Screening and Identification of GSH-Trapped Reactive Metabolites

Chem. Res. Toxicol. 2007, 20, 757-766

757

Screening and Identification of GSH-Trapped Reactive Metabolites Using Hybrid Triple Quadruple Linear Ion Trap Mass Spectrometry Joanna Zheng,† Li Ma,‡ Baomin Xin,† Timothy Olah,† W. Griffith Humphreys,‡ and Mingshe Zhu*,‡ Departments of Bioanalytical Research and Biotransformation, Bristol-Myers Squibb, Princeton, New Jersey 08543 ReceiVed October 17, 2006

The present study describes a new analytical approach for the detection and characterization of GSHtrapped reactive metabolites using multiple reaction monitoring (MRM) as the survey scan to trigger the acquisition of enhanced product ion (EPI) spectra on a triple quadrupole linear ion mass spectrometer. The MRM scan step was carried out following up to 114 MRM transitions from the protonated molecules of potential GSH adducts to their product ions derived from a neutral loss of 129 or 307 Da. MRM transition protocols were constructed on the basis of common bioactivation reactions predicted to occur in human liver microsomes (HLM). The effectiveness and reliability of the approach were evaluated using acetaminophen, diclofenac, and carbamazepine as model compounds. The total ion chromatograms of the MRM for the HLM incubations with these compounds and GSH clearly displayed a number of GSH adducts, including acetaminophen-GSH adducts and carbamazepine-GSH adducts that were not previously observed in HLM incubations. In addition, clomipramine and mefenamic acid that have the frame structures susceptible to P450-mediated bioactivation were investigated. As a result, the MRMEPI analysis revealed multiple GSH adducts of clomipramine and mefenamic acid in HLM incubations possibly mediated by epoxide and/or quinone imine intermediates. Compared with the neutral loss (NL) and precursor ion (PI) scanning analysis, the MRM-based approach provided superior sensitivity and selectivity for GSH adducts. It also enabled the sensitive acquisition of EPI spectra with rich fragmentation in the same LC/MS run, which were useful for the rapid structure elucidation of GSH adducts and the elimination of false positives. The MRM-EPI experiment can be employed for high throughput screening of reactive metabolites and should be especially applicable to compounds of the same chemotype. Also, it can be applied in conjunction with the PI or NL scan as a comprehensive method for the analysis of reactive metabolites in a drug discovery setting. Introduction Although the biochemical mechanisms of drug-induced idiosyncratic hepatotoxicity are unclear, it has been proposed that metabolic activation of drugs to reactive metabolites is associated with these toxicities (1-4). It was reported recently that 13 drugs formed reactive metabolites among the 21 drugs that since 1950 either have been withdrawn from the U. S. market or have a label that indicates a black box warning because of the hepatotoxicity (5). To minimize metabolic activation during the drug discovery process, a variety of analytical methodologies have been developed and applied to the early optimization of lead compounds, all the way to the characterization of development candidates (6-8). Analysis of protein covalent binding in human liver microsomes (HLM1) and in liver tissue from animals have been utilized for the quantitative determination of reactive metabolite formation of drug candidates (9, 10), which is useful for rank ordering compounds and predicting reactive metabolite exposure to * Corresponding author. Tel: 609-252-3324. Fax: 609-252-6802. Email: [email protected]. † Bioanalytical Research. ‡ Biotransformation. 1 Abbreviations: CBZE, carbamazepine-10,11-epoxide; EPI, enhanced product ion; HLM, human liver microsomes; IDA, information-dependent acquisition; MH+, protonated molecule; MRM, multiple reaction monitoring; NL, neutral loss; PI, precursor ion; TIC, total ion chromatogram.

