Effects of Pubertal Exposure to Butyl Benzyl Phthalate, Perfluorooctanoic Acid, and Zeranol on Mammary Gland Development and Tumorigenesis in Rats

Abstract

Endocrine-disrupting chemicals (EDCs)-including butyl benzyl phthalate (BBP), perfluorooctanoic acid (PFOA), and zeranol (α-ZAL, referred to as ZAL hereafter)-can interfere with the endocrine system and produce adverse effects. It remains unclear whether pubertal exposure to low doses of BBP, PFOA, and ZAL has an impact on breast development and tumorigenesis. We exposed female Sprague Dawley rats to BBP, PFOA, or ZAL through gavage for 21 days, starting on day 21, and analyzed their endocrine organs, serum hormones, mammary glands, and transcriptomic profiles of the mammary glands at days 50 and 100. We also conducted a tumorigenesis study for rats treated with PFOA and ZAL using a 7,12-dimethylbenz[a]anthracene (DMBA) model. Our results demonstrated that pubertal exposure to BBP, PFOA, and ZAL affected endocrine organs and serum hormones, and induced phenotypic and transcriptomic changes. The exposure to PFOA + ZAL induced the most phenotypic and transcriptomic changes in the mammary gland. PFOA + ZAL downregulated the expression of genes related to development at day 50, whereas it upregulated genes associated with tumorigenesis at day 100. PFOA + ZAL exposure also decreased rat mammary tumor latency, reduced the overall survival of rats after DMBA challenge, and affected the histopathology of mammary tumors. Therefore, our study suggests that exposure to low doses of EDCs during the pubertal period could induce changes in the endocrine system and mammary gland development in rats. The inhibition of mammary gland development by PFOA + ZAL might increase the risk of developing mammary tumors through activation of signaling pathways associated with tumorigenesis.

Keywords: RNA sequencing; Wnt signaling; endocrine-disrupting chemicals; estrogen signaling; mammary gland development; pubertal exposure; tumorigenesis.

Figure 1

Changes in rat mammary gland development induced by pubertal exposure to BBP, PFOA, and ZAL: (A) Experimental design for the study of rat mammary gland development. The rats were treated with 333 µL of working solution per 100 g of BW through oral gavage. (B) Representative images of rat mammary glands #4 and 5 whole mounts at an age of 100 days. Scale bar: 2000 µm. (C) Boxplots show the number of terminal end buds (TEBs) per cm² in the C area. TEB is a structure with a bulbous tip 80–120 µm in diameter on the perimeter of the gland. Each dot in the graph represents one rat. The samples with poor whole-mount preparation were excluded. Sample size is indicated by the numbers within parentheses after the name of each group; this rule is also applied to other figures. (D) Table shows the evaluation of mammary gland development by whole-mount qualitative rating, and IHC analyses of Ki67, ER alpha, and PR in the mammary glands. Data are presented as the mean ± STD. (E) Representative IHC images of the rat mammary glands. Magnification: 20× objective. Scale bar: 200 µm. One-way ANOVA or Kruskal–Wallis one-way analysis of variance on ranks were used to compare the phenotypic changes among 11 groups. Two-sample t-tests or Mann–Whitney rank sum tests were used to compare treated groups with Ctrl. These tests were also applied to Figure 2; * indicates p < 0.05, ** indicates p < 0.01 compared to Ctrl.

Figure 2

Changes in endocrine organs and serum hormones induced by exposure to BBP, PFOA, and ZAL: (A) Table shows the wet weight of uteri, ovaries, and adrenal glands at D50 and D100. The mean weight of two ovaries or adrenal glands of each rat was calculated and used to conduct the analysis. Sample size information is included in the Supplementary Materials. (B) Representative images of adrenal glands by H&E staining, showing necrosis in the adrenal cortex. Scale bar: 200 µm for 4×, 50 µm for 20× objective. (C) Quantification of the necrotic area in the adrenal gland cortex; each dot represents one rat. The samples with poor sectioning were excluded. (D) Serum E2 and P4 concentrations at D50 and D100; data are presented as the mean ± STD. (E) Association of serum P4 concentration with the phase of the estrous cycle. Only the samples that had enough cells for vaginal smear analysis were included. The Mann–Whitney rank sum test was used to perform the analysis; * indicates p < 0.05, ** indicates p < 0.01 compared to Ctrl for (A,C,D).

