Mutagenicity of the optical isomers of the diastereomeric bay-region chrysene 1,3-diol-3,4-epoxides in bacterial and mammalian cells.

The mutagenic activities of the four optically pure (+)- and (-)-enantiomers of the two diastereomeric bay-region chrysene 1,2-diol-3,4-epoxides were evaluated in histidine-dependent strains of Salmonella typhimurium and in cultured Chinese hamster V79 cells. In strain TA98 of S. typhimurium, (-)-1 alpha, 2 beta-dihydroxy-3 beta, 4 beta-epoxy-1,2,3,4-tetrahydrochrysene was 5 to 10 times more active than the other three optical isomers. However, in strain TA100 of S. typhimurium and in Chinese hamster V79 cells, (+)-1 beta, 2 alpha-dihydroxy-3 alpha, 4 alpha-epoxy-1,2,3,4-tetrahydrochrysene was the most mutagenic diol-epoxide and was from 5 to 40 times more active than the other three optical isomers. The bay-region (+)- and (-)-3,4-epoxy-1,2,3,4-tetrahydrochrysene isomers has identical mutagenic activities in all three systems. These studies indicate that the presence and orientation of the hydroxyl groups play an important role in modulating the mutagenic activity of bay-region epoxides of chrysene in both bacterial and mammalian cells.

Design and synthesis of novel alpha(1)(a) adrenoceptor-selective antagonists. 3. Approaches to eliminate opioid agonist metabolites by using substituted phenylpiperazine side chains.

Dihydropyrimidinones, such as 1, represent a novel class of alpha(1a) adrenoceptor antagonists with potential for the treatment of benign prostatic hyperplasia (BPH) (see part 1 of this series). Analysis of the metabolites of 1 revealed that 4-methoxycarbonyl-4-phenylpiperidine is formed as the major metabolite and is an agonist at the mu-opioid receptor. To circumvent any potential liability resulting from the metabolite, we decided to identify alternate templates devoid of agonist activity at the mu-opioid receptor to replace the 4-methoxycarbonyl-4-phenylpiperidine moiety. The present study describes the synthesis and SAR of dihydropyrimidinones linked to substituted 4-phenylpiperazine containing side chains. Compound (+)-38 was identified as a lead compound with a binding and functional profile comparable to that of 1. The putative metabolite 2-carboxamidophenylpiperazine has negligible affinity for the mu-opioid receptor.

Design and synthesis of novel alpha(1)(a) adrenoceptor-selective antagonists. 2. Approaches to eliminate opioid agonist metabolites via modification of linker and 4-methoxycarbonyl-4-phenylpiperidine moiety.

We have previously described compound 1a as a high-affinity subtype selective alpha(1a) antagonist. In vitro and in vivo evaluation of compound 1a showed its major metabolite to be a mu-opioid agonist, 4-methoxycarbonyl-4-phenylpiperidine (3). Several dihydropyrimidinone analogues were synthesized with the goal of either minimizing the formation of 3 by modification of the linker or finding alternative piperidine moieties which when cleaved as a consequence of metabolism would not give rise to mu-opioid activity. Modification of the linker gave several compounds with good alpha(1a) binding affinity (K(i) = < 1 nM) and selectivity (>300-fold over alpha(1b) and alpha(1d)). In vitro analysis in the microsomal assay revealed these modifications did not significantly affect N-dealkylation and the formation of the piperidine 3. The second approach, however, yielded several piperidine replacements for 3, which did not show significant mu-opioid activity. Several of these compounds maintained good affinity at the alpha(1a) adrenoceptor and selectivity over alpha(1b) and alpha(1d). For example, the piperidine fragments of (+)-73 and (+)-83, viz. 4-cyano-4-phenylpiperidine and 4-methyl-4-phenylpiperidine, were essentially inactive at the mu-opioid receptor (IC(50) > 30 microM vs 3 microM for 3). Compounds (+)-73 and (+)-83 were subjected to detailed in vitro and in vivo characterization. Both these compounds, in addition to their excellent selectivity (>880-fold) over alpha(1b) and alpha(1d), also showed good selectivity over several other recombinant human G-protein coupled receptors. Compounds (+)-73 and (+)-83 showed good functional potency in isolated human prostate tissues, with K(b)s comparable to their in vitro alpha(1a) binding data. In addition, compound (+)-73 also exhibited good uroselectivity (DBP K(b)/IUP K(b) > 20-fold) in the in vivo experiments in dogs, similar to 1a.

