- Email: [email protected]
Synergistic effect of MNU and DMBA in mammary carcinogenesis and H-ras activation in female Sprague–Dawley rats
Cancer Letters 120 (1997) 87–93
Synergistic effect of MNU and DMBA in mammary carcinogenesis and H-ras activation in female Sprague–Dawley rats Katsumasa Shirai a, Yoshiko Uemura a, Manabu Fukumoto b, Tetsuya Tsukamoto c, Rhett Pascual c, Satyabrata Nandi c, Airo Tsubura a ,* a
Department of Pathology, Kansai Medical University, Moriguchi, Osaka 570, Japan Department of Pathology, Graduate School of Medicine, Kyoto University, Sakyo-ku, Kyoto 606-01, Japan c Cancer Research Laboratory, Department of Molecular and Cell Biology, University of California, Berkeley, CA 94720, USA b
Received 26 May 1997; accepted 1 June 1997
Abstract The combined application of N-methyl-N-nitrosourea (MNU) and 7,12-dimethylbenz[a]anthracene (DMBA) was compared with the administration of each carcinogen alone as to the effectiveness of the induction of mammary carcinomas and the influence of H-ras oncogene activation in female Sprague–Dawley rats. At 50 days of age, group 1 received 30 mg/kg MNU intraperitoneally (i.p.), group 2 received 30 mg/kg DMBA i.p., group 3 received 60 mg/kg MNU i.p., group 4 received 60 mg/ kg DMBA i.p., group 5 received 30 mg/kg MNU followed by 30 mg/kg DMBA i.p., group 6 received 30 mg/kg MNU i.p. and then 30 mg/kg DMBA intravenously (i.v.) and group 7 remained untreated. Animals were killed when the largest mammary tumor reached 1–2 cm in diameter or were necropsied when they were 30 weeks of age. MNU i.p. produced no deaths (groups 1 and 3), however, the i.p. administration of DMBA induced death due to peritonitis (groups 2, 4 and 5), whereas the i.v. administration of DMBA suppressed the death (group 6). All of the tumors produced by MNU were adenocarcinomas of mammary origin. In contrast, DMBA produced tumors of other than mammary origin. The combined treatment with DMBA and MNU increased the mammary carcinogenic effect; it significantly increased the mean number of mammary cancers per rat. With either carcinogen alone and in combination, the mammary carcinomas produced identical adenocarcinoma histology. Of the mammary carcinomas induced by the combined application of MNU and DMBA (group 6), all 11 tumors from five rats showed the GGA to GAA transitional mutation in H-ras codon 12 (38%) and all 18 tumors from the other 10 rats remained as wild-type. An H-ras point mutation at codon 61 was not detected. 1997 Elsevier Science Ireland Ltd. Keywords: MNU; DMBA; Mammary carcinoma; H-ras; Sprague–Dawley rat
1. Introduction Carcinogenic action in the rat mammary gland occurs after a single exposure to a chemical carcinogen. The two most widely used experimental carcinoma systems are the models in which tumors are * Corresponding author. Tel.: +81 6 9921001; fax: +81 6 9925023.
induced in susceptible female rats at puberty by 7,12-dimethylbenz[a]anthracene (DMBA) [1] or by N-methyl-N-nitrosourea (MNU) [2,3]. The mammary tumors induced by DMBA and MNU are similar in terms of both hormone responsiveness and histology and are currently in widespread use in the study of human breast cancer [4]. In humans, it is not exactly known what causes breast cancer; probably, it is the result of a combination of factors, genetic as well as
0304-3835/97/$17.00 1997 Elsevier Science Ireland Ltd. All rights reserved PII S0304-3835 (97 )0 0293-0
88
K. Shirai et al. / Cancer Letters 120 (1997) 87–93
environmental (exposure to more than one carcinogen). In order to fully elucidate human mammary carcinogenesis an experimental system that mimics the human carcinoma more closely is needed. To date, the combined administration of DMBA and MNU to rats has not been precisely studied. In the present study, therefore, the tumor incidence, number of mammary carcinomas per rat, latency and histology of induced tumors were compared among rats administered a single carcinogen insult with the goal of establishing a model more closely resembling human breast cancer. Carcinogenesis is a multistep process associated with an accumulation of genetic alterations. Members of the ras gene family are the most commonly detected oncogenes in human tumors and in carcinogen-induced animal tumors and might play a causative role in the development of these tumors [5,6]. Animal models with well-defined etiological agents which demonstrate the reproducible induction of a specific tumor type provide the means of studying the molecular events involved in the tumorigenesis. Experimental model systems have revealed that a ras mutation occurs in a species- and tumor-specific pattern and site-specific activation of the H-ras gene is produced by different chemical carcinogens in rat mammary carcinogenesis [7]. Members of the ras gene family are known to acquire transforming properties by single-point mutations in two hot spots within their coding sequences, at the 12th and 61st codons [6]. In the majority of MNU-induced rat mammary carcinomas, the H-ras oncogene is activated by a G to A transition at the second nucleotide of codon 12 encoding glutamic acid in place of glycine [8,9]. In DMBA-induced rat mammary carcinomas, although there is also a conflicting report [10], point mutations within the 61st codon, the other hot spot for the activation of ras genes, have been demonstrated [9]. The activation of ras genes seems to depend on the type of carcinogen used. Our goal was to determine the site and type of ras gene mutation after the combined application of MNU and DMBA. Since the MNUand DMBA-induced mammary carcinomas show identical morphology, the purpose of the present study was to evaluate the effects of these two potent carcinogens on the same target organ to clarify the synergism between the two carcinogens and to detect the molecular events of ras gene mutation.