Figure 3

Transcriptomic changes in the rat mammary glands caused by BBP, PFOA, and ZAL exposure: (A) The number of DEGs induced by BBP, PFOA, and ZAL -alone or in combination- at D50 and D100; n = 6 glands per group. (B) Venn diagram representing the number of DEGs in each treatment (FC ≥ 2) and the common genes between/among the treatments. (C) Volcano plots of pairwise comparisons, displaying the relation between fold change and significance between the two groups, using a scatterplot view. The y-axis shows the negative log₁₀ of FDR-adjusted p-values (−log₁₀(p value)); a higher value indicates greater significance. The x-axis shows the difference in expression between the treated group and Ctrl, presented in log₂ fold change (log₂FC). Each gene is represented by one dot in the graph; red dots represent genes showing statistically significant changes (FDRp < 0.05) and log₂FC < 1, blue dots represent genes showing FDRp < 0.05 and log₂FC ≥ 1 (considered as DEGs), and black dots represent non-significant genes. (D) Heatmaps of the 40 top down- or upregulated DEGs (FC ≥ 2.0) by PFOA + ZAL compared to Ctrl at D50. Color code for downregulated genes: blue for overexpression, black for intermediate expression, and green for underexpression. Color code for upregulated genes: red for overexpression, black for intermediate expression, and blue for underexpression; n = 6 mammary gland samples per group, and each sample was sequenced twice.

Figure 4

Biological process and KEGG pathway analysis of DEGs induced by PFOA + ZAL: (A,C) Bar graphs represent genes categorized according to their most prominent biological function. Gene Ontology annotations were extracted using DAVID tools and the Shiny application in R version 3.5.3. (B,D) Involvement of the KEGG pathway genes was determined using KEGG web tools and the Shiny application in R version 3.5.3. Arrows indicate pathways discussed in the paper.

Figure 5

RT-PCR validation and IHC analysis for selected DEGs: (A) Graphs show qRT-PCR analysis of selected DEGs involved in mammary gland development or carcinogenesis. (B) IHC analysis of CD3 and CD8a in rat mammary glands. The Mann–Whitney rank sum test was used for statistical analysis. Sample size: n = 7. Magnification: 20× objective. Scale bar: 200µm; * indicates p < 0.05, ** indicates p < 0.01, @ indicates p < 0.1 compared to Ctrl.

Figure 6

Effects of exposure to PFOA_L, ZAL_L, and PFOA + ZAL on mammary tumorigenesis: (A) Experimental design for the mammary tumorigenesis study. Female Sprague Dawley rats were treated with low doses of PFOA, ZAL, or their combination from ages of 21 to 42 days, and were given one dose of DMBA at 30 mg/kg body weight at the age of 50 days. (B) Table shows tumor incidence and the numbers of tumors in tumor-bearing rats. (C) Dot plots show the number of mammary tumors developed in each rat; each dot represents one rat. One-way ANOVA, p = 0.674. The three lines in the dot plots indicate the lower quartile, median, and upper quartile. (D) Kaplan-Meier tumor-free survival calculated by the time period from DMBA administration to the appearance of the first tumor, indicating tumor latency. (E) Kaplan-Meier overall survival calculated by the time period from DMBA administration to the day of euthanasia. (F) Hazard ratio analysis of risk of death using Prism 8.0.

Figure 7

Histopathological analysis and molecular classification of rat mammary tumors: The histopathological type of each tumor was presented as single type when only one histopathological type was found on the tumor, or mixed type when two or more histopathological types were observed on a tumor. (A) Histological evaluation of the rat mammary tumors. (B) IHC analysis of Ki67, ER alpha, and PR in rat mammary tumors. Upper panel: representative image of one tissue microarray core of each staining. Bottom panel: one high-magnification (40× objective) field showing the staining of the cell nuclei. Scale bar: 20 µm. (C) Tables show sample size for IHC analysis and the percentage of positive cells for each staining. Data are presented as the mean ± STD. (D) Measurement of the BBP metabolite mBzP in urine, and serum concentrations of PFOA and ZAL at the age of 43 days—1 day after the last administration. NHANES: National Health and Nutrition Examination Survey; LOQ: limit of quantification; p50_NHANES and p95_NHANES represent the 50% and 95% percentile levels in the NHANES 2002–2012 data, respectively.