Associations of polygenic inheritance of physical activity with aerobic fitness, cardiometabolic risk factors and diseases: the HUNT Study

Abstract

ABSTRACTBackgroundPhysical activity (PA), aerobic fitness, and cardiometabolic diseases (CMD) are highly heritable multifactorial phenotypes. Shared genetic factors may underlie the associations between higher levels of PA and better aerobic fitness and a lower risk for cardiometabolic CMDs. Our aim was first to validate PA genotype against self-reported leisure-time PA (LTPA) and second to study how PA genotype associates with aerobic fitness and CMD risk factors. Finally, we investigated if PA genotype predicts CMD endpoints. We expected that these analyses would provide evidence for pleiotropic effects (same gene variants explaining both phenotypes) of PA and CMDs.Methods and findingsPolygenic risk score (PRS) for PA was constructed in the Trøndelag Health Study (N=47,148) using UK Biobank single nucleotide polymorphism-specific weights (N=400,124). The results showed that genotypes predisposing to higher PA were associated with greater self-reported PA (Beta [B]=0.282 MET-h·wk-1per SD of PRS for PA, 95% confidence interval [CI]=0.211, 0.354) but not with aerobic fitness. These genotypes were also associated with healthier profile in CMD risk factors (waist circumference [B=-0.003 cm, 95% CI=-0.004, -0.002], body mass index [B=-0.002 kg·(m2)-1, 95% CI=-0.004, -0.001], high-density lipoprotein cholesterol [B=0.004 mmol·L-1, 95% CI=0.002, 0.006]) and lower incidence of hypertensive diseases (HR=0.97, 95% CI=0.951, 0.990), stroke (HR=0.94, 95% CI=0.903, 0.978) and type 2 diabetes (HR=0.94, 95% CI=0.902, 0.970). When accounting for self-reported PA, the associations between PA genotypes and CMD risk factors remained statistically significant.ConclusionsGenotypes predisposing to higher PA were associated with higher amount of PA. However, in general the predictive value of the PRS for PA in predicting self-reported PA was low, possibly because of the acknowledged inconsistencies in assessing PA in cohort studies. Genotypes predisposing to higher PA were also associated with better cardiometabolic health and lower incidence of CMDs, and the observed associations were independent of self-reported PA. These results support earlier findings suggesting small pleiotropic effects between PA and CMDs and provide new evidence about associations with polygenic inheritance of PA and intermediate CMD risk factors. The observed association were however small, and results do not suggest clinical relevance in health promotion.