2. Materials and methods 2.1. Animals Eighty 6-week-old virgin female Sprague–Dawley rats were obtained from Clea Japan, Osaka. The animals were housed in plastic cages, three to five rats per cage with wood-chip bedding in a temperature (22 ± 2°C)- and humidity (60 ± 10%)-controlled animal room under a 12 h light/12 h dark cycle. They were fed a commercial pellet diet (CMF; Oriental Yeast, Kyoto, Japan) and water freely throughout the study. 2.2. Carcinogens N-methyl-N-nitrosourea (MNU) was obtained from Nacalai Tesque (Kyoto, Japan). Upon its arrival, the MNU was stored at −20°C in the dark. MNU solution was freshly prepared and dissolved in physiological saline containing 0.05% acetic acid just before use. Lipid emulsions containing 7,12-dimethylbenz[a] anthracene (DMBA) was synthesized by Upjohn (Kalamazoo, MI, USA). 2.3. Experimental procedures At 50 days of age, the animals were randomly divided into seven groups as follows: group 1, nine rats received a single intraperitoneal (i.p.) dose of 30 mg/kg MNU solution; group 2, nine rats received a single i.p. dose of 30 mg/kg DMBA emulsion; group 3, 10 rats received a 60 mg/kg i.p. dose of MNU solution; group 4, 10 rats received a 60 mg/kg i.p. dose of DMBA emulsion; group 5, 16 rats received a 30 mg/kg i.p. dose of MNU solution followed by a 30 mg/kg i.p. dose of DMBA emulsion after a 3 h interval; group 6, 20 rats received a 30 mg/kg i.p. dose of MNU solution and then a 30 mg/kg intravenous (i.v.) dose of DMBA 3 h later; group 7, six rats remained untreated. The i.p. injection of MNU solution and/or DMBA emulsion was made along the ventral midline of the rat abdomen [11,12] and the i.v. injection of DMBA emulsion was made in the tail vein. Following carcinogen treatment, all rats were weighed once weekly and palpated twice weekly for the detection of mammary tumors. The animals were killed when the largest tumor reached 1–2 cm in lar-
K. Shirai et al. / Cancer Letters 120 (1997) 87–93
gest diameter, or when the animals reached 30 weeks of age (when the experiments were terminated). Moribund animals were killed and autopsied. At necropsy, the rats were lightly anesthetized with ether and the mammary tumors as well as all other grossly abnormal organs or tissues were removed. All mammary tumors and suspected areas were fixed in 10% buffered formalin; parts of the large mammary tumors (.1 cm in diameter) were used for gene analysis. All tissues excised at necropsy were paraffinembedded, stained with hematoxylin and eosin and processed for routine histopathological examination. Mammary tumors were classified histopathologically according to previously published criteria [12–14]. 2.4. Gene analysis 2.4.1. DNA extraction DNA was obtained from fresh tumor tissue weighing approximately 50 mg using DNA zol reagent (Life Technologies, Grand Island, NY, USA) according to the manufacturer’s instructions. 2.4.2. Polymerase chain reaction (PCR) PCR reactions were performed using primer pairs to amplify the exon 1 covering codon 12 and exon 2 covering codon 61 of H-ras, respectively (Table 1). The reaction mixture contained 10 ml genomic DNA, 1.5 mM MgCl2, 200 mM dNTPs, 400 nM of each primer, 2 units of Taq polymerase (Toyobo, Osaka, Japan) and PCR reaction buffer provided by the manufacturer (Toyobo) in a total volume of 50 ml. The reaction performed was preheating at 94°C for 5 min followed by 30 cycles of denaturation at 94°C for 1 min, annealing at 55°C for 2 min and extension at 75°C for 2 min followed by an additional extension at 75°C for 5 min. Just before the preparation of the PCR mixture, TaqStart Antibody (Clontech, Palo Alto, CA, USA), which gave the advantage of ‘hot start’ PCR, was applied to the Taq polymerase. The
89
amplified products were confirmed by electrophoresis in 12% polyacrylamide gels and visualized with ethidium bromide staining. 2.4.3. Direct sequencing of amplified samples PCR products were pretreated by the combination of exonuclease I and shrimp alkaline phosphatase (Sequenase PCR product Sequencing Kit, Amersham Life Science, Arlington Heights, IL, USA). Direct nucleotide sequencing was then performed by the dideoxy chain termination method using Sequencing PRO (Toyobo), in which sequencing primers (exon 1, 5′-TGGTTTGGCAACCCCTGTAGA-3′ sense; exon 2, 5′-AGGACTCTACCGGAAACAGGTA-3′ sense) radiolabeled with [g-32P] ATP were used according to the manufacturer’s instructions. The reaction mixtures were denatured at 94°C for 5 min and then for exon 1, 20 cycles programmed with 95°C for 30 s, 62°C for 30 s and 72°C for 30 s and for exon 2, 30 cycles programmed with 95°C for 30 s and 72°C for 2 min were performed. After the reaction was completed, the products were fractionated by electrophoresis on an 8% polyacrylamide gel containing 8 M urea at 30 W for 2.5 h. The gel was dried on filter paper and exposed to Fuji Medical X-ray film (Fuji Photo Film, Tokyo, Japan) at −80°C for 24–36 h. 2.4.4. Allele-specific oligonucleotide hybridization (ASOH) ASOH was carried out as previously described [15,16]. 2.5. Statistics The comparisons of the mammary carcinoma incidence among the groups were performed by Fisher’s exact probability test and the mean number of mammary cancers per rat among groups were examined by Welch’s t-test.
