Substituent effects on the bioactivation of 2-(N-hydroxyacetamido)fluorenes by N-arylhydroxamic acid N,O-acyltransferase.

A series of 7-substituted analogues of 2-(N-hydroxyacetamido)fluorene (1) was subjected to bioactivation by a partially purified preparation of hamster hepatic AHAT, and the rates of methylthio adduct formation resulting from the reaction of the activated intermediates with N-acetylmethionine were determined. Electronegative substituents enhanced the amount of adduct formed; this finding contrasted with the results of a previous study in which it was found that electron-donating substituents facilitated the mechanism-based inactivation of AHAT by analogues of 1. The structures of the adducts formed from reaction of the activated forms of several of the 7-substituted compounds with N-acetylmethionine and with 2'-deoxyguanosine were determined; the types of adducts formed were similar to those formed with electrophiles generated by the AHAT-catalyzed activation of 1. Electronegative substituents enhanced the amount of adducts formed in the reaction with 2'-deoxyguanosine as well as with N-acetylmethionine.

Synthesis and evaluation of N-(phenylalkyl)acetohydroxamic acids as potential substrates for N-arylhydroxamic acid N,O-acyltransferase.

N-(4-Phenylcyclohexyl)acetohydroxamic acid and a series of N-(phenylalkyl)acetohydroxamic acids were synthesized and evaluated as substrates for partially purified rat and hamster hepatic arylhydroxamic acid N,O-acyltransferase systems (AHAT). The compounds were assayed for their abilities to function as acetyl donors in the AHAT-mediated transacetylation of 4-aminoazobenzene and for their abilities to participate in the AHAT-mediated conversion of N-arylhydroxylamines to electrophilic intermediates that form methylthio adducts upon reaction with N-acetylmethionine. None of the newly synthesized compounds displayed significant activity in either of the assays. The results of this study indicate that acetohydroxamic acids that have the nitrogen atom of the hydroxamic acid group attached directly to aliphatic or cycloalkyl groups are not likely to serve as substrates or inhibitors of AHAT.

Mechanism of S-(1,2-dichlorovinyl)glutathione-induced nephrotoxicity.

S-(1,2-Dichlorovinyl)glutathione and S-(1,2-dichlorovinyl)-DL-cysteine are potent nephrotoxins. Agents that inhibit gamma-glutamyl transpeptidase, cysteine conjugate beta-lyase, and renal organic anion transport systems, namely L-(alpha S,5S)-alpha-amino-3-chloro-4,5-dihydro-5-isoxazoleacetic acid (AT-125), aminooxyacetic acid, and probenecid, respectively, protected against S-conjugate-induced nephrotoxicity. Furthermore, S-(1,2-dichlorovinyl)-DL-alpha-methylcysteine, which cannot be cleaved by cysteine conjugate beta-lyase, was not nephrotoxic. These results strongly support a role for renal gamma-glutamyl transpeptidase, cysteine conjugate beta-lyase, and organic anion transport systems in S-(1,2-dichlorovinyl)glutathione- and S-(1,2-dichlorovinyl)cysteine-induced nephrotoxicity.

Renal cellular transport, metabolism, and cytotoxicity of S-(6-purinyl)glutathione, a prodrug of 6-mercaptopurine, and analogues.

