Deactivation of anti-cancer drug letrozole to a carbinol metabolite by polymorphic cytochrome P450 2A6 in human liver microsomes.

Cytochromes P450 (P450) involved in letrozole metabolism were investigated. Among 13 recombinant P450 forms examined, only P450 2A6 and 3A4 showed activities in transforming letrozole to its carbinol metabolite with small K(m) and high Vmax values yielding apparent Vmax/K(m) values of 0.48 and 0.24 nl min(-1) nmol(-1) P450, respectively. The metabolic activities of individual human liver microsomes showed a significant correlation with coumarin 7-hydroxylase activities (P450 2A6 marker) at a letrozole concentration of 0.5 microM, while a good correlation was also seen with testosterone 6beta-hydroxylase activities (P450 3A4 marker) at 5 microM substrate concentration with different inhibition by 8-methoxypsolaren. Significantly low carbinol-forming activities were seen in human liver microsomes from individuals possessing CYP2A6*4/*4 (whole CYP2A6 gene deletion) at a letrozole concentration of 0.5 microM. A Vmax/K(m) value measured for CYP2A6.7 (amino acid substitution type) in human liver microsomes, in the presence of anti-P450 3A4 antibodies, was approximately seven-fold smaller than that for CYP2A6.1 (wild-type). These results demonstrate that P450 2A6 and 3A4 catalyse the conversion of letrozole to its carbinol metabolite in vitro at low and high concentrations of letrozole. Polymorphic variation of CYP2A6 is considered to be relevant to inter-subject variation in therapeutic exposure of letrozole.

Optimal antidiarrhea treatment for antitumor agent irinotecan hydrochloride (CPT-11)-induced delayed diarrhea.

PURPOSE: An antitumor camptothecin derivative CPT-11 has proven a broad spectrum of solid tumor malignancy, but its severe diarrhea has often limited its more widespread use. We have demonstrated from a rat model that intestinal beta-glucuronidase may play a key role in the development of CPT-11-induced delayed diarrhea by the deconjugation of the luminal SN-38 glucuronide, and the elimination of the intestinal microflora by antibiotics or dosing of TJ-14, a Kampo medicine that contains beta-glucuronidase inhibitor baicalin, exerted a protective effect. In the present study, we assessed the efficacy of several potential treatments in our rat model to clarify which is the most promising treatment for CPT-11-induced delayed diarrhea. METHODS AND RESULTS: Oral dosing (twice daily from days -1 to 4) of streptomycin 20 mg/kg and penicillin 10 mg/kg (Str/Pen), neomycin 20 mg/kg and bacitracin 10 mg/kg (Neo/Bac), both of which inhibited almost completely the fecal beta-glucuronidase activity, or TJ-14 1,000 mg/kg improved the decrease in body weight and the delayed diarrhea symptoms induced by CPT-11 (60 mg/kg i.v. from days 1 to 4) to a similar extent. The efficacy was less but significant in activated charcoal (1,000 mg/kg p.o. twice daily from days -1 to 4). In a separate experiment using rats bearing breast cancer (Walker 256-TC), TJ-14, Neo/Bac, and charcoal at the same dose regimen improved CPT-11-induced intestinal toxicity without reducing CPT-11's antitumor activity. In contrast, oral dosing (twice a day) of cyclosporin A (50 mg/kg), a P-glycoprotein and cMOAT/MRP2 inhibitor or valproic acid (200 mg/kg), a UDP-glucuronosyltranferase inhibitor, exacerbated the intestinal toxicity without modifying CPT-11's antitumor activity. CONCLUSIONS: The result clearly demonstrated the ability of Neo/Bac, Str/Pen, and TJ-14, less but significant ability of activated charcoal, to ameliorate CPT-11-induced delayed-onset diarrhea, suggesting the treatments decreasing the exposure of the intestines to the luminal SN-38 are valuable for improvement of CPT-11-induced intestinal toxicity. In contrast, the treatments affecting the biliary excretion of CPT-11 and its metabolites might have undesirable results.

Species difference in N-hydroxylation of a tryptophan pyrolysis product in relation to mutagenic activation.

