Inhibition of E2-induced expression of BRCA1 by persistent organochlorines

Background Environmental persistent organochlorines (POCs) biomagnify in the food chain, and the chemicals are suspected of being involved in a broad range of human malignancies. It is speculated that some POCs that can interfere with estrogen receptor-mediated responses are involved in the initiation and progression of human breast cancer. The tumor suppressor gene BRCA1 plays a role in cell-cycle control, in DNA repair, and in genomic stability, and it is often downregulated in sporadic mammary cancers. The aim of the present study was to elucidate whether POCs have the potential to alter the expression of BRCA1. Methods Using human breast cancer cell lines MCF-7 and MDA-MB-231, the effect on BRCA1 expression of chemicals belonging to different classes of organochlorine chemicals (the pesticide toxaphene, 2,3,7,8-tetrachlorodibenzo-p-dioxin, and three polychlorinated biphenyls [PCB#138, PCB#153 and PCB#180]) was measured by a reporter gene construct carrying 267 bp of the BRCA1 promoter. A twofold concentration range was analyzed in MCF-7, and the results were supported by northern blot analysis of BRCA1 mRNA using the highest concentrations of the chemicals. Results All three polychlorinated biphenyls and 2,3,7,8-tetrachlorodibenzo-p-dioxin reduced 17β-estradiol (E2)-induced expression as well as basal reporter gene expression in both cell lines, whereas northern blot analysis only revealed a downregulation of E2-induced BRCA1 mRNA expression in MCF-7 cells. Toxaphene, like E2, induced BRCA1 expression in MCF-7. Conclusion The present study shows that some POCs have the capability to alter the expression of the tumor suppressor gene BRCA1 without affecting the cell-cycle control protein p21Waf/Cip1. Some POCs therefore have the potential to affect breast cancer risk.


Introduction
The incidence of breast cancer has been increasing steadily over the past 60 years [1], today affecting one in eight women in the United States [2]. Despite tremendous efforts to understand the disease, less than 50% of all cases are of known etiology such as genetic inheritance, first-degree relatives with breast cancer, and age of menstruation and menopause [3].
A large group of lipophilic organochlorines are found to persist in the environment and to biomagnify through the food chain. This group includes pesticides, polychlorinated biphenyls (PCBs) and polychlorinated dibenzo-dioxins/ polychlorinated dibenzo-furans [4][5][6][7][8]. Many of these compounds can interfere with a wide range of hormonal responses [9][10][11][12][13], including estrogen receptor (ER)mediated responses. Due to their lipophilic nature, (page number not for citation purposes) Breast Cancer Research Vol 4 No 6 Rattenborg et al. organochlorines accumulate in the adipose tissue, can cross the placenta and are found in the breast milk [14]. The chemicals give rise to life-long exposure and are supposed to disturb early development. There is, however, a great epidemiological disparity in the literature regarding whether an association exists between POCs and breast cancer [15,16].
The breast cancer susceptibility gene, BRCA1 [17], is linked to familial breast cancer [18], and it is found to be downregulated in sporadic breast and ovarian cancers [19,20]. Decreased expression of BRCA1 promotes cell proliferation and can cause cell transformation [20,21]. The function of BRCA1 remains unknown, but strong evidence suggests a role in cell-cycle control and DNA repair [22,23]. Furthermore, BRCA1 expression is regulated by 17β-estradiol (E2) [24,25].
To address whether POCs could affect BRCA1 expression, we studied the effect on BRCA1 promoter activity using a transient reporter gene assay and the effect on the BRCA1 mRNA level by northern blot analysis in two human breast cancer cell lines. Furthermore, the effect on the estrogen-responsive gene (pS2) was studied, as well the effect on the cell-cycle regulator (p21 Waf/Cip1 ), in order to shed some light on the mechanistic pathways involved.

Plasmids
The plasmid pGL12 contains 267 bp, spanning -224 to +43 bp of the BRCA1 promoter-α, subcloned upstream of the luciferase gene [26]. The plasmid pHEO1 contains full-length human ERα cDNA [27]. The plasmid pON249 encodes β-galactosidase under control of a CMV promoter [28], and was used as the internal standard for transfection efficiencies.

Total RNA isolation
On day 1, 2.2 × 10 6 cells were seeded in 75 cm 2 culture flasks and grown for 24 hours in fetal calf serum containing media. The cells were starved in serum-free media on day 2 to obtain G 1 -synchronization, and they were exposed to chemicals in charcoal-depleted fetal bovine serum media on day 3. Total RNA was harvested after 22 hours of chemical exposure (PUREScript; Gentra Systems, Minneapolis, Minnesota, USA).

