From aging to Alzheimer's disease: concordant brain DNA methylation changes in late life

Abstract

Aging is the strongest risk factor for Alzheimer's disease (AD), yet the molecular mechanisms linking aging to AD remain poorly understood. DNA methylation (DNAm) is an epigenetic modification that plays a critical role in gene regulation and has been implicated in both aging and AD. In this study, we performed a meta analysis of DNAm profiles in the prefrontal cortex using two large, independent postmortem brain cohorts, the Religious Orders Study and Memory and Aging Project (ROSMAP) and Brains for Dementia Research (BDR), to identify DNAm differences associated with aging in late life. We identified 3264 CpGs significantly associated with aging, the majority of which were hypermethylated and enriched in promoter regions and CpG islands. These aging associated DNAm changes were significantly overrepresented in genes involved in immune regulation and metabolic pathways. When compared with AD associated DNAm changes, we found a significant overlap, with nearly all CpGs and differentially methylated regions (DMRs) that were associated with both aging and AD Braak stage displaying concordant directionality. This supports the hypothesis that aging and AD are interconnected at the molecular level. Further integrative analyses indicated that a number of these DNAm variants may have functional relevance in AD. By integrating blood DNAm data, we identified multiple CpGs that showed significant brain to blood correlations and were involved in both aging and AD pathogenesis. Colocalization analyses with genomewide association study (GWAS) data revealed shared genetic regulation of DNAm and dementia at several AD risk loci. Out of sample validation using the Alzheimer's Disease Neuroimaging Initiative (ADNI) dataset demonstrated that, among 334 CpGs showing concordant DNAm changes in aging and AD, baseline DNAm levels at cg10752406 in the AZU1 promoter were significantly associated with AD progression at a 5% false discovery rate, even after adjusting for age, sex, APOE {varepsilon}4 allele status, years of education, and baseline MMSE. Notably, this CpG also showed a strong brain blood DNAm correlation, further supporting its potential as a peripheral biomarker for AD. Our study provides valuable insights into the epigenetic landscape of aging and its implications for AD, suggesting that aging related epigenetic modifications may provide a viable source of biomarkers for AD.