Oncogene (2003) 22, 5415–5426
& 2003 Nature Publishing Group All rights reserved 0950-9232/03 $25.00 www.nature.com/onc
Tumor formation in mice with conditional inactivation of Brca1 in epithelial tissues Thomas R Berton1, Takashi Matsumoto1, Angustias Page2, Claudio J Conti1, Chu-Xia Deng3, Jose´ L Jorcano2 and David G Johnson*,1 1 Department of Carcinogenesis, University of Texas MD Anderson Cancer Center, Science Park-Research Division, Smithville, TX 78957, USA; 2Cell and Molecular Biology CIEMAT, 28040, Madrid, Spain; 3Laboratory of Biochemistry and Metabolism, National Institute of Diabetes, Digestive and Kidney Diseases, Building 10, 9N105, Bethesda, MD, 20892, USA
The BRCA1 tumor-suppressor protein has been implicated in the regulation of transcription, DNA repair, proliferation, and apoptosis. BRCA1 is expressed in many proliferative tissues and this is at least in part due to E2Fdependent transcriptional control. In this study, inactivation of a conditional murine Brca1 allele was achieved in a variety of epithelial tissues via expression of the Cre recombinase under the control of a keratin 5 (K5) promoter. The K5 Cre:Brca1 conditional knockout mice exhibited modest epidermal hyperproliferation, increased apoptosis, and were predisposed to developing tumors in the skin, the inner ear canal, and the oral epithelium after 1 year of age. Overexpression of the E2F1 transcription factor in K5 Cre:Brca1 conditional knockout mice dramatically accelerated tumor development. In addition, Brca1 heterozygous female mice that had elevated E2F1 expression developed tumors of the reproductive tract at high incidence. These findings demonstrate that in mice Brca1 functions as a tumor suppressor in other epithelial tissues in addition to the mammary gland. Moreover, inactivation of Brca1 is shown to cooperate with deregulation of the Rb-E2F1 pathway to promote tumorigenesis. Oncogene (2003) 22, 5415–5426. doi:10.1038/sj.onc.1206825 Keywords: BRCA1; Cre-lox; epithelium; E2F1
Introduction A total of 50% of all familial breast cancers and 90% of all combined familial breast and ovarian cancers have germline mutations in the BRCA1 gene (Miki et al., 1994; Hill et al., 1997). The majority of the germline mutations are truncations, which result in termination of translation, producing a nonfunctional or unstable protein product (Couch and Weber, 1996). Somatic mutations in BRCA1 are very rare in sporadic breast and ovarian cancers although expression of BRCA1 may be inactivated by promoter methylation in some tumors *Correspondence: DG Johnson; E-mail:
[email protected] Received 25 October 2002; revised 12 May 2003; accepted 30 May 2003
(Futreal et al., 1994; Baldwin et al., 2000; Bianco et al., 2000; Rice et al., 2000; Esteller et al., 2001). Studies in mice have revealed that BRCA1 function is required for embryonic growth and development (Gowen et al., 1996; Hakem et al., 1996; Liu et al., 1996; Ludwig et al., 1997; Shen et al., 1998). Unlike humans, mice heterozygous for a mutant Brca1 allele have not been reported to be tumor prone (Hakem et al., 1996; Liu et al., 1996; Ludwig et al., 1997). However, using Cre-lox technology it has been shown that inactivation of murine Brca1 in the luminal epithelium of the mammary gland can lead to tumorigenesis after a long latency (Xu et al., 1999a). The mechanism by which loss of BRCA1 function leads to breast and ovarian cancer is unclear. BRCA1 is expressed not only in hormonally regulated tissues such as the mammary gland, uterus, and testes, but also in nonhormonally regulated tissues including the epidermis, thymus, and oral epithelium (Miki et al., 1994; Lane et al., 1995; Marquis et al., 1995). BRCA1 expression is cell cycle regulated and is highest in tissues undergoing rapid proliferation (Chen et al., 1996; Gudas et al., 1996; Vaughn et al., 1996). Consistent with this expression pattern, the BRCA1 promoter contains a functional binding site for the E2F family of transcription factors (Wang et al., 2000). The BRCA1 promoter responds to positive and negative regulators of the cell cycle, such as Rb, cyclin D1, and p16INK4a, through this E2F binding site (Wang et al., 2000). The phosphorylation status of BRCA1 is also cell cycle regulated. During late G1 and S phases, BRCA1 becomes hyperphosphorylated and in early M phase it is dephosphorylated (Chen et al., 1996; Vaughn et al., 1996; Ruffner and Verma, 1997; Ruffner et al., 1999). Cell cycle regulated phosphorylation of BRCA1 occurs through the interaction of cdk2 and possibly cdk6 and their associated cyclins A and D, respectively (Chen et al., 1996; Ruffner et al., 1999). Together, these findings imply a general role for BRCA1 in cell proliferation. BRCA1 can negatively regulate cell proliferation (Holt et al., 1996; Randrianarison et al., 2001) and this appears to be related to BRCA1’s ability to modulate the transcription of a number of cell cycle regulatory genes. A role for BRCA1 in transcriptional regulation was first suggested by the finding that BRCA1 is found
Inactivation of Brca1 in epithelial tissues TR Berton et al
5416
in RNA polymerase II-containing complexes (Scully et al., 1997a). Moreover, fusion proteins consisting of BRCA1 linked to the GAL4 DNA-binding domain are able to transcriptionally activate reporters containing GAL4 DNA-binding sites (Chapman and Verma, 1996; Monteiro et al., 1996). BRCA1 physically associates with several transcriptional coactivators and repressors, including CtIP, p300/CBP, Rb, HDAC1, RbAp46, and RbAp48 (Yu et al., 1998; Yarden and Brody, 1999; Pao et al., 2000). In addition, BRCA1 interacts with several DNA-binding transcription factors. Interaction of BRCA1 with c-myc and the estrogen receptor is associated with repression of transcription (Wang et al., 1998; Fan et al., 1999). On the other hand, interaction of BRCA1 with the p53 tumor suppressor enhances p53dependent transcriptional activation of genes such as p21 and bax (Ouchi et al., 1998; Zhang et al., 1998). BRCA1 can also regulate the expression of p21 and Gadd45 in a p53-independent manner (Somasundaram et al., 1997; Harkin et al., 1999). A role for BRCA1 in DNA repair and the maintenance of genomic integrity has also been established. The BRCA1 protein interacts with the DNA repair factor, RAD51, as well as the BRCA2 and BARD1 proteins (Wu et al., 1996; Scully et al., 1997c). These proteins colocalize in nuclear foci, termed ‘dots’, that dissipate upon DNA damage and may reappear at replication structures containing PCNA (Scully et al., 1997b). This response to DNA damage appears to involve the phosphorylation of BRCA1 by ATM and other checkpoint kinases (Cortez et al., 1999; Gatei et al., 2000, 2001; Lee et al., 2000; Tibbetts et al., 2000). Embryonic stem cells from Brca1/ embryos demonstrate genetic instability and accumulate chromosomal abnormalities that are thought to arise through unrepaired DNA damage (Shen et al., 1998; Xu et al., 1999b). BRCA1-deficient embryonic stem cells have also been shown to be defective in their ability to repair thymine glycols after oxidative damage, suggesting a role for BRCA1 in transcription-coupled repair (Gowen et al., 1998). In this study, we used a Brca1 allele in which exon 11 is flanked by loxP sites (floxed) (Xu et al., 1999a) and a transgene expressing Cre recombinase under the control of a K5 promoter to generate conditional knockout mice. The K5 promoter is active in a variety of epithelial tissues including the basal layer of the epidermis, epithelium of the oral and sinus cavity, esophagus, bladder, prostate, vagina, and the myoepithelium of the mammary gland (Ramirez et al., 1994; Pierce et al., 1998a, 1999). These tissues include many of the tissues that have been shown to express Brca1 (Lane et al., 1995; Marquis et al., 1995). We find that K5 Cre:Brca1 conditional knockout mice exhibit a modest increase in epidermal proliferation and apoptosis and develop spontaneous tumors in the skin, the epithelium of the inner ear canal, and the oral epithelium after 1 year of age. Given that Brca1 is a transcriptional target of E2F1 and that BRCA1 may regulate E2F-dependent transcription through its interaction with Rb, we also examined the effect of Brca1 disruption on the activity Oncogene
of E2F1 in this model system. We find that the inactivation of Brca1 does not significantly affect E2F1-induced proliferation or apoptosis, but that it does cooperate with deregulated E2F1 activity to promote tumorigenesis.
