Potent Hepatocarcinogen In Rats Research Articles

Induction of microRNAome deregulation in rat liver by long-term tamoxifen exposure

Micro RNAs (miRNAs) are small non-coding RNA molecules that function as negative regulators of gene expression. They play a crucial role in the regulation of genes involved in the control of development, cell proliferation, apoptosis, and stress response. Although miRNA levels are substantially altered in tumors, their role in carcinogenesis, specifically at the early pre-cancerous stages, has not been established. Here we report that exposure of Fisher 344 rats to tamoxifen, a potent hepatocarcinogen in rats, for 24 weeks leads to substantial changes in the expression of miRNA genes in the liver. We noted a significant up-regulation of known oncogenic miRNAs, such as the 17-92 cluster, miR-106a, and miR-34. Furthermore, we confirmed the corresponding changes in the expression of proteins targeted by these miRNAs, which include important cell cycle regulators, chromatin modifiers, and expression regulators implicated in carcinogenesis. All these miRNA changes correspond to previously reported alterations in full-fledged tumors, including hepatocellular carcinomas. Thus, our findings indicate that miRNA changes occur prior to tumor formation and are not merely a consequence of a transformed state.

Effect of long-term tamoxifen exposure on genotoxic and epigenetic changes in rat liver: implications for tamoxifen-induced hepatocarcinogenesis

Tamoxifen is a non-steroidal anti-estrogen used for the treatment of breast cancer and, more recently, as a chemopreventive agent in healthy women at high risk of developing breast cancer. On the other hand, tamoxifen is a potent hepatocarcinogen in rats, with both tumor-initiating and tumor-promoting properties. There is substantial evidence that hepatic tumors in rats are initiated as a result of formation of tamoxifen-DNA adducts; however, events subsequent to DNA adduct formation are not clear. Recently, it has been demonstrated that genotoxic carcinogens, in addition to exerting genotoxic effects, often cause epigenetic alterations. In the current study, we investigated whether or not the mechanism of tamoxifen-induced hepatocarcinogenesis includes both genotoxic and epigenetic components. Female Fisher 344 rats were fed a 420 p.p.m. tamoxifen diet for 6, 12, 18 or 24 weeks. Hepatic tamoxifen-DNA adduct levels, as assessed by high-performance liquid chromatography and electrospray tandem mass spectrometry, were 580 adducts/10(8) nt at 6 weeks, and increased to approximately 1700 adducts/10(8) nt by 18 weeks. Global liver DNA hypomethylation, as determined by an HpaII-based cytosine extension assay, was increased at all time points, with the maximum increase (approximately 200%) occurring at 6 weeks. Protein expressions of maintenance (DNMT1) DNA methyltransferase and de novo DNA methyltransferases DNMT3a and DNMT3b were decreased at all time points. Likewise, trimethylation of histone H4 lysine 20 was significantly decreased at all time points. In contrast, non-target tissues (i.e. mammary gland, pancreas and spleen) did not show any changes in global DNA methylation or DNA methyltransferase activity. These data indicate the importance of genotoxic and epigenetic alterations in the etiology of tamoxifen-induced hepatocarcinogenesis.

Species specific hepatocarcinogen inhibition of the glucocorticoid induction of tyrosine aminotransferase gene in mouse and rat liver.

3'-methyl-4-dimethylaminoazobenzene (3'-MeDAB) is a potent hepatocarcinogen in rats and a weak carcinogen in mice, whereas o-aminoazotoluene (OAT) is a potent hepatocarcinogen in mice but weak hepatocarcinogen in rats. They significantly suppress glucocorticoid induction of tyrosine aminotransferase (TAT) in the liver of sensitive animals and have minor effect on the induction of this enzyme in the liver of resistant animals (3'-MeDAB-treated mice and OAT-treated rats). The inhibitory effect of these carcinogens is realized at the level of gene transcription (decreased accumulation of TAT mRNA). This effect is mediated via reduction of DNA-binding activity of transcription factor HNF3 (without decrease of its content) without any involvement of the glucocorticoid receptor. It was shown that carcinogens influence DNA-binding activity of HNF3 via an unknown nuclear factor.

