A 26-Week Carcinogenicity Study of 2-Amino-3-Methylimidazo[4,5- f ]Quinoline in rasH2 Mice

Abstract

To evaluate the carcinogenic susceptibility of rasH2 mice to 2-amino-3-methylimidazo[4,5- f ]quinoline (IQ), 7-week-old rasH2 mice and their wild-type littermates (non-Tg mice) of both the sexes were fed a diet containing 0 or 300 ppm IQ for 26 weeks. Microscopical examinations revealed that the proliferative lesions of the forestomach, including squamous cell hyperplasias, papillomas, and carcinomas, were frequently encountered in male and female rasH2 mice fed with IQ. In non-Tg mice, no significant differences in the incidence of forestomach lesions were observed between the 0 ppm and 300 ppm groups. Histopathological changes such as periportal hepatocellular hypertrophy and oval cell proliferation in the liver were more apparent in female rasH2 and non-Tg mice than in males, and the incidence of hepatocellular altered foci significantly increased in female rasH2 mice in the 300 ppm group as compared to that in the 0 ppm group. These results suggest that the carcinogenic potential of IQ can be detected in rasH2 mice by a 26-week, short-term carcinogenicity test.

Introduction

The rasH2 mouse is a hemizygous transgenic mouse carrying 3 copies of the prototype human c-Ha-ras gene with its own promoter integrated into the genome in a tandem array (Saito et al., 1990; Suemizu et al., 2002). Based on previous studies, it is well recognized that rasH2 mice are very susceptible to genotoxic carcinogens; therefore, they are generally accepted as an alternative animal model and used in place of long-term carcinogenicity tests (Yamamoto et al., 1996; Usui et al., 2001; Morton et al., 2002). On the other hand, since rasH2 mice cannot always detect all types of carcinogens, it is necessary to validate the carcinogenic susceptibility of these mice to various chemicals for appropriate evaluation of carcinogenic substances.

2-Amino-3-methylimidazo[4,5- f ]quinoline (IQ) is one of the genotoxic and carcinogenic heterocyclic amines (HCAs) produced during the cooking of meat and fish (Ohgaki et al., 1984; Sugimura et al., 2004). It has been reported that long-term treatment with 300 ppm IQ induced tumors in the liver, lung, and forestomach of CDF1 mice (Ohgaki et al., 1984) and in the liver, small and large intestines, Zymbal gland, clitoral gland, and skin of F344 rats (Takayama et al., 1984). In addition, IQ has been demonstrated to possess carcinogenic potential in monkeys (Adamson et al., 1990). Although data with a direct relevance to the carcinogenicity of IQ to humans were unavailable, IQ is considered to be probably carcinogenic to humans (IARC, 1993). Recently, HCAs have been tested in several transgenic and knockout mice models, and it was expected that application of such models would provide useful information on HCAs-induced mutagenesis and car- cinogenesis and also on chemopreventive agents (Dashwood, 2003). However, these newly developed animal models were not always susceptible to IQ (Sorensen et al., 1996; Dashwood, 2003; Ogawa et al., 2005). In the present study, we performed a short-term carcinogenicity study of IQ in order to evaluate the carcinogenic susceptibility of rasH2 mice to this chemical compound.

Materials and Methods

Animals and Chemicals

Male and female rasH2 mice and their wild-type litter-mates (non-Tg mice) were purchased from CLEA Japan Inc. (Tokyo, Japan). They were housed in plastic cages (each cage contained 5 animals) with absorbent paper chip bedding in an animal room maintained under standard conditions (room temperature, 22 ± 2³C; relative humidity, 55% ± 5%; and light/dark cycle, 12 h). Food and water were made available ad libitum throughout the experimental period. The animals were acclimatized for 1 week prior to the beginning of experiments; at this time, the animals were 7 weeks of age. The experiment was performed in accordance with the guidelines for animal experimentation of the Faculty of Agriculture, Tokyo University of Agriculture and Technology.

