Relationship of DNA methylation and cognitive and neuropsychiatric symptoms among older adults

Abstract

<h3>Introduction</h3> Accelerated biological aging is related to worse brain aging. However, the interface of Alzheimer's Disease (AD) and the biology of aging is little understood. The study of epigenetics and epigenetic aging may yield advances in prevention and treatment of AD by identifying new therapeutic targets or approaches. Further, signals from epigenetic measures may provide novel mechanistic links between neuropsychiatric symptoms (NPS) and cognitive outcomes in aging; NPS are highly prevalent in Alzheimer's Disease and related dementia (ADRD) and even appear in pre-clinical and early stages. <h3>Methods</h3> We included 45 adults aged 60+ years who are participants in a depression prevention ancillary study (VITAL-DEP) to the VITamin D and OmegA-3 TriaL, a randomized trial of fish oil and vitamin D supplements. We identified the presence of NPS using established behavioral measures. Neuropsychiatric Inventory (NPI) severity scores were mapped from dozens of item-level NPS variables extracted from mood and anxiety questionnaires and structured psychiatric diagnostic interviews in the VITAL-DEP protocol. We defined 3 groups based on cognitive status: normal cognition with or without NPS (n=20), and mild cognitive impairment (MCI) with or without NPS (n=25). Genomic DNA was extracted from leukocytes, and DNA methylation was assayed using Illumina Infinium MethylationEPIC BeadArray technology (Methyl850K chip) at baseline and year-2 follow-up. Epigenetic aging metrics [DNA methylation age (DNAmAge), AgeAccel, DNAmGrimAge, DNAmPhenoAge] were computed using the DNA epigenetic clock - Horvath method. <h3>Results</h3> The mean age of participants was 69.8 (range=60.3-83.4) years; percentages of who were females=48.9% and with post-graduate education=62.2%. In preliminary analysis, epigenetic measures were significantly correlated with chronological age (Spearman correlation (rho): for DNAmAge=0.51, p<0.001; for DNAmGrimAge=0.84, p<0.01; for DNAmPhenoAge=0.65, p<0.01). Furthermore, DNAmAge was significantly inversely correlated with NPI score (rho=-0.36, p=0.02). Participants with MCI had higher DNAmPhenoAge and DNAmGrimAge compared to those with normal cognition [median (interquartile range): for DNAmPhenoAge=56.1 vs. 54.4 years; for DNAmGrimAge=69.1 vs. 68.5 years). Additional results of longitudinal 2-year change in DNAmAge and genome-wide differential methylation analyses among participants with MCI, with or without NPS, and normal cognition will be presented. <h3>Conclusions</h3> Our preliminary results suggest that accelerated biological aging may be related to cognitive and neuropsychiatric phenotypes.