Mechanistic Studies on the Metabolic Scission of Thiazolidinedione

880

Chem. Res. Toxicol. 2005, 18, 880-888

Mechanistic Studies on the Metabolic Scission of Thiazolidinedione Derivatives to Acyclic Thiols Vijay Bhasker G. Reddy,* Bindhu V. Karanam, Wendy L. Gruber, Michael A. Wallace, Stella H. Vincent, Ronald B. Franklin, and Thomas A. Baillie Department of Drug Metabolism, Merck Research Laboratories, P.O. Box 2000, Rahway, New Jersey 07065 Received February 10, 2005

Thiazolidinedione (TZD) derivatives have been reported to undergo metabolic activation of the TZD ring to produce reactive intermediates. In the case of troglitazone, it was proposed that a P450-mediated S-oxidation leads to TZD ring scission and the formation of a sulfenic acid intermediate, which may be trapped as a GSH conjugate. In the present study, we employed a model compound {denoted MRL-A, (()-5-[(2,4-dioxothiazolidin-5-yl)methyl]-2-methoxy-N[[(4-trifluoromethoxy)phenyl]methyl]benzamide} to investigate the mechanism of TZD ring scission. When MRL-A was incubated with monkey liver microsomes (or recombinant P450 3A4 and NADPH-P450 reductase) in the presence of NADPH and oxygen, the major products of TZD ring scission were the free thiol metabolite (M2) and its dimer (M3). Furthermore, a GSH conjugate of M2 (M4) also was formed when the incubation mixture was supplemented with GSH. Experiments with isolated M2 suggested that this metabolite was unstable and underwent spontaneous autooxidation to M3. A qualitatively similar metabolite profile was observed when MRL-A was incubated with recombinant P450 3A4 and cumene hydroperoxide. Because an oxygen atom is transferred to MRL-A under these conditions, these data suggested that S-oxidation alone may result in TZD ring scission and formation of M2 via a sulfenic acid intermediate. Also, because the latter incubation mixture did not contain any reducing agents, the formation of M2 may have occurred due to disproportionation of the sulfenic acid. When NADPH was added to the incubation mixture containing P450 3A4 and cumene hydroperoxide, the formation of M3 increased, suggesting that the sulfenic acid was reduced to M2 by NADPH and subsequently underwent dimerization to yield M3 (vide supra). When NADPH was replaced by GSH, the formation of M4 increased, consistent with reduction of the sulfenic acid by GSH. In summary, these results suggest that the TZD ring in MRL-A is activated by an initial P450mediated S-oxidation step followed by spontaneous scission of the TZD ring to a putative sulfenic acid intermediate; the latter species then undergoes reduction to the free thiol by GSH, NADPH, and/or disproportionation. Finally, the thiol may dimerize to the corresponding disulfide or, in the presence of S-adenosylmethionine, form the stable S-methyl derivative.

Introduction drugs,1

Thiazolidinedione (TZD)-containing also known as glitazones, are used widely in the treatment of type-2 diabetes. These drugs bind to peroxisome proliferatoractivated receptor γ, resulting in increased expression of genes encoding proteins that are involved in glucose and lipid metabolism (1). In vitro and in vivo studies of glitazone derivatives have demonstrated that the TZD ring may be subject to metabolic attack, although this pathway appears to be highly compound-dependent (2-15). For example, the metabolism of pioglitazone mainly involves hydroxylation of the ethyl side chain followed by glucuronidation, sulfation, or taurine conjugation (2-4). Similarly, the * To whom correspondence should be addressed. Tel: 732-594-7868. Fax: 732-594-1416. E-mail: [email protected]. 1 Abbreviations: MRL-A, (()-5-[(2,4-dioxothiazolidin-5-yl)methyl]2-methoxy-N-[[(4-trifluoromethoxy)phenyl]methyl]benzamide; TZD, thiazolidinedione; CID, collision-induced dissociation; SAM, S-adenosylmethionine; NAC, N-acetylcysteine; BME, β-mercaptoethanol.

major metabolism pathway of troglitazone, which was withdrawn from the market due to liver toxicity (1618), involves oxidation of the chroman moiety to a quinone followed by reduction and sulfation (6-9). In both cases, TZD ring scission was reported as a minor pathway of biotransformation (4, 9). However, in the case of troglitazone, interest in the TZD ring cleavage pathway stemmed from the observation that this route of metabolism led to several reactive intermediates that were trapped as GSH conjugates (8-11). More recently, it has been reported that the major in vitro biotransformation pathway of MK-0767 (Scheme 1) involves scission of the TZD ring, resulting in a thiol metabolite, which, in turn, undergoes S-methylation and further oxidation to methyl sulfoxide and methyl sulfone metabolites (13, 14). Similarly, the trifluoromethoxy analogue of MK-0767, designated (()-5-[(2,4-dioxothiazolidin-5-yl)methyl]-2-methoxyN-[[(4-trifluoromethoxy)phenyl]methyl]benzamide (MRLA) (Scheme 1), was shown in preliminary studies in our laboratories to undergo extensive TZD ring cleavage.

