Genome-wide meta-analysis of 241,258 adults accounting for smoking behaviour identifies novel loci for obesity traits

Abstract

Few genome-wide association studies (GWAS) account for environmental exposures, like smoking, potentially impacting the overall trait variance when investigating the genetic contribution to obesity-related traits. Here, we use GWAS data from 51,080 current smokers and 190,178 nonsmokers (87% European descent) to identify loci influencing BMI and central adiposity, measured as waist circumference and waist-to-hip ratio both adjusted for BMI. We identify 23 novel genetic loci, and 9 loci with convincing evidence of gene-smoking interaction (GxSMK) on obesity-related traits. We show consistent direction of effect for all identified loci and significance for 18 novel and for 5 interaction loci in an independent study sample. These loci highlight novel biological functions, including response to oxidative stress, addictive behaviour, and regulatory functions emphasizing the importance of accounting for environment in genetic analyses. Our results suggest that tobacco smoking may alter the genetic susceptibility to overall adiposity and body fat distribution.

Figure 1. Summary of study design and results.

Approach 1 uses both SNP and SMK in the association model. Approaches 2 and 3 use the SMK-stratified meta-analyses. Approach 4 screens loci based on Approach 1, then uses SMK-stratified results to identify loci with significant interaction effects (Methods section).

Figure 2. Forest plot for novel and GxSMK loci stratified by smoking status.

Estimated effects (β±95% CI) for smokers (N up to 51,080) and nonsmokers (N up to 190,178 ) per risk allele for (a) BMI, (b) WCadjBMI and (c) WHRadjBMI for novel loci from Approaches 1 and 2 (SNPadjSMK and SNPjoint, respectively) and all loci from Approaches 3 and 4 (SNPint and SNPscreen) identified in the primary meta-analyses. Loci are ordered by greater magnitude of effect in smokers compared to nonsmokers and labelled with the nearest gene. For the locus near TMEM38B, rs9409082 was used for effect estimates in this plot. (^¥loci identified for Approach 4, *loci identified for Approach 3).

Figure 3. Power comparison across Approaches.

Shown is the power to identify adjusted (Approach 1, dashed black lines), joint (Approach 2, dotted green lines) and interaction (Approach 3 and 4, solid magenta and orange lines) effects for various combinations of SMK- and NonSMK-specific effects and assuming 50,000 smokers and 180,000 nonsmokers. For (a,c,e), the effect in smokers was fixed at a small (=0.01%, similar to the realistic NUDT3 effect on BMI), medium (=0.07%, similar to the realistic BDNF effect on BMI) or large (=0.34%, similar to the realistic FTO effect on BMI) genetic effect, respectively, and varied in nonsmokers. For (b,d,f), the effect in nonsmokers was fixed to the small, medium and large BMI effects, respectively, and varied in smokers.

Figure 4. Stratum specific estimates of variance explained.

Total smoking status-specific explained variance (±s.e.) by SNPs meeting varying thresholds of overall association in Approach 1 (SNPadjSMK) and the difference between the proportion of variance explained between smokers and nonsmokers for these same sets of SNPs in BMI (a,b), WCadjBMI (c,d), and for WHRadjBMI (e,f).