Germ-line mutations in the BRCA1 gene increase the risk of breast cancer in women, but the precise mechanistic basis for this connection remains uncertain. One popular hypothesis to explain breast tissue specificity postulates a link between BRCA1 and the action of the ovarian hormones estrogen and progesterone. Given the relevance of progesterone for normal mammary development and breast cancer formation, we searched for a functional relationship between BRCA1 and progesterone receptor (PR) in the PR-positive breast cancer cell line T47D. Here, we report that BRCA1 inhibits the transcriptional activity of PR by at least 2 mechanisms involving the E3 ubiquitin ligase activity of BRCA1. First, BRCA1 has a direct effect on the cellular level of PR and, hence, on the extent of PR recruitment to target promoters through the promotion of its ligand-independent and -dependent degradation. Through in vitro and in vivo assays, we found that BRCA1/BARD1 may be the main E3 ubiquitin ligase responsible for ubiquitination and degradation of PR in the absence of hormone. Second, after hormone treatment of cells, the BRCA1/BARD1 complex is recruited via interaction with PR to the hormone-responsive regions of PR target genes, affecting local levels of monoubiquitinated histone H2A and contributing to epigenetic silencing of these promoters. The connections between BRCA1/BARD1 and PR activity suggested by our findings may help explain why host mutations in BRCA1 exert a tissue specificity in preferentially elevating the risk of breast cancer. Cancer Res; 71(9); 3422–31. ©2011 AACR.
Germ-line mutations in the tumor suppressor gene BRCA1 confer an estimated 80% to 85% lifetime risk of developing breast cancer and 54% of ovarian cancer (1–3). Although BRCA1 mutation carriers present also a 2-fold increased risk of developing cervix, uterus, or pancreas cancer, the predominant breast tissue specificity has awakened the interest of oncologists and biologists.
BRCA1 encodes a mainly nuclear protein with 2 highly conserved domains: a RING domain at N-terminus and 2 BRCT motifs at C-terminus. Several cancer-predisposing mutations have been found within these regions, indicating that they are of critical importance for tumor suppression (3). The RING domain of BRCA1 possesses E3 ubiquitin ligase activity (4), which is dramatically enhanced by the heterodimerization with BARD1 (5). The in vivo target proteins of BRCA1/BARD1 and their significance are just starting to be identified. For instance, the largest subunit of RNAPII is ubiquitinated and sent to degradation (6, 7).
One of the most meaningful hypotheses to explain the tissue specificity of BRCA1-related cancers proposes a role for BRCA1 in regulating the activity of ovarian hormones, estrogen and progesterone, on their target organs. Both hormones act in the breast through the activation of signaling pathways and transcriptional programs (8) leading to profound changes in the proliferation and differentiation of the mammary gland. Accordingly, many known breast cancer risk factors correlate with increased exposure to ovarian hormones. Progesterone has a generally stimulatory role in breast cancer, both by stimulating the rate of proliferation and by serving as a survival factor (9, 10).
BRCA1 was shown to inhibit the activity of the estrogen receptor (ER; refs. 11, 12) and, recently, also of the progesterone receptor (PR; ref. 13). Here we confirm that BRCA1 regulates the transcriptional activity of PR by means of its E3 ubiquitin ligase activity by using a proteasome-dependent and a proteasome-independent mechanism. On one hand, BRCA1 enhances the degradation of PR protein through its ubiquitination, having a direct effect on PR recruitment to regulated promoters. On the other hand, BRCA1/BARD1 is recruited to hormone-responsive regions (HRR) of PR target genes and affects monoubiquitinated histone H2A (uH2A) levels, linking BRCA1 action with chromatin silencing at PR-regulated promoters. These findings support a functional connection between BRCA1 and PR activity in breast cancer cells, which may contribute to explain the particular susceptibility to breast cancer in BRCA1 mutant carriers.
