The association between sleep quality and accelerated epigenetic aging with metabolic syndrome in Korean adults

Sleep patterns and DNA methylation age acceleration in middle-aged and older Chinese adults

BACKGROUND

Sleep is a biological necessity and fundamental to health. However, the associations of sleep patterns (integrating sleep determinants) with DNA methylation age acceleration (DNAm AA) remain unknown. We aimed to investigate the associations of sleep patterns with DNAm AA.

METHODS

This cross-sectional and prospective cohort study used data from the Dongfeng-Tongji cohort collected from 2013 to December 31, 2018. Sleep patterns were reflected by sleep scores (range 0-4, with higher scores indicating healthier sleep patterns) characterized by bedtime, sleep duration, sleep quality, and midday napping. DNAm AA was estimated by PhenoAge acceleration (PhenoAgeAccel), GrimAge acceleration (GrimAgeAccel), DunedinPACE, and DNAm mortality risk score (DNAm MS). Linear regression models were used to estimate β and 95% confidence intervals (CIs) for the cross-sectional associations between sleep patterns and DNAm AA. Mediation models were applied to assess the mediating role of DNAm AA in the associations between sleep patterns and all-cause mortality in a prospective cohort.

RESULTS

Among 3566 participants (mean age 65.5 years), 426 participants died during a mean 5.4-year follow-up. A higher sleep score was associated with lower DNAm AA in a dose-response manner. Each 1-point increase in sleep score was associated with significantly lower PhenoAgeAccel (β = - 0.208; 95% CI - 0.369 to - 0.047), GrimAgeAccel (β = - 0.107; 95% CI - 0.207 to - 0.007), DunedinPACE (β = - 0.008; 95% CI - 0.012 to - 0.004), and DNAm MS (β = - 0.019; 95% CI - 0.030 to - 0.008). Chronological age modified the associations between higher sleep scores and lower PhenoAgeAccel (p for interaction = 0.031) and DunedinPACE (p for interaction = 0.027), with stronger associations observed in older adults. Moreover, a slower DunedinPACE mediated 6.2% (95% CI 0.8% to 11.5%) of the association between a higher sleep score and a lower all-cause mortality risk.

CONCLUSION

In this cohort study, individuals with a higher sleep score had a slower DNAm AA, particularly in older adults. A slower DunedinPACE partially explained the association between higher sleep scores and lower all-cause mortality risk. These findings suggest that adopting healthy sleep patterns may promote healthy aging and further benefit premature mortality prevention, highlighting the value of sleep patterns as a potential tool for clinical management in aging.

Insomnia accelerates the epigenetic clocks in older adults

Insomnia is a common sleep disorder characterized mainly by poor sleep quality and insufficient sleep duration. It affects a significant proportion of the global population and is correlated with physical and mental consequences such as cognitive decline, anxiety, chronic fatigue, poor concentration, and memory impairment. Interestingly, it is also linked to ageing and age-related diseases (cardiovascular, metabolic, and neurodegenerative). On the other hand, as we age, DNA methylation patterns undergo significant changes. These have been used to develop the so-called epigenetic clocks that estimate the biological age linked to the environment and the risk of diseases. Few studies have evaluated the association between insomnia and epigenetic clocks, providing insight into the role of insomnia in ageing acceleration. Therefore, in the present study, we carried out an epigenetic analysis by using Illumina EPICv.2 array on 63 older adults (> 60 years old, n = 33 with insomnia vs. n = 30 control) to evaluate the relation between insomnia and epigenetic ages (HorvathAGE, HannumAGE, PhenoAGE, SkinBloodClock, GrimAGE, DunedinPACE, DNAmTL). As a result, we found an increased acceleration and correlation between GrimAGE and SkinBloodClock and a significant reduction in the DNAmTL in individuals with insomnia. An EWAS analysis showed a global pattern of hypomethylation and an enrichment of several proteostasis and oxidative pathways. In conclusion, our results suggest that insomnia increases GrimAGE and SkinBloodClock acceleration and may be participating in telomere shortening. Additionally, changes in DNA methylation patterns induced by insomnia impact proteostasis and oxidative stress.

Sleep traits causally affect epigenetic age acceleration: a Mendelian randomization study

Sleep disorders (SDs) are a common issue in the elderly. Epigenetic clocks based on DNA methylation (DNAm) are now considered highly accurate predictors of the aging process and are associated with age-related diseases. This study aimed to investigate the causal relationship between sleep traits and the epigenetic clock using Mendelian randomization (MR) analysis. The genome-wide association study (GWAS) statistics for epigenetic clocks (HannumAge, intrinsic epigenetic age acceleration [IEAA], PhenoAge, and GrimAge) and sleep traits were obtained from the UK Biobank (UKB), 23andMe and Finngen. Moreover, crucial instrumental variables (IVs) were evaluated. Inverse variance weighted (IVW), MR-Egger, weighted median (WM), weighted mode, and simple mode methods were employed to assess the causal relationship between them. Multiple analyses were performed for quality control evaluation. Our study showed that self-reported insomnia may speed up the aging process by GrimAge clock, while GrimAge acceleration could faintly reduce self-reported insomnia. Epigenetic clocks mainly influence sleep traits by PhenoAge and GrimAge with weak effects. This may indicate that early interventions of SDs could be a breaking point for aging and age-related diseases. Further studies are required to elucidate the potential mechanisms involved.