Metabolic activation of a tryptophan pyrolysate, 3-amino-1-methyl-5H-pyrido[4,3-b]indole (Trp-P-2), in liver microsomes from rats, mice, hamsters, guinea pigs, and rabbits was studied to know whether N-hydroxylation is a common obligatory step for mutagenic activation of Trp-P-2. Among hepatic microsomes obtained from untreated animals, the highest activity of Trp-P-2 N-hydroxylase was observed in microsomes from hamsters, followed by those from guinea pigs, mice, and rabbits. In rats, the activity was low, and there was no appreciable difference between the sexes. The activity of Trp-P-2 N-hydroxylase in microsomes was increased by pretreating the animals with a polychlorinated biphenyl mixture. The induction was most profound in rats, and the activity was enhanced 257-fold, as compared to that of untreated animals. The activity was also enhanced in microsomes from polychlorinated biphenyl mixture-treated rabbits, mice, and hamsters, while the activity was increased only slightly in guinea pigs and was thereby lowest among microsomes from the polychlorinated biphenyl mixture-treated animals. In bacterial mutagenesis test systems using Salmonella typhimurium TA 98, the number of revertants induced by Trp-P-2 was increased in parallel with the microsomal ability of the N-hydroxylation. These results indicate that in all species examined N-hydroxylation is an essential metabolic step for mutagenic activation of Trp-P-2.

Participation of cytochrome P-450 in reductive metabolism of 1-nitropyrene by rat liver microsomes.

Reductive metabolism of carcinogenic 1-nitropyrene by rat liver microsomes and reconstituted cytochrome P-450 systems was investigated. Under the nitrogen atmosphere, 1-aminopyrene was the only detected metabolite of 1-nitropyrene. The reductase activity in liver 105,000 X g supernatant fraction was ascribed to DT-diaphorase, aldehyde oxidase, and other unknown enzyme(s) from the results of cofactor requirements and inhibition experiments. The microsomal reductase activity was inhibited by oxygen, carbon monoxide, 2,4-dichloro-6-phenylphenoxyethylamine, and n-octylamine. Flavin mononucleotide markedly enhanced the activity, and 2-diethylaminoethyl-2,2-diphenylvalerate hydrochloride also enhanced it, but slightly. The microsomal activity was induced by the pretreatment of rats with 3-methylcholanthrene, sodium phenobarbital, or polychlorinated biphenyl, and the increments of the activity correlated well with those of the specific contents of cytochrome P-450 in microsomes. The reductase activity could be reconstituted by NADPH-cytochrome P-450 reductase and forms of cytochrome P-450 purified from liver microsomes of polychlorinated biphenyl-induced rats. Among four forms of cytochrome P-450 examined, an isozyme P-448-IId which showed high activity in hydroxylation of benzo(a)pyrene catalyzed most efficiently the reduction of 1-nitropyrene. The results of this study indicate the central role of cytochrome P-450 in the reductive metabolism of 1-nitropyrene in liver microsomes.

A new form of cytochrome P-450 responsible for mutagenic activation of 2-amino-3-methylimidazo[4,5-f]quinoline in human livers.

Antibodies to P-450IA2 strongly inhibited the mutagenic activation of 2-amino-3-methylimidazo [4,5-f]quinoline (IQ) and 3-amino-1-methyl-5H-pyrido[4,3-b]indole acetate but not aflatoxin B1 in human liver microsomes. The anti-rat P-450IA2 antibodies were capable of recognizing two proteins which show different mobilities on sodium dodecyl sulfate-polyacrylamide gel electrophoresis of human liver microsomes. A new form of cytochrome P-450 (designated P-450-HM4) cross-reactive with anti-rat P-450IA2 antibodies showing that the smaller molecular weight was purified from human liver microsomes by means of the fast-performance liquid chromatography system. The molecular weight of P-450-HM4 was estimated to be 49,000, which was apparently different from that of P-450PA (human P-450IA2). The antibodies to P-450-HM4 did not cross-react with P-450PA (human P-450IA2) but inhibited to various extents the mutagenic activation of IQ in microsomes from human livers. In addition, P-450-HM4 showed significant mutagen-producing activity from IQ in a reconstituted system. Together with these and other results reported previously, it is concluded that at least two forms of cytochrome P-450 [P-450-HM4 and P-450PA (human P-450IA2)] are involved in the mutagenic activation of IQ in human liver.

