Possible Application of Human c-Ha-ras Proto-Oncogene Transgenic Rats in a Medium-Term Bioassay Model for Carcinogens

Abstract

With the aim of developing a medium-term assay for screening of environmental carcinogens, we exposed mammary carcinogen sensitive human c-Ha-ras proto-oncogene transgenic (Hras128) rats to various carcinogens, including compounds that do not normally induce mammary tumors. Seven-week-old Hras128 rats and wild-type littermates received administrations of 3-methylcholanthrene (3-MC), benzo[a]pyrene (B[a]P), anthracene, pyrene, 2-amino-3-methylimidazo[4,5-f]quinoline (IQ), 2-amino-3,8-dimethylimidazo[4,5-f]quinoxaline (MeIQx), 4-(methyl-nitrosamino)-1-(3-pyridyl)-1-butanone (NNK), dimethylarsinic acid (DMA), diethylnitrosamine (DEN) or azoxymethane (AOM) and were sacrificed at week 12 (females) (at week 10 for the 3-MC group) or week 20 (males). Female Hras128 rats receiving NNK, DEN, or DMA showed a significant increase in mammary tumor incidence and/or multiplicity compared to the respective values with olive oil or deionized distilled water (DDW) vehicles. In male Hras128 rats, a significant increase in mammary tumors was also observed in groups administered 3-MC, B[a]P, anthracene, IQ, and NNK. Mutations of transgenes were observed in codons 12 and/or 61 in the induced tumors by PCR-RFLP except in the DEN group in female and in the MeIQx group in male Hras128 rats. Thus various carcinogens, not necessarily limited to those normally targeting the breast, were found to induce mammary carcinomas in Hras128 rats, especially in females, pointing to potential use for medium-term screening.

Introduction

We have generated human c-Ha-ras proto-oncogene transgenic (Hras128) rats that are highly sensitive to mammary carcinogens, rapidly developing carcinomas after exposure to N-methyl-N-nitrosourea (MNU), dimethylbenzo[a]anthracene (DMBA), or PhIP (Asamoto et al., 2000; Tsuda et al., 2001). Furthermore, the Hras128 rats are also highly susceptible to induction of lesions in the esophagus, bladder, skin and tongue (Ota et al., 2000; Asamoto et al., 2002; Park et al., 2004; Suzuki et al., 2006).

The incidence of spontaneous tumors in the mammary gland of Hras128 rats was 52.8% at 40 weeks and slightly increased compared to female Sprague–Dawley wild-type rats (Tsuda et al., 2005). Taking advantage of these characteristics, we have focused on whether our transgenic animals might have advantages for use in short- or medium-term assay systems for screening environmental carcinogens. One problem is that carcinogens generally have specific organotropic actions as initiating agents (Tsuda et al., 1999).

One way to overcome this is to use multi-organ carcinogenesis models (Imaida and Fukushima, 1996; Ito et al., 1988) in which animals are first treated with various carcinogens, initiating carcinogenesis in the major organs and then assaying promotion or other modulating effects. However, the established protocols require upwards of 30 weeks until tumors or preneoplastic lesions are induced. As a single organ model, the Ito approach in the liver has many advantages in terms of cost and duration, at 8 weeks, but requires partial hepatectomy to enhance carcinogenesis (Tsuda et al., 1980; Ito et al., 1989). While transgenic (rasH2) mice bearing a human c-H-ras proto-oncogene have attracted interest for testing purposes (Ando et al., 1992; Yamamoto et al., 1996), the assay takes 26 weeks and cannot be said to be short-term. Our Hras128 rats develop tumors within 8 weeks.

For validation in the present study, a number of known carcinogens were selected. These were genotoxic agents as the results could not be applied directly to nongenotoxic agents. The polycyclic aromatic hydrocarbons 3-methylcholanthrene (3-MC) and benzo[a]pyrene (B[a]P) and their parent nuclear substances anthracene and pyrene are included in exhaust gas and tobacco smoke. 3-MC and B[a]P in particular are known to be causative agents for lung cancer in humans and mammary cancers in rats (Bolasny et al., 1963; Gingell et al., 1981). The heterocyclic amines 2-amino-3-methylimidazo[4,5- f ]quinoline (IQ) and 2-amino-3, 8-dimethylimidazo[4, 5- f ]quinoxaline (MeIQx) are contained in broiled meat and fish (Sugimura, 1985) and 4-(methylnitrosamino)-1-(3-pyridinyl)-1-butanone (NNK) is found in tobacco smoke (Brown et al., 1999). Diethylnitrosamine (DEN) is an N-nitroso compound commonly used in liver cancer experiments (Ito et al., 1989), while azoxymethane (AOM) specifically induces aberrant crypt foci and tumors in the colon (Thorup et al., 1995). Dimethylarsinic acid (DMA) is an arsenic compound present in the environment (Braman and Foreback, 1973), which is known to cause urinary bladder cancers (Wei et al., 1999; Cohen et al., 2001).