Materials and Methods
R5020 was from Perkin Elmer. Doxycycline, lactacystin, and anti-FLAG antibody were from Sigma. E1 enzyme was from Calbiochem, E2 enzymes from Biomol, and ubiquitin from Upstate. BRCA1/BARD1 and I26ABRCA1/BARD1 proteins were kindly provided by Dr. Richard Baer (Institute for Cancer Genetics, NY). Adenovirus encoding wtBRCA1 were provided by Dr. Didier Marot (IGR, Villejuif, France). Antibodies to BRCA1 were from Oncogene (Ab1) and Santa Cruz (C20). PR (H190), ER (HC-20), BARD1 (H300), and α-tubulin antibodies were from Santa Cruz. Anti-H2A antibody was from Cell Signaling and anti–ubiquityl-H2A (uH2A) was from Upstate. Antibody anti–phospho-Ser294-PR was from NeoMarkers.
pcDNA3-wtBRCA1 was a gift from Dr. Barbara Weber (Novartis Institutes for Biomedical Research, Massachusetts) and pSG5-I26ABRCA1 was provided by Dr. Jeffrey Parvin (OSUMC, Columbus). pLVTHM, ptTR-KRAB-Red, pCMC-R8.91, and pMD.G were provided by Dr. Didier Trono (14). The plasmid for Ubiquitin was a gift from Dr. Timothy Thomson (IBMB, Barcelona, Spain).
T47D (from ATCC) and T47D-MTVL (T47D-M) were grown in RPMI 1640, 10% FBS, 2 mmol/L l-Glutamine, antibiotics and only in the case of T47D-M cells 700 mg/mL G418 (Invitrogen). T47DYV and its derivatives (T47DYV-empty and FLAG-PRB-T47DYV) were grown in MEM, 7% FBS, 2 mmol/L L-Glutamine, and antibiotics.
Cells were transfected with Lipofectamine Plus (Invitrogen) following manufacturer's instructions.
Whole cell extracts were prepared in SDS-lysis buffer, whereas histones were extracted as previously described (15). Equal protein amounts were resolved in 8 to 15% SDS-PAGE and transferred to nitrocellulose membranes, which were probed with indicated antibodies. Quantification of the signal was carried out with MultiGauge (FujiFilm).
Nuclear extracts were prepared as described (16) and were incubated with indicated antibody for 3 hours at 4°C. Protein G-beads (Sigma) were added and incubated overnight (o/n) at 4°C. Beads were washed extensively before being resuspended in SDS-loading buffer.
Quantitative real-time PCR
Total RNA was purified using RNeasy kit (Qiagen). cDNA was generated from 100 ng of RNA using Superscript First Strand system (Invitrogen). cDNA (1 μL) was used as the starting material for PCR. When indicated, quantitative real-time PCR (qRT-PCR) was carried out using LightCycler 480 SYBR Green Master Mix (Roche).
We used a lentiviral vector-mediated doxycycline-inducible system for the expression of short hairpin RNAs (shRNA) against BRCA1 in T47D-M cells. Cells were further transfected with an exogenous siRNA to assure a robust silencing. For the transient transfection of siRNAs we used Lipofectamine 2000 (Invitrogen).
Chromatin immunoprecipitation (ChIP) was carried out as described (17). Chromatin was sheared to an average 300 to 500 bp. Primer sequences and PCR conditions are available on request.
In vivo assays were carried out as follows. FLAG-PRB-T47DYV cells were transfected with ubiquitin and AdBRCA1. Cells were pretreated with lactacystin (10μmol/L, 1 hour) prior R5020 for 6 hours. Nuclear extracts were prepared and immunoprecipitated with anti-FLAG or control antibody. In vitro ubiquitination assays were carried out as described (18) using as substrate hormone-activated PRB isolated from baculovirus infected cells (19).
Results were analyzed with Student's t test. Differences with a P < 0.05 or P < 0.01 were regarded as significant.
