Induction of the BRCA2 Promoter by Nuclear Factor-κB*

BRCA2 is a tumor suppressor gene that has been implicated in response to DNA damage, cell cycle control, and transcription. BRCA2 has been found to be overexpressed in many breast tumors, suggesting that altered expression of the BRCA2 gene may contribute to breast tumorigenesis. To determine how BRCA2 is overexpressed in tumors, we investigated the transcriptional regulation of the BRCA2 promoter. Deletion mapping of the BRCA2 promoter identified three regions associated with 3-fold activation or repression and one upstream stimulatory factor binding site associated with 20-fold activation. Gel shift and cotransfection studies verified the role of USF in regulation of BRCA2 transcription. Analysis of the −144 to −59 region associated with 3-fold activation identified a putative NFκB binding site. Cotransfection of the p65 and p50 subunits of NFκB up-regulated the BRCA2 promoter 16-fold in a luciferase reporter assay, whereas mutations in the binding site ablated the effect. Gel shift and supershift assays with anti-p65 and -p50 antibodies demonstrated that NFκB binds specifically to the NFκB site. In addition, ectopic expression of NFκB resulted in increased levels of endogeneous BRCA2 expression. Thus, NFκB and USF regulate BRCA2 expression through the BRCA2 promoter.

BRCA2 is a tumor suppressor gene associated with familial predisposition to breast and ovarian cancer (1,2). Mutations in BRCA2 are thought to account for 20 -35% of all inherited breast cancers and are associated with a 37-85% lifetime risk of developing cancer (3,4). The great majority of disease-associated mutations in BRCA2 result in truncation of the BRCA2 protein, suggesting that loss of function of BRCA2 results in tumor susceptibility. However, the mechanisms by which the BRCA2 protein suppresses tumor cell growth are largely unknown.
The BRCA2 gene encodes a 3418-amino acid nuclear protein (2,5), that has been implicated in the cellular response to DNA damage. BRCA2 interacts directly with RAD51, a protein involved in meiotic and mitotic recombination, DNA doublestranded break repair, and chromosome segregation (6,7), through the BRC repeats and a C-terminal binding site. BRCA2 Ϫ/Ϫ animals die as early embryos (8 -11), and viable BRCA2 Ϫ/Ϫ early mouse embryos are highly sensitive to ␥-irradiation-induced DNA damage (9). Moreover, cells expressing mutant BRCA2 are more sensitive to methyl methanesulfonate-induced DNA damage than cells expressing wild type BRCA2 (12), and BRCA2 appears to be required for ionizing radiation-induced assembly of a RAD51 protein complex in vivo (13).
BRCA2 may be also involved in regulation of the cell cycle and genome instability. BRCA2 is expressed in a cell cycle-dependent manner with peak expression in the S and G 2 phases of the cell cycle. Low levels of expression are detected in G 0 , G 1 , and M phase (14). Cell cycle-dependent expression has recently been associated with binding of the upstream stimulatory factor (USF) 1 protein and Elf-1 transcription factor to the BRCA2 promoter (15). In addition, BRCA2 expression is elevated indirectly in response to the mitogenic activity of estrogen, which has been associated with progression of the cell cycle (16,17). Furthermore, recent studies of BRCA2 Ϫ/Ϫ mouse embryo fibroblasts identified extensive chromosomal rearrangement, centrosome amplification, and aneuploidy, consistent with abrogation of a mitotic checkpoint (18). Likewise, tumor cells expressing mutant BRCA2 have been shown to contain multiple chromosomal rearrangements (19). These data suggest that BRCA2 plays a key role in regulation of cell growth and proliferation in many cell types.
Several studies have attempted to define a role for BRCA2 in development of sporadic breast cancer (20 -25). Loss of heterozygosity of the BRCA2 locus has been detected in over 50% of sporadic breast tumors, suggesting a role for BRCA2 in sporadic breast cancer development (20 -22). However, no somatic mutations of BRCA2 have been found in sporadic breast cancers (23,24). Also, the BRCA2 promoter is not inactivated by methylation in breast tumors (25). Although no sequence alterations have been found in the BRCA2 gene in sporadic tumors, it remains possible that BRCA2 does contribute to sporadic breast cancer development, albeit not by inactivation of the BRCA2 protein through mutagenesis and methylation. One possible mechanism of BRCA2 involvement is through deregulated expression of the BRCA2 gene. Recently, it has been shown that BRCA2 is significantly overexpressed in many sporadic breast cancers (26). It is not known whether this overexpression of BRCA2 is due to induction of the BRCA2 promoter or is a result of an increased number of cells in S phase of the cell cycle. However, when combining this observation with the known relevance of BRCA2 function to regulation of cell proliferation, it seems likely that expression of the BRCA2 gene is tightly regulated and that altered expression of BRCA2 may contribute to breast cancer development.