Design and synthesis of novel alpha(1)(a) adrenoceptor-selective antagonists. 4. Structure-activity relationship in the dihydropyrimidine series.

We have previously disclosed dihydropyridines such as 1a,b as selective alpha(1a) antagonists as a potential treatment for benign prostatic hyperplasia (BPH). The propensity of dihydropyridines toward an oxidation led us to find suitable replacements of the core unit. The accompanying papers describe the structure-activity relationship (SAR) of dihydropyrimidinones 2a,b as selective alpha(1a) antagonists. We report herein the SAR of dihydropyrimidines such as 4 and highlight the similarities and differences between the dihydropyrimidine and dihydropyrimidinone series of compounds.

Biotransformation of lovastatin--III. Effect of cimetidine and famotidine on in vitro metabolism of lovastatin by rat and human liver microsomes.

The effects of the H2-receptor antagonists, cimetidine and famotidine, on the microsomal metabolism of [14C]lovastatin were investigated. Liver microsomes were prepared from control, phenobarbital- and 3-methylcholanthrene-pretreated rats and humans (male and female). Concentration-dependent inhibition of the metabolism of lovastatin (0.1 mM) was observed with cimetidine (0.1 to 1.0 mM). In contrast, famotidine at a similar concentration was a very weak inhibitor. The formation of 6'beta-hydroxy-lovastatin, the major microsomal metabolite of lovastatin, was similarly inhibited. The results suggest that in vivo metabolic interaction with concomitantly administered lovastatin is less likely with famotidine than with cimetidine. Phenobarbital pretreatment produced 58% stimulation in overall metabolism, whereas 3-methylcholanthrene pretreatment had no effect relative to control rats (5.4 nmol/mg protein/min). Liver microsomes from phenobarbital-pretreated rats produced 67% more of the 6'beta-hydroxy-lovastatin but 63-66% less of the 3''-hydroxy and 6'-exomethylene metabolites. Liver microsomes from 3-methylcholanthrene-treated rats also produced less 3"-hydroxy-lovastatin (49%) but similar quantities of the other two metabolites. 6'beta-Hydroxy-lovastatin was a major metabolite with human liver microsomes. Interestingly with these microsomes, hydroxylation at the 3''-position of the molecule was a negligible pathway and hydrolysis to the hydroxy acid form was not observed. The formation of 6'-exomethylene-lovastatin was also catalyzed by human liver microsomes (0.5 to 0.8 nmol/mg protein/min).

Pharmacokinetics of intravenous alendronate.

Alendronate is a potent bisphosphonate that has been studied for the treatment of osteoporosis and Paget's disease of the bone. To examine the pharmacokinetics of this drug, several groups of postmenopausal women were dosed intravenously in several studies. Twelve patients with metastatic bone disease were administered an intravenous dose of 10 mg of 14C-labeled alendronate (approximately 26 muCi), and plasma, feces, and urine samples were collected for 72 hours. Radioactivity was excreted almost exclusively in urine, and all of it was accounted for by alendronate. Overall recovery accounted for 47% of dose, with the remainder presumed to be retained in bone. Metabolism of alendronate was not observed. Renal clearance of alendronate was 71 mL/min. An additional 10 subjects were given repeated i.v. administrations of alendronate to demonstrate that previous exposure does not alter the pharmacokinetic behavior of the drug. Examination of the findings from these and other studies in which alendronate was administered intravenously revealed that disposition of single doses is linear in the range of 0.125 to 10 mg. With the possible exception of a somewhat greater skeletal retention of a systemically administered dose, the pharmacokinetics of i.v. alendronate were found to be similar to those of other bisphosphonates.

