Genome-Wide DNA Methylation in Pregnancy – Preliminary Results From the EARLI Study

Background:  Several lines of evidence point to a potential role for epigenetic mechanisms in autism etiology. Changes in DNA methylation (DNAm) during development or early life may be most relevant for ASDs. Thus, looking at DNAm in mothers during pregnancy and in their babies in early childhood provides an opportunity for understanding the role of epigenetics in ASD during critical windows of development. Further, DNAm is affected by environmental factors and therefore could be a biological mechanism linking environments during pregnancy and early life and risk for ASD.

Objectives:  As foundational pilot work towards understanding the role of epigenetics in autism, we have measured DNAm across the genome in blood samples from EARLI mothers taken at multiple pregnancy visits to assess changes in DNAm over the pregnancy interval and correlations between DNAm and maternal characteristics that may related to ASD risk.

Methods:  We performed genome-wide DNAm analysis on blood samples of 46 pregnant women from the EARLI study using CHARM 2.0, an array-based genome-wide approach containing over 4 million probes. Women contributed between 2 and 4 separate samples for DNAm measurements corresponding to the first, second, and third trimesters of pregnancy and 3 months post-pregnancy. We first searched for regions where DNAm is changing over time in the mothers, then looked for regions of the genome that are differentially methylated between mothers with particular exposures such as prenatal vitamin use or potential ASD risk factors such as auto-immune disorders.

Results:  We do not see large changes in DNAm within mothers over the pregnancy interval in this pilot sample. We do observe differences in DNAm between mothers with and without particular characteristics such as presence of autoimmune disorders. Results of comparisons with a larger array of factors will be presented.

Conclusions:  This work has allowed us to develop the laboratory pipeline to analyze DNAm in epidemiologic samples. It appears that DNAm marks in blood are stable throughout pregnancy, which has implications for interpretation of results relating DNAm with potential ASD risk factors and outcomes. We have developed a strategy for identifying differentially methylated regions related to risk factors and outcomes.  This strategy is being applied across a spectrum of variables with the goal of identifying epigenetic marks that may relate to environmental risk factors and thus elucidate mechanisms by which these risk factors influence ASD risk.