Existence of No Hepatocarcinogenic Effect Levels of 2-amino-3,8-dimethylimidazo[4,5- f ]quinoxaline with or without Coadministration with Ethanol

Abstract

There is increasing evidence of the existence of no effect levels for genotoxic carcinogens. However, only limited information is available regarding dose-response curves for combination effects of multiple carcinogens at low dose. In the present study, 280 male F344 rats were divided into 14 groups to determine the effects of co-administration of various doses of 2-amino-3,8-dimethylimidazo[4,5- f]quinoxaline (MeIQx) and 10% ethanol on the development of glutathione S-transferase placental form (GST-P)-positive foci in the liver. The results provided concrete evidence for the existence of no effect levels for hepatocarcinogenicity of MeIQx either in presence or absence of ethanol and, therefore, for a practical threshold for this genotoxic carcinogen.

Introduction

Dose response assessment to define relationships between doses of agents and the probability of carcinogenic effects is one of the most important components of risk assessment to humans. Despite a lack of definitive experimental evidence, the dose response curve for a genotoxic carcinogen is generally assumed to be linear without a threshold dose below which carcinogenic effects are absent; thus genotoxic carcinogens may pose some risk at any level of exposure. However, there is increasing experimental evidence that thresholds, at least in practical terms, can exist for genotoxic carcinogens. 2-Amino-3,8-dimethylimidazo[4,5-f]quinoxaline (MeIQx), a heterocyclic amine found in the cooked meat and fish, is a potent dietary genotoxic carcinogen (Terada et al., 1986; Johansson and Jagerstad, 1994), which induces hepatocellular carcinoma at high doses in male F344 rats (Kato et al., 1988; Kushida et al., 1994).

However, our recent low-dose studies showed the existence of no effect levels for MeIQx for induction of glutathione S-transferase placental form (GST-P)-positive foci, well-established preneoplastic lesions, in the livers of F344 and BN rats (Fukushima et al., 2002, 2003; Wei et al., 2005). GST-P-positive-foci have been accepted as a useful endpoint marker in assessment of carcinogenic effects of environmentally relevant concentrations of liver carcinogens in rats (Tsuda et al., 2003). Furthermore, in vivo mutagenicity assays revealed that low doses of MeIQx did not increase mutation frequencies, and the dose-response curve for mutation frequency was also non-linear in rat liver (Hoshi et al., 2004). These findings argue against the linear, no-threshold risk assessment model for genotoxic carcinogens and suggest that there is a practical threshold for hepatocarcinogenetic effects of MeIQx. Since there are many carcinogens in our environment, it is important to assess and manage potential risks associated with multiple carcinogen exposure. Simultaneous exposure to multiple carcinogens may cause synergistic effects and it is well known that carcinogenicities of genotoxic carcinogens can be vastly magnified in the presence of tumor promotors. All previous low-dose studies of MeIQx were designed to evaluate the effects of this carcinogen alone. Further studies are now necessary to evaluate effects of co-administration of MeIQx and liver tumor promoters to provide more insight into dose-response relationships.

Epidemiological studies have shown that the chronic intake of high levels of alcoholic beverages is associated with an increased risk of several cancers in humans, including liver cancer (Morgan et al., 2004; Poschl and Seitz, 2004). In experimental animal studies, ethanol per se has not been found to induce liver tumors (Holmberg and Ekstrom, 1995), but rather shows strong promoting activity on chemically induced hepatocarcinogenesis (Porta et al., 1985; Takada et al., 1986; Tanaka et al., 1989). We have previously reported ethanol to significantly enhance the development of GST-P-positive foci after initiation with 10 ppm MeIQx in rat liver (Karim et al., 2003). However, effects of ethanol on hepatocarcinogenic effects of low doses of MeIQx have not been hitherto investigated.