Table 1 Primers used for PCR H-ras exon 1 H-ras exon 2
5′-TGGTTTGGCAACCCCTGTAGA-3′ 5′-AGTGGGATCATACTCGTCCAC-3′ 5′-AGGACTCCTACCGGAAACAG-3′ 5′-ACCTGTACTGATGGATGTC-3′
3. Results The intraperitoneal injection of 30 or 60 mg/kg MNU solution into 50-day-old rats induced no death in any of the rats (groups 1 and 3). Death due to panperitonitis was seen in groups 2, 4 and 5 (four,
90
K. Shirai et al. / Cancer Letters 120 (1997) 87–93
Table 2 Mammary carcinoma incidence following MNU or DMBA treatment or combined application of both agents in female Sprague–Dawley rats Group
Treatment
No. of rats examined
Effective no. of rats
Mammary cancer-bearing rats (%)
Mean no. of mammary cancers/ effective rats (%)
Latency (weeks)
Other umors
1 2 3 4 5
30 mg/kg MNU i.p. 30 mg/kg DMBA i.p. 60 mg/kg MNU i.p. 60 mg/kg DMBA i.p. 30 mg/kg MNU i.p. + 30 mg/kg DMBA i.p. 30 mg/kg MNU i.p. + 30 mg/kg DMBA i.v. No treatment
9 9 10 10 16
9 5 10 6 14
4 1 6 3 11
(44%) (20%) (60%) (50%) (78%)
6/9 (0.7) 5/5 (1.0) 8/10 (0.8) 7/6 (1.2) 39/14 (2.8)
19.5 8 20.2 19.7 19.7
0 1a 0 3b 3c
20
20
16 (80%)
76/20 (3.8)
17.8
5d
6
6
–
–
0
6 7
0
a
One intraabdominal sarcoma. Three intraabdominal sarcomas. Two intraabdominal sarcomas and one forestomach squamous cell carcinoma. d One mammary carcinosarcoma, two fibroadenomas, one ear duct tumor and one nephroblastoma. b c
four and two rats, respectively) in which DMBA emulsions were injected intraperitoneally. These animals were excluded from the analysis. In contrast, i.v. DMBA in combination with i.p. MNU caused no death (group 6). The growth curves as determined by the body weight gain of the effective rats in each carcinogen-treated group (groups 1–6) were comparable to that of the control rats (group 7). Table 2 shows the effective numbers of rats, mammary cancer-bearing rats, mean number of mammary cancers per rat, latency and other tumors induced by DMBA, MNU or the combined application of both agents. When DMBA was administered alone or in combination with MNU, tumors of other than mammary origin, such as intraabdominal sarcoma, forestomach squamous cell carcinoma, ear duct tumor and nephroblastoma, were induced. The administration of each carcinogen alone at increasing doses (MNU, group 1 versus group 3; DMBA, group 2 versus group 4) resulted in slight but not significant enhancing effects; the incidence of mammary cancer and mean number of cancers per rat increased. The administration of the combination of MNU and DMBA (groups 5 and 6) resulted in greater enhancing effects with increased mammary cancer incidence and number of mammary cancers per rat in comparison with the administration of either 60 mg/kg MNU or 60 mg/kg DMBA alone (groups 3 and 4, respectively). Moreover, the i.v. administration of DMBA in combination with i.p. MNU (group 6) was more effective than the administration
of i.p. DMBA in combination with i.p. MNU (group 5). The mean number of mammary cancers per rat in group 6 compared to those in groups 3 and 4 was significantly higher (P , 0.05, respectively). However, the combined application did not shorten the tumor latency. Regardless of the type of carcinogen treatment, MNU or DMBA alone or in combination, all of the mammary cancers had similar adenocarcinoma histo-pathology. Fig. 1 shows a representative adenocarcinoma histology induced by the combined application of MNU and DMBA. The untreated rats showed no evidence of tumor formation at the end of the experiment (group 7). Thirty mammary tumors .1 cm in diameter, defined as adenocarcinoma by histology and taken from 16 rats that received both MNU and DMBA (group 6), were examined for H-ras codons 12 and 61 activation (Table 3). One to four tumors were examined from each rat. However, PCR amplification was unsuccessful in one tumor (rat 7) and was excluded from the examination. The results of nucleotide sequencing in representative cases are shown in Fig. 2. The G to A transition at the second position of codon 12 was detected in 11 tumors (38%) in which the mutation was heterozygous, while the other 18 tumors did not show a point mutation (Fig. 2). Notable, the G to A mutation was seen in all the above-mentioned 11 tumors from five rats (rats 1, 3, 5, 9 and 12), whereas all 18 tumors from the other 10 rats remained as wild-type (rats 2, 4, 6, 8,
91
K. Shirai et al. / Cancer Letters 120 (1997) 87–93 Table 3
H-ras mutation in mammary adenocarcinomas derived from the combined application of MNU and DMBA Rat
Fig. 1. Histological appearance of a mammary carcinoma induced by the combined application of MNU and DMBA. Histology shows solid and papillary structures (HE, ×200).
10, 11 and 13–16). The direct sequencing and ASOH showed similar results. In contrast, all 29 tumors showed wild-type CAA at codon 61 in the H-ras oncogene in which no point mutation was detected (Fig. 3).
4. Discussion Mammary carcinomas with identical histology were induced by two different carcinogens, MNU and DMBA alone, and by the combined application of both agents. The route of carcinogen administration is important for certain types of tumor induction. MNU, originally administered i.v. to female rats during puberty, induced a high yield of mammary carcinomas [,2]. However, subcutaneous injection [17] and direct applications [12] of MNU are also effective. The i.p. route is another reliable route for MNU administration [11]. In the present study, i.p. MNU alone at a dose of 30 or 60 mg/kg selectively induced adenocarcinoma of the mammary gland. The i.p. inoculation of rats with DMBA either alone or in combination with MNU evoked panperitonitis which led to death, to sarcoma induction originating in the peritoneal cavity, or to other tumors. The i.p. route of DMBA administration has previously shown targets other than the mammary gland [1]. The incidence of
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 a
No. of tumors examined 2 3 1 1 3 1 1 3 4 1 2 1 1 3 1 2
H-ras mutation Codon 12a
Codon 61
2 0 1 0 3 0 – 0 4 0 0 1 0 0 0 0
0 0 0 0 0 0 – 0 0 0 0 0 0 0 0 0
GGA to GAA translational mutation.
mammary carcinoma was greater in the present groups given the combined treatment of MNU and DMBA than in the groups given these agents singly. The i.p. administration of MNU combined with the i.v. administration of DMBA resulted in no deaths, no intraabdominal sarcoma was induced and the mean number of mammary carcinomas per rat was significantly higher (group 6 versus groups 3 and 4, respectively). Thus, since the dose and ratio of MNU and DMBA might be considered further [19], the dosing schedule of 30 mg/kg MNU i.p. followed by 30 mg/kg DMBA i.v. was the most effective method of inducing mammary adenocarcinomas. The present regime
Fig. 2. Nucleotide sequence around codon 12 of the H-ras gene. G to A transition was detected in the second position of codon 12 (arrowhead). Lanes 1 and 2, mammary carcinoma from rat 1 which showed a point mutation; lanes 3–5, mammary carcinoma from rat 8 without mutation.