humans (11). However, the protein covalent binding assays require radiolabeled compounds, which greatly limits their applications. LC/MS/MS analysis of reactive metabolites trapped by glutathione (GSH) is a common practice in screening for reactive metabolites when radiolabeled test compounds are not available (8, 12). The first LC/MS method implemented for screening for GSH-trapped reactive metabolites employed the neutral loss (NL) scanning technique on a triple quadrupole mass spectrometer (13). In the NL analysis, test compounds are incubated with HLM in the presence of GSH followed by monitoring for protonated ions that fragment to give a loss of 129 Da, corresponding to the pyroglutamic acid moiety. Detected GSH adducts are structurally characterized further by a product ion scan in a second LC/MS run. The NL scanning method provides a practical means for increased throughput and is especially useful when protonated molecules of GSH adducts cannot be predicted. However, this method suffers low LC/MS sensitivity and limited selectivity resulting from the interference of endogenous compounds and background noise (14). Furthermore, the effectiveness of the neutral loss scanning experiment varies among different classes of GSH adducts because the neutral loss fragmentation patterns are compound-dependent (14). To improve the selectivity of the GSH adduct screening by the neutral loss scan, methods using a mixture of GSH and stable-isotope labeled GSH (1:1 ratio) as the trapping agent (15-

10.1021/tx600277y CCC: $37.00 © 2007 American Chemical Society Published on Web 04/03/2007

758 Chem. Res. Toxicol., Vol. 20, No. 5, 2007

Zheng et al.

Figure 1. Structures of test compounds.

17) and high-resolution LC/MS (18) have been reported. Alternatively, a multiple reaction monitoring (MRM) method to monitor six transitions from protonated molecules (MH+) of potential reactive metabolites trapped by glutathione ethyl ester to their product ions [MH - 129]+ was developed for the sensitive detection of reactive metabolites (19). In addition, a precursor ion (PI) scanning method in the negative electrospray ionization mode has been employed for improving the effectiveness of the triple quadrupole LC/MS analysis (14). Recently, a new, updated triple quadrupole linear ion trap mass spectrometer (API 4000 Q-Trap) was introduced. The hybrid mass spectrometer not only retains the MS/MS scan functions of traditional triple quadrupole instruments, such as the NL, PI, and MRM scans, but also has the capability of linear ion trap instruments, including the full scan MS-based datadependent MS/MS analysis and the acquisition of MS3 spectra. Additionally, it allows for the use of NL, PI, or MRM as the survey scan to trigger the information-dependent acquisition (IDA) of enhanced product ion (EPI) spectra (20-22). The MRM survey scan can be set to follow up to 100 MRM transitions without a significant loss of sensitivity (23). These new MS/MS survey scanning capabilities, combined with the common functions of both triple quadrupole and ion trap instruments, have been utilized for the rapid detection and characterization of oxidative metabolites (24), glucuronides (25), and other conjugated metabolites (26). Additionally, these MS/MS scanning experiments have been applied to simultaneous drug quantification and metabolite screening in liver microsomal incubations (27) and plasma samples (23). The main objective of this study was to explore an MRM-based analytical approach for the detection and characterization of GSH-trapped reactive metabolites using a quadrupole linear ion trap mass spectrometer. The effectiveness of this approach was compared with that of the NL and PI scanning analyses. Additionally, the MRM-based approach was applied to the study of clomipramine and mefenamic acid bioactivation in HLM. Although the two compounds contain frame structures similar to that of imipramine or diclofenac, known to form reactive metabolites in vitro and in vivo, the P450-mediated bioactivation of clomipramine and mefenamic acid has not been reported in the literature.

Experimental Procedures Materials. Pooled HLM was purchased from BD Biosciences (Woburn, MA). GSH, NADPH, acetaminophen, carbamazepine, carbamazepine-10,11-epoxide (CBZE), diclofenac, clomipramine, and mefenamic acid were purchased from Sigma-Aldrich (Saint Louis, MO).

Figure 2. Flow chart of the use of triple quadrupole linear ion trap LC/MS for the detection and characterization of GSH-trapped reactive metabolites. The MRM-EPI approach can be employed alone for the high throughput screening of common and predicted GSH adducts (Strategy A), or used in conjunction with PI or NL prescreening for comprehensive analysis of GSH-trapped reactive metabolites (Strategy B).

