A Common OXTR Risk Variant Alters Regulation of Gene Expression by DNA Hydroxymethylation in Pregnant Human Myometrium

Abstract

Postpartum hemorrhage, or excessive bleeding after birth, is a leading cause of maternal morbidity. A major cause of postpartum hemorrhage is uterine atony, tiring of the uterus which leads to ineffective contractions. Uterine contractions depend on oxytocin signaling in the myometrium, which in turn depends on expression of the oxytocin receptor (OXTR). Both genetic and epigenetic factors related to the oxytocin receptor are associated with risk of postpartum hemorrhage, but a mechanism relating these factors to oxytocin receptor activity in myometrium remains unclear. We report a genetic by epigenetic interaction whereby the relationship between DNA hydroxymethylation and OXTR gene expression depends on a common OXTR gene variant (rs53576). We also provide evidence that a similar genetic by epigenetic interaction using blood-derived DNA methylation is associated with relevant clinical outcomes: quantity of oxytocin administration and odds for postpartum hemorrhage. These results provide new avenues for predicting how women will respond to pharmacological agents in the prevention and treatment of postpartum hemorrhage.

Keywords: DNA hydroxymethylation; DNA methylation; Epigenetic biomarker; Myometrium; Oxytocin receptor; Postpartum hemorrhage.

Fig. 1

OXTR DNA hydroxymethylation and gene expression in pregnant myometrium are positively associated and this relationship is disrupted in rs53576 risk allele carriers. a) Schematic of OXTR indicating where CpG -934 and SNP rs53576 are in the gene. Boxes are exons, lines are introns. Coding regions are in white and untranslated regions are in gray. b) Genotype at rs53576 interacts with CpG -934 methylation to predict OXTR expression (n = 6 A carriers, 10 G/G genotype, F_(1,11) = 4.74, p = 0.052). c) Similar genetic by epigenetic interaction based on DNA methylation in blood, though it does not significantly predict gene expression (n = 6 A carriers, 11 G/G genotype, F_(1,12) = 3.07, p = 0.105) d) Genotype at rs53576 interacts with CpG -934 DNA hydroxymethylation to predict OXTR expression (n = 6 A carriers, 11 G/G genotype, F_(1,10) = 5.20, p = 0.046). e) True DNA methylation does not interact with rs53576 genotype to predict OXTR gene expression (n = 6 A carriers, 11 G/G genotype, F_(1,10) = 0.69, p = 0.504). * p < 0.05, + p = 0.052, ns p > 0.052

Fig. 2

rs53576 A carriers with high OXTR methylation in the blood require more oxytocin during labor. In rs53576 A carriers, there is a significant positive relationship between DNA methylation at OXTR CpG -934 and total units oxytocin administered during labor (n = 57, ρ(55) = 0.4, p = 0.0026). In rs53576 G/G individuals, there is no relationship between DNA methylation at OXTR CpG -934 and total units oxytocin administered during labor (n = 36, ρ(36) = 0.06, p = 0.74). Regression lines are shown for visualization purposes