hide
Free keywords:
-
Abstract:
DNA methylation variation at HP / BP3 and TTC9B is modified by estrogen
exposure in the rodent hippocampus and was previously shown to be
prospectively predictive of postpartum depression (PPD) when modeled in
antenatal blood. The objective of this study was to replicate the
predictive efficacy of the previously established model in women with
and without a previous psychiatric diagnosis and to understand the
effects of changing hormone levels on PPD biomarker loci. Using a
statistical model trained on DNA methylation data from N = 51 high-risk
women, we prospectively predicted PPD status in an independent N = 51
women using first trimester antenatal gene expression levels of HP I BP3
and TTC9B, with an area under the receiver operator characteristic curve
(AUC) of 0.81 (95% CI: 0.69-0.92, p<5 x 10(-4)). Modeling DNA
methylation of these genes in N = 240 women without a previous
psychiatric diagnosis resulted in a cross-sectional prediction of PPD
status with an AUC of 0.81 (95% CI: 0.68-0.93, p = 0.01). TTC9B and HP I
BP3 DNA methylation at early antenatal time points showed moderate
evidence for association to the change in estradiol and allopregnanolone
over the course of pregnancy, suggesting that epigenetic variation at
these loci may be important for mediating hormonal sensitivity. In
addition both loci showed PPD-specific trajectories with age, possibly
mediated by age-associated hormonal changes. The data add to the growing
body of evidence suggesting that PPD is mediated by differential gene
expression and epigenetic sensitivity to pregnancy hormones and that
modeling proxies of this sensitivity enable accurate prediction of PPD.