Bioactivation of 2-amino-1-methyl-6-phenylimidazo[4,5-b]-pyridine by liver microsomes from three different rat strains

Collaborative study on fifteen compounds in the rat-liver Comet assay integrated into 2- and 4-week repeat-dose studies

A collaborative trial was conducted to evaluate the possibility of integrating the rat-liver Comet assay into repeat-dose toxicity studies. Fourteen laboratories from Europe, Japan and the USA tested fifteen chemicals. Two chemicals had been previously shown to induce micronuclei in an acute protocol, but were found negative in a 4-week Micronucleus (MN) Assay (benzo[a]pyrene and 1,2-dimethylhydrazine; Hamada et al., 2001); four genotoxic rat-liver carcinogens that were negative in the MN assay in bone marrow or blood (2,6-dinitrotoluene, dimethylnitrosamine, 1,2-dibromomethane, and 2-amino-3-methylimidazo[4,5-f]quinoline); three compounds used in the ongoing JaCVAM (Japanese Center for the Validation of Alternative Methods) validation study of the acute liver Comet assay (2,4-diaminotoluene, 2,6-diaminotoluene and acrylamide); three pharmaceutical-like compounds (chlordiazepoxide, pyrimethamine and gemifloxacin), and three non-genotoxic rodent liver carcinogens (methapyrilene, clofibrate and phenobarbital). Male rats received oral administrations of the test compounds, daily for two or four weeks. The top dose was meant to be the highest dose producing clinical signs or histopathological effects without causing mortality, i.e. the 28-day maximum tolerated dose. The liver Comet assay was performed according to published recommendations and following the protocol for the ongoing JaCVAM validation trial. Laboratories provided liver Comet assay data obtained at the end of the long-term (2- or 4-week) studies together with an evaluation of liver histology. Most of the test compounds were also investigated in the liver Comet assay after short-term (1-3 daily) administration to compare the sensitivity of the two study designs. MN analyses were conducted in bone marrow or peripheral blood for most of the compounds to determine whether the liver Comet assay could complement the MN assay for the detection of genotoxins after long-term treatment. Most of the liver genotoxins were positive and the three non-genotoxic carcinogens gave negative result in the liver Comet assay after long-term administration. There was a high concordance between short- and long-term Comet assay results. Most compounds when tested up to the maximum tolerated dose were correctly detected in both short- and long-term studies. Discrepant results were obtained with 2,6 diaminotoluene (negative in the short-term, but positive in the long-term study), phenobarbital (positive in the short-term, but negative in the long-term study) and gemifloxacin (positive in the short-term, but negative in the long-term study). The overall results indicate that the liver Comet assay can be integrated within repeat-dose toxicity studies and efficiently complements the MN assay in detecting genotoxins. Practical aspects of integrating genotoxicity endpoints into repeat-dose studies were evaluated, e.g. by investigating the effect of blood sampling, as typically performed during toxicity studies, on the Comet and MN assays. The bleeding protocols used here did not affect the conclusions of the Comet assay or of the MN assays in blood and bone marrow. Although bleeding generally increased reticulocyte frequencies, the sensitivity of the response in the MN assay was not altered. These findings indicate that all animals in a toxicity study (main-study animals as well as toxicokinetic (TK) satellite animals) could be used for evaluating genotoxicity. However, possible logistical issues with scheduling of the necropsies and the need to conduct electrophoresis promptly after tissue sampling suggest that the use of TK animals could be simpler. The data so far do not indicate that liver proliferation or toxicity confound the results of the liver Comet assay. As was also true for other genotoxicity assays, criteria for evaluation of Comet assay results and statistical analyses differed among laboratories. Whereas comprehensive advice on statistical analysis is available in the literature, agreement is needed on applying consistent criteria.

