Establishment and clinical application evaluations of a deep mining strategy of plasma proteomics based on nanomaterial protein coronas

• Published in: Analytica chimica acta 2023-09, Vol.1275, p.341569-341569, Article 341569
• Main Authors: Wang, Jianan, Xie, Wei, Sun, Longqin, Li, Jingli, Wu, Songfeng, Li, Ruibing, Zhao, Yan
• Format: Article
• Language: English
• Subjects: Biomarkers; Deep mining beads; Escherichia coli; Humans; LC-MS; Multiple system atrophy; Nanostructures; Pilot Projects; Plasma proteomics; Protein Corona; Proteome - analysis; Proteomics - methods
• ISSN: 0003-2670; 1873-4324
• DOI: 10.1016/j.aca.2023.341569

Research on plasma proteomics has received extensive attention, because human plasma is an important sample for disease biomarker research due to its easy clinical accessibility and richness in biological information. Plasma samples contain a large number of leaked proteins from different tissues in the body, immune proteins and communication signal proteins. However, MS signal suppression from high-abundance proteins results in a large number of proteins that are present in low abundance in plasma not being detected by the LC-MS method. This situation makes it more difficult to study neurological diseases, where tissue sampling is difficult and body fluid samples such as plasma or cerebrospinal fluid are both affected by signal suppression. A large number of methods have been developed to deeply mine plasma proteomics information; however, their application limitations remain to some extent. Traditional immuno- or affinity-based depletion, fractionation and subproteome enrichment methods cannot meet the challenges of large clinical cohort applications due to limited time efficiency. In this study, a deep mining strategy of plasma proteomics was established by combing the protein corona formed by deep mining beads (DMB beads, hereafter referred to as magnetic covalent organic frameworks Fe3O4@TpPa-1), DIA-MS detection and the DIA-NN library searching method. By optimizing the enrichment step, mass spectrometry acquisition and data processing, the evaluation results of the deep mining strategy showed the following: depth, the strategy identified and quantified results of 2000+ proteins per plasma sample; stability, more than 87% of the enriched low-abundance proteins had CV < 20%; accuracy, good agreement between measured and theoretical values (1.81/2, 8.68/10, 38.36/50) for the gradient addition of E. coli proteins to a plasma sample; time efficiency, the processing time was reduced from >12h in the traditional method to