Sex- and Dose-Specific Effects of Maternal Bisphenol A Exposure on Pancreatic Islets of First- and Second-Generation Adult Mice Offspring

Abstract

Background: Exposure to the environmental endocrine disruptor bisphenol A (BPA) is ubiquitous and associated with the increased risk of diabetes and obesity. However, the underlying mechanisms remain unknown. We recently demonstrated that perinatal BPA exposure is associated with higher body fat, impaired glucose tolerance, and reduced insulin secretion in first- (F1) and second-generation (F2) C57BL/6J male mice offspring.

Objective: We sought to determine the multigenerational effects of maternal bisphenol A exposure on mouse pancreatic islets.

Methods: Cellular and molecular mechanisms underlying these persistent changes were determined in F1 and F2 adult offspring of F0 mothers exposed to two relevant human exposure levels of BPA (10μg/kg/d-LowerB and 10mg/kg/d-UpperB).

Results: Both doses of BPA significantly impaired insulin secretion in male but not female F1 and F2 offspring. Surprisingly, LowerB and UpperB induced islet inflammation in male F1 offspring that persisted into the next generation. We also observed dose-specific effects of BPA on islets in males. UpperB exposure impaired mitochondrial function, whereas LowerB exposure significantly reduced β-cell mass and increased β-cell death that persisted in the F2 generation. Transcriptome analyses supported these physiologic findings and there were significant dose-specific changes in the expression of genes regulating inflammation and mitochondrial function. Previously we observed increased expression of the critically important β-cell gene, Igf2 in whole F1 embryos. Surprisingly, increased Igf2 expression persisted in the islets of male F1 and F2 offspring and was associated with altered DNA methylation.

Conclusion: These findings demonstrate that maternal BPA exposure has dose- and sex-specific effects on pancreatic islets of adult F1 and F2 mice offspring. The transmission of these changes across multiple generations may involve either mitochondrial dysfunction and/or epigenetic modifications. https://doi.org/10.1289/EHP1674.

Figure 1.

Mitochondrial driven insulin secretion in islets of F1 and F2 adult male offspring: (A, C) F1 and (B, D) F2 adult male offspring. The maximal KIC-stimulated insulin release was calculated as the insulin area under the curve for $15\text{mM}$ KIC. Data are individual litter data (one animal per litter), presented as $\text{mean}+\text{SEM}$, and analyzed using Dunnett’s test. p-Values are relative to Control. Note: KCL, potassium chloride; KIC, $\mathrm{\alpha }-\text{ketoisocaproate}$.

Figure 2.

Glucose-stimulated and mitochondrial associated insulin secretion in islets of F1 and F2 adult female offspring: (A–B, E–F) F1, and (C–D, G–H) F2 adult female offspring. The maximal glucose- and KIC-stimulated insulin release was calculated as the insulin area under the curve for $25\text{mM}$ glucose or $15\text{mM}$ KIC, respectively. Data are individual litter data (one animal per litter), presented as $\text{mean}+\text{SEM}$, and analyzed using Dunnett’s test. $p-\text{Values}$ are relative to Control. Note: KCL, potassium chloride; KIC, $\mathrm{\alpha }-\text{ketoisocaproate}$.

Figure 3.

Basal and maximal mitochondrial oxygen consumption in islets of F1 and F2 adult male offspring: (A–B) F1, and (C–D) F2 adult male offspring. Data are individual litter data (islets pooled from 2–3 males from the same litter) with mean superimposed. Data were analyzed using Dunnett’s test. p-values are relative to Control. Oxygen flow per cell at: ROUTINE, i.e., basal state of cell respiration; electron transport system [ETS; uncoupled; carbonyl cyanide-p-trifluoro-methoxyphenylhydrazone (FCCP)]; and residual oxygen consumption (rox; rotenone, antimycin A).

Figure 4.

Beta cell mass of F1 and F2 adult male offspring. Representative photomicrographs of pancreatic immunofluorescent staining in (A–C) F1 and (D–F) F2 Control, LowerB, UpperB male mice. All images have insulin (red), glucagon (green), somatostatin (yellow), and DAPI (blue). Image magnified $20\times$. Data are individual litter data (one animal per litter) with mean superimposed from (G) F1 and (H) F2 males, analyzed by Dunnett’s test. p-Values are relative to Control.

Figure 5.

Cell viability in islets of (A–C) F1 and (D–F) F2 male offspring. Data are individual litter data (one animal per litter) with mean superimposed. Each parameter is normalized to total protein concentration. Data were analyzed using Dunnett’s test performed on log-transformed data, where required (BCL2). Note: RFU: relative fluorescence unit. p-Values are relative to Control.

Figure 6.

IL6, MCP1 levels in (A–B) F1 and (C–D) F2 adult male offspring. Note: IL6, interleukin 6; MCPm: monocyte chemoattractant protein-1. Cytokine levels for each sample were normalized to total protein concentration. Data are individual litter data (one animal per litter) with mean superimposed. Data were analyzed by Dunnett’s test performed on log-transformed data, where required (F1 IL6, MCP1; F2 IL6). p-Values are relative to Control.

Figure 7.

Representative photomicrographs of pancreatic immunohistochemical staining in F1 and F2 adult male offspring: (A–H) F1 and (I–P) F2 Control, LowerB, UpperB mice, and Positive Ctrl (internal Control—lymph node). Image magnified $20\times$. Note: CD3, cluster of differentiation 3 (marker for T lymphocytes); F4/80, marker for macrophages. F1, $n=6$ litters per group (one animal per litter); F2, $n=5$ litters Control and UpperB, $n=4$ litters LowerB (one animal per litter).

Figure 8.

Percent DNA methylation at Igf2 DMR1 and Esr1 Exon A in islets of F1 and F2 adult male offspring: (A–B) F1 and (C–D) F2 adult male offspring. Data are percent CpG methylation values from individual litter (islets pooled from 2–3 males from the same litter) and presented as $\text{mean}+\text{SEM}$. Data were analyzed using Dunnett’s test; p-Values are relative to Control. ** $p<0.05$, * $p>0.05$ and $p<0.09$.

Figure 9.

RNA Seq data in islets of F1 male mice. (A) F1 Lower BPA male vs. Control male canonical pathways from Ingenuity Pathway Analysis (IPA); IPA parameters: 962 genes; $p<0.05$; inflammatory pathways highlighted by the rectangular box. (B) F1 Upper BPA male vs. Control male canonical Pathways from IPA; IPA parameters: 1,275 genes; $p<0.05$; inflammatory and mitochondrial dysfunction pathways highlighted by the rectangular box. (C) Venn diagram illustrating number of differentially expressed genes when comparing F1 Lower BPA (LowerB), Upper BPA (UpperB), and Control males; parameters: false discovery rate (FDR) $<10\%$, which is equivalent to $p$ adjusted (also known as q-value) $<0.1$; $n=3$ litters (islets pooled from 2–3 males from the same litter) per group.