Steed, H. | bioRxiv Science

bioRxiv2024

Mining the Proteome of Human Ovarian Cancer Extracellular Vesicles Using Thermolysin Proteolysis

This study investigates the utility of Thermolysin as a proteolytic enzyme to enhance the depth and coverage of proteomic analysis in ovarian cancer (OC) extracellular vesicles (EVs). EVs were isolated from OC cell lines and ascites fluid samples from women diagnosed with high-grade serous carcinoma. Proteins were digested using Thermolysin and Trypsin/LysC, followed by label-free data-dependent acquisition (DDA) and data-independent acquisition (DIA) mass spectrometry. The proteolytic efficiency, sequence coverage, peptide complexity, and proteomic depth were compared between Thermolysin and Trypsin/LysC digests. In silico analyses predicted theoretical benchmarks of these parameters using a core set of 22 proteins, and Gene set enrichment analyses (GSEA) highlighted the biological relevance of proteins identified throughout the study. Thermolysin digestion significantly increased the complexity or of peptide pools compared to Trypsin/LysC leading to limited peptide and protein identification, albeit total sequence coverage was increased through complementation to tryptic peptides. In both cell line and ascites EVs, Thermolysin identified unique proteins not detected by Trypsin/LysC that are known drivers of metastatic solid cancers, such as Ly6E. Offline strong cation exchange (SCX) fractionation improved proteomic depth and sequence coverage obtained with Thermolysin to generate a spectral library for DIA. Our DIA analysis of Thermolysin digests revealed the presence of NODAL in OC ascites EVs, a protein associated with poor clinical prognosis, which was not detected in Trypsin/LysC digests. The importance of NODAL was cross-validated in TGCA-OV and AOCS datasets by clinical cohorts by assessing RNA levels in solid tumors or ascites fluid, respectively. Collectively, we demonstrate that Thermolysin complements traditional enzymes like Trypsin/LysC to provide a more comprehensive proteomic landscape for biomarker discovery.

cancer biology↗

bioRxiv2022

Targeted Proteomics and Support Vector Classification Reveal Potential Biomarkers for the Early Detection of High-grade Serous Ovarian Cancer

PurposeThe five-year prognosis for patients with late-stage high-grade serous carcinoma (HGSC) remains dismal, underscoring the critical need for identifying early-stage biomarkers. This study explores the potential of extracellular vesicles (EVs) circulating in blood, which are believed to harbor proteomic cargo reflective of the HGSC microenvironment, as a source for biomarker discovery. Experimental DesignWe conducted a comprehensive proteomic profiling of EVs isolated from blood plasma, ascites, and cell lines of patients, employing both data-dependent (DDA) and data-independent acquisition (DIA) methods to construct a spectral library tailored for targeted proteomics. Our investigation aimed at uncovering novel biomarkers for the early detection of HGSC by comparing the proteomic signatures of EVs from women with HGSC to those with benign gynecological conditions. The initial cohort, comprising 9-10 donors, utilized DDA proteomics for spectral library development. The subsequent cohort, involving 30 HGSC patients and 30 control subjects, employed DIA proteomics for a similar purpose. Support vector machine (SVM) classification was applied in both cohorts to identify combinatorial biomarkers with high specificity and sensitivity (ROC-AUC > 0.90). ResultsNotably, MUC1 emerged as a significant biomarker in both cohorts when used in combination with additional biomarkers. Validation through an ELISA assay on a subset of benign (n=18), Stage I (n=9), and stage II (n=9) plasma samples corroborated the diagnostic utility of MUC1 in the early-stage detection of HGSC. ConclusionsThis study highlights the value of EV-based proteomic analysis in the discovery of combinatorial biomarkers for early ovarian cancer detection.

cancer biology↗