Development of a Medium-term Animal Model Using gpt Delta Rats to Evaluate Chemical Carcinogenicity and Genotoxicity.

In this study, the potential for development of an animal model (GPG46) capable of rapidly detecting chemical carcinogenicity and the underlying mechanisms of action were examined in gpt delta rats using a reporter gene assay to detect mutations and a medium-term rat liver bioassay to detect tumor promotion. The tentative protocol for the GPG46 model was developed based on the results of dose-response exposure to diethylnitrosamine (DEN) and treatment with phenobarbital over time following DEN administration. Briefly, gpt delta rats were exposed to various chemicals for 4 weeks, followed by a partial hepatectomy (PH) to collect samples for an in vivo mutation assay. The mutant frequencies (MFs) of the reporter genes were examined as an indication of tumor initiation. A single intraperitoneal (ip) injection of 10 mg/kg DEN was administered to rats 18 h after the PH to initiate hepatocytes. Tumor-promoting activity was evaluated based on the development of glutathione S-transferase placental form (GST-P)-positive foci at week 10. The genotoxic carcinogens 2-acetylaminofluorene (2-AAF), 2-amino-3-methylimidazo [4,5-f] quinolone (IQ) and safrole (SF), the non-genotoxic carcinogens piperonyl butoxide (PBO) and phenytoin (PHE), the non-carcinogen acetaminophen (APAP) and the genotoxic non-hepatocarcinogen aristolochic acid (AA) were tested to validate the GPG46 model. The validation results indicate that the GPG46 model could be a powerful tool in understanding chemical carcinogenesis and provide valuable information regarding human risk hazards.

Introduction

Environmental chemicals, including pharmaceuticals, agrochemicals and food additives, are important in various aspects of daily life. However, these chemicals may pose a risk to humans, and their toxicities have been extensively assessed in animal studies. In particular, carcinogenicity is a key component of safety assessments because the resulting lesions can be irreversible and are often fatal. The current gold standard for assessing the risk of cancer is a lifetime bioassay in rodents, but this method requires over 3 years to complete, including histopathological procedures[1]. It is estimated that only approximately 1500 chemicals have been tested over the past 30 years despite the addition of nearly 4000 new chemicals in the Chemical Abstracts Service (CAS) Registry database every day[2],[3]. Although conventional lifetime bioassays can provide data regarding the potential carcinogenicity and target organs of various chemicals, these assays do not provide any information about the associated mechanisms of action that influence carcinogenesis. The development of bioassays that can rapidly detect chemical carcinogenicity and provide information about the underlying mechanisms of action is currently being pursued.

Thresholds in dose-related chemical carcinogenicity curves depend on the involvement of genotoxic mechanisms[4]. Mutagenicity and carcinogenicity are important factors when determining risk assessments[5]. Although in vitro genotoxic assays, such as the Ames test, the micronucleus test and the chromosomal aberration test, are considered standard tools for investigating chemical mutagenicity, the results of these methods are not necessarily indicative of carcinogenicity[5]. Reporter gene mutation assays are promising genotoxic techniques because in vivo metabolic processes can be evaluated at the target organs[6]. Comprehensive toxicity studies and the measurement of DNA adducts, oxidative stress and enzymatic activities have been demonstrated in animal models using gpt delta rodents[7],[8],[9],[10]. Using the reliable preneoplastic marker glutathione S-transferase placental form (GST-P) foci, medium-term rat liver bioassays have been developed to rapidly detect tumor promoters because the liver is the most common target organ for carcinogenesis[11]. However, the conventional medium-term bioassays do not provide information regarding the involvement of genotoxic mechanisms in carcinogenesis as a result of exposure to test compounds.

In this study, we evaluated the possibility of developing a new animal model designed to rapidly detect chemical carcinogenicity and underlying molecular mechanisms using a reporter gene mutation assay and a medium-term liver bioassay. The conditions were optimized to establish a tentative experimental protocol, and validation of the model was confirmed using several carcinogens.

