The Combined Effects of Bisphenol A Exposure, Diet, and Physical Activity on the Aging Epigenome

Abstract

Increasing evidence supports the developmental origins of health and disease (DOHaD) hypothesis, which posits that exposure to environmental factors (e.g. diet, chemicals, stress etc.) during sensitive periods of life (e.g. pre-conception, gestation, infancy, adolescence) alters disease susceptibility later in life by influencing developmental plasticity. As support for DOHaD accumulates, it has been proposed that developmental exposures alter later-life gene regulation and subsequent phenotype through changes in heritable epigenetic marks – e.g. DNA methylation. Both biological aging and environmental exposures are associated with changes in DNA methylation, and it has been shown that developmental exposures can alter the rate of epigenetic aging. Based on these existing data, we defined a new term – environmental deflection – that refers to an environment- or toxicant-mediated shift away from the baseline rate of epigenetic aging within an organism. For this project, longitudinal animal model and human cohort studies were used to investigate whether developmental exposure to specific environmental factors – bisphenol A (BPA), Western high-fat diet (WHFD), and physical activity – would lead to environmental deflection of the aging epigenome. In the animal model study, matched blood and tail samples were collected from congenic a/a Agouti mice perinatally exposed to BPA (50 µg/kg diet) and/or WHFD. Linear mixed effects models were used to test for environmental deflection of epigenetic aging by dietary exposures. In the same mice, we used two next-generation sequencing methods – enhanced reduced representation bisulfite sequencing (ERRBS) and hydroxymethylated DNA immunoprecipitation sequencing (HMeDIP-seq) – to determine the contributions of 5-methylcytosine (5-mC) and 5-hydroxymethylcytosine (5-hmC) to the aging epigenome. In the Healthy Families Project, a human cohort, we investigated the effects of physical activity and diet quality on age-related methylation in longitudinal blood samples. DNA methylation was measured at obesity-related genes in matched neonatal bloodspot and childhood blood samples (12-24 months old, 3-5 years old, 10-12 years old). In an effort to test the utility of neonatal blood DNA methylation as a biomarker of obesity risk during childhood, we also investigated whether childhood obesity likelihood was associated with neonatal and/or childhood DNA methylation at a number of obesity-related genes. In both the mouse and human studies, we showed significant, gene-specific age-related DNA methylation. In mice, WHFD, but not BPA exposure, deflected age-related Esr1 methylation rates away from Control baseline. In the mouse blood sequencing data, we showed a locus-specific contribution of 5-hmC to age-related DNA methylation patterns in mice, and also demonstrated significant effects of BPA exposure on DNA hydroxymethylation in the gene bodies of imprinted loci. In the human cohort, environmental deflection modeling was limited by sample size, but there was some indication of deflection by childhood BMI z-score and physical activity levels. Neonatal bloodspot LINE-1 DNA methylation was significantly associated with obesity likelihood in preschool children, and childhood PPARA was also negatively associated with body mass index z-score. In this dissertation, we showed that both altered diet and physical activity have the potential to alter rates of epigenetic aging. Separately, we found that developmental BPA exposure stably alters DNA hydroxymethylation at murine imprinted genes, and that human neonatal bloodspot DNA methylation may be a useful biomarker for estimating childhood obesity risk. These results emphasize the importance of longitudinal study design in toxicoepigenetics research, and suggest that environmental factors play a key role in the developmental origins of adult disease.