Interplay between genetics and environment in obesity

Abstract: Background. Being one of the largest current global health problems, obesity is a result of interplay between genetic and environmental factors. In this thesis we studied the genetic underpinning of adipose tissue distribution; investigated causality between obesity and periodontitis using instrumental variable analyses in genetic epidemiology settings; studied interactions between obesity-associated loci and physical activity on obesity; and checked if loci identified in genome-wide association studies are good candidates for gene-environment interactions using heterogeneity of variance analyses.Methods. In Paper 1 we performed a meta-analysis of large-scale genetic association studies for waist traits in 224,459 participants followed by a variety of statistical and bioinformatics methods to obtain insights into the biology of the underlying adipose distributions. In Paper 2 we used genetic variants associated with body mass index (BMI) as instruments within a Mendelian randomization framework to study causality between obesity and periodontitis in 49,066 participants. In Paper 3 we studied the interaction between an FTO single nucleotide polymorphism (SNP) rs9939609 and physical activity in obesity outcomes through meta-analysis of 237,434 participants. In Paper 4 we evaluated if SNPs established through main-effects meta-analysis of genome-wide association studies represent strong or weak candidates for gene-environment interactions, inferred by the degree of phenotypic heterogeneity across genotypes at a given SNP locus.Results. In Paper 1 we identified 49 loci (33 novel) associated with waist-to-hip ratio adjusted for BMI (P<5×10-8), 20 of which displayed significant sexual dimorphism, of which 19 displayed a stronger effect in women than in men. We also detected 19 additional loci newly associated with related waist and hip circumference measures. The identified loci were enriched for genes expressed in adipose tissue and for putative regulatory elements in adipocytes. Pathway analyses implicated adipogenesis, angiogenesis, transcriptional regulation, and insulin resistance as processes affecting fat distribution, providing insights into potential pathophysiological mechanisms. In Paper 2 we confirmed observational associations between BMI and periodontitis with a pooled odds ratio (OR) of 1.13 per standard deviation increase in BMI (95%CI:1.03, 1.24) in all participants and 1.25 (95%CI:1.10, 1.42) in participants with clinical data. The instrumental variable meta-analysis yielded an OR of 1.05 (95%CI:0.80, 1.38) per BMI standard deviation, and 0.90 (95%CI:0.56, 1.46) in participants with clinical data. In Paper 3 we showed that physical activity attenuates the influence of FTO variation in rs9939609 on obesity (Pinteraction=0.001) with the A allele of rs9939609 increasing the odds of obesity less in the physically active group, with a per-allele OR of 1.22 (95%CI:1.19, 1.25), than in the inactive group, with a per-allele OR of 1.30 (95%CI:1.24, 1.36). In Paper 4 we show that rank-ordered distributions of P-values between marginal effects genome-wide association studies (GWAS) and heterogeneity of variance analyses for all SNPs for BMI and lipid traits have very weak correlations for most traits (Spearman rho of 0.0034 for total cholesterol and 0.0044 for BMI for the squared phenotype residuals regression method) indicating that variants with strong marginal effects are in general poor candidates for gene-environment interactions based on heterogeneity of variance analyses, although a small number of variants convey strong marginal and variance effects, such as those at the FTO locus, meaning that they are likely to convey both marginal and interaction effects.Conclusion. The work outlined in this thesis sheds light on the complexity of genetic and environmental factors in obesity. We have identified novel loci associated with waist traits and described pathways implicated in adipose distribution. We have shown that based on Mendelian randomization analyses the association between periodontitis and obesity is unlikely to be causal. We have confirmed interactions between a bona-fide obesity locus (FTO) and physical activity on obesity and have shown that, in contrast to FTO, the majority of genetic variants identified through GWAS are unlikely to be good candidates for gene-environment interactions based on heterogeneity of variance analyses.