10.1021/tx0500373 CCC: $30.25 © 2005 American Chemical Society Published on Web 04/14/2005

Metabolic Scission of Thiazolidinedione Derivatives Scheme 1. Structures of MK-0767, MRL-A, and Metabolites M1-M6

The objective of the present study was to investigate the mechanism of TZD ring scission leading to the formation of the thiol metabolite (M2, Scheme 1) using MRL-A as a model compound. Theoretically, cleavage of the TZD ring could involve oxidative, reductive, or hydrolytic processes (or a combination thereof). In studies on the bioactivation of troglitazone, Kassahun et al. (9) proposed that P450 3A-mediated oxidation of the sulfur atom of the TZD ring afforded a reactive sulfoxide intermediate, which underwent spontaneous ring opening to a highly electrophilic R-keto isocyanate derivative, which was trapped in the form of a GSH conjugate (9). Although the thiol metabolite of troglitazone (corresponding to M2 in Scheme 1) was not identified in these studies, a GSH conjugate of M2 was detected (9). Presumably, therefore, initial oxidative cleavage of the TZD ring was followed by a reductive process in order to generate the thiol. In the present communication, we report on the results of a series of in vitro studies to explore mechanistic aspects of the TZD ring cleavage, which support the operation of an initial S-oxidation step. Subsequent spontaneous ring opening and reduction of the putative sulfenic acid intermediate to the thiol metabolite then occurs and may be mediated by GSH, NADPH, and/or disproportionation.

Materials and Methods Materials. MRL-A and [14C]MRL-A were synthesized using trifluromethoxy benzylamine, instead of trifluromethyl benzylamine, following a similar procedure reported for the synthesis of MK-0767 (19). The specific activity of [14C]MRL-A was 55 mCi/ mmol, and the radiochemical purity, as determined by HPLC, was 99.6%. LC-MS analysis of [14C]MRL-A indicated the presence of MH+ ion at m/z 457, collision-induced dissociation (CID) of which yielded prominent ions at m/z 266 and 283, similar to that reported for MK-0767 (14). Acetonitrile and methanol (HPLC grade) were purchased from Fisher Scientific (Fair Lawn, NJ). L-R-Dilauroylphosphatidyl choline, S-adenosylmethionine (SAM), cumene hydroperoxide, NADPH, ammonium acetate, and GSH were obtained from Sigma (St. Louis, MO). Rat, dog, monkey, and human liver microsomes were prepared using literature procedures. Recombinant P450 3A4 and NADPHP450 reductase, provided by Dr. Magang Shou (Merck Research Laboratories, Department of Drug Metabolism, West Point, PA), were prepared as described previously (20-22). Incubation of [14C]MRL-A with Liver Microsomes and NADPH. [14C]MRL-A (4 µM, ∼0.12 µCi, in acetonitrile) was incubated with rat, dog, monkey, or human hepatic microsomes at a protein concentration of 2 mg/mL in 0.1 M KH2PO4 (pH 7.4) buffer containing 0.1 mM MgCl2, in the presence or absence of 1 mM GSH, N-acetylcysteine (NAC), dithiothreitol (DTT), or β-mercaptoethanol (BME). The final volume was 500 µL, and the final concentration of acetonitrile in the assay mixture was 1% (by volume). Reactions were initiated by the addition of 1