Microsomal Incubation. Test compounds (50 µM) (Figure 1) were incubated separately with HLM (1.0 mg/mL), GSH (1 mM), and NADPH (1 mM) in potassium phosphate buffer (100 mM, pH 7.4) for 30 min. The total incubation volume was 2 mL. The incubation reactions were initiated by the addition of a NADPH solution after a 3-min preincubation and were stopped by the addition of 300 µL of trichloroacetic acid (10%) (15). After centrifugation (13,000 rpm for 10 min), supernatants were loaded onto solid-phase extraction cartridges (Oasis extraction cartridges, Waters Corp., Milford, MA). The cartridges were washed with 1 mL of water and then eluted with 2 mL of methanol. The methanol fractions were dried and reconstituted with 200 µL of a wateracetonitrile mixture (v/v, 95/5). Aliquots (20 µL) of the reconstituted solutions were injected into LC/MS/MS. Triple Quadrupole Linear Ion Trap LC/MS. The HPLC system consisted of Shimadzu LC10ADvp bipumps (Columbia, MD), an HTC PAL autosampler (Leap Technologies, Cary, NC), and a Zorbax SB C-18 HPLC column (2.1 × 150 mm, 5 µm, Agilent Technologies, Palo Alto, CA). HPLC mobile phase A was formic acid in water (0.1%), and mobile phase B was acetonitrile (100%). The HPLC flow rate was 0.3 mL/min. The HPLC gradient started at 5% B for 2 min, ramped linearly to 70% B over 15 min, increased to 90% B over 2 min, and then returned to the initial condition over 0.1 min. The HPLC system was interfaced to an API 4000 Q-trap mass spectrometer (Sciex, Toronto, Canada) equipped with the Turboionspray source. The MRM-EPI analysis of GSH adducts followed preset MRM transitions, corresponding to the loss of the GSH or pyroglutamic acid moiety. The MRM transition protocols were prepared on the basis of potential GSH adducts predicted from common bioactivation reactions using an active spreadsheet that can be automatically transferred to an acquisition method via Analyst 1.41 (Sciex, Toronto, Canada). For example, the MH+ of an anticipated GSH adduct of acetaminophen via the quinone imine reactive metabolite was at m/z 457 (protonated acetaminophen + GSH - 2H). The MRM transitions targeting this specific GSH adduct were set from the MH+ to the product ion at both m/z 328 (MH+ - 129) and m/z 150 (MH+ - 307). To evaluate the effectiveness of the number of MRM transitions on the sensitivity and selectivity of GSH adducts, up to 114 MRM transitions, including some dummy MRM transitions, were performed. The dummy MRM transitions were randomly selected. In the MRM analysis, the source temperature was set at 450 °C, and

GSH Adduct Screening by Q-Trap LC/MS

Chem. Res. Toxicol., Vol. 20, No. 5, 2007 759

Table 1. Common P450-Mediated Bioactivation Reactions of Drugs and the Associated Mass Changes of the Drug Moiety of GSH Adductsa mass shift of the drug moietyb

GSH adduct compositionc

functional group

metabolic activation reaction loss of Cl to form aziridinium or episulfonium R-NH-C(dS)-NH-R′ f R-NH-C(SG)dN-R′ formation of hydroxylamine R-NH-NH2 f R-SG loss of N-acetyl group followed by formation of hydroxylamine formation of quinone imine followed by GSH attachment and loss of HCl oxidative deamination to formamide formation of quinone imine by demethylation formation of quinone formation of a furan epoxide formation of nitrenium formation of isocynate formation of quinone methide formation of quinone-imine formation of epoxide followed by GSH attachment and loss of water formation of N-hydroxy formation of quinone via oxidative deamination CH2dCH-C(dO)-R′ f GS-CH2-CH-C(dO)-R′

