg, Germany Cross-linking mass spectrometry (XL-MS) is capable of identifying protein-protein interactions (PPIs) in complex protein mixtures and providing valuable structural information of proteins in their native states. XL-MS is performed by using specifically designed cross-linkers that connect two amino acids in close proximity within the same or different proteins with a covalent bond. Upon enzymatic digestion, this generates cross-linked peptides, which are identifiable by liquid chromatography tandem mass spectrometry, (LC-MS/MS). Despite many recent outstanding studies employing XL-MS approaches, there are still several limitations that have to be overcome in order to establish XL-MS as a robust workflow that can be used in service infrastructures. Detecting and identifying cross-linked peptides is one of the challenging steps due to the broad dynamic range and high complexity of proteomes, which often interfere with the identification of low-abundance cross-linked peptides in complex mixtures. Therefore, careful consideration is very important when selecting a peptide separation approach in order to enhance sample recovery and cross-link identifications. This work aims to optimize four fractionation methods, Size Exclusion Chromatography (SEC), High-pH Reversed-Phase Chromatography, Porous Graphitic Carbon (PGC)-HPLC, and PGC stage tips, that are available at our facility. The gradient LC-MS/MS profiles and MS acquisition parameters are first optimized and then applied to validate these fractionation methods for separating cross-linked peptides from high-complexity samples (E. coli cell lysates) prior to MS analysis. In combination with the [MR1] enrichable cross-linker PhoX, approximately 2,000-5,800 unique cross- linked peptides and 100-700 protein-protein interactions (PPIs) from E. coli lysates are identified across the methods. From the comparison of four different fractionation approaches, SEC and PGC-HPLC methods are able to identify more crosslinks than other methods. Altogether, the validated fractionation method developed here will be useful as a guideline for the XL-MS field and can be applicable for several types of cross-linkers and other proteome-wide XL-MS studies. This work will lead to a robust and widely applicable XL-MS protocol in the EMBL proteomics core facility and EMBL researchers and researchers across the EMBL member states. 323

324

P-I-0176 Identification of biomarkers for diabetic kidney disease progression by utilizing a nanobinder-enabled deep plasma proteomics platform Ban Zhao1,2, Yonghao Zhang3, Xiehua Ouyang3, Le Shen3,4, Yi Wang5, Hao Wu3, Yonghui Mao1,2 1Beijing Hospital, Department of Nephrology, Beijing, China 2Institute of Geriatric Medicine, Chinese Academy of Medical Sciences, National Center for Genrontology, Beijing, China 3Nanomics Biotechnology Co., Ltd., Hangzhou, China 4The University of Chicago, Department of Surgery, Chicago, IL, United States 5China Resources Research Institute of Science & Technology, Hong Kong, Hong Kong Introduction: Diabetes presents a global health challenge, with diabetic kidney disease (DKD) being a major complication that can progress to end-stage kidney disease. Predicting which patients will develop DKD and experience rapid progression is crucial, but current blood biomarker methods lack sensitivity and specificity. To address this, we developed the Proteonano™ platform for deep, untargeted plasma proteomic profiling that allowed discovery of protein biomarkers that effectively predicts DKD progression. Methods: Proteonano™ technology was developed as an affinity-selective mass spectrometry platform, including the usage of nanoparticle-based affinity protein binders (nanobinders) to enrich low abundance proteins and the employment of an automated pre-treatment workstation for parallel sample preparation. Patients were serially enrolled under approved ethical review. Serum samples were collected at the time of enrollment, kidney functions were assessed for up to five years, and patients were stratified as DKD progressors and non- progressors. Samples were processed through the Proteonano™ Lite pipeline and analyzed with a ThermoFisher Exlporis 480 mass spectrometer at data-independent acquisition mode. Data were processed using DIA-NN and a customized bioinformatics pipeline. Results: Serum samples from 67 DKD patients (DKD progressors=38; non-progressors=29) were analyzed. 1393 protein groups were identified, with an average of 951±11 (AVG±SE) per sample. 1185 proteins were mapped to the human plasma protein project plasma protein catalog, with concentrations spanning eight orders of magnitude, and the lowest at 3.0 pg/mL. Differential expression analysis identified 45 proteins between progressors and non-progressors, with VWF being the most upregulated and PASK the most downregulated protein. VWF was the best single differentiator but not superior to the urine albumin to serum creatinine ratio (UACR). Thus, multivariate analyses were subsequently conducted. Eight feature selection methods were employed, and the top features selected in each method were subjected to Akaiki information criteria (AIC) based model selection. The best model, selected by random forest method, included five proteins (ROC- AUC=0.97, 95 % confidence interval: 0.94-1.00), outperforming UACR (ROC-AUC=0.87). Adding UACR to this improved discrimination (ROC-AUC=0.99). These results demonstrate Proteonano™ model marginally nanobinder assisted sample preprocessing improves untargeted proteomic analysis of blood samples, which allowed identification of a novel protein panel that effectively identifies patients with increased risk for DKD progression. 324

325

P-I-0177 Absolute quantification of fluorescent proteins by mass spectrometry Anna Shevchenko1, Archishan Ghosh1, Andrea Schuhmann1, Henrik Thomas1, Aliona Bogdanova1, T-Y. Dora Tang1, Andrej Shevchenko1 1MPI of Molecular Cell Biology and Genetics, Dresden, Germany Objectives: Fusions with fluorescent proteins (FPs) are widely used for targeted visualization of proteins in a variety of biological contexts. While microscopy monitors the localization, dynamics and interactions of FP- labelled proteins, their molar abundance remains unknown. Proteins fluorescence is affected by bleaching, slow chromophore maturation or autofluorescence and is sensitive to experimental conditions. To overcome these problems we developed fluorescence-independent method for the absolute (molar) quantification of a variety of FPs using LC-MS/MS and a single protein chimera standard. Method: We selected and validated quantitypic peptides from the sequences of six popular FPs (mScarlet-1, mKate2, mCherry, EGFP, mNeonGreen and Dendra2) and from two small self-labelling proteins HaloTag and Snap-Tag. The gene encoding for concatenated peptide sequences together with for five peptides from reference protein (BSA) was designed in-silico. Corresponding chimera protein named "qFP8" was expressed in ΔargA ΔlysA auxotrophic strain of E.coli in the media supplemented with 13C6,15N4-L-arginine and 13C6-L- lysine. An aliquot of E.coli lysate containing isotopically labelled chimera standard was either co-digested with the analysed material, or digested separately with known amount of BSA, and then spiked into the sample lysate and analysed by LC-MS/MS. Absolute quantification was performed based on the extracted abundance (FragPipe, SkyLine software) of peptides derived from chimeric standard protein and corresponding FPs (Fig.1). First, chimeric standard was quantified by BSA and then the amount of FP was calculated using corresponding isotopically labelled peptide proxies. The amount of FP-fused protein is equimolar to the FP. Results: We demonstrated that 30 proxy peptides included into the chimeric protein standard enabled targeted quantification of more than 70 known FPs, their fusions and proteins labelled with organic dyes (e.g. Alexa) via Snap-Tag and HaloTag at the low femtomole to attomole level. Each FP was independently quantified with 3 to 5 peptides with inter-peptide CV < 20%. Several FPs could be quantified simultaneously in a single experiment directly from whole cell or tissue lysate down to single cell level. Since the concentration of chimera is exactly known, it can be used as generic internal standard for absolute quantification of any protein detectable at MS1 spectra. We employed this method to monitor the expression of FPs and FP-labelled proteins in bacterial and human cells, and in cell-free in-vitro translation-transcription systems. Combination of MS and fluorescence readouts were applied to model expression kinetics of FP-labelled human ID-proteins. Fig. 1 (optional) 325

326

P-I-0178 Probe the protein conformation using top-down hydrogen exchange mass spectrometry at higher resolution with electron transfer dissociation Yuqi Shi1, Graeme McAlister1, Rosa Viner1 1Thermo Fisher Scientific, San Jose, CA, United States HDX has emerged as a powerful tool to study protein conformation in native states. When HDX is analyzed at the peptide or protein level, only the overall deuterium uptake of peptides or proteins is obtained. Determination of deuterium incorporation at single amino acid resolution requires fragmentation which may induce “hydrogen scrambling”. ETD, as a nonergodic, fragmentation technique could yield a low level of hydrogen scrambling and therefore allow single residue localization of incorporated deuterium. A model peptide was used to evaluate this technique to obtain single residue resolution. Calcium binding protein, calmodulin, was studied with top-down and bottom-up approaches to compare the protein conformation for its apo- and holo- forms. Apo- and holo- forms of calmodulin were studied by top-down and bottom-up HDX-ETD experiments using a fully automatic HDX workflow station with Vanquish Binary pump N™, HESI source and a modified Thermo Scientific™ Orbitrap Ascend™ mass spectrometer. A full scan (MS) and an ETD MS2 scan were both conducted at the same time for bottom-up experiments. The MS full spectra were used to probe region of significant change in deuterium incorporation, and the ETD MS2 spectra were then used to pinpoint deuterium incorporation at the single amino acid level for the specific peptides that showed significant deuterium change. The synthetic peptide P1, HHHHHHIIKIIK, was prepared in both H2O and D2O at 10 µmol/L and kept at 4 ºC over 24 hours allowing complete deuterium exchange. The effect of source and capillary temperature on hydrogen scrambling and deuterium back exchange was measured, and MS parameters were tuned to limit scrambling. The low deuterium incorporation between c2 and c6 is indication of a low level of deuterium scrambling. The deuterium content for the c fragments of peptide P1 is comparable to what was reported in the literature. Top-down experiment of MS full scan with targeted ETD MS2 was performed for apo-, holo- calmodulin at different labeling time points, using optimized ETD reaction time of 5 msec. The deuterium incorporation of the intact level was measured. High sequence coverage was obtained from identified fragments from N and C terminals. HDExaminer was used to calculate the deuterium incorporation for each fragment and to generate the mirror plots. The plots clearly showed that there was more deuterium incorporation for the C-terminal than N-terminal. Based on the crystal structure, there were more conformation changes on C-terminal side than N-terminal side. Nearly single amino acid resolution for N- and C-terminal was obtained to improve the resolution. Bottom-up hydrogen/deuterium exchange with five labeling time points experiments were also conducted Consistent results were obtained from the two approaches. Overall, the apo- calmodulin had more deuterium incorporation than the holo- calmodulin form. 326

327

P-I-0179 Profiling the proteome of tumor-derived exosomes in microliters of serum by liposome-mediated membrane fusion Qian Shi1, Xiaoni Fang1, Baohong Liu1 1Fudan University, Department of Chemistry, Shanghai, China Exosome protein markers have garnered significant attention as liquid biopsy targets in recent years. They play a crucial role in early cancer diagnosis, treatment monitoring, and prognosis assessment. However, existing studies predominantly focus on all extracellular vesicles or all exosomes, failing to accurately distinguish and analyze tumor-derived exosome subpopulations directly associated with cancer. Consequently, the current accuracy and specificity of exosome proteomics research remain inadequate, severely limiting the efficiency of clinical protein marker screening. The primary challenges in exosome protein marker research include highly specific identification and separation of tumor-derived exosome subpopulations from complex serum samples while excluding interference from protein complexes and free proteins. To address these challenges, we employed the tumor marker epithelial cell adhesion molecule (EpCAM) as the recognition target for tumor-derived exosomes. We designed a corresponding nucleic acid aptamer and modified it onto the surface of liposomes. By fusing these liposomes with exosomes, we achieved precise identification and separation of tumor-derived exosome subpopulations within the serum background. Subsequent downstream proteomic analysis further enhances our understanding. This innovative approach enables relative quantitative analysis of serum proteomes from both prostate cancer patients and healthy individuals. It holds promise for accurate prostate cancer diagnosis and serves as a powerful tool for disease monitoring and prognosis assessment. 327

328

P-I-0180 Chromatin-directed proteomics identifies cell vulnerabilities in DNA repair and elucidates the impact of signal transduction on the DNA-bound proteome Giancluca Sigismondo1, Huiyu Wang2,3, Lavinia Arseni4, Nicolàs Palacio-Escat3, Thomas G Hofmann5, Martina Seiffert4, Jeroen Krijgsveld3 1Heidelberg University Hospital and German Cancer Research Center (DKFZ), Neuropathology, Heidelberg, Germany 2Shanghai Jiao Tong University, State Key Laboratory of Microbial Metabolism, School of Life Sciences and Biotechnology, Shanghai, China 3German Cancer Research Center, Proteomics of Stem Cell and Cancer, Heidelberg, Germany 4German Cancer Research Center, Immune Modulation in Cancer, Heidelberg, Germany 5University Medical Center of the Johannes Gutenberg-University Mainz, Mainz, Germany Fundamental biological processes ranging from DNA replication and repair to cell cycle and gene transcription are tightly controlled in time and space through the coordinate interplay of among different chromatin layers including DNA accessibility, histone post-translational modifications (hPTMs), and chromatin-associated proteins. Yet a comprehensive understanding of chromatin dynamics downstream of signaling cascades is elusive. We here develop chromatin-directed proteomic approaches to provide an unbiased perspective of the chromatin composition dynamics upon signaling pathway perturbation. In particular we adopt our multi-layered chromatin proteomic strategy to quantify and dynamically profile different chromatin layers at high temporal resolution during DNA damage repair (DDR) signaling. In particular, we a) use ChIP to enrich for known repair proteins and characterize their functional chromatin-mediated interactions via the Selective Isolation of Chromatin-Associated Proteins (or ChIP-SICAP), b) dissect the composition of the DNA-bound proteome by the isolation of Proteins on Chromatin (or iPOC), and c) profile signatures of hPTMs at mononucleosome resolution with Native Chromatin Proteomics (N-ChroP). Through our multilayered approach we quantify a high number of both expected and novel determinants deregulated during the DDR, and validated them by imaging- and FACS-based orthogonal approaches. These results allow us to attribute novel candidates to the main DDR repair pathways. In particular, we identified a role for the histone reader ATAD2, the microtubule organizer TPX2 and the histone methyltransferase G9A as novel important regulators of the homologous recombination (HR) repair pathway, and showed that their depletion sensitizes to PARP inhibitors, thus prompting for innovative combinatorial treatments. Furthermore, during the DDR we profile 33 hPTMs at mononucleosomes bearing the DNA damage marker gamma H2AX (or γH2AX). Integration of these data with the complementary results from iPOC shed light on, among the others, G9A- mediated monomethylation of H3K56 in HR repair. Moreover, all multi-layered chromatin proteomic data can be easily queried via a shinyApp at chromatin-proteomics.dkfz.de. In addition, we further optimized and applied iPOC to investigate the DNA-bound proteome dynamics upon perturbation of the cancer-relevant PI3K/AKT/mTOR signaling pathway. Our results show how the selective inhibition of PI3K, AKT, or mTOR has a distinctive effect on the DNA-bound proteome. Moreover, we provide evidence that the PI3K/AKT/mTOR signaling cascade acts on the DNA-bound status of the epigenetic regulator SUZ12, thereby modulating H3K27me3 levels. Collectively, our results provide novel insight on the chromatin dynamics in highly studied signaling pathways, thereby demonstrating the added value of chromatin-directed proteomics in decoding the complexity of biological processes. 328

329

P-I-0181 Advancing single-cell proteomics for profiling tumour heterogeneity Joachim Smollich1, Shiri Karagach1, Ofir Atrakchi1, Tamar Geiger1 1Weizmann Institute of Science, Rehovot, Israel Tumour heterogeneity is the mayor cause of ineffective cancer therapies, contributing to treatment resistance and metastatic progression. The intricate cellular diversity within tumours, metastases, and patient populations underscores the need advanced measurement technologies at single cell resolution. Despite significant progress in genomic and transcriptomic analyses, the comprehensive characterization of tumour proteomic heterogeneity remains elusive. Contemporary methodologies, particularly in single-cell proteomics (SCP), are facing diverse challenges. Technical barries are the ultra-low sample amounts, lack of sample amplification and very limited high throughput sample acquisition options. Biological challenges include sample complexity and heterogeneity. As a prerequisite for biological meaningful and technical reliable (single cell) sample experiments tackling the tumour heterogeneity new methods for the acquisition of thousands of single (cell) samples are needed. To meet this technical demand, we exploited current state-of-the-art single-cell mass spectrometry acquisition methods and ultra-low sample processing workflows to overcome these challenges. By overcoming existing barriers, we established a robust and automated framework for single-cell mass spectrometry-based proteomics allowing mass spectrometry-based proteomics experiments of heterogeneous tumour samples. The established SCP workflow enables identification of more than 3000 protein ID from a single cell. Furthermore, usage of low- binding 1536 well culture microplates and semi-automated sample loading allows processing of hundreds of single cell samples in a robust and time-efficient manner. Finally, exploiting the concept of dimethyl labeling of single cell samples and the miniaturization of the labeling workflow doubles sample throughput without any significant decrease in protein identification. The presented advancements in mass spectrometry-based SCP like automated sample processing, advanced mass spectrometry-based workflows and multiplexed sample acquisition methods hold promise for future translational cancer research. The presented improvements may contribute to a more complete understanding of proteomic tumour heterogeneity at a single cell resolution. In addition, implemented automation and enhanced throughput capabilities will further expedite the analysis of large cohorts of heterogeneous tumour samples. 329

330

P-I-0182 Addressing weather-driven ion mobility drifts in diaPASEF-based proteomics Michael Steidel1, Marcus Bantscheff1, Nico Zinn1 1GSK, Omics Sciences, Target Disovery, Heidelberg, Germany Data-independent acquisition (DIA) is increasingly popular in mass spectrometry-based proteomics providing unbiased results and high data completeness. Adding an ion mobility dimension in diaPASEF on timsTOFpro mass spectrometers has proven effective for achieving deep proteome coverage, even with short liquid chromatography gradients commonly used in high throughput studies. However, for large sample cohorts, ensuring both mass stability (m/z) and ion mobility stability (1/K0) over time is crucial for consistent data quality. We discovered that even minor changes in environmental air pressure during measurements systematically affect TIMS vacuum pressure, leading to shifted ion mobilities. By comparing experimental ion mobilities with historical weather data, we found that drifts observed over a few days can be attributed to typical day-to-day fluctuations in ground air pressure. Even small ion mobility drifts negatively impact the quality of quantitative results. Ions at the mobility boundaries of the diaPASEF precursor isolation windows exhibit drift-dependent abundance changes and increased missing values, which cannot be corrected by post-processing. To address this issue, we applied an in-batch mobility autocalibration feature on a run-wise basis that robustly prevents weather-related ion mobility drifts. Fig. 1 (optional) 330

331

P-I-0183 Rapid and in-depth proteomic analysis of CHO cell bioprocesses Darina Stoyanova1, Karl Burgess1, Mark Rendall2, Jeff Keen2, Luke Johnston1, Lisa Imrie1, Leon Pybus2, David Mentlak3 1University of Edinburgh, School of Biological Sciences, Edinburgh, United Kingdom 2FUJIFILM Diosynth Biotechnologies, Billingham, United Kingdom 3University of York, York, United Kingdom Chinese hamster ovary cells (CHO) are the main platform for monoclonal antibody production in the growing biopharmaceutical market. Compared to any other production platform, CHO cells, being mammalian cells, have the major benefit of providing appropriate protein folding and glycosylation patterns. However, further knowledge on the effect of protein burden and optimisations aimed at improving the viability and culture longevity for higher protein production are required. Programmed cell death – apoptosis, is the predominant form of response occurring from biochemical stresses in bioreactor culturing. Cell longevity, yield and product quality are thus negatively affected. However, apoptosis is operated through multinodal pathways, not yet comprehensively understood. We used proteomics for global investigation of CHO cell bioprocesses and identification of pathways or markers associated with apoptosis that could be targets for cell process improvement. Due to the very large sample sets obtained per experiment, we introduced a short liquid chromatography gradient combined with a data- independent acquisition strategy, allowing for more sensitive and reproducible analysis of industrially relevant CHO cell lines. Additional utilisation of µPAC Neo HPLC columns reduced inter-column variability, increased the number of identified proteins and allowed increased number of samples processed per batch. Investigation of different monoclonal antibody producing cell lines, with different productivities, has allowed the identification of stress-induced pathways connected with increased protein burden. Additional comparison of bioreactor-adapted producer cell line and non-adapted non-producer cell line suggested mechanisms through which cells adapt to the stress occurring when culturing non-adapted lines in bioreactor conditions. Our high throughput methodology could also be used in both industrial scale experiments and in other spheres such as large-sample clinical cohorts. 331

332

P-I-0185 Getting faster, deeper biological insights using ZT Scan DIA Stephen Tate1, Patrick Pribil1, Naomi Diaz1, Claudia Alvarez1 1SCIEX, Research, Vaughan, Canada Techniques to identify proteins in samples are potentially reaching a limit where almost all proteins that are expressed are being measured, with many publications showing the identification of ~10,000 proteins. The ability to identify all peptides/proteins in a sample through stochastic data-dependent acquisition (DDA) approaches may never be achieved due to the extreme complexity of the proteome (proteoforms, modified peptides and abhorrently spliced variants). Latest data-independent acquisition (DIA) methods are significantly improving the ability to assign parent mass with fragments, enabling the identification of a wider range of peptides/proteins in a single analysis, providing better sequence coverage and higher quality quantitation. ZT Scan DIA, a Zeno trap-enabled scanning DIA workflow using better parent ion/fragment ion association, was implemented on a research-modified Zeno trap-enabled QTOF system. Both a mixture of 3 proteomes at variable ratios and mixtures of other species including multiple human cell line digests were used to evaluate this novel scan mode for identification of more peptides/proteins in a single analysis. Data was processed using library-free mode using a combined proteome FASTA. Custom Python scripts were used for analysis of the different outputs for further analysis/visualization of results. Utilizing multiple data aspects to reference a peptide in a complex LC-MS profile provides more confident peptide identification. By investigating the number of confidently identified protein groups within replicate runs and by then determining the sequence coverage for the different proteins, we show that the use of an additional dimensionality in DIA data provides an improvement in both the coverage as well as the number of protein groups identified using microflow gradients. We show a significant improvement in the sequence coverage of identified proteins. The coverage of the proteins is more significantly increased at the lower intensity range. 332

333

P-I-0186 Utilising data-independent acquisition mass spectrometry (DIA-MS) to enable rapid profiling of the human secretome Chloe Tayler1 1GSK, Stevenage, United Kingdom Existing cytokine assay formats are currently the gold standard for analysing secreted proteins from a given system. These assays utilise secondary antibody read-outs and are very sensitive, however they are limited in terms of depth, incredibly costly (thousands of pounds per plate) and are unable to perform an un-targeted global analysis of a systems secretome. Advances in MS-based proteomics sample preparation and technology make it an advantageous platform for studying the secretome at pace. In addition, an MS approach allows both un-targeted global analyses and targeted PRM analyses from a single sample. The aim of this work was to develop and apply a high-throughput compatible workflow for profiling the human secretome by MS-based proteomics. All experimental analysis was performed on an EvosepOne liquid chromatography system (Evosep Biosystems) coupled with the timsTOF Pro2 mass spectrometer (Bruker Daltonics). In the first step of prm-PASEF acquisition method development, a panel of cytokines, chemokines and growth factors were digested using Trypsin and analysed on the 100 SPD (11 minutes) LC gradient using standard DDA-PASEF acquisition. To create an initial prm-PASEF target list, a database search was conducted in FragPipe before filtering to remove non-unique peptides and those with missed cleavages or methionine oxidation. The initial prm-PASEF target list was tested using individual cytokine standards and the corresponding transition list was generated using Skyline software. Here, the top performing peptides for each protein were selected and collated into the final prm-PASEF target list. The linear detection range for each peptide was subsequently determined in prm-PASEF mode on the 100, 200 and 300 SPD (11, 5 and 3 minutes, respectively) LC gradients. Quantitation was performed using peptide standard curves spanning a 1000-fold concentration range over eleven concentration levels in both mobile phase and in a surrogate extracellular matrix. Finally, as a proof-of-concept study, we treated human iPSC derived macrophages with various inflammatory stimuli before harvesting cell pellets and supernatants for both global and targeted proteomic analysis. Here, we were able to demonstrate that this workflow can be used for discriminating between different inflammatory phenotypes. A parallel study was also set up in the same cell type, stimulating samples with various Toll-like receptor agonists before harvesting cell pellets and supernatants over 24 hours for proteomic analyses; enabling the generation of a kinetic profile of cytokine release over this time period. 333

334

P-I-0187 Characterizing the performance of midia-PASEF – maximizing information content in data-independent acquistion Ute Distler1, Mateusz Lacki1, Michał Piotr Startek1, David Teschner2, Jens Decker3, Stephanie Kaspar-Schoenefeld3, Thilo Schild1, Jonathan Krieger4, Florian Krohs3, Oliver Raether3, Andreas Hildebrandt2, Stefan Tenzer1,5 1University Medical Center, Johannes Gutenberg University Mainz, Immunology, Mainz, Germany 2Johannes-Gutenberg University Mainz, Institute for Informatics, Mainz, Germany 3Bruker Daltonics GmbH & Co. KG, Bremen, Germany 4Bruker Ltd, Milton, Canada 5Helmholtz Institute for Translational Oncology (HI-TRON), D191 Immunoproteomics, Mainz, Germany Here, we characterize the performance of midia-PASEF, a novel DIA scan mode using mobility-specific scanning of overlapping quadrupole windows to optimally cover the ion population in the ion mobility-mass to charge plane for various sample types, including phosphopeptidome and mixed proteome samples. Using diagonal scanning, midia-PASEF covers the entire precursor ion mass range without sacrificing cycle time, thereby maximizing information content in DIA acquisitions. Scan-specific encoding of quadrupole positions enables the determination of the precursor m/z of each fragment ion by dedicated precursor predoction algorithms. To process the resulting high-complexity datasets, we developed the Snakemake-based midiaID pipeline. midiaID integrates a set of specifically developed algorithms for multidimensional peak detection and for machine- learning-based classification of precursor-fragment relationships, which are stored in the format of a bipartite graph. Using machine learning approaches to refine the MIDIA graph based on initial database results, we significantly improved the specificity of precursor-fragment relationships, thereby surpassing the spectral purity of DDA in our deconvoluted MIDIA-MSMS spectra as indicated by PEAKS database search score distributions. The midiaID pipeline enables fully automated processing and multidimensional deconvolution of midia-PASEF files and exports highly specific DDA-like MSMS spectra with a fragment ion mass accuracy below 10 ppm. In contrast to DDA, midiaPASEF is non-stochastic and thus generates detailed detection profiles of each fragment ion in retention time, ion mobility and quadrupole isolation m/z dimensions, which facilitates the highly specific deconvolution and scoring of precursor-fragment relationships. Resulting deconvoluted midia-PASEF-MSMS spectra are exported as .mgf files which are suitable for de novo sequencing and can be searched directly with established tools including SAGE, PEAKS, FragPipe and Mascot. In short gradients, midia-PASEF acquisition identifies over 100 unique peptides per second and provides powerful library-free DIA analyses including phosphopeptidome and immunopeptidome samples. midiaPASEF data are also suitable for peptide centric approaches and can be interrogated using spectral libraries. In summary, midia-PASEF provides all benefits of DIA acquisitions, including efficient ion sampling, high duty cycle and excellent reproducibility. 334

335

P-I-0188 Improving protein identification and quantitation using ZT Scan DIA Wen Jin1, Leroi DeSouza1, Alina Dindyal-Popescu1, Doug Simmons1, Katherine Tran1 1SCIEX, Concord, Canada Sensitive, specific, and quantitatively precise detection of peptides and proteins is the cornerstone of proteomics, helping to drive biological insight for a wide range of disease research interests. Data-independent acquisition (DIA) has come to the forefront of MS-based proteomics research. With ZT Scan DIA, a Zeno trap- enabled scanning DIA workflow, the quadrupole moves continuously along the precursor mass range while MS/MS spectra are acquired. Fragment ion signals appear when the leading edge of the quadrupole window passes the precursor m/z and disappear when it falls out of the quadrupole window. A fragment ion signal detected in a TOF pulse is assigned to all the precursors associated with this TOF pulse. Data from each TOF pulse are summed and sectioned into narrow precursor bins (1–2 Da), resulting in a triangular shape of the quadrupole ion trace (QIT) of the fragment. The highest signal of a fragment ion (tip of the triangle) is in the precursor bin that contains the respective precursor ion. Zeno trap pulsing significantly improves fragment sensitivity and has been shown to be beneficial in quantitative proteomics. In this work, with analytical-flow chromatography, ZT Scan DIA with Zeno trap pulsing was used for the analysis of commercial K562 digest using a research modified Zeno trap-enabled QTOF system. The number of proteins and peptides identified and quantified were compared between variable window Zeno SWATH DIA and ZT Scan DIA methods with equivalent accumulation times and cycle times. In addition, the quantitation performance of ZT Scan DIA in terms of lower limits of quantitation (LLOQs), upper limits of quantitation (ULOQs) and linear dynamic ranges (LDRs) were evaluated by spiking various concentrations of isotopically labeled synthetic peptides into the K562 matrix. 335

336

P-I-0189 Novel multi-reflecting TOF instruments: a compact multi-reflecting TOFMS (MRT) shines in both – top-down and DIA MSMS. Another MRT with 1 million resolution is for BUP quantitation at 20 ppb accuracy Anatoly Verenchikov1 1Mass Spectrometry Consulting (MSC), R&D, Bar, Montenegro Unveiling new frontiers in proteomics, this work introduces two novel Multi-Reflecting TOF Mass Spectrometers (MRTs) designed to revolutionize protein exploration. These instruments offer a compelling combination of speed, dynamic range, and resolution, pushing the boundaries of current mass spectrometry capabilities. The first MRT emerges as a compact powerhouse for broad proteomic coverage. This instrument boasts impressive specifications: ultrafast acquisition reaching 500 spectra per second, exceptional dynamic range spanning five orders of magnitude (10^5), and high resolution exceeding 100,000. Notably, this translates to 0.1 ppm mass precision in DIA LC-MS/MS experiments. Furthermore, the cost-effective design, featuring a compact and budget-friendly build, makes it a highly accessible option for widespread adoption in clinical studies. Excelling at identifying a vast number of proteins (currently exceeding 8,000), this instrument maintains excellent isotopic resolution for proteins up to 80 kDa. Additionally, it delivers accurate monoisotopic mass measurements (100 ppb) for mid-sized proteins and achieves remarkable sequence coverage exceeding 80% in MS/MS spectra. The second high-resolution MRT breaks new ground with a resolution of 1 million. This instrument combines exceptional resolution with fast scanning, enabling rapid analysis, and full mass range analysis, capturing the entire proteome in a single run. Furthermore, it boasts high sensitivity, allowing for quantitation at 20 ppb mass precision. This translates to deep proteome coverage, identifying over 3,000 proteins at a false discovery rate (FDR) below 1% from single cell digests in a single LC-MS run. This instrument offers a powerful advantage for quantitative studies, allowing accurate protein quantification directly from MS1 spectra even in DIA MS/MS experiments. Notably, it maintains impressive protein identification capabilities (over 7,000 proteins) during DIA MS/MS analysis. While operator skill influenced proteomic study performance, our advanced instrumentation developed by MSC surpasses the speed and dynamic range of current platforms by significant margins. These groundbreaking instruments hold tremendous potential to expand the toolkit available for proteomic research and applications. A detailed discussion on the fundamental advantages and limitations of these novel instruments compared to existing MS platforms will be presented. 336

337

P-I-0190 Improved proteome coverage combined with reproducible quantitation on the timsTOF platform Stephanie Kaspar-Schoenefeld1, Dijana Vitko2, Andreas Schmidt1, Markus Lubeck1, Pierre-Olivier Schmit3, Torsten Mueller1, Scarlet Koch1 1Bruker Daltonics GmbH & Co. KG, Bremen, Germany 2Bruker Daltonics Inc., Billerica, MA, United States 3Bruker France S.A., Wissembourg, France In the realm of proteomics, accurate and precise relative protein quantitation is the key to unravel the complex secrets of biological processes. In recent years there has been a shift towards analyzing complex proteomics samples in shorter time to keep up with the increasing demand of sample throughput. Dia-PASEF is an advanced variant of Data-Independent Acquisition (DIA), capitalizing on the additional dimension of separation by trapped ion mobility separation (TIMS). We applied more advanced (py_diAID, vistaScan) window schemes to evaluate their performance on complex proteomics samples analyzed from different amounts in short gradient times. Tryptic digests of human cell lysate (in-house digest), Saccharomyces cerevisiae (Promega) and E. coli (Waters) were used to evaluate the performance of sophisticated DIA methods (py_diAID and VistaScan) for short gradients. Digests were loaded on a 15cm C18 column (75μm, 1.6μm, Aurora, IonOpticks) using a nanoElute 2 nano HPLC (Bruker) coupled to a timsTOF instrument (Bruker) via a CaptiveSpray source (Bruker) using a 15-min ACN gradient. Data were processed in Spectronaut (v19, Biognosys) using library-free mode (directDIA+™). For direct database identifications from dia-PASEF runs, we used human, E. coli, and yeast Uniprot fasta files. False discovery rate (FDR) was controlled at 1% for peptide and protein group level. In our study we found that using a variable window-based dia-PASEF method enables measurement of nearly the complete yeast proteome in 15 minutes by identifying more than 4500 protein groups using library-free data processing. 95% of the protein groups were identified and quantified with CV values below 20% from triplicate injections and 83.1% with CV <10%. On average 50,233 peptides were identified with very low variation of ±350 peptides between replicates. From a human protein digest sample, representing a higher complexity proteome, nearly 113,000 stripped peptide sequences from over 8000 protein groups have been identified with comparable reproducibility to the yeast proteome measurements. Samples mixed in defined ratios (HeLa, yeast, E. coli) were used to evaluate the setup for quantitative proteomics. Within a 15 min gradient we could identify and quantify over 13,000 from 400ng sample, with on average 10 peptides being identified per protein group. Investigation of the quantitative performance showed low median coefficient of variation for the replicate runs at around 5% on protein group level and quantitative accuracy with low standard deviation. Lastly, we investigated performance of the timsTOF platform using further optimized schemes based on VistaScan, a method allowing to scan the quadrupole synchronously with the TIMS ramp during the mobility scan to optimally cover the ion population in the ion-mobility-mass to charge plane. 337

338

P-I-0191 Nanobinder based plasma protein capture enabled deep proteomic profiling identifies novel biomarkers for early cognitive decline in an aging population Dan Liu1,2,3, Yonghao Zhang4, Xiehua Ouyang4, Shanshan Lv4, Yanting Meng4, Libing Wang4, Yi Wang5, Hao Wu4, Yan Zeng1,2,3 1Wuhan University of Science and Technology, Brain Science and Advanced Technology Institute, Wuhan, China 2Geriatric Hospital Affiliated to Wuhan University of Science and Technology, Wuhan, China 3Wuhan University of Science and Technology, School of Public Health, Wuhan, China 4Nanomics Biotechnology Co., Ltd., Hangzhou, China 5China Resources Research Institute of Science & Technology, Hong Kong, Hong Kong Introduction: Alzheimer's disease (AD) progresses over more than a decade, with brain damage occurring before noticeable behavioral symptoms. Early detection is crucial for slowing disease progression and enhancing patient quality of life. Existing biomarkers like amyloid β (Aβ40, Aβ42) are inadequate for early-stage detection, necessitating novel biomarkers. We developed the Proteonano™ platform for deep, untargeted plasma proteome profiling to detect AD early. Methods: Proteonano™ technology was developed as an affinity-selective mass spectrometry platform, including usage of nanoparticle-based affinity protein binders (nanobinders) to enrich low abundance proteins and employment of an automated workstation for parallel sample preparation. Patients were serially enrolled under approved ethical review. Plasma samples were collected at time of enrollment, and participants were stratified as normal (N), mild cognitive impairment (MCI), and dementia (D). Plasma samples were processed through the ProteonanoTM pipeline and analyzed by a ThermoFisher Orbitrap Astral mass spectrometer at data independent acquisition mode. Raw data were analyzed by using DIA-NN, normalized, further processed by using a customized biostastic and bioinformatic pipeline. Results: Plasma samples from 206 serially enrolled participants (N=142, MCI=35, D=29) were analyzed. 4347 protein groups were identified, with 2344 ± 37 (AVG±SE) proteins per sample. 2704 proteins matched the human plasma protein project published plasma protein database. Protein concentrations ranged over nine orders of magnitude, with the lowest at 1.6 pg/mL. Differential expression analysis showed 64, 91, and 159 proteins with different abundances between MCI vs. N, D vs. MCI, and D vs. N groups, respectively. MGP was the most upregulated and SPTBN1 was the most downregulated in MCI group relative to normal group. Eight feature selection methods were employed, and top features selected in each method were subjected to Akaike information criteria (AIC) based model selection. Identified features were then combined for another round of selection. Differentiating powers of these models were assessed by receiver operating curve (ROC) and precision and recall (PR) methods. The best model for differentiating MCI from N used nine proteins, achieving a ROC-AUC of 0.92 (95 % confidence interval: 0.89-0.98)and PR-AUC of 0.84, outperforming Aβ 40 (ROC-AUC=0.80). Models for differentiating D vs. N (eight proteins, ROC-AUC=0.99, PR-AUC=0.96) and D vs. MCI (nine proteins, ROC- AUC=0.98, PR-AUC=0.97) also performed well. This indicates that nanobinder-assisted untargeted proteomics can effectively identify protein features distinguishing different cognitive states, enhancing early AD detection. 338

339

P-I-0192 Proteonano™: a nanobinder-based proteomics platform for robust and standardized deep plasma proteomics study Xiehua Ouyang1, Libing Wang1, Shanshan Lv1, Xin Du1, Yanting Meng1, Yonghao Zhang1, Yi Wang2, Jie Jin3,4, Hao Wu1 1Nanomics Biotechnology Co., Ltd., Hangzhou, China 2China Resources Research Institute of Science & Technology, Hong Kong, Hong Kong 3Wuhan Biorun BioSciences Co.,Ltd., Wuhan, China 4Wuhan Institute of Biotechnology, Wuhan, China Introduction: Profiling the human plasma proteome is challenging due to the immense dynamic range of plasma protein concentration spanning ten orders of magnitudes between the most abundant protein and the least abundant proteins detected so far. Most of currently used workflows are labor-intensive and rely on abundant protein depletion and fractionation methods to improve detection depth. In addition, most of the pipelines involve manual sample processing, limiting throughput and assay consistency. We developed unique nanobinders that enriches low abundant proteins and semi-automatic workflow to increase plasma protein detection limit with increased throughput and decreased intersample variation. Methods: Superparamagnetic nanoparticles were synthesized at scale and conjugated with small molecules and peptides by using a proprietary deep learning pipeline. Abilities of these nanobinders to capture plasma proteins were determined by LC-MS/MS. Sample processing was performed by using a in-house designed liquid handling Nanomation™ G1 system optimized for parallel (up to 96 samples) nanobinder-based proteomic sample processing. Using Nanomation™ G1, nanobinders were incubated with plasma samples, washed, followed by denaturation, reduction, alkylation, digestion, and desalting. Digested samples were analyzed using high- resolution mass spectrometers, including ThermoFisher Exploris 480 and Orbitrap Astral, and Bruker timsTOF Pro2. Results: More than 500 nanobinders were synthesized and characterized. The ability of these nanobinders were determined by LC-MS/MS. The best nanobinder detected 4129 protein groups in a single run, compared to 810 without enrichment. Concentrations of detected proteins spanned nine orders of magnitude, with 70.2% of detected proteins had reported concentrations below 10 ng/mL. When human and porcine plasma were mixed at different proportions and subjected to LC-MS/MS, concentrations of detected human proteins at different mixing ratios displayed similar linearity between samples with or without Proteonano™ based pre-processing. For assay sample were parallelly processed by Nanomation™ instrument using two batches of the same nanobinder. 4057±38, 4159±30 protein groups were identified for each batch of nanobinders, respectively, with protein concentration coefficient of variation (CV) at 9.87% and 9.62%, without data normalization. In a real-world test, 12 replicates of the same quality control (QC) tests were interspersed with 206 experimental plasma samples, which detected 4348 protein groups in QC samples with a protein number CV of 1.47%. For commonly detected protein groups, a median CV for protein centration was 16.88 % without data normalization demonstrating high reproducibility. stability measurements, the same plasma Conclusions: Proteonano™ platform allows deep, rapid, reproducible proteomic analysis of plasma samples. 339

340

P-I-0193 TSAP: a method for targeted signal amplification of peptides in phosphoproteome profiling Fujia Wang1, Maarten Altelaar1, Kelly Stecker1 1Utrecht University, Biomolecular mass spectrometry and proteomics, Utrecht, Netherlands Introduction: Mass spectrometry-based phosphoproteomics is a powerful tool that aims to explore signaling network regulation and dysregulation in normal and pathological conditions. Due to the low stoichiometry of phosphorylated peptides and the high diversity of phosphorylated peptide species, the acquisition of sufficient coverage of regulatory phosphorylation sites is challenging, especially when analyzing samples with tiny amounts. In a common pipeline, a discovery method enables global profiling of proteome at first, due to many signals being missed in the profiling, a targeted method for the interested peptides has to be performed afterward. This pipeline suffers from the high cost of heavy-labeled standards, time-consuming to set up an experiment, and more sample consumption. Methods: Here, we developed a multiplexing method using tandem mass tags with a carrier channel to amplify the signals of interested phosphorylated peptides. Sixty phosphorylated peptides were spiked in as the carrier and an SPS-MS3 with real-time searching method was applied. The proof of concept was tested using breast cancer cell lysates analyzed on the Thermo Fisher Orbitrap Eclipse. Three approaches, MS2, SPS-MS3, and SPS- MS3 with real-time search, were compared in conditions with this strategy and DDA (Data-dependent acquisition) approach. After adjustment of the parameters, the performance of identification and quantification were tested by a two-proteome interference assay. Then, the TSAP with optimized parameters was compared to a targeted method. In addition, TSAP was used to detect the phosphorylation status of breast cancer cell models treated with insulin. Conclusions: In this project, we demonstrated a phosphoproteomics method for limited amounts of samples, which combined a multiplexing strategy and the SPS-MS3 with the real-time searching method. This approach is beneficial for applications involving limited quantities of samples. It enables targeted observation of the status of phosphoprotein peptides of interest in parallel with comprehensive global phosphoproteomics analysis in a single run. Moreover, the simple setup and high throughput of this method allow it to be used in a variety of situations. Fig. 1 (optional) 340

341

P-I-0194 Combining data independent acquisition with spike-in SILAC (DIA-SiS) improves proteome coverage and quantification Anna Sophie Welter1, Maximilian Gerwien1, Robert Kerridge1, Keziban Merve Alp1, Philipp Mertins1,2, Matthias Selbach1,3 1Max Delbrück Center, Berlin, Germany 2Berlin Institute of Health at Charité – Universitätsmedizin Berlin, Berlin, Germany 3Charité – Universitätsmedizin Berlin, Berlin, Germany Data Independent Acquisition (DIA) is increasingly preferred over Data Dependent Acquisition (DDA) due to its higher throughput and fewer missing values. Whereas DDA often utilizes stable isotope labeling to improve quantification, DIA mostly relies on label-free approaches. Efforts to integrate DIA with isotope labeling include chemical methods like mTRAQ and dimethyl labeling, which, while effective, complicate sample preparation. Stable isotope labeling by amino acids in cell culture (SILAC) achieves high labeling efficiency through the metabolic incorporation of heavy labels into proteins in vivo. However, the need for metabolic incorporation limits the direct use in clinical scenarios. Spike-in SILAC methods utilize an externally generated heavy sample as an internal reference, enabling SILAC-based quantification even for samples that cannot be directly labeled. Here, we combine DIA with spike-in SILAC (DIA-SiS), leveraging the robust quantification of SILAC without the complexities associated with chemical labeling. We developed and rigorously validated DIA-SiS through a mixed- species benchmark to assess its performance in proteome coverage and quantification. We demonstrate that DIA-SiS significantly improves proteome coverage and quantification compared to label-free approaches. Additionally, DIA-SiS proves effective in analyzing proteins in low-input formalin-fixed paraffin-embedded (FFPE) tissue sections. DIA-SiS combines the precision of stable isotope-based quantification with the simplicity of label- free sample preparation, facilitating simple, accurate and comprehensive proteome profiling. 341

342

P-I-0195 ESPRESSO: a universal proteomic workflow based on a novel non-toxic dewaxing agent for FFPE samples Rong Xu1,2,3, Yuanxuan Huang1, Chuanxi Huang2, Pingping Hu4, Aihua Sun1, Tiannan Guo4, Fuchu He1,2,3, Dongxue Wang1,2 1Beijing Proteome Research Center, State Key Laboratory of Proteomics, Beijing, China 2International Academy of Phronesis Medicine, The π-HuB Project Infrastructure, Guangzhou, China 3Guangzhou Laboratory, Guangzhou, China 4Westlake University, Key Laboratory of Structural Biology of Zhejiang Province, Hangzhou, China Introduction: FFPE tissue samples are invaluable resources for proteomics-based clinical and biomedical research. Formalin fixation and paraffin-embedding allow proteins to be stable for decades but make protein extraction and analysis more challenging than fresh frozen tissues. Traditionally, paraffin is removed by stepwise incubating samples with xylene, ethanol, several aqueous solutions with decreasing concentrations, and finally with water. Besides being time-consuming, researchers are unavoidably exposed to the hazards of xylene. Here, we presented a non-toxic, efficient, and cheap substitution of xylene which passed the careful evaluation of proteomic analysis. The streamlined proteomic method based on this substitution proves the feasibility of analyzing large tissue cohorts in a robust, timely, and unbiased manner. Methods: Based on this, a non-toxic green dewaxing reagent that can completely replace xylene was screened, and a complete method system (ESPRTSSO) in DIA mode was established and evaluated. Subsequently, completely symmetrical FFPE and OCT samples were prepared based on mouse kidney tissues, and the biological differences between FFPE samples and OCT samples were systematically evaluated using different DIA library search modes of different protein search software (MaxDIA, DIA-NN, and Spectronaut), as well as comparing the effects of different libraries on protein quantification. Results: The entire method reduces the processing time by approximately 4/5 compared to the traditional method, with the preparation process completed within 4 hours. The sensitivity of the method enables protein- level identification with a starting amount of 1µg protein. Furthermore, we tested the stability in human tumor FFPE samples, and the novel dewaxing agent can be adapted to other method systems, such as PCT. A systematic comparison of perfectly symmetrical FFPE and OCT samples from mice revealed that the difference between FFPE and OCT samples is much smaller than that of the biological samples. Furthermore, the construction of a hybrid library based on peptides from FFPE samples and peptides from OCT samples demonstrated that proteins lost during the preparation of FFPE samples can be retrieved. This improvement in peptide coverage per protein group enhances the overall reliability of the method. Conclusion: In conclusion, we've developed a mass spectrometry pre-processing procedure for FFPE samples using a new non-toxic green dewaxing agent and conducted a thorough evaluation of the method in DIA mode. 342

343

P-I-0196 Optimization of data-independent acquisition mass spectrometry coupled with FAIMS interface for highly quantitative proteomic analysis Yeji Yang1, Hea Ji Lee1, Ji Hyun Kang1, Jung Hoon Choi1, Ju Yeon Lee1,2, Heeyoun Hwang1, Jin Young Kim1,2 1Korea Basic Science Institute, Digital Omics Research Center, Cheongju, South Korea 2Research Institute of Bioscience and Biotechnology, Daejeon, South Korea DIA (Data Independent Acquisition) analysis is utilized to accurately identify and quantitatively measure proteins and other biological molecules in large-scale biological samples. It provides a more comprehensive and in-depth analysis, offering rapid and precise information in complex biological systems. Moreover, the FAIMS (high-field asymmetric waveform ion mobility spectrometer) interface reduces chemical noise and improves signal-to-noise ratios, particularly benefiting proteomics. To enhance the utility of DIA in the diagnostic field, we aimed to couple it with a FAIMS interface to maximize protein identification in human serum and optimize quantitative comparative analysis. Initially, we validated the effectiveness of DIA analysis before applying it to human serum and conducted CV (compensation voltage) optimization. Subsequently, we confirmed whether these results were consistent in human serum and compared the quantitative results of DIA-FAIMS with or without depletion of high abundant proteins. By integrating multiple results, we established an optimized method for human serum analysis that maximizes effectiveness. 343

344

P-I-0197 π-HelixNovoV2: Narrowing the gap between amino acid recall and peptide recall for de novo peptide sequencing Tingpeng Yang1,2, Tianze Ling3,4, Boyan Sun3, Zhendong Liang1,2, Fan Xu1, Xiansong Huang1, Linhai Xie3, Yonghong He1,2, Leyuan Li3, Fuchu He3,5, Yu Wang1, Cheng Chang3,5 1Peng Cheng Laboratory, Shenzhen, China 2Tsinghua Shenzhen International Graduate School, Shenzhen, China 3National Center for Protein Sciences (Beijing), State Key Laboratory of Proteomics, Beijing, China 4Tsinghua University, School of Life Sciences, Beijing, China 5Chinese Academy of Medical Sciences, Research Unit of Proteomics Driven Cancer Precision Medicine, Beijing, China in de novo peptide sequencing. Therefore, De novo peptide sequencing is the mainstream technique to identify novel peptides, highlighting the performance improvements for the state-of-the-art models. The gap between amino acid recall and peptide recall is usually huge because any amino acid parsing error in a peptide will result in the parsing error of that peptide, presenting a significant challenge improving the performance of de novo peptide sequencing models not only requires a higher amino acid recall but also a smaller gap between the amino acid recall and the peptide recall. Here, inspired by the idiom "Many hands make light work", we introduce the bidirectional parsing system (generating peptides in two directions: in two directions: from C-terminal to N-terminal and from N-terminal to C-terminal) to narrow the gap and enhance the peptide recall for de novo peptide sequencing models. We first developed two candidate bidirectional parsing systems, namely encoder-independent (EI) and encoder-shared (ES) systems, and demonstrated their effectiveness in enhancing the performance of de novo peptide sequencing models (i.e., Casanovo, π- HelixNovoV2, etc.) utilizing a series of ablation experiments. Meanwhile, we compared the performances of the EI and ES systems and confirmed the details of the bidirectional parsing system. Then, we applied the system to our previous study π-HelixNovoV1 and upgraded it to π-HelixNovoV2. Afterward, we demonstrated that π- HelixNovoV2 outperforms other state-of-the-art models (i.e., PEAKS, pNovo, DeepNovo, PointNovo, Casanovo, π-HelixNovoV1, NovoB, PepNet and GraphNovo) using a series of comparative experiments. Moreover, we trained a powerful π-HelixNovoV2 utilizing a larger training dataset, and as expected, π-HelixNovoV2 achieves unprecedented performance, even for peptide-spectrum matches with never-before-seen peptide sequences and the mass spectra with the missing fragmentation problem. We also used the powerful π-HelixNovoV2 to identify antibody peptides, multi-enzyme cleavage peptides, and non-enzymatic peptides, and π-HelixNovoV2 demonstrates robust and superior performance in these applications, with notable improvements achieved through fine-tuning. Finally, we proposed a quality control strategy for de novo peptide sequencing, and π- HelixNovoV2 significantly improves the sensitivity of peptide identification under the quality control strategy. We also utilized π-HelixNovoV2 with the quality control strategy to de novo gut metaproteome peptides and the results show π-HelixNovoV2 increases the identification coverage and accuracy of gut metaproteome and enhances the taxonomic resolution of gut metaproteome. Our results demonstrate the effectivity of the π- HelixNovoV2 and take a significant step forward in de novo peptide sequencing. Fig. 1 (optional) 344

345

P-I-0198 High-throughput discovery of differential subcellular protein localization during chemotherapy resistance in triple-negative breast cancer Shij Yang1, Liang Qiao1, Ling Lin1 1Fudan University, Shanghai, China Intracellular proteins localize and translocate to perform specific functions essential for maintaining physiological activities. Dysfunctional subcellular protein localization is implicated in various diseases, including heart failure, neurodegenerative diseases, and cancer. Traditionally, the study of differential protein translocation has relied on low-throughput methods such as immunofluorescence for screening and verification. Here, we established and optimized a high-throughput approach based on mass spectrometry to investigate the differential localization of subcellular proteins under varying conditions. Specifically, we utilized differential centrifugation to separate organelles, followed by label-free quantification through mass spectrometry and data- independent acquisition (DIA) to obtain systematic subcellular protein profiles. To identify differentially localized proteins, we employed the Bayesian ANalysis of Differential Localization Experiments (BANDLE), a Bayesian algorithm model designed to calculate the probability of protein localization changes under various experimental conditions. This model analyzes protein abundance across different cellular components, providing a robust method for identifying proteins with significant localization changes. Our experiments revealed that treatment with paclitaxel resulted in the subcellular differential localization of 198 proteins in triple-negative breast cancer (TNBC) cell line MDA-MB-231. Notably, proteins in the Golgi apparatus, mitochondria, and endoplasmic reticulum exhibited significant translocation. In addition,we found that the position of JAK-STAT signaling pathway related proteins in the cell was significantly changed under PTX, with O60674 protein transferred from an unknown location to the cell membrane and P40763 protein transferred to the cytoplasm. This result is consistent with previously reported results of changes in the signaling pathway under PTX drug treatment. This high-throughput method provides a novel perspective on the mechanisms underlying chemotherapy resistance in cancer and offers potential therapeutic strategies for targeting TNBC based on organelle protein translocation. Future applications of this method will explore the differential localization of proteins in cardiomyocytes under hypoxic conditions. Additionally, validating the function of key translocated proteins through in vivo and in vitro models using cellular and molecular biology techniques are also part of our next steps. Fig. 1 (optional) 345

346

P-I-0199 Optimization of DIA for Accurate and Precise Label-Free Quantification in Proteomics Pegah Zahedimaram1, Hannah Heinrich2, Natalie Widmann1, Elke Schaeffeler2, Ute Hofmann1, Thomas Mürdter1, Matthias Schwab3 1IKP, Chemical Analytics & Synthesis, Stuttgart, Germany 2IKP, Molecular Analytics, Stuttgart, Germany 3IKP, Stuttgart, Germany Recent advancements in Orbitrap mass spectrometry and AI-based data processing have significantly enhanced the depth and accuracy of proteomic analyses, enabling the identification of more proteins and improved quantification precision. In label-free quantification (LFQ), improving protein coverage, reproducibility of peptide abundances, and minimizing the coefficient of variation (CV) are crucial. This study aimed to evaluate different LFQ methods in proteomics for robust and reproducible results when comparing biological samples. To identify the optimal LFQ method, using HEK293T total cell lysate, we compared data-dependent acquisition (DDA) and data-independent acquisition (DIA) techniques using various sample preparation protocols, LC gradients, and data processing workflows. The study utilized the Vanquish Neo Nano-LC system coupled with the Exploris 480 Orbitrap mass spectrometer, with data processing performed using PD 3.1. We assessed data quality based on the number of proteins and peptides identified, CV% of measured abundances, and accuracy of abundance measurements using spiked-in peptide standards. We tested IW ranging from 35 to 200, using normal, variable, and staggered patterns. Variable window sizes were designed based on peptide counts in each m/z range from a DDA acquisition of the same sample, either manually or using the Encyclopedia wizard. Our results indicated that all tested acquisition methods demonstrated similar performance in terms of peptide and protein identifications and reproducibility. Interestingly, 200 IW resulted in the highest number of identified proteins (~ 7000) at a false discovery rate (FDR) of 1%, with an abundance CV of less than 20% for over 80% of quantified proteins, though, the CV% distribution tended toward higher values, indicating the importance of the number of data points for LFQ. To determine the most accurate method, a standard digest was spiked into the sample in varying amounts, allowing for the assessment of the method reporting the closest ratio. The DIA method, designed based on peptide counts in each m/z range from a DDA acquisition of the same sample, exhibited the highest accuracy in the full range of 400-1,000 m/z. This method utilized a variable isolation windows pattern with isolation widths of 10, 5.5, 10, and 15 Da, identifying 6,000 total proteins and quantifying 5,861 proteins, achieving a CV of less than 20% for 85% of the quantified proteins. Finally, it was tested to identify differentially expressed proteins in two biologically distinct samples to validate the method. This study underscores the critical role of optimizing DIA parameters, particularly isolation windows, in enhancing the accuracy and reproducibility of LFQ in proteomics. The findings suggest that careful method design, tailored to specific sample characteristics, can significantly improve proteomic analyses, paving the way for more precise and reliable biological insights. 346

348

P-I-0200 Assessing protein and payload stability of antibody drug conjugate brentuximab vedotin in monkey plasma using agilent assayMAP bravo and agilent 6495 triple quadrupole LC/MS systems Xi Qiu1, Andrea Ngai2, Kenda L.J. Evans3, Anthony Zerlin3, Muhammed Naqvi2 1Agilent Technologies, Wilmington, NC, United States 2Amador Bioscience, Pleasanton, CA, United States 3Agilent Technologies, Santa Clara, CA, United States By the end of 2022 there have been 14 antibody drug conjugates (ADC) approved by US Food and Drug Administration to treat a variety of cancers and more than 100 ADC candidates have been investigated in clinical stages at present. Unlike traditional drugs, ADC is typically composed of three parts, a monoclonal antibody (mAb) a cytotoxic drug (payload) and a chemical linker which links the mAb and toxic drug. The traditional analytical approach of these large molecules is via the ligand binding assay (LBA), but LBA cannot determine free payload concentration and drug to antibody ratio (DAR). Here we demonstrate a hybrid LBA/LC/MS workflow, which combines LBA and LC/MS technologies, to quantify the mAb and payload in monkey plasma using an Agilent AssayMAP Bravo, Agilent 1290 Infinity II Bio LC, and an Agilent 6495 triple quadrupole system. Our results show that this hybrid LBA/LC/MS workflow can be used for quantitative analysis of ADC without the need for a specific antibody, while providing excellent sensitivity, high specificity, and fast method development. These factors will play an important role in drug discovery and development. 348

349

P-I-0201 NovoCert: statistical validation of de novo peptide sequencing Shanji Zhang1, Seunghyuk Choi2, Seungjin Na3, Eunok Paek1,4 1Hanyang University, Computer Science, Seoul, South Korea 2Baylor College of Medicine, Houston, TX, United States 3Korea Basic Science Institute, Cheongju, South Korea 4Hanyang University, Institute for Artificial Intelligence Research, Seoul, South Korea De novo peptide sequencing via tandem mass spectrometry can be useful to discover novel, unknown peptides. Yet, its practical application is hampered by a lack of statistical validation. We introduce NovoCert, a novel method leveraging semi-supervised learning and statistical techniques to validate peptide spectrum matches (PSMs) inferred from de novo sequencing. NovoCert can independently validate high-confidence peptides, reducing reliance on scores from de novo peptide sequencing tools. Initially, PSMs were acquired by de novo peptide sequencing tandem mass spectra. The 'exact' group was determined by aligning de novo results with the reference protein sequences. For each PSM in the exact group, we generated a corresponding decoy PSM by de novo sequencing a 'reverse-shifted (RS)' spectrum: (1) a decoy peptide is produced by reversing each target peptide; (2) the target spectrum is converted into a decoy spectrum by adjusting the positions of its fragment ions to align with the computed m/z values expected for the decoy peptide. A 1% false discovery rate (FDR) for the exact group was estimated using Percolator. We used ProtomeTools synthetic peptide dataset (PXD004732) to perform decoy justification. We used Comet for database search and used PEAKS for de novo peptide sequencing. We compared distributions of XCorr, precursor mass error, IonFrac, deltaCn between results of searching target spectra using DecoyDB (reversed human protein sequences) and results from searching decoy spectra generated by RS and PS(Precursor Swap) methods against TargetDB (reference sequences of homo sapiens). We adopted the RS method because the RS method has distributions most similar to the target spectra searched against DecoyDB than PS method. Utilizing PEAKS, NovoCert identified 70% of PSMs (92.2% at the peptide level) within the exact group (2,438,270 total PSMs) at 1% FDR. We compared the results with the existing FDR estimation method for de novo sequencing, which we called ReverseDB method. The ReverseDB method generates a decoy PSM by matching de novo results with the reverse protein sequence DB to estimate FDR. We can see that NovoCert is a more conservative method and more reliable than the ReverseDB method. NovoCert demonstrated its effectiveness by improved identifications in terms of spectral angle, delta retention time, number of annotation peaks and the precursor mass error. 349

350

P-I-0202 Trace sample proteome analysis by data-dependent acquisition without dynamic exclusion Shen Zhang1, Jiao Lei1, Fei Meng1 1Reproductive and Genetic Hospital of CITIC-XIANGYA, Changsha, China Despite continuous technological improvements in sample preparation, mass-spectrometry-based proteomics for trace samples faces the challenges of sensitivity, quantification accuracy and reproducibility. Herein, we explored the applicability of turboDDA (a method by using data-dependent acquisition without dynamic exclusion) for quantitative proteomics of trace samples. After systematic optimization of acquisition parameters, we compared the performance of turboDDA with data-dependent acquisition with dynamic exclusion (DEDDA). By benchmarking the analysis of trace unlabeled human cell digests, turboDDA showed substantially better sensitivity in comparison with DEDDA, no matter for unfractionated or high pH fractionated samples. Furthermore, through designing an iTRAQ labeled three-proteome model (i.e., tryptic digest of protein lysates from yeast, human and E. coli) to document the interference effect, we evaluated the quantification interference, accuracy, reproducibility of iTRAQ labeled trace samples, and the impact of PIF (precursor intensity fraction) cutoff for different approaches (turboDDA and DEDDA). The results showed that improved quantification accuracy and reproducibility could be achieved by turboDDA while more stringent PIF cutoff resulted in more accurate quantification but less peptide identification for both approaches. Finally, the turboDDA strategy was applied to the differential analysis of limited amounts of human lung cancer cell samples, showing great promise in the trace proteomics sample analysis. 350

351

P-I-0203 Degradation of different ubiquitinated proteoforms of tau proteins by the 20s proteasome Gabriele Antonio Zingale1, Irene Pandino1, Mariapina D'Onofrio2, Michael Assfalg2, Giuseppe Grasso3, Grazia Raffaella Tundo4, Diego Sbardella1 1IRCCS Fondazione Bietti, Rome, Italy 2University of Verona, Department of Biotechnology, Verona, Italy 3University of Catania, Department of Chemical Sciences, Catania, Italy 4University of Rome “Tor Vergata”, Rome, Italy The structural and functional role of site-specific ubiquitylation of proteins is an emerging research topic in cell biology and molecular medicine. Therefore, this research was applied to figure out the susceptibility to proteasome degradation of: 1) an enzymatic multi-mono-Ub tau proteoform (Ub-K18) and the non-Ub variant (K18), covering the 4RD-MBD domain of tau (TauK18), which is mainly involved in the aggregating properties of the protein and contains all the acceptor lysine residues found ubiquitinated in brain biopsies of subjects deceased for tauopathies; 2) site-specific Ub-S-S-Tau proteoforms (Ub-S-S-Tau254, Ub-S-S-Tau311 and Ub-S-S- Tau353) generated by disulfide coupling chemistry and their non-Ub control, C291A/C322ATau4RD, also called TauΔC (a mutant devoid of Cys residues). In the view of the growing interest on free 20S mediated proteolysis, SDS-PAGE electrophoresis and Mass Spectrometry (MS) experiments were set-up to investigate the pattern of degradation of the Tau proteoforms (5 μM) by an affinity-purified human 20S proteasome (h20S, 10 nM) over 3h in 25 mM Tris-HCl, pH 7.5, 37°C. Proteins were then resolved by SDS-PAGE. For MS experiments, reaction mixtures harvested at each time-point were stopped by adding TFA and analysed with a UHPLC-MS ThermoFisher Ultimate3000 – Orbitrap 240 Exploris. Introduction of Ub by enzymatic reaction (Ub-Tau-K18) renders the protein susceptible to h20S mediated digestion, as non-Ub K18 was resistant. In the case of the digestion assays of Ub-S-S-proteoforms preparation, the position of the Ub moiety along the TauK18 sequence significantly affected the dynamics of digestion of the individual protein by the h20S. Indeed, the h20S showed preferences for Tau species with Ub installed at one out of the two termini, -254 and -353 whereas -311 was a poor substrate. Remarkably, Ub was cleaved by the h20S along with the substrate: Ub and Tau-Ub branched fragments were detected in the reaction mixture of -254, -353, but also of -311 (though at low rate) in the presence of the h20S. These data highlight that the h20S degrades mono-Ub tau species of pathological relevance in vitro. Furthermore, the position of Ub appears to be critical for catalytic preferences and, notably, Ub is degraded along with the substrate it is conjugated with, to generating a repertoire of peptides more diversified than expected based on actual knowledge of the 20S properties. 351

352

P-I-0204 Cell-based reactive cysteine profiling of an electrophile fragment library to explore novel target opportunities Melanie Holzner1, Sandra Poser1, Michael Zollo1, Maria Inês Faria1, Philipp Weber1, Christin Zasada1, Katelyn Cassidy2, Roman Agafonov2, Brent Appleton2, Johannes Krumm1, Jan Schülke1, Jan Muntel1, Hannes Hahne1, Götz Hagemann1 1OmicScouts GmbH, Freising, Germany 2C4 Therapeutics, Inc., Watertown, MA, United States Covalent drugs may facilitate expansion of the druggable proteome through their potential to target proteins at sites that lack well-defined binding pockets generally targeted by non-covalent small molecules. The discovery of novel covalent compounds is aided by technologies such as chemoproteomic reactive cysteine profiling that can assess functional cysteine sites across the proteome to identify potential novel chemical matter targeting the cysteinome. Here, we present a strategy that enables cell-based compound screening to provide a comprehensive overview of putatively ligandable cysteine residues in the proteome. Making use of a 96-well- format and fragment pools for treatment, we applied this assay to profile ~250 fragments in live cells. We quantified almost 30,000 reactive cysteine sites originating from more than 9,200 proteins and identified ~750 strongly liganded cysteine sites (competition ratio (CR) >4; corrected p-value <0.05) in 640 proteins of which ~80% are considered undruggable so far. To validate the fragment pooling strategy, we compared the covalent CDK7 inhibitor THZ1 as single compound or as part of a fragment pool and observed congruent THZ1 cysteine site labeling on its primary target CDK7 and on potential off-targets. This pilot data indicates that we have developed a scalable live cell chemoproteomic platform to screen large compound libraries to expand the portrait of the potentially ligandable cysteinome. 352

353

Population proteomics in health and disease (P-I-0205 to P-I-0232) P-I-0205 Plasma cytokine profiles in the SHIP-TREND-0 cohort – associations with age, sex and BMI Sabine Ameling1, Sandra Van der Auwera2, Stephan Michalik1, Silva Holtfreter3, Elke Hammer1, Barbara Bröker3, Matthias Nauck4, Uwe Völker1 1University Medicine Greifswald, Interfaculty Institute for Genetics and Functional Genomics, Greifswald, Germany 2University Medicine Greifswald, Department of Psychiatry and Psychotherapy, Greifswald, Germany 3University Medicine Greifswald, Institute of Immunology, Greifswald, Germany 4University Medicine Greifswald, Institute of Clinical Chemistry and Laboratory Medicine, Greifswald, Germany Background: The characterization of physiological immune signatures in a population-based cohort enables the identification of pathological immune signatures associated with inflammatory or autoimmune diseases. Methods: A total of 47 plasma cytokines, chemokines, and growth factors were quantified in 1175 individuals of the SHIP-TREND-0 cohort (532 men and 643 women, aged 20 to 81 years, with a BMI of 17.7 to 53.6) using a bead-based multiplex assay (Merck Milliplex HCYTA-60K) on the FlexMap3D instrument. The relationships between cytokine concentrations and age, sex, body mass index (BMI), season, and blood cell parameters (BCP) were investigated using multivariate regression models. Results: Higher levels of pro-inflammatory cytokines and chemokines, including IL-6, IL-8, CXCL9, CXCL10, IL- 12p40, CCL2, CCL4, CCL11, IL-27, and FLT3LG, were significantly associated with increasing age. Significant sex differences were observed for IL-18, CCL2, CCL4, CCL11, CCL22, CCL3, IL-27, IL-12p40, IL1RA, and TNFα levels. IL- 6, IL12p40, and IL-27 levels were affected by BCP. Furthermore, seven cytokines and chemokines, namely CCL4, CCL22, CXCL10, IL-6, IL1RA, IL-18, and TNFα, exhibited elevated levels with increasing BMI. Conversely, only CCL11 demonstrated lower levels with increasing BMI. Additionally, the levels of CCL4, CCL22, CXCL10, IL-6, IL1RA, and CCL11 were found to be significantly distinct between subjects categorised as normal and obese. No significant non-linear seasonal effects were observed for the cytokines under investigation. Conclusion: This study provides an overview of the circulating cytokines and chemokines and their associations with common phenotypes in the population-based cohort SHIP-TREND-0. This could help to identify the onset of pathological changes as a basis for subsequent studies on common diseases. Furthermore, blood cell parameters should be considered as potential confounders in association studies based on plasma cytokine levels. 353

354

P-I-0206 Pre-existing immunity effects on viral infections and vaccinations Carlota Arias-Hidalgo1, Pablo Juanes-Velasco1, Ana Nuno-Soriano1, Angela-Patricia Hernandez1, Javier de las Rivas1 1IBSAL-CIC-IBMCC-CSIC-USAL, Medicine and Proteomics Unit, Salamanca, Spain Infectious diseases are one of the major health issues worldwide, causing high morbidity, and mortality each year. Virus exposure, through infection or vaccination, can elicit cellular and humoral responses. For many viruses, these immune response are long-lasting; among cross-reactive pre-existing antibodies can exacerbate viral pathogenesis in some cases. The implication of pre-existing immunity to many pathogens ( viruses, bacteria, parasites, fungi, …) are not fully understood. Recently, several efforts have been focused on the potential of novel "omics" techniques to obtain information in a simple, faster, and more efficient way. Protein microarrays are one of these techniques useful for characterization of humoral response by multiplexed immunoassays with higy robustness and reproducibility. Methods: a new multiplex platform has been designed and developed for the simultaneous immune-detection and s of IgM and IgG against 37 microbial antigens (Salmonella, SARS-CoV-2, flu, adenovirus, HPV, CMV,...) prevalent in the population that reflect each donor's infection history and previous immunity. Results: In the case of the demographic variables such as sex, age range and ethnicity of the studied population, no distinctive profile was observed when comparing these characteristics. However, differences were observed in the distribution of antigens between the two immunoglobulin isotypes. The detection of IgM has been found to be predominant in the RBD protein and the nucleocapsid protein of SARS-CoV-2 and the outer membrane porin of Salmonella typhi[m1] . Humoral responses have been found to be highly heterogeneous between individuals to the same microbial antigens. Conclusions: Knowing in depth the natural variations of these proteins, depending on the different clinical- biological characteristics, could enable us to correlate these patterns with patients presenting different pathologies, the role of pre-existing immunity, cross reactive immune responses, and vaccine approaches that circumvent pre-existing immunity. In addition, it could reduce mortality or improve diagnosis and treatment progression in an individualized and personalized manner. 354

355

P-I-0207 Trans-pandemic population profiling in young adults Annika Bendes1, Sophia Björkander2, Maura Kere2, Simon Kebede Merid2, Alexandra Lövquist3, Amelie Vogt1, Anna Gardell1, Leo Dahl1, Olga Mukalova3, Olof Beck4, Benjamin Murrell5, Niclas Roxhed6, Erik Melén2, Jochen Schwenk1 1KTH Royal Institute of Technology, Protein Science, Solna, Sweden 2Södersjukhuset, Karolinska Institutet, Clinical Science and Education, Stockholm, Sweden 3Karolinska Institute, Institute of Environmental Medicine, Stockholm, Sweden 4Karolinska Institute, Clinical Neuroscience, Stockholm, Sweden 5Karolinska Institute, Microbiology, Tumor and Cell Biology, Stockholm, Sweden 6KTH Royal Institute of Technology, Intelligent Systems, Stockholm, Sweden Background: The COVID-19 pandemic posed significant challenges to the global community, and particularly to the healthcare systems. In response the virus, new vaccines and technologies were developed and applied to induce, manage, and understand the human immune response to the infection. Despite these efforts, little is known about how pre-pandemic conditions contributed to the immune response to the virus or vaccination. Materials and Methods: During 2020-2022, we collected three home-sampled dried blood spots (DBS) from ~800 participants from the Swedish BAMSE (Children, Allergy, Milieu, Stockholm, Epidemiology) cohort1. We then employed multi-analyte serology2 and affinity proteomics3 to measure immune response against 38 SARS- CoV-2 antigens and 365 inflammation-related proteins. We used various statistical models with data from pre- and peri-pandemic examinations and laboratory tests to elucidate the the effect of the pandemic on younger adults. Results: This study aimed to determine the effect of the pandemic on young adults from the general population by using longitudinal and self-collected DBS samples. Univariate and cluster-based analysis of the multi-analyte serology revealed the expected immune response to the virus infection, vaccination and their combination. It also highlighted that time between sampling and vaccination was indicative for the amplitude in immune response, and that a subset of donors remained uninfected and vaccinated. Our proteome analysis of self- sampled DBS revealed protein signatures that were stable over time, variable, or changed in response to vaccination and/or infection. Conclusion: Our findings demonstrate the technical feasibility of using home-sampled DBS for obtaining insights into immunologic and proteomics response to vaccines and infections. This allowed us to monitor molecular changes and understand how predispositions and acquired conditions contribute to phenotypes of health and disease. References: 1. Wang G et al. Plasticity of Individual Lung Function States from Childhood to Adulthood. Am J Respir Crit Care Med. 2023. doi: 10.1164/rccm.202203-0444OC 2. Roxhed N et al. Multianalyte serology in home-sampled blood enables an unbiased assessment of the immune response against SARS-CoV-2. Nat Commun. 2021. doi: 10.1038/s41467-021-23893-4 3. Fredolini C et al. Proteome profiling of home-sampled dried blood spots reveals proteins of SARS-CoV-2 infections. Commun Med (Lond). 2024 doi: 10.1038/s43856-024-00480-4 355

356

P-I-0208 In-depth plasma proteome from patients with post-viral neurocognitive disorders: a pilot study Jonas Bergquist1, Anna Widgren1, Roxana Martinez-Pinna2, Florian Marty3, Tabiwang Arrey4, Nicolaie Eugen Damoc4, Joanna Kirkpatrick5 1Uppsala University, Uppsala, Sweden 2Thermo Fisher Scientific, Dreieich, Germany 3Thermo Fisher Scientific, Reinach, Switzerland 4Thermo Fisher Scientific, Bremen, Germany 5Thermo Fisher Scientific, Hemel Hempstead, United Kingdom Background lasting sequels. The There are ample examples of outbreaks of post-infectious disorders with pathophysiology of those disorders is very much under-investigated and there is a lack of consensus diagnostic, prognostic and therapeutic guidelines. One of the more common, but still less understood disorders are myalgic encephalomyelitis / chronic fatigue syndrome (ME/CFS). Today, we think that at least 60-80% of the patients have been initially affected and triggered by an Ebstein Barr virus infection. It has also been estimated that more than 65 million individuals worldwide are suspected to be suffering from Long COVID, a complex multisystemic condition, wherein patients of all ages report fatigue, post-exertional malaise, and other symptoms resembling ME/CFS. With no current treatments or reliable diagnostic markers, there is an urgent need to define the molecular underpinnings of these conditions. long Methodology Using a small cohort of patients (8 individuals) and matched healthy controls (8 individuals), we investigated three common sample preparation procedures for plasma proteomes, with the aim to choose the best method for a larger-scale analysis (>10000 samples). These were: standard digestion with trypsin (neat), an antibody- based depletion method known to deplete the 14 most abundant plasma proteins (Top14 depletion), and perchloric acid depletion (per-CA). Reported protein biomarkers for long covid were used to verify the applied depletion methods. We employed the Thermo Scientific™ Orbitrap™ Astral™ mass spectrometer, using DIA to evaluate depth of coverage, quantitative performance, at different throughputs, to determine both the best sample preparation protocol and address the question of whether the larger cohort measurements could be achieved in ~1 yr or 2 yrs. All files were processed with Spectronaut 18. Principal component analysis (PCA) was used for downstream analysis, and significant protein biomarkers were manually verified. Results Analysing three replicates of the 16 samples for each depletion method we identified 2532 protein groups (PGs) for Top14, 1842 PGs for per-CA, and 1031 PGs for neat plasma. Protein group CVs followed the same trend as number of protein groups identified (best 7.5%, worst 12.6%). PCA was able to separate patients versus control for both neat and per-CA treated samples. Strikingly, for per- CA, PC1 contained 63% of the explained variance clearly separating the two groups. Further, we could observe a general trend of higher variability between disease samples compared to healthy controls. In contrast, top14 depletion resulted in no clear separation of the two groups as many of the assumed ME/CFS protein markers (e.g., immunoglobulins) are depleted from the sample. Overall, we conclude that per-CA depletion followed by high-throughput DIA analysis on Orbitrap Astral enables the investigation of large sample cohorts to study post-infectious disorders. 356

357

P-I-0209 Protein signatures across multiple neurodegenerative diseases Sofia Bergström1, Fredrik Edfors1, María Bueno Álvez1, Anna Månberg1, Mathias Uhlen1, Peter Nilsson1 1KTH Royal Institute of Technology, Protein Science, Stockholm, Sweden The identification of disease-associated protein signatures could contribute to an increased understanding of neurodegenerative disorders. Despite some key discoveries, much about disease pathogenesis remains unknown. We hypothesize that there are complex patterns and associations to be discovered throughout the various degree of heterogeneity within and similarities across different diseases. To uncover these patterns, it is essential to investigate many proteins across several independent cohorts. The Human Protein Atlas, one of the most visited biological databases in the world, is expanding to include protein signatures in plasma across many major diseases. This expansion includes 10,000 samples analyzed with Olink Explore 1,500 and another 10,000 samples analyzed with Olink Explore HT, covering 5,400 proteins. The Human Blood Disease Atlas will include protein profiles in plasma from patients with cancer, autoimmune diseases, infectious diseases, and neurological diseases. Specifically, we are analyzing over 2,200 samples from patients with neurological diseases, including Alzheimer’s disease, Parkinson"s disease, frontotemporal dementia, and amyotrophic lateral sclerosis, but also schizophrenia, bipolar disorder, and epilepsy. We are including well-characterized cohorts in collaboration with distinguished clinicians both in Sweden and internationally. This unique effort aims to identify disease-specific signatures as well as patterns common across several diseases. The protein levels across various neurological diseases will be made available as an open-access resource within the Human Protein Atlas (www.proteinatlas.org). In conclusion, analyzing large and independent cohorts across different neurodegenerative diseases is crucial for identifying disease-relevant protein signatures. These complex signatures could enhance our understanding of these diseases and aid in identifying relevant patient subgroups. 357

358

P-I-0210 Persistent antibody clones dominate the plasma antibody repertoire Albert Bondt1, Danique van Rijswijck1, Dina Raafat2, Silva Holtfreter2, Uwe Völker3, Barbara Bröker2, Albert J. R. Heck1 1Utrecht University, Biomolecular Mass Spectrometry and Proteomics, Utrecht, Netherlands 2Universitätsmedizin Greifswald, Institute of Immunology, Greifswald, Germany 3Universität Greifswald, Interfaculty Institute for Genetics and Functional Genomics, Greifswald, Germany The half-life of therapeutic monoclonal antibodies has been extensively studied, and has been found to range roughly between one and four weeks. In contrast, it has been reported that, for instance, viral antigen recognition in blood is extremely durable. Until recently, it was not possible to determine whether the sustained immune response resulted from ongoing production of identical clones, or if repeated activation of B cells led to the continuous emergence of new clones over time. We have developed mass spectrometry-based methods to analyze the antibody repertoire with clonal resolution. Initial analyses conducted on samples from severely ill patients over a period of up to 60 days demonstrated substantial changes in the antibody repertoire over time for some clones, while other clones exhibited remarkable stability. In this study, we examined the clonal repertoire stability in healthy individuals over a longer time period. We followed eleven healthcare workers for almost one year during which they did not experience any significant illness. However, they did receive three doses of a COVID-19 vaccine. Blood samples were collected at nine timepoints, and the plasma IgG1 antibody repertoire was recorded using Fab clonal profiling by mass spectrometry. We show that the plasma IgG1 repertoires of healthy donors are dominated by a few clones, and that these clones remain remarkably stable over at least one year. Even after administering three doses of the COVID-19 vaccine the detected repertoire remained mostly unchanged. This suggests that this large portion of the total repertoire associates with long term immunity, rather than acute infections. Moreover, given that these dominant clones do not appear to undergo affinity maturation, it is most likely that they are produced by long- lived rather than short-lived plasma cells. Fig. 1 (optional) 358

359

P-I-0211 In-depth crude plasma micro-flow LC-MS/MS profiling for disease understanding and biomarkers discovery using novel tandem UHPLC workflow and on-line coupled columns Alicia Grunow1, Klaus Zieroth1, Frank Sacher2, Martina Schmidt1, Zhipeng Wang1, Christian Schultewolter1, Oleksandr Boychenko1 1Bayer AG, Wuppertal, Germany 2Bayer AG, Berlin, Germany The recent emphasis on comprehensive profiling of biofluids, particularly easily accessible plasma samples, is driven by promises of understanding diseases, identifying biomarkers and drug targets, facilitating companion diagnostics development, predicting toxicology effects, and enabling patient stratification for precision medicine. Despite the remarkable progress in technologies such as ligand-binding proximity extension, aptamer-based panels, and ultra-fast and sensitive LC-MS instruments the translation of large cohort data to clinical applications remains challenging. This limitation can be attributed to the complexity of multiplexed data, which reveals population-level trends but lacks applicability to individual patients. Addressing this challenge is crucial for harnessing the full potential of clinical proteomics. Here we focused on conducting in-depth plasma LC-MS profiling proteome profiling. Our aim is to increase the number of identifications by exploring non-specific cleavage sites and fragments, as well as post-translational modifications of proteins (e.g. glycation, phosphorylation, citrullination, acetylation, glycosylation, etc.), and incorporate variety of protein isoforms. To enhance the dynamic range and mitigate biases in enrichment steps, we have improved the separation of peptides by coupling columns with different diameters: a high-capacity 1 mm ID column and a high-resolution 300 µm ID column. Additionally, we utilized a UHPLC system with a pressure during the run above 1000 bar to maintain micro-flow rates, enabling the use of a robust H-ESI source. The coupling of columns has led to an increased system void volume and reduced MS utilization time with common trap and elute or direct injection setups. Consequently, we have implemented a novel tandem UHPLC workflow that has increased MS utilization to above 95%, allowed for deep column wash in parallel to peptide separation, and eliminated delays in peptide elution. In combination with multistep and multi-search engine data processing, we have been able to drastically increase the identification rate of peptides in plasma, which is commonly observed in biofluids analysis. The optimized proteomics workflow was employed to enhance our understanding of cardiovascular, kidney, and skeletal muscle diseases by comparing plasma proteome profiles, thereby unveiling pathways enriched in patient samples compared to controls. We also conducted comprehensive phenotyping of multiple animal species to facilitate the translation of biomarkers during the drug development process. Additionally, we completed a comprehensive study of sample preparation conditions, particularly centrifugation speed on the abundance of molecular markers in plasma and linked proteomics data with metabolomics, mRNA and miRNA profiling. Overall plasma proteomics has achieved a level of robustness that enables the generation of essential data on a day-to- day basis to support the drug development process. 359

360

P-I-0212 Empowering robust population-scale MS plasma proteomics studies with nanoparticle enrichment Lee Cantrell1, Jian Wang1, Ryan W. Benz1, Harendra Guturu1, Eltaher Elgierari1, Evangelina Bahu1, Iman Motashemi1, Andrew Nichols1, Seth Just1, Janos Kis1, Theodore Platt1, Serafim Batzoglou1 1Seer, Inc., Redwood City, CA, United States Introduction: Measurement of plasma in large cohorts enables a variety of studies including biomarker discovery, disease progression tracking and genomic variant analysis. Mass spectrometry (MS) is a powerful technology for unbiased and species-agnostic protein identification. However, challenges for population scale MS measurement have persisted due to dynamic range of plasma protein abundance, multiple sources of analytical variability, and limitations in conventional computational frameworks. Here we demonstrate a sensitive, robust, and efficient integrated workflow for MS proteomics measurement at population scale. Methods: In a blinded study of 1,600 samples, we leveraged the Proteograph™ XT product suite to combine nanoparticle enrichment with plate-based automated sample preparation, MS measurement and cloud-scalable search and processing resources. MS measurements were made using the Orbitrap™ Astral™ MS in data- independent acquisition mode at a throughput of 40 samples per day with robust microflow chromatographic separation. Plate design included digestion, enrichment, cleanup, and MS controls to monitor independent components of platform performance. An additional pooled sample was deposited on every plate to characterize technical reproducibility of the workflow. Search of raw data files was enabled by a novel cloud- based implementation of DIA-NN version 1.8.1 match-between runs (MBR), titled scalable MBR, available in the Proteograph Analysis Suite (PAS) software. Results: In total, approximately 8,100 protein groups were represented in this cohort by measurement of more than 104,000 peptide sequences. Plate design enabled continuous quality control monitoring to support the importance of automated sample preparations. For each control, monitoring metrics included depth, quantitative robustness, outlier analysis, correlation analysis, and dimensional reduction sample representation. Across the study, technical reproducibility of preparation and MS was observed at approximately 30% precursor- level coefficient of variation, significantly below non-technical replicate variance. Using a novel metric, we were able to infer interpretable biological signal without prior knowledge of study design. Additionally, the implementation of scalable MBR in PAS enabled completion of search in under one day, with projected completion more than twenty-times faster than equivalent search on a high-performance workstation. Conclusions: The integration of automated enrichment and sample preparation, robust MS configuration and cloud informatics demonstrates the capacity of Proteograph XT to effectively scale to population-level studies. Furthermore, our on-plate controls validate the ability of this platform to generate reliable and reproducible measurements without laborious protocols or difficult to implement algorithmic pipelines. 360

361

P-I-0213 Large-scale protein-disease risk association analysis in the UK Biobank: Introducing an extensive and freely available research resource in Olink® Insight Ola Caster1, Linn Fagerberg1, Hilda Andersson1, Markus Sällman Almén1, Ida Grundberg1 1Olink Proteomics, Uppsala, Sweden Thanks to increased sample throughput and highly parallelized assaying capabilities, proteomics has emerged as an indispensable tool in biomarker discovery for the detection, prognosis, and treatment of disease. Its impact on healthcare in general and precision medicine specifically is expected to continue to grow. The UK Biobank (UKB) Pharma Proteomics Project (PPP) represents the prime example of large-scale proteomics, with protein level quantification of plasma samples from more than 50,000 individuals. Based on Olink® Explore technology, UKB-PPP has generated data on nearly 3,000 proteins for this vast cohort. When combined with e.g., genomic or healthcare data in UKB, the opportunities for biomarker research in biology and medicine are tremendous. This study aimed to estimate the future risk of a large and diverse set of diseases for all protein biomarkers available in UKB, thus generating a library of protein-disease risk associations freely available to researchers worldwide. In total, 107 diseases, including e.g., cancers, neurological, and cardiovascular diseases, were selected from the PheWAS ontology and mapped to diagnosis codes from longitudinal hospital records, cancer registries, and death registries in UKB. A cohort was created for each disease, consisting of all incident cases within 10 years from the time of blood sampling as well as a set of matched, randomly selected controls with no occurrence of the disease during follow-up. All individuals with a first occurrence of the disease prior to their blood sample were excluded. For each protein, the association between measured plasma levels and the time to first occurrence of the disease was estimated using Cox regression, adjusting for sex, age, body mass index, and smoking status. In total, over 300,000 protein-disease risk associations were quantified in terms of hazard ratios with associated p-values. Our results reveal a large heterogeneity in strength and number of associations both across diseases and proteins. Some proteins, for example growth differentiation factor 15 (GDF15) and WAP four-disulfide core domain 2 (WFDC2), have statistically significant associations to a high proportion of all included diseases, suggesting they might be less suitable as targeted biomarkers. Several other strong associations identified here, including TNF receptor superfamily member 13B (TNFRSF13B) with leukemia and neurofascin (NFASC) with type II diabetes, have been previously reported in independent research, which provides a basic level of validation. The complete set of results has been made freely available via Olink Insight, an online portal to support proteomic research. With its broad coverage of diseases and its large pool of UKB data as foundation, this new resource can enhance future studies. For example, our results can help guide biomarker selection for a given disease of interest or be used to cross-reference findings in the post-study analysis phase. 361

362

P-I-0214 Proteomics profiling of normal and tumoral human colon stem and epithelial cell populations Anniek L de Jager1, Sara Kassem1, Inge F de Laat1, Alesha Louis1, Kyra van der Pan1, Enrique De La Rosa2,3, A Stijn LP Crobach4, Koen CMJ Peeters5, Fabian A Holman5, Alex L Vahrmeijer5, Javier de las Rivas2,3, Jacques JM van Dongen1,6,7,2, Cristina Teodosio1,6,7,2, Paula Díez8,1,9 1Leiden University Medical Center, Department of Immunology, Leiden, Netherlands 2Institute of Biomedical Research of Salamanca (IBSAL), Salamanca, Spain 3Cancer Research Center, Bioinformatics Group, Salamanca, Spain 4Leiden University Medical Center, Department of Pathology, Leiden, Netherlands 5Leiden University Medical Center, Department of Surgery, Leiden, Netherlands 6Cancer Research Center, Translational and Clinical Research Program, Salamanca, Spain 7University of Salamanca, Department of Medicine, Salamanca, Spain 8University of Oviedo, Department of Functional Biology, Oviedo, Spain 9University of Oviedo, Department of Physical and Analytical Chemistry, Oviedo, Spain The colon epithelium is composed of diverse cell types with specific functions, originating from intestinal stem cells (ISCs) which are crucial in various intestinal disorders, including colorectal cancer (CRC). Previous studies on colon epithelial cells (EC), especially ISCs, have shown high expression of proteins like EpCAM (CD326), PGP-1 (CD44), and ALCAM (CD166). However, these studies have limitations in depth and analyte detection. Comprehensive proteomic analysis can offer a deeper insight into the colon epithelium and identify new CRC protein targets. Our goal was to analyze freshly purified ISCs and EC populations from paired normal and CRC samples to depict their proteomes. Paired healthy and tumoral tissue biopsies from seven CRC patients were dissociated. Single-cell suspensions were sorted into four subsets by flow cytometry: colon ISCs (CD326+/CD166+/CD44+), and three additional EC populations [CD166bright ECs (CD326+/CD166bright/CD44-), CD44+ ECs (CD326+/CD166-/CD44+) and other ECs (CD326+/CD166-/het/CD44-)]. Purified cell populations (>93% purity) were processed for mass spectrometry (MS) analysis using a 9 M urea-based lysis buffer, proteins were enzymatically digested and purified using a bead- based method. Then, 3 μg of each sample were labeled with tandem mass tags (TMT10plex+TMT131C) and measured on an Orbitrap Fusion™Lumos™Tribrid™ mass spectrometer. Functional enrichment analysis (FEA) was performed using DAVID v2021. Flow cytometry analyses indicated varying population distributions related to the anatomical origin of colon samples. Most cells belonged to the other EC subgroup across all samples, with right and left colon samples containing the CD166bright EC population (<0.1% frequency) absent in sigmoid and rectum samples. MS identified 3,991 distinct proteins in all colon samples, with 3,435 common proteins, including EC-specific sets of 4 to 5 proteins. While no unique proteins were found for any subset, different expression levels were observed between populations, reflecting different functional profiles. Other ECs exhibited typical colon functions such as absorption, microorganism interaction, and villus formation, while the CD166-/CD44+ population showed additional metabolism-related functions. ISCs displayed enhanced cell cycle and development functions, akin to the more active CD166bright population. Comparing normal and tumoral populations, the latter closely resembled the general EC profile, sharing 3,101 proteins without unique identifications. FEA showed functions highlighting metabolic, RNA, mutational processes, and protein transport functions in tumoral cells, indicative of increased activity and tumoral development. This study presents the proteomic landscapes of ISCs and EC populations from normal and tumoral colon samples, revealing distinct functional patterns that could aid in identifying new tumor biomarkers for CRC diagnosis, treatment, and monitoring when validated in additional cohorts. 362

363

P-I-0215 Mitochondrial dynamics in gastric cancer: contrasting diffuse and intestinal types Diana Lashidua Fernandez Coto1, Jeovanis Gil Valdés2, Sergio Manuel Encarnación-Guevara1, Guadalupe Ayala Aguilar3, György Marko-Varga2, Roberto Herrera Goepfert4 1Universidad Nacional Autónoma de México, Cuernavaca, Mexico 2Lund University, Lund, Sweden 3Instituto Nacional de Salud Pública, Morelos, Mexico 4Instituto Nacional de Cancerología, Mexico City, Mexico Introduction: Gastric cancer remains a significant public health challenge, ranking as the fourth leading cause of cancer-related mortality and the fifth leading cause of cancer incidence worldwide. Patients with diffuse-type gastric cancer (DGC) face a particularly poor prognosis compared to those with intestinal-type gastric cancer (IGC), which benefits from well-defined preneoplastic lesions. The lack of identifiable preneoplastic markers for DGC hinders effective surveillance and prevention. Thus, identifying molecular biomarkers that contribute to the risk of developing DGC is crucial for informing novel preventive strategies. Methods: Fresh frozen biopsies of lesions and adjacent tissues were collected from 42 patients diagnosed with chronic gastritis (CG), diffuse-type gastric cancer (DGC), or intestinal-type gastric cancer (IGC). These 84 samples, consisting of pairs from 15 DGC, 9 IGC, and 18 CG patients, were analyzed using quantitative proteomics. High- resolution variable window Data Independent Acquisition Mass Spectrometry (DIA-MS) was performed via LC- MS/MS. Peptide and protein identification was conducted using DIA-NN. Data processing and statistical analyses were carried out with Perseus v2.0.10.0, R, and GraphPad Prism. Additionally, pathway and biological processes enrichment analyses were performed in R and the Gene Set Enrichment Analysis program (GSEA). Results: Proteomic analysis identified and quantified 9,040 proteins. IGC exhibited relatively higher expression of immune response proteins, including those involved in the interferon response, as well as spliceosome proteins, compared to DGC. Conversely, DGC showed elevated expression of mitochondrial process proteins, particularly subunits of complex I of the electron transport chain, and proteins involved in the Krebs cycle and oxidative phosphorylation. Notably, there was an inverse correlation between the expression of mitochondrial pathways and immune response proteins. Differential analysis between cancerous and adjacent tissues revealed a decrease in mitochondrial processes in both DGC and IGC, with the loss of mitochondrial function being more pronounced in IGC. Conclusions: Our comprehensive proteomic profiling of DGC and IGC, compared with chronic gastritis, provides detailed molecular insights into the distinct and shared pathways underlying these gastric cancer types. The identification of an inverse correlation between mitochondrial pathway activity and immune response proteins highlights key differences in the tumor microenvironment of DGC and IGC. These findings significantly enhance our understanding of gastric cancer etiology and may inform the development of more effective, targeted therapeutic strategies. Furthermore, the distinct molecular signatures identified in this study could serve as potential biomarkers for early detection and personalized treatment of gastric cancer. 363

364

P-I-0216 Analyzing the proteome, N-glycoproteome and phosphoproteome of dried blood: method optimization and application to phenotyping of human sepsis Matt Foster1, Youwei Chen1, Deng Madut1, Robert Plumb2, Marlene Violette1, A. Ian Wong1, Lingye Chen1, Asia Hemed Kipengele3, James Ngocho3, Venance Maro3, Erik Soderblom1, Matthew Rubach1, Timothy McMahon1 1Duke University, Durham, NC, United States 2Waters Corporation, Milford, CT, United States 3Kilimanjaro Christian Medical University College (KCMUCo), Moshi, Tanzania, United Republic Of Introduction: Blood is a common matrix for large-scale phenotyping of disease. However, most proteomic studies are restricted to blood plasma, which can suffer from pre-analytical variability and may not capture relevant cellular pathobiology. We are exploring the utility of dried whole blood, collected by volumetric microsampling, for characterization of sepsis. Method development has focused on facile sample collection and storage, plate-based sample prep, and medium- to high-throughput LC-MS/MS analyses of proteins and post- translational modifications. Methods: For prospective sample collection, twenty microliters of venous whole blood was captured on Neoteryx Mitra devices, dried and transferred to Matrix tubes for -80 °C storage and subsequent processing. Mitra tips were subjected to deoxycholate-assisted trypsin digestion, and ~1 mg of digests were sequentially enriched for N-glycopeptides and phosphopeptides using tip-based hydrophilic interaction chromatography (HILIC) and magnetic nanoparticle-based immobilized metal affinity chromatography (IMAC), respectively. Analysis of unenriched digests used microflow-LC (Waters ACQUITY Premier 1 mm CSH C18) and DIA-MS/MS (Exploris 480), and N-glyco- and phospho-peptide enriched fractions utilized an Evosep One LC and 60 sample- per-day method, and stepped-collision energy DDA-MS/MS (Exploris 480) and DIA-MS/MS (Orbitral Astral), respectively. Spectronaut, Glyco-Decipher and Byonic/Proteome Discoverer were utilized to identify and quantify proteins and PTMs. Preliminary Results: We developed a streamlined workflow for plate-based processing of dried whole blood for bottom-up proteomics, N-glycoproteomics and phosphoproteomics that avoids sample cleanup or drying steps and can be completed in under 2 days. An optimized phosphoproteomic workflow utilized depletion of glycopeptides, magnetic nanoparticle-based enrichment and DIA using an Orbitrap Astral MS. In less than 2h of total analysis time per sample, approximately 1,100 proteins, 750 glycopeptides, and 6,000 site-localized phosphopeptide precursors were quantified in blood from healthy controls after filtering on a coefficient of variation of <20% in replicates of a study-specific pool. As proof of concept, samples were collected in Tanzania as part of an ongoing NIH-funded sepsis subtyping study. The analyses of these samples and integration with clinical and "omic data is in progress. Conclusions: A workflow has been developed for comprehensive and precise proteomic profiling of dried whole blood which utilizes minimal sample prep steps and relatively high-throughput LC-MS/MS analysis. Sample collection is easily integrating into typical bio-banking protocols. This overall strategy should enable the discovery of new insights into human sepsis. 364

365

P-I-0218 Exploring the spectrum of human encounters with bacterial pathogens by profiling pathogen-specific antibody repertoires in a population-based setting using the SHIP-TREND cohort Marie-Sofie Illenseher1, Dina Raafat2,3, Denise Dittmar1, Tanja C. Meyer1, Lisann Scheunemann1, Stefan Weiss1, Nele Friedrich4, Henry Völzke5, Thomas Kocher6, Christian Kohler7, Kristin Surmann1, Christian Hentschker1, Silva Holtfreter3, Lars Kaderali8, Stefan Simm8, Barbara Bröker3, Stephan Michalik1, Uwe Völker1 1Interfaculty Institute for Genetics and Functional Genomics, University Medicine Greifswald, Department of Functional Genomics, Greifswald, Germany 2Faculty of Pharmacy, Department of Microbiology and Immunology, Alexandria, Egypt 3Institute of Immunology, University Medicine Greifswald, Greifswald, Germany 4Institute of Clinical Chemistry and Laboratory Medicine, University Medicine Greifswald, Greifswald, Germany 5Institute for Community Medicine, University Medicine Greifswald, Greifswald, Germany 6Unit of Periodontology, University Medicine Greifswald, Greifswald, Germany 7Friedrich-Loeffler Institute of Medical Microbiology, University Medicine Greifswald, Greifswald, Germany 8Institute of Bioinformatics, University Medicine Greifswald, Greifswald, Germany The Study of Health in Pomerania (SHIP) is a comprehensive population-based project that investigates the health dynamics of the population of Northeast Germany, a region historically associated with lower life expectancy. SHIP"s extensive data collection, including omics data, provides a longitudinal perspective on health trends, enabling the analysis of the complex interplay of biological and environmental factors. As one example, recently the human immune response to Staphylococcus aureus (S. aureus) was measured using the dilution- based multiplexed bead array to monitor antigen-specific IgG and IgA responses across the SHIP-TREND-0 cohort. The findings revealed a highly heterogeneous antibody response influenced by factors such as colonization status, sex, age and health indicators. This uncovered the complexity of the immune response to S. aureus, which ranges from unnoticed colonization to severe infections. We now expanded the study to track the antibody response to 34 pathogens associated with human infections, such as pneumonia, sepsis, endocarditis and joint infections. We recorded the profile of the antibody response to these bacteria in the SHIP-TREND cohort using the Infection-Array and then later examined its correlation with the composition of the human microbiome and disease risks. The Infection-Array comprises antigens of the secretomes of these 34 pathogens, including staphylococci, streptococci and members of the Enterobacteriaceae, which are conjugated to individual bead species. Incubation of serum samples in seven-step dilution series with the multiplexed beads allow the binding of the serum antibodies to their respective presented antigens over a high dynamic range. This binding is detected through the addition of a secondary R-phycoerythrin-conjugated antibody, which induces a fluorescence signal, measured with the Luminex® Flexmap 3D® technology. In total, 1307 serum samples of the SHIP-TREND (0 and 1) and SHIP-START (2 and 3) cohorts from different time points (0/1 as baseline and 2/3 as follow-up, respectively) were included in the analysis to also gain a longitudinal perspective on antibody titers in the general population. These studies offer valuable insights into the dynamics of human health and immune responses in the general population as well as data on the association of antibody signals with microbiome composition and different phenotypes. 365

366

P-I-0219 Challenges of scientific project management in the SMART-CARE consortium Roman Ladig1, Jeroen Krijgsveld1 1DKFZ Heidelberg, Proteomics of Stem Cells and Cancer, Heidelberg, Germany The recurrence of cancer remains a leading cause of cancer-related deaths. Comprehensive genomic characterization has improved risk stratification, but novel devices, methods, and AI-driven analyses of proteome and metabolome data can significantly enhance personalized medicine. SMART-CARE focuses on two main objectives: (1) Establishing a robust, standardized mass spectrometry pipeline for reproducible proteome and metabolome analyses, and (2) Using this pipeline in a systems medicine approach to improve tumor recurrence prediction. The project emphasizes: Establishing standardized procedures for sample collection, analyte extraction, and mass spectrometry processing ensures consistency and reproducibility across different research sites and over time. Combining proteome data with other omics data (e.g., genomics, metabolomics) provides a comprehensive view of the molecular landscape of tumors, enhancing prediction models for cancer recurrence. Using cutting-edge mass spectrometry and bioinformatics tools to analyze complex proteome datasets, involving high-resolution mass spectrometry and sophisticated data analysis pipelines. Identifying and validating proteomic biomarkers that can predict tumor recurrence, ensuring clinical relevance for personalized treatment strategies. Managing a large-scale scientific project like SMART-CARE involves various challenges: Interdisciplinary Collaboration: Integrating expertise from proteomics, metabolomics, computational biology, and clinical practice requires seamless communication and collaboration among diverse teams. Data Management: Handling and analyzing vast amounts of proteome data generated by mass spectrometry poses significant challenges. Efficient data management systems and computational tools are necessary for processing and interpreting complex datasets. Regulatory Compliance: Adhering to ethical guidelines and regulatory standards, particularly when dealing with patient samples and clinical data, requires meticulous oversight. Translational Research: Bridging the gap between basic research findings and clinical applications requires effective translation strategies, including validating predictive models in clinical settings to ensure research outcomes are applicable to patient care. Innovation and Technological Integration: Continuously integrating new technologies and methodologies into the research pipeline is vital to maintain the consortium's cutting-edge work. Outcome Dissemination: Effectively disseminating research findings to the scientific community, stakeholders, and the public is necessary for the project's impact, involving strategic planning for publications, presentations, and outreach activities. Risk Management: Identifying potential risks and developing mitigation strategies is essential for navigating uncertainties and ensuring the project's success. The SMART-CARE through a structured approach emphasizing collaboration, consortium addresses standardization, and continuous leveraging interdisciplinary expertise, the consortium aims to make significant strides in fighting cancer recurrence. innovation. By fostering a collaborative environment and these challenges 366

368

P-I-0220 Proteomics as a bridge between genomics and human health: a systematic approach to identify therapeutic targets in large population health cohorts like the UK biobank Ryan Lamers1, Ray Chen2, Cindy Lawley2 1Olink Proteomics, Scientific Affairs, Waltham, MA, United States 2Olink Proteomics, Strategic Accounts, Waltham, MA, United States Understanding the dynamics of the human proteome is crucial for identifying biomarkers to be used as measurable indicators for disease severity and progression, patient stratification, and drug development. It can also help us translate the impact our genetics has on more real-time health. The Proximity Extension Assay (PEA) translates protein information into actionable insights across large sample sizes in both healthy and disease samples. We have combined the PEA technology with automated sample preparation and a high-throughput sequencing readout for parallel measurement of >5,400 proteins for up to 172 samples at a time. Coverage includes low abundant proteins (e.g., cytokines), tissue leakage proteins (e.g., troponins), and overlap with high abundant proteins well served by mass spectrometry (e.g., globins). Characterizing the proteome alongside genetic and clinical data enables a protein quantitative trait loci (pQTL) framework to not only validate known clinical targets and identify new clinical targets but to also suggest repurposing opportunities of clinical candidates for new indications. We will discuss goals and results of large population health studies integrating proteomics, genomics and clinical data, including the UK Biobank-Pharma Proteomics Project (UKB-PPP). 368

369

P-I-0221 Assessing the role of protein variants in modulating protein-protein interaction networks by a quantitative interactomics approach Yushi Men1 1IGBMC, Illkirch-Graffenstaden, France Cell activity is the result of a series of biological processes. In this context, the protein-protein interaction (PPI) networks of signalling pathways play a vital role in mediating the entirety of cell activities due to their complexity and specificity (1). Pathogenic protein variants in key signaling proteins may lead to impairments of signalling pathways, which in turn will affect the orderly progress of cell activities. Meanwhile, non-pathogenic variants may also cause human individual differences, thereby causing phenotypic polymorphisms. To assess the role of protein variants in modulating protein-protein interaction networks we focused on the PDZ domain family in signal proteins. The human proteome contains 266 PDZ domains from 155 proteins and more than 4,000 PDZ binding motifs (PBM). PDZ-PBM interaction networks regulate multiple biological processes such as transport, ion channel signalling, cell-cell junctions, cell polarity and adhesion, etc. Functional perturbations in these networks can cause the development of cancers, degenerations of the nervous system and other diseases (2-4). We rationally designed 52 different variants from 5 wt PDZ domains, based both on allele frequency and structural criteria. All variants are part of the natural polymorphism, including disease-associated variants, human natural polymorphic variants and orthologs from close organisms. Different from common high- throughput qualitative assessments, we used a high-throughput quantitative approach developed in our group, Holdup assay with fluorescence readout, which quantifies PPI affinities (Kd) at an unprecedented scale and precision. We measured the affinities of the different PDZ for a set of 374 PBMs, and thus quantified a network of ~20,000 wt and variant PDZ-PBM interactions. We observed different types of interactome perturbations: unaffected interactome, destabilised interactome, isolated interactome changes, broad interactome reshuffling, etc. (Figure 1). Furthermore, these interactome perturbations depended on the variant categories: likely pathogenic variants vs. likely benign variants, variants in protein binding or non-binding sites, variants with different alleles frequencies. The long-term plan of this project is to further understand the impact of single variants on cell activities through understanding the impact of single variants on PPI networks, and ultimately describe pathogenic mechanisms with better precision. (1) Giancotti, "Complexity and specificity of integrin signalling", Nature Cell Biology, 2000, 2: E13–E14. (2) Lee & Zeng, "PDZ domains and their binding partners: structure, specificity, and modification", Cell Communication and Signaling, 2010, 8: 8. (3) Gogl, et al., "Quantitative fragmentomics allow affinity mapping of interactomes", Nature Communications, 2022, 13: 5472. (4) Nardella, et al., "Targeting PDZ domains as potential treatment for viral infections, neurodegeneration and cancer", Biol Direct. 2021, 16: 15. 369

371

P-I-0222 Follicular fluid peptidomics: understanding Oocyte reproductive potential Inez Mruk1, Paulina Czaplewska1, Krzysztof Łukaszuk2 1University of Gdańsk, Intercollegiate Faculty of Biotechnology, University of Gdańsk and Medical University of Gdańsk, Gdańsk, Poland 2Medical University of Gdańsk, Institute of Nursing and Midwifery, Gdańsk, Poland In vitro fertilization (IVF) is a widely utilized assisted reproductive technology (ART) that assists infertile couples in achieving pregnancy. However, current methods for evaluating the quality of oocytes collected for fertilization primarily rely on morphological assessments, which are inherently subjective and can lead to variability in outcomes. Improving the accuracy of oocyte quality assessment would not only address ethical concerns by minimizing the number of oocytes retrieved but also enhance the likelihood of successful fertilization and pregnancy. Follicular fluid (FF), the environment in which oocyte maturation occurs, offers a promising avenue for more objective assessment of oocyte quality. While previous research on FF has largely focused on its proteomic composition, the peptide fraction remains underexplored. Peptides, which are short chains of amino acids, play crucial roles in various biological processes, and their detailed analysis could provide deeper insights into oocyte health and maturation. To address this gap, our study utilizes advanced mass spectrometry to analyze the peptide fraction of FF, specifically targeting sequences derived from non-canonical protein variants. These non-canonical variants are protein sequences that deviate from the standard structures, arising from unconventional synthesis or processing mechanisms. They are known to influence cellular functions and have implications in various health and disease contexts. We will collect FF samples from women with endometriosis and Polycystic Ovary Syndrome (PCOS), as well as from healthy egg donors, to identify key components involved in intrafollicular signaling. This comparative analysis will be conducted using LC-MS/MS followed by comprehensive bioinformatics analysis. The goal is to uncover potential biomarkers that correlate with healthy oocytes and reproductive capacity. In this abstract, we present the findings from our preliminary study, which focused on developing a method for peptide extraction from FF and establishing a robust bioinformatics pathway for analyzing the resulting data. Our preliminary results are promising, demonstrating the feasibility of our approach and laying the groundwork for a more detailed qualitative and quantitative analysis. The anticipated outcomes of our research include the identification of specific peptides and non-canonical protein variants associated with oocyte quality in the context of endometriosis and PCOS. These findings have the potential to significantly improve the evaluation processes in IVF by providing more objective biomarkers for oocyte quality, ultimately enhancing reproductive outcomes for women undergoing ART. By advancing our understanding of the peptidomic landscape of FF, we aim to contribute to the development of more precise diagnostic tools and therapeutic strategies in reproductive medicine. 371

372

P-I-0223 Facilitating large-scale proteomics by combining automated sample preparation with short gradient LC-MS acquisitions Luca Räss1, Sandra Schär1, Christopher Below1, Ino Karemaker1, Sebastian Müller1, Roland Bruderer1, Lukas Reiter1 1Biognosys AG, R&D, Schlieren, Switzerland Mass spectrometry (MS)-based proteomics allows the comprehensive identification and quantification of proteins in various biological specimen such as cell lines, body fluids or tissues. To study complex samples for applications like biomarker discovery usually requires large sample cohorts to achieve sufficient statistical power. Labor-intensive manual sample preparation, coupled with long high-performance liquid chromatography (HPLC) gradients, restricts throughput and introduces technical variability. Workflows with increased throughput, standardized sample processing and cost-effectiveness are required to facilitate large-scale discovery proteomics and yield generalizable results. We present the integration of the latest technological advancements in sample preparation along with short gradient LC-MS acquisitions to achieve remarkable speed, depth, and reproducibility in proteomic studies. An automated end-to-end sample preparation workflow designed to process diverse biological specimen as input to MS-ready peptide output is showcased. An integrated liquid handling platform is utilized, enabling efficient parallelized processing of up to two 96-well plates in a single run. The automated sample preparation is combined with optimized dia-PASEF® LC-MS/MS measurements on the timsTOF high-throughput (HT) mass spectrometer achieved a throughput of 38 samples per day. The data was analyzed using Spectronaut's® directDIA+® workflow. As proof-of-concept ten distinct human cell lines were processed in analytical triplicates. The entire study, from initiating the sample preparation to concluding LC-MS acquisitions, was completed in 31 hours while reducing human input by almost 90% compared to manual processing. This workflow has the potential to be scaled up to a processing throughput of 1000 samples within a single month. The analyzed cell lines resulted in a total of 11"789 identified protein groups with an average of 9,547 proteins identified per sample demonstrating a remarkable coverage with ultra-short gradients. The median coefficient of variation (CV) for the detected protein groups was calculated at 7.9%, with 87% of all identified proteins exhibiting a CV below 20%, indicating an outstanding quantitative precision and reproducibility. In this comprehensive dataset biologically relevant proteins for cancer cell lines which are common targets for pharmaceutical intervention, such as KRAS, were consistently detected in all samples with high quality peptide profiles. Taken together, the workflow described facilitates proteomics experiments at scale by unlocking optimal synergies between sample preparation, LC-MS acquisition and data analysis. It enables to process and measure a substantial number of samples in a short time frame at reduced costs per sample while retaining high depth and precision. Consequently, the presented approach further extends the limits of large-scale studies enabling the systematic exploration of the proteome. 372

373

P-I-0224 Pediatric proteomics: a systematic approach to deciphering disease and enhancing early intervention Ericka Itang1, Alicia-Sophie Schebesta1, Susanne Pangratz-Führer2, Vincent Albrecht1, Christoph Klein2, Matthias Mann1, Johannes Müller-Reif1 1Max-Planck Institute of Biochemistry, Proteomics and Signal Transduction, Martinsried, Germany 2Dr. von Hauner Children's Hospital, LMU Munich, Department of Pediatrics, Munich, Germany Proteomics is a state-of-the-art method for biomarker discovery, yet its application as an analytical diagnostic tool is currently limited due to technical challenges. Characterizing and diagnosing rare diseases in the pediatric field with traditional single-readout methods is challenging and could benefit significantly from the routine acquisition of proteomics data. Our aim is to unlock insights into pediatric disease mechanisms by characterizing proteomic profiles in the pediatric population, thereby enhancing early diagnosis and intervention strategies through advanced mass spectrometry (MS)-based proteomics. To enable the diagnostic application of proteomics for individual patients, a thorough understanding of the population proteome is required. By mapping covariate-dependent variances, we can define and differentiate proteome signatures of disease from physiological and developmental processes. Among other large cohorts, we are systematically profiling the blood plasma and urinary proteomes of over 5,000 pediatric subjects using MS- based proteomics to establish a comprehensive proteome profile database. This involves a detailed analysis of population variance, disease-related proteomic changes, and the impact of monogenetic diseases. The resulting data will serve as a reference for future proteome analyses, aiming to integrate proteomic information into routine diagnostic practices. For specific disease phenotypes, additional specimens are analyzed to link patterns across different matrices. For example, co-analysis of plasma and intestinal biopsies in conditions such as very early onset intestinal bowel disease provides deeper insights into the disease mechanisms. Parallel in-vitro investigations using induced pluripotent stem (iPS) cells and organoids help to characterize monogenetic drivers and delineate molecular disease etiologies at the proteome level. These studies complement our profiling efforts, allowing us to explore the functional implications of identified proteomic changes. Preliminary data from plasma and urine proteomes reveal distinct proteomic signatures within the pediatric population, defining both normal and aberrant proteomic landscapes. Initial findings suggest novel covariate- dependent protein biomarkers and adverse proteomic trajectories associated with specific pediatric diseases. The integration of comprehensive proteomic profiling with in-vitro studies advances our understanding of pediatric disease biology. Our research has the potential to refine the precision of early diagnosis and pave the way for innovative developments in systemic and disease-overarching pediatric research. We believe our findings will support the development of tailored diagnostic and therapeutic strategies in precision medicine, benefiting both pediatric and adult populations. Pediatric patients with rare disease conditions can act as key drivers of novel findings, with broad relevance for physiology and disease biology. 373

374

P-I-0225 Profiling of the peripheral blood mononuclear cells proteome by shot-gun proteomics identifies alterations of immune components, proteolytic balance and mithocondrial metabolism in Glaucoma Irene Pandino1, Sara Giammaria1, Gabriele Antonio Zingale1, Grazia Raffaella Tundo2, Giulia Coco2, Gianluca Manni2, Giuseppe Grasso3, Gloria Roberti1, Manuele Michelessi1, Carmela Carnevale1, Francesco Oddone1, Diego Sbardella1 1IRCCS Fondazione Bietti, Rome, Italy 2University of Tor Vergata, Rome, Italy 3University of Catania, Catania, Italy Glaucoma is a chronic optic neuropathy characterized by degeneration of the Retinal Ganglion Cells (RGCs) and the optic nerve. As of 2024, glaucoma is the second cause of irreversible blindness on a global scale. The pathogenesis of the disease is largely uncharacterized yet, but, several recent studies have pointed out phenotypical and molecular alterations of circulating immune cells, in particular T-lymphocytes, in preclinical models and in human subjects. Therefore, we have here undertaken a characterization of the proteome of peripheral blood mononuclear cells (PBMC) of subjects diagnosed with primary open angle glaucoma (POAG) and cataract, enrolled as healthy controls. The study was authorized by the local ethic committee and subjects asked to signa written informed consent. PBMCs (n=12 POAG, n=12 cataract; age: 71.8±7.5y POAG, 70.1±12.7 cataract; comparable gender ratio) were isolated from peripheral blood by Ficoll-Histopaque sedimentation and lysed in urea buffer. 200 µg of proteins were reduced, alkylated and digested with lysil-c endopeptidase (1:2000) and trypsin (1:50 ratio). Peptides (750 ng) were injected in a Orbitrap Exploris 240 online with a nanoUHPLC. The analysis was run by label free quantification using three different acquisition modalities: DDA, DIA and DIA-GPF. Proteins were searched by Proteome Discoverer (DDA) and DIA-NN (for DIA library free searches). In the case of GPF-refined library generation, staggered GPF (n=6) files were submitted to DIA-NN. Output data were analysed by R. The following discussion refers to DIA data under library free searches. Proteins were filtered according to: q- value (Lib.Q.Value in the output matrix) ≤0.01; identification with at least 1 proteotypic peptide; identification of the protein in ≥ 8 out of 12 subjects of the same experimental group. Quality controls (e.g., Coefficient of Variation, distribution of data, blood proteins contamination, PCA plot etc.) were checked as optimal. Even applying strict criteria, the approach identified 4001 proteins common to POAG and cataract groups. Additional 465 proteins were identified as POAG-specific and 118 as cataract-specific. Volcano plot (0.5<log2Fold Change<0.5, p<0.05) indicated that 353 proteins were found as upregulated in POAG PBMCs, while 112 proteins were identified as downregulated in the same experimental group. GeneOntology-enrichment and KEGG pathway analysis showed that POAG PBMCs suffered from possible alterations referable to mechanisms of immune system polarization, transcription and translation processes, extracellular proteolysis, autophagy, as well as mitochondrial and redox metabolism. Main proteins were studied by Western blotting to confirm proteomics data. Further studies are needed to figure out the distribution of PBMC subsets along with their polarization and metabolic trajectories to address the role of these factors in mediating the neuroinflammation state that accompanies glaucoma progression. 374

375

P-I-0226 A network analysis approach to identify protein clusters and multi-Omic determinants of childhood respiratory- related diseases Nicole Prince1,2, Rachel S Kelly1,2, Kevin M Mendez1,2, Ofer Levy1,3, Scott T Weiss1,2, Augusto A Litonjua4, Jessica A Lasky-Su1,2 1Harvard Medical School, Boston, MA, United States 2Brigham and Women's Hospital, Channing Division of Network Medicine, Boston, MA, United States 3Boston Children's Hospital, Precision Vaccines Program, Boston, MA, United States 4Golisano Children's Hospital at Strong, University of Rochester Medical Center, Division of Pediatric Pulmonary Medicine, Rochester, NY, United States The first year of represents a critical period of immune development that can impact health trajectories and the risk of developing respiratory-related diseases, such as asthma, recurrent infections, and eczema. However, the underlying biology between perturbed immune protein profiles and subsequent disease development remains unknown. Using weighted gene correlation network analysis (WGCNA), we derived clusters of immune- mediating proteins based on plasma protein profiling data from children at age 1 (N=294) in the Vitamin D Antenatal Asthma Reduction Trial (VDAART). We then used regression analysis to understand the relationships between protein clusters during early life immune development and subsequent clinical outcomes. We further characterized genomic and environmental determinants of clusters and investigated downstream biochemical impacts through the metabolome. WGCNA identified four protein clusters at age 1 year associated with incidence of childhood asthma and/or recurrent wheeze (Padj range: 0.02-0.03), respiratory infections (Padj range: 6.3x10-9-2.9x10-6), and eczema (Padj=0.01); three of these protein clusters were associated with at least one environmental exposure (Padj range: 2.8x10-10-0.03) and metabolomic alteration of at least one pathway (Padj range: 2.8x10-6-0.04). No significant genetic risk factors were identified for any protein cluster. Overall, these findings suggested protein clusters derived during the critical immune development window contribute to development of respiratory-related diseases, which may enlighten new strategies to clinically identify vulnerable groups based on immune protein profiling. Fig. 1 (optional) 375

376

P-I-0227 Plasma proteins associated with COVID-19 in unvaccinated and vaccinated Puerto Ricans Ana Rodriguez1,2, Lester Rosario1, Yadira Cantres2, Kelvin Carrasquillo2, Alexandra Rosa2, Veronica Rivera2, Eduardo Tosado-Rodríguez2, Loyda Méndez3, Abiel Roche-Lima2, Jorge Bertran4, Loyda Meléndez1 1University of Puerto Rico Medical Science Campus, Biomedical Sciences, San Juan, Puerto Rico 2University of Puerto Rico, Center for Collaborative Research in Health Disparities, San Juan, Puerto Rico 3Ana G. Mendez University, Science and Technology, Carolina, Puerto Rico 4Auxilio Mutuo Hospital, Infectious Diseases, San Juan, Puerto Rico Viral strains, age, and host factors are associated with variable immune responses against SARS-CoV-2 and disease severity. Puerto Ricans have a genetic mixture of races: European, African, and Native American. We hypothesized that unique host proteins/pathways are associated with COVID-19 disease severity in Puerto Rico and that vaccination reverts these proteins' dysregulation. Following IRB approval, 95 unvaccinated men and women aged 21-71 years-old were recruited in Puerto Rico from 2020-2021. Plasma samples were collected from COVID-19-positive subjects (n = 39) and COVID-19-negative individuals (n = 56) during acute disease. COVID-19-positive individuals were stratified based on symptomatology as follows: mild (n = 18), moderate (n = 13), and severe (n = 8). An additional subgroup of vaccinated participants before COVID-19 infection (n=11) was compared to vaccinated COVID-19-negative individuals (n=15). Quantitative proteomics was performed using Tandem Mass Tag (TMT) labeling in plasma samples. Labeled peptides were subjected to LC/MS/MS and analyzed by Proteome Discoverer, Limma software, and Ingenuity Pathways Analysis (IPA). Cytokines were also quantified using a human cytokine array. Proteomics analyses of severely affected COVID-19-positive individuals revealed 2 upregulated and 54 downregulated proteins. In vaccinated individuals we did not find downregulated proteins. Cytokine immunoassay showed that IL-1Ra, IP-10, and TNF-⍺ levels increased while PDGF decreased with COVID-19 in unvaccinated individuals while only IP-10 was upregulated in vaccinated individuals. This study demonstrates that proteomics studies help uncover potential host predictors of COVID-19 severity in Puerto Ricans modulated by vaccination. Supported in part by UPR COVID-19 Emergency Funds, National Institutes of Health National Institute of General Medical Sciences (NIGMS), PR-INBRE-Institutional Developmental Award (IDEA) P20GM103475, Research Infrastructure Core components – National Institute of Minority Health and Health Disparities (NIMHHD) U54MD007600, Translational Proteomics Center, and Bioinformatics and Health Informatics, "D-SPAN"- F99NS11345 and K00NS11345. 376

377

P-I-0228 Proteomic profiling of cancer-associated fibroblasts in squamous cell carcinoma Sibilla Sander1, Jörn Dengjel1 1University of Fribourg, Biology, Fribourg, Switzerland Cutaneous squamous cell carcinoma (cSCC) is the second most prevalent type of non-melanoma skin cancer, characterized by the abnormal, accelerated growth of squamous cells in the epidermis and metastatic potential. Understanding cancer progression and metastasis has shifted towards acknowledging the influence of the tumor microenvironment (TME) rather than solely relying on cancer cell-autonomous defects. The TME significantly influences tumorigenesis, development, and metastasis. It has been shown that among TME components, cancer-associated fibroblasts (CAFs) play a crucial role, exhibiting dual functions of promoting or restraining cancer progression. Several markers, including α smooth muscle actin (αSMA), platelet-derived growth factor receptor α(PDGFRα), and fibroblast activation protein (FAP), have been extensively used to isolate CAF populations. However, these markers have inherent limitations such as low specificity, raising concerns about their ability to identify all CAFs or merely a specific subset within the broader CAF population. Our goal is to gain insight into cSCC-associated CAFs and investigate the dysregulation of molecular pathways. For this, we aim to identify their proteomic profile and eventual subtypes. 377

378

P-I-0229 Genetic atlas of human protein complexes Xia Shen1,2,3, Ting Li2 1Karolinska Institute, Medical Epidemiology and Biostatistics, Stockholm, Sweden 2Fudan University, Shanghai, China 3University of Edinburgh, Edinburgh, United Kingdom Understanding protein complexes is crucial as they perform vital cellular functions through coordinated interactions among multiple proteins, unlike individual protein subunits. However, quantification of protein complexes is a highly challenging task. Here, we develop and apply both experimental assays and intelligent computational methods to generate the first high-throughput quantification of hundreds of protein complexes in human cohorts. This study significantly advances proteomics by offering a detailed genetic atlas of human protein complexes. We identified thousands of genetic variants associated with these complexes, revealing hundreds of cis- and trans-associations. These findings highlight the strong effects of protein complex QTL (pcQTL) and their role in complex genetic regulation. Our analysis showed that higher heritability correlates with more detected pcQTL and greater phenotypic variance explained. We discovered significant genetic links between protein complexes and human traits and diseases through colocalization and Mendelian randomization analyses. For instance, specific protein complexes showed protective effects against Alzheimer's disease and associations with immune-related conditions and cardiovascular diseases. To explore therapeutic potential, we investigated protein markers as drug targets. Among the identified protein complexes, many protein subunits were linked to various diseases and health conditions. We identified over 100 drugs targeting these proteins, suggesting many drug-protein-complex-phenotype combinations for potential repurposing and new therapeutic strategies. In summary, our research presents a comprehensive genetic atlas of human protein complexes, emphasizing the importance of studying protein complexes for understanding cellular functions, disease mechanisms, and therapeutic targets. This work paves the way for future innovations in diagnostics and precision medicine. Fig. 1 (optional) 378

379

P-I-0230 Every hour counts: mapping neonatal blood plasma proteome changes during the first 2 days of life at hourly resolution Hanno Steen1, Tue Bjerg Bennike1,2, Alec Plotkin3, Jensen Pak3, Arthur Viodé1, Christoph N. Schlaffner4,5, Olubukola T. Idoko6,7, Oludare A. Odumade3,8, Dylan Nicholas Tabang1, Rym Ben-Othman9, Joann Diray-Arce3,8, Alex Ozonoff3,8,10, Judith A. Steen4,8, EPIC-HIPC Consortium3, Tobias R. Kollmann9, Beate Kampmann6,7,11, Ofer Levy3,8,10, Kinga K. Smolen3,8 1Boston Children's Hospital/Harvard Medical School, Department of Pathology, Boston, MA, United States 2Aalborg University, Department of Health Science and Technology, Aalborg, Denmark 3Boston Children's Hospital, Precision Vaccines Program, Dept. of Pediatrics, Boston, MA, United States 4Boston Children's Hospital, F.M. Kirby Neurobiology Center, Boston, MA, United States 5Hasso Plattner Institute for Digital Engineering, Digital Engineering Faculty, Potsdam, Germany 6Medical Research Council Unit The Gambia at the London School of Hygiene and Tropical Medicine, Vaccines & Immunity Theme, Banjul, The Gambia 7London School of Hygiene and Tropical Medicine, Faculty of Infectious and Tropical Diseases, London, United Kingdom 8Harvard Medical School, Boston, MA, United States 9Telethon Kids Institute, Perth, Australia 10Broad Institute of MIT and Harvard, Cambridge, MA, United States 11Charité – Universitätsmedizin Berlin, Charité Centre for Global Health/Institute for Int'l Health, Berlin, Germany Birth represents a dramatic shift from a quasi-sterile intrauterine life to an antigen-rich extrauterine environment. The transition from the fetal state with induced maternal tolerance to a state where new immune challenges must be met, necessitates a significant immune system adaptation. The main components of the immune system circulate in plasma. Given the plasma"s major immune modulatory role, these adaptations are reflected by changes to the plasma proteome composition. We collected 906 peripheral blood samples from 679 healthy term newborns at day of life 0 and day of life 1 as part of an ongoing newborn vaccination study. Carefully recording the time of birth and time of blood collection allowed us to calculate the exact age of the newborn in hours, rather than the commonly applied daily resolution. We used a multiplexed antibody-based assay for monitoring cytokines and chemokines, and a targeted liquid chromatography-mass spectrometry assay for monitoring the plasma proteome. We utilized a sample-sparing strategy, which was a prerequisite for this study given the limited nature of newborn blood sample volumes. This 2-pronged strategy allowed us to monitor the postnatal dynamics of nearly 400 proteinaceous analytes in plasma during the first 48 hours of life at an hourly resolution, with a throughput of ~100 samples per day. The analysis unveiled distinct and dynamic changes in developmental trajectories of the proteome. The detailed study of these crucial first 48 hours post birth at an hourly resolution provided more accurate information about fold changes and dynamics, insights into the ordering of protein changes, and co-regulations – all these findings could not be gleaned from the conventional time course analysis at day resolution, which compress complex dynamics over 24 hours into a single measurement with large error bars. Notable observations included 10-fold increases of haptoglobin during the first 24 hours of life and complex dynamics for CXCL8, with little net change at conventional day resolution. This first-of-its-kind high resolution plasma proteome ontogeny study delineated postnatal immune system development in healthy newborns immediately after birth, and mapped the temporal molecular roadmap including several critical periods during the first 48 hours of life, which possibly demarcate important postnatal physiological processes such as the postnatal closure of the ductus arteriosus, which bypasses the liquid-filled non-functioning lungs in fetuses. 379

380

P-I-0231 Prediction of type 1 diabetes using proteomics Tomi Suomi1, M. Karoliina Hirvonen1, Inna Starskaia1, Tommi Välikangas1, Robert Moulder1, Riitta Lahesmaa1, Laura Elo1 1University of Turku, Turku, Finland Type 1 diabetes (T1D) is a chronic autoimmune condition where the immune system mistakenly attacks and destroys the insulin secreting β-cells in the pancreatc islets of Langerhans. T1D affects millions worldwide, with an onset often happening already in childhood. Existing methods to assess T1D are predominantly for later stages of the disease, assessing disease symptoms, or detecting specific autoantibodies. Detection of earlier markers is important for development of improved treatment strategies. Mass spectrometry proteomics provides a compelling option for such biomarker discovery, thanks to its capabilities to provide functional information beyond genomic studies, providing a more direct measure of the phenotype. Importantly, longitudinal proteome measurements allow revealing dynamic changes in the proteome over time, enabling early detection of diseases by identifying protein changes that precede clinical symptoms and allowing to account for within-individual variability. To fully utilize such longitudinal proteome measurements, we have developed a robust reproducibility optimization approach, which builds on our previously developed reproducibility-optimized statistical framework. Using this approach, we have discovered candidate protein markers associated with progression of T1D, including those associated with the very early stages of T1D. In particular, we have identified several proteins that are indicative of future decline in β-cell functionality. To furher bring the analysis to single-cell level, we have utilized mass cytometry, which can measure dozens of proteins simultaneously at the single-cell level, allowing for detailed profiling of complex biological samples. We have developed a comprehensive toolkit for the analysis of mass cytometry data, revealing different endotypes in children developing T1D, with distinct characteristics of early immune responses in blood, depending on the type of autoantibody that appears first. These findings the pave way for more personalized threatment of T1D in the future and highlight the potential of proteomics in biomarker discovery. 380

381

P-I-0232 Proteomic profiling of CSF and serum: non-linear progression and staging biomarkers for preclinical Alzheimer's disease Jiayi Zhang1, Yuanyu Huang1, Yu Guo2, Yaru Zhang2, Huali Shen1, Jintai Yu2 1Institutes of Biomedical Sciences, Fudan University, Shanghai, China 2Department of Neurology and National Center for Neurological Disorders, Huashan Hospital, Shanghai, China The disease spectrum of Alzheimer's Disease (AD) is characterized by a continuum that ranges from asymptomatic phases to more overt stages of cognitive decline. Early intervention during these initial stages can significantly improve patient outcomes. Staging AD based on biomarkers is crucial as it allows for the identification of disease changes before clinical symptoms manifest, thereby advancing precision medicine approaches for AD. In our study, we utilized data independent acquisition (DIA) mass spectrometry to analyze 174 paired samples of serum and cerebrospinal fluid (CSF). Our discovery cohort included samples from three distinct preclinical AD stages (Stage I, II, III), AD-related mild cognitive impairment (MCI), and healthy controls. From these samples, we identified 2,988 CSF protein groups and 414 serum protein groups. Notably, 965 CSF proteins and 22 serum proteins demonstrated significant changes and can be considered as potential biomarkers for monitoring AD progression or for its early diagnosis. These potential biomarkers offer insights into the molecular dynamics at play during the earliest phases of AD, suggesting potential targets for therapeutic intervention. These findings emphasize the non-linear path of pathological changes during the preclinical development of AD and provide a deeper molecular and systemic understanding that could help the management and treatment of Alzheimer's Disease. Future validation of these biomarkers in external datasets will be imperative to confirm their utility and reliability in clinical settings. 381

382

Spatial and imaging proteomics (P-I-0233 to P-I-0273) P-I-0233 Spatial dynamic proteotyping of surfaceome proteostasis using DynCSC Jonas B. Albinus1,2, Audrey van Drogen1,2,3, Patrick G. A. Pedrioli1,2,3, Arend Keller1,2, Sandra Goetze1,2,3, Sebastian N. Steiner1,2, Jacqueline Hammer1,2, Kathrin Kläsener4,5,6, Michael Reth4,5,6, Bernd Wollscheid1,2,3 1ETH Zürich, Department of Health Science and Technology, Institute of Translational Medicine, Zurich, Switzerland 2Swiss Institute of Bioinformatics (SIB), Lausanne, Switzerland 3ETH PHRT, Swiss Multi-Omics Center (SMOC), Zurich, Switzerland 4University of Freiburg, Biology III, Faculty of Biology, Freiburg, Germany 5University of Freiburg, Centres for Integrative Biological Signalling Studies CIBSS and BIOSS, Freiburg, Germany 6University of Freiburg, Department of Rheumatology and Clinical Immunology, Medical Center, Freiburg, Germany involves post-translational modifications, protein-protein Introduction: Protein homeostasis, known as proteostasis, ensures cellular functionality by maintaining a balanced and functional proteome. A complex network of regulatory proteostasis mechanisms maintains the proteome, which interactions, and subcellular localisations. Studies have shown that spatial proteostasis in subcellular localisations cannot be elucidated based on bulk proteome studies and that spatial enrichment strategies with high specificity are needed. Spatial dynamic studies of the cell surface residing protein pool, the surfaceome, have revealed their disappearance, an essential aspect of surfaceome proteostasis. However, these studies did not include the appearance of new cell surface proteins, which is vital for fully understanding the proteostasis mechanisms of the surfaceome. Here, we developed the Dynamic Cell Surface Capturing (DynCSC) strategy, which provides system-wide insights into surfaceome proteostasis and how this information can be leveraged for regulating the surfaceome. Method: The DynCSC strategy employs a modified triple-pulsed SILAC (Stable Isotope Labelling of Amino Acids in Cell culture) approach to capture the appearance and disappearance of cell surface receptors independently. The strategy is combined with a cell surface protein enrichment of either the global surfaceome or within a cell surface protein community using the light-mediated proximity technology LUX-MS. Using the B cell as a model system, we applied DynCSC to analyse surfaceome proteostasis using spatial proteotyping. Results: Our findings demonstrate that surfaceome proteostasis can not be fully understood based on bulk proteome data alone. Spatially resolved cell surface proteins exhibit slower dynamics than when captured without spatial resolution. With the application of the DynCSC strategy, we further reveal that cell surface proteins reside in different temporal states and can be clustered in distinct dynamic groups. These temporal clusters are not governed by protein size, number of N-glycosylation sites and transmembrane domains. Moreover, we identify different temporal turnover kinetics within members of the B cell receptor (BCR) community, indicating independent turnover regulation within this specific protein community. Upon knocking out BCR neighbours CD81 or CD19, we observe alterations in surfaceome composition and proteostasis globally and within the BCR community locally. We specifically highlight dynamic alterations found only within the BCR community, indicating community-specific alterations. Conclusion: DynCSC technology enables the analysis of spatially restricted proteostasis and provides new insights into the turnover of surfaceome protein communities. 382

383

P-I-0235 Integrating MALDI-MSI and imaging mass cytometry to reveal intratumoral heterogeneity in breast cancer subtypes induced by the tumor microenvironment Klára Brožová1, Silvester J. Bartsch1, Rebecca Herzog1, Brigitte Hantusch1, Thomas H. Helbich1, Katja Pinker- Domenig1,2, Lukas Kenner1,3, Klaus Kratochwill1 1Medical University of Vienna, Vienna, Austria 2Memorial Sloan Kettering Cancer Center, New York, NY, United States 3University of Veterinary Medicine, Vienna, Austria Breast cancer (BC) is a global health issue, affecting a high proportion of the female population. Spatial proteomics enables the study of the proteome within tissue context, offering a promising approach to unravel the biological processes contributing to BC heterogeneity that can lead to ineffective therapies. To study this heterogeneity, we use a combination of untargeted MALDI imaging (MSI), bulk tissue LC-MS/MS, and targeted imaging mass cytometry (IMC). This approach aids in identifying diagnostic and prognostic biomarkers masked by tissue heterogeneity and allows precise mapping and quantification of selected biomarkers. The co- registration of these techniques can enhance our ability to correlate molecular and spatial data, leading to comprehensive insights into tumor biology and potential therapeutic targets. We established an optimized protocol for sample preparation of gelatin and paraffin-embedded BC tissue for MALDI imaging, configured a high multiplex IMC-panel, and performed data integration to analyze BC subtypes. Human BC cell lines (MCF-7, SKBR-3, MDA-MB-231) were inoculated into female athymic BALB/c-nude mice. Excised tumors were embedded in gelatin, paraffin, and liquid nitrogen. For MSI, we optimized pre-analytical and analytical parameters to enhance the signal-to-noise (S/N) ratio and enable protein identification. Following MSI, samples were stained for standard histological evaluation. Adjacent tissue sections were used for IMC. Proteins were also identified by TMT-based LC-MS/MS from homogenized tissue. Following the establishment of the MSI protocol, peptide signals were analyzed using spatial segmentation and uniform manifold approximation and projection (U-MAP) plots. Segmentation maps based on MSI aligned with expert-based histological assessment were generated. Unsupervised U-MAP plots enabled visualization of multidimensional information, allowing segmentation of different tumor, stroma, and necrotic tissue regions. Molecular BC subtypes were clearly distinguishable by MSI. In the bulk MS analysis ~17 000 proteins were identified, providing a source for confident matching of peptide signals from MSI. Based on the expression analysis of the TMT data and comprehensive literature review, we selected 28 markers for IMC. The targeted analysis revealed distinct marker expression patterns, with some markers predominantly expressed in specific subtypes. This detailed mapping highlighted unique proteomic landscapes across the tumor subtypes, including variations in pathways related to cell proliferation, apoptosis, hypoxia and immune response, thereby advancing our understanding of BC heterogeneity. Using the established workflow, peptide signatures not only differentiate histologically confirmed tumor regions but also reveal sub-regions that are indistinguishable by standard methods, demonstrating the potential of MSI as a tool for translational BC research. 383

384

P-I-0236 Myelin-axon interface vulnerability uncovered by subcellular proteomics and imaging of Alzheimer's human brain Yifei Cai1, Iguaracy Pinheiro-de-Sousa2, Mykhaylo Slobodyanyuk3,4, Fuyi Chen5, Tram Huynh5, Jean Kanyo5, Peiyang Tang5, Lukas A. Fuentes5, Amber Braker5, Rachel Welch5, Anita Huttner5, TuKiet Lam6, Evangelia Petsalaki2, Jüri Reimand3,4,7, Angus C. Nairn8,9, Jaime Grutzendler10,1 1Yale University, Department of Neurology, New Haven, CT, United States 2European Molecular Biology Laboratory (EMBL), European Bioinformatics Institute, Cambridge, United Kingdom 3Ontario Institute for Cancer Research, Computational Biology Program, Toronto, Canada 4University of Toronto, Department of Medical Biophysics, Toronto, Canada 5Yale University, New Haven, CT, United States 6Yale University, Yale/NIDA Neuroproteomics Center, New Haven, CT, United States 7University of Toronto, Department of Molecular Genetics, Toronto, Canada 8Yale University, Department of Psychiatry, New Haven, CT, United States 9Yale University, Department of Pharmacology, New Haven, CT, United States 10Yale University, Department of Neuroscience, New Haven, CT, United States Myelin ensheathment is critical for rapid axonal electrical conduction, metabolic support and neuronal plasticity. Myelin and axonal pathology are observed in Alzheimer’s disease (AD), but the precise mechanisms are not well understood. Using novel proximity labeling subcellular proteomics of the myelin-axon interface in postmortem human brains, we discovered dysregulation of signaling pathways and ligand-receptor pairs in AD, including β-amyloid processing, axonal outgrowth and lipid metabolism. Expansion Microscopy revealed intrusion of amyloid fibers into the internodal peri-axonal space and paranodal channels. Additionally, we observed aberrant myelination and paranode positioning at plaque-associated axonal spheroids as well as an overall reduction in paranode density without overt myelin loss. These findings highlight the myelin-axon interface as a potential site of significant protein aggregation and disruption of neuro-glial communication in AD. The proteomics dataset has uncovered the molecular architecture of the myelin-axon interface, paving the way for future mechanistic studies in health and disease. 384

385

P-I-0237 Proteome dynamics in iPSC-derived human dopaminergic neurons Claudia Cavarischia-Rega1, Karan Sharma2, Julia C. Fitzgerald2, Boris Macek1 1University of Tübingen, Quantitative Proteomics, Tuebingen, Germany 2Hertie Institute for Clinical Brain Research, Department of Neurodegeneration, Tuebingen, Germany Dopaminergic neurons participate in fundamental physiological processes and are the cell type primarily affected in Parkinson"s disease (PD). Their analysis is challenging due to the intricate nature of their function, their involvement in diverse neurological processes, their heterogeneity and localization in deep brain regions. Consequently, most of the research on the protein dynamics of dopaminergic neurons has been performed in animal cells ex vivo. Here we use iPSC-derived, human mid-brain specific dopaminergic neurons to study general features of their proteome biology. In order to maximize the proteome coverage, we first performed in-depth identification of proteins in hDaN cultures by high-pH fractionation followed by measurement using Data- Independent Acquisition (DIA). With this approach, we cover the proteome to a depth of 9,409 proteins, which we estimate to present up to 80% of the total expressed cellular proteome. We next focused on investigation of protein turnover using dynamic SILAC labeling to measure the half-life of more than 4,300 proteins. We report uniform turnover rates of conserved protein complexes and identify several outliers in the mitochondrial outer membrane and mitophagy pathway. We use differential dynamic SILAC labeling in combination with microfluidic devices to analyze local protein synthesis and transport between axons and soma. Briefly, neurons were seeded in a two-well microfluidic device (Xona Microfluidics) and pulsed with heavy amino acids in both wells simultaneously: Lys4+Arg6 in the somatodendritic well and Lys8+Arg10 the axonal well. Samples were harvested after different time points (12- 120 hours), allowing temporal analyses of label incorporation. We report 105 potentially novel axonal markers and detect translocation of 269 proteins between axons and the soma in the time frame of our analysis (120 hours). Importantly, we provide evidence for local synthesis of 154 proteins in the axon and their retrograde transport to the soma, among them several proteins involved in RNA biology such as adenosine deaminase ADAR and RNA helicase DHX30, involved in the assembly of mitochondrial ribosomes. We are currently applying this workflow to investigate how oxidative stress, one of hallmarks of PD, influences protein synthesis, trafficking and turnover. Our study provides a workflow and resource for future applications of quantitative proteomics in iPSC- derived human neurons. 385

386

P-I-0238 Hypothesis-free discovery of the subcellular proteome: a total-sync ultra-content microscopic opto-biotinylation system Tina Chong1, Jung-Chi Liao1, Yi-De Chen1, Chantal Hoi Yin Cheung1, Hsiao-Jen Chang1, Yong-Da Sie1, Ellen Chung1, Jeng Ting Chen1 1Syncell Inc., Taipei, Taiwan Studying proteomes within subcellular structures presents significant challenges, particularly for structures that are membrane-less or cannot be isolated from cells. Addressing this hurdle, MicroscoopTM emerges as a groundbreaking technology capable of precisely capturing proteins at specified subcellular regions of interest (ROIs) with nanoscale precision. This sophisticated system integrates microscopy, optics, mechatronics, photochemistry, and deep learning to enable high-content in situ photolabeling. With MicroscoopTM, proteins are precisely biotinylated within user-defined cellular organelles, granules, or cell-cell contact surfaces under a microscope, utilizing directed photochemistry within each field of view (FOV). This process is automatically repeated across thousands of FOVs to photo-label cellular structures sharing similar morphological features. Subsequently, ample biotinylated proteins are obtained for streptavidin pulldown and mass spectrometry analysis. The robustness of this approach is demonstrated through the successful mapping of the human cellular nucleus proteome, identifying over 1000 nuclear proteins with a specificity exceeding 90%. Further analysis reveals comprehensive coverage of nuclear complexes and identification of low-copy-number proteins. In one study using MicroscoopTM technology to explore the nucleolus proteome, 97 of the top 100 abundant proteins are confirmed to be originated from the nucleolus. In another investigation focusing on stress granules (SG) proteome, we identified 2,614 proteins, including 200 with low copy numbers. Although the specificity is moderate, with only 66% of the top 50 abundant proteins mapped with the known stress granules database, functional enrichment analysis highlights 13 non-SG associated proteins as high-confidence core interactors within the stress granule network. Among them, 11 proteins (PDLIM7, EIF3CL, YWHAE, RPSA, MTA2, UGDH, DDX17, ANLN, PSMD3, PSMA6, and MCM2) are confirmed to co-localize with G3BP1 through immunostaining, thus elevating the specificity eventually to 96% SG association. Overall, our study illustrates that MicroscoopTM enables hypothesis-free, comprehensive mapping of subcellular proteomes at user-defined regions of interest, significantly advancing cell biology by revealing new proteins or biomarkers. 386

387

P-I-0239 High-confidence mapping of the human peroxisomal proteome through multifaceted spatial proteomics Hirak Das1, Ruth Carmichael2, Serhii Chornyi3, Sven Spielhaupter4, Claudio Costa5, Ann-Kathrin Richard1, Wignand W. D. Mühlhäuser6, Julian Bender1, Ann-Kathrin Staudt1, Bettina Knapp7, Silke Oeljeklaus1, Alexander Buchberger4, Marc Fransen5, Hans Waterham3, Michael Schrader2, Bettina Warscheid1 1University of Würzburg, Chair of Biochemistry II, Wuerzburg, Germany 2University of Exeter, Department of Biosciences, Exeter, United Kingdom 3Amsterdam University Medical Center, Dept. of Gastroenterology Endocrinology Metabolism, Amsterdam, Netherlands 4University of Würzburg, Chair of Biochemistry I, Wuerzburg, Germany 5KU Leuven, Department of Cellular and Molecular Medicine, Leuven, Belgium 6Scitaris GmbH & Co. KG, Berlin, Germany 7University of Freiburg, Freiburg, Germany Objectives: Peroxisomes perform vital metabolic functions, such as β-oxidation of fatty acids or synthesis of lipids, plasmalogens and bile acid. In addition to the detoxification of hydrogen peroxide, they also play role in regulating immune responses and cellular signalling. Importantly, they share metabolic pathways and form membrane contact sites with other organelles. Deficiencies in peroxisome biogenesis or function are linked to severe, often lethal human diseases (e.g. Zellweger spectrum disorders). To fully understand peroxisomal functions and their ways of communication and metabolite exchange, knowledge of the complete peroxisomal proteome is crucial. However, due to small size of peroxisomes, low abundance, membrane contact sites and multi-localized protein (MLP) content, defining the proteome is challenging. Since previous spatial proteomics studies did not properly address peroxisomes, we aim at establishing a multifaceted high-resolution spatial proteomics and data analysis pipeline tailored to human peroxisomes. Methods: We established a workflow for the separation of peroxisomes from human cells (HEK293) employing subcellular and density gradient fractionation. Using label-free LC-MS, we profiled the distribution of organelles across all fractions using organellar marker proteins and peroxisomal MLPs. Through high-resolution protein correlation profiling, we identify known and new peroxisomal proteins. To validate peroxisomal candidates, we employed a new SILAC-based comparative spatial proteomics strategy and fluorescence microscopy. We performed multi-modal integration of our dataset with auxiliary data from other spatial/interaction proteomics datasets and single protein studies to boost confidence of inference for our candidates, score them using an ensemble machine learning framework, and catalogue all proteins to make for the first time a comprehensive inventory of human peroxisomal proteins, which we termed PeroCarta. Results: We profiled more than 8,800 proteins and identified most known peroxisomal proteins and novel candidates. Based on orthogonal validation and integrated datasets from other studies, we compile a first draft of the human peroxisomal proteome with 242 proteins as peroxisomal of which 216 are identified in our study and includes 82 new candidates. Our data confirm that peroxisomal proteins are mainly MLPs, with the bulk of candidates shared with mitochondria or the cytosol. Selected candidates including potential contact- site/tethering proteins have been investigated, which provide new insight into peroxisome functions in a cellular context. Conclusion: Taken together, results from our multifaceted spatial proteomics study represent a milestone towards defining the human peroxisomal proteome including MLPs. Our newly established PeroCarta, which includes all known and new peroxisomal proteins, will serve as point of reference for future studies of human peroxisomes in health and disease. 387

389

P-I-0240 Mapping molecular complexity with spatially-resolved proteomics Simon Davis1, Connor Scott1, Philip Charles1, Aleema Iqbal1, Benedikt Kessler1, Olaf Ansorge1, Roman Fischer1 1University of Oxford, Oxford, United Kingdom Background: Spatial protein abundance profiles within tissues are a key factor in understanding disease pathology and cellular function. These profiles are typically studied by antibody-based imaging techniques that provide high spatial resolution, but can only probe a limited number of proteins. In order to more precisely define molecular phenotypes in disease, there is a need for unbiased, quantitative technology capable of mapping the expression of many hundreds to thousands of proteins within tissue structures. Laser capture microdissection (LCM) in combination with high-throughput LC-MS/MS-based proteomics is well placed to meet this need. Methods: Tissue voxels were isolated from 10 µm-thick sections of a human brain tumour in a gridded pattern as resolutions as low as 40 µm. Spatially variable proteins were detected using spatial auto- and co-correlations. Spatial protein profiles were used for unbiased spatial clustering by affinity network fusion and hierarchical clustering. Results: Spatially-variable protein patterns can be detected at a range of resolutions, but the process is sensitive to the handling of missing values. Unbiased spatial clustering generates clusters of samples which co-cluster in space, reflecting the prominent pathology visible in H&E-stained sections along with further nuances such as immune-cell infiltration and oxygen/nutrient gradients, potentially boosting biomarker discovery. Conclusions: Spatially-resolved proteomics methods which spatially profile thousands of proteins will push the boundaries of understanding tissue biology and pathology at the molecular level. Recent mass spectrometry technology developments were key to achieving this throughput and depth, and this approach stands to benefit from further technological developments. Fig. 1 (optional) 389

390

P-I-0241 Cartographic analysis of the human cell surface proteome using proximity labeling proteomics Brendan Floyd1,2, Nicholas Till1,2, Elizabeth Schmidt1,2, Jonathan Yang1,2, Pinyu Liao1,2, Carolyn Bertozzi1,2,3 1Stanford University, Department of Chemistry, Stanford, CA, United States 2Stanford University, Sarafan ChEM-H, Stanford, CA, United States 3Stanford University, Howard Hughes Medical Institute, Stanford, CA, United States The cell surface is a highly heterogenous environment that harbours proteins critical for immune cell activation, cell-cell interactions, and cellular signaling among many other tasks. . Despite the many critical roles of proteins found on the cell surface, the importance of protein arrangement on the cell surface has only recently come to be recognized. Methods such as high-resolution fluorescence microscopy and proximity labeling enable studies of protein arrangement on the cell surface. These methods have currently not been used to systematically analyze the arrangement of the entire cell surface proteome. In this study a combination of low- and high- resolution proximity labeling methods along with an optimized DIA mass spectrometry workflow were used to systematically map the spatial distribution of the cell surface proteome on human T- and B-lymphoblast cell lines. Antibodies for over 80 surface proteins were tested and used to bind surface proteins of interest and collect a proximity labeling dataset that encompasses the entire surface proteome. Network analysis and a high accuracy machine learning classifier allowed for the determination of cell surface protein colocalization and the identification of functional clusters on the cell surface. These clusters include regions of "unconventional" cell surface proteins previously unstudied. Overall, this study illuminates the arrangement of the cell surface proteome which provides insight for informed cell surface engineering among other goals. 390

391

P-I-0242 Click and detect: a new generation of tagged probes for customed targeted multiplex MS imaging Alexandre Goossen1, Laurine Lagache1, Cédric Lion2, Christophe Biot2, Michel Salzet1, Isabelle Fournier1 1University of Lille, PRISM Inserm U1192, Villeneuve d'Ascq, France 2University of Lille, UGSF UMR CNRS 8576, Villeneuve d'Ascq, France By bringing a spatial dimension, MS Imaging (MSI) has shown to be a game changer for biology and clinics. However, despite steady progress in sensitivity many analytes remain invisible to MSI. This issue can be circumvented by the development of a targeted approach where a modified probe enabling MS detection is used in combination with conventional affinity-based imaging methods. Hence, the TagMass concept was previously demonstrated with oligonucleotide probes for imaging mRNA through in situ hybridization (ISH) and antibodies for imaging proteins via immunohistochemistry (IHC). Henceforth we have developed a new generation of tags that can be easily used to customize various probes and perform multiplex targeted MSI thanks to biorthogonal conjugation. This novel generation of tags is not only easily coupled to any probe thanks to the click-chemistry but is also designed to enable both photo- or chemically cleavage which can then be used in MALDI- or DESI-MSI. Hence the produced probes are composed of a linker, a photo- or chemically cleavable group, and a tag chosen to show good MS detection. Based on this principle, more than 10 novel tags were synthesized with different, but close in mass, chemical functions to get a similar analytical behavior and an accurate quantification level These tags were characterized by MS and NMR, and their cleavage yield was assessed alone and mixed. Then different tagged probes were produced from primary antibodies using bioconjugation followed by strain-promoted azide-alkyne click chemistry (SPAAC) reaction to perform targeted multiplex in MALDI- and DESI-MSI. First, the different tags were compared on the same antibody (anti-GFAP). Then different primary antibodies directed against various targets present in the rat brain (including for example GFAP as an astrocytic marker, NeuN for neurons, Myelin for oligodendrocytes, and tubulin beta-3 as a marker of stem cells and neural progenitors) were modified to perform multiplexing and cross-validated by Fluorescence IHC (FIHC). Some of the antibody probes were also functionalized with a chemically cleavable tag and tested in MALDI and DESI. Additionally, some tags were used to functionalize aptamers considering the various advantages of aptamers over antibodies in terms of specificity, binding affinity, and cost. Finally, a proof-of- concept maw made by customizing other probes directed against markers for ovarian and breast cancers such as PDL-1, CD47, TROP-2, Reg-Alpha, HER2, and markers of the immune cells. Fig. 1 (optional) 391

392

P-I-0243 Cell type resolved tissue proteomics to identify drivers of cancer therapy resistance in head and neck cancer Sonja Fritzsche1, Simon Schallenberg2, José Nimo1, Rafael Deliz-Aguirre1, Ingeborg Tinhofer-Keilholz3, Konrad Klinghammer4, Fabian Coscia1 1Max-Delbrueck-Center, Spatial Proteomics, Berlin, Germany 2Charité – University Medicine Berlin, Institute of Pathology, Berlin, Germany 3Charité – University Medicine Berlin, Institute of Radiooncology, Berlin, Germany 4Charité – University Medicine Berlin, Institute of Hematology & Oncology, Berlin, Germany Head and neck squamous cell carcinoma (HNSCC) incidence rates are rising globally, mainly due to the consumption of carcinogen-containing products and HPV infections. However, treatment success especially in HPV-negative patients is limited [1]. Patients with recurrent and metastatic disease are often treated with anti- cancer immunotherapy targeting the PD-1 pathway. However, most patients do not respond to the treatment with checkpoint inhibitors despite being selected based on PD-L1 positivity in immunohistochemistry. This suggests that patient selection cannot rely entirely on PD-L1 alone due to alternative checkpoints or pathways that affect the immune response. To systematically profile the tumor-immune microenvironments of responders and non-responders with single-cell and spatial resolution, we selected a patient cohort of 90 individuals and applied multiplex immunofluorescence imaging of 25 markers to capture cellular phenotypes and their functional states. We employed machine learning based image analysis, which identified more than 1 million cells, covering various tumor and immune cell phenotypes, including T cells, B cells, macrophages as well as cancer cells and their states (epithelial, mesenchymal, proliferative). We identified distinct patterns of cellular organizations and interactions that significantly varied between responders and non-responders. Our data revealed strong differences in immune cell abundances and their spatial distributions especially at the tumor invasive front. In the future, we will combine these promising data with the deep visual proteomics workflow employing single-cell laser microdissection followed by deep mass-spectrometry based low-input tissue proteomics. By pooling phenotypically similar cells, we will link cellular neighborhood information to functional proteome sates. With this integrated, multimodal approach, we can combine the advantages of targeted and untargeted spatial proteomics, enabling us to identify potential new drug targets, prognostic biomarkers of treatment response and infer mechanisms of treatment resistance. [1] Hashibe, Mia et al., Journal of the National Cancer Institute (2007) 392

393

P-I-0244 Advancing spatial multi-omics analysis in brain tissue by integrating MALDI-MSI with LCM-MS Byoung-Kyu Cho1, Jessica Lukowski1, Minsoo Son1, Antonia Zamacona Calderon1, Young Ah Goo1 1Washington University in St. Louis, St. Louis, MO, United States Background: Spatial omics technology has revolutionized biomolecular analysis by preserving molecular spatial context in tissues, offering applications in neuroscience, cancer biology, and drug discovery. We introduce an advanced approach, MALDI-MSI-guided LCM-MS, which combines mass spectrometry imaging (MSI) with laser capture microdissection (LCM). This approach allows for comprehensive profiling and quantitative analysis of the proteome, metabolome, and lipidome, offering new insights into complex biological systems and molecular distribution in a wide range of tissue types. Methods: We employed fresh frozen mouse brain tissues (n=3) for method development. MALDI-MSI used timsTOF fleX MALDI-2, processed with SCiLS, METASPACE, and SwissLipids. LCM-proteomics utilized timsTOF Pro2, processed with Mascot and Scaffold Quant. LCM-metabolomics and lipidomics used Orbitrap IDX Tribrid Mass Spectrometer, processed with Compound Discoverer. We tested the method on six mouse brain tissues from old and young groups (n=3 each), which underwent MALDI-MSI and LCM-proteomics. Results: MALDI-MSI identified 300+ putative metabolites and lipids in mouse brain, with [PC (38:2) +K]+ and [PC (32:0) +Na]+ showing high abundance in the hippocampus and cortex. LCM-omics of MALDI-guided regions (1 million cells) detected 300+ small molecules and 2800 proteins. When applied to different age groups of mice, our approach identified 3900+ proteins in LCM-proteomics. Among these, 778 and 161 proteins exhibited differential expression in the cortex and hippocampus, respectively, between the two age groups. Biological network analysis revealed the involvement of differentially expressed proteins in aging-related pathways, such as mitochondrial dysfunction, glutamatergic synapse, and AMPK signaling. Conclusions: Our integrated approach of MALDI-MSI and LCM-MS in spatial omics provides a comprehensive and quantitative understanding of complex biological systems. This method holds great potential for advancing neuroscience research and gaining deeper insights into molecular mechanisms underlying age-related changes in the brain. 393

394

P-I-0245 Filling in missing pieces in the proteomic picture of amyloidosis by MALDI TIMS imaging supervised segmentation analysis and label-free intensity profiling Juliane Gottwald1, Janina Oetjen2, Romano Hebeler2, Corinna Henkel2, Christian Treitz3, Hans-Michael Behrens1, Daniel Becker1, Natalie Berghaus4, Sarah Schreiner4, Julian Baur5, Stefanie Huhn6, Ute Hegenbart4, Stefan O. Schönland4, Christian Haupt5, Andreas Tholey3, Christoph Röcken1 1Christian-Albrechts-University Kiel, Department of Pathology, Kiel, Germany 2Bruker Daltonics GmbH & Co. KG, Bremen, Germany 3Christian-Albrechts-University Kiel, Systematic Proteome Research and Bioanalytics, Institute of Experimental Medicine, Kiel, Germany 4Heidelberg University Hospital, Medical Department V, Amyloidosis Center, Heidelberg, Germany 5Ulm University, Institute of Protein Biochemistry, Ulm, Germany 6Heidelberg University Hospital, Medical Department V, Section of Multiple Myeloma, Heidelberg, Germany In recent years, a group of diseases has become the topic of advanced training for clinicians, i.e., amyloidoses. These are caused by misfolding of proteins into amyloid fibrils that deposit at different tissue- and organ-sites causing various diseases. Once considered a rare disease, it has become almost endemic. We have not yet fully understood the pathomechanisms responsible for the different clinical manifestations of amyloidoses. In search for key players in disease progression, recent research uncovered that the fibril conformation is likely not the sole reason for differential clinical manifestations. Hence, the analysis of the overall amyloid proteome and specifically co-deposition of amyloid-associated compounds may shed light upon the underlying disease mechanisms. Understanding these mechanisms will provide earlier diagnosis and advanced therapy for patients suffering from an often lately detected, and potentially fatal disease. For our studies, we selected formalin-fixed and paraffin-embedded (FFPE) heart explants diagnosed with the most common cardiac amyloid types, i.e., immunoglobulin light chain- (AL) and transthyretin-derived amyloid (ATTR). We coupled two methods to liquid chromatography tandem mass spectrometry (LC-MS/MS). At first, Congo red staining visualized the distribution of amyloid histologically and guided tissue sampling with varying amyloid load for the generation of tissue microarrays (TMAs). Selected samples of the TMAs were then processed for bottom-up LC-MS/MS followed by label-free intensity (LFI) profiling. This method correlates protein abundance to a target LFI profile such as the amyloid load, amyloid-associated or amyloidogenic proteins. One case each of AL and ATTR amyloidosis with comprehensive LFI profiling, were further investigated by mounting FFPE tissue sections directly on Bruker IntelliSlides. Tryptic digestion was followed by matrix- assisted laser desorption/ionization trapped ion mobility spectrometry (MALDI TIMS) imaging (timsTOF fleX, Bruker). After data acquisition, a list of peptides previously related to proteins associated with amyloid supervised a segmentation analysis of the imaging data and identified two clusters in each amyloidosis case (Fig.). Laser microdissection (LMD) sampled these selected clusters, which were then forwarded to LC-MS/MS for a more in-depth proteomic composition analysis. Both workflows were able to allocate tissue regions with variable amounts of amyloid and differences in the abundance of amyloidogenic- and amyloid-associated proteins. This enables us to discuss potentially disease- related mechanisms and amyloid-related compounds, and to compare AL- to ATTR deposition. In summary, both methods have proven their potential to characterize the cardiac amyloid proteome. Hence, these workflows will produce a comprehensive knowledge gain with each sample analyzed, which in turn will ideally help to reduce numbers of lately diagnosed amyloidosis in the near future. 394

396

P-I-0246 Glycocapture-based cell surface proteome analysis of human epidermal melanocytes and melanoma cells Lili Fruzsina Balogh1, Minh Ngoc Nguyen1, Emese Zsigrai1,2, Patrik Kovács1, David J Boocock3, Csaba Matta1, Tibor Hajdú1 1University of Debrecen, Department of Anatomy, Histology and Embryology, Debrecen, Hungary 2University of Debrecen, Department of Pediatrics, Debrecen, Hungary 3Nottingham Trent University, John van Geest Cancer Research Centre, Nottingham, United Kingdom Cell surface proteins collectively constitute the surfaceome, a special subset of the plasma membrane proteome. The surfaceome plays crucial roles in determining cellular phenotype and identity, and its qualitative and quantitative composition undergoes significant alterations during healthy and pathological differentiation. Currently, there are no data available concerning the complement of cell surface proteins of healthy and pathological human pigment cells. Therefore, this project is aimed at discovering the epidermal melanocyte and cutaneous melanoma-specific surfaceome. Three human pigment cell cultures were used in our experiments: primary epidermal melanocytes (from skin samples), WM35 cell line (from in situ melanoma), and A2058 cell line (from metastatic melanoma). Cell surface proteins were selectively labelled by aminooxy-biotin and then isolated by streptavidin-conjugated beads. Isolated proteins were trypsin-digested, and peptides were analysed by high-throughput shotgun mass spectrometry. Identified proteins were then evaluated by bioinformatics tools. For validation, we performed RT- qPCRs, western blots and immunocytochemistry reactions. The enrichment of the surfaceome was 70,1%, 63,1% and 56,9% in melanocytes, WM35 and A2058 cells, respectively. Based on GO annotations, the proteins constituting the surfaceome were further classified into functional subcategories (transporters, enzymes, receptors, adhesion, miscellaneous proteins). The combined list consisted of 470 different proteins, 112 of which were detected only in melanocytes, whereas 186 proteins were specific to the melanoma cells and 95 proteins were identified in all three cell types. Based on the proteomic analysis we found several proteins worth of further examination. Matrix metalloproteinase inducer protein, basigin (CD147) was one of them and based on mass spectrometry results it was only present in metastatic melanoma cells. However, qPCRs demonstrated basigin mRNA expression in all examined cell cultures, while western blots proved basigin protein expression only in melanoma cells. Surprisingly, immunocytochemistry reactions followed by confocal microscopic evaluation revealed plasma membrane associated basigin expression in all three cell cultures, although intensity of fluorescent signals seemed to increase with malignancy. Cell surface proteins are main targets of biomedical research due to their potential to become possible biomarkers and therapeutic alternatives. The differences found in the surfaceome of healthy vs malignant pigment cells can lead to the discovery of highly selective biomarkers of disease onset and progression. The protein expression analysis verified our proteomic findings as basigin may become a potent biomarker in melanoma diagnostics. This work was supported by the Young Researcher Excellence Programme (grant number: FK-134304) of the National Research, Development and Innovation Office, Hungary. 396

397

P-I-0247 A high-resolution spatio-temporal atlas of the testicular germ cell proteome Feria Hikmet Noraddin1, Loren Méar1,2, Jonas Gustavsson1, Borbala Katona1, Rutger Schutten1, Kalle von Feilitzen3, Mattias Forsberg3, Mathias Uhlen3,4, Cecilia Lindskog1 1Uppsala University, Dept. of Immunology, Genetics and Patholgy, Uppsala, Sweden 2Karolinska Institutet and Karolinska University Hospital, Division of Obstetrics and Gynecology, Department of Clinical Science, Intervention and Technology, Solna, Sweden 3KTH Royal Institute of Technology, Science for Life Laboratory, School of Engineering Sciences in Chemistry, Biotechnology and Health, Solna, Sweden 4Karolinska Institute, Department of Neuroscience, Stockholm, Sweden fluorescent-based multiplex immunohistochemistry pipeline, we performed an Spermatogenesis is a tightly regulated process to produce male gametes. It involves the sequential and coordinated activation and repression of thousands of genes and proteins. Many of these genes and proteins are testis-specific and are required across multiple cell states, for the successful development of spermatogonial stem cells into mature sperm cells. Understanding the intricate functions and mechanisms at each step of this process requires a multi-dimensional approach that integrates quantitative and qualitative methods. We used single-cell RNA sequencing data, spatial protein profiling, and automated image analysis to generate a high- resolution spatio-temporal map of the human testis. Through transcriptomics analysis, we identified 12 distinct states of germ cells corresponding to subsets of spermatogonia, spermatocytes, and spermatids. By utilizing a large-scale in-depth characterization of approximately 500 proteins expressed in different germ cells, mapping their exact spatial localization during spermatogenesis. The generated data allowed us to cluster the proteins into expression groups based on common patterns of cell state-specific expression, forming the basis for downstream functional analysis. In total, we identified 18 different protein clusters and demonstrated that the proteins in these groups were linked to germ cell-specific functions such as cell differentiation or cell cycle regulation at different phases of spermatogenesis. Mapping mRNA and protein levels in the germ cell states also allowed studying the correlation between mRNA and protein along spermatogenesis. During spermatogenesis, two distinct phases of inactive transcription have been identified and are known to i) be associated with early meiosis and ii) mRNA production arrest during the cellular transformation of haploid round cells into elongating spermatids. We identified several cases of high positive correlation between mRNA and protein expression in germ cell states, but interestingly, for more than one-third of the investigated proteins, the correlation coefficients were very low. Proteins exhibiting expression at a later cell state compared to their mRNA expression provided new insight into the complex spatio-temporal nature of spermatogenic function and mRNA translation dynamics. In summary, we present a spatio-temporal single-cell type reference map of adult human spermatogenesis. Our automated staining-to-image analysis pipeline reliably extracts quantitative measurements from state-specific germ cells in human testis tissue and can be used to characterize poorly defined proteins not previously described in the context of sperm development. This data also has the potential to contribute to valuable insights into the molecular functions and pathways of proteins previously linked to infertility and disorders of the male gonads. 397

398

P-I-0248 Comparative spatial proteomics as a tool to understand difficult-to-express molecules in CHO cell lines Alistair Hines1, Owen Vennard2, Nicholas Bond1, Claire Harris1, Lisa Breckels2, Kathryn Lilley2, Diane Hatton1 1AstraZeneca, Cambridge, United Kingdom 2University of Cambridge, Biochemistry/CCP, Cambridge, United Kingdom Chinese hamster ovary (CHO) cell line expression systems dominate the biotechnology industry for expressing therapeutic proteins but can still be plagued with unpredictable low production yields, which make for substandard manufacturability of the drug product. Proteins localise to discrete, subcellular locations within the cell to fulfil their location-specific functions and enable biological processes to occur. Movement between these niches (e.g., organelles) is essential for cellular processes. Modifications, mutations, and other variations that cause deviant protein localisation have been associated with a wide range of diseases, as well as impacting on protein expression. Determining the location of proteins in different cell types and states is essential to understanding the functions and processes involved with their differential localisation. Here, we have applied the subcellular spatial proteomic method, hyper-plexed localisation of organelle proteins by isotope tagging (hyperLOPIT), to determine differential localisations of proteins between a CHO cell line expressing an monoclonal antibody that is easy to express (ETE) and a cell line expressing a difficult to express protein (DTE). Nearly 5,000 proteins were mapped to 13 organellar compartments with excellent resolution using the hyperLOPIT workflow, serving as a significant resource for anyone utilising CHO cells as an expression platform. Little is know about the subcellular organisation of proteins specifically within CHO cells, or how this changes with the expression of therapeutic proteins. Therefore a bespoke approach, Bayesian ANalysis of Differential Localisation Experiments (BANDLE), was applied to compute probability of differential localisation of proteins between the the cells that have relative apparent ease or difficulty in expressing therapeutic proteins. Results from this extensive analysis identified many proteins that are differentially localised between cell lines expressing DTE and ETE molecules with an implication in various intracellular processes – particularly the upregulation of proteins involved in endocytosis, autophagy, and phagosome formation in the cell line expressing the DTE protein. This work provides novel insight into potential gene targets to enhance the manufacturing capability of biologics in CHO cell lines. 398

399

P-I-0249 MALDI image-based molecular classification of difficult-to-diagnose rare gastrointestinal tumors in dialog with pathological and clinical aspects Pia Hönscheid1,2,3, Linna Sommer1,2, Frieder Meier1, Christian Sperling1, Daniela Aust1,2,3, Patrick Jensen4,5, Jan Lellmann5, Therese Seidlitz6, Daniel Stange2,3,6, Herbert Thiele7, Gustavo Baretton1,2,3 1Institute of Pathology, Medical Faculty and University Hospital Carl Gustav Carus, Technical University Dresden, Dresden, Germany 2National Center for Tumor Diseases Dresden, Dresden, Germany 3German Cancer Consortium, Dresden, Germany 4Department of Applied Mathematics and Computer Science, Lyngby, Denmark 5Institute of Mathematics and Image Computing, Universität zu Lübeck, Luebeck, Germany 6Department of Visceral, Thoracic and Vascular Surgery, Medical Faculty and University Hospital Carl Gustav Carus, Technical University Dresden, Dresden, Germany 7Fraunhofer Institute for Digital Medicine MEVIS, Luebeck, Germany By using analytical methods such as MALDI imaging on formalin-fixed tumor tissues, we can examine cancer proteomes sensitively and rapidly from a diagnostic point of view. To reflect the heterogeneity, we analyzed a collection of gastrointestinal cancer samples, including rare gastric and ampullary cancers, according to WHO classification and MALDI imaging TOF classification algorithms. Redetermination of the cancer tissues by mass spectra demonstrated intra-tumoral heterogeneity as well as relatedness to tumor stroma and overlapping mass spectra of tumor subtypes - independent of patient. Interestingly, greater parallels were found with the detailed histo-morphological tumor classification than with other clinical parameters at the time of surgery. This also shows the good usability of this tool as a substitute for classical histological analyses with the addition of the representation of paracrine (cellular) and tissue group-related correlations. Tumor detection is successfully demonstrated on the tumor sample itself, metastasis or artificially generated samples with little surrounding material such as organoids. For example, out of this analysis mass peaks 1441, 1775 and 2886 m/z were exclusively related to gastric tumor areas of the stomach. Transferability and comparability from one sample to another was calculated by leave-one-out classification. Mass spectra of unknown tissue matched the reference tumor spectra yielded positive predictive values ranging from 52% -77%, and showed the highest migration from tumor specific pixels to tumor-altered material such as stroma or inflammation. Gastric cancer organoids with different histological subtypes had divergent molecular signatures to that extent, that could be traced back to the signatures of the original tumor out of the gastric intestinal cancer collection. Overall, the classification of organoids and donor tissues only was 54% successful on average and need to be improved by sample size and training method before relying on the representative restriction of organoids. In addition, subclasses that are particularly difficult to diagnose, such as cancer mixed types, showed relationships to subclasses of the pre- trained MALDI imaging classes similar to the pathological assessment. Using ampullary carcinoma as an example, we were able to identify all subgroups and reveal new features of those mixed ones. In summary, we can demonstrate the potential of a proteome-based classification system for the development of grading of gastrointestinal carcinomas. In addition to histological detection, MALDI TOF analysis reveals many other aspects of how tumor tissue can be characterized. Rare tumor types in particular can benefit from this robust and fast detection of similarities or differences to existing tumor subtypes. The use of AI analysis will bring further significant developments in this area. 399

400

P-I-0250 Intratumor heterogeneity causes spatial proteome variation in NSCLC patients Denise Jansen1,2,3,4, Anuar Makhmut5, Janett König5, Lisa Strotmann6,2,4, Martin B. Schneider1, Dario Frey2,4, Barbara Helm2,4, Fabian Coscia5, Marc Schneider6,4, Yifat Merbl3, Ursula Klingmüller2,4,7, Dominic Helm1 1German Cancer Research Center (DKFZ), Proteomics Core Facility, Heidelberg, Germany 2German Cancer Research Center (DKFZ), Division Systems Biology of Signal Transduction, Heidelberg, Germany 3Weizmann Institute of Science (WIS), Department of Immunology, Rehovot, Israel 4German Center for Lung Research (DZL), Translational Lung Research Center Heidelberg (TLRC), Heidelberg, Germany 5Max-Delbrück-Center for Molecular Medicine in the Helmholtz Association, Spatial Proteomics Group, Berlin, Germany 6Thoraxklinik at University Hospital Heidelberg, Translational Research Unit, Heidelberg, Germany 7Liver Systems Medicine against Cancer (LiSyM-Krebs), Heidelberg, Germany Non-small cell lung cancer (NSCLC) accounts for around 85% of all lung cancer instances, with patients encountering poor prognoses. It is subclassified as squamous (SQ) and non-SQ cell carcinoma, the latter of which splits again into the subtypes of large cell carcinoma and adenocarcinoma. NSCLC subtypes do not show particularly distinct genetic profiles and have mostly been identified by their morphological properties. NSCLC evolves in an environment with low selection pressure which maximizes the existence of more diverse cell clones, eventually leading to intratumor heterogeneity (ITH). ITH causes therapeutic resistance and unfavorable treatment outcomes in patients. Mass spectrometry (MS)-based proteomic analysis has become a key technology in systems biology and thus in cancer research. MS-based proteomics focussed, for many years, on analyzing "bulk" material, e.g. patient tumor samples or whole tissue sections. This also applies to clinical tumor samples, which usually contain a mixture of tumor and tumor-associated cells, resulting in an average proteome representation of the tumor. Hence, ITH blurs the picture of tumor cell diversity, which can only be addressed by low-input/single-cell approaches. The recently developed Deep Visual Proteomics (DVP) approach (Mund and Coscia et al. 2022) is an innovative technology that combines high-end microscopy with state-of-the-art high- sensitivity MS-based proteomics. It allows precise isolation of small areas /single cells from formalin-fixed and paraffin-embedded (FFPE) tissue samples via laser microdissection (LMD), based on their phenotypic appearance while maintaining spatial information. Subsequent liquid-chromatography (LC)-MS analysis will provide a clearer picture of the underlying proteome of phenotypically different tumor regions. This study serves as proof of relevance for the utilization of the DVP method in lung cancer studies. We compare the proteome of phenotypically different tumor and stroma regions to identify proteomic variances and uncover key aspects of spatial cellular regulation within the diverse tumor landscape of NSCLC subtypes. 400

401

P-I-0251 Mapping the functional protein nanoscale organization in T Cell bispecific antibody-induced immunological synapses using LUX-MS Arend Keller1, Antonio Sorrentino2, Claire Xu2, Marina Bacac2, Bernd Wollscheid1 1ETH Zürich, Department of Health Sciences and Technology (D-HEST), Zurich, Switzerland 2Roche Pharmaceutical Research & Early Development, Zurich, Switzerland T cell bispecific antibodies (TCBs) are potent cancer immunotherapies that redirect T cells to attack tumors. Recent studies reveal that TCBs can induce the formation of immunological synapses between T cells and cancer cells, mimicking physiological immune synapses. However, these TCB-induced synapses exhibit distinct spatial and temporal patterns, suggesting unique receptor-ligand interactions govern their formation. To identify new target cell surface proteins sustaining TCB-induced synapses, we employed the optoproteomic LUX-MS technology. LUX-MS enables the discovery of proximal protein interaction networks within (cis) and between (trans) living cells using light-controlled singlet oxygen generators (SOG). We established a co-culture model of naive T cells and OCI-LY18 lymphoma B cells, treated with either Glofitamab (CD20-TCB) or Gazyva (CD20- Antibody) as a negative control. After confirming the functionality of SOG-conjugated Glofitamab, we identified 1 hour as an optimal time point for synapse formation analysis through FACS-based time-course experiments. Our spatial proteotyping analysis revealed cell surface protein target candidates involved in the acute organization of the TCB-induced synapse between B lymphoma cells and primary T cells. Integrative network analysis showed that most identified proteins belonged to integrin-mediated signaling networks, heterotypic cell-cell adhesion networks, and extracellular structure organization pathways. Validation of the functional roles of these targets in sustaining TCB-mediated effects using CRISPR-Cas9 depleted knock-out cells is currently ongoing. Ultimately, our spatial proteotyping strategy enables time-resolved analysis of the TCB-induced surfaceome synapse at the molecular nanoscale level. The generated data provides a rational path forward to target validated surfaceome hits synergizing with TCB efficacy and decrease the threshold of TCB-induced T cell activation for therapeutic application. 401

402

P-I-0252 Spatial proteomics reveals the dynamic phenotypic landscape of vascular smooth muscle cells (VSMCs) in atherosclerosis Elena Kratz1, Ankit Sinha1, Trusha Adeshara2, Nadja Sachs2, Jessica Pauli2, Lars Maegdefessel2, Matthias Mann1 1Max-Planck-Insitute of Biochemistry, Proteomics and Signal Transduction, Martinsried, Germany 2University Hospital rechts der Isar, Technical University of Munich, Institute of Molecular Vascular Medicine, Munich, Germany Atherosclerosis, the primary cause of cardiovascular diseases, is marked by plaque accumulation in arteries, which consist of fatty and fibrous material. Identifying high-risk plaques and effective therapies remains difficult due to limited understanding of underlying mechanisms. Vascular smooth muscle cells (VSMCs) are critical in atherosclerosis progression and plaque stability, exhibiting remarkable plasticity with either atheroprotective or atherogenic roles. To explore the mechanisms behind these roles in plaque formation, we employed in vitro and spatial in situ proteomics. We initially treated primary carotid SMCs with various stimuli, inducing diverse phenotypes including contractile, synthetic, osteogenic, macrophage-like and inflammatory states. Subsequently, we applied an integrative proteomics approach, encompassing cellular, secreted and phospho proteomics, to generate an extensive quantitative profile of VSMC phenotypes in vitro. This quantitatively profiled more than 9,000 proteins, including regulatory members of TGFβ, JAK-STAT, NFκB and TNFα signaling pathways. Our findings reveal distinct shifts in protein expression patterns corresponding to VSMC phenotypes, revealing links between plasticity, key signaling pathways and extracellular matrix (ECM) remodeling. To extend our findings to clinically relevant samples, we analyzed formalin-fixed paraffin-embedded (FFPE) human carotid plaques by Deep Visual Proteomics (DVP). Our approach integrates immunofluorescence imaging, automated cellular segmentation and spatial clustering with laser microdissection and ultra-high sensitive quantitative mass spectrometry. It allowed us to individually dissect and in situ analyze the spatial distribution of VSMC phenotypes in plaque microtomes. For each in vivo VSMC population, we achieved a median protein depth of 4,000 protein groups, and hallmark VSMC proteins such as smooth muscle actin (ACTA2) and transgelin (TAGLN) were highly abundant, confirming proteomic specificity. Principal component analysis (PCA) revealed that dissected cellular populations cluster according to their spatial location in the plaque. Remarkably, unsupervised clustering identified two clusters, one associated with upregulation of smooth muscle contraction and ECM organization, and the other with inflammatory and lipid processes. This opens up for direct study of various VSMC phenotypes, such as contractile and inflammatory states in situ. Finally, we used phenotype- specific signatures of the in vitro experiments to assign proteomes to the in situ populations. In summary, our study leverages advanced proteomics to enhance understanding of VSMC biology in atherosclerosis, identifying novel proteins associated with VSMC phenotypic changes, potentially serving as therapeutic targets. 402

404

P-I-0253 Heterogeneity assessment and protein pathway prediction via spatial Lipidomic and proteomic correlation: advancing dry proteomics concept for human Glioblastoma prognosis Laurine Lagache1, Yanis Zirem1, Emilie Le Rhun1, Isabelle Fournier1, Michel Salzet1 1PRISM - Inserm U1192, Villeneuve d'Ascq, France Prediction of protein pathways from lipid analyses via MALDI MSI is a pressing challenge. We introduced "dry proteomics," using MALDI MSI to validate spatial localization of identified optimal clusters in lipid or protein imaging. Consistent cluster appearance across omics images suggests association with specific lipid and protein pathways, forming the basis of dry proteomics. The methodology was refined using rat brain cerebellum tissue as a model, then applied to human glioblastoma, a highly heterogeneous cancer. Sequential tissue sections underwent omics MALDI MSI and unsupervised clustering. A MatLab script was developed to process omics image datasets, with SVD data compression, providing unsupervised clustering, independently for each data set. The integration of the Silhouette criterion prediction allowed to determine the most optimized segmentation number of cluster. Differentiated lipid and protein clusters, with distinct spatial locations, were identified. Spatial omics analysis facilitated lipid and protein characterization, leading to a predictive model identifying clusters in any tissue based on unique lipid signatures and predicting associated protein pathways. Application to rat brain horizontal slices revealed diverse tissue subpopulations, including successfully predicted cerebellum areas with distinct lipid and protein pathways. Dry proteomic workflow was applied on a prospective and retrospective cohort of 50 glioblastoma patients tissues, re-using MALDI imaging and proteomic data from previous study. Resulting analysis confirmed lipid- protein associations on intricate and heterogeneous pathology samples. By correlating data obtained from lipid MALDI MSI cluster identification with patient stratification and prognosis data, we aimed to establish a robust link between the lipid and protein signatures identified through the optimized method and the clinical outcomes of glioblastoma patients. Thus, this study has culminated in the development of a sophisticated prognostic prediction model for glioblastoma patients, amalgamating lipid imaging clusters with corresponding proteomic data to stratify distinct prognostic categories. This crucial validation step not only enhances confidence in the reliability of this approach but also holds significant promise for advancing personalized medicine strategies in the management of this challenging disease. 404

405

P-I-0254 Deciphering pulmonary tumor microenvironment through MALDI imaging mass spectrometry: new approaches for tumor expansion inhibition Elena Guerra-Paes1, Cristina María López-Vázquez2, José Manuel Muñoz-Félix1, Ángela Peralbo-Molina2, Eduardo Chicano-Gálvez2, Fernando Sánchez-Juanes1 1Salamanca Biomedical Research Institute (IBSAL), Department of Biochemistry and Molecular Biology, University of Salamanca, Salamanca, Spain 2Maimonides Biomedical Research Institute of Cordoba (IMIBIC), IMIBIC Mass Spectrometry and Molecular Imaging Unit (IMSMI), Córdoba, Spain During tumor growth, cancer cells increase their oxygen and nutrients demand necessitating the formation of new blood vessels to supply these fuels. Tumor angiogenesis, the formation of new blood vessels from pre- existing ones, is considered the major mechanism of tumor vascularization. Over more than 40 years, numerous molecules have been studied in preclinical and clinical trials to inhibit tumor angiogenesis such as vascular endothelial growth factor (VEGF) antibodies. These anti-angiogenic strategies have shown numerous limitations in the clinical practice. One of these limitations is that cancer cells can use non-angiogenic mechanisms to obtain oxygen and nutrients. Vessel co-option (VCO) is a mechanism in which cancer cells hijack pre-existing blood vessels instead of developing new blood vessels through the angiogenesis process. Overcoming VCO as a resistance mechanism to anti angiogenic therapies is a new challenge in cancer research. Considering that pre-existing blood vessels are quiescent, the hypothesis is that the inhibition of vessel quiescence can be a good strategy for inhibiting VCO. The main goal of this project is to test if the inhibition of vessel quiescence induced by bone morphogenetic protein 9 (BMP9) through ALK1 receptor, promotes VCO inhibition and it is a good anti-tumor strategy to use in combination with chemotherapies, immunotherapies or anti angiogenic drugs. Moreover, potential biomarkers of non-angiogenic vs angiogenic growth are aimed to be identified using MALDI-Imaging (MSI). For this purpose, lung metastastes have been generated by injecting 4T1 cells (breast cancer) intravenously. Once pulmonary metastases are established, animals were treated with either placebo or PF-03446962 (an antibody against ALK1 receptor) three times during two weeks. At the endpoint of the in vivo procedure, lungs were harvested, stained with haematoxylin-eosin and analyzed by MALDI Imaging. Data processing and analysis with SCiLS™ software allowed the isolation of the tumor region of interest as well as the region close to the blood vessels. Preliminary results revealed differences between the spectra of the selected regions of interest (ROI) and allowed the characterization of a specific "biomolecular" signatures of them. Then, general tissue microextractions were done to identify some peptides present in the tissue surface by LCMS and assign these identifications to specific features present in the detected biomolecular signatures. Some of these peptides were found to be related to platelet activation, blood coagulation, the formation of fibrin clots and angiogenesis giving us a new perspective on the associated mechanisms of the VCO used by the tumors. 405

406

P-I-0255 High throughput proteomic profiling of tumor immunophenotypes by integrating microPOTS sample preparation, laser dissection, and implanted microtumor array Amanda Lorentzian1, Catherine Carbone2, Meena Choi1, Ira Mellman2, Christine Moussion2, Ying Zhu1 1Genentech, Inc., PGT, South San Francisco, CA, United States 2Genentech, Inc., CI, South San Francisco, CA, United States Tumours contain multiple cell types with distinct spatial organization, which forms heterogeneous microenvironments and immunophenotypes. It is increasingly recognized that the tumor immunophenotypes play key roles in tumor progression and therapeutic outcome, butit is still unclear what the main determining factors for these phenotypes are. A recently-established Skin Tumour Array by MicroPoration (STAMP) is an approach to generate hundreds of tumors in vivo to study immunophenotypes. Within a single mouse ear slice, there are typically 10-15 tumors originating from the same cell line that exhibit distinct levels of immune infiltration. Recent advancements in mass spectrometry-based proteomics have revolutionized the analysis of complex biological samples. Spatial proteomics has emerged as a promising tool for investigating cellular heterogeneity, tumor microenvironments, and subcellular protein localization, thus enhancing our understanding of disease mechanisms. Herein, we applied spatial proteomics to study diverse immunophenotypes by profiling of individual microtumours in STAMP. For high-throughput profiling, we employed laser capture microdissection to isolate individual microtumours from mouse skin tissue, followed by microPOTS sample preparation and LC-MS analysis. RNAscope analysis was first performed using T-cell and dendritic cell (DC) markers to characterize the inflammation of immune cells within the tumour. Three main phenotypes are established; EE (T-cell and DC excluded), II (T-cell and DC inflamed) and IE (T-cell inflamed and DC excluded). Tumor slices approximately 1 mm in diameter and 10 µm in thickness were collected from each phenotype using laser capture microdissection on a Zeiss LCM microscope and directly collected into a microPOTS chip. Sample preparation was performed in a volume of ~2 µL to improve sample recovery. The entire sample was injected for LC-MS/MS analysis using a 2-hr LC-gradient and data-dependent acquisition. FragPipe with IonQuant was used for database search and label- free quantification. An average of 6900 proteins were identified per sample with >90% of proteins robustly quantified in all samples, providing deep proteome coverage from ~ 1-mm-diamter tumour slice. We observed distinct enrichment of proteins and pathways among these phenotypes, including the identification of T-cell and DC surface markers such as CD3 and CD4, which were enriched in the II tumours. Additionally II tumours were enriched for immunological-related processes, as expected. While phenotype-specific patterns were evident, an underlying biological variation driven by mouse background was also noted. However, this also highlights the sensitivity of our approach. Together, this study demonstrated spatial proteomics coupled with STAMP as a high-throughput and high-efficiency platform for large-scale tumor phenotyping and mechanistic study. 406

407

P-I-0256 Spatial proteomics of ovarian cancer precursor lesions Anuar Makhmut1,2, Mihnea-Paul Dragomir2, Eliane Taube2, Janett König1, Fabian Coscia1 1Max Delbrück Center for Molecular Medicine, Berlin, Germany 2Charité – Universitätsmedizin Berlin, Berlin, Germany High-grade serous ovarian cancer (HGSOC) is the most prevailing and lethal subtype of epithelial ovarian cancer (EOC). Despite significant research efforts, the prognosis has remained unchanged for this disease. To a significant extent, this is due to a limited understanding of the early stages of this disease. Accumulating evidence suggests that serous tubal intraepithelial carcinoma (STIC) of the fallopian tube is the main precursor lesion giving rise to HGSOC. However, not much is known about the molecular pathways that drive the transition from fallopian tube epithelium (FTE), to STIC, and then to invasive fallopian tube carcinoma (invFTC). Genomic studies have provided important knowledge on driver mutations and the genetic, clonal relationship between STIC and invFTC. However, genomics alone does not explain the functional implications of the genetic alterations. Since proteins are a closer representation of cellular state and function, proteome profiling of STICs can provide insights into how the proteome changes during the early stages of cancer. The proteomic analysis of STICs has remained challenging because they are microscopic, consisting of only a few cells, compared to established tumors. Due to a lack of workflows tailored towards proteomics of ultra-low sample input from formalin-fixed paraffin-embedded (FFPE) tissues, proteomics has not been performed yet to systematically profile STICs. To better understand the molecular processes that drive the transition from fallopian tube epithelium (FTE), to STIC, and then to HGSOC, we for the first time performed mass spectrometry-based spatial proteomics on FFPE tissue sections of a cohort of 36 patients. We characterized proteomes to a depth of 7,000 proteins from STICs as small as 50 cells. Preliminary data suggests pronounced proteome changes between FTE and STICs, but a high level of similarity between STICs and invFTC. Clustering of precursor lesions and invFTC revealed phenotypically divergent tumors with distinct enriched pathways that might contribute to the clinical outcome. Conversely, we found pronounced differences in the stroma sampled from regions directly adjacent to FTE, STICs and invFTC, suggesting a critical role of the tumor microenvironment in promoting the full transformation of pre-cancerous STICs to invasive carcinoma. Our data provides new insights into the molecular alterations that define the early stages of HGSOC development. In the future, we plan to integrate our data with genomics, and functional studies. This could ultimately help to develop strategies to intervene early and prevent HGSOC in the long term, an approach crucial in improving the disease outcome. Fig. 1 (optional) 407

408

P-I-0257 Systematic evaluation of the performance metrics for spatial proteomics workflow and its application on colorectal cancer samples Giada Marino1, Thomas Gaitanos1, Sonja Neumaier1, Andreas Tebbe1, Nagarjuna Nagaraj1 1Evotec GmbH, Mass Spectrometry, Neuried, Germany Recent advances in mass spectrometry instrumentation have led to a great increase in sensitivity, thereby facilitating the analysis of single cells and other low input material sample types. Laser capture microdissection (LCM), where spatially-defined regions of interests (ROIs), such as individual or populations of cells, are extracted from a biospecimen under the microscope, generates highly-defined samples of low input material. Visual proteomics, which combines high sensitivity proteomics technology and high precision LCM sampling has been shown by pioneers in the field to yield differences at a spatial resolution within a single tissue slice; however, it remains a niche technique and is yet to be widely adapted and applied in an industry setting. Here, we investigate the basic performance metrics of visual proteomics, including sensitivity and reproducibility of the pipeline. First, we optimized the gradient and acquisition settings of the LC-MS/MS setup using a serial dilution of commercially available HeLa digest. Samples were measured in data independent acquisition (DIA) mode using the Bruker timsTOF Ultra mass spectrometer. We identified on average more than 5,000 protein groups from 250 pg of standard HeLa digest using a library free approach in Spectronaut. Next, we assessed the robustness, scalability, and minimal sample limit of a typical LCM workflow. Homogenous areas of murine liver FFPE tissue ranging from 150,000 µm2 to 1,250 µm2 were laser capture dissected using a Leica LMD7 microscope. We identified on average approximately 1,500 protein groups with a CV of 20%. As proof of concept, we applied our optimized workflow to analyze proteomic differences in colorectal cancer tumor tissue and near adjacent to tumor areas from clinical samples. Finally, we discuss the common practical issues typically associated with such cutting-edge technology. We highlight issues valuable to the community, from sample preparation to data acquisition and analysis. For example, classical bottom-up proteomics commonly involves system performance checks using HeLa or K562 tryptic peptide digests, whereas an equivalent control standard for visual proteomics currently does not exist. Due to the differences in scale and composition between classical MS and LCM-derived samples, a customized performance test for proper quality management in an industrial environment is a must. Scalability requirement provides another challenge for industry. The translation of systematic image analysis through laser capture to LC-MS/MS is a major bottle neck that needs to be overcome to reach the throughput required for a commercial setting. Ultimately, it is important to set these standards correctly as early as possible to ensure that the data produced matches the rapid technological advancements currently seen in this emerging field. 408

409

P-I-0258 The human Ovarian cortex: unraveling spatial dynamics across the lifespan and treatment exposures Loren Méar1,2,3, Filippa Bertilsson1, Ilmatar Rooda2,3, Pauliina Damdimopoulou2,3, Cecilia Lindskog1 1Uppsala University, Immunology, Genetics and Pathology, Uppsala, Sweden 2Karolinska Institute, Clinical Science, Intervention and Technology, Stockholm, Sweden 3Karolinska University Hospital, Gynaecology and Reproductive Medicine, Stockholm, Sweden The human ovary, pivotal to the endocrine and reproductive systems, undergoes dynamic structural changes throughout life, but the molecular mechanisms of puberty and menopause from a spatiotemporal aspect remain poorly characterized. This knowledge gap hinders advancements of infertility treatments, fertility preservation, and personalized medicine, which is of particular importance for childhood cancer survivors. Given that current options are limited to cryopreservation of ovarian tissue—a method developed for adult ovaries—a deeper understanding of ovarian cellular dynamics and cell type interactions across all ages is imperative. To date, there are no reports of live births following the transplantation of ovarian tissue collected during early infancy. The use of single-cell transcriptomics (scRNA-seq) has improved our understanding of ovarian biology and revealed exciting gene and protein candidates for genes and proteins involved in aging or treatment-induced damage. However, there is no clear understanding of the distribution of cells within the tissue, nor is there any distinction between follicle subpopulations. This study aims to construct a comprehensive spatiotemporal reference map of the human ovarian cortex, utilizing advanced transcriptomic and proteomic technologies, with a focus on unraveling the impact of chemotherapy and age-related changes. We analyzed a unique cohort of samples from pediatric patients undergoing fertility preservation (both treated and untreated), reproductive-age adults undergoing either gender reassignment surgery or cesarean section, and post-menopausal women. The goal was to understand patterns of cell localization, proportion changes, and associations with clinical features (age shift or treatment exposure). Utilizing a high-parameter fluorescent-based multiplex immunohistochemistry pipeline, we leveraged the expertise of the Human Protein Atlas (HPA) to identify five main cell types: oocytes, granulosa cells, immune cells, endothelial cells, and Schwann cells. This multiplexed immunofluorescence imaging provided a holistic view of the spatial organization of ovarian cortex across different age and therapy groups. Next, protein signatures for novel populations or subgroups of cells identified by scRNA-seq were stained alongside the multiplex panel outlining the five main cell types. Spatial analysis, based on the nearest neighbor method, unraveled the organization of the ovarian cortex, providing, for the first time, a spatiotemporal map of the human ovary across lifespan and treatment exposures. An in-depth view of this complex proteomic landscape offers a novel perspective that may shed light on uncharted territory and lead to breakthroughs in clinical interventions and therapeutics. 409

410

P-I-0259 Spatial proteomics characterizes the transition of precancerous lesions of the fallopian tube to invasive cancer Andreas Metousis1, Lisa Schweizer1, Agnes Bilecz2, Isabel Alcazar2, Rahul Krishnan2, Aasa Shimizu2, Sophia Steigerwald1, Thierry M. Nordmann1, Ricardo Lastra2, Ernst Lengyel2, Matthias Mann1 1Max Planck Institute of Biochemistry, Proteomics and signal transduction, Planegg, Germany 2The University of Chicago, Department of Obstetrics and Gynecology, Chicago, IL, United States Ovarian cancer is the deadliest gynecologic malignancy, affecting more than 300,000 women yearly. Most patients will be diagnosed at advanced stages, when the prognosis is poor. Earlier detection requires a better understanding of the precancerous and early stages of the disease. A prominent hypothesis on the origin of serous ovarian cancer is that it comes from early lesions in the fallopian tubes. These lesions are characterized by aberrant p53 expression and are divided into p53 signatures, Serous Tubal Intraepithelial Lesions (STILs) and Serous Tubal Intraepithelial Carcinomas (STICs), based on histopathological features. Such lesions are usually tiny in size, spanning only a few cells. This poses technical challenges, especially for input-limited approaches like mass spectrometry-based proteomics. Here we set out to characterize the transition of early lesions of the fallopian tubes to invasive tumors by applying Deep Visual Proteomics (DVP), an ultra-sensitive, cell-type resolved, spatial proteomics method that we developed. We started with staining fallopian tube FFPE tissue sections for markers of interest. Based on these markers, we segmented single cells by employing deep learning and classified them in normal ciliated, normal secretory, p53 signature, STIL, STIC, invasive tumor and metastatic cells. From each of these categories, we laser microdissected epithelial cells (15,000 square um per sample) and matching stroma (25,000 square um per sample) and prepared the resulting samples in a semi-automatic fashion. Last, we measured the samples on a Thermo Orbitrap Astral coupled to an Evosep One liquid chromatography system running a 40 samples per day gradient. We identified a median of 6562 and 5072 protein groups for epithelial and stromal samples, respectively. The corresponding PCAs demonstrated a partial separation between samples of different categories. Strikingly, the respective volcano plots uncovered hundreds of differentially regulated proteins in most of the pairwise comparisons. For example, members of the MCM complex, which is crucial for DNA replication, were upregulated in STIC epithelial cells vs normal secretory cells of the fallopian tube. On the stromal level, progression to STIC was associated with reduced amounts of collagens 1,3,5 and 6. Importantly, pseudotime analysis on single patients recapitulated and confirmed the hypothesized trajectory of early lesions to advanced disease. Along this progression, proteins such as ASPH and CNN2 gradually increase, while others like LAMB3 and WFS1 show the opposite trend, suggesting proteins with potential therapeutic or diagnostic implications. In conclusion, this first deep proteomic characterization of the transition of tiny early lesions of the fallopian tubes all the way to invasive and metastatic tumors, on a spatial and cell-type resolved manner, sheds unique insights into this deadly disease. 410

412

P-I-0260 Investigating chemoresistance in triple negative breast cancer, by imaging-guided MS-proteomics José Nimo1, Dražen Papić2, Stefan Florian2, Fabian Coscia1 1Max Delbrück Center, Berlin, Germany 2Charité – Universitätsmedizin Berlin, Pathology, Berlin, Germany Triple negative breast cancer (TNBC) is an aggressive disease whose standard treatment is neoadjuvant platinum-based chemotherapy. However, chemoresistance and relapse are common for this cancer subtype. While it is clear that the understanding the tumor microenvironment (TME) is essential for improving chemotherapeutic efficacy, systematic analyses with high spatial and phenotypic resolution have remained limited. My aim is to better understand the TME of TNBC with cell-type and spatial resolution by expanding the Deep Visual Proteomics framework. We analyzed a cohort of 90+ patient biopsies, before and after chemotherapy, to characterize key biomarkers predictive of the variable treatment outcomes. We performed multiplex imaging with a tissue-specific panel of 25+ markers to identify the majority of cell types of the TME. Images are stitched, registered, segmented, and cells classified with a Nextflow pipeline --- taking advantage of its transparency and replicability. Analyzing the spatial distribution of all cells allowed us to categorize distinct cellular niches; this perspective has the potential to identify novel associations between cellular distributions and diverse treatment outcomes. Lastly, these analyses will enable the prioritization of the most interesting cellular phenotypes, which will be collected by laser microdissection for deep proteomic profiling by mass spectrometry (4000+ proteins), characterizing their global, functional states. These multimodal data provide us with unique insights into spatially and cell-type resolved proteomic changes induced by chemotherapy of TNBC, potentially identifying novel response markers that could improve clinical decision making in the long term. 412

413

P-I-0261 Democratizing spatial proteomics: creative solutions for an affordable pipeline Thibault Courtellemont1, Romain Hamelin1, Florence Armand1, Jessica Sordet-Dessimoz1, Maria Pavlou1 1EPFL, Lausanne, Switzerland Over the past decade, the field of spatial biology has experienced remarkable growth, leading to a paradigm shift in our understanding of cellular organization and function. Initially dominated by spatial genomics and transcriptomics, spatial proteomics is now emerging as a pivotal discipline within this domain, providing unprecedented insights into the spatial distribution of proteins within cells and tissues. Despite significant technological advancements, spatial proteomics remains challenging due to limited cell quantities, diverse contaminants, and significant sample loss throughout the protocol steps, all of which constrain the dynamic range of proteins and result in a reduced number of quantified proteins. Overcoming these obstacles requires innovative strategies and resourcefulness. Recently, the combination of methods involving artificial intelligence-driven image analysis and segmentation, single-cell proteomics, and high-throughput techniques has paved a new road in spatial proteomics (1). This pioneering workflow can utilize formalin fixation and paraffin-embedding (FFPE) samples, providing access to billions of samples from patient biopsies stored in biobank archives, thus offering new hope for our understanding of some pathologies (2). In this study, we aim to generalize spatial proteomics by capitalizing on the competencies present at the EPFL-SV Research Core Facilities (Proteomic , Histology , Bioimaging and Optics and Flow Cytometry Core Facility). We have developed an accessible and user-friendly approach for manual tissue microdissection coupled with bottom-up MS-based proteomic analysis to investigate the spatial variability of the proteome in different samples. Leveraging creative solutions and do-it-yourself techniques, we have developed a streamlined workflow that can be easily implemented on most proteomic platforms without requiring substantial investments. Our approach combines laser-cut microscopy (LCM) with deep proteome profiling using a sensitive data-independent acquisition (DIA) on an Orbitrap Exploris 480 mass spectrometer. Inspired by a recent study (3), we adopted a high-resolution small mass range and large window DIA approach. This enabled us to quantify up to 7000 proteins from regions containing fewer than 100 mouse hepatocytes. Notably, our method delivers rapid results, with a turnaround time of less than a day from slide preparation to final data analysis. This method can also be easily adapted to cell sorting for FACS downstream analysis. We are now collaborating with research groups at EPFL to validate our assay in diverse biological studies, spanning various models from tissue organisms to organoid models. This collaborative effort aims to address fundamental questions related to organism development and pathology, thus advancing our understanding of complex biological systems. 1.https://doi.org/10.1038/s41587-022-01302-5 2.DOI: 10.1002/path.5420 3.doi: 10.1021/acs.jproteome.3c00736 413

414

P-I-0263 Microscopy-guided proteomics to explore non-genetic drug resistance Di Qin1, Sonja Fritzsche1, Vanessa-Deborah Sachse2, Ingeborg Tinhofer-Keilholz2, Fabian Coscia1 1MDC Berlin, Berlin, Germany 2Charité – University Medicine Berlin, Berlin, Germany Background: All cancer treatments face the problem of drug resistance. Genetic factors that lead to treatment failure have been extensively studied in past decades. However, up to 40% of recurrent and resistant tumors have failed to identify a clear genetic cause. Accumulating evidence indicates that non-genetic mechanisms play an important role in primary and acquired drug resistance, but these processes have been largely overlooked. Interestingly, even under identical genetic backgrounds and in the absence of drug exposure, resistant cell fates are often predetermined by the cell"s intrinsic molecular makeup. This suggests that identifying these discrete molecular programs functionally linked to primary drug resistance could ultimately allow us to better stratify patients with the highest chance of treatment success. Cell heterogeneity is a well-recognized reason that significantly contributes to drug resistance. Recently, imaging-based single-cell proteomic studies revealed that approx. one-fifth of the proteome [MOU1] shows cell-to-cell variability emphasizing the importance of protein layer information in studying cell diversity linked to drug resistance. We therefore hypothesize that the analysis of single-cell proteome heterogeneity is another critical aspect of understanding primary drug resistance mechanisms. Method and Results: We developed a new concept of imaging-based proteomics of single-cell derived colonies that combine the benefits from both single-cell and bulk proteomics concepts. In our approach, single-cell derived colonies can be imaged, isolated, and comprehensively profiled (6,000+ proteins) by proteomic profiling. Our method enables the integration of functional marker imaging (e.g. DNA damage marker) and cell morphology assessments with quantitative proteomics read-outs. As proof-of-concept experiment, we profiled a single-cell derived clonal head and neck cancer cell line to study sub-clonal proteome heterogeneity linked to primary drug resistance. Our data highlighted hundreds of proteins with subclone specific protein level changes including signatures previously associated with primary chemo-resistance. Interestingly, these findings also aligned with HNSCC patient tissue proteomes, underscoring the potential of this pipeline for bridging in vitro and in vivo results. Conclusion: In conclusion, we propose an imaging-guided proteomics method of single-cell derived colonies, harnessing the strengths of both single-cell and bulk proteomics approaches. Our method has the potential to investigate non-genetic protein-level changes associated with drug resistance. This versatile pipeline can be adapted to address various research questions. 414

415

P-I-0264 Label-free DIA-based workflow for single-cell proteomic analysis on an Orbitrap ascend tribrid mass spectrometer Fernanda Salvato1, Bernard Delanghe1, Julia Kraegenbring1, Amirmansoor Hakimi1, Tonya Pekar Hart1 1Thermo Fisher Scientific, San Jose, CA, United States Background: Recent advances in LC-MS have enabled label-free single-cell proteome analysis, revealing unexpected functional diversity in cells. However, there are still key challenges in this field, such as sensitivity, coverage, dynamic range, and throughput. To address some of these challenges, new method developments as well as optimization of existing LC-MS-based proteomics workflows are necessary. Here, we demonstrate the use of the Orbitrap Ascend Tribrid mass spectrometer and the Vanquish Neo UHPLC system for high-throughput single cell applications. Methods: Individual HeLa cells were sorted, followed by reduction, alkylation, and trypsin digestion using CellenONE as per the manufacturer"s protocols. Pierce HeLa digest was used for dilution series from 50 pg to 10 ng loaded on column. Single cell digests and the diluted standard HeLa digest samples were analyzed using the Orbitrap Ascend Tribrid MS with the FAIMS Pro interface coupled to the Vanquish Neo UHPLC system. Separation was performed on the Aurora Ultimate TS 25cm column. Data was acquired in a DIA mode and searched with a beta version of Proteome Discoverer Software 3.1 and Spectronaut 18. Results: The performance of this ultra-sensitive LC-MS workflow was first optimized using a dilution series of HeLa digests. From 250 pg of HeLa digest load, we could identify on average 3,200 protein groups by a library- free approach, whereas using DIA-library generated from 5 and 10 ng of HeLa digest, we were able to identify around 5,000 protein groups using a method with a throughput of 50 samples per day. HeLa single cells resulted in an average of 3,700 protein groups and 21,000 peptides using library-free searches and 4,400 protein groups and >26,000 peptides in library-based searches. Conclusion: End-to-end workflow for single-cell proteomics on an Orbitrap Ascend Tribrid MS with sensitive detections of proteins from single cell or low-input samples. 415

416

P-I-0266 Spatial proteomics provides new insight into proteostasis processes during acute mitochondrial stress Lakshita Sharma1, Silke Meiners2, Sven Dennerlein3, Bettina Warscheid1 1Julius-Maximilians-Universität Würzburg, Department of Biochemistry II, Wuerzburg, Germany 2Research Lung centre, Leibniz Lung centre, Borstel, Germany 3University Medical Center Göttingen, Department of Cellular Biochemistry, Goettingen, Germany Objectives: Mitochondria are crucial for metabolism in eukaryotic cells which primarily produce ATP through oxidative phosphorylation. To fulfill their various functions, mitochondria need a healthy proteome that is maintained through the import of mitochondrial precursor proteins, a process highly scrutinized by the ubiquitin-proteasome system (UPS). Mitochondrial dysfunction and defects in the UPS are linked to human diseases. Problems with mitochondrial protein import trigger a cellular stress response, with UPS quickly trying to remove the mislocalized precursor proteins. However, knowledge about the different factors and mechanisms involved in mitochondria-associated degradation (MAD) pathways is incomplete. This study aims to explore the UPS-dependent factors and processes that uphold mitochondrial and cellular proteostasis during acute mitochondrial dysfunction using spatial and functional proteomics approaches. Methods: We induced mitochondrial protein import stress in human HEK293T cells by inhibiting the presequence protein import pathway with a chemical probe. To study changes in protein abundance and localization, we perform global and spatial proteomics using quantitative MS technology. Subcellular fractions are generated by differential and density gradient centrifugation. Protein correlation profiling analysis with organellar marker proteins is used to reveal relocalization events. Selected candidates from proteomics data are further examined by high-resolution microscopy. Validated UPS factors are characterized for their role in MAD using functional proteomics, including loss-of-function analysis, protein proximity labelling, and ubiquitinome profiling. Results: Through spatial proteomics, we profiled different subcellular niches and analyzed the subcellular localization of proteins during mitochondrial dysfunction. Our spatial data indicate the involvement of novel components of the UPS in MAD processes. Notably, the dynamic relocalization of central subunits of the proteasome machinery appears to represent a direct mechanism for the efficient clearance of arrested and/or non-imported mitochondrial precursor proteins accumulating at the mitochondrial surface. Our initial screen has also identified several E3 ligases associated with dysfunctional mitochondria and first interactome data indicate a specific recruitment of UPS factors. We are currently validating these findings using multiple experimental approaches. Conclusion: Our spatial proteomics and microscopic analyses proved to be effective for identifying new UPS factors and processes involved in MAD pathways in human cells. Our data indicate that the specific relocalization of UPS factors to a subcellular niche where proteostasis is disturbed provides a direct, fast mechanism for maintaining a functional cellular proteome during acute mitochondrial stress conditions. 416

417

P-I-0267 Investigation of antiangiogenic tyrosine kinase inhibitors in primary and metastatic tumor models Anna Solyom-Tisza1,2, Szilvia Torok1, Yutaka Sugihara3, Aron Gellert4, Edina Bugyik1, Judit Berta1, Ildiko Kovacs1, Nandor Barany1, György Marko-Varga3, Jozsef Tovari5, Sandor Paku2, Balázs Döme1,4,6, Katalin Dezso2, Viktoria Laszlo1,6, Melinda Rezeli3 1National Korányi Institute of Pulmonology, Department of Tumor Biology, Budapest, Hungary 2Semmelweis University, Department of Pathology and Experimental Cancer Research, Budapest, Hungary 3Lund University, Department of Biomedical Engineering, Lund, Sweden 4National Institute of Oncology-Semmelweis University, Department of Thoracic Surgery, Budapest, Hungary 5National Institute of Oncology, Department of Experimental Pharmacology, Budapest, Hungary 6Medical University of Vienna, Department of Thoracic Surgery, Comprehensive Cancer Center, Vienna, Austria Background: A frequent and poor prognostic event in the progression of malignant tumors is the appearance of lung metastases. Patients with lung metastasis often receive antiangiogenic (AA) agents in combination with chemotherapy. However, the efficacy of these combinations falls far short of what would be predicted from preclinical animal studies, resulting in an overall survival benefit of only a few months. AA agents that are applied in the clinic are designed to inhibit the mechanism of endothelial sprouting, however, in highly vascularized organs (e.g. lung), alternative vascularization mechanisms are of greater importance. Since these organs are also the most common sites of metastasis of solid tumors, we hypothesize that primary and metastatic tumors may use distinct mechanisms to form their vascular network, and thus respond differently to AA therapy. Objectives: Our aim was to compare the efficacy (i.e. inhibition of tumor growth and angiogenesis) of an AA agent, sunitinib malate in in vivo models of primary and secondary breast and renal cell carcinomas. Materials and methods: The following cell lines were selected for our studies: 4T1 mouse mammary carcinoma, MDA-MB-231 human mammary carcinoma, RENCA mouse renal cell carcinoma. Cells were inoculated orthotopically in the case of primary tumor models, while injected into the tail vein to mimic migration and tumor formation of lung metastases. Mice were treated with 80 mg/kg sunitinib malate suspended in 1.5% carboxymethyl cellulose (CMC) or with 1.5% CMC in the control group for 2x5 days. The proportion of microvessels and proliferating cells, as well as the size of apoptotic and hypoxic areas were examined by immunofluorescence labeling on frozen tissue samples. The tissue distribution of sunitinib was investigated using matrix-assisted laser desorption/ionization mass spectrometry imaging (MALDI-MSI). Results: Treatment with sunitinib malate significantly reduced tumor size and weight in all primary models and in the RENCA metastasis model, but had no effect on tumor growth in metastasis models of breast carcinomas. Moreover, the area of the vascular network in the sunitinib treated tumors was significantly decreased compared to controls. Additionally, we found that sunitinib malate treatment increased the proportion of hypoxic and necrotic areas only in tumors of the primary models. Evaluation of the MSI data revealed that the mean sunitinib signal was higher in metastatic tumors than in the corresponding primary models, and this difference was significant in breast carcinomas. Furthermore, we found that in all primary models, a significantly higher sunitinib signal was detected in the viable tumor compartments compared to the non-viable areas. A similar trend was observed in the metastasis models, but no significant difference was found. Funding: Mrs. Berta Kamprad"s Cancer Foundation (FBKS-2020-22-(291), National Research, Development and Innovation Office (2021-1.2.4-TÉT-2021-00030). 417

418

P-I-0268 The proteomic landscape of synaptic diversity across brain regions and cell types Marc van Oostrum1,2, Thomas Blok2, Stefano L. Giandomenico2, Susanne tom Dieck2, Georgi Tushev2, Nicole Fürst2, Julian D. Langer2, Erin Schuman2 1Biozentrum, University of Basel, Basel, Switzerland 2Max Planck for Brain Research, Frankfurt, Germany Neurons build synaptic contacts using different protein combinations that define the specificity, function, and plasticity potential of synapses; however, the diversity of synaptic proteomes remains largely unexplored. We prepared synaptosomes from 7 different transgenic mouse lines with fluorescently labeled presynaptic terminals. Combining microdissection of 5 different brain regions with fluorescent-activated synaptosome sorting (FASS), we isolated and analyzed the proteomes of 18 different synapse types. We discovered 1,800 unique synapse-type-enriched proteins and allocated thousands of proteins to different types of synapses (https://syndive.org/). We identify shared synaptic protein modules and highlight the proteomic hotspots for synapse specialization. We reveal unique and common features of the striatal dopaminergic proteome and discover the proteome signatures that relate to the functional properties of different interneuron classes. This study provides a molecular systems-biology analysis of synapses and a framework to integrate proteomic information for synapse subtypes of interest with cellular or circuit level experiments. 418

419

P-I-0269 Reducing the impact of contamination to produce reliable single cell proteomics data Alienke van Pijkeren1, Kristoffer Basse1, Surina Chuckaree1, Jing Zheng1, Mohammed Hanzala Kaniyar2, Omar Rosas Bringas2, John LaCava2,3, Peter Horvatovich1, Guinevere Lageveen-Kammeijer1 1Groningen Institute of Pharmacy, University of Groningen, Analytical Biochemistry, Groningen, Netherlands 2University Medical Centre Groningen, European Research Institute for the Biology of Ageing, Groningen, Netherlands 3The Rockefeller University, Laboratory of Cellular & Structural Biology, New York, NY, United States Introduction: Single cell proteomics (SCP) is an emerging technique advancing our understanding of cellular protein heterogeneity. Hitherto, the community is pushed to report on the highest detected and quantified proteins, but it is important to note that the protein content of the desired sample is in the few-hundred picogram range and even a small amount of contamination can lead to large effects in the measured data. In this project, we aim to explore and identify the cause of potential contamination, by investigating a range of experimental parameters during cell isolation and sample preparation. Next, our aim is to assess technical and biological variability of SCP measurements as biological variability is key to assess single cell heterogeneity and is essential to understand molecular heterogeneity in cells. Material and Methods: All experiments were performed on the CellenONE cell isolation platform and the investigated parameters include the different cell types, the target plate type (96-well plate and Cellenion EVO96 proteoCHIP), the use of laminar flow cabinet, as well as extensive cleaning of the CellenONE prior to usage. Next to dummy samples (zero cells dispensed in the well; N>=16), each plate also contained single cells (N>=36) and a few samples with higher cell counts up to 50 cells. All samples were processed and analyzed using a nanoElute2 LC system coupled to a timsTOF SCP-MS. Additionally, various data analysis software packages (Bruker ProteoScape, Spectronaut V18, and DIANN 1.8.1) have been evaluated as the best practice for analyzing a cohort that contains a varying number of cells per cohort. Results: Preliminary data analysis has been performed on "0 cells" samples (treated identically to the single cell samples that were generated in parallel) from isolations of N2102Ep cells into either a standard 96-well plate or the CellenONE EVO96 proteoCHIP using both Spectronaut and DIANN. Overall, both tools showed that the usage of the EVO96 proteoCHIP resulted in lower contamination (57% and 97% decrease in Spectronaut and DIANN, respectively). While DIANN overall provided a lower identification of proteins, 7 (636 precursors) proteins compared to 284 (989 precursors) in Spectronaut. For more results see Figure. Further tests will include the bulk samples in the analysis runs to evaluate the effect of boosting (allow peptides that are identified in one sample, to be used in the search in other samples from the same cohort) and to filter the detected proteins based on the number of peptides associated with them to evaluate the reliability of the protein inference process as implemented in both Spectronaut and DIANN. Conclusions: In conclusion, as SCP is gaining momentum, we hope with our obtained knowledge to raise awareness and eventually provide a general recommendation to the SCP community on how contamination can be avoided or minimized, and how best to account for any contamination in the data analysis process. 419

421

P-I-0270 Location, location, location: exploration of the spatial proteome of plants Monique van Schie1, Mark Roosjen1, Catherine Albrecht1, Jarik van Marsdijk1, Dolf Weijers1 1Wageningen University, Laboratory of Biochemistry, Wageningen, Netherlands The development of organelles is an important hallmark of evolution. Organelles offer unique micro- environments in the cell where a broader range of proteins can function, and provide an additional layer of regulation by separating or concentrating proteins. Hence, many cellular processes involve movement of proteins between organelles and subcellular localization of proteins can often be linked to their functions. The systematic assessment of the localization of proteins is therefore insightful, but also technically challenging. In plants specifically, the presence of a rigid cell wall and organelles such as chloroplasts and large vacuoles add to the challenges. Hence, knowledge about the spatial proteome of plants is lacking compared to other eukaryotes, which complicates the assignment of functions to genes. To address this problem, we used our expertise on both plant biology and proteomics to adapt pre-existing methods from the broader spatial proteomics field for use in plants. We were not only able to obtain spatial data on the widely used model plant Arabidopsis thaliana, but also on the less well known evolutionary model Marchantia polymorpha, one of the first plants to grow on land. Furthermore, we compared the spatial proteome of cultured cells, organs and whole organisms to study the differences of the spatial proteome. Additionally, we used a cross-linking approach to study protein complexes in their spatial context. Lastly, we show that these approaches not only offer static maps of protein subcellular localization, but can also be used to detect proteins that move between organelles in different conditions or genetic backgrounds. As an example we treated Arabidopsis roots with the protein trafficking disrupting drug Brefeldin-a, identified the translocating proteins and confirmed a number of those translocations in planta. Summarizing, we will present not only an extensive resource of protein subcellular localization and complex formation in plants, but also a strategy to identify translocating proteins during a great variety of processes at the subcellular level. 421

422

P-I-0271 Neuromelanin granules show a characteristic proteomic profile in Parkinson's disease Maximilian Wulf1,2, Katalin Barkovits-Boeddinghaus1,2, Karin Schork1,2, Peter Riederer3,4, Manfred Gerlach5, Britta Eggers1,2, Katrin Marcus-Alic1,2 1Ruhr University Bochum, Medizinisches Proteom-Center, Bochum, Germany 2Ruhr University Bochum, Medical Proteome Analysis, Center for Proteindiagnostics (PRODI), Bochum, Germany 3University Hospital of Würzburg, Center of Mental Health, Department of Psychiatry, Psychosomatics and Psychotherapy, Wuerzburg, Germany 4University of Southern Denmark Odense University Hospital, Psychiatry Department of Clinical Research, Odense, Denmark 5University Hospital of Würzburg, Center of Mental Health, Department of Child and Adolescent Psychiatry, Psychosomatics and Psychotherapy, Wuerzburg, Germany Neuromelanin granules (NMGs) are organelle-like structures present in the human substantia nigra (SN) pars compacta. NMGs are of high clinical relevance, because neurons with a high NMG content are more vulnerable in the context of Parkinson"s disease (PD) and dementia with Lewy bodies (DLB). Nevertheless, the role of NMGs in health and disease remains incompletely understood. Since NMGs are absent in the brain of most common laboratory animals without genetic modifications, researchers are dependent on human post-mortem brain tissue, which is highly requested and therefore limited. To overcome this limitation, we established a protocol for the proteomic characterization of NMGs in comparison to the surrounding SN (SNSurr.) tissue, based on the use of laser microdissection and bottom-up proteomics (Wulf et al., 2021). Two follow-up studies have been conducted using this workflow, with the proteome of NMGs being characterized under physiological conditions (Wulf et al., 2022a) and neurodegenerative conditions using post-mortem brain tissue of DLB cases (Wulf et al., 2022b). A comparison of the proteomic profiles of NMGs from CTRL and DLB cases revealed that alpha-synuclein and protein S100A9 were significantly higher abundant in NMGs of DLB cases, indicating the presence of a DLB-specific proteomic profile in NMGs. In the current study, we analyzed post-mortem brain tissue from eight PD and ten CTRL cases and compared the proteomic content of NMG and SNSurr. samples. In concordance to our study on DLB-specific proteomic changes, alpha-synuclein was again significantly higher abundant in PD NMGs compared to CTRL NMGs. Despite this concordance, other upregulated proteins differed, indicating that proteins potentially co-accumulating in NMGs with alpha-synuclein may be disease-specific and thus indicate disease specific pathomechanisms. From a methodological perspective, we were able to drastically reduce the amount of tissue required and the analysis time after laser microdissection. By using the EvoSep One in combination with the TimsTOF Pro 2, we were able to reach yet unprecedentedly high numbers in protein identifications while reducing the sample input more than 10-fold. These results drastically demonstrate the benefits of state-of-the-art instrumentation for the research on NMGs. 422

423

P-I-0272 Spatially resolved ultra visual MicroProteomics via hydrogel-based tissue expansion Shuchang Zhao1, Kiryl Piatkevich2, Tianing Liu1, Zhien Rong1, Cuiji Sun1, Guanqiao Dong1, Tiannan Guo1 1Westlake University, Life science school, Hangzhou, China 2Westlake University, Life science school, Hangzhou, China Nanoscale spatially resolved proteomics is crucial for mapping the heterogeneity of proteomes within biological samples. However, this remains a technical challenge due to the complexity of tissue micro-sampling techniques and the mass spectrometry analysis required for nanoscale specimen volumes. Here, we introduce a novel method called microProteomEx, which combines spatially resolved visual proteomics, enabling MS-based analysis of intact tissue samples guided by super-resolution imaging. microProteomEx enables the quantitative profiling of spatial proteome variability in mammalian tissues with an impressive lateral resolution of approximately 57 µm. This is achieved through re-embedding micro-sampling, which does not require custom or specialized equipment. When paired with ultrasensitive liquid chromatography-mass spectrometry, microProteomEx can identify around 900 proteins from tissue samples as small as ~12 cells. Region of interest selection is guided by super-resolution imaging of the expanded samples under a conventional diffraction- limited microscope, allowing the identification of ultrastructural morphological features with an effective spatial imaging resolution of 47.0 nm. Compared to the original protocol, ProteomEx, we have developed a workflow that is faster, simpler, offers higher spatial resolution, and achieves greater protein identification. We have validated the utility of microProteomEx in profiling the proteomes and super-resolution imaging of various biological tissues, including the brain, liver, kidney, and heart. This method preserves both the spatial proteomic data and cellular structure imaging information within the tissue context, which has significant implications for the molecular profiling of clinical samples and mouse models of Alzheimer’s disease. The combination of super- resolution imaging with spatially resolved proteomic analysis of intact tissues enables a new perspective of deep visual proteomics where proteome and cytoarchitecture can be analyzed and correlated on the same sample. Fig. 1 (optional) 423

424

P-I-0273 Deciphering the complexity of spatial-temporal proteomics via subcellular location optimal transport Jun Zhu1,2, Hongchao Li2, Xiaowen Wang2,3,4, Fuchu He2,3,4, Cheng Chang2,3,4 1Tsinghua University, School of Life Sciences, Beijing, China 2National Center for Protein Sciences (Beijing), Beijing, China 3State Key Laboratory of Medical Proteomics, Beijing, China 4Beijing Proteome Research Center, Beijing, China The localization of intracellular proteins is highly dynamic, which segregates protein expression and plays a crucial role in the functional polarization of cells. Recent advancements in technology have facilitated the high- throughput measurement of subcellular spatially resolved protein expression. However, methods to analyse these data are not well established. We present a machine-learning-based approach called Subcellular Location Optimal Transport (SLOT) to decipher the complexity of spatial-temporally variable proteins. By comparing the spatial expression patterns of different proteins, SLOT identifies potential spatially co-localized proteins, shedding light on their functional relationships within the cell. SLOT graphs are constructed for proteins at different time points, and these temporal SLOT graphs are aligned across time series to elucidate dynamic changes in protein co-localization patterns over time. We applied SLOT to reveal spatial-temporal proteomic changes in Xenopus oocytes and early embryos, uncovering previously unexplored intricacies in protein localization dynamics during early developmental stages. These findings underscore the utility of SLOT in providing mechanistic insights into fundamental cellular processes, such as development and differentiation. Our study demonstrates that SLOT is a powerful analytical tool for deciphering the spatial-temporal dynamics of protein localization within cells. Furthermore, the insights gained from SLOT analysis enhance our understanding of key cellular processes, offering new avenues for research in developmental biology and beyond. 424

425

Clinical proteomics I (P-I-0275 to P-I-0383) P-I-0275 Proteomics multiplex technologies: an ideal clinical tool for biomarker discovery and precision medicine for CNS conditions Houari Abdesselem1 1Hamad Bin Khalifa University, Doha, Qatar Discovering novel biomarkers of Central Nervous System (CNS) Disorders and Neurodegenerative Diseases is a fundamental step towards developing new personalized diagnostic, prognostic and therapeutic tools for many life-threatening CNS conditions. As of now, modern medicine still lacks accurate personalized diagnostic and therapeutic tools, which will have great impact on future clinical trials research, and which remains an urgent need for clinical management of patients with CNS conditions. Here, we highlight recent studies we conducted on Autism Spectrum Disorder (ASD), Alzheimer’s Disease, post-Severe COVID-19 sequelae with neurological complications, and ongoing work on Multiple Sclerosis Disorder, using powerful multiplex proteomics technologies such as OLINK and SENGENICS to discover novel biomarker proteins and autoantibodies of CNS conditions in patients of Qatari population, to be used for future development of diagnostic and therapeutic purposes. 425

426

P-I-0276 Investigating molecular mechanisms and novel biomarkers of postoperative recurrence in hepatocellular carcinoma: proteomic analysis of FFPE samples Yuki Adachi1,2, Rei Noguchi1, Julia Osaki1, Shuhei Iwata1, Yomogi Shiota1, Mishie Tanino3, Kazuki Sasaki4, Sumio Ohtsuki5, Hideki Yokoo2, Tadashi Kondo1 1National Cancer Center Research Institute, Division of Rare Cancer Research, Tokyo, Japan 2Asahikawa Medical University Hospital, HBP and Transplant Surgery, Asahikawa, Japan 3Asahikawa Medical University Hospital, Diagnostic Pathology, Tokyo, Japan 4Sasaki Institute, Peptidomics, Tokyo, Japan 5Kumamoto University, Pharmaceutical Microbiology, Kumamoto, Japan Background: The molecular mechanism by which hepatocellular carcinoma (HCC) relapses after surgery is not well understood. Advances in mass spectrometry and data analysis techniques have enabled accurate and detailed proteomic analysis of formalin-fixed paraffin-embedded (FFPE) samples. This study aims to elucidate the molecular mechanisms behind postoperative recurrence in HCC and identify novel prognostic biomarkers through proteomic analysis of FFPE samples. Methods: We selected 43 cases with early recurrence (within one year) and 62 cases with no recurrence for over five years from 487 HCC patients who underwent initial curative hepatectomy between 2010 and 2022 in Asahikawa Medical University Hospital. Proteins were extracted from cancerous and non-cancerous areas of thin-sectioned FFPE samples using macro-dissection and ultrasonic treatment with Adaptive Focused Acoustics (AFA) technology to maximize extraction efficiency. Comprehensive proteomic analysis was performed using mass spectrometry after peptide digestion of the extracted proteins. Results: Approximately 8000 proteins were commonly identified in a quantitative way from all samples. Principal component analysis showed no obvious differences in protein profiles of non-cancerous areas between early recurrence and no recurrence groups, but proteomic differences were observed in cancerous areas of the patients with different prognoses. Notably, hundreds of differentially expressed proteins between the groups were detected in cancerous areas. Enrichment analysis revealed significant upregulation of multiple molecular pathways related to cell proliferation, such as DNA replication, cell cycle, and mismatch repair, in the early recurrence group, identifying several candidate biomarkers. Discussion: Several proteins and molecular pathways associated with postoperative recurrence of HCC were identified. These observations were consistent with previously reported cancer progression mechanisms, indicating the validity of our results. However, the complexity of recurrence mechanisms also highlights the limitations of predicting prognosis based on single biomarker changes. The development of a prognostic model integrating multiple biomarkers is currently underway. Conclusion: Proteomic analysis of FFPE samples identified protein groups involved in postoperative recurrence of HCC. However, developing precise prognostic biomarkers and novel therapeutic strategies requires a more comprehensive and detailed analysis approach that considers the interactions of multiple proteins and pathways. 426

427

P-I-0277 Identifying mass spectrometry-based biomarkers for predicting tumor recurrence in lung cancer Karim Aljakouch1,2, Qian-Wu Liao1, Marc Schneider3, Barbara Helm2, Dario Frey2, Christina Schmidt1, Thomas Naake4, Qiuqin Zhou5, Gernot Poschet6, Rüdiger Hell6, Hauke Winter3, Julio Saez-Rodriguez1, Wolfgang Huber4, Carsten Hopf5, Ursula Klingmüller2, Jeroen Krijgsveld1,2, Junyan Lu1 1Medical Faculty Heidelberg, Heidelberg, Germany 2German Cancer Research Center (DKFZ), Heidelberg, Germany 3Thoraxklinik Heidelberg, Heidelberg, Germany 4European Molecular Biology Laboratory (EMBL), Heidelberg, Germany 5Center for Mass Spectrometry and Optical Spectroscopy (CeMOS), Mannheim University of Applied Sciences, Mannheim, Germany 6Centre for Organismal Studies (COS), Heidelberg University, Heidelberg, Germany Cancer recurrence remains the main reason for cancer-related death, with nearly 50% of non-small cell lung cancer (NSCLC) patients experiencing recurrence after curative resection. As part of the SMART-CARE consortium, we aim to identify proteomic and metabolomic biomarkers to predict surgical outcomes and the necessity for chemotherapy to reduce recurrence risk. We collected snap-frozen tumor and adjacent tissue samples from 60 stage II adenocarcinoma patients, with half relapsing within two years post-surgery. Longitudinal plasma samples at baseline, at the point of recurrence, and after two years in patients without recurrence were also gathered. Subsequently, proteomic, targeted/untargeted metabolomic, and lipidomic profiles of those samples were acquired with state-of-the-art mass spectrometry technologies. Extensive bioinformatics quality control was performed to ensure high- resolution data free of environmental or technical variations, which are common in clinical omic data. We then applied various machine learning models to predict recurrence using these data modalities. Comparative analyses showed that linear XGBoost models with proteomic data from tumor-adjacent tissues most effectively predicted recurrence, with a cross-validated AUROC of 0.85, surpassing current genomic or transcriptomic models. Models based on baseline lipidomics performed slightly worse, with an AUROC of 0.78, but could be more accessible for clinical translation. We further applied multi-omics factor analysis (MOFA) to integrate biomarker candidates from all data modalities to understand their functional roles. This analysis revealed a gradient significantly associated with recurrence and shared between proteomic and lipidomic modalities. Enrichment analysis of the proteins associated with this gradient suggested a downregulation in cell- cell adhesion and upregulation in oxidative phosphorylation and fatty acid metabolism in tissues from patients prone to recurrence, highlighting the role of lipid metabolism in lung cancer prognosis. Our study presents viable protein and metabolite biomarkers for clinical use and offers a case study for future mass spectrometry studies aimed at clinical translation. 427

428

P-I-0278 Quantitative proteome analysis of normal and degenerated intervertebral disc tissue Divya Arunachalam1, Sharon Miracle Nayagam1, Narmatha Devi Paulraj1, Sunmathi Rajendran1, Murugesh Easwaran1, Karthik Ramachandran2, Sri Vijay Anand KS2, Chitraa Tangavel1, Rajasekaran Shanmuganathan2 1Ganga Orthopedic Research and Education Foundation (GOREF), Coimbatore, India 2Ganga Medical Center and Hospitals Pvt Ltd, Coimbatore, India Introduction: Intervertebral disc degeneration (IVDD) is a musculoskeletal disorder affecting people of all ages and causes low back pain. Factors causing IVDD includes genetics, trauma, ageing, life style, etc. resulting in excess catabolic over the anabolic responses. IVDD is also associated with modic changes as seen in MRI including inflammation and edema. Understanding the pathophysiology is crucial to manage the disease conditions. The present study explores the quantitative proteome analysis of normal and degenerated disc tissue to detect the differentially expressed proteins. Methods: Intervertebral disc tissues were obtained from brain-dead organ donors (TRANSTAN-214/2016; Transplant Authority Government of Tamil Nadu, India) and patients undergoing surgery with written informed consent. The study subjects include control, modic and non-modic intervertebral disc tissue. Proteins were extracted with 8M urea in 50mM TEAB buffer, sonicated for 5 cycles with 15 sec pulse on/off and isolated by centrifugation. After reduction and alkylation, proteins were digested with trypsin, labeled with TMT-6 plex kit and subjected to MS/MS acquisition in Q-Exactive plus Orbitrap quadrupole mass spectrometer. The experiments were performed as four independent experiments with three technical replicates to identify the differentially expressed proteins. Using Proteome Discoverer 2.5, the protein abundance were detected for further bioinformatic analysis. Results: Using mass spectrometry, 1200 proteins were identified in two out of four experiments (50% cut-off). Our earlier study using label free mass spectrometry revealed the upregulation of complement proteins and downregulation of ECM proteins in degenerated disc compared to normal disc (Rajasekaran et al., 2019). In this study, we specifically determined the expression of various complements and ECM related proteins from the TMT mass spectrometry data. In the complement pathway, complement proteins C3, C5 and MAC proteins (C6- 9) were upregulated in modic and non-modic tissue. Interestingly, complement regulators including CFH and CFI were increased in modic phenotypes. In ECM-receptor interaction pathway, CHAD, THBS3, TNXB were downregulated in modic and non-modic phenotypes compared to control. While the expression of collagens such as COL2A1, COL6A1, COL6A2 were decreased in modic. Conclusion: The present study revealed the alteration of complement and ECM proteins in degenerated disc compared to control. Although modic and non-modic condition activates complement pathway, the distinct expression of complement regulators CFH and CFI in modic, indicating the activation of alternate complement pathway. This findings hypothesize that bacterial infection triggers the alternate complement pathway and the infection leads to the downregulation of matrix proteins and collagens in modic. Further, the results should be validated in future by targeted PRM based mass spectrometry. 428

430

P-I-0279 Comprehensive proteomic and glycoproteomic profiling of Organoids derived from human hepatocellular carcinoma Geul Bang1,2, Eun Hee Han2, Jin Young Kim1 1Korea Basic Science Institute, Digital Omics Research Center, Cheungju, South Korea 2Korea Basic Science Institute, Cheungju, South Korea Organoids have emerged as a revolutionary ex vivo 3D culture model for medicinal research. Despite their promising applications, the detailed compositions of organoids, particularly their proteome and glycoproteome, remain underexplored, with limited studies comparing these profiles to those of the original tissues. Additionally, obtaining high-quality mass spectrometric data from the small sample amounts of organoids poses a challenge. In this study, we successfully generated proteomic and glycoproteomic profiles using only 1-5 μg of protein from hepatocellular carcinoma (HCC) organoids. Our global proteomics analysis identified 2,099 proteins, highlighting key metabolic processes such as RNA metabolism, nucleobase-containing small molecule metabolic processes, and carbon metabolism, which are reflective of the primary metabolic functions of liver tissue. Glycoproteomic analysis revealed 41 glycoproteins from 159 glycopeptides, including significant therapeutic targets and biomarkers for HCC, such as EGFR and AFP. Comparative analysis of glycan compositions between HCC tissue and HCC-derived organoids demonstrated their similarity, suggesting that organoids retain the essential characteristics of the actual tissues. 430

431

P-I-0280 Exploring the role of TMOD2 in colorectal cancer development and progression through functional proteomics Ana Montero-Calle1, Sofía Jimenez de Ocaña1, Raquel Rejas-Gonzalez1, Vivian de los Rios2, Javier Martínez- Useros3, María Jesús Fernandez-Aceñero4, Alberto Peláez-García5, Rodrigo Barderas1,6 1Instituto de Salud Carlos III, Chronic Disease Programme, Madrid, Spain 2CIB-CSIC, Madrid, Spain 3Translational Oncology Division, OncoHealth Institute, Fundación Jimenez Diaz University Hospital, Madrid, Spain 4Hospital Universitario Clínico San Carlos, Madrid, Spain 5Hospital Universitario La Paz, Madrid, Spain 6CIBERFES, Madrid, Spain Most colorectal cancer (CRC)-related deaths are associated to metastasis, with the liver being the primary organ for CRC colonization. Due to the lack of efficient treatments for cancer metastasis, the identification of new therapeutic targets either direct or upstream regulators of selected targets that focus on this metastasis is crucial to reduce CRC mortality. Tropomodulin-2 (TMOD2), a protein not previously linked to CRC, was found upregulated in highly metastatic to liver CRC cells. Here, we aimed to functionally characterize TMOD2"s role, and its interactome, and proteome in CRC. Methods involved inducing stable TMOD2 overexpression and transient silencing in low-metastatic and highly metastatic isogenic CRC cell models. In vitro and in vivo functional assays were performed to determine TMOD2"s role in CRC cells" tumorigenic and metastatic properties. Furthermore, a quantitative label free (LFQ) proteomics approach was performed to characterize the differential proteome associated with TMOD2 overexpression using an Orbitrap Astral mass spectrometer. In addition, an immunoprecipitation (IP) coupled with LC-MS/MS was performed to characterize the interactome associated with TMOD2 using a Q Exactive mass spectrometer.. Results suggested an important role of TMOD2 in inducing proliferation, migration, and anchorage-independent growth capacities of CRC cells. Moreover, TMOD2 was implicated in improving liver colonization and tumoral growth in metastatic niches and was observed to be involved in the regulation of focal adhesions. Quantitative proteomics identified 350 out of 5540 proteins as significantly dysregulated upon TMOD2 overexpression, with selected proteins whose dysregulation was confirmed by WB. By IP, 115 potential TMOD2 interacting proteins were identified, and 13 were validated as actual interactors by WB and immunofluorescence. These findings suggest a novel role of TMOD2 in CRC progression and metastasis, linking it to the induction of CRC cell migration, liver colonization, and proliferation. These enhanced capabilities characterized by functional proteomics may be associated with the increased number of focal adhesions induced by TMOD2 upregulation, and the differential proteome and interactome associated with TMOD2. 431

432

P-I-0281 High-throughput LC-MS for plasma proteomics in patients with stroke Teresa Barth1, Aileen Preuss1, Yasin Eshragi2, Teresa Wölfer2, Walter Viegener3, Jürgen Cox3, Steffen Tiedt2, Axel Imhof1 1Protein Analysis Unit, Biomedical Center Munich, Martinsried, Germany 2Institute for Stroke and Dementia Research, University Hospital LMU Munich, Munich, Germany 3Computational Systems Biochemistry, Max Planck Institute of Biochemistry, Martinsried, Germany As part of the CLINSPECT-M consortium (clinical mass spectrometry center Munich), we are focusing on diseases of the nervous system like stroke. Stroke is a leading cause of death and disability world-wide (1), with ischemic stroke as major form. Different etiologies lead to the formation of causative blood clots. Knowing the origin of the thrombus is essential for the strategy of secondary prevention medication. Proteins in blood might serve as biomarkers for stroke pathogenesis and thus allow a quick, low invasive diagnostic test. The PROMISE study (PRecisiOn Medicine In StrokE) includes 502 patients with ischemic stroke with serial high-frequency blood sampling in the acute phase in combination with deep clinical phenotyping. For a high-throughput measurement of a large number of plasma samples, we developed a pipeline to prepare samples in a 96-well format. We make use of automatization on the Agilent Bravo liquid handling system with the autoSP3 protocol (2). Peptides are analyzed on an Evosep One – Exploris 480 setup using data-independent acquisition (DIA). Subsequent data analysis is carried out using MaxDIA (3) and compared to other softwares in an unbiased, library-free manner. Sample throughput per week can be up to almost 280 samples. This pipeline now allows us to run thousands of plasma samples from the PROMISE study. (1) Global, regional, and national burden of stroke and its risk factors, 1990-2019: a systematic analysis for the Global Burden of Disease Study 2019 GBD 2019 Stroke Collaborators, Lancet Neurol. 2021, PMID: 34487721 (2) Automated sample preparation with SP3 for low-input clinical proteomics Müller et al., Molecular Systems Biology 2020, PMID: 32129943 (3) MaxDIA enables library-based and library-free data-independent acquisition proteomics Sinitcyn et al., Nat Biotechnol 2021, PMID: 34239088 432

433

P-I-0282 Impact of the tumor microenvironment in splenic metastasis in high-grade serous ovarian cancer revealed by quantitative proteomics Nicholas Bateman1, Suzanne Jockjtys1, Tamara Abulez1, Ryan Khan2, Brian Hood1, Kelly Conrads1, Allison Hunt1, David Bowtell3, Kathleen Darcy1, Neil Phippen1, Britta Weitgelt2, Dennis Chi2, George Larry Maxwell1, Thomas Conrads1 1Gynecologic Cancer Center of Excellence/ Women's Health Integrated Research Center, Annandale, VA, United States 2Memorial Sloan Kettering Comprehensive Cancer Center, New York, NY, United States 3Peter MacCallum Cancer Centre, Victoria, Australia Objectives: High grade serous ovarian cancer (HGSOC) patients often present with diffuse metastatic spread at diagnosis which correlates with poor prognosis. The relationship between tumor cell metastasis and the microenvironment remains poorly understood. Our study is focused on defining proteome profiles in tumor and stroma cell populations in primary and malignant tumors to improve our understanding of molecular profiles underlying metastatic spread. Methods: A quantitative mass spectrometry-based proteomic analysis was performed on laser microdissected (LMD) tumor and stroma cell populations from primary HGSOC tumors along with patient matched metastases from the liver, peritoneum, omentum, and spleen for 32 patients. Unique and selective proteome alterations in splenic metastases were validated from an independent analysis of LMD enriched tumor and stroma from 13 HGSOC patients. Results: We quantified >7,400 and >5,600 proteins from tumor and tumor-involved stroma collections, respectively, and unsupervised analyses showed that tumor cells cluster by patient while stromal cells largely by organ site (K-S p < 2.2e-16). A differential analysis showed that metastatic tumor cells from the spleen exhibited more uniquely altered proteins compared the omentum or liver, among which fourteen splenic-specific tumor cell proteins were validated in an independent set of LMD enriched primary (n=12) and splenic (n=6) tumor cell specimens (Spearman Rho = 0.793). Comparison of tumor-involved stroma collections from patient matched primary and splenic metastases identified 132 altered proteins that were highly correlated between these cohorts (Spearman Rho = 0.883, P<0.01). Comparison of tumor-involved stromal proteomes with benign spleen identified 30 proteins uniquely elevated in tumor-involved stroma from splenic metastases. Conclusions: Our study identified that proteome profiles of HGSOC tumor cells are highly conserved by patient while tumor-involved stroma correlates more with disease site. We also identified highly conserved protein alterations in primary and splenic metastasis in HGSOC patients, including a subset of features reflecting tumor- involved stroma in the spleen. These findings support improved molecular characterization of the tumor microenvironment with HGSOC. 433

434

P-I-0283 Saliva and stool proteomic profiling revealed potential biomarkers for inflammatory bowel diseases and colorectal carcinoma Ilaria Battisti1, Nicole Contran2, Giorgio Arrigoni2, Renata D'Incà3, Imerio Angriman3, Cinzia Franchin2, Maria Luisa Scapellato4, Andrea Padoan5, Stefana Moz5, Ada Aita5, Edoardo Savarino3, Greta Lorenzon3, Fabiana Zingone3, Gaya Spolverato3, Salvatore Pucciarelli3, Evelyn Nordi2, Paola Galozzi5, Daniela Basso5 1University Hospital of Padova, Operative Unit of Laboratory Medicine, Padova, Italy 2University of Padova, Department of Biomedical Sciences, Padova, Italy 3University of Padova, Department of Surgery, Oncology and Gastroenterology, Padova, Italy 4Univeristy of Padova, Department of Cardiac, Thoracic and Vascular Sciences and Public Health, Padova, Italy 5University of Padova, Department of Medicine, Padova, Italy Inflammatory Bowel Diseases (IBDs) are a heterogeneous group of chronic disorders, with Crohn"s disease (CD) and ulcerative colitis (UC) being the most frequent. Patients diagnosed with IBDs are at high risk of developing several neoplasia, including colorectal carcinoma (CRC), as a result of the inflammation. IBDs and CRC often exhibit shared pathogenetic mechanisms and display common non-specific symptoms, making the diagnostic process challenging. Moreover, traditional clinical practice based on blood laboratory tests and endoscopy can hardly detect such diseases at early onset. Hence, there is an urgent need for predictive novel biomarkers for both IBDs and CRC. Although serum is the preferred biological matrix, stools and saliva are also promising matrices to screen for biomarkers, considering their ease and non-invasive collection methods. Mass spectrometry-based proteomics is a high-throughput technology to study potential biomarkers on a large scale. In this study, we aimed to investigate the presence of biomarkers for IBDs and CRC in saliva and stool samples belonging to 20 healthy controls (CS), 12 CD, 13 UC, and 37 CRC, using a proteomic approach. Saliva samples were pooled and proteins were separated by SDS-PAGE before digestion, whereas endogenous peptides were extracted from pooled fecal samples using 10 kDa cut-off filters. LC-MS/MS analyses were performed with an LTQ Orbitrap XL coupled to a nano-HPLC Ultimate 3000 (Thermo Fisher Scientific), using a linear gradient of ACN / 0.1% FA to separate the peptide mixtures. A total of 152 proteins were identified in saliva samples, of which 73 were commonly found in CRC, CD, and UC, but not in CS. According to a Gene Ontology-based enrichment analysis, these proteins are involved in cell adhesion, regulator activity of peptidases, and glucose and nucleic acid metabolism. Of the remaining 79 proteins, those highly represented in CRC and CD, are related to inflammation, DNA stability, immunity, and oxidation. In stool samples, peptides belonging to 30 different proteins were identified, of which 4 were specific for CD. Among these, Basic Salivary Proline-Rich Protein 1 (PRB1) was highly abundant, whereas its isoform PRB2 was reduced in abundance in IBDs and CRC saliva samples. Moreover, two PRB1 peptides were shown to stimulate in vitro CRC cells" growth and the pro-proliferative signaling pathways Erk1/2, Akt, and p38. These results demonstrated that CRC and CD share common salivary proteins, and PRB1-derived peptides stimulate cancer cell growth, suggesting a potential role of such proteins in IBDs and cancer pathogenesis. Grant: PNRR-POC-2022-12375750 "Diagnostic innovation for inflammatory bowel diseases in children and adults: non invasive stool and salivary testing with ecologic new methods and biomaterials", funded by the European Union - NextGenerationEU – NRRP M6C2 - Investment: 2.1 "Enhancement and strengthening of biomedical research within the NHS. CUP I93C22000590006 434

435

P-I-0284 Phosphoproteomics and depletion studies in human serum – revealing the differences between healthy state, obesity and type 2 diabetes mellitus Petra Magdolna Bertalan1,2, Uladzislau Vadadokhau1,2, Ádám Pap3,4, Zsuzsanna Darula3,4, Gergő Kalló1, Miklós Káplár5, József Tőzsér1, Miklós Emri6, Éva Csősz1 1University of Debrecen, Department of Biochemistry and Molecular Biology, Debrecen, Hungary 2University of Debrecen, Doctoral School of Molecular Cell and Immune Biology, Debrecen, Hungary 3Hungarian Centre of Excellence for Molecular Medicine (HCEMM), Single Cell Omics Advanced Core Facility, Szeged, Hungary 4HUN-REN Biological Research Centre, Laboratory of Proteomics Research, Core Facility, Szeged, Hungary 5University of Debrecen, Department of Internal Medicine, Debrecen, Hungary 6University of Debrecen, Department of Medical Imaging, Debrecen, Hungary Introduction: Type 2 diabetes mellitus (T2D) is a worldwide known disease with an increasing incidence year by year. Extensive research has proven that besides the numerous risk factors, obesity is the main leading risk factor for the development of T2D. Regrettably, obesity also shows growing tendency because of an unhealthy lifestyle in the modern era. Aim: The aim of our study was to characterize proteins and phosphoproteins in serum samples from patients with obesity, T2D and healthy controls. To reveal the relevant biological processes between control, obese and T2D groups, network- and gene ontology (GO) analyses were performed. Materials and methods: 135 serum samples from obese, T2D and age- and sex-matched controls were collected by the Department of Internal Medicine, University of Debrecen. The protein concentration of the samples was determined by BCA assay. After it, the experiment was divided into two parts. Depletion was carried out, the top 14 most abundant proteins were depleted followed by digestion using iST kit (PreOmics) and two fractions were collected from the digested samples. Besides, non-depleted serum samples were used for TMT-labelling. The TMT-labelled peptides were fractionated using a high-pH reverse phase column and subjected to phosphopeptide enrichment. After the preparation of all samples, LC-MS analyses were performed using data- independent acquisition (DIA) method for depleted samples and data-dependent acquisition (DDA) for TMT- labelled samples. Data analyses were performed using DIA-NN and Proteome Discoverer. For network and GO analyses, Cytoscape and clueGO were used and the kinases having a role in the generation of the phosphorylated sites were identified. Results and conclusion: 232 proteins were found in the depleted samples whereas 268 proteins were identified by TMT-labelling experiments in the three patient groups. Protein clusters and pathways were characteristic to each condition were identified such as cell-cell connections, protein folding and endoplasmic reticulum stress. The complex analysis revealed proteins important in obesity and T2D. This research was funded by NKFIH FK134605, GINOP-2.3.3-15-2016-00020, KIM NKFIA 2022-2.1.1-NL-2022- 00005, European Union's Horizon 2020 research and innovation programme under grant agreement No 739593. 435

436

P-I-0285 Proteome signatures in urinary extracellular vesicles enable prostate cancer detection Irene Bijnsdorp1,2,3, Leyla A. Erozenci4,2,3, Jaco Knol4,2,3, Thang V. Pham4,2,3, Sander R. Piersma4,2,3, Berend Gagestein4,2,3, Gerald Verhaegh5, Jack Schalken5, Guido Jenster6, Martin E. van Royen7, Elena Martens- Uzunova6, R. Jeroen A. van Moorselaar1,2, Connie Jimenez4,2,3 1Amsterdam University Medical Center, Urology, Amsterdam, Netherlands 2Cancer Center Amsterdam, Amsterdam, Netherlands 3Amsterdam University Medical Center, OncoProteomics Laboratory, Amsterdam, Netherlands 4Amsterdam University Medical Center, Medical Oncology, Amsterdam, Netherlands 5Radboud University Medical Center, Urology, Nijmegen, Netherlands 6Erasmus MC Cancer Institute, Erasmus University Medical Center, Urology, Rotterdam, Netherlands 7Erasmus MC Cancer Institute, Erasmus University Medical Center, Pathology, Rotterdam, Netherlands Prostate cancer (PCa) diagnosis faces limitations, including high invasiveness and inaccuracy, highlighting the need for improved non-invasive diagnostic methods. Due to the distal location of the prostate to the urinary bladder, PCa markers can be detected in urine. Urine contains extracellular vesicles (EVs), which are secreted by various cell types, including PCa cells. The EV content in-part mirrors the content of the cells it originates from and has previously been proven to contain disease specific markers. To investigate whether PCa patients can be identified from the protein profiles in urinary EVs (uEVs), we performed in-depth LC-MS/MS-based proteomics profiling of uEVs from 222 individual patients, using multicenter-collected samples. We generated three datasets using DDA (n=43/D43, n=94/D94) and DIA (n=85/D85) and identified 2025, 2693, and 3891 proteins, respectively (30% data presence). Prostate-specific proteins (e.g. KLK3, FOLH1) and EV-associated proteins (e.g. CD9, CD63) correlated to an EV- score, calculated based on the top 100 EV proteins from vesiclepedia, confirming the presence of EVs and prostate-proteins within the uEVs. Gene Ontology analysis shows a highly similar function of protein in the 3 datasets, with proteins involved in immunity, metabolism and cell adhesion. In uEVs of PCa patients, 15 proteins were significantly upregulated, and 17 significantly downregulated, in 2/3 datasets. (Single-sample) Gene-Set- Enrichment-Analysis showed enrichment of immune-related processes, and decreased cell cycle pathways/signaling. To develop a protein-signature with high sensitivity/specificity, we calculated a score based on the number of proteins that were above the specified threshold of 2 times the value of the standard deviation. The top 10 proteins that were above this threshold in most of the PCa patients were included in the signature. We achieved a sensitivity of 81, 82, 82% with a specificity of 83,81,82% in D43, D94, and D85, respectively. In conclusion, uEVs are a great source for accurate PCa detection. This study was financially supported by the IMMPROVE Alpe d"HuZes grant of the Dutch Cancer Society (EMCR2015-8022), and Cancer Center Amsterdam and the Netherlands Organization for Scientific Research (NWO Middelgroot, #91116017) are acknowledged for support of the mass spectrometry infrastructure. 436

437

P-I-0286 Comprehensive in vivo multiomics analysis of Alzheimer's disease pathogenesis in plasma Sanae Benabou1, Paolo Bonini1, Sajjan Singh Mehta1 1Olobion, SL, Barcelona, Spain Alzheimer’s disease (AD), affecting over 24 million people worldwide, remains the leading cause of dementia, underscoring the critical need for new therapeutic targets. Conventional research has predominantly focused on the amyloid cascade hypothesis, which attributes AD pathogenesis to the aggregation of amyloid-β protein1. Yet, this framework inadequately captures the disease's multifaceted symptomology. The presence of neurofibrillary tangles, arising from the hyperphosphorylation and aggregation of tau protein, signals the need for a more expansive research approach. This study advances our understanding by integrating high-throughput mass spectrometry-based proteomics, metabolomics and lipidomics, offering a comprehensive examination of AD's lipidome, proteome, and metabolome. Analyzing two distinct genotype groups alongside their controls, utilizing mouse plasma samples, the study employed a dual-phase extraction method for metabolomics and lipidomics, the PreOmics ENRICH-iST sample preparation kit for proteomics, and the pioneering Zeno SWATH MS. This data-independent acquisition (DIA) technique, augmented by Zeno trap pulsing, enhances sensitivity, with proteomic data processed via DiaNN software and metabolomics and lipidomic data via oloMAP 2.02. Our results highlight the Zeno SWATH MS's ability to significantly improve protein identification at both micro and analytical flow-rate chromatographies, even with minimal sample volumes. The analysis delineates clear multiomics distinctions between the groups, notably between genotypes and their controls. Furthermore, the synergy of DDA with SWATH-DIA methodology markedly increases double annotation coverage in lipidomics. The identification of differentially expressed proteins, metabolites and lipids provides new insights into AD's molecular mechanisms, moving beyond the amyloid-centric paradigm. These findings emphasize the potential of uncovering novel AD therapeutic targets in preclinical drug development settings. References 1. Herrup, K. The case for rejecting the amyloid cascade hypothesis. Nat. Neurosci. 18, 794–799 (2015). 2. Tsugawa, H. et al. A lipidome atlas in MS-DIAL 4. Nat. Biotechnol. 38, 1159–1163 (2020). 437

438

P-I-0287 Enrichment strategies for fast and sensitive analysis of cerebrospinal fluid Eva Borràs1,2, Federica Anastasi3,4, Olga Pastor1,2, Marc Suárez-Calvet3,4, Eduard Sabidó1,2 1University Pompeu Fabra, Barcelona, Spain 2Center for Genomics Regulation, Barcelona, Spain 3BarcelonaBeta Brain Research Center, Barcelona, Spain 4Hospital del Mar Medical Research Institute, Barcelona, Spain Cerebrospinal fluid (CSF) has been proven a useful matrix for neurological diseases research, not only to identify biomarkers for diagnostic purposes, treatment prediction, or treatment response, but also to understand the molecular mechanisms involved in disease progression. However, CSF proteome profiling by mass spectrometry faces two main analytical challenges. Firstly, the broad dynamic range of all liquid biopsies, which hinders the sensitivity of protein measurements. Secondly, the need to analyze hundreds of human samples in a reproducible and timely manner. Indeed, there has been a trade-off among sensitivity, number of analytes, number of samples, sample amount, and data quality. Recently developed mass spectrometry instrumentation allows for the rapid acquisition of samples, aligning with the need of large patient cohorts essential for translational clinical projects. On the other hand, several commercial solutions have appeared to address the limitations associated with processing of liquid biopsies improving in-depth protein identification. One such solution is the ENRICH-iST kit by PreOmics, originally designed to diminish the dynamic range in plasma samples by enriching low-abundance proteins onto paramagnetic beads. In this study, we adapted the ENRICH-iST kit protocol to process cerebrospinal fluid. We assessed several amounts of CSF starting material, and evaluated the reproducibility and the sensitivity of this enrichment strategy by means of the coefficient of variation and the number of identified proteins and peptides detected. A pool of healthy human CSF sample was processed in triplicates and different amounts with either the iST-BCT or ENRICH-iST PreOmics kits. Digested samples were analyzed by liquid chromatography coupled to mass spectrometry using data-independent acquisition and short gradients across three instruments: an Orbitrap Eclipse Tribrid, an Orbitrap Astral, and a timsTOF HF. Chromatography settings and data-acquisition schema were tailored for each platform. Our findings highlight the suitability of the ENRICH-iST sample preparation kit for processing cerebrospinal fluid. The kit significantly increases both proteome coverage and depth in cerebrospinal fluid, while maintaining technical reproducibility and low coefficients of variation. Therefore, our work shows how using such sample preparation kits, like ENRICH-iST, ensures accurate and high-quality profiling of CSF samples. In summary, we show how new developments in sample preparation together with last generation mass spectrometry instrumentation allow for fast and sensitive analysis of CSF samples required for large-scope neurological disease studies. 438

439

P-I-0288 Machine learning identifies clinical sepsis phenotypes that translate to the plasma proteome level Thilo Bracht1,2, Malte Bayer1,2, Björn Koos1, Karin Schork2, Martin Eisenacher2, Tim Rahmel1, Matthias Unterberg1, Michael Adamzik1, Barbara Sitek1,2, Hartmuth Nowak1,3 1University Hospital Knappschaftskrankenhaus Bochum, Department of Anesthesiology, Intensive Care Medicine and Pain Therapy, Bochum, Germany 2Ruhr University Bochum, Medizinisches Proteom-Center, Bochum, Germany 3University Hospital Knappschaftskrankenhaus Bochum, Center for Artificial Intelligence, Medical Informatics and Data Science, Bochum, Germany Background: In Germany, around 280,000 patients develop sepsis every year and more than 150 patients die every day. Currently, effective therapeutic measures are limited to the treatment of the underlying infection and supportive measures. Specific treatment options that directly address the molecular changes in sepsis have not yet been identified. As sepsis manifests itself in individual courses, a generalised therapeutic approach is ruled out. With the aim of a future individualised sepsis therapy, we used machine learning (ML) to identify clinical sepsis phenotypes in a multicentre, prospective cohort and characterised them using plasma proteomics. Methods: Within the Sepsis- and CovidDataNet.NRW studies, routine clinical data from the patient data management systems was collected from critically ill sepsis patients. From this, clinical variables were selected from 384 patients on the day of sepsis diagnosis and a Principal Component Analysis (PCA) was carried out for dimension reduction. The sepsis phenotypes were identified by cluster analysis using the k-means algorithm on the PCA results. In addition, plasma samples from 276 patients were analysed by mass spectrometry-based plasma proteomics using an Ultimate3000 uHPLC coupled to an Orbitrap Exploris 240. Data was acquired in DIA mode and statistically analysed and interpreted in the context of the identified clinical phenotypes. Results: Three clinical sepsis phenotypes were identified and termed A-C. Cluster C (n=25) was characterised by the highest disease severity, a mortality rate of 92% and multi-organ failure with leading liver failure. Cluster B (n=156), with a mortality rate of 45%, was also characterised by relevant organ failure, with kidney injury being particularly striking in comparison to cluster A (n=202). At the level of the plasma proteome, 613 proteins were quantified and massive differences between cluster C and clusters A and B were apparent. These mainly reflected the organ failure of patients in cluster C, and prominently showed the excessive consumption of acute phase proteins, coagulation factors and proteins of the complement system. The difference between clusters A and B was less pronounced overall, but reflected increased disease severity as well as kidney failure in cluster B. The consumption of complement and coagulation factor was already apparent in cluster B in comparison to the mildest cluster A. Conclusions: The clinical phenotypes identified by an ML approach showed different severities of sepsis and could be translated to the plasma proteome level. The differential proteins allowed a deeper insight into the molecular processes of sepsis and open up the possibility of characterising the identified phenotypes in more detail on the molecular level. This approach may pave the way for future targeted therapeutic approaches to sepsis. 439

441

P-I-0289 High sensitivity analysis of tau phospho, non-phospho, and MTBR peptides in Alzheimer's disease CSF and plasma using Orbitrap Astral Gunnar Brinkmalm1, Johan Gobom1, Laia Montoliu-Gaya1, Nicolaie Eugen Damoc2, Tabiwang Arrey2, Bernard Delanghe2, Kaj Blennow1, Henrik Zetterberg1, Joanna Kirkpatrick3 1University of Gothenburg, Neuroscience and physiology, Moelndal, Sweden 2Thermo Fisher Scientific, Bremen, Germany 3Thermo Fisher Scientific, Hemel Hempstead, United Kingdom Introduction: Recently active treatments against Alzheimer’s disease (AD) have been made avaliable, creating an even greater need for biomarkers to diagnose and monitor the treatment of patients. Different phosphorylated forms of tau (pTau) have emerged as the most promising biomarkers for diagnosis and staging of AD. Their low plasma concentrations have so far required millilitres for mass spectrometry (MS) measurement, severely restricting the applicability of MS in clinical studies. In addition, there is an interest to measure the parts of tau that constitutes the core of the tangles that are formed in the brain, the microtubule binding region (MTBR). The tau fragments containing the MTBR are present at very low concentrations in cerebrospinal fluid (CSF) and plasma. Here, we evaluated the recently introduced Orbitrap Astral mass spectrometer for quantification of CSF and plasma tau in the context of Alzheimer’s disease. Methods: Both paired patient CSF and plasma samples submitted to the routine lab at the Sahlgrenska University Hospital/Mölndal [CSF biomarker (total-tau, pTau181, amyloid β1-42) defined AD and non-AD controls] cohort of plasma samples were analysed on an Orbitrap Astral (Thermo Fisher Scientific). Analysis was performed using parallel reaction monitoring (PRM) to measure tryptic tau peptides. Six phospho and eight non- phospho peptides (of which six were isoform specific reflecting contributions from both the central and peripheral nervous systems) were monitored: 45-68 0N, 131-138 PNS, 275-291 PNS, 175-190-p181, 195-209, 195-209-p199, 195-209-p 202, 195-209-p 205, 212-221, 212-221-p217, 225-240-p231, 275-280 4R, 275-286 3R, 282-290 4R. Pooled samples were used for method validation (determination of minimal sample volume, LLOQ, precision). Preliminary data: All 14 tau peptides in the assay were detectable both in CSF and plasma. Tau175-190-p181 could be measured down to 50 amol on-column, corresponding to 2 µL CSF and 212-221-p217 could be measured in the same CSF sample volume down to 10 amol on-column (CV=1.6%). In previous studies using Orbitrap Fusion Lumos, 250 µL CSF was used for analysis. For plasma, the analyses were performed using sample volumes of 250 µL, i.e., a quarter of the 1-mL-volume required previously. The high detection sensitivity of the Orbitrap Astral enabled detection of pTau forms in significantly lower CSF and plasma sample volumes compared with previously published studies, and with low technical variation. Moreover, because of the high sampling rate, the degree of multiplexing could be increased without compromising the analytical quality. Novel Aspect: Orbitrap Astral enabled detection by MS of pTau and MTBR biomarkers in significantly lower volumes than previously possible and allowed for a higher degree of multiplexing, with high technical precision. 441

442

P-I-0290 Targeted proteomics of synapse-associated pathological processes in neurodegenerative diseaeses Ann Brinkmalm1, Johanna Nilsson1, Elena Camporesi1, Juan Lantero Rodriguez1, Tammaryn Lashley2, Henrik Zetterberg1, Kaj Blennow1, Gunnar Brinkmalm1 1University of Gothenburg, Institute of Neuroscience and Physiology, Moelndal, Sweden 2University College London, Department of Neurodegenerative Disease, Queen Square Institute of Neurology, London, United Kingdom Aims: We have recently developed CSF biomarker panels targeting synaptic dysfunction as well as pathological alterations in the endo-lysosomal and ubiquitin-proteasome systems. Several of the proteins that we have found to be significantly altered in CSF in early stages of AD or PD are synapse-associated membrane-proteins and in some cases, we have identified specific processed protein forms that seem to be disease related. Our aim in this study is to develop a selected reaction monitoring (SRM) mass spectrometry-based panel to characterize and quantify specific forms of synapse associated proteins in brain for possible translation to fluid biomarkers. Methods: We have recently developed CSF biomarker panels targeting synaptic dysfunction as well as pathological alterations in the endo-lysosomal and ubiquitin-proteasome systems. Several of the proteins that we have found to be significantly altered in CSF in early stages of AD or PD are synapse-associated membrane- proteins and in some cases, we have identified specific processed protein forms that seem to be disease related. Our aim in this study is to develop a selected reaction monitoring (SRM) mass spectrometry-based panel to characterize and quantify specific forms of synapse associated proteins in brain for possible translation to fluid biomarkers. Results: Solid-phase extraction and parallel reaction monitoring mass spectrometry were used to prepare TBS soluble brain homogenate fractions and almost 2000 proteins were liquid chromatography and high-resolution mass spectrometry. High flow liquid chromatography and selected reaction monitoring were used to develop a panel including SNAP-25, syntaxin-1A and B, VAMP2, complexin-1 and 2, SV2A, synaptophysin, and synapsin-1. identified with low flow Conclusions: Differential patterns of CSF biomarkers associated to the synapse is probably due to differences in pathology mechanisms. The presented method could be used to compare the biomarkers" diagnostic and disease monitoring potential as well as to investigate specific pathological molecular patterns across and within neurodegenerative diseases. 442