Chem. Res. Toxicol., Vol. 18, No. 5, 2005 881 mM NADPH, and incubations proceeded for 30-60 min at 37 °C in a shaking water bath. For anaerobic experiments, reaction mixtures were evacuated under reduced pressure and flushed three times with argon to ensure removal of oxygen. The NADPH solution also was evacuated and purged with argon, separately, before addition to the incubation mixture. The reactions were stopped by the addition of half of the incubation volume of acetonitrile and chilled on melting ice. Following centrifugation at 10000g at 4 °C for 10 min, 50 µL aliquots of the supernatant were taken for analysis by LC-MS/MS (vide infra). Isolation of Thiol Metabolite (M2) and Thiol Dimer (M3). Incubations of [14C]MRL-A with monkey liver microsomes and NADPH were performed as described above in the presence of GSH. The final incubation volume was 2 mL. Following 60 min of incubation at 37 °C, reactions were stopped by the addition of 1 mL of acetonitrile and chilled on melting ice. Following centrifugation, 100 µL aliquots of the supernatant were subjected to preparative HPLC (vide infra). The fractions corresponding to metabolites M2 and M3 were collected and evaporated to dryness under a stream of nitrogen. Reduction of the Thiol Dimer M3. The isolated M3 metabolite (∼56000 dpm, 0.22 µg; Scheme 1) in 100 mM phosphate buffer (0.1 mL, pH 7.4) was incubated with 0.5 mM GSH for 20 min at 37 °C. At this point, 50 µL of acetonitrile was added and an aliquot of the resulting mixture was analyzed by LC-MS/MS. Incubation of [14C]MRL-A with Monkey Liver Microsomes and Cumene Hydroperoxide. Incubation mixtures contained 500 µL of 0.1 M potassium phosphate (pH 7.4), 0.1 mM MgCl2, 4 µM [14C]MRL-A, and 2 mg/mL microsomal protein in the presence or absence of 1 mM GSH, NAC, DTT, or BME. Reactions were initiated by the addition of 500 µM cumene hydroperoxide in water, and incubations proceeded for 2-30 min at 37 °C in a shaking water bath. The reactions were terminated by the addition of acetonitrile and analyzed by LC-MS/MS. Incubation of [14C]MRL-A with Recombinant P450 3A4, NADPH-P450 Reductase, and NADPH. Incubation mixtures contained 250 µL of 0.1 M potassium phosphate (pH 7.4), 0.1 mM MgCl2, 4 µM [14C]MRL-A, 100 pmol of recombinant P450 3A4, 200 pmol of NADPH-P450 reductase, and 5 µg of L-Rdilauroylphosphatidyl choline, with or without 1 mM GSH. Reactions were initiated by the addition of 1 mM NADPH solution, and incubations were performed for 30-60 min at 37 °C in a shaking water bath. The reactions were stopped as described above, and the products were analyzed by LC-MS/ MS. Incubation of [14C]MRL-A with Recombinant P450 3A4 and Cumene Hydroperoxide. Incubation mixtures contained 250 µL of 0.1 M potassium phosphate (pH 7.4), 0.1 mM MgCl2, 4 µM [14C]MRL-A, 100 pmol of recombinant P450 3A4, and 5 µg of L-R-dilauroylphosphatidyl choline, with or without 1 mM NADPH and/or GSH. Reactions were initiated by the addition of 500 µM cumene hydroperoxide. The samples were incubated for 2-30 min at 37 °C in a shaking water bath, and reactions were stopped as described previously and analyzed by LC-MS/ MS. Incubation of [14C]MRL-A with Monkey Liver Microsomes, NADPH, GSH, and SAM. Incubation mixtures contained 4 µM [14C]MRL-A, 2 mg/mL microsomal protein, 100 µM SAM, and 0.1 mM MgCl2 in 500 µL of 0.1 M potassium phosphate (pH 7.4), with or without 1 mM GSH. The reactions were initiated by the addition of 1 mM NADPH, and incubations were carried out for 2-30 min at 37 °C in a shaking water bath. The reactions were terminated as described above and analyzed by LC-MS/MS. Instrumentation. Metabolites were identified by electrospray LC-MS/MS analysis using a Finnigan LCQ Deca mass spectrometer (San Jose, CA), which was interfaced to a Shimadzu HPLC system (Columbia, MD) equipped with two Series LC-10ADVP micropumps and a Series SIL-10ADVP autosampler. The spray voltage was maintained at 4.1 kV, and the

882

Chem. Res. Toxicol., Vol. 18, No. 5, 2005

Figure 1. Radiochromatographic profiles of products formed upon metabolism of [14C]MRL-A in monkey liver microsomes. [14C]L-MRL-A (4 µM) was incubated with liver microsomes and NADPH in the absence (A) and presence (B) of GSH (1 mM). Incubation mixtures were extracted with acetonitrile and analyzed by HPLC with radiodetection as described in the text. (C) Reanalysis of metabolite M2 isolated from the assays performed under conditions B (depicting the spontaneous oxidation of M2 to M3). (D) Reduction of M3 to M4 and M2 by GSH. capillary temperature was set at 250 °C. Full scan spectra, from m/z 400 to 1000, were obtained in the positive ion mode, and product ion spectra were generated by CID of the MH+ ions of interest. Separation of metabolites was achieved on a phenomenex Luna 5 µm phenyl-hexyl column (4.6 mm × 150 mm; Torrance, CA) at a flow rate of 1 mL/min. The mobile phase consisted of water containing 0.1% formic acid (A) and acetonitrile/methanol (1:1, v/v; B). The column was eluted with a linear gradient from 40 to 70% B over 30 min followed by a wash with 90% B for 15 min. One-quarter of the column eluate was directed into the mass spectrometer, while the remainder was passed into a Packard radiometric detector (Downers Grove, IL) for on-line radioprofiling.

Results Metabolism of [14C]MRL-A in NADPH-Fortified Liver Microsomes. Incubation of [14C]MRL-A with liver microsomes from rats, dogs, monkeys, and humans in the presence of NADPH yielded qualitatively similar metabolite profiles. A representative radiochromatogram depicting metabolites of [14C]MRL-A formed upon incubation with monkey liver microsomes is shown in Figure 1A. LC-MS analysis indicated the presence of MH+ ions at m/z 473, 431, 859, and 736 for M1, M2, M3, and M4, respectively. In the case of M1, whose molecular mass corresponded to the addition of 16 Da to the parent compound, the CID of the MH+ ion at m/z 473 (M1) yielded a prominent fragment ion at m/z 282, resulting from the neutral loss of 191 Da (Figure 2), indicating

Reddy et al.