Based on our previous finding of a threshold level, at least a practical one, for the carcinogenicity of MeIQx (Fukushima et al., 2002, 2003; Wei et al., 2005), no effect levels for development of GST-P-positive foci might be expected to exist either in the presence or absence of ethanol. To test this hypothesis, the present study of the carcinogenic effects of various doses of MeIQx in combination with 10% ethanol was conducted. Since oxidative DNA damage and increased cell proliferation have been implicated in the mechanism of MeIQx carcinogenesis (Fukushima et al., 2002, 2003), we also examined the liver DNA level of 8-hydroxydeoxyguanosine (8-OHdG) formation, a useful marker of cellular oxidative stress (Kasai et al., 1989), as well as hepatocyte proliferation, to cast further light on mechanistic aspects of low-dose carcinogenicity of MeIQx in the present model.

Materials and Methods

Animals

Two hundred and eighty male F344 rats, 20 days old, were obtained from Charles River Japan, Inc. (Atsugi, Kanagawa, Japan). The animals were housed in polycarbonate cages (5/cage) in a room with a targeted temperature of 23 ± 2°C, humidity of 55 ± 5%, and a 12 h light/dark cycle and ad libitum access to food and tap water. The level of care provided to the animals met or exceed the basic requirement outlined in the Guide for the Care and Use of Laboratory Animals (NIH publication #86-23, 1996).

Chemicals and Diets

MeIQx (purity, 99.9%) was purchased from the Nard Institute, Nishinomiya, Japan, and ethanol (99.5%) from Wako Pure Industries, Ltd., Osaka. Basal diet (powdered MF) and MeIQx diets were prepared at Oriental Yeast Co., Tokyo, and concentrations of MeIQx in the diets were confirmed by HPLC. The lowest level of MeIQx fed in the diet was 0.001 ppm, equivalent to the daily intake of this carcinogen by humans (Fukushima et al., 2002).

Experimental Procedures

The following protocol was approved by the Institutional Animal Care and Use Committee of the Osaka City University Medical School. At 21 days of age, rats were randomized into 14 groups of 20 animals each. As shown in Table 1, the rats were fed diets containing 0, 0.001, 0.01, 0.1, 1, 10 or 100 ppm of MeIQx. Groups 1–6 were concurrently administered 10% ethanol in the drinking water while groups 7–14 were given tap water without ethanol. Body weight, water and food consumption were measured weekly. All rats were sacrificed under diethyl ether anesthesia after 16 weeks of treatment. At sacrifice, blood was drawn via the abdominal aorta from 5 rats of each group under anesthesia. Serum aspartate transaminase (AST) and alanine transaminase (ALT) levels were measured at FALCO Biosystems, Osaka, Japan. Livers were excised quickly, weighed and 3 slices each from the left lateral, medial, and right lateral lobes were fixed in 10% phosphate-buffered formalin. After fixation, they were processed for paraffin embedding and stained with hematoxylin and eosin for histological examination, and immunohistochemical analysis. The remaining liver tissues were snap-frozen in liquid nitrogen and stored at −80°C for 8-OHdG analysis.

Immunohistochemistry for GST-P-Positive Foci and Proliferating Cell Nuclear Antigen (PCNA)

Rat polyclonal anti-GST-P antibody (Medical & Biological Laboratories Co., Ltd., Nagoya, Japan) and mouse monoclonal anti-PCNA antibody (Dako EPOS, Dako Cytomation, CA, USA) were used for immunohistochemical staining of GST-P and PCNA, respectively, using the avidin-biotin-peroxidase complex (ABC) method. GST-P-positive foci comprising 2 or more cells in cross-section were counted under a light microscope (Karim et al., 2003; Hoshi et al., 2004; Wei et al., 2005). Total areas of livers were measured using a color image processor (IPAP, Sumica Technos, Osaka, Japan), and then the numbers of foci per cm² of liver tissue were calculated. PCNA labeling indices were calculated as positive cells per 100 hepatocytes.

Determination of 8-OHdG Formation in DNA of the Hepatocytes

DNA was extracted from frozen liver tissues from 5 rats of each group using a DNA Extractor WB Kit (Wako Pure Chemical Industries, Kyoto, Japan), and then digested into deoxynucleosides by combined treatment with nuclease P1 (Tamasa Shoyu, Chiba, Japan) and alkaline phosphatase (Sigma, St. Louis, MO). The levels of 8-OHdG were then determined by the HPLC-ECD according to the method of Nagake et al. (1997).