BRCA1 regulates the downregulation of PR
First, we measured the effect of BRCA1 on PR levels in the cell line FLAG-PRB-T47DYV [a PR-negative clonal derivative cell line from T47D engineered to express a FLAG-tagged form of PRB (20)]. Cell extracts were prepared from control and from cells transduced to overexpress BRCA1 (AdBRCA1;Fig. 1A, top), and the level of PR protein was determined by Western blot. We observed that overexpression of BRCA1 decreased significantly PR protein levels in basal conditions as well as after 8 hours of treatment with the synthetic progestin R5020 (10 nmol/L), when the receptor is naturally downregulated (Fig. 1A, bottom). We corroborated this observation in BRCA1-depleted cells. T47D-M cells were stably transduced with an doxycycline-inducible RNAi system against BRCA1 and were further transfected with an siRNA to assure a robust silencing (Fig. 1B top and Supplementary Fig. S2A). We observed an increased expression of the 2 PR isoforms correlating with BRCA1 silencing (Fig. 1B, bottom). To test whether the increased PR levels are functional, we measured the levels of phospho-Ser294-PR, a modification diagnostic of ligand-activated PR (21). The levels of phospho-Ser294-PRB [PRA gets poorly Ser294-phosphorylated (22)] are increased in R5020-treated cells (10 minutes) depleted of BRCA1 (Fig. 1B), indicating that the excess of PR is normally activated by phosphorylation. The rise in protein level was not concomitant to an increase in PR mRNA synthesis (Fig. 1B, bottom).
To explore if BRCA1 affects the rate of hormone-induced PR degradation, control and BRCA1-depleted cells (Sh524 and Sh664) were treated with R5020 for 0 to 12 hours, and PR protein levels were monitored along time by Western blot and densitometry (Fig. 1C and D). We observed a higher basal level (time 0 hour) of PR protein in BRCA1-depleted cells compared to control cell lines. PR was extensively degraded at 8 to 12 hours of hormone treatment in control cells, reaching 50% of the starting value after 4.5 to 5 or 6.5 to 8 hours. In the 2 BRCA1-depleted cells, the hormone-induced degradation clearly takes place at a slower rate; the half-life of PR is extended to 7 to 8 hours in Sh524 and to 11 to 12 hours in Sh664 (Table 1). Altogether, the data indicate that BRCA1 affects the basal level of PR content and the progestin-induced degradation of the receptor.
BRCA1 promotes the ubiquitination of PR
Ligand-independent and -dependent degradation of PR is mediated by the 26S proteasome by attachment to polyubiquitin chains (21). To determine if BRCA1 might be regulating PR degradation through its ubiquitination, we carried out an in vivo ubiquitination assay. We overexpressed ubiquitin and AdBRCA1 and pretreated the cells with the proteasome inhibitor lactacystin for 1 hour before R5020 induction for 6 hours. Under these conditions, a ladder of high molecular weight species of PR corresponding to ubiquitinated intermediates, Ubn-PR, formed (Fig. 2A, lane 2). The ladder was further increased when BRCA1 was overexpressed (lane 3), meaning that BRCA1 enhances the formation of ubiquitinated forms of PRB in vivo.
To verify if BRCA1 is directly responsible for the ubiquitination of PR, we carried out an in vitro ubiquitination assay. PRB and BRCA1/BARD1 were purified from Sf9 cells infected with the corresponding baculovirus. We prepared a reaction mix containing PRB, BRCA1/BARD1, ATP, ubiquitin, E1 ubiquitin-activating enzyme and E2 ubiquitin-conjugating enzyme (UbcH5c). After 1 hour of incubation at 37ºC, ubiquitinated PRB was detected (Fig. 2B, lane 4). Only UbcH5c out of the 6 different E2 enzymes (23) tested was active in the ubiquitination of PRB. To exclude the possibility that ubiquitination was due to contaminating ubiquitin ligases, we also included the I26ABRCA1 mutant. The mutation I26A in the BRCA1 RING domain does not alter its tertiary structure but specifically disrupts the contact site with the E2 enzyme (24), losing the ubiquitin ligase activity. The I26ABRCA1 mutant heterodimer failed to ubiquitinate PRB in vitro (Fig. 2B, lane 3), suggesting that the ubiquitination of PRB observed is specifically dependent on the ubiquitin ligase activity of wtBRCA1/BARD1.
