Report

BRCA1 Inhibition of Estrogen Receptor Signaling in Transfected Cells

See allHide authors and affiliations

Science  21 May 1999:
Vol. 284, Issue 5418, pp. 1354-1356
DOI: 10.1126/science.284.5418.1354

Abstract

Mutations of the breast cancer susceptibility gene BRCA1confer increased risk for breast, ovarian, and prostatic cancers, but it is not clear why the mutations are associated with these particular tumor types. In transient transfection assays, BRCA1 was found to inhibit signaling by the ligand-activated estrogen receptor (ER-α) through the estrogen-responsive enhancer element and to block the transcriptional activation function AF-2 of ER-α. These results raise the possibility that wild-type BRCA1 suppresses estrogen-dependent transcriptional pathways related to mammary epithelial cell proliferation and that loss of this ability contributes to tumorigenesis.

Germ line mutations of theBRCA1 gene (17q21) (1) account for 40% to 50% of hereditary breast cancers and confer increased risk for ovarian and prostatic cancers (2). The BRCA1 gene encodes an 1863-amino acid protein with a highly conserved NH2-terminal RING finger domain and a COOH-terminal acidic transcriptional activation domain (1, 3). The product of theBRCA1 gene is a 220-kD nuclear phosphoprotein (4) that has been implicated in regulation of cell proliferation, cell cycle progression (4, 5), apoptosis induction (6, 7), and DNA repair and recombination (7, 8). These functions of BRCA1have been observed in various human epithelial cancer cell types and mouse fibroblasts and do not explain the association ofBRCA1 mutations with specific tumor types, such as breast cancer. Estrogen stimulation of mammary epithelia is thought to be a major factor in promoting development of breast cancer, probably through the expansion of previously initiated mammary epithelial cell clones (9). Thus, we hypothesized that BRCA1 protein might function, in part, to regulate the cellular response to estrogen.

Estrogen responses are mediated by two closely related members of the nuclear receptor family of ligand-activated transcription factors, ER-α and ER-β [reviewed in (10)]. The 66-kD ER-α, the major estrogen receptor of mammary epithelia, contains NH2- and COOH-terminal transcriptional activation functions (AF-1 and AF-2, respectively), a DNA binding domain, a ligand binding domain associated with AF-2, and a cofactor binding domain. Estrogens bind and activate cytoplasmic ER-α, which translocates to the nucleus, dimerizes, binds to estrogen-responsive enhancer elements (EREs), and activates transcription. ER-α activation requires a ligand and cofactor-dependent conformational change that allows interaction between AF-1 and AF-2 and recruitment of coactivator proteins. These coactivators link ER-α to the basal transcriptional machinery and convert chromatin to the active state by histone acetylation.

To assess the effect of BRCA1 on estrogen response, we measured the ability of the BRCA1 gene to modulate the transcriptional activity of ER-α in transient transfection assays. Human prostate cancer cell line DU-145 lacks ER-α, but, when supplied with exogenous ER-α, these cells became competent for estrogenic signaling, as demonstrated by estradiol (E2)-stimulated activation of estrogen-responsive reporter ERE-TK-Luc (Fig. 1A). Cotransfection of a wild-typeBRCA1 gene (wtBRCA1) abolished activation of this reporter in a dose-dependent fashion, whereas empty BRCA1vector (pcDNA3) had little or no effect on reporter activity (Fig. 1A). Equally strong inhibition of ER-α signaling by wtBRCA1 was observed in breast cancer cell line T47D, in two other human breast cancer cell lines (MCF-7 and MDA-MB-231), and in two other human prostate cancer cell lines (LNCaP and TsuPr-1) (Fig. 1, A and B). However, three human cervical cancer cell lines showed relatively weak inhibition of ER-α signaling by wtBRCA1: E2-stimulated reporter activity in the presence of wtBRCA1 was about 40% to 50% that in the presence of empty vector (Fig. 1B).

