Genetic determinants of daytime napping and effects on cardiometabolic health

Abstract

Daytime napping is a common, heritable behavior, but its genetic basis and causal relationship with cardiometabolic health remain unclear. Here, we perform a genome-wide association study of self-reported daytime napping in the UK Biobank (n = 452,633) and identify 123 loci of which 61 replicate in the 23andMe research cohort (n = 541,333). Findings include missense variants in established drug targets for sleep disorders (HCRTR1, HCRTR2), genes with roles in arousal (TRPC6, PNOC), and genes suggesting an obesity-hypersomnolence pathway (PNOC, PATJ). Association signals are concordant with accelerometer-measured daytime inactivity duration and 33 loci colocalize with loci for other sleep phenotypes. Cluster analysis identifies three distinct clusters of nap-promoting mechanisms with heterogeneous associations with cardiometabolic outcomes. Mendelian randomization shows potential causal links between more frequent daytime napping and higher blood pressure and waist circumference.

Fig. 1. Plots for genome-wide association analysis results for daytime napping in the UK Biobank (n = 452,633) and replication in 23andMe (n = 541,333).

A Manhattan plot of daytime napping genome-wide association study in the UK Biobank (n = 452,633). Plot shows the −log₁₀P values (y-axis) for all genotyped and imputed single-nucleotide polymorphisms (SNPs) passing quality control (BOLT-LMM mixed-model association test P values) ordered by chromosome and base position (x-axis). Blue peaks represent genome-wide significant loci. Horizontal red line denotes genome-wide significance (P = 5 × 10⁻⁸). Top 8 loci are annotated with nearest gene. B Daytime napping signals’ effect estimates from UK Biobank (n = 452,633) plotted against effect estimates from 23andMe (n = 541,333). Error bars represent the 95% confidence intervals for each effect estimate. C Effect estimates of daytime napping signals from UK Biobank and 23andMe meta-analysis (total n = 993,966) plotted against minor allele frequency.

Fig. 2. Colocalization analysis reveals a shared causal variant reducing FADS1 gene expression in the frontal cortex and increasing napping liability, and a shared causal missense variant in HCRTR2 influencing daytime napping, chronotype, and ease of awakening.

A Regional association plots for daytime napping and FADS1 gene expression in the frontal cortex at rs174561 and variants within 400 kb on chromosome 11. The y-axis shows the −log₁₀ P value for each variant in the region, and the x-axis shows the genomic position. Each variant is represented by a filled circle, with the rs174561 variant colored purple, and nearby variants colored according to degree of linkage disequilibrium (r²) with rs174561. The lower panel shows genes located in the displayed region and the blue line corresponds to the recombination rate. B Forest plot of associations between the C allele of genetic variant rs174561 in FADS1 with daytime napping and gene expression of FADS1 in the frontal cortex. Units of daytime napping reflect an increase on the ordinal scale of the trait, and gene expression is in standard deviation units. P values are two-sided and were obtained using linear regression. Black box indicates the effect estimate and lines represent 95% confidence intervals. C Regional association plot for colocalized sleep phenotypes at rs2653349 and variants within 400 kb on chromosome 6. D Crystal structure of HCRTR2 (PDB ID 6TPJ) showing localization of rs2653349 that changes Isoleucine to Phenylalanine or to valine at the transmembrane domain of HCRTR2. Protein sequence was visualized using iCn3D (https://www.ncbi.nlm.nih.gov/Structure/icn3d/full.html). The variant rs2653349 was aligned with the sequence (arrows to Human Missense variant in Figure) and the previously published canine HCRTR2 mutations, which disrupt transmembrane and signaling domains or truncate the HCRTR2 protein are highlighted in cyan. E Forest plot of associations between the A allele of genetic variant rs2653349 in HCRTR2 and the colocalized sleep phenotypes. P values are two-sided and were obtained using linear regression. Black boxes show effect estimates, and surrounding lines display 95% confidence intervals.

Fig. 3. Tissue expression, single-cell, and pathway-based enrichment analyses for daytime napping.

