Mining the plasma proteome: Evaluation of enrichment methods for depth and reproducibility

Abstract

Plasma is a complex biological fluid containing extracellular vesicles (EVs), residual platelets, and soluble proteins. While conventional plasma proteomics typically identifies hundreds of proteins, recent enrichment strategies have expanded coverage to thousands. It is still unclear whether these methods enrich preferentially different classes of protein and whether they allow for reliable quantification. Here, we compared three common advanced proteomic workflows—Proteograph (Seer), Mag-Net (ReSynBio), and ENRICHplus (PreOmics) as well as EV enrichment obtained by centrifugation. We explore the content in soluble proteins, EV cargo, and platelet-derived proteins after the enrichments. Quantification was evaluated comparing each method to neat plasma using protein coefficient of variation and point-biserial correlation.

We quantified an average of ∼4500 proteins with EV centrifugation, ∼4000 with Seer, ∼2800 with ENRICHplus, ∼2300 with Mag-Net, and ∼ 900 with neat plasma. Each method enriched distinct sets of protein signatures: EV preparations were enriched with EV markers such as CD81; ENRICHplus predominantly captured lipoproteins; and Proteograph was enriched for cytokines and hormones. Platelet protein intensity was directly correlated with total protein identifications but did not compromise quantification of low-abundance proteins. Across 50 healthy individuals, Proteograph consistently demonstrated reproducible enrichment and depletion patterns, with some reported exceptions.

Our results highlight the strengths and biases of different plasma enrichment strategies.

Significance

This study benchmarks corona formation strategies for enriching low-abundance plasma proteins, including those from platelets and EVs. While enabling deeper proteome coverage compared to depletion methods, these approaches also reshape the intensity landscape and reveal method-specific patterns in protein class enrichment and in quantification repeatability.