Materials and methods Brca1 targeting and transgenic mice Mice with the mutant and conditional Brca1 alleles have been described (Shen et al., 1998; Xu et al., 1999a). The K5 Cre recombinase transgene was made by cloning the lambda bacteriophage Cre recombinase cDNA into a plasmid containing the bovine K5 promoter (Ramirez et al., 1994; Brakebusch et al., 2000). Founder transgenic mice were made by injecting the purified transgene into the pronuclei of zygotes, and then implanting the zygotes into pseudo-pregnant female mice. The established transgenic line was maintained in a 129SV/C57BL/6 background. The generation of K5 E2F1 transgenic mice has been described (Pierce et al., 1998a). Briefly, the transgene contains the bovine K5 promoter, the rabbit b-globin intron 2, the simian virus 40 polyadenylation signal, and the human E2F1 cDNA. Genotyping K5 E2F1 transgenic mice were genotyped by PCR amplification of sequences specific for the b-globin intron and/or human cDNA from tail genomic DNA. The K5 Cre transgenic mice were genotyped using primer a, 50 -GCCTGCATTACCGGTCGATGCAACGA-30 , and primer b, 50 -GTGGCAGATGGCGCGGCAACACCATT-30 , which amplify a 700 bp band by PCR. Mice carrying the mutant Brca1 allele and/or the conditional Brca1 allele were genotyped by PCR amplification as described (Shen et al., 1998; Xu et al., 1999a). K5 Cre: Brca1 þ / males or K5 E2F1:K5 Cre:Brca1 þ / males were bred with Brca1co/co females for generation of the conditional knockout mice. Southern blot analysis For Southern hybridization, DNA from the epidermis of 6–8-week-old mice was sequentially extracted with buffersaturated phenol (Boehringer Mannheim), phenol : chloroform : isoamyl alcohol (25 : 24 : 1), and chloroform : isoamyl alcohol (24 : 1), and precipitated with 0.1 volume of 3 m sodium acetate (pH 5.2)/1.0 volume isopropanol overnight at 201C. DNA (5 mg) was digested with Xba1 and run on 0.7% agarose gel overnight at 30 V. The gel was denatured in 1.5 m NaCl, 0.5 n NaOH for 45 min with gentle agitation, rinsed in water, and incubated in neutralizing solution (1 m Tris pH 7.4, 1.5 m NaCl) twice for 20 min. After an overnight transfer onto Hybond XL nylon membrane (Amersham) in 10 SCC, the membrane was crosslinked with a UV Stratalinker 2400. The membrane was prehybridized in 10 Denhardt, 4 SET (20 SET: 300 ml 5 m NaCl, 200 ml 1 m Tris-HCl (pH 8.0), and 20 ml 0.5 m EDTA (pH 8.0)), 0.1% SDS, 0.1% sodium pyrophosphate, and 100 mg/ml denatured salmon testes DNA (Sigma) for a minimum of 4 h at 651C. The membrane was hybridized in the same solution overnight at 651C with a cDNA probe for intron 12 (Xu et al., 1999a). Before exposing to film, the membrane was washed two times for 15 min at 651C with 0.4 SET, 0.1% SDS, and 0.1% sodium pyrophosphate.
Inactivation of Brca1 in epithelial tissues TR Berton et al
5417 RT–PCR analysis of BRCA1 expression
Activated caspase 3, keratin 5, and E2F1 immunohistochemistry
Mouse epidermal RNA was isolated from dorsal skin by submerging the skin in DEPC-treated H2O at 551C for 30 s and transferring the skin to DEPC-treated H20 at 41C for 30 s. The skin was placed in Tri Reagent (Molecular Research Center, Cincinnati, OH, USA) on ice in an RNAse-free glass 10-cm dish, and the epidermis was scrapped off using a scalpel. Epidermal scrapes were transferred to a corex tube and extracted as per the manufacturer’s protocol. Total RNA (10 mg) was DNase treated with 10 U of RQ1-RNase free DNase (Promega, Madison, WI, USA) at 371C for 30 min in a final volume of 20 ml (6.0 mm MgSO4). Stop solution (2 ml) was added according to the manufacturer’s protocol and heated to 651C for 10 min. To eliminate secondary structure, 2 mg of DNase-treated total RNA for each sample was heated to 751C for 5 min with 5 mm Random Decamers (Ambion, Austin, TX, USA) and 0.5 mm dNTPs, and then placed on ice. The reverse transcription was carried out with 300 U of Superscript III (Invitrogen, Carlsbad, CA, USA) at 501C for 90 min, with 1 U RNaseOUT (Invitrogen) (10.0 mm Tris-HCl, pH 8.3; 50.0 mm KCl; 1.5 mm MgCl2), then heated to 951C for 10 min. To control for loading, PCR analysis of the mouse RAS gene (192 bp) was performed (941C, 30 s; 551C, 45 s; 721C, 45 s; 30 cycles) using the following primers: rasF, 50 -GACTCCTACCGGAAACAGGT-30 and rasR, 50 -TGATGGATGTCCTCG AAGGA-30 . For BRCA1 PCR analysis, we used the following primers: exon 3F primer, 50 -AGGGAAG CACAAGGTTTAGTCAG-30 ; exon 6R primer, 50 -GGGTCTGCCTGTTTT TCTTCACT-30 ; exon 11F primer, 50 -CCAACAGCCTGGCATAGCAG-30 ; and exon 11R primer, 50 -GATGAAGTCCTCAGGTTGAAG-30 ; exon 6F primer, 50 GAAGAAAAA CAGGCAGACCCAAC-30 ; and exon 12R primer, 50 CTCTGCGAGCAGTCTTCAGAAAG-30 . Primers exon 3F and exon 6R (941C, 30 s; 581C, 45 s; 721C, 1 min; 34 cycles) amplify a 291 bp PCR product in the N-terminus of the mouse Brca1 mRNA. Primers exon 11F and exon 11R (941C, 30 s; 601C, 45 s; 721C, 45 s; 34 cycles) amplify a 619 bp exon 11-specific PCR product. Primers exon 6F and exon 12R (941C, 30 s; 581C, 45 s; 721C, 1 min; 34 cycles) flank exon 11 and amplify a 229 bp PCR product only when exon 11 is absent. PCR products were analysed by electrophoresis through 2.0% agarose gels containing ethidium bromide and the gels were photographed under ultraviolet illumination.
