[CANCER RESEARCH 59, 2438–2444, May 15, 1999] Elevated Frequency of Loss of Heterozygosity in Mammary Tumors Arising in Mouse Mammary Tumor Virus/neu Transgenic Mice1 Marc Cool and Paul Jolicoeur2 Laboratory of Molecular Biology, Clinical Research Institute of Montreal, Montreal, Quebec, H2W 1R7 Canada [M. C., P. J.]; Department of Microbiology and Immunology, University of Montreal, Montreal, Quebec, H3C 3J7 Canada [P. J.]; and Division of Experimental Medicine, McGill University, Montreal, Quebec, H3G 1A4 Canada [P. J.] ABSTRACT LOH. The number and identity of most tumor suppressor genes involved in human breast cancers are unknown, and only a few of Loss of heterozygosity (LOH) analysis was performed on 62 mammary them have been identified; p53 (16), RB1 (17, 18), PTEN/MMAC1 tumors that were induced in (BALB/c 3 C57BL/6)F mouse mammary 1 (19, 20), E-cadherin (21, 22), FHIT (23, 24), TSG101 (25), and tumor virus/neu transgenic mice. Eighty-six simple sequence length poly- INK4a morphism markers were used to cover all of the somatic chromosomes. p16 (26, 27) have all been found to be inactivated at various Frequency of LOH was observed to be significant for chromosomes 4 frequencies in sporadic breast carcinoma. In familial breast cancer, (50%), 19 (32%), and 8 (21%). On chromosome 4, at least three distinct inactivation of BRCA1 and BRCA2 is a relatively frequent event regions of allelic deletions could be identified: one proximal to 22 cM; the (28, 29). second close to the p16INK4a/p15INK4b locus, which is commonly deleted in To study the molecular nature of the erbB-2/neu transformation various tumors; and the third one in the proximity of Mom1. The fre- event of the mammary glands, we (30) and others (31) have generated quency of LOH on chromosome 19 was the same for the four markers Tg mouse animal models (MMTV/neu), which are more amenable to used. Our data suggested the presence of two distinct LOH loci, one experimentation. These mice harbor a transgene consisting of a full- proximal to 47 cM and the other at the distal region. On chromosome 8, length neu cDNA, with a Val-6643Glu mutation located in the possibly two distinct LOH loci could be recognized, one around 52 cM and transmembrane domain, expressed under the regulatory sequences of the other one at 67 cM or distal to it. These regions map close to E-cadherin (Cdh1) and M-cadherin (Cdh15) loci, respectively. Because the MMTV promoter (30). These MMTV/neu Tg mice developed LOH sites are thought to harbor tumor suppressor genes, this allelotype mammary carcinoma that was pathologically very similar to the screening has allowed the mapping of putative tumor suppressor genes human tumors. Although, mammary tumors have been reported to that may be implicated, in collaboration with the erbB-2/neu oncogene, in arise after a short latent period in some MMTV/neu Tg mice (31), the development of mammary tumors in these transgenic mice. most of these MMTV/neu Tg mice develop tumors stochastically and after a long latency (30, 32), strongly suggesting that the erbB-2/neu INTRODUCTION oncogene is not by itself sufficient for tumor formation and that other genetic events (such as activation of other oncogenes and inactivation The erbB-2/neu gene encodes a growth factor receptor transmem- of recessive oncogenes) are likely to contribute to the development of brane tyrosine kinase (1, 2). This gene is frequently amplified and these malignant tumors. overexpressed in human breast cancers, and this change is thought to In an attempt to identify novel tumor suppressor genes that may be contribute at least partially to malignancy (reviewed in Ref. 3). Other involved, in collaboration with the activated erbB-2/neu oncogene, in genetic modifications, such as activation of other oncogenes or inac- the appearance of these tumors, we used the LOH analysis on tumors tivation of tumor suppressor genes, in addition to the increased ex- 3 from (BALB/c C57BL/6)F1 MMTV/neu Tg mice. For this study, pression of the erbB-2/neu gene, are likely to be involved in the we used 62 mouse mammary tumors from MMTV/neu Tg animals development of these malignant tumors, as it has been elegantly 3 and performed a genome-wide analysis with 86 microsatellite SSLP demonstrated in other types of human tumors (4). In fact, LOH is the markers to search for LOH sites. A few specific regions of LOH on 3 most common type of mutation in human primary breast carcinoma chromosomes were identified. We are proposing that these regions (5). It affects at least 20 regions of the genome, encompassing the with allelic losses harbor tumor suppressor genes involved in the following human chromosome arms: 1p, 1q, 3p, 6q, 7q, 8p, 11p, 11q, development of these mammary tumors, in collaboration with the 13q, 16q, 17p, 17q, 18q, 22q (6–8), 2p (9), 7p (10), 8q, 9p, 15q (11), erbB-2/neu oncogene. 