BRCA1 affects the level of PR recruitment to the progestin-regulated mouse mammary tumor virus promoter
We wondered if the observed effect on PR protein levels would influence the extent of PR recruitment to a target promoter. We carried out ChIP assays in T47D-MTVL (T47D-M) cells, which carry a stably integrated single copy of a mouse mammary tumor virus (MMTV)-Luciferase reporter (25, 26). In these cells, the MMTV promoter is covered by positioned nucleosomes, one of which (NucB) covers the hormone-responsive elements (HRE) to which PR is recruited on hormone treatment (27, 28). Control cells and cells overexpressing BRCA1 (AdBRCA1) were exposed to R5020 for 0 to 30 minutes and ChIP experiments were carried out to detect PR at the NucB region (Fig. 3). We observed a rapid recruitment of PR at 5 minutes that is maintained for 30 minutes. In cells overexpressing BRCA1, we detected a slight though consistent decrease in PR protein recruited (lanes 6–8), showing that BRCA1 affects the level of PR recruited to a target promoter.
As expected given the reduction in PR recruitment, overexpression of BRCA1 led to the inhibition of the R5020-induced expression of PR target genes like DUSP1, 11β-HSD or EGF (Supplementary Fig. S1), as previously reported (13). It also had a certain inhibitory effect on the basal expression of some genes. Contrarily, depletion of BRCA1 led to an enhancement in the R5020-induced gene expression (Supplementary Fig. S2B and D). We conclude that BRCA1 is mainly acting to reduce the level of hormonal gene activation.
BRCA1 inhibitory effect on PR is dependent on its E3 ubiquitin ligase activity
We wondered whether the ubiquitin ligase activity of BRCA1 was indispensable for the inhibitory effect of BRCA1 on PR activity. We overexpressed wtBRCA1 or the mutant I26ABRCA1 in T47D cells (Fig. 4A, top) and analyzed the expression of progesterone target genes. Overexpressed wtBRCA1 inhibited PR target genes induction, whereas the mutant did not (Fig. 4A, bottom). These results indicate that the E3 ubiquitin ligase activity of BRCA1 is required for the inhibition of PR transcriptional activity.
Next, we wanted to discern if the repressive effect of BRCA1 was also proteasome dependent. We overexpressed BRCA1 in T47D cells and pretreated the cells for 30 minutes with lactacystin before adding vehicle or R5020 for 4.5 hours. We observed 2 different responses to the proteasome inhibition. In the case of the DUSP1 gene, the lactacystin treatment partially reverted the inhibitory effect of BRCA1 overexpression (Fig. 4B, top graph). However, in the case of the 11β-HSD gene the treatment with lactacystin had no consequence (Fig. 4B, bottom graph). This would indicate that the BRCA1 inhibitory effect on PR activity can be proteasome dependent or independent in a gene-specific manner.
BRCA1 regulates the levels of monoubiquitinated histone H2A at a progestin-regulated promoter
We investigated in more detail the proteasome-independent mechanism of action of BRCA1. BRCA1 was previously found associated to the promoter of regulated genes, like ERα target genes (11). We wondered if BRCA1 would be present at HRRs of PR target genes and regulate transcription locally. To test this possibility, we carried out ChIP assays against BRCA1 in T47D-M cells on the MMTV promoter. We detected BRCA1 bound to the NucB region already before addition of hormone; the signal diminished 5 to 10 minutes after hormone addition but reached higher levels after 30 to 60 minutes or even at later time points (4 hours; Fig. 5A). The overexpression of BRCA1 increased its recruitment to the NucB region at 10 and 30 minutes of hormone treatment (Fig. 5B), whereas the recruitment of PR was decreased, as previously observed. As controls, chromatin immunoprecipitated with a control antibody (Fig. 5A, lane 9) and chromatin prepared from BRCA1-depleted and R5020-treated cells immunoprecipitated with anti-BRCA1 antibody (Fig. 5A, lane 13) did not yield any signal.