IQ was purchased from Nard Institute (Osaka, Japan). The presence of contaminants in IQ measured by elementary analysis, infrared and mass spectra and HPLC was less than 0.4%. The preparation of IQ-containing diet was carried out in accordance with a previous report (Ohgaki et al., 1984), and the diet was stored at 4°C in a refrigerator prior to use. We used CA-1 (CLEA Japan, Inc.) as the basal diet.

Experimental Design

Twenty-five rasH2 mice and 15 non-Tg mice of both the sexes were fed a diet containing 300 ppm IQ, while 15 rasH2 mice and 10 non-Tg mice of both the sexes were fed basal diet for 26 weeks. As the standard protocol of short-term carcinogenicity test using rasH2 mice, 15 animals/sex/group were used in validation studies by ILSI/HESI (Robinson and MacDonald, 2001), and 20–25 animals/sex/group are recommended from the viewpoint of statistical power currently (Morton et al., 2002). To save the number of animals used in this study, we used the minimum number of animals based on this information. Body weight and food consumption were measured once a week during the experimental period. The mean actual intake of IQ was calculated from the mean body weight and mean food consumption. For complete necropsy and histopathological examinations, 3 male and female rasH2 mice in the 300 ppm group were sacrificed at week 13 as a preliminary study, and all surviving animals were sacrificed at the end of the 26-week examination period. All major organs and tissues from each animal were fixed with 10% neutral buffered formalin. The bones were decalcified in Plank Rychlo solution, which contained 7 g aluminum chloride, 8.5 ml formic acid, and 5 ml hydrochloric acid in 100 ml distilled water, prior to embedding. The tissues were processed routinely, embedded in paraffin, sectioned, and stained with hematoxylin and eosin (H&E). Eosinophilic changes in the epithelium of the nasal cavity were classified based on the severity. The evaluation of the severity of eosinophilic changes in the olfactory and respiratory epithelia was performed at the dorsal meatus and the nasal septum, respectively, in the section made at the level of incisive papilla.

Statistical Analysis

Intergroup differences in body weight, food consumption, and IQ intake were analyzed by the Student’s t-test. The incidence of major lesions, which include the proliferative, degenerative, and atypical lesions related to the IQ administration, was analyzed by Fisher’s exact probability test. The Mann–Whitney U-test was used to evaluate the severity of eosinophilic changes in the olfactory and respiratory epithelia.

Results

The data of survival rate, initial and final body weight, average food consumption, and mean actual intake of IQ are summarized in Table 1. One male rasH2 mouse and 4 female rasH2 mice fed with a diet containing 300 ppm IQ died during the experimental period. A male rasH2 mouse died in week 26, and the pathological examination revealed a subcutaneous hemangiosarcoma of the lower part of the abdomen, inflammation of the prostate, transitional cell hyperplasia of the bladder, and squamous cell papilloma of the forestomach. No obvious abnormalities were found in female rasH2 mice that died in week 16–23, and the cause of their death could not be determined. Body weight gain in all rasH2 and non-Tg mice in the 300 ppm group decreased as compared to the 0 ppm group from the early stage of the experiment; the final body weight of both mice in the 300 ppm group was significantly lower than those in the 0 ppm group. The food consumption of both rasH2 mice and non-Tg mice in the 300 ppm group slightly decreased when compared with those in the 0 ppm group. The mean actual intake of IQ of the rasH2 mice and non-Tg mice were 37.2 ± 4.3 mg/kg/day and 36.4 ± 8.8 mg/kg/day, respectively, in males, and 46.5 ± 9.4 mg/kg/day and 44.2 ± 9.6 mg/kg/day, respectively, in females.