443

P-I-0291 Characterizing inflammatory cytokines in plasma, tracheal aspirate, and urine samples of children hospitalized with tracheostomy-associated respiratory infections utilizing Olink technology Steven Bruzek1, Vera Ignjatovic1,2, Neil Goldenberg1,2,3, John Morrison4,5 1Johns Hopkins All Children's Hospital, Institute for Clinical and Translational Research, Saint Petersburg, FL, United States 2Johns Hopkins University, Department of Medicine, Baltimore, MD, United States 3Johns Hopkins University, Department of Pediatrics, Baltimore, MD, United States 4Johns Hopkins All Children's Hospital, Division of Pediatric Hospital Medicine, Saint Petersburg, FL, United States 5Johns Hopkins University School of Medicine, Department of Pediatrics, Baltimore, MD, United States Background: Children with tracheostomy represent a complex population that increasingly requires frequent hospitalization after receiving inpatient care for tracheostomy-associated respiratory infections (TARI). While certain conditions may contribute to the risk of readmission, it is currently unknown if specific factors such as inflammatory markers could define pathophysiologic mechanisms associated with recurrent hospitalization for TARI. Aims: To characterize cytokine signaling and inflammatory signatures across multiple specimen types in children hospitalized for TARI. Methods: We performed a prospective proof of concept study in children <17 years of age hospitalized for TARI. Blood, tracheal aspirate, and urine samples were collected within 48 hours of hospital admission and 24 hours before discharge. Platelet-poor plasma, tracheal aspirate, and urine were processed according to Olink® recommendations. The Olink Target 48 Cytokine Panel targeting 45 protein biomarkers of key pathways related to cytokine signaling and inflammatory processes was used. Olink normalized protein expression (NPX) values were uploaded into the Mass Dynamics platform for data interpretation. Principal component analysis (PCA) was employed separately for each sample type. Results: Forty samples from eight subjects were analyzed, consisting of 11 plasma, 24 tracheal aspirate, and five urine samples. PCA findings are shown in Fig 1A. Compared to plasma, tracheal aspirate specimens had a significantly higher abundance of IL-33, EGF, IL-1β, IL-6, CXCL8, IL-7, and OSM and a significantly lower abundance of LTA, CXCL12, FLT3LG, CCL11, IL-27, TSLP, CCL13, CCL19, and IL-17C (Fig. 1B). Compared to urine, tracheal aspirate specimens had a significantly higher abundance for 21 targets, with IL-1β, CSF3, CSCL8, and OSM comprising those with the largest difference (Fig. 1C). Compared to urine, plasma had a significantly higher abundance for 29 targets, with CCL19, MMP1, MMP12, CXCL11, and CSF3 comprising the largest difference, and a significantly lower abundance of EGF and OLR1 (Fig. 1D). Abundance of TNF, IFN-γ, IL-10, CCL8, and IL-17C in tracheal aspirate was decreased (Fig. 1E), and of TGFA, IL-7, CCL4, FLT3LG, and CCL11 in plasma was increased (Fig. 1F) for discharge vs. admission samples, although these differences were not statistically significant. Conclusions: Proinflammatory cytokines were detected in plasma, tracheal aspirate, and urine specimens of children hospitalized with TARI, with the highest abundance observed in tracheal aspirate and plasma. Tracheal aspirate specimens showed a potentially meaningful decrease and plasma specimens showed an increase in inflammatory signatures from admission to discharge. Future research will utilize marker levels in risk prediction models for hospital readmission in pediatric TARI patients. 443

445

P-I-0292 Plasma proteomic biomarkers of respiratory failure severity and mechanical ventilation duration among critically ill children Steven Bruzek1, Chan-Hyun Na2, Seongmin Ahn2, Sungtaek Oh2, Vera Ignjatovic1,3, Neil Goldenberg1,3,4, Anthony Sochet1,5 1Johns Hopkins All Children's Hospital, Institute for Clinical and Translational Research, Saint Petersburg, FL, United States 2Johns Hopkins University School of Medicine, Institute for Cell Engineering, Baltimore, MD, United States 3Johns Hopkins University, Department of Medicine, Baltimore, MD, United States 4Johns Hopkins University, Department of Pediatrics, Baltimore, MD, United States 5Johns Hopkins University School of Medicine, Department of Anesthesiology and Critical Care Medicine, Baltimore, MD, United States Background: Pediatric acute respiratory failure is associated with elevated mortality and morbidity. Prognostic indicators of poor clinical trajectory are limited to invasive mechanical ventilation (IMV) parameters such as oxygenation index (OI). Aims: Characterize the plasma proteome of children undergoing IMV via discovery proteomics and assess candidate markers for extended IMV duration and elevated OI. Methods: We performed a single-center, observational cohort study of critically ill children <18 years of age undergoing IMV for acute respiratory failure from 2018 to 2020. Blood was collected within 24- and 48-72 hours following endotracheal intubation in sodium citrate tubes and processed to platelet-poor plasma. Tandem Mass Tag (TMT)-labeled liquid chromatography mass spectrometry (LC-MS) was conducted. Protein abundance data analyses included normalization, log2-transformation, removal of proteins with >30% missing values, z-scoring, imputation of missing values from normal distribution, and normalization to remove batch effects. The Mass Dynamics platform was utilized for proteomic analyses. Cohorts were defined and compared by IMV duration (long duration [upper quartile, >16 days] versus short duration [lower quartile, <4 days]) and ARDS severity (high [OI>16] versus mild [OI<8]) to identify proteins of interest (POIs). Results: Eighty-one samples from 48 subjects were analyzed, with 759 total proteins identified. Compared to those with a short IMV duration, those with a long IMV duration had the following POIs noted <24 hours of IMV: YWHAZ, TKT, ALDOA, SERPINB1, TALDO1, and RNPEP with higher abundance and CDH13, ACAN, MMRN2, ICAM2, CTSA, IGLV1-47, and IGHV3-43 with lower abundance (Fig. 1A,B). Of those, SERPINB1, TALDO1, and ICAM2 were differentially abundant at both <24- and 48-72 hours of IMV. Compared to those with mild ARDS, those with severe ARDS had the following POIs noted <24 hours of IMV: TKT, LTA4H, ICAM1, PIGR, RNH1, INHBE, HSPA5, CD163 with higher abundance and KIT, TFRC, KRT77, SELENOP, APOM, and TTR with lower abundance (Fig. 1D,E). Of those, LTA4H, ICAM1, PIGR, and KIT were differentially abundant at both <24- and 48-72 hours of IMV. Univariate logistic regression analysis (yielding odds ratios with 95% confidence intervals and p-values for severe ARDS (OI>16) and long duration of IMV (>16 days), respectively) identified SERPINB1 (5.61 (1.05-29.85), p = 0.043; 7.0 (1.28-38.28), p = 0.025), TALDO1 (2.73 (1.04-7.16), p = 0.04; 3.48 (1.18-10.24), p = 0.023), TKT (3.61 (1.37-9.49), p = 0.009; 2.97 (1.19-7.37), p = 0.019), RNH1 (6.95 (1.5-32.11), p = 0.013; 3.96 (1.08-14.5), p = 0.038), and HSPA5 (9.48 (1.43-62.79), p = 0.02; 11.42 (1.59-81.72), p = 0.015) as proteins of interest with differential abundance at 24 hours (Fig. 1C,F). Conclusions: Using a TMT-LC-MS approach, we identified 5 POIs that after further validation may represent putative prognostic indicators of ARDS severity and poor clinical trajectory. 445

447

P-I-0293 Development and evaluation of a novel biomarker-based risk stratification platform for obesity-related complications using targeted proteomics Rachel Byrne1, Aleksandra Dudzik1, Stephen Pennington1, Fiona McGillicuddy1 1University College Dublin, Conway Institute, Medicine, Dublin, Ireland Introduction: Targeted mass spectrometry (MS) can be used for sensitive quantification of specific proteins in complex biological samples. By selectively monitoring predefined peptide ions, targeted MS enables accurate measurement of protein abundance, aiding biomarker discovery and development. However, it is crucial to analytically validate targeted MS assays to ensure their reproducibility. Analytical validation includes assessment of the stability of the assay, ensuring that the results are consistent over time. Stability is essential for the practical application of an assay. We aim to use targeted MS to measure proteins associated with high-density lipoprotein particles (HDL-P), to develop and validate a biomarker-based risk stratification platform for assessing metabolic health in individuals with obesity. Methods: A high-throughput method for HDL isolation has been developed. Proteotypic peptide assays for proteins of interest, related to metabolic health in people with or without obesity, were developed. The protein panel for multiple reaction monitoring (MRM) assays was selected based on literature searches, and in-house discovery proteomics experiments. MRM peptide analysis was performed using an Agilent 6490 triple- quadrupole MS with Jet-Stream electrospray source coupled to a 1290 Quaternary Pump HPLC system. The number of transitions used for assay development included all b and y ions, with the final number of transitions per peptide reduced to 3-4 based on characteristics such as intensity and peak shape. Peptide retention times (RT) and optimized collision energies were investigated using Skyline software (Version 23.1; MacCoss Lab) to establish a dynamic MRM-scheduled method based on input parameters of 600ms cycle times and 2-minute RT windows. Results: An MRM method has been developed to measure 82 proteins using 196 peptide MRM assays, represented by 596 transitions with a gradient time of 36 minutes. The reproducibility of multiple sample injections of the same HDL-P sample resulted in an overall coefficient of variation (CV) of less than 10%. The peptides associated with ApoA-1, the most abundant HDL-associated protein, had a CV <20%. Investigating the HDL-P associated peptides measured over a time course of 5 days showed minimal differences, indicating that HDL associated peptides are stable for at least 120hrs. Conclusions: The findings demonstrate that MRM can measure HDL-P associated proteins with high reproducibility, and show stability for up to 5 days. Upon completion of additional analytical validation experiments (within-day variability; and intra/inter operator), samples from patients will be analysed to delineate people with obesity high-risk, who are at high-risk of developing life-limiting complications from those who are low-risk and who can maintain their metabolic health despite excessive adiposity. 447

448

P-I-0294 A carrier-based quantitative proteomics method applied to biomarker discovery in pericardial fluid Amanda Jessica Campbell1,2, Samir Cakar1,2, Nicolai Bjødstrup Palstrøm1,2,3, Lars Peter Riber4,3, Lars Melholt Rasmussen1,2,5,3, Hans Christian Beck1,2,5,3 1Odense University Hospital, Department of Clinical Biochemistry, Odense, Denmark 2Odense University Hospital, Center for Clinical Proteomics (CPP), Odense, Denmark 3University of Southern Denmark, Department of Clinical Research, Faculty of Health Science, Odense, Denmark 4Odense University Hospital, Department of Cardiac, Thoracic and Vascular Surgery, Odense, Denmark 5Odense University Hospital, Center for Individualized Medicine in Arterial Diseases (CIMA), Odense, Denmark Data-dependent liquid chromatography tandem mass spectrometry (LC-MS/MS) is challenged by the large concentration range of proteins in plasma and related fluids. We adapted the SCoPE method from single-cell proteomics to pericardial fluid, where a myocardial tissue carrier was used to aid protein quantification. The carrier proteome and patient samples were labeled with distinct isobaric labels, which allowed separate quantification. Undepleted pericardial fluid from patients with type 2 diabetes mellitus and/or heart failure undergoing heart surgery was analyzed with either a traditional LC-MS/MS method or with the carrier proteome. In total, 1398 proteins were quantified with a carrier, compared to 265 without, and a higher proportion of these proteins were of myocardial origin. The number of differentially expressed proteins also increased nearly four-fold. For patients with both heart failure and type 2 diabetes mellitus, pathway analysis of upregulated proteins demonstrated enrichment of immune activation, blood coagulation, and stress pathways. Overall, our work demonstrates the applicability of a carrier for enhanced protein quantification in challenging biological matrices such as pericardial fluid, with potential applications for biomarker discovery. Mass spectrometry data are available via ProteomeXchange with identifier PXD052067. Fig. 1 (optional) 448

449

P-I-0295 Towards skin cohort proteomics studies: A fine-tuned MS-based workflow applied to Pachyonychia congenita Sara Ceccacci1,2, Lucile Marchal2, Hélène Ragot2, Janice Schwartz3, Alain Hovnanian2, Ida Chiara Guerrera4 1Necker Proteomics, Université Paris Cité, Structure Fédérative de Recherche Necker, INSERM US24/CNRS UAR3633, Paris, France 2Université Paris Cité, Inserm, UMR 1163, Institut Imagine, Laboratory of genetic skin diseases, F-75015, Paris, France 3Pachyonychia Congenita Project, Salt Lake City, UT, United States 4Necker Proteomics, Université Paris Cité, Structure Fédérative de Recherche Necker, INSERM US24/CNRS UAR3633, Paris, France Pachyonychia congenita (PC) is a rare skin disorder caused by heterozygous mutations in keratin genes (KRT6A, KRT6B, KRT6C, KRT16 or KRT17). The most severe clinical feature is the intense pain associated with palmoplantar keratoderma. To date, there are no satisfactory therapeutic options. An unbiased mass spectrometry (MS)-based proteomics analysis of patient skin biopsies could help to elucidate the elusive pathogenic mechanisms driving PC and to pinpoint new potential therapeutic targets. However, proteomic studies of skin samples present significant challenges. The most limiting step is protein extraction, as skin tissues are hard to homogenize due to the abundance of structural proteins like keratins and collagens. Furthermore, OCT-preserved skin samples also require the removal of OCT, which is incompatible with MS analysis, leading to longer extraction protocol and low protein yield. These issues are even more challenging in skin biopsies from patients with keratoderma. In our study, we optimized a high-throughput MS-based workflow for skin sample analysis. 5 mm snap-frozen skin tissues from healthy donors (controls) were divided by 4 to obtain the equivalent of 2 mm skin biopsies and then homogenized using BeatBox Tissue Kit 24x (PreOmics). 10µm slides from OCT tissues from the same controls and OCT tissues from three PC patients with KRT6A and KRT16 mutations (callus edges and adjacent non-lesional areas) were washed with PBS before BeatBox homogenization. Each sample was subjected to 96 well S-Trap plate (Protifi) digestion. 600ng of peptides were injected on an EvosepOne LC- timsTOF HT MS (Bruker) and eluted using a 30SDP method. MS data were acquired using diaPASEF and analyzed by DIA-NN v1.8.1. We obtained the highest skin tissue proteome coverage reported to date in a single-shot MS analysis, reducing dramatically the protein extraction time (24 samples in 1 hour). We identified an average of 6,800 proteins in the snap-frozen tissue samples and 5,200 proteins in the OCT tissues from the controls. Levels of outer epidermis (KRT1 and KRT14) and dermis markers (COL3A1) were similar in snap-frozen and OCT samples, confirming that the extraction was not biased. Despite the presence of keratoderma, OCT tissues from the three PC patients led to 6,800 PG. Of these, 196 proteins were deregulated in lesional versus non-lesional PC tissues (FDR=0.05), including over 30 candidates highlighted in a previous transcriptomics study. Enrichment analysis of the 85 upregulated proteins confirmed hyperkeratinization and revealed significant alterations in lipid biosynthesis pathways, which explain the clinical benefits of some of the treatments described in case reports. In conclusion, we propose a reproducible, miniaturized and in-depth proteomics workflow for analyzing large cohorts of both OCT-preserved and snap-frozen skin tissue samples. This enabled us to elucidate the pathways involved in the pathogenesis of Pachyonychia congenita. 449

450

P-I-0296 Proteomic profiling uncovers key proteins in ascites extracellular vesicles from Ovarian cancer patients: insight to tumor plasticity and prognosis Wei Dai1, Jianwei Zhou2, Ting Chen1 1Zhejiang University, Cancer Institute (Key Laboratory of Cancer Prevention and Intervention, China National Ministry of Education), The Second Affiliated Hospital, School of Medicine, Hangzhou, China 2Zhejiang University, Department of Gynecology, The Second Affiliated Hospital, School of Medicine, Hangzhou, China Over 75% of high-grade serous ovarian cancer (HGSOC) patients are found to have advanced disease at initial diagnosis, frequently presenting with peritoneal metastasis and ascites which are considered further promoting the metastasis of HGSOC. The proteins in fluid portion of ascites are mainly exsit in free style, small and large extracellular vesicles and the composition of them needs to be further explored for characterization. In this study, we quantitatively analyzed 5650 proteins in 39 samples from 13 patients with HGSOC using data- independent analysis (DIA) mass spectrometry, focusing on liquid ascites, small extracellular vesicles (sEVs) and large extracellular vesicles (lEVs). Our results revealed distinct proteomic profiles among the three categories and identified eight expression patterns through trend clustering analysis of the differential proteins. Through the integration of single-cell RNA-seq data of ascites cells, we discovered that proteins significantly enriched in sEVs were more frequently derived from fibroblasts rather than malignant cells, a pattern not observed in lEVs. Subsequently, we focused on a set of 242 proteins with expression trends of sEV>lEV>liquid ascites. To further analyze their correlation in metastasis and prognosis, we investigated plasticity-related genes in transcriptome datasets from GEO and MSigDB database, and performed prognostic analyses using data from GEO, EGA, and TCGA database, identifying four genes closely linked to tumor cell plasticity and patient survival. In conclusion, we comprehensively investigated the protein composition of HGSOC ascites, identifying proteins exhibiting enrichment in sEVs and are associated with tumor plasticity and prognosis. This offers a significant complementary to traditional investigations centered on patient ovarian cancer cells, deepening our understanding of the disease. 450

451

P-I-0297 Antibodyome analysis of schizophrenia by Escherichia coli proteome microarrays Chien-Sheng Chen1 1National Cheng Kung University, Food Safety, Tainan, Taiwan is Schizophrenia (SZ) influenced by genetic and environmental factors, and associated with chronic neuroinflammation. If the symptoms express after adolescence, environmental impacts are more substantial, and the disease is defined as adult-onset schizophrenia (AOS). Effects of environmental factors on antibody responses such as Escherichia coli (E. coli) to immunoglobulin G (IgG) and immunoglobulin M (IgM) might increase the severity of symptoms in SZ via the gut–brain axis. The purpose of this study is to reveal antibody profiles of SZ against bacterial protein antigens. We analyzed the IgG and IgM antibodies using E. coli proteome microarrays from 80 SZ patients and 40 healthy controls (HC). Using support vector machine to select panels of proteins differentiating between groups and conducted enrichment analysis for those proteins. We identified that the groL, pldA, yjjU, livG, and ftsE can classify IgGs in AOS vs HC achieved accuracy of 0.7. The protein yjjU, livG and ftsE can form the best combination panel to classify IgG in AOS vs HC with accuracy of 0.8. The enrichment results are highly related to ABC (ATP binding cassette) transporter in the protein domain and cellular component. We further found that the human ATP binding cassette subfamily b member 1 (ABCB1) autoantibody level in AOS is significantly higher than in HC. The findings suggest that AOS had different immunoglobulin production compared to early-onset schizophrenia (EOS) and HC. We also identified potential antibody biomarkers of AOS and found their antigens are enriched in ABC transporter related domains, including human ABCB1 protein. Fig. 1 (optional) 451

452

P-I-0298 Reducing inter-observer variability: harnessing molecular insights for clinical applications Garry Corthals1,2, Aleksandra Chojnacka1, Rianne Hofstraat1,2, Jesper Kers1,2,3 1University of Amsterdam, HIMS, Amsterdam, Netherlands 2Amsterdam University Medical Center, Pathology, Amsterdam, Netherlands 3Leiden University Medical Center, Leiden, Netherlands Pathological assessment of tissue samples is crucial for diagnosis and optimal treatment decisions. However, pathology often suffers from subjectivity, resulting in inter-observer variability, and has limitations in identifying subtle molecular changes. Proteomics offers a promising solution by providing both molecular evidence and unbiased classification, thereby increasing the quality and reliability of tissue assessments. For proteomics to be useful, the molecular information from formalin-fixed paraffin-embedded (FFPE) biopsy tissues must be consistent, quantitatively accurate, and rich in clinically relevant molecular information. Over the years, we have developed several methods for both fresh frozen (FF) and FFPE tissues, aiming to deliver results quickly (within 24 hours) and in line with current pathology practices. We will present a range of methods that work with tissues, albeit with varying success rates and identification numbers. Key factors guiding our choice of method include integration with current pathology procedures and the analytical repeatability on tens to hundreds of biopsies. Currently, we have achieved a throughput where, within the scope of 24 hours, the workflow reveals molecular differences and similarities of over 5000 proteins per biopsy, across hundreds of patient samples. The methods we present are based on data-independent acquisition mass spectrometry (diaMS), generating gigabytes of data per sample per time-point, with exceptionally high quantitative accuracy. The FFPE-based method we conclude with provides a coefficient of variation below 20%, analysing over 5000 proteins per sample in parallel. We will highlight findings from an ongoing large cohort study consisting of hundreds of transplant biopsies, revealing physical parameters such as the effects of size, time in the autosampler, and their impact on clinical validation. The findings presented here demonstrate the ease, speed, and robustness of the MS-based method. From minute tissue sections, a wealth of molecular data can assist and enhance pathology, potentially reducing inter- observer variability. Fig. 1 (optional) 452

453

P-I-0299 The in vivo measure and mathematical modeling of protein turnover in multiple biological compartments and across patient cohorts Sylvain Lehmann1,2,3, Pierre Giroux4,5,1, Jerome Vialaret1,3, Christophe Hirtz1,2,3,6, Jacques Colinge5,1,7 1University of Montpellier, Montpellier, France 2INM, Montpellier, France 3University Hospital (CHU), LBPC-PPC, Montpellier, France 4IRCM / University of Montpellier / ICM, Cancer bioinformatics and systems biology, Montpellier, France 5IRCM Inserm U1194, Cancer Bioinformatics and Systems Biology, Montpellier, France 6University Hospital (CHU), CMRR, Montpellier, France 7ICM, Montpellier, France Beyond the knowledge of protein abundance in various tissues, there is great interest in the turnover or dynamics of proteins. Protein turnover is the net rate at which proteins are produced or imported in a tissue, and simultaneously degraded or cleared. This complementary perspective to protein abundance is relevant in a number of applications. For example, abnormal turnover has been observed for specific proteins such as amyloid-β (Aβ), Tau, or sTREM2 in Alzheimer disease (AD) (Mawuenyega et al., 2010; Sato et al., 2018; Suárez- Calvet et al., 2016) or retinol-binding protein 4 (RBP4) in diabetes (Jourdan et al., 2009). Turnover data are typically acquired by mass spectrometry (MS) after introducing isotopic tracers to label the newly synthesized proteins. The ratio of labeled versus unlabeled protein peptide abundances is called the relative isotope abundance (RIA). Summarized at the protein level, the variation of RIA over time provides turnover information. We have developed both targeted and genome-wide methods (Lehmann et al., 2015 & 2019) based on the stable isotope labeling kinetics (SILK) protocol relying on 13C6-Leu intravenous injection (Bateman et al., 2006). Injection lasts for 9 hours and serial proteome acquisitions are performed from time 0 to 24 hours or longer. Here, applying concepts of population pharmacodynamics, we show how turnover data obtained from patient cerebrospinal fluid (CSF) and blood plasma can be modeled both at the cohort and individual levels. Data were obtained from 4 individuals in CSF and 7 in plasma (Fig. 1A). Mathematical modeling was based on differential equation coupled with Bayesian statistics. We discuss how such models can enable inter-cohort statistical tests to identify altered turnover. Then, applying notions of pharmacokinetics, we modeled genome-wide turnover data obtained from plasma and CSF simultaneously from a same individual. The new mathematical model entails the modeling of a hidden central nervous system (CNS) compartment to properly model passages from blood to CSF and vice versa (Fig. 1B). We show that the observed dynamics in plasma and CSF can be similar or rather different, and in certain configurations this strongly constrain the dynamics in the CNS, i.e. we can accurately infer an average CNS turnover. In other configurations, the CNS contribution seems less crucial and hence the CNS turnover is vaguely inferred only (Fig. 1C). Lastly, we will present unpublished turnover data of a cohort of Alzheimer patients modeled both at the cohort and 2-biological compartment levels. Figure 1. (A) Population and individual dynamics in plasma for 7 patients. (B) Barriers between CSF, plasma and CNS. (C) Simultaneous dynamics in CSF and plasma along with inferred CNS dynamics (k_ij = rates of passage, k_i = clearance rate, l_i = amplitude). 453

455

P-I-0301 Molecular basis of progressive familial intrahepatic cholestasis 3. A proteomics study Fernando Corrales1, Laura Guerrero1, Lorena Carmona-Rodríguez1, Fatima Milhano Santos1, Sergio Ciordia1, Luiz Stark2, Loreto Hierro2, Pablo Perez-Montero2, David Vicent2 1Centro Nacional de Biotecnología, CSIC, Macromolecular Structure, Madrid, Spain 2IdiPAZ, Hospital Universitario La Paz, Madrid, Spain Background and aims: Progressive familiar intrahepatic cholestasis type 3 (PFIC3) is a severe rare liver disease which affects between 1/50,000 to 1/100,000 children. In physiological conditions, bile is produced by the liver and stored in the gallbladder, then it flows to the small intestine to play its role in fat digestion. To prevent tissue damage, bile acids are kept into phospholipid micelles. Mutations in phosphatidyl choline transporter ABCB4 (MDR3) lead to intrahepatic accumulation of free bile acids that results in liver damage. PFIC3 onset occurs usually at early ages, progress rapidly and the prognosis is poor. Currently, besides the palliative use of ursodeoxycholate, the only available treatment for this disease is liver transplantation, which is really challenging for short-aged patients. Methods: To gain insight into the pathogenesis of PFIC3 we have performed an integrated proteomics and phosphoproteomics study in human liver samples to then validate the emerging functional hypotheses in a PFIC3 murine model of this disease. Results: We identified 6,246 protein groups, 324 proteins among them showing differential expression between control and PFIC3. The phosphoproteomic analysis allowed the identification of 5,090 phosphopeptides, from which 215 corresponding to 157 protein groups, were differentially phosphorylated in PFIC3, including MDR3. Regulation of essential cellular processes and structures, such as inflammation, metabolic reprograming, cytoskeleton and extracellular matrix remodeling and cell proliferation were identified as main drivers of the disease. Conclusion: Our results provide a strong molecular background that significantly contributes to a better understanding of PFIC3 and provides new concepts that might prove useful in the clinical management of patients. Fig. 1 (optional) 455

456

P-I-0302 A rapid and economical workflow for protein biomarker discovery using agilent 6495D triple quadrupole LC/MS system Linfeng Wu1, Daniel Cuthbertson2, Wolfgang Mayser3 1Agilent Technologies, Santa Clara, CA, United States 2Agilent Technologies, Denver, CO, United States 3Agilent Technologies Deutschland GmbH, Waldbronn, Germany The advancement of triple quadrupole LC/MS systems and the publicly available peptide MS/MS spectral libraries have enabled researchers to develop comprehensive MRM-based assays without performing extensive proteome identifications in the first place. In this study, we performed a proof-of-concept study in healthy human plasma samples using Uniprot human reference proteome, PeptideAtlas plasma spectral libraries, Skyline software and Agilent 6495D LC/TQ system. A comprehensive peptide quantitative assay was developed quickly based on publicly available libraries without using stable isotope-labeled standard peptides (SILs). The final optimized dynamic MRM method contain a 20- min linear LC gradient and covers 3604 MRM transitions presumptively matching to 390 plasma proteins and 846 peptides. This assay was applied to quantify plasma samples from 40 healthy human subjects (20 males and 20 females) for group comparison analysis. The results demonstrated that this rapid and economical workflow can discover protein biomarkers in complex cohort samples. 456

457

P-I-0303 Profiling the proteome response to tamoxifen Leo Dahl1, Marike Gabrielson2, Felix Grassmann2,3, Åsa Hedman2,4, Hampus Hagelin1, Kamila Czene2, Anders Mälarstig2,4, Per Hall2, Jochen Schwenk1 1KTH Royal Institute of Technology, Solna, Sweden 2Karolinska Institute, Solna, Sweden 3Health and Medical University, Potsdam, Germany 4Pfizer, Stockholm, Sweden Background: Breast cancer is the leading cause of cancer-related mortality in women world-wide. Research has been conducted to reduce breast cancer mortality mainly through better diagnostic tools, screening programs and targeted therapy. A measure to prevent the onset of the disease in high-risk women is to use selective estrogen receptor modulators, such as tamoxifen. The drug is, however, associated with many side effects, reducing life quality, adherence and uptake. We therefore seek to understand the effects of tamoxifen on a systemic level. Methods: We explored the response associated with tamoxifen use using 176 samples from the KARISMA phase II low-dose tamoxifen trial. The levels of > 2000 circulating proteins were measured using the Olink Explore 3k platform in age and BMI-matched women taking placebo (N=44) and 20 mg tamoxifen (N=44). The protein levels were measured at baseline and at 6 months in both treatment arms to study differential protein response by using a linear model adjusted for age and BMI at baseline. Results: Comparing the protein levels between treatment arms at baseline showed no proteins exhibiting significant differences (FDR > 0.05), and levels in the placebo group remained stable over 6 months. Contrasting the changes in protein levels between the treatment arms over the 6-month administration period revealed wide-spread changes in the proteome, summing to 193 (~10%) of the measured proteins showing statistically significant differences in protein level (FDR < 0.05). The protein response was spread across the different functional categories, showing a systemic effect associated with the intake of tamoxifen. Conclusions: This pilot study shows that intake of tamoxifen is linked to strong and wide-spread changes in the circulating proteome, warranting continued investigations into the functional pathways of treatment response. Future endeavours should focus on replication with more samples from different treatment arms, use of orthogonal technologies to confirm and expand our insights, study of metabolites of tamoxifen, and consider menopause status when interpreting the effects on breast cancer biology. 457

458

P-I-0304 Single cell and clinical proteomics profiling in non-small cell lung cancer (NSCLC) Hannah Demeke1,2,3, Corinna Friedrich1,2,3, Rebecca Fritz2,3, Liliana H. Mochmann4,3, Dražen Papić2,3, Oliver Popp1, Mohamed Haji1, Sylvia Niquet2,5, Janine Wolff2,5, Melissa Klingeberg6, Florian Mutschler7, Simon Schallenberg2,3, Fabian Coscia6, Frederick Klauschen2,4,5, Philipp Mertins1,5 1Max Delbrück Center for Molecular Medicine, Proteomics Platform, Berlin, Germany 2Charité – University Medicine Berlin, Pathology, Berlin, Germany 3BIFOLD – Berlin Institute for the Foundations of Learning and Data, Berlin, Germany 4Ludwig-Maximilians-Universität München (LMU), Pathology, Munich, Germany 5BIH – Berlin Institute of Health, Berlin, Germany 6Max Delbrück Center for Molecular Medicine, Spatial Proteomics, Berlin, Germany 7Max Delbrück Center for Molecular Medicine, Proteome Dynamic, Berlin, Germany Early detection of non-small cell lung cancer (NSCLC) is crucial for effective therapeutic intervention and improving the survival rate of patients. Herein we present our strategy and initial results on the comparison of lung cancer subtypes in clinical patient samples and lung cancer cell lines as part of the BIFOLD LungCAIRE project. Our aim is to use multiple proteomic approaches to study early-stage lung cancer, focusing on the identification of biomarkers and prognostic signatures within a diverse patient cohort including matched NSCLCpatient- derived plasma, fresh primary bulk tissue specimens, and single cells. The study highlights the latest developments in single-cell proteomics, from buffer comparisons to sorting techniques, and evaluates two MS measurement methods (DIA-PASEF and Slice-PASEF) for low-input samples. We conducted a comparative analysis of adenocarcinoma (ADC) and squamous cell carcinoma (SCC) subtypes using cell culture models, tissue samples and patient plasma. Our results reveal the differences between these subtypes, identifying distinct proteomic profiles correlating with disease state and co-regulation of oncogenic protein networks and candidate biomarkers. Our long-term goal is to improve the understanding of the mechanisms driving non-small cell lung cancer (NSCLC) by an innovative approach of profiling tumor-associated proteins in both bulk and single cell models, including plasma proteins from NSCLC-derived tumors. 458

459

P-I-0305 Comprehensive proteomics analysis of plasma from ApoE- and LDLR- deficient mice reveals atherosclerosis biomarkers with potential application in clinic Cristina Amparo Devesa Arbiol1, Inmaculada Jorge2,1, Estefanía Núñez2,1, Emilio Camafeita2,1, Almudena Ramiro1, David Sancho1, Jose Luis Martin Ventura3, Jesus Vázquez Cobos2,1 1Centro Nacional de Investigacions Cardiovasculares (CNCI), Madrid, Spain 2Centro de Investigación Biomédica en Red de Enfermedades Cardiovasculares (CIBERCV), Madrid, Spain 3Hospital Universitario Fundación Jiménez Díaz, Madrid, Spain Cardiovascular diseases originated by atherosclerosis are the main cause of death worldwide. The objective of this work is to study the mechanisms underlying atherosclerosis based on two atherosclerotic mouse models, as well as to find disease biomarkers with potential application in the clinical setting. A proteomics analysis has been carried out on two mouse models, ApoE KO (apolipoprotein E knockout) and LDLR KO (low-density lipoprotein receptor knockout), which were fed with different diets. Plasma protein digestion was followed by multiplexed stable isotope labelling of peptides, protein identification and protein function analysis. Results highlight that in both clinical atherosclerotic models functional categories such as lipid metabolism, complement system, immunoglobulin regulation, blood coagulation and extracellular matrix remodeling are altered. Comparison of the corresponding protein components between the two mouse models showed marked differences regarding complement activation components and blood coagulation regulators. ApoE KO model was further analyzed to assess differences between the clinical and the subclinical model and the effect of gut microbiota. This analysis revealed eight potential biomarkers to discriminate between clinical and subclinical atherosclerotic stages and seven pointer proteins that could understand the molecular mechanisms undergoing the impact of the gut microbiota metabolism on the atherosclerosis. Finally, this proteomics study has been compared with three human cohorts: individuals with clinical (carotid wall lesions, n=48) and subclinical (PESA (n=444) and AWHS (n=350) studies) atherosclerosis. It was observed that the ApoE KO plasma proteomic profile resembles more closely its human counterpart as compared to the LDLR KO model. Moreover, a link has been established between gut microbiota and atherosclerosis, and potential biomarkers have been proposed. 459

460

P-I-0306 Quantitative tissue analysis reveals AK2, COL1A1, & PLG protein signatures: targeted therapeutics for meningioma Shamjetsabam Nandibala Devi1 1SIR GANGA RAM HOSPITAL, DEPARTMENT OF BIOTECHNOLOGY AND RESEARCH, New Delhi, India Background: Meningioma is the most common primary intracranial brain tumor, accounting for one-third of all CNS tumors. It is a life-threatening tumor with a high recurrence rate, leading to an increasing healthcare burden globally. Current clinical approaches include surgical resection, chemotherapy, and radiation therapies, but they are not always effective. Targeted therapies and molecular markers offer better patient care compared to traditional methods. While there is proteomic data on meningioma, knowledge of potential biomarkers that differentiate between tumor grades is limited. It is essential to validate reported markers in large, independent sample cohorts in the future to identify the best set of biomarkers. Methods: A total of 12 samples were collected, with 3 samples for the control group (forming pool 1), 3 samples for Meningioma grade I (forming pool 2 and pool 3), and 3 samples for Meningioma grade II (forming pool 4). These samples were used for LC-MS/MS analysis. Following this, the expression of three proteins was assessed using immunocytochemistry, flow cytometry, and western blotting. Results: Three important proteins were found to be upregulated namely AK2, COL1A1, and PLG. The biomarker potential of these proteins was further checked by ICC, western blotting, and flow cytometry. All techniques presented higher expression of these proteins in grades as compared to control. Here, the trend obtained was control < MG grade I < MG grade II. Conclusions: This study aims to identify a suitable therapeutic marker for the early diagnosis and treatment of meningioma. The research labeling and detection, followed by flow cytometry, immunocytochemistry, and western blotting. The study also examines the effectiveness of specific proteins - AK2, COL1A1, and PLG - as potential targeted molecules. involves quantitative 460

461

P-I-0307 Proteome changes by P2X7 receptor stimulation of microvesicles and exosomes from human glioblastoma stem cells Vimal Di Virgilio1, Nicola Tupone1, Fabrizio Di Giuseppe1, Laura Pierdomenico1, Giuliano Ascani2, Patricia Giuliani1, Renata Ciccarelli1, Lucia Ricci Vitiani3, Stefania Angelucci1 1G.d'Annunzio University, Proteomics Lab at Center for advanced Studies and Technology_CAST, Pescara, Italy 2UOSD Maxillofacial surgery, Azienda Sanitaria Locale di Pescara, Pescara, Italy 3Istitto Superiore di Sanità, Rome, Italy Extracellular vesicles (EVs) are released from many tumors including glioblastoma multiforme (GBM), the most common and lethal brain tumor in adults characterized by high resistance to current therapies and poor patient prognosis. Since GSCs are deeply involved in GBM aggressiveness, recent studies have begun to investigate the compounds specifically transported by EVs secreted from these cells, which would be implicated in GBM malignancy. Given the high relevance of the information provided by cancer cell secretome, we performed a proteomic analysis of microvesicles (MVs) and exosomes (EXOs) secreted from GBM-derived stem cells (GSCs). These latter were obtained from surgical specimens of human GBM and expressed P2X7 receptors (P2X7R), which positively correlate with GBM growth and invasiveness.Here, we investigated the proteomic content of GSCs-derived MVs and EXOs following pharmacological stimulation of the ionotropic P2X7R with low affinity to ATP [10], based on the evidence that: i) extracellular ATP levels in glioma cells are in the high micromolar range due to the poor nucleotide metabolism [11]; ii) decreased ATP levels positively correlate with enhanced GBM growth [12]; iii) P2X7R sense high extracellular ATP levels and promote glioma progression [13]; iv) P2X7R are highly expressed in GSCs [14] and their stimulation increases cell aggressiveness by enhancing the expression of epithelial-mesenchymal transition (EMT)-related markers, cell migration/invasion and GSC survival.Our finding indicate that P2X7R stimulation of GSCs caused significant changes in the EV content, mostly ex-novo inducing or upregulating the expression of proteins related to cytoskeleton reorganization, cell motility/spreading, energy supply, protection against oxidative stress, chromatin remodeling and transcriptional regulation. Most of the induced/upregulated proteins has already been identified as GBM diagnostic/prognostic factors, while others have only been reported in peripheral tumors. Our findings indicate that P2X7R stimulation enhances the transport and, therefore, possible intercellular exchange of GBM aggressiveness-increasing proteins by GSC- derived EVs. Thus, P2X7R could be considered as a new druggable target of human GBM, although these data need to be confirmed in larger experimental sets. 461

462

P-I-0308 Proteomic analyses of the urothelial cancer landscape reveal highly distinct prognostic and predictive subtypes Franz Dressler1,2,3, Onur Dikmen1, Falk Diedrichs1, Deema Sabtan1, Sofie Hinrichs2, Christoph Krisp4, Paulina Mackedanz2, Mareile Schlotfeldt2, Martin Hennig5, Hartmut Schlüter4, Arkadiusz Miernik6, Sven Perner2, Philipp Wolf6, Roman Zubarev7, Ákos Végvári7 1Charité – Universitätsmedizin Berlin, Institute of Pathology, Berlin, Germany 2University Medical Center Schleswig-Holstein, Campus Lübeck, Institute of Pathology, Luebeck, Germany 3Berlin Institute of Health at Charité – Universitätsmedizin Berlin, Berlin, Germany 4Universitätsklinikum Hamburg-Eppendorf, Institut für Klinische Chemie und Labormedizin, AG Massenspektrometrie und Proteomics, Hamburg, Germany 5University Medical Center Schleswig-Holstein, Campus Lübeck, Department of Urology, Luebeck, Germany 6University Medical Center Freiburg, Department of Urology, Freiburg, Germany 7Karolinska Institutet, Division of Physiological Chemistry I, Department of Medical Biochemistry and Biophysics,, Stockholm, Sweden Questions: Urothelial cancer (UC) is a challenging disease with a wide tumor-biological spectrum. Most molecular classifications cover only muscle-invasive bladder cancer and are transcriptome-based, relating only indirectly to the therapeutically relevant and more stable protein level. Also, the advent of new targeted therapies requires quantitative data about the tumor specificity of their target proteins. We turned to the proteome to address these questions. Methods: We performed deep proteomic profiling of a comprehensive cohort by optimized tandem mass tag- labelled liquid chromatography-coupled tandem mass spectrometry. Data acquisition was validated internally with immunoblotting and externally by bioinformatic reclassification within existing, filtered transcriptomic data. After bioinformatics, the top cluster defining proteins were quantified immunohistochemically under real-world conditions. Protein profiles were individualized and separately evaluated with drug repurposing libraries. Cell viability assays were performed for a panel of twelve UC cell lines to validate these predictions in vitro. Results: We analyzed 434 samples with 242 tumors and 192 paired normal mucosae, covering all stages of UC from pTa to >pT2. 9542 proteins were quantified and revealed five distinct proteomic subtypes. These were validated internally and externally, showing relevant survival stratification also in the TCGA dataset. The proteomic subtypes were independent from pathological groups with relevant stratification of progression- free and overall survival (low vs. high-risk: median 103 vs. 27 months). Tumor specificity of all proteins was highly heterogeneous across stages and subtypes. As an example, the therapeutic target NECTIN4 was generally overexpressed only in non-muscle-invasive UC. Drug repurposing revealed several new candidate drugs, each specific to different proteomic subtypes. In vitro data showed increased sensitivity by subtype in line with four out of seven representative predictions. Conclusion: Proteomic subtypes add independent prognostic information and carry predictive value for several newly identified adjuvant drug candidates. The tumor specificity of biomarkers and drug targets is highly dependent on stage and subtype and calls for quantitative patient-specific testing. 462

463

P-I-0309 Targeted Proteomics goes clinical as primary endpoint: a case study Yvan Eb-Levadoux1, Catherine Pech1, Eloi Haudebourg1, Alberto Vezzelli2, François Autelitano1, Winfried Wunderlich3 1Evotec GmbH, Toulouse, France 2Evotec GmbH, Verona, Italy 3Evotec GmbH, Goettingen, Germany Introduction: While proteomics tools are increasingly used in the early steps of the drug development journey, its use remains limited to exploratory endpoints in clinical phase. Successful MS methods used in the exploratory and preclinical phases are adapted to antibody-based assays. For the present study, no antibody of the target engagement biomarker could be validated for animals in preclinical phase. An MS-based protein quantification method was therefore developed in Research Use Only (RUO) and ultimately transferred to a Good Clinical Practice (GCP) environment for validation and application to support preclinical development and clinical trial Phase 1 as a primary endpoint. Method: First, a high purity Stable Isotope Labelled (SIL) protein was produced in HEK293 cells cultured in SILAC medium and purified (anti-poly-histidine followed by IMAC and SEC). Second, a targeted MS assay was developed for the absolute quantification of biomarker of interest. Briefly 2 µL of plasma was processed on an iST kit (Preomics), peptides were injected using a microflow LC (M Class, Waters) on a C18-CSH column (Acquity, Waters) and quantified in MRM mode on a triple quadrupole (6500+, Sciex). Results: Acceptable protein purity (> 80 % based on MS intensity) and excellent heavy label incorporation (>98 %) was archived. A targeted proteomics assay was first developed in RUO and transferred to a GCP facility where the method was validated according to FDA validation of bioanalytical methods for Industry and ICH-M10 guidelines. Specifically, the method validation included the assessment of precision and accuracy of the assay, surrogate matrix equivalence, lower limit of detection (100 ng/ml), concentration range, short- and long-term stability of the samples. The validated, targeted MS method is routinely used to quantify a target engagement biomarker in a currently ongoing Phase 1 clinical trial. It is, to our knowledge, the first time a MS-based targeted proteomics is used to quantify proteins in clinical samples as primary endpoint. 463

464

P-I-0311 Integrating proteomics into diagnostic molecular pathology reports to support molecular tumor board decisions Matthias Fahrner1,2, Johanna Thiery1, Niko Pinter1, Elaine-Pashupati Dopfer1, Uta Matysiak1, Stepan Cysar1, Ulrike Wlokka1, Silke Laßmann1,3, Martin Werner1,2,3, Oliver Schilling1,2 1Institute for Surgical Pathology, Medical Center – University of Freiburg, Faculty of Medicine, University of Freiburg, Freiburg, Germany 2German Cancer Consortium (DKTK) and Cancer Research Center (DKFZ), Freiburg, Germany 3Comprehensive Cancer Center Freiburg, Medical Center, Freiburg, Germany Background and Aim: Patients suffering from particularly challenging tumor cases receive in-depth molecular diagnostics within the molecular tumor board (MTB). Based on the molecular findings the interdisciplinary MTB team collaboratively recommends the best therapeutic strategies for each patient. Current molecular diagnostics predominantly depend on genomic and transcriptomic methods. The continuous advancements in quantitative mass spectrometry (MS)-based proteomics enable robust and reliable proteome examination in diverse patient samples. Therefore, we set out to incorporate clinical proteomics into molecular diagnostics within the MTB, enhancing and broadening the current molecular diagnostic practices while offering additional biological insights. Methods: We apply automated MS-based quantitative proteomic workflows to perform reproducible and robust in-depth proteomics of formalin-fixed paraffin-embedded (FFPE) tissue from patients in the MTB. Our proteomic workflow is seamlessly integrated into the molecular pathology routine. We present and discuss molecular diagnostic results, including proteomics, for each patient during biweekly interdisciplinary meetings. Results: Our robust, reproducible, and automated sample preparation approach routinely identifies and quantifies more than 6000 proteins in patient-derived FFPE samples. We acquired protein profiles from over 300 MTB patients and in many cases, the proteomic analysis has contributed to the molecular pathology reports. In these cases, proteomics corroborated immunohistochemical staining results and yielded additional information on multiple antibody-drug conjugate targets and complex tumor markers such as the microsatellite stability status. We have established a phosphoproteomic protocol for FFPE samples yielding functional insights into pathway activity in selected MTB cases. Notable findings include identifying phospho-ERK signaling in a patient with a BRAF-KIAA1549 inhibition. More recently, we developed a proteogenomic workflow that enables the integration of patient-matched genomics and proteomics data, paving the way for more comprehensive personalized medicine. Our proteogenomic analysis provided additional insights by detecting the KRAS G12D peptide, supporting previously uncertain genomic results due to low tumor cell content. The complete data analysis pipeline was integrated into a local Galaxy instance, maximizing robustness and reproducibility by storing the raw data, analysis tools (including software versions), and analysis steps in designated histories. recommendation of MEK the fusion, providing evidence for Conclusion: We present the additional insights generated by integrating proteomics in the molecular profiling of particularly challenging tumor cases, such as rare and recurrent/resistant malignancies. Integrating proteomics into the MTB paves the way for further clinical proteomic applications complementing current routine molecular diagnostics. 464

465

P-I-0313 Proteomic characterisation of clear cell renal cell carcinoma in patients with Von Hippel-Lindau Syndrome Tobias Feilen1,2,3, Anna L. Koessinger1, Manuel Rogg1, Elke Neumann-Haefelin4, Athina Ganner4, Markus Grabbert5, Christoph Schell1, Oliver Schilling1,6 1Medical Center Freiburg, Germany, Institute of Surgical Pathology, Freiburg, Germany 2Spemann Graduate School of Biology and Medicine (SGBM), University of Freiburg, Freiburg, Germany 3Faculty of Biology, University of Freiburg, Freiburg, Germany 4Renal Division, Department of Medicine, Freiburg University Medical Center, Faculty of Medicine, University of Freiburg, Freiburg, Germany 5Department of Urology, Medical Center, Faculty of Medicine, University of Freiburg, Freiburg, Germany 6German Cancer Consortium (DKTK) and Cancer Research Center (DKFZ), Heidelberg, Germany Introduction: The Von Hippel-Lindau (VHL) syndrome describes a familial neoplastic condition, caused by germline mutations of the VHL tumor suppressor gene, resulting in a functional loss of the VHL protein (pVHL). pVHL, as part of a multiunit ubiquitin ligase complex, is involved in the proteasomal degradation of specific target proteins, notably hypoxia-inducible transcription factors (HIF-1α/2α). Impaired HIF degradation and the subsequent induction of a pseudo-hypoxic state predispose patients for the development of various tumors, including clear cell renal cell carcinoma (ccRCC). In addition to its role in tumorigenesis, dysregulated HIF signaling in VHL disease affects the extracellular matrix (ECM). We aim to elucidate the molecular alterations linked to hereditary VHL loss in ccRCC, with a focus on matrisomal proteins. Methods: The present study entails a comprehensive analysis of formalin-fixed paraffin-embedded (FFPE) patient samples (n = 134) from the Freiburg VHL register through LC-MS/MS. The bead-based single-pot, solid- phase-enhanced sample preparation (SP3) protocol combined with a semi-automated liquid handling platform were utilized for FFPE sample processing. The samples were measured on a timsTOF Flex mass spectrometer coupled to the Evosep One chromatography system. Our investigation involved the proteomic characterization of 54 tumor specimens, 45 pseudo-capsule (PC) specimens, and 35 samples of non-malignant adjacent tissue (NAT). Preliminary Data: In our study, we employed MS-based proteomics to analyze ccRCC patient samples, achieving a proteome coverage of up to 8500 proteins per sample. To corroborate our findings, we demonstrated the alignment of tumor and NAT samples with those of a Clinical Proteomics Tumor Analysis Consortium (CPTAC) dataset using a principal component analysis (PCA) after batch correction. Differential abundance analysis comparing tumor and NAT and subsequent gene ontology (GO) enrichment analysis unveiled an upregulation of proteins involved in anaerobic glucose metabolism within the tumor, alongside a downregulation of proteins associated with aerobic respiration. Moreover, we analyzed the PC proteome through comparison with both tumor and NAT and identified a predominance of upregulated proteins specifically associated with extracellular matrix (ECM) organization and fibrosis within the PC samples. These findings suggest the PC as a highly intriguing entity in the context of tumor microenvironment dynamics and ECM remodeling. Further, we employed a missingness statistics approach to identify proteins with a significantly higher missingness in either tumor, PC or NAT. In our analysis, we observed that chemokines (CCL14, CXCL12), semaphorins (SEMA3B, SEMA3C), Wnt signaling proteins (WNT9B, SFRP1), and collagens (COL4A3-A6) were significantly depleted in tumor samples. Conversely, angiogenic factors such as VEGF, ANGPTL2, and ANGPTL4 exhibited a lower presence in NAT. 465

466

P-I-0314 Impact of CFTR modulators on the blood proteome Kerstin Fentker1,2, Marieluise Kirchner1,3, Matthias Ziehm1,3, Sylvia Niquet1,3, Julia Duerr4,5, Jobst Röhmel4,5, Stephanie Thee4,5, Susanne Hämmerling6,7, Olaf Sommerburg6,7, Mirjam Stahl3,4,5, Simon Gräber3,4,5, Marcus Mall3,4,5, Philipp Mertins1,3 1Max-Delbrück-Center for Molecular Medicine, Proteomics Platform, Berlin, Germany 2Freie Universität Berlin, Berlin, Germany 3Berlin Institute of Health at Charité – Universitätsmedizin Berlin, Berlin, Germany 4Charité – Universitätsmedizin Berlin, Department of Pediatric Respiratory Medicine, Berlin, Germany 5German Center for Lung Research (DZL), Berlin, Germany 6University of Heidelberg, Department of Pediatrics, Heidelberg, Germany 7German Center for Lung Research (DZL), Translational Lung Research Center Heidelberg (TLRC), Heidelberg, Germany Cystic fibrosis (CF) is caused by a dysfunction in the cystic fibrosis transmembrane conductance regulator (CFTR) and affects multiple organs. It is characterized by mucus obstruction in the lungs and altered secretion in multiple organs including the pancreas, intestines, and liver. The novel CFTR modulator drug combination Elexacaftor/Tezacaftor/Ivacaftor (ELX/TEZ/IVA) has significantly improved symptoms for most patients with CF eligible for this therapy. Despite the notable benefits, it is crucial to note that modulator therapy does not constitute a cure for CF. This raises questions about the altered disease dynamics and ongoing clinical challenges in CF management. This study compared longitudinal systemic treatment responses in the blood proteomes of cystic fibrosis patients using mass spectrometry. From a clinical perspective, the double combination of Lumacaftor/Ivacaftor (LUM/IVA) is less effective than the new and improved ELX/TEZ/IVA therapy, as shown by improved lung function and CFTR function, for example measured by sweat chloride concentration. We could identify specific and common pharmacodynamic signatures for both CFTR modulator treatments, affecting systemic inflammation and metabolic processes. While both treatments induce alterations in the blood proteome that correlate with ion channel and lung functions, the effects of ELX/TEZ/IVA were more robust, with a larger magnitude across patients and directed towards healthy signatures. Additionally, we were able to compare the effects of ELX/TEZ/IVA treatment on the proteomes of lung-specific sputum samples with systemic effects in the blood. Among others, we observed counter-directional changes in the pulmonary surfactant-associated protein B, SFTPB. This development is a potential sign of lung repair that is absent in LUM/IVA treated patients. These findings indicate that ELX/TEZ/IVA not only inhibits further deterioration of lung function but also gradually improves the overall health status of patients. This systems medicine study provides a resource of pharmacodynamic drug proteome alterations for CFTR modulator treatments and contributes to our understanding of systemic and local protein regulation in CF. 466

467

P-I-0315 The proteogenomic subgroups of relapsed Osteosarcoma Joris Frenz1,2,3, Lianghao Mao1,2,3, Yongjie Wang1,2,3, Dibyendu Seal1,3, Noelle Jung1,2,3, Sirisha Muppa1,3, Michaela Nathrath4, Robert Autry5, Stefan Pfister6, Andreas von Deimling7, Andreas Kulozik1,2,3, Ashok Jayavelu1,2,3 1German Cancer Research Center (DKFZ), Proteomics and Cancer Cell Signalling, Heidelberg, Germany 2University Clinic Heidelberg, Pediatric Leukemia, Heidelberg, Germany 3Hopp Children's Cancer Center, Proteomics and Cancer Cell Signalling, Heidelberg, Germany 4Klinikum Kassel, Pädiatrische Hämatologie und Onkologie, Kassel, Germany 5Hopp Children's Cancer Center, Clinical Bioinformatics and Translational Genomics, Heidelberg, Germany 6Hopp Children's Cancer Center, Pediatric Neurooncology, Heidelberg, Germany 7University Clinic Heidelberg, Neuropathology, Heidelberg, Germany Osteosarcoma (OS) is one of the most common and a highly aggressive bone cancer in children. It typically occurs during childhood or adolescence and often develops in areas of bone growth. Standard front-line therapy achieves cure rate of 60-70% for non-metastatic OS and merely a 30% for metastatic OS. Unfortunately, the treatment of OS patients has not significantly improved in recent decades, in particular for relapsed OS, presenting an unmet clinical need. We performed advanced mass spectrometry (MS)-based proteomic profiling and functional investigations to identify novel drug targets. We proteome profiled a cohort of 154 samples comprising OS tissue samples from relapsed patients of the INFORM cohort, cell lines and healthy controls. Using our high-throughput proteomic platform, we quantified a total of >12,100 proteins by DIA-PASEF with exceptional reproducibility. In addition, we have performed RNAseq, WES, EPIC array and Phosphoproteome for most of these samples to allow a multiomics examination. Our analysis reveals that one of the proteomic clusters enriched for MYC targets, RNA processing and splicing in OS patients significantly associates with low overall survival, and displays a higher rate of TP53 mutation as well as MYC amplifications. A similar cluster could be identified in OS cell lines, indicating a so far undescribed OS subgroup and enabling functional validations. In line with the observation of lower overall survival in patients, cell lines falling into the respective subgroup show higher proliferation, underlining the increased aggressiveness of this subgroup. Further we found overexpression and hyperactivation of MYC is not associated with MYC amplification. Interestingly, no difference in phosphorylation status of the key regulating MYC Phosphosite S77 could be detected between MYC amplified and MYC non-amplified patients, strongly suggesting that MYC overexpression and hyperactivation is independent of the MYC amplification status. Importantly, several pan-spliceosomal inhibitors do not show differential activity between the OS subgroups, indicating existence of specific functional dependencies. We identified three splicing proteins to be upregulated in the respective patient and cell line subgroups that are more essential for the respective OS cell lines compared to cell lines from the other subgroup using the Depmap database. Additionally, overexpression of these three proteins correlates with low event-free survival in four publicly available datasets. Ongoing efforts in our lab aim at exploring the functional roles played by these proteins using CRISPR/Cas9 genome editing as well as knock-down and overexpression approaches. In summary, we present an unsupervised proteogenomic clustering of relapsed Osteosarcoma which reveals specific vulnerabilities in a subset of Osteosarcoma patients. These findings will help to further develop therapeutic options for patients showing bad clinical prognostic characteristics. 467

469

P-I-0316 Standardized, fully automated scalable plasma workflows enabled by the Evotip pure Dorte Bekker-Jensen1, Joel Vej-Nielsen1, Magnus Huusfeldt1, Stoyan Stoychev1, Maurine Fucito1, Michael Lund Nielsen1, Nicolai Bache1 1Evosep, Odense, Denmark Plasma is a rich source of potential biomarkers for monitoring disease onset and progression. The implementation of non-targeted plasma proteomics in disease diagnostics promises to improve both speed and accuracy, thereby contributing to improved patient prognosis. However, the lack of reproducible, robust, and high throughput proteomics workflows is a major limiting factor in the transition from biomarker discovery to verification and ultimately clinical application. Additionally, the large dynamic range of plasma proteins prevents the identification of many proteins with bottom-up proteomics in undepleted plasma. We have implemented automated end-to-end sample preparation methods, integrated with the Evotip Pure, for profiling of undepleted plasma and deep plasma proteomics. These advancements will significantly enhance scalability for large scale cohort analysis. For the undepleted plasma workflow, we combined on-robot reduction and alkylation with protein aggregation capture (PAC) based digestion. An approach for deeper profiling of a sub-proteome based on a MagNet enrichment step was added for deep plasma profiling to capture membrane-bound particles. Both methods were fully automated on a Biomek i5 liquid handling robot for processing 576 samples in parallel, incorporating all steps from neat plasma to peptides loaded on Evotips, ready-to-analyze. This is driven by efficient protocols based on hyper-porous magnetic beads directly followed by concentration of the digested peptides on Evotips to minimize losses otherwise associated with additional desalting steps. The standardized workflows can robustly quantify close to 1,500 protein groups from neat plasma, increasing to more than 5,000 protein groups from the MagNet workflow with a throughput of 100 samples analyzed per day using the Orbitrap Astral MS. 469

470

P-I-0317 The comparison and correlation of in vitro and in vivo UPEC proteomes Muhammad Naadir Ganief1 1University of Basel, Biozentrum, Infection Biology, Basel, Switzerland Urinary Tract Infections (UTIs), are one of the most common infections globally. To improve treatment, there is a need to understand the physiological state of bacteria involved in the infection. Escherichia coli accounts for a large proportion of UTIs, and in this context we aim to profile the proteomes of Escherichia coli (E. coli), isolated from the urine of patients with UTIs and compare and correlate those proteomes with proteomes from E. coli cultured in various conditions, to better understand the physiology of E. coli in vivo. By comparing and correlating in vitro and in vivo E. coli proteomes, we aim to understand critical features that describe in vitro and in vivo growth of E. coli in Urine. These include nutrient availability and host-immune properties. Here we present a workflow that allows the comparison and correlation of in vitro generated proteomes, with the aim of understanding in vivo E coli physiology, and benchmarking in vitro models. Our data suggests that the major processes that drive the differences between our current in vitro models and in vivo sample; are involved in amino-acid metabolism and carbon metabolism - including fatty acid metabolism, glucose metabolism and the glyoxylate cycle, environmental conditions such as low oxygen tension are also implicated. Furthermore, we also identify clusters of patient derived samples, that may be driven by the host state, such as the availability of glycolytic substrates, or host immune response. The workflow presented here shows a data-driven approach to generate hypotheses aimed at improving in vitro models to better mimic in vivo conditions.And shows particular value in iterative approaches in model design. It also allows the functional investigation of diverse clinical proteomes. Fig. 1 (optional) 470

471

P-I-0318 Urine proteomic analysis of patients with schizophrenia Chenyang Zhao1, Fuchun Zhou2, Yuhang Huan1, Jian Yang2, Youhe Gao1 1Beijing Normal University, Beijing, China 2Beijing Anding Hospital affiliated to Capital Medical University, Beijing, China We tried to explore the difference of urinary proteome between unmedicated schizophrenia patients and normal people through a small number of cases. The results showed that a total of 35 differential proteins were screened in the schizophrenia group compared with the healthy control group. Through random grouping evaluation, it has 91.4 % credibility. Fifteen of the differentially expressed proteins were reported to be related to schizophrenia mechanism, drug target or nervous system regulation. Among them, the aromatic amino acid decarboxylase related to the pathogenesis of schizophrenia can distinguish all 10 patients and 9 normal people with 100 % accuracy in this study, and the AUC value of 17 proteins is greater than or equal to 0.9. The biological pathways enriched by differentially expressed proteins include ephrin receptor signaling pathway, positive regulation of long-term potentiation ( LTP ), etc. This study shows that urine proteomics can reflect the difference between schizophrenia and healthy controls, and has the potential as a diagnostic marker. 471

472

P-I-0319 Urinary proteome profiling for children with autism using data-independent acquisition proteomics Wenshu Meng1, Yuhang Huan1, Youhe Gao1 1Beijing Normal University, Beijing, China Background: Autism is a complex neurodevelopmental disorder. Objective and reliable biomarkers are crucial for the clinical diagnosis of autism. Urine can accumulate early changes of the whole body and is a sensitive source for disease biomarkers. Methods: The data-independent acquisition (DIA) strategy was used to identify differential proteins in the urinary proteome between autistic and non-autistic children aged 3–7 years. Receiver operating characteristic (ROC) curves were developed to evaluate the diagnostic performance of differential proteins. Results: A total of 118 differential proteins were identified in the urine between autistic and non-autistic children, of which 18 proteins were reported to be related to autism. Randomized grouping statistical analysis indicated that 91.5% of the differential proteins were reliable. Functional analysis revealed that some differential proteins were associated with axonal guidance signaling, endocannabinoid developing neuron pathway, synaptic long-term depression, agrin interactions at neuromuscular junction, phosphatase and tensin homolog deleted on chromosome 10 (PTEN) signaling and synaptogenesis signaling pathway. The combination of cadherin-related family member 5 (CDHR5) and vacuolar protein sorting-associated protein 4B (VPS4B) showed the best discriminative performance between autistic and non-autistic children with an area under the curve (AUC) value of 0.987. Conclusions: The urinary proteome could distinguish between autistic children and non-autistic children. This study will provide a promising approach for future biomarker research of neuropsychiatric disorders. 472

473

P-I-0320 Abundant serum proteins potentially discriminate normal and GDM pregnancy: from peptidome to intact proteins Xi Gao1,2, Guixue Hou2, Liang Lin2, Siqi Liu2 1University of Chinese Academy of Sciences, College of Life Sciences, BeiJing, China 2BGI Genomics, Shenzhen, China Gestational diabetes mellitus (GDM) is a common pregnancy complication due to the onset of glucose intolerance, which presents a significant risk for both mother and offspring. However, GDM diagnosis at early state is still a tough challenge even though several modern technologies have been involved in this frontier. Herein, we propose that screening of the disease-related peptidome in blood is worth a try to explore the biomarker set of GDM, because the technology effectively avoids the interference of abundant proteins in blood, while its results lead to discovery of peptide biomarkers or verification of intact protein changes in response to GDM. A total of 179 sera were collected, including 83 GDM and 96 health pregnant women. The serum peptides were enriched by hydrophilic-lipophilic balanced materials coated on magnetic solid phase. The bound peptides were processed through a MGI liquid automatic transfer and were finally eluted by 70% acetonitrile. The peptides without treatment of trypsin digestion were directly loaded onto an UltiMate 3000 HPLC that was mounted on Lumos Orbitrap mass spectrometer with acquisition mode at DDA. And the acquired MS/MS raw data was analyzed using MaxQuant searching against databases of human proteome with default parameter of non- tryptic digestion. The quantities of serum proteins derived from the differential peptides were further confirmed by Acquity UPLC I that was coupled with SCIEX Triple Quad 6500 mass spectrometer with acquisition mode at MRM. Finally, a discriminator was built up to distinguish the health pregnant from the GDM sera. Totally 11849 peptides were identified by mass spectrometer, while after a strict cutoff (Fold>1.5, p<0.05, and Missing value<0.33), the abundance of 206 peptides in GDM was found significantly different from that in health pregnant sera. Focusing on these peptides, 151 peptides globally detected were further implemented to stridently quantitative measurement with two approaches at PRM and MRM mode. At PRM, 10 peptides were confirmed significant abundance differences between health and GDM, while these are derived from 6 intact human proteins, such as FIBA, FIBB, TTHY, KISS1, HEP2, and LEGL. At MRM, 151 peptides were theoretically checked their precursor proteins at first, resulting in them come from 65 proteins. Of these proteins, 27 proteins with specific MRM signals that have been evaluated by PQ500 study were selected for further experiments. In these serum samples, 20 proteins were satisfied with their MRM signals and were quantified based on calibration curves. Finally, these quantified proteins were taken into construction of a discriminator by machine learning. Combination of the quantitative signals of four serum proteins (A2MG, CO3, KNG1 and HEP2) in ROC could well distinguish the two serum samples of health pregnant and GDM with AUC at 0.871. All the data, from peptides to proteins, suggests the abundance of some serum proteins sensitive to GDM status. 473

474

P-I-0321 DIA, DDA and PRM strategies for biomarker identification and validation in plasma cohorts of COVID-19 patients with varying severity Marina Gay1, Antoni Berenguer-Llergo2, Antonia Odena1, Gianluca Arauz-Garofalo1, Ignasi Folch-i-Casanovas1, Mar Vilanova1, Carla Triguero1, Sònia Jarió1, Joan Calvet2, Marta Vilaseca1 1IRB Barcelona, Barcelona, Spain 2Institut d’Investigació i Innovació Parc Taulí (I3PT), Rheumatology Department, Barcelona, Spain High-throughput Mass Spectrometry (MS)-based proteomics has generated extensive data on COVID-19, enhancing our understanding of the disease. Studies on COVID-19 patients show significant impact of the infection on plasma and sera proteome [1–3]. These findings have spurred research into biomarkers for diagnosis, prognosis or therapy [4,5]. Improved MS methodologies for accurate proteome profiling and evaluation promise better quantity and quality of data, as well as the cost-benefit quantification [6]. These methods include label free approaches, based on Data-Independent Acquisition (DIA) or Data-Dependent Acquisition (DDA) strategies, to scan for the whole proteome or the Parallel Reaction Monitoring (PRM) strategy, for validating known proteins without needing specific antibodies. Despite extensive use of DIA, DDA and PRM in studying diseases like COVID-19 [7,8], the concordance between these technologies for biomarker identification has not been formally evaluated in clinical human samples. This study aimed to evaluate the concordance of these three MS proteomics approaches in identifying and assessing biomarkers discovery in the plasma of COVID-19 patients with varying grades of severity. We focused on how DIA-based strategy enhances candidate identification versus DDA and compared protein quantification reliability between DIA, DDA and PRM. Concordance between DIA and DDA were conducted on samples from 48 hospitalized and 48 non-hospitalized patients (N = 96). Among them, samples from 10 patients in each group were available for DIA-PRM comparisons (63 proteins). PRM data showed higher statistical power and better biomarker identification than DIA, which in turn was superior to DDA. All technologies exhibited excellent concordance in Fold-Change and good sample-level correlation. However, systematic differences between PRM and DIA, and between DIA and DDA, highlight the need for calibration to make their quantifications comparable. We achieved this with simple linear regression, which performed outstandingly. Finally, we analyzed by PRM a cohort of 170 individuals and we selected 10 proteins as clinically relevant for further validation as infection prognosis biomarkers. References 1. Shen, B. et al. Cell 182, 59–72.e15 (2020). 2. Messner, C. B. et al. Cell Syst 11, 11–24.e4 (2020). 3. Whetton, A. D. et al. J. Proteome Res. 19, 4219–4232 (2020). 4. Bouhaddou, M. et al. Cell 182, 685–712.e19 (2020). 5. Bojkova, D. et al. Nature 583, 469–472 (2020). 6. Bantscheff, M. et al. Anal. Bioanal. Chem. 404, 939–965 (2012). 7. Geyer, P. E. et al. EMBO Mol. Med. 13, e14167 (2021). 8. Demichev, V. et al. Cell Syst 12, 780–794.e7 (2021). 474

475

P-I-0322 Multi-omics analysis of tissue and serum samples from patients suffering from Crohn's patients identifies pathological mechanisms Christopher Gerner1, Andrea Bileck1, Rudolf Oehler2, Lukas Unger2 1University of Vienna, Faculty of Chemistry, Vienna, Austria 2Medical University of Vienna, Department of General Surgery, Vienna, Austria Crohn's disease (CD) is a chronic inflammatory bowel disease (IBD) that primarily affects the small intestine. Patients undergoing surgery provided blood and tissue samples for subsequent molecular profiling, including data-dependent and data-independent proteomics, data-dependent oxylipin analysis and data-independent metabolomics. Comparison of inflamed tissue samples with less inflamed matched controls revealed a significant loss of antimicrobial proteins, mitochondrial proteins, xenobiotic enzymes and enzymes involved in lipid metabolism in diseased tissue samples. Inflammation-related proteins, platelet-derived proteins and protein markers of hypoxia were found to be upregulated. These findings were corroborated by characteristic patterns of proteins, oxylipins and metabolites detected in blood serum and related to inflammation, platelet activation and hypoxic stress. The present data suggest a synergistic escalation of events involved in mitochondrial function, detoxification, gut microbiome control and blood coagulation, and may allow better prediction of disease risk factors. 475

476

P-I-0323 Developmental-specific proteomic vulnerabilities of Leukemia Sudip Ghosh1, Louise Weigel Olsson1, Mina Davoudi1, Joanna Watral1, Somdatta Pal1, Jenny Hansson1 1Lund University, Department of Experimental Medical Science, Lund, Sweden Introduction: The distinct development of fetal and adult blood cells, governed by unique molecular networks and extracellular signals, contributes to the separate onset of acute leukemia in children and adults. Both infant and adult MLL rearranged (MLLr) acute leukemia have poor prognoses due to limited treatment options. To develop age-tailored targeted therapies, a better understanding of the regulatory networks and their responses to external signals is needed. Previously, we identified age-specific proteomic features of normal and MLLr leukemia-initiating cells1-2. Here, we establish an in vitro approach to perturb these age-specific features in infant and adult leukemic cells by stimulating with IFNa or HDAC3 inhibitor, aiming to induce an age-specific proteomic response. The long-term goal is to identify proteins for age-tailored immune therapies targeting immunopeptides on pre-leukemic and leukemic cells. Methods: Hematopoietic progenitors were FACS-sorted from fetal and adult MLL-ENL-inducible mice, cultured for 3 days, treated with IFNa, and collected at 17, 40, and 64 hours. Infant (THP1, KOPN8) and adult (NOMO-1, MONO-MAC-6) human leukemic cell lines were treated with IFNa or HDAC3 inhibitor (HDAC3i) and collected after 17 hours. The cellular proteome was analyzed using in-depth data-independent acquisition-based proteomics. Results: We found that IFNa stimulation rapidly induces protein expression in both healthy and pre-leukemic cells, and this effect persists longer in wild-type cells than in pre-leukemic cells. IFNa induces upregulation of potential age and lineage-specific targets in both mice pre-leukemic and human leukemic cells. Conversely, HDAC3 inhibition reduces viability in infant leukemia cells harboring MLL-ENL fusion by reprogramming metabolic pathways. In addition, HDAC3 inhibition impacts regulators of proteasomal degradation and induces differential expressions of proteasomal units, predominantly in infant leukemia, which may cause indirect immunomodulatory effects through modulating antigen processing and presentation. Conclusions: Our study comprehensively describes IFNa-regulated proteome changes in healthy and MLLr hematopoietic progenitor cells, as well as IFNa- and HDAC3i-regulated proteomic changes in infant and adult human leukemic cells, highlighting age-specific vulnerabilities in MLLr leukemia. Combined with ongoing efforts to immunopeptides, this work contributes to future developments of combinatorial leukemia treatments utilizing proteomic perturbation and targeted immunotherapy. identify development-specific References 1. 2. Jassinskaja, M., et al. (2017). Cell Reports, 21(11), 3285–3297. Jassinskaja, M., Ghosh, S. et al. (2024) Leukemia, 38, 1115-1130. 476

477

P-I-0324 Proteomic signatures of Mitochondrial and immune responses as predictors of melanoma evolution Jeovanis Gil Valdés1, Jéssica Guedes1, Leticia Szadai2, Nicole Woldmar1, Natalia Pinto1, István B Németh2, György Marko-Varga1 1Lund University, Translational Medicine, Lund, Sweden 2University of Szeged, Dermatology and Allergology, Szeged, Hungary Melanoma, characterized by its high metastatic potential and recurrence, presents a complex interplay of genetic, metabolic, and immune mechanisms. Recent advances have elucidated the key roles of mitochondrial dysfunction and immune evasion in the progression and recurrence of melanoma1,2. These insights pave the way for novel therapeutic targets and prognostic biomarkers, enhancing our capacity for personalized treatment strategies. Recently, we have shown that mitochondrial dysregulation is a hallmark of aggressive melanoma phenotypes, particularly influencing tumor cell metabolism and resistance to apoptosis3. The mitochondrial translation machinery, crucial for biosynthetic processes, exhibits notable alterations in melanoma, leading to metabolic reprogramming. Enhanced mitochondrial biogenesis and function have been linked with increased tumor aggressiveness and poor prognosis due to their role invasive tumor characteristics4. Concurrent with mitochondrial alterations, melanoma cells deploy sophisticated strategies to evade immune surveillance5. Mutation and downregulation of major histocompatibility complex (MHC) molecules and upregulation of immune checkpoint ligands, such as PD-L1, allow melanoma cells to remain undetected by the immune system. Furthermore, the tumor microenvironment is modulated to suppress effective immune responses, facilitating tumor growth and metastasis. in sustaining proliferative and We have generated and leveraged data from extensive proteomic and genomic analyses on clinical samples. Our research underscores the significant interaction between mitochondrial dysregulation and immune responses. For instance, proteomic profiling has identified distinct mitochondrial and immune signatures that differentiate recurrent melanomas from non-recurrent ones. These signatures not only enhance our understanding of melanoma biology but also highlight potential biomarkers for early detection of recurrence. Targeting mitochondrial pathways offers a promising therapeutic avenue. We have shown that agents that disrupt mitochondrial functions, such as oxidative phosphorylation and mitochondrial translation inhibitors, have shown efficacy in cellular models by inducing metabolic crisis in melanoma cells. The integration of mitochondrial and immune signatures provides a comprehensive understanding of melanoma's aggressive nature and recurrence, supported by ongoing clinical trials focusing on these pathways. This presentation underscores the importance of multi-omics approaches in unraveling the complex biology of melanoma, setting the stage for future breakthroughs in melanoma management. References 1. Szadai, L. et al. Clin Transl Med 13, e1495 (2023). 2. Gil, J. et al. Cancers (Basel) 13, (2021). 3. Gil, J. et al. bioRxiv 2022.10.24.513481 (2022) 4. Pinto, N. et al. Submitted (2024). 5. Gil, J. et al. bioRxiv 2021.04.10.439245 (2021) 477

478

P-I-0325 CSF proteomic analysis across the Alzheimer's disease continuum Sophia Weiner1, Mathias Sauer1, Nicholas Ashton1, Pedro Rosa-Neto2, Kaj Blennow1,3, Henrik Zetterberg1,3,4,5, Johan Gobom1,3 1University of Gothenburg, Institute of Neuroscience and Physiology, Moelndal, Sweden 2McGill University, McGill Centre for Studies in Aging, Montreal, Canada 3SU/Moelndal Hospital, Clinical Chemistry/Neurocehmistry laboratory, Moelndal, Sweden 4UCL Institute of Neurology, Department of Neurodegenerative Disease, London, United Kingdom 5Hong Kong Center for Neurodegenerative Diseases, Hong Kong, Hong Kong Introduction: Alzheimer’s disease (AD) can be viewed as a continuum, starting with a preclinical phase that begins many years before onset of symptoms during which amyloid plaques start to accumulate in the brain, progressing to the clinical stage characterized by mild cognitive impairment, and finally on to Alzheimer’s disease dementia. Measuring proteomic changes in the cerebrospinal fluid (CSF) across the AD continuum can reveal at which point in the disease progression different pathological changes occur, identify staging biomarkers that can be used to monitor disease progression and the effect of treatments, and determine new therapeutic targets. Methods: We performed a cross-sectional TMT proteomic study of CSF from the Translational Biomarkers in Aging and Dementia (TRIAD) cohort (n = 214); a highly profiled cohort with clinical and neuropsychological assessments as well as quantitative amyloid and tau positron emission tomography (PET) data that includes participants within the entire Alzheimer’s disease spectrum, as well as patients with frontotemporal dementia, other dementias, and healthy elderly and young individuals. Peptide correlation network analysis was performed to identify peptide clusters, and linear models were used to assess their changes across the disease stages. Many of the findings were validated using data from two previous CSF proteomics studies. Results: Early proteomic changes were detected for proteins linked to endo-lysosomal dysfunction and APOE- related pathways (e.g. CTSD, SIAE, RELN, AGT) that increased in the beginning of the disease and returned to normal in the later stages. Many of these proteins correlated specifically with CSF Amyloid-beta 1-42, suggesting a link to amyloid pathology. Several proteins known to be associated with neurodegenerative processes (e.g. NEFL, YWHAG, ALDOA) showed a gradual increase with disease progression and correlated strongly with tau PET SUVR, as well as with many phospho-tau forms (e.g., pTau217, pTau181, pTau231). Protein C4A/C4B also showed an increase across the AD continuum but correlated more strongly with pTau205. Proteomic changes in the late disease stages involved proteins that reflect diverse neurodegenerative processes and an increase in blood-derived proteins (e.g., ALB, PLG) indicative of blood-brain barrier dysfunction, as well as a decrease in synaptic proteins that may reflect synaptic loss. Conclusions: The concentration of many CSF proteins involved in diverse cellular and pathological processes change across the AD continuum, following different trajectories throughout the course of the disease. Some of these proteins may therefore be better for early diagnosis while others are more suitable to monitor disease progression and treatment effects. Our results also highlight the usefulness of correlating proteome changes to imaging and fluid biomarker data in addition to clinical assessments. 478

479

P-I-0327 Discovery, verification and validation of urinary protein signatures in HIV-associated nephropathy in South Africa Ireshyn Govender1,2, Stoyan Stoychev2, Andrea Ellero2, Firdaus Nabeemeeah3, Neil Martinson3, Previn Naicker1 1CSIR – National Chemical Laboratory, Future Production Chemicals, Pretoria, South Africa 2ReSyn Biosciences, Johannesburg, South Africa 3University of the Witwatersrand, Perinatal HIV Research Unit, Johannesburg, South Africa Introduction: Approximately one in five South Africans is HIV positive, accounting for ~10% of the global burden of HIV/AIDS. South Africa administers the largest antiretroviral therapy (ART) programme worldwide. However, 8-10% of first-line ART patients experience HIV-associated nephropathy (HIVAN), possibly due to both ART and virus insult. Current kidney injury tests are unreliable and reflect following significant kidney damage. Therefore, it is necessary to identify sensitive and accurate markers for early detection of kidney injury. This study employed mass spectrometry-based urinary proteomics for HIVAN biomarker discovery in distinct South African cohorts. Methods: Two cohorts matched in age, race, and gender (case-control) were analysed in the first two phases. A simple sample collection procedure was coupled to a robust, in-house developed, magnetic microparticle-based method for protein capture, clean-up, and digestion. Proteome profiles were generated using an Evosep One coupled to a timsTOF HT in diaPASEF mode. The raw data were processed using DirectDIA+ in Spectronaut 18. Proteins common to both cohorts were filtered as HIVAN reporters and processed using Perseus (v1.6.12.0) and Cytoscape (v3.8.2). Machine learning (ML) was applied to protein abundance data. Results: Several common proteins (q-value ≤ 0.01, unique peptides ≥ 2) showed the same directional change in abundance between the two cohorts. These indicated that damaged kidneys may undergo renal hepatization showing increased levels of liver-associated proteins. Immune system, complement cascade and serine protease inhibitors, were the dominant protein families showing increased abundance in HIVAN patients. The XGBoost algorithm showed ≥ 95% sensitivity and specificity in classifying HIVAN and severity. Conclusions: The workflow applied here showed clinical robustness and can be used as a routine workflow for urinary proteomics studies. ML offers promise for use with complex urinary proteomics datasets for clinically applicable outcomes. Validation of identified protein biosignatures is ongoing in a large cohort (n = 1000) using short gradient DIA-MS, which is the first of its kind study on the African continent. Fig. 1 (optional) 479

480

P-I-0328 Mag-Net™ Urine: Improved and complementary workflow for clinical urinary proteomics Ireshyn Govender1,2, Previn Naicker2, Andrea Ellero1,3, Isak Gerber2, Justin Jordaan1, Stoyan Stoychev1,4, Michael J MacCoss5, Christine Wu5 1ReSyn Biosciences, Pretoria, South Africa 2CSIR – National Chemical Laboratory, Pretoria, South Africa 3University of Pretoria, Pretoria, South Africa 4Evosep Biosystems, Odense, Denmark 5University of Washington, Department of Genome Sciences, Seattle, WA, United States Intro: Urine is an ideal biofluid for clinical proteomics experiments due to its non-invasive ease of collection. Urinary microvesicles from both the urinary tract and systemic sources can provide a diverse source of information related to the health status of a patient. We present an automated, reproducible, and robust sample processing workflow for mass spectrometry-based urinary microvesicle analysis in clinical proteomics. This workflow has been adapted from the Mag-Net™ method (Wu et al. 2023) for deep plasma profiling through enrichment of membrane vesicles. The adapted workflow was compared with two published methods for urinary proteome profiling based on MagReSyn® HILIC microparticles, uHLC (Govender et al. 2023) and protein aggregation capture (PAC) (Batth et al. 2019). Methods: Combinations of existing and novel MagReSyn® beads were evaluated to achieve the greatest proteome coverage. Various capture methods were also tested. Captured vesicles were then lysed and the proteins were solubilised, reduced, and alkylated whilst bound on-bead. Following washes, using organic solvents, on-bead digest was performed through the addition of trypsin and LysC and 2 h incubation at 47°C. Resulting peptides were loaded onto Evotips and analysed using an Evosep One (60SPD) coupled to a Bruker timsTOF HT (default short gradient diaPASEF method). Data were searched in Spectronaut 18 using DirectDIA+™. Protein and peptide data were filtered at 0.01 q-value and processed using STRING-db. Custom data analysis scripts were used to assess microvesicle enrichment. Preliminary Data: The automated Mag-Net™ urine workflow generates peptides ready for mass spectrometry analysis in 6 hours. A large overlap was observed between the workflows with workflow-specific unique peptides and proteins, as expected. The PAC urine workflow showed the highest protein group identifications. Approximately 30% of the detected proteins were unique to urine samples processed with Mag-Net™. Gene ontology-based cellular component analysis showed strong enrichment of vesicles, exosomes and membrane- bounded organelle associated proteins, alluding to positive capture of urinary microvesicles by the Mag-Net™ Urine workflow. Both Mag-Net™ and uHLC enriched for proteins associated with extracellular exosomes, while PAC primarily enriched for extracellular proteins. The specificity of the Mag-Net™ urine method can be further improved through systematic optimization of binding buffer composition, microsphere functional chemistry and microsphere to urine ratio, which is currently on-going. Preliminary data suggest that the Mag-Net™ Urine workflow can be used alone or in conjunction with standard soluble protein urinary workflows, such as uHLC and PAC, to gain deeper insights into specific disease states in clinical proteomics studies. 480

481

P-I-0329 Quantitative SRM assay development for oral cancer prognosis in a clinical routine Daniela Campos Granato1, Carolina Moretto Carnielli1, Alan Santos-Silva2, Marcio Lopes2, Estela Tango3, Thais Brandão4, Ana Carolina Prado Ribeiro4, Adriana Paes Leme1 1Brazilian Center for Research in Energy and Materials - CNPEM, Brazillian Biosciences National Laboratory - LNBio, Campinas, Brazil 2University of Campinas, Dental School, Piracicaba, Brazil 3Unesp – Instituto de Ciência e Tecnologia (ICT), Dental School, São José dos Campos, Brazil 4Instituto do Câncer do Estado de São Paulo, São Paulo, Brazil Introduction: Mass spectrometry-based targeted proteomics for absolute peptide and protein quantification in biological matrices is a powerful tool and has many advantages that make it promising for clinical oncological applications. Previous study from our group1 has revealed through discovery and targeted proteomics a four- peptide signature (COL6A1-pep, LTA4H-pep1, LTA4H-pep2, CSTB-pep) capable of distinguishing low risk from high-risk oral cancer (OSCC) patients. To validate this signature, we have developed a methodology that allows an accurate quantification of a prognostic signature in saliva for further application in oral cancer patients cohorts. Methods: Saliva samples from healthy individuals were used for development and optimization of an absolute quantification (AQ)-MS isotopic labeled workflow2,3 which will be further applied to OSCC patients (N0, without and N+, with lymph node metastasis). LC-MS SRM analysis was performed on triple-quadrupole Xevo TQ-XS coupled with Acquity UPLC-Class M LC (Waters). Results: First, we have evaluated the ability of different reagents such as urea, TFE, DDM and SDS to extract, solubilize and recover targeted peptides from saliva. The effectiveness of the protocols was evaluated regarding endogenous peptide recovery, providing a 2-fold increase under LysC and trypsin sequential digestion. Protocol without any desalt or acetone precipitation also improved peptide recovery. For transition selection, 3 and 10 transitions per peptide were evaluated, and the total area was improved in 50% for both LTA4H and CSTB peptides, while no differences were observed for COL6A1-pep when considering 10 transitions per peptide for quantification. Collison energies (CE) were optimized in steps of 3eV, under a range of 4 different steps, and resulted in higher peak area in CE different from the default4. We also compared 20-, 30- and 60-minute total gradients (11.5 min, 22.5 min and 45min at 40% MeCN, respectively), and total peak area did not change considerably while number of points per peak was impaired in shorter gradients, but still obtained 8 points per peak in 30 min. We compared the performance in signal to noise ratio of BEH, CSH and HSS iKey separation devices, and HSS showed an improvement of 50%. The response curves with the range 500 amol to 200 fmol SIS peptide concentration was performed with the optimized sample preparation and SRM-MS method using saliva, HeLa and BSA matrices, for LOD calculation of COL6A1, LTA4H, and CSTB peptides. For all the optimization steps, simpler, cheaper, and faster protocols were prioritized to ease the transference to the clinical routine. Conclusions: In summary, the present study has been able to develop a quantitative SRM-MS quantification workflow applied to prognostic signature for oral cancer patients in the clinical routine. References: 1-doi:10.1038/s41467-018-05696-2; 2- doi: 10.1186/s12014-024-09452-1; 3- doi: 10.3389/fimmu.2021.765898. 4- doi: 10.1021/ac102179j 481

482

P-I-0330 Mass spectrometric comparison of blood matrices and sample preparation techniques. Benchmarking clinical sample SOPs and different proteolysis methods for plasma and serum by LC-MSMS in DIA mode Thomas Gronauer1, Christine von Toerne1, Juliane Merl-Pham1, Katharina Habler2, Daniel Teupser2, Stefanie Hauck1 1Helmholtz Munich, Metabolomics and Proteomics Core, Munich, Germany 2Ludwig-Maximilians-Universität München (LMU), Institute for Laboratory Medicine, Munich, Germany While plasma and serum are the most frequently studied sample types in clinical research, the susceptibility to minor changes during clinical sample collection and preparation such as centrifugation, type of collection tubes and sample temperature during handling remains understudied. The high dynamic range of plasma proteins further impairs the analytical depths and coverage when time and costs are critical. Here, we systematically explored the LC-MS/MS proteomic profiles of pooled blood samples from three healthy probands sampled by EDTA-, citrate- and heparin-anticoagulated blood as well as serum generated with and without separation gel. In addition, proteolysis prior to LC-MS/MS analysis was conducted with five different methods. We evaluated the commercially available iST and ENRICH-iST kits from Preomics as well as strong- anion exchange (SAX) beads from MagReSyn, the TFA-based approach SPEED and a depletion-based method which uses perchloric acid (perCA) for precipitation of large proteins. Mass-spectrometric measurements were performed on a QExactive HF-X and a timsTOF HT in DIA mode. Identification and quantification analysis of recorded spectra was performed with DIA-NN v1.8.1. The total dataset comprises 250 raw files (5 different blood matrices, proteolyzed with 5 different methods and measured in 5 replicates on two LC-MSMS set-ups). Over all possible combinations of matrices and proteolysis methods, the QExactive HF-X MS instrument achieved identifications between 250 and 600 protein groups whereas the timsTOF HT identifies between 275 and 1100 protein groups. Large differences on protein group ID level in the applied proteolysis techniques are evident. Workflows using SAX-beads, ENRICH-iST and the perCA precipitation showed highest identification rates but also highest variability between the different blood matrices. Here, EDTA-plasma and serum exhibited the highest ID numbers while citrated plasma showed the lowest. 116 protein groups overlap between all 25 sample sets for QExactive HF-X and 182 protein groups for timsTOF HT. Moreover, subsets of protein groups are uniquely identified in specific combinations of proteolysis method and blood matrix indicating important implications for clinical studies. In conclusion, this study represents a systematic approach to determine suitable sample preparation and matrix parameters for the robust identification of individual body fluid marker proteins by mass spectrometry. 482

483

P-I-0331 In-depth analysis of the plasma proteome: Are we enriching extracellular vesicles, platelets, or soluble proteins? Ines Metatla1, Kevin Roger1, Cerina Chhuon1, Sara Ceccacci1, Ida Chiara Guerrera1 1Necker Proteomics, Université Paris Cite, Structure Fédérative de Recherche Necker, INSERM US24/CNRS UAR3633, Paris, France Plasma is a complex milieu comprising extracellular vesicles (EVs), residual platelets, and soluble proteins, each carrying information about physiological and pathological states. Traditional neat plasma analysis yields identification of hundreds of proteins, but recent strategies allow to expand coverage to thousands of proteins. These strategies primarily utilize centrifugation and magnetic nanoparticles. In this study, we compared three different strategies for achieving an in-depth analysis of plasma proteomics to evaluate the contribution of soluble proteins, EV cargo proteins, and platelets. We compared the protein profiles obtained from EV preparations isolated by centrifugation and enriched plasma fractions isolated using magnetic nanoparticle- based and beads methods. We analyzed platelet-poor plasma (PPP) samples from five healthy donors. For each sample, we conducted neat plasma analysis and nanoparticle-based enrichment using the XT kit on the Proteograph assay (Seer), as well SCX magnetic beads (MagNet ResynBio) and multiple centrifugation steps to enrich the EV. Proteins were digested using STRAP and analyzed on the Evosep One coupled to a timsTOF HT (Bruker Datonics) for the MagNet assay and on the nanoElute2-timsTOF PRO (Bruker Datonics) using packed emitter column (C18, 25 cm × 75 μm 1.6 μm) (IonOpticks). Data were analyzed using DIA-NN in library-free mode. We combined ORA, GSEA and EV- specific markers, to assess the presence of EV cargo proteins in different preparations, providing insights into the functional relevance of EVs in plasma. Preliminary results indicate that although all three strategies allowed identification of 1200-4500 proteins, and that they were very distinguishable by PCA analysis. From the samples, we quantified ca. 4500 using EV preparation by centrifugation, ca. 4200 using Nanoparticles Seer and about 1200 proteins using SCX beads MagNET. Samples exhibited distinct signatures of EVs, platelets, and soluble proteins, with significant enrichment of EV cargo proteins in EV preparations compared to nanoparticle-enriched plasma fractions. EV markers, CD63 and CD9 were two times more abundant in EV preparation compared to Seer, and were not detected in neat plasma and in MagNET. CD81 was uniquely identified in EV preparations. Complement related proteins were not significantly enriched in any samples. Further data acquisitions and analysis are underway to confirm these findings and elucidate further the specific pattern of identified proteins in each sample preparation type. This analysis provides strategies, which can guide the choice of biomarker discovery approach. insights into the composition of plasma proteome enriched using different 483

484

P-I-0332 Unraveling the molecular mechanisms of Metabolic Dysfunction-Associated Steatotic Liver Disease (MASLD): Leveraging Proteomics and PTMs analysis for the development of precision medicine Victor M. Guerrero-Sanchez1, Enrique Calvo2, Alfonso Mora3, Guadalupe Sabio3, Jesús Vázquez1,4 1Centro Nacional de Investigaciones Cardiovasculares Carlos III (CNIC), Cardiovascular Proteomics Laboratory, Madrid, Spain 2Centro Nacional de Investigaciones Cardiovasculares Carlos III (CNIC), Proteomics Unit, Madrid, Spain 3Centro Nacional de Investigaciones Oncológicas (CNIO), Organ crosstalk in metabolic diseases Group, Madrid, Spain 4Centro de Investigación Biomédica en Red, Enfermedades Cardiovasculares (CIBERCV), Madrid, Spain Steatotic Liver Disease (SLD) is a pathology characterized by an excessive accumulation of fat in hepatic tissue. This abnormal concentration of lipids in hepatocytes produces damage and subsequently activates the inflammatory response. Persistent inflammation can lead to further liver damage and the development of more severe outcomes such as scarring (fibrosis), cirrhosis, or even hepatic cancer. Several risk factors have been associated with the development of this pathology, including obesity, heavy alcohol consumption, type 2 diabetes, high levels of triglycerides in the blood, and genetics, among others. Metabolic (dysfunction)- Associated Steatotic Liver Disease (MASLD) is a type of SLD in which the steatosis is not associated with high alcohol consumption, but it can also lead to other complications like cardiovascular disease. In hepatocytes, an excess of glucose and free fatty acids promotes de novo lipogenesis. As a result of excessive lipid accumulation, the b-oxidation decreases, and the mitochondria become dysfunctional, increasing the production of reactive oxygen species (ROS). This accumulation of oxidative stress damages lipids, proteins, and DNA, promoting inflammation and fibrosis, and in advanced stages, cellular apoptosis or necrosis. In this study, we analyzed the proteomic profile of liver biopsies from patients with MASLD under different conditions (Obesity, Type 2 Diabetes, Steatosis, and Fibrosis). The results revealed which key proteins involved in sucrose, lipid, and amino acid metabolism, as well as signaling, are differentially implicated across the various patient types. In addition, we utilized our lab-developed softwares (Comet-PTM, ReCom, iSanXoT, and a new PTMs analysis package) to conduct an open-search proteomic analysis, aiming to obtain a detailed map of the molecular alterations occurring under each of the conditions analyzed, including a comprehensive study of the post- translationally modified peptidome. This analysis provided an additional layer of information regarding the changes taking place in the studied proteomes. The majority of the differential PTMs were oxidative and exhibited varying abundances, enabling differentiation between patient groups. Overall, this study identified differential proteins with the potential for use as biomarkers for the diagnosis of MASLD. Additionally, the utilization of a new open-search analysis tool enabled the localization of PTM differences across the patients, providing a more comprehensive representation of protein changes in liver tissue. This will aid researchers in the development of targeted therapy strategies. 484

485

P-I-0333 Proteomic Characterization of Papillary Thyroid Carcinoma with Distant Metastasis Zhiqing Gui1, Yan Zhou2, He Wang2, Lu Li2, Pingping Hu2, Tiannan Guo2, Yaoting Sun2, Zhihong Wang1, Hao Zhang1 1The First Hospital of China Medical University, Liaoning, China 2Westlake University, Hangzhou, China Introduction: Papillary thyroid carcinoma (PTC) generally has a favorable prognosis, but distant metastases can occur in approximately 6% of patients. The disease-specific survival rates for PTC patients with distant metastases are only 35% at five years and 25% at ten years. Identifying biomarkers and understanding the mechanisms underlying distant metastasis in PTC are particularly important. In our study, we conducted analyses of exome and proteome in PTC patients with and without distant metastases to identify key biomarkers and pathways that promote tumor invasion and metastasis. Methods: We retrospectively collected FFPE tissues and clinical data from 34 PTC patients with distant metastases, treated at the First Hospital of China Medical University between 2011 and 2016. Additionally, we collected data from 35 PTC patients matched by clinical and pathological characteristics. Exome sequencing was performed on 52 of these patients, comparing gene expression profiles with those from the TCGA database to observe changes specific to Asian patients. Proteomic data from all samples were collected and analyzed using data-independent acquisition (DIA) mode on the Thermo Scientific Orbitrap Exploris 480. Preliminary Data: Exome sequencing did not identify significant differences between the two groups genes. Based on proteomics, a total of 52,169 precursors, 44,838 peptides, and 7,551 proteins were identified. Using limma and Welch's t-test, we found 207 consistently differentially expressed proteins (DEPs) in PTC patients with distant metastases (40 upregulated and 167 downregulated, log Fold Change > 0.5, adjusted P < 0.05). Functional and pathway enrichment analysis revealed significant clustering in mitochondrial dysfunction. By focusing on proteins localized to the mitochondria and with known drug targets, we identified 33 significantly different proteins, among which only LRP2 was upregulated. We further validated LRP2 expression in tissues through immunohistochemical staining across all samples. Next, data were randomly divided into a training set and a test set (2:1), and using the top ten proteins ranked by the Gini coefficient from the DEPs, we built a random forest model for classification. The model achieved an accuracy of 88.1% for identifying patients with distant metastasis. Novel Aspect: By leveraging exome sequencing and proteomic analysis, this study offers unique insights into the molecular mechanisms of metastatic PTC in Asian patients. It highlights mitochondrial dysfunction as a crucial factor and identifies LRP2 as a potential biomarker. Additionally, a highly accurate random forest model based on differentially expressed proteins predicts distant metastasis in PTC, paving the way for personalized treatment strategies. 485

486

P-I-0334 Sample preparation comparison for the investigation of unstable peptidomic biomarkers of hypertension in human plasma Elliot John Gyedu1, Dennis Bernieh1, Dan Lane1, Donald J.L Jones1, Pankaj Gupta1 1University of Leicester, Cardiovascular Sciences, Leicester, United Kingdom Introduction: The renin-angiotensin system (RAS) and kallikrein-kinin system (KKS) are a network of endogenous peptides with vasoconstrictive/vasodilative properties. They are associated with the development of hypertension, the most prevalent risk factor for cardiovascular diseases. However, they have not been studied extensively due to the instability of the peptides in human plasma. Method development of the RAS and KKS in human plasma requires either protease inactivation or the use of a surrogate matrix. In the former, there is a background concentration of the endogenous peptides which may lead to inaccurate quantitation and in the latter, a surrogate matrix may not mirror the matrix effects in human plasma. The aim of this project was to develop and test different sample preparation methods for accurate quantification of the peptides in human plasma without the need for a surrogate matrix and develop a quantitative assay using liquid chromatography tandem mass spectrometry (LC-MS/MS). Methods: Purchased human plasma were aliquoted into four 1.5 mL Eppendorf tubes and in the first tube, a protease inhibitor mix was added; the second aliquot was heated at 50 oC for 10 minutes and a protease inhibitor mix added after cooling; the third aliquot was only heated for 50 oC; the fourth aliquot was not heated and a protease inhibitor mix was not added. Nine calibration standards ranging from 0.06 to 200 fmol/µL were prepared for the light and heavy RAS and KKS peptides in each of the four plasma aliquots to prepare to obtain a final sample volume of 100 µL. Four-hundred microlitres of acetonitrile was added to the samples, vortexed briefly and centrifuged for 10 mins at 20000 rpm. Two-hundred microlitres of the supernatant was dried down, lyophilised and reconstituted with 50µL of 5% acetonitrile, 0.1% formic acid. Six microlitres of the resulting solutions were injected into the Waters Xevo UPLC- MS/MS and the results from the chromatographic data analysed. Results: The peptides had varying sensitivity and linearity in the four methods. With no heating/ addition of protease inhibitor mix, the peptides had low sensitivity and linearity, R2 < 0.67 and not all peptides were not detected (12/19 at LOD, 2.5 fmol/µL). Upon addition of protease inhibitor, the sensitivity and linearity of the assay increased, (19/19 at LOD = 0.125 fmol/µL), R2 > 0.93. With heating, the extraction method had high linearity however sensitivity did not improve, (19/19 at LOD = 0.125 fmol/µL), R2 > 0.95. Heating plus addition of protease inhibitor mix had the highest sensitivity and linearity for the peptides, (19/19 at LOD = 0.06 fmol/ µL), R2 > 0.98 In summary, we developed a novel assay for all endogenous angiotensin and kinin peptides in plasma. Assay linearity, and peptide stability was reserved upon heating and addition of protease inhibitor mix. Following validation, this assay has the potential to be utilised in the clinic for the management of patients with hypertension. 486

487

P-I-0335 Proteomic landscaping of high-grade serous ovarian carcinoma identifies stearoyl-CoA desaturase 5 as a potential predictive biomarker for poly(ADP-ribose) polymerase inhibitor response Hyeyoon Kim1,2, Se Ik Kim3,4, Kisoon Dan1, Hong-Beom Park5,6, Cheol Lee7,8, Dohyun Han5,9,1, Maria Lee3,4 1Seoul National University Hospital, Proteomics Core Facility, Seoul, South Korea 2Pacific Northwest National Laboratory, Biological Science Division, Richland, WA, United States 3Seoul National University Hospital, Department of Obstetrics and Gynecology, Seoul, South Korea 4Seoul National University College of Medicine, Department of Obstetrics and Gynecolgy, Seoul, South Korea 5Seoul National University Hospital, Department of Transdisciplinary Medicine, Seoul, South Korea 6Seoul National University Graduate School, Department of Biomedical Sciences, Seoul, South Korea 7Seoul National University Hospital, Department of Pathology, Seoul, South Korea 8Seoul National University College of Medicine, Department of Pathology, Seoul, South Korea 9Seoul National University College of Medicine, Department of Medicine, Seoul, South Korea High‐grade serous ovarian carcinoma (HGSOC) has shown high recurrence and mortality rates despite treatment comprising cytoreductive surgery and chemotherapy. However, the recent introduction of poly(ADP‐ribose) polymerase inhibitors (PARPi) in the management of HGSOC significantly improved the prognosis. As more HGSOC patients receive PARPi treatment, accurately predicting the treatment response becomes crucial. In this study, we first establish the proteomic landscape of ovarian cancer according to PARPi response. Protein signatures were validated in the independent cohort, and preliminarily investigated their potential roles in PARPi resistance. To identify protein signatures associated with PARPi resistance, we conducted an in‐depth quantitative proteomic analysis of formalin‐fixed paraffin‐embedded (FFPE) cancer tissues (n = 24) from platinum‐sensitive recurrent HGSOC patients using a tandem mass tag (TMT) 10‐plex quantification strategy. According to PARPi treatment response, a total of 187 differentially expressed proteins (DEPs) were identified (p < 0.05, |fold‐change| > 1.2), with 54 and 133 exhibiting higher expression in the good and poor response groups, respectively. Next, we conducted feature selection with leave‐one‐out cross‐validation to identify protein signatures that stratify the response to PARPi. Machine learning algorithms yielded lists of top proteins selected by each respective algorithm. Among these, we identified three common proteins: stearoyl‐CoA desaturase 5 (SCD5), NUDT4 and GRP107 which were upregulated in the poor response group. For validation, we performed data‐independent acquisition‐based proteomics using an independent set of FFPE tissues (three good and seven poor responders). The SCD5 showed high predictive performance in identifying poor responders (area under the receiver operating characteristic curve, 0.952). From a clinical viewpoint, for HGSOC patients who are identified as at high risk of showing poor response to PARPi maintenance therapy via pretreatment proteomics analyses, physicians may deter the use of PARPi, maximize the effect of PARPi combined with other therapy, or recommend intensive surveillance during PARPi maintenance to detect recurrence earlier. Our findings broaden biological insights into predicting PARPi responses 487

488

P-I-0336 Transmembrane protein 160 (TMEM160) is involved in cell proliferation and migration in lung and cervical cancer Gloria Angelina Herrera-Quiterio1, Heriberto A Valencia-González2, Karen G. De la Cruz-López2, Diana Lashidua Fernandez Coto1, Jeovanis Gil Valdés3, Magdalena Hernández-Ortiz1, Angel Gabriel Martínez-Batallar1, Alejandro Avilés-Salas2, Alejandro García-Carrancá2, Sergio Manuel Encarnación-Guevara1 1National Autonomous University of Mexico, Center for Genomic Sciences, Cuernavaca, Mexico 2National Cancer Institute, Research Laboratory 9 "Viruses and Cancer", Mexico City, Mexico 3Lund University, Department of Clinical Sciences Lund, Lund, Sweden Introduction: Transmembrane protein 160 (TMEM160) is localized to the mitochondrial membrane and is predicted to also localize to the nucleus, peroxisome membrane, and endoplasmic reticulum. It has been reported that in humans its expression is found in more than 200 tissues. It has been suggested that TMEM160 may suppress the generation of reactive oxygen species and stabilize mitochondrial proteins. In recent years, it has been reported that TMEM family proteins play roles as oncogenes or tumor suppressor genes. It is unknown if TMEM160 plays a role in cancer, we identified that it is overexpressed in the cell lines A549 of lung adenocarcinoma and HeLa of cervical adenocarcinoma, so we are interested in identifying the role that TMEM160 plays in lung cancer and cervical cancer. Methodology: To investigate the role of TMEM160 in cancer, we employed a variety of techniques. First, we silenced the TMEM160 gene using CRISPR/CAS9 in A549 and HeLa cell lines. We then conducted proliferation, migration, and invasion assays in wildtype (WT) and knockout (KO) cells. In vivo experiments involved xenografting mice with WT and KO cells. To gain further insight, we co-immunoprecipitated TMEM160 in protein extracts from A549 and HeLa cell lines using LC-MS/MS. Pathway enrichment analysis was performed to understand the processes involving TMEM160 and its partners. Finally, we used immunofluorescence to identify the subcellular localization of TMEM160. Results: TMEM160 shows higher expression in the A549 cell line versus the BEAS cell line (non-tumor bronchial epithelium) and in the HeLa cell line compared to HaCaT (non-tumor keratinocytes). In addition, TMEM160 knockout decreases cell proliferation and migration in vitro and in vivo in lung cancer and cervical cancer cells. It was identified that in co-immunoprecipitations in A549 cells, TMEM160 interacts with importins that mediate the entry of proteins into the nucleus. Likewise, the cellular processes that were enriched were DNA replication and spliceosome. By immunofluorescence we observed that TMEM160 has cytoplasmic and nuclear localization. Conclusions: This study provides an initial understanding of the function of TMEM160 in cancer cells. Our findings indicate that the absence of the protein in A549 cells significantly reduces cell proliferation and migration. Moreover, we discovered a novel interaction between TMEM160 and nuclear importins, shedding light on a potential mechanism in cancer progression. 488

489

P-I-0337 Targeted serum proteomics of longitudinal samples from newly diagnosed youth with type 1 diabetes distinguishes markers of disease and C-peptide trajectory Robert Moulder1, Tommi Välikangas1, M. Karoliina Hirvonen1, Tomi Suomi1, Caroline A. Brorsson2, Niina Lietzén1, Sylvaine F.A. Bruggraber3, Lut Overbergh4, David B. Dunger3, Mark Peakman5, Piotr J. Chmura2, Soren Brunak2, Anke M. Schulte6, Chantal Mathieu4, Mikael Knip7, Laura Elo1, Riitta Lahesmaa1 1University of Turku, Turku Bioscience Centre, Turku, Finland 2University of Copenhagen, Novo Nordisk Foundation Center for Protein Research, Copenhagen, Denmark 3University of Cambridge, Department of Paediatrics, Cambridge, United Kingdom 4Katholieke Universiteit Leuven/Universitaire Ziekenhuizen, Leuven, Belgium 5Sanofi, Immunology & Inflammation Research Therapeutic Area, Boston, MA, United States 6Sanofi-Aventis Deutschland, Frankfurt, Germany 7University of Helsinki and Helsinki University Hospital, Pediatric Research Center, Helsinki, Finland With growing concerns for the increasing worldwide incidence of type 1 diabetes, there is a need for markers that can be used to monitor the progression, treatment and remission of the disease. In the present study, selected reaction monitoring (SRM) mass spectrometry was used to measure the levels of 85 type 1 diabetes- associated proteins from longitudinal serum samples of youth newly diagnosed (ND) with type 1 diabetes. To understand associations with beta cell function, the measurements were compared with changes in fasting C- peptide/glucose levels. Additional comparisons were made between the data from ND individuals and cross- sectional measurements obtained from autoantibody-negative unaffected family members (UFMs). In the samples from under 18-year-old ND youth (n=86) recruited among the first 100 ND individuals, statistically significant associations with changes in fasting C-peptide glucose were found in eleven protein, and the levels of 13 proteins differed between ND and UFMs (n=194). Among these were apolipoproteins, insulin-like growth factor (IGF)-family members, coagulants and proteins involved in oxidative stress and beta cell function and integrity. These 21 significant proteins were further validated in ND youth (n=146) subsequently enrolled in the next 150 study and UFMs (n=272). Here, we were able to replicate the majority of the differences between ND and UFMs and confirm association with C-peptide/glucose for three out of the eleven targets. Our study highlights a panel of protein markers reflecting changes accompanying type 1 diabetes progression and their clinical potential to monitor beta cell function. 489

490

P-I-0339 Uncovering H. Pylori associated and independent Gastric cancer signature by tissue phosphoproteomic analysis Hsiang-En Hsu1, Juanilita Waniwan1, Tzu-Chan Hong2, Yi-Ju Chen1, Kuen-Tyng Lin1, Pei-Rong Huang1, Ming- Shiang Wu2, Yu-Ju Chen1 1Academia Sinica, Institute Of Chemistry, Taipei, Taiwan 2National Taiwan University College of Medicine, Department of Internal Medicine, Taipei, Taiwan Gastric cancer, the fifth most prevalent cancer worldwide, involves complex etiologies, including but not limited to H. pylori infection. While numerous genomic studies have found molecular signatures associated with pathogenic phenotypes, 40% patients, especially young male, do not associate with H. pylori that necessitates understanding of their underlying mechanism. To gain more insights into gastric cancer pathogenesis and therapeutic drug discovery, we performed a deep proteomic and phosphoproteomic profiling of paired tumors and adjacent normal tissues (NAT) from a prospectively collected gastric cancer cohort of 120 patients, predominantly male (65%), with an average age of 64. Global proteome and phosphoproteome profiles were conducted by TMT 11-plex-based proteomic workflow using pooled NAT and tumor as an internal standard to ensure long-term analysis reproducibility. A total of 11,498 proteins and 95,341 unique phosphosites from 110,454 phosphopeptides in 10,044 proteins were identified. The PCA analysis revealed separate patterns of normal and tumor profiles. Unsupervised hierarchical clustering of proteomic and phosphoproteomic profiles stratifies patients into three (PT1-PT3) and four (PS1-PS4) clusters, respectively. Proteomic cluster 2 (PT2, 37 patients) is enriched with male (84%), H. Pylori infection (73%) and distal location of the stomach (81%). Similarly, phosphoproteomic subtype 2 (PS2) is also enriched with H. Pylori infection (81%), including 17 common patients from the PT2 cluster with a higher proportion of male and H. Pylori infection. Further pathway enrichment and functional annotation indicate a strong association with bacterial invasion, cell adhesion, and aberrant ECM-receptor modulation. In addition, differential proteins in H. pylori negative and positive populations were compared in NATs and tumor proteome. The findings revealed distinctly differential proteins in tumor and NATs, showing enrichment of immune response and metabolic pathways regulation in the tumor proteome, while the NAT proteome is enriched with focal adhesion and PI3K-AKT signaling pathway. Taken together, the proteomic and phosphoproteomic profiling revealed new insight into the etiology of gastric cancer patients associated with H. Pylori infection, and may provide potential mechanisms for biomarker development and drug treatment. 490

491

P-I-0340 Multi-proteomics identifies knee osteoarthritis progression markers in hoffa's fat pad Luojiao Huang1, Kaj Emanuel2, Martijn van Griensven1, Pieter J. Emans3, Berta Cillero-Pastor1,4 1Maastricht University, MERLN, Maastricht, Netherlands 2Amsterdam University Medical Center, Department of Orthopaedic Surgery, Amsterdam, Netherlands 3Maastricht University, Department of Orthopedic Surgery, Maastricht, Netherlands 4Maastricht University, Maastricht MultiModal Molecular Imaging Institute (M4I), Maastricht, Netherlands Knee osteoarthritis (OA) is characterized by chronic, progressive degeneration impacting the entire joint tissue, often leading to total knee arthroplasty in the end stage. Understanding the pathological progression and identifying early intervention biomarkers is crucial to alleviating patient burden. This study focuses on Hoffa's fat pad (HFP), an adipose tissue in the knee's anterior compartment, known to influence OA inflammatory processes. While HFP's role in OA has been recognized, its potential in diagnosing or predicting disease progression is underexplored. Our research aims to bridge this gap by conducting a comparative proteomics analysis of HFP samples from patients with cartilage defects (CD) and those with end-stage OA, seeking to identify sensitive and specific protein alterations associated with OA progression. In this study, we analyzed HFP samples from 7 CD patients and 22 OA patients. OA patients were divided into 3 subgroups, including 10 non-obese, 8 obese, 4 diabetes. The HFP samples, collected postoperatively, underwent tissue homogenization, protein extraction, and single‐pot solid‐phase‐enhanced sample preparation on a Bravo robotic platform. We employed a multi-proteomics approach, integrating label-free DIA-based quantitative proteomics for extensive HFP proteome profiling and Olink proximity extension assay for targeted quantification of 92 cardiometabolic biomarkers. Statistical analysis was performed using one-way ANOVA analysis, with age factor correction and false discovery rate correction. Pathway enrichment analysis was performed in Cytoscape. Our study identified elevated levels of 28 proteins in all OA subgroups compared to CD patients (Fig 1.A). These proteins primarily function in binding and catalytic activity. Notably, the increased level of TGFBI was consistent between MS and Olink measurements. TGFBI, an extracellular matrix protein, is an important regulator of cartilage homeostasis. Additionally, other consistent proteins, including CRTAC1 and VASN, showed increased levels in non-obese OA. VCAM1 showed increased levels in diabetic OA (Fig 1.B). Previous studies have highlighted the promising role of CRTAC1 as a specific predictor of OA progression to joint replacement. Our results demonstrated its first-time deregulation in HFP and its association with OA. Pathway analysis showed significant changes in energy exploitation and immune response across all OA endotypes. (Fig 1.C). Our study leveraged a multi-proteomics approach and demonstrated the capability of HFP in identifying biomarkers that delineate the progression from CD to OA marked by chronic inflammation and degeneration. The alterations in key proteins like CRTCA1 and VCAM1 align with previous serum studies, suggesting a potential role of HFP in mediating their secretion within the joint environment. Future research will validate these potential markers and their expressions using tissue immunohistochemistry and fresh HFP secretome analysis. 491

493

P-I-0341 Effect of a western diet and age on liver proteome in male and female rats Isabel Huber1,2, Michal Rackiewicz1, Sorel Sagu3, Theresa Hofmeister3, Kristine Wolf1,4, Jens Raila1, Harshadrai Rawel1, Matt Fuszard5, Katja Witzel6, Andrea Henze3,1 1University of Potsdam, Institute of Nutritional Science, Potsdam, Germany 2Health and Medical University, Anatomy, Potsdam, Germany 3Martin-Luther-University Halle-Wittenberg, Nutrition Physiology, Halle (Saale), Germany 4Brandenburg Medical School Theodor Fontane, Brandenburg, Germany 5Martin-Luther-University Halle-Wittenberg, Core Facility Proteomic Mass Spectrometry, Halle (Saale), Germany 6Leibniz Institute of Vegetable and Cornamental Crops, Research Group Plant-Microbes, Grossbeeren, Germany Aging combined with an increasingly unbalanced diet in Western societies promote the manifestation of diseases such as obesity and non-alcoholic fatty liver disease. Proteome profiling, including the identification, quantification and qualitative characterization could provide a better understanding of the molecular mechanisms associated with these diseases. The general aim of the present project was to systematically investigate the potential role of diet and age on the liver proteome as a central metabolic organ. To this end, 84 Wistar rats of both sexes were fed with a standard diet (SD) and a moderate Western diet (WD, 36 kcal % fat, 13 kcal % sugar), respectively, up to the age of 26, 52 and 72 weeks. Rats were characterized among other things for weight changes, clinical and functional parameters as well as for liver histology. For proteome analysis, proteins were extracted from rat liver homogenates, digested, labelled using a TMT 11plex system and analyzed using an ion trap mass spectrometer (Quadrupol-Orbitrap QExactivePlus) coupled to liquid chromatography equipment. The collected data were processed using MaxQuant and Perseus. Looking at the global proteome, a total of 28,079 peptides were initially identified and could be assigned to 4,164 proteins. Following processing and refinement, 1,602 and 1,495 proteins were derived for female and male rats, respectively, with 1,310 proteins overlapping between the sexes. Statistical analysis revealed a significant difference in the relative amount of protein between SD and WD in both female and male rats. Taking age into account, results showed significant differences with 206 and 129 (at 26 weeks of age); 141 and 168 (at 52 weeks of age) and 200 and 195 proteins (at 78 weeks of age) between females and males, respectively. Furthermore, pathway enrichment analysis identified regulators of fatty acid metabolism in both sexes, proteins involved in oxidative energy production, carbohydrate and lipid metabolism were predominant in female groups, while in male rats, proteins involved in peroxisome metabolism and the NF-κB signaling pathway were predominantly identified. In addition, an age-dependent change in the relative abundance of proteins was observed for a total of 35 proteins in males compared to only 9 proteins in females. In contrast, no effects of WD could be observed for clinical and histological compounds, except for differences in body weight for males. These results may illustrate a tendency for females to adapt better than males to a Westernized diet as they age. The significant differences observed between the diets in both sexes may further indicate an adaptation in metabolism, and in particular a compensatory regulation prior to a probable clinical manifestation, mainly in the context of the Western diet. 493

494

P-I-0342 Proteomic strategy to mitigate off-target liabilities for targeted protein degraders Martha Ingola1, Therese Eriksson Faelker2, Stefanie Zimmermann1, Clara Lemos3, Philipp Cromm4, Clemens- Jeremias von Buehler1 1Bayer AG, Drug Metabolism and Pharmacokinetics, Wuppertal, Germany 2Bayer AG, Safety Pharmacology, Wuppertal, Germany 3Bayer U.S. LLC, RED Oncology, Cambridge, MA, United States 4Bayer AG, Drug Discovery Sciences, Wuppertal, Germany Androgen receptor (AR) degraders have emerged as promising therapeutic agents for the treatment of metastatic castration resistant prostate cancer. AR degraders are designed to induce the degradation of the AR protein, which plays a critical role in the growth and progression of prostate cancer. Compared to classical small molecules, novel modalities such as targeted protein degraders (TPD) provide several advantages, offering a potential strategy to overcome resistance to traditional anti-androgen therapies. However, the development and optimization of AR degraders are still in the early stages, and due to its specific mode of action they present unique considerations for safety assessment. The specificity and selectivity of AR degraders especially needs to be thoroughly evaluated to minimize off-target effects. Off-target liabilities can lead to undesired biological effects and potential adverse effects. Treatment effects were investigated in vitro across cell lines and in vivo using dogs following repeated intravenous infusions to overcome limited oral bioavailability. From these studies both on- and off-target cell and tissue proteomics approaches were established. For the analysis of on-target effects, samples from prostate were collected for tissue proteomics. In addition, 13 different tissue types were collected, including heart, lung, liver, muscle, kidney and brain, to analyze potential off-target effects, especially off-target degradation, using tissue proteomics. The various samples were lysed and the protein extracts digested using the Preomics iST kit in an automated manner using a Hamilton"s StarV automated liquid handling robot. The peptides were separated using a nano ultra-high-performance liquid chromatography (UHPLC) coupled to an Orbitrap mass spectrometry (Orbitrap-MS). The data was acquired in a data-independent acquisition (DIA) manner and data analysis was performed using Spectronaut. In summary, the utilization of proteomics offers valuable insights into potential off-target effects of the novel modalities mentioned above, thereby contributing to their preclinical risk mitigation and potential clinical application. Furthermore, ongoing investigation and innovation in this field are essential to fully leverage the therapeutic potential of degraders and to advance personalized treatment strategies for several cancers. 494

495

P-I-0343 Integrating urine and plasma proteomics to uncover age-specific molecular signatures of pediatric diseases Ericka Itang1, Susanne Pangratz-Führer2, Katharina Danhauser2, Anna-Lisa Lanz2, Christoph Klein2, Matthias Mann1, Johannes Müller-Reif1 1Max Planck Institute of Biochemistry, Department of Proteomics and Signal Transduction, Martinsried, Germany 2Dr. von Hauner Children's Hospital, LMU Munich, Department of Pediatrics, Munich, Germany Mass spectrometry-based proteomics is increasingly used in clinical settings due to its ability to identify and quantify hundreds to thousands of proteins. In pediatric medicine, MS-based proteomics not only elucidates disease mechanisms but also provides insights into disease progression in relation to children"s developmental maturation. Given the rapid developmental changes in children, identifying age-specific proteomic profiles is crucial for selecting biomarker panels of pediatric diseases. Here, we used MS-based proteomics to analyze paired urine and plasma samples obtained from the SCIVIAS cohort of pediatric participants to systematically characterize age-specific molecular profiles and infer their implications for pediatric health. Our cohort consists of 1000 children aged 3 to 17 years, balanced across sexes, with 112 healthy controls and 888 patients diagnosed with various pediatric diseases. Urine samples were processed using an on-bead SP3 protocol with MagReSyn® HILIC beads and analyzed using the dimethyl-based multiplexed-DIA (mDIA) workflow with pooled urine as the reference channel. Meanwhile, neat plasma samples were analyzed label-free. In both cases, an automated liquid handling system (Agilent Bravo) was used in a 384-well plate format to increase the throughput. The multiplexed urine samples were then measured using the 30SPD gradient of the EvoSep One system on a PepSep column (15 cm length, 75 μm ID, 1.5 μm C18 beads) coupled to a Bruker timsTOF HT mass spectrometer. Data analysis was done using DIA-NN for protein search, RefQuant for channel normalization, and directLFQ for protein intensity measurement. The neat plasma samples were measured using the 60SPD gradient of the EvoSep One system on a PepSep column (8 cm length, 150 μm ID, 1.5 μm C18 beads) coupled to a Bruker timsTOF HT mass spectrometer. Data analysis was performed using DIA-NN for protein search and directLFQ for protein intensity measurement. The preliminary urinary proteomics analysis identified up to 5000 distinct protein groups across the cohort using the mDIA workflow, with a median of 1200 protein groups per sample. Several proteins linked to reproductive system development positively correlated with age at an FDR-adjusted p-value (Benjamini-Hochberg) of <0.05. Case-control comparisons identified distinct proteomic signatures associated with prevalent pediatric diseases, including cystic fibrosis, celiac disease, and type 1 diabetes mellitus. Additionally, pairwise correlation between urine and plasma proteome will be performed to assess the diagnostic capability of urine proteomics in relation to plasma proteomics. This integration will enable the identification of robust biomarker panels and enhance our understanding of pediatric disease. This study aims to pave the way for MS-based proteomics application for early diagnostics and tailored therapeutic strategies in pediatric precision medicine. Fig. 1 (optional) 495

496

P-I-0344 Novel serum biomarkers in intrahepatic cholangio-carcinoma: a pilot proteomic exploration Maria Iacobescu1, Lavinia Patricia Mocan1, Cristiana Grapa1, Rareș Crăciun1, Ioana Ecaterina Pralea1, Alina Uifălean1, Andreea Maria Soporan1, Tudor Mocan2, Cristina Adela Iuga1 1"Iuliu Hatieganu" University of Medicine and Pharmacy Cluj-Napoca, Romania, Cluj-Napoca, Romania 2"Prof. Dr. Octavian Fodor" Regional Institute of Gastroenterology and Hepatology, Cluj-Napoca, Romania Cholangiocarcinoma (CCA) is a highly aggressive form of biliary tract cancer arising from the malignant transformation of cholangiocytes. A fundamental hurdle lies in comprehending the underlying biology of CCA tumorigenesis, as it holds significance for early detection and intervention. Monitoring alterations in protein abundance within serological proteomes throughout the course of disease progression can offer invaluable insights into the intricacies of CCA physiology and pathology. This study aimed to explore the serum proteome of patients with intrahepatic CCA (iCCA), hepatocellular carcinoma (HCC), liver cirrhosis (CIR), primary sclerosing cholangitis (PSC), to identify new biomarker candidates along with possible pathways involved in the development of these complications. By using high-throughput mass spectrometry proteomics analysis, significant differences were identified. Notably, serum amyloid A1 and A4 (SAA1, SAA4), along with Vascular cell adhesion molecule 1 (VCAM-1) and angiopoietin-1 receptor (TEK), were found to have discriminatory power between iCCA and HCC. Heat shock protein 90 (HSP90), Annexin A9 (ANXA9), alpha-1-antichymotrypsin (SERPINA3) and Leucine-rich alpha-2-glycoprotein (LRG1) were highlighted for evidencing risk factors, such as CIR and PSC. This first pilot proteomic analysis underscores the reliability of identifying promising novel biomarker candidates for iCCA. Further validation and exploration of these serum proteins' clinical utility are warranted. Fig. 1 (optional) 496

497

P-I-0345 Proteomic profiling of left ventricular hypertrophy regression in severe aortic valve stenosis after valve replacement Christoph F. Jacob1, Gabriel Christian Riedemann1, Justus B. Wegener1, Gerd Hasenfuß1, Bo E. Beuthner1, Christof Lenz2,3, Stephan E. Lehnart1, Miriam Puls1, Sören Brandenburg1 1University Medical Center Göttingen, Department of Cardiology and Pneumology, Goettingen, Germany 2University Medical Center Göttingen, Department of Clinical Chemistry, Goettingen, Germany 3Max Planck Institute for Multidisciplinary Sciences, Bioanalytical Mass Spectrometry Group, Goettingen, Germany Background: In severe aortic stenosis (AS), left-ventricular cardiomyocytes hypertrophy to compensate for increased left-ventricular pressure overload and to reduce wall stress. Hypertrophic remodeling, however, impairs the diastolic capacity of the left-ventricle. Reverse remodeling after transcatheter aortic valve implantation (TAVI) may restore the integrity of the left-ventricle but fails to occur in a subgroup of normal ejection-fraction high-gradient (NEF-HG) AS patients. Noteworthy, this is associated with worsening post- interventional outcome shown in the PARTNER trials. Objectives: To identify the molecular mechanisms of reversible versus fixed myocardial hypertrophy after TAVI to optimize and individualize treatment strategies in AS. Methods: In a new proteomic approach, out of 90 NEF-HG patients from the SFB1002 database, 20 small left- ventricular biopsies harvested from NEF-HG patients during transcatheter aortic valve replacement were chosen. Inclusion criteria were based on echocardiographic diagnosis of severe AS and guideline-defined AS-subtype classification. Samples from non-failing donor hearts served as control. Based on clinical follow-up, patients with left-ventricular mass index regression six-month post TAVI were grouped above and below the median of -22.7 g/m2 (above: mean regression: -34,4 g/m2, below: -1,39 g/m2, p < 0.001) and n= 10 vs. 10 samples were analyzed by small biopsy processing including pressure cycling technology and subsequent data-independent acquisition mass spectrometry (DIA-MS). Results: DIA-MS reliably detected 3.726 proteins in human left-ventricular tissue across all samples of which 24 proteins showed significant abundance changes between reversible versus fixed myocardial hypertrophy. Moreover, 827 proteins were significantly up-/ down-regulated compared to non-failing samples. Unsupervised hierarchical clustering of the significant proteins revealed proteomic signatures regarding the median regression, and gene ontology identified potential cellular mechanisms concerning cell cycle pathways and cellular component disassembly (NRAP, STMN1, HSPA2). Analysis of the clinical data of the entire NEF-HG cohort confirmed a worse outcome in fixed hypertrophy shown via Minnesota Living with Heart-Failure Questionnaire (MLHFQ): below median, baseline (BL): 30.34 points, 6 months follow-up (FU): 22.39 points; above median, BL: 28.74 points, FU: 16.26 points (t-test, p = 0.04). Conclusion: Our data identify the proteomic nature of significant left-ventricular hypertrophy regression in severe AS post-TAVI, which is associated with improved clinical outcome. These findings may have implications in terms of timing of valve replacement and the role of adjunctive medical therapy after TAVI. 497

499

P-I-0346 Acrivon Predictive Precision Proteomics (AP3)-guided development and prospective clinical Phase 2 validation of the response-predictive OncoSignature test for the CHK1/2 inhibitor, ACR-368 Magnus E. Jakobsson1, Michail Shipitsin2, Caroline Wigerup1, Ayesha Murshid2, Lei Shi2, Dorte Bekker-Jensen3, Sibgat Choudry2, Christina Noe2, Kailash Singh2, Jayakumar Nair4, James Dunyak2, David A. Proia2, J.-M. Lee4, S. MacLaughlan5, D. Matei6, M. Song7, L. Brubaker8, B. Rimmel9, R. Eskander10, C. Kyi11, H. Williams12, L. Duska13, F. Musa14, R. Coleman15, B. Slomovitz16, Eric Gamelin2, J.-M. Cuillerot2, M. Phadnis2, B. Guercio17, S. Crafton18, P. Konstantinopoulos19, Jesper Olsen3, Kristina Masson1, Peter Blume-Jensen2 1Acrivon AB, Lund, Sweden 2Acrivon Therapeutics Inc., Watertown, MA, United States 3Copenhagen University, Center for Protein Research, Copenhagen, Denmark 4National Cancer Institute, Bethesda, MD, United States 5University of Illinois Cancer Center, Chicago, IL, United States 6Northwestern University Feinberg School of Medicine, Chicago, IL, United States 7City of Hope Comprehensive Cancer Center, Duarte, CA, United States 8UCHealth Cancer Care – Anschutz Medical Campus - University of Colorado Cancer Center, Aurora, CO, United States 9Cedars Sinai Medical Center, Los Angeles, CA, United States 10Obstetrics, Gynecology and Reproductive Sciences, University of California San Diego - UCSD, La Jolla, CA, United States 11Gynecologic Oncology, MSKCC – Memorial Sloan Kettering Cancer Center, New York, NY, United States 12University of Arkansas for Medical Sciences, Little Rock, AR, United States 13The University of Virginia, Charlottesville, VA, United States 14Gynecologic Oncology, Swedish Cancer Institute - Medical Oncology - First Hill, Seattle, WA, United States 15Gynecologic Oncology Department, The US Oncology Network, The Woodlands, TX, United States 16Mount Sinai Comprehensive Cancer Center, Miami, FL, United States 17URMC – University of Rochester Medical Center, Rochester, NY, United States 18West Penn Hospital, Pittsburgh, PA, United States 19Dana-Farber Cancer Institute, Boston, MA, United States Introduction: ACR-368 (prexasertib) is a potent and selective CHK1/2 inhibitor with demonstrated durable, single-agent activity in patients with advanced solid tumors. Genomic biomarkers have been unsuccessful in predicting response to ACR-368 due in part to the complex genetic changes in cancer that translate into dysregulated protein signaling pathways. To address this challenge, we sought to identify and clinically validate protein-based predictive biomarkers that measure the ACR-368-sensitive dysregulated signaling driving tumorigenesis using our proprietary approach, termed Acrivon Predictive Precision Proteomics (AP3). Methods: Ovarian cancer cells with differential ACR-368 sensitivity were profiled with phosphoproteomics. Kinase activity inference and signaling pathway activity analyses were conducted to uncover ACR-368-regulated pathways associated with tumorigenesis and biomarkers predictive of drug sensitivity. A quantitative, multiplexed immunofluorescent (IF) assay was developed based on three biomarkers, termed ACR-368 OncoSignature™, and validated in prospective clinical studies. Results: AP3 phospho-proteomic profiling of ACR-368-sensitive and non-sensitive cells exposed to ACR-368 uncovered >17,000 confidently localized phospho-sites (localization probability > 0.75), with 8272 being significantly regulated by ACR-368 (FC > 1.5, Q-value < 0.05, Limma t-test). Upon ACR-368 treatment, annotated ATM/ATR and CDK1/2 substrate consensus sites were strongly upregulated and CHK1 substrates downregulated. Global phospho-pathway analyses revealed three functionally orthogonal biomarkers that were assembled into a predictive quantitative multiplex in situ assay, ACR-368 OncoSignature™, developed for routine-processed FFPE biopsy tissue to provide a direct readout of a tumor"s dependency on the signaling axis inhibited by ACR-368. The ACR-368 OncoSignature™ test was robustly validated in preclinical studies, including 2 prospectively designed, blinded studies on biopsies from past ACR-368 Phase 2 trials on platinum-resistant ovarian cancer. The test demonstrated segregation of responders from non-responders across studies, including an ORR enrichment to 47% (p=0.01, Wilcoxon), and significant increase in progression free survival (p=0.006, CoxPH) in biopsies from Phase 2 trials at NCI. Acrivon has shown initial prospective validation of the AP3-based ACR-368 OncoSignature™ test in the ongoing, registrational-intent Phase 2b clinical trial in ovarian and endometrial cancers (NCT05548296), demonstrating clear segregation of RECIST responders in the OncoSignature™-positive (50% confirmed ORR in 10 patients) versus OncoSignature™-negative (0% ORR in 16 patients) arms (p-value=0.0038). 499

501

P-I-0347 Advancing clinical diagnostics and prognostics through proteomics: insights from core facility perspective Mohamed Ali Jarboui1, Vanessa Hollfoth2, Arslan Ali2, Stephan Singer2, Marius Ueffing2, Karsten Boldt1 1University Clinic Tübingen, Core Facility Medical Proteomics, Tuebingen, Germany 2University Clinic Tübingen, Tuebingen, Germany Proteomics has emerged as a pivotal tool in clinical applications, providing profound insights into disease mechanisms and aiding in developing diagnostic and prognostic strategies. Clinical proteomics spans from pre- clinical discovery to applied diagnostics, leveraging proteomic technologies to address clinical questions and inform medical decision-making. The robust capabilities of Liquid Chromatography-Mass Spectrometry (LC-MS), the rise of Data-Independent Acquisition (DIA) methods, and the development of advanced sample preparation techniques have revolutionized proteomics. These technological advancements, coupled with enhanced bioinformatics tools, have significantly broadened the accessibility and applicability of clinical proteomics. This makes it suitable for analyzing complex biological samples, such as formalin-fixed paraffin-embedded (FFPE) tissues, serum, plasma, and extracellular vesicles (EVs). Integrating machine learning techniques enhances the identification of disease- relevant features from clinical datasets, improving predictive models and enabling personalized treatment approaches. The future of biomarker research lies in increased utilization of non-invasive or minimally invasive liquid biopsies, or FFPEs already available from histopathological diagnosis, which can provide insights into disease mechanisms and drug resistance through multiple sampling. Non-invasive diagnostics are becoming increasingly important, offering viable alternatives to more invasive samples and allowing the detection of perturbations in cellular signaling pathways caused by genetic abnormalities, presenting a shift in protein composition that can serve as diagnostic signatures for various diseases. In this work, we showcase our core facility's collaborative efforts to support clinical applications and highlight the potential of proteomics investigations in personalized medicine. We emphasize the role of proteomics in patient stratification and therapy outcome prediction across diseases, including amyloidosis, cancer, and neurodegenerative disorders. With the integration of our proteomic workflows into clinical practice, we aim to support the development of diagnostic and prognostic tools that can significantly improve patient care. Furthermore, we highlight the possibilities that proteomics offers for clinical applications, ultimately contributing to better health outcomes. Moreover, we underscore the transformative potential of LC-MS-based proteomics in clinical settings, illustrating how advancements in proteomic technologies and collaborative efforts with pathologists in a clinical setup are paving the way for improved diagnostic and therapeutic strategies in modern medicine. 501

502

P-I-0348 Proteomic and proteogenomic characterization of triple negative breast cancer Henrik Johansson1, Akram Emdadi1, Mahshid Zarrineh1, Ioannis Siavelis1, Haris Babacic1, Eduardo Araujo1, Rui Branca1, Johan Staaf2, Janne Lehtiö1 1Karolinska Institute, Solna, Sweden 2Lund University, Department of Clinical Sciences, Lund, Sweden Henrik J. Johansson1, Akram Emdadi1, Mahshid Zarrineh1, Ioannis Siavelis1, Haris Babacic1, Eduardo Araujo1, Rui M. Branca1, Johan Staaf2, Janne Lehtiö1. of Oncology Sweden 1Department 2Division of Oncology, Department of Clinical Sciences, Lund and CREATE Health Strategic Center for Translational Cancer Research, Lund University, Lund, Sweden Pathology, Karolinska and Institutet, SciLifeLab, Solna, Triple negative breast cancer (TNBC) has the worst prognosis in breast cancer with limited therapy options. Here we investigate the phenotypic proteome level of TNBC using mass spectrometry (MS) based proteomics methods, with the aim to foster a better biological understanding of TNBC, identify subtypes and link them to therapy response for proteome precision medicine. The multi-omics population-based discovery cohort consists of 225 TNBC samples from SCANB in Sweden with WGS, RNA-seq, data independent acquisition (DIA) and TMT based quantitative MS based proteomics data. The TMT based proteomics data is generated using peptide fractionation using immobilized pH gradient-isoelectric focusing (IPG-IEF) before nanoLC-MS/MS, which enable searching a 6-reading frame translation of the human genome to identify novel protein coding regions and potential neoantigens. In the discovery cohort we have identified subtypes with different biological characteristics and clinical outcomes. For validation of the subtypes and linking them to drug response we are developing proteome-based classifiers based on the DIA data and are acquiring DIA data from clinical trials with chemo- and immunotherapy. 502

503

P-I-0349 Optimization of SAX-Based enrichment of plasma extracellular vesicles for high-throughput deep phenotyping of the heart failure syndrome Anna Maria Jozefowicz1,2, Han Byul Yoo1,2, Dana Hein1,2, Priyadarshini Tilak1,2, David Gomez-Zepeda2,3,4, Ute Distler1,2, Stefan Tenzer1,2,3,4 1Research Center for Immunotherapy (FZI), University Medical Center of the Johannes-Gutenberg University, Mainz, Germany 2Institute of Immunology, University Medical Center of the Johannes-Gutenberg University, Mainz, Germany 3Helmholtz Institute for Translational Oncology (HI-TRON), German Cancer Research Center (DKFZ), Mainz, Germany 4German Cancer Research Center (DKFZ) Heidelberg, Division 191, Heidelberg, Germany Introduction: Heart failure syndrome (HF), affecting approximately 15 million Europeans, ranks as the leading cause of hospitalization. HF profoundly diminishes quality of life and life expectancy, imposing a substantial burden on public health. Treating patients with HF remains a significant challenge due to limited understanding of a pathophysiology of the heterogeneous HF syndrome, limited effectiveness of existing prevention strategies and its progressive nature. The aim of the study is to develop high throughput, DIA- based, deep coverage plasma proteomics workflow for screening of patients, to suggest novel hypotheses about the molecular mechanisms of HF syndrome that explain how the identified biomarkers contribute to the disease. Methods: We established a SAX-bead based protocol for enrichment of plasma extracellular vesicles in 96-well plate format using a Biomek i7 liquid handling station to enable digestion throughput of 450 samples per week. Initial method evaluation was performed on smaller cohort of 144 diabetes patients treated with empagliflozin vs. placebo, examined three times during the phase of 12 weeks. We evaluated DIA- based acquisition workflows on Orbitrap Exploris 480 coupled to Ultimate 3000 and timsTOF Pro2 coupled to nanoElute2, aiming to provide deeper knowledge about proteomics signatures associated with HF. The data were processed DIA-NN using both spectral library and library-free approaches. Results and conclusion: Label-free DIA analysis of the plasma samples using 30 min method (50 SPD) resulted in the quantification of 1282 unique protein groups (n=2) in Orbitrap and 1269 in timsTOF Pro2 in the library-free approach with the overlap of 92% between two platforms. Our established method shows excellent reproducibility with the median CV < 15% and Pearson correlation of 0.98 between the digestion QCs. Using a gas-phase fractionation generated spectral library in the timsTOF Pro 2 platform increased the number of quantified unique proteins to 1631. The SAX -beads based automated protocol has been demonstrated to ensure deep coverage of plasma proteome, but also to be reproducible, cost-effective and usable in high- throughput for clinical applications. 503

504

P-I-0350 Urinary proteomic analysis reveals new molecular subtypes of lupus nephritis Cheng Fang1, Zhan Dongdong1, Tan Meng1, Sun Xiaojing1, Tan Ying1, Wang Yi1, Jun Qin1 1Beijing Pineal Diagnostics Co., Ltd., Research and Development Department(R&D), Beijing, China Lupus nephritis (LN) is a common complication of systemic lupus erythematosus that presents a high risk of end- stage renal disease. Clinically, the kidney biopsy is used to guide initial therapy in patients with LN but still unsatisfactory. Thus, non-invasive tools stratifying LC patients are urgently needed. Here, we collected 112 urine samples from LN patients and performed in-depth urine proteomics test. Unsupervised clustering distinguished four molecular subtypes (S1-4). S1 was associated with higher acute index (AI) and increased leukocyte infiltration, and showed a high response rate (83%) to prednisolone combined with Mycophenolate mofetil. In S2, keratins were significantly upregulated, indicating their role in cell structure maintaining. S3 displayed the activation of complement system and elevated chronic index. Importantly, Complement C5 was detected in 92% of S3 cases and exhibited higher expression compared with the other subtypes, suggesting the potential therapeutic utility of C5 inhibitors. S4 was characterized by metabolic abnormalities and lower treatment response rate (55%). The overexpression of SLC5A1 emerged as a distinctive hallmark for S4, its inhibitors might logistic regression model that can prove effective within this context. Further, we also established a distinguished LN patients with higher AI three urinary biomarkers (A2M, CORO1A and ART3), which was replicated in an independent validation set. In summary, noninvasive urinary proteomics provides valuable insight into the molecular characteristics for LN and reliable predictors for AI. (AUC = 0.935) using index Keywords: Urine proteome, Molecular subtypes, Biomarker, Logistic regression model References 1. Yu, Chen et al. Lupus nephritis: new progress in diagnosis and treatment.Journal of autoimmunity vol. 132 (2022): 102871. 2. Fan, Guanjie et al. Urine proteomics identifies biomarkers for diabetic kidney disease at different stages. Clinical proteomics vol. 18,1 32. 29 Dec. 2021. 504

505

P-I-0351 Multi-tiered proteome analysis displays the hyper-permeability of the rheumatoid synovial compartment for plasma proteins Sofia Kalaidopoulou Nteak1,2, Eva Maria Stork3, Danique van Rijswijck1,2, Rene E.M. Toes3, Albert Bondt1,2, Tom W.J. Huizinga3, Albert J. R. Heck1,2 1Utrecht University, Biomolecular Mass Spectrometry and Proteomics, Utrecht, Netherlands 2Netherlands Proteomics Center, Utrecht, Netherlands 3Leiden University Medical Center, Leiden, Netherlands Rheumatoid arthritis (RA) is characterized by synovial hyperplasia, cartilage/bone destruction and stiffness. RA affects the joints, the synovial membrane and consequently also the proteome of synovial fluid. Most of the RA studies are focusing on the role of T cells, but antibodies as well as other proteins may also play a role in joint destruction. As the proteomic signature of liquid biopsies can provide insights into the underlying (inflammatory) processes, we analysed using the protein composition of plasma and synovial fluid at three distinct levels of depth using mass spectrometry, namely the total proteome, the clonal IgG1 repertoire and the RA-related clonal IgG1 ACPA autoantibody repertoire. We compared paired synovial/plasma liquid biopsies from 9 individual patients diagnosed with RA and assessed the similarity in protein composition between plasma and synovial fluid within each patient at these three molecular levels. Intriguingly, at the total proteome level, the synovial fluid proteome aligned qualitatively and quantitatively with the plasma proteome from the same patient, independent of the size and/or other biochemical features of the observed ~300 proteins. Although dominated by plasma proteins, the synovial fluid contained additional lower abundant proteins primarily related to cartilage and several, mainly neutrophil-derived, cellular proteins. Strikingly, also the synovial IgG1 repertoire and ACPA autoantibody repertoire, monitored quantitatively and with clonal resolution, containing several hundred of distinctive clones, resembled very well the plasma IgG1 repertoires within the same patient. Our comprehensive multilayer proteomics data reveals not only that the synovial fluid in RA patients is a direct, unbiased, filtrate from the plasma, but also that, within the same patient, all the dominant abundant IgG1 (auto)antibody clones are equivalently present between the two. We argue that the huge infiltration of plasma into the synovium, as reported here, may reduce the normal lubricating function of synovial fluid and partially clarify the key phenotype of RA stiffness 505

506

P-I-0352 The tape-stripping procedure for the identification of protein N-termini within skin specimens Patrick Kaleja1, Andreas Tholey1 1IEM, Christian-Albrechts-Universität zu Kiel, Systematic Proteome Research & Bioanalytics, Kiel, Germany Minimal invasive sampling techniques are an essential tool for all types of biological analyses. For the biopsy of the skin, the tape-stripping procedure has been established as a reasonable technique to sample the stratum corneum, the outermost section of the epidermis. Tape strips are polymer-based adhesive sheets, which are iteratively applied to the volunteer"s skin in order to remove layers of extracellular matrix and embedded cells. While the technique has many benefits, it provides certain analytical challenges, especially for protein-centered approaches. Extraction of the proteins requires to overcome non-covalent interactions with the adhesive, and highly cornified cells need to be lysed. We previously established a protocol, which utilized detergent-based buffers in combination with the SP3 protocol [1] to tackle this task and enabled the identification of hundreds of proteins from single strips [2]. We now expanded the analytics of tape-stripped specimens with the application of an N-terminomics strategy. For this, we combined our established procedure with the high-efficiency undecanal-based N-termini enrichment (HUNTER) protocol [3], and analyzed isolated protein N-termini by LC-MS methods. The technique enables the characterization of hundreds of termini from limited sample material and indicates a comprehensive processing of proteins within the stratum corneum. [1] Hughes, C.S., Moggridge, S., Müller, T., Sorensen, P.H., et al., Nat Protoc 2019, 14, 68. [2] Kaleja, P., Emmert, H., Gerstel, U., Weidinger, S., Tholey, A., Journal of Proteomics 2020, 217, 103678. [3] Weng, S.S.H., Demir, F., Ergin, E.K., Dirnberger, S., et al., Molecular & Cellular Proteomics 2019, 18, 2335. 506

507

P-I-0353 Salivary chemical barrier proteins in oral squamous cell carcinoma – alterations in the defense mechanism of the oral cavity Gergő Kalló1, Petra Magdolna Bertalan1, Ildikó Márton1, Csongor Kiss2, Éva Csősz1 1University of Debrecen, Department of Biochemistry and Molecular Biology, Debrecen, Hungary 2University of Debrecen, Division of Pediatric Hematology-Oncology, Department of Pediatrics, Debrecen, Hungary Introduction: Oral squamous cell carcinoma (OSCC) is one of the most frequent types of head and neck cancer and among the European countries; Hungary has the highest rate of incidence and mortality of OSCC. Despite the genetic and environmental risk factors, OSCC is also associated with microbial infections and/or dysbiosis. Saliva serves as the chemical barrier of the oral cavity since the most abundant salivary proteins belong to antimicrobial and immunomodulatory protein (AMP) family. Aims: Considering that saliva contains many chemical barrier proteins and the fact that OSCC can alter the protein composition secreted into saliva, our aim was to analyze the effect of OSCC on the salivary chemical barrier proteins. Materials and methods: Publicly available proteomics datasets regarding the analysis of salivary proteins from patients with OSCC and controls were collected and examined in order to identify differentially expressed chemical barrier proteins. Network analysis and gene ontology (GO) classification of the differentially expressed chemical barrier proteins were performed as well. Results: One hundred and twenty-seven proteins showing different expression pattern between the OSCC and control groups were found. Protein-protein interaction networks of up- and down-regulated proteins were constructed and analyzed. The main hub proteins (IL-6, IL-1B, IL-8, TNF, APOA1, APOA2, APOB, APOC3, APOE, and HP) were identified and the enriched GO terms were examined. Our study highlights the importance of the chemical barrier of saliva in the development of OSCC. This research was funded by NKFIH K143021 and GINOP-2.3.3-15-2016-00020. 507

508

P-I-0354 Exosomal protein biomarkers for early prediction for dengue severity in pediatric population using quantitative proteomics approach Ravi Kant1, Anushka Singh1, Pratik Pathade1, Dinesh Kaul1, Anil Sachdev1, Nirmal Kumar Ganguly1, Rashmi Rana1 1Sir Ganga Ram Hospital, New Delhi, India Background: Globally, the incidence of pediatric dengue has been increasing on large scale, resulting in high number of morbidity and deaths. In 2009, WHO classified dengue into three different categories i.e. dengue without warning sign, dengue with warning sign and severe dengue. The issue with the current classification is that classification is not specific for pediatric population and also the classification is so complex that sometimes it is not possible for clinicians to classify the patient dengue category on the basis of symptoms of the patient, which results in improper diagnosis and management of the dengue cases. Material and Methods: In this study, exosomal proteomics study was performed using LC/MS technique. The study population included 48 samples (12 samples from each dengue category and 12 samples from healthy individuals) for discovery phase and 200 samples (50 samples from each dengue category and 50 samples from healthy individuals) for verification phase. Exosomes were isolated from plasma of dengue patients as were from healthy individuals for control and exosomes were characterized by using different characterization techniques such as western blotting, Nanoparticle tracking analysis, dynamic light scattering and flow cytometry. After characterization quantitative proteomic study was performed. Results: Comparative data analysis was performed by combing the Mass-spectrometric data with different machine learning algorithms and analysis revealed the identification of 529 protein groups. Statistical approach such as student, limma, Anova were applies to identify the differentially altered significant proteins which identified 96 significant proteins. Further analysis found that 26 proteins were following the trend of sequential upregulation or downregulation in all the three defined categories of dengue with respect to control samples. Out of these 26 proteins, 12 number of proteins were found to be upregulated and 14 Number of proteins were found to be downregulated. 6 proteins were carried out for verification phase on the basis of In-silico analysis which included Gene ontology analysis and KEGG pathway analysis. On the basis of verification phase results and correlation of these results with clinical details of the patients,3 proteins are proposed as potential biomarkers for early prediction of dengue severity in pediatric population. Conclusion: The utilization of exosomes for the diagnosis of various infectious disease has gained significant attention due to its potential for early detection, disease progression tracking, non-invasiveness, and cost- effectiveness. After further verification of these outcomes by different approaches these markers could be potentially used for early prediction of dengue severity which could allow for reduction in morbidity and mortality of pediatric dengue patients. Also, this study could help in proper classification of Dengue cases on the basis of these potential proteomic biomarkers. 508

509

P-I-0355 Benchmarking sample preparation and data analysis for clinical proteomics of human biofluids Oliver Kardell1, Till Kindel1, Frank Rolfs1, Samira Vautrin1, Barbara Kracher1, Thomas Gronauer2, Christine von Toerne2, Andreas Tebbe1, Stefanie Hauck2, Carleen Kluger1, Giada Marino1 1Evotec GmbH, Neuried, Germany 2Helmholtz Zentrum München, Munich, Germany Liquid biopsies like plasma, cerebrospinal fluid (CSF), or urine are a less invasive and inexpensive alternative to tissue biopsies. Additionally, these body fluids are known to contain a multitude of known and yet to be revealed biomarkers that can be detected using mass spectrometry (MS) based proteomics. There are several distinct advantages of this technology: (I) it gives unbiased access to the biofluid proteome, (II) it is species independent, (III) it can reveal information about isoforms and post-translational modifications and (IV) it does not require availability of antibodies. Furthermore, MS based proteomics has been rapidly evolving over the last years due to the development of new generations of MS instruments such as timsTOF or Thermo Astral, advancements in sample preparation like nanoparticle-based protein enrichment, as well as novel AI based tools for data processing. The continuous progress results in increased depth and throughput of clinical proteomics. However, this highly dynamic nature of the field can also be a drawback, as applications of mass spectrometry- based proteomics for larger clinical trials is often hindered by a lack of standardization across laboratories and studies. To address this problem and to enhance the integration of MS-based proteomics into medical research the CLINPSECT-M consortium and its MS laboratories conducted a round robin study on clinical specimens such as human plasma and CSF [1]. Here, the data from the round robin study is compared to internal benchmark studies on human biofluids that were performed at Evotec. In addition, while the focus of the round-robin study was to harmonize workflows for undepleted plasma and CSF and track interlaboratory reproducibility, the Evotec study also measured longitudinal platform stability and shows how nanoparticle-based proteomics using the ProteographTM can provide deeper coverage of the plasma proteome and at the same time overcome some of the challenges associated with undepleted workflows. [1] Multicenter Collaborative Study to Optimize Mass Spectrometry Workflows of Clinical Specimens Oliver Kardell, Christine von Toerne, Juliane Merl-Pham, Ann-Christine König, Marcel Blindert, Teresa K. Barth, Julia Mergner, Christina Ludwig, Johanna Tüshaus, Stephan Eckert, Stephan A. Müller, Stephan Breimann, Pieter Giesbertz, Alexander M. Bernhardt, Lisa Schweizer, Vincent Albrecht, Daniel Teupser, Axel Imhof, Bernhard Kuster, Stefan F. Lichtenthaler, Matthias Mann, Jürgen Cox, and Stefanie M. Hauck, Journal of Proteome Research Article ASAP, DOI: 10.1021/acs.jproteome.3c00473 509

510

P-I-0356 Providing an analytical platform for state-of-the-art absolute apolipoprotein quantification in human blood plasma using prm-PASEF Stephanie Kaspar-Schoenefeld1, Christian Albers1, David Kotol2 1Bruker Daltonics GmbH & Co. KG, Bremen, Germany 2ProteomEdge, Stockholm, Sweden Introduction: With proteomics making a move into precision medicine, there is an increasing pressure on quantitative data quality in studies of blood profiles. Despite the endeavors to develop new data-analysis pipelines, options for improving quantification results early on and prior to the sample preparation are limited. Heavy labelled protein standards are promising solution to this challenge with potential to ensure precise quantification when combined with targeted mass spectrometry. Apolipoproteins are proteins that bind to lipids to form lipoproteins. They have broad diagnostic importance in cardiovascular risk assessment, identification of lipoprotein abnormalities, monitoring treatment efficacy and genetic disorders. We utilized prm-PASEF a powerful targeted mass spectrometry technique to develop a reference target assay for absolute quantitation of apolipoproteins in human blood plasma. Methods: The ApoEdge panel (ProteomEdge, Sweden), consisting of heavy Lys and Arg 13C and 15N quantitative recombinant protein standards (qRePS) covering all 18 human apolipoproteins, was either digested and analyzed directly or spiked-in to non-depleted plasma, digested and used for absolute quantification. Tryptic digests were loaded onto 25cm C18 column (75μm, 1.6μm, Aurora, IonOpticks) using nanoElute HPLC (Bruker) coupled to timsTOF HT (Bruker) via a CaptiveSpray source (Bruker) using a 35-min ACN gradient. Target method generation was done by analyzing standards using dia-PASEF, followed by library and panel generation in SpectroDive (v11, Biognosys). Generated prm-PASEF panel had a retention time tolerance set to 2 min and ion mobility tolerance set to 0.05 1/K0. Subsequent targeted data processing was done in SpectroDive. Results: In the presented study we developed a prm-PASEF assay for analysis of Apolipoproteins in human plasma background. Targeted quantitation was based on 19 qRePS spiked into the sample of interest, which allowed determination of absolute concentrations. Assay development was done by analyzing pure labelled samples without background using dia-PASEF for library generation. In total, 225 peptides from all 18 panel proteins were identified using a 35-min gradient. Panel refinement based on initial measurements done in plasma background resulted in a final panel of 114 targets to be measured. In prm-PASEF, several precursors could be selected within one IMS scan and consequently fragmented in the collision cell. For the developed assay up to 3 precursors were scheduled per TIMS frame. Data processing of 8 replicates revealed excellent reproducibility with a median coefficient of variation of 7% for peptides spiked in complex plasma background. We further estimated the concentration of the endogenous proteins by single-point calibration with the reference spike-in standard. In the next step, we will further expand the assay to 175 clinically relevant proteins. 510

511

P-I-0357 Streamlined extraction of proteins from clinical Trizol samples via aggregate capture on magnetic beads Aigerim Kassym1,2, Mario M. Müller1,2, Denys Oliinyk1,2, Felix Schneidmadel1,2, Jonas Marx1,2, Caroline Tschernjawski1,2, Maximilian Böhme3,2, Jenny Rinke3,2, Andreas Hochhaus3,2, Thomas Ernst3,2, Florian Meier-Rosar1,2 1Jena University Hospital, Functional Proteomics, Jena, Germany 2Comprehensive Cancer Center Central Germany, Jena, Germany 3Jena University Hospital, Innere Medizin II, Hämatologie und Onkologie, Jena, Germany Tumor biopsies are routinely sequenced and archived, representing a rich source of specimens for biomedical research. However, these samples are typically preserved in buffers optimized for nucleic acid analysis such as phenol and guanidine isothiocyanate buffer ("Trizol"), which presents challenges for the extraction of proteins for mass spectrometry (MS) analysis. This is particularly true for small input amounts and large numbers of samples. To unlock this resource for clinical proteomics, we present a streamlined workflow for Trizol-lysed cells via protein aggregate capture (PAC/SP3) on magnetic beads. Human epithelial cervix carcinoma HeLa, human histiocytic lymphoma U937 and human chronic myelogenous leukemia K562 cells were lysed in Trizol or 4% SDC to establish, optimize and benchmark the Trizol-PAC workflow. Proteins were aggregated on SIMAG-carboxyl (Chemicell) beads in 96-well format and digested with trypsin/LysC in 4% SDC buffer. Purified peptides were analyzed via nanoflow liquid chromatography on a 15 cm homemade column coupled to a trapped ion mobility-quadrupole time-of-flight mass analyser (Bruker timsTOF- HT) operated in dia-PASEF acquisition mode. Raw files were processed in DIA-NN v1.8.1 and analyzed in Python scripts. First, we assessed the protein recovery for the standard PAC protocol with different concentrations of HeLa cells lysed in Trizol, which did not provide reproducible yields from the phenolic phase. To overcome this challenge, we carefully adjusted solvents, pH and bead mass to aggregate proteins on beads. Next, we tested various washing steps to efficiently remove phenol traces. Increasing the number of washing steps lowered miscleavage rates in K562 cell lysates from 38% to 25% on average, hinting at a negative effect of Trizol components on the digestion efficiency. Having established a suitable bead-based workflow for Trizol samples, we benchmarked it against established workflows in our laboratory. This resulted in an average number of ~7000 proteins identified with the Trizol-PAC protocol, on par with lysis buffers optimized for MS. We found a high quantitative reproducibility, minimal variability within workflow replicates and Pearson correlation coefficients >0.92 between workflows. While we observed a large overlap in protein identifications, GO cellular compartment analysis at the quantitative level revealed a slightly higher abundance of cytosolic and a lower abundance of membrane proteins in the Trizol-PAC workflow as compared to in-solution digestion, suggesting a somewhat less efficient aggregation of hydrophobic proteins on magnetic beads from Trizol. Taken together, we demonstrate a scalable and efficient protocol for MS-based proteomics from Trizol samples. We envision that this will allow to seamlessly integrate proteomics analysis in clinical studies, and open up for parallel analysis of nucleic acids and proteins from the same samples. 511

512

P-I-0358 Identification of novel prognostic protein biomarkers from maternal serum of women with recurrent pregnancy loss Drosos Katsavelis1,2, Salwan Al-Nasiry3, Judith Bons4, Marieke van der Hart5,2, Hjalmar Permentier6, Peter Horvatovich6, Thomas Cremers5,2 1University of Groningen, Analytical Biochemistry, Groningen, Netherlands 2Quantall, Groningen, Netherlands 3Maastricht University, Department of Obstetrics and Gynaecology, Maastricht, Netherlands 4Maastricht University, Central Diagnostic Laboratory, Maastricht, Netherlands 5University of Groningen, Analytical Biochemistry, Groningen, Netherlands 6University of Groningen, Analytical Biochemistry, Groningen, Netherlands Background: The underlying cause of recurrent pregnancy loss (RPL) is often not understood. Abnormal receptivity of the endometrium or chromosomal aberrations of the embryo are the usual determining factors in the cascade leading to early miscarriages. The risk of miscarriage declines with ongoing pregnancy, however the anxious wait of couples until the end of the first trimester is an important cause of stress which can be minimized if reliable prediction models exist for fetal viability. Methods: Serum samples were obtained at gestational age of 5-7 post-menstrual weeks from 8 women that suffered subsequently a miscarriage within the trimester of pregnancy, and 10 healthy control subjects (HC) with continuing viable pregnancy. Using bottom-up untargeted proteomics with LC-MS/MS in data-independent acquisition (DIA) mode, maternal serum samples were analyzed in order to investigate their proteome profiles. We utilized Spectronaut 18 for protein quantification and a combination of analytical parameters, including statistical significance, fold change, number of unique peptides detected, number of missing values and the coefficient of variation of the candidate biomarkers among the QC samples to detect reliable prognostic biomarkers which were further evaluated using receiver operating characteristic curve (ROC) curve analysis. Results: In total, 411 proteins from 3698 peptides were quantified in the studied serum samples. Among them, 13 proteins were significantly downregulated in RPL cases and 2 proteins were significantly upregulated. Two low-abundance proteins involved in extracellular matrix remodeling and in antioxidant defense mechanisms were selected after filtering for further analysis. The combination of the two proteins showed excellent diagnostic accuracy (AUC=0.95), while individual AUC values were 0.9 and 0.763, respectively. Conclusion: Shotgun proteomics revealed that two low-abundance proteins have the potential to serve as prognostic biomarkers for the early screening of RPL. These promising findings will be further validated in an independent cohort by enzyme-linked immunosorbent assay (ELISA), providing strong evidence for the reliability of the new biomarkers discovered. Fig. 1 (optional) 512

513

P-I-0359 Profiling the blood proteome in autoimmune disease using proximity extension assay Josefin Kenrick1, Fredrik Edfors1, Mathias Uhlen1, Peter Nilsson1, Sofia Bergström1, Elisa Pin1 1KTH Royal Institute of Technology, Protein Science, Solna, Sweden Autoimmune diseases are heterogeneous diseases characterized by dysregulation of the immune system. They often result in chronic inflammation and damage to overall health. Due to the complex nature of these diseases, they are frequently difficult to diagnose and present with comorbidities which increase mortality risk. There is a pressing need for the discovery of novel biomarkers to facilitate early diagnosis, stratification and treatment evaluation of patients within these disease populations. In this study, six autoimmune diseases were selected for plasma profiling as part of the Human Disease Blood Atlas program, including multiple sclerosis (n=234), myositis (n=209), rheumatoid arthritis (n=84), scleroderma (n=100), Sjögren"s syndrome (n=99), and systemic lupus erythematosus (n=97). Additionally, subgroup analysis of these diseases has been performed for multiple sclerosis and myositis. In total, 823 plasma samples were analysed using the Olink Explore 1536 platform, a highly sensitive and multiplexed antibody-based technology, resulting in expression data of 1463 unique proteins. Differential expression and supervised machine learning analysis with glmnet lasso identified potential prognostic biomarkers; some of these have previously been found to be associated with autoimmune disease, and others are novel. Interestingly, certain proteins exhibited patterns suggesting patient subgroups separate from what is identified in the clinic. Pathway analysis provides further insights into the underlying biological processes and molecular interactions involved in the pathogenesis of these autoimmune disorders. Many identified proteins are involved in pro- inflammatory response and have suggested immune system functions. A portion of identified proteins have strong associations with cancer as well as infectious disease. In summary, this study provides a comprehensive, exploratory analysis with the aim to identify distinct protein profiles both within and across six autoimmune diseases and their subgroups. 513

514

P-I-0360 Impact of non-aqueous fluid flow on protein profile in glaucoma drainage device implants Alka Khera1, Surinder Singh Pandav1, Faisal TT1 1Postgraduate Institute of Medical Education and Research, Ophthamology, Chandigarh, India Glaucoma surgery, particularly trabeculectomy, has high failure rates. Failure is typically defined by postoperative intraocular pressure (IOP) exceeding the target IOP. Glaucoma drainage devices (GDDs) also have high failure rates, often leading to a thick fibrous capsule. A new model, the "CERA" implant, has been used to study bleb capsule porosity in rabbits, showing that NZW rabbits developed thick capsules with low porosity at 4 weeks. The study aims to understand the effect of non-aqueous fluid (BSS) flow on the proteome responsible for capsular porosity. Methodology: Twenty-eight days after tube surgery, animals were anesthetized, and the implant's capsule porosity was tested by the flow of BSS into the capsule measured at 12 mm Hg for 5 minutes, followed by a 30- minute fluid challenge at the same pressure. The study utilized Metaboanalyst to identify proteins and analyze their expression patterns at different times. A heat map was used to visualize the data. Results: Many proteins showed upregulation at the 24 and 48-hour time points. These proteins were involved in oxygen transport, regulating calcium homeostasis, blood coagulation, and immune and defense responses. Conclusion: The study revealed a time-dependent control of proteins associated with immunity and defense. The findings highlight the dynamic nature of the immune response and the complex interactions between proteins and immune response pathways. These proteins maintain immunological processes, and calcium homeostasis, and limit excessive bleeding and wound healing. The study also suggests that the immune system is important for tissue regeneration and repair. 514

515

P-I-0361 Proteomic strategies uncover novel aging and age-related disease signatures in the human post-menopausal ovary Christina King1, Shweta S. Dipali2, Joanna Bons1, Hannah Anvari2, Jacob P. Rose1, Mark Watson1, Judith Campisi1, Francesca E. Duncan1,2, Birgit Schilling1 1Buck Institute for Research on Aging, Schilling Lab, Novato, CA, United States 2Northeastern University, Department of Obstetrics and Gynecology, Chicago, IL, United States Reproductive decline begins when women are in their mid-thirties, increasing the risk for infertility, miscarriages, and ovarian cancer. This is characterized by molecular changes within the ovarian microenvironment, including fibrosis, accumulation of collagen, and loss of hyaluronan. Cellular senescence is a pro-inflammatory stress response that results in the production of a Senescence-Associated Secretory Phenotype (SASP). The presence of senescent cells is consistent with the fibro-inflammation signature in the aging ovary, thus giving rise to the importance of studying senescence in this tissue. Here, the postmenopausal ovary was studied to identify 1) the ovarian SASP in an induced model of senescence and 2) to understand age-associated changes that may contribute to the increased incidence of high-grade serous ovarian cancer (HGSOC) in non-Hispanic Black women, which disproportionally affects this population.Explant cultures of cortex and medulla regions from one donor were treated with doxorubicin (doxo) exposure to induce senescence or DMSO as control. Conditioned media from human explants were collected and processed by label-free quantitative proteomics strategies. Upon senescence induction by doxorubicin, 135 and 184 SASP factors were significantly-altered in cortex and medulla explants, respectively. Upregulated biological processes in both the cortex and medulla explants are related to chromatin remodeling and assembly. Significantly-altered proteins in both cortex and medulla explants include Serpin H1, Tenascin, and Periostin, which are well established SASP factors. To understand ovarian aging in the context of ovarian cancer risk, human ovarian cortex and medulla tissues (N=42) were acquired prospectively from six non-Hispanic black (NHB) and five non-Hispanic white (NHW) non-ovarian cancer post-menopausal patients. Ovarian tissue was prepared for analyses by having native and decellularized conditions and samples were prepared using label free- quantitative proteomics strategies. In this analysis, 4,257 quantifiable protein groups with ≥2 unique peptides were detected in the native cortex (NHB/NHW). Significantly-altered proteins include SASP factors Serpin H1, Tenascin, and Periostin. Upregulated biological processes are related to metabolism whereas downregulated processes are related to coagulation, ECM organization, and wound healing. Several significantly-altered SASP proteins in decellularized cortex are also present in the native condition, thus confirming the preservation of the extracellular matrix. Up- and downregulated biological processes in the decellularized are consistent with the native condition. These analyses provide great insight into the ovarian SASP and how the SASP may contribute to the higher incidence of HGSOC in non-Hispanic Black women. With this information, the mechanisms that underlie ovarian aging will be better understood and potential therapeutics may be developed to improve health outcomes. 515

516

P-I-0362 Proteomic profiling of circulating tumor cells for precision oncology Melissa Klingeberg1, Stephanie Staudte2, Ingeborg Tinhofer-Keilholz2, Fabian Coscia1 1Max Delbrück Center for Molecular Medicine, Berlin, Germany 2Charité – Universitätsmedizin Berlin, Berlin, Germany Shedding of circulating tumor cells (CTCs) from solid tumors into the bloodstream contributes to metastasis and is estimated to be responsible for 90 % of cancer-related deaths. The scarcity of these cells makes their isolation and analysis a particular challenge. Until now, genomic and transcriptomic methodologies have characterized these rare cells, revealing great cellular heterogeneity. However, proteomic analyses of CTCs have only recently become in reach due to recent advances in ultrasensitive mass spectrometry based proteomics. Gaining comprehensive proteomics data from CTCs is essential for a holistic understanding of the mechanisms of metastasis, comprehensive cell characterization, and the identification of novel treatment targets. We established a first-of-its-kind microscopy-guided ultrasensitive LC-MS workflow to detect and profile CTCs. The combination of microfluidic cell enrichment, cytospin centrifugation with immunofluorescent staining and high- resolution imaging enables optimal differentiation of tumor cell phenotypes and lymphocytes. Using a cancer- specific antibody panel allowed us to identify various tumor subpopulations to capture different cellular states undergoing epithelial-mesenchymal transition (EMT). This approach goes beyond assessing only epithelial cells, which is insufficient to capture high phenotypic tumor heterogeneity. Further, we combined imaging data with the deep visual proteomics workflow employing single-cell laser microdissection followed by robotic sample preparation and mass-spectrometry based low-input proteomics. We will apply our unique workflow to analyze CTCs from liquid biopsies from patients with locally advanced head and neck squamous cell carcinoma. Studying molecular properties on the protein level will supplement and expand the knowledge about CTCs almost exclusively from genomic and transcriptomic perspectives to unveil valuable means of metastatic mechanisms. We anticipate that our findings will provide valuable insights into disease progression and may have implications for the development of novel targeted therapies for anti-metastatic cancer treatment. 516

517

P-I-0363 Nanoparticle-based proteomics analysis of human biofluid samples from different patient cohorts Carleen Kluger1, Giada Marino1 1Evotec GmbH, Clinical Proteomics, Neuried, Germany The advance of nanoparticle-based technologies (e.g. ProteographTM) in combination with high-end mass spectrometry (MS) has increased the depth of proteome coverage for biofluids like plasma, serum or cerebrospinal fluid (CSF), making it possible to routinely analyze 2.000-3,000 proteins per individual biofluid sample. Over the last two years we have extensively used this technology at Evotec to screen multiple different patient cohorts, each containing up to a few hundred serum or plasma samples from patients with very different disease backgrounds, including patients with multiple co-morbidities and a dedicated cohort of healthy subjects that were hospitalized over-night to ensure maximum control of biofluid and data collection. While in the past, most activities in the field of Clinical Proteomics were limited to a specific patient cohort in a clearly defined disease context, the next challenge is to integrate proteomics signatures from multiple cohorts into a unified framework. Here, we show how this can be achieved by minimizing technical variability during sample collection, sample preparation and MS measurement using highly automated nanoparticle-based proteomics approaches. In addition, we present how PanHunter – a multi-omics tool optimized for collaborative data analysis – can support management of patient metadata as well as integration of multiple studies. 517

518

P-I-0364 Proteomic alterations in early-stage breast cancer: insights from tumor-adjacent normal tissues analysis Po-Hsin Kong1,2, Yu-Chun Tsao1, Yen-Chun Huang1, Ling-Ming Tseng3,4, Chi-Cheng Huang3,5, Chen-Chung Liao6 1Marker Exploration corp., Taipei, Taiwan 2National Sun Yat-sen University, Institute of Biomedical Sciences, Kaohsiung, Taiwan 3Taipei Veterans General Hospital, Division of Breast Surgery, Department of Surgery, Taipei, Taiwan 4School of Medicine, College of Medicine, National Yang Ming Chiao Tung University, Taipei, Taiwan 5College of Public Health, National Taiwan University, Institute of Epidemiology and Preventive Medicine, Taipei, Taiwan 6National Yang Ming Chiao Tung University, Cancer and Immunology Research Center, Taipei, Taiwan Background: Molecular characterization, such as genomic and transcriptomic profiling, has been incorporated into clinical practice for classifying and selecting treatments for breast cancer patients. However, at the protein level, conventional immunohistochemistry (IHC) is limited to analyzing a specific protein at a time, whereas untargeted proteomic analysis offers a complementary approach to IHC, revealing potential insights into cancer biology and therapeutic targets. Methods: Paired tumor and adjacent normal tissue samples were collected from 17 breast cancer patients with invasive ductal carcinoma staged I or II. Nanoflow liquid chromatography-high resolution mass spectrometry- based proteomic techniques and label-free quantitation were employed for protein profiling. The differences between tumor and adjacent normal portions were revealed by paired analysis, and the differences in the expression levels of tumor across samples were adjusted by the adjacent normal portions. Differentially expressed proteins (DEPs) were identified by at least 2-fold changes with p-values of paired t-test smaller than 0.05. The functional implications were explored using Gene Ontology (GO) analysis. Results: More than eight thousand proteins were identified at a false discovery rate of 1 %, and 3,885 DEPs were found between tumor and adjacent normal tissues. GO analysis revealed that the tumor tissues were highlighted in the processes in gene expression, pathways related to cadherin binding, and functions involved in DNA repair, characterizing the cancerous nature in the alterations of focal adhesion, cellular motility, and the evolvement with genomic instability. Additionally, in the comparison of tumor tissues across patients, upregulations of programmed cell death and mismatch repair protein families were observed in some cases, implying potential benefits from immunotherapies. Conclusion: This study demonstrates the value of untargeted proteomics in complementing IHC for comprehensive understanding of early-stage breast cancer. The proteins and pathways identified offer valuable insights into cancer biology and potentially improve diagnostic and therapeutic strategies. 518

519

P-I-0365 Ubiquitin: characterization of a host cell protein covalently attached to a monoclonal antibody product by LC- MS/MS Regina Kufer1, Vincent Larraillet1, Sabrina Thalhauser1, Tobias Graf1, Manuel Endesfelder1, Stefanie Wohlrab1 1Roche Diagnostics, Penzberg, Germany Host cell protein (HCP) characterization is a crucial quality parameter for biotherapeutic drug safety and stability. With a liquid chromatography-tandem mass spectrometry (LC-MS/MS) approach, we identified ubiquitin in ultrafiltration/diafiltration (UF/DF) pools of one of our monoclonal antibody (mAb) products. Since ubiquitin occurs physiologically as a post-translational modification (PTM) involved in many cellular functions, we suspected the possibility that if identified as an HCP, it may occur as a covalent modification on the mAb. In fact, in this study we characterized and quantified the ubiquitin modification on the Fc domain of mAbX by data dependent acquisition (DDA) and data independent acquisition (DIA) – MS workflows. Covalent binding and site localization were confirmed by identifying a characteristic diglycine motif on the modified peptide. Initially observed reduced detectability of ubiquitin in samples prepared with native digestion was attributed to impaired digestion and subsequent removal along with the mAb in the precipitation step. Our work has contributed to a better understanding of ubiquitin as an HCP considering its specific features such as occurrence in different topologies and provided insight into how covalent binding to a drug product can affect its identification by MS when native digestion conditions are used. Fig. 1 (optional) 519

520

P-I-0366 Targeted proteomic analysis for monitoring of multiple myeloma progression Rudolf Kupcik1, Ales Syrovy1,2, Katerina Hrochova3, Marie Vajrychova1, Jakub Radocha4 1Univeristy Hospital Hradec Kralove, Biomedical Research Centre, Hradec Kralove, Czech Republic 2University of Pardubice, Faculty of Chemical Technology, Pardubice, Czech Republic 3University Hospital Hradec Kralove, The Fingerland Department of Pathology, Hradec Kralove, Czech Republic 4Charles University and University Hospital Hradec Kralove, 4th Department of Internal Medicine – Hematology, Faculty of Medicine on Hradec Kralove, Hradec Kralove, Czech Republic Background: Multiple myeloma (MM) is a cancer of plasma cells that produce monoclonal immunoglobulins (mAb). Despite the development of new therapeutics, MM is associated with a high probability of persistent minimal residual disease (MRD) and relapse. Relapse in MM is driven by clonal heterogeneity and evolution, which can lead to the selection of drug-resistant malignant plasma cell (PC) clones. Monitoring of these residual PC clones is currently limited due to the invasiveness of bone marrow biopsy and the low sampling frequency. The primary objective was to test sensitive method for detection of malignant clones based on mAb analysis in human serum to monitor MM progression. Methods: Our approach utilizes human serum or selectively isolated mAb from this serum as an initial material. Subsequently, the samples are digested using proteolytic enzymes and analysed using a targeted approach focused on the variable (CDR3) region of mAb. The analysis is performed using nanoLC separation hyphenated to a high-resolution mass spectrometer (Exploris 480, Thermo Scientific). Specific sequence of CDR3 was determined using sequencing of the corresponding DNA regions from myeloma cells. Patient samples before and after treatment were included in this study. The efficacy of the method was also evaluated on mAb standards spiked into human sera. Results: The analytical workflow demonstrated the ability to detect mAb in serum with an exceptionally low limit of detection (≤5 mg.L-1). The analysis of digested preisolated mAb using targeted nanoLC-MS/MS detection proved to be more effective than the same analysis of crude human sera. Consequently, the method is deemed suitable for the sensitive detection of HC corresponding to malignant clones. The high sensitivity of the workflow allows for the sensitive monitoring of patients after treatment when analysis of the unique CDR3 region prior to MRD is conducted. Conclusions: These findings permit the careful monitoring of patients in short time intervals and the detection of disease progression. 520

521

P-I-0367 Proteomic profiling of preeclampsia Magdalena Kuras1, Axel Broman1, Charlotte Welinder2, Lena Erlandsson3, Thomas Laurell1, Stefan R Hansson3,4, Melinda Rezeli1 1Lund University, Department of Biomedical Engineering, Lund, Sweden 2Lund University, Division of Mass Spectrometry, Department of Clinical Sciences Lund, Lund, Sweden 3Lund University, Division of Obstetrics and Gynecology, Department of Clinical Sciences Lund, Lund, Sweden 4Lund University, Department of Obstetrics and Gynecology, Skåne University Hospital, Lund, Sweden Preeclampsia (PE) occurs in about 5% of all pregnancies. It is a complex disease characterized by sudden-onset hypertension, and it is harmful to the mother and the growing baby, accounting for 40% of fetal deaths worldwide. The overall objective of this project is to identify molecular subtypes of PE and to further apply these subtypes to liquid biopsies, including plasma and isolated placental syncytiotrophoblast-derived extracellular vesicles, enabling the identification of biomarker panels for PE prediction and early diagnosis. For tissue proteomics, a pilot cohort of 24 placentas (eight early- and eight late-onset PE and eight controls) was processed according to our published method (Kuras et al., 2018). Samples were acquired using two protocols: a more standardized one and one with longer processing times and unspecified collection sites. Proteomic analysis was performed using DIA–PASEF on an Evosep LC coupled to a timsTOF HT mass spectrometer. Following off-line fractionation of pooled placenta samples, a spectral library was built in FragPipe, and the raw data was searched using DIA-NN. Further analyses were performed in Perseus and RStudio. Extracellular vesicles were isolated on an acoustic trapping platform, described in detail previously (Broman et al., 2021). Using the acoustic trap, extracellular vesicles (EVs) were effectively isolated from 10 ml of plasma. A project-specific spectral library consisting of 9,917 proteins was built, of which 7,202 proteins were quantified in the 24 samples, 6,700–7,000 per sample. Sample collection and storage-related alterations in proteome integrity were revealed in the data analysis. We observed a significantly lower level of mitochondrial and ribosomal proteins in the compromised samples. In contrast, acute and chronic inflammation and hemoglobin protein chains showed higher expression, emphasizing the importance of the chain of custody during sample collection and storage for reliable protein quantification. When investigating the samples collected in a standardized manner (stored at 4°C and processed within 4 h), early-onset PE differed more from late-onset PE and controls and displayed upregulation of proteins related to inflammation, hypoxia, and metabolism. In contrast, extracellular matrix and translation-related proteins showed lower expression levels in early-onset PE. A workflow enabling high-throughput, large-scale proteomic characterization of placenta tissue was developed and successfully applied to a small cohort in which we, to the best of our knowledge, identified and quantified more proteins compared to previously published studies. Moreover, this study provides insight into the importance of sample collection and storage. Additionally, the successful isolation of EVs from minute plasma volumes enables the utilization of large cohorts of biobank samples. 521

522

P-I-0368 Quantitative assessment of multiple pathogen exposure, autoimmunity and immune dynamics at scale Joshua LaBaer1, Femina Rauf1, Ching-Wen Hou1, Ji Qiu1, Vel Murugan1, Yunro Chung1, Huafang Lai1, Deborah Adam1, D. Mitchell Magee1, Guillermo Trivino Soto1, Milene Peterson1, Karen S. Anderson1, Stephen G. Rice1, Benjamin Readhead1, Jin G. Park1 1Arizona State University, Biodesign Institute, Tempe, AZ, United States Serological responses reveal recent and historical exposure to pathogens, as well as the state of autoimmune and other chronic conditions, including cancer. Serological tests either assess one or a few antigens in many people, or multiple antigens for modest sample sizes. Here, we describe a multiplexed serology method that evaluates samples at the scale of thousands1. This molecular epidemiology tool operates at a population scale sufficient to evaluate broadly how various infectious exposures or autoimmune responses affect health. The method employs full-length folded proteins, is quantitative over a wide dynamic range and performs favorably compared with commercial clinical assays. Responses to 39 bacteria species/strains and 99 viruses in 2400 people were evaluated. Subjects with longitudinal data showed quantitative stability of response to all antigens over the six-month time window, enabling the detection of intervening clinical events. Variation over this time period was low enough to be compatible with clinical use. These data revealed intervening infections, vaccinations, and age and gender-related prevalence data for EBV, HPV and Rubella. The Multiplexed In-Solution Protein Array (MISPA) platform employs a solution-based protein library, each protein antigen covalently linked to a unique DNA barcode. The use of barcodes, in lieu of attaching proteins to a surface, allowed antibody and target antigen to interact with solution phase kinetics, and provided a quantitative dynamic range over several orders of magnitude. The platform is highly scalable; a recent study included testing 11,000 samples for responses to 180 different targets, yielding more than 2 million data points. We expect this highly adaptable method will find broad application in immune profile tracking. 522

523

P-I-0369 Predictive biomarker discovery for immune checkpoint inhibitor in metastatic MMRd endometrial cancer through deep proteomic profiling of FFPE tissue samples Juan Francisco Grau-Béjar1,2, Antoine Amaury Lachaud3, Martin Mehnert3, Yuehan Feng3, Alexandra Leary1, Despina Stefanoska3 1Institut Gustave Roussy, Villejuif, France 2Vall d’Hebron Institute of Oncology, Barcelona, Spain 3Biognosys AG, Schlieren, Switzerland Mismtach repair deficiency (MMRd) status is a robust predictive biomarker for immune checkpoint inhibitor (ICI) in endometrial cancer (EC). Nevertheless half of the selected patients do not respond to the therapy. Most studies exploring biomarkers of response to ICI have focused on tumor or immune cell factors. In the present study, we aimed to identify novel response biomarker by exploring the proteomic profiles of baseline patient FFPE samples with known treatment outcome. Clinical data and outcomes of metastatic MMRd EC patients, treated with ICI at Gustave Roussy Institute (2016- 2021), were retrospectively collected. Patients were classified as responders/Rs (complete response/CR, partial response/PR, or stable disease/SD ≥12 months) or non-responders/NRs (progressive disease/PD or SD <12 months). 25 Pre-ICI treatment FFPE tumor samples (17 Rs and 8 NRs) were subjected to Biognosys UltraDeep TrueDiscovery™ data-independent acquisition (DIA)-Mass Spectrometry (MS) analysis and resulted data were analyzed using Spectronaut software. In total, 11"187 proteins were quantified in all samples with an average of 10101 proteins quantified per sample. Further in-depth analysis of the data unveiled that extracellular matrix modeling is one of the major pathways driving the separation of Rs and NRs. For instance, multiple proteins from the collagen family are significantly upregulated in the NRs. To further validate this finding, several orthogonal technologies were employed to investigate the relationship between collagen and response to ICI: 1) Collagen density was determined on whole FFPE slides using Masson"s trichrome and quantified by Qupath as % positive surface in the intratumoral area. 2) Second-harmonic generation (SHG) microscopy combined with polarimetry (p-SHG) was used to estimate collagen content and describe the spatial orientation of fibrillar collagen in the intratumoral area. 3) intraepithelial (ie) and stromal (s) CD3+ and CD8+ cells were quantified by Qupath. The spatial imaging results indicate that collagen density in the intratumoral area was significantly higher in NRs compared to Rs. Intratumoral collagen density was positively correlated with collagen type V expression, and negatively correlated with ie and s CD3+ cell and ie and s CD8+ cell infiltration. SHG confirmed that fibrillar collagen content within the tumor area was significantly greater in NRs and p-SHG revealed that NR tumors tended to show a highly disorganized collagen matrix. Taken together, MS-based discovery proteomics was instrumental in identifying ECM as one of the drivers of treatment response. Two different spatial approaches consistently demonstrated that increased collagen expression or density may serve as a potential predictive biomarker of resistance to ICI in MMRd metastatic EC, contributing to create an immunosuppressive tumor microenvironment (TME). 523

524

P-I-0370 Plasma protein biomarker discovery for ovarian cancer Ryan Lamers1, Ray Chen2, Cindy Lawley2, Ida Grundberg3, Karin Sundfeldt4, Karin Stalberg5, Stefan Enroth6, Ulf Gyllensten6 1Olink Proteomics, Scientific Affairs, Waltham, MA, United States 2Olink Proteomics, Strategic Accounts, Waltham, MA, United States 3Olink Proteomics, Product Management, Uppsala, Sweden 4University of Gothenburg, Department of Obstetrics and Gynecology, Gothenburg, Sweden 5Uppsala University, Department of Women’s and Children’s Health, Uppsala, Sweden 6Uppsala University, Department of Immunology, Genetics, and Pathology, SciLifeLab Uppsala, Uppsala, Sweden Ovarian cancer (OvCa) is the fifth deadliest cancer in females with over 300,000 new cases and over 200,000 deaths per year. OvCa has an overall 5-year survival of only 30-50%. Survival increases to over 90% when detected in stage I, but is only 20% when detected in stage IV. Thus, there is a pressing clinical need for accurate biomarkers that enable the early detection of OvCa. This study measured plasma protein levels in two independent cohorts with women surgically diagnosed with benign or malign conditions after suspicion of OvCa. In total, plasma protein levels of 404 women were measured using the Olink® Explore HT (5400+ unique proteins) platform and its predecessor, Olink® Explore 3072 (~3000 unique proteins). Technical comparisons of 2,844 assays overlapping between the two platforms revealed a strong median R correlation of 0.91, demonstrating that Olink Explore HT delivers the same high data quality as Olink Explore 3072, even with its expanded proteome coverage. Using one of the investigated cohorts as discovery, we searched for biomarkers that were indicative of early (stage I-II), late (stage III-IV), or any stage cancer compared to benign tumors, and after strict adjustment for multiple hypothesis testing, we found a total of 327 such associations. These associations corresponded to 191 unique proteins, including early-stage cancers compared to benign and with both increased or decreased circulating levels in malign cases compared to benign. Using the second cohort as validation, 99.7% (326) of these associations were then replicated. All of the biomarker candidates were found to have similar fold change (Pearson"s R=0.93, p < 3.0 x 10-140) between the two cohorts. Thirty-six of the 191 proteins (19%) were unique to the Olink Explore HT platform. MUCIN-16 (CA-125), a well-known OvCa marker used in the clinic, was not found to be significantly different between early-stage cancers and benign tumors and was not among the top 5 ranked biomarker candidates. This study highlights the potential of Olink Explore HT to enhance the discovery of biomarkers for early-stage OvCa detection, deepen our understanding of disease biology, and open new avenues for potential therapeutics. 524

525

P-I-0371 ZURÜCKGEZOGEN: Personalized tissue and liquid biopsy proteomics identify treatment targets and inform patient care Philipp Lange1 1University of British Columbia, Vancouver, Canada Molecularly targeted precision treatments have significant potential to improve therapy options for hard-to- treat cancers. Genome sequencing has laid the foundation for precision medicine, yet, clinical success remains moderate. Identifying therapeutic targets at the protein and pathway levels holds great promise but real-life clinical evidence is scarce. Here we will demonstrate that DIA proteome analysis of FFPE biopsies can identify treatment targets and inform patient care in a clinically meaningful timeframe. As proof of concept, we describe the case of an adolescent with metastatic, progressive spindle epithelial tumor with thymus-like differentiation. DIA proteomics on a QE- HF Orbitrap mass spectrometer using a gas phase fractionation library quantified >6,500 proteins from macro- dissected tumor and adjacent normal from formalin fixed paraffin embedded sections. Tissue annotation and dissection by a trained pathologist, sample processing by ASAP in triplicates, LC-MS/MS, data analysis with target identification and validation by IHC was completed within two weeks of biopsy and reported to a molecular tumor board as preliminary assessment. Subsequently, target pathway stability was evaluated by DIA proteomics on archived prior biopsies from the same patient. Target levels were validated and compared across AYA tumors by immunohistochemistry, the response to targeted inhibition was tested in chicken chorioallantoic membrane and larval zebrafish xenografts generated from the patient, and on target effect was confirmed by stable isotope tracing. Following failure of cytotoxic chemotherapy and second-line treatment, an 8 week innovative therapy trial was initiated by the patient. Over the short treatment course clinical RECIST scoring indicated progressive disease but a reduced tumor growth rate and no new lesions suggested in-patient cellular response. We then present the expansion to a rapid Laser Capture Microdissection based workflow coupled to DIA-PASEF quantification on a timsTOF Ultra system to provide rapid assessments on target heterogeneity and demonstrate the potential for rapid MagNet based Extracellular Vesicle (EV) enrichment based liquid biopsy proteomics to provide molecular tumor boards with personalized pre-clinical data without invasive procedures. Lastly, we present an update on the Canada-wide integration of pre-clinical proteomics evidence into pediatric precision oncology within the Canadian Pediatric Cancer Consortium ACCESS and the PROFYLE study to promote equitable access to proteomics analyses. Significance: Overall, we demonstrate in a proof-of-concept case that pre-clinical proteomics analysis can be conducted in a clinically meaningful timeframe and provide pre-clinical data to support medical decision-making. To determine if this is generalizable, we further outline a road map for advancing access to pre-clinical proteomics and innovative molecular assays across Canada. 525

526

P-I-0372 Exploring proteomic signatures of high-grade serous carcinoma of the ovary Matthias Fahrner1,2, Patrick Bernhard2,3,4, Tannith Noye5,6, Tilman Werner2,3,4, Tobias Dreyer7, Eva Brombacher8,9, Roberto Leguizamon2, Clemens Kreuz8,9, Miguel Conseza-Contreras2,4,10, Frank Hause2,11,12, Viktor Magdolen7, Holger Bronger7, Carmela Ricciardeli5,6, Oliver Schilling2 1German Cancer Consortium (DKTK) and Cancer Research Center (DKFZ), Freiburg, Germany 2University Medical Center, Institute for Surgical Pathology, Freiburg, Germany 3Spemann Graduate School of Biology and Medicine (SGBM), Freiburg, Germany 4University of Freiburg, Faculty of Biology, Freiburg, Germany 5University of Adelaide, Adelaide Medical School, Obstetrics and Gynaecology, Adelaide, Australia 6University, The Robinson Research Institute, Adelaide, Australia 7Technical University of Munich, Munich, Germany 8University, Institute of Medical Biometry and Statistics, Faculty of Medicine and Medical Center, Freiburg, Germany 9University, Centre for Integrative Biological Signaling Studies (CIBSS), Freiburg, Germany 10University, MeInBio Graduate School, Freiburg, Germany 11University, Department of Pharmaceutical Chemistry & Bioanalytics, Institute of Pharmacy, Martin Luther University, Halle (Saale), Germany 12Martin-Luther-University Halle-Wittenberg, Institute of Molecular Medicine, Halle (Saale), Germany Introduction: Ovarian High-Grade Serous Carcinoma (HGSOC) is the most common ovarian carcinoma and is an aggressive disease that is frequently detected at advanced stages. Median 5-year survival rates are between 15% and 55%, with stage and degree of debulking being the major prognostic factors. Despite high initial responses to chemotherapy in about 65% of patients, most cases recur. Early detection and prognosis are urgently needed; this is where Mass spectrometry-based proteomics plays a key role in biomarker detection. Methods: Treatment-naïve, formalin-fixed paraffin-embedded (FFPE) tissues from 77 HGSOC patients were measured via liquid-chromatography mass-spectrometry (LC-MS/MS) based proteomics in data independent acquisition mode (DIA). Results We identified more than 4000 proteins per sample, 3500 of which with less than 20% missing values. In a hierarchical clustering analysis, we could highlight distinct proteomic differences between three interconnected clusters: Cluster 1: Upregulation of proteins linked to innate immunity and extracellular matrix (ECM) organization, with reduced metabolic activity. Cluster 2: High aerobic respiration and antigen processing/presentation, with decreased cell motility. Cluster 3: Lower immune infiltration and upregulated organic acid synthesis. We found that immune cell markers were enriched in Cluster 1, as were extracellular matrix proteins. Conversely, epigenetic regulators were increased in Cluster 2, suggesting increased genomic plasticity. We also detected antibody-drug conjugate (ADC) targets, such as FOLR1 and MSLN with differential abundance. Conclusion: Our proteomic analysis revealed three interconnected clusters within the cohort, reflecting continuous variations rather than separate biological entities. These findings highlight the complex proteomic landscape of HGSOC and identify potential biomarkers and therapeutic targets. 526

527

P-I-0373 Decoding CRC progression: impact of patient individuality on the proteomic phenotype Agnes Lengenfeld1, Thorben Sauer1, Julia Horn1, Katja Klempt-Gießing1, Jan Nolde2, Claudia Benecke2, Jutta Kirfel3, Timo Gemoll1 1University of Lübeck, University Hospital Schleswig Holstein, Campus Lübeck, Section of Translational Surgical Oncology and Biobanking, Department of Surgery, Luebeck, Germany 2University Hospital Schleswig-Holstein, Campus Lübeck, Department of Surgery, Luebeck, Germany 3University Hospital Schleswig-Holstein, Campus Lübeck, Institute of Pathology, Luebeck, Germany Colorectal cancer (CRC) is one of the most prevalent malignant tumors and the second most common cause of cancer-related mortality worldwide. Overall, the prognosis for CRC with liver metastasis is inferior and little is known about the patient-individual metastasis. The study aimed to assess significant differences in protein expression using a unique set of paired and unpaired patient samples respecting tumor heterogeneity. We analysed formalin-fixed paraffin-embedded (FFPE) tissue samples derived from healthy colon mucosa (NM), colorectal carcinoma (T) and liver metastases (LM, n=78 patients). Paired tissue samples (NM-T-LM) were obtained from 18 patients. For global proteome analysis, we employed DIA-PASEF (data-independent acquisition - Parallel Accumulation- Serial Fragmentation) mass spectrometry coupled with neural-network-based, library-free protein identification. For global proteomic profiling, we employed principal component analyses, differential expression analyses and gene set enrichment analyses. Protein profiling revealed 6.673 identified proteins after strict raw data processing. Based on the group comparison results, clear separation between the NM from T and LM samples was detected in a principal component analysis. To assess the impact of patient individuality, matched and unmatched T were compared to LM and revealed 298 (matched) and 246 (unmatched) differentially expressed proteins (q-values ≤ 0.05, net log2 fold-change ≥ 1, Figure 1A-B). Interestingly, 94 and 42 proteins were unique for the matched and unmatched test, respectively (Figure 1C). Our research results highlight the importance of individual, patient-paired evaluations for clinical studies. In addition to changes among groups reflecting CRC progression, we identified significant expression differences between NM, primary T, and LM samples from individuals, which might accelerate implementation of precision oncology in the future. Figure 1: Differential expression analysis results. A: Volcano plot visualizing the differential expression analysis results of LM samples compared to their T pairs. Significance was considered at q-value ≤ 0.05 and net log2 fold change ≥ 1. B: Volcano plot visualizing the differential expression analysis results of LM compared to unpaired T samples. For group size balancing, 18 T samples were randomly selected. C: UpSet plot visualizing set intersections and unique sets of differentially expressed proteins for paired and unpaired testing setups. 527

529

P-I-0374 Proteomics uncovers the underlying mechanism of distant metastasis in follicular thyroid carcinoma Lu Li1,2, Wenhao Jiang1, Jianbiao Wang3, He Wang1, Zhiqiang Gui1,4, Zhihong Wang4, Hao Zhang4, Hanqing Liu5, Chuang Chen5,6, Jiafei Shen6, Dong Xu6, Wenjun Wei7, Yu Wang7, Wanyuan Chen8, Haixia Guan9, Tiannan Guo1, Jianqing Gao2, Yaoting Sun1 1Westlake University, School of Medicine, Hangzhou, China 2Zhejiang University, College of Pharmaceutical Sciences, Hangzhou, China 3Zhejiang University, Hangzhou, China 4The First Hospital of China Medical University, Shenyang, China 5Renmin Hospital of Wuhan University, Wuhan, China 6Cancer Hospital of the University of Chinese Academy of Sciences, Hangzhou, China 7Fudan University, Shanghai Medical College, Shanghai, China 8Zhejiang Provincial People's Hospital, Hangzhou, China 9Guangdong Provincial People's Hospital, Guangzhou, China Background and objective: Over the past two decades, the occurrence of thyroid cancer has shown a consistent upward trend. While the majority of thyroid cancers tend to be relatively slow-growing, it is noteworthy that distant metastases occur in 1-4% of cases, often leading to fatal outcomes. Among differentiated thyroid cancers, follicular thyroid carcinoma (FTC) stands out as having a higher risk of developing distant metastases, accounting for approximately 25% of such cases. Despite this, the underlying mechanism behind FTC distant metastases remains elusive. Materials and Methods: To describe the molecular changes, we collected 66 formalin-fixed paraffin-embedded (FFPE) slides obtained from 66 FTC patients with (N=25) and without (N=41) distant metastasis. The distant metastasis group can be further divided into three subgroups: 8 cases of bone metastasis, 11 cases of lung metastasis, and 6 cases of combined lung and bone metastasis. The samples were prepared through a pressure cycling technology-assisted pipeline, proteomic data were acquired by PulseDIA mode on timsTOF Pro and searched by DIA-NN (v1.8.1) against a thyroid-specific spectral library. Preliminary results: In this dataset, we quantified 11,044 proteins and identified 135 deferentially expressed proteins (DEPs) between the patients with and without distant metastases. These two groups of samples can be partially differentiated by uniform manifold approximation and projection analysis using the DEPs. Moreover, the pathway and function analysis revealed that extracellular matrix organization, cell matrix adhesion and collagen fibril organization were significantly dysregulated in the thyroid cancer metastasis. Based on the DEPs, we established an Xgboost model for the protein feature selection and stratification. Our model achieved an AUC of 0.848 under 10 features. Next, we delve deep into the intricate mechanisms underlying tumor metastases across various organs. Through differential protein screening and cluster analysis, we uncovered that lung metastases are intimately linked to ECM perturbations, bone metastases are associated with immune dysregulation, and combined metastases exhibit complex interactions involving both metabolic alterations and immune dysregulation. Conclusion: This study will provide clues for the understanding of the molecular mechanisms of distant metastasis in FTC, and provide a theoretical basis for improving clinical patient management and potential therapeutic options. 529

530

P-I-0375 Mitochondrial Proteomic Characterization of Thyroid Oncocytic Tumors: Diagnostic Potential of IDH2 as a Biomarker Lu Li1, Wenhao Jiang2, Likun Zhang3, Zhiqiang Gui4, Zhihong Wang4, Hao Zhang4, Tiannan Guo2, Zhiyan Liu3, Yaoting Sun2, Jianqing Gao1 1Zhejiang University, Hangzhou, China 2Westlake University, Hangzhou, China 3Shanghai Sixth People's Hospital Affiliated to Shanghai Jiao Tong University School of Medicine, Shanghai, China 4The First Hospital of China Medical University, Shenyang, China Background and objective: Oncocytic thyroid neoplasm (OTN), also known as Hürthle cell tumor, is characterized by the abundant expression of mitochondria and is typically insensitive to radioactive iodine treatment. Compared to canonical thyroid cancer, oncocytic thyroid carcinoma (OTC) exhibits more aggressive biological behavior with a higher rate of lymph nodes and distant organs. Our 8-gene panel results indicate a low frequency of driver gene mutations of OTN. Hence, we explore the alterations based on the proteome. Preliminary Data: 1. Differences in mitochondrial protein expression between oncocytic and non-oncocytic tumors and the development of the oncocytic scoring system To identify oncocytic protein features, we initially analyzed the proteome data from the Singapore dataset (N=64) which we published previously. Proteomic data were obtained by data-independent acquisition mode and searched through DIA-NN (1.8.1) against a thyroid-specific spectral library containing 136,821 precursors and 10,547 proteins. We next filtered the protein matrix by extracting proteins only expressed in mitochondria (n=386). To get robust protein features, we compared oncocytic adenomas versus non-oncocytic adenomas and oncocytic carcinomas versus non-oncocytic carcinomas through three statistical testing methods. As a result, 64 proteins were distilled. The pathway enrichment analysis showed the generation of precursor metabolites and energy was altered most. 2. IDH2 as a biomarker to identify OTC Upon a closer examination of the data, we discovered that four out of five IDH family proteins exhibited a remarkable upregulation, exceeding a four-fold change. Notably, IDH2, a classical drug target protein, not only serves as a marker to distinguish oncocytic cells from non-oncocytic cells but also functions as a discriminator between malignant and benign oncocytic tumors. This observation aligns positively with the malignancy grade, indicating its potential diagnostic significance. Furthermore, our proteomics findings, based on mass spectrometry, were corroborated by immunohistochemistry in an independent cohort of 80 samples, reinforcing the reliability of our observations. Novel Aspect: We have established a scoring system based on 64 mitochondrial proteins, which serves as a quantitative measure for thyroid oncocytic character. Furthermore, we propose that the protein marker IDH2 holds significant translational potential, promising to be a clinically valuable diagnostic tool. 530

531

P-I-0376 Serum proteomic analysis reveals molecular features for stratification of chronic obstructive pulmonary disease Wen-Yi Liu1, Sheng-Ming Wu2,3,4, Kang-Yuan Chen2,3,4, Kang-Yun Lee2,3,4, Chia-Li Han1,4 1Taipei Medical University, master program in clinical genomics and proteomics, Taipei, Taiwan 2Taipei Medical University-Shuang Ho Hospital, Division of Pulmonary Medicine, Department of Internal Medicine, New Taipei City, Taiwan 3Taipei Medical University, Division of Pulmonary Medicine, Department of Internal Medicine, School of Medicine, College of Medicine, Taipei, Taiwan 4Taipei Medical University, TMU Research Center for Thoracic Medicine, Taipei, Taiwan Chronic obstructive pulmonary disease (COPD) is a lung disease characterized by airway limitation and persistent respiratory symptoms. Tobacco smoking and exposure to air pollution have been reported as major risk factors of COPD, leading to diverse clinical presentations of COPD with unclear pathogenesis and no efficient treatments are available for now. Identification of biomarkers for stratification of COPD phenotypes will benefit the discovery of potential treatment strategies. In this study, we develop a data-independent acquisition mass spectrometry (DIA-MS) platform assisted by an in-house serum peptide spectral library for in-depth profiling of serum proteome in COPD. Individualized proteomic analyses of serum samples from 88 COPD patients identified 1,062 proteins, of which 37 down-regulated and 81 up-regulated proteins involved in nuclear receptor signaling, cytokine signaling and innate immune system pathways were filtered in the late-stage COPD. A number of stage- dependent proteins was discovered; the expression levels of ADIPOQ and FAM20C are negatively correlated with forced expiratory volume (FEV1%), while CNTN3, MEGF8, CFHR2 and DAG1 are positively correlated (p<0.05). Furthermore, the differentially expressed serum proteins in smoker COPD are associated with regulation of cell migration and epithelial cell differentiation. The clustering analysis of serum proteome profiles identified two molecular subtypes in the 88 COPD patients, however, no significant differences were found in clinical characteristics between the two clusters of patients. These data suggest a diverse pathogenesis mechanism in different phenotypes of COPD patients. Based on this finding, we will further extract a reliable serum protein panel that could offer efficient stratification of COPD for clinical utility and explore potential treatment strategies toward each phenotype of COPD. 531

532

P-I-0377 Revolutionizing Proteomics: Advanced nanoESI Ion Source with NanoElectrospray Emitters for Unmatched Sensitivity and Robustness Daniel Lopez Ferrer1 1Montara Biolabs, Las Matas, Spain While high-throughput, robust, and highly sensitive mass spectrometry (MS) analysis is essential for proteomic breakthroughs, current ion sources often grapple with achieving all three simultaneously. Existing options might provide good sensitivity but require complex emitter setups and meticulous maintenance, limiting accessibility. Conversely, user-friendly alternatives may compromise sensitivity, hindering deep proteome coverage. A groundbreaking nanoESI ion source disrupts this status quo by integrating advanced sensors and "plug & play" nanoelectrospray emitters. These emitters boast an ultra-sharp cone angle and a hydrophobic silica nanoparticle coating, significantly increased sensitivity and reduced sample consumption. The "plug & play" design eliminates complex emitter setups, making high-performance emitters accessible to all researchers. Integrated sensors continuously monitor critical conditions, ensuring optimal performance and safeguarding instruments. This system was tested using the Direct Infusion Shotgun Proteomics Analysis (DISPA) approach on a Thermo Fisher Scientific Orbitrap Astral mass spectrometer, achieved remarkable results: identifying approximately 2,000 proteins from HeLa cells at a rate exceeding 8 proteins per second, along with detecting a broad spectrum of biomarkers in human plasma samples. This innovative approach, combining unmatched sensitivity and robustness with user-friendly operation, paves the way for significant advancements in clinical diagnostics and biomarker discovery, transforming the clinical landscape. improving electrospray ionization for 532

533

P-I-0378 Revealing protein profiles in Parkinson's disease by mass spectrometric analysis of living brain tissue during deep brain stimulation Saskia Leonie Ludwig1,2,3, Ulrike Resch4, Michael Saynisch1,2, Pablo Andrade-Montemayor3, Marcus Krüger1,2 1Cologne University, Institute for Genetics, Cologne, Germany 2CECAD Cologne, Cologne, Germany 3University Clinic Cologne, Stereotactic Surgery, Cologne, Germany 4University of Vienna, Vienna, Austria Introduction: In an ageing population, the increase in neurodegenerative disorders such as Parkinson's disease (PD) poses a major challenge. A deeper understanding of the pathophysiology and identification of biomarkers is therefore of utmost interest to develop new diagnostic and therapeutic interventions. PD manifests through a spectrum of motor and non-motor symptoms, rooted in the selective loss of dopaminergic neurons in the substantia nigra. Current diagnostic approaches rely on post-mortem examinations of brain samples, underscoring the urgency for novel methodologies. Single-use deep brain stimulation (DBS) electrodes provide a promising tool for acquiring living brain tissue, enabling proteomic analysis to elucidate disease mechanisms and identify biomarkers, advancing our understanding of PD's progression. Methods: Immediately after DBS surgery, electrodes were collected and associated proteins were added to lysis buffer. Protein concentration was determined and 20 µg of the samples were digested using the SP3 method. Peptides were cleaned and desalted using C18 stop and go extraction tips. Liquid-chromatography tandem mass- spectrometry (LC-MS/MS) was performed with an ion mobility TOF instrument using a 44min gradient and measured in DIA-PASEF approach. Spectra were analysed using DIA-NN and bioinformatic data analysis was performed using Perseus and InstantClue to obtain disease-specific protein profiles from different disease entities. Preliminary data: We collected 38 samples from 12 patients undergoing DBS surgery to explore recovering neuronal cells from surgical. The patient cohort exhibited various neurological conditions, including Alzheimer’s Disease (AD), dystonia, essential tremor, Tourette Syndrome, and PD. Protein concentrations ranged from 25 µg to 390 µg per test electrode and LC-MS/MS identified up to 1182 proteins per sample, with a total of 3276 proteins from living brain tissue. The most abundant proteins were neurofilament and myelin-related proteins (PLP1, CA1, MBP), indicating the presence of neurons and glial cells. Principal component analysis (PCA) didn't show patient group segregation, possibly due to sample size limitations. However, a-synuclein (SNCA) showed a significant enrichment in PD whereas the amyloid precursor protein (APP) or tau protein (MAPT) showed and increased expression in AD patients. The preliminary findings suggest feasibility in acquiring sufficient material from DBS electrodes for analysis and reveal PD and AD-related proteins. In future experiments we will expand the patient cohort with non- neurodegenerative conditions like Tourette Syndrome and pain as a control. This approach enables comprehensive examination of protein changes in specific and intact brain regions of living patients, overcoming the challenges of post-mortem tissue use, such as protein degeneration. This method provides a valuable tool for studying the molecular mechanisms underlying neurodegenerative diseases. 533

534

P-I-0379 Single-shot mass spectrometry-based phosphoproteomic analysis of patient-derived formalin-fixed, parrafin- embedded samples reveal tissue specific fingerprints Maximilian Maldacker1,2,3, Daniel Vogele1,2,4, Johanna Thiery1,2,5, Matthias Fahrner1,2,5,6, Raphael Kirschstein1, Nadine Meier1, Peter Bronsert1,5,6, Oliver Schilling1,2,3,7,4,5,6 1University Hospital Freiburg, Institute for Surgical Pathology, Freiburg, Germany 2Albert-Ludwigs University of Freiburg, Faculty of Biology, Freiburg, Germany 3Albert-Ludwigs University of Freiburg, Oncogene-driven Immune Escape (OncoEscape), Freiburg, Germany 4Albert-Ludwigs University of Freiburg, RTG ProtPath, Freiburg, Germany 5Partner Site Freiburg, German Cancer Consortium (DKTK), Freiburg, Germany 6University Medical Center Freiburg, Medical Faculty, Freiburg, Germany 7Institute for Surgical Pathology, Freiburg, Germany refining current automated phosphoenrichment protocols, we aim Routine in molecular tumor boards for patient therapy evaluation mainly relies on formalin-fixed and paraffin- embedded (FFPE) tissues for multiple screening strategies. While the analysis of the proteome from FFPE has been implemented just recently in MTB screenings, the profiling of post-translational modifications like phosphorylation is not applicable from FFPE samples. Previous approaches for this sample type involve TMT- labeling that is not suitable for individualized and independent patient-level profiling. In addition, validation for a robust analysis of the cell state, fate and functionality on phosphoproteome level remains to be demonstrated. By label-free phosphoproteomic workflow for FFPE-derived tissues. After metal-ion affinity chromatography, we analyze the phosphoproteome of a multi-patient, multi-organic non-cancer cohort on a timsTOF HT in data-independent acquisition. In the explorative analysis more than 6000 phosphopeptides are identified from 50 ug of single samples which enables the accurate tissue-specific profiling in accordance to previously published clinical phosphoproteome data sets. By testing for the coverage of relevant oncogenic signaling pathways, we showcase that the transfer of state-of-the art workflows from research to clinical routine adds valuable insights to personalized medicine. In addition, we aim to control for routinely error-prone tissue preparation steps from the surgery to MS and their effect on the phosphoproteomic level in a time-to-fixation experiment allowing us to implement a pipeline for sample quality assessment. to establish a 534

535

P-I-0380 Pathophysiological subtypes of mild cognitive impairment due to Alzheimers disease identified by CSF proteomics Bruno Manadas1 1University of Coimbra, Center for Neuroscience and Cell Biology, Cantanhede, Portugal Alzheimer’s disease (AD) is a heterogeneous neurodegenerative disorder and detailed understanding of its pathophysiologic processes at earlier phases is crucial for effective treatment development and prevention. In the present study, we investigated within the Cognitive Complaints Cohort (CCC) the changes in the CSF proteome of patients with Mild Cognitive Impairment (MCI) fulfilling criteria of high likelihood for AD (n=45), compared to MCI patients without any biomarkers of Aβ deposition or neuronal injury (n=23). This research strategy intended to control for nonspecific changes that might influence the proteome and thus sort out the alterations directly related to AD pathogenesis. CSF proteomics could discriminate patients with MCI due to AD from patients with MCI without biomarkers of Aβ deposition or neurodegeneration. In a second step, we looked at the existence of putative subtypes in patients with MCI due to AD using a clustering non-negative matrix factorization method. Results were validated by applying the same methodology on MCI patients with (n=92) and without (n=102) AD pathology from the European Medical Information Framework for Alzheimer’s Disease (EMIF-AD) cohort. Using the 164 and 170 selected proteins in the CCC and EMIF-AD cohorts, respectively, that differed significantly between patients with MCI due to AD and patients with non-degenerative MCI, two consistent clusters of AD patients at the MCI stage were found. Twenty (47.6%) patients were labelled in Cluster 1 and 22 (52.4%) were labelled in Cluster 2 in the CCC cohort, and 46 (50%) participants were labeled in Cluster 1 and 46 (50%) labeled in Cluster 2 in the EMIF-AD cohort. Distinct Gene Ontology profiles were found for the two clusters. Cluster 1 showed increased levels of proteins related to biological processes such as the immune system, acute inflammatory response, coagulation and lipid metabolism, whereas Cluster 2 showed involvement of neuronal and neurodevelopmental processes. There was a statistically significant concordance between the two cohorts regarding the contribution of different gene ontologies to Cluster 1 and Cluster 2 (Cohen"s kappa= 0.398, p<0.0001). Cluster 2 had higher levels of CSF tTau, CSF pTau and CSF Aβ42 levels as compared to Cluster 1. The results show that, in spite of slightly distinct diagnostic criteria and different mass spectrometry approaches used in the two cohorts, two consistent subtypes of AD patients emerge that could be described as (i) immune and blood-brain barrier dysfunction (Cluster 1) and (ii) hyperplasticity (Cluster 2), as previously proposed. These findings might be important for the design and interpretation of clinical trials since there might be an association between response to the therapy and the AD subtype to which the patients belong. Fig. 1 (optional) 535

536

P-I-0381 Addressing inter-individual variability in CSF levels of brain-derived proteins across neurodegenerative diseases Sára Mravinacová1, Sofia Bergström1, Jennie Olofsson1, Nerea Gómez de San José2, Sarah Anderl-Straub2, Janine Diehl-Schmid3, Klaus Fassbender4, Holger Jahn5, Johannes Kornhuber6, G Bernhard Landwehrmeyer2, Martin Lauer7, Johannes Levin8, Albert Ludolph2, Johannes Prudlo9, Anja Schneider10, Matthias L Schroeter11, Jens Wiltfang12, Petra Steinacker13, Markus Otto2,13, Peter Nilsson1, Anna Månberg1, FTLD Consortium2 1KTH Royal Institute of Technology, SciLifeLab, Solna, Sweden 2University Hospital Ulm (UKU), Ulm, Germany 3Technical University of Munich, Munich, Germany 4Saarland University, Homburg, Germany 5University Hospital Hamburg Eppendorf, Hamburg, Germany 6Friedrich-Alexander University Erlangen-Nuremberg, Erlangen, Germany 7University Hospital of Würzburg, Wuerzburg, Germany 8LMU University Hospital, Munich, Germany 9Rostock University Medical Center and German Center for Neurodegenerative Disease, Rostock, Germany 10University of Bonn and DZNE Bonn, Bonn, Germany 11University Clinic Leipzig, and Max Planck Institute for Human Cognitive and Brain Sciences, Leipzig, Germany 12University Medical Center Goettingen, and DZNE, Goettingen, Germany 13Martin-Luther-University Halle-Wittenberg, Halle (Saale), Germany Background: Accurate diagnosis and monitoring of neurodegenerative diseases requires reliable biomarkers. Cerebrospinal fluid (CSF) proteins hold promise for reflecting brain pathology, yet their utility may be compromised by natural variability between individuals, weakening their association with disease and impacting their diagnostic accuracy. Method: We measured the levels of 69 proteins in CSF from 499 individuals using an antibody-based suspension bead array technology. In this multi-disease cohort including participants with Alzheimer’s disease, behavioural variant frontotemporal dementia, primary progressive aphasias, amyotrophic lateral sclerosis, corticobasal syndrome, primary supranuclear palsy and healthy controls, we investigated inter-individual variability in CSF protein levels. Further, we explored protein profiles across the diseases, after adjusting for this variability using linear modelling. Results: Correlation and PCA analysis on the measured CSF proteins revealed the presence of inter-individual variability in overall CSF levels of brain-derived proteins, which could not be explained by disease associations. Using linear modelling, we show that adjusting for median CSF levels of brain-derived proteins increases the diagnostic accuracy of many proteins, including those previously identified as altered in CSF in the context of neurodegenerative disorders. We further demonstrate a simplified approach for the adjustment using pairs of correlated proteins with opposite alteration in the diseases. With this approach, the proteins adjust for each other and further increase the biomarker performance through additive effect. Moreover, we find that many CSF proteins show similar alteration across the studied diseases, indicating that these proteins likely reflect shared processes. Conclusion: CSF protein profiles are affected by inter-individual variability which partially hinders the diagnostic potential of CSF proteins. Thus, we propose that this variability should be accounted for in future studies. Further, we demonstrate high overlap in CSF protein patterns across neurodegenerative diseases, highlighting the need for multi-disease studies in order to identify biomarkers that are specific to or common between different diseases. 536

537

P-I-0382 Mining proteomic signature of soft tissue sarcoma by mass-spectrometry to find new candidate drug targets Veronica De Giorgis1, Vittoria Federica Borrini2, Maria Rescigno3, Marcello Manfredi1 1University of Piemonte Orientale, Translational medicine, Novara, Italy 2University of Piemonte Orientale, Sciences and innovation technologies, Novara, Italy 3Humanitas IRCCS, Milan, Italy Background: Soft-tissue sarcoma is an aggressive cancer of the mesenchymal tissue. It is a heterogeneous disease since comprises several histotypes and can arise in different areas of the body. Its aggressiveness is mostly related to its high metastasis rate (40-50%) and the unsatisfactory efficacy of traditional therapies. Given this, it is mandatory to deepen the biological knowledge of this group of malignancies and to find out new therapeutic strategies based on their specific molecular patterns. To this aim, drug repurposing is one of the most powerful tool for therapeutic discovery because it allows the reduction of time and costs. In this research, we used an advanced proteomic and bioinformatic approach on sarcoma and healthy tissue, to identify new potential therapeutic targets. Methods: We performed proteomic analysis of 26 tumors and healthy samples from the same sarcoma patient, including different soft-tisse sarcoma histotypes. Tissue biopsies were lysed, digested, and analyzed on an Ultimate 3000 RSLC nano coupled to an Orbitrap Exploris 480 with a High-Field Asymmetric Waveform Ion Mobility Spectrometry System. The acquired raw MS data files were processed and analyzed using Proteome Discoverer with CHIMERYS algorithm. Bioinformatic analysis were performed on modulated proteins using DAVID and IPA software. Druggable proteome was also investigated using the Drug Bank database. Potential drug targets were further mapped to UniProt, Therapeutic Target Database (TTD) and Cancer/testis antigens database. In addition, investigation of clinical trials" status on ClinicalTrials.gov was performed. Results: The proteomic analysis allowed the identification of almost 10000 unique proteins and the quantification of more than 8000 proteins. Sarcoma cancer tissue was characterized by a specific proteomic signature: an enrichment of pathways related to PPAR signaling, carbon metabolism and antigen processing and presentation was identified. Regarding molecular functions, an enrichment of proteins related to RNA binding, chromatin and ribosome was found. The comparison of cancer/healthy proteome allowed the identification of specific over-expressed/under-expressed drug targets for each patient / histotype. Among all the drug targets, 370 were present in at least 88% of the samples, of which 314 were upregulated and 56 were down regulated. A more in-depth analysis allowed the identification of six drugs potentially effective against sarcoma, and that were already employed in the treatment of other diseases. Conclusion: In the presented work we exploited mass spectrometry to create a proteomic mapping of 26 sarcoma biopsies of different soft tissue sarcomas and we were able to characterize the biological signature of soft tissue sarcoma and to obtain a list of reliable new drug targets to develop new therapeutic options. The in vitro validation of the drugs is ongoing. 537

538

P-I-0383 Investigation of inter- and intra-day variability of tear fluid in terms of flow rate, protein concentration and protein composition Annika Guntermann1,2, Oluwaseun Fatoba1,2, Marc Kronenberg1,2, Sabrina Reinehr3, Pia Grotegut3, Marc Schargus3,4, Teresa Tsai3, Svetlana Ivanova3, Bettina Serschnitzki1,2, Nina Kumowski5,6, Christoph Maier7, Katrin Marcus-Alic1,2, H. Burkhard Dick3, Stephanie C. Joachim3, Caroline May1,2 1Ruhr University Bochum, Medical Proteome Center, Bochum, Germany 2Ruhr-University Bochum, Center for Protein Diagnostics (ProDi), Medical Proteome Analysis, Bochum, Germany 3University Eye Hospital, Ruhr-University Bochum, Experimental Eye Research Institute, Bochum, Germany 4University Hospital Duesseldorf, Heinrich-Heine-University Duesseldorf, Department of Ophthalmology, Duesseldorf, Germany 5BG University Hospital Bergmannsheil GmbH, Ruhr-University Bochum, Department of Pain Medicine, Bochum, Germany 6RWTH Aachen University Hospital, Medical Clinic I - Cardiology, Angiology and Internal Intensive Care Medicine, Aachen, Germany 7University Children's Hospital, Ruhr-University Bochum, Bochum, Germany, Bochum, Germany Aim: The aim of this study was to determine the intra- and inter-day variability of tear flow rate, tear protein concentration and tear protein composition for biomarker studies 1. Methods: The study was performed by non-invasive collection of tear fluid from 18 healthy volunteers who repeated the Schirmer's tests at four different time points over a period of two days. The flow rate of the tears fluid was quantified using Schirmer test strips. Tear proteins were extracted from test strips and protein concentration determined. Tear protein composition was analysed by DIA-based mass spectrometry. Subjects underwent a comprehensive neurological and ophthalmological examination to exclude the possibility of health impairment. Results: The results showed that sampling tear fluid from either the right or left eye had no effect on tear flow rate or protein concentration in individual subjects. In addition, protein concentration was found to be independent of tear volume. A comparison between study days revealed significant differences in tear flow rate and protein concentration. Mass spectrometric analysis of the tear fluid revealed eleven proteins that were differentially regulated between the two examination days. Conclusions: Our results provide evidence for diurnal variations in tear flow rate, tear proteome concentration and composition in healthy subjects. This suggests that diurnal variations should be taken into account when investigating tear biomarkers. 1. Guntermann, A. et al. Investigation of Inter- and Intra-Day Variability of Tear Fluid Regarding Flow Rate, Protein Concentration as well as Protein Composition. Invest Ophthalmol Vis Sci 64, 13 (2023). https://doi.org/10.1167/iovs.64.13.13 538

539

New technology: Sample preparation (P-II-0384 to P-II-0434) P-II-0384 Overnight enzymatic digestion? Try it at room temperature! Lazaro Hiram Betancourt Nunez1, Anahita Bakochi1, Aniel Sanchez2, Indira Pla2, Sven Kjellstrom1 1Lund University, Lund, Sweden 2Northeastern University, Chicago, IL, United States Background: The bottom-up proteomic approach predominantly relies on trypsin or Lys-C/trypsin digestion to convert proteins into peptides, which are then analyzed by LC-MS/MS. Typically, proteomic results are derived from overnight digestions at 37°C, considered the optimal temperature for trypsin activity. However, we previously discovered that overnight urea in-solution Lys-C/trypsin digestion at room temperature (RT) outperforms the classical digestion at 37°C (1). This method not only reduces carbamylation at α- and ε-amino groups but also enhances the number and relative intensity of peptides and proteins. Additionally, it demonstrates higher reproducibility and less formation of other post-translational modifications (PTMs) such as pyroglutamic acid. Building on these results, we compared the performance of several samples preparation methods when the enzymatic digestions were carried out at 37°C and at room temperature. Materials and Methods: BT-549 cell lysates and plasma samples were processed and analyzed in triplicates using five different approaches: Urea- and GuCl in-solution digestion, FASP, MED-FASP, and S-trap methods, with both 37°C and room temperature overnight digestions. Peptide mixtures were separated and analyzed using an Evosep One LC system (Evosep Biosystems) coupled to a timsTOF Pro ion mobility mass spectrometer (Bruker Daltonics). MS data was acquired using the 60 samples per day DD-PASEF method. Database searches were performed using FragPipe (v20.0). False-discovery rate (FDR) was filtered at 1% at both the precursor and protein levels. All statistical analyses were conducted in Perseus (v.1.6.13.0) Results: Our study indicates that, regardless of the sample type and preparation method, overnight enzymatic digestion at room temperature with trypsin, Lys-C or Lys-C/trypsin outperforms digestion at 37°C. This was consistently demonstrated by an increase in the number of identified peptides (up to 32%) and proteins (up to 16%), alongside higher peptide signal intensities, a significant number of proteins with higher abundance (70- 80%), and comparable or lower coefficients of variation for quantified proteins. These advantages come with only a slight increase (up to 4%) in the number of missed cleaved peptides. To uncover the underlying causes of these results, we evaluated several parameters, including nonspecific cleavages, proteome physico-chemical properties, and protease enzymatic activities. Our findings suggest that the significantly reduced autolysis of trypsin and Lys-C is a crucial factor underlying the enhanced efficiency of overnight digestions at room temperature. Conclusion: This study provides strong evidence supporting the use of room temperature for all overnight enzymatic digestions with trypsin, Lys-C, or Lys-C/trypsin, given its superior performance across several methods and samples. (1) J. Proteome Res. 2018, 17, 2556−2561. 539

541

P-II-0385 Robust sample preparation workflows for protein analysis: removing complexities and variability while ensuring high-quality data Debadeep Bhattacharyya1, Ulrich Thomann1, Sameer Vasantgadkar1, Patrick McCarthy1, Eugenio Daviso1 1Covaris, Woburn, MA, United States Robust, efficient, and reproducible protein extraction and sample processing is a key step for bottom-up proteomics. While Liquid Chromatography (LC) coupled to Mass Spectrometry (MS) based workflows are sophisticated multistep processes, the resultant data quality essentially depends on the characteristics and limitations of each step employed, from sample preparation to LC-MS analysis. The sample preparation workflow is a key determinant in defining what proportion of the protein/proteome is available for end detection and analysis. Moreover, the robustness of sample preparation workflows control variation and systematic bias. In this report, we demonstrate the use of a comprehensive sample preparation workflow for generation of high- quality proteins and peptides from mammalian cells and Formalin Fixed Paraffin Embedded (FFPE) tissue samples by leveraging class-leading Adaptive Focused Acoustics® (AFA®) Technology. The AFA-enabled workflows also employ truPREP® Protein kits for cells and FFPE samples, respectively. Cells: A wide range of mammalian cells (0.1-0.5 M) were harvested followed by aliquoting 25 mL of washed and prepared cell pellets in each well of the 8 AFA-TUBE TPX strips. 25 mL of lysis buffer was added to each sample followed by AFA and centrifugation. Reduction and alkylation of the extracted proteins were performed by addition of reducing agent to each sample followed by heating at 50 °C (1 hr) and addition of alkylation reagent. Purification of samples was done using an optimal amount of magnetic bead mixture. The beads were then suspended in Trypsin/Elution buffer for accelerated digestion. Both purification and digestion processes were enabled by AFA. FFPE: 10 µm FFPE scrolls were deparaffinized with 50 µL of rehydration buffer and 450 µL Deparaffinization Solution in a prepTUBE TPX, following incubation at 56 °C for 5 minutes. After centrifugation, Deparaffinization Solution was discarded and 100 µL lysis buffer was added followed by homogenization of tissues with AFA in a Covaris R230 Focused-Ultrasonicator. De-crosslinking of proteins was conducted at 90 °C/ 90 min prior to another AFA treatment, followed by reduction and alkylation. Protein purification was conducted following the Protein Aggregation Capture (PAC) protocol with magnetic beads. The beads were suspended in trypsin buffer for overnight digestion. The purification process was enabled by AFA. The results shown in this study exhibit robust and reliable workflows using kits that can address a wide array of mammalian cells and FFPE samples resulting in high-quality proteins and peptides. The FFPE workflow and kit use a deparaffinization step that does not use any toxic and harsh, protein damaging chemicals, and it generates consistent yields of proteins from only one 10 mm FFPE scroll. For cells and FFPE samples, both protein and peptide yields were found to be very reproducible, with low CVs for intact proteins (~7%) and peptides (~9%). 541

542

P-II-0386 Scalable and comprehensive protein analysis sample preparation workflows: analyzing fresh frozen tissue samples made easy Debadeep Bhattacharyya1, Sameer Vasantgadkar1, Ulrich Thomann1, Eugenio Daviso1, Patrick McCarthy1 1Covaris, Woburn, MA, United States Human tissues are an important link between organ-specific spatial molecular information and understanding pathology, which in turn helps in optimizing treatment options. Fresh frozen (FF) tissue is often the preferred sample for proteomics because it preserves proteins in their natural state. However, patient tissues are uniquely obtained by time and location, and limited in their availability and size. Currently, little knowledge exists about appropriate and simplified protocols for routine Liquid Chromatography (LC) – Mass Spectrometry (MS)-based analysis of various types and sizes of tissues. In this report, we demonstrate the use of a comprehensive sample preparation workflow resulting in generation of high-quality proteins and peptides from FF tissues by leveraging class leading Adaptive Focused Acoustics® (AFA®) Technology. The AFA enabled workflows also employ dedicated consumables that ensure complete homogenization of tissues to extraction of proteins followed by purification and expedited digestion to peptides. The protein analysis workflow detailed in this study highlights use of a novel consumable with R230 focused- ultrasonicator that enhances homogenization efficiency of FF tissues with AFA. For sample preparation, FF tissue samples (2 mg, 5 mg, and 10 mg) were treated with AFA after they were loaded in the tube (microTUBE-130 with Bead Snap-Cap) with 120 mL of buffer (two different buffers were used - 8M Urea and Tissue Lysis Buffer). In addition to liver samples, small intestine and lung tissue samples were tested using the same workflow. Post-homogenization, the samples were centrifuged using the prototype rack followed by collection of the supernatant. BCA assay was performed with the supernatant to determine total protein concentration. The downstream protein cleanup was performed with either S-Trap or Protein Aggregation Capture (PAC) methods followed by AFA-enabled expedited digestion for generation of peptides. The homogenization tests were performed at 10 °C and in continuous mode with 5 mins of AFA treatment per column. Different time intervals were used with the above-mentioned AFA settings to ensure complete homogenization of FF liver tissue samples. While 5 mins of AFA treatment was sufficient for complete homogenization of 2 mg and 5 mg samples, 15 mins was necessary for 10 mg of samples. Average protein yield (mg) for every mg of FF tissue samples were 142 (2 mg tissue), 133 (5 mg tissue), and 102 (10 mg tissue) with the %CV not exceeding 9.2. Similar results were observed for Lung (13 mg, 15 mins) and small intestine (11 mg, 15 mins). Results in this study demonstrates an efficient, reliable, scalable workflow for a wide array of FF tissues focused on complete homogenization of samples and maximizing protein yield, while ensuring high reproducibility. The specific tubes used in this study can be used in a 96-plate format with the appropriate rack that is compatible with a host of focused ultrasonicators 542

543

P-II-0387 Protein and RNA co-extraction using RNAprotect® cell reagent Klara Borrmann1,2, Kamil Grzyb1, Gunnar Dittmar1,2 1Luxembourg Institute of Health, Department of Infection and Immunity, Strassen, Luxembourg 2University of Luxembourg, Esch-sur-Alzette, Luxembourg Proteomic and transcriptomic data are often analyzed side-by-side to yield complementary information. To minimize variation between the proteomic and transcriptomic datasets, it is essential to be able to "freeze" the proteins and mRNA to prevent any further modification or degradation during the sample preparation. RNA protection agents like RNAprotect® offer a simple way to block any modifications and are easy to use. To avoid the introduction of variations from parallel experiments for proteomics and transcriptomics, it is desirable to develop a quick and easy sample preparation method to extract proteins and RNA simultaneously from the same sample. The described method using RNAprotect® Cell Reagent (Qiagen, Belgium) can minimize variances due to multiple sampling and help increase the throughput in the analysis of large measurement cohorts. By adding RNAprotect®, the RNA and proteins are rapidly extracted and stabilized for months before the sample can be further processed for the actual measurement. The method is widely suitable for any cell culture experimental setups, including cells growing on membrane inserts. We developed a workflow for the rapid measurement of proteomic and transcriptomic data, which allows the deep analysis of proteome and transcriptome from the same sample. 543

544

P-II-0388 Optimization of Extracellular Vesicles Isolation Protocol for Proteomics Analysis from Limited Cell Media Samples Asia Botto1, Elisa Giovannetti1,2, Liam A. McDonnell1 1Fondazione Pisana per la Scienza ONLUS, San Giuliano Terme, Italy 2VU University, Department of Medical Oncology, Amsterdam, Netherlands Extracellular vesicles (EVs) are lipid bilayer-enclosed particles secreted by cells that contain proteins, lipids, and nucleic acids reflective of their cell-of-origin. These vesicles, including exosomes, microvesicles, and apoptotic bodies, are pivotal in cell communication and have potential applications in liquid biopsies and therapeutic delivery. For proteomics analysis, isolating EVs from limited biological samples poses a challenge due to the limited amount of EV proteins and the complexity of media contaminants. Here we sought to optimize an EV isolation workflow tailored for proteomics from small volumes (15 mL) of cell culture media. We compared two workflows: a combination of polymer-based precipitation followed by size exclusion chromatography (PPT+ExoSpin) and an ultrahigh-performance liquid chromatography-size exclusion chromatography (UHPLC-SEC) method. Cell media from cell lines of pancreatic ductal adenocarcinoma (PDAC) were processed, and EVs were isolated using both workflows. The PPT+ExoSpin method eliminated the need for a concentration step of the EV fractions after SEC separation, significantly reducing EV loss. Isolated EVs were characterized via transmission electron microscopy (TEM) and western blot. The PPT+ExoSpin workflow demonstrated superior sensitivity, identifying a greater number of EV-associated proteins compared to the UHPLC-SEC method. Proteomics analysis revealed that the PPT+ExoSpin method identified 97 out of the top 100 proteins listed in the ExoCarta database, as well as all previously reported markers of PDAC EVs (EGFR, AXL, GPC1, EPCAM) in all replicates. An analogous analysis for EVs isolated using the UHPLC-SEC method identified only 68 or the 100 proteins in the ExoCarta database and either did not identify PDAC EV markers (EGFR and AXL) or did so with poor reproducibility (GPC1 – 2 replicates; EPCAM – 1 replicate). The optimized PPT+ExoSpin workflow provides a reproducible and efficient approach for isolating EVs suitable for proteomics, enabling more comprehensive protein profiling from limited cell culture media volumes. 544

545

P-II-0389 A new sample preparation workflow for proteome profiling of very limited-volume murine cerebrospinal fluid Clarissa Braccia1, Cristina Lanni2, Clara Ciampi3, Francesca Fagiani1, Lorenzo di Cesare Mannelli3, Annapaola Andolfo4 1IRCCS Ospedale San Raffaele, Translational Neuropathology Unit, Division of Neuroscience, Milan, Italy 2University of Pavia, Department of Drug Sciences (Pharmacology Section), Padova, Italy 3University of Florence, NEUROFARBA Department, Florence, Italy 4IRCCS Ospedale San Raffaele, Center for Omics Sciences, Proteomics and Metabolomics Facility (ProMeFa), Milan, Italy Cerebrospinal fluid (CSF) is a clear, colourless body fluid present in the brain and spinal cord. CSF performs vital functions since it protects the brain from injury and helps in maintaining a stable chemical environment for the central nervous system (CNS), acting as waste removal. Because cerebrospinal fluid (CSF) is in direct contact with the extracellular space of the brain, it provides insight into the status of the CNS. The analysis of changes in CSF proteome of murine models could help researchers in understanding the pathophysiology of several neurological diseases, such as Alzheimer's disease, Parkinson's disease, multiple sclerosis, and amyotrophic lateral sclerosis. However, the small volume of CSF in mice has limited individual mouse proteome characterization. Moreover, the wide dynamic range of protein concentrations in CSF requires specific sample preparation for label-free untargted proteomics; the removal of the most abundant proteins, such as albumin and immunoglobulins, from CSF samples significantly enhances the ability to detect and quantify low-abundance proteins, providing a deeper and more accurate proteomic profile. This process is known as depletion, and several immunodepletion kits are available on the market, specifically designed for the depletion of high- abundance proteins in murine CSF. To overcome issues related to the low protein concentration in murine CSF (typically ranges from 0.3 to 0.5 mg/mL), and to the limited amount of CSF which can be collected from each mouse, pooling CSF from multiple mice is a common practice in proteomics to increase the protein content available for the analysis. Recently, PreOmics, a company specialized in sample preparation solutions for proteomics research, developed the ENRICH-iST kit, that enables the identification and the quantification of low-abundant proteins by using only 20 µL of murine CSF. In this way, we were able to perform label-free quantitative proteomic analysis on CSF collected from single biological replicate thus preserving the inter-individual variability of the experiment. We identified and quantified a total of 502 proteins by comparing 20 µL of CSF collected from 8 colitis-like mice to the same volume collected from control mice. Of these proteins, corresponding to 465 genes, we found 431 genes that are already known to be present in human CSF. Moreover, we found some proteins (about 1/5 of the total quantified proteins) that are significantly dysregulated in the condition of interest in comparison to the control. With these results, we demonstrated that the Enrich-iST is perfectly suited for the analysis of 20 µL murine CSF, without the need of pooling samples. 545

546

P-II-0390 Assessment of membrane peptide loss in common bottom-up proteomic workflows leads to an improved membrane-inclusive sample preparation protocol Jana Brezinova1, Michal Korecky1, Adela Pravdova1, Martin Hubálek1 1Institute of Organic Chemistry and Biochemistry of the CAS, Proteomics, Prague, Czech Republic Analysis of membrane proteins in bottom-up proteomics often relies on characterisation of their water-soluble regions. Identification coverage of transmembrane domains through widely used methods remains poor. Membrane spanning regions are difficult to analyse due to lack of trypsin cleavage sites and abundance of nonpolar amino acids. Even though membrane proteins are key drug targets, are crucial to cell transport and communication, their analysis remains a challenging task. To develop a more membrane inclusive method tailored to the analysis of membrane peptides, their losses were closely assessed during sample treatment. Three common sample preparation protocols were compared – eFASP(1), SP3(2) and SP4(3) with respect to membrane peptide identifications. Additional fractions generated during sample treatment were collected and analysed regarding membrane peptide loss. These include molecular-weight-cut-off filter washes employed in eFASP, or bead and test tube washes in SP3 and SP4 protocols, respectively. All these washes contained valuable source of membrane peptides overlooked by standard sample processing protocols. Further method modifications comprise liquid chromatography by using isopropanol/acetonitrile mixture in 0.1% formic acid as a strong eluent and dissolution of samples prior to analysis in 30% ACN. improved elution of hydrophobic peptides during Addressing challenges in hydrophobic peptide analysis in a newly modified sample preparation method enabled us to analyse underrepresented membrane regions more comprehensively in proteomic experiments. 1. Erde J, Loo RR, Loo JA. Enhanced FASP (eFASP) to increase proteome coverage and sample recovery for quantitative proteomic experiments. J Proteome Res. 2014;13(4):1885-95. 2. Hughes CS, Moggridge S, Müller T, Sorensen PH, Morin GB, Krijgsveld J. Single-pot, solid-phase- 3. enhanced sample preparation for proteomics experiments. Nature Protocols. 2019;14(1):68-85. Johnston HE, Yadav K, Kirkpatrick JM, Biggs GS, Oxley D, Kramer HB, et al. Solvent Precipitation SP3 (SP4) enhances recovery for proteomics sample preparation without magnetic beads. 2021. 546

547

P-II-0391 Automated systems to accelerate large-scale phosphoproteomics studies James Burgess1, Hannah Huckstep1, Sean Humphrey1, Lina Le1 1Murdoch Childrens Research Institute, Functional Phosphoproteomics, Melbourne, Australia James Burgess1, Hannah Huckstep1, Lina Le1, Sean J. Humphrey1 1Functional Phosphoproteomics, Murdoch Children"s Research Institute, Royal Children"s Hospital, Melbourne, Victoria 3052, Australia Protein phosphorylation is involved in almost all cellular processes. The ability to study phosphorylation in a global and unbiased manner using mass spectrometry (MS-based phosphoproteomics) is transforming our understanding of cell biology. Advances in mass spectrometry instrumentation together with streamlined and efficient workflows for performing phosphopeptide enrichment now make it feasible to process hundreds of phosphoproteome samples in parallel. However, workflows for upstream sample generation present a bottleneck, limiting the throughput of large-scale phosphoproteomic studies. Here we describe an automated platform for mass spectrometry-based phosphoproteomics, enhancing reproducibility and throughput through three key innovations: automated and semi-automated human-in-the-loop experimental manipulation, multi- instrument control by programming of a robotic arm with high degrees of freedom, and sustained high- throughput operation. 1) Automated Experimental Manipulation: Advanced liquid handling instruments by vendors like Tecan and Opentrons enable standardisation of sample preparation process, including the application of compounds or ligands, cell lysis, and protein digestion. Automating sample handling using these instruments minimises human error, ensuring consistent and precise manipulation and facilitating large batch processing. 2) Robotic Arm Programming: Open-source libraries for motion planning, collision aware inverse kinematics and environmental simulations enable programming of collaborative robotic arms to perform complex sample transfers in tandem with liquid handling and cell incubators, ensuring precise, timely movements between machines and reducing the risk of cross-contamination. 3) Sustained High-Throughput Operation: Our platform is designed to achieve sustained throughput in phosphoproteomics studies. This sustained operation will facilitate comprehensive phosphoproteome characterisation, greatly increasing the quantity and quality of data gathered as compared to manual operations in a similar timeframe at a fraction of the cost. Conclusion: Our platform improves existing phosphoproteomics workflows by providing an economical, reliable and high-throughput solution for the generation and handling of biological samples. By integrating liquid handling machines and incubators using robotic arm programming, we enable studies otherwise out of reach. We will first apply our platform to study the impact of a library of small molecules on the phosphoproteome. We plan to subsequently use this data to elucidate novel pathways and effector proteins specifically responsible for observed cellular phenotypic effects of the compounds in a disease context. 547

548

P-II-0392 Evaluation of different depletion methods and development of an optimized LC-MS/MS method for murine deep plasma proteomics in preclinical safety assessment studies Benita M. Burghardt1, Johannes R. Schmidt1, Jörg Lehmann1,2, Stefan Kalkhof1,3 1Fraunhofer Institute for Cell Therapy and Immunology IZI, Department of Preclinical Development and Validation, Leipzig, Germany 2Fraunhofer Cluster of Excellence Immune-Mediated Diseases CIMD, Frankfurt/Main, Hannover, Leipzig, Germany 3Institute for Bioanalysis, University of Applied Science Coburg, Coburg, Germany Introduction: Plasma proteomics emerges as a powerful tool for identifying potential off-target effects, unwanted on-target effects and safety concerns associated with novel therapeutic approaches in preclinical studies. Besides the high potential to detect systemic biological changes at an early stage, there are challenges to proteomic plasma analysis. The main limiting factor often displays the large dynamic range of abundance of plasma proteins, which can lead to masking of low abundant proteins, and thus a limited number of proteins that can be identified and quantified. To address this challenge, various methods exist to deplete the high abundant protein fraction, thus enabling deep proteomic profiling. The aim of the study was to apply a dual approach to optimize the proteomic analysis of murine plasma for safety assessment in preclinical animal studies. Methods: For this purpose, different depletion methods (immunodepletion, selective acid precipitation and ProteoMiner™ protein enrichment technology) were evaluated in comparison to the analysis of neat plasma based on the identification (ID) rates, the number of reproducibly quantified proteins as well as the technical variance and the correlation of protein abundances in depleted and neat plasma. Based on these results, a suitable sample preparation method for preclinical toxicological safety assessment in mice was selected. The second part of the study comprised the optimization of the LC-MS/MS method as well as the generation of a spectral library to further deepen the proteomic analysis. Sample preparation included the depletion of the high abundant protein fraction or dilution of the neat plasma followed by protein digestion using a filter-aided sample preparation protocol. Downstream analysis was performed using a nano-LC system (nanoElute 2, Bruker) online coupled to timsTOF Pro 2 mass spectrometer (Bruker) with parallel accumulation and serial fragmentation. Data were mainly analyzed by library-free as well as library-based searches using FragPipe and DIA-NN computational platforms. Results: The evaluated depletion methods differ in terms of ID rates, reproducibility and number of exclusively quantifiable proteins. While the highest ID rates were determined with depletion by acid precipitation and ProteoMiner™ protein enrichment technology, the measurement of neat plasma is characterized by its high reproducibility. Optimization of the MS and tims parameters (mass range, ion mobility range and window size as well as sample amount and ramp time) further deepened the proteomic coverage and enabled the identification of more than 800 murine plasma proteins with a 70-minute gradient. Conclusion: The presented method is suitable for deep profiling of the murine plasma proteome and can be used in preclinical safety assessment studies in future. 548

549

P-II-0393 Optimized automated workflow for BioID improves reproducibility and identification of protein-protein interactions Emilio Cirri1, Hannah Knaudt1, Domenico Di Fraia1,2, Nadine Pömpner1, Norman Rahnis1, Ivonne Heinze1, Maleen Hoffmann1, Alessandro Ori1,3, Therese Dau1,2 1Leibniz Institute on Aging – Fritz Lipmann Institute, Jena, Germany 2University of Rochester, Rochester, NY, United States 3Genentech, Inc., San Francisco, CA, United States Proximity dependent biotinylation is an important method to study protein-protein interactions in cells, for which an expanding number of applications has been proposed. The laborious and time consuming sample processing has limited project sizes so far. Here, we introduce an automated workflow on a liquid handler to process up to 96 samples at a time. A sample preparation workflow for enrichment and digestion of biotinylated proteins (Bartolome, A et al. bioRxiv 2022) was adapted and implemented on a liquid handler (Agilent Bravo). Here, streptavidin beads are acetylated prior to loading of the samples to decreases streptavidin contamination after on-bead digestion with LysC. In a second elution, a mixture of acetonitrile and trifluoroacetic acid is used to retrieve biotinylated peptides. As this step is time sensitive, automatization on a liquid handler improves reproducibility. In addition, the impact of the sample input as well as the LC gradient was analysed. The sample input could be reduced from 20 Mio to 8 Mio cells and the gradient shortened from 120 minutes to 21 minutes without compromising the efficiency of biotinylated protein enrichment. Taken together we were able to implement a new workflow that improves reproducibility and speeds up sample processing and measuring time for proximity labelling experiments considerably, and we successfully applied this workflow to optimize the detection of proteasome substrates by proximity-dependent labeling. 549

550

P-II-0394 Overcoming the dynamic range difficulties in CSF samples through a novel enrichment step for biomarker discovery studies Katharina Limm1, Zuzana Demianova1, Katrin Hartinger1, Nils Kulak1 1PreOmics GmbH, Planegg, Germany Introduction: Cerebrospinal fluid (CSF) is an essential body fluid that is particularly important for investigating brain disorders. However, CSF not only exhibits a very high dynamic range, but also has a very low protein concentration, which poses a particular challenge for proteome analysis. ENRICH-iST is a novel high-throughput workflow that addresses the dynamic range challenge in plasma/serum samples by providing a fast and straightforward sample preparation solution for LC-MS-based proteomic studies. It is based on the enrichment of low-abundance proteins, resulting in a significant increase in protein identifications and improved proteome coverage. Thus, we evaluated and optimized the ENRICH-iST workflow for CSF samples to provide an easy-to-use and standardized workflow for in-depth proteome analysis. Methods: The ENRICH-iST workflow is based on the novel ENRICH technology that provides efficient dynamic range compression by enriching in biofluid samples onto non-functionalized paramagnetic microbeads. To address the challenge of low protein concentration in CSF samples, the ENRICH step was optimized by adjusting the starting volumes. For subsequent LC-MS sample preparation, the iST-BCT protocol optimized for biofluids (PreOmics) was employed for on-bead denaturation, reduction, alkylation, digestion, and peptide cleanup. Peptides were analyzed by nanoLC coupled to a timsTOF instrument (Bruker) using dia-PASEF® acquisition mode. low-abundance proteins Results: Considering the low concentration of CSF samples (~0.1 mg/mL) compared to plasma/serum samples (~60-80 mg/mL), the ENRICH step must be adapted to ensure successful dynamic range compression. Therefore, different input volumes of CSF were compared, and the workflow was optimized accordingly. Of note, the ENRICH-iST workflow has been successfully applied to CSF samples with input volumes ranging from 20 µl to 200 µl, resulting in improved proteome coverage. The workflow is therefore suitable for a wide range of starting volumes and can be scaled to the available CSF volume, providing maximum flexibility. Processing 200 µL of rat CSF with ENRICH-iST resulted in the identification of ~2300 protein groups; this corresponds to a 1.8-fold increase in protein identifications compared to the iST-BCT workflow without enrichment. Furthermore, as an antibody-free approach, the ENRICH-iST workflow is not specific to human samples and can be applied to other mammalian species, making it particularly interesting for preclinical studies. Conclusion: The optimized ENRICH-iST workflow enables in-depth proteome analysis for large CSF cohorts, simplifying and improving biomarker discovery studies. 550

551

P-II-0395 An efficient and robust protocol for quenching TMT-labelling reactions for reversal of over-labelling at serine, threonine, and tyrosine residues Yana Demyanenko1, Andrew Giltrap1, Eduardo Kitano1, Shabaz Mohammed1 1The Rosalind Franklin Institute, Didcot, United Kingdom Tandem mass tags (TMT) are widely used for relative quantification of protein abundance. The labels are introduced onto primary amines of lysine residues and N-termini of tryptic peptides via amide formation upon reaction with their N-hydroxy succinimide (NHS) ester moiety. Although otherwise rare products, under the conditions required for complete labelling, seryl-, tyrosyl-, and threonyl- esters of the TMT label can form on up to 15% of peptides in addition to the expected amide derivatives. These peptides are typically classed as over- labelled and are disregarded for quantification. Their presence also unnecessarily increases sample complexity, which reduces identification rates. Hydroxylamine was shown to reverse the formation of tyrosyl esters and has since been universally used to quench TMT-labelling reactions. However, our data shows that hydroxylamine treatment has limited efficacy against seryl- or threonyl- esters. Through systematic screening of nucleophilic aminolysis reagents and reaction conditions, we developed a robust method to remove all three ester types from over-labelled peptides across a wide range of reaction scales. The new method reliably reduces the proportion of over-labelled peptides in the sample to less than 1% without affecting the labelling rate or introducing other modifications, leading to an increase in identification rates from single-shot analysis. 551

552

P-II-0396 Quantitative proteomics of small numbers of closely-related cells: selection of the optimal method for a clinical setting Kyra van der Pan1, Sara Kassem1, Indu Khatri1,2, Fadi al Makindji1, Eftychia Stavrakaki3, Anniek L de Jager1, Brigitta AE Naber1, Inge F de Laat1, Alesha Louis1, Wouter BL van den Bossche3, Lisette B Vogelezang3, Rutger K Balvers3, Martine LM Lamfers3, Peter van Veelen4, Alberto Orfao5,6,7, Jacques JM van Dongen1,5,6,7, Cristina Teodosio1,5,6,7, Paula Díez1,5,6,7 1Leiden University Medical Center, Department of Immunology, Leiden, Netherlands 2Leiden University Medical Center, Leiden Computational Biology Center, Leiden, Netherlands 3Erasmus University Medical Center, Department of Neurosurgery, Erasmus, Netherlands 4Leiden University Medical Center, Center for Proteomics and Metabolomics, Leiden, Netherlands 5Cancer Research Center, Translational and Clinical Research Program, Salamanca, Spain 6University of Salamanca, Department of Medicine, Salamanca, Spain 7Institute of Biomedical Research of Salamanca (IBSAL), Salamanca, Spain In the era of single-cell analysis, the application of mass spectrometry (MS)-based proteomics to paucicellular samples remains limited. While new approaches have emerged for single-cell studies, most have not been standardized and/or often rely on specialized, custom-built equipment, hindering widespread implementation, particularly in non-specialized settings. Hence, it is necessary to select an optimal MS-oriented proteomics approach suitable for clinical settings. To this aim, we assessed ten different sample preparation procedures in paucicellular samples of closely-related cell types, using monocytic cells as a model system. Five cell lysis protocols with varying chemistries and mechanical forces were combined with two sample clean-up techniques (C18 filter- and SP3-based), followed by tandem mass tag (TMT)-based protein quantification. The assessment was performed in three phases: i) selection of the sample preparation approach, employing hematopoietic and non-hematopoietic cell lines. This analysis identified the combination of a urea-based lysis buffer with the SP3 bead-based clean-up system as the best performer (allowing the reliable characterization of proteomes using only 2,500 cells), considering parameters such as reproducibility, accessibility, spatial distribution, ease of use, processing time and cost. ii) Evaluation of the method on maturation-related cell populations: classical, intermediate and non-classical monocytes from peripheral blood and macrophages/microglia (MAC) from glioblastoma samples. Closer protein expression profiles were observed between non-classical monocytes and MAC, with the latter showing the co- expression of M1 and M2 markers, although pro-tumoral and anti-inflammatory proteins were more represented. iii) Validation of results by high-end spectral flow cytometry on paired monocyte/MAC samples to determine the sensitivity of the selected MS approach. The feasibility of the urea-based lysis and SP3 bead-based clean-up method was proven in 194 additional samples corresponding to 38 different cell types, including e.g. distinct tissue origins, lineages, phenotypes and stimulus conditions. In summary, we selected a reproducible, easy-to-implement sample preparation method for MS-based proteomic characterization of paucicellular samples, applicable to functionally closely-related cell populations. This approach addresses the current limitations in single-cell proteomics and enables broader implementation in clinical settings. 552

553

P-II-0397 Recombinant production and surface immobilization of thermostable trypsin for optimized antibody digestion Sarah Döring1, Birte Wulfes1 1Bundesanstalt für Materialforschung und -prüfung (BAM), Analytical Chemistry; Reference Materials, Berlin, Germany Therapeutic monoclonal antibodies are the fastest-growing class of biological agents generating yearly sales of approx. USD 220 billion in 2023 which is expected to triple over the next decade [1]. With the growing market significance of therapeutic antibodies, analytical methods allowing reproducible and metrologically traceable quantification need to be developed. The serin protease trypsin has become the gold standard enzyme for digesting intact protein into peptides for LC-MS analysis. However, many protocols exist that often lead to different results depending on the digestion procedure [2]. In particular, the type and amount of detergents added during sample preparation for protein unfolding before digestion is known to create undesirable structure modifications of proteins (e.g. carbamylation) and significant bias in measurement results by inhibiting enzyme activity [3]. The enzymes themselves can also have an influence, whereby immobilized variants are increasingly being used to improve their stability, minimize self-digestion and enable a higher substrate-enzyme ratio [4]. Here, we studied the implementation of a thermostable[5] and surface-functionalized trypsin-variant for improved antibody digestion. Applying this enzyme should allow digestion at elevated temperatures where the protein is naturally unfolding, thereby increasing enzyme accessibility and reducing the impact of detergents on the overall degree of unfolding. For this purpose, the inactive proenzym was recombinantly produced in inclusion bodies of E. coli and successfully folded back into functional conformation via multi-stage dialysis using a cysteine-cystine redox system. Only the enzyme showed activity after cleavage of the propeptide, which strongly reduces the possibility of self-digestion during production and subsequent immobilization steps. To further enhance enzyme stability, the enzyme was coupled covalently onto magnetic particles and investigated with regard to changes in activity. Compared to wild-type and a non-immobilized enzyme, the surface- functionalized thermostable variant showed increased activity at temperatures above 80°C. Thus, this novel type of trypsin particles promises to not only improve antibody digestion at high temperatures, but also to prevent self-digestion and enable the separation of trypsin from target peptides in LC–MS analysis. References [1] Monoclonal Antibodies Market Outlook (2023 to 2033). https://is.gd/d9oVOT. [2] Burkhart, J. M., et al. J Proteomics 75.4 (2012): 1454-1462. DOI: 10.1016/j.jprot.2011.11.016. [3] Proc, J. L., et al. J Proteome Res 9.10 (2010): 5422-5437. DOI: 10.1021/pr100656u. [4] Naldi, M. et al. J Pharm Biomed Anal, (2018): 160, 222-237. DOI: 10.1016/j.jpba.2018.07.051. [5] Xiao, T., et al. Process Biochemistry 130 (2023): 227-235. DOI: 10.1016/j.procbio.2023.04.024. Fig. 1 (optional) 553

554

P-II-0398 Comparison of plasma and serum abundant protein depletion methods Félix Elortza1, Javier Beaskoetxea1, Ibon Iloro1, Iraide Escobés1, Mikel Azkargorta1 1CIC bioGUNE, BRTA, Proteomics Platform, Derio, Spain Blood is the most collected biofluid globally for clinical purposes and is used in approximately 40% of all completed tests. Consequently, plasma and serum are among the preferred sources for potential biomarker discovery and are of major interest in liquid biopsy research. Identification of protein biomarkers in plasma has been limited, among others, because the 22 most abundant proteins in plasma account for 99% of all protein content, preventing the detection of less abundant proteins using discovery mass spectrometry (MS)–based proteomics. Therefore, depletion of the most abundant proteins is a common way to circumvent the dynamic range issue and different immunodepletion-based resins have been used for this purpose. Steen"s laboratory has recently published a method based on perchloric acid precipitation that is described to be fast-forward and cost-effective (1). With the aim of sizing the overall performance of different depletion methods, we have accomplished a comparison among them: a) Immunodepletion by a commercial Top 14 protein depletion cartridge (Thermo Fischer); b) Perchloric acid depletion; c) Trichloro acetic acid (TCA) depletion. The depletions performance was checked by DDA and DIA in an EVOSEP + TimsTOF Pro system. The overall results show that the three methods tested work in similar but complementary ways. The number of proteins identified by each approach is comparable, with some proteins being exclusive to each method. Precipitation-based approaches are methodologically easier to carry out in a high-throughput manner and are significantly cheaper. Consequently, we expect these methods to become routine in plasma-based biomarker discovery. References Viode A, van Zalm P, Smolen KK, Fatou B, Stevenson D, Jha M, Levy O, Steen J, Steen H; IMPACC Network. A simple, time- and cost-effective, high-throughput depletion strategy for deep plasma proteomics. Sci Adv. 2023 Mar 29;9(13)eadf9717. 554

555

P-II-0399 High throughput abundant protein depletion for plasma proteomics Amarjeet Flora1, Anastasia Klenke1, Bhavin Patel1, Ryan Bomgarden1, Nikki Jarret1 1Thermo Fisher Scientific, R&D, Rockford, IL, United States Biological fluids like plasma, etc. are widely used for proteomics-based disease biomarker discovery. However, its analysis is challenging due to the complexity and wide dynamic range of the proteome, ranging over 10 orders of magnitude. The high abundant proteins like albumin, IgGs etc. constitute about 90% of the plasma and their presence masks the identification of the low abundant proteins in the sample. Several workflows have been developed around depleting these abundant proteins to enable the deep penetration into the proteome to identify low abundant proteins of interest. These workflows are laborious with multi-step protocols which greatly increases hands on- time while processing multiple sample types. To address these issues, we developed a 96 well plate format of the depletion resin to support higher sample throughput. This format can deplete 96 samples at a time with a depletion efficiency of ≥ 95% and showed similar performance to our spin column format in terms of protein/peptide identifications and Top14 abundant proteins depletion in the plasma samples improving the proteome coverage by 2.4-fold. We have also evaluated our format with different biological fluids, such as CSF etc. In addition, our depletion resin in 96 well format combined with our optimized EasyPep chemistry can process these samples in 7-8 hours with high reproducibility and <5% CVs. These two combined can significantly eliminate the hands-on-challenges while handling a large number of samples. Moreover, combining our depletion resin in plate format with EasyPep96 magnetic using an automated KingFisher platform generated ready-to-inject samples in less amount of time and showed similar protein/peptide identifications with ≥ 95% depletion efficiency compared to other workflows. Differentially expressed proteins in a cancer vs. normal human plasma samples were identified and quantified using a label-free data-independent acquisition (DIA) strategy. Overall, we demonstrate that our depletion resin in a 96 well format enables the depletion of Top14 abundant proteins from the biological fluids with a greater depletion efficiency for high throughput plasma proteomics. 555

556

P-II-0400 Evaluation of novel chemical derivatization methods for N-Terminal sequencing of proteins Peter Gehrig1, Simone Wüthrich1, Ralph Schlapbach1, Paolo Nanni1 1University of Zürich, Functional Genomics Center Zurich, Zurich, Switzerland Introduction: N-terminal protein sequencing permits reliable identification of real N-termini of mature proteins, and it provides essential insights into protease substrates and specificity. A critical step in MS-based N- terminomics workflows is the modification of primary amines present at the protein N-terminus and in lysine residues. Reductive methylation is such a strategy, which has been widely used for N-terminal protein sequencing. Chemically, two methyl groups are added to primary amines in a mixture of formaldehyde and sodium cyanoborohydride. In this study, new methods for N-terminal modification of proteins were introduced and applied to commercially available proteins. For the dimethylation reaction, we replaced the commonly applied, but highly poisonous reducing reagent sodium cyanoborohydride by pyridine borane, which showed a similar reactivity. Furthermore, we modified free amino groups with nicotinic acid N-hydroxysuccinimide (NHS) ester as an alternative to the dimethylation reaction. Finally, we evaluated the one-step reaction with 2- pyridinecarboxaldehyde (2PCA), which is specific for protein N-termini due to a unique cyclization reaction involving the second amino acid. Methods: Selected proteins (BSA, myoglobin, α-casein, α-lactalbumin, α-casein, ß-galactosidase) were modified with a mixture of formaldehyde and pyridine borane, nicotinic acid NHS ester, or 2PCA. The reagents were quenched and removed, and the labeled proteins were cleaved with trypsin or chymotrypsin using the SP3 approach. The resulting peptides were subjected to LC-MS and MALDI-TOF analysis. Results and Conclusions: Important criteria for chemical derivatization methods are complete reactions and the absence of undesired side reactions. In addition, chemical modifications of peptides should have a favorable influence on MS signal intensities and on the fragmentation spectra. The results for the dimethylation reaction showed that pyridine borane promotes dimethylation at levels comparable to those recommended for the standard reducing agent sodium cyanoborohydride. Nearly complete dimethylation of N-termini and lysines was achieved, and we recommend replacement of the highly toxic cyanoborohydride by pyridine borane for this application. Nearly quantitative reactions and only minor side reactions were observed for the nicotinic acid NHS ester modification of proteins. Interestingly, tandem mass spectra acquired from peptides carrying this modification gave higher database search scores due to more intense b ions including b1 ions, which are not detectable otherwise. Our preliminary results confirm that the reaction with 2PCA is largely specific for N-terminal amines. Further optimization is required to increase the yield beyond approximately 50%. 2PCA labeling is complementary to most N-terminal labeling methods since lysine residues remain unmodified and are available for cleavage by proteases. 556

557

P-II-0401 Optimization of sample preparation strategies for high-throughput plasma proteomics Cristina Gil Gonzalez1, Paola Pisano1, Michael Wierer1 1Proteomics Research Infrastructure/University of Copenhagen, Copenhagen, Denmark The discovery of biomarkers is essential for combating diseases and maintaining public health. Biomarkers enable prevention, early diagnosis, and personalized treatment of diseases. Plasma serves as a primary source for biomarker discovery due to its ease of accessibility, and reliability in assessing overall health status. Despite a significant increase in biomarker studies in recent years, only a small number of biomarkers have been approved and translated into clinical practice. This is mainly due to the numerous challenges that persist in identifying reliable biomarkers, most predominantly the ability to run large sample cohorts, and the proteomic depth that can be achieved. The aim of this study is to enhance the efficiency and throughput in plasma proteomics by implementing and systematically comparing different sample preparation workflows. This includes (i) various lysis conditions with optional clean-up steps for non-enzyme-compatible buffers, (ii) different enzyme combinations for protein digestion, and (iii) different peptide purification strategies. All samples will be loaded onto Evotips and analyzed on a Evosep- Astral LC-MS setup. The ultimate goal is to find the optimal method for large-scale analysis of neat plasma, surpassing the current widely-used protocol consisting of heat denaturing, trypsin-lys-c digestion and Evosep desalting. Coupled with the exceptional sequencing speed of the Thermo Astral instrument, this will pave the way for accurate and precise high-throughput analysis of large plasma cohorts, facilitating the identification of new, relevant biomarkers. 557

558

P-II-0402 Advancing plasma proteome profiling through cost effective, robust, high-throughput and orthogonal workflows based on hyper-porous magnetic beads Stoyan Stoychev1,2, Ireshyn Govender1,3, Adele Nel1,4, Andrea Ellero1,4, Previn Naicker3, Isak Gerber3, Justin Jordaan1, Christine Wu5, Michael J MacCoss5 1ReSyn Biosciences, Pretoria, South Africa 2Evosep Biosystems, Odense, Denmark 3CSIR – National Chemical Laboratory, Pretoria, South Africa 4University of Pretoria, Pretoria, South Africa 5University of Washington, Department of Genome Sciences, Seattle, WA, United States Plasma proteome profiling is challenging due to the absence of efficient workflows that (i) allow for a deep dive into the plasma proteome, (ii) utilize existing automation infrastructure, (iii) are scalable in terms of ability to process large patient cohorts in a reproducible, and cost-effective manner. Here we present an evolution of the Mag-Net method for deep plasma profiling (Wu et al, 10.1101/2023.06.10.544439) by employing combinations of hyper-porous magnetic microparticles as well as further optimizing the workflow by integrating additional orthogonal steps for plasma depletion and/or enrichment aimed at further increasing proteome depth. Plasma from a single donor was collected using EDRN SOP (Tuck et al., 10.1021/pr800545q), a range of hyper- porous beads with varying functional groups, together with MagReSyn® SAX as a control, were utilized with the Mag-Net protocol in different combinations. Further depletion and enrichment steps were integrated with Mag- NetTM. The methods were performed in semi-automated manner using a King Fisher magnetic handling platform or fully automated on an Opentrons OT2 robot. Peptides were loaded onto Evotips and analyzed using an Evosep One coupled to a Bruker timsTOF HT. Data was processed using DIA-NN (version 1.8.1) in library-free mode. Post data processing was performed in R for the comparison of each experimental condition. Preliminary data indicates that the use of novel mixed mode magnetic beads increases plasma proteome depth by approximately 25%. Whilst the addition of complementary enrichment steps utilizing bead combinations increase plasma proteome coverage by 30%, identifying up to 3,600 protein groups at a throughput 100SPD. This was achieved without sacrificing robustness or the requirement for additional LCMS time. The evaluation of additional bead chemistries and workflow variations are ongoing. 558

559

P-II-0404 Efficient and repeatable peptide desalting with AttractSPE®Disks tips C18 in manual and fully automated proteomics workflows Florine Hallez1, Michel Arotçarena1, Sami Bayoudh1, Kaynoush Naraghi1, Florent Dingli2, Damarys Loew2, Célia Jardin3, Soumia Hamada3, Cédric Pionneau3, Martin Technau4, Solenne Chardonnet3 1Affinisep, Le Houlme, France 2Institut Curie, PSL Research University, Paris, France 3Sorbonne University, Inserm, UMS PASS, P3S platform, Paris, France 4CEM GmbH, Kamp-Lintfort, Germany Bottom-up approaches are commonly used for proteomics analysis in LC-MS/MS. However, salts present in the digestion buffers can generate ion-suppression that limits peptide detection and greatly impacts proteins identification. Several AttractSPE®Disks Tips C18 were evaluated for peptide desalting. These SPE Tips, packed with small sorbent beads embedded in a thin and mechanically stable membrane that combines high capacity and small dead volume, are particularly adapted for centrifugation or positive pressure assays. First, 200µL AttractSPE®Disks Tips C18 with different binding capacities were tested for the manual desalting of 100ng of HeLa digest by centrifugation. Five replicates of each assay were carried out and the nanoLC-MS/MS analyses of peptides were compared to those obtained after direct injection of a non-purified loading solution. This comparison showed that 97% of proteins could be identified after desalting on the C18 tips, with high recovery of all peptides on the whole polarity range. Low variability was obtained for intra- and inter-tips assays (RSD < 3%), highlighting the Tips reliability. With the aim to automate peptide desalting, the P3S plateform has asked CEM (previously Intavis) to create a custom needle to run home-made StageTips on the DigestPro MSi robot. The program was adapted from manual StageTips protocol and further optimized by playing with liquid aspirate and dispense speeds and volumes. Moreover, P3S has evaluated AttractSPE® C18 Tips compared to home-made StageTips (with Empore C18 SPE disks) for 2µg and 100ng of protein digest. The comparison showed little difference between the two kinds of tips when working with 2µg of peptides but showed a 10% increase in the number of identified proteins with the AttractSPE® C18 Tips when working with 100ng of peptides. Moreover, the peptide intensities were higher with the AttractSPE® C18 Tips. In both cases, 2µg and 100ng starting material, the AttractSPE® C18 Tips performed better in the hydrophilic range, thus demonstrating than this C18 sorbent offers a wider spectrum of interactions compared to home-made StageTips. In addition, the P3S team has optimized the program on the DigestPro MSi robot to reduce the time from 40min down to 10min per sample. This study has shown that the AttractSPE®Disks Tips C18 can be used both for manual or automated desalting workflows, and are perfectly adapted to single cell-like analysis as well as high peptide amounts, thus covering a wide range of applications. 559