We also analyzed the presence of the BRCA1 protein partner, BARD1. BARD1 was absent from the NucB region before hormone addition, was recruited 10 minutes after hormone addition, and remained detectable along time (Fig. 5A, fourth panel). Thus, our data supports the recruitment of the BRCA1/BARD1 complex to the MMTV promoter within 10 to 30 minutes of hormone treatment.
We then wondered if PR could be the protein tethering the BRCA1/BARD1 complex to the promoter of target genes on hormone addition. We studied the interaction between BRCA1 and PR by coimmunoprecipitation in nuclear extracts from T47D-M cells grown in medium without hormones or in medium containing 10nmol/L R5020. Independently of hormone addition, BRCA1 was coprecipitated with PRB, PRA, and even ERα (Fig. 5C lane 2). Therefore, PR could target BRCA1 to the genomic PR binding sites that mediate progestin regulation. To test this possibility, we examined the presence of BRCA1 at different HRRs by ChIP in 2 cell lines: PR-negative T47DYV containing the MMTV-luc transgene and its derivative FLAG-PRB-T47DYV. We carried out ChIP analyses against FLAG-PRB and BRCA1 at the NucB and at the HRRs of 2 endogenous progesterone target genes: 11β-HSD (29) and EGF (Cecilia Ballaré, personal communication). In cells expressing FLAG-PRB, PR was found bound to these promoters 30 minutes after hormone treatment. The levels of BRCA1 at the HRRs were increased by R5020 addition in cells expressing FLAG-PRB (Fig. 5D, lane 4), whereas no signal was detected in the PR-null cell line (lanes 1 and 2). Therefore, our ChIP experiments indicate that the progestin-enhanced recruitment of BRCA1 is PR dependent, in agreement with the immunoprecipitation assays showing the physical interaction of the 2 proteins.
Given that BRCA1 is present at the promoters of progestin-regulated genes, we considered the possibility that BRCA1/BARD1 could modify histones by ubiquitination. In humans, H2A ubiquitination is mediated by at least 2 ubiquitin ligases, Ring1B and 2A-HUB, both associated with transcriptional silencing (30, 31). BRCA1 itself can catalyze the ubiquitination of H2A in vitro, although its in vivo relevance has not been elucidated (32). First, we studied if BRCA1 affected the total levels of monoubiquitinated histone H2A (uH2A) in T47D cells. Total histones were acid-extracted from control and BRCA1 (Supplementary Fig. S2C) or BARD1-depleted cells and were probed with an antibody against total histone H2A and an antibody recognizing monoubiquitinated histone H2A (uH2A). We observed no significant changes in the total amount of uH2A (Fig. 6A). However, we analyzed the effect of BRCA1 depletion on the local instead of total levels of uH2A by carrying out ChIP assays at a regulated promoter (Fig. 6B). We detected the presence of uH2A at the NucB of the MMTV promoter before hormone induction. This repressive mark disappears 5 minutes after hormone addition and returns after 60 minutes, just coinciding with the recruitment of BRCA1. Moreover, the presence of uH2A at the HRR is markedly decreased following depletion of BRCA1. These data indicate that BRCA1 affects the level of uH2A at the MMTV promoter, and this effect might be implicated in the proteasome-independent transcription regulation exerted by BRCA1 at PR target genes.