4-Hydroxytamoxifen sulfation metabolism.

Tamoxifen (TAM) is an important chemotherapeutic agent for the treatment of breast cancer. It has also been shown to decrease breast cancer incidence in healthy women at high risk for the disease. The increased risk of endometrial cancer in women has raised concerns in the use of the drug. Tamoxifen has also been shown to be a potent hepatocarcinogen in rats. The oxidative metabolites of TAM include alpha-hydroxytamoxifen (alpha-OH-TAM) and 4-hydroxytamoxifen (4-OH-TAM). The studies on the sulfation of these metabolites are very limited. It has been reported that alpha-OH-TAM is a substrate for rat hydroxysteroid sulfotransferase a (STa). Our studies on the sulfation of 4-OH-TAM demonstrated that 4-hydroxytamoxifen can be sulfated by human liver and human intestinal cytosols. Human phenol-sulfating sulfotransferase and human estrogen sulfotransferase are the major enzymes for the sulfation of 4-OH-TAM. Human dopamine-sulfating sulfotransferase also has sulfation activity for 4-OH-TAM. In contrast, rat liver and intestine cytosols have no detectable sulfation activity for 4-OH-TAM. The results suggest that the alpha-OH-TAM sulfation pathway leads to bioactivation of TAM, and the 4-OH-TAM sulfation pathway leads to detoxification of TAM. This agrees with the fact that TAM is more toxic for rats than for human beings.

Localization of tamoxifen-induced DNA adducts in rat liver

Tamoxifen is a non-steroidal anti-estrogen and is one of the most widely used and successful agents in the treatment of breast cancer. Tamoxifen competes with endogenous estrogens for estrogen receptors in mammary tissue, and thereby inhibits transcription of estrogen-responsive genes. The success of tamoxifen in the adjuvant therapy of breast cancer has resulted in its current widespread use as a prophylactic in healthy women who are at high risk of developing breast cancer. However, the prophylactic use of tamoxifen is controversial because of mixed results from clinical trials and, in particular, the increase in the incidence of endometrial cancer associated with tamoxifen therapy.The reason for this increase in endometrial cancer in humans is a matter of considerable debate. One theory is that tamoxifen is a genotoxic carcinogen – that is, it is metabolized to electrophilic, DNA-reactive intermediates that damage the genetic material and form DNA ‘adducts’. In support of this hypothesis, tamoxifen is a potent hepatocarcinogen in rats, and adducts are indeed found in the liver DNA of rats treated with tamoxifen. In humans, however, there are inconsistencies in the literature concerning the presence of DNA adducts in endometrial tissue from women undergoing tamoxifen therapy. It is therefore unclear whether tamoxifen is a genotoxic carcinogen in humans, or whether it exerts its carcinogenic effects by another mechanism.A recent report by Divi et al. 1xImmunohistochemical localization and semi-quantitation of hepatic tamoxifen-DNA adducts in rats exposed orally to tamoxifen. Divi, R.L. et al. Carcinogenesis. 2001; 10: 1693–1699CrossrefSee all References1 might help shed light on this contentious issue. The authors have developed a sensitive immunohistochemical assay to detect tamoxifen–DNA adducts in rat liver tissue sections. An advantage of this technique over conventional assays for adduct detection, in which DNA must first be isolated from tissue or cells, is that adducts can be localized to particular regions of the tissue in situ. In a beautiful series of immunohistochemical staining images, Divi and colleagues demonstrate nuclear localization of DNA adducts in liver sections from rats treated orally with tamoxifen at various doses. In subsequent experiments, the authors located regions within the liver tissue that expressed an altered form of glutathione-S-transferase, a phase II detoxification enzyme that reduces adduct formation. It was found that, in these regions, levels of tamoxifen–DNA adducts were much lower. Most striking of all is the presence of discrete foci of cells, positive for tamoxifen–DNA adducts, surrounded by areas of tissue that are relatively adduct free; this demonstrates localization of adducts at single-cell resolution within the tissue.The authors raise the possibility that this technique might be suitable for the analysis of human endometrial samples from women undergoing tamoxifen therapy, and might reveal specific regions or cells within the endometrial tissue that contain significant levels of tamoxifen–DNA adducts. This technique might well be a more suitable and sensitive method of analysis than other adduct-detection methods in which DNA must be isolated, and in which the presence of adducts in certain regions of the tissue are masked by the vast majority of surrounding tissue that contains no DNA adducts. If this is indeed the case, analysis of the localization of tamoxifen–DNA adducts in human tissues might help us understand the mechanism of tamoxifen-induced carcinogenesis in humans.

