Human cytosolic enzymes involved in the metabolic activation of carcinogenic aristolochic acid: evidence for reductive activation by human NAD(P)H:quinone oxidoreductase

Aristolochic acid (AA), a naturally occurring nephrotoxin and carcinogen, has been associated with the development of urothelial cancer in humans. Understanding which human enzymes are involved in AA metabolism is important in the assessment of an individual's susceptibility to this carcinogen. Using the 32P-postlabeling assay we examined the ability of enzymes of cytosolic samples from 10 different human livers and from one human kidney to activate the major component of the plant extract AA, 8-methoxy- 6-nitro-phenanthro-(3,4-d)-1,3-dioxolo-5-carboxylic acid (AAI), to metabolites forming adducts in DNA. Cytosolic fractions of both organs generated AAI-DNA adduct patterns reproducing those found in renal tissues from humans exposed to AA. 7-(Deoxyadenosin-N6-yl)aristolactam I, 7-(deoxyguanosin-N2-yl)aristolactam I and 7-(deoxyadenosin-N6-yl)aristolactam II, indicating a possible demethoxylation reaction of AAI, were identified as AA-DNA adducts formed from AAI by all human hepatic and renal cytosols. To define the role of human cytosolic reductases in the activation of AAI, we investigated the modulation of AAI-DNA adduct formation by cofactors or selective inhibitors of the NAD(P)H:quinone oxidoreductase (NQO1), xanthine oxidase (XO) and aldehyde oxidase. We also determined whether the activities of NQO1 and XO in different human hepatic cytosolic samples correlated with the levels of AAI-DNA adducts formed by the same cytosolic samples. Based on these studies, we attribute most of the activation of AA in human cytosols to NQO1, although a role of cytosolic XO cannot be ruled out. With purified NQO1 from rat liver and kidney and XO from buttermilk, the major role of NQO1 in the formation of AAI-DNA adducts was confirmed. The orientation of AAI in the active site of human NQO1 was predicted from molecular modeling based on published X-ray structures. The results demonstrate for the first time the potential of human NQO1 to activate AAI by nitroreduction.     

The binding of aristolochic acid I to the active site of human cytochromes P450 1A1 and 1A2 explains their potential to reductively activate this human carcinogen

Aristolochic acid (AA), a naturally occurring nephrotoxin and carcinogen, has been associated with the development of urothelial cancer in humans. Using the 32P-postlabeling assay we showed that AAI is activated by human recombinant cytochrome P450 (CYP) 1A1, CYP1A2 and NADPH:CYP reductase to species generating DNA adduct patterns reproducing those found in renal tissues from humans exposed to AA. 7-(Deoxyadenosin-N6-yl)aristolactam I, 7-(deoxyguanosin-N2-yl)aristolactam I and 7-(deoxyadenosin-N6-yl)aristolactam II were identified as AA-DNA adducts formed from AAI by the enzymes. The formation of these AA-derived DNA adducts indicates that all the human enzymes reduce the nitro group of AAI to the putative reactive cyclic nitrenium ion responsible for adduct formation. The concentrations of AAI required for its half-maximum DNA binding were 38, 65 and 126 microM AAI for reductive activation by human CYP1A2, CYP1A1 and NADPH:CYP reductase, respectively. CYP1A1 and 1A2 homology modeling followed by docking of AAI to the CYP1A1 and 1A2 active centers was utilized to explain the potential of these enzymes to reduce AAI. Models of human CYP1A1 and 1A2 were constructed on the basis of the crystallographic structure of truncated mammalian CYP enzymes, CYP2B4, 2C5, 2C8, 2C9 and 3A4. The in silico docking of AAI to the active sites of CYP1A1 and 1A2 indicates that AAI binds as an axial ligand of the heme iron and that the nitro group of AAI is in close vicinity to the heme iron of CYP1A2 in an orientation allowing the efficient reduction of this group observed experimentally. The orientation of AAI in the active centre of CYP1A1 however causes an interaction of the heme iron with both the nitro- and the carboxylic groups of AAI. This observation explains the lower reductive potential of CYP1A1 for AAI than CYP1A2, detected experimentally.     

Carcinogenic aristolochic acids upon activation by DT-diaphorase form adducts found in DNA of patients with Chinese herbs nephropathy