Materials and Methods

Chemicals

Diethylnitrosamine (DEN) and safrole (SF) were purchased from Tokyo Kasei Kogyo (Tokyo, Japan). Phenobarbital (PhB), 2-acetylaminofluorene (2-AAF), piperonylbutoxide (PBO), and phenytoin (PHE) were obtained from Wako Pure Chemical Industries (Osaka, Japan), and acetaminophen (APAP) was purchased from MP Biomedicals (Irvine, CA, USA). 2-Amino-3-methylimidazo [4,5-f] quinolone (IQ) and aristolochic acid (AA) were obtained from Toronto Research Chemicals (North York, ON, Canada) and Sigma-Aldrich (St. Louis, MO, USA), respectively.

Experimental animals and housing conditions

The protocol was approved by the Animal Care and Utilization Committee of the National Institute of Health Sciences. Five- or nine-week-old specific pathogen-free F344/NSlc rats or five-week-old specific pathogen-free F344/NSlc-Tg (gpt delta) rats carrying approximately five tandem copies of the transgene lambda EG10 per haploid genome were obtained from Japan SLC (Shizuoka, Japan) and acclimated for 1 week prior to testing. The rats were housed in polycarbonate cages (two or three rats per cage) with hardwood chips for bedding in a conventional animal facility. Animals were maintained under controlled temperature (23 ± 2°C), relative humidity (55 ± 5%), air changes (12 times/h), and lighting (12 h light-dark cycle) conditions with free access to a basal diet (CRF-1; Oriental Yeast Co., Ltd, Tokyo, Japan) and tap water. At the end of each experiment, the rats were euthanized by exsanguination via transection of the abdominal aorta under deep anesthesia.

Animal treatments

Experiment I: The effects of a single administration of DEN on the development of GST-P-positive foci were evaluated. A partial hepatectomy (PH) was performed on ten-week-old male F344/NSlc rats (n=5 rats per dose). After 18 h, an intraperitoneal (ip) injection of DEN was administered at doses of 0, 10, 50, and 100 mg/kg. Six weeks after the start of the experiment, the rat livers were fixed in 10% neutral-buffered formalin. The fixed tissues were embedded in paraffin, sectioned and evaluated using immunohistochemistry for the quantitative analysis of GST-P-positive foci.

Experiment II: Changes in the development of GST-P-positive foci over time following administration of PhB after a PH and single dose exposure to DEN were examined. Six-week-old male F344/NSlc rats (n=10 rats per dose) were fed PhB at concentrations of 0 and 500 ppm in their basal diets. This dose was selected based on a previous carcinogenicity test[12]. After 4 weeks, a PH was performed. An ip injection of DEN at a dose of 10 mg/kg was administered 18 h after the PH. The rats continued to feed on a diet containing PhB until they were sacrificed at 10, 12, or 14 weeks after the start of the experiment. The livers were fixed in 10% neutral-buffered formalin, and the tissues were embedded in paraffin, sectioned and evaluated using immunohistochemistry for the quantitative analysis of GST-P-positive foci.

Experiment III: Validation of the animal model was confirmed using genotoxic, non-genotoxic carcinogens and a non-carcinogen. Six-week-old male F344/NSlc-Tg (gpt delta) rats (n=15 per dose) were fed 20 ppm 2-AAF, 12000 ppm PBO or 6000 ppm APAP in their basal diets. A control group was fed the basal diet without chemical supplementation. The 2-AAF dose was selected based on a preliminary study in which no toxic effects were observed in rats treated with 20 ppm (data not shown). The doses of PBO and APAP were selected based on previous carcinogenicity tests[13],[14]. The animal model was further validated using genotoxic and non-genotoxic carcinogens and a genotoxic non-hepatocarcinogen. Six-week-old male F344/NSlc-Tg (gpt delta) rats (n=15 per dose) were fed 20 ppm IQ, 5000 ppm SF or 2400 ppm PHE in their basal diets. The rats treated with AA received 0.3 mg/kg body weight in 1% sodium bicarbonate by gavage once a day. A control group was fed the basal diet without chemical supplementation. The IQ dose was selected based on a preliminary study in which no toxic effects were observed in rats treated with 20 ppm (data not shown). The doses of SF and PHE were selected based on previous carcinogenicity tests[15],[16], and the dose of AA was determined based on a previous report in which the gpt mutant frequencies (MFs) were increased in rats treated with AA for 4 weeks[17]. The carcinogenic properties of the test chemicals are summarized in Table 1. A PH was performed on all rats after 4 weeks, and an ip injection of DEN at a dose of 10 mg/kg was administered 18 h after the PH. The excised liver tissues were perfused with saline to remove residual blood and stored at –80°C for the gpt assay. The rats continued to feed on the basal diets containing the various chemicals. Ten weeks after the start of the experiment, the livers were fixed in 10% neutral-buffered formalin. The fixed tissues were embedded in paraffin, sectioned and evaluated using immunohistochemistry for the quantitative analysis of GST-P-positive foci.