addition of oxygen on the benzyl-TZD region of the molecule. The MS/MS spectrum also exhibited a fragment ion at m/z 455 due to loss of the elements of water from the MH+ species. The fragment ion at m/z 222 was consistent with the loss of 60 Da (SCO) from the precursor at m/z 282, as shown in Figure 2. On the basis of these data and by comparison with the MS/MS fragmentation of a similar metabolite of MK-0767 (14), metabolite M1 was identified as a hydroxy-TZD derivative, as shown in Scheme 1. Further experiments with this purified metabolite suggested that M1 was not involved as an intermediate in the process leading to TZD ring opening (data not shown); hence, M1 will not be discussed further. The MH+ ion at m/z 431 (M2) indicated that this metabolite was formed by net loss of 26 Da from the parent compound. The CID of m/z 431 yielded an abundant fragment ion at m/z 240, resulting from the same amide bond cleavage observed with M1, associated with the neutral loss of 191 Da (Figure 3). The fragment ion at m/z 195 is likely due to the elimination of the elements of formamide from the ion at m/z 240, while the fragment at m/z 414 corresponded to the loss of NH3 from the MH+ species. On the basis of this fragmentation pattern and by comparison of this CID spectrum with that reported for a similar metabolite of MK-0767 (14), M2 was assigned the thiol structure shown in Figure 3. Metabolite M3, whose MH+ ion was present at m/z 859, represented the dimer of M2. The CID of m/z 859 yielded fragment ions at m/z 668 (MH+ - 191) and m/z 842 (MH+ - NH3), as shown in Figure 4. The ion with m/z 668 underwent further fragmentation by loss of 191 Da (to give m/z 477), loss of CO (to give m/z 640), and loss of formamide (to give m/z 623). The fragment ion at m/z 431 is the result of reductive scission of the disulfide bond. The CID of the MH+ ion at m/z 736 (M4) yielded fragment ions with m/z 607, corresponding to the neutral loss of 129 Da (pyroglutamate), and m/z 661, resulting from loss of 75 Da (glycine), as shown in Figure 5; these neutral losses are characteristic of GSH conjugates (23, 24). In addition, the CID spectrum exhibited an ion at m/z 416, likely due to the combined loss of 191 and 129 Da. The neutral loss of 191 Da suggested that the lefthand portion of the compound, as depicted in Figure 5, remained intact. On the basis of these data, the structure of M4 was assigned as the mixed disulfide GSH conjugate of thiol metabolite M2, as shown in Figure 5. Further evidence for the presence of a disulfide bond in metabolites M3 and M4 is provided below. When [14C]MRL-A was incubated with monkey liver microsomes and NADPH in the presence of GSH, the relative proportions of M2 and M4 increased, while that of the dimer M3 decreased (Figure 1B), as compared to the incubation conducted in the absence of GSH (Figure 1A). Finally, when [14C]MRL-A was incubated with monkey liver microsomes and NADPH under anaerobic conditions, in the presence or absence of GSH, no turnover of the parent compound was detected (data not shown). Autooxidation of the Thiol Metabolite (M2). Figure 1C depicts a radiochromatogram obtained upon reanalysis of purified metabolite M2 (isolated from the incubation of [14C]MRL-A with monkey liver microsomes, NADPH, and GSH). The chromatogram showed a single peak, which, surprisingly, did not correspond to M2 but to M3, suggesting that M2 was susceptible to autooxi-

Metabolic Scission of Thiazolidinedione Derivatives

Chem. Res. Toxicol., Vol. 18, No. 5, 2005 883

Figure 2. Product ion spectrum obtained by CID of the MH+ ion of metabolite M1 at m/z 473.


dation to the corresponding disulfide M3 during sample handling. Reduction of the Thiol Dimer (M3) by GSH. Figure 1D depicts a radiochromatogram of the products (M2 and M4) formed upon treatment of the disulfide M3 with GSH, further supporting the dimeric nature of M3. Metabolism of [14C]MRL-A by Recombinant P450 3A4, NADPH-P450 Reductase and NADPH. The radiochromatograms depicted in Figure 6 illustrate the metabolite profiles obtained upon incubation of [14C]MRL-A with recombinant P450 3A4, NADPH-P450 reductase, and NADPH, in the presence or absence of GSH. In the absence of GSH, metabolites M2 and M3 were formed (panel A), while in the presence of GSH, M4 was

formed in addition to M2 and a trace of M3 (panel B). The hydroxylated product, M1, was formed in both incubations. These metabolite profiles were similar to those obtained upon incubation of [14C]MRL-A with liver microsomes and NADPH in the presence or absence of GSH. Metabolism of [14C]MRL-A by P450 3A4 and Cumene Hydroperoxide. Metabolites profiles obtained upon incubation of [14C]MRL-A with recombinant P450 3A4 in the presence of cumene hydroperoxide were qualitatively similar to those produced in the presence of monkey liver microsomes and NADPH and included a small amount of thiol metabolite (M2) and the dimer M3 (Figure 7A). When NADPH was added to the incubation

884


Reddy et al.



mixture containing recombinant P450 3A4 and cumene hydroperoxide, the relative amounts of thiol dimer (M3) increased, as shown in panel B. Finally, when GSH was added to the incubation, in place of NADPH, the major metabolites detected were M2 and its GSH conjugate, M4 (panel C), and only a small amount of M3 was formed. The hydroxylated metabolite M1 was present in all postincubation mixtures. Metabolism of [14C]MRL-A by Liver Microsomes, NADPH, GSH, and SAM. Incubation of [14C]MRL-A with monkey liver microsomes in the presence of NADPH, GSH, and SAM yielded metabolites M5 (methylated

thiol) and M6 (methyl sulfoxide), in addition to M2. However, the thiol dimer M3 was not detected in these incubations (data not shown). LC-MS analysis of M5 indicated the MH+ ion at m/z 445, the CID of which produced fragment ions at m/z 254 (neutral loss of 191 Da), m/z 400 (loss of formamide), and m/z 428 (loss of NH3). On the basis of these data and by comparison with the MS/MS spectrum of the corresponding metabolite of MK-0767 (14), the structure assigned to M5 was the TZD ring-cleaved methyl sulfide derivative shown in Scheme 1. LC-MS analysis of M6 indicated the MH+ ion at m/z 461, which is 16 Da higher than that of M5 metabolite.