The disposition of S-(6-purinyl)glutathione (6-PG) and its metabolites, including the antitumor agent 6-mercaptopurine (6-MP), was characterized in freshly isolated renal cortical cells from male F344 rats to assess the ability of the kidney to convert 6-PG to 6-MP. The intracellular transport and accumulation of 6-PG and 6-MP, the metabolism of 6-PG to 6-MP, and the potential cytotoxicity of 6-MP, 6-thioxanthine (6-ThXan), and 6-thioguanine (6-ThGua) were determined. 6-PG and 6-MP were accumulated by renal cortical cells by time- and concentration-dependent processes, reaching maximal levels of 14.2 and 1.52 nmol/10(6) cells, respectively, with 1 mM concentrations of each compound. Treatment with acivicin, an inhibitor of 6-PG metabolism by gamma-glutamyltransferase, increased accumulation of 6-PG, and treatment with alpha-keto-gamma-methiolbutyrate, a keto acid cosubstrate that stimulates activity of the cysteine conjugate beta-lyase (beta-lyase), which generates 6-MP, decreased accumulation of 6-PG. Incubation of renal cells with 10 mM 6-PG generated 6-MP at a rate of 2.4 nmol/min per 10(6) cells, demonstrating that the beta-lyase pathway forms the desired product from the prodrug within the intact renal cell. Preincubation of cells with acivicin or aminooxyacetic acid, an inhibitor of the beta-lyase, decreased the net formation of 6-MP, demonstrating further the function of the beta-lyase. 6-MP, 6-ThXan, and 6-ThGua exhibited approximately equivalent cytotoxicity (45-55% release of lactate dehydrogenase with 1 mM at 2 hr) in isolated renal cells. Based on the known antitumor potency of these agents, this suggests that cytotoxicity and antitumor activity occur by distinct mechanisms. The high amount of accumulation of 6-PG and its subsequent metabolism to 6-MP, as compared with the relatively low amount of accumulation of 6-MP, in renal cells suggest that 6-PG can function as a prodrug and is a more effective delivery vehicle for 6-MP to renal cells than 6-MP itself. Administration of 6-PG may be an effective means of treating renal tumors or suppressing renal transplant rejection.

Methimazole as a protectant against cisplatin-induced nephrotoxicity using the dog as a model.

The protective effect of methimazole, a commonly used antithyroid drug, on cisplatin-induced nephrotoxicity was studied. Eight dogs received 80 mg/m2 cisplatin i.v. without saline prehydration. Dogs were randomized into two groups of four dogs each: one group received 40 mg/kg methimazole i.p. at 30 min prior to and 4 h after cisplatin delivery, and the other group received saline placebo i.p. Methimazole protected dogs against the in vivo nephrotoxicity elicited by cisplatin as evidenced by clinicopathologic and histopathologic indices. Protection was not complete, as methimazole-treated animals developed mild histopathologic renal changes. Measures of renal oxidative stress did not differ between the two groups at day 5 following cisplatin treatment. No difference was noted for serum thyroxine concentrations before or after therapy in either group; however, serum levels of 3,5,3'-triiodothyronine were significantly higher on day 5 in both groups of dogs receiving cisplatin, regardless of whether they received methimazole or not. Methimazole as used in this study was found to be well tolerated in dogs over the short term, with no significant clinical or clinicopathologic toxicity being observed. The results of this study support the additional evaluation of methimazole as a protectant against cisplatin-induced nephrotoxicity using the dog as a model.

Biochemistry of 1,3-butadiene metabolism and its relevance to 1,3-butadiene-induced carcinogenicity.

Recently, the roles of specific P450 isoforms, myeloperoxidase (MPO), GSH-S-transferase and epoxide hydrolase in the metabolism of 1,3-butadiene, and its major oxidative metabolite, butadiene monoxide (BM), were investigated. The results provided evidence for P450s 2A6 and 2E1 being major catalysts of 1,3-butadiene oxidation in human liver microsomes. cDNA-expressed human P450s 2E1, 2A6, and 2C9 catalyzed BM oxidation to meso- and (+/-)-diepoxybutane (DEB), but the rates of BM oxidation in mouse, rat, or human liver microsomes were much lower than the rates of 1,3-butadiene oxidation in these tissues. Human MPO catalyzed 1,3-butadiene oxidation to BM, but MPO incubations with BM did not yield DEB. Rates of BM formation in mouse and human liver microsomes were similar and were nearly 3.4-fold higher than that obtained with rat liver microsomes. However, rat liver epoxide hydrolase activity was nearly 2-fold higher than that of mouse liver microsomes. Rat and mouse liver GSH-S-transferases exhibited similar BM conjugation kinetics, but rats excreted more BM-mercapturic acids compared to mice given low equimolar doses of BM. BM reacted with guanosine and adenosine to yield N7-, N2-, and N1-guanosinyl and N6-adenosinyl adducts, respectively. These results may contribute to a better understanding of the biochemical basis of 1,3-butadiene-induced carcinogenicity.

Potential role of the flavin-containing monooxygenases in the metabolism of endogenous compounds.