Activated caspase 3, K5, and E2F1 immunohistochemistry was performed on formalin-fixed, paraffin-embedded skin sections. Tissue sections were deparaffinized and then boiled in 10 mm citrate buffer (pH 6.0) for 10 min for antigen retrieval. Endogenous peroxidase activity was blocked with 3% hydrogen peroxide in methanol for 10 min. Tissue sections were washed with water. Slides were preincubated with normal goat serum for 30 min before the addition of the primary antibody. The slides were incubated for 30 min with either a rabbit antihuman/anti-mouse caspase 3 antibody (1 : 1000, R&D Systems, Inc.), a rabbit anti-mouse K5 antibody (1 : 500, BAbCO), or a rabbit anti-human E2F1 antibody (1 : 500, Santa Cruz Biotechnology). Afterwards, the sections were rinsed with PBS, and then incubated with biotinylated anti-rabbit IgG for 30 min. The slides were rinsed with PBS and developed with streptavidin–horseradish peroxidase conjugate and DAB substrate. The tissue sections were rinsed again and counterstained before mounting. Caspase 3-positive basal epidermal cells were counted along 10 mm of linear skin at 40.
BrdU incorporation Mice were injected with 170 ml of 20 mm 50 -bromo-30 -deoxyuridine (BrdU, Sigma), allowed to incorporate for 30 min and sacrificed. Skin samples were collected, fixed in 10% neutral buffered formalin, and paraffin-embedded. Tissue sections were then deparaffinized and fixed in methanol containing 1% hydrogen peroxide for 20 min. Tissue sections were rinsed with PBS containing 0.1% BSA three times for 5 min each. Slides were preincubated with normal goat serum for 20 min before addition of primary BrdU antibody (1 : 500, Becton Dickinson). The sections were incubated for 20 min, rinsed with PBS/BSA, and then incubated with biotinylated anti-rabbit IgG for 30 min. The slides were rinsed with PBS/BSA and developed with streptavidin– horseradish peroxidase conjugate and diaminobenzidine (DAB) substrate. The tissue sections were rinsed again and counterstained before mounting. To determine percent incorporation, the number of unstained and DAB-stained interfollicular basal cells (approximately 1000) were counted at 40.
Results Development of K5 Cre:Brca1 conditional knockout mice A K5 Cre recombinase transgene was generated by subcloning the bacteriophage Cre recombinase cDNA into a vector containing the bovine K5 promoter, the rabbit b-globin intron and the SV40 polyadenylation signal (Ramirez et al., 1994; Pierce et al., 1998a; Brakebusch et al., 2000). The K5 promoter drives expression of the Cre recombinase to the basal cell layer of the epidermis and several other epithelial tissues that have been shown to express Brca1 (Ramirez et al., 1994; Lane et al., 1995; Marquis et al., 1995; Pierce et al., 1998a). To generate K5 Cre:Brca1 conditional knockout mice (K5 Cre:Brca1co/), the K5 Cre transgene was first introduced into a Brca1 þ / background. The Brca1 null allele is disrupted by insertion of the pGKneo cassette into intron 10 deleting 330 bp of intron 10 and 407 bp of exon 11, including the splice acceptor for exon 11 (Shen et al., 1998). Male K5 Cre:Brca1 þ / mice were then mated to Brca1co/co homozygous female mice. The Brca1co/co mice contain two copies of the Brca1 conditional allele, where loxP sites flank exon 11 (Xu et al., 1999a). In the presence of Cre recombinase, exon 11 is deleted resulting in an alternative Brca1 transcript lacking exon 11. Exon 11 encodes approximately 60% of the BRCA1 protein and contains the two nuclear localization signals (Lane et al., 1995; Thakur et al., 1997; Wilson et al., 1997), as well as the sites of interaction for many of the proteins that associate with BRCA1 (Scully et al., 1997b, c; Wang et al., 1998; Aprelikova et al., 1999; Zhong et al., 1999). The BRCA1 protein product lacking exon 11-encoded amino acids is inactive for many BRCA1 functions, but has been shown to modestly extend the viability of Brca1 mutant embryos (Thakur et al., 1997; Wilson et al., 1997; Xu et al., 1999b; Huber et al., 2001). We verified Cre-mediated excision of exon 11 in the epidermis of Brca1 conditional knockout mice by Oncogene
Inactivation of Brca1 in epithelial tissues TR Berton et al
5418
mice were born at the expected ratio, were completely viable, and did not appear to have any gross abnormalities. Increased epidermal proliferation and apoptosis in K5 Cre:Brca1co/ mice Skin samples were taken from 6–8-week-old K5 Cre: Brca1co/ mice as well as K5 Cre:Brca1co/ þ , Brca1co/, and Brca1co/ þ littermates to analyse proliferation and apoptosis in the epidermis. For the purpose of these studies, the conditional allele was treated as a wild-type allele when Cre was not present. In Brca1co/ þ mice, 3.1% of interfollicular basal layer keratinocytes were in the S phase, as measured by BrdU incorporation (Figure 2a). This is consistent with previous measurements of epidermal proliferation in wild-type mice. In mice heterozygous for a functional Brca1 allele, either Brca1co/ or K5 Cre:Brca1co/ þ , the average percent of basal keratinocytes was slightly increased compared to wild-type mice, but this was not statistically significant. In contrast, K5 Cre:Brca1co/ conditional knockout mice did have a statistically significant (Po0.01) increase in the level of epidermal proliferation, which was double (6.6%) the level observed in wild-type mice. It should be noted that the presence of the Cre transgene appeared to
Figure 1 Cre-mediated recombination of Brca1 in the epidermis. (a) Southern blot analysis was performed using a murine cDNA probe for intron 12 of Brca1 on genomic epidermal DNA (5.0 mg) from 6–8-week-old mice from the following genotypes: Brca1co/co (lane 1), Brca1co/ (lane 2), and K5 Cre:Brca1co/ (lane 3). The wildtype and null alleles are 10.5 kb (a), the conditional allele is 8.0 kb (b), and the recombined allele is 3.5 kb (c). (b) RT–PCR analysis was performed for Brca1 exons 3–6, exon 11, and Cre-mediated exon 11 deletion (Dexon 11), as indicated using 2.0 mg of total RNA isolated from 6–8-week-old mice from the following genotypes: Brca1 þ / þ (lane 2), Brca1 þ / (lane 3), and K5 Cre:Brca1co/ (lane 4). Lane 1 is Kb-plus ladder. PCR analysis of Ras expression was used as an RNA quality control
Southern blot analysis using an intron 12-specific probe (Figure 1a). Expression of Brca1 was analysed by RT– PCR using cDNA made from mouse epidermal RNA. As expected, all genotypes expressed a Brca1 transcript that contained exons 3–6 from the 50 end of the Brca1 gene (Figure 1b). Mice that were wild type or hemizygous for Brca1 expressed transcripts that contained exon 11, as indicated by a PCR product specific to exon 11 sequences (Figure 1b, lanes 2 and 3). However, K5 Cre:Brca1co/ conditional knockout mice lacked detectable levels of transcript containing exon 11 (Figure 1b, lane 4). Instead, a transcript lacking exon 11, D exon 11, was detected in K5 Cre:Brca1co/ mice but not in wild-type or hemizygous mice. Previous studies have demonstrated that the conditional Brca1 allele functions similar to the wild-type allele in the absence of Cre recombinase (Xu et al., 1999a). K5 Cre:Brca1co/ Oncogene