Aristolochic acid (AA), a naturally occurring nephrotoxin and rodent carcinogen, has recently been associated with the development of urothelial cancer in humans. Understanding which enzymes are involved in AA activation and/or detoxication is important in the assessment of an individual susceptibility to this natural carcinogen. We examined the ability of enzymes of rat renal and hepatic cytosolic fractions to activate AA to metabolites forming DNA adducts by the nuclease P1-enhanced version of the (32)P-postlabeling assay. Cytosolic fractions of both these organs generated AA-DNA adduct patterns reproducing those found in renal tissues from humans exposed to AA. 7-(Deoxyadenosin-N(6)-yl)aristolactam I, 7-(deoxyguanosin-N(2)-yl)aristolactam I and 7-(deoxyadenosin-N(6)-yl)aristolactam II were identified as AA-DNA adducts formed from AAI and 7-(deoxyguanosin-N(2)-yl)aristolactam II and 7-(deoxyadenosin-N(6)-yl)aristolactam II were generated from AAII by hepatic cytosol. Qualitatively the same AA-DNA adduct patterns were observed, although at lower levels, upon incubation of AAs with renal cytosol. To define the role of cytosolic reductases in the reductive activation of AA, we investigated the modulation of AA-DNA adduct formation by cofactors, specific inducers or selective inhibitors of the cytosolic reductases, DT-diaphorase, xanthine oxidase (XO) and aldehyde oxidase. The role of the enzymes in AA activation was also investigated by correlating the DT-diaphorase- and XO-dependent catalytic activities in cytosolic sample with the levels of AA-DNA adducts formed by the same cytosolic sample. On the basis of these studies, we attribute most of the cytosolic activation of AA to DT-diaphorase, although a role of cytosolic XO cannot be ruled out. With purified DT-diaphorase, the participation of this enzyme in the formation of AA-DNA adducts was confirmed. The binding orientation of AAI in the active site of DT-diaphorase was predicted by computer modeling based on published X-ray structures. The results presented here are the first report demonstrating a reductive activation of carcinogenic AAs by DT-diaphorase.     

Sudan I is a potential carcinogen for humans: evidence for its metabolic activation and detoxication by human recombinant cytochrome P450 1A1 and liver microsomes

1-Phenylazo-2-hydroxynaphthol (Sudan I, C.I. Solvent Yellow 14) is a liver and urinary bladder carcinogen in mammals. We compared the ability of hepatic microsomal samples from different species including human to metabolize Sudan I. Comparison between experimental animals and human cytochromes P450 (CYP) is essential for the extrapolation of animal carcinogenicity data to assess human health risk. Human microsomes generated the pattern of Sudan I metabolites reproducing that formed by hepatic microsomes of rats. Using hepatic microsomes of rats pretreated with specific CYP inducers, microsomes from Baculovirus-transfected insect cells expressing recombinant human CYP enzymes, purified CYP enzymes, and selective CYP inhibitors, we found that rat CYP1A1 and recombinant human CYP1A1 are the most efficient enzymes metabolizing Sudan I. Microsomes from livers (the target of Sudan I carcinogenicity) of different human donors were used to estimate whether authentic human CYPs oxidize Sudan I. Using Western blot analysis and NH(2)-terminal sequencing, we were able to detect and quantify CYP1A1 in human hepatic microsomes. The sequence of nine amino acids of the protein band cross-reacting with antirat CYP1A1 in human microsomes, LFPISMSAT, matched the sequence of human CYP1A1 perfectly (residues 2-10). CYP1A1 expression levels varied significantly among the different human microsomes (0.04-2.4 pmol/mg protein), and constituted <0.6% of the total hepatic CYP complement. All of the human hepatic microsomal samples oxidized Sudan I to C-hydroxymetabolites. Moreover, using the nuclease P1-enhanced version of the (32)P-postlabeling assay, we found that human microsomes were competent in activating Sudan I to form adducts with DNA. The role of specific CYP enzymes in the human hepatic microsomal metabolism was investigated by correlating the CYP-catalytic activities (or CYP contents) in each microsomal sample with the levels of individual metabolites and/or Sudan I-DNA adducts formed by the same microsomes, and by examining the effects of agents that can inhibit specific CYP in Sudan I metabolism. On the basis of these studies, we attribute most of Sudan I metabolism in human microsomes to CYP1A1, but participation of CYP3A4 cannot be ruled out. These results, the first report on the metabolism of Sudan I by human CYP enzymes, strongly suggest a carcinogenic potency of this rodent carcinogen for humans.     

Human enzymes involved in the metabolic activation of the environmental contaminant 3-nitrobenzanthrone: evidence for reductive activation by human NADPH:cytochrome p450 reductase

Determining the capability of humans to metabolize the suspected carcinogen 3-nitrobenzanthrone (3-NBA) and understanding which human enzymes are involved in its activation are important in the assessment of individual susceptibility to this environmental contaminant found in diesel exhaust and ambient air pollution. We compared the ability of eight human hepatic microsomal samples to catalyze DNA adduct formation by 3-NBA. Using two enrichment procedures of the (32)P-postlabeling method, nuclease P1 digestion and butanol extraction, we found that all hepatic microsomes were competent to activate 3-NBA. DNA adduct patterns with multiple adducts, qualitatively similar to those found recently in vivo in rats, were observed. Additionally one major DNA adduct generated by human microsomes was detected. The role of specific cytochromes p450 (p450) and NADPH:p450 reductase in the human hepatic microsomal samples in 3-NBA activation was investigated by correlating the p450- and NADPH: p450 reductase-linked catalytic activities in each microsomal sample with the level of DNA adducts formed by the same microsomes. On the basis of this analysis, most of the hepatic microsomal activation of 3-NBA was attributed to NADPH: p450 reductase. Inhibition of DNA adduct formation in human liver microsomes by alpha-lipoic acid, an inhibitor of NADPH: p450 reductase, supported this finding. Using the purified rabbit enzyme and recombinant human NADPH: p450 reductase expressed in Chinese hamster V79 cells, we confirmed the participation of this enzyme in the formation of 3-NBA-derived DNA adducts. Moreover, essentially the same DNA adduct pattern found in microsomes was detected in metabolically competent human lymphoblastoid MCL-5 cells. The role of individual human recombinant p450s 1A1, 1A2, 1B1, 2A6, 2B6, 2D6, 2C9, 2E1, and 3A4 and of NADPH: p450 reductase in the metabolic activation of 3-NBA, catalyzing DNA adduct formation, was also examined using microsomes of baculovirus-transfected insect cells containing the recombinant enzymes (Supersomes). DNA adducts were observed in all Supersomes preparations, essentially similar to those found with human hepatic microsomes and in human cells. Of all of the recombinant human p450s, p450 2B6 and -2D6 were the most efficient to activate 3-NBA, followed by p450 1A1 and -1A2. These results demonstrate for the first time the potential of human NADPH: p450 reductase and recombinant p450s to contribute to the metabolic activation of 3-NBA by nitroreduction.     