To begin to assess the contribution of altered expression of * This work was supported by Grant DAMD-97-7048 (to F. J. C.) from the United States Army Medical Research and Materiel Command. 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.
ʈ BRCA2 to breast tumorigenesis, we investigated the transcriptional regulation of the BRCA2 promoter. Here we provide evidence for direct induction of the BRCA2 promoter through binding of the nuclear factor-B (NFB) transcription factor, and we verify the role of USF in regulation of basal activity of the promoter.

EXPERIMENTAL PROCEDURES
Cell Culture-Human breast adenocarcinoma MCF-7 cells were obtained from American Type Culture Collection, propagated in the Dulbecco's modified Eagle's medium supplemented with 10% bovine calf serum (HyClone) and maintained at 37°C with 5% CO 2 . Cell culture reagents were obtained from Life Technologies, Inc.
BRCA2 Reporter Constructs-A BAC clone (B489G) containing the 5Ј end of the BRCA2 gene was isolated from a BAC library (27) using a polymerase chain reaction (PCR)-generated hybridization probe consisting of bases 72-560 of the BRCA2 cDNA. B489G DNA was digested with SacI and PstI enzymes, and the resulting fragments were subcloned into the pGL3 basic vector containing a firefly luciferase reporter gene (Promega) and plated. Colonies containing the 5Ј end of the BRCA2 gene were identified by hybridization with the 72-560-bp cDNA probe. Plasmid DNA from positive colonies was prepared and sequenced using vector specific primers. Sequences were then matched against the complete genomic sequence of this region in GenBank TM . A clone with an 8-kb insert (pGL3Prom) was found to include 4.3 kb of sequence upstream of the putative BRCA2 transcription start site and 3.7 kb downstream of the transcription start site including exons 1, 2, and 3 of BRCA2. The entire 8-kb insert was then sequenced by the Molecular Biology Core of the Mayo Clinic.
Deletion Mutants of the BRCA2 Promoter-A series of deletion mutants (see Figs. 1 and 2) of the BRCA2 promoter were generated by restriction enzyme digestion with a variety of restriction enzymes followed by religation and also by direct PCR amplification. The Del-1 construct was generated by digesting the pGL3Prom construct with HindIII and PstI and religating the pGL3Prom plasmid. Del-2 resulted from religation following digestion with MluI and PstI. Del-9 was generated by subcloning a 1249-bp fragment of pGL3Prom, resulting from KpnI and MluI digestion, into the pGL3 basic promoter. Del-2 was then digested by combinations of SacI with NdeI, HindIII, EcoRI, and BbrPI, and the linearized plasmids were blunt-ended with Klenow enzyme (New England BioLabs) and religated to form Del-3, Del-4, Del-5, and Del-16, respectively.
Point Mutants of the BRCA2 Promoter-Site-directed mutagenesis of the Del-15 construct was performed using the QuikChange site-directed mutagenesis kit (Stratagene) to prepare constructs containing mutations in predicted cis-elements within the promoter. Specifically, mutations were introduced into putative DNA-binding sites for the ATF, USF, MLTF, and c-Myc transcription factors (see Fig. 2). Mutations were confirmed by DNA sequencing.
Luciferase Reporter Assays-Plasmid DNA for transient transfection was isolated using the plasmid maxi kit (Qiagen). MCF-7 cells were plated at a density of 1 ϫ 10 5 cells/well of 6-well plates and grown in Dulbecco's modified Eagle's medium with 10% bovine calf serum overnight prior to transfection. All transfections were carried out using Fugene-6 (Roche Molecular Biochemicals) according to the manufacturer's instructions. A total of 2 g of BRCA2 promoter construct and 0.1 g of pRL-TK Renilla luciferase vector (Promega) with 4 l of Fugene-6 was used for each transfection. The pRL-TK Renilla luciferase activity was used to control for transfection efficiency. Each transfection experiment was performed in duplicate and repeated a minimum of three times. For cotransfection experiments, cells received 0.5 g of BRCA2 promoter construct, 0.1 g of pRL-TK Renilla luciferase vector, and 0.5 g of the indicated expression plasmids and carrier DNA. Expression plasmids included pCMV-USF, pCMV-USF-VP16, pCMV-VP16, pCMV, pcDNA3.1-p65, pcDNA3.1-p50, pcDNA3.1, pCMV-CREB, pCMV-Myc, and pCMV-Max. Firefly luciferase and Renilla luciferase assays were performed using the Dual-Luciferase Reporter Assay System (Promega). Approximately 48 h after transfection, cells were washed twice with 1ϫ phosphate-buffered saline and harvested with 600 l of passive lysis buffer (Promega). Cell lysates were cleared by centrifugation, and 5 l was added to 100 l of firefly luciferase substrate, and light units were measured in a luminometer. Renilla luciferase activities were measured in the same tube after addition of 100 l of Stop and Glo reagent.