Mutagenicity of the potent rat hepatocarcinogen 6BT to the liver of transgenic (lacI) rats: consideration of a reduced mutation assay protocol.

6-(p-dimethylaminophenylazo)benzothiazole (6BT) is an unusually potent rat hepatocarcinogen, producing large malignant liver tumours after only 2-3 months of dietary administration in a riboflavin-deficient diet. This azocarcinogen has been evaluated in a Big Blue F344 transgenic rat (lacI) gene mutation assay. In a reproduction of the early stages of the carcinogenesis bioassay of this agent, rats were maintained on a riboflavin-deficient diet and were given 10 consecutive daily doses of 6BT (10 mg/kg) by oral gavage. The animals were killed and the livers examined 11 days after the final dose. The livers of 6BT-treated rats showed evidence of hepatocellular hypertrophy in centrolobular areas, with some indication of an increased incidence of mitotic figures. An approximately 10-fold increase in the mutation frequency of DNA isolated from an aliquot of the combined liver homogenates of 6BT-treated rats was observed over that obtained from an equivalent aliquot from control animals. Examination of DNA samples isolated from the livers of individual animals confirmed that 6BT was mutagenic in Big Blue rat livers. These data extend the sensitivity of this transgenic assay to include azo hepatocarcinogens. The determination of mutation frequencies using pooled tissue samples represented a major resource-saving adaptation of the assay protocol in the present study; the general advantages and disadvantages of this practice are discussed.

Clastogenicity of methapyrilene hydrochloride in cultured Chinese hamster cells

The antihistaminic agent methapyrilene hydrochloride (MPH, CAS No. 135-23-9) has been reported as a potent rat hepatocarcinogen. However, neither the action mechanism nor mutagenicity of MPH have been clearly elucidated yet. Chromosomal aberration (CA) tests were performed with MPH in CHO-K1-BH4 and CHL cells without a metabolic activation system. Isochromatid break-like aberrations were observed in metaphases of CHO-K1-BH4 cells treated with MPH. MPH induced CA more effectively at higher pH in equitoxicity base. In contrast to CHO-K1-BH4 cells, typical chromatid type exchanges were observed in CHL cells treated with MPH. These results clearly support the results of other researchers that MPH induce chromosomal aberration and imply that MPH induces cancer in animal via other mechanisms than mere DNA damage or gene mutation which had been nearly proven to be negative by many researchers. It was suggested that pH variation might increase the sensitivity of CA tests for certain category of chemicals with dissociation potential.

Strong intensification of mouse hepatic tamoxifen DNA adduct formation by pretreatment with the sulfotransferase inhibitor and ubiquitous environmental pollutant pentachlorophenol

