Placental and Cord Blood DNA Methylation Changes Associated With Gestational Diabetes Mellitus in a Marginalized Population: The Untold Role of Saturated Fats

Abstract

The role of DNA methylation (DNAm) and its modulation by dietary factors in gestational diabetes mellitus (GDM) remains underexplored, particularly in marginalized populations. This study investigates DNAm alterations in GDM-exposed cord blood and placenta and their association with maternal dietary quality and single nutrient intake in a low-income population from the Myanmar-Thailand border. A matched case-control design (GDM: n = 38, controls: n = 34) was selected from a Myanmar-Thailand pregnancy cohort. Dietary intake was assessed via 24-h recalls and analyzed using Nutritionist Pro, with dietary quality evaluated by the healthy eating index (HEI). DNAm was profiled in 72 cord blood and 72 placental samples using the Infinium MethylationEPIC array. Significant differences in dietary vitamin D, total folate, and saturated fat intake were observed between the groups. RnBeads analyses revealed hypomethylation as the predominant DNAm pattern in GDM, particularly at ADORA2B (placenta) and ZFP57 (cord blood) promoters. The excessive intake of saturated fats was associated with GDM hypomethylation profiles and negatively correlated with ZFP57 methylation levels. This study highlights the influence of saturated fat intake on epigenetic changes in pregnancy, revealing potential biomarkers for GDM and emphasizing the need for tailored, population-specific nutritional interventions to mitigate transgenerational health impacts.

Keywords: DNA hypomethylation; gestational diabetes mellitus; marginalized populations; maternal diet; saturated fats.

FIGURE 1

The differences in maternal dietary quality and nutrient intake between GDM and controls. There was no significant difference in dietary quality, which was assessed by the HEI, between GDM and controls (A). Differences were only detected in (B) SF consumption, (C) vitamin D intake, and (D) total folate intake. Sample sizes were as follows: GDM group (n = 37), control group (n = 31). For non‐normally distributed data (A), statistical analysis was performed using the Mann–Whitney U‐test; center values represent the median; and error bars represent the interquartile range (IQR). For normally distributed data (B, C, and D), statistical analysis was performed using the Student's t‐test, center values represent the mean, and error bars represent the standard deviation (s.d.). Replicates are biological, each representing an independent sample from a different individual. p < 0.05 is considered significant (*), while ns indicates no significant differences.

FIGURE 2

A hypomethylation profile in placental samples of GDM. Scatter plots for (A) the top 10 000 most differentially methylated CpG sites and the top 500 most differentially methylated (B) genes, and (C) promoters between GDM (x‐axis) and controls (y‐axis). The sparsest 1% of points are explicitly plotted, with a maximum of 10 000 points displayed, while colored points indicate differentially methylated sites or regions. mean.mean.diff (GDM vs. Controls), log2 of the mean quotient in means (mean.mean.quot.log2), and the combined p value that is computed using Fisher's method are ranked for all regions, to calculate the combinedRank.

FIGURE 3

Identification of (A) the most differentially methylated (hypomethylated) promoter of ADOAR2B in placental samples of GDM, which showed (B) significantly lower β values in GDM compared to controls following Unpaired t‐test with Welch correction and Holm–Šídák method for multiple comparisons. (C) is a graphical representation of all β values of the corresponding CpG sites, which showed lower methylation levels in GDM compared to that in controls, and (D) represents the top 3 CpG sites in ADOAR2B promoter based on the highest methylation difference. Sample sizes were as follows: GDM group (n = 38), control group (n = 34). Center values represent the mean, and error bars represent the standard deviation (s.d.). Replicates are biological, each representing an independent sample from a different individual. Adjusted p value < 0.05 is considered significant (*). Chm indicates chromosome; id, promoter id; num.sites, number of sites associated with the region; Start & End, start and end coordinates of the promoter; symbol, associated gene symbol to the given promoter.

FIGURE 4

Gene ontology of the significant hypomethylated promoters in GDM specific to placental samples.