Several xenobiotics and their corresponding cysteine S-conjugates are metabolized in vivo to cysteine S-conjugate sulfoxides and/or N-acetylcysteine S-conjugate sulfoxides. Homocysteine S-conjugates, such as methionine and ethionine, are also metabolized in vivo to sulfoxides. The enzymatic basis for these metabolic reactions is not known. Recently, the rat liver and kidney S-benzyl-L-cysteine S-oxidase activities were found to be associated with flavin-containing monooxygenases that are structurally and immunochemically related to known FMO1 isoforms. Further evidence for FMO1 being the major FMO isoform involved in S-benzyl-L-cysteine sulfoxidation was obtained from kinetic studies with cDNA-expressed rabbit FMOs. Endogenous cysteine S-conjugates, e.g. cysteinylcatecholamines, cysteinylleukotrienes, lanthionine and djenkolic acid may also be substrates for FMOs, since S-benzyl-L-cysteine can be considered a model for these compounds. Methionine, an endogenous homocysteine S-conjugate, was shown to be a substrate for cDNA-expressed rabbit FMO1, FMO2, and FMO3, however, the methionine sulfoxidation reaction was preferentially catalyzed by FMO3. These results suggest that FMOs may also play a role in the in vivo metabolism of endogenous homocysteine S-conjugates.

Cellular and molecular basis for species, sex and tissue differences in 1,3-butadiene metabolism.

Species differences in 1,3-butadiene (BD) bioactivation and detoxication have been implicated in the greater sensitivity of mice to the carcinogenic effects of BD compared to rats, but the molecular basis for species differences in BD metabolism is not well understood. Previous and recent work conducted in this laboratory has examined the relative rates of BD oxidation to epoxybutene (EB) in male and female B6C3F1 mouse tissues, characterized the major cytochrome P450 enzymes involved in BD bioactivation in these tissues, and determined the potential utility of the freshly isolated hepatocyte model to investigate species differences in metabolism of BD and related compounds. Collectively, the results suggest a role for P450s 2E1, 2A5, and 4B1 in sex and tissue differences in BD bioactivation in the mouse. When coordinated metabolism of EB was investigated in male B6C3F1 mouse and Sprague-Dawley rat hepatocytes, the hepatocytes from both species were found to catalyze EB oxidation to meso- and (+/-)-diepoxybutane (DEB), EB hydrolysis to 3-butene-1,2-diol (BDD), and EB conjugation to form GSH conjugates (GSEB). The metabolite area under the curve (AUC) exhibited dependence on the EB concentration used. However, the EB activation/detoxication ratios with the mouse hepatocytes were much higher than the ratios obtained with the rat hepatocytes. These results illustrate the potential utility of the hepatocyte model for estimating flux through competing metabolic pathways and predicting in-vivo metabolism of EB. Collectively, the results may allow a better understanding of the molecular and kinetic basis of species differences in BD metabolism and may lead to a more accurate assessment of human risk.

Advances in the mass spectrometry of hemoglobin adducts: global analysis of the covalent binding of butadiene monoxide.

A common method to assess exposure to 1,3-butadiene through both occupational and environmental routes involves the detection of hemoglobin adducts formed by the primary reactive metabolite butadiene monoxide (EB). This assay is a modification of the Edman degradation procedure, which was developed to determine adducts formed specifically at the amine group of the N-terminal valine of hemoglobin. The goals of the current research are to determine the global modification of alpha- and beta-globin chains by EB and to localize the primary reactive residues to specific regions of the globin polypeptides. The degree of modification was monitored by electrospray mass spectrometry, which was used to measure the formation of EB-hemoglobin adducts (up to ten adducts per globin). Structural analysis of these modifications was performed by peptide mapping of globin peptides after trypsin digestion using liquid chromatography-mass spectrometry. These experiments provided information as to the relative reactivity of alpha- and beta-globin towards EB, as well as to the localization of adducts to specific peptide sequences. The results reveal variable reactivities of alpha- and beta-globin towards EB and also show the formation of multiple adducts at several alpha- and beta-globin sites. In addition, it is established that the N-terminal valine residues are not the first to be modified by EB.

Mutagenicity of amino acid and glutathione S-conjugates in the Ames test.