Figure 2 Epidermal proliferation and apoptosis is elevated in K5 Cre:Brca1co/ mice. (a) The average percent of interfollicular keratinocytes labeled with BrdU is presented for Brca1co/ þ (wild type), Brca1co/, K5 Cre:Brca1co/ þ , and K5 Cre:Brca1co/ mice. (b) The average number of apoptotic cells per 10 mm of interfollicular skin from the same four genotypes. At least five mice in each group were used to perform each of the above calculations
Inactivation of Brca1 in epithelial tissues TR Berton et al
5419
slightly increase epidermal proliferation, even in a Brca1 þ / þ background, but again this was not statistically significant (data not shown). Deletion of Brca1 exon 11 leads to embryonic lethality associated with elevated apoptosis (Gowen et al., 1996; Xu et al., 2001), and conditional inactivation of Brca1 has been shown to elevate apoptosis in the epithelium of the mammary gland (Xu et al., 1999a). To determine the extent of apoptosis in the epidermis of K5 Cre:Brca1co/ mice, the expression of activated caspase 3 was assessed by immunohistochemistry. The number of cells staining positive for activated caspase 3 was significantly elevated in the epidermis of mice with increasing loss of Brca1 gene dosage (Po0.01) (Figure 2b). Despite these observed changes in epidermal homeostasis, no obvious histological difference was observed in the skin of conditional knockout mice and the thickness of the epidermis was similar among genotypes (data not shown). Spontaneous tumor development in K5 Cre:Brca1co/ mice To determine if Brca1 inactivation in K5-expressing tissues would contribute to tumor development, K5 Cre:Brca1co/ mice, along with K5 Cre:Brca1co/ þ , Brca1co/ , and Brca1co/ þ littermates, were maintained without treatment for approximately 20 months (88 weeks). K5 Cre:Brca1co/ mice developed tumors beginning after 1 year of age and by 88 weeks of age, 13 of 18 mice (72%) had developed tumors (Figure 3), which was significantly different from Brca1 þ /, K5 Cre:Brca1co/ þ , and Brca1co/ þ mice (Po0.001). Less than 5% of the wildtype or heterozygous littermates developed a tumor by 88 weeks of age. One Brca1co/ mouse developed squamous cell carcinomas (SCC) of the stomach at 83 weeks of age, while no K5 Cre:Brca1co/ þ mice developed a tumor by 88 weeks of age. The majority of the tumors that formed in K5 Cre:Brca1co/ mice were SCC of the inner ear canal (Figure 4a and c) and oral epithelium (Figure 4e) (Table 1). In one case, a tumor from the
Figure 3 The conditional inactivation of Brca1 in K5-expressing tissues accelerates tumorigenesis. K5 Cre:Brca1co/ mice were generated and monitored along with Brca1co/ þ , Brca1co/, and K5 Cre:Brca1co/ þ mice for tumor formation. Tumor appearance and incidence in the K5 Cre:Brca1co/ mice (n ¼ 18) were significantly different (Po0.001) from Brca1co/ þ (n ¼ 17), Brca1co/ (n ¼ 21), and K5 Cre:Brca1co/ þ mice (n ¼ 13) using Kaplan–Meier statistical methods with pairwise log-rank test
inner ear canal had the appearance of a basal cell carcinoma. Analyses of the tumors from the inner ear canal indicated that they originated from K5-expressing epithelium (Figure 4b and d). Tumors from the oral epithelium also originated from K5-expressing tissue (Figure 4f). The remaining K5-expressing tumors from K5 Cre:Brca1co/ mice occurred in the skin and vagina. Two K5 Cre:Brca1co/ mice developed tumors that were not epithelial in origin, a sarcoma of the gastrointestinal tract and a diffused lymphoma. Six K5 Cre:Brca1co/ mice died from unknown causes during the 88-week period and were excluded from this study. Inactivation of Brca1 does not modulate E2F1- induced proliferation or apoptosis The Brca1 gene is a transcriptional target of the E2F1 transcription factor (Wang et al., 2000; Polager et al., 2002). This was initially discovered when it was found that Brca1 expression is elevated in the epidermis of K5 E2F1 transgenic mice using the RAGE technique (Wang et al., 1999). Similar to K5 Cre:Brca1co/ mice, K5 E2F1 transgenic mice have elevated levels of epidermal proliferation and apoptosis and develop tumors in the skin and some other epithelial tissues after 1 year of age (Pierce et al., 1998a, b, 1999). To examine the effect of Brca1 inactivation on E2F1 activities in this model system, the K5 E2F1 transgene was introduced into K5 Cre:Brca1co/ mice. To generate K5 E2F1:K5 Cre: Brca1co/ mice, male K5 Cre:Brca1 þ / mice were bred to K5 E2F1 transgenic female mice from the lowexpressing E2F1.1 line (Pierce et al., 1998a). Male K5 E2F1:K5 Cre:Brca1co/ mice were then bred to Brca1co/co homozygous female mice. From this mating, it was expected that 12.5% of the offspring would be K5 E2F1:K5 Cre:Brca1co/, but the observed ratio of this genotype was actually 6.25% (10 out of 160). Of these 10 mice, five were killed at 6–8 weeks of age to examine the phenotype of the epidermis and five were maintained without treatment. Skin samples from K5 E2F1:K5 Cre:Brca1co/ mice and their littermates were used to analyse epidermal proliferation and apoptosis as described previously. Inactivation of one or both alleles of Brca1 in the epidermis of K5 E2F1 transgenic mice had no significant effect on the percent of basal layer keratinocytes in the S phase compared to K5 E2F1 transgenic mice with wild-type Brca1 status (Figure 5a). The level of apoptosis, as measured by activated caspase 3 staining, was modestly elevated in K5 E2F1 transgenic mice when one or two copies of Brca1 were inactivated (Figure 5b). However, the level of apoptosis in K5 E2F1:K5 Cre:Brca1co/ mice was similar to the level of apoptosis observed in K5 Cre:Brca1co/ mice (Figure 2b). We conclude that inactivation of Brca1 does not have a significant effect on the proliferative or apoptotic activities of E2F1 in this model system. Conversely, overexpression of E2F1 has no significant effect on the proliferative and apoptotic phenotype of epidermis from Brca1 conditional knockout mice. Oncogene
Inactivation of Brca1 in epithelial tissues TR Berton et al
5420
Figure 4 K5 Cre:Brca1co/ mice develop tumors of the inner ear canal and oral epithelium. (a) Low-magnification photomicrograph of a normal inner ear canal (ec) and sebaceous gland (sg) stained with H&E. (b) The same tissue in (a) showing K5 expression in the inner ear canal and in a sebaceous gland. (c) An SCC of the inner ear canal stained with H&E and a keratin cyst located in the sebaceous gland next to the SCC. (d) The same SCC in (c) showing K5 expression. (e) An SCC of the oral epithelium stained with H&E. (f) The same tumor in (e) showing K5 expression. The bar in (a) is 100 mm and the same scale was used for panels (b–d). The bar in (e) is 100 mm and the same scale was used in (f)
