Invited Commentary: Is DNA Methylation an Actionable Mediator of Prenatal Exposure Effects on Child Health?

A substantial body of literature has shown robust associations between prenatal smoking exposure and DNA methylation levels. The pattern of DNA methylation can be used as a molecular signature of past prenatal smoking exposure and might also provide mechanistic insights into associations between prenatal smoking exposure and adverse health outcomes. In this issue of the Journal, Cardenas et al. (Am J Epidemiol. 2019;188(11):1878–1886) evaluated whether DNA methylation mediates the association between prenatal smoking and low birth weight in a tissue that is mechanistically relevant to birth weight—the placenta—using formal mediation analyses. They found that methylation levels, at 5 loci, mediated smoking exposure effects on birth weight but only among children whose mothers smoked during pregnancy. Given the use of formal mediation methods and measurement in a mechanistically relevant tissue, this work has the potential to inform novel directions for intervention. Replication of these findings in larger and more racially and ethnically diverse samples, repeated measures to better tease apart the timing of DNA methylation changes with respect to exposure and birth weight, and continued use of intervention-focused mediation methods are needed before the impact of these findings will be fully realized.