32P-post-labelling analysis of DNA adducts formed by aristolochic acid in tissues from patients with Chinese herbs nephropathy

Recently, we reported that aristolochic acid (AA) a naturally occurring nephrotoxin and carcinogen is implicated in a unique type of renal fibrosis, designated Chinese herbs nephropathy (CHN). Indeed, we identified the principal aristolochic acid-DNA adduct in the kidney of five such patients. We now extend these observations and demonstrate the presence of additional AA-DNA adducts by the 32P-post-labelling method not only in the kidneys, but also in a ureter obtained after renal transplantation. Using the nuclease P1 version of the assay not only the major DNA adduct of aristolochic acid, 7-(deoxyadenosin-N6-yl)- aristolactam I (dA-AAI), but also the minor adducts, 7-(deoxyguanosin- N2-yl)-aristolactam I (dG-AAI) and 7-(deoxyadenosin-N6-yl)-aristolactam II (dA-AAII) were detected, and identified by cochromatographic analyses with TLC and HPLC. Quantitative analyses of six kidneys revealed relative adduct levels from 0.7 to 5.3/10(7) for dA-AAI, from 0.02 to 0.12/10(7) for dG-AAI and 0.06 to 0.24/ 10(7) nucleotides for dA-AAII. The detection of the dA-AAII adduct is consistent with the occurrence of aristolochic acid II (AAII) in the herb powder imported under the name of Stephania tetrandra and confirms that the patients had indeed ingested the natural mixture of AAI and AAII. 32P-post- labelling analyses of further biopsy samples of one patient showed the known adduct pattern of AA exposure not only in the kidney, but also in the ureter, whereas in skin and muscle tissue no adduct spots were detectable. In an attempt to explain the higher level of the dA-AAI adduct compared to the dG-AAI adduct level in renal tissue even 44 months after the end of regimen, the persistence of these two purine adducts was investigated in the kidney of rats given a single oral dose of pure AAI. In contrast to the dG-AAI adduct, the dA-AAI adduct exhibited a lifelong persistence in the kidney of rats. Our data demonstrate that AA forms DNA adducts in human tissue by the same activation mechanism(s) reported from animal studies. Thus, the carcinogenic/mutagenic activity of AA observed in animals could also be responsible for the urothelial cancers observed in two of the CHN patients.      

Identification of a genotoxic mechanism for the carcinogenicity of the environmental pollutant and suspected human carcinogen o-anisidine

2-methoxyaniline (o-anisidine) is an industrial and environmental pollutant and a bladder carcinogen for rodents. The mechanism of its carcinogenicity was investigated with 2 independent methods, 32P-postlabeling and 14C-labeled o-anisidine, to show that o-anisidine binds covalently to DNA in vitro after its activation by human hepatic microsomes. We also investigated the capacity of o-anisidine to form DNA adducts in vivo. Rats were treated i.p. with o-anisidine (0.15 mg/kg daily for 5 days) and DNA from several organs was analyzed by 32P-postlabeling. Two o-anisidine-DNA adducts, identical to those found in DNA incubated with o-anisidine and human microsomes in vitro, were detected in urinary bladder (4.1 adducts per 10(7) nucleotides), the target organ, and, to a lesser extent, in liver, kidney and spleen. These DNA adducts were identified as deoxyguanosine adducts derived from a metabolite of o-anisidine, N-(2-methoxyphenyl)hydroxylamine. This metabolite was identified in incubations with human microsomes. With 9 human hepatic microsomal preparations, we identified the specific CYP catalyzing the formation of the o-anisidine metabolites by correlation studies and by examining the effects of CYP inhibitors. On the basis of these analyses, oxidation of o-anisidine was attributed mainly to CYP2E1. Using recombinant human CYP (in Supersomes) and purified CYPs, the participation of CYP2E1 in o-anisidine oxidation was confirmed. In Supersomes, CYP1A2 was even more efficient in oxidizing o-anisidine than CYP2E1, followed by CYP2B6, 1A1, 2A6, 2D6 and 3A4. The results, the first report on the potential of the human microsomal CYP enzymes to activate o-anisidine, strongly suggest a carcinogenic potential of this rodent carcinogen for humans.     