The mutagenicity of the glutathione S-conjugate S-(1,2-dichlorovinyl)glutathione (DCVG), the cysteine conjugates S-(1,2-dichlorovinyl)-L-cysteine (DCVC) and S-(1,2-dichlorovinyl)-DL-alpha-methylcysteine (DCVMC), and the homocysteine conjugates S-(1,2-dichlorovinyl)-L-homocysteine (DCVHC) and S-(1,2-dichlorovinyl)-DL-alpha-methylhomocysteine (DCVMHC) was investigated in Salmonella typhimurium strain TA2638 with the preincubation assay. DCVC was a strong, direct-acting mutagen; the cysteine conjugate beta-lyase inhibitor aminooxyacetic acid decreased significantly the number of revertants induced by DCVC; rat renal mitochondria (11,000 X g pellet) and cytosol (105,000 X g supernatant) with high beta-lyase activity increased DCVC mutagenicity at high DCVC concentrations. DCVG was also mutagenic without the addition of mammalian activating enzymes; the presence of low gamma-glutamyltransferase activity in bacteria, the reduction of DCVG mutagenicity by aminooxyacetic acid, and the potentiation of DCVG mutagenicity by rat kidney mitochondria and microsomes (105,000 X g pellet) with high gamma-glutamyltransferase activity indicate that gamma-glutamyltransferase and beta-lyase participate in the metabolism of DCVG to mutagenic intermediates. The homocysteine conjugate DCVHC was only weakly mutagenic in the presence of rat renal cytosol, which exhibits considerable gamma-lyase activity, this mutagenic effect was also inhibited by aminooxyacetic acid. The conjugates DCVMC and DCVMHC, which are not metabolized to reactive intermediates, were not mutagenic at concentrations up to 1 mumole/plate. The results demonstrate that gamma-glutamyltransferase and beta-lyase are the key enzymes in the biotransformation of cysteine and glutathione conjugates to reactive intermediates that interact with DNA and thereby cause mutagenicity.

Pharmacokinetics and short-term clinicopathologic changes after intravenous administration of a high dose of methimazole in dogs.

A bolus dose of methimazole (MMI) was administered IV over 1 minute to 5 healthy adult dogs at a dosage (40 mg/kg of body weight) known to impart protection against cisplatin-induced renal disease. Blood and urine samples for pharmacokinetic analysis were collected over a 24-hour period. Physical examination, CBC, determination of serum thyroid hormone concentrations, and serum biochemistry analysis were performed over a 10-day period to evaluate short-term toxicoses. At this dosage, MMI appears to be safe and well tolerated in dogs; only 1 of the 5 dogs had mild and transient increases in serum activity of hepatic enzymes. In addition, MMI did not alter serum thyroid hormone concentrations. Half-life of 8.82 hours and mean residence time of 12.18 hours were determined for MMI. Renal clearance of native MMI, along with sulfate and glucuronide conjugates, represented only 20% of total systemic clearance. Results of this study provide further information concerning clinical use of MMI in dogs and may contribute to better understanding of the mechanism of MMI protection against chemically induced nephrotoxicosis.

Determination of p-nitrophenol hydroxylase activity of rat liver microsomes by high-pressure liquid chromatography.

p-Nitrophenol hydroxylation to p-nitrocatechol is a useful metabolic marker for the presence of functional cytochrome P450 2E1 in mammalian cell microsomes, but the assay is limited by the sensitivity of the spectrophotometric method used to monitor p-nitrocatechol formation. In this paper, a reverse-phase high-pressure liquid chromatography method, which is nearly 20 times more sensitive than the spectrophotometric method and more specific for p-nitrocatechol determination, is described. The method involves monitoring the presence of p-nitrocatechol in the trifluoroacetic acid-quenched reaction mixtures at 345 nm. The utility of the method was demonstrated with rat liver microsomes, where p-nitrocatechol formation was found to be NADPH dependent, was linear with incubation times (2.5 to 30.0 min) and protein concentrations (0.03-0.48 mg/incubation), and exhibited typical Michaelis-Menton kinetics (Km = 197 microM, Vmax = 2.8 nmol/mg protein/min).

Detection and mechanisms of formation of S-(6-purinyl)glutathione and 6-mercaptopurine in rats given 6-chloropurine.