92
K. Shirai et al. / Cancer Letters 120 (1997) 87–93
Acknowledgements The authors thank Ms T. Akamatsu for her technical support and Ms M. Fukuchi for preparing the manuscript. This work was supported in part by the Science Research Promotion Fund of the Japanese Private School Promotion Foundation (1996) and by the Katano Foundation (1995). Fig. 3. Nucleotide sequence around codon 61 of the H-ras gene. Wild-type CAA genotype is seen. Lanes 1 and 2, mammary carcinoma from rat 1.
showed a syncarcinogenic effect; although the latency was not shortened, the two agents acted together in the carcinogenesis with the result that tumors developed more frequently at the predetermined site [18]. A high frequency (.80%) of the G to A transition mutation in H-ras codon 12 is a hallmark of MNUinduced rat mammary carcinogenesis [8,10] and 21% of DMBA-induced rat mammary tumor shows the activation of H-ras codon 61 [10]. In the present study, the same type of tumor was induced by different carcinogens which carried different H-ras activations. However, the H-ras codon 61 mutation was not seen in any of the 50 DMBA-induced rat mammary tumors in inbred Sprague–Dawley rats [9]. Following the present combined application of MNU and DMBA, the point mutation at H-ras codon 61 was not detected. Since we used outbred Sprague–Dawley rats, the strain may have influenced the H-ras activation at codon 61. In contrast, among the 29 mammary adenocarcinomas examined by the PCR/direct sequencing method and by the PCR/ASOH method, 11 tumors (38%) showed a GGA to GAA transitional mutation in H-ras codon 12, i.e. the pattern frequency evoked by MNU, but in lower frequency. Although p53 mutations seem not to be involved in MNUinduced mammary carcinogenesis [20], other mutations might be necessary for rat mammary carcinogenesis. However, among the .1 cm mammary adenocarcinomas examined, the activation was hostdependent in that all 11 tumors derived from five rats showed the H-ras codon 12 mutation (GGA/GAA) and the other 18 tumors obtained from 10 other rats were wild-type (GAA/GAA). This phenomenon should be further investigated and since H-ras is not the only oncogene activated in rat mammary tumors [21], the other oncogene(s) involved must also be investigated.
References [1] C.B. Huggins, Experimental Leukemia and Mammary Cancer, University of Chicago Press, Chicago, IL, 1979. [2] P.M. Gullino, H.M. Pettigrew, F.H. Grantham, N-Nitrosomethylurea as mammary gland carcinogen in rats, J. Natl. Cancer Inst. 54 (1975) 401–414. [3] D.L. McCormick, C.B. Adamowski, A. Fiks, R.C. Moon, Lifetime dose-response relationships for mammary tumor induction by a single administration of N-methyl-N-nitrosourea, Cancer Res. 41 (1981) 1690–1694. [4] J. Russo, B.A. Gusterson, A.E. Rogers, I.H. Russo, S.R. Wellings, M.J. van Zwieten, Comparative study of human and rat mammary tumorigenesis, Lab. Invest. 62 (1990) 244–278. [5] J.L. Bos, ras oncogenes in human cancer: a review, Cancer Res. 49 (1989) 4682–4689. [6] S. Sukumar, ras oncogenes in chemical carcinogenesis, Curr. Topics Microbiol. Immunol. 148 (1989) 93–114. [7] H. Zarbl, S. Sukumar, A.L. Arthur, D. Martin-Zanca, M. Barbacid, Activation of H-ras-1 oncogenes by chemical carcinogens, Basic Life Sci. 38 (1986) 385–397. [8] S. Sukumar, V. Notario, D. Martin-Zanca, M. Barbacid, Induction of mammary carcinomas in rats by nitroso-methylurea involves malignant activation of H-ras-1 locus by single point mutations, Nature 306 (1983) 658–661. [9] H. Zarbl, S. Sukumar, A.V. Arthur, D. Martin-Zanca, M. Barbacid, Direct mutagenesis of H-ras-1 oncogenes by Nnitroso-N-methylurea during initiation of mammary carcinogenesis in rats, Nature 315 (1985) 382–385. [10] V. Waldmann, B. Suchy, H.M. Rabes, Cell proliferation and prevalence of ras gene mutations in 7,12-dimethylbenz[\ a]anthracene (DMBA)-induced rat mammary tumors, Res. Exp. Med. 193 (1993) 143–151. [11] H.J. Thompson, H. Adlakha, Dose-responsive induction of mammary gland carcinomas by the intraperitoneal injection of l-methyl-l-nitrosourea, Cancer Res. 51 (1991) 3411–3415. [12] H. Takahashi, Y. Uemura, I. Nakao, A. Tsubura, Induction of mammary carcinomas by the direct application of crystalline N-methyl-N-nitrosourea onto rat mammary gland, Cancer Lett. 92 (1995) 105–111. [13] A. Tsubura, N. Shikata, T. Inui, N. Sakaida, S. Morii, Morphological and biological characteristics of mammary tumours induced by the direct application of DMBA powder to rat mammary glands, Virchows Arch. A. 414 (1988) 1– 7.
K. Shirai et al. / Cancer Letters 120 (1997) 87–93 [14] J. Russo, I.H. Russo, A.E. Rogers, M.J. van Zwieten, B. Gusterson, Tumours of the mammary gland, in: V. Turusov, U. Mohr, Pathology of Tumours in Laboratory Animals. Vol. 1, Tumours of the Rat, second ed., IARC Scientific Publication No. 99, IARC, Lyon, 1990, pp. 47–78. [15] R.C. Guzman, R.C. Osborn, S.M. Swanson, R. Sakthivel, S.I. Hwang, S. Miyamoto, S. Nandi, Incidence of c-Ki-ras activation in N-methyl-N-nitrosourea-induced mammary carcinomas in pituitary-isografted mice, Cancer Res. 52 (1992) 5732–5737. [16] S.-I. Hwang, R.C. Guzman, S.M. Swanson, S. Nandi, Hormone dependent and independent mammary tumor development from N-methyl-N-nitrosourea-treated rat mammary epithelial cell xenografts in the nude mouse: multiple pathways and H-ras activation, Cancer Lett. 101 (1996) 123–134.
93
[17] H.J. Thompson, L.D. Meeker, Induction of mammary gland carcinomas by the subcutaneous injection of l-methyl- lnitrosourea, Cancer Res. 43 (1983) 1628–1629. [18] M. Habs, D. Schma¨hl, Synergistic effects of N-nitroso compounds in experimental long-term carcinogenesis studies, Oncology 37 (1980) 259–265. [19] M. Berger, K. Ku¨nstler, Experiments on the influence of combined application of N-methylnitrosourea and 7,12dimethylbenzanthracene in female Sprague–Dawley rats, Arch. Geschwulstforsch. 53 (Suppl. 5) (1983) 423–427. [20] K. Ogawa, Y. Tokusashi, I. Fukuda, Absence of p53 mutations in methylnitrosourea-induced mammary tumors in rats, Cancer Detect. Prev. 20 (1996) 214–217. [21] S. Sukumar, K. McKenzie, Y. Chen, Animal models for breast cancer, Mutat. Res. 333 (1995) 37–44.