Table 1.

Summary of the carcinogenic properties of the test chemicals used in the validation study

In vivo mutation assays

6-Thioguanine (6-TG) was used according to the method described in Nohmi et al. [18]. Briefly, genomic DNA was extracted from each liver, and the lambda EG10 DNA (48 kb) was rescued in phages by in vitro packaging. For 6-TG selection, the packaged phages were incubated with Escherichia coli YG6020, which expresses Cre recombinase, and converted to plasmids carrying genes encoding gpt and chloramphenicol acetyltransferase. The infected cells were mixed with molten soft agar and poured onto agar plates containing chloramphenicol and 6-TG. To determine the total number of rescued plasmids, the infected cells were poured on plates containing chloramphenicol without 6-TG. The plates were incubated at 37°C for the selection of 6-TG-resistant colonies. Positive colonies were counted on day 3 and collected on day 4. The gpt MFs were calculated by dividing the number of gpt mutants by the number of rescued phages.

Immunohistochemical staining for GST-P

Immunohistochemical staining was performed using polyclonal antibodies against GST-P (1:1000 dilution; Medical & Biological Laboratories Co., Ltd., Nagoya, Japan). The number and area of GST-P-positive foci consisting of 5 or more nucleated hepatocytes in a crosssection were evaluated using an image analyzer (IPAP, Sumika Technoservice, Hyogo, Japan) [19].

Statistics

The number and area of GST-P-positive foci in experiment I were analyzed using ANOVA followed by Dunnett’s multiple comparison test. The number and area of GST-P-positive foci in experiments II, III and IV and the gpt MFs in experiments III and IV were analyzed by assessing the variance for homogeneity using the F-test. The Student’s t-test and Welch’s t-test were used for homogeneous and heterogeneous data, respectively. The gpt MFs in the rats treated with SF in experiment IV were analyzed using the Mann-Whitney U test.

Results

Experiment I

Two of the rats in the control group died due to surgical complications of the PH and were eliminated from further evaluation. Treatment with DEN increased the number and area of GST-P-positive foci in a dose-dependent manner compared with the control group (Table 2), although the differences were not significant in the rats that were treated with 10 mg/kg and 50 mg/kg.

Table 2.

Quantitative analysis of GST-P-positive foci

Experiment II

Two rats from the 14-week control group, one rat from the 10-week PhB group and one rat from the 12-week PhB group died due to surgical complications of the PH and were eliminated from further evaluation. The number and area of GST-P-positive foci were significantly increased in the rats treated with PhB in each experimental time period (Table 2).

Experiment III

Three rats in the control group, one rat in the group treated with 2-AAF, five rats in the group treated with PBO and one rat in the group treated with APAP died due to surgical complications of the PH and were eliminated from further evaluation. Table 3 shows the MFs in the excised livers of gpt delta rats that were treated with 2-AAF, PBO or APAP for 4 weeks. The MFs in the rats treated with 2-AAF were significantly increased compared with the rats in the control group. No significant changes were observed in the rats treated with PBO or APAP. In the gpt mutation spectra, GC:TA and GC:CG transversions and single base pair deletions were significantly increased in the rats treated with 2-AAF (Table 4). The number and area of GST-P-positive foci were significantly increased in livers of the rats treated with 2-AAF or PBO and significantly decreased in the livers of the rats treated with APAP (Table 2).