Macroscopic observations at the end of the 26-week treatment period revealed white nodules/masses with a diameter of 1–3 mm on the forestomach mucosa in approximately half of the rasH2 mice in the 300 ppm group. Histopathologically, these nodules/masses were diagnosed as squamous cell hyperplasias, papillomas, and carcinomas (Figure 1). In our diagnostic criteria, squamous cell hyperplasia is characterized by focal or diffuse thickening of epithelial layer. Squamous cell papillomas consist of a stalk protruding into the lumen with multiple finger-like projections arising from the stalk, and squamous cell carcinomas show invasive proliferation into the submucosa. Squamous cell papillomas were frequently detected in male mice, whereas squamous cell hyperplasias were often detected in female mice (Table 2). The incidence of total proliferative lesions in the forestomach, including hyperplasias, papillomas, and carcinomas, in all rasH2 mice in the 300 ppm group was significantly higher than that in the 0 ppm group. In non-Tg mice, no significant differences in the incidences of forestomach lesions were observed between the 0 ppm and 300 ppm groups.

The results of histopathological examinations of the liver, spleen, lung, and skeletal muscles are shown in Table 3. The histopathological changes in the liver were more apparent in female mice than in male mice. Hepatocyte hypertrophy of periportal regions was observed only in female rasH2 mice and non-Tg mice in relation to IQ treatment and was accompanied by oval cell proliferation in many cases (Figure 2A–D, Table 3). Periportal hepatocyte hypertrophy was also seen in the liver of female rasH2 mice at week 13. Hepatocellular altered foci were detected in male and female mice; most of these foci were classified as the basophilic cell type. The incidence of altered foci in female rasH2 mice in the 300 ppm group was significantly increased when compared with that in the 0 ppm group. In the spleen and lung, which are the tumor target organs in short-term carcinogenicity studies of other carcinogens in rasH2 mice, no significant differences in the incidence of lesions relevant to IQ treatment were observed between rasH2 and non-Tg mice. Additionally, IQ-related histopathological changes were not detected in other organs. Skeletal myopathy was detected in almost all the rasH2 mice with or without IQ treatment, as reported by Tsuchiya et al. (2002).

The severity of eosinophilic changes in the epithelium of the nasal cavity varied between rasH2 mice and non-Tg mice. Based on the severity, the eosinophilic changes in the olfactory and respiratory epithelia were classified into slight (+), moderate (++), and severe (+++) (Figure 2E–H, Table 4). In the olfactory epithelium, the severity of eosinophilic changes significantly increased in all rasH2 mice when compared with that in non-Tg mice, although the incidence of this lesion in both mice was similar. In the respiratory epithelium, both the incidence and severity of eosinophilic changes in rasH2 mice were higher than those in non-Tg mice. The severity of these changes in both rasH2 and non-Tg mice did not correlate with IQ treatment.

Discussion

It has been reported that in CDF1 mice, tumors in the liver, forestomach, and lung could be induced by feeding 300 ppm IQ for 96 weeks (Ohgaki et al., 1984). In the present study, IQ induced proliferative lesions in the forestomach of male and female rasH2 mice within 26 weeks of treatment; this suggests that rasH2 mice show carcinogenic susceptibility to IQ. Forestomach papillomas and carcinomas have been reported as one of the spontaneous tumors in rasH2 mice, although the incidences of these tumors were generally low (Mitsumori et al., 1998). In addition, the forestomach is extremely sensitive to various carcinogens, and numerous reports indicate that forestomach papillomas and/or squamous cell carcinomas were frequently induced by treatment with various carcinogens (Yamamoto et al., 1996; Usui et al., 2001). The E mu-pim-1 transgenic mice, which overexpresses the pim-1 oncogene in the lymphoid tissues, were reported to show high susceptibility to the induction of lymphoma by 2-amino-1-methyl-6-phenylimidazo[4,5-b]pyridine (PhIP), which induced lymphoma in the long-term carcinogenicity study in mice (Sorensen et al., 1996). However, these mice were not susceptible to the IQ treatment that the lymphoid tissue is not a carcinogenic target organ (Sorensen et al., 1996). Thus, the induction of forestomach tumors in this study may be attributed to the finding that the target organs of IQ were consistent with the organs that are sensitive to carcinogens in rasH2 mice. The incidence of the forestomach tumors induced by IQ in male CDF1 mice was higher than that in female mice (Ohgaki et al., 1984), and in this study, the carcinogenic susceptibility of male rasH2 mice was higher than that of female mice; this suggests that there is a sex difference in the carcinogenicity of IQ.