Expression of cytochrome P450 1A1 and its contribution to oxidation of a potential human carcinogen 1-phenylazo-2-naphthol (Sudan I) in human livers

Cytochrome P450 1A1 (CYP1A1) is one of the most important enzymes implicated in the metabolic activation of carcinogens. To date, there is still conflicting evidence for the expression of enzymatically functional CYP1A1 in human liver. In the present work, we clearly demonstrate that CYP1A1 capable of metabolizing a carcinogen 1-phenylazo-2-naphthol (Sudan I) is expressed in livers of eight American Caucasian donors. Using two independent methods (immunoblotting and N-terminal sequencing), CYP1A1 protein was detected and quantified in all human hepatic microsomes tested in the study. Its levels, ranging from 0.97 to 3.0 pmol/mg protein, correlated with activities catalyzed by this enzyme [7-ethoxyresorufin O-deethylation (EROD) and oxidation of Sudan I], indicating the presence of enzymatically active CYP1A1. Even though levels of CYP1A1 expression are low, <0.7% of total hepatic CYP, the CYP1A1 contribution to oxidation of carcinogenic Sudan I in the test set of human liver microsomes ranges from 12 to 30%.     

Metabolic activation of benzo[a]pyrene in vitro by hepatic cytochrome P450 contrasts with detoxification in vivo: experiments with hepatic cytochrome P450 reductase null mice

Many studies using mammalian cellular and subcellular systems have demonstrated that polycyclic aromatic hydrocarbons, including benzo[a]pyrene (BaP), are metabolically activated by cytochrome P450s (CYPs). In order to evaluate the role of hepatic versus extra-hepatic metabolism of BaP and its pharmacokinetics, we used the hepatic cytochrome P450 reductase null (HRN) mouse model, in which cytochrome P450 oxidoreductase, the unique electron donor to CYPs, is deleted specifically in hepatocytes, resulting in the loss of essentially all hepatic CYP function. HRN and wild-type (WT) mice were treated intraperitoneally (i.p.) with 125 mg/kg body wt BaP daily for up to 5 days. Clearance of BaP from blood was analysed by high-performance liquid chromatography with fluorescence detection. DNA adduct levels were measured by (32)P-post-labelling analysis with structural confirmation of the formation of 10-(deoxyguanosin-N(2)-yl)-7,8,9-trihydroxy-7,8,9,10-tetrahydrobenzo[a]pyrene by liquid chromatography-tandem mass spectrometry analysis. Hepatic microsomes isolated from BaP-treated and untreated mice were also incubated with BaP and DNA in vitro. BaP-DNA adduct formation was up to 7-fold lower with the microsomes from HRN mice than with that from WT mice. Most of the hepatic microsomal activation of BaP in vitro was attributable to CYP1A. Pharmacokinetic analysis of BaP in blood revealed no significant differences between HRN and WT mice. BaP-DNA adduct levels were higher in the livers (up to 13-fold) and elevated in several extra-hepatic tissues of HRN mice (by 1.7- to 2.6-fold) relative to WT mice. These data reveal an apparent paradox, whereby hepatic CYP enzymes appear to be more important for detoxification of BaP in vivo, despite being involved in its metabolic activation in vitro.     

Bioactivation of 3-aminobenzanthrone, a human metabolite of the environmental pollutant 3-nitrobenzanthrone: evidence for DNA adduct formation mediated by cytochrome P450 enzymes and peroxidases

3-Nitrobenzanthrone (3-NBA) is a suspected human carcinogen found in diesel exhaust and ambient air pollution. The main metabolite of 3-NBA, 3-aminobenzanthrone (3-ABA), was detected in the urine of salt mining workers occupationally exposed to diesel emissions. We evaluated the role of hepatic cytochrome P450 (CYP) enzymes in the activation of 3-ABA in vivo by treating hepatic cytochrome P450 oxidoreductase (POR)-null mice and wild-type littermates intraperitoneally with 0.2 and 2mg/kg body weight of 3-ABA. Hepatic POR-null mice lack POR-mediated CYP enzyme activity in the liver. Using the (32)P-postlabelling method, multiple 3-ABA-derived DNA adducts were observed in liver DNA from wild-type mice, qualitatively similar to those formed in incubations using human hepatic microsomes. The adduct pattern was also similar to those formed by the nitroaromatic counterpart 3-NBA and which derive from reductive metabolites of 3-NBA bound to purine bases in DNA. DNA binding by 3-ABA in the livers of the null mice was undetectable at the lower dose and substantially reduced (by up to 80%), relative to wild-type mice, at the higher dose. These data indicate that POR-mediated CYP enzyme activities are important for the oxidative activation of 3-ABA in livers, confirming recent results indicating that CYP1A1 and -1A2 are mainly responsible for the metabolic activation of 3-ABA in human hepatic microsomes. No difference in DNA binding was found in kidney and bladder between null and wild-type mice, suggesting that cells in these extrahepatic organs have the metabolic capacity to oxidize 3-ABA to species forming the same 3-ABA-derived DNA adducts, independently from the CYP-mediated oxidation in the liver. We determined that different model peroxidases are able to catalyse DNA adduct formation by 3-ABA in vitro. Horseradish peroxidase (HRP), lactoperoxidase (LPO), myeloperoxidase (MPO), and prostaglandin H synthase (PHS) were all effective in activating 3-ABA in vitro, forming DNA adducts qualitatively similar to those formed in vivo in mice treated with 3-ABA and to those found in DNA reacted with N-hydroxy-3-aminobenzanthrone (N-OH-ABA). Collectively, these results suggest that both CYPs and peroxidases may play an important role in metabolizing 3-ABA to reactive DNA adduct forming species.     