Although negative in assays for mutagenicity, the clinically important antiestrogen tamoxifen induces hepatic DNA adduct formation in mice, rats and hamsters, as indicated by 32P-postlabeling, and is a potent hepatocarcinogen in rats. Both phenolic and alcoholic metabolites of tamoxifen have been reported. As these metabolites are potential candidates for sulfate conjugation, we examined whether the sulfotransferase inhibitor pentachlorophenol, a ubiquitous environmental contaminant, modulates hepatic tamoxifen adduct formation in vivo. Female ICR mice were given tamoxifen (45 mg/kg) daily per os for up to 4 days, with and without i.p. pretreatment with pentachlorophenol (20 mg/kg) 1 h before dosing with tamoxifen. At days 1, 2 and 4, liver DNA was analyzed 5 h after tamoxifen administration by a modified monophosphate version of the 32P-postlabeling assay. At day 4, pentachlorophenol pretreatment led to a large increase (13- to 17-fold) of the levels of four tamoxifen adduct fractions, while two adducts appeared unaffected, resulting in an approximately 7-fold enhancement of overall adduct formation. Significant pentachlorophenol related increases were also observed at day 1 and day 2. The mechanism of this effect has not yet been determined, but may involve the inhibition of sulfation of a tamoxifen metabolite(s) involved in the detoxication of the drug to nonelectrophilic derivatives. It was also apparent that there are two pathways of metabolic activation of tamoxifen, one being sensitive and the other resistant to pentachlorophenol.

The evaluation of methapyrilene for bacterial mutation with metabolic activation by Aroclor-induced, methapyrilene-induced and noninduced rat-liver S9

The antihistamine methapyrilene (MP) has been shown to be a potent hepatocarcinogen in rats. However, it has demonstrated little genotoxic activity in a wide variety of short-term tests. In this study, Fischer 344 rats were fed a carcinogenic dose of 0.1% methapyrilene in the diet for 10 weeks prior to sacrifice. S9 was prepared from the livers of the control, MP-treated and Aroclor-induced Fischer 344 rats. Each type of S9 was analyzed for mixed function oxidase activity, cytochrome P-450, and protein content. MP was then evaluated for mutagenicity in 6 strains of S. typhimurium (TA1535, TA1537, TA98, TA100, TA2638 and TA104) and one strain of E. coli (WP2uvrA −) using the standard plate-incorporation assay. MP was not mutagenic in any of the 7 bacterial strains when tested at concentrations ⩽ 10 mg/ml in the presence of each type of S9. However, in the absence of metabolic activation, an approximate 2-fold increase in revertants was noted with strain TA1535. The data from this study show that MP was not converted to a mutagenic metabolite by any of the three S9 types examined. However, the “weak” positive response with strain 1535 in the absence of metabolic activation indicates that further research is needed to elucidate the mechanism of action of this rat carcinogen.

Relationship of hepatocarcinogenicity and hepatocellular proliferation induced by mutagenic noncarcinogens vs carcinogens: II. 1- vs 2-nitropropane

2-Nitropropane (2-NP) is mutagenic in a number of short-term mutagenicity assays in vitro and in vivo, and is a potent hepatocarcinogen in rats. A structural isomer, 1-nitropropane (1-NP), is mutagenic in V79 cells and can induce unscheduled DNA synthesis in rat hepatocytes, yet did not induce tumors in rats following chronic exposure. We examined the correlation of cell proliferation and hepatocarcinogenesis induced by this mutagenic noncarcinogen-carcinogen pair in a rat liver proliferation model. Rats were exposed to gavage doses of 0.5, 1, or 2 mmol/kg of 1-NP or 2-NP daily for 10 days; the highest two dose groups were similar to the doses used in the carcinogenesis bioassay. Cell proliferation was quantitated by incorporation of bromodeoxyuridine, detected immunohistochemically, into newly synthesized DNA. Animals exposed to the vehicle exhibited a labeling index (LI) of approximately 1.9% and animals exposed to CCL 4 had a LI of approximately 30%. Rats exposed to the hepatocarcinogen 2-NP exhibited a dose-related increase in LI to 6.3 and 11% at the 1 and 2 mmol/kg doses, respectively, and no increase above control at the 0.5 mmol/kg exposure level. Animals exposed to the noncarcinogenic isomer 1-NP showed no statistically significant increase in LI above controls at any dose level tested. Serum chemistries were consistent with mild to moderate decreases in hepatocellular function, cholestasis, and necrosis following 2-NP exposure, but only minimal effects were observed, probably due to slight dehydration resulting from 1-NP exposure. These data indicate a positive association between increased cell proliferation and hepatocarcinogenesis induced by these two nitropropane isomers.

