Abstract 4365633: The 5,000+ Plasma Proteome of an Atherosclerosis Mouse Model Identifies Known and Potential Disease Drivers

Abstract

Background: Mutations in the LDL receptor (LDLR) are causal for familial hypercholesterolemia, inciting the Ldlr-/- model for atherosclerosis. Despite >3,000 reports using this model, technical limitations have impeded efforts to identify candidate disease drivers in plasma. We capitalized on recent advances in mass spectrometry technologies to investigate whether dyslipidemia impacts plasma proteins implicated in human disease. Methods: Ldlr-/- mice were fed a chow or high-fat diet (HFD) for 3 (n= 27 per diet) or 6 months (n=12 per diet). Plasma samples were processed using a nanoparticle workflow (Seer), and peptides were analyzed using the Orbitrap TM Astral TM (Thermo) ( Fig.A ). Proteomes were queried against the Tabula Muris mouse single-cell and Gene Ontology (GO) databases; and queried against a GWAS list of 419 risk loci for coronary artery disease (CAD). Results: We sequenced 5,080 plasma proteins surpassing previous reports by 10-fold. The most intense proteins were the prototypical apolipoproteins whereas proteins associated with cytokine/chemokine signaling represent the previously unchartered mouse plasma proteome ( Fig.B) . To monitor sweeping changes over time, we divided the proteome into quartiles (Q1-Q4). Proteins with a sustained enrichment in HFD (n=705 remain within Q1) are indicative of liver cell subtypes (Tabula Muris, q<7.7E-0.02; Fig.C ). Whereas proteins that moved up from the lower quartiles - Q2 (n=228), Q3 (115) and Q4 (63) - indicate leukocytes and fibroblasts (Q2, q<3.5E-0.03), and endothelial cells (Q3 and Q4, q<2.9E-0.02) ( Fig.C ); demonstrating that signatures of inflammation and endothelial activation increase with disease progression. We also cross-referenced these interquartile changes to CAD risk loci and confirmed 24 genes including LPL that associate with disease progression, and another 104 genes confirmed in plasma. We focused on the 86 and 146 proteins that increased with a log2FC>1 (q<5.0E-0.02) at each timepoint ( Fig.D ). Fifty-one were common including MMP12, a candidate atherosclerosis biomarker in humans. Most of these increased proteins are mitochondrial or peroxisomal (q<5.0E-0.02), demonstrating that organelle components for fatty acid oxidation circulate and increase with disease progression. Conclusions: Mass spectrometry now permits us to identify novel biomarkers or targets in Ldlr-/- mice plasma; and to monitor candidate proteins associated with human disease in preclinical interventional therapeutic studies.