Proteomic Profiling of Advanced Melanoma Patients to Predict Therapeutic Response to Anti-PD-1 Therapy.

PURPOSE: Despite high clinical need, there are no biomarkers that accurately predict the response of patients with metastatic melanoma to anti-PD-1 therapy. EXPERIMENTAL DESIGN: In this multicenter study, we applied protein depletion and enrichment methods prior to various proteomic techniques to analyze a serum discovery cohort (n = 56) and three independent serum validation cohorts (n = 80, n = 12, n = 17). Further validation analyses by literature and survival analysis followed. RESULTS: We identified several significantly regulated proteins as well as biological processes such as neutrophil degranulation, cell-substrate adhesion, and extracellular matrix organization. Analysis of the three independent serum validation cohorts confirmed the significant differences between responders (R) and nonresponders (NR) observed in the initial discovery cohort. In addition, literature-based validation highlighted 30 markers overlapping with previously published signatures. Survival analysis using the TCGA database showed that overexpression of 17 of the markers we identified correlated with lower overall survival in patients with melanoma. CONCLUSIONS: Ultimately, this multilayered serum analysis led to a potential marker signature with 10 key markers significantly altered in at least two independent serum cohorts: CRP, LYVE1, SAA2, C1RL, CFHR3, LBP, LDHB, S100A8, S100A9, and SAA1, which will serve as the basis for further investigation. In addition to patient serum, we analyzed primary melanoma tumor cells from NR and found a potential marker signature with four key markers: LAMC1, PXDN, SERPINE1, and VCAN.

Proteomic-based stratification of intermediate-risk prostate cancer patients.

Gleason grading is an important prognostic indicator for prostate adenocarcinoma and is crucial for patient treatment decisions. However, intermediate-risk patients diagnosed in the Gleason grade group (GG) 2 and GG3 can harbour either aggressive or non-aggressive disease, resulting in under- or overtreatment of a significant number of patients. Here, we performed proteomic, differential expression, machine learning, and survival analyses for 1,348 matched tumour and benign sample runs from 278 patients. Three proteins (F5, TMEM126B, and EARS2) were identified as candidate biomarkers in patients with biochemical recurrence. Multivariate Cox regression yielded 18 proteins, from which a risk score was constructed to dichotomize prostate cancer patients into low- and high-risk groups. This 18-protein signature is prognostic for the risk of biochemical recurrence and completely independent of the intermediate GG. Our results suggest that markers generated by computational proteomic profiling have the potential for clinical applications including integration into prostate cancer management.

Functional and spatial proteomics profiling reveals intra- and intercellular signaling crosstalk in colorectal cancer.

Precision oncology approaches for patients with colorectal cancer (CRC) continue to lag behind other solid cancers. Functional precision oncology-a strategy that is based on perturbing primary tumor cells from cancer patients-could provide a road forward to personalize treatment. We extend this paradigm to measuring proteome activity landscapes by acquiring quantitative phosphoproteomic data from patient-derived organoids (PDOs). We show that kinase inhibitors induce inhibitor- and patient-specific off-target effects and pathway crosstalk. Reconstruction of the kinase networks revealed that the signaling rewiring is modestly affected by mutations. We show non-genetic heterogeneity of the PDOs and upregulation of stemness and differentiation genes by kinase inhibitors. Using imaging mass-cytometry-based profiling of the primary tumors, we characterize the tumor microenvironment (TME) and determine spatial heterocellular crosstalk and tumor-immune cell interactions. Collectively, we provide a framework for inferring tumor cell intrinsic signaling and external signaling from the TME to inform precision (immuno-) oncology in CRC.

Integrative solution structure of PTBP1-IRES complex reveals strong compaction and ordering with residual conformational flexibility.