Rat oesophageal cytochrome P450 (CYP) monooxygenase system: comparison to the liver and relevance in N-nitrosodiethylamine carcinogenesis.

N-nitrosodiethylamine (NDEA) is able to induce tumours in the rat oesophagus. It has been suggested that this could be due to tissue specific expression of NDEA activating cytochrome P450 enzymes. We investigated this by characterizing the oesophageal monooxygenase complex of male Wistar rats and comparing it with that of the liver. Total amount of cytochrome P450, NADPH P450 reductase, cytochrome b5 and cytochrome b5 reductase of the oesophageal mucosa was approximately 7% of what was found in the liver. In addition, major differences were found in the cytochrome P450 isoenzyme composition between these organs: CYP 2B1/2B2 and CYP3A were found only in the liver, whereas CYP1A1 was constitutively expressed only in the oesophagus. Of the two well-known nitrosamine metabolizing enzymes, CYP2A3 was found only in the oesophagus whereas CYP2E1 was exclusively expressed in the liver. Catalytic studies, western blotting and RT-PCR analyses confirmed the expression of CYP2A3 in the oesophagus. CYP2A enzymes are known to be good catalysts of NDEA metabolism. Oesophageal microsomes had a K(m) for NDEA metabolism, which was about one-third of that of hepatic microsomes, but they showed similar activities when compared per nmol of total P450. NDEA activity in the oesophagus was significantly increased by coumarin (CO), which also induced oesophageal CYP2A3. Immunoinhibition of the microsomal NDEA activity showed that up to 70% of this reaction is catalysed by CYP2A3 in the oesophagus, whereas no inhibition of the hepatic NDEA activity could be achieved by the anti-CYP2A5 antibody. NDEA, but not N-nitrosodimethylamine (NDMA) inhibited the oesophageal metabolism of CO. The results of the present investigation show major differences in the enzyme composition of the oesophageal and hepatic monooxygenase complexes, and are in accordance with the hypothesis that the NDEA organotropism could, to a large extent, be due to the tissue specific expression of the activating enzymes.     

Characterization of cytochrome P450-dependent dimethylhydrazine metabolism in human colon microsomes

Polyclonal antibodies to components of the rat liver cytochrome P450 system were used to examine the composition and function of the microsomal cytochrome P450-dependent monooxygenase system of human colonic mucosal cells. Anticytochrome P450 reductase antibody gave a strong band of immunocross-reactivity in human colon microsomes at the same molecular weight level as purified cytochrome P450 reductase from rat liver, as well as hepatic microsomes isolated from untreated or phenobarbital-treated rats. These results demonstrate the presence of cytochrome P450 reductase in human colon cells. Similarly, cytochromes P450 IIB1 and IIA1 also appear to be present in Western blots of human colon microsomes. These antibodies, as well as antibodies to reductase and cytochrome b5, inhibit dimethylhydrazine metabolism in human colon microsomes to varying degrees. These data argue for a functional P450-dependent drug metabolism system in colon capable of activating/metabolizing the colon-specific model carcinogen, 1,2-dimethylhydrazine.     

Aristolochic acid binds covalently to the exocyclic amino group of purine nucleotides in DNA

The plant extract aristolochic acid (AA) has been used as a herbal drug in many cultures since antiquity. In 1982 AA was shown to be mutagenic and a strong carcinogen in Wistar rats. The crude mixture consists of five nitrophenanthrene carboxylic acid derivatives with aristolochic acid I [AA I; 8-methoxy-6-nitro-phenanthro-(3,4-d)-1,3-dioxolo-5-carboxyli c acid] being the major component. The isolated compound has been found to be mutagenic in the Ames assay. The major metabolite of AA I formed under anaerobic conditions in vitro and excreted in vivo in several species including man, is the reduction product aristolactam I. Using the 32P-postlabeling assay, we could show that AA I forms covalent DNA adducts upon metabolic activation in vitro and in vivo in different organs in the rat. Xanthine oxidase, a mammalian nitroreductase, has served as a sufficient model system mimicking the reductive route of in vivo activation of carcinogenic nitroarenes. This paper reports on two major fluorescent adducts of AA I formed by in vitro reaction of AA I with xanthine oxidase and deoxyguanosine or deoxyadenosine. After isolation and purification by preparative HPLC the adducts were characterized by 1H-NMR, FAB mass, UV/Vis and fluorescence spectroscopy. Their structures were elucidated as 7-(deoxyguanosin-N2-yl)-aristolactam I and 7-(deoxyadenosin-N6-yl)-aristolactam I. These findings are in marked contrast to the results reported for other nitroaromatic carcinogens, where C8-modified deoxyguanosine adducts predominate and N2-substituted deoxyguanosine derivatives are found as minor reaction products. Our results suggest a cyclic N-acylnitrenium ion with delocalized positive charge as the ultimate carcinogenic species, binding preferentially to the exocyclic amino group of purine nucleotides in DNA.     

