Effects of low doses of bisphenol A on the metabolome of perinatally exposed CD-1 mice

Abstract

Background: Bisphenol A (BPA) is a well-known endocrine disruptor used to manufacture polycarbonate plastics and epoxy resins. Exposure of pregnant rodents to low doses of BPA results in pleiotropic effects in their offspring.

Objective: We used metabolomics--a method for determining metabolic changes in response to nutritional, pharmacological, or toxic stimuli--to examine metabolic shifts induced in vivo by perinatal exposure to low doses of BPA in CD-1 mice.

Methods: Male offspring born to pregnant CD-1 mice that were exposed to vehicle or to 0.025, 0.25, or 25 µg BPA/kg body weight/day, from gestation day 8 through day 16 of lactation, were examined on postnatal day (PND) 2 or PND21. Aqueous extracts of newborns (PND2, whole animal) and of livers, brains, and serum samples from PND21 pups were submitted to (1)H nuclear magnetic resonance spectroscopy. Data were analyzed using partial least squares discriminant analysis.

Results: Examination of endogenous metabolic fingerprints revealed remarkable discrimination in whole extracts of the four PND2 newborn treatment groups, strongly suggesting changes in the global metabolism. Furthermore, statistical analyses of liver, serum, and brain samples collected on PND21 successfully discriminated among treatment groups. Variations in glucose, pyruvate, some amino acids, and neurotransmitters (γ-aminobutyric acid and glutamate) were identified.

Conclusions: Low doses of BPA disrupt global metabolism, including energy metabolism and brain function, in perinatally exposed CD-1 mouse pups. Metabolomics can be used to highlight the effects of low doses of endocrine disruptors by linking perinatal exposure to changes in global metabolism.

Figure 1

Two-dimensional PLS-DA score plot of integrated ¹H NMR spectra of PND2 whole body extracts. (A) PLS‑DA results for all treatment treatment groups (control, n = 20; 0.025 µg BPA, n = 18; 0.25 µg BPA, n = 14; and 25 µg BPA, n = 11) (first and second latent variable of three components: R²Y = 71.5% and Q² = 0.557). (B) PLS‑DA results for 0.025 µg BPA and 0.25 µg BPA taken separately (first and second latent variable of three components: R²Y = 73.1% and Q² = 0.731).

Figure 2

Two-dimensional PLS-DA score plot of integrated ¹H NMR spectra of PND21 serum. (A) PLS‑DA results for all four treatment groups [control, n = 11; 0.025 µg BPA, n = 12; 0.25 µg BPA, n = 14; and 25 µg BPA, n = 12 (two components): R²Y = 55.3% and Q² = 0.450]. (B) PLS‑DA results for control and 0.025 µg BPA (three components: R²Y = 99.6% and Q² = 0.826) (C) PLS‑DA results for control and 0.25 µg BPA (four components: R²Y = 98.9% and Q² = 0.932).

Figure 3

Two-dimensional PLS-DA score plot of integrated ¹HNMR spectra of PND21 liver extracts. (A) PLS‑DA results for all four treatment groups [control, n = 11; 0.025 µg BPA, n = 11; 0.25 µg BPA, n = 13; 25 µg BPA, n = 14 (first and second latent variable of three components: R²Y 48.3% and Q² = 0.421]. (B) PLS‑DA results for control and 0.025 µg BPA (one component: R²Y = 89.7% and Q² = 0.822); red and green lines indicate ± 3 SD and ± 2 SD, respectively. (C) PLS‑DA results for 0.025 µg BPA and 0.25 µg BPA (three components: R²Y = 99.5% and Q² = 0.896). (D) PLS‑DA results for 0.025 µg BPA and 25 µg BPA (three components: R²Y = 99.4% and Q² = 0.950).

Figure 4

Two-dimensional PLS-DA score plot of integrated ¹H NMR spectra of PND21 brain extracts from control (n = 11), 0.025 µg BPA (n = 11), 0.25 µg BPA (n = 13), and 25 µg BPA (n = 14) groups (first and second latent variable of three components: R²Y = 78.9% and Q² = 0.564).