Electrophoretic Mobility Shift Assays-Double strand oligonucleotides generated from the single strand oligonucleotides listed in Table I and II were used as electrophoretic mobility shift assay probes. The upper strand (sense) oligonucleotide (30 ng) was 5Ј end labeled using polynucleotide kinase with [␥-32 P]dATP (Amersham Pharmacia Biotech). After the labeling reaction, 2-fold excess of lower strand (antisense) oligonucleotide was annealed to the upper strand. Doublestranded DNA probes were purified from the reaction mixture using a Bio-Gel P-100 column (Bio-Rad). Whole cell extracts were isolated from cultured MCF-7 cells. DNA-protein binding was performed in 0.5ϫ Dignam buffer D (20 mM HEPES, pH 7.9, 100 mM KCl, 20% glycerol, 0.2 mM EDTA) supplemented with 0.5 mM phenylmethylsulfonyl fluoride, 0.5 mM dithiothreitol, 10 mM MgCl 2 , and 100 g/ml poly(dI-dC). Binding reactions were initiated by addition of 30,000 cpm DNA probe in TE buffer (10 mM Tris⅐HCl, pH 7.5, 1 mM EDTA) to 5-10 l of whole cell extracts. Electrophoresis was performed in acrylamide gels, gels were dried and exposed to film for 16 -48 h.
Competition experiments were carried out in the same way as described above except that increasing amounts of double-stranded wild type oligonucleotide were mixed with 30,000 cpm of M-1 probe (see Table I) and added to the binding reaction. For optimized antibody mediated supershift experiments, increased DNA probe (60,000 cpm) and decreased whole cell extracts (5 l) were applied. The binding reaction included 1-4 l of antibodies against ATF2 (Santa Cruz Biotech), c-Myc (Santa Cruz Biotech) or USF-1 (kindly provided by Dr. Michele Sawadogo, M.D. Anderson Cancer Center).
DNA binding assays for NFB were also performed using electrophoretic mobility shift assays. Whole cell extracts were prepared from MCF-7 cells 48 h after transfection with pcDNA 3.1 or NFB p65 and p50 subunit expression constructs. Components of NFB proteins were identified by supershift assay using antibodies against p50 and p65 (Santa Cruz Biotech).
Western Blotting-48 h after transfection with pcDNA 3.1 or p50 and p65 expression constructs, MCF-7 cells were washed with 1ϫ phosphate-buffered saline, and cell lysates were prepared with RIPA buffer containing COMPLETE proteinase inhibitor mixture (Roche Molecular Biochemicals). Equal amounts of protein lysate from each transfection were subjected to electrophoresis, transferred to membrane, and probed with primary antibodies and alkaline phosphatase-conjugated secondary antibodies (Jackson ImmunoResearch Laboratories). Signals were developed by ECL detection system.
RNA Isolation and Northern Blotting-Total RNA was isolated from MCF-7 cells 48 h after transfection with pcDNA 3.1 or p50 and p65 expression constructs, or pcDNA3.1, dominant negative IB␣ 32A/36A (dn-IB␣), NFB p65, IB␣ wild type, p65 with dn-IB␣, or p65 with wild type IB␣ using TRIzol reagent (Life Technologies, Inc.), according to the manufacturer's instructions. Total RNA samples (20 g/lane) for pcDNA 3.1, p50, and p65 transfected cells were resolved on 0.8% agarose-formaldehyde gels and transferred to nylon membranes. The membranes were prehybridized at 62°C for 1 h in ExpressHyb Hybridization Solution (CLONTECH) and then hybridized for 1 h in the same solution with [␣-32 P]ATP random labeled full-length human BRCA2 cDNA. After hybridization, the membrane were washed (three times for 15 min each time at room temperature) with 2ϫ SSC, 0.05% SDS and washed (three times for 15 min each time at 62°C) with 0.5ϫ SSC, 0.1% SDS. Membranes were then exposed in a PhosphorImager. Each membrane was also hybridized as described above with a GAPDH probe for normalization of mRNA levels.