IQ is metabolically activated and then forms bulky adducts in DNA, which is considered as a major factor of carcino-genesis by IQ (Schut and Snyderwine, 1999; Sugimira et al., 2004). DNA adducts produced by IQ are repaired mainly via the nucleotide excision repair mechanism, in contrast to DNA damage by alkylating agents, which are repaired mostly via the base excision repair (BER) and DNA strand break repair mechanisms (Sobol et al., 1999; Wu et al., 2003). Poly(ADP-ribose) polymerase-1 knockout mice, which show decreased BER efficiency, demonstrated no increase in the incidence of tumors induced by IQ; however, it showed an elevated susceptibility to carcinogenesis induced by alkylating agents (Ogawa et al., 2005). Since rasH2 mice showed increased susceptibility to IQ as well as alkylating agents such as N-bis(2-hydroxypropyl)nitrosamine (Okamura et al., 2004a), it

Currently, mutation and overexpression of the transgene are considered to be the most probable mechanisms of enhanced carcinogenesis in rasH2 mice (Tamaoki, 2001). Point mutations in ras genes have been detected in tumors induced by IQ in rats and mice (Herzog et al., 1993; Nagao et al., 1997). Therefore, mutation of the transgene might occur in the forestomach tumors in rasH2 mice after treatment with IQ. On the other hand, overexpression of the transgene was detected in the forestomach and lung tumors in rasH2 mice, irrespective of the mutation frequency of the transgene (Tamaoki, 2001; Toyosawa et al., 2003; Okamura et al., 2004b). Accordingly, the overexpression of the transgene might be involved in the tumorigenesis by IQ.

Histopathological changes such as periportal hepatocellular hypertrophy and oval cell proliferation in the liver were more apparent in female mice than in male mice, and the hepatocellular altered foci were increased in female rasH2 mice fed with IQ. In CDF1 mice, the sensitivity of females to the induction of liver tumor was approximately twice that of males (Ohgaki et al., 1984). Because it is well known that oval cell proliferation and basophilic foci are induced by hepato-carcinogens (Goldfarb et al., 1983; He et al., 1994), female rasH2 mice are considered to be more sensitive to hepatocar-cinogenesis induced by IQ than male rasH2 mice, as in the case of CDF1 mice. IQ is metabolized by cytochrome P450 (CYP) 1A2 mainly in the liver and induces its activating enzyme (Nerurkar et al., 1993). Thus, there is a possibility that periportal hepatocyte hypertrophy is probably caused by the induction of enzyme. However, since the localization pattern of CYP 1A2 induction is not always specific to the periportal region (Oinonen and Lindros, 1998), further studies are required to clarify the causes of hepatocyte hypertrophy.

In the present study, we found that the severity of eosinophilic changes in the olfactory and respiratory epithelia varied between rasH2 mice and non-Tg mice; the severity of eosinophilic changes in the olfactory epithelium significantly increased in rasH2 mice as compared to that in non-Tg mice, and both the incidence and severity of eosinophilic changes in the respiratory epithelium in rasH2 mice were higher than those in non-Tg mice. It is known that these eosinophilic changes occur spontaneously in aged rats and mice, although the etiology of these changes is uncertain (Nagano et al., 1997). In rasH2 mice, skeletal myopathy, which is characterized by variation in muscle fiber size, centrally placed nuclei, regenerating fibers, and interstitial fibrosis, has been reported to be a spontaneous histopathological lesion with no difference with respect to gender (Tsuchiya et al., 2002). The myopathic changes were observed with aging and were detected in almost 100% of 34-week-old rasH2 mice (Tsuchiya et al., 2002; Tsuchiya et al., 2005). Although the integration of the c-Ha-ras gene is thought to be crucial, the underlying mechanism of its pathogenesis remains to be elucidated. Similarly, the severe eosinophilic changes in the nasal epithelial cells are considered to be one of the spontaneous histopathological lesions in rasH2 mice, and further studies are necessary to clarify the rationale behind a high severity of such lesions in rasH2 mice as compared to the non-Tg mice.