Figure 1

BRCA1 selectively inhibits ER-α signaling in transfected human cancer cells in culture. (A) Wild-type BRCA1 gene (wtBRCA1) strongly inhibits ER-α signaling in prostate (DU-145) and breast cancer (T47D) cells. Cells were assayed for stimulation of ERE-TK-Luc reporter activity by 17β-estradiol (E2) (13). ERE-TK-Luc is composed of the vitellogenin A2 ERE controlling a minimal thymidine kinase promoter (TK81) and luciferase in plasmid pGL2 (11). E2 induced a 100- to 1000-fold increase in reporter activity when expression vector pSG5-ER-α (ER-α) was provided. Cotransfection of wtBRCA1 expression vector (13), but not empty vector (pcDNA3), caused dose-dependent inhibition of the E2-stimulated reporter activity in DU-145 and T47D cells. A control reporter missing the ERE (TK-Luc) gave no E2-stimulated activity. (B) wtBRCA1 weakly inhibits ER-α signaling in human cervical cancer cells. In three cervical cancer cell lines (C33A, CaSki, SiHa), wtBRCA1reduced the E2-stimulated ERE-TK-Luc reporter activity to 38% to 51% of control. In contrast, wtBRCA1 nearly abolished reporter activity in two other prostate (LNCaP, TsuPr-1) and two other breast (MCF-7, MDA-MB-231) cancer cell lines. (C)wtBRCA1 inhibits ER-α activation function AF-2. DU-145 and T47D cells were transfected with pCMV-gal4-ER (gal4-ER), which encodes the yeast gal4 DBD fused to the ER-α AF-2/ligand binding domain, and reporter gal4-Luc. Gal4-Luc is composed of a gal4 site (pentameric upstream activating sequence) controlling a minimal promoter (adenovirus E1B TATA box) and luciferase (11). In the presence of E2, gal4-Luc was activated by gal4-ER but not by control plasmid pCMV-gal4 (gal4). Gal4-ER/gal4-Luc activity was ablated bywtBRCA1 but not by empty pcDNA3 vector. A control reporter missing the gal4 site (TATA-Luc) yielded no luciferase activity. (D) wtBRCA1 does not inhibit E2F1 or Sp1 signaling. The effect of wtBRCA1 on E2F1 and Sp1 activity was tested with expression vectors pCMV-E2F1 (E2F1) and pCMV-Sp1 (Sp1) and reporters E2F-TK-Luc and Sp1-TK-Luc, respectively. These reporters are composed of the E2F site from adenovirus E2a and the Sp1 site from the cyclin D1 promoter (−127 to −99), respectively, linked to TK81 and luciferase (14). Reporter activity was stimulated by E2F1 or Sp1 vector. Neither E2F1- nor Sp1-stimulated reporter activity was inhibited by wtBRCA1. (E)wtBRCA1 does not inhibit the transcriptional activation functions of c-Jun and E2F1. Expression plasmids encoding gal4 DBD linked to the activation domains of c-Jun and E2F1 (15) activated the gal4-Luc reporter, but wtBRCA1 failed to inhibit gal4-jun and gal4-E2F1 stimulated reporter activity.

We used the gal4-ER/gal4-Luc assay system (11) to determine whether BRCA1 directly inhibits the highly conserved COOH-terminal activation function (AF-2) of ER-α, independently of any effect it might have on the DNA binding activity of the receptor. When linked to the yeast gal4 DNA binding domain (DBD), the AF-2 domain of ER-α stimulated activity of a reporter containing a multimeric gal4 binding site in an E2-dependent fashion (Fig. 1C). In both DU-145 and T47D cells, wtBRCA1 reduced gal4-ER/gal4-Luc activity to background levels, whereas pcDNA3 vector had little or no effect on reporter activity. These findings suggest that BRCA1 inhibits the AF-2 function of ER-α, but they do not rule out the possibility that BRCA1 targets additional steps in ER-α activation.

In both the ER-α and gal4-ER assays, basal reporter activity was very low, so it was not possible to determine whether BRCA1 acted as a nonspecific transcriptional repressor. We tested the ability ofwtBRCA1 to regulate the activity of other transcription factors. Expression vectors for cell cycle–regulated transcription factors E2F1 and Sp1 caused activation of luciferase reporters controlled by the corresponding binding elements, but thewtBRCA1 gene did not inhibit activation of either reporter in DU-145 or T47D cells (Fig. 1D). wtBRCA1 also failed to repress activation of the gal4-Luc reporter by vectors encoding the gal4 DBD linked to the transcriptional activation domains of c-Jun and E2F1 (Fig. 1E). In fact, in some of these assays, reporter activity was slightly higher in the presence of wtBRCA1. These results suggest that BRCA1 is not an indiscriminate repressor of transcription.

Previous studies have linked expression of theBRCA1 gene with differentiation of specific cell types, including mammary epithelium (12), but the molecular targets of BRCA1 for mammary differentiation have not been identified. Our findings suggest that the wild-type BRCA1 protein may function, in part, to suppress estrogen-dependent mammary epithelial proliferation by inhibiting ER-α-mediated transcriptional pathways related to cell proliferation. Conversely, inactivation of the ability to repress ER-α transcriptional activity by mutation or by loss ofBRCA1 expression may contribute to mammary carcinogenesis. Interestingly, the relatively weak inhibition of ER-α signaling in human cervical cancer cell lines by the wtBRCA1gene correlates with the absence of an association of cervical cancer, another estrogen-dependent tumor type, with germ line mutations ofBRCA1. To establish the validity of these hypotheses, it is necessary to confirm the ability of BRCA1 to regulate the ER-α response in in vivo models.

REFERENCES AND NOTES

View Abstract

Navigate This Article