In a series of experiments we administered these chemical carcinogens to Hras128 rats and made gross pathological and histopathological assessment of lesion induction. Furthermore, transgene mutations were examined to determine whether the exogenous gene copies were targeted by the carcinogens. The results indicated that the Hras128 rat may indeed have potential for use as a medium-term assay model.

Materials and Methods

Animals and Chemicals

Sprague–Dawley rats (Clea Japan, Inc., Tokyo, Japan) were used for creating the human c-Ha-ras proto-oncogene transgenic rats (Hras128) (Asamoto et al., 2000), with the human c-Ha-ras proto-oncogene established by Sekiya et al. (1985). The animals were kept under constant conditions with a 12-hour light/dark cycle, a room temperature of 22 ± 2°C, and a humidity of 55 ± 10%. They were allowed access to a basal diet (Oriental MF, Oriental Yeast Co., Tokyo, Japan) and tap water. All rats, transgenic and wild-type littermates, were treated the same. 3-MC, B[a]P, pyrene, and AOM were purchased from Sigma Chemical Co., St Louis, USA; IQ and MeIQx from Nard Institute, Osaka, Japan; NNK from Toronto Research Chemicals Inc., Ontario, Canada; DEN from Tokyo Kasei, Co., Tokyo, Japan; and DMA and olive oil from Wako Pure Chemical Industries, Osaka, Japan. Anthracene (purity >99.9%) was provided by Dr. Matsushima of Japan Bioassay Research Center, Hadano, Japan.

The experiments were conducted according to the “Guidelines for Animal Experiments in the National Cancer Center Japan” promulgated by the Committee for Ethics of Animal Experimentation.

Experimental Protocol (Figure 1)

3-MC, B[a]P, anthracene, pyrene, IQ, MeIQx, and NNK were dissolved in olive oil, and DEN, DMA, and AOM in deionized distilled water (DDW). Two hundred mg/kg of 3-MC, B[a]P, anthracene, and pyrene, 80 mg/kg of IQ and MeIQx, and 100 mg/kg of NNK and DMA were administered by gastric intubation to 7-week-old Hras128 rats and their littermates (wild-type) once a week for 3 weeks. One hundred mg/kg of DEN was given once a week for 2 weeks, and 50 mg/kg of AOM once. The control group received 5 ml/kg of olive oil or DDW. Females were sacrificed at week 12, except for the 3-MC treated group (week 10 due to a moribund condition caused by multiple mammary carcinomas), and males at week 20. Numbers, weights, and sizes of all mammary tumors were then recorded.

Histological Study and DNA Isolation

All mammary tumors were removed and, after measurements, were immediately fixed in ice-cold acetone. Tissues were embedded in paraffin and stained with hematoxylineosin, followed by histopathological examination. DNA was extracted using DEXPAT (Takara, Otsu, Japan) from paraffin sections 10 μm in thickness.

Mutation Analysis

Mutation analysis of transgene codons 12 and 61 was performed using the PCR-restriction fragment length polymorphism (RFLP) approach (Asamoto et al., 2002). The primers for codon 12 were hHras1F (5′-GC AGGCCCCTGAGGAGCGAT-3′), and hHras1RN (5′-AGC AGCTGCTGGCACCTGGA-3′), and for codon 61 were hHras2F (5′-AGCCCTGTCCTCCTGCAGGAT-3′), hHras2 R (5′-GGCCAGCCTCACGGGGTTCA-3′), and H61/2A2 (5′-CGCATGGCGCTGTACAGCTC-3′). After 5 minutes at 95°C, thermocycling conditions were: 1 minute at 95°C, 1 minute at 60°C, 3 minutes at 72°C for 35 cycles, with a final extension of 10 minutes at 72°C. The thermal cycler was a Gene Amp PCR System 9600 (Perkin-Elmer Corp. Norwalk, USA), with MSP I (Takara, Otsu, Japan) for codon 12 and AlwN I (New England BioLabs, MA, USA) for codon 61 as restriction enzymes. After confirming mutations in codons 12 and 61 with PCR-RFLP, DNA lengths of 167 bp for codon 12 and 93 bp for codon 61 were extracted from 4% agarose gels (NuSieve GTG agarose, BMA, USA) using a Min Elute Gel Extraction Kit (QIAGEN, USA) and sequenced using Big Dye Terminator v3.1 (Applied Biosystems, Japan) and an ABI PRIZM3100-Avant Genetic Analyzer (Applied Biosystems, Japan).