Semi-quantitative Reverse Transcription-PCR Analysis-1 g of total RNA from cells transfected with pcDNA3.1, dn-IB␣, NFB p65, IB␣ wild type, p65 with dn-IB␣, or p65 with wild type IB␣ was used for preparation of cDNA with random hexamer primers and superscript II reverse transcriptase (Life Technologies, Inc.). After treatment with DNase, 2 l from a total of 100 l was used for semi-quantitative PCR with BRCA2 and GAPDH PCR primers. The sequences of forward (F) and reverse (R) PCR primers were as following: BRCA2, 5Ј-GCAGT-GAAGAATGCAGCAGA-3Ј (F, within the exon 21 of human BRCA2) and 5Ј-CAATACGCAACTTCCACACG-3Ј (R, within the exon 22 of human BRCA2); GAPDH, 5Ј-CAACTACATGGTTTACATGTTC-3Ј (F) and 5Ј-GCCAGTGGACTCCACGAC-3Ј (R). Each PCR amplification was performed using Taq DNA polymerase (Promega) with both PCR primers for BRCA2 and GAPDH under the following conditions: 1 cycle for 2 min at 94°C; 25 cycles at 94°C for 30 s, 54°C for 30 s, and 72°C for 30 s; and 1 cycle at 72°C for 10 min. The GAPDH product was used as a normalization control for the amount of cDNA in the PCR reactions. PCR products were subjected to electrophoresis using 6% polyacrylamide gels and stained by Sybr Green for 1 h. Results were analyzed with a Molecular Dynamics PhosphorImager system.

Identification of Regulatory Domains in the BRCA2
Promoter-To analyze transcriptional regulation of the BRCA2 gene and to define functionally important cis-DNA elements in the 5Ј-flanking region of this gene, an 8-kb region of human genomic DNA containing the BRCA2 putative promoter was isolated from BAC clone B489G and subcloned into the pGL3 basic luciferase reporter vector. The 8-kb fragment contained 4.3 kb of sequence upstream of the putative transcription start site (2) and 3.7 kb downstream of the transcription start site as far as the 3Ј donor splice site of exon 3. This pGL3Prom and the pGL3 parent vector were transiently transfected into MCF-7 cells, and luciferase activity was measured after 48 h. All activities were normalized by activity measurements from co-transfected pRL-TK Renilla luciferase vector. The pGL3Prom construct yielded 100-fold more luciferase activity than pGL3, suggesting that the pGL3Prom construct contained the BRCA2 promoter.
To identify the minimal BRCA2 promoter, a series of deletion constructs (Fig. 1A) derived from pGL3Prom were generated as described above. Firefly luciferase expression was assayed following transient transfection of MCF-7 cells with these BRCA2 promoter constructs. The normalized luciferase activities for each deletion construct of the promoter relative to pGL3Prom activity are shown in Fig. 1B. The results indicate that the BRCA2 promoter is regulated in a complex fashion. No change in activity was detected when comparing the Del-2 construct with pGL3Prom, suggesting that the ϩ668 to ϩ3678 region has no influence on promoter activity. Deletion of the Ϫ4328 to Ϫ583 region caused a 2-fold increase in luciferase activity. Further deletion from Ϫ582 to Ϫ516 resulted in 2.5-fold activation of the promoter, whereas a 3-fold reduction in activity was detected following deletion of the Ϫ144 to Ϫ59 region. However, a 20-fold loss of luciferase activity was observed following deletion of a 40-bp region (Ϫ58 to Ϫ19), suggesting that the region contains cis-elements that are critical for positive regulation of basal transcription activity in the BRCA2 promoter.
Analysis of the Minimal BRCA2 Promoter-To more accurately map the cis-element within the Ϫ58 to Ϫ19 region that regulates BRCA2 basal transcription, a further series of deletions were constructed using the Del-15 construct as a template ( Fig. 2A). Deletion of the Ϫ34 to Ϫ19 region (Del-16) resulted in a 12-fold reduction in luciferase activity in comparison to Del-19 ( Fig. 2A). Sequence analysis of a 20-bp region from Ϫ34 to Ϫ14 was carried out in an effort to identify putative transcription factor binding sites that might regulate BRCA2 basal transcription activity. The region was found to contain a tan- To control for transfection, efficiency cells were cotransfected with pRL-TK, and the activity associated with each construct was normalized relative to Renilla luciferase activity. The luciferase activity for each construct is shown relative to the wild type pGL3Prom construct. dem GCGTCACG repeat ( Fig. 2A) that encodes several predicted transcription factor binding sites including cis-elements for c-Myc, USF, MLTF, and ATF transcription factors. A number of point mutations were introduced into the 16-bp repeat sequence in the Del-15 reporter construct in an effort to identify the cis-element, which was regulating basal transcription from the promoter. Substitution of nucleotides from repeat 1 resulted in a 4-fold loss in activity, whereas substitution of repeat 2 led to a 12-fold loss in luciferase activity ( Fig. 2A). Thus, both 8-bp repeats appear to be involved in regulation of basal transcription. Further mutation studies eliminated the ATF, c-Myc, and MLTF binding sites from consideration and determined that BRCA2 basal transcription is predominantly regulated through the USF binding site.