Northern blot and hybridization
A BRCA1 probe was produced by PCR on genomic DNA against exon 11 (5′-GAGAGGCATCCAGAAAAGTATC-AGG-3′ and 5′-CTCTGGGAAAGTATCGCTGTCATG-3′) of the BRCA1 gene (882-2216 bp of BRCA1 cDNA [17]). Twenty micrograms of total RNA was used for northern blot analysis (Zeta-probe membrane; Bio-Rad, Hercules, California, USA). The 32 P-labeled BRCA1 probe was made by random primed labeling (Megaprime; Amersham Pharmacia Biotech), and was hybridized overnight at 42°C in 50% formamide (as described by the manufacturer). The pS2-specific probe and the internal control (the GADPH probe) were used as described elsewhere [13,29]. The blots were visualized by a PhosphorImager and quantified with ImageQuant (Molecular Dynamics Amersham Biosciences, Sunnyvale, California, USA).

p21 Waf/Cip1 protein level
The p21 Waf/Cip1 protein was quantified on 24 hour exposure of MCF-7(BUS) to the given chemicals using the Waf1 Elisa kit (Oncogene Research Products, San Diego, California, USA).

Statistics
Statistical analysis was carried out in the Microsoft Excel standard diagram for independent samples test. P values were calculated using the two-tailed t-test two sample, assuming unequal variances.

Reporter gene assay
The dose-response effects of the chemicals on the BRCA1 promoter-luciferase construct (pGL12) in MCF-7(BUS) are presented in Table 1. In MCF-7(BUS), TCDD, ICI 182780, and the PCBs all significantly reduced the basal (ethanol) and the E2-induced promoter activity at the maximum concentrations tested (TCDD, 100 nM; ICI 182780 and PCBs, 10 µM). However, 10 µM toxaphene significantly induced basal promoter activity. TCDD and the pure ER antagonist ICI 182780 had the potential to decrease basal and E2-induced promoter activity at 100 and 10 pM, respectively, while PCB#138 had the potential to decrease E2-induced promoter activity at 100 pM. All chemicals (TCDD, 10 nM; ICI 182780 and PCBs, 10 µM) significantly reduced the BRCA1-promoter activity in MDA-MB-231. Cotransfection of the ER into MDA-MB-231 showed that the BRCA1 promoter is indeed induced by E2 (Table 1); however, only twofold induction occurred compared with the fivefold induction in MCF-7(BUS).

Northern blot analysis
To assess whether the findings in the reporter gene assay are also valid for genomic BRCA1 mRNA expression, northern blot analysis was carried out to measure the effect on the BRCA1 mRNA level with each compound at 10 µM, except for TCDD at 10 nM and ICI 182780 at 100 nM. Figure 1 shows a PhosphorImager image of BRCA1, pS2 and GADPH hybridizations to a northern blot. The analyzed data for BRCA1 are presented in Fig. 2. E2 induced BRCA1 mRNA expression twofold, and all compounds except toxaphene significantly decreased the E2-induced BRCA1 expression in MCF-7. Toxaphene increased the basal BRCA1 mRNA expression, while ICI 182780 downregulated basal and E2induced expression. No effect of the chemicals on BRCA1 mRNA expression was measured in the ER nonresponsive MDA-MB-231 cells.
To further address whether ER mediates the observed deregulation of BRCA1 mRNA, expression of the ER responsive gene pS2 was measured in parallel (Fig. 3a). A 2.5-fold induction of pS2 by E2 was observed, and a twofold induction by toxaphene confirmed its ER agonistic behavior. ICI 182780 significantly decreased basal and E2-induced pS2 expression, whereas TCDD significantly reduced E2-induced pS2 mRNA expression. The three PCBs did not significantly affect the pS2 expression.

p21 Waf/Cip1 study
To address whether the observed response could be a direct consequence of changes in cell-cycle regulation, we determined the effect of the POCs on the negative cell-cycle regulator p21 Waf/Cip1 . As shown in Fig. 3b, the POCs generally did not affect the p21 Waf/Cip1 protein level. Toxaphene, however, decreased the basal p21 Waf/Cip1 level like E2. Similar to ICI 182780, TCDD elicited an antiestrogenic effect on cotreatment with E2.

