Data integration, analysis, and interpretation of eight academic CLARITY-BPA studies

Abstract

"Consortium Linking Academic and Regulatory Insights on BPA Toxicity" (CLARITY-BPA) was a comprehensive "industry-standard" Good Laboratory Practice (GLP)-compliant 2-year chronic exposure study of bisphenol A (BPA) toxicity that was supplemented by hypothesis-driven independent investigator-initiated studies. The investigator-initiated studies were focused on integrating disease-associated, molecular, and physiological endpoints previously found by academic scientists into an industry standard guideline-compliant toxicity study. Thus, the goal of this collaboration was to provide a more comprehensive dataset upon which to base safety standards and to determine whether industry-standard tests are as sensitive and predictive as molecular and disease-associated endpoints. The goal of this report is to integrate the findings from the investigator-initiated studies into a comprehensive overview of the observed impacts of BPA across the multiple organs and systems analyzed. For each organ system, we provide the rationale for the study, an overview of methodology, and summarize major findings. We then compare the results of the CLARITY-BPA studies across organ systems with the results of previous peer-reviewed studies from independent labs. Finally, we discuss potential influences that contributed to differences between studies. Developmental exposure to BPA can lead to adverse effects in multiple organs systems, including the brain, prostate gland, urinary tract, ovary, mammary gland, and heart. As published previously, many effects were at the lowest dose tested, 2.5μg/kg /day, and many of the responses were non-monotonic. Because the low dose of BPA affected endpoints in the same animals across organs evaluated in different labs, we conclude that these are biologically - and toxicologically - relevant.

Keywords: Bisphenol A; CLARITY-BPA; EDC; Endocrine disruptor; GLP; Guideline study; Systemic effects.

Fig. 1

Dorsolateral prostate pathology and prostasphere numbers in rats treated with vehicle or increasing doses of BPA. A—B) Prostate pathology in 1-year-old rats treated with bisphenol A (BPA) during gestation to weaning (stop-dose) and given implants of testosterone + estradiol (T + E) at postnatal day 90 to elevate circulating estradiol levels. A) Severity scores of lateral lobe prostate (LP) intraepithelial neoplasia (PIN) lesions were significantly elevated in rats given ethinyl estradiol (EE) or 2.5, 250, or 25,000 μg/kg/day BPA during development as compared to vehicle controls. *P < 0.05, **P < 0.01 vs controls. B) Multiplicity of dorsolateral prostate (DLP) ductal adenocarcinoma was significantly increased in rats treated with 2.5 μg/kg/day BPA during early life as compared to vehicle controls. **P < 0.01 vs controls. CD—) Prostaspheres in 6-month old rats treated continuously with vehicle or BPA from gestational day 6 through time of tissue collection. Number of prostaspheres cultured from DLPs of rats exposed to BPA from gestation through 6 months of age. Daily exposure to EE or 2.5 μg/kg/day BPA doubled the spheroid numbers as compared to vehicle controls (ANOVA = 0.02; *P < 0.02 vs vehicle). Treatment with 250 μg BPA also doubled prostasphere numbers, but this was not significant due to high variance. D) Prostasphere sizes showed a significant increase in large spheroids (>80 μm) from DLPs exposed in vivo to EE or 25 μg/kg/day BPA as compared to vehicle controls (ANOVA < 0.01; ^†P < 0.01, *P < 0.02 vs vehicle). N for each group: vehicle=4; EE=5; 2.5 μg BPA=5; 25 μg BPA=3; and 250 μg BPA= 5.

Fig. 2

3D serial section reconstruction of the urogenital sinus (UGS) from gestation day (GD) 19 male CD-1 mice exposed to low doses of bisphenol A (BPA) and ethinyl estradiol (EE) from gestation day 14-18 via feeding the pregnant dam. UGS depicted for each treatment was closest to the group mean. There was a marked alteration in urethra shape, particularly at the junction of the bladder and urethra, which is constricted (*) in mice exposed to estrogenic chemicals compared to controls. In addition, the UGS region (prostatic sulcus or colliculus, arrow) is significantly enlarged by BPA compared to controls, based on data published in Timms et al. 2005 [59].