Microsomal activation of 2-amino-3-methylimidazo[4,5-f]quinoline, a pyrolysate of sardine and beef extracts, to a mutagenic intermediate.

The mechanism involved in the metabolic activation of 2-amino-3-methylimidazo[4,5-f]quinoline, which is a pyrolysate isolated from broiled foods, to a mutagenic intermediate was studied in vitro. In a system containing hepatic microsomes and reduced nicotinamide adenine dinucleotide phosphate, 2-amino-3-methylimidazo[4,5-f]quinoline was converted to a product which was directly mutagenic to Salmonella typhimurium. The structure of the mutagenic metabolite was determined as the 2-N-hydroxy derivative on the basis of the chemical properties and the mass spectral evidence of the azoxy adduct with o-nitrosotoluene. The activation reaction was mediated by microsomal enzymes and was inhibited by carbon monoxide, 7,8-benzoflavone, and other chemicals which were known to inhibit the cytochrome P-450-dependent reaction. With the use of four forms of purified cytochrome P-450, the N-hydroxylation of 2-amino-3-methylimidazo[4,5-f]quinoline and the induction of the reverse mutation of the bacteria were clearly demonstrated to be catalyzed mainly by a high-spin form of cytochrome P-450, P-448 II-a.

Metabolic activation of mutagenic tryptophan pyrolysis products by rat liver microsomes.

In an attempt to determine whether cytochrome P-450 metabolizes tryptophan pyrolysis products, 3-amino-1,4-dimethyl-5H-pyrido[4,3-b]indole (Trp-P-1) and 3-amino-1-methyl-5H-pyrido[4,3-b]indole (Trp-P-2) to active forms, studies were done using microsomes and the 9000 x g supernatant (S-9) from polychlorinated biphenyl mixture-treated rat livers together with Salmonella typhimurium TA 98 as a tester strain. The number of revertants increased with increments in the amount of S-9 fraction added. The highest mutation was seen with an amount of S-9 that is equivalent to 2.5 mg of liver tissue; however, further increase in the S-9 fraction resulted in a sharp decline in the number of revertants. There was a similar biphasic response when a relatively large amount of S-9 fraction was incubated for an increasing length of time. Microsomes showed parallel biphasic responses, and, in addition, the 105,000 x g supernatant fraction was ineffective in increasing the number of revertants. These results suggested that most if not all of the ability of the S-9 fraction to convert Trp-1 and Trp-P-2 to active forms resided in the microsomes. The involvement of microsomal cytochrome P-450 in this process was further confirmed by the following evidence. The treatment of rats with polychlorinated biphenyl mixture and 3-methylcholanthrene resulted in a marked increase in the ability of microsomes to activate Trp-P-1 and Trp-P-2, and the activation reaction required reduced nicotinamide adenine dinucleotide phosphate as a cofactor and was inhibited by carbon monoxide, 7,8-benzoflavone, n-octylamine, and 2-diethylaminoethyl-2,2-diphenylvalerate. The mutagenic metabolite formed as a result of microsomal metabolism of Trp-P-2 was fairly stable and survived heating at 90 degrees for 7 min. With a high-performance liquid chromatography, an active metabolite of Trp-P-2 was purified. A preliminary analysis showed this molecule to be an N-hydroxylated derivative of Trp-P-2.

Fetus-specific CYP3A7 and adult-specific CYP3A4 expressed in Chinese hamster CHL cells have similar capacity to activate carcinogenic mycotoxins.

