In utero Bisphenol A Exposure Is Linked with Sex Specific Changes in the Transcriptome and Methylome of Human Amniocytes

Abstract

Context: Prenatal exposure to bisphenol A (BPA) is linked to obesity and diabetes but the molecular mechanisms driving these phenomena are not known. Alterations in deoxyribonucleic acid (DNA) methylation in amniocytes exposed to BPA in utero represent a potential mechanism leading to metabolic dysfunction later in life.

Objective: To profile changes in genome-wide DNA methylation and expression in second trimester human amniocytes exposed to BPA in utero.

Design: A nested case-control study was performed in amniocytes matched for offspring sex, maternal race/ethnicity, maternal age, gestational age at amniocentesis, and gestational age at birth. Cases had amniotic fluid BPA measuring 0.251 to 23.74 ng/mL. Sex-specific genome-wide DNA methylation analysis and RNA-sequencing (RNA-seq) were performed to determine differentially methylated regions (DMRs) and gene expression changes associated with BPA exposure. Ingenuity pathway analysis was performed to identify biologically relevant pathways enriched after BPA exposure. In silico Hi-C analysis identified potential chromatin interactions with DMRs.

Results: There were 101 genes with altered expression in male amniocytes exposed to BPA (q < 0.05) in utero, with enrichment of pathways critical to hepatic dysfunction, collagen signaling and adipogenesis. Thirty-six DMRs were identified in male BPA-exposed amniocytes and 14 in female amniocyte analysis (q < 0.05). Hi-C analysis identified interactions between DMRs and 24 genes with expression changes in male amniocytes and 12 in female amniocytes (P < 0.05).

Conclusion: In a unique repository of human amniocytes exposed to BPA in utero, sex-specific analyses identified gene expression changes in pathways associated with metabolic disease and novel DMRs with potential distal regulatory functions.

Figure 1.

Amniotic fluid bisphenol A (BPA) concentration of amniocytes assayed by enhanced reduced representation bisulfite sequencing. Male BPA-exposed samples = black squares (n = 8). Female BPA-exposed samples = black circles (n = 7).

Figure 2.

Genes with significant changes in expression identified via RNA-seq. (A) Genes with q < 0.05. (B) Genes with P < 0.01, absolute value logFC > 1, logCPM+. X axis represents the number of genes with increased (positive values, black bars) or decreased (negative values, grey bars) expression compared to control for each bar. Y axis represents the comparison pairs: all bisphenol A (BPA) versus control, male BPA versus male control, and female BPA versus female control, as well as total number of differentially expressed genes for each comparison pair.

Figure 3.

Confirmation of RNA-sequencing gene expression changes via quantitative polymerase chain reaction. (A) Relative fold change of COL5A1 in bisphenol A (BPA) male versus control male. (B) Relative fold change INSIG1 in BPA female versus control female. Relative fold change was determined by 2^ΔΔCt using ITCH as the housekeeping gene. Data are statistically analyzed by unpaired t-test; P < 0.05.

Figure 4.

CpG sites with significant changes in DNA methylation. (A) CpGs with absolute value (AV) change in percentage methylation >15% compared to controls. q < 0.05. (B) CpGs with AV percentage methylation > 20%, q < 0.05. (C) CpGs with AV percentage methylation > 25%, q < 0.05. X axis represents each comparison pair: all bisphenol A (BPA) versus control, male BPA versus male control, and female BPA versus female control, including the total number of differentially methylated CpGs for each pair. Y axis represents number of significant differentially methylated CpGs with increased (positive, black bars) or decreased (negative, grey bars) percentage methylation levels. (D–F) Distribution of location of single CpGs. Values represent percentage of single CpGs located within a specific genomic regulatory region.

Figure 5.

Genes predicted to interact with differentially methylation regions (DMRs) through 3D chromatin looping. (A) Number of genes with potential interactions with DMRs through 3D chromatin looping identified by in silico Hi-C analysis. Gene expression directionality change identified from RNA-sequencing (RNA-seq; some interacting genes had no change on RNA-seq). (B) Number of genes with potential interactions with DMRs through 3D chromatin looping identified by in silico Hi-C that also had significant changes in gene expression as measured by RNA-seq (P < 0.05). Number of interacting genes with increased expression shown with positive solid black bars and genes with decreased expression identified with negative grey bars.

Figure 6.

Gene expression changes in human adipose derived stem cells assessed via quantitative polymerase chain reaction. Relative fold change of (A) KLF2, (B) GATA1, and (C) GATA3, after treatment with 1nM, 5nM, and 100nM bisphenol A in undifferentiated human adipose-derived stem cells. Relative fold change is determined by 2^ΔΔCt using 18s as the housekeeping gene. Data are statistically analyzed by one-way analysis of variance; P values are relative to vehicle on post hoc analysis.