USF Regulates BRCA2 Basal Transcription-The Ϫ34 to Ϫ15 region has recently been reported to be responsible for regulation of the basal activity of the BRCA2 promoter (15). The USF binding site was shown to regulate promoter activity in a cell cycle-dependent manner, with binding of USF resulting in 3-fold induction of luciferase activity. In addition, the Elf-1 transcription factor was shown to bind to the Ets consensus binding site (Ϫ61 to Ϫ54) and to induce activity 3-fold. To verify these observations, we carried out gel shift assays with wild type and mutant probes from the Ϫ34 to Ϫ15 region. Four oligonucleotide probes, as shown in Table I, were synthesized and used for gel shift assays with MCF-7 total cell extracts. A specific DNA-protein complex was detected with wild type probe (Fig. 2B). The complex binds to repeat 2 and is ablated by mutant forms of this 8-bp sequence. The remaining complexes bind to all probes and most likely represent nonspecific binding. These findings were further confirmed by competition experiments. A 90-fold excess of unlabeled wild type oligonucleotide probe effectively blocked binding of labeled probe to the protein complex but had little effect on the nonspecific complexes (Fig. 2C). To verify that this protein complex contained a member of the USF transcription factor family, as previously suggested, supershift assays were performed using specific antibody against USF1. USF1 antibody efficiently supershifted the complexes formed with the wild type and M-1 DNA probes. In comparison, antibodies against c-Myc and ATF failed to  (Table I) and MCF-7 whole cell protein extracts were performed. M-1 contains substitution mutations in the first 8-bp repeat, and M-2 contains mutations in the second 8-bp repeat. The single protein complex is indicated. C, a protein complex binds specifically to the second 8-bp repeat. A competition assay was performed using increasing concentrations of unlabeled wild type oligonucleotide probe. D, the USF transcription factor binds to the second 8-bp repeat. Supershift assays were performed with whole cell lysates of MCF-7, anti-USF-1, c-Myc, and ATF antibodies, and wild type and mutant oligonucleotides (WT, M-1, and M-2). The supershifted complex containing the anti-USF antibody and the gel shift complex are indicated. E, USF requires a transactivating cofactor to activate the BRCA2 promoter. MCF7 cells were transfected with pGL3Prom, Del-15, Del-15-1 (substitution in the first repeat), Del-15-2 (substitution in the second repeat), or Del-15-1ϩ2 (substitutions in both repeats) constructs along with a pCMV-USF-VP16 construct containing a USF and VP16 fusion gene or a vector control. Luciferase activities were normalized by protein concentration and are shown relative to the activity from the pGL3Prom wild type construct.
supershift the complex (Fig. 2D). The combined data strongly suggest that USF binds to the BRCA2 promoter.
To address the role of USF in regulation of BRCA2 basal transcription, a series of expression assays were performed. Cotransfection of USF1 or USF2 expression constructs with the Del-15 luciferase reporter construct had no significant effect on luciferase activity in MCF-7 (data not shown). As a control, CREB, c-Myc, and c-Myc plus Max expression constructs were also cotransfected with the reporter constructs into the various cell lines. Ectopic expression of these transcription factors failed to induce luciferase activity (data not shown). Recent studies of USF-dependent promoters containing USF consensus binding sites in a variety of cell lines have determined that USF cooperates with transactivating factors to induce expression. In fact, ectopic expression of USF1 or USF2 in most epithelial tumor cell lines, such as MCF-7, results in minimal induction of USF-dependent promoters (28). However, in normal mammary epithelial cell lines such as human mammary epithelial cells (Clonetics) and MCF10A, and in the Saos-2 osteosarcoma cell line, ectopic expression of USF1 or USF2 induced a substantial increase in reporter gene expression and activity (28,29), suggesting a requirement for a cell line-specific transactivating factor. To evaluate whether USF must interact with a transactivating partner to induce the BRCA2 promoter, we ectopically expressed a USF-VP16 fusion construct in MCF-7 cells. As shown in Fig. 2E, the USF-VP16 fusion protein induced a 4-fold increase in luciferase activity from the full-length BRCA2 promoter and from the minimal promoter (Del-15). In addition, mutations in the repeat 2 USF binding site ablated the increased luciferase activity. This suggests that USF interacts with other transactivating proteins to regulate basal transcription from the BRCA2 promoter.