Table 3.

gpt MFs in livers of F344 gpt delta rats treated with 2-AAF, PBO and APAP

Table 4.

Mutation spectra of gpt mutant colonies in livers of F344 gpt delta rats treated with 2-AAF, PBO and APAP

One rat in the control group, four rats in the group treated with IQ, eight rats in the group treated with SF, three rats in the group treated with PHE and two rats in the group treated with AA died due to surgical complications of the PH and were eliminated from further evaluation. Table 5 shows the MFs in the excised livers of gpt delta rats that were treated with IQ, SF, PHE or AA for 4 weeks. The MFs in the rats treated with IQ, SF and AA were significantly increased compared with the rats in the control group. In the gpt mutation spectra, GC:TA transversions, GC:AT transitions and single base pair deletions were significantly increased in the rats treated with IQ, and AT:TA transversions were significantly increased in the rats treated with AA (Table 6). No significant changes were observed in the rats treated with SF. The number and area of GST-P-positive foci were significantly increased in the livers of the rats treated with IQ, SF and PHE (Table 2).

Table 5.

gpt MFs in livers of F344 gpt delta rats treated with IQ, SF, PHE and AA

Table 6.

Mutation spectra of gpt mutant colonies in livers of F344 gpt delta rats treated with IQ, SF, PHE and AA

Discussion

Chemical carcinogenesis involves multiple gene alterations, which can be divided into initiation and promotion phases. A medium-term rat liver bioassay involving the quantitative analysis of GST-P-positive foci following cell proliferative stimuli via PH was established to detect the tumor promoting activities of various chemicals. Reporter gene mutation assays using transgenic animals have been developed to detect in vivo mutagenicity. Because this assay can be performed under conditions that are similar to the conventional long-term bioassay, the results may represent the tumor initiation phase of chemical carcinogenesis. GST-P-positive foci have been analyzed in gptdelta rats[7],[20],[21]. The GPG46 animal model described in this study can detect the tumor-initiating and tumor-promoting activities of various chemicals by combining the reporter gene mutation assay and the medium-term liver bioassay.

In this animal model, gpt delta rats were exposed to chemicals, and a PH was performed to collect liver samples for an in vivo mutation assay. The rats were subsequently administered a single ip injection of DEN, and the tumor promoting activity of the chemical was evaluated based on the development of GST-P-positive foci. The World Health Organization (WHO) guidelines state that 4 weeks of exposure is sufficient for detecting mutations in the reporter gene[6], which is supported by additional data[8],[22]. Therefore, the period of exposure prior to PH in this study was determined to be 4 weeks. Initial exposure to a potent genotoxic carcinogen is necessary to detect tumor promoting activities over a short period of time. In this model, DEN was selected because correlations between the administration of DEN and the induction of GST-P foci in the rat liver have been extensively reported[23],[24],[25],[26]. However, the dose of DEN should be as low as possible to avoid any effects on the metabolism of the test chemical because DEN has been shown to influence various parameters, including the induction of cytochrome P450 (CYP) and glutathione S-transferase[27],[28]. We took advantage of the rapid induction of cell proliferation following PH because genotoxic compounds can effectively induce gene mutations under conditions of high cell proliferation[4]. Tsuda et al. [29] reported that the initiator should optimally be administered 18 h after PH to effectively enhance initiation. Based on these data, appropriate dosages of DEN were investigated in a dose-response study consisting of single ip injections of DEN 18 h after PH at doses of 10 mg/kg and higher. The optimal dosage of DEN was established as 10 mg/kg based on the quantitative analysis of GST-P-positive foci. PhB, a liver tumor promoter in rodents[30], was used to determine the optimal duration of exposure following a PH in experiment II. The results of this study demonstrate that treatment with PhB at 500 ppm in the diet for 6 weeks is effective in detecting the effects of tumor promotion. The tentative protocol for the GPG46 animal model is shown in Fig. 1.