Statistics

Analysis of the incidences of mammary tumors and their sizes and multiplicities was conducted using the JMP software package (version 3.1)(SAS Institute, Cary, NC). Chi squared tests were conducted for tumor incidence data and the Dunnett’s t-test with ANOVA for tumor size and multiplicity.

Results

Incidences and Multiplicity of Mammary Tumors

Female Rats

(see Table 1) All tumors taken (larger than 3 mm in longer diameter) were adenocarcinomas with obvious invasion of surrounding mammary and stromal tissue. In female Hras128 rats, mammary tumors developed in 7 of 7 rats (100%) given 3-MC, 8/8 (100%) with B[a]P, 4/7 (57.1%) with anthracene, 3/7 (42.9%) with pyrene, 2/10 (20%) with NNK, 7/10 (70%) with IQ, 6/10 (60%) with MeIQx, 3/9 (33.3%) with DEN, 6/9 (66.7%) with AOM, and 1/9 (11.1%) with DMA. There was a significant increase in the tumor incidence in female Hras128 rats in the 3-MC and B[a]P groups at p < 0.001, the IQ group at p < 0.01, and the anthracene, MeIQx, and AOM groups at p <0.05. The pyrene group also exhibited a significantly increased number of tumors in comparison with the olive oil group (p < 0.05). Among the littermate rats (wild-type), single tumors were found in 2 rats of the 3-MC group and 1 rat of the IQ group, but there were no significant differences from the control (olive oil) group. No tumors other than mammary gland were found in Hras 128 rats. No tumors were detected in any other groups of wild-type rats.

Male Rats (Table 2)

The percentages of male rats with mammary tumors for each carcinogen were as follows: 3-MC, 87.5%; B[a]P, 62.5%; anthracene, 42.9%; pyrene, 10%; NNK, 25%; IQ, 16.7%; MeIQx, 8.3%; AOM, 25%; DEN and DMA, 0%. The incidences were significantly increased in the 3-MC, B[a]P (p < 0.001), and anthracene (p < 0.05) groups. The multiplicity was significantly increased in the NNK (p < 0.05) group. Mammary tumor size (7.8 ± 15 mm) was significantly greater in the IQ group than in the olive oil group (p < 0.05). No significant difference from controls was seen in tumor development in the littermate wild rats. In Hras128 rats, zymbal gland tumors occurred in 3 rats, colonic polyps in 3 rats and scrotal squamous cell papillomas in 2 rats with AOM (Figure 3A), a scrotal squamous cell papillomas in 1 rat and a malignant lymphoma in 1 rat with DMA, zymbal gland tumors in 2 rats with NNK, scrotal squamous cell papillomas in 2 rats with DEN, a scrotal squamous cell papilloma in 1 rat with IQ and a back skin squamous cell papilloma in 1 rat with pyrene (Table 3). Sarcomas, composed of spindle shaped tumor cells, were found only in male Hras128 rats at lower incidences. These cells were negative for antibodies for pankeratin, S-100 protein and alfa-smooth muscle actin (Figure 3B). Sarcomas occurred in 2 rats with 3-MC, 1 rat with B[a]p, and 1 rat with MeIQx.

Mutation Analysis of the Transgenes

The tumors mutation results of PCR-RFLP for codons 12 and/or 61 in the Hras128 rats are shown in Tables 4 and 5. Codons 12 and/or 61 in female rats were as follows: 3-MC, 84.8%; B[a]P, 75%; anthracene, 66.7%; NNK, 100%; IQ, 83.3%; and AOM, 100%. Mutations in both codons 12 and 61 were present in 18.2% of the 3-MC group and 28.6% of the B[a]P group (Table 4). Codons12 and/or 61 in male rats were 3-MC, 66.7%; B[a]P, 100%; anthracene, 66.7%; and AOM, 100%. Mutations on both codons 12 and 61 were present in 5.6% of the 3-MC group and 50% of the B[a]P group (Table 5).

Direct Sequencing of Mutated Bands

The results of direct sequencing of DNA are summarized in Table 6. Figures in Table 6 show the numbers of mutation type in mammary tumors in Hras128 rats combined for female with male. In codon 12 there were transversion mutations of GGC to GTC and GGC to TGC (mutation underlined) at rates of 95.3% (61/64) and 4.7% (3/61), respectively. In codon 61 there were transition mutations of CAG to CGG and transversion mutation of CAG to CAT, CAG to AAG, CAG to CTG and CAG to CGT (mutation underlined) at rates of 58.3% (21/36), 33.3% (12/36), 2.8% (1/36), 2.8% (1/36) and 2.8% (1/36), respectively.