To assess whether CYP3A4 and CYP3A7 have a similar capacity to activate carcinogenic mycotoxins, we established cell lines stably expressing human CYP3A4 and CYP3A7, which are adult- and fetal-specific forms of cytochrome P450 in human livers, respectively. Each cDNA was introduced into CR-119 cells which had been established by introducing guinea pig NADPH-cytochrome P450 reductase cDNA into Chinese hamster lung cells. The cell lines (4-line and 7-line) stably expressed the mRNA and the protein corresponding to CYP3A4 and CYP3A7, respectively. The concentration-response for aflatoxin B1 (AFB1) cytotoxicity in 4-line and 7-line, respectively, was compared. 4-10 and 7-40 cells were approximately 17- and 20 times more sensitive to AFB1 than the parental CR-119 cells, respectively. In addition, the sensitivities to AFB1 of both 4-10 and 7-40 cells were enhanced approximately seven times by the addition of 10 microM alpha-naphthoflavone, a known activator of CYP3A enzyme, while the sensitivities were suppressed approximately four times by the addition of 100 microM troleandomycin, which forms a metabolite intermediate complex with CYP3A enzyme. Moreover, both cell lines showed approximately 10 and 2 times higher sensitivity to sterigmatocystin and aflatoxin G1 than CR-119 cells, respectively. These results indicate that CYP3A4 and CYP3A7 have essentially similar capacities to activate AFB1, sterigmatocystin, and aflatoxin G1 to produce toxic metabolites.

Stable expression of human CYP1A2 and N-acetyltransferases in Chinese hamster CHL cells: mutagenic activation of 2-amino-3-methylimidazo[4,5-f]quinoline and 2-amino-3,8-dimethylimidazo[4,5-f]quinoxaline.

In order to investigate the metabolic activation pathway of food-derived heterocyclic amines, 2-amino-3-methylimidazo[4,5-f]quinoline (IQ) and 2-amino-3,8-dimethylimidazo[4,5-f]quinoxaline (MeIQx), cultured cell lines which stably expressed human cytochrome P4501A2 (CYP1A2) and N-acetyltransferases (NATs) were developed by the method of complementary DNA (cDNA) transfection. First, a cell line expressing CYP1A2, designated A2R-5, was established from the cell line CR-68, which was previously established by introducing NADPH-cytochrome P-450 reductase cDNA into Chinese hamster CHL cells. The expression of CYP1A2 in the transfected cells was confirmed by determining sensitivity to aflatoxin B1. As the next step, the A2R-5 as well as CR-68 cells were further transfected with human monomorphic NAT (NAT1) or polymorphic NAT (NAT2) cDNAs. The expression of NAT in the transfected cells was confirmed using p-aminobenzoic acid and sulfamethazine as substrates, while no activity was seen in parental CR-68 and A2R-5 cells. The cell line, ANP-25, which expressed both CYP1A2 and NAT2, was approximately 370- and 100-fold more sensitive to IQ and MeIQx, respectively, than parental CR-68 cells in cytotoxicity assays. There were no clear differences in sensitivity to both compounds among CR-68, A2R-5, and the cell lines which expressed NAT1 alone, NAT2 alone, and CYP1A2 plus NAT1. Mutagenicity of IQ and MeIQx at the hypoxanthine-guanine phosphoribosyltransferase locus was also detectable only in ANP-25 cells but not in A2R-5 or the cell line expressing CYP1A2 plus NAT1. From these results, it is proposed that both CYP1A2 and NAT2 (but not NAT1) are required for mutagenic activation of these compounds, implying that acetylator polymorphism may be an important risk factor in the carcinogenicity of these compounds.

Acetyl coenzyme A dependent activation of N-hydroxy derivatives of carcinogenic arylamines: mechanism of activation, species difference, tissue distribution, and acetyl donor specificity.