Induction of the BRCA2 Promoter by NFB-Following validation of the role of USF in regulation of BRCA2 basal transcription, we began to systematically map other transcription factor binding sites within the BRCA2 promoter that contribute to regulation of the promoter. Initially, we focused on the Ϫ144 to Ϫ59 region that was shown to induce basal transcription 3-fold. Sequence analysis of this region identified several putative transcription factor binding sites including an NFB consensus binding site located at positions Ϫ116 to Ϫ107 in the 8-kb BRCA2 promoter. To examine the role of NFB in regulation of the BRCA2 promoter, the effect of overexpression of NFB on luciferase activity was studied. Cotransfection of expression constructs of the p65 and p50 subunits of NFB with the pGL3Prom reporter construct containing the wild type BRCA2 promoter resulted in significant induction of luciferase activity. Expression of p65 alone and in combination with p50 increased activity 9-and 16-fold, respectively (Fig. 3A). However, expression of p50 alone resulted in a small reduction in activity in comparison to a vector control.
To determine whether this NFB site was required for regulation of BRCA2 basal transcription, the consensus GGAATT-TCCT site was substituted by TAACTTTCCT in the Del-14 BRCA2 promoter reporter construct. The Del-14 construct and the Del-14 mutant construct were transfected into MCF-7 cells, and the luciferase activity was measured as before. As shown in Fig. 3B, expression of p65 or p65 with p50 induced a 3-6-fold increase in luciferase activity from the wild type Del-14 promoter in MCF-7 cells but had little activating effect on the mutant promoter. These data suggest that the NFB p65 subunit can induce BRCA2 promoter activity by forming a heterodimer with endogeneous or ectopically expressed p50.
NFB Binds to BRCA2 Promoter-To determine whether NFB subunit proteins bind to the NFB site in the BRCA2 promoter, we performed gel shift assays of MCF-7 whole cell protein extracts with wild type (WT-B) and mutant (MT-B) oligonucleotide probes containing the NFB site from the BRCA2 promoter (Table II). Whole cell extracts were prepared from MCF-7 cells 48 h after transfection with pcDNA 3.1 vector and with NFB p65 plus p50 expression constructs. Gel shift analysis demonstrated that a protein complex specifically binds to the wild type NFB probe but not to the mutant probe following overexpression of p50 and p65 (Fig. 3C). No significant complex formed in the absence of overexpression of these genes. Addition of 100-fold excess of cold competitor DNA probe completely eliminated protein binding to labeled DNA probe (data not shown), suggesting that the protein complex binds specifically to the NFB site in the BRCA2 promoter. The complex was also supershifted by anti-p50 antibody, indicating that the NFB p50 subunit formed part of the complex (Fig.  3C). Although an anti-p65 antibody did not supershift the complex, a significant decrease in the amount of labeled complex was observed (Fig. 3C). Thus, the anti-p65 antibody may be binding to p65 in the complex, resulting in reduced access of the DNA probe to the p50 DNA-binding subunit of NFB. These data suggest that a p50/p65 NFB heterodimer directly interacts with the NFB-like site in the BRCA2 promoter, resulting in direct induction of the promoter.
In Vivo Induction of BRCA2 by Overexpression of NFB-To demonstrate an in vivo effect of NFB on BRCA2 promoter function, we studied the effect of overexpression of p50 and p65 NFB subunits on endogeneous BRCA2 expression. As before, p50 and p65 constructs were transfected into MCF-7 cells, and Northern blots of RNA from the cells were hybridized with a full-length BRCA2 cDNA probe. Substantial increases in BRCA2 mRNA expression were observed following ectopic expression of p65 and p50 plus p65 (Fig. 4A). In addition, Western blots of whole cell extracts were hybridized with anti-p50 and anti-p65 antibodies (Santa Cruz Biotech) to verify expression of the NFB subunits and with 9D3 anti-BRCA2 antibody (Gene-Tex) to determine protein levels of BRCA2 in response to expression of NFB subunits. Transfections of MCF7 cells with p65 and p50 plus p65 constructs resulted in substantially increased levels of these proteins (Fig. 4B). BRCA2 protein levels were also significantly increased in response to p65 and p50 plus p65 expression, whereas BRCA2 levels remained low in vector control transfected cells (Fig. 4B). This result verifies that NFB expression results in induction of BRCA2 expression.
Dominant Negative and Wild Type IB␣ Inhibit NFB-dependent Induction of BRCA2-To further demonstrate the role of NFB in regulation of the BRCA2 promoter, the effect of signaling from the NFB signaling pathway on BRCA2 promoter induction was assessed. In this study, transfection with a wild type IB␣ or a dominant negative mutant IB␣ expression construct was used to block signaling through the NFB  pathway. The dominant negative IB␣ mutant (IB␣ 32A/36A) (30) is mutated at two phosphorylation sites and cannot be degraded following IKK-dependent phosphorylation, resulting in retention of NFB in the cytoplasm. MCF-7 cells were cotransfected with expression constructs for p65, dn-IB␣, IB␣, p65 plus dn-IB␣, p65 plus IB␣, and vector controls along with the pGL3Prom reporter construct. Expression of dn-IB␣ or IB␣ in combination with p65 resulted in a significant reduction in luciferase activity when compared with the effect of p65 alone, as shown in Fig. 5A. In addition, quantitative reverse transcription-PCR analysis of RNA from these transfected cells demonstrated that ectopic expression of dn-IB␣ or IB␣ significantly reduced the level of expression of BRCA2 (Fig. 5B). These data suggest that inhibition of nuclear translocation of NFB by dn-IB␣ or IB␣ substantially inhibits BRCA2 promoter activity.