560

P-II-0405 Non-small cell lung cancer biomarker discovery from unfractionated blood cell pellets Ben Herbert1, Cameron Hill1, Elisabeth Karsten1, Dana Pascovici1,2, Rosalee McMahon1, Natasha Lucas1 1Sangui Bio, R&D, St Leonards, Australia 2CSIRO, Sydney, Australia Blood biomarker studies are almost exclusively performed on plasma or serum. The blood collection and sample preparation steps are optimised to avoid haemolysis where possible because of the additional dynamic range issues arising from haemoglobin and the potential for artefactual results in immunoassays. In some clinical studies the cell pellet after plasma removal is stored frozen, although the lysis of cells exacerbates the sample preparation challenges for proteomics. In this study we obtained plasma and frozen cell pellets from 16 patients with clinical stages II – IV non-small cell lung cancer (NSCLC) and 18 age and sex matched healthy controls. For proteomics analysis we have focused on the cell pellets and used a novel sample preparation technique to enable broad coverage of soluble, intracellular and membrane proteins. In a previous study we demonstrated detection and robust quantification from single- shot shotgun-LC-MS analysis of dried whole blood1. We used volumetric absorptive micro-sampling (VAMS) dried blood spot devices and loaded 30µL aliquots of thawed cell pellet sample. The samples were dried, washed, and trypsin digested in situ. Peptides were separated with a one hour gradient and quantified using DIA on a QE-HFX Orbitrap, producing ~3,700 protein IDs for each sample. There were 508 differentially expressed proteins initially identified. Ingenuity pathway analysis revealed these identified proteins were involved with a variety of functional pathways covering adhesion and migration, as well as a strong network of known cancer-associated cytokines and enzymes. To reduce complexity, proteins were ranked based on area under the curve (AUC), and separately using a boosted regression importance filter. The first filter produced a short list of markers that were both differentially expressed for the model and had an AUC > 0.9. The second filter calculated the importance rank of each protein using gradient boosting methods. The procedure was repeated 100 times, and the markers were ranked in terms of the number of times they were selected in the top 10 importance rank was recorded for each protein. Using the methods described above a set of 14 markers were identified using the AUC filter, and 13 markers were identified using the importance rank filter that discriminate between NSCLC and healthy controls with 3 markers common to both analyses. Our rapid and reproducible methods enable the production of high-quality data from small aliquots of complex samples that are typically seen as requiring significant fractionation prior to proteomic analysis. 1. Molloy, M. P., et al. Proteomic analysis of whole blood using volumetric absorptive microsampling for precision medicine biomarker studies.Journal of Proteome Research. (2022). 560

561

P-II-0406 Comparative analysis of recombinant trypsin variants for enhanced MS-based peptide identification and quantification Cristina Hernandez Rollan1, Tanveer Singh Batth1, Jesper Olsen1, Morten Nørholm2 1University of Copenhagen, NNF Center for Protein Research, Copenhagen, Denmark 2Technical University of Denmark, NNF Center for Biosustainability, Lyngby, Denmark Introduction: Trypsin is the most popular protease in proteomics for the digestion of proteins into smaller peptides, making them ideal for liquid chromatography-tandem mass spectrometry (LC-MS/MS) analysis. One of the most used Trypsin proteases in proteomics is the native Trypsin purified from pig pancreas. However, native Trypsin is prone to autolysis, leading to the formation of pseudotrypsin, which can interfere with peptide spectra analysis. To mitigate this, native Trypsin is modified by reductive methylation of lysine residues, enhancing its activity by preventing autolytic digestion. In addition, native Trypsin is further treated with TPCK to inhibit possible chymotrypsin contamination activity. Due to these extensive modifications, the purification of Trypsin from porcine pancreas remains challenging. Recombinant Trypsin produced by genetic engineering offers high purity and batch-to-batch consistency, presenting distinct advantages in proteomics studies, such as the absence of contaminating chymotrypsin, increased stability, minimized autolysis, and enhanced reproducibility across experiments. In this study, we compared the performance of commercial Trypsin with recombinant Trypsins from Sus scrofa and Streptomyces griseus, both produced in the yeast Pichia pastoris. We evaluated digestion efficiencies, missed cleavage rates, and substrate efficiency at different time points and conditions. Methods: Codon-optimized sequences for S. scrofa and S. griseus Trypsin genes were synthesized and inserted into Pichia pastoris. Induced under controlled fermentation conditions, recombinant Trypsins were harvested and purified. Protein extracts from HeLa cell line underwent a robust proteomics sample preparation employing the Protein Aggregation Protocol (PAC) protocol. Recombinant and commercial sequencing-grade Trypsins were compared for protein digestion under different digestion conditions. Mass spectrometric analysis was performed using an EvoSep One liquid chromatography system coupled to an Astral mass spectrometer operated in data- dependent acquisition (DDA) mode. MaxQuant software suite was utilized for peptide identification and quantification prior to further downstream data processing using R. Results: Recombinant Sus scrofa Trypsin demonstrated comparable performance and efficiency at digesting cellular extracts as commercial Trypsin, with no significant impact from the absence of lysine methylation, even during extended digestion times (24 hours). Recombinant Streptomyces griseus trypsin exhibited good cleavage efficiency but identified fewer peptides than porcine Trypsins. Additionally, we evaluated the performance of recombinant trypsin in combination with LysC, another commonly used enzyme in proteomics. Our results show that the digestion efficiency improves simultaneously as the miss cleavages decrease when both enzymes are combined. 561

562

P-II-0407 A sequential (phospho)proteomics enrichment approach reveals distinct protein phosphorylation patterns of mesothelial cells exposed to high glucose containing PD fluid or heat stress Rebecca Herzog1, Elisabeth Lang1,2, Anja Wagner1,2, Markus Unterwurzacher1,2, Klaus Kratochwill1,2 1Medical University of Vienna, Pediatric Nephrology, Vienna, Austria 2Medical University of Vienna, Proteomics Core Facility, Vienna, Austria Peritoneal dialysis (PD) is a kidney function replacement therapy, where the peritoneum is used as a semi- permeable membrane to remove water, toxins, and solutes from patients by a glucose-based PD-fluid. Exposure to PD-fluid triggers changes in peritoneal mesothelial cells and has been shown to impair cellular stress response (CSR) mechanisms. Glucose degradation products (GDPs) in the PD-fluids lead to diabetes-like vasculopathy and are speculated to block critical elements of the CSR which is highly regulated via PTMs. This study aims to identify system-wide protein phosphorylation impairing the CSR leading to deterioration of the peritoneal membrane and poor clinical outcome. Human mesothelial cells exposed to heat, PD-fluids or GDPs were analyzed with a quantitative (phospho)proteomics approach. 68 samples were labeled with TMTpro-18-plex. Phosphopeptides were enriched in a sequential approach, first using TiO2, followed by FeNTA enrichment of the flow-through. The two phospho- peptide enriched eluates were combined and samples were high-pH RP offline-fractionated. Both sample types were analyzed by nano-LC /Orbitrap-Exploris480 with FAIMSpro. An internal pooled standard consisting of equal parts of all samples was used to link multiple TMT-runs. Database search and quantification were performed using Proteome Discoverer. The approach enabled identification of >10,100 proteins (>9,000 with abundance in all runs), 46,848 phospho- peptides and 38,336 unique phospho-sites. 83% phospho-sites were serine residues, 16% threonines and 1% tyrosines. The analysis of the total proteome showed a shift to biological processes related to chaperone activity and CSR in heat but not in PD-fluid conditions. The quantitative phosphoproteomics approach revealed differentially de/activated players in the HSF-1 activation pathway and its effector kinases depending on the presence of GDPs. This approach enables high coverage of the total and phosphoproteome from the same sample and thereby enabling analysis of relevant processes of the mesothelial reaction to PD-fluid induced stress possibly leading to identification of new therapeutic targets. 562

563

P-II-0408 Improved proteomic methods enabled by a fast and efficient arginine-specific protease Christopher Hosfield1, Jessica Wohlfahrt2, Jennifer Guergues2, Marjeta Urh1, Michael Rosenblatt1, Stanley Stevens Jr.2 1Promega, Madison, WI, United States 2University of South Florida, Tampa, FL, United States Introduction: In bottom-up proteomic studies, proteins are proteolytically digested into peptides that are subsequently identified using LC-MS/MS. Ideally, the protease used should be both highly efficient and specific, since poor digestion due to low specificity and/or efficiency complicates LC-MS/MS analysis by increasing sample complexity and reducing sensitivity. Here we describe the application of an ultra-efficient and ultra-specific protease that cleaves C-terminal to arginine residues and can produce near-zero missed-cleavage rates. This Arg-C Ultra protease can be used in combination with Lys-C to produce "tryptic" digests that are superior to those produce by trypsin or trypsin combined with Lys-C. Methods: Human K562 extracts were digesed with combinations of Arg-C Ultra, Lys-C and Trypsin at a variety of digestion times and enzyme:substrate ratios to analyze digestion efficiency and specificity. Digested peptides were analyzed on an Orbitrap Exploris 240 (ThermoFisher) and data were searched with Byonic (Protein Metrics). Ubiquitination analysis utilized Protac-treated cells from which total protein was extracted, digested with Arg-C Ultra or Trypsin, followed by C18 cleanup and subsequent enrichment of peptides with the PTMScan HS Ubiquitin remnant motif (K-Ɛ-GG) magnetic beads (Cell Signaling). Histones were extracted from adult- derived immortalized mouse microglia (IMG cells) using the EpiQuik Total Histone Extraction Kit (EpiGentek). Samples were reduced before 2hr digestion with Arg-C Ultra followed by alkylation and desalting. Histone data were collected using a Bruker timsTOF Pro instrument and analyzed with FragPipe (using LFQ-MBR and open search workflows). Results: Digestion of human K562 cell extract with Arg-C Ultra demonstrated superior cleavage efficiency and specificity over any protease tested to-date. The combination of Arg-C with Lys-C was shown to outperform Trypsin or Trypsin/Lys-C in terms of both digestion efficiency and specificity. Histone analysis was completed without prior protein and peptide derivatization (e.g., propionylation of lysine/N-termini). High sequence coverage of histones was obtained, which included the basic N-terminal tail regions of H2A, H2B, H3, and H4. Specifically, overall sequence coverage for the core histone variants was approximately 80% or greater (greatest at ~96% for H2B) with high digestion efficiency observed (<1% missed cleavages). Numerous PTMs were identified using either targeted or open search strategies including phosphorylation as well as lysine mono-, di-, tri-methylation, acetylation, formylation, ubiquitination (Gly-Gly tag), and butyrylation Conclusions: Advanced proteomic sample preparation enabled by an ultra-efficient arginine-specific protease. 563

564

P-II-0409 Magnetic bead-based SPEED enables rapid and high-throughput proteomics sample preparation: A case study in clinical microbiology for species identification and antibiotic resistance detection Joshua Kanus1, Madita Winkler1, Andy Schneider1, Peter Lasch1, Jörg Döllinger1 1Robert Koch Institut, ZBS 6, Berlin, Germany The throughput of proteomics is constantly increasing, which enables the development of new applications focusing on large-scale sample cohorts. These applications range from answering fundamental biological questions based on the analysis of gene knockout libraries to the analysis of population-scale plasma cohorts or the diagnostics of clinical bacterial isolates. Upscaling proteomics challenges all stages of the workflow including sample preparation, data acquisition and data analysis. Initially, comprehensive, accurate and deep proteome analysis relies on the reproducible generation of peptides from the entirety of proteins present in each sample. Most sample preparation protocols employ detergents or chaotropic agents for protein extraction and support sample lysis by physical disruption methods. As many extraction reagents inhibit enzymatic digestion and are incompatible with LC-MS/MS, the idea behind most sample preparation methods is to remove interfering substances before digestion. Recently, we introduced a detergent-free method, named Sample Preparation by Easy Extraction and Digestion (SPEED), which consists of three steps, acidification, neutralization and digestion. SPEED is a universal method for peptide generation from various sources and is easily applicable even for lysis-resistant sample types. As the scale of proteome studies increases, magnetic-based methods (PAC, SP3) are gaining attraction because of their automation potential. This motivated us to develop and evaluate an improved bead-based SPEED protocol to enable high-throughput sample preparation in 96-well plates with efficient and versatile one-step pure chemical-based sample lysis and direct reagent-free protein quantitation. The performance of magnetic bead-based SPEED was evaluated in an analysis of bacterial clinical isolates using our workflow for concomitant species identification and antibiotic resistance (AMR) detection. Samples were analyzed using a 100 SPD method (Evosep One) and diaPASEF data acquisition (timsTOF HT). The results show, that proteomics provided high specificity and sensitivity for AMR detection, while allowing species identification from very large sequence databases with high accuracy. This workflow has the ability for the successful implementation of proteomics in clinical microbiology. The magnetic bead-based sample preparation (SPEED) is able to prepare all ESKAPE bacteria for measurement in less than 2h without protocol adaptations. The method also has great potential for automation to prepare native samples for mass spectrometry without human intervention in the future. 564

565

P-II-0410 Automated on bead derivatization and digestion for the analysis of histone acetylation dynamics Tobias Kipura1, Madlen Hotze1, David Moritz1, Patrick Knjisa1, Anna-Sophia Egger1, Marcel Kwiatkowski1 1University of Innsbruck, Institute of Biochemistry and Center for Molecular Biosciences Innsbruck, Innsbruck, Austria Histones are highly basic proteins that interact with DNA to form nucleosomes and belong to the most highly modified proteins. Each modification not only changes the chemical composition of the histones, but can also alters the function of the histone species. The N-terminal region of histones is rich in lysine and arginine residues, of which lysines in particular are the targets for reversible acetylation and methylation, regulating DNA interaction, chromatin state and recruitment of transcription factors and associated proteins. Histone acetylation is therefore a hallmark of epigenetics and plays an important role in cell homeostasis and the regulation of various cellular processes, including metabolism and cell growth. Modulating histone deacetylases (HDACs) holds significant therapeutic potential for treating cancer, inflammatory, metabolic, cardiovascular, and neurological diseases. However, HDAC inhibition can cause multiple effects and toxicity due to the varied roles of histone acetylation. Therefore, selective HDAC inhibitors like benzamides are of great interest. For the characterization of the kinetic effects of HDAC inhibition on individual histone acetylation sites, we recently developed a mass spectrometric approach combining metabolic and chemical labeling (CoMetChem) generating chemically equivalent acetylated protein and histone species for quantitative comparison using in solution derivatization and digestion [1]. Employing multiple acetylation steps, each requiring manual pH adjustment per sample, to ensure quantitative conversion from lysine to acetyl lysine residues, in solution acetylation of proteins represents an elaborate and time-consuming protocol. To overcome the limitations of in solution acetylation, we present an automatized protocol for protein acetylation and digestion using a robotic liquid handling platform for on bead derivatization with acetic anhydride followed by tryptic digestion. In combination with metabolic labelling, our automatized sample preparation workflow was applied to decipher the kinetic specificity and metabolic dependency of selective and pan- HDAC inhibitors at individual histone acetylation sites. Acknowledgements This work was supported by the Tyrolian Science Fund (TWF; P18903 to M.K.), the Promotion Program for Young Scientists at University of Innsbruck (P316826 to M.K.) and the Early-Stage Funding at the University of Innsbruck (P415586 to T.K.). References [1] A. van Pijkeren et. al, Combined Metabolic and Chemical (CoMetChem) Labeling Using Stable Isotopes-a Strategy to Reveal Site-Specific Histone Acetylation and Deacetylation Rates by LC-MS, Analytical chemistry 93 (2021) 12872–12880. https://doi.org/10.1021/acs.anal-chem.1c01359. 565

566

P-II-0411 Enhancing plasma proteomics and comparing mass spectrometry with affinity-based technology Ann-Christine König1, Thomas Gronauer1, Andreas Schmidt2, Marcel Blindert1, Fabian Gruhn1, Zuzana Demianova3, Juliane Merl-Pham1, Stefanie Hauck1 1Helmholtz Center Munich, Metabolomics and Proteomics Core, Munich, Germany 2Bruker Daltonics GmbH & Co. KG, Bremen, Germany 3PreOmics GmbH, Munich, Germany Blood is a crucial reservoir of clinically relevant information, necessitating a comprehensive understanding of its protein composition. Exploring plasma proteins is highly competitive within the field of proteomics. However, it faces challenges due to a dynamic range spanning over 10 orders of magnitude. Our objective was to enhance mass spectrometry-based plasma protein identifications using ENRICH-iST (PreOmics) for sample preparation and a timsTOF HT instrument (Bruker) connected to an Ultimate 3000 nano-RSLC (Thermo Fisher Scientific). The acquired data were compared with plasma data from the CLINSPECT-M consortium, originating from various laboratories across Munich, and additionally across platforms using OLINK® Explore 3072 data. We compared four sample preparation methods: employing ENRICH-iST kit (PreOmics), in solution digest, iST- BCT kit (PreOmics) and perchloric acid precipitation. By applying the ENRICH-iST sample preparation, we increased the protein identifications by about 50%. Gradient length analysis (30min, 70min, and 90min) revealed an approximate 30% increase between 30min and the two longer gradients, with negligible differences between 70min and 90min gradient length. The use of a tailored py_diAID method (Skowronek et al., 2022), using varying window sizes depending on the amount of eluting precursor ions, yielded an additional 10% more protein identification compared to the standard DIA method provided by Bruker. After selecting the most effective combination, we used plasma ring trail raw data from the CLINSPECT-M consortium to assess the scope of our improved plasma proteomics. We reanalyzed DIA raw files from six different setups, each employing its best practices in plasma processing. DIA_NN was used for data analysis, utilizing a predicted human spectral library to enable comparative evaluations through mpwR (R package for mass spectrometry data, Kardell et al., 2023). To compare data across platforms, we utilized identical plasma pool samples for measurements on both the OLINK platform and our own. The results revealed distinctiveness for each platform, with approximately 2000 unique protein IDs identified, respectively. The overlap of mass spectrometry and OLINK data was comparably small, with around 850 protein IDs. In conclusion, our performance in plasma proteomics surpasses the average standards in the field of mass spectrometry. The cross-platform comparison with OLINK indicates an additive effect on protein identifications, potentially enhancing the comprehensiveness of the overall perspective on plasma proteomics. 566

567

P-II-0412 Efficiencies gained in targeted serum proteomics using NRicher™ Beads – simplified and diversified workflows for sub-proteome and biomarker enrichment Matt Kuruc1, Swapan Roy1 1Biotech Support Group LLC, Monmouth Junction, NJ, United States The need for new biomarkers to support personalized healthcare, has fostered numerous proteomic innovations. Still, a number of challenges remain. One is the preponderance of high abundance proteins and, concurrently in targeted proteomic workflows, efficiency and consistency in quantifying target biomarker peptides from different sample cohorts. This is in part due to the changing landscape of proteins/peptides not associated with the selected biomarker targets. A solution for both these challenges is now available through a suite of products called NRicher™. This bead-based technology is derived from experience of over 10 years at the forefront of manufacturing beads (i.e., ionic, hydrophobic, hydrogen bonding, aromatic, polymeric) with differential proteome binding properties. NRicher™ now offers a solution that stands out in its simplicity and versatility. The NRicher™ workflow follows a bead-based bind/wash protocol using a standard microfuge to separate the buffer solutions from the beads. The NRicher™-derived sub-proteomes (different for each product application) are bound to the beads, followed by Bead-Assisted Sample Prep (BASP™), performing reduction, alkylation and digestion directly on the bead-bound proteome. The use of detergents or denaturants is not required. Once complete, the derived peptides seamlessly integrate to the LC-MS analysis. NRicher™ products were evaluated by LC-MS/MS for specific purposes, including: enrichment of low abundance human serum proteins for improving overall coverage, and for targeted proteomics – decreasing LC gradient time to improve productivity. Comparisons to pooled neat (without enrichment) serum are evaluated. Using NRicher™, enrichment of low abundance protein targets (<1 µg/ml) reach MS signal levels typically observed for mid-abundance proteins (1-100 µg/ml) and with less background noise, effectively suited for quantitation of target peptide signals. Consequently, NRicher™ sub-proteome enrichment can minimize acquisition time, collectively improving overall throughput, cost, and productivity. For these reasons, sample prep workflows combining NRicher™ sub-proteome enrichment with on-bead digestion will become especially desirable. We now introduce the NRicher™ Knowledgebase with a purpose so that researchers can enrich for particular protein biomarker(s) of interest. The Knowledgebase contains over 2000 proteins with corresponding relative signal intensities that can help select for the best NRicher™ bead/method combination for the biomarker(s) of interest; all freely accessible on a non-confidential basis. The Knowledgebase identifies and annotates > 100 prospective biofluid biomarkers associated with disease conditions, with references. Also, annotation of >200 soluble membrane proteins, derived from ectodomain shedding, a common dysregulated disease mechanism, and an important source of precision biomarkers are curated. Finally, we describe strategies to use the Knowledgebase to compartmentalize and profile systemic chronic inflammation, and associated diseases. 567

568

P-II-0413 NRicher™ Bead Platform Provides Unique Sub-Proteome Biases Creating Fit For Purpose Opportunities for Biomarker Discovery and Targeted LC-MS Quantification Matt Kuruc1, Swapan Roy1, Hugo Gagnon2, Jean-Philippe Couture2, Jean-Francois Noel2, Victoria Miller2 1Biotech Support Group LLC, Monmouth Junction, NJ, United States 2Allumiqs, Halifax, Canada For over 10 years BSG has been developing synthetic beads with differential proteome binding properties. Derivative products are well established with numerous citations in leading journals. Following the BSG lead to circumvent immuno-affinity, corona-based separations adapted to magnetic nano-particles has been introduced. However, this necessitates a specialized instrument-based workflow, and the need to perform LC-MS analysis on multiple bead-derived sub-proteomes. Nevertheless, singular BSG products support low abundance serum/plasma, enrichment, and different levels of Albumin removal. These products offer solutions that stands in terms of simplicity and versatility. To further their utility, LC-MS data characterization of the binding biases towards certain families or structurally related features of the sub-proteomes enriched is advantageous. Here we compared the Platelet rich plasma proteome using NRicher™, AlbuVoid™ and AlbuVoid™ PLUS by DIA LC- MS/MS. We characterized and evaluated the derived proteomes, vs. neat (not enriched) standard workflows. The bead-based workflow follows a bind/wash protocol using a standard microfuge to separate the buffer solutions from the beads. The derived sub-proteomes (different for each product) are bound to the beads, and then eluted. Proteins were reduced, alkylated, and digested with Trypsin/LysC and precipitated. Samples were acidified with 2% formic acid and purified by SPE. Peptides were resuspended in H2O, 0.1% FA, and analyzed on a TripleTof (Sciex), using a 45 min. gradient. AlbuVoid™ products provide the most proteome coverage over the 3 enriched) sample. AlbuVoid™ and AlbuVoid™ PLUS reduced Albumin specific signal to around 85%. AlbuVoid™ PLUS also reduces Immunoglobulin signal. Importantly, regardless of the different efficiencies of Albumin removal, all threeproducts generate a signal enrichment of around 1.6X of the low abundance sub-proteome; those proteins below the top 100. Notably, the one key characteristic from the NRicher™ data is a very reproducible enrichment of specific sets of sub-proteomes, with low %CV across technical replicates. Using clustering analysis, network analysis and Gene ontology we have extracted those specific sub-proteomes to determine their communalities. In particular, enzymes, proteases and membrane protein families are especially enriched using NRicher™ making it an interesting tool for biomarker discovery targeted towards those specific families. Also, for NRicher™, we noted an increase coverage of Complement proteins, especially C4A, which is usually hard to detect and quantify in plasma biofluids. Overall, from this analysis and our experience with other common workflows, these products provide a cost-effective alternative, comparable or better than commonly used depletion methods. Individual product biases, thus create fit for purpose workflows with applications in both discovery, and targeted LC-MS quantification, products, with identifications total neat (not of 1600 vs 900 for the 568

569

P-II-0414 Automated workflow for deep proteomic profiling of cervical fluid Guilherme Lanfredi1, Irene de la Calle2, Silvia Cabrera3, Antonio Gil-Moreno2,3, Vicente Bebia3, Francesc Serra- Marín2, Zehan Hu4, Torsten Pirch4, Vincenzo Romaniello4, Nils Kulak4, Eva Colas2, Gunnar Dittmar1 1Luxembourg Institute of Health, Department of Infection and Immunity, Strassen, Luxembourg 2Vall d’Hebron Research Institute, Barcelona, Spain 3Vall d’Hebron University Hospital, Barcelona, Spain 4PreOmics GmbH, Martinsried, Germany Introduction: Endometrial cancer is the most common gynecological cancer in developed countries, with increasing incidence. Its diagnosis relies on the invasive sampling of endometrial tissue after abnormal vaginal bleeding (AVB). However, only 5 to 10% of sampling leads to a diagnosis of endometrial cancer, rendering the majority of procedures unnecessary. An alternative diagnostic test for endometrial cancer could be based on the analysis of cervical fluid, which is non-invasive. To overcome this challenge, we present a deep proteomic profiling of cervical fluid, combined with automated sample preparation, using Automated ProteinPrep 96 (APP96TM) and LC-MS-based measurement pipeline, enabling high-throughput analysis for large-cohort studies. Methods: A cohort comprising 130 cervical fluid samples collected from patients with abnormal vaginal bleeding or cervical pathology. The samples were categorized into three groups: patients presenting AVB diagnosed with endometrial cancer confirmed by histopathological examination, patients with AVB diagnosis of benign conditions, and patients with other cervical pathologies. For each sample, an equivalent of 10 µg of protein was automatically processed using the APP96 platform, following the PreOmics iST protocol, including lysis, digestion, and cleanup steps. Chromatography was performed using a 44-minute gradient, separating the equivalent of 200 ng peptides on the Evosep One system with a 15 cm column and a 30 samples per day method. Mass spectrometry data were acquired in dia-PASEF mode on a timsTOF Pro mass spectrometer and processed using DIA-NN library-free search. Additionally, a subset of samples was selected to compare reproducibility and performance between the APP96 and our standard sample preparation workflow. Results: The precision tests showed that the APP96 platform resulted in high reproducibility and precise chromatography between replicates, with an average peak length of 10.2 seconds. This approach enabled the identification of 5,600 protein groups across all samples, with an average of 3,000 protein groups identified per sample. Notably, some proteins were differentially expressed, suggesting potential targets for further analysis to identify pathological states. Conclusions: Precise sample preparation with the APP96 automation system resulted in high-quality chromatographical profiles, which translates to a high number of identifications in the analysis of cervical fluid. The high-throughput setup makes the system particularly suitable for analyzing large cohorts. 569

570

P-II-0416 A robust and streamlined enrichment for delineating comprehensive small protein atlas in iPS cells derived from Parkinson's diseases Bo-Yu Chen1, Pei-Shan Wu2, Miao-Hsia Lin1 1National Taiwan University, Department of Microbiology, Taipei, Taiwan 2Chang Gung University, Chemistry Division, Center for Education, Taoyuan, Taiwan Small proteins, composed of fewer than 100 amino acids, have been historically neglected due to challenges in their annotation and biochemical detection. Despite their small size, they are crucial for essential cellular processes such as gene expression regulation, signal transduction, enzymatic activity, structural integrity, stress response, and development. In current proteomic analysis, the comprehensive identification of small proteins is hampered by signal suppression from larger proteins, their inherent low abundance, incomplete genome annotation, and inappropriate database search criteria. While ribosomal profiling and bioinformatic adjustments can overcome some impediments in genome annotation and database searches, bottlenecks in the front end of mass spectrometry (MS)-based proteomic analysis, such as sample preparation and getting rid of suppression from larger proteins in LC-MS/MS analysis, remain unsolved. To address these challenges, we developed a homemade StageTip for small protein purification before LC-MS/MS analysis, resulting in a 1.5-fold increase in small protein coverage and a 2.5-fold increase in unique peptides compared to the total proteome. In line with in silico predictions, using Lys-C instead of trypsin expanded the small protein dataset from 239 to 270 in E. coli, the largest resource so far. Further investigating the role of small proteins in the pathogenesis of Parkinson"s" diseases, a total of 26 small proteins were significantly changed between LRRK2 p.G2019S iPSCs and mutation- corrected (LRRK2 p.G2019G) iPSCs. Small proteins related to mitochondrial functions and presynaptic function were upregulated in LRRK2 p.G2019S, while stress response and protein translation pathways were enriched in corrected iPSCs. Interestingly, 6 out of these 26 small proteins were mitochondrial proteins and interacted, suggesting involvement in Complex III formation. Notably, most small proteins were annotated with membrane- bound organelle localization, consistent with their roles in regulating cellular functions through interactions with larger protein complexes. These findings underscore the feasibility of our novel approach in exploring the critical roles of small proteins in cellular physiology and disease mechanisms. 570

571

P-II-0417 Ever simpler: streamlining proteomics with S-Trap turbo and BCA-no-more Stefen Loroch1, Alexandre Zougman2, Sandra Wilson1, John Wilson1 1ProtiFi LLC, Fairport, NY, United States 2University of Leeds, Leeds, United Kingdom Introduction: Proteomics advancements necessitate streamlined workflows. Traditional protein assays, like BCA, hinder throughput with their time consumption and susceptibility to interference, waste sample and are not suited to low-level input. Concurrently, sample dry-down is a variable and laborious step. We propose two transformative solutions: the S-Trap Turbo, which minimizes elution volumes for rapid LC-MS or MALDI analysis, and BCA-no-more, a method for direct protein quantification on S-Trap plates via intrinsic fluorescence, bypassing additional assays. Methods: The S-Trap Turbo employs new polymeric traps with high-density surface derivatizations for sample processing in low volumes: elute and shoot. BCA-no-more utilizes the protein-binding capabilities of S-Trap plates to quantify proteins directly on plate, removing BCA assays. We evaluate the technologies on diverse samples from serum to cell lysates to brain acetone powder, explore equivalence of performance to standard S- Trap plates and BCA assays, and explore the limit of detection, reproducibility, and dynamic range across a protein range of 1-100 µg. Preliminary Data: BCA-no-more significantly outperforms traditional BCA assays in speed while improving sensitivity within a reasonable dynamic range. It successfully integrates into automated workflows and demonstrated better reproducibility. The S-Trap Turbo performed robustly across a range of protein concentrations (1-100 µg), with elution volumes as low as 5 µL. Comparative analyses with traditional S-Traps on various sample types including serum, cell lysates and brain acetone powder showed no loss in digestion completeness or peptide yield, identified peptides and proteins, and improved reproducibility. Importantly, the time to analysis-ready samples was substantially reduced by removing the need for speedvacing. S-Trap Turbos and standard S-Traps showed no difference in their ability to handle hydrophobic peptides. The adaptability of BCA-no-more and S-Trap Turbo to immediate autosampler loading, with or without prior acidification, was confirmed, though acidification led to peptide precipitation in some samples. These innovations represent a shift in proteomic sample preparation to maximum efficiency by providing a streamlined, user-friendly approach that does not compromise on quality or thoroughness of analysis. Novel Aspect: Protein quantification and sample preparation techniques that reduce time, steps and sample loss in proteomics workflows. 571

572

P-II-0418 New tools for biologics analyses and QC: the shredder and S-Trap turbo MAM Stefen Loroch1, Alexandre Zougman2, Sandra Wilson1, John Wilson1 1ProtiFi LLC, Fairport, NY, United States 2University of Leeds, Leeds, United Kingdom Introduction: Quality control (QC) is paramount in proteomics and biologics manufacturing and traditional methods can be limited by specificity and fragmentation behaviors; comprehensive sequence coverage is often lacking, especially for novel species or post-translational modifications (PTMs). Additionally, Multiple Attribute Monitoring (MAM) applied to different manufacturing stages and products requires robust tools that can handle different contaminants such as surfactants and stabilizers. To address these challenges, we present two innovative solutions: the Shredder, for in-depth, rapid proteomic sequencing with full sequence coverage, and a standardized, automatable MAM kit, based on the S-Trap Turbo 96-well plate, for streamlined biologics QC. Methods: The Shredder applies a pseudo-enzymatic exceedingly low-specificity cleavage at all points along a peptide backbone to generate a series of highly overlapping peptides for complete sequence coverage. The MAM kit, based on a purpose-built version of the S-Trap Turbo, is tested under diverse conditions including surfactant-rich environments. The Shredder and MAM kit are applied to model proteins such as the NIST mAb RM8671 in the presence of contaminants like surfactants and polymers. Preliminary Data: The novel Shredder yielded unparalleled sequence coverage, producing an average of >10-100 peptides per amino acid position in a mere 30 minutes. This depth of analysis is akin to 'depth' in genomic sequencing and offers a granular view of the proteome, including elusive PTMs. We show the Shredder's efficiency compared to multiple enzymatic digestions on both model proteins and those of pharmaceutical interest, and subsequently propose possible applications in both academic and industry settings where verified PTM site localization and/or full sequence coverage may be essential. The S-Trap Turbo MAM kit was validated using the NISTmAb RM 8671 monoclonal antibody across challenging matrices, including viscous solutions and those with high surfactant content. It was robust in manual, semi- automated and fully automated workflows, and demonstrated high reproducibility including detection of PTMs and glycosylations, as well as residual host cell proteins (HCPs). This consistency was observed when comparing different automation platforms, with variations within the range of technical replicate injections. The standardized MAM kit worked successfully on samples with a myriad of contaminants and the Shredder provides unusually great sequence coverage. Together, they present a tool kit that address analytical needs at multiple stages of biopharmaceutical development. The technologies help ensure the detection of critical quality attributes to facilitating broad MAM adoption in therapeutic manufacturing from early research to final product release testing. Novel Aspect: Full-sequence peptide mapping and a universal MAM kit for standardized robust biotherapeutic QC. 572

573

P-II-0419 Comparison of SDS-PAGE to automated parallel capillary electrophoresis for protein size and purity assessments Kyle Luttgeharm1, Ralph Beneke1 1Agilent Technologies, Ankeny, IA, United States Traditionally, purified proteins have been analyzed using SDS-PAGE to confirm the presence of the correct manufactured products and identify any impurities present in the sample. While this technique is widely used it is also manual, time consuming, and requires conversion into a digital format for accurate and thorough analysis. Staining procedures for traditional SDS-PAGE, such as Coomassie blue, further hamper purity assessments due to their limited quantitative linear dynamic range. With the digitization of laboratory records becoming more common and increasingly stringent quality standards, alternatives to SDS-PAGE are needed. To address these problems with SDS-PAGE, Agilent has developed a 12-channel, parallel Capillary Electrophoresis-SDS (CE-SDS) instrument and reagents. Samples are prepared directly in a 96-well plate and covalently labeled with a fluorescent dye allowing for a 3-log dynamic range. The system automatically cleans and conditions the capillary environment prior to filling with fresh sieving gel to prevent protein carryover and buildup. Internal standards allow precise sizing through 240 kDa and quantification/purity results for a variety of reduced and non-reduced proteins. This automatic cleaning procedure allows for the analysis of samples ranging from crude cell lysates to purified proteins. The data can be individually analyzed as a high-resolution electropherogram and as a digital gel image for quick comparisons of multiple samples. This poster provides sizing and purity comparisons between the new Agilent ProteoAnalyzer system and SDS-PAGE using commercially available DNA Polymerases, Restriction Enzymes, and other purified proteins. The system demonstrates equivalent sizing to traditional SDS- PAGE while also providing impurity detection down to 0.1% due to the increased quantitative dynamic range. 573

574

P-II-0420 Designing a fully integrated microfluidic proteomic sample preparation workflow for clinical applications Maximilian Maldacker1,2,3, Manuel Metzger4, Ines Derya Steenbuck5,6, Bettina Wehrle5, Peter Bronsert1,6,7,8, Martin Werner5,1,6,7,8, Niklas Klatt4, Oliver Schilling5,1,2,3,6,7,8,9 1University Hospital Freiburg, Institute for Surgical Pathology, Freiburg, Germany 2Albert-Ludwigs University of Freiburg, Faculty of Biology, Freiburg, Germany 3Albert-Ludwigs University of Freiburg, Oncogene-driven Immune Escape (OncoEscape), Freiburg, Germany 4Hahn-Schickard, Freiburg, Germany 5Institute for Surgical Pathology, Freiburg, Germany 6Center for Personalized Medicine, Freiburg, Germany 7Partner Site Freiburg, German Cancer Consortium (DKTK), Freiburg, Germany 8University Medical Center Freiburg, Medical Faculty, Freiburg, Germany 9Albert-Ludwigs University of Freiburg, RTG ProtPath, Freiburg, Germany Mass spectrometry (MS) is pivotal in tumor characterization from clinical, patient-derived paraffin-embedded and formalin-fixed tissues (FFPE). However, implementation of the proteomic analysis in medium throughput clinical laboratories necessitate a lot of effort and is - although being partially automated – not cost effective. Widely used automated workflows on robotic platforms still cannot provide an intervention-free sample preparation optimized for clinical routine. Furthermore, transferring sample preparation workflows into clinics affords standardization and increases the bureaucratic workload while being not interoperable between laboratories. Thus, we opt to design a fully integrated miniaturized laser-microdissection-compatible sample preparation workflow using microfluidics following a "set-it-and-forget-it" principle. We validate our approach by elucidating proteomic intratumoral heterogeneity in breast cancer across different intrinsic subtypes. The automated processing is coupled to high throughput liquid chromatography and mass spectrometry (LC-MS). The herein developed workflow is applied for dissection and analysis of central, adjacent and peripheral regions of breast cancer tumors from 40 patients. With our labile surfactant-based protocol for miniaturization, we are able to extract up to 3000 proteins from microdissected paraffin-embedded and formalin-fixed tissue slices while reducing the variation across samples. Furthermore, our laser-microdissection and microfluidics-based workflow enables the proteomic analysis of breast cancer tumors for which alternative dissection approaches are not applicable. By dissection of central, adjacent and peripheral, we aim to display the intratumoral diversity on the proteome level. Thereby we currently opt to implement mass spectrometry-based approaches that spatially deconvolute the mechanisms underlying therapeutic resistance and recurrence in breast cancer. In our highly collaborative developmental process, we are able to show that proteomics is well on the way to resolve heterogeneity from patient derived FFPE samples with increased depth, easy handling combined in a robust workflow. 574

575

P-II-0421 Protein aggregation capture assisted profiling of the thiol redox proteome Ana Martinez-Val1,2, Jorge Lumbreras1,2, Irene Rodríguez1,2, Estefanía Núñez1,2, Inmaculada Jorge1,2, Jesús Vázquez1,2 1CNIC – Centro Nacional de Investigaciones Cardiovasculares, Cardiovascular proteomics, Madrid, Spain 2Centro de Investigación Biomédica en Red de Enfermedades Cardiovasculares (CIBERCV), Madrid, Spain Oxidative damage is central to multiple diseases, ranging from cardiovascular to neurological pathologies. In this context, thiol oxidation-reduction (redox) reactions are known to be tightly regulated and can act as sensors of oxidative stress, with implications for protein structure and function. Redox proteomics using mass spectrometry aims to evaluate thiol redox status globally in cells and tissues. Thiol redox status can be quantified using differential alkylation, which involves labeling cysteines before and after reduction with cysteine-reactive chemicals. Here, we present a novel strategy to perform differential alkylation, which is based on the protein aggregation capture (PAC) protocol[1]. In contrast to previous strategies, such as FASP-based (FASILOX)[2] or gel-based (GESILOX)[3] strategies, PAC-assisted redox profiling significantly reduces sample preparation time and costs without decreasing overall thiol or proteome coverage. We tested our novel PAC-assisted redox profiling protocol using Data Independent Acquisition on primary cell lines (HUVEC) and tissues (pig heart). For that, 1ug of sample was acquired in an Evosep (15SPD) coupled to a Thermo orbitrap Eclipse. Data was analysed using DIANN (v1.8.1) with a previously generated experimental library. We were able to quantify 3,000 proteins in pig heart and >20,000 modified peptides, including ~4,000 cysteine-containing peptides, 25% of which corresponded to reversibly-oxidized cysteines. Interestingly, we observed a change iodoacetamide and methyl- methanethiosulfonate that could hamper identification of the later when using library-free DIA in DIANN. in retention time between cysteines labelled with Finally, we benchmark PAC-assisted redox profiling against a previous study using FASILOX to evaluate oxidative damage in a pig myocardial infarction model[4]. Overall, we present a high-throughput and streamlined sample preparation PAC-based strategy for redox proteomics, that is compatible with isobaric labeling or can be combined with DIA analysis using label-free quantitative workflows. 1. Batth, T.S., et al., Protein Aggregation Capture on Microparticles Enables Multipurpose Proteomics Sample Preparation. Mol Cell Proteomics, 2019. 18(5): p. 1027-1035. 2. Bonzon-Kulichenko, E., et al., Improved integrative analysis of the thiol redox proteome using filter-aided sample preparation. J Proteomics, 2020. 214: p. 103624. 3. Martinez-Acedo, P., et al., A novel strategy for global analysis of the dynamic thiol redox proteome. Mol Cell Proteomics, 2012. 11(9): p. 800-13. 4. Binek, A., et al., Oxidative Post-translational Protein Modifications upon Ischemia/Reperfusion Injury. Antioxidants (Basel), 2024. 13(1). 575

576

P-II-0422 Contaminant removal from plant digests prior to LC-MS/MS: SP2 vs Ethyl acetate extraction Petra Martinková1,2, Hana Konečná1,2, Petr Gintar1,2, Karolína Kryštofová1,2, David Potěšil1,2, Martin Trtílek3, Zbynek Zdrahal1,2 1Masaryk University, National Centre for Biomolecular Research, Faculty of Science, Brno, Czech Republic 2Masaryk University, Mendel Centre for Plant Genomics and Proteomics, CEITEC, Brno, Czech Republic 3Photon Systems Instruments, Spol. s r.o, Drásov, Czech Republic The presence of contaminants in a peptide sample can have a detrimental impact on the LC-MS/MS analyses. Thus, the availability of efficient methods for the removal of contaminants is crucial. Here, we evaluated the efficacy of single-pot solid-phase-enhanced peptide sample preparation on carboxylate-modified paramagnetic beads (termed SP2) for the removal of sodium dodecyl sulphate (SDS) or polyethylene glycol (PEG) from Arabidopsis thaliana tryptic peptides. A 40-minute SP2 protocol was proposed and subjected to a comprehensive benchmarking against the ethyl acetate extraction protocol. Arabidopsis thaliana rosette leaves were harvested after four weeks of growth, and ground in liquid nitrogen. The plant tissue powder was solubilised in SDT buffer (4% SDS, 100 mM DTT; 100 mM Tris/HCl, pH 7.6). The proteins were digested with trypsin (an enzyme-to-protein ratio of 1:50) using the filter-aided sample preparation (FASP) method. The obtained peptides were subsequently contaminated with SDS and/or a mixture of PEGs (PEG 200 to PEG 8000) to various final concentrations. Aliquots of 250 ng and 10 µg of contaminated peptide samples were subjected to either ethyl acetate extraction or the SP2 protocol in order to remove the contamination. The peptide samples were then analysed by LC-MS/MS using the Ultimate 3000 RSLCnano system connected to the Thermo ScientificTM Exploris 480 mass spectrometer operated in data-independent acquisition mode (DIA). The acquired data were processed in DIA-NN software (version 1.8.0). The efficacy of the proposed SP2 and ethyl acetate extraction protocols was evaluated in the removal of SDS contamination from A. thaliana tryptic peptide samples. The ethyl acetate extraction protocol demonstrated the ability to effectively remove SDS contamination up to 1%. However, our optimized SP2 protocol demonstrated superior performance, efficiently removing SDS contamination up to 5% from 250 ng and 10 µg peptide inputs. The SP2 protocol was also found to be effective in the removal of PEG contamination up to 1%, and even for simultaneous 5% SDS and 1% PEG contamination without significant sample losses. Conversely, incorporating a small amount of PEG into the sample could potentially prevent losses throughout the clean-up process. The optimised SP2 protocol outperformed the ethyl acetate extraction in terms of the effectiveness of contamination removal and its versatility. The findings indicate that the SP2 method is capable of efficient performance with various bead and peptide ratios, suggesting that our protocol is versatile enough to be applied to a wide range of peptide inputs without requiring adjustments. The study was supported by European Regional Development Fund—Project "SINGING PLANT" (CZ.02.1.01/0.0/0.0/16_026/0008446), CIISB, Instruct-CZ Centre of Instruct-ERIC EU consortium, funded by MEYS CR infrastructure (LM2023042), and the e-INFRA CZ project (ID:90254), supported by MEYS CR. 576

577

P-II-0423 Enhanced characterization of a Lysine-conjugated ADC using the Arginine-specific protease GingisREX® Jantje Mittwollen1, Laurent Rieux2, Lucy Fernandes3, Andreas Nägeli1, Krisztina Radi3, Olivia Eliasson1, Marshall Bern3, St. John Skilton3 1Genovis AB, Kävelinge, Sweden 2Genovis Inc., Cambridge, MA, United States 3Protein Metrics, Cupertino, CA, United States Bottom-up mass spectrometry approaches in the biopharmaceutical industry typically rely on lysine-specific proteases, such as trypsin and/or Lys-C, for sample preparation. However, molecules which contain modified lysine residues, such as lysine-conjugated antibody-drug conjugates (ADCs) and glycated proteins, pose a significant challenge to these traditional digestion methods. Modification at lysine residues is detrimental to the digestion efficiency of lysine-specific proteases and can result in highly complex digestion patterns due to the heterogeneity of the modifications. Digestion with trypsin resulted in incomplete coverage of lysine-containing peptides during the analysis of trastuzumab emtansine, a model ADC. Also, since conjugation prevents digestion at lysine, the modified and unmodified lysines are digested at different rates, therefore, not generating comparable peptides required for easy label-free quantification of site occupancy. Here we present an alternative and improved bottom-up strategy for the analysis of molecules containing modified lysine residues. Utilizing an arginine-specific protease, GingisREX® , we were able to achieve full coverage of all lysine conjugated sites for easy quantification of site occupancy in trastuzumab emtansine. As GingisREX is highly specific for arginine residues, with no detectable activity towards lysine, the digestion efficiency is independent of potential lysine modifications and for trastuzumab emtansine, the same set of peptides are generated regardless of conjugation level. In addition to site occupancy determination, we were also able to characterize the changes in drug-antibody ratio (DAR) which may occur during forced degradation studies following incubation of the ADC in serum. Finally, intact reconstruction from peptide mapping data was used to correlate the bottom-up site occupancy measurements with DAR determinations performed at the intact level. 577

578

P-II-0424 Enhancing biomarker discovery through the analysis of plasma depleted samples by DIA and PRM strategies using the Orbitrap astral mass spectrometer Ana Montero-Calle1, Rodrigo Barderas1,2 1Instituto de Salud Carlos III, Functional Proteomics Unit, Madrid, Spain 2CIBERFES, Madrid, Spain Non-infectious chronic diseases are currently the most common cause of disability worldwide, and responsible for 74% of total deaths worldwide. Most chronic disease are underdiagnosed or diagnosed at late stages, primarily due to the late onset of their first clinical symptoms. Thus, the identification of biomarkers in minimally invasive biological fluids with diagnostic ability of these pathologies is mandatory to improve their diagnosis, treatment, and progression. In this context, proteomics analyses of plasma or serum play a crucial role for the identification of biomarkers that could be implemented in clinical routine. However, mass spectrometry analyses yet face challenges due to the wide dynamic range in protein concentration of these samples. Thus, the depletion of high-abundant proteins, such as albumin or immunoglobulins, is essential to reduce sample complexity and enhance the detection of low-abundant proteins. We aimed at identifying plasma biomarkers for colorectal cancer (CCR) and Alzheimer" disease (AD), two of the most common chronic diseases worldwide. Plasma samples were depleted using three different strategies to enrich them in low-abundant proteins that might be useful as biomarkers of these diseases: immunoaffinity chromatography, ENRICH-iST kit, and Mag-Net method. Subsequently, the depleted and 1 µL of neat plasma samples were trypsin digested and analyzed using data independent acquisition (DIA) in the Orbitrap Astral and Orbitrap Eclipse Tribrid mass spectrometers, with 15-minutes and 2-hours gradients, respectively. Nearly 4000 plasma proteins were identified in the plasma samples depleted by the non-conventional immunoaffinity methods with the Astral mass spectrometer. Additionally, the efficacy of plasma enrichment methods was assessed using parallel reaction monitoring (PRM) in the Orbitrap Astral and Orbitrap Eclipse Tribrid mass spectrometers, with 15-minutes and 2-hours gradients, respectively, to detect previously described CRC and AD plasma biomarkers. The superior performance of the combination of plasma depletion and the Orbitrap Astral mass spectrometer was further demonstrated for targeted proteomics analyses. The implementation of plasma protein depletion strategies is crucial for enhancing the sensitivity and depth of mass spectrometry analyses, enabling the identification of a broader range of proteins and potential biomarkers for various diseases. This approach benefits both discovery and targeted proteomic analyses, offering valuable insights into disease mechanisms and potential diagnostic markers. 578

579

P-II-0425 High-dimensional phosphoproteomics with multiplexed µPhos Denys Oliinyk1, Ilka Klose1, Felix Schneidmadel1, Florian Meier1 1University Klinikum Jena, Jena, Germany Modern mass spectrometry (MS)-based phosphoproteomics quantifies thousands of phosphorylation sites. However, most of them have no experimentally derived biological function or cognate kinase. A promising strategy to unravel phosphorylation networks in more detail is to study their dynamic response to multiple internal or external perturbations in a systematic manner. To facilitate such high-dimensional experimental designs, we recently introduced µPhos, a scalable and sensitive platform for phosphopeptide enrichment (Oliinyk et al. 2024). By minimizing processing volumes and transfer steps, µPhos achieves lossless processing of 96-well cell culture plates or other input-limited samples within one day. Combined with trapped ion mobility – mass spectrometry, we quantified >10,000 class I phosphosites with excellent reproducibility (median CV < 15%) and data completeness (>99%) from 10 µg protein starting amount, which equates the protein mass of ~40,000 human cancer cells cultured in a 96-well format. Diluting the input amount to 250 ng and 62.5 ng, we still identified about 2,000 and 700 phosphosites with a localization probability >0.75. Considering that many emerging proteomics applications demand such or even lower input amounts, we here explored strategies to further miniaturise our setup. Specifically, we tested non-isobaric labelling-based multiplexed data-independent acquisition to simultaneously increase throughput and sensitivity. Pilot experiments indicated that 96-well plates are not ideal for phosphopeptide enrichment from low (<1µg) protein starting inputs. We reasoned that this is because of the volume-to-area ratio in 96-well plates, which lowers the concentration of peptides and results in adsorptive losses at plastic surfaces. To solve this problem, we explored a combination of 384-well plate sample preparation and stable isotope labelled dimethyl-based multiplexing. Depending on the combination of labels and digestion enzyme, this should offer an up to five-fold increase in sensitivity and sample throughput. To make µPhos compatible with dimethyl labelling, we optimized cell lysis and alkylation/reduction buffers to achieve a high labelling efficiency (>95%) of tryptic peptides. To evaluate the quantitative accuracy with multiplexed diaPASEF, we performed benchmarking experiments of samples mixed in known ratios and also test the influence of gradient length and MS instrumentation for different input amounts in the sub-µg range. Collectively, these results demonstrate the feasibility of high-dimensional phosphoproteomics experiments, for example to screen drug response in a time- and dose-dependent manner. 579

580

P-II-0426 A complete and fast workflow for single cell proteomics analysis combining the Tecan UNO cell dispensing platform, Vanquish Neo LC and exploris 480 DIA MS Amanda Suárez-Fernández1, Judit Bestilleiro2, Gonzalo Sanchez-Duffhues2, Ignacio Ortea1,2 1ISPA, Proteomics Unit, Oviedo, Spain 2Centro de Investigación en Nanomateriales y Nanotecnología (CINN-CSIC), Instituto de Investigación Sanitaria del Principado de Asturias (ISPA), Oviedo, Spain Single-cell proteomics (SCP) has gained significant attention in recent years. This technique allows for the differentiation of proteomic profiles at the single-cell level, which is not possible with traditional bulk proteomics, where the information of each protein is averaged across all cells in the sample. This enables the exploration of diversity and state among subpopulations of cells with distinct expression profiles. The potential applications of SCP are highly diverse, including cancer research, where it can characterize the mechanisms involving different cell types; stem cell biology, to understand the processes of pluripotency and differentiation into various cell types; and precision medicine, to understand individual patient variability in response to treatments. Given the extremely low amount of protein contained in a single cell, the two critical points in SCP are sample preparation and LC-MS analysis. Only recently have mass spectrometers been developed with sufficient sensitivity to achieve acceptable cellular proteome coverage. Consequently, SCP is a very recent field, and most studies conducted to date focus on the development and optimization of methods for individual cell dispensing, processing (cell lysis and protein digestion), and LC-MS analysis. The use of automated platforms increases robustness and decreases variability in the results, in addition to promoting the standardization of protocols. In this study, we used a relatively affordable automated platform, the Tecan UNO (HP D100), for the dispensing of single cells and their subsequent processing in a single step (cell lysis and rapid enzymatic digestion). Human mesenchymal cell digests were prepared in 96-well plates, with a total processing time of less than 75 minutes (including enzymatic digestion with Trypsin and Lys-C). The plates were loaded directly, or after freezing, into the autosampler of a Vanquish Neo LC system (Thermo Scientific), avoiding sample transfers. Various commercial chromatographic columns were tested in standard setups as used for bulk proteomics, with a throughput of approximately 48 samples per day. MS analysis was performed in an Orbitrap Exploris 480 (Thermo Scientific), with a nanoFlex source and FAIMS Pro Duo ion mobility interface. Several DIA acquisition methods, optimized for low-amount protein samples and using one FAIMS compensation voltage, were tested, and the runs were processed with Spectronaut v17 in the directDIA (library-free) workflow. For functional analysis and assessment of the molecular pathways covered by the identified proteins, STRING v12.0 and Reactome v88 were used to obtain pathway and process coverage for the identified protein groups. The results obtained regarding the single cell dispensing success, identification and quantification of proteins with the different variables tested will be presented, along with the GO terms and molecular pathways covered by the optimized workflow. Fig. 1 (optional) 580

581

P-II-0427 Evaluation of a transfer-free proteomics approach for the analysis of few dispensed cells: a core facility perspective Bernd Roschitzki1, Sibylle Pfammatter1, Rafaela Truffer2, Ralph Schlapbach1, Paolo Nanni1 1ETH Zürich / University of Zürich, Functional Genomics Center Zurich (FGCZ), Zurich, Switzerland 2Tecan Group Ltd., Maennedorf, Switzerland The field of single-cell proteomics has witnessed significant advancements in recent years. However, despite these advances, single-cell proteomics remains a challenging area for laboratories with limited resources, including personnel, instrumentation, and investments. This is due to the inherent difficulties associated with achieving optimal sample preparation, throughput, and sensitivity, which are essential for obtaining a comprehensive proteome profile. A recent study by Ye et al. [1] introduced a novel workflow for direct cell dispensing in an Evotip (Evosep) followed by on-tip lysis, protein digestion, and clean-up using an Uno Single Cell DispenserTM. This study aims to investigate its application in a core facility setting for samples ranging between 1 and 25 cells. HeLa cells at near confluency were washed twice with DPBS and the cell suspension was filtered to remove debris. The cell viability and concentration were determined and diluted to a working concentration of 200 cells/μl. Twenty microliters of solution were transferred to a C1 Dispensehead Cassette (Tecan) and loaded in the Uno Single Cell Dispenser. The cells were then directly dispensed on a pre-equilibrated Evotip with an aqueous layer. Enzymatic digestion was performed on the Evotip and subsequently prepared for LC-MS analysis on an Evosep (EvoSep Biosystems) coupled either to an Orbitrap Astral (Thermo) or a timsTOF (Bruker). Both datasets will be analyzed using the directDIA approach in Spectronaut (Biognosys). Evotips Pure were equilibrated by preparing a buffering layer on the upper part of the tip. Cells were dispensed directly into the Evotip using the Uno Single Cell Dispenser and a single-cell success dispensing rate >95% was achieved. For processing 96 samples, cells were dispensed in under 5 min resulting in an overall sample preparation time of 5h, including 4h enzymatic digestion. Once the settings are optimized, the workflow is universally applicable to cell types within diameters of 9-25 μm. To benchmark the workflow, single cells were also dispensed with the Uno Single Cell Dispenser as well as FACS in 384-well plates, digested with the help of a Mosquito HV (SPT Labtech) and manually loaded into pre-conditioned Evotips. Data were acquired on the Orbitrap Astral mass spectrometer and will be acquired on a timsTOF in DiaPASEF mode both coupled to an Evosep One LC device. Preliminary results demonstrate that a higher level of reproducibility is achieved when dispensing single cells in an aqueous layer compared to an empty Evotip. However, the on-Evotip dispensing approach led to lower identifications than the in-plate methods. The preliminary results show that similar data are obtained from cells processed in-plate after dispensing with the Uno Single Cell Dispenser or the FACS sorter demonstrating that the Uno Single Cell Dispenser is well-suited for implementation in a core facility. [1] Ye, Z., et al., Nat Commun 15, 2474 (2024) 581

582

P-II-0428 Evaluation and optimization of quantitative mass spec sample prep workflows Michael Rosenblatt1, Christopher Hosfield1, Virginia Kincaid1, Ellen Crummy1, Marjeta Urh1 1Promega, Research and Development, Madison, WI, United States Introduction: With the increase in proteomic sample preparation, instrumentation, and development of improved software tools, the number of biological studies utilizing Mass Spec data has increased in the last 5-10 years. Workflows including whole proteome screening after small molecule treatment (i.e. PROTACS, inhibitors, etc.), Immunoprecipitation, and PTM enrichment are commonly utilized. As these workflows are implemented, optimization of specific steps is needed. In this work we will focus on several workflows: PROTAC and kinase inhibitor treatment of cancer cells, enrichment of HiBiT tagged proteins, PTM enrichment (ubiquitin and phosphorylation), enrichment of exosomes, and automation. Improvements to specific steps in these workflows was examined with a focus on improvements to lysis bufferm proteolysis, desalting (i.e. detergent and surfactant removal), and steps to reduce time and increase overall workflow efficiency. Methods: Human cancer cells were lysed with various lysis buffers desalted (SPE, precipitation, non-specific capture on magnetic beads, etc) and subjected to proteolysis. PTM"s or epitope tags were analyzed using magnetic beads specific for the appropriate target. Samples were analysed by nanoflow LC-MS/MS analysis using a capillary C18 column (0.075 x 250 mm; Acclaim PEPmap) coupled to an Exploris 240 Mass Spectrometer (Thermo) using both DDA and DIA methods. Data were processed using Byonic (Protein Metrics), PD 3.1 (Thermo) or Maxquant (Max Planck Institute). Results: In our efforts to optimize our ability to quantify thousands of proteins in a single Mass Spec experiment, several factors were considered. These include selection of optimal lysis buffer, optimization of proteolysis, and desalting methods. In our evaluation , we have found that surfactant based lysis buffers produce more proteins and peptides relative to denaturant and solvent based lysis. Implementation of a sequential trypsin, Lys-c, and Arg-c Ultra protocol, in small ratios (1:10 versus 1:50) results in improved proteolytic efficiency at both lysine and arginine. With regard to desalting, we find that the use of magnetic beads, regardless of bead surface, performs as well or better than precipitation or filter-based methods, but with increased efficiency and without any technical challenges. This observation is particularly applicable for small samples (i.e. less than 5000 cells) where precipitation and filtration based methods are particularly challenging. These optimization steps were applied to the multiple workflows described above, resulting in improved analytical results (i.e. more quantifiable proteins) produced in shorter time periods. In many cases, because of the properties of magnetic beads, we can automate many of the workflows under study. 582

583

P-II-0429 Targeted image-guided 3-D tissue sampling with nanoliter resolution for bottom-up and top-down proteomics Jan Hahn1, Manuela Moritz1, Bente Siebels1, Paula Nissen1, Antonia Gocke1, Ali Biabani1, Hartmut Schlüter1 1UKE, Section Mass Spectrometry and Proteomics, Hamburg, Germany Proteomics of tissues using mass spectrometry-based bottom-up or top-down approaches requires sampling the area of interest, followed by homogenization. Typically, tissues consist of various cell types, each with distinct proteomes. Consequently, distinguishing areas composed of different cell types is essential to avoid mixed proteomes during sampling. In the second step, homogenizing the selected tissue breaks the compartments of the cells, releasing their molecules, including enzymes. These enzymes can alter the original composition of proteoforms, resulting in false positives and negatives. We have developed a new instrument, the 3D-miTi-LAb (3-Dimensional micro-Tissue Laser Ablation system), which allows the selection of areas of interest using microscopes and optical coherence tomography (OCT). Sampling a minimal volume of 0.5 nano-liters is achieved using a single shot of laser light with a wavelength of 2.9 µm, emitted by a nanosecond infrared laser (NIRL). During irradiation, water molecules in the tissue absorb the laser"s energy and rapidly move out, causing an explosion. This explosion converts the irradiated tissue area (80 µm diameter and 20 µm depth) into an aerosol containing all molecules previously within the tissue. Post-centrifugation, the absence of a visible pellet demonstrates effective homogenization. This is evidenced by the significantly higher yields of proteins—both in total amount and number of identified proteins—obtained through tissue sampling with NIRL compared to classical mechanical homogenization. Since NIRL-based tissue sampling is extremely fast, enzymatic degradation—typically occurring during mechanical homogenization—is minimized. Additionally, NIRL sampling is gentle, avoiding the fragmentation of labile proteins, as the laser energy is not transferred to proteins, and the exploding water molecules lose their energy via adiabatic cooling. In summary, the new 3D-miTi-LAb technology is ideal for bottom-up and top-down proteomics of tissues, offering 3-dimensional spatial resolution. It helps get closer to the original composition of proteoforms in intact tissue and minimizes mixed proteomes originating from different cell types. 583

584

P-II-0430 Deep middle-down proteomics using dissolvable BAC cross-linked Polyacrylamide gel electrophoresis Ayako Takemori1, Philipp T. Kaulich2, Andreas Tholey2, Nobuaki Takemori1 1Ehime University, Toon, Japan 2Christian-Albrechts-Universität zu Kiel, Kiel, Germany In the analysis of proteoforms by mass spectrometry (MS)-based proteomics, it is difficult to glean comprehensive information about proteoforms using the bottom-up approach because short peptide fragments, obtained by enzymatic digestion such as trypsin, are used for analysis. As an alternative, the top-down approach, in which proteoforms are introduced into a mass spectrometer in intact form allowing analysis of their intact chemical structures, has become the mainstream. However, most of the proteoforms that can be measured in current top-down proteomics are 50 kDa or smaller, and the analysis of higher molecular weight proteoforms has remained a challenge. As a middle-ground approach, middle-down proteomics, which analyzes middle-down peptides obtained by limited proteoform digestion, is attracting attention as a practical solution for obtaining proteoform information from a wide range of molecular weight regions. We have developed a multidimensional separation workflow, GeLC-FAIMS-MS, which combines high-resolution intact proteoform fractionation by SDS-PAGE with an LC-FAIMS-MS system, and have succeeded in achieving in- depth top-down proteomics from small sample amounts. GeLC-FAIMS-MS has also demonstrated its superior performance in Glu-C-based middle-down proteomics, dramatically increasing the number of detectable middle- down peptides and enabling broader sequence coverage. In this presentation, we report on 2D-GeLC-FAIMS-MS, a novel workflow that combines GeLC-FAIMS-MS with pre-fractionation of middle-down peptides using dissolvable BAC cross-linked polyacrylamide gel electrophoresis (BAC-PAGE) to achieve even deeper middle- down analysis. 584

585

P-II-0431 Influence of different sample preparation approaches on proteoform identification by top-down proteomics Philipp T. Kaulich1, Kyowon Jeong1,2, Oliver Kohlbacher2, Andreas Tholey1 1Christian-Albrechts-Universität zu Kiel, AG Proteomics & Bioanalytik, Kiel, Germany 2University of Tübingen, Institute for Bioinformatics and Medical Informatics, Tuebingen, Germany Top-down proteomics (TDP) analyzes intact proteoforms, i.e., all molecular forms in which proteins can be present. Despite significant advances that have been made in the past, the proteome-wide identification of proteoforms faces several challenges, e.g., limited sensitivity and complex spectra. A typical TDP sample preparation workflow for analyzing cellular proteoforms includes cell lysis, prefractionation of proteoforms smaller than approximately 30 kDa, optional reduction/alkylation, and multidimensional fractionation approaches. In the literature, a plethora of sample preparation strategies have been presented. Here, we systematically examined the influence of the various sample preparation steps on identifying proteoforms and proteins by TDP from human Caco-2 cells. Besides the number of identifications, their physicochemical properties and the occurrence of artificially introduced modifications were evaluated. Each step in sample preparation influenced the number, confidence, and physicochemical properties (mass, isoelectric point, GRAVY score) of the identifications. Lysis conditions using acidic pH tended to extract a higher number of proteoforms with an alkaline pI and vice versa. Moreover, the lysis affected the occurrence of artificially modified proteoforms (covalent modifications/adducts and artificial truncation). Reduction and alkylation increased the number and residue cleavage of proteoform identifications, with the effect on cysteine- containing proteoforms being the most significant. However, disulfide reduction resulted in the identification of artificially introduced proteoforms due to hydrolysis of peptide bonds C-terminal to aspartate residues, likely due to elevated temperature during the reduction. Moreover, information about disulfide bridges is lost. Prefractionation of proteoforms smaller than 30-50 kDa could significantly increase the number of identifications compared to full lysate analyses. However, different biases for subgroups of proteoforms (e.g., small, acidic, and hydrophilic) were observed depending on the sample preparation strategies. Specific steps in sample preparation led to the generation of artificial proteoforms, e.g., the use of formic acid and β-mercaptoethanol resulted in covalent adducts. Elevated temperatures can result in hydrolysis of peptide bonds C-terminal to aspartate residues. Multidimensional fractionation schemes significantly increased the number and confidence of proteoform identifications but on the cost of the measurement time. Furthermore, the longer sample handling time increases the risk of artificially introduced modifications and hydrolysis events. In summary, this study provides a comprehensive overview of the influence of different sample preparation steps in TDP on proteoform and protein identifications. For in-depth characterization of cells/organisms, performing multiple complementary sample preparation methods provides a straightforward approach to increase the number of identifications. Each variation in the various sample preparation steps has distinct advantages and limitations, and the specific research objectives should guide the selection of sample preparation steps based on the presented results. 585

586

P-II-0432 Improved protocol for preparation of urine native peptides Tomohiro Uchimoto1, Keiko Yamamoto1, Kengo Yanagita1, Amr Elguoshy1, Tadashi Yamamoto1,2 1Niigata University, Biofluid Biomarker Center (BBC), Niigata, Japan 2Shinrakuen Hospital, Clinical laboratory, Niigata, Japan Background: As an emerging field of omics, peptidomics is focused on analysis of endogenous native peptides (NP), which are produced from precursor proteins. The NP could be crucial for understanding physiological conditions or disease mechanisms. We developed an extraction protocol of NP from urine for LC-MS/MS based peptidomics. Since urine contains various substances, which are not so easy to remove, urine components eluted from C18 MonoSpin column still have something else other than NP. Methods: Urinary proteins were isolated under a Methanol / Chloroform precipitation method and NP was then purified from the supernatant (SUP) by passing through 30K MWCO filter to remove high molecular weight components. Then, NP was precipitated by adding x5 volume of cold EtOH at -20C for overnight. After 30 minute centrifugation, NP in a pellet was washed by adding 80% EtOH and centrifuged for other 20 minutes. The NP pellet was lightly dried and dissolved in 8M urea buffer, then was reduced and alkylated by using DTT and IAA. In our protocol, C18 MonoSpin column was finally used for purification of NP. Hereby, we improved the purification method of NP by stepwise elution from C18 MonoSpin column with acetonitrile. Results: Our results demonstrated the improved NP preparation protocol from urine provided stable and pure NP extraction than previous one. It makes sample loading easier for mass spectrometry analysis. NP measurement done by NanoDrop equipment provided a consistent scan shape as one shown by a standard peptide of BSA. Usage of ESI spry tip on mass spectrometry was enhanced due to removing of contaminants from the NP samples. Conclusion: This protocol of NP preparation from urine could purify NP and remove other contaminants efficiently. The purified NP of high quality reduced troubles of LC-MS/MS instruments and made a comprehensive and unbiased creation of the native peptide profiles in urine possible and facilitated reproducible and accurate quantitation of the endogenous native peptides. 586

587

P-II-0433 Automated sample preparation for in-depth plasma proteomics Fabian Wendt1, Claudia Martelli2, Zehan Hu3, Sandra Schaer4, Roland Bruderer4, Katrin Hartinger3, Nils Kulak3, Manuel Bauer2 1Tecan Group Ltd., Maennedorf, Switzerland 2Bruker Switzerland AG, Faellanden, Switzerland 3PreOmics GmbH, Martinsried, Germany 4Biognosys AG, Zurich, Switzerland The blood plasma proteome is a promising source of biomarkers, offering the advantage of minimal invasive collection and widespread clinical availability. However, sample-heterogeneity and wide dynamic range of protein abundances hinders comprehensive profiling within the necessary throughput for large cohort studies. To address this challenge, we present a fully automated sample preparation on the Fluent® applying the standardized ENRICH-iST technology that facilitates the identification and quantification of low abundant proteins from plasma. Combined with the next-generation dia-PASEF® MS acquisition, it is designed for high- throughput yet profound characterization of the plasma proteome of large cohorts. The automation platform consists of the Fluent® liquid-handling system (Tecan) with an integrated positive pressure and evaporator module, enabling automation from sample to dried peptides. The automated sample processing covers lysis, enrichment of low abundant proteins, tryptic digestion and clean-up including peptide drying by evaporation for LC-MS ready peptides employing ENRICH-iST technology (PreOmics). A total of 96 samples/run can be prepared for LC-MS analysis within one working day processing 20 µL of blood plasma/ sample. For mass spectrometry- based analysis 300 ng of peptides were separated on the nanoElute® 2 system and analyzed using the dia- PASEF® acquisition program on the timsTOF HT (Bruker). The data were searched with Spectronaut using directDIA+ (Biognosys AG). The high-throughput proteomic plasma preparation underwent a comprehensive evaluation, comprised of a comparative analysis between the automated and manual ENRICH-iST workflow, processing plasma samples of three healthy donors. Remarkably, identification rates and quantitative performance of the automated workflow were found to be equivalent to those obtained through the manual process. When comparing against neat plasma preparation, both manual and automated workflow demonstrated efficient dynamic range compression, yielding an approximate two-fold increase in identifications with application of ENRICH-iST over neat plasma preparation. To further assess the robustness of the automated pipeline, an inter-plate comparison spanning three days was conducted. The Inter-Day study revealed low variability in protein group identifications and their abundance across independent sample preparation runs, affirming the reliability of the automated plasma pipeline. The workflow flexibility, accommodating samples sizes ranging from 1 to 96, enables seamless processing of both small- and large-scale cohorts. This adaptability was exemplified by processing an age-dependent differences, yielding high-quality data to provide valuable biological insights across diverse experimental settings. This automated plasma proteome pipeline combines scalable throughput with profound proteomic characterization, bringing simplicity into the field of plasma proteomics. 587

588

P-II-0434 Unlocking the proteomic potential of FFPE tissues with BeatBox® and iST: a xylene-free, high-throughput workflow for in-depth proteome analysis Silvia Wuertenberger1, Jasmin Johansson1, Katharina Limm1, Katrin Hartinger1, Jonathan Krieger2, Nils Kulak1 1PreOmics GmbH, Planegg, Germany 2Bruker Ltd, Milton, Canada Introduction: Access to formalin-fixed, paraffin-embedded (FFPE) tissue is relatively easy, as it is evaluated, preserved, and stored in routine clinical diagnostics. FFPE tissue sample preparation for proteomic analysis, however, is extremely challenging and harsh conditions are required to remove cross-links and paraffin for efficient protein extraction. By combining the BeatBox® tissue homogenizer with an optimized iST sample preparation workflow, it became possible to process up to 96 samples of FFPE-curls per working day easily, efficiently, and reproducibly without using toxic xylene-based deparaffinization, opening up a new avenue for retrospective proteomic studies. Proteomic labs typically receive FFPE tissue in various formats and amounts. To address this issue and to further increase workflow flexibility, we now aimed to optimize and adapt the workflow to higher FFPE input amounts. Methods: FFPE samples (full curls, non-deparaffinized) from mouse tissues (cardiac muscle, kidney, and liver) and matching snap-frozen tissue samples were processed in plate or tube format using BeatBox homogenization coupled to iST sample preparation. For FFPE samples, an optimized workflow was established: FFPE curls were homogenized in the BeatBox (10 minutes), followed by an one-hour heat incubation, to de-crosslink, extract, reduce and alkylate proteins. Applying the iST sample preparation protocol, tryptic digestion was followed by an optimized peptide clean-up with an additional washing step to remove last traces of paraffin. To extend the workflow for larger FFPE input material, the workflow was adjusted accordingly. Peptides were analyzed on a nano-LC coupled to a timsTOF mass spectrometer. Raw files were analyzed using Bruker ProteoScapeTM. Results: This novel xylene-free BeatBox-iST workflow was applied to FFPE curls in plate or tube format and compared to a traditional workflow with xylene-based deparaffinization and bead-based sonication. For all tested tissue types, BeatBox homogenization outperformed bead-based sonication, revealing an increase in protein IDs of >10%, with a remarkable increase of 43% for mouse cardiac muscle tissue. Notably, the BeatBox- iST protocol achieved excellent intra- and interday repeatability with median CVs of <10% (n=4). Besides, it could be ensured that paraffin was completely removed from FFPE samples and no contamination was introduced into the LC-MS instrumentation. Comparing FFPE tissues with the corresponding fresh-frozen tissues revealed more than 10´000 proteins for both sample formats with up to 73% shared proteins and a similar dynamic range, confirming the superior performance of the FFPE BeatBox-iST workflow. Conclusion: This BeatBox-iST workflow simplifies large-scale retrospective proteomic studies by providing a xylene-free, robust and high-throughput solution for FFPE tissue samples and can now also be applied to larger FFPE input material. 588

589

New technology: Chromatography (P-II-0435 to P-II-0444) P-II-0435 Repurposed 3D printer allows economical and programmable fraction collection for proteomics of nanogram scale samples Eduardo Kitano1,2, Gareth Nisbet3, Yana Demyanenko1,2, Katarzyna M Kowalczyk1,2, Louisa Iselin4,5, Stephen Cross1, Alfredo Castello4, Shabaz Mohammed1,6,7 1The Rosalind Franklin Institute, Didcot, United Kingdom 2University of Oxford, Department of Pharmacology, Oxford, United Kingdom 3Diamond Light Source Ltd, Didcot, United Kingdom 4MRC-University of Glasgow Centre for Virus Research, Glasgow, United Kingdom 5University of Oxford, Nuffield Department of Medicine, Oxford, United Kingdom 6University of Oxford, Department of Biochemistry, Oxford, United Kingdom 7University of Oxford, Department of Chemistry, Oxford, United Kingdom Introduction: Two-dimensional liquid chromatography (2DLC) is a powerful tool for the analysis of complex proteomes. However, it suffers from sample losses during transfer between separation steps, compromising sensitivity and quantitative accuracy in LC-MS analysis, restricting the use of 2DLC to studies where abundant material is readily available. For studies involving samples with limited availability, achieving adequate analytical depth while minimizing sample losses presents a significant challenge. To address this issue, in this work, we built and applied a low loss fractionation system comprised of a reversibly reconfigured Evosep One LC system for the first dimension and a repurposed 3D-printer as a fraction collector in a range of Proteomics workflows covering both data dependent and data independent approaches as well as phosphoproteomics experiment. Methods: Peptides from human cell lysates were analysed by 2D RP-RP. An Evosep One LC system was used for the first-dimension separation in combination with a modified 3D printer as a fraction collector. Fractionated and non-fractionated samples were analysed by low-pH LC-MS/MS on an Ultimate 3000 RSLCnano system coupled to an Orbitrap Ascend Tribrid mass spectrometer (Thermo Fisher Scientific) using Data Dependent (DDA), Data Independent (DIA) or Wide Window Acquisition (WWA) modes. The resulting MS raw data were analysed by FragPipe, MaxQuant, DIA-NN or Proteome Discoverer. Data analysis was performed using the tools available in Perseus and GraphpadPrism. Results: The 2D RP-RP system used in this work demonstrated superior proteome coverage over single-shot experiments, achieving high sensitivity at peptide level for low amount of input material. More importantly, increase in protein level intensity was observed in 2D RP-RP, which indicates the complexity reduction afforded by the system outweighs the sample losses endured. The application of DIA and WWA to fractionated samples allowed further improvements in peptide and protein IDs in comparison to DDA, demonstrating that the use of data independent acquisition methods also benefit from the complexity reduction caused by fractionation. The utility of the 2D system was further investigated in phosphoproteomics workflow, showing significant improvements in phosphosites and phosphoproteins over single-shot analysis. Conclusions: Our 2D workflow showed significant increases in peak capacity and identification rates over single- shot experiments in the analysis of human cell lysates, exhibiting superior proteome coverage using limited sample amounts. Moreover, substantial improvements in coverage were also observed when the workflow was applied to the analysis of subproteomes. We envision that the 2D system showed here will have broad applicability across various proteomics workflows and could be particularly useful for isobaric multiplexed experiments applied to small cell populations. 589

591

P-II-0437 Versatile and bidirectional use of microfluidic LC columns, optimized configuration for achieving significant increases in proteome coverage at low nanoLC flow rates Jeff Op de Beeck1, Natalie Van Landuyt1, Robert van Ling1, Paul Jacobs1 1Thermo Fisher Scientific, Zwijnaarde, Belgium LC-MS-based proteomics relies on HRAM mass spectrometers for resolution, speed, and sensitivity, critical for accurate identification and quantification after LC separation. To match the increasing speed of MS instruments, nanoLC column development has focused on reducing dead volume and ensuring optimal elution, particularly at ultra-low flow rates. Pillar array technology, with its precise order and high permeability, demonstrates outstanding performance, although connectivity quality can significantly impact results. An optimal strategy is proposed to enhance connectivity and diminish post-column dead volumes, resulting in notable performance enhancements and enabling more thorough proteomic analysis. Moreover, its unique capabilities facilitate high- throughput, ultra-sensitive analysis by swiftly adapting to programmed flow rates, enabling 100 samples per day at elution rates as low as 100 nL/min. Utilizing a column heating device directly attached to the ESI source of the mass spectrometer enabled close integration of the nanoLC column, reducing the post-column volume by a factor of 3 to as little as 50 nL. Ensuring a low dead volume connection from the outlet fitting to a suitable nanoelectrospray emitter preserved the "on-chip" separation performance, resulting in significantly improved chromatography and deeper proteome coverage across all tested conditions. In our initial assessment, we explored the benefits for conventional bottom-up proteomics sample loads (50-500 ng of tryptic digest injected) using customized LC methods. These methods achieved sample throughput rates of 60 and 30 samples per day, respectively, with peptide elution at conventional nanoLC flow rates (250 nL/min), reaching instrument productivity levels of up to 85%. From triplicates at a throughput rate of 60 samples per day, we identified 5970 protein groups, with 94% quantified below a coefficient of variation of 20%. High- resolution LC separation and reproducible elution allowed for quantification of significantly more protein groups (7% below 20% CV and up to 30% below 10% CV) compared to conventional pulled-tip emitter nanoLC column types. Anticipating greater performance gains at lower flow rates, we conducted benchmarking with sample loads ranging from 50 pg to 10 ng. Methods were optimized for elution at 100 nL/min, achieving sample throughput rates of 96, 72, and 48 samples per day with instrument productivities of 70%, 77.5%, and 85%, respectively. Post-column volume reduction led to a performance increase of up to 25% (reducing average FWHM by 1.5s), increasing proteome coverage by up to 22% on the protein group level and identifying up to 2413 protein groups from as little as 250 pg. 591

592

P-II-0438 Improved proteome coverage and reproducibility in large cohort analyses using packed emitter columns Vincent Albrecht1, Tim Heymann1, Greta Briedyte2, Jarrod Sandow2, Johannes Müller-Reif1 1Max Planck Institute of Biochemistry, Martinsried, Germany 2IonOpticks, Fitzroy, Australia Proteomic analyses of large cohorts of human patient samples are used to understand the biological mechanisms of disease and to discover novel biomarkers for diagnosis and response to treatment. Deep proteome coverage and reproducibility of results is crucial for the success of these studies. We developed a novel packed emitter chromatography column, the IonOpticks Aurora Rapid 8x150 (8 cm length x 150 μm inner diameter, 1.7 μm C18), that improves the results from patient cohort analyses by ensuring high and consistent proteome coverage and reproducible quantitation. This study evaluates the efficiency of the column for high- throughput proteomic analysis of large patient plasma cohorts coupled with an Evosep One and a Thermo Fisher Orbitrap Astral mass spectrometer. As part of this study, patient plasma samples were analysed using a 100 samples per day method with approximately 700 samples run between quality control (QC) injections. To ensure consistent performance and reproducibility, the system underwent weekly QC injections of HeLa tryptic peptides using a 60 samples per day method. The system achieved robust protein identifications, with over 9,300 proteins and 100,000 unique peptides consistently identified per QC run across multiple columns. To assess the reproducibility of protein quantification across multiple QC runs and columns, protein intensities were used to calculate a coefficient of variation (CV) and Pearson correlation, revealing uniform performance with low CV values and high correlation between runs and between columns. This consistent performance, along with consistently narrow peak widths averaging 2.9 seconds FWHM, indicates good column performance and reproducibility, important for large patient cohort analyses. Furthermore, to ensure reliable data analysis, we evaluated retention time stability. This analysis demonstrated a high level of consistency, with a 2.52% median CV of retention times. Additionally, a four-month experiment on 16 selected peptides demonstrated stable retention times. Taken together, these results demonstrate that our novel packed emitter column combined with the Evosep One and Thermo Fisher Orbitrap Astral Mass Spectrometer enabled maximum protein and peptide identifications whilst also ensuring robust and reproducible protein quantification across a large patient plasma cohort analysis. 592

593

P-II-0439 High-throughput plasma proteomics with cap-flow LC separation and dia-PASEF-MS/MS detection Andreas Schmidt1, Claudia Martelli2, Leander Runtsch3, Raphael Schuster1, Markus Lubeck1, Bernd Müller1 1Bruker Daltonics GmbH & Co. KG, Application Development, Bremen, Germany 2Bruker Switzerland AG, R&D, Faellanden, Switzerland 3PreOmics GmbH, R&D, Martinsried, Germany Plasma represents a major source for clinical diagnosis and research which is obtained minimally invasive. Proteomics analysis of plasma is challenged by the high dynamic range of protein abundance as well as the variability between individual donations and donors. Major issues that hinder widespread application of mass spectrometry-based plasma proteome analysis is the low sample throughput of nanoLC-MS applications and lower sensitivity of HPLC separation at high flow rates. Capillary flow rate chromatography (1-10ul/min) provides a good option to combine high sensitivity and high sample throughput. The protein depth of cap-LC is similar to that observed with nanoLC applications using similar column dimensions and gradient length, however durations for sample loading and column equilibration are minimized due to the increased flow rate. Two sample preparations of plasma proteins were performed for 15 individual plasma donations, direct trypsination for a "neat" proteome composition and the ENRICH-iST kit for a deep plasma proteome. HPLC and MS settings were optimized at a flow rate of 2 µl/min to surpass 110 samples/day. MS acquisition parameters for dia-PASEF on the TimsTOF HT mass spectrometer were adapted to maintain reliable quantitation and to maximize the number of peptide and protein hits. For cap-LC separation, the separation gradient and the injection amount were programmed for highest sample throughput at highest level of data coverage as well as minimal sample carry-over and column loading and equilibration periods. A pooled sample was generated for LC method optimization. First, dia-PASEF parameters were optimized with nanoLC separation by probing the number of IM ramps, the position and extraction width of the windows within a 600 Th range for precursor isolation and limiting the ion mobility range to allow for optimal positioning of the windows or the precursor ion cloud. Using the optimized dia-PASEF cycle, injection amount and gradient length were tested for nanoLC and cap-LC flow rates. For nanoLC, we aimed at highest proteome coverage achievable to generate a reliable precursor library for subsequent fast gradient analysis. For cap-LC measurements, gradients for 75 and 140 samples/day were programmed and the reproducibility and throughput were confirmed. Results were searched in Spectronaut 19 against the spectral library generated by nanoLC analysis as well as in the directDIA+ node. For neat plasma, more than 400 protein groups with 140 SPD were quantified reliably. Enriched plasma samples covered more than 2000 protein groups in 7 min active gradient time. 593

594

P-II-0440 Long-term stability and reproducibility of nano-, capillary- and micro-flow LC-MS separations Christopher Pynn1, Runsheng Zheng1, Tabiwang Arrey2, Amirmansoor Hakimi3, Alec Valenta1, Oleksandr Boychenko1, Martin Samonig1 1Thermo Fisher Scientific, Germering, Germany 2Thermo Fisher Scientific, Bremen, Germany 3Thermo Fisher Scientific, San Jose, CA, United States Introduction: Nano-, capillary- and micro-flow LC hyphenation with MS detection is irreplaceable in a high- sensitivityanalysis of complex bio-samples. But concerns remain around long-term stability, day-to-day as well assystem-to-system results reproducibility. The transition of low-flow LC-MS applications from academic andR&D to translational research laboratories demands to prove the suitability of LC and MS technologies forthe analysis of large sample cohorts. In addition, the consistency of results across multiple laboratories andoperators is essential for pooling datasets for joint statistical analysis. Here we describe the output ofcomprehensive evaluation of proteomics data collected over several months of continuous operation atdifferent laboratories and discuss the impact of hardware and consumables on the reproducibility of results. Methods: Low-flow LC-MS and/or LC-UV data were collected in Germering (Germany), Bremen (Germany), and SanJose (USA) by different operators using identical LC and MS hardware and different batches of separationcolumns. The Vanquish Neo UHPLC system was used for peptides separation on 75 μm ID EASY- SprayPepMap Neo columns with 15, 50, and 75 cm length and 1 mm ID columns with 15 cm length. OrbitrapExploris 480 and Orbitrap Exploris 240 MS instruments were used to detect and fragment peptides anddata-dependent acquisition mode. Data were processed with and without resolving of chimeric spectra andreported with 1% FDR on peptide and protein level. Preliminary Data: Long-term robustness. The continuous injections of bovine serum albumin protein digest under typical fornanoLC-MS bottom-up proteomics experimental conditions were conducted for at least 6 months in twolaboratories. The longevity results of column backpressure, retention time variation, and full- width half-max(FWHM) for selected peptide peaks showed the suitability of low-flow separations LC for the uninterruptedlong-term analysis. Novel Aspect: A comprehensive evaluation of analytical variability in low-flow LC-MS analysis provides guidelines forachieving reproducible results across instruments and laboratories 594

595

P-II-0441 A tandem capillary and micro-flow LC workflow for high-throughput quantitative proteomics at near 100% mass spectrometer utilization Runsheng Zheng1, Martin Rendl1, Cristina Jacob2, Philip Remes2, Christopher Pynn1, Alec Valenta1, Ece Aydin1, Maksim Daniliuk3, Robert van Ling4, Scott Peterman2, Wim Decrop1, Neloni Wijeratne2, Martin Samonig1, Anne Morgenstern1 1Thermo Fisher Scientific, Germering, Germany 2Thermo Fisher Scientific, San Jose, CA, United States 3Thermo Fisher Scientific, Vilnius, Lithuania 4Thermo Fisher Scientific, Breda, Netherlands Introduction: Pushing the limits of low-flow LC for high-throughput quantitative workflows often comes at the cost of decreasing mass spectrometer utilization, data quality (e.g. proteome depth, column carryover and quantitative accuracy/precision), and method robustness. For fast methods, such trade-offs must be made to maximize acquisition time and minimize method overhead. Here, we present a novel tandem LC workflow for high-throughput quantitative proteomics that eliminates method overhead while improving on traditional column washing and equilibration cycles. Methods: A Vanquish Neo UHPLC system was configured for the tandem direct injection workflow coupled to either an Orbitrap Exploris 480, TSQ Altis Plus Triple Quadrupole or Stellar mass spectrometer for ultra-fast DDA, DIA and PRM analysis. Human plasma and cell lysates were utilized to demonstrate the performance in bottom- up proteomics at flow rates of 1.5 - 100 µL/min with 150 µm – 1 mm ID columns connecting to only one emitter, offering minimal analytical variation. Preliminary results: The tandem LC-MS workflow was tested using gradients down to 8 min (corresponding to 180 samples/day) to demonstrate throughput, run-to-run reproducibility, and close to 100% MS utilization for quantitative proteomics. Using the 150 µm x 15 cm column operated at 1.5 µL/min, we reproducibly identified >1,200 protein groups in DDA at 180 samples/day throughput with high quantification accuracy between columns. While moving to the PRM analysis, the tandem direct injection workflow along with the Stellar mass spectrometer directly supports the expansion of large-scale targeted experiments covering > 3500 peptides and >60 samples per day from 300 ng human plasma digest with negligible carryover permitting high-confidence peptide verification. Overall, this new workflow is a promising approach to improving instrument productivity offering high sensitivity, high sample throughput, and maximum quantification confidence. (For Research Use/Purposes Only – Not for Diagnostic Procedures) 595

596

P-II-0442 Achieving close-to-complete proteome coverage in single-shot deep dive DIA workflows Robert van Ling1, Amirmansoor Hakimi2, Santosh Renuse2, Paul Jakobs3, Jeff Op de Beeck3 1Thermo Fisher Scientific, Breda, Netherlands 2Thermo Fisher Scientific, San Jose, CA, United States 3Thermo Fisher Scientific, Ghent, Belgium Throughout the history of LC-MS based proteomics, comprehensive identification and quantification of all proteoforms present in biological samples has been the ultimate ambition. Several approaches have been close to achieving this goal, with the deepest coverage commonly obtained by fractionation of digested proteins using a separation mechanism orthogonal to low-pH reversed phase separation (SCX, high-pH C18) and subsequent LC-MS analysis of isobaric labeled peptide samples. As recent innovations in LC-MS instrumentation have leveraged significant increases in analysis speed, resolution and sensitivity, we investigate the possibility to achieve similar coverage using single-shot label-free analyses with ultra-high peak capacity LC-MS. Hereby significantly reducing the labor-intensiveness, overall cost and required analytical skill needed to achieve close to fulll proteome coverage. Human proteome samples were analyzed using an Orbitrap mass spectrometer coupled to a Vanquish Neo UHPLC instrument. For spectral library generation, protein digest samples were fractionated into 8 concatenated fractions at pH 5.5. For single-shot analyses, lyophilized and unfractionated cell digests were resuspended in 40 µl 0.1 % FA to obtain a stock solution of 500 ng/µl. Reversed phase LC separation was performed on a 110 cm µPAC Neo column with gradient lengths up to 180 min. Both DDA and DIA acquisition strategies were used to collect data. LC-MS data were searched using Spectronaut 17 or a beta version of Thermo Scientific Proteome Discoverer 3.1. In single-shot deep dive proteomic experiments, highest possible peak capacities are key to reduce sample complexity as much as possible prior to MS/MS analysis. To increase separation performance for long gradient analyses, a 110 cm long microfluidic channel uniformly filled with 2.5 µm pillars was used, significantly increasing peak capacities to well above 1600 in single-shot analyses. Median peak widths as low as 4.6 and 7.2 s have been obtained for respective gradient lengths of 90 and 180 min, which is 30% better than reported up to now. Evaluating the effect of increased separation performance on proteome depth for single species tryptic digests,following systematic optimization of MS acquisition settings, demonstrated that significant increases in depth were achieved by operating in DIA mode, with up 20% deeper coverage as compared to the best performing DDA method we tested. Using a library free approach, close to 10.000 protein groups were consistently identified with 180 min gradient times. We will further optimize LC-MS methods towards increased throughput and evaluate the impact of high-field asymmetric waveform ion mobility spectrometry (FAIMS) on the proteome coverage that can be achieved within single-shot LC-MS analyses. 596

597

P-II-0443 A novel selective comprehensive two-dimensional online nanoLC-CZE-MS platform for proteoform characterization Tobias Waldmann1, Alexander Stolz1, Christian Neusüß1 1Aalen University, Aalen, Germany CE-MS has been used to separate proteoforms on intact protein levels, however, selectivity and sensitivity are often not sufficient for complex biological samples. We therefore developed a two-dimensional (2D) heart-cut nanoLC-CZE-MS platform and have shown that this allows the pre-separation of intact proteins from a complex matrix and a 280-fold increased sensitivity compared to a one-dimensional CZE-MS approach. Furthermore, with this 2D approach, a higher proteoform selectivity can be gained [1]. While the transfer of a peak from the first to the second dimension is efficient using our two-dimensional heart-cut approach, the characterization of multiple chromatographic peaks or partly separated proteoforms is time-consuming, requires high sample amounts, and might lead to incomplete proteoform characterization. Therefore, we present here a novel nanoLC-CZE-MS platform to perform selective comprehensive nanoLC-CZE-MS. There, the RPLC column is connected to a storage capillary by a 10-port valve to decouple the storage capillary from the first dimension while performing the second dimension analysis. This allows the storage of a fraction from the first dimension for subsequent 2D analysis. The volume of the stored fraction depends on the volume of the storage capillary and can be easily adjusted. The storage capillary is also connected to an 8-port nanoliter valve with four 20 nL internal loops. Hence, 20 nL fractions can be transferred from the storage capillary to the CZE as the second dimension. Here, we show the essential parameters of our selective comprehensive two-dimensional nanoLC-CZE-MS platform. In addition, we show initial results from the analysis of a cell lysate using our selective comprehensive two- dimensional online nanoLC-CZE-MS platform, demonstrating its highly selective and sensitive analysis of proteoforms. References [1] A. Stolz, C. Neusüß, Characterisation of a new online nanoLC-CZE-MS platform and application for the glycosylation profiling of alpha-1-acid glycoprotein, Anal. Bioanal. Chem. 414 (2022) 1745–1757. 597

598

P-II-0444 Enhancing sample throughput for deep dive proteomics with Tandem LC Runsheng Zheng1, Martin Rendl1, Alec Valenta1, Tabiwang Arrey2, Yuan Lin3, Christopher Pynn1, Maksim Daniliuk4, Ece Aydin1, Robert van Ling5, Wim Decrop1, Felix Josef6, Till Reinhardt2, Nagarjuna Nagaraj6, Andreas Tebbe6, Nicolaie Eugen Damoc2, Martin Samonig1, Anne Morgenstern1 1Thermo Fisher Scientific, Germering, Germany 2Thermo Fisher Scientific, Bremen, Germany 3Thermo Fisher Scientific, Sunnyvale, CA, United States 4Thermo Fisher Scientific, Vilnius, Lithuania 5Thermo Fisher Scientific, Breda, Netherlands 6Evotec GmbH, Munich, Germany Introduction: Low-flow UHPLC coupled with mass spectrometry (MS) is the gold standard for deep quantitative analysis of complex proteomes. However, the unmatched sensitivity of low-flow LC-MS typically comes at the cost of low MS utilization and sample throughput. A high proportion of idle MS time results from long sample injection, loading, column washing, and equilibration cycles. Here, we present a novel, facile, and robust workflow for deep dive proteomics that eliminates idle MS time by employing a tandem LC. Methods: The novel LC-MS configuration, comprising a Vanquish Neo UHPLC system configured for tandem direct injection, was coupled to either an Orbitrap Exploris 240/480 or an Orbitrap Astral mass spectrometer for deep proteome profiling using library-free DIA. Human cell lysates were analyzed to demonstrate the performance of untargeted bottom-up TMT and label-free proteomics on 75 µm (≤ 300 nL/min) and 150 µm ID (1.5 µL/min) columns with throughput up to 180 samples/day. To minimize post-column dispersion the Sonation double-barrel column oven was used for all experiments. Preliminary results: The tandem LC-MS workflow was tested using gradients from 8 to 120 min (corresponding to 180 to 12 samples/day, respectively) to demonstrate high throughput, run-to-run reproducibility, and close to 100% MS utilization for discovery proteomics. For example, a 65-min gradient with a 75 µm x 75 cm (2 µm particles) column at 300 nL/min enabled the identification of >11,000 protein groups per run in DIA from 200 ng HeLa digest with > 91% MS utilization on the Orbitrap Astral MS. With more than 90% of protein groups overlapped among runs and <5% overhead time, this configuration delivers excellent reproducibility for large- cohort deep-dive proteomics analysis. (For Research Use/Purposes Only – Not for Diagnostic Procedures) 598

599

New technology: AI and bioinformatics in mass spectrometry (P-II-0445 to P-II-0504) P-II-0445 Extending PSM rescoring for timsTOF data to MSFragger with Oktoberfest Charlotte Adams1, Wassim Gabriel2, Mario Picciani2, Kris Laukens1, Mathias Wilhelm2, Kurt Boonen3, Wout Bittremieux1 1University of Antwerp, Computer Science, Antwerp, Belgium 2Technical University of Munich, Computational Mass Spectrometry, Freising, Germany 3University of Antwerp, Biomedical Sciences, Antwerp, Belgium Peptide-spectrum match (PSM) rescoring is a powerful enhancement of standard searching that boosts the spectrum annotation performance. Recently we fine-tuned the deep learning-based fragment ion intensity prediction model, Prosit, on a tryptic and non-tryptic ground-truth dataset acquired on timsTOF instruments, expanding the utility of Prosit to this data type. Implementing the new model into the PSM rescoring tool Oktoberfest resulted in an up to 3-fold improvement in the identification of immunopeptides and an improved identification rate of samples digested with AspN, GluC, and trypsin, demonstrating a broader biological and biomedical application. The high sensitivity of timsTOF instruments makes them ideal for studying samples containing low-abundance peptides. The parallel accumulation-serial fragmentation (PASEF) mode enables repeated re-targeting of low- abundance precursors. These individual scans of the same precursor are then summed to improve MS/MS spectrum quality. However, this relies on search engine-specific implementations of scan summing functionality. Until now, Oktoberfest for timsTOF data was limited to search engine outputs from MaxQuant, as the timsTOF Prosit 2023 model relies on MaxQuant for summing the MS/MS scans. Here, we present an extension of Oktoberfest to support MSFragger, a faster and well-established search engine, thereby significantly increasing its usability. Before implementing MSFragger into Oktoberfest, we will evaluate whether the timsTOF Prosit 2023 model can be used on spectra resulting from a different scan summing method. In addition, we will compare the results from MSFragger followed by Oktoberfest with those previously obtained with MaxQuant and Oktoberfest, as well as results from MSFragger and MSBooster. We anticipate that this improvement will especially benefit researchers investigating low-abundance peptides or working with large datasets. 599

600

P-II-0446 Tree-based quantification retrieves proteoforms from bottom-up proteomics data Constantin Ammar1, Marvin Thielert1, Eugenia Voytik1, Caroline Weiss1, Edwin Rodriguez1, Maximilian T. Strauss2, Florian Rosenberger1, Wen-Feng Zeng1, Matthias Mann1 1Max Planck Institute of Biochemistry, Proteomics and Signal Transduction, Planegg, Germany 2University of Copenhagen, NNF Center for Protein Research, Copenhagen, Denmark Recent algorithmic improvements in mass spectrometry (MS)-based proteomics have greatly increased the number of identifications. However, less attention has focused on proper quantification, which arguably provides the ultimate value of a proteomics experiment. We argue that much of the potential to improve quantification is still untapped. In particular: 1) Much of the data that could contribute to quantification is neglected and condensed into single "protein quantity" values. 2) Comparisons between linked ions (e.g. fragment ions vs. MS1 precursor ions) enable the extraction of rich information that is underutilized. 3) Information about the acquisition such as peak quality and retention time is currently not integrated into the quantitative feature selection. 4) Peptides of a protein potentially belong to different proteoforms and should be separately quantified but this is generally not the case. We present AlphaQuant, an end-to-end software pipeline, which offers novel solutions for the above issues. AlphaQuant introduces a "tree-based quantification" approach, where all data is organized into a hierarchical tree which systematically integrates information along the different levels of quantification from fragment ions, MS1 isotopes, charge states, modifications, peptides, proteins to proteoforms. Quantities at the fragment and precursor level are integrated with information about the experimental design and acquisition. Rigorous statistical analyses are performed on all available data, boosting statistical power. Clustering along the tree allows us to assess the consistency along its different levels. To optimally integrate this information, we perform machine learning on features derived from the tree structure. For differential expression analysis, we show massively boosted sensitivity in proteome, phosphoproteome as well as single cell data with up to 50-fold increases compared to a current state-of-the-art approach. We further address the notoriously difficult challenge of scoring quantification accuracy (as opposed to precision), i.e. whether the fold change of a peptide reflects the true abundance change. We benchmark on a mixed-species dataset and show superior classification abilities compared to standard metrics, removing biased peptides with an AUC of 0.94. We furthermore statistically compare and cluster peptides with similar quantitative behavior within our tree, allowing us to resolve proteoform profiles from bottom-up proteomics data. We combine this with deep learning classification of peptide sequences to infer regulated phospho-peptides from non-enriched proteome data alone. Finally, we apply our method to tissue measurements and shed light on the global proteoform diversity of the mouse proteome. AlphaQuant will be distributed as an open-source Python package within the AlphaX ecosystem, containing a graphical user interface and a one-click installer. Fig. 1 (optional) 600

601

P-II-0447 Zero-shot retention time prediction for unseen post-translational modifications with molecular structure encodings Ceder Dens1, Darien Yeung2, Oleg Krokhin2, Kris Laukens1, Wout Bittremieux1 1University of Antwerp, Computer Science, Antwerp, Belgium 2University of Manitoba, Manitoba Centre for Proteomics and Systems Biology, Winnipeg, Canada Introduction: Mass spectrometry-based proteomics has advanced our understanding of cellular processes, disease mechanisms, and drug target screening. However, identifying proteoforms carrying diverse post- translational modifications (PTMs) remains challenging. PTMs regulate protein activity and interactions and impact stability and localization. Limited understanding of the myriad PTMs and their influence on experimental properties like liquid chromatography (LC) behavior presents a significant hurdle. Peptide property prediction, such as retention times (RTs), has improved peptide identification but primarily for unmodified peptides or those with limited diversity in modifications. Methods: We present a Molecular Structure Transformer (MST), a transformer-based RT prediction model for peptides with any PTM. Our novel input representation encodes each amino acid and its PTM as a molecular structure graph, which is processed by a molecule encoder to obtain a separate embedding for each residue. Next, a transformer encoder is trained to predict the RT from these embeddings, enabling accurate predictions, even for peptides containing PTMs not seen during training. To enhance accuracy, we introduce a second model employing a two-step strategy. First, a regular transformer encoder predicts the RT of the unmodified sequence. Then, the MST predicts the RT shift induced by the modification. This strategy leverages the high prediction accuracy for unmodified peptides and the superior input representation for modified peptides. Results: DeepLC, our MST, and our two-step MST were trained on the Chronologer dataset. This dataset contains the RT of 1.3M unmodified and 913K modified peptides with 9 different PTMs. Subsequently, the models were tested on a collection of external datasets from various experiments, including ProteomeTools. The test dataset contains 70K peptides with 16 PTMs that were not included during training. Each test peptide contained at least one of these PTMs. We calculate the absolute error (AE) of the predicted iRT to compare the model"s performance. Over the complete test dataset, DeepLC achieves a mean AE (MAE) ± standard deviation of 24.07 ± 13.44, our MST does worse with 26.71 ± 23.19 but our 2-step model does significantly better with 12.83 ± 11.83. As a second metric, we calculate the macro average PTM MAE, which averages the MAE per unseen PTM. DeepLC achieves 12.54 ± 9.17, our MST 12.18 ± 9.71 and the 2-step model outperforms both models again with 10.57 ± 6.57. This marks a big improvement over previous approaches and constitutes state-of-the-art RT prediction performance on peptides containing previously unseen modifications. Fig. 1 (optional) 601

602

P-II-0448 Mixed-species based computational simulation of proteoform groups Isabell Bludau1, Constantin Ammar2, Lukas Käll3, Veit Schwämmle4, Felix Sahm1 1University Hospital Heidelberg, Neuropathology, Heidelberg, Germany 2Max Planck Institute of Biochemistry, Department of Proteomics and Signal Transduction, Martinsried, Germany 3KTH Royal Institute of Technology, Science for Life Laboratory, Solna, Sweden 4University of Southern Denmark, Department of Biochemistry and Molecular Biology, Odense, Denmark Many genes are known to give rise to multiple different protein products, also called proteoforms, that differ in their primary amino acid sequence and associated modifications. Although this expansion of molecular diversity on the protein level is playing a crucial role for functional diversification, proteoforms are challenging to detect in classical bottom-up proteomics workflows and are typically neglected. While this facilitates downstream interpretability of results, it is also known to introduce biases and valuable information from the peptide-level is lost. To address this shortcoming, we and others have previously developed computational tools that leverage information of peptide expression across large datasets to infer proteoform groups. A key challenge in the development of such tools is the lack of available ground truth data that can be used for training and benchmarking. Here, we propose a new strategy to computationally simulate proteoform data based on mixed-species experiments with various mixing ratios. The core concept is to create proteoform groups by combining peptides from proteins of the different species. Importantly, peptides shared between organisms mimic peptides shared between different proteoforms. The proposed approach conserves the natural variation and noise of proteomics measurements while also reflecting the key characteristics of differential proteoform expression. To enable easy adoption by the community, we developed ProteoformMixer, an open-source Python library for mixed-species based proteoform simulation. Users can either take a deposited or load their own mixed-species data. Proteoforms are subsequently simulated according to user-defined characteristics such as the desired number of proteoform groups per protein and the peptide distribution across them. ProteoformMixer also offers basic benchmarking functionality. Users can load their proteoform grouping results to retrieve the sensitivity and specificity for overall detection as well as statistics on correct peptide assignment. To showcase the strategy and functionality of ProteoformMixer, we utilized a previously published mixed-species dataset (Lou et al., 2023), comprising 35 DIA runs with 6 different mixing ratios of mouse and yeast proteomes. We simulated proteoform datasets with diverse characteristics to reflect a broad range of applications compatible with various proteoform detection algorithms, including PeCorA (Dermit et al., 2021), COPF (Bludau et al., 2021), and TPP-specific implementations (Kurzawa et al., 2023). While these tools provide an initial glimpse into the proteoform diversity detectable by bottom-up proteomics, further advancements are needed to enhance robustness, applicability, sensitivity, and specificity. We believe our new mixed-species based proteoform simulation approach will be a valuable model for training and benchmarking emerging proteoform detection tools, thereby supporting future developments in the field. 602

603

P-II-0449 Towards open modification search with MS Andrea Louise Marie Buur1, Viktoria Dorfer1 1University of Applied Sciences Upper Austria, Hagenberg, Austria In contrast to conventional closed database searches, open modification searches make it possible to search for hundreds or even thousands of different post-translational modifications (PTMs) in a single search, potentially increasing the identification rate of mass spectra. However, the inclusion of numerous PTMs exponentially expands the search space, leading to lengthy run times and a higher risk of matching false positive hits. Here we present MS Andrea, a new open modification search engine that utilizes sequence tags to reduce the number of peptide candidates to tackle the expanding search space. MS Andrea is able to accurately identify peptides and their PTMs, allowing insight into the modification landscape of proteins. The first step of the MS Andrea search algorithm is spectral pre-processing, encompassing the removal of precursor peaks and standard deconvolution. Subsequently, peaks are selected from 100 m/z windows in the spectrum. Using these picked peaks, sequence tags with a length of three and four residues are extracted from each spectrum. Peptides from the protein database are then ordered according to their sequence tags making it possible to extract only those peptides that contain at least one sequence tag found in the spectrum. This peptide candidate pool is then further filtered using a wide precursor mass tolerance and subsequently scored in two rounds using our in-house developed MS Amanda scoring function. In the first search round all the peptide candidates that remain after filtering based on the sequence tags and the precursor tolerance are matched to the spectrum, considering only fixed modifications. Subsequently, in the second search round, the ten highest scoring peptides are re-scored, this time considering combinations of PTMs that correspond to the delta mass between the uncharged precursor mass and the mass of the matched peptide. Currently, it is possible to identify up to two different types of variable modifications per peptide with MS Andrea. To evaluate the performance of MS Andrea, we conducted an analysis on two different datasets— phosphopeptides and peptides derived from HeLa cells. We compared the identification results with those obtained using the well-established closed search engine MS Amanda. Initial findings indicate a substantial overlap in peptide-spectrum matches between the two search engines across both datasets at 1% estimated false discovery rate. Our ongoing efforts involve continuous refinement and development of MS Andrea to achieve more efficient and accurate identification of modified peptides. This also includes comparing MS Andrea to other open modification search engines such as MS Fragger. 603

604

P-II-0450 DIAgui a shiny app to process the output from DIA-NN Luc Camoin1, Samuel Granjeaud1, Marc Antoine Gerault1 1Centre de Recherche en Cancérologie de Marseille - INSERM, Marseille Proteomics, Marseille, France Data-independent acquisition (DIA) proteomics is a recently-developed global mass spectrometry (MS)-based proteomics strategy. In DIA method, precursor ions are isolated into pre-defined isolation windows and then fragmented; unlike in data-dependent acquisition (DDA) where they are isolated for a specific m/z. Fragmented ions in each window are then analyzed by a high-resolution mass spectrometer. DIA technology has been increasingly utilized in various proteomics studies since it offers a broad protein coverage, high reproducibility, and accuracy. The analysis of these chimeric MS2 spectra is now facilitated thanks to AI. Many software dedicated to DIA are now available including DIA-NN which is increasingly popular in the proteomics community [1]. However, it is strongly advised to filter the DIA-NN output under R to get better quantification. For this, users must be comfortable with the R language. In addition, for the moment DIA-NN does not offer the possibility to get a TOP3 or iBAQ quantification. To overcome these problems, we offer DIAgui an R package based on Demichev's diann-rpackage that contains a user-friendly interface to process the output of DIA-NN. DIAgui contains two main functions: report_process which allows to process your report file output from DIA-NN with one R command and runDIAgui which launch the shiny app to process your file in an interactive way. After loading the report file, you can change the names of your fractions which are by default the path to the raw files used. Then, you can choose to extract the precursor, peptide, protein group or gene datasets from your report. You will have the possibility to filter according to some q values, to keep only proteotypic and to eliminate or not the modified peptides. For the protein group file, the Max-LFQ algorithm will be used to quantify proteins. You have two options: use the diann-rpackage or iq package method. These two methods are completely equivalent but the algorithm from iq package is much faster. You can also choose to get the Top3 and iBAQ quantification. For iBAQ, you can either load a FASTA file or use seqinr package which will make a query to SwissProt database. Use a FASTA file is way faster since the shiny app doesn"t have to make a query for each protein. For the other datasets, the app uses the function diann_matrix which is based on the function of the same name from diann- rpackage. Nevertheless, we added the features offers by the shiny app which are the possibility to obtain the Top3 quantification and the number of peptides used for the quantification (for the gene centred dataset), and also to take the sum or the max of the intensities of same ID. In the last tab of the app, you can visualize your data with an interactive heatmap, a density and MDS plot or others, and this for each your dataset (either the ones obtained in the app or one you uploaded). 604

605

P-II-0451 Evaluation of different rescoring platforms to increase peptide identifications Jesus Castano1,2, Francis Beaudry1,2 1University of Montreal, Department of Veterinary Biomedicine, Saint-Hyacinthe, Canada 2Centre de recherche sur le cerveau et l’apprentissage (CIRCA), Montreal, Canada The accurate identification of proteins and comprehensive coverage of the proteome are fundamental aspects of proteomics research. This can hugely impact the detection of proteins within complex biological samples, becoming a critical factor for paradigm-altering research in areas such as medicine, biofuels, or agriculture. Bottom-up proteomics using mass spectrometry relies on accurate comparison, facilitated by a search engine, between experimental and in silico spectra obtained from proteome databases. Over the last decade, the emergence of machine learning (ML) and deep learning (DL) approaches revolutionized this task; traditional comparisons that involved spectra with unit intensity peaks were transformed due to the use of ML and DL to predict spectral intensities. Furthermore, ML approaches integrated multiple matching features that were sub- utilized by search engines such as missed cleavages, retention time, and mass accuracy into a single score, in a process known as rescoring. Multiple rescoring functions have been developed in the past years, showing their potential to increase peptide identification. In this research, we compared several rescoring functions, offering some insights into their performance and discussing some caveats that might guide their selection for a specific proteomic experiment. We utilized HeLa digest samples, and we conducted a top 20 DDA experiment using a 50- cm separation column. The results were analyzed with the MaxQuant software at a 100%FDR, and the outputs were rescored using three different rescoring platforms: Oktoberfest, MS2Rescore, and inSPIRE. The number of peptides identified was substantially increased with all the platforms (40-53%) – with inSPIRE yielding the highest increase. Most of the peptides originally identified by MaxQuant were conserved after rescoring. However, a small percentage (3.5-5%) was lost, mainly because these peptides contained post-translational modifications that were not taken into account by the rescoring function. This could be an important limitation depending on the user"s needs. Another drawback is that rescoring increases processing time by up to 80%, not including the time spent in manual manipulations. Importantly, the need for command line-based software for this step can discourage less experienced users. Our results indicated that the peptide identification enhancement by the rescoring functions was related to their ability to detect a higher proportion of highly charged precursors and peptides with more missed cleavages, which mainly leveraged features from intensity peak prediction. In summary, the use of rescoring functions constitutes an important tool to significantly boost peptide identifications, which makes their implementation essential in proteomic pipelines. Their wide application, however, is still limited due to the low number of amino acid modifications included, increased time analysis, and the lack of integrated user-friendly platforms. 605

606

P-II-0452 pi-SPECFormer: pretraining improves spectrum embedding through contrastive learning with transformer Tianze Ling1, Cheng Chang1 1National Center for Protein Sciences (Beijing), Beijing, China Mass spectra are crucial as the start and the basic unit of shotgun proteomics data. However, spectra exhibit significant noise level and heterogeneity, limiting the efficiency of current spectrum analysis algorithms. Fortunately, large-scale pretrained models have introduce a new paradigm in artificial intelligence. Through pretraining on extensive datasets, these models develop a foundational understanding of data, leading to enhanced performance across downstream tasks. Consequently, developing a foundation model for spectrum data to decode the basic "words" of proteins and promoting proteomics data analysis, is challenging yet highly promising. Here, we introduced the first pretrained model in the proteomics field, pi-SPECFormer, designed to generate insightful spectrum embeddings for various spectrum analysis tasks. During its pretraining phase, pi-SPECFormer utilized a dictionary lookup task to identify analogous spectrum pairs in the representation space, achieving a comprehensive understanding of peak distribution patterns within spectra. In downstream applications, pi- SPECFormer was applied to de novo sequencing, presenting the state-of-the-art performance on the widely- used nine-species benchmark dataset, surpassing leading algorithms Casanovo (Yilmaz, M. et al.) and PointNovo (Qiao, R. et al.) by up to 14% and 28% in relative peptide recall. In the spectrum clustering task, pi-SPECFormer achieved relative 99% error reduction while maintaining a consistent clustering degree compared to the recognized spectrum embedder GLEAMS (Bittremieux, W. et al.). In the PTMs detection task, pi-SPECFormer demonstrated exceptional accuracy in identifying the characteristics of various PTMs, outperforming the leading algorithm AHLF (Altenburg, T. et al.) by 10-22% in relative accuracy for phosphorylation detection. Furthermore, it achieved over 95% accuracy across all 21 PTMs types in a synthesized dataset. These benchmark studies demonstrated pi-SPECFormer was a pioneering foundation model for MS data analysis. Our work highlighted the value of pretrained models for MS data, exploring and pushing the boundaries of foundation models in the proteomics field. 606

607

P-II-0453 MassDash: a web-based dashboard for data-independent acquisition mass spectrometry visualization Joshua Charkow1, Justin Sing1, Mohammed AlHigaylan1, Ira Horecka1, Leon Xu1, Hannes Rost1 1University of Toronto, Molecular Genetics, Toronto, Canada Introduction: In bottom-up mass spectrometry-based proteomics, Data Independent Acquisition (DIA) is a robust strategy for reproducibly quantifying the proteome. Due to its popularity, DIA methods continue to be refined and applied to new instruments and applications. Even though DIA proteomics is usually analyzed with automated software solutions, considering the vast set of applications, manual inspection of identified peptide features is necessary to ensure that algorithms are functioning as expected. Although well-established visualization tools exist, many are platform-specific or limited to one automated analysis tool which prevents comparison between tools. We sought to create a cross-platform, easy-to-use, and extendable framework for DIA proteomics visualization tool to be used post-automated analysis. Here, we present MassDash, a Python package, and Web-based interactive viewer, which currently supports OpenSwath, DIA-NN, and dreamDIA. MassDash is designed for tool comparison, validation, parameter optimization, and novel method development. Methods: MassDash is implemented as a Python package with a web-based GUI. MassDash ships with various feature detection algorithms including the OpenSwath and the Conformer deep-learning peak pickers. Interactive plotting is built using bokeh and plotly. MassDash has been tested with numerous datasets including, the SWATH-MS gold standard dataset, a diaPASEF dataset, and a synthetic-phospho dataset. Results: Using a raw mass spectrometry file and its corresponding identification results from OpenSwath, DIA- NN, or dreamDIA, MassDash allows for visualization of any given peptide precursor and the reported feature boundaries. MassDash supports various visualization options including 2D heatmap visualization across retention time and ion mobility and 3D visualization plots. Furthermore, users can also compare peptide quantification boundaries across different runs and software tools. Visual inspection of several peptide precursor examples identified by OpenSwath and DIA-NN demonstrates that peptide precursors come in varying quality and disagreement between software tools can occur with low-quality features. Moreover, MassDash can be used for realtime visualization of peak picking on the extracted ion chromatograms. Current built-in methods allow for on-the-fly parameter optimization which could then be used to reanalyze the experiment. Conclusions: MassDash is a versatile application supporting a on-the-fly feature detection and a variety of visualizations that can be used for parameter optimization and validation of results of multiple software tools. MassDash at https://github.com/Roestlab/massdash. More information can be found in our recent publication (DOI: 10.1021/acs.jproteome.4c00026). at https://massdash.streamlit.app. of MassDash open-source license and under a BSD 3-Clause accessed can be installed can is A demo version be 607

608

P-II-0454 imputomics 2.0: bridging data gaps in proteomics with missing value imputation algorithms and their benchmark Jarosław Chilimoniuk1, Krystyna Grzesiak1,2, Jakub Kołodziejczyk1, Dominik Nowakowski3, Adam Kretowski1, Rafał Kolenda4,5, Michał Ciborowski1, Michał Burdukiewicz1,6 1Medical University of Białystok, Clinical Research Centre, Białystok, Poland 2University of Wrocław, Faculty of Mathematics and Computer Science, Wrocław, Poland 3Medical University of Białystok, Department of Biostatistics and Medical Informatics, Białystok, Poland 4Norwich Research Park, Quadram Institute Biosciences, Norwich, United Kingdom 5Wrocław University of Environmental and Life Sciences, Faculty of Veterinary Medicine, Wrocław, Poland 6Autonomous University of Barcelona, Institute of Biotechnology and Biomedicine, Barcelona, Spain in proteomics data pose significant challenges, impacting downstream analysis and Missing values interpretation. To address this issue, we introduce imputomics 2.0, a comprehensive wrapper that integrates a wide array of missing value imputation algorithms (MVIAs) tailored specifically for proteomic data. imputomics 2.0 will integrate multiple state-of-the-art imputation algorithms, including but not limited to k-nearest neighbors (KNN), random forest (RF), expectation-maximization (EM), and singular value decomposition (SVD) and more, offering researchers a comprehensive toolkit to effectively handle missing data. imputomics 2.0 will provide a user-friendly web interface that streamlines the application of different imputation strategies, allowing researchers to effortlessly experiment with various algorithms and parameter settings. This flexibility will enable to customize and optimize imputation methods to suit the specific characteristics of individual proteomic datasets, ultimately enhancing the accuracy and reliability of downstream analyses. Central to our approach is the development of a unified framework that streamlines the use of different imputation algorithms. Therefore, any implemented algorithm must satisfy three conditions. Uniformity: the wrapper function standardizes input and output to 'data.frames' for tidy data processing, preserving original data structure and auxiliary functions designed specifically for 'data.frames'. Input data verification: input data is automatically verified for missing values ('NAs') or zeros/ones, ensuring all numeric, non-negative values are exclusively used, with optional additional integrity checks if recommended by the original imputation function authors. Passing arguments to function calls: relevant arguments are passed to underlying imputation functions while retaining default values, with any alterations to default behavior clearly documented, and each function enhanced with a boolean 'verbose' argument to suppress unnecessary screen prompts. Furthermore, we will conduct an extensive benchmark to evaluate the efficacy of each MVIA across diverse proteomic datasets. Additionally, we will examine the computational efficiency and scalability of each algorithm, crucial considerations for large-scale proteomic studies. Our tool will be available as web server, shiny application (GUI) and R package (CLI). Fig. 1 (optional) 608

609

P-II-0455 Phosphorylation site prediction model using structural context Yujin Choo1, Eunok Paek1,2,3, Seungjin Na4 1Hanyang University, Department of Artificial Intelligence, Seoul, South Korea 2Hanyang University, Department of Computer Science, Seoul, South Korea 3Hanyang University, Institute of Artificial Intelligence Research, Seoul, South Korea 4Korea Basic Science Institute, Cheongju, South Korea Post-translational modifications are covalent proceesing events that occur after protein synthesis and change the properties of a protein. In particular, phosphorylation is essential for cellular function and signaling and therefore, their identification is crucial for understanding the protein mechanisms. While various machine learning and deep learning models have been proposed to predict phosphorylation sites, the advent of AlphaFold2, with its high-accuracy protein structure predictions, presents new opportunities. We have developed a prediction model that leverages AlphaFold2"s structural prediction, enhancing sequence embeddings with structural information through a Transformer-based cross-attention mechanism. Our phosphorylation site prediction model was benchmarked against MusiteDeep, a sequence-based predictor, using an independent test set with an equal number of positive and negative phosphosite instances (5,074 each) from proteins not included in training/validation sets. MusiteDeep achieved an area under the ROC of 0.9118, whereas our model demonstrated an improved result of 0.9417. Additionally, our model showed enhanced precision (0.8321 over MusiteDeep"s 0.8042) and improved recall (0.9253 over Musitedeep"s 0.9034), at the same time. We also explored kinase-specific prediction through transfer learning to assess how effectively structural information is integrated into the embeddings. The results from our transfer learning approach surpassed those from fine-tuning and DeepPhos model, a sequence-based prediction tool, across all kinase- specific datasets. 609

610

P-II-0456 Tesorai Search: Large pretrained model boosts peptide identifications without the need for Percolator Peter Cimermancic1, Maximilen Burq1, Bryan Crampton1, Dejan Stepec1, Juan Restrepo2, Shivani Tiwary1, Ioana Clotea3, Jure Zbontar1, Shamil Urazbakhtin2, Beatrix Ueberheide3, Jürgen Cox2 1Tesorai, San Diego, CA, United States 2Max Planck Institute, Biochemistry, Martinsried, Germany 3New York University, Department of Biochemistry and Molecular Pharmacology, New York, NY, United States Current search algorithms for mass spectrometry proteomics typically fail to identify up to 75% of spectra. To mitigate the issue, machine learning (ML) has become a staple of modern database search engines for mass- spectrometry proteomics. Second-generation software such as Percolator, PeptideProphet, Prosit, AlphaPeptDeep, MSBooster, Peaks, Chimerys and others show impressive increases in the number of peptides and proteins identified, when compared to first-generation tools (eg, Sequest, X!Tandem, Comet, Andromeda, MS-GF+, and Sage). However, recent evidence has shown that the increased identifications come at the cost of a significant underestimation of the FDR estimate. They observe that this is inherent to how these tools leverage ML, where a new model is re-trained to separate targets from decoys for every run. This approach stands in sharp contrast with the recent trend in other fields of ML, where large models are trained only once and then applied on a wide range of modalities and use-cases. We hypothesized that training a single large peptide-spectrum matching (PSM) model can improve upon the gains from using Percolator-like tools, while solving the FDR control issue. To that end, we built a novel approach to peptide-spectrum matching, based on a pre-trained large deep learning model, which does not utilize decoys during training and does not require training a new model for every new sample. We trained the Tesorai model on over 100M real peptide-spectrum pairs, demonstrating that the approach performs robustly across a wide range of use-cases including standard trypsin-digested samples, metaproteomics, single-cell, and isobaric- labeled samples. In addition to providing robust FDR control, our method consistently increases identifications when compared to state-of-the-art approaches. For example, in immunopeptidomics, our approach leads to up to 90% increase in peptide identification, compared to MSFragger + Percolator. We are in the process of validating the new peptide identification, in a blinded reader study by human experts. The preliminary results indicate strong performance of the model, with its unique PSMs aligning with expert reviews better than PSMs unique to other search algorithms. With the significant improvement of the immunopeptidome coverage, we next set to analyze the landscape of the HLA-I-bound peptide sequences, re-analyzing several publicly available datasets. Interestingly, we find that our model enriches for non-tryptic peptides, suggesting trypsin-digest biases in current software, and offering new insights into immunopeptidome. Finally, to facilitate new discoveries in MS proteomics more broadly, we incorporated our model into a cloud-native, end-to-end system which can process 500 samples in less than an hour, and made it publicly available at www.tesorai.com. Fig. 1 (optional) 610

611

P-II-0457 DIA-NN: enabling peptidoform confidence in DIA proteomics Vadim Demichev1 1Charité – University Medicine Berlin, Quantitative Proteomics Lab, Berlin, Germany In recent years, data-independent acquisition (DIA) proteomics has become a widespread quantitative method, useful not only for large high-throughput screens, but also for applications that require high sensitivity, such as spatial and single cell proteomics, as well as phosphoproteomics and ubiquitinomics. Much of this progress in DIA technologies has been driven by improvements and new functionality in data processing software, such as support for ion mobility-resolved acquisition, PTM identification and multiplexing. However, wide-isolation window DIA methods still possess a fundamental limitation, that is the lack of inherent ability to confidently identify and quantify arbitrary peptidoforms. Not only can peptidoform misidentification affect protein quantities, it also holds back a range of applications of DIA proteomics, including metaproteomics and the emerging profiling of common sequence variants in population proteomics. Here I will introduce a new neural network-based peptidoform-scoring module in our DIA-NN software suite for DIA data processing. The new module allows the use of any DIA proteomics method to confidently identify and quantify peptidoforms with statistical error-rate control. Moreover, DIA can be optimised to maximise peptidoform confidence. In fact, the high proteomic depth, characteristic of DIA methods, is largely retained when the analysis is restricted solely to peptidoform-confident identifications. The new peptidoform-scoring module in DIA-NN, therefore, addresses the long-standing limitation of DIA proteomics and paves the way for new applications of DIA proteomics that require peptidoform confidence. 611

612

P-II-0458 IM2Deep predicts collisional cross-sections of modified and multiconformational peptide ions Robbe Devreese1,2, Alireza Nameni1,2, Arthur Declerq1,2, Alexander Kensert1,2, Ralf Gabriels1,2, Sven Degroeve1,2, Lennart Martens1,2, Robbin Bouwmeester1,2 1Ghent University, Center for Medical Biotechnology, Department of Biomolecular Medicine, Ghent, Belgium 2Ghent University, Center for Medical Biotechnology (CMB), Ghent, Belgium Instruments capable of measuring the ion mobility of peptides are now often part of the standard proteomics workflows. This opens up opportunities for the incorporation of ion mobility information in the peptide identification pipeline. As a result, there has recently been substantial progress in the development of collisional cross-section (CCS) prediction models. However, most of these recently developed CCS predictors struggle to accurately predict CCS for peptidoforms with modifications. This is especially problematic for increasingly popular open searches that must deal with extreme peptide identification ambiguity. Additionally, existing CCS prediction models overlook peptide ions that can adopt multiple conformations in the gas phase, since these are typically trained on only the most abundant conformer within a given dataset. However, this obviously removes valuable knowledge and creates a potential bias towards the most frequently occurring conformers. Addressing these limitations, we present IM2Deep, a deep learning-based CCS predictor capable of handling both peptides with any modification, as well as peptide ions exhibiting multiple conformations. IM2Deep uses the same principles as DeepLC, a retention time predictor, by encoding peptidoforms on the atomic composition level. This enables accurate CCS prediction for peptides with modifications unseen during training. To allow multi-output prediction for multiconformational peptide ions, we adapted IM2Deep"s architecture to support multi-output predictions, applying transfer learning to refine predictions on a dataset enriched with multiconformational peptide ions. Our results demonstrate that IM2Deep not only predicts CCS for multiple conformations with high accuracy but also improves the baseline performance for single-conformational peptide ions. Furthermore, we explore the application of graph neural networks (GNNs) in predicting CCS for peptide ions. GNNs leverage the full structural information of molecular graphs, making them particularly suited for CCS prediction, which is closely tied to the ion's structural configuration. By computing saliency maps from these trained networks, we can better understand the significance of specific molecular sub-structures in CCS prediction, enhancing interpretability of our predictions. 612

613

P-II-0459 Faster and more accurate intensity-based PTM localization in Chimerys Daniel Zolg1, Tobias Schmidt1, Siegfried Gessulat1, Florian Seefried1, Michael Graber1, Samia Ben Fredj1, Patroklos Samaras1, Markus Schneider1, Layla Eljagh1, Vishal Sukumar1, Michelle T. Berger1, Alexander Hogrebe1, Igor Bronshtein1, Pedro Navarro2, Kai Fritzemeier2, Yovany Cordero Hernandez2, Frank Berg2, Carmen Paschke2, David Horn3, Philip Loziuk3, Bernard Delanghe2, Christoph Henrich2, Martin Frejno1 1MSAID, Garching, Germany 2Thermo Fisher Scientific (Bremen) GmbH, Bremen, Germany 3Thermo Fisher Scientific, San Jose, CA, United States Background: Acetylation and ubiquitination dynamically regulate protein function and lifespan. Mass spectrometry-based bottom-up proteomics excels at detecting and quantifying modified peptides. Initially, data dependent acquisition (DDA) was often used due to its simplicity and precise precursor isolation. However, data independent acquisition (DIA) is gaining momentum despite challenges in post-translational modification (PTM) identification and localization. We integrated intensity-based PTM localization for DDA, WWA and DIA data into Chimerys and revamped its architecture to substantially reduce DIA data processing time. Methods: Chimerys, a spectrum-centric and acquisition method-agnostic search algorithm, deconvolutes chimeric spectra by maximizing the amount of explained experimental intensity with a minimal peptide set. Predictions stem from Inferys, a deep learning framework, formerly limited to unmodified, oxidized, and phosphorylated peptides. To support acetylation and ubiquitination, we included 16 M PSMs with common PTMs from public repositories in our training data foundation, crafting a new deep learning model. Integrated into our algorithm, this model powers an intensity-based PTM localization module, and an architecture overhaul improves CHIMERYS" DIA data processing speed. This version is integrated into Thermo Scientific™ Proteome Discoverer™ 3.2. Results: Chimerys' capabilities depend on the peptide tandem mass spectra our model was trained on. To surmount this, we deployed a cloud-native pipeline to access and process PRIDE datasets, merging them with our training data. The updated model now supports various PTMs, like N-terminal and lysine acetylation, lysine ubiquitination, TMT0 & TMTpro 'zero' and 'super heavy', unmodified cysteine residues, and the PreOmics™ iST- NHS cysteine modification. With extra training data, prediction accuracy increased, notably for phosphorylated peptides. Leveraging these advancements, we reanalyzed a public dataset, yielding a ~30% increase in identified modified peptides versus Sequest HT. Moreover, we introduced an innovative intensity-based algorithm for PTM localization. It compares predicted spectra of all potential isomers against experimental data, furnishing a localization score for each modified residue. This algorithm surpasses ptmRS on DDA datasets and adeptly localizes modified residues even in complex DIA spectra, embracing a spectrum-centric methodology. Furthermore, to facilitate large-scale DIA experiments, we fine-tuned the CHIMERYS workflow. Individual raw files are now submitted separately, expediting job execution. Additionally, a scoring logic overhaul slashed individual DIA raw file processing time twofold. These improvements, along with a revamped workflow in Proteome Discoverer, expedite processing and scale up experiments. Conclusion: A major update to a spectrum-centric, acquisition-method-agnostic search algorithm, enhancing PTM compatibility, and user experience. 613

614

P-II-0460 Altimeter and pioneer: fast and open-source software for analyzing narrow-window data-independent acquisition proteomics experiments Nathan Wamsley1, Ben Major1, Dennis Goldfarb1 1Washington University School of Medicine in St. Louis, Cell Biology & Physiology, Saint Louis, WA, United States We present Altimeter and Pioneer, cross-platform and open-source tools to predict peptide fragmentation patterns (Altimeter) and identify and quantify proteins and peptides from data independent acquisition (DIA) experiments given either in silico or empirically generated spectral libraries (Pioneer). DIA experiments deliberately co-isolate and co-fragment multiple peptides during each MS2 scan. Given a spectral library of predicted fragment ion intensities and precursor retention times, Pioneer quantifies the relative contributions of likely candidate precursor ions to each MS2 spectrum. In our benchmarks, Pioneer quantified 17% more precursor ions as DIA-NN 1.8.1 on an Astral three-proteome experiment, but in less than half the computation time and with more conservative FDR control. Pioneer is written in the Julia programming language and is entirely open-source. Altimeter is a deep learning model utilizing the UniSpec architecture (Lapin et al. 2024) trained to predict a fragment"s total abundance rather than the monoisotopic abundance. Pioneer searches an MSFragger-style fragment ion intensity index to identify all precursors for which the most abundant and at least two of the top three ranked library ions match to the spectrum. Subsequent expensive calculations are restricted to these candidates. Pioneer then computes expected isotope distributions for fragments from the spectral library based on the given isolation window and regresses them onto each empirical spectrum. It does so by a linear model that minimizes the pseudo-Huber loss under a non-negativity constraint. Pioneer then quantifies precursors using these regression coefficients. Pioneer scores precursors using several features and spectral similarity scores, most notably spectral entropy (Li et al. 2021), scribe score (Searle, Shannon, and Wilburn 2023), and Manhattan distance. We compared Pioneer to DIA-NN (default settings) with respect to data processing time and the number of quantifiable identifications using a publicly available three-proteome benchmark data set from Guzman et al. 2024. The data were searched against an Altimeter generated spectral library of tryptic H. sapien, S. cerevisiae, and E. coli, proteomes using Pioneer and an internally generated library with DIA-NN. Excluding library generation and formatting and including searching, quantitation, and scoring, data processing time for the six raw files was 19.5 min and 52.5 min for Pioneer and DIA-NN respectively. Benchmarking was carried out on a desktop computer (Windows 10, 3.00 GHz 18-Core Intel i9-10980XE CPU, 64 GB 2666 MHz DDR4) using 24 threads. For the comparison, we retained identifications at 1% FDR or better. Precursors identified in at least 3 of 3 samples in both conditions with a coefficient of variation of less than 20% were considered quantifiable. By these definitions, Pioneer quantified 112,750 precursors and DIA-NN quantified 96,405 precursors. 614

615

P-II-0461 Effect of sample size on statistical accessibility of proteins and enrichment analysis in proteomics experiments Francesco Greco1,2, Aldo Pastore2, Michele Emdin1, Liam A. McDonnell2 1Fondazione Toscana Gabriele Monasterio, Pisa, Italy 2Fondazione Pisana per la Scienza ONLUS, Pisa, Italy Background: Sample variability in proteomics comes from both biological and technical sources, but it can be overcome by an increase in the sample size. The estimation of the sample size necessary for (pre-)clinical experiments is essential to ensure its success. Despite the fact that each protein detected in a proteomics experiment is measured with its own coefficient-of-variation, too frequently sample size estimations are made using CV values from single proteins or using an average CV value. Here we calculate the number of samples needed to observe statistically significant differences for each protein in a dataset, to estimate the adequate sample size to statistically access most of the quantified proteins. Methods: We performed sample size estimation on publicly available datasets by using 1) the power analysis formula applying the Bonferroni correction; 2) generating an artificial dataset of true positives differences and repeating the t-test after sample size reduction; and 3) reducing sample size in empirical two-group statistical comparisons. Proteins accessible with smaller sample sizes were subjected to enrichment analysis to investigate the bias that small sample size has on gene ontology analysis. Results: The fraction of statistically accessible proteins varied with the characteristic variance of each dataset. In general, 10 vs 10 replicates per groups was the minimum number to be able to detect an effect size of 1 (log 2 transformed data) in 67-93% of the proteins of the datasets. Several KEGG pathways were found to be highlighted simply by the choice of a small sample amount, including Proteasome, Protein processing in endoplasmic reticulum, Prion disease, Parkinson and Alzheimer disease, Endocytosis, and Amyotrophic lateral sclerosis. Conclusions: The choice of larger sample sizes is crucial since it allows to statistically access larger fractions of the quantified proteins. Insufficient sample size may result in lack of power, but also might introduce bias into the enrichment analysis. Fig. 1 (optional) 615

616

P-II-0462 Feature screening and model regularization for k-mer representations of biological sequences Krystyna Grzesiak1,2, Jakub Kała2, Małgorzata Bogdan1, Michał Burdukiewicz2 1University of Wrocław, Faculty of Mathematics and Computer Science, Wrocław, Poland 2Medical University of Białystok, Clinical Research Centre, Białystok, Poland Proteins and peptides function effectively because they adopt specific spatial conformations. However, experimentally determining these structures is both costly and time-consuming. To accelerate this process, computational methods are often employed, leveraging the readily available amino acid sequences of proteins. Training deep learning models on such data, however, necessitates a large volume of annotated sequences. For peptide-specific models, where experimentally labeled sequences are scarce, classical statistical methods are preferred. Unfortunately, current state-of-the-art methods for peptide property prediction often fail to fully utilize the information embedded in amino acid sequences due to inefficient feature representations. Classical methods like Generalized Linear Models (GLMs) require structured data, leading us to focus on k-mer representations of proteins. These representations can be formatted as binary sequences indicating the presence of consecutive k-mers, or as integer count sequences. The protein properties we aim to predict span various data types, including binary (e.g., presence of disordered regions), categorical (e.g., subcellular location), and continuous (e.g., minimum inhibitory concentration of antimicrobial peptides). Our objective is to clarify the relationship between these protein properties and their k-mer representations. To achieve this, we introduce an advanced data simulation framework and methods for motif identification from k- mer data using information criteria like mBIC2 and regularization models. We conducted comprehensive benchmarking of various feature selection techniques, including Fast Correlation-Based Filter Solution (FCBF) and QuiPT, in conjunction with GLM-based regularization techniques. Our benchmarks identified the optimal feature selection methods for high-dimensional k-mer data, establishing a robust framework for accurately predicting protein properties and detecting motifs. 616

617

P-II-0463 NA deconvolution analysis(NADA) promotes excavation of unique proteins and protein-protein interactions from proteomics datasets Weiheng Guo1, Wenyi Jin1, Liang Zhang1 1City University of Hong Kong, Department of Biomedical Sciences, Hong Kong, Hong Kong The fast-advancing mass spectrometry and related technologies have greatly extended the depth of coverage in various proteomics studies. However, it remains challenging to handle proteins with missing values that are often presented as "NA" (not available) as sample numbers grow rapidly. We can interpret "NA" as evidence of no expression, low expression below the detection threshold, or as a false negative detection due to technical issues. Delineating these issues will help us analyze the complex proteome organizations in various biological contexts. In the current study, we first performed a comprehensive profiling of NA values in the proteomic datasets of over one thousand cell lines. We observed hundreds of non-NA proteins that are universally present in all cell lines. Such universal proteins are enriched in pathways of RNA processing, suggesting their essential roles in diverse cell types. Interestingly, handling NA values with a binary transformation had a robust performance in stratifying cell lines of suspension vs attachment culture conditions, and it also effectively separated non-small cell lung cancers (NSCLC) cell lines from small cell lung cancers (SCLC) ones. This prompted us to develop NA deconvolution analysis (NADA), a machine learning algorithm that extracts and integrates the features of neighboring samples/proteins to determine the nature of an NA value as "biological" or "technical". NADA could successfully identify characteristic proteins with binary expression patterns that classify cell lines of different tissue origins. For example, hematological and lymphoid cell lines are specifically missing proteins involved in cell adhesion and cell junction assembly pathways, but uniquely enrich proteins that are immune-related receptors, chemokines, and transcriptional factors. hematological origin from ones that grow on solid support. Moreover, NADA also facilitated the development of a missing-value-weighted method that improved protein-protein interaction (PPI) analyses via co-expression profiling. Finally, we applied NADA to analyze the proteomics datasets of extracellular vesicles (EVs) and identified differential PPI networks in EVs derived from cancerous and normal tissues. In conclusion, we present NADA as a powerful tool to delineate NA values and extract biological insights from the expanding datasets of proteomics. Fig. 1 (optional) 617

618

P-II-0464 Deep learning for MALDI-MSI: towards a common framework for transferable tissue diagnostics Patrick Jensen1,2, Jan Lellmann2, Herbert Thiele3, Pia Hönscheid4,5,6, Christian Sperling4,5, Gustavo Baretton4,5,6, Oliver Klein7, Carsten Tschöpe7, Karin Klingel8 1Technical University of Denmark, Department of Applied Mathematics and Computer Science, Lyngby, Denmark 2University of Lübeck, Institute of Mathematics and Image Computing, Luebeck, Germany 3Fraunhofer Institute for Digital Medicine MEVIS, Luebeck, Germany 4Technical University Dresden, Institute of Pathology, Medical Faculty and University Hospital Carl Gustav Carus, Dresden, Germany 5National Center for Tumor Diseases, Dresden, Germany 6German Cancer Consortium, Dresden, Germany 7BIH Center for Regenerative Therapies, Berlin, Germany 8University Hospital Tübingen, Institute for Pathology and Neuropathology, Tuebingen, Germany Matrix-assisted-laser-desorption/ionization mass spectrometry imaging (MALDI MSI) allows us to determine the spatial distribution of tissue analytes such as peptide signatures via their charge-to-mass ratio. This highly sensitive description of the tissue makes it attractive for clinical use as a supplement to existing diagnostic procedures and automated tools. Here, methods based on deep learning are emerging as a powerful paradigm for the analysis of MALDI MSI data. However, as the application to MALDI MSI data is still fairly nascent, unified methodologies transferable from one laboratory to another have yet to emerge. This makes adoption by clinical practitioners challenging, as a substantial amount of domain-level expertise is required to apply deep learning methods successfully. This work presents a common pipeline for deep learning-based MALDI MSI analysis of human tissue samples. We focus on the case of spectrum classification. Our pipeline is summarized in the figure. First, each spectrum is preprocessed by baseline removal, normalization (e.g., by total ion count), and downsampling. Note that, in contrast to most established approaches, we do not need to make use of any peak picking or other dimensionality reduction techniques. Then, we train a deep learning model to predict the class for a given spectrum. Specifically, we use a 1-dimensional Visual Transformer and optimize the cross-entropy loss with the AdamW optimizer. To increase the robustness of the model, we augment the spectra with Gaussian noise, random intensity scaling and mixup during training. We demonstrate that our pipeline is generally applicable by testing it on three MALDI MSI datasets from different clinical sources. For all datasets, our results are based on 3-fold cross-validation and we report the mean balanced accuracy (b.acc.) on the test sets for each cross-validation fold. We chose balanced accuracy as it is well-suited for imbalanced datasets. For the first dataset, the task is to distinguish between pancreatobiliary or intestinal type ampullary cancer (2 classes). Here, the pipeline achieves a b.acc. of 0.854. The second is a dataset where the task is to distinguish between AL and ATTR amyloidosis myocardial tissues (2 classes). Here, the pipeline achieves a b.acc. of 0.733. In the last dataset, the task is to distinguish between spectra from tissue containing pancreatic ductal adenocarcinoma or other pancreatic cancer types (4 classes). Here, the pipeline achieves a b.acc of 0.676. Importantly, all training hyperparameters remain the same for all datasets: While one may achieve better scores by tuning these to each datasets, our goal is specifically to show that a simple pipeline can achieve promising results without further user interaction. Our aim is that this may serve as a common baseline and, importantly, a good starting point for further research. To this end, we have also made the source code available at: https://github.com/patmjen/maldi_dl 618

620

P-II-0465 FLASHTagger: an open-source web application for ion type- and precursor mass-free protein identification in top-down mass spectrometry Kyowon Jeong1, Wonhyeuk Jung2, Tom Müller1, Jaywon Lee2, Aniruddha Panda2, Jared Shaw3, Louise Marie Buur4, Viktoria Dorfer4, Oliver Kohlbacher1, Kallol Gupta2 1University of Tübingen, Tuebingen, Germany 2Yale School of Medicine, New Haven, CT, United States 3University of Nebraska-Lincoln, Lincoln, NE, United States 4University of Applied Sciences Upper Austria, Hagenberg, Austria The growing capacity to detect proteins and protein complexes in MS pose computational challenges in identifying them through Top-DownMS (TDMS). While alternative fragmentation methods such as ECD and UVPD open up multiple fragmentation pathways increasing sequence coverage, they also complicate the interpretation of fragment spectra. Recently developed protocols like complex-down MS often isolate protein complexes at once yielding multiplexed fragment spectra. Together with complex signal structure of TDMS spectra and frequent errors in deconvolution, they present challenges in correct precursor ion interpretation. Addressing these issues, we present FLASHTagger, a high-sensitivity protein identification tool for TDMS platforms. Unlike most conventional database searches that work with a predefined set of fragment ion types and assume precursor ions of monomeric proteoform of a single kind, FLASHTagger employs a de novo sequence tag-based protein filtration algorithm that does not need ion type and precursor mass specifications. FLASHTagger takes deconvolved TD-MS/MS scans (currently deconvolved by FLASHDeconv) and a protein sequence database as inputs. On the deconvolved spectrum, high-scoring sequence tags are efficiently generated using multi-dimensional dynamic programming. The tags are used for fast searches to find the candidate proteins. For each matched protein, protein-level false discovery rates (FDRs) are calculated with a decoy database. FLASHTagger can be configured for both targeted and discovery studies by specifying the database file input. For both modes, however, deconvolution and tag generation are done blindly (e.g., without sequence information). We first validated the targeted mode using monoclonal antibody standards, where FLASHTagger found tags corresponding to both the heavy and light chain CDR3 regions, demonstrating it can robustly identify heterodimers. To benchmark discovery mode, we analyzed three nTD (native TD) datasets which were generated by isolating and fragmenting (EChcD) native oligomeric states of VAMP2 (monomer, rat), semiSWEET (dimer, yeast), and AqpZ (tetramer, E. coli), respectively, directly from intact liposomes. The data was searched against a proteome database containing 1,775 E. coli membrane proteins plus the target proteins. For all cases, target proteins were identified as the top hits. The matched tags revealed various ion types including terminal b/y and c/z ion types and even internal ions. For AqpZ, we could identify the unmodified, as well as dimethylated proteoforms. Current implementation of FLASHTagger focuses on the low complexity datasets, but analysis of complex datasets will be made available in near future as a part of our new proteoform search engine. The precursor independent feature of FLASHTagger would open up the gate toward data independent acquisition in TDP. FLASHTagger is deployed at https://abi-services.cs.uni-tuebingen.de/flashviewer/ (Figure 1). 620

622

P-II-0466 Inferring dynamic interaction network landscape with deep graph contrastive representation learning Wenyi Jin1, Weiheng Guo1, Liang Zhang1 1City University of Hong Kong, Biomedical Sciences, Hong Kong, Hong Kong Multi-protein complexes carry out a large amount of essential cellular functions. Dynamic protein-protein interactions (PPIs) play a crucial role in regulating the assembly/disassembly of protein complexes. The fast accumulation of proteomics datasets has enabled the computational integration of large-scale protein abundance and known interaction features to construct/refine PPI networks. However, the dynamic PPI changes that underlie transitions of cellular states are often obscured in noise and susceptible to heterogeneity across samples. To address this challenge, we developed DsINE (Dynamic sample Interaction Network Examination), a deep learning framework that utilizes contrastive graph representation to calculate fine-grained molecular networks of individual samples in large biological datasets. DsINE integrates the co-expression degrees of interacting partners with various features of cellular states to determine the dynamic landscape of PPI complexes. It maximized mutual information across specific sub-graphs and homogenized background to capture graphic structures and define state-specific PPIs. In practice, DsINE could illustrate the refined landscape of PPIs in breast cancer cell lines and accurately determine the featuring molecular complexes of an individual sample. Moreover, DsINE dissected neuroblastoma subtypes' PPIs at the single-cell level, revealing dynamic alterations of molecular networks during neuroblastoma differentiation and development. Finally, we exploited DsINE to analyze extracellular vesicles (EV) and discovered distinctive PPI landscapes in EVs derived from cancer and normal tissues, which demonstrated robust diagnostic performance. In conclusion, DsINE is a powerful tool that exploits fast-growing omics datasets to delineate the PPI networks in various biological contexts. Fig. 1 (optional) 622

623

P-II-0467 XL-Prosit: accurate fragment intensity prediction and enhanced cross-linked peptide identification for protein structural and interaction studies Mostafa Kalhor1, Falk Schimweg2, Mario Picciani1, Lutz Fischer2, Juri Rappsilber2, Mathias Wilhelm1 1Technical University of Munich, Freising, Germany 2Technical University of Berlin, Berlin, Germany Introduction: Chemical cross-linking (XL) mass spectrometry (MS) is an effective tool for analyzing protein structure and protein-protein interactions. Here, we extend Prosit to predict the fragment ion intensities for cleavable MS2 (CMS2), cleavable MS3 (CMS3), and non-cleavable MS2 (NMS2) cross-linked peptides covering the most well known crosslinkers (DSSO, DSBU, and DSS/BS3). Integrating XL-Prosit into Oktoberfest, a data- driven rescoring pipeline, and utilizing Percolator for rescoring resulted in up to 450%, 233%, and 265% improvement in confidently identified crosslink-spectrum matches (CSMs), unique cross-linked peptides (UXLs), and protein-protein (PP) interactions levels compared to xiSEARCH, respectively. XL-Prosit is available via the public prediction service Koina, and our rescoring pipeline is readily available, allowing immediate use by the community. Methods: We systematically evaluated methods to fine-tune a pre-trained model of Prosit to allow the prediction for unknown modifications using the concept of transfer learning. In addition, we improved the accuracy of the prediction model by calibrating collision energy across datasets and augmenting training data by swapping the position of the two cross-linked peptides in the context of a model that focuses on the prediction of the intensity pattern of one. XL-Prosit was integrated into Oktoberfest (github.com/wilhelm-lab/oktoberfest) via Koina (koina.wilhelmlab.org) to extract novel intensity-based spectral similarity features for each unfiltered identified CSM obtained from a search engine (Figure A). Subsequently, Percolator was applied to efficiently discriminate correct from incorrect CMSs. This was achieved by training Percolator on PSM level (alpha and beta peptides separately) rather than at CSM level and generating CSM-, UXL-, and PPI-level scores for the final FDR calculation. Results: Raw files were downloaded from PRIDE and analyzed using Plink 2 and XlinkX. Currently, 124k MS2 and 32k MS3 spectra for cleavable and 36k MS2 spectra for non-cleavable cross-linked peptides were collected. Our model's performance, evaluated on a test dataset, reveals an accuracy exceeding 0.96, 0.94 and 0.93 on CMS3, CMS2, and NMS2 based on Pearson correlation coefficient (PCC) (Figure B), respectively. To verify if our FDR estimate is well calibrated, we utilized XL-Prosit on one well-controlled XL-MS dataset standard comprising hundreds of recombinant proteins that are systematically mixed for cross-linking. Our pipeline achieves lower experimentally validated FDRs (0.67%, 0.88%, and 1.52%) compared to Scout alone (1.18%, 1.13%, and 2.3%), while improving identification rates by 66%, 63%, and 30% for CSMs, UXLs, and PPs, respectively, at an estimated 1% FDR. Last, we applied XL-Prosit to another large-scale dataset, which yielded a remarkable improvement, of 450% increase in CSMs, 233% in UXLs, and 265% in PPs compared to the xiSEARCH alone (Figure C), all below 1% FDR. Fig. 1 (optional) 623

624

P-II-0468 TIMSrescore: timsTOF-optimized PSM rescoring boosts identification rates for immunopeptidomics Jonathan Krieger1, Tharan Srikumar1, Dennis Trede2, George Rosenberger3, Arthur Declerq4, Ralf Gabriels4, Robbin Bouwmeester4, Sven Degroeve4, Lennart Martens4, Michele Genangeli5 1Bruker Ltd, Mississauga, Canada 2Bruker Daltonics GmbH & Co. KG, Bremen, Germany 3Bruker Switzerland AG, Faellanden, Switzerland 4Ghent University, CompOmics, Ghent, Belgium 5Bruker Nederland BV, Leiderdorp, Netherlands Introduction: The immunoproteome, the subset of protein involved in immune response, are key components of various disease mechanisms and recent progress in MS instrumentation and analysis have gained considerable attention for this special application. However, because immunopeptides are non-enzymatically cleaved, the necessary search space dramatically increases, resulting in lower peptide identification rates. It has been demonstrated that the introduction of orthogonal scores comparing predicted vs observed peptide properties can increase both sensitivity and specificity in these scenarios. Here, we introduce TIMSrescore, a timsTOF- optimized algorithm based on MS2Rescore, that provides these benefits for the timsTOF family of instruments. Methods and Results: The peptide fragmentation (MS2PIP) and collisional cross section (im2deep) predictors employed in TIMSrescore have been trained using a diverse set of timsTOF dda-PASEF PSMs. The database search engine Sage and MS2Rescore have been extended to support timsTOF data. For benchmarking, three different types of datasets were used: 1) a single dda-PASEF run of an in-house whole cell lysate proteome mixture (human, yeast, E. coli), 2) a single dda-PASEF run of a fractionated, phosphopeptide-enriched dataset (Skowronek et al., MCP, 2022), 3) a single dda-PASEF run of an HLA-I immunopeptidomic dda-PASEF dataset (Phulphagar et al., MCP, 2023). Sage and MS2Rescore were run with default parameters for standard, phosphoproteomic and immunopeptidomic analyses (8-15 AA). For the whole cell lysate and phosphoproteomic datasets, TIMSrescore increased recovery of peptides by 2.1 and 2.0%, respectively. However, 12.5 and 15.9% of peptides were reprioritized, due to higher confidence after rescoring. For the immunoproteomic dataset, a considerable increase of 21.8% could be observed, indicating that rescoring is particularly impactful for large search spaces. Particularly the timsTOF optimizations of the fragmentation predictors were critical, improving median Pearson correlation to 0.88 from 0.53 (standard MS2PIP HCD model). Conclusions: TIMSrescore represents a timsTOF-optimized rescoring approach that can improve recovery of peptide identifications on immunopeptidomic applications substantially. 624

625

P-II-0469 MHC validator boosts unbiased discovery of conserved T-Cell epitope vaccine candidates through improved immunopeptidomics sensitivity Kevin Kovalchik1, David J. Hamelin1,2, Peter Kubiniok1, Benoîte Bourdin1, Fatima Mostefei2, Raphael Poujol2, Zhaoguan Wu1, Bastien Pare1, Shawn Simpson1, John Sidney3, Eric Bonneil4, Mathieu Courcelle4, Sunil Kumar Saini5, Bayrem Gharsallaoui1, Saketh Kapoor1,6, Jean-Christophe Grenier2, Loïze Maréchal1, Christopher Savoie1, Alesandro Sette3, Pierre Thibault4,7, Isabelle Sirois1, Martin Smith1, Hélène Decaluwe1, Julie Hussin2,8, Mathieu Lavallée-Adam2,9, Etienne Caron1,6,10 1St Justine Research Centre, Montreal, Canada 2University of Montreal, Montreal Heart Institute, Bioinformatics, Montreal, Canada 3Center for Infectious Disease and Vaccine Research, La Jolla Institute for Immunology, La Jolla, CA, United States 4Institute of Research in Immunology and Cancer, Biochemistry, Montreal, Canada 5Department of Health Technology, Section of Experimental and Translational Immunology, Technical University of Denmark, Kongens Lyngby, Denmark, Experimental and Translational Immunology, Kongens Lyngby, Denmark 6Yale Center for Immuno-Oncology, Yale School of Medicine, New Haven, CT, United States 7University of Montreal, Department of Chemistry, Montreal, Canada 8University of Montreal, Department of Biochemistry and Molecular Medicine, Montreal, Canada 9University of Ottawa, Department of Biochemistry, Ottawa, Canada 10Yale University, Department of Immunobiology, New Haven, CT, United States Next-generation T-cell-directed vaccines against COVID-19 aim to induce long-lasting T cell immunity to protect against circulating and future hypermutated SARS-CoV-2 variants. Despite the potential of mass spectrometry (MS)-based immunopeptidomics in guiding T-cell vaccine design against rapidly mutating viruses, persistent computational challenges hinder the sensitive, accurate and unbiased identification of conserved, vaccine- relevant T-cell epitopes. To address these challenges, we present a comprehensive framework that integrates a novel machine learning algorithm, immunopeptidomics, intra-host data, epitope immunogenicity, and global CD8+ T-cell epitope conservation analyses. At the forefront of our approach is MHCvalidator, an innovative artificial neural network-based algorithm designed to increase the sensitivity of MS-based immunopeptidomics experiments by modeling antigen presentation and sequence-feature properties, enhancing the unbiased discovery of both self and viral peptide antigens. Notably, the application of MHCvalidator led to the identification of a new form of nonconventional SARS-CoV-2 T-cell epitope presented by B7 supertype molecules. This discovery stems from a +1-frameshift in a truncated version of the Spike (S) antigen, supported by ribo-seq data. Remarkably, analysis of SARS-CoV-2 proteomes from ~80,000 COVID-19 patients unveiled a prevalent S antigen truncation in ~65% of cases, revealing a rich source of frameshifted and untapped viral antigens. Furthermore, integral to our framework, we developed a new computational pipeline to track the global mutational dynamics of MHCvalidator-identified SARS-CoV-2 CD8+ epitopes from the T-cell-directed vaccine BNT162b4, currently undergoing clinical trials. While most CD8+ epitopes encoded by the vaccine exhibit global conservation from January 2020 to October 2023, a highly immunodominant A*01-associated epitope, especially in hospitalized patients, experiences substantial mutations in the Delta and Omicron variants. Collectively, our approach unveils unprecedented SARS-CoV-2 T-cell epitopes, elucidates novel antigenic features, and highlights the non-trivial mutational dynamics of immunodominant, vaccine-relevant epitopes. Our approach can be applied to any viruses and underscores the need for continuous vigilance in T-cell vaccine development against the evolving landscape of circulating and future hypermutated SARS-CoV-2 variants. Fig. 1 (optional) 625

626

P-II-0470 midiaID: a flexible pipeline for the analysis of midiaPASEF data Mateusz Lacki1, Ute Distler1, Michał Piotr Startek1,2, David Teschner3, Tanja Ziesmann1, Stephanie Kaspar- Schoenefeld4, Jens Decker4, Florian Krohs4, Jonathan Krieger5, Oliver Raether4, Stefan Tenzer1 1University Medical Center of the Johannes Gutenberg-University Mainz, Institute for Immunology, Mainz, Germany 2University of Warsaw, Faculty of Mathematics, Informatics, and Mechanics, Warsaw, Poland 3Johannes-Gutenberg University Mainz, Institut for Informatics, Mainz, Germany 4Bruker Daltonics GmbH & Co. KG, Bremen, Germany 5Bruker Ltd, Milton, Canada Background: midiaPASEF is a novel Data-Independent Acquisition (DIA) method implemented on Bruker timsTOF devices. midiaPASEF offers the benefit of recording the precursor origins of fragment measurements. Due to its novelty, the method's advantages cannot be fully utilized with existing data analysis solutions. Therefore, we present midiaID: a flexible pipeline for the analysis of midiaPASEF data. Methods: midiaID is a multistep procedure organized into a Snakemake pipeline. This allows for easy investigation of different pipeline setups, optimization, and scalability depending on the hardware used. The pipeline proceeds along the following lines: 1. Precursor and fragment data is clustered 2. For each fragment cluster we predict its precursor position roughly matching it with potential precursors. 3. Obtained fragment-precursor matches intensity patterns are compared and unlikely matches filtered out. 4. Data is then cast into an MGF format and searched with a DDA search engine, SAGE. 5. Search results are mapped back to the original graph, allowing for recalibration of measured quantities and the training of machine learning model for edge quality assessment. 6. Low scoring edges are filtered out and remaining one are searched again and findings rescored. The two-searches approach allow cleaning of the precursor-fragment graph, providing the search engine with less noisy spectra and thus -- enhancing the quality of matches. Conclusions: midiaID is a flexible and scalable pipeline for the analysis of midiaPASEF data. 626

627

P-II-0471 A novel supervised learning algorithm for real-time collision energy selection to optimize peptide fragmentation in mass spectrometry Yun-En Chung1, Matthew Willetts2, Jens Decker2, Nagarjuna Nagaraj2, Jonathan Krieger3, Tharan Srikumar3, Mathieu Lavallée-Adam1 1University of Ottawa, Ottawa, Canada 2Bruker Daltonics Inc., Bremen, Germany 3Bruker Ltd, Milton, Canada The ability to identify peptides, proteins and their associated post-translational modifications using mass spectrometry is directly linked to the fragmentation level of peptide precursor ions. Typically, a well-fragmented precursor ion generates a tandem mass spectrum with a high signal-to-noise ratio, which facilitates peptide sequence identification. Precursor ion properties, such as mass-to-charge ratio (m/z), charge state, and ion mobility coefficient influence the level of collision energy required for optimal fragmentation. Nonetheless, most mass spectrometers do not make use of all these pieces of information when attempting to determine the optimal collision energy for a given peptide. Furthermore, current approaches optimizing precursor ion fragmentation, typically rely on knowing peptide sequences prior to analysis or attempting many fragmentations with different collision energies during the experiment. To address these limitations, we present a novel supervised learning strategy that uses precursor ion property measurements from mass spectrometry to determine in real-time the collision energy level that optimizes precursor ion fragmentation. Specifically, we designed an artificial neural network using precursor ion properties such as m/z, charge state and ion mobility coefficient to predict precursor ion fragmentation level when a given collision energy is applied. To train our network, we built a dataset of data-dependent acquisition (DDA) mass spectrometry analyses of commercial human cell lysates (Promega) on a timsTOF Pro (Bruker) mass spectrometer using fixed collision energy values in each analysis, ranging from 5 to 100eV. This algorithm is then used to determine in real-time, during mass spectrometry analysis, the optimal collision energy for a given precursor ion. We show that our algorithm accurately predicts precursor ion relative fragmentation based on the collision energy applied in DDA mass spectrometry analyses of whole cell lysates (r2=0.72). When our tool selects the optimal collision energy level for a given precursor ion, peptide and protein identifications from PEAKS de novo sequencing increase by 15% and 5%, respectively, compared to the instrument"s default collision energy settings (false discovery rate (FDR)<1%). Despite being trained on DDA data, our algorithm can determine the optimal collision energy for peptide ions in each isolation window in real-time during data-independent acquisition (DIA) mass spectrometry experiments and increase the number of peptide identifications by 9.0% (FDR<1%). Finally, we show that when applied to human cell lysates enriched for phosphorylated peptides, our method identified 12% more post-translationally modified peptides (FDR<1%). Overall, our novel precursor ion fragmentation optimization method improves peptide, protein and post- translational modification identifications, thereby providing a more comprehensive characterization of samples analyzed by mass spectrometry. 627

628

P-II-0472 GPMelt: a versatile and interpretable Gaussian process-based statistical framework for analyzing thermal proteome profiling datasets Cecile Le Sueur1, Mikhail Savitski1, Magnus Rattray2 1EMBL Heidelberg, Genome Biology unit, Heidelberg, Germany 2University of Manchester, Division of Informatics, Imaging & Data Sciences, Manchester, United Kingdom Thermal proteome profiling (TPP) is a proteome-wide technology combining the cellular thermal shift assay with quantitative mass spectrometry to provide insights into protein interactions and states. Statistical analysis of temperature range TPP (TPP-TR) datasets involves comparing protein melting curves, describing the amount of non-denatured proteins as a function of temperature, between different conditions (e.g. presence or absence of a drug). Current state-of-the-art models are restricted to sigmoidal melting behaviors, yet unconventional melting curves represent up to 20% of TPP-TR datasets. As shown in the literature, important biological information is likely to be carried by non-sigmoidal melting curves. We recently introduced GPMelt, a novel statistical framework based on hierarchical Gaussian Process (GP) models, to make TPP-TR datasets analysis unbiased with respect to the proteins' melting profiles. The model robustly integrates replicates information, accommodates multiple conditions and handles any melting curve shape. By further formulating the hierarchical GP model into a multi-task GP regression, we provide an interpretable statistical model which can easily be extended to more complex TPP protocols. As an illustration, the analysis of peptide-level TPP-TR datasets, considering melting curves of tryptic peptides instead of protein averages, is implemented using deeper hierarchies. Unbiased analyses of peptide-level TPP-TR datasets, of high value for the study of protein post- translational modifications, were yet hindered due to the abundance of unconventional melting curves. A second example is the extension of the GPMelt model to 2D-TPP datasets. 2D-TPP datasets compare protein thermal stability across a larger number of conditions, using a distinct sample multiplexing strategy that previously impeded melting curve reconstruction. Retaining the melting curve shape, and hence key biological information, strengthens the accuracy of 2D-TPP discoveries. Collectively, GPMelt extends the analysis of TPP-TR datasets for both protein and peptide levels melting curves, offering access to thousands of previously excluded melting curves. The hierarchical GP model's versatility and interpretability permits its extensions for application to nearly any TPP-based protocol, paving the way for groundbreaking biological discoveries in protein interactions, localisation and functions. 628

629

P-II-0473 A Shrinkage-based statistical method for testing group mean differences in quantitative bottom-up proteomics Namgil Lee1,2, Hojin Yoo1, Heejung Yang1,3, Juhyeong Kim2 1Bionsight, Seoul, South Korea 2Kangwon National University, Information Statistics, Chuncheon, South Korea 3Kangwon National University, Department of Pharmacy, Chuncheon, South Korea A statistical method for testing group mean differences in quantitative bottom-up proteomics is proposed. A probabilistic graphical model is introduced that explains the performance degradation of traditional statistical testing methods by random variations in ionization efficiency. The proposed method enhances a traditional feature-based test statistic by integrating nonparametric shrinkage estimation of covariance matrices and bootstrap estimation of degrees-of-freedom. Numerical experiments, based on simulated and real quantitative tandem mass spectrometry data, illustrate that the proposed method surpasses traditional statistical methods in specificity and sensitivity, particularly under conditions of small sample sizes. 629

630

P-II-0474 Transforming de novo peptide sequencing by explainable AI Zhendong Liang1, Tianze Ling2, Tingpeng Yang1, Yonghong He1, Yu Wang3, Cheng Chang4 1Tsinghua University, Tsinghua Shenzhen International Graduate School, Shenzhen, China 2Tsinghua University, School of Life Sciences, Beijing, China 3Peng Cheng Laboratory, Shenzhen, China 4Chinese Academy of Medical Sciences, Research Unit of Proteomics Driven Cancer Precision Medicine, Beijing, China De novo peptide sequencing is essential for identifying novel proteins, yet its broader application is constrained by the lack of a robust quality control system. In response, we have developed a transformer-based model, π- xNovo, which significantly enhances sequencing accuracy. Through a detailed analysis of the model's attention matrix, we have elucidated the contribution of mass spectrometry peaks to amino acid predictions. Leveraging these insights, we designed the π-xNovo-QCS system, which discern peptide predictions with accuracies exceeding 80% and sensitivities above 90%. Applying this model to a large-scale deep human proteome dataset resulted in the identification of 1,931,761 additional peptides, marking a 137% increase over traditional database search results. These newly identified peptides facilitated a 17.9% increase in protein identification, a 23.59% increase in the detection of single amino acid polymorphism events, and a 20.02% increase in exon- skipping splicing events. The deployment of this explainable AI system holds significant potential for expanding the application of de novo peptide sequencing, particularly in exploring the darker matter of the human proteome universe. 630

631

P-II-0475 MS-BERT: a pre-trained model for data-independent acquisition mass spectrometry-based proteomics data analysis Zhiwei Liu1, Pu Liu2, Zongxiang Nie2, Yingying Sun1, Yuqi Zhang1, Yi Chen2, Tiannan Guo1 1Westlake University, Hangzhou, China 2Westlake Omics (Hangzhou) Biotechnology, Hangzhou, China Analyzing data-independent acquisition mass spectrometry (DIA-MS) based proteomics data remains a formidable challenge. DIA identification tools, exemplified by DIA-NN, typically employ a multi-stage approach, initially scoring peak groups and then refining the selection through iterative filtering. This iterative process harnesses an expanding array of features, progressively enhancing the discriminative power of the scoring model to bolster selection accuracy. Nevertheless, extant DIA identification tools encounter constraints. They employ independent scoring models for each mass spectrometry file, and owing to the restricted training samples from each file, these tools frequently resort to conventional machine learning models with limited capacity to avert overfitting. Consequently, they often fall short in adequately capturing intricate peptide-peak-group matching patterns. The prevalent reliance on manually engineered features and the isolation of feature learning from classifier training in current methodologies hinder optimization and limit overall analytical performance. BERT, an acronym for Bidirectional Encoder Representations from Transformers, stands at the forefront of natural language processing (NLP) breakthroughs. Here, we introduce MS-BERT, a transformer-based pre-trained AI model tailored for analyzing DIA-MS proteomics data. This paper unveils a pioneering DIA identification algorithm leveraging an encoder-only transformer model. The models in the initial and subsequent stages undergo training on extensive datasets comprising 233 million and 7.4 million training instances sourced from 600 mass spectrometry files, respectively. The breadth of these datasets equips our models with exceptional generalization prowess, facilitating precise identification without necessitating separate training on test files. Harnessing the robust data modeling capabilities inherent in convolutional and transformer networks empowers our models to adeptly capture intricate peptide-peak-group matching patterns. We have obtained a novel human proteome dataset, comprising three samples each from human tissues and human serum, for performance evaluation. In comparison to the prevailing DIA-NN software, MS-BERT demonstrates the capability to identify over 4,000 human serum proteins, resulting in a remarkable enhancement of 45% in protein identification depth and 22% in precursor coverage within human proteomes. For quantification assessment, we utilized a three-species dataset (human, yeast, and C. elegans) sourced from the PRIDE database. Notably, MS-BERT exhibits quantification performance on par with DIA-NN, evidenced by a Pearson"s correlation coefficient of 0.91 for protein quantification. This study introduces a robust new software tool for DIA proteomic identification and quantification, underscoring the substantial potential of pre-trained models and synthetic datasets in propelling advancements in DIA proteomics analysis. 631

632

P-II-0476 MsBayesImpute: a versatile framework for handling missing values in mass-spectrometry data through a combined Bayesian factorization and probabilistic dropout model Junyan Lu1, Jiaojiao He1 1Heidelberg University, Medical Faculty Heidelberg, Heidelberg, Germany Advancements in mass spectrometry (MS) technologies have revolutionized the quantification of proteins, their modifications, and metabolites in biological samples, significantly accelerating biomedical research. However, MS-based proteomic and metabolomic datasets are often plagued by missing values due to a complex interplay of missing at random (MAR) and missing not at random (MNAR) mechanisms. Such missing data, if not properly addressed, can result in both false positive and false negative results in downstream analyses and interpretations. Existing imputation methods typically focus on either MAR, using Bayesian PCA, random forest, and k-nearest neighbors approach, or MNAR, using left-censored methods like quantile regression imputation of left-censored data (QRILC), deterministic minimum (MinDet), and probabilistic minimum (MinProb). However, few techniques effectively address both MAR and MNAR within a single framework and those that do often require manual tuning of mixture percentages between MNAR and MAR or rely on specific two-group experimental designs. This is inadequate for many biomedical studies, which involve large, heterogeneous patient samples and often require the usage of machine learning techniques on complete data matrices. To improve the handling of missing values in MS-based proteomic and metabolomic data from biomedical samples, we developed msBayesImpute, a novel computational method that combines Bayesian factorization with a probabilistic dropout model. We benchmarked msBayesImpute against several popular imputation methods using both simulated and real-world datasets, demonstrating superior performance in reconstructing the missing values or identifying differentially expressed proteins/metabolites, across various levels of missingness. Our model employs stochastic variational inference, offering considerable speed advantages over Bayesian PCA and missForest, especially in large datasets. Importantly, msBayesImpute does not require predefined experimental design information, such as group structures, making it highly adaptable to various study designs and compatible with any downstream analytical tools. This versatility makes msBayesImpute an effective and robust tool for enhancing the utility of MS datasets in biological research. 632

633

P-II-0477 Identifying and quantifying the unknown. Two workflows combined to search for variant/mutation information and to quantify un-/identified spectra in MS-based proteomics Dominik Lux1, Julian Uszkoreit2, Brit Mollenhauer3,4, Katalin Barkovits-Boeddinghaus1, Martin Eisenacher1, Katrin Marcus-Alic1 1Ruhr University Bochum / Medical Proteome Center, Bochum, Germany 2Ruhr University Bochum / Medical Bioinformatics, Bochum, Germany 3University Medical Center Göttingen, Goettingen, Germany 4Paracelus-Elena Klinik, Kassel, Germany Introduction: Today's data analysis in MS-based proteomics primarily relies on typical identification and quantification workflows. These workflows usually utilize a FASTA database where mostly only canonical sequences of proteins are provided. In such workflows a theoretical spectrum is matched to an actual measured spectrum (search engine) returning identified spectra, which further are used to quantify the found peptide/protein. While these workflows usually suffice for the selected use-cases, not all spectra are identified and only a part of all measured molecules in a mass spectrometer are quantified, thus not reported in the final data analysis. Our workflows aim to shed some light on the unknown, trying to identify non-canonical sequences and by quantifying charged molecules, regardless of whether they are identified or not. Methods: In our lab, we developed two workflows: a sophisticated FASTA generator (1) and a MS1-based quantification workflow (2). (1) generates a peptide FASTA databases containing already digested peptides, complemented with signal-, pro- and other specially cleaved peptides as well as peptides containing variants and mutations. This is achieved using protein graphs generated via ProtGraph, the MS2 precursors and a sophisticated traversal algorithm implemented in C++ to retrieve the peptide entries. (2) utilizes OpenMS to find features (FeatureFinderCentroided). Label-free matching across multiple measurements is done via MapAlignerTreeGuided and FeatureLinkerUnlabeled, using identified features as anchor points. The actual quantitative values for all features is then extracted via the ThermoRawFileParser XIC- Extraction. These two workflows are implemented in Nextflow, with various Python scripts used as intermediate steps. Identification in this combined workflow is done via Comet and Percolator with a q-value cut-off of 1%, using the human proteome database as plain text format, and allowing for up to 5 variants per peptide. Results: We applied this combined workflow on measured CSF samples (two groups, DDA). Our results demonstrate the benefit of using a custom tailored FASTA database from workflow (1), which includes entries typically not searched for, thereby increasing the number of identified spectra in the CSF dataset. Workflow (2) shows that charged molecules can be quantified, regardless of whether an identification or even a MS2 spectrum is present. By combining and normalizing the results of each workflow, we generated a volcano plot, containing unidentified and non-canonical entries alongside of usual canonical entries, illustrating the additional data points gained by this combined workflow. 633

634

P-II-0478 Statistical benchmark of Label-free DIA quantification using the generic integration algorithm with iSanXoT Ana Martinez-Val1,2, José Manuel Rodríguez1,2, Jesper Olsen3, Jesús Vázquez1,2 1Centro de Investigación Biomédica en Red de Enfermedades Cardiovasculares (CIBERCV), Madrid, Spain 2Centro Nacional de Investigaciones Cardiovasculares Carlos III (CNIC), Cardiovascular Proteomics Laboratory, Madrid, Spain 3Novo Nordisk Foundation Center for Protein Research, Copenhagen, Denmark iSanXoT(1) is a standalone application that processes relative abundances between MS signals and integrates them sequentially to upper levels using the Generic Integration Algorithm (GIA)(2,3), which has been benchmarked for label-based and label-free quantitative proteomics analysis using Data Dependent Acquisition (DDA). Lately, the proteomics community is pivoting towards the use of Data Independent Acquisition (DIA) for label-free quantification analysis due to its reproducibility and consistency. The performance of DIA for protein quantification depends on the quality of the MS-data and the software used for signal extraction, identification, quantification and integration of quantitative data. This is especially relevant for latest-generation mass spectrometers, such as the Thermo Orbitrap Astral, that offer unprecedented peptide and protein coverage using DIA. Therefore, determination of the quality of quantification is essential to determine which peptides actually provide meaningful information that can be used in quantitative experiments. iSanXoT allows user-defined integrations of any quantification level (i.e. features, peptides, modified peptides, proteoforms, protein groups, functional categories) while performing differential analysis among experimental conditions defined in the workflow. The GIA algorithm implemented in iSanXoT accurately models the error structure of the data at any level of quantitative information, allowing its sequential integration using error propagation theory. GIA assigns a statistical weight to all the elements, obviating the need to apply quality filters and allowing a full control of quantitative errors. Hence, iSanXoT provides an optimum environment to test the performance of DIA-based quantification. Here, we have benchmarked the integration to protein level using iSanXot versus the integration to protein performed directly in Spectronaut (MaxLFQ or Quant2.0), and DIANN (legacy or quantUMS). For that purpose, we have used mixed-species mixture datasets from previous publications and different MS-platforms: Orbitrap-Exploris(4), TripleTOF 5600(5), timsTOF Pro(6) and Orbitrap- Astral(4). Overall, this work highlights the advantages of iSanXot to process quantitative data obtained using DIA, independently on the MS-platform used. 1. Rodríguez, J. M. et al. Comput Struct Biotechnol J 23, 452–459 (2024). 2. Navarro, P. et al. J Proteome Res 13, 1234–47 (2014). 3. García-Marqués, F. et al. Mol Cell Proteomics 15, 1740–60 (2016). 4. Guzman, U. H. et al. Nat Biotechnol (2024) doi:10.1038/s41587-023-02099-7. 5. Navarro, P. et al. Nat Biotechnol 34, 1130–1136 (2016). 6. Szyrwiel, L., Gille, C., Mülleder, M., Demichev, V. & Ralser, M. Proteomics 24, (2024). 634

635

P-II-0479 SpectroPipeR – an R package for the downstream analysis of Spectronaut DIA-MS data Stephan Michalik1, Uwe Völker1 1University Medicine Greifswald, Department of Functional Genomics, Greifswald, Germany During proteome studies, researchers frequently face various challenges, some of which can be mitigated while others cannot. A prevalent issue is the bottleneck in downstream data analysis, which arises due to a limited number of bioinformaticians, rapid generation of raw data, and variations in data analysis methods or workflows. To tackle this problem, we developed an R package named SpectroPipeR. This pipeline is designed to i) simplify data analysis tasks, significantly reduce the workload for scientists, ii) be easily expandable, iii) be user-friendly even for those with minimal bioinformatics knowledge, iv) generate standardized analysis, outputs, and reports for each project, and v) produce publication-ready tables, figures, and reports. SpectroPipeR comprises a set of R functions that facilitate a comprehensive, fully automated, and standardized data analysis of Spectronaut DIA-MS data. These functions include ID rate summary, ON/OFF analysis, normalization, batch or covariate adjustment, iBAQ and maxLFQ quantification, multivariate analysis, peptide- centric statistical analysis (ROPECA or modified t-test), and interactive HTML report generation. The output is presented through a variety of clear graphs and tables in a well-structured folder system. The comprehensive standalone HTML report is extremely useful for existing Electronic Laboratory Notebooks (ELN) or Laboratory Information Management Systems (LIMS) to quickly obtain a project-specific overview. SpectroPipeR consists of a global parameter setting and four analysis modules and one reporting module that are executed sequentially. This modular approach allows flexibility where specific analyses like ID- and intensity plots can be run independently or as part of the complete pipeline. After each module execution, dynamic console feedback is provided and written to a log file to help identify errors early on. Upon completion of all modules, a comprehensive set of tables and plots categorized in different folders is generated to summarize the project from various perspectives. Beyond the analysis modules, a XIC plotting module assists the user in examining the raw data in a user-friendly manner. In conclusion, SpectroPipeR serves as a valuable tool in proteome projects by enhancing data analysis efficiency and providing researchers with detailed insights into their experimental data. Fig. 1 (optional) 635

636

P-II-0480 Systems-wide functional characterization of time-resolved protein dynamics by mass spectrometry Till Möcklinghoff1, Barbara Helm1, Marcel Schilling1, Jacques Hermes2,3,4, Marcus Rosenblatt2,3,4, Franziska Gödtel1, Yannik Dieter1, Katharina Büchner1, Jens Timmer2,3,4, Ursula Klingmüller1 1German Cancer Research Center (DKFZ), Division Systems Biology of Signal Transduction, Heidelberg, Germany 2University of Freiburg, Institute of Physics, Freiburg, Germany 3University of Freiburg, Freiburg Center for Data Analysis and Modelling (FDM), Freiburg, Germany 4University of Freiburg, Centre for Integrative Biological Signaling Studies (CIBSS), Freiburg, Germany Disease progression is governed by dynamic protein abundance changes altering molecular and phenotypic state of the cell. The usually acquired snapshot measurements are prone to random effect errors and might miss the most informative timepoint in the specific context. Contrary, time-resolved data acquisition is immensely helpful for the identification of genuinely altered abundance changes and these dynamics contain valuable mechanistic information. However, there is a lack of easy-to-use bioinformatic methods to apply on proteomic datasets identifying proteins that are differentially regulated over time and subsequently characterizing their dynamics. To generate a generally applicable approach, we investigated the effects induced by a key modulator of innate and adaptive immune responses, interferon alpha 2a (IFNα2a), on hepatoma cells in context of viral infections and cancer. To identify and characterize protein dynamics at a global scale, we stimulated HepG2hNTCP cells with multiple doses of IFNα2a and acquired data across nine timepoints by DIA proteomics. A novel and unbiased data analysis method was applied identifying the induction of 74 proteins, including known IFNα-induced proteins, but also novel candidates, and the repression of 17 proteins, e.g. a negative regulator of the mTOR pathway. To characterize the identified differential protein dynamics, a retarded transient function was fitted to the experimental data enabling to retrieve information regarding the time delay until induction, the timepoint of maximal induction and the extend of transient behavior. The trajectories of protein dynamics follow a similar sequential induction as previously observed in the transcription factor dynamics obtained from predictions of a mathematical model on IFNα2a signal transduction. The curve characteristics were compared to dynamic profiles previously acquired by RT-qPCR. Overall, the distinct time clusters were confirmed at the protein level, however, a delayed maximal induction was observed compared to the mRNA profiles. Surprisingly, different from the mRNA profiles, for a large proportion of proteins the abundance was maintained after reaching maximal induction and only a few proteins, including the DNA damage response protein PARP10, showed a transient dynamic behavior. We demonstrated a novel approach to analyze time-resolved proteome data as a proof-of-concept on IFNα- induced protein dynamics which now allows for application on other datasets. 636

637

P-II-0481 Enhancing peptide-spectrum match identification: leveraging mokapot with nonlinear machine learning algorithms to increase PSMs, with consideration for false positives Alireza Nameni1, Robbin Bouwmeester1, Lennart Martens1, Sven Degroeve1 1Ghent University, Center for Medical Biotechnology, Department of Biomolecular Medicine, Ghent, Belgium In MS-based proteomics, search engines are used to match peptides to observed tandem MS spectra. Liquid chromatography coupled with mass spectrometry (LC-MS) generates complex data requiring specialized tools for interpretation. These tools calculate the peptide-spectrum-match (PSM) score, estimating the likelihood of a peptide being the source of an observed spectrum. Machine learning rescoring methods combine several scoring metrics into a single PSM score to differentiate true target PSMs from decoys (false matches). While more complex models like gradient tree boosting (XGBoost) increase PSM identification rates, they also risk overfitting the target proteome database, resulting in overly optimistic false discovery rate (FDR) statistics. To investigate the potential overfitting of complex models, we used an entrapment sequence database approach. This method expands the target database with peptides from nine randomly shuffled versions of the target proteome to attract the majority of random (incorrect) target PSMs. This ensures that the majority of PSMs aligning with the true target database are indeed correct. We also implemented a feature subset selection algorithm to assess individual feature contributions to the overfitting of the rescoring model. Different models were evaluated using spectrum files from eleven LC-MS experiments of varying sizes. We performed searches with four different search engines: Andromeda, Comet, MSGF, and MSGF against the human UniProt database. Rescoring was conducted using five different algorithms: Logistic Regression, multi- layer perceptron, linear support vector machine (LSVM), XGBoost models, and Random Forest that increase in their complexity of fitting the rescoring function. Increasing model complexity enhances PSM rescoring sensitivity, resulting in an average identification increase of about 11%. However, this sensitivity comes at the cost of higher entrapment FDR, with an average increase of about 500%. This suggests that complex models introduce a bias towards the target proteome database, leading to overly optimistic FDR estimates. While complex models improve PSM identification, they also increase bias towards (including random wrong matches) from the target proteome database, leading to overly optimistic FDR estimates. Our study underscores the need for careful evaluation of model complexity and feature selection to mitigate overfitting in PSM rescoring. Fig. 1 (optional) 637

638

P-II-0482 Direct spectrum-centric analysis of diaPASEF data using diaTracer and FragPipe computational platform Kai Li1, Guo Ci Teo1, Fengchao Yu1, Alexey Nesvizhskii1 1University of Michigan, Ann Arbor, MI, United States Data-independent acquisition (DIA) has become a widely used strategy for peptide and protein quantification in mass spectrometry-based proteomics studies. The integration of ion mobility separation into DIA analysis, such as the diaPASEF technology available on Bruker"s timsTOF platform, further improves the quantification accuracy and protein depth achievable using DIA. However, the complex data presents new challenges for computational analyses. Creating the spectra library built by additional data-dependent acquisition (DDA) experiments requires additional effort. And the in-silico predicted spectral library is limited in the post- translational modification and no-specific searches. An alternative computational strategy, first proposed by us and implemented in the DIA-Umpire algorithm, generates DDA-like pseudo-MS/MS spectra by detecting extracted ion chromatogram (XIC) features directly from DIA data. However, DIA-Umpire cannot process diaPASEF data due to the challenges of detecting XIC features with the additional ion mobility dimension. To address the need for an efficient and versatile method for analyzing diaPASEF data, we developed diaTracer, a novel direct spectrum-centric tool. diaTracer has been designed to perform highly efficient processing of three- dimensional information (m/z, retention time, ion mobility) to detect precursor and fragment ion features, followed by the generation of pseudo-MS/MS spectra. It is fully integrated into the widely used FragPipe computational platform for direct analysis of diaPASEF data. We demonstrate the performance of diaTracer using diaPASEF data from cerebrospinal fluid (CSF) and plasma proteome datasets, as well as data from phosphoproteomics, immunopeptidomics, and low-input spatial proteomics studies. We also show that our strategy enables unrestricted PTM searches by applying the open/mass offset mode of MSFragger to diaTracer- extracted files. Overall, our preliminary analysis shows that diaTracer, coulpled with FragPipe computational platform, provides a one-stop, fast, and sensitive solution for analyzing of diaPASEF data from a wide range of biological applications. 638

639

P-II-0483 Artificial Intelligence in mass spectrometry: fascinating past, exciting present and promising future Magnus Palmblad1 1Leiden University Medical Center, Center for Proteomics and Metabolomics, Leiden, Netherlands The histories of computing, artificial intelligence (AI) and mass spectrometry are intertwined, going back at least to the work of Sibyl Rock at Consolidated Engineering Corporation (CEC) compiling the first computing manual for mass spectrometry 1946, and together with Clifford Berry developing the first commercially available electronic computer. This work was driven by the need to solve problems in mass spectrometry. Another intersection is the development of an early expert system, Dendral, by Joshua Lederberg and colleagues at Stanford, for identifying chemical structures from mass spectra, representing a major milestone not only in mass spectrometry, but also in the development of AI. In recent decades, machine learning, in particular deep neural networks trained on thousands or millions of spectra, has significantly improved our ability to identify peptides and small molecules by predicting their observability, chromatographic behavior and fragmentation in tandem mass spectrometry. This presentation will highlight some of these recent developments, including our own contributions. The integration of machine learned models can already be used to generate realistic synthetic data. This has many potentially beneficial applications in system suitability testing, experimental design and method optimization, and providing ground truth for algorithm and computational workflow benchmarking. It also has some significant risks, which must be addressed. Large language models are now also able to learn from millions of documents describing mass spectrometers and mass spectrometry data may provide further assistance to human experts in data interpretation, resuming the pioneering work of the DCRT/CIS conversational AI mass spectrometry search developed at NIH in the early 1970s. Finally, some recent work on using generative AI to explore data analysis workflows for mass spectrometry will be presented. By building and running ensembles of workflows rather than one particular workflow, it may be possible to increase the confidence in biological interpretations from the proteomics data, as well as learn something about the combined tools themselves. 639

640

P-II-0484 Optimization of peptide selection for absolute protein quantification using QconCATs Jorge Peinado-Izaguerri1, Robert Crawford1, Blanca Navarrete1, Mark Ashe1, Chris Grant1, Simon Hubbard2, Graham Pavitt1 1University of Manchester, Division of Molecular and Cellular Function, Manchester, United Kingdom 2University of Manchester, Division of Evolution, Infection and Genomics, Manchester, United Kingdom Absolute protein quantification is one of the most relevant applications of LC-MS based proteomics. Absolute quantification studies offer the opportunity to unravel complex biological incognitas such as the stoichiometric relation of proteins in subcellular complexes, but also presents higher costs and complexity than relative quantification studies. As a result, development of experimental and bioinformatic solutions that can help simplify absolute protein quantification are of great importance. Among available strategies, the use of heavy- isotope labelled internal standards, such as labelled peptides or full-length proteins, has proven to be one of the most accurate approaches to determine protein concentration and stoichiometric relations. QconCATs (quantification concatamers) are synthetic proteins constituted by a series of concatenated tryptic peptides which can act as heavy-isotope standards for the absolute quantification of multiple proteins in a complex biological sample using a single LC-MS run. Optimal selection of peptides to be included in QconCATs is of utmost importance to enable accurate protein quantification as ideally they are both proteotypic (routinely detected by MS experiments where the parent protein is present) and quantotypic (showing a good correlation between their measured signal intensity and that of the parent protein). Phenomena such as peptide modifications, whether is derived from post translational modifications or sample preparation, or presence of miscleaved species enhanced by conflictive flanking regions can degrade quantification accuracy. Recently, a software tool "AlacatDesigner" was developed to score candidate peptides using algorithms and publicly available data. In this study we selected peptide candidates to be included in QconCATs for the quantification of proteins in translation initiation complexes within whole cell lysates making use of a range of theoretical criteria as well as empiric data that allowed the assessment of candidate peptide quantotypicness. We compared the selected candidates with the top-scoring peptides identified by AlacatDesigner. AlacatDesigner proved to be a useful tool in aiding candidate selection but its top-scoring peptides were not always optimal choices when additional factors were considered such as likelihood of post translational modifications and other empirical evidence. Our analysis demonstrates the importance of using data acquired under equivalent experimental conditions that will be used for quantification experiments when selecting ideal peptide candidates. Information from this study will be used to refine the AlacatDesigner peptide scoring algorithm. 640

641

P-II-0485 Aiproteomics: a Python package to facilitate the use of deep learning for DIA-MS-based phosphoproteomics data processing Thang V. Pham1, Robin Richardson2, Alex Henneman1, Connie Jimenez1 1Amsterdam University Medical Center, location Vrije Universiteit Amsterdam, OncoProteomics Laboratory, Medical Oncology, Amsterdam, Netherlands 2Netherlands eScience Center, Amsterdam, Netherlands Introduction: A mass spectral library is typically required for analysis of data-independent acquisition mass spectrometry (DIA-MS) data. It has been shown that a deep neural network can learn from experimental data to predict all essential components of a spectral library including mass spectra, retention time, and ion mobility from input peptides sequences. However, the use of different deep learning models makes it hard to use in practice. Hence, we aim to develop a software tool that supports all components of a spectral library. Methods: We use a transformer-based deep learning model for all prediction tasks. The transformer model enables parallel processing with high accuracy and has demonstrated state-of-the-art performance for a wide range of problems, including retention time prediction of phosphopeptides (Pham et al. Proteomics 2023 Apr;23(7-8):e2200041). We develop a spectral library generation module to produce a spectral library that can be readily used by the DIA-MS processing tool DIA-NN (Demichev et al. Nat Methods. 2020 Jan;17(1):41-44). open-source Python package The Results: at https://github.com/aiproteomics/aiproteomics, is being developed with a view to sustainability through application of best practices for research software. It is easily installable using the standard Python package manager, pip. Tutorial notebooks are provided for each task to provide a quick start for users. The output spectral library can be directly used by DIA-NN for DIA-MS data processing. Finally, we are benchmarking the accuracy of the transformer models against state-of-the-art methods for DIA-MS-based phosphoproteomics. aiproteomics, available Conclusions: The open-source Python package aiproteomics enables spectral library generation for DIA-MS data analysis using state-of-the-art transformer-based deep learning models. We also provide a guideline for contribution from the community. 641

642

P-II-0486 CURTAIN – a unique web-based tool for exploration and sharing of MS-based proteomics data Toan Phung1,2, Kerryn Berndsen1,2, Tran L. C. H. B. Phan1, Miratul M. K. Muqit1,2, Dario R. Alessi1,2, Raja S. Nirujogi1,2, Rosamund Shastry1,2 1University of Dundee, MRC-PPU, Dundee, United Kingdom 2Aligning Science Across Parkinson's Collaborative Research Network, Chevy Chase, MD, United States To facilitate analysis and sharing of mass spectrometry (MS)-based proteomics data, we created online tools called CURTAIN (https://curtain.proteo.info) and CURTAIN-PTM (https://curtainptm.proteo.info) with an accompanying series of video tutorials (https://www.youtube.com/@CURTAIN-me6hl). These are designed to enable non-MS experts to interactively peruse volcano plots and deconvolute primary experimental data so that replicates can be visualized in bar charts or violin plots and exported in publication-ready format. They also allow assessment of overall experimental quality by correlation matrix and profile plot analysis. After making a selection of protein "hits", the user can analyze known domain structure, AlphaFold predicted structure, reported interactors, relative expression as well as disease links. CURTAIN-PTM permits analysis of all identified PTM sites on protein(s) of interest with selected databases. CURTAIN-PTM also links with the Kinase Library to predict upstream kinases that may phosphorylate sites of interest. We provide examples of the utility of CURTAIN and CURTAIN-PTM in analyzing how targeted degradation of the PPM1H Rab phosphatase that counteracts the Parkinson"s LRRK2 kinase impacts cellular protein levels and phosphorylation sites. We also reanalyzed a ubiquitylation dataset, characterizing the PINK1-Parkin pathway activation in primary neurons, revealing data of interest not highlighted previously. CURTAIN and CURTAIN-PTM are free to use and open source, enabling researchers to share and maximize the impact of their proteomics data. We advocate that MS data published in volcano plot format be reported containing a shareable CURTAIN weblink, thereby allowing readers to better analyze and exploit the data. Fig. 1 (optional) 642

643

P-II-0487 DIGging into the DIGestion Jasmína Mária Portašiková1,2, Zuzana Kalaninova1,2, Daniel Kavan1, Petr Novák1, Petr Man1 1Institute of Microbiology, Czech Academy of Sciences, BioCeV, Prague, Czech Republic 2Charles University, Faculty of Science, Prague, Czech Republic Protein digestion is an essential tool in both classical bottom-up proteomics and structural proteomics approaches such as chemical cross-linking, radical labelling or hydrogen/deuterium exchange. While proteomics uses specific proteases with predictable digestion patterns and minimal missed cleavages, HDX relies on non- specific or semi-specific acidic proteases. These are highly promiscuous in terms of amino acid preferences and tend to generate a lot of overlaps. Although this hampers predictability and adds complexity, it also has several advantages, such as a higher number of experimental data points and increased spatial resolution. As more proteolytic columns become available, the need for rigorous methods to compare digestion results increases. To meet these needs, we have developed a Java-based software tool that takes the exported raw search results (e.g. csv from MASCOT) and the corresponding database and performs an instant extraction of the key digestion parameters. This tool, called DigDig, provides digestion metrics, reproducibility assessment, coverage maps, extraction of cleavage preferences up to the N-th position, peptide length distribution, missed cleavage analysis and data export for IceLogo. It works with single proteins or complex mixtures, where it can plot data for either selected protein(s) or the entire identified ensemble. We will show how this software has helped us to identify novel cleavage preferences, optimise the HDX-MS workflow, but also its applicability in the evaluation of specific proteases, the key tool of the proteomic pipeline. Financial support from the MEYS/EU OP JAK project PHOTOMACHINES - CZ.02.01.01/00/22_008/0004624 is gratefully acknowledged. 643

644

P-II-0488 An infinite well: harmonizing all public proteomics data for machine learning Tobias Schmidt1, Siegfried Gessulat1, Michael Graber1, Vishal Sukumar1, Markus Schneider1, Alexander Hogrebe1, Samia Ben Fredj1, Daniel Zolg1, Martin Frejno1 1MSAID, Garching, Germany Background: Synthetic proteomics datasets, like ProteomeTools, are invaluable for machine learning but finite, especially regarding spectra from post-translational modifications. In contrast, public repositories abound with data, yet harmonizing them poses challenges due to diverse measurement environments, methods, software, and calibrations, lacking shared QC peptides. To address this, we developed a distributed, standardized, and cost-effective workflow for searching and harmonizing public data with minimal manual intervention. Processing over 3000 files containing ubiquitinated, phosphorylated, and acetylated peptides, we used a novel machine learning-based calibration. This resulted in ~1 million modified peptides from >33 million high-quality PSMs, complements synthetic public data such as ProteomeTools. Methods: The workflow takes a PRIDE or MASSIVE identifier and a FASTA file as input, producing a standardized PSM table with optional collision energy calibration and fragment ion annotation. Results are stored as parquet files on S3 and executed on a 30-node Kubernetes cluster. Utilizing ppx for spectrum file downloading, msconvert for transformation, and Sage for search, along with custom algorithms for calibration and annotation, we achieve automation. A web interface displays QC metrics, aiding project selection. Collision energy is calibrated by generating curves to determine offsets, and retention times are normalized for transfer learning. Results: The goal is to unlock publicly available proteomics data for machine learning, costing <$0.10 per spectrum file and taking 36 seconds on average, processing 100 files per hour. Unique features include model- based collision energy calibration, yielding high correlations (>0.85) compared to QC-based methods. Processing 60 PRIDE projects yielded >400,000 unique peptides and >10 million PSMs, similar results from the Atlantic NCI60 dataset and ProteomeTools pools. We processed 60 PRIDE projects with >1000 files containing ubiquitinated peptides from human samples, yielding >400,000 unique peptides and >10 million PSMs. Processing additional species is expected to increase this yield by ~75%. In addition, we processed the Atlantic NCI60 dataset containing phosphorylated peptides from ~500 files and identified >200,000 unique peptides and ~6 million PSMs. Similarly, searching two ProteomeTools pools for N-terminally acetylated peptides (an FMOC- synthesis artifact) identified ~350,000 unique peptides and ~16 million PSMs. Using the harmonized dataset, we achieved high correlations (>0.85) on modified peptides, surpassing non-calibrated training. In summary, our fast, cost-effective workflow produces a high-quality dataset, complementing synthetic datasets for training peptide fragmentation prediction models. Conclusion: Hands-off, inexpensive workflow to search, annotate and calibrate arbitrary public proteomics datasets and integrate them into harmonized machine learning datasets. 644

645

P-II-0489 AI-powered Onco-Knowledge Graph (OncoKG) for Precision Medicine Dibyendu Seal1, Yongjie Wang1, Andreas Kulozik1, Ashok Jayavelu1 1German Cancer Research Center (DKFZ), A400, Heidelberg, Germany Clinical knowledge graphs have profound usage in biomedical research due to their ability to integrate and structure enormous amount of heterogenous biological information scattered across various databases. These graphs can serve as prime resources for querying gene-gene, gene-disease, drug-disease and drug-target associations as well as kinase-substrate interactions. Drug discovery being expensive and time-consuming, these knowledge bases can be leveraged to identify novel interactions that enhance decision-making in drug repurposing and precision medicine. However, to make maximum utilization of such a knowledge repository, queries related to drugs or drug targets need to be precise and focused. Consequently, knowledge graphs can be utilized to provide evidence-based recommendations for tailoring treatment in patients based on their unique molecular profile. In oncology, standard frontline chemotherapy is often not curative, and the disease frequently relapses. Relapse, attributed to drug resistance and refractoriness, severely limits the patient's chances of being cured. Patients often exhaust therapy options, leaving fewer effective therapies available. Despite the increasing availability of next generation sequencing and omics information for these patients, identifying actionable and highly effective targets remain challenging. This underscores the unmet need of a model capable of identifying molecular targets and rank potent drugs in order of their efficacy for a more effective personalized treatment. To address this gap, we introduce an AI-powered framework with active learning designed to predict drugs for precision medicine based on patients" omics profiles. This framework includes an Onco-Knowledge Graph (OncoKG) and an AI module to identify novel drug targets from multi-omics data along with a Large Language Model (LLM)-enabled interface to query the OncoKG. To assess the tool, we utilized publicly available multi- omics datasets of Pancreatic Ductal Carcinoma (PDAC) from Clinical Proteomic Tumor Analysis Consortium (CPTAC) and Acute Myeloid Leukemia (AML). Preliminary results validate targets identified for PDAC and AML to be clinically relevant. Importantly, drugs suggested by the AI-OncoKG framework show promise and warrant further investigation invitro and invivo. With further pre-clinical examinations, this tool has the potential to revolutionize personalized treatment with an outlook to facilitate drug choices and repurposing. 645

646

P-II-0490 Optimizing search engines to detect marginal peptide matches using large-scale matched-matrix truth datasets Brian Searle1,2 1The Ohio State University Medical Center, Biomedical Informatics, Columbus, OH, United States 2The Ohio State University Comprehensive Cancer Center, Pelotonia Institute for Immuno-Oncology, Columbus, OH, United States Detecting peptides with data-independent acquisition (DIA) analysis relies on search engines that identify signals associated with those peptides. Search engine algorithms have been developed using training datasets typically built using existing computer annotations, which can reinforce bias from older-generation search engines. One of the goals of new search engines is to correctly detect more peptides from marginal data, but these matches are also the most likely to include errors. Additionally, poor fragmentation or low signal peptide matches can also be the most challenging to curate manually. While large-scale peptide libraries such as ProteomeTools eliminate computational bias because they are built off of synthesized peptides, they also include biases from peptide selection and synthesis limitations. We developed an approach to generate large-scale DIA "truth sets" for training search engines using matrix- matched calibration curves. Peptides in undiluted points in the calibration curve produce the highest signal and are the easiest for both existing computer engines and scientists to interpret. As those peptides are diluted into a matrix-matched background, increasing interference and decreasing signal make matches more challenging to validate. However, because the matrix stays constant, retention times and fragmentation patterns remain the same, allowing us to use accurate mass and retention time tags to confirm the signal of those peptides without needing detections from existing search engines. To test this approach, we used a DIA-based matched-matrix curve generated with a linear ion trap (LIT). LITs are affordable and robust mass analyzers that provide fast scanning speeds with high sensitivity, where their only significant disadvantage is inferior mass accuracy compared to time-of-flight or Orbitrap mass analyzers. However, the decreased mass accuracy of LITs means that low signal-to-noise detections are challenging to measure confidently. We developed a rapid manual validation tool to confirm tens of thousands of detections of the undiluted sample and then used the dilution curve to provide hundreds of thousands of observations with lower signal-to-noise. We further demonstrate the benefits of new training sets on improving DIA scoring functions in EncyclopeDIA. 646

647

P-II-0491 ProteomicsDB: connecting proteomes across species Armin Soleymaniniya1,2, Miriam Abele3,4, Sarah Brajkovic2,4, Christina Ludwig2,3, Bernhard Küster2,3,4, Mathias Wilhelm1,2 1Technical University of Munich, Computational Mass Spectrometry, Freising, Germany 2Technical University of Munich, Elite Network of Bavaria, Freising, Germany 3Technical University of Munich, Bavarian Center for Biomolecular Mass Spectrometry (BayBioMS), Freising, Germany 4Technical University of Munich, Chair of Proteomics and Bioanalytics, Freising, Germany Introduction: ProteomicsDB has evolved over the years into a multi-omics and multi-organism resource for life sciences research, encompassing more than 198 projects totaling over 29k LC-MS/MS experiments and providing access to all this data with FAIR standards. The latest developments focus on cross-species analysis at the protein level. By leveraging the principle that proteins with common ancestry tend to exhibit resemblances despite evolutionary divergence, homology analysis empowers researchers to glean insights into the functions of uncharted proteins in lesser-known organisms. In addition to empowering the transfer of knowledge from well- studied organisms to less studied ones, this approach can provide insights into phylogenetical lineage-enriched traits of organisms. Methods: To increase the number of covered organisms, we integrated the proteomics baseline data of 305 bacterial species into ProteomicsDB. The data was searched with MaxQuant (100% FDR) and rescored with Prosit via the Oktoberfest pipeline. The false discovery rate was controlled at 1% by the picked-group FDR algorithm. Label-free quantification was done using the iBAQ method to estimate protein abundances. The homology inference was done with OrthoFinder, which assigns proteins to "orthogroups" that reveal the paralogy/orthology relations among them. The functional annotation of protein sequences was analyzed using InterPro. Results: We integrated data from the "Proteomics landscape across the bacterial kingdom of life", an atlas encompassing the baseline proteome of 305 bacterial species. This is the most comprehensive collection of prokaryote draft proteomes to this date, identifying 14,140,252 PSMs and 7,005,416 peptides (1% FDR). Furthermore, we substantially improved the PSM identification by 11.89% and peptides by 9.69% (on average) by rescoring the search results with Prosit. This data covers, on average, 50% of the theoretical proteome of studied species. Homology analysis with OrthoFinder on the theoretical proteomes assigned a total of 1,131,620 (96.9% of overall) proteins to 34,979 orthogroups with a median size of four proteins. Among these, only 99 orthogroups cover all of the studied organisms, and 26 consist of only one protein per species. The inferred orthogroups were analyzed to detect possible enrichments on specific phylogenetic levels, which can be a proxy for studying the lineage-specific functions of bacteria. Furthermore, the retrieved GO terms via the functional annotation analysis were enriched at the orthogroup level to provide insights into the functions of protein groups. We developed a new interactive visualization that enables users to compare the expression of proteins within one orthogroup across participating species in a phylogenetic-aware fashion. Also, we implemented a histogram visualization of protein expressions of the corresponding species level to replace the anatomograms on the protein expression page for single-cell organisms. 647

648

P-II-0493 A CCS-centric HLA-specific trained de novo module for precise and accurate real-time immunopeptide identification on the Bruker ProteoScape platform Rui Zhang1, Qixin Liu1, Tharan Srikumar2, George Rosenberger3, Jonathan Krieger2, Dennis Trede4, Bin Ma1, Angela Paul5 1Rapid Novor Inc., Kitchener, Canada 2Bruker Ltd, Milton, Canada 3Bruker Switzerland AG, Faellanden, Switzerland 4Bruker Daltonics GmbH & Co. KG, Bremen, Germany 5Bruker UK Ltd, Coventry, United Kingdom Since its conception, Bruker ProteoScape (BPS) has transformed into a comprehensive proteomics data analysis platform that can integrate third-party tools while utilizing the concept of data streaming to realize fully customizable real-time processing workflows. To expand the capabilities of the BPS platform for immunopeptidomics, and other applications, we previously developed and integrated a timsTOF optimized de novo sequencing engine from Rapid Novor Inc., called BPS Novor. To further develop this tool and support the extremely rapidly growing field of timsTOF based immunopeptidomics, we have retrained this module with >1.4 million spectra mapping to >150,000 HLA-presented peptides from MHCI and MHC II classes. Methods and Results: Novor was re-trained on a variety of timsTOF acquired data, where ground truth is taken from the ProLuCID database search results filtered to 1% PSM FDR. Previously we have shown that BPS Novor is highly accurate and precise on a variety of datasets. On amino acid level, at 75% precision, BPS Novor achieved between 40-60%, whereas standard Novor achieved between 25-50%. We have also shown that BPS Novor is extremely fast by evaluating the processing speed across 5 datasets, with an average processing speed of 1338±226 spectra/second. Here we compare the re-trained BPS Novor (MHC model) across two immunopeptidomics datasets using standardized numbers of computing cores. We focused on two recently published immunopeptidomics datasets (Feola et al., 2021 and Phulphagar et. al., 2023) to show the newly optimized BPS Novor MHC model increases precision and accuracy over the previous versions for immunopeptidomics applications, while retaining the processing speed advantage over other algorithms. We show in the Feola et al., dataset modest improvement of 2% in amino acid correctness and 1% at the peptide correctness, with a marginal increase in processing time. For the 10 melanoma derived samples from Phulphagar et. al. A ~6% increase in amino acid and peptide correctness was observed. The MHC model showed on average 8.5% increase in the number of 8-11mer peptide sequences in the dataset. Together, these datasets confirm that the MHC optimized scoring model for BPS Novor provides greater accuracy for immunopeptidomics data. Conclusion: We here introduce a purposefully optimized BPS Novor module, providing on-the-fly real-time de novo sequencing for timsTOF immunopeptidomics data. Conflict of Interest Disclosure: R.Z., Q.L., M.X., B.M. are employees of Rapid Novor, Inc. Q.L., M.X., B.M. are co- founders of Rapid Novor, Inc. D.T., T.S., J.K., G.R. are employees of subsidiaries of Bruker Corp. Novor is a product of Rapid Novor, Inc. BPS Novor is a product of Rapid Novor, Inc. sold and distributed by subsidiaries of Bruker Corp. 648

649

P-II-0494 The benefit of in silico predicted spectral libraries in data-independent acquisition data analysis workflows An Staes1,2,3, Teresa Maia1,2,3, Sara Dufour1,2,3, Robbin Bouwmeester1,3, Ralf Gabriels1,3, Lennart Martens1,3, Kris Gevaert1,3, Francis Impens1,2,3, Simon Devos1,2,3 1Ghent University, Center for Medical Biotechnology (CMB), Zwijnaarde, Belgium 2Ghent University, Proteomics Core, Zwijnaarde, Belgium 3Ghent University, Center for Medical Biotechnology, Department of Biomolecular Medicine, Zwijnaarde, Belgium Data-independent acquisition (DIA) has become a well-established method for MS-based proteomics. However, the list of options to analyze this type of data is quite extensive, and the use of spectral libraries has become an important factor in DIA data analysis. More specifically the use of in silico predicted libraries is gaining more interest. By working with a differential spike-in of human standard proteins (UPS2) in a constant yeast tryptic digest background, we evaluated the sensitivity, precision and accuracy of the use of in silico predicted libraries in DIA data analysis workflows compared to more established workflows. Three commonly used DIA software tools, DIA-NN, EncyclopeDIA and SpectronautTM, were each tested in spectral library mode and spectral library- free mode. In spectral library mode, we used independent spectral library prediction tools PROSIT and MS2PIP together with DeepLC, next to classical data-dependent acquisition (DDA)-based spectral libraries. In total, we benchmarked twelve computational workflows for DIA and one for DDA. Our comparison showed that DIA-NN reached the highest sensitivity while maintaining a good compromise on the reproducibility and accuracy levels in either library-free mode or using in silico predicted libraries pointing to a general benefit in using in silico predicted libraries. Fig. 1 (optional) 649

650

P-II-0495 TimSim – a simulation framework for PASEF-like timsTOF proteomics mass spectrometry raw data David Teschner1, David Gomez-Zepeda2,3, Mateusz Krzysztof Łacki4, Tim Maier1, Stefan Tenzer2,4,3, Andreas Hildebrandt1 1Johannes Gutenberg-Universität (JGU) Mainz, Institute of Computer Science, Mainz, Germany 2Helmholtz Institute for Translational Oncology Mainz (HI-TRON Mainz), Mainz, Germany 3German Cancer Research Center (DKFZ), Heidelberg, Germany 4University Medical Center of the Johannes Gutenberg-University Mainz, Institute of Immunology, Mainz, Germany Mass spectrometry continues to advance as a crucial technology for analyzing and quantifying the chemical composition of biological samples. Data-Independent Acquisition (DIA) techniques have become key for handling complex data from large-scale studies. The timsTOF platform has gained prominence in the field of DIA, utilizing ion-mobility separation to innovative acquisition strategies collectively known as Parallel Accumulation-Serial Fragmentation (PASEF). implement The latest PASEF advancements offer expanded experimental possibilities but also necessitate the optimization of sample acquisition parameters by researchers. Additionally, these advancements call for the development of new data processing algorithms to fully leverage the rich information contained in the resulting raw data. To address these challenges, we present TimSim, a simulation framework for generating in-silico proteomics data specific to timsTOF devices. TimSim reproduces the complete spectrum of raw proteomic mass spectrometry data generated by timsTOF instruments, allowing for thorough investigation of how different acquisition methods impact dataset information content. Furthermore, TimSim supports the generation of fully labeled data, down to individual peak annotations, providing a resource that complements wet-lab experimental data for the rapid development and evaluation of new data processing algorithms. We leverage a two-language approach, with lower-level Rust code implementing computationally demanding tasks such as frame generation but provide the tool with a user-friendly Python interface. Built with extensibility in mind, TimSim also enables the integration of emerging machine and deep learning models, aiding continued innovation in mass spectrometry data analysis. 650

651

P-II-0497 SWAPS: a joint peptide identification and quantification approach to increase the depth and coverage of LC-MS experiments Zixuan Xiao1, Mathias Wilhelm1 1Technical University of Munich, Professur für Computational Mass Spectrometry, Freising, Germany Liquid Chromatography coupled to tandem mass spectrometry (LC-MS/MS) is widely employed in proteomics studies. Most commonly, the tandem mass spectra (MS2) serve for identification, while MS1 is used for quantification. However, relying solely on MS2 introduces biases and results in incomplete coverage of the available signal in MS1. In addition, MS2 scans make up around 90% of all acquired scans, equivalent to ~75% of measurement time, however, yielding redundant information in multi-run experiments. To optimize the MS1 data utilization rate, we introduce Scan-Wise Activation and Peak Selection (SWAPS) for MS1-based joint identification and quantification of peptides, with the aim of reducing the dependency on MS2 scans, freeing up measurement time and increasing throughput. In SWAPS, MS1 signals are de-convoluted in a scan-by-scan manner. For each MS1 scan, we construct a dictionary of potential observable peptides from either an in silico digestion or prior deep proteome measurements, described by their accurate mass, isotope pattern, predicted retention time (RT), charge state, and ion mobility (IM). The joint identification and quantification are formulated as a sparse coding problem, i.e. finding a sparse representation of the spectrum in the form of a linear combination of dictionary peptides, termed scan-wise activation. Scan-wise activations are concatenated for each peptide to form activation curves, which represent the inferred elution profile of a peptide. As a post-processing step, a Convolution Neural Network (CNN) is trained to detect and classify peaks in an activation curve in case multiple activation peaks are generated by the sparse coding. Our results reveal that with accurate RT from single-shot experiments, the correlation between quantification outcomes from SWAPS (solely utilizing MS1 data) and MaxQuant (utilizing both MS1 and MS2 data) achieves a Pearson correlation coefficient of 0.95. Moreover, SWAPS successfully identified and quantified 82% of all peptide sequences detected in a corresponding deep proteome measurement, compared to only 28% identified in the same data when using MaxQuant. With predicted RT, SWAPS tends to recover multiple activation peaks for each peptide, leading to overestimation in peptide quantification. However, with the CNN-based peak selection model, the true peak can be identified for 85% of the peptides, resulting in a final Pearson correlation of 0.78. We are currently working towards further improvement on peak selection accuracy by integrating ion mobility, as well as charge distribution and ionization efficiency. In conclusion, we highlight the feasibility and potential of leveraging MS1 information with SWAPS, demonstrating the potential of achieving higher throughput and higher measurement depth without additional experimental effort, enhancing efficiency in proteomics analyses. 651

652

P-II-0498 Benchmarking deep learning models and classical de novo sequencing tools for immunopeptidomics Liang Xue1, Mykola Bordyuh1, Djork-Arné Clevert2, Robert Stanton1 1Pfizer, Pfizer Research and Development, Cambridge, MA, United States 2Pfizer, Pfizer Research and Development, Berlin, Germany The identification of neo-peptide antigens is essential for the development of immunotherapies like chimeric antigen receptor T-cell therapy (CART), mRNA, and peptide vaccine. While mass spectrometry-based peptidomics is powerful for identifying peptides, its utility in discovery of non-canonical antigens is yet to achieve the full potential. Some challenges of such application include the constraints of the matching the resulting spectra against reference databases which may not represent the neo-peptide sequence, lower statistical power with exploded searching space, as well as low throughput due to computation power requirement. Inspired by fast growing deep learning methods in various industry, we developed a general-purpose language model learning a joint representation of peptides and spectrum, and further applied such method to peptide identification for immunopeptidomics. Furthermore, the model was benchmarked with other machine learning models as well as classical de novo sequencing softwares. Our general-purpose model demonstrated comparable results on neo-peptide identification against classical analytical methods and other machine learning based model, while offering speed, lower dependencies on heuristic decision, and the versatility towards downstream tasks. 652

653

P-II-0499 Benchmarking predictions models for improved peptide identification with MSBooster and Koina Kevin Li Yang1, Ludwig Lautenbacher2, Fengchao Yu1, Mathias Wilhelm2, Alexey Nesvizhskii1 1University of Michigan, Ann Arbor, MI, United States 2Technical University of Munich, Computational Mass Spectrometry, Freising, Germany Peptide-spectrum-match (PSM) rescoring in bottom-up proteomics synthesizes multiple characteristics per PSM to better separate true positive target PSMs from decoy and false positive target PSMs. Multiple models (e.g. DIA-NN, Prosit, AlphaPeptDeep, MS2PIP/DeepLC) have been trained to predict MS/MS spectra and retention time (RT) and have been leveraged to improve PSM rescoring by comparing predicted to experimental values, leading to increased peptide identifications. It has been difficult to benchmark these prediction models, as they are usually run within their individual analysis pipelines, so the number of peptide identifications is influenced by factors external to the prediction model. To enable fair comparisons of the models while keeping all other factors constant, we have integrated the Koina prediction web server into MSBooster as part of the FragPipe computational pipeline, giving us access to a growing list of models designed by the proteomics community. We provide an easy way to choose/mix-and-match models. Furthermore, we benchmark the models on various data types, namely phosphoproteomics, immunopeptidomics, DIA data from the Thermo Scientific Astral mass spectrometer, and TMT11 data. We find that DIA-NN and the Koina models each have certain data types they perform best on, with Koina models improving peptide identifications by as much as 6% over the FragPipe default DIA-NN. To decrease the burden that comes with having to find the optimal models for one"s dataset out of a growing list of available Koina models, we propose a heuristic best model search that predicts which MS2 and RT models will perform best on the data at hand. Our approach selects RT models that on average identifies 90% of the peptides found by the empirically best RT model; for MS2 models, our heuristic approach"s performance rises to 98%. When designing the best model search algorithm, we realized that prediction models achieving the highest median MS/MS similarity with experimental spectra or lowest median RT differences does not always translate to that model getting the most peptide identifications. Koina is integrated into FragPipe, allowing users to visualize spectra from Koina models in FragPipe-PDV mirror plots. Various quality control plots are also made available. MSBooster code is freely available at https://github.com/Nesvilab/MSBooster. Fig. 1 (optional) 653

654

P-II-0500 Enhancing metaproteomics analysis with deep learning prediction to uncover new microbial communities in deep-sea hydrothermal vents Xinpei Yi1 1Shanghai Jiao Tong University, Shanghai, China Metaproteomics based on tandem mass spectrometry is a crucial tool for studying microbial communities. A major challenge in identifying microbial proteins from metaproteomics data lies in the extensive protein sequence search space, which significantly affects the accuracy and identification rate of microbial peptides. To overcome this issue, we developed a new workflow, DeepMeta, which integrates deep learning-predicted retention time and fragment ion intensity information to enhance the identification rate and accuracy of metaproteomic peptide identification. We evaluated the performance of DeepMeta on three sets of in-silico datasets, including two sets generated by the entrapment strategy and one real-world simulated dataset. Applying DeepMeta to two search engines, Comet and MS-GF+, and comparing it with traditional one-step and two-step search methods, DeepMeta demonstrated an improvement of 25%-45% in the number of identified peptides. Under the condition of maintaining accuracy, the recall rate of DeepMeta averaged around 98%, whereas the traditional methods averaged only 85%-90%. The results obtained by DeepMeta also enhanced the intersection of identified peptides between the two search engines, further proving the accuracy of DeepMeta. Applied to real metaproteomics data from deep-sea hydrothermal vents, DeepMeta significantly increased the peptide identification rate (90% for one-step and 76% for two-step) and uncovered new microbial communities in the deep-sea hydrothermal vents. Integrating deep learning prediction in DeepMeta improves peptide identification in metaproteomics and facilitates biological discoveries in microbial. 654

655

P-II-0501 Unleash the power of Hybrid-DIA data analysis with AI-driven software for biomarker discovery and validation in translational research Qing Zhang1, Zia Rahman1, Baozhen Shan1, Yue Xuan2 1Bioinformatics Solutions Inc., Waterloo, Canada 2Thermo Fisher Scientific, Bremen, Germany Introduction: Hybrid-DIA is a powerful mass spectrometry method that intelligently combines the strengths of Data Independent Acquisition (DIA) and Parallel Reaction Monitor (PRM) for biomarker discovery and validation in one single experiment for unbiased proteome profiling, as well as targeted monitoring/quantifying of endogenous biomarkers using spiked-in counterparts, making it particularly valuable in translational research. It allows [1,2]. However, there is currently a lack of standardized software for processing Hybrid-DIA data. Existing publications require users to navigate through multiple software platforms to obtain DIA results, extract PRM signals, and perform signal normalization[1,2]. This fragmented approach is time-consuming and prone to errors, highlighting the need for a unified and automated software tool. To address this, we propose an AI-driven software tool that simplifies Hybrid-DIA data analysis. Our tool integrates AI algorithms for DIA analysis, PRM signal extraction, and normalization, streamlining the process. Additionally, it utilizes DIA-de novo technology for novel peptide identification. Our software tool enhances efficiency and accuracy in Hybrid-DIA data analysis, unlocking its full potential for biomarker discovery and validation in translational research. Methods: We used published dataset from MSVXXXX to benchmark our proposed workflow. Preliminary Data or Plenary Speakers Abstract: For fitting Hybrid-DIA, we used AI-driven predictors and decision tree to select suitable fragment ions for PRM and DIA quantification, this increased sensitivity and reduce intense labour work. First, we evaluated our workflow using simulated samples, which consist of human Hela cells spiked with isotope-labeled peptides representing 185 tumor-associated antigens (TAA). The results, depicted in Figure X, demonstrate that Hybrid-DIA outperformed conventional DIA in identifying and quantifying TAAs. Hybrid-DIA successfully identified and quantified 167 out of the 185 TAAs, especially excelling in low input samples where DIA failed to extract signals, highlighting the comparable sensitivity of Hybrid-DIA to Parallel Reaction Monitor (PRM). Furthermore, real-world data from patient samples was also analyzed. Our workflow identified an average of 6500 protein groups in both DIA and Hybrid-DIA data. Principal Component Analysis (PCA) plots and protein heatmaps revealed significant differences across different patients. Finally, DIA de novo also reported some interesting peptides which is not listed in the fasta. We check one peptide which may derived from mutation and validate its sequence by synthetic peptide. In summary, our streamlined Hybrid-DIA workflow offers enhanced sensitivity and performance for biomarker discovery and quantitation analysis, particularly in low input samples. Additionally, DIA de novo helps to find potential mutations whose sequence is not recorded in protein database. 655

656

P-II-0502 SPverse: an R package for data analysis of spatial proteomics derived from micro-FFPE tissues Haichao Zhou1, Linyuan Fan1, Guixue Hou2, Siqi Liu2 1University of Chinese Academy of Sciences, Shenzhen, China 2BGI Genomics, Shenzhen, China Spatial omics is a frontier that provides plentiful information regarding gene expression and functions with characteristics of spatial distribution, of which the technology and bioinformatics related to spatial transcriptomics have reached a maturation level, whereas that in spatial proteomics is still at an infant phase. As spatial proteome sampling from tissues and identifying as well as quantifying on peptides are so different from spatial transcriptome, all the procedures of data analysis are reasoned a proteome specific mode. Besides, the ideas on mining the spatial information of transcriptomes and the datasets of single cell transcriptomes should be used as references to develop informatic tool of spatial proteomics. Herein, we have elaborated a workflow of data analysis towards spatial proteomics with R package, termed as SPverse. SPverse is consisted of four layers for data analysis. To gain proteomic data with high quality, the data with poor quality was removed according to protein identification numbers and density distribution then the filtrated data was globally normalized. As proteome differences due to individual diversity interfere sample clustering in entire proteomes of a cohort, Harmony was taken to attenuate these interferences. After such debatching, the treated data was iteratively implemented by consensus clustering until a stable result. To find out the proteomic characteristics in specific regions, the typically representative proteins were extracted through region comparison guided by HE-staining images, and then the index of protein features in a special region was assessed by GSVA. The scores of protein features were mapped onto the tissue slides to construct a spatial proteome image. To compare of spatial feature changes responding to a stress, the differential expression protein (DEP) related to the stress was first derived from Volcano analysis, and the DEPs were put into GSVA to make an index of protein features that functions a scoring system towards spatial distribution of stress-related proteomes. In the proteome database of FFPE micro-tissues, a total of 1144 proteomic data was collected, including 144 from specifically and 1000 from continuously picking up. During pre-treatment to the raw data, 138 from specific spots and 971 from continuous spots were remained. Of these specific spots, 278 feature proteins of tumor and 139 that of stroma were defined, while according to the feature proteins, the tumor scores were generated by GSVA. Based on the proteomic information, these continuous spots were broadly divided into 6 groups by consensus clustering, while upon the tumor scores, the spatial distribution of scores on tissue was delineated. Impressively, the proteomic clusters were spatially overlapped with the distribution of tumor scores. Furthermore, the proteins in the tumor or stroma tissues from the patients who were sensitive or insensitive to chemo-therapy were explored using Wilcoxon and T-test. 656

657

P-II-0503 Enhancing protein analysis with the confident PTM/Mutation algorithm Yandong Zhu1, Wenting Li1, Zia Rahman1, Baozhen Shan1 1Bioinformatics Solutions Inc., Waterloo, Canada Confident PTM/Mutation is an innovative algorithm designed to enhance the accuracy of protein post- translational modification (PTM) and mutation identification. The importance of accurate PTM and mutation identification in proteomics cannot be overstated. PTM plays crucial roles in regulating protein function, activity, localization, and interaction with other biomolecules. Mutations, on the other hand, can lead to altered protein function and are often implicated in various diseases, including cancer. Understanding these modifications and mutations at a high confidence level is essential for elucidating biological processes and disease mechanisms. Furthermore, reliable identification of PTMs and mutations can aid in the development of targeted therapies and personalized medicine approaches, highlighting the critical need for advanced algorithms like Confident PTM/Mutation in the field of proteomics. Traditional methods generate extensive lists of potential modifications, many of which lack confident experimental evidence. This excessive output includes numerous unconfident PTMs and mutations, which contaminate the current identification results. This algorithm improves upon conventional methods by incorporating a filtering mechanism based on fragment ion intensity. It requires a consecutive fragment ions series, including the modified position, to exceed a certain intensity threshold for a modification to be considered confident. By focusing on modifications supported by strong ion intensity, the algorithm significantly reduces the number of unconfident PTMs and mutations. Leveraging the premise that higher ion intensity indicates more reliable fragment ion existence, the Confident PTM/Mutation algorithm provides a streamlined and accurate set of modification candidates. This approach enhances the reliability and accuracy of proteomic analysis. To demonstrate the correctness of the Confident PTM/Mutation algorithm, we applied it to the iPRG 2012 data set for PTM identification. The algorithm successfully reduced the number of reported PTMs from 300 to 16 confident PTMs, all of which have been manually validated. Additionally, we tested the algorithm on the iPRG 2013 data set for mutation identification. Out of 23 peptides, the algorithm confidently identified 6 mutations, while the remaining 17 were also verified manually. These results underscore the algorithm's capability to significantly enhance the accuracy and reliability of PTM and mutation identification in proteomic studies. In summary, the Confident PTM/Mutation algorithm significantly improves the clarity of results compared to traditional methods without filtering out any important modifications or mutations during proteomic analysis. This advancement ensures that critical PTMs and mutations are identified with high confidence, providing a robust tool for advancing our understanding of protein function and disease mechanisms. 657

658

P-II-0504 Deep learning prediction of Glycopeptide tandem mass spectra powers Glycoproteomics Yu Zong1 1Fudan University, Chemistry, Shanghai, China Protein glycosylation plays a significant role in numerous physiological and pathological cellular functions. Glycoproteomics based on liquid chromatography-tandem mass spectrometry (LC-MS/MS) studies the protein glycosylation on a proteome-wide scale to get combinational information on glycosylation site, glycosylation level and glycan structure. However, the current database searching-based methods for glycoproteomics often fall short in glycan structure determination due to the limited occurrence of structure-determining ions. While spectral searching methods can utilize fragment intensity information to facilitate the identification of glycopeptides, its application is hindered by the difficulties in spectral library construction. In this work, we present DeepGP, a hybrid deep learning framework based on Transformer and graph neural network (GNN), for the prediction of MS/MS spectra and retention time of glycopeptides. Two GNN modules are utilized to capture the branched glycan structure and predict glycan ions intensity respectively. Additionally, a pre-training strategy is implemented to alleviate the insufficiency of glycoproteomics data. Testing on multiple biological datasets, we demonstrate that DeepGP can predict MS/MS spectra and retention time of glycopeptides closely aligning with the experimental results. Comprehensive benchmarking of DeepGP on synthetic and biological datasets validates its effectiveness isomeric glycopeptides using MS/MS spectra without diagnostic ions. Based on various decoy methods, we demonstrated that DeepGP in combination with database searching can significantly increase the detection sensitivity of glycopeptides. We outlook that DeepGP can inspire extensive future work in glycoproteomics. in distinguishing similar glycoforms . Remarkably, DeepGP can differentiate 658

659

Multiomics approaches (P-II-0505 to P-II-0558) P-II-0505 Exploring sex differences in zebrafish livers using a novel targeted discovery metabolomics approach Michaela Schwaiger-Haber1,2,3, Bashar Amer4, Darshak Gadara1,2,3, Cristina Jacob4, Philip Remes4, Susan S. Bird4, Gary J. Patti1,2,3, Scott Peterman4 1Washington University in St. Louis, Department of Chemistry, St. Louis, MO, United States 2Center for Metabolomics and Isotope Tracing, Washington University in St. Louis, St. Louis, MO, United States 3Washington University in St. Louis, Department of Medicine, St. Louis, MO, United States 4Thermo Fisher Scientific, San Jose, CA, United States Targeted workflows using triple quadrupole (QQQ) instruments have recently gained more attention in the field of metabolomics. The targeted aspect, namely having to decide which metabolites to measure prior to data acquisition, is a barrier for untargeted metabolomics. However, we have shown previously that even a QQQ instrument can be used to acquire data with the same level of metabolite coverage as conventional untargeted metabolomics experiments. As new nominal mass instruments with high acquisition rates, sensitivity, and advanced features like MSn emerge, our goal is to expand the boundaries of what can be achieved through a targeted approach, and propose an alternative to untargeted analysis. Using a targeted discovery metabolomics experiment, we measured liver extracts from male and female zebrafish. Livers from adult male and female zebrafish (Danio rerio) were harvested. Polar metabolites were extracted with 2:2:1 acetonitrile/methanol/water (40 µL per mg wet weight). A pooled sample was prepared using an aliquot of all samples. Untargeted metabolomics data were acquired using hydrophilic interaction liquid chromatography coupled to a high-resolution mass spectrometer. After feature detection and grouping using Compound Discoverer 3.4, a list of identified and unidentified compounds was generated and used as a precursor list for the new hybrid nominal mass instrument. Data were acquired in parallel reaction monitoring (PRM) mode using fast polarity switching and stepped collision energies. Our data revealed significant sex-based differences in metabolite levels: over 30% of the compounds had a p- value < 0.05 after Benjamini-Hochberg correction and a fold change (|FC|) ≥ 1.5. Threonine levels were higher in female livers, uridine 5'-diphosphoglucuronic acid was seven times higher in females, and methylhistidine was twice as high in males. Additionally, using the instrument"s MSn fragmentation capability and an MS3 method, we characterized the fatty acid composition of triglycerides. Most lipids were increased in female livers, except for those containing fatty acid 18:2, which were lower compared to males. 659

660

P-II-0506 TurboOmics: a web-based platform for the analysis of transcriptomics, proteomics and untargeted metabolomics data using a multi-Omics integrative approach Rafael Barrero-Rodríguez1, José Manuel Rodríguez1, Annalaura Mastrangelo1, Alessia Ferrarini1, Jesús Vázquez1 1Centro Nacional de Investigaciones Cardiovasculares Carlos III (CNIC), Madrid, Spain Multi-omics integrative analysis is an emerging field that allows for a deeper and more comprehensive profiling of interactions among molecular entities across various functional levels. Untargeted metabolomics holds particular promise in this context due to its unbiased approach and its ability to capture the dynamic interactions of the metabolome with the proteome and genome. However, the integration of these omics datasets is often hindered by the lack of tools that combine putative annotations of metabolomic features with multivariate statistical techniques for multi-omics integration. To address this issue, we present TurboOmics (https://proteomics.cnic.es/TurboPutative/TurboOmicsApp.html), a web-based platform that facilitates the integration of proteomics and transcriptomics data with untargeted metabolomics. TurboOmics accepts quantitative data from these three omic fields along with metadata describing the experimental design. Metabolomics features uploaded to the platform are annotated based on their experimental mass/charge ratio using Ceu Mass Mediator, and the resulting annotations are simplified using TurboPutative. Additionally, TurboOmics provides basic functionalities for preprocessing quantitative data and sections for the exploratory data analysis. Multi-omics integration is performed using Multi-Omics Factor Analysis coupled with linear regressions between the projections and the metadata of the observations. This approach enables the fast and intuitive detection of biologically relevant factors and, consequently, the identification of highly correlated molecular entities (transcripts, proteins and metabolites) exhibiting differential behavior among sample groups. Furthermore, TurboOmics performs functional enrichment analysis of relevant proteins and transcripts using g:Profiler. Combined with the obtained putative annotations of the metabolomics features, this comprehensive analysis enables a deep and integral understanding of the biological system under investigation. Reanalysis of the proteomics and metabolomics data from the Lorenzo et al. (2021) study with TurboOmics allows results to be obtained in significantly less time, providing more complete results and demonstrating protein-metabolite associations that were previously unappreciated. Hence, this platform will aid the transcriptomic, proteomic, and metabolomic communities in performing simple, fast, and intuitive integrative analyses with omics data, thereby gathering valuable information to discern associations between biomolecules and disease phenotypes and establish novel disease biomarkers. Fig. 1 (optional) 660

661

P-II-0507 Integrative multi-omic analysis reveals conserved cell-projection deficits in human down syndrome brains Mohit Rastogi1, Martina Bartolucci2, Marina Nanni1, Michelangelo Aloisio3, Diego Vozzi3, Andrea Petretto2, Andrea Contestabile1, Laura Cancedda1,4 1Istituto Itliano di Tecnologia, Brain Development and Disease Laboratory, Genova, Italy 2IRCCS Istituto Giannina Gaslini, Core Facilities of clinical Proteomics and Metabolomics, Genova, Italy 3Istituto Itliano di Tecnologia, Central RNA Laboratory, Genova, Italy 4Dulbecco Telethon Institute, Rome, Italy Down syndrome is a chromosomal alteration characterised by a complex symptomatology affecting several organs including the brain. It is the most frequent cause of intellectual disability of genetic origin and among the cognitive processes most affected are memory and language. While there have been advances in surgery and rehabilitation to improve physical symptoms, there are no pharmacological approaches to date for intellectual ones. The precise mechanisms by which the triplication of HSA21 genes leads to neuroanatomical and neurobehavioral phenotypes in individuals with DS are still poorly understood. In this work, we took a multi-omic approach and simultaneously analyze by RNA-seq and proteomics the expression signature of hippocampus and cortex of human postmortem DS and control brains to identify dysregulated expression networks in DS. Our data provided molecular information that consistently repeated at gene, transcript, and protein levels and identify new possible critical players in the pathophysiology of DS. We observed alteration of RNA splicing specifically altering the expression of genes involved in cytoskeleton and axonal dynamics in DS brains. Accordingly, we found alteration in axonal polarization in neurons from DS human induced-pluripotent-stem cells and mice. Thus, our study provides an integrated multi-layer expression database capable of identifying new potential targets to possibly aid design future clinical interventions in DS capable of enhancing existing cognitive abilities, improving learning and memory skills. 661

662

P-II-0508 Plasma instead of serum avoids critical confounding of clinical metabolomics studies by platelets Andrea Bileck1, Gerhard Hagn1, Samuel Meier-Menches1, Günter Plessl-Walder1, Thomas Mohr1, Gerhard Garhöfer2, Christopher Gerner1 1University of Vienna, Department of Analytical Chemistry, Vienna, Austria 2Medical University of Vienna, Department of Clinical Pharmacology, Vienna, Austria Metabolomics is an emerging and powerful bioanalytical method supporting clinical investigations. Serum and plasma are commonly used without rational prioritization. Serum is collected after blood coagulation, a complex biochemical process involving active platelet metabolism. This may affect the metabolome and increase the variance as platelet counts and function may vary substantially in individuals. A multi-omics approach systematically investigating the suitability of serum and plasma for clinical studies demonstrated that metabolites correlated well (n=461, R2=0.991), whereas lipid mediators (n=83, R2=0.906) and proteins (n=322, R2=0.860) differed substantially between specimen. Independently, analysis of platelet releasates identified most biomolecules significantly enriched in serum when compared to plasma. A prospective, randomized, controlled parallel group metabolomics trial with acetylsalicylic acid administered for 7 days demonstrated that the apparent drug effects significantly differ depending on analyzed specimen. Only serum analyses of healthy individuals suggested a significant downregulation of TXB2 and 12-HETE, which were specifically formed during coagulation in vitro. Plasma analyses reliably identified acetylsalicylic acid effects on metabolites and lipids occurring in vivo such as an increase in serotonin, 15-deoxy-PGJ2 and sphingosine-1-phosphate and a decrease in polyunsaturated fatty acids. The present data suggests that plasma should be preferred above serum for clinical metabolomics studies as the serum metabolome may be substantially confounded by platelets. 662

663

P-II-0509 Development of a multi-omics sample preparation workflow for comprehensive metabolomics, lipidomics and proteomics datasets using a single tissue sample Lana Brockbals1,2, Maiken Ueland2,3, Shanlin Fu2, Matthew P. Padula4 1Zurich Institute of Forensic Medicine, University of Zurich, Department of Forensic Pharmacology and Toxicology, Zurich, Switzerland 2University of Technology Sydney, Centre for Forensic Science, School of Mathematical and Physical Sciences, Sydney, Australia 3University of Technology Sydney, Hyphenated Mass Spectrometry Laboratory, School of Mathematical and Physical Sciences, Sydney, Australia 4University of Technology Sydney, School of Life Sciences, Sydney, Australia Introduction: Multi-omics approaches have gained popularity in recent years aiming to wholistically investigate the biochemical processes in the human body. A frequent problem is the requirement for specialized sample preparation for the different compound classes. This leads to issues with the comparability between analyses (e.g. tissue homogeneity, additional freeze-thaw cycles) and difficulties combining results for biological interpretation. The aim of the current study was to develop a multi-omics sample preparation workflow using a postmortem tissue sample to produce comprehensive Metabolomics, Lipidomics and Proteomics datasets. Methods: Human postmortem muscle and liver tissue samples were used to evaluate different tissue homogenizations- (solvents (n=5), pulsing frequencies (n=5), sample material-to-solvent ratios (n=4) and sample amounts (n=4)) and extraction parameters (extraction solvents/mixtures (n=10)). The general sample preparation workflow included bead homogenization, extraction and compound class specific analysis with LC- MS/MS-based analytical methods as well as 1D-SDS PAGE and a BCA protein assay. Suitability of homogenization parameters were assessed based on a targeted evaluation of the Metabolomics data (microflow-LC-ion mobility separated qTOF analysis, DIA) and based on the numbers/intensities of protein bands visible on the 1D SDS- PAGE (protein fraction). The suitability of the extraction solvents was evaluated and ranked based on the following parameters: targeted and untargeted evaluation of the Metabolomics (see above) and Lipidomics dataset (high flow-LC-orbitrap analysis, DDA), untargeted bottom-up Proteomics data (microflow-LC-ion mobility separated qTOF analysis, DIA), 1D SDS-PAGE (protein- and metabolite fraction) and BCA assay results. Results and Discussion: A methanol-to-water mixture (2:1) added in a 1:10 sample material-to-homogenization solvent ratio was found to lead to the best Metabolomics and visual protein results. A 20 mg tissue sample, homogenized at 3x30 s with a 1 min pause on ice in between pulses, was found to be sufficient to result in comprehensive datasets. The use of a small sample amount is ideal for studies with access to only limited sample amounts. Commonly used 2-phase extraction solvent mixtures for Proteomics studies lead to significantly weaker or less reproducible Metabolomics and Lipidomics results compared to 1-phasic extraction solvents more commonly seen in small molecule studies. A simple protein precipitation of the homogenized tissue samples with a methanol:acetone mixture (9:1) showed the best compromise while achieving comprehensive results. Conclusion: In summary, a multi-omics sample preparation workflow was established that can lead to comprehensive targeted and untargeted Metabolomics, Lipidomics and Proteomics datasets from a single postmortem tissue sample (20 mg). This method can serve as a basis for multi-omics studies in other matrices. 663

664

P-II-0510 Single-Cell proteomics and transcriptomics: unveiling cellular complexity Michael Lewandowski1, Shad Morton1, Aeronn Bevan2, Katharina Meyer1, Jenny Tam1, Ahmad Rushdy1, George Church1, David Walt1, Bogdan Budnik1 1Wyss Institute for Biologically Inspired Engineering, Harvard University, Boston, MA, United States 2Tecan Group Ltd., Zurich, Switzerland Background: Single-cell methodologies have revolutionized biology by allowing us to explore cellular heterogeneity and complexity at unprecedented resolution. The emergence of single-cell proteomics (SCP) over the last five years has helped us understand the function of the cell as a complementary method to deep sequencing and imaging modalities, which have primarily focused on transcriptomics. SCP is a promising tool for studying complex biological processes, such as cell differentiation, drug response, and disease development. It can also provide valuable insights into the role of post-translational modifications in the regulation of cellular function. Methods: Despite the challenge that proteins cannot be amplified like transcripts, single-cell proteomics has already led to exciting discoveries. Among them are identifying rare cell types, analyzing cellular heterogeneity, and examining signaling pathways. Utilizing small sample volumes remain challenging, necessitating robust statistical methods for data interpretation. To address this issue, we have implemented very precise liquid sample handling robots for the separation of material for two different analytical workflows in order to both transcriptomics and proteomics within the same cell. In addition, it is essential to establish standards for widespread accessibility. By using industry-developed hardware that can be easily integrated into most analytical laboratories worldwide, we advocate rapid problem-solving to integrate SCP into a high-throughput, scalable multi-omics platform for diagnosing and treating diseases. Conclusion: Integrating single-cell proteomics with transcriptomics provides a more detailed view of cellular function. As we continue to elucidate single cell behavior within the context of the whole cell population, this rapidly evolving technology promises to revolutionize biological research. The relationship between the transcriptome and the proteome machinery in cell function and disease can be fully understood by using and integrating both transcriptomics and proteomics at single-cell resolution. References: Ahmad, R., & Budnik, B. (2023). A review of the current state of single-cell proteomics and future perspective. Analytical and Bioanalytical Chemistry, 415(2023), 6889–6899. Straubhaar, J; D'Souza, A; Niziolek, Z; Budnik, B (2024) Single cell proteomics analysis of drug response shows its potential as a drug discovery platform Molecular Omics Vol1, issue 20, 1-5. Specht, H., Emmott, E., Petelski, A. A., Huffman, R. G., Perlman, D. H., Serra, M., Kharchenko, P., Koller, A., & Slavov, N. (2021). Single-cell proteomic and transcriptomic analysis of macrophage heterogeneity using SCoPE2. Genome Biology, 22, 50. 664

665

P-II-0511 Delineation of signaling routes that underlie differences in macrophage phenotypic states Marija Buljan1,2, Tiberiu Totu1,2,3, Jonas Bossart1,2,3, Katharina Hast1,3, Vanesa Ayala1 1Empa, Materials science and technology, Health and performance, St. Gallen, Switzerland 2Swiss Institute of Bioinformatics (SIB), Lausanne, Switzerland 3ETH Zürich, Zurich, Switzerland Macrophages represent a major immune cell type in tumor microenvironments, they exist in multiple functional states and are of a strong interest for therapeutic reprogramming. While signaling cascades defining pro- inflammatory macrophages are better characterized, pathways that drive polarization in immunosuppressive macrophages are incompletely mapped. We performed an in-depth characterization of signaling events in primary human macrophages in different functional states using mass spectrometry-based proteomic and phosphoproteomic profiling. Analysis of direct and indirect footprints of kinase activities has suggested PAK2 and PKCα kinases as important regulators of better studied in vitro immunosuppressive macrophages (IL-4/IL-13 or IL-10 stimulated) and has highlighted several other major regulatory kinases in less well studied macrophages with immunosuppresive roles. Network integration of these data with the corresesponding transcriptome profiles has, among others, further highlighted FOS and NCOR2 as central transcription regulators in immunosuppressive states. Furthermore, we retrieved single cell sequencing datasets for tumors from cancer patients and found that the unbiased signatures identified through proteomic analysis were able to successfully separate different macrophage populations in a clinical setting and could thus be used to expand state-specific markers. This study contributes to in-depth multi-omics characterizations of macrophage phenotypic landscapes, which could be valuable for assisting future interventions that therapeutically alter immune cell compartments. 665

666

P-II-0512 Expanding proteomics horizons through metadata Tine Claeys1, Yasset Perez Riverol2, Kris Gevaert1, Juan Vizcaino2, Lennart Martens1 1Ghent University, Center for Medical Biotechnology (CMB), Ghent, Belgium 2EMBL – EBI, Proteomics, Cambridge, United Kingdom Large-scale data analysis in omics is becoming increasingly prevalent, driven by advancements in technical and data analysis methods. These advancements produce vast datasets with significant potential for AI applications and multi-omics pipelines, uncovering novel biological knowledge. One essential factor in these endeavours is the presence of comprehensive metadata, ensuring interpretability. This is however notably lacking within proteomics. The Sample and Data Relationship Format (SDRF) for Proteomics, introduced in 2020, maps data files to sample characteristics using ontology terms. Despite its potential, its complexity and the absence of easy- to-use annotation tools have limited its use. To address this issue, we have developed lesSDRF, a user-friendly, web-accessible application facilitating the SDRF annotation process. esSDRF is designed for ease of use, requiring no installation and available online via Streamlit. It includes all necessary ontologies for SDRF-Proteomics, eliminating the need to navigate the Ontology Lookup Service and ensuring adherence to standardized terminology. The application guides users through five steps: species selection, metadata file upload, labeling information, and input of required and additional columns using ontology terms. This structured approach ensures SDRF-Proteomics compliance and user convenience. Importantly, lesSDRF relies heavily on community input, contributing to the development of new SDRF- Proteomics standards including single-cell proteomics, metaproteomics, and metabolomics. The application is available at https://lessdrf.streamlit.app/ for emerging subfields within proteomics, Despite only being published in Nature Communications in October 2023, lesSDRF has already demonstrated a clear impact, with numerous datasets featuring lesSDRF-generated metadata annotation submitted to PRIDE in May 2024. Current developments include integration into the PRIDE submission pipeline, creation of a desktop application, and incorporation into automated proteomics reanalysis pipelines like quant-ms, open-ms, and wombat-p. To minimize user effort, search engine developers such as FragPipe, PeptideShaker, ionbot, and MSAmanda are integrating the export of a Technical SDRF file into their tools. This feature provides a partially filled SDRF-Proteomics file with technical details in the desired format. Ongoing discussions aim to expand this effort, while MSStats and MSqRob support SDRF-Proteomics compatibility for experimental design files. This metadata initiative extends beyond proteomics. Recognizing the potential of multi-omics analysis, there is a need for a comprehensive metadata standard across various fields. Collaborating with the Research Data Alliance, we are working to establish a multi-omics metadata standard that includes proteomics, genomics, metabolomics, phenomics, and beyond. 666

667

P-II-0513 Ligand-receptor cellular networks from bulk and single-cell proteomics data Jean-Philippe Villemin1,2,3, Pierre Giroux1,2,3, Jacques Colinge1,2,3 1IRCM Inserm U1194, Cancer Bioinformatics and Systems Biology, Montpellier, France 2University of Montpellier, Montpellier, France 3ICM, Montpellier, France The dialog of cells in a tissue through the secretion of ligands and sensing by receptors plays an essential role in development, homeostasis, and diseases (Ramilowski et al., 2015). The advent of single-cell omics has led to remarkable progresses in the analysis of the cell composition and ligand-receptor networks within a tissue. Nevertheless, single-cell data on patient cohorts remain limited compared with bulk datasets, especially in proteomics. To bring the inference of ligand-receptor interactions (LRIs) to regular expression (bulk) proteomics from tissues, we have introduced a sophisticated R library: BulkSignalR (Villemin et al., 2023). BulkSignalR enables reliable LRI inference through a statistical model that integrates observed correlations between the ligand and the receptor, and between the receptor and downstream pathways over a global, reference molecular network (Fig. 1A). Many analytical and graphical functions are available to represent and further manipulate the obtained cellular networks (Fig. 1BC). Very recently, we extended BulkSignalR statistical model to also analyze single-cell data, in particular single-cell proteomics. To the best of our knowledge, this is the first LRI tool adapted to single-cell proteomics. Since this new extension of BulkSignalR can also deal with single-cell transcriptomics, we will show results illustrating the complementarity of the two modalities when available for the same samples. In Specht et al. (2021) data, a significant overlap is observed, but each modality uncovers otherwise missed and relevant cellular interactions in comparable proportions, i.e., there is no obvious superiority but clear complementarity (Fig. 1D). This phenomenon certainly relates to molecule abundances but also proteins with longer half-life that do not necessitate constant gene transcription. Figure 1. (A) Tabular output with triples (ligand, receptor, downstream pathway) along with statistical significance. (B) Classical chord diagram representation of LRIs. (C) Since we model receptor downstream pathways, we can extract networks relating a LRI to its regulatory targets, i.e., to link cellular networks with intracellular networks. (D) Example application of the single-cell extension to Specht et al. data illustrating the complementarity of single-cell transcriptomics and proteomics on such an inter-/intracellular network for differentiated U937 cells. Fig. 1 (optional) 667

668

P-II-0514 Multi-proteomics unveils multi-level dysregulations in methylmalonic acidemia Margherita Ruoppolo1,2, Sabrina Bianco1,2, Lucia Santorelli3, Armando Cevenini1,2, Laxmikanth Kollipara4, Giulia Capitoli5, Marianna Caterino1,2, Albert Sickmann4, Michele Costanzo1,2 1University of Naples Federico II, Department of Molecular Medicine and Medical Biotechnology, Naples, Italy 2CEINGE – Biotecnologie Avanzate Franco Salvatore, Naples, Italy 3Telethon Institute of Genetics and Medicine (TIGEM), Pozzuoli, Italy 4Leibniz-Institut für Analytische Wissenschaften – ISAS – e.V., Dortmund, Germany 5University of Milano-Bicocca, Bicocca Bioinformatics Biostatistics and Bioimaging B4 Center, Monza, Italy Methylmalonic acidemia (MMA) is a rare metabolic disorder due to deficiency of the methylmalonyl-CoA mutase (MUT) enzyme. MMA patients show aberrant degradation of odd-chain fatty acids, cholesterol, and branched- chain amino acids due to the block in the conversion of the methylmalonyl-CoA into succinyl-CoA catalyzed by MUT, with subsequent accumulation of toxic metabolites into body fluids, such as methylmalonic acid and methylcitric acid. Despite the prompt diagnosis of MMA is feasible owing to the mass spectrometry-based newborn screening, actually the damage and stress mechanisms in MMA are not fully clarified and the current therapies are ineffective. To shed light on MMA pathogenesis, diverse MMA cell models were employed, including an engineered MUT- knockout HEK-293 cell line and dermal fibroblasts derived from patients. Multi-proteomic analyses were performed on the MMA models to characterize the whole cell proteomes and their sub-proteomes. Global proteome analysis mainly showed alterations in autophagy- and lysosomal-related pathways, demonstrating the compromission of the lysosomal/autophagic machinery in MUT-deficient cells, which are not able to fully degrade accumulated matter. Both the cell models also showed profound dysregulations in the mitochondrial routes. In fact, the analysis of the mitochondrial proteome enriched specific pathways related to fatty acid and carbohydrate metabolism, revealing unbalances in energetic metabolism. The profiling of the nuclear proteome highlighted epigenetic and structural modifications affecting the MMA cells. Finally, an XL-MS-based approach was applied to study the interactions of the membranome of MUT-deficient cells, indicating novel actors important for cell-cell connections and structural architecture. These data suggest that multi-level dysregulations occurring in MMA patients trigger the cellular and organ damage, providing hints towards novel potential pathways to be targeted to develop efficient therapeutic intervention in MMA. 668

669

P-II-0515 Multi-omics approach for unraveling novel Gaucher disease biomarkers on dried blood spot Maria Concetta Cufaro1,2, Silvia Valentinuzzi1,2, Giorgia Spalluto1,2, Ilaria Cicalini1,2, Fabio Di Ferdinando1,2, Federica Di Marco1,2, Alberto Frisco1,2, Maria Lucia Tommolini1,2, Claudia Rossi1,2, Luca Federici1,2, Vincenzo De Laurenzi1,2, Piero Del Boccio1,3, Damiana Pieragostino1,2 1University of Chieti, Center for Advanced Studies and Technology (CAST), Chieti, Italy 2University of Chieti, Department of Innovative Technologies in Clinical Medicine & Dentistry (DTIM&O), Chieti, Italy 3University of Chieti, Department of Pharmacy, Chieti, Italy leading to a deficiency of a Background: Gaucher disease (GD) is a rare, inherited, and autosomal recessive disorder caused by mutations in lysosomal enzyme, acid β-glucocerebrosidase, and the GBA1 gene glucosylsphingosine (Lyso-Gb1) toxic overload. GD diagnosis and its subsequent management are very challenging due to the wide spectrum of symptoms and complications. The introduction of enzyme replacement therapy (ERT) 30 years ago was considered a great revolution in the treatment of GD patients. However, the exact cause remains unknown, thus a deeper comprehension of the underlying pathogenetic mechanisms is required to provide valuable alternatives for its management. In this context, a "multi-omics" approach could be useful to enable novel biomarker discovery directly on dried blood spot (DBS) by using untargeted metabolomics and proteomics combination that can profile the phenotypical status produced by the genetic background, metabolic and protein characterizations. Methods: Metabolites and proteins from DBS specimens of a homozygous GD patient, before and after ERT, and a healthy control at pediatric age were analysed with an orbitrap platform (Orbitrap ExplorisTM 120 and Orbitrap FusionTM Tribrid for metabolomics and proteomics strategies, respectively). For both "omics" analyses, the investigated first for statistics through Thermo ScientificTM Compound molecular features have been DiscovererTM and Proteome DiscovererTM, then for further combined functional analysis using Ingenuity Pathway Analysis (IPA® tool, QIAGEN). Results: Overall, more than 500 metabolites and 200 proteins were robustly quantified. Among the known GD biomarkers, routinely used in the clinical practice, LysoGb1 and chitotriosidase-1 were confirmed as overexpressed at baseline before ERT, while they decreased after ERT, demonstrating the robustness of our method as proof of concept. Simultaneously, among the other features with the same trend, a new putative biomarker emerged from the metabolomics approach, that is likely compatible with 2-deoxy-2,3-dihydro-N- acetylneuraminic acid, a ganglioside derivative. According to proteomics analysis, galectin-3 binding protein was quantified in GD DBS with the same trend of the metabolites explained above suggesting it as a new putative marker since galectin-3, together with chitotriosidase-1, is already described as sign of macrophage activation. Moreover, functional analysis revealed the modulation of inflammatory pathways involving several classes of sphingolipids, which are reverted by ERT. In conclusion, the DBS "multi-omics" strategy, can help unravel the pathogenesis of GD, the most common sphingolipidosis in the Caucasian population, by identifying new biomarkers and therapeutical targets, and supporting the disease management over time. 669

670

P-II-0516 Direct probing of the energy-information axis in embryonic stem cells and cancer using a multi-omics histone- metabolome-proteome workflow Laura Corvelyn1, Sigrid Verhelst1, Jolien Vanhooren2, Ruben Almey1, Hendrik Marks3, Vincent Pasque4, Dieter Deforce1, Simon Daled1, Maarten Dhaenens1 1Ghent University, ProGenTomics, Ghent, Belgium 2Ghent University Hospital, Pediatric Hematology-Oncology, Ghent, Belgium 3Radboud University Medical Center, Molecular Biology, Nijmegen, Netherlands 4KU Leuven, Stem cell biology, Leuven, Belgium All organisms need energy for their survival, and an intricate network of proteins has evolved to harvest the available energy sources. However, energy availability and necessity intrinsically fluctuate throughout life history and between different tissues and cells and therefore not all available enzymes can be present at every given time. This would be too energy intensive, surpassing the very goal of obtaining more energy. Therefore, life had to evolve a mechanism to regulate protein expression in a fast and dynamic way in response to changes in resource availability. Possibly, such mechanism arose with the first Eukaryote merger, where a prokaryote provided the energy system and the Archaeal cell provided histones as an information management system. Chemical modifications on histones are called histone posttranslational modifications (hPTMs) and they influence the degree of DNA compaction, making genes more or less accessible to the transcription machinery. This regulation of gene expression is what coordinates the production of proteins involved in the various metabolic processes. While still relatively new, the perception is growing that histones could potentially be reading out the energy need and availability in a cell by directly binding the metabolites as PTMs to allow more efficient energy harvesting through altering protein expression. In multicellular organisms, each cell thus can tap into its optimal energy source in this way. Yet, when this energy-information system is hacked, cells can start harvesting energy uncontrollably, potentially fueling disease and carcinogenesis. Therefore, we have developed a workflow to directly measure the expression of over 4000 proteins and the abundance of over 100 histone PTMs and close to a 1000 metabolites from a single cell pellet, effectively allowing us to directly probe this energy-information axis. First, we use a human and mouse embryonic stem cell (ESC) model in an effort to correlate the metabolic conversion between naïve and primed ESCs to the changes in hPTMs. Next, we verify the potential implications in acute myeloid leukemia (AML), where recently a new "Mito- High" phenotype was discovered through proteomics, which is characterized by a high expression of mitochondrial proteins that could in turn affect the histone code. Here, we present some of the most intricate reciprocal correlations between hPTMs and correlate them directly to protein expression in ESCs. Additionally, for 18 AML cell lines, we are in the process of supplementing a similar perspective with their respective metabolomics profile, allowing us to reflect on the available evidence for a direct energy-information axis in Eukaryote cells. 670

671

P-II-0517 Exploring cisplatin resistance in cancer using mass spectrometry-based bioanalytical tools and patient-derived xenograft cell models Carlos López-Portugués1,2, Lucía Gutiérrez-Romero1,2, Borja Gallego3, E Alejandro Sweet-Cordero4, René Rodríguez3, Maria Montes-Bayón1,2, Paula Díez5,1,3 1University of Oviedo, Department of Physical and Analytical Chemistry, Oviedo, Spain 2Health Research Institute of Asturias, Oviedo, Spain 3Health Research Institute of Asturias, Sarcomas and Experimental Therapies, Oviedo, Spain 4University of California San Francisco, Division of Pediatric Oncology, San Francisco, CA, United States 5University of Oviedo, Department of Functional Biology, Oviedo, Spain Osteosarcomas (OST) and ovarian cancer (OC) are heterogeneous aggressive malignancies affecting the bones and the female reproductive system, respectively. Their limited treatment options, mainly consisting of chemotherapeutic agents like cisplatin and doxorubicin drugs, often result in the development of drug-resistant phenotypes and metastasis, compromising the treatment efficacies and therefore the patient's survival. Here we aimed at applying bioanalytical tools (mass spectrometry, MS, and inductively coupled plasma MS, ICP-MS) to understand better the cellular mechanisms mediating drug resistance in cancer. Five commercial and eight primary OST patient-derived xenograft cell lines (PDXCL), and four commercial OC cell lines were subjected to cisplatin-resistance induction by continuous exposure to stepwise increasing drug concentrations. The viability of cell lines was assayed (n=3) after 72h of drug exposure using a cell proliferation reagent to determine the resistance index (RI=IC50resistant/IC50parental). The cellular incorporation of cisplatin within the parental and resistant cell lines and the formation of platinum-DNA adducts were investigated by single cell- and flow injection analysis (FIA)-ICP-MS approaches, respectively. Additionally, MS-based proteomics analyses (LC-Q-TOF) were performed to identify treatment-response predictors at the protein level. After 1-8 months of cisplatin exposure, ~47% of the cell lines (6/9 commercial and 2/8 PDXCL) developed a stable drug resistance with similar RI values (3.82±1.21) for cisplatin treatment. Up to 10.4 times more platinum was detected in parental cellular cells compared to their cisplatin-resistant counterparts (0.228 ft Pt/cell vs 0.022 ft Pt/cell, respectively). Proteomics analysis reported 8,033 proteins in common for parental and cisplatin- resistant models and a 192-protein treatment response signature. Different proteomics phenotypes were also observed for each cell line and cancer type. In summary, high heterogeneity for drug resistance development was observed, with the OST PDX cells showing greater drug sensitivity. Moreover, both OST and OC parental cells depicted higher Pt cellular incorporation, which might relate to increased cell death rates. Finally, high-throughput proteomics identified cisplatin- treatment response signatures for OST and OC, which could serve as potential markers for targeted screening. Altogether, this study evidenced the need to employ patient-derived models to better resemble the clinical settings, together with the co-existence of common and case-dependent phenotypic features associated with developing drug resistance in cancer. 671

672

P-II-0518 Urine biomarkers for diagnosis of pancreatic cancer Anna E Lokshin1, Randal Brand1, Denise Prosser1, Liudmila Velikokhatnaya1, Lynette Smith1 1University of Pittsburgh, Medicine, Pathology, Ob/Gyn, Pittsburgh, PA, United States Multiplexed bead-based immunoassay was used for development of biomarkers associated with resectable pancreatic ductal adenocarcinoma (PDAC). Seventy-eight candidate biomarkers were measured in urine samples from patients with resectable PDAC (rPDAC) (n=88) in comparison with healthy (n=46), benign pancreatic disease (n=41), and unresectable PDAC (urPDAC) (n=36). Following biomarkers were altered in rPDAC vs. non- PDAC (healthy+benign) with high statistically significance at p<0.001: bHCG, Thrombospondin-2, HGF MIF, VCAM1, LRG1, Cyfra21.1, NCAM, ICAM1, KIM1, IGFBP1, CathD, CalbD, ErbB2, Tie2, PAI1, suPAR, SCF, MCSF, IGFBP2, Fn, GDF15, AFP, HE4, PDGFAB/BB, RANTES, PEDF, CXCL10, IL-6, MMP2, EGFR. Only bHCG was statistically different between rPDAC vs urPDAC. Classification algorithm comprised of bHCG, CXCL10, PDGFAB/BB offered 90%SN for rPDAC at 50%SP for healthy+benign. 672

673

P-II-0519 Proteo-Metabo-Flux – linking Acetyl-CoA and acetylation dynamics for accurate determination of histone acetylation rates Anna-Sophia Egger1, Eva Rauch2, Suraj Sharma3, Thomas Mair4, Alina Hohenegg1, Tobias Kipura1, Madlen Hotze1, Philipp Kobler1, Ines Heiland3, Marcel Kwiatkowski1 1University of Innsbruck, Institute of Biochemistry and Center for Molecular Biosciences Innsbruck, Innsbruck, Austria 2Medical University of Innsbruck, Institute of Cell Biology, Biocenter, Innsbruck, Austria 3University of Bergen, Department of Biomedicine, Bergen, Norway 4University Medical Center Hamburg-Eppendorf, Section / Core Facility Mass Spectrometry and Proteomics, Hamburg, Germany Introduction: Histone acetylation is an important post-translational modification that regulates DNA accessibility and gene translation. The substrate for this modification is the small molecule acetyl-CoA, which plays a central role in carbon metabolism. To fully understand the biological function of this reversible post-translational modification it is not sufficient to determine static acetylation levels but it is necessary to capture individual acetylation and deacetylation reaction rates. We have developed a quantitative proteo-metabo-flux approach and pipeline (Proteo-Metabo-Flux-Quant, PMF-Quant) including the simultaneous extraction of acetyl-CoA and histones from cell culture as well as normalisation and correction algorithms to quantify site-specific histone acetylation rates and to combine them with acetyl-CoA turnover dynamics. Results: Here, we present our global Proteo-Metabo-Flux-Quant (PMF-Quant) pipeline for the analysis of acetyl- CoA and histone acetylation dynamics. We use metabolic labelling with [U-13]-C glucose followed by a simultaneous extraction of acetyl-CoA and histones [1]. Histones are tchemically acetylated with 13C4,D6-acetic anhydride and digested using trypsine to obtain chemically equivalent peptides [2]. Ordinary differential equations are used to build an integrated model of label incorporation into both acetyl-CoA and histone acetyl groups. We find that 13C from glucose is incorporated into the acetyl- as well as the ribose and adenine moieties of acetyl-CoA over a of 6 hour period, with label incorporation plateauing at approximately 70%. Taking this partial incorporation into account is important when calculating histone acetylation rates, which are otherwise greatly underestimated. As a proof of principle, we treat cells with the Akt inhibitor MK-2206, which inhibits glycolysis. This leads to an apparent change in histone acetylation rates when the correction is omitted, which is shown to be an artefact of upstream effects. References [1] van Pijkeren, A. et al, Proteome Coverage after Simultaneous Proteo-Metabolome Liquid–Liquid Extraction. Journal of Proteome Research 2023, 22, 3, 951-966 [2] van Pijkeren, A. et al., Combined Metabolic and Chemical (CoMetChem) Labeling Using Stable Isotopes—a Strategy to Reveal Site-Specific Histone Acetylation and Deacetylation Rates by LC–MS. Analytical Chemistry 2021, 93, 38, 12872-12880 673

674

P-II-0521 Assessment of data-independent acquisition mass spectrometry for the identification of single amino acid variants Ivo Fierro-Monti1,2, Christian Schori2, Klemens Fröhlich2, Alexander Schmidt2 1EMBL – EBI, Cambridgeshire, United Kingdom 2University of Basel, Biozentrum, Basel, Switzerland Assessment of data-independent acquisition mass spectrometry for the identification of single amino acid variants. Introduction: Data-independent acquisition mass spectrometry (DIA-MS) holds promise for proteogenomic analysis, particularly in identifying single amino acid variants (SAAVs) crucial for understanding diseases like cancer or neurodegeneration. Here, we evaluate the performance of different commonly used DIA-MS search tools regarding number of correct SAAV identifications and error rates using HeLa cell lysates. We further show a comparison to DDA and confirm the identity of the confidently identified SAAV peptides using SID-PRM targeted MS. Methods: We used HeLa cells and devised a custom sequence search database integrating 233 SAAV peptides derived from HeLa cell single nucleotide polymorphisms (SNPs). We employed DIA-MS with search engines DIA- NN, Spectronaut (SN), and Fragpipe-MSFragger-DIA (FP-DIA). Our evaluation focused on identifying true positive SAAV peptides and false positives through entrapment databases to assess false discovery match ratio (FDMR) for SAAV peptide identifications. We performed stable isotope dilution and parallel reaction monitoring (SID- PRM) to validate the identification of SAAV peptides. Results: FP-DIA showed the lowest FDMR among DIA search engines. DIA-MS demonstrated increased FDMR with larger entrapment databases, yet maintained relatively constant true positive SAAV peptide counts. In addition, integrating DIA and data-dependent acquisition (DDA) MS data enhanced SAAV peptide identification, with a lower false discovery rate (FDR) observed in DDA searches. DIA-MS outperformed DDA-MS in identifying true positive HeLa SAAV peptides albeit displaying greater mean FDMR. Validation using stable isotope dilution and parallel reaction monitoring (SID-PRM) confirmed SAAV peptides identified by DIA-MS and DDA-MS searches, highlighting the reliability of our approach. Conclusion: Our study underscores the capability of DIA-MS for extensive and confident SAAV identification, offering insights into its potential for proteogenomic analyses. The comparative assessment of DIA search engines makes FP-DIA the tool with the least error, suggesting its utility for comprehensive SAAV profiling in disease contexts. 674

675

P-II-0522 Exploring the ghost proteome in ovarian cancer cells by proteogenomic characterization and cross-linking mass spectrometry Diego Fernando Garcia del Rio1,2, Amelie Bonnefond3, Isabelle Fournier2, Michel Salzet2, Tristan Cardon2 1Martin-Luther-University Halle-Wittenberg, Center for Structural Mass Spectrometry, Halle (Saale), Germany 2Université Lille, Laboratoire Protéomique, Réponse Inflammatoire et Spectrométrie de Masse (PRISM)- INSERM U1192, Lille, France 3Pasteur Institute of Lille, European Genomic Institute for Diabetes, Lille, France Recent studies in bioanalytics, biology, and biomedicine have increasingly focused on alternative open reading frame (ORF)-encoded proteins (AltProts), in addition to the traditional reference proteins (RefProts). Previously overlooked, these proteins have also been classified as microproteins, cryptic proteins, and small ORF-encoded proteins. The OpenProt database reveals that AltProts originate from various sources, such as 3" and 5" UTRs, +1 and +2 frame shifts in the main ORF, and non-coding RNAs (ncRNAs). Overall, this ghost or un-referenced proteome enlarges the complexity of proteomes in all eukaryotic organisms. Despite being a new field of research, advanced techniques like mass spectrometry-based proteomics, Next Generation Sequencing, RNA sequencing (RNAseq), and ribosome profiling have facilitated large-scale identification of these proteins. However, their functions remain largely unknown. Some are associated with cell regulation and development, while others have been linked to pathological processes, including cancer. Therefore, understanding their role in cancer-related molecular processes is vital. In this study, we developed a methodology combining cross-linking MS (XL-MS), subcellular fractionation, and proteogenomics of two ovarian cancer cell lines (PEO-4 and SKOV-3 cells) and one normal ovarian epithelial cell line (T1074 cell). By using subcellular fractionation, we could reduce the complexity of the samples analyzed by liquid chromatography-tandem mass spectrometry (LC-MS/MS). With RNA-seq data, we generated customized protein databases for each cell line, which helped identify protein-protein interactions (PPIs). We validated these cross-linked PPIs by modeling the 3D structures of the AltProts, performing docking studies, and measuring the corresponding crosslink distances. Network analysis suggested potential roles for AltProts in biological functions and processes. Our approach revealed an interaction between POLD3 and the AltProt IP_183088 which, after molecular docking, was located between POLD3-POLD2 binding sites, suggesting a possible role in DNA replication and repair. Other benefits of this workflow include non-targeted identification of 597 AltProts based on the predictions from RNAseq experiments, variant annotation, and the differential expression at their subcellular localization. This study underscores the considerable potential of proteogenomics in revealing new aspects of ovarian cancer biology. It enables the identification of previously unexplored proteins and variants that may have functional significance. The use of customized protein databases and the crosslinking approach has shed light on the "ghost proteome," a hitherto unexplored area. 675