Statistical Analysis

Differences between mean values (mean ± SD) for the MeIQx alone groups and the non-treatment control groups, MeIQx plus ethanol co-administration groups and the ethanol alone groups were analyzed using the Dunnett two-tailed post hoc test (Fukushima et al., 2002, 2004, 2005; Wei et al., 2005). Student’s t-test or Welch’s t-test were applied for identifying differences between MeIQx and the corresponding MeIQx plus ethanol coadministration groups. p < 0.05 were considered to be statistically significant. All statistical analyses were performed using the StatView-J5.0 program (Abacus Concepts, Inc., Berkeley, CA).

Results

General Findings

All of the rats survived to the end of the study in good condition and no treatment-related clinical signs were observed in any group. As shown in Table 1, significant decreases in final mean body weights were observed in the 100 ppm MeIQx group compared to the non-treatment group and the 100 ppm MeIQx plus 10% ethanol compared to the 10% ethanol alone group. There was no ethanol treatment-related body weight decrease. MeIQx alone had no effects on absolute weight of livers evidenced by the finding of no significant difference among groups administered various doses of MeIQx alone, as well as among groups administered MeIQx plus 10% ethanol. Significant increases in relative liver weights observed in the 100 ppm MeIQx group compared to the non-treatment group and the 100 ppm MeIQx plus 10% ethanol compared to the 10% ethanol alone group appear to be due to decreases in the body weight in 100 ppm MeIQx-treated groups. The 10% ethanol alone significantly increased absolute and relative liver weights compared to the non-treatment group. Similarly, significant increases in absolute and relative liver weights were also seen in groups co-administered 10% ethanol with MeIQx when compared to the corresponding MeIQx alone group. There were no significant differences in absolute and relative kidney weight among the groups.

Serum Transaminase Levels

There were no inter-group differences in AST and ALT (data not shown).

Histopathology

There was no evidence of treatment-related histological changes in rat administered various doses of MeIQx and/or 10% ethanol by histological examination with H&E staining. Neither toxic change such as hypertrophy nor prenoplastic lesions such as foci of cellular alteration and focal hepatocellular hyperplasia was observed in any group.

Induction of GST-P-Positive Foci in the Liver

No difference in staining characteristics of GST-P-positive foci observed among groups. Numbers of GST-P-positive foci per unit area of the rat livers are summarized in Table 2. 100 ppm MeIQx caused significant increases in their numbers compared to the non-treatment control group, whereas 10 ppm and lower doses of MeIQx had no effect. Likewise, numbers of GST-P-positive foci were significantly increased in groups co-administered 10% ethanol and MeIQx at doses of 10 and 100 ppm but not 1 ppm or lower, compared with the ethanol alone group. Co-administration of ethanol with MeIQx significantly increased the number of GST-P-positive foci compared to the corresponding MeIQx alone groups.

PCNA Labeling Indices in the Livers

Hepatocyte proliferation was determined immunohistochemically by PCNA labeling. As shown in Figure 1, PCNA labeling indices were significantly increased in the ethanol-treated groups compared to corresponding MeIQx alone groups. Furthermore, a significant increase was noted for the group co-administered ethanol and 10 or 100 ppm MeIQx but not for 1 ppm or lower compared to the ethanol alone group. There were no significant differences in PCNA labeling indices among the MeIQx alone groups. As no toxic change such as liver hypertrophy was observed, the increased liver weights ethanol-treated groups were possibly due to the significant increases in hepatocellular proliferation compared to corresponding MeIQx alone groups.

