Enhanced polygenic risk score incorporating gene–environment interaction suggests the association of major depressive disorder with cardiac and lung function

Abstract

Background: Depression has been linked to an increased risk of cardiovascular and respiratory diseases; however, its impact on cardiac and lung function remains unclear, especially when accounting for potential gene–environment interactions. Methods: We developed a novel polygenic and gene–environment interaction risk score (PGIRS) integrating the major genetic effect and gene–environment interaction effect of depression-associated loci. The single nucleotide polymorphisms (SNPs) demonstrating major genetic effect or environmental interaction effect were obtained from genome-wide SNP association and SNP-environment interaction analyses of depression. We then calculated the depression PGIRS for non-depressed individuals, using smoking and alcohol consumption as environmental factors. Using linear regression analysis, we assessed the associations of PGIRS and conventional polygenic risk score (PRS) with lung function (N= 42 886) and cardiac function (N= 1791) in the subjects with or without exposing to smoking and alcohol drinking. Results: We detected significant associations of depression PGIRS with cardiac and lung function, contrary to conventional depression PRS. Among smokers, forced vital capacity exhibited a negative association with PGIRS (β = −0.037, FDR = 1.00 × 10−8), contrasting with no significant association with PRS (β = −0.002, FDR = 0.943). In drinkers, we observed a positive association between cardiac index with PGIRS (β = 0.088, FDR = 0.010), whereas no such association was found with PRS (β = 0.040, FDR = 0.265). Notably, in individuals who both smoked and drank, forced expiratory volume in 1-second demonstrated a negative association with PGIRS (β = −0.042, FDR = 6.30 × 10−9), but not with PRS (β = −0.003, FDR = 0.857). Conclusions: Our findings underscore the profound impact of depression on cardiac and lung function, highlighting the enhanced efficacy of considering gene–environment interactions in PRS-based studies.