12th Annual IGSS Conference • October 28-29, 2021

Integrating Genetics and the Social Sciences 2021

Can DNA methylation clocks function as a surrogate endpoint for the most important risk factors for mortality?

David Rehkopf, Department of Epidemiology and Population Health, Stanford University

DNA methylation is hypothesized to be an important biological mechanism temporally situated between environmental exposures and chronic disease risk. Thus far, however, investigations of DNA methylation have taken two broad approaches, and have generally not addressed this question directly. The approach to this problem that we take here is to examine the relationship between environmental factors and DNA methylation clocks, since these clocks have a well-established relationship with a wide range of health outcomes and mortality. We thus conceptually link DNAm through aging clocks with relevant health outcomes, and focus on an empirical analysis of environmental factors and DNAm within our data. We a priori select 57 risk factors for mortality to use as the most critical environmental factors to examine, across domains of (A) adverse socioeconomic and psychosocial experiences during childhood, (B) socioeconomic conditions, (C) health behaviors, (D) social connections, (E) psychological characteristics, and (F) adverse experiences during adulthood. We examined these in relation to thirteen of the most commonly used DNAm clocks, all available in the Health and Retirement Study data. We find limited evidence for most associations between risk factors for mortality and DNAm clocks. However, for a selected number of risk factors, the Zhang and MPOA clocks emerged as being most strongly related. Our findings are thus consistent with the fact that the Zhang and MPOA clocks likely mediate some important components of environmental factors in relation to chronic disease outcomes, but that such relationships will need to be confirmed when true three time point datasets are available for a more formal mediation analysis. Our findings also suggest caution for interpreting DNAm as a mediator between risk factors and chronic disease outcomes as most associations we observe are consistent with no associations between DNAm clocks and the most important risk factors for mortality.

Presenter's website