Rizatriptan, a novel 5-HT1B/1D agonist for migraine: single- and multiple-dose tolerability and pharmacokinetics in healthy subjects.

Rizatriptan is a novel 5-HT1D/1B agonist for relief of migraine headache. The pharmacokinetics, metabolite profiles, and tolerability of rizatriptan were examined in a multiple-dose study in healthy subjects. Rizatriptan (N = 24) (or placebo, N = 12) was administered as a single 10 mg dose, followed 48 hours later by administration of one 10 mg dose every 2 hours for three doses on 4 consecutive days, corresponding to the maximum daily dose for a migraine attack. The AUC of rizatriptan and its active N-monodesmethyl metabolite after three 10 mg doses was approximately threefold greater than the plasma concentrations following a single 10 mg dose. Metabolite profiles were similar after single and multiple doses. Adverse events during rizatriptan were mild and transient; similar events occurred during placebo, with a somewhat reduced incidence. Diastolic blood pressure tended to increase compared with placebo (approximately 5 mmHg), particularly on the first multiple-dose day (p < .01 vs. placebo). In conclusion, rizatriptan is well tolerated by healthy subjects during multiple-dose administration, with no unexpected accumulation of drug in plasma.

Effect of inhibitors of alpha-naphthylisothiocyanate-induced hepatotoxicity on the in vitro metabolism of alpha-naphthylisothiocyanate.

The role of S-oxidation in the toxic bioactivation of alpha-naphthylisothiocyanate (ANIT) was investigated. The effects of several thione compounds, inhibitors and an inducer of the cytochrome P-450-dependent mixed function oxidase systems on the in vitro metabolism of ANIT and aminopyrine were determined. Ethionamide, sodium diethyldithiocarbamate (Na-DDTC) and S-methyl diethyldithiocarbamate (Me-DDTC), three agents known to undergo metabolism by an S-oxidative pathway and diminish ANIT's toxicity, inhibited the in vitro enzymatic metabolism of ANIT by rat liver microsomes. Methimazole failed to alter either the hyperbilirubinemic response of ANIT or the in vitro metabolism of ANIT. All four thione compounds (i.e., ethionamide, Me-DDTC, Na-DDTC and methimazole) inhibited the enzymatic metabolism of aminopyrine by rat liver microsomes. Me-DDTC was the most potent, whereas methimazole was the least potent inhibitor of aminopyrine metabolism. Phenobarbital, which potentiates, and SKF-525A, which inhibits the hepatotoxicity of ANIT in vivo, correspondingly stimulated or inhibited the NADPH-dependent metabolism of ANIT and aminopyrine by liver microsomes. N-Decylimidazole (NDI), another classical inhibitor of cytochrome P-450-dependent monooxygenase system, inhibited both the in vivo toxicity and in vitro metabolism of ANIT. NDI also diminished the enzymatic metabolism of aminopyrine by liver microsomes. Thus the results of this study indicate that metabolism of ANIT is intimately related to its toxicity and that ANIT probably undergoes its toxic bioactivation via a cytochrome P-450-dependent S-oxidative pathway.

Metabolism and hepatotoxicity of the naturally occurring benzo[b]pyran precocene I.