-34

P + GSH - HCl

-32 -30 -26

P + GSH - S - 2H P + GSH - 2O - 4H P + GSH - 2N - 4H P + GSH + O - CH2CO - 2H

β-Cl to a nitrogen or sulfur thiourea aromatic nitro hydralazine N-acetyl aromatic amine

-18

P + GSH + O - HCl

aromatic chlorine

-15

P + GSH - NH3

benzylamine

-14

P + GSH - CH2 - 2H

aromatic ether

-12 -2 0

P + GSH - C - 2H P + GSH - 4H P + GSH - 2H

methylenedioxy monoterpene ketone azaheterocycle formamide substituted phenol derivative anilide (p-aminophenol) thiophene, furan aromatic amine aromatic amine

+1

P + GSH + O - NH - 2H

+2

P + GSH

+4

P + GSH + O - CH2

R,β -unsaturated cabonyl benzopyran

+14

P + GSH + O - 4H

benzylamine

+16

P + GSH + O - 2H

+18

P + GSH + O

+28 +32 +34

P + GSH + CO - 2H P + GSH + 2O - 2H P + GSH + 2O

aromatic amine phenol thiophene furan, benzene, thiophene benzylamine benzene 1,3-conjugated diene derivative

example

demethylaion and expoxidation formation of hydroxylamine followed by oxidation to nitrile oxide formation of nitroso formation of quinone formation of S-oxide formation of epoxide followed by GSH attack formation of isocyanate formation of quinone formation of R,β-unsaturated carbonyl

phenoxybenzamine (35) SDZ HDL 376 (3) metronidazole (35) hydralazine (3) phenacetin (4) diclofenac (32) benzuylamine (3) eugenol (3) paroxetine (3) R-(+)-pulegone (45) clozapine (3) dopamine (3) acetaminophen (4) thiophene derivative (2) procainamide (4) 5-aminosalicylic acid (35) CI-1033 (3) precocene I (46) DPC 423 (3)

aryl amine derivatives (35) estradiol (3) tienilic acid (2) aflatoxin B1 (3) benzylamine (3) bromobenzene (3) butadiene (3)

a The listed common bioactivation reactions of drugs that may occur in HLM or other in vitro systems were mainly summarized on the basis of the data presented in recent reviews in the literature (2-5, 34, 35). The list may not include reactive metabolites formed via uncommon bioactivation reactions or multistep pathways, or reactive metabolites that are not trapped by GSH. b The mass shift represents the mass change of the drug moiety of a GSH adduct. The mass shift listed is defined as the MH+ of the GSH adduct - 305 (GSH - 2H). One mass shift may represent multiple bioactivation reactions. c P is the parent drug.

the ionspray voltage was set to 4.8 kV. Nitrogen was used as the nebulizer and auxiliary gas. The dwell times for MRM analysis (86 to 114 transitions) were 5-7 ms, and the interscan pause time for all MRM analysis was 5 ms. The same declustering potential (60 V), collision energy (40 eV), and a collision energy spread ((15 eV) were applied for all potential GSH adducts in both NLEPI and MRM-EPI modes. The NL experiment was performed by monitoring losses of 129 and 307 Da under the same MS/MS conditions used in the MRM analysis. The PI scan of m/z 272 was run in the negative electrospray mode at a scan range from m/z 350 to m/z 800. The ionspray voltage was set to -4.5 kV, and the declustering potential was at -60 V. The collision energy and spread for EPI in the negative mode were set at -35 and 15 eV, respectively. Triple Quadrupole LC/MS. The same HPLC instrument and solvents described above were used. The HPLC gradient started at 2% B for 3 min, ramped linearly to 90% B over 27 min, held at 90% B over 5 min, and then returned to the initial condition over

2 min. The HPLC flow rate was 0.25 mL/min. The HPLC system was interfaced to a Finnigan TSQ quantum triple quadrupole mass spectrometer (Thermo Finnigan, San Jose, CA). The NL scans of 129 and 307 Da were performed at a scan range from 350 to 800 Da and in the positive ion electrospray mode. The spray voltage was set at 4.0 kV, and the capillary temperature was set at 350 °C. The MS/MS analyses were operated using argon as the collision gas with the collision energy at 26 eV.