10q (9), 16p (12), Xp, and Xq (13). Some and maybe all of these regions of allelic loss are likely to harbor a tumor suppressor gene. This concept has now been validated by the cloning of several tumor MATERIALS AND METHODS suppressor genes in regions of LOH (14). This idea stems from the two-hit model of genetic inactivation of tumor suppressor genes by Mice. The MMTV/neu Tg mice line MN-10 has been described previously (30). They were initially produced in a (C57BL/6 3 C3H)F background and Knudson (15), stating that inactivation of a tumor suppressor gene 2 were subsequently maintained on a BALB/c background by crossing with male requires mutations affecting both alleles independently, the first event or female BALB/c mice (Charles River, St-Constant, Quebec, Canada) for 10 being a small germ-line mutation of a single allele, in the case of generations. At the 10th generation, the MMTV/neu Tg males were bred with familial cancer, and the second being a mutational event on the other C57BL/6 females (Charles River, St-Constant) to obtain (BALB/c 3 C57BL/ allele, often consisting of a large chromosomal deletion scored as 6)F1 MMTV/neu Tg female mice, which developed spontaneous mammary tumors. The frequency and characterization of these tumors have been de- Received 10/5/98; accepted 3/18/99. scribed (30). The costs of publication of this article were defrayed in part by the payment of page DNA Preparation. Mammary tumors and kidneys were collected from F1 charges. This article must therefore be hereby marked advertisement in accordance with Tg females. Kidneys were used as source for normal DNA. DNA was extracted 18 U.S.C. Section 1734 solely to indicate this fact. as described previously (33) in 10 mM Tris (pH 7.5), 0.4 M NaCl, 2 mM EDTA, 1 This work was supported by a grant from the Canadian Breast Cancer Research Initiative (National Cancer Institute of Canada; to P. J.). 0.5% SDS, and 0.5 mg/ml Pronase buffer for overnight at 65°C, followed by 2 To whom requests for reprints should be addressed, at Clinical Research Institute of phenol extraction, then chloroform/isoamyl alcohol (24:1) extraction. DNA Montreal, 110 Pine Avenue West, Montreal, Quebec, H2W 1R7 Canada. Phone: was precipitated in 2 volumes of ethanol. (514) 987-5569; Fax: (514) 987-5794; E-mail: [email protected]. Microsatellites. The 86 microsatellite markers (SSLPs) used (Table 1) 3 The abbreviations used are: LOH, loss of heterozygosity; Tg, transgenic; MMTV, mouse mammary tumor virus; SSLP, simple sequence length polymorphism; AI, allelic have been described by Dietrich et al. (34, 35). They encompass all 20 somatic imbalance; Naad, neu-associated allelic deletion. chromosomes and exhibit polymorphism at the C57BL/6 and BALB/c alleles. 2438 Downloaded from cancerres.aacrjournals.org on September 25, 2021. © 1999 American Association for Cancer Research. LOH ANALYSIS IN neu-INDUCED MOUSE MAMMARY TUMORS Table 1 Characteristics of the 86 microsatellite markers used in the study of 62 background for 20 generations. In addition, in one region of chromosome 7, 3 mammary tumors from (BALB/c-neu C57BL/6)F1, mice whose a large portion has remained heterozygote, the transgene may have cM position cM position cM position integrated in the C57BL/6 allele and been selected for. Strain bias was SSLP marker (MGD) SSLP marker (MGD) SSLP marker (MGD) observed to deviate from the expected 50% distribution and was confirmed by D1Mit70 17.8 D7Mit55 15.0 D13Mit16 10.0 a x2 statistical test. D1Mit132 43.1 D7Mit62 44.0 D13Mit20 35.0 D1Mit187 62.0 D7Mit68 60.0 D13Mit226 59.0 D1Mit106 85.0 D7Mit259 72.4 D13Mit151 71.0 RESULTS D1Mit113 92.3 D1Mit361 101.5 D8Mit4 14.0 D14Mit50 6.0 Characteristics of the Mammary Tumors. The MMTV/neu Tg D8Mit69 31.0 D14Mit60 15.0 D2Mit45 53.0 D8Mit86 52.0 D14Mit37 27.5 mice (line MN-10) described previously (30) and used in the present D2Mit340 68.9 D8Mit91 67.0 D14Mit34 40.0 study have been bred on a BALB/c background for 10 generations. D2Mit456 86.3 D8Mit280 72.0 D14Mit75 54.0 Both male and female MMTV/neu Tg mice were crossed with normal D2Mit230 107.0 D15Mit226 11.6 C57BL/6 mice to produce F1 mice. The F1 Tg female mice were bred D3Mit164 2.4 D9Mit247 17.0 D15Mit159 49.6 continuously to multiply the number of pregnancies and were allowed D3Mit182 23.3 D9Mit162 30.0 to age until they spontaneously develop mammary tumors between the D3Mit29 45.2 D9Mit269 43.0 D16Mit9 4.0 D3Mit189 49.7 D9Mit18 71.0 D16Mit4 27.3 ages of 10 and 26 months.