Figure 6. Radiochromatographic profiles of products formed upon metabolism of [14C]MRL-A by recombinant P450 3A4. [14C]MRL-A (4 µM) was incubated with P450 3A4, NADPH-P450 reductase, and NADPH in the absence (A) and presence (B) of GSH. Assay mixtures were extracted with acetonitrile and analyzed by HPLC with radiodetection as described in the text.

Figure 7. Radiochromatographic profiles of products formed upon metabolism of [14C]MRL-A by recombinant P450 3A4. [14C]MRL-A (4 µM) was incubated with P450 3A4 and cumene hydroperoxide (A) and in the presence of NADPH (B) or GSH (C). Assay mixtures were extracted with acetonitrile and analyzed by HPLC with radiodetection as described in the text.

The proposed methyl sulfoxide structure for this metabolite is shown in Scheme 1. The CID of the MH+ ion at m/z 461 produced fragments at m/z 397 (loss of the methylsulfoxide moiety), m/z 380 (additional loss of NH3), and m/z 206 (loss of 191 Da and methylsulfoxide). These data were in good agreement with the MS/MS characteristics of the corresponding metabolite formed from MK-0767 (14).

Discussion The process by which TZD-containing compounds undergo metabolism to products of TZD ring scission has been reported to involve the formation of chemically reactive intermediates, which may be trapped as conjugates with GSH (8, 9). In the case of MK-0767, TZD ring scission results in the formation of a free thiol, which


undergoes facile S-methylation and oxidation (13, 14). The purpose of the present study was to gain an understanding of the mechanism of TZD ring cleavage and formation of the free thiol derivative and employed MRL-A, a close analogue of MK-0767, as a model substrate (Scheme 1). The metabolism of [14C]MRL-A proved to be closely similar to that of MK-0767 in rat, dog, monkey, and human liver microsomes and produced metabolites largely by way of TZD ring scission. Because the turnover of [14C]MRL-A was comparatively higher in monkey liver microsomes than in corresponding preparations from other species, monkey liver microsomes were used as the enzyme source in the present study. When [14C]MRL-A was incubated with monkey liver microsomes under anaerobic conditions, no turnover was observed (data not shown), suggesting that an oxidative process rather than a hydrolytic or reductive pathway was involved in TZD ring opening, with the sulfur atom representing the most likely site for oxidation. Indeed, the oxidation of the sulfur of TZD, thiazole, and thiophene ring systems has been proposed as the initial step in the bioactivation of troglitazone, L-766,112, and tienilic acid, respectively (9, 25, 26). In the case of MRL-A, oxidation of the sulfur likely would result in spontaneous scission of the TZD ring, producing a reactive R-keto isocyanate derivative, as shown in Scheme 4. Hydrolysis of the isocyanate and nucleophilic attack by GSH on the sulfenic acid moiety would account for the formation of the mixed disulfide conjugate M4. This mechanism of oxidative opening of the TZD ring is same as that proposed for troglitazone (9). While the mixed disulfide GSH conjugate corresponding to M4 was identified as a metabolite of troglitazone (9), GSH adducts similar to those derived from the R-keto isocyanate intermediate of troglitazone were not detected in the present study with MRL-A. Sulfenic acids are known to react with protein thiols and reduced GSH. Thus, the diuretic thiosteroid, spironolactone, was oxidized to spironolactone sulfenic acid by rat liver microsomes and trapped as a glutathionyl-spironolactone disulfide (27, 28). In the present studies, it was demonstrated that the mixed disulfide conjugate (M4) was reduced efficiently to the free thiol (M2) in the presence of GSH, and this provided a convenient explanation for the generation of M2 from MRL-A in liver microsomal preparations fortified with GSH. However, it was noted that M2 was formed in the in vitro preparations even in the absence of added GSH (Figure 1A). Moreover, trace amounts of the mixed disulfide conjugate (M4) also were generated in these experiments, suggesting that the microsomal incubations were not entirely “GSH-free”. It is interesting that a dimer of the thiol metabolite (M3) also was formed, in addition to M2 and M4, when [14C]MRL-A was incubated with liver microsomes. As expected, levels of M3 were much lower when GSH was added to the incubation mixtures (Figure 1A), reflecting the ability of GSH to reduce the disulfide bond in M3 according to the two-step mechanism outlined in Scheme 2. Separate experiments with purified M2 (the free thiol) demonstrated that this compound underwent facile autooxidation to regenerate M3, and thus, M2 and M3 may be considered to exist in an equilibrium, with the relative levels of the two metabolites being dictated by the redox state of the biological system in question. It is also possible that the thiol metabolite, M2, could be oxidized by P450 or FMO in microsomal incubations to a thiyl

