Metabolism of 3-methylindole by vaccinia-expressed P450 enzymes: correlation of 3-methyleneindolenine formation and protein-binding.

The toxicity of 3-methylindole (3 MI), a selective pneumotoxin, is dependent upon cytochrome P450-mediated bioactivation 3. Using vaccinia-expressed P450 enzymes, the metabolites of radiolabeled 3 MI produced by 14 individual P450s were identified and quantified by high performance liquid chromatography. Indole-3-carbinol was produced from incubations of 3 MI with only four enzymes. Although 3-methyloxindole was a product of several P450s, human 1A2 was most efficient in producing this metabolite. The toxic intermediate of 3 MI is believed to be a reactive methylene imine, 3-methyleneindolenine. In this study, this intermediate was detected as its mercapturate adduct, when N-acetylcysteine was added to the incubations. 3-Methyleneindolenine was produced by CYP2A6 at a rate of 50.9 +/- 8.9 pmol/mg protein/hr and by CYP2F1 at a rate of 205.7 +/- 12.5 pmol/mg/hr. The mouse 1a-2 and rabbit 4B1 enzymes produced the reactive intermediate in amounts that exceeded that of the human 2F1 enzyme by 1.4-fold and 1.9-fold, respectively. The toxicity of 3 MI is believed to be due to covalent binding of a P450-generated intermediate to critical pulmonary proteins. Comparison of covalent binding studies to the formation of the metabolites revealed a strong correlation between the amount of the 3 MI adduct detected and covalent binding. This study showed that the methylene imine electrophile is produced by only a few P450 enzymes and is the metabolite responsible for the covalent binding and presumably, the toxicity of 3 MI. Remarkable product preferences between the desaturation pathway to form the methyleneindolenine by CYP2F1 and the ring epoxidation pathway to form the oxindole by CYP1A2, were observed.

Identification of beta-glucuronidase-resistant diastereomeric glucuronides of 3-hydroxy-3-methyloxindole formed during 3-methylindole metabolism in goats.

Goats were jugularly infused with the pneumotoxin 3-methylindole (3MI; 15 mg/kg, 0.5 microCi/kg) dissolved in cremophor-EL to characterize the urinary metabolites of 3MI in a ruminant specie. Urine was collected for 36 hr after the beginning of a 2-hr infusion period, and 3MI metabolites were purified using reversed-phase HPLC. Goats excreted 3MI as at least 11 distinct metabolites. Metabolites were characterized using a combination of UV spectroscopy, 1H- and 13C-NMR spectroscopy, and negative-ion FAB/MS. Two of the metabolites (E1 and E2), representing approximately 30% of the urinary radioactivity, were unambiguously identified as diastereomeric glucuronides of 3-hydroxy-3-methyloxindole [HMOI; 3-(beta-D-glucosiduronic acid)-3-methyloxindole]. Glucuronide conjugates were investigated using enzymatic and chemical hydrolysis. These ethereal glucuronides were unique in that they were not readily hydrolyzable with bovine beta-glucuronidase, although one of the diastereomers was hydrolyzed sparingly by beta-glucuronidase from Helix pomatia. Treatment of the glucuronides with 6 M HCI for a 2-hr period liberated unconjugated HMOI. Treatment of each diastereomer with dilute acid (pH 3) or dilute alkali (pH 10) was ineffective at hydrolyzing the conjugates. Goats form HMOI from 3MI and extensively glucuronidate the metabolite before excreting it, as opposed to mice that do not conjugate HMOI before excretion. These ethereal glucuronic acid conjugates seem to be unique in that they are essentially resistant to beta-glucuronidase-catalyzed hydrolysis.

Oxidation at C-1 controls the cytotoxicity of 1,1-dichloro-2,2- bis(p-chlorophenyl)ethane by rabbit and human lung cells.

Isolated rabbit Clara cells and a transformed human bronchial epithelial cell line, BEAS-2B, were used to investigate the mechanism of cytotoxicity of 1,1-dichloro-2,2-bis(p-chlorophenyl)ethane (DDD), a persistent insecticide and stable metabolite of 1,1,1-trichloro-2,2- bis(p-chlorophenyl)ethane. Both BEAS-2B cells and rabbit Clara cells were highly susceptible to DDD toxicity and were partially protected by 1-aminobenzotriazole, a suicide substrate inhibitor of cytochrome P450 enzymes. DDD (0.05 mM) killed 47 +/- 1.8% of rabbit Clara cells and 42 +/- 7.9% of BEAS-2B cells after 3 hr and 84 +/- 3.0% of rabbit Clara cells and 80 +/- 14% of BEAS-2B cells after 6 hr. Consequently, DDD is the most potent Clara cell toxicant recognized to date. The cytotoxicity of DDD to these cells was decreased by deuterium substitution at the C-1 position. Rabbit Clara cells and pulmonary microsomes incubated with 14C-DDD produced the fully oxidized acetic acid metabolite 2,2'-bis(p- chlorophenyl)acetic acid (DDA), but DDA was not formed by Clara cells when DDD was coincubated with 1-aminobenzotriazole. These results support the hypothesis that the cytotoxicity of DDD to susceptible subpopulations of rabbit and human lung cells is, at least in part, caused by cytochrome P450-mediated oxidation of DDD at C-1. A required step for the production of the cytotoxic intermediate is proposed to be the formation of a highly reactive acyl halide intermediate that is readily hydrolyzed to a stable, nontoxic metabolite, DDA.