The in vitro metabolism of precocene I by liver microsomes from control and treated rats and the effects of precocene I on the function and histology of the rat liver were examined. The major metabolites (80-90% of total metabolites) from all microsomal preparations were the cis and trans 3,4-diols of precocene I produced with a cis/trans isomer ratio of 1:2. These diols appear to arise mainly by spontaneous hydrolysis of precocene I 3,4-oxide. (+)-(3R,4R)-cis- and (-)-(3R,4S)-trans-precocene I 3,4-diols were the predominant enantiomers of the 3,4-diol formed. The enantiomeric excess of these diols (2-50%) is dependent on the microsomal preparation, with microsomes from control rats exhibiting the highest stereoselectivity and microsomes from phenobarbital-treated rats the least. 6-Hydroxyprecocene I was the next major metabolite and was formed to the extent of 5% (control), 10% and 17% (phenobarbital and 3-methylcholanthrene treatment, respectively) of total metabolites. Treatment of rats with a single i.p. dose of precocene I (300 mg/kg) resulted in extensive hepatic damage as evidenced by a marked increase of plasma glutamic pyruvic transaminase levels and histologic observation in liver sections of severe centrolobular necrosis. Although phenobarbital treatment of rats increased the rate of liver microsomal metabolism of precocene I by approximately 50% (nmol products/nmol cytochrome P-450/min) compared to liver microsomes from control rats, hepatic damage caused by precocene I was not significantly affected. Depletion of glutathione levels in the rats with diethyl maleate prior to precocene I treatment dramatically increased the severity of hepatic insult, whereas treatment of the rats with the mixed function oxidase inhibitor piperonyl butoxide prior to treatment with precocene I blocked hepatic damage. Treatment of rats with cysteamine prior to treatment with precocene I protected the animals against the toxic effects. Neither cis nor trans precocene I 3,4-diol nor 3,4-dihydroprecocene I elicited impaired liver function or cellular damage. The above results are consistent with the view that precocene I 3,4-oxide is the metabolite responsible for the hepatotoxic effects observed when precocene I is injected into rats.

Disposition of DMP 811 (L-708,404), a potent angiotensin II receptor antagonist, in the rat, monkey, and chimpanzee.

DMP 811 {4-ethyl-2-n-propyl-1-[(2'-(1H-tetrazole-5-yl)biphenyl-4-yl) methyl]imidazole-5-carboxylic acid; L-708,404} is a highly potent angiotensin II receptor antagonist. The physiological disposition of DMP 811 was examined in the Sprague-Dawley rat, rhesus monkey, and pan troglodytes chimpanzee. Plasma concentrations of DMP 811 were determined by an HPLC assay with fluorescence detection. After intravenous administration of DMP 811 to rats (1 mg/kg), monkeys (0.5 mg/kg), and chimpanzees (0.5 mg/kg), the plasma clearance was 1.3, 1.8, and 0.3 ml/min/kg, respectively. The corresponding volumes of distribution were 0.16, 0.10, and 0.10 liters/kg, and the value for the terminal half-life was 3.0, 2.4, and 10.1 hr in the respective species. After oral administration to rats (5 mg/kg), monkeys (2 mg/kg), and chimpanzees (2 mg/kg), DMP 811 was 8.4%, 10.2%, and 8.0% bioavailable, respectively. The mass balance of [14C]DMP 811 was investigated in rats and monkeys. In rats, the radiolabeled dose was excreted primarily in feces (79% intravenous; 99% oral) with <1% of the dose in urine. In monkeys, the intravenous radiolabeled dose was excreted in both urine (48%) and feces (42%), whereas the oral dose was excreted largely in feces (79%), with an additional 6% in urine. In summary, DMP 811 was cleared slowly in all three species. The oral bioavailability of DMP 811 was low, but consistent across species. Pharmacokinetic data suggest that the low oral bioavailability was not caused by first-pass metabolism, but probably caused by limited absorption.

Regioselectivity and stereoselectivity in the metabolism of HMG-CoA reductase inhibitors.