DISCUSSION
Evidence for involvement of BRCA2 in regulation of cellular response to DNA damage (9,12), in cell proliferation (8), in cell cycle regulation (18), and in transcriptional regulation (31)(32)(33) has been accumulating. The variety of functions of BRCA2 suggests that regulation of expression levels of this gene may play an important role in regulation of a number of important cellular processes and that alterations in BRCA2 expression may contribute to tumorigenesis. Interestingly, although no somatic mutations have been identified in BRCA2, apparent overexpression of BRCA2 has been detected in a significant proportion of sporadic breast cancers.  A, ectopic expression of the p65 and p50 NFB subunits activates the pGL3Prom wild type BRCA2 promoter in MCF-7 cells. The pGL3Prom wild type BRCA2 promoter reporter gene construct was transfected into MCF-7 cells with a pRL-TK Renilla luciferase construct and either pcDNA 3.1 (vector), p50 expression construct (p50), p65 expression construct (p65), or p50 and p65 expression constructs (p50ϩp65). Luciferase activities were normalized by the Renilla luciferase activity and are presented relative to the pcDNA3.1 control. B, ectopic expression of the p65 and p50 NFB subunits activates the minimal BRCA2 promoter in MCF-7 cells. The Del-14 wild type and Del-14 mutant (substitution in the NFB consensus binding site) reporter constructs were transfected with p50 and p65 expression constructs or a vector control. Luciferase activity was normalized as before and is shown relative to activity from the Del-14 wild type. C, NFB p50/p65 heterodimers bind to the NFB consensus binding site in the BRCA2 promoter in MCF-7 cells. Gel shift assays for the p50 and p65 NFB subunits with wild type (W) and mutant (M) oligonucleotide probes for the NFB consensus binding site in the BRCA2 promoter are shown. Gel shifts were performed using extracts from pcDNA3.1 or p65 and p50 transfected MCF-7 cells. Supershift assays were performed with anti-p50 and anti-p65 antibodies.
In this study, we have shown that the NFB transcription factor binds to the BRCA2 promoter and induces expression of the BRCA2 gene. Deletion mapping of the promoter determined that the Ϫ144 and Ϫ59 region, which contains an NFB binding site (GGAATTTCCT), is associated with 3-fold activation of the promoter. A combination of gel shift and supershift assays confirmed that NFB binds to the NFB cis-element. In addition, ectopic expression of NFB subunits p65 or p65 plus p50 resulted in induction of the BRCA2 promoter and increased levels of BRCA2 mRNA and protein within MCF-7 cells, whereas substitution mutations in the NFB binding site ablated these effects. These data strongly suggest that NFB can activate the BRCA2 promoter and induce increased expression of the BRCA2 gene.
The NFB transcription factor consists mostly of p50/p65 heterodimers, which are complexed to IB␣ in the cytoplasm of unstimulated cells. Upon activation of the NFB signaling pathway, degradation of IB␣ exposes nuclear localization signals on the p50/p65 heterodimer leading to nuclear translocation and transcriptional activation of a number of promoters. In this study, we have shown that overexpression of the p50 DNA-binding domain of NFB does not result in up-regulation of the BRCA2 promoter. Ectopically expressed p50 most likely forms a heterodimer with endogeneous p65, but because p65 levels are low and the NFB nuclear localization signals are present in p65, relatively little heterodimer translocates to the nucleus and binds to the BRCA2 promoter. Conversely, expression of the p65 subunit with or without ectopic p50 significantly induced luciferase activity, suggesting that the transactivating p65 subunit is necessary for induction of the BRCA2 promoter. Overexpressed p65 most likely binds to endogeneous p50, saturates IB, and translocates to the nucleus resulting in upregulation of the BRCA2 promoter. In this case, endogeneous levels of p50 appear to be sufficient to facilitate increased binding of the p50/p65 heterodimer to the promoter. Although only the p50 and p65 NFB subunits were analyzed in this study, it is likely that the other subunits such as c-Rel, p52, and RelB are also capable of contributing to induction of the BRCA2 promoter.