Synergistic effect of MNU and DMBA in mammary carcinogenesis and H-ras activation in female Sprague–Dawley rats Katsumasa Shirai a, Yoshiko Uemura a, Manabu Fukumoto b, Tetsuya Tsukamoto c, Rhett Pascual c, Satyabrata Nandi c, Airo Tsubura a ,* a
Department of Pathology, Kansai Medical University, Moriguchi, Osaka 570, Japan Department of Pathology, Graduate School of Medicine, Kyoto University, Sakyo-ku, Kyoto 606-01, Japan c Cancer Research Laboratory, Department of Molecular and Cell Biology, University of California, Berkeley, CA 94720, USA b
Received 26 May 1997; accepted 1 June 1997
Abstract The combined application of N-methyl-N-nitrosourea (MNU) and 7,12-dimethylbenz[a]anthracene (DMBA) was compared with the administration of each carcinogen alone as to the effectiveness of the induction of mammary carcinomas and the influence of H-ras oncogene activation in female Sprague–Dawley rats. At 50 days of age, group 1 received 30 mg/kg MNU intraperitoneally (i.p.), group 2 received 30 mg/kg DMBA i.p., group 3 received 60 mg/kg MNU i.p., group 4 received 60 mg/ kg DMBA i.p., group 5 received 30 mg/kg MNU followed by 30 mg/kg DMBA i.p., group 6 received 30 mg/kg MNU i.p. and then 30 mg/kg DMBA intravenously (i.v.) and group 7 remained untreated. Animals were killed when the largest mammary tumor reached 1–2 cm in diameter or were necropsied when they were 30 weeks of age. MNU i.p. produced no deaths (groups 1 and 3), however, the i.p. administration of DMBA induced death due to peritonitis (groups 2, 4 and 5), whereas the i.v. administration of DMBA suppressed the death (group 6). All of the tumors produced by MNU were adenocarcinomas of mammary origin. In contrast, DMBA produced tumors of other than mammary origin. The combined treatment with DMBA and MNU increased the mammary carcinogenic effect; it significantly increased the mean number of mammary cancers per rat. With either carcinogen alone and in combination, the mammary carcinomas produced identical adenocarcinoma histology. Of the mammary carcinomas induced by the combined application of MNU and DMBA (group 6), all 11 tumors from five rats showed the GGA to GAA transitional mutation in H-ras codon 12 (38%) and all 18 tumors from the other 10 rats remained as wild-type. An H-ras point mutation at codon 61 was not detected. 1997 Elsevier Science Ireland Ltd. Keywords: MNU; DMBA; Mammary carcinoma; H-ras; Sprague–Dawley rat
1. Introduction Carcinogenic action in the rat mammary gland occurs after a single exposure to a chemical carcinogen. The two most widely used experimental carcinoma systems are the models in which tumors are * Corresponding author. Tel.: +81 6 9921001; fax: +81 6 9925023.
induced in susceptible female rats at puberty by 7,12-dimethylbenz[a]anthracene (DMBA) [1] or by N-methyl-N-nitrosourea (MNU) [2,3]. The mammary tumors induced by DMBA and MNU are similar in terms of both hormone responsiveness and histology and are currently in widespread use in the study of human breast cancer [4]. In humans, it is not exactly known what causes breast cancer; probably, it is the result of a combination of factors, genetic as well as
0304-3835/97/$17.00 1997 Elsevier Science Ireland Ltd. All rights reserved PII S0304-3835 (97 )0 0293-0
88
K. Shirai et al. / Cancer Letters 120 (1997) 87–93
environmental (exposure to more than one carcinogen). In order to fully elucidate human mammary carcinogenesis an experimental system that mimics the human carcinoma more closely is needed. To date, the combined administration of DMBA and MNU to rats has not been precisely studied. In the present study, therefore, the tumor incidence, number of mammary carcinomas per rat, latency and histology of induced tumors were compared among rats administered a single carcinogen insult with the goal of establishing a model more closely resembling human breast cancer. Carcinogenesis is a multistep process associated with an accumulation of genetic alterations. Members of the ras gene family are the most commonly detected oncogenes in human tumors and in carcinogen-induced animal tumors and might play a causative role in the development of these tumors [5,6]. Animal models with well-defined etiological agents which demonstrate the reproducible induction of a specific tumor type provide the means of studying the molecular events involved in the tumorigenesis. Experimental model systems have revealed that a ras mutation occurs in a species- and tumor-specific pattern and site-specific activation of the H-ras gene is produced by different chemical carcinogens in rat mammary carcinogenesis [7]. Members of the ras gene family are known to acquire transforming properties by single-point mutations in two hot spots within their coding sequences, at the 12th and 61st codons [6]. In the majority of MNU-induced rat mammary carcinomas, the H-ras oncogene is activated by a G to A transition at the second nucleotide of codon 12 encoding glutamic acid in place of glycine [8,9]. In DMBA-induced rat mammary carcinomas, although there is also a conflicting report [10], point mutations within the 61st codon, the other hot spot for the activation of ras genes, have been demonstrated [9]. The activation of ras genes seems to depend on the type of carcinogen used. Our goal was to determine the site and type of ras gene mutation after the combined application of MNU and DMBA. Since the MNUand DMBA-induced mammary carcinomas show identical morphology, the purpose of the present study was to evaluate the effects of these two potent carcinogens on the same target organ to clarify the synergism between the two carcinogens and to detect the molecular events of ras gene mutation.
