Mass Spectrometry-Based Plasma Proteomics: Considerations from Sample Collection to Achieving Translational Data

Abstract

The proteomic analysis of human blood and blood-derived products (e.g., plasma) offers an attractive avenue to translate research progress from the laboratory into the clinic. However, due to its unique protein composition, performing proteomics assays with plasma is challenging. Plasma proteomics has regained interest due to recent technological advances, but challenges imposed by both complications inherent to studying human biology (e.g., interindividual variability) and analysis of biospecimens (e.g., sample variability), as well as technological limitations remain. As part of the Human Proteome Project (HPP), the Human Plasma Proteome Project (HPPP) brings together key aspects of the plasma proteomics pipeline. Here, we provide considerations and recommendations concerning study design, plasma collection, quality metrics, plasma processing workflows, mass spectrometry (MS) data acquisition, data processing, and bioinformatic analysis. With exciting opportunities in studying human health and disease though this plasma proteomics pipeline, a more informed analysis of human plasma will accelerate interest while enhancing possibilities for the incorporation of proteomics-scaled assays into clinical practice.

Keywords: Human Plasma Proteome Project (HPPP); Human Proteome Project (HPP); bioinformatic analysis; blood; data acquisition; data processing; mass spectrometry (MS); plasma; plasma processing workflows; quality metrics; sample collection; serum; study design.

Figure 1.. Frequency of protein identification in relation to plasma and serum concentration.

Using data collected for the 2017 draft of the human plasma and serum proteome (hosted by PeptideAtlas), proteins (in red) are plotted as a function of their concentration rank (x-axis) and the number of studies in which they were identified (y-axis, left). The identified proteins (solid black line) are also plotted as a function of their concentration rank (x-axis) and their estimated concentration (y-axis, right). The data are compiled from 178 samples, 71% of which are from plasma, 14% from serum, and 15% of unclear origin.

Figure 2.. Components of a plasma proteomics workflow.

Profiling proteins in plasma begins with collecting the samples in a standardized manner, reporting pre-analytical variables related to the sample and information about the blood donors. After protein digestion and peptide purification, the peptides are separated by liquid chromatography (LC) and ionized by electro spray (ES) for the analysis in the mass spectrometer (MS). Appropriate MS workflows and peptide identification and quantification tools are then applied. For protein identifications, the HPP guidelines recommend a protein-level FDR of ≤1%. Lastly, data analysis should consider how many peptides and proteins were identified and their consistency across samples.