The rat-liver carcinogen N-nitrosomorpholine initiates unscheduled DNA synthesis and induces micronuclei in the rat liver in vivo

Alkylation of DNA is generally accepted as the primary event in the carcinogenicity of nitrosamines. However, the cyclic nitrosamine N-nitrosomorpholine (NMOR), a potent rat hepatocarcinogen, has been reported as binding at very low levels to the liver DNA of treated rats. This led us to investigate the activity of NMOR in two in vivo rat-liver genotoxicity assays — for the induction of unscheduled DNA synthesis (UDS) and the production of micronucleated hepatocytes in the liver micronucleus assay (LMN). Rats treated with oral doses of NMOR (10–200 mg/kg) gave a positive liver UDS response either 2.5 h or 12 h after dosing. Similarly, treatment with oral doses of NMOR (10 or 100 mg/kg) followed by mitogenic stimulation with 4-acetylaminofluorene (4AAF) resulted in high incidences of micronucleated hepatocytes in the LMN assay. These data confirm that the genotoxicity reported for NMOR in vitro can be reproduced in vivo and that NMOR interacts with liver DNA of treated rats. Earlier reports of only very weak binding of radiolabelled NMOR to rat liver DNA in vivo are discussed within the context of these data.

Methapyrilene is a genotoxic carcinogen: Studies on methapyrilene and pyrilamine in the L5178Y/TK +/− mouse lymphoma assay

Methapyrilene (MP), a sedating antihistamine, is a potent rat hepatocarcinogen which has been thought to be non-genotoxic on the basis of the negative results in a small number of short-term mutagenicity tests. The present studies show that MP is a moderately active mutagen in the L5178Y/TK +/− → TK −/− mouse lymphoma assay (MLA) in the presence of Aroclor-induced rat-liver S9, and that it induces predominantly small-colony thymidine kinase-deficient (TK −/−) mutants of demonstrated chromosomal origin. 10 of 12 small colony TK −/− mutants analyzed by banded karyotype (230-band level of resolution) show aberrations to chromosome 11b, the known location of the single functional TK gene in these cells. The observed aberrations from nine of the mutants included insertions, deletions and translocations while the tenth mutant had highly rearranged, multiple copies of chromosome 11 segments. By varying the concentrations of the S9 protein and cofactors it was shown that our standard S9 composition was close to optimum for activating MP to a mutagen. The activity and stability of various lots of S9 prepared in-house or purchased from a contract laboratory revealed significant differences. The ability of 2 lots of in-house S9 to activate a standard concentration of MP increased rapidly over the first 4 weeks of liquid nitrogen storage then declined slowly over the next 16 weeks. Three separate lots of purchased S9 were essentially inactive for the first 2 weeks of liquid nitrogen storage then increased in activity thereafter; these were the only occasions in which MP was not mutagenic in our hands. The mutagenic activity of pyrilamine (PYR), a structurally related antihistamine which is far less carcinogenic in rats, but easily detected in short-term tests as being genotoxic, was also investigated in the MLA. PYR was slightly less mutagenic than MP over a comparable range of concentrations, and also induced predominantly small-colony mutants. These studies fail to adequately explain the great carcinogenic differences between these two compounds, but are consistent with the potent hepatocarcinogenicity of MP resulting through a mutagenic mechanism.