Spontaneous tumor development in K5 E2F1:K5 Cre: Brca1co/ mice To determine if Brca1 inactivation would cooperate with increased E2F1 activity in tumor development, a group of K5 E2F1:K5 Cre:Brca1co/ mice along with K5 E2F1:Brca1co/ and K5 E2F1:Brca1co/ þ littermates were maintained for over a year. K5 E2F1 transgenic mice with wild-type Brca1 developed tumors after 1 year of age, and by 70 weeks 50% had developed tumors (Figure 6). This is highly consistent with previous tumor incidence studies of K5 E2F1 transgenic Oncogene
mice (Pierce et al., 1999). In contrast, by 45 weeks of age, all K5 E2F1:K5 Cre:Brca1co/ mice had developed at least one skin tumor, with an average of four tumors per mouse (Figure 6). The tumor latency and incidence of K5 E2F1:K5 Cre:Brca1co/ mice were also significantly different when compared to K5 Cre:Brca1co/ mice without the K5 E2F1 transgene (compare Figure 3 with Figure 6) (Po0.001). Thus, inactivation of Brca1 cooperates with deregulated E2F1 activity to promote tumorigenesis. Inactivating one Brca1 allele in K5 E2F1 mice also significantly accelerated tumorigenesis
Inactivation of Brca1 in epithelial tissues TR Berton et al
5421 Table 1
Spontaneous tumors in K5 Cre: Brca1co/ mice
ID
Sex
Age (weeks)
Location
Histological appearance
1
M
53
2 3 4
M M F
59 61 62
5 6
M M
64 64
7 8 9
F F M
65 65 70
10 11 12 13
M F M F
74 75 80 86
Oral epithelium Skin Oral epithelium Ear canal Vagina Skin Ear canal Ear canal Oral epithelium Skin Oral epithelium Oral epithelium Ear canal Oral epithelium Multiple organs Multiple organs Skin
Squamous cell carcinoma Squamous cell carcinoma Squamous cell carcinoma Basal cell carcinoma Squamous cell carcinoma Squamous cell carcinoma Squamous cell carcinoma Squamous cell carcinoma Squamous cell carcinoma Squamous cell carcinoma Squamous cell carcinoma Squamous cell carcinoma Squamous cell carcinoma Squamous cell carcinoma Undifferentiated sarcoma Lymphoma Squamous cell carcinoma
compared to K5 E2F1 transgenic mice with wild-type Brca1 (Figure 6). The majority of tumors that develop in K5 E2F1 transgenic mice occur in the skin, with a relatively low incidence of tumors in the oral cavity and vagina (Pierce
Figure 5 Conditional inactivation of Brca1 does not affect E2F1induced epidermal proliferation and apoptosis. (a) The average percent of interfollicular keratinocytes labeled with BrdU is presented for Brca1co/ þ , K5 E2F1:Brca1co/ þ , K5 E2F1:Brca1co/, and K5 E2F1:K5 Cre:Brca1co/ mice. (b) The average number of apoptotic cells per 10 mm of interfollicular skin from the same four genotypes. At least five mice in each group were used to perform each of the above calculations
Figure 6 The conditional inactivation of Brca1 in K5 E2F1 transgenic mice accelerates spontaneous tumor development. K5 E2F1:K5 Cre:Brca1co/ mice were generated and monitored along with K5 E2F1:Brca1co/ and K5 E2F1:Brca1co/ þ mice for spontaneous tumor formation. Tumorigenesis in K5 E2F1:K5 Cre:Brca1co/ mice (n ¼ 5) was significantly different (Po0.001) from K5 E2F1:Brca1co/ (n ¼ 19) and K5 E2F1:Brca1co/ þ mice (n ¼ 6). K5 E2F1:Brca1co/ mice are significantly different from K5 E2F1:Brca1co/ þ mice (Po0.01) using Kaplan–Meier statistical methods with pairwise log-rank test
et al., 1999). Tumors in the K5 E2F1:K5 Cre:Brca1co/ mice also developed primarily in the skin. These tumors were all SCC of the skin, with the exception of one mouse that also had an SCC of the oral epithelium (Table 2). The majority of tumors from five male K5 E2F1 transgenic mice heterozygous for Brca1 (K5 E2F1:Brca1co/) were also SCC of the skin, with two mice having an SCC of the oral epithelium. In contrast, female K5 E2F1:Brca1co/ mice appeared to be highly predisposed to tumors of the reproductive tract (Table 2). All of the female K5 E2F1:Brca1co/ mice acquired a tumor. The majority of tumors were from the cervix (two papillomas and six SCC) and vaginal epithelium (seven SCC). Of the eight K5 E2F1:Brca1co/ mice that developed SCC of the cervix, four also developed SCC of the vagina. The remaining female mice developed tumors in the oral epithelium. Only one K5 E2F1:Brca1co/ mouse developed a tumor in a nonK5-expressing tissue, which was an adenocarcinoma of the mammary gland. We also generated K5 E2F1:K5 Cre:Brca1 þ / and K5 E2F1:K5 Cre:Brca1co/ þ littermates from the K5 E2F1:K5 Cre:Brca1co/ and Brca1co/co breeding. Some of these littermates were maintained without treatment for over a year, but were not included in Figure 6 and Table 2, even though they developed tumors around the same time and in the same tissues as K5 E2F1:Brca1co/ mice. Given the fact that K5 E2F1:Brca1co/ mice develop tumors of the cervix and K5 expression in the human reproductive tract occurs in the ectocervical epithelium and endocervical epithelium (Smedts et al., 1992), we examined the pattern of expression of K5 in the cervix of wild-type mice. In a normal adult female reproductive tract, we observed positive K5 expression throughout the cervix including the basal layer (Figure 7b). K5 expression was also observed in the reserve cells of the squamocolumnar junction (SCJ) (Figure 7d). Cervical Oncogene
Inactivation of Brca1 in epithelial tissues TR Berton et al
5422 Table 2 Tumor profile for Brca1 conditional mice Genotype
# Of mice
Locaton of tumors
a
w/tumors
Brca1co/+ Brca1co/ K5 Cre:Brca1co/ K5 E2F1:Brca1co/+ K5 E2F1:Brca1co/ K5 E2F1:K5 Cre:Brca1co/
# Of micea
M
F
% Of mice w/tumors
Skin
Oral epithelium
Inner ear canal
Vagina
Cervix
17 21 21 6 19 5
0 0 6 1 5 2
0 0 3 2 13 3
0.0% 0.0% 42.9% 50.0% 94.7% 100%
0 0 3 2 9 5
0 0 5 1 6 1
0 0 4 0 0 0
0 0 1 1 7 0
0 0 0 1 8 0
a
At 70 weeks of age
Figure 7 Keratin 5 is expressed in the cervix and the reserve cells of the squamocolumnar junction. Low-power magnification of a normal cervix stained with (a) H&E and (b) immunohistochemically stained for K5 expression. The same normal cervix as in (a, b) shown at high power magnification (c) stained with H&E and (d) immunohistochemically stained for K5 expression. Black arrows in (c, d) indicate the squamocolumnar junction (scj) and the reserve cells (rc) stained positive for K5. The bar in (a) is 100 mm and the same scale was used for panel (b). The bar in (c) is 10 mm and the same scale was used for panel (d)
tumors from the K5 E2F1:Brca1co/ mice appeared to originate from K5-expressing squamous epithelium since all tumors stained positive for K5 expression (Figure 8b and e). We also confirmed expression of the E2F1 transgene in the squamous epithelium of the cervix and in the tumors (Figure 8c, f ).