Discussion

The present study demonstrated that the mammary tissue of our transgenic rats is sensitive to the carcinogenic actions of chemicals such as IQ, MeIQx, NNK, and AOM, the last two not normally inducing breast tumors (Reddy et al., 1975; Thorup et al., 1995; Masumura et al., 2003). Furthermore, positive results were also obtained with 3-MC and B[a]P, along with their parent compounds, pyrene, rated as Group 3 in the IARC Monograph series (1983), and anthracene. It should be noted that anthracene, which has been generally considered as a noncarcinogen, also gave positive results in a 2-year chronic feeding test (personal communication from Dr. Matsushima of the Japan Bioassay Research Center).

Although the mouse model harboring the same human c-Ha-ras proto-oncogene as in our Hras128 rats has been extensively examined for susceptibility to various carcinogens and has found application as a medium-term assay system with lung tumors as the endpoint lesions, the experimental protocol required in 26 weeks (Mitsumori et al., 1998; Yamamoto et al., 1998). The duration with the current model, 12 weeks for females and 20 weeks for males, has clear advantages in terms of practical application. Indeed, based on our recent observation of development of mammary cancers 15 and 20 days after the administration of MNU (Matsuoka et al., 2003), it may be possible to shorten the experimental period by histopathological detection of early carcinomas in abdominal mammary glands.

Tumors observed in Hras128 rats were mammary carcinomas and squamous cell papillomas in the back and the scrotum skin. Histological types of mammary carcinomas were tubular with a cribriform arrangement, solid tubular or papillary tubular (Figure 2), all of which are similarly found after treatment with N-methyl-N-nitrosourea, and, importantly, resemble those found in humans (Asamoto et al., 2000). Acinar cell type tumors were not observed. Areas of differing morphology were often found mixed within the same mammary tumors. Furthermore, there was no tendency for specific types to be localized in different mammary glands. No treatment related incidence of any specific histological type or localization was observed. Fibrosarcomas, composed of spindle-shaped, irregular-shaped tumor cells, were found only in male Hras128 rats at lower incidence. Metastasis from adenocarcinomas was not found. Although we have conducted histological examinations of all major organs, including the esophagus, forestomach, tongue, and urinary bladder, which were also found to be highly susceptible to chemical carcinogens in Hras128 rats, no tumors were found, possibly due to the relatively shorter duration of the observation period and low doses of carcinogens. It appears that carcinomas induced in Hras128 rats are not as variable as those observed in transgenic mice (Cardiff et al., 2000).

Since high incidences of the transgene mutation are observed in the mammary tumors in this transgenic rat induced by typical mammary carcinogens (Asamoto et al., 2000), it is clearly of interest whether the same situation might exist with regard to various other carcinogens. Our present studies clearly indicated that the transgenes, but not the endogenous rat c-Ha-ras gene, demonstrate mutations at relatively high incidence, suggesting an important role in carcinogenesis. Although the number of tumors used for mutation analysis was low except for the B[a]P and 3-MC cases, the results are highly suggestive that the compounds commonly cause mutation of the transgenes. The c-Ha-ras gene was also observed in mice with the same transgene (Ando et al., 1992).

In our recent studies, such mutations were already evident in endbuds (Hamaguchi et al., 2004), postulated tissue targets of carcinogens (Russo et al., 1979, 1983), before obvious proliferative change occurred. Thus, it is possible that test compounds including nonmammary carcinogens might also cause mutation of the transgenes, a possibility which we are presently exploring. In the present study, most mutations were of transversion type in codon 12, GGC to GTC predominating, irrespective of the chemical carcinogen. Clearly, it is necessary to analyze whether transversion clustering is dependent on the carcinogen administered or the organ in which the tumor appears. Establishment of short-term assay models is essential to reduce the cost and increase the number of compounds that can be tested (Tennant et al., 1995; Tsuda et al., 1999). From our present review, the human c-Ha-ras proto-oncogene transgenic rat is a good candidate for this purpose.

The assay model is advantageous because the endpoint is mammary carcinomas that can be grossly observed. Furthermore, this model can be used for the assay of modifying agents including chemopreventive compounds (Matsuoka et al., 2003) and also nongenotoxic promoting agents (Fukamachi et al., 2004; Tsuda et al., 2005). Given the number of compounds released in our environment, further validation studies using Hras128 rats will be necessary.