Fig. 1.

Tentative protocol for the GPG46 model. Six-week-old male F344 gpt delta rats were exposed to various chemicals for 10 weeks. A partial hepatectomy (PH) was performed at week 4, and the rats were administered a single ip injection of 10 mg/kg diethylnitrosamine (DEN) 18 h after PH. The gpt assay, which is an indicator of tumor initiation, was performed using the liver samples excised via PH at week 4. Tumor promoting activities were evaluated based on the development of GST-P-positive foci induced by DEN at week 10.

The animal model was validated in experiment III. 2-AAF, IQ and SF are genotoxic murine liver carcinogens that produce deoxyguanine adducts via metabolic activation and play a key role in liver carcinogenesis[31],[32],[33],[34]. A significant increase in the MFs of the gpt genes in the rats treated with 2-AAF, IQ and SF was shown using the GPG46 model. Spectrum analysis in the gpt mutant colonies revealed that guanine-related mutations and single base pair deletions were induced by 2-AAF and IQ, but not SF, which is in agreement with previous reports[35],[36],[37]. In the conventional medium-term bioassay, 2-AAF, IQ and SF exposure induced a marked increase in the development of GST-P-positive foci[38], implying that these chemicals also exert a strong tumor promoting action. The GPG46 animal model showed that the development of GST-P-positive foci at 10 weeks was markedly increased in the livers of rats treated with these carcinogens. PBO and PHE were reported to act as hepatocarcinogens in F344 rats fed a diet containing 12000 ppm and 2400 ppm for 2 years, respectively[13],[16]. These compounds are classified as non-genotoxic carcinogens based on the results of variousgenotoxicity studies[16],[39]. An increase in the development of GST-P-positive foci was observed in rats treated with PBO or PHE in a conventional medium-term bioassay[38],[40]. Treatment with PBO and PHE at the carcinogenic dose in the GPG46 animal model did not increase the gpt MF, although the development of GST-P-positive foci was significantly increased. APAP was not reported to behepatocarcinogenic in F344 rats fed a diet containing 6000 ppm for 2 years[14]. In the present study, treatment with APAP in the GPG46 model at a dose of 6000 ppm did not increase the gpt MF and inhibited the development of GST-P-positive foci. Ito et al.[38] showed that APAP had an inhibitory effect on the development of GST-P-positive foci in a conventional medium-term bioassay. AA has been reported to be carcinogenic in the kidney and the stomach of rodents[41]. In an in vivo genotoxicity study in Big Blue transgenic rats, AA exposure elevated cII MFs and produced AA-specific deoxyadenine and deoxyguanine adducts in the kidney and the liver[42]. A significant increase in gpt MFs in rats treated with AA was observed in the GPG46 model, and AT:TA transversions were the predominant mutation in the mutation spectra analysis, which is similar to a previous report[42]. AA did not have an enhancing effect on the development of GST-P-positive foci, which may reflect the fact that AA exerts initiation activity, but not carcinogenicity, in the liver.

Overall, the validation results show the possibility of developing a new animal model using gpt delta rats. However, a possible limitation of the tentative protocol is that the test chemicals are co-administered simultaneously with DEN. Although there did not appear to be any mutual effects between DEN and the test chemicals, this treatment regimen may modify the detoxification or metabolic activation of DEN. Several isoforms of CYP have been reported to participate in the metabolic activation of DEN, with CYP2E1 in particular playing an essential role[43]. Because many liver tumor promoters in rodents can induce several types of CYPs and/or modify the expression of phase II enzymes, we are working toward improving the timing of the regimen to avoid the possibility of mutual effects. Validation studies of the revised protocol based on changes in the timing of chemical administration are currently in progress.

In conclusion, the potential development of a GPG46 medium-term animal model to evaluate the tumor-initiating and tumor-promoting activities of various chemicals in a single study was demonstrated. In this assay, additional analyses, such as quantification of DNA modifications, the activities of metabolic enzymes and the mRNA levels of tumor-associated genes, are valuable for understanding the modes of action of various test chemicals.