676

P-II-0523 Computational analysis of proteome and secretome of human pluripotent stem cells Daniel Giesel1, Sofia Hakala2, Nikolaos Giannareas2, Sara Zanivan3, Jim Norman3, Elisa Närvä2, Tomi Suomi1, Laura Elo1,2 1Turku Bioscience Center, University of Turku and Åbo Akademi University, Turku, Finland 2Institute of Biomedicine, University of Turku, Turku, Finland 3CRUK Beatson Institute, Glasgow, United Kingdom Human pluripotent stem cells (hPSCs) are invaluable tools in regenerative medicine, drug development, and disease modelling due to their ability to differentiate into any cell type in the human body and self-renew indefinitely. Although the primary factors that control pluripotency have been identified, there are still significant knowledge gaps regarding the regulation of pluripotency, particularly through proteomic signaling and interactions within the microenvironment. To bridge this gap, this study leverages computational proteomics and proteogenomics analysis of hPSCs, using both the total proteome and the secretome. We have established a computational proteogenomics workflow that integrates DNA- and RNA-seq data with mass spectrometry proteomics data from the same samples, allowing construction of a tailored database to enhance the interpretation of the mass spectra. The workflow enables revealing pluripotency-associated protein variants that may be missed using conventional methods. In particular, our analysis revealed several differentially expressed proteins in the total proteome and in the secretome of hPSCs, indicating multiple uncharacterised pluripotency-associated factors. These findings emphasize the value of computational techniques in proteogenomics to provide deeper insights into the regulation of hPSCs. 676

677

P-II-0524 Multi-omics analysis of host responses in tracheal tissues of pigs after infection with a human- and swine- adapted influenza A viruses Chrysillis Hellemann Polhaus1, Helena Aagaard Layborn1, Charlotte Kristensen2, Louise Brogaard1, Betina Lyngfeldt Henriksen1, Ramona Trebbien3, Lars Erik Larsen2, Konstantinos Kalogeropoulos1, Ulrich auf dem Keller1, Kerstin Skovgaard1 1Technical University of Denmark, Bioengineering, Kongens Lyngby, Denmark 2Copenhagen University, Copenhagen, Denmark 3Statens Serums Institut, Copenhagen, Denmark Influenza is a global respiratory disease, which annually leads to 3-5 million cases of severe illness, resulting in 290,000-650,000 deaths. During the past century, four global IAV pandemics have claimed millions of human lives. The emergence of IAV pandemics is attributed to zoonotic transmission from avian or swine sources, followed by reassortment events. To understand the underlying molecular mechanisms behind infection, we need more knowledge in how the host reacts to different types of influenza strains, which can be both host- adapted and non-adapted. Even though most infections are assumed to be recognized and controlled by the innate immune system, viral recognition and immune responses in the respiratory tract are poorly understood. To study the molecular landscape of infection and host response, we inoculated two groups of 12 pigs with 10^7 TCID50/ml of swine- (swH1N1) or human-adapted (huH1N1) influenza A virus, and one control group of 6 pigs with culture medium. We collected tracheal tissues three days post infection and investigated the host response by global proteomics, RNA-seq, high-throughput microfluidic qPCR, and immunohistochemistry. With our results, we describe a classical antiviral immune response with induction of IFNs and ISGs in tracheal tissues after infection with both swH1N1 and huH1N1, though with differences in dynamics and expression levels after infection with the two strains. We find that infection with the swine-adapted strain results in necrotizing tracheitis and regulation of other host metabolic factors in the trachea, such as cytoskeleton elements, the pyrimidine metabolism, apoptosis, and factors involved in the JAK-STAT pathway. We conclude that host factors centrally involved in the early orchestration of the antiviral immune response are tightly regulated in the trachea post infection. Expression levels, dynamics, histological findings, and regulation of metabolic host factors were altered depending on adaptation level of the virus, highlighting that host adaptation and the risk of cross-species infection might be connected to modulation of host factors by the virus. 677

678

P-II-0525 Groningen proteogenomics workflow for detecting a wide range protein variants in LC-MS/MS data Yanick Paco Hagemeijer1,2, Yuan Hu1,2, Tessa Gilette1,2, Frank Klont1, Cornelia Verhoeven3, Justina Clarinda Wolters4, Gyorgy Halmos3, Victor Guryev2, Peter Horvatovich1 1University of Groningen, Analytical Biochemistry, Groningen Research Institue for Pharmacy, Groningen, Netherlands 2European Research Institute for the Biology of Ageing, Laboratory of Genome Structure and Ageing, Groningen, Netherlands 3University Medical Center Groningen, Department of Otorhinolaryngology, Head and Neck Surgery, Groningen, Netherlands 4University Medical Center Groningen, Laboratory of Pediatrics, Groningen, Netherlands Introduction: Common proteomics workflow uses manually curated public protein sequences, which either use canonical sequences (one sequence per protein coding genes) or a limited number of well-known high frequency occurring isoforms and protein variants. This limits the potential to study the presence and effects of protein variants in proteomics studies in context of basic biology and clinical research. Transcriptomics are often used in research as proxy for protein variants, however, only fraction of these variants can be detected at sufficient levels in the proteome. In this work we present a generic proteogenomics data integration workflow1 called Groningen Proteogenomics Workflow (GPW), which predicts protein sequence variants from polyadenylated transcriptomics data obtained from the same sample and connects that to the generated proteomics results. GPW is designed to predict a wide variety of protein variants beside the well-known annotated sequences such as single amino acid variants, indels, splice isoforms and translated sequences from new open reading frames (ORFs). Approach: GPW has been implemented in python and nextflow2 and include mRNA quality assessment tool (FastQC) and alignment tool for reference genome (STAR), splice isoform prediction (StringTie), de novo assembly of unmapped reads (Trinity), variant calling (GATK, bcftools), functional annotation using VEP and translation of mRNA sequences to protein amino acid sequences with transdecoder. The translated sequences are used for peptide and protein identification in data dependent and independent (DDA/DIA) LC-MS/MS proteomics data are analyzed with the FragPipe3 toolset. GPW produces tables and plots summarizing the different protein variant categories. Results and discussion: The power of GPW is demonstrated on laryngeal squamous cell carcinoma and controls of healthy mucosa of the same site showcasing the effect of protein variants in tumor characteristics of young and elderly patients as well as to identify the genes with missense variants in laryngeal cancer and healthy tissue controls of the same patients. References 1. Hagemeijer YP, Guryev V, Horvatovich P. Accurate Prediction of Protein Sequences for Proteogenomics Data Integration. Methods Mol Biol. 2022;2420:233-260. 2. Di Tommaso P., et al. Nextflow enables reproducible computational workflows. Nature Biotechnology 35, 2017, 316–319. 3. Yu, F, Teo, G. C., Kong, A. T., Fröhlich, K., Li, G. X. , Demichev, V, Nesvizhskii, A..I. (2023). Analysis of DIA proteomics data using MSFragger-DIA and FragPipe computational platform, Nature Communications 14:4154. 678

679

P-II-0526 The comprehensive SARS-CoV-2 "hijackome" knowledge base Sini Huuskonen1, Xiaonan Liu1, Taras Redchuk1, Kari Salokas1, Rickard Lundberg2, Sari Maljanen2, Milja Belik2, Arttu Reinholm2, Pekka Kolehmainen2, Garima Tripathi1, Pia Laine1, Sergei Belanov1, Petri Auvinen1, Maria Vartiainen1, Pamela Österlund3, Larissa Laine3, Ilkka Julkunen4,2,3, Laura Kakkola4,2,3, Markku Varjosalo1 1University of helsinki, Institute of Biotechnology, Helsinki, Finland 2University of Turku, Clinical Microbiology, Turku, Finland 3Finnish Institute for Health and welfare, THL, Helsinki, Finland 4University of Turku, InFlames Research Flagship Center, Turku, Finland The continuous evolution of SARS-CoV-2 has led to the emergence of several variants of concern (VOCs) that significantly affect global health. This study aims to investigate how these VOCs affect host cells at proteome level to better understand the mechanisms of disease. To achieve this, we first analyzed the (phospho)proteome changes of host cells infected with Alpha, Beta, Delta, and Omicron BA.1 and BA.5 variants over time frames extending from 1 to 36 hours post-infection. Our results revealed distinct temporal patterns of protein expression across the VOCs, with notable differences in the (phospho)proteome dynamics that suggest variant-specific adaptations. Specifically, we observed enhanced expression and activation of key components within crucial cellular pathways such as the RHO GTPase cycle, RNA splicing, and ER-associated degradation (ERAD)-related processes. We further utilized proximity biotinylation Mass Spectrometry (BioID-MS) to investigate how specific mutation of these VOC influence viral-host protein interactions. Our comprehensive interactomics dataset uncovers distinct interaction profiles for each variant, illustrating on how specific mutations can change viral protein functionality. Overall, our extensive analysis provides a detailed proteomic profile of host cell for each variant, offering valuable insights into how specific mutations may influence viral protein functionality and impact therapeutic target identification. These insights are crucial for the design of new targeted interventions, aiming to enhance the efficacy of treatments against evolving SARS-CoV-2 variants. Fig. 1 (optional) 679

680

P-II-0527 Ubiquitinomic profiling of CYLD cutaneous syndrome skin tumours identifies an insulin signalling axis Joseph Inns1, Andrew Frey2, Elise Bennett1, Georgie Holt1, Matthias Trost2, Neil Rajan1 1Newcastle University, Translational and Clinical Research Institute, Newcastle upon tyne, United Kingdom 2Newcastle University, Biosciences Institute, Newcastle upon tyne, United Kingdom CYLD cutaneous syndrome (CCS) is an inherited hair follicle tumour syndrome driven by biallelic loss of function pathogenic variants in the deubiquitinase (DUB) CYLD. CCS tumour cells hence represent a patient relevant model to gain insights into the ubiquitinomic consequences of loss of function of this important DUB. We utilised ubiquitinomic profiling to detect enriched ubiquitin tagged substrates in CCS tumours. We complemented these data with CCS tumour keratinocyte proteome and single cell transcriptome data to develop a multi-omic model which encapsulated the CCS tumour microenvironment. These data led us to discover an altered insulin signalling axis in CCS tumours. CCS tumour tissue and healthy skin control samples were collected and flash frozen for ubiquitinomic analysis. Tissue slices were lysed, and trypsin digested with the resulting peptides undergoing di-glycine immuno- purification. Additionally, we processed fresh tumour tissue and control skin, by enzymatic digestion and flow cytometric sorting. We assessed the proteome of CCS tumour initiating CD45- CD200+ keratinocytes and prepared CD45- cells for single cell transcriptomic analysis using the 10X workflow. Tryptic peptides were analysed by liquid chromatography tandem mass spectrometry (LC-MS/MS) on a Bruker Tims-ToF HT mass spectrometer, utilising a data independent acquisition method. Peptide identification was performed by searching MS data against an in-silico neural network derived spectral library with di-glycine modification. We identified 1273 ubiquitination sites across 649 proteins in CCS tumour and control samples through ubiquitinomic analysis. We found 305 ubiquitination sites that were either unique to their respective condition or differentially regulated after performing t-tests (p<0.05). Analysis of ubiquitinated peptide sequences uncovered cell surface receptors with increased ubiquitination in CCS tumour samples including the insulin receptor which had ubiquitination modifications at its kinase domain. Immunofluorescence staining of cylindroma sections revealed localisation of the insulin receptor to central regions of cylindroma "islands". We interrogated a CCS single cell transcriptomics dataset (n=3 tumours; k=19,754) and a CCS CD45- CD200+ proteomic dataset (n=5 tumours) to delineate insulin signalling. Across these datasets we demonstrated evidence of an insulin signalling axis, with increased expression of IGF1 in CCS tumour associated fibroblasts, and IGF1R, INSR, IGFBP3, and MMP7 in CCS tumour keratinocytes. Our findings delineate the ubiquitinome of CCS keratinocytes and highlight the in the CCS tumour microenvironment. insulin receptor as a putative target of CYLD 680

681

P-II-0528 Factor analysis based methods for an integrated unsupervised analysis of multi-omics data Bernard Isekah Osang'ir1,2, Surya Gupta1, Jürgen Claesen2,3, Ziv Shkedy2 1Belgium Nuclear Research Centre (SCKCEN), Mol, Belgium 2Hasselt University, Hasselt, Belgium 3Amsterdam University Medical Center, Amsterdam, Netherlands Multi-omics data integration is essential in life sciences for comprehensive insights into complex biological systems. With high-throughput technologies, the amount and diversity of bio-molecular data generated have grown substantially and often, multiple (omics) datasets (platforms) are available in a single experiment. To identify common biological processes across different platforms, various approaches such as network-based, cluster-based, correlation-based, and unsupervised factor-based methods have emerged in uncovering underlying patterns and interdependence within large and heterogeneous datasets. We proposed a new method and pipeline for Integrated data analysis based on Factor Analysis for Biclustering Acquisition (FABIA)[1, 2],— and conducting a benchmark analysis of this method together with other three unsupervised factor-based methods—Multi-Omics Factor Analysis (MOFA)[3], Multiple Factor Analysis (MFA)[4], Group Factor Analysis (GFA)[5]—to assess their effectiveness in integrating proteomics and other diverse bio-molecular data types. The methods were tested using both the real datasets and the data from the simulation study. Three multi-omics datasets are used: (1); an experimental dataset from the invivo radiation experiment conducted by Belgian Nuclear Research Centre (SCK•CEN) microbiology lab. to study the impact of radiation on the cognitive abilities of the newborn males mice exposed to Gy ionizing radiation. The dataset comprises of proteomics data and transcriptomics data, (2); an open-source proteomic and transcriptomics datasets, and (3); a simulated high- dimensional datasets (HDD) with pre-specified signals (latent factors). For the HDD, we examine 100 replicates, adjusting noise levels from very low to extremely high while monitoring the performance of the method's ability to identify the true signal. Our primary objective was to determine if these methods consistently identified the same patterns in terms of features and factors. The findings of our analysis revealed that all the methods uncovered similar patterns. This was assessed through pairwise-Pearson correlation analysis which examined the relationship between the feature weights and factor scores across methods. Additionally, we used Jaccard similarity indices (JSI) and performance metrics such as sensitivity, specificity, and accuracy to further evaluate the results. References 1. Hochreiter, S., et al 2010. FABIA: factor analysis for bicluster acquisition. 2. Kasim, A., et al 2016. Applied biclustering methods for big and high dimensional data using R. 3. Argelaguet, R., et al 2020. MOFA+: a statistical framework for comprehensive integration of multi- modal single-cell data. 4. Abdi, H., et al 2007. Multiple factor analysis (MFA). Encyclopedia of measurement and statistics. 5. Klami, A., et al 2014. Group factor analysis. IEEE transactions on neural networks and learning systems. 681

682

P-II-0529 Inference of mechanistic insights from multi-omics data facilitating the drug discovery process Mathias Kalxdorf1, Trisevgeni Rapakoulia2, Holger Franken1, Emma Laing2 1GSK, Heidelberg, Germany 2GSK, Stevenage, United Kingdom Background: Drug development is a costly and challenging process, among others, due to high failure rates at late stages of the drug discovery process. Lack of comprehensive knowledge of disease mechanism and of causal effects induced by perturbation of selected drug targets are key causes for failure. Increasing availability of multi-omics data provides opportunities to address these lacks through quantitative assessment of mechanistic hypotheses from such data. Method: A two-step approach is proposed to infer mechanisms from multi-omics data. First, agreements between observed data and prior-knowledge molecular interaction graphs are identified. Second, measures of confidence are added to mechanistic hypotheses by joining evidence from causal-reasoning with evidence from multi-omics-based protein activity estimations. Results: The approach is evaluated using proteomics, phosphoproteomics, and transcriptomics data from pro- and anti-inflammatory macrophages. Results demonstrate how omics layers complement each other to provide mechanistic insights e.g. for key regulators like STAT1 and STAT6. Conclusion: The presented approach enables quantitative inference of mechanistic insights from complex biological systems, linking disease-causing genes to measured phenotypes and explaining causal routes from drug targets to perturbation-induced effects. This approach can help to identify novel high-confidence drug targets, reveal unfavorable off-target mechanisms, and thereby facilitate the drug discovery process. 682

683

P-II-0530 A comprehensive multi-OMICS integration framework to address sample heterogeneity and identify causative genes and proteins in the pathogenesis of Keratoconus Ramaraj Kannan1,2, Rohit Shetty3, Pooja Khamar3, Marlies Gijs2, Rudy Nuijts2, Arkasubhra Ghosh1, Vrushali Deshpande1 1Narayana Nethralaya Foundation, GROW Research lab, Bangalore, India 2Maastricht University Medical Center, University Eye Clinic Maastricht, Maastricht, Netherlands 3Narayana Nethralaya, Cornea and Refractive surgery, Bangalore, India Purpose: Keratoconus (KC), a complex corneal disorder, requires a comprehensive understanding due to its multifactorial nature. Limited sample availability often confines us to single omics strategies. This study introduces an innovative framework for integrating multi-omics data, addressing the challenges of heterogeneity in sample type and aims to uncover causative mechanisms behind KC pathogenesis. Methods: WES was conducted using blood samples from familial KC cohort (n = 9 families). Transcriptomic analysis, based on microarray technology, was performed on corneal epithelium samples (KC = 60, control = 18) in a prior study, and mass-spectrometry-based Proteomics was conducted on tears (KC = 107, control = 40). An in-house gene-protein interaction network-based framework was developed using Python 3.7. Results: Individual omics analyses were performed, revealing genomics data identifying seven novel genes, including immunomodulatory metalloproteinases and transcription regulatory factors. Transcriptomics uncovered elevated reactive oxygen species expression, while proteomics identified proteins associated with ROS, inflammatory processes, and ECM remodelling. The inhouse integration framework revealed that the gene with pathogenic variant TPPP plays a regulatory role in various other genes associated with apoptotic functions, including S100A8 and S100A6. CTBP2 governed genes related to the ECM and proteins in oxidative stress, while the immune modulator KIR2DL1 induced alterations in IQGAP1 expression. Conclusion: Our framework was able to map dysregulated gene expression in corneal epithelium and corresponding protein expression in tears. These findings contribute novel insights into the potential mechanisms driving the pathogenesis of KC, offering a promising avenue for further research and clinical applications. Fig. 1 (optional) 683

684

P-II-0531 Human embryonic stem cells (hESC) epitranscriptome forms epitranslatome and translation profiles Anna Kozlova1, Sergey Radko2, Elena Ponomarenko1, Ekaterina Ilgisonis1 1Institute of Biomedical Chemistry, Postgenomic Data Analysis Laboratory, Moscow, Russian Federation 2Institute of Biomedical Chemistry, Laboratory of Protein Biochemistry and Pathology, Moscow, Russian Federation Methylation of m6A mRNA is involved in important regulatory functions such as proper splicing, translation initiation, and translation arrest. Recently, quite a lot of information has been accumulated on transcriptome methylation, but few people have mentioned methylation in the context of the translatome. To study the effect of m6A methylation on translation, we obtained the translatome and transcriptome profiles for the same human embryonic stem cell samples on the Oxford Nanopore sequencing platform. About 42% of expressed splice- forms were detected at the translatome level. The m6A modification was detected in about 50% of transcripts at the transcriptome level, but only in 3% (476 transcripts) at the translatome level. In the translatome, 106 m6A- modified transcripts were detected, which were not found in the transcriptome. Comparison of epitranslatome data with ribosomal pause sites obtained from Ribo-seq data showed that methylation plays an important role in the regulation of translation rate. This work was supported by RSF No.24-14-0006. 684

685

P-II-0532 Proteomic profiling of paired primary tumors and brain metastases from lung adenocarcinoma patients uncovers phenotypic changes during disease progression Magdalena Kuras1, Nicole Woldmar1, Beata Szeitz2, Viktoria Laszlo3, Judit Moldvay2, Attila Marcell Szasz4, Johan Malm5, Lilla Reiniger2, Peter Horvatovich6, György Marko-Varga1, Zsolt Megyesfalvi3, Balázs Döme3, Melinda Rezeli1 1Lund University, Department of Biomedical Engineering, Lund, Sweden 2National Korányi Institute of Pulmonology, Budapest, Hungary 3Medical University of Vienna, Vienna, Austria 4Semmelweis University, Budapest, Hungary 5Lund University, Department of Translational Medicine, Malmö, Sweden 6University of Groningen, Department of Analytical Biochemistry, Groningen, Netherlands At the time of lung cancer diagnosis, approximately half of the patients have existing metastases. In lung adenocarcinoma (LADC), brain metastases are the most common, which have a significant negative impact on patient outcomes. The main objective of this study was to characterize surgically removed primary tumors and brain metastases from patients with LADC to better understand tumor heterogeneity and plasticity during disease progression. Formalin-fixed and paraffin-embedded primary and brain tumor tissues from 58 patients were subjected to proteomic analysis using a previously established sample preparation protocol (Kuras et al., 2021). Peptides were labeled with TMTpro 16-plex, and the proteomic analysis was performed using high-performance liquid chromatography coupled with tandem mass spectrometry (Ultimate 3000 UPLC, QExactive HF-X). Protein identification and quantification were carried out using Proteome Discoverer 2.4 software. Statistical and bioinformatic analyses were conducted using R and Perseus software. The proteomic analysis identified 10,862 proteins, from which 7,418 proteins were quantified across all 116 samples. Principal component analysis separated the samples based on tissue origin, revealing a greater heterogeneity among the brain metastases than the primary tumors. Four proteomic subtypes were detected by consensus clustering in both primary tumors and brain metastases, of which the primary subtypes showed an association with progression-free survival (PFS). The primary subtype with worse PFS was characterized by fast proliferation, which was indicated by higher expression of Ki67, Myc targets, and cell-cycle-related pathways. In contrast, the subtype with the best PFS showed higher expression of immune-related proteins. Downregulation of protective immune mechanisms was observed in brain metastases compared to primary tumors. Additionally, there was no significant association between primary and metastatic subtypes, which is indicative of a phenotypic change during disease progression. Our results provide insight into the molecular heterogeneity of LADC tumors and the development of brain metastasis. Better disease stratification could be obtained by proteomic profiling of resected primary tumors and metastatic lesions. Our results indicate a phenotypic change occurring during disease progression, which is why the effectiveness of metastasis treatments based only on examining primary tumors is uncertain. Funding: Royal Physiographic Society in Lund, Mrs. Berta Kamprad"s Cancer Foundation (FBKS-2020-22-(291)), National Research, Development and Innovation Office (2021-1.2.4-TÉT-2021-00030). 685

686

P-II-0533 Alternative splicing derived proteoforms show tissue-specific expression and rewire intrinsically disordered regions Boomathi Pandi1, Stella Brenman1, Alexander Black1, Dominic Ng1, Edward Lau1, Maggie Lam1 1University of Colorado, Anschutz Medical Campus, Medicine, Aurora, CO, United States Alternative splicing (AS) contributes to the functional diversity of the genome by allowing multiple transcript variants to be encoded by one gene. Prior work using transcriptomics has shown that AS is highly tissue specific and has strong potential to alter protein interaction networks. In contrast, knowledge into the protein-level existence and function of the translated protein isoforms remains lagging. To address this gap, we and others have employed a proteogenomics approach that combines information from transcriptomics and proteomics to identify hundreds of AS-derived isoforms in mammalian tissues and cell lines. We have used our in-house software JCAST to translate RNA-seq-derived splicing information into protein isoform sequence databases and then generated new mass spectrometry data to discover AS-derived isoforms, leading to the discovery of protein-level evidence of hundreds of AS-derived protein isoforms. More recently, we found 29 non-canonical isoforms that show signs of regulation based on statistically significant preferences in different chambers of the human heart from public proteomics datasets. To investigate the potential function of these non-canonical isoforms, we focused on the atrium-enriched isoform 2 of PDZ and LIM Domain 3 (PDLIM3-2). Using recent deep learning based software tools including AlphaFold-Multimer, we examined the function of the expressed protein isoforms and found that the alternative region of the protein impacts PDLIM3 binding to its interaction partner. Moreover, on a proteome-wide scale, we find that variant regions that differ between alternative and canonical isoforms are highly enriched in intrinsically disordered regions (IDR), and over two-thirds of such regions are predicted to function in protein binding and/or RNA binding. Our analysis lends further credence to the notion that AS diversifies the proteome by rewiring IDRs, which are increasingly recognized to play important roles in the generation of biological function from protein sequences. 686

687

P-II-0534 Comprehensive Solution of Ultrasensitive Proteome Technology (CS-UPT) combined with low-input multi-omics analysis provides new insights and clues for mouse maternal-to-zygotic transition Chen Li1 1Shanghai Jiao Tong University School of Medicine, Shanghai, China Single-cell and low-input multi-omics techniques have revolutionized the study of preimplantation embryo development. However, single-cell and low-input proteomics researches are relatively underdeveloped. To get enough depth proteome, hundreds of mouse embryos often need to be collected for each sample. Maternal-to-zygotic transition (MZT) is a fundamental and conserved process, during which the maternal environment of the oocyte transitions to the zygotic genome driven expression program. N6-methyladenosine (m6A) is the most prevalent internal modification present in the mRNA across higher eukaryotes. However, the potential regulation role of m6A modification on RNA metabolism during MZT remains largely unknown. Therefore, we developed a Comprehensive Solution of Ultrasensitive Proteome Technology (CS-UPT) for single- cell and low-input mouse oocytes and preimplantation embryos. Both deep coverage and high-throughput routes significantly reduced the starting material and provided a full set of options for investigators" demands. Using the deep coverage route, we provided a large-scale snapshot of the very early stage of mouse MZT, including almost 5,500 protein groups from 20 mouse oocytes or zygotes for each sample. Moreover, significant protein regulatory networks centered on transcription factors and kinases between the MII oocyte and 1-cell embryo provided rich insights into minor zygotic genome activation. We also investigated m6A dynamics during mouse MZT through low-input multi-omics analysis. Our results showed that m6A could be maternally inherited or de novo gained after fertilization. By integrating SLIM-seq, LiRibo-seq, low-input proteomics and RIP-qRT-PCR data, we demonstrated that a group of maternal mRNAs with m6A mark were maintained throughout the MZT, and most of these genes were actively translated, indicating a role of m6A in safeguarding RNA stability for early embryo development. All these findings provided new insights and clues for the study of critical biological events during mouse preimplantation development. 687

688

P-II-0535 Integral-Omics: serial profiling of metabolome, lipidome, genome, transcriptome, whole proteome, and phosphoproteome in biopsy tissue Wei Li1, Yujuan Wei2, Rui Sun1, Junke Zheng2, Yi Zhu1, Tiannan Guo1 1Westlake University, School of Medicine, Hangzhou, China 2Shanghai Jiao Tong University, School of Medicine, Shanghai, China Integrative multi-omics studies often analyze different tissue pieces or homogenized bulk specimens, neglecting the limited size of biopsy-level specimens ( lipidome, genome, We extensively reviewed established methods for the profiling of metabolome, transcriptome, proteome, and phosphoproteome, and developed a novel integral workflow for extracting metabolites, lipids, genomic DNA, total RNA, proteins, and phosphopeptides from a biopsy-level tissue specimen (10 mg). It does not require specialized instruments. The workflow begins with lipidome and metabolome extraction, followed by gentle sonication to release RNA and proteins in a lysis buffer, while DNA remains in the precipitate. Optional phosphopeptide enrichment can be performed for phosphoproteome characterization. From a 10 mg mouse liver sample, the Integral-Omics pipeline successfully identified 1138 lipids and 997 metabolites using MRM-based targeted mass spectrometry on a Sciex 6500+ instrument. Additionally, the pipeline yielded a total RNA extraction of 0.18% with RIN/RQN values ranging from 8.1 to 8.6, and a genomic DNA extraction of 0.14% with a main fragment size exceeding 20,000 bp. Peptides were extracted with a yield of 9.7%. Subsequently, a 30-minute DIA analysis on a Q Exactive HF instrument resulted in the identification of 48,337 peptides derived from 4652 proteins. For phosphoproteome analysis, 80 µg of peptides underwent enrichment, leading to the identification of 18,238 phosphopeptides using a 90-minute DDA LC-MS acquisition strategy on a Bruker timsTOF Pro instrument. We further applied this method to 6 colorectal adenoma and carcinoma tissue specimens (~ 10 mg each), and demonstrated its practicality in analyzing the multiomics of clinical biopsy specimens. In summary, our approach enables the sequential profiling of metabolome, lipidome, genome, transcriptome, whole proteome, and phosphoproteome using biopsy tissue samples. This integrated methodology provides a comprehensive analysis of biomolecules, offering insights into tissue composition and biological processes. Fig. 1 (optional) 688

689

P-II-0536 Time-resolved metabolite and lipid profiling depicts macrophage continuum with apoptotic and ferroptotic heterogeneity along foam cell formation Ling Lin1, Yiwen Wang1, Yunzeng Zou1, Liang Qiao1 1Fudan University, Shanghai, China Atherosclerosis is a chronic inflammatory disease driven by the maladaptive lipid metabolism and inflammatory response of macrophages. In the progression of atherosclerosis, macrophages work as a lipid scavenger and immune regulator by dynamic lipid and metabolic reprogramming in response to the microenvironment. Nevertheless, we are only starting to characterize the function of the heterogeneity in metabolite and lipid compositions. Herein, we developed a four-dimensional micro-manipulation platform for single-cell sample preparation, and observed the macrophage heterogeneity along foam cell formation via a single-cell time- resolved metabolite and lipid profiling. By integrating bulk metabolomic and lipidomic data with single-cell metabolomic and lipidomic data, we were surprised to uncover that macrophages have different fate outcomes during the late foam cell formation process. Some foam cells were more prone to apoptosis, while others are more susceptible to ferroptosis, suggesting heterogeneity in cell fate determination could have important implications for understanding atherosclerosis progression. Single-cell transcriptome sequencing was further performed on late-stage foam cells derived from both THP-1 and peripheral blood mononuclear cells, jointly confirming the divergent cell fates toward apoptosis or ferroptosis. Lastly, we validated the heterogeneity of late-stage foam cell differentiation through a series of in vitro experiments, including caspase activity assay and lipid peroxidation assay. In summary, single-cell multi-omics depict the molecular choreography that dictates the cell death in late atherosclerosis. Keywords: Single-cell analysis; Metabolomics; Lipidomics; Foam cell; Ferroptosis. Fig. 1 (optional) 689

690

P-II-0537 Metaproteomic profiling of pediatric patients microbiota with acute lymphoblastic leukemia: impact of chemotherapy Sameh Magdeldin1,2, Shahd Ezzeldin1,2, Ashraf Essam1,2, Aya Osama1,2, Ahmed Karam1,2, Asmaa Ramzy1,2, Taghreed Khaled Abdelmoneim1,2, Maha Mokhtar1,2 1Children's Cancer Hospital CCHE-57357, Proteomics and Metabolomics research program, Cairo, Egypt 2Children's Cancer Hospital CCHE-57357, Proteomics and metabolomics, Cairo, Egypt This study aims to assess the impact of chemotherapy on the gut microbiota of pediatric patients diagnosed with acute lymphoid leukemia (ALL). We will analyze 90 stool samples collected from these patients during the initial and induction phases of chemotherapy treatment. The microbiota composition will be evaluated by extracting metaproteins from the stool samples, followed by peptide fingerprinting techniques to identify microbial proteins. This approach will allow for a detailed assessment of how chemotherapy influences the microbial communities in the gut, which could potentially affect patient health and treatment outcomes. Understanding these changes could provide insights into the management of microbiota-related complications and aid in developing microbiota-friendly therapeutic strategies for ALL patients undergoing chemotherapy. 690

691

P-II-0538 Multi-omics characterization of the monkeypox virus infection Yiqi Huang1, Vincent Grass1, Quirin Emslander1, Valter Bergant1, Sabri Hamad1, Philipp Hubel2, Julia Mergner3, Antonio Piras1, Karsten Krey1, Alexander F. Henrici1, Rupert Oellinger4, Yonas Tesfamariam5, Till Bunse1, Gerd Sutter6, Florian I. Schmidt5, Michael Way7, Roland Rad4, Andrew Bowie8, Gregor Ebert1, Ulrike Protzer1, Andreas Pichlmair1 1Technical University of Munich, Institute of Virology, Munich, Germany 2University of Stuttgart, Core Facility Hohenheim, Stuttgart, Germany 3Technical University of Munich, Bavarian Center for Biomolecular Mass Spectrometry at Klinkum rechts der Isar, Munich, Germany 4Technical University of Munich, Institute of Molecular Oncology and Functional Genomics, Munich, Germany 5University Hospital Bonn, Institute of Innate Immunity, Bonn, Germany 6Ludwig-Maximilians-University Munich, Institute of Infectious Diseases and Zoonoses, Munich, Germany 7The Francis Crick Institute, London, United Kingdom 8Trinity College Dublin, School of Biochemistry and Immunology, Dublin, Ireland Due to its exceptional scale and spread, the 2022-2023 outbreak of monkeypox was declared a global emergency by the World Health Organization. The global spread of the monkeypox virus has demonstrated the need for in-depth characterization of the virus infection process and virus-host interactions. Multi-omics studies have helped to understand the fundamental principles of virus pathophysiology and thus guide the selection of virus- and host-directed therapies. In this study, we present a time-resolved multi-omics comparison of monkeypox viral infection in primary human cells. Primary human foreskin fibroblast cells were infected with monkeypox virus for 6, 12, and 24 hours and subsequently processed for transcriptome and proteome analysis. Phosphopeptides were enriched using TiO2 beads. Liquid chromatography coupled mass spectrometry-based proteomics measurements were performed in data-dependent acquisition mode and searched using the MaxQuant/Andromeda peptide search engine. We could show that crucial cellular processes are affected at different regulatory layers with distinct temporal profiles during monkeypox infection. To better understand the virus biology and to identify potentially druggable hot spots that the virus relies on, we performed an in-depth systems analysis by projecting the multi-omics data onto the known host biology. Using existing knowledge about host signaling cascades and interactome maps, we extracted infection-elicited molecular fingerprints and applied a graph-based approach to match these with potential anti-monkeypox virus drug targets. A selection of 52 drugs with diverse molecular modes of action was subsequently used in a proof-of-concept drug target validation screen. With this approach, we showcase the clear benefit of integrating multi-omics dataset information to elucidate the molecular patterns of viral infection. These patterns, in turn, can be used to predict potential targets of host- directed antivirals. 691

692

P-II-0539 Uncovering new proteins in Alzheimer's disease patients by quantitative proteomics: blood-based markers and their connection to Amyloid-beta plaques Ana Montero-Calle1, Raquel Coronel2, Vivian de los Rios3, Alberto Peláez-García4, María Jesús Fernandez- Aceñero5, Javier Martínez-Useros6, Isabel Liste2, Marta L. Mendes7, Rodrigo Barderas1,8 1Instituto de Salud Carlos III, Functional Proteomics Unit, Madrid, Spain 2Instituto de Salud Carlos III, Unidad de Regeneración Neural, Madrid, Spain 3Centro de Investigaciones Biológicas, CSIC, Madrid, Spain 4Hospital Universitario La Paz, Madrid, Spain 5Hospital Universitario Clínico San Carlos, Madrid, Spain 6OncoHealth Institute, Fundación Jimenez Diaz, Madrid, Spain 7Luxembourg Institute of Health, Strassen, Luxembourg 8CIBERFES, Madrid, Spain Alzheimer’s disease (AD) is the most common neurodegenerative disease-causing dementia. The disease mechanisms are still unclear, and thus, further research is needed for a better understanding of the pathology associated to the disease and for its early diagnosis. Studying proteins involved in AD pathogenesis can provide insights and identify potential AD markers, as well as therapeutic targets of intervention. Here, we have analyzed protein dysregulation in AD tissues compared to healthy individuals and patients with Vascular Dementia (VD) and Frontotemporal Dementia (FTD)"s tissues to identify key proteins in AD pathogenesis. Two proteomics strategies were used: TMT 10-plex quantitative proteomics using an Orbitrap Q Exactive and label free (LFQ) analysis of a pull-down with in vitro synthesized Aβ fibers using an Orbitrap Astral mass spectrometer. For proteomics analyses, frozen left prefrontal cortex tissue samples from AD, VD, and FTD patients, and healthy individuals were used. TMT 10-Plex quantitative proteomics was performed with AD patients and healthy individuals and VD and FTD patients as controls. For the pull-down assays, in vitro synthesized biotinylated Aβ fibers were incubated with protein extracts from AD patients and healthy individuals. In addition, fibers induced from scramble Aβ peptides were used as negative controls of the assays. Among the 3281 proteins identified by TMT, 15 and 154 proteins were found statistically upregulated and downregulated, respectively, in AD patients compared to controls. After bioinformatics analysis, 10 dysregulated proteins in AD were selected for further validation and to elucidate their role as blood-based biomarkers. Additionally, about 332 proteins were observed as potential interactors of Aβ fibers, with 18 proteins selected for validation. The dysregulation of selected targets in AD patients was confirmed at mRNA and protein level. Importantly, two proteins showed potential as blood-based diagnostic biomarkers of AD, and seven candidate proteins were validated in vitro by WB and ex vivo using FFPE tissue of AD patients and controls and immunofluorescence and immunohistochemistry as novel Aβ plaques interactors, highlighting their role in AD development and progression. 692

693

P-II-0540 Proteogenomic profiling to identify therapeutic targets for acute myeloid leukaemia Heather Murray1, Anoop Enjeti1,2, Joshua Brzozowski1, Kasey Miller1, Jonathan Sillar1,2, Saumya Samaraweera3, Matt Dun1, Richard D'Andrea3, Nicole Verrills1 1The University of Newcastle, Callaghan, Australia 2Calvary Mater Newcastle, Newcastle, Australia 3University of South Australia, Centre for Cancer Biology, Adelaide, Australia Acute myeloid leukaemia (AML) is the most common and fatal acute leukaemia, with a 5-year survival rate of 25%. AML is a heterogenous cancer, with multiple recurring mutations identified, driving activation of complex signalling cascades. Conventional single-agent therapies targeting AML-associated mutations provide only short- term therapeutic benefit. Adding proteomic information to genomic platforms – "proteogenomics" – holds promise to identify druggable pathways. We have expanded this concept to incorporate phosphoproteomic profiling and ex vivo drug screening, enabling identification of activated signalling pathways in individual patients. Aim: To perform multi-omic and drug sensitivity profiling of primary AML patient samples, to identify novel molecular targets for improved therapies. Ex vivo drug sensitivity profiling has been performed on 71 patients. Genomic profiling was performed using the SOPHiA Genetics Myeloid Panel. Transcriptomic and proteomic analysis was performed on bone marrow mononuclear cells from 24 AML patients at diagnosis, and 5 healthy control samples. Paired-end 150bp sequencing was performed on poly-A enriched RNA using an Illumina NextSeq500 Quantitative shotgun proteomics was performed using TMT16 labelling and mass spectrometry analysis. Analysis was performed using Perseus, MoCluster, and Kinase Enrichment Analysis. Results were compared to clinical genomic information. We identified both known and novel associations between drug sensitivities and mutation subgroups. Consistent with previous reports, NPM1 mutations were associated with higher sensitivity to venetoclax. Samples with NPM1 mutation were also significantly more sensitive to Aurora kinase/CDK inhibitor JNJ7706621 compared to NPM1 wildtype. NPM1 has documented roles in cell cycle regulation, and these results suggest that mutant NPM1 may sensitise to cell cycle targeted therapies. Multi-omic analyses of 23 intermediate/high risk patients, 9 of which had no actionable mutations, identified further drug targets. Integrated multi-omic clustering defined 4 molecular subtypes of AML, each with distinct therapeutic vulnerabilities; including DNA repair and metabolic pathways. Proof-of-principle follow-up using in vitro, ex vivo and in vivo preclinical models shows that pharmacological inhibition of DNA-protein kinase combined with standard AML therapy is a novel approach for some AML subtypes including FLT3-mutant AML. Validation of this approach was exemplified by individual case studies where our profiling platform correlated with clinical outcome. The results of these and further case studies will be presented. This study presents a unique pharmaco-phospho-proteogenomic characterisation of AML, providing unprecedented insight into the biology of AML. This is a powerful approach for functional characterisation and therapeutic target identification in AML, both for discovery and precision medicine approaches. 693

694

P-II-0541 Multi-omics characterization of childhood Rhabdomyosarcoma Aya Osama1, Maha Mokhtar1, Asmaa Ramzy1, Taghreed Khaled Abdelmoneim1, Ahmed Karam1, Sameh Magdeldin1 1Children's Cancer Hospital CCHE-57357, Cairo, Egypt Sarcomas are a heterogeneous group of malignant neoplasms of mesenchymal origin. Soft tissue sarcoma (STS) are tumors arising from mesenchymal cells that harbor de-regulation of pathways and chromosomal aberrations governing the morphogenesis of tissue. STS accounts for ~7% of cancers in children and 1% of cancers in adults. About half of the children that have soft tissue sarcoma are with rhabdomyosarcoma (RMS), which is a high- grade, aggressive and highly malignant form of cancer that develops from skeletal (striated) muscle cells and failed to fully differentiate. Two major molecular and histopathological subtypes of RMS, "alveolar" RMS (ARMS) and "embryonic" RMS (ERMS), which are driven by fundamentally different mechanisms. We utilized multi-omics technology to unleash differences between both subtypes. Our findings revealed initially a substantial metabolic and proteomics differences that enables a quick and rapid identification of RM subtypes. The study provides a platform for better predicting the biomarkers of different RMS subgroups based on their omics profiles. 694

695

P-II-0542 Relation between cardiovascular health and Alzheimer: unraveling the role of lipid metabolism and related proteins through DIA-PASEF Laura Mourino-Alvarez1,2, Ángela Peralbo-Molina3, Cristina Juarez-Alia1,2, Tamara Sastre-Oliva1,2, Inés Perales- Sánchez1,2, Cristina María López-Vázquez3, Germán Hernández-Fernández1,2, Felipe Madruga4, Emilio Blanco- López5, Teresa Tejerina6, Luis R Padial7, Eduardo Chicano-Gálvez3, María G. Barderas1,2 1Hospital Nacional de Parapléjicos, SESCAM, Department of Vascular Physiopathology, Toledo, Spain 2Hospital Nacional de Parapléjicos, IDISCAM, Department of Vascular Physiopathology, Toledo, Spain 3IMIBIC, Mass Spectrometry Unit, Córdoba, Spain 4Hospital Virgen del Valle, SESCAM, Department of Geriatrics, Toledo, Spain 5Ciudad Real General University Hospital, Department of Cardiology, Ciudad Real, Spain 6School of Medicine, Universidad Complutense de Madrid, Department of Pharmacology, Madrid, Spain 7Hospital General Universitario de Toledo, SESCAM, Department of Cardiology, Toledo, Spain Dementia of the Alzheimer's type (DAT) is the most common cause of dementia worldwide. Cardiovascular risk factors and established cardiovascular disease (CVD) are known to increase the risk of suffering DAT, although the molecular mechanisms involved in that relationship remain unclear. Here, we set out to define specific proteomic and metabolomic profiles of CVD in patients with DAT to better understand the relationship between these conditions, to unravel the mechanisms underlying the high risk of developing DAT in CVD patients and to define new markers of early phases of this disease. Plasma samples from patients with DAT alone, or patients with DAT and CVD, were analyzed through a multiomics approach that centered on DIA-PASEF technology, and that integrated the proteomics and metabolomics datasets using the OmicsNet web-based tool. ABCA1 and APOH were identified in this way as potential biomarkers for the combined condition, as verified through immunoblotting in an independent cohort with two additional study groups: controls and patients with CVD alone. The metabolomics results showed an enrichment in lipids and lipid-like molecules, especially in fatty acids and sterol lipids. Similarly, the most significant cluster identified in the proteomics study was formed by 5 proteins related to lipoprotein and cholesterol metabolism, including ABCA1 and APOH. After integration and functional enrichment according to the KEGG database, we found glycerolipid metabolism, fatty acid degradation and sphingolipid metabolism to be among the most significant functions. Finally, differential expression of ABCA1 and APOH was verified through immunodetection, with higher levels in patients with DAT and CVD, and positively correlated with phospho-Tau (181), a classical hallmark of DAT. Different molecular profiles appear to exist in patients with DAT, with and without CVD. Alterations can be detected in both diseases but such dysregulation is exacerbated in patients in which DAT and CVD co-exist. This information may help to define new biomarkers like ABCA1 and APOH that identify patients with cardiovascular dysfunction that are at high risk of developing DAT. Such markers will allow more personalized patient interventions to be selected, a further step towards precision medicine for individuals whose molecular profiles might indicate a distinct response to the same management strategies. 695

696

P-II-0543 APOC2 protein as a cornerstone in the development of Aortic stenosis, in patients with Type 2 diabetes mellitus Nerea Corbacho-Alonso1,2, Tamara Sastre-Oliva1,2, Inés Perales-Sánchez1,2, Ángela Peralbo-Molina3, Cristina María López-Vázquez3, Germán Hernández-Fernández1,2, Cristina Juarez-Alia1,2, Emilio Blanco-López2,4, Luis F. López-Almodóvar5, Jorge Solís6,7,8, Luis R Padial9, Teresa Tejerina10, AE López-Jiménez10, Laura Mourino- Alvarez1,2, Eduardo Chicano-Gálvez3, María G. Barderas1,2 1Hospital Nacional de Parapléjicos, SESCAM, Department of Vascular Physiopathology, Toledo, Spain 2Hospital Nacional de Parapléjicos, IDISCAM, Department of Vascular Physiopathology, Toledo, Spain 3IMIBIC, Mass Spectrometry Unit, Córdoba, Spain 4Ciudad Real General University Hospital, Department of Cardiology, Ciudad Real, Spain 5Hospital General Universitario de Toledo, SESCAM, Cardiac Surgery, Toledo, Spain 6Hospital Universitario 12 de Octubre, Instituto de Investigación Sanitaria Hospital 12 de Octubre (imas12), Department of Cardiology, Madrid, Spain 7AtriaClinic, Madrid, Spain 8Instituto de Salud Carlos III, Centro de Investigación Biomédica en Red de Enfermedades Cardiovasculares (CIBERCV), Madrid, Spain 9Hospital General Universitario de Toledo, SESCAM, Department of Cardiology, Toledo, Spain 10School of Medicine, Universidad Complutense de Madrid, Department of Pharmacology, Madrid, Spain Aortic stenosis (AS) and diabetes mellitus (DM) are both progressive diseases that if left untreated, result in significant morbidity and mortality. Several studies revealed that the prevalence of DM is substantially higher in patients with AS and, thus, the progression from mild to severe AS is greater in those patients with DM. Our goal is presenting a method for the treatment, prognosis and/or diagnosis of both diseases. A multi-omic strategy was performed in a cohort of 47 patients with AS and with or without DM. In the discovery phase, we analyzed aortic valve tissue, calcified and non-calcified, by Tandem Mass Tag (TMT) and, by Transcriptome sequencing using RNA-Seq. Then, For the quantitative analysis, the ratio of calcified:non-calcified tissue of each patient was compared. After the statistical analysis of the proteome and transcriptome of the AVs from patients with DAS, with or without DM, the results were compared to evaluate the correlation between them and to identify robust biomarkers for possible clinical use. The APOC2 protein and APOC2 gene gave a p-value < 0.05 and a fold change (FC) > 1.5 in both analyses, and these values moved in the same direction between the study groups Thus, the APOC2 protein was considered the most robust potential biomarker since it was identified in both the proteomic and transcriptomic analysis, and moreover, it was confirmed in plasma samples from an independent cohort of patients when studied by ELISA and turbidimetry. Using both these techniques, the amounts of this protein were higher in the group of diabetic patients relative to the non-diabetic patients (ELISA p-value = 0.018; turbidimetry p-value = 0.038). Therefore, the present study aims to resolve an unmet medical need to achieve a diagnosis of AS in patients with T2DM. 696

697

P-II-0544 Unraveling the protein phosphorylation and glycosylation landscape of diffuse large B-Cell lymphoma using multilayer proteomics and integrative multiomics Sebastian Perner1,2,3, Yanlong Ji1,4, Chih-Hsuan Yeh1,4, Ralf Pflanz1, Sabine Koenig1, Monika Raabe1, Julius Enssle2, Bjoern Haeupl2,5, Thomas Oellerich2,3,5, Kuan-Ting Pan2,3, Henning Urlaub1,4 1Max-Planck-Institute for Multidisciplinary Sciences, Bioanalytical Mass Spectrometry, Goettingen, Germany 2University Hospital Frankfurt a.M., Department of Hematology/Oncology, Frankfurt, Germany 3Frankfurt Cancer Institute, Frankfurt, Germany 4University Medical Center Göttingen, Bioanalytics – Department of Clinical Chemistry, Goettingen, Goettingen, Germany 5German Cancer Consortium (DKTK) and Cancer Research Center (DKFZ), Heidelberg, Germany Introduction: Diffuse large B-cell lymphoma (DLBCL) is the most common type of non-Hodgkin lymphoma encountered clinically. Extensive genomics studies have enabled the stratification of DLBCL subtypes which exhibit significantly different clinical outcomes and responses to the most common treatment regimes. However, the functional consequences of these heterogeneous genetic aberrations on protein dynamics remain largely unknown. We applied multilayer proteomics and integrative multi-omics on DLBCL cell lines to investigate the underlying mechanisms of the disease. Methods: 20 DLBCL cell lines were quantitatively profiled using our streamlined TMT-based pipeline, allowing for multiplexed proteomics, phospho- and glycoproteomics being analyzed from one consecutive sample preparation workflow. Data was analyzed on single layer level (e.g. differential expression and functional enrichment) and afterwards combined in an integrative approach using other layers of omics data (e.g. RNAseq and drug sensitivity) to identify interlayer-correlations. Results: Our data clustered the cell lines into 4 subgroups showing distinct proteomic patterns of 772 proteins total. Two clusters showed significantly regulated proteins in B-cell receptor and NF-kappaB signalling pathways, respectively, in agreement with the well-established DLBCL genetic subtypes. Quantification of 18278 phospho- sites revealed distinct molecular regulations in the cell line clusters and allowed for activity-based prediction of upstream kinases and phosphatases, further stratifying the clusters. Quantification of 7800 glycoforms showed a distinct pattern compared to proteomics-based clustering, establishing three glycosylation-specific clusters exhibiting unique characteristics. The integrative multi-omics approach allowed us to identify additional pathways of interest in the respective clusters and provides further validation for our findings. Conclusion: We established characteristic clusters of DLBCL cell lines with distinct biological and functional features. Integrated multi-omics analysis provides a valuable insight into the underlying mechanisms of different DLBCL subtypes and identifies potential targets for functional analysis guiding towards new treatment approaches, e.g. interfering with critical upstream enzymes of the respective subtypes. Soon, more DLBCL cell lines will be characterized using other mass spectrometry approaches and more in-depth integration is to be established. 697

698

P-II-0545 Precision through Networks: AI-driven Targeting of Protein-Protein Interactions in Lysosomal Storage Disorders Harikrishnan Ramadasan1, Lennart Martens1, Sven Eyckerman1, Sven Degroeve1 1Ghent University, Center for Medical Biotechnology (CMB), Ghent, Belgium interactions Understanding protein-protein is crucial for deciphering cellular processes and diseases. Perturbation data offers insights into phenotypic alterations resulting from modifications to protein interactions, revealing functional relationships between proteins and pathways. These interactions shape cellular phenotypes and influence disease susceptibility or progression. Most protein-protein interactions are non-additive, adding to the complexity of phenotypic variations. Integrating diverse data modalities, such as gene expression profiles, genomic and protein sequences, and biological knowledge graphs, our proposed multi-modal artificial intelligence (AI) model aims to predict cellular responses to perturbation. Our approach utilizes advanced machine learning algorithms to analyze and integrate high-dimensional datasets, enabling the identification of intricate patterns and interactions that traditional methods might overlook. By uncovering complex protein-protein interactions, our approach aims to identify novel therapeutic opportunities, particularly in rare lysosomal storage diseases affecting 1/5000 live births. These diseases often result from defects in lysosomal enzymes, leading to the accumulation of undigested substrates and subsequent cellular dysfunction. Current therapies face significant challenges due to the systemic nature of these conditions, often requiring lifelong treatment. Our multimodal AI framework offers a promising solution by providing a comprehensive understanding of disease mechanisms at the molecular level. This can lead to the development of more effective, targeted therapies that address the root causes of lysosomal storage disorders rather than just managing symptoms. This project addresses the lack of effective therapeutic strategies for such diseases by revealing novel opportunities for intervention advancing targeted therapeutics. 698

699

P-II-0547 NCI's clinical proteomic tumor analysis consortium: analysis tools and resources Ana I. Robles1, Eunkyung An1, Xu Zhang1, Jasmin Bavarva1, Mehdi Mesri1, Tara Hiltke1, Henry Rodriguez1 1National Cancer Institute, Office of Cancer Clinical Proteomics Research, Rockville, MD, United States The National Cancer Institute"s Clinical Proteomic Tumor Analysis Consortium (NCI"s CPTAC) is a national effort to accelerate the understanding of cancer biology through the integration of large-scale proteome and genome analysis, or proteogenomics. The program which is composed of a Tumor Characterization Program and a Translational Research Program (proteomics.cancer.gov) aims to leverage the investments made in cancer research by understanding the molecular changes by proteogenomically characterizing cancers and accelerating the basic discovery toward clinical impact. All data and analytical tools are made broadly available to the research community through public databases to maximize utility and public benefit. CPTAC teams have characterized a plethora of treatment-naïve tumor types, including colorectal, ovarian, breast, clear cell renal cell carcinoma, uterine corpus endometrial carcinoma, lung adenocarcinoma, lung squamous cell carcinoma, brain (including adult, pediatric, and adolescent and young adult), head and neck squamous cell carcinoma, and pancreatic ductal adenocarcinoma (proteomics.cancer.gov/resources/milestones- and-publications). Ongoing projects include acute myeloid leukemia (AML), cutaneous melanoma, sarcoma, oligodendroglioma, gastric, prostate, liver, and thyroid cancers. In the PTRC Program, CPTAC is partnering with NCI-sponsored clinical trials to support clinically relevant research projects that would elucidate biological mechanisms of therapeutic response, resistance, and/or toxicity in cancers including AML, multiple myeloma, melanoma, and lung cancers. All mass spectrometry-based proteomic, genomic, and imaging data (histopathology and radiology) are made publicly available at the CPTAC Proteomic Data Commons (pdc.cancer.gov), Genomic Data Commons (gdc.cancer.gov), and the Cancer Imaging Archive (cancerimagingarchive.net) / Imaging Data Commons (imaging.datacommons.cancer.gov) respectively. CPTAC is also supporting development of new proteogenomic data analysis tools (proteomics.cancer.gov/resources/computational-tools). In addition, the CPTAC Assay Portal (assays.cancer.gov) is a public resource populated with mass spectrometry-based targeted proteomic assays developed by the consortium for quantitatively measuring proteins of interest, including those discovered through comprehensive tumor characterization. Lastly, well-characterized monoclonal antibodies targeting cancer-specific proteins and peptides are also made available at CPTAC"s Antibody Portal (antibodies.cancer.gov). 699

700

P-II-0548 Using established biorepositories for emerging research questions: a feasibility study Sadr Shaheed1, Lente JS Lerink2, Christopher Sutton3, Henny G Otten4, Letizia Lo Faro1, Jan HN Lindeman2, Rutger J Ploeg2,1 1Nuffield Department of Surgical Sciences, University of Oxford, Oxford, United Kingdom 2Department of Surgery, Transplant Centre, Leiden University Medical Centre, Leiden, Netherlands 3Institute of Cancer Therapeutics, University of Bradford, Bradford, United Kingdom 4Central Diagnostic Laboratory/Center for Translational Immunology, Central DiagnUniversity Medical Centre Utrecht, Utrecht, Netherlands Background: Proteomics and metabolomics offer substantial potential for advancing kidney transplant research by providing versatile opportunities for gaining insights into the biomolecular processes occurring in donors, recipients, and grafts. To achieve this, adequate quality and numbers of biological samples are required. Whilst access to donor samples is facilitated by initiatives such as the QUOD biobank, an adequately powered biobank allowing exploration of recipient-related aspects in long-term transplant outcomes is missing. Rich, yet unverified resources of recipient material are the serum repositories present in the immunological laboratories of kidney transplant centers that prospectively collect recipient sera for immunological monitoring. However, it is yet unsure whether these samples are also suitable for -omics applications, since such clinical samples are collected and stored by individual centers using non-uniform protocols, and undergo an undocumented number of freeze-thaw cycles. Whilst these handling and storage aspects may affect individual proteins and metabolites, it was reasoned that incidental handling/storage artifacts will have a limited effect on a theoretical network (pathway) analysis. To test the potential of such long-term stored clinical serum samples for pathway profiling, we submitted these samples to discovery proteomics and metabolomics. Methods: A mass spectrometry-based shotgun discovery approach was used to obtain an overview of proteins and metabolites in clinical serum samples from the immunological laboratories of the Dutch PROCARE consortium. Parallel analyses were performed with material from the strictly protocolized QUOD biobank. Results: Following metabolomics, more than 800 compounds could be identified in both sample groups, of which 163 endogenous metabolites were found in samples from both biorepositories. Proteomics yielded more than 600 proteins in both groups. Despite the higher prevalence of fragments in the clinical, non-uniformly collected samples compared to the biobanked ones (42.5% vs 26.5% of their proteomes, respectively), these fragments could still be connected to their parent proteins. Next, the proteomic and metabolomic profiles were successfully mapped onto theoretical pathways through integrated pathway analysis, which showed significant enrichment of 79 pathways. long-term stored serum samples from clinical Conclusions: This feasibility study demonstrated that biorepositories can be used for qualitative proteomic and metabolomic pathway analysis, a notion with far- reaching implications for all biomedical, long-term outcome-dependent research questions and studies focusing on rare events. 700

701

P-II-0549 Characterisation of redox post-translational modifications upon acute antioxidant treatment in the diabetic heart, using quantitative mass spectrometry Molly Talbot1,2, Alexander Rookyard2,3, Desmond Li3,4, Stuart Cordwell2,3,5, Melanie White1,2 1University of Sydney, School of Medical Science, Camperdown, Australia 2University of Sydney, Charles Perkins Centre, Camperdown, Australia 3University of Sydney, School of Life and Environmental Sciences, Camperdown, Australia 4Heart Research Institute, Newtown, Australia 5University of Sydney, Sydney Mass Spectrometry, Camperdown, Australia Overproduction of reactive oxygen species (ROS) is a key determinant in the pathogenesis of Type 2 Diabetes (T2D) and cardiovascular disease (CVD) linked to obesity, hyperglycaemia, and loss of insulin sensitivity. Cellular redox levels directly reduce antioxidant capacity promoting protein alterations, manifesting as cardiac functional deficiencies. Current pre-clinical evidence reveals the potential for antioxidant therapies to reduce or attenuate ROS derived damage and dysfunction and restore redox balance. However, translation to clinical settings has been unsuccessful thus driving further investigation. We aimed to characterise modified cysteine-containing peptides in response to pathological levels of ROS and potential sites protected by acute antioxidant treatment. Using an 8-week rat model of T2D that combines a low-dose streptozotocin (STZ) and high-fat diet to model the T2D phenotype. Hearts are excised for Langendorff isolated ex vivo perfusion, a system enabling antioxidant intervention. This study featured two subsets, the first received 5-minute perfusion to define native redox balance; and the second received 40-minute perfusion with a global antioxidant, N-propionylglycine (MPG; 1mM). To determine how MPG was affording such protection, we performed quantitative redox proteomics using thiol-disulfide exchange to specifically identify and quantify cysteine-containing peptides using tandem mass spectrometry (MS/MS). In the presence of MPG, contractile performance was maintained in T2D myocardium upon ex vivo perfusion, contrary to our previous studies which showed a gradual decline in function. Percentage maintenance of rate pressure product of the T2D cohort increased by 13% with acute MPG treatment during ex vivo perfusion (84% vs 97%). Proteomics results had 23.48% of proteins in our native redox balance hearts change significantly by ANOVA and z-score. Increasing modified cystines were observed within cardiac antioxidant (e.g. Peroxidasin), mitochondrial, and contractile proteins. We have identified MPG treatment rescues reversibly modified cysteines in endogenous antioxidant mechanisms, mitochondrial proteins, and contractile proteins. Within this dataset 30.22% of proteins were observed significantly changing by ANOVA and z-score. Aforementioned peroxidasin was observed decreasing reversible cysteine modification at cysteine 264 upon MPG exposure, the same site reported in the native state as significantly increasing. Conversely, MPG showed little capacity to protect against redox changes to proteins regulating metabolic pathways. Metabolomics confirmed these observations with dysregulation of TCA intermediates unable to be salvaged by MPG-treatment. This suggests that redox changes in metabolic pathways occur during the development of the T2D phenotype (prior to ex vivo), while contractile dysfunction during ex vivo perfusion is specifically driven by redox alterations in the mitochondria and contractile apparatus. 701

702

P-II-0551 Exploring serine metabolism in Alzheimer's disease and astrocyte differentiation Elisa Maffioli1, Valentina Rabattoni2, Farida Tripodi3, Simona Nonnis1, Zoraide Motta2, Silvia Sacchi2, Anna Maria Rinaldi4, Paola Coccetti3, Loredano Pollegioni2, Gabriella Tedeschi1 1University of Milano, Veterinary Medicine and Animal Science, Milan, Italy 2University of Insubria, Biotechnology and Life Science, Varese, Italy 3University of Milano-Bicocca, Biotechnology and Biosciences, Milan, Italy 4University of Rome “Tor Vergata”, System Medicine, Rome, Italy During aging, the brain experiences metabolic shifts, particularly involving glucose metabolism and insulin signaling. In Alzheimer's disease (AD), there is a known correlation between cerebral hypometabolism and synaptic activity decline. Astrocytes play a pivotal role in this context, as the glycolytic pathway in these cells supports the production of L-serine (L-Ser), necessary for the synthesis of biomolecules crucial for neuronal function, such as glycine (Gly) and D-serine (D-Ser), which act as co-agonists of N-methyl-D-aspartate (NMDA) receptors. In the brain, de novo L-Ser synthesis proceeds via the phosphorylated pathway (PP), involving three key enzymes: 3-phosphoglycerate dehydrogenase (PHGDH), phosphoserine aminotransferase (PSAT), and phosphoserine phosphatase (PSP) (1). Multi-omics investigations comparing hippocampal samples from healthy individuals (CTR) and AD patients revealed metabolic alterations, prominently affecting serine metabolism and disclosing for the first time a gender difference. Increased levels of PHGDH and PSAT in AD were associated with a significant decrease of L- Ser levels and an increase of D-/(D+L)-serine ratio (2). This aligns with prior research linking serine dysfunction to various neurological and psychiatric disorders. Since AD pathology also involves astrocyte impairment, the results prompted us to investigate into astrocyte differentiation and D-Ser's role in astrocyte-central nervous system communication. Multi-omics analysis of NSC-derived human astrocytes underline metabolic changes during astrocytes differentiation and reveal restricted D-Ser synthesis in differentiated astrocytes (3). Moreover, the overexpression of the three enzymes of the PP in hiPSC-derived astrocytes leads to significant alterations not only in serine metabolism pathways but also in folate and nucleotide metabolism and TCA cycle, suggesting a complex metabolic rewiring. This study is aimed to develop new approaches to address brain diseases related to alterations of D-Ser concentration. This work was supported by PRIN 2017 2017H4J3AS-Dissecting serine metabolism in the brain. (1) Murtas et al. (2020) Cell Mol Life Sci. 77(24):5131-5148 (2) Maffioli et al. (2022) Cell Rep. 40(10): 111271 (3) Tripodi et al. (2023) FEBS J. 290(18):4440-4464. 702

703

P-II-0553 Multi-proteomics in bleeding and hemostasis Tirsa van Duijl1, Eva Smit1, Alette Kooiker1, Iris Kreft1, Jessica Del Castillo-Alferez1, Stijn Groten1, Pieter Langerhorst1, Diana Mundo Sandoval1, Carmen van der Zwaan1, Jan Voorberg1,2, Herm-Jan Brinkman1, Joost C. M. Meijers2, Arjan Hoogendijk1, Maartje van den Biggelaar1 1Sanquin Research, Laboratory of Proteomics, Molecular Hematology, Amsterdam, Netherlands 2Amsterdam University Medical Center, Department of experimental Vascular Medicine, Amsterdam, Netherlands Hemostasis is the process to prevent bleeding and encompasses the interplay of endothelial cells in the vessel wall, platelets and plasma proteins. Hemostatic imbalance in one of these components can result in thrombosis or bleeding problems. Mass spectrometry based proteotyping has the potential to address the unmet clinical needs in biology-driven diagnosis of bleeding or thrombotic disorders, prediction of hemostatic events and therapy response monitoring. In our research studies, we use complementary proteomics and peptidomics approaches combined with differential mass spectrometry-based acquisition strategies to unravel molecular aetiologies underlying thrombotic or bleeding episodes. High-resolution Orbitrap MS platforms and TimsTOF MS technology empowered high-throughput profiling of the endothelial, platelet and plasma proteome. Based on this multi-proteomics approaches, we constructed a proteomic landscape of hemostatic disorders, including defects in primary hemostasis (platelet function disorders, Von Willebrand disease and endothelial dysfunction) and secondary hemostasis (coagulation factor deficiencies and bleeding disorders of unknown cause). Proteogenomic analysis of Von Willebrand facilitated Von Willebrand Disease subclassification without a priori genotypic information. Peptidomics on coagulated plasma highlighted in-depth insights on proteolytic events in secondary hemostasis, including zymogen to enzyme transitions and regulation of serpin activity. Finally, within a framework of clinicians, clinical chemists and hemostasis experts, we designed a targeted proteomics assay for absolute quantification of coagulation and fibrinolysis proteins in harmonization with functional hemostatic assays of the clinical chemistry laboratory. In conclusion, our studies highlight the advances in multi-proteomics for molecular footprinting in bleeding and hemostasis. factor proteoforms 703

704

P-II-0554 The development of SDRF-Metaproteomics metadata standards: harmonizing efforts from proteomics and metagenomics Tim Van Den Bossche1, Kay Schallert2, Robert Heyer2, Jean Armengaud3 1Ghent University, CompOmics, Ghent, Belgium 2Leibniz-Institut für Analytische Wissenschaften – ISAS – e.V., Dortmund, Germany 3CEA, Bagnols-sur-Cèze, France Sharing data and resources has fundamentally transformed the life sciences, particularly in the fields of proteomics and metaproteomics. Public data availability has enabled novel reanalyses and interpretations, but the lack of comprehensive metadata limits the potential of these shared resources. The development of the Sample and Data Relationship Format for Proteomics (SDRF-Proteomics), co-developed by EuBIC-MS and HUPO- PSI, aimed to address this, though this format is incomplete for the protein analysis of microbial communities. Therefore, the Metaproteomics Initiative is proposing SDRF-Metaproteomics metadata standards tailored for three commonly studied microbial environments: the human gut, soil, and water. It"s important to note that these standards serve as pilot projects and can be further extended to include other environments as well. Each of these metadata standards contains attributes that are not included in SDRF-Proteomics, but are present in the metadata standards used in metagenomics, commonly known as the minimum information about any (x) sequence (MIxS) specifications developed by the Genomics Standards Consortium. Examples are details about the diet from the patient (human gut), and the depth or elevation of the sampling site (soil and water). We aim to, once approved, integrate these standards in lesSDRF, a tool developed to simplify the process of metadata annotation. To conclude, the introduction of SDRF-Metaproteomics metadata standards marks, to our knowledge, the first attempt of standardizing metadata reporting in metaproteomics. By harmonizing SDRF-Proteomics (what is known but incomplete), with what closely related fields do (MIxS in metagenomics), we address the critical gap in comprehensive data annotation for microbial community protein analysis. The tailored metadata standards for the human gut, soil, and water environments not only enhance data consistency and usability but also pave the way for more accurate and meaningful interpretations and comparisons of metaproteomic data. As these standards gain acceptance and integration through tools like lesSDRF, this will only increase. 704

705

P-II-0555 Multi-omics profiling of bone regeneration in diabetes reveals association of increased mast cell activity with healing impairment Vivien Wiltzsch1, Daniela B. Dias2, Johannes R. Schmidt1, Klaudia Adamowicz3, Wing Lee Chan2, Tanja Laske3, Jan Baumbach3,4, Jörg Lehmann1,5, Jennifer A. Kirwan6,7, Patrina S.P. Poh2, Stefan Kalkhof1,8 1Department Preclinical Development and Validation, Fraunhofer Institute for Cell Therapy and Immunology IZI, Leipzig, Germany 2Julius Wolff Institute, Berlin Institute of Health at Charité - Universitätsmedizin Berlin, Berlin, Germany 3Institute for Computational Systems Biology, University of Hamburg, Hamburg, Germany 4Computational Biomedicine Lab, Department of Mathematics and Computer Science, University of Southern Denmark, Odense, Denmark 5Fraunhofer Cluster of Excellence Immune-Mediated Diseases CIMD, Frankfurt/Main, Hannover, Leipzig, Germany 6Metabolomics Platform, Berlin Institute of Health at Charité − Universitätsmedizin Berlin, Berlin, Germany 7Max-Delbrück-Center for Molecular Medicine in the Helmholtz Association, Berlin, Germany 8Institute for Bioanalysis, University of Applied Science Coburg, Coburg, Germany Introduction: Conventional treatments for critical-sized bone defects are often insufficient in individuals comorbid with Type 2 diabetes mellitus(T2DM) due to their increased risk of prolonged healing and complications. To optimize personalized treatments based on bioactive and resorbable biomaterial-guided bone regeneration, a comprehensive understanding of the molecular mechanisms involved is essential. This study investigates the progression of diabetic bone healing over 42 days through imaging and multi-omics analyses of tissue as well as plasma proteomics for systemic evaluation. Methods: We developed the first sequential multi-omics extraction protocol for bone tissue and regenerating callus. Explants of biomaterial-supported regenerated and contralateral tissue after 21- or 42-days post- implantation from a diabetic rat model with a critical-size femur defect were extracted successively for their metabolome and proteome. In addition to analysis of central carbon metabolites using GC-MS, quantitative targeted metabolomics was performed using a mixture of flow injection analysis and LC-MS applying the Biocrates Quant MxP500 kit. TMT-based LC-MS/MS proteomic analyses of the same explants were conducted on a quadrupole-Orbitrap MS. Differential expression profiles of proteins and metabolites were determined by pairwise comparisons, providing a holistic view of molecular changes occurring during the healing process. Finally, results were integrated with imaging information from µCT scans and histological data to correlate molecular findings with clinical outcomes. Results: Employing this novel multi-omics extraction protocol on regenerating bone callus enabled the quantification of about 4000 proteins and over 500 metabolites, thus providing an in-depth perspective on bone healing under diabetic conditions. Integration of multi-omics with histological data revealed several key insights, including decreased levels of structural proteins essential for soft callus formation, a disturbed metabolism considering main energy production pathways (glycolysis, TCA cycle), and extended inflammatory processes. Furthermore, we discovered important links highlighting an imbalanced population of mast cells (MC) within the regenerating tissue of diabetic individuals. This imbalance was characterized by clusters of MC proteases, elevated levels of the MC mediator histamine, and increased MC presence, rendering MCs a potential target population to address in compromised diabetic bone healing. Conclusion: This study demonstrates the performance of a new multi-omics extraction protocol for bone, providing critical insights into molecular disruptions in diabetic bone regeneration. It underscores the importance of multi-omics approaches for deep molecular profiling to understand T2DM's negative impact on bone healing. Our ongoing preclinical study will evaluate interventions targeting the observed dysregulated mast cell activity to enhance bone regeneration. 705

706

P-II-0556 Microfluidics coupled with mass spectrometry-based multi-omics for antimicrobial resistance analysis Dongxue Zhang1, Liang Qiao1 1Fudan University, Department of Chemistry, Shanghai, China Bacterial antimicrobial resistance (AMR) is one of the most significant challenges to current human society. Exposing bacteria to antibiotics can activate their self-saving responses, e.g., filamentation, leading to the development of bacterial AMR. Understanding the molecular changes during the self-saving responses can understand the mechanism of AMR and reveal new inhibition methods of drug-resistant bacteria. In this abstract, we integrated microfluidics with mass spectrometry-based omics to investigate the mechanism of bacterial AMR during bacterial filamentation when bacterial cells were treated with antibiotics. First, we used a gradient microfluidic chip to observe the morphological changes of bacteria under the impact of antibiotics. Differential proteins were laser desorption/ionization time-of-flight mass spectrometry, and further validated by transcriptome. Combined proteomics and transcriptomic results, three potential protein biomarkers were identified, including fimbrial subunit type 3, penicillin-binding protein activator LpoB, and 30S ribosomal protein S14 from carbapenem-resistant Klebsiella pneumoniae. identified by matrix assisted We also used an online microfluidics mass spectrometry system for real-time characterization of metabolic changes of bacteria during filamentation under the stimulus of antibiotics. The proteome of bacteria in the chip was also analyzed, and significant pathways, e.g., nucleotide metabolism and coenzyme A biosynthesis, from extended-spectrum beta-lactamase-producing Escherichia coli (ESBL-E. coli) were comprehensively discovered to related to the bacterial filamentation. We further developed a method to inhibit antibiotic-resistant bacteria by combining traditional antibiotics and chemical inhibitors against the enzymes involved in the bacterial self-saving responses. Besides, we developed single bacterial cell metabolic profiling by mass spectrometry to study bacterial AMR at single-cell level. By utilizing a microprobe controlled by a microoperation platform, single filamentous ESBL-E. coli cells stimulated by antibiotic can be extracted and spray-ionized for mass spectrometry analysis. Heterogeneous among ESBL-E. coli cells under the same antibiotic stimulus condition was observed from mass spectra as well as cell morphology. The metabolic profiles by mass spectrometry of different individual cells can be clustered into subgroups well in accordance with bacterial cell length. In summary, we combined microfluidics with proteomics, metabolomics, and to comprehensively investigate the mechanism of AMR, discovery biomarkers, as well as develop new inhibition method of antibiotic-resistant bacteria. transcriptomics 706

707

P-II-0557 Multi-omics analysis reveals mechanosensitive alterations in cardiac Metabolome during pathological remodeling Hong Lin1, Ling Lin1, Yunzeng Zou1 1Zhongshan Hospital, Fudan university, Shanghai, China Mechanical overload is a major trigger of pathological cardiac remodeling (PCR), ultimately leading to heart failure. Understanding how cardiomyocytes respond to mechanical force signals and alter cellular states is crucial for developing treatments for PCR. However, the mechanosensitive alterations in myocardial proteome and metabolome involved in PCR remain unclear. In this study, we established a cell model of mechanical stress-induced cardiac remodeling by applying cyclic mechanical stretch to adult mouse primary cardiomyocytes (AMCM). By adjusting different mechanical stretch parameters, we induced both physiological and pathological cardiac remodeling states. Excessive mechanical stress resulted in sarcomere disorganization, mitochondrial damage, and lipid accumulation in AMCM. We employed a multi-omics approach—including untargeted proteomics, metabolomics, and lipidomics—utilizing mass spectrometry to analyze the effects of varying mechanical stretch parameters on the mechanotransductive response of cardiomyocytes. Comparative proteomic data between AMCM and myocardial cell line AC16 revealed for the first time that the high abundance of structural proteins (such as actin and myosin) as well as mitochondrial-related proteins (such as ATP synthase and citrate synthase) in adult primary cardiomyocytes significantly inhibited the detection of low-abundance proteins. Consequently, using timsTOF Pro 2 mass spectrometry platform, 4,885 proteins were identified in AMCM, which was significantly fewer than 7,790 proteins identified in AC16. Further quantification of 432 differentially expressed proteins from AMCM suggested a decrease in protein vesicle transport capacity, mitochondrial respiratory chain damage, and inhibition of oxidative phosphorylation with increasing mechanical stress. Using the AlteredPQR algorithm for membrane protein interaction network analysis, we revealed the activation of mechanosensitive ion channels such as Piezo and TRPC, leading to an imbalance in cardiac calcium homeostasis. Integrating metabolomics and lipidomics data from stretched AMCM under different mechanical parameters, we observed aberrant lipid metabolism associated with amino acid metabolism remodeling. Construction of protein- metabolite interaction networks suggested that mechanical stress mediated myocardial lipid deposition, enhanced glutamine decomposition metabolism, increased arginine synthesis, and reduced alanine synthesis metabolism. These changes were possibly related to alterations in cell membrane composition, calcium homeostasis imbalance, and mitochondrial oxidative stress. In summary, our findings reveal that mechanical stress alters cell membrane lipid composition and activates cardiac mechanosensitive ion channels, leading to calcium homeostasis imbalance, amino acid metabolism remodeling, inhibition of mitochondrial oxidative phosphorylation, ultimately contributing to pathological cardiac remodeling. 707

709

P-II-0558 Spatially resolved proteogenomics on human melanoma samples reveals evolutionary trajectories in tumor progression Maximilian Zwiebel1, Thierry M. Nordmann1, Maria Wahle1, Katrin Kerl-French2, Rudolf Stadler3, Reinhard Dummer4, Matthias Mann1 1Max-Planck-Institute of Biochemistry, Proteomics and Signal Transduction, Martinsried, Germany 2Ludwig-Maximilians-University Munich, Department of Dermatology and Allergology, Munich, Germany 3Johannes Wesling Klinikum Minden, Ruhr University Bochum, Department of Dermatopathology, Minden, Germany 4University Hospital Zürich (UHZ), Deparment of Dermatology, Zurich, Switzerland Although melanoma is less common than other types of skin cancer, its aggressiveness and strong ability to metastasize render it a particularly dangerous neoplastic disease. The molecular evolution from benign precursor lesions to malignant melanoma is a complex and dynamic process, driven by the acquisition of genetic alterations and leading to increased progressivity and invasiveness. While the underlying genetic changes of this transformation have been investigated extensively over the past decades, pathophysiological details of tumor progression and metastasis on a phenotypic level are less well understood. To this end, we set out to apply our recently developed workflow for spatially resolved cell-type specific proteogenomics based on Deep Visual Proteomics (DVP) on a cohort of microscopic formalin-fixed paraffin-embedded (FFPE) sections of melanoma patients. Our aim was to relate genomic mutations to protein levels, providing novel insights to metastatic cancer evolution. We retrospectively collected FFPE biopsies from melanoma patients displaying at least two different disease progression stages and performed automated immunofluorescence staining for SOX10 and PRAME using an Agilent Dako Omnis platform. AI-driven image classification identified distinct pathological disease stages. Based on this, we excised shapes of similar phenotypic identity using a Leica LMD7 laser microscope and separated DNA and peptides on Evotips, aiming to acquire fully matched genomic and proteomic data sets. We also conducted Next Generation Sequencing (NGS) on an Illumina NovaSeq 6000 and mass spectrometry (MS)-based proteomics on a Thermo Fisher Orbitrap Astral. Our workflow yielded high-quality whole exome sequences, permitting robust variant calling, and enabled the identification of more than 5000 proteins per single cell shape type on average. We predicted the consequences at the protein level by creating patient-specific FASTA files containing the genetic mutations, increasing the identification of proteoforms by more than 10% compared to the standard UniProt search file. This strategy allowed us to quantify the differential expression of mutated alleles on protein level and molecularly distinguish developmental stages of melanoma. We further conducted gene set enrichment analysis on the proteomic data and related genetic changes to the activation of pathways involved in cell proliferation and metastasis. Applying our workflow on a clinical melanoma cohort serves as first demonstration of the benefits of integrating deep spatially resolved genomics and proteomics data directly in single cell types in vivo. Specifically, spatial proteogenomics will enable us to investigate trajectories of clonal cancer cell evolution. More broadly, our interdisciplinary technology promises to reveal unique insights into the mechanisms of tumor pathogenesis and beyond. 709

711

Structural proteomics (P-II-0559 to P-II-0594) P-II-0559 LiP-MS guided drug discovery using the cancer driver Ras as a model system Foroughsadat Absar1, Brandon Novy2, Evgeniy V. Petrotchenko3, Konstantin I. Popov2, Roopa Thapar4, Jason B. Cross4, Christoph H. Borchers3,5,6,7 1McGill University, Lady Davis Institute, Experimental Medicine, Montreal, Canada 2Division of Chemical Biology and Medicinal Chemistry, University of North Carolina at Chapel Hill, Chapel Hill, North Carolina, Chapel Hill, NC, United States 3Jewish General Hospital, Segal Cancer Proteomics Centre, Lady Davis Institute for Medical Research, Montreal, Canada 4M.D. Anderson Cancer Center, Institute for Applied Cancer Science, Therapeutics Discovery Division, Houston, Texas, USA, Houston, TX, United States 5Jewish General Hospital, Gerald Bronfman Department of Oncology, Montreal, Canada 6McGill University, Division of Experimental Medicine, Montreal, Canada 7McGill University, Department of Pathology, Montreal, Canada Limited proteolysis combined with mass spectrometry (LiP-MS) is an informative technique for studying changes in protein structure upon ligand binding. LiP-MS utilizes proteases to cleave proteins under native conditions, revealing a structure-specific proteolytic pattern. This method distinguishes between free and drug-bound states by the variations in susceptibility to proteolysis, where small molecule binding stabilizes the protein structure, resulting in fewer exposed cleavage sites. Here, we have characterized the interaction of drug candidates with G12D mutant KRas protein by LiP-MS. We used intact protein mass spectrometry and top-down analysis to detect and identify KRas cleavage products, respectively, and therefore to locate the cleavage site in the protein sequence. Responsive to the drug binding, cleavage sites corresponded to the location of the drug binding site. Limited proteolysis yield can be estimated quantitatively using intact protein mass spectrometry and we found that the degree of cleavage depends on the binding affinity and presence of particular functional groups in the drug molecule structure. The approach was applicable to characterizing the drug binding with affinities in the micromolar range. This makes it potentially applicable for the screening of the drug leads libraries. Molecular dynamics simulations showed that the responsive to the drug binding cleavage sites were located in the disordered regions of the protein structure, which were stabilized upon drug binding. It also sheds light on the contribution of the drug molecule functional groups to the stabilization of the protein structure and consequently to the degree of cleavage. The approach can be used for the identification of the drug binding sites, quantitative characterization of the binding, and drug design. 711

712

P-II-0561 Aggrescan4D: structure-informed analysis of pH-dependent protein aggregation Oriol Bárcenas1, Aleksander Kuriata2, Mateusz Zalewski2, Valentín Iglesias1,3, Carlos Pintado-Grima1, Grzegorz Firlik2, Michał Burdukiewicz1,3, Sebastian Kmiecik2, Salvador Ventura1 1Autonomous University of Barcelona, Biochemistry and Molecular Biology, Bellaterra, Spain 2University of Warsaw, Faculty of Chemistry, Warsaw, Poland 3Medical University of Białystok, Białystok, Poland Protein aggregation is a multifactorial process, directed by the intrinsic properties of proteins and heavily influenced by environmental factors. This phenomenon accounts for the onset of highly debilitating proteinopathies, which pose a significant challenge to human health. Additionally, it imposes constraints on developing and implementing protein-based biotechnological and biomedical applications. Classical computational methods that predict protein aggregation rely on sequential information to assess the propensity of a protein to aggregate. While highlighting APRs provides information on the regions driving aggregation, this strategy assumes that the proximity between residues is directly proportional to their sequential distance. This presumption might apply to disordered or denatured protein states but is an imprecise approximation for the native state of globular proteins. Moreover, these methods treat the aggregation contribution of amino acids residing in collapsed or inaccessible regions (such as the hydrophobic core) equally to exposed residues. To overcome these limitations, we developed a three-dimensional structure aggregation predictor, Aggrescan3D (A3D), which projects classical Aggrescan scoring to proteins" native structure. Over the years, we have been advancing the A3D method, aiming to deepen our comprehension of protein aggregation and assist the engineering of protein solubility. Since its inception, A3D has become one of the most popular structure-based aggregation predictors because of its performance, modular functionalities, RESTful service for extensive screenings, and intuitive user interface. Expanding on our previous work, we introduce Aggrescan4D (A4D), a tool that significantly enhances A3D"s capabilities. A4D is designed to predict pH-dependent protein structure aggregation and incorporates an evolutionary-informed automatic mutation protocol for engineering protein solubility and maintaining structure and stability. It also integrates precalculated results from the A3D Model Organisms Database, covering nearly 500,000 jobs, and allows structure retrieval from the AlphaFold database. Globally, A4D is a comprehensive tool for understanding, predicting, and addressing specific protein aggregation challenges. The A4D web server and extensive documentation are accessible at https://biocomp.chem.uw.edu.pl/a4d/. Fig. 1 (optional) 712

713

P-II-0563 Structural analysis of a plant glucosyl ceramidase (GBA2)-like enzyme from arabidopsis thaliana by cryogenic electron microscopy (CRYO-EM) Sunaree Choknud1, James Ketudat Cairns1, Chamaipon Beagbandee1, Puey Ounjai2, Wasinee Rungsarityotin3 1Suranaree University of Technology, Nakhon Ratchasima, Thailand 2Mahidol University, Biology, Bangkok, Thailand 3Mahidol University, Molecular Biosciences, Nakhon Pathom, Thailand Glucosylceramide (GlcCer) is a common glycosphingolipid, that accumulates in cells in response to Gaucher disease, diabetes, and skin disorders in human but it is not well understood in plants. In animals, GlcCer is catabolized by glucosylcerebrosidase 1 and 2 (GBA1 and GBA2). GBA1 is a lysosomal enzyme in glycoside hydrolase (GH) family GH30, while GBA2 is a membrane-associated cytoplasmic protein in family GH116. Currently, there are no experimental structures of eukaryotic GH116 homologues. Although the bacterial TxGH116 β-glucosidase structure was determined by x-ray crystallography, it does not hydrolyze glucosylceramides, unlike the animal and plant enzymes. Therefore, we have investigated the structure of plant GH116 (At4GH116) by Cryo-EM single particle analysis. The recombinant At4GH116 protein was produced in E. coli with a 6x His-tag and purified by immobilized metal affinity chromatography followed by S200 size exclusion chromatography. Cryo-EM conducted at Mahidol University, Thailand, enabled the capture of detailed images of At4GH116, indicating strong potential for high-resolution structure determination. Initial map generation indicated a unique hexameric structure, consistent with previous high resolution size exclusion chromatography analysis. Solving the first eukaryotic structure of a GH116 enzyme will generate a template for improved modeling of human GBA2, with implications for treating human diseases, such as Gaucher disease and hereditary spastic paraplegia. 713

714

P-II-0564 Natural inhibitors of snake venom toxins: a promising source of therapeutic peptides for toxin-centric next- generation antivenoms Yago Inocencio Vieira de Pinho1, Barbara da Silva Soares1, Surza Lucia Gonçalves da Rocha1, Teresa Escalante2, Borries Demeler3, Richard Hemmi Valente1, Francisco Gomes-Neto1, Ana Gisele da Costa Neves Ferreira1 1Oswaldo Cruz Institute – Fiocruz, Laboratory of Toxinology, Rio de Janeiro, Brazil 2Codomiro Picado Institute, San José, Costa Rica 3University of Lethbridge, Department of Chemistry and Biochemistry, Lethbridge, Canada analytical ultracentrifugation. By 2030, the World Health Organization aims to halve the global rates of death and disability caused by snake envenomation. To meet this challenge, a comprehensive strategy has been recently devised, which includes the need to accelerate research into new therapies. In this context, the use of effective inhibitors of tissue-damaging toxins [phospholipase A2 (PLA2) and metalloendopeptidase (SVMP)] constitutes an attractive adjunctive therapy. Circulating antitoxins observed in animals naturally resistant to snake envenomation represent a rich source of molecular scaffolds for developing novel peptide-based therapeutics. This study aimed to characterize the interaction between the anti-hemorrhagic protein DM43 from the opossum Didelphis aurita (SisGen AF0A111, CEUA/IOC L-035/2018) and the SVMP BaP1 from Bothrops asper venom, using an integrated Structural Biology approach. The stoichiometry of the interaction between DM43 and BaP1 was analyzed using sedimentation velocity BS3 [(bis(sulfosuccinimidyl)suberate], and all cross-linked peptides (intra- and inter-proteins) were identified by nLC- MS/MS on the QExactive HF-X mass spectrometer, using the SIM-XL software, followed by manual validation. The five best DM43 models generated by AlphaFold2 were evaluated according to the level of global validation of the cross-links, using the Topolink software to measure the topological distances between the cross-linked residues identified (Lys or Ser). Molecular dynamics simulation strategies were also applied to evaluate the stability of the models and the rate of cross-linking validation as a function of time. DM43 formed a non-covalent equimolar complex with BaP1, inhibiting its proteolytic and hemorrhagic activities. DM43 structural models revealed a low validation of intra-DM43 cross-links, indicating that the computational models do not accurately reflect the conformational structure of the inhibitor in solution. Most inter-protein links observed involved the third domain of DM43 and the regions of BaP1 between residues 60-80 and 150-170. These distance constraints will be used to guide the molecular docking of the crystal structure of BaP1 with an optimized model of DM43. With these strategies, we intend to topologically characterize the toxin-antitoxin complex, defining the regions of DM43 potentially involved in its anti-ophidic properties. stabilized with The toxin-antitoxin complex was 714

715

P-II-0565 Optimized XL-MS workflows for heterobifunctional crosslinkers SDA and DizSEC Yi He1, Erum Raja2, Leigh Foster2, Anthony Ciancone3, Anneliese Faustino4, Piyoosh Sharma4, Stephen Fried4,5, Francis O’Reilly3, Ryan Bomgarden2, Rosa Viner1 1Thermo Fisher Scientific, San Jose, CA, United States 2Thermo Fisher Scientific, Rockford, IL, United States 3NIH, National Cancer Institute, Frederick, MD, United States 4Johns Hopkins University, Department of Chemistry, Baltimore, MD, United States 5Johns Hopkins University, Department of Biophysics, Baltimore, MD, United States Introduction: Cross-linking mass spectrometry (XL-MS) has grown dramatically as a key workflow for elucidating protein higher-order structure and mapping protein-protein or protein-nucleic acid interaction networks on a proteome-wide scale. Photoactivatable crosslinkers SDA and DizSEC (cleavable) link lysine with any residue in proteins, providing comprehensive information about protein structures. However, challenges remain in both instrument methods and data analysis for these crosslinkers. In this work, we optimized XL-MS workflows for SDA and DizSEC crosslinked samples and developed a new MS2-MS3 method to improve DizSEC XL identifications. Methods: Heterobifunctional photoactivatable -crosslinkers SDA (succinimidyl 4,4'-azipentanoate), sulfo-SDA (sulfosuccinimidyl 4,4'-azipentanoate) and DizSEC (2-(3-methyl-3H-diazirin-3- yl)ethyl)carbamate) were used to crosslink standard proteins. Separation was achieved using Thermo Scientific™ Vanquish™ Neo LC system with a 60 min gradient using EASY-Spray™ PepMap™ Neo column. Following separation, the peptides were detected on Thermo Scientific™ Orbitrap Ascend™ or Exploris 480 ™ mass spectrometers, and data were analyzed using XlinkX node in Proteome Discoverer v3.1 (Thermo Scientific) and xiSEARCH/xiFDR. Identified XLs were visualized using the XMAS plug-in for ChimeraX. (2,5-Dioxopyrrolidin-1-yl Results: Photoactivatable, heterobifunctional crosslinkers like SDA can link lysine residues with any residue providing detailed information for protein structures but also pose challenges due to lack of MS-cleavability and increased search space during data analysis. Recently, a new cleavable crosslinker, DizSEC, was developed to overcome this issue. DizSEC links the same residues as SDA, but DizSEC crosslinked peptides can generate unique peptide pairs in MS/MS to facilitate data analysis. We compared three photo-crosslinkers (SDA, sulfo-SDA, and DizSEC) using monomeric, dimeric, and tetrameric standard proteins. We developed three FAIMS- MS methods: MS2 SCE, MS2 SCE prioritizing charge states +4 to +6, and a new MS2-MS3 method for DizSEC. Compared to the standard MS2 SCE method, the instrument method prioritizing high charge states increased the number of crosslinks by 30-50%. When using the MS2-MS3 workflow for DizSEC, the instrument identifies peptide pairs with a 26 Da difference in MS/MS and selects them for MS3 fragmentation. The MS2-MS3 method generated similar numbers of crosslinks to the standard MS2 SCE method but improved FDR. For data analysis, we successfully created the new DizSEC templates in Proteome Discoverer 3.1 for both MS2 and MS2-MS3 data analysis. Three different SDA crosslinkers provided a similar number of crosslinked peptides for monomeric protein HSA; however, SDA and sulfo-SDA outperformed DizSEC for multimeric proteins probably due to the length of the linker. Novel aspect: Optimized XL-MS workflows for SDA and the new cleavable photo-crosslinker DizSEC 715

716

P-II-0566 Protein–RNA crosslinking combined with MS becomes quantitative at the peptide and amino-acid level Yi He1, Nicolaie Eugen Damoc2, Luisa Welp3,4, Olexandr Dubkov3, Alexander Chernev3, Timo Sachsenberg5,6, Monika Raabe3, Arslan Siraj5, Oliver Kohlbacher5,6,7, Henning Urlaub3,4, Rosa Viner1, Daniel Hermanson1 1Thermo Fisher Scientific, San Jose, CA, United States 2Thermo Fisher Scientific, Bremen, Germany 3Max Planck Institute for Multidisciplinary Sciences, Bioanalytical Mass Spectrometry Group, Goettingen, Germany 4University Medical Center Göttingen, Bioanalytics, Group Institute of Clinical Chemistry, Goettingen, Germany 5University of Tübingen, Institute of Bioinformatics and Medical Informatics, Tuebingen, Germany 6University of Tübingen, Applied Bioinformatics, Dept. of Computer Science, Tuebingen, Germany 7University Hospital Tübingen, Institute of Translational Bioinformatics, Tuebingen, Germany Mass Spectrometry can resolve UV- and chemically induced protein-RNA crosslinking at the peptide and amino- acid level. Our search engine, NuXL, improves the identification of protein-RNA UV/chemical crosslinks at amino- acid resolution, irrespective of sample complexity (isolated complexes or cellular entities). However, the quantification of crosslinks remains a challenge for the sensitive and precise detection of the abundance changes among peptides crosslinked under different cellular conditions or states. UV crosslinks various amino acids to mainly uracil, whereas chemical crosslinking connects specific amino acids to guanosine and adenosine. Accurate MS-based quantification requires monitoring of the same crosslinked species derived from different states, and this can be hampered by changes in crosslinking patterns. Here, we explored a label-free quantification (LFQ) approach to analyze UV and chemical crosslinks using the NuXL node in Thermo Scientific™ Proteome Discoverer™ software. We used the E. coli ribosome to monitor UV- or chemically-induced protein-RNA crosslinking quantitatively by MS. Different amounts of E. coli ribosomes were exposed to UV or chemical crosslinkers and enriched using different reagents. After the sample preparation, samples were separated using a Thermo Scientific™ Vanquish™ Neo LC system with a 60 min gradient using an IonOpticks Aurora Ultimate™ (75 um x 25 cm) column. The yield of crosslinked peptide-RNA oligonucleotides was measured by their intensities in MS1 (label-free) using a Thermo Scientific™ Orbitrap™ Astral™ mass spectrometer. The NuXL node in Protoeome Discoverer 3.1 software, based on the OpenMS software package, was used for analyzing XLMS data from UV or chemically crosslinked protein–RNA samples. In this study, we extend the use of this tool to include the quantitative analysis of crosslinks. We investigated the feasibility of quantitatively analyzing UV or chemical crosslinking with different amounts of E. coli ribosomes. Although quantitative label- free MS is generally straightforward, it is challenged here by a final TiO2 enrichment step before LC-MS analysis during sample-processing of the crosslinked peptide-RNA oligonucleotide species. We evaluated and annotated the quantitative label-free results with respect to (i) the background of non-crosslinked species in the input, i.e., before the final enrichment step; (ii) non-bound peptides after TiO2 enrichment, and (iii) residual non- crosslinked peptide species in the TiO2-enriched sample, in order to optimize UV or chemical crosslinking sample preparation and enrichment protocols. We were able to identify and quantify more than 360 UV and 1400 chemically crosslinked peptides for over 300 unique peptide sequences. In conclusion, label-free quantification of crosslinked peptide-RNA oligonucleotides can be achieved accurately and reproducibly using the NuXL node embedded in Proteome Discoverer 3.1 software for complex samples. 716

717

P-II-0567 The regulation role of host cell calmodulin in late phase of retroviral life cycle Petra Junková1, Jan Prchal2, Markéta Častorálová2, Karolína Burešová2, Tomáš Ruml2, Martin Hubálek1 1Institute of Organic Chemistry and Biochemistry of the CAS, Prague, Czech Republic 2UCT Prague, Department of Biochemistry and Microbiology, Prague, Czech Republic Retroviruses comprise a group of viruses characterized by the use of reverse transcriptase to convert their genome ssRNA to DNA in host cells. The mechanisms that retroviruses use in host cells to proliferate are similar to all retrovirus species, but not identical. Enormous efforts have been devoted to uncovering the mechanisms that the human immunodeficiency virus (HIV) uses in individual parts of its life cycle. Nevertheless, there are still unclear processes in the life cycle of these viruses, for example the exact mechanism of the interaction of viral structural polyprotein Gag with the cytoplasmic membrane or the way of regulation of virus maturation. To better understand the just-mentioned processes in the life cycle of retroviruses, we focused on the interaction between the matrix protein (MA) of Mason-Pfizer monkey virus (M-PMV) and host cell calmodulin (CAM). Matrix protein is N-terminal domain of structural polyprotein Gag that mediates the binding of Gag to cytoplasmatic membrane prior to budding of newly formed viral particles from host cell and their maturation. The impact of the interaction between MA and human CAM for the binding of retroviral precursor Gag to the cytoplasmic membrane has already been reported for HIV. However, the structure of the MA-CAM complex has not been described yet. In our study, we found out that the presence of CAM increases the interaction of M- PMV myrMAPPHis construct (myristoylated N-terminal part of Gag polyprotein with preserved cleavage site for viral protease) to cytoplasmatic membrane. Our data also shows that the oligomerization of M-PMV myrMAPPHis as well as its tendency to be cleaved by viral protease are more effective in the presence of calmodulin. These findings lead us to hypothesize that CAM interaction with MA stimulate the conformation changes of MA to enhance the aforementioned processes and this way serves as the regulation factor in the retroviral life cycle. To confirm our hypothesis, we applied protein cross-linking that enabled us to confirm the direct interaction between M-PMV myrMAPPHis and human CAM together with revelation the protein regions which are in the complex in close proximity. Interestingly, the N-terminal and C-terminal parts of MA were identified to be in close proximity to N-terminal region of human CAM. On the N-terminus of MA there is post-translationally bound myristic acid which is supposed to play a role in the interaction with the membrane through the myristic switch mechanism and on C-terminal part the specific cleavage site for viral protease occur. 717

718

P-II-0568 A novel cross-linking strategy for aromatic residues Michael Karpisek1,2, Lukas Fojtik1,2, Zdenek Kukacka1, Petr Novák1,2 1Institute of Microbiology of the Czech Academy of Sciences, Prague, Czech Republic 2Faculty of Science, Charles University, Prague, Czech Republic Chemical cross-linking in combination with mass spectrometry (CXMS) has been developed into a powerful method for mapping protein structures, dynamics and interaction networks including molecular interfaces in protein-protein and protein-nucleic acid complexes. Although many cross-linkers have been developed in last two decades, majority of CXMS analyses still utilizes lysine-specific cross-linking reagents based on N- hydroxysuccinimide esters. Other cross-linking reagents have only limited use for various reasons such as low reactivity, low occurrence of targeted residues, unwanted side products etc. In this study we show a new generation of cross-linkers based on recently published Fast FluoroAlkylation of Proteins (FFAP) technology that enables targeting of aromatic amino acids residues. Cross-linking reaction is based on two step FFAP mechanism which includes activation of hypervalent iodine using metal ions or lewis acid and subsequent attack of aromatic residues by the resulting radical. The studied proteins were analyzed by bottom up approach using mass spectrometry (timsTOF SCP, Bruker Daltonics) where samples were digested by trypsin protease, separated on reverse phase column online coupled to mass spectrometer. The results on several protein models such as horse heart myoglobin clearly demonstrate the potential of our novel cross-linking strategy. The cross-links of aromatic residues generated by fluoroalkyl radicals nicely correspond with protein crystal structures of studied protein and provide information not attainable by lysine- specific cross-links. Our data lead to assumption the cross-linker based on fluoroalkyl radicals could be use as fast and powerful method to study protein structure and dynamics in solution. This work was mainly supported by the Czech Science Foundation (grant number 22-27695S) and European Commission H2020 (The European Proteomics Infrastructure Consortium providing access EPIC-XS – project agreement No. 823839). 718

719

P-II-0569 Fast FluoroAlkylation of Proteins (FFAP) as a new tool for characterization of protein therapeutics Zdenek Kukacka1, Lukas Fojtik1,2, Zuzana Kalaninova1,2, Jan Fiala1,2, Petr Halada1, Josef Chmelik1,2, Petr Man1,2, Petr Novák1,2 1Institute of Microbiology of the Czech Academy of Sciences, Prague, Czech Republic 2Faculty of Science, Charles University in Prague, Prague, Czech Republic Covalent labeling combined with mass spectrometry is a robust method for studying protein structure, interactions, and dynamics. Recently, a new technique called Fast FluoroAlkylation of Proteins (FFAP) has expanded the toolbox of covalent labeling techniques. FFAP is a novel protein labeling method that uses fluoroalkyl radicals generated from hypervalent Togni reagents. These radicals target aromatic aminoacids and cystein and are beneficial for structural characterization of monoclonal antibodies (and biosimilars) as well as epitope mapping. The Togni reagents are decomposed by ascorbic acid to form fluoroalkyl radicals, which then label the solvent- accessible aromatic amino acid side chains of both Trastuzumab and the human HER2 receptor, alone and in complex, for a short duration using a quench-flow microcapillary apparatus. Following this, Trastuzumab and the HER2 receptor are reduced, alkylated, and digested by trypsin/Lys-C. The resulting peptides are separated on a reversed-phase column (Luna Omega Polar C18, 100Å, Phenomenex) in a water-acetonitrile gradient (Agilent 1290, Agilent Technologies) and analyzed by high-resolution mass spectrometry (solariX XR 15T, Bruker Daltonics). The extent of modification is determined from the intensity of the first isotope at the peak maximum using DataAnalysis software (Bruker Daltonics). Initially, different batches of Trastuzumab (a monoclonal anti-HER2 antibody marketed as Herceptin) were labeled with fluoroalkyl radicals, revealing consistent solvent-accessible regions in both batches, particularly in the Fab fragment including the antigen-binding sites. These results align with the X-ray structural model (1N8Z). Further testing of our novel fluoroalkyl radical technology for epitope mapping on the Trastuzumab-HER2 complex identified differences in the modification of free proteins and the complex, uncovering residues at the interaction interface that correspond with previous structural analyses. Additionally, the data revealed a minor structural rearrangement of the HER2 receptor upon Trastuzumab binding, explained by the remodeling of the β-sheet structure into an unstructured loop, a behavior previously observed in rat HER2 (1N8Y) and the human HER2 receptor–Pertuzumab complex. This work was mainly supported by the Czech Science Foundation (22-27695S), NPO-NEURO-EXCELLES (LX22NPO5107) and by European Commission H2020 (EPIC-XS grant agreement ID: 823839; EU FT-ICR MS grant agreement ID: 731077), and the Grant Agency of Charles University (grant number 359221) and, in part, by the Czech Academy of Sciences (RVO61388971). We also acknowledge the Structural Mass Spectrometry Core Facility of CIISB, Instruct-CZ Centre, supported by MEYS CR (LM2018127) and European Regional Development Fund Project "UP CIISB" (No. CZ.02.1.01/0.0/0.0/18_046/ 0015974). 719

720

P-II-0570 Quantitative assessment of protein structural changes and drug binding affinities using limited proteolysis mass spectrometry Namgil Lee1,2, Hojin Yoo1, Heejung Yang1,3 1Bionsight, Seoul, South Korea 2Kangwon National University, Information Statistics, Chuncheon, South Korea 3Kangwon National University, Department of Pharmacy, Chuncheon, South Korea Protein structure is subject to alteration through interactions with various molecules, including protein–protein interactions, aggregation, and binding to other molecular entities. Limited proteolysis coupled with mass spectrometry (LiP-MS) has emerged as a powerful technique for accurately quantifying these structural changes at the peptide level in response to such perturbations. This method enables the simultaneous probing of structural changes across thousands of proteins within a sample. Despite its potential, there is limited validation to confirm whether the quantitative structural changes observed through LiP-MS consistently correspond with these perturbations. In this study, we investigate the relationship between protein structural changes under drug treatment and the binding affinities of proteins for the drug. More precisely, we demonstrate that the effective dose in dose- response curves derived from LiP-MS data exhibits a linear correlation with drug-protein binding affinity. Moreover, we propose conditions for selecting peptides that show a high linear correlation to binding affinity. The linear relationship is validated through extensive LiP-MS data analysis using various kinase inhibitors with known binding affinities and multiple cancer cell lines. This work shows the utility of LiP-MS not only as a tool for mapping structural changes in proteins but also as a quantitative approach for assessing drug-protein interactions. By establishing a linear correlation between LiP- MS-derived dose-response data and binding affinities, our findings provide a framework to utilize LiP-MS in drug discovery and development. 720

721

P-II-0571 Fast photochemical oxidation of protein: DIA workflow and the impact of the peptide identification framework Marek Zakopcanik1,2, Daniel Kavan1, Petr Novák1, Dmitry Loginov1 1Institute of Microbiology CAS, Prague, Czech Republic 2Charles University, Department of Biochemistry, Faculty of Science, Prague, Czech Republic Methods of structural mass spectrometry facilitate structural studies of protein folding and interaction. Fast photochemical oxidation of protein (FPOP) utilizes •OH for the protein labelling occurring on a microsecond timescale and possess a broad reactivity. Accurate interpretation of MSMS spectra is a key point for the FPOP analysis. This task is solved by using different search engines which differ in technique for spectrum identification and scoring systems, providing unique identifications in addition to set of mutual identifications. In this study we compared the reliability of peptide identifications by two different search engines and evaluated the perspectives of using spectral libraries in combination with DIA acquisition to increase the precision of FPOP modification assignment. A comparison of the search engines Mascot and PEAKS X+ was conducted using the data set PXD021621 (ProteomeXchange). The results demonstrated that the choice of search engine significantly affects the assignment of FPOP modifications to detected peptides, with only 31% of identifications common to both engines. PCA revealed distinct clusters of modified peptides based on score, site probabilities, and intensities. Mascot showed more confident site determination, with a clear correlation between fragmentation spectrum quality and assigned site confidence. Consequently, Mascot required less manual validation due to the majority of identifications belonging to a reliable top cluster. In contrast, PEAKS ID clustering lacked a clear connection to identification reliability, necessitating extensive manual validation. We propose a novel approach that combines validated spectral libraries with DIA data. First, DDA data searched by FragPipe are validated using Skyline software to form a spectral library. This library is then matched against the DIA data to filter out non-representative IDs. Out of 421 accepted identifications, 346 contained FPOP modifications, primarily oxidations and dioxidations. Comparing the manually validated library with Mascot search results revealed that the library offers a wider range of unique modifications, increased spatial resolution, and better coverage of biologically relevant areas of the protein complex. For the first time, a DIA workflow was used for FPOP data with a manually validated spectral library. This work was supported by Czech Scientific Foundation (22-27695S), Technology Agency of the Czech Republic (ODEEP-TH86010001) Photomachines (CZ.02.01.01/00/22_008/0004624). Programme Comenius and by Johannes Amos 721

722

P-II-0572 Structural proteomics of human serum proteins Petr Man1, Dmitry Loginov1, Pavla Vankova2, Petr Pompach2, Petr Novák1 1Institute of Microbiology, Czech Academy of Sciences, BioCeV, Prague, Czech Republic 2Institute of Biotechnology, Czech Academy of Sciences, BioCeV, Vestec, Czech Republic Hydrogen/deuterium exchange mass spectrometry is a well-established tool of structural biology, providing dynamic insights into the functioning of isolated protein systems. One of the often-mentioned advantages is the "no size limit" feature, allowing to study larger protein assemblies. Here we employed HDX-MS to get detailed insight into the heterogeneous complex assemblies of haptoglobin and follow its interaction with hemoglobin. Haptoglobin is one of the abundant plasma proteins and is responsible for protection again oxidative stress arising from freely circulating hemoglobin. It exists in three major phenotypes, depending on the combination of alleles 1 and 2 (1-1, 2-1 or 2-2). The phenotype 1-1 forms dimers, while 2-1 and 2-2 are characterized by higher order oligomers (dimer, trimer up to at least pentadecamer) with different architecture. The existing high- resolution structures however used only the homogeneous, fully dimeric 1-1 phenotype and employed its stabilization via addition of other interacting proteins. Here all three phenotypes were studied in the free form in solution and their interaction with hemoglobin was followed for various molar ratios. Next, the role of N- glycosylation was probed and the whole study was taken to proteome wide level by following the interaction in crude as well as albumin/Ig depleted serum. Once the serum HDX was established, interactions of additional serum proteins with small molecule ligands were followed and compared with the experiments done on isolated proteins. The present study demonstrates how HDX-MS can be taken to more complex levels and how it provides insights into the protein interactions on the level of individuals. Financial support from the MEYS/EU OP JAK project Talking microbes - CZ.02.01.01/00/22_008/0004597 is gratefully acknowledged. 722

723

P-II-0573 Unveiling hidden protein dynamics: a multi-proteomic approach to study new players in metastatic melanoma progression Alessandra Morelli1, Francesco Farris1, Vittoria Matafora1, Angela Bachi1 1IFOM, Milan, Italy In our recent study, we uncovered amyloid-related proteins and fibrillar aggregates in the extracellular environment of metastatic melanoma cells. Intriguingly, there is only a known physiological amyloidogenic process in humans occurring in melanocytes, where BACE2 processes PMEL protein into amyloidogenic peptides to create a melanin scaffold. This process seems deregulated in metastatic melanoma, leading to aberrant fibril deposition. BACE2 is upregulated in many tumors, and our findings showed that inhibiting its activity in metastatic melanoma cells decreased cell proliferation and aggregate deposition.The Limited Proteolysis coupled to Mass Spectrometry (LiP-MS) approach was employed to analyze the metastatic melanoma secretome following BACE2 inhibition. This approach permitted the study of protein aggregation in the biological environment, highlighting conformational perturbations. BACE2 is a sheddase located both intracellularly and at the plasma membrane and regulates many proteolytic processes, potentially causing aberrant shedding of many transmembrane proteins. The N-Tails degradomics approach assessed the global proteolytic process mediated by BACE2. LiP-MS secretome analysis revealed a conformational perturbation in four known amyloidogenic proteins, including PMEL. Conformational rearranged peptides were found in two domains within the amyloidogenic core, and N-Tails showed cleavages close to these domains. This suggested that amyloidogenic secreted proteins were in a fibrillar conformation. Interestingly, recombinant PMEL amyloid fibrils added to the extracellular environment promoted migration and invasion in metastatic melanoma cells, indicating that deregulated fibrillation supports tumor progression. Through N-Tails, we observed other transmembrane proteins affected by BACE2-related cleavage, and we are now characterizing these targets in relation to their activity in metastatic melanoma. These findings demonstrate that BACE2 significantly rearranges the extracellular environment by influencing the secretion of pro-tumoral fibrils. Moreover, these data open up the possibility of decoding strategies employed by the tumor to regulate cellular processes through BACE2 cleavage. Using a multi-proteomic approach, we gained insights into post-translational and conformational modifications of proteins, beyond the abundance perturbations detected by shotgun proteomics, thus facilitating our understanding of underlying biological mechanisms. Fig. 1 (optional) 723

724

P-II-0574 Trapped-ion mobility spectrometry with PASEF for analysis of RNA-protein crosslinks in Escherichia coli ribosomes and HeLa cell line nucleic extract Sergey Moshkovskiy1, Olexandr Dybkov1, Timo Sachsenberg2, Ralf Pflanz1, Monika Raabe1, Vadim Demichev3, Oliver Kohlbacher2,4, Henning Urlaub1,5 1Max Planck Institute for Multidisciplinary Sciences, Bioanalytical Mass Spectrometry, Goettingen, Germany 2University of Tübingen, Tuebingen, Germany 3Charité – Universitätsmedizin Berlin, Berlin, Germany 4University Hospital Tübingen, Tuebingen, Germany 5University Medical Center Göttingen, Goettingen, Germany Covalent crosslinking through UV exposure or chemical agents is a method used to analyze the interaction between nucleic acids and proteins. This method involves RNA-protein crosslinking, followed by the analysis of the resulting nucleotide-peptide crosslinks using high-resolution mass spectrometry. It can be applied on a proteome-wide scale to discover new RNA-binding proteins, or on isolated macromolecular complexes to complement unresolved structures determined by other structural methods. However, reliably identifying and quantifying nucleotide-peptide crosslinks using mass spectrometry remains challenging. We demonstrate the use of trapped-ion mobility spectrometry in the timsTOF mass spectrometer, first in the dda-PASEF mode to characterize ion mobility of the crosslinks, and then in the dia-PASEF mode with a customized spectral library to quantify the crosslinks. We conducted crosslinking experiments using E. coli ribosomes and HeLa nucleic extract, with both samples crosslinked using UV 254 nm and ribosomes additionally using nitrogen mustard chemical (NM). The samples were treated with nucleases and trypsin, and the resulting crosslinked nucleotide peptides were enriched using titanium dioxide beads. Crosslinks were identified from dda-PASEF data using OpenNuXL software. Spectral libraries were created for ribosome results, and dia-PASEF data were analyzed using DIA-NN. Collisional cross-sections of identified crosslinks were generally lower than those of peptides of the same m/z, most likely due to the more compact structures of the crosslinks. Additionally, it was found that mono- or dinucleotide additions consistently increased the collisional cross-sections (CCS) compared to corresponding linear peptides, with the presence of uridine monophosphate leading to a decrease in CCS of crosslinks. To conduct quantitative analysis, we used different UV exposure times and NM concentrations to model various crosslink concentrations. We quantified 39 UV crosslinks at all five irradiation time points ranging from 30 seconds to 10 minutes, compared to 112 crosslinks in the 10-minute treatment. In addition, we quantified 463 linear peptides at all time points. The relative intensity of crosslinks returned by DIA-NN followed a logarithmic function with saturation close to the 5-minute exposure point. Similar results were obtained for NM crosslinks. When using NM concentrations of 0.1, 1, 5, and 10 mM, we quantified significant numbers of crosslinks, with 223 species compared to 314 peptides at all four points. In both treatments, crosslinks exhibited high fitness and steep slopes in the linearized logarithmic regression curves, unlike peptides. This quantitative analysis of RNA- protein crosslinks using data-independent acquisition opens up new possibilities for system biology analysis in relevant biological conditions and models, such as viral infections and translation-targeting drug treatments. 724

726

P-II-0575 Determination of M-PMV matrix protein-calmodulin interaction interface using hydrogen deuterium exchange mass spectrometry Tereza Nesporova1, Petra Junková2, Jakub Sýs2, Jan Prchal1, Martin Hubálek2, Tomáš Ruml1 1University of Chemistry and Technology, Department of Biochemistry and Microbiology, Prague, Czech Republic 2Institute of Organic Chemistry and Biochemistry of the CAS, Prague, Czech Republic Determining the structure of protein complexes is crucial for a deeper understanding of their function in biological systems. Therefore, specifying the interacting interface of complexes that forms retroviral proteins with host cell machinery is essential for understanding the individual steps in the life cycle of retroviruses. Traditional techniques including X-ray crystallography, cryo-electron microscopy and nuclear magnetic resonance (NMR) that are commonly used to determine protein structure are often limited by sample quantity, complex size and, not least, laboratory instrumentation. Hydrogen deuterium exchange mass spectrometry (HDX-MS) is an elegant complementary method to study the protein interaction interface. For single proteins and protein complexes, HDX-MS can be used to obtain information on structure, protein-protein or protein- ligand interaction sites, allosteric effects and conformational changes induced by posttranslational modifications. HDX-MS has the added advantage of not being limited by the size of proteins or protein complexes, is highly sensitive and is able to detect coexisting protein conformations. Calmodulin (CAM) is a small, ubiquitous, and evolutionarily highly conserved protein regulating basic cellular functions in all eukaryotes. As well known, calmodulin interacts with structural components of retroviruses in their life cycle, namely with N-terminal domain of Gag structural polyprotein termed matrix protein (MA), which is responsible for many key steps in interaction of the retroviral particle with the cytoplasmatic membrane within the infected cell. However, the mechanism of interaction still remains unclear. For deeper clarification the MA-CAM interaction interface, myrMAPPHis construct (myristoylated N-terminal part of Gag polyprotein with preserved cleavage site for viral protease) of the Mason-Pfizer monkey virus (M-PMV) was used. The structure of the MA-CAM complex is been extensively studied in our laboratory group by NMR, cross linking mass spectrometry and also by HDX-MS. The data obtained by HDX-MS indicate strongly influence of calmodulin to the dynamics of several parts of myrMAPPHis, especially by changing the availability of residues in the region of the second and third helix and the loop between them. Moreover, the residues in these regions are also significant for interaction of MA with phospholipids on the membrane. Furthermore, the data also show that calmodulin also affects the availability of residues in the viral protease cleavage site region of the fifth helix. According to the results, the interaction of myrMAPPHis with calmodulin and subsequently with phospholipids on the membrane may be the right stimulus for the myristoyl switch leading to the reorganization of the myrMAPPHis structure and its cleavage from the polyprotein precursor Gag as part of the maturation process. 726

727

P-II-0576 Investigation of protein dynamics in immortalized mouse macrophages stimulated by lipopolysaccharide (LPS) using limited proteolysis-coupled mass spectrometry Aleksandra Nita-Lazar1, Doeun Kim1 1National Institute of Allergy and Infectious Diseases, NIH, Laboratory of Immune System Biology, Bethesda, MD, United States TLR4 is expressed on the endosomal membranes of innate immune cells, acting as one of the sensors of danger signals. Upon TLR4 activation, signaling cascades are initiated through MyD88-dependent and TRIF-dependent mechanisms. These pathways induce the expression of proinflammatory cytokines and type 1 interferons. However, molecular mechanisms controlling the formation, interaction, and disassembly of protein complexes remain unclear. To figure it out, we have optimized limited proteolysis-coupled mass spectrometry (LiP-MS) using quantitative proteomics analysis, following a time-course LPS stimulation in immortalized mouse macrophages and subcellular fractionation into cytoplasm and nuclear. Conformational changes in protein dynamics were observed mainly in the cytoplasmic fraction after 1 hour of LPS stimulation. Within the TLR signaling pathway, inhibitor of nuclear factor kappa-B kinase subunit beta (IKKb) displayed significant conformational changes. In the protease degradation assay, IKKb and Elongator complex protein 1 (Elp1) showed a degradation pattern change between the control and LPS stimulation groups. Elp1 is predicted to act as a scaffold protein that assembles active IKK-MAP3K14 complexes based on protein sequence similarity. Our research indicates that IKKb may be involved in so far unidentified interactions with uncharacterized binding partners, investigating. This research was supported by the Division of Intramural Research of NIAID, NIH. immunity which we are currently indicating novel mechanisms of innate 727

728

P-II-0577 A novel opportunities for the structural characterization of the biomolecular assemblies Petr Novák1, Michael Karpisek1, Lukas Fojtik1, Jasmína Mária Portašiková1, Petr Man1, Zuzana Kalaninova1, Zdenek Kukacka1 1Institute of Microbiology, Prague, Czech Republic Introduction: Covalent labeling and chemical cross-linking in combination with mass spectrometry is a powerful approach for structural biologists to approach proteins" structure, interaction and dynamic. Recently, the toolbox of covalent labeling techniques has been expanded by Fast FluoroAlkylation of Proteins (FFAP). FFAP is a novel protein labeling method relying on fluoroalkyl radicals generated from hypervalent Togni reagents. These reactive species target Cysteine, Histidine, Phenylalanine, Tryptophane and Tyrosine and are beneficial for structural characterization of proteins and protein complexes. Methods: The mono- or bi-valent Togni reagents are decomposed by the ascorbic acid at the native environment to form a fluoroalkyl radicals that decorate the solvent accessible aromatic and sulfhydryl amino acid side chains of any protein at a short time period utilizing quench-flow microcapillary apparatus. Further, the protein samples are reduced, alkylated and digested by trypsin/Lys-C. Resulting peptides are separated on reversed-phase column (Luna Omega Polar C18, 100Å, Phenomenex) in water-acetonitrile gradient (Agilent 1290, Agilent Technologies) and directly analyzed by high resolution mass spectrometry (solariX XR 15T or timsTOF SCP, Bruker Daltonics). For mono-valent probe, the regular and modified peptides are identified by FragPipe (v.20.0) and the extent of modification is calculated from the intensity of the first isotope at the maximum of the chromatographic peak using DataAnalysis software (Bruker Daltonics). In case of bi-valent probe, the cross- linked peptides are fished out from the complex fragment spectra thanks to MeroX 2.0. Results: First, the mono- and bi-valent probes shed a light on complex formation of human Hemoglobin- Haptoglobin complex. Second, mono-valent radicals were successfully tested for epitope mapping on the complex of Trastuzumab with Her2. The differences in the modification of free proteins and in the complex uncovered residues located on interaction interface that nicely correspond with previous structural analysis. Moreover, these data revealed a minor structural rearrangement of HER2 receptor upon Trastuzumab binding which is explained by remodeling of β-sheet structure to unstructured loop. Similar behavior was already reported for rat Her2 (1N8Y) and human HER2 receptor – Pertuzumab complex. Further, the FFAP was subsequently used for the study of the structural differences between E. coli chloride channel (CLC-ec1) and its QQQ mutant. Results showed a critical connection between protonation and structural changes in the ion transport path and conformational change in the transmembrane helixes. 728

729

P-II-0578 Novel, specific protein structural biomarkers for Alzheimer's disease and dementia with Lewy bodies Anna Pagotto1, Luise Nagel2, Jan-Philipp Quast1, Wilma D J van de Berg3, Charlotte E Teunissen3, Natalie de Souza1,4, Andreas Beyer2, Paola Picotti1 1ETH Zürich, Biology – Institute of Molecular Systems Biology, Zurich, Switzerland 2University of Cologne, Cologne, Germany 3Amsterdam University Medical Center, Amsterdam, Netherlands 4University of Zürich, Zurich, Switzerland Neurodegenerative diseases are increasingly prevalent in aging human societies, causing significant suffering and a growing socio-economic burden. Despite having distinct underlying molecular mechanisms, these diseases still overlap in signs, symptoms and pathological features. This hinders accurate diagnosis, negatively affecting disease management, patient enrolment in clinical trials, and development of specific therapies. Robust and disease-specific biomarkers are essential tools for improving the diagnosis and gaining better insights into the molecular pathways involved in these pathologies. While biomarkers for Alzheimer's disease (AD) (e.g., Aβ42/Aβ40 ratio, total and phosphorylated Tau levels) are currently used in diagnostics, reliable biomarkers for Parkinson"s disease (PD) are still missing. For example, the use of the oligomeric:total alpha-synuclein ratio, a PD hallmark, is impeded by inconsistencies across studies and lack of specificity. The situation is even worse for Dementia with Lewy Bodies (DLB), an under-diagnosed disorder that shares features with AD and PD, and for which no biomarkers are available. In a previous work (1), we have shown that the global analysis of protein structures in cerebrospinal fluid (CSF) of PD patients constitutes a novel, orthogonal readout for the disease state. Protein structural information could indeed better discriminate between PD patients and healthy individuals than protein abundance information, the typical readout for protein-based biomarker screens. Furthermore, the newly identified protein structural biomarkers were linked to disease mechanisms and provided complementary information to the gold standard alpha-synuclein. In this study, we extended this approach to the study of AD and DLB, aiming to assess the specificity of our previously identified CSF biomarkers for PD and to discover novel structural biomarker specific to AD and DLB. We used Limited Proteolysis coupled to Mass Spectrometry (LiP-MS) to generate structural fingerprints of proteins in the CSF of 50 individuals with AD, 50 with DLB and 50 with subjective cognitive decline (SCD) as a reference. After correcting for covariates, we are now performing a pair-wise comparison to identify structurally altered proteins in the CSF of AD and DLB patients. I will present the results of these analyses, the identified candidate biomarkers and discuss the emerging concept of structural disease biomarkers and its impact in the study of pathological processes. 1. Mackmull, Marie-Therese et al. "Global, in situ analysis of the structural proteome in individuals with Parkinson's disease to identify a new class of biomarker." Nature structural & molecular biologyvol. 29,10 (2022): 978-989. doi:10.1038/s41594-022-00837-0 729

730

P-II-0579 In silico identification of potential inhibitors against Galectin-3-Binding Protein: A potent Key factor for glioma progression Rashmi Rana1 1Sir Ganga Ram Hospital, Biotechnology and Research, New Delhi, India Glioma account for the majority of the brain tumors worldwide. Glioblastoma is one of the most lethal type of Gliomas. The treatment of the brain tumor is greatly dependent on the stage of its detection. Galactin-3-binding protein (LGALS3BP) protein serves as a novel circulatory biomarker for the detection of glioma at an early stage. This protein is also responsible for metastasis, proliferative signalling, angiogenesis and immune system evasion in case of brain tumors. Inhibition of LGALS3BP can help in reduction of metastasis and progression of the disease. Currently there are no drugs available which can completely treat glioma, hence in this study we have virtually screened large number of drugs from database of National Cancer Institute, USA against LGALS3BP followed by induced fit docking and molecular dynamics simulations. We found that on the basis of the binding free energy calculations using MMPBSA three compounds 627861 (-16.69 kcal/mol), 329090 (-13.66 kcal/mol), 627855 (-10.01 kcal/mol) were best when compared with the co-crystallised ligand taken as control (-9.58 kcal/mol). After extensive computational analysis we can conclude that these three molecules 627861, 329090 and 627855 can be the potential inhibitors of LGAL3SBP and form the basis of the treatment of the brain tumor. The structural scaffolds of these molecules can also lead to the optimization of better compounds against LGALS3BP in future. Although In-vitro and In-vivo studies are required before concluding any potential beneficial effects of these compounds against the glioma Fig. 1 (optional) 730

731

P-II-0580 Use of singlet oxygen to analyze protein structure Jan Rasl1,2, Marek Polák1,2, Lukas Fojtik1,2, Jasmína Mária Portašiková1,2, Daniel Kavan1,2, Petr Man1,2, Michael Volný1,2, Petr Novák1,2 1Institute of Microbiology of the CAS, Prague, Czech Republic 2Faculty of Science, Charles University, Prague, Czech Republic Proteins must undergo various types of conformational changes to fulfill their biological function. Current mass spectrometry techniques are used in structural biology and can provide a variety of information about protein structure, conformational changes or specific parts of the protein involved in interaction with other biomolecules. The idea behind the integration of mass spectrometry into structural techniques is based on the labeling of the protein by a suitable chemical probe. The structure, the accessibility of the reaction sites and other factors determine the modification of the protein by the probe. These labeling methods are usually referred to as "protein footprinting" and reflect the protein structure. The rapid reactivity of the probing agent modifies the parts of the protein accessible to the solvent and records the molecular "footprint". The probing agent can be bound to the protein either unstably (as in the hydrogen- deuterium exchange technique) or stably (e.g. by radical modification of the protein in the Fast Photochemical Oxidation of Proteins technique). In this work we show that singlet oxygen (1O2) is a suitable covalent labeling agent. Singlet oxygen is generated under mild conditions with a photosensitizer and red light of 650 nm. The photosensitizer is excited by the red light and generates singlet oxygen, which immediately reacts with the protein under investigation, preferably with the amino acid residues Cys, His, Met, Trp and Tyr. We investigated various conditions of the labeling process, including buffer composition, specificity and modification of the target residues, and tested the usefulness of the data for structural mass spectrometry. 731

732

P-II-0582 A multi-proteomics approach identifies non-catalytic effects of ATP-competitive kinase inhibition Viviane Reber1, Sabrina Keller1, Stefanie Loosli1, Paola Picotti1, Matthias Gstaiger1 1ETH Zürich, Institute of Molecular Systems Biology, Zurich, Switzerland Kinases are essential regulators of most cellular processes, and their activity is tightly controlled. Aberrant kinase function can contribute to the development and progression of various diseases. Small molecule inhibitors offer a means to correct this dysregulated kinase activity. Type I inhibitors are the largest group of kinase inhibitors and target the ATP-binding pocket of kinases in an ATP-competitive manner. While this mode of action has been well studied, it is largely unclear how these inhibitors could affect non-catalytic kinase functions by remodeling the structural dynamics and subsequent complex formation of kinases. Here we present a proteomics workflow based on affinity purification (AP) and limited proteolysis (LiP) coupled with mass spectrometry (MS) to probe changes in complex formation and kinase structure induced by these compounds. We show that inhibition with SGC-CAMKK2-1 induces CAMKK2 complex formation with known and candidate substrates, including the metabolic regulator AMPK, typically associated with activation of CAMKK2. To understand the mechanisms underlying inhibitor induced protein sequestration, we explored structural changes upon inhibition of CAMKK2 with the inhibitors SGC-CAMKK2-1 and staurosporine using AP-LiP-MS. Both inhibitors induced allosteric changes at the autoinhibitory domain, which is consistent with a relief of autoinhibition upon inhibitor binding. We propose that the inhibitor locks the kinase in a pseudoactive conformation that resembles the active, non- autoinhibited conformation, promoting stable complex formation with substrates and other proteins, potentially altering their activity. We observed similar structural changes upon inhibition of the related kinases CHEK1, DCLK1, CAMK1 and PKCα, suggesting a more general mechanism underlying Type I inhibition of kinases with autoinhibitory domains. We propose that holistic characterization of the effects of Type I inhibition can reveal crucial new insights into the molecular mechanisms that drive cellular phenotypes of existing and future kinase therapeutics beyond inhibition of kinase mediated substrate phosphorylation, to allow future expansion of the activities that could be targeted by ATP-competitive kinase inhibitors. 732

733

P-II-0583 Deep proteomics reveals incorporation of unedited proteins into mitochondrial protein complexes in Arabidopsis Nils Rugen1, Michael Senkler1, Hans-Peter Braun1 1Hannover Leibniz University, Plant Proteomics, Hannover, Germany The mitochondrial proteome consists of numerous types of proteins that are either encoded and synthesized in the mitochondria, or encoded in the nucleus, synthesized in the cytoplasm and imported into the mitochondria. Their mitochondrial synthesis depends on editing of primary transcripts at specific sites. This process, known as "RNA editing", is mainly caused by deamination of cytidine to uridine and can lead to changes in the amino acid sequences of the encoded proteins. Since RNA editing occurs at different frequencies at these sites, partially edited transcripts could also be used for translation. Indeed, such transcripts have been shown to be loaded onto mitoribosomes. Translation of partially edited transcripts could lead to protein polymorphisms that affect protein activity or protein-protein interactions. However, the effects of differential RNA editing have not been systematically investigated at the proteome level. Overall, mitochondrial RNA editing suggested to be a repair process that compensates for defective sequences within mitochondrial genes(1). Here, we have comprehensively characterized the Arabidopsis mitochondrial proteome to analyze its editing status. The vast majority of mitochondrially encoded proteins are derived from edited mRNAs, suggesting that complete maturation of the transcripts is a prerequisite for the biosynthesis of the corresponding proteins. However, we report exceptions for subunits of the ATP synthase complex and mitoribosomes. Furthermore, we present strong evidence for the incorporation of these incompletely edited proteins into their corresponding protein complexes (2). The results of this project as well as preliminary results from mutant studies will be presented and discussed. (1) Small, I. D., Schallenberg-Rüdinger, M., Takenaka, M., Mireau, H. & Ostersetzer-Biran,O. (2020) Plant organellar RNA editing: what 30 years of research has revealed. Plant Journal.$ (2) Rugen, N., Senkler, M. & Braun, H.-P. (2024) Deep proteomics reveals incorporation of unedited proteins into mitochondrial protein complexes in Arabidopsis. Plant Physiology., in press. 733

734

P-II-0585 Comparison of different glycoproteomics software of glycoproteomics including GRable, a new web-based software using a MS1-based glycopeptide detection method Hiroaki Sakaue1, Chiaki Nagai-Okatani1, Azusa Tomioka1, Hiroyuki Kaji2, Atsushi Kuno1 1National Institute of Advanced Industrial Science and Technology, Cellular and Molecular Biotechnology Research Institute, Tsukuba, Japan 2Nagoya University, Institute for Glyco-core Research, Nagoya, Japan Glycosylation is a major post-translational modification of proteins, which plays a vital role in cell-cell interactions and signal transduction. Since the cell surface is covered with glycans and their structures vary greatly depending on the cell type and condition, disease-related alteration in protein glycosylation is expected to be a pathologically relevant biomarker. Recent technological advances in large-scale analysis of glycopeptides, i.e., glycoproteomics using mass spectrometry (MS), have enabled us to discover unique candidates. In particular, software development for MS-based glycoproteomics allows us to determine the detailed glycan structures on proteins. However, there is no standardized way of thinking for evaluating which assignment is truly "identified". To address this issue, we compared commercial (Byonic) and open-resource software (MSFragger-glyco[1], pGlyco3[2], Glyco-Decipher[3], GRable[4]) to evaluate the differences in glycopeptides identified by each software, using a cultured cell line derived from lymphoma (BJAB cells) as a crude glycoprotein sample. In this study, proteins were extracted from BJAB cells using phase transfer surfactant (PTS) buffer, and approximately 1 mg of the resulting protein was reduced and alkylated. The protein solution was diluted 5-fold and then treated with trypsin at 1/100 of the protein weight at 37°C overnight. Trifluoroacetic acid was then added to a final concentration of 1% to precipitate and remove PTS, and the glycopeptides were purified using an amide column. One-tenth of the sample was used for LC-MS analysis. The obtained data were analyzed using commercial and open resource software, and the differences in glycopeptides identified by each software were evaluated. Among all 2,002 identified glycopeptides, 236 glycopeptides were commonly identified by all software, supporting their plausibility. On the other hand, a total of 799 glycopeptides were identified by only one software, indicating that many glycopeptides are uniquely identified by each analysis software. In this presentation, we will discuss how to evaluate the certainty of these solely found assignments. References [1] Polasky et. al., Nat Methods, 17, 1125-1132 (2020), [2] Zeng et. al., Nat Methods, 18,1515-1523 (2021), [3] Fang et. al. Nat. Commun., 7 (2022), [4] Nagai-Okatani et. al.,BioRxiv, (2024) 734

735

P-II-0586 Photo-crosslinked peptide-RNA hybrids with tunable length map protein-RNA interfaces across the proteome and transcriptome Shuyao Sha1, Jakob Trendel1, Bernhard Küster1 1Technical University of Munich, Chair of Proteomics and Bioanalytics, Freising, Germany Protein-RNA interactions are crucial for gene expression regulation, with their interface providing key structural insights into the underlying mechanisms. Photo-crosslinking can be applied to generate covalently connected peptide-RNA hybrids at the protein-RNA interaction interface. To date, the analysis of peptide-RNA hybrids by liquid chromatography-tandem mass spectrometry (LC-MS/MS) has been challenged by the inefficient isolation of these hybrids, low mass spectra quality, and complex data analysis. Recent studies have attempted to address these issues by chemically digesting the RNA moiety down to a single nucleotide, yet this approach compromises the sequence information of the RNA at the crosslink sites. This study reveals that non-crosslinked RNA is a major contaminant in the sample of photo-crosslinked protein-RNA hybrid, interfering in their LC-MS/MS detection in many ways. Here, we present a unifying workflow that efficiently removes free RNA from samples while preserving the crosslinked RNA, which can be tuned to the appropriate length for LC-MS/MS or single- molecule RNA sequencing. Using a novel high-intensity UV irradiation device, we achieved highly efficient protein-RNA crosslinking in living cells in their original culture medium within seconds of irradiation. Photo- crosslinked peptide-RNA hybrids were then isolated using a systematically optimized workup, which massively improved the LC-MS/MS identification and quantification of peptide-RNA hybrids. We showcase the performance of this method for tracking dynamic protein-RNA interactions upon drug treatment, recapitulating structural changes in known protein-RNA complexes. Overall, our study identifies confounding problems in the sample quality of peptide-RNA hybrids and presents an optimized workflow for their isolation that enables the analysis of protein-RNA interaction sites with a temporal resolution of seconds. 735

736

P-II-0587 Advancing Crosslinking Mass Spectrometry: Integrating Novel Data Acquisition Strategies and Enhanced Work- Up Techniques Barbara Steigenberger1 1Max-Planck-Institute of Biochemistry, Mass Spectrometry Core Facility, Munich, Germany Crosslinking mass spectrometry (XL-MS) has emerged as a powerful tool for investigating protein interactions and complex biological systems. However, its widespread application has been hindered by challenges in data acquisition and analysis, particularly in complex biological samples. We present a comprehensive approach that integrates cutting-edge data acquisition strategies and advanced work-up techniques to enhance the output and feasibility of XL-MS in complex systems biology studies. Our approach incorporates several innovative methodologies, including High field asymmetric waveform ion mobility spectrometry (FAIMS), fractionation prior to Liquid Chromatography-Mass Spectrometry (LC-MS), and a combination of Data-Dependent Acquisition (DDA) and Data-Independent Acquisition (DIA) approaches. These strategies enable enhanced sensitivity, resolution, and throughput, facilitating the identification and characterization of crosslinked peptides in complex mixtures. Moreover, we make use of the enrichable crosslinking reagent PhoX, which offers distinct advantages in XL-MS experiments by providing improved selectivity and efficiency in crosslink identification. Through the integration of these advancements, our work demonstrates a significant enhancement in the capabilities of XL-MS, making it more accessible and practical for investigating intricate protein networks and dynamic interactions within complex biological systems. 736

737

P-II-0588 Multimodal mass spectrometry identifies a conserved protective epitope in S. pyogenes Streptolysin O Di Tang1, Carlos Gueto-Tettay1, Elisabeth Hjortswang1, Joel Ströbaek1, Simon Ekström2, Lotta Happonen1, Lars Malmström1, Johan Malmström1,2 1Lund University, Department of Clinical Sciences Lund, Lund, Sweden 2SciLifeLab, Integrated Structural Biology Platform, Lund, Sweden Introduction: Antibody-guided vaccine design leverages functional antibodies to identify protective epitopes in their native three-dimensional conformation, an essential process for effective vaccine development. Despite its huge potential, this approach faces significant analytical challenges in epitope mapping and managing the complexity of bacterial pathogens, which have impeded its success against bacterial infections. In this study, we present a novel multimodal mass spectrometry (MS) strategy to characterize antigen−antibody complexes in- depth, using the bacterial immunogen Streptolysin O (SLO) from Streptococcus pyogenes as a showcase. Methods: Our approach began with the identification of a previously uncharacterized monoclonal antibody that neutralizes SLO-mediated haemolysis. Using MS-based de novo sequencing, we determined the primary structure of the antibody's light and heavy chains respectively. Next, chemical cross-linking mass spectrometry (XL-MS) was applied to generate informative distance constraints between the antibody's antigen-binding region and SLO. Integrative computational modelling subsequently identified a discontinuous epitope within domain 3 of SLO. Validation of this epitope was achieved through hydrogen-deuterium exchange mass spectrometry (HDX- MS) and reverse engineering of the targeted epitope. Results and Conclusion: Our results demonstrate that the protective antibody inhibits SLO-induced cytolysis by binding to a conformational epitope in domain 3, which likely inhibits the SLO oligomerization and secondary structure transitions necessary for pore formation. Remarkably, this epitope is conserved in over 98% of S. pyogenes isolates analysed, making it an attractive target for therapeutic and vaccine strategies against severe streptococcal infections. This study underscores the importance of integrating advanced MS-based techniques in structural proteomics to elucidate immune protection mechanisms and further guide rational vaccine design. Fig. 1 (optional) 737

738

P-II-0589 A structural proteomics approach to quantify protein reorganization during TDP-43 liquid-to-solid phase transitions Federico Uliana1, Yelyzaveta Zadorozhna1, Emre Pekbilir1, Carla Schmidt1, Dorothee Dormann1,2 1Johannes Gutenberg-Universität (JGU) Mainz, Mainz, Germany 2Institute of Molecular Biology, Mainz, Germany Neurodegenerative disorders such as Alzheimer's disease (AD), amyotrophic lateral sclerosis (ALS), and frontotemporal dementia (FTD) are characterized by an accumulation of protein aggregates in the affected brain regions. These protein aggregates are thought to arise from biomolecular condensates undergoing a liquid-to- solid phase transition. Despite the significant achievement in the identification of proteins partitioning into biomolecular condensates and the driving forces behind phase separation and aggregate formation, the molecular interactions regulating condensates/aggregates remains elusive. The primary factors contributing to this knowledge gap include the lack of a defined 3D structure for the flexible proteins involved and the absence of biochemical tools to quantify protein rearrangements during the liquid-to-solid condensate transition. To gain insight into the relationship between structure and physical properties, we study TDP-43 as a protein model. Although neuropathological studies have linked TDP-43 mis-localization and aggregation to major neurodegenerative diseases (ALS, FTD and AD), the knowledge about TDP-43"s structural organization is still limited to few protein domains/regions and cannot recapitulate the dynamic re-arrangement of the whole protein during the aggregation process. Combining multiple structural proteomics approaches (crosslinking coupled to mass spectrometry, limited proteolysis, and native mass spectrometry) we probed TDP-43 protein conformational states in-vitro using a set of TDP-43 variants characterized by different condensation properties (wild-type, condensation deficient or aggregation-prone TDP-43 variants). The generated data provide insight into intra-molecular structural rearrangements occurring in TDP-43 condensates and their interconnection with biophysical properties. We identified that the increase flexibility in the C-terminal IDR of the protein foster the interaction between the IDR C terminal region and the RRM1 and RRM2 regulating the process of protein rearrangement within liquid and solid-like condensates. By applying this panel of structural techniques to the field of neurodegeneration and in vitro and in-vivo condensation/aggregation systems, we aim to describe the molecular mechanisms regulating the aggregation processes, pinpointing interactions uniquely identified in pathological states, and offering potential targets for intervention strategies in neurodegenerative diseases. Fig. 1 (optional) 738

739

P-II-0590 Structure and interactions of the endogenous human commander complex Esa-Pekka Kumpula1, Saara Laulumaa1, Juha Huiskonen1, Markku Varjosalo1 1University of Helsinki, HiLIFE, Helsinki, Finland The recently discovered multiprotein Commander complex is a master regulator of cellular homeostasis within endosomal transport and the trans-Golgi network. The Commander complex has been linked to the regulation of several cellular functions, including ion and lipid homeostasis, embryogenesis, immune response, cell growth, and cell cycle. It consists of ten homologous Copper-metabolism Murr1 Domain proteins (COMMD1–10), coiled- coil domain-containing proteins CCDC22 and CCDC93, vacuolar protein sorting-associated proteins VPS26C, VPS35L, and VPS29, and a DENN (differentially expressed in normal and neoplastic cells) domain-containing protein 10 (DENND10). COMMD proteins 1–10 are highly conserved sequence homologs of COMMD1 found in mammals and some in lower metazoans. COMMD1 was named after its initially discovered function in copper homeostasis regulation. Knockouts of individual COMMD proteins cause severe reduction in protein levels of all COMMD proteins, suggesting that they assemble into a larger complex. Knockdowns of COMMD proteins have been associated with developmental disorders in lower vertebrates, while in humans, missense mutations in the COMMD genes have been linked to various human diseases such as Wilson"s disease, Parkinson"s disease, atherosclerosis, as well as viral protein recycling and cancer. Despite several studies focusing on individual Commander subunits, further systematic studies that assess the complete complex are necessary to understand how the multitude of its different functions are regulated. Similarly, the structure of the Commander complex, essential for understanding its molecular function, has remained elusive. In this study, we address these critical gaps in our knowledge of the Commander complex. We report the structure of the endogenous human Commander complex determined by cryogenic electron microscopy (cryo-EM) and comprehensively map the molecular context and interactions of the Commander and individual complex interactions in human cells using affinity purification mass spectrometry (AP-MS) and proximity labelling (BioID). These findings provide a structure-based blueprint for understanding the function of the Commander complex in endosomal sorting and other essential cellular processes. Fig. 1 (optional) 739

740

P-II-0591 Structural proteomics for the development of thrombolytic drug Naina Verma1,2,3, Josef Kucera1,2,3, Alan Strunga2,3, Linda Kasiarova2,3, Martin Marek2,3, Zbynek Prokop2,3, Jiri Damborsky2,3, Lenka Hernychova1,2,3 1Masaryk Memorial Cancer Institute, Proteomics, Brno, Czech Republic 2Loschmidt Laboratories, Department of Experimental Biology and RECETOX, Faculty of Science, Masaryk University, Brno, Czech Republic 3International Clinical Research Center, St. Anne’s University Hospital, Brno, Czech Republic The structural proteomic approach, using hydrogen-deuterium exchange combined with mass spectrometry (HDX-MS), can provide extensive insights into protein behaviour by serving as a link between structure, conformational dynamics, and function. It can help us identify the sites involved in the protein-protein interactions. Stroke is currently the second leading cause of death, with alteplase as the sole FDA-approved drug for its treatment. Alteplase, however, has several side effects, including intracranial bleeding and neurotoxicity. The development of the new generation of thrombolytics is highly desirable. Our research focuses on developing an alternative thrombolytic drug, staphylokinase (SAK), which is highly promising in terms of its thrombolytic activity, fibrin specificity, and low production cost. SAK is a 15.5 kDa prokaryotic protein derived from lysogenic strains of Staphylococcus aureus. It is a well-recognized thrombolytic agent that has demonstrated its clinical efficacy at par with alteplase. We aim to improve the thrombolytic efficiency of SAK by understanding its molecular structure, biochemical properties, and interaction sites. The structure of SAK has been studied through X-ray crystallography and nuclear magnetic resonance, revealing that its conformation is crucial for driving thrombolysis by activating plasminogen and interacting with plasmin. Using the HDX-MS approach, we aim to understand the complete in-solution conformation of wild-type and rationally designed SAK proteins. This analysis identifies the solvent-exposed regions of folded SAK proteins to detect structural changes. Subsequently, its interaction with plasmin and assistance in the activation of plasminogen will be studied in the holo form of SAK proteins. Thus far, we have conducted a comparative HDX study of a few non-immunogenic SAK mutants with 3-point, 8-point, and 12-point mutations. The results indicate significant differences in deuterium uptake in the bulge region comprising residues Leu40-Glu46 and the C- terminus that are known for their role in plasminogen activation. Deuterium uptake is higher in the solvent- exposed residues which is directly related to protein flexibility and affinity. The extent of their involvement in plasminogen activation will be studied through interactive HDX analysis. These results, complemented by biochemical assays, will help explain the impact of SAK conformation on the thrombolytic mechanism. 740

741

P-II-0592 Novel XL-MS analysis workflows for membrane protein characterization Yi He1, Weijing Liu1, Donggyun Kim2, Vadim Cherezov2, Gregory Dodge3, Barbara Imperiali3, Susan Leonhardt4, Mark Yeager4, Rosa Viner1 1Thermo Fisher Scientific, San Jose, CA, United States 2University of Southern California, Department of Chemistry, Los Angeles, CA, United States 3Massachusetts Institute of Technology, Department of Biology, Cambridge, MA, United States 4University of Miami, Coral Gables, FL, United States Membrane proteins (MPs) are involved in almost all cellular processes and comprise approximately 30% of the proteome, which makes them an ideal choice for 60% of drug targets. Although the study and characterization of MPs" structure, function, and interactions are clearly essential, they represent a challenge for current analytical methods, owing mostly to low production yields and the need for membrane mimetics to solubilize the proteins before analysis. Cross-linking mass spectrometry (XL-MS) has grown dramatically into a routinely utilized strategy for characterizing protein higher-order structure and mapping protein-protein interaction networks on a proteome-wide scale. However, the XL-MS analysis of membrane proteins is still a significant challenge due to their hydrophobic properties as well as the lipid-rich environment. In this work, we optimized sample preparation and detection of crosslinked peptides, including disulfide mapping for membrane proteins in SMALPs or detergents. Different amine-reactive crosslinkers, including DSS, DSSO and tBuPhoX were used to crosslink membrane proteins in SMALPs or detergents. Cross-linked samples were reduced, alkylated, and digested with trypsin or pepsin. For disulfide mapping, proteins were hydrolyzed with microwave-assisted acid hydrolysis (MAAH) and desalted using C18 stage tips. All data were acquired using the Thermo Scientific™ Vanquish™ Neo system connected to an EASY-Spray™ PepMap™ RSLC C18 column (0.075 mm x 250 mm, (Thermo Fisher Scientific)) at 250 nL/min flow rate and a Thermo Scientific™ Orbitrap™ Ascend Structural Biology Tribrid™ mass spectrometer coupled with a FAIMS Pro Duo interface. Data were analyzed using Thermo Scientific™ Proteome Discoverer™ 3.1 software and XlinkX 3.1 node. Our structural understanding of membrane proteins is very limited, largely due to their instability outside of the native lipid environment. They are usually extracted or solubilized in detergents or styrene maleic acid (SMA) lipid particles (SMALPs). It has been a challenge to develop XL sample preparation methods for membrane proteins that are compatible with mass spectrometry. In this study, we developed end-to-end XL-MS/S-S workflows for membrane proteins using different amino-reactive crosslinkers. In the optimized workflow, crosslinking was performed in detergents or SMALPs to keep their native conformations, followed by buffer exchange to 1% DDM or 0.1% RapiGest. Trypsin digestion was carried out using 0.1% RapiGest . Pepsin digestion was carried out in 1% formic acid . MAAH was performed in the original sample buffer with 25% TFA. With this new sample preparation methods, we were able to achieve a sequence coverage of more than 90% of the membrane proteins in SMALPs or detergents. More importantly, we identified crosslinked peptides and provide critical information for downstream structural analysis of GPCRs, HIV proteins, and bacterial enzymes by X-Ray and cryo-EM. 741

742

P-II-0593 In vivo protein footprinting reveals the dynamic conformational changes of Proteome of multiple tissues in progressing Alzheimer's disease John Yates1, Ahrum Son1, Hyunsoo Kim1,2, Jolene Diedrich1, Casimir Bamberger1, Daniel McClatchy1 1Scripps Research, Moleculara Medicine, La Jolla, CA, United States 2Chungnam National University, Convergent Bioscience and Informatics, Daejeon, South Korea Numerous studies have investigated changes in protein expression at the system level using proteomic mass spectrometry, but only recently have studies explored the structure of proteins at the proteome level.We developed covalent protein painting (CPP), a protein footprinting method that quantitatively labels exposed lysine, and have now extended the method to whole intact animals to measure surface accessibility as a surrogate of in vivo protein conformations. We investigated how protein structure and protein expression change as Alzheimer's disease (AD) progresses by conducting in vivo whole animal labeling of AD mice. This allowed us to analyze broadly protein accessibility in various organs over the course of AD. We observed that structural changes of proteins related to 'energy generation,' 'carbon metabolism,' and 'metal ion homeostasis' preceded expression changes in the brain. We found that proteins in certain pathways undergoing structural changes were significantly co-regulated in the brain, kidney, muscle, and spleen. 742

743

P-II-0594 Structural mass spectrometry captures residue-resolved comprehensive conformational rearrangements of a G protein-coupled receptor Bingjie Zhang1, Hongyue Liu1, Wenqing Shui1 1ShanghaiTech University, Shanghai, China GPCR structural studies with in-solution spectroscopic approaches have offered distinctive insights into GPCR activation and signaling that highly complement those yielded from structural snapshots by crystallography or cryo-EM. While most current spectroscopic approaches allow for probing structural changes at selected residues or loop regions, they are not suitable for capturing a holistic view of GPCR conformational rearrangements across multiple domains. Herein, we develop an approach based on limited proteolysis mass spectrometry (LiP- MS) to simultaneously monitor conformational alterations of a large number of residues spanning both flexible loops and structured transmembrane domains for a given GPCR. To benchmark LiP-MS for GPCR conformational profiling, we studied adenosine 2A ligand binding (agonist/antagonist/allosteric modulators) and G protein coupling. Systematic and residue-resolved profiling of A2AR conformational rearrangements by LiP-MS precisely captures structural mechanisms in multiple domains underlying ligand engagement, receptor activation and allostery, and may also reflect local conformational flexibility. Furthermore, these residue-resolution structural fingerprints of A2AR protein allow us to readily classify ligands of different pharmacology and distinguish the G protein-coupled state. Thus, we establish a new structural MS approach that would help address ligand or transducer-induced conformational transition and plasticity, a long-standing challenge for GPCR biology and rational drug design. to different response receptor (A2AR) in 743

744

One health approaches (plant, food, nutrition, animal, environment) (P-II-0595 to P-II-0617) P-II-0595 Interactome of Ambrosia artemisifolia pollen allergens and nanoplastics in simulated lung fluids Jelena Aćimović1, Mirjana Radomirović1, Maja Krstć Ristivojević1, Katarina Smiljanić1, Raquel Portela2, Lea Ann Dailey3, Dragana Stanić Vučinić1, Tanja Ćirković Veličković1,4 1University of Belgrade – Faculty of chemistry, Department of Biochemistry, Belgrade, Serbia 2Consejo Superior de Investigaciones Cientificas, Instituto de Catalisis y Petroleoquimica, Madrid, Spain 3University of Vienna, Division of Pharmaceutical Technology and Biopharmaceutics, Vienna, Austria 4Serbian Academy of Sciences and Arts, Belgrade, Serbia Interaction of nanoplastics (NP) with proteins produce a complex. Surrounded by protein coat - corona, NP develop new biological properties such as escaping from immune system, prolonged persistence in circulation and interference with cellular and molecular processes. After initial binding of molecules with higher affinity and formation of hard corona (HC), soft corona (SC) finalise biofilm NP encapsulation. The pollen of Ambrosia artemisifolia is widespread cause of respiratory allergic reactions, a single plant may produce about a billion grains of pollen per season. Ambrosia a. allergen was chosen as model system for investigation of NPs interactions with respiratory allergens. Ambrosia pollen extract (APE) was prepared by a short protocol, mimicking environmental exposure (ie. rainfall) and/or lung fluid extraction. APE was incubated in vitro with NPs of respirable size (nPP and nPET, 100 nm). Corona formation and composition were analysed by SDS PAGE, Schiff glycoprotein staining and immunoblotting with patients" sera. The profile of proteins bound to NPs with the highest affinity to be embedded into corona due to Vroman effect was analysed by mass spectrometry. These proteins are the most likely to be carried to respiratory system by NPs and to induce sensitization/allergic response. In SC of both nPET and nPP several main bands from APE are present, profile is similar to the bulk and extract profiles. HC shows different protein profile from SC for both NPs. There is enrichment of band at around 26-30 kDa (Mw corresponds to Amb a 4 allergen). This heavily glycosylated allergen consists of several isoforms. Schiff staining of glycoproteins confirmed that there is a glycopeptide in HC of both nPET and nPP. It seems that this glycopeptide, representing fragments of several proteins, Mw around 5 kDa preferentially binds to the surface of NPs. IgE-reactive protein at around 16 kDa in soft corona (and bulk) of both nPP and nPET was identified as nsLTP (Amb a 6) and plastocyanin (Amb a 3). In shotgun proteomics of Ambrosia elatior the major hit according to the number of peptides detected was PCC13-62-like protein, so far never shown at the protein level. Plant basic secretory protein, PBSP, also detected with high confidence in the extracts, was just a predicted protein, too. These two proteins have been detected in HC of nPP particles, major IgE-reactive band at 34 kDa. Specific adsorption of PCC13-62-like protein to NPs allows us to suggest novel protein PCC13-62-like protein to be lead biocorona component of NPs in the simulated lung fluids and a novel protein of Ambrosia pollen identified at the proteome level in this study. Acknowledgment. Funding from Horizon 2020, No965173, IMPTOX, by Ministry of Science, Innovation and Technological Development, Republic of Serbia, grant No451-03- 66/2024-03/200168 744

745

P-II-0596 Root proteomics and protein interaction studies showed a shift in biosynthetic pathways to different salt stress levels in Pistachio Mohammad Akbari1,2, Rakesh Singh3, Dalia Vishnudasan4, Elena Andriunaite5, Prashanth Survajhala4, Ramesh Katam1 1Florida A&M University, Department of Biological Sciences, Tallahassee, FL, United States 2Nazari Business Group, Pistat Research Center, Tehran, Iran 3Georgia Tech Institute, Atlanta, GA, United States 4Amrita Vishwa Vidhyapeetham, Amrita School of Biotechnology, Kerala, India 5Lithuanian Research Centre for Agriculture and Forestry, Institute of Horticulture, Babtai, Lithuania Pistachio (Pistacia vera L.) is an economically important tree nut that commonly thrives in semi-arid and arid environments. P. vera is highly adaptable to various abiotic stresses, including drought and salinity, making it suitable for reforesting arid and salinized zones. This study aimed to investigate the physiological and molecular mechanisms underlying stress tolerance in the UCB-1 cultivar. Five one-year-old pistachio rootstocks were subjected to four saline water regimes for 100 days. The rootstocks adopted the Na+ exclusion strategy to resist the salinity stress. Total proteins were isolated from the pistachio roots exposed to different NaCl concentrations. These proteins were characterized using high-throughput LC-MS/MS spectrometry, with the results searched against the Citrus database. Over 1600 protein IDs were detected, with comparative analysis identifying 245 proteins that were more abundant and 190 that were less abundant across three stress levels. Proteins associated with amino acid metabolism, cell wall organization, protein homeostasis, response to stress, signal transduction, TCA cycle, and vesicular trafficking were constantly overexpressed at all stress levels. At low and moderate stress levels, proteins involved in chromatin and cytoskeleton organization and lipid metabolism were overexpressed, while these proteins remained unaffected at higher salt concentrations. Transcription and translation processes were down-regulated across all stress levels. The protein interaction network, mapped to Arabidopsis thaliana orthologs, revealed clusters associated with these proteins. Under control conditions, the highest number of proteins were involved in chloroplast and cytosolic pathways. Under low salinity stress, there was a slight shift, with the most interconnected node activity assigned to mitochondrial and cytosolic pathways. As salinity increased, the pathways shifted focus, emphasizing carbohydrate metabolism under moderate salinity stress and amino acid metabolism under severe salinity stress. 745

746

P-II-0597 Scientometrics evaluation of published scientific papers on the use of proteomics technologies in mastitis research Maria Bourganou1, Dimitrios Chatzopoulos1, Daphne Lianou2, George Tsangaris3, George Fthenakis2, Angeliki Katsafadou1 1University of Thessaly, Department of Public and One Health, Karditsa, Greece 2University of Thessaly, Veterinary Faculty, Karditsa, Greece 3Biomedical Research Foundation of the Academy of Athens, Proteomics Research Unit, Athens, Greece The objective of the study was the presentation of quantitative characteristics regarding the scientific content and bibliometric details of the relevant publications. This is the first ever scientometrics evaluation of proteomics studies, the results of which highlighted the characteristics of published papers on mastitis and proteomics. In total, 156 papers were considered. Most papers presented original studies (n = 135), and fewer were reviews (n = 21). Most original articles (n = 101) referred to work involving cattle, most described work related to the diagnosis (n = 72) or pathogenesis (n = 62) of the infection and most included field work (n = 75). The tissue assessed most frequently in the studies was milk (n = 59). Milk was assessed more frequently in studies on the diagnosis (61.1% of relevant studies) or pathogenesis (30.6%) of the infection, but mammary tissue was assessed more frequently in studies on the treatment (31.0%). In total, 47 pathogens were included in the studies described, most frequently Staphylococcus aureus (n = 55 articles), Escherichia coli (n = 31) or Streptococcus uberis (n = 19). The proteomics technology employed more often in the respective studies was liquid chromatography-tandem mass spectrometry (LC-MS/MS), either on its own (n = 56) or in combination with other technologies (n = 40). The median year of publication of articles involving bioinformatics or LC-MS/MS and bioinformatics was the most recent: 2022. The 156 papers were published in 78 different journals, most frequently in the Journal of Proteomics (n = 16 papers) and the Journal of Dairy Science (n = 12). The median number of cited references in the papers was 48. In the papers, there were 1143 co-authors (mean: 7.3 ± 0.3 co- authors per paper, median: 7, min.–max.: 1–19) and 742 individual authors. Among them, 15 authors had published at least seven papers (maximum number of published papers per one author: 10). There were 218 individual authors who were the first or last authors in the papers. Most papers were submitted for open access publication (n = 79). The median number of citations received by the 156 papers was 12 (min.–max.: 0–339), and the median yearly number of citations was 2.0 (min.–max.: 0.0–29.5). The h-index of the papers was 33 and the m-index was 2. The increased number of cited references in papers and international collaboration in the respective to published papers. In conclusion, the use of proteomics in mastitis research has focused on the elucidation of pathogenesis and diagnosis of the infection; LC-MS/MS has been established as the most frequently used proteomics technology, although the use of bioinformatics has also emerged recently as a useful tool. associated with most study were the variables citations Fig. 1 (optional) 746

747

P-II-0598 Measurement of gluten peptides in human duodenal fluids using immuno-liquid chromatography-mass spectrometry Loic Dayon1,2, Antonio Núñez Galindo1, Julien Chevalier3, Michèl Aquarius4, Britt Otten5,6, Freddy J. Troost5,6, Peter Duncan3, Michael Affolter1 1Nestlé Research, NIFSAS, Lausanne, Switzerland 2Ecole Polytechnique Fédérale de Lausanne, Institut des Sciences et Ingénierie Chimiques, Lausanne, Switzerland 3Nestlé Research, NIHS, Lausanne, Switzerland 4VieCuri Medical Center, Department of Gastroenterology, Venlo, Netherlands 5Maastricht University, Department of Human Biology, Maastricht, Netherlands 6Maastricht University, Food innovation and health, Venlo, Netherlands Gluten proteins found in wheat are known to have a certain resistance to gastrointestinal digestion. To address the issue of accidental gluten ingestion in individuals with gluten sensitivity, it is crucial to understand how gluten peptides behave in biological samples such as gastrointestinal juices, blood plasma, or urine. However, current immunoassays used for gluten detection in food have limitations when applied to complex biological samples like duodenal fluids. In this study, our objective was to develop a mass spectrometry (MS)-based method for accurately measuring gluten peptides in human duodenal fluids. After conducting a literature review and characterizing a gluten-containing product, we selected seven gluten peptides, including the well-documented 33-mer gluten peptide. Isotopically labeled peptides were used as references, and we designed a targeted liquid chromatography (LC) MS assay utilizing high-resolution parallel reaction monitoring (PRM). Despite extensive optimization of the LC-PRM-MS method, the low levels of endogenous gluten peptides in human duodenal fluid samples made direct detection challenging. Therefore, we incorporated an initial immunoprecipitation (IP) step. We tested several antibodies and identified one that reliably enriched the 33-mer gluten peptide, along with a few additional gluten peptides. We assessed the performance of the immuno-LC-PRM-MS assay, focusing on quantification accuracy and precision. Our MS-based method successfully measures the 33-mer gluten peptide in human duodenal fluids, overcoming the limitations of current classical immunoassays. By combining IP and LC-PRM-MS analysis, and utilizing isotopic dilution, our approach provides sensitive, quantitative, and reproducible measurements. 747

748

P-II-0599 Proteomics allows differentiation of growing location of faba bean varieties Leigh Donnellan1, Clifford Young1, Peter Hoffmann1, Arineh Thamasian2 1University of South Australia, Clinical Health Sciences, Adelaide, Australia 2Edith Cowan University, Australian Research Council Centre of Excellence for Innovations in Peptide and Protein Science, Perth, Australia It is estimated that global food systems will need to double production by 2050 to meet the requirements of a growing population. To meet this demand and mitigate the environmental impacts of meat production, alternative protein sources such as pulse crops are being investigated for their sustainability. Faba bean (Vicia faba) is an important pulse crop to Australian farmers, with approximately 678,000 tonnes produced in 2020- 2021. In addition to being a cash crop, it plays important roles in nitrogen fixation, disease break and weed control within crop rotations. Nutritionally, faba bean is desirable because of its high-quality plant protein and dietary fibre. It is therefore one of the few crops that could satisfy the criteria for sustainable alternative protein sources, while advancing the current market-trade for Australia. Most research on the faba bean has focussed on the removal of vicine and convicine, which are antinutritional compounds that cause favism (haemolytic anaemia) in glucose-6-phosphate dehydrogenase deficient individuals. Conversely, only a few studies have investigated attributes such as crop quality, yield, resistance to environmental stressors or protein content. Furthermore, proteomic investigation to potentially address these questions have been extensively hampered by the low number of annotated proteins in the corresponding UniProt database (125 proteins). Herein, we describe the first large scale investigation of the faba bean proteome from 20 varieties grown at two different locations in Australia. Utilising theoretical protein sequences generated from a recently sequenced faba bean genomes, we produced the first extensively annotated faba bean protein database. From this, we identified ~2800 proteins and ~27,000 precursors using a 35-minute gradient at 5 µl/min on a ZenoTOF 7600 acquired with 64 variable window SWATH. Proteomics data allowed discrimination of the growing location, suggesting proteome adaptions to site-specific environmental influences. These results may allow for the identification of protein markers for suitability of different varieties grown at specific sites. This study is the first step in generating a comprehensive proteomic resource to aid both industry and researchers in identifying proteomic markers which can help guide the suitability of different varieties across growing locations 748

749

P-II-0600 MALDI-TOF MS for determination of sand fly trophic preferences from dried blood spots Petr Halada1, Ida Hanusniakova2, Daniel Kavan1, Alexandra Chaskopoulou3, Petr Volf2, Vit Dvorak2 1BioCeV, Institute of Microbiology of the Czech Academy of Sciences, Vestec, Czech Republic 2Faculty of Science, Charles University, Department of Parasitology, Prague, Czech Republic 3USDA–ARS, European Biological Control Laboratory, Thessaloniki, Greece Leishmaniases transmitted by bites of females of phlebotomine sand flies (Diptera: Psychodidae) belong among important and yet still neglected vector-borne diseases. Knowledge of feeding preferences towards potential reservoir animals is essential for understanding interactions of sand fly vectors with their hosts and setting correct and efficient control measures in endemic areas. Recently, MALDI-TOF mass spectrometry of host- specific hemoglobin peptides generated by trypsin digestion of the engorged blood was introduced as a rapid and cost-effective method for reliable blood meal identification in medically important arthropods including sand flies. For routine use in field surveys, a protocol for identification of blood meals crushed on filter paper (dried blood spots) was developed and its performance assessed using laboratory-reared sand fly females of three genera (Phlebotomus perniciosus, Sergentomyia schwetzi, Lutzomyia longipalpis) fed on two hosts (mouse, rabbit), sampled at different time points after feeding and stored for varying times. The approach provided correct host assignment of experimentally acquired blood meals in the interval 0 – 48 hours post blood feeding in all tested sand fly species and from blood spots preserved for up to one year at room temperature. Practical usefullness of the protocol was further evaluated using naturally engorged sand flies (Phlebotomus tobbi, Ph. perfiliewi, Ph. simici) from an endemic locality in Greece, successfully analyzing 33/34 females and identifying human and domestic animal hosts (sheep, goats, pigs, horses, dogs and chickens) including a mixed blood meal. MALDI-TOF MS peptide mass mapping of blood meals spotted on filter paper proved to be a powerful and robust method for identification of blood meal origin of sand flies and potentially a valuable tool in entomological surveys targeting vectors of sand fly-borne pathogens. Moreover, dried blood spots represent simple and low-cost alternative sample storage enabling easy shipment of the blood meal samples at ambient temperatures from regions of collection to MS facilities for the analysis. 749

750

P-II-0601 In-depth proteomics to elucidate pathways associated with cross-tolerance to drought and aflatoxin tolerance in peanut seed Ramesh Katam1, Caleb Faboye1, Rakesh Singh1 1Florida A&M University, Biological Sciences, Tallahassee, FL, United States Droughts severely affect peanut yields and predispose the crops to Aspergillus invasion. Proteome analysis plays a critical role in functional genomics to better understand the molecular mechanisms of plant stress responses. This study aims to (i) determine differentially expressed proteins in drought-tolerant and susceptible genotypes in response to drought and fungal infestation and (ii) elucidate the proteins associated with cross-tolerance to both stresses. Drought-tolerant and susceptible cultivars, ICGS76 and K1341, were subjected to drought stress via deficit irrigation. Additionally, the plants were challenged with A. parasiticus. Samples of peanut seeds were harvested from control, water-stressed, and fungal inoculated plants at regular time intervals to extract total proteins. The samples were subjected to overnight trypsin digestion, following LC-MS/MS analysis. Protein identification was carried out using ProteinPilot™ 5.0 software. Tandem mass spectrometry data collected for the peanut samples were searched against in silico tryptic digests of Viridiplantae proteins of the UniProt. Gene ontology analysis of identified proteins was carried out using Blast2GO v2.4.0.37 to annotate the protein sequences. The study demonstrates that the ICGS76 genotype has 57 abundantly expressed proteins. In contrast, K1341 has 37 abundantly expressed proteins in double stressed treated peanut seed (drought and Aspergillus) and T1 and T2. ICGS-76 revealed 11 unique proteins such as ubiquitin, phosphoglycerate kinase, and late embryogenesis proteins involved in plant defense. Comparatively, seven unique proteins, including phosphoglucomutase, sucrose synthase, and tubulin, involved in metabolic and glycolytic related processes, were detected in K1341. Protein-protein interactions have revealed the presence of physical interactions between proteins involved in essential biological processes, including signal transduction, stress response, plant defense, metabolism, glycolysis, phosphorylation, and the activation and deactivation post-translational modification as well as assembly of the cytoskeleton. The expression and interaction of defense proteins in response to drought and Aspergillus infestation suggests cross-tolerance to both stresses. 750

751

P-II-0602 The effect of GABA on liver proteomics profiles in high fat diet-induced obese mice Haejin Kwon1, Ji Young Mun1, Kwang-Won Lee2, Hyun-Mee Park1 1Korea Institute of Science and Technology, Advanced Analysis and Data Center, Seoul, South Korea 2Korea University, Department of Biotechnology, Seoul, South Korea Obesity is characterized by the excessive accumulation of body fat. And this is recognized as a leading risk factor for various chronic diseases, including cardiovascular diseases, type 2 diabetes, certain cancers, respiratory disorders and etc. Gamma-aminobutryic acid (GABA) diet is considered to have anti-obesity effect. In this study, we performed proteomic profiling of liver in obese mice non-treated or treated with GABA. High-fat diet-induced male C57BL/6J mice, 18 week old, were randomly allocated to two groups; administrated without, or with oral GABA at 14.3 mg/kg/day for 5 weeks. At the end of experiment, mice were euthanized using CO2, and then tissue samples were collected. We performed proteomic analysis using liquid chromatography and tandem mass spectrometry (LC–MS/MS) analysis. Proteomic profiling revealed 136 proteins statistically significant changes (p- value < 0.05, fold-change > 2); 54 up-regulated and 82 down-regulated for GABA-treated mice. Also we performed analysis for gene ontology (GO, p-value < 0.05) to assess the effect of obesity. The GO analysis for biological processes showed that differentially expressed proteins (DEPs) participated in metabolic process and catabolic process. And the GO terms for molecular function indicated that DEPs mainly enriched in the binding and activity. 751

752

P-II-0603 Coupling proteomics and lipidomics to unravel lipid synthesis in oats (Avena sativa L.) Darren Lau1, Leigh Donnellan1, John C. Harris2,3, Julie E. Hayes3, Janine Croser2, Peter Hoffmann1 1University of South Australia, Clinical and Health Sciences, Adelaide, Australia 2South Australian Research & Development Institute, Department of Primary Industries and Regions, Adelaide, Australia 3University of Adelaide, School of Agriculture, Food and Wine, Adelaide, Australia Oat (Avena sativa L.) is famously known for its high oil content as compared to other cereal grains, such as barley and wheat. It also has a unique nutritional profile with high β-glucan fiber, low gluten, high protein and avenanthramides, which have been shown to provide a range of health benefits. Hence, there is a high marketing appeal for oats to be made into nutrient-enriched products. However, the high oil content of oats contributes to the formation of clumps during milling and fractionation processes, which renders oat processing very inefficient due to clogging of machinery. Therefore, there is interest from research and industry in the breeding of low-oil oats to overcome this processing barrier. To achieve this, greater depth of knowledge on the regulation of oil synthesis in oats is essential. We optimized a complementary proteomics and lipidomics approach to investigate differences in the regulation of oil synthesis in a large cohort of genotypically and phenotypically diverse oat lines, as well as grain development series of two milling-grade variety. Prior to analysis, lipids and proteins were sequentially extracted from minute amounts of a single sample. LC-MS/MS analyses were performed on the SCIEX ZenoTOF 7600 system. Proteomics and lipidomics data were processed using Spectronaut™ and MS-DIAL respectively. From the oat cultivars, we observed correlations between their oil contents and the abundances of several proteins involved in the lipid synthesis pathway (eg. acetyl-CoA carboxylase, 3-hydroxyacyl-ACP dehydratase and long-chain acyl-CoA synthetase). While the expression levels of triacylglycerol species were also positively correlated with oil levels, phospholipid expression exhibited an opposite trend. To provide further insights into oat oil synthesis, we analysed a grain development series (4 to 20 days after pollination) of two varieties and identified significant changes in protein and lipid profile at different stages that reflected their difference in oil content. Findings from this study can help oat breeders to better understand oat oil regulation and potentially inform breeding strategies for low-oil oats. 752

753

P-II-0604 Discovery of plasma biomarkers in dogs afflicted with mammary gland tumors through proteograph analysis Jeong-Woon Lee1,2, Je-Yoel Cho1,2 1College of Veterinary Medicine, Seoul National University, Department of Biochemistry, Seoul, South Korea 2Science Research Center, Seoul National University, Comparative Medicine and Disease Research Center(CDRC), Seoul, South Korea To identify differential proteins in the blood of tumor-bearing patients, numerous studies have employed various rigorous proteomics assays. However, challenges in sensitivity and time efficiency, especially in detecting low- abundance proteins, persist. In this study, we addressed these challenges by utilizing Proteograph technology. Through the analysis of plasma samples from eight dogs, Proteograph identified a higher number of proteins, including low-abundance ones, and detected more differentially expressed proteins (DEPs) compared to previous studies using conventional LC-MS/MS. Further exploration of cancer-upregulated DEPs revealed enrichments in immune and complement-related gene ontology terms in GO analysis, while proteins related to proteasome and complement activation clustered in PPI analysis. Notably, proteins involved in epigenetic mechanisms, such as methylation, were identified as cancer-upregulated proteins, a finding not observed in previous studies. Comparison with published human studies revealed a higher consistency in the directionality of common proteins among cancer-upregulated ones. Additionally, by utilizing the TCGA and GTEx databases, we identified CDH11 and PSMA5 as proteins upregulated at both RNA and protein levels in breast cancer, with high levels associated with poor prognosis, suggesting their potential as biomarkers in cancer progression. We plan to further investigate CDH11, identified as a DEP exclusively by Proteograph, to understand how circulating cadherins might contribute to tumor prognosis in patients. 753

754

P-II-0605 Mechanistic insights into the combined hepatotoxic effects of the deoxynivalenol and the emerging mycotoxin Enniatin B using multiomics Thibaut Léger1, Sylvie Huet2, Kevin Hogeveen2, Valérie Fessard2, Denis Habauzit2 1French Agency for Food, Environmental and Occupational Health & Safety (ANSES), Toxicology of contaminant Unit - Proteomics and metabolomics facility (PROMETHEE), Fougères, France 2ANSES Fougères, Toxicology of contaminant Unit, Fougères, France Cereal and crop production throughout the world is facing contamination by natural substances such as mycotoxins. Consequently to climate change and its impact on the development of toxin-producing fungus, mycotoxin contamination of food and feed is increasing in the last decades, raising consumer health concern worldwide. The genus Fusarium can produce a wide range of mycotoxins. Among the main Fusarium mycotoxins, deoxynivalenol (DON), has been found in a broad panel of foodstuffs while the enniatins, with enniatin B (ENNB) as the major produced variant, are considered as emerging mycotoxins. Enniatins can be found in mixture with DON in food. However, if DON is regulated in Europe, no regulation for enniatins is currently available due to a lack of information on toxicity. Nevertheless, the mechanisms of action involved in DON toxicity and its combined effect with enniatins are not well characterized. Therefore, the aim of this study was to characterize the toxic effects of these two mycotoxins, alone and in combination, on the HepaRG liver cell model using a multiomic approach. First, the toxicity of the 2 mycotoxins (ENNB and DON) alone and in mixture at equivalent concentration was assessed on HepaRG cells by High Content Analysis. Several toxicological endpoints were investigated such as viability, DNA damage, apoptosis, cell cycle analysis, and pro-inflammatory response. Proteomic and metabolomic analyses were performed, in order to decipher the mechanisms involved in their toxicity. DON induced a dose-dependent decrease of cell number, and increased caspase 3 and γH2AX as well as IL-8 release. In contrast to DON, ENNB decreased the viability only of 20 % and did not affect caspase 3 and γH2AX staining. However, similarly to DON, ENNB increased IL-8 release. The combination of these two mycotoxins rescues HepaRG from DON-induced cell death and inhibited the DON-increase of γH2AX and caspase 3 induced. The mixture slightly decreased the IL-8 release provoked by each toxin. The molecular mechanisms involved in the single toxins and the mixture responses were compared using proteomics (on cell content) and metabolomics (on cell culture medium). We showed that the protective effect of ENNB observed on DON mortality involved lipids and mitochondria. Indeed, DON treatment increased the abundance of long chain saturated dicarboxylic acids, signs of a defective fatty acid oxidation (FAO). DON also increased the abundances of histones and proteins associated to mitochondrial respiration and decreased specifically abundances of proteins implicated in FAO. These effects were alleviated by ENNB in the cells treated with the mixture, suggesting opposite effects of the two toxins on HepaRG cells. This study provided new findings on the molecular effects of DON and ENNB on liver cells and highlighted that their combined effects should not be ruled out in risk assessment. 754

755

P-II-0606 Proteomic and metabolomic insights into the cellular pathway changes in Choysum (Brassica rapa) during postharvest storage Qingsong Lin1, Xin Ee Yong1, Qifeng Lin1, Renee Chin1, Teck Kwang Lim1, Ritu Bhalla1, Deivanai Subramanian1 1National University of Singapore, Biological Sciences, Singapore, Singapore After being harvested, green leafy vegetables such as Choysum (Brassica rapa) begin to lose their freshness and nutritional properties, leading to significant postharvest losses. These losses are a major challenge in agriculture, with rates estimated at 25% in developed countries and up to 50% in developing nations, imposing substantial economic, social, and ecological burdens. Employing suitable postharvest storage methods, such as cold storage, is a common approach to slow down deterioration. However, the intricate cellular pathways involved in vegetable deterioration are not fully understood. To address this, we aimed to elucidate the impact of postharvest storage on the cellular pathways involved in degradation and senescence in Choysum using proteomics and metabolomics approaches. Protein and metabolite extractions were performed on Choysum samples stored at 4°C for various time points, followed by SWATH quantitative proteomics and key metabolite quantification to analyze changes in cellular pathways. Preliminary data from duplicate protein extracts identified and quantified 2372 proteins using SWATH MS, with 20 proteins found to be differentially abundant based on a 1.5-fold change and a Q-value of less than 0.05. Gene annotation of these proteins revealed several cellular pathways involved in the degradation of Choysum during postharvest storage, with proteins involved in organic metabolic processes, amino acid biosynthesis, and translation decreasing in abundance, while proteins involved in carbohydrate and ROS metabolic processes increased in abundance. Latest results of this ongoing project will be presented during the conference. This study will enhance our understanding of the plant's response to storage by elucidating the cellular pathways involved in postharvest quality loss of vegetables. 755

756

P-II-0607 Environmental temperature changes and BDNF depletion induce neurotoxicity in Danio Rerio Francesca Grassi Scalvini1, Simona Nonnis1, Elisa Maffioli1, Armando Negri1, Gabriella Tedeschi1, Flavia Fabretti2, Mattia Toni3 1University of Milano, Veterinary Medicine and Animal Science, Milan, Italy 2University of Bologna, Department of Medical and Surgical Sciences – DIMEC, Bologna, Italy 3Sapienza University, Department of Biology and Biotechnologies "Charles Darwin", Rome, Italy Anthropic action amplifies global warming taking place on our planet, causing the release greenhouse gases, including CO2. Physiology and behavior of ectothermic animals are influenced by temperature. We recently demonstrated that environmental temperature variation heavily alters locomotor activity but also complex behaviors such as anxiety, social behavior, aggression, learning, and cognitive behaviors in Danio rerio [1, 2]. In summary, our data suggested that temperature variation can be counted among the factors that generate neurotoxicity. To confirm such hypothesis, the different proteomic datasets used in our previous studies (18°C, 26°C and 34°C) for 4 days (acute) or 21 days (chronic treatment) [1, 2], and BDNF+/- and BDNF-/-zebrafish (kept at 26°C or 34°C for 21 days) [3] were re-analysed by an integrated proteomic approach by means of the software Ingenuity Pathway Analysis (IPA). The output of our analysis provides three protein panels related to 18°C, 34°C and BDNF depletion that can be linked to anxiety-like or boldness behavior compared to control (26 °C) and identifies proteins that can be used as hallmarks of neurotoxic processes common to all the treatments applied [4]. To further characterize the impact of environmental changes on the zebrafish brain, we are now analyzing the lipid component of the brain using a targeted and untargeted lipidomic approach and we are studying the eye proteome, as the eye is anatomical extension of the brain which is often affected in neurological disorder. Previously published in: 1. Toni M et al. (2019) J Proteomics. 30;204:103396. 2. Nonnis et al. (2021) Sci Rep. 28;11(1):2521. 3. Maffioli et al. (2022) Int J Mol Sci. 17;23(10):5606 4. Maffioli et al. (2023) Int J Mol Sci. 24(21):15735. 756

757

P-II-0608 Profiling antimicrobial and resistance proteins in milk from diverse cow breeds Cristian Piras1, Rosario De Fazio1, Antonella Di Francesco2, Vincenzo Cunsolo2, Domenico Britti1, Rainer Cramer3, Paola Roncada1 1University "Magna Græcia" of Catanzaro, Department of Health Sciences, Catanzaro, Italy 2University of Catania, Laboratory of Organic Mass Spectrometry, Department of Chemical Sciences, Catania, Italy 3University of Reading, Department of Chemistry, Reading, United Kingdom Milk is a complex fluid containing nutritional and bioactive compounds, including antimicrobial peptides (AMPs) and resistome proteins. AMPs, crucial for innate immunity, protect against pathogens in the mammary gland and infant gut. Resistome proteins, linked to antimicrobial resistance (AMR), pose a global health concern. Understanding these proteins in milk from different bovine breeds, such as the Podolica breed, can shed light on antimicrobial properties and bioactive potential. Monitoring resistome proteins is crucial due to the spread of AMR genes in animal production. This study used LAP-MALDI MS, proteomics, and metaproteomics to classify Podolica milk, identify AMPs, and investigate resistome proteins across various breeds and environments. Eight farms in Calabria (Italy) were studied: four Podolica farms and four with other cattle breeds. Milk samples were collected twice from each farm, 7 days apart, using a sanitized dipper after 5–10 minutes of agitation. Each 250 mL sample was partially used for somatic cell count (SCC). Samples were transported at 4°C to the Centro Interdipartimentale di Servizi Veterinari, Magna Graecia University of Catanzaro, Italy, and stored at -80°C. LAP-MALDI MS profiling differentiated Podolica milk from other breeds. Milk samples were processed via centrifugation and TCA precipitation, followed by hybrid quadrupole-time of flight mass spectrometer analysis. Protein concentrations were measured, and samples were digested with trypsin. LC-MS/MS analysis was performed using a ThermoFisher Scientific Orbitrap Fusion Tribrid® mass spectrometer. Data analysis involved PEAKS XPpro and MaxQuant software for label-free quantification (LFQ), using Bos taurus, bacteria, and resistome databases. LAP-MALDI MS profiling coupled with LDA analysis yielded 86% accuracy in classifying Podolica versus control milk samples in the training set. The test set achieved 98.4% accuracy. Significant ion species for classification were identified, showing notable abundance differences. Bottom-up proteomics identified about 220 proteins from LC-MS/MS searches. Cathelicidins and annexins were more abundant in control cow milk. Bacterial taxonomy analysis showed different bacterial presences between groups. The CARD database analysis detected beta-lactamases and tetracycline resistance proteins in both sample groups, with an isoform of beta-lactamase more abundant in Podolica milk, highlighting the importance of monitoring AMR. In conclusion, this study analyzed milk composition from various bovine breeds, focusing on AMPs and resistome proteins. Using LAP-MALDI MS profiling and bottom-up proteomics, we distinguished Podolica milk, identifying around 220 proteins. Both control and Podolica samples contained beta-lactamases and tetracycline resistance proteins, emphasizing the need for routine AMR monitorin 757

758

P-II-0609 Leveraging LC-MS-based proteomics for improved wheat breeding: unveiling genetic and environmental factors on the expression of flour proteins that influence end-use quality and tolerability Malte Sielaff1,2, Khaoula El Hassouni3, Muhammad Afzal3, Franz Pfleger4, Melanie Ruhrländer4, Norbert Huintjes4, Stefan Tenzer1,2,5,6, Carl Friedrich Horst Longin3 1University Medical Center of the Johannes Gutenberg-University Mainz, Institute of Immunology, Mainz, Germany 2University Medical Center of the Johannes Gutenberg-University Mainz, Research Center for Immunotherapy (FZI), Mainz, Germany 3University of Hohenheim, State Plant Breeding Institute, Stuttgart, Germany 4DIGeFa GmbH, Detmold Institute for Grain and Fat Analysis, Detmold, Germany 5Helmholtz Institute for Translational Oncology (HI-TRON), German Cancer Research Center (DKFZ), Mainz, Germany 6German Cancer Research Center (DKFZ) Heidelberg, Division 191, Heidelberg, Germany Wheat is one of the world's most important crops, providing around 20% of human calorie and protein intake. Its popularity is attributed to its distinct dough and baking characteristics, which are essential for bread and pasta making. These unique properties are largely determined by the amount and composition of flour proteins. On the downside, many of those are associated with wheat-related diseases like allergies or intolerances. Breeding for higher yield and better resistances is of great importance in order to meet the increasing demand and to adapt to climate change. Improving quality is also of interest but limited by the negative correlation of yield with protein content, used as a measure for baking quality. Further methods are thus required to estimate protein quality beyond mere amounts. To this end, we developed a high-throughput LC-MS-based bottom-up proteomics workflow for the analysis of wheat flour and analyzed 282 varieties grown under 4 environmental conditions (1200 samples). Proteins were extracted under denaturing and reducing conditions and transferred to microtiter plates. Semiautomatic sample preparation including protein assay, protein clean-up on magnetic beads and tryptic digestion were performed using a Beckman Biomek i7 liquid handling robot. Peptides were loaded onto Evotips and analyzed using an Evosep One (60 samples/day) coupled to a Bruker timsTOF HT (diaPASEF). Raw data were processed in DIA-NN (library-free mode). Results were filtered to a run-specific and global FDR of 1% at the precursor and protein level. Protein intensities were batch effect-corrected using the R package limma. Average protein intensities per variety, genetic and environmental variance components affecting the protein levels and heritabilities were modelled using the R package ASreml. Statistical analyses were performed using the R packages limma, rstatix and fgsea. Overall, 6567 protein groups were quantified in the flours, >6000 with a data completeness >90% across all wheat varieties. For 3461 proteins high heritabilities (proportion of observed variation that can be attributed to genetic differences rather than environmental factors) h2>0.5 were detected, representing candidates whose levels can be effectively modified through selective breeding. Many showed significant and biologically meaningful correlations with conventional measures like loaf volume or falling number, indicating that proteomic data could indeed help to predict baking quality. A comparison with allergen databases identified 305 potentially immunogenic proteins. Of those, many showed h2>0.5 and no correlations with agronomic and quality parameters, indicating that a reduction of allergens through breeding seems to be possible. With the advantage of the low input required and the prospect of faster LC-MS methods, our data show that proteomics has the potential to become a valuable tool in future wheat breeding and routine testing along the supply chain. 758

759

P-II-0610 A proteogenomic investigation of the role of the Type VI Secretion System on the dynamics of Campylobacter jejuni and Bacillus subtilis interaction Katarina Šimunović1, Gabriele Meroni2, Giulia Laterza2, Paola Roncada3, Sonja Smole Možina4, Luigi Bonizzi2, Ines Mandić Mulec1, Alessio Soggiu2 1University of Ljubljana, Department of Microbiology, Ljubljana, Slovenia 2University of Milan, Department of Biomedical, Surgical and Dental Sciences, One-Health Section, Milan, Italy 3University‘ Magna Græcia’ of Catanzaro, Department of Health Sciences, Catanzaro, Italy 4University of Ljubljana, Department of Food Science and Technology, Ljubljana, Slovenia Campylobacteriosis is the predominant zoonotic infection in Europe and poses a significant public health concern due to its potential to cause severe sequelae, including peripheral neuropathies, Guillain–Barré syndrome, Miller-Fisher syndrome, and functional bowel disorders such as irritable bowel syndrome. The primary causative agents, Campylobacter coli and Campylobacter jejuni, inhabit domestic animals like poultry, cattle, and pigs, along with wild birds, acting as commensals. Poultry meat is identified as the major source of infection and, when transmitted to humans through ingestion, these bacteria follow a pathogenic route. Despite various strategies at both management and treatment levels to reduce Campylobacter in poultry, the effectiveness of many probiotics as feed additives, particularly Bacillus subtilis PS-216, requires comprehensive investigation. The aim of this study was to investigate at the proteome level the in vitro interaction between C. jejuni and B. subtilis, with a specific focus on comparing C. jejuni strains possessing the virulence factor Type 6 Secretion System (T6SS) and strains not possessing it. T6SS is a relevant element to consider since it can function as a potent weapon against host cells and other bacteria, adding a layer of complexity to Campylobacter pathogenicity. In this study, a proteogenomics approach was used to evaluate in parallel the growth dynamics in coculture and the production of antimicrobial and signalling proteins by B. subtilis in response to the T6SS presence or absence. We initially monitored the growth differences of C. jejuni strains through bacterial monoculture and coculture with B. subtilis in Mueller Hilton Broth in microaerobic conditions. In parallel, the study assessed the functional proteome remodelling of B. subtilis during the interaction with C. jejuni with and without T6SS. In summary, this investigation aimed to advance our understanding of Campylobacteriosis by exploring the dynamics of C. jejuni and B. subtilis interaction, shedding light on potential avenues for targeted interventions and preventive measures. 759

760

P-II-0611 Optimizing recombinant protein production: comparative analysis of different CHO cell lines Sujay Subramaniam1, Alexander Reder1, Ulrike Mäder1, Mark Smales2, Jan Maarten van Dijl3, Uwe Völker1 1University Medicine Greifswald, Department of Functional Genomics, Greifswald, Germany 2University of Kent, School of Biological Science, Canterbury, United Kingdom 3University Medical Center Groningen, Medical Microbiology, Groningen, Netherlands Increasing the secretory capacity of various host systems for the production of high-quality recombinant proteins, particularly therapeutic proteins (biotherapeutics) and industrial enzymes is crucial for a sustainable bio-based economy. The high production costs of biotherapeutics limit patient access to novel therapies. Chinese Hamster Ovary (CHO) cells are the most commonly used production system, accounting for over 85% of current market products. Our research focuses on enhancing protein production in CHO cells by utilizing different cell lines and evaluating the burden on these cell lines by overexpressing a model protein. The main objectives are to compare the protein production capacity of different CHO cell lines and to generate an optimized vector system for stable high-yield protein expression system. The cell lines studied include GIBCO- CHO-S, CHO-K1-SmaF, Freestyle CHO-S, and CHO-S. Protein estimation was performed for all cell lysates and secretome samples, followed by mass spectrometry (MS) analysis for protein identification and quantification. Transient expression studies with optimized plasmids were assessed using western blotting. Initial protein estimation revealed substantial differences in total amounts of secreted proteins across cell lines. MS analysis identified 5700-5800 proteins in cell lysates, with secretome protein counts ranging from 1100 to 1600, of which only 20-25% were secreted proteins. The CHO-K1-SmaF cell line showed notable differences in protein secretion compared to others. Optimized plasmid constructs demonstrated higher protein expression in transient studies. These findings suggest that selecting appropriate CHO cell lines and a better plasmid design are critical for enhancing protein production. This approach can improve industrial efficiency, reduce waste and lower costs benefiting numerous industrial applications. Ongoing experiments aim to develop more effective expression vectors and evaluate their performance in various CHO cell lines to further enhance production yields and streamline biotherapeutic manufacturing processes. 760

761

P-II-0612 Maternal urinary proteome revealed pregnant process Shuxuan Tang1 1Beijing Ditan Hospital, Capital Medical University, Beijing, China., Beijing, China Each year, there are an average of 200 million pregnancies, 300,000 maternal deaths related to pregnancy and birth, and 7 million fetal perinatal deaths worldwide. Maternal biological fluids used to describe the state of a healthy pregnancy and fetal development were important to ensure a successful monitoring. Urine, as a filtrate of the blood, bears no need nor mechanism to be stable, and tolerates changes to a higher degree, which are associated with physiological or pathophysiological processes of the body. Therefore, changes were more easily detected in urine than in blood. In this study, liquid chromatography-mass spectrometry(LC/MS) was used to monitor the pregnancy process through maternal urinary proteome. Pregnant rats urine samples were collected throughout pregnancy (from GD 0d to GD 21 d), and urine samples were collected from control group at corresponding time points. By comparing with the urinary proteome within groups on the first day of pregnancy, the differential proteins were involved in embryo implantation and trophoblast differentiation. Liver, kidney, and bone development started early to be enriched in the pregnant group, but not in the control group. Interestingly, the developmental processes of the fetal heart such as heart looping and endocardial cushion formation could be seen in urine of pregnant rats. Moreover, the timings were consistent with embryological studies. The timing of fetal lung surfactant appearance in urine was right before birth. The differential proteins related to pancreas development appeared time of pancreatic cell proliferation and differentiation. Furthermore, coagulation-associated pathways were found to be increasingly prominent before labor. in urine at reported the time during Our results indicated that maternal urinary proteome can reflect the process of pregnancy, even embryonic and fetal development. This study provides a new perspective for early fetal monitoring. Fig. 1 (optional) 761

762

P-II-0613 The bee gut microbiota bridges pathogens and virulence traits in a one health perspective Bruno Tilocca1, Viviana Greco2, Mariachiara Paonessa1, Andrea Urbani2, Paola Roncada1 1University of Catanzaro, Health Science, Catanzaro, Italy 2Catholic University of the Sacred Heart, Rome, Italy The microbiota of the gastrointestinal tract of bees plays a crucial role in the services they provide to the environment, as well as to humans and animals. Moreover, microorganisms harbored in the bee gut are moved, with the bee flies, between seemingly distant habitats. This promotes microbial heterogeneity of the visited microcosms and facilitates the exchange of genetic material among microbes at both intra- and inter-kingdom level, with strong influences on the functional array of the entire microbial community. Thus, studying the functional concern of the bee gut microbiota might represent a suitable strategy for monitoring and forecasting microbial dynamics and evolutions. Here, only a paucity of studies are available in the literature, most of which deal with the definition of the bee gut microbiota composition; whereas, very few aim at the functional investigation of the sole bacterial fraction. In this study, we employed a metaproteomics approach to assess the functional dynamics of the bacteria, fungi, and single-celled parasites residing in the bee gut. Our results describe a dynamic composition of the bacterial community, listing species of relevance to both humans and animals. Bacterial functional concern underlines the vehiculation of virulence and antimicrobial resistance traits. The fungal and single-celled parasite fractions consist of specimens commonly related to the environmental and animal spheres, whose metabolic activities support the role of bees in the diffusion of functional traits across ecological niches. Additionally, the host proteome reflects essential physiological activities of bees, supporting the use of metaproteomics to study multiple microbial species simultaneously and their interactions with the host. Altogether, this study enhances our understanding of the structure and function of the bee gut microbiota, underscoring its potential suitability in addressing various pressing issues such as antimicrobial resistance diffusion, environmental pollution, and the promotion of environmental health. 762

763

P-II-0614 The secret(ome) behind microbial survival in nutrient poor aquatic environments Claudia Tugui1, Martin Pabst1, Dimitry Sorokin1,2, Mark van Loosdrecht1 1Delft University of Technology, Biotechnology, Delft, Netherlands 2Winogradsky Institute of Microbiology, Moscow, Russian Federation Water scarcity has become a prevalent challenge worldwide due to climate change and poor water management. Therefore, safe, and clean water becomes of utmost importance for governments and major efforts are made to ensure bacteriological stable and high-quality drinking water. Bacteria of the genus Aeromonas, which include potential pathogens, are ubiquitous in aquatic environments and are also commonly found in drinking water distribution systems. Despite the nutrient-poor environment of these systems, Aeromonas species can survive and thrive, posing a health risk to vulnerable groups such as children, the elderly, and immunocompromised patients. Studies have shown that the presence of Aeromonas is associated with loose deposits and the presence of invertebrates, specifically Asellus aquaticus. Chitin, the structural shell component of these invertebrates, may serve as a potential nutrient source in these otherwise nutrient-poor environments. Biological polymers are among the most abundant and rich carbon sources on Earth. However, their complex structures and chemistry often make them resistant to microbial degradation and utilization as a carbon source. In this study, we demonstrate that two Aeromonas strains, commonly found in drinking water distribution systems, can efficiently degrade and utilize chitin as a sole carbon and nitrogen source. Through quantitative proteomics analysis of the cell biomass and secretome, we have identified a dedicated and diverse array of hydrolytic enzymes and pathways for chitin uptake and metabolism.1 An extended study on a broader range of natural biopolymers revealed that Aeromonas species are capable of degrading and utilizing a variety of polymers, including starch and collagen. In summary, this study illustrates how microbes like Aeromonas, by linking to invertebrates and other biopolymer carbon sources, can sustain themselves in nutrient-poor drinking water distribution systems. This insight lays the groundwork for developing more effective water sanitation strategies. References: 1. Tugui, Claudia, et al. "Exploring the metabolic potential of Aeromonas to utilise the carbohydrate polymer chitin." bioRxiv (2024): 2024-02. Fig. 1 (optional) 763

764

P-II-0615 Comparison of urinary proteome in the first two days after mating in male rats Haitong Wang1, Chenyang Zhao1, Youhe Gao1 1Beijing Normal University, Beijing, China Objective: To compare urine proteome changes in the first two days after mating behavior in male rats. Methods: The urine samples of Sprague-Dawley rats on the day of mating and the day after mating were collected and analyzed by non-label quantitative proteomics by high performance liquid chromatography- tandem mass spectrometry (LC-MS/MS), and the differential proteins in the urine proteome were screened protein function and biological pathways were analyzed. Results: 43 differential proteins were identified by comparing the urine proteome of rats on the next day of mating with that on the day of mating, and nearly two-thirds of the differential proteins were related to spermatogenesis. Conclusions: There were changes in the urinary proteome on the next day of mating compared with the day of mating, and some of the proteins whose functions were known to be related to spermatogenesis. Urine proteome has the potential to reflect spermatogenesis. 764

765

P-II-0616 Effect of autonomous consumption of different sweet substances on the urinary proteome of mice Haitong Wang1, Youhe Gao1 1Beijing Normal University, Beijing, China Objective: To explore the possible effects on the organism by analysing changes in the urinary proteome of mice after autonomous consumption of different sweet substances. Methods: Urine samples were collected from C57BL/6l mice before and after active consumption of sweet substances. The sweet substances included sucrose, stevia glycosides, acesulfame, and sucralose, which are more widely used worldwide and can elicit a preference response in mice, and the concentrations of the non- nutritive sweeteners were chosen to be those that have been shown by existing studies to have the strongest preference response in mice. The analyses were performed by the non-labelled quantitative proteomics technique of high performance liquid chromatography tandem mass spectrometry (LC-MS/MS), and groups were screened for differential proteins in the urinary proteome for the analysis of protein functions and biological processes; comparisons of urinary proteomes before and after consumption of sweeteners by a single mouse were carried out; and side-by-side comparisons of different sweeteners were made. Results and Conclusions: Urine proteome could reflect the changes in the organism of mice after voluntary consumption of sweeteners, and the effects of different sweeteners on the urine proteome were not the same.Among the four sweeteners, sucralose and sucrose induced the most similar changes in the organism, and steviol glycosides induced the furthest changes in the organism; the changes induced by sucrose, acesulfame, and sucralose were similar, and steviol glycosides induced changes different from those of the other sweeteners. Stevia glycosides induced changes that were different from the other sweeteners. The urinary proteomic proteins that differed among the four sweeteners consumed autonomously by mice included proteins that have been reported to be associated with brain reward circuits, whereas only the urinary proteomic proteins that differed among the four sweeteners consumed autonomously by mice were associated with metabolic processes in large quantities after the voluntary consumption of sucrose, acesulfame, and sucralose, and the urinary proteomic proteins that differed among the four sweeteners consumed autonomously by mice were associated with the assembly of nucleosomes, gene expression, and other processes after the voluntary consumption of stevia glycosides. 765

766

P-II-0617 Comparison of urine proteome between obese people and normal weight people Haitong Wang1, Lilong Wei2, Yun Zhou2, Yongtong Cao2, Youhe Gao1 1Beijing Normal University, Beijing, China 2China-Japan Friendship Hospital, Beijing, China Objective: To explore whether urine proteome can reflect the difference between obese and normal weight people. Methods: Urine samples from obese and normal weight people were collected and identified by non-label quantitative proteomics using high performance liquid chromatography tandem mass spectrometry (LC-MS/MS). The difference proteins of urine proteome between obese and normal weight people were screened for protein function and biological pathway analysis. The urine proteome of obese individuals was compared with that of normal weight people, and the common differential proteins were counted to analyze the protein function and biological pathways. Reported biomarkers of obesity were searched in the urine proteome of obese individuals. Results: 38 different proteins can be identified in the urine proteome of obese people compared with normal weight people, some of which have been reported to be related to metabolism and obesity, and the biological processes of differential proteins are also related to metabolism and other processes. 8 common differential proteins in the urine proteome of obese individuals and normal weight people, among which some proteins have been reported to be related to metabolism and obesity, and the biological processes of differential proteins are also related to metabolism and other processes. Among the differential proteins in the urine proteome of obese individuals compared with the normal weight people, the reported obesity biomarkers can be matched. Conclusions: The urine proteome can distinguish the obese people from the normal weight people, and the differential proteins in the urine proteome have key proteins that are known to be related to obesity and metabolism, and the biological processes of differential proteins also related biological processes such as nutrition and metabolism. Urine proteome has the potential to explore the pathogenesis of obesity and provide personalized treatment. 766

767

Clinical proteomics II (P-II-0618 to P-II-0723) P-II-0618 Proteomic characterization of Meningioma tissue and serum Julieta Maria Medaglia1, Tilman Werner1,2,3, Agnes Schäfer4, Jörg Walter Bartsch4, Oliver Schilling1,5 1Institute for Surgical Pathology, Faculty of Medicine, University Medical Center Freiburg, Freiburg, Germany 2University of Freiburg, Faculty of Biology, Freiburg, Germany 3University of Freiburg, Spemann Graduate School of Biology and Medicine (SGBM), Freiburg, Germany 4Philipps University Marburg, Department of Neurosurgery, Marburg, Germany 5German Cancer Consortium (DKTK) and Cancer Research Center (DKFZ), Heidelberg, Germany Introduction: Meningioma is a primary tumor of the central nervous system (CNS) that accounts for approximately one-third of all CNS neoplasms. Despite the fact that these tumors are mostly benign, meningiomas exhibit complex histologic features and can be located in compromised sites, which has proven to be challenging in terms of management and predicting prognostic outcomes. Advances in molecular characterization of meningiomas have revealed potential applications for addressing the lack of standardization in diagnostic and therapeutic choices, which could improve staging and prediction of prognosis. However, further studies are required to gain a molecular understanding of this cancer. Methods: 24 fresh-frozen meningioma tissue samples were digested using the S-Trap protocol and measured via liquid-chromatography mass-spectrometry (LC-MS/MS) based proteomics in data independent acquisition mode (DIA). Matching patient serum was digested via a depletion-based protocol and measured in DIA as well. Results: More than 7000 unique proteins were identified per sample on average, more than 6900 of these with a consistency of less than 30% missing values. Principal Component Analysis was unable to differentiate between subtypes or WHO grading based on the data alone. However, Partial Least Squares Discriminant Analysis (PLS-DA) was able to separate the meningothelial and fibrous subtypes. Nevertheless, most subtypes still exhibited overlapping characteristics. Further analysis is required to determine whether the tissue heterogeneity permits the identification of a defined proteomic profile. The objective of our future research is to compare different meningioma subtypes and correlate tissue samples with matching serum samples, which has yielded over 750 identified proteins, over 650 of which were identified with less than 30% missingness. Conclusion: In conclusion, proteomic characterization of meningioma tissue samples holds great promise for addressing the diagnostic challenges of this diverse tumor. Despite progress, the lack of reliable biomarkers remains a significant hurdle. Our research aims to identify proteins that could be potential biomarkers using advanced proteomic techniques. 767

768

P-II-0619 Quantitative proteomics identifies proteins and pathways involved in the progression from precancerous gastric lesions in a population of high-cancer gastric risk Lizeth Mejia1, Andres Castillo1, Jeovanis Gil Valdés2 1Universidad del Valle, Biology, Cali, Colombia 2Lund University, Lund, Sweden Gastric cancer (GC) is responsible for approximately 763,000 annual deaths worldwide, ranking fifth in mortality rates and sixth in terms of cancer types with the highest incidence. The most common type of GC is intestinal- type gastric cancer. In most cases, this cancer develops from a series of well-defined pathological states, starting with non-atrophic gastritis (NAG) that can progress towards precancerous lesions (PL), including atrophic gastritis (AG), intestinal metaplasia (IM), dysplasia, and finally, GC. In Colombia, GC is the leading cause of cancer-related deaths, and in the Andean region in southwestern Colombia, the incidence and mortality rates of GC are among the highest in the country. In this region, a cohort of patients has been followed since 1991. We examined 63 gastric tissue samples from patients who were followed up for 16 years. Biopsies from three stages - NAG, IM, and dysplasia - were collected and analyzed using liquid chromatography coupled with tandem mass spectrometry (LC-MS/MS). Additionally, three samples of normal gastric tissue were included for analysis. We have identified and reported quantitative data for 8351 proteins that were confidently quantified. This revealed dysregulated pathways and processes at different stages. Differentially expressed proteins (DEPs) increased as the lesions progressed. The most significant dysregulation was found in proteins in the dysplasia stage compared to the IM and NAG stages. We also found four protein clusters with similar trajectories, indicating dynamic and progressive changes in proteomic profiles from non-atrophic to advanced gastric lesions. Three clusters showed a decrease in expression as they progressed to PL. The protein AP1B1 exhibited a clear tendency to decrease its expression, significantly reducing dysplasia. Another protein cluster showed increased expression in NAG, which remained stable until MI but markedly increased expression in dysplasia. Proteins SFN, GOLM1, ABRACL, ANKLE2, STMN1, MYL12A, S100A10, TAGLN2, and EXO1 tended to increase from normal tissue to NAG, then to IM, with a significant increase observed in dysplasia. IM was characterized by the downregulation of proteins related to nonsense-mediated decay (NMD) and the Unfolded Protein Response, with overexpression of proteins associated with the immune system. The transformation to dysplasia involves overexpression of oxidative phosphorylation, apoptosis, oxidative stress, and the downregulation of proteins involved in DNA repair, telomere maintenance, chromatin remodeling, regulation of cell-cell adhesion, and different signaling pathways in carcinogenesis. Differentially expressed proteins (DEPs) in dysplasia, namely STMN1, SRI, CLIC1, and SPARC1, which have a prior association with GC, show a gradual increase from NAG to GLP. Together, these findings suggest that changes in the expression of specific proteins are associated with the progression of gastric lesions towards malignancy. 768

769

P-II-0620 Assessment of Hemoglobin-related clinical markers using volumetric absorptive microsampling and data- independent acquisition mass spectrometry Valdemir Melechco Carvalho1,2, Debora Lima1,2, Rodrigo Andrade Schuch1, Jessica Salgueiro1, Maria Carolina Pintão1 1Fleury Group, Research and Development, São Paulo, Brazil 2University of São Paulo, Phaculty of Pharmaceutical Sciences, Graduate Program in Clinical Analysis and Toxicology, São Paulo, Brazil Data-independent acquisition (DIA) offers extensive proteome coverage and serves as a unified protocol for quantifying numerous blood proteins. With its high specificity, mass spectrometry significantly reduces interference commonly encountered in other blood marker assays. In this study, we integrated DIA with volumetric absorptive microsampling (VAMS) for whole blood collection and storage with automated proteomics sample processing to evaluate clinical markers. As a proof of concept, we assessed two hemoglobin-related biomarkers: glycated hemoglobin (HbA1c) and hemoglobin (Hb) variants. Our results demonstrated that HbA1c quantification via DIA correlated well with the reference method, though the method's imprecision did not meet the required quality standards for this biomarker. Additionally, we developed proposed an approach for hemoglobin variant identification based on a customized hemoglobin database combined with a workflow for DIA data extraction and rigorous evaluation of peptides including multiple qualifiers. 769

770

P-II-0621 Harmonization of quantitative proteomics data collected across the international cancer proteogenome consortium Gennifer E. Merrihew1, Aaron Maurais1, Julia E. Robbins1, Brian Connolly1, Simona Colantonio2, Joshua J. Reading2, Joseph K. Knotts2, Rhonda R. Roberts2, Stephanie Garcia-Buntley3, Hongyan Ma3, William Bocik3, Jesse Stottlemyer3, John Hamre3, Eunkyung An4, George E. Craft5, Peter G. Hains5, Roger R. Reddel5, Phillip J. Robinson5, Qing Zhong5, Vincent Richard6, Christoph H. Borchers7, Kiminori Hori8, Hiroki Shinchi8, Koji Ueda8, Hiroshi Nishida9, Kosuke Ogata9, Yasushi Ishihama9, Kyu Jin Song10, Jae-Won Oh10, Kwang Pyo Kim10, Hazara Begum Mohammad11, Ye-Ji Do11, Min-sik Kim11, Shinyeong Ju12, Hankyul Lee12, Cheolju Lee12, Jingi Bae13, Chaewon Kang13, Sang-Won Lee13, Ignasi Jarne14, Joan Josep Bech-Serra14, Tatiani Brenelli Lima14, Carolina De La Torre Gomez14, Lazaro Hiram Betancourt Nunez15, Nicole Woldmar15, Roger Appelqvist16, György Marko-Varga16, Sandra Goetze17,18,19, Bernd Wollscheid17,19, Yi-Ju Chen20, Hsiang-En Hsu20, Hao Fang20, Yu-Ju Chen20, Yu- Tsun Lin21, Kun-Yi Chien21,22, Jau-Song Yu21,22,23, Sara Ten Have24, Angus Lamond24, Nathan J. Edwards25, Ana I. Robles4, Henry Rodriguez4, Michael J MacCoss1 1University of Washington, Department of Genome Sciences, Seattle, WA, United States 2Frederick National Laboratory for Cancer Research, Antibody Characterization Laboratory, Cancer Research Technology Program, Frederick, MD, United States 3Frederick National Laboratory for Cancer Research, Proteomics Characterization Laboratory, Cancer Research Technology Program, Frederick, MD, United States 4National Cancer Institute, National Institutes of Health, Office of Cancer Clinical Proteomics Research, Division of Cancer Treatment and Diagnosis, Rockville, MD, United States 5The University of Sydney, ProCan, Children’s Medical Research Institute, Faculty of Medicine and Health, Westmead, Australia 6Jewish General Hospital, Segal Cancer Proteomics Centre, Lady Davis Institute for Medical Research, Montreal, Canada 7Jewish General Hospital, Gerald Bronfman Department of Oncology, Montreal, Canada 8Japanese Foundation For Cancer Research, Cancer Proteomics Group, Cancer Precision Medicine Center, Tokyo, Japan 9Kyoto University, Division of Medicinal Frontier Sciences, Graduate School of Pharmaceutical Sciences, Kyoto, Japan 10Kyung Hee University, Department of Applied Chemistry, Institute of Natural Science, Yongin, South Korea 11Daegu Gyeongbuk Institute of Science and Technology, Department of New Biology, Daegu, South Korea 12Korea Institute of Science and Technology, Chemical & Biological Integrative Research Center, Seoul, South Korea 13Korea University, Department of Chemistry, Center for Proteogenome Research, Seoul, South Korea 14Josep Carreras Leukaemia Research Institute (IJC), Proteomics Unit, Badalona, Spain 15Lund University, Skåne University Hospital, Department of Translational Medicine, Malmö, Sweden 16Lund University, Department of Biomedical Engineering, Lund, Sweden 17ETH Zürich, Institute of Translational Medicine at the Department of Health Sciences and Technology, Zurich, Switzerland 18ETH PHRT Swiss Multi-Omics Center (SMOC), Zurich, Switzerland 19Swiss Institute of Bioinformatics (SIB), Lausanne, Switzerland 20Academia Sinica, Institute of Chemistry, Taipei, Taiwan 21Chang Gung University, Molecular Medicine Research Center, Taoyuan, Taiwan 22Chang Gung University, Graduate Institute of Biomedical Sciences, College of Medicine, Taoyuan, Taiwan 23Chang Gung Memorial Hospital, Department of Otolaryngology-Head and Neck Surgery, Taoyuan, Taiwan 24University of Dundee, Laboratory of Quantitative Proteomics, Division of Molecular, Cell, and Developmental Biology, School of Life Sciences, Dundee, United Kingdom 25Georgetown University, Department of Biochemistry and Molecular & Cellular Biology, Washington, DC, United States The International Cancer Proteogenome Consortium (ICPC) is a collaboration among leading cancer and proteogenomics research centers. This study had the goal to assess whether different laboratories could obtain comparable quantitative proteomics data independent of the protocol, instrument platform, or analysis strategy. To perform this analysis, nine samples were derived from the NCI-7 cell line panel (Clark et al, J. Prot Res, 2018). The NCI-7 Cell Line Panel is composed of the NCI-H23, RPMI-8226, T47D, A549, COLO205, NCI-H226, and CCRF CEM cell lines which represent multiple tissues and of a variety of genetic mutations. Cell lines were grown, lysed and aliquoted at the Frederick National Laboratory for Cancer Research and sent to the University of Washington to coordinate distribution of lysates to 16 labs representing 10 countries. Each lab was sent lysates from each of the seven individual cell lines, a 4-pool lysate (NCI-H23, T47D, A549, and CCRF CEM) and the 7-pool lysate, the NCI-7 universal proteomic reference. The use of an additional reference material of each lab"s choice was encouraged. Each lab analyzed these lysates with their preferred sample preparation protocol, mass spectrometry acquisition type and instrumentation. Sample preparation protocols used in this study included S- trap, PAC/SP3, in-solution digestion (SDS or urea), in-gel digestion, and FASP. Liquid chromatography systems used include Sciex nanoLC 425, Evosep One, Thermo Easy nLC, Thermo Neo Vanquish, and Thermo UltiMate 3000. Mass spectrometers used in this analysis included Bruker timsTOF HT or Pro2, SCIEX 6600 Triple TOF, Thermo Q-Exactive (HF, HF-X, or Plus), Thermo Orbitrap Fusion Lumos, Thermo Orbitrap Eclipse, Thermo 770

771

Orbitrap Fusion Lumos-FAIMS Pro, and Thermo Exploris 480. Quantitative acquisition methods were either DDA- LFQ, DDA-TMT, DIA or DIA-PASEF. A total of 20 datasets were collected as some labs performed multiple analyses. All DDA-TMT data was analyzed using the NCI Common Data Analysis Pipeline (CDAP). All DIA data were analyzed with a DIA CDAP that makes use of either EncyclopeDIA or DIA-NN as the search engine and Skyline for quantification. The DDA-LFQ analyses were analyzed with MSAmanda and Skyline. Protein parsimony was performed using Skyline and normalized protein quantities derived using directLFQ. Batch correction was performed between each dataset using ComBat. Following normalization and batch correction, each cell line clustered together, regardless of location, sample preparation methods or instrumentation. The 4-pool lysate also clusters as expected centered amongst the 4 individual cell lines. Fig. 1 (optional) 771

772

P-II-0622 Integrative proteomic and AI-based histopathologic profiling reveals new prognostic subclusters in early-stage non-small cell lung cancer Corinna Friedrich1,2, Simon Schallenberg2, Matthias Ziehm1,3, Gabriel Dernbach2, Philipp Keyl2,4, Hannah Demeke1,2, Keziban Merve Alp1, Dražen Papić2, Rebecca Fritz2, Mohamed Haji1, David Horst2, Frederick Klauschen2,4, Philipp Mertins1,3 1Max Delbrück Center for Molecular Medicine, Proteomics Platform, Berlin, Germany 2Charité – Universitätsmedizin Berlin, Institute of Pathology, Berlin, Germany 3Berlin Institute of Health at Charité – Universitätsmedizin Berlin, Proteomics core unit, Berlin, Germany 4Ludwig Maximilians University Hospital Munich, Institute of Pathology, Munich, Germany Understanding the early stages of lung cancer progression at the molecular and cellular levels is crucial for improving patient outcomes. Our study employed an innovative integrative approach combining advanced deep proteomics profiling with cutting-edge histopathology foundation models. This method facilitated a comprehensive analysis of protein regulation, cell composition, and clinical outcomes in a formalin-fixed, paraffin-embedded (FFPE) cohort comprising over 1,000 non-small cell lung cancer cases. We identified a novel proteomic subcluster associated with a favorable prognosis, characterized by an enrichment of cytotoxic T cells. Our analysis differentiated between three-year relapse-free survivors and non- survivors of lung squamous cell carcinoma, revealing that non-survivors exhibited specific upregulation of the inflammation-promoting complement-based membrane attack complex, alongside higher neutrophil granulocyte counts. In contrast, adenocarcinomas in non-survivors showed upregulation of the minichromosome maintenance (MCM) replication complex, accompanied by higher tumor-associated macrophage counts at diagnosis. Our discovery-driven proteomics approach successfully identified both tissue-resident and systemic protein markers linked to poor outcomes in lung cancer, alongside distinctive protein signatures correlating with cell types identified through histopathological analysis. Notable potential biomarkers, including complement protein C7 and the MCM2 component of the replication complex, were further validated using multiplexed immunohistochemistry. This research underscores the potential of combining high-throughput proteomics with histopathology foundation models to elucidate molecular and cellular interactions that correlate with prognosis in extensive cohorts of solid tumors. These findings not only enhance our understanding of lung cancer biology but also pave the way for the development of new diagnostic strategies. 772

773

P-II-0623 Understanding the proteome changes arising throughout 5-Fluorouracil chemoresistance development in colorectal cancer cells, with special focus on the methyl proteome Isaac Micallef1,2, Byron Baron1, Norie Araki2,3 1Centre for Molecular Medicine and Biobanking, Msida, Malta 2Graduate School of Medical Sciences, Kumamoto University, Tumor Genetics and Biology/Molecular Genetics, Kumamoto, Japan 3Graduate School of Medical Sciences, Faculty of Life Sciences, Kumamoto University, Kumamoto, Japan Introduction: Comprehensive proteomics strategies for analysing post-translational modifications (PTMs) have showcased the potential use of protein modifications for defining novel cancer-driving mechanisms and drug- resistance markers. Recent studies highlight the aberrant methyl proteome changes arising in the development of various cancers, but the changes arising throughout the development of chemoresistance in most cancers has not been explored in depth. This study is currently making use of colorectal cancer (CRC) cell lines as a cancer model to uncover the changes in the global proteome and methyl proteome arising following development of 5- fluorouracil (5FU) chemoresistance. Method: Three CRC cell lines (HCT116, Caco2, DLD1) were cultured by regular exposure to 5FU over a long period (3 months to 1 year), and stable 5FU-resistant cells were successfully established and isolated for downstream analysis. Whole cell lysates from sensitive and resistant cells (periodically exposed for 6 months) were used, and the peptide digests extracted were measured in label-free quantification (LFQ) DDA-mode using an Easy-nanoLC-Orbitrap-Fusion-Tribrid system, equipped with Nikkyo-C18-nano-Column. The data was processed using Proteome Discoverer (v.2.4), statistical analysis and cluster analysis was performed with Perseus, while Gene Ontology and network analyses were assisted by DAVID, KEGG, and KeyMolnet. Results: Through global proteomics, approximately 7500 proteins (FDR < 1%) were quantitatively identified, with each cell line exhibiting more than 800 dysregulated proteins (-1.5 ≤ Log2 FC ≥1.5,p < 0.05). Network analysis for the common dysregulated proteins revealed that signalling pathways related to transcriptional and translational regulatory factors and amino acid metabolism, were significantly altered. Furthermore, more than 70 types of methyltransferases were identified, with over 1000 methylated peptides being detected (without enrichment). Manual filtering resulted in a list of just over 600 methylated peptides in each cell line, with di-methylated residues being among the most identified (>300 sites) in each cell line pair. The common proteins with variable methylation sites revealed dysregulated signalling pathways related to alternative splicing. Conclusions: From the combined data of global proteomics and methyl proteomics, the preliminary data collected has revealed significantly altered methylated proteins and methyltransferases in 5FUR CRC cell lines. It is anticipated that following further validation, the data generated will provide a better understanding of how to improve the effectiveness of anti-cancer drugs from a methyl proteomic standpoint. 773

774

P-II-0624 ACR-2316: a potent, selective WEE1/PKMYT1 inhibitor rationally designed for superior single agent activity using Acrivon Predictive Precision Proteomics (AP3) for biological SAR Caroline Wigerup1, Helén Nilsson1, Lei Shi2, Joon Jung2, Joelle Baddour-Sousounis2, Ruban Cornelius1, Nina Lipjankic1, Uthira Muralitharan1, Valentina Siino1, Georgia Mitsa1, Ignacio Arribas Diez1, Zachary Best2, Martina Pasetto1, William Dahlberg2, Shahrzad Rafiei2, Portia Lombardo2, Magnus E. Jakobsson1, Reina Improgo2, Christina Scherer2, John van Duzer2, David A. Proia2, Kristina Masson1, Peter Blume-Jensen2 1Acrivon AB, Lund, Sweden 2Acrivon Therapeutics Inc., Watertown, MA, United States WEE1 and PKMYT1 kinases play essential roles in cell cycle checkpoints and DNA damage response. Genetic interaction has been documented between WEE1 and PKMYT1, and clinical trials have reported clinical activity with WEE1 inhibitors adavosertib (adavo), azenosertib (azeno) and the PKMYT1 inhibitor lunresertib (lunres). Here, we report the discovery and characterization of ACR-2316, a dual inhibitor of WEE1 and PKMYT1 specifically designed for optimal selectivity through co-crystallography and superior single agent activity uniquely enabled by Acrivon Predictive Precision Proteomics (AP3). Mass spectrometry-based AP3 profiling was conducted across several novel WEE1 and PKMYT1 inhibitor leads generated through co-crystallography-based rational drug design. Selective leads originating from one series were optimized using AP3 for biological structure-activity relationship analysis. AP3 profiling revealed WEE1 inhibitor-upregulated phosphorylation sites across a subset of phosphoproteins that were oppositely regulated by PKMYT1 inhibition, including CDK1-T14, a direct PKMYT1 phosphorylation site, as well as CHK1-S296. A particular lead compound, ACR-2316, demonstrated a desirable potent, balanced ratio of cellular WEE1 and PKMYT1 target engagement (95% inhibition of pCDK1/2-Y15 and 14% inhibition of pCDK1-T14 at 30 nM in ACHN cells), resulting in superior activation of the mitotic kinases CDK1, CDK2, and PLK1 compared to adavo and lunres, based on annotated kinase substrate relationships. ACR-2316 is more selective than adavo, azeno, and lunres based on >200 kinases profiled by AP3, and 468 kinases assessed by KINOMEscan. ACR-2316 induced a drastic S-G2/M accumulation distinct from adavo or lunres. In a panel of 20 human tumor cell lines, ACR-2316 demonstrated >5-fold greater potency in all cell lines tested (CellTiterGlo) compared to azeno and lunres (mean IC50 = 70, 414 and ≥496 nM, respectively). Superior anti-cancer activity of ACR-2316 was observed in 16 ovarian cancer patient-derived xenograft (PDX) models tested ex vivo (CellTiter-Glo 3D) compared to azeno and lunres (median IC50 = 11, 264 and 2260 nM, respectively). ACR-2316 demonstrated superior, durable, dose-dependent efficacy compared to azeno and lunres in human cell line derived xenograft and PDX models and was well tolerated. Tumor regression with ACR-2316 was associated with strong WEE1 (>85%) and intermediate PKMYT1 (30%) inhibition in tumors at 60 mg/kg. Additionally, ACR-2316 resulted in rapid and deep regression in xenograft tumors progressing on azeno. In conclusion, ACR-2316 is a potentially first-in-class, selective dual WEE1/PKMYT1 inhibitor with superior preclinical single-agent activity compared to clinical benchmark WEE1 or PKMYT1 inhibitors. ACR-2316 is progressing through IND-enabling studies in preparation for clinical monotherapy development. 774

775

P-II-0625 Targeted quantitative proteomics of anticoagulated patients Eleonora Camilleri1,2,3, Bart van Vlijmen3, Manfred Wuhrer1, Christoph H. Borchers2, Suzanne Cannegieter3, Nienke van Rein3, Yassene Mohammed1,2 1Leiden University Medical Center, Center for Proteomics and Metabolomics, Leiden, Netherlands 2McGill University, Montreal, Canada 3Leiden University Medical Center, Leiden, Netherlands Oral anticoagulants (OACs) are commonly prescribed medications to treat and prevent thromboembolic complications. The two most common indications are the prevention of ischemic stroke in patients with atrial fibrillation (AF) and the treatment or prevention of venous thromboembolism (VTE). Despite their beneficial antithrombotic effect, OACs treatment also leads to severe adverse events, such as occurrence of major bleeding in around 3% of patients each year. Moreover, OACs are not always effective, as breakthrough strokes or VTE still occur despite treatment. The frequency of these adverse events calls for more research towards the individualization of anticoagulant therapy, through better characterization of the population and understanding of differences in risks of developing adverse events. While a few proteomics studies in patients on OAC treatment were published recently, such as patients with AF (PMID: 34353540, 34390530) or VTE (PMID: 29296750), the potential of proteomics characterization for these patients is currently underutilized. We used targeted proteomics with internal standards to measure 270 proteins, including 64 in the complement and coagulation cascade, in a prospective case-cohort from BLEEDS (PMID: 27935941), which enrolled 16,570 patients initiating VKAs between 2012-2014. The case-cohort included 370 random subcohort members, 250 major bleeding, and 59 stroke cases occurring during follow-up. Plasma concentrations of 270 proteins were measured at baseline (i.e., before the bleeding) using quantitative targeted proteomics with internal standards. To investigate subgroups differences, protein concentrations in the subcohort were compared by age, sex, treatment indication, comorbidities and comedications with Wilcoxon rank-sum test (p-value≤0.05 adjusted for false discovery rate). 139 proteins passed all our quantification criteria. The measurements were not influenced by patients" characteristics or occurrence of clinical outcomes. Proteomic profiles were age- and sex-specific in the 370 subcohort patients. Older patients had higher levels of alpha-2-macroglobulin, fibulin and complement factors. VTE patients had elevated levels of apolipoprotein C-I, B-100, L-1 and A-II compared with AF patients. Patients with pulmonary embolism had a similar profile compared to deep vein thrombosis patients. Neuropilin-2, insulin-like growth factor-binding protein 2, protein AMBP, β-2-microglobulin and leucine-rich α-2-glycoprotein were, among others, ≥15% increased in the major bleeding patients compared to the subcohort. Our results show how precise proteomic characterization of patient plasma samples on anticoagulation treatment can identify novel potential biomarkers for VTE and major bleeding. The attractiveness of using targeted assays with internal standard is due to the ability of faster adaptation of assays in clinical settings after required validation. 775

777

P-II-0626 Proteomic mapping of spatiotemporal dynamics in metastatic colorectal cancer Vishnu Mohan1, Ofir Atrakchi1, Sophie Piwko1, Iina Takala1, Mirie Zerbib2, Roni Oren2, Tomer Meir Salame3, Ekaterina Kopitman3, Eviatar Weizman4, Ben Boursi5, Tamar Geiger1 1Weizmann Institute of Science, Department of Molecular Cell Biology, Rehovot, Israel 2Weizmann Institute of Science, Department of Veterinary Resources, Rehovot, Israel 3Weizmann Institute of Science, Department of Life Sciences Core Facilities, Rehovot, Israel 4Weizmann Institute of Science, The Mantoux Bioinformatics Institute of the Nancy and Stephen Grand Israel National Center for Personalized Medicine, Rehovot, Israel 5Sheba Medical Center, Department of Oncology, Tel HaShomer, Israel Metastatic colorectal cancer (mCRC) presents a significant health challenge, with poor patient outcomes and limited treatment options. Untreated mCRC carries a bleak prognosis, with a 5-year survival rate below 10% and a median survival of only 5 months. Metastatic spread to the liver, due to its anatomical proximity, and to the lungs is common in mCRC, often indicating advanced disease and impacting treatment strategies. This study aims to elucidate the temporal and spatial dynamics of cell populations and their protein expression through the metastatic process, utilizing both a murine metastasis model and human primary and metastatic tumor samples. We employed a multi-modal approach, combining CyTOF characterization of immune and stromal landscapes within murine tumors with in-depth proteomic analysis of isolated cell populations. Additionally, we investigated human tumor samples to gain insights into the proteomic landscape of CRC in the spatial dimension. Our findings reveal a complex interplay between different cell types in tumor progression, highlighting the tissue specific role played by the immune cells in shaping the tumor microenvironment. Late-stage tumors are dominated by immunosuppressive macrophages, monocytes, dendritic cells, and cancer-associated fibroblasts, while early-stage tumors mostly encounter tissue-specific immune surveyors. Proteomic analysis identified distinct protein expression profiles in liver, lung, and primary human colorectal cancer (hCRC). Liver metastases showed enrichment in pathways related to drug metabolism, fatty acid metabolism, amino acid biosynthesis, and ECM-receptor interaction/focal adhesion. Compared to primary and lung tumors, liver metastases had lower levels of cell cycle and nucleotide biosynthesis proteins, but higher levels of certain metabolic and innate immune proteins. These findings emphasize the importance of studying organ-specific protein expression patterns and cellular heterogeneity in guiding therapies for metastatic CRC. 777

778

P-II-0628 The promise of the Orbitrap ASTRAL mass-spectrometer for proteomics analysis Jean-Marc Monneuse1, Josephine Abi Ghanem2, Célie da Silva2, Sangeetha Covindarassou1, Thomas Ribette1, Magali Sarafian1, Adrien Saliou3 1BIOASTER, OMICS - Metabolomics & Proteomics, Lyon, France 2BIOASTER, OMICS - Computational Biology, Lyon, France 3BIOASTER, OMICS, Lyon, France Over the last two decades, the advent of high-throughput sequencing has revolutionized our access to genomic and transcriptomic data, and thus our understanding of biological systems. On the other hand, mass spectrometry (MS)-based proteomics has undergone a slow steadily technology advancements in terms of sensitivity and robustness. However, the last innovations allow complete proteome coverage, with short turnaround time and reduced sample material. They will facilitate the implementation of the proteomics methods in preclinical and clinical routine practice. In that context, the introduction of the Orbitrap asymmetric track lossless (ASTRAL, Thermo Fisher) analyzer is a major technological leap for MS-based analysis with a scan speed up to 200 Hz, combining sensibility, reproducibility, robustness and high throughput. Following our recent acquisition of the ASTRAL, we performed multiple experiments to assess the performances of the system in terms of identification of proteins at bulk and single cell levels. The results were compared with our gold standard Exploris 480 (E480). Both mass spectrometers were coupled to the Vanquish Neo using an EasySpray source. Data were analyzed using Proteome Discoverer (PD 3.1, Thermo Fisher). After injection of 250 ng HeLa digest, the ASTRAL identified 10,000 proteins in only 30 minutes, whereas the E480 identified 7,000 proteins in 60 minutes. By increasing both the sample material 500 ng and the gradient 120 minutes, the E480 could identify up to 8,500 proteins. We then analyzed different mixtures of 3 commercial digests (HeLa (Thermo), yeast (Thermo) and E. coli (Waters)) to measure the accuracy of the quantification with a fixed and variable proportion for HeLa and the two other species, respectively. We used a 9-minute gradient, corresponding to about 160 samples analyzed per day, to monitor the ASTRAL performance in a high-throughput analysis. The three proteomes were each time identified in the different mixtures with a total of 10,000 proteins identified per mixture. The theoretical ratios were also conserved with a minimum precision of 80%. Last but not least, we tested the ASTRAL coupled to vanquish Neo in single cell conditions against E480 equipped with field asymmetric ion mobility spectrometry (FAIMS) and Loload column. 250 pg of the previous digest mixtures were injected on the ASTRAL, allowing the identification of 3,500 proteins, despite the absence of the FAIMS. On the other hand, E480 identified up to 2,093 proteins but with four times more material (1 ng). Altogether, our preliminary results showed that the ASTRAL outperformed the performances of the E480 in terms of depth, identification of proteins and coverage of the proteomes, requiring lower amount of material and shorter turnaround time. We consider that the ASTRAL, and preferentially coupled to the FAIMS, will help to discover new proteins of interest that were never seen and to implement the proteomics in clinical routine. 778

779

P-II-0629 Circulating plasma and salivary extracellular vesicles proteomes indicate prognostic signature in oral cancer Carolina Moretto Carnielli1, Daniela Campos Granato1, Sami Yokoo1, Fábio Malta de Sá Patroni1, Ana Carolina Prado Ribeiro2,3, Thais Brandão2, Miyuki Uno4, Adriana Paes Leme1 1Brazilian Center for Research in Energy and Materials - CNPEM, Brazillian Biosciences National Laboratory - LNBio, Campinas, Brazil 2University of São Paulo, ICESP, Serviço de Odontologia Oncológica, São Paulo, Brazil 3Universidade Brasil, Fernandópolis, Brazil 4University of São Paulo, ICESP, Faculdade de Medicina, São Paulo, Brazil Oral squamous cell carcinoma (OSCC) is the most common type of head and neck malignant tumor and the fifth most frequent cancer in Brazil with high prevalence and morbidity. Research efforts have focused on the discovery of signature prognostic markers using minimally invasive liquid biopsy to guide the decision of the appropriate treatment for OSCC patients. In particular, extracellular vesicles (EVs) are important mediators in the intercellular communication with critical role in the establishment of pre-metastatic niche. In this study, we compared the proteome of circulating EVs from saliva (n = 32) and plasma (n = 31) between patients with (N+) or without lymph node metastasis (N0) to investigate metastasis signature in EVs. The majority of the 1,071 and 1,414 proteins quantified in salivary and plasma EVs were associated with the enrichment of antigen processing and cell-cell adhesion biological processes, and platelet degranulation and complement activation, respectively. From the 49 salivary EV proteins with difference in abundance between N0 and N+ samples (p<0.05, Student's t test), 46 were able to stratify these patients (AUC>70%) using logistic regression (LR) model or random forest (RF). Likewise, 130 proteins were differential between the plasma EVs proteomes, from which 114 were able to classify these patients in LR and RF models. Moreover, nearly 20 proteins from each dataset were associated to clinicopathological features (Linear regression, P value < 0.05, R < −0.7 or 0.7 < R and R2 > 0.5) and were further verified using parallel reaction monitoring in independent cohorts (salivary EVs n= 20 N+ and 7 N0, plasma EVs n= 6 N0 and 11 N+; 54 proteins and 223 proteotypic peptides). In conclusion, salivary and plasma EV proteins are promising as prognosis signature markers to the management of OSCC patients. 779

780

P-II-0630 The forgotten success story of clinical proteomics: how autoantibodies discovered via proteomics tools are now key diagnostic biomarkers in neurology Christian Moritz1 1University Hospital of Saint-Étienne, Neurology Department, Saint-Étienne, France One major goal of proteomics is to detect biomarkers that may serve as diagnostic or prognostic tools. In this way, biomarkers might be used in clinical routine to accelerate diagnosis or to assist treatment decisions. Unfortunately, examples of clinically used biomarkers that were originally detected via proteomics are scarce. However, in one scientific domain, namely in autoantibody-related autoimmunity, proteomics has led to discoveries of several biomarkers that are applied in clinical routine diagnostics. These biomarkers are autoantibodies that can be found in serum or cerebrospinal fluid, for example. Interestingly, this research domain is underrepresented in the proteomics community. Here, my goal is to demonstrate and discuss autoantibodies that are used in clinical routine and that have been identified via proteomics tools. The current version of the autoantibody encyclopedia lists 200 known autoantibodies. It is extremely difficult to estimate how many of them are clinically used in routine diagnostics because some autoantibody biomarkers are only applied in a few countries or even hospitals. Hence, in this presentation, I want to focus on autoantibodies used in neurology. One of the first autoantibodies that were discovered in neurological diseases were paraneoplastic neurological syndromes, i.e., diseases of the nervous system that are secondary to cancer and involve T or B cell-related autoimmunity. Some of the autoantibodies have been identified via proteomics-like methods of the pre- proteomics era. More recently, important biomarkers have been discovered in demyelinating inflammatory polyneuropathies. The presence of anti-neurofascin or anti-contactin autoantibodies defines the entity of autoimmune nodopathies. Hence, their detection is mandatory for disease diagnosis. Regarding anti-neurofascin antibodies, the corresponding band of interest was detected in 2007 using the isolated glycoprotein fraction from human myelin upon affinity chromatography. The actual protein was identified via 2D gel electrophoresis and detection of the autoimmune targets by incubating the patient serum, followed by mass spectrometry. Anti- contactin autoantibodies have been detected in 2012 via immunoprecipitation of the autoantibody targets, one- dimensional gel separation, and identification of the protein in the additional band via mass spectrometry. Interestingly, these biomarkers have even changed the clinical perspective towards some diseases leading to entirely novel disease entities. In conclusion, autoantibodies are important diagnostic tools in neurology but also in other clinical domains. Until today, proteomics is a promising tool to discover novel autoantibody biomarkers. The discovery of clinically relevant autoantibodies is one of the forgotten but major success stories of our proteomics community. 780

781

P-II-0631 Optimizing capivasertib response prediction in cancer: unveiling resistance mechanisms through precision proteomics Neginsadat Mostolizadeh1, Negarsadat Mostolizadeh1, Gerald Batist1,2,3, Mark Basik2,3, Christoph H. Borchers1,4,3,5 1Jewish General Hospital, Segal Cancer Proteomics Centre, Lady Davis Institute for Medical Research, Montreal, Canada 2Jewish General Hospital, Gerald Bronfman Department of Oncology, Montreal, Canada 3McGill University, Division of Experimental Medicine, Montreal, Canada 4MRM Proteomics, Montreal, Canada 5McGill University, Department of Pathology, Montreal, Canada Introduction: The PI3K/AKT/mTOR pathway plays a role in more than half of breast cancers, making it a crucial focus for new treatments. Capivasertib (AZD5363), an effective inhibitor of all AKT isoforms, has shown promise in Phase III trials for both breast and prostate cancers. However, pinpointing the right patients for this treatment is challenging due to gene-based uncertainties. To address this, we're using advanced mass spectrometry (MS) techniques to explore how Capivasertib affects the proteomic landscape. Our goal is to gain insights that will improve patient selection and refine precision medicine approaches for cancer treatment. Methods: We've developed and validated MS-based targeted assays to quantify proteins associated with Capivasertib in FFPE tumor blocks and breast cancer cell lines. These assays use synthetic peptides for calibration and stable isotope-labeled standard peptides for quantitation, ensuring accurate and reproducible results. Additionally, we've created targeted MS-based assays to investigate off-target effects, helping us understand Capivasertib"s mechanisms and resistance, and refine patient selection criteria. Results: In a previous study using untargeted proteomic methods, distinct protein profiles were identified in PIK3CA-mutant tumors from patients receiving Capivasertib. Pathways like EIF2 signaling, eIF4/p70S6K, and the mTOR pathway were linked to these profiles and helped distinguish patients who benefited from Capivasertib. To confirm the association, targeted MRM-MS assays for 50 proteins were developed, distinguishing between Capivasertib-resistant and sensitive PIK3CA-altered breast cancer cell lines. However, in patient samples, the lower limit of quantification (LLOQ) in FFPE samples posed challenges. Thus, we developed more sensitive Evosep-MRM-MS and Evosep-PRM-MS assays using Evosep-Agilent 6495C and Evosep-timsTof-HT, respectively. To further understand Capivasertib"s effects, we are developing assays for absolute quantitation of cancer proteins like EGFR, PI3K, and KRAS, and for 1000 mutation events in these proteins. We have optimized 522 assays, enabling deeper exploration of Capivasertib"s molecular effects. Conclusions: Our study uses advanced MS techniques to investigate how Capivasertib interacts with cellular proteins. While there are challenges in quantifying proteins from patient samples, our targeted assays offer a more sensitive approach, needing only 1 μg of protein, making them suitable for clinical use. Moving forward, our research aims to delve deeper into Capivasertib"s impact on various cancer-related proteins and pathways. By expanding our targeted assays to cover a broader range of protein networks, we hope to better understand the complex interactions between Capivasertib and cellular processes, ultimately enhancing its therapeutic potential and application. Fig. 1 (optional) 781

782

P-II-0632 Quantification of Her2 expression in Her2-low tumors by targeted mass spectrometry Negarsadat Mostolizadeh1,2, Neginsadat Mostolizadeh1,2, Mark Basik1,3,4, Gerald Batist1,3,2,4, Christoph H. Borchers1,3,2,4 1McGill University, Division of Experimental Medicine, Montreal, Canada 2Jewish General Hospital, Segal Cancer Proteomics Centre, Lady Davis Institute for Medical Research, Montreal, Canada 3McGill University, Lady Davis Institute, Montreal, Canada 4Jewish General Hospital, Gerald Bronfman Department of Oncology, Montreal, Canada Introduction: In approximately 15% of breast cancer cases, ERBB2 gene amplification leads to an overexpression of human epidermal growth factor receptor 2 (HER2) protein, resulting in a worse prognosis unless patients receive HER2-targeted therapy. However, roughly half of breast cancer cases exhibit low levels of HER2 expression without ERBB2 amplification. These are termed HER2-low. Unfortunately, these cases are excluded from HER2-targeted therapy. Nevertheless, there is hope with the emergence of new drugs targeting HER2, which have shown promising results in treating HER2-low breast cancers. This underscores the need for more accurate methods of determining HER2 status, given the limitations of current testing techniques such as IHC. To address these challenges, our efforts have been focused on developing a targeted mass spectrometry-based assay for the HER2 peptide, GLQSLPTHDPSPLQR. Our hypothesis is that this approach will provide the necessary assay sensitivity to quantify HER2 in HER2-low tumors, thereby refining patient selection for treatment with HER2-targeting antibody-drug conjugates. Method: In this ongoing study, we refined and validated a PRM-PASEF assay to measure HER2 levels using clinical FFPE samples from both patient-derived xenografts (PDX) and cancer patients. After standard assessments such as H&E staining and IHC, we used nanoflow liquid chromatography and PRM-mass spectrometry to quantify HER2 in diverse HER2 status samples. The final step involves correlating the HER2 status (based on our assay) with the IHC results, ensuring its accuracy and reliability. Result: The PRM-PASEF assay, targeting the GLQSLPTHDPSPLQR peptide, was developed for HER2 protein quantification. Characterization through response curves used synthetic peptides for calibration and isotope- labeled standards for quantitation, covering almost 3 orders of magnitude across 5 concentration points. The quantification of endogenous concentrations of the HER2 peptide revealed that the majority of values were above the lower limit of quantification (LLOQ). Moreover, a significant positive correlation was found between normalized HER2 concentrations and their IHC results. Conclusion: Our targeted assay for HER2 quantification shows potential in improving HER2 status assessment, especially in distinguishing HER2 low cases from negative ones. This could reduce the need for repeat IHC or FISH tests, lowering diagnosis time and costs for breast cancer patients. The simplified assay procedure also can lower the barriers to entry for the clinical application of the targeted assay.. 782

784

P-II-0633 Serum proteomics of mother-infant dyads carrying HLA-conferred type 1 diabetes risk Robert Moulder1,2, Santosh Bhosale1, Tomi Suomi1, Terhi Ruohtula3, Jarno Honkanen3, Suvi Virtanen4, Ilonen Jorma3, Laura Elo5, Knip Mikael1, Riitta Lahesmaa5 1University of Helsinki, Faculty of Medicine, Turku, Finland 2University of Turku, Turku Bioscience, Turku, Finland 3University of Turku, Institute of Biomedicine, Turku, Finland 4DepartmeFinnish Institute for Health and Welfarent of Public Health and Welfare, Department of Public Health and Welfare, Helsinki, Finland 5University of Turku, Turku Bioscience Centre, Turku, Finland During pregnancy progression and early childhood, in-utero and dietary factors can make important contributions towards health and development. In relation to studies of childhood diseases, serum proteomics of maturing infants can provide important insights, which, together with perinatal proteomes could reveal further biological perspectives. In the present study, serum proteomics analyses of mother-infant dyads with HLA-conferred susceptibility to type 1 diabetes (n = 22), weaned to either an extensively hydrolyzed or regular cow"s milk formula, were made to determine differences between feeding groups and changes in infancy. Sera from the beginning of third trimester, the time of delivery, 3 months postpartum, cord blood and collected from the infants at 3, 6, 9 and 12 months were analyzed by LC-MS/MS. Correlations were detected within the dyads on the basis of comparison of the ranked protein intensities. In addition, a range of perinatal and age-related changes were distinguished. A number of significant correlations with intestinal permeability were also noted. 784

785

P-II-0634 Proteomic signatures of the eutopic endometrium of patients with deep bowel endometriosis Lidia Myung1, Rosanne Kho1, Raphaela Oliveira1, Wager Fontes1, Carlos Ricart1, Marcelo Sousa1, Isabelle Luz1, Denise Guerreiro1, Maria Julia Bezerra1, Kathiane Augusto1, Leonardo Bezerra1, Andrea Lomagno1, Ishak Yosuf1, Dario Silvestre1, Aline Maria Araujo Martins1, Arlindo Moura1, Mauricio Abrão1 1Federal University of Ceará, Animal Science, Fortaleza, Brazil Endometriosis is characterized by the presence of endometrial stroma and/or glands outside the uterine cavity, affecting women with severe dysmenorrhea and infertility. Deep bowel endometriosis relates to acute symptoms due to intestinal infiltrative lesions and high density of nerve fibers. The present study analyses the proteome of eutopic endometrium of patients with stage II (EII, n = 8) and stage IV (EIV, n = 6) deep bowel endometriosis as compared to healthy women (HW, n = 7). University ethics committees approved the research. Samples were collected during laparoscopic surgery of EII and EIV patients, and from women having tubal ligation (HW). Digested endometrium proteins were analyzed by mass spectrometry, and resulting files, processed by SEQUEST HT algorithm and Proteome Discoverer 2.5 (Thermo Fisher Scientific, USA). Protein profile comparisons and definition of differentially expressed proteins (DEPs) were based on Peptide Spectrum Matches (PSMs), Linear Discriminant Analysis (LDA), hierarchical clustering (NW vs EII; HW vs EIV and EII vs EIV), and integrated intensity area of the peaks (Progenesis QI; Nonlinear Dynamics, USA). MetaboAnalyst platform was used for multivariate analysis, followed by Tukey's test (RStudio). Bioinformatics tools evaluated protein functional annotations, protein-protein interaction networks, hubs and bottlenecks. There were 1448, 1623 and 1616 proteins identified in eutopic endometria of HW, EII and EIV patients, respectively. Considering 144 DEPs, 19 showed the same trend of expression based on PSMs, LDA, and intensity values. In this group, 18 proteins were upregulated and C3 had lower content in EII and EIV subjects. Represented modules associated with protein profiles protein synthesis and translation, chromatin organization, RNA splicing/processing, cytoskeleton organization, redox homeostasis, metabolism, protein folding and stress response, immune system, vesicle- mediated transport and proteolysis. From healthy to EII and EIV subjects, most modules showed increased expressions. However, apolipoproteins, peptidase/protease inhibitors and immune system components were downregulated in endometriosis patients. Transition from stage II to stage IV was characterized by upregulation of proteins from the SnRNP complex, involved in RNA splicing/processing. Proteins such as SNCA, APOB and B2M were found in HW hubs/bottlenecks, while KRAS, in hub/bottlenecks for EII patients. Progesterone receptor and MAOA resulted in EII-specific bottlenecks, while RELA was among EII-specific hubs and CASP3 was present in EIV- specific hubs/bottlenecks. SEC13 and SEC31A were common in EII and EIV hubs/bottlenecks, characterizing the early stages of the disease. In conclusion, this study highlights the proteome landscape of endometria of patients with deep bowel endometriosis. Proteins characterizing the disease relate to vital functional modules, being potential targets for early diagnosis and development of new therapies. 785

786

P-II-0635 Evaluation of new sample preparation methods associated with DIA analysis on Exploris 480 and Tims Tof SCP mass spectrometers to increase proteomic depth on plasma Emmanuelle Mouton-Barbosa1, Marlene Marcellin2, Clément Robin2, Karima Chaoui2, Odile Bruley-Schiltz2, Anne Gonzalez de Peredo2 1University Toulouse / CNRS, Proteomics, Toulouse, France 2UPS CNRS IPBS, Proteomics, Toulouse, France Human plasma contains biological information from all organs of the human body and therefore offers invaluable possibilities for clinical biomarker detection. However, detecting low-abundant proteins of interest in plasma samples is challenging. The heterogeneity and complexity of plasma, with high dynamic range in protein concentrations, represent strong hurdles for LC-MS-based proteomics, limiting access to full proteome information. Therefore, in recent years, sample preparation methods based on the depletion of highly abundant proteins, on the enrichment of low abundant proteins, or on the selective enrichment of plasma microvesicles, have been reported in order to increase plasma proteomic coverage. In addition, DIA-based methods on modern fast-sequencing mass spectrometers also provide better proteomic depth for such applications. We used plasma samples obtained from five different patients and evaluated three commercial protocols for human plasma preparation, including: - depletion columns for single-step removal of 14 high-abundant proteins representing approximately 95% of plasma total protein content (ThermoScientific) - the ENRICH-iST kit described to enrich low abundant proteins onto paramagnetic beads (Preomics) - the Mag-Net method based on MagReSyn® strong anion exchange (SAX) magnetic beads (ReSyn Biosciences) to capture extracellular vesicles from plasma We implemented DIA analysis workflows based either on the Exploris TM 480 from Thermo Fisher Scientific, or on the timsTOF SCP from Bruker, associated with DIA NN 1.8 data processing, and compared the results obtained using these three protocols with the direct analysis of neat plasma. All three methods improved significantly the proteomic characterization of plasma, allowing the identification of 800 to 2900 proteins/run, depending on the protocol and the patient, and giving access to the detection of some very low abundant molecules. The two preparation methods based on enrichment of proteins on magnetic beads (ENRICH-iST and Mag-Net) proved to be more efficient than the depletion approach, and they also appeared to be quite complementary, each of them improving the detection of specific classes of proteins. Altogether, more than 3000 plasmatic proteins could be identified and quantified from one patient using a combination of those sample preparation methods. 786

787

P-II-0636 Why is the peripartum heart a substrate for dilated cardiomyopathy? A data driven investigation of molecular and cellular remodelling. John Mulvey1, Carolin Sailer1, Jonathan Achter1, Konstantin Kahnert1, Seçil Erbil Bilir1, Gregory Milburn2, Ross Bretherton3, Charles Pyke3, Finn Gustafsson4, Kenneth Campbell2, Kate Herum3, Alicia Lundby3 1University of Copenhagen, Copenhagen, Denmark 2University of Kentucky, Lexington, KY, United States 3Novo Nordisk, Måløv, Denmark 4Rigshospitalet, Department of Cardiology, Copenhagen, Denmark Peripartum cardiomyopathy (PPCM) is a rare form of acute heart failure with a high mortality rate that develops in women towards the end of pregnancy or early in the postpartum period. Heart transplantation is currently the only cure for end stage disease, and clinical practice is focused on the management of symptoms. The mechanisms through which the disease progresses are not well understood, and despite similar clinical characteristics to dilated cardiomyopathy (DCM) of other aetiologies it is not known how the molecular remodeling that drives the pathology differs between these two groups. We obtained heart tissue from female patients with end stage disease receiving either heart transplantation or implantation of a left ventricular assist device, or from organ donors without a history of heart disease as a control group. Groups were matched for BMI and haemoglobin A1c status. We utilised high resolution mass-spectrometry based bulk proteomics using TMTpro labelling. At the proteomic level, we observed regulation of canonical markers of end-stage heart failure in both PPCM and DCM hearts, consistent with similarities in the clinical phenotype. Discriminant analysis identified proteins regulated specifically in PPCM rather than in heart failure in general, and intersection with cellular gene expression profiles suggested altered mast cell activation in PPCM which was confirmed by immunohistochemistry. In addition, we performed droplet based single nucleus RNA sequencing and spatial transcriptomics to define the repertoire of molecular and cellular modelling in PPCM, in DCM and in contrast to control human heart tissue. Integrating transcriptomic data modalities allows us to interrogate not only shifts in gene expression patterns within cell types distinct to PPCM, but also to define the importance of the local spatial context - both in terms of the cellular environment and the local gene expression. By elucidating specific molecular mechanisms we aim to enable the development of therapeutic strategies specifically for PPCM. 787

788

P-II-0637 Peptidomic analysis of cerebrospinal fluid samples reveals new biomarker candidates for amyotrophic lateral sclerosis Besnik Muqaku1, Johannes Dorst2, Maximilian Wiesenfarth2, Simon Witzel2, Markus Otto3, Albert Ludolph1,2, Patrick Oeckl1,2 1German Center for Neurodegenerative Diseases (DZNE), Ulm, Germany 2Ulm University Hospital, Department of Neurology, Ulm, Germany 3Martin-Luther-University Halle-Wittenberg, Department of Neurology, Halle (Saale), Germany Introduction: Amyotrophic lateral sclerosis (ALS) is a devastating and progressive neurodegenerative disease that affects motor neurons and leading to death within 3 years. Diagnosis is based on clinical symptoms, the pathophysiology is still unclear for most patients and treatment options are limited. Therefore, there is an urgent need for fluid biomarkers to support basic research, diagnosis and drug development in ALS. Methods: We applied mass-spectrometry-based peptidomic analysis in cerebrospinal fluid (CSF) samples of ALS and non-neurodegenerative control patients (Con) from a discovery (24 ALS, 24 Con) and validation cohort of patients (67 ALS, 42 Con). Peptides were isolated and purified from CSF for unbiased peptidomics in the discovery cohort and analyzed via mass spectrometry (MS). Selected peptide candidates were evaluated by targeted parallel reaction monitoring (PRM) in the validation cohort including stable isotope-labelled peptides. Analyses were performed with an UltiMate 3000 RSLCnano system and an Orbitrap Exploris 480 MS equipped with FAIMS interface. Screening data were processed with the PEAKS software and targeted PRM data were analyzed in Skyline. Peptide abundance changes in the validation cohort were correlated with clinical parameters. Results: We identified 33605 peptides in CSF samples from the discovery cohort and 56 of them were significantly regulated in ALS compared to controls. A coefficient of variation (CV) of 6.1% was obtained for the number of identified peptides over all patients and the abundance dynamic range of peptides in patient samples spanned eight orders of magnitude. Systematic selection for the best candidates revealed a targeted PRM method with 8 peptides, which was validated in terms of intraassay variation (CV 5-12%), dilution stability (accuracy 80-119%) and freeze-thaw stability (accuracy 89-148%). The targeted PRM confirmed the abundance changes of the discovery experiment for all peptides. Combination of all peptides in a logistic regression model led to an area under the curve value of 98.4% for the discrimination of ALS from controls. PRM data of the neurofilament light chain (NfL) peptide strongly correlated with an established NfL immunoassay (Ella) (r=0.968, n=109). Conclusion: We report eight peptides derived from seven proteins as biomarker candidates for ALS. The peptides are derived from proteins with different function and their determination with our PRM method provides the opportunity for simultaneous investigation of key processes in ALS and other neurodegenerative diseases (neurodegeneration, metabolic alteration, muscle weakness/atrophy and synaptic function or the secretory pathway). Future studies must clarify their clinical value for diagnosis, monitoring disease progression and treatment response, their changes in the preclinical phase of ALS and whether the peptide level changes are indicative of the parent protein concentration or reflect changes in protein processing. 788

789

P-II-0638 Differentially expressed protein biomarkers in early and advanced stage of young triple-negative breast cancer patients Shamim Mushtaq1, Moazzam Shahid1 1Ziauddin University, Biochemistry, Karachi, Pakistan Breast cancer (BC) claims the lives of half a million women every year and is the most common cause of death in the developing world. In 2019, it was estimated that BC alone accounts for 15% of all cancer deaths in younger women (aged < 45 years old) with advanced stage with lung metastasis. According to World Health Organization & International Union against Cancer, in Asia high cancer related deaths will be observed in 2020 whereas burden will be reduced in western countries due to awareness about the disease, better health facilities and advanced treatments. In the last 15 years, it has been reported that incidence of BC is increasing by 1.1% among Asian compared to US population from 2003 to 2012. To date, several BC biological subtypes are reported so far which are associated with different treatment responses. The heterogeneity and diversity of BC reflect with these different subtypes, includes Luminal A (23.7% prevalence) and B (38.8% prevalence) that has pathological estrogen receptor (ER+)-positive tumors, the human epidermal growth factor receptor 2 (HER2) (11.2% prevalence) and triple-negative breast cancer (TNBC) (25% prevalence). According to Shaukat Khanum Memorial Cancer Hospital and Research Centre – Pakistan, ten years data showed that among 636 BC patients, 30.5% had TNBC who were Therefore, there is a dire need to explore and develop therapeutic targets for the treatment of early TNBC. Since the last decade unfortunately, there has been little success to understand the complexity of TNBC and to discover new biological therapeutic targets, however conventional chemotherapy is the only choice of treatment for TNBC patients. Many investigators revealed advances in multiomics (multiple "omes", e.g genome, proteome, transcriptome, epigenome, and microbiome) which were later identified as actionable targets and increase prevalence in TNBC patients. However, various drugs have been identified so far which are related to a particular diagnostic and prognostic biomarker. For example Epidermal growth factor receptor ( EGFR or ErbB-1), HER-2/neu (ErbB-2), HER-3 (ErbB-3), and HER-4 (ErbB-4). Protein Transglin-2 (TAGLN 2 ) and Profilins-1 (Pfn-1 ) are the ubiquitously expressed large family of proteins present in all eukaryotes enabling actin cytoskeletal re- organization. It is known that oncogenic transformation of cells is accompanied by alteration in the actin cytoskeleton. There are causal connections between altered expression of actin cytoskeletal regulators and cancer progression. Our case control study identified TAGLN-2 and Pfn-1 proteins for the first time in TNBC blood by mass spectrometry. Both TAGLN-2 and Pfn-1 proteins are differentially expressed in early and advance stages of TNBS patients which could be potential predictors or therapeutic target for TNBC. 789

790

P-II-0639 Plasma proteome profiling revealed altered levels of complements in the plasma of interverterbal disc degeneration patients Sharon Miracle Nayagam1, Narmatha Devi Palraj1, Sunmathi Rajendran1, Murugesh Easwaran1, Ganesh Selvaraj1, Karthik Ramachandran2, Srivijay Anand K S2, Chitraa Tangavel1, Rajasekaran Shanmuganathan2 1Ganga Orthopedic Research and Education Foundation (GOREF), Coimbatore, India 2Ganga Medical Center and Hospitals Pvt Ltd, Spine Surgery and Orthopedics, Coimbatore, India Background: The complement system, a network of proteins regulating immune and inflammatory responses, has been implicated in intervertebral disc degeneration (IDD) pathogenesis. Tissue-specific complement activation in degenerated discs is well-established, but systemic expression in IDD patients' plasma is less understood. This study investigates the levels of complement proteins in disease plasma to understand their implications during the disease progression. Purpose: To understand the (dis)similarities between the proteome profiles of HV-plasma (healthy volunteers) and DD-plasma (patients with degenerative disc disease). Materials and Methods: Eleven healthy volunteers (HV) and 39 patients with disc degeneration (DD) were recruited. Plasma samples from the HV group served as controls. A high-throughput mass spectrometric approach was used for plasma proteome profiling. Results: Demographic data showed comparable sex ratios, BMI, blood profiles, and dietary patterns between healthy volunteers (HV) and DD patients. Elevated levels of C-reactive protein (CRP) and erythrocyte sedimentation rate (ESR) in the DD group indicated significant inflammation (p < 0.05). Proteomic analysis identified 691 proteins in DD-plasma and 526 in HV plasma, with 470 proteins common to both. The corresponding DD-tissue samples revealed 1842 proteins. The PLS-DA model demonstrated a 19.4% variance between HV-plasma and DD-plasma proteins. Further Pathway enrichment analysis highlighted the involvement of 47 complement proteins in the complement and coagulation cascades. Analysis of complement pathways revealed significant upregulation of classical pathway components, C4BPA, C4a, C4b, C2, C3, and C3/C5 convertase, with a notable downregulation of C1q. The lectin pathway showed significant upregulation of MASP1. The alternate pathway exhibited upregulation of MAC constituents (C5, C6, C7, C8A, C8B) and regulatory proteins (CFH, CFI), suggesting their role in immune defense. Interestingly in the corresponding disease tissue, we observed downregulation of all the complement proteins discussed above except for C1q which is upregulated. Conclusion: This study profiled the proteomes of HV-plasma, DD-plasma, and DD-tissue, revealing altered complement activation (C1q, C1r, and C1s) during degeneration. Upregulation of MAC constituents and regulatory proteins suggests involvement in inflammation. The correlation between complement levels inside the DD-tissue and in the DD-plasma has to be studied further to understand the actions of complement proteins in an IDD context. Clinical Significance: The inverse correlation between C1q levels in plasma and tissue of IVDD patients could serve as a promising candidate for the prognosis of IVDD progression. C1q expression levels need to be studied in various disease stages in a larger population to be recommended as a therapeutic target. Keywords: IVDD, proteomics, complement system, plasma proteins, inflammatory markers, CRP, ESR 790

792

P-II-0640 Plasma proteome profiling revealed altered levels of complements in the plasma of interverterbal disc degeneration patients Sharon Miracle Nayagam1, Narmatha Devi Palraj2, Sunmathi Rajendran2, Murugesh Easwaran2, Ganesh Selvaraj2, Karthik Ramachandran3, Srivijay Anand K S3, Chitraa Tangavel2, Rajasekaran Shanmuganathan3 1Ganga Orthopedic Research and Education Foundation (GOREF), Proteomics and Metabolomics, Coimbatore, India 2Ganga Orthopedic Research and Education Foundation (GOREF), Coimbatore, India 3Ganga Medical Center and Hospitals Pvt Ltd, Spine Surgery and Orthopedics, Coimbatore, India Background: The complement system, a network of proteins regulating immune and inflammatory responses, has been implicated in intervertebral disc degeneration (IDD) pathogenesis. Tissue- specific complement activation in degenerated discs is well-established, systemic expression in IDD patients' plasma is less understood. This study investigates the levels of complement proteins in disease plasma to understand their implications during the disease progression. Purpose: To understand the (dis)similarities between the proteome profiles of HV-plasma (healthy volunteers) and DD-plasma (patients with degenerative disc disease). Materials and Methods: Eleven healthy volunteers (HV) and 39 patients with disc degeneration (DD) were recruited. Plasma samples from the HV group served as controls. High-throughput mass spectrometric approach was used for plasma proteome profiling. Results: Demographic data showed comparable sex ratios, BMI, blood profiles, and dietary patterns between healthy volunteers (HV) and DD patients. Elevated levels of C-reactive protein (CRP) and erythrocyte sedimentation rate (ESR) in the DD group indicated significant inflammation (p < 0.05). Proteomic analysis identified 691 proteins in DD-plasma and 526 in HV plasma, with 470 proteins common to both. The corresponding DD-tissue samples revealed 1842 proteins. The PLS-DA model demonstrated a 19.4% variance between HV-plasma and DD-plasma proteins. Further Pathway enrichment analysis highlighted the involvement of 47 complement proteins in the complement and coagulation cascades. Analysis of complement pathways revealed significant upregulation of classical pathway components, C4BPA, C4a, C4b, C2, C3, and C3/C5 convertase, with a notable downregulation of C1q. The lectin pathway showed significant upregulation of MASP1. The alternate pathway exhibited upregulation of MAC constituents (C5, C6, C7, C8A, C8B) and regulatory proteins (CFH, CFI), suggesting their role in immune defense. Interestingly in the corresponding disease tissue, we observed downregulation of all the complement proteins discussed above except for C1q which is upregulated. Conclusion: This study profiled the proteomes of HV-plasma, DD-plasma, and DD-tissue, revealing altered complement activation (C1q, C1r, and C1s) during degeneration. Upregulation of MAC constituents and regulatory proteins suggests involvement in inflammation. The correlation between complement levels inside the DD-tissue and in the DD-plasma has to be studied further to understand the actions of complement proteins in an IDD context. Clinical Significance: The inverse correlation between C1q levels in plasma and tissue of IVDD patients could serve as a promising candidate for prognosis of the IVDD progression. C1q expression levels need to be studied in various disease stages in a larger population to recommend as a therapeutic target. Keywords: IVDD, proteomics, complement system, plasma proteins, inflammatory markers, CRP, ESR 792

794

P-II-0641 Comparative proteomic analysis of breast cancer and normal tissues with deep plasma profiling using the Mag- Net™ workflow Adele Nel1,2, Andrea Ellero3,2, Sindisiwe Buthelezi4, Ireshyn Govender2,4, Previn Naicker4, Iman van den Bout1, Justin Jordaan2, Stoyan Stoychev2, Christine Wu5, Michael J MacCoss5 1University of Pretoria, Department of Physiology, Pretoria, South Africa 2Resyn Biosciences, Johannesburg, South Africa 3University of Pretoria, Department of Pharmacology, Pretoria, South Africa 4Council for Scientific and Industrial Research, Pretoria, South Africa 5University of Washington, Seattle, WA, United States With breast cancer being responsible for the most cancer deaths worldwide, improvements in understanding of the molecular underpinnings of the disease are paramount. Plasma proteomic profiling has emerged as an invaluable tool in the medical research field due to plasma"s accessibility as a biofluid. However, robust profiling of the plasma proteome is challenged by its high dynamic range of protein abundances and the lack of high- throughput methods with sufficient depth. Mag-Net™, a high-throughput, cost-effective workflow for deep profiling of the plasma proteome based on enrichment of membrane-bound vesicles has been developed and automated (Wu et al., 2023). Here we set out to compare protein signatures between normal and breast cancer tissue samples and subsequently, using Mag-Net™, profile the plasma proteome of breast cancer patients to find tumour reporting proteins. Blood, tumour tissue and normal tissue samples (n=15) were collected from consenting patients diagnosed with infiltrative ductal carcinoma undergoing surgery at Kalafong Tertiary Provincial Hospital. Plasma isolation was performed directly after blood collection following the EDRN SOP and stored at -80°C. Following homogenisation of fresh frozen tumour and normal tissue samples, protein extraction, capture, clean-up and digestion were performed using a semi-automated workflow on a KingFisher Flex with hydrophilic affinity beads (MagResyn HILIC). High abundant protein depletion and capture of membrane-bound vesicles were performed on patient plasma samples using the semi-automated Mag-Net™workflow on a KingFisher Flex system with MagResyn SAX beads. Peptides were loaded onto Evotips and analysed using an Evosep One coupled to a Bruker timsTOF HT system. We previously reported using the Mag-Net™ workflow on plasma collected from patients with pancreatic ductal adenocarcinoma. Here we showed that, compared to a neat plasma workflow, Mag-Net™ identified double the number of proteins, provided clear separation of disease versus control patient groups and identified more cancer relevant pathways. Experiments on the breast cancer patient cohort is ongoing. 794

795

P-II-0642 Protein-profiling of liquid biopsies for the differential diagnosis of pancreatic diseases Thao Vi Nguyen1, Thorben Sauer1, Kim Honselmann2, Rüdiger Braun2, Katharina Kern1, Michael Kohl1, Ulrich Wellner2, Timo Gemoll1 1University of Lübeck, Section for Translational Surgical Oncology and Biomaterial Banks, Luebeck, Germany 2University Medical Center Lübeck, Department of Surgery, Luebeck, Germany Pancreatic ductal adenocarcinomas (PDAC) form the largest subset of pancreatic cancers and are considered one of the most lethal diseases with a global overall 5-year survival rate of 9 %. For the differential diagnosis of benign neoplasia, chronic pancreatitis or carcinomas, often an invasive, histopathologic examination needs to be performed. Consequently, the development of a non-invasive liquid biopsy-based method for the differential diagnosis of pancreatic diseases would represent a significant clinical advance. The objective of this study was to identify differentially expressed serum proteins by comparing protein profiles between healthy controls and patients with pancreatic diseases (n=73, e.g. PDAC, IPMN, …) using LC-MS/MS (liquid chromatography-mass spectrometry/mass spectrometry) in data-independent acquisition mode (SWATH). 280 serum proteins were identified, quantified and revealed distinct separations between PDAC and benign pancreatic lesions such as pseudocysts, cystadenomas, and IPMNs (intraductal papillary mucinous neoplasia) using clustering analyses. Statistical evaluation revealed eight significant protein expression differences between the healthy, benign, and malignant group (FC > 1.5 & q-value < 0.05). Analysis through Panther pathway analysis showed that significant protein expressions of the malignant group were associated with processes like the coagulation cascade or inflammatory processes. This study demonstrates significant differences in the global serum protein expression between patients with malignant and benign pancreatic diseases. Protein targets were identified by quantitative mass spectrometry and will be further validated for their potential as biomarkers. 795