To verify the role of NFB in BRCA2 transcriptional regulation we also evaluated the effect of the NFB signaling pathway on BRCA2 expression. IB is a component of the NFB signaling pathway that binds to NFB and prevents nuclear translocation of NFB. Thus, overexpression of IB␣ or a dominant negative form of IB␣ that is resistant to IKK dependent degradation is expected to inhibit NFB nuclear translocation and NFB dependent promoter induction. In this study, ectopic expression of both dn-IB␣ and IB␣ abrogated NFB-dependent BRCA2 promoter induction and down-regulated BRCA2 mRNA levels, suggesting that expression of the BRCA2 tumor suppressor can be regulated by modulation of the NFB signaling pathway.
NFB is known to regulate expression of a large number of genes that play critical roles in regulation of apoptosis, tumorigenesis, and inflammation. In breast cancers, alterations in DNA binding activity, gene expression, and/or nuclear translocation of NFB proteins have been observed. More specifically, increased NFB DNA binding activity has been correlated with expression of the c-erbB-2 gene (34), and high levels of NFB/ Rel binding have been observed in carcinogen-induced primary rat mammary tumors (35). Because NFB appears to regulate BRCA2 expression, it seems likely that alterations in NFB expression and DNA binding (34,35) contribute to the observed overexpression of BRCA2 in breast tumors (26). Thus, alteration of expression of the BRCA2 tumor suppressor gene may be one mechanism by which aberrantly regulated NFB contributes to tumorigenesis. Interestingly, a 3-fold difference in luciferase activity between the pGL3Prom construct and the Del-14 construct was detected in the presence of ectopically expressed p65. This result suggests that other elements within the BRCA2 promoter are directly or indirectly responsive to NFB. One other NFB consensus binding site (GAGAAACCCC) was identified in the promoter at positions Ϫ808 to Ϫ799. However, using deletion constructs and by overexpressing p65, we have shown that this NFB site does not play a role in regulation of the BRCA2 promoter (data not shown). Thus, other cis-elements that are indirectly affected by NFB may contribute to regulation of the BRCA2 promoter.
In addition to NFB responsive elements, we have also identified another activation domain that results in 3-fold reduction in activity when removed and a single repression domain that results in 2.5-fold activation when removed (Fig.  1B). The transcription factor binding sites from these regions and the associated transcription factors that contribute to regulation of the BRCA2 promoter are not yet known. Recently a repression domain associated with 10-fold downregulation of the BRCA2 promoter was reported (36). This domain is associated with two Alu repeats and is located in the Del-7 and Del-8 clones and is deleted from the Del-9 clone shown in Fig. 1. However, in the current study the repression domain in the BRCA2 promoter maps to a different location (Del-9; Fig. 1). Further studies are needed to explain the differing results from the two studies.
The role of the USF transcription factor in regulation of basal transcription from the BRCA2 promoter was also verified in the course of this study. A critical 20-bp regulatory sequence (Ϫ34 to Ϫ15), which is predominantly controlled by binding of USF and is responsible for the majority of BRCA2 transcription, was identified. The critical 20-bp region contains a tandem repeat sequence (GCGTCACG) ( Fig. 2A) and consensus DNA-binding motifs for transcription factors such as c-Myc, ATF, and USF. Gel shift, supershift, and cotransfection studies demonstrated that only USF binds to the second repeat and regulates the BRCA2 promoter. Recently, Davis and colleagues (15) reported that USF binds to the BRCA2 promoter as a heterodimeric complex of USF-1 and USF-2 and regulates basal transcription in a cell cycle-dependent manner. In this earlier study, overexpression of USF in MCF-7 cells induced only a 2.5-fold increase in promoter activity. However, the addition of IE62, a varicella zoster viral protein that binds to USF proteins (37), resulted in 12-fold induction of promoter activity. Similarly, we have demonstrated that expression of a USF-VP16 fusion protein enhanced induction of the promoter. These results suggest that a co-activating factor is needed for USF activation of the BRCA2 promoter.
The USF family of basic helix-loop-helix leucine zipper transcription factors were originally named MLTF because of their involvement in transcription from the adenovirus major later promoter (38). It is noteworthy that many USF target genes such as p53 (39), transforming growth factor ␤2 (40), and cyclin B1 (41) are involved in regulation of proliferation and the cell cycle.Moreover,USFoverexpressionsignificantlyinhibitsc-Mycdependent cell transformation (42) and proliferation of certain transformed cells (29). Thus, the activation of the BRCA2 tumor suppressor gene promoter by USF is consistent with the anti-proliferative effect of this transcription factor. The observation that USF transcriptional activity is lost in breast cancer cell lines but not in normal breast epithelial cells (28) further supports a role for USF as a key regulator of breast cancer development. The combination of these studies and our data suggests that regulation of BRCA2 promoter activity by USF may serve an essential role in the prevention of breast cancer development.