Discussion
BRCA1 mRNA expression is frequently downregulated in sporadic breast and ovarian cancers [19,20]. This altered regulation is generally not correlated with mutations within the gene [30]. The present study demonstrated that POCs in vitro had the potential to deregulate BRCA1 promoter activity and mRNA expression in human breast cancer cell lines. Using an E2-responsive construct carrying the key positive regulatory element of the BRCA1 promoter upstream of the luciferase gene [26,31], we found that TCDD, PCB#138, PCB#153, and PCB#180 can downregulate the basal as well as the E2-induced BRCA1 promoter activity in both the ER-positive cell line (MCF-7(BUS)) and in the ER-negative cell line (MDA-MB-231). However, toxaphene induced basal promoter activity ( Table 1).
The inhibiting effect of TCDD and the three PCBs were confirmed by northern blot analysis on the E2-induced BRCA1 mRNA level in MCF-7 (Fig. 2). E2 induced the BRCA1 mRNA level twofold, and the observed estrogenlike activity of toxaphene supports earlier findings [32,33]. At concentrations of 10 µM, the three PCBs had the capacity to downregulate both basal and E2-induced BRCA1 promoter activity, while they only decreased the E2-induced BRCA1 mRNA level. Furthermore, the PCBs antagonized E2-induced pS2 mRNA expression, but it was not statistically significant (Fig. 3a). The antiestrogenic potential of the three PCBs mediated via the ER, causing an antiproliferative effect in MCF-7, has previously been published by our laboratory [9].
Jeffy et al. found BRCA1 to be downregulated by benzo[a]pyrene in an ER-dependent fashion, probably through an aryl hydrocarbon receptor-mediated pathway [34]. Using RT-PCR, however, Jeffy et al. did not observe TCDD to downregulate BRCA1 in MCF-7. In the present study, TCDD significantly reduced both BRCA1 promoter activity as well as E2-induced BRCA1 (Fig. 2) and pS2 mRNA expression in MCF-7 ( Fig. 3a; see [13]).
All the POCs except toxaphene significantly decreased E2-induced BRCA1 mRNA expression in MCF-7. No effect was measured on the mRNA level in MDA-MB-231 (Fig. 2), which indicates the mechanism of action to be via the ER. In MDA-MB-231, however, the POCs also decreased the BRCA1 promoter activity, suggesting that other pathways than the ER-mediated pathway should be considered. We have shown that the three PCBs in MCF-7(BUS) are antiproliferative and anti-estrogenic at 10 µM [9], as has also been reported for TCDD [34], whereas toxaphene has been shown to increase MCF-7 proliferation [33,35]. In the present study, the negative cell-cycle regulator p21 Waf/Cip1 was not affected by the PCBs, while the effect of E2 was as expected [36] and toxaphene decreased the p21 Waf/Cip1 protein level. ICI 182780 and TCDD increased the E2 reduced level to its normal level (Fig. 3b). This suggests that the deregulation of BRCA1 observed in this study is not a simple consequence of altered cell-cycle regulation. Since the reporter gene assay also showed an effect of the POCs in MDA-MB-231, further research is needed to elucidate the pathways of the BRCA1 deregulation as well as the involvement of other regulatory sites outside of the 267 bp key regulatory element of the BRCA1 promoter cloned in the pGL-12 vector. In addition, a possible effect on BRCA1 expression of the chemicals in combination and in a mimic of the concentrations found in the adipose tissue should be addressed. (page number not for citation purposes) Table 1 Effect All values are corrected with β-galactosidase values (MCF-7(BUS)) or protein (MDA-MB-231) and then normalized to 17β-estradiol (E2)-induced cells (set to 100). The data are mean ± standard deviations of at least five or six measurements from two independent experiments. pGL12/ER, cotransfection with the pHEO1-expressing estrogen receptor alpha; Tox.Tech., toxaphene (technical mixture); TCDD, 2,3,7,8 tetrachlorodibenzo-pdioxin; PCB, polychlorinated biphenyl. * P < 0.05 relative to the given control (ethanol/DMSO or E2). (page number not for citation purposes)

Conclusion
It was found that TCDD, PCB#138, PCB#153, and PCB#180 decreased E2-induced BRCA1 promoter activity and mRNA expression in human breast cancer cell lines. Toxaphene induced the basal BRCA1 expression, probably via inducing cell proliferation since it was found to downregulate the cell-cycle protein p21 Waf/Cip1 . The presented data show that the ER is likely to play an essential role in the deregulation of BRCA1, but the data indicate a possible involvement of other pathways as well. The effect of these POCs on BRCA1 expression may, however, for a large part be a consequence of their anti-estrogenic, and hence antiproliferative, effect. By deregulating BRCA1, therefore, as has been found in sporadic breast cancers, the POCs could impair the DNA repair machinery, the cellcycle control and the stress-induced apoptosis [37]. POCs consequently could affect the overall risk of breast cancer. (page number not for citation purposes)