A macrophage gene-regulatory network linked to clinical severity of coronary artery disease : The STARNET and...

Coronary artery disease (CAD) is a major cause of global morbidity and mortality. Macrophages play a central role in orchestrating this disease process. In 2016, we initiated the STARNET primary blood macrophage study, followed by the multi-ethnic NGS-PREDICT primary blood macrophage study in 2018. We applied integrative systems genetics analysis to explore and validate the role of macrophage gene regulatory co-expression networks (GRNs) in clinically significant CAD. This study included 318 CAD cases and 134 CAD-free controls in STARNET, and 95 CAD cases and 35 CAD-free controls in NGS-PREDICT. Primary leukocytes were isolated from blood and differentiated into macrophages in vitro, followed by RNA extraction and deep sequencing (RNAseq). In STARNET, we analyzed differentially expressed genes, inferred macrophage GRNs, assessed the phenotypic associations and functions of these GRNs, and determined their key driver genes. Integrative analysis of STARNET expression quantitative traits (eQTLs) with genotype data from genome-wide association studies was performed to determine the content of CAD candidate genes in these GRNs, and their contributions to CAD heritability. Five independent RNAseq datasets were used to retrospectively validate CAD-associated macrophage GRNs, followed by prospective validation in the NGS-PREDICT study. Using the STARNET datasets, we identified 23 macrophage GRNs. Of these, GRN_GREEN stood out as being causally associated with CAD severity (SYNTAX score) and comprised 729 genes and 90 key drivers, with the top key driver being NEIL1. GRN_GREEN accounted for 3.73% of CAD heritability and contained 34 candidate genes previously identified by GWAS of CAD. Functional analysis of GRN_GREEN revealed a large portion of genes involved in the biological process of SRP-dependent co-translational protein targeting to the membrane. GRN_GREEN replicated retrospectively in five independent human arterial wall RNAseq datasets, and prospectively in the NGS-PREDICT study. To prevent clinically significant CAD, GRN_GREEN and its top key driver NEIL1 may be suitable therapeutic targets to modify SRP-dependent co-translational targeting of proteins to the endoplasmic reticulum in macrophages.