PR requires local remodeling of chromatin to get access to less exposed HREs 2 and 3 at the NucB region (19). As reported by Vicent and colleagues (33), this remodeling includes the displacement of histones H2A/H2B dimers at 30 minutes of hormone treatment. We wanted to examine the displacement of histone H2A at the NucB in the context of BRCA1 depletion and the accompanying decrease in H2A monoubiquitination. For this purpose, ChIP assays were carried out at nucleosome resolution with anti-H2A antibody in control and BRCA1-depleted cells (Fig. 6C). We observed the expected displacement of histone H2A at 30 minutes of hormone treatment in control cells. The signal was appreciably decreased in BRCA1-depleted cells before and after hormone addition, indicating that BRCA1 may preclude histone H2A displacement from the NucB region and, thus, interfere with factor accessibility and transcriptional activation.
As a possible explanation for the tissue specificity of tumors associated with BRCA1 mutations, BRCA1 was shown to inhibit the transcriptional activity of ERα and PR in breast cancer cell lines (11–13). We have confirmed that BRCA1 overexpression results in the inhibition of endogenous PR target promoters, whereas BRCA1 depletion leads to upregulation of progestin-induced genes. Our findings point to 2 ubiquitination-related mechanisms by which BRCA1 interferes with the transcriptional activity of PR. Ubiquitination may also be involved in the BRCA1 inhibition of ERα function since mutants disrupting the RING domain of BRCA1 (containing the ubiquitin ligase activity) failed to inhibit ERα transcriptional activity (34).
BRCA1 has a direct effect on the basal amount of PR protein and affects the rate of hormone-triggered degradation of the receptor. Our in vivo and in vitro experiments showed that the BRCA1/BARD1 heterodimer specifically mediates the polyubiquitination of PR with the E2 enzyme UbcH5c. The usage of this specific E2 enzyme, with a trend toward polyubiquitination, might be substrate specific and determine protein fate, that is, protein degradation. To date, only the protein CUEDC2 has been described as directly implicated in the ubiquitination of PR for its subsequent hormone-driven degradation (35). However, it is unclear how unliganded PR is targeted for degradation. In its unliganded form, PR forms a complex with several heat-shock proteins, which maintain the unstable ligand binding conformation (36). The disruption in the assembly of the mature complex by the use of geldanamycin leads to the rapid proteasome-driven degradation of the receptor (21). We propose BRCA1/BARD1 as the main E3 ubiquitin ligase responsible for the degradation of unstable PR-chaperone complexes, and as a secondary enzyme for the degradation of the ligand-bound receptor. Consequently, the lack of functional BRCA1 results in the accumulation of PR protein and increased recruitment to target promoters. Our data strongly supports this mechanism of action instead of the sequestering of PR molecules proposed in another study (37).
In addition, BRCA1/BARD1 is present at the HRRs of PR target genes. Basal BRCA1 could be tethered to the promoter by the basal transcription machinery (6), where it could exert control over the basal transcription of the genes, as well. As BRCA1 and PR interact, we propose that liganded PR drives the progestin-enhanced recruitment of BRCA1/BARD1 to replace preexisting BRCA1 at HREs. In this way, hormone treatment increases the ubiquitin ligase activity at target promoters. Similarly, BRCA1 is associated to the pS2 gene promoter prior to estradiol treatment, leaves the promoter on hormone addition (11) and returns at later times (38). Mutations located at the interacting surface between BRCA1 and ERα disrupted the ability of BRCA1 to bind to and repress ERα transcriptional activity. Therefore, as in the case of BRCA1-ERα, we propose that the interaction BRCA1-PR is also functionally important.
BRCA1 was shown to drive the ubiquitination of histones H2A and H2B in vitro as isolated proteins and on the nucleosome particle (32, 39). Here we show that BRCA1 ubiquitinates histone H2A on the promoter of a PR-regulated gene. Most importantly, the level of uH2A correlates inversely with the local displacement of histone H2A/H2B dimers. How H2A ubiquitination leads to gene repression is largely unknown. The data by Zhou and colleagues (31) suggested that regulation of H2A ubiquitination occurs in a gene- and enzyme-specific fashion. They propose that histone H2A ubiquitination serves to pause RNAPII at the promoter-proximal region by preventing the recruitment of FACT, an elongation factor that acts as a histone chaperone and displaces dimers of histones H2A/H2B. This idea is consistent with our observations. Displacement of H2A/H2B dimers is needed for the transcriptional activation of some PR target genes (33), such as the MMTV promoter.