Discussion Inheritance of a mutant BRCA1 allele predisposes women to breast and ovarian cancer (Miki et al., 1994; Oncogene
Hill et al., 1997). In mice, inactivation of one allele of Brca1 does not produce a phenotype and has limited impact on neoplasia, while the deletion of both alleles results in embryonic lethality (Gowen et al., 1996; Hakem et al., 1996; Liu et al., 1996; Ludwig et al., 1997). Recent advances in mouse-modeling technology have made it possible to create conditional knockout mice to overcome embryonic lethality. This was achieved with the Brca1 gene in mice by flanking exon 11 with loxP sites and mating them to mice that overexpress the Cre recombinase in the mammary gland by using the Wap or MMTV promoters (Xu et al., 1999a). Wap-Cre:Brca1
Inactivation of Brca1 in epithelial tissues TR Berton et al
5423
Figure 8 Keratin 5 and E2F1 expression in tumors from K5 E2F1:Brca1co/ mice. (a) H&E stained papilloma of the uterine cervix. The same papilloma in (a) was stained for expression of (b) K5 and (c) E2F1. (d) H&E stained SCC of the cervix. The same SCC in (d) was stained for expression of (e) K5 and (f) E2F1. The bar in (a) is 100 mm and the same scale was used for panels (b, c). The bar in (d) is 100 mm and the same scale was used for panels (e, f)
and MMTV-Cre:Brca1 conditional knockout mice were viable, but had blunted ductal development of the mammary glands and increased mammary epithelial apoptosis. Mammary tumor formation occurred between 10 and 13 months of age in approximately 20% of female mice with conditional inactivation of Brca1 in the luminal epithelium (Xu et al., 1999a). Mammary tumors from these mice contain several genetic abnormalities that are similar to human BRCA1-associated tumors (Weaver et al., 2002). Thus, the Wap-Cre:Brca1 and MMTV-Cre:Brca1 conditional knockout mice appear to be good models for BRCA1-associated breast cancer in humans. Expression of Brca1 in the developing embryo and adult mouse occurs in a variety of proliferative tissues including the epidermis, uterus, and oral epithelium (Lane et al., 1995; Marquis et al., 1995). To better understand the role of Brca1 in regulating tissue homeostasis and tumor suppression in these tissues, we have targeted the conditional deletion of Brca1 exon 11 using a K5 Cre transgene. The K5 promoter is expressed in the basal layer of the epidermis and in many of the same epithelial tissues where Brca1 is expressed (Ramirez et al., 1994; Lane et al., 1995; Marquis et al., 1995). An additional advantage of directing the Cre recombinase to the epidermis is the ability to assess easily the effects of Brca1 inactivation on proliferation, apoptosis, and tumorigenesis all in the same primary tissue. We found that the conditional inactivation of Brca1 in the epidermis results in increased proliferation and apoptosis of basal keratinocytes. Although an increase in proliferation due to the conditional inactivation of Brca1 in the mammary gland was not observed (Xu et al., 1999a), in vitro evidence suggests that the loss of BRCA1 expression can lead to increased cell proliferation. An increase in proliferation has been observed
using antisense oligonucleotides to BRCA1 in both normal and malignant breast epithelial cells (Thompson et al., 1995), or when antisense Brca1 cDNA was transfected into NIH3T3 cells (Rao et al., 1996). Conversely, ectopic expression of wild-type BRCA1 in mammary or ovarian tumor cells can inhibit cell proliferation (Holt et al., 1996; Randrianarison et al., 2001). Moreover, when the same tumor cells are transfected with either a 50 truncation of exon 11 or an internal exon 11 deletion of BRCA1, cell proliferation is not inhibited (Holt et al., 1996). Consistent with the previous study of adult mammary glands lacking functional Brca1 (Xu et al., 1999a), the conditional inactivation of Brca1 in the epidermis also increased the levels of apoptosis. K5 Cre:Brca1co/ mice were predisposed to developing spontaneous tumors, which is also similar to the conditional inactivation of Brca1 in the mammary gland (Xu et al., 1999a). In agreement with the Wap-Cre:Brca1co/ and MMTVCre:Brca1co/ mice, K5 Cre:Brca1co/ mice developed tumors in tissues that expressed the Cre recombinase after a long latency. Spontaneous tumors developed in the K5 Cre:Brca1co/ mice beginning at 53 weeks of age, and by 88 weeks of age approximately 72% of mice had developed tumors. The majority of the tumors in the K5 Cre:Brca1co/ mice were SCC of the inner ear canal, oral epithelium, and skin. Tumors did not develop from the myoepithelium of the mammary gland, where K5 expression also occurs. These data demonstrate that Brca1 functions as a tumor suppressor in tissues other than the mammary gland in mice. A previous study in which a hypomorphic, truncating allele of Brca1 was introduced into mice also supports a broad role for BRCA1 in suppressing tumorigenesis in a variety of epithelial, as well as nonepithelial, tissues (Ludwig et al., 2001). Interestingly, population studies on BRCA1Oncogene
Inactivation of Brca1 in epithelial tissues TR Berton et al
5424
associated kindreds and mutational analysis of loss of heterozygosity for BRCA1 in various tumors suggest that BRCA1 loss could contribute to the development of other tumor types in humans, including SCC of the skin as well as the head and neck (Johannsson et al., 1999; Tong et al., 2000). Deregulation of the p16INK4a-cyclin D-Rb-E2F pathway is a very common event in human cancers (Sherr, 1996). A number of links between the Rb-E2F pathway and BRCA1 have been observed. The human BRCA1 and mouse Brca1 gene promoters contain a conserved E2F DNA-binding site that regulates Brca1 expression in response to Rb, cyclin D1, and p16INK4a (Wang et al., 2000). Upregulation of Brca1 expression by E2F1 has been shown to occur in cell culture and in K5 E2F1 transgenic mice (Wang et al., 1999; Polager et al., 2002). BRCA1 has also been found to physically associate with Rb though a domain encoded by exon 11 of the BRCA1 gene (Aprelikova et al., 1999). Furthermore, this BRCA1–RB complex interacts with the histone deacetylase complexes through the BRCA1 carboxyl- terminal domain and the Rb-binding proteins RbAp46 and RbAp48 (Yarden and Brody, 1999). The interaction between BRCA1 and Rb may be important for BRCA1’s ability to regulate proliferation since BRCA1 cannot induce growth arrest in cells lacking Rb (Aprelikova et al., 1999). These findings suggest that BRCA1 may regulate proliferation by cooperating with Rb to inhibit E2F-dependent transcription and that upregulation of BRCA1 by E2F1 may function as a negative feedback mechanism to limit the E2F-proliferative signal. Despite the evidence that BRCA1 may regulate E2F activity, inactivation of Brca1 had no effect on keratinocyte proliferation in the K5 E2F1 transgenic mouse model system. Brca1 inactivation also had little effect on aberrant apoptosis in K5 E2F1 transgenic epidermis. Nonetheless, inactivation of Brca1 in K5 E2F1 transgenic mice did significantly accelerate tumorigenesis. The mechanism for this cooperation between Brca1 inactivation and E2F1 overexpression is unclear but may be related to the decreased genomic stability