RNA-binding proteins (RBPs) are crucial regulators of gene expression, often composed of defined domains interspersed with flexible, intrinsically disordered regions. Determining the structure of ribonucleoprotein (RNP) complexes involving such RBPs necessitates integrative structural modeling due to their lack of a single stable state. In this study, we integrate magnetic resonance, mass spectrometry, and small-angle scattering data to determine the solution structure of the polypyrimidine-tract binding protein 1 (PTBP1/hnRNP I) bound to an RNA fragment from the internal ribosome entry site (IRES) of the encephalomyocarditis virus (EMCV). This binding, essential for enhancing the translation of viral RNA, leads to a complex structure that demonstrates RNA and protein compaction, while maintaining pronounced conformational flexibility. Acting as an RNA chaperone, PTBP1 orchestrates the IRES RNA into a few distinct conformations, exposing the RNA stems outward. This conformational diversity is likely common among RNP structures and functionally important. Our approach enables atomic-level characterization of heterogeneous RNP structures.

A computational framework for the inference of protein complex remodeling from whole-proteome measurements.

Protein complexes are responsible for the enactment of most cellular functions. For the protein complex to form and function, its subunits often need to be present at defined quantitative ratios. Typically, global changes in protein complex composition are assessed with experimental approaches that tend to be time consuming. Here, we have developed a computational algorithm for the detection of altered protein complexes based on the systematic assessment of subunit ratios from quantitative proteomic measurements. We applied it to measurements from breast cancer cell lines and patient biopsies and were able to identify strong remodeling of HDAC2 epigenetic complexes in more aggressive forms of cancer. The presented algorithm is available as an R package and enables the inference of changes in protein complex states by extracting functionally relevant information from bottom-up proteomic datasets.

Proteomic Dynamics of Breast Cancer Cell Lines Identifies Potential Therapeutic Protein Targets.

Treatment and relevant targets for breast cancer (BC) remain limited, especially for triple-negative BC (TNBC). We identified 6091 proteins of 76 human BC cell lines using data-independent acquisition (DIA). Integrating our proteomic findings with prior multi-omics datasets, we found that including proteomics data improved drug sensitivity predictions and provided insights into the mechanisms of action. We subsequently profiled the proteomic changes in nine cell lines (five TNBC and four non-TNBC) treated with EGFR/AKT/mTOR inhibitors. In TNBC, metabolism pathways were dysregulated after EGFR/mTOR inhibitor treatment, while RNA modification and cell cycle pathways were affected by AKT inhibitor. This systematic multi-omics and in-depth analysis of the proteome of BC cells can help prioritize potential therapeutic targets and provide insights into adaptive resistance in TNBC.

A DNA-PK phosphorylation site on MET regulates its signaling interface with the DNA damage response.

The DNA damage response (DDR) is intertwined with signaling pathways downstream of oncogenic receptor tyrosine kinases (RTKs). To drive research into the application of targeted therapies as radiosensitizers, a better understanding of this molecular crosstalk is necessary. We present here the characterization of a previously unreported MET RTK phosphosite, Serine 1016 (S1016) that represents a potential DDR-MET interface. MET S1016 phosphorylation increases in response to irradiation and is mainly targeted by DNA-dependent protein kinase (DNA-PK). Phosphoproteomics unveils an impact of the S1016A substitution on the overall long-term cell cycle regulation following DNA damage. Accordingly, the abrogation of this phosphosite strongly perturbs the phosphorylation of proteins involved in the cell cycle and formation of the mitotic spindle, enabling cells to bypass a G2 arrest upon irradiation and leading to the entry into mitosis despite compromised genome integrity. This results in the formation of abnormal mitotic spindles and a lower proliferation rate. Altogether, the current data uncover a novel signaling mechanism through which the DDR uses a growth factor receptor system for regulating and maintaining genome stability.