8-OHdG Formation in the Liver DNA

Data for 8-OHdG formation in liver DNA are shown in Figure 2. 100 ppm MeIQx alone significantly increased the formation of 8-OHdG compared to non-treatment controls, whereas 10 ppm and lower doses of MeIQx had no significant effects. Similarly, among groups co-administered MeIQx and 10% ethanol, no significant increases were observed with 1 ppm and lower doses of MeIQx group compared to the ethanol alone group. 10% ethanol alone had no effect on the level of 8-OHdG compared to the non-treatment control group. However, co-administration of ethanol significantly increased 8-OHdG formation in the 10 and 100 ppm MeIQx groups compared to the corresponding MeIQx alone groups.

Discussion

The data in the present study demonstrated that administration of MeIQx alone at doses of 1 ppm or less had no apparent effect on induction of GST-P-positive foci, a well-established biomarker of rat hepatocarcinogenesis (Ito et al., 1988; Tsuda et al., 2003). These results are consistent with our previous finding of no-observed effect levels for hepatocarcinogenicity of MeIQx in rats (Fukushima et al., 2002; Hoshi et al., 2004; Wei et al., 2005). We also established that the numbers of GST-P-positive foci were significantly increased in groups co-administered ethanol and 10 or 100 ppm MeIQx compared to the ethanol alone group, whereas no significant differences were observed in groups co-administered ethanol and with 1 ppm MeIQx or lower.

Furthermore, significant increases in hepatocyte proliferation and levels of DNA oxidative damage were only noted in the 10 or 100 ppm MeIQx groups compared to the ethanol alone group. Thus, the results indicate that ethanol lacks promoting effects on the hepatocarcinogenicity of low doses of MeIQx (1 ppm or less), and the most plausible explanation for this phenomenon is that MeIQx is not carcinogenic at these doses. Our findings clearly suggest that the dose-response curve of MeIQx carcinogenicity is non-linear, even in the presence of ethanol, and provide further evidence for the existence of a practical threshold.

On the basis of a body of epidemiological studies, consumption of alcoholic beverages has been linked to an increased risk of liver cancer in humans (Morgan et al., 2004; Poschl and Seitz, 2004). However, the exact mechanisms of alcohol carcinogenicity remain unclear. In experimental animal studies, ethanol per se has not been found to induce liver tumors but rather to promote chemically induced hepatocarcinogenesis (Porta et al., 1985; Takada et al., 1986; Tanaka et al., 1989; Holmberg and Ekstrom, 1995; Poschl and Seitz, 2004). We have previously proposed that treatment with ethanol might promote 10 ppm MeIQx-induced hepatocarcinogenesis through elevated cell proliferation and oxidative stress (Karim et al., 2003). In the present study, ethanol alone also significantly increased cell proliferation in all co-administration groups and formation of 8-OHdG in hepatic DNA with high doses of MeIQx. In contrast, there were no differences in the levels of 8-OHdG in hepatic DNA between the ethanol alone and the non-treatment control group, suggesting that ethanol per se does not induce oxidative DNA damage. Ethanol has been shown to affect the metabolism of carcinogens when given concurrently. Therefore, one possible explanation for why co-administration of ethanol only increased the 8-OHdG levels in groups given relatively high doses of MeIQx (1 ppm or higher) is that ethanol affected the metabolism of high but not low doses of MeIQx that may increase the generation of hydroxyl radicals (Maeda et al., 1995), thereby causing more oxidative DNA damage.

In summary, the results presented here confirm the existence of apparent no effect levels for hepatocarcinogenicity of MeIQx in rats even in the presence of the promotor ethanol. Several other genotoxic carcinogens have also been shown to have apparent no effect levels for their carcinogenic effects, including the liver carcinogens diethylnitrosamine (Fukushima et al., 2002, 2005) and 2-acetylaminofluorene, as well as the colon carcinogen 2-amino-1-6-phenolimidazo [4,5-b] pyridine (Fukushima et al., 2004) in our previous studies. It is likely that many processes may result in non-linearity for the dose response for MeIQx carcinogenicity in the low-dose region, including detoxification reactions, cell cycle arrest, DNA repair, apoptosis, and immune surveillance. The present work thus provides further evidence in support of the existence of a practical threshold for genotoxic carcinogens.