796

P-II-0643 Cell lines recapitulate the kinase activity profiles of colorectal tumor tissue and predict drug response Rei Noguchi1, Yuki Adachi1, Takuya Ono1, Yuki Yoshimatsu1, Kazuki Sasaki2, Tadashi Kondo3 1Tochigi Cancer Center Research, Department of Patient-Derived Cancer Model, Utsunomiya, Japan 2Tochigi Cancer Center Research, Department of Oncopeptidemics, Utsunomiya, Japan 3National Cancer Center Research Institute, Division of Rare Cancer Research, Tokyo, Japan Background and Aims: Colorectal cancer (CRC) is the 3rd most common cancer in the world. Various protein and genetic markers have been reported to predict chemotherapy efficacy. However, treatment response prediction poses a challenge because of CRC heterogeneity. Therefore, precise and individualized treatment strategies with approaches different from proteomics and genomics are needed. Given the kinase dysregulation during oncogenesis, a battery of kinase inhibitors has been developed and used in clinics. Since kinase network dysregulation is caused by a set of kinases and influenced by various mechanisms, direct measurement of global kinase activity represents a practical way to infer identify upstream kinases. We aimed to analyze to what extent the kinase activity profiles of CRC are maintained in established CRC cell lines and whether kinase activity reflects the drug response to kinase inhibitors. Methods: Comprehensive kinase activity assay was conducted on six CRC cell lines, CRC tumors (n=16), and matched normal tissues (n=16) with a three-dimensional peptide array. Kinase-substrate enrichment analysis was performed to compare kinase activity values between tumor and normal tissues and to infer the relative activities of kinases. A drug sensitivity test was performed on the six CRC cell lines with 60 FDA-approved tyrosine kinase inhibitors. The kinase activity profiles and drug sensitivity data were integrated to identify and characterize target kinases and corresponding kinase inhibitors. Results: A total of 43 tyrosine kinases were inferred through the phosphorylation intensity data from 181 substrate peptides. Increased kinase activity was found in tumors relative to their corresponding normal tissues. Six kinases including VEGFR1, PDGFRB, JAK3, SYK, MAP2K4, and INSR were aberrantly activated in the tumors. CRC cell lines mirrored the kinomic aberrations observed in tumor tissues. All the samples were grouped into two clusters depending on the overall kinase activity status, hyper-activated and lower-activated. This was also the case with the cell lines. Drug screening data indicated that nine drugs targeting 13 kinases strongly suppressed cell viability. Upon profiling the response of six CRC cell lines to 60 tyrosine kinase inhibitors, two clusters emerged that showed different degrees of drug sensitivity, responsive and non-responsive groups. Integrating the kinase activity profiles and the drug sensitivity data, global kinase activity was found to be relatively high in the responsive groups, while it was relatively low in the non-responsive groups. Conclusions: Our pharmacokinomic approach revealed that kinase activity profiles of the cell lines recapitulate those of tumors. The kinase activity profile was able to predict sensitivity to tyrosine kinase inhibitors. Our integrative approach lends itself to personalized cancer medicine, demonstrating the potential of kinase activity profiling. 796

797

P-II-0644 A proteomic skin disease atlas for accurate and rapid AI-driven clinical decision making Thierry M. Nordmann1, Pia-Charlotte Stadler1, Matthias Neulinger2, Georg Wallmann1, Maximilian Zwiebel1, Laurie Eicher2, Sonja Senner2, Annesophie Boehm2, Mohammed Mitwalli2, Werner Kempf3, Michael Flaig2, Rudolf Stadler4, Doris Helbig5, Katrin Kerl-French2, Maximilian T. Strauss6, Takashi Satoh2, Lars French2, Matthias Mann1 1Max Planck Insitutue of Biochemistry, Proteomics and Signal Transduction, Planegg, Germany 2Ludwig-Maximilians-University Munich, Dermatology, Munich, Germany 3Kempf und Pfaltz Histologische Diagnostik, Zurich, Switzerland 4University of Bochum, Dermatology, Skin Cancer Center, Minden, Germany 5University Hospital of Cologne, Dermatology and Venereology, Cologne, Germany 6University of Copenhagen, Proteomics Program, Novo Nordisk Foundation Center for Protein Research, Copenhagen, Denmark A quarter of the world"s population suffers from chronic inflammatory skin diseases (ISDs). These diseases encompass a broad spectrum of heterogenous entities and a wide range of different diagnoses. As a consequence, ISDs are particularly challenging to diagnose accurately and timely, which is a prerequisite for the effective treatment of the underlying disease. Omic technologies in combination with artificial intelligence hold the promise to contribute substantially to clinical decision making and patient care. Here, we performed mass spectrometry-based proteomics with the latest generation mass spectrometer (Orbitrap Astral) from Thermo Fisher to generate a comprehensive proteomic atlas of all major inflammatory skin diseases. For this, we collected retrospective FFPE skin tissue biopsies and associated clinical metadata from 459 patients, so far. These FFPE tissue sections were processed for MS-based proteomic analysis using a semi-automated (Opentrons) workflow and integrated normalization strategies, with the ultimate goal in mind to enable this workflow at other locations in the near future. All samples were rapidly measured using the standardized 21 min gradient (60SPD) on the Evosep One connected to the Orbitrap Astral (Thermo) using an 8cm PepSep column. This robust setup yielded 10,364 proteins across all diseases measured, spanning five orders of magnitude and contained many low-abundant cytokines. With our comprehensive proteomic dataset and an independent additional cohort at hand we assessed the capability of our atlas to improve clinical decision making in "Erythroderma" (n = 115). Erythroderma is a potentially life-threatening condition defined by redness that affects more than 90% of the skin and is the result of an underlying inflammatory skin condition, including psoriasis, atopic dermatitis, pityriasis rubra pilaris, cutaneous T-cell lymphoma and drug reactions. The diagnosis of the underlying cause usually takes many weeks to months, despite the need of rapid and disease specific treatment initiation. Through integration of machine learning algorithms, our proteomic dataset effectively distinguished between the erythrodermic diseases with high accuracy and an AUC of up to 0.92, using up to 50 proteins. Feature selection was performed across these two independent cohorts by bootstrapping and permutation testing with 5-fold cross-validation. In addition, we detected novel and disease specific biology, including the induction of the IL17C/IL17RE Axis in pityriasis rubra pilaris. This study successfully highlights the potential of proteomics for AI-driven disease stratification in cutaneous inflammatory disorders and clinical decision making in general. Fig. 1 (optional) 797

798

P-II-0645 A translational, high-throughput proteomics platform for the deep analysis of plasma samples Estefanía Núñez1,2, Enrique Calvo1,2, María Gómez-Serrano3, Leticia Fernández-Friera4, Antonio Fernández-Ortiz1,5, Borja Ibañez1,2,6, Jesús Vázquez1,2 1Centro Nacional de Investigaciones Cardiovasculares Carlos III (CNIC), Cardiovascular Proteomics Laboratory, Madrid, Spain 2CIBER de Enfermedades Cardiovasculares (CIBERCV), Madrid, Spain 3Zentrum für Tumor- und Immunbiologie, Philipps Universität Marburg, Marburg, Germany 4Hospital Universitario HM Montepríncipe, Madrid, Spain 5Hospital Universitario Clínico San Carlos, Madrid, Spain 6IIS-Fundación Jiménez Díaz Hospital, Madrid, Spain Translational Proteomics, crucial for protein analysis in clinical samples, aims to achieve three key objectives: i) analyze extensive cohorts rapidly; ii) quantify a high number of proteins per sample using hypothesis-free and non-targeted analysis; and iii) integrate robust statistical models for quantitative data analysis. In an effort to attain these goals, we report here a novel, rapid and hypothesis-free workflow using multiplexed labelling that allows the analysis of thousands of plasma samples without the need for high-stable mass spectometers. This approach enables accurate and reproducible quantification of over a thousand of proteins. The described workflow successfully analyzed more than 1300 non-depleted plasma samples (across 130 TMT experiments), yielding highly reproducible results. The analysis of 600 proteins per experiment was achieved without peptide fractionation in a swift 6-hour timeframe per experiment, completing the entire process in just two weeks. Moreover, peptide fractionation, while extending the analysis to over 1000 proteins, required 30 hours per experiment, spanning a two-month duration. Although protein yield was higher in depleted samples, depletion introduced a quantification bias affecting approximately half of the plasma proteome, as revealed by hierarchical clustering analysis. Notably, a high correlation between biochemical quantitation and mass spectrometry measurements was observed for several proteins (p<1e-17). Additionally, the selection of a longitudinal prospective cohort (baseline and three-years follow up) enabled us to analyze the long-term stability of the plasma proteome. We observed that proteins from 84% of the individuals showed a significant correlation (FDR<5%) along time. Finally, the study identified gender-discriminating proteins. These results highlight the robustness of the new method for high-throughput quantitative analysis of the deep plasma proteome that could facilitate and improve clinical proteomics discovery in human blood plasma. 798

799

P-II-0646 ZenoSWATH proteomics and clustering analysis reveal different modulation patterns in fibroblasts from cystinosis patients by cysteamine Ignacio Ortea1, Lorena Rodríguez-Martínez2, Mónica Carrera3, Miguel González-Barcia2, Anxo Fernández- Ferreiro2, Jesús Mateos2 1ISPA, Proteomics Unit, Oviedo, Spain 2Health Research Institute of Santiago de Compostela (IDIS), Santiago de Compostela, Spain 3Spanish National Research Council (CSIC), Institute of Marine Research (IIM), Vigo, Spain Introduction: Cystinosis is a rare metabolic disease, where mutations in the CTNS gene, which codifies for a symporter that mediates export of cystine from lysosomes, result in the accumulation of cystine crystals in the cells. The metabolic imbalance caused by intra-lysosomal cystine accumulation potentially leads to serious complications leading to kidney transplantation. Up to one-third of the patients die before age 30 if untreated. in various organs, especially kidneys, eyes and muscles, eventually Cysteamine (CTA) therapy, which reduces cystine levels, is the primary treatment for cystinosis. CTA ameliorates the symptoms and delays the progression, improving patient quality of life. But it does not cure the disease, has serious undesirable side effects, and fails in preventing the end-stage renal failure. Thus, it is crucial to characterize the cellular events occurring in the affected cells, identifying the mechanisms involved in both the beneficial and adverse effects of cysteamine. Methodology: Skin fibroblasts from cystinosis patients and healthy controls were cultured with increasing concentrations of CTA (0, 15, 30, 75, 150 and 300 µM). Cells were lysated and the protein samples were prepared using the iST kit (PreOmics). 400 ng of the resulting peptidic digests were loaded on Evotips (Evosep) and analyzed by LC-MS in a ZenoTOF 7600 (SCIEX) coupled to a Evosep One (Evosep). A 60 SPD method (21 min total run time), using an 8 cm x 150 μm Performance column (Evosep) was used for LC separation. A variable ZenoSWATH DIA method, was used, combining a 25 ms TOF MS scan with 63 20 ms MS/MS scans in variable isolation windows (range 400-991 m/z). The ZenoSWATH runs were processed in a library-free workflow using DIA-NN v1.8.1, and the quantitative profiles of the quantified proteins were analyzed using VSClust v1.2 for obtaining the protein clusters that showed similar expression profiles. For functional analysis, STRING v12.0 and Metascape v3.5 were used for obtaining the detailed pathway and process enrichment for each protein cluster. Results: Six protein clusters exhibiting similar expression profiles across the increasing CTA concentrations were identified by VSClust after ZenoSWATH quantitation. Interestingly, for a great number of the protein changes found in cystinotic versus non-cystinotic patients, CTA is not reversing these changes, and in some case is even exacerbating these differences, providing an explanation for the shortcomings and the side effects in the therapeutic action of CTA. However, for some proteins CTA does seem to correct the effect of the disease, and we report the potential cellular mechanisms by which this beneficial effect of CTA could take place. 799

801

P-II-0647 Pharmaco-proteogenomics for synovial sarcoma towards novel therapy Julia Osaki1, Rei Noguchi1, Yomogi Shiota1, Yuki Adachi1, Shuhei Iwata1, Kazuki Sasaki2, Masaki Matsumoto3, Tadashi Kondo1 1National Cancer Center, Division of Rare Cancer Research, Tokyo, Japan 2Tochigi Cancer Center Research, Tumor Peptidomics Project, Tochigi, Japan 3Niigata University, Department of Omics and Systems Biology, Niigata, Japan Background: Sarcomas originate in various mesenchymal tissues, consisting of more than 100 subtypes with different histological, biological, and clinical characteristics. Given that these tumors generally lack targetable mutations, genomic technology has obvious limitations in sarcoma research. With this in mind, we conduct pharmaco-proteogenomics to develop novel treatments for sarcomas. Synovial sarcoma (SS) is defined by the SS18-SSX fusion gene, with no effective anti-cancer drug available for patients with advanced SS. The genomic aberrations useful for treatments have not been reported in SS, probably because the SS18-SSX fusion drives oncogenesis dominantly. Therefore, phenomic and proteomic approaches are worth attempting to develop novel therapy in SS. Objective: This pharmaco-proteogenomic study aims to discover innovative seeds for novel therapy in SS. Methods: Patient-derived cancer models, such as cell lines, organoids, and xenografts were established from fresh surgical specimens. Drug sensitivity tests were conducted using a drug library consisting of 214 anti-cancer drugs. DIA mass spectrometry, RNA-seq, and target sequencing were performed for proteomics, transcriptomics, and genomics, respectively. Results: In vitro models (cell lines and organoids) and in vivo models (xenografts) were established from six and three patients, respectively. We found that five anti-cancer drugs suppressed the proliferation of all cell lines examined at extremely low concentrations. Although the five drugs included a tyrosine kinase inhibitor approved for the treatment of a particular type of lung cancer with unique mutations, there were no corresponding mutations in the SS cells. Proteomic and transcriptomic backgrounds were collectively investigated to reveal the molecular signatures preserved in the models. Discussion: We need to consider the utility of omics techniques other than genomics, as genome medicine is limited in malignancies with no targetable mutations. The combination of drug sensitivity screening and the proteogenomic approach is powerful, because we can simultaneously obtain phenotype, proteotype, and genotype data. In such studies, the adequate and appropriate use of patient-derived cancer models is critical, and we successfully established such models. Our approach has three limitations. Firstly, we need to clarify how precisely our models recapitulate their original tumors. Secondly, the results of drug screening should be validated in animal models, and the link among different models should be optimized. Thirdly, we need to study multiple aspects of the proteome, such as expression, activity, and protein interactions. By addressing these challenges, our pharmaco-proteogenomics strategy will have versatile utility and can be applied to various types of malignancies. 801

802

P-II-0648 The European cancer moonshot "Biobank" – the heart of cancer studies Henriett Oskolás1, Henrik Lindberg1, Roger Appelqvist2, Johan Malm1, György Marko-Varga2 1Lund University, Department of Translational Medicine, Clinical Chemistry, Lund, Sweden 2Lund University, Biomedical Engineering, Lund, Sweden European Cancer Moonshot Lund Biobank: A Seven-Year Journey Over the past seven years, we have been developing the European Cancer Moonshot Biobank, amassing a significant archive of biological samples to advance cancer research. Our efforts have yielded over 1 million samples through clinical studies conducted at seven prominent clinical centers worldwide. Sample Handling, and Processing: • Unique Identification:Each sample is assigned a unique 2D barcode to ensure precise tracking and identification. • Automated Workflow:Sample processing is fully automated, with blood fractions aliquoted into 384- sample tube plates, enhancing efficiency and accuracy. The entire workflow from patient to stored processed samples takes less than 2 hours. 1D-coded primary tubes and 1D-coded 384-well racks with 2D-coded vials. All information is scanned at each processing step to record the information about each blood fraction. Each donor generates 192 vials of blood and blood components, each with 70µL. All fractioning is done on a Hamilton Microlab STAR Line workstation. • Robotic Sample Storage: Samples are stored at -80°C LiCONiC Kiwi 1k5 ULT automated freezer using state-of-the-art robotic systems, maintaining optimal conditions for long-term preservation and integrity. Each aliquoted fractions samples can be picked individually and retrieved from the freezer. • Data Storage: All information regarding patients" clinical parameters and demographics are stored in a SQL –database using RedCap as a Web based Electronic Data Capture tool and all other information about the processed samples are stored in our own biobank tracking system. Our biobank collections have been intense and play a crucial role in advancing medical research by providing access to large amounts of our biological specimens with the respective associated clinical data. This robust infrastructure supports our ongoing research activities, providing invaluable resources for cancer studies and contributing to the global fight against cancer. 802

803

P-II-0649 Predicting abdominal aortic aneurysms by combining mass spectrometry-based proteomics with clinical data Nicolai Bjødstrup Palstrøm1,2, Amanda Jessica Campbell1,2, Lars Melholt Rasmussen1,2, Mette Sørensen3, Jes Sanddal Lindholt4, Hans Christian Beck1,2 1Odense University Hospital, Department of Clinical Biochemistry, Odense, Denmark 2Odense University Hospital, Centre for Clinical Proteomics, Odense, Denmark 3University of Southern Denmark, Epidemiology, Biostatistics and Biodemography - Department of Public Health, Odense, Denmark 4Odense University Hospital, Department of Cardiothoracic and Vascular Surgery, Odense, Denmark Background: Abdominal aortic aneurysms (AAA) are in most cases asymptomatic and pose a life-threatening risk due to unexpected rupture of the aortic wall. AAA are most often detected accidently when patients undergo CT scans for other purposes, and in many countries AAA screening is considered too costly. Therefore, a simple blood test discriminating individuals with an AAA from non-AAA individuals is urgently needed. Aim: The aim of the present study was to identify specific plasma proteins that discriminate individuals with an AAA from non-AAA individuals by applying mass spectrometry-based proteomics. Methods: Our study involved 7082 participants, including 545 patients with AAA, and 6537 control subjects from the population-based Danish Cardiovascular Screening (DANCAVAS) trial. Plasma samples were prepared and labelled with 18-plex TMTpro prior to LC-MS/MS analysis using data-dependent acquisition on an Orbitrap Eclipse mass spectrometer. Data processing was performed using a combined Sequest HT, MSPepSearch, CHIMERYS®, and MS Amanda search in Proteome Discoverer 3.0. For MSPepSearch our recently developed TMTpro-specific spectral library was applied. For statistical model development, participants were randomly divided into either a training set (80%) or a test set (20%). Gene Ontology enrichment analysis of significant differentially regulated proteins was made with ShinyGo (v. 0.80). Logistic regression combining proteomics data and clinical data was used for the identification of individuals with AAA. Performance of our model was measured using the area under the curve (AUC) for the detection of AAAs on the independent test set. Results: Mass spectrometry-based proteomics identified 46 proteins as significantly differentially regulated between AAA patients and control subjects (all FDR adj. p < 0.05). Up-regulated pathways included complement and coagulation cascades, cholesterol metabolism, and platelet activation, while vitamin digestion and absorption, cholesterol metabolism and fat digestion and absorption were among the down-regulated pathways. A standard model trained on clinical variables only achieved an AUC = 0.81. Inclusion of a panel of twenty-two proteins in our model significantly improved the ability to discriminate between control subjects and AAA patients with an AUC = 0.85 (DeLong, p = 0.0384). Conclusion: We identified a panel of plasma proteins that – in combination with clinical variables - improves the prediction of individuals with an AAA and has the potential as blood protein markers for the detection of AAA outside a hospital setting. 803

804

P-II-0650 Proteome analysis with combined DIA and Glyco-DIA approach of laser microdissected precursor lesions from pancreatobiliary cancer to improve early cancer diagnostics Stella Maris Pauls1 1Molecular Proteomics laboratory, Heinrich-Heine Universität, Duesseldorf, Germany Proteomic analysis of FFPE tissue samples is important for the detection of proteins that are involved in tumor progression, with the potential to find new disease-specific biomarker to improve diagnostics. For pancreatobiliary (PB) cancer a specific biomarker for diagnostics is not yet known, except the tumor markers CA 19-9 and CEA, which are used for follow-up purposes but cannot be used for early diagnosis of PB cancer due to their elevation also in benign conditions, such as pancreatitis. The early stage in tumor progression is associated with a transition of normal cells into so-called precursor lesions, which seem to develop along similar pathways and show numerous morphological and molecular similarities. The group of Prof. Esposito at the Institute of Pathology developed methods for morphomolecular characterization of PB precursors as well as molecular subtyping of different precursor stages applied on formalin-fixed, paraffin-embedded (FFPE) tissues. For the characterization of PB cancer, several proteome studies have been attempted to detect proteins involved in progression. However, they provided only limited insight into the tumorigenic mechanism of PB cancer, because the results were biased by signals from different cell types such as acinar, endocrine, stromal and inflammatory cells. In this study we optimized and applied sensitive methods combining microdissection and quantitative, label-free proteomics to gain a more comprehensive understanding in the development of PB cancer. By using a data independent acquisition (DIA) mode for LC-MS/MS analysis the goal is a combined proteomic and glycoproteomic (specific O-glycans) analysis to unravel differences in malignant and benignant tissue transformation of pancreatic precursor lesions. These findings can support clinical decisions and improves early cancer diagnostics as well as give insights to regulatory processes provided by glycosylation patterns. Pathway analysis reveals enrichment of transport processes from ER to Golgi, RNA splicing and glycolytic processes in malignant progression. We observed one protein candidate, arising from differential analysis of the proteomic dataset that is strongly associated with ER activity. Herein we confirm RNA expression data from literature and give contradictory results to the current biomarker validation via antibody staining. Within the Glyco-DIA approach we found overall enrichment of sialylated glycoproteins, as well as glycoproteins carrying the core-1 glycan in malignant progression. Differences in the glyco-modification type is observed for 14-3-3 theta, in which the peptide carrying the core-1 glycan (T antigen) is enriched in malignant progression. For this modification, literature evidence is not yet given and the results need further validation by external methods. Fig. 1 (optional) 804

805

P-II-0651 Clinical proteomics elucidated the complex molecular mechanisms in the progression of dry eye syndrome with different severity levels: a step forward to personalized diagnosis Natarajan Perumal1, Anna Lindner1, Adina Glasmacher1, Bettina Multani1, Alfred Francis Pakkam John Francis1, Caroline Manicam1 1University Medical Center of the Johannes Gutenberg-University Mainz, Department of Ophthalmology, Mainz, Germany Dry eye syndrome (DES) is a collective term used to define a pathology of the ocular surface with pleiotropic etiology that affects the tear film homeostasis and results in perturbations in the visual function. However, specific clinical biomarkers associated with the vicious cycle involving multiple interacting pathways that perpetuate and exacerbate the progression of DES are still largely lacking. Therefore, this study was undertaken to unravel the intricate proteome changes in individuals with increasing severity of different DES subgroups. Subjects underwent a battery of clinical examinations including the basic secretory test, tear break-up time and ocular surface disease index. Subsequently, the subjects were divided into evaporative (DRYlip) and aqueous- deficient (DRYaq), as well as the combination of both DES (DRYaqlip) with increasing severity (normal, moderate and severe) and healthy (CTRL) based on distinct clinical attributes. Next, the individual tear samples (N=322) were analyzed employing a rapid and robust mass spectrometry (MS)-based proteomics strategy, followed by bioinformatics analyses to elucidate the underlying molecular processes and protein functionalities. A total of 71, 218 and 306 proteins were found to be significantly (p < 0.05) differentially regulated in DRYlip, DRYaq and DRYaqlip vs. CTRL, respectively. The hallmark of this study was the identification of 30 lacrimal gland- specific proteins comprising mainly LTF, LYZ, LACRT and PROL1 secreted primarily via the exocytotic pathway, which were found to be progressively down-regulated in DRYaq and DRYaqlip vs. CTRL (e.g. LTF: moderate = Δlog2 -1.8; severe = Δlog2 -3.4). Inflammatory response was found to be increasingly activated in DRYlip (p=3.8×10-4), DRYaq (p=4.2×10-9) and DRYaqlip (p=2.1×10-11). Likewise, proteins involved in apoptosis (p=1.0×10- 15), antibacterial response (p=2.3×10-5) and synthesis of reactive oxygen species (p=1.3×10-12) were also found to be increasingly activated in all three DES subgroups. Notably, glycolysis (p=2.6×10-8), actin cytoskeleton signaling (p=1.0×10-3) and complement system (p=4.5×10-6) were found to be activated only in patients with moderate and severe DRYaqlip. In summary, our study took advantage of the MS-based proteomics strategy to analyze the largest number of tear samples in a clinical setting to date, considering different DES subgroups and varying levels of disease severity. Notably, this study has demonstrated that the protein markers involved in the exocytotic pathway, which play major roles in the production of tears, were significantly impaired. This resulted in the gradual activation of the vicious cycle involving the interplay between various pathways in the progression of DES. Importantly, the key clinical protein markers for DES were unraveled for the first time and could be potentially used for improved clinical diagnosis and targeted therapeutic interventions in ameliorating DES in a personalized manner. 805

806

P-II-0652 Development of a plasma proteomics methodology for biomarker discovery of immunocytokine therapy in Glioblastoma Lucrezia Principi1, Sebastian Oehler1, Teresa Hemmerle1, Domenico Ravazza1, Dario Neri2,3, Ettore Gilardoni1 1Philochem AG, Otelfingen, Switzerland 2Philogen S.P.A., Sovicille, Italy 3Swiss Federal Institute of Technology, Zurich, Switzerland Glioblastoma is the most common and deadliest primary brain malignancy in adults, with a median survival of only 16 months from diagnosis. Chemotherapy with Lomustine is currently the standard of care treatment for glioblastoma at progression with a median progression-free survival of only 1.5 months. Novel therapeutic approaches are urgently needed. L19TNF is a fully human tumor-targeting antibody-cytokine fusion protein, consisting of TNF fused to the L19 antibody, specific to the EDB domain of fibronectin. Phase I and II clinical trials of L19TNF for newly diagnosed glioblastoma or at recurrence are ongoing, investigating the benefit for the patients. Drug-specific response biomarkers might substantially shorten clinical development time and increase the success rate at the latest phase. Mass spectrometry (MS)-based proteomics is a potent tool for the in-depth analysis of the plasma proteome. To support response assessment and to improve the decision-making processes in clinical drug development, we are developing an LC-MS-based plasma proteomic approach for the discovery of pharmacodynamic biomarkers and immune modulation effects of L19TNF in glioblastoma patients. Several methodological parameters to identify the suitable conditions for our instruments and the clinical goals were investigated. Particularly, plasma depletion or enrichment during sample preparation is being evaluated to understand the benefits or the disadvantages compared to neat plasma analysis. Moreover, a dedicated Data Independent Acquisition method is being developed. Instrument parameters are being investigated to increase the sensitivity, precision, and total number of quantified proteins. The best combination of sample preparation and data analysis parameters will be used to perform exploratory studies in mice models to validate the platform developed and in clinical samples to monitor the drug response of Glioblastoma patients. 806

807

P-II-0653 Saliva peptidome as source of possible biomarkers of salivary glands tumor Michał Puchalski1, Marcel Thiel1, Stanisław Ołdziej1 1University of Gdańsk, Intercollegiate Faculty of Biotechnology of the University of Gdansk and the Medical University of Gdansk, Gdańsk, Poland Salivary glands tumors account for approximately 3% of all tumors in the head and neck region. This type of tumors is highly variable in clinical manifestation and histology. The World Health Organization (WHO) classifies 22 types of malignant and 11 types of benign tumors of the salivary glands. The multiplicity of types of such tumors poses enormous problems for diagnosis, which is largely based on imaging (ultrasound, magnetic resonance is based on the histopathological examination of the removed tumor tissue. Saliva as the product of salivary glands thus appears to be an ideal material for the possible search for biomarkers. imaging) and fine-needle aspiration biopsy; however, the final diagnosis The saliva peptidome of 24 patients diagnosed with salivary gland tumor was analyzed. For 11 patients, samples of tumor tissue as well as healthy tissue were also available; tissue samples were obtained during surgical removal of tumors. The control group consisted of 8 saliva samples from healthy subjects and data obtained from the PRIDE repository (PXD020211) on saliva samples from 23 patients diagnosed with oral squamous cell carcinoma (cancer of the head and neck region other than salivary gland tumors). More than a hundred unique peptides found only in the saliva of patients with salivary gland tumors were identified. However, none of these peptides was identified in more than two-three samples. Perhaps the lack of identification of universal markers is related to the significant heterogeneity of the tumors studied. In our research group, we developed a bioinformatics tool (AliceDB) to identify natural amino acid sequence variants. In each sample (saliva is not very rich in proteins), from a few to a dozen natural variants of various proteins were potentially identified. We are currently at the stage of analyzing the occurrence of natural variants, especially those variants that may be markers of the cancer lesions under study. For example, to our surprise in about 60% of the samples (samples from all analysed groups) we identified peptides containing the R49H variant for products of the PRB4 gene encoding Basic salivary proline-rich protein 4. The result is all the more surprising because gnomic data suggest that this type of variant is quite rare (less than 0.1% of population - GenBank-NCBI). The identified peptides containing the above-mentioned variant cannot be part of any other protein (from any species) whose sequence is deposited in the UNIPROT database. Therefore, we speculate that the identified peptide is unlikely to be derived from oral bacterial flora or food debris, and is a rather non-canonical product of the PRB4 gene. We hope that, further detailed analysis of the protein sequence variants will enable the identification of salivary gland tumor markers, which is the goal of this project. 807

808

P-II-0654 Time-resolved plasma proteomics predicts survival outcome of COVID-19 patients during hospitalization Mauricio Quiñones-Vega1,2,3, Patricia Sosa-Costa1,2,3, Giovanna C.Giudicelli4, Mariléa F. Feira4, Nathan A. Cadore4, Fernanda S. L. Vianna4,5, Fabio Cesar Sousa Nogueira1,2,3, Gilberto B. Domont1,2 1Federal University of Rio de Janeiro, Institute of Chemistry, Rio de Janeiro, Brazil 2Precision Medicine Research Center, Proteomics Department, Rio de Janeiro, Brazil 3Federal University of Rio de Janeiro, LADETEC (Insitute of Chemistry), Laboratory of Proteomics (LabProt), Rio de Janeiro, Brazil 4Centro de Pesquisa Experimental, Laboratório de Medicina Genômica, Rio de Janeiro, Brazil 5Universidade Federal do Rio Grande do Sul, Genetics Department, Laboratory of Immunogenetics, Porto Alegre, Brazil Introduction: The need to understand the progression of COVID-19 and SARS-CoV-2 infection is essential to identifying potential biomarkers and therapeutic targets. The goals of this study are to comprehend the trajectory of the disease by characterizing changes in the plasma proteome over time and to propose putative biomarkers proteins that can differentiate between non-survivor and survivor phenotypes. Methods: During hospital admission, plasma samples were collected weekly from 44 patients over a six-week period from Hospital de Clinicas de Porto Alegre Biobank (total = 198). Plasma was processed using S-TrapTM digestion protocol and acquired in 90 min gradient using a Vanquish™ Neo UHPLC System coupled to an Orbitrap Exploris™ 480 Mass Spectrometer. Proteins were identified using DIA-NN software. Results: Proteomic analysis identified 1600 proteins, of which 851 were selected for quantitative analysis. It was observed that the acute phase response and the inflammatory reaction decreased over time returning to normal levels by the sixth week. Other processes such as antigen binding, endocytosis, and vesicle-mediated transport increased during hospital admission related to the first week. In the global correlation map, a cluster of proteins associated with cytokine stimulation was observed. Furthermore, the comparison between patients who survived and those who died revealed differentially abundant proteins even in the first weeks of hospitalization. In non-survivor patients were found more abundant proteins related to complement activation and the innate immune response, while the humoral response and antibody-related annotations decreased in these patients, especially in the first three weeks. From the fourth week onwards, despite of decreasing in time, acute inflammatory response was increased in patients who died related to survivors, accompanied by a reversal of previously diminished processes, such as leukocyte chemotaxis and complement activation. To identify potential biomarkers that could predict survivor or non-survivor phenotypes a linear SVM model was trained with proteins HPR, PF4, LPA, IGFALS, COLEC11 resulting in a ROC curve with AUC 0.916. In addition, the model correctly classifies 85% of the patients. Conclusions: This study provides important insights into the pathophysiology and progression of COVID-19 and highlights proteomics as a powerful tool for understanding the host response during viral infection and correlating and validating the clinical information. 808

809

P-II-0655 Profiling of the pathophysiological mediators of a rare neurodegenerative disease, Friedreich's Ataxia (FRDA) Jannatul Rafeya1, Kathrin Reetz2, Katharina Schütt3, Joachim Jankowski1,4, Vera Jankowski1 1RWTH Uniklinikum Aachen, The Institute for Molecular Cardiovascular Research, Aachen, Germany 2RWTH Uniklinikum Aachen, The Department of Neurology, Aachen, Germany 3RWTH Uniklinikum Aachen, Clinic for Cardiology, Angiology and Internal Intensive Care Medicine (Med. Clinic I), Aachen, Germany 4Maastricht University, School for Cardiovascular Diseases, Maastricht, Netherlands Introduction: Friedreich's ataxia is a neuro-degenerative, autosomal recessive disease, which is clinically characterized by loss of coordination, ataxia, along with cardiomyopathy, scoliosis, and an increased risk of diabetes [1]. The disease is manifested due to a GAA expansion mutation on the intron of FXN gene, resulting in a reduction of functional frataxin in cells [2]. The project aims to identify the unknown pathophysiological mediators involved in the genesis and progression of Friedreich ataxia along with generating a proteomic profile of the patients. Method: The samples were split into four major cohorts of healthy, patient, fibrotic patient, and non-fibrotic patients. Proteins from the plasma and urine samples were separated with gel-electrophoresis. Mass spectrometry (MS1 & MS2) analyses were performed to determine the proteomic profile of the samples with the Mascot Search Engine. A targeted MS analysis focused was conducted by focusing on frataxin and Neurofilament light chain (NFL) in sample cohorts. Result: The patient cohort's frataxin level was reduced compared to the healthy cohort (Figure 1, A). However, there was no significant change in frataxin levels between the fibrotic and non-fibrotic patients (Figure 1, B). Finally, frataxin was detected in urine (Figure 1, C) at a reduced level compared to plasma. NFL level had been increased in almost all FRDA patients, except for five patients (Figure 1, D). There was also no significant change in the NFL level between the fibrotic and non-fibrotic patients (Figure 1, E). NFL was also detected in urine (Figure 1, F) and the protein level was lower compared to the plasma NFL level. Discussion: Frataxin reduction in the patient cohort correlates with the notion that FRDA is a frataxin deficiency disease [3]. An increase in NFL in most of the patients indicated increased axonal damage associated with the disease. The relatively lower NFL in five samples can be because the NFL level stabilizes toward the later stage of a neurodegenerative disease even though it increases during the early phase of the disease [4]. None of the frataxin and NFL had a significant difference between the fibrotic and non-fibrotic cohorts indicating irrelevance of these proteins to fibrosis progression. The detection of frataxin and NFL in the urine, albeit at a reduced level, was one of the most interesting findings, as no prior literature article describes this phenomenon. Focusing on this might lead to a new revelation associated with the pathophysiology of the FRDA disease. References 1. Lynch et al, J. Multidiscip. Healthc., pp.1645-1658, 2021. 2. Gavriilaki et al, The Cerebellum, 1-20, 2023. 3. Doni et al, Cell Death Dis, 14(12), p.805, 2023. 4. Zhang et al, Front Neurol., 13, p.833507, 2022. 809

811

P-II-0656 Requirements and limitations for the preparation of targeted mass spectrometry assays enabling therapeutic drug monitoring Yvonne Reinders1, Andreas Henstchel1, Leon Welter1, Cornelia Schumbrutzki1, Peter Findeisen2, Albert Sickmann1 1ISAS e.V., Dortmund, Germany 2MVZ Labor Dr. Limbach & Kollegen, Heidelberg, Germany The integration of mass spectrometric methods in the field of therapeutic drug monitoring (TDM) represents a promising approach. Through the application of mass spectrometry, selective and accurate measurements of monoclonal antibodies (mAB´s) in the blood are obtainable, enhancing the reliability and accuracy of TDM analysis and minimizing the effects of autoantibodies on test results, thereby preventing patients" over-and underdosing during therapy. Additionally, mass spectrometry offers the capability of multiplexing, allowing simultaneous detection and quantification of multiple mAB"s in a patient´s sample, especially pivotal for patients under therapy using a combination of antibodies. For those reasons, it was our aim to establish a targeted mass- spectrometric assay facilitating reproducible absolute quantification of peptides from the hypervariable and interacting regions of selected antibodies. Therefore, tryptic peptides and their corresponding stable isotopologue were analysed using UHPLC coupled to a TSQ Altis Triple-Quad mass spectrometer. In this study, exemplary for the antibody infliximab, we demonstrate the substantial influence of peptide selection, choice of hydrolase for digestion, and digestion time on digestion efficiency and absolute peptide yield. Additionally, we showed the robustness and reproducibility of the generated calibration curves including key figures such as LOD and LLOQ. Finally, we have developed a semi-automated platform enabling the rapid (2 days of sample preparation and 30 min of measurement time per sample) and robust absolute quantification of Infliximab antibody concentration in patients" blood samples (Hentschel A et al.; Clin Proteomics. 2024 Feb 29;21(1):16. doi: 10.1186/s12014-024-09464-x.). The use of mass spectrometry also facilitates the straightforward expansion of the method to include additional antibody peptides. 811

812

P-II-0657 Proteomics signature predictive of early biochemical progression and distant progression after stereotactic body radiotherapy in oligometastatic prostate cancer Ombretta Repetto1, Filippo Sperti1, Giulia Brisotto1, Matteo Turetta1, Fabio Del Ben1, Alessandra Donofrio2, Lucia Fratino2, Roberto Bortolus2, Agostino Steffan1, Fabio Matrone2 1Centro di Riferimento Oncologico di Aviano (CRO), IRCCS, Immunopathology and Tumour Biomarkers Unit, Department of Research and Advanced Tumour Diagnostics, Aviano, Italy 2Centro di Riferimento Oncologico di Aviano (CRO), IRCCS, National Cancer Institute, Department of Radiation Oncology, Aviano, Italy Background: In oligometastatic prostate cancer (oligo-mPC), metastasis-directed therapy (such as stereotactic body radiotherapy, SBRT) can improve local control and progression free survival. However, a part of oligo-mPC patients present an early biochemical progression (BP) which may develop to a polymetastatic scenario. At present, early clinical predictors of BP after SBRT in oligo-mPC are mostly unknown, and plasma proteomics investigations of oligo-mPC undergoing SBRT are still to be performed. Aims and study design: The main objectives were to identify a plasma proteomics signature associated to early- BP or distant progression (OLIGO or POLI) in oligo-mPC before SBRT at T0. The identification and relative quantitation of differential plasma proteins in early-BP versus no-BP, dist-prog-OLIGO or dist-prog-POLI versus no-dist-prog were investigated by mass spectrometry-based untargeted Data-dependent (DDA) and Data- independent (DIA) proteomics. Patients and plasma: Oligo-mPC patients (patients with ≤3 bone or lymph node metastases) undergoing SBRT were enrolled at CRO Institute. Oligo-mPC patients were classified on the basis of biochemical progression free survival (BPFS) after SBRT into "early-BP" (n=7, BPFS Results: LC-MS/MS identified a total of around 300 and 500 distinct proteins by DDA and DIA approaches, respectively. Plasma protein profiles of early-BP and no-BP differed for some differentially abundant proteins (FDR < 0.01, Score Sequest ≥ 1, |fold change|≥ 2, Padjusted < 0.01 and abundance grouped CV% <30). In particular, early-BP showed a lower abundance of extracellular superoxide dismutase [Cu-Zn] (SOD3), carbonic anhydrase (CA1) and nucleophosmin (NPM1) proteins and an higher abundance of serine/threonine-protein kinase Chk1 (CHEK1). Moreover, dist-prog-OLIGO versus no-dist-prog showed an increase in HLA class I histocompatibility antigen, C alpha chain (HLA-C) and a decrease in complement factor H-related protein 5 (CFHR5), while both dist-prog-OLIGO and POLI versus no-dist-prog were characterized by an increase in cadherin-1 (CDH1) and integrin alpha-6 (ITGA6). Conclusions and perspectives: This study identified some candidate protein markers associated to early-BP or distant progression in oligo-mPC undergoing SBRT. Their differential abundance will be validated by Western Blotting in the same cohort of patients and by further MS analysis in a verification cohort consisting of newly enrolled patients. By means of the expected results, CRO Aviano may benefit for patient stratification and early identification of oligo-mPC patients early ongoing to BP, who thus represent potential candidate for a more aggressive therapeutic approach. 812

813

P-II-0658 Comparative proteomic analysis of human ventricular tissue of over 100 patients undergoing heart transplantation Ulrike Resch1,2,3, Luisa Schmidt2, Philipp Antczak1,4, Stefan Müller5, Jan-Wilm Lackmann5, Artur Lichtenberg3, Hug Aubin3, Marcus Krüger2,5, Elivra Weber3 1Medical University of Vienna, Vascular Biology and Thrombosis Research, Vienna, Austria 2University of Cologne, Institute for Genetics, CECAD, Cologne, Germany 3Heinrich-Heine University Düsseldorf, Medical Faculty and University Hospital Düsseldorf, Department of Cardiac Surgery, 3D Cardiovascular Regenerative Medicine & Tissue Engineering, Duesseldorf, Germany 4University of Cologne, Center for Molecular Medicine Cologne CMMC, Cologne, Germany 5University of Cologne, CECAD/ZMMK Proteomics Facility, Cologne, Germany Background: Heart failure remains a major cause of morbidity and mortality worldwide and its prevalence steadily increases. Despite improved pharmacologic and surgical treatment options knowledge of molecular mechanisms underlying the progressive pathologic changes in the heart still needs to be improved. Ischemic and dilated cardiomyopathy (ICM/DCM) are two major types of cardiomyopathies which often exhibit similar clinical symptoms despite essentially different etiologies. To characterize disease-differentiating ventricular proteome signatures and molecular patterns, we performed in-depth, label-free quantitative proteome profiling of left and right ventricles (LV/RV) biopsies from 68 ICM patients (51 male, 17 female, 60.2±6.0 years, BMI 26.2±4.3) and 68 DCM patients (45 male, 23 female, 52.7±11.7 years, BMI 26.0±5.1) admitted to heart transplantation. Methods: A spectral library and an optimized 30SPD dia-PASEF acquisition on a Evosep-timsTOF Pro2 system was utilized. Results: We successfully established a feasible and robust proteomic workflow in sample and data analysis time and identified 3000 to 5000 across all samples. LMMA-bioinformatic analysis revealed a significantly different proteome in LV and RV tissue, which is furthermore differentiating ICM form DCM. Our protein-network analysis suggests a substantial proteotoxic, oxidative-reductive stress in ICM´s ventricles which translates into distinguished defects in fatty acid mobilization and mitochondrial function, specifically complex I, resulting into a vicious, energy-depleting cycle. This perpetuates into malfunctioning cytoskeletal and extracellular matrix assembly, myocardial scarring and tissue remodeling. Furthermore, our large dataset allowed to uncover ventricular proteins specific to physiological covariates such as sex, BMI and comorbidities like diabetes. Conclusions: Comparative ventricular proteome profiling of ante-mortem human heart tissue in such a large patient cohort has not been described so far. More than 300 significant proteins discriminating ICM and DCM provide new molecular links for basic research and precision medicine-based intervention strategies. 813

814

P-II-0659 A high-throughput, in-depth investigation into the immunological effects of Lipopolysaccharide challenged mice in neat plasma on Orbitrap astral mass spectrometer Jana Richter1, Nicolas Hartel2, Amirmansoor Hakimi2, Stephanie Samra2 1Thermo Fisher Scientific, Bremen, Germany 2Thermo Fisher Scientific, San Jose, CA, United States Lipopolysaccharide (LPS) inhalation stimulates an immune response in mice and serves as a preclinical model for investigating potential therapeutics for inflammatory related disorders such as acute bronchitis. Here mice were treated and challenged with LPS and compared to healthy, unchallenged mice to gain insights into immunological response and identify key down regulated and upregulated proteins of importance. Traditionally, neat plasma requires long runtimes of over 1hr to achieve the depth of coverage needed for meaningful insights by mass spectrometry. Here the Thermo Scientific™ Orbitrap™ Astral™ mass spectrometer was optimized and assessed for high-throughput analysis while still maintaining depth of coverage. Neat plasma from five biological replicates of LPS challenged and unchallenged mice were prepared by automated sample preparation on the Thermo Scientific™ AccelerOme™ with LysC/trypsin digestion that yielded peptides ready for separation by Thermo Scientific™ Vanquish™Neo™ and detection by Orbitrap Astral™ mass spectrometer. Separation was performed by reverse phase gradient with Ionopticks columns for various sample throughputs of 20SPD, 60SPD, and 180SPD to compare depth of coverage and quantitative accuracy. Three technical replicates were acquired to assess instrument reproducibility. A gas phase fractionation library was created for the 60SPD method and used to search the data for maximum protein identification. Data collected was analyzed by Thermo Scientific™ Proteome Discoverer™ 3.1, DIANN, and Spectronaut®18 software for comparison. Over 3600 protein groups and 25400 peptides were detected with the 20SPD method searched with DIANN with no library applied. This was over 40% and 86% improvement over the 60SPD and 180SPD method respectively. Results showed over 93% quantifiable IDs and median coefficient of variance (CV) below 10% for technical replicates for different run times. The gas phase fractionation library for 60SPD was used to process results and yielded a 44% improvement from 1680 protein IDs to 2427 protein IDs. The five biological replicates for the LPS challenged mice versus the unchallenged mice yielded 2169 and 1914 protein IDs respectively and were reproducible and robust across all five biological replicates. The LPS challenged results for the 60SPD method were 76% quantifiable in all five replicates and 91% quantifiable in three out of five replicates. Proteins were ranked on a ranked protein list and spanned nearly six order of magnitude and were clustered on a PCA plot to show the differences observed between the two treated and untreated groups. Next, differential expression was performed with a p-value <0.001 and +/- 2 Log2FC as cut-off, revealing 35 upregulated and two downregulated proteins in LPS challenged plasma versus unchallenged plasma. Orbitrap Astral mass spectrometer optimized for high-throughput and depth of coverage yielded >3600 protein groups in LPS challenged neat plasma. 814

815

P-II-0660 Quality control for clinical proteomics Svitlana Rozanova1,2, Julian Uszkoreit1,2, Karin Schork1,2, Bettina Serschnitzki1,2, Martin Eisenacher1,2, Lars Toenges3, Katalin Barkovits-Boeddinghaus1,2, Katrin Marcus-Alic1,2 1Medizinisches Proteom-Center, Ruhr University Bochum, Bochum, Germany 2Ruhr University Bochum, Medical Proteome Analysis, ProDi, Bochum, Germany 3Ruhr University Bochum, St Josef-Hospital, Bochum, Germany Proteomic studies using mass spectrometry (MS)-based quantification are a main approach to the discovery of new biomarkers. However, a number of analytical conditions in front and during MS data acquisition can affect the accuracy of the obtained outcome. Therefore, comprehensive quality assessment of the acquired data plays a central role in quantitative proteomics, though, due to the immense complexity of MS data, it is often neglected. Here, we address practically the quality assessment of quantitative MS data, describing key steps for the evaluation, including the levels of raw data, identification and quantification. With this, four independent datasets from cerebrospinal fluid, an important biofluid for neurodegenerative disease biomarker studies, were assessed, demonstrating that sample processing-based differences are already reflected at all three levels but with varying impacts on the quality of the quantitative data. Specifically, we provide guidance to critically interpret the quality of MS data for quantitative proteomics. Moreover, we provide the free and open source quality control tool MaCProQC, enabling systematic, rapid and uncomplicated data comparison of raw data, identification and feature detection levels through defined quality metrics and a step-by-step quality control workflow. 815

816

P-II-0661 An in-depth and high throughput plasma proteomics workflow powered by Orbitrap Exploris 480 mass spectrometer using multi nanoparticle-based workflow Kevin Yang1, Nicholas Mucci1, Evangelina Bahu1, Andrea Cerda1, Lee Cantrell1, Julian Saba2, Khatereh Motamedchaboki3, Stephanie Samra3, Amirmansoor Hakimi3 1Seer, Inc., Redwood City, CA, United States 2Thermo Fisher Scientific, Winnipeg, Canada 3Thermo Fisher Scientific, San Jose, CA, United States Plasma proteomics using mass spectrometry (MS) remains a promising method to discover disease biomarkers. However, for large-scale plasma proteomics studies, a robust liquid chromatography-mass spectrometry (LC-MS) setup that does not compromise protein identification, sequence coverage, dynamic range, or analytical precision is required. Here we present a high-throughput and maximum identification (Max-ID) workflow on a Thermo ScientificTMOrbitrap ExplorisTM 480 mass spectrometer for in-depth analysis of plasma. The plasma samples were processed with Seer"s Proteograph™ XT Assay utilizing multi-nanoparticles (NPs) for unbiased and deep proteomics enrichment at scale. [SJ1] Samples were processed using 240 μL aliquots of each plasma sample, mixed with each of the two NP wells included in the Proteograph XT Assay Kit. The enriched protein coronas were reduced, alkylated, and digested with Trypsin/Lys-C to generate tryptic peptides for LC-MS analysis. All steps were performed in a one-pot reaction directly on the NPs. The in-solution digestion mixture was desalted, and all detergents were removed using a mixed-media filter plate and positive pressure (MPE) system. Clean peptides were eluted in a high- organic buffer into a deep-well collection plate. 500 ng of tryptic peptides were analyzed on the Orbitrap Exploris 480 MS with both nanoflow and capillary flow LC methods. For the capillary flow high-throughput method, two individual injections of 24 minutes each were used for a total instrument time of 48 minutes per sample. For the nanoflow Max ID method, the two peptide wells were pooled, and LC-MS analysis was done in a single injection using a 102-minutes total gradient. The optimized high-throughput DIA method achieved the identification of more than 3,000 protein groups and over 20,000 peptides from the Proteograph Analysis Suite with the match between run (MBR) feature. Of these, 87.2% presented a coefficient of variation (CV) below 20%. For the Max ID method, we successfully identified more than 3,800 protein groups and over 27,000 peptides, with 89.2% of protein groups showing a CV under 20%, highlighting the excellent quantitation performance. In addition, our data revealed that 95% and 99% of protein groups were consistently identified across all replicates from the high-throughput and Max ID methods, respectively. The results underscored the completeness of the data resulting from the LFQ-DIA workflow. Together, we demonstrated a comprehensive plasma proteomics workflow on an Orbitrap Exploris 480 mass spectrometer with Seer"s Proteograph XT suite to offer high-throughput analysis with in-depth proteome coverage and excellent quantitation performance. 816

817

P-II-0662 Deep plasma proteomics profiling of healthy humans at two different stages of adulthood Federica Anastasi1,2,3, Eva Borràs4,5, Olga Pastor4,5, Marc Suárez-Calvet1,2,3, Eduard Sabidó4,5 1BarcelonaBeta Brain Research Center, Barcelona, Spain 2Pasqual Maragall Foundation, Barceona, Spain 3Hospital del Mar Research Institute, Barceona, Spain 4Centre for Genomic Regulation, Barcelona, Spain 5Universtitat Pompeu Fabra, Barcelona, Spain Mass spectrometry-based proteomics is a valuable strategy for plasma proteome profiling that has successfully been used in many personalized medicine and health monitoring studies. Despite its potential, analyzing plasma samples by mass spectrometry poses some analytical challenges, particularly due to the broad dynamic range of plasma, which limits access to its complete proteome. To overcome these limitations and enhance in-depth protein identification, several commercial solutions have been developed to increase the number of identified proteins and the portion of the accessible plasma proteome. However, beyond merely increasing the number of identified peptides and proteins, it is also crucial to demonstrate that these new approaches maintain the fidelity of protein relative quantification, thereby reproducing and extending previous biological insights obtained with neat plasma. In this study, we compared the relative quantitation and biological information retrieved from neat and enriched human plasma in a cohort of healthy individuals at two different stages of adulthood. Plasma samples from 60 healthy individuals (30 aged 18-25, 30 aged 70-84) were analyzed by mass spectrometry using a data- independent acquisition (DIA) workflow. The study utilized neat and enriched plasma from the same individuals, processed with either the iST-BCT or ENRICH-iST PreOmics kits, and analyzed using a 60-minute gradient on an Orbitrap Eclipse instrument. Acquired data were analyzed using a library-free approach with DIA-NN 1.8.1 software. Enrichment procedures resulted in more than a threefold increase in protein identification compared to neat plasma, with most proteins identified in neat plasma also detected in enriched samples. Moreover, our results show that the enrichment process does not increase quantitative variability and can reproduce most protein fold-changes previously observed. Finally, this approach provides deeper insights into age-related differences in the plasma proteome, adding valuable information from enriched samples. 817

818

P-II-0663 Quantification of intra-amniotic inflammation in late preterm prelabor rupture of membranes associated with the response of amniotic fluid proteome Marie Vajrychova1, Michaela Sadibolova1, Jaroslav Stranik2, Rudolf Kupcik1, Rudolf Kukla3, Radka Bolehovska3, Marian Kacerovsky2 1University Hospital Hradec Kralove, Biomedical Research Center, Hradec Kralove, Czech Republic 2University Hospital Hradec Kralove, Department of Obstetrics and Gynecology, Hradec Kralove, Czech Republic 3University Hospital Hradec Kralove, Institute of Clinical Microbiology, Hradec Kralove, Czech Republic Introduction: Preterm prelabor rupture of membranes is associated with the risk of intra-amniotic inflammation (IAI) even beyond 34 weeks of gestation (late PPROM). Currently, interleukin-6 (IL-6) concentration of 2500 pg/mL in amniotic fluid is frequently discussed as a marker of the presence of IAI in PPROM. However, there is a lack of information about the level of IL-6 associated with the global change of amniotic fluid proteome as a response to IAI. The correlation of IL-6 with the global change of amniotic fluid proteome might better define the cut-off value for IAI and help reveal women with the highest risk of excessive inflammatory response in late PPROM. Methods: In total, 60 amniotic fluid samples were collected and the concentration of IL-6 was determined in each of them using Cobas ECL analyzer. Protein quantification was based on Tandem Mass Tag (TMT) isobaric peptide labeling followed by 2D nanoLC-MS/MS. To verify proteomic data, target quantification (ELISA) of selected proteins (S100A12, MMP8, MPO, Lipocalin-2) was performed. Results: Using 2D TMT nanoLC-MS/MS, 925 proteins were efficiently quantified in all samples and multiple comparison analysis (ANOVA, B-H FDR 5%) revealed 254 proteins mostly upregulated towards IL-6 concentration above 10 000 pg/mL, which is far above the concentration of IL-6 2500 – 3000 pg/mL being previously presumed as the cut-off value for IAI. As examples, proteins associated with oxidative burst and redox imbalance (MPO, SOD2, TXN, PRDX1, PRDX5, GAPDH), neutrophil antimicrobial activity (CAMP, DEFA3, ENO1, histones, HSP, HMGB2, MMP9, MMP8, LCN2), and neutrophil metabolic activity (GPI, LDHA, LDHB, MDH1) were found as upregulated in the group of samples with IL-6 concentration higher than 10 000 pg/mL. ROC curves designed based on ELISA protein quantification confirmed that the amniotic fluid proteome shift was associated with IL-6 concentration higher than 5000 pg/mL. Novel aspect: Correlation of global amniotic fluid proteome and IL-6 concentration was used to quantify IAI in late PPROM. 818

819

P-II-0664 Biomarker discovery through in-depth characterisation of the serum protein composition of infant victims of silent Abusive Head Trauma (AHT) Tatjana Sajic1, Kim Wiskott2, Federica Gilardi1, Alexandre Hainard3, Patrick M. Kochanek4, Aurelien Thomas1, Rachel P. Berger4, Tony Fracasso2 1Lausanne University Hospital, University Center of Legal Medicine, Lausanne, Switzerland 2Geneva University Hospital, University Center of Legal Medicine, Lausanne, Switzerland 3University of Geneva, Proteomics Core Facility, Faculty of Medicine, Geneva, Switzerland 4University of Pittsburgh, University of Pittsburgh Medical Center Children’s Hospital of Pittsburgh, Pittsburgh, PA, United States Background: AHT including shaken baby syndrome (SBS), is the most common form of traumatic brain injury (TBI) in infants1,2. Nearly 40 infants per 100,000 live births are affected each year, and one third of cases are misdiagnosed at the first medical visit1,2. In fact, in cases of mild to moderate AHT (Glasgow Coma Scale (GCS): 9- 15), which we have termed silent AHT, clinical symptoms may overlap non-specifically with those of infants with acute malaise (such as non-febrile convulsions or Brief Resolved Unexplained Event (BRUE)). Currently, there is no screening test to support standard clinical practice in identifying children with brain injury who are at risk of silent AHT. Our study in the infants with severe AHT (GCS: 3-4) compared to atraumatic controls highlighted the serum increase of proteins physiologically enriched in the brainstem, eye and spinal cord2. Based on our most recent analysis of the large pediatric AHT blood cohort (UPMC Children's Hospital of Pittsburgh, USA)1, we sought to test the hypothesis that serum protein expression profiles provide the basic knowledge for computational machine learning to obtain signatures that discriminate silent AHT. Methods: We used data-independent acquisition mass spectrometry (DIA-MS) proteomics and analysed the cohort of 246 pediatric sera, including the cases of silent and severe AHT, as well as atraumatic control and accidental pediatric TBI cases, approved by the University of Pittsburgh IRB. We analysed 50 pediatric serum samples (i.e. mainly AHT cases) and two samples of adult cerebrospinal fluid by data-dependent acquisition (DDA-MS), generating a high-quality project-specific spectral library. The library contains information for 1725 Protein Groups (Qvalue ≤ 0.01) of which 232 brain-associated proteins according to the Human Protein Atlas. We performed targeted data analysis of raw DIA-MS files using commercial software Spectronaut (Biognosys) Results: We quantify 1539 Protein Groups in the unique pediatric AHT dataset, comprising nearly 300 MS injections including technical and biological replicates, and observe excellent technical reproducibility (replicates Spearman's rho>0.97; p < 2.2e-16). To investigate the biological data quality, we divided the samples into i) low GCS cases (3-8), moderate GCS cases (9-15) and other cases (non-communicating GCS or atraumatic). We performed a supervised PLS-DA analysis and the first two variants distinguish atraumatic samples (Fig.1, red circles) from low and moderate GCS cases (green/black circles). Among the top selected markers contributing to the serum variability based on the GCS, we detected proteins elevated in the anatomical regions of the brain(e.g. Alpha-synuclein, Fig1.B). Conclusions: This project aims to improve the diagnosis of silent AHT in infants through the proteomic identification of a panel of less-invasive serum markers. 1Berger, R. P. et al. JAMA Pediatr 171, e170429 (2017). 2Wiskott, K. et al. Proteomics 23, e2200078 (2023). 819

821

P-II-0665 Search for novel diagnostic markers of pancreatic ductal adenocarcinoma using secretome and glycoproteom analysis Yoshino Sakuma1,2, Shigetsugu Takano3, Yoshio Kodera4, Masayuki Ohtsuka3, Kazuyuki Sogawa1,2 1Azabu University, Biochemistry, Sagamihara, Japan 2Graduate School of environmental health, Azabu University, Molecular Diagnosis, Sagamihara, Japan 3Graduate School of Medicine, Chiba University, General Surgery, Chiba, Japan 4Kitasato University School of Science, Center for Disease Proteomics, Kanagawa, Japan Pancreatic ductal adenocarcinoma (PDAC) accounts for 85–90% of all pancreatic tumours. The median survival of all PDAC patients is less than 6 months, and the recent 5-year-survival rate is approximately 8%. One of the most crucial reasons for the poor prognosis is the lack of early diagnostic markers for PDAC. To overcome this and improve the outcomes of patients with PDAC, there is an urgent need to identify highly sensitive and specific markers for early detection. The widely used serum-circulating markers for PDAC, carbohydrate tumour– associated antigen 19-9 (CA19-9) and carcinoembryonic antigen (CEA), are not sufficiently accurate to be used as early diagnostic markers. In the present study, we employed the TMT method to generate comparative protein profiles of PDAC cell lines. Furthermore, we compared the serum levels of candidate proteins to evaluate the ability to discriminate between PDAC and healthy controls. Using tandem mass tag labelling and LC-MS/MS, we performed comparative analyses of secreted proteins in culturing media obtained from human PDAC cell lines (MIA PaCa-2, PANC-1, and BxPC-3) to identify serum biomarkers for PDAC. In validation studies, we evaluated the discriminatory power of candidate proteins. Of 27 proteins, 13 proteins were identified with the increase of protein levels among these all three cell lines, and receptor expression-enhancing protein 2 (REEP2) was selected for further analysis. The sera levels of REEP2 were significantly higher in the preoperative PDAC patients than those in the postoperative ones (p<0.001). In future study, we will verify this result with more number of samples, and pursue the usefulness of REEP2 as a novel diagnostic marker in PDAC. 821

822

P-II-0666 Vasodilator-stimulated phosphoprotein acts as a potential plasma-derived exosomal protein marker for early detection of colorectal cancer Juthaporn Sangwallek1, Jisnuson Svasti2, Chantragan Srisomsap2, Churat Weeraphan2, Daranee Chokchaichamnankit2 1Chulabhorn Royal Academy, Princess Srisavangavadhana College of Medicine, Bangkok, Thailand 2Chulabhorn Research Institute, Department of Biochemistry, Bangkok, Thailand Colorectal cancer (CRC) has the third most common cancer incidence and the second most frequent cause of cancer death worldwide, with greater mortality rate when detected at advanced stages. The earlier the cancer is detected, the better the chance of survival. Colonoscopy is currently the gold standard for CRC diagnosis but is invasive and costly. Therefore, there is a pressing need for a highly sensitive, specific, and minimally invasive biomarker for the detection of CRC at asymptotic stages. This study aims to identify blood-borne exosomal protein markers for the early detection of high-risk adenoma and CRC. Proteomics data (LC-MS/MS) were collected on plasma-derived exosomes from Normal (N), Low-risk adenoma (LRA), High-risk adenoma (HRA), and Colorectal cancer (CRC) groups. We found 16 exosomal proteins significantly upregulated in the HRA or CRC group compared to the N or LRA group. Validation of P-selectin (LYAM3, Log2FC -2.50, p-value 0.039), vasodilator-stimulated phosphoprotein (VASP, Log2FC -1.75, p-value 0.001), and Multimerin-1 (MMRN1, Log2FC -0.93, p-value 0.006) as potential biomarkers for high-risk adenomas and CRC was performed using Western blot analysis in plasma-derived exosomes extracted from individual samples. VASP was significantly more abundant in high-risk adenoma or CRC samples than in low-risk adenoma or healthy controls, indicating that VASP could distinguish the high-risk adenoma or CRC groups from the low-risk adenoma or healthy controls. We therefore conclude that VASP can potentially act as a blood-based exosomal protein marker for the early detection of high- risk adenomas and CRC. Fig. 1 (optional) 822

823

P-II-0667 Sample storage as pre-analytical factor on serum proteome quality Thorben Sauer1, Julia Horn1, Martina Oberländer2, Regina Maushagen2, Rüdiger Braun3, Kim Honselmann3, Jens Habermann1, Tobias Keck3, Silke Szymczak4, Timo Gemoll1 1University of Lübeck, University Hospital Schleswig Holstein, Campus Lübeck, Section of Translational Surgical Oncology and Biobanking, Department of Surgery, Luebeck, Germany 2University of Lübeck, Interdisciplinary Center for Biobanking-Lübeck (ICB-L), Luebeck, Germany 3University Hospital Schleswig-Holstein, Campus Lübeck, Department of Surgery, Luebeck, Germany 4University of Lübeck, Institute of Medical Biometry and Statistics, Luebeck, Germany The validated use of biomarkers for clinical diagnostics and therapy necessitates the availability of a substantial number of high-quality samples including the complete set of clinical data. To reduce the variability of sample quality, it is necessary to establish standards for sample collection, storage, and quality control. This study evaluated the impact of a single freeze-and-thaw (FT) cycle on the quality of liquid nitrogen-stored clinical serum aliquots by data-independent acquisition mass spectrometry (DIA-MS) on two independent sample cohorts. The pilot study included repeated measurements of 25 patient serum samples (total number of samples = 99), whereas the independent validation study included single-shot measurements of 109 serum samples. A longitudinal observational study monitoring the stability of liquid nitrogen-stored serum samples in monthly intervals over 12 months was conducted with pooled serum samples. Following the library-free data analysis and implementation of strict data preprocessing, the relative abundance of 213 (pilot study) and 248 (validation study) human serum proteins were analyzed by comparing paired fresh (never frozen) and FT samples. Of 39 target proteins observed in the pilot study, 21 proteins (53.8%) were successfully validated in the validation cohort. These targets exhibited a significant change (q-value <0.05) in measurable protein abundance after just one FT cycle, indicating FT sensitivity. The longitudinal study ruled out a considerable storage time dependence, as a profound variance (coefficient of variation >20%) in protein quantity was only observed in only 1.4% of the quantified proteins. In conclusion, distinct protein abundance patterns can be discerned between fresh serum samples and samples that have been stored in liquid nitrogen and thawed for analysis. While measurements in the clinical routine are typically carried out on fresh samples, but biomarker discovery studies often utilize liquid nitrogen stored samples, a FT-mediated pre-analytical impact factor could explain the global challenge translating protein-based biomarker candidates from the discovery stage to clinical use. 823

824

P-II-0668 Profiling of the global proteome and the repertoire of N-termini of vitreous humor uncovers alterations of cell adhesiveness and extracellular matrix composition in rhegmatogenous retinal detachment Diego Sbardella1, Gabriele Antonio Zingale1, Sara Giammaria1, Irene Pandino1, Peter A. Bell2, Pamela Cosimi3, Luca Placentino3, Guido Ripandelli3, Tommaso Rossi3 1IRCCS Fondazione Bietti, Proteomics Research in Ophtalmology, Rome, Italy 2University of British Columbia, Vancouver, Canada 3IRCCS-Fondazione G.B. Bietti, Vitreo-Retinal Surgery, Rome, Italy Rhegmatogenous retinal detachment (RRD) is a severe eye condition characterized by the detachment of photoreceptors layer, and of the whole retina indeed, from the retinal pigment epithelium (RPE). The disease carries a heavy burden for the quality of life and for social and health costs, as a significant fraction of subjects experiences a second detachment few months after the first episode, this occurrence being associated with sight loss. Under healthy conditions, the adhesiveness between the two layers is mainly promoted by the mechanical compression exerted by the vitreous humor (VH), a fluid which shapes the eye globe and nourishes the tissues. Hence, alterations of VH composition and of the extracellular matrix remodeling are thought to underscore the disease onset, but the molecular mechanisms are largely unknown yet. Based on these premises, profiling of the perturbations of the VH proteome, may provide clues for deciphering the early molecular events triggering the RRD, thereby helping the clinicians in predicting the fraction of subjects who are more vulnerable to a second detachment. Therefore, we have here undertaken a shot-gun/bottom-up proteomic characterization of the VH isolated from primary RRD subjects (n=9) and subjects diagnosed with macular holes or pucker (controls, n=9), which are two pathologies not classifiable as retinal detachment but which are treated by VH surgery as well. Age and gender ratio were comparable between the two experimental groups as well as the general inclusion and exclusion criteria. Tryptic digests were then injected into a Orbitrap Exploris 240 (2 injections for each individual sample) coupled with a nano-HPLC system. Raw data were searched against a human proteome for exploring the composition of global VH proteome (Sequest enabling Inferys rescoring, in Proteome Discoverer 2.5) and the repertoire of natural endogenous N-termini (semi-tryptic search using FragPipe). By setting FDR<0.01 (target-decoy) and filters for identification of unique proteins, 869 proteins were identified in the VH, and up to 721 N-termini were identified. After having applied a quantile normalization, and verified quality controls (e.g., data distribution, etc.), analysis of differentially expressed proteins and N-termini, followed by data clustering and rationalization, revealed a significant perturbation of key pathways for cell adhesiveness (Wnt signalling), matrix components, stress responses and immune responses. Data obtained suggest that RRD is effectively sustained by global alterations of the VH proteome and that aberrant proteolytic events on extracellular matrix components may serve roles for disease onset and progression. 824

825

P-II-0669 Endoscope-based cold atmospheric Plasma (PLASMASKOP) as a novel therapeutic tool for the treatment of head and neck cancer Christian Scharf1, Martin Wilhelm1, Görries Fabian1, Corvin Scheffler1, Nils Rodenburger1, Achim Beule2, Jörn Winter3, Leonie Huckschlag1, Sara Rosenwirth1, Jose Gonzalez1, Chia-Jung Busch1 1University Medicine Greifswald, Otorhinolaryngology, Head and Neck Surgery, Greifswald, Germany 2University Clinic Münster, Otorhinolaryngology, Head and Neck Surgery, Muenster, Germany 3Leibniz Institute for Plasma Science and Technology (INP), Greifswald, Germany Background: Over the past decade, cold atmospheric plasma (CAP), a near room temperature ionized gas has shown its promising application in cancer therapy. In contrast to conventional anti-cancer approaches and drugs, CAP is a selective anti-cancer treatment modality. Previous experimental studies have demonstrated that CAP treatment results in significant growth inhibition of tumor cells and is able to induce apoptosis and to modulate the immune response. In contrast to already existing plasma sources, a new endoscope-based plasma source first had to be developed for applications in head and neck surgery. Methods: Cold atmospheric plasma (CAP) was generated via a new developed endoscope-based atmospheric pressure plasma jet (PLASMASKOP). Physiological and molecular effects of CAP treatment on various human head and neck squamous cell carcinoma (HNSCC) cell lines compared to human epithelial cell lines were analyzed by cell kinetic analyses, FACS analyses, COMET-Assays, Proteome and Western blot analyses. 3D respiratory epithelial organoids with embedded tumors were subsequently treated by a site-specific plasma treatment, and the molecular and biochemical effects were analyzed using transcriptomic and proteomics approaches. Results: CAP treatment effectively attenuates malignant cell growth. The results demonstrated that even a single application of a short-term CAP treatment led to an attenuation of HNSCC cell growth and motility. We studied the detailed cellular adaptation reactions for a specified plasma intensity by time-resolved comparative proteome analyses of CAP treated vs. nontreated cells to elucidate the molecular mechanisms and to define potential biomarkers and networks for the evaluation of plasma effects on human tumor cells. Results were consistent in various HNSCC cell lines. CAP treatment-specific induced post translational protein modifications that promote tumor cell death were additionally identified and characterized by mass spectrometry (e.g. modified protein activity by reactive oxygen species). Conclusion: In summary, the CAP application in head and neck surgery leads to strong anti-proliferative effects and opens up novel opportunities for the treatment of HNSCCs. As an intraoperative application, CAP may represent a promising option particularly for the treatment of tissue regions that are close to critical structures (e. g. nerves, adjacent organs). 825

826

P-II-0670 Plasma proteome profiling reveals immunological dynamics in a longitudinal yellow FeverVaccination cohort Alicia-Sophie Schebesta1, Ericka Itang1, Vincent Albrecht1, Frank Dahlström2, Johannes Müller-Reif1, Simon Rothenfusser2, Matthias Mann1 1Max Planck Institute of Biochemistry, Proteomics and Signal Transduction, Martinsried, Germany 2LMU University Hospital, Division of Clinical Pharmacology, Munich, Germany We conducted plasma proteome profiling on a cohort of 250 patients vaccinated against yellow fever (YF), monitored across five timepoints before and after receiving the YF17D vaccine. Our objective is to elucidate the immunological dynamics induced by this live-attenuated vaccine, the only active YF vaccination currently available. Blood samples were collected at defined intervals: immediately before vaccination and on days 3, 7, 14, and 28 post-vaccination. This approach enabled a detailed longitudinal analysis of the plasma proteome, resulting in 1250 samples. The YF17D vaccine is known for eliciting a robust and transient innate immune response, characterized by the activation and subsequent replenishment of circulating dendritic cells (DCs) and monocytes. This coordinated response is crucial for the development of long-lasting adaptive immunity. For proteome analysis, blood plasma samples were aliquoted into 96-well plates and prepared semi- automatically using the Bravo liquid handling platform. Briefly, 1 µL of plasma per sample was transferred to a 384-well plate containing 14 µL of denaturation buffer (40 mM CAA, 10 mM TCEP, 100 mM Tris, pH 8.4) and incubated at 95°C for 10 minutes for protein denaturation, reduction, and alkylation. Proteins were digested with Trypsin and Lys-C at a 1:100 ratio (total of 0.5 µg per sample) and incubated at 37°C overnight. Peptides were desalted and prepared for measurement using Evotips, following the manufacturer's protocol. Samples were analyzed on an Evosep One HPLC with a 100 SPD gradient, coupled to a timsTOF HT mass spectrometer using an Evosep performance column and a Bruker captive spray emitter. The mass spectrometer operated in DIA-PASEF mode, with a method optimized by pydiaid. Raw data were analyzed using DIA-NN (1.8.1). Preliminary results reveal significant protein regulation between pre- and post-vaccination timepoints. Inflammation-associated proteins such as CRP, SAA1, and ORM1 peaked at 7 days post-vaccination, while neutrophil-related proteins S100A8 and S100A9 decreased over time. Unsupervised clustering of the longitudinal data indicated that samples from the same patient clustered closely together, serving as a quality control and highlighting the benefits of paired testing methods for future analyses. Additionally, sex-dependent proteins SHBG and PZP showed high variance between patients, indicating the necessity to account for these and other anthropometric data as covariates in future analyses. Our findings contribute to a broader understanding of the immune response elicited by live-attenuated vaccines and may inform strategies for improving vaccine design and implementation. 826

827

P-II-0671 Enriching the diagnostic portfolio of the molecular tumor board with proteomics and phosphoproteomics of FFPE samples Oliver Schilling1 1University Medical Center Freiburg, Institute of Surgical Pathology, Freiburg, Germany Background and Aim: Proteome profiling still has to enter the realm of clinical pathology on a wider scale. To facilitate the integration of mass spectrometry-based proteomics into routinely executed workflows of molecular diagnostics, the usage of formalin-fixated, paraffin-embedded (FFPE) specimens is key, as this is the mainstay approach to sample handling and conservation in clinical pathology. In molecular diagnostics, the quest for multiplexed protein analytics and deep proteome-level insight into therapy-guiding markers is gaining momentum and is ideally based on the tight integration into pre-existing diagnostic workflows. These approaches extend beyond oncology to include inflammatory diseases and amyloidosis typing. Methods: We have established the routine in-depth proteomic investigation using FFPE tissue samples as well as other clinically relevant sample specimens from as little as 0,01 mm³ of tissue rendering most patient-derived samples accesible, including small needle biopsies. Additional laser capture microdissection (LMD) enables tissue and cell-specific proteome profiling in a spatial context. Sample preparation and LC-MS/MS data analysis have been standardized and automated while robust LC-MS/MS instrumentation and acquisition strategies enable fast and reproducible data acquisition to meet the needs for clinical and diagnostic applications. High-coverage phospho-proteomic profiling from FFPE specimens has been established. Results: We have integrated mass-spectrometry-based proteomics and phosphoproteomics into the diagnostic portfolio of the Molecular Tumor Board (MTB) Freiburg. To date, we have provided proteome profiling for > 250 MTB cases to support clinical decision-making. The usage of single-shot, label-free quantitation approaches has provided the required flexibility and speed. In addition, we have assembled an in-house reference dataset of > 3000 proteomes covering a variety of malignancies and leading towards a growing reference database of cancer- related proteomes as well as inflammatory diseases. Comparing individual cases against molecular- and/or entity-matched references might facilitate the identification of patient-specific tumor-driving factors. The proteomic detection of clinically relevant sequence variants such as KRAS G12C or BRAF V600E has proven applicable to foster proteo-genomic corroboration and target prioritization. Phosphoproteomics has corroborated oncogenic signaling activity in select MTB cases. Integration of these approaches into the Centers for Personalized Medicine Baden-Württemberg facilitated diagnostically meaningful interlaboratory tests that underline the robustness and reproducibility of proteome profiling. Conclusion: MS-based clinical proteomics and proteogenomics has advanced as a sensitive and robust approach to complement current molecular profiling and paves the way towards precision medicine. 827

828

P-II-0672 Phosphoproteomics adds value to treatment recommendations in molecular tumor boards Annika Schneider1, Cecilia Jensen1, Julia Woortman1, Amirhossein Sakhteman1, Christoph Stange1, Firas Hamood1, Matthew The1, Moritz Resch1, Bernhard Küster1, Stefan Fröhling2, Veronica-Maria Teleanu2, Peter Horak2, Jennifer Huellein2, Daniel Hübschmann2, Stefan Pfister3, Olaf Witt3, David Jones3 1Technical University of Munich, Chair of proteomics and bioanalytics, Freising, Germany 2NCT, Heidelberg, Germany 3KITZ, Heidelberg, Germany Precision oncology approaches employing genomics-guided targeted therapies for individual patients have provided significant survival benefits in several cancer types. However, varying response rates in solid malignancies, many patients without actionable genomic lesions, and increasing evidence that non-genomic mechanisms may play an important role in tumors indicate that genomics alone is often insufficient to inform and guide the clinical care of patients. Since most targeted anti-cancer drugs inhibit the activity of protein kinases, measuring the tumor phosphoproteome as a direct readout of kinase activity is poised to enhance molecular stratification. We established a clinical (phospho)proteomic workflow for integration into the molecular tumor board (MTB) workflow of the Germany-wide INFORM (for children with relapsed cancers) and MASTER (for young adults with refractory cancers and patients with rare tumors) registry studies and profiled > 1300 tumor tissue specimen from patients enrolled in the DKFZ/NCT/DKTK MASTER study or the INFORM registry trial. To assess the aberrant activity of druggable signaling pathways, we developed a new tumor pathway activity (TUPAC) scoring methodology using expression data of > 8,000 proteins and > 20,000 phospho- sites per patient in a heterogeneous pan-cancer cohort. TUPAC scores integrate protein expression and phosphopeptide abundance data at several levels to detect aberrant signaling pathway activities of several interdependent oncogenic signaling pathways within one tumor specimen. We show for the first time that comprehensive (phospho)proteome profiling is feasible and informative in a real-world prospective precision oncology setting. Discussion of (phospho)proteomic data of > 500 prospective patients in weekly MTB meetings revealed that adding a (phospho)proteomic layer can supply critical information for personalized therapies that is not discernable from genomic and transcriptomic data. In an independent value evaluation (329 target recommendations in 104 patients), the phosphoproteome layer was decisive in 22% of all recommended targets. Based on anecdotal cases with available clinical follow-up, we provide evidence that (phospho)proteome profiling carries important diagnostic value by detecting actionable tumor-driving kinase signaling in patients without actionable genomic lesions or by functionalizing genomic variants of unknown significance. Overall, our work demonstrates the utility of the additional phosphoproteome data layer to enhance therapeutic recommendations in molecular tumor boards. 828

829

P-II-0673 Workflow for rapid proteome and phosphoproteome profiling of aging in mouse tissues Felix Schneidmadel1, Denys Oliinyk1, Aigerim Kassym1, Caroline Tschernjawski1, Patrick R. Winterhalter2, Andreas Simm2, Florian Meier-Rosar1 1Jena University Hospital, AG funktionelle Proteomeanalyse, Jena, Germany 2Halle (Saale) University Hospital, Department for Cardiac Surgery, Halle (Saale), Germany Aging is a primary contributor to the development of late-life diseases, yet the fundamental processes and complex molecular changes involved remain underexplored. Investigating tissue-specific changes during the aging trajectory on multiple molecular levels could provide valuable insights into the aging process and altered molecular pathways. Here, we present initial results from a multi-omics study that aims to analyze organ-specific changes in a cohort of aging mice at multiple time points over the course of two years. BL/6N murine tissues were dissected, flushed with cold PBS, and snap-frozen in liquid nitrogen prior to homogenization and lysis in TRIzol. Samples were prepared for mass spectrometry (MS)-based proteomics using an adapted protein aggregate capture (PAC)-based protocol (Kassym et al., in preparation). To add a second layer of information, we enriched phosphopeptides using our recently described µPhos protocol (Oliinyk et al., bioRxiv 2023). Purified peptides were separated with a Bruker nanoElute liquid chromatography (LC) system on a homemade 15 cm C18 reversed-phase column within 30 min. The LC was coupled via a CaptiveSpray nano- electrospray source to a TIMS quadrupole time-of-flight mass spectrometer (Bruker timsTOF HT), and all data were acquired in dia-PASEF mode (Meier et al., Nat. Methods 2020). Raw data were processed in Spectronaut (v19.0) without experimental spectral libraries ("directDIA+"). Our first objective was to establish a robust and scalable workflow for fresh-frozen tissue samples lysed in TRIzol. Pilot experiments from brain, colon, heart, kidney, liver and muscle tissue resulted in 3,100 (muscle) to 6,400 (colon) quantified protein groups in 30-min dia-PASEF runs with high reproducibility. Note that while we found a slightly higher proteome coverage (up to 7200 proteins) with a standard sodium deoxycholate (SDC) lysis protocol, TRIzol offers the ability to perform DNA and RNA sequencing from the same sample making it an attractive choice for multi-omics studies. Next, we tested the possibility to enrich and quantify phosphopeptides from TRIzol tissue samples. Starting from ~20 µg protein input material, we identified approximately 6,500 protein phosphorylation sites in colon with an enrichment selectivity of ~85%. From these results, we conclude that our workflow is very well suited for the large-scale analysis of multiple molecular layers from tissue samples and we envision that longitudinal proteome and phosphoproteome profiles across organs will provide novel insight into molecular mechanisms of aging. 829

830

P-II-0674 Tumour cell-induced platelet aggregation revealed different phenotypes between PDAC plasticity subtypes – A proteomic comparison between two pancreatic cancer cell lines Katharina Schulz1, Thorben Sauer1, Antje Rackisch1, Julia Horn1, Caroline Gruner1, Emma Neumann1, Katja Klempt-Gießing1, Frank Gieseler2, Timo Gemoll1 1UKSH Lübeck, Department of Surgery, Section for Translational Surgical Oncology and Biobanking, Luebeck, Germany 2UKSH Lübeck, Division of Experimental Oncology, Department of Hematology and Oncology, Luebeck, Germany Platelets have been shown to play an important role in metastasis and tumour-cell proliferation. In this context, the interaction of platelets with tumour cells is another prerequisite for successful metastatic dissemination. To investigate the interplay between pancreatic cancer cells and platelets (TCIPA), we compared the tumour-cell and platelet proteome during platelet aggregation respecting distinct plastic PDAC subtypes. We isolated platelets from healthy donors and followed their crosstalk after calcification and incubation with either TRAP (+ control), only CaCl2 (- control), and one of the SILAC-labelled pancreatic cancer cell lines BxPC3 or PanC1. Aggregation was analysed using aggregometry, immunochemistry, and nanoparticle tracking analysis (NTA). Samples were subsequently prepared for data-independent acquisition parallel accumulation-serial fragmentation (diaPASEF) mass spectrometry to perform proteomic analysis. Machine learning algorithms, principal component analyses, and gene set enrichment analyses were employed to identify differential expressed protein features. Although both cell lines were able to induce aggregation, we found that BxPC3 (epithelial subtype) induced faster and earlier platelet aggregation compared to PanC1 (mesenchymal subtype). After TCIPA, 22 proteins were significantly upregulated in BxPC3 compared to PanC1, whereas 58 proteins were significantly upregulated in PanC1 compared to BxPC3 (q-value £ 0.05 and log2 fold-change ³ 1. We also found 106 proteins significantly upregulated in both cell lines after TCIPA, which are part of platelet adhesion and aggregation pathways. The results highlight the identification of important proteins involved in the interaction between platelets and pancreatic cancer cells for the first time. Additionally, different pancreatic subtypes revealed distinct TCIPAs that could be used to characterise PDAC progression. 830

831

P-II-0675 Exploring extracellular vesicle proteins as source of biomarkers for early detection of meningioma and predicting their therapeutic implications Swati Sharma1, Rashmi Rana1, Kirti Chauhan1, Ankita Singh1, M.H. Yashavarddhan Mohanty1, Kriti Jain1, Poonam Gautam1, Prem Prakash1, Mahesh J. Kulkarni1, Reema Banarjee1, Satnam Singh Chhabra1, Rajesh Acharya1, Samir Kumar Kalra1, Anshul Gupta1, Sunila Jain1, Nirmal Kumar Ganguly1 1Sir Ganga Ram Hospital, New Delhi, India Background: Meningioma (MG) is a most common primary intracranial brain tumor. Patients survival and wellbeing can be increased by using immunological diagnostic assays, which are able to detect cases earlier and be used frequently. These techniques, in particular, depend on specific antibodies to identify the multitude of blood proteins known as biomarkers. Proteomic techniques are used to identify disease related Biomarkers in body fluids offering low-cost benefits and are less invasive. Cells release exosomes in both healthy and unhealthy circumstances. They are good source of biomarkers for clinical diagnostics because they transport proteins and nucleic acids from their host cells that are indicative of pathophysiological abnormalities. Exosomal proteins have the ability to control various aspects of tumor growth, such as metastasis, angiogenesis, the epithelial- mesenchymal transition, immunological modulation associated with tumors, and microenvironment reconstruction. Developing a drug to promote or inhibit a certain chemical route or cycle in the body generally requires a thorough understanding of the pathways linked to disease. Drugs are made to target particular proteins inside the cellular network. Here, efforts have been made to show the application of exosomal proteome in meningioma detection using mass spectrometry and study drug development process by molecular docking simulations. Methods: Exosomes were isolated from 36 blood plasma samples which included 27 samples from MG grade 1 and 9 from healthy control samples. Isolated exosomes were characterized using techniques western blotting, NTA and flow cytometry. After characterization, samples were prepared for Mass-spectrometry (MS) and results were analyzed. For validation, ELISA of 3 selected proteins alongwith their in silico analyses was performed. Molecular Docking Simulations were also conducted for the target receptor (PI3K) and selected compounds. Results: Based on the literature search, fold change of exosomal proteins obtained from MS and in silico analyses, 3 proteins were selected. Validation for 3 proteins was done by performing ELISA. When compared to healthy people, we found that meningioma (Gr I) patients' individual plasma specimens had considerably higher levels of Galectin-3BP, TGFBI, and POSTN. Further by using in silico approach to find FDA compounds as small inhibitors against PI3K/Akt pathway (common pathway of three proteins) using molecular docking simulations we obtained 4 drugs namely Cladribine > Fexofenadine > Dobutamine > Labeltalol. Conclusion: Elevated levels of three exosomal proteins; Galectin-3BP, TGFBI and POSTN were identified and validated through MS analysis and ELISA respectively. These results alongwith in silico analysis proved these proteins as biomarkers of meningioma. Here, we have also found a few FDA-approved drugs through various computational techniques and checked their binding affinity with our target PI3K. 831

833

P-II-0676 Search for novel diagnostic markers of chronic kidney disease in cats using proteome analysis Momori Shimizu1, Hiroto Maeda2, Yoshio Kodera3, Kazuyuki Sogawa1 1Azabu University, Biochemistry, Sagamihara, Japan 2Maeda Veterinary Hospital, Hokkaido, Japan 3Kitasato University School of Science, Center for Disease Proteomics, Kanagawa, Japan Chronic kidney disease (CKD) is a common disorder and cause of death in cats. In the classification proposed by the International Renal Interest Society (IRIS), stage 1 and 2 CKD are difficult to diagnose accurately using markers, in comparison with normal controls. We recently described a simple and highly reproducible tandem mass tag labelling method for identifying potential disease-marker candidates among low-abundance plasma proteins. We identify novel plasma biomarkers that are potentially superior to classical markers for early and specific detection of CKD stage 1 cats. In the current study, Plasma samples were obtained from 43 normal controls and from 19 cats with CKD (stage 1 and 2). To identify new serum biomarkers for CKD, six plasma samples (two normal control samples, two CKD stage 1 samples and two CKD stage 2 samples) were extracted with LCA lectin magnetic beads, differentially labelled with TMT, digested with trypsin and subjected to analysis using LC-MS/MS. Apolipoprotein A-IV was identified as a protein with lower levels in CKD stage 1 cats. An ELISA of plasma sialylated Apolipoprotein A-IV showed within-run (6.0-8.2%) and between-day (9.1–10.4%) reproducibility. Plasma sialylated Apolipoprotein A-IV levels measured with this assay were significantly greater in normal controls than in CKD stage 1 cats (0.17±0.06 vs. 0.07±0.01 AU/mL, p<0.01). These results indicate that sialylated Apolipoprotein A-IV may be useful as a complementary marker with plasma creatinine, BUN and SDMA for detection of CKD stage 1 cats. 833

834

P-II-0677 Search for novel diagnostic markers of pancreatic ductal adenocarcinoma using secretome and proteome analysis Noriaki Shinoda1, Shigetsugu Takano2, Yoshio Kodera3, Kazuyuki Sogawa1, Masayuki Ohtsuka2 1Azabu University, Biochemistry, Sagamihara, Japan 2Graduate School of Medicine, Chiba University, General Surgery, Chiba, Japan 3Kitasato University School of Science, Center for Disease Proteomics, Kanagawa, Japan Pancreatic ductal adenocarcinoma (PDAC) accounts for 85–90% of all pancreatic tumours. The median survival of all PDAC patients is less than 6 months, and the recent 5-year-survival rate is approximately 8%. One of the most crucial reasons for the poor prognosis is the lack of early diagnostic markers for PDAC. To overcome this and improve the outcomes of patients with PDAC, there is an urgent need to identify highly sensitive and specific markers for early detection. The widely used serum-circulating markers for PDAC, carbohydrate tumour– associated antigen 19-9 (CA19-9) and carcinoembryonic antigen (CEA), are not sufficiently accurate to be used as early diagnostic markers. In the present study, we employed the TMT method to generate comparative protein profiles of PDAC cell lines. Furthermore, we compared the serum levels of candidate proteins to evaluate the ability to discriminate between PDAC and healthy controls. Using tandem mass tag labelling and LC-MS/MS, we performed comparative analyses of secreted proteins in culturing media obtained from human PDAC cell lines (BxPC‐3, PANC‐1, MIA PaCa‐2, Capan‐2, Capan‐1, and AsPC‐1) to identify serum biomarkers for PDAC. In validation studies, we evaluated the discriminatory power of candidate proteins. Of 33 proteins, One protein was identified with the increase of protein levels among these all six cell lines, and Armadillo repeat-containing X- linked protein 4 (ARMCX4) was selected for further analysis. The sera levels of ARMCX4 were significantly higher in the preoperative PDAC patients than those in the postoperative ones (P<0.001). In future study, we will verify this result with more number of samples, and pursue the usefulness of ARMCX4 as a novel diagnostic marker in PDAC. 834

835

P-II-0678 Intra-sample three-point quantitative calibration using metabolically labelled QconCATs Florian Christoph Sigloch1, Michael Forchheim1, Mario Hofweber1, Michelle Batsch1 1PolyQuant GmbH, Bad Abbach, Germany Quantitative concatamers (QconCATs) have been used since almost 20 years as internal quantification standards for multi-protein panels in targeted and untargeted proteomics assays. Classical approaches use heavy stable isotope labelled QconCATs for single-point internal calibration. The most common labelling strategy uses the labelled amino acids arginine and lysine (SILAC). Here, we show the advantages and drawbacks of metabolic QconCAT labelling using 15N salt and 13C glucose compared to classical SILAC labelling. By spiking multiple differently labelled QconCATs at increasing concentrations into the same sample, we establish an intra-sample three-point calibration curve. Next, we compare the quantitative performance of intra-sample calibration in a multi-protein MRM panel with one-point calibration and inter-sample multi-point calibration. 835

836

P-II-0679 Transmembrane glycoprotein GPNMB is a candidate predictive biomarker and potentially a therapeutic target in metastatic renal cell carcinoma Jan Simonik1, Pavla Bouchalova1, Petr Lapcik1, David Potěšil2, Jan Podhorec3,4, Lucia Janacova1, Alice Hlobilkova5, Katerina Juraskova6, Alexandr Poprach3,4, Milan Hora7, Ondrej Fiala8,9, Pavel Bouchal1 1Masaryk University, Biochemistry, Brno, Czech Republic 2Masaryk University, Proteomics Core Facility, CEITEC, Brno, Czech Republic 3Masaryk Memorial Cancer Institute, Comprehensive Cancer Care, Brno, Czech Republic 4Masaryk University, Comprehensive Cancer Care, Brno, Czech Republic 5Masaryk Memorial Cancer Institute, Oncological Pathology, Brno, Czech Republic 6Masaryk University, Core Facility Bioinformatics, CEITEC, Brno, Czech Republic 7Charles University, Faculty of Medicine in Pilsen, Urology, University Hospital, Pilsen, Czech Republic 8Charles University, Faculty of Medicine in Pilsen, Oncology, University Hospital, Pilsen, Czech Republic 9Charles University, Faculty of Medicine in Pilsen, Biomedical Center, Pilsen, Czech Republic Background: Metastatic renal cell carcinoma (mRCC) is a serious disease with adverse prognosis often treated with receptor tyrosine kinase inhibitors (TKI). However, approximately a half of mRCC patients do not profit from TKI, and there is no clinical molecular marker identifying these non-responders. Methods: To address this issue, we performed a retrospective proteomic study on 53 mRCC tumors treated with TKI (30 non-responders vs. 23 responders) using next-generation, data independent acquisition mass spectrometry. To investigate the potential of key identified protein as an alternative therapeutic target, we knocked-out its expression using CRISPR/Cas9 and performed migration and invasion studies in 786-0 RCC cell line. Results: 6183 protein groups (FDR 0.01) were consistently quantified in the proteomics dataset. Analysis of differential protein abundance identified 12 proteins associated with poor treatment response, of which 5 were successfully confirmed in a validation cohort (12 non-responders vs. 10 non-responders). Of these, transmembrane glycoprotein B (GPNMB) exhibited the best predictive value for treatment response (AUC=0.753, p=1.4.10-9), was associated with progression-free survival (HR=1.403, padj=0.0207) and trend of increased GPNMB levels was visible in an independent cohort (n=40) using immunohistochemistry. Comparison of parental and GPNMB-/- cells showed that GPNMB supports migration capacity of 786-0 cells in scratch (p=0.0073) and Transwell (p < 0.0001) assays, and also cell invasiveness in Transwell (p=0.0009) and 3D invasion (p < 0.0001) assays. Pathway analysis associated GPNMB deregulation with enrichment of INFLAMMATORY_RESPONSE pathway in both mRCC tissues and 786-0 cells. Conclusions: Our data show that GPNMB has potential to serve as predictive biomarker of poor TKI response and hypothetically also as mRCC therapeutic target since GPNMB blockage might overcome TKI resistance. Supported by Ministry of Health of the Czech Republic, grant nr. NV19-08-00250, all rights reserved. CIISB, Instruct-CZ Centre of Instruct-ERIC EU consortium, funded by MEYS CR infrastructure project LM2023042 and European Regional Development Fund-Project "UP CIISB" (No. CZ.02.1.01/0.0/0.0/18_046/0015974), is gratefully acknowledged for the financial support of the measurements at the CEITEC Proteomics Core Facility. Supported by the project National Institute for Cancer Research (Programme EXCELES, ID Project No. LX22NPO5102)— Funded by the European Union—Next Generation EU. 836

837

P-II-0680 FFPE tissue proteomics (LC-MS/MS): applications in diagnostic molecular pathology Arslan Ali1, Sven Mattern1, Vanessa Hollfoth1, Eyyub Bag1, Mohamed Ali Jarboui2, Karsten Boldt2, Boris Macek3, Marius Ueffing2, Stephan Singer1 1University Hospital Tübingen, Institute of Pathology, Tuebingen, Germany 2University of Tübingen, Institute for Ophthalmic Research, Tuebingen, Germany 3University of Tübingen, Proteome Center Tübingen, Tuebingen, Germany Formalin-fixation and paraffin-embedding (FFPE) is the standard preservation process of diagnostic tissue specimens in pathology institutes worldwide. FFPE provides excellent tissue architecture for histopathological evaluation and allows sample storage at low costs for decades. Optimization of sample preparation protocols rendered even low amounts of FFPE tissue (including endoscopic biopsies or punch biopsies) amenable to liquid chromatography - tandem mass spectrometry (LC-MS/MS). With millions of specimens being stored in pathology archives the value of FFPE tissue in proteomics-based identification of diagnostic, prognostic, and predictive biomarkers cannot be overestimated. Here, we present various diseases in which we use FFPE LC-MS/MS as a powerful tool in diagnostic molecular pathology either complementing or potentially replacing other techniques in routine pathology. Among these, amyloidosis subtyping may serve as a paramount example for which FFPE- LC-MS/MS can be considered as the diagnostic gold standard with better performance than the widely used immunohistochemical based approach. The term "amyloidosis" defines a disease, sharing the deposition of misfolded proteins in the interstitium of various tissues by formation of insoluble fibrils. These deposits damage the tissue structure and impair cell or organ function with varying severity and dynamics, depending on the amyloidosis subtype. Thus, the clinician relies on a precise amyloidosis subtyping for adequate patient care. Other examples in which we use FFPE-proteomics to support the clinical decision making are patients discussed in the molecular tumor or inflammation board (MTB/MiB) of the center for personalized medicine (ZPM). Here, FFPE proteomics also in combination with AI can clarify the diagnosis in borderline or unclear cases (e.g. Crohn"s disease vs. ulcerative colitis or cancer of unknown primary (CUP)). Furthermore, proteomic profiles may also reveal or confirm (as suggested by genomic or transcriptomic analyses) predictive biomarkers for individualized treatment decisions. Taken together, FFPE-LC-MS/MS has already entered the stage of advanced molecular pathology and will further exploit its full potential in precision diagnostics and personalized medicine. 837

838

P-II-0681 Proteomics-driven biomarker discovery for M3913 in multiple myeloma xenograft models Jasjot Singh1, Jens Baumgärtner1, Richard Schneider2, Frank Czauderna2, Frank Fischer1 1Merck, Proteomics, Darmstadt, Germany 2Merck, RU ONC In Vivo Pharmacology, Darmstadt, Germany Introduction: At Merck, we utilize proteomics for unbiased biomarker discovery across various therapeutic areas. This study focuses on the mechanism of action (MoA) of M3913, a promising therapeutic candidate for multiple myeloma (MM). Using two xenograft models—OPM2 as a responder and MM1.S as a non-responder—we aim to identify potential biomarkers and elucidate the molecular pathways affected by M3913. Our overall objective is to enhance therapeutic strategies for MM by understanding the differential effects of M3913 in these models. Methods: Samples underwent proteolytic digestion using the Protein Aggregation Capture (PAC) protocol on the KingFisher Flex system by Thermo Fisher. We employed two multiple myeloma cell line models, OPM2 (responder) and MM1.S (non-responder), for this study. Doses of 30mg/kg and 100mg/kg were administered. Time points were carefully selected: 1 hour, 4 hours, 24 hours, and 48 hours post-dose. Each experimental condition was conducted with biological replicates (N=5). Following digestion, samples were analyzed using an UltiMate 3000 RSLCnano UHPLC system coupled to a Bruker timsTOF Pro mass spectrometer, operated in DIA- PASEF mode. Data analysis was carried out using DIANN (version 1.8.1), with further downstream analysis utilizing our in-house R-scripts. Results: Our proteomic analysis identified 9647 unique proteins from 90 MM tumor samples, with a 79% overlap between OPM2 and MM1.S cell lines. Principal component analysis demonstrated clear separation between treatment groups, and differential statistical analysis highlighted significant protein regulation, especially at later time points. Gene ontology enrichment revealed significant regulation of ER-related proteins, proteotoxic stress pathways, and the unfolded protein response (UPR) in response to M3913, particularly at early time points (1h and 4h). Key proteins, including GBP2 and XBP1, were differentially regulated, confirming the precision of our proteomics approach. The regulation of IGKC was consistent with pharmacokinetic/pharmacodynamic plasma studies, underscoring the robustness of our findings. Conclusion: This study underscores the power of proteomics in uncovering the complex MoA of M3913 in MM. M3913 significantly impacts critical biological processes, including proteotoxic stress pathways, UPR, and ER stress, and modulates immunoglobulin synthesis rates. Our findings pave the way for developing biomarkers to monitor drug activity, track therapeutic responses, and enable patient selection in MM treatment. These insights are invaluable for guiding future research and therapeutic strategies at Merck. 838

839

P-II-0682 Proteomic remodeling of the heart following transplantation: the signature of cardiac allograft vasculopathy Mikkel Skjoldan Svenningsen1, Jonathan Achter1, Thomas Hartvig Lindkær Jensen2, Julie Lyng Forman3, Finn Gustafsson2, Alicia Lundby1 1University of Copenhagen, Department of Biomedical Sciences, Copenhagen, Denmark 2Copenhagen University Hospital – Rigshospitalet, The Heart Centre, Copenhagen, Denmark 3University of Copenhagen, Section of Biostatistics, Copenhagen, Denmark The major detriment to long-term survival with a transplanted heart is cardiac allograft vasculopathy (CAV). The molecular pathophysiology of CAV is poorly understood. Once CAV is diagnosed, treatment is ineffective, but early (preangiographic) detection would enable a more timely treatment strategy. The aim of the study is to investigate the early changes in the heart proteome following a transplantation and to identify differences associated with CAV development. As part of routine clinical care following heart transplantation, the graft is monitored by collecting endomyocardial biopsies at defined time points. These are currently evaluated by histology, but some formalin- fixed and paraffin-embedded (FFPE) tissue remains, enabling research into new diagnostic strategies. We have established a state-of-the-art workflow to perform deep quantitative proteomics measurements from cardiac FFPE samples on a timsTOF HT instrument (diaPASEF). To appropriately analyze proteomics data with repeated measurements taken from the same subject, we have implemented a regression strategy using a linear mixed model with an unstructured covariance pattern for data analysis. We identified more than 5000 unique proteins across 200 endomyocardial biopsies, corresponding to 40 patients and 5 time points per patient. We observed a substantial change in the proteome over time as the graft adapts to its new environment in the transplant recipient. Notably, we identified proteins associated with the development of CAV. This study adds to the understanding of the early adaptation of a transplanted human heart and could potentially lead to identification of biomarkers for early CAV detection. 839

840

P-II-0683 Search for novel diagnostic markers of pancreatic ductal adenocarcinoma using secretome analysis Kazuyuki Sogawa1,2, Momori Shimizu1, Haruka Takedomi1, Shigetsugu Takano3, Yoshio Kodera4, Masayuki Ohtsuka3 1Azabu University, Biochemistry, Sagamihara, Japan 2Graduate School of environmental health, Azabu University, Molecular Diagnosis, Sagamihara, Japan 3Graduate School of Medicine, Chiba University, General Surgery, Chiba, Japan 4Kitasato University School of Science, Center for Disease Proteomics, Kanagawa, Japan Pancreatic ductal adenocarcinoma (PDAC) accounts for 85–90% of all pancreatic tumours. The median survival of all PDAC patients is less than 6 months, and the recent 5-year-survival rate is approximately 8%. One of the most crucial reasons for the poor prognosis is the lack of early diagnostic markers for PDAC. To overcome this and improve the outcomes of patients with PDAC, there is an urgent need to identify highly sensitive and specific markers for early detection. The widely used serum-circulating markers for PDAC, carbohydrate tumour– associated antigen 19-9 (CA19-9) and carcinoembryonic antigen (CEA), are not sufficiently accurate to be used as early diagnostic markers. The interplay between cancer cells and stromal components, including soluble mediators released from cancer cells, contributes to the progression of PDAC. Here, we set out to identify key secreted proteins involved in PDAC progression. In the present study, we performed secretome analyses of culture media of mouse pancreatic intraepithelial neoplasia (PanIN) and PDAC cells using Stable Isotope Labeling by Amino acid in Cell culture (SILAC) with click chemistry and liquid chromatography-mass spectrometry (LC- MS/MS). Furthermore, we compared the serum levels of candidate proteins to evaluate the ability to discriminate between PDAC and healthy controls. The sera levels of Protein X were significantly higher in the preoperative PDAC patients than those in the postoperative ones (p 840

841

P-II-0684 The PATH biobank – fostering breast cancer research Franziska Sommermeyer1, Kathrin Lerchl1, Heinz Bodenmüller1 1Stiftung PATH, PATH Biobank, Munich, Germany Introduction: For proteomics analysis, high quality of biomaterials is mandatory. The PATH biobank can offer best quality of tissue and blood specimen and annotating data sets in the field of breast cancer. PATH Biobank is a non-profit organization and coordinates a network of five certified breast cancer centers in Germany. There, breast cancer samples are collected according to SOPs and stored in the gas phase of liquid nitrogen. Scientists from academic institutions and industry can receive samples and annotating data sets after application with a scientific rationale and a positive vote on ethics. Materials and methods: The PATH Biobank operates a centralized database and a decentralized biorepository. The samples are stored at the five participating institutes of pathology. Tumor tissue as well as normal (benign) adjacent tissue and blood serum aliquots are prepared within a cold ischemia time < 45 min, labeled, and stored at -160 °C according to SOPs. The donors have given broad informed consent for the use of the samples and data for cancer research. With an extensive long-term follow-up program, disease progression and breast cancer therapy are being monitored. Results: Biomaterial from more than 12,800 breast cancer patients was collected since 2004. A classification of breast cancer into the intrinsic molecular subtypes is possible for 90 % of the donors based on the annotating data sets. Information about disease progression is available for 65 % and overall survival for 90 % of the donors. Since 2008 samples and data have been used to support 43 research projects. In total, more than 20 publications have been published so far, with results based on samples or data from the PATH biobank. Feedback from researchers using PATH samples indicate an excellent quality of the specimens. Analyzed PATH samples have a high tumor content and a good RNA integrity. This also emphasizes the importance of sample storage at -160 °C and demonstrates the high standard in PATH"s pre-analytical sample processing. Conclusions: PATH"s adequately processed and stored breast cancer specimens, including the annotating data sets, are a valuable source for successful proteomics research. 841

842

P-II-0685 Analysis of women's health-related proteins in dried blood spots William Stauch1, Johan Olausson2,3,4, Annika Bendes1, Mikael Ström4, Olof Beck5, Ingvar Rydén6,7, Jochen Schwenk1 1KTH Royal Institute of Technology, Science for Life Laboratory, Department of Protein Science, Stockholm, Sweden 2Östersund Hospital, Department of Laboratory Medicine, Östersund, Sweden 3Clinical Chemistry, Department of Medical Sciences, Uppsala, Sweden 4Capitainer AB, Solna, Sweden 5Karolinska Institute, Department of Clinical Neuroscience, Stockholm, Sweden 6Linköping University, Department of Biomedical and Clinical Sciences, Linköpng, Sweden 7Region Kalmar County, Department of Research, Kalmar, Sweden Female fertility is tightly regulated by hormones circulating in the blood. Currently, the primary blood sampling procedure for such analysis is a venous puncture, necessitating the involvement of health care professionals. A minimally invasive alternative to obtain and measure blood protein levels can be offered by self-sampling devices that generate quantitative dried blood spots (qDBS). The aim was to assess the potential and accuracy of determining fertility-related blood biomarkers in quantitative DBS (qDBS, Capitainer AB) using multiplex immunoassays. Nine proteins were evaluated for their detectability and precision in quantifying protein levels in qDBS. In a pilot study with samples from anonymous donors, all proteins were detectable above the background, but only five of the nine proteins could be quantified accurately in qDBS. The measured protein concentrations in qDBS were close to levels previously reported in the literature for serum and plasma levels, supporting the potential of the results. A second pilot with 100 qDBS samples confirmed the utility of the five quantifiable proteins. The results showed differences between males and females and three proteins associated with age ranges related to the onset of menopause. With a third study of 100 qDBS and plasma samples, we replicate our observations and compare the qDBS data with plasma levels obtained in a clinical chemistry lab. The levels correlate well, across the techniques and sample types, for three of the hormones (Pearson R2 0.84-0.94). So far, our results have shown noticeable and reliable differences in protein levels between the donors, sexes, and ages. Their utility to support clinical diagnostics is being tested. 842

843

P-II-0686 Identifying new immunotherapy targets for pediatric leukemia using cell surface proteome profiling. Jorik van Rijn1,2, Maarten Altelaar1, Stefan Nierkens2, Kelly Stecker1 1Utrecht University, Pharmaceutical Sciences, Utrecht, Netherlands 2Princess Maxima Center for Pediatric Oncology, Utrecht, Netherlands Overview: Immunotherapies that direct cytotoxic immune cells to tumor targets, such as CAR-Ts, have shown tremendous clinical success for certain B-cell malignancies(1). However, effective CAR-T strategies for pediatric T cell lymphoblastic leukemia/lymphoma (T-LBLL) are limited by the lack of tumor-specific antigens targeting malignant progenitor T cells while preserving healthy mature T cells(2). Here we apply cell-surface proteomic profiling of T-LBLL patient samples, Patient-Derived Xenografts (PDXs), cell lines and healthy-donor mature T cells to identify new tumor-specific antigens for T-LBLL CAR-T therapy. Furthermore, we perform method optimization to improve plasma-membrane protein enrichment in surfaceome profiling. Methods: Sialylated cell surface glycoproteins were labeled with biotin followed by streptavidin enrichment and on-bead digestion(3). Samples were analyzed by LC-MS using DIA acquisition on an Exploris-480. Flow cytometry was used to isolate progenitor T cell populations in healthy thymocytes for validating expression profiles of candidate immunotherapy targets. Results: Direct surface proteome profiling of pediatric patient samples resulted in the quantification of approx. 1500 proteins per patient, with strong enrichment for plasma membrane proteins (Fig. 1A-D). Patient samples displayed both distinct and overlapping profiles with T-LBLL PDX and cell-line models. From our identification list, 233 putative T-LBLL surface protein immune therapy targets were selected based on their conserved enrichment across patient samples and absence in the surface profiles of healthy-donor mature T cells (Fig. 1E). We further filtered our list to exclude proteins with high RNA expression profiles in healthy tissues and identified 80 candidate immune therapy targets. We validated the expression of one putative T-TBLL antigen for CAR-T therapy, CD1b, that displays higher expression across patient samples (Fig. 1F) and greater specificity to immature T-cells than the current leading T-TBLL targets in clinical trials(4-6) (Fig. 1G). Additionally, by implementing enrichment steps for cell viability prior to surface protein labeling, we significantly improve plasma membrane protein enrichment in our surfaceome profiling. Conclusions: MS-based surfaceome proteomic profiling is a powerful technique to elucidate new targets for cell therapy. By improving surface proteome enrichment in our optimized workflow, we can confidently identify more surface proteins with higher sensitivity and specificity. Our data elucidates an exciting new potential CAR-T target for T-TBLL treatment, addressing an unmet clinical need in pediatric oncology. References: 1. Rowe JM, et. al. Blood. 2005 2. Li Z, et. al. Blood. 2023 3. Wollscheid B, et. al. Nat Biotechnol. 2009 4. Pinz K, et. al. Leukemia. 2016 5. Cwynarski K. et. al. Blood. 2022 6. Sánchez-Martínez D, et. al. Blood. 2019 843

845

P-II-0687 Single Cell Deep Visual Proteomics (scDVP) of human hepatocytes in alpha-1 antitrypsin deficiency Sophia Steigerwald1, Florian Rosenberger1, Katrine Holtz Thorhauge2,3, Sönke Detlefsen3,4, Maria Wahle1, Maximilian Zwiebel1, Caroline Weiss1, Malin Fromme5, Peter Boor6, Aleksander Krag2,3,7, Pavel Strnad5, Matthias Mann1,8 1Max Planck Institute of Biochemistry, Department of Proteomics and Signal Transduction, Martinsried, Germany 2Centre for Liver Research, Department of Gastroenterology and Hepatology, Odense, Denmark 3University of Southern Denmark, Department of Clinical Research, Faculty of Health Sciences, Odense, Denmark 4Odense University Hospital, Department of Pathology, Odense, Denmark 5University Hospital Aachen, Department of Internal Medicine III and IZKF, Gastroenterology, Metabolic Diseases and Intensive Care, Aachen, Germany 6University Hospital Aachen, RWTH Aachen University, Institute of Pathology, Aachen, Germany 7Danish Institute of Advanced Study (DIAS), University of Southern Denmark, Odense, Denmark 8Novo Nordisk Foundation Center for Protein Research, Copenhagen, Denmark Single-cell Deep Visual Proteomics (scDVP) enables the resolution of the context-dependent, spatial proteome of single cells in intact tissue across biological conditions and diseases. In this study, we focus on alpha1-antitrypsin deficiency (AATD), a genetic disorder caused by an autosomal, co-dominant inherited mutation in the SERPINA1 gene. Misfolded alpha1-antitrypsin (AAT) protein accumulates within hepatocytes, which induces cellular stress and ultimately leads to liver fibrosis and cirrhosis in affected individuals. As these liver complications often advance silently and disease manifestation differs greatly across affected individuals, timely detection and management of the disease are hindered. While current understanding suggests a correlation between the severity of liver damage and the extent of AAT aggregation, the precise molecular events in time and space are not well understood. Here, we leverage the mass resolution, speed, and sensitivity of the Thermo Scientific Orbitrap Astral mass spectrometer (MS) and our Deep Visual Proteomics (DVP) workflow. Decreasing sample loss, by minimal sample transfer steps and ensuring high LC/MS sensitivity with a combination of the Evosep Whisper40 low flow gradient, Ionopticks Aurora columns and the Orbitrap Astral MS. We first acquired deep proteomes of liver hepatocytes from samples containing 10-15 hepatocytes (~100 hepatocyte shapes) each to characterize the relationship between aggregate load and hepatocyte health. We then applied single cell DVP to precisely map the proteome of individual hepatocytes in regions with striking spatial disease manifestations, such as cells with high and no load of AAT in direct proximity. This data uniquely enabled us to characterize the proteome of a single hepatocyte shape (~1/3 to 1/2 of a hepatocyte) at a depth of up to 3600 proteins. We find that the hepatocyte proteome was strongly dependent on both fibrosis stage and AAT load. We observed significant changes in the abundance of proteins produced in the ER and targeted for secretion, highlighting an unfolded protein response (UPR) involving all canonical UPR pathways and a unique reactive oxygen species (ROS) defense in hepatocytes with high AAT aggregate load. The combination of high-sensitivity proteomics by mass spectrometry with the spatial resolution of (sc)DVP thus offers clinical insights in AATD and resilience mechanisms of affected cells, harboring the potential for tailored AATD management strategies. Fig. 1 (optional) 845

846

P-II-0688 MultiOmics landscape of pathotypes in rheumatoid arthritis – enroute to digital health using AI clinical diagnostics by nanoproteomics and AI based imaging diagnostics Allan Stensballe1, Jacob Skallerup Andersen1, Mikkel Eggert Thomsen1, Christopher Aboo1, Søren Andreas Just2 1Aalborg University, Gistrup, Denmark 2Sygehus Svendborg, Svendborg, Denmark Introduction and Aim: Rheumatoid arthritis (RA) is a chronic autoimmune disease that primarily affects synovial joints. Despite significant advances in our understanding of RA, the underlying molecular mechanisms within synovial tissue of RA at the protein level remain largely unexplored. Additionally, the factors contributing to synovial histological heterogeneity, which may impact treatment outcomes, also remain largely unknown at the protein level. Mapping the proteomic landscape of RA-affected synovial tissue, extracellular matrix (ECM), and identifying determinants of histological pathotypes, could provide valuable insights into the molecular mechanisms of RA and identify new therapeutic targets. This knowledge could also pave the way for personalized medicine options, ultimately leading to improved treatment outcomes for RA patients. Methods: Synovial tissue by ultrasound-guided nano-biopsy and matched plasma samples were collected from patients with early untreated RA and longstanding RA as well as healthy controls. All samples were characterized using optimized clinical discovery proteomics profiling (EvosepOne-timsTOF PRO2), autoantibody profiling and clinical immunological profiling. Next, data integration - of the synovial tissue proteome, plasma proteome, autoantigen profiles (Sengenics), clinical scores including Krenn synovitis score, immune cell scores, synovial immune infiltration and number of circulating immune cells - was conducted to identify the molecular and cellular determinants of synovial histological pathotypes. Results: In the SYPRORA study we performed a multimodal integration of data types (mixOMICs) including deep plasma protein (>400 proteins) and synovial tissue nano-proteome (>3400 proteins) were identified and quantified after data preprocessing. Data integration identified distinct molecular and cellular signatures underlying synovial pathotypes that suggests the difference between lymphoid, myeloid, and fibroid pathotypes is a continuous spectrum that is linearly modellable. The nanoproteomic findings correlated with ARTHUR - the world's first fully automated clinical robot that performs ultrasound scans on RA patients and assesses the disease with the help of artificial intelligence. Conclusion: Proteomic characterization of nano-proteomics synovial tissue biopsies elucidates RA pathogenesis at different stages of disease and identifies molecular and cellular determinants of synovial heterogeneity and impact to ECM. Fig. 1 (optional) 846

847

P-II-0689 An end-to-end (phospho)proteomics data analysis platform for precision oncology of 1000 prospective pan- cancer patients Cecilia Jensen1, Amirhossein Sakhteman1, Firas Hamood1, Julia Woortman1, Annika Schneider1, Bernhard Küster1, Matthew The1 1Technical University of Munich, Freising, Germany In precision oncology, DNA and RNA sequencing are increasingly applied to profile rare and difficult tumors. However, the therapeutically relevant (phospho)proteome is rarely included. This can, in part, be attributed to the need for deep, reproducible profiling along with the absence of comprehensive data analysis software that manages the entire processing from raw data to patient-specific oncogenic activity. We implemented an end-to-end clinical proteomics workflow into two existing molecular tumor board programs and have been profiling >1400 rare and difficult cancer patient samples for >3 years. Here, we present the automated analysis pipeline of (phospho)proteomic data, enabling therapeutic recommendations within a turnaround time of two weeks. Methods: 16 patients (2 TMT-11 batches) are processed each week, resulting in expression data for ~8,000 proteins and ~30,000 phospho-sites per patient. Raw data is searched per batch and then combined into a single data matrix using several in-house developed tools that have been optimized to handle the large volumes of data (>4TB of raw data). This is followed by tools that enhance data completeness and reduce batch effects. Single patient proteomes are then analyzed at protein abundance and phospho activity level using z-scores calculated across patients to find aberrant activity. This information is also integrated into a kinase substrate and protein phosphorylation signaling scoring strategy. Moreover, abundances and scores are combined into TUmor Proteome ACtivity scores (TUPAC) for 24 receptor tyrosine kinases that typically are druggable targets. Preliminary data: Our activity scoring relies on relative protein and phosphopeptide quantification compared to a pan cancer background cohort to eliminate the need of healthy controls for comparison. We are continuously expanding our patient cohort for increased statistical significance and broader coverage of rare cancer types and different oncogenic mechanisms. Currently, the cohort consists of >1400 patient samples from various entities covering >100 different tumor types from >20 different tissues. >1000 prospective samples have been discussed in molecular tumor boards. To evaluate each patient"s proteome efficiently, an interactive web portal containing the cohort data was created which also generates patient-specific reports in Excel format. These contain information on the level of phosphopeptide, protein, kinase activity and our own TUPAC scores. Quantification measures are given as rank, fold change and z-scores, each computed relative to the background cohort. The portal further includes tools for batch effect assessment and quality control. We investigated the effects of our normalization procedures and observed a substantial reduction of batch effects. The outcome of our software has shown to be clinically useful, easily interpretable through the portal, representing relevant biology and to be both robust and reproducible. 847

849

P-II-0690 Insights from proteomic analysis in Crohn's disease Johanna Thiery1,2,3, Matthias Fahrner1, Adrian Fritsch4, Felix Röttele4, Peter Hasselblatt4, Oliver Schilling1 1University Medical Center, Institute for Surgical Pathology, Freiburg, Germany 2German Cancer Consortium (DKTK), Freiburg, Germany 3University, Faculty of Biology, Freiburg, Germany 4University Medical Center, Clinic for Internal Medicine II, Freiburg, Germany Crohn's Disease (CD) presents a challenging inflammatory bowel disease with rising incidences. Although biologic agents have improved treatment outcomes, response rates remain variable, and predictive biomarkers are lacking. This study investigates the proteomic profile of CD patients undergoing treatment, aiming to uncover molecular mechanisms during treatment and predict therapeutic outcomes. We analysed two patient cohorts: one comprising 34 samples classified as either inflammation or remission, and another with 47 samples taken pre- or post-therapy, with treatment response assessed via clinical and endoscopic criteria. Formalin-fixed, paraffin-embedded (FFPE) ileum biopsies were subjected to robotic sample preparation and submitted to liquid chromatography-tandem mass spectrometry (LC-MS/MS) using data- independent acquisition – parallel accumulation serial fragmentation (dia-PASEF). The proteomic analysis revealed a rich proteomic landscape, identifying an average of 6,700 proteins per sample. Cluster analyses differentiated inflammation from clinical remission, revealing a notable upregulation of 2,100 proteins during acute inflammation, including markers like myeloperoxidase and calprotectin. Treatment interventions showed significant attenuation of inflammatory pathways in responders, contrasting with minimal changes in non-responders, highlighting heterogeneous treatment responses. Gene Ontology Enrichment Analysis indicated an upregulation of metabolic and cellular respiration processes after successful treatment, suggesting a restoration of cellular homeostasis. Integrating proteomic data with clinical parameters provides a comprehensive understanding of CD pathophysiology and treatment response. Our next steps will focus on deeper analysis of the proteomic dataset, comparing different treatment strategies, and examining proteolytic activity as well as the metaproteome. 849

850

P-II-0692 Towards a noninvasive diagnosis of Peri-implantitis: proteomic analysis of Peri-implant crevicular fluid Annika Topitsch1,2,3,4, Tim Halstenbach4, Gerhard Iglhaut4, Katja Nelson4, Oliver Schilling1, Tobias Fretwurst4 1Institute for Surgical Pathology, University Medical Center Freiburg, Freiburg, Germany 2Spemann Graduate School of Biology and Medicine (SGBM), University of Freiburg, Freiburg, Germany 3Faculty of Biology, University of Freiburg, Freiburg, Germany 4Department of Oral- and Craniomaxillofacial Surgery, University Medical Center Freiburg, Freiburg, Germany Peri-implantitis, the inflammatory process affecting both the soft and hard tissues surrounding dental implants, has a prevalence of approximately 20 %. In current clinical procedure, diagnosis of peri-implantitis relies on the detection of an increased periodontal pocket depth and the radiological recording of bone loss, both of which are symptoms of an already advanced stage of the disease. To date, neither the cause of the disease nor the mechanisms of its progression have been clarified, which makes adequate therapy a major challenge. Therefore, the detection of peri-implantitis is of great importance to prevent progression and to ensure long-term success of dental implants. Crevicular fluid, localized in the gap between dental implants and the gingiva, was proposed as a noninvasive source for the identification of potential diagnostic, prognostic, or therapeutic markers in the field of periodontology and implantology. In this study, we used explorative LC-MS/MS analysis with a semi-automated sample preparation workflow to identify and quantify >3000 proteins, revealing distinct proteomic profiles for crevicular fluids of healthy and diseased dental implants. Among the upregulated proteins in peri-implantitis- affected implants are proinflammatory proteins, predominantly driven by neutrophil activity, and antimicrobial proteins. The crevicular fluid of diseased implants also exhibited higher abundances of semi-specific peptides, indicative of increased proteolytic activity. 850

851

P-II-0693 Comparative proteomic analysis of laser microdissected human glomeruli and tubules in IgA nephropathy Yudai Tsuji1, Sei Saitoh2, Yukako Ohyama1,3, Masaya Hirayama1,4, Naotake Tsuboi3, Kazuo Takahashi1,3 1Fujita Health University School of Medicine, Department of Biomedical Molecular Sciences, Aichi, Japan 2Fujita Health University, Department of Disease Systems Analysis Medicine, School of Medical Sciences, Aichi, Japan 3Fujita Health University School of Medicine, Department of Nephrology, Aichi, Japan 4Fujita Health University, Department of Morphology and Diagnostic Pathology, School of Medical Sciences, Aichi, Japan Background: IgA nephropathy (IgAN) is the most common glomerulonephritis worldwide, and 30-40% of cases progress to end-stage kidney disease. Although the glomerular mesangium is the leading site of IgA deposition in IgAN, tubulointerstitial lesions vary among patients. A new workflow for laser microdissection (LMD) coupled with mass spectrometry (MS) was developed. Using this workflow for glomeruli and tubules, we attempted to detect molecules associated with the pathogenesis and disease progression in IgAN. Methods: Using the treatment protocol for IgAN at Fujita Health University Hospital, patients with IgAN were divided into immunosuppressive (tonsillectomy and steroid pulse therapy, n=19) and conservative (n=14) therapy groups. Glomerular and tubular cross-sections were microdissected from stored formalin-fixed paraffin- embedded kidney biopsy tissues and control kidney tissues (n=10) purchased from OriGene Technologies. Samples were analyzed using Orbitrap FusionTM Tribrid mass spectrometer (liquid chromatography-mass spectrometry) coupled to EASY-nLC system, and the relative amounts of glomerular and tubular proteins were compared between the groups. The data analysis was performed using Proteome Discoverer 2.4 and Epithelial-Mesenchymal Transition (EMT) Genes Database. Results and Discussion: Glomerular proteomic analysis quantified 1807 proteins, of which 85 had significantly different abundances in patients with IgAN than in controls. Comparing the different treatment groups of patients with IgAN, 78 proteins showed significantly different abundances. These proteins are mainly a part of the complement system, such as factors H and H-related proteins. Several international genome-wide association studies have shown that these complement proteins are associated with IgAN. Therefore, these proteins are likely involved in IgAN pathogenesis and progression. Tubular proteomic analysis quantified 2354 proteins, of which 73 were significantly differentially expressed between patients with IgAN and controls. When comparing the different IgAN treatment groups, 75 proteins showed significantly different abundances. These proteins include extracellular matrix and EMT-related proteins. In patients with IgAN, EMT induction in the tubules can cause fibrosis and loss of kidney function. Conclusions: Using LMD-coupled MS, we demonstrated that the complement system in the glomerular lesion and the epithelial-mesenchymal transition system in the tubular interstitial lesion differed between different treatment groups. Investigating the relationship between these molecules and kidney prognosis is crucial for discovering new biomarkers to determine kidney prognosis. 851

852

P-II-0694 Proteogenomic profiling of mutated proteins in urine extracellular vesicles for non-invasive bladder cancer liquid biopsy Yuji Hakozaki1,2, Kazuma Sugimoto1,2, Yuta Yamada2, Haruki Kume2, Koji Ueda1 1Japanese Foundation For Cancer Research, Cancer Precision Medicine Center, Koto-ku, Japan 2The University of Tokyo, Department of Urology, Tokyo, Japan Extracellular vesicles (EVs), including exosomes, microvesicles, and apoptotic bodies, are small particles released by cells into bodily fluids like blood, urine, and saliva. A liquid biopsy using EVs is a promising and less invasive method for detecting cancer. These vesicles carry molecular information from their cell of origin, including proteins, lipids, RNA, and DNA, making them valuable biomarkers for various diseases, including cancer. We have previously identified a group of proteins with quantitatively cancer-specific high expression in plasma EVs of cancer patients for the purpose of enabling liquid biopsy of renal or colorectal cancer (Int J Cancer (2018);142:607)(Mol Cancer Res (2021);19:834). However, in many cases, these biomarker candidates were also found to be expressed in several normal organs, and their cancer specificity was insufficient. Therefore in this study, we established an original proteogenomics analysis method and developed a cancer liquid biopsy technology targeting proteins with cancer-specific somatic mutations, which are expected to have 100% cancer specificity. Bladder cancer is a cancer with a high risk of recurrence and requires regular follow-up using cystoscopy and contrast-enhanced CT after endoscopic resection. In this study, we aimed to develop a new liquid biopsy technique by detecting mutated proteins in urinary extracellular vesicles as a new, minimally invasive method of diagnosing recurrence. Seven patients with urothelial carcinoma who underwent transurethral resection of bladder tumor at the University of Tokyo Hospital were included in the study. Genomic DNA was extracted from the resected tumors and blood samples from the patients, and somatic mutations were identified by whole exon sequencing. As the result, 55-444 nonsynonymous mutations and 5-16 frameshift mutations were identified. Based on this information, the personalized proteogenomic analysis of bladder cancer tissue was performed, resulting in identification of 43 mutated proteins. More importantly, analysis of the patients" urinary EVs revealed 4 mutated proteins in common with tissues. An absolute quantification system using stable isotope-labeled peptide standards was established for these urinary EV mutated proteins, and the quantitative changes in the patient's treatment prognosis with disease progression were also monitored. In conclusion, we introduce the potential of EV mutant proteins as a new cancer liquid biopsy modality. Fig. 1 (optional) 852

853

P-II-0695 Early detection of liver cancer – the role of extracellular matrix proteins Henry Unger1, Sherin Al Aoua2, Sebastian Burbano De Lara1, Barbara Helm1, Dominic Helm3, Agustin Rodriguez Gonzalez1, Madlen Matz-soja4, Heike Bantel2, Marc Schneider5, Marcel Schilling1, Thomas Berg4, Ursula Klingmüller1 1Deutsches Krebsforschungszentrum (DKFZ), Systems Biology of Signal Transduction, Heidelberg, Germany 2Hannover Medical School, Department of Gastroenterology, Hannover, Germany 3German Cancer Research Center (DKFZ), MS-Based Protein Analysis Core Facility, Heidelberg, Germany 4Leipzig University Hospital, Clinic and Polyclinic for Oncology, Gastroenterology, Hepatology and Pneumology, Leipzig, Germany 5Thorax Clinic Heidelberg University Hospita, Translational Research Section, Heidelberg, Germany Hepatocellular carcinoma (HCC) is a leading cause of cancer deaths worldwide, primarily linked to chronic liver damage such as fibrosis and cirrhosis. During cirrhosis, activated hepatic stellate cells cause changes in the extracellular matrix. In general, it is unresolved why patients with liver cirrhosis cross the tipping point towards HCC and why some patients are affected earlier than others. To determine proteomic parameters that correlate with structural changes in human cirrhotic liver tissue and therefore could be used for the early detection of HCC, we employed a data-independent acquisition mass spectrometry approach. In addition to analyzing patient-derived liver tissue, we utilized this approach for the analysis of the blood proteome of cirrhotic patients with the aim to identify HCC-relevant signatures. For the large-scale mass spectrometric analysis of human serum and plasma, we developed a standardized protocol that includes automated sample preparation and a screening step to assess sample quality and comparability. Based on this protocol, we were able to determine the extent to which the proteomic information between serum and plasma samples differs. The initial analysis of proteomic changes in longitudinal serum samples from cirrhotic patients who developed HCC during the observation time provided first evidence for HCC-dependent protein clusters. 853

854

P-II-0696 Efficient Human Plasma Proteome Profiling at up to 240 Samples Per Day Natalie Van Landuyt1, Robert van Ling1, Paul Jacobs1, Jeff Op de Beeck1 1Thermo Fisher Scientific, Zwijnaarde, Belgium LC-MS-based proteomics stands as a vital tool for the analysis of intricate biological and clinical samples. Within the separation core, the LC column assumes a pivotal role by reducing sample complexity before its injection into the MS. The micro pillar array column (µPAC™) emerges as an innovative LC column featuring uniformly arranged pillars within a microfluidic channel, offering both highly efficient separation and minimal back pressure. In this study, we used a 5.5 cm long µPAC Neo High Throughput column that has channels filled with rectangular-shaped silicon pillars, coupled with HRAM mass spectrometry. Our goal was to expedite bottom-up proteomics profiling of neat human plasma samples while leveraging pillar arrays' benefits, such as high reproducibility and minimal sample carry-over. Using a 3-minute gradient 240 SPD method, the amount of protein groups that could be identified from neat human plasma steadily increased according to the sample load, up to a load of 200 ng plasma on column, where a plateau of about 250 protein groups was achieved. Increasing beyond 200 ng didn't significantly improve protein identification, suggesting 200 ng is sufficient for favorable outcomes in this timeframe. This was also confirmed by checking the basepeak chromatograms and overlaying the different sample loads, where 200 ng was identified as the onset of overloading. Given the extremely high throughput of this method, fair quantitation could be achieved resulting in up to 205 protein groups quantified at CV below 20% for 500 ng of plasma analyzed in triplicate. Similar observations were done across different SPD settings (170, 100), where increasing sample loading beyond 500 ng did not increase proteome depth. By extending the gradient length to respectively 5.5 and 11 min, increased depth and especially an increase in number of quantifyable proteins could be obtained. The µPAC Neo High Throughput column achieved a pragmatic compromise at 170 SPD, quantifiying 248 protein groups (below 20% CV). The impact of extending gradients beyond the 5.5 min used for the 170 SPD method appeared to be very limited, as barely any depth could be gained. Ongoing efforts are focusing on extended performance testing and comparative analysis of the susceptibility to sample related carry-over related to different LC and injection configurations. 854

855

P-II-0697 Ultra-deep MS based plasma proteomics in a hand foot and mouth disease cohort Iolanda Vendrell1,2, Georgina Berridge1, Than Tran Tan Thanh3, Le Nguyen Thanh Nhan4, Garwin Pichler5, Zuzana Demianova5, Benedikt Kessler1,2, Le Van Tan3,6, Roman Fischer1,2 1TDI/CMD, University of Oxford, NDM, Oxford, United Kingdom 2CAMS/COI, University of Oxford, Oxford, United Kingdom 3OUCRU, Ho Chi Minh City, Viet Nam 4Children's Hospital, Ho Chi Minh City, Viet Nam 5PreOmics GmbH, Planegg, Germany 6CTMGH, University of Oxford, Oxford, United Kingdom Hand foot and mouth disease (HFMD) is usually a mild illness affecting children ≤5 years old, commonly caused by coxsackieviruses A16, A10 and A6, and enterovirus A71 (EV-A71). However, since 1997, large outbreaks of severe EV-A71 associated HFMD, involving millions of infections and substantial numbers of deaths, have been reported in the Asia Pacific region. Increased detections of EV-A71 and associated neurological disease have also been documented outside of Asia. Currently, there are no clinically proven effective antivirals to treat severe HFMD. Inactivated EV-A71 vaccines have been successfully developed in China and most recently in Taiwan, but have only been implemented in China. The identification of biomarkers that could differentiate between clinical severity, predict disease progression and help in prognosis and patient stratification has become one of the main goals for this project. Recent developments in mass spectrometry-based plasma proteomics which combines new protein enrichment strategies with more sensitive and fast instruments, address the major analytical challenge in mass spectrometry-based analysis of the deep plasma proteome by decreasing the dynamic range of protein abundance in the analysed sample. This enables the detection of proteins beyond the functional plasma proteome and get into the realm of tissue leakage and cell signalling, now addressing biological pathways relevant to disease. We have profiled the proteomes of 20 HFMD samples from two different severity groups (n of 10 per group) in a direct comparison of neat plasma to bead-based enrich plasma using the PreOmics ENRICH plasma and the new ENRICHplus. Samples have been analysed by LC-MS/MS using the EvosepOne Tims-Tof-Pro platform using the 60 samples per day (spd) and DIA-PASEF acquisition. DIA-NN library free has been used for data analysis. We show that we can detect around 3200 and 4200 protein groups when using the ENRICH plasma and ENRICHplus, respectively. This means we can increase the plasma proteome depth by a 4.5 and 6-fold change compared to neat plasma when using the PreOmics ENRICH plasma and ENRICHplus, respectively. 855

856

P-II-0698 Multiplex mass spectrometry-based analysis of inflammatory proteins and their isoforms in serum samples of analog astronauts Veronika Vidova1,2, Lucie Rackova1, Margot Issertine3, Elliott James Price1,2 1Masaryk University, RECETOX, Brno, Czech Republic 2EIRENE-CZ, Brno, Czech Republic 3University of Montpellier, DMeM lab, Montpellier, France Inflammation is a complex mechanism of host defense response to stimulus (e.g. infection, tissue damage, or cancer development), tissue repair and recovery. Activated cells of the immune system release signal proteins that activate systemic response via translation of acute phase proteins (APPs). The quantification of selected highly conserved APPs, such as C-reactive protein (CRP), α-1 antitrypsin (A1AT), serum amyloid A (SAA), and α-1- acid glycoprotein (A1AG), calprotectin, haptoglobin (HP) and adaptive immunity effectors (i.e., immunoglobulins, myeloperoxidase) presumably allows for stratification among specific types of inflammation, and for differentiation between physiological and pathological responses. APPs are routinely analyzed in clinics using high throughput immunoassays based on immunochemical interaction between protein and antigen. However, immunoassays rarely quantify isoforms of APPs. Although the function of isoforms is not known, the quantification of individual isoforms may help to characterize state of inflammation. We have developed a multiplex mass spectrometry method for the quantification of APPs, adaptive immunity effectors and apolipoproteins in serum samples, including exact protein isoforms (i.e. SAA1, SAA2, CRP1, A1AT1), and to detect single nucleotide polymorphism (SNP). The method has been applied to serum samples of 6 analog astronauts. Samples were collected before and after 6-day long APICES (Astroland Project Inside Caves for Earth-based Space exploration) Space Analogue Mission, studying the effect of space travel on astronaut"s health. All analog astronauts showed low concentration of APPs before and after the mission. A slight decrease in Apolipoprotein A1 was observed after the mission. Although the samples were "blind" for analysis, the quantification of IgG isoforms meant the specific isoform ratio per astronaut was able to be identified, and samples paired. Furthermore, two analog astronauts showed a polymorphism in haptoglobin sequence TEGDGVYTLNNEK, possibly indicating haptoglobin phenotype Hp2-2 (missing a1 unit HP). 856

857

P-II-0699 Loss of SDHB manifests differently in the proteomes of neuroendocrine tumors and a tumor-derived cell line model hPheo-1 Ondrej Vit1, Sarka Dvorakova2, Jiri Neuzil2, Ales Vicha3, Tomas Zelinka4, Karel Pacak5, Jiri Petrak1 1Charles University, First Faculty of Medicine, BIOCEV, Vestec, Czech Republic 2Czech Academy of Sciences, Institute of Biotechnology, BIOCEV, Vestec, Czech Republic 3Charles University, Second Faculty of Medicine, Department of Paediatric Haematology and Oncology, Prague, Czech Republic 4Charles University, First Faculty of Medicine, Third Department of Internal Medicine, Endocrinology and Metabolism, Prague, Czech Republic 5National Institutes of Health, Eunice Kennedy Shriver National Institute of Child Health and Human Development,, Section on Medical Neuroendocrinology, Bethesda, MD, United States Pheochromocytomas and paragangliomas (PPGL) are rare neuroendocrine tumors derived from chromaffin cells. PPGL with germline mutations in the succinate dehydrogenase subunit B (SDHB), an enzyme participating in the TCA cycle and electron transport chain (ETC), are associated with a high rate of malignancy and metastasis. Unfortunately, the absence of a reliable in vitro or in vivo human PPGL model limits functional research, including drug testing. Only a single stable cell line of human PPGL origin—hPheo1—is currently available. An SDHB knockout cell sub-line hPheo-1 SDHB KO has been recently generated as a model of human SDHB-mutated PPGL. To evaluate its relevance, we asked whether this model recapitulates the specific proteome changes found in SDHB-mutated human PPGL tumors. Using 2D-LC-MS/MS with TMT-based quantitation, we performed proteomic analysis of 1) hPheo1 cells vs. hPheo-1 cells SDHB KO and b) human PPGL samples, including tumors with and without SDHB mutation. In the cell line model, we identified and quantified over 7200 proteins. The SDHB loss was associated with almost 250 differentially expressed proteins, including marked downregulation of the remaining succinate dehydrogenase subunits SDHC and SDHD but not SDHA. In addition to the complex II of ETC, most observed components of complex I, III, and IV were downregulated in the SDHB KO cell model. However, no marked changes were observed in the expression of the other TCA cycle enzymes. synthase and along with upregulated monofunctional C1-tetrahydrofolate The overlap between the changes caused by SDHB deficiency in cell line versus human PPGL was rather limited. Arginase 2 and ornithine aminotransferase upregulation was associated with SDHB defect in both tumors and cells, bifunctional methylenetetrahydrofolate dehydrogenase. This may suggest a compensatory metabolic remodeling of amino acid utilization and of the folate cycle. While the SDHB-deficient cell line downregulated the expression of all ETC complexes, SDHB mutation in tumors was generally associated with the opposite effect- mildly upregulated ETC components (except for SDHB-containing Complex II). Similarly, in contrast to the cell model, SDHB-deficient tumors showed increased expression of enzymes of pyruvate metabolism and TCA cycle (except SDHB). These energo-metabolic changes seem to point toward different modes of metabolic compensation for the SDHB loss in the complex tumor environment versus cell media in the cell line model. Alternatively, it may indicate some residual SDHB activity in the tumors (missense mutations of SDHB) compared to the effect of the complete SDHB knock-out in the cell model. Based on our observations, we conclude that any in vitro studies with hPheo-1 SDHB KO cells as a model of human SDHB-mutated PPGL should be interpreted with caution, considering the molecular differences between the model and the human PPGL. 857

858

P-II-0700 Urinary proteome profiling of rat and human unveils biomarkers of LRRK2 activity Duc Tung Vu1, Matthias Mann1, Özge Karayel1 1Max Planck Institute of Biochemistry, Proteomics and Signal Transduction, Munich, Germany in the immune processes and levels of 94 proteins associated with Pathogenic mutations in Leucine-rich repeat kinase 2 (LRRK2), such as G2019S, represent the predominant genetic cause of Parkinson's disease (PD), frequently resulting in heightened kinase activity. Consequently, LRRK2 kinase inhibitors have emerged as promising treatments for PD. Despite ongoing clinical and pre-clinical trials of these inhibitors, specific, sensitive, and non-invasive biomarkers for measuring drug target engagement or stratifying mutation carriers remain elusive. Utilizing our scalable and sensitive Orbitrap Exploris 480 mass spectrometer connected to an EASY-nLC 1200 system, we reproducibly quantified over 3,000 proteins in urine samples obtained from LRRK2 knock-out and G2019S rats, as well as from over 1,000 individuals with and without LRRK2 mutations across multiple independent cohorts. Loss of LRRK2 and pathogenic mutations led to significant changes lysosome- glycosphingolipid metabolism in rat urine. To classify individuals as bearing LRRK2 mutations or not, we employed a random forest model that narrowed down the list of 94 significantly regulated rat proteins to the 22 most important features in humans, thereby defining a panel of LRRK2 activity markers. Interestingly, the machine learning model trained using these features excelled in classifying the mutation status in the largest human urine cohort (Parkinson"s Progression Markers Initiative - PPMI[MOU1] : 634 WT and 343 G2019S), achieving an AUC of 0.86 and an average precision of 0.91. Some of the panel proteins, such as ENPEP, FUCA1, and CTSZ, also correlated strongly with disease severity scores such as UPDRS and the Hoehn and Yahr PD progression score, underscoring the multimodality of these proteins in evaluating the severity and stratification of familial PD. Lastly, we further validated the utility of our biomarker panel by demonstrating that the elevated levels of marker proteins in the urine of rats with the LRRK2 G2019S mutation were ameliorated in the urine of rats treated with LRRK2 inhibitors. Together, our study underscores the potential of proteomics for identifying LRRK2 biomarkers and for the non-invasive evaluation of LRRK2 inhibitor efficacy in human urine, thereby aiding in the development of therapeutics for PD. Fig. 1 (optional) 858

859

P-II-0701 Proteomic profiling of progressive multiple sclerosis patient brains reveals regional and cell-type protein differences related to disease processes Henry Wang1, Niall Pollock2, Hajar Mahabadi2, Eman Moussa1, William Branton2, Christopher Power2,3, Olivier Julien1 1University of Alberta, Biochemistry, Edmonton, Canada 2University of Alberta, Medicine (Neurology), Edmonton, Canada 3University of Alberta, Medical Microbiology and Immunology, Edmonton, Canada Introduction: Multiple sclerosis (MS) is a chronic autoimmune disease of the central nervous system (CNS) characterized by the loss of protective myelin surrounding nerve fibers. Patients with progressive MS exhibit gradual declining neurologic function and increasing disability over time, in contrast to relapsing-remitting MS where patients experience defined relapsing attacks of neurologic symptoms. MS pathogenesis is marked by disruption of the blood-brain barrier and inflammatory damage to the brain and spinal cord; however, the responsible molecular mechanisms remain poorly understood. Here, we use mass spectrometry-based proteomics to study region-distinct brain tissue from progressive MS patients to gain insights into disrupted biological pathways at the protein level. Methods: We profiled the proteomes of 8 post-mortem progressive MS patient brains in different regions (normal appearing white matter, white matter lesions, and cerebral cortex) against 8 other disease control patient region-matched brains. We also profiled the proteomes of isolated primary fetal human cultures of CNS cell types (astrocytes, neurons, microglia) and MO3.13 (oligodendrocyte-like) cells. Brain tissue and cell-type lysate samples underwent protein precipitation, digestion, and desalting before being analyzed by liquid chromatography-tandem mass spectrometry (LC-MS/MS) using data-independent acquisition (DIA) mode. Label- free protein identification and quantification were done using Spectronaut software to calculate the statistical significance of protein fold changes associated with MS. Results: We found 518 significantly changed proteins (306 increased, 212 decreased) in the white matter of MS patients and 340 proteins (60 increased, 280 decreased) in the cortex of MS patients compared to controls. Cell- type profiling revealed significantly changed proteins primarily derived from astrocytes and microglia. Gene ontology analysis of these proteins revealed biological processes related to MS, such as the upregulation of proteins in the white matter involved in immune cell migration and the downregulation of mitochondrial proteins in the cortex. Similar trends in protein changes were also observed in our previous proteomic analysis of mouse brains in a demyelinating MS model (cuprizone). Conclusion: We will be performing future work using targeting drugs, recombinant protein delivery, and gene knockouts in MS mouse models to investigate the potential role of these proteins in MS. Altogether, we present datasets that provide insight into possible molecular determinants of MS pathogenesis and potential targets for preclinical drug testing. 859

860

P-II-0702 Dynamic proteomic changes in tumor and immune organs reveal systemic immune response to tumor development Yi Wang1, Zhike Li2, Shuwen Liu2, Jun Qin2 1China Resources Research Institute of Science & Technology, Hong Kong, Hong Kong 2National Center for Proteome Research, Beijing, China In orthotopic mouse tumor models, tumor progression is a complex process, involving interactions among tumor cells, host cell derived stromal cells and immune cells. Much attention has been focused on the tumor and its tumor microenvironment, while the host’s macroenvironment, including immune organs, in response to tumorigenesis is poorly understood. Here, we report a temporal proteomic analysis on a subcutaneous tumor and three immune organs (LN, MLN and spleen) collected on Days 0, 3, 7, 10, 14 and 21 after inoculation of mouse forestomach cancer (MFC) cells in a syngeneic mouse model. Bioinformatics analysis identified key biological processes during distinct tumor development phases, including an initial acute immune response, the attack by the host immune system, followed by the adaptive immune activation and the build-up of extracellular matrix. Proteomic changes in LN and spleen largely recapitulated the dynamics of the immune response in the tumor, consistent with an acute defense response on D3, adaptive immune response on D10 and the immune evasion by D21. In contrast, the immune response in MLN showed a gradual and sustained activation, suggesting a delayed response from a distal immune organ. Combined analyses of tumors and host immune organs allowed the identification of potential therapeutic targets. A proof-of concept experiment demonstrated that significant growth reduction can be achieved by dual inhibition of MEK and DDR2. Together, our temporal proteomic dataset of tumors and immune organs provides a useful resource to understand the interaction between tumors and the immune system, and has the potential for identifying new therapeutic targets for cancer treatment. 860

861

P-II-0703 TMT proteomics reveals distinct cerebrospinal fluid signatures across genetic frontotemporal dementia subtypes Sophia Weiner1, Aitana Sogorb-Esteve2, Joel Simrén1, Kaj Blennow3,1, Henrik Zetterberg3,2,1,4,5,6, Jonathan Rohrer2, Johan Gobom3,1 1University of Gothenburg, Institute of Neuroscience and Physiology, Moelndal, Sweden 2UK Dementia Research Institute at UCL, UCL Queen Square Institute of Neurology, University College London, London, United Kingdom 3Clinical Neurochemistry Lab, Institute of Neuroscience and Physiology, Sahlgrenska University Hospital, Moelndal, Sweden 4UCL Institute of Neurology, Department of Neurodegenerative Disease, London, United Kingdom 5Hong Kong Center for Neurodegenerative Diseases, Hong Kong, Hong Kong 6Wisconsin Alzheimer’s Disease Research Center, University of Wisconsin School of Medicine and Public Health, University of Wisconsin- Madison, Madison, WI, United States Background: Frontotemporal dementia (FTD) encompasses a group of complex disorders characterized by progressive neurodegeneration in the brain"s frontal and temporal lobes. The multifaceted clinical and neuropathological nature of FTD has so far hampered diagnostic and therapeutic advances. Notably, the two major brain pathologies encountered in FTD are protein aggregates composed of either TAR DNA-binding protein 43 (TDP-43) or tau. To date, the scientific community lacks established biomarkers to distinguish between pathologies antemortem as well as markers to predict or track disease progression. Aim: This study set out to explore proteomic differences in the cerebrospinal fluid (CSF) of genetic FTD cases, exhibiting a clear relationship between genetic mutation and underlying pathology, to discern FTD pathology- specific proteomic signatures. We further aimed to identify proteins associated with disease onset and progression. Methods: We analysed 238 CSF samples from C9orf72 (TDP-43 pathology; n=71), GRN (TDP-43 pathology; n=55), and MAPT (tau pathology; n=36) presymptomatic and symptomatic mutation carriers, as well as 76 controls using TMTpro 18-plex proteomics. TMT multiplex samples were fractionated into 24 concatenated fractions by reversed-phase chromatography at basic pH and analysed by LC-MS on an Orbitrap hybrid mass spectrometer (Lumos Tribrid, Thermo Scientific). Protein correlation network analysis (WGCNA) was performed to detect subtype-specific proteomic and pathophysiological alterations. Results: Leveraging WGCNA, we identified several biological processes affected across all symptomatic mutation carriers and pathologies (Figure 1): Proteins associated with neurodegeneration ("Core markers", e.g. YWHAG), actin binding (e.g. MSN), and cellular stress response were increased in symptomatic mutation carriers compared to presymptomatic individuals and controls. These protein clusters strongly correlated with clinical measures of disease severity (FTLD-CDR-SOB, p<0.001) and expected year of symptom onset (EYO, p<0.001), suggesting their potential utility as prognostic biomarkers. Besides common proteomic alterations, we observed that subpopulations of proteins were uniquely associated with each genetic mutation: GPI and HK1, for instance, showed a stepwise increase in abundance from presymptomatic to symptomatic C9orf72 mutation carriers, hinting at early changes in glucose metabolism within this genetic group. Notably, we also found a cluster of proteins with the potential to distinguish between tau and TDP-43 pathology: lysosomal proteins (e.g. DNASE2 and PLBD2) were lower in abundance in symptomatic MAPT mutation carriers (p<0.05) compared with controls, while their levels remained unchanged in GRN and C9orf72 mutation carriers. Conclusion: We discovered shared and distinct proteomic signatures of genetic FTD subtypes that may have prognostic value and distinguish between FTD-pathologies. 861

863

P-II-0704 Proteomic profiling of IDH-wildtype glioblastoma reveals prognostic subtypes with matching abundances of semi-specific peptides in serum Tilman Werner1,2,3, Agnes Schäfer4, Michael Hennes1, Miguel Cosenza-Contreras1,5,2, Guadalupe Espadas6, Eduard Sabidó6, Lena Cook4, Axel Pagenstecher7, Niko Pinter1, Tobias Feilen1,2,3, Christopher Nimsky4, Jörg Walter Bartsch4, Oliver Schilling4,8 1University Medical Center Freiburg, Institute for Surgical Pathology, Freiburg, Germany 2University of Freiburg, Faculty of Biology, Freiburg, Germany 3University of Freiburg, Spemann Graduate School of Biology and Medicine (SGBM), Freiburg, Germany 4Philipps University Marburg, Department of Neurosurgery, Marburg, Germany 5University of Freiburg, MeInBio Graduate School, Freiburg, Germany 6Universitat Pompeu Fabra, Centre for Genomic Regulation (CRG), Barcelona, Spain 7Philipps University Marburg, Department of Neuropathology, Marburg, Germany 8German Cancer Consortium (DKTK) and Cancer Research Center (DKFZ), Heidelberg, Germany Background: Glioblastoma multiforme (GBM) is the most common primary brain cancer, with a dire 5-year survival rate of less than 10%. Despite comprehensive initial treatment, recurrences occur in nine out of ten patients. Key challenges in GBM therapy include the lack of targeted treatment options and diagnostic markers for recurrent tumors. Methods: This study includes a cohort of 55 GBM patients, examining tissue samples from primary tumors and five matching recurrences. Together with each tissue specimen, matching serum samples were collected before and after surgery. All samples were measured using mass-spectrometry-based proteomics in data-independent acquisition mode. Results: In tissue, we identified 6300 proteins on average per sample. Hierarchical clustering of primary tumors revealed four distinct subgroups, independent of previous pathological annotations. The subgroups included: A neuronal cluster marked by elevated mature neuron markers An innate immunity cluster with increased protease expression A mixed cluster with high translational activity A stem-cell cluster with increased stem-cell marker expression The primary clustering drivers were neurodevelopmental and inflammatory processes, with semi-specific peptide levels indicating proteolytic activity intensifying alongside inflammation levels. This pattern was expanded and confirmed through an extra analysis including proteins of lower coverage. Patients in the neuronal cluster showed significantly longer survival compared to those in the stem-cell cluster. In patient-matched differential expression analysis, recurrent tumors exhibited significant changes in protein expression highlighting the proteomic plasticity of recurrent GBM. Serum proteome analysis, utilizing a depletion-based protocol, yielded 730 protein identifications. It revealed stable, patient-specific proteome compositions despite increased inflammation markers post-surgery. This finding was consistent with an independent control cohort of serum from patients undergoing surgery for meningioma (MNG). Several proteins decreased in GBM serum after tumor removal, with some remaining unchanged in MNG serum. Additionally, circulating proteolytic products matched to tissue proteolytic activity, and one fragment of proteolysis-activated receptor 2 (PAR2) consistently decreased after the removal of tumors assigned to the innate immunity cluster. Conclusion: We identified distinct tumor subgroups in a large GBM cohort together with circulating marker candidates. This may enable recurrence risk prediction and personalized therapies in the future. 863

864

P-II-0705 Monitoring disease status and treatment performance in multiple myeloma Charissa Wijnands1, Peter Karel1, Jolein Gloerich2, Vincent Bonifay3, Luciano Di Stefano3, Theo Luider4, Martijn Vanduijn4, Sandra Croockewit5, Elizabeth de Kort5, Daan Castelijn6, Claudia Stege6, Alain van Gool2, Niels van de Donk6, Joannes Jacobs1 1Radboud University Medical Center, Laboratory medicine, Nijmegen, Netherlands 2Radboud University Medical Center, Translational Metabolic Laboratory, Department of Human genetics, Nijmegen, Netherlands 3Sebia, Lisses, France 4Erasmus University Medical Center, Department of Neurology, Rotterdam, Netherlands 5Radboud University Medical Center, Department of Hematology, Nijmegen, Netherlands 6Amsterdam University Medical Center, Department of Hematology, Amsterdam, Netherlands Introduction: Multiple Myeloma (MM) is a common blood cancer caused by malignant plasma cells that overproduce a monoclonal immunoglobulin (M-protein). The M-protein is a personalized biomarker. Since current M-protein quantification methods lack in sensitivity, we have developed a LC-MS/MS method to quantify clonotypic peptides from the variable region of the M-protein (MS-MRD). Therapeutic monoclonal antibodies (t- MAbs) play an important role in MM treatment but often lead to a depletion of total immunoglobulin (Ig) levels which requires intravenous IgG administration. Monitoring t-MAb and polyclonal Ig levels could provide important information for therapy administration. Here, we demonstrate the monitoring of M-proteins, t-MAbs, and Ig constant regions in one LC-MS/MS run providing a holistic view of therapy response kinetics. Methods: Clonotypic peptides for M-proteins and t-MAbs were selected based on linearity and sensitivity in serial dilutions. Ig constant regions of all classes and subclasses were monitored using (sub)class-unique peptides. With a multi-enzyme de novo protein sequencing approach, we obtained the M-protein sequences (M- inSight, Sebia). Patient sera collected between 2013 and 2024 from four MM patients who have received multiple treatment lines were analyzed with MS-MRD. Sera were analyzed using PASEF DDA on a timsTOF pro - PaSER platform (Bruker). Subsequently, patient-specific libraries and dia-PASEF methods were generated for sensitive disease and therapy monitoring. Peak areas of the selected clonotypic, t-MAb, and Ig constant region peptides were filtered from raw dia-PASEF data files and further processed for quantification. Results: Clonotypic peptides were identified in all four patients and serial dilutions demonstrated an assay sensitivity of 1 mg/L, which is 1,000-fold more sensitive compared to routine M-protein evaluation. MS-MRD analyses in patient 1 and 2 showed the lowest M-protein concentrations since diagnosis in 2013 and 2016 with teclistamab (a novel anti-CD3 and -BCMA bispecific antibody) given under compassionate use as the 8th and 6th line of therapy respectively. Monitoring all administered t-mAbs was feasible in all patients and the measured concentrations corresponded to the administration dates. A 1,000-fold lower concentration of teclistamab compared to daratumumab and isatuximab (anti-CD38 t-MAbs) was observed, corresponding to the significantly different dosing of these drugs. MS-MRD revealed rapid and deep depletion of all Ig classes which is a side-effect of teclistamab, followed by a selective increase in IgG due to intravenous IgG administration to counter this side- effect. Conclusion: This case series demonstrated the robustness of the M-protein as personalized biomarker during follow-up periods of up to 11 years. Quantification of the M-protein, t-mAbs, and all Ig (sub)classes within a single mass spectrometry run enables dynamic monitoring of novel therapeutics and therapy response. 864

865

P-II-0706 dia-PASEF proteomic analysis of HNSCC tumor and stroma enriched sections from FFPE samples prepared with laser capture microdissection Matthew Willetts1, Diego Assis1, Aswini Panigrahi2, Radoslav Goldman2, Allison Hunt3, Thomas Conrads3, Pierre- Olivier Schmit4 1Bruker Scientific LLC, Protoemics, Billerica, MA, United States 2Georgetown University, Washington, DC, United States 3Inova Health System, Falls Church, VA, United States 4Bruker France SAS, Paris, France Introduction: Head and neck squamous cell carcinoma (HNSCC), an epithelial cancer is the most common type of head and neck cancer. HNSCC cells first invade the basement membrane of native epithelium, and in >50% cases proceed to lymph node metastasis, which is associated with poor survival. Overall, the response to available treatments has been moderate. The genomic and transcriptomic landscape of HNSCC (The Cancer Genome Atlas) has been defined, but pinpointing the genetic aberrations linked to tumor phenotypes remains elusive. Deep proteome analysis of tumor and matched normal adjacent tissues (NATs) has been performed (Clinical Proteomic Tumor Analysis Consortium). Proteomic comparison of the cancer cells and its neighboring microenvironment may help identify novel targets for early detection, and intervention of HNSCC. Methods: In this study, laser capture microdissection (LCM) was used to collect tumor and stroma enriched sections from formalin-fixed paraffin-embedded (FFPE) tissues. The samples were processed and digested with trypsin. Data independent LC-MS/MS analysis was performed using the timsTOF HT mass spectrometer connected to nanoElute 2 LC system. Each sample was analyzed in triplicate using a 32-minute gradient (500 ng peptide per injection, 40 min total run time), resulting in a throughput of 24 samples per day. The data analysis was performed using the directDIA+ workflow (Spectronaut 18 software) and the Uniprot-Human-reviewed database (20,383 protein entries). Preliminary Data: We have completed analysis of 20 samples, and in total 8,300 protein groups were identified corresponding to ~96,000 precursors. Excellent technical reproducibility was observed between sample runs with median CV of <8% at protein quantitation level. In paired tumor and stroma samples 6,676 proteins were quantified across all samples. All stroma samples clustered together and clearly separated from tumor group by hierarchical analysis. Volcano plot analysis extracted the proteins that are significantly more abundant in stroma vs. tumor enriched samples. GO functional and pathway enrichment analysis of these proteins identified several functional groups relevant to stromal and tumor regions, e.g., relative high abundance of growth factor binding, collagen binding, heparin binding proteins, and ECM structural constituents in the stromal region. Overall, in this study we were able to achieve excellent proteome coverage in LCM FFPE samples. This is especially important for stromal samples where available amount of protein is very low. The methodology allows for comparative deep proteome analysis of tumor and its adjacent microenvironment in a scalable format. 865

866

P-II-0707 Morphological and molecular profiling as a pilar of precision medicine: tailoring treatment in late-stage cancer patients Nicole Woldmar1, Jéssica Guedes1, Leticia Szadai2, András Kriston3, Ede Migh3, Henriett Oskolás1, Roger Appelqvist4, Zsolt Megyesfalvi5,6,7, Peter Horvath3, István B Németh2, Balázs Döme1,5,6,7, Johan Malm1, Jeovanis Gil Valdés1, György Marko-Varga4 1Lund University, Department of Translational Medicine, Malmö, Sweden 2University of Szeged, Department of Dermatology and Allergology, Szeged, Hungary 3Biological Research Centre, Synthetic and Systems Biology Unit, Szeged, Hungary 4Lund University, Department of Biomedical Engineering, Lund, Sweden 5Medical University of Vienna, Department of Thoracic Surgery, Vienna, Austria 6Korányi Institute of Pulmonology, Budapest, Hungary 7National Institute of Oncology-Semmelweis University, Department of Thoracic Surgery, Budapest, Hungary Treatment of metastatic cancer is often a challenge, thus accounting for over 90% of cancer deaths. Main treatment options are only partially effective, leading to resistance and failure, as there are no well-established markers to predict response. Molecular characterization of individual tumor samples can provide information on actionable targets, allowing selection of the most beneficial treatment for the patient. We developed a workflow including morphological and molecular profiling for aiding treatment decision making in late-stage cancer patients. AI-based morphological characterization of the FFPE resected tissues was applied to identify areas of interest for further laser microdissection (LMD) of tumor, stroma components, and different subclones. Molecular profiling of the tissue samples included multi-omics strategies, such as gene mutation panels, transcriptomics, and proteomic analysis. The established workflow was applied to multiple metastatic cancer patients with advanced clinical stages. In all cases, we were able to identify targetable molecular signatures, proposing new treatment directions for each patient. Patient 1 (metastatic melanoma) presented multiple subclones with different molecular signatures, indicating high intra-tumor heterogeneity. Overall, this patient presented low immune cell infiltration and high metabolic activity in both primary and metastatic tumors, suggesting a non-response to the current immunotherapy and possible response to targeted treatment. In Patient 2 (metastatic prostate cancer), we identified high mitochondrial metabolic activity and presence of immune system response. LMD analysis revealed that mitochondria activation was more present in tumor compared to stroma regions, where PI3K/AKT/MTOR signaling pathway was upregulated. Our data suggested the use of mitochondrial complex inhibitors could be effective in combination with targeted treatment. Patient 3 (metastatic melanoma) also revealed multiple subclones in the morphological profile, suggesting high intra-tumor heterogeneity in the primary. However, its molecular profile compared with the metastases did not show high inter-tumor heterogeneity. In general, this patient exhibited a high proliferative profile in all tumors but indicated a higher metabolic activity in the metastases. These results implied that the patient could respond well to inhibitors of key metabolic components. In all case studies, we have proved that high-throughput molecular profiling allied with morphology can be essential elements for implementing precision medicine and selecting the most beneficial treatment for individual patients with advanced metastatic disease. Our data predicted treatment response and aided clinical decision making, prolonging patient survival from the predicted prognosis, and setting the ground for omic analysis inside the clinical surroundings. 866

868

P-II-0708 Leveraging the Orbitrap astral mass spectrometer to enable the analysis of low tissue content sample sources Dylan Xavier1, Phillip J. Robinson1, Peter G. Hains1 1The University of Sydney, ProCan, Children’s Medical Research Institute, Faculty of Medicine and Health, Westmead, Australia Analysis of cancer cohorts can face sample limitations due to insufficient sample tissue. This may be due to tumour specimens collected for pathology stains to assist in disease characterisation during cancer diagnosis. Further, the analysis of tissue samples from Laser Capture Microdissection (LCM) and tissue microarrays (TMA) that can provide a valuable source of proteomics information are limited due to insufficient tissue for proteomics. The sensitivity and speed of current generation mass analysers for proteomics have been a limiting factors in analysing smaller tissue samples. Our aim was to utilise the enhanced sensitivity of the Orbitrap Astral mass spectrometer to determine its ability to analyse smaller tissue samples, with minimal loss in peptide and protein identification while maintaining quantitative accuracy. To test the Orbitrap Astral sensitivity and scalability, HEK-290 cell line digests were initially used at varying sample loads (2 ng-500 ng). While low sample loads can benefit from nanoflow, the extended un-utilised idle time at the start of a run and the challenges associated with nanoflow robustness suggested use of microflow over nanoflow. To balance the robustness of microflow with the sensitivity gains of nanoflow, a low microflow rate of 1.5 μL/min and a 150 μm ID column were used. With a 500 ng peptide load of HEK-293 digest and a 13-minute gradient, DIA analysis identified nearly 9,000 protein groups and over 150,000 peptides. This decreased to 8,500 proteins and 120,000 peptides with a 200 ng load. At minimal loads of 10 ng and 2 ng, more than 4,500 and 2,500 proteins, respectively, were identified. To enhance peptide and protein identification for low on-column load samples, such as LCM and TMA samples, we tested the impact of different accumulation times, DIA window sizes, and precursor mass ranges using HEK- 293 digest at low on-column loads (2ng-10ng). After optimization, there was a 30% increase in the number of identified proteins. This emphasizes the significance of precisely adjusting these parameters, especially when dealing with low quantity samples. The quantitative accuracy of the Orbitrap Astral was then assessed at low loads using a three-species mix of Human, E. coli and Yeast at varying ratios. Loads ranging from 2 -50 ng were compared with a 200 ng load. The instrument consistently maintained good quantitative accuracy across the low loads, while increasing on-column loads demonstrating improved quantitative performance especially for low abundant peptides and proteins. The data demonstrates the effectiveness of the Orbitrap Astral to achieve high sensitivity and accuracy with far smaller sample requirements, expanding sample access to smaller specimens such as those acquired from LCM or TMAs. 868

869

P-II-0709 Proteomic characterization of epithelial ovarian cancer delineates molecular signatures and therapeutic targets in distinct histological subtypes He-Li Xu1 1China Medical University, Shengjing Hospital, ShenYang, China Clear cell carcinoma (CCC), endometrioid carcinoma (EC), and serous carcinoma (SC) are the major histological subtypes of epithelial ovarian cancer (EOC), whose differences in carcinogenesis are still unclear. Here, we undertake comprehensive proteomic profiling of 80 CCC, 79 EC, 80 SC, and 30 control samples. Our analysis reveals the prognostic or diagnostic value of dysregulated proteins and phosphorylation sites in important pathways. Moreover, protein co-expression network not only provides comprehensive view of biological features of each histological subtype, but also indicates potential prognostic biomarkers and progression landmarks. Notably, EOC have strong inter-tumor heterogeneity, with significantly different clinical characteristics, proteomic patterns and signaling pathway disorders in CCC, EC, and SC. Finally, we infer MPP7 protein as potential therapeutic target for SC, whose biological functions are confirmed in SC cells. Our proteomic cohort provides valuable resources for understanding molecular mechanisms and developing treatment strategies of distinct histological subtypes. Fig. 1 (optional) 869

870

P-II-0710 Comparative analysis of EV isolation method for urinary extracellular vesicles proteomics in cancer Lijie Xu1, Berend Gagestein1, Thang V. Pham1, Jaco Knol1, Alex Henneman1, Ka Wan Li2, Irene Bijnsdorp1,3, Connie Jimenez1 1Cancer Center Amsterdam, Department of Medical Oncology, Amsterdam, Netherlands 2VU University of Amsterdam, Department of Neurosciences, Amsterdam, Netherlands 3Amsterdam University Medical Center, Department of Urology, Amsterdam, Netherlands Urinary extracellular vesicles (uEVs) have emerged as a promising source of biomarkers for non-invasive diagnostic applications. To this end, a robust, scalable and reproducible uEV isolation method is required. In recent years several methods were reported, however a comprehensive comparison of uEV isolation methods is missing. In this study, we optimized and compared four EV isolation methods for uEV profiling by mass spectrometry-based proteomics. Furthermore, as a proof of concept, we applied the best performing method to profiling urine from lung cancer patients. Finally, to explore urinary pellet as an alternative sample source for cancer detection, uEV protein profiles were compared to urine fractions. The following EV isolation methods were compared using 15 mL and 2 mL urine: VN96 peptide-based capture, Strong Anion Exchange (SAX) beads, Titanium Dioxide (TiO2), and Size Exclusion Chromatography (SEC). Comparison was performed using Western Blot for EV markers, Coomassie staining and proteomics using a Tims- TOF HT platform as read-out. Our results so far showed that EV markers are best captured using the VN96 method with 2 mL as input. For 15 mL input, VN96-based EV capture and TiO2 method performed equally well. Proteomics results will be presented. Profiling of uEVs isolated using VN96 from cancer patients (n=60; prostate, bladder, colorectal, cervical and lung cancer) and healthy controls (n=13) demonstrated that cancer type-specific biology and proteins can be detected in urinary EV proteomes. Finally, we are currently exploring whether a simpler protocol could provide similar uEV-carried information or more comprehensive data for cancer biomarker discovery. The comparative analysis of uEV, urine pellet, supernatant and full urine will be reported using samples from prostate and kidney cancer patients and healthy donors. In conclusion, our study provides a comprehensive comparison of four uEV isolation methods and their applications to cancer urine, underscoring the potential of uEVs as a rich source for noninvasive protein biomarkers for cancer detection. 870

871

P-II-0711 Proteomics-guided personalized therapy for advanced primary uterus diffuse large B-cell lymphoma: a case report Xizhen Xu1, Dongdong Zhan2, Zitian Huo3, Fang Cheng2, Yi Wang4, Xiaolong Tian5, Jun Qin6, Qian Chu1 1Tongji Hospital, Huazhong University of Science and Technology, Department of Oncology, Wuhan, China 2Beijing Pineal Diagnostics Co., Ltd., Beijing, China 3Tongji Hospital, Huazhong University of Science and Technology, Institute of Pathology, Wuhan, China 4China Resources Research Institute of Science & Technology, Hong Kong, Hong Kong 5Xiangyang Central Hospital, Department of Hematology, Xiangyang, China 6State Key Laboratory of Medical Proteomics, Beijing Proteome Research Center, National Center for Protein Sciences (Beijing), Beijing Institute of Lifeomics, Beijing, China The primary uterine diffuse large B cell lymphoma (DLBCL) is an extremely rare and poorly studied disease. Clinically, the treatment options and management for primary uterine DLBCL cases have not been standardized. Here, we reported the first clinical practice of proteomics-guided precision treatment in a primary uterine DLBCL patient. The case, a 55-year-old female, was pathologically confirmed as non-specific, germinal center B-cell-like phenotype. Oncologists initially recommended an R-CHOP regimen. After LC-MS/MS-based proteomic analysis for the Formalin-fixed paraffin-embedded (FFPE) tissue samples from the patient, the overexpressed JAK1, STAT3, BTK, and NFKB2 were identified and compared with in-house protein expression reference ranges established using 4,500 para-cancer tissues, which indicated activation of the JAK-STAT3 and BCR signaling pathways. These results suggested an increased risk of R-CHOP resistance and a possibility of targeting BTK. Thus, the oncologist optimized the treatment plan by combining R-CHOP and BTK inhibitor (Zanubrutinib). Expectably, after 4 cycles, the patient received a PET-CT evaluation, which revealed no signs of hypermetabolic tumor recurrence and achieved complete remission (CR). By far, the patient has remained stable to date, achieving a total disease-free survival (DFS) time exceeding 12 months. Collectively, this case exemplifies the emerging role of proteomics in guiding oncological treatment, showcasing how precision medicine can enhance therapeutic outcomes in clinical oncology. Fig. 1 (optional) 871

872

P-II-0712 Reproducibility and reliability of protein quantity analyzed by DIA quantitative proteomics for urine biomarker discovery Keiko Yamamoto1, Kengo Yanagita1, Amr Elguoshy1, Tomohiro Uchimoto1, Tadashi Yamamoto1,2 1Niigata University, Biofluid Biomarker Center, Niigata, Japan 2Shinrakuen Hospital, Clinical laboratory, Niigata, Japan Introduction: Currently, quantitative proteomics has improved its accuracy and sensitivity by using highly advanced mass spectrometry and analysis tools, enabling comprehensive quantification of proteomes. In our laboratory urine biomarkers for diseases have been searched by proteomics. In general, quantification of highly abundant proteins is reliable however, biomarker proteins may be trace in urine and quantification of these proteins were not reliable. In addition, plasma proteins are often massive in urine, especially in kidney disease patient urine, making discovery of biomarkers of trace quantities difficult. In this study we investigated the dynamic range of quantities of urinary proteins by recently advanced technologies of quantitative proteomics. Furthermore, we challenged to find biomarkers among the low-abundant proteins. Methods: Proteins were extracted from a mixture of urine from six healthy volunteers (HV) or six diabetic patients (DM) by methanol/chloroform precipitation. After digestion with trypsin, the peptides were purified with C-18 column and 200 ng each were measured 5 times by Bruker's TimsTOFpro with DIA mode. Then quantitative analysis was performed with DIA-NN analysis software. After normalization of peptides, peptide quantity was estimated after multipling by 10^6 (ppm). Then protein amounts were calculated by summing the peptides quantities (ppm), which were assigned to a given protein. The reproducibility of the quantification for each protein across runs was evaluated by a coefficient of variation (CV). In addition, parallel fluctuating changes of quantities of the peptides assigned to a protein also demonstrated the reliability of the quantification of each protein. Results: The quantitative data of approximately 3,000 urinary proteins were obtained with the dynamic range from 10^4 to 10^-1ppm. The most abundant protein in urine was serum albumin; 2.24x10^4ppm in HV and 3.25x10^4ppm in DM. The CV of albumin quantities (ppm) was 4.9％ in HV and 3.1％ in DM. The CVs of top 1,000 highly abundant proteins (10^3~10^5ppm), approximately one-third of all proteins, were less than 20%. Interestingly, more than half of the proteins of below 10 ppm, which accounted for about 20% of the total, had CVs of 20% or less, indicating that the quantitative values of even very trace amounts of proteins are also accurate. For proteins with high CVs, we examined whether the quantitative values of multiple peptides detected as those assigned to a given protein were fluctuated synchronously, and examined the cause of the high CVs. The analysis suggested that one of the reasons of the high CVs may be mismatches of peptides, which were shown by unsynchronized fluctuation of the quantities of a few peptides assigned to a given protein among different urine samples. 872

873

P-II-0713 Urine bank in biofluid biomarker center and its usability for biomarker discovery Tadashi Yamamoto1,2, Kengo Yanagita1, Amr Elguoshy1, Tomohiro Uchimoto1, Keiko Yamamoto1 1Niigata University, Biofluid Biomarker Center, Niigata, Japan 2Shinrakuen Hospital, Niigata, Japan Introduction: Urine is a good biofluid to obtain non-invasively for health check and medical examination. Many trials have been done to find biomarkers in urine by proteomics and to apply the biomarker detection to clinic. For the purpose, we have been collecting urine samples from healthy volunteers and patients since 2013. In this presentation, we will introduce the numbers of samples in our urine bank in Biofluid Biomarker Center (BBC), Niigata University, Japan and how to select urine samples for proteomic analysis. Methods: Urine samples have been collected at several hospitals and facilities according to the guideline for urine collection proposed from HUPO Kidney and Urine Proteome Project Initiative and stored in freezers at BBC http://www.hkupp.org/Urine%20collectiion%20Documents.htm). The number of peoples who donated urine is more than 15.000 and the number of urine samples is more than 160.000, indicating more than 10 urine samples have been collected serially from a person for several years. In this presentation, we focused on urine samples from diabetic patients who might develop kidney impairment detected by microalbuminuria/proteinuria and decrease of the glomerular filtration rate (GFR) during urine collection, Results: In BBC urine bank, 105,592 urine samples have been collected from 8,346 diabetic patients. Among them, 3,353 patients were in good GFR (>60 ml/min) in 2018 and were followed for 5 years to the present. The patients who still keep the good GFR was 2,760 (82.3%) but the others (17.7%) had decline of the GFR in these 5 years. In addition, 2,777 (61.1%) patients out of 3,353 had no microabluminuria or proteinuria in 2018, but 1,081 (38.9%) patients developed microabluminuria or proteinuria. These urine samples may be used to find1by quantitative proteomics. Conclusion: We have collected a world largest number of urine samples from various disease patients in BBC for more than 5 years. These urine collection may be useful for discovery of urine biomarkers to predict or find diseases in early stages. 873

874

P-II-0714 Discovery and validation of urine biomarkers for kidney nephron/site-specific damages by proteomics and immunoassay Kengo Yanagita1, Keiko Yamamoto1, Amr Elguoshy1, Tomohiro Uchimoto1, Chizuko Kitabayashi2, Yoshio Konishi2, Tadashi Yamamoto1,3 1Biofluid Biomarker Center, Niigata University, Niigata, Japan 2Osaka City General Hospital, Nephrology and Hypertension, Osaka, Japan 3Shinrakuen Hospital, Clinical laboratory, Niigata, Japan Introduction: The kidney is a complicated organ constituted with nephrons. The nephron has several parts, which have different functions and are damaged by various injuries. To know the whole kidney damages, kidney biopsy tissues are examined histologically. We aimed to select urine biomarkers for the kidney nephron part damages by proteomics and immunoassay. Methods: Proteins were extracted from urine and nephron tissues, which were laser-micro-dissected from formalin-fixed paraffin-embedded kidney. The extracted proteins were digested into peptides with trypsin and identified using LC-MS in a Data Dependent Acquisition (DDA) manner. Amounts of the proteins selected as kidney nephron/site-unique ones were measured in urine samples obtained from healthy volunteers and patients with various kidney diseases (IgA nephropathy, diabetic kidney diseases, nephrosclerosis, and other chronic kidney disease) by surface plasmon resonance (SPR) using antibodies. Then, quantitative a Data Independent Acquisition (DIA) proteomics was also applied for quantification. Results: By LC-MS, ten proteins were selected as nephron/site-unique ones in the human kidney. SPR assay using antibodies against these proteins demonstrated their up-regulation in urine from kidney disease patients, indicating that they were candidates of urine biomarkers for kidney-site specific injuries. Discussion: We performed DDA analysis of urine proteins from kidney disease patients using LC-MS, and succeeded to select candidate proteins as biomarkers for kidney-site specific injuries. However, DIA quantitative proteomics with recently advanced MS technology was useful to quantify many urinary proteins accurately for biomarker discovery and validation. 874

875

P-II-0715 Unlocking biomarker discovery with novel high-throughput LC-MS workflows for translational research on an Orbitrap Astral mass spectrometer Kevin Yang1, Amarjeet Flora2, Bhavin Patel2, Khatereh Motamedchaboki1, Stephanie Samra1, Amirmansoor Hakimi1 1Thermo Fisher Scientific, San Jose, CA, United States 2Thermo Fisher Scientific, Rockford, IL, United States Cancer claims millions of lives globally every year. Current tests, such as invasive biopsies or costly imaging scans, can lack sensitivity or selectivity and are not readily available. Thus, analyzing blood plasma is widely accepted as a promising technique for biomarker detection. Here, we demonstrate the performance of a high-throughput, in-depth plasma proteomics workflow using different sample preparation technologies coupled with a state-of- the-art Orbitrap Astral mass spectrometer in a small group of diverse cancer and non-cancer samples, which sets the stage for larger-scale cancer cohort studies and translational research. In the mini-cohort study, the plasma samples of various cancers, including melanoma, non-small cell lung cancer (NSCLC), B-cell lymphoma, ovarian, and pancreatic cancer were compared with age- and ethnicity-matched normal plasma. Plasma samples were processed with Seer"s Proteograph XT Assay or Pierce"s Top 14 abundant protein depletion plate for enrichment and depletion, respectively. Different throughput needs were met by separating the resulting peptides using either EASY-Spray PepMap Neo HPLC column or IonOpticks Aurora Frontier UHPLC column on a Vanquish Neo UHPLC system. The separation was coupled to an Orbitrap Astral mass spectrometer operated in DIA mode at a rate of either 16 or 60 samples per day (SPD). Quantitation accuracy was accessed by comparing mixtures of chicken and human plasma at different ratios. Proteomics data was analyzed by Chimerys in Proteome Discoverer software and DIA-NN software. Protein group intensity was used for the unsupervised classification of patients. In total, we were able to identify approximately 1500 protein groups from the depleted plasma samples using the high-throughput 60 SPD method. For enriched plasma samples, we identified approximately 10,000 [ND1] and 7000 protein groups from a throughput of 16 and 60 SPD, respectively. The results demonstrate the unprecedented plasma proteome coverage from Orbitrap Astral mass spectrometer. With in-depth proteome coverage and excellent quantitation, we further conducted principal component analysis. Our results indicate that we were able to classify the samples based on cancer types, highlighting the performance of the workflow in biomarker discovery. Quantitation accuracy was further assessed by comparing the data of human and chicken plasma mixed at different ratios. The quantitative results of the observed ratio accurately reflected the expected ratio, which demonstrates excellent quantitation accuracy. Altogether, our LC-MS workflows on Orbitrap Astral offer extensive coverage along with excellent throughput and quantitation accuracy for proteomics, empowering enhanced classification and biomarker discovery. The mini cancer cohort results demonstrate the advancements in unprecedented depth of coverage and biomarker detection from the Orbitrap Astral mass spectrometer. 875

876

P-II-0717 Discrepancies in proteomics data: autopsy versus fresh frozen samples Xiaoke Yin1, Alicia Beele2, Konstantinos Theofilatos3, Ferheen Baig3, Maria Hasman3, Lukas Schmidt4, Moritz von Scheidt2, Adam Turner5, Clint Miller5, Gerard Pasterkamp6, Stefan Stojkovic4, Johann Wojta4, Michael Joner2, Manuel Mayr1 1Imperial College London, National Heart and Lung Institute, London, United Kingdom 2German Heart Centre Munich, Munich, Germany 3King’s College London, London, United Kingdom 4Medical University of Vienna, Vienna, Austria 5University of Virginia, Charlottesville, VA, United States 6University Medical Center Utrecht, Utrecht, Netherlands Proteomic analyses of human tissues are sometimes conducted on autopsy samples. However, no comparative analysis between proteomic data derived from autopsy samples and fresh frozen samples has been undertaken, nor has there been an assessment of the post-mortem interval (PMI) influences on protein quantification. In the current study, 94 human left anterior descending (LAD) coronary artery were collected from deceased patients. Proteins were analysed using nanoflow liquid chromatography-tandem mass spectrometry. Data were processed with Proteome Discoverer and Mascot. The correlations between the protein abundances and the PMI were calculated. DAVID software was used for pathway and GO annotation enrichment. Among consistently quantified proteins, approximately 40% of the protein abundances exhibited significant correlations with PMI, most of which being inverse. Notably, smooth muscle cell markers displayed substantial reduction with prolonged PMI. Conversely, positive correlations with PMI were observed for immunoglobulins, coagulation factors, and complement factors, including coagulation factor XII, plasminogen, and lactotransferrin. Comparative analyses of sex differences between autopsy LAD samples and fresh-frozen LAD samples (n=65) showed no concordance in protein quantification (Figure A). However, a robust correlation was observed within a sex comparison conducted between fresh-frozen carotid endarterectomies (CEA) from 2 different cohorts (n=104 and n=200) (Figure B). This study represents the first large-scale proteomics investigation into the influence of PMI on the protein composition of human vasculature. We observed significant correlations with PMI for nearly 40% of the consistently quantified proteins. Our findings underscore potential discrepancies in the quantitative accuracy of proteomics data derived from autopsy samples. Consequently, results obtained from post-mortem specimens may not be reproducible in fresh-frozen samples. Fig. 1 (optional) 876

877

P-II-0718 Proteomics identifies circulating TIMP-1 as a prognostic biomarker for diffuse large B-Cell Lymphoma Xiaobo Yu1 1National Center for Protein Sciences (Beijing), Beijing, China Diffuse large B-cell lymphoma (DLBCL) is a heterogeneous disease, although disease stratification using in-depth plasma proteomics has not been performed to date. By measuring more than 1000 proteins in the plasma of 147 DLBCL patients using data-independent acquisition mass spectrometry and antibody array, DLBCL patients were classified into four proteomic subtypes (PS-I-IV). Patients with the PS-IV subtype and worst prognosis had increased levels of proteins involved in inflammation, including a high expression of metalloproteinase inhibitor- 1 (TIMP-1) that was associated with poor survival across two validation cohorts (n = 180). Notably, the combination of TIMP-1 with the international prognostic index (IPI) identified 64.00% to 88.24% of relapsed and 65.00% to 80.49% of deceased patients in the discovery and two validation cohorts, which represents a 24.00% to 41.67% and 20.00% to 31.70% improvement compared to the IPI score alone, respectively. Taken together, we demonstrate that DLBCL heterogeneity is reflected in the plasma proteome and that TIMP-1, together with the IPI, could improve the prognostic stratification of patients. Reference: Lou N, Wang G, Wang Y, Xu M, Zhou Y, Tan Q, Zhong Q, Zhang L, Zhang X, Liu S, Luo R, Wang S, Tang L, Yao J, Zhang Z, Shi Y, Yu X, Han X. Proteomics Identifies Circulating TIMP-1 as a Prognostic Biomarker for Diffuse Large B-Cell Lymphoma. Mol Cell Proteomics. 2023 Sep;22(9):100625 Fig. 1 (optional) 877

878

P-II-0719 Structure-based protein biomarkers for early detection of breast cancer Boris Zaslavsky1 1Cleveland Diagnostics, Cleveland, OH, United States The low sensitivity of immunoassays for quantifying known cancer antigens hinders the discovery of biomarkers for early cancer detection. Highly sensitive R&D immunoassays are not as reliable as FDA-approved ones. We define the diagnostic biomarker as a protein whose structure is altered in the presence of a given pathology, resulting in changes in the nature or 3D arrangement of the solvent-exposed groups or interactions of the protein with other proteins. The discovery of such protein biomarkers with solvent interaction analysis (SIA) technology will be described. One example of this technology's application is the new commercially available isoPSA test. The preliminary results of the technology's application for early detection of breast cancer will be presented. 878

879

P-II-0720 High-throughput peptidome profiling reveals Ischemic stroke-specific proteolysis changes in plasma Andreas Zellner1, Steffen Tiedt2, Chien-Yun Lee1 1Technical University of Munich, Mass Spectrometry in Systems Neurosciences, Freising, Germany 2Ludwig-Maximilians-University Munich, Institute for Stroke and Dementia Research, Munich, Germany large-scale untargeted Introduction: Endogenous peptides play crucial roles in various biological functions and are extensively used in medical diagnostics and therapy. Although targeted analysis of individual peptides has been employed in clinical studies, a identification of endogenous plasma peptides (peptidome) by mass spectrometry is still challenging. The primary obstacle in sample preparation is the predominance of high abundant proteins, which hinders the detection of low-abundance peptide biomarkers. Additionally, the low identification rate of MS2 spectra suggests the presence of non-tryptic or modified peptides that are missed by standard methods. Our goal is to develop a robust, high-throughput workflow to improve the identification of endogenous peptides in plasma and apply it to a large clinical cohort of ischemic stroke patients, where early diagnosis is essential for tailored treatment. Methods: We combined methanol-based precipitation to deplete abundant plasma proteins (e.g., albumin, globulins) with subsequent ultrafiltration for peptide separation (10kDa MWCO) and SCX/C18-based purification on a 96-well plate format. Peptides were analysed by Orbitrap LC-MS/MS and searched using MSFragger (ver. 4.0) with unspecific search parameters. For clinical applicability, we applied this method to plasma samples from a cohort of 24 ischemic stroke patients (collected on Day 1, 3, and 7 post-stroke) and 24 healthy controls, acquiring both proteome and peptidome data. Results: Our optimized workflow enables the identification of over 6,000 endogenous plasma peptides from 100µl sample input, largely expanding the peptidome of previous studies. Many low abundant small proteins and peptides, such as protease inhibitors, peptide hormones, and neuropeptides were more frequently detected compared to standard proteomic analysis. Notably, in the ischemic stroke cohort, we observed a significant up- regulation of various endogenous peptides, while the proteome remained largely unaffected. Many of these peptides were derived from proteins associated with the coagulation cascade, such as Fibrinogen (A and B), Coagulation Factor IX, and Serum Amyloid A1. Mapping the peptidome to the sequence localization of the underlying proteins revealed a preference for specific peptide termini, suggesting stroke-specific cleavage events possibly induced during the acute phase. Conclusions: Our peptidome workflow facilitates the untargeted identification of the largest number of endogenous peptides in plasma to date, suitable for large clinical cohorts in the 96-well format, and easily transferable to other biofluids. Complementary to proteome measurement, peptidome profiling reveals disease- specific proteolysis in clinical samples, serving as a potential biomarker signature. 879

880

P-II-0721 Cross-platform proteomic profiling of plasma extracellular vesicles to identify biomarkers of malignant pulmonary modules Liang Zhang1 1City University of Hong Kong, Biomedical Sciences, Hong Kong, Hong Kong Extracellular vesicles (EVs) are nano-sized lipid bilayer structures released by most cells, and their distinct proteomic contents offer significant potential for disease biomarker discovery. A crucial yet frequently overlooked aspect is the robustness of EV proteomic profiling across different liquid chromatography-mass spectrometry (LC-MS) platforms. Determining key reproducibility parameters among LC-MS platforms/workflows can streamline downstream validation and expedite the clinical translation of EV protein biomarkers. To address this, we evaluated the performance of the high-throughput EvoSep ONE platform and the conventional EASY nLC 1200 platform in profiling the serum EV proteome. Using samples from patients with pulmonary nodules (PNs) patients, we compared the sensitivity and accuracy of these two platforms, as well as their performance in stratifying benign vs. malignant PNs individuals. For qualitative profiling, EASY nLC 1200 had deeper coverage of EV proteins than the high-throughput EvoSep workflow, demonstrating superior sensitivity. Despite this difference, both platforms exhibited robust performance in stratifying malignant and benign PN patients using the label-free quantification (LFQ) abundance of EV proteins. We determined that EV proteins of mid and high LFQ abundance exhibit strong intra- and inter-platform reproducibility. Notably, a common panel of differential EV proteins was identified by both platforms to be significantly upregulated in patients with malignant PNs. The malignant upregulation of ORM1, a high-LFQ differential EV protein, was confirmed through ELISA validation in an independent patient cohort. Conversely, ELISA assays failed to validate the differential expressions of COMP, MVP, and RBP4, which displayed low LFQs in the proteomics data. Collectively, these results indicate that EV proteins of mid-to-high LFQ abundance are more reliable in differential analysis. Our findings underscore the importance of cross-platform validation in proteomic profiling of EVs and suggest a threshold of EV protein LFQ abundance for developing reliable biomarkers. This study provides valuable insights that could guide future research in the field. 880

881

P-II-0722 The molecular insights into the glomerulus injury and prognosis of patients with diabetic kidney disease by integrating proteomics and single-sell RNA sequencing Tingting Zhao1, Weisong Qin1, Dongdong Zhan2, Xuefei Zhao2, Chunxia Zheng1, Zhihong Liu1 1Nanjing University School of Medicine, Jinling Hospital, National Clinical Research Center of Kidney Diseases, Nanjing, China 2National Center for Protein Sciences (Beijing), State Key Laboratory of Proteomics, Beijing Proteome Research Center, Beijing, China Background: As the leading cause of the end-stage renal disease (ESRD), the clinical and pathological trajectories as well as prognosis of diabetic kidney disease (DKD) patients vary between individuals and within a patient over time. Therefore, it is urgently needed to reveal the molecular insights into the glomerular injury and long-term kidney outcomes in DKD patients. Methods: Glomeruli obtained from 50 biopsy-proven DKD patients underwent proteomics profiling and quantitative morphometric analysis. The clinical follow-up data for ten-years were collected. And the single- nucleus RNA sequencing (scRNA-seq) of diabetic nephropathy patients was used for integrative analysis. Results: First, 1,047 glomeruli proteins were differential expression, among which 143, 39 and 18 ones were specifically expressed in podocyte, mesangial cell and endothelial cell. Six podocyte-specific ligands and ten mesangial cell-specific ligands were upregulated, those ligands could exert important function via interacting with corresponding receptors on receiver cells. Meanwhile, Mfuzz analysis showed that pathological progression of DKD glomerular was not gradual and monotonic, but passed through nonlinear and drastic transitions at the pathologic classification of class II and IV. The obviously upregulated proteins in class I-II mainly fall into the categories of metabolize, immunity, inflammation and apoptosis, and the proteins in class III-IV tend to executed functions of extracellular matrix deposition, collagen formation, complement and translation regulation. Furthermore, WGCNA analysis identified 191 proteins significantly associated with glomerular clinical or pathological phenotypes. Specially, seven phenotype-related proteins expressed in podocyte simultaneously associated with eGFR, proteinuria, serum creatinine, mesangial fraction and GBM thickness. Finally, by integrating the long-term kidney outcomes, we found that those glomerular cell-specific ligands and phenotype- related proteins were obviously related with kidney prognosis. Conclusion: we have integrated clinical measurements, structural morphometry, long-term kidney outcomes and proteomics profiling of glomeruli from research kidney biopsies combined with scRNA-seq of diabetic nephropathy patients to explore the cell-specific molecular insights into glomerular injury as well as kidney prognosis. Thus, this study provides potential avenue for identification of urgently needed therapeutic targets and prognostic markers of diabetic kidney disease. 881

882

P-II-0723 Proteomic and gene panel-based profiling of medullary thyroid cancer for prognosis prediction Yan Zhou1, Yingrui Wang1, Xiao Shi2, Jiatong Wang1, Zelin Zang1, Likun Zhang3, Zhiqiang Gui4, Honghan Chen1, Jiaxi Wang5, Hanqing Liu6, Zhihong Wang4, Haixia Guan7, Xiaohong Wu8, Cenkai Shen2, Yi He9, Bo Huang10, Hao Zhang4, Jianbiao Wang11, Yijun Wu6, Chuang Chen5, Yaoting Sun1, Zhiyan Liu3, Yu Wang2, Tiannan Guo1 1Westlake University, Hangzhou, China 2Fudan University Shanghai Cancer Center, Shanghai, China 3Shanghai Sixth People's Hospital Affiliated to Shanghai Jiao Tong University School of Medicine, Shanghai, China 4The First Hospital of China Medical University, Shenyang, China 5Renmin Hospital of Wuhan University, Wuhan, China 6The First Affiliated Hospital, School of Medicine, Zhejiang University, Hangzhou, China 7Guangdong Provincial People's Hospital, Guangzhou, China 8Zhejiang Provincial People's Hospital, Hangzhou, China 9The Second Hospital of Dalian Medical University, Dalian, China 10Cancer Hospital of Dalian University of Technology, Shenyang, China 11The Affiliated Sir Run Run Shaw Hospital, School of Medicine, Zhejiang University, Hangzhou, China Introduction: Thyroid carcinoma ranks 7th globally in cancer incidence, with 821,173 new cases in 2022. Medullary thyroid carcinoma (MTC), a rare neuroendocrine tumor originating from C cells, represents 1~2% of thyroid neoplasms and is relatively aggressive due to metastasis and treatment resistance. Effective stratification and high-risk identification are crucial. Our study aims to identify molecular prognosis biomarkers and AI models from proteomic and gene panel data for personalized postoperative management. Methods: We retrospectively collected 573 MTC FFPE samples (#patient=542) from 10 Chinese hospitals covering different graphical areas, with a discovery cohort from 6 centers (n=455, #patient=424) and an independent test cohort from 4 centers (n=118, #patient=118) (Fig1A). All samples were reviewed by at least two experienced pathologists. Non-MTC samples and samples lacking clinical information and loss to follow-up were removed, and 389 (#patient=358) and 106 (#patient=106) samples remained eligible in the discovery and the test cohort, respectively. A semi-auto sample preparation method was used to process samples to peptides. Proteomic data were acquired using data-independent acquisition (DIA) on a timsTOF Pro. MS/MS spectra were searched against a self-built thyroid-specific library. A 28-gene panel was sequenced, and MTC tumor grade was scored by HE and ki67 staining. Preliminary Data: We quantified 10,092 proteins from the discovery cohort. We applied a nonnegative matrix factorization algorithm to identify MTC proteomic subtypes and extracted 52 proteins from proteins with high CV and differently expressed proteins. Based on the 52 proteins, discovery samples were separated into three subtypes (Fig1B). The structural recurrence significantly differs among the three clusters (P<0.0001). Cluster #2 has a poor prognosis with 70.7% of 5-year recurrence-free survival (RFS) and 14.6% of disease-specific death (DSD), while samples in cluster #3 have a favorable prognosis with 94.6% 5-year RFS and 0.9% DSD. The 52- protein classifier was validated in an independent testing cohort (P=0.019) and 102 MTC samples (P=0.014) from published resources, showing the robustness of our 52-protein classifier in distinguishing patients into different prognosis groups. We further explored the tumor immune microenvironment in the three clusters by analyzing the immune infiltration by CIBERSORTx and conducting multiplex immunofluorescence imaging. To predict the recurrence risk, an AI model trained using proteomic, gene mutation and clinical data achieved AUC=0.99 in the discovery cohort and validated in the test cohort (AUC=0.87). Novel Aspect: This is the largest multi-center multi-omic study of MTC to date. The MTC subtyping classifier and recurrence risk prediction model enable individualized risk stratification and offer resources for drug target exploration. 882

884

Infectious biology insights (P-II-0724 to P-II-0746) P-II-0724 Protective factors in the elderly: comparative iPSC-derived astrocytes proteome in Brazilian centenarians vs critically ill adults with COVID-19 Gilberto B. Domont1, Luis Ariel Espinosa1, Mateus Vidigal2, Mauricio Quiñones-Vega1, Danyllo Felipe de Oliveira2, Amanda Faria Assoni2, Mayana Zatz2, Fabio Cesar Sousa Nogueira1 1Federal University of Rio de Janeiro, Biochemistry, Rio de Janeiro, Brazil 2University of São Paulo, Genetics and Evolutionary Biology, São Paulo, Brazil COVID-19 has posed significant challenges globally, with Brazil being one of the hardest-hit countries due to the emergence of the Gamma.P1 variant of the SARS-CoV-2 virus. Despite the grim statistics of COVID-19-related deaths, there were notable cases of centenarians exhibiting resilience to severe illness caused by the virus. In this study, we aimed to elucidate the proteomic signatures associated with SARS-CoV-2 infection and individual factors that confer protection or susceptibility in human astrocytes derived from iPSC originated from mild infected centenary patients (n=4, CENT) and critically ill adults (n=3, ADU). Our analysis was focused on the Gamma.P1 (GAM) and Wuhan (WU) infection variants but also compared the response to age. Each patient- derived sample in the study has three conditions: cells infected with WU and GAM variant of SARS-CoV-2 virus, or the non-infected cells used as Control (Ctrl) for 21 samples in the cohort. DIA-MS analysis of astrocyte cells identified 6054 proteins and >3000 were quantified for each comparison group. Differential protein expression analysis was conducted to compare infected cells vs Ctrl considering age, virus type, and global infection as variables, revealing 1214 differentially regulated proteins (p<0.05, log |FC|>1). Functional enrichment analysis of CENT-derived astrocytes exhibited robust activation of gene expression regulation and cellular stress response pathways in response to infection, potentially contributing to their resilience against severe COVID-19. Notably, the enrichment in processes related to RNA stability, DNA damage response, and immune activation suggests a coordinated cellular defense mechanism against viral infection. In contrast, ADU-derived astrocytes showed a different response profile to viral infection. While they also exhibited immune activation, there was a prominent enrichment in proteolysis, metabolic processes, and protein complex organization. This indicates a more pronounced metabolic and proteolytic response in adult astrocytes, which may contribute to the pathogenesis of severe COVID-19 in this age group. Furthermore, functional enrichment analysis highlighted specific pathways modulated by SARS-CoV-2 infection. In CENT-derived astrocytes, pathways related to RNA stabilization, p53- mediated signaling, and cellular stress response were significantly enriched, indicating a concerted effort to mitigate the effects of viral stress. In ADU-derived astrocytes, processes associated with proteolysis, macromolecular complex assembly, and immune activation were predominant, reflecting a distinct cellular response profile. Overall, our findings provide valuable insights into the molecular mechanisms underlying COVID-19 pathology and the differential response of astrocytes to infection in mild centenarians vs critically ill adults for two virus strains. Keywords: COVID-19, Wuhan, Gamma.P1, Brazilian centenarians, astrocytes, proteomic, DIA-MS 884

885

P-II-0725 Lactylation-dependent polarization of macrophages and its role in inflammatory diseases Lars Borgards1, Olga Shevchuk1, Bente Siebels2, Hannah Voss1,2, Christoph Krisp3, Stephanie Schaefer-Tautges1, Devon Siemes1, Philippa Spangenberg1, Jenny Dick1, Sibylle von Vietinghoff4, Jessica Schmitz5, Jan Hinrich Bräsen5, Hartmut Schlüter2, Lisa Schwartz6, Florian Wagenlehner6, Daniel Robert Engel1 1University Hospital Essen, Immunodynamics, Essen, Germany 2University Hospital Hamburg Eppendorf, Core Facility Mass Spectrometric Proteomics, Hamburg, Germany 3Bruker Daltonics GmbH & Co. KG, Bremen, Germany 4University Hospital Bonn, Nephrology Section, Medical Clinic 1, Bonn, Germany 5Hannover Medical School, Nephropathology unit, Department of Pathology, Hannover, Germany 6Justus-Liebig-University Giessen, Clinic for Urology, Pediatric Urology and Andrology, Giessen, Germany During inflammation, i.e., in the context of cancer and infection, metabolites serve as key mediators of immunoregulatory processes. For instance, adenosine, itaconate, and lactate were shown to modulate immune cell recruitment and activity. In a preliminary study, kidney tissue from pyelonephritis and control patients was analyzed by label-free LC-MS/MS on a quadrupole-ion trap-orbitrap MS (Orbitrap Fusion, Thermo Fisher) coupled to a nano-UPLC (Dionex Ultimate 3000 UPLC system, Thermo Fisher). Here, metabolic factors, such as lactate dehydrogenase A, were found to be positively correlated with CD163, a marker of macrophage polarization. This potential co-regulation of metabolic and immune processes might be facilitated through lactylation, a post-translational modification affecting protein activity and gene expression patterns. To further study the polarizing effect of lactate, we employed a model of murine bone marrow-derived macrophages (BMDMs). Here, exposure to titrated concentrations of sodium-DL-lactate resulted in elevated levels of lactylation and CD163 expression, as assessed by flow cytometry. Additional LC-MS/MS-based proteomic analysis will allow in-depth characterization of both lactate-dependent macrophage polarization and the cell-specific lactylome. Future measurements will be performed on a timsTOF-HT (Bruker Daltonics) coupled to a nano-Elute- 2 in DIA mode. Moreover, results of in vitro studies, i.e. regarding macrophage lactylation, will be validated in tissue from pyelonephritis and melanoma patients, further elucidating the interrelationship between metabolism and immunity in vivo. These findings will improve our understanding of the molecular factors that shape macrophage differentiation in inflammatory disorders. 885

886

P-II-0726 Toxin exposure and HLA alleles determine serum antibody binding to toxic shock syndrome toxin 1 (TSST-1) of Staphylococcus aureus Silva Holtfreter1, Stefan Weiss2,3, Tanja C. Meyer2, Frieder Schmiedeke1, Clemens Cammann4, Marcus Dörr3,5, Stephan B. Felix3,5, Hans J. Grabe6, Georg Homuth2, Christian Kohler4, Cedric Mahncke4, Stephan Michalik2, Matthias Nauck3,7, Nele Friedrich3,7, Stefanie Samietz8, Henry Völzke3,9, Uwe Völker2,3, Barbara Bröker1 1University Medicine Greifswald, Institute of Immunology, Greifswald, Germany 2University Medicine Greifswald, Interfaculty Institute for Genetics and Functional Genomics, Greifswald, Germany 3German Center for Cardiovascular Research, Greifswald, Germany 4University Medicine Greifswald, Friedrich Loeffler Institute for Medical Microbiology, Greifswald, Germany 5University Medicine Greifswald, Department of Internal Medicine B, Greifswald, Germany 6University Medicine Greifswald, Department of Psychiatry and Psychotherapy, Greifswald, Germany 7University Medicine Greifswald, Institute of Clinical Chemistry and Laboratory Medicine, Greifswald, Germany 8University Medicine Greifswald, Department of Prosthetic Dentistry, Gerodontology and Biomaterials, Greifswald, Germany 9University Medicine Greifswald, Institute for Community Medicine, Greifswald, Germany Toxic shock syndrome (TSS) is an acute life-threatening disease caused by superantigens (SAg) of the pathobiont Staphylococcus aureus, foremost by toxic shock syndrome toxin-1 (TSST-1). While most adults are protected by TSST-1-neutralizing antibodies, up to 15% of the population remains at risk. Are they unable to mount an effective antibody response to TSST-1? To identify determinants of the anti-TSST-1 antibody response, we examined 976 participants of the German population-based epidemiological Study of Health in Pomerania (SHIP- TREND-0). We profiled the anti-S. aureus antibody response, including anti-TSST-1 antibody levels, using immuno-proteomics, analyzed the nasal colonization with TSST-1-encoding S. aureus strains, and performed a genome-wide association analysis of genetic risk factors. TSST-1-specific serum IgG levels varied over a range of 4.2 logs and were elevated by a factor of 12.3 upon nasal colonization with TSST-1-encoding S. aureus. Moreover, the anti-TSST-1 antibody levels were strongly associated with single nucleotide variants in the HLA class II gene loci. HLA typing revealed that HLA-DRB1*03:01 and HLA-DQB1*02:01 were positively, and HLA- DRB1*01:01 as well as HLA-DQB1*05:01 negatively associated with the anti-TSST-1 antibody levels. We also demonstrate that the HLA haplotype can be accurately determined from array-based genotyping data using the newly published Four-digit Multi-ethnic HLA v2 (2022) algorithm. To conclude, both toxin exposure and HLA alleles affect the human antibody response to TSST-1. These findings suggest that immunizing susceptible individuals with a detoxified TSST-1 vaccine could provide protection against TSS in the future, encouraging efforts to develop such a vaccine. Moreover, the array-based HLA prediction promises new insights into the genetics of infectious diseases, autoimmune disorders, immune deficiencies, and inflammatory disorders. 886

887

P-II-0727 A proteomic model to investigate the effect of ClpX deficiency in Staphylococcus aureus in infection processes Larissa Milena Busch1, Hannes Wolfgramm1, Supradipta De2, Christian Hentschker1, Manuela Gesell Salazar1, Celina Hopp1, Meike Kröber1, Gina Wockenfuß1, Alexander Ganske1, Ulrike Mäder1, Sven Hammerschmidt2, Ulf Gerth3, Stephan Michalik1, Alexander Reder1, Kristin Surmann1, Uwe Völker1 1University Medicine Greifswald, Department of Functional Genomics, Greifswald, Germany 2University of Greifswald, Department of Molecular Genetics and Infection Biology, Greifswald, Germany 3University of Greifswald, Department for Microbial Physiology and Molecular Biology, Greifswald, Germany Host-mediated stressors affect bacterial pathogens during infection processes. In response to those stressors, bacteria adapt their gene expression as well as ultimately proteome profile and activity. In Staphylococcus aureus protein homeostasis is largely mediated by the Clp system consisting of the ClpP peptidase, the unfoldases ClpX and ClpC, and the chaperones ClpB and ClpL. The proteases ClpXP and ClpCP are crucial for general and targeted proteolysis, which rely on the unfoldases interacting with specific targets. ClpX is particularly involved in overall regulation of bacterial virulence and fitness and it is the most conserved Clp unfoldase. However, the majority of S. aureus ClpX targets remains elusive. For investigation of the effect of ClpX deficiency the S. aureus mutant strain HG001 ∆clpX::km and the complemented strain HG001 ∆clpX::km pTripleTREP_clpX were constructed using the newly developed plasmid systems pSauSE (allele exchange plasmid) and pTripleTREP (controllable expression plasmid). When tracking infection processes, the S. aureus mutant strain HG001 ∆clpX revealed a less virulent phenotype in a Galleria model compared to the HG001 wild type and drastically low intracellular replication in in-vitro HBE cell culture infection experiments. The effect of ClpX deficiency under infection-relevant conditions was recorded after cultivating HG001 wild type, the HG001 ∆clpX mutant and the complemented strain HG001 ∆clpX pTripleTREP_clpX under iron limitation and microaerobic conditions by mass spectrometry in data independent mode. To address differences in proteome composition caused by the strain background and growth conditions, a normalization strategy based on an external standard was employed. The standard consists of 15N-labeled Bacillus subtilis proteins enabling reliable distinction between standard and sample proteins and providing robust relative quantification of sample proteins. The proteomic profiles revealed specific ClpX- and condition-dependent changes as well as entanglement of those. An iModulon-inspired analysis resulted in 36 independent components, unravelling the different influences. Known ClpX-targets such as Spx and Sle1 were confirmed as robust targets and new targets such as NikA and SAOUHSC_00671 were suggested. Out of the two investigated infection-relevant conditions, iron limitation generally induced the Fur regulon, and under microaerobic condition the Rex regulon was induced. Moreover, ClpX-dependent condition-specific effects, such as a subregulon of the known Fur regulon, which is less responsive to the provoked iron limitation in the ∆clpX mutant, were also observed. With our proteomics data, we provide a global insight into ClpX-dependent adaptation of S. aureus physiology under infection-relevant conditions. 887

888

P-II-0728 The protein-polysaccharide complex Venetin-1 from earthworm coelomic fluid as a potential natural remedy in the fight against yeast infections and human lung cancer Paulina Czaplewska1, Natalia Musiał1, Marc Muller1, Katarzyna Macur1, Magda Rybicka1, Weronika Ścibek- Rejmontowska1, Aleksandra Bogucka2, Marcin Okrój3, Anna Felberg-Miętka3, Joanna Sadowska4, Wioleta Dudzińska5, Anna Lubkowska6, Beata Tokarz-Deptuła7, Marta Fiołka8 1University of Gdańsk, Intercollegiate Faculty of Biotechnology, Gdańsk, Poland 2Medical University of Gdańsk, Department of Physiology, Gdańsk, Poland 3University of Gdansk and Medical University of Gdansk, Department of Cell Biology and Immunology, Gdańsk, Poland 4West Pomeranian University of Technology, Department of Applied Microbiology and Human Nutrition Physiology, Szczecin, Poland 5University of Szczecin, Department of Physiology and Biochemistry, Szczecin, Poland 6Pomeranian Medical University in Szczecin, Department of Functional Diagnostics and Physical, Szczecin, Poland 7University of Szczecin, Institute of Biology, Szczecin, Poland 8Curie-Skłodowska University, Department of Immunobiology, Lublin, Poland The human body functions efficiently thanks to the synchronised work of many vital cells. However, during our lifetime, each of them is exposed to numerous infections and cell changes like cancer. Resistance of human pathogens and cancer cells to commonly used therapies is also increasingly observed. That is why it is essential to constantly search for substances of natural origin, that allow for the strengthening or targeting of modern therapies. Among the many natural compounds with great potential, preparations of earthworm origin attract attention. They have been used in therapy in Asia since ancient times as powders, liquid extracts of dried earthworms or pastes. Currently, research has confirmed antimicrobial, proteolytic, hemolytic, hemagglutinating, antifungal, and antitumor properties of the coelomic fluid (CF) naturally filling the body cavity of the earthworm. The problem is the toxicity of CF to the vertebrate cells. It turns out that the processing of the fluid associated with heating for a short time at 70°C changes the toxic properties of the CF. It becomes non-toxic to normal cells. We characterized the processed CF under the name Venetin 1 in terms of composition and biological activity. Using proteomic techniques, the protein composition of the preparation was characterized, which showed the presence of proteins from the lysenin, lysozyme group and sugars. Venetin 1 exhibits documented anti-fungal and anti-cancer activity. Proteomics studies conducted on Candida albicans cells and cancer cell lines (A549) using the SWATH-MS approach showed the proteome changes in both cases. In the case of Candida albicans cells, Venetin 1 showed effects on the cell wall and on mitochondria, activating the processes of apoptosis and autophagy. The lung cancer line A549 strongly responds to its presence by overexpressing proteins related to the activity of the mitochondrion and the endoplasmic reticulum and decrease of the levels of proteins associated with glycolysis/gluconeogenesis. The reduced expression of several proteins associated with radio- sensitization and chemotherapy of cancer cells was also observed. Studies conducted on mice treated with cyclophosphamide have shown that sequential administration of CPAm and Venetin 1 positively affects lung tissue. Proteomic analysis showed activation of processes related to regeneration, autophagy, and anti-inflammatory processes in the mouse lung tissue. There is also an apparent effect on the pathways related to Rho GTPases and the activation of the complement system. The presented results provide grounds for further research into using Venetin 1 as a preparation in the fight against candida infections and as a potential support in the treatment of lung cancer in humans. Our thanks go to the OPUS19 grant from the National Science Center 2020/37/B/NZ7/00763, UMO-2020/37/B/NZ7/00763. in Poland, the project 888

889

P-II-0729 Virus detection by untargeted proteomics enabled by a spectral library of the human virome built for diagnostics Marica Grossegesse1, Fabian Horn2, Peter Lasch2, Andreas Nitsche1, Jörg Döllinger2,1 1Robert Koch Institute, ZBS1, Berlin, Germany 2Robert Koch Institute, ZBS6, Berlin, Germany Viruses are ubiquitous in the environment. Almost all living organisms are known to host certain viruses, including bacteria, fungi, plants and animals. A small proportion of these viruses can in turn infect humans and cause disease. Viral infections are primarily diagnosed by gene or protein detection of viral components using targeted methods such as PCR and immunoassays. Metagenomics (mNGS), on the other hand, enables the untargeted detection of viral genomes, while there is no well-established comparable proteome-based method. In this study, we aim to introduce proteomics as a complementary approach for the untargeted detection of human-pathogenic viruses. The viral proteomics (vPro) workflow is based on an in-silico derived peptide library covering the human virome in UniProt (331 viruses, 20,386 genomes, 121,977 peptides), which was designed for diagnostic purposes. The reliability of virus identifications is monitored by a confidence score ("vPro score") specifically adapted to proteomics. In combination with high-throughput diaPASEF-based data acquisition, this workflow enables the analysis of up to 60 samples per day with a specificity of > 99.9 % for virus detection, which was determined in an analysis of 221 samples including plasma and swab samples. The sensitivity of this approach for the detection of SARS-CoV-2 in nasopharyngeal swabs was determined to be ct 27 (CI = 95 %). The comparison of mNGS and proteomics for the detection of SARS-CoV-2 in swab samples revealed, that proteomics is already able to provide the same specificity on species-level and quantitative accuracy with respect to qPCR as mNGS at similar or even higher throughput. Currently, the limitation for virus detection using proteomics is the sensitivity especially compared to targeted virus detection methods but already less pronounced compared to mNGS. This study further demonstrates the potential to improve the sensitivity on various stages of the proteomics workflow in the future. Taken together, this study provides the basis to introduce proteomics as an alternative approach for untargeted detection of human viruses. Further improvements of the sensitivity should enable its wider adoption for high- throughput studies. The data analysis workflow of vPro can further be integrated into large-scale proteome studies of biofluids such as human plasma to explain outliers due to acute infections and to determine the prevalence of persistent infections, such as SARS-CoV-2. Fig. 1 (optional) 889

890

P-II-0730 Prevalent and persistent new-onset autoantibodies in mild to severe COVID-19 August F. Jernbom1, Lovisa Skoglund1, Elisa Pin1, Ronald Sjöberg1, Hanna Tegel2, Sophia Hober2, Anna Månberg1, Peter Nilsson1 1KTH Royal Institute of Technology, SciLifeLab, Protein Science, Stockholm, Sweden 2KTH Royal Institute of Technology, AlbaNova, Protein Science, Stockholm, Sweden Autoantibodies have been shown to be implied in COVID-19 but the emerging autoantibody repertoire remains largely unexplored. We investigated the new-onset autoantibody repertoire in 525 healthcare workers and hospitalized COVID-19 patients in five time points over 16 months using proteome-wide and targeted protein and peptide arrays. Our results show that prevalent new-onset autoantibodies against a wide range of antigens emerged following SARS-CoV-2 infection in relation to pre-infectious baseline samples and remained elevated for at least 12 months. We found an increased prevalence of new-onset autoantibodies after severe COVID-19 and demonstrated associations between distinct new-onset autoantibodies and neuropsychiatric symptoms post-COVID-19. Using epitope mapping, we determined the main epitopes of selected new-onset autoantibodies, validated them in independent cohorts of neuro-COVID and pre-pandemic healthy controls, and identified sequence similarities indicative of molecular mimicry between main epitopes and the conserved fusion peptide of the SARS-CoV-2 Spike glycoprotein. Our work describes the complexity and dynamics of the autoantibody repertoire emerging with COVID-19 and supports the need for continued analysis of the new-onset autoantibody repertoire to elucidate the mechanisms of the post-COVID-19 condition. 890

891

P-II-0731 ProteoGenomics characterization reveals a new molecular classification of chronic lymphocytic leukemia: novel insights in personalized therapeutical response Manuel Fuentes1, Pablo Juanes-Velasco1, Carlota Arias-Hidalgo1, Ana Nuno-Soriano1, Angela-Patricia Hernandez1, Almudena Navarro-Bailon1, Miguel Alcoceba1, Marcos Gonzalez1, Javier de las Rivas1 1IBSAL-CIC-IBMCC-CSIC-USAL, Medicine and Proteomics Unit, Salamanca, Spain Cancer heterogeneity at the proteome level may explain differences in therapy response and prognosis beyond the currently established genomic and transcriptomic-based diagnostics. B-cell chronic lymphocytic leukemia (B-CLL) is a blood cancer with highly heterogeneous genomic alterations and altered signaling pathways. The relevance of proteomics for disease classifications remains to be established in clinically heterogeneous cancer entities such as B-CLL. Here, we characterized the proteome, phosphoproteome, tumor microenviroment profiling (TME), immunocheckpoint profiling and differential serological profiles (against 121 auto-antigen & 32 microbial antigens) alongside B-CLL and monoclonal B cell lymphocytosis (MBL Hi) in a discovery cohort (n= 67). From proteome and phosphoproteome dataset, it is revealing 7 subgroups, which 5 of them are correlated with genomic and transcriptomic alterations; meanwhile 2 new groups are only detectable at proteomics level. Classified developed to identify these new groups on tis characteristic proteome in one independent cohort ( n= 197) confirm that these new groups comprised 25% of B-CLL patients. The overall survival rate is highly different in this new group and is independent of the conventional target therapeutic algorithm based on TP53 mutation and IGHV mutation status. Based on multi-omics factor analysis is established a novel molecular classification of B-CLL and proteomics is providing novel insights to improve patient stratification beyond genetic alterations and heterogeneity. 891

892

P-II-0732 Dual proteomics of infected macrophages reveals bacterial and host players involved in the Francisella dissemination Heloise Rytter1, Kevin Roger2, Cerina Chhuon2, Xiongqi Ding1, Mathieu Coureuil1, Anne Jamet1, Thomas Henry3, Alain Charbit1, Ida Chiara Guerrera2 1INSERM UMR-S1151, Institut Necker Enfants Malades, Université Paris Cite, Paris, France 2Necker Proteomics, Université Paris Cite, Structure Fédérative de Recherche Necker, INSERM US24/CNRS UAR3633, Paris, France 3INSERM U1111, CIRI Centre International de Recherche en Infectiologie, Université Lyon, Lyon, France Facultative intracellular bacteria adapt and replicate within host cells, while host cells develop mechanisms to eliminate them. Using a dual proteomic approach, we characterized the intra-macrophage proteome of the facultative intracellular pathogen, Francisella novicida. We combine cell sorting to enrich infected host cells and in depth proteomics analysis to observe bacterial and host players involved in the Francisella intracellular life cycle and dissemination. Macrophages were infected with GFP-expressing bacteria for 10 hours, then sorted using the FACS Sony MA900 and compared to non-infected cells (five biological replicates). Cells were then lysed and digested using S-Trap spin column. Tryptic peptides were analysed on nanoElute HPLC - timsTOF Pro MS (Bruker Daltonics) using PASEF DDA method over the m/z range of 100 to 1700 Th with ion mobility values from 0.75 to 1.25 V·s/cm² (1/k0). The total cycle time was 1.17 seconds with 10 PASEF MS/MS scans per cycle. Ramp time was set to 100 ms, with target intensity and intensity threshold set to 15,000 and 500, respectively. The MS files were processed with the MaxQuant v2.0.1.0. Gene Ontology Over Representation Analysis (GO ORA) was performed on the signifcant up- and downregulated proteins using ClusterProfler R package (v4.6.2). Bacterial multiplication was monitored in the fully automated microscope Incucyte® S3 (Essen BioScience). Over 900 bacterial proteins were identified in infected macrophages after a 10 hour infection, representing more than 50% of Francisella estimated proteome. Biotin biosynthesis-related proteins were found upregulated, emphasizing the role of biotin-associated genes in Francisella replication. Conversely, most of the proteins encoded by the Francisella pathogenicity island were downregulated, supporting the importance of the Francisella Type VI Secretion System during the early phase of infection but not during cytosolic replication. In the host cell, over 300 proteins showed differential expression among the 6200 identified during infection. The most upregulated host protein was cis-aconitate decarboxylase IRG1, known for itaconate production with antimicrobial properties in Francisella. Surprisingly, disrupting IRG1 expression did not impact Francisella's intracellular life cycle, suggesting redundancy with other immune proteins or inclusion in larger complexes. GO ORA highlighted cell-cell contact and actin polymerization as the main deregulated pathways in macrophage. Using flow cytometry and live cell imaging, we demonstrated that cell-to-cell contacts and actin polymerization- dependent processes are essential for dissemination and we illustrated the merocytophagy as one of the mechanisms of dissemination. These findings lay the groundwork for further exploration of merocytophagy and its molecular mechanisms in a variety of host-pathogen models. 892

893

P-II-0733 Proteogenomics identifies conserved and lineage-specific novel small proteins in clinical Mycobacterium tuberculosis reference strains Benjamin Heiniger1, Christian Schori1, Mohammad Arefian2, Amir Banaei-Esfahani3, Martin Schuler1, Sonia Borell4, Daniela Brites4, Ruedi Aebersold3, Sébastien Gagneux4, Ben Collins2, Christian Ahrens1 1Agroscope, Zurich, Switzerland 2Queen's University Belfast, School of Biological Sciences, Belfast, United Kingdom 3ETH Zürich, Department of Biology, Zurich, Switzerland 4University of Basel, Swiss Tropical and Public Health Institute, Basel, Switzerland Objectives: Accurate and comprehensive prediction of all protein coding genes is still an unresolved issue. Genes encoding small proteins are often missed in genome annotations, even in well-studied bacterial model organisms such as Escherichia coli [1] and Pseudomonas aeruginosa [2]. Yet, these small proteins can carry out many important functions [3]. Using proteogenomics, we aim to identify protein coding genes missed by conventional genome annotation methods in six Mycobacterium tuberculosis clinical reference strains from lineages 1 and 2. M. tuberculosis is among the top bacterial infectious diseases world-wide, with the modern lineage 2 being especially virulent. Methods: We de novo assembled complete genomes of six clinical reference strains from long read PacBio data. By hierarchically integrating reference annotations, ab initio gene predictions and a modified six-frame translation that considers alternative start codons, we created large but minimally redundant integrated proteogenomics search databases (iPtgxDB) [4], where ~95% of peptides unambiguously identify one protein [5]. Total cell extracts were analyzed with Parallel Accumulation-Serial Fragmentation (PASEF) mass spectrometry (MS). After strict FDR filtering and validation we prioritized novel small proteins based on functional predictions and conservation. Results: The assembly of complete genomes allowed us to overcome drawbacks of fragmented, short-read based assemblies that can even miss essential genes [2]. Comparative genomics identified a large core genome (94% of genes), reflecting the lack of horizontal gene transfer in Mycobacterium. Proteogenomics allowed us to identify 35 novel proteins which were enriched in proteins shorter than 100 amino acids and had a significantly higher average of pI values than annotated proteins. Interestingly, the novel candidates included strain and lineage specific proteins and three candidates with a predicted function in toxin-antitoxin systems. Conclusion: We successfully applied our proteogenomics framework for prokaroytes, which is available as a public web server (https://iptgxdb.expasy.org) [3], to six clinical reference strains of a major human pathogen [7]. By leveraging state of the art tandem MS we were able to identify novel small proteins with potential roles in pathogenicity without the need for sub-cellular fractionation. References [1]. Storz G, et al. EcoSal Plus 2020, 9:10. [2]. Varadarajan AR et al. NPJ Biofilms & Microbiomes 2020, 6:46. [3]. Storz G, et al. Annu Rev Biochem 2014, 83: 753-777. [4]. Omasits U et al. Genome Res 2017, 27: 2083-2095. [5]. Qeli E & Ahrens CH. Nat Biotechnol 2010, 28:647-650. [7]. Heiniger B et al. (in preparation). 893

895

P-II-0734 The bottleneck of interactome analysis: comparison of different ways to treat coimmunoprecipitated complexes Martin Hubálek1, Aleš Zábranský1, Alena Křenková1, Michal Korecky1, Iva Pichová1 1Institute of Organic Chemistry and Biochemistry, CAS, Proteomics Core Facility, Prague, Czech Republic The interactome analysis relies on robust, reliable, reproducible sample preparation to capture the interacting protein partners. At the same time, the sample should be free of contaminants, the proteins unrelated to the studied protein complex. Therefore, balancing the stringency of conditions is essential to minimize background noise while preserving weak interactions. In this context, we evaluated the effectiveness of various procedures typically employed at the final stage of interactome capture before protein digestion. In the presented work we compared several approaches for processing co-immunoprecipitated protein complexes on affinity magnetic beads. We typically use direct digestion on the beads for this task, but it may lead to the loss of weak interactions due to washes required to remove detergents after the pull-down. Thus, we modified the washing procedure to capture all interactors. In addition, we have modified the conditions for the release of antigen-antibody binding either by competitive binder or by efficient wash conditions to be able to release the protein complexes and not need to deal with antibody peptides. We focused on Transitional endoplasmic reticulum ATPase (VCP) protein complexes with well-characterized protein interactors for the procedure performance testing. This protein is present in the nucleus and cytosol, including many subcellular compartments. Its known interacting partners range from strong binders to weak and transient interactors. Therefore, it represents an optimal model to study the best approach for capturing a wide spectrum of the interaction network. This protein plays an important role also in host-pathogen interaction and is known to be involved in the virus-cell cycle of the Hepatitis B Virus. The sample digestion was performed in several replicates, applied through a traditional LC-MS pipeline ending on Orbitrap Fusion LUMOS, and data were searched, processed, and statistically evaluated. 895

896

P-II-0735 Molecular insights into BCG vaccine efficacy: Proteomic, Phosphoproteomic, and Lipidomic characterization of three clinical strains john Iradukunda1 1University of Cape Town, Integrated Biomedical Sciences, Cape Town, South Africa Tuberculosis (TB) is a communicable disease caused by Mycobacterium tuberculosis (Mtb) that is of global importance, particularly in low-income countries. It is spread through the nasal pharyngeal route and localizes in the alveolar macrophages of the lung, with 5-10% of infected individuals developing active TB disease. To combat this chronic airborne disease, there is an urgent need for better therapeutic prophylaxis and measures that are both effective and affordable. Vaccines are a cost-effective and safe prophylaxis, and Mycobacterium bovis bacillus Calmette-Guerin (M. Bovis BCG) is the only anti-tuberculosis vaccine available. There are 13 BCG strains in use today, but it is unclear why they elicit different immunological responses following vaccination. To better understand the immunological differences between the BCG vaccine strains, this study aims to use proteomics, Phosphoproteomics and Lipidomics to investigate the extent in which three clinical M. Bovis BCG vaccine strains (BCG Danish, BCG Pasteur and BCG Russia) differ from each other. We will analyze the molecular complexity of BCG - mammalian systems through mass spectrometry and correlate the observed changes with the differing immunological responses found in infants vaccinated with the different BCG strains. We hypothesize that the BCG strains will induce distinct immune responses and that these differences can be measured through mass spectrometric techniques. We observe differential protein expression in macrophage cells infected with the different BCG strains; Enriched pathways highlighting various lipid classes primarily driving the different macrophage immune responses. This research has the potential to inform decision-making around which BCG strain to use and to inform future vaccine development efforts. 896

897

P-II-0736 Phospho- and redox proteomic analysis of the A. fumigatus response to hypoxic stress Thomas Krüger1, Lia Ivanova1, Arite Bigalke1, Thorsten Heinekamp1, Michael Cyrulies2, Miriam Agler- Rosenbaum3,2, Axel A. Brakhage3,1, Olaf Kniemeyer1 1Leibniz-HKI, Department of Molecular and Applied Microbiology, Jena, Germany 2Leibniz-HKI, Bio Pilot Plant, Jena, Germany 3Institute of Microbiology / Friedrich Schiller University Jena, Jena, Germany The filamentous fungus Aspergillus fumigatus is an opportunistic human pathogen, which can cause multiple diseases ranging from life-threatening invasive pulmonary aspergillosis in immunocompromised patients to chronic, noninvasive forms of infection and allergies. The ability of the fungus to sense and adapt to low environmental oxygen concentrations is an important virulence trait. The transcription factor SrbA is the central regulator of the fungal hypoxia response. However, little is known about the adaptation to hypoxia on the protein and posttranslational level. Therefore, we performed quantitative proteomics, including phospho- and redox proteomic analyses by comparing fungal mycelium grown under either normoxic (21% O2) or hypoxic (0.2% O2) conditions in an oxygen-controlled fermenter. The goal was to find novel regulatory circuits in hypoxic adaptation. Proteins were extracted from A. fumigatus mycelium and digested in solution with a trypsin/LysC mixture. Subsequently, phosphopeptides were enriched using a TiO2/ZrO2 solid phase extraction protocol. For the identification of oxidative thiol modifications, we applied the thiol trapping OxICAT technique. All samples were analyzed in triplicates by nanoLC-MS/MS (Thermo QExactive HF). The pTmRS algorithm was used to calculate phosphosite probabilities and the Minora algorithm for precursor ion quantification. We quantified in total 4607 proteins, of which 407 proteins and 1391 phosphopeptides showed significantly different abundance upon hypoxia (fold change >2, ratio-adjusted p-value <0.05). In particular proteins involved in mitochondrion organization, amino acid metabolism, and lipid metabolic processes increased in abundance under hypoxia. The phosphoproteomic data indicated further that the mitotic cell cycle and autophagy processes are differentially regulated under hypoxic growth. The OxICAT method revealed 44 cysteine-containing peptides with a decreased level of thiol oxidation and 36 peptides with an increased level of thiol oxidation under hypoxic growth. A cysteine in the osmotic stress regulating mitogen-activated protein kinase Hog1/SakA exhibited a particularly drastic level of thiol oxidation under hypoxic growth. Based on these findings the identified regulatory mechanisms will be investigated further in A. fumigatus on the genetic level. 897

898

P-II-0737 Unveiling disease dynamics: longitudinal plasma protein profiling in a Plasmodium falciparum malaria cohort for enhanced patient stratification and severity assessment Maximilian Julius Lautenbach1,2, Katja Wyss1, Victor Yman1, Fariba Foroogh1, Donya Satarvandi1, Zaynab Mousavian1, Klara Sondén1, Jun Wang3, María Bueno Álvez4, Fredrik Edfors4, Petter Brodin3, Mathias Uhlen4, Christopher Sundling1, Anna Färnert1,2 1Center for Molecular Medicine – Karolinska Institutet, Medicine Solna, Solna, Sweden 2Karolinska University Hospital, Department of Infectious Diseases, Stockholm, Sweden 3Science for Life Laboratory – Karolinska Institutet, Department of Women's and Children's Health, Stockholm, Sweden 4Science for Life Laboratory – KTH Royal Insitutet of Technology, Department of Protein Science, Stockholm, Sweden Introduction: Malaria remains a major global burden with more than 200 million symptomatic cases leading to 600 000 deaths every year. Infection with the malaria-causing parasite Plasmodium falciparum can result in mild or severe forms of the disease, with less severe symptoms associated with previous repeated infections. The pathology itself is partly a result of the immune response to the blood stage of the disease, which triggers a strong inflammatory reaction. Here, we aim to use a systems-level approach to comprehensively describe the plasma protein profiles in clinical malaria and their association with the disease. Methods: Here, we investigated the peripheral immune response of symptomatic patients in a well- characterized cohort of 72 returning travelers treated for P. falciparum malaria at Karolinska University Hospital. Using Proximity Extension Assay (PEA) by Olink, we profiled 1463 proteins in plasma at acute disease and follow- up samples up to one year after treatment without reinfection. To infer the cellular source and possible targets of the proteins, we used a single-cell multi-omics approach on PBMCs to assess the transcriptional profiles in a subset of donors. Finally, the proteomic data enabled a cohort stratification that associated proteomic profiles with disease severity in patients. Results: Clinical malaria caused a strong perturbation of the plasma protein profile, resulting in more than 800 differentially abundant proteins compared with convalescent samples. Many of these proteins are predicted to be secreted to blood, indicating a specific role in the immune response, especially in cytokine–cytokine receptor interaction. Analysis of single-cell profiles revealed that innate immune cells seem to be the main source and target of several highly affected proteins. The plasma proteomic profiles stratified the patients into three groups associated with disease severity as determined by the sequential organ failure assessment (SOFA) score. We propose a protein signature that is distinct for severe malaria compared to other febrile infections. Conclusion: Next-generation plasma profiling combined with transcriptomics on immune cells and extensive clinical data enabled us to investigate the immune response to natural symptomatic malaria to an unprecedented level and identify proteins associated with clinical presentation. Our work confirms, extends, and provides a useful resource for better understanding of the systemic immune response to malaria and connects it to clinical outcomes and other molecular parameters. Importance: Thanks to the development of novel omics technologies, precious patient samples can be profiled for increasing numbers of parameters. This allowed us to study the host response in patients with malaria to an unprecedented level, enabling us to deepen our understanding of the response, identify novel drivers of disease severity, and suggest new therapeutic targets. 898

899

P-II-0738 The RBPome of influenza A virus mRNA reveals a role for TDP-43 in viral replication Maud Dupont1, Thibaut Douché2, Tim Krischuns2, Quentin Giai Gianetto2, David G Courtney3, Joel Perez-Perri4, Mathias W. Hentze4, Stephen Cusack5, Catherine Isel1, Nadia Naffakh1, Mariette Matondo2 1Institut Pasteur, Paris, France 2Institut Pasteur, DBSC, Paris, France 3Wellcome-Wolfson Institute for Experimental Medicine, Queen's University Belfast, Belfast, United Kingdom 4European Molecular Biology Laboratory (EMBL), Heidelberg, Germany 5European Molecular Biology Laboratory (EMBL), Grenoble, France Recent technical advances, particularly in mass spectrometry-based proteomics, have significantly improved our understanding of the RNA-binding protein (RBP) repertoire present within eukaryotic cells, with particular emphasis on RBPs that interact with cellular polyadenylated mRNAs. However, recent studies using the same technologies have distinguished the RBP interactome of viral mRNAs, including SARS-CoV-2, revealing both similarities and differences between the RBP profiles of viral and cellular mRNAs. In our study, we combined recent protein-RNA crosslinking, RNA interactome capture using a sequence-specific antisense RNA probe with recent quantitative proteomics approaches with high-resolution mass spectrometry, and dedicated bioinformatics and statistical analysis to identify the RBP proteins that are associated with the NP mRNA of an influenza A virus (IAV). In our study, we combined recent methods of protein-RNA cross-linking, RNA interactome capture using a sequence-specific antisense RNA probe, and mass spectrometry to identify RBP proteins associated with NP mRNA from an influenza A virus (IAV). These technologies, combined with a robust quantitative proteomic approach and dedicated bioinformatics and statistical tools, allowed us to identify 51 RBP proteins in infected human cells that interact directly with the NP mRNA of an influenza A virus. By combining our proteomic data with loss-of-function experiments, we discovered cellular factors important for productive viral infection, among which the host protein TDP-43 is recruited by the transcribing viral polymerase through an RNA-independent interaction between FluPol and TDP-43. Overall, our data provide the first insights into the composition and function of influenza mRNPs and reveal a critical role for viral transcriptase in the recruitment of cellular RBPs to viral mRNPs. Fig. 1 (optional) 899

900

P-II-0739 Proteomic signatures uncover phenotypic plasticity of susceptible and resistant genotypes by wall remodelers in rice blast Kanika Narula1, Arunima Sinha1, Atreyee Sengupta1, Niranjan Chakraborty1, Subhra Chakraborty1 1National Institute of Plant Genome Research, Lab105, New Delhi, India including, malondialdehyde production, electrolyte Synergistic action of shared and conserved extracellular matrix (ECM) signaling network determines the fate of pathogen attack and triggers innate immunity in plants. Rice blast, caused by Magnaporthe oryzae adversely affects rice productivity. As a major communication interface, ECM senses external perturbation and provides selective biochemical barrier to segregate cellular organization and signaling pathways from external environment Extensive research has shed light on how plant immune components are activated, regulated and function in response to environmental perturbation. However, two key questions remain largely unresolved: (i) how does ECM dynamics govern susceptibility and disease resistance, (ii) what are the components that underpin these phenomena? Rice cultivars were inoculated with Magnaporthe oryzae, scanning electron microscopy and biochemical analysis leakage, was performed. ECM proteins was isolated by mechanical disruption, CaCl2 extraction, followed by purity assessment using transmission electron microscopy and enzyme assays. To develop disease and immune responsive ECM proteome, TMT-10Plex labeling, trypsin digestion and reverse phase fractionation was performed. LC-MS/MS of 12 fractions was carried out using Orbitrap Fusion Tribrid interfaced with Easy-nLC II nanoflow. Data were analyzed in MaxQuant v.1.6.0.1 using concatenated Oryza sativa and Magnaporthe oryzae protein database. ECM cellulose, pectin and lignin was determined. Real-time PCR of putative markers was performed. Protein-protein identify regulatory disease/immune regulatory hubs. To understand ECM regulated genotype-phenotype plasticity in blast disease, we temporally profiled two contrasting rice genotypes in disease- and immune-state. Morpho-histological, biochemical and electron microscopy analyses revealed that increased necrotic lesions accompanied by electrolyte leakage governs disease-state. Wall carbohydrate quantification showed that changes in pectin level was more significant in blast-susceptible (BS) compared to blast-resistant (BR) cultivar. Temporally-resolved quantitative disease- and immune-responsive ECM proteomes identified 308 and 334 proteins, respectively involved in wall remodeling and integrity, signaling and disease/immune response. Pairwise comparisons between time and treatment, mRNA expression, diseasome- and immunome-networks revealed novel blast-related functional modules. We observed accumulation of wall deformation- and homeostasis-related components in BS cultivar, whereas deregulation of pattern-triggered immunity, wall remodeler and calcium signaling components were observed in BR cultivar. Data demonstrated accumulation of α-galactosidase and phosphatase were associated with disease-state, while ROS, induction of Lysin motif proteins, CAZymes and extracellular Ca-receptor protein govern immune-state. interaction network was developed to 900

901

P-II-0740 Transcriptional regulation of the murine cytomegalovirus m152 protein during acute infection Assel Nurbekova1, Ute Distler1, Niels Lemmermann2, Stefan Tenzer1, Sara Becker3 1University Medical Center of the Johannes Gutenberg-University Mainz, Institute of Immunology, Mainz, Germany 2University Hospital Bonn, Institute of Virology, Bonn, Germany 3Fred Hutch Cancer Center, Vaccine and Infectious Disease Division, Seattle, WA, United States In eons of co-evolution, the species-adapted cytomegaloviruses (CMVs) have evolved multiple strategies dedicated to subvert the immune response. Most of the research done so far focused on the viral side, whereas much less is known on how the host counters viral immune evasion and to which extent viral and host proteins interact. The protein m152/gp40 was the first confirmed MHC-I-like immune evasion molecule encoded by mCMV and plays a central role in subverting antiviral immune responses. The transcription of m152 is controlled by an 'IFN regulatory factor element (IRFE)' located in the promoter region of the m152 gene. During acute infection, the IRFE is targeted by suppressive IFN response factors (IRFs) resulting in a delayed transcription of m152. This identified the m152 promotor region as a target for host countermeasures. To identify transcription factors that regulate m152 gene expression, we have generated synthetic biotinylated DNA fragments, containing sections of the viral m152 promotor with potential transcription factor binding sites. The DNA-binding proteins were captured using an adapted pulldown protocol, which includes the best ratio of nuclear extract vs streptavidin-coated magnetic beads and target oligonucleotide, as well as optimized washing and eluting steps to reduce the background and increase the specificity of the pulldown. In brief, the biotinylated DNA was incubated with nuclear lysates of a murine fibroblast cell line (NIH3T3), followed by immobilization on streptavidin-coated magnetic beads. The DNA-binding proteins were eluted by digestion of the DNA using a nuclease. Afterwards, the proteins were digested using a Single-Pot, Solid-Phase-enhanced Sample Preparation (SP3) protocol and analyzed by LC-MS on a Bruker timsTOF Pro 2 mass spectrometer coupled to a nanoElute LC system in diaPASEF mode. Library-free search of raw data using UniProtKB/Swiss-Prot FASTA databases of Mus Musculus and Murid herpesvirus 1 (strain Smith) proteins and LFQ was performed in DIA-NN. Current results from our DNA-protein pulldowns revealed several interferon regulatory factors (IRFs) that bind to the viral promotor sequence of m152 with IRFE region. These IRFs most likely regulate the transcription of the m152 during in the early stages of infection as a host countermeasure against viral infection. In addition, several more proteins were found to be binding to the oligonucleotide sequences, such as different subunits of nuclear factor NF-kappa-B and transcription factors, which may be forming complexes together with IRFs that regulate the transcription of m152.In conclusion, our current investigation has helped to shed light into the factors involved in transcriptional regulation of the viral m152 protein during the early stages of infection. 901

902

P-II-0741 Protein-level evidence of specific IGHV chain usage in immune responses to SARS-CoV-2 using an undepleted proteomic profiling workflow Shaufa Shareef1, Eleanor Matthews1, Matthew E. Daly1, Chris Hughes2, Stephen Fowler1,3, Timothy Felton1,3, Angela Simpson1,3, Lee A. Gethings1,2, E.N.C. Mills1,4 1University of Manchester, School of Biological Sciences, Manchester, United Kingdom 2Waters, Wilmslow, United Kingdom 3Wythenshawe Hospital, Manchester, United Kingdom 4University of Surrey, Guildford, United Kingdom Antibodies play a critical role in directing responses of the adaptive immune system towards infectious organisms. They exhibit great diversity as a consequence of genetic rearrangements and somatic hypermutation in the antibody binding sites which affects their functionality in vivo. Convergent responses have been identified to a number of pathogens using transcriptomics and isolated which identified convergent usage of certain variable heavy chain mutations. However, protein level evidence for such variants can be challenging to obtain. In a pilot study we have compared depleted and undepleted proteomics workflows using less than 10 µL of plasma from two patients with a severe current SARS-CoV-2 infection and two convalescent patients. Proteomics analysis was performed using a rapid 15-minute gradient with a high-sensitivity cyclic ion mobility mass spectrometer. Progenesis QI for proteomics and R 4.3.1 were used to analyse mass spectral data. To validate this workflow, the relative normalized abundance of CRP was compared against hospital measurements and the relative proportion of abundant proteins in the undepleted workflow was compared against expected proportions in plasma. While more low abundance and non-secretory proteins were identified in the depleted workflow, immunoglobulin variable regions could only be consistently identified in the undepleted samples. Of the 306 International ImMunoGeneTics Information System, 62 were identified. This included two proteoforms with neutralising activity in COVID-19 patients identified using RNA-Seq workflows, VH3-53 and VH1-2. VH1-2 was found in all patients, whilst VH3-53 was found in both patients with a severe current infection. This approach has provided protein-level evidence of specific variable heavy chain usage in COVID-19. Further analysis of the ManARTS COVID-19 cohort of 184 patients will provide further insights into the presence and level of expression of particular immunoglobulin isoforms which can be linked to patient phenotypes. immunoglobulin heavy variable region sequences present functional in the 902

903

P-II-0742 Development of proteomics-based assays for detection and prognosis of Malaria and COVID-19 Sanjeeva Srivastava1 1IIT Bombay, Biosciences and Bioengineering, Mumbai, India We used a quantitative proteomics pipeline to systematically investigate plasma host and parasite proteome alterations in falciparum and vivax malaria patients. Proteomic analysis of human plasma and parasite lysates from P. vivax patients identified five parasite proteins—PVX_090265, PVX_083555, PVX_094303, PVX_003545, and PVX_101520—expressed in human plasma and the parasite proteome. Bioinformatics analysis confirmed these candidates as specific for P. vivax. Further validation through targeted proteomics (multiple reaction monitoring; MRM) suggested their potential as biomarkers for next-generation rapid diagnostic kits for P. vivax infection. Comprehensive proteomics analysis of severe falciparum malaria revealed significant modulations in critical physiological pathways, including lipid metabolism, cytokine signaling, complement cascades, platelet degranulation, and coagulation cascades. Key blood markers—Leucine-rich glycoprotein, Ceruloplasmin, Histidine-rich glycoprotein, Zinc-alpha-2 glycoprotein, Alpha-2HS glycoprotein, and Alpha-1 antichymotrypsin— were identified as effective in defining different complications of severe falciparum malaria, such as cerebral syndromes and severe anemia. These findings could lead to the development of clinical tests for monitoring malaria severity and understanding its pathobiology. For COVID-19, we used FTIR and high-resolution MS-based proteomics on swab and plasma samples from over 200 patients, identifying host peptides that correlated with viral load, indicating their potential as prognostic biomarkers. We further employed SARS-CoV-2 Proteome microarray, containing ~5000 peptides representing the viral proteome and mutant peptides, to study IgA and IgG responses in COVID-19 patients. Data analysis identified 204 and 676 peptides for IgA and IgG, respectively, with 6 and 443 peptides showing discriminatory potential based on disease severity. Notably, we discovered 5 (IgA) and 10 (IgG) peptides with mutations from B1.1.7, 501.V2, and P.1 eliciting immunogenic responses, particularly an increased response against the N501Y mutation in Spike and decreased response against K417T and D138Y mutations. These findings present novel IgA and IgG epitopes with disease severity discriminatory potential and altered immune responses against mutant viral epitopes. We are currently developing assays for clinical translation of these findings, aiming for efficient diagnosis and elimination of malaria, as well as improved prognosis for COVID-19. 903

904

P-II-0743 Proteomic and phosphoproteomic analyses to unravel relevant aspects of Bordetella pertussis pathogenesis Kristin Surmann1, Carlos Baroli2, Hanna Schifter1, Martina Debandi2, Marie-Sofie Illenseher1, Mariela Carrica2, Christian Hentschker1, Thomas Sura1, Larissa Milena Busch1, Hannes Wolfgramm1, Alexander Reder1, Yanina Lamberti2, Maria Eugenia Rodriguez2, Uwe Völker1 1University Medicine Greifswald, Department of Functional Genomics, Greifswald, Germany 2Universidad Nacional de La Plata, Facultad de Ciencias Exactas, La Plata, Argentina Bordetella pertussis is the etiological agent of whooping cough (or pertussis), a highly contagious, though vaccine-preventable, respiratory illness that is especially life-threatening to young children and is experiencing a global resurgence. Pertussis resurgence can be attributed to several factors, including waning immunity after vaccination with current vaccines and the emergence of B. pertussis strains with an adapted genome and virulence factor repertoire. Furthermore, intracellular persistence allows B. pertussis to evade the host's immune system, contributing to infection and transmission. Therefore, understanding the mechanisms underlying B. pertussis virulence regulation during intracellular persistence is essential for developing novel preventive strategies, to avoid intracellular persistence upon infection and subsequently, transmission. Previously, we used a proteomics approach to investigate B. pertussis" adaptation to human macrophages. Currently, we are monitoring the proteome adaptation of this pathogen, as well as the host adaptation after bacterial infection, in polarized human bronchial epithelial 16HBE14o- cells. To achieve this, we use a combined approach of fluorescence-based cell sorting and mass spectrometry in data independent mode. These data indicate a niche specific adaptation of B. pertussis, e.g. by different abundance changes of proteins involved in bacterial stress response and of some virulence factors upon internalization. However, the mechanisms of how B. pertussis adapts its proteomic profile to host-provided environmental conditions, remain poorly understood. In bacteria, genetic adaptation is primarily governed by two-component systems (TCS) with the BvgAS system serving as the master regulator of virulence genes in B. pertussis. Besides BvgAS, only two more (PlrRS, RisA/RisK) out of about 20 potential TCS have been functionally characterized in B. pertussis. Recently, we found that BP1092, a histidine kinase putatively belonging to a TCS, is involved in the fine-tuning of virulence factor regulation in B. pertussis. In order to understand the function of until now poorly characterized TCS, we apply phosphoproteomics of B. pertussis under different infection-mimicking conditions. To do so, we cultivate B. pertussis under standard conditions as well as e.g., during iron and oxygen limitation and harvest bacterial samples in the late exponential growth phase. Then, we apply phosphoproteomic protocols adapted to reproducibly detect histidine- and aspartate phosphorylation at TCS components (e.g., Potel et al., Nat Methods, 2018). By doing so, we expect to characterize the activity of yet undescribed TCS. To summarize, we show that B. pertussis is able to adapt to several niches of the human host by specific proteome adaptations and we provide a method to further improve our understanding of the regulation of adaptation processes relevant during pertussis infection. 904

905

P-II-0744 Mapping the global landscape of HIV protein complexes in CD4+ T-Cells and macrophages using Size Exclusion Chromatography Mass Spectrometry (SEC-MS) Manisha Ummadi1,2,3, Michael McGregor1,2,3, Monita Muralidharan1,2,3, Ujjwal Rathore1,2,3, Nevan Krogan1,2,3, Robyn Kaake1,2,3 1University of California San Francisco, Quantitative Biosciences Institute (QBI), San Francisco, CA, United States 2J. David Gladstone Institutes, San Francisco, CA, United States 3University of California San Francisco, HIV Accessory & Regulatory Complexes (HARC) Center, San Francisco, CA, United States Human immunodeficiency virus (HIV) relies on host-virus protein-protein interactions (PPIs) to establish viral replication and latency. Comprehensive analyses of protein complex assembly and composition are crucial to understanding how HIV-infected cells coordinate diverse functions, underscoring the need for techniques to dynamically map PPIs1. Here, we propose size exclusion chromatography mass spectrometry (SEC-MS) as an unbiased method to identify and map endogenous HIV-host PPIs associated with host innate immune responses in primary human immune cells2. Our method provides a streamlined experimental and computational workflow for rapid profiling of protein complex organization in primary human CD4+ T-Cells and macrophages (M0, M1, and M2) using SEC-MS in combination with Data-Independent Acquisition (DIA) and Sequential Windowed Acquisition of All Theoretical Mass Spectra (SWATH-MS)1,3. This pipeline offers a fast, condition-specific methodology to study global protein complex dynamics in the context of HIV infection of primary human immune cells. We also demonstrate that complexing SEC-MS with previously established methods including affinity purification mass spectrometry (AP-MS), crosslinking mass spectrometry (XL-MS), and CRISPRa/n genetic screens provides a powerful multi-pronged approach to characterizing elusive endogenous host-HIV PPIs in CD4+ T-Cells and macrophages4,5,6. Our study highlights the advantages of using SEC-MS, in conjunction with other established proteomics techniques, to dynamically investigate host-virus PPIs involved in HIV infection of primary human innate immune cells, potentially unveiling novel targets for future therapeutic intervention. References 1Hiatt et al. Nat Commun. 2022. 2Fossati et al. Methods Mol Biol. 2021. 3Rathore et al. mBio. 2024. 4Jäger et al. Nature. 2011. 5Kaake et al. Mol Cell Proteomics. 2011. 6Kao et al. Mol Cell Proteomics. 2011. 905

906

P-II-0745 Proteomic and transcriptomic characterization of a new cell culture model to study viral infections Johanna Wallner1,2, Lara Unger3, Tobias Pfeil3, Lukas Peintner4, Rosina Ehmann3, Roman Wölfel3, Jan Muntel1, Hannes Hahne1 1OmicScouts GmbH, Freising, Germany 2Technical University of Munich, School of Natrual Sciences, Chair of Organic Chemistry II, Garching, Germany 3Bundeswehr, Institute of Microbiology, Munich, Germany 4University of Freiburg, Institute of Molecular Medicine and Cell Research, Freiburg, Germany Viral infections are on the rise, prompting researchers to swiftly investigate the underlying molecular mechanisms. This task is challenging due to the reliance on easily accessible cell lines, which often originate from abnormal cancer cells, or the necessity for time-consuming mouse experiments. We aim to address this issue by developing an easy-to-handle model that represents an original lung phenotype and has an increased susceptibility to viral infections. Given that 3D cell culture can reprogram cells to a more physiological phenotype, we employed a 3D cultivation approach using Calu3, an established respiratory lung adenocarcinoma cell line. To identify the optimal 3D cultivation method and assess the molecular changes induced by environmental shifts, we conducted qPCR and transcriptomic analyses, along with in-depth proteomic analysis of cells cultivated in both 2D and 3D environments. Proteomic data was acquired using data-independent acquisition combined with gas-phase fractionation on an LC-MS/MS system (Vanquish Neo, Orbitrap Exploris 480, ThermoFisher). Our data clearly demonstrates that 3D cultivation induces changes in the transcriptome and proteome. However, these changes are not sustained when the cells are transferred back to 2D, where several proteins involved in cell growth and cell cycle regulation are upregulated again. Building on these key findings, we aim to develop an adapted lung cell model to facilitate virus cultivation. Additionally, we are conducting infection experiments with our 3D cell model to further validate its utility. 906

907

P-II-0746 N-terminomics identifies substrates of the secreted Staphylococcus aureus protease Jep missed by classical label-free bottom-up proteomics Hannes Wolfgramm1, Jöran Tebben1, Shruthi S. Peringathara2, Christopher Saade2, Leif Steil1, Stephan Michalik1, Alexander Reder1, Manuela Gesell Salazar1, Liliane M. Fernandes Hartzig2, Patricia Trübe2, Barbara Bröker2, Keenan Lacey3, Victor J. Torres3, Kristin Surmann1, Silva Holtfreter2, Uwe Völker1 1University Medicine Greifswald, Department of Functional Genomics, Greifswald, Germany 2University Medicine Greifswald, Department of Immunology, Greifswald, Germany 3St. Jude Children's Research Hospital, Department of Host-Microbe Interactions, Memphis, TN, United States The identification of protease substrates in complex samples, such as plasma or cell culture supernatants, can be challenging, particularly when the substrate is not degraded but cleaved at specific sites by proteolytic processing. In such cases, classical label-free bottom-up mass spectrometry reaches its limits in identifying proteins of interest, as the abundance of most peptides remains unchanged. However, N-terminomics can overcome this issue by enriching the N-termini prior to mass spectrometric analysis1. With this approach, the full potential of mass spectrometry can be focused on the peptides of particular interest. Moreover, in addition to identifying protease substrates, this method can provide information on protease-specific cleavage motifs. In our study, we focused on the novel serine protease Jep, which is found almost exclusively in mouse-adapted Staphylococcus aureus strains. In contrast to clinical S. aureus isolates, mouse-adapted strains were demonstrated to persistently colonise mice under physiological conditions. Consequently, these strains are crucial for improving current mouse models for investigating S. aureus colonization and infection2, which is necessary for the development of new prophylactic and therapeutic strategies against this notorious hospital pathogen. To understand the role of Jep as a highly host-specific factor is a pivotal step in defining the model. We have shown that the deletion of jep in the mouse-adapted S. aureus strain JSNZ led to hypervirulence in a murine bacteraemia model. However, no substrates were known so far, and a classical label-free bottom-up proteomic approach did not reveal any differences in the proteome pattern comparing the deletion strain with the wild-type strain. Various potential substrate groups were subsequently screened for Jep substrates using N- terminomics. Tested samples include bacterial culture supernatants of JSNZ wild-type and a jep deletion mutant as well as mouse plasma and murine lung epithelial cells incubated with purified Jep protease. Most interestingly, this revealed alterations in the N-termini of a number of secreted bacterial proteins, including known virulence factors such as the subunits of LukAB. Our results suggest that the protease Jep influences virulence rather by targeted proteolytic processing of specific secreted virulence factors than by protein degradation. This example illustrates the power of N-terminomics in the investigation of proteases, to reveal effects that cannot be covered by classical label-free bottom-up proteomic approaches. (1) Kleifeld O, Doucet A, auf dem Keller U, et al. Isotopic labeling of terminal amines in complex samples identifies protein N-termini and protease cleavage products. Nat Biotechnol. 2010. (2) Trübe P, Hertlein T, Mrochen DM, et al. Bringing together what belongs together: Optimizing murine infection models by using mouse-adapted Staphylococcus aureus strains. Int J Med Microbiol. 2019. 907

908

New technology: Non MS-based proteomics (P-III-0747 to P-III-0762) P-III-0747 Development of Auto-2D: a fully automated two-dimensional electrophoresis system, and application for the top-down proteomics to identify proteoforms in tumor samples Norie Araki1, Manasaporn Sukmak1, Akira Sasao1 1Kumamoto University, Faculty of Life Sciences, Kumamoto, Japan With the advancement of mass spectrometry, large-scale bottom-up proteomics has become common. However, it is still extremely difficult to elucidate in detail of proteoforms, i.e., the structural changes caused by the shift of various post-translational modification sites and splicing in biological proteins, using only bottom-up shotgun analysis at the peptide level. The two-dimensional electrophoresis method, which attracted much attention and was widely used in the early years of proteomics, is still the most useful method to easily and visually separate nearly 10,000 proteins from biological tissues, cells, body fluids, etc. However, it is difficult to perform because it requires techniques to separated proteins quantitatively with high reproducibility, and it is time-consuming, skill-consuming, and thus the use of 2D electrophoresis for proteomics has gradually declined. Against this background, we have previously succeeded in developing a fully automated 2D electrophoresis system (Auto2D), which is simple and easy to use with high resolution, sensitivity, and reproducibility in a short time, and proposed to apply it to proteoform analysis using top-down proteomics, which has recently attracted attention. In this presentation, we will show some successful experimental results obtained by using Auto-2D, with tumor samples including cranial nerve tumors such as neurofibroma, malignant peripheral schwannoma, drug-resistant malignant gliomas, glioma stem cells, and treatment-resistant prostate cancer, in order to identify their target proteins and molecules. In addition, the post translational modifications such as phosphorylation, glycosylation, proteolysis, alternative splicing, and other changes in the target molecule proteoforms involved in malignant transformation and drug resistance have been successfully identified. We present how the proteoforms identified with these strategies using Auto2D are useful for the development of drugs and clinical markers, and will discuss future perspectives of the top-down proteomics. Fig. 1 (optional) 908

909

P-III-0748 Revolutionizing multiplexed protein profiling: dual-reporter capabilities of Luminex xMAP® technology Ronald Backer1, Sherry Dunbar1, Stephen Angeloni1 1Luminex Corporation, Applications, Austin, TX, United States Luminex® xMAP® Technology is a flow cytometry-based multiplexing platform that utilizes colored magnetic microspheres (beads) coupled with target-specific molecules such as antigens, antibodies, receptors, enzyme substrates, as well as oligonucleotides. This technology facilitates the simultaneous detection of up to 500 analytes in a single reaction. The xMAP® Technology platform offers a wide range of preconfigured or custom applications, delivering results with sensitivity and reliability comparable to conventional singleplex assays like ELISA, western blot, and qPCR, but with greater efficiency and throughput. Consequently, it has become an invaluable tool in fields such as proteomics, genomics, clinical diagnostics, and drug discovery. Here we introduce the xMAP INTELLIFLEX® System, featuring innovative dual-reporter capabilities, as a novel tool for multiplexed protein profiling applications. This system enables the detection of two parameters per analyte on the same bead in a single well. The dual-reporter functionality has been successfully implemented in antibody isotyping assays to monitor immune responses to specific infections or vaccinations, facilitating the detection of two immunoglobulin (Ig) isotypes for multiple antigens simultaneously. Examples include the concurrent detection of IgG/IgM, IgA/IgM, and IgG/IgA responses to various pathogens, such as multiple Borrelia species (Häring et al. 2023), SARS-Cov-2 (Fhied et al 2022), Epstein-Barr Virus (Schrieber et al. 2023) and Neisseria gonorrhoeae (Stover et al. 2024). In addition to monitoring antibody titers, dual reporter assays can also determine antibody neutralization potential, as demonstrated by the ability of the ACE2 receptor to block antibody binding to S and RBD (Cameron et al. 2022). and between free and pathways, distinguishing The dual-reporter capabilities extend to various assay formats where simultaneous measurements of analyte pairs are critical. Examples include analyzing post-translational modifications, such as phosphorylation in protein signaling in pharmacokinetic/pharmacodynamic (PK/PD) studies. Additionally, dual-reporter assays can effectively characterize bi-specific/fusion proteins, used for profiling of extracellular vesicles, and bioparticles, like viral particles (Sahi et al. 2024), and detect mutant proteins in cell and tissue lysates. Moreover, xMAP® dual-reporter capabilities facilitate protein-protein interaction studies, allowing determination of interaction affinities or execution of competition experiments. Notably, this latter application has been proven sensitive in assessing vaccine-induced antibody blockade of PD-1/PD-L1 interactions (Overholster et al. 2023). drug forms bound In conclusion, Luminex xMAP® Technology, combined with intelligent workflow automation, revolutionizes protein identification and characterization. It elevates proteomics research in your laboratory by maximizing the number of results while minimizing sample input. 909

910

P-III-0749 Massively parallel single-molecule protein quantitation using protein identification by short-epitope mapping (PrISM) Taryn Gillies1, Vivek Badamagunta1, Giovanni Bellesia1, Rob Grothe1, Katie Winters1, Maureen Newman1, Maria Villancio-Wolter1, James Joly1, Sonal Tonapi1, Steven Tan1, Isaac Sprague2, Kathy Lazaruk1, Maryam Jouzi1, Sheri Wilcox1, Torri Rinker1, Jarrett Egertson1, Kara Juneau1, Parag Mallick1 1Nautilus Biotechnology, San Carlos, CA, United States 2Nautilus Biotechnology, Seattle, WA, United States Introduction: Analyzing protein samples at single-molecule resolution holds the promise of achieving improved quantitative dynamic range, reproducibility, and sensitivity. By employing multi-affinity probes to interrogate immobilized, intact proteins, we aim to use Protein Identification by Short-epitope Mapping (PrISM) to analyze >95% of the proteome with broad dynamic range. PrISM has the potential to revolutionize proteomics by delivering highly accurate, large-scale analyses of protein samples. Methods: Proteins were conjugated to DNA nanoparticles and deposited on a high-density patterned flow cell at optically resolvable locations. PrISM utilizes non-traditional affinity reagents called multi-affinity probes, designed with high affinity and low specificity in order to bind short epitopes present in multiple proteins across the proteome. Multi-affinity probes were applied to proteins to generate binding patterns for each single- molecule protein, which are translated to protein identifications and quantities using machine learning. We acquired PrISM data on a human osteosarcoma cell line (U2OS), protein mixtures, and control samples using hundreds of multi-affinity probes. Results: We demonstrate detection and quantification of proteins from the U2OS cell line using PrISM data comprising billions of single-molecule binding measurements per sample. The detected proteins span a wide dynamic range of abundance in the U2OS cell line sample. We compare PrISM protein identification to mass spectrometry analysis of the same cell line. Improvements in instrumentation, reagents, and data processing approaches yield improved quantitative performance defined by sensitivity, accuracy, reproducibility, and dynamic range. An in-depth analysis of binding data shows a high diversity among patterns generated and indicates an opportunity to identify more proteins by adding more multi-affinity probes. Conclusions: Steady improvement in the Nautilus platform driving the PrISM approach has resulted in detection of proteins from a human cell line sample at single-molecule resolution. 910

911

P-III-0751 Single-molecule proteoform analysis of EGFR in cancer cells in response to therapeutic drug treatment Zhengjian Zhang1, James Joly1, Vivek Budamagunta1, Ishwar Kohale2, Daryl Bulloch2, Mathew Rardin2, Andreas Huhmer1, Greg Kapp1, Parag Mallick1 1Nautilus Biotechnology, San Carlos, CA, United States 2Amgen, South San Francisco, CA, United States Alterations in both abundance and activation of the epidermal growth factor receptor (EGFR) have been correlatesd with poor prognosis in numerous cancers. Extensive studies of EGFR signaling have revealed dozens of phosphorylation sites that are believed to govern its role as a potent oncogene. Though numerous phosphorylation sites have been identified, the coordinated action of these sites within heterogeneous mixtures of proteoforms has yet to be elaborated. For example, it is not yet known how triple phosphorylation of EGFR at sites 1045,1086 and 1173 impacts cell behavior differently than triple phosphorylation of EGFR at sites 974, 992 and 1068. Here we describe the development and application of a high throughput single-molecule analysis platform to study EGFR proteoforms in cancer cells in response to ligand stimuli and drug treatments. The goal of the assay is to quantify the abundances, with single-molecule resolution, of a discrete set of proteoforms of EGFR, as defined by a pre-determined set of site-specific post-translational modifications. This measurement is achieved through first immobilizing billions of individual protein molecules onto a patterned flow cell. In parallel, each molecule is iteratively probed with fluorescently labeled affinity reagents that bind to site-specific post-translational modifications. A computational framework, then takes these measurements, accounts for cross-reactivity, false-positive and false-negative binding and infers the abundance of each proteoform. Titration studies have demonstrated that the assay is able to detect proteoforms that are present at less than 0.1% of the total population. To further enhance sensitivity, affinity purification strategies are used to enrich EGFR proteins from TCP 1016 lung cancer cells that have been treated with a monoclonal antibody therapeutic. The assay was first validated using in vitro phosphorylated recombinant EGFR (GST fusion of the intracellular domain, residuals 668-1210) as compared to un-phosphorylated control. Next, the assay was used to examine the impact of drug treatment on EGFR proteoform molecular heterogeneity. Ultimately, this novel assay may enable a deeper understanding of the regulation and dysregulation of EGFR during oncogenesis and in response to therapeutic intervention. 911

912

P-III-0752 Characterization of exosome surface proteins related to colon cancer using proximity barcoding assay to profile single-exosomes Yuxin Li1, Yanling Cai2, Guixue Hou3, Siqi Liu3 1University of Chinese Academy of Sciences, Faculty of Life Sciences, Shenzhen, China 2Shanghai Secretech Co., Ltd., Shanghai, China 3BGI Genomics, Shenzhen, China Colorectal cancer (CRC) is a common disease with high incidence around the world. A large body of evidence reveals exosomes play vital roles in regulating CRC proliferation, metastasis, and prognosis. It is generally acceptable that exosomes are heterogeneous in size, membrane composition, content, and function. Proximity barcoding assay (PBA) is a mass spectrometry-free approach that can resolve exosome surface proteomes at a single exosome level. For the first time, we have initiated a systematic investigation to profile and cluster the CRC-related exosomes in vitro using PBA. The study goals at classifying exosome subtypes secreted from CRC cell lines and clarifying differences in exosome subtypes among cancer cell lines. A total of 9 CRC and 2 gastric cancer (GC) cell lines were cultured and the corresponding exosomes were harvested through differential ultracentrifugation. After assessment with nanoparticle tracking analysis and transmission electron microscopy, the prepared exosomes were loaded onto microtiters followed by adding PBA probes with 262 antibodies. Then the mixture incubated with RCA products was put into a PCR machine to amplify the tag signals. The MGI T7 sequencer was employed for acquiring sequence and readout information. The sequencing data was treated by EVisualizer, and the treated data was further statistically evaluated by GraphPad Prism. The exosomes measured by PBA in individual cancer cell lines varied very much, from 0.2 to 5.2 million. Of 262 exosome biomarkers, only 27 proteins were identified in the exosomes of these cell lines if a cutoff set at S/N >= 30. To inquire about whether a cell line secrets similar exosomes or not, the qualitative and quantitative surface proteins in the exosomes from individual cell lines were clustered under unsupervised mode using Partitioning Around Medoids. Of these CRC cell lines, 5 exhibited simply hierarchical subgroups of exosomes with 2-4 clusters, while the other cell lines had exosome clusters around 20. The next question was whether CRC cell lines shared similar exosome subgroups. There were 5 proteins generally detected in these exosomes, EPCAM (58.7%), CD151 (50.0%), ITGB1 (46.7%), ITGA6 (27.2%) and ITGB4 (23.9%). Thus, the patterns of their abundance distribution could decide the exosome subgroups. For instance, based on the expression level of CD151, the exosomes in CRC cell lines were broadly divided into two categories, while the other four exosome biomarkers were decisive factors to several exosome subgroups. With the same approach, the single exosomes from two GC cell lines were examined as well, indicating that CD151 was consistently identified in these exosomes, however, THBD and EGFR were the key proteins to define the exosome subgroups of GC cell lines. This study, therefore, offers the convincing evidence that the CRC cell lines indeed share some exosome subgroups, leading to a hypothesis of exosome subgroups involved in cancer metastasis and therapy. 912

913

P-III-0753 Quantum-Si's next-generation protein sequencer™ enables protein detection and peptide characterization from biological samples Chengman Lun1, Kendrick Nguyen1, John Kudolo1 1Quantum-SI, Branford, CT, United States Studies of the proteome would greatly benefit from methods that enable direct protein detection and peptide characterization with exceptional sensitivity. In this context, we present the unique capabilities of Next- Generation Protein Sequencing™ on Quantum-Si"s Platinum® instrument, which utilizes single-molecule protein sequencing to achieve detection of unknown proteins and protein variants. Individual peptides are digested from proteins, immobilized on a semiconductor chip, and probed by dye-labeled N-terminal amino acid (NAA) recognizers, followed by subsequent aminopeptidase cleavage to expose each NAA in the peptide for recognition. Through the recording and analysis of fluorescent intensity, lifetime, and binding kinetics of each NAA binding event, we can successfully identify proteins and detect PTMs. To demonstrate the power of protein sequencing, we identified proteins from serum without prior knowledge, identified pathological PTMs, and utilized protein barcodes to screen and characterize proteins. We isolated proteins from human serum via immunoprecipitation or SDS-PAGE and correctly identified them from the sequencing data with high confidence by mapping to an 8,000-protein reference panel. Additionally, we demonstrated the power of Platinum to detect PTMs based on kinetic changes by detecting citrullination and dimethylation of arginine—two PTMs that play key roles in disease states such as cardiovascular disease, autoimmune disease, and cancer. Furthermore, we identified both the wild-type and single-amino-acid variants of both hemoglobin alpha and beta from clinical samples of heterozygous patients. Finally, we showcased the detection of peptide biomarkers along with the utilization of barcoding techniques to streamline protein engineering, antibody engineering, and enzyme engineering. These results demonstrate the transformative potential of single-molecule protein sequencing using Platinum for comprehensive proteome analysis. Our technology enables the detection and characterization of proteins and peptides in complex mixtures and biofluids, as well as the detection of critical PTMs implicated in human pathophysiology. As proteomics continues to advance, Quantum-Si's platform offers a promising approach to drive new insights into protein research and disease mechanisms. 913

