Prenatal maternal depressive symptoms and infant DNA methylation: a longitudinal epigenome-wide study

Abstract

Background: Prenatal maternal stress increases the risk of offspring developmental and psychological difficulties. The biological mechanisms behind these associations are mostly unknown. One explanation suggests that exposure of the fetus to maternal stress may influence DNA methylation. However, this hypothesis is largely based on animal studies, and human studies of candidate genes from single timepoints.

Aim: The aim of this study was to investigate if prenatal maternal stress, in the form of maternal depressive symptoms, was associated with variation in genome-wide DNA methylation at two timepoints.

Methods: One-hundred and eighty-four mother-child dyads were selected from a population of pregnant women in the Little-in-Norway study. The Edinburgh Postnatal Depression Scale (EPDS) measured maternal depressive symptoms. It was completed by the pregnant mothers between weeks 17 and 32 of gestation. DNA was obtained from infant saliva cells at two timepoints (age 6 weeks and 12 months). DNA methylation was measured in 274 samples from 6 weeks (n = 146) and 12 months (n = 128) using the Illumina Infinium HumanMethylation 450 BeadChip. Linear regression analyses of prenatal maternal depressive symptoms and infant methylation were performed at 6 weeks and 12 months separately, and for both timepoints together using a mixed model.

Results: The analyses revealed no significant genome-wide association between maternal depressive symptoms and infant DNA methylation in the separate analyses and for both timepoints together.

Conclusions: This sample of pregnant women and their infants living in Norway did not reveal associations between maternal depressive symptoms and infant DNA methylation.

Keywords:

• Prenatal maternal depression
• maternal depressive symptoms
• early life stress
• infant outcome
• DNA methylation

Acknowledgements

We would like to thank the participants for their contribution to the study, Martin Hammerø for his work in the lab and Roald Skoelv for proofreading.

Disclosure statement

No potential conflict of interest was reported by the authors.

Additional information

Funding

This work is supported by Norges Forskningsråd.