886


Scheme 2. Mechanism of Reduction of M3 and M4 to M2 by GSH

Reddy et al. Scheme 4. Proposed Mechanism of Formation of the Metabolites M2-M4 from MRL-A

Scheme 3. Proposed Mechanism of Disproportionation of the Sulfenic Acid Intermediate to Produce M2

radical, which, in turn, would form the dimer M3 (27). In addition, as expected, M3 was converted to M4 and, finally, to M2 when treated with GSH (Figure 1D and Scheme 2). However, the question remained as to whether M2 could be formed in the absence of GSH and, if so, by what mechanism? Because it was difficult to completely remove endogenous GSH from microsomal incubations, a recombinant P450 3A4 system was used to confirm unambiguously the role of GSH in the formation of M2 from [14C]MRL-A. The results showed that incubations of [14C]MRL-A with P450 3A4, NADPH-P450 reductase, and NADPH also generated both M2 and M3, suggesting that GSH was not required for the formation of M2. From a mechanistic standpoint, therefore, it seemed reasonable to conclude that following initial S-oxidation of the TZD ring, the formation of M2 must have proceeded via a reduction step, which was not necessarily dependent upon GSH. To investigate whether NADPH might serve as a source of reducing equivalent for this reaction, we exploited the “peroxide shunt pathway” of P450 enzymes. It is known that purified P450, free of electron carriers, catalyzes the hydroperoxide-dependent oxidation of a variety of substrates in the absence of NADPH, NADPH-P450 reductase, and molecular oxygen (29-32). Indeed, when incubated with liver microsomes in the absence of NADPH, [14C]MRL-A was found to undergo oxidation in the presence of added hydroperoxide, with cumene hydroperoxide serving as a better oxygen donor than tert-butyl hydroperoxide or hydrogen peroxide, to yield products similar to those formed with microsomes and NADPH (data not shown). Also, incubation of [14C]MRL-A with recombinant P450 in the presence of cumene hydroperoxide (but without GSH and NADPH) again resulted in the formation of both M2 and M3. To account for this somewhat unexpected observation, it was necessary to

consider the possibility of a disproportionation reaction involving the putative sulfenic acid intermediate of TZD ring scission. Generally, sulfenic acids are unstable and reactive, although a few sulfenic acids have been isolated in pure form (33). It has been reported that purine-6sulfenic acid, a metabolite of the antineoplastic agent 6-thiopurine, disproportionates to give 6-thiopurine and purine-6-sulfinic acid (34-36). By analogy, it is possible that the sulfenic acid intermediate derived from [14C]MRL-A also may undergo disproportionation to give M2 and a sulfinic acid, as shown in Scheme 3. The sulfinic acid thus formed would be expected to undergo further oxidation to a sulfonic acid, and indeed, this compound was identified by LC-MS/MS (data not shown). Moreover, incubation of [14C]MRL-A with P450 3A4 and cumene hydroperoxide in the presence of NADPH resulted in higher levels of M3 (Figure 7B). Because NADPH cannot transfer electrons to P450 enzymes in the absence of NADPH-P450 reductase, the latter observation suggested that NADPH may have reduced the sulfenic acid intermediate directly to M2, which subsequently dimerized to M3 (vide supra). Finally, incubation of [14C]MRL-A with P450 3A4 and cumene hydroperoxide in the presence of GSH produced M2 and M4, as expected, likely due to the chemical reaction of the sulfenic acid inter-


mediate with GSH. The ability of other thiol reagents to reduce the sulfenic acid intermediate also was investigated in liver microsomal incubations. NAC, DTT, and BME all were found to be effective in reducing the intermediate to M2, and a conjugate with BME (equivalent M4) also was identified (data not shown). It was reported that the corresponding free thiol metabolite of MK-0767 is further metabolized by methyl transferase to a methyl sulfide (13). Indeed, when [14C]MRL-A was incubated with liver microsomes in the presence of SAM and NADPH, the thiol M2 was metabolized effectively to the corresponding methyl sulfide M5, and autooxidation of M2 to M3 was not observed (data not shown). In conclusion, the results of this study have provided an insight into the mechanism by which TZD derivatives undergo metabolic ring scission to acyclic thiols and related products. Thus, initial P450-mediated S-oxidation appears to lead to an unstable TZD sulfoxide, which undergoes spontaneous cleavage to a reactive R-keto isocyanate (Scheme 4). Hydrolysis of the isocyanate moiety generates a sulfenic acid intermediate, which serves as the source of the free thiol (through reduction and/or disproportionation reactions), as well as a mixed disulfide conjugate with GSH. The free thiol, in turn, may dimerize or be further processed by S-methylation to stable end products. The toxicological significance of the reactive intermediates involved in the metabolic cleavage of TZD-containing compounds remains to be determined.

Acknowledgment. We thank the following people from Merck Research Laboratories: Drs. David C. Evans and George A. Doss for reviewing the manuscript and Dr. Conrad E. Raab for helpful discussions.


