Evaluating the Effects of Diet-Gut Microbiota Interactions on Sleep Traits Using the UK Biobank Cohort

Previous studies showed that diet and gut microbiota had a correlation with sleep. However, the potential interaction effects of diet and gut microbiota on sleep are still unclear. The phenotypic data of insomnia (including 374,505 subjects) and sleep duration (including 372,805 subjects) were obtained from the UK Biobank cohort. The Single Nucleotide Polymorphisms (SNPs) associated with 114 gut microbiota, 84 dietary habits, and 4 dietary compositions were derived from the published Genome-wide Association Study (GWAS). We used Linkage Disequilibrium Score Regression (LDSC) to estimate the genetic correlation and colocalization analysis to assess whether dietary habits and insomnia/sleep duration shared a causal variant in a region of the genome. Using UK Biobank genotype data, the polygenetic risk score of gut microbiota, dietary habits, and dietary compositions were calculated for each subject. Logistic regression and linear regression models were used to assess the potential effects of diet-gut microbiota interactions on sleep phenotypes, including insomnia and sleep duration. Insomnia and sleep duration were used as dependent variables, and sex, age, the Townsend Deprivation Index scores, and smoking and drinking habits were selected as covariates in the regression analysis. All statistical analyses were conducted using R-3.5.1 software. Significant genetic correlations were discovered between insomnia/sleep duration and dietary habits. Further, we found several significant dietary compositions-gut microbiota interactions associated with sleep, such as fat x G_Collinsella_RNT (p = 1.843 x 10(-)(2)) and protein x G_Collinsella_HB (p = 7.11 x 10(-)(3)). Besides, multiple dietary habits-gut microbiota interactions were identified for sleep, such as overall beef intake x G_Desulfovibrio_RNT (p = 3.26 x 10(-4)), cups of coffee per day x G_Escherichia_Shigella_RNT (p = 1.14 x 10(-3)), and pieces of dried fruit per day x G_Bifidobacterium_RNT (p = 5.80 x 10(-3)). This study reported multiple diet-gut microbiota interactions associated with sleep, which may provide insights into the biological mechanisms of diet and gut microbiota affecting sleep.