914

P-III-0754 Analytical validation of NULISAseq inflammation panel 250 for multiplexed detection of inflammation-related proteins in autoimmune diseases Xiao-Jun Ma1, Qinyu Hao1, Helena Sun1, Vasudha Murlidhar1, Anna Lee1, Ming Yu1, Ishara Ariyapala1, Sean Kim1, Joanne Beer1, Wei Feng1, Yiyuan Yin1, Dwight Kuo1, Lei Fang1, Yuling Luo1 1Alamar Biosciences, Inc., Fremont, CA, United States The plasma proteome holds great significance in tracking the development, progression and therapeutic response of autoimmune diseases. However, current methodologies lack the sensitivity and dynamic range to measure the diverse array and abundance of proteins in blood. NULISA™, Nucleic acid-linked immunosandwich assay, integrates proximity ligation-based specificity with background suppression and immunocomplex purification using a dual capture and release mechanism to achieve attomolar sensitivity and 12 log dynamic range of detection for multiplexed proteomic analysis. Here, we report the analytical validation of fully automated NULISA on ARGO HT™ with the NULISAseq Inflammation Panel 250. Limit of detection and dynamic range were determined for each assay from standard curves generated by serial dilution of pooled recombinant proteins. Intra- and inter-assay precision and reproducibility were evaluated across instruments and with different lots of conjugated antibodies for the inflammation panel. Dilutional linearity of each assay was evaluated using a pooled plasma sample. Assay specificity was evaluated by testing for cross reactivity both within the panel of targeted proteins and to untargeted proteins with high homology to targeted proteins. Assay sensitivity was further assessed for target detectability in plasma. The NULISA Inflammation Panel demonstrated a 12-log dynamic range spanning low abundance targets such as IFNL1 at attomolar concentrations and higher abundance targets like CRP at micromolar concentrations without sample dilution or enrichment. The median intra- and inter-plate CV across the panel were <10%, and the median total CV including different reagent lots was <15%. Dilutional linearity studies demonstrated excellent relative quantitation linearity for 97% of the targets. Cross reactivity studies demonstrated high specificity with <1% cross reactivity for 95% of the targets in the panel. Overall detectability in plasma samples from healthy and diseased donors was over 95%. Overall, the fully automated NULISA on the ARGO HT platform demonstrated high analytical performance and enables both single-plex and multiplexed detection of a wide dynamic range of proteins in blood for discovery research and validation studies. 914

915

P-III-0755 Single-molecule library preparation of a complex cell lysate Julia Robinson1, Brittany Nortman1, Maureen Newman1, Hunter Boyce1,2, Kathy Lazaruk1, Parag Mallick1 1Nautilus Biotechnology, San Carlos, CA, United States 2Boyce Data Insights, Amsterdam, Netherlands Protein Identification by Short-epitope Mapping (PrISM) is a novel, single-molecule based proteome analysis approach that is designed to quantify nearly the entire human proteome in a single experiment. A key attribute of this approach includes the ability to provide comprehensive coverage of sample constituents and a wide dynamic range suitable to quantify high- and low-abundant proteins in the proteome. The first step in PrISM involves the generation of a single-molecule library from the sample, which must be simple-to-execute, efficient, reproducible and comprehensive to consistently provide a representative portion of the sample for subsequent deposition and detection on the billions of landing pads in the flowcell used for analysis. Using both a model-protein mixture and a U2-OS cell line, we prepared a library of each mixture by denaturing the sample using SDS and functionalizing the lysine residues and N-terminal primary amines using an STP methyltetrazine group (mTz), which puts a click reagent onto each of the proteins. To complete the single pot reaction, we reduced and alkylated the disulfide bonds. Cleanup steps removed the reactants, resulting in protein molecules labeled with the click group mTz. These labeled proteins were then conjugated to a proprietary nanoparticle scaffold for deposition onto our hyper dense flowcell. We first investigated the efficiency of labeling model-protein mixture and lysate samples with STP-mTz by applying bottom-up LCMS to the modified samples. As we expect many lysine residues to be modified, samples were digested with both trypsin and arginase-C to maximize sequence coverage. We obtained a broad coverage of the U2-OS proteome, providing a direct read out of the lysine residue and N-terminal primary amine modifications on the lysate. The results indicate that, on average, more than one lysine residue per protein is modified by mTz allowing for the efficient covalent reaction with the complimentary click group TCO on the nanoparticle scaffolds. We did not see any obvious classes of proteins that were unable to be modified. We next investigated the efficiency of conjugation between mTz modified proteins and nanoparticle scaffolds through LCMS analysis of the proteins that had been successfully conjugated to the nanoparticles. Comparing our U2-OS LCMS analysis with previously published U2-OS proteome analysis under similar LCMS conditions indicates that our simple, straightforward library preparation method enables the U2-OS proteome to be deposited as single proteins at high density on the flowcell for subsequent analysis by PrISM. 915

916

P-III-0756 A deep-learning based approach for spot segmentation of indigenous protein microarray images Arka Ray1, Pritam Sil2, Chaitanya Tuckley1, Sanjeeva Srivastava3, Siddhartha Duttagupta4 1Indian Institute of Technology Bombay, Centre for Research in Nanotechnology and Science, Mumbai, India 2Indian Institute of Technology Bombay, Department of Computer Science and Engineering, Mumbai, India 3Indian Institute of Technology Bombay, Department of Biosciences and Bioengineering, Mumbai, India 4Indian Institute of Technology Bombay, Department of Electrical Engineering, Mumbai, India Introduction: Microarray is a high-throughput platform for identifying autoantibodies and humoral immune response for viral diseases like COVID-19. The microarray chip and their scanners are very expensive. Hence, we have developed an indigenous and low-cost array and a fluorescence based microarray instrument. The microarray data extraction for traditional cases like DNA array, protein array or peptide array having circular features (spot) with the help of GenePix Array List (GAL) file is quite time consuming and considered to be a major bottleneck. The goal of this work is to implement a light weight deep learning model for spot segmentation of indigenous protein arrays with circular receptacles. In this work, we have developed an indigenous microarray instrument prototype followed by implementation of a robust algorithm for data extraction. Methodology: For the in-house microarray instrument prototype in our laboratory, we have made our own indigenous microarray chip consisting of fluorophore molecules (Cy3 and Cy5) for testing the working of the instrument. Gradually, we are approaching towards high throughput approaches to check the performance of the instrument. The instrument works based on the fluorescence imaging principle. The components controlling the excitation source and the detector unit is AI-powered. For pixel wise segmentation of different biomedical images researchers used many traditional thresholding approach like OTSU and clustering approach like fuzzy c- means (FCM). But U-Net was found to be more significant for automatic spot segmentation as per the performance metrics compared to other methods. We have taken 236 real protein array images from the database and split the dataset in 60%, 20% and 20% for training, testing and validation purposes. The model has been developed using PyTorch. The batch size and the number of epochs used were 8 and 50 respectively. A default learning rate was set at 0.001. Results: From the performance metric of the test images (shown in Table in the figure attached), it was found U- Net provided a better segmentation accuracy, precision, recall and F1 score. The dice coefficient for U-Net was also found to be higher compared to the other segmentation approaches representing a higher similarity between predicted and the ground truth values. Conclusion: The U-net architecture implemented has been tested on the Indigenous and protein microarray images. This will replace the laborious manual process of aligning grid on the image that tends to be erroneous. References 1. A. Acharjee, A. Ray, A. Salkar et al., Viruses. 15, 248 (2023). 2. A. Ray, S. Srivastava, S. P. Duttagupta, in 2024 IEEE 3rd International Conference on Control, (2024; Communication (CIEC) 3. R. S. Sengar, A. K. Upadhyay, B. Jain, IEEE Transactions on Instrumentation and Measurement. 71, 1–9 Instrumentation, https://ieeexplore.ieee.org/document/10468123), pp. 118–123. Energy & (2022). 916

918

P-III-0758 Integrated sequencing of single transcript and protein molecules discerns alternatively spliced tropomyosin proteoforms Kenneth Skinner1, Natchanon Sittipongpittaya2, Gloria Sheynkman2, Meredith Carpenter1 1Quantum-SI, Branford, CT, United States 2University of Virginia, Charlottesville, VA, United States isoforms and alternatively spliced proteoforms is a key goal Coupling detection of distinct RNA in proteogenomics. For example, actin-binding tropomyosin (TPM1-4) genes generate over 40 spliceoforms in mammals, several of which are functionally distinct. Notably, genetic and transcriptomic studies in human models have associated several TPM spliceoforms with complex diseases such as osteoporosis and coronary artery disease. However, subtle amino acid variations hamper unequivocal identification of TPM proteoforms. Here we integrate LR-RNA seq (PacBio) of transcript isoforms with Platinum (Quantum-Si), a benchtop platform that reveals the primary structure of single protein molecules, to distinguish TPM1/2 (87% amino acid identity). We created a database for long read-guided interpretation of protein isoforms and constructed a map of protein isoforms for splicing events associated with bone mineral density in the human fetal osteoblast (hFOB) cell line. We derived a list of TPM1/2-informative peptides and TPM2 isoform-informative peptides from our long-read transcript database, then utilized Platinum to differentiate synthetic peptides that correspond to TPM1/2 paralogs and spliceoforms. In the Platinum workflow, derivatized peptides are conjugated via C-termini and loaded into nanoscale reaction chambers of a semiconductor chip, with the goal of one peptide per aperture. During sequencing, dye-labeled N- terminal amino acid (NAA) recognizers reversibly bind to cognate NAAs, generating distinct binding and pulsing patterns. Aminopeptidases sequentially cleave NAAs to expose subsequent amino acids for recognition. Upon completion of a sequencing run, data is transmitted for cloud-based analysis. Traces are segmented into contiguous groups of pulses, or recognition segments (RSs), each representing interactions between recognizers and NAAs. The order of recognizer binding and kinetic properties of RSs are aligned to reference sequences to determine peptide sequence. Platinum sequencing discriminates TPM peptides with similar physicochemical properties. NAA recognizers that bind glutamate (E) distinguish SLEAQAEK (TPM1) and SLEAQADK (TPM2). Also, differential pulse durations (PDs) discern isobaric TPM1/2 peptides. At the exon level, NAA recognizers for aromatic residues reveal an isoform-informative peptide in TPM2. Overall, single molecule sequencing with orthogonal methods facilitates proteogenomics and the transition from the discovery to proteoform scoring phase. 918

920

P-III-0761 Development of a software platform for studying the protein expression of microarrays Chaitanya Tuckley1, Arka Ray1, Sanjeeva Srivastava1, Siddhartha Duttagupta1 1Indian Institute of Technology Bombay, Centre for Research in Nanotechnology and Science, Mumbai, India Introduction: Microarrays can house thousands of assays. The multiplexed output from each ligand is used to study the protein expression for a particular disease. The intensity extraction exercise, however, has to be performed manually for each spot. This not only delays the experiment, but is also prone to human errors. This work presents an automated technique to segregate the spot intensities from the background that has been embedded in an application software. Methodology: The user interface also enables the operator to attach a file containing ligand position details. This generates an output file containing the protein names along with their respective intensities and other statistical parameters. The "Project" mode allows the user to load multiple images pertaining to the same assay and label them in subgroups as per the cohort or severity. In this mode, significant proteins or peptides are found and required visualizations are generated. User is given an option to set various parameters in the tasks of intensity extraction, statistical analysis and visualization. The imaging process involves the denoising of the image to remove the features containing abrupt intensity changes. Then the image is segmented into object and background using the local threshold instead of global. Out of the extracted objects, microarray spots are selected based on the criteria of convolution. The row and column details of the spots are identified to match them with the provided file. Statistical process begins with the preprocessing steps like background correction followed by the normalization or standardization techniques. The processed data is subjected to parametric or non-parametric according to the user preference to identify the novel biomarkers. Conclusion: The said technique has not only been tried on protein and peptide microarrays scanned under high- end instrument but it can also extract data from microarray chips imaged under indigenous instrument. Unlike the contemporary methods, the feature pixels are not enclosed inside regular grid cells like circle or rectangle but their boundary contours are extracted ignoring the background area. References 1. Ray, Arka, Sanjeeva Srivastava, and Siddhartha P. Duttagupta. "MicroNET: Microarray Image Orientation Predictor and Automated Data Extraction." International Conference on Control, Instrumentation, Energy & Communication (CIEC), pp. 118-123. IEEE, 2024. IEEE 3rd In 2024 2. Acharjee, Arup, Arka Ray, Akanksha Salkar, Surbhi Bihani, Chaitanya Tuckley, Jayanthi Shastri, Sachee Agrawal, Siddhartha Duttagupta, and Sanjeeva Srivastava. "Humoral immune response profile of COVID-19 reveals severity and variant-specific epitopes: Lessons from SARS-CoV-2 peptide microarray." Viruses 15, no. 1 (2023): 248. 920

922

P-III-0762 Autoantibody reactome profiling of health and disease using AAgAtlas 1.0 microarray platform Bomiao Yu1, Zuo Wang1,2, Xiaomei Zhang1, Ran Zhang1, Meng Xu1, Di Hu2, Yan Jia2, Xiaobo Yu1 1National center for protein science-Beijing (PHOENIX Center), Beijing, China 2ProteomicsEra Medical Ltd. Co., Beijing, China Each individual possesses a unique repertoire of autoantibodies, termed the "autoantibody reactome," which exerts a range of biological effects, leading to both pathogenic and protective clinical outcomes. This diversity may serve as a valuable source for monitoring disease-specific pathology, facilitating earlier diagnosis and risk stratification [1-2]. However, the physiological significance of autoantibodies may have been underappreciated due to the absence of unbiased, high-throughput, and quantitative detection methods that could be leveraged among surveys at a proteome-wide scale. Here, we introduce an AAgAtlas 1.0 microarray that enables the profiling of autoantibody reactome in healthy individuals and patients with different diseases [3-6]. First, we developed a method to quantify autoantibody binding on microarray surfaces, enhancing sensitivity and accuracy over conventional methods. It boasted a linear dynamic range spanning three to four orders of magnitude, with the lowest limit of detection (LOD) of 68 pg/mL and an average coefficient of variation (CV) for quantification results of less than 5% intra-assay and 10% inter-assay. Second, we compared our platform with Chemiluminescence Immunoassay (CLIA) in 267 plasma samples. The classification accuracy achieved an average of over 95%, with a correlation coefficient of 0.94 with clinical quantitative data. Afterward, we designed and prepared an AAgAtlas 1.0 microarray according to the AAgAtlas 1.0 database [6], which was employed to screen ~1,000 serums of healthy controls and patients with Hepatitis B virus-infected hepatitis (CHB), liver cirrhosis (LC), and hepatocellular carcinoma (HCC). Machine learning algorithms were used to generate classification models to identify the signatures that could distinguish well between healthy and patient groups. This study underscores the potential of humoral immunity in detecting health and disease. Reference J Proteome Res. 2023 Jun 2;22(6):1800-1815. 1. Science. 2024 Feb 16;383(6684):705-707. 2. 3. Proteomics Clin Appl. 2022 Nov;16(6):e2100132. 4. Cancer Immunol Immunother. 2022 Jul;71(7):1681-1691. 5. Theranostics. 2020 May 16;10(14):6399-6410. 6. Nucleic Acids Res. 2017 Jan 4;45(D1):D769-D776. 922

923

Data integration: With bioinformatics to biological knowledge (P-III-0763 to P-III-0852) P-III-0763 Overcoming notorious challenges in prokaryotic genome annotation: identifying missed small proteins with proteogenomics and Ribo-seq and consolidating genome annotations Benjamin Heiniger1, Ulrich Omasits1, Lydia Hadjeras2, Christian Schori1, Jakob Meier-Credo3, Elena Evguenieva- Hackenberg4, Cynthia Sharma2, Julian D. Langer3, Wolfgang Hess5, Christian Ahrens1 1Agroscope & SIB Swiss Institute of Bioinformatics, Zürich, Switzerland, Zurich, Switzerland 2University of Würzburg, Institute of Microbiology and Infection Medicine (IMIB), Wuerzburg, Germany 3Proteomics, Max Planck Institute of Biophysics, Frankfurt, Germany 4Institute of Microbiology and Molecular Biology, University of Giessen, Giessen, Germany 5Genetics and Experimental Bioinformatics, Faculty of Biology, University of Freiburg, Freiburg, Germany Complete prokaryotic genome sequences represent an optimal basis for the comprehensive annotation of all protein-coding genes. Yet, in stark contrast to the bounty of genome data, our ability to accurately predict all protein-coding genes is still severely limited [1]. This is best illustrated by the growing evidence that so far unannotated, small ORF encoded proteins (SEPs) carry out numerous key biological functions [2,3], including roles in cell-cell communication and as peptide antibiotics, shaping microbiome composition. The genome annotation problem covers i) the large differences of the number of coding sequences (and their precise start sites) predicted by different annotation centers for an identical genome sequence, ii) the severe under-representation of SEPs, and iii) the continuously changing annotation over time, which is difficult to track. To address the first two issues, we developed a generic proteogenomics approach that can integrate and consolidate different reference annotations, ab initio gene predictions and in silico ORF predictions in a large but minimally redundant integrated proteogenomics (iPtgxDB) search database, that covers the entire protein coding potential of a prokaryotic genome [1]. Such iPtgxDBs have been successfully applied in numerous projects to uncover proteomic evidence for novel SEPs. Today, I will illustrate the succesful application of our solution, which is available as a public web server (https://iptgxdb.expasy.org), to some highly relevant use cases, e.g. to find novel SEPs not only in single bacteria but also in synthetic model communities [4] and clinical strains. By integrating data from top-down [5] and PASEF mass spectromety approaches, we can increase the number of novel SEPs identified. Importantly, we show that incroporating top Ribo-seq candidates can help to minimize the search database size and thereby improve detection of known and novel SEPs [6]. Finally, we adapt the basis of our proteogenomics approach to address the much more common issue -and unmet need of the research community- to integrate and consolidate differing genome annotations in the context of experimental data. Such solutions promise to enable research communities that work with different model organisms to unlock the full potential of functional genomics technologies to captialize on the value of genome sequence data. References [1]. Omasits U, et al. Genome Res 2017, 27: 2083-2095. [2]. Storz G, et al. Annu Rev Biochem 2014, 83: 753-777. [3]. Orr MW, et al. Nuclei Acids Res 2020, 48 :1029-1042. [4]. Petruschke H, et al. Microbiome 2021, 9:55. [5]. Meier-Credo J, et al. Anal Chem 2023, 95:11892-11900. [6]. Hadjeras L, et al. Microlife 2023, 4:uqad012. 923

924

P-III-0764 Evaluating Phospho-peptides quantification across various acquisition methods and bioinformatic tools: novel standards and their utility in algorithm development Pinar Altiner1, Anne Gonzalez de Peredo1, Carine Froment1, Veit Schwämmle2, Odile Schiltz1, David Bouyssié1 1Institut de Pharmacologie et de Biologie StruUniversité Toulouse III – Paul Sabatier (UT3), Toulouse, France 2University of Southern Denmark, Department of Biochemistry and Molecular Biology, Odense, Denmark Phospho-proteomic analysis seeks to identify, localize, and quantify phospho-peptides purified from complex protein samples. Despite advances localization of phosphorylation site(s) for a given phospho-peptide remains challenging due to the ambiguity in experimental fragmentation patterns within MS/MS spectra. Evaluating existing localization algorithms is crucial to assess their relative sensitivity and specificity. Numerous studies were published to that aim, but they solely focused on localization accuracy at the identification level. However, the precise evaluation of the relationship between localization and label-free quantification accuracy remains still unclear to our knowledge. in analytical techniques, determining the exact Thus, we performed several experiments designed for a quantitative evaluation across various acquisition methods and bioinformatic tools. We used either a library of 180 synthetic phospho-peptides of known phosphosite localization, or phospho-peptides enriched from mouse T-cells. Both phosphorylated standards were spiked at various concentrations into an E. coli background. Those samples were analyzed using different MS instruments (Thermo Exploris, Bruker timsTOF) and various acquisition methods (Data Dependent Acquisition (DDA), Data Independent Acquisition (DIA)), with/out ion mobility separation. Several bioinformatic tools were used for the post-processing analysis: Proline, MaxQuant, ProteomeDiscoverer (PD), DIA-NN, and Spectronaut. As we hypothesized, we could show that label-free quantification errors are increased for phospho-peptides present in the sample as different isomers, compared to non-isomeric peptides, for both DDA and DIA experiments analyzed with a given software. We thus provide a ground truth phosphoproteomic dataset, which illustrates how positional isomers with different eluting times can be erroneously matched and quantified in classical label-free strategies. This could constitute a useful tool for the future improvement of processing methods in quantitative phosphoproteomics. This benchmarking study also enabled us to assess the relative quantitative performance of existing tools and acquisition methods. The DIA data analyzed with Spectronaut exhibited the best sensitivity/specificity ratio among all tested approaches, while the PD results led to the highest sensitivity, at the expense of a higher number of wrongly quantified phospho-peptides. The spiked T- cells dataset provided higher complexity to complement this benchmarking study and illustrated for example the benefit of ion-mobility (e.g. the FAIMS module on the Thermo Exploris instrument) for the identification and quantification of phospho-peptides. Finally, the created scripts, automating the whole comparison of the spiked-in dataset will be inside WOMBAT-P, which is a platform for benchmarking MS-based proteomics pipelines. This should ease further evaluations of the provided datasets by the proteomics community. 924

925

P-III-0765 Complex portal: a resource for functionally annotated macromolecular complexes Sucharitha Balu1, Kalpana Paneerselvam1, Juanjo Medina1, Eliot Ragueneau1, Kevin Drew2, Georg Kustatscher3, Sandra Orchard1, Henning Hermjakob1 1EMBL – EBI, Cambridge, United Kingdom 2University of Illinois at Chicago, Chicago, IL, United States 3University of Edinburgh, Edinburgh, United Kingdom The Complex Portal is a manually curated encyclopaedic resource of macromolecular complexes from 28 species which includes over 2150 human complexes. The resource enables the linking of knowledge on protein complexes across multiple databases, providing a reference set of component molecules with stable, versioned accession numbers. The Complex Portal defines a macromolecular complex as a stable set of (two or more) interacting protein and/or RNA molecules which can be co-purified and have been shown to exist as a functional unit in vivo. The degree of evidence for a complex is defined using the Evidence and Conclusion Ontology (ECO). Complexes are curated to a high level of granularity to include details on isoforms, variants, stoichiometry, and tissue-specificity. The Complex Portal is an active collaborator in the UniPlex project, a collaborative effort to expand the database and map the human complexome based on HuMap3.0 (http://proteincomplexes.org/) and Proteome HD 2.0 (https://www.proteomehd.net/). The newly generated data will be incorporated into the Complex Portal and regularly updated. The data derived will be consolidated from a combination of machine- learning predicted complexes and protein co-expression data, and further validated by manual curation. Data generated by the UniPlex project will greatly enrich the database, allowing for further exploration of the role of complexes through IntAct"s Cytoscape application using GO enrichment tags and disease ontology terms. The project will also enable the comparison of complexes under diverse cellular conditions for a better understanding of their role in biological functions and processes. The data in the Complex Portal is freely available formats (https://www.ebi.ac.uk/complexportal/download). XML, MI-JSON downloaded and tab-delimited and can be in PSI-MI 925

926

P-III-0766 Autoprot: a modular package for processing, analysis and visualization of mass spectrometry proteomics data in Python Julian Bender1, Johannes P. Zimmermann1, Wignand W. D. Mühlhäuser2, Friedel Drepper2, Bettina Warscheid1,2 1University of Würzburg, Biochemistry II, Theodor Boveri-Institute, Biocentre, Wuerzburg, Germany 2University of Freiburg, Biochemistry and Functional Proteomics, Institute of Biology II, Freiburg, Germany The rising complexity tandem mass spectrometry (MS/MS)-based proteomics data sets require standardized and reliable data analysis workflows. In this regard, scripted languages such as Python and R provide documentation of the analysis and, when combined with interactive notebooks such as Jupyter, interactive data inspection. Despite its central role for machine learning, there is only a limited repertoire of Python software available for the analysis of MS/MS data generated in data-dependent acquisition mode. In contrast, several specialised analysis tools have been developed, most prominently, in the R language. To leverage the power of both languages in data analysis, proficiency in Python and R is often required. However, this increases the hurdle for scripted data analysis to be widely adopted. We developed autoprot, a Python module for analysing proteomics search results generated with the MaxQuant software. Autoprot offers functionalities for common data processing tasks such as initial filtering, normalization, and statistical analysis all based on the popular pandas dataframe object. Moreover, it provides an interface to specialised functions so far only available in R thereby increasing the robustness of the analysis without reinventing the wheel. Furthermore, it generates dynamic javascript-based charts that can be integrated into interactive web applications. We show the application of autoprot using publicly available MS datasets, highlight functions of the submodules for data preprocessing, analysis and visualisation and showcase interactive plots generated with the software. Autoprot is written with collaboration and code customisation in mind using best practices for code writing as well as automatically build documentation and continuous testing via GitHub actions. In summary, autoprot provides standardised, fast, and reliable proteomics data analysis while ensuring a high customisability needed to tailor the analysis pipeline to specific experimental strategies. Fig. 1 (optional) 926

927

P-III-0767 Cross talk between Parkinson's disease and Alzheimer's disease in identification of concordant and discordant marker Harsheet Singh1, Deeptarup Biswas2, Ankit Halder2, Sanjyot Shenoy3, Dewaansh Vijayvargiya4, Aparna Chauhan2, Sanjeeva Srivastava2 1IIT Bombay, Aerospace Engineering, Mumbai, India 2IIT Bombay, BSBE, Mumbai, India 3IIT Bombay, Mathematics, Mumbai, India 4IIT Bombay, Civil Engineering, Mumbai, India Introduction: Neurodegenerative diseases (NDDs) such as Alzheimer's Disease (AD) and Parkinson's Disease (PD) present significant challenges, often progressing stealthily and causing irreversible neuronal damage before symptoms appear. Despite being classified under NDDs, AD and PD exhibit distinct molecular pathologies. However, there is limited research investigating the concordance and discordance between AD and PD on a global omics scale. Conducting a meta-analysis of proteomic and transcriptomic data to identify commonalities between these diseases can provide insights into their pathobiology and uncover markers for classification, offering potential therapeutic insights. Methods: The methodology involved data mining of proteomics and transcriptomics datasets from public repositories such as ProteomeXchange Consortium, PRIDE, ArrayExpress, Gene Expression Omnibus (GEO), and Omics Discovery Index (OmicsDI). A standardized approach was used for data preprocessing, quality assessment, and statistical analysis across all omics datasets. Significance annotation was performed using parametric T-tests to calculate p-values, and differential expression analysis was conducted comparing Disease versus Control groups. Common markers and unique features were identified to investigate the interplay between AD and PD. Results: Seven transcriptomics datasets for Alzheimer's and fourteen for Parkinson's, as well as seven proteomics datasets for Alzheimer's and three for Parkinson's, were analyzed, chosen for their balanced sample distribution. Among the top 1000 markers identified from each transcriptomics and proteomics analysis, 231 were common across diseases in transcriptomics, and 181 proteins were common in the proteomics workflow, highlighting shared molecular signatures. Through feature selection techniques, key features were identified across both diseases. Remarkably, 13 common markers emerged, underscoring both shared and distinctive aspects of these NDDs. The selection process confirmed the relevance of well-known markers, corroborating with BrainProt's BDMC score, validating their importance in disease classification and pathobiology. Conclusion: This study has shed light on the molecular landscape of Alzheimer's and Parkinson's diseases, revealing both shared and distinct features across transcriptomic and proteomic datasets. Through meticulous analysis and feature selection, a subset of common markers has emerged, providing insights into the pathobiology of these neurodegenerative disorders. The identification of shared molecular signatures suggests potential converging pathways and therapeutic targets, while the recognition of disease-specific markers underscores the heterogeneity within these conditions. These findings hold promise for guiding future research aimed at better understanding, diagnosing, and treating Alzheimer's and Parkinson's diseases. 927

928

P-III-0768 Integrating proteomics and PTM data for comprehensive multi-omics analysis Tim Blokker1, Christian Schiffmann1, Micael Fernandes dos Reis1, Christopher Bruhn1, Christiane Honisch1, Carleen Kluger1, Barbara Kracher1, Erik Schliep1 1Evotec International GmbH, PanHunter, Hamburg, Germany PanHunter is a user-oriented and interactive web application allowing for in-depth data discovery across multi- omics datasets without the need for coding skills. The platform was originally built by Evotec's bioinformatics team to handle large transcriptomics datasets generated by our industrialized high-throughput transcriptomics platform. In recent years, with the advancement of high-throughput mass spectrometry-based proteomics, major advances have been made to tailor PanHunter towards handling massive datasets from different proteomics data formats. The strength and novelty of proteomics analysis in PanHunter is founded in the holistic approach to integrate multiple levels of data (e.g. gene, protein, and post-translational modification (PTM)) through the whole workflow starting from quality control (QC) over exploratory analysis, differential expression analysis down to functional analysis, and incorporation of biological knowledge from various sources. PanHunter can map between PTM IDs, UniProt IDs, and on transcriptomics level Ensemble IDs. This mapping allows the user to analyze features simultaneously in linked plots, with data stemming from transcriptomics, proteomics or PTM data. Here, we provide examples to highlight how the expansion of PanHunter's proteomics capabilities facilitates data integration and interpretation in the context of precision medicine, thus driving easier access to novel medicines across various therapeutic areas. E.MPD, Evotec's molecular patient database, combining patient and omics data across multiple disease areas has been extended to cover autoimmune diseases in recent years. Integration of proteomics data from Evotecs's high-throughput mass spectrometry-based proteomics platform with patient data in PanHunter enables novel insights into the regulation of autoimmune diseases on the level of the proteome. Evotec's ScreenPep platform allows for high-throughput screening of compounds of interest. To enable seamless data analysis after the screening, PanHunter's signature database has been broadened with proteomics signatures from Mitchell et al. 2023. These signatures can be used to investigate toxicological effects and help decipher the mechanism of action of novel compounds. Mitchell, D.C., Kuljanin, M., Li, J. et al. A proteome-wide atlas of drug mechanism of action. Nat Biotechnol 41, 845–857 (2023). https://doi.org/10.1038/s41587-022-01539-0 928

929

P-III-0769 ProteoBench: a community-curated platform for comparison of proteomics data analysis workflows Robbin Bouwmeester1, Holda A. Anagho2, Sven Degroeve1, Nadezhda T Doncheva2, Viktoria Dorfer3, Klemens Fröhlich4, Ralf Gabriels1, Vedran Kasalica5, Caroline Lennartsson2, Matthias Mattanovich2,6, Emmanuelle Mouton-Barbosa7, Martin Rykaer2, Veit Schwämmle8, Maximilian T. Strauss2, Julian Uszkoreit9, Bart Van Puyvelde10, Tim Van Den Bossche1, Jakub Vašíček11, Henry Webel6, Witold Wolski6,12, Marie Locard- Paulet7 1Ghent University, Zwijnaarde, Belgium 2University of Copenhagen, Novo Nordisk Foundation Center for Protein Research (CPR), Copenhagen, Denmark 3University of Applied Sciences Upper Austria, Hagenberg, Austria 4University of Basel, Biozentrum Basel, Basel, Switzerland 5Netherlands eScience Center, Amsterdam, Netherlands 6University of Copenhagen, Novo Nordisk Foundation Center for Basic Metabolic Research, Copenhagen, Denmark 7Université de Toulouse, Institut de Pharmacologie et de Biologie Structurale (IPBS), Toulouse, France 8University of Southern Denmark, Department of Biochemistry and Molecular Biology, Odense, Denmark 9Ruhr University Bochum, Medical Faculty, Bochum, Germany 10Ghent University, Laboratory of Pharmaceutical Biotechnology, Ghent, Belgium 11University of Bergen, Department of Clinical Science, Bergen, Norway 12ETH Zürich / University of Zürich, Functional Genomics Center Zurich, Zurich, Switzerland Mass spectrometry (MS)-based proteomics is a crucial method for analyzing complex biological mixtures. With the abundance of dedicated data analysis pipelines, continuously updated and developed, the community would benefit from a platform to compare their performance in an objective and unbiased manner. Here we propose ProteoBench (proteobench.cubimed.rub.de/), a comprehensive and open web platform that allows for easy and controlled comparison of proteomics tools developed or used by the participants to other state-of-the-art pipelines. ProteoBench originated as a community project by the European Bioinformatics Community for Mass Spectrometry (EuBIC-MS) and its design and development are open to all interested researchers. Its goal is to provide a centralized resource, which is easy to use for non-coders. ProteoBench will comprise several benchmarking modules dedicated to the comparison of specific analysis tools developed for identification, quantification (DDA or DIA), and statistical analysis. Users shall be able to provide their analysis results for comparison, with the option of making these results publicly available. Currently, we have five modules in preparation, of which two are in active development and one is already available as a prototype. For this module you need to download a set of six publicly-available runs containing 3 species with known quantities. Any user can analyse these data with any workflow, and then upload analysis outputs back to ProteoBench, which then evaluates sensitivity and quantification accuracy (mean difference between expected and measured ratio) for precursor ions. It allows the quick visualisation of the corresponding results in the cloud of already-computed analysis outputs. The submitted data will continuously grow and should remain up-to-date with the latest developments. It will create a frame of reference for evaluating the performance of new and/or custom tools when publishing results and should increase transparency and reproducibility between data analysis pipelines developed in the field. It will allow end-users to select a well-performing workflow to meet their needs; developers to discern the strengths and limitations of their workflow, thereby directing further development; publishers and reviewers to easily position workflows in the context of existing state-of-the-art workflows. 929

930

P-III-0770 The annotation of post-translation modifications in UniProtKB Emily Bowler-Barnett1, Jun Fan1, Kerry Ramsbottom2, Oscar Martin Camacho2, Ananth Prakash1, Sun Zhi1, Yasset Perez Riverol1, Eric Deutsch3, Juan Vizcaino1, Andrew Jones2, Sandra Orchard1, Maria Martin1, UniProt Consortium1 1EMBL – EBI, Cambridge, United Kingdom 2University of Liverpool, Liverpool, United Kingdom 3Institute of systems biology, Seattle, WA, United States UniProtKB is a comprehensive, expert-led, publicly available database of protein sequence, function and disease variation information. Incorporation of data from a wide variety of resources, either imported from collaborative sources or added from the scientific literature by manual curation, provides a wealth of protein sequence, peptide modification, interaction, and protein function data. Post-translational modifications (PTMs) increase proteome diversity and allow a protein to alter many aspects of its biological state, including its function, cellular localization, interaction partners, and stability. PTMs can be fluid throughout the life cycle of a protein and contribute to the proteomes" dynamic and robust response to stimuli and biological pressures. PTMs therefore play a key role in both endogenous cellular systems and disease signaling pathology. Manual curation of the scientific literature facilitates the detailed capture of how proteins are modified within a cell and additionally how this might modify the function, cellular location, or interactions of that protein. The PTMeXchange project, in conjunction with the University of Liverpool, Peptide Atlas and PRIDE aims to re- analyse publicly available PTM-enriched proteomics datasets and control for accurate modification site localization, whilst abiding by FAIR data principles. We present an update to the protein post-translational modification landscape in UniProtKB, including new visualization tools, and multi-query search functionality both on the website as well as the UniProt Proteins API which allows access and complex querying of the whole database. 930

931

P-III-0771 PROTzilla – an open-source analysis platform for proteomics and peptidomics data for paper-ready visualizations Selena Braune1, Janni M. Roebbecke1, Grit L. Manssen1, Henning Gaertner1, Kuganash Ravishanker1, Luca J. Treide1, Anton G.A. Neubauer1, Lilly A. Zintl1, Hannes M. Spitz1, Fynn M. Kroeger1, Sara Grau-i-Blade1, Belana Zwadsich1, Pia Francesca Rissom1, Linea Schmidt1, Clara Uktar1, Lars Masanneck1, Hendrik Raetz1, Bernhard Y. Renard1, Christoph N. Schlaffner1 1Hasso Plattner Institute, University of Potsdam, Potsdam, Germany Over the last few years, mass spectrometry (MS) studies of the proteome and peptidome have increased exponentially in numbers and also in complexity. To gather insight into the underlying biology, researchers often rely on historically grown platforms that suffer from complexities requiring developer guidance to use. Additionally, some researchers manually string together multiple single-task tools or even fewer create custom pipelines, requiring programming knowledge. Challenges with data management between analysis steps, sharing of the used parameters, replicability of the analyses across large dataset cohorts, and general accessibility are a result. To address these challenges, a user-friendly tool that provides a one-stop-shop for downstream proteomic and peptidomic analysis in an easy-to-use graphical interface is required. Here, we introduce an open-source tool named PROTzilla in response. Our intuitive tool offers flexibility to customize data processing, ensures the storage of applied parameters for reproducible sharing of analysis workflows, and enables automated analysis through a command line interface for large-scale studies. PROTzilla comprises a Python package with preprocessing, analysis, high-quality plotting capabilities, and integration of external knowledge bases alongside a user-friendly and browser-based interface. Besides proteomic and peptidomic analysis workflows, PROTzilla also provides analysis methods for post-translational modification (PTM) centric research questions. Furthermore, our tool, for the first time, enables the direct continued analysis of proteins of interest and their associated peptides and PTMs, including a graph-based sequence coverage model for different isoforms. Utilizing our cross-platform tool PROTzilla researchers can effortlessly perform analysis tasks and generate publication-ready visualizations, eliminating the need for programming knowledge. The sharable analysis workflows in PROTzilla are reproducible as well as automatable and community members can easily contribute new functionality. With PROTzilla, we aim to simplify proteomics and peptidomics analysis in the field though enhanced flexibility, transparency, and reproducibility of analysis and reporting. We anticipate that this open-source tool will help researchers to effectively generate knowledge from their proteomics and peptidomics experiments and improve current practices for analysis sharing and reporting. 931

932

P-III-0772 Linking the human metabolome and proteome in UniProtKB through Rhea Lionel Breuza1, Lucila Aimo1, Ghislaine Argoud-Puy1, Kristian B. Axelsen1, Cristina Casals-Casas1, Elisabeth Coudert1, Nadine Gruaz-Gumowski1, Anne Morgat1, Nevila Hyka-Nouspikel1, Lucille Pourcel1, Shyamala Sundaram1, Alan Bridge1, the UniProt Consortium1,2,3 1Swiss Institute of Bioinformatics (SIB), Swiss-Prot group, Geneva, Switzerland 2European Molecular Biology Laboratory (EMBL), Cambridge, United Kingdom 3Protein Information Resource, Washington, DC, United States The UniProt Knowledgebase (UniProtKB, at www.uniprot.org) is a reference resource of protein sequences and functional annotation that is commonly used for proteomics, transcriptomics, and genomics analyses. Here we describe work that extends the domain of applications of UniProtKB to include integrated analyses of metabolomics and other chemical data, through the reannotation of all human enzyme and transporter functions using Rhea, an expert curated knowledgebase of biochemical reactions (www.rhea-db.org) based on the ChEBI ontology of small molecules (www.ebi.ac.uk/chebi/). With over 3,600 human enzymes and transporters now linked to Rhea (of 29 million sequences in UniProtKB linked to Rhea), this work is part of a wider effort to improve small molecule data in UniProtKB for all species that also includes annotation of ligand binding sites, and PTMs, using ChEBI. It improves interoperability with other data and knowledge resources dealing with small molecules, such as the metabolomics data repository MetaboLights, the protein structure repository PDB, the pathway knowledgebase Reactome, and the Gene Ontology, and provides enhanced support for metabolic modeling, multi-omics data integration and analysis, and the use of advanced machine learning approaches to predict enzyme function and biosynthetic and bioremediation pathways. 932

933

P-III-0773 Confident protein datasets for liquid-liquid phase separation studies Michał Burdukiewicz1,2, Carlos Pintado-Grima2, Oriol Bárcenas2, Valentín Iglesias2, Eva Arribas2, Salvador Ventura2 1Medical University of Białystok, Clinical Research Centre, Białystok, Poland 2Autonomous University of Barcelona, Institute of Biotechnology and Biomedicine, Barcelona, Spain Background: Proteins self-organize in dynamic cellular contexts by assembling into reversible biomolecular condensates by liquid-liquid phase separation (LLPS). These condensates can be composed of single or multiple proteins with different roles in the ensemble's structural and functional integrity. Driver proteins form condensates autonomously, without the requirement of any additional partner, while client proteins are later recruited into them and are not essential for maintaining the condensate architecture. Although several databases have been developed to catalog proteins undergoing LLPS, they often contain divergent data, reflecting distinct conceptions of the LLPS phenomenon. This inconsistency impedes interoperability between these resources, hindering the integrative usage of their contents. Moreover, there is an apparent missing consensus on selecting proteins without any explicit experimental association with condensates (negative data). These two aspects have prevented the generation of reliable predictive models and fair benchmarks. Results: To alleviate these constraints, in this work, we have carefully explored protein information from all relevant LLPS databases to generate confident datasets of client and driver proteins through an integrated biocuration protocol. Besides, to the best of our knowledge, we introduce for the first time standardized negative datasets, including both globular and disordered proteins. To validate our datasets, we investigated specific physicochemical properties related to LLPS across different sets of sequences and observed that specific features can be used to distinguish drivers and clients of LLPS but also against negative proteins (non- participants of condensates). Conclusions: The datasets presented here provide a reliable means for confidently assessing the specific roles of proteins in LLPS and for identifying key differences in physicochemical properties underlying phase separation. These highly confident datasets are poised to train a new generation of multilabel models, build more standardized benchmarks, and mitigate sequential biases associated with the presence of IDRs. Fig. 1 (optional) 933

934

P-III-0774 Enhancing biomolecule detection with a multi-Omic co-clustering model in mass spectrometry: incorporating additional spatial constraints for detecting latent patterns in imaging Giulia Capitoli1, Vanna Denti1, Michele Costanzo2,3, Lucia Santorelli4 1University of Milano-Bicocca, Department of Medicine and Surgery, Vedano al Lambro, Italy 2University of Naples Federico II, Department of Molecular Medicine and Medical Biotechnology, Naples, Italy 3CEINGE – Biotecnologie Avanzate Franco Salvatore, Naples, Italy 4Telethon Institute of Genetics and Medicine (TIGEM), Pozzuoli, Italy The constant innovation of novel measurement technologies has granted the availability of multiple omics information that can offer a comprehensive view of biological systems. Our research focuses on integrating multi-omics data, a crucial step to improve biomarker discovery and study the interplay of various molecular layers. Therefore, developing computationally scalable algorithms and statistical models that can provide accurate biological insights is of utmost importance. Modern mass spectrometry (MS) techniques analyze the abundance of a wide range of molecules from a specific biological sample. The MS imaging (MSI) technique also generates imaging data, with each pixel containing a mass spectrum. This spatial information reveals the localization of biomolecules (including lipids, N-glycans, and tryptic peptides) within the tissue, enabling the identification of patterns that may not be visible in traditional morphological images. Traditional statistical methods do not fully address the spatial dependencies of pixels and molecular networks. To overcome this, we developed a multi-omic co-clustering statistical model (based on Non-Negative Matrix Tri- factorization) that partitions the spatial expression profiles of multiple molecular layers. This model accounts for the spatial dependencies between neighboring pixels to infer the latent block structure of the data, resulting in two types of clustering: 1) clustering molecules based on their tissue-wide expression and 2) clustering image areas based on pixel coordinates. Our approach allows for the integration of data from sequential analyses of various molecular classes from the same tissue, uncovering hidden patterns due to spatial correlations and relationships between molecular layers. We validated this method on murine brain and clear cell renal cell carcinoma (ccRCC) tissues, aiming to deepen our understanding of biomolecular functions and interactions in different tissue regions, which is crucial for comprehending key biological mechanisms. This methodology is broadly applicable to any multi-omics approach that excludes spatial dependencies when handling MS data from solution samples. As a proof of concept, we tested the model on samples of hereditary metabolic diseases analyzed using Liquid Chromatography coupled with tandem mass spectrometry (LC-MS/MS). Specifically, we evaluated cellular models of methylmalonic acidemia by integrating proteomics, metabolomics, and lipidomics data to uncover specific molecular networks. Additionally, we will integrate the biostatistical model into a web application for multi-omics data integration, providing a user-friendly interface accessible to the scientific community. This work serves as a proof of concept, laying the foundation for interpreting clinical variables in pathophysiological contexts. 934

935

P-III-0776 ZURÜCKGEZOGEN: Peptide ranking strategy towards targeted workflow verification in brain disorders Aparna Chauhan1, Deeptarup Biswas1, Sampurna Dutta1, Sanjeeva Srivastava1 1Indian Institute of Technology Bombay, Department of Biosciences and Bioengineering, Mumbai, India Brain diseases are increasingly burdening global healthcare systems despite significant advancements in neurobiology research. The major challenges remain in accurate diagnosis, effective prognosis, and developing robust treatment strategies. Although numerous proteomics studies have been conducted, identified biomarkers have not yet achieved clinical application due to the need for further validation at the peptide level. Peptide databases such as PeptideAtlas and SRMAtlas serve as extensive repositories, cataloging peptides identified across various organisms, including humans, through mass spectrometry-based experiments. Despite their comprehensiveness, these databases lack disease-specific peptide scoring and the integration of qualitative and quantitative information. Currently, no database exists that leverages user data to prioritize peptides, resulting in the generic nature of the existing resources. The methodology involved the utilization of 15 datasets derived from six distinct brain diseases. These datasets were processed in MaxQuant with uniform parameters. Subsequently, the same datasets underwent brain sample-specific spectral angle calculation post-MaxQuant processing. Each peptide scan yielded a corresponding spectral angle the median for each scan of peptide gave the spectral angle score. Additionally, brain-specific score was generated from frequency and intensity of peptide abundance table. A combination of these scores along with PSS from PeptideAtlas were used to rank the peptides using statistical methods. The ranking was verified using MRM experiment on meningioma tissue and brain tumor cell line samples. Upon comparing the experimental results with the combined statistical rankings, it became evident that peptides with higher scores exhibited favorable peak characteristics, including dotp and peak area, across all 3 brain tumor cell line samples. Conversely, peptides with lower scores displayed suboptimal peak properties, validating the accuracy of the scoring strategy. These findings, facilitated by Skyline analysis shed light on peptide stability across various sample types in MS experiments. Similarly, in Meningioma tissue samples, peptides with higher scores demonstrated superior peak attributes. The developed peptide ranking strategy serves as a proof-of-concept to facilitate the targeted validation of protein markers and therapeutic targets within MS-based proteomics. The novelty of our peptide selection strategy was corroborated through experimental validation across multiple tumor tissues and cell lines, demonstrating a strong correlation between scores and experimental outcomes. This method enabled the curation of a robust peptide list suitable for designing targeted experiments tailored to brain disease samples. The implementation of this ranking strategy promises significant time savings in data and literature mining, and experiment planning for brain disease targeted proteomics experiments. 935

936

P-III-0777 QSample: a quality control system to monitor user's samples in proteomics core facilities Cristina Chiva1,2, Roger Olivella1,2, Eva Borràs1,2, Eduard Sabidó1,2 1UPF, Barcelona, Spain 2CRG, Proteomics Unit, Barcelona, Spain Quality in the proteomics laboratory has different facets including instrument performance, sample preparation, tracking of reagents and the establishment of standard operation procedures. We developed some years ago QCloud, a comprehensive automated QC system that supports proteomics laboratories in daily quality assessment using a user-friendly interface, easy setup, automated data processing and archiving, and unbiased instrument evaluation. Following the success on the implementation of QCloud we have now developed QSample, a system that helps to increase the quality and efficiency of the proteomics analyses of user"s samples automatically monitoring the key quality indicators of each sample processed in a proteomics laboratory in almost real time. Specifically, QSample is designed to support proteomics laboratories in the daily quality assessment of different proteomics applications including the analysis of regular proteomes, phospho-proteomes, and protein interactions, using either data-dependent or data-independent acquisition methods. For these workflows, QSample is currently capable of extracting several key quality control parameters that substantially simplify the quality assessment of the proteomics experiments. These parameters include, among others, number of proteins and peptides identified, number of missed cleavages, polymeric contamination and extension of the most common secondary reactions. Furthermore, the system is connected to QCloud to assess the instrument performance during the acquisition of the user"s samples. We have also containerized the QSample code to encapsulate all required dependencies to allow other laboratories to implement it. In summary, QSample enables for fast evaluation of users" samples, fast-decision making, and therefore, guarantees that proteomics core facilities acquire and deliver data conforming to the best research practice in the scientific community. 936

937

P-III-0778 Serving your proteomics analysis needs – the Core Unit for Bioinformatics (CUBiMed.RUB) in Bochum and de.NBI 2.0 / ELIXIR-DE Karin Schork1,2,3, Maike Weber1,2,3, Svitlana Rozanova1,2,3, Dirk Winkelhardt3,4, Sven Berres3,4, Julian Uszkoreit3,4, Martin Eisenacher1,2,3 1Medizinisches Proteom-Center, Ruhr University Bochum, Medical Bioinformatics, Bochum, Germany 2Ruhr University Bochum, PRODI research building, Medical Proteome Analysis, Bochum, Germany 3Ruhr University Bochum, Core Unit for Bioinformatics CUBiMed, Bochum, Germany 4Medical Bioinformatics, Ruhr University Bochum, Bochum, Germany Background: As a recently established core unit within the Medical Faculty of the Ruhr-University Bochum we are succeeding the Service Center "Bioinformatics for Proteomics - BioInfra.Prot" of the "German Network for Bioinformatic Infrastructure (de.NBI)", which was a seven year-BMBF funded project ending December 2021. Since March 2024 there is additionally a continued BMBF funding (de.NBI 2.0) via the "Forschungszentrum Jülich", which realizes the German node ELIXIR-DE of the European ELIXIR infrastructure. We provide a comprehensive portfolio of Bioinformatics cooperation and infrastructure possibilities to clinical, biological, and mass spectrometry users. Materials and Methods: On the one hand we provide well-established tools and workflows for the analysis of mass spectrometry data regarding peptide inference and peptide-to-protein quantification, as well as finding protein variants. On the other hand, we provide univariate and multivariate statistical methodology such as t-tests, multi-group tests, clustering and classification / machine learning. This portfolio is provided by web services / github projects, but also as Bioinformatics and Biostatistical consulting or even analysis (free for academic cooperation partners). All topics are also covered in training and education activities. identification, protein Results: Bioinformatics tools for Proteomics such as a quality control workflow, PIA (Protein Inference Algorithm), CalibraCurve or MaCPepDB (Mass Centric Peptide Database) have been developed, are maintained, and will be extended by user needs. Virtual servers and docker containers can be started by users in a high-core and / or high-RAM cloud infrastructure. Clustering, classification with machine learning and even deep learning are applied to Proteomics data and can be discussed with interested researchers. Training courses, including both beginner and expert R courses, are held regularly. The Proteomics cooperation possibilities are complemented by a multi-omics department of the core unit, that can also work with proteogenomics data. Conclusion: By providing free, high-quality infrastructure and cooperation opportunities, we play a significant role in the bioinformatics infrastructure within the proteomics research landscape. With that we complement the existing mass spectrometry core facilities in Germany and Europe. Fig. 1 (optional) 937

938

P-III-0779 A shiny application to create a library from DDA data for DIA-NN analysis and its visualization Amr Elguoshy1,2, Keiko Yamamoto1, Kengo Yanagita1, Tomohiro Uchimoto1, Tadashi Yamamoto1,3 1Biofluid Biomarker Center(BBC), Niigata University, Niigata, Japan 2Faculty of Agriculture, Al-Azhar University, Cairo, Egypt 3Shinrakuen Hospital, Niigata, Japan We present a comprehensive Shiny application designed to create a library for Data-Independent Acquisition (DIA) by DIA-NN, a state-of-the-art software tool for proteomics data analysis. from Data-Dependent Acquisition (DDA) mzid files, and to facilitate the post-processing and visualization of DIA-NN DIA results. This application integrates several critical functionalities into a user-friendly graphical interface, empowering users with minimal coding expertise to execute complex workflows. Key features include the parsing and extraction of relevant information at precursor and fragment ion levels from mzid files to construct high-quality DDA libraries, essential for accurate peptide identification in DIA workflows. Additionally, the application supports robust post-processing of DIA-NN output, including data normalization, filtering, and statistical analysis. Additionally, our tool is providing interactive and customizable plots for matched MS/MS profile between DDA library and DIA data identified by DIA-NN. Moreover. Users may generate and download summary statistics, visualize protein and peptide quantification data, and perform comparative analyses between experimental conditions. Our Shiny application offers a versatile solution for proteomics researchers, enhancing the reproducibility, accessibility, and efficiency of DDA library creation and DIA data analysis. 938

939

P-III-0780 Investigating the phenotypic proteome of Triple Negative Breast Cancer (TNBC) for precision medicine Akram Emdadi1, Henrik Johansson1 1Karolinska Institute, Oncology-pathology, Stockholm, Sweden Triple-negative breast cancer (TNBC) represents a subgroup of breast tumors characterized by the absence of the markers ESR1, PGR, and ERBB2. TNBC is a heterogeneous group of tumors with unmet needs due to its aggressive behavior and limited therapy options. Historically, chemotherapy has been the primary treatment for TNBC, until the introduction of PARP inhibitors and immune checkpoint inhibitors (ICI). However, certain subgroups of TNBC still rely solely on chemotherapy, highlighting the need for further patient stratification to optimize the use of ICI therapy. The primary goal of this study is to develop comprehensive knowledge of the proteome and establish proteomics assays to facilitate proteome-based precision medicine for TNBC. In this study, we aim to identify tumor subtypes and correlate them with therapy responses, thereby enhancing the biological understanding of triple-negative breast cancer (TNBC). A more comprehensive understanding of TNBC biology can facilitate the development of targeted therapies and improve patient outcomes. We are developing protein-based classifiers for validation in independent cohorts and identifying specific immune infiltration subgroups associated with distinct tumor characteristics. Additionally, we apply EcoTyper, a machine learning framework, to identify and characterize cell states using tandem mass tag (TMT) and data-independent acquisition (DIA) proteomics data. Our analysis includes ligand-receptor interactions using permutation tests to further elucidate the molecular landscape of TNBC. 939

940

P-III-0782 A comprehensive and accurate representation of human gene functions from large-scale evolutionary modeling and experimental gene ontology annotation Marc Feuermann1, Huaiyu Mi2, Pascale Gaudet1, Anushya Muruganujan2, Suzanna Lewis3, Dustin Ebert2, Tremayne Mushayahama2, Paul D Thomas2 1Swiss Institute of Bioinformatics (SIB), Swiss-Prot group, Geneva, Switzerland 2University of Southern California, Department of Population and Public Health Sciences, Los Angeles, CA, United States 3Lawrence Berkeley National Laboratory, Environmental Genomics and Systems Biology, Berkeley, CA, United States Understanding the set of functions performed by the protein-coding genes of the human genome has been a longstanding goal of biomedical research. The last two decades has seen substantial progress towards achieving this goal, with continued improvements in the annotation of the human genome sequence, and dramatic advances in the experimental characterization of human genes and their homologs from well-studied model organisms. Here, we describe the first attempt to create a draft human "functionome" through a comprehensive synthesis of functional data obtained for human genes and their homologs in non-human model organisms. All relevant function information in the Gene Ontology knowledgebase has been synthesized using an evolutionary framework based on phylogenetic trees, creating curated models of function evolution for 6333 gene families. This allowed the association of at least one functional characteristic to 82% of human protein-coding genes, each of which can be individually traced to experimental evidence in human and/or non-human model systems. Our evolutionary models provide an insight into how functions have evolved over different time periods to build up the repertoire of human gene functions. The evolutionary models and resulting GO annotations are useful in numerous applications from gene set enrichment analysis to understanding genetic evolution. 940

941

P-III-0783 To be or not to be – a non-canonical ORF MS-detectable? Ivo Fierro-Monti1, Eric Deutsch2 1University of Basel, Biozentrum, Basel, Switzerland 2Institute of System Biology, Seattle, WA, United States We present two statistical neural network learning models aimed at classifying non-canonical open reading frame (ncORF) peptides and microproteins or micropeptides into detectable or undetectable categories by mass spectrometry. For ncORF predicted HLA-binding peptides, a Multilayer Perceptron (MLP) Classifier model was trained on approximately equal number of detected and undetected peptides, including 22 features (selected using the Boruta algorithm). The MLP Classifier model achieved an accuracy of 0.699 and ROC AUC of 0.694, with the highest variable of importance showing correlation, thus suggesting potential for ncORF peptide detectability. For ncORF microproteins or micropeptides, a TensorFlow-Keras model was trained on 7264 instances, with 25% detectable, and 43 features (of which 36 Boruta-selected), yielding an accuracy of 0.60 and ROC AUC of 0.676. The model's performance is narrowly favourable to classify ORFs as detectable or undetectable, the ratio of Number of predicted HLA-binding peptides to the corresponding ncORF amino acid length emerged as the most important attribute for their detectability, followed by two attributes with similar importance: the designated orf_biotype4 - lncRNA category and the microprotein ncORF length in number of amino acids. The model's most important features allow to a certain extent prediction of ORF microprotein/micropeptide detectability. These models provide insights into the discriminatory features influencing ncORF peptide and microprotein detectability, facilitating a deeper understanding of HLA-I peptide and ncORF biology. Future work should focus on refining model performance with additional data e.g. on peptide abundance. 941

942

P-III-0785 Paradigm shift in biomarker translation: an innovative and automated pipeline to generate and select clinical- grade signature peptides during DIA discovery Qin Fu1, Manasa Vegesna1, Niveda Sundararaman1, Nicolaie Eugen Damoc2, Tabiwang Arrey2, Anna Pashkova2, Sandy Joung1, Susan Cheng1, Dalin Li3, Philip Debbas3, Dermot McGovern3, Yue Xuan2, Jennifer Van Eyk1 1Cedars Sinai Medical Center, Heart Institute, Los Angeles, CA, United States 2Thermo Fisher Scientific, Bremen, Germany 3Cedars Sinai Medical Center, Los Angeles, CA, United States The transition of biomarker candidates from discovery to the validation is an expensive, challenging, and time- consuming process. The Obitrap Astral MS platform achieves extensive plasma proteome coverage with data completeness and quantification repeatability. In combination with the innovative TEAQ (targeted extraction assessment of quantification) data analysis software solution, we have developed an automated pipeline to generate and select clinical-grade signature peptides that can directly be applied in clinical validation studies. Our test study is a 493 subjects Inflammatory bowel diseases (IBD) case-control cohort, identifying 1,116 peptides signature peptides fulfilling the clinical chemistry criteria. Pooled plasma samples at different loading amounts 12.5ng to 500ng were analyzed at three different throughputs (180SPD, 100SPD, and 60SPD) with five technical replicates. The throughput of 60 SPD shows the best balance between identification and quantification, identifying 1662 proteins with 1% FDR. Remarkably, 649 unique proteins with 97% of the precursors quantifiable in all 5 technical replicates. Notably, 80% of these plasma proteins were quantifiable across all technical replicates with a median coefficient of variance (CV) of less than 20%, required for clinical-grade assays. Therefore, 60SPD method was applied to analyze the IBD cohort (492 samples). To automate the selection of signature peptides representing. biomarker candidates that meet clinical chemistry criteria, TEAQ, was developed and deployed as a two-step process: 1) Assignment of all quantifiable precursors or peptides based on loading curve experiments. 2) Assignment of only those correlated quantifiable precursors or peptides within the linear range. The candidate biomarkers which are altered in IBD cohort were selected based on statistical significance and fold of change. TEAQ identified a total of 9481 precursors representing 1661 proteins, 781 and 346 of which has characterizable lower limits of detection and quantitation, respectively determined from 5 injections each of 11 loading points (12.5 ng to 1000ng) at 60SPD. To accelerate the identification of rigorous signature peptides for clinical validation phase, TEAQ selected 1,225 precursors and 1,116 peptides representing 327 proteins. This included CRP (with 1 signature peptide), a general inflammation marker which has been reported to elevate in response to inflammation in IBD patients. CRP was elevated in the IBD cohort with P=0.03. In addition, we have also observed two more acute phase proteins elevated in IBD: AGP1(Orosomucoid1) with 5 correlated signature peptides (P<0.0001) and AGP2 (Orosomucoid 2) with 3 correlated signature peptides (p=0.018). We have developed an innovative and automated pipeline to generate clinical grade signature peptides that can directly be applied in clinical validation studies. 942

943

P-III-0786 Deciphering cross-cohort metabolic signatures of immune responses and their implications for disease pathogenesis Jianbo Fu1,2, Nienke van Unen1, Andrei Sarlea3, Nhan Nguyen1, Martin Jaeger3, Javier Botey-Bataller1, Valerie Koeken3, L. Charlotte de Bree3, Vera P. Mourits3, Simone J.C.F.M. Moorlag3, Godfrey Temba3, Vesla I. Kullaya4, Quirijn de Mast3, Leo A.B. Joosten3, Cheng-Jian Xu1, Mihai G. Netea3, Yang Li1 1Centre for Individualised Infection Medicine (CiiM), a joint venture between the Helmholtz Centre for Infection Research (HZI) and Hannover Medical School (MHH), Hannover, Germany 2ETH Zürich, Department of Health Sciences and Technology, Zurich, Switzerland 3Radboud University Medical Center, Nijmegen, Netherlands 4Kilimanjaro Christian Medical University College (KCMUCo), Department of Medical Biochemistry and Molecular Biology, Moshi, Tanzania, United Republic Of Background: The intricate interplay between circulating metabolites and immune responses, though crucial to disease pathophysiology, remains poorly understood and underexplored in systematic research. Methods: We performed a comprehensive analysis of the relationships between 4,361 metabolic features and 172 different cytokine responses to stimuli in two Western European cohorts (534 and 324 healthy individuals) and one sub-Saharan Africa (323 healthy donors). To elucidate the metabolic networks underlying immune phenotypes, we specifically assessed the co-expression networks of metabolites and immune responses (IL-1β, IL-6, TNF, IFN-γ) following Staphylococcus aureus stimulation. In analyzing the relationship between metabolites and immune cytokines, we initially filtered features based on their Spearman correlation p-values, retaining only those that passed a specific threshold (0.05) for further analysis. Subsequently, we conducted in vitro experiments to further validate these relationships. To explore the potential role of metabolites in disease, we examined the differential expression of metabolites infectious diseases and used the Mendelian randomization (MR) to study the causal relationships between these metabolites and the disease. in Results: Our study identified glycerophospholipid metabolism as a consistent link between metabolic pathways and immune responses, highlighting its role in regulating infection-induced immune responses, while revealing sex-specific differences in the correlation between phosphatidylcholine and cytokine responses to ex-vivo stimulations. Notably, a significant negative correlation was observed between sphingomyelin and the monocyte-derived cytokine response in response to Staphylococcus aureus stimulation, a relationship validated through functional experiments. Subsequently, employing MR analysis, we established a link between sphingomyelin and COVID-19 severity, providing compelling evidence for a modulatory effect of sphingomyelin on immune responses during human infection. Collectively, Conclusions: (https://lab-li.ciim- hannover.de/apps/imetabomap/) for exploring metabolic signatures associated with immune function in different populations, highlighting sphingomyelin as a potential target in treating inflammatory and infectious diseases. represent a unique our results resource 943

945

P-III-0787 The uncharted territories of EGF signaling: a deep-dive via short-time phosphoproteomics and network models Martin Garrido-Rodriguez1,2, Clement Potel1, Mikhail Savitski1, Julio Saez-Rodriguez2,3 1EMBL – EBI, Genome Biology Unit, Heidelberg, Germany 2Heidelberg University, Faculty of Medicine, and Heidelberg University Hospital, Institute for Computational Biomedicine, Heidelberg, Germany 3EMBL – EBI, Cambridge, United Kingdom Cellular signaling processes are driven by protein interactions, which transmit information through various mechanisms. Phosphorylation-driven activation and inhibition of proteins is a fundamental aspect of signaling, and computational models can make this knowledge actionable. However, classic models were limited by the proteome coverage of traditional biochemistry techniques and the small amount of well characterised kinase- substrate interactions. Recent advances in phosphoproteomics have enabled measurement of tens of thousands of phosphosites per experiment at short time scales and subcellular resolution. Additionally, screenings of kinase-substrate interactions have expanded our knowledge of serine/threonine and tyrosine kinases, while computational models have improved with the application of modern transformer architectures through protein language models. In this study, we combine these new resources with an ultra-deep characterization of the EGF signaling response using short-term phosphoproteomics. We integrate modern data and knowledge into a network model that considers the dynamic constraints measured at proteome scale. Using this model, we uncover known and propose new mechanisms associated with EGF signaling. This study not only uncovers new EGF-related signaling mechanisms, but also aims to expand the community"s understanding of the complexity of signaling pathways, demonstrating that they are at least an order of magnitude larger than traditionally described. Fig. 1 (optional) 945

946

P-III-0788 Integrating statistical tests for differential analysis in quantitative proteomics: enhancing power while controlling false discoveries Quentin Giai Gianetto1, Louka Chenal1, Mariette Matondo1 1Institut Pasteur, Paris, France Highlighting the proteins/peptides exhibiting significant quantitative differences between two biological conditions is the cornerstone of many quantitative proteomic experiments. Typically, this involves employing a statistical strategy across the quantitative dataset to extract relevant proteins/peptides. Basic approaches include fold-change calculations or standard Student's t-tests. Various statistical tests have emerged to refine such analyses and to allow for better statistical power. Some adjust the denominator of the Student's t-statistic (e.g., Welch, SAM, LIMMA, BayesT, VarMixT, ROTS), while others are rank-based (e.g., Wilcoxon, LPE, RankProduct). Recent methods from explainable artificial intelligence were also developed to provide p-values (e.g., PIMP, NTA). Since the 2000s reproducibility crisis, skepticism about p-values has grown due to their reliance on hypothetical distributions and the need for many replicates to be reliable. This issue is notable in proteomics with limited experiment replicates. Alternatives like the « Bayesian framework », using measures like the region of practical equivalence (ROPE) or probability of direction (PD), have recently emerged. Herein, we undertook a comprehensive analysis, comparing many statistical tests and strategies on various simulated and real data to assess their similarities and differences. Additionally, we explored ways of combining strategies to maximize the statistical power (or true non-discovery rate) while controlling the FDR (false discovery rate), using methods like p-value fusion (e.g., Fisher's, Stouffer's,…) or data-driven strategies (machine learning classification). For this, we investigated the estimation of the FDR when combining statistical strategies by estimating joint empirical cumulative distribution functions. Our study proposes to exploit a variety of different tests to extract the best possible strategy in difficult contexts, such as to detect small variations in abundances between conditions (small effect sizes). We present the most successful statistical strategies and the combinations providing the best statistical power while controlling the false discovery rate under a fixed threshold. Our comprehensive approach yields valuable insights applicable to any quantitative dataset characterized by a "large p, small n" framework, commonly encountered in proteomics but also in other quantitative omics datasets, such as transcriptomics or metabolomics datasets. 946

947

P-III-0789 Estimating the taxonomic composition of complex samples using metaproteomics and machine learning Hamid Hachemi1 1Li2D, CEA, Codolet, France Metaproteomics is a powerful tool to characterize how microbiota function by analyzing their proteic content by tandem mass spectrometry. Given the complexity of these samples, accurately assessing their taxonomical composition without prior information and based solely on peptide sequences remains a challenge. Here, we present LineageFilter, a new python-based AI software for refined proteotyping of complex samples using metaproteomics raw data and machine learning. Given a tentative list of taxa, their abundance, and the scores associated to their identified peptides, LineageFilter computes a comprehensive set of features for each identified taxon at all taxonomical ranks. Its machine-learning model assesses the likelihood of each taxon's presence based on these features, enabling efficient filtration of false-positive taxa. LineageFilter represents a robust and powerful tool to improve identification of the taxa composition of complex multispecies samples, paving the way for a better proteins" annotation and functional annotation for metaproteomics. The versatility of LF is showcased by its effectiveness in handling noisy data, allowing for the extraction of a higher number of proteins validated at FDR 1%. As presented, the LF method could be easily implemented in external tools and pipelines supporting large-scale data analysis and currently using Unipept, to improve proteotyping sensitivity and recall. Furthermore, despite the recommended default settings, LF could easily be customized with other parameters, as long as the training is performed with training datasets filtered at the required PSM confidence. In conclusion, the use of LF, prior to further metaproteomics analysis, enhances the efficiency and particularly the precision of standard metaproteomics tools. When applied to the proteotyping of metaproteomics datasets, it allows a better description of how microbiota function by enlarging the list of proteins identified and certifying their origins. 947

948

P-III-0790 Integrated Omics-Based Knowledgebase BrainProt™ and BDDF: Advancing the Identification of Druggable Proteins Ankit Halder1, Sabyasachi Samantaray1, Sahil Barbade1, Sanjeeva Srivastava1 1IIT Bombay, Biosciences and Bioengineering, Mumbai, India In this study, we intend to dig deeper The development of integrated Omics-based knowledgebase BrainProt™ (https://www.brainprot.org/) has facilitated our understanding of markers associated with brain diseases. Further, the development of BDDF (Brain Disease Drug Finder) domain has led to the understanding of the availability of current drugs targeting these disease and their associated markers. But, huge indistinctness lies in the mind of the researchers in choosing a certain drug target due to lack of comprehensive models for identifying the best available targets.We emphasise that beyond screening the potential inhibitors, an understanding of the viable drug targets is also crucial. The unavailability of reliable structures for a large portion of the druggable proteome leads to the inaccurate results in in silico screening. Previous studies attempting to understand the properties of the druggable proteome using simple physicochemical properties have failed to achieve optimum accuracy via reliable methods. into protein characteristics,beyond basic physicochemical properties of the protein, and incorporate a wealth of SwissProt-verified information extracted from UniProt, including subcellular locations, protein-protein interactions and peptide domains. We employ several feature selection strategies to disseminate the major criteria for a protein druggability and evaluate these features comprehensively. For non interpretable models like Neural Networks, we utilise Shapley values to promote explainability and as an interpretation to feature importance. The inherent skewness in the number of druggable versus non-druggable proteins (according to DrugBank (https://go.drugbank.com/) poses a challenge for model development. We address this imbalance using oversampling strategies like SMOTE. Our classifier has achieved approximately 80% accuracy, 0.8 F1-score and AUC 0.9 in predicting the druggable and non druggable proteins on the held-out test set. Looking ahead, we hypothesise that employing Recursive Neural Networks(RNNs) surface accessibility(RSA), ASA, PDB secondary structure information etc. can further improve the accuracy. We anticipate that these models can be deployed at a large scale to facilitate the drug discovery process more effectively. to encode sequential properties flexibility, hydrophobicity, like relative 948

949

P-III-0791 From post-translational modifications to proteoforms Yannick Hartmaring1, Bernhard Y. Renard1, Christoph N. Schlaffner1 1Hasso Plattner Institute, University of Potsdam, Potsdam, Germany A post-translational modification (PTM) is an alteration on a protein that changes the amino acid sequence into a functional proteoform. It can regulate structure, stability, cellular localization and function of protein and, therefore, affect almost all cellular processes. However, with over 200 different known PTMs, the overall complexity of the proteome is exponentially increased impeding confident identification and quantification using biochemistry and mass spectrometry. In recent years, the heterogeneity across samples and interactions between PTMs, called cross-talk, have garnered broader interest through associations with diseases such as cancer, diabetes, and Alzheimer’s disease. Identifying dependencies between PTMs on the same protein has been a challenge both in the shotgun and top-down approaches of proteomic mass spectrometry. While approaches for identification of various PTMs exist, PTM-centric proteoform assembly methods are lacking to further investigate PTM cross-talk. To address this issue, we have developed an approach which is able to reconstruct the proteoforms from shotgun proteomics experiments. Using an expectation maximization algorithm, we iteratively exclude unlikely combinations of peptides with and without PTMs to maximize the probability for the most likely set of proteoforms supported by the underlying peptide and PTM data. Thereby, we provide the most likely set of only few proteoforms that can be synthesized in a laboratory and used for targeted analyses. We tested our approach by simulating shotgun mass spectrometry experiments for a given set of ground truth proteoforms. Our results show high accuracy and precision. And additional benefit of our approach is the post-hoc processing capability to vast amounts of publicly available proteomics data, making cross species, cross tissue, and cross condition comparisons of PTM-centric proteoforms possible. With the knowledge gained from reconstructing the most likely set of proteoforms and therefore the identified co-expression of PTMs, we hope to provide researchers with a tool to perform targeted analyses with biochemistry and synthesizes versions of the proteoforms. This will provide new insights into cellular protein regulation and, furthermore, highlight sets of PTMs that are of particular interest in a variety of human tissues and conditions. Additionally, personal PTM patterns could be identified that can help develop new and targeted treatments for various diseases. 949

950

P-III-0792 Exploring dynamic behavior with SmartPhos: an integrated pipeline for low-input, high-throughput Phosphoproteomics in clinical samples Barbara Helm1,2, Dario Frey1,2, Cong Quan Ta1,3, Yueyang Xie1,3, Shubham Agrawal3, Marc Schneider2,4, Thomas Muley2,4, Laura Klotz2,4, Hauke Winter2,4, Dominic Helm1, Marcel Schilling1, Junyan Lu3, Ursula Klingmüller1,2 1German Cancer Research Center (DKFZ), Heidelberg, Germany 2German Center for Lung Research (DZL), Translational Lung Research Center (TLRC), Heidelberg, Germany 3Heidelberg University, Institute of Computational Biomedicine, Heidelberg, Germany 4Thoraxklinik at University Hospital Heidelberg, Heidelberg, Germany Protein phosphorylation, a transient modification, plays a pivotal role in intracellular information processing. Changes in these processes significantly influence disease onset and development. Comprehensive studies examining changes in phosphorylation sites are crucial to unravel these complex, non-linear interrelations. However, a robust mass spectrometry-based phosphoproteomics workflow is challenging, given the low abundance of phosphorylated proteins in highly complex samples. In response to these challenges, we introduce SmartPhos, a cohesive multistep phosphoproteomics sample preparation and data analysis pipeline designed for longitudinal cell culture experiments and snapshot tissue samples. It reduces the need for starting material and aids in generating time-resolved data on phosphorylation sites. The user-friendly SmartPhos software streamlines data its effectiveness, we used the SmartPhos pipeline to study the dynamics of growth factor-induced phosphorylation of the receptor tyrosine kinases EGFR and MET in non-small cell lung cancer (NSCLC) cells. In two days, we processed 156 samples, generating a report that included over 9,000 proteins and 40,000 class I phosphosites for each time point. The reliability and robustness of our pipeline were confirmed by comparing the phosphoproteome results with those from quantitative immunoblotting on the dynamics of selected phosphorylation sites. The SmartPhos pipeline enables a system-wide analysis to unravel dynamic cellular responses, while mathematical models facilitate the understanding of disease progression in individual patients. Together, they provide a holistic perspective on cellular information processing and yield critical insights for mathematical modeling to dissect disease-promoting mechanisms at a personalized level. import, quality control, preprocessing, and exploratory analysis. To demonstrate 950

951

P-III-0793 Reactome biomolecular pathways database – update Eliot Ragueneau1, Alexander Grentner2, Chuqiao Gong1, Cristoffer Sevilla1, Peter D'Eustachio3, Guanming Wu4, Lincoln Stein5, Johannes Griss2, Henning Hermjakob1 1European Molecular Biology Laboratory (EMBL), European Bioinformatics Institute (EMBL-EBI), Cambridge, United Kingdom 2Medical University of Vienna, Vienna, Austria 3NYU Grossman School of Medicine, New York, NY, United States 4Oregon Health and Science University, Portland, OR, United States 5Ontario Institute for Cancer Research, Toronto, Canada The Reactome database of curated human pathways (Milacic et al, 2024) covers 11,442 proteins in 15,212 reactions supported by 38,549 literature references (Release 88, March 2024). Building on our successful collaboration with the DiseaseMaps community (Ostaszewski et al, 2021), we are increasing our efforts in pandemic preparedness, among others through the recent release of the Respiratory Syncytial Virus (RSV) Infection Pathway. We are now also providing "early access" to pathways still awaiting external review, to extend the available data, and to invite community collaboration around specific curation topics. Pathway analysis capabilities range from simple gene enrichment to quantitative multi-omics analysis with ReactomeGSA (Griss et al. 2020). Analysis can be submitted through an R interface, and results are presented interactively in a visually attractive genome-wide results overview based on Voronoi maps. ReactomeGSA now includes a function that enables users to directly search for public datasets in Expression Atlas (Moreno et al., 2022) and GREIN (Mahi et al., 2019) from within ReactomeGSA, that can then be loaded into a ReactomeGSA analysis. These new features are available through a completely re-developed web interface as well as the ReactomeGSA Bioconductor R package. In the updated ReactomeGSA we are also piloting a new look-and-feel, and integrated user guidance, features planned for the next version of the Reactome home page, which will be released in 2025. Reactome is open source, open data, open graphics, all components are distributed under permissive licenses (CC0, CC-BY). Fig. 1 (optional) 951

952

P-III-0794 Spectra-sum method for protein qunatification from LC-MS/MS data with TMT-labeling Heeyoun Hwang1, Yubin Ko2, Geul Bang1, Jin Young Kim1 1Korea Basic Science Institute, Digital OMICs Research Center, Cheongju, South Korea 2Yonsei University, Chemical Genomics Leader Research Initiative, College of Life Science & Biotechnology, Seoul, South Korea There are two methods used to calculate data for protein quantification using LC-MS/MS with TMT labeling. Spectra-sum method is calculated as a sum of the intensities of TMT report ions in spectra level. In other hand, Spectra-ratio method is calculated as a median of the ratio intensities with the global reference ion. In order to find a best quantification method, we prepared Hela standard peptides with TMT-6-plex labeling of difference protein content ratio as 0.5:0.5:1.0:1.0:2.0:2.0. First, we identified 3,684 proteins with FDR < 1% as spectra and global protein level using Integrated Proteomic Pipeline and ProteinInferencer. Of 30 replicated experiment of day by day, 16,384 is median number of identified spectra. We calculated the ratio value between the different protein content samples using the TMT reporter ions. As a result of spectra-sum method, the mean value of the ratio of number of proteins with CV < 30% was 0.865, and its standard deviation (SD) was low (0.07). The other hand of spectra-ratio method, the mean value was 0.45 and SD was high (0.11). It is indicated that spectra-sum method is more accurate quantification, where 87% of proteins could be quantifiable. 952

953

P-III-0795 Machine learning unlocks new insights in ovarian cancer proteomics Ekaterina Ilgisonis1, Anna Kliuchnikova1, Anna Kozlova1, Elizaveta Sarygina1, Svetlana Tarbeeva1, Elena Ponomarenko1 1IBMC, Laboratory of postgenomic data analysis, Moscow, Russian Federation Proteomics has revolutionized cancer research, offering profound insights into the molecular underpinnings of oncogenesis. In this study, we reanalyze an ovarian cancer proteomics dataset using advanced machine learning (ML) techniques to uncover novel biomarkers and therapeutic targets. The dataset, comprising 189 proteomic profiles from the PRIDE database, includes samples from serous ovarian cancer, non-cancerous tissues, and plasma. Data were split into 80% for training and 20% for validation. Our approach integrates unsupervised clustering, supervised classification, and feature selection algorithms. Supervised ML models, including deep learning networks, achieve an impressive 94% accuracy in predicting and diagnosing ovarian cancer, significantly outperforming traditional methods. Feature selection techniques identify a ranked list of 50 key proteins, used for the further pathway analysis. In addition to standard ovarian cancer biomarkers presented in the OVA1 diagnostic test (CA125, transthyretin (prealbumin), apolipprotein A1 (Apo A-1), beta-2 microglobulin, and transferrin) the list of proteins exhibiting specific expression patterns includes von Willebrand factor, actin, cytoplasmic 1, attractin and immunoglobulin lambda-like polypeptide 1. This reanalysis not only provides a comprehensive proteomic profile of ovarian cancer but also underscores the potential of deep learning models for its early prediction and diagnosis. Our findings demonstrate the power of ML in transforming proteomics data into actionable biological insights, paving the way for improved preclinical cancer screening and personalized management of ovarian cancer. In summary, this work represents a significant step forward in the application of ML to cancer proteomics, demonstrating the potential to unlock new dimensions of understanding in ovarian cancer research. 953

954

P-III-0796 Spatial proteomic analysis of breast cancer heterogeneity unravels mechanisms of cancer cell phenotypic plasticity Shiri Karagach1, Mariya Mardamshina2, Vishnu Mohan1, Ofra Golani3, Iris Barshack4, Gabriella Jacobson- Eisenberg1, Tamar Geiger1 1Weizmann Institute of Science, Molecular Cell Biology, Rehovot, Israel 2Tel Aviv University, School of Medicine, Tel Aviv, Israel 3Weizmann Institute of Science, Department of Life Sciences Core Facilities, Rehovot, Israel 4Sheba Medical Center, Pathology Institute, Ramat Gan, Israel Tumor heterogeneity is the primary driver of drug resistance and relapse. Treatment-resistant clones lead to tumor relapse and metastasis, even years after an initial cancer diagnosis, and variability in the interactions with the immune system can lead to immune evasion and tumor progression. Despite its critical importance, intratumor heterogeneity (ITH) has been studied primarily at the genomic level over the last decade. In contrast, the functional outcomes of heterogeneity and the interplay with the tumor microenvironment have not been fully addressed. In this study, we integrated multi-region spatial mass spectrometry (MS)-based proteomics of 280 tumor regions, exome sequencing, and imaging to investigate spatial proteomic heterogeneity in breast cancer. We identified 8,263 proteins with relatively uniform receptor expression patterns within each region. Using this multi-layered data, our findings reveal increased proteomic heterogeneity with tumor progression, independent of genomic heterogeneity but closely associated with microenvironmental differences. Integration with immune and stromal imaging highlighted a dynamic interplay where low-grade tumors exhibit constrained immune infiltration, and upon progression, macrophages and T-cells infiltrate. However, pathways involving kynurenine and prostaglandins impede anti-tumorigenic activities. Integration with the global protein network provides potential novel targetable mediators of immune evasion in breast cancer, which can serve as the basis for future development of personalized cancer treatments. 954

955

P-III-0797 Evaluating knowledge graphs for disease prediction and therapeutic drug recommendation using proteomics data Daniel Kautzner1, Cassandra Königs1, Marcel Friedrichs1, Robert Heyer1,2 1Bielefeld University, Faculty of Technology, Bielefeld, Germany 2Leibniz-Institut für Analytische Wissenschaften – ISAS – e.V., Dortmund, Germany Introduction: Proteomics is important for understanding disease mechanisms and identifying therapeutic targets by linking proteins to various biological aspects. However, linking proteins with their function and further information (e.g., protein-drug interactions) requires more sophisticated approaches and data structures. Implementing knowledge graphs (KGs) could systematically organize and standardize protein data and further information, enhancing the accessibility and clarity of protein connections. Therefore, KGs potentially make automated decision processes possible such as automated diagnosis or treatment customization for individual patients. Recent KGs, such as the Clinical Knowledge Graph, BioDWH2-KG, BioKG, PharMeBINet, and PrimeKG, have been developed to incorporate proteomics information and connect them with various biological aspects. Material and Methods: Our objective is to test these five KGs with clinical proteomics data to evaluate their current capabilities and identify limitations. Specifically, we aim to determine if KGs can connect patient proteomics data with corresponding diseases, identify key disease processes based on the patient data, and assist in identifying therapeutic drugs based on dysregulated proteins involved in key disease processes. This evaluation should provide an overview of how these KGs perform in these tasks and assess whether it is possible to automate diagnosis and treatment customization for individual patients, or what additional developments are needed to achieve this. For testing, we utilized diverse patient datasets, including results from a meta-study on inflammatory bowel disease (IBD) biomarkers, proteomics data from pre- and post-drug treatment for IBD, and data from patients with polyneuropathy and platelet disorders. Results: Our first results indicate that PharMeBINet provides the most comprehensive information for connecting proteins and drugs, resulting in the best recommendations for therapeutic drugs. However, a combined approach using PharMeBINet, Clinical Knowledge Graph, and BioDWH2-KG significantly enhances the quality of insights due to their varied data integration methods and diverse database coverage. Predicting diseases using solely proteomics data faces challenges, primarily due to a bias stemming from better-researched diseases having more connections and a higher chance to connect to the patient data. Consequently, disease prediction for less-studied diseases, which are poorly connected, tends to yield poorer outcomes. Conclusion: This study will comprehensively evaluate existing KGs with incorporated proteomics data, examining their potential to automate disease diagnosis and drug recommendation processes. 955

956

P-III-0798 TIMS DIA-NN 3.0: improved algorithms to enhance peptide detection with confidence and accuracy Jonathan Krieger1, Robin Park2, Dennis Trede3, Tharan Srikumar1 1Bruker Ltd, Mississauga, Canada 2Bruker Daltonics GmbH & Co. KG, San Jose, CA, United States 3Bruker Daltonics GmbH & Co. KG, Bremen, Germany Due to the rising popularity of data-independent acquisition (DIA), we previously developed and launched an automated spectral library generation tool and the CCS-aware TIMS DIA-NN 2.0 for DIA analysis, for the purpose of analyzing high-throughput mass spectrometry data. Subsequent to this initial launch, we have continuously put effort into creating TIMS DIA-NN 3.0 by redesigning and implementing new algorithms. The primary focus in this iteration has been on substantially improving the confidence, accuracy, and reliability of both identification and quantification processes. In processing raw files, the TIMS DIA-NN effectively identifies local maxima by clustering and merging peaks, employing a 4D-proteomics approach. This method integrates key dimensions - collision cross-section (CCS), retention time, mass-to-charge ratio (m/z), and MS/MS fingerprinting - to optimize peak detection. Enhancements have been made with the introduction of multiple new algorithms, including new mass drift and mass tolerance calibration algorithms, which improve the detection of peptide candidates and the precision of chromatogram construction. Additionally, we have integrated novel algorithms, including a Gaussian correlation score for peak detection in neural network analysis, peak boundary detection in chromatograms, and fragment ion selection for MBR to improve reproducibility. These improvements collectively contribute to the improvement of the consistency of the coefficient of variation (CV) across replicates, thereby elevating the robustness and accuracy of the data analysis. We ran both TIMS DIA-NN 3.0 and 2.0 on the Hela DIA triplicate data, which was acquired from the Bruker timsTOF Pro, for benchmarking purposes. The analysis was performed using a spectral library generated from a ProteoScape DDA search. While TIMS DIA-NN 2.0 identified 62,402 peptides with a CV of 16.61, TIMS DIA-NN 3.0 identified 64,507 peptides with a reduced CV of 12.59. Furthermore, when using a standard tri-species Human/Yeast/EColi; of the total 10304 proteins identified, 9698 (94%) are quantified with a CV <20%, and 7778 (75%) are quantified with a CV <10%, indicating excellent quantitation. The TIMS DIA-NN 3.0 has been integrated into the Bruker ProteoScape platform and is available to all timsTOF and ProteoScape users. 956

957

P-III-0799 The PRIDE Affinity-proteomics Archive (PRIDE-AP): making affinity proteomics datasets FAIR Deepti Jaiswal Kundu1, Selvakumar Kamatchinathan1, Chakradhar Bandla1, Jingwen Bai1, Suresh Hewapathirana1, Kathy Nevola2, Lina Hultin Rosenberg2, Eva Schmidt2, Juan Vizcaino1, Yasset Perez Riverol1 1European Bioinformatics Institute (EMBL-EBI), Proteomics, Cambridge, United Kingdom 2Olink Proteomics, Uppsala, Sweden Background: The PRIDE database at the European Bioinformatics Institute, also a founder of ProteomeXchange, is the most popular resource supporting open science practices in the mass spectrometry-based proteomics field. The PRIDE Affinity Proteomics (PRIDE-AP) Archive, is a new dedicated section within PRIDE that aims to revolutionize affinity proteomics by centralizing openly accessible datasets generated using affinity proteomics platforms. PRIDE-AP Archive aims to establish an inclusive technology-agnostic repository representing all companies. As a first step, data submission of Olink Proteomics Proximity Extension Assay (PEA) (https://olink.com/) has been realized. By adhering to the principles of Findability, Accessibility, Interoperability, and Reusability (FAIR), PRIDE-AP seeks to enhance collaboration, expedite research, and ensure the high-quality dissemination of affinity proteomics data. Methods: The PRIDE-AP Archive defines a set of guidelines for the submission, validation and dissemination of affinity proteomics datasets while maintaining stringent adherence to FAIR principles. PRIDE-AP employs a robust submission (using the PRIDE submission tool) and validation process to ensure datasets meet comprehensive metadata and file requirements. Each dataset is assigned a unique PAD identifier (e.g., PAD000001) for citation. Submitters must provide essential metadata, including title, description, sample and data analysis protocols, and information on data submitters and principal investigators. Currently supported technologies include Olink Proteomics (e.g. Explore & Target technologies) submissions with their specific 'RAW' file type requirements (.parquet for Olink Proteomics). Additional files, such as experimental design documents and quality control reports, can also be included. The PRIDE-AP Archive infrastructure facilitates data submission, review, and public release, and the data remains unreleased to the public until publication. Results and Discussion: Since its inception, PRIDE-AP has successfully integrated several high-quality affinity proteomics datasets, making them accessible to the scientific community. Implementing unique PAD identifiers has streamlined dataset citation and potential reuse in the scientific literature. The structured metadata requirements and stringent file validation processes have resulted in well-annotated, interoperable datasets. The PRIDE-AP Archive significantly advances the field of affinity proteomics by adhering to the FAIR principles. By centralizing datasets from technologies such as Olink Proteomics, PRIDE-AP facilitates collaboration, accelerates research, and promotes the widespread dissemination of high-quality data. Moving forward, PRIDE-AP will evolve, supporting new technologies and expanding its impact on the affinity proteomics field. 957

958

P-III-0800 Input normalization and PTM stoichiometry calculations in Spectronaut® Veronique Laforte1, Damiano Robbiani1, Monika Puchalska1, Anamarija Pfeiffer1, Tejas Gandhi1, Lukas Reiter1 1Biognosys AG, Research and Development, Schlieren, Switzerland Introduction: Post-translational modifications (PTM) such as protein phosphorylation can dramatically alter the function of proteins and have important roles in disease and health as biomarkers or pharmaceutical targets. An important aspect of PTMs is the rate of site occupancy or PTM stoichiometry, which can provide biological information irrespective of the underlying protein abundance. In proteomics experiments where modifications are studied using an enrichment step, it is important to decouple the changes in PTM site quantity from the changes in protein abundance in the original sample. This is achieved by performing input normalization of PTM site quantity. There is currently a need for a software tool that supports input normalization and PTM site occupancy calculations using label-free data-independent acquisition (DIA) measurement of samples which provide simple, deep, and accurate quantification of PTM sites. In this abstract, we present the novel support for input normalization and PTM site occupancy in Spectronaut® for label-free DIA data, in addition to the existing PTM localization support. Methods: Two DIA phosphoproteome datasets were obtained from a public repository (1). The first dataset used to validate input normalization consists of two mixtures of yeast and human cell extract in 0.25:1 and 2:1 ratios (yeast:human), enriched with Ti-IMAC beads and measured by DIA. The second dataset used to validate PTM stoichiometry consists in enriched yeast phosphopeptides treated with phosphatase or untreated. Phosphopeptides and phosphatase-treated phosphopeptides were mixed in knowns ratios at 0.1%, 1%, 10%, 50%, 90%, 99% and 99.9% stoichiometry with a background of human cell extract. For both datasets, we analyzed the datasets with default settings and phospho (STY) in Spectronaut® 19. Runs were analyzed once to generate reference saved experiments, which were used as input normalization in a second analysis using "All peptides" as input normalization strategy. Stoichiometry was calculated with maximum Q-value of 1% and minimum fold change of 5. Results: In the first dataset, we measured a 7.32 median fold change between the two samples for yeast PTM site quantities and a 1.03 median fold change after input normalization, effectively showing that the difference in PTM site quantities is due to the difference in protein abundance and not in PTM site stoichiometry. In the second dataset, we achieve 8.8% accuracy and 24.2% precision in the reported PTM stoichiometry compared to the expected PTM site stoichiometries. Conclusion: The new support for input normalization and PTM site occupancy in Spectronaut® allows scientists and researchers to quickly access relevant and meaningful biological information from their enrichment experiments calculate PTM site occupancy even when protein abundance varies between samples. References (1) Bekker-Jensen D. B et al., Nature Communications 11:787 (2020) 958

959

P-III-0801 A protein sequence-based deep transfer learning framework for identifying human proteome-wide deubiquitinase-substrate interactions Dong Li1 1Beijing Proteome Research Center, Beijing, China in comparison to the well-studied E3-substrate Protein ubiquitination is a highly prevalent post-translational modification that regulates a wide range of cellular processes. The degree of protein ubiquitination is determined by the delicate balance between specific ubiquitin ligase (E3)-mediated ubiquitination and deubiquitinase (DUB)-mediated deubiquitination. These two types of interactions have been elegantly leveraged for drug development in the forms of PROTAC and DUBTAC technology. However, interactions, the DUB-substrate interactions (DSIs) remain insufficiently investigated. To address this challenge, we introduce a protein sequence-based ab initio method, TransDSI, which transfers proteome-scale evolutionary information to predict unknown DSIs based on insufficient training datasets. An explainable module is integrated to suggest the critical protein regions for DSIs while predicting DSIs. The performance of TransDSI outperforms multiple machine learning strategies against both cross-validation and independent test. Two predicted DUBs (USP11 and USP20) for FOXP3, along with two predicted substrates (AR and TP53) for USP22, are validated by our "wet lab" experiments, laying the groundwork for future investigations into the discovery of drug targets related to tumor immune evasion and the precise application of anti-tumor agents, respectively. TransDSI also provides a new perspective for disease omics data analysis by identifying regulatory DUBs for significantly dysregulated proteins. To facilitate the usage of TransDSI, we made the predicted human proteome-wide DSI dataset and the corresponding program codes available on github (https://github.com/LiDlab/TransDSI). 959

960

P-III-0802 Reproducibility-optimized multi-group statistic and survival analysis Markus Lindén1, Laura Elo1,2, Tomi Suomi1 1University of Turku and Åbo Akademi, Turku Bioscience Centre, Turku, Finland 2University of Turku, Institute of Biomedicine, Turku, Finland Bioinformaticians are faced with increasingly large and complex datasets as the costs of performing high- throughput omics are constantly coming down. Selecting differentially expressed features (e.g. genes or proteins) from these data remains vital to address biological questions, and for example to discover viable biomarkers. The methods used in differential expression analysis of omics datasets remain important, as it has been shown that the traditionally used classical methods are suboptimal when applied to high dimensional data. We have previously introduced a widely used Bioconductor R package ROTS for calculating reproducibility- optimized test statistic. It has been successfully applied to multiple types of data, including quantitative mass spectrometry proteomics and both bulk and single-cell RNA sequencing data. To further encourage the use of reproducibility-optimized test statistic in more complex settings, we have now added support for performing analysis between multiple groups and to perform survival analysis. In two-group analysis, ROTS adjusts a modified t-statistic according to the underlying data. In multi-group and survival settings, ROTS uses modified f- statistics and Cox score test, respectively. The new statistical tests were benchmarked using gold standard spike-in datasets and clinical gene expression data of breast cancer patients. ROTS offered better or comparative performance to other state-of-the-art methods. The updated software is freely available from Bioconductor. 960

961

P-III-0803 Semares – a semantic and FAIR technology framework for proteomics and multi-omics data integration, analysis and visualization Thomas Lingner1, Markus Fricke1, Muhammad Ahmad1 1Genevention GmbH, Goettingen, Germany Integrative Bioinformatics and Systems Medicine research projects usually require the management and integration of heterogeneous (pre-)clinical and experimental big data. Often, such projects involve complex data analyses from bioinformatics, machine learning or mathematical modeling. Furthermore, the FAIR principles suggest rich and harmonized annotation, controlled access, interoperability and sustainability of data and results. Finally, analyzing, exploring, sharing and discussing data and results among the project partners creates iterative cycles, which can be substantially accelerated by a homogenized digital platform with intuitive and interactive interfaces for the aforementioned tasks. Semares™ is a software technology framework designed for the configuration and development of integrative bioinformatics and systems medicine digital platforms. It supports data management and integration according to FAIR principles through a configurable semantic layer and user-friendly interfaces. The platform includes a workflow subsystem that enables the integration of complex bioinformatics analyses as push-button apps. Preconfigured and customizable interactive visualization plugins allow users to explore data and results and connect to external knowledgebases. Semares™ is easy to set up and maintain, featuring a configurable data privacy layer. It is utilized in various national and international research projects to identify biomarkers from multi-omics data, connect clinical and experimental data, and facilitate end-to-end research use cases. Fig. 1 (optional) 961

962

P-III-0804 Functional analysis of (phospho)-proteomic data with rROMA enables per-sample pathway and kinase activity quantification Stephane Liva1, Matthieu Najm1, Michael Richard1, Patrick Poullet1, Laurence Calzone1, Loredana Martignetti1 1Institut Curie, PSL Research University, Inserm U900, Paris, France Mass spectrometry based proteomic approaches have enabled the quantification of thousands of proteins and phospho-sites in a single experiment. The volume and complexity of the data generated by this technology require dedicated computational methods for converting the data into relevant biological knowkledge. An important part of proteomics data analysis is moving from focusing on individual proteins to a more integrative approach that looks at the entire system. This shift emphasizes the activity of groups of related proteins rather than just the differences in individual protein abundance. Here, we present the rROMA software for fast and accurate computation of the activity of sets of features with coordinated abundance. Sets of features can correspond to proteins with the same functional activities, proteins belonging to the same signaling pathway or target phospho-sites of a common kinase. Based on known protein assignments to pathways and kinase-subtrate interactions, it assesses the extent to which a particular biological pathway is active or dysregulated in each individual sample. Subsequently, these samples are analyzed and stratified according to their profiles across these active signalling pathways and kinases. The rROMA algorithm has been demonstrated to successfully identify disease-related active signaling pathways using transcriptomic and proteomic data [1]. It is an open-source package available on github at: www.github.com/sysbio-curie/rROMA. Here, we applied rROMA to multiple publicly available proteomic and phospho-proteomic breast cancer datasets, to characterize tumor subtypes in terms of signalling and kinases activity. Results indicate that rROMA performed well with the proteomic and phosphoproteomc datasets, revealing distinct clusters of breast cancer samples and higher disease heterogeneity compared to clinical stratification based on the PAM50 subtypes. The pathways identified as active by rROMA within these clusters are consistent with established knowledge regarding signaling pathways relevant to the subtypes, such as high activation epithelial to mesenchymal transition and inflammatory pathways in the basal subtype, ATM kinase dysregulation in basal tumors, high activation of estrogen response in the luminal subtype. Furthermore, it highlights new pathways and kinases that appear deregulated in multiple datasets, uncovering intriguing proteins that warrants further investigation. [1] Representation and quantification of module activity from omics data with rROMA. Najm M, Cornet M, Albergante L, Zinovyev A, Sermet-Gaudelus I, Stoven V, Calzone L, Martignetti L. NPJ Syst Biol Appl. 2024 Jan 19;10(1):8. 962

963

P-III-0805 Functional analysis of MS-based proteomics data: from protein groups to networks Marie Locard-Paulet1, Nadezhda T. Doncheva2, John H. Morris3, Lars Juhl Jensen2 1Institut de Pharmacologie et de Biologie StruUniversité Toulouse III – Paul Sabatier (UT3), Toulouse, France 2Novo Nordisk Foundation Center for Protein Research, University of Copenhagen, Copenhagen, Denmark 3Resource on Biocomputing, Visualization, and Informatics, University of California, San Francisco, CA, United States In high-throughput mass spectrometry (MS), proteins are digested into peptides and the peptide MS signals are then used to infer protein relative quantities across samples. Proteins that cannot be unambiguously distinguished based on the available set of peptides are reported as protein groups containing several protein accessions. However, typical follow-up analysis such as gene-set enrichment and protein interaction networks are based on gene-level annotation. Thus, they can only be performed on single proteins or genes, rendering such analysis incompatible with protein group outputs. Currently, there is no best practice on how to handle this and its impact on functional analysis has not been studied yet. Here, we investigate the composition of protein groups identified in 14 published proteomics data sets, including deep proteomes, phosphoproteomics data, single-cell proteomics, and pull downs from different species. We show that gene-set enrichment and network analysis can be affected to a different extent by the choice of which single protein is selected from each protein group, and that this selection should not be overlooked. To this end, we developed the Cytoscape app Proteo Visualizer that can complement the widely used stringApp by creating STRING networks from protein groups input instead of single protein accessions. In the resulting networks, each protein group is represented as a single node that inherits all existing edges of the group members, without discarding information from any protein accession. This app opens new avenues for performing network analysis with protein groups from bottom-up MS studies. 963

964

P-III-0806 PhosX: data-driven kinase activity inference from phosphoproteomics experiments Alessandro Lussana1, Evangelia Petsalaki1 1EMBL – EBI, Cambridge, United Kingdom Methodologies to infer differential kinase activities from global phosphoproteomics datasets have provided important insight into kinase functions in health and disease, and have catalysed the discovery of new kinase- substrate associations, the description of drug mechanisms of action and many other discoveries. However, these methods rely on previously annotated kinase-substrate associations. Given that the vast majority of identified phosphosites have no known upstream kinase and that approximately a third of human kinases have none or less than 5 known substrates, to date, these methods have been blind to large parts of the human signalling networks. With the recent availability of unbiased kinase substrate specificity maps for all human kinases, we are presented with the exciting opportunity to study kinase activities in a purely data-driven way. However, robust and appropriately benchmarked methods to take advantage of these kinase specificity maps for kinase activity inference are still lacking. Here we present such a method, called PhosX. Our method performs better than the best method that was based on known kinase-substrate associations with the added benefit that it can infer activities for all kinases. PhosX also performs better than the only other method to date that can use these kinase specificity maps. The addition of machine-learned kinase-substrate relationships and the integration of graph-based algorithms is currently ongoing to further improve PhosX's differential kinase activity inference. We expect PhosX to become an important tool in the data-driven studies of human cell signalling, allowing discoveries relating to context-specific functions of well-studied kinases but also shedding light on the vast understudied signalling space. Fig. 1 (optional) 964

965

P-III-0807 Modeling multipartite networks integrating high-throughput phospho-proteomics (ModPhosphoNet) Anne-Claire Kroger1, Marco Ruscone1, Michael Richard1, Stephane Liva1, Patrick Poullet1, Laurence Calzone1, Loredana Martignetti1 1Institut Curie, PSL Research University, Inserm U900, Paris, France Scientific background: Network models are a common and powerful formalism for studying cell functioning and its deregulations during cancer. They allow the integration of a large number of entities into a single complex network. However, the analysis of these large networks remains a challenge. Notably, not many integrative studies include phosphoproteomics data, an important layer of information for understanding signaling pathways, due to the challenges in analyzing phosphoproteome data. In Medulloblastoma (MB), the most common malignant pediatric brain tumor, we have revealed that key signaling pathways are controlled at the post-transcriptional level in two out of four disease subgroups [1]. Thus, we propose here a novel integrative and modeling strategy to accurately integrate information about phosphoproteome and its potential functional consequences. Results: Two main computational approaches have been explored: (i) stochastic blockmodeling (SBM) of multipartite networks and (ii) dynamical simulation of multi-layer networks. Blockmodeling is an increasingly popular class of models in statistical analysis of graphs that can be used to discover or understand the latent structure of a network, as well as for clustering purposes [2]. Each sample and each gene have as an output a certain probability to belong to a given block. This can be interpreted as a classification of patients into molecular similar clusters (subgroups) on the basis of similarly expressed genes (signatures). The output of a stochastic block modeling algorithm is a probability distribution of membership, i.e., the probability of a given sample to be associated to a subgroup and, at the same time, the probability of a gene to characterize a given subgroup. This mixed probabilistic membership is a major advantage of stochastic block modeling with respect to standard clustering approaches. First, we proposed to use SBM algorithms developed for multipartite graphs to identify clusters of patients based on their connection behavior in multiple molecular networks (proteomics, phosphoproteomics and transcriptomics). The output of the SBMs analyses have been interpreted as weighted signatures per MB subgroups encompassing transcription factors, proteins and phosphorylated states of kinases. Second, we used this information to build network models specific for each disease subgroup. These networks can be used as basis for mathematical modeling and dynamical simulations to predict possible points of intervention. [1] Forget A et al, Aberrant ERBB4-SRC Signaling as a Hallmark of Group 4 Medulloblastoma Revealed by Integrative Phosphoproteomic Profiling. Cancer Cell 10;34(3):379-395.e7 (2018) [2] Lee C et al, A review of stochastic block models and extensions for graph clustering. Applied Network Science 4(1)(2019) 965

966

P-III-0808 A Common Data Analysis Pipeline (CDAP) for data independent acquisition mass spectrometry data Aaron Maurais1, Michael Riffle1, Nathan J. Edwards2, Gennifer E. Merrihew1, Brian Connolly1, Matthew Chambers1, Brendan MacLean1, Julia E. Robbins1, Ratna Thangudu3, Brian Searle4, Christine Wu1, Paul Rudnick5, Michael J MacCoss1 1University of Washington, Genome Sciences, Seattle, WA, United States 2Georgetown University, Department of Biochemistry and Molecular & Cellular Biology, Washington, DC, United States 3ICF International, Reston, VA, United States 4Ohio State University, Department of Chemistry and Biochemistry, Columbus, OH, United States 5Spectragen Informatics, Bainbridge Island, WA, United States We've developed a fast and scalable NextFlow pipeline for the analysis of large Data Independent Acquisition (DIA) datasets. The pipeline starts from vendor raw files which can be stored on either PanoramaWeb, the NCI Proteomic Data Commons (PDC), or a local workstation. The raw files are downloaded if necessary and converted to mzML using MSConvert. Each file is analyzed with EncyclopeDIA in parallel then merged into a single library. The merged library and mzMLs are then imported into a Skyline document for quantification and visualization. Peptide and protein or gene level matrices are exported from the Skyline document and used to generate a quality control (QC) report to enable an assessment of data quality. To run the pipeline, the only software which must be installed is the NextFlow engine. The docker containers necessary to run each step are downloaded on the fly as the workflow is running. The pipeline can be run either in the cloud with AWS Batch or on a local workstation. We illustrate the utility of the pipeline with three use cases. In the first use case the pipeline is used to download and analyze raw files and metadata from several DIA studies in the PDC. To facilitate data harmonization between existing DDA studies, gene level reports are automatically generated in the same format used by the DDA CDAP. In the second use case, lysates from the NCI-7 panel of reference cancer cell lines were sent to 16 groups in the International Cancer Proteogenome Consortium (ICPC). The pipeline was used to analyze data from 11 of the 21 datasets which were acquired by DIA. After normalization and batch correction, each cell type clustered together, regardless of the lab the samples were acquired in. A third use case was to analyze data for a challenge from the Intelligence Advanced Research Projects Activity (IARPA) with 745 samples. 966

967

P-III-0809 Molecular complex navigator: a new way to visualise related complexes in complex portal Juan Jose Medina Reyes1 1EMBL – EBI, Cambridge, United Kingdom The EMBL-EBI Complex Portal is an encyclopaedic resource providing high quality, manually curated, functional biomolecular complexes. Those complexes consist mainly of proteins but they may also contain small molecules, nucleic acids and subcomplexes. The Molecular Complex Navigator facilitates the visualisation of large sets of related complexes in a context-specific manner. The Complex Navigator is implemented as a matrix, with the complexes represented as rows, and the interactors as columns. Users can access detailed information on both complexes and interactors via clickable links on their names. Selecting a complex link directs users to its detailed Complex Portal page, while selecting an interactor link provides access to a page with information on the interactor, which could be from UniProt, ChEBI or Complex Portal page depending on the interactor type. The matrix also features an expandable menu to view subcomplexes" interactors, and each matrix cell displays the stoichiometry of the complex interactors. The Molecular Complex Navigator is designed to simplify and enhance the visualisation of multiple closely related complexes and their interactors. 967

968

P-III-0810 A pipeline for designing proteogeno-typic peptide standards to quantify variants: the potential of neutrophil elastase in proteogenomics Karl A. T. Makepeace1, Pallab Bhowmick1, David Goodlett1, Peter van Veelen2, Christoph H. Borchers3, Yassene Mohammed2,3 1University of Victoria, Victoria, Canada 2Leiden University Medical Center, Center for Proteomics and Metabolomics, Leiden, Netherlands 3McGill University, Montreal, Canada Proteogenomics integrates additional genomic information with proteomic characterization to enable the identification of protein variants specific to a genotype. To that end, proteogenomic characterization of, for example, a human sample, is the personalized proteomic analysis of the sample based on the genomic data of that person. We have developed Proteogenomic Improved and Guided Quantification Pipeline – PIGQpipe, a software tool to design targeted proteogenomic experiments. PIGQpipe draws upon our experimentally-derived assay development experience and integrates that with data from public online databases to facilitate targeted proteogenomics assay design. PIGQpipe features interfaces for Genomic Data Commons (GDC) and Ensembl data. GDC curates and provides access to somatic mutations identified in cancer patients, while Ensembl documents a broader range including both germline and somatic mutations, across multiple species (Figure 1). Our tool facilitates also analysis of COSMIC Cell Lines Project (CLP) data. To process the genomic data, we developed a set of routines packaged as an R library. This library features parsing functionality for Human Genome Variation Society (HGVS) nomenclature, enabling the prediction of potential protein consequences from standardized genetic variant descriptions. This is crucial, as variants like frameshift, termination, and untranslated region (UTR) variants, require more advanced handling logic than simple single amino acid substitution. This feature allows our pipeline to integrate with various databases and pipelines focused on genomic variants. Using the Jurkat cell line with TP53 as a target protein and considering three proteases (glutamyl endopeptidase; neutrophil elastase; trypsin) analysis takes approximately 20 seconds on a consumer laptop to produce all possible theoretical targeted assays that can be used to quantify known TP53 mutations specific to that cell line. We have processed all known human germline mutations from Ensembl to generate an overview, and we have also processed all somatic mutations in all 1000+ sequenced cell lines available for research, as hosted by the COSMIC knowledgebase (Figure 1). The results suggest to use neutrophil elastase to increase coverage of detecting mutation consequences on the protein level. To validate PIGQpipe, we reanalyzed publicly available deep proteomics datasets from 9 cell lines representing the 9 cancer tissues of the NCI-60 [PMID: 23933261]; available from PRIDE. We have also acquired in-depth proteomics datasets from two melanoma patient-derived cells using trypsin and neutrophil elastase as proteases. The experimental data affirmed the presence of protein variants predicted by our pipeline. Although our goal has focused on protegeno-typic peptides for targeted proteogenomics assays, PIGQpipe is generic in its use and can help general proteogenomics data analysis. 968

970

P-III-0811 PTMNavigator: interactive visualization of differentially regulated post-translational modifications in cellular signaling pathways Julian Müller1, Florian P. Bayer1, Bernhard Küster1, Mathias Wilhelm2, Maximilian Schuh3, Matthew The1 1Technical University of Munich, Chair of Proteomics and Bioanalytics, Freising, Germany 2Technical University of Munich, Computational Mass Spectrometry, Freising, Germany 3Technical University of Munich, Chair of Organic Chemistry II, Garching, Germany Introduction: A central element of cellular signaling are protein post-translational modifications (PTMs). Upon a perturbation such as a drug treatment, many proteins undergo a change on PTM level, which can have fundamental consequences for the phenotype. While PTM data nowadays is routinely acquired, it is still hard to interpret the regulated PTMs in the context of signaling cascades. Here, we present PTMNavigator, a web-based software tool to explore PTM proteomics datasets from a pathway-centric viewpoint. Our software is able to render pathway diagrams interactively and project experimental PTM data onto them. A number of enrichment analyses can be performed to identify the most relevant pathways, gene sets, and kinases for a dataset. The tool is hosted on ProteomicsDB (proteomicsdb.org/analytics/ptmNavigator), and users can either browse through what is available there or upload their own data. Methods: PTMNavigator was developed as a Vue.js application. The core visualization component is a WebComponent (github.com/kusterlab/biowc-pathwaygraph). A set of canonical pathway diagrams was created in ProteomicsDB by importing pathways of 10 model organisms from KEGG and WikiPathways. To perform enrichment analyses, a Flask-based web service was implemented. PTMNavigator integrates the visualization and enrichment results together with additional information (e.g. from PhosphoSitePlus) in a web platform. Results: Users of our tool can choose from the ~3,000 available pathway diagrams and project their PTM data onto them. In addition, they can create their own diagrams to visualize bespoke interactions. Multiple functional enrichment algorithms (e.g. PTM-SEA, KSEA) are automatically run after data upload and the results are summarized in PTMNavigator. The results are also used to score and rank the canonical pathways to facilitate manual inspection. Optionally, kinase-substrate relationships from PhosphoSitePlus can be visualized as additional links in the pathway diagrams. Also, the software PHONEMeS is used to construct a data-driven signaling network using prior knowledge from STRING and PhosphoSitePlus. Visualizations can be exported to create publication-level figures with ease. Using a number of recently published datasets, we demonstrated some applications of our software. First, we showed how pathway visualization and enrichment analysis can be combined to find out which pathways are actually relevant for a dataset. Second, we visualized how kinase inhibitors with different targets perturb the PI3K/AKT/mTOR phospho-signaling cascade at different stages. Third, PTMNavigator helped us to extend an existing pathway by uncovering previously unknown relationships between PTMs and pathway components. Conclusion: We present an easy-to-use web tool for the study of PTM signaling networks. To our knowledge, PTMNavigator is the first tool that enables comprehensive and in-depth pathway-centric analyses of PTM datasets within a single application. 970

971

P-III-0812 KPOP: Korea ProteOme rePository for open access proteomics data Seungjin Na1, Hokeun Kim2, Jingi Bae2, Heeyoun Hwang1, Shinyeong Ju3, Cheolju Lee3, Jin Young Kim1, Sang- Won Lee2, Eunok Paek4 1Korea Basic Science Institute, Cheongju, South Korea 2Korea University, Seoul, South Korea 3Korea Institute of Science and Technology, Seoul, South Korea 4Hanyang University, Seoul, South Korea With the surge in mass spectrometry (MS)-based proteomics, there is a growing abundance of data from proteomics projects worldwide. Since 2012, the ProteomeXchange (PX) consortium has played a key role in standardizing the submission and dissemination of public MS proteomics data. To promote the global sharing and reuse of proteomic resources, we introduce KPOP repository (https://kbds.re.kr/KPOP), a public platform facilitating open access to a wealth of proteomic resources, encompassing various types of MS data, analysis results, and metadata. KPOP has implemented user-friendly interfaces, accelerated file transfers, and flexible file management, significantly enhancing the usability and value of proteomics data for scientific research. In addition to sample The primary goal of KPOP is to curate meticulously maintained data sets, supporting prevalent "big data" approaches in proteomics, including machine learning and computational modeling. These approaches enable researchers to decipher complex data patterns and associations from proteomics data, driving innovation in the field. information, KPOP retains metadata related to fractionation, digestion, quantification, modifications, experiment type, and the MS instrument employed. KPOP is a member of K-BDS (Korea BigData Station), which serves as a comprehensive repository for various biological research data, promoting integrated "multi-omics" analyses. This integration is particularly beneficial when datasets, such as whole exome and RNA sequencing data, are simultaneously available, originating from the same project. Researchers can combine proteomics data with other omics data to gain deeper insights into biological processes and disease mechanisms. KPOP complies with PX Consortium"s guidelines, allowing registered users to submit their proteomics data in public or private modes. Private mode allows sharing with designated users, ideal for pre-publication collaboration or peer review. Public mode offers unrestricted access to datasets. KPOP"s structure is user- centric, organizing data by projects and data sets, with each set encapsulating a distinct experimental setup, such as biological replicates, technical replicates, or clinical samples. In accordance with PX guidelines, KPOP supports two submission workflows, Complete or Partial submission. 971

972

P-III-0813 Liver proteome analysis in 111 knock-out mouse models provides crucial references for the function of genes influencing metabolic processes Areum Nam1,2,3, Heeyoun Hwang4, Sanghyuk Yoon1,2, Hoonyoung Yoon1,2, Haeju Lee1,2, Yeo-Jin Im1,2, Je-Yoel Cho1,2 1Seoul National University, Comparative Medicine Disease Research Center, Seoul, South Korea 2Seoul National University, College of Veterinary Medicine, Seoul, South Korea 3Seoul National University, Graduate School of Public Health, Seoul, South Korea 4Korea Basic Science Institute, Digital OMICs Research Center, Cheongju, South Korea The liver serves as a metabolic orchestra, playing a pivotal role in fundamental homeostasis, including lipid, carbohydrate, and protein metabolism. Understanding how the liver copes with overloaded metabolic processes is challenging due to the numerous factors involved in the metabolic pathway. We conducted a large-scale isobaric labeling proteomics analysis using TMT on liver tissues from 111 knock-out mouse models. From this liver proteomics profiling, we discerned the informative relationships between each of the 111 genes (randomly selected by KMPC) and metabolism. Using differentially expressed proteins, we established liver-specific gene scores, named the 'Liver Impact Factor'. This factor indicates the degree of a gene's relevance to the liver. To verify the specific metabolic phenotype present in the knock-out mouse model, we turned to clinical blood chemistry analysis data. Both the proteomics data and the clinical blood chemistry data suggested a focus on cholesterol and lipid metabolism in the knock-out mouse models. We identified differentially expressed proteins (DEPs) associated with cholesterol metabolism. We performed a correlation analysis between protein expression and serum cholesterol levels to further pinpoint candidate target proteins related to cholesterol metabolism. The resulting data offers a reference for normal and abnormal mouse model liver proteomes. This correlation between liver-associated metabolic values and protein expression provides insights into potential candidate proteins related to these metabolic processes. 972

973

P-III-0814 Are we there yet? Assessing the readiness of single-cell proteomics to answer biological hypotheses Alyssa A Nitz1, Jose Humberto Giraldez Chavez1, Zachary G Eliason1, Samuel H Payne1 1Brigham Young University, Biology, Provo, UT, United States Single-cell measurements are crucial in investigating biological hypotheses that involve cell-to-cell heterogeneity. Bulk measurements, generated with hundreds or thousands of cells, represent sample averages and can obscure functionally important cell-to-cell differences. Mass-spectrometry-based single-cell proteomics (MS-based SCP) has seen a lot of work in method development to make it a viable investigative option. Improvements in sample preparation and methods for data acquisition have led to greatly-increased proteomic depth from a single cell (~3,000-5,000 proteins per cell). Our goal was to assess the readiness of SCP to participate in biological investigations, such as: in a sample, even rare cells (cell annotation); reconstructing trajectories representing a continuous arc of development for a cell type of interest (developmental trajectory inference); and analyzing cells in relation to their location in a microenvironment (spatial mapping).We identified sample throughput as the primary limitation of SCP. Additionally, we identified important experimental design and data analysis considerations specific to the SCP data type, including: sample size, proteomic depth, batch effects, double dipping, assumptions of available analysis tools, and spatial resolution tradeoffs. identifying all subpopulations 973

974

P-III-0815 Single nucleus RNA sequencing analysis combined with in vitro validation in human dopaminergic neurons reveals aspects of idiopathic Parkinson's disease pathology Sofia Notopoulou1, Georgia Katsanopoulou1, Georgios Gavriilidis1, Vasileios Vasileiou1, Spyros Petrakis1, Fotis Psomopoulos1 1INAB-CERTH, Thessaloniki, Greece The number of cases of Idiopathic Parkinson's disease (IPD) has doubled in the past 25 years, with this increase expected to continue its rapid incline, as the population ages. During the disease"s onset, a variety of clinical symptoms gradually arise, including both motor and nonmotor, the latter manifesting earlier, before an observation of the motor symptoms or an exact diagnosis can be made. Nonetheless, our understanding of the exact molecular mechanisms associated with the generation and progression of this disease has yet to be fully understood. To gain detailed insights into IPD cellular pathology, we utilized a publicly available dataset and analyzed 41,435 single-nuclei transcriptomes of post-mortem midbrain from IPD patients (n=6) and healthy donors (n=5). Overall, alteration in a plethora of subtypes was observed, with oligodendrocytes, microglia, and dopaminergic neurons (DANs) being the more substantial ones. Differential analysis of activated transcription factors revealed dysregulation of dopaminergic-related TFs in IPD samples, with NR4A2 being one of the most evident ones. Similarly, Trail and p53 pathways were significantly altered, indicating a malfunction in pathways associated with cell death. To validate these results in vitro, we generated an IPD-like cellular model of human DANs that inducibly expressed the EYFP-SNCA WT transgene. Specifically, human neural progenitor cells (hNPCs) were stably transfected with the TetOn EYFP-SNCA WT cassette and directly differentiated into midbrain dopaminergic neurons. Interestingly, the generated DANs exhibited high levels of tyrosine hydroxylase (TH), dopamine transporter (SLC6A3), and vesicular monoamine transporter 2 (SLC18A2). Additionally, they were characterized by the absence of increased expression of Aldehyde Dehydrogenase 1 (ALDH1A1), a pattern specific to IDP-prone DaNs. After inducible expression of EYFP-SNCA WT for 5 days, DaNs exhibited a significant activation of apoptosis as well as impairment in the patterns of transcription factors involved in dopaminergic neuron maintenance, particularly that of NR4A2, which was also found to be dysregulated in our computational analysis. Interestingly, in silico knockout of targeted key players had a significant impact on IPD-related neurodegeneration, indicating potential therapeutic interventions. Our findings propose a potent workflow where a combination of snRNA-seq analysis along with the in vitro validation in a reliable cellular model might reveal crucial aspects of IPD pathology. lipase (LPL) and the lipoprotein 974

975

P-III-0816 The UniProt KnowledgeBase – supporting proteomics research Sandra Orchard1, Emily Bowler-Barnett1, Jun Fan1, Maria Martin1, UniProt Consortium1 1EMBL – EBI, Cambridge, United Kingdom The UniProtKB database (www.uniprot.org) is the world's leading high-quality, comprehensive and freely accessible resource of protein sequence and functional information. A proteome comprises of all the proteins produced by the transcripts translated from a genome, including isoforms and proteins/peptides produced by cleavage of the initial transcript into smaller peptides and polypeptides. For each experimentally-studied protein in a proteome UniProtKB presents the user with a summary of experimentally verified, or computationally predicted, functional information added by our expert biocuration team. This includes additional modifications in both sequence and function caused by nonsynonymous variants and posttranslational modifications. Computationally transferred or predicted annotations are added to those proteins which have not been functionally studied (unreviewed entries). We present here revisions to our proteomics peptide mapping pipeline, in line with HPP3.0 guidelines, a summary of recent updates to the UniProt resource, including visualization of proteins in the context of protein complexes in collaboration with the Complex Portal, and an overview of the current state of the human proteome. 975

976

P-III-0817 SimpliFi in a box: multiomics data analysis on private servers Jim Palmeri1, John Wilson1 1ProtiFi LLC, Fairport, NY, United States Introduction: The rapid growth of cohort sizes and multiomics analyses generates increasingly vast amounts of data. The analysis and interpretation of this data, along with the ability to precisely replicate analyses and share results, is crucial. To address these needs, we developed SimpliFi (simplifi.protifi.com), a cloud-based, browser- driven data-to-meaning engine designed with an intuitive, user-friendly interface accessible to users of all experience levels. Additionally, we introduce SimpliFi Box, a private deployment of SimpliFi for organizations requiring stringent data privacy and security for internal network use or secure virtual private cloud environments. Methods: SimpliFi models biology using exclusively nonparametric statistics, with biological replicates defining their own distributions. P-values and fold-changes are determined based on biological variation, sample numbers, observations, and measurement error. SimpliFi accepts and integrates all types of omics data and provides confidence intervals for all values, including p-values. Importantly, SimpliFi does not transform data and accommodates increased data variance at low or high intensities. Its user interface allows for the efficient analysis of large sample sets. Projects can be shared via private or public URLs, facilitating collaborative data exploration. Preliminary Data: SimpliFi employs non-parametric statistics where biological replicates yield their own empirical distributions, resulting in p-values that can differ significantly from those derived from parametric tests like T- tests. This is due to the non-Gaussian nature of biological and omics data. Sample size variations, over- or under- sampling of variability, and outliers can lead to false negatives or positives. SimpliFi presents analyses through interactive displays of pathways, tissue states, diseases, cells, and molecular-level classifications. Front-end displays are optimized for large cohorts; quality control features automatically flagging unusual samples. Users can filter analytes (proteins, metabolites, etc.) by various parameters to quickly identify biomarkers or other significant analytes, and then perform pathway and other analyses on the selected subgroups. Differential expression analysis is available at a population-wide level, with visualizations such as distributions, violin plots, and box-and-whisker plots enabling immediate understanding of experiments with hundreds to thousands of samples. Graphs and diagrams can be saved as vector or raster images with a right- click, and customized reports can be generated easily. SimpliFi operates on a cloud-based model, allowing users to upload data to a server equipped with GPUs for rapid statistical analysis. However, with SimpliFi Box, organizations can run SimpliFi on-site, ensuring that private data remains on premises. Projects can be shared internally, enabling users to explore data collaboratively while accessing all features. 976

977

P-III-0818 Complete quantitative data-independent acquisition mass spectrometry workflow using Nextflow Mats Daniel Perk1, Sami Pietilä1, Tommi Välikangas1, Tomi Suomi1, Laura Elo1,2 1University of Turku and Åbo Akedemi University, Medical Bioinformatics Centre, Turku, Finland 2University of Turku, Institute of Biomedicine, Turku, Finland A key high-throughput proteomics technology to study proteins is mass spectrometry proteomics, whose data analysis involves the identification and quantification of proteins. In data-independent acquisition (DIA) mode the mass spectrometer systematically scans over a wide range of mass-to-charge (m/z) values, resulting in more reproducible protein abundances compared to other methods, making it a reliable method for data analysis and comparison. To address the computational challenges associated with DIA data analysis, a workflow management tool Nextflow can be employed. Nextflow is particularly well-suited for proteomics data analysis workflows, as it allows for the handling of complex and dynamic workflows in a reproducible and scalable manner. In this context, we introduce glaDIAtor-nf, a free and open-source proteomics data analysis tool powered by Nextflow, which aims to assist researchers in the field of proteomics. To demonstrate its capabilities, we showcase the performance of glaDIAtor-nf using publicly available datasets, including gold- standard spike-in datasets and real-world biological datasets. 977

978

P-III-0819 Uniquome: construction and decoding of a novel proteomic atlas that contains new peptide entities Vasileios Pierros1, Evangelos Kontopodis1, Dimitrios Stravopodis2, George Tsangaris1 1Biomedical Research Foundation of the Academy of Athens, Proteomics Research Unit, Athens, Greece 2National & Kapodistrean University of Athens, Section of Cell Biology and Biophysics, Department of Biology, Athens, Greece Cellular and molecular uniqueness has recently gained eminent importance, due to the large amount of data produced by "-omics" technologies. Herein, we have constructed and decoded the "Uniquome", by introduction of the new peptide entities: (a) "Core Unique Peptide" (CrUP), defined as the peptide whose sequence is accommodated, specifically and exclusively, only in one protein in a given proteome, and also bears the minimum length of amino acid sequence; and (b) "Composite Unique Peptide" (CmUP), defined as the peptide composed by the linear unification of CrUPs, when two or more successive in order CrUPs overlap one another. We have analyzed the human reviewed proteins that are numbered 20,422 in all. We found that 7,271,607 peptides from 4 to 100 amino acids in length, were characterized as CrUPs, in 20,258 human proteins, whereas 164 reviewed proteins (0.80%) were not found to contain any CrUP. Further analysis, revealed that these CrUPs could generate 66,447 CmUPs. Analysis of CrUP structure in relation to their length (number of aa they are composed of) showed that the majority (69.03%) of human CrUPs consist of 6 aa, while 20.6 % and 8.62% of CrUPs are characterized by 5 and 7 aa, respectively. For CmUP species, the analysis unveiled that 11-peptides and 12-peptides strikingly represent the higher CmUP numbers (5.71% and 4.57%, respectively), followed by 10- peptides and 13-peptides (1.81% and 2.56%, respectively).The distribution of CrUPs onto human chromosomes proved that chromosomes exhibit CrUP-coverage values from 90% to 96%, while the mitochondrial chromosome presents 100% -respective- coverage, whereas the Y sex-chromosome is characterized by the lowest CrUP coverage of 32%. The detection of CrUPs in human proteome dictates our emerging ability to reliably identify a protein by use of only one peptide. Towards this direction, we found that 564,148 peptides of the 815,026, trypsin digest-derived, peptides of 20,258 proteins, (69.22%) contained at least one CrUP species herein defined as "Tryptic digest Unique Peptides" ("TUPs"). The detected TUPs originated from 20,136 proteins (99.39%), whereas 122 proteins were found to lack TUP species. To strengthen the biological significance of CrUPs, we unveiled that 89% of Cancer Antigenic Peptides ("CAPs") and 87% of Immune Epitope Peptides ("IEPs") contain at least one CrUP species, while it is demonstrated that 97% of bioactive peptides being included in the PeptideDB database contain at least one CrUP. Taken together, our findings strongly support the potentially beneficial applications of human CrUPs in tissue pathology, therapeutic oncology, and translational medicine. Furthermore, these novel peptide entities seem to play a crucial role for protein identification, protein-function prediction, cell physiology and tissue pathology. 978

979

P-III-0820 OutCyte UPS 2.0: a representation learning based new tool for leaderless protein secretion prediction Jan van Grimbergen1,2, Sara Schulte2, Gunnar Klau2, Kai Stühler1, Gereon Poschmann1 1Heinrich Heine University Düsseldorf, Institute of Molecular Medicine Proteome Research, Duesseldorf, Germany 2Heinrich Heine University Düsseldorf, Algorithmic Bioinformatics, Duesseldorf, Germany To communicate with their environment, cells developed different strategies to release proteins into the extracellular space. Classical protein secretion via the endoplasmic reticulum / Golgi route is based on N-terminal signal peptides. While prediction of signal peptides and classically secreted proteins has been well established by several groups, the prediction of leaderless secreted proteins is still challenging because a clear sequence motif is lacking and the number of proven leaderless secreted proteins is limited. One available tool to overcome these challenges is OutCyte, which has previously been developed by our group. OutCyte is a feature-based deep- learning method where the features are calculated from the protein sequence. To overcome the limited number of available leaderless secreted proteins, OutCyte used experimental secretome data for training. Here, we present OutCyte UPS 2.0, a new version of OutCyte with two major improvements: First, we extended the source of experimentally proven secreted proteins from 10 to 53 different cell types. Second, instead of using a feature-based model in which relevant features might not be included, we developed an artificial neural network based on a representation learning approach for which no features are necessary. For this, we used representations from the transformer protein language model ESM2 (evolutional scale model). Using 5-fold cross-validation, we observed an F1 score of 0.64 for the previous OutCyte version, which was surpassed by OutCyte UPS 2.0 with a score of 0.83. Out of 18 proteins, described as leaderless secreted in the literature, our model is able to predict 15 correctly. OutCyte UPS 2.0 is publicly available at www.outcyte.com. 979

980

P-III-0821 Integrating quantitative proteomics and pharmacoproteomics for reliable biomarker discovery in Alzheimer's disease Murih Pusparum1,2, Gökhan Ertaylan1 1VITO Belgium, Unit Health - Environmental Intelligence, Mol, Belgium 2Hasselt University, Data Science Institute, Hasselt, Belgium Quantitative proteomics has emerged as a pivotal technique in precision medicine, offering significant potential to support clinical diagnostics and disease treatments. Despite its promise, the clinical validation of protein biomarkers remains costly, limiting their integration into routine clinical practice. Addressing this gap, our study presents a novel data-driven approach designed to enhance the discovery and validation of protein biomarkers, making the process more efficient and cost-effective. In our study, the proposed data-driven approach combines parametric and nonparametric statistical techniques and machine learning algorithms to process high-dimensional quantitative proteomics data. We propose reference intervals for plasma proteomics value, critical for identifying deviations associated with disease states. We validate our method using a case study on Alzheimer's disease (AD) in the UK Biobank. We also performed pharmacoproteomics analysis, where we focus on individual proteomics profiles in response to medications. This ensures the relevance and reliability of the identified biomarkers, as medication data can precisely indicate the disease onset and progression. In addition to proteomics data, we incorporate clinical biochemistry tests focusing on lipid profiles, such as apolipoproteins and cholesterol levels, which are known to be associated with a higher risk of AD. This holistic approach allows us to explore the interplay between protein and lipid profiles. Our study identified five candidate protein biomarkers for AD. The individual protein profiles of these biomarkers are aligned with both medication responses and lipid profiles, providing a comprehensive understanding of their potential clinical relevance. The integration of lipid profile data further substantiates the link between these biomarkers and AD risk. The proposed data-driven approach offers a rapid, efficient, cost-effective alternative for advancing proteomics biomarker studies. Our findings highlight the potential of this approach to revolutionize the field of precision medicine, particularly in the context of AD, facilitating the transition of protein biomarkers from research to routine clinical practice. 980

981

P-III-0822 Leveraging MD 2.0 to support proteomics-based biomarker discovery Anna Quaglieri1, Aaron Triantafyllidis1, Bradley Green1, Mark Condina1,2, Paula Burton1, Giuseppe Infusini1, Andrew Webb1,3,4 1Mass Dynamics, Melbourne, Australia 2University of South Australia, Clinical & Health Sciences, Adelaide, Australia 3WEHI, Melbourne, Australia 4University of Melbourne, Department of Medical Biology, Melbourne, Australia Typical proteomics-based Biomarker discovery approaches aim to define a small subset of biological markers able to predict biological groups of interest. While biomarker discovery spans a broad range of applications, a key use is the detection of markers able to signal the early onset of a disease, increasing the potential to inform and highly improve the treatment of patients. Despite the potential of detecting early diagnosis markers, the number of well-validated and clinically established markers remains low. A biomarker discovery workflow requires careful consideration when defining the experiment design, the data preparation, the downstream bioinformatics analysis, and integration with knowledge databases for external validation. Furthermore, validation of the defined set of markers across independent datasets is required to prove their quality and biological reproducibility. Therefore, once proteins or peptides have been identified and quantified, a number of quality control, statistical analyses, and validation steps routinely follow. However, these are often developed for and by computational biologists. Mass Dynamics 2.0 (MD 2.0) exists to make current and new computational methodologies accessible to all researchers in an intuitive modular, web, and cloud-based collaborative environment. In addition, MD 2.0 enables seamless results comparison and multi-dataset management so that a researcher can easily organize and access their multiple discovery and validation datasets and analyses in one place. Leveraging publicly available cancer proteomics datasets, we showcase the end-to-end reproducible biomarkers discovery analysis suite in MD 2.0. This includes: 1) interactive modules for quality control, including PCA, t-SNE, and heatmaps to verify the similarities between the biological groups of interest. 2) Different statistical approaches to define a subset of discriminative markers between conditions of interest using independent approaches. These methods include differential expression analysis, discriminant analysis, and penalized logistic regression. 3) Interactive upsets plot to easily define intersections between sets of interest and ROC curves to validate the predictive power of one or subsets of the defined markers. The workflow presented here highlights the ability to leverage MD 2.0 as a platform for guided assessment of multi-proteomics data to improve the confidence in identifying biomarkers for further assessment. The platform provides a collaborative space that improves the ability for research teams to make key insights into complex proteomics data and lowers the ability for all stakeholders to access and compare various processing approaches and the influence of metadata on results. 981

982

P-III-0823 AlphaFold3 and protein modifications: obvious strengths and significant challenges Pathmanaban Ramasamy1,2, Jasper Zuallaert1, David Bickel2, Lennart Martens1, Wim F. Vranken2 1Ghent University, Center for Medical Biotechnology, Department of Biomolecular Medicine, Ghent, Belgium 2Vrije Universiteit Brussel, Department of Bio-engineering Sciences, Brussels, Belgium Proteins are fundamental macromolecules in cells, serving a diverse range of functions. Understanding how cells orchestrate transitions between states is crucial for understanding organismal development. Proteins undergo various post-translational modifications (PTMs), such as phosphorylation, acetylation and ubiquitination, which dynamically alter their structure, function, folding, and interactions. Dysregulation of these modifications can disrupt cellular homeostasis, leading to diseases such as cancer and Alzheimer's, and affecting protein proteostasis. Despite their significance, our understanding of the structure-function relationship of proteins and PTMs is very limited due to limited experimental data on protein structures, the diversity of PTMs, and the dynamic nature of proteins. Deep-learning methods such as AlphaFold and RoseTTAFold have revolutionized protein structure prediction and design, yet their current capabilities are limited to specific systems like covalent modifications and mutations. The latest releases, RoseTTAFold-All-Atom and AlphaFold 3, demonstrate a significant increase in accuracy in predicting protein complexes, including nucleic acids, small molecules, and post-translational modifications. In this work, we highlight the strengths, weaknesses, and major challenges associated with state-of-the-art protein structure prediction methods in relation to protein modifications, using phosphorylation as a case study. 982

983

P-III-0824 Beyond the reference code: unveiling multicoding genes and their alternative proteins and microproteins Xavier Roucou1, Frédérick Comtois1, Jean-François Jacques1, Davy Ouedraogo1, Aïda Ouangraoua1 1University of Sherbrooke, Biochemistry and functional genomics, Sherbrooke, Canada In humans, protein-coding genes are typically annotated with a single functional open reading frame (ORF) or coding sequence, typically the longest ORF. The corresponding transcripts contain the reference ORF or a variant of the ORF after alternative splicing, resulting in the expression of the reference protein and different isoforms. However, there is increasing evidence that many genes have more than one ORF and produce transcripts with more than one translated ORF. There is currently no consensus on the terminology of these new proteins. They are presently labeled small ORF-encoded peptides, microproteins, or alternative proteins. In this study, we use the term "alternative ORFs and proteins" (altORFs and altProts) in comparison with reference ORFs and proteins (refORFs and refProts). The lack of annotation of altORFs and altProts represents a significant challenge to their detection and the study of their functions. Several studies and proteogenomic resources combining ribo-seq or/and proteomics have been developed to help annotate altORFs and altProts to address this issue. We created the OpenProt proteogenomic resource, which enables the identification of multiple ORFs per transcript and the annotation of all ORFs exceeding 29 codons in the transcriptome of various species, along with their corresponding proteins. The reanalysis of large-scale ribo-seq and proteomics data with OpenProt enables the retrieval of evidence of expression at the translatome (altORFs) and the proteome (altProts) levels, respectively, and the identification of potential multicoding genes. The P53-Induced Protein With A Death Domain gene (PIDD1) encodes a 910-amino-acid protein that engages with a variety of interactors and stimulates the DNA-damage response, centrosome surveillance, NFkB activation, and cell death. It has been established that several mutations in PIDD1 are linked to cerebral cortex malformations and intellectual disability. However, the underlying molecular pathophysiology remains unclear. We show here that PIDD1 is among the potential multicoding genes annotated by OpenProt with strong evidence of expression by ribo-seq and MS-based proteomics for a 171 amino acid-long alternative protein termed altPIDD1. AltPIDD1 and PIDD1 are co-expressed from two overlapping ORFs in the same transcript. The relative quantification of PIDD1 and altPIDD1 indicates that altPIDD1 is the primary product of translation, with a ratio of 1:40. This previously unknown protein is found in actin-rich cytoskeletal structures and is cleaved by caspase 3 and/or 7 during apoptosis. In contrast to other non-canonical proteins that emerged in primates and are relatively new, AltPIDD1 emerged in placental mammals. The findings of this study underscore the importance of further investigating the potential for multicoding genes, reinforcing the necessity of more detailed and accurate descriptions of the genome's coding potential. 983

984

P-III-0825 Human proteome size as a function of development of experimental technologies and bioinformatics methods Elizaveta Sarygina1, Anna Kozlova1, Elena Ponomarenko1, Ekaterina Ilgisonis1 1Institute of Biomedical Chemistry, Moscow, Russian Federation Our research presents a retrospective analysis of changes in information on the number of human proteoforms, post-translational modification (PTM) events, alternative splicing (AS), single-amino acid polymorphisms (SAP) associated with protein-coding genes in the NextProt database. In 2016, our group proposed three mathematical models for predicting the number of different proteins (proteoforms) in the human proteome. Eight years later, we compared the original data of information resources and their contribution to the prediction results, comparing the differences with new approaches to experimental and bioinformatics analysis of protein modifications. This work is designed to update information on the status of records in databases on identified proteoforms since 2016, as well as to identify trends in changes in the numbers of these records. It was found that, according to various information models, modern experimental methods allow us to identify from 5 to 125 million different proteoforms – proteins formed as a result of alternative splicing, implementation of single-nucleotide polymorphisms and post-translational modifications in various combinations at the proteomic level. This result reflects an increase in the size of the human proteome by 20 or more times over the past 8 years. The work was carried out within the framework of the Fundamental Research Program in the Russian Federation for the long-term period 2021-2030 (No122030100168-2). 984

985

P-III-0826 Standardization efforts in metaproteomics Kay Schallert1, Robert Heyer1 1Leibniz-Institut für Analytische Wissenschaften – ISAS – e.V., MdOA, Dortmund, Germany Background and Goals: New proteomics studies are continuously conducted, and their data is uploaded to ProteomeExchange during the publication process. In the field of metaproteomics meta-data is particularly diverse and not standardized. However, meta-analyses rely on the availability of meta data. Meta-data is readily available for laboratory procedures and data processing parameters but is unfortunately lacking in the most critical areas: a thorough description of the samples and the results. Meta data annotation is often considered a tedious process and additional work. Therefore, our long-term goal is to enable easy and fast meta data annotation and integrate the annotation tasks into the day-to-day workflows through research data management solutions. This should lead to better annotated data, while the workload is reduced. Meta data annotation web service: MetaForge: Here we present MetaForge, an easy-to use web service that also converts user input files into the standard formats mzML, mzIdentML and the emerging standard of proteomics SDRF. We combine three methods to conveniently annotate meta data. First, we show a questionnaire tree, that narrows down the required meta data to a particular research area and fills in general meta data. Second, optionally a large language model can be used to extract meta data from full text such as the method section of a paper. Third, an Excel-like table can be edited and supplemented with existing meta data from already prepared spreadsheets. Furthermore, research data management is increasingly becoming a requirement, and for this purpose we are developing a data management workflow for (meta-)proteomics with openBIS that fully integrates MetaForge, drastically reducing the time researchers need to spend on documentation. Conclusion and Outlook: The current level of documentation of raw data and meta data in (meta-)proteomics studies is still insufficient to fully the integrate the knowledge that is generated using automatic means. Our effort to enable convenient metadata annotation and integrated research data management paves the way for better integration and ultimately reduces the workload from documentation tasks. Better annotated data is crucial for modeling biological systems, improving and curating public databases such as UniProt, and conducting meta-analyses to consolidate knowledge. Fig. 1 (optional) 985

986

P-III-0827 Systematic proteomic meta-analysis of osteoblasts, bone, and blood: identifying druggable targets, active factors, and biomarkers for bone biomaterial development Johannes R. Schmidt1, Klaudia Adamowicz2, Lis Arend2,3, Jörg Lehmann1,4, Markus List3,5, Patrina S.P. Poh6, Jan Baumbach2,7, Stefan Kalkhof1,4,8 1Fraunhofer Institute for Cell Therapy and Immunology IZI, Preclinical Development and Validation, Leipzig, Germany 2University of Hamburg, Institute for Computational Systems Biology, Hamburg, Germany 3Technical University of Munich, Data Science in Systems Biology,, Freising, Germany 4Fraunhofer Cluster of Excellence Immune-Mediated Diseases CIMD, Frankfurt/Main, Hannover, Leipzig, Germany 5Technical University of Munich, Munich Data Science Institute (MDSI), Garching, Germany 6Berlin Institute of Health at Charité – Universitätsmedizin Berlin, Julius Wolff Institute, Berlin, Germany 7University of Southern Denmark, Computational Biomedicine Lab, Odense, Denmark 8University of Applied Science Coburg, Institute for Bioanalysis, Coburg, Germany Non-healing bone defects represent a significant public health concern, leading to decreased life expectancy and quality of life, particularly among an aging population with increasing comorbidities. Conventional treatments for these defects are ineffective in 5-10% of fractures, exacerbating patient burden and complicating medical interventions. This underscores the need for more effective treatment strategies and early identification of at- risk patients. Our systematic proteomic meta-analysis addresses this issue by identifying universally affected proteins and functions crucial for bone regeneration. These insights can be leveraged to develop novel bioactive biomaterials, therapeutic targets, and diagnostic biomarkers. To this end, we developed and applied the first high-quality meta-analysis and standardized proteomics workflow that integrates multi-disciplinary, multi-species studies assessing bone healing at the local and systemic level in multiple cellular, tissue and (pre)-clinical experimental set-ups. This integrated framework involves cross-studie data harmonization, protein selection, network construction, module mining, functional enrichment, drug repurposing, and protein scoring metrics. By applying this framework to synthesize 29 proteomics studies in the context of bone healing, we identified 51 key proteins potentially essential for bone healing, including well-known ECM components such as collagens, fibronectin, and periostin, alongside eleven less-studied proteins like YWHAE, HSPG2, and TGFBI. These proteins offer new opportunities for advancing bone biomaterial development. Moreover, we discovered compounds such as Trifluoperazine and Quercetin that target critical proteins involved in bone regeneration, presenting new therapeutic options to mitigate the socio-economic impact of non-healing bone defects. Our findings represent a significant step forward in bridging fundamental research and clinical application, providing a foundation for improved diagnostic, treatment, and predictive measures in bone healing. The integration of multidisciplinary and multi-species data highlights the robustness of our approach, enabling the identification of novel molecular mechanisms and potential drug targets. This work not only advances the field of tissue engineering but also illustrates the power of interdisciplinary research connecting fundamental and translational research with computational sciences in developing clinical solutions. This framework, detailed in our published tutorial, is designed to be broadly applicable to other research areas. 986

987

P-III-0828 Improved protein quantification by using bipartite peptide-protein graphs Karin Schork1,2,3,4, Julian Uszkoreit2,5, Michael Turewicz6, Jörg Rahnenführer4, Martin Eisenacher1,2,3 1Ruhr University Bochum, Medizinisches Proteom-Center, Bochum, Germany 2Ruhr University Bochum, Core Unit Bioinformatics - CUBiMed.RUB, Medical Faculty, Bochum, Germany 3Ruhr University Bochum, Medical Proteome Analysis, Center for Protein Diagnostics (PRODI),, Bochum, Germany 4TU Dortmund University, Department of Statistics, Dortmund, Germany 5Ruhr University Bochum, Medical Bioinformatics, Medical Faculty, Bochum, Germany 6German Diabetes Center (DDZ), Leibniz Center for Diabetes Research at the Heinrich Heine University Düsseldorf, Institute for Clinical Biochemistry and Pathobiochemistry, Duesseldorf, Germany In bottom-up proteomics, proteins are enzymatically digested to peptides (smaller amino acid chains) before measurement with mass spectrometry (MS), often using the enzyme trypsin. Because of this, peptides are identified and quantified directly from the MS measurements. Quantification of proteins from this peptide-level data remains a challenge, especially due to the occurrence of shared peptides, which could originate from multiple different protein sequences. The relationship between proteins and their corresponding peptides can be represented by bipartite graphs. A novel protein quantification method bppgQuant is proposed that is based on the bipartite peptide-protein graphs. The bipartite graphs were constructed from four different quantitative data sets stemming from different organisms (quantitative graphs) as well as theoretical graphs from an in silico digestion of the corresponding protein databases (database graphs). These graphs were characterized and it was shown that a large fraction of protein nodes does not have any unique peptides, which are difficult to quantify and would not be considered by most existing protein quantification methods. This highlights the potential of a novel protein quantification method that can handle those cases. bppgQuant calculates protein ratios from peptide ratios. It uses the structures of the bipartite peptide-protein graphs to build an equation system and subsequent optimization problem to calculate estimates for the protein ratios. bppgQuant was evaluated on the four above mentioned test data sets containing known protein ratios. Criteria were the MAD and median compared to the expected protein ratios, as well as the number of quantified protein nodes and the ability to distinguish different organisms within one data set or spike-in proteins. In summary, bppgQuant showed good results, especially when unneeded protein nodes were filtered before optimization. Fig. 1 (optional) 987

988

P-III-0829 OmicsQ: a user-friendly and interactive platform for quantitative proteomics data analytics Xuan-Tung Trinh1, David Bouyssié1, Ole N Jensen1, Adelina Rogowska-Wrzesinska1, Veit Schwämmle1 1University of Southern Denmark, Department of Biochemistry and Molecular Biology, Odense, Denmark Quantitative analysis and biological interpretation of proteomics data require appropriate statistical methods and careful assessment of their outputs. There is a lack of user-friendly, interactive toolkits for supervised analysis of quantitative peptide and protein expression profiles. Additionally, common tools often rely on value imputation, potentially leading to overestimation of statistical power and inclusion of erroneous values. We developed a user-friendly software toolkit for analyzing quantitative peptide or protein profiles from textual or Excel files. The toolkit includes modern methods for statistical testing, appropriate handling of missing data, quantitative QC, clustering of quantitative profiles, and investigating the quantitative behavior of protein complexes. Features include consistency checks of uploaded data, filtering unsuitable samples, batch correction, statistical power harmonization, and automatic sample group assignment. The underlying tools have been updated to work with missing data without need for imputation. Protein complex analysis with ComplexBrowser and CoExpresso now integrate recent data from Complex Portal and CORUM. Statistical testing with PolySTest and clustering with VSClust have been turned into Bioconductor packages. The platform is fully open source and is can via https://computproteomics.bmb.sdu.dk/app_direct/OmicsQ. deployed available as a locally. OmicsQ also be web service The OmicsQ toolkit enables deep proteomics analysis with minimal user effort, facilitating extensive biological interpretation and visualization. It features statistical testing without imputation, variance-sensitive clustering for improved recognition of co-expressed feature groups, and comprehensive analysis of protein complex behavior through the co-regulative behavior of their subunits. Running as a web browser application, it requires no local installation. Broad parameterization options and extensive visualizations provide full control over the analysis, easing the exposure of relevant content for a given biological context. We demonstrate the power of the toolkit on various proteomics datasets, showcasing its ability for deep characterization in different biological systems. In summary, OmicsQ offers a user-friendly, powerful solution for in-depth investigation of biological processes in multi-dimensional proteomics datasets. 988

989

P-III-0830 MultiOmicsAgent: guided extreme gradient-boosted decision trees-based approaches for biomarker-candidate discovery in multi-omics data Jens Settelmeier1,2, Sandra Goetze1,2,3, Julia Boshart1, Jianbo Fu1,2,3, Sebastian N. Steiner1, Martin Gesell1, Peter J. Schüffler4, Diyora Salimova5, Patrick G. A. Pedrioli1,2,3,6, Bernd Wollscheid1,2,3 1ETH Zürich, Institute of Translational Medicine at the Department of Health Sciences and Technology, Zurich, Switzerland 2Swiss Institute of Bioinformatics (SIB), Lausanne, Switzerland 3ETH PHRT Swiss Multi-Omics Center (SMOC), Zurich, Switzerland 4Technical University of Munich, Institute of Pathology, TUM School of Medicine and Health, Munich, Germany 5Albert-Ludwigs University of Freiburg, Department for Applied Mathematics, Freiburg, Germany 6ETH Zürich, Department of Biology, Zurich, Switzerland Introduction: MultiOmicsAgent (MOAgent) is an innovative, Python based open-source tool for biomarker discovery, utilizing machine learning techniques specifically extreme gradient-boosted decision trees to process multi-omics data. With its cross-platform compatibility, user-oriented graphical interface and a well-documented API, MOAgent not only meets the needs of both coding professionals and those new to machine learning but also addresses common data analysis challenges like data incompleteness, class imbalances and data leakage between disjoint data splits. MOAgent"s guided data analysis strategy opens up data-driven insights from digitized clinical biospecimen cohorts and makes advanced data analysis accessible and reliable for a wide audience. Method: A monte-carlo-like multi step approach reduces the initial features to a handalbe amount of the most phenotype-relevant features of quantitative omics expression tables for further downstream analysis, utilizing machine learning and sampling techniques. MOAgent provides a wide range of comprehensive output for further downstream analysis and interpretations. The most phenotype-predictive features are evaluated due their classification performance given by the decision threshold independent area under the receiver operating curve on a separate test set and Shapley (SHAP) values a widely used approach from cooperative game theory in explainable AI (XAI) that comes with desirable properties like fairness and efficiency. Preliminary results: We applied MOAgent in five case studies, including metabolomics, transcriptomics and proteomics abundances, as well as combinations of their of. MOAgent retrieves well known disease relevant features and further features which play a significant role for the phenotype classification and thus potentially are included in crucial biological processes related to the investigated diseases. Novel aspect and conclusion: MOAgent enables innovative integration of multi-omics data using machine learning and facilitates follow-up experiments involving the list of biomarker candidates provided. 989

991

P-III-0831 Systematic discovery and validation of degrader targets using deep proteomics screening combined with bioinformatics Bachuki Shashikadze1, Björn Schwalb1, Uli Ohmayer1, Martin Steger1, Anastasia H. Bednarz1, Sophie Machata1, Tobias Graf1, Denis Bartoschek1, Jutta Fritz1, Henrik Daub1 1NEOsphere Biotechnologies GmbH, Planegg, Germany Targeted protein degradation (TPD) using molecular glue degraders (MGD) represents a groundbreaking therapeutic strategy to eliminate disease-related proteins previously considered undruggable. However, despite its huge potential, systematic discovery of new molecular glues and their cellular degradation targets has been challenging. We introduce a deep proteomic screening followed by advanced biostatistics analysis to improve TPD drug discovery at all stages. This deep proteomic screening utilizes data-independent acquisition (DIA) based mass spectrometry to analyze compound libraries of potential molecular glues against cellular proteomes with very high throughput, coverage, and sensitivity. It can identify and quantify over 11,000 proteins per sample from cell lines treated with molecular glues, facilitating comprehensive proteomics-based drug and drug target discovery. Our pipeline begins with highly automated sample preparation workflows, ensuring efficient and reproducible screening. Following this, single-shot DIA-based mass spectrometry is performed. The quality of each sample is monitored in parallel before the final analysis, with wide range of quality control measures. For visualization, statistical analysis, and data interpretation, we utilize internal R Shiny applications optimized for efficiency and scalability, enhancing our ability to manage and analyze very large screens effectively. Statistical workflows are tailored to maximize the yield of putative neosubstrates, through optimized filtering, normalization and batch correction procedures. Likely direct degrader targets are identified using a sophisticated and comprehensive scoring system. Initially, targets are prioritized based on statistical test results. Then, the disease relevance of each target is assessed using available evidence of its causal relationship with the disease of interest. Further evaluation includes checking for the presence of structural motifs relevant to E3 ubiquitin ligase recruitment. Additionally, cheminformatics approaches are used to classify compounds based on structural similarity and/or presence of chemical substructures relevant to E3-MGD-neosubstrate complex formation and correlate this information with proteome changes. To further demonstrate degrader-induced modifications, we are performing ubiqutinomics experiments and are reliably quantifying up to 50,000 ubiquitination sites. Such an approach offers rapid validation of cellular downregulations being due to E3 ligase-neosubstrate relationships, without the need for pharmacological intervention or genetic modification. As part of mechanistic validation, we are also performing interactomics experiments to detect degrader induced E3 ligase binding. 991

992

P-III-0832 Digital proteome: a resource for target expression across 22 tissues in human and preclinical species Polina Shichkova1, Sandra Schär1, Marco Tognetti1, Jan Muntel1, Christopher Below1, Roland Bruderer1, Yuehan Feng1, Lukas Reiter1 1Biognosys AG, Schlieren, Switzerland Proteins are the primary targets and off-targets of therapeutics such as small molecule and biologics. Thus, a thorough understanding of their expression profiles across various human tissues in health and disease is essential through all stages of drug development. However, generating such a comprehensive proteome profile for each tissue poses a major analytical challenge due to the large dynamic range of protein abundance within a tissue and the enormous complexity introduced by distinct proteoforms. These proteoforms may arise (e.g. mutations), posttranscriptional events (e.g. isoforms from alternative splicing) or posttranslational modifications, resulting in more than a million protein species from 20,000 protein-coding genes. from genetic variations Thanks to advancements in sensitivity and speed in the recent years, mass spectrometry (MS)-based proteomics can now be routinely applied to quantify ten thousand of proteins in biological samples. Taking advantage of MS being an unbiased and matrix-agnostic method, we pivoted the generation of a high quality and uniform digital proteome resource of 22 healthy tissues from human, mouse and rat origin. We constructed the atlas in two parts aiming to address the following two aspects respectively: 1) a deep library providing evidence for protein/proteoform presence in each of the 22 tissues; 2) quantitative profiles for cross-species comparison in all tissues. For the first part, each tissue sample was subjected to an extensive fractionation process followed by MS analysis of each of the fractions. In total, we identified 20"025 protein groups mapped to 17"715 protein-coding genes, providing protein-level evidence for > 90% of all protein-coding genes annotated in the neXtProt repository. Among all identified proteins, 10"755 were detectable across all tissues. Notably, we identified 37 proteins mapping to putative genes. In part two of the atlas, quantitative expression profiles were generated for 15"772 protein groups using TrueDiscovery DIA-MS across the 22 tissues. These data allow us to investigate ubiquitous vs. tissue-restricted protein expression of various protein classes such as surface receptors or kinases. Moreover, we performed correlation analysis between protein expression and mRNA abundance across 19 tissues, demonstrating that for a significant portion of the proteome, mRNA level-information is not an adequate proxy for protein expression. Collectively, this MS-based resource alleviates dependency on antibody-based reagents in human and preclinical models. We envision that the tissue atlas can be expanded 1) to additional preclinical model species and 2) to cover various disease indications, providing valuable insights for target discovery and assessment of target/off- target tissue distribution. 992

993

P-III-0833 Significantly faster library-free dia-PASEF analysis with a Spectronaut integrated workflow in ProteoScape Grzegorz Skoraczynski1, Tejas Gandhi1, Lukas Reiter1, Heidi Vitrac2, Sven Brehmer3, Javier Lopez3, Jose Abuin3, Tharan Srikumar3, Dennis Trede3 1Biognosys AG, Schlieren, Switzerland 2Bruker, San Jose, CA, United States 3Bruker Scientific LLC, Billerica, MA, United States Introduction: ProteoScape, formerly PaSER, has been transforming into a comprehensive proteomics data analysis platform that can integrate third-party tools while utilizing the concept of data streaming to realize fully customizable real-time processing workflows including on-the fly decision making based on the data generated. Recent advances in technology have decidedly allowed the field to migrate towards fast LC gradients, allowing for increasingly larger sample cohorts in studies in both data-dependent acquisition (DDA) and data-independent acquisition (DIA) modes. "Library-free" analysis of DIA data has also grown to be an increasingly alternative to generating project specific spectral library from DDA acquisitions. A significant downside of these "library-free" workflows is longer processing times due to the increased search space, especially when PTMs are considered. Methods and Results: Here we"ve integrated a workflow including a Spectronaut module in Bruker ProteoScape (BPS) to take advantage of the synergistic capabilities of the two software. We show that BPS is a natural platform for faster library free analysis of dia-PASEF data using Spectronaut"s directDIA+ workflow by: 1) Allowing data to be processed at a single injection level, before combining at the project level. 2) Providing same results, including FDR control, for both single injection and project level analysis, and 3) Providing an intuitive user interface for the entire workflow. We utilize this new workflow on several dataset, including a phosphorylation enriched dataset and a multi-species dataset with know ratios to help evaluate precision and accuracy of the workflow and can show more than 80% savings in time-to-results vs analysis fully after the acquisition. Conclusion: ProteoScape has workflows integrating academic algorithms, such as DIA-NN, or 3rd party algorithms, such as Novor, and now Spectronaut greatly reducing time-to-results for large projects while simultaneously providing detailed results at the single injection level. 993

994

P-III-0834 BrainProt™: unraveling insights into meningioma tumorigenesis and neurological signaling cascades Sanjeeva Srivastava1, Deeptarup Biswas1, Ankit Halder1, Aparna Chauhan1 1IIT Bombay, Biosciences and Bioengineering, Mumbai, India The burgeoning interest in understanding the complexities of the human brain and its associated diseases has spurred the creation of diverse consortia, databases, and repositories. Innovations in Omics technologies have been instrumental in understanding underlying mechanisms, resulting in an abundance of brain-related data dispersed across multiple platforms. Our initiative aimed to develop an integrated platform, BrainProt™ (https://www.brainprot.org/), facilitating user-friendly and swift exploration of brain functions and diseases. This platform encompasses six interconnected domains, such as the Inter-Hemispheric Brain Proteome Map and Brain Phospho Map, and includes extensive data on disease markers, drug interactions, clinical trials, and multi- omics information for over 50 human brain diseases. By using BrainProt™, we understood Meningioma tumorigenesis, highlighting the platform's potential to reveal novel insights into brain tumors. Meningioma, constituting a significant proportion of brain tumors, has posed challenges to healthcare due to its rising incidence in recent years. Current diagnostic and treatment approaches largely rely on MRI and surgical interventions, highlighting the need for early diagnostic markers and tailored therapeutic strategies. Our study utilized BrainProt™'s Brain Disease Marker Curator (BDMC) section, coupled with our own quantitative clinical proteomics experiments, to identify and validate grade-specific markers for Meningioma in a cohort of Indian patients. Integrating Multiple Reaction Monitoring (MRM) validation and Machine Learning (ML) techniques, we successfully delineated segregation patterns between low-grade and high-grade Meningiomas. Furthermore, through meta-omics analysis, we identified altered signaling pathways and therapeutic targets, particularly highlighting Integrin Linked Kinase (ILK) as a key regulator in Meningioma tumorigenesis. Experimental inhibition of ILK demonstrated anti-proliferative effects and potential for therapeutic intervention. These findings highlight the significance of BrainProt™ in identifying disease markers and elucidating signaling cascades, showcasing its value in clinical research and neurobiology. References 1. Biswas, D. et al. Srivastava S. (2021). Deciphering the interregional and interhemisphere proteome of the human brain in the context of the human proteome project. JPR, 20(12), 5280-5293. 2. Biswas, D. et al. Srivastava S. (2022). Deep Phosphoproteome Landscape of Interhemispheric Functionality of Neuroanatomical Regions of the Human Brain. JPR, 22(4), 1043-1055. 3. Biswas, D. et al. Srivastava S. (2023). BrainProt(TM) 3.0: Understanding Human Brain Diseases using comprehensively curated & Integrated OMICS datasets. bioRxiv, 2023.06.21.545851 4. Biswas, D. et al. Srivastava S. (2023). Integrated Meta-Omics Analysis Unveils the Pathways Modulating Tumorigenesis and Proliferation in High-Grade Meningioma. Cells, 12(20), 2483. 994

995

P-III-0835 Relative quantification of proteins and of post-translational modifications with shared peptides: a weight-based approach Mateusz Staniak1,2, Ting Huang3, Amanda Figueroa-Navedo3, Devon Kohler3, Meena Choi4, Trent Hinkle4, Tracy Kleinheinz4, Robert Blake4, Christopher Rose4, Małgorzata Bogdan1,5, Olga Vitek3 1University of Wrocław, Wrocław, Poland 2Hasselt University, Hasselt, Belgium 3Northeastern University, Boston, MA, United States 4Genentech, Inc., San Francisco, CA, United States 5University of Lund, Lund, Sweden Bottom-up mass spectrometry-based proteomics studies changes in protein abundance and structure across conditions. Since the currency of these experiments are peptides, i.e. subsets of protein sequences that carry the quantitative information, conclusions at a different level, e.g., at the level of proteins or of post-translational modifications, must be computationally inferred. The inference is particularly challenging in situations where the peptides are shared by multiple proteins or post-translational modifications. While many approaches infer the underlying abundances from unique peptides, there is a need to distinguish the quantitative patterns when peptides are shared. From a statistical perspective, inclusion of shared peptides into the estimates of abundances or proteins or post- translational modifications models induces a data structure, in which observations (peptide intensities) may belong to multiple groups defined by proteins or modification sites. Typically, shared peptides are removed from analysis of MS data, which leads to loss of information, and in particular some proteins or post-translational modifications may be lost. Alternatively, proteins that share peptides are grouped together, eliminating the possibility of estimating their distinct quantitative patterns. We introduce a statistical approach for estimating protein abundances, as well as site occupancies of post- translational modifications, based on quantitative information that includes shared peptides. This approach extends the existing MSstatsTMT framework for labeled MS data summarization and differential analysis by treating the quantitative patterns of shared peptides as convex combinations of abundances of individual proteins or modification sites, and estimating the abundance of each source in a sample together with the weights of the combination. Abundances estimated using this method can serve as input to statistical models that estimate differences between experimental conditions. We demonstrate the utility of this new summarization method using computer simulations and examples based on data from experiments with diverse biological objectives, including protein degradation and to changes in protein post-translational modifications. 995

996

P-III-0837 AliceDB – database and bioinformatic tool for identification of natural protein variants based on mass spectrometry measurements Marcel Thiel1, Michał Puchalski1, Alicja Różycka1, Stanisław Ołdziej1 1University of Gdańsk, Intercollegiate Faculty of Biotechnology UG&MUG, Gdańsk, Poland The results of proteomic/peptidomic studies are highly dependent on the amino acid sequence databases used during analysis. In recent years, due to a deeper understanding of the biological systems functions (by interpretation and information processing of the data related to the DNA sequences), amino acid sequence databases have been enriched with sequences resulting from alternative RNA splicing (isoforms), products of short ORFs or pseudogenes. However, even these modifications of amino acid sequence databases do not capture all of the diversity of the proteome. Amino acid sequence databases offer only canonical sequence representation of the product (or products) of a given gene. Moreover, the genetic material within a species is not identical for all individuals. Genomes (as well as proteomes) of individuals of the same species contain a huge number of natural variants. The occurrence of natural variants in proteomic/peptidomic studies is virtually impossible to identify due to the lack of systematized and readily available information, organized in a way suitable to use in available software used to process mass spectrometry data. Based on public available data resources (Uniprot and NCBI's ClinVar databases), we built a database of natural variants of proteins as well ecosystem of bioinformatic tooling named AliceDB. It is used in identification of natural variants based on mass spectrometry data. Currently, AliceDB covers human proteome only (with more than 3,3 million records). Even that, it is worth to mention that software allows to build a database for investigation for any proteome. AliceDB is designed to work with commercially and non-commercially available software used to process and analyze mass spectrometry data. The performance of the developed tool was tested on two datasets acquired from the PRIDE repository (PXD048837 and PXD024347 respectively). Preliminary results indicate that inclusion of natural variants in the processing of peptidomic/proteomic data allows to increase the number of peptide identifications by 2-11% (depending on the sample type). A higher number of identified peptides leads to the higher number of protein identifications (increase by 2-6%). As a result of that, statistical significance of amino acid sequence identifications also improves. It is worth to mention that significant potential for AliceDB usage lies in data analysis of medical importance. At present, information about possible natural variants comes mainly from DNA sequencing data (genomics) or mRNA sequencing data (transcriptomics). Identification of natural variants in protein products of genes, especially those of marker or diagnostic importance, was practically limited only to targeted analyses focused on a few predefined ones. AliceDB enables the identification of all natural variants, greatly expanding the application of mass spectrometry in studies related to medical diagnostics. 996

997

P-III-0838 Multilayer data integration and cross-validation of human protein atlas datasets: a generative artificial intelligence driven roadmap to validate DNA target gene predictions of KRAB ZNF proteins Hans-Juergen Thiesen1,2, Felix Steinbeck1,2, Christa Geisser2, Claudia Roewer2, Volker Stadler3 1University of Rostock, Rostock, Germany 2Gesellschaft für Individualisierte Medizin mbH, Rostock, Germany 3PEPperPRINT GmbH, Heidelberg, Germany Our computational roadmap is based on the following hypothesis: The transcriptional expression of KRAB ZNF target genes is assumed to be inversely regulated in respect to KRAB ZNF expression. The Human Protein Atlas (HPA) offers a valuable resource for validating expression and localization of KRAB ZNF proteins in immunohistochemical (IHC) images across normal and cancer tissue sections by providing RNA seq data of tissues and cell lines. The KRAB domain initially identified in KOX1/ZNF10 (December 27th, 1988) [PMID2288909] is encoded in about 400 human C2H2 zinc finger genes [PMID37716846]. The KRAB domain of KOX1 fused to tetracycline repressor as TETR-KRAB showed in 1995 [PMID7891684] that transcriptional repression is an active process mediated by SMP1/TRIM28 utilized in CRISPRi applications [PMID 25448066]. Modern KRAB C2H2 zinc finger protein evolved during tetrapod evolution by insertion of an additional amino acid in the leucine zipper structure of ancestral KRAB domains (PRDM7) [PMID35162997]. Our team efforts outline a comprehensive roadmap for validating DNA target genes of human KRAB ZNF genes utilizing HPA data sets. Specificities and cross-reactivities of 42 antibodies (29 KRAB and 13nonKRAB) were assessed by high-density-peptide-array analysis. 57120 immunohistochemical (IHC) images of 20 cancer and 48 normal tissues using 269 antibodies/240 KRAB proteins, 11 non-KRAB antibodies/10 proteins were semi-quantitated by Definiens Developer XD 1.1 defining 5 subsets of KRAB ZNF expression categories and compared to RNA seq data available from the same tissue [PMID24309898]. Opposed to individual pairwise comparisons of RNA to protein levels, a combination of selected KRAB ZNF IHC subsets showed reduced expression in cancer tissues on RNA and protein levels. Data integration of published KRAB ZNF targets genes [PMID37730438, PMID28273063, PMID33326746] identified a gene set of about 100 genes that harbor from 100 to 833 KRAB ZNF binding sites, whereas 2256 genes had only one KRAB ZNF binding site. Our computational analyses reveal technical challenges and experimental pitfalls regarding strengths and weaknesses of IHC images related to limited linearities of IHC images and antibody specificities within KRAB and non-KRAB C2H2 multigene zinc finger families. Limitations of RNA seq data sets are mirrored such as splice variants, transcriptional and posttranscriptional processing, and posttranslational modifications. Besides obviously required technical QC improvements, our Polygenic Expression Network (PEN) Score selects individuals that share non-coding and coding single nucleotide polymorphisms (SNPs), see our animation (http://www.indymed.net/#interferon/ to streamline diversities in population genetics. Our AI toolbox approach presents combined multilayer efforts for improving DNA target gene predictions of KRAB zinc finger (ZNF) proteins in human tissues in phylo- and ontogenesis asking the HPA initiative for whole genome information. Fig. 1 (optional) 997

998

P-III-0839 deTELpy: a Python package for high throughput detection of amino acid substitutions in mass spectrometry datasets Agnes Toth-Petroczy1, Cedric Landerer1, Jonas Poehls1, Maxim Scheremetjew1, HongKee Moon1, Lena Hersemann1 1Max Planck Institute of Molecular Cell Biology and Genetics, Dresden, Germany Errors in protein synthesis in the cell can lead to phenotypic mutations, such as amino acid substitutions and extended polypeptides due to stop codon readthrough. These proteoforms remain invisible and unannotated during conventional mass spectrometry data processing. Here, we provide a Python package implementation of a high throughput pipeline to detect phenotypic mutations in hundreds of mass spectrometry datasets. deTELpy can facilitate the systematic understanding of translation errors, and study phenotypic mutations across organisms and stress conditions such as drug treatment or disease conditions. Fig. 1 (optional) 998

999

P-III-0840 Disease-specific antibody databases for bottom-up proteomics Xuan-Tung Trinh1, Veit Schwämmle1 1University of Southern Denmark, Biochemistry and Molecular Biology, Odense, Denmark Identification of human antibodies using bottom-up proteomics approaches relies on database searches that compare experimental peptide values and their fragments to theoretical values derived from protein sequences in a database. Conventional protein databases like UniProt contain only a fraction of the antibody sequences present in the human body, with just 1,095 reviewed sequences and 37,848 unreviewed sequences as of May 2024, despite the human body's capacity to produce quintillions of antibodies. This limitation underscores the urgent need for new solutions to identify expressed antibodies. We propose using antibody sequence databases based on previous genomic studies to address this limitation. Leveraging billions of human antibody sequences from 70 genomic studies (contained in the Observed Antibody Space database: https://opig.stats.ox.ac.uk/webapps/oas/), we created disease-specific antibody databases for bottom-up proteomics to identify new antibodies associated with relevant diseases such as SARS-CoV-2, HIV, Ebola, and chronic lymphocytic leukaemia. Instead of directly using antibody sequences, which have considerable redundancy due to antibody constant regions, we used peptide sequences corresponding only to the variable regions of antibodies. We performed in silico digestion of antibody sequences to obtain peptide sequences and selected peptides in the variable regions of antibodies. The peptides already present in the two major human protein databases, UniProt and NCBI-RefSeq, were filtered out. Subsequently, we merged the resultant peptides with the UniProt human proteome and the common Repository of Adventitious Proteins (cRAP) to form disease-specific peptide databases for bottom-up proteomics. We demonstrated the use of these databases in identifying new SARS-CoV-2 antibody peptides in previously measured blood samples, validated by running the same analysis on negative control samples. Additionally, we created an online resource for researchers to download disease-specific peptide databases in fasta format. This study provides researchers with antibody resources comprising millions of antibodies for each disease, ready to be used as part of common proteomics data analysis workflows. Our databases will aid researchers in discovering novel disease-specific antibodies in mass spectrometry data, adding a new layer of valuable information, not provided when using commonly used databases such as UniProtKB and NCBI-RefSeq. 999

1000

P-III-0841 Identification of candidates from high-throughput proteomic profiling of PROTAC-induced degradation Nikita Verheyden1, Giulio Ferrario1, Jasmin Schäfer1, Bhavesh Parmar1, Christian Münch1 1University Clinic Frankfurt, Molecular Systems Medicine, Frankfurt, Germany Advances in mass spectrometry-based high-throughput proteomics have made it possible to perform rapid quantitative proteomic screens. These screens can be particularly useful for small molecules in pharmacology. Their effects are sometimes difficult to predict and many variants can be designed and produced cheaply. A specific application is their use in targeted proteolysis by facilitating proximity between the E3 ligase complex and the target protein. This leads to ubiquitination and subsequent degradation of the target protein. These molecules are known as proteolysis targeting chimeras (PROTACs). To determine both degradation levels and translational effects, we employ multiplexed enhanced protein dynamics (mePROD). In this method, newly translated proteins are labelled with SILAC, allowing them to be distinguished from previously present and differently labelled proteins. Using a 96-well plate to screen multiple compounds allows us to rapidly generate proteome-wide data of both steady-state and newly translated protein levels for a large number of candidates. However, this large amount of data requires an effective processing pipeline to find relevant effects within the screening data. We have developed an automated pipeline for processing SILAC-TMT tagged proteomics data, including statistical analysis using linear mixed models. Changes in effects are visualized using complex plots, such as circus or hive plots, to show overall trends without losing much detail. However, these plots can quickly become complex and not every data point may be visible. To improve the process of discovering relevant effects in our datasets, we developed a scoring system to identify new candidates. This score includes several layers of information, such as fold changes for all proteins across samples, and determines direct and indirect effects of the PROTAC. The result is an easy-to-read indicator of efficacy that retains the nuances of the data. In addition, this system be applied not only to small molecule screens but also to other targeted degradation systems such as siRNAs. 1000

1001

P-III-0842 Metadata matters: enhancing TBI proteomics Melissa S Vorster1, David Tabb2, Hugh G Patterton1, Ben Loos3, Bernhard Küster4 1Stellenbosch University, Centre for Bioinformatics and Computational Biology, Stellenbosch, South Africa 2University Medical Center Groningen, Groningen, Netherlands 3Stellenbosch University, Physiology, Stellenbosch, South Africa 4Technical University of Munich, Proteomics and Bioanalytics, Munich, Germany Meta-analysis is a cornerstone of scientific research, enabling the synthesis of data across multiple studies to derive comprehensive conclusions and gain insights into connections between the different studies. However, the utility of meta-analyses is significantly compromised by the lack of comprehensive metadata. This study explores the importance of detailed metadata in proteomics meta-analyses, with a focus on understanding the proteomic alterations in Traumatic Brain Injury (TBI). TBI is a complex condition involving a wide range of pathological processes and diverse proteomic changes. Despite being a leading cause of morbidity and mortality worldwide, its underlying molecular mechanisms remain poorly understood. To address this, a meta-analysis of publicly available proteomic TBI datasets was designed. We conducted a systematic review of proteomics TBI studies, assessing the quality and completeness of their metadata. Our analysis revealed significant deficiencies in metadata reporting, including the absence of and inconsistencies in study designs, sample characteristics, and data processing methods that were uploaded to the PRIDE repository. These gaps hinder accurate meta-analyses, leading to reduced reproducibility of findings. By addressing these deficiencies and adhering to the FAIR (Findable, Accessible, Interoperable, and Reusable) principles, the proteomics community can significantly improve the transparency, reproducibility, and overall quality of meta-analyses. Despite the lack of comprehensive metadata, our study identified key TBI proteins present in the datasets at an identification level. Enhanced metadata practices would allow us to analyze the datasets at a quantification level, thereby advancing our understanding of TBI and facilitating the translation of proteomic discoveries into clinical applications. In conclusion, prioritizing meticulous metadata collection and reporting (like enforcing the SDRF or MIAPE initiatives) is essential for unlocking deeper insights, identifying reliable biomarkers, and developing more effective diagnostic and therapeutic strategies. This study exemplifies the transformative potential of data integration and bioinformatics in translating raw proteomic data into actionable biological knowledge, with significant implications for advancing TBI research and ultimately improving patient outcomes. Keywords: Meta-analysis, Metadata, Traumatic Brain Injury, Data Integration, Bioinformatics, Reproducibility 1001

1002

P-III-0843 A meta-analysis of brain and cerebrospinal fluid proteomic profiles of Huntington's disease animal models and human patients Eleni Voukali1, Petr Vodicka1 1Czech Academy of Sciences, Laboratory of Applied Proteome Analyses, Institute of Animal Physiology and Genetics, Libechov, Czech Republic Mass spectrometry and proteomics are continuously evolving methods in systems biology carrying the enormous potential for accurately profiling the proteome of diseased tissues and exploring biomarker molecules. In Huntington"s disease (HD), identifying novel effective biomarkers for monitoring the disease progression from the early asymptomatic stage would be extremely useful for clinical decision-making. Animal models allow us to delve into the molecular profile of asymptomatic HD stages that would not be possible by studies of post-mortem human brains. Several proteomic studies in HD animal models focusing on the elucidation of huntingtin (HTT) and mutant HTT (patho)physiological roles, rather than on biomarker discovery are impeded by heterogeneity of proteomic methods and thus inconclusive overlap of their findings. Integrating the datasets from multiple studies can provide valuable information that a single analysis does not offer. In this study, we performed a meta-analysis of publicly available in-depth proteomic datasets from HD's transgenic and knock-in animal models. The primary objective of our meta-analysis was to identify the degree of overlap and correlation between the different animal models of HD based on the protein identifications and expression data. Our secondary objective was to compare these outcomes with the available clinical proteomic data from the brain and cerebrospinal fluid. We identified four datasets in transgenic and knock-in HD mice using label-free proteomic methods to quantify protein expression. The raw liquid chromatography-mass spectrometry (LC/MS) data were retrieved from the PRIDE repository and re-analysed using Max Quant and the most recent reference proteomes. The datasets from these parallel studies are then combined and meta-analysed using the MetaOmics pipeline that allows the data preprocessing, integration of differential expression data and pathway and network analyses. Two available datasets in human HD patients are correlated with these data. The results of this meta-analysis will be presented and discussed in relevance to the overlap with the original findings and preliminary data from our recently published systematic review summarising the protein identifications from HD patients and experimental models, reporting biomarker candidates verified by independent methods or independent studies. This is the first study attempting to explore the publicly available LS/MS datasets in the context of HD and its conclusions might be beneficial for discovering interesting biomarkers. This work was supported by the Czech Science Foundation project 22-24983S. 1002

1003

P-III-0844 Integrated proteomics analysis of baseline protein expression in pig tissues Shengbo Wang1, Andrew Collins2, Ananth Pride1, Silvie Fexova1, Irene Papatheodorou1, Andrew Jones2, Juan Vizcaino1 1EMBL – EBI, Cambridge, United Kingdom 2University of Liverpool, Systems, Molecular and Integrative Biology, Liverpool, United Kingdom Sus scrofa, the domestic pig, is a model organism relevant for food production and for human biomedical research. The availability of an increasingly large amount of public proteomics datasets in ProteomeXchange and in the PRIDE database (https://www.ebi.ac.uk/pride/) in particular, presents an opportunity for performing combined meta-analyses to generate comprehensive organism-wide protein expression maps across different organisms and biological conditions. To enable the access to proteomics data by the wider scientific community, the PRIDE team is developing data dissemination and integration pipelines with existing popular resources so that proteomics data is made more accessible. We have already performed combined analyses of baseline (without any perturbation) protein expression for human, mouse and rat tissues. Here we reanalysed 14 public proteomics datasets from the PRIDE database coming from pig tissues to assess baseline (without any biological perturbation) protein abundance in 14 organs, encompassing a total of 20 healthy tissues from 128 samples. The analysis involved the quantification of protein abundance in 599 mass spectrometry runs. We compared protein expression patterns among different pig organs and examined the distribution of proteins across these organs. The number of canonical proteins per dataset ranged from 789 (dataset PXD002918, biceps femoris) to 6,062 (dataset PXD012636, heart). Then, we studied how protein abundances compared across different datasets and studied the tissue specificity of the detected proteins. Of particular interest, we conducted a comparative analysis of protein expression between pig and human tissues (from a previous study performed following the same methodology), revealing a high degree of correlation in protein expression among orthologs, particularly in brain, kidney, heart, and liver samples. Additionally, we analysed the differences in quantitative expression of proteins across organs between human and pig organs. integrated We have resource (https://www.ebi.ac.uk/gxa/) for easy access and visualisation of the protein expression data individually or alongside gene expression data. into EBI"s Expression Atlas the protein expression results 1003

1004

P-III-0845 Prolfquapp: protein differential expression analysis with advanced R tools and dynamic visual reporting Witold Wolski1,2,3, Jonas Grossmann1,2,3, Peter Leary1,2,3, Leonardo Schwarz1,2,3, Bernd Roschitzki1,3, Laura Kunz1,3, Paolo Nanni3, Christian Panse1,2,3, Ralph Schlapbach1,3 1University of Zürich, FGCZ, Zurich, Switzerland 2Swiss Institute of Bioinformatics (SIB), Lausanne, Switzerland 3ETH Zürich, FGCZ, Zurich, Switzerland The prolfquapp R package [github.com/prolfqua/prolfquapp] is a powerful tool for simplifying protein differential expression analysis and reporting by streamlining data processing and enhancing result visualization through dynamic HTML reports. Developed for label-free and labeled proteomics, this command-line R package leverages prolfqua [https://doi.org/10.1021/acs.jproteome.2c00441] to offer a streamlined workflow from data input to insightful outputs. advanced statistical package models the in R One of the most intriguing features of prolfquapp is its ability to facilitate the study of how various factors influence protein expression and how these factors interact. To illustrate this, we will use an experiment with repeated measurements and two explanatory variables as an example. We will guide you through the process of specifying group differences and their interactions using the annotation file. We will also demonstrate how the visualizations in the dynamic HTML report make it easier to understand the results. Our aim is to ignite your curiosity in the capabilities of factorial designs and prolfquapp. The prolfquapp package is designed with the user in mind, offering a user-friendly command-line tool. It enhances research efficiency by performing essential tasks like data aggregation, summarization, normalization, and differential expression analysis using just a few input files. These include the parameter file, the sample annotation file, and a folder with the outputs of quantification software like DIANN, MaxQuant, Spectronaut, or Compound Discoverer. It produces dynamic HTML reports with quality control and visualizations, along with multiple output formats, including PDF and XLSX. Its seamless integration with the interactive exploreDE [https://zenodo.org/records/10571252] Shiny application enhances the user's ability to visualize and interpret the data and create publication-ready figures. This user-friendly and efficient solution empowers researchers in protein differential expression analysis. Prolfquapp is similar to other R applications, such as Einprot, LFQAnalyst, or MSDap, in that it reuses statistical models implemented in other R-packages, and allows non-bioinformaticians to quickly generate QC and DEA results for label-free and labeled proteomics quantification experiments. It is unique because it generates interactive HTML documents, integrates seamlessly with the Shiny exploreDE application, and allows easy analysis from https://github.com/prolfqua/prolfquapp. experiments with Prolfquapp installed factorial designs. can be of Fig. 1 (optional) 1004

1005

P-III-0846 Enhancing Mass Spectrometry Imaging Accuracy via Spatially Informed Mass Recalibration Leonardo Schwarz1,2,3, Christian Panse1,2,3, Witold Wolski1,2,3, Jonas Grossmann1,2,3, John Abbey4,5, Mengze Zhang4,5, Bernd Bodenmiller4,5, Ralph Schlapbach1,3 1University of Zürich, Functional Genomics Center Zürich, Zurich, Switzerland 2Swiss Institute of Bioinformatics (SIB), Lausanne, Switzerland 3ETH Zürich, Functional Genomics Center Zürich, Zurich, Switzerland 4University of Zürich, Department of Quantitative Biomedicine, Zurich, Switzerland 5ETH Zürich, Institute for Molecular Health Sciences, Zurich, Switzerland Mass spectrometry imaging (MSI) is a powerful analytical technique that allows for visualizing the spatial distribution of molecules directly on biological samples. MSI data acquired using MALDI-TOF spectrometers can suffer from significant mass shifts of up to 1 Dalton, jeopardizing data interpretation. Despite efforts in sample preparation, practical limitations tied to the specific mass spectrometer and sample may persist. Our work introduces a computational pipeline developed in the Python language that recalibrates proteomics MSI data. It estimates these shifts, uses the spatial structure to smooth the problem, and uses the estimates to mitigate the effect of these shifts effectively. We study two scenarios: targeted, where known target masses serve as calibration references, and untargeted, where we create a reference mass list using an average peptide mass model. We applied the pipeline to several example datasets and developed and implemented visualizations to study the effects of the calibration. We compared the newly developed error smoothing method with previously published methods for mass calibration of proteomics MALDI-TOF data. Correcting mass errors can yield high-quality images and facilitate MSI data interpretation. We also show some of the current limitations of the approach. The newly developed Python package supports several methods for mass calibrating proteomics MSI datasets and visualization methods to assess calibration quality. Fig. 1 (optional) 1005

1006

P-III-0847 Integrating MSstats with INDRA for enhanced interpretation of proteomic differential analysis results Anthony Wu1, Klas Karis1, Lindsay Pino2, Benjamin M. Gyori1, Olga Vitek1 1Northeastern University, Boston, MA, United States 2Talus Bioscience, Seattle, WA, United States Interpreting results for MS-based proteomic experiments, particularly those from analyses of differential protein abundance, often involves statistical metrics such as lists of p-values that are devoid of biological context. There remains a challenge to contextualize these metrics with respect to existing biomolecular knowledge. Manually curated biological pathway databases offer valuable insights but they capture only a fraction of available literature, leaving many results unexplained. Especially in the case of early-stage discovery experiments, where few differentially abundant proteins are to be expected, the need for interpretation tools becomes critical to derive any meaningful hypotheses for future experiments. To address this gap, we introduce a software integration between MSstats, a family of open- source packages for detecting differentially abundant proteins, and INDRA, an automated system leveraging natural language processing to extract biomolecular network information from biomedical literature at scale. Additionally, INDRA incorporates data from existing protein interaction databases, such as Phospho.ELM and Reactome. INDRA facilitates access to the extracted data by offering a convenient database client that enables users to programmatically query for relations. We implemented a new algorithm within MSstats that sends a list of differentially abundant proteins to INDRA"s REST API to retrieve a list of relevant biological networks. Using the list of biological networks, the algorithm constructs a network visualization that contains annotations from both MSstats and INDRA results, enabling further exploratory analysis. With this new integration, users can explore the list of proteins with significant p-values in the context of surrounding mechanisms, linked directly to context- specific literature evidence. We demonstrate this integration through a data set measuring the effect of small molecule compounds on the activity of transcription factors. Intact, live cells from the THP-1 cell line were treated with one of eight compounds, then chromatin-bound proteins were quantified using data independent acquisition while identification and quantification were performed with DIA- NN. There were a total of 270 MS runs, 132 runs for the DMSO group and 16-18 runs for each compound. We processed the DIA results through an MSstats pipeline, incorporating an additional step to retrieve data from INDRA and visualize the results. Through this case study, we demonstrate that the new integration between MSstats and INDRA can uncover plausible biological explanations in a seamless way, even in cases where statistical results contain few differentially abundant proteins. Our work highlights an essential step to bridging the gap between statistical analysis and biological interpretation, allowing researchers to derive meaningful insights from proteomics experiments. 1006

1007

P-III-0848 BioLadder: a bioinformatic platform primarily focused on proteomic data analysis Songfeng Wu1,2, Yupeng Zhang1,2,3, Chunyuan yang1,2,3,4, Yunping Zhu1,2,3,4, Jingli Li1,2,3,4 1BPRC, Beijing, China 2Beijing Qinglian Biotech Co., Ltd., Beijing, China 3Beijing Qinglian Biotech Co., Ltd., Beijing, China 4State Key Laboratory of Proteomics, Beijing Proteome Research Center, National Center for Protein Sciences (Beijing), Beijing Institute of Lifeomics, Beijing, China Data analysis is an essential component of large-scale life omics research. Online data analysis platforms allow researchers to complete data analysis without coding the program, significantly speeding up the scientific research process. While there are many online data analysis tools available, there has not yet been a dedicated professional website for proteomics data analysis. We have developed BioLadder, a proteomics bioinformatics data analysis platform, with the aim of providing protein researchers with easy-to-use online data analysis tools. BioLadder consists of over 50 analysis modules, belonging to 7 classes (2 main categories), covering experimental data visualization, data preprocessing, qualitative and quantitative analysis, functional analysis, and so on. BioLadder also incorporates four user-friendly design features to assist users in accurately and conveniently completing data analysis the website https://www.bioladder.cn/ and is free to use after registration. tasks. BioLadder can be accessed through 1007

1008

P-III-0849 Cell type deconvolution of bulk proteome samples Måns Zamore1, Sergio Mosquim Junior1, Sebastian L Andree1, Can Altunbulakli1, Tova Hermodsson1, Malin Lindstedt1, Fredrik Levander1,2 1Lund University, Department of Immunotechnology, Lund, Sweden 2Lund University, National Bioinformatics Infrastructure Sweden, Lund, Sweden Determining the types and amounts of cells present in bulk proteome samples can be of clinical relevance, for example by aiding in stratification of cancer treatment groups. Traditionally, gene expression data has been used for this purpose, but the potential of protein data has been largely unexplored. This study investigates computational methods for estimating immune cell composition from bulk sample protein data. By collecting proteome data of immune cell populations with DIA mass spectrometry, as well as simulating mixtures of collected sample data, we evaluate the feasibility of cell type deconvolution algorithms for estimation of cell type content. We furthermore explore the use of either peptide level data or protein level data for this purpose. Our findings demonstrate how high accuracy in estimation of mixed cell populations can be achieved, with some estimates closely matching the expected cell levels depending on sample complexity. Moreover, we provide an R package to streamline the preprocessing of proteomics data and facilitate the interpretation of results. This study emphasizes the potential of utilizing proteome data for deciphering cell-type composition in complex biological samples. 1008

1009

P-III-0850 DORSSAA: Drug-target interactOmics resource based on stability/solubility alteration assay Ehsan Zangene1, Elham Gholizadeh1,2, Salma Ibourki1,2,3, Amir Ata Saei1,2,3,4, Mathias Wilhelm1,2,3,4,5, Mohieddin Jafari1,2,3,4,5 1University of Helsinki, Biochemistry and Developmental Biology, Helsinki, Finland 2University of Helsinki, Helsinki, Finland 3University of Helsinki, Lille, France 4Karolinska Institute, Stockholm, Sweden 5Technical University of Munich, Munich, Germany The field of drug development faces significant challenges in assessing cellular-level target and off-target engagement and the mechanism of action of drugs. To address these challenges, innovative approaches like Thermal Proteome Profiling and Proteome Integral Solubility Alteration have emerged. These high-throughput techniques leverage mass spectrometry to characterize drug-protein interactions on a large scale, holding the potential to revolutionize drug discovery. However, these methods still lack a unified platform for comprehensive data analysis and comparison across different studies, hindering the ability to cross-validate findings and fully understand the implications of drug interactions. Here, we introduce DORSSAA, an integrative resource specifically designed for the comparative analysis of stability and solubility alteration assay datasets. DORSSAA hosts an extensive database containing an impressive 258,238 drug-protein interactions, including data from 18 distinct cell lines and organisms, 33 unique compounds, and a vast collection of 32,934 proteins. With the incorporation of multiple analysis tools, DORSSAA provides a user-friendly web application to support the comprehensive study of drug-target interactions. In a comprehensive case study, DORSSAA's utility is demonstrated by identifying and exploring protein-drug interactions, such as the prominent DHFR-Methotrexate interaction, across diverse cell lines and experimental conditions. This case study highlights the significance and cross-verification of drug targets using DORSSAA database. This comprehensive resource accelerates drug discovery, enhances our understanding of protein target behavior, and empowers drug discovery and optimization research. DORSSAA is available at https://dorssaa.it.helsinki.fi/. Fig. 1 (optional) 1009

1010

P-III-0851 A data processing pipeline to streamline and evaluate report tables from different mass spectrometry data processing software Tanja Ziesmann1, Maximilian Sprang2,3, Dana Hein1, David Gomez-Zepeda1,4,5, Stefan Tenzer1,4,5, Ute Distler1 1University Medical Center of the Johannes Gutenberg-University Mainz, Institute for Immunology and Research Center for Immunotherapy (FZI), Mainz, Germany 2Johannes-Gutenberg University Mainz, Computational Biology and Data Mining, Institute of Molecular Ecology, Institute for Quantitative & Computational Biosciences, Faculty of Biology, Mainz, Germany 3Deutsches Forschungszentrum für Künstliche Intelligenz GmbH, Data Science and its Applications (DSA) Research Department, Kaiserslautern, Germany 4Helmholtz Institute for Translational Oncology (HI-TRON), Mainz, Germany 5Deutsches Krebsforschungszentrum (DKFZ), Heidelberg, Germany Background: Proteomics data acquired through mass spectrometry needs further processing to be human readable. However, the output tables produced by processing software can require large amount of resources, may not be readable with Excel, and might include information not required for further analysis, such as creating plots or doing statistical analysis. Creating a workflow customised to a lab"s need helps solving this problem. Methods: Data acquired using timsTOF (Bruker Daltonics) or Orbitrap (Thermo Fisher Scientific) instruments were analysed with the processing software DIA-NN, MaxQuant, or Spectronaut. The resulting tabular outputs in software specific formats were further analysed with a downstream processing pipeline programmed in Python and R. The user provides a design file, linking the instrument file with the experiment condition. Using the information provided by the user in the config file the pipeline executes the desired analyses. Most analysis steps are separate from each other and can be chosen individually, however some plots are dependent on the results of the statistical analysis, e.g. t-test, as they visualise the significantly enriched proteins. Results: With minimal user input, the pipeline unifies the different output formats and produces standardized output formats as well as an html-report. The report includes the description of the different plots and output files, as well as information about the settings used for the statistical analysis. This report is easily shareable between collaboration partners. The pipeline is built modular, so the user can either obtain a standardized tabular output or additional plots and tables as specified in the config file. The user can chose among different statistical tests for either pairwise or groupwise comparisons. The plots include volcano plots, pca plots, UpSet plots, violin plots and more. Other outputs include an Excel file consisting of the filtered output in wide format for easy use in e.g. R, another Excel sheet containing the results of the statistical tests for each protein and comparison. Additional filters can be applied, e.g. for the removal of contaminants of proteins with low peptide support. Conclusions: Our pipeline takes the output tables of different processing software and creates a unified table useful for further processing, as well as a detailed report, including several different plots, descriptions, and statistical analysis. It allows a fast analysis and visualization of proteomic datasets, reducing user time to a minimum. 1010

1011

P-III-0852 The sky is the limit: a cloud-based proteomics platform for the masses Daniel Zolg1, Markus Schneider1, Patroklos Samaras1, Samia Ben Fredj1, Florian Seefried1, Dulguun Bold1, Layla Eljagh1, Tobias Schmidt1, Siegfried Gessulat1, Martin Frejno1 1MSAID, Garching, Germany Background: Laboratories dealing with bottom-up proteomics data often encounter computational hurdles in the journey from raw data to conclusive insights. Challenges arise from the absence of automated pipelines and disjointed local infrastructure for data storage, processing, systematic result management, and interpretation. The advent of fast-scanning instruments exacerbates these issues by overwhelming local infrastructure with a multitude of files and large raw data sizes. Here, we introduce a highly scalable, fully automatable, cloud-based proteomics platform designed to streamline the entire workflow. Methods: Our cloud-native platform comprises microservices operated on AWS™ and orchestrated by Kubernetes. Users can access the platform through either a command line client or a browser-based interface, both interacting with a API governing all platform functionalities. Raw data undergoes processing using Chimerys 4 on an elastic compute cluster. Results are stored in a data lake and can be explored directly in the browser or downloaded. Metadata annotation facilitates navigation and contextualization of numerous files. Platform access is available through subscription or self-hosted deployment. Results: We present a comprehensive, managed solution for proteomics data management, obviating the need for user-managed pipelines and infrastructure. The platform offers an intuitive web interface for collaborative data upload, management, and processing. File transfer occurs at speeds up to 100 MB/s into scalable object storage. Raw data can be annotated with metadata via a searchable tag system, simplifying organization and retrieval. A scalable compute cluster enables simultaneous processing of DDA, DIA, and PRM data from thousands of files. The platform is algorithm-independent, currently supporting Chimerys 4.0 with plans for additional search engines. We demonstrate the scalability by processing 1k files without significant increase in processing time compared to single file processing. Processed data can be organized using the same tag system employed for raw data, with the processing overview providing immediate insight into key parameters for data quality assessment. A fast post-processing workflow combines individually searched raw files, facilitating longitudinal data acquisition and processing without overheads. Results can be accessed via API- or browser- based download, direct API access to the result data lake, browser-based data exploration, or a customizable visualization dashboard featuring common data analyses and visualizations.In conclusion, this managed, automated proteomics data pipeline promises to streamline the journey from raw data to insights, particularly benefiting laboratories lacking the resources to develop and maintain in-house solutions. 1011

1012

Chemical biology insights (P-III-0853 to P-III-0875) P-III-0853 Development of chemical proteomics enabling tissue-selective proteomic and interactomic analysis Xiucong Bao1,2, Siyue Huang2, Qiao Ran3, Xiaomeng Li2, Chaogu Zheng3, Xiang Li2 1HKU, Biomedical Sciences, Hong Kong, Hong Kong 2HKU, Chemistry, Hong Kong, Hong Kong 3HKU, Biology, Hong Kong, Hong Kong Multicellular lives, such as plants and animals, are complex because different cells inside the organism must differentiate into cell types with distinct morphologies and functions. This differentiation process relies on the expression of different sets of proteins in various cell types and their subsequent assembly into molecular machinery through protein-protein interactions. However, our current understanding of such protein-protein interactions often lacks cellular contexts because they were usually studied in an in vitro heterozygous system or in cultured cells isolated from the tissue environment. Effective methods to interrogate protein-protein interactions in a tissue-specific manner are largely missing. To close this technology gap, we developed a novel chemical biology approach to label proteins from specific cells with a bifunctional amino acid probe that carries both an affinity tag for proteome extraction and a photo- crosslinker for capturing protein-protein interaction. This new method, termed Methionine Analog-based Cell- Specific Proteomics and Interactomics (MACSPI), enables the profiling of tissue-specific proteomes and the identification of tissue-specific protein interactors in intact organisms and can be applied to a wide range of biological problems, such as the study of organ development and disease pathogenesis. We designed and synthesized an unnatural amino acid that is structurally similar to methionine but has two additional modules, an alkyne group that can be used as a handle for protein purification through "click chemistry" and a diazirine group that mediates covalent linkages of interacting proteins upon photoactivation. This bifunctional probe is named "photo-ANA". Next, we engineered a methionyl-tRNA synthetase (MetRS) variant that is capable of attaching photo-ANA to methionyl-tRNA. By controlling the expression of this MetRS mutant in specific tissues, only proteins from the tissue of interest are labeled by the photo-ANA probe and extracted by affinity purification. Moreover, with photo-crosslinking, protein complexes from specific tissues can be captured and isolated. We applied MACSPI in Caenorhabditis elegans, a model multicellular organism, to selectively label, capture, and profile the proteomes of the body wall muscle and the nervous system, which led to identifying new tissue-specific proteins. We also successfully profiled HSP90 interactors in muscles and neurons and identified tissue-specific interactors and stress-related interactors. Our study demonstrates that MACSPI can be used to profile tissue-specific proteomes and interactomes in intact multicellular organisms. Fig. 1 (optional) 1012

1013

P-III-0854 nanoBRET: shedding light on binding kinetics of covalent inhibitors in living cells Elena Borregan Ochando1, Bas Housmans1, Martine Vrints1, David Moreno Delgado1 1Galapagos, Mechelen, Belgium Covalent drugs feature a reactive functional group that establishes covalent bonds with protein targets, augmenting binding affinity beyond non-covalent interactions. Many early covalent drugs were discovered serendipitously and are known to bind active sites to inhibit enzymatic activity (e.g. Aspirin). Over the last three decades, there has been growing interest in the rational design of covalent drugs, with a focus on targeting certain amino acids to enhance selectivity. The prolonged engagement of target proteins by covalent drugs offers unique pharmacodynamic profiles. Despite concerns regarding reactivity, the potential advantages of covalent drugs have motivated medicinal chemists to explore this domain. Through careful optimization balancing reactivity, selectivity, and potency, several safe and efficacious drugs have been developed (e.g. ritlecitinib, ibrutinib and Osimertinib). By combining the power of mass spectrometry to decipher the mechanism of action of study drugs with nanoBret proximity assay, we were able to assess the specificity and affinity of compounds to the target proteins in a biochemical and cellular context. Fig. 1 (optional) 1013

1014

P-III-0855 Igniting drug discovery: pushing CETSA®-MS to scale Jana Flegel1, Shiva Ahmadi1, Tommaso Mari1, Rieke Berger1, Chiara Sinsteden1, Alana Gerlach1, Hannah Bock1, Sonja Anlauf1, Sebastian Essig1 1Bayer AG, Life Science Technologies / Chemical Biology, Wuppertal, Germany For the development of effective and safe drugs, a clear understanding of the direct drug-protein interactions and the mode of action in a physiologically relevant biological system is of highest relevance. A high number of drug candidates fail throughout all stages of drug development due to lack of efficiency or undesired side effects potentially caused by off-target binding. These failures lead to an immense loss of resources and time, which ultimately translates into a prolonged suffering time for patients from their diseases. Consequently, understanding of the molecular interactions of biologically active small molecules at very early stages is crucial for early de-risking, compound prioritization and data-driven allocation of research resources to the most promising drug development projects. Over the past years, chemical proteomics approaches have increasingly gained popularity within early drug discovery for unbiased target and off-target identification, target engagement assays as well as to facilitate compound mode of action deconvolution. Among others, the CEllular Thermal Shift Assay (CETSA®) technology is emerging as a key technology for drug discovery programs. Multiple studies have shown that CETSA® coupled to mass spectrometry (MS) is a powerful tool that allows for label-free target deconvolution in complex samples such as cell lysates, intact cells and tissues. However, although multiplexing approaches were reported over the past years, CETSA®-MS workflows remain laborious. Additionally accounting for the required MS measurement time, the throughput of this technology is still very limited. Here, we will provide an overview about the compressed CETSA® MS platform that we established at Bayer AG. The improvements we implemented led to a largely automated workflow that enables proteome-wide target engagement studies at higher throughput. Here, we showcase how this platform is used for flexible and fast evaluation of compound-protein interactions at scale, and how we use it to support ongoing drug discovery projects with drug-candidate prioritization and early de-risking of drug-development programs. 1014

1015

P-III-0856 Comprehensive characterisation of CDK4/6 inhibitor trilaciclib in acute myeloid Leukaemia using proteomics approaches Amy George1, Andy Frey1, Maria Emilia Dueñas1, José Luis Marín-Rubio1, Matthias Trost1 1Newcastle University, Newcastle upon Tyne, United Kingdom Cyclin-dependent kinases 4 and 6 (CDK4/6) are important mediators of cell cycle progression and are essential for the growth of many cancer types. Inhibition of these kinases can elicit varied phenotypic responses that are contingent on cell type, determining clinical utility. Acute myeloid leukaemia (AML) with FLT3 internal tandem duplication (FLT3-ITD) mutation is an aggressive type of haematological malignancy that exhibits hypersensitivity to CDK4/6 inhibition pre-clinically, but the biological mechanism underlying this effect remains largely unexplored. Here, we aim to delineate the mode of action of trilaciclib, a CDK4/6 inhibitor, for repositioning in the treatment of AML through comprehensive proteomic analyses. MOLM-13 cells (FLT3-ITD+) were treated with 1 µM trilaciclib or vehicle control for 24 hours, with sample collection at 0hr, 1hr, 6hr and 24 hr time points to investigate (i) initial responses by thermal proteome profiling, (ii) cellular reprogramming by phosphoproteomics, and (iii) expression proteomics reflecting the induced phenotype after 24 hours. At 1hr, thermal proteome profiling was conducted to identify proteins exhibiting altered stability due to ligand- binding, post-translational modifications, and protein-protein interactions within live cells. Cells were aliquot, heated at 8 temperatures between 40-68 °C and then lysed by freeze-thawing. The soluble protein fraction was quantified, and equal amounts digested using 96-well S-Trap™ plates. Peptides (500ng) underwent single-shot analysis by 40SPD diaPASEF using Evosep One LC coupled to a TimsToF-HT mass spectrometer, identifying more than 6000 proteins. Thermal shifts in CDK4 and CDK6 from ligand binding were not detectable, but AP2 associated kinase 1 (AAK1) established a clear significant thermal shift in all four replicates. To complement this, cell pellets were collected at 1hr, 6 hr and 24 hr and lysed with 500µL 5% SDS, 50mM TEAB. Protein (3 mg) was digested using S-Trap™ Midi-columns and 1 µg injected for whole-cell proteomic analysis. Phosphopeptides were enriched from the remaining digest using in-house TiO2-based microcolumns, then labelled using TMTpro 16plex (Fisher Scientific, UK), pooled, separated into 12 fractions by basic-pH reversed- phase offline-LC, and analysed with an Orbitrap Fusion Lumos Tribrid mass spectrometer. Raw DIA data was analysed using DIA-NN version 1.8 in library-free mode, while TMT-DDA MS files were analysed with MaxQuant. Trilaciclib treatment resulted in the downregulation of many proteins associated with S-Phase transition and DNA synthesis reflecting inhibited CDK4/6 activity and consequential cell cycle arrest in G1 phase. We highlight signalling pathways implicated in cellular response to CDK4/6 inhibition in leukaemia with constitutive FLT3 activity. This study provides a detailed understanding of trilaciclib's effects in AML, elucidating its mode of action through advanced proteomic analyses. 1015

1016

P-III-0857 Global analysis of metabolically regulated nucleic acid-protein interactions Maximilian Hornisch1, Claudio Bassot1, Eliska Fuerst1, Tomas Koudelka1, Ilaria Piazza1 1Max Delbrück Center for Molecular Medicine (MDC), Berlin, Germany Metabolite-protein interactions control a wide range of cellular processes. One prominent example is the allosteric network that regulates the activity of metabolic enzymes to ensure that the cell controls the biochemical reactions and maintain homeostasis. In bacteria this network not only includes metabolic enzymes but expands to proteins that act on gene expression such as the lac repressor whose DNA-binding activity is inhibited by the metabolite allolactose. Similar mechanisms, in which a metabolite-protein interaction regulates the activity of RNA binding proteins, have also been reported in mammals for individual cases. However, it is not clear how prevalent these mechanisms are in mammalian cells. Here, we generalized the limited-proteolysis mass spectrometry (LiP-MS) structural proteomics method to study the crosstalk between metabolite-protein and nucleic acid-protein interactions. Our results provide evidence that ATP binds to RNA-binding domains and that Acetyl-CoA interacts with the DNA-damage- repair complex DNA-PK. We examine how ATP affects RNA binding and how Acetyl- CoA affects the DNA-PK complex activity, highlighting the importance of metabolism in controlling cellular programs. 1016

1017

P-III-0858 Deciphering the complexity of combinatorial therapies through advanced target engagement deconvolution assays in acute myeloid leukemia Elham Gholizadeh1, Ehsan Zangene1, Rabah Soliymani1, Caroline Heckman1, Amir Ata Saei2, Mohieddin Jafari1 1Helsinki University, Biochemistry and Developmental Biology, Helsinki, Finland 2Karolinska Institute, Stockholm, Sweden Combinatorial therapy (CT) has demonstrated significant potential in overcoming the limitations of single-drug treatments and mitigating cancer cell resistance in patients. Despite advancements in CT, a substantial knowledge gap persists in deconvoluting drug targets and identifying downstream pathways or mechanisms of action (MoA) due to the inherent complexity of these combinations. This study aims to develop a computational workflow to identify drug targets and elucidate the MoA for two recently proposed CTs for acute myeloid leukemia (AML): LY3009120-Sapanisertib (LS) and Ruxolitinib-Ulixertinib (RU). We employed a comprehensive proteome integral stability and solubility assay (CoPISA), an advanced computational development of PISA, to deconvolute and characterize the protein targets of these drug combinations. Using four AML-related cell lines (MOLM13, SKM, NOMO, and MOLM16), we analyzed protein extracts and intact cells after single and paired drug treatments, compared to control. Our study successfully identified both primary and secondary targets of the aforementioned drug combinations for AML, as well as for the individual drugs. For LS, we identified 564 direct and indirect protein targets, with 55 being primary targets after considering single and paired drug treatments in addition to control. For RU, 793 direct and indirect protein targets were identified, with 65 primary targets implicated in this combinatorial therapy. Pathway analysis revealed significant involvement of apoptotic signaling and cell cycle regulation pathways, suggesting mechanisms through which these drug combinations exert their effects, resulting in reduced toxicity on normal cells and enhanced efficacy on cancer cells. Our workflow, CoPISA, demonstrates substantial potential in the high-throughput analysis of drug-target interactions and the elucidation of drug MoA by facilitating comprehensive comparisons of target sets. This approach is instrumental in understanding the mechanisms behind the reduced toxicity and enhanced efficacy of LS and RU compared to first-line treatments for AML. Furthermore, the proteomics-based insights derived from CoPISA are pivotal in refining combinatorial therapy strategies, thus contributing to the optimization of therapeutic regimens and improving clinical outcomes for AML patients. This advancement in proteomics not only enhances our understanding of drug interactions at the molecular level but also paves the way for personalized medicine approaches in oncology. Fig. 1 (optional) 1017

1018

P-III-0859 Comprehensive glycation profiling of Glargine by high-resolution mass spectrometry Swaraj Jathar1,2, Ashtamy MG1, Shivani Palkar1, Shiva Shankar S1,2, Reema Banarjee1, Mahesh J. Kulkarni1,2 1CSIR – National Chemical Laboratory, Biochemical Sciences, Pune, India 2Academy of Scientific and Innovative Research, Ghaziabad, India Background: Methylglyoxal is a highly reactive α-oxoaldehyde metabolite formed during glycolysis that reacts with proteins in circulation leading to the formation of methylglyoxal-derived hydroimidazolones (MG-H) which can further be modified into advanced glycation end products (AGEs). AGEs activate the RAGE signaling pathway which leads to increased oxidative stress, inflammation, and insulin resistance. In diabetes, hyperglycemia and oxidative stress lead to increased plasma methylglyoxal concentrations and consequently, increased AGE formation. Glargine, the first long-acting insulin analog for diabetic therapy, has a prolonged plasma half-life that allows once a day injection. However, this also makes it susceptible to glycation which can affect the biological activity of the drug. Glargine being one of the most widely prescribed insulin analogs for diabetic therapy, it is crucial to determine how its function is affected by glycation. In this study, we have aimed to characterize glycated glargine to identify possible modifications. Methods: Glargine was glycated in vitro with methylglyoxal at different pH for 4 days. Samples were collected every 24 hours and intact protein mass spectrometry was performed using a Q-Exactive High-Resolution Mass Spectrometer (HRMS) coupled to a UHPLC system. The mass spectra were deconvoluted using UniDec software to determine the mass shift and adduct formation. The observed mass shifts were validated by an extracted ion chromatogram (XIC). Results: In this study, we observed that the acidic buffer condition was most suitable for the Glargine glycation study. 50% of glargine was found to be modified within 24 hours of methylglyoxal treatment with carboxyethyl lysine/arginine being the most abundant glycated proteoform. The amount of unmodified glargine gradually decreased to 18% after 96 hours of reaction. Day one did not show more than two sites of modification which increased to four on day two. Only after 72 hours of reaction, there is an appearance of five MG-H modifications. Ours is the first study to identify the glycation of glargine by methylglyoxal. Glycation can lead to structural changes in glargine leading to reduced binding to insulin receptor and loss of activity. Furthermore, glycated glargine can bind to RAGE and aggravate insulin resistance. References P. B. Walke, S. B. Bansode, N. P. More, A. H. Chaurasiya, R. S. Joshi, and M. J. Kulkarni, "Molecular investigation of glycated insulin-induced insulin resistance via insulin signaling and AGE-RAGE axis," Biochimica et Biophysica Acta (BBA)-Molecular Basis of Disease, 2021. R. Hilgenfeld, G. Seipke, H. Berchtold, and D. R. Owens, "The evolution of insulin glargine and its continuing contribution to diabetes care," Drugs, 2014. J. Xue et al., "The receptor for advanced glycation end products (RAGE) specifically recognizes methylglyoxal- derived AGEs," Biochemistry, 2014. 1018

1020

P-III-0860 POWER-QC – assessment of covalent enantiomer libraries for fragment-based drug design by harnessing the power of proteome-wide, site-specific chemoproteomics Karsten Krug1, Laura Marholz1, Emily Lachtara1, Elizabeth Admirand1, Daniel Erlanson1, Robert Everley1 1Frontier Medicines, Boston, MA, United States Frontier"s drug discovery engine combines chemoproteomics, covalent fragment-based approaches, and AI and allows us to target >90% of the proteome. Screening fragments has emerged as a powerful approach in drug discovery, especially when utilizing a covalent strategy, in which a covalent bond can irreversibly capture an otherwise weak ligand. Frontier"s covalent fragment library is custom-built and leverages the concept of enantiomers or "enantiopairs", molecules with identical atomic composition but spatially arranged as non- superimposable mirror images of each other. Because the chemical reactivity of enantiomers is identical, discovery of enantiomer-specific (ES) hits provides immediate structure-activity relationships (SAR). Careful quality control (QC) of enantiomer libraries is crucial to minimize identification of false-positive ES hits, e.g. if one fragment has lost activity due to synthetic impurities or degradation. However, compound QC would not capture downstream problems such as sample preparation or enantio-selective interactions with efflux transporters. Here we introduce ProteOme-Wide Enantio-Ratio QC (POWER-QC) to complement standard compound QC by leveraging high-throughput mass spectrometry-based chemoproteomics. We apply POWER-QC to a subset of our covalent enantiomer library, assess its ability to flag compounds with potential QC problems, and demonstrate its utility to differentiate between true and false positive ES hits. Proteome-wide mapping of site-level protein-compound engagements was accomplished by an optimized streamlined cysteine activity-based protein profiling (SLC-ABPP) protocol utilizing TMT-based quantification of DMSO/compound competition ratios (CR). Selectivity and reactivity scores were calculated from proteome-wide chemoproteomics data as hit ratios and as 95th percentiles of the CR distribution. Derived scores should be similar for the two members of an enantiopair, and metrics to quantify deviations were assessed. The significance of the extent of deviations between enantiopair members was determined empirically using known QC-failed examples. We calculated proteome-wide scores for each compound which exhibited the expected distributions across different warhead classes, and which correlated well with estimated promiscuity scores derived from molecular physical properties of the compounds. 95% of a subset of our enantiomer fragment library passed POWER-QC, suggesting most enantiopairs showed concordant selectivity and reactivity. Comparison to compound purity assessments revealed only partial agreement, suggesting POWER-QC as a complementary QC method. We discovered several examples of ES hits to active sites of enzymes passing POWER-QC, providing promising starting points to modulate the catalytic activity of these enzymes. POWER-QC is a valuable addition to Frontier"s toolbox and is routinely used in our hit triage process for cellular screening campaigns and orthogonal biochemical assays. 1020

1021

P-III-0861 Development of a mass spectrometry-compatible chemical probe for protein citrullination enrichment Rebecca Meelker Gonzalez1, Sophia Laposchan1, Erik Reidel1, Chien-Yun Lee1 1Technical University of Munich, School of Life Sciences, Freising, Germany Citrullination is a post-translational modification (PTM) on arginine catalyzed by peptidylarginine deiminase (PAD) enzymes. Despite the significance of citrullination in health and disease, the absence of efficient MS- compatible enrichment tools has hindered the exploration of how this modification influences the functions of proteins at the proteome-wide level. Here, we present a cleavable, MS-compatible chemical probe for the enrichment of citrullinated peptides, based on two commercially available compounds, making it readily accessible to the community to explore functional implications in diverse biological contexts. Citrullinated peptides were derivatized by 4-azidophenyl glyoxal, followed by a click reaction with dde-biotin alkyne that adds a biotin residue with a cleavable linker. After enrichment by streptavidin beads, the peptides are cleaved off, leaving a mass tag of 212 Da. The peptides were then measured in LC-MS/MS and analyzed by MSFragger (v20.0) using a custom variable modification. First, we optimized reaction conditions such as incubation time, temperature, probe ratio, etc., using synthetic citrullinated peptides. Then, we spiked them into tryptic HeLa peptides to assess platform sensitivity in a complex background. The detection limit of the single peptide is ~7.8 nmol/µg. Next, we generated in vitro citrullinated HeLa peptides with 0.1 to 2% citrullination content to assess enrichment efficiency. As a result, we identified a 2 to 7-fold increase in citrullinated peptides, and the enrichment selectivity (intensity-based) is up to 90% for the sample with the highest citrullination content. Finally, we extended the capacity to a 96-well format to demonstrate its high-throughput ability and quantitative reproducibility. In conclusion, this study presents a robust enrichment workflow for protein citrullination at the peptide level. This enhances the depth of citrullination proteomics analysis using commercially available probes and enables high-throughput citrullination analysis in various sample types from diverse biological contexts. 1021

1022

P-III-0863 Profiling the targetable ubiquitin system with proteomics and chemical probes Matthew Maitland1, Jonathan St. Germain2,3, Tanner Tessier4,5, Shili Duan1,2, Cheryl Arrowsmith1,2,3, Brian Raught2,3, Dalia Barsyte-Lovejoy1,6 1University of Toronto, Structural Genomics Consortium, Toronto, Canada 2University Health Network, Princess Margaret Cancer Centre, Toronto, Canada 3University of Toronto, Medical Biophysics, Toronto, Canada 4The Children's Hospital of Philadelphia, Protective Immunity, Philadelphia, PA, United States 5University of Pennsylvania Perelman School of Medicine, Pathology and Laboratory Medicine, Philadelphia, PA, United States 6University of Toronto, Pharmacology & Toxicology, Toronto, Canada The ubiquitin machinery, including E3 ligases and deubiquitinases, are critical regulators of cellular processes and can be chemically targeted with therapeutic potential. For example, PROteolysis TArgeting Chimeras (PROTACs) are bifunctional compounds that recruit an E3 ligase to a disease-relevant neo-substrate for targeted protein degradation. To uncover the intricacies of this system and identify new ubiquitin pathways for disease intervention, comprehensive profiling using mass spectrometry-based proteomics is essential. In this study, we present an extensive proteomic profiling of the ubiquitination machinery and investigate cellular responses to their targeted modulation. Using proximity-dependent interactomics and quantitative proteomics, we profiled deubiquitinases and E3 ligase substrate receptors. Our analyses, conducted in the presence and absence of selective and potent ligands, revealed a broad spectrum of non-degradative and degradative substrates. These findings offer insights into the dynamic regulatory roles these enzymes play beyond their well-known function in protein degradation. Moreover, we developed a novel serial enrichment strategy for ubiquitin proteomics, which enables the identification of ubiquitin sites induced by PROTAC treatment on a neo-substrate. This innovative approach demonstrates high specificity and sensitivity, and can be easily extended to other post-translational modifications and applications. Our study highlights the versatility and complexity of the ubiquitination machinery and underscores its potential as a therapeutic target. The methodologies we have established provide robust tools for dissecting ubiquitin- related pathways and their implications in health and disease. 1022

1023

P-III-0864 Phenotypic drug discovery screen in human iPSC-derived macrophages using novel MALDI-TOF MS workflows and in-depth diaPASEF proteomics Leonie Mueller1, Roland Annan2, Maria Emilia Dueñas1, Matthias Trost1, Rachel Peltier-Heap2 1Newcastle University, Biomedical Sciences, Newcastle upon Tyne, United Kingdom 2GlaxoSmithKline, Stevenage, United Kingdom In the past, there has been a significant demand to develop rapid, label-free, and physiologically relevant drug discovery assays; anticipating growing need for high-throughput (HT) compatible cellular assays. MALDI-TOF MS has played a key role in expanding the label-free HT screening environment of biochemical assays and it is routinely used for rapid phenotyping of prokaryotic cells. I will present a novel workflow that distinguishes human iPSC-derived macrophage phenotypes by MALDI-TOF MS based on cell-specific fingerprints. The workflow is utilised to characterise ~200 small molecule compounds and to analyse macrophage response to different bacterial stimuli. Accompanying proteomics analysis was carried out with HT compatible gradients on a timsTOF instrument. We utilised iPSC-derived macrophages from multiple human donors and polarised them into a pro-inflammatory phenotype with bacterial (LPS, M.tb, E.coli, Staph) and cytokine (IFNγ) stimuli in the presence or absence of compounds. The supernatant was collected to measure secreted cytokine levels and the frozen cell pellets were utilised for MS analysis. Samples were prepared with S-Trap and analysed on a Evosep coupled to the timsTOF HT; acquired with diaPASEF and a 60 sample per day (60SPD) method. For MALDI-TOF MS, the lysed cell pellet and matrix were spotted onto the target plate and acquired with a speed of one second per sample on the rapifleX MALDI PharmaPulse. We implemented workflow automation at different stages which enabled cell culture in 96 well plates and data acquisition from as little as 5000 cells. We can show robust and reproducible separation between non-inflammatory and pro-inflammatory macrophage phenotypes across multiple human donors in a PCA, and with "biomarker" peaks from the lipid and metabolite region of the mass spectrum. As a first proof of concept screen, the Broad Institute JUMP compound set was utilised; demonstrating assay robustness and identification of common phenotypic screen hitters. The second set contained ~100 well-annotated compounds from a Chemogenomics library, some with macrophage inflammatory pathway targets. The MALDI-TOF MS results were compared against the conventional label-based approach: secreted pro- and anti-inflammatory cytokine levels. A large overlap between the identified hit compounds in the screens, and a correlation between MALDI-TOF MS "biomarker" peaks and the cytokines were observed. The proteomics workflow was optimised to allow HT measures of all compounds to investigate the underlying signalling pathways. Further, macrophage clustering was observed according to the bacterial stimuli by MALDI-TOF MS and investigated by proteomics, indicating M.tb infection unique targets. We describe a novel MALDI-TOF MS based HT compatible cell phenotyping workflow and its combination with proteomics to address needs in the drug discovery field. 1023

1024

P-III-0865 Deconvolute the controversial antiviral activity of Hydroxychloroquine through target identification with chemical proteomics Jin Shang1 1The HK polyU, ABCT, Hong Kong, Hong Kong Hydroxychloroquine (HCQ), a traditional clinical anti-malarial medicine, has shown broad-spectrum activities, including anticancer, anti-inflammatory, and antiviral effects. Since the outbreak of coronavirus disease 2019 (COVID-19), HCQ has been extensively investigated due to its antiviral properties. However, the therapeutic effect and mechanism of how HCQ inhibits coronavirus are unclear, which limits clinical applications of HCQ and poses risks to its ongoing clinical trials. In this study, we utilized chemical biology and proteomics approaches to reveal a lysosomal acid protease as a binding partner of HCQ and related antiviral mechanisms. A photo-crosslinking probe mimicking HCQ was synthesized to purify proteins that physically bind to HCQ. Next, multiple mass-spectrometry-based analyses were conducted to validate the interactions. Furthermore, HCQ significantly inhibited the enzymatic activity of the target protein and, thus, the endosomal entry pathway of coronavirus, as shown in a pseudovirus assay. Finally, by comparing HCQ"s inhibitory effects on different SARS-CoV-2 variants with or without gene knockdown, we demonstrated that the anti-COVID-19 mechanism of HCQ was based on the inhibition of the target protein. Our study not only revealed a novel binding target of HCQ but also provided a promising antiviral therapy agent against endosomal pathway-dependent viruses. Fig. 1 (optional) 1024

1025

P-III-0866 Target identification, selectivity profiling and binding site mapping of small molecule and peptide drugs by LiP- MS Liliana Malinovska1, Polina Shichkova1, Fabio Sabino1, Luca Räss1, Ino Karemaker1, Roland Bruderer1, Yuehan Feng1, Lukas Reiter1 1Biognosys AG, Schlieren, Switzerland Target identification in the cellular context is a critical step for both target-based and phenotypic drug discovery. Limited proteolysis coupled with mass spectrometry (LiP-MS) has emerged as a powerful technique for target deconvolution of small molecules or peptides in cell lysate without compound modification or genetic manipulation of cell lines [1,2,3,4]. Utilizing a non-specific protease under well-controlled conditions, LiP-MS exploits drug-induced structural alteration or steric hindrance on protein targets and utilizes quantitative mass spectrometry to probe over 250"000 peptides covering more than 9000 proteins in the proteome. To aid target ID, this workflow incorporates a 7-concentration dose response experiment and a machine-learning framework [1] to compute a LiP-score to rank identified target proteins, as well as predict potential binding site. In this study, we present target deconvolution results on four small molecule compounds and one peptide compound with very distinct pharmacological profile: 1) staurosporine, a broad-specific kinase inhibitor; 2) roniciclib, a failed clinical-stage pan-CDK inhibitor; 3) selumetinib, an allosteric, non-ATP-competitive MEK 1 and 2 inhibitor; 4) NST-628, a newly developed non-degradating molecular glue inhibiting the RAS/MAPK pathway; and 5) Acetyl-calpastatin, a 24 amino-acid polypeptide inhibitor of calpain 1 and 2. Target ID experiment with staurosporine and roniciclib yielded > 150 and 68 kinase targets were identified respectively. For the pan-kinase inhibitor roniciclib, both cell-cycle CDKs (CDK1/cyclin B, CDK2/cyclin E, CDK4 cyclin D) and transcriptional CDK9 were among the targets. For Selumetinib, LiP-score ranking identified MEK1 and MEK2 as the top two targets, and only kinases among a total of 23 targets. Furthermore, mapping the LiP- peptides from this experiment onto MEK1 showed that they are in close proximity to the selumetinib allosteric binding site on co-crystal structure. Last but not least, we assessed the target landscape of NST-628, a novel non- degrading molecular glue and identified MEK1 and MEK2 clearly as the primary targets. This finding is in line with orthogonal data obtained from X-ray crystallography [5]. Collectively, this data demonstrates that LiP-MS can be deployed to effectively identify protein drug targets and predict binding sites in complex cellular millieu independent of the compound"s mechanism of action and without compound modification or labeling. These capabilities make LiP-MS a powerful addition to the target deconvolution toolbox. In addition, LiP-MS protocol can be fully automated to enable small molecule target ID following phenotypic drug screening or at lead optimization stage with higher throughput. [1] Beaton et al. Nature Comm 2020 [2] Hendricks et al. ACS Chem Bio 2021 [3] AACR 2022 [4] AACR 2023 [5] AACR-NCI-EORTC 2023 1025

1026

P-III-0867 Endogenous cell membrane interactome mapping for GLP-1 receptor via ligand-based proximity labeling Ting Dang1, Jie Yu1, Min Zhuang1, Wenqing Shui1 1ShanghaiTech University, School of Life Science and Technology, Shanghai, China GLP-1 receptor, one of the most successful targets for therapeutic drug development against type 2 diabetes and obesity, is known to engage multiple intracellular proteins to initiate different signaling pathways. However, due to technical challenges, it remains much less explored how the receptor interaction with proteins on the cell membrane mediates its signaling activity and physiological effects. Here, we present a ligand-based proximity labeling approach to be integrated with quantitative proteomics so as to interrogate the native cell membrane interactome for the GLP-1 receptor upon agonist simulation. With this approach, our study identified a number of unreported cell membrane interactors for the endogenous receptor in both a pancreatic β cell line and a neuronal cell line, which revealed interaction landscapes distinct from those previously mapped in less physiological cellular systems. We further demonstrate this strategy can be exploited to uncover new regulators of GLP-1 receptor-mediated signaling and insulinotropic response in β cells. Additionally, we obtain a time- resolved cell membrane interactome map for the receptor in β cells. Therefore, our study provides a new approach that is generalizable to map endogenous cell membrane interactomes for GPCRs so as to decipher the molecular basis of their cell-type-specific functional regulation. Fig. 1 (optional) 1026

1027

P-III-0868 The timsTOF Ultra enables deep global ubiquitinomics of ultra-low protein input samples for validating degraderdrug targets Philipp Strohmidel1, Martin Steger2, Uli Ohmayer2, Björn Schwalb2, Torsten Müller1, Christian Albers1 1Bruker Daltonics GmbH & Co. KG, Bremen, Germany 2NEOsphere Biotechnologies GmbH, Planegg, Germany Targeted protein degradation (TPD) is a promising new drug modality to remove specific cellular proteins. The assessment of degraders for TPD via mass spectrometry (MS)-based proteomics is particularly appealing for the unbiased discovery of novel targets for molecular glues, since rational drug design proves challenging for those molecules. Once identified by global proteomics, such drug target candidates need to be mechanistically validated, for example by demonstrating drug-induced target protein ubiquitination. We present a global ubiquitinomics workflow for detecting degrader drug induced protein ubiquitinations of ultra-low protein input samples, by combining an improved ubiquitin-remnant peptide enrichment workflow with single-shot LC-MS on the timsTOF Ultra, and a custom-made statistical analysis pipeline. All samples were analyzed using a nanoElute® 2 (Bruker Daltonics) equipped with an Aurora Ultimate CSI 25 cm column (IonOpticks) coupled to either a timsTOF HT (Bruker Daltonics) or timsTOF Ultra (Bruker Daltonics). Chromatographic separation was done at 50°C on a 38 min active gradient with 45 min total runtime and a flow rate of 250 nL/min. The mass spectrometers were operated in dia-PASEF® mode using a slice-PASEF (1 frame - 15 windows) scheme. Raw data processing was performed using dia-NN 1.8.2 beta 22 and an in-house developed pipeline was used for downstream statistical analysis. Quadruplicates of degrader drug-treated samples of low protein input (100 µg, no proteasome inhibitor added) were enriched for ubiquitin-remnant peptides (diGly of K-GG peptides) and split into three before data acquisition by slice-PASEF, on both the timsTOF HT and the timsTOF Ultra. Additionally, injection triplicates of samples with higher protein input (300 µg) were acquired on both systems. Peptides were separated on an Aurora Ultimate CSI 25 cm column and a nanoElute 2 nano-LC in both cases. Using a predicted spectral library in dia-NN, sample measurements on the timsTOF HT resulted in 14,500-17,600 identified K-GG peptides when excluding the samples of higher protein input. By including the high input samples to the data processing with match between runs (MBR), the number of identified K-GG peptides was increased to 16,600-23,500. Strikingly, when comparing the low-input samples without MBR, the number of identified K-GG peptides was increased by 52% on average on the timsTOF Ultra. Inclusion of high-input samples for MBRs led to increased K- GG peptide identifications to 27,000-34,000, corresponding to an increase of 46% as compared to the timsTOF HT. The high scanning speed (FWHM in scans: 4.5) of the timsTOF instruments resulted in an excellent quantitative precision (CVs between 7.1% and 8.6% per condition, on protein level) Finally, a direct comparison of both LS-MS platforms revealed a much better sensitivity of the timsTOF Ultra for detecting degrader drug-induced ubiquitination events. 1027

1028

P-III-0869 Spatiotemporal and global profiling of DNA-protein interactions and substrates of lysine-modifying enzymes in living cells Minjia Tan1, Hao Hu1, Xiaohua Chen1, Linhui Zhai1 1Shanghai Institue of Materia Medica, Chinese Academy of Sciences, Shanghai, China Precise dissection of DNA-protein and enzyme-substrate interactions is essential for elucidating the recognition basis, dynamics, signaling transduction and gene regulation mechanism. However, global profiling of weak and dynamic DNA-protein and enzyme-substrate interactions remains a long-standing challenge. Here, we establish a light-induced lysine (K) enabled crosslinking (LIKE-XL) strategy for spatiotemporal and global profiling of DNA- protein interactions. Harnessing unique abilities to capture weak and transient DNA-protein interactions, we demonstrate that LIKE-XL enables the discovery of low-affinity transcription-factor/DNA interactions via sequence-specific DNA baits, determining the binding sites for transcription factors that have been previously unknown. More importantly, we successfully decipher the dynamics of the transcription factor subproteome in response to drug treatment in a time-resolved manner and find downstream target transcription factors from drug perturbations. Using a similar light-induced chemical crosslinking tool, we further develop a strategy to directly capture substrates of lysine-modifying enzymes via PTM-acceptor residue crosslinking in living cells, enabling global profiling of substrates of PTM-enzymes and validation of PTM-sites in a straightforward manner. By integrating enzymatic PTM-mechanisms, and genetically encoding residue-selective photo-crosslinker into PTM-enzymes, our strategy expands the substrate profiles of both bacterial and mammalian lysine acylation enzymes, including bacterial lysine acylases PatZ, YiaC, LplA, TmcA, and YjaB, as well as mammalian acyltransferases GCN5 and Tip60, leading to discovery of distinct yet functionally important substrates and acylation sites. Together, our strategies offer complementary approaches to expand the DNA-protein profiling and substrates of PTM-enzyme identification that were previously inaccessible via non-covalent strategies. References: 1. Hu H, et al. Spatiotemporal and direct capturing global substrates of lysine-modifying enzymes in living cells. Nat Commun. 2024, 15:1465. 2. Guo A, et al. Spatiotemporal and global profiling of DNA-protein interactions enables discovery of low- affinity transcription factors. Nat Chem 2023, 15:803-814 Keywords: light-induced crosslinking; DNA-protein interactions; transcription factors; lysine-modifying enzyme substrates; proteomics analysis 1028

1029

P-III-0870 Proteomics signature in mouse brain cortex at different for post-mortem intervals: a preliminary study for forensic biomarker identification Nicola Tupone1, Vimal Di Virgilio1, Martina Bonelli2, Fabrizio Di Giuseppe1, Patricia Giuliani2, Gianluigi Giammaria3, Giuliano Ascani4, Enrica Rosato5, Maurizio Locatelli5, Renata Ciccarelli2, Christian D'Ovidio2, Stefania Angelucci1 1G.d'Annunzio University, Proteomics Lab at Center for advanced Studies and Technology_CAST, Pescara, Italy 2G.d'Annunzio University, Center for advanced Studies and Technology_CAST, Chieti, Italy 3Ravenna Medical Center (GVM Care and Research), Ravenna, Italy 4UOSD Maxillofacial surgery, Azienda Sanitaria Locale di Pescara, Pescara, Italy 5University “G. d’Annunzio” of Chieti-Pescara, Department of Pharmacy,, Chieti, Italy Post-mortem interval (PMI) estimation is still based on empirical analysis of macrodata not always endowed with the reliability required in forensic field. Death promotes tissue decomposition due to biomolecules degradation, especially proteins. In order to improve the accurate PMI estimation, identifying sensitive post-mortem biomarkers, was applied herein for the first time a gel-based proteomics approach on mouse model over post mortem time course. A specific and reproducible evaluation of tissues for thanatochronological purposes is strictly dependent on the condition of biological source, so that tissue handling should be optimized. To this aim, we investigated the proteome changes in cortex samples of 6-weeks old female SAMR1 mice over a post- mortem time course. After death, brain tissue was removed immediately (T0) and after 4, 8, 12, 24 and 32 h, maintaing animals at 4°C. Dissected tissues were frozen at -80C°until processed. Proteomic analysis, performed on sampes related to early and late PMIs (24 h post-mortem, respectively) showed protein level changes as compared to T0 samples, with a remarkable increase in Calpain11 in the early PMI, as well as in Caspases 7 and 8 together Gasdermin 3 in late PMI. These findings were confirmed by LIFT mass spectrometry technology and western blot analysis and, although requiring further investigation in other biological samples, suggest that these proteins could be considered as putative biomarkers of different PMIs. The primary aim was to identify potential new biomarkers useful for PMI assessment as well as additional parameters for sample quality measurements. 1029

1030

P-III-0871 Addressing retromer dysfunction via PPI-orchestrating small molecules as a therapeutic strategy for neurodegeneration Emily Samuela Turilli-Ghisolfi1,2, Nicholas Bergna1, Marta Lualdi1, Valentina Cappelletti2, Paola Picotti2, Bruno Figadère3, Tiziana Alberio1, Mauro Fasano1 1Università degli Studi dell'Insubria, Department of Science and High Technology, Busto Arsizio, Italy 2ETH Zürich, Institute of Molecular Systems Biology, Zurich, Switzerland 3Université Paris-Saclay, BioCIS, Faculté de Pharmacie, CNRS, Orsay, France Alzheimer's disease (AD) and Parkinson's disease (PD) are the most frequent neurodegenerative conditions worldwide, and the urgency to discover disease-modifying therapies represents a strong unmet medical need. Both proteinopathies involve pathological protein-protein interactions. To unveil proteome-wide different protein interactions in cellular models of neurodegeneration, we set up a structural proteomics method combining Limited Proteolysis-coupled Mass Spectrometry (LiP-MS) with Terminal Amine Isotopic Labeling of Substrates (TAILS) in the SH-SY5Y cellular model of altered dopamine homeostasis, an early step in PD pathogenesis. Among the proteins that changed their surface exposure, dopamine showed to affect retromer-mediated trafficking, and in particular the cargo-recognition core (composed by the vacuolar sorting protein VPS35-VPS26-VPS29 heterotrimer). Validation by indirect immunofluorescence of VPS35 allowed us to hypothesize cytoplasmic accumulation of dysfunctional vesicles. Our observations are in line with increasing evidence highlighting dysfunctional endolysosomal trafficking as a potential pharmacological target in neurodegenerative diseases. Here, we propose the rational design of therapeutic molecular glues, i.e., small molecules to enhance the stability and restore the functionality of the retromer complex, by stabilizing the VPS35-VPS29 and the VPS35- VPS26 protein-protein interfaces. Two distinct small molecule libraries were considered: a laboratory library, and one generated and refined through a computational pipeline based on physicochemical properties favoring oral bioavailability (Lipinski rules), cell permeability and blood brain barrier crossing, and, when available, considering cytotoxicity, binding profiles, and additional early Absorption, Distribution, Metabolism and Excretion (ADME) data or further pharmacokinetics information in the case of more advanced compounds. After a computationally-enabled determination of the binding pockets of interest on the VPS35-VPS29 and the VPS35-VPS26 complexes, molecular-docking based in silico screening was performed with flexibility conditions for the amino acid side chains of the most relevant residues in the protein sites identified, and complete flexibility for the ligand candidates. After collecting SAR insights and performing hit optimization, the mechanism of action of the resulting lead compounds will be verified in vitro. In conclusion, this approach offers a new therapeutic strategy for treating neurodegenerative diseases by leveraging molecular glues to stabilize intracellular protein-protein complexes. 1030

1031

P-III-0872 Real-time Isotopic signature targeted profiling Chu Wang1 1Peking University, College of Chemistry and Molecular Engineering, Beijing, China Data-dependent acquisition (DDA) mode is widely applied in shotgun proteomics. However, restricted by the scanning speed of mass spectrometers, it"s difficult for DDA to directly detect peptides with low abundance. Herein, we introduce isoSTAR as a real-time isotopic signature targeted profiling platform, which identifies target peptides by their unique isotopic signatures during the full-MS scanning stage and schedule targeted MS/MS scans in no time. The isoSTAR showed dramatical improvement of sensitivity and identification of targets compared to traditional acquisition methods, which significantly enhanced the discovery of novel PTMs and the profiling of biomacromolecule interactions in practice. We envision that isoSTAR can become a convenient and powerful tool in shot-gun proteomics, providing enhanced and accurate peptide profiling. 1031

1032

P-III-0873 Functional immunoproteomic modification of human antigen processing enzymes by bactericidal small molecule and allergen methylisothiazolinone Sebastian Wolf1,2, Hermann-Josef Thierse1,2 1German Federal Institute for Risk Assessment, Chemical and Product Safety, Berlin, Germany 2Heidelberg University, Medical Faculty Mannheim, Mannheim, Germany Objective: Contact allergies to isothiazolinones in humans are a common health problem worldwide, affecting hundreds of thousands of consumers and workers in their daily quality of life. Adaptive immune responses to the widely used preservative methylisothiazolinone (MI) require cellular uptake and the involvement of antigen processing and presentation pathways. Electrophilic MI molecules form disulfide bonds with cysteine thiols of self-proteins (haptenation) which are processed to generate T-cell epitopes. In the case of antigen processing cysteine proteases, such as lysosomal cathepsins, cysteine thiol residues provide structural protein integrity to ensure catalytic activity. Possible MI-induced disruption of these structures may impair homeostatic antigen processing and lead to unexpected immunomodulation. In this immunoproteomic study, we examined (1) potential exogenous, post-translational MI modifications of cysteine-proteases of the cathepsin family (B, S), which play a crucial role in antigen processing and presentation; and (2) cathepsin enzyme activity and enzyme kinetics, potentially affected by transient or persistent MI-protein modifications. Methods: Classical Orbi-Trap LC-MS/MS analyses were applied to identify position-specific MI-modified cysteine residues on cathepsin B (& S), which is essential for MHC class II peptide loading. Data of chemical-specific modifications were analyzed in more detail by Peaks software. Complementary functional enzyme assays were performed and kinetic characteristics compared in the presence or absence of MI-protein modifications. Moreover, complementary immune cell effects were studied (FACS analyses). Results: (1) LC-MS/MS studies revealed several specific binding sites for MI, apart from the catalytic center. (2) MI-protein modification demonstrated significant cathepsin B (& S) enzyme inhibition, which was not neutralized by lysosomal and physiological concentrations of reducing agent glutathione (GSH). (3) Enzyme kinetics with moderate MI concentrations (MI-protein modification) indicated e.g. a non-competitive inhibition of cathepsin B by MI, which supported the LC-MS/MS observation (1). Conclusion: MI-protein modifications of antigen processing enzymes may cause altered immune cell responses due to novel generated T cell epitopes, independent of initially modified self-proteins, by directly interfering with the homoeostatic antigen processing machinery in human immune cells. In particular, the MI-specific inhibition of cathepsin B (& S) might contribute to novel T cell epitopes by (1) generation of non-homoeostatic immunopeptidomes and (2) by loading incompletely processed peptides into MHC class II molecules for antigen presentation (neo-antigens, immunogenic peptides, low affinity peptides) to variable human T cell receptors. 1032

1033

P-III-0874 High-throughput PROTAC compound screening workflow for targeted protein degradation on an Orbitrap Astral mass spectrometer with accurate label-free quantitation Kevin Yang1, Tonya Pekar Hart1, Amirmansoor Hakimi1 1Thermo Fisher Scientific, San Jose, CA, United States Targeted protein degradation (TPD) is a transformative strategy in drug discovery that utilizes cellular protein degradation processes to selectively eliminate deleterious proteins. Directly designing small molecules that promote induced proximity to E3 ligase to facilitate protein degradation with selective is challenging in practice, Thereby, compound screening with reliable quantitation accuracy is a crucial phase in drug discovery in the TPD space. The need for screening large numbers of compounds with accurate quantitation makes high-throughput mass spectrometry-based workflows an obvious choice for ensuring accurate lead identification. In this study, LFQ-DIA workflows of various gradients with highly accurate quantitation on Orbitrap Astral mass spectrometer were successfully developed to cater the needs of compound screening. In our pilot experiments, VCaP prostate cancer epithelial cells were treated with different concentrations of ARCC-4, a PROTAC protein degrader for androgen receptor. Cells were lysed and digested with trypsin using an AccelerOme automated sample preparation platform. The ensuing peptides were loaded onto a 15-cm EASY- Spray™ or 5-cm Ionopticks Aurora UHPLC column and separated with nano or capillary flowrate in trap and elute workflow using a Vanquish Neo UHPLC system for a throughput of 60, 100, 180 and 300 samples per day. Eluted peptides were ionized with an Easy-Spray source and transferred into the Orbitrap Astral mass spectrometer for DIA analysis. Acquired data was processed by Spectronaut, DIA-NN or Proteome Discoverer™ software using CHIMERYS™ intelligent search algorithm by MSAID. The 100 SPD method enabled over 9000 proteins to be reproducibly identified and quantified with a protein group CV of approximately 10%. To explore the possibility of ultra-fast compound screening, we further increased the throughput to 180 and 300 SPD, which led to the identification of over 8500 proteins and close to 8000 protein groups, respectively. To achieve maximized proteome coverage, an essential component for the validation phase, we further extended this workflow to 60 SPD. We successfully identified approximately 10,000 proteins with the 60 SPD method, which signifies a near-complete proteome identification within the cell. In all the throughputs investigated herein, we observed decreased androgen receptor expression upon increasing ARCC-4 concentration while the vast majority of proteome remained unaffected. In addition, the quantitation accuracy for workflows of various throughputs was validated by three proteomes with different ratios. The observed ratio was found to align closely with the theoretical ratio, thereby affirming the reliability of our workflow. Together, our data demonstrate that the Orbitrap Astral mass spectrometer with Vanquish Neo UHPLC system enable high-throughput compound screening and comprehensive on-target/off-target validation in the development of targeted protein degradation. 1033

1034

P-III-0875 Aspirin inhibits proteasomal degradation and promotes protein aggregate clearance through K63 ubiquitination Yaoyang Zhang1, Jing Gao1, Yang Liu1, Kaiwen He1 1Interdisciplinary Research Center on Biology and Chemistry, Chinese Academy of Sciences, Shanghai, China Aspirin is a potent lysine acetylation inducer, but its impact on lysine ubiquitination and ubiquitination-directed protein degradation is unclear. Herein, we developed the reversed-pulsed-SILAC strategy to systematically profile protein degradome in response to aspirin. By integrating degradome, acetylome, and ubiquitinome analyses, we showed that aspirin impairs proteasome activity to inhibit proteasomal degradation, rather than directly suppressing lysine ubiquitination. Interestingly, aspirin increases lysosomal degradation-implicated K63- linked ubiquitination. Accordingly, we found that aspirin is able to reduce a-syn aggregates, the major pathologic protein of Parkinson's disease, in cultured cells, neurons, and PD model mice with a rescued locomotor ability. We further revealed that the a-syn aggregate clearance induced by aspirin is K63-ubiquitination dependent in both cells and PD mice. These findings suggest two complementary mechanisms by which aspirin regulates the degradation of soluble and insoluble proteins, providing new insights into its diverse pharmacological effects that can aid in future drug development efforts. 1034

1035

Organisation of the proteome (PPI) (P-III-0876 to P-III-0890) P-III-0876 Deciphering the Agrin-mediated signaling network of MuSK by proximity-dependent labeling Cinzia Barresi1, Alexandra Wallner1, Markus Hartl2, Weiqiang Chen2, Ruth Herbst1 1Medical University of Vienna, Center for Pathophysiology, Infectiology and Immunology, Vienna, Austria 2Max Perutz Labs Vienna, Mass Spectrometry Facility, Vienna, Austria The muscle-specific kinase MuSK is a receptor tyrosine kinase that regulates the formation and maintenance of the neuromuscular junction (NMJ), the synapse between the motor nerve and skeletal muscle fiber. MuSK activity is tightly regulated, and alterations of its signaling is responsible for serious diseases, such as congenital myasthenic syndromes or myasthenia gravis. MuSK activation requires both the nerve-derived Agrin signal and the co-receptor Lrp4, a member of the low-density lipoprotein receptor-related protein family. During development, Agrin is released from the motor nerve terminal into the synaptic basal lamina, and forms a complex with Lrp4 to indirectly induce MuSK homodimerization and activation of the intracellular kinase domain. The docking protein Dok-7 is recruited to MuSK to stabilize its activity and in turn is phosphorylated by MuSK to propagate the signal through a series of downstream effectors. Indeed, phosphorylated Dok-7 interacts with Crk and CrkL, ultimately leading to AChR clustering at the postsynaptic membrane. Although several proteomics studies have improved our understanding of MuSK signaling, many aspects that follow MuSK activation and particularly their temporal regulation remain unknown. Here we propose to decipher the MuSK proximity proteome using proximity-dependent biotin identification (BioID) coupled to mass spectrometry (MS). We have established a muscle cell line stably expressing MuSK fused to TurboID, a highly active biotin ligase, which allows experiments with high temporal resolution. Muscle cells were stimulated with Agrin for 2 h, 4 h and 8 h to uncover the early and late events in the MuSK signaling cascade. Altogether, we identified 74 proteins including known binding partners of MuSK as well as novel interactors, which were clustered according to their temporal regulation. Proteins enriched during early signaling events reveal that CrkL functions as a signaling hub to recruit different molecules for distinct signaling outcomes. These proteins are mainly involved in small GTPase- mediated signal transduction. Within this group, we identified the Crk binding partner Rapgef1 (Rap guanine nucleotide exchange factor 1). Interestingly, we found that Rapgef1 is required for the formation of AChR clusters, as knockdown experiments showed reduced phosphorylation of AChRs with consequences for the ability to form full-sized clusters in Agrin-stimulated myotubes. In conclusion, our proteomic study provides new insights into the proximity network of MuSK, uncovers the role of Rapgef1 in MuSK downstream signaling, and contributes to a better understanding of the complex mechanisms that regulate NMJ formation. 1035

1036

P-III-0877 Temporal dissection of NLRP3 protein-protein interactions in inflammasome regulation Emma Mckay1, Mohammad Arefian1, Rebecca Coll2, Ben Collins1 1Queen's University Belfast, School of Biological Sciences, Belfast, United Kingdom 2Queen's University Belfast, Wellcome Wolfson Institiute for Experimental Medicine, Belfast, United Kingdom The NLRP3 inflammasome is a central signalling hub of the innate immune system which mediates inflammation through pro-inflammatory cytokine release and inflammatory cell death (pyroptosis). Dysregulated NLRP3 inflammasome signalling underlies many human diseases. The NLRP3 pathway undergoes a complex two-step programme of priming and activation modulated by dynamic protein-protein interactions (PPIs) and post- translational modifications (PTMs). However, the contextual and temporal dynamics of these mechanisms, and how they act to regulate NLRP3, are not completely understood. Previous studies have shown that temporal dynamics of PPI networks and corresponding PTMs can be assessed by affinity purification coupled to quantitative mass spectrometry (AP-MS) (1,2) and correlations between protein complex formation and post- translational modifications can be assessed, indicative of functional relationships (2). This project aims to build a comprehensive temporal map of PPI/PTM dynamics in NLRP3 inflammasome activation identifying novel regulatory mechanisms and their functional consequences. To derive a time-resolved quantitative map of reorganization of protein-protein interactions and corresponding PTMs we affinity purified NLRP3 from cells across a time course of priming and activation stimuli. We have established and validated protocols for AP-MS of endogenously expressed twin-strep tagged NLRP3 following stimulation of 5x106 THP-1 monocytes cells. The enhanced sensitivity and speed of diaPASEF (3) enabled quantitative analysis over many conditions and time-points including 150+ samples from 38 distinct experimental conditions. Time points were selected to focus on poorly understood early events (i.e. within 15 minutes of stimulation) in both priming and activating stimulation conditions and examine NLRP3 interactome immediately following stimulus detection. High confidence interactors have been identified for each stimulation timepoint via filtering against untagged (wild type) NLRP3 background controls. These interactions have been validated by the presence of known NLRP3 interacting proteins, and include putative novel interacting partners. Ongoing analysis is examining the relationships of detected PPIs and PTMs of NLRP3 and will attempt to define their functional significance for NLRP3 priming and activation and its phenotypic consequences. We show that AP-MS using diaPASEF enables in depth quantitative analysis of NLRP3 protein interaction dynamics across finely resolved time points and diverse stimulation conditions providing new insights into this complex and clinically important innate immune signalling pathway. References 1. Collins BC, et al. Nature Methods. 2013;10(12):1246–53. 2. Zheng Y, et al. Nature. 2013;499(7457):166–71. 3. Meier F, et al. Nature Methods. 2020;17(12):1229–36. 1036

1037

P-III-0878 In situ membrane protein and flagellar enrichment increase the depth of large-scale bacterial interactomics studies using ion mobility and cross-linking mass spectrometry (XL-MS) Ashleigh Dale1, Dylan Harney1, Mark Larance1, Stuart Cordwell1 1University of Sydney, Camperdown, Australia Cross-linking mass spectrometry (XL-MS) is an invaluable tool for capturing protein-wide interactomes in an in vivo context, however it remains challenging for complex protein mixtures, particularly those including membrane-embedded and/or membrane-associated proteins. The Gram negative bacterial membrane consists of a variety of sub-structures that include a cytoplasmic inner phospholipid membrane containing hydrophobic transmembrane proteins, a liquid phase periplasm containing soluble proteins, the cell shape and rigidity determinant peptidoglycan (PG) layer and finally, an asymmetric outer membrane (OM) lipid bilayer consisting of phospholipids, OM proteins and external lipopolysaccharide. The membrane facilitates interactions with the environment, including the human host, and is critical for all aspects of virulence, including adherence, host cell invasion, secretion, flagellar motility and antimicrobial resistance. We developed membrane enrichment strategies for in situ cross-linking of membrane proteins and the flagella apparatus from the helical human gastrointestinal pathogen Campylobacter jejuni. We investigated different fractionation approaches, including offline size-exclusion chromatography (SEC) and online gas-phase fractionation with high-field asymmetric waveform ion mobility spectrometry (FAIMS), and then investigated how different protein digestion approaches and gradient lengths affect the success of these strategies in identifying cross-linked peptide species at both the sample preparation and acquisition levels. A hybrid MS2-MS3 fragmentation strategy with or without an internal-stepping compensation voltage scheme showed the benefits of FAIMS for XL-MS increases with sample complexity and gradient length by reducing interference and increasing signal:noise. Significantly, single mg injections were able to yield more cross-links that 8 mg of SEC fractions over the same acquisition time. The use of FAIMS and SEC with different proteases suggests that protease-specific cross-linked peptides respond differently to ion-based separation across several physicochemical properties. In situ membrane protein cross- linking revealed extensive interactions between the major human adherence factor CadF, PG-associated lipoprotein Pal and major OM protein MOMP providing evidence of how C. jejuni maintains membrane integrity. Flagellar apparatus cross-linking revealed flagellin multimerisation and conformational dynamics for the first time, including multimeric interfaces and filament stabilisation. In conclusion, proteome-wide interactome interrogation with XL-MS can be tailored to the needs of the researcher in terms of sample type, preparation, fractionation and acquisition. Optimisation of these parameters can allow, for example, identification of sub- proteome specific interactomes, provide greater depth and coverage of the proteome and interactome, and lead to increased cross-link identification and quality of interactome data. 1037

1039

P-III-0879 hu.MAP3: Data integration of 25k mass spectrometry experiments accurately identifies protein complexes Erin Claussen1, Samantha Fischer1, Sandra Orchard2, Henning Hermjakob2, Georg Kustatscher3, Kevin Drew1 1University of Illinois at Chicago, Department of Biological Sciences, Chicago, IL, United States 2European Bioinformatics Institute (EMBL-EBI), Wellcome Genome Campus, Cambridge, United Kingdom 3University of Edinburgh, Wellcome Centre for Cell Biology, Edinburgh, United Kingdom Macromolecular protein complexes carry out most functions in the cell including essential functions required for cell survival. The identification of protein complexes in human cells is a long-standing challenge. To address this challenge several high throughput techniques have been developed including affinity purification mass spectrometry (AP-MS), co-fractionation mass spectrometry (CF-MS), proximity labeling strategies, among others, aimed at identifying protein interactions across the human proteome. As a result of these advances, there exists large scale datasets (>25,000 mass spectrometry experiments) using these methods that cover different cell types, tissues, and parts of the proteome. This presented an opportunity to integrate these data to build a more complete and accurate human protein complex map, which we call hu.MAP3 (Fig1A). We identify over 15,000 protein complexes covering ~3/4ths of the human proteome, a substantial increase over previous complex maps. Of note, we place >1,200 highly uncharacterized proteins into complexes providing testable hypotheses as to their function. Further, we refine our complexes by integrating proteomics co-variation data (ProteomeHD) as well as AlphaFold structural models to identify mutually exclusive protein interactions (Fig1B). We expect our resource to be valuable to the broader research community. Fig. 1 (optional) 1039

1040

P-III-0880 Deep Interactome Profiling MS (DIP-MS), a novel interaction proteomics method for the deconvolution of immunoprecipitated protein complexes Fabian Frommelt1,2, Andrea Fossati2,3,4,5, Federico Uliana2,6,7, Fabian Wendt8, Peng Xue2,9, Moritz Heusel2, Bernd Wollscheid8, Ruedi Aebersold2, Rodolfo Ciuffa2, Matthias Gstaiger2 1CeMM, Vienna, Austria 2ETH Zürich, Institute of Molecular Systems Biology, Department of Biology, Zurich, Switzerland 3University of California San Francisco, Quantitative Biosciences Institute (QBI), San Francisco, CA, United States 4University of California San Francisco, Department of Cellular and Molecular Pharmacology, San Francisco, CA, United States 5J. David Gladstone Institutes, San Francisco, CA, United States 6ETH Zürich, Institute of Biochemistry, Department of Biology, Zurich, Switzerland 7JGU University, Mainz, Germany 8ETH Zürich, Institute of Translational Medicine (ITM), Department of Health Sciences and Technology (D-HEST), Zurich, Switzerland 9Guangzhou National Laboratory, Guang Zhou, China Proteins organize into dynamic modules to carry out most cellular processes. Ever-advancing analytical capabilities to study modular proteome organization holds great promise to pinpoint molecular lesions driving disease phenotypes. Affinity purification combined with LC-MS/MS is a powerful method to identify a protein"s set of direct and indirect interaction partners with high sensitivity, but cannot resolve the sum of purified proteins into instances of specific protein complexes within a single AP-MS experiment. We devised an integrated computational/wet lab method Deep Interactome Profiling by Mass Spectrometry (DIP-MS) which identifies distinct protein complexes from an AP sample at high sensitivity, This is accomplished by the integration of targeted affinity enrichment with blue native-PAGE separation, quantitative high-throughput DIA-MS and deep-learning-based signal processing. 96-well based MS-sample processing combined with high-throughput DIA-MS allows protein profiling from up to 70 BN-PAGE fractions within one day. In addition, we introduced PPIprophet for efficient data driven neural network based complex deconvolution from the acquired data. We applied DIP-MS to probe the modular structure of the human prefoldin (PFD) family of complexes, resolving distinct PFD holo- and sub-complex variants, complex-complex interactions and complex isoforms with new subunits that were experimentally validated. When benchmarked against AP-MS and SEC-MS, we could demonstrate the superior resolution and coverage of our approach. DIP-MS identified a novel low abundant PDRG1 based complex isoform of canonical PFD, which we validated by in-silico structural modelling and reciprocal AP-MS. Finally, we found that DIP-MS was sensitive enough to identify folding substrate for the complex-complex interactions of CCT/TRiC-PFD and the PAQosome complex, a super complex involved in the assembly of protein complexes. The results were validated by orthogonal DIP-MS experiments, literature derived and experimental AP-MS data. The exceptional sensitivity, resolution, and speed of DIP-MS promise to significantly advance our ability to link specific proteome organizational states to phenotypic outcomes in healthy and diseased conditions. Frommelt F, Fossati A, Uliana F, Wendt F, Xue P, Heusel M, Wollscheid B, Aebersold R, Ciuffa R, Gstaiger M. DIP- MS: ultra-deep interaction proteomics for the deconvolution of protein complexes. Nat Methods. 2024 Apr;21(4):635-647. doi: 10.1038/s41592-024-02211-y. Epub 2024 Mar 26. PMID: 38532014; PMCID: PMC11009110. 1040

1042

P-III-0881 Proximity proteomics reveals a co-evolved LRRK2-regulatory network linked to centrosomes Marita Eckert1,2, Pasquale Miglionico3, Francesca Izzi1, Natalia De Oliveira Rosa3, Marius Ueffing4, Francesco Raimondi3, Christian Johannes Gloeckner1 1German Center for Neurodegenerative Diseases (DZNE), Tuebingen, Germany 2University of Tübingen, Graduate School of Cellular and Molecular Neuroscience, Tuebingen, Germany 3Scuola Normale Superiore, Laboratorio di Biologia Bio@SNS, Pisa, Italy 4University of Tübingen, Center for Ophthalmology, Institute for Ophthalmic Research, Tuebingen, Germany The Leucine-rich repeat kinase 2 (LRRK2) not only plays a vital role in familial forms of Parkinson"s disease (PD) but is also considered as a risk factor for developing idiopathic PD. So far, no causative treatment for PD exists, ameliorating the progressive loss of dopaminergic neurons within the substatia nigra. The multi-domain protein LRRK2 represents a promising drug target for PD treatment. Besides a kinase and a G-domain, LRRK2 contains multiple scaffolding domains, not only being involved in keeping inactive LRRK2 in an auto-inhibited state but are also involved in protein-protein interactions making LRRK2 to a signaling hub possibly integrating multiple upstream signals or enabling a context-specific activation of downstream pathways by controlling the spatial localization of LRRK2. Therefore, an in-depth understanding of the LRRK2 protein-protein interaction (PPI) networks not only allows illuminating a deeper understanding of the associated signaling pathways but potentially also reveals new drug targets. Here, we present new BioID proximity proteomes of LRRK2 covering both known as well as yet unknown interactors. By clustering based on co-evolution within the PPI network combined with functional enrichment, we identified a subcluster showing highest co-evolution to LRRK2 which is enriched in cytoskeletal components linked to the centrosome and microtubules as well as primary cilia. In addition, by this approach, we could demonstrate that the resulting BioID datasets are enriched in direct interactions. Furthermore, structural prediction of binary interactions via AlphaFold-multimer revealed distinct groups of interactors engaging through specific interfaces and LRRK2 conformations. Finally, we identified distinct changes in the LRRK2 proximity proteome induced by the type I kinase inhibitor MLi2 and by RAB29 co- expression. These functional state-specific interactomes link LRRK2 to defined cellular sub-compartments. Noteworthy, MLi2-induced LRRK2 complexes are characterized by a significantly higher co-evolution and are not predicted to interact with the catalytic domains, particularly the kinase domain. In conclusion, our novel approach combining proximity proteomics with the analysis of co-evolution and structural modelling helped to stratify the primary BioID datasets and lead to the identification of molecular networks linking LRRK2 to centrosomal biology and primary cilia. 1042

1043

P-III-0883 Mass spectrometry-centric approaches identify the proteins and neuropeptides interacting with amyloid-β Chenxi Jia1 1National Center for Protein Sciences (Beijing), Department of Proteomics, Beijing, China Amyloid-β (Aβ) aggregates are recognized as initiators of Alzheimer’s disease, and their interaction with the nervous system contributes to the progression of neurodegeneration. Therefore, there is an urgent need for researchers to discover the biomolecules that interact with Aβ peptides to determine the molecular basis. To enable profiling of Aβ-interacting proteins, we developed the novel Peptide-ligand-induced Abundance Change of proTeinS (PACTS)-assisted TPP strategy for the identification of peptide-interacting proteins based on the peptide-ligand-induced change in protein abundance. The interactome of amyloid beta (Aβ) 1-42 in THP-1 cells was described and resulted in the identification of 103 interacting proteins. Validation experiments indicated that Aβ1-42 interacted directly with fatty acid synthase (FAS) and inhibited its enzymatic activity, providing insights into fatty acid metabolic disorders in Alzheimer's disease. Furthermore, we applied the PACTS- assisted TPP approach to analyze clinical samples to identify Aβ1-42-interacting proteins. We detected 115 proteins that interact with Aβ1-42 in human frontal lobe tissue. Further orthogonal validation revealed that Aβ1- 42 interacted with the AD-associated protein mitogen-activated protein kinase 3 (MAPK3), and knockdown of the Aβ1-42 amyloid precursor protein (APP) inhibited the MAPK signaling pathway, suggesting potential functional roles for Aβ1-42 in interacting with MAPK3. To enable profiling of Aβ-interacting neuropeptides, we established a native mass spectrometry (MS)-centric workflow for screening Aβ-interacting neuropeptides, and six out of twelve neuropeptides formed noncovalent complexes with Aβ species in the MS gas phase. Thioflavin-T fluorescence assays and gel separation indicated that leptin and cerebellin decreased Aβ aggregation, whereas kisspeptin increased this process. In addition, leptin and cerebellin attenuated Aβ-induced cytotoxicity, which was independent from the influence of metal ions. Leptin can chelate copper from copper-bound Aβ species, reducing cytotoxicity caused by the aggregation of Aβ and metal ion complexes. Overall, our study demonstrated that neuropeptides frequently interact with Aβ and revealed that leptin and cerebillin are potential inhibitors of Aβ aggregation, providing great insight into understanding the molecular mechanism of Aβ interacting with nervous system and facilitating drug development. Fig. 1 (optional) 1043

1044

P-III-0884 An integrative interactomic approach to assess the function of CASK and its involvement in neurodevelopmental disorders Elodie Monsellier1, Yushi Men1, Gilles Trave1, Etienne Coyaud2, François Delalande3, Christine Carapito3 1IGBMC, Integrated Structural Biology, Illkirch-Graffenstaden, France 2PRISM, University of Lille, Lille, France 3LSMBO, Strasbourg, France Proteins act in intertwined regulated networks and are the main effectors of biological processes, as well as the principal targets of drug treatments. Identification all the protein-protein interactions (PPIs) established by a protein in a particular context, its interactome, is a powerful approach to uncover its normal and pathological functions. We implemented a multipronged interactomic approach to decipher the functions of CASK (Ca2+/calmodulin-dependent serine protein kinase) and its involvement in some neurodevelopmental disorders (NDDs). CASK is a large multi-domain protein achieving multiple functions in the central nervous system (Figure 1A). Several severe NDDs are caused by CASK mutations whose dysfunctional consequences remain sketchy characterized, yet underpinning PPI perturbations are suspected. We generated a contextual and quantitative interactome of the CASK wt protein of unprecedent depth and robustness, by combining four complementary interactomic approaches: affinity-purification mass spectrometry (AP-MS), in vitro AP-MS (pouring cell lysates on immobilized recombinant bait column), proximity-dependent biotinylation (BioID), and our recently developed Holdup assay (a robust and precise chromatographic assay for high-throughput quantification of protein-ligand affinities, Kd). Our CASK interactome is highly relevant since significantly enriched in (i) previously identified CASK partners; (ii) GO terms associated to the known functions of the protein, and (iii) NDD- or ASD- (Autism Spectrum Disorders) associated genes (Figure 1B). Strikingly, this interactome also reveals a potential new, previously undescribed function of the protein, which calls for further investigation. Finally, we systematically and finely investigated the complementarity between the four interactomic methods we used, providing a rigorous analysis of the strengths and weaknesses inherent to each approach (Figure 1C). In the second step, we aim to demonstrate that some CASK-associated mutations exert their pathological mechanisms through global interactome perturbations (Figure 1D). We will analyze how these interactome perturbations correlate to phenotypic alterations. In the future, this pathological fingerprint of interactome perturbations will allow us to assess the pathogenicity of Variants of Unknown Significance (VUS), a critical question for diagnosing NDD. Together, our results will provide an in-depth understanding of CASK mutation- associated mechanisms. 1044

1046

P-III-0885 Human protein-protein interactome: IMEx's metadata on host tissue specificity and genetic variability of proteins Kalpana Panneerselvam1, Pablo Poras1,2, Noemí del-Toro3, Margaret Duesbury1, Livia Perfetto1,4, Sucharitha Balu1, Eliot Ragueneau1, Juan Jose Medina Reyes1, Sandra Orchard1, Henning Hermjakob1, IMEx Consortium Curators1 1European Bioinfomatics Institute, Cambridgeshire, United Kingdom 2AstraZeneca, Cambridge, United Kingdom 3Healx, Cambridge, United Kingdom 4La Sapienza University, Rome, Italy Molecular interaction networks provide maps to explore cellular processes from a systems perspective. Access to the experimental evidence and contextual metadata which influence the interaction outcome are critical for accurate interpretation of interaction dynamics. The IMEx consortium (www.imexconsortium.org) is an international collaboration that compiles molecular interaction data in full experimental detail from scientific literature and direct submissions. This information is freely available to the public through an open source, open data model. Each interaction receives an IMEx-level confidence score based on interaction type, detection methods, and the publications reporting the interaction. Factors affecting interaction strength are annotated at experimental, interaction, and participant levels, including clinical and experimental mutations, required or sufficient binding regions, post-translational modifications (PTMs), and external modulators such as agonists and antagonists, derived from low-throughput experimental evidence. Our curation strategy allows us to map interactions to the specific host tissue or cell type where they were detected. Recently, we completed mapping our host tissues and cell types to the Experimental Factor Ontology (EFO). Modern high-throughput methods like affinity purification-mass spectrometry (AP-MS) can generate protein-protein interactions (PPIs) in their close-to native state of expression. IMEx contains over 11,800 n-ary interactions, mostly AP-MS evidence captured in human neurons, including more than 3,000 interactions affected by mutations. Approximately 2,200 interactions have been detected in embryonic stem cells. In the bronchial epithelium, more than 3,800 interactions have been detected, with 3,500 of these involving mutation data. Additionally, 500 interactions have been identified in bronchial epithelial cells with a cystic fibrosis phenotype. Close to 500 interactions have been curated in human retinal tissue, over 500 in blood immune cells, and around 350 in human blood plasma and serum. Most references feature PPI data studied in cancer cell lines: over 8,500 interactions have been curated in colorectal cancer cell lines, 6,000 in human breast cancer cell lines, and more than 4,400 in human glioblastoma cell lines and tissue samples. We also capture proximity assays to expand our coverage and provide data on proximal protein partners in different cellular compartments. IMEx"s comprehensive approach to building metadata for each interaction, with consistent use of controlled vocabularies and ontologies, can generate interaction networks specific to various layers of information. This facilitates the detection of macromolecular assemblies under different conditions. 1046

