|
Abstract INTRODUCTION
The term “blood proteomics” is most often associated with the analysis of plasma or serum and ignores the contribution of abundant cell-types. We have recently developed methods for mass spectrometry-based quantification of proteins, N-glycopeptides and phosphopeptides from dried blood collected by volumetric microsampling, and we applied these methods to a longitudinal study of sepsis (Foster MW et al. biorxiv 2025). Numerous acute phase and neutrophil-derived derived proteins, and related phosphosites and glycoforms, were reduced between presentation and recovery at d28-42. Here, we sought to replicate those findings in a cross-sectional sepsis cohort and to further extend these methods to dried blood spots (DBS).
METHODS
Venous EDTA-blood from adult healthy controls, patients with sepsis-3 criteria was collected on 20 µL Mitra devices. Alternatively, venous EDTA-blood was obtained from infants with or without Pompe disease (GSD type II) diagnosis based on deficient acid alpha-glucosidase (GAA) enzyme activity, and adult healthy controls, and spotted onto dried blood spot cards. Mitra tips, or 6 mm punch from DBS cards, were processed using deoxycholate-assisted in solution trypsin digestion. Approximately 1 mg of tryptic digests were enriched for N-glycopeptides using hydrophilic ion chromatography (HILIC), and the flow through was enriched for phosphopeptides using immobilized metal affinity chromatography (IMAC). The unenriched proteome was analyzed using microflowLC and Orbitrap Astral data-independent acquisition, and the N-glycoproteome and phoshoproteomes were analyzed using an Evosep LC and stepped collision energy Orbitrap data-dependent acquisition or Orbitrap Astral DIA. Data was analyzed using Spectronaut, FragPipe and Skyline.
RESULTS
Approximate 10,000 analytes (1,600 proteins, 4,000 glycopeptides and 4,600 phosphosites) were quantified from Mitra devices loaded with venous EDTA-blood. Multi-omics factor analysis (MOFA) separated healthy control and sepsis-3 patients on a latent factors plot, with factor 1 weighting acute phase (e.g. LBP, CRP and FGL1) and immune cell proteins (e.g. MMP8, IFM3, PERM), glycoforms of AACT and phosphosites of OSTP, GOLM1 and GAS7. These data validated longitudinal changes observed in a discovery cohort of sepsis-2 patients at presentation versus recovery (day 28-42). This approach was scaled to process a 6 mm DBS punch. Comparable numbers of proteins and PTMs were quantified from DBS versus Mitra devices. A comparison of the proteomes of newborns (n=4 controls and n=3 Pompe disease) and unaffected adult controls (n=5) showed many differentially-abundant proteins, including expected higher levels of fetal hemoglobin and lower IGM complex (IGHM, CD5L and ApoJ) in newborns.
CONCLUSION
Signatures of recovery from sepsis, reflected in the blood proteome, glycoproteome and phosphoproteome, are replicated in a cross-sectional analysis of sepsis versus healthy controls. Methods for processing and “multi-proteome” analysis of dried venous blood are easily translated to DBS. Discovery of blood-based markers of GAA enzyme activity and Pomp disease-causing variants may require a larger number of cases due to heterogeneity of disease and inter-individual blood proteomes.
|
= Discovery stage. (53.14%, 2025)
= Translation stage. (22.33%, 2025)
= Clinically available. (24.53%, 2025)