(10)

(11)

(12)

(13)

(14)

(15)

(16)

(17)

(18) (19)

(20)

References (1) Mudaliar, S., and Henry, R. R. (2001) New oral therapies for type 2 diabetes mellitus: The glitazones or insulin sensitizers. Annu. Rev. Med. 52, 239-257. (2) Krieter, P. A., Colletti, A. E., Doss, G. A., and Miller, R. R. (1994) Disposition and metabolism of the hypoglycemic agent pioglitazone in rats. Drug Metab. Dispos. 22, 625-630. (3) Kiyota, Y., Kondo, T., Maeshiba, Y., Hashimoto, A., Yamashita, K., Yoshimura, Y., Motohashi, M., and Tanayama, S. (1997) Studies on the metabolism of the new antidiabetic agent pioglitazone: Identification of metabolites in rats and dogs. Arzneim.Forsch./Drug Res. 47, 22-28. (4) Shen, Z., Reed, J. R., Creighton, M., Liu, D. Q., Tand, Y. S., Hora, D. F., Feeney, W., Szewczyk, J., Bakhtiar, R., Franklin, R. B., and Vincent, S. H. (2003) Identification of novel metabolites of pioglitazone in rat and dog. Xenobiotica 5, 499-509. (5) Bolton, G. C., Keogh, J. P., East, P. D., Hollis, F. J., and Shore, A. D. (1996) The fate of a thiazolidinedione antidiabetic agent in rat and dog. Xenobiotica 26, 627-636. (6) Kawai, K., Kawasaki-Tokui, Y., Odaka, T., Tsuruta, F., Kazui, M., Iwabuchi, H., Nakamura, T., Kinoshita, T., Ikeda, T., Yoshioka, T., Komai, T., and Nakamura, K. (1997) Disposition and metabolism of the new oral antidiabetic drug troglitazone in rats, mice and dogs. Arzneim.-Forsch./Drug Res. 47, 356-368. (7) Yamazaki, H., Shibata, A., Suzuki, M., Nakajima, M., Shimada, N., Geungerich, F. P., and Yokoi, T. (1999) Oxidation of troglitazone to a quinone-type metabolite catalyzed by cytochrome P-450 2C8 and P-450 3A4 in human liver microsomes. Drug Metab. Dispos. 27, 1260-1266. (8) Tettey, J. N., Maggs, J. L., Rapeport, W. G., Pirmohamed, M., and Park B. K. (2001) Enzyme-induction dependent bioactivation of troglitazone and troglitazone quinone in vivo. Chem. Res. Toxicol. 14, 965-974. (9) Kassahun, K., Pearson, P. G., Tang, W., McIntosh, I., Leung, K., Elmore, C., Dean, D., Wang, R., Doss, G., and Baillie, T. A. (2001) Studies on the metabolism of troglitazone to reactive intermediates in vitro and in vivo. Evidence for novel biotransformation

(21)

(22)

(23)

(24)

(25)

(26)

(27)

(28)

