Many heritable cardiovascular disease (CVD) risk traits tend to be correlated suggesting the underlying presence of pleiotropic effect genes. Genome-wide association studies (GWAS) aimed at composite CVD traits may reveal such genes, which would remain latent using single phenotype analyses. Analysis of large pedigrees offers the added advantage of assessing the heritability of composite traits, which can help prioritise genetically influenced phenotypes for GWAS analysis.
In this study we performed a principal components analysis (PCA) and subsequent heritability (H2) estimation of n=37 CVD-related phenotypes in 330 related individuals forming a large pedigree from the Norfolk Island genetic isolate (a phenome scan). We then performed a pedigree-based GWAS using Illumina 610K quad chips to search for pleiotropic effect loci.
PCA revealed 13 components explaining >75% of the total variance within the sample space. Nine components yielded statistically significant H2 values ranging from 0.22 to 0.54 (P<0.05). The most heritable component was loaded with 7 phenotypic measures including; % body fat, waist-to-hip ratio, systolic and diastolic blood pressure, creatine, urea, uric acid. A GWAS of this composite phenotype revealed statistically significant associations for 3 adjacent SNPs on chr. 1p22 (P<1*10-8). These SNPs form a 42kb haplotype block and explain 11% of the genetic variance indicating a major susceptibility locus for this phenotype. No associations were observed for the phenotypes assessed individually. Nor were any associations observed for the other 6 heritable component traits.
Our results support the ‘phenome scanning’ approach to search for pleiotropic effect loci associated with correlated CVD phenotypes. Further research is now underway to elucidate the causative gene variant(s) within the susceptibility region and explain the functional impact on this composite CVD phenotype.