Biotransformation of three analogs of simvastatin, L-672,201, L-157,012 and L-672,220, by rat liver microsomes has been examined. These compounds differ from each other at the 6' position of the hexahydronaphthalene system. When 6'-substituents were in the alpha configuration, rat liver microsomes catalysed biotransformation primarily at the 6' position. Hydroxylation was stereoselective giving 6' beta-hydroxy derivatives as major metabolites. In contrast, when the 6'-substituent had a beta-configuration, metabolism at this site was blocked. Rates of metabolism (nmols/mg protein/min) also indicated that 6' beta-derivatives were poorer substrates than their 6' alpha-counterparts. The results indicate that cytochrome P-450 exhibits a high degree of regio- and stereoselectivity in the metabolism of HMG-CoA reductase inhibitors.

A radioimmunoassay for the anticonvulsant and neuroprotective agent, MK-801.

A radioimmunoassay is described for MK-801, a potent anticonvulsant and neuroprotective agent. Two immunogens were prepared from N-glutaryl- and N-carboxyethyl-MK-801 by coupling through their carboxyl groups to bovine serum albumin. Radioligands were I-125-iodotyramine conjugates of the same derivatives. Both types of antisera displayed bridge recognition which could be circumvented. In the first case, specificity for N-acyl derivatives was satisfied by acetylating the analyte prior to measurement. Antisera to the N-alkyl derivative yielded a satisfactory assay for MK-801 when the heterologous radioligand was employed. The first of these strategies was adopted for the routine assay. Specificity relative to hydroxylated metabolites was a function both of antiserum selectivity and sample preparation. High plasma concentrations of drugs concomitantly administered to epileptics posed special analytical problems. Assay sensitivity is 40 pg/ml in plasma and the interassay CV is about 5%.

Effect of thiocarbonyl compounds on alpha-naphthylisothiocyanate-induced hepatotoxicity and the urinary excretion of [35S]alpha-naphthylisothiocyanate in the rat.

The effect of disulfiram (DSF), sodium diethyldithiocarbamate (DDTC), methyl diethyldithiocarbamate (Me-DDTC), and ethionamide on the hepatotoxic response of alpha-naphthylisothiocyanate (ANIT) was studied in the rat. The hyperbilirubinemic response of ANIT was significantly inhibited by ip or po DSF pretreatment. A more marked inhibition of toxicity occurred when DSF was given via ip injection. DDTC, Me-DDTC, and ethionamide significantly inhibited ANIT-induced hyperbilirubinemia. Me-DDTC is approximately three times more potent than DDTC as an inhibitor of toxicity. Approximately 16% of a dose of [35S]ANIT was excreted in the urine as inorganic sulfate 48 hr after dosing. Me-DDTC administered simultaneously with [35S]ANIT significantly reduced urinary [35S]sulfate excretion in the first 24 hr. Ethionamide reduced urinary [35S]sulfate excretion. Pretreatment with phenobarbital which stimulates toxicity in vivo increased urinary [35S]sulfate excretion 300% in the first 12 hr. Thus, this study shows that agents which sensitize or protect rats from the toxic effects of ANIT, correspondingly stimulate or inhibit the oxidative desulfuration of [35S]ANIT in vivo.

Disposition of MK-852, a fibrinogen receptor antagonist, in rats and dogs.

MK-852, an antagonist of the platelet fibrinogen receptor GPIIb/IIIa, is the cyclic disulfide N-acetyl-cys-asn-(5,5-dimethyl-4-thiazolidine-carbonyl)- (4-aminomethyl-phe)-gly-asp-cys, monoacetate (all L-amino acids). Radiolabeled MK-852 was synthesized with either a 3H label in the N-acetyl group of the cystine residue or a 14C label in the aminomethyl group. Plasma concentrations of unchanged MK-852 in five rats declined with a mean terminal half-life of 0.92 hr after a 2.5 mg/kg i.v. dose of MK-852; plasma clearance and Vd were 23.1 ml/min/kg and 1.81 liters/kg, respectively. More label was excreted in urine (74-76%) than in feces (7-15%) when either [3H]MK-852 or [14C]MK-852 was given intravenously to groups of four rats (2.5 mg/kg). High concentrations of 3H in rat kidney were consistent with high renal clearance of MK-852, and MK-852 accounted for virtually all of the urinary 3H (and 14C) label. Following a 0.6 mg/kg i.v. dose, the half-life, plasma clearance, and Vd of MK-852 in four dogs were 0.84 hr, 3.93 ml/min/kg, and 0.28 liters/kg, respectively. In dogs, the excretion patterns of radioactivity were similar to those of rats, except that 14C urinary recoveries (79%) were higher than 3H (63%). Unchanged MK-852 represented essentially all of the urinary 3H label. Fractionation of dog 14C urinary radioactivity yielded one major and several minor polar labeled species. The major species was unchanged [14C]MK-852 (quantitated by radioimmunoassay as approximately 80% of the label).(ABSTRACT TRUNCATED AT 250 WORDS)

