Uncoupling associations of risk alleles with endophenotypes and phenotypes: insights from the ApoB locus and heart-related traits.

Abstract

Traditionally, genomewide association studies (GWAS) have emphasized the benefits of large samples in the analyses of age-related traits rather than their specific properties. We adopted a realistic concept of genetic susceptibility to inherently heterogeneous, age-related traits driven by the elusive role of evolution in their properties. We analyzed in detail the associations of rs693 and rs562338 polymorphisms representing the Apolipoprotein B locus with endophenotypes (total cholesterol [TC] and high-density lipoprotein cholesterol) and phenotypes (myocardial infarction [MI] and survival) in four large-scale studies, which include 20 748 individuals with 2357 MI events. We showed that a strong, robust predisposition of rs693 and rs562338 to TC (β = 0.72, P = 7.7 × 10(-30) for rs693 and β = -1.08, P = 9.8 × 10(-42) for rs562338) is not translated into a predisposition to MI and survival. The rs693_A allele influences risks of MI and mortality after MI additively with lipids. This allele shows antagonistic effects-protecting against MI risks (β = -0.18, P = 1.1 × 10(-5) ) or increasing MI risks (β = 0.15, P = 2.8 × 10(-3) ) and mortality after MI, in different populations. Paradoxically, increased TC concentrations can be protective against MI for the rs693_A allele carriers. Our results uncouple the influences of the same alleles on endophenotypes and phenotypes despite potential causal relationships among the latter. Our strategy reveals virtually genomewide significance for the associations of rs693 with MI (P = 5.5 × 10(-8) ) that is contrasted with a weak estimate following the traditional, sample-size-centered GWAS strategy (P = 0.16) in the same sample. These results caution against the use of the traditional GWAS strategy for gaining profound insights into genetic predisposition to healthspan and lifespan.