ABSTRACT: Genome-wide association studies (GWAS) have suggested that sequence variation in cis-regulatory elements (CREs) modulate common disease risk and trait variation. However, identification of the majority of these causal variants and their mechanism of action have remained significant challenges. We used reporter assays for all common variants at the QT interval associated SCN5A GWAS locus with the goal of identifying causal variants for the association. A target region of ~500kb containing SCN5A was defined based on recombination hotspots (rate>10cM/Mb; estimated from HapMap) flanking the 5 independent QT interval GWAS hits. Within this region, all common variants (minor allele frequency >5%), from the 1000 Genomes European ancestry populations, in moderate linkage disequilibrium (LD; r2>0.3) with any of the 5 sentinel hits, were selected for analyses. Of a total 121 variants selected, 112 variants in 104 amplicons passed primer design (amplicon size 256-617bp; median 397bp), with both alleles of 106 variants successfully cloned along with flanking sequences into 98 amplicons upstream of a minimal promoter-driven firefly luciferase gene in pGL4.23. Cardiomyocyte cells, AC16 (human) and HL1 (mouse), were transfected with test constructs and Renilla luciferase vector (for transfection normalization) in triplicate; luciferase assays were performed 24h later. All cloning and reporter assays were performed in 96- and 24-well plates. In reporter assays across the two cell lines, compared to empty vector, 37 amplicons showed enhancer activity (z-score>99 %tile), with a concordance rate of ~70%. Of these 37 enhancer CREs, 9 overlapped open chromatin regions (DNase-seq peaks) observed in adult human heart tissue, largest among all human tissues evaluated by the RoadMap Epigenomics project. Of these 37 enhancer CREs, 12 showed allelic difference in reporter activity (P<0.05), thus identifying at least one enhancer CRE variant in high-to-moderate LD with each of the 5 sentinel hits. Using GTEx heart left ventricle (n=190) gene expression data, we showed correlation between SCN5A expression and the number of QT interval prolonging alleles across the 5 index SNPs. We are currently assessing the binding of cardiac nuclear factors at the 12 enhancer CRE variants by gel-shift assays and the effect of CRISPR-Cas9 mediated CRE deletion on SCN5A expression in AC16 and HL1 cells, an analyses helped by evaluation of AC16 and HL1 cells by RNA-seq, ATAC-seq and karyotyping. Thus, independent of the publicly available epigenomic data, which are of limited cell-type relevance, an unbiased in vitro reporter screen for CREs overlapping all common variants associated with QT interval at the SCN5A GWAS locus identified 12 common cis-regulatory variants that map to cardiac open chromatin regions and correlate with SCN5A cardiac expression.