The anticancer drug ellipticine forms covalent DNA adducts, mediated by human cytochromes P450, through metabolism to 13-hydroxyellipticine and ellipticine N2-oxide

Ellipticine is an antineoplastic agent, the mode of action of which is considered to be based on DNA intercalation and inhibition of topoisomerase II. We found that ellipticine also forms the cytochrome P450 (CYP)-mediated covalent DNA adducts. We now identified the ellipticine metabolites formed by human CYPs and elucidated the metabolites responsible for DNA binding. The 7-hydroxyellipticine, 9-hydroxyellipticine, 12-hydroxyellipticine, 13-hydroxyellipticine, and ellipticine N(2)-oxide are generated by hepatic microsomes from eight human donors. The role of specific CYPs in the oxidation of ellipticine and the role of the ellipticine metabolites in the formation of DNA adducts were investigated by correlating the levels of metabolites formed in each microsomal sample with CYP activities and with the levels of the ellipticine-derived deoxyguanosine adducts in DNA. On the basis of this analysis, formation of 9-hydroxyellipticine and 7-hydroxyellipticine was attributable to CYP1A1/2, whereas production of 13-hydroxyellipticine and ellipticine N(2)-oxide, the metabolites responsible for formation of two major DNA adducts, was attributable to CYP3A4. Using recombinant human enzymes, oxidation of ellipticine to 9-hydroxyellipticine and 7-hydroxyellipticine by CYP1A1/2 and to 13-hydroxyellipticine and N(2)-oxide by CYP3A4 was corroborated. Homologue modeling and docking of ellipticine to the CYP3A4 active center was used to explain the predominance of ellipticine oxidation by CYP3A4 to 13-hydroxyellipticine and N(2)-oxide.     

Relationship between content and activity of cytochrome P450 and induction of heterocyclic amine DNA adducts in human liver samples in vivo and in vitro.

This study was designed to estimate a correlation between metabolic activation phenotypes and formation of DNA adducts by heterocyclic amines (HCA) in 15 liver samples from healthy donors. The correlation between the amount of endogenous DNA adducts and the content of cytochrome P450 in human liver samples in vivo was statistically significant at r(2) = 0.71 and P < 0.005. Furthermore, the isolated human liver microsomes were treated in vitro with two HCAs, 2-amino-1-methyl-6-phenylimidazo[4,5-b]pyridine (PhIP) and 2-amino-9H-pyrido[2,3-b]indole (A alpha C), which have been recognized to induce two DNA adducts: 3',5'-diphosphate-N-(2'-deoxyguanosin-8-yl)-PhIP (3',5'-pdGp-C8-PhIP) and 3',5'-diphosphate-N-(2'-deoxyguanosin-8-yl)-A alpha C (3',5'-pdGp-C8-A alpha C). The correlations between the amount of DNA adducts induced by both compounds in vitro and the content of cytochrome P450 in human microsomes are statistically significant at r(2) = 0.69 and r(2) = 0.62 (P < 0.001), respectively. Furthermore, the level of DNA adducts after treatment with PhIP and A alpha C correlated with the activities of three isozymes of cytochrome P450: CYP1A1, CYP1A2, and CYP3A4. Therefore, three chemical inhibitors were used in the experiments: ellipticine against CYP1A1, furafylline against CYP1A2, and troleandomycin against CYP3A4. The highest inhibition levels in the formation of 3',5'-pdGp-C8-PhIP and 3',5'-pdGp-C8-A alpha C adducts were estimated to occur in the presence of furafylline at 56% and 69%, respectively. Ellipticine was involved in the inhibition of 40% of 3',5'-pdG-C8-PhIP adducts and in only 18% of the inhibition of 3',5'-pdGp-C8-A alpha C adducts. Troleandomycin did not significantly inhibit the formation of 3',5'-pdGp-C8-PhIP adducts under these conditions, but it inhibited the formation of 31% of the 3',5'-pdGpC8-A alpha C adducts. We conclude that the formation of DNA adducts can be used as a relevant marker of interindividual variability in the metabolic activation of HCAs in humans.     

Cytochromes P450 in cynomolgus monkeys mutagenically activate 2-amino-3-methylimidazo(4, 5-f)quinoline (IQ) but not 2-amino-3, 8-dimethylimidazo(4, 5-f)quinoxaline(MeIQx)