Microsomal metabolism of the 5-lipoxygenase inhibitor L-739,010: evidence for furan bioactivation.

The novel 5-lipoxygenase inhibitor [1S,5R]-3-cyano-1-(3-furyl)-6-(6-[3-(3 alpha-hydroxy-6,8-dioxabicyclo[3.2.1]octanyl)]pyridin-2-yl- methoxyl)naphthalene (L-739,010), when administered to rats and rhesus monkeys, was found to produce metabolites which appeared to be covalently bound to plasma proteins. Incubation of [14C]L-739,010 with rat liver microsomes did not yield appreciable amounts of soluble metabolites but resulted in covalent binding to microsomal proteins. The covalent binding was NADPH-dependent and was enhanced by 1.5- and 2-fold in liver microsomes from rats, pretreated with phenobarbital and dexamethasone, respectively. Addition of triacetyloleandomycin and diethyldithiocarbamate to the incubation mixture inhibited the covalent binding by 60% and 46%, respectively. These findings suggest that the cytochrome P450 3A family of enzymes play an important role in the bioactivation of L-739,010. The presence of GSH attenuated the covalent binding by 50%, while methoxylamine, an aldehyde trapping agent, blocked the covalent binding completely and, concurrently, produced several soluble metabolic adducts. Subsequently, major methoxylamine adducts were identified by LC-MS/MS and NMR as O-methyloximes of the ring-opened furan moiety of L-739,010. Incubation of L-739,010 with methoxylamine and hepatic microsomes from dog, rhesus monkey, and human produced similar metabolic adducts as those formed by rat liver microsomes. Therefore, under these experimental conditions, the furan moiety, which undergoes oxidative cleavage to the highly reactive 2-butene-1,4-dialdehyde, represents the major site of L-739,010 biotransformation. This putative reactive intermediate could react with microsomal proteins in vitro and physiological proteins in vivo. Since furan bioactivation is believed to be responsible for the toxicity of many furan-containing compounds, the furan moiety of L-739,010 may be regarded as undesirable.

Comparative studies of drug-metabolizing enzymes in dog, monkey, and human small intestines, and in Caco-2 cells.