Results Analytical Strategy. The MRM-based approach for the detection and characterization of GSH-trapped reactive metabolites using the hybrid quadrupole triple linear ion trap LC/MS (Q-Trap) is illustrated in Figure 2. In the analysis, MRM is employed as the survey scan by following more than 100 MRM transitions, all targeting potential GSH adducts. The protonated molecules of the GSH adducts targeted are calculated on the


Figure 3. MRM-EPI analysis (86 MRM transitions) of acetaminophenGSH adducts formed in the HLM incubation. (A) TIC of MRM; (B) TIC of EPI.

basis of common metabolic activation reactions predicted to occur in HLM (Table 1) and specific bioactivation reactions observed from the compounds that have similar sub-structures. Once a GSH adduct is detected by the MRM survey scan, a dependent acquisition of EPI is triggered. GSH adducts detected are displayed in the total ion chromatogram (TIC) of the MRM or the EPI scan (Figure 3). Product ion spectra acquired in the MRM-EPI analysis are utilized for both the structure elucidation of the GSH adducts and the elimination of false positive peaks. In the current study, two analytical strategies that utilized the MRM-EPI approach were explored: (A) MRM screening of common and predicted GSH adducts and (B) MRM screening following an initial PI or NL prescreen for comprehensive analysis of GSH-trapped reactive metabolites (Figure 2). Acetaminophen, carbamazepine, and diclofenac, all known to undergo metabolic activation, were used as model compounds for the validation of strategy A. In addition, the bioactivation of clomipramine and mefenamic acid in HLM was investigated using strategy B. Analysis of GSH Adducts Formed in HLM Incubations with Acetaminophen. The MRM-EPI analysis of an HLM incubation of acetaminophen with GSH revealed four GSH adducts (Figure 3). The total ion chromatograms of the MRM (Figure 3A) and EPI scan (Figure 3B) following 86 MRM transitions were very comparable. The EPI spectra (Figure 4) of the detected acetaminophen-GSH adducts (AM1-AM4) were directly retrieved from the TIC of the EPI scan (Figure 3B). AM2 and AM3 had the same protonated molecule at m/z 457, corresponding to the direct addition of GSH to acetaminophen (Table 2). The two GSH adducts showed very similar product ion spectra (Figures 4B and 4C), consistent with the previously reported product ion spectra of 2′-GS-APAP and 3′-GS-APAP (28). AM2 is tentatively assigned as 3′-GS-APAP (Figure 4B) on the basis of its relative abundance. Consequently, the minor GSH adduct AM3 is tentatively assigned as 2′-GS-APAP (Figure 4C). The protonated molecules of AM1 (m/z 473) and AM4 (m/z 489) indicate that AM1 was a GSH adduct of monohydroxyacetaminophen, and AM4 was a GSH adduct of dihydroxyacetaminophen (Table 2). The product ion spectra of AM1 (Figure 4A) and AM4 (Figure 4D) suggest that the sites of the mono- and di-hydroxylation occurred on the aromatic ring of acetaminophen. The NL scanning of 129 and 307 Da was only able to detect AM2 in the HLM incubation (Table 2), which is consistent with previous observations from the NL scanning analyses (13, 15, 18). The signal-to-noise ratios of the

Zheng et al.