1047

P-III-0886 Detailed investigation of the interactions of the GADD34 (growth arrest and DNA damage [GADD]‐inducible34) protein Aladár Pettkó-Szandtner1, Norbert Gyöngyösi2, Eszter Kálmán2, Orsolya Kapuy2, Krisztina Futosi3, Anna Hajdara4, Krisztina Veszelyi5, Mariam Vakhtangishvili2, Nikolett Szilárd2 1HUN-REN BRC Szeged, Core Facility, Proteomics Research Group, Szeged, Hungary 2Semmelweis University, Faculty of Medicine, Department of Molecular Biology at the Institute of Biochemistry and Molecular Biology, Budapest, Hungary 3Semmelweis University, Faculty of Medicine, Department of Physiology, Budapest, Hungary 4VRG Therapeutics Plc., Budapest, Hungary 5Semmelweis University, Faculty of Medicine, Institute of Translational Medicine, Budapest, Hungary Protein phosphatase 1 (PP1) is one of the most important serine-threonine phosphatases in eukaryotic cells, owning approximately two hundred regulatory subunits which are responsible for the specificity of its substrates. Among its regulatory subunits, GADD34 plays a well-known substantial role by influencing translation via dephosphorylation of the eucaryotic initiation factor 2α (eiF2α) which is a key reaction for recovery of protein expression after stress. Guanabenz, an approved antihypertensive drug acting on α-2 adrenergic receptors, inhibits GADD34 as well. Guanabnez also exhibits therapeutic potential for neurodegenerative disorders with repurposing Phase 1 studies ongoing. We aimed to investigate the interactome of GADD34 protein under different ER stress conditions. We performed proximity-dependent biotin labelling method using wt and mutant GADD34-TURBO-2XFLAG chimera proteins in parallel to co-IP experiments. The mutant fusion protein (GADD34(Δ578-617)-TURBO-2XFLAG) lacked an evolutionary conserved region, including the known binding site for eiF2α. We generated 3-3 independent stable cell lines for both types of chimaeras from adherent human embryonic kidney cells (adHEK). We compared the interactome of wt GADD34-TURBO to that of the deletion mutant both under control and ER stress condition, i.e after treatment with the SERCA-inhibitor thapsigargin, or mTOR-inhibitor rapamycin. We labelled potential interaction partners in vivo, followed by identification via mass spectrometry (MS). In addition, co-immunoprecipitation studies with both chimaeras followed by MS were conducted. Our investigation found numerous novel interaction partners providing new insights including e.g. cell cycle, apoptosis, autophagy, cytoskeletal system/motor proteins and vesicular trafficking. Significant Gene Ontology enrichment revealed the involvement of the GADD34/PP1 complex in different immune functions and neurodegenerative diseases as well. Our findings suggest potential new indications for guanabenz beyond its current approvals, while also opening new perspectives for drug research focusing on GADD34 inhibition. A new computational model based on our hits fits well with literature data for the regulation of autophagy via the mTOR-AMPK-GADD34 triangle. We suggest that the GADD34/PP1 complex is a main switch in the cell to counteract stress-induced phosphorylation in main cellular functions, promoting cell proliferation, that has been validated by functional tests as well. Treatment of human T cells and neutrophils by Guanabenz also validates our proteomic finding. We suggest new mechanisms of action for antiinflammatory effects and beneficial outcomes in neurodegenerative diseases of Guanabenz treatment. In summary, our data contributes to a deeper understanding of various cellular processes, including cell cycle regulation, autophagy, apoptosis and functions of immune cells besides its pharmacologic relevance regarding GADD34 inhibition. into different cellular functions, 1047

1048

P-III-0887 Introducing TIE-UP-SIN: a novel method for enhanced identification of protein-protein interactions Maximilian Schedlowski1, Stephan Michalik1, Tilly Hoffmüller1, Christian Hentschker1, Manuela Gesell Salazar1, Uwe Völker1, Alexander Reder1 1Center for Functional Genomics Universitymedicine of Greifswald, Functional Genomics, Greifswald, Germany Proteins are indispensable for life due to their versatility and specificity. They are involved in virtually every cellular process from catalytic reactions, structural integrity, signaling and communication to immune defense and pathogenicity. Their ability to adopt diverse structures and functions makes them central to the complexity and adaptability of all living organisms. As they usually interact with other proteins or cellular components to carry out complex biological processes they rarely function alone. Thus, identifying Protein-Protein-Interactions (PPIs) helps elucidating the complex networks and pathways that underlie cellular functions and organismal biology. Methods for studying PPIs include yeast two-hybrid screening, co-immunoprecipitation, mass spectrometry, fluorescence resonance energy transfer, computational approaches and many others. Although each method has its specific advantages, they all have disadvantages such as a high false positive rate, a high technical complexity and low efficiency, which makes precise identifications of specific interaction partners somtimes difficult. Here we present our TIE-UP-SIN method which represents a Targeted Interactome Experiment for Unknown Proteins by Stable Isotope Normalization. This method provides an in vivo approach for the identification of PPIs based on stable isotope labeling, reversible crosslinking and affinity purification coupled with highly sensitive PPI identification by mass spectrometry (MS). The isotope labeling is essential for normalizing MS sample data as it addresses variability, ensures accurate quantification, and enhances the reliability of comparative DIA-MS proteomic analyses. This leads to precise and reproducible results, facilitating robust biological insights eliminating high false discovery rates. The combination with a quick and cheap reversible crosslinking and affinity purification protocol makes this technique a powerful tool to stabilize transient or weak in vivo interactions and to specifically capture them in a highly specific purification process. Overall, TIE-UP-SIN is a robust and favorably priced method for studying PPIs, offering quick and reliable insights into the composition and architecture of protein complexes in a biological in vivo context. The utility of the approach will be presented using the well-described and essential housekeeping sigma factor SigA from Bacillus subtilis as an example. 1048

1049

P-III-0888 High-resolution complexome profiling identifies auxiliary subunits and supercomplexes of renal transporters Uwe Schulte1,2,3, Catrin Müller1,4, Alexander Haupt1, Wolfgang Bildl1, Maciej Kocylowski1, Akos Kulik1, Gerd Zolles1, Bernd Fakler1,2 1University of Freiburg, Institute of Physiology II, Freiburg, Germany 2Signaling Research Centers BIOSS and CIBSS, University of Freiburg, Freiburg, Germany 3Logopharm GmbH, March-Buchheim, Germany 4Novartis, Basel, Switzerland The kidney plays a central role in fluid and electrolyte homeostasis as well as for the excretion of xenobiotics. For this purpose, a multitude of transporters and other transmembrane proteins is expressed along nephrons with high density and distinct cellular localization. Peptides, metabolites and inorganic ions are mainly resorbed in the proximal tubules by epithelial cells carrying tightly packed microvilli (together forming the brush border). These membrane structures are densely packed with transporters, but how these are molecularly organized and functionally cooperate is largely unknown. To resolve the molecular composition of renal transporters we combined two proteomic approaches: (i) we used high-resolution complexome profiling (based on cryo-slicing Blue Native Gel mass spectrometry (csBN-MS); Müller et al., 2016, Schulte et al., 2023) to analyse a preparation of brush border-enriched membranes from mouse kidney. After native separation of gently solubilized protein complexes, the respective gel was cut into 280 slices à 0.3 mm which were tryptically digested and comprehensively analyzed by LC-MS/MS on a Orbitrap Exploris mass spectrometer. Proteins were identified with PEAKS and quantified using an elaborate in-house developed processing and profile building pipeline (Schulte et al., 2023). (ii) The sodium-phosphate transporter SLC34A1 (NPT2A) and other selected examples of newly identified transporter complexes were further investigated by multi-epitope affinity purification mass spectrometry (meAP-MS), i.e. using multiple target- specific antibodies in wildtype and target knockout kidney membrane solubilisates as well as target-unrelated antibody controls. Complexome profiling analysis provided more than 5000 detailed abundance-size distributions of proteins with strong representation of bona-fide brush-border transmembrane transporters. Inspection of the profile peaks revealed that the majority of transmembrane proteins are not existing as monomers but are assembled into several higher molecular weight complexes (as in Schulte et al., 2023). Apart from known homo- and heteromeric transporter complexes, a number of peak correlations indicated the formation of novel transporter supercomplexes and assembly of transporters with yet unknown subunits. These results were corroborated by subsequent meAP-MS analyses. Thus, the obtained NPT2A interactome comprised the known interaction partners PDZ1L and NHRF3, in addition to 20 proteins that had not yet been associated with sodium-phosphate transporters and their cell biology. Most notably, NPT2A was found to interact with a subset of transporters coincident with high-molecular weight peak clusters associated with NPT2A in the brush border complexome. The functional and structural implications of these findings are currently investigated by electrophysiological experiments (upon heterologous reconstitution) and ultrastructural methods (like freeze-fracture immuno- electron microscopy). 1049

1050

P-III-0890 Mapping the human proteome with physical access to DNA Jakob Trendel1, Simon Trendel1, Shuyao Sha1, Bernhard Küster1 1Technical University Munich, Proteomics and Bioanalytics, Freising, Germany In a human cell, DNA is packed in histones, RNA, and chromatin-associated proteins, forming a cohesive gel. At any given moment, only a specific subset of the proteome has physical access to the DNA and organizes its structure, transcription, replication, repair and other molecular functions essential to the way the genome is read and maintained. We have developed a "zero-distance" photo-crosslinking approach to quantify proteins in direct contact with DNA in living cells. Collecting DNA interactomes from human breast cancer cells, we present an atlas of over one thousand proteins with physical access to DNA, and hundreds of peptide-nucleotide crosslinks pinpointing protein-DNA interfaces with single amino-acid resolution. Differential comparisons of DNA interactomes from cells undergoing treatment with estrogen or genotoxic chemotherapy recapitulated the recruitment of key transcription factors and DNA damage proteins. This opens a direct way to explore genomic regulation in a hypothesis-free manner, applicable to many organisms and systems. Fig. 1 (optional) 1050

1051

Glycobiology insights (P-III-0891 to P-III-0916) P-III-0891 Characterization of the glycosylation pattern of HAVCR2 and OSTP following a glycoproteomics approach Mikel Azkargorta1, Mirane Florencio2, Iris Bermejo2, Ibon Iloro1, Javier Beaskoetxea1, Iraide Escobés1, Ana Ardá2, Luca Unione2, Jesús Jiménez-Barbero2, Félix Elortza1 1CIC bioGUNE, BRTA, Proteomics Platform, Derio, Spain 2CIC bioGUNE, BRTA, Chemical Glycobyology Lab, Derio, Spain Protein glycosylation is a common post-translational modification, involving glycan linkage to proteins via N- or O- linkages. It affects protein folding and stability, trafficking, differentiation, and cell–cell interactions, among others. In addition, aberrant protein glycosylation is a hallmark of cancer, crucial in malignant transformation, tumor development, invasiveness, and metastatic disease. The characterization of the glycosylation pattern of proteins directly involved in immune response-related processes, such as recognition of pathogens or tumoral cells, might provide valuable information for the understanding of this interaction process, helping drug or vaccine design. In this regard, MS-based approaches generate complementary information to NMR-based studies, providing specific site-assigment and occupancy information. In this work we present data on the glycan profile characterization of two biologically and clinically relevant overexpressed and purified proteins (HAVCR2 and OSTP), with important roles in immune response modulation and cell-cell interaction. Trypsin and chymotrypsin were used for their digestion, and the resulting peptides were analysed in a timsTOF Pro mass spectrometer (Bruker). Specific acquisition methods were used for the analysis of unmodified peptides and for the glycopeptides. The data was analysed using Byonic software (Protein Metrics). Altogether, our results reveal that the use of different proteases, together with specific MS acquisition methods, and efforts on the bioinformatic side (scan merging, wildcard search for unexpected glycan modifications or variants in Byonic) provide deep information on the glycosylation pattern of HAVCR2 and OSTP, that complements and details information obtained by NMR. 1051

1052

P-III-0892 DeGlyPHER-ing site-specific distribution of N-glycans on candidate vaccines against HIV Sabyasachi Baboo1, Jolene Diedrich1, Jon Steichen2, Torben Schiffner2, Kimmo Rantalainen2, Olivia Swanson2, Christopher Cottrell2, Oleksandr Kalyuzhnyi2,3, Alessia Liguori2,3, William Schief2,3,4,5, James Paulson1,2, John Yates1 1The Scripps Research Institute, Molecular Medicine, La Jolla, CA, United States 2The Scripps Research Institute, Immunology and Microbiology, La Jolla, CA, United States 3The Scripps Research Institute, IAVI Neutralizing Antibody Center, La Jolla, CA, United States 4The Ragon Institute of Mass General, MIT, and Harvard, Cambridge, MA, United States 5Moderna Inc., Cambridge, MA, United States Introduction: Site-specific N-glycosylation on many highly glycosylated candidate vaccines must be determined. This helps to estimate the impact of changing glycosylation on antigenicity and immunogenicity when developing vaccines and provides a measure for monitoring the consistency of GMP-grade vaccine production. With substantial number of vaccines being designed, it is essential to rapidly select candidates with glycan landscapes that most closely resemble a naturally-occurring immunogen. DeGlyPHER (Deglycosylation-dependent Glycan/Proteomic Heterogeneity Evaluation Report) is developed as a highly sensitive, rapid, and reproducible approach, that perfectly fits these needs. Methods: Using Proteinase K we generate overlapping peptides mapping to every N-glycosylation site, sequentially deglycosylate these peptides with Endo H and then PNGase F, hence creating residual mass signatures that are identified by LC-MS/MS, thus quantifying the degree of glycan occupancy, and the broad characterization of glycan processing – constituting the transformation of high mannose forms into complex forms owing to mannose residues being replaced by other "apical" monosaccharides. Results: DeGlyPHER has evaluated N-glycan distribution on >200 immunogens that are being developed as vaccines against HIV, SARS-CoV-2, Influenza and Ebola. We have DeGlyPHER-ed >150 vaccines being developed using the highly-promising germline-targeting approach against every important immunogenic epitope on HIV Env/spike-protein that elicits broadly-neutralizing antibodies (bNAb). E.g., [1] we scanned boost-vaccine candidates against the V3-glycan epitope that elicits BG18-like antibodies, to inform V1-loop design to cause high glycan occupancy, potentially limiting off-target responses; [2] we discovered that in 10E8 priming- immunogens comprising of membrane proximal external region (MPER) epitope grafted into a protein-scaffold to create a multi-valent nanoparticle, the nanoparticle component has much higher glycan occupancy than the epitope scaffold; and [3] we confirmed the plugging of the N276 "glycan-hole" as part of a "prime-boost" strategy based on VRC01 epitope. We also found that HIV Env gp41 is significantly better shielded with N-glycans when HIV Env is membrane-tethered – its native state on intact virus and in most mRNA vaccine strategies, thus promising a means to reduce unwanted base effects stemming from partial glycan-shielding of gp41. All these findings have contributed substantially to downstream non-human primate (NHP) studies and clinical trials. Conclusion: DeGlyPHER is informing HIV vaccine development in real time, being a much more rapid, sensitive, and robust quantitative alternative to "intact-glycan" glycoproteomic approaches, with excellent potential for adapting it to high-throughput platforms for bulk glycoprotein analysis. Funding: This work was supported by the NIH grants P41GM103533, UM1AI100663, UM1AI144462, and R01/R56AI113867. 1052

1053

P-III-0893 Proteomic, N-glycoproteomic and glycosaminoglycan comparison of A549 and BEAS-2B cell line-derived small extracellular vesicles Mirjam Balbisi1,2, Tamás Langó3, Virág Horváth1, Zoltán Varga4, Kinga Ilyés4, Nikolett Nagy3, Otília Tóth3, Beáta G. Vértessy3, Lilla Turiak1 1HUN-REN Research Centre for Natural Sciences, MTA-TTK Lendület (Momentum) Glycan Biomarker Research Group, Budapest, Hungary 2Semmelweis University Doctoral School, Budapest, Hungary 3HUN-REN Research Centre for Natural Sciences, Institute of Molecular Life Sciences, Budapest, Hungary 4HUN-REN Research Centre for Natural Sciences, Biological Nanochemistry Research Group, Budapest, Hungary Small extracellular vesicles (sEVs) are lipid-bound particles ranging from 30 to 150 nm in size, released from cells to facilitate cell signaling processes. While glycosylation is a significant post-translational modification often linked with cell signaling, exploring the glycosylation of extracellular vesicle proteins remains challenging, with limited studies in this area. In the present study, we developed a methodology to investigate the proteomic, N-glycopeptide, and chondroitin sulfate glycosaminoglycan profiles of sEVs derived from A549 lung adenocarcinoma and BEAS-2B non-tumorigenic cell lines. Before isolation, A549 cells were cultured in non-completed F12 medium for 72 hours, while BEAS-2B cells were cultured in BEGM medium for the same duration. The supernatant was then collected and subjected to centrifugation and filtering steps to remove cell debris, apoptotic bodies, and microvesicles. Small EVs were isolated by mini-size exclusion chromatography on custom-made columns, with fractions 3, 4, and 5 showing enrichment in sEVs and combined for further analysis. Characterization of sEVs included transmission electron microscopy for shape assessment and microfluidic resistive pulse sensing for size distribution analysis. Isolated sEVs were subjected to repeated freeze-thaw cycles and were divided into two parts. One part underwent trypsin digestion followed by acetone precipitation to enrich glycopeptides in the pellet, while the supernatant contained non-glycosylated peptides. The other part was subjected to digestion with chondroitinase ABC enzyme to release chondroitin sulfate disaccharides. Following purification, the collected peptides, N-glycopeptides, and GAG disaccharides were analyzed by nanoUHPLC-MS(MS). Six independent sEV isolations are currently performed from each cell line, with pure mediums serving as controls to exclude components not of cellular origin. Data analysis and subsequent statistical comparisons may reveal differences in proteomic, N-glycomic, and chondroitin sulfate profiles between the sEVs, suggesting potential clinical relevance of sEV glycosylation. Support of the Lendület (Momentum) Program of the Hungarian Academy of Sciences and Semmelweis 250+ Excellence PhD Scholarship is acknowledged. The Authors are grateful to Markus Ralser for donating the nanoAcquity HPLC to the group. 1053

1054

P-III-0894 Influence of N-Linked glycosylation on proteolytic processing Diana Campos1, Michael Girgis2, Mikell Paige2, Miloslav Sanda1 1Max Planck Institute for Heart and Lung Research, Scientific Service Group Mass Spectrometry, Bad Nauheim, Germany 2George Mason University, Department of Bioengineering, Fairfax, VA, United States The proteolytic process is crucial for a multitude of proteomic analysis by Mass Spectrometry (MS). One of the factors influencing this process is the presence of Post Translational Modifications (PTMs), such as N-linked as well as O-linked glycosylation, on specific amino acid residues. The N-linked glycosylation and proteolytic cleavage processes are explored in their individual contexts and interrelated dynamics. We aim to elucidate how N-linked glycosylation influences protease cleavage, especially within the framework of proteomics bottom-up approach. Here, we have systematically investigated how differently N-glycosylated proteins and synthetic peptides influence trypsin cleavage. The results provide evidence for this influence and hint for the direct effects of specific glycostructures near to protease recognition sites. This work identifies current challenges in understanding this complex relationship and suggests future research directions that could significantly impact widely used glycoproteomics workflows. By providing a detailed analysis of the interplay between N-linked glycosylation and protease cleavage, this work also contribute to a deeper understanding of their roles in protein processing and molecular biology, underscoring the importance of these processes in cellular functions and disease mechanisms. 1054

1055

P-III-0895 Glycoproteomics of plasma immunoglobulin A in patients with colorectal cancer Voraratt Champattanachai1, Chris Verathamjamras1, Amnart Khongmanee1, Juthamard Chantaraamporn1, Somchai Chutipongtanate1, Chantragan Srisomsap1, Jisnuson Svasti1 1Chulabhorn Research Institute, Talat Bang Khen, Lak Si, Thailand Analysis of intact glycopeptide and glycoproteins remains challenging because of the complexity of sugar attachment and difficulties in data interpretation. Immunoglobulin A (IgA), the second most antibody in blood, has two subclasses (IgA1 and IgA2) which contains two (N144 and N340) and five (N47, N92, N131, N205, and N327) N-liked glycosylation sites, respectively. In this study, we development a strategy for differential analysis of the intact site-specific N-glycopeptides of serum IgA between colorectal cancer (CRC) and healthy controls (HC). IgA was purified using Peptide-M from individual samples of patients with non-metastatic (NM-CRC, n=10) and metastatic CRC (M-CRC, n=10), and HC (n=11), and followed by in-solution tryptic digestion. The digested peptide samples were subjected to a nano liquid chromatography coupled tadem mass spectrometer with Orbitrap Mass Spectrometer using higher-energy collisional dissociation (HCD) for MS/MS fragmentation. Identification and quantitation of site-specific of N-glycopeptides were performed by label-free analysis of Progenesis QI and Byonic software using an in-house human IgA and N-glycan database containing 148 glycan forms. In total, several glycoforms in each sites were identified including 45 (N144/N131), 12 (N340), 6 (N92 and N205), 18 (N131) glycoforms, while none was detected in N47 and N327. The relative N-linked glycopeptides dominantly existed at N144/N131 with the average of 80% of all observed N-linked sites. Among 45 glycoforms at N144/N131, differential quantitation analysis showed that the level of HexNAc(5)Hex(3) was increased in a disease-dependent manner with statically different between HC and M-CRC (p<0.01). This strategy enables us to discover the detailed profiles of N-glycopeptides and the relative quantitation to determine the level changes in site-specific N-glycopeptides of patients with CRC. Fig. 1 (optional) 1055

1056

P-III-0896 Comparative study of salivary secretory IgA N-glycosylation of COVID-19 infected and vaccinated patients Eniko Gebri1, Anna Farkas2, Kinga Ágnes Bágyi3, Petra Polyák4, Andras Guttman2 1Faculty of Dentistry, University of Debrecen, Oral Medicine, Debrecen, Hungary 2Faculty of Medicine, University of Debrecen, Horváth Csaba Memorial Laboratory of Bioseparation Sciences, Research Center for Molecular Medicine, Doctoral School of Molecular Medicine, Debrecen, Hungary 3Faculty of Dentistry, University of Debrecen, Department of Operative Dentistry and Endodontics, Debrecen, Hungary 4Faculty of Dentistry, University of Debrecen, Oral Medicine Department, Debrecen, Hungary Albeit the vaccines developed during the pandemic caused by the SARS-CoV-2 virus reduced the number and severity of diseases, the course of the disease showed a high degree of variability between individuals. Due to the frequent mutation of the virus, there are many variants of it even today, which differ in their infectivity. Oral immunity plays a fundamental role in the susceptibility to infection and the severity of the disease. The aim of our research was to examine the changes in the N-glycosylation profile of salivary secretory immunoglobulin A (sIgA) as a result of infection with or without vaccination. Six people from 3 age groups (25-34 years; 35-59 years; ≥60 years) without acute or chronic diseases were included in the study. Six patients had confirmed SARS- CoV-2 infection within 1-6 months before sample collection, 2 were asymptomatic contacts; 5 patients were vaccinated and 1 was unvaccinated. The N-glycosylation pattern of sIgA, which was purified by affinity chromatography from resting mixed saliva samples obtained from patients before and after the pandemic (2018 and 2022, respectively) was analyzed using capillary electrophoresis with laser-induced fluorescence detection (CE-LIF). Student"s t test was used for statistical analysis. All experiments were approved by the Regional Institutional Research Ethics Committee, Clinical Centre, University of Debrecen (Ethical license: DE RKEB/IKEB 4948-2018, 6051-2022; 5570-1/2018/EKU). The sIgA N-glycosylation profiles of vaccinated patients differed significantly (p≤0.05) from the unvaccinated patient at both time points. Changes in the peak areas of some N- glycan structures (increase or decrease) were observed between unvaccinated and vaccinated patients after infection. The sIgA N-glycosylation profile of the vaccinated contacts without clinical symptoms showed the most significant changes. It can be concluded that both the SARS-CoV-2 infection and the vaccination resulted in a permanent, detectable changes in the N-glycosylation pattern of salivary sIgA. Changes in the resting mixed saliva sIgA N-glycan profile may reflect the severity of the course of the disease, susceptibility to infection, and effectiveness of vaccination. Further investigation on a larger patients cohort is planned. 1056

1057

P-III-0897 Glycoproteomic analysis of renal cell carcinoma FFPE tissue reveals differential glycosylation signature compared to normal tissue Hector Huang1, Rachel Rice1, Maryam Baniasad1, Gege Xu1 1InterVenn Biosciences, Translation and Development, South San Francisco, CA, United States Formalin-fixed, paraffin-embedded (FFPE) tissues are banked in great numbers as a cost effective and simple way to preserve clinical specimens for study at a later time. The preservation method cross-links proteins and dehydrates cells, but primarily maintains the structural integrity of the tissue, making it a valuable resource for clinical assessment. Molecular analysis of FFPE tissue can be challenging due to potential chemical artifacts leftover from the preservation process, although recent developments in sample preparation and analytical techniques have helped to make FFPE tissue an asset for biomarker discovery. In particular, advancements in mass spectrometric techniques have allowed a more comprehensive characterization of a tissue"s proteome, including post-translational modifications (PTM). Glycosylation is one class of PTM that has been shown to be informative of certain disease outcomes. Global profiling of the glycoproteome can help to elucidate the complex signature specific to a biological state. Here, we present a method to capture and analyze the glycoproteome from FFPE tissue samples that includes mechanical disruption and heat induced un-crosslinking of the tissue, detergent-based extraction and digestion of the proteome, and enrichment of glycopeptides through hydrophilic interaction chromatography. We estimate that ~ 10-12 mm3 of tissue, sliced into 10 μm sections, is sufficient material to produce a single replicate analysis of the glycoproteome. We compare, in three technical replicates, the glycoproteomic signature for a piece of FFPE tumor tissue from renal cell carcinoma (RCC) to a control of normal adjacent tissue (NAT) for the same individual subject and quantify over 2,000 unique protein groups and over 1,500 unique N- glycopeptides (derived from 169 glycoproteins) by LC-MS/MS through data dependent acquisition (DDA). We find more than 200 glycopeptides increased with statistical significance (p-value ≤ 0.05) in abundance in RCC compared to NAT, and more than 150 glycopeptides decreased with statistical significance (p-value ≤ 0.05) in abundance. Aggregate analysis of glycan subtypes reveal increased fucosylation in RCC compared to NAT. Altogether, we demonstrate a platform for glycoproteomic analysis of FFPE tissue with reproducible, quantitative results. This presents a unique opportunity to generate valuable insights into a number of diseases through the analysis of archived specimens. 1057

1058

P-III-0898 Integrated proteomic and Glycoproteomic characterization of diffuse large B cell lymphoma cell lines Yanlong Ji1,2, Sebastian Perner1, Chih-Hsuan Yeh1,2, Henning Urlaub1,2 1Max-Planck-Institute for Multidisciplinary Sciences, Goettingen, Germany 2University Medical Center Göttingen, Goettingen, Germany Diffuse large B-cell lymphoma (DLBCL) is the most common subtype of non-Hodgkin lymphoma (NHL) with a poor prognosis. DLBCL is also clinically heterogeneous, as demonstrated by genomic, transcriptional, translational and post translational molecular characterization. Although several studies have revealed disparate molecular profiles in DLBCL on gene expression, protein abundance, glycosylation occupancy levels, a large scale site-specific N-glycoproteome of DLBCL is still missing, which is crucial to determine N-glycan changes in DLBCL subtypes and explore the functional roles of glycosylation biosynthesis enzymes in DLBCL development. In this study, we perform proteomic and site-specific N-glycoproteomic quantitative analyses of 20 DLBCL cell lines (7 activate-B-cell-like (ABC) and 13 germinal center B-cell-like (GCB) subtypes). Our results identified and quantified 7800 glycopeptides bearing 457 glycans on 1074 glycosites from 636 proteins. Subsequent analysis based on the glycopeptide abundances classified the DLBCL cell lines into three major subtypes. Integrated proteomic and glycoproteomic analysis identified N-glycosylation changes, which correlated with subtype- specific expression of glycan biosynthesis enzymes. Furthermore, we proposed 3 glyco-processing models regulated by different glycan biosynthesis pathways in DLBCL. Specifically, in model 1 glycopeptides bearing larger glycan with more branches are highly enriched, in model 2 highly sialyated glycopeptides are the most abundant, in model 3 glycopeptides whose glycans own incomplete poly-N-acetyllactosamine (over/under- expressed galactose) show higher expression. Our results provided a valuable resource and shed light on understanding glycosylation regulations in DLBCL cells and identifying glycan biomarkers for DLBCL diagnosis. 1058

1059

P-III-0899 Sulfoglycomic identification of sulfated glycotopes underlies subsequent development of target-driven glycoproteomic workflows Kay-Hooi Khoo1, Pao-Yuan Wang1, Hua-Chien Chang1, Chu-Wei Kuo1 1Academia Sinica, Institute of Biological Chemistry, Taipei, Taiwan Glycans carried on proteins in rodent and human brains are well characterized by the expression of terminal GlcNAc, LacNAc, sialyl-LacNAc, LeX, and a low level of disialylated LacNAc in two different settings. In contrast, despite the well-known presence of the HNK-1 epitope, the sulfated glycotopes have rarely been analyzed and reported in present-day glycomic efforts, presumably due to their low abundance and unfavorable ionization properties for efficient mass spectrometry detection. Similarly, none of the many reported brain N- glycoproteomics to date have confidently identified any specific protein carriers of sulfated N-glycans, which severely impeded progress in gaining further glycobiology insights into how additional sulfation on glycans may modulate functions mediated by protein site-specific glycosylation. By our unique sulfoglycomic workflow based on negative mode MS analysis of permethylated glycans, we have, over the years, resolved and definitively identified a myriad of sulfated glycotopes carried on the mouse and human brains, including internal 6-sulfo- GlcNAc of a LacNAc, terminal 3 or/and 6-sulfo Gal, terminal 4-sulfo-GalNAc of a LacdiNAc, and HNK-1. Intriguingly, we have unambiguously shown that the core fucose of a subset of N-glycans, from simple truncated paucimannose to complex type multiantennary structures, could be sulfated, by a combination of diagnostic MS2 ions. Furthermore, glial and neuron cells isolated from murine cortex/striatum displayed cell type-specific expression of core Fuc-sulfation, whereby neurons carried the highest level of core Fuc-sulfated N-glycan structures, with lesser amount in astrocytes, and not detected in microglia. The regional and cell type-specific sulfoglycomic profiles are largely supported by the differential expression of the mRNA transcripts of various known and putative carbohydrate sulfotransferases. These could be manipulated to derive model cells and glycopeptides enriched with the target sulfated glycotopes, thereby serving as invaluable standards and models for our ongoing development of target-driven sulfo-glycoproteomic workflows. 1059

1060

P-III-0900 O-Glycan analysis by CE-MS Karthika Korumadathil Shaji1, Jing Zheng1, Hector Rhault Loponte1, Peter Horvatovich1, Guinevere Lageveen- Kammeijer1 1University of Groningen, Faculty of Science and Engineering, Analytical Biochemistry — Groningen Research Institute of Pharmacy, Groningen, Netherlands One of the most important co- and post-translational modification of a protein is glycosylation (the attachment of a sugar moiety group, also known as glycans). This modification plays a vital role in various biological events such as cellular communications, their signaling process, immune system, protein folding and stability. Minor changes in the glycosylation pattern are related to various diseases such as cancer. As glycosylation is involved in almost all biological events, it is an ideal biomarker candidate. Glycosylation is divided in various types, where protein glycosylation is either N- or O-linked. The latter is the least studied of the two due to the lack of generic enzymes, their often complex structure, large heterogeneity and the variety in core structures. Moreover, glycans can exist in various isomeric structures, which cannot readily be distinguished by solely mass spectrometry (MS). The requirement for a separation technique capable of analyzing low-abundant species, highlights the significance of capillary electrophoresis (CE) in recent years, because of its high sensitivity, ability to separate isomers, and the linkage analysis. Until now, the N-glycan class is the main form being explored by CE-MS at a released level, leaving O-glycosylation relatively unexplored. This is mainly related to the separation mechanism of CE, as it is important that the analytes carry a charge, which in glycomic workflows is often introduced by labelling the reducing end of the glycan. As common release strategies (β elimination) for O- glycans make the reducing end unavailable for labelling, the analytes are more complicated to analyze by CE-MS. Just recently, a new releasing procedure was introduced that protects the reducing end, opening up a range of new opportunities. Notably, this method allows the sequential release of both N- and O-glycome from the same sample. Therefore, this study will focus on extending the glycoanalytical CE-MS portfolio by adding this layer and comparing it to the already developed LC-MS workflow. Both N- and O-glycan were sequentially released from total plasma and fetuin. After the release, an internal standard (DP7) was spiked to the sample, followed by a labeling procedure with either 2AB or GirP for LC-MS or CE-MS, respectively. The first experiments, revealed a successful release of N- and O-glycans from fetuin, as the most abundant forms could easily be detected at 30 ng by CE-MS. To ensure the highest sensitivity possible, further experiments will focus on further optimalization the workflow by critically evaluating the sample preparation step. Additionally, the workflow will be assessed on different type of biospecimens (e.g. plasma, tissue and cells). Eventually, this will allow us to gain a deeper understanding of the glycome and, eventually, decipher glycomic intra-tumoral heterogeneity in, for example, head and neck squamous cell carcinoma (HNSCC). 1060

1061

P-III-0901 Comparative glycan analysis of 6 rAAV serotypes using multimodal glycan omics technologies Atsushi Kuno1, Yuki Yamaguchi2, Kentaro Ishii2, Sachiko Koizumi3,4, Hiroaki Sakaue1, Takahiro Maruno2,5, Mitsuko Fukuhara2,5, Risa Shibuya2, Yasuo Tsunaka2, Aoba Matsushita2, Karin Bandoh2, Tetsuo Torisu2, Chie Murata-Kishimoto3, Azusa Tomioka1, Saho Mizukado1, Sayaka Fuseya1, Hiroyuki Kaji6, Yuji Kashiwakura7, Tsukasa Ohmori7, Susumu Uchiyama2 1National Institute of Advanced Industrial Science and Technology, Cellular and Molecular Biotechnology Research Institute, Tsukuba, Japan 2Osaka University, Graduate School of Engineering, Suita, Japan 3GlycoTechnica Ltd., Yokohama, Japan 4Precision System Science Co. Ltd., Matsudo, Japan 5U-Medico Inc., Suita, Japan 6Nagoya University, Institute for Glyco-core Research, Nagoya, Japan 7Jichi Medical University School of Medicine, Shimotsuke, Japan Glycan analysis of viral vectors is important from the viewpoint of quality evaluation of pharmaceuticals. Due to recent advances in analytical technology, glycosylation of several adeno-associated viruses (AAVs) has been reported. However, the obtained glycosylation pattern and structure are varied and still controversial, suggesting that data acquisition and interpretation using a standardized analytical method are required [1]. In this study, we developed an in-depth analytical method for glycosylation of several serotypes of AAVs using multimodal glycan omics technologies [2]. In brief, 25 recombinant AAVs (rAAVs) among 6 different serotypes were initially subjected to an ultrasensitive lectin microarray we recently improved [3]. Significant signals on O- glycan binding lectins were observed only in rAAV6. The subsequential liquid chromatography-tandem mass spectrometry (LC-MS/MS), combined with the pre-enrichment step using an O-glycan binding lectin and with the post-MS step for glycopeptide identification using a software GRable [4], enabled us to find the mucin-type glycans on viral protein 2 of rAAV6. Interestingly, the glycosylated rAAV6 had a much lower VP1 to VP2/VP3 ratio than non-glycosylated rAAV6. The difference in component proportion leads to the reduced transduction efficiency of rAAV6 [5]. This study was supported by a grant-in-aid from the 'Research and development of core technologies for gene and cell therapy' supported by the Japan Agency for Medical Research and Development (AMED) [grant number JP18ae0201001 and JP18ae0201002]. References [1] Rumachik N.G., et al., Mol. Ther. Methods Clin. Dev., 18, 98-118 (2020). [2] Hiono T., et al., J. Proteome Res., 23, 1408-19 (2024). [3] Boottanun P., et al., Anal. Bioanal. Chem., 415, 6975-84 (2023). [4] Nagai-Okatani C., et al., bioRxiv (2024). doi:10.1101/2023.10.30.564073. [5] Yamaguchi Y., et al., Mol. Ther. Methods Clin. Dev., 32, 101256 (2024). 1061

1062

P-III-0902 Decoding the glycoproteome of Alzheimer's disease by network-based glycoproteomics Qi Zhang1, Cheng Ma2, Sheng Pan2, Lih-Shen Chin1, Lian Li1 1Emory University School of Medicine, Atlanta, GA, United States 2University of Texas Health Science Center at Housto, Houston, TX, United States that integrates Alzheimer's disease (AD) is the most common neurodegenerative disorder with no effective means of prevention or treatment, highlighting the need for novel biomarkers and therapeutic targets to improve AD diagnosis and treatment. Protein glycosylation plays critical roles in controlling brain function, but current knowledge of human brain glycoproteome and its alterations in AD is limited. To address the gap in knowledge, we established a network-based glycoproteomics approach intact glycopeptide-based quantitative glycoproteomics with network biology for large-scale, site-specific analysis of human brain glycoproteome and its changes in AD. We used this approach to analyze human brain tissue samples from neuropathologically confirmed AD cases and their age-matched controls and identified over 10,000 human brain N-glycoforms from nearly 1200 glycoproteins with 164 glycan compositions across more than 2500 in vivo N-glycosylation sites. Data-driven network analyses reveal previously unknown co-regulation relationships among different glycan modifications in human brain and highlights an involvement of glycan-specific elevations in oligomannosylation and paucimannosylation and reductions in sialylation and N-glycan branching in AD pathophysiology. Furthermore, network analyses show that human brain glycoproteome is organized into 21 modules of co- regulated protein glycoforms, 7 of which are associated with AD clinical phenotype, amyloid-b accumulation, and tau pathology. This study uncovers a number of dysregulated glycosylation-affected processes in AD, including altered cell adhesion, matrisome dysfunction, neuroinflammation, synaptic dysfunction, signaling dysregulation, endocytic trafficking alteration, and lysosomal dysfunction. Our findings provide mechanistic insights into AD pathogenesis and pave the way forward for developing glycosylation-based therapies and biomarkers to combat this devastating disease. 1062

1063

P-III-0904 High-throughput discovery of functional N-glycosylation sites for protein stability by thermal glycoproteome profiling Shulei Liu1, Benjamin Schulz1 1University of Queensland, School of Chemistry and Molecular Biosciences, Brisbane, Australia N-glycosylation, a crucial post-translational modification (PTM) referring to the attachment of glycans to asparagine residues, is important in facilitating protein folding, controlling intrinsic protein structure, and modulating protein interactions. Despite its importance for fundamental biology and the biopharmaceutical industry, our understanding of the impact of site-specific N-glycosylation on protein stability is still lacking. Sequential Window Acquisition of all Theoretical Mass Spectra (SWATH-MS) facilitates global analysis of proteoforms and site-specific measurement of PTMs, providing the possibility to map and quantify site- specific N-glycosylation on proteins at large scale and high sensitivity. Our study developed a high-throughput strategy to discover specific N-glycosylation sites that impact protein stability using thermal glycoproteome profiling of together with partial site-specific de-glycosylation, and allowed us to discover functional N- glycosylation sites. The established method provides an efficient strategy to reveal functional PTMs affecting protein stability at a large scale, with potential applications in different systems including exploring the impact of site-specific glycosylation on protein-protein or protein-ligand interactions, which will benefit quality enhancement of biologics and discovery of biomarkers. 1063

1064

P-III-0905 Glycoproteomic characterization of endogenously cleaved chondroitin sulfate-glycopeptides from aggrecan in the urine of children Jonas Nilsson1, Fredrik Noborn2, Maria Blomqvist2, Göran Larson2 1University of Gothenburg, Proteomics core facility, Gothenburg, Sweden 2University of Gothenburg, Department of Laboratory medicine, Gothenburg, Sweden Glycosaminoglycans (GAGs) of the chondroitin sulfate (CS) class are composed of anionic polysaccharide chains that are O-glycosidically linked, via a linkage region, to specific Ser residues of extracellular glycoproteins to form the CS proteoglycans (PGs). CSPGs are widely distributed in the extracellular matrix (ECM) and on cell surfaces. CSPGs are involved in many aspects of cell signalling and in organism development; and are major building blocks of cartilage and bone. A major human CSPG is aggrecan, composed of 2530 amino acid residues potentially carrying >100 CS chains, which is an important structural constituent of cartilage. Significantly, protease degradation of aggrecan leading to a loss of cartilage is a key feature in the development of osteoarthritis. To better understand the structure, function and biomedical importance of expressed and post- translationally modified ECM proteins, the matrisome, it is crucial to structurally characterize and quantitate them. In the emerging field of glycoproteomics, N- and O-glycopeptides are LC-MS/MS analysed, after protease digestion and enrichment procedures, to facilitate the site-specific glycan analysis of glycoproteins. In line with this, we have developed glycoproteomic methods to investigate the linkage region glycan structure of CSPGs and to identify novel CS-glycosites from human urine samples. Since children have a relatively high concentration of urine GAGs we have now started to analyse urine samples from 0.5-19 year old children. Thereby a multitude of endogenously cleaved CS-glycopeptides from aggrecan were identified that were not present in urine samples from adults. Two examples are S.GLPS.G and G.VEDIS.G. By using the Byonic software, 719 unique glycopeptides from 231 different peptides encompassing up to 151 out of the 184 possible SG and SA CS-glycosites of aggrecan were identified. Thus, thanks to the presence of the endogenously cleaved CS-glycopeptide fragments in the urine from children we were able to characterize the site-specific CS glycosylation of aggrecan that previously had been inaccessible for analysis. Several new LC-MS/MS approaches were used for the structural analyses. For instance, fragmentation analysis after negative mode higher-energy collision dissociation (HCD) was used to characterize the sulfate distribution pattern of the GAG chains of the dominating GLPS glycopeptides. The presented methodologies will be of importance in future CS glycoproteomics projects, and it will be important to investigate whether the aggrecan glycopeptides can be used as biomarkers in adults, for instance in osteoaerthritis or in reumatic disorders. 1064

1065

P-III-0906 Chasing O-glycopeptides: comparison of two sample preparation methods Ádám Pap1,2, Zsuzsanna Darula1,2, Katalin Medzihradszky F.2, Éva Hunyadi-Gulyás2, Aladár Pettkó-Szandtner2 1Hungarian Centre of Excellence for Molecular Medicine (HCEMM), Single Cell Omics Advanced Core Facility, Szeged, Hungary 2HUN-REN Biological Research Centre, Laboratory of Proteomics Research Core Facility, Szeged, Hungary Introduction: The successful analysis of post-translational modifications depends on multiple factors, such as complete digestion of the sample and efficient enrichment of the peptides of interest. This work focuses on the former factor by comparing two sample preparation methods: filter-aided sample preparation (FASP) and suspension trapping (S-Trap) for the characterization of O-glycosylation of human urinary proteins. Method: Urine samples were pooled and 3-3 aliquots (5mg protein per aliquot) were processed using the FASP and the S-Trap protocol, respectively. Samples were digested with trypsin and subsequently with PNGaseF. After the enzymatic removal of N-glycans, O-glycopeptides were enriched using a two-round lectin affinity chromatography using wheat germ agglutinin. The enriched samples were analysed using LC-MS/MS and applying a HCD product ion-dependent EThcD data acquisition method. Glycopeptides were identified using the Protein Prospector Batch Tag Web search engine, semiquantitative comparison of the identified glycopeptides was performed using spectral counting. Results: We observed no difference in the identified O-glycan pool between the results of the two sample preparation methods. The top 50% of the identified MS/MS spectra for both sample preparation methods belonged to O-glycopeptides carrying additive masses of 656 (HexNAcHexNeuAc), 947 (HexNAcHexNeuAc2) and 1312 (HexNAc2Hex2NeuAc2). The most interesting finding was that the S-Trap method covered a subset of the O- glycoproteins compared to the FASP protocol. This difference in the identified glycoproteins was observed for the HCD and the EThcD data as well. Acknowledgements: This work was supported by the following grants: EU Horizon 2020 Grant No.: 739593; KIM NKFIA 2022-2.1.1-NL-2022-00005; GINOP-2.3.2-15-2016-00001; GINOP-2.3.2-15-2016-00020. 1065

1066

P-III-0907 The synergy of DeepGP Platform and GlycanFinder promote the deep characterization of protein N/O- glycosylation Chao Peng1,2, Jinzhi Zhao1, Ping Wu1, Guangqin Cheng1, Wenting Li2, Lei Xin2, Baozhen Shan1,2 1Baizhen Biotechnologies, Inc., R&D, Shanghai, China 2Bioinformatics Solutions Inc., R&D, Waterloo, Canada The glycosylation of proteins is pivotal for their structural integrity and significantly impacts a variety of biological processes. The detailed characterization of protein glycosylation is essential for optimizing biological functions, understanding pharmacokinetics, managing immunogenicity, ensuring quality control, and complying with regulatory standards. However, the microheterogeneity of potential N/O-glycosylation sites and the clustering of multiple adjacent O-glycans present a considerable challenge in the analysis of glycosylation. In our study, we have developed a novel, highly efficient parallel enrichment strategy, named DeepGP, capable of simultaneously enriching both N and O-glycopeptides. We conducted a systematic comparison of five commonly used enrichment columns for their effectiveness in enriching intact N-glycopeptides from human IgG and mouse plasma. Our results demonstrated that DeepGP not only identified the highest number of glycopeptides but also achieved the highest efficiency. Moreover, we showcased DeepGP's ability to effectively enrich minimal sample quantities; for instance, it successfully enriched as little as 20ug of HeLa peptides, leading to the identification of 829 glycopeptides. Impressively, with 200ug of HeLa protein, DeepGP identified 4582 glycopeptides, with a consistent coefficient of variation (CV) of less than 7.8% across three replicates. We also employed DeepGP in conjunction with GlycanFinder to characterize N and O-glycosylation qualitatively and quantitatively in Etanercept. Our innovative approach offers several distinct advantages over existing methods: (1) A significant enhancement in the detection of O-glycosylation sites, with the identification of up to 11 previously unreported sites; (2) An increased diversity of glycans at various O-glycosylation sites; (3) Improved accuracy through the incorporation of additional corroborating evidence, such as b/y or c/z ions generated via Electron Transfer/Higher-energy Collision Dissociation (EtHCD) mode, which bolsters the confidence level of our findings. Fig. 1 (optional) 1066

1067

P-III-0908 N-linked glycopeptide identification using nanoflow LC and electron activated dissociation (EAD) MS/MS fragmentation Patrick Pribil1, Remco van Soest2 1SCIEX Canada, Technical Marketing, Concord, Canada 2SCIEX, Technical Marketing, Redwood City, CA, United States Protein glycosylation is a critical post-translational modification (PTM) that affects protein folding and stability. It is also essential for cell-cell adhesion and, as such, plays a role in immune response, cancer, and numerous diseases. Mass spectrometry (MS) instrument sensitivity is one of the significant limitations when analyzing glycopeptides due to the heterogeneity of glycan structures, resulting in multiple peptide isoforms with much lower abundances than their non-glycosylated forms. The challenges associated with low glycopeptide abundances can be overcome with strategies like enrichment of glycopeptides, a lower flow rate LC separation regime (i.e., nanoflow LC separation), and high-performance MS systems for glycopeptide detection and characterization. Another challenge for glycopeptide characterization comes from the labile nature of the glycosylation modification on peptides. Common fragmentation methods such as collision-induced dissociation (CID) can provide peptide backbone information but often result in fragments lacking the labile side chain modifications. Alternative fragmentation methods, such as electron activated dissociation (EAD)-based MS/MS, have been shown to provide site-specific PTM localization due to the retention of these modifications on the resulting fragment ions. In this work, nanoflow LC separation and EAD fragmentation were used to demonstrate the identification of N- glycopeptides in depleted, digested and glycopeptide-enriched human plasma. Using 60-minute nanoflow LC gradients, EAD-based data-dependent acquisition (DDA), and data processing with PEAKS GlycanFinder software, >1000 glycopeptides were identified from samples derived from 37.5 µg of plasma. 1067

1068

P-III-0909 Surface glycoproteome responses of primary mucosal immune cells in infection Tina Rietschel1, Qile Yang1, Paul Beltau1, Marco van Ham1, Mario Müller2, Hortense Slevogt3, Lothar Jänsch1 1Helmholtz-Centre for Infection research, Cellular Proteome Research, Braunschweig, Germany 2Universitätsklinikum Jena, Functional Proteomics, Jena, Germany 3Helmholtz-Centre for Infection research, Dynamics of Respiratory Infections, Braunschweig, Germany Surface receptor proteins are crucial for understanding and targeting immune responses. They are central to (i) the cluster of differentiation (CD) concept for classifying immune cells and phenotypes, (ii) coordinating immune signaling and effector functions, and (iii) serving as prime diagnostic and therapeutic targets in human diseases. Virtually all surface proteins are N- and/or O-glycosylated, with these glycosylated domains significantly influencing intra- and intermolecular interactions. Untargeted glycoproteomics using LC-MS is the gold standard for providing comprehensive and unbiased information on immune-regulatory surface proteins. Here, we present the first surfaceome study of primary human mucosal-associated invariant T cells (MAITs) and their glycosylation response under infection-like conditions. MAITs are the most abundant anti-infective T cell population in the liver, lung, and gut. These innate-like effector cells can directly recognize, via their T cell receptor, bacterial-derived metabolites presented by MR1 on antigen-presenting cells. Activated MAIT cells are multifunctional, providing regulatory, inflammatory, or cytotoxic responses enabling direct killing of infected cells as well as maintenance of organ homeostasis. Recently, we provided the first proteome study of MAITs indicating their unique effector phenotype (Bulitta et al., 2018), and discovered their antiviral properties by translatomics (Jakob et al., 2023). In order to elucidate the possibilities for selective regulation of their different immune responses, we established a robust glycoproteome pipeline to study surfaceomes of primary human immune cells. Bacterial-activated MAITs from six individuals were characterized by flow cytometry and total glycoproteomes were examined using an adapted hydrazide chemistry approach (Zhang et al. 2003). This allowed the enrichment of N- and O-glycosylated proteins which were characterized donor-dependently by LC- MS/MS (Dionex 3000 RSLC-Orbitrap Exploris 480 MS). In total, we identified 2300 glycoproteins containing over 200 surface proteins including known (e.g. CD25, CD26, CD45) and novel surface receptors, which we started to confirm by flow cytometry. Additionally, PNGase treatment of glycosylated proteins at hydrazide beads allowed the selective release and investigation of N-glycosylated asparagines. Interestingly, infection-like conditions caused the regulation of only 6 N-glycosylation sites suggesting a functional role in anti-infective MAIT cell responses. In conclusion, we established robust surfaceome analytics which (i) unravels the human MAIT cell surfaceome and receptors as novel targets for anti-infective strategies and (ii) is now used for characterizing additional immune cell types such as monocytes/macrophages. 1068

1069

P-III-0910 Peptide mapping of tryptic digests for mAbs using a novel ExD cell on the 6545XT Q-TOF mass spectrometer Stephen Sciuto1, Maozi Liu1, Rachel Franklin1, Jerry Han1 1Agilent Technologies, Santa Clara, CA, United States Introduction: Post-translation modifications (PTMs) on monoclonal antibodies (mAb) play an important role in the safety, efficacy, and binding of a therapeutic to its target. Common examples of PTMs that seek to be identified include glycosylation and phosphorylation. One challenge in the identification of these PTMs is that dissociation techniques such as collision-induced dissociation (CID) can break apart these fragile modifications. Here, we describe the use of an electron-based dissociation technique (ExD) that supplies low-energy electrons for the fragmentation of tryptic digests that contain glycosylation at asparagine consensus sites (NXS/T). The fragmentation of trastuzumab tryptic digest on a 6545XT AdvanceBio LC/Q-TOF using CID is compared to illustrate that CID alone will fragment these labile PTMs. Experimental: Tryptic digests for mAbs were generated using the Agilent AssayMAP Bravo system using either an in-solution digestion or single-pot (SP3) protocol. Following digestion, peptides were reconstituted in 0.1% formic acid in water at a concentration of ~0.5 µg/µL. Peptides were added to a G7167B multisampler at 6oC inside a 1290 Infinity II LC system. LC/MS data was collected on a 6545XT equipped with an ExD cell and searched using a pre-released version of MassHunter BioConfirm 12.1 software. For tuning, the Extended Dynamic Range (2 GHz) mode was used with the 100 – 3000 m/z range. Preliminary data: A) Sequence coverage of trastuzumab and NIST: LC/MS data was acquired on the 6545XT AdvanceBio LC/Q-TOF system operated in ExD or CID mode. Following peptide mapping search in MassHunter BioConfirm software, the sequence coverage was 99% for the light- and heavy-chains of NIST and trastuzumab. Similarly, the sequence coverage of mAbs in CID mode was 99%, however this fragmentation technique resulted in glycopeptide fragmentation along the peptide backbone. B) NIST and Trastuzumab glycosylation identification from MS/MS The peptide TKPREEQYNSTYRVVSVLTVLHQDWLNGKEYK with G1F (NIST) and G0F (Trastuzumab) was examined because this glycopeptide had a strong MS1 response that would provide amino-acid resolution along the peptide backbone around the glycosylation site at the N9 position. Sequence coverage of these glycopeptides were found to be 100%, providing unambiguous assignment of the glycosylation PTM. In particular, the 2+ monoisotopic fragment ion was identified at m/z 1930.4256 for Trastuzumab, resulting in a deisotoped z21 2+ monoisotopic mass of 3858.85 Da. Moving one amino acid to the right (toward the C-terminus), the z20 fragment ion was identified at m/z 1150.6100, resulting in a deisotoped mass of 2299.22 Da. The deisotoped mass shift between z21 and z20 (1559.63 Da) is consistent with the G0F+Asparagine residue with H2O loss. A mass shift of 1721.63 Da is observed for NIST with the G1F+Asparagine residue with H20 loss. 1069

1070

P-III-0911 Unveiling Glycosylation features in the oseltamivir-resistant NA protein of the H1N1pdm09 strain Joy Seitanidou1, Chris McMillan1, Andrea Corner1, Cassandra Pegg1, Benjamin Schulz1 1University of Queensland, School of Chemistry and Molecular Biosciences, Brisbane, Australia Influenza A virus remains a persistent health threat in our society, despite ongoing efforts to develop more effective vaccines. The virus's ability to infect hosts relies on the coordinated actions of its surface antigens, haemagglutinin (HA) and neuraminidase (NA), which facilitate viral binding and spreading respectively. Traditionally, HA has been the primary focus of vaccine development. However, following the emergence of the H1N1pdm09 strain, which exhibited resistance to the NA inhibitor Tamiflu® (oseltamivir, OC) due to a single amino acid mutation (H275Y), there is increasing support among scientists to include NA antigens in annual influenza vaccines to enhance their effectiveness and coverage. NA is a glycosylated protein crucial for budding off newly synthesised virions and facilitating the effective viral spreading. Therefore, understanding how glycans mediate the above process might be crucial for future vaccine design. NA is a tetrameric protein with a cytoplasmic tail, a transmembrane domain, and a stalk that anchors the catalytic head in the viral membrane. Given that the enzymatic activity resides in the head, most studies on NA's glycosylation have concentrated on this region. However, there is growing evidence of the potential antigenic role of the stalk region, although information on the glycan structures at the corresponding sites remains limited. In this study, we aimed to characterize the glycan variation and occupancy in both the stalk and head regions of the wild-type and oseltamivir-resistant recombinant NA (rNA) from the H1N1pdm09 strain. We conducted site- directed mutagenesis to introduce mutations associated with OC resistance (H275Y) and mutations restoring viral fitness (V241I and N369K) and expressed rNA in mammalian cells (HEK293). Purified proteins were subjected to proteolytic digestion, and the resulting glycopeptides were analysed using the ZenoTOF and cyclic SELECT mass spectrometers. Identification of glycan formulas was performed in Byonic, and all glycopeptide spectra were manually curated for the presence of structure-specific fragments. Analysis of ion mobility data was performed in DriftScope with the retention times associated with glycopeptides manually extracted. The above analysis resulted in the identification of all four glycosites in the head domain (N88, N146, N235 and N386) and all three conserved sites in the stalk (N50, N58, N63 and N68). The majority of the glycans identified were complex, with fucose being either at the core or antenna of the glycan structure. Neither sialylated- nor sulphated-glycopeptides were identified. To the best of our knowledge, this study represents one of the initial attempts to provide a comprehensive glycosylation profile of a pandemic-emerged NA, and therefore could offer valuable insights for future vaccine designs. 1070

1071

P-III-0912 Glycomics and glycoproteomics analysis of lung adenocarcinoma tissue samples Simon Sugár1, Mirjam Balbisi1,2, Fanni Bugyi1, Domonkos Pal1,2, Balázs Molnar1, Gabor Kecskemeti3, Zoltán Szabó3, Ibolya Laczo4, Tünde Harkó5, Judit Moldvay6, Lilla Turiak1 1HUN-REN Research Centre for Natural Sciences, Glycan Biomarker Research Group, Budapest, Hungary 2Semmelweis University, Doctoral School, Budapest, Hungary 3University of Szeged, Department of Medical Chemistry, Szeged, Hungary 4Békés County Central Hospital, Gyula, Hungary 5National Korányi Institute of Pulmonology, Budapest, Hungary 6University of Szeged, Department of Pulmonology, Deszk, Hungary Lung adenocarcinoma (LUAC) accounts for approximately 40% of all lung cancer cases. Several driver mutation(s) contribute to the genetic heterogeneity of LUAC. LUAC that is wild type for the EGFR, KRAS, and ALK genes is commonly referred to as "triple wild-type" (triple wt). Triple wt lung cancer has a poor prognosis, so the molecular characterization of this cancer type is especially important. Increasing evidence suggests that aberrant protein glycosylation plays a crucial role in the pathogenesis and progression of lung cancer. Alterations in glycosylation patterns have been previously observed between healthy and cancerous samples as well as between different lung cancer subtypes. Therefore, our ultimate goal is to compare the glycosaminoglycan and N-glycoproteomic profiles of triple wt LUAC samples and those harbouring different mutations (EGFR, KRAS and ALK), approximately 100 samples altogether. In the present pilot study our aim was the detailed proteomic and N-glycoproteomic characterization of the tumor and tumor adjacent regions of formalin-fixed paraffin embedded human lung tissue samples from a cohort of triple wt lung cancer patients (n=12). Following dewaxing and antigen retrieval of the tissue sections, on-surface tryptic digestion was performed on regions of interest, then glycopeptide enrichment was carried out using acetone precipitation. The pellet and supernatant fractions were analyzed separately, using different nanoHPLC-MS methods. Altogether 3376 proteins were quantified using MaxQuant, and 1626 unique glycopeptides were identified corresponding to 865 glycosylation sites on 365 glycoproteins using GlycReSoft. The protein expression profiles of the triple wt tumor and adjacent normal regions were found to be markedly different. Significant differences were identified in the expression levels of 1066 proteins, and several biological processes were found to be enriched (for both up- and downregulated genes) based on Fast Gene Set Enrichment Analysis (FGSEA). We have identified 28 glycopeptides with differences in abundance between tumor and adjacent normal tissue. Several of the corresponding glycoproteins showed significant differences in protein expression levels suggesting that changes detected in glycopeptide abundance can be attributed to a combination of changes in both glycoprotein expression, and glycosylation. To the best of our knowledge this the first study to analyze the N-glycosylation of glycoproteins of triple wt lung cancer and could be a valuable resource for further research regarding the potential of diagnostic glycoprotein markers, and drug targets as well. Support of the National Research, Development and Innovation Office (OTKA FK 131603, and OTKA K-147226) and the Lendület (Momentum) Program of the Hungarian Academy of Sciences is acknowledged. 1071

1072

P-III-0913 Simultaneous serum immunoglobulin A and G glycopeptide profiling sheds light into IgA nephropathy mechanism Anastasia Tzasta1, Tomas Post2, Anna M. Wasynczuk3, Steinar Gijze3, David Falck3, Nils Rother2, Ilse Rood2, Elmar Pietersen2, Alain van Gool4, Jonathan Barratt5, Raphaël Duivenvoorden2, Manfred Wuhrer3, Joannes Jacobs1 1Radboud University Medical Center, Laboratory Medical Immunology, Nijmegen, Netherlands 2Radboud University Medical Center, Nephrology, Nijmegen, Netherlands 3Leiden University Medical Center, Center for Proteomics and Metabolomics, Leiden, Netherlands 4Radboud University Medical Center, Translational Metabolic Laboratory, Nijmegen, Netherlands 5University of Leicester, Cardiovascular Sciences, Leicester, United Kingdom Introduction: IgA nephropathy (IgAN) is the most prevalent primary glomerular disease worldwide. The disease course ranges from mild to progressive, with a high lifetime risk for the development of kidney failure for the majority of patients. Although the disease mechanisms are not yet fully understood, a four-hit hypothesis for IgAN pathogenesis has been proposed, emphasizing the critical role of galactose-deficient IgA1 (gd-IgA1). According to this hypothesis, an overproduction of gd-IgA1 leads to elevated gd-IgA1 serum levels (hit 1) which are recognized by anti-gd-IgA1 autoantibodies (hit 2), resulting in the formation of immunocomplexes (hit 3). These complexes deposit in the mesangium (hit 4) causing glomerular inflammation and kidney injury. Current literature on IgAN pathology is based on gd-IgA detected by lectin-based immuno-assays. In this study, we aim to further elucidate the role of serum immunoglobulin glycosylation in IgAN pathology using a novel mass spectrometric technology that allows detailed site-specific glycoprofiling. Methods: We employed a nanoliquid chromatography (LC)- mass spectrometry (MS)-based workflow for simultaneous analysis of IgG and IgA glycopeptides generated from serum of 41 patients with IgAN in different disease stages and 41 age- and kidney function-matched controls. Using ELISA-based methods, we measured gd- IgA, IgG autoantibodies against gd-IgA, and IgA/IgG immunocomplexes. Results: We reliably identified 92 N- and O- glycopeptides from different peptide backbones (IgG1, IgG2/3, IgA1, IgA2). Calculation of structural features including galactosylation, sialylation, and N-glycan complexity, confirmed previous observations that high gd-IgA ELISA results were not caused by low expression of O-linked GalNAcs or due to shielding of galactose epitopes by sialic acids. Gd-IgA levels measured with ELISA correlated well with the amount of galactoses as measured with LC-MS (p=1.3e-09). O-glycosylation in the hinge region differed significantly between IgAN patients and controls and was similar across different disease stages. Sulfation of N- glycans was increased in non-IgAN patients and correlated with kidney function as assessed with the estimated glomerular filtration rate (eGFR). Furthermore, IgG N-glycosylation correlated with eGFR regardless of having IgAN. Finally, IgG autoantibodies against gd-IgA and IgA1/IgG immunocomplex levels were similar between controls and IgAN patients. Conclusions: Our comprehensive workflow combining glycome profiling of both IgG and IgA provides a detailed signature of IgAN and points towards candidate glycan motifs recognized in current immunoassays detecting gd- IgA and gd-IgA mediated immune-complexes. This provides insight into the nature and role of skewed immunoglobulin glycosylation in IgAN pathogenesis. Our results support the role of O-glycosylation in the formation of immune complexes, crucial in driving the inflammatory response in IgAN. 1072

1073

P-III-0914 The distinctive structural isomers of IgG Fc-glycans are subclass-specific Weiwei Wang1,2, Joshua Maliepaard1,2, Gestur Vidarsson1,2, Albert J. R. Heck1,2, Karli Reiding1,2 1Utrecht University, Utrecht, Netherlands 2Netherlands Proteomics Center, Utrecht, Netherlands Immunoglobulins G (IgGs), are dominant immunoglobulins in the humoral immune response and exhibit dual functionalities: the Fab-domains target antigens, whereas the Fc-domains modulate immune activities through interactions with receptors. In human serum, IgGs are classified into four subclasses, IgG1, IgG2, IgG3 and IgG4, each characterized by unique features in their sequences, disulphide bridges and glycosylation signatures. Fc N- glycosylation plays a key role in modulating the functions of IgGs. Taking IgG1 as an example, changes in the glycans may alter action from inhibitory/anti-inflammatory to activating the complement system and promoting antibody-dependent cellular cytotoxicity. The functionality of protein glycosylation depends heavily on the precise nature of the linkages and branching between monosaccharides, as well as their specific positions within proteins. Glycosylation research typically involves the analysis of released glycans or glycopeptides. Structural information of glycans is mostly accessible at the released glycan level, yet, the lack of a peptide prevents knowing the proteins of origin (e.g., different IgG subclasses) or the specific glycosylation sites (e.g., the Fc or Fab regions). Glycopeptide analysis, on the other hand, does allow site-specificity, but it greatly underestimates the complexity of the involved glycan structures as it is primarily performed at the compositional level. Method development into studying glycopeptide structure is urgently needed, the chromatographic dimension of LC-MS/MS being a promising direction. Here, we developed a high-resolution nano-HILIC-LC-MS/MS method capable of quantifying IgG glycan structures at the glycopeptide level. Using this methodology, we demonstrated that IgG subclasses carry differential yet defined structural glycan characteristics, in both recombinant and endogenous IgGs. For instance, our analyses revealed distinct structural glycosylation signatures for each IgG subclass, e.g., IgG1 and IgG3 showing predominant galactosylation of the 6-branched antenna, IgG2 of the 3-branched antenna, and IgG4 a balance of the two. These and other subclass-specific glycostructural elements proved observable in recombinant IgG and in human plasma, in which inter-individual variability and temporal stability could be demonstrated. The here-described onset of structural glycoproteomics is expected to fundamentally alter the way in which we study IgG, opening up a new layer of functional investigation and biomarker development. Fig. 1 (optional) 1073

1074

P-III-0915 Genotype-dependent N-glycosylation and novel O-glycosylation affect the activation of histidine-rich glycoprotein by plasmin Yang Zou1, Matti Pronker1, Karli Reiding1, Albert J. R. Heck1 1Utrecht University, Utrecht, Netherlands Introduction: Histidine-rich glycoprotein (HRG) is an abundant plasma glycoprotein that regulates various biological processes, including coagulation and immune response. HRG has been identified as a biomarker for multiple diseases, such as COVID-19, AIDS, asthma, renal disease, pulmonary disease, and thrombotic disorders. In humans, HRG is highly polymorphic, with five variants having a maximum allele frequency (MAF) over 10%, affecting its structure and function. The Pro204Ser mutation, notably, introduces a putative N-glycosylation site at Asn202, which is associated with reduced pregnancy rates, aging, and mortality. Other N-glycosylation sites in HRG also impact functions such as cell proliferation and cancer regulation. Despite HRG's significant role in homeostasis, the precise effects of its mutations and glycosylation remain unclear. Methods: We optimized an immobilized metal affinity chromatography (IMAC) purification protocol, leveraging HRG's histidine-rich region to efficiently enrich the protein from serum. We then used LC-MS/MS-based (glyco)proteomics to characterize HRG mutants and glycosylation in the sera of COVID-19 infected and healthy individuals, as well as in recombinant variants with mutations at positions 204 and 493. To further explore HRG mutations, we extracted data from the 1000 Genomes Project, assessing the frequency of these mutants. Additionally, we examined plasmin-induced cleavage of HRG to understand the functional and structural interplay between HRG function, mutations, and glycosylation. Preliminary Data: Our (glyco)proteomics approach identified the genetic make-up of HRG in the sera of all donors, revealing their homozygotic or heterozygotic allotypes and the glycosylation profiles of each purified HRG sample. We found strong correlations between specific HRG allotype mutations, some of which were either mutually exclusive or supportive. These findings were partially corroborated by data from the 1000 Genomes Project. We also measured the glycosylation profiles of endogenous plasma HRG and recombinantly produced HRG (in CHO and HEK293 cells), showing distinctive patterns. The Pro204Ser-specific Asn202 site, with a ~50% MAF frequency, had high occupancy by complex-type N-glycans. Additionally, we identified novel O-glycosylation sites at HRG Thr273 and Thr274, with over 90% occupancy across all HRG variants. As these O-glycosylation sites are near a proposed plasmin-cleavage site (Lys275), we assessed plasmin-induced cleavage kinetics and specificity. The assay showed that plasmin-induced cleavage was significantly influenced by sialylation, particularly of the O- linked glycans. Moreover, the Pro204Ser mutation notably impacted plasmin-induced cleavage. Our findings reveal the intricate interplay of glycoproteogenomic features within HRG, suggesting how these factors may contribute to HRG's regulatory role in homeostasis in vivo. 1074

1075

P-III-0916 In-depth N-glycoproteomic analysis reveals strategies to elucidate unanticipated N-glycans in the low-abundant human blood plasma N-glycoproteome Frania J. Zuniga-Banuelos1,2, Marcus Hoffmann1, Udo Reichl1,3, Erdmann Rapp1,2 1Max Planck Institute for Dynamics of Complex Technical Systems, Bioprocess Engineering, Magdeburg, Germany 2glyXera GmbH, Magdeburg, Germany 3Otto-von-Guericke University Magdeburg, Chair of Bioprocess Engineering, Magdeburg, Germany To understand the implications of protein glycosylation for clinical diagnostics and biopharmaceuticals, innovative glycoproteomic technologies are required. Recently, significant advances were made, particularly towards structure-focused N-glycoproteomic analyses. The mass spectrometric analysis of intact N-glycopeptides using stepped collision fragmentation and glycan oxonium ion profiling can assist the discrimination between different N-glycan structures to a large extent. Still, there are weaknesses that current N-glycoproteomic approaches must overcome. These concern the recognition and handling of incorrect identifications, the identification of rare and modified N-glycans, and the insufficient glycoproteomic coverage, especially for complex samples. To address these shortcomings, we have developed an innovative N-glycoproteomic workflow that provides comprehensive site-specific and structural N-glycoproteomic data on human blood plasma glycoproteins. The sample preparation workflow features protein depletion, various fractionation strategies, and high-resolution mass spectrometry with stepped collision energy fragmentation. Furthermore, we have established a decision- tree strategy for data validation that significantly improves the description of the N-glycan micro-heterogeneity. Our data analysis workflow shows the reliable differentiation of ambiguous N-glycan structures like antenna- versus core-fucosylation plus modified and rare N-glycans such as phosphorylated, sulfated, and glucuronidated ones. With this workflow, we were able to advance in the analysis of human blood plasma glycoproteins to concentrations as low as 10 pg/mL. A total of 1,929 N-glycopeptides and 942 N-glycosites derived from 805 middle- to low-abundant human glycoproteins were identified. These include prothrombin (110 μg/mL), extracellular matrix protein 1 (0.78 μg/mL), integrin subunit alpha M (1 ng/mL), and cysteine-rich secretory protein LCCL domain-containing 1 (8.2 pg/mL) – protein concentrations reported for human blood plasma. Overall, the presented in-depth N-glycoproteomic workflow offers strategies to conduct a refined analysis on complex N-glycoproteomes, and expands the list of N-glycans to be assessed as possible blood plasma biomarkers. Applying the strategies shown in this work can significantly enhance our understanding of glycoproteins of clinical interest and contribute to general health care. 1075

1076

Cell biology insights (P-III-0917 to P-III-0997) P-III-0917 In-depth phosphoproteomic profiling of the insulin signaling response in heart tissue and cardiomyocytes unveils canonical and specialized regulation Jonathan Achter1, Estefania Torres Vega1, Andrea Sorrentino1, Konstantin Kahnert1, Katrine Douglas Galsgaard1, Pablo Hernandez-Veras2, Michael Wierer3, Jens Juul Holst1,4, Jørgen Wojtaszewski5, Robert William Mills1, Rasmus Kjøbsted5, Alicia Lundby1 1University of Copenhagen, Department of Biomedical Sciences, Copenhagen, Denmark 2University of Copenhagen, Core Facility for Integrated Microscopy, Copenhagen, Denmark 3University of Copenhagen, Proteomics Research Infrastructure, Copenhagen, Denmark 4University of Copenhagen, NNF Center for Basic Metabolic Research, Copenhagen, Denmark 5University of Copenhagen, Department of Nutrition, Exercise and Sports, Copenhagen, Denmark Background: Insulin signaling regulates cardiac substrate utilization and is implicated in physiological adaptations of the heart. Alterations in the signaling response within the heart are believed to contribute to pathological conditions such as type-2 diabetes and heart failure. While extensively investigated in several metabolic organs using phosphoproteomic strategies, the signaling response elicited in cardiac tissue in general, and specifically in the specialized cardiomyocyte cells, has not yet been investigated to the same extent. Methods: Insulin or vehicle was administered to male C57BL6/JRj mice, and ventricular tissue was analyzed by quantitative phosphoproteomics to evaluate the insulin signaling response. Cardiomyocytes from a cardiac and skeletal muscle-specific Tbc1d4 knockout mouse and wildtype littermates were used to study Tbc1d4"s role in insulin signal transduction. Phosphoproteomic studies involved isobaric peptide labeling with Tandem Mass Tags (TMT), enrichment liquid chromatography, and high-resolution mass spectrometry measurement. for phosphorylated peptides, fractionation via micro-flow reverse-phase sites on events, including phosphorylation Results: We quantified 10,399 phosphorylated peptides from ventricular tissue and 12,739 from isolated cardiomyocytes, localizing to 3,232 and 3,128 unique proteins, respectively. In cardiac tissue, we identified 84 insulin-regulated Receptor (Insr;Y1351, Y1175, Y1179, Y1180) itself as well as the Insulin receptor substrate protein 1 (Irs1;S522, S526). Predicted kinases with increased activity in response to insulin stimulation included Rps6kb1, Akt1 and Mtor. Tbc1d4 emerged as a major phosphorylation target in cardiomyocytes. Despite limited impact on the global phosphorylation landscape, Tbc1d4 deficiency in cardiomyocytes attenuated insulin-induced Glut4 translocation and induced protein remodeling. We observed 15 proteins significantly regulated upon knockout of Tbc1d4. While Glut4 exhibited decreased protein abundances consequent to Tbc1d4-deficiency, Txnip levels were notably increased. Stimulation of wildtype cardiomyocytes with insulin led to the upregulation of 262 significant phosphorylation events, predicted to be regulated by kinases such as Akt1, Mtor, Akt2, and Insr. In cardiomyocytes, the canonical insulin signaling response is elicited in addition to regulation on specialized cardiomyocyte proteins, such as Kcnj11;Y12 and Dsp;S2597. Details of all phosphorylation sites are provided. the Insulin 1076

1077

P-III-0918 Kinome-profiling using patient-derived cell lines to discover novel therapeutic targets for treating dermatofibrosarcoma protuberans with resistance to imatinib Taro Akiyama1,2, Takuya Ono1, Shuhei Iwata1,2, Yuki Adachi1, Julia Osaki1, Yuki Yoshimatsu3, Rei Noguchi1, Seiji Ohtori2, Tadashi Kondo1 1National Cancer Center Research, Rare cancer Research, Tokyo, Japan 2Chiba university, Orthopedic surgery, Chiba, Japan 3Tochigi Cancer Center Research, Patient derived cancer model, Tochigi, Japan Introduction: Dermatofibrosarcoma protuberans (DFSP) is a rare sarcoma, characterized by a unique fusion gene, COL1A1-PDGFB. Imatinib, a kinase inhibitor which blocks PDGF receptors, is included in the standard treatment of DFSP, and 50% of patients with DFSP demonstrate resistance. The purpose of this study is to reveal the response to the imatinib treatments at the molecular level, and identify the kinases, which could be predictive biomarkers or therapeutic targets in the resistant patients. We hypothesized that the alternative kinases pathways are activated in the patients exhibiting the resistance. Examining the kinase activity in the DFSP cell lines with the different sensitivity to the treatments with imatinib, we investigated the kinases, which contribute to the resistance. These kinases and the activated pathways will be candidates of predictive biomarkers and therapeutic targets in DFSP. Materials and Method: Five cell lines established from tumor tissues of five patients with DFSP were used in the study. The tumor tissues were obtained from the surgically resected tumor prior to imatinib treatment. The cell lines were exposed to different concentrations of imatinib, and the half maximal inhibitory concentration (IC50; μM) was calculated. The cell lines were classified into two groups; Group S as sensitive to imatinib treatments with IC50 < 10 μM, and Group R as resistant to imatinib treatments with IC50 > 10 μM. Activity of 98 kinases was examined by in vitro kinase assay (PamStation, PamGene). All cell lines were treated with imatinib with concentration of each IC50 or dimethyl sulfoxide as control. The activity of kinases was subjected to the comparative study, and the effects of imatinib treatments were evaluated by student-t analysis with the statistical significance of 0.05. The kinases exhibiting the fold activity change greater than 2 with statistical significance were considered as those affected by the exposure of imatinib. We determined the kinases whose activity was differently regulated by the imatinib exposure between the cell lines in Group R and Group S. Result: Two and three DFSP cell lines were classified into Group S (IC50: 2.4-4.6 μM) and Group R (IC50: 13 – 42 μM), respectively. All cell lines underwent the kinome-profiling after exposure to imatinib or dimethyl sulfoxide. We observed the activity from 142 phosphorylated substrate peptides, and compared them between both groups. By the statistical comparison, we found that the kinases that are the targets of Ponatinib, which was approved for the treatments of the patients with chronic myelogenous leukemia, were activated in Group R, but not in Group S. Discussion: The kinases whose activity contributes to the resistance to imatinib treatments can be potent candidates for novel therapeutic targets. Ponatinib and its target kinases are worth investigating for treating of the patients resistant to imatinib. 1077

1078

P-III-0919 Neprilysin protease substrates: degradomics reveals novel roles in mitochondria and in neurodegeneration Tiziana Alberio1,2, Giulia Gaiferri1, Vittoria Monaco3,4, Maria Monti3,4, Mauro Fasano1,2, Marta Lualdi1,2 1Università degli Studi dell'Insubria, Department of Science and High Technology, Busto Arsizio, Italy 2Center for Neuroscience Research, Busto Arsizio, Italy 3University Federico II of Naples, Department of Chemical Sciences, Naples, Italy 4CEINGE Advanced Biotechnologies “Franco Salvatore”, Naples, Italy The role of proteases in neurodegeneration is central and extensively investigated. Recently, by using the mitochondrial dimethylation-TAILS degradomics approach in a cellular model of dopamine (DA) imbalance in neuroblastoma SH-SY5Y cells, we demonstrated that Neprilysin is a candidate protease involved in mitochondrial dysfunction related to Parkinson"s disease (PD)1. Neprilysin is a Zinc-dependent endopeptidase, which exists in two catalytically active forms, i.e., transmembrane, and soluble. It is a recognized biotarget, being responsible for the cleavage of biologically relevant substrates such as angiotensin, bradykinin, substance P, glucagon, insulin β-chain, oxytocin, enkephalins, and amyloid-beta. To investigate the role of neprilysin as a candidate mitochondria-localized protease, we first verified its presence and analysed its localization inside mitochondria, by sub-fractionation experiments and fluorescence staining. Moreover, we observed a markedly increased co-localization between neprilysin and mitochondria upon the induction of DA imbalance in SH-SY5Y cells. Then, to obtain the complete set of neprilysin substrates, we applied the dimethylation-TAILS approach on the total proteome of SH-SY5Y cells. A list of 155 candidate substrates emerged, and twenty of them resulted to be mitochondria-associated proteins, including some ribosomal proteins, heat shock proteins, membrane transporters, and metabolic enzymes. The functional enrichment analysis highlighted "protein translation", "ribosomal function", "processing of RNAs and proteins", and "neurodegeneration" as significantly enriched pathways in the whole substrates list. In conclusion, we demonstrated for the first time that neprilysin exists as a mitochondrial protease and plays a role in mitochondrial dysfunction linked to DA imbalance in PD. These processes are central to understand early events in PD development. 1 Lualdi, M.; Ronci, M.; Zilocchi, M.; Corno, F.; Turilli, E. S.; Sponchiado, M.; Aceto, A.; Alberio, T.; Fasano, M. Exploring the Mitochondrial Degradome by the TAILS Proteomics Approach in a Cellular Model of Parkinson"s Disease. Front. Aging Neurosci. 2019, 11, 195. https://doi.org/10.3389/fnagi.2019.00195. 1078

1079

P-III-0921 SureQuant-based absolute quantification of the Lysosomal mouse proteome Dhriti Arora1, Peter Mosen1, Biswajit Moharana1, Dominic Winter1 1Institute of Biochemistry and Molecular Biology, University of Bonn, Bonn, Germany The lysosome, which acts in several ways as the cell's metabolic center, is crucial for macromolecule recycling, nutrient sensing, antigen processing, cell death, and metabolic signaling. Dysfunction of lysosomes can lead to severe conditions, including lysosomal storage disorders, neurodegenerative diseases, cancer, and other metabolic syndromes. For studying lysosomal pathology in these diseases and cellular processes, accurate and reproducible quantification of lysosomal proteins is essential. Due to the low abundance of lysosomes, it is challenging to detect and quantify these proteins using whole proteome shotgun proteomics approaches. One way to overcome this is lysosome enrichment, which is typically limited to certain cell lines, hindering studies in other sample types like body fluids or tissues. Furthermore, so-far only relative quantification data are available for lysosomal proteins, and absolute amounts and stoichiometric ratios remain largely unknown. To enable absolute quantification of mouse lysosomal proteins, we developed a targeted strategy using medium/heavy isotope-labeled internal standards which we generated following the Quantification Concatemer (QconCAT) approach. The QconCATs were designed to represent 143 lysosomal proteins, for which three tryptic peptides each were included, organized into 12 QconCAT proteins. Following construct optimization, we synthesized lysoQconCAT genes that incorporated an isobaric standard peptide for normalization purposes and a His-Tag to facilitate affinity purification. We validated the generated QconCAT standards using shotgun proteomics, assessing the quality, sequence coverage, and labeling efficiencies of the isotope-labeled QconCAT peptides. We successfully expressed and purified these QconCAT standard proteins. Using these standards we established a SureQuant assay on an Orbitrap Fusion Lumos. As SureQuant typically uses high levels of heavy-isotope labeled peptides as trigger for PRM-based label free quantification, absolute quantification is not possible as the difference in intensity between these standards and sample peptides does not match the dynamic range of the mass spectrometer. To address this isotope-labeled standards for quantification at close to endogenous levels. This 3plex approach, utilizing the heavy-isotope labeled standards as trigger and medium standards for abundance estimation, allows automated absolute quantification with SureQuant. issue, we further added medium With this strategy, we determined absolute abundances of lysosomal proteins in mouse cell line. We optimized the spike-in amount of both heavy and medium isotope-labeled standards to maximize the number of covered peptides for absolute quantification, identifying and quantifying 136 verified lysosomal proteins. This method offers a reliable and sensitive approach to study lysosomal proteins, with possible application in the analysis of cell lines and mouse tissues, to advance our understanding of lysosomal function and disease implications. 1079

1080

P-III-0922 Developing a model to assess the impact of dysregulated protein methyltransferase domains using EHMT2 in colorectal cancer Byron Baron1 1University of Malta, Centre for Molecular Medicine and Biobanking, Msida, Malta Protein methylation,mainly on lysine and arginine, is the smallest post-translational modification but can modulate protein properties such as structure, function and stability. Dysregulatedprotein methylation can initiate abnormal biochemistry, leading to disease. Methyltransferases have been found to be dysregulated in various cancers including colorectal cancer (CRC). However, very little is known regarding the impact this has on specific cancers because the methylated targets are not fully known, particularly when the methyltransferase domain is mutated. Since EHMT2 is one of the 5 most dysregulated lysine methyltransferases in CRC (as well as various other cancers) it is an important research target. It catalyses the addition of methylation via the Su(var)3-9, Enhancer- of-zeste and Trithorax (SET) domain, composed of 130- to 140-amino acids, flanked by two cysteine-rich regions called the pre-SET and post-SET domains. These 3 essential domains only cover residues 972-1180 of the massive EHMT2 protein. The aim of this project was to use a yeast two-hybrid set up to isolate interactors of EHMT2 using only the 3 catalytic domains and then validate which of the isolated interactors were dysregulated in CRC. A clone spanning the pre-SET, SET and post-SET domains of EHMT2 was generated and used to create a bait fusion protein for the isolation of interactors from a protein coding library through mating in a yeast-two hybrid system. Following the determination of the interaction strength of the isolated interactions using a CPRG assay, a number of successful strong interactors were sequenced. The differential expression of the isolated EHMT2 interactors is now being investigated in CRC spheroids, since 3D cell culture better mimics in-vivo protein expression patterns. Once this model is fully validated, specific misense mutation especially those listed as "likely pathogenic" within the catalytic region of EHMT2 can be mutated in the bait construct to understand their impact on target interactions. This workflow is also being extended to methyltransferases of other families such as methyltransferase-like (METTL) enzymes. The data generated will be a useful guide for future diagnostic and prognostic testing of different CRC sub-types, particularly prior to chemotherapy administration or to determine the metastatic potential of tumours, especially in cases where the methyltransferase is dysregulated due to mutation of the catalytic domain. 1080

1081

P-III-0923 Matrix stiffness regulates the protein profile of extracellular vesicles of pancreatic cancer cell lines Benedetta Ferrara1,2, Sandrine Bourgoin-Voillard3,4,5, Damien Habert1, Benoit Vallée1, Alba Nicolas-Boluda6, Isidora Simanic7, Michel Seve3,4,5, Benoit Vingert8,9, Florence Gazeau6, José Cohen1,10, José Courty1,10, Ilaria Cascone1,10 1Université Paris-Est, Immunorégulation et Biothérapie, INSERM U955, Hopital Henri Mondor, Créteil, France 2Diabetes Research Institute, IRCCS San Raffaele Scientific Institute, Milan, Italy 3Univ. Grenoble-Alpes, CNRS UMR 5525, Grenoble INP, TIMC, EPSP, 38000 Grenoble, France, La Tronche, France 4Univ. Grenoble-Alpes, CNRS, Grenoble INP, CHU Grenoble Alpes, TIMC, EPSP, 38000 Grenoble, France, Grenoble, France 5Univ. Grenoble Alpes, LBFA et BEeSy, Inserm, U1055, CHU Grenoble Alpes, PROMETHEE Proteomic Platform, 38000 Grenoble, France, Grenoble, France 6Matière et Systèmes Complexes MSC, CNRS, Université Paris Cité, 45 rue des Saints Pères, Paris, France 7Modèles de cellules souches malignes et therapeutiques, INSERM UMR-S 935, Université Paris-Saclay, Villejuif, France 8Etablissement Français du Sang, Créteil, France 9Inserm, U955, Equipe 2, Créteil, France 10AP-HP, Groupe hospitalo-universitaire Chenevier Mondor, Centre d’investigation clinique Biotherapie,, Créteil, France The fibrotic stroma characterizing pancreatic ductal adenocarcinoma (PDAC) derives from a progressive tissue rigidification, which induces epithelial mesenchymal transition and metastatic dissemination. The aim of this study was to investigate the influence of matrix stiffness on PDAC progression by analyzing the proteome of PDAC-derived extracellular vesicles (EVs). PDAC cell lines (mPDAC and KPC) were grown on synthetic supports with a stiffness of 3 and 12 kPa and the protein expression levels in cell-derived EVs were analyzed by a quantitative MSE label-free mass spectrometry approach. Our analysis figured out 15 differentially expressed proteins (DEPs) in mPDAC-EVs and 20 DEPs in KPC-EVs in response to matrix rigidification. Up-regulated proteins participate to the processes of metabolism, matrix remodeling, and immune response, altogether hallmarks of PDAC progression. A multimodal network analysis revealed that the majority of DEPs are strongly related to pancreatic cancer. Interestingly, among DEPs, 11 related genes for mPDAC-EVs and 9 genes for KPC-EVs were significantly overexpressed in tumor-derived tissues according to gene expression profiling interaction analysis (GEPIA). In particular, investigating the potential clinical relevance of these data, the cluster of 3 genes displayed an adverse effect (p < 0.05) on the disease-free survival of PDAC patients. 1081

1082

P-III-0924 Diving into the hidden world of AltProts: unraveling functions and interactions Tristan Cardon1, Diego Fernando Garcia del Rio1,2, Isabelle Fournier1, Michel Salzet1 1INSERM U1192 Protéomique, Réponse Inflammatoire et Spectrométrie de Masse (PRISM), Villeneuve d'Ascq, France 2Martin-Luther-University Halle-Wittenberg, Pharmaceutical Chemistry and Bioanalytics, Halle (Saale), Germany For several years, we have focused on identifying an extended proteome beyond the traditionally referenced proteins (RefProt) contained in databases. While most proteomic studies concentrate on these known proteins, there is another layer to the proteome called the "non-referenced or Ghost proteome," which includes "Alternative proteins" (AltProt). These proteins originate from regions typically annotated as non-coding in transcripts, such as the 5' and 3' untranslated regions (UTRs), alternative reading frames (+1 or +2), and non- coding RNAs (ncRNAs, lncRNAs, circRNAs). Because AltProts are not included in standard databases, they are difficult to detect, and their functions are not well understood. Our research aims to uncover these proteins and elucidate their roles in cellular processes. This raises the critical question: "How can we determine the function of AltProts given their lack of molecular tools?" To address this, we are developing innovative strategies to identify and characterize the functions of AltProts. By leveraging advanced techniques such as crosslinking mass spectrometry (XL-MS), we aim to reveal protein-protein interactions involving AltProts, thereby predicting their roles in various cellular pathways and processes. This approach has already shown promise, as we have identified interactions between AltProts and known proteins, providing insights into their potential functions and contributions to cellular mechanisms. Using the XL-MS approach, we can identify protein-protein interactions in an untargeted manner. This technique allows us to reveal the presence of AltProts within protein interaction networks and predict their roles in signaling pathways. For instance, using this approach, IP_166911.1 (AltATAD2) was found interacting with RPL10 and AUF1, possibly regulating the binding of 5S ribosomal RNA in the ribosome. Similarly, in glioblastoma cells (NCH82), several AltProts were identified interacting with TPM4, and some of these interactions were confirmed by Immunofluorescence and TurboID. We have optimized the detection of crosslinked peptides through subcellular fractionation techniques, enabling the identification of AltProts specific to cellular localization and their interactions. This approach revealed AltProts interacting with HLA-B, particularly at the B2M interaction site. AltProt IP_2284785 might be involved in the stability or inhibition of the MHC-I complex. Our ongoing research aims to expand the understanding of AltProts and their roles in cellular processes, potentially leading to new insights into cellular mechanisms and the development of novel therapeutic targets. Fig. 1 (optional) 1082

1083

P-III-0925 Large-scale proteomics as a tool to disclose molecular mechanisms of the phenolic compound gibbilimbol B Carolina Cardona1, Gina Mendez1, Diego Muñoz1 1Universidad de Ciencias Aplicadas y Ambientales, Medical school, Bogotá, Colombia Cancer research has made significant progress in recent years, particularly with the application of new omics technologies. By addressing proteomics, it is possible to gather vast amounts of data that offer a comprehensive view of the molecular mechanisms involved in potential treatments. This research aims to use label-free quantitative proteomics coupled with pathways and networks-based bioinformatics analysis, to uncover the molecular mechanisms of the natural alkylphenol Gibbilimbol B isolated from P. eriopodon. Dynamic proteomic profiling of breast tumor cell lines treated with Gibbilimbol B showed the impact of several molecular pathways in a time and cell-dependent manner. The activation of the apoptotic pathway was confirmed for both MCF-7 and MDA-MB-231 cells, nevertheless, the proteomic profile provided new insights into molecular mechanisms triggered by ubiquitin-dependent protein catabolic process and the identification of P53-Independent DNA damage checkpoint specific for MCF-7 cells among others. On the contrary, in MDA-MB-231, the pattern of the protein profile was correlated with a possible response to endoplasmic reticulum stress, the activation of nuclear receptors in response to pro-inflammatory processes, and an immune response related to antigen processing. Overall, these advances provide detailed insights at the cellular level for understanding potential treatment strategies. 1083

1084

P-III-0926 The inadequate expression of LCRMP-1 leads to sinoatrial dysfunction through disrupting calcium cycling Xuan-Ren Chen1,2, Chung-Lieh Hung3,2, Ya-Ting Yang1, Szu-Hua Pan1,4,5 1College of Medicine, National Taiwan University, Graduate Institute of Medical Genomics and Proteomics, Taipei, Taiwan 2Mackay Medical College, Institute of Biomedical Sciences, Taipei, Taiwan 3MacKay Memorial Hospital, Division of Cardiology, Departments of Internal Medicine, Taipei, Taiwan 4Academia Sinica and National Taiwan University, Genome and Systems Biology Degree Program, Taipei, Taiwan 5National Taiwan University, Doctoral Degree Program of Translational Medicine, Taipei, Taiwan Introduction and Objectives: The sinoatrial (SA) node acts as the heart's natural pacemaker, regulating muscle contractions for blood flow. Sick sinus syndrome (SSS), caused by SA node dysfunction, can lead to various heart issues. Diagnosis involves electrocardiography and may require an implanted pacemaker. However, there is a need to find new biomarkers for diagnosis and develop alternative therapeutic approaches to address associated limitations and risks. Previously, we found that LCRMP-1 can enhance cancer metastasis and promote tumor angiogenesis. Recently, we observed signs of sinoatrial arrhythmia in both young and older LCRMP-1 knockout (KO) mice, along with bi-atrial enlargement in the older mice. This suggests LCRMP-1 may regulate sinoatrial heart activity, requiring further investigation for a comprehensive understanding. Methods: To investigate the impact of LCRMP-1 on cardiac function, we created specific LCRMP-1 KO mice using two different methods. We used echocardiography and MRI to evaluate the cardiac structures and functions and monitored the rhythmic activity in LCRMP-1 KO mice using ECGs. Subsequently, we established LCRMP-1 silenced HL-1 stable atrial cardiomyocytes and assessed their electrophysiological properties using a microelectrode array (MEA). To gain insight into the detailed mechanism of LCRMP-1-regulated heart rhythm, we conducted a spatial interactome analysis of LCRMP-1 protein in the tissues. We established a specific LCRMP- 1 protein linkage map in the SA node and right atrium regions. Additionally, we investigated the dynamic changes and detailed action mechanisms of LCRMP-1 on calcium cycling in vitro. Results and Discussion: The presence of occasional absent P waves, particularly noticeable in younger mice, suggests a potential manifestation of SSS due to LCRMP-1 deficiency. This indicates a direct impact on the function of the sinoatrial node, leading to irregular heart rhythms. In our spatial IP-MS analysis, we revealed a network of LCRMP-1 interacting proteins associated with metabolic pathways and ion homeostasis, especially those related to calcium ion channels. This analysis provided a comprehensive view of the proteins that interact with LCRMP-1 and their potential roles in cardiac function. Reciprocal Co-IP assays confirmed the binding of LCRMP-1 to calcium-regulating proteins, emphasizing its role in modulating calcium release channels. Calcium cycling studies in HL-1 cardiomyocytes demonstrated that LCRMP-1 knockdown disrupted calcium dynamics, leading to irregular calcium waves. Conclusion: LCRMP-1 is critical in maintaining cardiac structure, electrical rhythm, and calcium cycling. LCRMP-1 KO mice showed right atrial dilation and SSS, which were linked to disrupted calcium cycling. These findings emphasize the importance of LCRMP-1 as a key regulator of the sinoatrial area and offer new insights into potential pathways for SSS. Keywords: Sick sinus syndrome, LCRMP-1, Spatial IP-MS 1084

1085

P-III-0927 Characterisation of the MSS-CRC surface proteome for novel immunotherapeutic target discovery Daryl Chin1,2, Ziliang Ma1,3, Wei Wu1,3,2 1A*STAR Singapore Immunology Network, Singapore, Singapore 2National University of Singapore, Department of Pharmacy and Pharmaceutical Sciences, Singapore, Singapore 3Utrecht University, Biomolecular Mass Spectrometry and Proteomics, Utrecht, Netherlands Cancer immunotherapies represent an emerging forefront in the treatment of various terminal malignancies. Unfortunately, existing single-arm and combinations of immunotherapies for microsatellite-stable colorectal cancer (MSS-CRC) are poorly efficacious due to the low tumour mutational burden and an immunosuppressive tumour microenvironment. An emerging strategy to elucidate novel immune-druggable targets involves characterising the "surfaceome", as the cell membrane is a crucial and therapeutically accessible interface. Here, we focused on deconvoluting the MSS-CRC "surfaceome" for novel immunotherapeutic target identification by performing extracellular biotinylation of intact patient-derived MSS-CRC organoids and quantitative proteomics on affinity-enriched surface proteins, comparing healthy, primary and metastatic MSS-CRC. From a total pool of 487 highly confident surface proteins, we first demonstrate the reliability and clinical relevance of our dataset through the detection of high levels of prominent CRC progression markers CDH17, LY75 and CA9. Importantly, we shortlisted 2 potential targets DPEP1 and SCARB1 which displayed differentially upregulated protein expression in primary MSS-CRC and further validated using tissue microarray and gene silencing approaches. The discovery of these new MSS-CRC-specific prognostic markers can potentially guide clinical staging and disease stratification in a multi-biomarker panel setting. By adopting an empirical biochemical approach, we ultimately aim to progress targeted therapeutic strategies in MSS-CRC. The success achieved here could also be directly applied extensively to study other forms of advanced solid tumours, and tumour subtypes. 1085

1086

P-III-0929 In-depth proteomic profiling of eosinophils and neutrophils to unveil molecular insights into immune system function Byoung-Kyu Cho1, Chatsuda Thamrongsak1, Carol Brenner-Webster1, Kathrine E. O’Connor1, Rachel A. Idol1, Antonia Zamacona Calderon1, Michelle A. Gill1, Young Ah Goo1 1Washington University School of Medicine, St. Louis, MO, United States Eosinophils (Eos) and neutrophils (PMNs) are crucial components of the immune system, defending against infections and participating in immune responses. They collectively belong to a category of white blood cells known as granulocytes, characterized by the presence of cytoplasmic granules. Eos and PMNs are each linked with various inflammatory diseases and thus represent specific therapeutic targets being evaluated in ongoing clinical trials. There are gaps in knowledge, however, regarding the full range of differences between these cells in health as well as how these individual cell types are specifically impacted by inflammatory states and diseases. In this study, we employed an in-depth proteomics approach using highly purified Eos and PMNs to categorize the proteome present in each of these important granulocyte subsets in healthy adults as a first step in defining how Eos and PMNs proteomes can subsequently be studied in disease states. Eosinophils and neutrophils were each purified from blood samples of four healthy adult donors enrolled in a local immune cell study (IRB # 202207037). Granulocyte fractions were first isolated by density gradient-based centrifugation and dextran sedimentation and subsequently utilized to purify Eos and PMNs by negative selection with antibody-labeled magnetic beads (STEMCELL Technologies). The purity of each subset was > 95%. For proteomics sample preparation, 500,000 cells per sample were used, representing < 5% of the amount used in previously reported proteomics studies of these cells. LC-MS analysis was performed on the timsTOF Pro2, coupled with the nanoElute system using dia-PASEF acquisition. The data were processed using directDIA analysis with Spectronaut 18 software, and functional enrichment analysis was performed using STRING (version 12.0). As a result, 5322 protein groups were identified, comprising 5246 eosinophil and 5044 neutrophil proteins. To specify the enriched proteome in each cell type, label-free quantification (LFQ) was conducted through a t-test utilizing log2 protein quantities. Proteins meeting the criteria of q-value < 0.05 and fold change > 2 were designated as significantly enriched, identifying 628 eosinophil-enriched and 543 neutrophil-enriched proteins. Functional enrichment analysis of the enriched proteins for each cell type highlighted shared categories involved in the innate immune system, infectious diseases, endocytosis, lipid metabolism, and neurodegeneration, all associated with the hallmark of the immune system. Our findings provide insights into the specific proteomes of eosinophils and neutrophils, innate immune cells with critical roles in both health and disease. Further analysis of these data will enable a deeper understanding of differences in these cell types and provide the foundation for future comparisons to Eos and PMNs proteomes in inflammatory states where these cell types are specifically associated with clinical disease. 1086

1087

P-III-0930 Comprehensive proteomic profiling of converted Adipocyte-like cells from normal human dermal fibroblast using Data-Independent Acquisition Mass Spectrometry (DIA-MS) Sunkyu Choi1,2, Rajaa S.D. Dalloula1,2, Praveen Babu Vemulapalli1,2, Sondos Yousef1,2, Neha Goswami1,2, Frank Schmidt1 1Weill Cornell Medicine in Qatar, Research, Doha, Qatar 2Weill Cornell Medicine in Qatar, Doha, Qatar Adipocytes play an important role in the regulation of systemic energy homeostasis and are closely related to metabolic disorders such as type-2 diabetes and inflammatory bowel diseases. Particularly, there is an increasing need for a human adipocyte model for studying metabolic diseases and obesity study. However, utilizing human primary adipocyte culture and stem-cell based models presents several practical limitations due to their time- consuming nature, requirement for relatively intensive labor, and high cost. Here, we applied direct conversion of normal human dermal fibroblast (NHDF) into adipocyte-like cells using adipogenic cocktail containing IBMX, dexamethasone, insulin, and rosiglitazone and confirmed prominent lipid droplet accumulation in the converted cells. For profiling of the proteome changes in the converted cells, we conducted comprehensive quantitative proteome analysis in intracellular proteome fraction and extracellular proteome fraction. We observed several proteins, which are known to be highly expressed in adipocytes specifically, were dominantly increased in the converted cells. In this study, we suggest that NHDF can be converted into adipocyte-like cells by adiopogenic cocktail and can serve as a useful tool to address human adipocyte study and its metabolism study. Fig. 1 (optional) 1087

1088

P-III-0931 Mechanical response proteomics of cancer cells Jiawei Dai1, Huali Shen1 1Fudan University, IBS, Shanghai, China Cancer cells encounter various mechanical stresses from the extracellular microenvironment. Understanding how cancer cells adapt to these stresses is critical for developing innovative anticancer therapies. In this study, we designed cell culture systems that imposed five types of mechanical stresses-fluid viscosity, fluid shear, matrix adhesion, stretching, and confinement pressure, and studied the proliferation and migration responses of 8 cell lines representing 6 cancer types under these conditions. We also performed proteomic analysis of the 8 cancer cell lines under different mechanical force conditions. Among them, we identified 10,790 proteins in the fluid viscosity and matrix adhesion stress tests, of which 2,931 showed significant mechanical force responses. These included 1,413 proteins that changed under high fluid viscosity stress, 1,264 proteins that changed under high matrix adhesion stress, and 286 proteins that changed in both conditions. Correlation analysis showed that there were different mechanical response pathways between glioma and liver cancer cells compared with lung cancer, breast cancer, and colorectal cancer. In addition, differential protein expression was not only related to cell migration, but also to mitochondrial inner membrane function, matrix ATP binding, and energy metabolism, highlighting the important role of mitochondria in managing cellular processes under stress. Notably, ITIH4 and A2MG were upregulated in all cells under fluid viscosity stress, with the increase in ITIH4 further validated by WB, indicating its role as a mechanosensitive factor. In addition, FN1 protein was significantly upregulated in cancer cells facing high matrix adhesion stress, indicating its involvement in responding to this specific stress. The comprehensive approach of this study combining mechanical stress systems with proteomics facilitated the discovery of novel mechanoresponsive molecules. Future studies will aim to identify key proteins in stress response pathways, explore their molecular functions, and elucidate the mechanosensitivity mechanisms of cancer cells, thereby facilitating the development of targeted anticancer therapies. 1088

1089

P-III-0932 Targeted proteomics reveals the regulation of the polycomb repressive complex 1 by USP7 and its involvement in neurodevelopment Jeroen Demmers1 1Erasmus University Medical Center, Proteomics Center, Rotterdam, Netherlands Ubiquitin-specific protease 7 (USP7) is a deubiquitylating enzyme that is involved in the regulation of multiple key cellular processes, including tumor suppression, transcription, epigenetics, the DNA damage response, and DNA replication. For debiquitinating enzymes, the regulatory mode of action is at the posttranslational level and proteomics tools are particularly suited to study this. Here, we took a multi-omics approach with a heavy targeted quantitative proteomics component to define the core USP7 network and to study its roles in regulation of the Polycomb system and in neurodevelopment. We performed interactomics, in-depth global proteomics and ubiquitinomics to identify the USP7 interaction network and its regulation of neurodevelopment. Targeted proteomics using parallel reaction monitoring (PRM) was performed to accurately quantify relevant players of the Polycomb system and other suspected USP7 targets in a label free manner. Integrating unbiased proteomics, genomic, and functional molecular approaches, we determined the core USP7 regulatory circuitry. This multi-angle analysis establishes USP7 as a regulatory hub in a multinodal network involved in tumor biology, protein (de)ubiquitylation, genome regulation and neurodevelopment. We found that USP7 modulates the ncPRC1 axis at the posttranslational level through stabilization of ncPRC1.6 and ncPRC1.1. At the transcriptional level, USP7 silences the autism related protein AUTS2, the subunit that suppresses H2A ubiquitylation by ncPRC1.3/5. Collectively, these USP7 activities increase the genomic deposition of H2AK119ub1. The connection of USP7 to the Polycomb system suggests that its role in cancer extends beyond regulation of p53. Our interactomics assay shows that USP7 has a remarkable range of interaction partners, of which only a portion appears to be stabilized by USP7. Next, we focused on the functions of these interaction partners and of USP7 itself in neurodevelopment. Neuronal cells in which USP7 was knocked out did not differentiate and no neurite outgrowth was observed, in contrast to WT cells. Global in-depth DIA-MS based proteomics suggested that proteins that were not expressed or only at lower levels in USP7 knockout compared to WT cells are involved in neuronal development and axon regulation. In contrast, expression of proteins involved in various metabolic processes was downregulated or switched off in differentiating WT cells. Current studies address the molecular mechanisms involved in the regulation of neurodevelopment by USP7. Combined, our multi-omics results provide a resource for future studies on the role of USP7 (neuro)development and cancer. in Fig. 1 (optional) 1089

1090

P-III-0933 Comprehensive proteomics landscape of human myeloid antigen-presenting cells Anniek L de Jager1, Sara Kassem1, Inge F de Laat1, Alesha Louis1, Kyra van der Pan1, Brigitta AE Naber1, Bas de Mooij1, Jacques JM van Dongen1,2,3,4, Cristina Teodosio1,2,3,4, Paula Díez5,1,6 1Leiden University Medical Center, Department of Immunology, Leiden, Netherlands 2Cancer Research Center, Translational and Clinical Research Program, Salamanca, Spain 3University of Salamanca, Department of Medicine, Salamanca, Spain 4Institute of Biomedical Research of Salamanca (IBSAL), Salamanca, Spain 5University of Oviedo, Department of Functional Biology, Oviedo, Spain 6University of Oviedo, Department of Physical and Analytical Chemistry, Oviedo, Spain Antigen-presenting cells (APCs) are key players in response to pathogens, bridging innate and adaptive immune responses. This heterogeneous set of populations includes dendritic cells (myeloid, myDC; plasmacytoid, pDC), monocytes (classical, cMo; intermediate, iMo, and non-classical, ncMo), and macrophages (MAC), which perform specialized functions while sharing the potential to recognize antigens. However, there is a lack of studies addressing how distinct stimuli elicit different responses in the same population and the impact of the cell background on the response to antigens. Furthermore, the limitations associated with studying ex vivo MAC populations led to the use of in vitro cell models, but little is known regarding the comparability of these models to their ex vivo counterparts. This study aimed to perform a thorough proteomics analysis of ex vivo human APC populations under viral and bacterial stimulated and unstimulated conditions, alongside distinct in vitro cell models. cMo, iMo, ncMo, pDC, and myDC were isolated from human peripheral blood (PB) samples (n=3-7) by cell sorting. cMo, pDC, and myDC subsets were stimulated for 18 h with either viral (R848, CpG ODN) or bacterial (LPS) ligands to evaluate the stimulation effects. To assess MAC models, THP1 cells were differentiated into MAC (dTHP1) (200nM PMA, 72h), and CD14+ monocytes were differentiated with M-CSF (M0 MAC) and polarized into M1 and M2 using IFNg and IL-4, respectively. Tissue MAC were isolated from healthy colon (n=5), skin (n=4), and peritoneal dialysate (n=3) samples. Cell samples were lysed, and 4 or 20 μg of protein were enzymatically digested and purified using a bead-based method. Peptide samples were labeled with tandem mass tags (TMTpro16) for quantitative analysis, measured on an Orbitrap Fusion™Lumos™Tribrid™ mass spectrometer, and analyzed using Proteome Discoverer v2.4. A total of 4,503 different proteins were identified across ex vivo APC populations, with 3,207 (71%) common proteins. Despite the lack of unique proteins due to their close maturational relationship, a modulation of protein patterns was observed. In vitro stimulation of PB populations led to changes in protein abundance, with a role for both the type of stimulus and the cellular background being observed e.g. proteases, like CTSZ, CPQ, and CTSH, potentially leading to different protein cutting patterns and antigen presentation processes. The comparison of in vitro and ex vivo MAC showed that there is not an optimal universal model. Proteome profiles of colon MAC were more similar to M0 and anti-inflammatory M2 MAC, suggesting a more homeostatic/regulatory role in the colon environment. In contrast, skin MAC were more akin to pro- inflammatory M1 MAC, and PD MAC displayed features closer to dTHP1 cells. Overall, this study shed insight on the processes involved in responses to antigens by APCs and highlighted the importance of selecting the appropriate MAC models when studying these cells. 1090

1091

P-III-0934 GPX1 expression promotes stemness and aggressiveness in sarcomas Carmen Huergo1,2, Juan Tornín1,2, Óscar Estupiñán1,2, Paula Díez1,2,3, Jun Gao4, M Victoria González1,2,5,6, Dzohara Murillo1,2, Borja Gallego1,2, Verónica Rey1,2,7, Cristina Viera1,2, Verónica Blanco1,2,6, Jessie Lavoie4,8, René Rodríguez1,2,7 1Instituto de Investigación Sanitaria del Principado de Asturias, Sarcomas y terapias experimentales, Oviedo, Spain 2Instituto Universitario de Oncología del Principado de Asturias, Oviedo, Spain 3University of Oviedo, Department of Immunology, Oviedo, Spain 4Centre for Oncology, Radiopharmaceuticals and Research, Health Canada, Biologic and Radiopharmaceutical Drugs Directorate, Ottawa, Canada 5University of Oviedo, Department of Surgery, Oviedo, Spain 6University Hospital of Asturias (HUCA), Department of Pathology, Oviedo, Spain 7Ciber en Oncología (CIBERONC), Madrid, Spain 8University of Ottawa, Department of Biochemistry, Microbiology and Immunology, Faculty of Medicine, Ottawa, Canada The sarcomagenic process initiates upon the tumoral transformation of mesenchymal stromal/stem cell (MSCs) or MSC-derived cell types. Experimental evidence also supports that sarcoma evolution is in part driven by the emergence of subpopulations of cancer stem cells (CSCs) after the accumulation of further epigenetic and/or genetic alterations. These CSCs are strongly associated with the evolution of tumors towards more aggressive behaviors. Therefore, the characterization of these subpopulations will contribute to develop more effective therapies against these tumors. With this aim, we used tandem mass spectrometry to the compare the proteome of adherent and CSC-enriched tumorsphere cultures in a tumor progression model of myxoid liposarcoma composed by three cell lines showing increasing aggressiveness after being serially transplanted in mice. In this analysis we found that the expression of the antioxidant enzyme GPX1 increased constantly during the CSC-enrichment process in all the cell lines of the model. Depletion of GPX1 resulted in decreased proliferation, tumorsphere-forming potential and migration capability in vitro and dramatically reduced tumor- formation ability in vivo. Conversely, GPX1 overexpression resulted in increased in vivo tumorigenicity. In agreement with this findings, GPX1 expression in sarcoma patients was associated to advanced tumor stages, more aggressive phenotypes, and a worse prognosis. To gain insight into the molecular basis behind the anti-tumor effects observed after the inhibition of GPX1 in sarcoma cells, we performed a proteomic LC-MS analysis in a hybrid Q-TOF mass spectrometer using triplicate samples of control and GPX1-depleted cells. ZenoSWATH data-independent acquisition (DIA) was used as the MS method. Comparing sh vs control conditions, we detected 116 differentially expressed proteins, associated with a downregulation of interferon-mediated response, the IL6/JAK/STAT3 axis and the NFκB-mediated signaling in GPX1-silenced cells. Overall, these results suggest that this factor may be a functional marker of aggressive CSCs subpopulations in sarcoma with prognostic and therapeutic implications. 1091

1092

P-III-0935 Enhancing high-grade serous ovarian cancer treatment through proteomic and phosphoproteomic analysis Parisa Esmaeili1, Fredrik Levander1, Kazi Uddin2, Ahmad Nasimian2, Anna Sandström Gerdtsson1, Sergio Mosquim Junior1 1Lund University, Immunotechnology, Lund, Sweden 2Lund University, Division of Translational Cancer Research, Lund, Sweden High-grade serous ovarian carcinoma (HGSOC) is the most lethal form of ovarian cancer, accounting for the majority of ovarian cancer-related deaths. It typically presents at advanced stages, often with metastasis, complicating treatment efforts. Despite an initial response to chemotherapy, resistance frequently develops, emphasizing the need for new therapeutic strategies. Identifying signature proteins that could serve as potential drug targets for overcoming resistance in metastasized high-grade ovarian cancer would be of immense value in addressing this challenge. In this study, we investigated the interaction between MTDH and SND1, both of which have been shown to be overexpressed in various types of cancer by others and us. We aimed to examine their roles and potential as therapeutic targets in ovarian cancer. We performed comprehensive proteomics and phosphoproteomics using cell line models to investigate the changes resulting from the inhibition of MTDH-SND1 interaction. Additionally, we conducted proteomics and phosphoproteomics on cells with silenced MTDH or SND1 to observe the effects of gene silencing, allowing us to compare these effects with those observed from direct inhibition. The data were analysed using enrichment analysis of gene ontologies and pathways to gain insights into the underlying biological processes and pathways involved. Inhibiting the MTDH and SND1 interaction resulted in a reduction in cell growth and motility-related pathways, as well as in the molecular functions of the cancer cells. These effects were comparable to those observed when the target genes were silenced. This similarity suggests that both inhibition and silencing of the drug targets disrupt pathways and molecular functions necessary for cancer cell proliferation and progression. Additionally, we observed a significant reduction in cell viability in cisplatin-resistant cells in both monolayer and spheroid models when the interaction between the drug targets was blocked. This finding highlights the potential of targeting the MTDH-SND1 interaction to overcome chemoresistance and improve therapeutic outcomes in ovarian cancer. Furthermore, the study underscores the crucial role of proteomic and phosphoproteomic analyses in identifying, characterizing, and validating novel therapeutic targets for effective treatment strategies in ovarian cancer. 1092

1093

P-III-0936 Characterizing surfacome and secretome dynamics upon cellular perturbation Martin Fehmer1, Christian Münch1 1University Hospital Frankfurt a.M., Institute of Molecular Systems Medicine, Frankfurt, Germany Efforts towards characterizing the cellular surface proteome and secretome with the help of LC-MS approaches has proven to yield valuable resources for the identification of biomarkers as well as uncovering compositional changes upon perturbation and alteration of cellular homeostasis. Although comparative analyses are highly effective in revealing differences in protein abundances between separate conditions, they do not fully encompass potential effects that could manifest within the translation as well as shuttling of constituents that make up the plasma membrane proteome as well as the extracellular secretome. In order to investigate this additional aspect of proteostasis, this project established a time-resolved assay to measure the accumulation rates of newly synthesized proteins at and outside the cell surface over time. To this end, a cell surface capture approach (CSC) is combined with TMT-SILAC multiplexing to allow mass spectrometry-based time-resolved quantitation of newly-synthesized plasma membrane protein populations. Secretome samples harvested from cell culture supernatant are processed in a similar manner. We build upon a method previously established within our group called mePROD which enables quantification of proteins with a low heavy-to light ratios. With this approach, we hope to accurately delineate and characterize dynamic changes in cell surface proteome and secretome subpopulations upon general or specific cellular disturbance. Through establishing a bioinformatics analysis pipeline, we furthermore strive towards translating initial findings towards certain disease models in the future. 1093

1094

P-III-0937 High-throughput profiling of PROTAC-induced protein degradation using mePROD quantitative proteomics Giulio Ferrario1, Jasmine Schäfer1, Bhavesh Parmar1, Nikita Verheyden1, Christian Münch1 1Goethe University Frankfurt, Frankfurt, Germany Targeted protein degradation (TPD) is an effective and potent strategy in drug discovery that involves the elimination of proteins of interest (POIs) using small-molecules as degraders. Proteolysis targeting chimeras (PROTACs) are a major class of such compounds that catalytically mediate the transfer of ubiquitin units from an E3 ligase complex to a protein of interest thereby marking the POI for subsequent proteasomal degradation. Since the major perturbation induced by the degraders is at the protein level, mass spectrometry-based quantitative proteomics serve as a one of the most direct methods of assessing their efficacy and selectivity. These methods are essential to illustrate the downstream effects of PROTACs, especially as they do not require prior knowledge of the specific target of the selected molecule. The disadvantage of these routine proteomics approaches, such as Label Free, Tandem Mass Tag or even SILAC, is that they provide insights into global proteome changes induced by degraders, without clear discrimination between primary and secondary targets. Moreover, such analyses are normally applied to a small subset of molecules, often when the final drug candidate has already been selected and partially optimized. In addition, the high dynamic rate of the changes at the proteome level, is often investigated to obtain details into the activity of PROTAC candidates but these approaches require long labelling times, which can be counterproductive in this context of understanding PROTACs impact on protein translation, modification and degradation. To address these limitations, we present the use of our multiplexed enhanced protein dynamics (mePROD) proteomics approach to obtain information on protein degradation and translation upon treatment with different PROTACs molecules. As a higher throughput approach would be beneficial to discriminate and confirm putative targets and to investigate the specificity of newly synthesized molecules, we have developed a pipeline to carry out the entire procedure in a 96-well format, limiting input and handling times to allow simultaneous proteome, degradome and translatome testing of over 20 PROTACs in replicates in a single plate. This comprehensive strategy promises to increase the throughput of analyzed small molecule degraders while providing a standardized method to generate large datasets for a more rigorous analysis of on- and off- targets effects of multiple known and unknown degraders, improving compound evaluation and optimization. By accelerating the discovery of therapeutic targets and deepening our understanding of protein dynamics in this cellular context, this approach is expected to facilitate the advancement of drug discovery and development of new PROTAC compounds. 1094

1095

P-III-0938 The study of fundamental biological determinants of the plasma proteome using proteographTM rise and murine models Matthew Chang1, Mark Flory1, Aaron Grossberg1, Jessie May Cartier1, Elise Manalo1, Arnaud Quentel1, Terry Morgan1 1Oregon Health and Science University, Portland, OR, United States Background: Seer ProteographTM now unlocks the ability to proteomically profile complex liquid biopsy types with deep sampling, scalability, and automation. While this platform is now being employed for discovery of disease-associated biomarkers in human specimens, the genetic malleability and low biological variance of murine models provides a unique opportunity to investigate the relationship between disease biology and biomarker dynamics. We employ the platform here to systematically probe the effect of cancer stage, driver mutations, and tissue of origin on plasma proteome composition in murine cancer. The primary objective of this study was to test the feasibility of the ProteographTM platform in the detection and quantification of murine plasma proteins. Methods: Plasma from multiple autochthonous genetic models of Kras-mutant cancers was processed using ProteographTM Rise. Matched tissue from select arms was cryomilled and prepared using PreOmics iST. Peptides were analyzed with 60 min total acquisition length liquid chromatography-mass spectrometry (LC-MS) on a Bruker timsTOF Pro with dia-PASEF. Leftover tissue peptides were pooled and fractionated for analysis by DDA- PASEF to generate a spectral library. Raw data were searched using DIA-NN on Seer PAS. Results: An average of 4440 protein groups and 25481 peptides were observed per sample with 6682 proteins and 51710 peptides identified across the entire study. Low technical assay and within-arm biological variability were observed with median coefficient of variation from protein intensities 40.8%. Statistical analysis revealed numerous differentially expressed proteins with striking differences across several models. Functional enrichment profiling revealed novel insights into the specific ontology associated with these differences. Conclusion: Deep sampling into the plasma proteomic dynamic range with low within-arm variability and resulting biological insights of this data demonstrate the power and feasibility of ProteographTM in a murine model and how this unique application can be leveraged to (a) systematically probe multiple determinant factors affecting plasma proteome composition and (b) potentially unravel the mechanisms linking these proteomic changes with the underlying disease biology. 1095

1096

P-III-0940 High throughput affinity pull-down interactomics Louisa Grauvogel1, Andre Michaelis1, Matthias Mann1 1Max-Planck Institute of Biochemistry, Proteomics, Munich, Germany Protein-protein interactions are of great value in biology as they place proteins in a functional context. Interactomics has long been a cornerstone of the proteomics tool chest and is routinely performed in laboratories worldwide. Based on a streamlined and high sensitivity technology capable of acquiring 60 interactomes per day, we recently reported the first substantially complete interactome of an organism – baker"s yeast. We found that almost all proteins are "social" meaning that they have at least one interaction partner and that the mean degree of connection between proteins is surprisingly similar to that in human social networks. Here we report on our efforts to further scale up and optimize out interactomics workflow. We compare the timsTOF and the new Orbitrap Astral platforms for this purpose. We also investigate different affinity tags, including the recently introduced ALFA tag which is a new, very short epitope that is rationally designed to form a stable alpha helix, against which a highly optimized nanobody with low picomolar affinity exists. We test and compare custom made nanobody-coated plates for this purpose. Furthermore, we take advantage of even faster Evosep gradients to increase the number of interactomes to at least 120 per day, thereby doubling the already high throughput. This opens up the capability to acquire highly standardized interactomes as a function of various perturbations. We apply our improved interactomics technology to a variety of human cellular systems. We are particularly interested in the interaction of a broad range of viral factors with the human intracellular host machinery. 1096

1097

P-III-0941 Spatiotemporal dissection of protein localization to human primary cilia Jan N. Hansen1,2, Kalliopi Tzavlaki2, Huangqingbo Sun1, Casper Winsnes2, Emmie Pohjanen1,2, Christian Gnann2, Mathias Uhlen2,3, Diana Mahdessian2, Ulrika Axelsson2, Emma Lundberg1,2,4,5 1Stanford University, Department of Bioengineering, Stanford, CA, United States 2Science for Life Laboratory, School of Engineering Sciences in Chemistry, Biotechnology and Health, KTH - Royal Institute of Technology, Stockholm, Sweden 3Karolinska Institute, Department of Neuroscience, Stockholm, Sweden 4Stanford University, Department of Pathology, Stanford, CA, United States 5Chan Zuckerberg Biohub, San Francisco, CA, United States Primary cilia are solitary, filamentous membrane protrusions emanating from the surface of most human cells. Most of what we know about their function stems from hereditary diseases that are caused by cilia dysfunction, so-called ciliopathies. Ciliopathy patients have a broad spectrum of phenotypes, underlining that cilia fulfil pleiotropic and tissue-specific functions. However, the molecular mechanisms underlying primary cilia function and cell-type specific roles of primary cilia remain poorly understood. This is mainly due to the lack of tools to reveal the protein composition of individual cilia. We address this challenge with antibody-based, spatial proteomics, and develop a cilia atlas as part of the subcellular section of the Human Protein Atlas (HPA) (https://www.proteinatlas.org/humanproteome/subcellular). We co-stained 1800 validated antibodies from the HPA with markers for nuclei, basal bodies, and primary cilia in three human cell lines of different tissue origin: hTERT-RPE1 cells from the retina, RPTEC/TERT1 cells from the kidney proximal tubules, and the mesenchymal stem cell line ASC52telo from the adipose tissue. We acquired high-resolution confocal 3D images and manually annotated as well as computationally analyzed each image. Our preliminary analysis revealed approximately 500 proteins to localize to primary cilia, with 200 of these proteins also detected at the basal body. We detected an additional 250 proteins that localized to basal bodies but not to cilia. The identified ciliary proteins highlight primary cilia as cellular antennae: At least 170 of the identified ciliary proteins are known to drive signal transduction and engage, e.g., in Hedgehog, GPCR, Calcium, TGFB, RTK, Rho GTPase, MAPK, or mTOR signaling. Furthermore, we observed a remarkable heterogeneity, both between cell lines and within supposedly identical cell populations. For approximately 75% of the identified ciliary proteins, we observed variable amounts in cilia of the same cell line. Furthermore, we identified only about 20% of the ciliary proteins in cilia of all three cell lines. Beyond that, for about 90% of the ciliary proteins, we observed co-localization to other subcellular compartments, with cytosol, basal body, microtubules of the cell body, and nucleoplasm being the most common other compartments. By computational analysis of the images, we relate the dynamic ciliary protein localization to cellular states. In conclusion, we reveal new ciliary proteins and demonstrate that the primary cilium is the most dynamic and diverse cellular compartment. Our cilia atlas will pave new avenues for basic and clinical research on cilia and ciliopathies. 1097

1098

P-III-0942 Mapping of the synaptic mitochondrial interactome by crosslinking mass spectrometry (XL-MS) Carina Hansohn1,2, Charlotte Krawczyk1, Sofia Ainatzi1, Ivan Silbern3, Olexandr Dybkov1, Henning Urlaub1,2 1Max Planck Institute for Multidisciplinary Sciences, Bioanalytical Mass Spectrometry, Goettingen, Germany 2University Medical Center Göttingen, Clinical Chemistry, Goettingen, Germany 3Evotec GmbH, Munich, Germany Neuronal communication is based on synaptic signal transmission with neurotransmitter release from synaptic vesicles at presynaptic compartments upon stimulation of the synapse. The complex molecular processes involved in synaptic signal transmission are subject to various studies, yet they are still not fully understood. Furthermore, it has not yet been established in which ways the high energy levels necessary for these processes are provided. High numbers of mitochondria, which are responsible for generation of the vast majority of cellular energy among various other functions, are also present in distal axons and presynaptic compartments. However, relatively few studies have been dedicated to the role of mitochondria in neuronal cells and especially at the synapse. The aim of this study is to provide an unbiased, deep mapping of the synaptic mitochondrial protein interactome using chemical crosslinking mass spectrometry (XL-MS). Chemical crosslinking of proteins is based on the formation of covalent bonds between amino acid residues in close spatial proximity upon reaction with the crosslinker. Subsequent MS analysis of endo-proteolytically digested protein samples results in identification of the crosslinked amino acid residues and their position within the peptide or protein. This provides insight into protein interactions based on the spatial restraints of crosslinking reactions. We performed chemical crosslinking under near-physiological conditions on synaptosomes (pinched-off synaptic terminals), which are common model systems for the study of synaptic functions. Synaptosomes were isolated and purified from rat brain tissue, followed by crosslinking and further subfractionation of crosslinked synaptosomes with enrichment for (intra-synaptosomal) mitochondria. LC-MS acquisition was performed upon proteolytically digested peptide samples pre-fractionated with reverse-phase C18 chromatography at basic pH. The XL-MS workflow was systematically optimized to maximize the number of identified inter-protein crosslinks. As results obtained by XL-MS studies are still limited by the complex computational analysis of MS acquisition files for crosslink identifications, two crosslinkers (disuccinimidyl suberate (DSS) and disuccinimidyl sulfoxide (DSSO)) with different properties and different subsequent search strategies were applied in this study and compared for quantity and quality of crosslink identifications. The resulting inventory of inter-crosslinks of mitochondrial proteins, as well as between mitochondrial and other synaptosomal proteins, were visualized in an interaction map, which can hopefully provide the basis for follow-up experiments to explore selected protein- protein interactions of mitochondria at the synapse. 1098

1099

P-III-0943 Reducing ribosome biogenesis upregulates interferon-stimulated gene translation Brenna N Hay1, Rachel Smid1, Nathan Louie1, Stephane Flibotte2, Eric Jan1, Leonard J Foster1 1University of British Columbia, Biochemistry and Molecular Biology, Vancouver, Canada 2University of British Columbia, Bioinformatics Core Facility, Vancouver, Canada The interferon response is a signaling pathway unique to vertebrates that links the innate and adaptive immune responses. Interferons signal through a cascade of factors, including the JAK-STAT pathway, to induce the transcription of hundreds of interferon signal transduction pathways and ISGs have been elucidated, translational regulation of ISG transcripts has not been fully investigated. Prior work demonstrated that ribosomal protein RPL28 negatively regulates a subset of ISGs. Recent findings suggest that ribosome biogenesis may impact the translational regulation of ISGs. The aim of this project is to examine the relationship between ribosome perturbations and ISG translation under interferon stimulation. interferon-stimulated genes (ISGs). Although the main To determine how ribosome biogenesis and the interferon response are associated, we knocked down one of the key ribosome biogenesis factors, BOP1, in A549 cells, with and without interferon treatment. The resulting samples were analyzed by mass spectrometry and RNA-seq. Differential expression analysis revealed a subset of ISGs which had higher protein abundance in the BOP1-depleted condition, further supporting that ISGs are uniquely regulated. To validate that this was regulation at the translational level, the RNA-seq data revealed no significant differential expression in the candidate ISGs in the knockdown condition compared to control with interferon stimulation. The protein- and gene-level differential expression results of interferon-stimulated BOP1 knockdown were aligned for further analysis. Specifically, the codon adaptation index (CAI) was calculated to investigate the codon optimality of genes with protein differential expression greater than the transcript differential expression. In both the overall dataset and an ISG dataset, genes that had higher protein differential expression had a lower CAI, indicating codon non-optimality. Canonically, codon non-optimality results in mRNA instability due to ribosome stalling and subsequent degradation; however, we propose that in instances of ribosome perturbation, such as ribosome biogenesis knockdown, the reduction in ribosome abundance prevents ribosome queuing and therefore increases mRNA stability and ultimately protein production (Figure 1). In times of stress, such as virus infection, ISGs and other immune-related proteins still need to be rapidly translated despite known reductions to host translation. It is possible that the reduced codon optimality of critical immune pathway transcripts is a method of ensuring robust translation of the immune response. Future work will clarify the role of codon optimality in the upregulation of both ISG and non-ISG proteins to further elucidate the proposed mechanism of translational regulation. Figure 1: A proposed model for the increased protein levels of ISGs during BOP1 knockdown and IFN stimulation (created with Biorender). 1099

1101

P-III-0944 Unravelling the role of reactive oxygen species in glucose-stimulated insulin secretion: insights from oxidative and phosphoproteomic analysis Theresa Hofmeister1, Fabian Tilz1, Michal Rackiewicz2, Chantal Greitsch1, Luise Nato1, Isabel Huber2,3, Sorel Sagu1, Matt Fuszard4, Andrea Henze1,2 1Martin-Luther-University Halle-Wittenberg, Institute of Agricultural and Nutritional Sciences, Potsdam, Germany 2University of Potsdam, Institute of Nutritional Science, Potsdam, Germany 3Health and Medical University, Anatomy, Potsdam, Germany 4Martin-Luther-University Halle-Wittenberg, Core Facility Proteomic Mass Spectrometry, Halle (Saale), Germany Insulin-secreting beta cells are essential in regulating blood glucose levels and carbohydrate metabolism. However, β-cell dysfunction is a primary factor in the pathogenesis of diabetes. The process of insulin secretion is tightly regulated and involves shifts in cellular redox status, with hydrogen peroxide (H2O2) identified as a key component in glucose-stimulated insulin secretion. But, excessive production of reactive oxygen species (ROS) leads to β-cell exhaustion and loss. Therefore, understanding the redox proteome is crucial for understanding the precise interplay between ROS and insulin secretion. As changes in cysteine redox status can reflect alterations in redox signalling pathways and kinases and phosphatases are key regulators of insulin secretion and frequently modulated by redox signalling, in this study we analysed chances within the redox- and phosphor- proteome of beta-cells after glucose stimulation. Therefore, INS-1E cells were incubated with varying glucose concentrations following a period of starvation, lysed and non-oxidized cysteines were labelled using N- ethylmaleimide (NEM). Sequential alkylation enabled labeling of reversibly oxidized cysteines with cysteine- specific phosphonate adaptable tag (CysPAT), allowing simultaneous titanium dioxide-based enrichment of cysteine-oxPTM and phospho sites after tryptic digestion and labelling with TMT. Finally, samples were analysed using high-resolution LC-MS/MS (Orbitrap). This protocol was complemented by characterizing the oxidative status of β-cells during insulin secretion using 2',7'-dichlorodihydrofluorescein diacetate (H2DCFDA). INS-1E cells are demonstrating a typical biphasic insulin secretion pattern and a strong responsiveness to glucose and other secretagogues such as amino acids or ketone bodies. Furthermore, our findings reveal that major reversible oxidation events occur in proteins associated with localization, cellular, and metabolic processes. Notably, there were time-dependent alterations in glucose-stimulated insulin secretion and corresponding changes in cellular oxidative status. This highlights dynamic protein oxidation and increased ROS production upon glucose stimulation of beta cells, which also impacts the oxidative status of regulatory proteins. Through this comprehensive analysis, we aim to understand the complex signalling pathways governing insulin secretion and provide insights into the role of oxidative stress in beta-cell function and dysfunction. 1101

1102

P-III-0945 A comprehensive study of the effect of IDH1 mutation in glioblastoma multiforme progression combining proteome, transcriptome and clinical data Jiwon Hong1, Tae-gun Park2, Seunghoon Back1, Do Young Hyeon3, Chaewon Kang1, Junho Hur4, Junseok Hur2, Daehee Hwang3, Sang-Won Lee1 1Korea University, Department of Chemistry, Seoul, South Korea 2Korea University College of Medicine, Department of Neurosurgery, Seoul, South Korea 3Seoul National University, Lab of System biology and medicine (SBM), Seoul, South Korea 4Hanyang University College of Medicine, Department of Genetics, Seoul, South Korea Glioblastoma multiforme (GBM), an aggressive type of cancer, is known to occur in the brain or spinal cord. A genome-wide mutational analysis revealed a common mutation, IDH1 mutation, shown in GBM patients to interestingly show association with survival increase. Despite its importance, difficulties in IDH1 mutation cell line construction have limited extensive proteomic research. In this study, we performed comprehensive global and phosphoproteome profiling on successfully constructed IDH1 R132H point mutated cell-lines to investigate the effect of the mutation on the progression of GBM. With wild type U-87, early (day1) and late (day 6) samples of two successful IDH1 mutated U-87 cell-line batches, three replicates of 6-plex TMT labeled peptides were obtained. Proteome profiling using DO-NCFC-RP/RP-MS/MS platform [2016, Lee et al] resulted in averages of 51,595 distinct phosphopeptides with 41,603 phosphorylation sites, corresponding to 7,352 phosphoprotein groups, and 259,061 distinct peptides corresponding to 11,868 protein groups. Along with gene ontology patterning analysis of the proteome data, RNA-seq expression results were combined to seek overlap of differentially expressed proteins/genes. These genes were compared with selected signature genes from GBM patient subtyping from TCGA public data and analyzed microarray results regarding TMZ resistance in GBM patients to extract 12 core genes, for a network model construction. From these results, our model centered in the decrease of TGF-beta signaling pathway and DNA damage response to explain the better survival of GBM patients with IDH1 mutation, hoping to provide insight for future GBM treatments. 1102

1103

P-III-0946 Exploring the role of dynamic protein profiling in bone Marrow-Derived mesenchymal stem cells and their influence on metastatic behavior in non-Malignant non-small cell lung cancer cells Yuan-Ling Hsu1, Tang-Chi Li1, Wei-Hsin Lin2, Chia-Che Lee2, Szu-Hua Pan1,3,4 1National Taiwan University, College of Medicine, Graduate Institute of Medical Genomics and Proteomics, Taipei, Taiwan 2National Taiwan University Hospital, Department of Orthopedic Surgery, Taipei, Taiwan 3Academia Sinica and National Taiwan University, Genome and Systems Biology Degree Program, Taipei, Taiwan 4National Taiwan University, Doctoral Degree Program of Translational Medicine, Taipei, Taiwan Introduction and objectives: Lung cancer often spreads to the bones in about 50% of lung cancer patients, leading to poor clinical outcomes with standard treatment. Therefore, preventing bone metastasis is crucial for lung cancer patients. Recent research suggests that bone marrow-derived mesenchymal stem cells (BM-MSCs) have the ability to differentiate into osteoblasts and adipocytes. Moreover, these cells can be attracted to primary tumor sites by cancer cells, where they transform into activated myofibroblasts and contribute to the tumor microenvironment, thereby fueling cancer progression. Methods: Multiple primary BM-MSCs were isolated from patients with fractures. An in vitro cell co-culture system was established using primary BM-MSCs and CL1-0, a low-invasive lung cancer cell line. We evaluated the impact of BM-MSCs on lung cancer cells in vitro by examining proliferation, independent colony formation, migration, invasion, adhesion, and changes in cell morphology. Additionally, we delved into the secretome within the bone niche using a cytokines array. We performed integrated proteomic analyses on tumor cells by LC- MS/MS to unveil the detailed mechanisms behind BM-MSC-induced lung cancer-related bone metastases. Results and Discussion: The study's findings suggest that BM-MSCs enhance the invasion ability and distal colonization of low-invasive lung cancer cells. CL1-0 cells were co-cultured with BM-MSCs for extended periods to investigate the regulatory mechanisms involved. Cell lysates were then subjected to total proteomic analysis, and the conditioned medium was collected for secretomic dissection. This analysis identified critical pathways, including wound healing and intermediate filament cytoskeleton organization, associated with metastasis and affecting cellular morphology, migration, invasion, and adhesion. The results indicated that low-invasive cancer cells increase their malignancy by regulating epithelial-mesenchymal transition (EMT) within the bone microenvironment. Conclusion: Low-invasive lung cancer cells' migration and invasion abilities increased after co-culturing with BM- MSCs. These cancer cells tended to adhere to bone cells and exhibited morphological changes towards a mesenchymal-like appearance. Proteomics analysis identified proteins associated with cell migration, cell shape, and cytoskeleton organization, indicating that BM-MSCs enhance the malignancy and metastatic potential of low-invasive lung cancer cells. In summary, the malignancy of low-invasive cancer cells is elevated through EMT regulation within the bone microenvironment. These findings offer insights into the changes occurring in lung cancer cells under the influence of cytokines from bone cells, potentially guiding the development of new treatment strategies to effectively prevent bone metastasis from lung cancer. Keywords: lung cancer, bone metastasis, bone microenvironment, paracrine 1103

1104

P-III-0947 Using high-throughput expression and spatial proteomics to characterise remodelling of HEK293 cells during the production of recombinant adeno-associated virus (rAAV)-8 Charlotte Hutchings1, Kathryn Lilley1, Nicholas Bond2, Claire Harris2, Sara Conde Rodriguez2 1University of Cambridge, Department of Biochemistry, Cambridge, United Kingdom 2AstraZeneca, Cambridge, United Kingdom The potential of recombinant adeno-associated viruses (rAAVs) as therapeutic DNA delivery agents has been established with many rAAV-based gene therapies available on the market. Such rAAVs are most commonly manufactured via transient transfection of the necessary genetic components into HEK293 cells. Unfortunately, low yield and recovery during this process currently limits the wider application of rAAV delivery vehicles. Hence, to improve the manufacturing of rAAVs and ensure the long-term success of rAAV gene therapies will require a better understanding of the molecular mechanism(s) by which rAAVs are produced inside HEK293 cells. In this project, rAAV8 was produced in HEK293-VP cells using a standard transient triple transfection method. The rAAV8 yield was similar to those previously reported with 2.64 x 1011 viral genomes/mL at 96-hours post- transfection, as determined by qPCR. Using high-throughput mass spectrometry, the global abundance of cellular proteins was quantified and compared between control (mock transfected) and rAAV8-producing HEK293-VP cells at 48- and 96-hours post-transfection using a TMT workflow. By 48-hours post-transfection, 347 proteins were found to have significantly differential abundance in rAAV8-producing cells compared to control cells. This early response included the downregulation of cell cycle transition factors and DNA damage response proteins. At 96-hours post-transfection, 1,517 proteins displayed significant changes in abundance. Proteins with increased abundance were enriched for gene ontology terms relating to structural organisation of the cell whilst proteins relating to normal cellular metabolism were decreased in abundance. To further explore the effect of rAAV8-production on cellular organisation and protein localisation, global subcellular proteome maps were acquired for both mock transfected and rAAV8-producing HEK293-VP cells using the well-established Localisation of Organellar Proteins by Isotope Tagging after Differential Centrifugation (LOPIT-DC) method. By exploiting semi-supervised Bayesian classification methods, over 6,000 proteins were systematically localised to 12 distinct subcellular organelles and protein complexes in each condition. Comparison of these maps revealed that substantial subcellular remodelling takes place during rAAV8-production with over 50 proteins displaying differential localisation. The subcellular niches with the greatest degree of change were the nucleus, nucleolus and organelles of the secretory pathway. Further validation of these proteomic datasets using targeted assays and microscopy will provide insight into the key responses of HEK293-VP cells to rAAV8-production. Further, these datasets provide a rich resource for discovery of cellular factors which may promote or limit rAAV production for gene therapies. 1104

1105

P-III-0948 Intra- and extracellular proteome dynamics during normal and malignant hematopoietic stem cell expansion Maria Jassinskaja1, Helena Kooi1,2, Daniel Bode1, Sudip Ghosh3, Monika Gonka1, Juan Antonio Rubio Lara1, Alyssa Cull1, Lilia Cabrera Cosme1, Rachel Popplewell1, Alexander J. Hogg1, Elise Bennett1, Joanna Milek1, Fiona Bain1, Sophie Davies1, Samuel Elberfeld1, Adam Wilkinson4, Satoshi Yamazaki5, Jenny Hansson3, David G. Kent1 1University of York, York Biomedical Research Institute, York, United Kingdom 2Sanquin Research, Amsterdam, Netherlands 3Lund University, Lund Stem Cell Center, Lund, Sweden 4University of Oxford, Oxford, United Kingdom 5University of Tokyo, Tokyo, Japan Delineating intrinsic and extrinsic drivers of hematopoietic stem cell (HSC) self-renewal is critical to improve efforts in ex vivo HSC expansion and to better understand leukemia cell biology. Loss-of-function mutations in the TET2 gene are frequently observed in myeloproliferative neoplasms (MPNs) and confer HSCs with increased self-renewal potential leading to clonal outgrowth of mutant cells. Relatively few genes mediating MPN initiation and progression have been identified to date and transcriptomic approaches have had limited success in identifying novel therapeutic targets. The low number of HSCs within mouse and human tissues has historically precluded global characterization of the cellular and extracellular proteome of HSCs; however, parallel advances in ex vivo HSC expansion and low-input proteomics puts us on the cusp of breaking through this decades-old barrier. Here, our aim was to perform proteomic profiling of freshly isolated or ex vivo-expanded normal and Tet2-deficient HSCs, and to characterize soluble factors in the cells" microenvironment in vivo and in culture to generate a comprehensive molecular map of intra- and extracellular self-renewal drivers in normal and pre- malignant HSCs. We applied miniaturized, multiplexed sample preparation protocols in combination with mass spectrometry (MS)-based quantitative proteomics to compare normal and Tet2-deficient HSCs, and data- independent acquisition (DIA)-MS to characterize the extracellular environment of normal and pre-leukemic HSCs in vivo and during ex vivo expansion. We show that the cellular and secreted proteome accurately stratify HSCs based on functional potency and mutational status and identify novel molecular components not captured in transcriptomic analyses. On the pre-leukemia side, we reveal that Tet2-deficient HSCs have altered expression of extracellular matrix (ECM) proteins and that interaction with these proteins in artificial niches affects cellular function. Extracellular proteomic analysis further reveals that Tet2-deficient cells create a microenvironment that is pro-inflammatory and pro-thrombotic even in young, asymptomatic animals. In HSC expansion assays, proteomics identifies the requirement for intact DNA repair pathways as key components of HSC clones capable of extensive self-renewal compared to unsuccessful expansion cultures. In line with this, functional assays demonstrate that DNA repair enzyme Parp1 is a key player in ex vivo HSC expansion. Analysis of the secretome of unsuccessful cultures further identifies mast cell proteins as predictive of failure to expand engraftable HSCs. Collectively, our study highlights novel proteins to which transcriptomic studies are blind, and open new opportunities for HSC expansion and pre-leukemic biology, paving the way for future ex vivo and in vivo modulation of HSC function via manipulation of the cells and their extracellular environment. 1105

1106

P-III-0949 Understanding telomere-driven genomic instability through proximity labelling proteomics Grishma Rane1, Zi Wayne Sin1, Naasyidah Zulkaflee1, Yu Bin Ng1, Peh Fern Ong2, Oliver Dreesen2, Dennis Kappei1,3,4 1National University of Singapore, Cancer Science Institute of Singapore, Singapore, Singapore 2A*STAR Skin Research Labs, Singapore, Singapore 3National University of Singapore, Department of Biochemistry, Singapore, Singapore 4National University of Singapore, NUS Center for Cancer Research, Singapore, Singapore Telomeres and its associated proteins are essential for genome stability and limit the proliferative potential of cells. Cancer cells achieve replicative immortality either by re-expression of the enzyme telomerase or activation of a recombination-based mechanism called Alternative Lengthening of Telomeres (ALT). Hence, telomere maintenance is a universal hallmark of cancer and in addition telomere-driven genomic instability fuels oncogenic genomic alterations in the majority of cancers. In Previous work we had established that the zinc finger protein ZBTB48 directly binds to telomeric DNA and acts as a negative regulator of telomere length in both telomerase-positive and ALT-positive cancer cells. In the latter, ZBTB48 is particularly abundant at telomeres and its overexpression had been reported to cause rapid telomere shortening, but the exact molecular mechanism has not been definitively established. Here, we describe that ZBTB48 overexpression induces chromatin bridge formation as the primary cause of telomere attrition, eventually culminating in chromothripsis. ZBTB48-induced chromatin bridge formation is exclusive to ALT-positive cellular contexts, dose-dependent and leads to concomitant C-circle accumulation while the bridges persist for hours into interphase as observed by live cell imaging. Using ZBTB48 separation-of-function mutants, we profiled both proximity partners and protein interactions contributing to the chromatin bridge formation by BioID as well as locus-specific CasID coupled to label-free quantitative mass spectrometry. Our data encourage how ZBTB48-dependent recruitment of epigenetic remodelling complexes causes chromatin bridge formation as a driver for genomic instability unique to ALT-positive contexts. 1106

1107

P-III-0950 Deep proteomic characterization of more than 120 cell lines Johanna Wallner1, Alexander Betz1, Janine Sequeira1, Theresa Keil1, Jan Muntel1, Götz Hagemann1, Stephan Sieber2, Hannes Hahne1 1OmicScouts GmbH, Mass Spectrometry and Proteomics, Freising, Germany 2Technical University of Munich, School of Natural Sciences, Chair of Organic Chemistry II, Garching, Germany Biomedical research heavily relies on the indispensable utility of cell lines. Due to their ease of handling, widespread availability, and facilitation of straightforward investigations into diverse treatments, cell lines play a crucial role in exploring the mechanisms of action during drug development. To ensure the efficacy of such studies, a profound characterization of cell lines becomes imperative, aiding in the selection of the most appropriate cell line for a specific drug. In our comprehensive study, we conducted a deep proteomic characterization encompassing over 120 distinct cell lines originating from six different organisms. Employing a semi-automated workflow, all samples were processed with a KingFisher Apex robot and subsequently subjected to 2-hour data-independent acquisition (DIA) runs utilizing cutting-edge LC-MS/MS systems, including the Vanquish Neo UHPLC System, Orbitrap Eclipse Tribrid Mass Spectrometer, and Orbitrap Exploris480 Mass Spectrometer from Thermo Fisher Scientific. Additionally, 12 cell lines underwent an extensive characterization involving six gas-phase fractions (GPF). In human cell lines, single-shot DIA measurements yielded an average identification of more than 8,000 unique proteins. The implementation of GPF increased this number, resulting in the identification of over 10,000 unique protein groups. Mouse cell lines exhibited an average identification of 7,500 proteins, while rat cell lines displayed an identification of 8,100 proteins. A key focus of our analysis was the expression profiling of different kinase groups within the characterized dataset. Human cell lines consistently exhibited the unique identification of 300 to 436 kinases, with 140 to 220 falling under the category of E3 ubiquitin ligases. Additionally, we identified 40 to 78 proteins belonging to the dark kinome. This rich dataset empowers researchers to make informed decisions regarding the selection of optimal cell lines tailored to specific research inquiries. In summary, our extensive proteomic characterization of diverse cell lines provides a valuable resource for the scientific community, facilitating the identification of suitable cell lines for a myriad of research questions and enhancing the understanding of cellular responses to various treatments in the context of drug development. 1107

1108

P-III-0951 The essential roles of checkpoint kinase 1 in DNA damage checkpoint response Suyeon Kim1, Charlotte Hutchings1, Philip Zegerman2, Kathryn Lilley1 1University of Cambridge, Department of Biochemistry, Cambridge, United Kingdom 2Tropic Biosciences Ltd., Norwich, United Kingdom The DNA damage checkpoint response is a critical surveillance mechanism to ensure that the genome is accurately replicated before undergoing cell division. Defects in cell cycle checkpoints are a significant underlying cause of genomic instability, which drives the development of cancer. Checkpoint kinases are central in the response to oncogene activation, chemo- and radiotherapy-induced DNA damage and replicative stress. As a result, many inhibitors targeting this pathway have been developed, yet none are in clinical use due to adverse side effects in normal cell survival, particularly during cellular differentiation and haematopoiesis. The lack of clarity surrounding the checkpoint response poses a challenge in developing checkpoint inhibitors without dose- limiting toxicity. My research focuses on investigating the essential functions of Checkpoint kinase 1 (Chk1) in healthy non-cancer cells, both in the presence and absence of DNA damage, using proximity labelling and spatial proteomics techniques. By tagging Chk1 with APEX2, identification of Chk1 interactors in a native physiological setting of the human hTERT-immortalized Retinal Pigment Epithelial-1 (RPE-1) cell line is achieved. The spatial proteome map of the RPE-1 cell line established through Localisation of Organelle Proteins by Isotope Tagging after Differential ultraCentrifugation (LOPIT-DC), provides additional validation and information on Chk1 interactors, especially those within the nucleus. Additionally, it reveals the dynamics of protein re-localization in response to DNA damage, offering a valuable and comprehensive understanding of the checkpoint response. Comprehending the role of checkpoint kinases in genome maintenance and cell division will provide fundamental insights for developing innovative approaches to target enzymes of the DNA damage response pathways in cancer patients. 1108

1109

P-III-0952 Pharmacoproteomic profiling identifies secreted markers for aberrant drug action Sascha Knecht1, Mathias Kalxdorf2, Johanna Korbeń1, Toby Mathieson1, Daniel Sevin1, Bernhard Küster3, H. Christian Eberl1, Marcus Bantscheff1 1GSK, Functional and chemical proteomics & metabolomics, Heidelberg, Germany 2GSK, Genomic Technologies, Heidelberg, Germany 3Technical University of Munich, Chair of Proteomics and Bioanalytics, Munich, Germany Adverse drug reactions (ADRs) contribute significantly to late-stage attrition in drug discovery due to their unpredictability and enigmatic underlying mechanisms. Here we applied mass spectrometry-based proteomics to assess the effects of 46 drugs and 9 tool compounds with various levels of concerns for drug induced liver injury on the secretome of a hepatocyte model system. We observed distinct clusters of non-canonical secretion and intracellular thermal proteome profiling linked dysregulated mechanisms to extracellular markers. Functional follow-up confirmed lysosomal alterations by cationic-amphiphilic drugs, connected damage of the respiratory chain to Rab7-dependent secretion of mitochondrial proteins, and linked drug-induced endoplasmic reticulum stress to reduced basal secretion. Perturbation of sphingolipid biosynthesis pathways specifically induced secretion of the cargo sorting protein SDF4 whilst suppressing secretion of its cargo proteins. Thermal stability changes of clusters of membrane proteins in distinct subcellular compartments suggest local accumulation as important driver for unexpected drug effects through direct and indirect interactions. Fig. 1 (optional) 1109

1110

P-III-0953 RNF213 plays a role in NRF2 signaling by regulating the DNA damage response Aneesha Kohli1, Oliver Hartmann2, Markus E. Diefenbacher2, Katharina Imkeller3, Christian Münch1 1Goethe University Frankfurt, Molecular Systems Medicine, Frankfurt, Germany 2Ludwig-Maximilians-University Munich, Munich, Germany 3Goethe University Frankfurt, Frankfurt, Germany RNF213 is a unique AAA+ ATPase and E3 ligase protein encoded by RNF213 gene which is a susceptibility factor for the cerebrovasculopathy Moyamoya. Recent studies have suggested that RNF213 plays a role in infection and inflammation, lipid metabolism, angiogenesis and even cancer. However, a clear mechanistic insight into its basal function and how this is altered in stressed states has not been explored in depth. We use IFNγ stress as an endogenous inducer of RNF213 and to further study its role in interferon and infection-specific stressed states. Using an acute knockdown system coupled with global proteomics approaches, we identify a novel role for RNF213 in the regulation of NRF2 signaling. Co-immunoprecipitation coupled mass spectrometry analysis of RNF213"s interaction partners reveals critical regulators of the replication, ribosomal biogenesis, transcription and splicing machinery. Furthermore, phosphoproteomic analysis of these conditions shows increased phosphorylation of critical regulators of the DNA damage response pathway. These findings indicate towards a role of RNF213 in the regulation of NRF2 signaling potentially via an altered DNA damage response initiated by perturbations of the replication and translation machinery complexes and interaction dynamics. RNF213-NRF2 regulation appears to be particularly important for its downstream functions in angiogenesis, infection and inflammation. Furthermore, it may also contribute towards novel aspects of RNF213"s role in cancer and senescence. 1110

1111

P-III-0954 Acetylsalicylic acid effects on malignant melanocyte cellular proteome, secretome and extracellular vesicle proteome Helena Kupcova Skalnikova1, Jakub Cervenka1, Karolina Strnadova2, Lukas Lacina2, Karel Smetana2, Petr Prikryl3, Aleksi Sedo1 1First Faculty of Medicine, Charles University, Institute of Biochemistry and Experimental Oncology, Prague, Czech Republic 2First Faculty of Medicine, Charles University, Institute of Anatomy, Prague, Czech Republic 3First Faculty of Medicine, Charles University, Institute of Pathological Physiology, Prague, Czech Republic Acetylsalicylic acid (ASA), also known as Aspirin, is a commonly used drug known for more than a century. While its anti-inflammatory and anti-thrombotic effects have been well documented, several newer beneficial effects of ASA, including cardioprotective and tumour growth-inhibiting functions were demonstrated. Recently, several studies showed a lower risk of various cancers in patients regularly using ASA, however, the precise anti-cancer mechanisms of ASA are yet to be elucidated. Melanoma is a highly malignant skin cancer arising from transformed melanocytes. We and others have recently shown that malignant melanocytes produce various bioactive molecules, such as cytokines (e.g., IL-6), chemokines (such as CXCL1 and IL-8) and growth factors that modify the tumour microenvironment with impact on malignant cell growth and spreading. The aim of this study was the explanation of protein changes in malignant melanocytes initiated by ASA. Cells of the human melanoma G361 line were treated with 5 mM ASA. Cell lysates, conditioned medium and exosomes were analysed by multiplex immunoassay for levels of IL-2, IL-6, IL-8, IFNγ, CXCL1, TNFα, HGF, CCL2, SDF1α, and VEGF-A. In addition, global cellular and exosome proteomes were analysed by LC-MS/MS (timsTOF HT) using the dia-PASEF method, followed by data analysis in Spectronaut. ASA treatment led to a reduction of exosome secretion, as documented among others by the decreased amount of isolated exosomes and by the highly significant decrease of all 8 subunits of exocyst complex in exosomes. Moreover, ASA also reduced IL-8 secretion. On the other hand, IL-6 levels in cells, conditioned medium and exosomes were increased upon ASA treatment. The following experiments will focus on the function of selected molecules in ASA-induced anti-inflammatory and anti-cancer effects. Acknowledgement: This study was supported by the National Institute for Cancer Research (LX22NPO5102) financed by the European Union – Next Generation EU, by the project Center for Tumor Ecology (CZ.02.1.01/0.0/0.0/16_019/0000785), and Charles University Cooperatio Program, research area "Oncology and Haematology". 1111

1112

P-III-0955 Protein turnover during PINK1/Parkin-dependent Mitophagy measured by dynamic SILAC Bianca Lemke1, Katharina Zittlau1, Anna Lechado-Terradas2, Philipp J. Kahle2, Boris Macek1 1University of Tübingen, Quantitative Proteomics, Department of Biology, Institute for Cell Biology, Tuebingen, Germany 2University of Tübingen, Functional Neurogenetics, Laboratory of Neurodegeneration, Hertie Institute for Clinical Brain Research and German Center for Neurodegenerative Diseases, Tuebingen, Germany Mitochondria play a major role in the maintenance of cellular homeostasis, e. g. by synthesis of ATP through oxidative phosphorylation. Because of their importance, mitochondrial quality control is extremely important for the cell and damaged mitochondria must be quickly eliminated. One of the most studied processes for removal of damaged mitochondria is PINK1/Parkin-dependent mitophagy, where PINK1 recruits the cytosolic E3 ubiquitin ligase parkin to damaged mitochondria. Parkin then mediates the ubiquitylation of mitochondrial proteins, leading to their degradation. The regulation of parkin during later stages of mitophagy is yet to be determined. This process is linked to Parkinson"s disease (PD), as both, PINK1 and PARKIN genes are linked to recessive, early onset PD. Thus, a thorough insight on the regulation of the main participants can potentially lead to important novel therapeutic targets. Our recent work identified an outside-in progression of this process and ubiquitylation of parkin itself during late stages of mitophagy. In this study we measure protein turnover during PINK1/Parkin-dependent mitophagy using dynamic SILAC (stable isotope labelling with amino acids in cell culture). HeLa cells expressing WT parkin and/or ligase-dead parkin (C431A) were grown in DMEM medium supplemented with "light" amino acids (Lys0, Arg0). Upon induction of mitophagy by addition of the protonophore carbonyl cyanide m-chlorophenyl hydrazone (CCCP), the medium was exchanged to DMEM containing "heavy" isotopes (Lys8, Arg10). Cells were harvested at six time points ranging from 0 to 24 hours post-induction. Of about 3,500 protein groups identified, protein turnover was determined for about 1,800 protein groups. Upon CCCP treatment, turnover was overall decreased in both cell lines with a few exceptions. Interestingly, several proteins exhibiting an increased turnover differed between WT and C431A parkin cells. One protein that was found to be significantly upregulated during late stages of mitophagy and thus exhibits a faster turnover in the WT cells is ISG15, a small, ubiquitin-like protein. Its upregulation was further confirmed by western blotting as well as by analysis of previous data from our lab. ISG15 is known as a central player in antiviral immunity but its role is not limited to this process. Although conjugation of ISG15 to mitochondrial proteins has been reported, the role of this modification remains to be studied. We are currently investigating potential targets of ISGylation during mitophagy by further examination of data as well as co-immunoprecipitation assays. 1112

1113

P-III-0956 Towards deciphering the evolutionary process of endosymbiont – organelle transition of chromatophores in Paulinella chromatophora by MS-based spatial proteomics (LOPIT-DC) Thomas Lenz1, Lawrence Rudy Cadena2, Eva Nowack2, Gereon Poschmann1, Kai Stühler1 1Heinrich Heine University Düsseldorf, Proteome Research (MPL/BMFZ), Duesseldorf, Germany 2Heinrich Heine University Düsseldorf, Institute of Microbial Cell Biology, Duesseldorf, Germany Paulinella chromatophora stands out for its fascinating genetic, proteomic, and endosymbiotic characteristics.[1] Unlike most eukaryotes that acquired photosynthesis through primary endosymbiosis long (~1.6 billion years) ago, in which cyanobacteria were engulfed and evolved into plastids, Paulinella is the only known case in which this primary endosymbiotic event was repeated. This latest cyanobacterium uptake occurred much more recently (~0.09–0.14 billion years ago) and led to the formation of chromatophores, specialized photosynthetic organelles with a distinct genetic makeup, and provides a unique case of convergent evolution to study organellogenesis, i.e., the transition from endosymbiont to organelle. The transport mechanisms and protein targeting signals in chromatophores also differ from those in conventional plastids, underscoring their independent genetic evolution and functional specialization. Chromatophores in P. chromatophora exhibit reductive genome evolution, reducing their coding capacity to ~25% of their initial one (still being 5–10 times larger than a typical plastid genome) and necessitating the import of nucleus-encoded proteins for various metabolic pathways. This integration of chromatophore genes into the nuclear genome, alongside the loss of specific metabolic pathways within the chromatophore itself, indicates a complex genetic and proteomic relationship between the organelle and the host cell. Here, we performed mass spectrometry based proteomic profiling of P. chromatophora cellular fractions obtained by high-resolution separation of organelles and subcellular compartments employing the Localisation of Organelle Proteins by Isotope Tagging after Differential ultraCentrifugation (LOPIT-DC)[2] approach. Studying the spatial proteome of P. chromatophora provided valuable insights into the host - endosymbiont interplay and the evolutionary processes of organellogenesis. 1. Macorano L, Nowack ECM: Paulinella chromatophora. Curr Biol 2021, 31(17):R1024-R1026. 2. Geladaki A, Kočevar Britovšek N, Breckels LM, Smith TS, Vennard OL, Mulvey CM, Crook OM, Gatto L, Lilley KS: Combining LOPIT with differential ultracentrifugation for high-resolution spatial proteomics. Nature Communications 2019, 10(1):331. 1113

1114

P-III-0957 Investigating protein and organelle degradation selectivity by autophagy using targeted proteomics Alexandre Leytens1, Rocío Benitez Fernandez2, Carlos Jiménez-Garciá2, Carole Roubaty1, Michael Stumpe1, Patricia Boya2, Jörn Dengjel1 1University of Fribourg, Biology department, Fribourg, Switzerland 2University of Fribourg, Department of Neuroscience and Movement science, Fribourg, Switzerland Autophagy is a crucial eukaryotic degradation pathway which is known to be perturbed in a wide range of diseases. This pathway involves the growing of a newly formed membrane-enclosed vesicle inside the cell, capturing parts of the cytoplasm and transporting them to the lysosome for hydrolysis. Autophagy provides the cell with the means to degrade large components such as damaged organelles or protein aggregates that cannot be degraded by the proteasome. Although it has long been assumed that autophagy is a bulk process capturing its cargo in a random manner, research from the last decade has uncovered numerous specialized selective autophagy pathways, targeting specific cargoes under certain stimuli for degradation. Accurate quantification of these pathway markers is crucial as it can be used as a proxy for the degradation of autophagy targets. Unfortunately, some markers are low-abundant proteins making them difficult to detect and to quantify in standard bottom-up proteomics approaches. Common methods to investigate these processes generally rely on the quantification of single protein markers often failing to address the complexity and crosstalk of these selective mechanisms. Additionally, these methods often rely on the (over-)expression of fluorescent proteins making them unpractical with certain sample types. Here, we introduce a targeted proteomics approach quantifying most characterized selective autophagy receptors and additional autophagy-relevant proteins as a tool to investigate autophagy selectivity. We compared our proteomic approach with state-of-the-art fluorescence-based approaches and reproduced respective results while also capturing more complex dynamics of organelle degradation. We next screened a panel of Melanoma-derived cell lines using this approach combined with data-independent expression proteomics to investigate how different cancers remodel their proteome in response to stress by selective autophagic activity. 1114

1115

P-III-0958 Secretome analysis deciphers microglia-derived inflammatory signals and identifies an inducer of neuronal deficit Yi Li1, Hanqing Zhang2, Xingchao Yang2, Hui Ye2 1China Pharmaceutical University, School of Pharmacy, Nanjing, China 2China Pharmaceutical University, Jiangsu Provincial Key Laboratory of Drug Metabolism and Pharmacokinetics, Nanjing, China Microglia are dominant immune cells in central nervous system (CNS). Upon inflammatory stimulation, they secrete rich signaling proteins including classic cytokines to interfere with neighbouring neurons. Such deleterious signals have been linked to multiple neurodegenrative diseases. Deciphering the signals released from activated microglia thus holds promise in unveiling underlying mechanisms for neurodegenrative diseases. To profile the microglia-derived signals, we first used secretome analysis to characterize the secretome of microglia cell line and primary murine microglia after Lipopolysaccharide (LPS) stimulation. The detected secretome repertoire consists of classic cytokines as well as proteins not previously associated with microglia activation, among which the top responder is a proteoglycan (PG4). Through biochemical manipulation with brefeldin A, we demonstrated that PG4 is secreted from microglia via the conventional secretory pathway Moreover, we confirmed that the secretion of PG4 also occurs in vivo by imunofluoresence, RT-qPCR analysis of mouse brain and ELISA analysis of LPS by intraperitoneal (i.p.) and intracerebroventricular (i.c.v.) injection. Notably, clinical big data analysis of Alzheimer's patients also showed upregulation of PG4 in CSF and human autopsy brain tissue, suggesting its clinical relevance. treated with from mice cerebrospinal fluid (CSF) These findings propels us to explore the potential mechanistic association of PG4 with neurodegenerative diseases such as AD. We found that, upon administering neurons with exogenous recombinant protein PG4, neurite outgrowth was inhibited. We then used Target-Responsive Accessibility Profling (TRAP) approach, a target identification approach previously developed in our lab, to elucidate how PG4 regulates neuronal degeneration. We tentatively identified a neuronal membrane glycoprotein as a dose-responsive target of PG4. Knockdown of this target attenuated PG4-inducted axonal loss in cortical neurons. Together, using secretome, chemoproteomics and molecular biology, we discover a signaling protein from activated microglia that exerts neurotoxicity. This finding may reveal novel pathological mechanisms of neuroinflammation-related CNS diseases and inspire therapeutic interventions. 1115

1116

P-III-0959 Testing the robustness of a BONCAT strategy for detection of rapamycin-induced small changes in protein translation after short-term treatment of HEK293 cells Patricia Luckner1, Aysa Yelboga1, Hanna Wimmer1, Judit Vonneguth1, Eduard Hochmuth1, Astrid Bruckmann1 1University of Regensburg, Regensburg, Germany BONCAT (Bioorthogonal non-canonical amino acid tagging) is widely used for identification of nascent proteins. We aimed to test the robustness of this technique for the investigation of rapamycin-induced changes in protein translation after different incubation times. Initial experiments for enrichment of Azidohomoalanine-tagged proteins showed that in our hands clicking of newly synthesized proteins (NSPs) directly to alkyne agarose outperformed enrichment via alkyne biotin derivatives. We then chose to treat HEK293 cells for six hours with rapamycin, inhibitor of mTOR (Serine/threonine-protein kinase mTOR), as proof of concept for our protocol to specifically enrich for NSPs. After clicking of NSPs to the alkyne agarose, we performed an on-bead tryptic digest and peptides were analyzed by LC-MSMS. MaxQuant analysis resulted in detection of >2600 proteins in total, 451 and 519 proteins uniquely identified in the untreated and treated sample, respectively. As expected, GO Enrichment analysis (PANTHER, ShinyGO) of differentially newly synthesized proteins (|log2(FC)|≥1) could link the results specifically to rapamycin treatment since major known cellular processes are affected, such as ribosome biogenesis and translation as well as splicing. Next, HEK293 cells were treated again with rapamycin but this time for a shorter period of only two hours. With these results we want to demonstrate the suitability of our BONCAT protocol to robustly identify even small changes in protein translation. 1116

1117

P-III-0960 Quantitative proteomic analysis of Carboplatin resistance in 2D and 3D cultured high-grade serous ovarian cancer cells Jimmy Maillard1, Theo Roumeliotis1, Ekta Paranjape1, Lisa Pickard1, Jyoti Choudhary1, Udai Banerji1 1Institute Cancer Research, Cancer therapeutics, Sutton, United Kingdom Drug discovery and development has historically been performed on cells cultured as flat monolayers that do not mimic an in-vivo environment. Thus, there is currently a growing interest in 3D cell culture systems for drug screening and discovery that are more likely to provide physiologically reliable information. Here, we present the first quantitative profiling of carboplatin-induced proteome comparing 2D and 3D models of High-grade Serous Ovarian Cancer (HGSOC) cells. With over 300,"000 cases and approximately 200,"000 deaths per year, ovarian cancer is the leading cause of death from gynaecological malignancies and high-grade serous ovarian cancer (HGSOC) is the most common form of ovarian cancer. We investigated the carboplatin effects on the proteome of two2 sets of commercially available and biologically- related cell lines with different sensitivity to carboplatin: UWB1/UWB1-BRCA and PEO1/PEO4 that were isolated from the same patient before and after developing chemoresistance. Cells were cultured as 3D spheroids in a basement membrane (Cultrex) and as 2D monolayers for comparison, half the replicateswith or without were exposed exposure to sublethal micromolar concentrations IC50) of carboplatin. Liquid chromatography coupled to tandem mass spectrometry (data-dependent acquisition mode) enabled protein quantitation based on a bottom-up approach relying on labelling of trypsin-digested proteins with isobaric mass tags (TMT) for direct comparison across samples and replicates. Isobaric labelling-based LC-MS analysis with Real Time Search MS3 acquisition enabled the relative quantification of We identified more thanover 6500 proteins across each all samples and achieveddemonstrating deep proteome coverage with high accuracy and precision. Using a 3-way ANOVA statistical analysis, we further identified more than 900 differentially regulated proteins differentially expressed (-log10 ANOVA value > 1.30) and were able to compare proteinwith variations in among the experimental factors groupings based on : 1)encompassing carboplatin response, culture dimension and carboplatin treatment. resistant vs sensitive phenotype 2) 2D vs 3D and 3) treated vs non-treated criteria. (twice the We found important significant changes differences in proteins expression levels involved inof proteins involved in DNA damage responses as well as in DNA packing (nucleosome) and regulation. H1-4 histone linker and isoforms (H1-3, H1-5) were downregulated (p value ≤ 0.05) in sensitive and treated 3D models, while being upregulated (p value ≤ 0.05) in resistant and treated 3D models;, while opposite results were found in 2D models (p value ≤ 0.05). Overall, our work highlights identifies major differences among 2D and 3D cultured samples and highlights the need testifies of the crucialness of implementing near patient ovarian cancer models that are closer to tumour structure characteristics for drug discovery. 1117

1118

P-III-0961 Proteome signatures of the aging retina reveal sexual dimorphism in mice with targeted Ephx2 gene knockout Caroline Manicam1, Elahe Zare1, Norbert Pfeiffer1, Natarajan Perumal1 1University Medical Center of the Johannes Gutenberg-University Mainz, Department of Ophthalmology, Mainz, Germany Aging is a pivotal driving force behind various degenerative processes that culminates in functional impairment and stochastic molecular changes, which predispose to chronic disease conditions. In the eye, progressive deterioration of the ocular neurovascular function is exacerbated by aging, particularly in neurodegenerative diseases such as glaucoma, which is the second cause of vision loss worldwide. To add to the complexity, there is gender predilection that is not clearly demarcated in retinopathies. We recently deciphered the neuro- vasculoprotective roles of bioactive lipids following inhibition of soluble epoxide hydrolase (sEH/ Ephx2) in the CYP pathway in male mice, alluding to a potential druggable target to mitigate the impact of aging on visual dysfunction. This study tested the hypothesis that there are distinct gendered differences in the proteome of aged retina of the Ephx2 gene knockout (KO) mice. Retinae were isolated from 160 young (3-5 months) and old (12-24 months) wild type (WT) and KO mice of both sexes. Samples were pooled (n=5 mice/group/replicate) with 4 biological replicates per group and subjected to label-free mass spectrometry-based proteomics using a Thermo Q Exactive Plus Hybrid Quadrupole-Orbitrap MS coupled to an EASY-nLC 1000 uHPLC system. The MS data analysis were followed by bioinformatics analysis with Ingenuity Pathway Analysis (IPA) to unravel protein functionalities and corresponding upstream regulators that govern the protein expression. We identified a total of 3706 retinal proteins with at least 2 unique peptides and FDR < 1 %. A significant (p < 0.05) number of differentially expressed protein clusters in the aging retina were implicated in the eIF2 and sirtuin signaling pathways in both sexes. Intriguingly, proteins related to neurodegenerative processes (p=9.9×10-4) that cause retinal diseases (p=9.6×10-4) and oxidative stress (p=2.6×10-4) were significantly regulated only in aged WT female. However, KO resulted in significant activation of cell-cell contact (p=1.2×10-3) and neuritogenesis (p=5.4×10-5) in both genders. The female sex showed striking resilience against ferroptosis, while the male counterparts demonstrated highly significant activation of this signaling pathway in both WT (z- score: 1.3) and KO mice (z-score: 2). Likewise, XBP1, an upstream transcription regulator with ER stress- ameliorating function, was found in only male retinae. Remarkably, this regulator was found to be significantly activated in the WT mice (p=3.9×10-4, z-score: 0.5) but inhibited in the KO male mice (p=8.1×10-3, z-score: -1.2). Taken together, our study characterized for the first time the impact of sex differences in the retinal proteome of KO mice, which is a key determinant in understanding the pathomechanisms underlying age-related retinal diseases. Our findings will be instrumental for development of personalized intervention targeting sEH for retinopathies that present a predilection for a particular sex. 1118

1119

P-III-0962 Identification of drug target engagement of repurposing candidates for polycystic kidney disease by thermal proteome profiling Alina Meyer1, Paraskevi Sotiropoulou2, André Mateus3, Daniel Dietrich2, Per Artursson1 1Uppsala University, Department of Pharmacy, Uppsala, Sweden 2University of Konstanz, Department of Biology, Konstanz, Germany 3Umeå University, Department of Chemistry, Umeå, Sweden Autosomal dominant polycystic kidney disease (ADPKD) is a chronic, progressive hereditary disease characterized primarily by fluid-filled cysts in the kidneys and caused by mutations in the polycystin-1 or -2 gene. The treatment is currently limited to a single approved drug worldwide. In the search for additional drugs, drug repurposing provides an alternative. With this approach, clinically validated drugs with established pharmacokinetic and safety profiles are used for new indications. Here, 2D-thermal proteome profiling (2DTPP) was used to validate known and identify novel targets of ADPKD repurposing candidates. The repurposing candidates birinapant, bardoxolone methyl and salicylic acid were selected based on their previously demonstrated attenuation of cyst growth in vitro, in clinical trials and in preclinical studies, respectively. To mimic in vivo conditions, whole cell experiments were performed under physiological oxygen tension in the renal proximal tubular epithelial cell line RPTEC/TERT1. Our results confirmed the expected modes of action (MoA) for each of these three repurposing candidates. We also identified shared effects on mRNA translation, monocarboxylate transport, and mitochondrial nucleotide transport and energy metabolism. In addition, we discovered focal adhesion kinase, an activator of the proliferative PI3K/Akt/mTOR pathway, as a novel potential target that may contribute to the anti-proliferative effect of salicylic acid. Furthermore, we observed thermal destabilization of the glucosidase GANAB, an ADPKD-causing gene involved in polycystin maturation and trafficking, in response to salicylic acid treatment, highlighting the broad range of mechanisms that salicylic acid exerts on cyst growth. Finally, we corrected the observed dose-dependent changes in thermal stability with measured unbound drug concentrations at the target site, defined as the intracellular drug bioavailability. Overall, our findings highlight the utility of 2DTPP for evaluation of drug target engagement of repurposing candidates, enabling complex MoA validation and target discovery. Salicylic acid emerges as a particularly promising repurposing candidate for ADPKD due to its anti-proliferative effects and favorable safety profile. Fig. 1 (optional) 1119

1120

P-III-0963 Relative quantification of (phospho)proteoforms reveals important common disease pathways in Retinitis pigmentosa mouse models Ahmed Montaser1, Katri Vainionpää1, Henri Leinonen1 1University of Eastern Finland, School of Pharmacy, Kuopio, Finland Introduction: Inherited retinal degenerative diseases (IRDs) affect approximately 1 in 2000 worldwide, leading to significant blindness among young and working-age people due to a lack of effective treatments 1,2. Retinitis pigmentosa (RP) is one major subtype of IRD arises from over 150 identified mutations 3. Targeting each specific mutation with gene therapy is unfeasible, so broader disease-modifying treatments offer a potential strategy to halt disease progression. The aim of the study is to find common drug targets in highly translationally relevant mouse models of RP such as Pde6βRd10 (rd10) and RhoP23H/WT (P23H). Results: Using LC-MS and label-free protein quantification 4,5, over 7000 proteins were quantified, with 600 common differentially expressed in rd10 and P23H models, implicating approximately 90 KEGG pathways. Notably, nucleotide metabolism and cAMP signaling pathways emerge as significant drug targets alongside the well-known cGMP pathway in retinal degenerative diseases 6. Moreover, in the rd10 mouse model, over 26,000 phosphorylation sites were identified and quantified. Following the normalization of phosphorylation levels to protein abundance, the sites were refined to feature those with a ratio exceeding 1.5 or falling below 0.5. Dephosphorylation was predominant, notably in proteins that regulate phototransduction and cGMP binding including GRK1, RHO, PDC, SLC24A1 & TULP1, as well as in those that play a role in nucleotide metabolism such as DNM3, HSPA1A, SERBP1, and HSP90AB1. Conclusion: Although the rod photoreceptors die via distinct pathways in the phosphodiesterase (rd10 model) and rhodopsin (P23H model) associated RP models, the downstream proteomic changes are largely convergent. Targeting some of these shared downstream pathways in RD could offer a unified treatment approach for retinal degenerative diseases caused by idiopathic factors or multiple genetic mutations, such as RP. References 1)doi.org/10.1016/j.preteyeres.2010.03.004 2)doi.org/10.1080/13816810.2021.1913610 3)doi.org/10.1016/j.preteyeres.2017.10.00 4)doi.org/10.1074/mcp.M113.031591 5)doi.org/10.1038/s41592-019-0638-x 6)doi.org/10.1073/pnas.1718792115 1120

1122

P-III-0964 Exploring the molecular factors governing the unconventional protein secretory pathway using quantitative proteomics Cristian Munteanu1, Gabriela Chiritoiu2, Gheorghita Isvoranu3, Stefana Petrescu2, Marioara Chiritoiu2 1Institute of Biochemistry of the Romanian Academy (IBRA), Bioinformatics and Structural Biochemistry (DBSB), Bucharest, Romania 2Institute of Biochemistry of the Romanian Academy (IBRA), Molecular Cell Biology, Bucharest, Romania 3"Victor Babeș” National Institute of Pathology, Department of Animal Husbandry, Bucharest, Romania To ensure tissue homeostasis, eukaryotic cells secrete factors involved in various biological processes such as defense against foreign pathogens, extracellular matrix remodeling, damaged cells etc. While most proteins are exported to the extracellular space via the classical secretory pathway (ER-Golgi route), an increasing category of proteins are being released independent of this pathway, via the unconventional secretory pathway (UPS). UPS cargoes are proteins with apparently unrelated functions and include growth cytokines (IL-1β, IL-1α, IL-18), neuropeptides (ACBP), redox enzymes (SOD1), factors (FGF2), ion channels (CFTR) and in general, few common factors have been identified between the molecular processes involved in their release. GRASP55 (Golgi- reassembling stacking proteins 55), has been linked with the UPS of several such cargoes and alterations in intracellular GRASP55 level results in perturbed UPS. Amongst these, the secretion of the pro-inflammatory cytokine IL-1β was shown to be GRASP55-dependent from primary macrophages under physiologically relevant conditions. We used a GRASP55 knock-out (G55-/-) mouse strain to underpin the molecular mechanisms involved in unconventional protein secretion from primary Mouse Embryonic Fibroblasts (MEFs) subjected to tunicamycin treatment, an N-glycosylation inhibitor which results in endoplasmic reticulum (ER) stress. We undertook a quantitative proteomics approach to obtain a global picture of the proteome relative changes from both WT and G55-/- MEF cells by combining offline high pH peptide fractionation with LC-MS/MS analysis to discover novel factors participating in the unconventional secretory pathway. We characterized the proteome of the primary cell cultures up to a depth of over 5000 proteins. Ablation of GRASP55 resulted in alterations of molecular factors involved in protein degradation, including the modification of the protein expression level of some ER and Golgi resident factors such as Emc4 or Lman2, involved in the sorting of the glycoproteins from the early secretory pathway. Moreover, we observed that during ER stress treatment BiP, Dnajc3 and the molecular chaperon Hsp90b1 are up-regulated compared with the control cells for both the WT and G55-/- primary cell lines. In general, tunicamycin treatment resulted in the modification of the relative expression level of proteins from the secretory pathway involved in various cell to cell communication signaling. Our results provide a detailed image of the GRASP55 dependent molecular effectors from the unconventional secretory pathway, with a potential-role of molecular hubs in the GRASP55-dependent secretome. 1122

1123

P-III-0965 Interplay between photosynthetic electron flux and organic carbon sinks in sucrose-excreting Synechocystis sp. PCC6803 revealed by omics approaches Dorota Muth-Pawlak1, Lauri Kakko1, Pauli Kallio1, Eva-Mari Aro1 1University of Turku, Department of Life Technologies, Turku, Finland Advancing the engineering of photosynthesis-based prokaryotic cell factories is important for sustainable chemical production and requires a deep understanding of the interplay between bioenergetic and metabolic pathways. Rearrangements in photosynthetic electron flow to increase the efficient use of the light energy for carbon fixation must be balanced with a strong carbon sink to avoid photoinhibition. In the cyanobacterium Synechocystis sp. PCC 6803, the flavodiiron protein Flv3 functions as an alternative electron acceptor of photosystem I and represents an interesting engineering target for reorganizing electron flow in attempts to enhance photosynthetic CO2 fixation and increase production yield. We have shown that inactivation of Flv3 in engineered sucrose-excreting Synechocystis (S02:Δflv3) induces a transition from photoautotrophic sucrose production to mixotrophic growth sustained by sucrose re-uptake and the formation of intracellular carbon sinks such as glycogen and polyhydroxybutyrate. The growth of S02:Δflv3 exceeds that of the sucrose-producing strain (S02) and demonstrates unforeseen proteomic and metabolomic changes over the course of the nine-day cultivation. To determine the rearrangement of the metabolic pathways important for energy source – carbon sick balance, the proteomes from whole cell lysates were analyzed with LC-MS/MS DIA technology, and the accumulation of selected metabolites were assessed in both S02 and S02:Δflv3. In the absence of Flv3, a down-regulation of proteins related to photosynthetic light reactions and CO2 assimilation occurred concomitantly with up-regulation of those related to glycolytic pathways, before any differences in sucrose production between S02 and S02:Δflv3 strains were observed. Over time, increased sucrose degradation in S02:Δflv3 led to the upregulation of respiratory pathway components, such as the plastoquinone reductase complexes NDH-11 and NDH-2 and the terminal respiratory oxidases Cyd and Cox, which transfer electrons to O2. While glycolytic metabolism is significantly up-regulated in S02:Δflv3 to provide energy for the cell, the accumulation of intracellular storage compounds and the increase in respiration serve as indirect sinks for photosynthetic electrons. Our results show that the presence of a strong carbon sink in the engineered sucrose-producing Synechocystis S02 strain, operating under high light, high CO2 and salt stress, cannot compensate for the lack of Flv3 by directly balancing the light transducing source and carbon fixing sink reactions. Instead, the cells immediately sense the imbalance, leading to extensive reprogramming of cellular bioenergetic, metabolic and ion transport pathways that favor mixotrophic growth rather than enhancing photoautotrophic sucrose production. 1123

1124

P-III-0966 Mass spectrometry-based proteomics to explore intratumoral heterogeneity in PAX3::FOXO1 fusion positive rhabdomyosarcoma Satoshi Nakano1,2, Konrad Kurowski2, Ebrahem Hamed3, Hizuru Amano1, Akinari Hinoki1, Peter Bronsert2, Hiroo Uchida1, Oliver Schilling2, Simone Hettmer3,4 1Nagoya University Graduate School of Medicine, Department of Pediatric Surgery, Nagoya, Japan 2University of Freiburg, Institute of Surgical Pathology, Freiburg, Germany 3University of Freiburg, Division of Pediatric Hematology and Oncology, Freiburg, Germany 4University Medicine Halle, Department for Surgical and Conservative Pediatric and Adolescent Medicine, Halle (Saale), Germany Introduction: Rhabdomyosarcoma (RMS) is the most common soft tissue tumor in childhood and adolescence. In PAX3::FOXO1 (P3F)-positive RMS, cell-to cell heterogeneity in fusion oncoprotein expression appears to play a crucial role in tumor progression and facilitate drug resistance. This study aims to elucidate intratumoral heterogeneity via Mass spectrometry (MS)-based proteomics. Methods: 41 xenograft tumors – established from 6 different, low-passage patient-derived cell cultures - were harvested. Two peripheral tissue cores and one central core sample were collected from each tumor for a total of 123 samples. These samples were lysed with 1% SDS, followed by double digestion with Trypsin and Lys-C via the single-pot, solid-phase-enhanced sample-preparation (Sp3) protocol. MS-based proteomics were processed using Evosep One chromatography and timsTOF flex with long gradient and data independent acquisition (DIA)- Parallel Accumulation Serial Fragmentation (PASEF). Data analysis was performed using DIA-NN and R software. Results: Overall, 7235 unique human and murine proteins were detected. Following sparsity reduction, 4339 human (tumor) proteins and 1365 murine (stromal) proteins were further analyzed. The proteomic profiles of peripheral and core samples were similar. As expected, no differences were detected among murine proteins. For human (tumoral) proteins, the proteins belonging to plasma membrane and coated membrane pathways were significantly upregulated in peripheral regions, whereas the chromosome and centromeric region pathways were enriched in core regions. Conclusions: MS-based proteomics was used to investigate intratumoral heterogeneity of protein expression in P3F positive RMS. Mixed-species proteomics of a PDX model provided new insight into the tumor stroma components. Our future studies will validate differences in protein expression and explore the biological impact of such differences. 1124

1126

P-III-0967 Proteomic and biochemical investigation of the effects of ERAP1 functional disruption in melanoma cells Martha Spyridoula Nikopaschou1,2, Martina Samiotaki3, Ellie Stylianaki4, Vassilis Aidinis4, Angeliki Chroni1, George Panayotou3, Efstratios Stratikos1,2 1National Centre for Scientific Research Demokritos, Athens, Greece 2National & Kapodistrean University of Athens, Department of Chemistry, Athens, Greece 3Biomedical Sciences Research Center "Alexander Fleming", Institute for Bioinnovation, Vari, Greece 4Biomedical Sciences Research Center "Alexander Fleming", Institute for Fundamental Biomedical Research, Vari, Greece Endoplasmic reticulum aminopeptidase 1 (ERAP1) is located in the endoplasmic reticulum (ER) and plays a crucial role in shaping adaptive immune responses. Its main function involves the removal of N-terminal amino acids from antigenic peptide precursors, aiding in the optimal presentation of these peptides by Major Histocompatibility Complex I (MHC I) molecules. While ERAP1's activity is necessary for generating many antigenic peptides, excessive trimming of some peptides has also been observed. In the case of cancer-specific epitopes, over-trimming could potentially lead to diminished cytotoxic T-cell responses and cancer evasion. This pivotal role in immune evasion, coupled with observations that inhibiting ERAP1 alters the immunopeptidome of cells, has highlighted ERAP1 as a promising target for cancer immunotherapy. Besides its role in antigen processing and presentation, ERAP1 has also been suggested to affect ER stress and to participate in innate immunity. The potential multi-faceted role of ERAP1 necessitates an increased understanding of the cellular processes that may be affected by ERAP1 inhibition before its establishment as a pharmacological target. To investigate the effects of ERAP1 inhibition on cellular processes, we employed a data-independent acquisition (DIA) approach, followed by library-free search, to examine alterations in expressed proteins following genetic manipulation (ERAP1 knockout) or allosteric inhibition of ERAP1 in A375 melanoma cells. Proteomic analysis revealed the identification of 5575 proteins in A375 cells. Disruption of ERAP1 led to significant shifts in the proteome, with 1375 proteins showing differential expression in inhibitor-treated and knockout cells compared to wild-type cells (FDR= 0.05, S0=0.5). Interestingly, inhibitor treatment induced less pronounced proteome changes, with most of these changes also observed in knockout cells. Investigation of the differentially expressed proteins indicated alterations in cellular processes such as antigen processing and presentation, cellular response to stress and the metabolic pathways of glycolysis and oxidative phosphorylation. Western blot analysis confirmed the upregulation of protein disulfide isomerase A3, which is part of the peptide loading complex that participates in MHC I mediated antigen presentation. Moreover, DCF assay suggested a small alteration in the levels of reactive oxygen species after ERAP1 functional inhibition, while extracellular flux analysis suggested limited effects on mitochondrial metabolism. Our results suggest that the reduction of ERAP1-trimming activity in cancer cells may affect peptide homeostasis in the ER and lead to limited compensatory changes in multiple cellular pathways. 1126

1128

P-III-0968 Systematic identification of degrader drug targets by deep proteomic screening and high-sensitivity ubiquitinomics Uli Ohmayer1, Martin Steger1, Björn Schwalb1, Bachuki Shashikadze1, Jutta Fritz1, Gisele Nishiguchi2, Kevin McGowan2, Zoran Rankovic2,3, Henrik Daub1 1NEOsphere Biotechnologies GmbH, Mass Spectrometry, Planegg, Germany 2St. Jude Children's Research Hospital, Department of Chemical Biology & Therapeutics, Memphis, TN, United States 3The Institute of Cancer Research, Centre of Protein Degradation, London, United Kingdom Targeted Protein Degradation (TPD) is an emerging therapeutic modality in which disease-causing proteins are selectively degraded by endogenous E3 ubiquitin ligases through interactions with small molecules, such as molecular glues or Proteolysis Targeting Chimeras (PROTACs). One bottleneck in the discovery of molecular glue degrader drugs is the lack of rational drug design strategies and therefore the absence of a well-defined target space. To systematically screen such degrader drugs in a target-agnostic manner, we have developed deep proteomic screening, an MS-based screening technology with high throughput capabilities. This technology allows us to screen extensive libraries of (potential) molecular glue degrader drugs against a substantial portion of the expressed cellular proteome, generating comprehensive data matrices containing robust quantification information for over 10,000 proteins. As proof of concept, we screened a collection of 100 molecular glue compounds, identifying numerous previously uncharacterized degrader target candidates. Following screening, degrader drug target candidates (neosubstrates) are validated by demonstrating modification by ubiquitin in cells. For this, we have developed a highly sensitive one-pot ubiquitinomics workflow that is coupled to slice-PASEF mass spectrometry. Our screening platform thus connects neosubstrate ubiquitination to protein degradation, thereby facilitating the identification of degrader targets with high confidence. 1128

1129

P-III-0969 Elucidating protease substrate specificities with enhanced secretome profiling techniques Gözde Ozcelik1,2, Karsten Nalbach1,2, Sarah Tschirner1,2, Anna Berghofer1,2, Stephan Müller1,2, Stefan Lichtenthaler1,2 1German Center for Neurodegenerative Diseases (DZNE), Neuroproteomics, Munich, Germany 2Technical University of Munich, School of Medicine and Health,Klinikum rechts der Isar, Munich, Germany The secretome is crucial for many biological processes, including development and neurodegeneration. Mass spectrometry-based proteomics offers profound insights into the cellular secretome, highlighting quantitative changes in disease contexts. To enhance secretome analysis, our group implemented the High-performance Secretome Protein Enrichment With Click Sugars (hiSPECS) method, incorporating lectin-based glycoprotein enrichment, direct azide glycoprotein coupling to alkyne beads, and on-bead proteolysis (Tushaus et al., 2020). Now, we further improved sensitivity of our method through the use of advanced magnetic beads and automated washing steps. Proteolytic cleavage of membrane proteins, known as shedding, releases ectodomains that become integral components of the secretome. Therefore, secretome analytics is ideally suited to identify physiologically relevant substrates of membrane proteases like ADAM10. ADAM10, a widely expressed transmembrane protein, functions as a "molecular scissor," mediating ectodomain shedding by cleaving extracellular domains from membrane protein substrates. With over 100 substrates, including Notch, amyloid precursor protein, cadherins, and growth factors, ADAM10 is crucial in both health and disease, particularly cancer and Alzheimer's. By understanding the substrates and activity of ADAM10, we can gain insights into its diverse functions and contributions to disease mechanisms. Using our improved secretome analysis, we explored the substrate specificities of ADAM10 in various cell types. Our findings reveal new dimensions of ADAM10's role in cellular processes, highlighting its potential as a therapeutic target and providing insights into the molecular mechanism of related diseases. 1129

1130

P-III-0971 Analysis of keratin post-translational modifications during cell division Nurhan Özlü1,2, Busra Harmanda1, Ceyda Seren Ceyhan1, Büşra Aytül Akarlar1, Venkatesha Basrur2, Alexey Nesvizhskii2 1Koç University, Molecular Biology and Genetics, Istanbul, Turkey 2University of Michigan, Department of Pathology, Ann Arbor, MI, United States Cell division is one of the most basic cellular processes which is required for the propagation of life. Any mistakes during cell division may lead to cell death, developmental disorders, abortion, or complex diseases like cancer. Mitotic kinases are the main drivers of mitosis and cytokinesis. It is estimated that 30% of the mitotic proteome is phosphorylated. Keratins are the members of intermediate filaments that are functionally regulated by post- translational modifications (PTM) including phosphorylation, glycosylation, sumoylation and acetylation. Keratins form highly stable polymers in epithelial cells that need to be cut through during cell division. However, it is not known how PTMs of Keratins are regulated and how the spatial organizations of Keratins are controlled during cell division. In this study, we profiled the cell cycle-dependent phosphorylation sites on Keratins in Hela cells. We used TiO2- based phosphopeptide enrichment to isolate the phosphopeptides from mitosis, cytokinesis, and interphase cell lysates, and then analyzed them using HF Q Exactive Mass spectrometry. Our analysis revealed multiple phosphorylation sites specific to mitosis and cytokinesis, particularly in simple epithelial keratins like K8 and K18. We conducted in vitro kinase assays and identified eleven phosphorylation sites on Keratin 8 that are dependent on Aurora B kinase. One of these, Keratin 8-Ser 34, is specifically phosphorylated at the cleavage furrow and remains present at the midzone until the end of cytokinesis. Inhibiting Aurora B kinase or using a non- phosphorylatable Keratin 8 mutant resulted in residual keratin bundles at the cleavage furrow and a failure in furrow ingression. Next, we systematically examined O-glycosylation which can compete with phosphorylation for the same residue on Keratins. For this, glycosylated peptides from mitotic and interphase lysates of Hela cells are enriched with phenyl-boronic acid-based resins. The samples were run in Orbitrap Ascend Tribrid Mass Spectrometry with EThcd based fragmentation and analyzed in MSFragger-Glyco in Fragpipe. We identified previously reported and also novel O-GlcNAc sites on Keratins, some of which are regulated in a cell cycle-dependent manner. Both co- regulated and counter-regulated glycosylation and phosphorylation sites are present when interphase and mitosis cell cycle stages are compared. Molecular analysis of these dynamic sites will provide insight into PTM crosstalks during cell division. This study offers new insights into the remodeling of Keratins through post- translational modifications in the cell division of epithelial lineages. 1130

1131

P-III-0972 The specificity of macrophage's proteome: a comparison with HeLa cell line Mariana Pereira Massafera1, Francielle Aguiar Gomes1, Graziella Eliza Ronsein1 1University of São Paulo, Institute of Chemistry, São Paulo, Brazil Introduction: Sample preparation is a key step of the proteomics workflow, determining the overall reproducibility and robustness of the analysis. Although previous studies have assessed the performance of different sample preparation protocols, the majority of them focused on HeLa cell line as model. Also, there is a scarcity of proteomic optimization workflows tailored for macrophages – a model cell line for studies involving the immune system, inflammation, atherosclerosis and obesity. Methods: THP-1 monocytes were cultured as usual, then differentiated to macrophages with 200 nM phorbol myristate (PMA), followed by incubation at 37 °C and 5% CO2 for three days. Adherent macrophages were lysed with SDC buffer (100 mM ammonium bicarbonate, 1% SDC and cOmplete protease inhibitor cocktail). The resulting proteins were quantified and digested following in-solution, in-gel or Filter-Aided Sample Preparation (FASP) protocols, then analyzed by nanoLC-MS/MS (ThermoScientific EASY-nLC 1200 coupled to an Orbitrap Fusion Lumos mass spectrometer) in a data-dependent acquisition mode, in 104 min runs. HeLa protein digest standard was acquired from ThermoFisher and analyzed by nanoLC-MS/MS under the same conditions as for the macrophage digests. For both sample types, we analyzed at least three replicates. Raw files were processed in MaxQuant then analyzed using Perseus, String, and WebGestalt. Results: Key differences in protein composition between HeLa and differentiated THP-1 (macrophages) cell lines can be visualized (Figure 1). HeLa cell"s proteome shows enrichment in biological processes related to DNA replication, telomerase activity and transcription. The most enriched proteins are related to the nucleus cellular compartment and biological processes – all characteristic of its origin as a cervical cancer cell line. On the other hand, an analysis of macrophages" proteome reveals their role in immunity and infection, with enrichment in proteins associated with cellular responses to bacterial infection or inflammation, and a relationship with NADPH oxidase complex, indicating an important role in oxidative stress. Conclusions: The striking differences between the proteome of HeLa and differentiated THP-1 cells reinforce the need for an optimized sample preparation protocol for macrophages. Our comprehensive assessment of digestion workflows for macrophages identified proteins specific to this cell line, enabling further studies to understand the role of these immune cells in inflammatory conditions. 1131

1133

P-III-0973 Large clostridial toxins reorganize the proteome and phosphoproteome of target cells Andreas Pich1, Theresa Schweitzer1, Florian Stieglitz1, Ralf Gerhard1, Harald Genth1, Ingo Just1 1Hannover Medical School, Toxicology, Hannover, Germany Large clostridial toxins (LCTs) have been identified as primary virulence factors in various animal and human diseases such as C. difficile infection (CDI), wound-associated gas gangrene or toxic shock syndrome. Classification of toxins as LCTs was made basing on a high sequence homology and similar structural set up as well as their ability to induce pronounced changes in cell morphology. The group of LCTs include toxins TcdA and TcdB from Clostridioides difficile, TcsL and TcsH from Paeniclostridium sordellii, TcnA from Clostridium novyi and Clostridium perfringens" TpeL. Proteome and phosphoproteome responses of C. difficile toxins and C. novyi"s TcnA were analyzed in cell culture to elucidate unknown functions of these pathogens and their toxins. A combinational approach of biochemical analysis and different MS techniques were used to investigate effects of TcdA, TcdB, CDT and TcnA on target cells in vitro. Cell-cell junction, cytoskeleton organization and cell death were found to be toxin-responsive. Toxic effects of TcdA could be traced back to its glucosyltransferase activity while in case of TcdB a catalytically inactive variant also altered the proteome of target cells and induced early cell death. By infection of mice with C. difficile spores various biological processes related to immune response were identified as enriched while metabolic processes were found to be down-regulated. Utilizing various MS techniques it was possible to illuminate so far unknown effects of C. difficile and its toxins as well as C. novyi"s TcnA on cellular processes of the host. 1133

1134

P-III-0974 Mapping heterogeneity in the protein landscape of human sperm Emmie Pohjanen1,2, Jan N. Hansen1,2, Jannika Patz3, Christoph Brenker3, Alexandra Johannesson1, Jenny Fall1, Mathias Uhlen1,4, Timo Strünker3, Ulrika Axelsson1, Emma Lundberg1,5,2,6 1Science for Life Laboratory, School of Engineering Sciences in Chemistry, Biotechnology, and Health, KTH - Royal Institute of Technology, Stockholm, Sweden 2Stanford University, Department of Bioengineering, Stanford, CA, United States 3Centre of Reproductive Medicine and Andrology, University Hospital Münster, University of Münster, Muenster, Germany 4Karolinska Institute, Department of Neuroscience, Stockholm, Sweden 5Chan Zuckerberg Biohub, San Francisco, CA, United States 6Stanford University, Department of Pathology, Stanford, CA, United States Sperm are among the most specialized cell types in the human body. Not only do they lack most of the organelles that are present in other cell types, but they also have a flagellum. Sperm cells have a massive task of navigating and traveling through the female reproductive tract towards the ovum and utilize the flagellum, which is a specialized motile cilium, as their motor, rudder, and sensor. For instance, sensory proteins on the flagellar membrane, such as ion channels, detect environmental signals and initiate a downstream signaling cascade that orchestrates the flagellar beat. Little is known about the signaling pathways and mechanisms that drive the flagellar beat, including the spatial information of signaling components in the flagella. Additionally, only a few of the hundred million of sperm cells ejaculated, reach the ovum. This indicates that there might be a certain heterogeneity in sperm cells that only allow some of them to complete the journey. In this study, we perform a comprehensive mapping of the subcellular spatial distribution of the protein repertoire in the human sperm cell, on the single-cell level. By utilizing antibody-based spatial proteomics with validated antibodies from the Human Protein Atlas (HPA) library, we are able to distinguish 12 subcellular structures of the sperm cell. We observe that roughly 60% of the sperm proteome shows heterogeneity on the single-cell level. By comparing the variable proteome to infertility databases, we will gain better insight to the factors contributing to male infertility cases. We believe that mapping the human sperm flagellum proteome will greatly contribute to our understanding of sperm diversity, signaling and function. We anticipate that this research will pave new avenues for research on how diseases translate into infertility phenotypes or for the development of in vitro diagnostic tools. 1134

1135

P-III-0975 Characterization of breast cancer tumoroids subclones by proteomics for personalize medicine Antonella Raffo Romero1, Laurine Lagache1, Nawale Hajjaji1,2, Marie Duhamel1, Isabelle Fournier1, Michel Salzet1 1Laboratoire PRISM, Biology, Villeneuve d'Ascq, France 2Breast Cancer Unit, Oscar Lambret Center, Lille, France Introduction: Breast cancer (BC) is the most common cancer among women causing the highest number of cancer-related deaths. Currently, BC research is mainly conducted using 2D cultures or mouse models. Both have proven to have limitations: 2D culture lacks cell-cell or cell-matrix interactions and mice are associated with ethical problems and poorly mimic the human diseases. There is an unmet need to develop novel more accurate, realistic, standardisable and robust experimentation models. Methodology: We have developed BC tumoroids made from a mixture of tumor cells isolated from human and canine BC tissues. In the tumoroid model, we conducted studies to isolate tumoroid-derived subclones with heterogeneous profiles. We analyzed these subclones using proteomics and mass spectrometry imaging to compare them with their parent tumoroids. Our goal was to determine whether the study of these subclones could lead to a more comprehensive treatment strategy that targeting all subclones, potentially reducing post- treatment resistance compared to standard hospital treatment. Results: Our results show that intratumoral heterogeneity is maintained in the tumoroid model. While many tumoroids exhibited a homogeneous profile, others showed heterogeneity within tumoroids derived from the same patient, which is critical for recapitulating in vivo drug responses. This led us to investigate different subclones within the tumoroids that represent intratumoral heterogeneity. We isolated different subpopulations and used mass spectrometry imaging to demonstrate their similarities to the parent tumoroids. Additionally, proteomic profiling revealed distinct differences between the subclones. Finally, we showed that not all subclones grow and respond uniformly to treatments. The next step is to propose new treatments based on our proteomics analysis that can affect all subclones, thereby preventing relapse in patients. Conclusion: Our knowledge of biology, combinedwith our skills in proteomics and mass spectrometry imaging helped us to better understand the BC tumoroids and to study the different intratumoral populations that are so problematic today. 1135

1136

P-III-0976 Hypoxia and Hyperglycemia-induced proteomic changes in retinal pigment epithelial cells: implications for tight junctions and vascular integrity Divya Ranga1, Ramandeep Singh1, Mohit Dogra1, Surya Prakash Sharma1, Nirbhai Singh1 1Post graduate institute of medical education and research, Department of Ophthalmology, Chandigarh, India Purpose:The retina is characterized by the presence of both an outer and inner blood-retinal barrier, which are essential for establishing immune privilege within the ocular environment. Outer blood-retinal barrier (oBRB) is composed of retinal pigment epithelium (RPE) cells that are interconnected by tight junctions. In Diabetic Macular Edema (DME), integrity of these cellular connections in retinal blood vessels are compromised due to hypoxia and hyperglycemia resulting in the leakage of fluid. We investigated molecular and pathway alterations in the proteome of RPE cells upon exposure to hypoxia, hyperglycaemia, and their combination. Method: In order to examine the molecular environment of oBRB, we grew RPE cells under normal glucose concentration of 5.5mM and glycaemic conditions 25 mM for 5 days. The cells were further either exposed to normoxia (20% O2) or hypoxia (1% O2) for 24 hours to mimic DME conditions in oBRB. The proteins extracted from cell pellets were subjected to enzymatic digestion, quantification and analysis using shotgun LC-MS/MS. Proteome Discoverer 2.0 used for identification of proteins. The proteins were further analysed with MetaboAnalyst 6.0, Reactome, string etc. Results: We have got more than 5000 proteins, upon analysis 4081 proteins were with high FDR and 1,658 of them being statistically significant within three groups. Reactome analysis revealed that proteins linked to tight junctions and cell adhesion pathways were disrupted in hypoxic and hyperglycemic cells as compared to normoxic cells. This disruption points to a possible stress-related degradation in the tight connection integrity of RPE cells. Analysing further showed the presence of angiogenin/F-actin pathway and its interactors in normoxic cells, validating its role in preserving vascular integrity. Notably, we found two novel proteins in RPE cells: junction plakoglobin and cofilin. Consistent with earlier research in rat models that linked this protein with neovascularization, cofilin was markedly increased under hypoxia and hyperglycemia. On the other hand, normoxic conditions increased junction plakoglobin, emphasizing its function in preserving cellular integrity under normal physiological situations. Conclusion: Our results show that tight junction proteins and RPE cell adhesion are altered by hypoxia and hyperglycemia. Subsequent analysis has revealed that angiogenin/F-actin provides protection to the blood vessels in the retina against stress. Cofilin and junction plakoglobin were formerly unknown in RPE cells. DME studies on rats showed increased level of cofilin, similar observation seen in RPE stressed cells validating connection between neovascularization and this protein. Increased levels of junction plakoglobin levels in normoxic conditions, demonstrates its significance for cell integrity. Fig. 1 (optional) 1136

1137

P-III-0977 Proteomics reveals key differences in neutrophils activation by PMA and Ionomycin Rafaela Oliveira Nascimento1, Lorenna Rocha Reis1, Mariana Pereira Massafera1, Graziella Eliza Ronsein1 1University of São Paulo, Biochemistry, São Paulo, Brazil Neutrophils play a crucial role defending the organism against pathogens and orchestrating inflammatory responses through diverse mechanisms, including granule content release, reactive species generation, and formation of neutrophil extracellular traps (NETs). This web-like structures are composed of chromatin and microbicidal proteins, and capable of killing pathogens, but are also involved in autoimmune responses. Many different stimuli are able to generate NETs, and recent evidences suggest the mechanisms of NETs release might be distinct. In this study, employing live-imaging fluorescence microscopy and high-resolution proteomics, we investigated neutrophil activation by comparing two stimuli: phorbol-12-myristate-13-acetate (PMA), related to generation of reactive species, and ionomycin, a calcium ionophore. Both stimuli induced NETs formation, although ionomycin-activated cells showed a faster kinetics. Proteomic analysis of the neutrophils" response before NETs release shows differential protein secretion profiles. Thus, PMA-stimulated neutrophils released proteins from secretory vesicles, and the enriched processes were associated with adhesion, cell killing, and immune response. The ionomycin-treated cells exhibited biological processes related to azurophilic granule proteins, actin filament organization and metabolism. The stimuli also led to different post-translational modifications, increased protein citrullination, with citrullinated proteins pertaining to classes of cytoskeleton organization, nucleus stability, and NADPH oxidase activation. On the other hand, due to NADPH oxidase activation, reactive oxygen species and HOCl formation, PMA treatment correlated with protein chlorination, including proteins related to defense response. In conclusion, our findings showed stimulus-specific mechanisms driving neutrophil activation, revealing distinct pathways for PMA or ionomycin induced NETs formation. ionomycin activation showed 1137

1138

P-III-0978 Investigating Dynamic Cellular Crosstalk and Pathomechanisms in Experimental Peritoneal Dialysis through Secretomics Analysis Juan Manuel Sacnun1, Rebecca Herzog1, Klaus Kratochwill1 1Medical University of Vienna, Vienna, Austria Peritoneal Dialysis (PD) is a life-saving renal replacement therapy. Yet, the used of PD-fluids induce adverse effects leading to detrimental changes in the peritoneal membrane (PM), reducing the therapy efficacy. Cross- talk among different peritoneal cell types modulates PD-associated deterioration. Currently, there is no model available to study the interactions of these cells in close proximity. Here, we aimed to develop a co-culture model to explore cell-to-cell communication by analyzing the cellular proteome and secretome. For modelling the PM, mesothelial and endothelial cells were co-cultured in transwell plates under optimized conditions for simultaneous culturing under non-starving conditions (5% FCS). Cells were exposed to PD-fluids in either co- or single-culture conditions. To overcome current limitation on secretomics analysis in non-starvation conditions, an equalizer approach was used to deplete high abundant proteins in combination with SILAC for tracing the cell origin of secreted proteins. For quantitative analysis of cellular and secreted protein abundances, LC-MS (TMT-18plex) was performed. Co-cultured cells yielded differently regulated pathways following PD-fluid exposure compared to individual cultures. Analysis of the secretome revealed cell-specific pattern of secretion and low correlation with intra- cellular abundances. Combined proteome and secretome analysis revealed different ligand-receptor pairs expressed uniquely in co-culture. Furthermore, the secretome was able to capture different types of secretion. Using protein-protein interaction analysis, the identified cell-secreted proteins (~1900) formed 11 functional clusters, interacting with different receptors presented by the cells. The resulting interactome between cells through the secreted proteins in combination with differentially expressed cellular and secreted proteins revealed modulation of the cross-talk by PD-fluids. Identified secreted proteins had a high overlap with clinical samples from PD-patients and largely correlated with their clinical parameters. The combination of the modulation of the cross-talk by PD-fluids and clinical information revealed 3 novel candidates regulating pathways related to angiogenesis, EMT and NOTCH, which were validated in a chronic mouse model and patient samples. This study shows that harmful effects of PD-fluid exposure on mesothelial also affect endothelial cells. Co- cultures revealed different responses compared to individual cultures, highlighting the importance of models that allow interactions between multiple cell types. Secretomics analysis showed to be essential to understand cell-to-cell communication, and showed a more complex modulation than classical ligand-receptor interaction. We further identified novel potential signaling axes between the cell types explaining pathophysiological changes of the PM during PD that may allow identifying therapeutic targets to reduce current limitations of PD. 1138

1139

P-III-0979 Highlighting the fate of brain pericytes under the blood-brain barrier-associated conditions and neuropathological contexts by mass spectrometry approaches Julien Saint-Pol1 1Université d'Artois, Blood-Brain Barrier Lab (BBB Lab, UR2465), Lens, France Despite their importance in the induction and maintenance of the blood-brain barrier (BBB) main features throughout life, brain pericyte physiology is understudied as well as their fate under neuropatholigical conditions, i.e. once exposed to pro-inflammaroty cytokines produced by brain cells under neurodegenaritive processes. This presentation aims (i) to characterize brain pericyte physiological features and the proteimic switch of this cell type induced by BBB endothelial cells, and (ii) to highlight the pathological events in brain pericytes leading to a disease propagation in autocrine and paracrine ways. All the data have been generated through mass spectrometry approaches (MS): unlabelled quantitative Sequential Window Acquisition of all Theoretical Mass Spectra (SWATH)-MS and Tandem Mass Tag (TMT)-MS. Alltoghether, these data suggest a proteomic switch of brain pericyte to aswer to BBB endothelial cells needs, and confort the role of cytokines- mediated inflammaroty processes under neurodegenerative disorders since brain pericyte is the first cell type of the neurovascular unit to undergo stress conditions and is able to diffuse it in its neighborhood. This study confirms the key role of brain pericyte in BBB homeostasis in physiological and pathological contexts. 1139

1140

P-III-0980 Proteomic evidence of the inhibition in astrocytes of Heimdall, an alternative protein related to a variable light chain kappa, switches astrocytes to a neural stem cell phenotype Michel Salzet1 1Univ.Lille, Inserm, CHU Lille, U1192 – Proteomics Inflammatory Response Mass Spectrometry- PRISM, Biology, Villeneuve d'Ascq, France The dogma "One gene, one protein" is clearly obsolete since cells use alternative splicing and generate multiple transcripts which are translated into protein isoforms, but also use alternative translation initiation sites and termination sites on a given transcripts. Alternative open reading frames for individual transcripts give proteins (the alternative Proteins: AltProts) originate from the 5'- and 3'- UTR mRNA regions, frameshifts of mRNA ORFs or from non-coding RNAs. To gain insight into the role of these newly identified alternative proteins in the regulation of cellular functions, it is crucial to assess their dynamic modulation within a framework of altered physiological modifications such as experimental spinal cord injury (SCI). Here, we carried out a longitudinal proteomic study on rat SCI from 12h to 10 days. Based on AltProt predictions, it was possible to identify a plethora of newly predicted protein hits. Among these proteins, some presented a special interest due to high homology with variable chain regions of immunoglobulins. We focus our interest on the one related to Kappa variable light chains which is similarly highly produced by B-cells in the Bence jones disease, but here expressed in astrocytes. This protein, name Heimdall is an Intrinsically disordered protein which is secreted under inflammatory conditions. We investigated Heimdall's function in these cells using advanced neuroproteomics supported by molecular cell biology and electrophysiology assays and CRISP-CAS9 invalidation. Immunoprecipitation experiments showed that the Heimdall interactome contained proteins related to astrocyte fate keepers such as "NOTCH1, EPHA3, IPO13". However, when Heimdall protein was neutralized utilizing a specific antibody or its gene knocked out by CRISPR-Cas9, sprouting elongations were observed in the corresponding astrocytes. Interestingly, depolarization assays and intracellular calcium measurements in Heimdall KO, established a depolarization effect on astrocyte membranes KO cells impacting the astrocyte phenotype sustained by the decrease of NOTCH2 and switch to neural progenitor cells. Taken together, Heimdall is a novel neural key player involved in astrocytes gatekeeper phenotype. References Capuz A, et al., Heimdall, an alternative protein issued from a ncRNA related to kappa light chain variable region of immunoglobulins from astrocytes: a new player in neural proteome, Cell Death Dis. 2023 Aug 16;14(8):526 Capuz A, et al., Astrocytes express aberrant immunoglobulins as putative gatekeeper of astrocytes to neuronal progenitor conversion. Cell Death Dis. 2023 Apr 4;14(4):237. Capuz et al., The Antibody Dependant Neurite Outgrowth Modulation Response Involvement in Spinal Cord Injury. Front Immunol. 2022 Jun 16;13:882830. Fig. 1 (optional) 1140

1141

P-III-0981 Cross-Linking Mass Spectrometry to Capture Protein Network Dynamics of Cell Membranome Lucia Santorelli1, Michele Costanzo2,3, Sara Petrosino1,4, Michele Santoro1, Giulia Capitoli5, Margherita Ruoppolo2,3, Paolo Grumati1,4 1Telethon Institute of Genetics and Medicine (TIGEM), Pozzuoli, Italy 2CEINGE – Biotecnologie Avanzate Franco Salvatore, Naples, Italy 3University of Naples Federico II, Department of Molecular Medicine and Medical Biotechnology, Naples, Italy 4Federico II University, Department of Clinical Medicine and Surgery, Naples, Italy 5University of Milano-Bicocca, Bioinformatics Biostatistics and Bioimaging B4 Center, School of Medicine and Surgery, Monza, Italy Protein-protein interactions (PPIs) regulate every aspect of cell life, playing a pivotal role in terms of accurate governement of multicellular organisms through fine regulation mechanisms. The consistent organization of PPIs networks increases exponentially the cell capability to control a plethora of molecular phenomena, including signal transduction, endocytosis, secretion and even cell functioning or death. However, when this knotty balance between assembly and disassembly is hampered, the effects can impact on organs and tissue functionality. It is a matter of fact that the identification of PPIs networks as potential pharmacological target in the treatment of several diseases represents a consolidate strategy in personalized medicine. Furthermore, advancements in biochemical field continuously unravel the PPIs complexities, providing deeper insights into cellular function and discovering potential therapeutic candidates. To dissect that molecular intricacy, cross-linking mass spectrometry (XL-MS) emerged as one of the most promising strategies to capture the dynamics of interactomes from small scale, such as macromolecular complexes to large scale, including vesicles, organelles and cells. Since the plasma membrane proteins (PMPs) constitute ~60% of all drug targets, have access to the complete "membranome" PPIs network would represent an unbiased approach to depict regulatory pathways controlling cell behavior. Despite the biochemical impact, knowledge gaps regarding protein interactome at PM persist. In this challenging context, we applied an XL-MS approach combined with PM purification pipeline to profile, on a system-wide scale, PM interactome of He-La cells used as cellular reference model. Specifically, we enriched PM-fractions from total cell lysates cross-linked before (native) and after (disrupted) PM dissolution. We defined the best experimental settings in terms of MS-cleavable cross-linker concentration, cross-linking reaction time, PM enrichment, MS acquisition and elaboration. Based on all detected inter-links, we assembled a protein global network composed of unique residue-to-residue pairs originating from more than 300 proteins assigned to the PM compartment. Then, we compared our experimental PMPs map to already annotated interactions and we revealed an overlap of 25%, confirming the accuracy of our dataset. With respect to PM-associated proteins, we targeted the highest number of PPIs for specific PM member families, counting ATP-binding cassette transporters, solute carrier transporters family, ATPase Na+/K+, Cu2+ and phospholipid transporters. Finally, we identified the remaining cross-links as part of proteins which are currently not described as interacting with each other, presenting potentially de novo partners. Currently, we are implementing our XL-MS cell membranome pipeline to unravel the existence of hidden disease-specific protein subnetworks in cell models of genetic metabolic disorders. 1141

1142

P-III-0982 Identifying novel engineering targets from evolved host populations using proteomics Cristina Sayago Ferreira1, Kathryn Lilley2, Nicholas Bond1, Clare Lovelady1 1AstraZeneca, Cambridge, United Kingdom 2University of Cambridge, Biochemistry, Cambridge, United Kingdom Chinese Hamster Ovary (CHO) cells are the cell type of choice for the production of recombinant biopharmaceuticals, specifically antibodies. Thus, there have been many efforts to increase recombinant protein productivity and product quality. Directed evolution of CHO cells has been shown to improve their functionality, in terms of higher antibody expression or product quality. In this process, host cells are subjected to a physical or chemical stress and cells that survive these conditions are able to persist and form a resistant population. Here, we have compared the proteome landscape of several evolved hosts to identify differentially regulated pathways that may explain phenotypic differences observed between the hosts. Expressing and null transfectant pools of seven evolved hosts were analysed and compared to the parental host. Using a multiplexing strategy, we found that both groups, expressing and null, showed distinct protein profiles. This indicates that the proteomes of expressing and null cell lines diverge when cells are expressing the antibody. This analysis revealed several proteins commonly downregulated in the antibody expressing pools, including proteins involved in degradative autophagy, endoplasmic-reticulum-associated protein degradation (ERAD) pathways and response to endoplasmic reticulum stress. Additionally, proteins related to organelle assembly and cellular localization were upregulated in all of the antibody expressing pools. Moreover, our results demonstrate that the evolved hosts with higher antibody expression and specific productivity (qP) undergo similar changes in their proteome profiles, and importantly, this response was different to the hosts with lower expression and qP. Particularly, proteins related to Golgi apparatus, cell cycle, endosome or nucleolus were found to be upregulated among these evolved hosts. These results suggest that the evolved hosts with higher antibody expression undergo similar mechanisms that are mostly associated with pathways involved in protein secretion. Finally, given that changes in the distribution of proteins between organelles may be fundamental in the regulation of the secretory pathway, we conducted a LOPIT-DC approach to identify changes in the protein subcellular location of a subset of pools which exhibited more changes in the total protein expression levels. Altogether, the work presented provides an insight into the fundamental proteomic changes underlying phenotypic differences in these manufacturing host cell lines. 1142

1143

P-III-0983 High fat diet promotes increased protein expression of MTPF1 and enhanced mitochondrial fission in fast type IIa skeletal muscle fibers Luisa Schmidt1, Marcus Krüger1, Michael Saynisch1, Philipp Antczak1, Ximena Hildebrandt1, Nieves Peltzer1, Christiana Zollo1, Ana Garcia-Saez1, Katharina Neuser1 1University of Cologne, CECAD, Institute for Genetics, Cologne, Germany Skeletal muscle is responsible for nearly 70 % of insulin-stimulated glucose processing, making it the body's largest energy reservoir. Severe obesity and type 2 diabetes mellitus reduce glucose uptake and processing in skeletal muscle, impairing muscle performance. Muscle glucose, fatty acids, and lipids processing is closely linked to its fiber type composition, which includes slow (type I) or fast (type IIa, IIx, IIb) fibers, each with distinct metabolic characteristics. Shifts in fiber type composition significantly impact whole-body energy homeostasis. Mitochondria, the most abundant organelles in both fiber types, have their activity regulated by fusion and fission processes responsive to nutritional and metabolic signals. Understanding how obesity and pre-diabetic stages affect individual muscle cells is vital for comprehending metabolic changes in muscle tissue. To investigate metabolic changes between fiber types in response to high-fat diet (HFD), we isolated 50 single soleus muscle fibers of wild-type mice subjected to a control diet (CD) or HFD (n=4) for 16-weeks. Single myofibers were transferred to 96-well-plates and subjected to tryptic digestion according to the SP3 protocol. Desalted peptides were analysed with short 21-min LC-MS gradients in data-independent acquisition mode with an ion mobility mass spectrometer. In total we identified ~2,600 proteins and ~2,000 proteins per single muscle cell. Marker proteins of metabolic pathways, including the citric acid cycle, ß-oxidation and fatty acid metabolism, were significantly altered, confirming the pre-diabetic phenotype in response to the HFD. The significant decrease in the fibre type-specific protein MYH7, consequently type I fibres, complicated comparative analysis between CD and HFD conditions. To address this, we used over 50 known fiber type proteins for cluster analysis, resulting in four clusters representing type I, type IIa fibers, mixed type I/IIa, and mixed type IIa/IIx fibers under normal and HFD conditions. Our analysis indicated an increase in mitochondrial content in pre-diabetic mice, with an increased abundance of the mitochondrial fission protein 1 (MTFP1) exclusively in type IIa fibers. Immunostaining against MTPF1 and TOM20 substantiated our proteomics findings and revealed an elevated mitochondrial fission in type IIa fibers compared to slow type I fibers in the soleus. Our analysis revealed significant metabolic changes in type IIa compared to type I fibers, suggesting greater sensitivity of type IIa fibers to stress conditions such as increased fatty acid intake. This study also introduced an alternative bioinformatic approach for classifying muscle fiber types independently of MYH, enabling more accurate comparability between fiber types. We are convinced that analysing individual muscle fibers will provide a more precise view of metabolic changes during aging, in metabolic syndromes and neuromuscular diseases such as amyotrophic lateral sclerosis. 1143

1145

P-III-0984 Characterizing global translation and mitochondrial protein import in senescence using mePROD and mePRODmt Jonas Martin Schmidt1, Rebecca George Tharyan1, Bhavesh Parmar1, Nikita Verheyden1, Süleyman Bozkurt1, Christian Münch1 1Institute of Molecular Systems Medicine, Faculty of Medicine, Goethe University, Frankfurt, Germany During aging, senescent cells accumulate in the human body. These cells are characterized by irreversible cell- cycle arrest, pro-inflammatory phenotype, and decline in proteostasis. Senescence also leads to changes in mitochondria, including enlargement, increased mass, reduced ATP production, and decreased membrane potential. These alterations likely affect mitochondrial protein import (MPI) that relies partly on membrane potential and ATP. Consequently, MPI impairment is highly plausible in aged mitochondria. Despite the significance of MPI, the impact of senescence on this process is largely unknown. This project aims to address this knowledge gap using mePRODmt, a SILAC-based proteomics approach for quantifying protein uptake into mitochondria. Since import and translation are closely related, we measured translation using the global SILAC- based proteomics approach mePROD. We induced senescence in three different ways in human lung fibroblasts (IMR90) and measured import and translation over the course of senescence transition. To further investigate alterations in MPI during senescence and examine affected signaling pathways in the senescent phenotype, we performed phospho-proteomics. Understanding the impact of senescence on translation and mitochondrial protein import is crucial for unraveling the complex interplay between aging, cellular homeostasis, and neurodegenerative diseases. 1145

1146

P-III-0985 Mapping the proteomic landscape of healthy human brains over the aging process Ernst Heinz Pulido1, Katrin Schmitt1, Rong Chi1, Jennifer Okamoto2, Alina Isakova2, Emma Lundberg1 1Stanford University, Bioengineering, Stanford, CA, United States 2Stanford University, Brain Resilience Laboratory, Stanford, CA, United States As our brains age, they undergo continues changes. Human and animal models suggest that even in healthy brains age-related tissue atrophy, alteration in neurotransmitters and damage accumulation in the cellular environment lead to cognitive decline. Aging processes, especially on the protein level remain incompletely understood. We aim to systematically map proteomic changes in the brain over the aging process with a particular focus on protective factors against neurodegenerative processes. We leverage high-resolution mass spectrometry on the Thermo Fisher Astral of fresh frozen brain tissue samples, to compare the proteome of different brain regions from male and female healthy human individuals over different age groups. We strive to identify molecular signatures of healthy brain aging. This will pave new avenues for developing strategies to prevent and or treat cognitive decline. 1146

1147

P-III-0986 Integrative study of patient-derived in-vitro model toward pharmaco-proteogenomics Yomogi Shiota1,2, Julia Osaki1, Rei Noguchi1, Shuhei Iwata1,3, Yuki Adachi1,4, Daiki Kato5, Tatsuya Usui2, Tadashi Kondo1 1National Cancer Research Center / Japan, Division of Rare Cancer Research, Tokyo, Japan 2Tokyo University of Agriculture and Technology, Department of Veterinary Pharmacology, Tokyo, Japan 3Chiba University Graduate school of Medicine, Department of Orthopaedical Sugery, Chiba, Japan 4Asahikawa Medical University Hospital, Department of surgery, Hokkaido, Japan 5The University of Tokyo Graduate School of Agricultural and Life Sciences, Faculty of Agriculture, Tokyo, Japan Background: Pharmaco-proteogenomics using patient-derived cancer models is a promising approach for the development of predictive biomarkers as well as the discovery of therapeutic targets. The critical resource of pharmaco-proteogenomics is the patient-derived cancer models, which are established from tumor tissues and maintain the characteristics of original tumors. Various in-vitro models have been established from malignant tumors and applied for the study of cancer biology as well as translational research. Previous reports revealed that cancer stem cells played key roles in drug resistance, metastasis, and invasiveness of clinical tumors. With this regard, the organoids are widely used as in-vitro models for evaluating drug sensitivity. However, the clinical utility has not been established in any models, and further investigation is required to improve the quality of pharmaco-proteogenomics. The utility of cancer models depends on how their phenome, proteome, and genome recapitulate the features of original tumors. Object: This study aims to reveal the similarity of phenotype and proteotype among the in-vitro models derived from identical tumor tissue and cultured in different methods. Stemness, tumorigenicity, and drug sensitivity are examined and used to interpret the proteomic difference among the individual cell models. Methods: Tumor tissues of canine lung cancer were used to establish the monolayer, spheroid, and organoid cultures. The cells were exposed to multiple drugs, and the response to the treatments was observed. The population of stem cells in the cultured cells was examined by Fluorescence Activated Cell Sorting (FACS). The expression of proteins and mRNA in the molecular pathways were studied by mass spectrometry and next- generation sequencing, respectively. The protein expression was confirmed by immunohistochemistry. The tumorigenic potential was evaluated in xenograft models. Results and Discussion: We found that the population of stem cells and the response to drug treatments were diverse in the monolayer, spheroids, and organoids. Moreover, the cells exhibited the unique expression of proteins and mRNAs under different cultured conditions. These observations were consistent with the previous reports, where individual experiments were performed in the different studies. The integration of whole data in our experiment system will generate novel biology as well as clinical applications. 1147

1148

P-III-0987 Dietary influence on protein post-translational modifications: insights on protein acetylation and malonylation Kyle Hoehn1, Riya Shrestha1, Ling Zhong2, Mark Raftery2 1University of New South Wales, School of Biotechnology and Biomolecular Sciences, Kensington, Australia 2University of New South Wales, Bioanalytical Mass Spectrometry Facility, Sydney, Australia Post-translational modification (PTM) of amino acid residues regulates protein function, activity, structure, and localization. Acetylation and malonylation are two major acylated modifications occurring in various proteins, especially on the ε-amine group in lysine residues. It has long been known that these lysine acylations are modulated by SIRT enzymes; however, they can also be regulated non-enzymatically. Both acetylation and malonylation PTM have been identified in histone and non-histone proteins. Since both acetyl-CoA and malonyl- CoA, substrates for acetylation and malonylation respectively, are by-products of the Krebs cycle, lysine acetylation and malonylation are commonly found in proteins involved in metabolic processes. Diet composition is known to regulate many catabolic and anabolic metabolic processes such as glycolysis, fatty acid oxidation, lipogenesis, and oxidative phosphorylation. In our study, we used C57BL/6 female mice and fed them three different diets from 8 weeks of age for 8 weeks: normal chow diet (74% carb (67% from starch), 20% protein, and 6% fat), high sugar diet ( 74% Carb (57% from sucrose), 20% protein, and 6% fat), and ketogenic diet (1% carb, 9% protein, and 90% fat). Immunoaffinity purification followed by label-free detection of acetylated and malonylated lysine peptides in the mice livers on the three different diets and compared the fold changes of these peptides across the diets. This study provides invaluable information on how diets can modify different acylation PTMs and offers insights into microscopic changes in proteins related to these diets, with implications for various diseases and diet therapies. Fig. 1 (optional) 1148

1149

P-III-0988 Kingdom of reproductive life; core sperm proteome David Skerrett-Byrne1,2,3,4, Brett Nixon2,3, Timothy Karr5,6, Taylor Pini7 1Helmholtz Munich, Institute of Experimental Genetics, Munich, Germany 2University of Newcastle, Priority Research Centre for Reproductive Science, School of Environmental and Life Sciences, Newcastle, Australia 3Hunter Medical Research Institute, Infertility and Reproduction Research Program, Newcastle, Australia 4German Center for Diabetes Research, Munich, Germany 5Arizona State University, Biosciences Mass Spectrometry Core Research Facility, Knowledge Enterprise, Tempe, AZ, United States 6Arizona State University, ASU-Banner Neurodegenerative Disease Research Center, The Biodesign Institute, Tempe, AZ, United States 7The University of Queensland, School of Veterinary Science, Gatton, Australia Reproductive biology is often considered in the three siloed research areas, namely humans, domesticated animals and wildlife. Yet, there are common needs across these silos, notably treatment of subfertility, biomarkers for fertility/gamete selection, development of assisted reproductive technologies and effective contraception. To efficiently develop solutions applicable to all species, we must improve our understanding of the common biology underpinning reproductive processes. Accordingly, here, we performed an in-silico analysis of publicly available sperm proteomic data across 12 vertebrate species to consolidate these proteomic data and to define the core sperm proteome; a collection of highly conserved proteins that are critical for sperm structure and function. Over 2TB of RAW mass spectrometry data was sourced from ProteomeXchange and processed through a stringent and uniform search protocol to provide high confident protein identifications (FDR ≤ 0.01). A total of 13,853 proteins residing in the sperm of those species studied herein were identified, with the most significant contributions to this inventory being from humans and mice. Proteomic characterisation of non- traditional model species revealed >90% of their sperm proteins are currently curated as predicted to exist or inferred from homology, indicating that experimental evidence for their existence remains poorly defined. Despite variations in sperm proteome size, pathway analyses showcased functional relationships between species that differed from that of their evolutionary distances. Moreover, we report proteins that are highly conserved at the species (45 proteins) and order (135 proteins) taxonomic levels. Such proteins, in turn, mapped to critical pathways and molecular functions including protein folding and recycling, acrosome function and energy generation. In addition, we demonstrated that despite the fundamental biological role of spermatozoa, differing physiological adaptations such as alterations in sperm metabolic preferences, external testes or even a history of selective breeding, are reflected in differences in the sperm proteome Fig. 1 (optional) 1149

1150

P-III-0989 Relative quantification of membrane associated EGFR from human cancer cell lines via a SILAC based method Jason Tonillo1, Johanna Meder1, Roland Kellner1 1Merck Healthcare KGaA, ADCs & Targeted NBE Therapeutics, Darmstadt, Germany Antibody-drug conjugates (ADCs) demonstrate a promising approach in cancer treatment strategies. Generally, ADCs are composed of a monoclonal antibody and a cytotoxic molecule that is conjugated by a linker to the antibody. After specific binding of the monoclonal antibody to the target protein located on the cell surface, the ADC-target protein complex is internalized and releases the drug by proteolytic digestion intracellularly. This causes selective killing of the targeted cancer cell. Therefore, ADCs are specific and toxic at the same time. Since ADC binding to the target protein is highly dependent on the receptor density on the cell surface, data of receptor densities provide valuable information about the biological background of ADC internalization, ADC efficacy and potency. To address the evaluation of receptor densities in human cancer, the exemplary target protein EGFR was selected and relatively quantified by applying a SILAC based approach. Hence, membrane EGFR from four different cell lines were relatively quantified. The cell lines BxPC3, A431, A673 and MDA-MB-468 originated from pancreatic cancer, epidermoid carcinoma, Ewings Sarcoma and breast cancer, respectively. Because previous performed fluorescence activated cell sorting (FACS) investigations have evaluated that the highest EGFR receptor density is present on the MDA-MB-468 cells, this cell line has been defined as reference and was cultivated in heavy medium, which contains isotope labeled lysine and arginine. After sample preparation and subsequent LC-MS/MS analysis, 11 top EGFR peptides were identified and further used for the evaluation process. Relative membrane EGFR quantities were calculated by using the max. intensities of the top 11 EGFR light and heavy peptides. By comparing the EGFR amounts in the membrane fractions of three cell lines with the reference cell line, approximately sixfold more EGFR was detected in A431 cells and twofold more EGFR was present in the BxPC3 cells. No EGFR was discovered in the membrane fraction of the A673 cells. It has been demonstrated that the SILAC based method is suitable for accurate and precise quantification of EGFR located in the membrane fraction of four human cancer cell lines. 1150

1151

P-III-0990 Proteomic strategy to identify the cell-surface progesterone-receptor mediating nongenomic steroid-signaling in human sperm cells Leander Van der Hoeven1, Tanveer Singh Batth1, Anna-Kathrine Pedersen2, Anders Aagaard Rehfeld3, Jesper Olsen1 1Novo Nordisk Foundation Center for Protein Research, Jesper Olsen Group, Copenhagen, Denmark 2Novo Nordisk, Copenhagen, Denmark 3Rigshospitalet, Klinisk Biokemi, Copenhagen, Denmark For a human sperm cell to successfully fertilize a human egg, the functions of the sperm cell must be precisely regulated during its journey through the female reproductive tract. Many of these functions are controlled via intracellular Ca2+ concentration, e.g., sperm motility, and acrosome reaction. Interestingly, all channel-mediated Ca2+ influx in human sperm cells occurs via CatSper, the calcium/cation channel that is exclusively expressed in sperm cells. It has been known for decades that the steroid progesterone induces a rapid Ca2+ influx in the transcriptionally silent human sperm cell. While it has been revealed that the binding site of progesterone in sperm cells must be located in close proximity of the CatSper channel in the plasma membrane, the actual receptor is yet to be identified. As CatSper is exclusively expressed in sperm cells, it, and its main endogenous ligand progesterone, present a promising specific and safe pharmacological target for both novel on-demand non-hormonal unisex contraceptives and for novel treatments to enhance male fertility. Here we employed a strategy based on a combination of state-of-the-art mass spectrometry-based proteomic techniques to identify and characterize the progesterone receptor. We used a combination of Thermal Proteome Profiling, Limited Proteolysis coupled with Mass Spectrometry, and crosslinking combined with a biotin tagged progesterone for Mass affinity purification coupled with Mass spectrometry. Using offline high-pH (HpH) reversed-phase peptide fractionation and short online liquid chromatography (LC) gradients, we profiled the sperm proteome analyzing the head and the tail separately and identified more than 9000 proteins. This is an increase of 40% in protein coverage compared to previous attempts. . Using the aforementioned techniques, we created a shortlist of potential cell-surface progesterone-receptor candidates, which will be validated through human sperm cell-based assays to determine perturbation of the progesterone-induced Ca2+ signals upon inhibition of these proteins. 1151

1152

P-III-0991 Intestinal organoids – a potential screening tool for advanced drug delivery systems Dinh Son Vo1, Ana Lopes2, Kajsa Björner3, Johan Vessby3, Mikael Sellin2, Per Artursson1, Madlen Hubert1 1Uppsala University, Department of Pharmacy, Uppsala, Sweden 2Uppsala University, Department of Medical Biochemistry and Microbiology, Uppsala, Sweden 3Uppsala University Hospital, Uppsala, Sweden Introduction: Nano-sized formulations for targeted oral drug delivery show promise in treating local gastrointestinal diseases, offering a patient-friendly alternative to predominantly parenteral treatments1. However, assessing advanced delivery systems requires more comprehensive intestinal models. Bridging the gap between conventional 2D cell-based systems and in vivo models, 3D organoids are able to partially recapitulate a cellular phenotype closer to the normal human intestinal epithelium2. Reversing the polarity of organoids from apical-in ("standard configuration") to an apical-out orientation enables direct access to the apical surface of the epithelium3. Thus apical-out organoids represent a powerful new tool for the precise evaluation of drug delivery system efficacy. Aim: To establish an in vitro 3D human organoid model derived from stem cell-containing crypts from ileal biopsies and characterize the model in terms of morphology, differentiation and protein expression. Method: Culture conditions were established to generate 3D organoids by isolating stem cell-containing crypts from human ileal tissue. Organoids were initially derived as apical-in cultures in an extracellular matrix (ECM). To access the apical surface of the epithelium, the polarity of the organoids was reversed by removing the ECM. Apical-in organoids were characterized for structural morphology (size, budding formation, shape) using brightfield and fluorescence microscopy. Mass spectrometry-based proteomics was utilized to evaluate cell composition (differentiation and immaturity marker expression), functionality (tight junction protein expression), and the expression of proteins involved in the cell uptake of drug delivery vehicles and intracellular transport. Results: Microscopic analysis revealed that the apical membrane of the organoids faced the suspension environment, confirming that the polarity was successfully reversed (Fig 1A). Apical-out organoids maintained consistent morphological structures over a period of eight weeks (eight passages). Proteomic analysis indicated expression of ileum-specific biomarkers and fundamental ADME machinery. Furthermore, expression of regulatory proteins of major endocytic pathways such as clathrin-mediated endocytosis (CLTC), caveolin- mediated endocytosis (CAV1) and macropinocytosis (CTBP1) and markers for intracellular compartments (Rab5a, Rab7, LAMP1) in organoids was comparable to biopsies (Fig 1B). Conclusion: Apical-out ileal organoids were successfully established with physiology-similar properties and will in the next step be evaluated as a screening tool for advanced drug delivery systems. Acknowledgement: The authors are grateful to the European Union for financial support (GENEGUT, GA 101057491). References: 1 Chu, J. N. et al. Nat. Rev. Gastroenterol. Hepatol. 19, 219–238 (2022); 2 Gunti, S. et al. Cancers 13, 874 (2021); 3 Co, J. Y. et al. Cell Rep. 26, 2509-2520.e4 (2019) 1152

1154

P-III-0992 Insight into communication mechanisms in MLL-rearranged Leukemia Joanna Watral1, Sudip Ghosh1, Ariana Seira Calderon Moreno1, Mina Davoudi1, Qinyu Zhang1, Maria Jassinskaja1, David Bryder1, Charlotta Böiers1, Jenny Hansson1 1Lund University, Lund, Sweden Although much progress has been made in the treatment of some types of blood cancers, others, particularly leukemias associated with MLL rearrangements, are still associated with dismal prognosis in both children and adults. New therapeutic options that are tailored to the age-specific biology of the disease are urgently needed. While numerous intrinsic age-specific molecular features of leukemia-initiating cells have been identified, understanding the interactions between leukemic cells and their microenvironment remains far from comprehensive. The aim of this study was to establish methods to investigate proteomic and cellular changes to the local extracellular microenvironments of hematopoietic stem and progenitor cells upon leukemia initiation. Biological materials were isolated from mice with an inducible MLL oncogene system and fusion partners. Flow cytometry and proteomic analysis were used to monitor changes in the fetal liver and adult bone marrow leukemic microenvironments. For ex vivo studies, polyvinyl alcohol-based (PVA) culture system was used. Proteomics data were acquired using FAIMSPro coupled to Exploris 480 in DIA mode, followed by statistical and bioinformatic analysis to elucidate potential communication mechanisms between leukemic cells and their environment. We established a protocol to study the proteomic composition of secretome using a PVA culture system and noticed marked differences in the fetal and adult pre-leukemic cells’ potential to influence and communicate with their microenvironment. While adult hematopoietic progenitors increase their secretion upon leukemia induction, fetal cells instead suppress secretion. Additionally, we found that the secretion pattern could be reversed by treatment with proteins that are highly abundant in the normal extracellular fluids of the niche. However, these changes were more specific to adult cells compared to their fetal counterpart. The findings influence the microenvironment, showing greater vulnerability to reversion of changes occurring upon leukemia initiation compared to fetal cells. These differences highlight significant variations in disease biology between infant and adult leukemias, offering insights for improving age-tailored therapeutic approaches to enhance the outcomes for leukemia patients. indicate that adult pre-leukemic cells more actively communicate with and 1154

1155

P-III-0993 Proteomic and high-resolution microscopy analysis of the role of Dysferlin in membrane remodelling after myocardial infarction Justus B. Wegener1,2, Yannik Zuehlke1,2, Carolin Fleischhacker1, Nora Josefine Paulke1, Gabriel Christian Riedemann1,2, Tobias Kohl1, Tobias Moser3,4,2, Gerd Hasenfuß1,2, Christof Lenz2,5,6, Stephan E. Lehnart1,2, Sören Brandenburg1 1University Medical Center Göttingen, Clinic of Cardiology and Pneumology, Goettingen, Germany 2University of Göttingen, Cluster of Excellence "Multiscale Bioimaging: from Molecular Machines to Networks of Excitable Cells" (MBExC), Goettingen, Germany 3University Medical Center Göttingen, Institute for Auditory Neuroscience and InnerEarLab, Goettingen, Germany 4Max Planck Institute for Multidisciplinary Science, Auditory Neuroscience and Synaptic Nanophysiology Group, Goettingen, Germany 5University Medical Center Göttingen, Department of Clinical Chemistry, Goettingen, Germany 6Max Planck Institute for Multidisciplinary Sciences, Bioanalytical Mass Spectrometry Group, Goettingen, Germany Introduction: Dysferlin, a large 230 kDa transmembrane protein with multiple Ca2+-binding C2-domains, is thought to aid in membrane repair by sealing sarcolemmal lesions upon Ca2+ entry. Mutations in Dysferlin are linked to dilated cardiomyopathy. However, its role in protecting membrane nanodomains in ventricular myocytes after myocardial infarction remains unclear. Methods: Myocardial infarction (MI) in adult wild-type (WT) and Dysferlin knockout (KO) mice was induced by ligating the left anterior descending artery. Sham-operated animals served as controls. Following heart extraction one week post-MI in n=5 mice per group, left ventricles were dissected into infarction zone, infarction border zone and remote zone under microscopic control. Left ventricular (LV) samples were lysed by pressure cycling technology, digested using 2% SDS and amine magnetic beads, and analysed by data-independent acquisition mass spectrometry (DIA-MS) on the Bruker timsTOF Pro. Additionally, we investigated the cardiac Dysferlin interactome in absence and presence Ca2+ via co-immunoprecipitation and DIA-MS of short run SDS- PAGE purified and in gel trypsin digested LV samples. Ultimately, super-resolution stimulated emission depletion (STED) microscopy was used to investigate the subcellular localization, function and interaction partners of Dysferlin in the MI border zone. Results: One week post-MI, echocardiography revealed larger infarct sizes and decreased LV ejection fraction in KO compared to WT mice (mean±SEM: 21±2% vs. 28±2%, n=20/30 mice, P<0.05). We reproducibly quantified 4.360 proteins including large transmembrane proteins across all LV samples, thereby highlighting genotype- and MI zone-specific proteomic changes. In the MI border zone, 725 differentially abundant proteins were identified in WT vs. KO samples, underscoring Dysferlin's crucial role in LV remodelling post-MI. Co- immunoprecipitation experiments from WT vs. KO LV tissues revealed previously unknown cardiac protein interactions with the Ca2+ release channel Ryanodine receptor type 2 and the gap junctional hemichannel Connexin-43. Consequently, STED microscopy was applied to study the subcellular localization and function of Dysferlin at Ca2+ release units and the intercalated disc membrane folds, revealing Dysferlin to protect specific membrane nanodomains in the MI border zone. Conclusions: In a spatial proteomics approach, we uncovered the proteomic basis of specific LV zones in MI by DIA-MS. Dysferlin deficiency further worsened LV remodelling post-MI as shown in deep proteomic profiling. Co- immunoprecipitation experiments and DIA-MS analysis characterized the cardiac Dysferlin interactome, including protein interactions with crucial Ca2+ release and membrane proteins in cardiomyocytes, which were confirmed by super-resolution microscopy. Therefore, Dysferlin emerges as a novel therapeutic target to maintain membrane integrity in cardiac tissue and prevent from cell death post-MI. 1155

1156

P-III-0994 Multilayered omics reveal sex- and depot-dependent adipose progenitor cell heterogeneity Yibo Wu1, Rana Gupta2 1University of Geneva, Geneva, Switzerland 2Duke University, Durham, NC, United States Single-cell RNA sequencing (scRNA-seq) has revealed that adult white adipose tissue (WAT) harbors functionally diverse subpopulations of mesenchymal stromal cells that differentially impact tissue plasticity. To date, the molecular basis of this cellular heterogeneity has not been fully defined. Here, we describe a multilayered omics approach to dissect adipose progenitor cell heterogeneity in three dimensions: progenitor subpopulation, sex, and anatomical localization. We applied state-of-the-art mass spectrometry methods to quantify 4,870 proteins in eight different stromal cell populations from perigonadal and inguinal WAT of male and female mice and acquired transcript expression levels of 15,477 genes using RNA-seq. Our data unveil molecular signatures defining sex differences in preadipocyte differentiation and identify regulatory pathways that functionally distinguish adipose progenitor subpopulations. This multilayered omics analysis, freely accessible at http://preadprofiler.net/, provides unprecedented insights into adipose stromal cell heterogeneity and highlights the benefit of complementary proteomics to support findings from scRNA-seq studies. Fig. 1 (optional) 1156

1157

P-III-0996 Investigating the regulatory role of Palmitoylation during cell divisions Gamze Nur Yapıcı1,2, Nazlı Ezgi Özkan1,2, Nurhan Özlü1,2 1Koç University, MOLECULAR BIOLOGY AND GENETICS, Sarıyer, Turkey 2Koç University, Sarıyer, Turkey Background/aim: Protein palmitoylation is a reversible lipid modification that plays a crucial role in regulating the localization and function of membrane-associated proteins. The modification has been shown to affect protein- protein interactions, membrane trafficking, and signaling pathways.SILAC-based proteomic analysis and the process of cell division to study the role of palmitoylation and enzymes that transfer palmitoyl groups during cell division. Methods: To prepare the cells for analysis, we synchronized HeLa cells at two key cell cycle stages: interphase and mitosis, using a well-established protocol.SILAC labeling was applied to create "heavy" and "light" cell populations, where heavy cells were metabolically labeled with 17-ODYA and light cells with palmitic acid, and vice versa in biological replicates.A clickable palmitic acid analog, 17-ODYA (17-octadecynoic acid), is incorporated into palmitoylation sites through a metabolic labeling method.Subsequently, the cells are lysed, and the click reaction is performed using biotin azide or rhodamine azide.The cells were cultured in DMEM medium devoid of lysine and arginine amino acids, supplemented with isotopically labeled lysine and arginine for heavy labeling and natural isotopes for light labeling. Subsequently, we performed click chemistry on the lysed SILAC-labeled cells, followed by biotinylation of palmitoylated proteins using streptavidin-coated beads. After enrichment, proteins were subjected to trypsin digestion, and resulting peptides were analyzed using mass spectrometry. The acquired data were processed using MaxQuant and Proteome Discoverer software against the Swiss-Prot human proteome database. Results and conclusions: As our prilimenery data using SILAC-labeled heavy and light samples, we have identified proteins with significantly altered SILAC ratios in samples labeled with 17-ODYA and palmitic acid. Upon further analysis of these proteins in the SwissPalm database, it was observed that the palmitoylated protein ratios exhibited a substantial increase. These findings confirm the successful identification of palmitoylated proteins. This data provides insights into the impact of palmitoylation on mitosis. Fig. 1 (optional) 1157

1158

P-III-0997 Proteomic analysis of PACS2 syndrome in a murine model: insights into molecular pathways and calcium regulation Arkadiusz Zbikowski1, Tomasz Kowalczyk1, Michał Ciborowski1, Adam Kretowski1,2, Napoleon Waszkiewicz3, Dominik Cysewski1, Kacper Lukasiewicz3,4 1Medical University of Białystok, Metabolomics and Proteomics Laboratory, Clinical Research Center, Białystok, Poland 2Medical University of Białystok, Department of Endocrinology, Diabetology and Internal Medicine, Białystok, Poland 3Medical University of Białystok, Department of Psychiatry, Białystok, Poland 4Medical University of Białystok, Experimental Medicine Centre, Białystok, Poland is an ultra-rare PACS2 Syndrome, resulting from E209K or E210K mutations (DEE). Despite patient disease classified among developmental and epileptic encephalopathies include early onset seizures, neurodevelopmental symptoms, heterogeneity, common characteristics impairment, and facial dysmorphism. Research on this syndrome limited due to the behavioral extremely low number of patients and the absence of standardized experimental models. To date, only three research groups have published data on the impact of the mutated protein on cellular function, all utilizing lines. To better understand the disease pathogenesis, more complex models bridging basic science and clinical studies are required. in vitro models and selected cell in the PACS2 protein, is Therefore, in this study we aim to insight into total proteome of brain and liver in mice model of PACS2 Syndrome. in Center of Experimental Medicine in collaboration with Czech Center of Phenogenomics. PACS2 Syndrome mouse model was obtain (Bialystok, Poland). Following Adult B6 mice were breed genotyping confirmation, six wild-type and six E209K heterozygote mice were the study. Mice were sacrificed and chosen brains regions and livers were collected. Due to proteome analysis, cerebellum, frontal cortex, hippocampus, other brain tissue and liver were homogenate. We performed proteomic analyzes using the Data Independent Analysis (DIA) mode of data collection. Tryptically aged proteins were separated on a nano-LC system coupled to Orbitrap Fusion MS, followed by data analysis using Spectronaut software. The analyzes allowed the identification of approximately 6,000 proteins differentiating the brain tissue of mice with the PACS2 mutation and the control group. included in is multifunctional sorting protein strongly associate with mitochondrial function. PACS2 protein In our study we detected different up and down regulation of molecular pathways dependent of studied tissue. Moreover, we observed changes in expression of proteins associate with calcium in neurotransmission and management neurodevelopment present in cytoplasm may lead to glutamate release from neurons and epilepsy. However, although our research provides new data on the molecular mechanisms of the disease, further analysis is required. light on changes levels in PACS2 Syndrome. Elevated calcium result may put some in cells. This, 1158

1159

Human health insights (neurobiology, cardiovascular, liver, kidney etc.) (P-III-0999 to P-III-1083) P-III-0999 Acetylome mapping of tibialis anterior muscle from mdx mice upon pharmacological Sirtuin1 activation Annapaola Andolfo1, Clarissa Braccia1, Matteo Giovarelli2, Silvia Zecchini2, Clara De Palma3 1IRCCS Ospedale San Raffaele, Center for Omics Sciences, Proteomics and Metabolomics Facility (ProMeFa), Milan, Italy 2University of Milan, Department of Biomedical and Clinical Sciences, Milan, Italy 3University of Milan, Department of Medical Biotechnology and Translational Medicine, Milan, Italy Duchenne muscular dystrophy (DMD) is an X-linked muscular disease, characterized by dystrophin deficiency due to a mutation in the gene encoding for this protein. Dystrophin deficiency leads to a cascade of several events, such as membrane instability, chronic inflammation, muscle degeneration and myofiber necrosis. As muscle regeneration processes decrease, muscle fibers are gradually replaced by connective and adipose tissue. While remarkable progress has been made in genetic approaches to restore dystrophin, or its function, targeting secondary pathological mechanisms remains an important issue to improve the efficacy of therapeutic interventions depending on tackling either the lacking gene or the downstream consequences. Besides, genetic approaches seem to be effective only in a limited number of patients and for this reason, it is necessary to focus on pharmacological strategies applicable to all DMD patients.An emerging therapeutic target for DMD seems to be the activation of Sirtuin1 (SIRT1). SIRT1 belongs to a class of NAD+-dependent class III histone/deacetylase proteins, the sirtuins. These enzymes modulate the activity of different nuclear and cytoplasmatic proteins, which in turn regulate several cellular processes. Sirtuins, and especially the SIRT1 isoform, promote longevity by delaying cellular aging, inhibit apoptosis, regulate the cell cycle, the glucose homeostasis and the insulin secretion, and are involved in inflammation, oxidative stress and mitochondrial biogenesis. SIRT1 is found both in the nucleus and in the cytosol and exerts its action by deacetylating histone and non- histone proteins. SIRT1 is expressed in different tissues including skeletal muscle, where it deacetylates and activates peroxisome proliferator-activated receptor-gamma coactivator (PGC-1α) thus improving muscle resistance, enhancing muscle fiber-type switching and decreasing the process of muscle wasting. Hence, the activation of the SIRT1 in skeletal muscle ameliorates the phenotype of the X-linked recessive, muscle-wasting disease DMD. Therefore, we decided to map the differences in the acetylome of tibialis anterior muscle samples from mdx mice (a common DMD murine model): samples derived from five mdx mice were compared to samples derived from other five mdx mice treated with an activator of SIRT1. We used an unbiased label-free quantitative proteomic approach, upon enrichment for acetylated peptides with PTMScan Acetyl-Lysine motif. In this way, out of 141 unique acetylation sites we identified 28 sites that were differentially abundant in the two conditions of interest; moreover, we analysed the corresponding proteins to map them in GO biological processes using David software. In conclusion, our results indicate that the modulation of SIRT1 activity leads to beneficial effects in skeletal muscles from mdx mice, especially confirming SIRT1 activation impact on muscle contraction. 1159

1160

P-III-1000 Exploring protein translation dynamics in autism spectrum disorder-associated mutations Jose Ignacio Astorga Rebolledo1, Matthias Selbach1, Marija Mihailovic1, Robert Kerridge1 1Max Delbrück Center for Molecular Medicine, Protein Dynamics, Berlin, Germany Autism spectrum disorder (ASD) is a complex and highly prevalent neurodevelopmental disorder characterized by significant heterogeneity in genotypes and phenotypes. Recent research has identified dysregulation of protein synthesis as a potential converging mechanism in ASD. To explore altered translation in ASD, we used induced pluripotent stem cell (iPSC)-derived iNeurons and cortical brain organoids. CRISPR/Cas9 technology was employed to introduce ASD-causing mutations in the FMR1, PTEN, and TSC2 genes in iPSCs from healthy donors. These modified iPSCs were then differentiated into either iNeurons or cortical brain organoids. Using label-free data-independent acquisition (DIA), we identified about 8,000 proteins. Principal component analysis (PCA) revealed that mutations in TSC2 and PTEN caused a wide range of consistent changes in both iNeurons and organoids. These changes included up-regulated translation and mitochondrial processes and down-regulated transcriptional processes. Notably, the down-regulated proteins were strongly enriched in genes known to be associated with ASD, indicating that the proteomic data can help explain the broad genetic diversity of ASD. In summary, our integrated proteomic analyses of iPSC-derived iNeurons and organoids shed new light on the pathophysiological processes underlying ASD. 1160

1161

P-III-1001 Proteomic analysis of cerebral microvascular endothelial cells under physiological normoxia and hyperoxia: implications for drug discovery Sarah Barakat1, Fan Yang2, Kivanc Kok1,3, Giovanni E. Mann2, Emrah Eroglu4 1Istanbul Medipol University, Regenerative and Restorative Medicine Research Center (REMER), Research Institute for Health Sciences and Technologies (SABITA), Istanbul, Turkey 2King’s College London, King’s British Heart Foundation Centre of Research Excellence, School of Cardiovascular and Metabolic Medicine & Sciences, Faculty of Life Sciences & Medicine, London, United Kingdom 3Istanbul Medipol University, Department of Biostatistics and Medical Informatics, International School of Medicine, Istanbul, Turkey 4Sabanci University, Molecular Biology, Genetics, and Bioengineering Program, Faculty of Engineering and Natural Sciences, Istanbul, Turkey Traditional cell culture is performed under atmospheric oxygen levels (18 kPa), which do not accurately reflect the lower oxygen levels experienced by most cells in vivo (Keeley & Mann, Physiol. Reviews 2019;99:161-234). This is particularly important for brain endothelial cells, which are exposed to oxygen levels no higher than 5 kPa. Hyperoxia leads to sustained oxidative stress with significant implications for drug discovery and screening. In this study, we investigated the proteomic profiles of cerebral microvascular endothelial cells (hCMEC/D3) adapted for 5 days to either physiological normoxia (5 kPa O2) or standard hyperoxic conditions (18 kPa O2). Using an untargeted label-free proteomics approach, we identified significant alterations in protein expression profiles between 5 and 18 kPa O2. The observed differential expression profile was associated with an enrichment of pathways related to energy metabolism, oxidative stress and protein degradation. Furthermore, we investigated the effects of the NRF2 activator sulforaphane (SFN) in cells adapted to 5 or 18 kPa O2. Proteomic analysis indicated that SFN (2.5μM, 24h) had negligible effects in cells adapted to 5 kPa O2, whereas under 18 kPa O2 SFN treatment led to an enrichment of pathways related to cell cycle regulation and cell proliferation. These findings highlight the critical importance of culturing cells under physiologically relevant O2 levels to improve the clinical translation of in vitro models. Understanding how pericellular oxygen levels affect the phenotype of cells is essential for the design of drug discovery and high throughput screening protocols. 1161

1162

P-III-1002 Analysis of mouse plasma proteome reveals systemic effects of Bothrops jararaca snake envenomation Bianca C S C Barros1, Alison F A Chaves1, Miguel Cosenza-Contreras2, Mariana S L C Morone1, Ana T A Sachetto1, Niko Pinter2, Marcelo L Santoro1, Oliver Schilling2, Solange M T Serrano1 1Butantan Institute, Laboratory of Applied Toxinology, São Paulo, Brazil 2Freiburg University, Institute of Clinical Pathology, Freiburg, Germany Introduction: Envenoming by viperid snakes is a critical public health problem in Brazil and it is recognized by the World Health Organization as an important Neglected Tropical Disease, causing a high percentage of permanent disability and death. In Brazil, Bothrops snakes are responsible for approximately 20,000 snakebites annually. Although it is well established that systemic effects of Bothrops envenomation comprise coagulopathy, hemorrhage and renal failure, the molecular mechanisms involved in the disturbance of plasma proteins are poorly understood. Proteomic analysis based on mass spectrometry has been used as a powerful tool in order to better understand systemic effects related to snakebite envenomation. Objectives: i) to characterize changes in plasma protein abundance in animals injected with venom in the thigh muscle, using mass spectrometry; ii) to characterize the impact of bothropic antivenom upon the plasma proteome. Methods: In vivo experiments were carried out after approval by the animal research ethics committee of Butantan Institute under no. CEUA 9991131219. Swiss mice were injected with B. jararacavenom (1.6 mg/kg) in the gastrocnemius muscle, mimicking a snakebite, and 1 h later with antivenom (1.6 mg/kg; i.v. tail). After 3 h, 6 h, and 24 h, citrated blood samples were collected and centrifuged to obtain plasma, which was further processed to deplete three abundant proteins (albumin, IgG and transferrin) using a multiple affinity removal column. Samples of protein-depleted plasma were submitted to the single-pot, solid-phase-enhanced sample preparation (sp3) protocol followed by trypsin digestion and analysis by LC-MS/MS using a QExactive mass spectrometer, and data-independent acquisition. Data were searched in Spectronaut 18, and MSstats R package was used for statistical analysis. Results: An average of 7,000 peptides and 600 unique proteins were identified. The number of significantly differentially abundant proteins increased over time in response to envenomation. After 6 h of venom injection, an increase of acute phase proteins was detected, including lipocalin 2, apolipoproteins and serpins, whereas fibrinogen, vWF and factor XIII were decreased. After 24 h, several inflammatory markers, such as complement C3, alpha-2-macroglobulin, serum amyloid A1 and A2, serum amyloid P component (Apcs), creatine kinase and lactate dehydrogenase were detected as increased. Conclusions: Murine envenomation by B. jararaca resulted in changes in the plasma protein profile compatible with inflammation in addition to alterations in hemostasis, and indicated leakage of proteins due to muscle damage, and possibly from kidney tissue. Under the experimental conditions of this study, antivenom injection did not reduce the number of differentially abundant proteins. jararaca, snakebite, envenomation, anti-bothropic antivenom, proteomics, mass Keywords: Bothrops spectrometry. 1162

1163

P-III-1003 Systemic changes in adults living with a fontan circulation – insights from 3000 plasma proteins Tue Bjerg Bennike1, Benjamin Kelly2,3, Maren Ravndal4, Peter Bach2, Niklas Telinius2, Vibeke E. Hjortdal4 1Aalborg University, Health Science and Technology, Gistrup, Denmark 2Aarhus University, Department of Clinical Medicine, Aarhus, Denmark 3Aarhus University Hospital, Department of Cardiothoracic Surgery, Aarhus, Denmark 4Copenhagen University Hospital – Rigshospitalet, The Heart Centre, Copenhagen, Denmark A Fontan procedure is used to manage complex congenital heart defects where only one ventricle is functional. These conditions include tricuspid atresia, hypoplastic left heart syndrome, and double inlet left ventricle. In the Fontan circulation the venous blood is rerouted directly to the pulmonary arteries, allowing blood to bypass the heart and flow directly to the lungs, and the single functional ventricle pump oxygenated blood to the body. While this approach improves survival and quality of life, a univentricular circulation is associated with significant residual morbidity and mortality throughout life. In the current study, we aimed at providing a comprehensive characterization into the biological impact of living with a Fontan circulation. We obtained blood-plasma samples from 58 adults with a Fontan circulation having undergone the procedure early in life (systemic right ventricle due to any diagnosis, or systemic left ventricle due to tricuspid atresia or double inlet left ventricle), and matched then 2:1 with healthy controls. To characterize the systemic impact of the Fontan circulation, we performed a relative quantification of the levels of 2943 plasma proteins using the Olink® Explorer 3072 platform. Following quality control and filtering 2605 plasma proteins were quantifiable, of which 513 had significantly changed levels in the Fontan group (424 increased and 89 decreased). Categorizing the significant proteins by their biological function(s) in a pathway enrichment analysis, we found that the changes to biological processes in the Fontan group include angiogenesis, lipoprotein synthesis, metabolism, bone and calcium homeostasis, and muscle structure and function. The unprecedentedly high number of plasma proteins constitutes the hitherto most detailed characterization of the systemic impact of a Fontan circulation and provides a unique perspective into the Fontan physiology and clinical presentations. This study provides an elaborate and unprecedented insight into the biological impact of living with a Fontan circulation, and we speculate local hypoxia to be a central driver for multiple of the determined changes. This study may provide the foundation and directions for future studies wanting to examine changes in detail and/or explore possible therapeutic targets. 1163

1164

P-III-1004 Health and aging trajectories of human lung proteome: senescence, fibrosis, and aging signatures across lifespan in Naïve tissues and explant models Joanna Bons1, Delphine Beaulieu2, Jacob P. Rose1, Mark Watson1, Christina King1, John Sembrat2, Oliver Eickelberg2, Birgit Schilling1, Melanie Königshoff2 1Buck Institute for Research on Aging, Novato, CA, United States 2University of Pittsburgh, Division of Pulmonary, Allergy, Critical Care and Sleep Medicine, Department of Medicine, Pittsburgh, PA, United States Aging is a complex biological process associated with multi-factorial molecular alterations, leading to tissue deterioration and increased disease vulnerability. Human lungs are exposed to various stressors throughout lifespan. Age is a major risk factor for pulmonary physiological function impairment and chronic lung disease (e.g., fibrosis, COPD) that develop through poorly characterized mechanisms. There is a critical need to understand the underlying biology to develop signatures of healthy lung aging that can be used to predict aging trajectories. To explore healthy lung aging, frozen naïve lung tissue was collected from 20 participants – 8 female and 12 male – aged between 4 months and 95 years old (10 old and 10 young). After tissue homogenization and protein digestion with trypsin, samples were analyzed using an Orbitrap Eclipse mass spectrometer (Thermo Fisher Scientific) operated in data-independent acquisition (DIA) mode. DIA data were processed using directDIA (Biognosys), yielding 5,558 identified and quantified protein groups (≥ 2 unique peptides). The lung proteome was dramatically remodeled with age with 729 significantly changing protein groups (q-value < 0.05 and |old vs young| > 1.5), featuring up-regulated immune and inflammatory responses and altered extracellular matrix homeostasis. We identified a panel of proteins that significantly correlated with age, specifically COL6A6, MGP, TIMP2, and TIMP3 showed a positive correlation, while DHP, NID1, and TNR negatively correlated. In addition, we observed a strong sex-specific effect on lung aging with 83% female-specific and 56% male-specific proteome changes. The female lung aging signature was typified by a loss of SIRT2 and a dramatic increase of MPO and MUC5B, while MUC5B and TNC were lost in male lungs. We are expanding our proteomic analysis to human precision-cut lung slices (PCLS) i) from healthy donors for ex vivo senescence induction with bleomycin and ii) from donors with idiopathic pulmonary fibrosis (IPF) to gain insight into human lung senescence and diseases. Preliminary results for senescent and matched control PCLS explants (3 donors) demonstrated that we can successfully quantify 2,464 protein groups (≥ 2 unique peptides), and that we were able to identify 205 proteins that changed upon senescence induction (q-value < 0.01 and |bleomycin vs control| > 1.5). This included senescence-associated factors GDF-15, CRYAB, and DDP4, and featured up-regulated proteins associated with ECM organization and inflammation and down-regulated metabolism pathways. A panel of 64 proteins composed of NID1, MPO, LCN2, MMPs, SERPINA1, and laminins, was commonly associated with aging and senescence. Altogether, these findings provide novel insights into protein changes with healthy lung aging, upon senescence induction, and with fibrotic conditions to decipher and predict various aging trajectories in the human lung proteome. 1164

1165

P-III-1005 Characterization and pathophysiological contribution of the proteome redox status in pseudomyxoma peritonei Florina Iulia Bura Balajan1,2, María Torres Martínez1,2, Melissa Granados-Rodríguez1,2, María del Carmen Vázquez Borrego1,2, Francisca Valenzuela Molina1,3, Blanca Rufián Andújar1,3, Lidia Rodriguez-Ortiz1,3, Ana Moreno- Serrano1,2, Laura Morón Márquez1,2, Carmen Michán1,2, José Alhama1,2, Alvaro Arjona-Sanchez1,3, Antonio Romero-Ruiz1,2 1Maimonides Biomedical Research Institute of Cordoba (IMIBIC), GE09 Group, Córdoba, Spain 2University of Córdoba, Biochemistry, Córdoba, Spain 3Reina Sofia University Hospital of Cordoba, Unit of Surgical Oncology, Department of Surgery, Córdoba, Spain Pseudomyxoma peritonei (PMP), a poorly investigated condition, is a rare carcinomatosis in which mucus from various internal organs accumulates in the abdomen. The current treatment option is a radical cytoreductive surgery combined with hyperthermic intraperitoneal chemotherapy. Although outcomes are favourable, many patients experience tumour recurrence and progression. In addition, this pathology is not fully characterised at the molecular level, particularly at the protein level, due to its low incidence and the difficulty posed by the mucin-rich mucus. Furthermore, PMP, like other cancers, has been described to occur in a hypoxic manner, theoretically leading to an imbalance in redox homeostasis, which could play an important role in the development and maintenance of this disease. Therefore, this study aims to improve the understanding of the pathophysiology of this tumour by defining the redox state of its tissues, both low-grade (LG) and high-grade (HG), in comparison with control tissues (C), and to study its contribution to the aetiopathogenesis of PMP. To this end, we assessed the cellular redox state by measuring the ratio of oxidized or reduced thiol groups using an optimized protocol involving cysteine alkylation with N-ethylmaleimide, quantification by mass spectrometry and data analysis. Next, in vitro functional studies using antioxidants and prooxidants will be used to unveil the contribution of the redox status to the PMP malignancy by analysing changes in cell proliferation rate, enzyme activities, glycoprotein production and the expression levels of key genes. The results showed significant changes in the oxidised/reduced ratios of more than 100 proteins in the LG vs. C comparison, and about 40 proteins in both the HG vs. C, and LG vs. HG comparisons. Bioinformatic analyses performed with different software tools revealed associations between the observed changes and intracellular functional processes, such as oncogenic pathways, negative regulation of apoptotic pathways, protein phosphorylation, response to stress, etc. In conclusion, the results of the present study, together with the ongoing in vitro assays, indicate that the redox state is significantly disturbed in PMP patients, paving the way for further research to understand in depth the functional significance of the observed changes. Keywords: redox proteomic; oxidative stress; Pseudomyxoma peritonei; antioxidants; prooxidants. Funding resources: This work is supported by "Instituto de Salud Carlos III (ISCIII)" (PI19/1603 and PI22/001213; exp. FI23/00028), co-funded by the European Union, and by "Asociación Española contra el Cáncer" (PRYES223170ARJO). * FI. B-B and M. T-M contributed equally to this work and should be considered first authors. 1165

