Abstract P521: Sleep Timing and Duration in Relation to DNA Methylation in Adolescents: An Epigenome-Wide Analysis

Abstract

Background: Accumulating evidence suggests that both sleep duration and sleep timing can impact metabolic health of adolescents, potentially through epigenetic modification. The objective of this study was to investigate the relationship between sleep duration and timing and DNA methylation, one type of epigenetic modification that may be responsive to changes in sleep. Methods: Cross-sectional epigenome-wide analysis of DNA methylation was conducted to identify sleep-related CpG sites in 269 females and 233 males (14.4±2.1 years on average) participating in the Early Life Exposure in Mexico to ENvironmental Toxicants (ELEMENT) cohort study. Sleep duration and midpoint on weekdays and weekends were assessed using7-day wrist actigraphy (Actigraph GTX-BT), and DNA methylation in blood leukocytes was measured using the Illumina Infinium Methylation EPIC BeadChip. Linear regression was conducted to assess the relationship between sleep variables and DNA methylation at each locus on the array (using beta values), adjusting for sex, age, maternal education, batch effects, monocytes, natural killer cells, B lymphocyte, granulocyte, and naïve CD8 T cells. A false discovery rate (FDR) using the Benjamini-Hochberg method was applied to correct for multiple testing. Sex-stratified associations were also examined. Results: The average sleep duration was 8.5±1.3 hours on weekdays and 9.1±1.3 hours on weekends, and the average sleep midpoint was 3:48AM±1.4 hours on weekdays and 4.48AM±1.3 hours on weekends. At false discovery rate <0.05, there were no significant associations at any of the 762,657 CpG sites. However, at a false discovery rate of <0.2, there were positive associations between weekday sleep duration and 4 CpG sites (i.e., longer sleep duration associated with higher % DNA methylation), including a site in or near CpG islands annotated to the gene DENND2C. Sex-stratified analyses revealed associations were more evident in males. Conclusion: Epigenome-wide analysis identified possible associations between sleep duration and multiple CpG sites, and in a sex-specific manner. Findings suggest that sleep duration may impact the epigenome in a distinct manner from sleep timing.