Acetyl coenzyme A dependent activation of 2-hydroxyamino-6-methyldipyrido[1,2-a:3',2'-d]imidazole (N-OH-Glu-P-1) and 3-hydroxyamino-1-methyl-5H-pyrido [4,3-b]indole (N-OH-Trp-P-2) was investigated using cytosols from hepatic and extrahepatic tissues of various animal species in comparison with that of N-hydroxy-2-aminofluorene. N-OH-Glu-P-1 and N-OH-Trp-P-2 were metabolized to the reactive species capable of binding to transfer RNA through a putative O-acetylation process by liver cytosols. Kidney, small intestinal mucosa, lung, and bladder from hamsters and rats also mediated the reaction, although their activities were lower than that in the liver. Marked species differences in the enzymatic activities of livers were observed. Hamsters showed the highest ability in the activation for N-OH-Glu-P-1 and N-OH-Trp-P-2, followed by rats. Rabbits with a rapid acetylator phenotype, which showed a high activity in the N-acetylation of arylamines, activated N-OH-Glu-P-1 but scarcely N-OH-Trp-P-2. A rabbit with a slow acetylator phenotype, mice, guinea pigs, and a dog showed marginal or nondetectable activities with N-OH-Glu-P-1 and N-OH-Trp-P-2. A typical nonheterocyclic N-hydroxyarylamine, N-hydroxy-2-aminofluorene was also activated by the acetyl coenzyme A dependent system to an intermediate which bound to transfer RNA. However, the acetyl-CoA dependent binding of N-hydroxy-2-aminofluorene was markedly different from those observed with N-OH-Glu-P-1 and N-OH-Trp-P-2 concerning the order of activities among animal species used. In addition to short chain acyl coenzyme As, N-hydroxy-2-acetylaminofluorene also served as an acetyl donor for the activation of N-OH-Glu-P-1 and N-OH-Trp-P-2 in liver cytosol systems. The formation of N-acetyl-N-OH-Glu-P-1, however, was not detected in the cytosolic system of N-OH-Glu-P-1 with acetyl-CoA, suggesting the direct O-acetylation at the N-hydroxy group as a major pathway for the activation of N-hydroxyarylamines.

Metabolism and activation of aflatoxin B1 by reconstituted cytochrome P-450 system of rat liver.

Metabolism and activation of aflatoxin B1, a potent hepatocarcinogenic and mutagenic mycotoxin of Aspergillus flavus, were investigated in the reconstituted enzyme system composed of purified NADPH-cytochrome P-450 reductase and cytochrome P-450 or P-448 of rat liver. The aflatoxin M1 formation was strictly mediated by P-448 purified from the liver microsomes of polychlorobiphenyl- and 3-methylcholanthrene-treated rats, while the aflatoxin Q1 formation, as well as the binding of DNA, were catalyzed by both P-450 and P448. Differences between the kinetic data on metabolism and activation of aflatoxin B1 obtained with the reconstituted cytochrome systems and those obtained with the microsomal and nuclear systems were discussed, and the significance of these biochemical data in the in vivo carcinogenicity of aflatoxin B1 was evaluated.

Evidence for the involvement of N-hydroxylation of 3-amino-1-methyl-5H-pyrido[4,3-b]indole by cytochrome P-450 in the covalent binding to DNA.

The involvement of N-hydroxylation of 3-amino-1-methyl-5H-pyrido[4,3-b]indole (Trp-P-2) by cytochrome P-450 in the formation of covalent binding of Trp-P-2 to DNA, the induction of his+ revertant in the Ames test, and the formation of the active metabolite were confirmed. Among four cytochrome P-450 preparations, PCB-P-448 and MC-P-448 purified from liver microsomes of polychlorinated biphenyl (PCB)- and 3-methylcholanthrene-treated rats, respectively, showed higher activities for induction of mutation by Trp-P-2 than did the other two preparations, PCB-P-450 and PB-P-450 purified from PCB-and phenobarbital (PB)-treated rats, respectively. PCB-P-448 was more active than was PB-P-450 in metabolizing Trp-P-2 to N-hydroxylated Trp-P-2 (N-hydroxy-Trp-P-2). Cytochrome P-450 with higher capacity to form the N-hydroxylated metabolite induced a larger number of his+ revertants. Larger amounts of [1-14C]Trp-P-2 bound covalently to DNA were also seen when PCB-P-448 was incubated with calf thymus DNA and reduced nicotinamide adenine dinucleotide phosphate than with PCB-P-450 and PB-P-450. The direct binding of N-[ring-3H]hydroxy-Trp-P-2 isolated by high-performance liquid chromatography to calf thymus DNA was also demonstrated. These results indicate that N-hydroxylation of Trp-P-2 is an obligatory step for the covalent binding to DNA and mutagenesis of Trp-P-2. Based on these results, we propose that N-hydroxy-Trp-P-2 produced by cytochrome P-450 is important in the exertion of the mutagenicity of Trp-P-2 as it binds to DNA.