Liver Tumors in Rodents: Extrapolation to Man

Man is a poor model for the prediction of agents that are hepatocarcinogenic for laboratory rodents. Relatively few agents are known to cause any form of primary liver cancer in man. The most important is hepatitis B virus, for which there is possibly a model in the woodchuck but not one in rats or mice. The only other agents known to cause primary liver cancer in man are certain steroid hormones, vinyl chloride, and thorium dioxide. There are animal models for the first two of these and a reasonable expectation that thorium dioxide would produce liver tumors in animals if the appropriate experiments were done. Aflatoxin, a potent hepatocarcinogen in rats and other species but not mice, is strongly suspected of being an important human hepatocarcinogen in certain geographical areas of the world, but the evidence is circumstantial. There is no more than a weak association between the nutritional type of cirrhosis secondary to excessive intake of alcohol and increased primary liver cancer in man, and no evidence at all that ethanol per se causes liver tumors in mice, rats, hamsters, or mastomys. By contrast, a very large number of chemicals to which people in the West have been exposed for many decades have been found to be hepatocarcinogens in laboratory rodents. In most cases the levels of exposure required to produce liver tumors in rodents far exceed those to which man is normally exposed. The problem is to guess whether low-level exposure to such rodent hepatocarcinogens poses any real liver cancer threat to man?The mortality from primary liver cancer is very low in countries such as England and Wales where there is widespread exposure to low doses of both natural and synthetic agents which, in high dosage, cause liver tumors in rodents. This suggests that, if there is any risk, it can only be very small. Death rate data collected in England and Wales by the Registrar General are consistent with there having been a small increase in the incidence of primary liver cancer in England and Wales during the past 20 years, but the apparent increase might well be a consequence of revisions in the International Classification of Diseases system and not real. During the first half of the present century the age-standardized incidence of primary liver cancer in England and Wales was falling.(ABSTRACT TRUNCATED AT 400 WORDS)

Dinitrotoluene isomer-specific enhancement of the expression of diethylnitrosamine-initiated hepatocyte foci.

Technical grade dinitrotoluene (TDNT) was shown to be a potent hepatocarcinogen in rats; however, ambiguous results were obtained from bioassays which evaluated 2,4-DNT, the principal isomer in the technical mixture. The present studies were designed to determine the relative hepatocyte foci promoting activity of TDNT and the primary isomers present in the technical mixture, 2,4- and 2,6-DNT. A rat hepatic initiation-promotion protocol was used in which male Fischer-344 rats were initiated with a single dose of diethylnitrosamine (DEN) (150 mg/kg, i.p.) and permitted to recover for 2 weeks. Following the recovery period, the animals were fed diets containing TDNT,2,4-DNT,2,6-DNT or phenobarbital (PB). The TDNT animals were killed after 3 or 6 weeks of feeding and the 2,4-DNT-,2,6-DNT-and PB-treated animals were killed after 6 or 12 weeks of feeding. Appropriate DEN and DNT controls were also killed at each time point. Sections from three liver lobes of each animal were stained for gamma-glutamyl transferase (GGT) and the number of GGT+ foci per cm3 (Nv) was calculated using stereological methods. TDNT feeding produced a dose-dependent increase in GGT+ foci (Nv) at 3 and 6 weeks relative to both DEN and TDNT controls. Administration of 2,4-DNT and PB produced time-dependent increases in GGT+ foci (Nv), while 2,6-DNT produced dose- and time-dependent increases. Unlike 2,4-DNT and PB, the high dose (14 mg/kg) of 2,6-dNT produced an increase in GGT+ foci in the control groups and in mean foci volume relative to the DEN control. These results establish that TDNT, 2,4-DNT and 2,6-DNT have hepatocyte foci promoting activity and that 2,6-DNT is approximately 10 times more potent than 2,4-DNT. In addition, TDNT administration neither alters hepatocyte replication following partial hepatectomy nor acts as a 'growth selection' agent in the Solt-Farber 'growth selection' model suggesting that differential effects on hepatocyte replication are not responsible for the promoting activity of TDNT. Based on previous initiation study results from our laboratory and the present data, 2,6-DNT appears to be a complete hepatocarcinogen while under the conditions of these studies 2,4-DNT is an apparent 'pure promoter'. These results provide an explanation for the conflicting DNT bioassay results.

Use of an in vivo/in vitro rat liver DNA repair assay to predict the relative rodent hepatocarcinogenic potency of 3 new azo mutagens.