The relative weight of the proteasome-dependent versus the proteosome-independent mechanism of action would depend on the combination of pathways used by PR in regulating different target genes. For instance, the 11β-HSD gene is regulated by PR directly through the STAT5A-mediated recruitment of PR to a distal promoter region and through signaling activation (29), whereas DUSP1 is probably induced indirectly, through the activation of signaling pathways and secondary transcription factors (20). It should be emphasized that the effect of BRCA1 on PR transcriptional activity is not caused by nonspecific effects on the RNAPII since other genes like GAPDH (glyceraldehyde 3 phosphate dehydrogenase) are not affected by BRCA1.
Our findings provide 2 mechanisms for how the ubiquitin ligase activity of BRCA1 participates in the biological functions attributed to BRCA1 (4, 5). Which would be the biological implications of the relationship BRCA1-PR in the mammary gland? Progesterone, along with other factors, regulates mammary gland development. It was proposed the existence of steroid receptor-positive progenitor stem cells (40), which would respond to ovarian hormones by secreting paracrine factors to feedback on steroid receptor-negative cells and induce their proliferation (41). In a BRCA1 mutant scenario, steroid receptor-positive cells would respond exaggeratedly to ovarian hormones, which would induce increased proliferation in surrounding negative cells, besides exposing these cells to the risks due to the lack of functional BRCA1. The steroid receptor-negative stem cells would then be good candidates for starting a tumorigenic overgrowth. The protective effect of ovariectomy on breast cancer development in BRCA1 mutant carriers (42) might be due to the elimination of these paracrine proliferative signals. In effect, the deficiency of Brca1 confers an exaggerated progesterone-induced growth response in the mammary glands of adult female mice (43). This could explain, as well, the steroid receptor negativity of BRCA1-related tumors since the bulk of cells composing the tumor would originate from those primitive negative cells.
In conclusion, we show that BRCA1 regulates PR transcriptional activity and propose a double mechanism for this function. By means of the control of PR turnover BRCA1 controls the level of PR recruitment to the promoter of specific target genes. Through its presence at the promoter of PR-regulated genes, BRCA1 can ubiquitinate histone H2A serving as a signal for transcription repression. We defend that one of the reasons for the tissue specificity associated with BRCA1-related cancer is the lack of inhibition of progesterone action, as it represents one of the most important hormones regulating the mammary gland biology.
Disclosure of Potential Conflicts of Interest
No potential conflicts of interest were disclosed.
This work was supported by Scientific Foundation of the Spanish Assotiation Against Cancer, the Spanish Ministry for Science and Innovation (BMC 2003-02902 and Consolider CSD2006-00049); Departament d'Innovació Universitat i Empresa of the Generalitat de Catalunya; Carlos III Institute of Health, Spanish Ministry for Science and Innovation (V. Calvo).
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
The authors thank D. Trono (GHI, Lausanne, Switzerland) for providing the inducible siRNA expression system; R. Baer and S. Wang (ICG, NY) for providing BRCA1/BARD1 protein and sharing their knowledge; D. Marot for the AdBRCA1 adenovirus; K. Horwitz (University of Colorado, Denver) for providing T47DYV cells; C. Ballaré, J. Clausell, and A. Subtil-Rodríguez for helpful discussion.
Note: Supplementary data for this article are available at Cancer Research Online (http://cancerres.aacrjournals.org/).
- Received October 8, 2010.
- Revision received February 25, 2011.
- Accepted March 3, 2011.
- ©2011 American Association for Cancer Research.