2. Materials and methods 2.1. Animals Eighty 6-week-old virgin female Sprague–Dawley rats were obtained from Clea Japan, Osaka. The animals were housed in plastic cages, three to five rats per cage with wood-chip bedding in a temperature (22 ± 2°C)- and humidity (60 ± 10%)-controlled animal room under a 12 h light/12 h dark cycle. They were fed a commercial pellet diet (CMF; Oriental Yeast, Kyoto, Japan) and water freely throughout the study. 2.2. Carcinogens N-methyl-N-nitrosourea (MNU) was obtained from Nacalai Tesque (Kyoto, Japan). Upon its arrival, the MNU was stored at −20°C in the dark. MNU solution was freshly prepared and dissolved in physiological saline containing 0.05% acetic acid just before use. Lipid emulsions containing 7,12-dimethylbenz[a] anthracene (DMBA) was synthesized by Upjohn (Kalamazoo, MI, USA). 2.3. Experimental procedures At 50 days of age, the animals were randomly divided into seven groups as follows: group 1, nine rats received a single intraperitoneal (i.p.) dose of 30 mg/kg MNU solution; group 2, nine rats received a single i.p. dose of 30 mg/kg DMBA emulsion; group 3, 10 rats received a 60 mg/kg i.p. dose of MNU solution; group 4, 10 rats received a 60 mg/kg i.p. dose of DMBA emulsion; group 5, 16 rats received a 30 mg/kg i.p. dose of MNU solution followed by a 30 mg/kg i.p. dose of DMBA emulsion after a 3 h interval; group 6, 20 rats received a 30 mg/kg i.p. dose of MNU solution and then a 30 mg/kg intravenous (i.v.) dose of DMBA 3 h later; group 7, six rats remained untreated. The i.p. injection of MNU solution and/or DMBA emulsion was made along the ventral midline of the rat abdomen [11,12] and the i.v. injection of DMBA emulsion was made in the tail vein. Following carcinogen treatment, all rats were weighed once weekly and palpated twice weekly for the detection of mammary tumors. The animals were killed when the largest tumor reached 1–2 cm in lar-
K. Shirai et al. / Cancer Letters 120 (1997) 87–93
gest diameter, or when the animals reached 30 weeks of age (when the experiments were terminated). Moribund animals were killed and autopsied. At necropsy, the rats were lightly anesthetized with ether and the mammary tumors as well as all other grossly abnormal organs or tissues were removed. All mammary tumors and suspected areas were fixed in 10% buffered formalin; parts of the large mammary tumors (.1 cm in diameter) were used for gene analysis. All tissues excised at necropsy were paraffinembedded, stained with hematoxylin and eosin and processed for routine histopathological examination. Mammary tumors were classified histopathologically according to previously published criteria [12–14]. 2.4. Gene analysis 2.4.1. DNA extraction DNA was obtained from fresh tumor tissue weighing approximately 50 mg using DNA zol reagent (Life Technologies, Grand Island, NY, USA) according to the manufacturer’s instructions. 2.4.2. Polymerase chain reaction (PCR) PCR reactions were performed using primer pairs to amplify the exon 1 covering codon 12 and exon 2 covering codon 61 of H-ras, respectively (Table 1). The reaction mixture contained 10 ml genomic DNA, 1.5 mM MgCl2, 200 mM dNTPs, 400 nM of each primer, 2 units of Taq polymerase (Toyobo, Osaka, Japan) and PCR reaction buffer provided by the manufacturer (Toyobo) in a total volume of 50 ml. The reaction performed was preheating at 94°C for 5 min followed by 30 cycles of denaturation at 94°C for 1 min, annealing at 55°C for 2 min and extension at 75°C for 2 min followed by an additional extension at 75°C for 5 min. Just before the preparation of the PCR mixture, TaqStart Antibody (Clontech, Palo Alto, CA, USA), which gave the advantage of ‘hot start’ PCR, was applied to the Taq polymerase. The
89
amplified products were confirmed by electrophoresis in 12% polyacrylamide gels and visualized with ethidium bromide staining. 2.4.3. Direct sequencing of amplified samples PCR products were pretreated by the combination of exonuclease I and shrimp alkaline phosphatase (Sequenase PCR product Sequencing Kit, Amersham Life Science, Arlington Heights, IL, USA). Direct nucleotide sequencing was then performed by the dideoxy chain termination method using Sequencing PRO (Toyobo), in which sequencing primers (exon 1, 5′-TGGTTTGGCAACCCCTGTAGA-3′ sense; exon 2, 5′-AGGACTCTACCGGAAACAGGTA-3′ sense) radiolabeled with [g-32P] ATP were used according to the manufacturer’s instructions. The reaction mixtures were denatured at 94°C for 5 min and then for exon 1, 20 cycles programmed with 95°C for 30 s, 62°C for 30 s and 72°C for 30 s and for exon 2, 30 cycles programmed with 95°C for 30 s and 72°C for 2 min were performed. After the reaction was completed, the products were fractionated by electrophoresis on an 8% polyacrylamide gel containing 8 M urea at 30 W for 2.5 h. The gel was dried on filter paper and exposed to Fuji Medical X-ray film (Fuji Photo Film, Tokyo, Japan) at −80°C for 24–36 h. 2.4.4. Allele-specific oligonucleotide hybridization (ASOH) ASOH was carried out as previously described [15,16]. 2.5. Statistics The comparisons of the mammary carcinoma incidence among the groups were performed by Fisher’s exact probability test and the mean number of mammary cancers per rat among groups were examined by Welch’s t-test.