Drug-metabolizing enzymes were studied in subcellular fractions of dog, monkey, and human small intestines, and in the human adenocarcinoma cell line Caco-2, a commonly used in vitro absorption model. Immunoblot analysis indicated the presence of enzymes related to cytochrome P450 (CYP) 1A1/CYP1A2, CYP2D6, CYP3A, and carboxylesterases (ESs) in human and monkey intestines, and of CYP3A and ES in dog intestines. Catalytically, human and monkey intestines exhibited significant and comparable testosterone 6 beta-hydroxylase, (+)-bufuralol 1'-hydroxylase, and ES activities. In contrast, dog intestine possessed moderate testosterone 6 beta-hydroxylase, much lower ES, and undetectable bufuralol hydroxylase activities. In addition, low tolbutamide methylhydroxylase activity was observed in human and monkey intestines, but not in dog intestines. Of the phase I enzymes investigated, only ES was detected immunologically and functionally in Caco-2 cells. With respect to phase II enzymes, human and monkey intestines contained relatively high intestinal glucuronyltransferase, N-acetyltransferase (NAT), sulfotransferase, and glutathione S-transferase activities. Except for NAT, all phase II enzymes studied were detectable in dog intestines. In Caco-2 cells, acetaminophen sulfation activity was below the limit of detection, whereas all other conjugating activities were evident. Studies of enzyme kinetics and inhibition by known inhibitors of testosterone 6 beta-hydroxylase activity, the major intestinal mono-oxygenase in all species, revealed some similarities between the responsible enzymes. Comparative studies with human liver microsomes suggested the possible involvement of CYP3A enzymes in the intestinal catalysis of testosterone 6 beta-hydroxylation similar to those observed with human hepatic CYP3A. Further studies on ESs, however, revealed multiplicity and species and/or tissue differences in the microsomal and cytosolic enzymes. Based on kinetic studies, monkey intestines and Caco-2 cells possessed NAT activities, with properties similar to those in human intestine and liver. Overall, the results demonstrated that both the preparations of small intestines and Caco-2 cells exhibited significant drug-metabolizing enzyme activities, although several differences were noted between the intestinal enzymes in the animals or in the Caco-2 cells and those found in humans.

Radioimmunoassays for the enantiomeric components of indacrinone and their phenolic metabolites.

MK-286 is a 90:10 mixture of the (+) and (-)-enantiomers of indacrinone, a combination which induces diuresis while maintaining isouricemia. The principal (phenolic) metabolites also possess pharmacological activity and assays for the four entities were needed for clinical studies. Antisera were produced with the stereospecificity required to measure one enantiomer in the presence of the others, but significant cross-reactivity between drug and metabolite necessitated the separation of these two species by means of Sep-Pak cartridges. Recovery was assessed by concurrently fractionated reference samples. The radioligand in each case was a [I-125]-L-iodotyrosine conjugate of the analyte. Absolute assay sensitivities ranged from 2-20 pg/assay tube, corresponding to 0.2-2.0 ng/ml in serum samples. The assay was also applied to urine with analytes at concentrations greater than 16 ng/ml.

Metabolism of antiparkinson agent dopazinol by rat liver microsomes.

Metabolism of dopazinol (DZ) by liver microsomes from control and phenobarbital- and 3-methylcholanthrene-treated rats has been investigated. Liver microsomes from control and treated rats metabolized DZ to N-despropyl-DZ (39-53% of total metabolites); 8-hydroxy-DZ, a catechol metabolite (32-39%); and 5- or 6-hydroxy-DZ (12-20%). The last metabolite was identified as its dehydration product 5,6-dehydro-DZ. N-Dealkylation was favored only slightly over catechol formation (ratio = 1.2) by liver microsomes from control and phenobarbital-treated rats, whereas with liver microsomes from 3-methylcholanthrene-treated rats, N-dealkylation predominated (ratio = 1.7). Liver microsomes from control rats metabolized DZ at a rate of 0.86 nmol/nmol cytochrome P-450/min. Pretreatment of rats with phenobarbital or 3-methylcholanthrene stimulated rates of metabolism by 2.4- and 3-fold, respectively. Metabolism of DZ was inhibited by SKF 525-A, methimazole, and thiobenzamide. SKF 525-A completely inhibited metabolism of DZ, while methimazole and thiobenzamide, two alternate substrates of the microsomal flavin-containing monooxygenase (MFMO) inhibited N-dealkylation only. These results indicated that while the cytochrome P-450-dependent monooxygenase is the primary enzyme system in DZ oxidation, the MFMO also catalyzes the N-dealkylation reaction. The catechol metabolite was converted to isomeric O-methylated derivatives in approximately 1:1 ratio by purified catechol-O-methyl transferase or 105,000g liver cytosol. The late eluting isomer was 8-methoxy-DZ.