M. tuberculosis relies on trace oxygen to maintain energy homeostasis and survive in hypoxic environments.

The bioenergetic mechanisms by which Mycobacterium tuberculosis survives hypoxia are poorly understood. Current models assume that the bacterium shifts to an alternate electron acceptor or fermentation to maintain membrane potential and ATP synthesis. Counterintuitively, we find here that oxygen itself is the principal terminal electron acceptor during hypoxic dormancy. M. tuberculosis can metabolize oxygen efficiently at least two orders of magnitude below the concentration predicted to occur in hypoxic lung granulomas. Despite a difference in apparent affinity for oxygen, both the cytochrome bcc:aa3 and cytochrome bd oxidase respiratory branches are required for hypoxic respiration. Simultaneous inhibition of both oxidases blocks oxygen consumption, reduces ATP levels, and kills M. tuberculosis under hypoxia. The capacity of mycobacteria to scavenge trace levels of oxygen, coupled with the absence of complex regulatory mechanisms to achieve hierarchal control of the terminal oxidases, may be a key determinant of long-term M. tuberculosis survival in hypoxic lung granulomas.

Quantitative proteomic landscapes of primary and recurrent glioblastoma reveal a protumorigeneic role for FBXO2-dependent glioma-microenvironment interactions.

BACKGROUND: Recent efforts have described the evolution of glioblastoma from initial diagnosis to post-treatment recurrence on a genomic and transcriptomic level. However, the evolution of the proteomic landscape is largely unknown. METHODS: Sequential window acquisition of all theoretical fragment ion spectra mass spectrometry (SWATH-MS) was used to characterize the quantitative proteomes of two independent cohorts of paired newly diagnosed and recurrent glioblastomas. Recurrence-associated proteins were validated using immunohistochemistry and further studied in human glioma cell lines, orthotopic xenograft models, and human organotypic brain slice cultures. External spatial transcriptomic, single-cell, and bulk RNA sequencing data were analyzed to gain mechanistic insights. RESULTS: Although overall proteomic changes were heterogeneous across patients, we identified BCAS1, INF2, and FBXO2 as consistently upregulated proteins at recurrence and validated these using immunohistochemistry. Knockout of FBXO2 in human glioma cells conferred a strong survival benefit in orthotopic xenograft mouse models and reduced invasive growth in organotypic brain slice cultures. In glioblastoma patient samples, FBXO2 expression was enriched in the tumor infiltration zone and FBXO2-positive cancer cells were associated with synaptic signaling processes. CONCLUSIONS: These findings demonstrate a potential role of FBXO2-dependent glioma-microenvironment interactions to promote tumor growth. Furthermore, the published datasets provide a valuable resource for further studies.

An oncogene addiction phosphorylation signature and its derived scores inform tumor responsiveness to targeted therapies.

PURPOSE: Oncogene addiction provides important therapeutic opportunities for precision oncology treatment strategies. To date the cellular circuitries associated with driving oncoproteins, which eventually establish the phenotypic manifestation of oncogene addiction, remain largely unexplored. Data suggest the DNA damage response (DDR) as a central signaling network that intersects with pathways associated with deregulated addicting oncoproteins with kinase activity in cancer cells. EXPERIMENTAL: DESIGN: We employed a targeted mass spectrometry approach to systematically explore alterations in 116 phosphosites related to oncogene signaling and its intersection with the DDR following inhibition of the addicting oncogene alone or in combination with irradiation in MET-, EGFR-, ALK- or BRAF (V600)-positive cancer models. An NSCLC tissue pipeline combining patient-derived xenografts (PDXs) and ex vivo patient organotypic cultures has been established for treatment responsiveness assessment. RESULTS: We identified an 'oncogene addiction phosphorylation signature' (OAPS) consisting of 8 protein phosphorylations (ACLY S455, IF4B S422, IF4G1 S1231, LIMA1 S490, MYCN S62, NCBP1 S22, P3C2A S259 and TERF2 S365) that are significantly suppressed upon targeted oncogene inhibition solely in addicted cell line models and patient tissues. We show that the OAPS is present in patient tissues and the OAPS-derived score strongly correlates with the ex vivo responses to targeted treatments. CONCLUSIONS: We propose a score derived from OAPS as a quantitative measure to evaluate oncogene addiction of cancer cell samples. This work underlines the importance of protein phosphorylation assessment for patient stratification in precision oncology and corresponding identification of tumor subtypes sensitive to inhibition of a particular oncogene.