Table 1 Primers used for PCR H-ras exon 1 H-ras exon 2
5′-TGGTTTGGCAACCCCTGTAGA-3′ 5′-AGTGGGATCATACTCGTCCAC-3′ 5′-AGGACTCCTACCGGAAACAG-3′ 5′-ACCTGTACTGATGGATGTC-3′
3. Results The intraperitoneal injection of 30 or 60 mg/kg MNU solution into 50-day-old rats induced no death in any of the rats (groups 1 and 3). Death due to panperitonitis was seen in groups 2, 4 and 5 (four,
90
K. Shirai et al. / Cancer Letters 120 (1997) 87–93
Table 2 Mammary carcinoma incidence following MNU or DMBA treatment or combined application of both agents in female Sprague–Dawley rats Group
Treatment
No. of rats examined
Effective no. of rats
Mammary cancer-bearing rats (%)
Mean no. of mammary cancers/ effective rats (%)
Latency (weeks)
Other umors
1 2 3 4 5
30 mg/kg MNU i.p. 30 mg/kg DMBA i.p. 60 mg/kg MNU i.p. 60 mg/kg DMBA i.p. 30 mg/kg MNU i.p. + 30 mg/kg DMBA i.p. 30 mg/kg MNU i.p. + 30 mg/kg DMBA i.v. No treatment
9 9 10 10 16
9 5 10 6 14
4 1 6 3 11
(44%) (20%) (60%) (50%) (78%)
6/9 (0.7) 5/5 (1.0) 8/10 (0.8) 7/6 (1.2) 39/14 (2.8)
19.5 8 20.2 19.7 19.7
0 1a 0 3b 3c
20
20
16 (80%)
76/20 (3.8)
17.8
5d
6
6
–
–
0
6 7
0
a
One intraabdominal sarcoma. Three intraabdominal sarcomas. Two intraabdominal sarcomas and one forestomach squamous cell carcinoma. d One mammary carcinosarcoma, two fibroadenomas, one ear duct tumor and one nephroblastoma. b c
four and two rats, respectively) in which DMBA emulsions were injected intraperitoneally. These animals were excluded from the analysis. In contrast, i.v. DMBA in combination with i.p. MNU caused no death (group 6). The growth curves as determined by the body weight gain of the effective rats in each carcinogen-treated group (groups 1–6) were comparable to that of the control rats (group 7). Table 2 shows the effective numbers of rats, mammary cancer-bearing rats, mean number of mammary cancers per rat, latency and other tumors induced by DMBA, MNU or the combined application of both agents. When DMBA was administered alone or in combination with MNU, tumors of other than mammary origin, such as intraabdominal sarcoma, forestomach squamous cell carcinoma, ear duct tumor and nephroblastoma, were induced. The administration of each carcinogen alone at increasing doses (MNU, group 1 versus group 3; DMBA, group 2 versus group 4) resulted in slight but not significant enhancing effects; the incidence of mammary cancer and mean number of cancers per rat increased. The administration of the combination of MNU and DMBA (groups 5 and 6) resulted in greater enhancing effects with increased mammary cancer incidence and number of mammary cancers per rat in comparison with the administration of either 60 mg/kg MNU or 60 mg/kg DMBA alone (groups 3 and 4, respectively). Moreover, the i.v. administration of DMBA in combination with i.p. MNU (group 6) was more effective than the administration
of i.p. DMBA in combination with i.p. MNU (group 5). The mean number of mammary cancers per rat in group 6 compared to those in groups 3 and 4 was significantly higher (P , 0.05, respectively). However, the combined application did not shorten the tumor latency. Regardless of the type of carcinogen treatment, MNU or DMBA alone or in combination, all of the mammary cancers had similar adenocarcinoma histo-pathology. Fig. 1 shows a representative adenocarcinoma histology induced by the combined application of MNU and DMBA. The untreated rats showed no evidence of tumor formation at the end of the experiment (group 7). Thirty mammary tumors .1 cm in diameter, defined as adenocarcinoma by histology and taken from 16 rats that received both MNU and DMBA (group 6), were examined for H-ras codons 12 and 61 activation (Table 3). One to four tumors were examined from each rat. However, PCR amplification was unsuccessful in one tumor (rat 7) and was excluded from the examination. The results of nucleotide sequencing in representative cases are shown in Fig. 2. The G to A transition at the second position of codon 12 was detected in 11 tumors (38%) in which the mutation was heterozygous, while the other 18 tumors did not show a point mutation (Fig. 2). Notable, the G to A mutation was seen in all the above-mentioned 11 tumors from five rats (rats 1, 3, 5, 9 and 12), whereas all 18 tumors from the other 10 rats remained as wild-type (rats 2, 4, 6, 8,
91
K. Shirai et al. / Cancer Letters 120 (1997) 87–93 Table 3
H-ras mutation in mammary adenocarcinomas derived from the combined application of MNU and DMBA Rat
Fig. 1. Histological appearance of a mammary carcinoma induced by the combined application of MNU and DMBA. Histology shows solid and papillary structures (HE, ×200).
10, 11 and 13–16). The direct sequencing and ASOH showed similar results. In contrast, all 29 tumors showed wild-type CAA at codon 61 in the H-ras oncogene in which no point mutation was detected (Fig. 3).
4. Discussion Mammary carcinomas with identical histology were induced by two different carcinogens, MNU and DMBA alone, and by the combined application of both agents. The route of carcinogen administration is important for certain types of tumor induction. MNU, originally administered i.v. to female rats during puberty, induced a high yield of mammary carcinomas [,2]. However, subcutaneous injection [17] and direct applications [12] of MNU are also effective. The i.p. route is another reliable route for MNU administration [11]. In the present study, i.p. MNU alone at a dose of 30 or 60 mg/kg selectively induced adenocarcinoma of the mammary gland. The i.p. inoculation of rats with DMBA either alone or in combination with MNU evoked panperitonitis which led to death, to sarcoma induction originating in the peritoneal cavity, or to other tumors. The i.p. route of DMBA administration has previously shown targets other than the mammary gland [1]. The incidence of
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 a
No. of tumors examined 2 3 1 1 3 1 1 3 4 1 2 1 1 3 1 2
H-ras mutation Codon 12a
Codon 61
2 0 1 0 3 0 – 0 4 0 0 1 0 0 0 0
0 0 0 0 0 0 – 0 0 0 0 0 0 0 0 0
GGA to GAA translational mutation.
mammary carcinoma was greater in the present groups given the combined treatment of MNU and DMBA than in the groups given these agents singly. The i.p. administration of MNU combined with the i.v. administration of DMBA resulted in no deaths, no intraabdominal sarcoma was induced and the mean number of mammary carcinomas per rat was significantly higher (group 6 versus groups 3 and 4, respectively). Thus, since the dose and ratio of MNU and DMBA might be considered further [19], the dosing schedule of 30 mg/kg MNU i.p. followed by 30 mg/kg DMBA i.v. was the most effective method of inducing mammary adenocarcinomas. The present regime
Fig. 2. Nucleotide sequence around codon 12 of the H-ras gene. G to A transition was detected in the second position of codon 12 (arrowhead). Lanes 1 and 2, mammary carcinoma from rat 1 which showed a point mutation; lanes 3–5, mammary carcinoma from rat 8 without mutation.
