[Enriching plasma exosomes for proteomic analysis using a phosphatidylserine-imprinted polymer]

Exosomes are 40-160 nm vesicular nano-bodies secreted by most cells that carry large amounts of biologically active substances originating from the parent cell. Proteins in exosomes are protected by phospholipid bilayer membranes that protect them from degradation by enzymes within body fluids. Along with nucleic acid, proteins and metabolites, exosomes are biomolecules that are considered to be among the most important for discovering tumor markers. Plasma is among the most commonly used body fluids in clinical settings; it is highly complex and contains many proteins and metabolites that interfere with exosome isolation. Consequently, the development of methods for effectively isolating exosomes is a key challenge prior to their use in clinical research. In this study, we used a phosphatidylserine molecularly imprinted polymer (PS-MIP) to enrich plasma exosomes. Subsequent immunoblotting analyses for the CD9, TSG101, and CD81 exosome marker proteins showed that signals can be detected using only 5 μL of plasma, thereby demonstrating the efficiency and specificity of the enrichment protocol. Transmission electron microscope (TEM) and nanoparticle tracking analysis (NTA) data revealed that the enriched vesicles are 30-100 nm in size with elliptical or cup-shaped structures, consistent with the morphology and particle-size-distribution characteristics of the exosomes, suggesting that PS-MIP is capable of successfully isolating exosomes. Nanoflow cytometry revealed that 75.4% of the multi-angle laser scattering (MALS) signal is derived from the PS-MIP-enriched exosomes, which indicates that these enriched exosomes are highly pure and free of interference from impurities, such as aggregated protein particles that are similar in size to the exosomes themselves. This method was used to analyze the proteomes and potential exosomal protein markers of clinical plasma samples from three pancreatic-cancer patients and three healthy volunteers. A total of 1052 proteins and 4545 peptides were identified in the plasma exosomes of healthy volunteers, with a total of 972 proteins and 4096 peptides identified in the plasma exosomes of the pancreatic-cancer patients. Further bioinformatics analyses revealed that the Vesiclepedia database covered 84% of the proteins identified in the plasma exosomes isolated using the PS-MIP method; these proteins comprise 77 of the 100 most frequently identified exosomal proteins in the ExoCarta database. The identified proteins from the cellular components were subjected to gene ontology (GO) analysis, which revealed that they are mainly derived from the exosomes, thereby demonstrating the high selectivity of the PS-MIP method for enriching plasma exosomes and providing specificity for subsequent tumor-marker screening. Label-free quantitative analysis showed that 11 proteins were upregulated and 24 proteins were downregulated in the plasma exosomes of patients with pancreatic cancer compared to those of healthy volunteers. The highly expressed and lowly expressed proteins in the plasma exosomes of patients with pancreatic cancer were subjected to GO, which showed that highly expressed proteins related to the positive regulation of metabolic and biological processes were found in the plasma exosomes of patients with pancreatic cancer compared to those of healthy volunteers, whereas the most significantly under-expressed proteins are related immune-system processes, followed by stimulus-responsive, multicellular bioprocesses, bioregulatory, and interspecies-interacting biological-process-related proteins. The top three proteins, which are relatively highly correlated through protein-protein interaction networks (PPI) analysis, were determined to be complement factor D (CFD), complement component 3 (C3), and von Willebrand factor (VWF). Among the upregulated proteins in the exosomes of patients with pancreatic cancer, exostosin-like glycosyltransferase 2 (EXTL2), α-2-macroglobulin like 1 (A2ML1), and Parkinson's disease protein 7 (PARK7) were the most significantly overexpressed. Hence, these proteins are potential biomarkers for the diagnostic and prognostic assessment of pancreatic cancer and may provide support for further clinical studies into pancreatic cancer.