Sex dependent genetic architecture of biochemically verified tobacco use

Tobacco use differs by genetics and sex, and dose-dependently increases the risk for numerous diseases. Nicotine is metabolized to cotinine (COT) which is further metabolized to 3'hydroxycotinine (3HC). COT and COT+3HC are biomarkers which capture tobacco intake more accurately than self-reported measures such as cigarettes/day. It is currently not known whether genetic risk factors for heavier tobacco intake, measured using these biomarkers, differ by sex.

We conducted a genome-wide genotype-by-sex (GxS) interaction analysis of COT and COT+3HC measured from blood in European treatment-seeking smokers (n = 541 males, n = 389 females) (NCT01314001). Linear regression models included Genotypes (coded additively), Sex, a GxS interaction term, covariates, and all covariate-by-genotype and covariate-by-sex interaction terms.

For COT, five suggestive (P < 5 × 10^-6) loci on chr 4, 15, 19, 12, and 1 were identified; the top variant was rs11520555 (5' of SPOCK3; beta=0.38, se=0.08, GxS P = 7.39 × 10^-7). For COT+3HC, eight suggestive loci on chr 21, 18, 17 (2 loci), 13, 5, 8, and 19 were identified; the top variant was rs73157714 (3' of HSPA13; beta=0.33, se=0.06, GxS P = 3.48 × 10^-7). Overall, 26 genes were mapped, with 9 showing moderate to high expression in brain, and 5 showing prior associations with psychiatric traits in the GWAS Catalog.

Our findings suggest that the genetic architecture of tobacco intake, measured accurately using biomarkers, differs between women and men. A more granular understanding of factors influencing tobacco intake in women versus men may identify risk factors for heavier use and sex-specific opportunities to promote smoking cessation and mitigate disease risk.

This genome-wide interaction study suggested that some of the genetic influences on tobacco intake, measured accurately using biomarkers, differ by sex. The loci identified in our study could be a starting point for developing new genetic biomarkers that predict sex-specific differences in tobacco intake and disease risk.