Fig. 3

Colliculus measurements. Colliculus angle (CAo) was defined as the angle the colliculus makes at the juncture with the cranial urethra (A) and analyzed by treatment (B). The colliculus size (CS), shaded blue was determined by measuring the colliculus distance between the lowest caudal point of the colliculus and the lowest caudal point drawn on the urethra and taking the reciprocal (C) and analyzed by treatment (D). (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

Fig. 4

C. Urethra 3D reconstruction using BioVis3D demonstrating how this software can isolate and quantify individual regions of the UGS to calculate length and width (μM), surface area (μM²), and volume (μM³) of the urethra. Shown is urethral distance measured straight across from the dorsal (most cranial point of urethra on most dorsal section) to ventral aspect: dorsal-ventral distance (I); cranial-caudal distance (J); urethral volume (K); and surface area (L). A. The width of the urethral lumen was significantly smaller in the BPA 25 and EE 0.05 groups relative to vehicle controls. B. the length of the urethra was significantly shorter in the BPA 2.5 and EE 0.05 groups relative to negative controls.

Fig. 5

(A) Body weight data on postnatal day (PND) 1 from 81 male rats collected and weighed by the NCTR staff to study effects of BPA and EE on urethra and bladder morphology. (B) From 81 males, 46 met the weight range criteria of 5 – 8.1 g and were provided blind for our urethra study by staff at the NCTR (Fig. 5-B). There were 5 blocks of matings, but males exposed to 2500 μg/kg/day BPA were all selected from treatment block 2, which tended to be lighter than pups from other blocks.

Fig. 6

Non-monotonic responses in (A) depth (measured in μm) of the epithelial gland compartment and (B) mean of average branch width (measured in μm) to bisphenol A (BPA) doses in postnatal day (PND) 21 animals. In both graph the x axis represents BPA doses in ug/kg/day. Non-linear regression illustrates a breaking point between 25 and 250 μg/kg/day doses. Graphs represent mean and standard deviation for each dose, and fit with a combination of linear and step functions. This pattern was observed for the majority of endpoints measured.

Fig. 7

Box plots of (A) roundness (ratio between smallest and largest axes of gland) and (B) log of average length of longest branches (length > 75 μm) of postnatal day (PND) 21 animals treated with control, 250 μg BPA/kg/day, or 0.5 μg/kg/day EE (n = 8–10 animals per group). P-values correspond to pairwise t-test.

Fig. 8

Effects of EE and BPA on ovarian morphology at postnatal day (PND) 21. On PND 21, rats from each group were euthanized, and one ovary from each animal was fixed for histological evaluation of ovarian follicle types. Graph represents mean ± SEM of number of follicles. *Significant difference between control group and BPA or EE groups (n = 8–10; P ≤ 0.05).

Fig. 9

Effects of continuous exposure to EE and BPA at 1 year. At 1 year, rats from each group were euthanized and serum was collected from the blood to measure sex steroid hormones. Graph represents means ± SEM of the amount of estradiol present in serum. *Significant difference between control group and BPA or EE groups (n = 3–9; P ≤ 0.05); ^P > 0.05.

Fig. 10

Concordance of BPA-induced ER mRNA expression changes in the amygdala and hypothalamus across NCTR-based studies. A. Estrogen receptor beta (ESR2) expression in the medial amygdala (MePD) is sexually dimorphic at birth, with higher levels in females, but switches at approximately PND 4, demonstrating how expression can change across development. Significant differences in expression compared to PND 0 levels are represented by **p < 0.01; significant sex differences are represented by †<0.05. B. A representative autoradiogram depicting the sex‐specific expression of ESR2 in the MePD and the neighboring central portion of the ventrolateral region of the ventromedial nucleus (cVMNvl) on PND2. As the sex difference in MePD expression diminishes, the one in the cVMNvl remains pronounced demonstrating that sex differences in ER expression are age and region-specific. C. An example from Study 1 revealing how dramatically different ESR2 expression differed between the gavaged (vehicle) and naïve controls. BPA exposure elevated ESR2 expression in both sexes but not to the level of the naïve controls. ***p < 0.001; ###<0.001 compared to male vehicle controls; §§§<0.001 compared to female vehicle controls. The direction of ER (ESR1 and ESR2) expression changes by dose and study are summarized in Table 5 for the hypothalamus and amygdala. Images adapted and compiled from prior Patisaul publications [61,121,122].