92
K. Shirai et al. / Cancer Letters 120 (1997) 87–93
Acknowledgements The authors thank Ms T. Akamatsu for her technical support and Ms M. Fukuchi for preparing the manuscript. This work was supported in part by the Science Research Promotion Fund of the Japanese Private School Promotion Foundation (1996) and by the Katano Foundation (1995). Fig. 3. Nucleotide sequence around codon 61 of the H-ras gene. Wild-type CAA genotype is seen. Lanes 1 and 2, mammary carcinoma from rat 1.
showed a syncarcinogenic effect; although the latency was not shortened, the two agents acted together in the carcinogenesis with the result that tumors developed more frequently at the predetermined site [18]. A high frequency (.80%) of the G to A transition mutation in H-ras codon 12 is a hallmark of MNUinduced rat mammary carcinogenesis [8,10] and 21% of DMBA-induced rat mammary tumor shows the activation of H-ras codon 61 [10]. In the present study, the same type of tumor was induced by different carcinogens which carried different H-ras activations. However, the H-ras codon 61 mutation was not seen in any of the 50 DMBA-induced rat mammary tumors in inbred Sprague–Dawley rats [9]. Following the present combined application of MNU and DMBA, the point mutation at H-ras codon 61 was not detected. Since we used outbred Sprague–Dawley rats, the strain may have influenced the H-ras activation at codon 61. In contrast, among the 29 mammary adenocarcinomas examined by the PCR/direct sequencing method and by the PCR/ASOH method, 11 tumors (38%) showed a GGA to GAA transitional mutation in H-ras codon 12, i.e. the pattern frequency evoked by MNU, but in lower frequency. Although p53 mutations seem not to be involved in MNUinduced mammary carcinogenesis [20], other mutations might be necessary for rat mammary carcinogenesis. However, among the .1 cm mammary adenocarcinomas examined, the activation was hostdependent in that all 11 tumors derived from five rats showed the H-ras codon 12 mutation (GGA/GAA) and the other 18 tumors obtained from 10 other rats were wild-type (GAA/GAA). This phenomenon should be further investigated and since H-ras is not the only oncogene activated in rat mammary tumors [21], the other oncogene(s) involved must also be investigated.
References [1] C.B. Huggins, Experimental Leukemia and Mammary Cancer, University of Chicago Press, Chicago, IL, 1979. [2] P.M. Gullino, H.M. Pettigrew, F.H. Grantham, N-Nitrosomethylurea as mammary gland carcinogen in rats, J. Natl. Cancer Inst. 54 (1975) 401–414. [3] D.L. McCormick, C.B. Adamowski, A. Fiks, R.C. Moon, Lifetime dose-response relationships for mammary tumor induction by a single administration of N-methyl-N-nitrosourea, Cancer Res. 41 (1981) 1690–1694. [4] J. Russo, B.A. Gusterson, A.E. Rogers, I.H. Russo, S.R. Wellings, M.J. van Zwieten, Comparative study of human and rat mammary tumorigenesis, Lab. Invest. 62 (1990) 244–278. [5] J.L. Bos, ras oncogenes in human cancer: a review, Cancer Res. 49 (1989) 4682–4689. [6] S. Sukumar, ras oncogenes in chemical carcinogenesis, Curr. Topics Microbiol. Immunol. 148 (1989) 93–114. [7] H. Zarbl, S. Sukumar, A.L. Arthur, D. Martin-Zanca, M. Barbacid, Activation of H-ras-1 oncogenes by chemical carcinogens, Basic Life Sci. 38 (1986) 385–397. [8] S. Sukumar, V. Notario, D. Martin-Zanca, M. Barbacid, Induction of mammary carcinomas in rats by nitroso-methylurea involves malignant activation of H-ras-1 locus by single point mutations, Nature 306 (1983) 658–661. [9] H. Zarbl, S. Sukumar, A.V. Arthur, D. Martin-Zanca, M. Barbacid, Direct mutagenesis of H-ras-1 oncogenes by Nnitroso-N-methylurea during initiation of mammary carcinogenesis in rats, Nature 315 (1985) 382–385. [10] V. Waldmann, B. Suchy, H.M. Rabes, Cell proliferation and prevalence of ras gene mutations in 7,12-dimethylbenz[\ a]anthracene (DMBA)-induced rat mammary tumors, Res. Exp. Med. 193 (1993) 143–151. [11] H.J. Thompson, H. Adlakha, Dose-responsive induction of mammary gland carcinomas by the intraperitoneal injection of l-methyl-l-nitrosourea, Cancer Res. 51 (1991) 3411–3415. [12] H. Takahashi, Y. Uemura, I. Nakao, A. Tsubura, Induction of mammary carcinomas by the direct application of crystalline N-methyl-N-nitrosourea onto rat mammary gland, Cancer Lett. 92 (1995) 105–111. [13] A. Tsubura, N. Shikata, T. Inui, N. Sakaida, S. Morii, Morphological and biological characteristics of mammary tumours induced by the direct application of DMBA powder to rat mammary glands, Virchows Arch. A. 414 (1988) 1– 7.
K. Shirai et al. / Cancer Letters 120 (1997) 87–93 [14] J. Russo, I.H. Russo, A.E. Rogers, M.J. van Zwieten, B. Gusterson, Tumours of the mammary gland, in: V. Turusov, U. Mohr, Pathology of Tumours in Laboratory Animals. Vol. 1, Tumours of the Rat, second ed., IARC Scientific Publication No. 99, IARC, Lyon, 1990, pp. 47–78. [15] R.C. Guzman, R.C. Osborn, S.M. Swanson, R. Sakthivel, S.I. Hwang, S. Miyamoto, S. Nandi, Incidence of c-Ki-ras activation in N-methyl-N-nitrosourea-induced mammary carcinomas in pituitary-isografted mice, Cancer Res. 52 (1992) 5732–5737. [16] S.-I. Hwang, R.C. Guzman, S.M. Swanson, S. Nandi, Hormone dependent and independent mammary tumor development from N-methyl-N-nitrosourea-treated rat mammary epithelial cell xenografts in the nude mouse: multiple pathways and H-ras activation, Cancer Lett. 101 (1996) 123–134.
93
[17] H.J. Thompson, L.D. Meeker, Induction of mammary gland carcinomas by the subcutaneous injection of l-methyl- lnitrosourea, Cancer Res. 43 (1983) 1628–1629. [18] M. Habs, D. Schma¨hl, Synergistic effects of N-nitroso compounds in experimental long-term carcinogenesis studies, Oncology 37 (1980) 259–265. [19] M. Berger, K. Ku¨nstler, Experiments on the influence of combined application of N-methylnitrosourea and 7,12dimethylbenzanthracene in female Sprague–Dawley rats, Arch. Geschwulstforsch. 53 (Suppl. 5) (1983) 423–427. [20] K. Ogawa, Y. Tokusashi, I. Fukuda, Absence of p53 mutations in methylnitrosourea-induced mammary tumors in rats, Cancer Detect. Prev. 20 (1996) 214–217. [21] S. Sukumar, K. McKenzie, Y. Chen, Animal models for breast cancer, Mutat. Res. 333 (1995) 37–44.