1166

P-III-1006 Elevated Markers of ECM Reorganization, Fibrosis, and Inflammation Uncovered in Accelerated Aging and Renal Dysfunction of Kidney Cortex and Medulla in Non-Human Primates Exposed to Ionizing Radiation Jordan Burton1, Christina King1, George Schaaf2, Stephen Kritchevsky2, J. Mark Cline2, Jamie Justice2,3, Ellen Quillen2, Birgit Schilling1 1Buck Institute for Research on Aging, Novato, CA, United States 2Wake Forest University, School of Medicine, Winston-Salem, NC, United States 3XPRIZE Foundation, Culver City, CA, United States Whole-body ionizing radiation (IR) causes damage to DNA, lipids, and proteins which leads to oxidative stress, inflammation, and cellular senescence, processes that accelerate biological aging. Slowly reproducing tissues, including kidney, are more likely to react to DNA damage, which may exacerbate aging processes. Kidney cortex and medulla tissues (N=38 each) were obtained from the Wake Forest Non-Human Primate Radiation Late Effects Cohort and experiments were conducted in compliance with state and federal animal welfare laws. Tissues were homogenized, proteins were digested using S-Trap, desalted by HLB C18-cartridges, then analyzed on a timsTOF HT mass spectrometer (Bruker) in data-independent acquisition (DIA) mode using parallel accumulation serial fragmentation (PASEF) to quantify dynamic protein changes across the following conditions: control with no kidney disease (CTL), irradiated with no kidney disease (IR), irradiated with kidney disease (IR KD), and control with kidney disease (KD). Raw data were analyzed using Spectronaut v18 (Biognosys) and significantly-altered proteins were subjected to bioinformatic analysis. We leveraged this NHP cohort, a unique national resource, to study accelerated aging and heterogeneity in resilience and aging in the kidney after ionizing radiation exposure. In this analysis, 5,472 and 5,411 M. mulatta protein groups with at least two unique peptides were identified in the kidney cortex and medulla, respectively. As the M. mulatta proteome is not well annotated, proteins from this dataset were mapped to homologous human proteins, thus providing a better annotation of the biological processes and associated gene ontology. Groupwise comparisons were performed to study the influence of 1) kidney disease in irradiated individuals (IR KD vs. IR), 2) only kidney disease (KD vs. CTL), and 3) only irradiation (IR vs. CTL). Interestingly, in almost all comparisons in both the cortex and medulla, over 75% of significantly-altered proteins (i.e., Q value < 0.01 and absolute average log2fold change > 0.58) are significantly upregulated. This includes the extracellular matrix and SASP proteins tenascin, upregulated in all conditions in the cortex and medulla, and periostin, upregulated in all conditions in the cortex. In both tissues, proteins that were significantly-altered and associated with kidney disease in irradiated and non-irradiated animals included SASP proteins and were linked to biological processes involving proteolysis, EGF pathways, integrin signaling, fibrosis, and inflammation. In both tissues, irradiation exposure, regardless of kidney disease status, was linked to significantly-altered SASP proteins involved in fibrosis and proteolytic pathways. Similarities in the fibrotic and proteolytic processes from significantly-altered proteins in either kidney disease or irradiation highlight a potential synergistic effect of irradiation on kidney disease. Overall results from this study will be presented. 1166

1167

P-III-1007 The NAA10-related neurodevelopmental syndrome: proteomic insights from patient-derived neuronal iPSCs Sam Calis1, Gholson Lyon2, Tom Rusielewicz3, Josephine Wesely3, Kris Gevaert1 1Ghent University, Center for Medical Biotechnology, Department of Biomolecular Medicine, Ghent, Belgium 2New York State Institute for Basic Research in Developmental Disabilities, Department of Human Genetics, Staten Island, NY, United States 3The New York Stem Cell Foundation Research Institute, New York, NY, United States A family of Nα-terminal acetyltransferase (NAT) complexes is responsible for the mainly co-translational acetylation of the nascent protein N-termini of ~80% of the human proteome. Although the functional implications of this modification remain elusive, a role in protein folding and stability has been suggested. Mutations in the catalytic subunit NAA10 of the Nα-terminal acetyltransferase A (NatA) complex give rise to series of rare X-linked developmental disorders collectively termed the NAA10-related neurodevelopmental syndrome. The symptoms observed in the predominantly male patients vary from mild to very severe mental and physical disabilities. Patient-derived induced pluripotent stem cells (iPSCs) harboring the NAA10 R83C mutation were generated, and have consequenty been re-differentiated into GABAergic neurons. The R83C mutation deactivates the acetyl CoA-binding pocket and renders the NatA complex catalytically inactive. We characterized the proteome and N-terminal acetylome of these neuronal cell lines with disturbed NatA activity and gained the disease mechanisms of this particular NAA10-related neurodevelopmental syndrome. fundamental insights in 1167

1168

P-III-1008 Exploring proteoform biomarkers to explain variability in pre-clinical cardiovascular disease risk among male and female young adults with high exposure to adverse childhood experiences Katrina Carbonara1, Deborah D. O'Leary1, Terrance J. Wade1, Adam J. MacNeil1 1Brock University, Health Sciences, St. Catharines, Canada Background: Around 15-20% of the population grow up in a toxic home environment exposed to high levels of adverse childhood experiences (ACEs), such as severe household dysfunction and maltreatment. Research indicates that high exposure to ACEs trigger physiological responses, leading to an array of later conditions including cardiovascular disease (CVD). The pathways linking ACEs to CVD likely involve lifestyle, psychosocial, and inflammatory factors that alter health trajectories through complex molecular mechanisms. While the conceptual linkage between biological, lifestyle, and psychosocial factors as determinants of health is recognized, understanding the precise mechanisms remains challenging. One potential avenue to elucidate these mechanisms is to examine overall differences in biologically active proteins (proteoforms) that link ACEs with pre-clinical markers of CVD, specifically central arterial stiffness (carotid-femoral pulse wave velocity; cfPWV). Comparing young adults with higher vs. lower cfPWV among both males and females with high ACE exposure can identify specific proteoforms that may contribute to variation in pre-clinical CVD risk markers. Methods: Using 2-dimensional gel electrophoresis (2DE) coupled with liquid chromatography and tandem mass spectrometry (LC/MS/MS), we analyzed the total serum proteoform content separately for male and female participants aged 19-26 with high ACE levels (≥4), stratified across the highest and lowest quartile of cfPWV. Targeted ELISA-based detection platforms will measure the abundance of identified proteoforms in participants with low ACE levels (0) and high ACE levels (≥4). These measurements will then be compared against previously collected data on lifestyle, psychosocial, and inflammatory factors to determine correlations. Results: Initial findings reveal over 100 resolved serum proteoforms differing significantly between high and low cfPWV groups. Moreover, a significant interaction between cfPWV and sex was identified for some proteoforms indicating potential sex-specific effects. All identifications belonged to various classes, predominantly defense/immunity, protein-binding activity modulators, metabolite interconversion enzymes, and protein modifying enzymes. Notably, some were found in more than one fixed MW and pI gel region, suggesting the presence of different proteoforms. Conclusion: These proteoform identifications offer insights into pathways and molecular mechanisms linking ACEs to CVD risk. Determining to what extent the concentrations of these markers are influenced by lifestyle, psychosocial, and inflammatory factors may enable the identification of protective and/or risk factors. This will facilitate the development of targeted protein detection tools to validate and evaluate protein-specific biomarkers in ACE-driven development of CVD, contributing to a deeper understanding of this complex relationship. 1168

1169

P-III-1009 Proteomic analysis of extracellular matrix produced by pericyte-like cells isolated from human glioblastomas Jakub Cervenka1,2, Petr Vymola1, Elena Garcia-Borja1, Eva Balaziova1, Barbora Vymolova1, Jana Veprkova1, Petr Vodicka2, Helena Kupcova Skalnikova1,2, Robert Tomas3, David Netuka4, Petr Busek1, Aleksi Sedo1 1Charles University, First Faculty of Medicine, Institute of Biochemistry and Experimental Oncology, Prague, Czech Republic 2Czech Academy of Sciences, Institute of Animal Physiology and Genetics, Libechov, Czech Republic 3Na Homolce Hospital, Department of Neurosurgery, Prague, Czech Republic 4Charles University, First Faculty of Medicine and Military University Hospital, Department of Neurosurgery and Neurooncology, Prague, Czech Republic Glioblastoma (also known as glioblastoma multiforme, GBM), is the most common and the most aggressive type of primary brain cancer with a worldwide prevalence between 2 to 5 individuals per 100,000 population. GBM mostly affects people between 45 to 70 years with a median age at diagnosis of 64 years and higher incidence in men. The overall prognosis is very poor as life expectancy for GBM patients is 12-18 months (when treatment is provided) and only four months without treatment. Less than 10% of patients survive more than five years. GBMs contain significantly higher amounts of extracellular matrix (ECM) in comparison with normal brain tissue. This ECM is enriched in fibrillar proteins such as collagen I and fibronectin and the abundance of these proteins is associated with poor prognosis of GBM patients. The highest levels of fibrillar proteins are found in perivascular niches, resided by fibroblast activation protein (FAP)-expressing pericyte-like cells. We therefore focused our proteomic analysis on the ECM produced by glioma cells and patient-derived FAP+ pericyte-like cells. Firstly, we employed data-dependent acquisition (DDA) to create sample-specific spectral library of proteins present in decellularized ECM derived from glioma cells, human brain vascular pericytes, and patient-derived FAP+ pericyte-like cells cultivated in vitro. Then, we performed quantitative measurements using data- independent acquisition (DIA) approach, followed by data analysis in Mascot, Skyline, and MSstats. In total, we quantified 3,232 distinct proteins, including about 30% of known human core matrisome proteins. The ECM produced by patient-derived FAP+ pericyte-like cells contained significantly higher amounts of fibrillar proteins, such as collagen I and fibronectin, than ECM from glioma cells. Additionally, the composition of the ECM produced by these FAP+ pericyte-like cells was markedly different from that produced by glioma cells and human brain vascular pericytes, particularly in its higher amount of basement membrane proteins. Follow-up experiments demonstrated that the ECM produced by patient-derived FAP+ pericyte-like cells supports the migration and adhesion of glioma cells in vitro, likely via activation of focal adhesion kinase (FAK) signalling. Therefore, the FAP+ pericyte-like cells seem to be critical players in GBM development and clinical outcome. This work was supported by the Ministry of Education, Youth and Sports of the Czech Republic projects National Institute for Cancer Research (Programme EXCELES, ID Project No. LX22NPO5102) funded by the European Union—Next Generation EU, EATRIS-CZ (LM2023053), the project Center for Tumor ecology (reg. n. CZ.02.1.01/0.0/0.0/16_019/0000785) supported by the Operational Program Research, Development and Education, Ministry of Health of the Czech Republic (grant NV19-03-00501), and Charles University Cooperatio Program, research area" "Oncology and Haematology" and project GA UK no. 1084120. 1169

1170

P-III-1010 Label-free proteomic analysis of P7C3-mediated protection in a rat model of Alzheimer's disease Kalyani Chaubey1,2,3,4, Edwin Vázquez-Rosa1,2,3,4, Jaymie Voorhees5, Kathryn Franke1,2,3,4, Coral Cintrón-Pérez1,2,3,4, Sonny Caradonna1,2,3,4, Feixiong Cheng6,7,8, Bindu Paul9,10,11,12, Andrew Pieper1,2,3,4 1Case Western Reserve University, Departments of Psychiatry, Neurosciences, and Pathology,, Cleveland, OH, United States 2Harrington Discovery Institute, Brain Health Medicines Center, Cleveland, OH, United States 3Case Western Reserve University, Institute of Transformative Molecular Medicine, Cleveland, OH, United States 4Louis Stokes Cleveland VA Medical Center, Geriatric Psychiatry, GRECC, Cleveland, OH, United States 5Inotiv, St. Louis, MO, United States 6Cleveland Clinic, Genomic Medicine Institute Lerner Research Institute, Cleveland, OH, United States 7Case Western Reserve University, Department of Molecular Medicine Cleveland Clinic Lerner College of Medicine, Cleveland, OH, United States 8Case Western Reserve University School of Medicine, Case Comprehensive Cancer Center, Cleveland, OH, United States 9Johns Hopkins University School of Medicine, Department of Pharmacology and Molecular Sciences, Baltimore, MD, United States 10Johns Hopkins University School of Medicine, The Solomon H. Snyder Department of Neuroscience, Baltimore, MD, United States 11Johns Hopkins University School of Medicine, Department of Psychiatry and Behavioral Sciences, Baltimore, MD, United States 12Lieber Institute for Brain Development, Baltimore, MD, United States Alzheimer's disease (AD) is a severe neurodegenerative condition that affects millions of people worldwide and currently lacks disease-modifying treatments. Our observations have shown that AD patients with lower brain nicotinamide adenine dinucleotide (NAD+)/NADH levels exhibit strong correlations with several secondary neurodegenerative changes, including blood-brain barrier (BBB) damage, oxidative stress, neuroinflammation, neurodegeneration, and impaired synaptic function. We have found that early treatment with the nicotinamide (NAD+)/NADH-normalizing P7C3-series of aminopropyl carbazoles prevents the adenine dinucleotide development of AD-like pathology and behavioral impairments in animal models of AD. Additionally, normalization of brain NAD+/NADH in symptomatic advanced-disease animal models enables the brain to recover from secondary pathological events, including oxidative stress, neuroinflammation, DNA damage, BBB deterioration, neurodegeneration, and impaired cognitive function. While this has provided a first proof of principle for therapeutic recovery from AD, the underlying mechanisms of how normalization of NAD+/NADH achieves these effects are not yet known. To investigate this question, we conducted an unbiased label-free proteomic analysis of TgF344 AD rat brains from animals that were treated daily with either P7C3 compounds or vehicle for 6 months. This analysis identified 855 differentially expressed proteins in AD vehicle rats compared to WT littermate vehicle rats. Notably, 125 of these proteins returned to normal WT levels in the AD animals that were treated with P7C3 compounds. Pathway analysis of these 125 proteins revealed enriched inclusion in pathways related to oxidative phosphorylation, glycolysis, tricarboxylic acid cycle, autophagy, tight junction proteins, and steroid and bile acid biosynthesis. This information provides important direction towards identifying physiologic mechanisms that may be therapeutically targeted in AD. 1170

1172

P-III-1011 Extracellular LGALS3BP: a potential contributor in glioma progression and can serve as biomarker for early diagnosis Kirti Chauhan1, Rashmi Rana1 1Sir Ganga Ram Hospital, Department of Biotechnology & Research, New Delhi, India Background: Gliomas are the most prevalent form of malignant brain tumor. They account for 70% of all primary malignant brain tumors. The only procedures available for diagnosis of grades are surgical incision and biopsy; however, our work focuses on finding a minimally invasive way for early glioma diagnosis. Identification of proteins released by cancer cells is of particular interest in this area, as it could lead to a better understanding of tumor growth. Methods: Extracellular vesicles (EVs) were extracted from pooled plasma of healthy individuals and glioma patients of various grades (Grade I, II, or III). The size and concentration of Plasma derived-EVs marker were determined using nanoparticle tracking analysis, western blot, and flow cytometry. In furthermore, an iTRAQ- based LC-MS/MS analysis of EVs protein was performed followed by validation of identified marker through ELISA, Immunohistochemistry, Immunocytochemistry and flow cytometry. Results: Total 123 proteins were identified from plasma derived pooled EVs. A total of seven proteins—namely, CRP, SAA2, SERPINA3, SAA1, C4A, LV211, and LGALS3BP — showed differential abundance in all the three grades. LGALS3BP is the only protein which found to be strikingly high in all three grades in a progressive manner. Conclusion: LGALS3BP was shown to be elevated across all grades, the enhanced expression of LGALS3BP further validated in glioma patients (Tissue & plasma). That shows the possible biomarker for early diagnosis of glioma and can influence the treatment for the improvement of patient survival. This study would provide the information about LGALS3BP involvement in glioma progression that will lead to better pathogenesis of tumor condition.. Keywords: LGALS3BP, Biomarker, Extracellular Vesicles, Gliomas, ICC, IHC 1172

1173

P-III-1012 The identification and mechanism exploration of GDF15 proteoforms in aging Yuling Chen1, Jinyu Li1, Haiteng Deng1 1Tsinghua University, School of Life Science, Beijing, China With the global shift towards aging populations, the prevalence and economic burden of aging-related diseases are escalating. It is imperative to develop prediction biomarkers and intervention targets of aging-related diseases to enhance the overall healthy lifespan of the people. GDF15 stands out as a prominently upregulated protein in aging processes and various age-related diseases. This protein also falls under the category of senescence-associated secretory phenotype (SASP), which is strongly correlated with increased mortality risk. As a member of the TGFβ superfamily, GDF15 is produced by a variety of immune and tissue cells in response to a broad range of stressors. It can act through the GFRAL receptor expressed in the hindbrain to suppress appetite and induce weight loss. Extensive research has revealed GDF-15's immunomodulatory effects on macrophages, mitigating inflammation by discouraging M1-like macrophage polarization while encouraging M2-like polarization. However, the pathophysiological role of GDF15 during the occurrence and development of aging- related diseases remains poorly understood. In our experiments we found that GDF15 precursor (pro-GDF15), instead of mature GDF15, was significantly increased in the plasma of old people and mice. Through mass spectrometry based targeted protein quantification, we examined the distribution of pro-GDF15 and mature GDF15 within SASPs of senescent cells under various induced conditions, as well as in plasma samples across different age groups. To further our understanding, we treated M0 macrophages with purified pro-GDF15 and mature GDF15, unearthing that pro-GDF15 exhibits a more robust anti-inflammatory impact compared to its mature counterpart. Interactome analysis further demonstrated that pro-GDF15 interacts with gap junction proteins, while mature GDF15 plays a more significant role in macrophage glucose and fatty acid metabolism. These discoveries shed new light on the functions and mechanisms of GDF15 at the proteoform level in the aging process, setting the stage for prospective clinical trials exploring GDF15-based anti-aging interventions. 1173

1174

P-III-1013 Sleep deprivation disrupts energy and lipid dynamics via the brain-brown adipose axis Peipei Chen1, Heyu Zhao1, Zhili Huang1, Huali Shen1 1Fudan University, Shanghai, China in synaptic functions and neurotransmitter Humans allocate approximately one-third of their lifespan to sleep. Chronic sleep deficiency is implicated in a spectrum of health complications and societal issues, and can precipitate grave pathophysiological outcomes, including mortality. Fundamental physiological processes such as cognition, immune function, and metabolism are reliant on adequate sleep. Despite its critical role, the precise functions and underlying mechanisms of sleep remain subjects of ongoing scientific debate.We established models of acute and chronic sleep deprivation in mice and found the activation of dopaminergic neurons under conditions of sleep deficiency. Utilizing proteomics, we identified dysregulated proteins in the NAc/SN/VTA brain regions' dopaminergic neurons following acute and chronic sleep deprivation. Bioinformatics analyses indicated significant changes of the dopaminergic neurons in the VTA region; following acute deprivation, the changed proteins were primarily involved transport, while chronic deprivation shifted proteins involving fatty acid metabolism, energy metabolism, and thermogenesis. Further study indicated that chronic sleep deprivation enhances respiratory entropy, oxygen consumption, basal metabolism, and thermogenesis, while no significant changes were observed after acute deprivation. Chronic deprivation also resulted in decreased adipose tissue mass and altered lipid metabolism. Neurotracing and co-localization immunostaining confirmed trans-synaptic connections between dopaminergic neurons in the VTA and cells within the brown adipose tissue. Overexpression of the Eci1, a differential protein discovered in the proteomic experiments, in VTA dopaminergic neurons decreased sleep duration and increased respiratory entropy in mice.Overall, our findings delineate the proteome changes in dopaminergic neurons under acute and chronic sleep deprivation and reveal that the Eci1 molecule in VTA dopaminergic neurons orchestrates the metabolic imbalance induced by chronic sleep deprivation. We further explored that the brain-fat axis directly regulates lipid metabolism disorders in sleep deprivation. Our study thus demonstrated how the brain encodes sleep deprivation caused metabolic dysregulation. 1174

1175

P-III-1014 Proteomics into pathological mechanisms of Chronic Progressive External Ophthalmoplegia (CPEO): implications of the pathogenic variability Michela Cicchinelli1,2, Guido Primiano2,3, Serenella Servidei2,3, Andrea Urbani1,2, Federica Iavarone1,2 1Catholic University of the Sacred Heart, Department of Basic Biotechnological Sciences, Intensive and Perioperative Clinics, Rome, Italy 2Fondazione Policlinico Universitario Agostino Gemelli IRCCS, Rome, Italy 3Catholic University of the Sacred Heart, Department of Neurosciences, Rome, Italy Chronic Progressive External Ophthalmoplegia (CPEO) is a primary mitochondrial myopathy (PMM) characterized by defects in oxidative phosphorylation (OXPHOS) and marked by genetic and clinical heterogeneity [1]. Defective OXPHOS leads to a lack of ATP production, compromising tissues with high-energy demand. This condition primarily affects muscles controlling eye and eyelid movement, resulting in ptosis, ophthalmoplegia, and myopathy. The clinical presentation of CPEO can also include impairment of multiple organs and varying degrees of systemic involvement. CPEO can arise from direct mitochondrial DNA (mtDNA) variants, such as single large-scale mtDNA deletions, or nuclear DNA (nDNA) mutations secondarily affecting mtDNA function [2]. Single large-scale deletions can vary in size and location, and remove multiple mtDNA genes essential for mitochondrial function. Secondary mtDNA defects are due to mutations in nDNA genes with a critical role in mtDNA replication and maintenance, such as POLG and TWNK. POLG encodes the catalytic subunit of mitochondrial DNA polymerase, and its mutations impair mtDNA replication and repair. TWNK mutations, which encodes a helicase essential for mtDNA replication, lead to mtDNA instability and multiple deletions. Understanding the pathogenic mechanisms of CPEO remains a challenge due to factors such as its genetic basis and the spectrum of symptoms, which result in inter-individual phenotypic variability. The aim of the study was to explore the urinary proteome of 10 CPEO patients with pathogenic variants in POLG and TWNK or single large-scale mtDNA deletions, sex and age matched with 10 healthy controls to individuate alteration in the proteomic profile related to pathological and mutation-dependent metabolic mechanisms. A bottom-up proteomic approach was performed with an Orbitrap Fusion Lumos Tribrid mass spectrometer (Thermo Scientific) followed by a label-free quantification. The exploratory phase, conducted with a Principal Component Analysis (PCA), showed a clear separation between CPEO patient group and the control group, and a mutation-guided sub-grouping within CPEO. Differential expression and pathways analyses confirmed this evidence, revealing how the clustering was driven by proteins and pathways differing according to mutation. The proteomic data obtained documented significant metabolic differences attributable to the pathogenic mechanisms in the distinct mutation CPEO sub-groups and how the proteomic approach could represent a valid contribution in unveiling the mechanisms of this specific mitochondrial disease. References [1] Mancuso, M.; et al. International Workshop : Outcome Measures and Clinical Trial Readiness in Primary Mitochondrial Myopathies in Children and Adults. Consensus Recommendations. 2017, 27, 1126–1137. [2] Alston, C. L. et al. The Genetics and Pathology of Mitochondrial Disease. Journal of Pathology. 2017, 241, 236–250. 1175

1176

P-III-1015 Psychedelic-induced proteome changes in rodent brain regions Dominik Cysewski1, Tomasz Kowalczyk1, Kacper Łukasiewicz1, Arkadiusz Zbikowski1, Napoleon Waszkiewicz1, Adam Kretowski1, Michał Ciborowski1 1Medical University of Białystok, Laboratory of Metabolomics and Proteomics, Białystok, Poland In recent decades, significant progress has been made in the field of psychedelics research. However, the short- and long-term cellular responses to psychedelics at the level of proteome remodeling remain largely unexplored. In this study, we utilized a rodent model to investigate alterations in the proteome of four brain regions involved in the response to psychedelic administration, including psilocybin. Psychedelic substances, including psilocybin, were administered to rodents at doses consistent with those reported in the literature as acute dosages, mirroring their application in psychotherapy. Animals were sacrificed, and their brains were promptly extracted, snap-frozen, and sectioned using a cryotome, and the selected brain regions were isolated. The tissues were processed, digested, and analyzed using an ultra-high- performance liquid chromatography (uHPLC) system coupled with a ThermoScientific Tribrid Orbitrap mass spectrometer. Data was collected in data-independent acquisition (DIA) mode and analyzed using PEAKS and FragPipe software. Our preliminary data highlight specific changes in protein abundance that are intricately tied to distinct brain regions. Notably, these alterations are predominantly observed in structural and membrane proteins, underscoring their significance in the context of our study. The lack of comprehensive proteomics research in this area emphasizes the need for further investigations to enhance the identification of membrane proteins within our samples. An intriguing aspect of our findings is the annotation of proteins exhibiting increased levels, which suggests a potential induction of neural plasticity. These observed changes may be linked to associations with receptors commonly involved in long-term potentiation (LTP), a crucial mechanism underlying learning and memory processes. Furthermore, these proteins may interact with other molecules integral to neural plasticity, indicating a complex network of interactions that warrant further investigation to elucidate their precise role in psychedelics-induced alterations in brain proteomics. 1176

1177

P-III-1016 FACS-Proteomics strategy toward extracellular vesicles single-phenotype characterization in biological fluids: insights into Leukocyte-derived EVs protein cargo and immunity Maria Concetta Cufaro1, Alice Di Sebastiano2,1, Ilaria Cicalini1, Marianna Gabriella Rispoli3, Giulia Catitti1, Domenico De Bellis1, Serena Pilato2, Antonella Fontana2, Valentina Tomassini3, Serena Veschi2, Alessandro Cama2, Paola Lanuti1, Luca Federici1, Damiana Pieragostino1, Piero Del Boccio2,1 1University "G. d'Annunzio", Center for Advanced Studies and Technology (CAST), Chieti, Italy 2University, Pharmacy, Chieti, Italy 3University "G. d'Annunzio", Department of Neurosciences, Imaging and Clinical Sciences, Chieti, Italy Background: it is now established that Extracellular Vesicles (EVs) display potent regulatory functions in biological systems. Circulating EVs can enable extreme reduction of the high dynamic range of detectable protein concentration in whole biofluids, paving the way for the biomarker discovery in the so-called "dark proteome". The best strategy of EVs isolation from biofluids remains an open question especially considering their heterogeneity in terms of phenotype and their suitability for proteomic application, once purified. Within such a complex and dynamic scenario, we have recently optimized an innovative "FACS-Proteomics" workflow for the isolation and subsequent proteomics characterization of EVs directly from untouched biological fluids taking advantage of a lipophilic cationic dye (LCD) able to probe intact EVs. This purification method offers the great opportunity to carry out proteomics on circulating and specific EV populations by sub-typing them with an appropriate panel of antibodies. Methods: FACS strategy was used to purify Leukocyte-derived EVs (Leuko EVs, LCD+/CD45+/Phalloidin- events) from blood and tears of healthy volunteers and patients affected by Multiple Sclerosis (MuS). After purification, Atomic Force Microscopy and Nanoparticle Tracking Analysis were used to investigate Leuko EVs size distribution, morphology and number in both blood and tears extracts. Label free proteomics approach was carried out by NanoLC-Orbitrap-Fusion-Tribrid Mass Spectrometer; raw data were processed by MaxQuant and loaded on Ingenuity Pathway Analysis (IPA) for functional investigation. Results: FACS isolated CD45+ EVs from blood and tears (1.4x105 and 2x106 events, respectively) displayed a high level of purity (>90%) and more than 90% of proteins (78 in tear and 155 blood) identified in both biological fluids were mainly referred to the leukocyte phenotype. The same workflow was applied to carry out proteomics on purified Leuko EVs from tears of Multiple Sclerosis (MuS) patients, demonstrating that they were enriched with an encoded protein cargo, reflecting the neuroinflammatory condition of MuS together with an increase of vascular networks (AGNPT2 and VEGF, upstream regulators activated) which reflects pro-angiogenic processes predicted as downstream effects. As leuko-EVs are still a heterogeneous group, we investigated it further, analysing the protein cargo of EVs released from neutrophils (the most abundant sub-group of white cells) and their activated counterpart the polymorphonuclear myeloid-derived suppressor cells (PMN-MDSC) (LOX1±), that in tourn seem to act in counteracting neuroinflammation although their role is still controversial. Our data showed that the combined FACS-Proteomics strategy for EV single phenotype characterization could have groundbreaking avenue for biomarker discovery, exalting the clinical value of tears EVs and helping in a better understanding of the EV-mediated processes in vivo. 1177

1178

P-III-1017 Proteolytic signatures of coagulation identified and monitored by plasma peptidomics Jessica Del Castillo-Alferez1,2, Alexander B. Meijer2, Herm-Jan Brinkman2, Arjan Hoogendijk2, Joost C. M. Meijers3, Maartje van den Biggelaar2, Tirsa van Duijl2 1Utrecht University, Utrecht, Netherlands 2Sanquin Research, Molecular Hematology, Amsterdam, Netherlands 3Amsterdam University Medical Center, Experimental Vascular Medicine, Amsterdam, Netherlands Background: Coagulation comprises a network of proteolytic processes initiated by tissue factor (TF) that regulate the function of plasma proteins and result in the formation of a fibrin clot. Altered proteolysis of coagulation proteins is implicated in hemostatic and fibrinolytic disorders. Here, we employed plasma peptidomics to identify the proteolytic products of coagulation and subsequently monitor the impact of tissue factor concentrations and thrombin inhibition in proteolytic processing. Methods: Pooled citrated plasma from healthy donors was clotted in vitro by recalcification and addition of increasing concentrations of tissue factor (TF), in the absence or presence of a thrombin inhibitor hirudin. Endogenous peptides were enriched by solid phase extraction (SPE), analysed by LC-MS/MS in DDA mode (Orbitrap Fusion Lumos) and identified using PEAKS X. Thrombin generation in clotting plasma was measured using the Calibrated Automated Thrombogram (CAT) assay with a microtiter plate reading fluorometer (Fluoroskan Ascent, ThermoLab systems) and Thrombinoscope® software (Thrombinoscope BV). Results: Plasma coagulation revealed a distinct and richer endogenous peptidome in which fibrinopeptides A and B were most abundant. Unexpected proteolytic hotspots were identified in the C-terminal region of fibrinogen α chain. The activation peptides of coagulation factors prothrombin (316TAT-DGR363) and FXIIIA (6TAF-VPR38) were recognized as well as their proteolytic degradation products. Peptides derived from protease inhibitors mapping within their internal bait (e.g. 705VGF-MGR715 of alpha 2 macroglobulin) or C-terminal regions (e.g. 426SLN-CVK464 of antithrombin) were observed exclusively upon coagulation and were sensitive to increasing TF concentrations. Thrombin inhibition blocked the formation of coagulation activation peptides and the proteolytic processing of hemostatic regulators, but not inflammatory proteins such as acute phase SAA4. Finally, quantitative monitoring of coagulation with calibrated automated thrombograms showed rapid cleavage of fibrinopeptides correlated with initial thrombin generation followed by cleavages in prothrombin, FXIIIA, and protease inhibitors. Conclusion: Plasma peptidomics exposed a distinct endogenous peptidome upon clotting, enriched with activation peptides from zymogen-to-enzyme transitions and proteolytic products of protease inhibitors. Monitoring changes in the peptidome under different coagulating conditions provided insights into thrombin and tissue factor sensitive proteolytic processing in plasma. This peptidomics strategy can be leveraged to study proteolytic processing of protease networks in bleeding or thrombosis. 1178

1179

P-III-1018 Spatial lipidomics via High Resolution MALDI-MSI highlights a cerebellum-specific molecular signature in the Marinesco-Sjogren mouse model Fabio Di Ferdinando1, Fabio Bellia1, Ilaria Cicalini1, Maria Concetta Cufaro1, Silvia Valentinuzzi1, Michele Sallese1, Vincenzo De Laurenzi1, Piero Del Boccio1, Damiana Pieragostino1 1University of Chieti-Pescara, Center for Advanced Studies and Technology (CAST), Chieti, Italy Background: Matrix-Assisted Laser Desorption/Ionization Mass Spectrometry Imaging (MALDI-MSI) can guide a more comprehensive knowledge in terms of identification, abundance, and distribution of biomolecules within tissues. The hyphenation of such ionization technique with the High Resolution Orbitrap MS analyzer improved performance and made the discovery phase more robust and reliable. Furthermore, the non-destructiveness of MALDI source enables the recycling of intact tissue slices for staining techniques to drive a comparison of structural features with the MS data. The Marinesco-Sjogren Syndrome (MSS) is an infantile-onset disease mainly characterized by ataxia, cerebellum atrophy and muscle weakness linked to a loss-of-function mutation of the gene Sil1 encoding for a nucleotide exchange factor of BiP3, a crucial Endoplasmic Reticulum (ER) chaperone. Although Sil1 being widely expressed in the brain, the mutation cause protein accumulation, chronic activation of the Unfolded Protein Response (UPR) and cell death only in the cerebellum Purkinje cells and even more specific it spares the vestibulocerebellum which include lobule X and the caudal region of lobule IX. Given that MALDI-MSI can play a crucial role to understand the reason of the mutation cell-specific effect showing eventual metabolic and signaling rewiring allowing some Sil1-deficient cells to avoid the neurodegeneration. We carried out a high resolution spatial lipidomics on the Marinesco-Sjogren Syndrome (MSS) mouse model woozy (wz) cerebellum. Methods: The MassTech™ Atmospheric-Pressure MALDI source with Ultra-High spatial Resolution (UHR) has been coupled with an Orbitrap Exploris 120 mass spectrometer for high resolution in situ analysis on cerebellum slices collected from homozygote and heterozygote mice for wz mutation, the latter considered the control. The analysis was conducted on both positive and negative ionization mode and the data were processed with LipostarMSI software. Results: An evaluation of the optimal sample preparation of mouse cerebellum tissues for MALDI-MSI-based lipidome analysis was conducted, in terms of embedding polymer, size of the raster, matrix and impact of aqueous wash before matrix deposition. Then, we applied the optimized method to compare the lipid signature of cerebellum between the two conditions for both ionization modes. The analysis of the cerebellar arbor vitae revealed a significative difference in the spatial distribution of many lipids between the two conditions and among the different areas of the same cerebellum coherent with the literature and the phenotype observed. These preliminary data pave the way for a better comprehension of the metabolic adaptation that cells pursue to ease the ER stress and avoid the neurodegeneration. 1179

1180

P-III-1019 Toward precision serology: resolving human polyclonal antibody repertoires with proteomics Zoe Turner1, Yasmine Rais1, Weize Tang1, Andrei Drabovich1 1University of Alberta, Laboratory Medicine and Pathology, Edmonton, Canada Introduction: Human polyclonal antibodies constitute a highly diverse and dynamic universe of clonotypes, isotypes, and subclasses circulating in serum and contributing to immune response. Traditionally, the analysis of human polyclonal antibodies has relied on indirect immunoassays which suffered from semi-quantitative measurements and lack of standardization. Recent advances in mass spectrometry (MS) and immunoaffinity proteomics, including our studies [1-4], facilitated selective, sensitive, and reproducible quantification of low- abundance proteins in serum and proximal fluids. Here, we will introduce a proteomic toolbox for the quantification and characterization of endogenous human polyclonal antibodies. We will demonstrate the complexity of the antibody response to viral antigens and cancer auto-antigens and discuss pathways toward Precision Serology. Methods: The proteomic toolbox included shotgun, targeted and intact protein MS assays to identify, quantify and sequence human polyclonal antibodies [5]. High-throughput immunoenrichments combined with targeted MS and carefully designed standards enabled absolute quantification of antibody isotypes (IgG, IgA, IgM, IgE, IgD) and subclasses (IgG1-4, IgA1-2) circulating in patient sera (>100 samples/day). Restricted proteolytic cleavage generated Fab, Fd, and L fragments for intact protein MS. Alternative proteases enabled increased sequence coverage of repertoires of antibody variable chains. Results: Significantly elevated levels of anti-SARS-CoV-2 RBD IgG1 (1.6 µg/mL), IgG3, IgM and IgA1 in patient sera (N=231) suggested a conventional Th1 immune response [4]. Respiratory Syncytial Virus antigens revealed a mixed Th1/Th2 immune response with elevated levels of IgG1 (2.6 µg/mL), IgM, IgA1, IgG2 and IgG4 in sera (N=69). The unexpected discovery of antigen-specific secreted IgD suggested a distinct immune response in certain infections. Profiling of the repertoire diversity of the heavy variable chains revealed the frequent use of only ~30 IGHV genes. In prostate cancer, a panel of prostate-specific antigens demonstrated distinct IgG1- and IgA1-mediated auto-antibody response in serum (N=50), with anti-PSA IgG1 found in 70% of cancer patients at 3-14 ng/mL. Conclusions: Our proteomic toolbox will facilitate quantification and characterization of the endogenous human polyclonal antibodies, antibody sequencing, investigation of immunoglobulin subclass cooperation in immune response, precision approaches in serology diagnostics, and immunology studies of infectious diseases and cancer [6]. References: [1] Fu Z… Drabovich AP. Mol. Cell. Proteomics 2021, 20, 100075; [2] Zhang J… Drabovich AP. Mol. Cell. Proteomics 2023, 22, 100556; J. Proteome Res. 2024, 10.1021/acs.jproteome.4c00027; [4] Fu Z… Drabovich AP. Anal. Chem. 2022, 94, 12990; [5] Wize T, Drabovich AP. bioRxiv 2023, 10.1101/2023.10.27.564451; [6] Walter J… Drabovich AP. Clin. Proteomics 2023, 20, 42. Y, Drabovich AP. [3] Rais 1180

1181

P-III-1020 Proteomic characterization of acquired PARP inhibitor resistance in high-grade serous ovarian cancer Jolene Duda1, Xiaonan Hou2, Ann Oberg2, Krista Goergen2, John Weroha2, Stefani Thomas1 1University of Minnesota, Minneapolis, MN, United States 2Mayo Clinic, Rochester, NY, United States High-grade serous ovarian cancer (HGSOC) is the deadliest gynecologic malignancy in women. The incorporation of poly (ADP-ribose) polymerase inhibitors (PARPi) into the HGSOC treatment regimen can increase progression- free survival; however, most patients develop PARPi resistance, limiting the use of this targeted therapy. Therefore, it is imperative to further elucidate the mechanisms driving PARPi resistance in HGSOC. In this project, we sought to elucidate the proteomic alterations that occur with the development of PARPi resistance using patient-derived xenograft (PDX) mouse models of HGSOC. The PDX models used in this study were homologous recombination deficient (HRD) via a BRCA2 mutation (PH077) or a Rad51C methylation (PH039) and had acquired PARPi resistance through repeated exposure to Niraparib. In the PH077 model, 5,231 proteins were quantified across all tumors (PARPi sensitive (n=3) and resistant (n=2)) based on label-free quantification. Global proteomic analysis resulted in the identification of 32 proteins with significant differences in relative abundance when comparing the sensitive and resistant tumors (fold-change > 2; p < 0.05). Interestingly, Principal Component Analysis revealed that the PARPi sensitive tumors exhibited a higher degree of proteomic similarity, whereas the resistant tumors did not exhibit this characteristic. This suggests that acquired PARPi resistance is tumor-specific and does not occur via common pathways.When carrying out a grouped analysis, we also identified proteins that were unique to only the sensitive or resistant tumors. Significant biological pathways found to be enriched from these unique proteins include: oncogenic MAPK signaling and sumoylation (sensitive), RNA processing and IL-12 signaling (resistant). To investigate the functional role of HDACi in sensitizing HGSOC tumors to PARPi treatment, next, we explored the differential abundance of HDACs within the PARPi sensitive and resistant groups. Increased HDAC6 expression is associated with poor prognosis and drug resistance in HGSOC. Differential abundances of HDACs 1,2,3,6 and 7 were observed in the HGSOC PDX mice. The relative abundance of HDAC6 was significantly increased in one of the PARPi resistant tumors compared with two of the sensitive tumors, suggesting that decreased levels of lysine acetylation are functionally implicated in refractoriness to PARPi treatment. Ongoing work is includes analysis of the PH039 model and a differential analysis of lysine acetylation modifications between the PARPi sensitive vs. resistant tumors using acetyl-lysine antibody-based peptide enrichment to further elucidate the molecular mechanisms implicated in the responsiveness to PARPi treatment. The results from this study will lead to a more thorough understanding of the mechanisms driving drug response in HGSOC towards an overall goal of establishing therapeutic vulnerabilities to improve patient survival. Fig. 1 (optional) 1181

1182

P-III-1021 Unraveling the proteomic and phosphoproteomics early events molecular effectors and signaling pathways of the Anti-CD19 CAR-T cell therapy John Teibo1,2, Virginia Silvestrini1,2, Izadora Archiolli1,2, Ana Paula Masson1, Renata Silvestre2, Virginia Picanço Castro2, Lucas Botelho de Souza2, Dominic Helm3, Raja S. Nirujogi4, Dario R. Alessi4, Vitor Faça1,2 1University of São Paulo, Biochemistry and Immunology - Ribeirão Preto Medical School, Ribeirão Preto, Brazil 2University of Sao Paulo, Center for Cell Based Therapy, Ribeirão Preto, Brazil 3German Cancer Research Center (DKFZ), Proteomics Core Facility, Heildelberg, Germany 4University of Dundee, Protein Phosphorylation and Ubiquitylation Unit, Dundee, United Kingdom Chimeric Antigen Receptor T-cell (CAR-T) therapy is a leading therapy in the field of cell immunotherapy, which utilizes genetically engineered T-lymphocyte cells expressing receptors against specific targets on tumor cells, such as CD-19, to induce cytotoxicity and the death of malignant cells. Using a locally-developed third-generation CAR-CD19, we generated an Anti-CD19 CAR-Jurkat, a T-cell line, which we used to characterize early events of proteomic and phosphoproteomic alterations important for understanding the signaling pathways and mechanism of action for this CAR. Anti-CD19 CAR-Jurkat and SILAC-heavy labeled RAJI B-cells were co-cultured for 10 minutes. The proteomic and phosphoproteomic profiles were obtained using the DIA approach on the Orbitrap Astral LC-MS/MS platform. The data were processed by DIA-NN and organized to discriminate the proteins and phosphopeptides from the effector CAR-Jurkat and the target RAJI cells based on the SILAC isotopically labeled peptides. The analysis of six biological replicates provided very deep coverage of the proteome and phosphoproteome, resulting in the identification and quantification of approximately 8,800 proteins and 20,000 phosphorylation sites, respectively. As expected, there were minor early proteomic alterations in the effector CAR-Jurkat cells, but few statistically significant alterations indicated special processes of antigen presentation by CD74 and other MHC molecules. On the other hand, the target RAJI B-cells had more statistically significant changes, including alterations in adhesion molecules and CD28, which are responsible for T-cell activation and survival. Additionally, other proteins with undisclosed direct connections with T-cell engagement were observed, which might represent interesting molecules that modulate the CAR response. Regarding the phosphoproteomic alterations, a large number of regulated phosphosites were observed for both the effector CAR-Jurkat and target RAJI cells. Interestingly, for both cells, a network analysis indicated a very significant alteration in proteins associated with splicing and the spliceosome. While the association of splicing and T-cell activation has been recently described, it is striking how early these events take place in CAR-T cell function. Based on our study, we uncovered proteins and phosphosites that play roles as molecular effectors of Anti-CD19 CAR-T cell therapy at early stages of CAR-T-target cell engagement and are expanding the understanding of the mechanism and signaling pathways that will aid the further development of CARs and improve the efficacy of such a revolutionary cancer therapy strategy. Keywords: CAR-T cell, Proteome, Phosphoproteome, Signaling pathways, Molecular Effectors. Support: CNPq (316934/2021-6) FAPESP (2020/13463-2, 2013/08135-2), RSC (C23-6685539336). 1182

1183

P-III-1022 Ultrasensitive profiling of > 300 plasma proteins for predicting biological age Xiao-Jun Ma1, Niyati Jhaveri1, Li Wang1, Aparna Sahajan1, Karl Garcia1, Tsz Tam1, Sean Kim1, Henry Huang1, Jesse R. Poganik2, Mahdi Moqri2, Dane Gobel3, Seth Paulson3, Vadim N. Gladyshev2, Dwight Kuo1, Bingqing Zhang1, Yuling Luo1 1Alamar Biosciences, Inc., Fremont, CA, United States 2Methuselah Foundation, Springfield, VA, United States 3Brigham and Women’s Hospital, Harvard Medical School, Division of Genetics, Department of Medicine, Boston, MA, United States Biological aging underlies almost all major chronic diseases such as cancer, autoimmune diseases and Alzheimer’s disease. Elucidating the biological mechanisms of aging and monitoring the aging process noninvasively holds great promise for the early detection and ultimately the prevention of aging-related diseases. Current approaches to quantify biological age have mostly focused on epigenetic changes in the genome. Advances in proteomic technologies can provide an alternative and arguably more actionable measure of biological aging at the protein level. Here we apply the recently developed Nucleic acid-linked immunosandwich assay (NULISA) to measure >300 proteins to assess age-related changes in plasma protein levels. In this study, we profiled a cohort of 500 individuals spanning over 7 decades of chronological age from a biobank maintained by a major metropolitan academic medical center. The cohort consisted of an even balance of male and female subjects and is generally representative of the racial/ethnic distribution of the collection site. Plasma samples from this cohort were assayed with the NULISAseq 250-plex Inflammation and 120-plex CNS Disease Panels targeting key inflammation-related cytokines and chemokines and proteins involved in all major hallmarks of neurodegeneration such as pTau and β-amyloid proteins. Overall target detectability in plasma samples from this cohort was 98.8% with the Inflammation Panel and 96.8% with the CNS Disease Panel. Preliminary bioinformatics analyses identified >150 targets from the Inflammation Panel that were increased with age including proinflammatory markers such as IL6, TNFα, LIF, TGFβ and IL18. Analysis with the CNS markers further identified age-dependent increases in pTau-217, pTau-181, pTau-231, Aβs (38, 40 and 42) as well as NEFL and NEFH among many other markers implicated in neurodegeneration. Additional analyses are planned to build a protein-based age predictor and assess its correlation with clinical outcomes to predict mortality and common diseases. In conclusion, this study applied the recently developed NULISA platform to profile key inflammation and neurodegeneration-related proteins in plasma in a well curated clinical cohort. The large number of proteins identified to be associated with age indicates that this dataset may provide a rich and high-quality dataset for the aging research community. 1183

1184

P-III-1023 Investigation of Parkinson's disease heterogeneity and driving factors based on DIA proteomics analysis of patient PBMCs Dani Flinkman1, Ali Mostafa1, Ye Hong1, Veronica Fagerholm1, Prasannakumar Deshpande1, Kari Lindholm2, Valtteri Kaasinen2, Peter James3, Eleanor Coffey1 1Turku Bioscience, University Of Turku And Åbo Akademi, Turku, Finland 2Turku University Hospital, Turku, Finland 3Lund University, Lund, Sweden Introduction: Parkinson's disease (PD) is the second most common and the fastest growing neurodegenerative disease. The disease is characterized by loss of dopaminergic neurons in mid brain and debilitating motor symptoms. Non-motor symptoms such as loss of smell, constipation and sleep disturbances are also present, and their appearance precede the motor symptoms by years. Cause of the disease is unknown, but it is thought that environmental factors, risk factors such as head injury, as well as disease causing genetic variants and proteins work together in pathogenesis of PD. Roughly 10% of individuals with PD carry a mutation and ~15% have a PD family history. The most common mutations are in genes coding for Leucine-rich repeat kinase 2 (LRRK2) and Glucocerebrosidase (GBA). Significant heterogeneity exists in PD progression, motor and non-motor symptoms and biomarkers. To better understand the heterogeneity and how proteins contribute to disease, progression we seek to find biomarkers and biology behind the PD using peripheral blood mononuclear cells (PBMC) from two independent PD patient cohorts. These contain samples from sporadic and genetic PD patients, motor disease patients without PD diagnosis (nonPD), prodromal individuals +/- disease-associated mutations and healthy individuals. Methods: PBMC samples in the first cohort were obtained from South-Western Finland catchment area (32 PD, 25 nonPD and 32 healthy). Second cohort is from Parkinson's Progression Marker Initiative consortium (PPMI) from 180 sporadic PD, 72 GBA+PD, 91 LRRK2+PD and 79 healthy individuals. Additionally, PPMI cohort includes 53 prodromal individuals without a mutation, 151 GBA and 139 LRRK2 mutation carriers without PD diagnosis. Total number of samples from PPMI cohort is 1181 with up to three time points per individual. High throughput digestion with S-trap™ was used to generate tryptic peptides. Each digestion batch contained two digestion controls from a standard sample. Resulting tryptic peptides were analysed with Evosep 30 sample per day method, with a daily standard sample on Oribitrap Lumos interfaced with FAIMS-pro with a single compensation voltage, in data independent acquisition mode. Results: MS data analysis is performed with Spectronaut® 18 yielding information on ~6700 protein groups after filtration and batch correction. PTMs that are quantifiable without PTM enrichment are also analysed. Statistical analysis is done with an in house developed MS data analysis tool PhosPiR. Weighted correlation network analysis is used to identify patient subgroups and protein modules, with clinical feature association. Preliminary findings will be shown at the poster. Conclusions: Insight into PD disease progression and what are the disease driving changes at the prodromal stage were obtained with associated clinical features. The results are also useful for multiomic analysis of the same individuals from PPMI cohort. 1184

1185

P-III-1024 Amelioration of fibrosis in vitro through downregulation of TRPA1 channel Jessa Flores1,2, Nammi Park2, Jubert Marquez2, Jeong Rim Ko2, Maria Victoria Faith Garcia2, Hyoung Kyu Kim1,2, Jin Han1,2 1Inje University, Department of Physiology, Busan, South Korea 2Cardiovascular and Metabolic Diseases Core Research Center, Busan, South Korea Background: Transient receptor potential Ankyrin 1 (TRPA1) is a non-selective cation channel primarily studied for its nociceptive role in neurons. However, more recent studies confirmed its expression in other tissues such as cardiac tissues where it facilitates ion transport. This study aimed (1) to evaluate the regulation of TRPA1; and (2) to determine the role of TRPA1 in the physiology of cardiac fibrosis in vitro. Methods: Western blot was performed to check the TRPA1 protein levels in vitro (mouse cardiac fibroblasts) fibrotic model. RT-PCR analysis was used to measure gene expression while immunocytochemistry was used to visualize the protein expression in cells. To create an in vitro fibrotic model, mouse cardiac fibroblasts (MCF) were treated with rhTGFβ1 (5 ng/mL). On the other hand, to create a TRPA1 knockdown model, siRNA targeting TRPA1 was treated to the cells. Results: TRPA1 levels in MCF were found to increase under fibrotic conditions. Meanwhile, knocking down TRPA1 using siRNA decreased the expression of fibrotic markers, α-SMA and col1a1. Accordingly, the decreased levels of TRPA1 under fibrotic conditions attenuated the phosphorylation of SMAD2 and ERK1/2 proteins which led to decreased fibrotic markers. Conclusions: The findings of this study showed that inhibition of TRPA1 is a potential strategy in addressing against TGF-β1-induced cardiac fibrosis. The study also provided evidence that TRPA1 affects the progression of fibrosis through the fibrotic ERK1/2 signaling pathway. Fig. 1 (optional) 1185

1186

P-III-1025 Unbiased complexome profiling and global proteomics analysis reveals mitochondrial and intercalated disk alterations in presymptomatic R14Δ/+-PLN mice hearts Brian Foo1,2, Hugo Amedei3, Surmeet Kaur3, Samir Jaawan1,2, Angela Boshnakovska4, Tanja Gall4, Rudolf A. de Boer5,6, Herman H.W. Silljé6, Henning Urlaub2,3,7, Peter Rehling4, Christof Lenz2,3,7, Stephan E. Lehnart1,2 1University Medical Center Göttingen, Heart Research Center Göttingen, Goettingen, Germany 2University of Göttingen, Cluster of Excellence “Multiscale Bioimaging: from Molecular Machines to Networks of Excitable Cells” (MBExC), Goettingen, Germany 3University Medical Center Göttingen, Department of Clinical Chemistry, Goettingen, Germany 4University Medical Center Göttingen, Department of Cellular Biochemistry, Goettingen, Germany 5Erasmus University Medical Center, Department of Cardiology, Rotterdam, Netherlands 6University Medical Center Groningen, Department of Cardiology, Groningen, Netherlands 7Max Planck Institute for Multidisciplinary Sciences, Bioanalytical Mass Spectrometry Group, Goettingen, Germany Background: Phospholamban (PLN) is a cardiac sarco-endoplasmic reticulum (SER) membrane protein that regulates cardiac contraction/relaxation by reversibly inhibiting the SERCA2a Ca2+-reuptake pump. The R14Δ- PLN mutation, most prevalent in North-West Europe and Greece, causes severe arrhythmogenic and/or dilated cardiomyopathy that is resistant to conventional treatment. Despite extensive clinical characterization, the cellular mechanisms underlying R14Δ-PLN pathogenesis remain unknown. Protein complexes and higher-order supercomplexes such as intercalated disk components and PLN-regulated Ca2+ release units underlie many critical cardiac functions, a subset of which, we hypothesize, may be disrupted by R14∆-PLN. Methods: We developed an improved complexome profiling (CP) workflow specifically geared towards systematically analyzing very high molecular-weight (>2 MDa) protein complexes and supercomplexes in R14Δ/+ mice hearts. Ventricular tissues were harvested from young (9wk- and 28wk-old) R14Δ/+ mice, significantly before the onset of symptoms at ~80wks of age. Tissues were homogenized under non-denaturing conditions, fractionated based on apparent molecular weight (MW) by size-exclusion chromatography (SEC) and subjected to quantitative data-independent acquisition mass spectrometry (DIA-MS) proteomics analysis. Conventional strategies for systematic analysis of CP data often rely on curated protein complex databases to provide ground- truth for analysis; however, cardiac-specific protein complexes (such as intercalated disk components and Ca2+ release units) and higher-order supercomplexes (such as the mitochondrial CI/CIII/CIV respirasome) are poorly annotated in these databases; consequently, conventional analytical workflows failed to generate significant insights from our cardiac tissue data. We thus developed PERCOM: a novel data analysis strategy that does not rely upon curated protein complex databases and can, furthermore, be implemented on widely available spreadsheet software. Results: SEC-DIA-MS coupled with PERCOM identified 296 proteins (out of ~3055 detected) with disrupted MW- elution profiles in presymptomatic 28wk-old R14Δ/+ mice. Hits were significantly enriched for mitochondrial and intercalated disk (ICD) components. Alterations to mitochondrial and ICD supercomplexes were observed in mice as young as 9wks of age, and were associated with reduced expression of mitochondrial proteins and maximal oxygen consumption rate. Conclusion: Demonstrating a novel CP-PERCOM workflow, we identify mitochondrial alterations as a very early- stage R14Δ-PLN event and provide preliminary data showing effects at the ICD which may contribute to the R14Δ conduction phenotype. These molecular components clearly underlie critical cardiac functions, hence their alteration at a young age may progressively contribute to R14∆-PLN associated arrhythmogenic and/or dilated cardiomyopathy. Preprint: https://doi.org/10.1101/2024.03.21.586093. 1186

1187

P-III-1026 Early candidate urine Biomarkers for detecting Alzheimer's disease before Amyloid-β Plaque deposition in an APP (swe)/PSEN1dE9 transgenic mouse model Fanshuang Zhang1, Jing Wei2, Xundou Li1, Chao Ma1, Youhe Gao2,1 1Peking Union Medical College, Beijing, China 2Beijing Normal University, Beijing, China Alzheimer’s disease (AD) is an incurable age-associated neurodegenerative disorder that is characterized by irreversible progressive cognitive deficits and extensive brain damage. The identification of candidate biomarkers before amyloid-β plaque deposition occurs is therefore of great importance for the early intervention of AD. Urine, which is not regulated by homeostatic mechanisms, theoretically accumulates changes associated with AD earlier than cerebrospinal fluid and blood. In this study, an APP (swe)/PSEN1dE9 transgenic mouse model was used to identify candidate biomarkers for early AD. Urine samples were collected from 4-, 6-, and 8-month-old transgenic mouse models, and the urinary proteomes were profiled using liquid chromatography coupled with tandem mass spectrometry (LC-MS/MS). The levels of 29 proteins differed significantly between wild type and 4-month-old mice, which had not started to accumulate amyloid-β plaques. Among these proteins, 13 have been associated with the mechanisms of AD, while 9 have been suggested as AD biomarkers. Our results indicated that urine proteins enable detection of AD before amyloid-β plaque deposition, which may present an opportunity for intervention. 1187

1188

P-III-1027 Preserved phosphorylation of mitochondrial creatine kinase 2 (CKMT2) confers cardioprotection against hypoxia/reoxygenation injury Maria Victoria Faith Garcia1, Nammi Park1, Jubert Marquez1, Jessa Flores1,2, Jeong Rim Ko2, Hyoung Kyu Kim1,2, Jin Han1,2 1Inje University, Department of Physiology, College of Medicine, Busan, South Korea 2Inje University, Cardiovascular and Metabolic Disease Core Research Support Center, Busan, South Korea Heart failure due to ischemic cardiomyopathy (ICM) is considered as one of the main causes of cardiovascular disease-related deaths worldwide. Patient mortality due to ICM still remains high. Ischemic preconditioning (IPC) has been found as an effective mitigator of ICM. This is done through short-time ischemia applied before ischemia/reperfusion injury (I/R) called ischemic preconditioning. The brief time applied for IPC may contribute to a rapid change in protein expression and regulation wherein protein function modulation by post-translational modifications is important. Mitochondria play a significant role in heart disease progression and so, it is a good target for ICM treatment. In this study, we focused on mitochondrial creatine kinase (CKMT2) under I/R injury following phosphoproteomic analyses. Ex vivo Langendorff system on Sprague-Dawley rat hearts were used to simulate normal perfusion, I/R, and IPC conditions and used for phosphoproteomic analysis. Phosphoproteomic analysis demonstrated a loss of phosphorylation in mitochondrial creatine kinase (CKMT2) coinciding with ischemia and ischemia/reperfusion (I/R) injury, whereas IPC maintained CKMT2 in its phosphorylated state. In vitro study using human cardiomyocyte AC16 cells were used to determine the cardioprotective role of mitochondrial creatine kinase through overexpression and how CKMT2 site-directed mutagenesis can be detrimental to CKMT2 protein activity, mitochondrial function, and protein expression in the cells. CKMT2 overexpression shows increased cell viability and mitochondrial ATP level against hypoxia/reoxygenation confirming the cardioprotective effect of CKMT2. Conversely, there was decreased cell viability and increased ROS production during H/R when CKMT2 is phosphomutated, specifically in Y368. We also confirmed increased mitochondrial function via the proliferator-activated receptor γ coactivator-1α/estrogen-related receptor-α pathway during CKMT2 overexpression. Our study identifies Y159, Y255, and Y368 of CKMT2 as critical phosphorylation sites situated at the nexus of the creatine-phosphocreatine (Cr-PCr) and ADP/ATP cycles. This phosphorylation likely facilitates the sustained generation of cellular ATP, thereby promoting cell survival. Our findings and hypoxia/reoxygenation (H/R) injury. Unlike the traditional focus on infarct size reduction, this study proposes CKMT2 phosphorylation as a potential therapeutic target such that CKMT2 regulation and phosphorylation may be used for future ICM therapeutics. cardioprotective mechanism suggest a novel during ischemia/reperfusion (I/R) Fig. 1 (optional) 1188

1189

P-III-1028 Salivary biomarkers in dementia Eniko Gebri1, Ferenc Tóth2, Tímea Kulcsár1, Anna Farkas3, Andras Guttman3,4, Tibor Hortobágyi4,5 1Faculty of Dentistry, University of Debrecen, Oral Medicine Department, Debrecen, Hungary 2Faculty of Dentistry, University of Debrecen, Biomaterials and Prosthetic Dentistry, Debrecen, Hungary 3Faculoty of Medicine, University of Debrecen, Horváth Csaba Memorial Laboratory of Bioseparation Sciences, Research Center for Molecular Medicine, Doctoral School of Molecular Medicine, Debrecen, Hungary 4Faculty of Medicine, University of Debrecen, Horváth Csaba Memorial Laboratory of Bioseparation Sciences, Research Center for Molecular Medicine, Doctoral School of Molecular Medicine, Debrecen, Hungary 5Faculty of Medicine, University of Debrecen, Neurology Clinic, Debrecen, Hungary Introduction: The incidence of dementia is increasing yearly. Alzheimer's disease accounts for about 70% of all cases. Neurodegenerative dementias are currently incurable. The first significant symptoms occur in an advanced stage. Caring for patients places a heavy burden on both the families and the healthcare system. During the COVID-19 pandemic, SARS-CoV-2 infection caused further rapid progression in the condition of patients with dementia. The aim of our study was to identify effective, early, non-invasive biomarkers in the diagnosis of dementia. Materials and methods: 25 SARS-CoV-2 negative patients with moderate to severe dementia participated in the study and peripheral blood, saliva and oral mucosa smears were taken. The severity of dementia was classified based on the MMSE (mini mental state examination-mini mental test), and patient data were systematically organized. The level of immunoglobulin A (IgA), secretory IgA (sIgA), interleukin-1β (IL-1β), tumor necrosis factor-α (TNF-α), interleukin-6 (IL-6), interleukin-8 (IL-8) and osteopontin (OPN) were determined from serum and rest mixed saliva with ELISA and ECLIA methods. Additional tests were performed on peripheral blood (e.g., APOE genotyping). A questionnaire summarizing COVID vaccinations and past infections was also completed. Cariological and periodontal status of the patients was also recorded. The obtained results were compared with age- and sex-matched healthy controls (n=10). Kolmogorov-Smirnov, Wilcoxon, Mann-Whitney and t-tests were used to analyze the data. We had the required ethical approvals for the experiments (DE RKEB/IKEB: 5395-2019). Results: Levels of both serum IgA and salivary sIgA, OPN, IL-6, IL-8, IL-1β, TNF-α differed significantly (p≤0.05) between each investigated groups (control vs moderate and severe dementia; moderate dementia vs severe dementia). Conclusion: Immunoglobulins and interleukins in saliva reflect well the acute and chronic inflammatory processes in the body. Taking other local and systemic modifying factors into account, they may be suitable as part of oral diagnostics to judge the severity of dementia and follow its progression. 1189

1190

P-III-1029 Unraveling the proteome of human embryo peri-implantation development Océane Girard1, Régis Lavigne2,3, Awa Bousso Gueye1, Jenna Lammers1,4, Emmanuelle Com2,3, Charles Pineau2,3, Laurent David1,5, Thomas Freour1,4 1Nantes University, CHU Nantes, Inserm, CR2TI, Nantes, France 2Univ Rennes, Inserm, EHESP, Irset - UMR_S 1085, Rennes, France 3Univ Rennes, CNRS, Inserm, Biosit UAR 3480 US_S 018, Protim Core Facility, Rennes, France 4CHU Nantes, Service de Biologie de la Reproduction, Nantes, France 5Nantes University, CHU Nantes, Inserm, CNRS BioCore, Nantes, France Infertility concerns 1 out 6 couples wishing to conceive and the worldwide use of assisted reproduction technology (ART) increases by 5 to 10% each year. Nevertheless, 10% of infertilities remain idiopathic. Understanding the window of human embryo peri-implantation development is essential to appreciate the beginning of human life but also for multiple clinical approaches such as in vitro fertilization, regenerative medicine and placenta physiopathology. Therefore, in order to investigate the mysteries of human embryo peri-implantation development, we studied the proteome of human embryos at different stages of development using a DIA mass spectrometry approach. MS analyses were performed on pre-human implantation embryos with morula, pre-implantation blastocyst stages, and human embryos miming the post-implantation blastocyst stage at days 8 and 10. It allowed the identification of 3,070 to 4,389 proteins depending on the stage of development. We could characterize significant changes during the human embryo peri-implantation development. Interestingly, the profile of proteins involved in metabolism appears different and depends on the embryo stage, highlighting the different requirements of the human embryo during his development in the maternal organism. Our work lead to the first proteome of the human embryo peri-implantation development including post- implantation embryo. It opens up new horizons for research not only on the human embryo, but also on stem cells and on the blastoid (pre-implantation blastocyst organoid) model. Altogether, theses models could contribute to improve the rate of in vitro fertilization success. Keywords: DIA analysis, human embryo, peri-implantation development 1190

1191

P-III-1030 Use of proteomics to develop a novel diagnostic approach in pseudomyxoma peritonei Melissa Granados-Rodríguez1,2, María C Vázquez-Borrego1,2, Florina Iulia Bura Balajan1,2, Blanca Rufian- Andujar1,3, Lidia Rodriguez-Ortiz1,3, Maria Torres-Martinez1,2, Ana Moreno-Serrano1,2, Francisca Valenzuela Molina1,3, Carmén Michán1,2, José Alhama1,2, Alvaro Arjona-Sanchez1,3, Antonio Romero-Ruiz1,2 1Maimonides Biomedical Research Institute of Cordoba (IMIBIC), GE09 Research in peritoneal and retroperitoneal oncological surgery, Córdoba, Spain 2University of Córdoba, Department of Biochemistry and Molecular Biology, Córdoba, Spain 3Reina Sofia University Hospital of Cordoba, Unit of Surgical Oncology, Department of Surgery, Córdoba, Spain Keywords: Pseudomyxoma peritonei, cancer, molecular characterization, proteomic analysis, mucins. Pseudomyxoma peritonei (PMP) is a rare malignant disease characterized by a massive accumulation of mucin and secreting tumour cells inside the peritoneal cavity. The only therapeutic option is a complex surgery combined with chemotherapy. Unfortunately, a considerable percentage of patients suffer tumoral recurrence with a fatal outcome. The only diagnosis method is limited to an invasive sampling followed by histological analysis. Additionally, mucin physicochemical properties difficult the molecular characterization of PMP and the development of novel diagnostic and prognostic approaches. We previously described the first protocol that allowed us to break the mucin barrier and isolate proteins from mucinous tumours. This method is based on the depletion of glycoproteins, immunoglobins, and albumins by affinity chromatography followed by mass spectrometry. This protocol led us to establish the first mucin isoform profile of PMP. Thanks to these results, we have designed a panel to measure the gene expression levels of all mucin isoforms present in tumour tissues and plasma samples, and generated an algorithm for predicting PMP disease grade using a machine learning model. In conclusion, our study provides essential information to develop a minimally invasive approach based on plasma samples to diagnose and classify PMP patients. Funding sources: This work is supported by "Instituto de Salud Carlos III (ISCIII)" (PI19/1603 and PI22/001213), co-funded by the European Union and by "Asociación Española contra el Cáncer" (PRYES223170ARJO). 1191

1192

P-III-1031 Understanding the impact of genetic variants on Alzheimer's disease with mass spectrometry proteogenomics Benjamin Lacar1, Harendra Guturu1, Amir Alavi1, Shadi Ferdosi1, Alexey Stukalov1, Guhan R. Venkataraman1, Matthijs de Geus2, Hiroko Dodge2, Chao-Yi Wu2, Pia Kivisakk2, Sudeshna Das2, Brad Hyman2, Serafim Batzoglou1, Steven E. Arnold2, Asim Siddiqui1 1Seer, Inc., Redwood City, CA, United States 2Massachusetts General Hospital, Charlestown, MA, United States Introduction: Proteogenomics is an emerging area of research that may provide novel insights into Alzheimer’s pathophysiology and therapeutic targets. While protein quantitative trait loci (pQTL) analysis has recently been performed with cerebrospinal fluid and brain samples, identification of pQTLs in easily accessible sample types like plasma could enable access to larger scale studies and provide insight to systemic factors influencing disease progression. Here, we identify pQTLs in an Alzheimer’s disease cohort using Proteograph™ XT and mass spectroscopy (MS)-based proteomics from plasma samples. With this study, we aim to examine the impact of genetic variation on protein abundance and reveal the potential functional implications in Alzheimer’s disease. Methods: Blood samples were collected from 1,005 participants in the Massachusetts Alzheimer’s Disease Research Center. Participant metadata includes demographics, cognitive scores, and clinical diagnoses. We also collected the plasma biomarker pTau-217 which has been shown to have strong diagnostic accuracy. Proteomics data were obtained by processing plasma samples with early release Proteograph XT (Seer, Inc.) before analysis on an Orbitrap Exploris 480 MS. Participant genotypes were obtained with low pass sequencing by Gencove. MS data was searched by DIA-NN in Library Free mode with scalable match between runs. The low pass sequencing data was mapped and imputed using GLIMPSE. We identified pQTLs and differential protein abundances based on Alzheimer's disease status and rate of progression. We then applied Mendelian randomization methods to identify putative causal protein markers of Alzheimer’s disease and cognitive decline. Preliminary Data: To reveal candidate proteins associated with Alzheimer’s disease, we performed differential abundance analysis and identified 138 protein groups that were different between healthy and AD groups. Additionally, eight protein groups were associated with dementia progression using Cox regression analysis. For pQTL identification, we performed a preliminary study with a pilot subset of the samples (N = 161) with array genotypes. We performed pQTL analysis both at peptide abundances for each nanoparticle as well as protein level abundance inferred by MaxLFQ rollup for nanoparticle-peptide abundances due to increased sensitivity of peptide level measurements. With nanoparticle-peptide level quantification, we identified 55 Bonferroni- corrected protein group pQTLs (16 cis and 39 trans) while panel protein quantification revealed 10 Bonferroni- corrected protein group pQTLs (5 cis and 5 trans). We will evaluate the reproducibility of the results from both quantification levels in the full data for defining our final set of pQTLs generated across all 1,005 participants. Novel Aspect: Identifying putative causality between protein abundance and AD progression with low pass sequencing, MS pQTLs, and Mendelian randomization. 1192

1194

P-III-1032 Characterisation of the cardiac function of alternative splicing-derived isoforms of synaptopodin 2-like protein (CHAP) Elke Hammer1,2, Sabine Ameling1,2, Stephanie Könemann2,3, Alexander Benkner2,3, Stephan Michalik1, Anja Wiechert1, Manuela Gesell Salazar1, Olga Tsoy4, Jan Baumbach4, Linda Weder5,6, Derk Frank5,6, Uwe Völker1,2 1University Medicine Greifswald, Functional Genomics, Greifswald, Germany 2German Center for Cardiovascular Research, partner site Greifswald, Greifswald, Germany 3University Medicine Greifswald, Department for Internal Medicine B, Greifswald, Germany 4University Hamburg, Faculty of Mathematics, Informatics and Natural Sciences, Computational Systems Biology, Hamburg, Germany 5University Hospital Schleswig-Holstein, Department of Internal Medicine III (Cardiology and Angiology), Kiel, Germany 6German Center for Cardiovascular Research, Partner site Kiel, Lübeck, Hamburg, Kiel, Germany Background: Heart failure (HF) of different origin is associated with alternative splicing of synaptopodin 2-like protein transcripts also named as cytoskeletal heart-enriched actin-associated protein (CHAP). In HF myocardial tissue display increased protein levels of the short isoform ChapB, while the amount of the canonical longer isoform ChapA remains unchanged. However, the cardiotropic function of the different isoforms is still largely unknown. Methods: ChapA and ChapB were cloned and overexpressed (OE) in neonatal rat ventricular cardiomyocytes (NRVCM) by transfection. Microscopy was used to determine the protein localisation and the influence of the isoform levels on cell size. The impact of altered isoform levels on the protein pattern was analysed by LC- MS/MS, affected pathways were identified and the results were compared with protein and transcript data of cardiac tissue from in-house mouse HF models. Results: WesternBlot and LC-MS/MS confirmed the successful OE of ChapA and ChapB. Changes in isoform levels were not associated with hypertrophic growth of the cells. Both isoforms were found to be localised in the cytoplasm as well as the nucleus. The cardiotropic influence of the ChapA OE was stronger than that of the ChapB OE. OE of the isoform ChapA, representing the major isoform in the healthy heart, shows a positive effect on cell viability and adhesion of cardiomyocytes. Higher levels of ChapB, indicative of pathological states, are associated with alterations in mitochondrial oxidases and ADP/ATP translocators, as well as changes in basal membrane proteins including Trim72, which is involved in membrane repair. Some of the findings were verified by correlating analyses of transcript and protein levels in murine HF models. Conclusion: Global protein profiling has been employed to provide molecular data on proteins that appear to be directly affected by Chap isoform levels in cardiomyocytes. Apparently, high ChapB levels in HF have an acute impact on membrane structure and mitochondrial ATP transport into the cytoplasm, thereby affecting the energy supply of the cells. Further investigation is required to ascertain biological significance which can be achieved by protein-protein-interaction studies and measurement of ATP levels in cells. 1194

1195

P-III-1033 Towards precision proteomics in the pathology of focal segmental glomerulosclerosis Tianen He1,2, Simon Davis1,3, Aneesha Bhandari2, Jessica Kepple2, Frederik Lassen1,4, Sarah Flannery1, Iolanda Vendrell1,3, Svenja Hester1, Sara Falcone2, Benedikt Kessler1,3, Richard Cornall2,3, Katherine Bull2,3,5, Roman Fischer1,3 1University of Oxford, Target Discovery Institute, Centre for Medicines Discovery, Nuffield Department of Medicine, Oxford, United Kingdom 2University of Oxford, Centre for Human Genetics, Nuffield Department of Medicine, Oxford, United Kingdom 3University of Oxford, Chinese Academy for Medical Sciences Oxford Institute, Nuffield Department of Medicine, Oxford, United Kingdom 4University of Oxford, Big Data Institute, Oxford, United Kingdom 5Oxford University Hospitals NHS Foundation Trust, Oxford Kidney Unit, Oxford, United Kingdom Focal segmental glomerulosclerosis (FSGS) is an understudied glomerular disease that leads to kidney failure associated with injury and loss of podocytes, specialised glomerular cells essential to the filtration barrier. Using spatial, single cell, and temporal proteomic techniques, this study aims to elucidate the pathology of FSGS, an essential first step towards the development of effective and targeted treatments. Using doxorubicin nephropathy in mice as a model for FSGS, the glomerular proteome signature was characterised with spatial proteomics based on laser capture microdissection. A single-cell proteomics workflow was showcased in human and mouse podocyte cell lines and piloted in primary mouse glomerular cells. The temporal proteome in podocytes derived from patients with genetic FSGS and healthy controls was measured across in vitro differentiation. Glomerular proteomics in an FSGS mouse model indicates damage to glomerular cells and extracellular matrix, and involvement of signalling pathways including JAK/STAT and ERK1/2 cascades, highlighting proteins for further functional validation. Marker proteins were identified from single mouse glomerular cells in preliminary experiments, suggesting the feasibility of characterising cell-type specific proteome and pathological mechanism. The differentiation proteome of human podocyte cell lines revealed pathways involved in disease podocytes, such as mitotic catastrophe, dysregulation of ATP synthesis and ion transmembrane transport. In summary, our research investigates the glomerular and cellular pathology of FSGS with multifaceted proteomics approaches, in mouse model, characterising podocyte defects in patients, and optimising single cell proteomics to understand renal pathology at higher precision. identifying potential therapeutic targets 1195

1196

P-III-1034 Cross-Omics Analysis of Proteome, Metabolome, and Transcriptome Dynamics during Peritoneal Dialysis Rebecca Herzog1, Fabian Eibensteiner1, Florian Wiesenhofer1, Klaus Kratochwill1,2 1Medical University of Vienna, Pediatric Nephrology, Vienna, Austria 2Medical University of Vienna, Proteomics Core Facility, Vienna, Austria Peritoneal dialysis (PD) as life-saving renal replacement therapy offers patients a home-based continuous therapy and higher quality of life compared to conventional center-based haemodialysis. PD uses the peritoneal membrane and a highly osmotic PD-fluid instilled into the peritoneal cavity to remove toxins, water, and solutes from the patient. The resulting waste product (PD effluent) is not only a rich source of markers for therapy monitoring and investigation of deregulated processes during PD it is also surprisingly underexplored and therefore poorly defined. For understanding PD transport dynamics and pathomechanisms on a systemic level, a multi-level omics approach is particularly attractive. Samples were obtained from stable patients chronically treated with PD at different time points of standard 4h peritoneal equilibration tests (PETs). Effluent was collected after the pre-PET (overnight) dwell and at 0h, 1h, and 4h dwells and plasma samples were also taken during this period. Effluent was separated into a cellular and cell- free component. Soluble metabolites in the cell-free compartment were processed using a targeted LC-MS workflow, proteins were enriched using equalizer beads, digested using the SP3 protocol and analysed using 36 off-line fractions and TMT-18plex with internal standard channel. The cellular material was subjected to RNA sequencing. The plasma proteome database was used for referencing plasma proteins and estimating plasma concentration. A bioinformatic workflow conjoined information from the datasets to reveal novel insights into the "PD-effluentome", especially unravelling the origin of proteins and metabolites in PD effluent. 207 unique metabolites were detected in cell-free PD effluent. A mixed-effect ANOVA of all metabolites demonstrated dwell time-dependent concentration changes in 173 metabolites. Post-hoc testing revealed most metabolites to be changed between 1h and overnight time points, followed by 114 and 46 differently concentrated metabolites between 4h and overnight and 1h and 4h, respectively. We quantified 9,797 transcripts in PD-effluent cells and 2,729 proteins in PD effluent. 342 proteins were filtered from plasma, while 800 proteins were attributable to local origin or production. A quantitative analysis of the interaction proteome and cellular transcripts of ~1700 protein-transcript pairs showed clusters of proteins explained by over- expression in peritoneal cells compared to plasma concentrations. Cross-omic profiling of PD effluent can be a valuable approach for revealing small molecule-related changes during PD. The exploitation of PD effluent on multiple improve the understanding of pathophysiological molecular processes and transport dynamics in the peritoneal cavity and their role in the development of PD complications. levels could 1196

1197

P-III-1035 Microsampling in a clinical setting enables stratification of heart failure patients Andreas Hober1, Tasso Miliotis1 1AstraZeneca R&D, Translational Science & Experimental Medicine, Moelndal, Sweden With the rapid progress in plasma proteomics and the ability to quantify proteins in a high throughput setting. The list of potential biomarkers is continuously expanded. To validate these biomarkers and show their potential as clinical tools for patient stratification, there is a need for improved sampling strategies. With recent understandings of the uniqueness amongst patients at a proteome level it has become evident that a lot of these potential biomarkers rely on longitudinal sampling. This is where microsampling has its biggest potential. Due to its non-invasive nature and good sample stability, microsampling allows for more frequent sample collection than conventional blood draws as well as a good tool for reaching a large population. In this study, we have collected paired plasma samples and Mitra tips from a cohort of 70 heart failure patients. With NT-proBNP being a highly interesting biomarker in heart failure studies, we have analyzed all paired samples using the Roche cobas assay targeting NT-proBNP. Additionally, all samples have been analyzed through data-independent acquisition (DIA) using both a SCIEX ZenoTOF 7600 and a Bruker timsTOF Pro 2, to further characterize the samples. All data processing of the MS results was done using DIA-NN followed by targeted data assessment using Skyline. The results from the Roche cobas measurements and the DIA analysis will be presented and highlight the possibility of using microsampling for patient stratification. The data proves how microsampling can be a useful asset for the future of precision medicine, where disease biomarkers can be assessed in a high throughput manner with a very limited patient burden. This way of screening for disease can improve the chances of early discovery of diseases while minimizing the cost of sampling. 1197

1198

P-III-1036 Investigating the cardioprotective effects of SGLT-2 inhibitors Julia Höhlschen1, Tamara Tomin1, Ruth Birner-Grünberger1,2 1TU Wien, Institute of Chemical Technology and Analytics, Vienna, Austria 2Medical University of Graz, Institute of Pathology, Graz, Austria Beyond their initial approvement for treating diabetic conditions, Sodium-Glucose co-transporter 2 (SGLT-2) inhibitors have recently been found to also offer cardioprotective properties. Since cardiovascular problems are one of the major complications that diabetic patients face, this finding was groundbreaking. Meanwhile it was shown that the cardioprotective effects are not limited to diabetic patients and the inhibitors" approvements have been expanded respectively. Observations from clinical studies hint in the direction that those effects exceed the glucose lowering effect without offering an explanation for the underlying mechanisms. Since the drug"s target protein SGLT-2 is exclusively expressed in the early proximal tubule where it is responsible for the reabsorption of glucose it remains inconclusive why the positive effects are observed in the heart. To identify off-targets of the drug, thermal proteome profiling is applied. In this method the basic principle is that the thermal stability of proteins is increased when binding to the drug, reflected by an increased stability of the respective protein when applying a temperature gradient. In order to target the off targets in the cardiac environment those experiments are carried out in the human cardiac cell line AC16 (cardiomyocytes derived from left ventricular heart tissue). 1198

1199

P-III-1037 Multi-omics analysis reveals phenotypic changes and energetic remodeling of VSMCs under Ang II treatment Yiwei Hu1, Liang Qiao1, Ling Lin2 1Fudan University, Department of Chemistry, Shanghai, China 2Zhongshan Hospital, Shanghai, China The process of vascular remodeling is a complex phenomenon in which the equilibrium between vascular cells and the internal environment is disrupted, particularly in the context of cardiovascular diseases. Angiotensin II (Ang II) has been demonstrated to induce vascular remodeling and contribute to the development of vascular diseases by regulating functions of vascular smooth muscle cells (SMCs). The objective of this study is to elucidate the pathophysiological mechanisms underlying vascular remodeling in greater detail. We obtained vascular specimens from clinical coronary artery bypass grafting (CABG) surgeries and successfully established a human primary aortic SMC model. By using varying concentrations of Ang II, we observed that Ang II promotes the proliferation of SMCs and slightly affects their migration ability. We further employed a multi- omics approach—including untargeted proteomics, metabolomics, and lipidomics—utilizing mass spectrometry (MS) to analyze the alterations induced by different dosage of Ang II in human primary SMCs. Multi-omics analysis identified a total of 8,382 proteins, 239 metabolites, and 797 lipids. Among the 528 differentially expressed proteins (DEPs), we observed abnormal deposition of extracellular matrix (ECM) proteins, along with perturbations of RhoA and Rac1 proteins, revealing alterations in the cytoskeleton and cell motility of SMCs. The multi-omics data obtained from Ang II-treated and control SMCs were integrated to identify significantly altered molecules. Subsequent analyses, including pathway enrichment and interaction network assessments, were conducted to pinpoint critical pathways underlying Ang II-induced vascular remodeling. Notably, the upregulation of numerous pro-inflammatory cytokines, including IL6 and TGFB1, indicated a heightened inflammatory response. This inflammatory response was further confirmed through lysophosphatidylcholine (LPC) accumulation in the lipidomic dataset. By combining the lipidomic and proteomic datasets, we observed an upregulation of the DNA replication pathway and a downregulation of the phosphatidylserine (PS), suggesting increased cell proliferation and inhibition of apoptosis. Furthermore, the integration of proteomic and metabolomic data collectively revealed significant alterations in energy metabolism, specifically in the pentose phosphate pathway (PPP), indicating an altered energy pattern during cell proliferation process. These findings revealed phenotypic changes and energetic remodeling in VSMCs under Ang II treatment, highlighting a regulatory role for VSMCs in the vascular remodeling process. Fig. 1 (optional) 1199

1200

P-III-1038 Revealing molecular subtypes and pathways in large vessel Occlusion stroke: a comprehensive quantitative proteomic analysis Ignasi Jarne Sanz1, Joan Josep Bech-Serra1, Helena Quesada2, Tatiani Brenelli Lima1, Julia Bauzá-Martinez1, Bernat Cucurull1, Sonia Aixut Lorenzo2, Alexandre Lüttich Uroz2, Laura Dorado3, Jesús Juega4, Carlos A. Molina4, Christina Ludwig5, Pere Cardona2, Jorge Pagola4, Carolina De La Torre Gomez1 1Josep Carreras Leukaemia Research Institute (IJC), Proteomis Unit, Badalona, Spain 2Bellvitge University Hospital, Fundació Institut d'Investigació Biomèdica de Bellvitge (IDIBELL), Department of Neuroradiology, L'Hospitalet de Llobregat, Spain 3Germans Trias I Pujol University Hospital, Stroke Unit, Department of Neurology, Badalona, Spain 4Vall d'Hebron Research Institute, Valld'Hebron University Hospital, Stroke Unit Department of Neurology, Barcelona, Spain 5School of Life Sciences, Technical University of Munich, Bavarian Center for Biomolecular Mass Spectrometry (BayBioMS), Freising, Germany Acute ischemic stroke is the first cause of disability and the second cause of mortality worldwide. In large vessel occlusion (LVO) stroke, the occlusion of cerebral arteries is most caused by a thrombus of arterial (arteriosclerotic, AT) or cardiac (cardioembolic, CAR) origin, although, in some cases the potential cause of occlusion is not clear, and patients are classified as presenting embolic stroke of undetermined source (ESUS). Currently, the LVO stroke diagnosis is based on integration of clinical examinations, neuroimaging analysis and physiologic monitoring, these often fail to determine with high specificity stroke etiology, translating into a lack of treatment plans tailored to the individual patient. Therefore, leveraging on advances on clinical proteomics, it would be valuable to apply translational and precision medicine approaches to move towards more accurate, efficient, and effective treatments. Recently, endovascular therapies, such as mechanical thrombectomy (MT), have demonstrated great benefit regarding long-term clinical outcomes on LVO stroke patients. MT allows the extraction of the occlusive thrombus, providing a unique opportunity to characterize its proteome and thus, to improve our understanding on the molecular mechanisms that drive stroke disease. In recent years, there has been a growing interest in defining a molecular signature of thrombi capable to explain acute ischemic stroke etiology. However, multiple studies attempting this did not find a robust or clinically significant molecular signature. In this study, we employ a multiplexing depth quantitative proteomics, and unsupervised analysis to profile the proteomic composition of 56 thrombi samples obtained from LVO stroke patients. Our investigation uncovers four distinct thrombi proteomic signatures that transcend stroke aetiology offering insights into specific molecular subtypes and dysregulated pathways, including the upregulation of proteins linked to the acute phase response; the proteasome and catabolism; mitochondrial dysfunction and platelet activation. Our results may open a new way this disease is characterized and subsequently permit the development of new therapies. 1200

1201

P-III-1039 Targeted MS screen to characterize the complete human protein tyrosine phosphatase repertoire in liver metabolic diseases Elisavet Kalaitsidou1,2, Ziliang Ma2, Esteban N. Gurzov3, Wei Wu1,2 1National University of Singapore, Department of Pharmacy and Pharmaceutical Sciences, Singapore, Singapore 2Agency for Science, Technology and Research (A*STAR), Singapore Immunology Network (SIgN), Singapore, Singapore 3Université libre de Bruxelles, Signal Transduction and Metabolism Laboratory, Brussels, Belgium Tyrosine phosphorylation is a critical post-translational modification (PTMs) regulated by protein tyrosine phosphatases (PTPs). Dysregulation in PTPs activity is often associated with human diseases. Recently, much focus has been placed on PTPs in the context of liver metabolic diseases and liver cancer, as there is growing evidence suggesting that PTPs can regulate processes associated with the progression of metabolic dysfunction- associated steatotic liver disease (MASLD). PTP enzymatic activity is dependent on the cysteine residue, that resides within the conserved [I/V]HCSxGxGR[T/S]G motif, shared exclusively among all the PTPs. Cysteine is susceptible to oxidation by reactive oxygen species (ROS); thus, oxidation is an intrinsic mechanism that maintains PTP activity dynamics within a required range. Understanding the PTP activity across MASLD progression could shed light on the phosphorylation patterns, offering strategy that can reverse dysregulated kinase activities. Here, we perform sequential steps of chemical derivatization to lock physiological redox- regulated PTP states and exploit the conserved [I/V]HCSxGxGR[T/S]G motif to facilitate PTP purification. To achieve complete and specific PTP detection we utilized a targeted MS acquisition in a parallel reaction monitoring (PRM) mode for selective PTP identification, based on the upstream sequence of the [I/V]HCSxGxGR[T/S]G motif. The therapeutic success of PTMs inhibitors underscores potential of PTPs as targets for therapeutic interventions. Thus, the developed method is applied to human liver biopsies representing different stages of MASLD to uncover the PTPs repertoire, enabling dissection of intricate phosphorylation signaling pathways in the liver and guide clinical therapeutic interventions in MASLD. 1201

1202

P-III-1040 Proteome changes in lung cancer cell lines treated with grape extracts: a potential study for biomarkers Chaquayla Katam1, Virginia Gottschalk1 1Florida Agricultural and Mechanical University, Biological Sciences, Tallahassee, FL, United States Muscadines grapes (Vitis rotundafolia) are rich in phenolic compounds and antioxidants. Studies on grape extracts have shown that several of these compounds possess cytotoxic effects against various human cancer cells. The goal of this research is to (1) determine the cytotoxicity of human lung cancer cells against a gradient concentration of grape pericarp and seed extracts, (2) identify differentially expressed proteins in cancer cells treated with the grape extracts, and (3) determine biosynthetic pathways associated with the inhibition of cancer cell growth. Seed and pericarp extracts were prepared separately and incubated with A209 human lung cancer cell lines for varying durations. The cells were harvested and tested for viability at 24 and 48 hours to determine the cell growth inhibition. Total proteins were isolated from both control and treated cells and resolved using 2-dimensional electrophoresis. The gels were scanned to detect differentially expressed proteins. Protein spots were eluted and subjected to LC-MS/MS for peptide sequences. The peptides were matched to protein sequences in the human database, which mapped the proteins' functions. A total of 38 proteins were identified with significantly altered expression in the cell lines treated with the extracts. Pericarp and seed extracts showed different responses on the protein abundance. Further analysis revealed that various proteins including disulfide-isomerase, thioredoxin, enolase, and peroxiredoxin were low in abundance in the lung cancer cells treated with grape extracts for 48 hours. The proteins mentioned serve a specific purpose attributing to the treatment of lung cancer cells. Thioredoxin plays a significant role in breast cancer cell lines because it inhibits apoptosis and stimulates cell growth. Enolase was found to be positively correlated with tumor size and venous invasion. Disulfide-isomerase has roles in protein folding and forming disulfide bonds and is abundant in metastasized breast cancer cells. Peroxidase is a molecular chaperone that is overexpressed in cancer cell lines. It was observed that the suppression of this protein resulted from an increase in the peroxide-induced cytotoxicity of the cells. This data shows the antiproliferative effect of muscadine extract on the downregulation of these proteins that play a major role in the metabolism of the cancer cells. These findings suggest that the proteins involved in the formation of the cytoskeleton, cell signaling, and cellular integrity were significantly affected when the cells were treated with muscadine extracts. This research provides insight into the discovery of potential protein biomarkers for the onset of human cancers. 1202

1203

P-III-1041 Mass-spectrometry based proteomics reveals tissue specific metabolic perturbations in a mitochondrial disease mouse model Glen Khumalo1, Jeremie Zander Lindeque1, Marianne Venter1 1North-West University, Focus Area for Human Metabolomics, Potchefstroom, South Africa Background: Mutations in the gene encoding NDUFS4, an accessory subunit of the complex I of the mitochondrial oxidative phosphorylation (OXPHOS) system, is a common cause of Leigh Syndrome (LS), an infantile heterogenous neurodegenerative disorder with variable clinical phenotypes. Understanding how different tissues/organs respond to complex I deficiency might help unravel the mechanisms involved in the heterogenous manifestation of mitochondrial disease. Methods: To investigate tissue-specific proteomic in LS, we performed quantitative mass-spectrometry based proteomics on a LS mouse model (Ndufs4 KO) using the SWATH-MS data acquisition mode. Six distinct tissues (brainstem, cerebellum, olfactory bulb, heart, kidney, and liver) were collected from Ndufs4 KO mice (n =9) and compared to age/sex matched controls (n =9). Proteins were extracted from whole tissue lysate, digested into peptides using trypsin and analysed on the Evosep One LC system coupled to a Sciex 5600 TripleTOF mass spectrometer. Results: Our SWATH-MS data acquisition mode resulted in the identification of more than 1, 000 proteins in each tissue based on 1% FDR. Differential expression analysis revealed significant downregulation of the NDUFS4 across the tissues. Functional enrichment analysis using gene ontology and pathway analysis on differentially expressed proteins (DEPs) unique to each tissue revealed varied cellular responses. Moreover, pathway enrichment analysis pointed towards enrichment of DEPs related to translation, neurotransmitter release cycle, fatty acid metabolism, biological oxidations, and phase II conjugations. Conclusion: This study reveals tissue-specific cellular responses to NDUFS4 mutations in LS, providing valuable insights into the molecular mechanisms underlying LS and other mitochondrial related disorders. Understanding these tissue-specific responses can inform the development of rational therapeutic strategies for LS. 1203

1204

P-III-1042 Utilizing proteomic approaches to study the molecular effects of emerging plastic additives in human adipocytes Jasmin Kleißen1, Isabel Karkossa1, Alix Sarah Aldehoff1, John Thomas Heiker2,3, Martin von Bergen1,4,5, Kristin Schubert1 1Helmholtz-Centre for Environmental Research - UFZ, Molecular Toxicology, Leipzig, Germany 2Helmholtz Institute for Metabolic, Obesity and Vascular Research (HI-MAG) of the Helmholtz Zentrum München at the University of Leipzig and University Hospital Leipzig, Leipzig, Germany 3University of Leipzig Medical Center, Medical Department III - Endocrinology, Nephrology, Rheumatology, Leipzig, Germany 4University of Leipzig, Institute of Biochemistry, Faculty of Biosciences, Pharmacy and Psychology, Leipzig, Germany 5German Centre for Integrative Biodiversity Research (iDiv) Halle-Jena-Leipzig, Leipzig, Germany Bisphenol A (BPA) is well-known for its disruptive effects on the endocrine system, as well as the metabolism. Adipose tissue, beyond its role in energy storage, is also an important endocrine organ. BPA has been shown to alter insulin sensitivity, which can subsequently lead to the development of type 2 diabetes. Due to the classification of BPA as an endocrine disruptive chemical its use in many everyday consumer products has been restricted in the European Union. Consequently, novel BPA alternatives are now prevalent in various consumer products, and so is our exposure to these chemicals. The effects of these novel plastic additives on the endocrine system and metabolism remain poorly understood, rendering the hazard assessment and the understanding of their molecular modes of action in the adipose tissue extremely relevant. Therefore, this research project aims to evaluate eight BPA alternatives of high concern for their impact on human adipocytes within the framework of the Partnership for the Assessment of Risk from Chemicals (PARC). Specifically, the analysis will focus on the critical endpoints such as lipid accumulation, cell toxicity, binding effects to cellular proteins, and alterations in the proteome. The treatment of adipocytes with brominated bisphenols TBBPA (tetrabromobisphenol A) and TBBPS (tetrabromobisphenol S), commonly used as flame retardants, induced significant changes in the adipocyte"s lipid accumulation. This effect is mediated through interference with nuclear receptors responsible for adipocyte differentiation and was demonstrated at concentrations ranging from environmentally relevant to bioaccumulative concentrations. Simultaneously, the cytotoxic effects of the halogenated bisphenols were determined at higher concentrations. Additionally, we were able to use thermal proteome profiling to identify potential chemical-protein interactions. Preliminary results have shown a significant shift in melting points of proteins that are associated with stress response and lipid metabolism. Additionally, global proteome analysis was applied to give insights into the changes in protein abundance, thereby revealing disrupted signalling pathways and shedding light on their potential modes of action. In conclusion, our findings could unveil molecular mechanisms triggered in human adipocytes and simultaneously aid in the hazard assessment, thereby supporting necessary restrictions for chemicals, humans and the environment are permanently exposed to. By comprehensively examining these various endpoints, our study seeks to elucidate the potential health risks posed by these novel plastic additives. 1204

1205

P-III-1043 Cereblon as potential biomarker for diabetic cardiomyopathy in type 2 diabetes-induced mice Jeong Rim Ko1, Sae Hwan Park1, Jin Han1 1Inje University, Busan, South Korea The high incidence of heart failure in patients with type 2 diabetes mellitus (T2DM) can be attributed to myocardial fibrosis, weakening of ventricular function, and damage to myocardial muscles, such as mitochondrial dysfunction. We studied physiological changes in the heart, which are early characteristics of diabetic myocardial disease, using 6- and 8-week old obese T2DM model mice (db/db, BKS.Cg- Dock7m+/+Leprdb/J) and wild type mice. We confirmed mitochondrial dysfunction in 8 weeks old db/db mice, which was not accompanied by any changes in heart function. Metabolomic analysis performed on heart tissues revealed that the levels of 12 metabolites changed substantially. The levels of glucose and leucine increased considerably, and lipid metabolites changed. Protein expression analysis of the blood and heart tissues confirmed a remarkable increase in cereblon (CRBN) levels and a decrease in AMP-activated protein kinase, a negative regulator. Furthermore, in silico studies on data in the blood of diabetic cardiomyopathy (DCM) patients. These results show that an increase in CRBN, without any abnormal heart function, in early T2DM mice roles an important role in the reduction of mitochondrial function and metabolomic changes. Therefore, CRBN level in blood and heart tissues may serve as a diagnostic biomarker for the detection of early DCM. 1205

1206

P-III-1044 Peptidomics analysis of mouse tissues Yoshio Kodera1,2, Yusei Okuda1, Yuzuru Nakagawa1, Takashi Matsui1,2, Makoto Itakura2,3 1School Of Science, Kitasato University, Physics, Sagamihara, Japan 2School Of Science, Kitasato University, Center for Disease Proteomics, Sagamihara, Japan 3School of Medicine, Kitasato University, Biochemistry, Sagamihara, Japan Bioactive peptides secreted by tissues circulate in the blood and perform various functions. Therefore, bioactive peptides are used as drugs and biomarkers. To identify novel bioactive peptides, we developed the Differential Solubilization (DS) method, a highly efficient peptide extraction method for plasma. As a result, we were able to identify more than 10,000 peptides. Howverer, most of the identified peptides were fragments of major proteins in plasma. It was considered necessary to identify peptides in plasma based on tissue-derived peptides for the discovery of novel bioactive peptides. Therefore, we started peptidome analysis in mouse tissues with the aim of proposing novel bioactive peptides from tissue-derived peptides in plasma. A mouse tissue was frozen immediately after dissection. Peptides were extracted from a frozen tissue using the modified differential solubilization (DS) method1,2 optimized for tissue and analyzed using Q-Exactive equipped with an EASY-nLC 1000 system (Thermo Fisher Scientific). LC-MS/MS data were searched using PEAKS Studio (Bioinformatics Solutions Inc.). For the hypothalamic tissue, LC-MS/MS analysis resulted in the identification of 1,535 peptides derived from 297 proteins. Approximately 45% of the identified peptides (690 of 1,535 peptides) belonged to prohormone precursor protein groups. Within the prohormone precursor protein group, 35 bioactive peptides listed in the database, including substance P, neurokinin A, neuropeptide K, neuropeptide Y, lipotropin γ and α-MSH, were identified. Furthermore, as for the neuropeptides secreted from the hypothalamus, Orexin B, β-endorphin and three opioids α-neoendorphin, dynorphin A and dynorphin B were detected. In the presentation, the results of peptidome analysis of the liver, kidney and brain tissue other than the hypothalamus will be presented and discussed in comparison with the results of previous peptide analyses. 1. Kawashima Y., et al. 2010 J. Proteome Research. 9:1694-705 2. Nakagawa Y., et al. 2021 Biochem Biophys Res Commun. 548:155-160 1206

1207

P-III-1045 Proteomic workflows for deep phenotypic profiling of 3D organotypic liver models Stefania Koutsilieri1, Evgeniya Mickols1,2, Ákos Végvári3, Volker M. Lauschke1,4,5 1Karolinska Institute, physiology and pharmacology, Solna, Sweden 2Uppsala University, Department of Pharmacy, Uppsala, Sweden 3Karolinska Institute, Division of Chemistry I, Department of Medical Biochemistry and Biophysics,, Solna, Sweden 4Dr Margarete Fischer-Bosch Institute of Clinical Pharmacology, Stuttgart, Germany 5University of Tübingen, Tuebingen, Germany Our publication addresses the pressing need for reliable and efficient methods in drug discovery and development by focusing on organotypic human tissue models. These models hold immense potential for advancing pharmacokinetic and toxicity studies, as well as phenotypic screening, due to their ability to maintain native cellular phenotypes and functions over extended periods. While proteomic profiling is recognized as the gold standard for tracing phenotypic changes in these tissue models, the impact of various proteomic workflows on analytical results and biological conclusions remains largely unexplored. Our study addresses this gap by comparing the performance of twelve mass spectrometry- based proteomic workflows on primary human liver spheroids. We investigated differences in the total number and subcellular compartment bias of identified proteins, crucial for accurate quantification of transporters and drug metabolizing enzymes. Our findings reveal significant variations between protocols, highlighting the importance of selecting an appropriate workflow for proteomic analysis in organotypic tissue models. Notably, we demonstrate that a standardized, high-throughput-compatible workflow based on thermal lysis yields robust results, even with minimal cellular input. This was exemplified in a model of metabolic dysfunction-associated steatotic liver disease, where critical disease pathways were reliably identified using individual spheroids as input. Our study significantly advances the field by increasing the applicability of proteomic profiling to phenotypic screens in organotypic microtissues. Furthermore, we provide a scalable platform for deep phenotyping from limited biological material, thus enhancing the efficiency and effectiveness of drug discovery and development processes. 1207

1208

P-III-1046 Profiling the proteome and phosphoproteome of testicular cancers (Germ Cell Tumors) Tomasz Kowalczyk1, Dominik Cysewski1, Adam Kretowski1, Michał Ciborowski1, Grzegorz Młynarczyk1 1Medical University of Białystok, Białystok, Poland Testicular cancer (TC) accounts for approximately 1% of adult cancers and 5% of all urological cancers diagnosed, with germ cell tumors (GCTs) being the most common histopathological subtype (90-95% of cases). The rapid metastasis of TC to retroperitoneal lymph nodes, mediastinum, lung, brain, liver, and bones necessitates a deeper understanding of the mechanisms driving this aggressive behavior. Elucidating these mechanisms could highlight significant changes in cell signaling or modulation of the immune system. To explore the molecular alterations underlying TC, we conducted comprehensive proteomics and phosphoproteomics analyses on tissue samples. Proteins were trypsin digested, separated using a nano-LC system coupled with Orbitrap Fusion MS, and processed with PEAKS 11 software. Phosphoproteins were enriched using Ti-IMAC beads and similarly analyzed. Our study quantified over 5000 proteins and more than 1000 phosphoproteins across 10 samples. Our findings reveal a significant upregulation in the tumor of biochemical pathways associated with the cell cycle, cell signaling, and cell-cell interactions mediated by tyrosine kinase receptors. Proteins responsible for neutrophil activation as well as the phagocytosis process also show increased expression. However, in cancer we observe reduced coagulation and prothrombin activation as well as reduced activity of processes related to cholesterol management. Phosphoproteomics identified 400 proteins differentiating cancerous tissue from controls. The identified proteins are responsible for the regulation of cell structures, cell metabolic processes, mRNA splicing, regulation of mRNA processing. Phosphorylated proteins indicate the involvement of small GTPases, as well as MAPK family proteins in the carcinogenesis process. The obtained phosphoproteomics results confirm the global observation of changes occurring in cancer cells affecting cell signaling, disturbances in glycolysis or the cell cycle, as well as cell-cell interaction resulting from disturbances in Gap junctions. These insights enhance our understanding of Testicular Cancer at molecular level and may inform the development of targeted therapies and diagnostic/prognostic markers. 1208

1209

P-III-1047 Decoding Glycosylation Changes in the Prader-Willi Syndrome Mouse Brain Delf-Magnus Kummerfeld1,2, Johanna Schrader3, Boris V. Skryabin2, Sergey Vakhrushev4, Timofey S. Rozhdestvensky2 1Leibniz Institute on Aging – Fritz Lipmann Institute, Jena, Germany 2Münster University, Transgenic animal and genetic engineering Models (TRAM), Muenster, Germany 3Leibniz University Hannover, L3S Research Center, Hannover, Germany 4University of Copenhagen, Department of Cellular and Molecular Medicine, Copenhagen, Denmark Prader-Willi syndrome (PWS) is a developmental neurogenetic disorder caused by the deletion or inactivation of imprinted genes within the PWS locus on the paternally inherited human chromosome 15. This locus contains two tandemly repeated non-protein-coding RNA (snoRNA) gene clusters, SNORD116 and SNORD115. Deletion of the SNORD116 region is associated with PWS in humans and results in growth retardation with 15% postnatal lethality in mice (PWS-like model). PWS is a multisystem disorder characterized by endocrine, neurological, and behavioral/psychiatric abnormalities. Glycosylation plays a crucial role in numerous cellular functions, and disruptions in glycosylation processes could significantly contribute to the pathophysiology of PWS. Previous studies have shown that glycoprotein-related dysfunctions may be involved in the development of the biological phenotype of PWS. For instance, attempts to analyze protein glycosylation in the blood serum of children with PWS have indicated hyposialylation, reflecting abnormalities in O-linked protein glycosylation pathways. However, comprehensive analyses of the glycoproteome in PWS-affected tissues, including the brain, have not been performed until now. For the first time, we have constructed a quantitative map of the O-GalNAc type glycoproteome in the pituitary gland, hypothalamus, thalamus, hippocampus, and olfactory bulb brain areas of a postnatal day 28 PWS-mouse model compared to wild-type siblings. To ensure deep mapping of protein glycosylation, we performed label free quantitative glycoproteomics with the GlycoDIA method. Our analysis revealed, on average, 7628±1461 unique glycopeptide precursors as well as 3799±300 distinct proteins per compartment, with the most significant differences in glycosylation patterns observed in the hypothalamus and pituitary gland. Altered glycosylation patterns in these brain areas may affect functional brain pathways by altering protein interactions, cell adhesion, and signaling. These changes can affect neuronal development, synaptic plasticity, and network formation that are critical for cognitive function and behavior. In the context of PWS, such dysregulation may contribute to the characteristic cognitive deficits, behavioral problems, developmental delay, and growth retardation. Understanding the underlying molecular mechanisms may provide insight into potential therapeutic targets to mitigate these symptoms. This study represents a significant step forward in understanding the role of glycosylation in PWS and underscores the importance of investigating glycosylation patterns to unravel the complexities of this disorder. 1209

1210

P-III-1048 Comprehensive plasma proteomics in heart failure patients: findings and future directions Uros Kuzmanov1,2, Vallijah Subasri3, Karem Abdul-Samad3,2, Bo Wang3,2, Douglas Lee1,3,2, Filio Billia1,3,2, Anthony Gramolini1,3,2 1University of Toronto, Ted Rogers Centre for Heart Research, Toronto, Canada 2University of Toronto, Toronto, Canada 3University Health Network, Toronto, Canada Patients with heart failure (HF) experience a variety of incapacitating symptoms, including breathlessness, fatigue, fluid accumulation and a significantly impaired quality of life with a 20% 1-year mortality. Globally, prevalence of HF has reached epidemic proportions, affecting an estimated 64 million individuals, and posing a escalating public health crisis with substantial financial implications for healthcare systems. Plasma samples from 512 heart failure patients were subjected to proteomic analysis using SEER Bio's proprietary nanoparticle platform employing custom-made nanoparticles to enrich low abundance plasma proteins, enhancing coverage of the plasma proteome. The mass spectrometry (MS) data was recorded on the latest Thermo Astral instrument with Data Independent Acquisition methodology and searched against an in-silico generated spectral library using DIA-NN 1.81. Processing, normalization, univariate and multivariate analyses (PCA, UMAP, t-SNE) of the search results were performed using a combination of R and Perseus. Functional enrichment analyses, based on clinical parameters including NYAH grade, LV ejection fraction, and HD etiology, were conducted to detect abnormal molecular pathways and other annotated biological processes using a compendium of publicly available databases such as Gene Ontology, KEGG, MSigDB, NCI Nature PID, NetPath, Panther, and Reactome. In excess of 7500 plasma proteins were identified. Statistical and bioinformatic analyses revealed hundreds of potential candidate proteins and biological pathways. These findings highlighted differences in patient characteristics such as NYHA grade, LV ejection fraction, and HF etiology. Consequently, this study provides a powerful resource and reference dataset for an unbiased global analysis of the circulating proteome in patients with heart failure, offering valuable insights into disease mechanisms and potential biomarkers. 1210

1211

P-III-1049 Evolution of proteomics of bioengineered human cardiac microtissues Uros Kuzmanov1,2, Homaira Hamidzada1,3,2, Erika Yan Wang2, Milica Radisic1,3,2, Slava Epelman1,3,2, Anthony Gramolini1,3,2 1University of Toronto, Ted Rogers Centre for Heart Research, Toronto, Canada 2University of Toronto, Toronto, Canada 3University Health Network, Toronto, Canada Organ-on-a-chip models for cardiac tissue are crucial for advancing biomedical research by providing human- relevant data, enabling real-time monitoring, reducing reliance on animal models, and facilitating personalized medicine and high-throughput drug screening. The evolution of the Biowire heart-on-a-chip system for proteomics analysis has significantly advanced our understanding of myocardial fibrosis and other cardiac pathologies. The initial strategy employed Q Exactive HF orbitrap nano LC mass spectrometry for comprehensive proteomic profiling across various cardiac models, including the Biowire system and patient-derived cardiac tissues. By mapping the relative abundance of proteins, key signaling pathways driving fibrosis were identified, offering new insights and a platform for anti-fibrotic drug screening. Additionally, the application of the Biowire system to model Angiotensin II-induced cardiomyopathy highlighted the limitations of rodent models and presented a controllable, human-relevant platform for drug evaluation. This model captured acute and chronic responses to Ang II, facilitating the assessment of cardioprotective compounds such as losartan, relaxin, and saracatinib. Finally, innovations in the Biowire system, incorporating human embryonic stem cell (hESC)-derived macrophages, defined critical interactions in cardiac microtissues, enhancing maturation and function through proteomic analyses. These advancements underscored the role of macrophages in reducing cytotoxicity and promoting cardiomyocyte maturation, with significant implications for in vitro studies and in vivo cell-based therapies. Together, these studies demonstrate the powerful capabilities of Biowire proteomics and organ-on- a-chip systems in advancing cardiac research, providing robust platforms for drug screening and disease modeling. Here, we compare the evolution over the years of the Biowire proteomic technical approaches utilizing isobaric labeling, label free data dependent and data independent acquisition methodologies and their performance in regard to depth of proteomic coverage, dynamic range, technical and biological reproducibility, robustness of bioinformatic and statistical analyses, and integration of findings into biological pathways and networks. 1211

1212

P-III-1050 Chlordecone-induced hepatotoxicity and fibrosis are mediated by the proteasomal degradation of septins Thibaut Léger1, Sarah Alilat2, Pierre-Jean Ferron3, Léonie Dec2, Tahar Bouceba4, Rachelle Lanceleur2, Sylvie Huet2, Yoann Devriendt-Renault5, Julien Parinet5, Bruno Clément3, Valérie Fessard2, Ludovic Le Hégarat2 1French Agency for Food, Environmental and Occupational Health & Safety (ANSES), Toxicology of contaminant Unit - Proteomics and metabolomics facility (PROMETHEE), Fougères, France 2ANSES Fougères, Toxicology of contaminant Unit, Fougères, France 3INSERM, University of Rennes, INRAE, Institut NuMeCan (Nutrition, Metabolisms and Cancer), Rennes, France 4Sorbonne University, CNRS, Institut de Biologie Paris-Seine (IBPS), Protein Engineering Platform, Molecular Interaction Service, Paris, France 5ANSES Maisons-Alfort, Pesticides and Marine Biotoxins (PBM) unit, Maisons-Alfort, France Background and aims: Chlordecone (CLD) is a pesticide used in the French West Indies (FWI) until the 1990s. Despite its ban, CLD persists in soils and contaminates food webs. In 2014, CLD was detected in 92 to 95% in the human population of the FWI. In humans, CLD is sequestered in the liver and poorly metabolized into chlordecol (CLDOH) and glucuronide conjugates. As it is thought to be involved in hepatic fibrosis and cancer besides other pathologies, it is a major public health concern. The aim of this study was to compare the fate and mechanistic effects of CLD and CLDOH in hepatotoxicity as well as their impact on metabolic dysfunction-associated steatotic liver disease (MASLD) susceptibility using in vitro liver cell models. Methods: A concentration-response and a kinetic study were performed in human HepaRG cells to investigate the impact of CLD and CLDOH on liver response using metabolomics and proteomics. The responses of HepaRG cells were compared with those obtained in primary human hepatocyte (PHH) cultures. Docking experiments, surface plasmon resonance (SPR), and high-content screening were used to confirm some specific effects. A 3D- cell model of MASLD was also used to investigate whether CLD and CLDOH can affect the MASLD process. Results: Hepatocytes were more sensitive to CLD than CLDOH. CLDOH was intensively and quickly metabolized into a glucuronide conjugate, whereas CLD was mainly sequestered in liver cells. CLD and CLDOH impacted OXPHOS and slightly decreased ATP levels after a few hours of exposure but they were restored at 24 hours. CLD but not CLDOH induced a drastic depletion of Septin-2, -7, -9, -10, and -11 due to proteasomal degradation. SPR experiments indicated that CLD and CLDOH could bind to Septin-2. CLD disrupted liquid droplet (LD) location and increased saturated long-chain dicarboxylic acid (DCA) production and secretion by inhibiting fatty acid oxidation (FAO) and stearoyl-CoA desaturase (SCD) abundance. Neither CLD nor CLDOH induced steatosis, but CLD induced fibrosis in a 3D co-culture model of MASLD. Conclusions: CLD hepatoxicity is specifically driven by the degradation of septins, involved in liver steatosis, fibrosis and cancer. In contrast, its major metabolite, CLDOH, was too rapidly metabolized to induce septin degradation. Septin depletion impacted LDs, resulting in the inhibition of mitochondrial β-oxidation favoring ω- oxidation and the production of saturated long-chain DCAs. These effects, associated with the perturbation of bile acid (BA) secretion, induced hepatotoxicity and susceptibility to fibrosis. Impact and implications: The mechanisms involved in CLD hepatotoxicity via its interaction with septins were deciphered. The conversion of CLD to CLDOH reduced hepatotoxicity and fibrosis in liver organoids. These results suggest that protective strategies could be explored to reduce the hepatotoxicity of CLD. 1212

1213

P-III-1051 A high definition proteome of the human intervertebral disk: understanding lower back pain Christof Lenz1,2, Lisa Neuenroth1, Jeanes Strebe1, Oliver Mohr1, Henning Urlaub1,2, Maximilian Reinhold3 1University Medical Center Göttingen, Department of Clinical Chemistry, Goettingen, Germany 2Max Planck Institute for Multidisciplinary Sciences, Bioanalytical Mass Spectrometry, Goettingen, Germany 3University Medical Center Göttingen, Clinic for Trauma Surgery, Orthopedics & Plastic Surgery, Goettingen, Germany Introduction: Chronic lower back pain is an extremely common musculoskeletal condition that severely impacts quality of life and frequently leads to disability. Indeed, it is the highest ranking musculoskeletal disorder as measured by years lived with disability worldwide [1]. One of its leading causes is the degeneration of the lumbar intervertebral disk [2]. While both pharmaceutical and surgical options are available for symptomatic treatment, there is no clear understanding of the cause(s) of the condition nor its pathogenesis, partially due to the lack of suitable disease models. Global proteomics has been used to interrogate human samples from both intervertebral-healthy and -degenerate donors, however met with limited success due to limitations in sampling and analytical methodology [3]. In the light of recent advances in tissue proteomics [4], we have applied DIA-MS analysis to defined regions of lumbar disks from intervertebral-healthy donors to generate a high definition spatially resolved proteome as a basis to study intervertebral disc degeneration in patient samples. Methods: Disk tissue from intervertebral-healthy donors was washed to remove blood, then sectioned to represent different functional regions. Samples were lysed by pressure cycling in a Barocycler 2320XT (PBI), digestion with trypsin using an optimized SP3 protocol in a Kingfisher Duo Prime (Thermo Fisher Scientific) and analysed by 120 min diaPASEF acquisitions in a timsTOF Pro 2 mass spectrometer (Bruker). Data were analysed by directDIA using Spectronaut v18.5 (Biognosys) against the human UniProtKB reference proteome v08/23. Differentially abundant proteins were analysed for functional enrichment using the ShinyGO v0.80 tool (SD State University). Results: Proteomic profiling of samples (n=4) from the Nucleus pulposus (NP) and the Annulus fibrosus (AF) as well as from adjacent regions of the disk led to detection and quantitation of a total of 8721 protein groups evidence by 153856 precursors, all at 1% FDR, at good quantitative reproducibility, with in-group CVs of 23.8- 34.3%. Differential abundance analysis showed highly regionalized proteomes, with a strong enrichment of extracellular matrix (ECM), granule- and vesicle- associated proteins in NF, and of proteins representing cellular organization (ribosomal, mitochondrial, focal adhesion) in AF. The results show that deep spatial proteomics of the intervertebral disk is readily possible, and can be utilized to study processes of degeneration. We plan to extend our approach to analysis of the glycoproteome in the future. Literature [1] Hoy et al., PMID: 24665116; [2] Kirnaz et al., PMID: 34929784; [3] Rajasekaran et al., PMID: 32973250; [4] Neuenroth et al., bioRxiv (2024). 1213

1214

P-III-1052 Isolation and Identification of mediators of heart failure in patients with cardiovascular disease James Eliott Lohman1, Katharina Schütt1, Joachim Jankowski1, Vera Jankowski1 1RWTH-UK Aachen, Institut für Molekulare Herz-Kreislaufforschung, Aachen, Germany Introduction: Cardiovascular Disease (CVD) covers a wide range of disorders and dysfunctions, and a growing global issue as the leading cause of death worldwide. Heart Failure (HF) is recognized as the most serious CVD complication, carrying a high rate of mortality. Risk of CVD and HF are correlated to several factors, but the rate of incidence is not fully explained by those risks. The underlying biochemical mechanism responsible for HF and it is probable that HF in relation to CVD may be the result of one or more unidentified proteins in the blood. In previous work from this lab, key peptides were identified in the plasma of pigs which proved to have a strong bradycardiac effect which could lead to HF, primarily those from the protein Peptide YY (PYY). The frequency in hearts of rats were severely reduced in each case when exposed to key peptides from PYY, indicating that they have a negative inotropic effect. Methods: Figure 1: Characteristic MS mass spectrogram of isolated proteins from a A) decompensated and B) compensated CVD patient. Samples of blood plasma from patients were split into 3 cohorts: patients with CVD experiencing cardiac episodes (decompensated), patients with CVD after those episodes, in absence of symptoms (compensated), and healthy patients with no history of CVD. Proteins in the blood plasma from each patient were first deprotonated, then separated and isolated using a series of chromatographic techniques including size exclusion, ion exchange, and reverse-phase separation. Mass spectrometry analyses (MS and MS/MS) were then performed to screen the samples for the targeted peptides. Upon detecting the presence of the targeted peptides with MS, the identities were confirmed via MS/MS fragmentation techniques and correlated to the contributing parent proteins. Results: Figure 2: MS/MS of 687 Da mass peak found in decompensated patient. Fragments of PYY2 have been observed in decompensated samples. One fragment of interest can be observed in Figure 1A, corresponding to m/z peak at 687 Da. In Figure 2 the MS/MS confirmation for the fragment is displayed, identifying it as a peptide with the sequence PEAPGEDA unique within the neuropeptide-Y family of proteins to PYY2. The presence of sub-fragments of the parent peak indicates that the peptide sequence is present and among other sequences identified with the same technique, providing strong evidence for the presence of PYY2, within the decompensated patient sample. It is notable that peaks at 687 Da are absent in compensated samples (Figure 1B). Discussion The hormone PYY2 was observed to be prevalent within decompensated patients. The PYY2 sequence is largely concurrent with the peptide observed to affect heart frequency in pigs. Determining the concentration of PYY2 is an on-going task and investigation of the cardiovascular effects of the protein will be elucidated using physiological assays. 1214

1216

P-III-1053 Unveiling the distinct microproteome in cell lines and tissues derived from diverse stages of colorectal cancer Alyssa Zi-Xin Leong1, Pey Yee Lee1, M. Aiman Mohtar1, Teck Yew Low1 1UKM Medical Molecular Biology Institute, Kuala Lumpur, Malaysia Recent research has shown that genomic regions once considered noncoding may contain short open reading frames encoding functional microproteins. Microproteins such as PIGBOS1 and NoBody are associated with cancer hallmarks. PIGBOS1 increases resistance to ER stress and apoptosis, while NoBody is crucial for nonsense- mediated mRNA decay, both regulating essential biological processes in cancer cells. In this study, microproteins were enriched from various stages of colorectal cancer (CRC) cell lines, including SW1116 (stage 1), SW480 (stage 2), SW48 (stage 3), COLO320DM (stage 3), COLO205 (stage 4), and CCD841CoN (noncancer control cells), using 0.25% acetic acid precipitation. These samples were analyzed using a nanoElute® system coupled with a Bruker tims-TOF Pro mass spectrometer. Additionally, publicly available TMT-10 plex MS raw data from CRC tissues (stages 1 to 4) were downloaded from CPTAC and reanalyzed. From the CRC cell lines, 207 microproteins were identified, with 167 showing differential expression. These microproteins exhibited significant motifs associated with functions like DNA binding and morphogenesis. SwissProt-identified microproteins were linked to cellular biosynthesis, RNA splicing, and mitochondrial activities, participating in cancer pathways. Using BLASTp, InterPro domain, and BioPlex interactome analyses, OpenProt (unannotated) microproteins were characterized, revealing sequence similarities with domains linked to RNA recognition and TCP-1-like chaperonins. Interactome analysis uncovered connections to growth factor binding, the NOD-like receptor pathway, and SCF-dependent proteasomal processes. Reanalysis of CPTAC data identified 205 microproteins, with 66 showing differential expression. In this dataset, SwissProt-identified microproteins were immune response suppression, suggesting roles in cancer extracellular matrix (ECM) remodeling and immune evasion. KEGG pathways highlighted their involvement in IL-17 and NOD-like receptor signaling. OpenProt microproteins resembled actin proteins and were part of actin family proteins and ribosomal subunits, with interactome data associating them with actin cytoskeleton pathways. in muscle system processes and involved Notably, PIGBOS1 and NoBody were identified and quantified in both cell lines and CPTAC data. With a cutoff value of -1 < |Log2FC| < 1, and p=0.05, PIGBOS1 was downregulated in all stages of CRC cell lines relative to the control except stage 1 (SW1116). Conversely, NoBody was downregulated in early-stage CRC cell lines (SW1116, SW480, SW48). However, both PIGBOS1 and NoBody were not significantly differentially expressed in the CPTAC CRC tissue data given the same cutoff. 1216

1217

P-III-1054 Phosphorylation of MDH2 at S246 renders cardioprotection against hypoxia/reoxygenation injury Jubert Marquez1,2, Nammi Park1, Maria Victoria Faith Garcia1, Jessa Flores1, Hyoung Kyu Kim1, Jin Han1, Jeong Rim Ko1 1Inje University, Busan, South Korea 2De La Salle University, Biology, Manila, Philippines Mitochondrial malate dehydrogenase (MDH2) regulates the conversion of malate/NAD+ to oxaloacetate/NADH. Proteomic studies show that MDH2 is involved in cardiovascular diseases wherein the expression of specific mitochondrial proteins is altered under ischemia/reperfusion conditions. Screening of phosphorylation sites of MDH2 revealed several posttranslational modifications occurring in the protein. However, it is still unclear how alterations of these sites lead to the altered activity and function of the system under pathologic models. This study focuses on the importance of MDH2 phosphorylation and how it regulates mitochondrial biogenesis and mitochondrial function. Phosphorylation occurs at various conserved sites, specifically Y56, Y80, Y161 and S246. Phosphorylation mutants based on these sites were manufactured using expression vectors and were transfected into cellular models for analysis. Phosphomutants decreased cell viability, ATP production, and MDH2 activity. Mitochondrial biogenesis marker expressions were also altered by phosphomutants under hypoxia/reoxygenation conditions. Overall, this study demonstrates the novel findings suggesting the unique role of the phosphorylation of mitochondrial MDH2 in the cardiac setting. We have identified phosphomutations in MDH2 which can be potential targets for future in vivo cardiac investigations. The protective mechanism of how MDH2 confers cardioprotection still remains unclear. Further studies on the role of MDH2 phosphorylation in mitochondrial biogenesis can elucidate the importance of MDH2 in cardiovascular disease models and treatment modalities. Fig. 1 (optional) 1217

1218

P-III-1055 Identification of potential urinary protein biomarkers in colorectal cancer: a pilot study using a proteomic approach Mickaella Michelson Martins1, Bruno Augusto Alves Martins2, Raphaela Menezes de Oliveira1, Kelly Grace Magalhaes1, Paulo Gonçalves de Oliveira2, Aline Maria Araujo Martins2,3 1University of Brasilia, Instituto de Ciências Biológicas, Brasília, Brazil 2University of Brasilia, Surgery, Brasilia, Brazil 3Scripps Institute, Molecular Medicine, San Diego, CA, United States Colorectal cancer (CRC) is among themost diagnosedmalignancies worldwide, and it is also the second leading cause of cancer-related deaths. Despite recent progress in screening programs, noninvasive accurate biomarkers are still needed in the CRC field. In this study, we evaluated and compared the urinary proteomic profiles of patients with colorectal adenocarcinoma and patients without cancer, aiming to identify potential biomarker proteins. Urine samples were collected from 9 patients with CRC and 9 patients with normal colonoscopy results. Mass spectrometry (label-free LC—MS/MS) was used to characterize the proteomic profile of the groups. Ten proteins that were differentially regulated were identified between patients in the experimental group and in the control group, with statistical significance with a p value ≤ 0.05. The only protein that presented upregulation in the CRC group was beta-2-microglobulin (B2M). Subsequent studies are needed to evaluate patients through different analysis approaches to independently verify and validate these biomarker candidates in a larger cohort sample. Fig. 1 (optional) 1218

1219

P-III-1056 Archaic NOVA1 alters FOXP2-expressing neurons upon lead exposure Aline Maria Araujo Martins1, Janaina Sena de Souza2, Luiz Pedro Petroski3, Sandra M. Sanchez-Sanchez2, Jose Oviedo4, Roberto H. Herai3, Bernardo Lemos4,5, Blake L. Tsu6, Christopher A. Barnes6, John Yates1, Alysson R. Muotri2 1Scripps Research, Molecular Medicine, San Diego, CA, United States 2University of California, San Diego, Department of Pediatrics and Cellular, San Diego, CA, United States 3Pontifícia Universidade Católica do Paraná, School of Medicine and Life Sciences, Curitiba, Brazil 4University of Arizona, Department of Pharmacology and Toxicology, Tucson, AZ, United States 5Harvard T. H. Chan School of Public Health, Program in Molecular and Integrative Physiological Science and Department of Environmental Health, Boston, MA, United States 6NGeneBioAI, Inc., San Diego, CA, United States A complex interplay between genetics and environmental factors shapes the evolution of cognitive function and brain architecture. Notably, 61 protein-coding genes, including neuro-oncological ventral antigen 1 (NOVA1), distinguish modern humans from extinct hominins like Neanderthals and Denisovans. When reintroduced into human brain organoids, the archaic variant of NOVA1 altered neural development. However, the selective pressures that favored the modern NOVA1 allele remain elusive. Single-cell proteomics-based mass spectrometry and single-cell transcriptomics analysis from dissociated brain organoids of Homo sapiens model and Neanderthal control and treated organoid were used in this study. Here, we demonstrate that lead exposure disproportionately impacts cell populations in brain organoids carrying the archaic NOVA1 variant. Importantly, this exposure suppresses FOXP2 expression in cortical neurons, implicating language and speech abnormalities in humans carrying the archaic genetic variant. Single-cell proteomics data corroborate those findings as proteins related to apoptosis induced DNA fragmentation, RHO GTPases activate PAKs, NABA ECM regulators, dopamine receptor signaling pathway, substantia nigra development, protein localization to chromatin, antiviral mechanism by IFN-stimulated genes and synapse organization were impaired in lead treat brain organoids. Our findings unveil a potential evolutionary advantage of the modern NOVA1 allele, mitigating the deleterious effect of lead toxicity. Fig. 1 (optional) 1219

1220

P-III-1057 Quantitative analysis of the extracellular matrix in murine heart tissue Juliane Merl-Pham1, Manar Elkenani2,3, Fabian Gruhn1, Stefanie Hauck1, Karl Toischer2,3, Belal A. Mohamed2,3 1Helmholtz Munich, Metabolomics and Proteomics Core, Neuherberg, Germany 2Universitätsmedizin Göttingen, Klinik für Kardiologie und Pneumologie, Goettingen, Germany 3DZHK (German Centre for Cardiovascular Research), Goettingen, Germany The cardiac extracellular matrix (ECM) plays a crucial role in providing structural, biochemical, and physiological support to the heart. Detecting ECM proteins through unbiased proteomic methods is challenging due to their solubility issues and relatively low abundance compared to the more prevalent cytoplasmic and mitochondrial proteins in the cardiomyocytes. To delineate proteome-wide and ECM-specific proteome composition in wild type (C57BL/6N) murine hearts, we conducted label-free quantitative mass spectrometric analyses of both soluble proteins and ECM extracted from myocardial tissues. The heart tissue was mechanically lysed, sonicated to solubilize non-ECM proteins efficiently, and separated by centrifugation into soluble and insoluble fractions before undergoing tryptic digestion for subsequent analysis on a QExactive HF-X mass spectrometer. Data-independent acquisition was used for the soluble fraction, while data-dependent acquisition was used for the insoluble/ECM fraction. Protein identification and quantification were carried out using Spectronaut 18 (soluble fraction) or Proteome Discoverer 2.5 (insoluble/ECM fraction), followed by statistical analysis and data visualization in Perseus software. In the soluble fraction we quantified over 4000 proteins, providing a detailed and comprehensive view of the molecular processes involved in the myocardial tissue. In the insoluble ECM fraction, we quantified more than 2000 proteins, including 155 known ECM proteins. This included 20 different collagen chains, 57 ECM glycoproteins, and 10 proteoglycans, providing a comprehensive understanding of the core matrisome in murine heart. The high-quality MSMS spectra also allowed for studying collagen chain stoichiometries and site-specific oxidations, giving insights into triple-helix structures and stability. In conclusion, our study demonstrates the successful application of our established ECM enrichment protocol previously used on lung tissue to heart tissue. This comprehensive proteomics analysis contributes to our understanding of the physiological structure of the cardiac proteome and ECMome, offering a valuable resource for future studies on diseased hearts. 1220

1221

P-III-1058 Proteomic profiling of remodeling in hearts of an animal model of cardiometabolic syndrome Emily Meyer1, Carolin Sailer1, Finn Benned Hansen1, Peter M. H. Heegaard2, Alicia Lundby1 1University of Copenhagen, Biomedical Institute, Copenhagen, Denmark 2Technical University of Denmark, Department of Health Technology, Kongens Lyngby, Denmark The wild pigs of Ossabaw Island, USA, exhibit a remarkable propensity for obesity and develop metabolic syndrome (MetS) when subjected to prolonged high-energy diets: closely resembling the state observed in metabolically unhealthy obese humans. Within nine weeks on a high-fat, high-fructose (HFHF) diet, Ossabaw pigs nearly double their body fat percentage and exhibit symptoms of MetS including a cluster of risk factors for cardiovascular disease. Consequently, the Ossabaw pig presents as an animal model for studying obesity and MetS associated diseases. Herein, we set out to investigate the proteomic changes in hearts associated with cardiometabolic syndrome. To this end, we investigated cardiac protein remodeling consequent to diet-induced obesity in the Ossabow pig model. Thirteen Ossabaw pigs were fed either a HFHF diet or a chow diet for 12 weeks. By the end of the 12- week period, the pigs on the HFHF diet exhibited a significant increase in body weight compared to the pigs on the chow diet. Heart tissue samples were collected from the ventricular wall and analyzed by quantitative proteomics using a TMT labeling approach. The approach led to measurement of more than 6,000 proteins. The protein remodeling in the hearts of the obese animals is evaluated for regulation of proteins involved in insulin signaling pathway, as well as protein remodeling observed in hearts of obese humans, in addition to global remodeling analyses. By analyzing the proteomic differences between pigs fed on the different diets, we outline the cardiac protein remodeling in the Ossabaw pig model for cardiometabolic syndrome. Fig. 1 (optional) 1221

1222

P-III-1059 Global proteomic profiling of a patient-derived 3D cyst model and reference tissue from polycystic kidney disease Alina Meyer1, Bola Khalil2,3, Margarita Iljin4, Hester Bange4, Leo Price4, Dorien Peters5, Per Artursson1 1Uppsala University, Department of Pharmacy, Uppsala, Sweden 2Johnson & Johnson, Computational Chemistry, Beerse, Belgium 3Leiden Academic Centre for Drug Research, Division of Drug Discovery and Safety, Leiden, Netherlands 4Crown Bioscience Netherlands B.V., Leiden, Netherlands 5Leiden University Medical Center, Department of Human Genetics, Leiden, Netherlands Autosomal dominant polycystic kidney disease (ADPKD) is the most common hereditary kidney disease eventually leading to end-stage renal disease. The progressively expanding, fluid-filled renal cysts are associated with proliferation, inflammation, fibrosis and Warburg-like metabolic reprogramming. Despite the relatively high prevalence, treatment options are currently limited to a single approved drug. While transcriptomic studies have elucidated deregulated pathways and potential drug targets, comprehensive proteomic profiling of human patient-derived samples remains largely unexplored. Here, we investigated the utility of human healthy and late- stage ADPKD patient-derived three-dimensional cyst cultures used in phenotypic drug screens as an in vitro model for ADPKD research. First, we analyzed the proteomes of 3D cyst cultures derived from healthy and diseased tissues using mass spectrometry-based quantitative global proteomic profiling. We then compared the protein expression profiles with those of reference tissues derived mainly from the same patients. We complemented our enrichment analyses with a database-based knowledge graph that links our differentially expressed proteins with diseases, pathways, biological processes and drugs. In addition, we quantified proteins affecting drug disposition in ADPKD as well as proposed drug targets for treatment of ADPKD. Our results show that proteomic profiling of human 3D cyst cultures reflects previously reported pathophysiological alterations, including aberrant protein expression in inflammation and metabolism. In addition, downregulated protein synthesis was observed in ADPKD. The knowledge graph allowed us to explore differentially expressed proteins in a multidimensional network and facilitated the biological interpretation of the data. While the 3D cyst cultures partially recapitulated the disease phenotype in vitro, transporter expression was notably reduced compared to in vivo conditions. Finally, known target proteins for promising drug repurposing candidates such as the AMP- activated protein kinase showed similar expression in vitro and in tissues, supporting the suitability of the 3D cyst model for ADPKD drug screenings. In summary, our results provide a multifaceted insight into the ADPKD proteome and offer a starting point for further research to improve treatment options for affected individuals. Fig. 1 (optional) 1222

1223

P-III-1060 Global proteomics for routine assessment of in vitro 3D liver models used for drug development Evgeniya Mickols1, Niklas Handin1, Alina Meyer1, Stefania Koutsilieri2, Volker M. Lauschke2, Maria Karlgren1, Per Artursson1 1Uppsala University, Pharmacy, Uppsala, Sweden 2Karolinska Institute, Physiology and Pharmacology, Stockholm, Sweden Introduction: 3D spheroids of primary human hepatocytes (3D PHH) retain a differentiated phenotype with largely conserved metabolic function and proteomic fingerprint over weeks in culture. As a result, 3D PHH are gaining importance as a model for drug discovery. However, different protocols are applied and comprehensive characterization of 3D PHH phenotypes is typically missing in the studies. One such gap is the quantification of human absorption, distribution, metabolism, and excretion (ADME) proteins across diverse primary human donors and culture conditions. In this regard, global proteomics serve as an indispensable tool for the large scale characterization of ADME protein expression along with detailed description of 3D PHH phenotype. Aim: To integrate global proteomics analysis in the routine phenotypical assessment of the 3D PHH cultures, which subsequently will improve drug development applicability and mechanistic understanding of these cultures. Methods: The 3D PHH were cultured for one to three weeks in a chemically defined physiologically relevant medium. Applicability for the drug development studies was demonstrated by ADME protein expression pattern. In brief, bottom-up global proteomics analysis constituted of controlled digestion by LysC and trypsin, peptide separation on an EASY-spray C18-column, and analysis on an Orbitrap Q Exactive HF mass spectrometer operating in a data-dependent mode. Proteins were identified by searching MS and MS/MS data of peptides against a reference human proteome using MaxQuant software. ADMExtract, an in-house developed streamlined workflow for mining of ADME protein was used to summarize the expression pattern of ADME proteins Results: We demonstrate that 3D PHH proteomes largely vary dependent on the culture conditions like cell culture medium or drug exposure, and these quintessential phenotypic differences are hard to detect with other basic functional culture assessment tests.1,2 We further confirm quantitative gene expression analysis routinely applied in drug exposure studies does not provide a correct representation of expression of drug metabolizing enzymes and transporters.2 To further integrate global proteomics analysis in drug development pipelines, we have optimized the analytical workflow for 3D PHH. We have demonstrated that the amount of initial material and sample preparation method directly affects the understanding of ADME proteins expression.3 Lastly, we have optimized a user-friendly in-house developed streamlined workflow for mining of ADME proteins and simplification of global proteomics analysis for phenotypic description of 3D PHH, that could be used by non- proteomics experts. Conclusion: Here we demonstrate that global proteomics analysis is indispensable for understanding 3D PHH phenotypic features. It provides a comprehensive understanding of ADME proteins expression and directly reflects conclusions drawn from 3D PHH experiments performed in drug development. 1223

1225

P-III-1061 Changes in the cellular proteome and secretome in response to exposure of mouse renal mesangial cells to Bothrops jararaca snake venom Jorge Moschem1, Alison F A Chaves1, Bianca C S C Barros1, Solange M T Serrano1 1Butantan Institute, Laboratory of Applied Toxicology, São Paulo, Brazil Background & aims: Snake venoms are composed of several families of protein toxins, in addition to bioactive peptides. In Brazil, Bothrops snakes are responsible for ~90% of human accidents, in which the clinical outcomes after the bite involve local and systemic effects, including kidney injury, which may progress to renal failure. To better understand the mechanisms of venom-induced renal toxicity, the aim of the present study is to evaluate the effects of B. jararaca venom on murine glomerular cells in culture with a data-independent acquisition (DIA)- mode proteomics approach. Methods: 5x106 SV40 MES 13 cells, derived from the murine renal glomerulus, were incubated with increasing concentrations of B. jararaca venom (0.25; 0.5; 1.0; 2.0 and 4.0 μg/mL) for 2 h. After cell viability assays, proteins of the cell content and secretome were analyzed by LC-MS/MS using a Vanquish Neo/Orbitrap Exploris 480 system. Results: 8,157 unique proteins were identified in all samples of the cell content, of which 92% (7,489 proteins) were quantified in terms of their relative abundance in label-free analysis. The data showed a decrease in cell viability and abundance of proteins related to cell adhesion, cytoskeleton structure, DNA repair and metabolism. Furthermore, at venom concentrations of 2.0 and 4.0 μg/mL, proteins related to apoptosis, inflammation and oxidative stress were observed with increased abundance. In the secretome, 5,866 unique proteins were identified, and 4,580 proteins were quantified. At venom concentrations of 2.0 and 4.0 μg/mL, proteins involved in several cellular pathways were identified in higher abundance, such as the inflammatory response. Conclusions: Taken together, these results indicate important changes in kidney glomerular cells upon treatment with B. jararaca venom in vitro and also demonstrate the potential of LC-MS/MS-based proteomics in studying the cellular response to a complex mixture of toxins. 1225

1226

P-III-1062 Proteomic investigation of pathogenesis mechanisms and the role of ITIH4 in PM2.5-induced lung injury Manh Quan Nguyen1,2, Jing-Ping Liou2, Chia-Li Han3,4 1Can Tho University of Medicine and Pharmacy, Faculty of Pharmacy, Can Tho, Viet Nam 2Taipei Medical University, PhD. Program in School of Pharmacy, Taipei, Taiwan 3Taipei Medical University, Master Program in Clinical Genomics and Proteomics, College of Pharmacy, Taipei, Taiwan 4Taipei Medical University, Research Center of Thoracic Medicine, Taipei, Taiwan Exposure to fine particulate matter (PM2.5) air pollution has detrimental effects on human health with increased risk of mortality and progression of pre-existing illnesses. Our previous studies have reported that the expression level of inter-alpha-trypsin inhibitor heavy chain H4 (ITIH4) was negatively correlated to PM2.5 exposure and emphysema severity and may have a role in inflammation. However, how the PM2.5 causes damages in lung at molecular level and the detailed role of ITIH4 in this pathogenesis remain largely unknown. Thus, we aim to elucidate the underlying mechanisms of PM2.5-caused lung injury and examine the molecular interactions of ITIH4 in BEAS-2B, a lung epithelial cell, by using quantitative proteomic analyses. Two doses (high and low doses) of PM2.5 were selected to treat BEAS-2B in comparison with untreated control. In the results, we observed dysregulation of idiopathic pulmonary fibrosis (IPF) and the macroautophagy signaling pathway which is mediated by misfolded collagens. Activation of ferroptosis, acute phase response signaling, pulmonary fibrosis idiopathic signaling pathway, and hepatic fibrosis as well as suppression of NAD and sirtuin signaling pathways were observed at both low- and high-dose treatments. The protection mechanisms such as NRF2-mediated oxidative stress response and wound healing signaling pathways were triggered only under high-dose PM2.5 treatment. In BEAS-2B cells, we explored the ITIH4-interacting network by using the optimized affinity purification mass spectrometry workflow and detected 49 proteins confidently interacted with ITIH4, including ITIH1, ITIH2, and ITIH3 as well as proteins involved in complement activation, coagulation, and intermediate filament cytoskeleton organization. Based on these findings, we are able to construct the detailed molecular mechanisms underlying PM2.5-induced lung injury and clarify the ITIH4-mediated cellular functions. Potential targets could be proposed to seek possible therapeutic strategies to restore the cell repair or protection ability against PM2.5. 1226

1227

P-III-1063 CRBN degrades Cav1.2α to regulate cardiac dysfunction Nammi Park1, Jubert Marquez1, Maria Victoria Faith Garcia1, Jessa Flores1, Jeong Rim Ko1, Jin Han1 1Inje University, Busan, South Korea Cereblon (CRBN) is a substrate receptor of the E3 ubiquitin ligase complex, which has been reported to target ion channel proteins. Modulation of the L-type voltage-dependent Ca2+ channel α1c subunit (Cav1.2α) is a critical player in heart failure associated with reduced ejection fraction (HFrEF), but the underlying cellular mechanisms are unclear. Here, we explored the role of CRBN in HFrEF by investigating the direct regulatory role of CRBN in Cav1.2α activity, and how it can serve as a target to address myocardial dysfunction. Cardiac tissues from HFrEF patients exhibit increased levels of CRBN, suggesting its involvement in heart failure (HF). In vivo and ex vivo animal model studies demonstrate that whole body CRBN-knockout (CRBN-/-) and cardiac-specific knockout mice (Crbnfl/fl/Myh6Cre+,) enhanced cardiac contractility with increased L-type calcium channel current (ICaL), modulated by the direct interaction of CRBN with Cav1.2α subunit. Mechanistically, we discovered that the Lon domain of CRBN directly interacts with the N-terminal of Cav1.2α. Increasing CRBN levels enhanced ubiquitination and proteasomal degradation of Cav1.2α and decreased L-type Ca2+ channel currents. By contrast, genetic or pharmacological depletion of CRBN via TD-165, a novel CRBN-targeting degrader, increased the surface expression of Cav1.2α and enhanced L-type Ca2+ channel currents. We have also discovered that CRBN-/- mice exhibited resistance to doxorubicin-induced HF. findings indicate that CRBN selectively degrades Cav1.2α, which in turn facilitates cardiac dysfunction, suggesting that reducing CRBN levels could serve as a new target of cardioprotective therapeutic strategy for HF. 1227

1228

P-III-1064 Exercise mitigates cisplatin-induced neurotoxicity in murine hippocampus via attenuation of neuroinflammation and synaptic plasticity SeHwan Park1, Jeong Rim Ko1, Nammi Park1, Jubert Marquez1, Maria Victoria Faith Garcia1, Jessa Flores1,2, Jin Han1,2 1Inje University, Cardiovascular and Metabolic Disease Core Research Support Center, Busan, South Korea 2Inje University, Department of Physiology, College of Medicine Graduate School, Busan, South Korea Chemotherapy-induced cognitive impairment is recognized as the most typical symptom in patients with cancer that occurs during and following the chemotherapy treatment. Recently many studies focused on pharmaceutical strategies to control the chemotherapy side effects, however it is far from satisfactory. There may be a need for more effective treatment options. The aim of this study was to investigate the protective effect of exercise on cisplatin-induced neurotoxicity. Eight-week-old C57BL6 mice were separated into three group: normal control (CON, n=8); cisplatin injection control (Cis-CON, n=8); cisplatin with aerobic exercise (Cis- EXE, n=8). Cisplatin was administered intraperitoneally at a dose of 3.5 mg/kg/day. The Cis-EXE group exercise by treadmill running (14~16m/min for 45 min daily, 3 times/week) for 12 weeks. Compared to the CON group, the cisplatin injection groups showed significant decrease in body weight and food intake, indicating successful induction of cisplatin toxicity. The Cis-CON group showed significantly increased levels of pro-inflammatory cytokines including IL-6, IL-1β, and TNF-a in the hippocampus, while the Cis-EXE group was significantly decreased in the expression of IL-6, IL-1β, and TNF-a. In addition, compared to the CON group, the levels of synapse-related proteins including synapsin-1 and -2 were significantly reduced in the Cis-CON group, and there was a significant difference between the Cis-CON and Cis-EXE groups. Antioxidant and apoptosis factors were significantly improved in the Cis-EXE group compared with the Cis-CON group. This study suggest that exercise could be meaningful approach to prevent or improve cisplatin-induced cognitive impairment. 1228

1229

P-III-1065 Proteomic mapping of different forms of amyotrophic lateral sclerosis Anna Percio1,2, Mariagrazia Summo1, Michela Cicchinelli1,2, Mario Sabatelli3, Andrea Urbani1,2, Viviana Greco1,2 1Università Cattolica del Sacro Cuore, Department of Basic Biotechnological Sciences, Intensive and Perioperative Clinics, Rome, Italy 2Fondazione Policlinico Universitario Agostino Gemelli IRCCS, Department of Laboratory Diagnostic and Infectious Diseases, Rome, Italy 3Fondazione Policlinico Universitario Agostino Gemelli IRCCS, Nemo Clinical Center, Rome, Italy Amyotrophic Lateral Sclerosis (ALS) is a neurodegenerative disease characterized by the progressive death of motorneurons with still no convincing therapy available. Despite considerable research efforts, the exact mechanism triggering the pathological process is not fully understood yet. A complex interaction of genetic and environmental factors is involved in motorneuron damage. Variants in several genes, including SOD1(Superoxide dismutase 1), TARDBP (TAR DNA-Binding Protein 43), as well as C9orf72 (Chromosome 9 open reading frame 72), and FUS (Fused in Sarcoma) have been associated with ALS, however familial ALS (fALS) accounts for about 10% of all cases, while the remaining 90% of cases are sporadic ALS (sALS). Both forms, familiar and sporadic, are clinically indistinguishable with the same phenotype. In this perspective, this study aims at an in-depth characterization of sALS and SOD1 and TARBP-related fALS proteomes to highlight common and different molecular features depending on gene variants. To achieve this purpose, fibroblasts derived from human skin biopsies deposited in the AISLA ALS National Biobank at the Nemo Clinical Center, (Fondazione Policlinico A. Gemelli, Rome, Italy), have been used. Fibroblasts from patients with sALS (n=6), fALS (including 6 SOD1 and 6 TARBP variants), and healthy donors (n=6) have been selected. MS analysis on protein extracts has been performed using HR-DDA approach on Orbitrap Fusion Lumos Tribrid Mass Spectrometer (Thermo Scientific). Protein identification and label free quantification have been performed by Peaks and Proteome Discoverer software; statistics and bioinformatics analysis have been carried out by R, Reactome and ClueGO software. Significant protein hallmarks have been shown by making comparisons between ALS and healthy patients, as well as within the ALS sub-groups. Impaired autophagy and dysregulation of antioxidant mechanisms have been highlighted in sALS and, mainly, in SOD1- fALS. On the other hand, enhanced dysregulation of RNA metabolism seems a notable signature of TARDBP variants. Studying disease subtypes could pose as a helpful route in resolving the conundrum of ALS etiopathogenesis and finding new pharmacological targets. 1229

1230

P-III-1066 Tau protein hyperphosphorylation during hibernation of golden hamsters: insights from a brain IP-MS study Diana Piotrowska1, Elena Camporesi1, Wagner S. Brum1, Jens T. Stieler2, Carsten Jäger3, Max Holzer2, Johan Gobom1, Henrik Zetterberg1, Kaj Blennow1, Thomas Arendt2, Gunnar Brinkmalm1 1Sahlgrenska Academy, University of Gothenburg, Department of Psychiatry & Neurochemistry, Moelndal, Sweden 2Paul Flechsig Institute, University of Leipzig, Leipzig, Germany 3Max Planck Institute for Human Cognitive and Brain Sciences, Leipzig, Germany Alzheimer’s disease (AD) originates from two major brain changes, extracellular plaques of amyloid beta and intraneuronal tau protein accumulations called tangles. Whereas presence of amyloid beta plaques is a defining feature of AD and is necessary for AD diagnosis, cognitive decline correlates better with tau protein accumulation rather than with amyloid plaque load. In healthy conditions, the microtubule-binding region (MTBR) of tau protein binds to microtubules and stabilises the intracellular scaffold that allows for neurons" characteristic shape. Neurofibrillary tangles (NFTs) are composed of highly phosphorylated, misfolded tau protein and are the main cause for neuronal death in AD progression. Nowadays, phosphorylated tau peptides detected in CSF and plasma serve as diagnostic tool for classifying patients with potential AD for disease modifying treatments. Although in tauopathies hyperphosphorylation of tau lies at the core if the pathology, it is a phenomenon that also occurs under physiological conditions. Reversible hyperphosphorylation of tau has been shown during hibernation i.e., in European ground squirrels, and brown bears. Moreover, hibernating bears accumulate tau aggregates during the hibernation period. We have developed a multiplex assay combining immunoprecipitation and high-resolution mass spectrometry (IP-MS) that allows for detecting and quantifying peptides from different tau protein isoforms as well as non- phosphorylated and phosphorylated peptides, carrying up to three phosphate groups. The method can be used for detailed analysis of tau, not only in AD cases but also in research on other tauopathies and animal models. Here we have analysed tau in golden hamsters. Brain tissue (cortex) was collected from hamsters during hibernation and from those who were not hibernating at the time. The tissue was homogenised and both the soluble (tris(hydroxymethyl)aminomethane (tris)-buffered saline, TBS) fraction and the insoluble (sarkosyl- insoluble, SI) fraction were collected. Subsequently, IP was performed using two anti-tau antibodies, HT7 and 77G7. After IP, samples were trypsinased overnight. Tryptic peptides were then analysed using LC-MS using internal, heavy-labelled standards. Preliminary data shows that during hibernation hamsters have increased levels of phosphorylated tau; pTau-217, pTau-231, and pTau231+235 were all increased 3-5 times. However, there was no increase for MTBR peptides, indicating that golden hamsters do not form tau aggregates during hibernation. The data set also provides information on the truncation pattern of tau as well as other post-translational modifications. 1230

1231

P-III-1067 Do caloric restriction and aging affect mitochondrial protein oxidation in heart and brain tissue? Shipan Fan1, Carina Ramallo-Guevara2, Norbert Dencher3, Ansgar Poetsch1,2 1Nanchang University, Nanchang, China 2Ruhr University Bochum, Bochum, Germany 3Technical University Darmstadt, Darmstadt, Germany in understanding, the process of aging remains an Despite advances irreversible, complex biological phenomenon, with numerous enigmatic aspects and disputed theories/factors. The prominent mitochondrial free radical theory of ageing (MRFA) postulates an increase of (oxidatively) damaged biomolecules in mitochondria as driver for cellular ageing. Our objective is to gain insight into oxidative protein changes in mitochondria during the aging process, which are influenced by caloric restriction (CR), a crucial factor for organs such as the heart and brain. Crude mitochondrial fractions were isolated from the heart and brain tissues of same rats (Rattus norvegicus), ages 6.5 months (young) and 27 months (old), fed ad libitum or on a CR diet (CR: intake about 60 % of ad libitum fed animals). We have utilized an untargeted proteome approach, which enabled us to comprehensively compare the amount of identified oxidized peptides with non-oxidized peptides under DDA mode. For cerebrum, the total changes in oxidative posttranslational modifications (oxPTMs) were found to be relatively minor. The age of levels of heart transmembrane/mitochondrial oxPTMs, but there was a decline in cerebrum transmembrane oxPTMs with age, which was further reduced by CR. The changes in heart transmembrane/mitochondrial oxPTMs were significant, with a general trend of reduced amount of oxPTM due to CR. significantly affect the the subjects did not It is of great importance to study individual oxPTM modifications to identify aging mediators. Here, monooxidation, methionine oxidation, and carbonylation were the most frequently observed oxPTMs. Long- term and short-term CR significantly decreased monooxidation and carbonylation in heart. For cerebrum, there was no uniform trend – for instance monooxidation decreased for short-term CR, but increased under long-term CR in comparison to AL. In conclusion, these results revealed that the alterations of oxPTMs are affected by age, diet, and tissue. Concerning the MRFA theory for aging and effect of CR, the differing outcomes for cerebrum and heart suggest that other factors, such as mitochondrial proteome composition or structure, affect ROS production and/or accumulation of oxPTMS. 1231

1233

P-III-1068 Proteomic profiling of grey and white matter in the human brain using astral mass spectrometry Ernst Heinz Pulido1, Emma Lundberg1 1Stanford University, Biophysics, Stanford, CA, United States The brain is composed of two main types of tissue: grey matter and white matter. Grey matter, containing neuronal cell bodies, is primarily responsible for processing information, while white matter consists of nerve fibers that connect different brain regions, facilitating communication. This study aims to investigate the distinct roles of grey and white matter by analyzing their protein profiles, which could provide insights into their functions and involvement in neurological diseases. Fresh-frozen human brain samples from various regions were precisely dissected to separate grey and white matter. Protein extraction was carried out using a standardized protocol, followed by quantification with the advanced Thermo Fisher Astral Mass Spectrometer. Peptides were then analyzed using high-resolution tandem mass spectrometry to ensure accurate and comprehensive protein identification. Additionally, rigorous quality control measures were implemented to validate the reliability of the proteomic data obtained. Comparative bioinformatics analyses were conducted to identify significant differences in protein expression between the two tissue types. The results revealed distinct protein compositions that correlate with the specific functions of grey and white matter, enhancing our understanding of their unique contributions to brain function. Differential expression analysis and pathway enrichment were performed to uncover biological pathways and processes significantly represented in each tissue type. These findings highlight potential mechanisms underlying brain diseases like Alzheimer's disease and multiple sclerosis, offering a foundation for future studies to explore how these tissues change over time and contribute to brain health and disease. Continued research in this area could lead to significant advances in diagnosing, preventing, and treating neurological conditions, ultimately improving brain health and patient outcomes. Fig. 1 (optional) 1233

1234

P-III-1069 Uncovering proteomic alterations in Parkinson's disease using patient-specific iPSC-derived brain cells Magdalena Kuras1, Erika Velasquez1,2, Elissavet-Kalliopi Akrioti2, Yuriy Pomeshchik2, Laurent Roybon2,3, Melinda Rezeli1,4 1Lund University, Dept. of Biomedical Engineering, Lund, Sweden 2Lund University, Department of Experimental Medical Science, Lund, Sweden 3Van Andel Institute, Department of Neurodegenerative Science, The MiND Program, Grand Rapids, MI, United States 4Lund University, BioMS, Lund, Sweden Parkinson's disease (PD) is primarily associated with motor impairment due to the loss of dopaminergic neurons located in the substantia nigra pars compacta of the brain. Despite intense research, its initiation, pathogenicity, and development remains uncertain. PD can be divided into familial or idiopathic forms. Monogenic forms of PD account for ~ 30% of known familial cases but less than 10% of all cases. This means that we currently do not understand the origin of more than 90% of PD cases. In addition to specific hallmarks such as neuronal cell type- specific loss and the formation of protein inclusions, PD has intrinsic heterogeneity and many atypical cases. Patient brain cell types are not accessible during the disease. However, these can be generated from induced pluripotent stem cells (iPSCs) to identify dysfunctions in cellular pathways. Preclinical findings have suggested that iPSC-based patient stratification may be key to patient management and therapeutic interventions, similar to what is currently used in certain cancers, where omics data are expected to play a major role. Moreover, iPSC- derived brain cell types generated from patients with idiopathic disease have been shown to have the intrinsic capability to exhibit disease phenotypes that resemble those present in brain cells of well-defined genetic cases. Our goal is to identify early disease mechanisms leading to neuronal injury in PD by uncovering the relationship between the genetic makeup of a patient and its brain cells' dysfunctional networks. To this end, we performed MS-based proteomic and phosphoproteomic analyses of brain cells (dopaminergic neurons and astrocytes) and midbrain spheroids generated from patient iPSCs (n=50) available from the Parkinson's Progression Markers Initiative (PPMI) cohort. Proteomic analysis resulted in the quantification of more than 7500 proteins and 9000 phosphopeptides per sample in FACS sorted cells, and higher numbers were obtained from spheroid samples. Based on their proteomic and phosphoproteomic profiles, different brain cell types and patient groups with distinct genetic makeup can be clearly distinguished from each other. Dysregulation of certain processes was observed in all PD groups compared to controls (e.g., mitochondrial translation, integrin 3 pathway), but patient group-specific changes were also identified. In addition, we also found significant differences between the disease groups in terms of kinase activity. The use of patient iPSC-derived brain cells may help to identify initial cellular dysfunctions and their development before neuronal damage prevails and leads to the appearance of the symptoms. Our results provide insight into the pathogenic processes of PD, which may contribute to better patient stratification and the development of personalized treatment strategies. Funding: The Michael J. Fox Foundation for Parkinson's Research (MJFF-021548) 1234

1235

P-III-1070 Global protein turnover rates of primary Astrocytes in a mouse model of Parkinson's disease Giulia Rocca1, Letizia Bernardo1, Veronica Giusti2, Ilaria Battisti1, Cinzia Franchin1, Elisa Greggio2, Laura Civiero2, Giorgio Arrigoni1 1University of Padova, Department of Biomedical Sciences, Padova, Italy 2University of Padova, Department of Biology, Padova, Italy Parkinson"s disease (PD) is a complex neurodegenerative disorder characterized by the progressive loss of dopaminergic neurons in the substantia nigra. One of its main pathological hallmarks is the presence of intracellular protein inclusions (Lewy Bodies) in neuronal and glial cells. While several genes have been associated with PD etiology, point mutations in leucine-rich repeat kinase 2 (LRRK2)-encoding gene stand out as the most common cause of the familial form. LRKK2 is a multidomain kinase involved in many cellular processes, from vesicle trafficking to the regulation of lysosome homeostasis. All pathogenic mutations seem to directly increase its kinase activity, resulting in abnormal modulation of its substrates. However, how this impacts PD progression is still not completely understood. In addition, the role of LRRK2 in glial cells such as astrocytes and the involvement of these cells in the degeneration of neurons and in the neuroinflammation process have been underexplored, as most studies focused on neurons. Since impaired protein homeostasis and accumulation of protein aggregates are key features of PD pathogenesis, we investigated the potential contribution of G2019S-LRRK2, the most frequently observed mutated form, on the global turnover of proteins in astrocytes at the early stages of the disease. To systematically estimate protein turnover rates, we carried out a dynamic-SILAC experiment on primary astrocytes of wildtype and G2019S-LRRK2 knocked-in mice. A pre-treatment for 24 hours with pre-formed fibrils of alpha-synuclein, the main component of Lewy Bodies, was performed to evaluate possible differences in turnover rates in simulated pathological conditions. Cells were collected at three different time points (1 day, 4 days, 6 days) and proteins were analyzed by data-dependent acquisition (DDA) LC-MS/MS. The turnover rate was determined for each protein by applying a linear model at the peptide-level data. Our results indicate that the G2019S mutation of LRRK2 deeply affects the half-lives of proteins involved in the ER-Golgi transport and cell adhesion. As expected, some of these proteins are known substrates of LRRK2. Interestingly, the treatment with alpha-synuclein does not appear to modify protein turnover rates of wildtype astrocytes significantly. In contrast, the same treatment in astrocytes carrying the G2019S-LRRK2 mutation strongly alters the half-life of proteins, in particular those belonging to the respiratory chain complexes and those that participate in vesicle-mediated transport. Moreover, the estimation of protein abundance levels suggests a possible inhibition of autophagy through the activation of mTORC1, as already observed for other PD models. 1235

1236

P-III-1071 Proteomic insights into MrgD receptor deficiency: motor hyperactive- and compulsive-like phenotype, and synaptic dysfunction in the nigrostriatal pathway Lucas Rodrigues-Ribeiro1,2, Kivia B. S. Santos3, Bruna S. Oliveira3, Caroline A. Machado3, Nícia P. Soares2, Arkadiusz Nawrocki1, Maria José Campagnole-Santos2, Aline S. Miranda3, Cristina Guatimosim3, Robson A. S. Santos2, Martin Larsen1, Thiago Verano-Braga2 1University of Southern Denmark, Department of Biochemistry and Molecular Biology, Odense, Denmark 2National Institute of Science and Technology in Nanobiopharmaceutics (INCT-Nanobiofar), Department of Physiology and Biophysics, Belo Horizonte, Brazil 3Federal University of Minas Gerais, Department of Morphology, Belo Horizonte, Brazil The renin-angiotensin system is composed by a proteolytic cascade of angiotensinogen leading to the formation of several bioactive peptides, including alamandine, which acts as an endogenous agonist of the MAS-related G protein-coupled receptor D (MrgD). Notably, this receptor is highly expressed in brain regions associated with the control of reward, cognition, and motor functions, including the nigrostriatal pathway. Behavioral studies done by our research group showed that MrgD receptor-deficient mice (KO) exhibit motor hyperactivity and a high compulsive-like phenotype, without anxiety-like phenotype or alterations in short or long-term memory. This study aimed to assess the proteome, phosphoproteome, and N-glycoproteome of the nigrostriatal pathway in C57Bl6/J (WT) and KO mice to uncover the molecular role of MrgD in the nigrostriatal pathway. Thus, substantia nigra (SN) and striatum (ST) regions were collected for proteomics (n=4). Proteins were extracted using the acid labile detergent SDC. Tryptic peptides were labeled with TMT-16plex and submitted to TiO2 to allow the simultaneous enrichment of phosphorylated and sialic acid-containing glycopeptides. Non-modified and modified peptides samples were pre-fractionated into 20 and 12 fractions in high pH reverse-phase chromatography, respectively. Samples were analyzed using LC-MS/MS (Orbitrap Exploris 480). Raw data was deconvoluted using Maxquant against the SwissProt Mus musculus FASTA file were used. To validate proteome findings, digested peptides were loaded into the Evosep One coupled with timsTOF Pro and analyzed using a 30 SPD gradient (n=5). Data was deconvoluted in DIA-NN using the same FASTA file. An extensive functional validation was done. A total of 5,878 unique proteins, 19,209 class 1 phosphosites, and 5,301 N-glycosylated sites were identified, with 468 proteins, 190 phosphosites, and 12 N-glycosylated sites regulated (p<0.05, FC±1.2). KO mice showed alterations in biological processes related to dopaminergic synapses, motor activity control, and synaptic vesicle exocytosis. A remarkable down-regulation in phosphorylation and/or abundancy of proteins regulating Ca2+ dynamics such as Protein Kinase C (PKC), Calcium voltage-gated channel (Cav2), Ca2+/calmodulin-dependent protein kinase (Camk2α) was observed. In addition, key proteins in exocytosis such as Syntaxin (Stx1a), Syntaxin-binding protein (Stxbp1), and Complexin 2 (Cplx2), were down-regulated and glycosylated in KO group. In line with data, functional validation showed a reduction in exocytosis in the KO group, leading to an accumulation of neurotransmitters in the SN and decreased levels in the ST of KO mice. In contrast, the treatment of SN and ST with alamandine in WT increased exocytosis. In conclusion, the Alamandine/MrgD signaling has an important role in vesicle trafficking and release in the nigrostriatal system, highlighting its relevance to neuropsychiatric disorders involving motor control and impulsivity. 1236

1237

P-III-1072 Proteomic basis for pancreatic acinar cell carcinoma and pancreatoblastoma as similar yet distinct entities and their contrast to pancreatic ductal adenocarcinoma Michael Roehrl1 1Beth Israel Deaconess Medical Center, Harvard Medical School, Pathology, Boston, MA, United States Acinar cell carcinoma (ACC) and pancreatoblastoma (PBL) are rare pancreatic malignancies of acinar differentiation. We conducted proteogenomic profiling of patient tumor cohorts with comparison to pancreatic ductal adenocarcinoma (PDAC) and benign tissue. The proteome depicts ACC and PBL as similar but separate entities and distinct from PDAC. ACC and PBL are enriched with proteins involved in RNA processing and transport, chromosome organization, and the mitoribosome, whereas PDACs overexpress proteins related to actin-based processes, extracellular matrix, and immune-active stroma. We examined pathway differences in metabolic adaptation, epithelial to mesenchymal transition, and DNA repair. We discovered proteomic changes in both Wnt-CTNNB1 and -IGF2 pathways. We identified 17 ACC-specific markers that highlight ACC as a metabolic disease with mitochondrial dysfunction and 34 PBL-specific proteins that mark this pediatric cancer with an embryonic stem cell phenotype and changes in chromosomal proteins and the cell cycle. This study offers a first proteomic view of two poorly understood pancreatic neoplasms and provides new pathway and protein targets for future diagnostic and therapeutic development. 1237

1238

P-III-1074 Deep proteomic profiling of matched CSF and plasma samples for aging biomarker discovery beyond protein Aida Kamalian1, Polina Shichkova2, Marco Tognetti2, Sara Ho1, Christopher Below2, Roland Bruderer2, Yuehan Feng2, Lukas Reiter2, Michael Lutz3, Abhay Moghekar1 1Johns Hopkins University, Baltimore, MD, United States 2Biognosys AG, Schlieren, Switzerland 3Duke University, Durham, NC, United States Aging is the leading risk factor for many neurodegenerative disorders [1]. However, the transition between healthy and diseased state in advanced age is poorly understood. Furthermore, neurodegeneration related changes are present in the brain years before the symptoms' onset [2]. To elucidate potential early signs of degeneration-related changes in healthy aging, we aimed to find differential cleavage, alternative splicing and phosphorylation events in aging CSF and plasma of cognitively normal adults. To this end, we applied data- independent acquisition mass spectrometry (DIA-MS) with Biognosys" TrueDiscovery Ultra Deep offering and analysed raw data using Spectronaut® software. We quantified 5599 and 3112 protein groups, and 127560 and 94415 peptides in CSF and plasma respectively. We performed differential abundance analysis at both peptide and protein levels. Then we applied a strict filtering for contribution of technical factors and found 27 genes with potential differential cleavage and alternative splicing events in CSF, including well-known APOE, APP and APLP1 among others. Proteins of these genes are not among significantly differentially abundant. About half of these genes are among aging and longevity associated genes and about half (with intersection of 6 genes) are known to have more than one transcript. Plasma data appeared to be noisier with no genes passing the same strict criteria as for CSF. In the second part of this work, we compared fold changes of peptide abundances in aging for phosphorylated and non-phosphorylated peptides. This way, we found differentially phosphorylated proteins of 32 genes in CSF data. About one third of these genes are known as related to aging and longevity. Protein phosphorylation plays important regulatory role and is known as being affected by aging. To sum up, in this work we focused on the peptide differential abundance to perform large-scale analysis of protein cleavage, splicing and phosphorylation changes in aging. We found differential cleavage, potential alternative splicing and phosphorylation both known aging and neurodegeneration associated genes as well as novel genes for which association with aging and neurodegeneration has not yet been established. References: 1. Hou Y, Dan X, Babbar M, Wei Y, Hasselbalch SG, Croteau DL, Bohr VA. Ageing as a risk factor for neurodegenerative disease. Nat Rev Neurol. 2019 Oct;15(10):565-581. doi: 10.1038/s41582-019-0244- 7. Epub 2019 Sep 9. PMID: 31501588. 2. Beason-Held LL, Goh JO, An Y, Kraut MA, O'Brien RJ, Ferrucci L, Resnick SM. Changes in brain function occur years before the onset of cognitive impairment. J Neurosci. 2013 Nov 13;33(46):18008-14. doi: 10.1523/JNEUROSCI.1402-13.2013. PMID: 24227712; PMCID: PMC3828456. 1238

1239

P-III-1075 A Study on the expression profile of genes coding for secretory proteins using in vitro model of Alzheimer's disease Ravinder Singh1, Aditya Sunkaria1 1Guru Nanak dev University, Amritsar, BIOTECHNOLOGY, Amritsar, Punjab, India Background: Age-related neurological illnesses like Alzheimer's disease (AD) are steadily becoming more prevalent among the elderly population in India and around the world. Cognitive deficits are caused by a progressive loss of normal brain functions. Increased production of amyloid (Aβ) and the development of neurofibrillary tangles (NFTs) are the two most significant pathogenic events that take place during AD progression. Until today, many attempts have been made towards discovering novel targets that are more precise and could identify the progression of the disease at a very early stage. Method: In the present study, the Human Protein Atlas (HPA) database was used to filter out the total human secretome. Using the single-cell type proteome of the HPA database, neuronal proteins were compared with the total human secretome to discover overlapping proteins and their respective genes. Further, Alzheimer's disease-associated genes were extracted using the DisGeNet database and ranked according to their gene- disease association (gda) score. Further, different proteoforms of Aβ were used to treat neuronal cells in vitro to mimic the early stages of AD-like pathology. The expression of secretory protein genes in an in vitro model system has been analysed. Result: Out of 1678 genes, 24 neuronal genes were identified that encoded secretory proteins. These genes were further categorized according to their protein function. Specific primers for the top GDAS score genes were designed, and expression analyses were conducted. Analyses revealed modulation in the expression of the APOE, MAPT, NPY, NGF, and RELN genes. Conclusion: The aim of this study was to discover early-stage biomarkers for Alzheimer's disease. In the present study, we have tried to identify modulation in the genes coding for secretory proteins, which might be reflected at protein levels as well. Currently, our lab is working on fully understanding the modulation of the AD secretome, which would enable us to identify AD-specific early-stage biomarkers. Fig. 1 (optional) 1239

1240

P-III-1077 Analysis of lysine metabolism in a rat model of type 2 diabetes reveals modulation of endocytosis and glucose metabolism Yifan Tan1, Lashodya Dissanayake2, Olha Kravtsova2, Fatih Demir1, Vladislav Levchenko2, Alexander Staruschenko2,3, Markus Rinschen1,4 1Aarhus University, Department of Biomedicine, Aarhus, Denmark 2University of South Florida, Department of Molecular Pharmacology & Physiology, Tampa, FL, United States 3James A. Haley Veterans’ Hospital, Tampa, FL, United States 4University Medical Center, Hamburg Eppendorf, III Department of Medicine, Hamburg, Germany Lysine, an essential amino acid crucial for protein synthesis and metabolic processes, has shown promise in ameliorating kidney diseases associated with hypertension. However, its role in diabetic nephropathy (DN) remains poorly understood. This study aims to investigate lysine's impact on kidney function in DN and elucidate its underlying mechanisms. We hypothesize that lysine affects the kidneys by mediating protein endocytosis, altering protein abundance, and disturbing metabolic pathways. We employed proteomic approaches to investigate the effects of lysine on renal proximal tubule endocytosis and its impact on other organs in Type 2 Diabetic Nephropathy (T2DN) rats. Additionally, we used targeted metabolomics to study changes in lysine conjugates and lysine metabolic products. Human urinary proteomics indicated lysine's potential inhibition of megalin and cubilin, key receptors in endocytosis, along with effects on proteins involved in glucose metabolism. In T2DN rats, lysine exhibited tissue-specific impacts, suppressing oxidative phosphorylation and the tricarboxylic acid (TCA) cycle in kidneys, and inhibiting cholesterol synthesis and peroxisomal lipid metabolism in the liver. Metabolomic analysis revealed reduced glucose levels in both blood and urine following lysine administration. Our findings highlight lysine's regulatory role in protein composition and metabolic pathways in nephropathic models. Further experiments will be conducted to validate these mechanisms and explore lysine's therapeutic potential in kidney diseases. Fig. 1 (optional) 1240

1241

P-III-1078 Investigation of the influence of steroid receptor ligands on the neuronal proteome as a possible trigger for depression Sam Thilmany1,2, Andreas Thomas1, Yvonne Reinders3, Matthias Vogel2, Catharina Scholl2, Albert Sickmann3, Mario Thevis1,4 1German Sport University Cologne, Institute of Biochemistry, Cologne, Germany 2Federal Institute for Drugs and Medical Devices, Bonn, Germany 3Leibniz-Institut für Analytische Wissenschaften – ISAS – e.V., Dortmund, Germany 4European Monitoring Center for Emerging Doping Agents, Cologne, Germany Background: Since its introduction in 1960, hormonal contraception has been a groundbreaking advancement in women's sexual autonomy and a cornerstone in preventing unintended pregnancies. Even today, after over six decades, oral hormonal contraception remains a pivotal method in many Western nations. However, hormonal contraception has also presented challenges. Early concerns about severe adverse reactions, such as thromboembolic events and an elevated risk of breast cancer, emerged. Additionally, emerging evidence suggests a potential link between hormonal contraception and adverse effects on mental health, including mood disturbances. Nevertheless, epidemiological studies have been divergent, suggesting that group-specific factors may complicate interpretation. Consequently, a definitive causal relationship has yet to be established. This project seeks to deepen our understanding of the pharmacological effects of steroid receptor ligand administration on the neuronal proteome, aiming to elucidate potential functional changes within the central nervous system that may be associated with depressive disorders. Methods: The ReNcell VM neuronal progenitor cell line was treated with estrogenic (ethynylestradiol), progestogenic (levonorgestrel), and androgenic (S-23) steroid receptor ligands. Treatment effects were analyzed by label-based and label-free quantitative bottom-up proteomics. Statistical analysis utilized Thermo Scientific Proteome Discoverer with Sequest HT and Chimerys and a tailored R script. Results: Among all the quantified proteins, several hundred exhibited differential expression upon treatment with steroid receptor ligands. Gene Ontology analysis revealed significant regulation in biological processes, including regulation of neurotransmitter uptake, cell-substrate adhesion, and gliogenesis. Essential molecular functions were integrin binding and S100 protein binding. Regulated cellular components included the collagen- containing extracellular matrix and focal adhesion. These findings provide insights into molecular mechanisms underlying depression-related changes, highlighting pathways involved in cell adhesion and neurotransmitter regulation. 1241

1242

P-III-1079 Characterization of brain α-synuclein in neurodegenerative diseases by high-resolution mass spectrometry Kim-Thanh VAN1, Nicolas Villain2, Foudil Lamari2, Susana Casas Boluda3, François Fenaille1, François Becher1 1CEA Paris-Saclay, Gif sur Yvette Cedex, France 2Hôpitaux Universitaires Pitié Salpêtrière, Paris, France 3Sorbonne Université, ICM, Inserm, CNRS, APHP, Hôpital de la Pitié-Salpêtrière, Paris, France Introduction: α-synuclein is the main protein involved in the pathology of a number of neurodegenerative diseases, including Parkinson"s disease, dementia with Lewy bodies (LBD), multiple system atrophy, and about half of Alzheimer’s disease cases. A primary hallmark of these synucleinopathies is the presence of Lewy bodies - insoluble particles made of proteinaceous materials in which α-synuclein is the main component. The literature has reported numerous post-translational modifications (PTMs) along the sequence of α-synuclein. Most of these PTMs, however, were detected using site-specific antibodies in vitro or on animal models. In this study, we used Western blot and developed novel high-resolution mass spectrometry methods, including wide proteome and immunoprecipitation, to characterize the proteoforms of α-synuclein from post-mortem brains. Methods: We used frozen brain samples (temporal cortex) from two LBD patients and two control cases. Brain tissue (∼100 mg) was homogenized in TBS buffer containing protease and phosphatase inhibitors, using a Dounce homogenizer. We then used ultracentrifugation (100,000 g for 1 h) to fractionate brain proteins into three fractions using buffers with increasing stringency: soluble fraction (TBS), detergent-soluble fraction (2% Triton-X-100), and detergent-insoluble fraction (8 M urea + 2% SDS). Antibody-free approaches were prepared with SP3 protocol, then phosphopeptide enrichment was performed with Zr-IMAC beads. In parallel, optimized immunoprecipitation (IP) conditions were applied to the three fractions. Samples were analyzed in parallel by nanoLC-HRMS/MS and Western blot. Results: A brain fractionation protocol was adapted from literature for isolation of insoluble and soluble species of α-synuclein. Western blots showed that while truncations seemed to appear ubiquitously, high molecular weight bands and phosphorylation were intensively visible in the detergent-insoluble fraction of disease brains. To characterize these α-synuclein species, we first developed a wide proteome analysis of brain fractions by PAC, combined or not with Zr-IMAC. To improve the sensitivity of bottom-up proteomics and get more insight into the α-synuclein heterogeneity, we optimized an IP protocol with different types of magnetic beads, elution buffers, and digestion enzymes. Various antibodies with different epitopes were also evaluated. Amongst 20 antibodies tested, only 5 can capture α-synuclein during IP. However, none of them can capture α-synuclein species corresponding to the high molecular weight bands. We applied bottom-up proteomics for both wide proteome and IP approaches, to the insoluble and insoluble fractions of brain extracts. In total, we detected 9 truncated sites, 4 phosphorylation, and 4 ubiquitinated sites. Our plan is to identify and quantify these PTMs in the cerebrospinal fluid of LBD patients and controls in large cohorts. We expect that these PTMs could become robust biomarkers for the diagnosis of LBD. 1242

1244

P-III-1080 The dysfunctional and immune reactive protein profile of ALS astrocytes rescued by FGF4 is abrogated by pro- inflammatory cytokine TNFa Erika Velasquez1, Ekaterina Savchenko2, Sara Marmolejo-Martínez-Artesero3, Désiré Challuau3, Aline Aebi4, Yuriy Pomeshchik1, Nuno Jorge Lamas5, Mauno Vihinen2, Melinda Rezeli2, Bernard Schneider4, Cedric Raoul3, Laurent Roybon6 1Lund University, Lund, Sweden 2Lund University, Lund, Sweden 3INM, Univ Montpellier, INSERM, Montpellier, France 4Bertarelli Platform for Gene Therapy, Ecole Polytechnique Fédérale de Lausanne, Geneva, Switzerland 5Centro Hospitalar e Universitário do Porto, Largo Professor Abel Salazar, Anatomic Pathology Service, Pathology Department, Porto, Portugal 6Department of Neurodegenerative Science, the MiND program, Van Andel Institute, Grand Rapids, MI, USA., Grand Rapids, MI, United States Astrocytes play a critical role in the onset and progression of amyotrophic lateral sclerosis (ALS), a fatal disorder marked by the rapid degeneration of motor neurons (MNs) in the central nervous system. Despite evidence that ALS astrocytes are toxic to MNs, the earliest pathological changes leading to their neurotoxic phenotype remain unclear. In this study, we generated human astrocytes from induced pluripotent stem cells (iPSCs) of patients with ALS-associated A4V mutation in superoxide dismutase 1 (SOD1) and controls to investigate early disease- related protein alterations. Proteomic analysis revealed significant changes in proteins previously linked to ALS pathology. Notably, ALS astrocytes exhibit an increment of proteins associated with an aberrant activation of the innate immune system. The proteome profile of ALS astrocytes transduced with AAV particles encoding fibroblast growth factor 4 (FGF4) was largely reversed to resemble that of control astrocytes, suggesting a reversion of the pathological protein profile. However, trying to reduce astrocyte reactivity with AAV-FGF4 in vivo fails to prevent MN death in ALS. To investigate the underlying potential mechanism behind the treatment failure, we evaluate the response of ALS and control astrocytes in a pro-inflammatory environment. Our analysis indicated that ALS astrocytes show an exacerbated expression of proteins related to immune reactivity, indicating that TNFα could abrogated the rescuing effects of FGF4 on ALS astrocytes. Our data demonstrate that ALS astrocytes sustain elevated levels of proteins related to the cGAS-STING pathway, such as XRCC5/XRCC6, and scaffold elements for IRF3 phosphorylation after TNFa+FGF4 treatment. In conclusion, this study shows the potential of the patient-derived iPSC models using proteomics as molecular readout to capture the cell- autonomous contribution of astrocytes in the pathogenesis of ALS and to test new therapeutic approaches to fine-tune the responsiveness of astrocytes and their contribution to neuroinflammation in ALS. 1244

1245

P-III-1081 Characterization of an adipocytes: macrophages co-culture to model healthy and inflamed adipose tissue using proteomics Greta Westermeier1, Alix Sarah Aldehoff1, Martin Wabitsch2, Kristin Schubert1 1Helmholtz Centre for Environmental Research (UFZ), Molecular Toxicology (MOLTOX), Leipzig, Germany 2Universitätsklinikum Ulm, Pädiatrische Endokrinologie und Diabetologie, Ulm, Germany According to the World Health Organization, the current prevalence of obesity is considered pandemic, with 43% of adults globally being overweight in 2022, defined as having a body mass index (BMI) greater than 25, and 16% being obese (BMI>30), with a rising trend. Apart from genetic predisposition, hypercaloric nutrition, and low physical activity, endocrine-disrupting chemicals (EDCs), e.g. plastic additives have been linked to obesity and the development of metabolic syndrome. Adipocytes and macrophages are essential targets for EDC-mediated effects in adipose tissue, with plastic additives such as bisphenols or phthalates shown to impair both cell types, causing adipose tissue dysfunction and disruption of homeostasis. To elucidate the effects of EDCs on the interplay of adipocytes and macrophages and their function, a co-culture system was established, and interactions were investigated. For this purpose, cells were seeded in the following ratios: 1:1, 2:1 and 10:1 (SGBS adipocytes:THP-1 macrophages), to model healthy and inflamed adipose tissue. The global protein profiles were determined using LC-MS/MS-based proteomics, revealing significant changes in both cell types. When co-cultivating SGBS and THP-1 cells in a ratio of 2:1, THP-1 macrophages exhibited an altered glucose metabolism (pentose phosphate pathway, glycolysis, gluconeogenesis, galactose and fructose mannose metabolism). These pathways were slightly upregulated. While pathway analysis of SGBS adipocytes displayed a significant upregulation of unsaturated fatty acid biogenesis, fatty acid metabolism, glycolysis, citrate cycle, and PPAR signaling pathway when cells were co-cultivated 2:1 and 10:1. These findings demonstrate that the established co-culture system is a valuable tool for investigating the interaction between adipocytes and macrophages, and the molecular effects of EDCs that may be involved in the development of on obesity. 1245

1246

P-III-1082 The odd one out – Mass spectrometric detection of the ever-changing IgG4 landscape Linus Wollenweber1, Albert Bondt1, Albert J. R. Heck1 1Utrecht University, Biomolecular Mass Spectrometry and Proteomics, Utrecht, Netherlands Immunoglobulin G4 (IgG4) is the least abundant subclass of the four human IgG antibodies in plasma, but the second most used format for therapeutic antibodies. IgG4s exhibit unique structural properties that diminish a) binding to most Fcy receptors, b) binding to C1q and enable c) facile Fab-arm exchange (FAE). These properties result in low antibody dependent cellular cytotoxicity (ADCC) activation, reduced complement activation, and inability to cross-link antigens. The ability of IgG4 to bind inhibitory FcyIIB in combination with the reduced binding to activating Fcy receptors leads to a shift towards cellular inhibition. While this anti-inflammatory character has benefits in allergy and parasitic infections, studies in recent years have shed light on IgG4- mediated pathologies, such as IgG4-autoimmune diseases (IgG4-AID) and IgG4-related diseases (IgG4-RD). Studying immunoglobulin repertoires in plasma with clonal resolution can provide novel insights into basic questions in immunology. Ig repertoire profiling allows to follow on antibody plasticity, to get novel insights into antibody maturation, and to detect mature Ig sequences directly from donors. In recent years, our group has presented new mass spectrometric approaches to profile antibody repertoires. Therefore, we capture all IgG1 and/or IgA1 and employ specific heavy chain (HC) hinge-directed proteases to generate their Fab fragments. Subsequently, liquid chromatography coupled online to mass spectrometry (LC- MS) is applied to measure plasma Ig clones in a quantitative manner while preserving information on heavy and light chain pairing. Here we build further on this approach by introducing plasma IgG4 clonal profiling at the level of the intact Fabs. Our initial attempts on IgG4 profiling were focused on the intact Fab2 fragments, which would allow extensive studies on FAE. In middle-up MS experiments, most of the measured signal was however dispersing in unusually high noise. We deduced, that the inflated complexity originated from FAE facilitated stochastic Fab:Fab pairings. This assumption was supported by the observation that reduction of the same Fab2 samples resulted in far less complex data. By that, the complexity originating from stochastic Fab:Fab pairings was reduced to the mere light chain and Fd diversity. Notably, this is in line with earlier claims that almost all serum IgG4 undergo FAE. One way to reduce the spectral complexity is to measure IgG4 Fabs instead of Fab2s. For that we take advantage of the IgG4 property that a fraction of all HCs is held together exclusively by non-covalent interactions in the CH3 domain. This unique structural property gives rise to two distinct digestion products after hinge digestion; Fab2 for the molecules with covalent HC:HC interaction, and Fabs for those with only non-covalent interaction. These products can then be separated by SEC to enrich for the lower abundant Fabs prior to MS analysis. 1246

1247

P-III-1083 Eml1 depletion impairs the microtubule cytoskeleton and the binding affinity of its partners in cortical heterotopia Berfu Nur Yiğit1, Dilaray Tüfekci1, Donia Zaidi2,3,4, Büşra Aytül Akarlar1, Valeria Viola2,3,4, Fiona Francis2,3,4, Nurhan Özlü1,5 1Koç University, Department of Molecular Biology and Genetics, Istanbul, Turkey 2Institut du Fer à Moulin, Paris, France 3Institut National de Santé et de Recherche Médicale (INSERM, UMR-S 1270), Paris, France 4Sorbonne Université, Paris, France 5Koç University, Research Center for Translational Medicine (KUTTAM), Istanbul, Turkey Background: The layered structure of the cerebral cortex is formed through a complicated sequence of highly controlled stages. During this process, the perturbations of neuronal migration and cell division can result in a rare disorder called cortical heterotopia. Heterotopia patients can have recurrent epileptic seizures, developmental delays, and mild intellectual disabilities. Studying heterotopia has been challenging because human mutations linked to the disease often do not result in heterotopia formation in mouse models. EML1 is a microtubule-binding protein, and it stands out as the first heterotopia-associated gene where mutations lead to heterotopia formation in both humans and mice. In this study, we conducted a comparative proteomic analysis of the Eml1 cKO heterotopic mice cortices and neuronal progenitor primary cells during cerebral cortex development. Method: We performed labeling-based quantitative proteomic approaches, and microtubule pelleting assays to understand how Eml1 depletion disrupts protein networks in cortical tissue and neuronal progenitor primary cells during cerebral cortex development. label-free and dimethyl Results: Our extensive analysis of Eml1-depleted E15.5 cortices showed dysregulation of multiple synaptic proteins. However, despite achieving high coverage of quantified proteins in our cortex analysis, the proportion of dysregulated proteins was relatively low. EML1 perturbations are closely linked to heterotopia formation with phenotypes in progenitor cells, we hypothesized that deregulations in progenitor cells might be masked in the complexity of cell types at the level of the cortex. To address this, we specifically isolated Pax6+ neuronal progenitor primary cell cultures from control and Eml1 cKO mice at E14.5. This focused approach revealed that Eml1 depletion disrupts various networks of biological processes, including cytoplasmic microtubule organization. Enrichment analysis of downregulated proteins highlighted the significant influence of EML1 on microtubule cytoskeleton organization, microtubules, and structural constituents of the cytoskeleton in neuronal progenitor cells. By using microtubule pelleting assays, we demonstrated that Eml1 depletion dramatically impairs the microtubule cytoskeleton and reduces the binding affinity of its partners. Conclusion: Our comprehensive proteomic mapping of cortices and progenitor cells has identified protein signatures specific to Eml1 cKO heterotopic mouse models. These findings offer a valuable resource for investigating the underlying mechanisms involved in heterotopia formation and brain development. 1247

1248

Immunopeptidomics (P-III-1084 to P-III-1108) P-III-1084 Combining FAIMS-DIA and targeted immunopeptidomics to enhance neoepitope discovery Jonas P. Becker1,2, Sven Blobner3,4, David Weber5, Sebastian Uhrig6, Jonas D. Förster1,2, Annika Baude3,4, Johanna Wagner3,4, Isabel Poschke7,8,9, Michael Volkmar10, Michael Platten7,8,9,11, Stefan Fröhling3,4,6, Peter Horak3,4, Angelika B. Riemer1,2 1German Cancer Research Center, Division of Immunotherapy and Immunoprevention, Heidelberg, Germany 2German Center for Infection Research, Molecular Vaccine Design, Heidelberg, Germany 3German Cancer Research Center, Division of Translational Medical Oncology, Heidelberg, Germany 4National Center for Tumor Diseases, Division of Translational Medical Oncology, Heidelberg, Germany 5TRON Mainz, Expression & Cell Biology, Mainz, Germany 6National Center for Tumor Diseases, Molecular Precision Oncology Program/Computational Oncology, Heidelberg, Germany 7National Center for Tumor Diseases, Immune Monitoring Unit, Heidelberg, Germany 8German Cancer Consortium, Heidelberg, Germany 9German Cancer Research Center, Clinical Cooperation Unit Neuroimmunology and Brain Tumor Immunology, Heidelberg, Germany 10HI-TRON Mainz, TCR Discovery Platform, Mainz, Germany 11Heidelberg University Mannheim Center for Translational Neuroscience, Department of Neurology, Heidelberg, Germany Liposarcomas are malignant soft-tissue tumors of which only a subset responds to conventional cytostatic drugs and there is currently no approved targeted therapy available. One of the most common histologic subtypes, dedifferentiated liposarcoma (DDLS), is characterized by focal amplifications on chromosome 12. We hypothesized that these structural changes generate open reading frames and thus presumably transcribed and translated chimeric genes. The encoded neoepitopes are promising targets for epitope-centric individualized therapies if they are presented by HLA class I molecules. Using two different variant calling pipelines – Arriba and EasyFuse – we identified gene fusions from whole genome/exome and RNA sequencing data of eight DDLS patients. Next, we performed a comprehensive mass spectrometry (MS) analysis of the patient's tumor immunopeptidome by combining untargeted and targeted methods in order to identify fusion-derived neoepitopes. First, we performed data-independent acquisition with ion mobility separation by high-field asymmetric waveform ion mobility spectrometry (FAIMS-DIA) on an Orbitrap Exploris 480 connected to a Ultimate3000 RSLCnano system. MS raw data was analyzed using the Spectronaut 17 software combining library generation directly from DIA data with PROSIT in silico predicted libraries for putative neoepitopes. We identified up to 16500 unique peptides per patient, including several neoepitope candidates. Next, we utilized our recently established optiPRM workflow. Here, candidate neoepitopes identified by FAIMS-DIA as well as additional candidates from variant calling with excellent predicted binding properties were synthesized as stable isotope-labeled (SIL) variants and assay-relevant parameters such as normalized collision energy (NCE) and exact retention times were determined relative to iRT peptides. Targeted analysis by PRM not only confirmed several of the FAIMS-DIA candidates but also yielded additional detections in 5 out of 6 analyzed patients. Using our complimentary FAIMS-DIA and PRM immunopeptidomics workflows, we have identified fusion-derived neoepitopes in 5 out of 8 DDLS patient samples. Importantly, the identifications contained both neoepitopes which would have been missed by the less sensitive FAIMS-DIA approach or which would have been excluded from the targeted analysis due to their poor predicted binding properties, highlighting the advantage of the combined approach. Currently, we are testing all identified fusion-derived neoepitopes in epitope-specific expansion cultures (ESPEC) from the respective patients for reactive T cell populations. Additionally, we plan to perform TCR repertoire analysis to identify fusion neoepitope-reactive T cell receptors. Both detected neoepitopes and cognate T cell receptors identified will provide starting points for the clinical development of personalized treatment options in liposarcoma such as cancer vaccines or TCR-based therapies. 1248

1249

P-III-1085 An optimized mild acid elution (MAE) sample preparation workflow enabling MS-based immunopeptidome analysis from less than 1x106 cells Julian Beyrle1, Megan Ford2, Ute Distler3, David Gomez-Zepeda1,2, Stefan Tenzer1,3 1Helmholtz-Institute for Translational Oncology (HI-TRON), German Cancer Research Center (DKFZ), Mainz, Immunoproteomics Unit D191, Mainz, Germany 2Helmholtz Institute for Translational Oncology (HI-TRON), German Cancer Research Center (DKFZ), Immunopeptidomics Platform D190, Mainz, Germany 3Insitute for Immunology and Research Center for Immunotherapy (FZI), University Medical Center of the Johannes Gutenberg-University Mainz, Core Facility for Mass Spectrometry, Mainz, Germany The analysis of the Human Leukocyte Antigen (HLA)-associated peptide (HLAp) ligand repertoire by mass spectrometry (MS)-based immunopeptidomics has become indispensable for modern vaccine design and personalized immunotherapy.1,2 Peptide antigens from endogenous proteins are typically presented via HLA class I complexes to immune cells. These are particularly relevant for defining the cancer-associated or -specific HLA ligandome which can be exploited as immunotherapy targets. However, human sample material is scarce and precious, and the identification of HLAps is challenging due to their individual specificity, high diversity, low abundance, and high dynamic range.1,2,3 Therefore, there is a strong need to develop high-sensitivity immunopeptidomics workflows to profile the immunopeptidome and identify potential disease-relevant HLAps. The two main methods for the enrichment of HLA class I immunopeptides are Immunoprecipitation (IP) and Mild Acid Elution (MAE). IP typically uses bead-coupled antibodies to enrich HLA-peptide complexes from lysed cells. MAE utilizes a mild acid to strip HLA-class I bound (HLAIps) ligands from the surface of whole cells.4 While the latter is often regarded as less specific, we show that HLAIp enrichment using an optimized MAE HLA class I sample preparation workflow enables immunopeptidome analysis from less than 1x106 cells with high specificity. More specifically, we show that our protocol enriches a high proportion (>80%) of potential HLAIp peptides (8- 13mers) comprising 75% predicted HLA ligands with the respective binding motifs. Moreover, further optimization increased the identification of 8-13mers by 50% while maintaining the HLAIp enrichment specificity. Using two cell lines JY and Raji harbouring distinct HLA-I Alleles and thus also ligand motifs, we further evaluated our optimized MAE protocol by comparing MAE and IP bulk preparations (40x106 cells per sample). Irrespective of the cell line, IP and MAE enriched on average over 7,000 potential HLAIps with high specificity (>85% 8-13mers) comprising mainly predicted binders (>75%). Importantly, >80% of predicted binders were shared between both protocols. This suggests that no evident bias is introduced using our optimized MAE instead of IP for HLAIp enrichment. Utilizing an Evosep LC coupled to a timsTOF Ultra MS, we identified more than 2500 HLAIps per 1x106 cells of JY and Raji in Thunder-DDA-PASEF5 and library-based HLA-tailored DIA analyses further leveraged HLAIp identification 1.5-fold. In summary, we optimized MAE sample preparation and MS methods to enable immunopeptidomics profiling of samples using less than 1x106 cells. Although the last steps for method optimization and validation are still ongoing, this workflow may enable the analysis of sorted cell populations to study HLAIp presentation in the heterogenous environment of the tumor. 1249

1250

P-III-1086 Impact of dipeptidyl peptidase 9 on the MHC class I immunopeptidome Ruth Fiestas Cueto1,2, Adrianna Seredynska1,3, Samuel Zolg2,4, Ruth Geiss-Friedlander2,4, Oliver Schilling1,2,3 1Institute for Surgical Pathology, Faculty of Medicine, Medical Center, University of Freiburg, Freiburg, Germany 2ProtPath Research Training Group, Freiburg, Germany 3German Cancer Consortium (DKTK) and Cancer Research Center (DKFZ), Heidelberg, Germany 4Institute of Molecular Medicine and Cell Research, Faculty of Medicine, University of Freiburg, Freiburg, Germany Introduction: Dipeptidyl peptidase 9 (DPP9) is an intracellular amino-dipeptidase with a wide-ranging physiological role including DNA repair, mitochondria homeostasis, and immune system regulation. The dysregulation of DPP9 has been implicated in immune-associated defects and cancer, including acute myeloid leukemia, non-small-cell lung cancer, colorectal cancer, clear cell renal cell carcinoma, and breast cancer. Notably, DPP9's first identified natural substrate was an antigen peptide, linking it to the presentation of antigens on MHC class I molecules for cytotoxic T lymphocytes. This study aims to deepen our understanding of DPP9's impact on peptide presentation by MHC I molecules through a comprehensive immunopeptidome analysis using mass spectrometry. By conducting high-coverage immunopeptidomic analysis, we can examine in which way modulation of DPP9 activity affects the diversity of peptides bound to MHC class I molecules. Methods: We established and optimized a workflow for enriching MHC I bound antigens, utilizing automated affinity purification with crosslinked monoclonal antibodies on AssayMAP Bravo platform. This optimization included successful downscaling of antibody use and protein input while maintaining detection efficiency. The workflow, adaptable to a variety of sample types, was applied to different HEK cell line conditions, including DPP9 knock-out, overexpression, and wild-type. This comparison gives the possibility to investigate DPP9's role in MHC I-bound peptide presentation. Results and Outlook: The MHC I enrichment method successfully identified over 1000 peptides in each cell line sample, without additional enhancement of the immune response. This high coverage facilitates robust analysis of the immunopeptidome, revealing significant differences across the three cell line conditions. Preliminary results indicate substantial variations in the peptide repertoire linked to DPP9 modulation. Further studies integrating additional proteomic data will provide deeper insights into DPP9's role in antigen processing and presentation dynamics on MHC I molecules. Fig. 1 (optional) 1250

1251

P-III-1087 Integrating immunopeptidomics and RNA sequencing for neoantigen discovery in NSCLC Daniel Flender1,2, Geert Baggerman3, Kurt Boonen1, Evelien Smits4, Eline Berghmans1,2 1University of Antwerp, Centre for Proteomics, Antwerp, Belgium 2VITO Belgium, Unit Health, Mol, Belgium 3University of Antwerp, Department Informatics, Antwerp, Belgium 4University of Antwerp, Center for Oncological Research, Antwerp, Belgium Non-small cell lung cancer (NSCLC) accounts for 80-85% of lung cancer cases and is a leading cause of cancer- related deaths. Despite advancements in immunotherapy, current treatments fall short due to tumor cell heterogeneity. There is a pressing need for new targets, especially for CAR T cell therapy. Neoantigens, novel peptide sequences from tumor-specific mutations presented by MHC on tumor cells, offer a promising target due to their specificity, which is crucial to prevent autoimmune reactions. This study aims to identify highly specific neoantigens in NSCLC patients using patient-specific multiomic approaches, including immunopeptidomics and RNA sequencing. Immunopeptidomics, a mass spectrometry- based technique, identifies immunopeptides presented by MHC on the cell surface, providing direct insight into the immunoepitope landscape. Samples were collected from 12 NSCLC patients, including both tumor and benign tissues. RNA was extracted and sequenced to identify patient-specific mutations using DeepVariant, detecting single nucleotide variants (SNVs) and insertions/deletions (INDELs). These mutations were used to create patient-specific FASTA databases, essential for matching mass spectrometry data to specific mutations using PEAKs and MSFragger software. Immunopeptidomics analysis included benign tissues to ensure identified neoantigens were tumor-specific, preventing potential autoimmunity in therapeutic applications. Additionally, RNA-Seq data allowed us to predict MHC subtypes for each patient, facilitating further bioinformatic analysis. This included predicting the binding affinity of neoantigens to the patient"s MHC subtypes, validating them as true neoantigens. As a proof of concept, we initially searched each patient"s raw data against a database extended with immunopeptide sequences from previously described actionable neoantigens (dbPepNeo), helping to benchmark and refine our method. Proteomic analysis was integrated into the study, including data from both tumor and benign tissues to investigate the pathway of mutated proteins towards their presentation within the immunoepitope. Comparative analysis at proteomic and immunopeptidomics levels provided deeper insights into molecular mechanisms and the tumor microenvironment. In conclusion, we established and applied an advanced immunopeptidomics platform to NSCLC patients, analyzing both tumor and benign tissue samples. Creating patient-specific databases allows us to pinpoint neoantigens unique to tumor tissue, potentially paving the way for personalized immunotherapy. Comparative analysis of proteomic and immunopeptidomics data between tumor and benign tissues enhances our understanding of molecular dynamics within the tumor microenvironment. Our study underscores the potential of combining multiomic approaches to discover actionable neoantigens, significantly impacting the development of targeted therapies for NSCLC. 1251

1253

P-III-1088 Benchmarking of high-throughput immunopeptidomics workflows Megan Ford1, Annica Preikschat2, Julian Beryle1, Lucas Kleinort1, Yannic Chen1, Stefan Tenzer1,2, David Gomez- Zepeda1 1HI-TRON DKFZ, Mainz, Germany 2University Medical Centre Mainz, Mainz, Germany Introduction: Immunopeptidomics is the gold standard method to profile MHC-bound peptides. These MHC peptides are monitored by immune cells and may elicit a response against potential pathogenic or malignant anomalies. This has an important application for cancer treatment as these peptides can be tumour-specific, allowing hijacking the immune system against cancer cells. However, immunopeptidomics methods require high input material and large amounts of anti-MHC antibody, and they are low-throughput, resulting in a bottleneck for patient cohort studies. To solve this, we aim to optimise a high-throughput method capable of processing up to 96 well samples (one plate) in one week from samples within the range of 50 to 5 million cells. This would result in an 8-fold increase in throughput and a 100-fold decrease in sample and antibody requirements. Methods: To select the workflow providing the highest coverage and reproducibility, we will benchmark three plate-based methods: (1) An adaptation of our in house protocol (Tenzer lab HI-TRON Mainz), (2) a flow-through method (Bassani-Steinberg lab, LICR), and (3) a workflow using magnetic beads (Mann lab, MPIB). First, we focused on transferring our protocol (1) to a plate format. After immunoprecipitation, peptides were ultra- filtered and the eluates were desalted before LC-MS analysis in an Evosep One coupled to TimsTOF Ultra using Thunder-DDA-PASEF. Data was analysed using Peaks X Pro boosted by MS2Rescore, and MHC binding was predicted by NetMHCpan via MhcVizPipe. Results: To transfer our protocol (1) to a plate format, we scaled-down the input material from 100 to 50, 10, and 5 million JY cells. On average, 84.5% of peptides had the expected size of HLA class I ligands (8-13 amino acids), and 88% of these were predicted MHC binders. From 50 million cell input and injection of 5 million cell equivalents we identified 3050 HLA binding peptides. Furthermore, from 5 million cell input and injection of 0.25 million cell equivalents we were able to identify 545 HLA binding peptides. Conclusion: The initial in-plate method provided promising results, indicating that it can be further optimised. After optimization, we will benchmark it to other methods previously described to select the optimal workflow. Altogether, this optimisation will enable us to process larger numbers of samples therefore profiling the immunopeptidome of patient cohorts which could result in the detection of cancer-specific neo-epitopes from patient plasma or tumour samples. 1253

1254

P-III-1089 Thunder-DDA-PASEF enables the identification of mRNA splicing-derived immunopeptides in human tumor cell lines David Gomez-Zepeda1, Franziska Lang2, Yannic Chen1, Patrick Sorn2, Alina Henrich2, Anne Kölsch2, Barbara Schrörs2, David Weber2, Martin Löwer2, Ugur Sahin2,3,4, Stefan Tenzer1,3, Jonas Ibn-Salem2 1HI-TRON DKFZ, Immunopeptidomics Platform, Mainz, Germany 2TRON – Translational Oncology at the University Medical Center of Johannes Gutenberg University Mainz gGmbH, Mainz, Germany 3Institute of Immunology, University Medical Center of the Johannes-Gutenberg University, Mainz, Germany 4BioNTech SE, Mainz, Germany Introduction: Human leukocyte antigen (HLA, or MHC) peptide ligands, known as immunopeptides, play a crucial role in immune surveillance by triggering immune responses when presented epitopes are recognized by T-cells. This can be exploited to develop immunotherapies targeting immunogenic tumor-specific immunopeptides, termed neoepitopes originating from diverse sources. For instance, somatic mutations can result in single nucleotide variations (SNVs) or RNA processing changes. However, identifying neoepitope candidates by LC-MS is challenging due to the high diversity, large dynamic range, and individual-specificity of the immunopeptidome. To address these challenges, we profiled patient-matched melanoma and normal cell lines with our previously optimized LC-IMS-MS immunopeptidomics method, Thunder-DDA-PASEF. Additionally, we used our R-package splice2neo to detect splice junctions in tumor RNA-sequencing data. Methods: Cell pellets of patient-derived cell lines were lysed and HLA class I-peptide complexes were immunoaffinity-enriched using W6/32 anti-HLA-A, -B, -C antibody. Then, peptides were eluted (0.2% TFA), ultrafiltered (10 kDa cutoff), and desalted (HLB Oasis) before LC-MS-IMS analysis in a nanoElute coupled to timsTOF-Pro-2. Peptides were separated within a 47-minute gradient in an Aurora Ultimate RP column, and MS data was acquired using Thunder-DDA-PASEF. Peptides were identified employing PEAKS 11 Online with personalized protein databases including single-nucleotide variants (SNV) and splice junctions obtained by splice2neo. Results: LC-MS immunopeptidomics profiling resulted in 5,000 to 15,000 peptides per cell line. Tumor cells with lower HLA transcript expression also resulted in lower immunopeptidome coverage than the corresponding normal cells. Nevertheless, we still detected multiple tumor-specific immunopeptides derived from SNVs, constituting candidate neoepitopes. Furthermore, we identified several immunopeptides derived from RNA splicing as predicted from tumor RNA-seq. Conclusions: Thunder-DDA-PASEF provides high-coverage immunopeptidome profiling of cancer cells and enables the validation of neoepitope candidates from SNVs and immunopeptides derived from splice junctions detected in RNA-seq data using splice2neo. This workflow represents a promising strategy to identify tumor- specific targets for personalized cancer immunotherapies. 1254

1255

P-III-1090 Kinetics of HLA antigen presentation in transition from latent to Lytic stage of EBV Nicolas Grinberg1, Florian Huber1, Huisong Pak1, Michal Bassani-Sternberg1 1Ludwig Institute, Centre Hospitalier Universitaire Vaudois, Oncologie, Lausanne, Switzerland Epstein-Barr Virus (EBV) is a gamma herpes virus associated with esophageal cancer, nasopharyngeal cancer and lymphomas. In the latent stage, EBV lies dormant in epithelial and B-cells with eight viral anti-apoptotic and anti- inflammatory proteins lowly expressed . Thus, few latent proteins are presented as peptides by the human leukocyte antigen (HLA). However, in the lytic stage, the cells massively produce virion particles. To our knowledge, there hasn"t been an effort to map out HLA-bound peptides presented in host cells in the various stages of lytic mode. By inducing virus production using 12-O-tetradecanoylphorbol-13-acetate (TPA) and sodium butyrate (NaBu) in six EBV-positive B-cell lines, we analyzed their antigenic landscape at eight different time points after induction, using sequential HLA-I and HLA-II high-throughput immunopeptidomics aided by the Agilent Assaymap™ automated liquid handling platform. HLA bound peptides were measured on an Eclipse™ mass spectrometer in hybrid DDA/DIA using 1-hour non-linear gradient resulting in more than 30,000 HLA-I and 15,000 HLA-II peptides presented at various stages in virus activation per cell line. We later combined the HLA-I and HLA-II data with transcriptomics and quantitative shotgun proteomics to create the first map of the lytic EBV immunopeptidome from transcript to protein to antigen. Tens of HLA-I peptides derived from key EBV-proteins were detected throughout viral activation. Among others, peptides from lytic proteins such as BCRF1 and HEPA were identified. These assays will shed light on antigen biogenesis in induced cells and unveil novel EBV peptides for development of therapeutic cancer vaccines and T cell-based immunotherapies. 1255

1256

P-III-1091 Neoantigen discovery in renal cell carcinoma through comparative immunopeptidome profiling of RENCA xenograft models Kyle Hoffman1 1Bioinformatics Solutions Inc., Waterloo, Canada Renal cell carcinoma (RCC) is a highly aggressive tumor with a poor prognosis. Despite advancements in therapeutic approaches, the current treatments often exhibit limited efficacy, underscoring the urgent need for novel strategies. Considering the tumor's immune environment, we hypothesize that RCC could be an ideal candidate for immunotherapeutic interventions, particularly those aimed at enhancing the activity of cytotoxic (CD8+) T cells. In this study, we explore the immunopeptidomic landscape of RCC using the RENCA xenograft model to compare tumor tissue with benign tissue samples. We employ an immune affinity purification method coupled with LC-MS/MS acquisition for peptide identification, as well as gene identification, and label-free quantification. To identify novel canonical and non-canonical tumor-specific antigens resulting from genetic and epigenetic alterations during tumor progression, we employ a LC-MS/MS data analysis workflow with a specific focus on both discovery and validation steps. For discovery steps, the workflow integrates database search and peptide de novo sequencing. To ensure the accuracy of peptide identifications, a group-specific False Discovery Rate (FDR) was calculated to control for false positives from canonical and non-canonical databases searching, as well as those from de novo sequencing. loci integrating chromosome information with our quantification analysis, which compares the By immunopeptidomes of xenograft model tumor and benign tissue samples, we identify both canonical and non- canonical peptides unique or upregulated in RCC. Binding affinity assays and immunogenicity predictions suggest that most of the peptides identified from RCC tumor cells bind to class I major histocompatibility complex (MHC- I) and a subset may elicit an immune response. In addition, our findings reveal that certain peptides are uniquely presented on RCC tumors, some of which are derived from non-coding regions of the genome, potentially serving as novel therapeutic targets. Discovering these peptides enhances our understanding of MHC-I peptide presentation and the tumor's antigenic profile. Overall, the study underscores the potential of immunopeptidomics in revealing novel targets for immunotherapy in RCC. This approach not only opens new avenues for treatment but also contributes to the broader understanding of RCC's immune landscape, paving the way for advancements in cancer immunotherapy. Fig. 1 (optional) 1256

1257

P-III-1092 Screening covalent small molecules that modulate HLA-I and HLA-II expression and their effect on the HLA immunopeptidome repertoire Shelly Kalaora1, Joao Paulo1, Steven Gygi1 1Harvard Medical School, Cell Biology, Boston, MA, United States HLA presentation plays a critical role in immune surveillance. While our current understanding of the antigen presentation pathway and machinery has advanced in the past decade, there is currently a lack of small molecules capable of modulating antigen presentation. Identifying such molecules can expand our research toolbox and open the opportunity for drug development. To identify small molecules that can modulate HLA-I and HLA-II expression in tumor cells, we screened a library of 1,500 cysteine reactive electrophilic compounds using flow cytometry. We identified molecules that regulate HLA-I, HLA-II or both HLA-I and HLA-II expression, suggesting that different molecules modulate HLA expression by distinct mechanisms. We utilized a multi-proteomic approach to uncover the mechanism by which the compounds modulate HLA expression and understand their effect on the cells. Activity-based protein profiling (ABPP) was performed to identify which cysteines are targeted by each compound, and in-depth quantitative whole proteome and phospho-proteome analyses to identify the proteome-wide changes by the compound. Combining the information gathered by the different approaches, allowed us to identify the pathways involved in the compound's mechanisms of action. Interestingly, from the changes in cysteine availability (ABPP) results, we could infer not only the cysteine position that the compound binds to, but also changes in protein-protein interactions resulting from the compound binding. Further, we investigated how these compounds affect the HLA immunopeptidome repertoire of the cells using immunopeptidomics. For the immunopeptidome analysis, we harnessed the advanced capabilities of the Orbitrap Astral Mass Spectrometer platform towards characterizing the HLA peptide repertoire of the treated and untreated cells. The Orbitrap Atral"s high resolution and sensitivity allow accurate peptide identification, even of low-abundance peptides. This is particularly useful for HLA immunopeptidome analysis. We explored and optimized several data acquisition parameters including mass analyzers (Orbitrap vs. Astral) and FAIMS compensation voltages (CV). Our optimized approach enabled the identification of thousands of unique HLA-I peptides in a single experiment, with peptides conforming to expected lengths and HLA allele motifs. 1257

1258

P-III-1093 Immunopeptidomics for the identification of microbiota-derived antigens in colorectal cancer Louise Laurijssen1 1University of Antwerp, Antwerp, Belgium Colorectal cancer (CRC) is the third most diagnosed cancer and the second leading cause of death by cancer worldwide. In recent years, more attention has been directed toward the role of the gut microbiota in the initiation and progression of CRC. Several bacterial species have been associated with CRC carcinogenesis, like F. nucleatum, B. fragilis, and pathogenic E. coli strains, for which the mechanisms involved in their carcinogenic effects have partially been unraveled. However, the link between many intestinal bacteria and CRC initiation and progression remains poorly understood. Several recent studies have demonstrated T cell cross-reactivity against bacteria-derived antigens and tumor-derived antigens by molecular mimicry. Cases where this cross-reactivity resulted in enhanced anti-tumor response and improved response to immune checkpoint inhibitors have been observed. However, the notion that T cell cross-reactivity results in a suppressed anti-tumor response because of T cell exhaustion has also been proposed. Therefore, research is needed to fully understand the effects of gut microbial antigen presentation on immune reactivity during CRC development. In this project, we aim to address whether and how the presentation of gut microbial antigens may contribute to immune reactivity against colorectal tumors. To do so, we developed a mass spectrometry-based immunopeptidomics pipeline that allows us to successfully identify both MHC I and MHC II-associated peptides from murine samples, using a TimsTOF SCP (Bruker) hyphenated with an Evosep One chromatography system. This setup allowed us to identify multiple class I and class II microbial antigens in OligoMM12 mice (elevated germ-free, gnotobiotic mice colonized with a cocktail of 12 microbial strains). Following these proof-of-concept experiments, microbial antigen presentation will be examined in healthy murine colon tissue versus colon tissue from CRC mouse models. For these studies, we can make use of a unique mouse model of invasive and microbiota-driven CRC (Zeb2IEC-Tg). Observed differences in microbial antigen presentation may offer new insights into its effects on CRC development. These data will also aid in the correct interpretation of data obtained from human CRC samples and allow the extrapolation of the observations obtained from the mouse models. 1258

1259

P-III-1094 MSRescue improves De Novo candidate selection Hanqing Liao1,2, Nicola Ternette1,3,2 1University of Oxford, Oxford, United Kingdom 2University of Dundee, Dundee, United Kingdom 3Utrecht University, Utrecht, Netherlands Background: Successful immunotherapy against cancer necessitates the precise characterization of human leukocyte antigen class I (HLA-I) binding peptides to identify suitable targets. Liquid chromatography-tandem mass spectrometry (LC-MS) based immunopeptidomics offers high throughput and accuracy in identifying HLA-I binding peptides from canonical protein degradation products, but still lacks streamlined procedures for non- canonical antigen identification. Previously, we developed the MARS algorithm that proposes to incorporate HLA binding prediction and retention time differences to improve de-novo candidate selection, which improved identification of non-canonical neoantigens from aberrant or unknown RNA translations. Here, we propose an alternative strategy, which incorporates sequence clustering for improved identification of peptide sequences. Method: We trained our previously published neural network model MSRescue using sequence clusters from database search-identified peptides with low false discovery rates (FDR) to predict HLA binding ranks. We used the pancreatic immunopeptidome of Human HLA Atlas UDN02 with 3017 unique peptides, to predict the de novo candidates of the lung immunopeptidome from the same subject using Peaks X de novo, Peaks 11 DeepNovo, and NovoR candidates. Fractional ranks (F-Ranks) of true candidates were used to measure overall improvement, and full sequence recall (FSR) was used to measure performance of a simple identification scheme by picking top candidates per spectrum. Results: We constructed a set of ground truth spectra using Peaks X and Peaks 11 database searches against the SwissProt human proteome, resulting in 7097 unique peptides from 27561 spectra. By allowing up to 20 candidates per spectrum, DeepNovo reported 2765 peptides as the first candidate from 6007 spectra containing the correct sequence in the candidate list, Peaks X de novo recovered 3792 peptides / 7899 spectra, and NovoR identified 5502 peptides / 12514 spectra. Both our original MARS method and our new approach significantly improved F-Ranks of true candidates when applied to Peaks and NovoR candidates, but the FSR improvements were not significant. Conclusion: Applying MSRescue improves de novo candidate ranking, and enables post-processing of de novo candidate data without prior knowledge on HLA haplotype. In contrast to our previously published MARS method, this technique can be applied in samples where HLA typing is not possible or permitted, facilitating the exploration of non-canonical neoantigens. 1259

1260

P-III-1095 Dia-PASEF for in-depth immunopeptidomics analysis: challenges and new opportunities Kristina Marx1, Naomi Hoenisch Gravel2,3, Juliane Walz2,3,4,5, Torsten Mueller1, Pierre-Olivier Schmit6, Daniel Hornburg7 1Bruker Daltonics GmbH & Co. KG, Bremen, Germany 2University and University Hospital, Department of Peptide-based Immunotherapy, Tuebingen, Germany 3University, Cluster of Excellence iFIT (ECX2180) "Image-Guided and Functionally Instructed Tumor Therapies", Tuebingen, Germany 4Partner site, German cancer Consortium (DKTK) and German Cancer Research Center (DKFZ), Tuebingen, Germany 5University Hospital, Clinical Collaboration Unit Translational immunology, DKTK, Department of Internal Medicine, Tuebingen, Germany 6Bruker France SAS, Wissembourg, France 7Bruker Scientific LLC, San Jose, CA, United States Introduction: The identification of peptide antigens presented by the major histocompatibility complex (MHC) provides important insights for the understanding of cancer, infectious or autoimmune diseases and for the design and development of the corresponding immunotherapies. However, the analysis of low-abundance peptides and their diverse sequences is still a challenge. In recent years, the tremendous increase of MS instruments sensitivity, coming along with more effective, multi-dimensional ("4D") acquisition modes allowing to efficiently target singly and multiply- charged precursors, has really boosted the number of immunopeptides that can be detected in an immunopeptidomics experiment. Nowadays, most of the proteomics applications use data independent acquisition for improved identification. Here we demonstrate the use of dia-PASEF methods to further deepen the analysis of immunopeptides. Methods: For method optimization varying concentrations of acute myeloid leukemia cell line EoL (HLA-A*24:02; A*298:02; B*35:03; B*44:03; C*04:01; C*16:01) were used. HLA class I ligands were isolated via immunoaffinity chromatography using the pan HLA class I-specific W6/32 monoclonal antibody. Chromatographic separation was done with a nanoElute 2 (Bruker) with gradient time of 22 or 45 minutes on a 25 cm / 75 µm ID column, pre- fitted nanoZero connection and integrated emitter tip (Aurora, IonOptics) coupled to a CaptiveSpray Ultra ionization source (Bruker). Samples were analyzed using 300Hz PASEF and dia-PASEF methods with varying parameters for method optimization on a timsTOF Ultra (Bruker). Raw data were processed in Spectronaut 19. Preliminary Data: Data Dependent Analysis (DDA) is still widely used for immunopeptidomics studies for their compatibilities with DeNovo sequencing and their ability to be used with very large antigen sequence databases. Using a high sensitivity PASEF method for low sample amounts on the timsTOF SCP with additional inclusion of singly charged precursors (>700 m/z) resulted in improved identification of class I immunopeptides. We have further optimized the method on the timsTOF Ultra to improve the sensitivity and speed of the analysis. This opens the door for the use of shorter gradients (45 min and 22 min) without losing identifications and sensitivity compared to previous set-ups. Data independent analysis acquisition methods, however, are becoming the standard approach for proteomics analysis and can be used for immunopeptidomics when the database size allows for it. Here, we evaluate the use of dia-PASEF methods for immunopeptidomics to increase immunopeptide identifications especially on low sample inputs. In addition, we discuss the challenges of using unspecific searches in dia workflows. 1260

1262

P-III-1096 Characterization of bioactive peptides from Tuna waste fermentation by Lactobacillus plantarum and Bacillus subtilis using mass spectrometric analysis Mohadeseh Montazeri Shatouri1, Anwar Sunna1, Igor Pirozzi1, Paul A Haynes1 1Macquarie University, Natural Sciences, Sydney, Australia Lactobacillus plantarum and Bacillus subtilis, known for their versatility in fermentation studies, were used to ferment tuna processing waste, leveraging their ability to hydrolyse protein sequences and generate bioactive peptides. Following optimization of the fermentation process, peptides were filtered using a 10 kDa size cutoff and analysed by nanolC-MS/MS using a Thermo Exploris 480 mass spectrometer. Analysis of the raw data via MSFragger 21.1 employing a non-specific search identified reproducible peptides observed in three biological replicate experiments. This resulted in the identification of 74 peptides from L. plantarum fermentation and 31 peptides from B. subtilis fermentation, none of which were found in the control (unfermented) experiments. In silico analysis revealed that out of the 74 peptides produced by L. plantarum, only 4 exhibited a bioactive probability exceeding 0.7, while 14 out of 31 peptides produced by B. subtilis surpassed this threshold. Peptides identified in the control group did not exhibit a bioactive probability greater than 0.7. Interestingly, most of the highly bioactive peptides identified in the fermentation experiments were found to originate from collagen. Keywords: Tuna waste, Protein, Bioactive peptides, Mass spectrometry, Fermentation 1262

1263

P-III-1097 Sensitive and high-throughput immunopeptidomics using the latest generation Orbitrap astral and ascend Annalisa Nicastri1, Zuza Kozik1, Sonja Radau2, Eleni Adamopoulou3, Joanna Kirkpatrick4, Tabiwang Arrey2, Nicolaie Eugen Damoc2, Alan Melcher1, Jyoti Choudhary1 1Institute Cancer Research, London, United Kingdom 2Thermo Fisher Scientific, Bremen, Germany 3University of Oxford, Oxford, United Kingdom 4Thermo Fisher Scientific, Hemel Hempstead, United Kingdom Quantitative mass spectrometry is a central technology for characterizing human leukocyte antigen (HLA) landscape and represents a key data source for the development of therapeutic vaccines. However, a combination of faster throughput, higher sensitivity and accuracy is required for deeper coverage of peptides due to the recovery of low abundance HLA-peptides and complexity of samples. Here, we evaluate and extend DDA, DIA and TMT labelling methods for the detection of HLA-I peptides using next generation of Thermo ScientificTM OrbitrapTM mass spectrometers (Orbitrap AscendTM and the Orbitrap AstralTM) instruments. We used immunopeptidomics-like standard (RPE-1 protein Lysate digested with Elastase) for the optimisation of the DDA and DIA methods evaluating different LC methods, precursor selection, collision energies, intensity threshold, target intensity and retention time. We validated the optimized methods using synthetic peptide libraries and benchmarked the sensitivity using serial dilutions. Finally, we applied our optimized methods for the detection of HLA-I peptides from HCT116 cells line and Human Plasma samples. Our results suggest that the latest generation Orbitrap mass spectrometers increase the sensitivity, throughput time, and quality of the analysis for a deeper coverage of HLA-I peptides at low sample input. 1263

1264

P-III-1098 Deciphering the MHC-I/II Immunopeptidome landscape of murine tumor with normal tissues for neoantigen discovery Chao Peng1,2, Ping Wu1, Haofei Miao1, Guangqin Cheng1, Ling Li1, Wenting Li2, Lei Xin2, Baozhen Shan1,2 1Baizhen Biotechnologies, Inc., R&D, Shanghai, China 2Bioinformatics Solutions Inc., R&D, Waterloo, Canada The Major Histocompatibility Complex (MHC) displays numerous peptides on the surface of cells, particularly neoplastic cells. These MHC-associated peptides are crucial not only for functional research on adaptive immunity but also for defining neoantigens that can be harnessed for cancer immunotherapy through vaccines or cell therapies. Mass Spectrometry (MS) has emerged as a powerful tool for identifying MHC-presented peptides (MIPs) from tumor cells, offering a direct method to detect peptides exactly presented by MHC molecules, a capability that is distinct from Next-Generation Sequencing (NGS). Despite the advantages of MS, there have been relatively few reports on the direct identification of neoantigens just by MS. We have developed a highly sensitive and reproducible immune-precipitation procedure that enables the direct identification of neoantigens through MS denovo sequencing algorithm—DeepNovo. Considering the widespread prevalence of cancer types, we designed an experiment involving seven tumor cell lines implanted into C57/Bl6 mice, including MC-38, Hepa 1-6, LLC1, B16-F10, Panc02, MFC, and Renca. The subcutaneous tumor tissues were subjected to immunoprecipitation to capture MHCI/II-associated peptides to discover neoantigens using an in-house developed integrated platform named DeepImmu.2000-10000 MHC I binders have been identified from these tumor/benign tissues. And 60 neoantigens were selected from around 10000 of the CRC tumor tissues according to the immunogenicity prediction by DeepSelf. 12 peptide pools were generated by 60 synthetic peptides to assess their immunogenicity by Elispot assay. 4 out of 12 peptide pools were confirmed to be positive for immunogenicity. In summary, the discovery of neoantigens can be achieved solely by MS data, relying on accurate peptide denovo-sequencing. This method has the potential to significantly accelerate the neoantigen discovery process in less than one week, indicating promising clinical applications. Using DeepImmu Platform, we also constructed a landscape of a pan-tissue immunopeptidomics library from both tumor and benign tissues, which could be a good resource for future neoantigen discovery. 1264

1265

P-III-1099 High sensitivity combined proteomic and immunopeptidomic analysis of synovial tissue of rheumatoid arthritis patients Annica Preikschat1,2, David Gomez-Zepeda3,4, Johan Verhagen5, Gerhard Krönke5, Stefan Tenzer1,2,3,4 1University Medical Center of the Johannes Gutenberg-University Mainz, Immunology, Mainz, Germany 2University Medical Center of the Johannes Gutenberg-University Mainz, Research Center for Immunotherapy (FZI), Mainz, Germany 3Helmholtz Institute for Translational Oncology (HI-TRON), Mainz, Germany 4German Cancer Research Center (DKFZ), Heidelberg, Germany 5Charité – University Medical Center, Rheumatology and clinical immunology, Berlin, Germany Introduction: Rheumatoid arthritis is an autoimmune disease with a 0.5-1% prevalence in developed countries. Currently, there is no cure and only treatment available, resulting in reduced life expectancy and quality of life. The disease mainly affects the joints, which makes it important to analyse those inflamed tissues for possible auto-antigens. Here, we focus on the characterization of the immunopeptidome, particularly on self-peptides presented on MHC class I or II that elicit an auto-immune response against synovial tissues. However, synovial tissue is very elastic and thus difficult to dissociate, posing a challenge for efficiently isolating MHC-peptide complexes. By combining efficient tissue digestion, high recovery, and multi-level sample preparation protocols with optimized methods for acquisition and data analysis of liquid chromatography mass spectrometry, we present a workflow enabling in-depth characterization of synovial tissue on both the immunopeptidome and proteome levels. Methods: 70-160 mg of synovial tissue was mechanically disintegrated and lysed by physical forces and a mild detergent. 1/10 of the lysate was digested by FASP for whole proteomics analysis. The remainder of the sample was used for sequential immunoprecipitation, including a blank, an HLA-DR, and a pan-MHC class I immunoaffinity enrichment. Peptides were eluted under acidic conditions, cleaned with a 10 kD MWCO filter, and desalted on an Oasis HLB plate. The MWCO filter retentate (enriched MHC proteins) and an aliquot of the unbound fraction depleted of MHC proteins were subjected to FASP digestion. This resulted in four different sample types derived from each tissue sample. Immunopeptides were analyzed using a previously optimized ion mobility enhanced data-dependent acquisition (Thunder-DDA-PASEF) with adjusted fragmentation isolation polygons respective to each MHC class. Peptide search was performed in PEAKS. Proteome samples were measured on an Orbitrap Exploris 480 in data-independent mode for high reproducibility and processed in DIANN. Preliminary results: Our optimized sample preparation and LC-MS workflow allows the identification of more than 1,000 MHC-bound peptides from 100 mg synovial tissue, increasing sensitivity by approximately 100-fold compared to previous studies requiring 10 g of synovial tissue.to reach a similar coverage. Using the established multi-level workflow, we can additionally infer knowledge from the tissue to the identified immunopeptides and vice versa. We can also assess the quality of the immunoprecipitation due to the filter retained fraction, and the relative composition of HLA proteins which can be used to evaluate differences on their expression. This workflow enables the identification of auto-immunopeptides that can be either targeted for therapy or provide information on the development and onset of rheumatoid arthritis. 1265

1266

P-III-1100 Fine tuning of ddaPASEF methods improves the immunopeptidomics coverage and throughput, which is further boosted by ML-assisted rescoring Jeewan Babu Rijal1, Aurélie Hirschler1, Arthur Declerq2,3, Pauline Perdu-Alloy1, Charline Keller1, Lennart Martens2,3, Christine Carapito1 1BioOrganic Mass Spectrometry Laboratory (LSMBO), IPHC UMR 7178, University of Strasbourg, CNRS, ProFI, Strasbourg, France 2Ghent University, Center for Medical Biotechnology (CMB), Ghent, Belgium 3Ghent University, Center for Medical Biotechnology, Department of Biomolecular Medicine, Ghent, Belgium Introduction: Identifying ligands of the human leukocyte antigen (HLA), known as immunopeptides, is vital for developing vaccines and immunotherapies. LC-MS-based immunopeptidomics can directly identify and quantify HLA peptides. However, this approach faces significant challenges, requiring optimization at each step of the workflow, from sample preparation through MS analysis to data treatment. Throughout, the non-tryptic nature of HLA adds specific complications. We optimized the MS acquisition method on the timsTOF Pro instrument but also implemented it on the timsTOF Ultra to achieve highest possible detection sensitivity and throughput. Additionally, we integrated MS²Rescore into the data analysis, which uses AI and machine learning-based peptide fragmentation (MS²PIP) and retention time (DeepLC) predictions to rescore search engine results. Methods: MS method optimizations were performed on a nanoElute timsTOF Pro and Ultra system (Bruker) with immunopeptides from 5.10e8 HL60 cells. Key acquisition parameters evaluated included ion mobility range (0.6 to 1.75 Vs/cm²), accumulation time (100 to 200 ms), collision energy (CE) ramp (10 to 55 eV and 20 to 59 eV), number of PASEF scans (6 or 10), and polygon filters (standard or HLA-tailored). We tested 5 cm and 25 cm Aurora3 columns (IonOptiks) with gradients ranging from 10 to 100 minutes. Peptides were identified using Proline or Sage software and rescored with MS²Rescore. Results: We benchmarked eight different MS methods using ddaPASEF acquisition on a timsTOF Pro to finely tune a method dedicated to non-tryptic immunopeptides. Crucial MS acquisition parameters were evaluated, including ion mobility, collision energy ranges, accumulation time, number of PASEF scans and tailored polygon filter. The optimized parameters increased identified immunopeptides from 4548 to 5793 compared to the classical method. Implementing Sage and MS²Rescore further boosted the number of identified peptides to 10257, a 77% increase. An HLA-tailored polygon filter improved the total number of identified peptides by an additional 10%, allowing the identification of singly charged peptides (14%), which had previously been discarded. When applied to the timsTOF Ultra, similar performance could be achieved with only half the injection amount and only one-fifth of the gradient time. A 100-minute gradient using 25 cm Aurora3 columns identified as many as 19,376 peptides, showing the highest level of performance. More than 5000 peptides were still detected using a shorter 5 cm column in 10 minute gradient, demonstrating great potential for increased throughput. Conclusion: Our results demonstrate a significant improvement in the immunopeptidomics workflow, achieving unprecedented performance with the latest timsTOF Ultra. Various gradient tests showed potential for increased throughput while maintaining adequate coverage. Overall, the enhanced coverage and depth of analysis are promising for detecting neo-antigens in cancer tissues. 1266

1267

P-III-1101 Enhancing immunopeptide profiling with Orbitrap astral mass spectrometer for unbiased discovery of neoantigens Fernanda Salvato1, Lilian Heil1, Amirmansoor Hakimi1, Tonya Pekar Hart1 1Thermo Fisher Scientific, San Jose, CA, United States Peptides presented on the cell surface by the major histocompatibility complex (MHC) play a vital role in immune response, making identification of neoantigens critical for immunotherapy. Mass spectrometry (MS) serves as a valuable tool for immunopeptidomics. These measurements can be challenging due to low starting material and low sample preparation recovery. The Orbitrap Astral™ mass spectrometer enables researchers to delve deeper, with the new levels of sensitivity and selectivity, providing comprehensive insights into the immunopeptidome and neoantigen discoveries. Class I MHC peptides were obtained by immunocapture with W6/32-conjugated resin on 100 million HCT-116 cells. For proof-of-concept studies, performance on low-loads was performed by diluting the starting material down to lower amounts. The MHC-bound peptides were eluted and subjected Liquid Chromatography (LC) coupled to MS analysis on an the Orbitrap Astral mass spectrometer interfaced with a Thermo Scientific™ Vanquish™ Neo LC and FAIMS Pro™ Interface. Preliminary findings suggest that the Orbitrap Astral MS enables deep immunopeptidome coverage from low starting inputs without compromising data quality. Over 3,500 peptides from 1e5 cell equivalents, >9,000 peptides from 1e6 cell equivalents were identified using our platform. Our initial data also suggests that the peptide identifications are of high quality. Results are validated on two different search engines and show similarly high coverage. We successfully achieved high reproducibility between runs with over 75% of peptides detected across all three technical injections. Overall, our preliminary results suggest that the Orbitrap Astral produces deep coverage with exceptional data quality, enabling vital biological discovery in the field of immunopeptidomics. 1267

1268

P-III-1102 The SysteMHC Atlas v2.5 Xiaoxiang Huang1, Wenguang Shao1 1Shanghai Jiao Tong University, Shanghai, China Comprehensive characterization of the entirety of major histocompatibility complex (MHC)-bound peptides, collectively referred to as the immunopeptidome, is essential to advance our understanding of immunology and provide the foundation to design vaccines effectively and safely. The SysteMHC Atlas v1.0 was the first public repository dedicated to mass spectrometry-based immunopeptidomics. Here, we introduce a new version of the SysteMHC Atlas (https://systemhc.sjtu.edu.cn), with an extensive collection of over 7000 MS measurements and the incorporation of a few novel computational tools. With all these features, we believe that the atlas can serve as a more useful community resource to provide key insights to the immunology and proteomics community, and will accelerate the development of vaccines and immunotherapies. First, we extended and optimized a new computational pipeline that allows the identification of MHC-bound peptides carrying on unexpected post-translational modifications (PTMs). In addition, the SysteMHC Atlas v2.5 introduces several new features, including the inclusion of non-UniProt peptides, and the incorporation of several novel computational tools for FDR estimation, binding affinity prediction and motif deconvolution. Moreover, we enhanced the user interface, upgraded website framework, and provided external links to other resources related. Finally, we built and provided various spectral libraries as community resources for data mining and future immunopeptidomic and proteomic analysis. Preliminary Data The current release of the atlas contains >180 million tandem mass spectra obtained from >7000 mass spectrometric raw files. The new computational pipeline includes the major updates: (i) identification of MHC- bound peptides carrying on unexpected PTMs; (ii) accurate FDR estimation by three different methods and (iii) incorporation of novel computational tools for binding affinity prediction and motif deconvolution. By re- analyzing the whole atlas using the computational pipeline, we identified a total of >2 million unique peptides at a peptide level FDR of 1%, which represents a >10-fold increase in comparison with the atlas v1.0. Among them, we identified >450K MHC class I binders and >350K MHC class II binders. These binders can be attributed to 153 MHC class I allotypes and 149 MHC class II allotypes. Regarding PTMs, our analysis identified >450K modified peptides, a >20-fold increase to that in v1.0. In total, the atlas contains 61 distinct modification types (i.e. 422 distinct amino acid - PTM combinations) found on MHC-bound peptides. 1268

1269

P-III-1103 Venturing into the unknown: non-canonical antigen discovery Nicola Ternette1 1University of Oxford, Centre for Immuno-Oncology, Oxford, United Kingdom - uProteins session contribution - The identification of non-canonical antigens has emerged as an essential aspect of immunopeptidomics, driving advancements in our understanding of immune responses and disease mechanisms which were missed with conventional approaches and hence left unexplored. As the field evolves, stringent quality control measures are essential to ensure high confidence in peptide identifications, particularly for peptides encoded by non-canonical genome regions and uncharacterised RNA variants. Emerging strategies including machine learning models, Omics data integration, and improved de novo sequencing algorithms enhance our ability of non-canonical antigen identification using mass spectrometry. Streamlining these approaches will enable us to systematically explore non-canonical, HLA-presented peptide repertoires, and facilitate the development of therapeutics targeting this novel space of antigens. 1269

1270

P-III-1104 Personalized de novo peptide sequencing to identify non-canonical HLA peptides from individual-patient immunopeptidome Ngoc Hieu Tran1, Rui Qiao1, Zeping Mao1, Shengying Pan1, Qing Zhang1, Wenting Li1, Lei Xin1, Baozhen Shan1 1Bioinformatics Solutions Inc., Waterloo, Canada Background: De novo peptide sequencing provides a direct and unbiased solution to identify non-canonical HLA peptides from mass spectrometry (MS) based immunopeptidomics. However, de novo peptides may often contain sequencing errors while neoantigen applications ultimately require completely accurate peptide sequences. Recent studies have developed new de novo sequencing methods and dedicated models that were optimized specifically for HLA data. Here we go one step further and propose a personalized deep learning model to learn from the immunopeptidome of each individual patient and then to predict non-canonical HLA peptides of that patient. Methods: Given an MS-based immunopeptidomics of an individual patient, PEAKS DB search was performed to identify canonical peptide-spectrum matches (PSMs) at false discovery rate (FDR) of 1%. A universal GraphNovo model was applied to perform de novo sequencing and predict 100 candidate peptides per spectrum. The model was further fine-tuned on the PSMs identified by PEAKS DB to obtain a personalized de novo sequencing model. The personalized GraphNovo model was finally combined with other retention time and binding prediction models to select the best peptide out of 100 candidates per spectrum. Thus, the final peptide was strongly supported by multiple dimensions of evidence, including its spectrum, retention time, binding motif, and the personal HLA profile of the patient. Results: We evaluated our personalized de novo sequencing approach on the datasets of ten cancer patients from three previously published studies. We also compared our personalized models to the universal GraphNovo model and to PEAKS De novo algorithm. The evaluation results show that our personalized models improved the peptide accuracy by 5-10% and increased the number of identified HLA peptides by 20-30%. Furthermore, the identified peptides also exhibited high binding levels and accurate binding motifs, especially at the second amino acid position, which often cause de novo sequencing errors due to the missing fragmentation problem of HLA peptides. Conclusions: Personalized de novo peptide sequencing models represent the most advantageous and ideal solution to identify non-canonical HLA peptides of each individual patient. 1270

1271

P-III-1105 Unipept next: next-generation peptide analysis tool for HLA-peptide identification Pieter Verschaffelt1,2,3, Tibo Vande Moortele2, Bram Devlaminck2, Peter Dawyndt2, Bart Mesuere2 1Ghent University, Center for Medical Biotechnology (CMB), Ghent, Belgium 2Ghent University, Department of Applied Mathematics, Statistics and Computer Science, Ghent, Belgium 3Ghent University, Center for Medical Biotechnology, Department of Biomolecular Medicine, Ghent, Belgium (https://doi.org/10.1021/acs.jproteome.8b00716), a pioneering Unipept in metaproteomics, has significantly advanced the analysis of complex ecosystems by facilitating both taxonomic and functional insights from environmental samples. Initially, Unipept's capabilities focused on tryptic peptides, utilizing the predictability and consistency of trypsin digestion to efficiently construct a protein reference database. However, the evolving landscape of proteomics and emerging fields like immunopeptidomics necessitate a more versatile approach that extends beyond the analysis of tryptic peptides. tool In metaproteomics, samples are typically prepared using trypsin which cleaves proteins at specific cleavage sites (https://doi.org/10.1038/s41467-021-27542-8). Occasionally, trypsin skips a cleavage site, resulting in fragments composed of multiple tryptic peptides, referred to as peptides with missed cleavages. Semi-tryptic peptides, produced when one terminus is generated by trypsin cleavage and the other is not, were previously undetectable by Unipept. Accurately identifying semi-tryptic peptides is crucial for understanding the complex proteolytic activities of the gut microbiome (https://doi.org/10.1186/s40168-020-00967-x). Immunopeptidomics (https://doi.org/10.1016/j.smim.2023.101727), an upcoming research discipline, focuses on characterizing HLA-peptides presented by MHC molecules, which are vital to study our immune responses. In the context of immunopeptidomics, the peptides presented by MHC molecules are typically between 8 and 25 amino acids long (depending on the class of the MHC molecule) and are derived from both self and non-self (e.g. viral or bacterial) proteins. By extracting the peptides presented by MHC molecules and identifying the organism they originate from, researchers can gain a deep understanding of how the immune system recognizes and responds to pathogens, tumors and other abnormal cells, and eventually help to identify peptide biomarkers for diseases. The past limitation of Unipept to match only tryptic peptides constrained its utility in characterizing HLA- peptides, studied by immunopeptidomics researchers. Over the past year, we have integrated a range of sophisticated index structures from the field of metagenomics, such as Enhanced Suffix Arrays (ESA), FM-indices, and the R-index, to enhance Unipept's capability in identifying substrings in large text bodies efficiently. Unipept Next is the next-generation of Unipept, capable of analyzing peptides regardless of how their originating proteins were cleaved. Utilizing an Enhanced Suffix Array, Unipept Next can identify the proteins from which a peptide originates in milliseconds, enabling rapid taxonomic and functional analysis of peptide samples. 1271

1272

P-III-1106 Plasma immunopeptidomic landscape in patients with alpha-1 antitrypsin deficiency Maria Wahle1, Florian Rosenberger1, Maximilian Zwiebel1, Sophia Steigerwald1, Katrine Holtz Thorhauge2, Aleksander Krag2,3,4, Matthias Mann1 1MPI of Biochemistry, Munich, Germany 2Center for Liver Research, Department of Gastroenterology and Hepatology, Odense, Denmark 3University of Southern Denmark, Department of Clinical Research, Odense, Denmark 4Danish Institute of Advanced Study (DIAS), Odense, Denmark Alpha-1 antitrypsin deficiency (AATD) is an autosomal monogenic defect that causes misfolding and a secretory defect of alpha-1 antitrypsin (AAT) in the producing hepatocyte cell. This can lead to variable symptoms both in the liver as well as in the periphery, predominantly the lung. Consequences can be severe liver fibrosis or even cirrhosis as well as pulmonary emphysema. In a previous study we discovered that a healthy human has a stable soluble human leukocyte class I (HLA-) immunopeptidomic signature, representative of various organ systems (Wahle et al. PMID 38043703). AATD is associated with the remodeling of various intracellular processes potentially reflected by different proteome compositions and as a consequence by the presented immunopeptides. We here apply our recently introduced IMBAS-MS immunopeptidomics workflow for the efficient enrichment of soluble immunopeptides from plasma of 120 homozygous AATD patients and their heterozygous genetic relatives. We leverage the speed and flexibility of the Orbitrap Astral mass spectrometer in analyzing immunopeptidomics samples separated over a 31 min nanoflow gradient (Whisper 40) using the Evosep One liquid chromatography system. We also acquired the neat plasma proteome of all patients on a Orbitrap Exploris 480. Raw files were analyzed with DIANN 1.8.1 or AlphaDIA using predicted spectral libraries and peptide quality was assessed using MHCVizPipe. We identify on average more than 4000 HLA-I immunopeptides per patient. We now aim to assess whether the soluble immunopeptidome can be indicative of cellular stress and hepatocyte damage as a consequence of AAT accumulation in those cells. Findings of biomarker candidate would enhance the precise monitoring of the disease status in patients with alpha-1 antitrypsin deficiency with clinical impact for this large disease group. 1272

1273

P-III-1107 Maximizing immunopeptide identification and reproducibility from minute pediatric solid tumor biopsy samples Matthew Willetts1, Diego Assis1, Peiyao Li2, Richa Kapoor2, John Maris2,3, Hossein Fazelinia2, Stephan Kühne4, Jasmin Meltretter4 1Bruker Scientific LLC, Protoemics, Billerica, MA, United States 2Children's Hospital of Philadelphia, Philadelphia, PA, United States 3University of Philadelphia, Philadelphia, PA, United States 4Bruker Daltonics GmbH & Co. KG, Bremen, Germany Introduction: Immunopeptidomics, a rapidly growing field of proteomics, plays a vital role in identifying and quantifying peptides presented by the major histocompatibility complex (MHC) to the immune system. Peptides presented by MHC molecules are typically present at very low levels especially in pediatric patient samples where MHC expression is also often low. Maximizing peptide yield, sensitivity, and intra- and inter-sample reproducibility with mass spectrometry is essential to discovering peptide-MHC (pMHC) targets for modern cancer immunotherapies. Here we compare sample preparation methodologies incorporating Sepharose bead columns and streptavidin magnetic beads combined with the ultra-high sensitivity mass spectrometry for the analysis of immunopeptides from pediatric solid malignancies. Methods: HLA class-I peptide samples were prepared using two immunoprecipitation (IP) methods: CNBr- activated Sepharose 4B column-based capture and streptavidin-biotin based capture. Briefly, frozen tissues and frozen cell pellets were pulverized by liquid nitrogen cooled cryomilling and then resuspended in lysis buffer to perform IP with using pan-HLA class I monoclonal antibody W6/32. Peptides were eluted from HLA molecules and then purified with a 5 kDa MWCO filter and C18 resin. Resulting peptides were separated by HPLC (nanoElute 2, Bruker Daltonics) on a 250 mm x 75 µm, 1.7 µm column.60 min gradients were analyzed in parallel on a timsTOF Pro 2 and timsTOF Ultra (Bruker Daltonics) in PASEF (Parallel Accumulation and Serial Fragmentation) mode. Data were processed using PEAKS software. Results: HLA class-I peptides from 1e8 cells of cancer cell lines yielded over 20,000 unique canonical peptides using CNBr-activated Sepharose 4B method, demonstrating the robustness of our large-scale IP approach. Osteosarcoma patient-derived xenograft (PDX) samples revealed an average of 4986 unique peptides (a range from 397 to 14,823 peptides), with variations linked to tumor input mass. In comparing small-scale IPs, the CNBr- activated Sepharose 4B method outperformed the streptavidin magnetic bead protocol, yielding 7710 and 4692 peptides, respectively, from 2e7 SKNAS neuroblastoma cells. While CNBr-activated Sepharose 4B offered higher overall peptide yields, it exhibited lower reproducibility compared to the streptavidin magnetic bead protocol, with approximately 40% and 50% overlap of peptide IDs between three technical replicates for the two sample preparation protocols run on the timsTOF Ultra. Moreover, the Bruker timsTOF Ultra identified > 40% more peptides compared to the timsTOF Pro 2. Importantly, an average of 13,098 unique HLA-I peptides, ranging from 3,879 to 29,261, were identified from pediatric glioma tissue samples of less than 200 mg using the timsTOF Ultra. These findings provide critical insights into optimizing reproducible immunopeptide identification in pediatric solid cancers, guiding future research and immunotherapy development. 1273

1274

P-III-1108 Overcoming low tumor mutational burden in cancer antigen discovery Wei Wu1 1A*STAR Singapore Immunology Network, Immuno-Oncology, Singapore, Singapore In recent years, personalized therapy with neoantigen-based vaccination has become a feasible reality at least in high mutational burden cancers. Propelled by ever-improving analytical capabilities and sensitive mutation calling, it is now possible to vaccinate a patient within a month of tissue biopsy confirming diagnosis. Yet, the prevalence of mutations remains a key determinant of which patients can benefit from vaccination. For instance, in colorectal cancer, >85% of patients are microsatellite stable with much fewer tumor-specific mutations. Most pediatric cancers are not high in mutational burden, considering the short time these young patients have to accumulate DNA aberrations. Here, we investigate non-genome templated RNA edits as potential sources of universal neoantigens. Adenosine deaminases acting on RNA (ADARs) are enzymes that perform A-to-I editing at the RNA level and are widely over-expressed in most cancers. We show here that RNA edited antigens largely do not alter antigen loading, but can either increase or decrease the immunogenicity of presented antigens. This expands the neoantigen space even in low tumor mutational burden cancers. Coupled to the emerging capabilities to predict RNA editing sites, we envision ADAR generated antigens to be possibly universal and important in overcoming the bottleneck of tumor mutational burden. 1274

1275

Late breaking poster (LP-01 to LP-10) LP-01 A hybrid DDA/DIA-PASEF based assay library for a deep proteotyping of triple-negative breast cancer Petr Lapcik1, Klara Synkova1, Lucia Janacova1, Pavla Bouchalova1, David Potesil2, Rudolf Nenutil3, Pavel Bouchal1 1Masaryk University, Faculty of Science, Biochemistry, Brno, Czech Republic 2Masaryk University, Central European Institute for Technology, Core facility Proteomics, Brno, Czech Republic 3Masaryk Memorial Cancer Institute, Clinical and Experimental Pathology, Brno, Czech Republic Triple-negative breast cancer (TNBC) is the most aggressive subtype of breast cancer, and deeper proteome coverage is needed for its molecular characterization. We present comprehensive library of targeted mass spectrometry assays specific for TNBC and demonstrate its applicability. Proteins were extracted from 105 TNBC tissues and digested. Aliquots were pooled, fractionated using hydrophilic chromatography and analyzed by LC- MS/MS in data-dependent acquisition (DDA) parallel accumulation-serial fragmentation (PASEF) mode on timsTOF Pro LC-MS system. 16 individual lysates were analyzed in data-independent acquisition (DIA)-PASEF mode. Hybrid library was generated in Spectronaut 16.0 software and covers 244,464 precursors, 168,006 peptides and 11,564 protein groups (FDR = 1%). Application of our library for pilot quantitative analysis of 16 tissues increased identification numbers in Spectronaut 18.5 and DIA-NN 1.8.1 software compared to library-free setting, with Spectronaut 18.5 achieving the best results represented by 190,310 precursors, 140,566 peptides, and 10,463 protein groups. In conclusion, we introduce assay library that offers, to our knowledge, the deepest coverage of TNBC proteome to date. The TNBC library is available via PRIDE repository (PXD047793). and European Regional "UP This work was supported by Ministry of Health of the Czech Republic (project NU22-08-00230), all rights reserved. CIISB, Instruct-CZ Centre of Instruct-ERIC EU consortium, funded by MEYS CR infrastructure project LM2023042 (No. CZ.02.1.01/0.0/0.0/18_046/0015974), is gratefully acknowledged for the financial support of the measurements at the CEITEC Proteomics Core Facility (by DP). Supported by the project BBMRI.CZ no. LM2023033. Supported by the project National Institute for Cancer Research (Programme EXCELES, ID Project No. LX22NPO5102)— Funded by the European Union—Next Generation EU. Development Fund-Project CIISB" 1275

1276

LP-02 Monitoring the dynamics of endoplasmic reticulum stress-response pathways in a dual-species co-culture of human aging neurons through DIA mass spectrometry Lea Barny1, Sarah Gracia2,3, Burak Uzay4, Ege Kavalali4, Lars Plate1,2,3 1Vanderbilt University, Chemical and Physical Biology Program, Nashville, United States 2Vanderbilt University, Biological Sciences, Nashville, United States 3Vanderbilt University, Department of Chemistry, Nashville, United States 4Vanderbilt University, Department of Pharmacology, Nashville, United States The proper folding of secretory proteins in the endoplasmic reticulum (ER) is vital to a variety of cellular and organismal functions. However, when ER protein folding capacity is compromised or ER luminal conditions become unfavorable, aggregated and/or misfolded proteins accumulate in the ER resulting in "ER stress". To restore ER protein homeostasis ("proteostasis") and prevent the trafficking of non-native protein conformations to secretory environments, cells have evolved a highly conserved tripartite signal transduction pathway known as the Unfolded Protein Response (UPR). Activation of each arm of the UPR (IRE1, ATF6, and PERK) results in the production of distinct transcription factors which transcriptionally remodel ER proteostasis pathways to combat ER stress. UPR activation has been shown to contribute to the pathogenesis of cancer, metabolic and neurological disorders. Recently, in a co-culture/ dual-species model of human aging neurons (grown on a feeder layer of mouse glia), UPR activation resulted in impaired action potential evoked synchronous neurotransmitter release, a hallmark of neurodegenerative disease. To comprehensively characterize temporal changes in UPR target protein expression throughout neuronal aging, we have implemented a DIA LC-MS/MS proteomics approach. A pipeline in R using DIA-NN result files as an input was then developed and validated to obtain species-specific protein quantification from the coculture model system. In comparison to western blot analysis, this method provides speciesspecific information to understand how the proteomes of neurons and glia change with age and how protein expression in glia changes in the presence and absence of neurons. To assess the status of UPR, previously described and novel proteins were then selected to monitor aggregate activation of these defined target protein sets, resulting in a high throughput assay to assess branch-specific pathway activation. Lastly, to explore how an increase in intracellular calcium affects neurons with age, an NMDAR antagonist was applied to modulate calcium concentration in neurons. Overall, a systematic proteomics evaluation of the three branches of the UPR throughout neuronal aging will inform studies to pharmacologically manipulate the UPR to restore synaptic transmission in the aging brain. 1276

1277

LP-03 Improved top-down analysis of antibody subunits using the Omnitrap-Orbitrap-Booster platform Camille Garcia1, Konstantin Nagornov2, Athanasios Smyrnakis3, Tingting Fu1, Anton Kozhinov2, Dimitris Papanastasiou3, Yury Tsybin2, Julia Chamot-Rooke1 1Institut Pasteur, Paris, France 2Spectroswiss, Lausanne, Switzerland 3Bruker, Athens, Greece A novel instrumental platform called Omnitrap-Orbitrap-Booster with enhanced high-resolution top-down proteomics capabilities was recently introduced by our team. This platform is composed of a Q-Exactive HF Orbitrap interfaced with a multimodal MS/MS Omnitrap, offering ECD and CID, and an external high- performance data acquisition and processing system, FTMS Booster X2. The latter allows to record time-domain transients of any length. Utilizing optimized experimental parameters and refined data processing, we show that this platform outperforms state-of-the-art instruments and allows the complete sequencing of a clinical antibody light chain on an LC time scale. Our platform's efficacy is also showcased through the analysis of other antibody subunits including mixtures. LC-MS/MS experiments were performed on the Omnitrap-Orbitrap-Booster platform (ECD, CID) as well as on an Eclipse Tribrid Orbitrap (EThcD) for comparison. Samples included an antibody light chain extracted from the urine of a multiple myeloma patient as well as commercial mAbs digested into various subunits. The time- domain transients collected by the FTMS Booster X2 (Spectroswiss) were converted into absorption mode FT (aFT). Conventional eFT mass spectra (.RAW) were collected in parallel. Data were analyzed with Proteome Discoverer, Peak-by-Peak (Spectroswiss), and in-house developed software tools. few % to The experimental parameters of the Omnitrap-Orbitrap-Booster platform were first optimized, and we found ECD (100 ms) to lead to the best sequence coverage for the LC-MS/MS analysis of our reduced/alkylated clinical light chain (25 kDa). The analysis of data acquired from the fragmentation of a single charge state (23+) demonstrated a sequence coverage increase of about 10% for aFT data in comparison with the eFT mass spectra acquired in parallel. Averaging time-domain data across five technical replicates further improved sequence coverage but only by a the same sample. We then acquired LC-MS/MS ECD datasets for charge states 15-25 with a single charge state resolution. Combining these results provided a sequence coverage of 100% already from the eFT mass spectra at 3 ppm mass accuracy. The time-domain data averaging produced additional benefits including enhanced confidence in product ion assignment. Our data acquisition strategy was optimized to allow the sequential fragmentation of multiple charge states in a single LC-MS/MS run. This approach circumvents the problem of overlap between charge-reduced species and product ions. Our optimized pipeline is also now currently applied to the top-down analysis of other antibody subunits (25/50 kDa) previously studied with the most recent platforms. Our objective is to achieve a 100% sequence coverage in a minimum of LC-MS/MS runs, which paves the way to the analysis of antibody mixtures and opens the attractive possibility of de novo sequencing from top-down data. reach 91% sequence coverage for 1277

1278

LP-04 Effects of cigarette smoke extracts on vascular smooth muscle cells: a functional and interactomic overview Rossana De Salvo1, Isabella Damiani2, Alfonso Carleo3, Silvia Castiglioni2, Clara Rossi2, Alberto Corsini2,4, Luca Bini1, Stefano Bellosta2, Laura Bianchi1 1University of Siena, Life Science, Siena, Italy 2University, Pharmacological and Biomolecular Sciences “Rodolfo Paoletti”, Milan, Italy 3University, Medicine, Surgery and Dentistry "Scuola Medica Salernitana", Salerno, Italy 4Hospital, IRCCS MultiMedica, Milan, Italy Cigarette smoke (CS) is the primary cause for morbidity and mortality worldwide. Alternative next generation products, such as electronic cigarette (E-cig) and Tobacco Heated Products (THP), are developed with the purpose to reduce the adverse effects of traditional cigarette (TC). CS contains several lipo- and hydrophilic components thought to imbalance cardiovascular system homeostasis. Aim of our study was to define phenotypic changes and molecular pathway modulation caused by CS in vascular smooth muscle cells (VSMC). Firstly, we investigated the effects of CS condensate (CSC) lipophilic components from TC on mouse and human VSMC, then we extended the study to aqueous extracts (AEs) from TC, E-cig, and THP. We performed cell proliferation and wound healing assays, gene expression, western blot, and mass spectrometry (MS)-based differential proteomic analyses. Bioinformatics combination of the acquired results allowed a holistic and interactomic overview on adverse effects of smoking on VSMC. The addition of CSC on VSMC cultures significantly reduced the expression of SMC-specific markers and increased the expression of inflammation-related ones. Our data supported the hypothesis that the acquisition of inflammatory state and the conversion of VSMC into foam-like cells depends on downregulation of Myocardin/Kruppel-like factor 4 (KLF4) axis and may lead to the formation of atherosclerotic plaques. CSC also drives the VSMC phenotypic switch by inducing migration through the increased expression of metalloproteases (MMPs). Also the exposure of VSMC to TC AE was observed to increase the expression of inflammatory markers. On the contrary, E-cig and THP AEs significantly induced the expression of contractile markers and genes involved in extracellular matrix (ECM) remodeling, for several of which we also proved the up-regulation of corresponding proteins. E-cig AE treatment was the most effective in inducing proliferation and migration of VSMC, while THP AE showed a minimal effect on VSMC migratory activity. Finally, the first-shell interactomic analysis highlighted as both CSC and AEs effects on VSMCs converged on and act through KLF4, which resulted, along with MMPs, EGF-containing fibulin like extracellular matrix protein 1 (EFEMP1), protein kinase RNA-activated (PKR), and interleukin-1 beta (IL-1b) the main players in the VSMC phenotype switching induced by smoke extract treatments. 1278

1279

LP-05 Discovery of a novel player in ER-Mitochondria interactions by DARTS-LC-MS/MS proteomics analysis Minjeong Ko1, Jungmin Ha1, Sun-Yeong Kwon2, Ji Young Mun2, Ju Yeon Lee3, Jin Young Kim3, Ho Jeong Kwon1 1Yonsei University, Seoul, South Korea 2Korea Brain Research Institute, Daegu, South Korea 3Korea Basic Science Institute, Ochang, South Korea Mitochondria-associated ER membranes (MAMs) have attracted attention for their important roles in physiological processes such as calcium transport, lipid metabolism, and organelle dynamics. In this study, we observed increased ER-mitochondria contacts and impaired autophagy in vascular endothelial cells stimulated by oxidized low-density lipoprotein (oxLDL) and in the apoE-/- mouse model of atherosclerosis. We found that CTS, a natural active compound , exhibited anti-atherosclerotic effects both in vitro and in vivo. In addition, CTS induced autophagy and reduced ER-mitochondria association. To elucidate the molecular mechanism of CTS underlying this effect, we performed DARTS-LC-MS/MS analysis combined with quantitative SWATH analysis to identify the target protein of CTS. A list of candidate targets showing more than 10% resistance to proteolysis upon CTS treatment was generated, revealing that most were localized to the ER or mitochondria. Based on high sequence coverage, we identified CTS-BP, an ER-resident chaperone protein, as the binding partner of CTS. Knockdown of the CTS-BP gene confirmed its role in mediating CTS effects, leading to MAM disruption and increased autophagy. Our findings provide novel insights into how an anti-atherosclerotic natural product modulates ER-mitochondria interactions to promote autophagy for its pharmacological activity. 1279

1280

LP-06 Inhibition of the ER-associated protein degradation mediator, VCP/p97, regulates organelle interaction Yubin Ko1, Minjeong Ko1, Marius Ueffing2, Jin Young Kim3, Geul Bang3, Ho Jeong Kwon1 1Yonsei university, Biotechnology, Seodaemun-gu, Seoul, South Korea 2Institute for Ophthalmic Research, Ophthalmology, Tübingen, Germany 3Korea Basic Science Institute, Biomedical Omics Group, Ochang, Chungbuk, South Korea To mitigate ER stress, three major protein quality control systems are initiated to maintain proteostasis: the unfolded protein response (UPR), ER-associated protein degradation (ERAD), and autophagy. During ERAD, misfolded proteins must be exported from the ER to the cytosol by a dislocon complex. This process is energy- dependent, and VCP/p97, a multifunctional transitional endoplasmic reticulum ATPase (TER ATPase) acts as a protein unfolding machine with hundreds of substrates in diverse cellular pathways. VCP/p97 is critical for the regulation of proteostasis and the degradation of signaling related proteins, and is implicated in several diseases. Moreover, VCP is essential for maturation of ubiquitin-containing autophagosomes. Here, we show that VCP inhibition at low doses (~1 μM) has cytostatic effects in human umbilical vein endothelial cells (HUVECs) and that autophagy, a self-degradation process, plays a critical role as a survival mechanism. Furthermore, we identified an increase in cytosolic Ca2+ as a key trigger for autophagy, and propose Ca2+ translocation between ER-plasma membrane and ER-mitochondria (Mitochondria-associated ER membranes, MAMs). Interestingly, we observed differences in cell proliferation inhibition between VCP inhibitor treatment and siRNA-mediated VCP knockdown, suggesting multiple-targets of the VCP inhibitor. To identify these new potential binding targets of the VCP inhibitor, we performed a label-free target protein identification method, cellular thermal shift assay (CETSA)-LC- MS/MS analysis. In this presentation, we will focus on elucidating the novel mechanisms by which the VCP inhibitor and its target proteins regulate autophagy and organelle interaction. 1280

1281

LP-07 A transformational palmitoyl transferase chemical genetic system for the deconvolution of specific ZDHHC substrates and S-acylation sites Jack Houghton1, Edward Tate1 1Imperial College London, Chemistry, London, United Kingdom ZDHHC S-acyltransferases catalyse long-chain S-acylation at cysteine residues of >1000 proteins, including >140 oncoproteins. However, protein S‐acylation presents an analytical challenge with no approaches that directly identify or modulate substrates specific to a given ZDHHC. Here we present a transformational chemical genetic system for the deconvolution of ZDHHC substrates and S‐acylation sites that circumvents the problems of lability, LC-MS:. https://www.nature.com/articles/s41587-023-02030-0. This system enables direct proteome‐wide profiling of cognate ZDHHC substrates and S‐acylation sites, and identification and validation of novel substrates specific or common to sets of ZDHHCs for the first time: ionisation, S-acylated of native peptides in and solubility Through structure guided design and homology modelling we developed and optimised a ZDHHC20 "hole" mutant and a "bumped" chemically tagged fatty acid probe. HEK293T, MDA-MB 231 and PANC1 cells transfected with WT ZDHHC20 or mutant ZDHHC20 were treated with a bumped 18-carbon fatty acid probe and proteins recovered following lysis were subjected to CuAAC ligation with azido biotin capture reagent. Non-acylated cysteines were blocked by N-ethylmaleimide treatment and labelled proteins enriched on resin. Thioester linkages were hydrolysed on-bead by hydroxylamine treatment and previously S-acylated cysteine residues capped with chloroacetamide. Followed by off-resin trypsin digestion and nanoLC-MS/MS analysis. This approach was expanded to two other ZDHHCs (7 and 15). Mutant ZDHHCs efficiently load and transfer "bumped" probes whilst WT ZDHHCs do not. Lipidomic analysis shows that bumped probes are activated in cells without disrupting endogenous biosynthetic pathways. 99 potential S-acylation sites were identified, including ZDHHC20 auto-S-acylation and 28 sites previously reported in S-acylation studies. Known ZDHHC20 substrate IFITM3 was among the most enriched proteins, demonstrating sensitivity sufficient to identify substrates at endogenous expression levels. Versatility and adaptability were demonstrated by identification of substrates and sites of modification in MDA-MB-231 and PANC1 cell lines. ZDHHC7 and ZDHHC15 systems identified substrates across three cell lines, alongside sites of modification across multiple proteins. Among the 301 unique substrates identified, we observed common and distinct substrates between ZDHHCs with a degree of redundancy across the family. Comparisons with conventional chemical proteomics and proximity labelling approaches reveals chemical genetics as a complementary approach by circumventing redundancy within ZDHHC substrate networks, whilst enhancing specificity and sensitivity towards substrates. Our system unlocks the potential of drug development from specific ZDHHCs through deepening our understanding of the pathways regulating S‐acylation. 1281

1283

LP-08 Production of antimicrobial peptides is associated with the reduction of severe kidney infections in human patients Hannah Luise Voß1, Stephanie Schaefer-Tautges1, Lars Borgards1, Lisa Schwartz2, Anne Aust2, Florian Wagenlehner2, Daniel Robert Engel1, Olga Shevchuk1 1Institute for Experimental Immunology and Imaging, Department of Immunodynamics, Essen, Germany 2Clinic for Urology, Pediatric Urology and Andrology, Justus-Liebig-University, Giessen, Germany Pyelonephritis (PN) is a common inflammatory disorder of the renal parenchyma, associated with severe morbidity, particularly among elderly and immunodeficient patients. While antibiotics are the most common treatment, the emergence of multi-drug-resistant strains of uropathogenic bacteria necessitates optimized treatment strategies. To unravel the molecular and cellular heterogeneity in PN patients, we analyzed the proteome profile of both the soluble and cellular components of the urine. The measurements were performed using a TimsTOF flex HT coupled with a nano-Elute2 (Bruker). Gene Ontology Biological Process (GOBP)-based gene set enrichment analysis (GSEA) revealed a clear upregulation of biological processes involved in the immune and defense response, including significant upregulation of proteins with antimicrobial functions (AMPs) in PN patients compared to healthy donors. Monte Carlo reference-based consensus clustering of soluble urine proteins demonstrated a clear differentiation of PN patients into two groups, correlating with clinical characteristics. For the first group, showing a less severe and systemic infection, upregulation of reactive oxygen species, complement activation, leukocyte immune response, and positive regulation of AMP secretion were detected. The other cohort with more severe infections showed alterations in extracellular matrix response to growth factors and renal system development. Protein profiles of AMPs, serving as the first line defense of the innate immune system upon infection, along with recently described encrypted AMPs were the most significant factors in distinguishing the two groups of PN patients. Our data, for the first time, decrypts the AMP landscape of the urine in health and PN subgroups and highlights the high diagnostic potential of AMPs as potential alternate antibiotics and new biomarkers in a non- invasive approach for the early detection of tissue damage and the assessment of PN infection severity. 1283

1284

LP-09 Diagnostic application of proteomics: testing a new sample preparation for the typing of amyloid proteins in biopsies Maria Voumvouraki1,2, Panagiota Efstathia Nikolaou3,4, Foteini Theodorakalou4, Giorgos Kanellis5, Nikolaos Thomaidis1,2, Guillaume Medard1,2, Efstathios Kastritis4, Julie Courraud1,2,4 1National and Kapodistrian University of Athens, Department of Chemistry, Athens, Greece 2National and Kapodistrian University of Athens, Proteomics Core Facility, Athens, Greece 3National and Kapodistrian University of Athens, Department of Pharmacy, Athens, Greece 4National and Kapodistrian University of Athens, Department of Clinical Therapeutics, Athens, Greece 5"Evangelismos" General Hospital, Department of Hematopathology, Athens, Greece Amyloidosis refers to a group of heterogeneous clinical conditions caused by the deposition of misfolded proteins in the extracellular matrix of tissues and organs. A key feature of these proteins is their tendency to form antiparallel beta-pleated sheets, which then become non-branching fibrils resistant to degradation. These fibrils cause mechanical damage and oxidative stress in organs like the heart, liver, kidneys, nervous system, and gastrointestinal tract. Τhe gold standard for diagnosing amyloidosis is Congo red staining of sections from formalin fixed paraffin embedded biopsies, showing amyloid deposits as green birefringent areas under polarized light. More than 40 precursor proteins are known to misfold and self-assemble as amyloids. Therefore, upon histological confirmation of the amyloid deposits, the identity of the amyloidogenic (precursor) protein must be determined, given that several amyloidoses have disease-specific therapies. This is called amyloid typing. Mass spectrometry-based bottom-up proteomic analysis of laser microdissected biopsy areas is currently the most reliable method to type amyloid. But sample preparation, as developed 10-15 years ago, is somewhat tedious as the fibrils are not easily lysed. We conducted experiments to compare the state-of-the-art sample preparation used for proteomic typing of amyloid to a simpler procedure inspired from the latest developments in other domains of proteomics. More specifically, we compared a protocol involving lysis with TrisBase, EDTA, Zwittergent (TEZ) buffer to a more user-friendly protocol featuring lysis with trifluoroacetic acid (TFA). Tryptic peptides were analysed on a timsTOF fleX (Bruker) using Data Independent Acquisition - Parallel Accumulation– Serial Fragmentation (DIA-PASEF). Raw data were processed with the DIA-NN library-free software. This study discusses the differences between the two methods in terms of 1) practicalities, cost and time, 2) number of protein groups detected and overlapping profiles, 3) capacity to conclude on the diagnosis. 1284

1285

LP-10 Probing the role of protein S-acylation in biomolecular condensate regulation Amina Nigmatulina1, Jack Houghton1, Matthew White1, Anne Schuhmacher1, Edward Tate1 1Imperial College London, Chemistry, London, United Kingdom Biomolecular condensates have emerged as novel membraneless compartments that regulate cellular biochemistry. Whilst the role of protein post-translational modifications in condensate regulation is an area of intensive interest, the potential role of protein fatty acylation remains largely unexplored. Given the inherent hydrophobicity of this class of PTMs, we hypothesise that protein S-acylation - a reversible PTM involving the attachment of a fatty acyl chain (usually C16) onto >3000 cysteine residues in the human proteome - regulates biomolecular condensate formation and disassembly through multivalent hydrophobic interactions. We are investigating this hypothesis in a model of stress granule (SG) formation to undertake the first comprehensive analysis of regulated S-acylation in a biomolecular condensate and investigate how changes in S-acylation affect SG dynamics. We first developed a novel site-specific S-Acylation Profiling by Phosphonate taggIng pRotEomics platform (SAPPHIRE), enabling sensitive and quantitative mass spectrometry-based analysis of S-acylation. Using SAPPHIRE for global S-acylation profiling in HEK293T cells, we identified 2419 S-acylation sites, 2226 of which are not previously reported in SwissPalm S-acylation database. We then couple SAPPHIRE to APEX2 proximity- dependent biotinylation (APEX-AX) to map changes in the S-acylation state of the SG proteome within a single proteomics workflow. These findings will enhance our understanding of the molecular grammar governing condensates, offering novel insights into basic biology and mechanisms of disease, and in the longer term enable the development of tools to regulate condensate formation through protein lipidation, for research and therapeutic purposes. 1285