Artificial intelligence defines protein-based classification of thyroid nodules.

Determination of malignancy in thyroid nodules remains a major diagnostic challenge. Here we report the feasibility and clinical utility of developing an AI-defined protein-based biomarker panel for diagnostic classification of thyroid nodules: based initially on formalin-fixed paraffin-embedded (FFPE), and further refined for fine-needle aspiration (FNA) tissue specimens of minute amounts which pose technical challenges for other methods. We first developed a neural network model of 19 protein biomarkers based on the proteomes of 1724 FFPE thyroid tissue samples from a retrospective cohort. This classifier achieved over 91% accuracy in the discovery set for classifying malignant thyroid nodules. The classifier was externally validated by blinded analyses in a retrospective cohort of 288 nodules (89% accuracy; FFPE) and a prospective cohort of 294 FNA biopsies (85% accuracy) from twelve independent clinical centers. This study shows that integrating high-throughput proteomics and AI technology in multi-center retrospective and prospective clinical cohorts facilitates precise disease diagnosis which is otherwise difficult to achieve by other methods.

Proteomic identification of proliferation and progression markers in human polycythemia vera stem and progenitor cells.

Polycythemia vera (PV) is a stem cell disorder characterized by hyperproliferation of the myeloid lineages and the presence of an activating JAK2 mutation. To elucidate mechanisms controlling PV stem and progenitor cell biology, we applied a recently developed highly sensitive data-independent acquisition mass spectrometry workflow to purified hematopoietic stem and progenitor cell (HSPC) subpopulations of patients with chronic and progressed PV. We integrated proteomic data with genomic, transcriptomic, flow cytometry, and in vitro colony formation data. Comparative analyses revealed added information gained by proteomic compared with transcriptomic data in 30% of proteins with changed expression in PV patients. Upregulated biological pathways in hematopoietic stem and multipotent progenitor cells (HSC/MPPs) of PV included mammalian target of rapamycin (MTOR), STAT, and interferon signaling. We further identified a prominent reduction of clusterin (CLU) protein expression and a corresponding activation of nuclear factor-κB (NF-κB) signaling in HSC/MPPs of untreated PV patients compared with controls. Reversing the reduction of CLU and inhibiting NF-κB signaling decreased proliferation and differentiation of PV HSC/MPPs in vitro. Upon progression of PV, we identified upregulation of LGALS9 and SOCS2 protein expression in HSC/MPPs. Treatment of patients with hydroxyurea normalized the expression of CLU and NF-κB2 but not of LGALS9 and SOCS2. These findings expand the current understanding of the molecular pathophysiology underlying PV and provide new potential targets (CLU and NF-κB) for antiproliferative therapy in patients with PV.

Progress Identifying and Analyzing the Human Proteome: 2021 Metrics from the HUPO Human Proteome Project.

The 2021 Metrics of the HUPO Human Proteome Project (HPP) show that protein expression has now been credibly detected (neXtProt PE1 level) for 18 357 (92.8%) of the 19 778 predicted proteins coded in the human genome, a gain of 483 since 2020 from reports throughout the world reanalyzed by the HPP. Conversely, the number of neXtProt PE2, PE3, and PE4 missing proteins has been reduced by 478 to 1421. This represents remarkable progress on the proteome parts list. The utilization of proteomics in a broad array of biological and clinical studies likewise continues to expand with many important findings and effective integration with other omics platforms. We present highlights from the Immunopeptidomics, Glycoproteomics, Infectious Disease, Cardiovascular, Musculo-Skeletal, Liver, and Cancers B/D-HPP teams and from the Knowledgebase, Mass Spectrometry, Antibody Profiling, and Pathology resource pillars, as well as ethical considerations important to the clinical utilization of proteomics and protein biomarkers.