Fig. 11

Overall hazard ratio for (A) females and (B) males in each treatment group to locate the escape hole in a Barnes maze. Note that increasing ratio equates to shorter latency. For both graphs, upper, middle, and lower bars represent ratio of locating the correct escape hole at 95 % upper confidence limit, mean, and 95 % lower confidence limit, respectively, for each group. Hazard ratio was used to account for those individuals who did not locate the escape hole in the allotted time (5 min). Comparisons of the significant two-way interaction for treatment ∗ sex are shown. Fig. reproduced with permission from Hormones and Behavior.

Fig. 12

Effect of early-life exposure to EE and BPA on methylation and expression of Bdnf in the adult rat hippocampus. (A) both males and females, (B) only females, and (C) only males with early-life exposure to vehicle control (white), 0.5 μg/kg/day EE (orange), or 2500 μg/kg/day BPA (violet). Data expressed as mean ± SEM. Each circle represents average percent methylation of each corresponding CG site from 6–7 individual colonies (N = 5 males and N = 5 females per group). For methylation analysis, *P < 0.05, ***P < 0.001, and ****P < 0.0001 vs vehicle by two-way ANOVA and Tukey’s multiple test comparison. For gene expression analysis, ***P < 0.001 vs vehicle by one-way ANOVA and Tukey test. Gene expression levels are expressed as gene expression level in treatment group relative to control group using the 2^−ΔΔCT method. Reproduced with permission from the Epigenetics.

Fig. 13

Correlation analyses for hippocampal Bdnf expression. (A) Overall correlation between expression of Bdnf and percent promoter methylation of Bdnf in rats using all data points. (B) Overall correlation between expression of Bdnf and rate of sniffing the correct hole in the Barnes maze in female rats. Gene expression levels are expressed in log 2 values and as CT value of Rpl19 relative to target gene. Percentage methylation of the putative 5′ end region of Bdnf is in log value. Correlation between velocity in the Barnes maze and hippocampal expression of (C) Avp, (D) Esr2, (E) Oxt, and (F) Otr in control rats. Gene expression levels expressed in log 2 values and as CT value of Rpl19 relative to target gene. Percentage methylation of putative 5′ end region of Bdnf is in log value. Velocity is expressed as mean from seven observations. P < 0.05 was considered statistically significant. P < 0.05 was considered statistically significant. Only samples with detectable CT values were used in correlation analysis. Reproduced with permission from the Epigenetics.

Fig. 14

R plot correlations across 6-month-old female data for low-dose BPA exposure (2.5 μg/kg/day) and controls. This analysis compares white adipose tissue, ovarian follicle, heart, mammary gland, and uterine assessments with peptide and steroid hormone data, and splenic function results. Values with strong correlation (r ≥ 0.7) are highlighted. Representative PCA diagrams with corresponding r-values are also delineated. In the PCA diagrams, each individual replicate is represented with a circle, and the blue and orange circles represent those exposed to BPA or vehicle control, respectively. N = 6 for BPA and vehicle control, respectively. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

Fig. 15

A) Positive and (B) negative circos plot correlations between white adipose tissue, ovarian follicle, heart, mammary gland, and uterine assessments with peptide and steroid hormone data in 6-month-old females exposed to low-dose BPA (2.5 μg/kg/day) or controls. Results for BPA-exposed females are indicated with a blue line outside of the circle; orange line indicates results for control females. Color of the line further from the circle indicates treatment group where these results are greater. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

Fig. 16

R plot correlations across 6-month-old male data for low-dose BPA exposure (2.5 μg/kg/day) and controls. This analysis compares white adipose tissue results, macrophages in white adipose tissue, adipose tissue weight, heart result, peptide hormones, prostate results, and splenic function results. Values with strong correlation (r ≥ 0.5) are highlighted. Representative PCA diagrams with corresponding r-values are also delineated. In the PCA diagrams, each individual replicate is represented with a circle, and the blue and orange circles represent those exposed to BPA or vehicle control, respectively. N = 7 for BPA and 6 for vehicle control, respectively. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

Fig. 17

(A) Positive and (B) negative circos plot correlations of white adipose tissue, heart, and prostate assessments with peptide hormone data in 6-month-old males exposed to low-dose BPA (2.5 μg/kg/day) or controls. Results for BPA-exposed males are indicated with a blue line outside the circle; orange line indicates results for control males. Color of the line further from the circle indicates treatment group where these results are greater. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)