Exposure to brominated flame retardants in utero and through lactation delays the development of DMBA-induced mammary cancer: potential effects on subtypes?

Abstract

Introduction: Brominated flame retardants (BFRs) are chemical compounds used to reduce the flammability of various products; some BFRs exhibit endocrine-disrupting properties and can leach into the environment leading to human and wildlife exposure. The mammary gland has specific vulnerability windows during which it is more sensitive to the effects of endocrine disrupting compounds (EDCs), such as the in utero life, puberty and pregnancy. Our previous studies revealed precocious mammary gland development, disruptions in junctional proteins, and altered proliferation-apoptosis balance during puberty in rats exposed to BFRs in utero and through lactation. Such effects have been associated with increased mammary cancer risk.

Objective: The current study aimed to determine if in utero and lactational exposure to BFRs renders the mammary gland more susceptible to 7,12-dimethylbenz[a]anthracene (DMBA)-induced mammary cancer.

Methods: Dams were exposed to a BFRs mixture (0. 0.06 or 60 mg/kg/day), and mammary cancer was induced in pups using DMBA at post-natal day 46. Tumors onset and growth were monitored, and tumors were characterized using histology and molecular biology.

Results: Although BFRs exposure did not significantly affect mammary tumor number or burden, it showed significant delay in mammary tumor onset and growth in BFR-exposed animal. These effects could potentially be due to BFRs' impact on cellular responses, DMBA metabolism, or mammary gland shift of the sensitivity window. Molecular analysis of mammary tumors showed a shift in the ratio of luminal A, luminal B, and (HER2)-enriched tumors, and an increase in triple-negative breast cancer (TNBC) subtypes in BFR-exposed animals. Additionally, BFRs exposure showed lung lesions indicative of inflammation, independent of mammary cancer development.

Conclusion: Our study highlights the complex relationship between BFRs exposure and mammary cancer risk, emphasizing the need for further investigation into underlying mechanisms and long-term effects of BFRs on mammary gland development and carcinogenesis.

Keywords: 7,12-dimethylbenz[a]anthracene (DMBA); brominated flame retardants; cancer; endocrine disrupting compounds (EDCs); gestational-lactational exposure; mammary gland.

Figure 1

Effects on tumor onset and measurements after an exposure in utero and through lactation to a brominated flame retardant (BFR) mixture and a gavage with DMB at PND46. No significant differences were observed for number of tumors (A), average tumor size (B) and tumor burden sum (considered as the sum of the estimated volume of all tumors per animal) (C) between treatments (0, 0.06 and 60 mg/kg of body weight/day). There was significant delay in tumor onset for the pups exposed to the high dose of BFRs, as measured by the number of days between DMBA exposure and the first tumor palpation (p=0.0039) (D). A delay was observed in survival in BFR-treated group, as defined by the time between DMBA- exposure (PND46) and the endpoint for the animal (tumor reaching 1.5 cm²) (E). The median survival was identified at 57 days for 0 mg/kg of body weight/day, 71 for 0.06 mg/kg of body weight/day and 142 for 60 mg/kg of body weight/day (p=0.0133 control vs low dose and p < 0.0001 control vs high dose) (E). Additionally, there was a significant decrease in the speed of tumor growth (from tumor onset until the endpoint, cm² per day) for the high dose (60 mg/kg/day) when compared to the control diet (p = 0.0014) (F). p-values were calculated using Mantel-cox test, Kruskal-Wallis statistical test or ANOVA. (n ≥ 10 pups, 1 per litter).