6-Chloropurine (CP) has antitumor activity against animal and human neoplasms, but the mechanism is unclear. Recently, we have shown that S-(6-purinyl)glutathione (PG), a putative metabolite of CP, is metabolized in vivo to yield the antitumor drug, 6-mercaptopurine (6-MP). In this study, CP metabolism to PG and 6-MP was investigated in an effort to provide further insights into the mechanism of CP antitumor activity. Rat hepatic and renal glutathione S-transferases metabolized CP to PG; Vmax values for liver and kidney cytosol were 166 and 24 nmol/mg of protein/min, respectively. PG was isolated and characterized by fast atom bombardment mass spectrometry from the bile of rats given CP. When rats were given CP (14 mumol/kg), PG excretion was linear with time for up to 1 hr; nearly 80% of the PG excreted at 2 hr was excreted at 1 hr. Rats given CP (10-1200 mumol/kg) excreted at 1 hr into bile nearly 18% of the dose as PG; rats given CP (400-1200 mumol/kg) excreted at 24 hr into urine nearly 4% of the dose as 6-MP and its further metabolites, 6-methylthiopurine and 6-thiouric acid. CP, PG, 6-MP, 6-methylthiopurine and 6-thiouric acid were also detected in plasma, liver and kidney of rats given CP (1200 mumol/kg); in these tissues, maximum CP concentrations were observed at 30 min, as compared to 60 to 180 min, and plasma CP concentrations were higher than those detected in liver or kidney. Liver or kidney CP metabolite concentrations at 30 to 120 min were, however, higher than those detected in plasma.(ABSTRACT TRUNCATED AT 250 WORDS)

Chloroperoxidase-mediated oxidation of 1,3-butadiene to 3-butenal, a crotonaldehyde precursor.

Previously, we have shown that 1,3-butadiene, a rodent and possibly a human carcinogen, can be oxidized by chloroperoxidase-H2O2 at pH 7.4 to yield the potent mutagens, butadiene monoxide and crotonaldehyde. Because the crotonaldehyde/butadiene monoxide ratio from reactions with chloroperoxidase was higher than that obtained from reactions with myeloperoxidase or cytochrome P450 enzymes, in the present study, the chloroperoxidase reaction was further investigated in an attempt to define optimal conditions for catalysis and to possibly obtain direct evidence for the formation of the crotonaldehyde precursor 3-butenal in these incubations. The results showed that butadiene monoxide and crotonaldehyde formation was optimal at pH 6.0. As opposed to incubations carried out at pH 7.4, GC analyses of incubations carried out at pH 4.5, 5.0, and 6.0 demonstrated the presence of a new peak which had a retention time different from that of butadiene monoxide and crotonaldehyde. The new peak was identified as 3-butenal by comparison of its retention time and mass spectrum with those of reference material. Evidence for 3-butenal being a precursor of crotonaldehyde was obtained by the findings that 3-butenal was not simply a decomposition product of butadiene monoxide or crotonaldehyde under the incubation or assay conditions, and that the 3-butenal/crotonaldehyde ratio decreased when the incubation time was increased between 5 and 30 min or when the incubation temperature was increased between 10 and 45 degrees C. The combined 3-butenal and crotonaldehyde concentrations remained constant at the various incubation temperatures. Furthermore, 3-butenal conversion to crotonaldehyde was faster at pH 7.4, compared to pH 6.0.(ABSTRACT TRUNCATED AT 250 WORDS)

Reactivity of cysteine S-conjugate sulfoxides: formation of S-[1-chloro-2-(S-glutathionyl)vinyl]-L-cysteine sulfoxide by the reaction of S-(1,2-dichlorovinyl)-L-cysteine sulfoxide with glutathione.

S-(1,2-Dichlorovinyl)-L-cysteine (DCVC) sulfoxide, a putative metabolite of the toxic cysteine S-conjugate DCVC, was synthesized by the reaction of DCVC with H2O2 and characterized by fast atom bombardment mass spectrometry (FAB-MS) and proton nuclear magnetic resonance spectroscopy. DCVC sulfoxide was stable when kept at room temperature overnight in phosphate buffer (pH 6.8-7.8) or when heated in phosphate buffer (pH 7.2 or 7.6) or H2O (pH 3.5 or 10.5) for 20 min at 37 degrees C. However, in the presence of glutathione (GSH), DCVC sulfoxide was readily converted to S-[1-chloro-2-(S-glutathionyl)vinyl]-L-cysteine sulfoxide (I), a product formed by the Michael addition of GSH to DCVC sulfoxide followed by the loss of HCl. Evidence for the mechanism of this reaction was obtained by the finding that DCVC, which cannot act as a Michael acceptor, did not react with GSH under conditions similar to those used with DCVC sulfoxide. When the reaction of DCVC sulfoxide with GSH was carried out at room temperature and pH 7.4, formation of I was complete at 5 min, but when the reaction was carried out for 2 h at pH 6.0 or 4.4 at 37 degrees C, product formation was nearly 37 or 3% of that formed at pH 7.4, respectively; product formation did not increase when the reaction was carried out at pH 8.5. When DCVC sulfoxide (100 mg/kg) was administered to rats, hepatic and renal reduced nonprotein thiol concentrations were decreased at 1 h to 74 and 27% of that in control rats, respectively.(ABSTRACT TRUNCATED AT 250 WORDS)