A MAGMA tissue expression analysis using gene expression per tissue based on GTEx RNA-seq data for 53 specific tissue types. Significant tissues (P < 9.43 × 10⁻⁴) are shown in red. B Significant single-cell types from single-cell enrichment analyses using human brain datasets in FUMA. C Top pathways determined from analysis using MAGMA gene sets and Pascal (gene-set enrichment analysis using 1077 pathways from KEGG, REACTOME, BIOCARTA). Significant pathways are shown in red (P_adj < 0.05). All pathway and tissue expression analyses in this figure can be found in tabular form in Supplementary Table 10, Supplementary Data 8, 9.

Fig. 4. Genome-wide genetic architecture of daytime napping correlations and associations with diseases and traits.

A Shared genetic architecture between daytime napping and cardiometabolic diseases and traits. Linkage disequilibrium (LD) score regression estimates of genetic correlation (r_g) were obtained by comparing genome-wide association estimates for daytime napping (without and with BMI adjustment) with summary statistics estimates from 257 publicly available genome-wide association studies. Blue indicates positive genetic correlation and red indicates negative genetic correlation; r_g values are displayed for significant correlations. Larger colored squares correspond to more significant P values. Asterisk denotes significant false discovery rate (FDR) corrected P values. Full genetic correlations for all 257 traits can be found in Supplementary Data 10. B Manhattan plot of phenome-wide association findings for daytime napping genome-wide polygenic score in Mass General Brigham Biobank (n = 23,561). The x-axis is color-coded phecodes organized by broad disease categories and the y-axis is P value of association (−log₁₀ P). The horizontal red line depicts phenome-wide significance using Bonferroni correction for all tested diseases (951 diseases), and the horizontal blue line depicts phenome-wide significance using FDR correction. Upward arrows denote positive associations (OR > 1), and downward arrows denote inverse associations (OR < 1). Full results for all 951 diseases can be found in Supplementary Data 11. C Cross-sectional association between quartile 10 and quartile 1 (reference group) of daytime napping genome-wide polygenic score and essential hypertension, obesity, and chronic nonalcoholic liver disease in the Mass General Brigham Biobank (n = 23,561). Error bars represent the 95% confidence intervals for association.

Fig. 5. Mendelian randomization supports a causal effect of daytime napping on higher blood pressure and waist circumference.

The MR estimates were calculated using the random-effects inverse-variance weighted method and represent the effect of a one-unit increase in napping category (never, sometimes, usually). Sample sizes reflect either the total sample size (for continuous outcomes) or number of cases and controls (for binary outcomes). A IVW effect estimates for more frequent daytime napping on cardiometabolic outcomes and risk factors. A unit increase in the adiposity and blood pressure measurement represents a standard deviation increase in the corresponding trait. Black boxes show effect estimates, and surrounding lines display 95% confidence intervals. All P values are two-sided. B IVW effect estimates for the effect of adiposity traits on daytime napping frequency. Black boxes show effect estimates, and surrounding lines display 95% confidence intervals. All P values are two-sided. * significant at Bonferroni-corrected alpha threshold and robust in sensitivity analyses. BMI body-mass index, CAD coronary artery disease, CI confidence interval, DBP diastolic blood pressure, HOMA homeostatic model assessment of insulin resistance, HOMAB homeostasis model assessment of β-cell function, LDL low-density lipoprotein, HDL high-density lipoprotein, OR odds ratio, SBP systolic blood pressure, SNP single-nucleotide polymorphism, T2DM type 2 diabetes mellitus, WC waist circumference, WHRadjBMI waist-to-hip ratio adjusted for BMI.

Fig. 6. Cardiovascular risk factor and disease associations of missense variants in HCRTR1 (rs2271933) and HCRTR2 (rs2653349), which encode targets of Suvorexant, an FDA-approved sleep medication with an unknown cardiovascular safety profile.

Sample size either reflects the total number of subjects (for continuous traits), or the number of cases and controls (for binary traits) that were included in each of the genome-wide association studies. All associations are oriented to the napping-increasing allele of the variants. Additional details regarding the included studies are provided in Supplementary Table 13 and Supplementary Data 13. Black boxes show Mendelian randaomization effect estimates and surrounding lines display 95% confidence intervals. BMI body-mass index, CI confidence interval, CVD cardiovascular disease, HDL high-density lipoprotein cholesterol, LDL low-density lipoprotein cholesterol, OR odds ratio, WMH white matter hyperintensities, WHR waist-to-hip ratio.