Metabolomics profiling and pathway analysis of human plasma and urine reveal further insights into the multifactorial nature of coronary artery disease

Abstract

Background: Coronary artery disease (CAD) claims lives yearly. Nuclear magnetic resonance ( 1 H NMR) metabolomics analysis is efficient in identifying metabolic biomarkers which lend credence to diagnosis. We aimed to identify CAD metabotypes and its implicated pathways using 1 H NMR analysis. Methods: We analysed plasma and urine samples of 50 stable CAD patients and 50 healthy controls using 1 H NMR. Orthogonal partial least square discriminant analysis (OPLS-DA) followed by multivariate logistic regression (MVLR) models were developed to indicate the discriminating metabotypes. Metabolic pathway analysis was performed to identify the implicated pathways. Results: Both plasma and urine OPLS-DA models had specificity, sensitivity and accuracy of 100%, 96% and 98%, respectively. Plasma MVLR model had specificity, sensitivity, accuracy and AUROC of 92%, 86%, 89% and 0.96, respectively. The MVLR model of urine had specificity, sensitivity, accuracy and AUROC of 90%, 80%, 85% and 0.92, respectively. 35 and 12 metabolites were identified in plasma and urine metabotypes, respectively. Metabolic pathway analysis revealed that urea cycle, aminoacyl-tRNA biosynthesis and synthesis and degradation of ketone bodies pathways were significantly disturbed in plasma, while methylhistidine metabolism and galactose metabolism pathways were significantly disturbed in urine. The enrichment over representation analysis against SNPs-associated-metabolite sets library revealed that 85 SNPs were significantly enriched in plasma metabotype. Conclusions: Cardiometabolic diseases, dysbiotic gut-microbiota and genetic variabilities are largely implicated in the pathogenesis of CAD.