Absence of genomic hypomethylation or regulation of cytosine-modifying enzymes with aging in male and female mice

Niran Hadad, Dustin R. Masser, Sreemathi Logan, Benjamin Wronowski, Colleen A. Mangold, Nicholas Clark, Laura Otalora, Archana Unnikrishnan, Matthew M. Ford, Cory B. Giles, Jonathan D. Wren, Arlan Richardson, William E. Sonntag, David R. Stanford, Willard Freeman

Changes to the epigenome with aging, and DNA modifications in particular, have been proposed as a central regulator of the aging process, a predictor of mortality, and a contributor to the pathogenesis of age-related diseases. In the central nervous system, control of learning and memory, neurogenesis, and plasticity require changes in cytosine methylation and hydroxymethylation. Although genome-wide decreases in methylation with aging are often reported as scientific dogma, primary research reports describe decreases, increases, or lack of change in methylation and hydroxymethylation and their principle regulators, DNA methyltransferases and ten-eleven translocation dioxygenases in the hippocampus. Furthermore, existing data are limited to only male animals. Through examination of the hippocampus in young, adult, and old male and female mice by antibody-based, pyrosequencing, and whole-genome oxidative bisulfite sequencing methods, we provide compelling evidence that contradicts the genomic hypomethylation theory of aging. We also demonstrate that expression of DNA methyltransferases and ten-eleven translocation dioxygenases is not differentially regulated with aging or between the sexes, including the proposed cognitive aging regulator DNMT3a2. Using oxidative bisulfite sequencing that discriminates methylation from hydroxymethylation and by cytosine (CG and non-CG) context, we observe sex differences in average CG methylation and hydroxymethylation of the X chromosome, and small age-related differences in hydroxymethylation of CG island shores and shelves, and methylation of promoter regions. These findings clarify a long-standing misconception of the epigenomic response to aging and demonstrate the need for studies of base-specific methylation and hydroxymethylation with aging in both sexes.