associated with the loss of Brca1. A number of studies have shown that inactivation of Brca1 leads to genetic instability characterized by aneuploidy and chromosomal rearrangements (Shen et al., 1998; Xu et al., 1999a, b; Weaver et al., 2002). Loss of p53 function is also associated with tumors lacking BRCA1 from both mice (Xu et al., 1999a, 2001; Weaver et al., 2002) and humans (Crook et al., 1997, 1998; Eisinger et al., 1997; Rhei et al., 1998; Tong et al., 2000). It has been demonstrated that p53 plays an important role in suppressing tumor development in response to deregulation of the Rb-E2F pathway (Howes et al., 1994; Symonds et al., 1994; Williams et al., 1994; Pierce et al., 1998b). It is therefore quite possible that Brca1 inactivation leads to an increased rate of genetic mutations, such as p53 loss, that cooperate with increased E2F1 activity in tumorigenesis. In summary, we have shown that the conditional inactivation of Brca1 in K5-expressing tissues increased cell proliferation and apoptosis in the epidermis and contributed to tumor formation in epithelial tissues other than the mammary gland. We also found that overexpression of the E2F1 transcription factor and conditional inactivation of Brca1 cooperated to accelerate tumor formation in a variety of epithelial tissues. These findings support roles for BRCA1 in regulating cell proliferation and apoptosis, and as a tumor suppressor that may be important for maintaining genomic stability in response to deregulation of the Rb-E2F pathway. Acknowledgements We thank Dr Dennis Johnston for his expert statistical analysis, Pam Blau and Stephanie Clifford for their technical assistance, Dr Lezlee Coghlan, Dale Weiss, and co-workers for animal care, Dr Irma Gimenez-Conti, Nancy Abbey, and coworkers for the immunohistochemistry, Chris Yone and Joi Holcomb for artwork. We also thank Dr Xiaoling Xu for the Brca1 probe for Southern analysis. This work was supported by grants from the National Institute of Health (U01 ES11047 and a T-32 National Service Award Training Grant CA 09480 to TRB).
References Aprelikova ON, Fang BS, Meissner EG, Cotter S, Campbell M, Kuthiala A, Bessho MJ, Jensen RA and Liu ET. (1999). Proc. Natl. Acad. Sci. USA, 96, 11866–11871. Baldwin RL, Nemeth E, Tran H, Shvartsman H, Cass I, Narod S and Karlan BY. (2000). Cancer Res., 60, 5329–5333. Bianco T, Chenevix-Trench G, Walsh DC, Cooper JE and Dobrovic A. (2000). Carcinogenesis, 21, 147–151. Brakebusch C, Grose R, Quondamatteo F, Ramirez A, Jorcano JL, Pirro A, Svensson M, Herken R, Sasaki T, Timpl R, Werner S and Fassler R. (2000). EMBO J., 19, 3990–4003. Chapman MS and Verma IM. (1996). Nature, 382, 678–679. Chen Y, Farmer AA, Chen CF, Jones DC, Chen PL and Lee WH. (1996). Cancer Res., 56, 3168–3172. Cortez D, Wang Y, Qin J and Elledge SJ. (1999). Science, 286, 1162–1166. Oncogene
Couch FJ and Weber BL. (1996). Hum. Mutat., 8, 8–18. Crook T, Brooks LA, Crossland S, Osin P, Barker KT, Waller J, Philp E, Smith PD, Yulug I, Peto J, Parker G, Allday MJ, Crompton MR and Gusterson BA. (1998). Oncogene, 17, 1681–1689. Crook T, Crossland S, Crompton MR, Osin P and Gusterson BA. (1997). Lancet, 350, 638–639. Eisinger F, Jacquemier J, Guinebretiere JM, Birnbaum D and Sobol H. (1997). Lancet, 350, 1101. Esteller M, Corn PG, Baylin SB and Herman JG. (2001). Cancer Res., 61, 3225–3229. Fan S, Wang JA, Yaun R, Ma Y, Meng Q, Erdos MR, Pestell RG, Yuan F, Auborn KJ, Goldberg ID and Rosen EM. (1999). Science, 284, 1354–1356. Futreal PA, Liu Q, Shattuck-Evans D, Cochran C, Harshman K, Tavtigian S, Bennett LM, Haugen-Strano A, Swenson J,
Inactivation of Brca1 in epithelial tissues TR Berton et al
5425 Miki T, Eddington K, McClure M, Frye C, WeaverFeldhaus J and Ding W. (1994). Science, 266, 120–122. Gatei M, Scott SP, Filippovitch I, Soronika N, Lavin MF, Weber B and Khanna KK. (2000). Cancer Res., 60, 3299–3304. Gatei M, Zhou BB, Hobson K, Scott S, Young D and Khanna KK. (2001). J. Biol. Chem., 276, 17276–17280. Gowen LC, Avrustskaya AV, Latour AM, Koller BH and Leadon SA. (1998). Science, 281, 1009–1012. Gowen LC, Johnson BL, Latour AM, Sulik KK and Koller BH. (1996). Nat. Genet., 12, 191–194. Gudas JM, Li T, Nguyen H, Jensen D, Rauscher III FJ and Cowan KH. (1996). Cell Growth Differ., 7, 717–723. Hakem R, de la Pompa J, Sirard C, Mo R, Woo M, Hakem A, Wakeman A, Potter J, Reitmair A, Billia F, Firpo E, Hui CC, Roberts J, Rossant J and Mak TW. (1996). Cell, 85, 1009–1023. Harkin PD, Bean JM, Miklos D, Song YH, Truong VB, Englert C, Christians FC, Ellisen LW, Maheswaran S, Oliner JD and Haber DA. (1999). Cell, 97, 575–586. Hill AD, Doyle JM, McDermott EW and O’Higgins NJ. (1997). Br. J. Surg., 84, 1334–1339. Holt JT, Thompson ME, Szabo C, Robinson-Benion C, Arteaga CL, King MC and Jensen RA. (1996). Nat. Genet., 12, 298–302. Howes KA, Ransom N, Papermaster DS, Lasudry JG, Albert DM and Windle JJ. (1994). Genes Dev., 8, 1300–1310. Huber LJ, Yang TW, Sarkisian CJ, Master SR, Deng CX and Chodosh LA. (2001). Mol. Cell. Biol., 21, 4005–4015. Johannsson O, Loman N, Moller T, Kristoffersson U, Borg A and Olsson H. (1999). Eur. J. Cancer, 35, 1248–1257. Lane TF, Deng C, Elson A, Lyu MS, Kozak CA and Leder P. (1995). Genes Dev., 9, 2712–2722. Lee JS, Collins KM, Brown AL, Lee CH and Chung JH. (2000). Nature, 404, 201–204. Liu CY, Flesken-Nikitin A, Li S, Zeng Y and Lee WH. (1996). Genes Dev., 10, 1835–1843. Ludwig T, Chapman DL, Papaioannou VE and Efstratiadis A. (1997). Genes Dev., 11, 1226–1241. Ludwig T, Fisher P, Ganesan S and Efstratiadis A. (2001). Genes Dev., 15, 1188–1193. Marquis ST, Rajan JV, Wynshaw-Boris A, Xu J, Yin GY, Abel KJ, Weber BL and Chodosh LA. (1995). Nat. Genet., 11, 17–26. Miki Y, Swenson J, Shattuck-Eidens D, Futreal PA, Harshman K, Tavitigian S, Liu Q, Cochran C, Bennett LM and Ding W. (1994). Science, 266, 66–71. Monteiro AN, August A and Hanafusa H. (1996). Proc. Natl. Acad. Sci. USA, 93, 13595–13599. Ouchi T, Monterio A, August A, Aaronson SA and Hanafusa H. (1998). Proc. Natl. Acad. Sci. USA, 95, 2302–2306. Pao GM, Janknecht R, Ruffner H, Hunter T and Verma IM. (2000). Proc. Natl. Acad. Sci. USA, 97, 1020–1025. Pierce AM, Fisher SM, Conti CJ and Johnson DG. (1998a). Oncogene, 16, 1267–1276. Pierce AM, Gimenez-Conti IB, Schneider-Broussard R, Martinez LA, Conti CJ and Johnson DG. (1998b). Proc. Natl. Acad. Sci. USA, 95, 8858–8863. Pierce AM, Schneider-Broussard R, Gimenez-Conti IB, Russell JL, Conti CJ and Johnson DG. (1999). Mol. Cell. Biol., 19, 6408–6414. Polager S, Kalma Y, Berkovich E and Ginsberg D. (2002). Oncogene, 21, 437–446. Ramirez A, Bravo A, Jorcano JL and Vidal M. (1994). Differentiation, 58, 53–64.