Phenotypic characterization of two novel cell line models of castration-resistant prostate cancer.

BACKGROUND: Resistance to androgen deprivation therapies is a major driver of mortality in advanced prostate cancer. Therefore, there is a need to develop new preclinical models that allow the investigation of resistance mechanisms and the assessment of drugs for the treatment of castration-resistant prostate cancer. METHODS: We generated two novel cell line models (LAPC4-CR and VCaP-CR) which were derived by passaging LAPC4 and VCaP cells in vivo and in vitro under castrate conditions. We performed detailed transcriptomic (RNA-seq) and proteomic analyses (SWATH-MS) to delineate expression differences between castration-sensitive and castration-resistant cell lines. Furthermore, we characterized the in vivo and in vitro growth characteristics of these novel cell line models. RESULTS: The two cell line derivatives LAPC4-CR and VCaP-CR showed castration-resistant growth in vitro and in vivo which was only minimally inhibited by AR antagonists, enzalutamide, and bicalutamide. High-dose androgen treatment resulted in significant growth arrest of VCaP-CR but not in LAPC4-CR cells. Both cell lines maintained AR expression, but exhibited distinct expression changes on the mRNA and protein level. Integrated analyses including data from LNCaP and the previously described castration-resistant LNCaP-abl cells revealed an expression signature of castration resistance. CONCLUSIONS: The two novel cell line models LAPC4-CR and VCaP-CR and their comprehensive characterization on the RNA and protein level represent important resources to study the molecular mechanisms of castration resistance.

On the feasibility of deep learning applications using raw mass spectrometry data.

SUMMARY: In recent years, SWATH-MS has become the proteomic method of choice for data-independent-acquisition, as it enables high proteome coverage, accuracy and reproducibility. However, data analysis is convoluted and requires prior information and expert curation. Furthermore, as quantification is limited to a small set of peptides, potentially important biological information may be discarded. Here we demonstrate that deep learning can be used to learn discriminative features directly from raw MS data, eliminating hence the need of elaborate data processing pipelines. Using transfer learning to overcome sample sparsity, we exploit a collection of publicly available deep learning models already trained for the task of natural image classification. These models are used to produce feature vectors from each mass spectrometry (MS) raw image, which are later used as input for a classifier trained to distinguish tumor from normal prostate biopsies. Although the deep learning models were originally trained for a completely different classification task and no additional fine-tuning is performed on them, we achieve a highly remarkable classification performance of 0.876 AUC. We investigate different types of image preprocessing and encoding. We also investigate whether the inclusion of the secondary MS2 spectra improves the classification performance. Throughout all tested models, we use standard protein expression vectors as gold standards. Even with our naïve implementation, our results suggest that the application of deep learning and transfer learning techniques might pave the way to the broader usage of raw mass spectrometry data in real-time diagnosis. AVAILABILITY AND IMPLEMENTATION: The open source code used to generate the results from MS images is available on GitHub: https://ibm.biz/mstransc. The raw MS data underlying this article cannot be shared publicly for the privacy of individuals that participated in the study. Processed data including the MS images, their encodings, classification labels and results can be accessed at the following link: https://ibm.box.com/v/mstc-supplementary. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

The protein landscape of chronic lymphocytic leukemia.