Human liver microsomes are efficient catalysts of 1,3-butadiene oxidation: evidence for major roles by cytochromes P450 2A6 and 2E1.

Previously, we provided evidence for the involvement of multiple cytochrome P450 enzymes in the metabolism of 1,3-butadiene, a rodent and possibly a human carcinogen, to butadiene monoxide in mouse and rat liver microsomes. In this study, 1,3-butadiene oxidation by seven cDNA-expressed human P450 enzymes and by human, mouse, and rat liver microsomes was characterized. Incubations with cDNA-expressed human P450 1A2, 2A6, 2B6, 2D6, and 2E1 resulted in the formation of butadiene monoxide, whereas incubations with P450 1A1 and 3A4 did not lead to the detection of any metabolite. Of the active isozymes examined, P450 2A6 and 2E1 were the most active when butadiene monoxide formation rates were normalized for the P450 content of the microsomes. With six human liver microsomal samples, 1,3-butadiene oxidation exhibited nearly threefold individual variation in the amounts of butadiene monoxide detected, and butadiene monoxide formation was NADPH- and time-dependent and was inhibited by the addition of 1-benzyl-imidazole or 4-methylpyrazole, known cytochrome P450 inhibitors. Correlation studies provided evidence for major roles by P450 2A6 and 2E1 in 1,3-butadiene oxidation in human liver microsomes. Butadiene monoxide formation rates in human liver microsomes were similar, or higher, than the rate obtained in mouse liver microsomes, whereas 1,3-butadiene oxidation rates in human and mouse liver microsomes were higher than the rate obtained in rat liver microsomes. These results provide direct evidence that 1,3-butadiene is a substrate for multiple P450 enzymes and suggest that humans may be at higher risk of expressing 1,3-butadiene toxicity compared to mice or rats. In addition, these results suggest that the mouse may be the more appropriate animal model to assess human risk.

Glutathione-dependent metabolism of cis-3-(9H-purin-6-ylthio)acrylic acid to yield the chemotherapeutic drug 6-mercaptopurine: evidence for two distinct mechanisms in rats.

cis-3-(9H-Purin-6-ylthio)acrylic acid (PTA) is a structural analog of azathioprine, a prodrug of the antitumor and immunosuppressive drug 6-mercaptopurine (6-MP). In this study, we examined the in vitro and in vivo metabolism of PTA in rats. Two metabolites of PTA, 6-MP and the major metabolite, S-(9H-purin-6-yl)glutathione (PG), were formed in a time- and GSH-dependent manner in vitro. Formation of 6-MP and PG occurred nonenzymatically, but 6-MP formation was enhanced 2- and 7-fold by the addition of liver and kidney homogenates, respectively. Purified rat liver glutathione S-transferases enhanced 6-MP formation from PTA by 1.8-fold, whereas human recombinant alpha, mu, and pi isozymes enhanced 6-MP formation by 1.7-, 1.3-, and 1.3-fold, respectively. In kidney homogenate incubations, PG accumulation was only observed during the first 15 min because of further metabolism by gamma-glutamyltranspeptidase, dipeptidase, and beta-lyase to yield 6-MP, as indicated by the use of the inhibitors acivicin and aminooxyacetic acid. Based on these results and other lines of evidence, two different GSH-dependent pathways are proposed for 6-MP formation: an indirect pathway involving PG formation and further metabolism to 6-MP, and a direct pathway in which PTA acts as a Michael acceptor. HPLC analyses of urine of rats treated i.p. with PTA (100 mg/kg) showed that 6-MP was formed in vivo and excreted in urine without apparent liver or kidney toxicity. Collectively, these studies show that PTA is metabolized to 6-MP both in vitro and in vivo and may therefore be a useful prodrug of 6-MP.