AM2 peak in the NL, PI, and MRM ion chromatograms determined by the Q-trap were 4, 29, and 292, respectively. Analysis of GSH Adducts Formed in HLM Incubations with Carbamazepine. The TIC (Figure 5A) and extracted ion chromatograms (Figures 5B-D) of the MRM-EPI analysis for an HLM incubation of carbamazepine with GSH displayed seven GSH adducts (CM1-CM7) with no significant false positive peaks. The NL scanning analysis of the same carbamazepine incubation sample by triple quadrupole LC/MS detected CM2, CM3, CM4, CM6, and CM7, but not CM1 and CM5 (Table 2). CM1, CM2, and CM4 had the same protonated molecule at m/z 560 (protonated carbamazepine + GSH + O) and displayed similar MS/MS spectra (Figure 5B and Table 2), suggesting that these adducts were formed after an epoxidation and ringopening by a nucleophilic attachment of GSH. The same GSH adducts were observed in the HLM incubation of CBZE with GSH in the absence of NADPH (data not shown). The HLM incubation of CBZE with GSH did not generate CM3, CM5, CM6, and CM7. The MS/MS spectra of CM2 and CM4, the major GSH adducts of CBZE, showed fragmentation patterns (Table 2) similar to those of the previously reported diastereomers of 10-hydroxy-11-glutathionyl-carbamazepine that were formed from the chemical reaction of CBZE with GSH (29). CM5, CM6, and CM7 had the same protonated molecule at m/z 558 (protonated carbamazepine + GSH + O - 2H) (Figure 5C), suggesting that an oxygen atom and a GSH molecule were incorporated into carbamazepine. Most likely, other epoxide intermediates rather than CBZE mediated the formation of these three GSH adducts. The protonated molecule of CM3 was at m/z 576 (protonated carbamazepine + GSH + 2O) (Figure 5D), suggesting that CM3 may be formed via an expoxide-derived form of a mono-hydroxylated carbamazepine (Table 2). Analysis of GSH Adduct Formed in HLM Incubations with Diclofenac. The MRM-EPI analysis of an HLM incubation of diclofenac with GSH revealed two GSH-trapped reactive metabolites that had protonated molecules at m/z 583 (DM1) and m/z 617 (DM2) (Table 2). The same GSH adducts were detected in the HLM incubation by the NL scanning analysis in this present study (Table 2) and previously studies (30, 31). The EPI spectrum of DM1 (Table 2) was very similar to that of 4′-OH-2′-glutathione-deschloro-diclofenac (32). Thus, DM1 was tentatively assigned as 4′-OH-2′-glutathione-deschlorodiclofenac. It was previously proposed that DM1 was derived from the diclofenac-1′,4′-quinone imine intermediate followed by GSH attack and a loss of HCl (32). DM2 could be one or a mixture of diclofenac-GSH adducts formed via the 1′,4′-quinone imine reactive metabolite and/or the diclofenac-2,5-quinone imine intermediate (30, 33). Analysis of GSH Adducts Formed in HLM Incubations with Clomipramine. To evaluate the utility of analytic strategy B (Figure 2), the neutral loss scanning analysis of 129 and 307 Da was first conducted to screen for GSH-trapped reactive metabolites of clomipramine in the HLM incubations on a triple quadrupole linear ion trap mass spectrometer. Two GSH adducts, LM1 and LM2, were detected by the NL scan of 307 Da (Figure 6B) but not by the NL scan of 129 Da (Figure 6A). The subsequent MRM-EPI analysis showed very intense LM1 and LM2 peaks with no other significant components at the TIC level (Figure 6C). The signal-to-noise ratios of the LM1 and LM2 peaks detected by the MRM analysis were 13- and 24-fold greater than those of the analogous peaks detected by the NL scan of 307 Da (Figure 6). The EPI spectra of LM1 and LM2 were identical and showed that the major product ions



Figure 4. Product ion spectra and proposed structures of acetaminophen-GSH adducts acquired by the MRM-directed EPI scan. (A) AM1; (B) AM2; (C) AM3; (D) AM4.