FIGURE 5

A hypomethylation profile in cord blood samples of GDM. Scatter plots for (A) the top 10 000 most differentially methylated CpG sites and the top 500 most differentially methylated (B) genes, and (C) promoters between GDM (x‐axis) and controls (y‐axis). The sparsest 1% of points are explicitly plotted, with a maximum of 10 000 points displayed, while colored points indicate differentially methylated sites or regions. mean.mean.diff (GDM vs. Controls), log2 of the mean quotient in means (mean.mean.quot.log2), and the combined p value that is computed using Fisher's method are ranked for all regions, to calculate the combinedRank.

FIGURE 6

Identification of (A) the most differentially methylated (hypomethylated) promoter of ZFP57 in cord blood samples of GDM, which showed (B) significantly lower β values in GDM compared to controls following Unpaired t‐test with Welch correction and Holm–Šídák method for multiple comparisons. (C) is a graphical representation of all β values of the corresponding CpG sites, which showed lower methylation levels in GDM compared to that in controls, and (D) represents the top 2 CpG sites in ZFP57 promoter based on the highest methylation difference. Sample sizes were as follows: GDM group (n = 38), control group (n = 34). Center values represent the mean, and error bars represent the standard deviation (s.d.). Replicates are biological, each representing an independent sample from a different individual. Adjusted p value < 0.05 is considered significant (*). Chm indicates chromosome; id, promoter id; num.sites, number of sites associated with the region; Start & End, start and end coordinates of the promoter; symbol, associated gene symbol to the given promoter.

FIGURE 7

Gene ontology of the significant hypomethylated promoters in GDM specific to cord blood samples.

FIGURE 8

SF intake shows an association with specific methylation sites in placental samples and a hypomethylation profile in high SF consumers from the GDM group. Panel A shows the negative correlation between the methylation levels (β values) of ZFP57 promoters in cord blood samples and SF intake. Spearmen r is computed, and p < 0.05 is considered significant (*), while ns indicates no significant differences. Scatter plots for (B) the top 10 000 most differentially methylated CpG sites and the top 500 most differentially methylated (C) genes, and (D) promoters between GDM (x‐axis) and controls (y‐axis). The sparsest 1% of points are explicitly plotted, with a maximum of 10 000 points displayed, while colored points indicate differentially methylated sites or regions. mean.mean.diff (GDM vs. Controls), log2 of the mean quotient in means (mean.mean.quot.log2), and the combined p value that is computed using Fisher's method are ranked for all regions, to calculate the combinedRank. (E) represents the most differentially methylated (hypomethylated) promoter (NOX5) in placental samples of high SF consumers. Sample sizes were as follows: high SF intake group (n = 19), normal SF intake group (n = 18). Chm indicates Chromosome; id, promoter id; num.sites, number of sites associated with the region; Start & End, start and end coordinates of the promoter; symbol, associated gene symbol to the given promoter.

FIGURE 9

A hypomethylation profile in cord blood samples of high SF consumers belonging to GDM group. Scatter plots for (A) the top 10 000 most differentially methylated CpG sites and the top 500 most differentially methylated (B) genes, and (C) promoters between GDM (x‐axis) and controls (y‐axis). The sparsest 1% of points are explicitly plotted, with a maximum of 10 000 points displayed, while colored points indicate differentially methylated sites or regions. mean.mean.diff (GDM vs. Controls), log2 of the mean quotient in means (mean.mean.quot.log2), and the combined p value that is computed using Fisher's method are ranked for all regions to calculate the combined rank. (D) represents the most differentially methylated (hypomethylated) promoter (ANTXR2) in cord blood samples of high SF consumers. Sample sizes were as follows: high SF intake group (n = 19), normal SF intake group (n = 18). Chm indicates Chromosome; id, promoter id; num. sites, number of sites associated with the region; Start & End, start and end coordinates of the promoter; symbol, associated gene symbol to the given promoter.