The in vivo/in vitro rat liver DNA repair assay described by Mirsalis and Butterworth has been employed to compare the relative genotoxicity to the rat liver of three mutagenic analogues of the potent rat hepatocarcinogen 6-dimethylamino-phenylazobenzthiazole (6BT). The compounds evaluated were 6BT, its monomethyl analogue (6-monomethylamino-phenylazobenzthiazole; MA6BT), an analogue in which the -NMe2 group of 6BT is replaced by a piperidinyl group (6-[4-N-piperidinylphenyl]azobenzene; 6PT) and the N-cyanoethyl analogue of MA6BT (6-[p-(N-beta-cyanoethyl-N-methylamino)-phenylazo]benzthiazole; CNEt6BT). The order of relative carcinogenic potency predicted by the Salmonella mutation data was CNEt6BT much greater than MA6BT = 6BT much greater than 6PT. In contrast, that inferred from the in vivo liver DNA repair data was MA6BT greater than 6BT much greater than CNEt6BT much greater than 6PT. This divergence of predictions is discussed in terms of the differing solution properties of the four test chemicals.

An attempt to activate styrene by erythrocytes in bacterial mutagenicity assays

The antihistamine methapyrilene (MP) has been shown to be a potent hepatocarcinogen in rats. However, it has demonstrated little genotoxic activity in a wide variety of short-term tests. In this study, Fischer 344 rats were fed a carcinogenic dose of 0.1% methapyrilene in the diet for 10 weeks prior to sacrifice. S9 was prepared from the livers of the control, MP-treated and Aroclor-induced Fischer 344 rats. Each type of S9 was analyzed for mixed function oxidase activity, cytochrome P-450, and protein content. MP was then evaluated for mutagenicity in 6 strains of S. typhimurium (TA1535, TA1537, TA98, TA100, TA2638 and TA104) and one strain of E. coli (WP2uvrA−) using the standard plate-incorporation assay. MP was not mutagenic in any of the 7 bacterial strains when tested at concentrations ⩽ 10 mg/ml in the presence of each type of S9. However, in the absence of metabolic activation, an approximate 2-fold increase in revertants was noted with strain TA1535.The data from this study show that MP was not converted to a mutagenic metabolite by any of the three S9 types examined. However, the “weak” positive response with strain 1535 in the absence of metabolic activation indicates that further research is needed to elucidate the mechanism of action of this rat carcinogen.

Role of gut flora in the genotoxicity of dinitrotoluene.

Dinitrotoluene (DNT), an important industrial nitroaromatic compound, is a potent hepatocarcinogen in rats. Male Fischer-334 rats fed technical-grade DNT in the diet for 12 months showed a 100% incidence of hepatocellular carcinomas1. However, various in vitro and in vivo short-term tests using several genotoxic end points have failed to show activity with DNT or purified DNT isomers2–6, including unscheduled DNA synthesis (UDS) in primary hepatocytes in vitro7. In the in vivo–in vitro hepatocyte DNA repair assay, which measures chemically induced UDS in hepatocytes isolated from rats treated in vivo8, DNT was found to be a potent UDS inducer9 having a peak of activity 12 h after a single dose. A possible explanation for the discrepancy between the in vivo and in vitro results is that metabolism by gut flora may be required for formation of the proximate or ultimate carcinogenic metabolites of DNT. The importance of biotransformation of toxicants by intestinal microflora has been demonstrated for cycasin10,11 and other xenobiotics12. We have used the in vivo–in vitro hepatocyte DNA repair assay to study the role of intestinal microflora in DNT genotoxicity. We report here extensive DNT-induced DNA repair in rats having the normal complement of gut flora, but not in rats which have no gut flora. These results indicate that metabolism of DNT by gut flora is a necessary step in the genotoxicity of this compound and illustrate the value of the in vivo–in vitro hepatocyte DNA repair assay in assessing the carcinogenic potential of nitroaromatic compounds.