were derived from losses of 129 (-Glu) and 307 (-GSH) Da (Figure 6D). Analysis of GSH Adducts Formed in HLM Incubations with Mefenamic Acid. The ion chromatogram from the NL scanning analysis of 129 Da for an HLM incubation of mefenamic acid with GSH only displayed several peaks that are not associated with GSH adducts (Figure 7A). The ion chromatogram from the PI scanning analysis of m/z 272 in the negative ion mode exhibited a GSH adduct peak (MM1) along with some false positive peaks (Figure 7B). The MRM analysis enabled the detection of MM1 and two additional GSH adducts (MM2 and MM3) (Figure 7C). The EPI spectra (Figure 8A-C) of MM1, MM2, and MM3, acquired by the dependent EPI scan, displayed the product ions associated with the cleavage of the GSH moiety (a loss of 75 (-Gly), 129 (-Glu) or 307 Da (-GSH)) and the carboxylic acid moiety of mefenamic acid (a loss of 18 (-H2O) or 44 Da (-COO)). The MH+ of MM1 was at m/z 563 (protonated mefenamic acid + GSH + O - 2H) (Figure 8A), indicating that an oxygen atom was incorporated into the mefenamic acid moiety. MM2 and MM3 had the same MH+ at m/z 547 (protonated mefenamic acid + GSH - 2H) (Figures 8B and C). On the basis of the data from the MRM-EPI analyses, tentative structures and formation pathways of MM1, MM2, and MM3 were proposed (Figure 9).

Discussion The hybrid linear ion trap triple quadruple mass spectrometer provides the capabilities of using NL, PI, or MRM as a survey scan to trigger the dependent acquisition of EPI spectra (20). These acquisition functions are not available on traditional triple quadrupole, linear ion trap, or high-resolution mass spectrometers. The analytical methodology developed in this study (Figure 2) takes advantage of the capability by utilizing the MRM-EPI scan for sensitive, selective, and rapid detection and structural characterization of GSH-trapped reactive metabolites. To facilitate the MRM analysis of reactive metabolites, the mass shifts of reactive metabolite-GSH adducts derived from common bioactivation reactions were summarized (Table 1). For searching for a potential GSH adduct, two MRM transitions from the protonated molecule of the GSH adduct to its product ions derived from a neutral loss of 129 or 307 Da were performed. As previously reported, the NL scans of 129 and 307 Da are complementary in the detection of different classes of GSH adducts (14). Two analytical strategies of using the MRM-based approach have been developed in this study. To rapidly screen for reactive metabolites of compounds, the MRM-EPI scan using predicted adduct mass shifts (Table 1) can be directly employed for both GSH adduct detection and acquisition of MS/MS spectral data in a single LC/MS run (Strategy A, Figure 2). If a related series


Zheng et al.

Table 2. Summary of GSH Adducts Detected by MRM

compound (MH+)

GSH adducta

MH+ and (major fragment) of GSH adductb

acetaminophen (152)

AM1

473 (440, 398, 344, 327, 312, 224, 164, 156) 457 (382, 336, 328, 311, 208, 166, 140) 457 (382, 336, 328, 311, 208, 166, 140 489 (414, 360, 184, 172) 560 (542, 499, 485, 431, 413, 370, 208, 180) 560 (542, 499, 485, 476, 431, 413, 370, 208, 180) 576 (558, 501, 447, 413, 370, 208, 180) 560 (542, 499, 485, 476, 431, 413, 370, 208, 180) 558 (540, 483, 429, 411, 366, 212, 180) 558 (540, 483, 429, 411, 393, 366, 212, 180) 558 (540, 483, 429, 411, 393, 366, 212, 180) 583 (508, 490, 454, 436, 419, 315, 262) 617 (542, 488, 470, 452, 331, 296) 620 (491, 313) 620 (491, 313) 563 (488, 434, 416, 288, 272, 272, 244, 240, 226) 547 (400, 240, 222, 194) 547 (529, 472, 240, 222, 196, 181)

AM2 e AM3 carbamazepine (237)

AM4 CM1 CM2 CM3 CM4 CM5 CM6 CM7

diclofenac (296)

DM1 DM2

clomipramine (316) mefenamic acid (242)