Dose-response effects of lycopene on selected drug-metabolizing and antioxidant enzymes in the rat

The administration of lycopene to female rats at doses ranging from 0.001 to 0.1 g/kg b.w. per day for 2 weeks was found to alter the drug-metabolizing capacity and antioxidant status of the exposed animals. An investigation of four cytochrome P450-dependent enzymes revealed that benzyloxyresorufin O-dealkylase activity in the liver was significantly induced in a dose-dependent fashion at all lycopene doses investigated. Likewise, ethoxyresorufin O-dealkylase activity was induced, although only at the two highest lycopene concentrations tested. An investigation of selected phase 2 detoxification enzymes provided evidence that lycopene was capable of inducing hepatic quinone reductase, approximately two-fold, at doses between 0.001 and 0.05 g/kg b.w. per day, whereas no effect was observed at the remaining doses tested. Glutathione transferase, using the two substrates, 2,4-dichloronitrobenzene and 1-chloro-2, 4-dinitrobenzene, was significantly induced at the 0.1 g/kg b.w. per day dose, whereas no effect was observed at the remaining lycopene doses. Analysis of the antioxidant status of the blood compartment revealed that three out of four antioxidant enzymes were affected by lycopene treatment. The activity of superoxide dismutase was thus significantly induced at lycopene doses of 0.005 and 0.05 g/kg b.w, whereas glutathione reductase and glutathione peroxidase was only induced at the 0.005 g/kg b.w. per day dose. For all antioxidant enzymes investigated, the activities seemed to return to the control level after exerting peak induction at doses between 0.005 and 0.05 g/kg b.w. per day. The explanation for this remains unknown. The plasma concentration of lycopene at dietary levels of 0.001, 0.005, 0.05 and 0.1 g/kg b.w. per day was estimated to be 16, 32, 71 and 67 nM, which is barely within the lower range of the mean human plasma concentration of lycopene, which ranges from 70-1790 nM. Oxidative stress induced by the heterocyclic amine, 2-amino-1-methyl-6-phenylimidazo [4,5-b]pyridine (PhIP), and investigated by analyzing for malondialdehyde in plasma, was not found to be affected by prior lycopene exposure. The level of PhIP-DNA adducts in the liver or colon was likewise not affected by lycopene at any dose. Overall, the present study provides evidence that lycopene administered in the diet of young female rats exerts minor modifying effects toward antioxidant and drug-metabolizing enzymes involved in the protection against oxidative stress and cancer. The fact that these enzymatic activities are induced at all of these very low plasma levels, could be taken to suggest that modulation of antioxidant and drug-metabolizing enzymes may indeed be relevant to humans, which in general exhibit a plasma lycopene level several fold above the effective levels observed in this study.

Binding of 2-amino-1-methyl-6-phenylimidazo[4,5-b]pyridine (PhIP) to protein- and low molecular weight thiols and its role in ring hydroxylation

The N-oxidized species of 2-amino-1-methyl-6-phenylimidazo[4,5-b]pyridine (PhIP) have been shown to react with thiols. We have previously characterized a glutathione conjugate of PhIP linked via the C2 of PhIP with apparent loss of the amino group, in rat hepatocytes and PhIP exposed rats. This metabolite was possibly formed from 1-methyl-2-nitro-6-phenylimidazo[4,5-b]pyridine (nitro-PhIP). Upon reacting nitro-PhIP with rat albumin, both in the presence and absence of a reducing system, four products were observed after enzymic proteolysis. One of them was markedly increased after 2-mercaptoethanol treatment of the protein. This adduct was linked to a cysteine-S via C2 of PhIP. Using N2-acetoxy-PhIP as a starting material, an unstable protein adduct was observed which degraded to 50% of the original concentration (t 1/2) after 3 days. This is compatible with the finding that serum PhIP adducts decline rapidly in PhIP exposed rats. Unstable adducts were also formed following the reaction of N2-acetoxy-PhIP with glutathione or cysteine. Based on mass spectroscopy and UV spectra analysis, the suggested structures were RS(-S-)-(H)N2-PhIP. In all cases a degradation product identified as 5-hydroxy-PhIP was formed as characterized by mass spectrometry and NMR spectroscopy. 5-hydroxy-PhIP and its glucuronyl derivative were also observed in rat hepatocytes incubated in vitro with PhIP. In bile of PhIP-exposed rats, only the glucuronyl derivative was observed. Depletion of glutathione reduced the amount excreted in bile and experiments with microsomes indicate that hydroxylation directly at the 5 position is not mediated by cytochrome P-450 mono-oxygenase system. This indicates that 5-hydroxy-PhIP may be formed from N-acetoxy-PhIP via binding to thiols also in cells.