THE PROMUTAGENIC AND PROCARCINOGENIC HETEROCYCLIC AMINES (HAS) FOUND IN COOKED MEATS ARE N-HYDROXYLATED BY MICRO-SOMAL CYTOCHROME P450 ENZYMES AS THE FIRST STEP IN THEIR METABOLIC ACTIVATION. IN CYNOMOLGUS MONKEYS, ONE OF THE HAS, 2-AMINO-3-METHYLIMIDAZO[4, 5-F: quinoline (IQ), has been shown to be a potent hepatocarcinogen.However, the structurally similar HA 2-amino-3, 8-dimethylimidazo[4,5-fquinoxaline (MelQx) lacks this potency to induce hepato-cellularcarcinoma in monkeys. Liver microsomes from cynomolgus monkeysshow a striking substrate specificity for the metabolic activationof IQ and MelQx, the former being a far better substrate forN-hydroxylation. Western blot analysis showed that cynomolgusmonkey hepatic microsomes constitutively express P450s immunologicallyrelated to the human CYP3A, CYP2C, and low levels of CYP1A1.For comparison, Western blot analysis of rat, human and patasmonkey microsomes was also carried out. Treatment of cynomolgusmonkeys with rifampicin induced hepatic cytochromes P450 relatedto human CYP3A4 and CYP2C9/10 without inducing CYP1A1 or CYP1A2.Immunoblot analysis also showed that chronic exposure of cynomolgusmonkeys to IQ induced hepatic microsomal cytochrome CYP1A1 andCYP1A2, similarly but lesser in magnitude to that observed with2, 3, 7, 8-tetrachlorodibenzo-p-dioxin (TCCD) induction. Usingthe Ames Salmonella mutagenicity assay, we examined the effectof the inducers on the mutagenic activation (i.e. N-hydroxylation)of IQ and MelQx by cynomolgus monkey hepatic microsomes. Wealso examined the mutagenic activation of these HAs by rat,human and patas monkey liver microsomes. Microsomes from cynomolgusmonkeys treated with rifampicin showed a 3-fold increase inthe mutagenic activation of IQ but showed no increase in themutagenic activation of MelQx. Since cytochromes P4503A and/orP4502C are constitutively expressed in cynomolgus monkey hepaticmicrosomes, and upon induction with rifampicin are associatedwith an increased metabolic activation of IQ but not MelQx,it appears that CYP3A and/or CYP2C are the isoform(s) showingthe selective substrate specificity in the metabolic activationof IQ over MelQx. Treatment of monkeys with TCDD significantlyincreased the mutagenic activation of both IQ and MelQx, concomitantwith an induction of CYP1A isozymes. Thus, it appears that TCDD-inducibleCYP1A enzymes W-hydroxylate both substrates without selectivity.Together, these findings suggest that CYP3A and CYP2C are theprincipal isoforms in the cynomolgus monkey, associated withthe metabolic activation implicated in the induction of hepatocarcinogenicityby IQ. Furthermore, the poor metabolic activation of MelQx byCYP3A and CYP2C, coupled with low constitutive levels of CYP1Aisozymes, provide a metabolic explanation for the low hepatocarcinogenicpotency of MelQx in cynomolgus monkeys.     

Human hepatic CYP1A1 and CYP1A2 content, determined with specific anti-peptide antibodies, correlates with the mutagenic activation of PhIP

Mono-specific antibodies targeted to human CYP1A1 and CYP1A2have been produced by immunizing rabbits with protein conjugatesof short synthetic peptides corresponding to residues 290–297and 284–296 respectively, of these enzymes. The antibodytargeted to CYP1A1 bound in immunoblotting to the recombinantprotein expressed in yeast but did not bind to any human hepaticmicrosomal protein, whereas the antibody targeted to CYP1A2bound only to this enzyme in immunoblotting of human hepaticmicrosomal fractions and did not recognize recombinant humanCYP1A1. The intensity of hepatic microsomal CYP1A2 immunoreactivity(n = 5) correlated significantly with a number of activitiescharacteristic of this enzyme: phenacetin O-deethylase (POD),ethoxyresorufin O-deethylase (EROD) and methoxyresorufin O-demethylaseactivities and the ability to activate the dietary carcinogen2-amino-3, 8-dimethylimidazo[4,5-f]quinoxaline (MeIQx), to amutagen. The anti-CYP1A2 anti-peptide antibody consistentlyinhibited both POD and EROD activities, but inhibition was incomplete(28%). In view of the known (>90%) contribution of CYP1A2to these activities and the correlation with antibody binding,this is consonant with an epitope for the anti-CYP1A2 anti-peptideantibody that forms the edge of a functionally important proinhibitorysurface region previously identified in rat cytochromes CYP1A.CYP1A2 immunoreactivity determined by immunoblotting correlatedsignificantly with the ability of human hepatic microsomal fractionsto activate 2-amino-1-methyl-6-phenylimidazo[4,5-b] pyridine(PhIP), another dietary carcinogen, to a mutagen. It is concludedthat CYP1A1 is absent from human liver and that CYP1A2 is likelyto be a major catalyst in the hepatic activation of PhIP.     