Randrianarison V, Marot D, Foray N, Cabannes J, Meret V, Connault E, Vitrat N, Opolon P, Perricaudet M and Feunteun J. (2001). Cancer Gene Ther., 8, 759–770. Rao VN, Shao N, Ahmad M and Reddy ES. (1996). Oncogene, 12, 523–528. Rhei E, Bogomolniy F, Federici MG, Maresco DL, Offit K, Robson ME, Saigo PE and Boyd J. (1998). Cancer Res., 58, 3193–3196. Rice JC, Ozcelik H, Maxeiner P, Andrulis I and Futscher BW. (2000). Carcinogenesis, 21, 1761–1765. Ruffner H, Jiang W, Craig AG, Hunter T and Verma IM. (1999). Mol. Cell. Biol., 19, 4843–4854. Ruffner H and Verma IM. (1997). Proc. Natl. Acad. Sci. USA, 94, 7138–7143. Scully R, Anderson SF, Chao DM, Wei W, Ye L, Young RA, Livingston DM and Parvin JD. (1997a). Proc. Natl. Acad. Sci. USA, 94, 5605–5610. Scully R, Chen J, Oches RL, Keegan K, Hoekstra M, Feunteun J and Livingston DM. (1997b). Cell, 90, 425–435. Scully R, Chen J, Plug A, Xiao Y, Waever D, Feunteun J, Ashley T and Livingston DM. (1997c). Cell, 88, 265–275. Shen SX, Weaver Z, Xu X, Li C, Weinstein W, Guan XY, Reid T and Deng CX. (1998). Oncogene, 17, 3115–3124. Sherr CJ. (1996). Science, 274, 1672–1677. Smedts F, Ramaekers F, Troyanovsky S, Pruszczynski M, Robben H, Lane B, Leigh I, Plantema F and Vooijs P. (1992). Am. J. Pathol., 140, 601–612. Somasundaram K, Zhang H, Zeng YX, Houvras Y, Peng Y, Wu GS, Licht JD, Weber BL and El-Deiry WS. (1997). Nature, 389, 187–190. Symonds H, Krall L, Remington L, Saenz-Robles M, Lowe S, Jacks T and Van Dyke T. (1994). Cell, 78, 703–711. Thakur S, Zhang HB, Peng Y, Le H, Carroll B, Ward T, Yao J, Farid LM, Couch FJ, Wilson RB and Weber BL. (1997). Mol. Cell. Biol., 17, 444–452. Thompson ME, Jensen RA, Obermiller PS, Page DL and Holt JT. (1995). Nat. Genet., 9, 444–450. Tibbetts RS, Cortez D, Brumbaugh KM, Scully R, Livingston D, Elledge SJ and Abraham RT. (2000). Genes Dev., 14, 2989–3002. Tong D, Kucera E, Schuster E, Schmutzler RK, Swoboda H, Reinthaller A, Leodolter S and Zeillinger R. (2000). Int. J. Cancer, 88, 319–322. Vaughn JP, Davis PL, Jarboe MD, Huper G, Evans AC, Wiseman RW, Berchuck A, Iglehart JD, Futreal AP and Marks JR. (1996). Cell Growth Differ., 7, 711–715. Wang A, Pierce AM, Judson-Kremer K, Schneider-Broussard R, MacLeod MC and Johnson DG. (1999). Nucleic Acids Res., 27, 4609–4618. Wang A, Schneider-Broussard R, Kumar AP, MacLeod MC and Johnson DG. (2000). J. Biol. Chem., 275, 4532–4536. Wang Q, Zhang H, Kajino K and Greene M. (1998). Oncogene, 17, 1939–1948. Weaver Z, Montagna C, Xu X, Howard T, Gadina M, Brodie SG, Deng CX and Ried T. (2002). Oncogene, 21, 5097–5107. Williams BO, Remington L, Albert DM, Mukai S, Bronson RT and Jacks T. (1994). Nat. Genet., 7, 480–484. Wilson CA, Payton MN, Elliott GS, Buaas FW, Cajulis EE, Grosshans D, Ramos L, Reese DM, Slamon DJ and Calzone FJ. (1997). Oncogene, 14, 1–16. Wu LC, Wang ZW, Tsan JT, Spillman MA, Phung A, Xu XL, Yang MC, Hwang LY, Bowcock AM and Baer R. (1996). Nat. Genet., 14, 430–440. Xu X, Qiao W, Linke SP, Cao L, Li WM, Furth PA, Harris CC and Deng CX. (2001). Nat. Genet., 28, 266–271. Oncogene
Inactivation of Brca1 in epithelial tissues TR Berton et al
5426 Xu X, Wagner KU, Larson D, Weaver Z, Li C, Reid T, Hennighausen L, Wynshaw-Boris A and Deng CX. (1999a). Nat. Genet., 22, 37–43. Xu X, Weaver KU, Linke SP, Li C, Gotay J, Wang XW, Harris CC, Reid T and Deng CX. (1999b). Mol. Cell, 3, 389–395. Yarden RI and Brody LC. (1999). Proc. Natl. Acad. Sci. USA, 96, 4983–4988.
Oncogene
Yu X, Wu LC, Bowcock AM, Aronheim A and Baer R. (1998). J. Biol. Chem., 273, 25388–25392. Zhang H, Somasundaram K, Peng Y, Tian H, Bi D, Weber BL and El-Deiry WS. (1998). Oncogene, 16, 1713–1721. Zhong Q, Chen CF, Li S, Wang CC, Xiao J, Chen PL, Sharp ZD and Lee WH. (1999). Science, 285, 747–750.