Metabolic activation of aflatoxin B1 and 2-amino-3-methylimidazo[4,5-f]-quinoline by human adult and fetal livers.

The mutagenic activation of promutagens by human adult and fetal livers was investigated using the umu test system. Among the promutagens studied, aflatoxin B1 (AFB1) and 2-amino-3-methyl-imidazo[4,5-f] quinoline (IQ) were efficiently activated to mutagens by both adult and fetal livers. 7,8-Benzoflavone inhibited the activation of IQ by fetal livers, but the inhibition observed in fetal livers was much less than that observed in adult livers. Antibodies to P450HM1 (P450111A4) and P450HFLa markedly inhibited the activation of AFB1 by adult and fetal livers, respectively. The formation of genotoxic product(s) from IQ in human adult livers was almost completely inhibited by anti-P448H (P4501A2) antibodies but not by anti-P450HM1 antibodies, whereas that in fetal livers was inhibited by both anti-P450HFLa and anti-P450IA2 antibodies. P450HFLa catalyzed the mutagenic activation of both AFB1 and IQ in a reconstituted system. On the contrary, P450HM1 catalyzed the mutagenic activation of AFB1 but not IQ. A preparation of cytochrome P450 partially purified from human fetal livers and cross-reactive with anti-P450IA2 antibodies was found to be active for mutagenic activation of IQ in a reconstituted system. These results indicate that P450HFLa and P450HM1 are mainly involved in the genotoxic product formation from AFB1 in fetal and adult livers, respectively, and that the metabolic activation of IQ in fetal livers is catalyzed by two forms of cytochrome P450, P450HFLa, and cytochrome P450 immunochemically related to P450IA2 but that in adult livers it is mainly catalyzed by cytochrome P450 related to P450IA2.

Catalysis of the covalent binding of 3-hydroxyamino-1-methyl-5H-pyrido[4,3-b]indole to DNA by a L-proline- and adenosine triphosphate-dependent enzyme in rat hepatic cytosol.

An enzymatic mechanism involved in the activation of 3-hydroxyamino-1-methyl-5H-pyrido[4,3-b]indole (N-hydroxy-Trp-P-2), a mutagenic intermediate of a tryptophan pyrolysate, was studied in vitro. In hepatic cytosol supplemented with adenosine triphosphate and L-proline, N-hydroxy-Trp-P-2 was converted to a form which reacts readily with DNA. The enzyme responsible for the activation was partially purified and identified as prolyl transfer RNA synthetase as judged by their cofactor requirements, inhibition by pyrophosphate or adenosine monophosphate, and copurification of their activities. The prolyl transfer RNA-dependent covalent binding of N-hydroxy-Trp-P-2 to DNA of hepatic cytosol was highest in rats, followed by mice, hamsters, rabbits, and guinea pigs in that order. The capacity for the binding of N-hydroxy-Trp-P-2 was largely consistent with their prolyl transfer RNA synthetase activity. With regard to the ultimate form of N-hydroxy-Trp-P-2 for the covalent binding, a possible formation of N,O-prolyl-3-amino-1-methyl-5H-pyrido[4,3-b]indole was proposed.

In vivo activation of aflatoxin B1 in C57BL/6N mice carrying a human fetus-specific CYP3A7 gene.