Many functional consequences of mutations on tumor phenotypes in chronic lymphocytic leukemia (CLL) are unknown. This may be in part due to a scarcity of information on the proteome of CLL. We profiled the proteome of 117 CLL patient samples with data-independent acquisition mass spectrometry and integrated the results with genomic, transcriptomic, ex vivo drug response, and clinical outcome data. We found trisomy 12, IGHV mutational status, mutated SF3B1, trisomy 19, del(17)(p13), del(11)(q22.3), mutated DDX3X and MED12 to influence protein expression (false discovery rate [FDR] = 5%). Trisomy 12 and IGHV status were the major determinants of protein expression variation in CLL as shown by principal-component analysis (1055 and 542 differentially expressed proteins, FDR = 5%). Gene set enrichment analyses of CLL with trisomy 12 implicated B-cell receptor (BCR)/phosphatidylinositol 3-kinase (PI3K)/AKT signaling as a tumor driver. These findings were supported by analyses of protein abundance buffering and protein complex formation, which identified limited protein abundance buffering and an upregulated protein complex involved in BCR, AKT, MAPK, and PI3K signaling in trisomy 12 CLL. A survey of proteins associated with trisomy 12/IGHV-independent drug response linked STAT2 protein expression with response to kinase inhibitors, including Bruton tyrosine kinase and mitogen-activated protein kinase kinase (MEK) inhibitors. STAT2 was upregulated in unmutated IGHV CLL and trisomy 12 CLL and required for chemokine/cytokine signaling (interferon response). This study highlights the importance of protein abundance data as a nonredundant layer of information in tumor biology and provides a protein expression reference map for CLL.

Systematic detection of functional proteoform groups from bottom-up proteomic datasets.

To a large extent functional diversity in cells is achieved by the expansion of molecular complexity beyond that of the coding genome. Various processes create multiple distinct but related proteins per coding gene - so-called proteoforms - that expand the functional capacity of a cell. Evaluating proteoforms from classical bottom-up proteomics datasets, where peptides instead of intact proteoforms are measured, has remained difficult. Here we present COPF, a tool for COrrelation-based functional ProteoForm assessment in bottom-up proteomics data. It leverages the concept of peptide correlation analysis to systematically assign peptides to co-varying proteoform groups. We show applications of COPF to protein complex co-fractionation data as well as to more typical protein abundance vs. sample data matrices, demonstrating the systematic detection of assembly- and tissue-specific proteoform groups, respectively, in either dataset. We envision that the presented approach lays the foundation for a systematic assessment of proteoforms and their functional implications directly from bottom-up proteomic datasets.

Structural and functional dissection of reovirus capsid folding and assembly by the prefoldin-TRiC/CCT chaperone network.

Intracellular protein homeostasis is maintained by a network of chaperones that function to fold proteins into their native conformation. The eukaryotic TRiC chaperonin (TCP1-ring complex, also called CCT for cytosolic chaperonin containing TCP1) facilitates folding of a subset of proteins with folding constraints such as complex topologies. To better understand the mechanism of TRiC folding, we investigated the biogenesis of an obligate TRiC substrate, the reovirus σ3 capsid protein. We discovered that the σ3 protein interacts with a network of chaperones, including TRiC and prefoldin. Using a combination of cryoelectron microscopy, cross-linking mass spectrometry, and biochemical approaches, we establish functions for TRiC and prefoldin in folding σ3 and promoting its assembly into higher-order oligomers. These studies illuminate the molecular dynamics of σ3 folding and establish a biological function for TRiC in virus assembly. In addition, our findings provide structural and functional insight into the mechanism by which TRiC and prefoldin participate in the assembly of protein complexes.

Mapping specificity, cleavage entropy, allosteric changes and substrates of blood proteases in a high-throughput screen.

Proteases are among the largest protein families and critical regulators of biochemical processes like apoptosis and blood coagulation. Knowledge of proteases has been expanded by the development of proteomic approaches, however, technology for multiplexed screening of proteases within native environments is currently lacking behind. Here we introduce a simple method to profile protease activity based on isolation of protease products from native lysates using a 96FASP filter, their analysis in a mass spectrometer and a custom data analysis pipeline. The method is significantly faster, cheaper, technically less demanding, easy to multiplex and produces accurate protease fingerprints. Using the blood cascade proteases as a case study, we obtain protease substrate profiles that can be used to map specificity, cleavage entropy and allosteric effects and to design protease probes. The data further show that protease substrate predictions enable the selection of potential physiological substrates for targeted validation in biochemical assays.