Identification of goat and mouse urinary metabolites of the pneumotoxin, 3-methylindole.

1. Urine from goats dosed i.v. with 3-methylindole (3MI; 15 mg/kg) or [methyl-14C] 3MI (15 mg/kg, 0.5 microCi/kg) contained at least 11 metabolites of 3MI. 2. Goat metabolized 3MI to sulfate conjugates of 4- or 7-hydroxy-3-methyloxindole, 5- or 6-hydroxy-3-methyloxindole, and 3,5- or 6-dihydroxy-3-methyloxindole; glucuronic acid conjugates of indole-3-carboxylic acid and 4- or 7-hydroxy-3-methyloxindole; and unconjugated 3-hydroxy-3-methyloxindole. Diastereoisomeric glucuronic acid conjugates of 3-hydroxy-3-methyloxindole were also identified in goat urine. 3. Urine from mice dosed i.p. with 3MI (400 mg/kg) or [ring-UL-14C] 3MI (400 mg/kg, 125 microCi/kg) contained at least six metabolites of 3MI. 4. Mice metabolized 3MI to glucuronic acid conjugates of 3,5- or 6-dihydroxy-3-methyloxindole, 5- or 6-hydroxy-3-methyloxindole, and indole-3-carboxylic acid; and unconjugated indole-3-carboxylic acid. Unconjugated 3-hydroxy-3-methyloxindole was identified in mouse urine in a previous report. 5. Both goats and mice metabolized 3MI to a mercapturate, 3-[(N-acetyl-L-cystine-S-yl)methyl]indole, which has been previously identified and was confirmed in this study. 6. 3-Methyloxindole was not identified in the urine of either goats or mice. 7. The major pathways of 3MI biotransformation in goats and mice is the formation of mono- and dihydroxy-3-methyloxindoles and their subsequent conjugation with glucuronic acid or sulfate. 8. There are no apparent qualitative differences in the biotransformation of 3MI between goats and mice that can account for their different sensitivities to 3MI-induced lung injury.

Metabolism of 3-methylindole in human tissues.

Bioactivation of 3-methylindole (3MI), a highly selective pneumotoxin in goats, was investigated in human lung and liver tissues in order to provide information about the susceptibility of humans to 3MI toxicity. Human lung microsomes were prepared from eight organ transplantation donors and liver microsomes from one of the donors were utilized. The 3MI turnover rate with human lung microsomes was 0.23 +/- 0.06 nmol/mg/min, which was lower than the rate with the human liver microsomes (7.40 nmol/mg/min). The activities were NADPH dependent and inhibited by 1-aminobenzotriazole, a potent cytochrome P-450 suicide substrate inhibitor. Covalent binding of 3MI reactive intermediates to human tissues was determined by incubation of 14C-3MI and NADPH with human lung and liver microsomal proteins. Although human lung microsomes displayed measurable covalent binding activity (2.74 +/- 2.57 pmol/mg/min), the magnitude of this reaction was only 4% as large as that seen with human liver microsomes (62.02 pmol/mg/min). However, the covalent binding was protein dependent and also was inhibited by 1-aminobenzotriazole. Therefore, the bioactivation of 3MI to covalently binding intermediates is catalyzed by cytochrome P-450 in human pulmonary tissues. These activities were compared to those activities measured with tissues from goats. Proteins from goat and human pulmonary and hepatic microsomal incubations were incubated with radioactive 3MI, and radioactive proteins were analyzed by SDS-PAGE and HPLC and visualized by autoradiography and radiochromatography, respectively. The results showed that a 57-kDa protein was clearly the most prominently alkylated target associated with 3MI reactive intermediates.(ABSTRACT TRUNCATED AT 250 WORDS)

Isolation of a mercapturate adduct produced subsequent to glutathione conjugation of bioactivated 3-methylindole.

Bioactivation of the pneumotoxin 3-methylindole (3MI) to a methylene imine intermediate has been demonstrated previously by trapping the electrophile with glutathione in goat lung microsomal incubations. To determine whether the same bioactivation process occurs in whole animals, 3MI was administered to goats, mice, and rats, and the urinary metabolites from these three species were analyzed by HPLC for the presence of the mercapturate that would be expected as the processed and excreted form of the 3MI-glutathione adduct. The mercapturate, 3-[(N-acetylcysteine-S-yl)-methyl]indole (3MI-NAC), was identified in the urine from all three species and was isolated from rat urine for structural identification by uv, NMR, and mass spectrometry. Synthetic 3MI-NAC had uv, NMR, and chromatographic characteristics identical to the isolated metabolite. The presence of this mercapturate in the urine of treated animals unequivocally demonstrates that 3MI is bioactivated to the methylene imine in vivo and that the glutathione adduct is also formed, presumably to detoxify the methylene imine.