The in vivo activation of aflatoxin B1 (AFB1) was assessed by using two transgenic mouse lines, M2 and M10, in which the human fetus-specific CYP3A7 was expressed in the kidney (M2) and the liver (M10), respectively. Male mice of 8 weeks old from these two lines were treated with a single i.p. injection of AFB1 (4 mg/kg body weight). AFB1-N7-guanine adduct was quantified by high-performance liquid chromatography. DNA damage was measured using the alkaline elution technique 2 and 6 h after AFB1 treatment. Administration of AFB1 resulted in a significantly higher level of AFB1-N7-guanine in the livers of M10 transgenic mice compared with their nontransgenic littermates (16.5 +/- 4.2 versus 10.4 +/- 1.2 ng/mg DNA, P < 0.01). The level of this biomarker was also significantly higher in the kidney of the M2 mice compared with control mice (73.0 +/- 6.3 versus 50.2 +/- 9.5 ng/mg DNA, P < 0.01). Similar results were also observed with DNA damage expressed as normalized area above curve (NAAC) in the two transgenic lineages, e.g., NAAC values were significantly higher in the livers of M10 and the kidneys of M2 mice. A dose-response relationship of NAAC values was observed in the livers of M10 mice when treated with AFB1 at different doses ranging from 1 to 16 mg/kg body weight, whereas in nontransgenic mice, only slight but not statistically significant increases of NAAC values were observed. Both the mouse CYP3A11 and GST-Yc subunit were expressed at identical levels in these transgenic lines. The results of this study present further evidence that human fetuses are also under the carcinogenic attack of AFB1 as adults if exposed to this potent carcinogen.

Establishment of a Salmonella tester strain highly sensitive to mutagenic heterocyclic amines.

Heterocyclic amines (HCAs) that are present in cooked foods require metabolic activation to exert their genotoxicity. They undergo activation via N-hydroxylation by cytochrome P450 1A2 (CYP1A2), followed by O-esterification by O-acetyltransferase (OAT). To develop a Salmonella tester strain that is highly sensitive to mutagenic HCAs, we introduced a coexpression plasmid (p1A2OR) carrying human CYP1A2 and NADPH-CYP reductase cDNAs and an expression plasmid (pOAT) carrying Salmonella OAT to Salmonella typhimurium TA1538 to yield a TA1538/ARO strain. The TA1538/ARO strain was proven to express the enzymes, as indicated by high activities of 7-ethoxyresorufin O-deethylase and isoniazid N-acetylase. The TA1538/ARO strain exhibited very high sensitivity to mutagenic HCAs 2-amino-3,4-dimethylimidazo[4,5-f]quinoline, 2-amino-3-methylimidazo[4,5-f]quinoline (IQ), and 2-amino-3,8-dimethylimidazo[4,5-f]quinoxaline and a somewhat higher sensitivity to 2-amino-1-methyl-6-phenylimidazo[4,5-b]pyridine compared with the parent Ames tester strain TA1538. The minimum concentrations of 2-amino-3,4-dimethylimidazo[4,5-f]quinoline, IQ, 2-amino-3,8-dimethylimidazo[4,5-f]quinoxaline, and 2-amino-1-methyl-6-phenylimidazo[4,5-b]pyridine giving positive results were defined by evidence that the number of colonies increased in a dose-dependent manner and reached a number two times higher than that obtained by vehicle alone as a control in the TA1538/ARO strain at concentrations of 0.3, 3, 30, and 1000 pM, respectively. When the membrane and cytosol fractions prepared from TA1538/ARO were added to a mixture containing the parental TA1538, the sensitivity of TA1538 to IQ was much lower than that seen with TA1538/ARO. These results indicate that the intracellular expression of drug-metabolizing enzymes makes the established strain of Salmonella highly sensitive to mutagenic HCAs.

Involvement of beta-glucuronidase in intestinal microflora in the intestinal toxicity of the antitumor camptothecin derivative irinotecan hydrochloride (CPT-11) in rats.

