Maternal dietary glycaemic change during gestation influences insulin-related gene methylation in the placental tissue: a genome-wide methylation analysis.

Background: Studies have shown that the effects of maternal nutrition exposure during gestation influence metabolic risk in early life through an epigenetic mechanism. Low glycaemic index (GI) diets benefit both maternal and neonatal gestational outcomes. We hypothesize that maternal dietary GI or glycaemic load (GL) changes during pregnancy impact placental DNA methylation, especially in insulin resistance-related genes.

Methods: From a clinical trial of overweight pregnant women, 12 subjects who successfully reduced their GI and another 12 whose GI increased despite the intervention were selected. A genome-wide differential methylation analysis of placental tissue DNA was conducted, followed by bioinformatic annotation and validation analysis. The distribution of genome-wide differentially methylated regions (DMRs) and CpG sites was described. Six CpG sites in regulatory regions of four insulin-related genes (PLIN1, CPT1B, SSTR4, and CIDEA) were selectively validated by pyrosequencing. Pairwise Spearman correlation analysis was performed to test methylation-phenotype association in an additional 153 subjects from the same trial. Correlation between methylation of significant sites and placental mRNA expression of SSTR4 was also analysed.

Results: Dietary GI decreased by 24.3 (26.2-20.1) in the group who responded appropriately to the intervention and increased by 19.6 (15.2-29.1) in the comparison group. Epigenome-wide analysis identified 108 DMRs and 365 CpG sites with P < 0.05 adjusted by false discovery rate, distributed over all chromosomes. The methylation level of cg05009389 in the 3' UTR of PLIN1 was negatively correlated with maternal weight gain (ρ = - 0.21, P = 0.027) and increase in insulin levels (ρ = - 0.24, P = 0.015) during gestation. Methylation levels of cg17586860 and cg18197392 in the 5' UTR region of SSTR4 were negatively correlated with changes in dietary carbohydrate intake (ρ = - 0.24, Ps ≤ 0.006) and GL across gestation (ρ = - 0.23, Ps ≤ .008). This correlation survived the adjustment for maternal factors such as dietary GI, body mass index, and gestational diabetes. Up to 89% of cg18197392 methylation was explained by GL change. Cg14631053 methylation correlated positively with mRNA expression of SSTR4 in the placenta (ρ = 0.20, P = 0.037).

Conclusions: We provide the first evidence that maternal dietary GI changes during gestation may impact placental DNA methylation of insulin regulation genes. This supports the hypothesis that placental methylation may be the epigenetic mechanism through which maternal diet influences the metabolic health of offspring.