Physical activity and DNA methylation–based markers of ageing in 6208 middle-aged and older Australians: cross-sectional and longitudinal analyses

Epigenetic age quantifies biological age using DNA methylation information and is a potential pathway by which physical activity benefits general health. We aimed to assess the cross-sectional and longitudinal associations between physical activity and epigenetic age in middle-aged and older Australians. Blood DNA methylation data for 6208 participants (40% female) in the Melbourne Collaborative Cohort Study (MCCS) were available at baseline (1990-1994, mean age, 59 years) and, of those, for 1009 at follow-up (2003-2007, mean age, 69 years). Physical activity measurements (weighted scores at baseline and follow-up and total MET hours per week at follow-up) were calculated from self-reported questionnaire data. Five blood methylation-based markers of ageing (PCGrimAge, PCPhenoAge, bAge, DNAmFitAge, and DunedinPACE) and four fitness-related markers (DNAmGaitspeed, DNAmGripmax, DNAmVO2max, and DNAmFEV1) were calculated and adjusted for age. Linear regression was used to examine the cross-sectional and longitudinal associations between physical activity and epigenetic age. Effect modification by age, sex, and BMI was assessed. At baseline, a standard deviation (SD) increment in physical activity was associated with 0.03-SD (DNAmFitAge, 95%CI = 0.01, 0.06, P = 0.02) to 0.07-SD (bAge, 95%CI = 0.04, 0.09, P = 2 × 10^-8) lower epigenetic age. These associations were attenuated after adjustment for other lifestyle variables. Only weak evidence was found for the longitudinal association (N = 1009) of changes in physical activity and epigenetic age (e.g. DNAmFitAge: adjusted β = - 0.04, 95%CI = - 0.08, 0.01). The associations were not modified by age, sex, or BMI. In middle-aged and older Australians, higher levels of self-reported physical activity were associated with slightly lower epigenetic age.