Figure 2

Effects of exposure in utero and through lactation to a brominated flame retardant (BFR) mixture on protein levels of cancer-related markers in DMBA-induced cancer in female pups. Semi-quantitative Western Blot analysis of total proteins extracted from the mammary tumors after indirect exposure to 0, 0.06 or 60 mg/kg of body weight/day and gavage by DMBA at PND46. Graphs show average expression of Estrogen Receptor alpha (ERα) (A), beta (ERβ) (B), Progesterone Receptors isoforms A and B (PR- A, PR-B) (C), Human epidermal growth factor receptor 2 (HER2) (p = 0.0225) (D) and Proliferating cell nuclear antigen (PCNA) (E). Histograms represent the means ± SEM (n = 10 pups, 1 per litter) for each band normalized to the total protein level. p-values were calculated with a Kruskal-Wallis statistical test or ANOVA. Oil-treated animals are showed in Supplementary Figures .

Figure 3

Effects of exposure in utero and through lactation to a brominated flame retardant (BFR) mixture and DMBA-treated on junctional and keratin protein levels of female pups. Semi-quantitative Western Blot analysis of total proteins extracted from the mammary tumors after indirect exposure to 0, 0.06 or 60 mg/kg of body weight/day and gavage by DMBA at PND46. Graphs show average expression of E-cadherin (A), β-catenin (B), connexin-43 (Cx43) (C), keratin-14 (K14) (D) and keratin-18 (K18). (E). Histograms represent the means ± SEM (n = 6-10 pups, 1 per litter) for each band normalized to the total protein level. p-values were calculated with a Kruskal-Wallis statistical test or ANOVA. Oil-treated animals are showed in Supplementary Figures .

Figure 4

Histology of mammary tumors using Masson's trichome staining. Exposure to BFRs did not affect the histological class of breast cancer. Cribriform carcinoma was the most represented breast cancer in the animals of the study independently to the exposure to BFRs. Images of representative tumors from DMBA-treated female pups upon gestational-lactational exposure to a mixture of BFRs at 0 mg/kg/day (A, B), 0.06 mg/kg/day (C, D) and 60 mg/kg/day (E, F). Tumors, excised at necropsy, were formalin fixed and paraffin embedded, and 5 µm sections were stained with Masson's trichrome. Scale = 100 μm.

Figure 5

Visualization of clustered protein expression profile per pup. Clustered heatmap of markers for breast cancer subtype classification among the animals from different treatments (0, 0.06 and 60 mg/kg of body weight/day). Each line represents relative levels of the markers for each animal (identified by the numbers in brackets). The interval of protein expression of animals from the group exposed to 0 mg/kg of body weight/day and given oil by gavage at PND46 was considered as a for this heatmap. The graph was coded with the pheatmap Rstudio function (n = 10 pups, 1 per litter).

Figure 6

Visualization of breast cancer subtypes based on their protein expression. The classification was performed based on Figure 5 profiles and Supplementary Figure S6 description. DMBA-induced tumors from animals exposed to control diet were composed of 30% luminal A, 10% luminal B, 60% HER2-enriched and 0% Triple-Negative (TN) subtypes (A). For the low dose of BFRs (0.06 mg/kg of body weight/day), an increase in the presence of TN (10%) and of luminal A, and a decrease of HER2-enriched subtypes were observed, while luminal B percentage remained unchanged (B). In animals exposed to the high dose of the BFRs mixture (60mg/kg body weight/day), there was an increase of TN, and a slight decrease of HER2-enriched and luminal A tumors percentages compared to control; no luminal B subtype tumors were identified (C).

Figure 7

Representation of lung lesions after BFRs exposure in utero and through lactation. The animals exposed to BFRs showed macroscopic and microscopic lesions, resulting in a significant difference between BFRs-exposed and control groups for the macroscopic lesions (A, B). Quantitative analysis of microscopic lesions was done by giving a grade to each slide [0 normal tissue (C), 1 abnormal cell accumulation (D) and 3 = lesions (E)]. Histograms represent the means ± SEM (n ≥ 6 pups, 1 per litter). p-values were calculated with a Kruskal-Wallis statistical test or ANOVA. Variables are statistically indistinguishable if they share at least one letter. Scale = 100 µm.