DNA adduct formation by the environmental contaminant 3-nitrobenzanthrone in V79 cells expressing human cytochrome P450 enzymes

Diesel exhaust is known to induce tumours in animals. Of the compounds found in diesel exhaust 3-nitrobenzanthrone (3-NBA) is particularly a powerful mutagen. Recently we showed that 3-NBA is genotoxic in vivo in rats by forming specific DNA adducts derived from nitroreduction. In this study a panel of genetically engineered V79 Chinese hamster cell lines expressing various human cytochrome P450 (CYP) enzymes (CYP1A1, CYP3A4) and/or human NADPH:CYP oxidoreductase (CYPOR) was used to identify CYP enzymes involved in the metabolic activation of 3-NBA. We analyzed the formation of specific DNA adducts by 32P-postlabelling after exposing cells to 1 microM 3-NBA. A similar pattern with a total of four distinct 3-NBA-DNA adducts was found in all cells, identical to those detected previously in DNA from rats treated with 3-NBA in vivo. Total adduct levels ranged from 75 to 132 using nuclease P1 and from 103 to 220 adducts per 10(8) nucleotides, using butanol enrichment. Comparison of DNA binding between different V79MZ derived cells revealed that human CYPOR and CYP3A4 were involved in the metabolic activation of 3-NBA. Furthermore, dose-dependent high adduct levels were detected after exposure to 0.01, 0.1 or 1 microM 3-NBA in the subclone V79NH which exhibits high activities of nitroreductase and N,O-acetyltransferase. Our results suggest that nitroreduction is the major pathway in the human bioactivation of 3-NBA. Moreover, acetylation of the initially formed N-hydroxy arylamine intermediates may contribute to the high genotoxic potential of 3-NBA.     

Androgen-dependent renal microsomal cytochrome P-450 responsible for N-hydroxylation and mutagenic activation of 3-methoxy-4-aminoazobenzene in the BALB/c mouse

A murine renal microsomal enzyme responsible for the mutagenic activation of 3-methoxy-4-aminoazobenzene (3-MeO-AAB) was characterized by its catalytic activity for the mutagenic and metabolic conversion of 3-MeO-AAB. Incubation of 3-MeO-AAB with a renal or hepatic microsome fraction from male BALB/c mice in the presence of NADPH and NADH yielded N-hydroxy and 4'-hydroxy metabolites of 3-MeO-AAB as determined by two-dimensional thin layer chromatography, and the enzyme responsible for the N-hydroxylation was named 3-MeO-AAB N-hydroxylase. A mutagenicity test using Salmonella typhimurium TA98 bacteria as a tester strain has revealed that N-hydroxy-3-MeO-AAB is a potent direct mutagen but that 4'-hydroxy-3-MeO-AAB is not mutagenic. Although 3-MeO-AAB N-hydroxylase activity in liver microsomes showed no sex difference, the enzyme activity in the kidney was detected from male mice but not from females. However, administration of testosterone to female mice induced the enzyme in the kidney. Castration of male mice depressed the activity of 3-MeO-AAB N-hydroxylase in renal microsomes but it little affected the hepatic activity, and on administration of testosterone to the castrated mice the depressed renal microsomal activity recovered to a normal level. The activity of 3-MeO-AAB hydroxylase and the amount of cytochrome P-450 in renal microsomes showed a close correlation. Both renal and hepatic microsomes required NADPH as a main cofactor to mutagenize 3-MeO-AAB and to yield N-hydroxy-3-MeO-AAB from 3-MeO-AAB, and the enzyme activity was strongly inhibited by 7,8-benzoflavone. When the activities of renal and hepatic 3-MeO-AAB N-hydroxylase were compared on the basis of the amount of cytochrome P-450, the renal type enzyme showed about 8 times greater activity than hepatic type enzyme. These results indicate that the kidney contains an androgen-dependent microsomal 3-MeO-AAB hydroxylase which is different from an isozyme present in the liver and which is a new type of cytochrome P-450 isozyme.     

Cytochrome P450 species involved in the metabolism of quinoline

Quinoline is a hepatocarcinogen in rats and mice and a well-knownmutagen in bacteria after incubation with rat liver microsomes.The specific cytochrome P450 enzymes involved in quinoline metabolismin human and rat liver microsomes were determined using cDNA-expressedcytochrome P450s, correlations with specific cytochrome P450-linkedmonooxygenase activities in human liver microsomes and inhibitionby specific inhibitors and antibodies. CYP2A6 is the principalcytochrome P450 involved in the formation of quinoline-1-oxidein human liver micro-somes (correlation coefficient r = 0.95),but is formed in only minute quantities in rat liver microsomes.CYP2E1 is the principal cytochrome P450 involved in the formationof 3-hydroxyquinoline (r = 0.93) in human liver microsomes andis involved in the formation in rat liver microsomes. A highcorrelation coefficient (r = 0.91) between CYP2A6 activity andquinoline-5,6-diol formation in human liver microsomes was observed,but this most likely reflects the involvement of CYP2A6 in theformation of quinoline-5,6-epoxide, from which the quinoline-5,6-diolis formed, as conversion of quinoline-5,6-epoxide to quinoline-5,6-diolon incubation of the epoxide with CYP2A6 could not be demonstrated.A cDNA-expressed human microsomal epoxide hydrolase, however,efficiently converted the epoxide to the diol and the microsomalepoxide inhibitor cyclohexene oxide inhibited quinoline-5,6-diolformation in rat liver microsomes. A preliminary kinetic analysisof quinoline metabolism in human liver microsomes was carriedout and Eadie-Hofstee plots indicate that the formation of quinoline-5,6-diolis monophasic, while that of quinoline-1-oxide and 3-hydroxyquinolineis biphasic.