pathways involving quinone methide formation and thiazolidinedione ring scission. Chem. Res. Toxicol. 14, 62-70. Prabhu, S., Fackett, A., Lloyd, S., McClellan, H. A., Terrell, C. M., Silber, P. M., and Li, A. P. (2002) Identification of glutathione conjugates of troglitazone in human hepatocytes. Chem-Biol. Interact. 142, 83-97. He, K., Talaat, R. E., Pool, W. F., Reily, M. D., Reed, J. E., Bridges, A. J., and Woolf, T. F. (2004) Metabolic activation of troglitazone: Identification of a reactive metabolite and mechanisms involved. Drug Metab. Dispos. 32, 639-646. Cox, P. J., Ryan, D. A., Hollis, F. F., Harris, A. M., Miller, A. K., Vousden, M., and Cowley, H. (2000) Absorption, distribution, and metabolism of rosiglitazone, a potent thiazolidinedione insulin sensitizer, in humans. Drug Metab. Dispos. 28, 772-780. Karanam, B. V., Hop, C. E. C. A., Liu, D., Mike, W., Dean, D., Toriumi, C., Domuro, M., Taga, F., Kusajima, H., and Vincent, S. H (2004) In vitro metabolism of MK-0767, a peroxisome proliferator-activated receptor R/γ agonist. I. Role of cytochrome P450, methyltransferases, FMOs and esterases. Drug Metab. Dispos. 32, 1015-1022. Liu, D. Q., Karanam, B. V., Doss, G. A., Sidler, R. R., Vincent, S. H., and Hop, C. E. C. A. (2004) In vitro metabolism of MK-0767, a peroxisome proliferator-activated receptor R/γ agonist. II. Identification of metabolites by liquid chromatography-tandem mass spectrometry. Drug Metab. Dispos. 32, 1023-1031. Reddy, G. V. B., Doss, G. A., Creighton, M., Kochansky, C., Vincent, S. H., Franklin, R. B., and Karanam, B. V. (2004) Identification and metabolism of a novel dihydrohydroxy-glutathionyl conjugate of MK-0767. Drug Metab. Dispos. 32, 11541161. Gitlin, N., Julie, N. L., Spurr, C. L., Lim, K. N., and Jurabe, H. M. (1998) Two cases of severe clinical and histologic hepatotoxicity associated with troglitazone. Ann. Intern. Med. 129, 38-41. Neuschwander-Tetri, B. A., Isley, W. L., Oki, J. C., Ramrakhiani, S., Quiason, S. G., Phillips, N. J., and Brunt, E. M. (1998) Troglitazone-induced hepatic failure leading to liver transplantation. Ann. Inter. Med. 129, 36-38. Watkins, P. B., and Whitcomb, R. W. (1998) Hepatic dysfunction associated with troglitazone. N. Engl. J. Med. 338, 916-917. Maeda, T., Nomura, M., Awano, K., Kinoshita, S., Satoh, H., Murakami, K., and Tsunoda, M. (2000) N-benzyldioxothiazolidylbenzamide derivatives and process for producing the same. U.S. Patent 6,030,990. Shou, M., Dai, R., Cui, D., Korzekwa, K. R., Baillie, T. A., and Rushmore, T. H. (2001) A kinetic model for the metabolic interaction of two substrates at the active site of cytochrome P450 3A4. J. Biol. Chem. 276, 2256-2262. Buters, J. T., Shou, M., Hardwick, J. P., Korzekwa, K. R., and Gonzalez, F. J. (1995) cDNA-directed expression of human cytochrome P450 1A1 using baculovirus. Purification, dependency on NADPH-P450 oxidoreductase, and reconstitution of catalytic properties without purification. Drug Metab. Dispos. 23, 696701. Shen, A. L., Porter, T. D., Wilson, T. E., and Kasper, C. B. (1989) Structural analysis of the FMN binding domain of NADPHcytochrome P-450 oxidoreductase by site-directed mutagenesis. J. Biol. Chem. 264, 7584-7589. Baillie, T. A., and Davis, M. R. (1993) Mass spectrometry in the analysis of glutathione conjugates. Biol. Mass Spectrom. 22, 319325. Murphy, C. M., Fenselau, C., and Gutierrez, P. L. (1992) Fragmentation characteristics of glutathione conjugates activated by high-energy collisions. J. Am. Soc. Mass Spectrom. 3, 815822. Trimble, L. A., Chauret, N., Silva, J. M., Nicoll-Griffith, D. A., Li, C.-S., and Yergey, J. A. (1997) Characterization of the in vitro oxidative metabolites of the COX-2 selective inhibitor L-766,112. Bioorg. Med. Chem. Lett. 7, 53-56. Valadon, P., Dansete, P. M., Girault, J. P., Amar, C., and Mansuy, D. (1996) Thiophene sulfoxides as reactive intermediates: Formation upon microsomal oxidation of a 3-aroylthiophene and fate in the presence of nucleophiles in vitro and in vivo. Chem. Res. Toxicol. 9, 1403-1413. Decker, C. J., Cashman, J. R., Sugiyama, K., Maltby, D., and Correia, M. A. (1991) Formation of glutathionyl-spironolactone disulfide by rat liver cytochromes P450 or hog liver flavincontaining monooxygenases: A functional probe of two-electron oxidations of the thiosteroid? Chem. Res. Toxicol. 4, 669-677. Decker, C. J., Rashed, M. S., Baillie, T. A., Maltby, D., and Correia, M. A. (1989) Oxidative metabolism of spironolactone: Evidence for the involvement of electrophilic thiosteroid species in drug-mediated destruction of rat hepatic cytochrome P450. Biochemistry 28, 5128-5136.

888


(29) Nordblom, G. D., White, R. E., and Coon, M. J. (1976) Studies on hydroperoxide-dependent substrate hydroxylation by purified liver microsomal cytochrome P450. Arch. Biochem. Biophys. 175, 524-533. (30) Rahimtula, A. D., and O’Brien, P. J. (1974) Hydroperoxide catalyzed liver microsomal aromatic hydroxylation reaction involving cytochrome P450. Biochem. Biophys. Res. Commun. 60, 440-447. (31) Hrycay, E. G., Gustafsson, J.-A., Ingelman-Sundberg, M., and Ernster, L. (1975) Sodium periodate, sodium chlorite, organic hydroperoxides, and hydrogen peroxide as hydroxylating agents in steroid hydroxylation reaction catalyzed by partially purified cytochrome P450. Biochem. Biophys. Res. Commun. 66, 209-216. (32) Hrycay, E. G., and O’Brien, P. J. (1971) Cytochrome P450 as a microsomal peroxidase utilizing a lipid peroxide substrate. Arch. Biochem. Biophys. 147, 14-27.

Reddy et al. (33) Kice, J. L. (1980) Mechanisms and reactivity of organic oxyacids of sulfur and their anhydrides. In Advances in Physical Organic Chemistry (Gold, V., Bethell, D., Eds.) Vol. 17, pp 65-181, Academic Press, New York. (34) Abraham, R. T., Benson, L. M., and Jardine, I. (1983) Synthesis and pH-dependent stability of purine-6-sulfenic acid, a putative reactive metabolite of 6-thiopurine. J. Med. Chem. 26, 15231526. (35) Block, E., and O’Connor, J. (1974) The chemistry of alkyl thiosulfinate esters. VII. Mechanistic studies and synthetic applications. J. Am. Chem. Soc. 96, 3929-3944. (36) Doerr, I. L., Wempen, I., Clarke, D. A., and Fox, J. J. (1961) Thiation of nucleosides. III. Oxidation of 6-mercaptopurines. J. Org. Chem. 26, 3401-3409.

TX0500373

Mechanistic Studies on the Metabolic Scission of Thiazolidinedione

Recommend Documents