Irinotecan hydrochloride (CPT-11), an antitumor camptothecin derivative, causes severe forms of diarrhea clinically. We characterized CPT-11-induced diarrhea histologically and enzymologically and assessed the relationships between intestinal toxicity and the activity of the enzymes that play a key role in the major metabolic pathway of CPT-11 in rats. CPT-11 (60 mg/kg i.v. for 4 days) induced intestinal toxicity characterized by severe chronic diarrhea, loss of body weight, and anorexia. Histological damage was most severe in the cecum. The segmental difference in the degree of the damage showed good correlation with the beta-glucuronidase activity in the contents of the lumen in each case, but not with the intestinal tissue carboxylesterase activity, which converts CPT-11 to its active form (7-ethyl-10-hydroxycamptothecin). Inhibition of the beta-glucuronidase activity in the intestinal microflora by antibiotics (1 mg penicillin and 2 mg streptomycin per ml of drinking water) markedly ameliorated the diarrhea and reduced cecal damage. Analysis of CPT-11 and its metabolites in the feces indicated that antibiotics completely inhibited the deconjugation of the glucuronic conjugate of 7-ethyl-10-hydroxycamptothecin by beta-glucuronidase. It is suggested that CPT-11-induced diarrhea would be attributable to the damage to the cecum, and that the inhibition of the beta-glucuronidase activity in the intestinal microflora is a major protective effect of antibiotics.

Mouse NADPH-cytochrome P-450 oxidoreductase: molecular cloning and functional expression in yeast.

We published isolation of a mouse NADPH-cytochrome P-450 oxidoreductase cDNA and afterward ascribed the cDNA to the guinea-pig instead of the mouse (Ohgiya, S. et al. (1992) Biochim. Biophys. Acta 1171, 103-105 and Corrigendum (1993) Biochim. Biophys. Acta 1174, 313). We report here nucleotide and deduced amino acid sequences of an NADPH-cytochrome P-450 oxidoreductase cDNA isolated from the ddY mouse. The mouse cytochrome P-450 oxidoreductase shares 98.4% identity with its rat counterpart. In particular, clusters of acidic residues that presumably participate in interaction with cytochrome P-450 are highly conserved in primary structures of mammalian cytochrome P-450 oxidoreductases. The mouse cytochrome P-450 oxidoreductase was functionally expressed in yeast using a modified cDNA clone lacking whole noncoding regions.

Molecular cloning and sequence analysis of hamster CYP2E1.

Two cDNA clones, pHSj31 and pHSj3, coding for CYP2E1 were isolated from a hamster liver cDNA library. The sequence analyses revealed that they encoded the same polypeptide of 493 amino acid residues (M(r) = 56,616) and differed from the length of their 3'-untranslated region. The deduced amino acid sequence of hamster CYP2E1 showed approx. 90% identities with those of the rats and mice, and approx. 80% identities with those of the rabbits, monkeys and humans. The NH2-terminal 35 deduced amino acid sequences of hamster CYP2E1 were completely identical with purified protein, ha P-450j. The Northern blot analysis showed that CYP2E1 was expressed in livers and to lesser extents in kidneys and lungs.

Molecular cloning of monkey P450 1A1 cDNA and expression in yeast.

Monkey P450 1A1 cDNA (MKah1) was isolated from the lambda gt11 cDNA library of a liver from a 3-methylcholanthrene (3MC)-treated crab-eating monkey using a dog P450 1A1 cDNA fragment as a probe. MKah1 was 2453 bp long and contained an entire coding region for a polypeptide of 512 residues. The nucleotide and deduced amino acid sequences of MKah1 displayed 95% and 94% identity with those of the human P450 1A1 gene, respectively. Even in the 3' noncoding region, MKah1 showed 94% homology with human P450 1A1, whereas it showed less than 69% homology with other mammalian P450 1A1. Monkey P450 1A1 mRNA was not detectable in untreated livers, but was induced by polychlorinated biphenyl and 3MC. The expression plasmid (designated as pMKC-1) was constructed by introduction of the coding region of MKah1 into a yeast expression vector (pAM82) containing the promoter of acid phosphatase (APase). Northern blot analysis revealed that monkey P450 1A1 mRNA was expressed in yeast under the control of the APase promoter. Microsomes from yeast transformed by pMKC-1 catalyzed 7-ethoxycoumarin O-deethylation, benzo(a)pyrene hydroxylation and the mutagenic activation of 2-amino-3-methyl-imidazo[4,5-f]quinoline (IQ), 3-amino-1-methyl-5H-pyrido(4,3-b)-indole acetate (Trp-P-2) and 2-amino-6-methyldipyrido(1,2-a:3',2'-d)imidazole acetate (Glu-P-1).