LM1f LM2f MM1g MM2 MM3

mass shift of the drug moiety of the GSH adductc

GSH adduct detected by NLd

+ 16

-

P + GSH - 2H

0

+

P + GSH - 2H

0

-

P + GSH, + 2O - 2H P + GSH + O

+ 32 + 18

-

P + GSH + O

+ 18

+

P + GSH + 2O

+ 34

+

P + GSH + O

+ 18

+

P + GSH + O - 2H

+ 16

-

P + GSH + O - 2H

+ 16

+

P + GSH + O - 2H

+ 16

+

P + GSH + O - HCl

- 18

+

P + GSH + O - 2H

+ 16

+

P + GSH - 2H P + GSH - 2H P + GSH + O - 2H

0 0 +16

+ + -

0 0

-

postulated adduct composition P + GSH + O - 2H

P + GSH - 2H P + GSH - 2H

a The GSH adducts detected were the peaks displayed in total ion chromatograms obtained from MRM analyses. The MRM-EPI analysis targeted expected and common reactive metabolites listed in Table 1. In addition, some MRM transitions that are not associated with bioactivation were included to evaluate the effects of the number of MRM transitions on the effectiveness of the MRM scanning method. The numbers of MRM transitions were 86 (acetaminophen), 88 (carbamazepine), 114 (diclofenac), 14 (clomipramine), and 44 (mefenamic acid). b MS/MS spectra were obtained by the MRM-EPI scan. Product ions listed were of greater than 5% relative intensity. The boldface type denotes the product ions raised from a loss of either 129 or 307 Da. c The mass shift of the drug moiety of GSH adduct is defined as the protonated GSH adduct molecule - 305 (GSH - 2H) (Table 1). GSH adducts were detected by the MRM transitions from protonated adduct molecules (MH+ of drug + the mass shift listed + 305) to the product ions derived from a loss of either 120 or 307 Da. d The + denotes GSH adducts displayed in total ion chromatograms of NL scanning of 129 and 307 Da on the triple quadrupole instrument; the - denotes those not detected. e AM2 was also detected by the NL and PI scans on the Q-Trap instrument. f LM1 and LM2 were also detected by the NL scan of 307 Da on the Q-Trap instrument (Figure 6). g NM1 was also detected by the negative ion PI scan of m/z 272 on the Q-Trap instrument (Figure 7).

of compounds are being screened, then MRM transition protocols are modified on the basis of the specific bioactivation pathways occurring for lead compounds. To accomplish the same analytical tasks on a triple quadrupole mass spectrometer, two LC/MS runs including a NL, PI, or MRM scan and a subsequent product ion scan are required (14, 15, 18, 19). The utility of this approach has been demonstrated in the analysis of GSH adducts formed in the HLM incubations with acetaminophen (86 MRM transitions, Figure 3), carbamazepine (88 MRM transitions, Figure 5), and diclofenac (114 MRM transitions, Table 2). Additionally, the results demonstrate that the MRM analyses were highly effective, even though more than 100 MRM transitions were performed. Table 1 lists 19 different mass shifts of drug moieties of GSH adducts derived from common bioactivation reactions (2-5, 34, 35). Thus, an MRM protocol with a total of 38 MRM transitions is able to detect all of the potential GSH adducts listed. To search for the GSH adducts with mass shifts that are not listed in Table 1, more MRM transitions can be added in the MRM analysis. As indicated in Table 1, the mass shifts of drug moieties of common GSH adducts are all even numbers except for two GSH adducts derived from the bioactivation that are associated with a loss a nitrogen atom (the mass shift of -15 or +1). Therefore, if a loss of a N atom is not expected in the bioactivation of a test

compound, a generic MRM protocol containing all even numbers of mass shifts over a range from +50 to -40 Da around the m/z values of drug + GSH (approximately 90 MRM transitions) can be simply applied to the screening for potential GSH-trapped reactive metabolites. For comprehensive analyses of reactive metabolites, the use of the NL or PI scan as a prescreen followed by the MRM-EPI scan can be applied (Strategy B, Figure 2). The NL or PI approach is used to search for uncommon or unpredicted reactive metabolites, whereas the MRM analysis provides better sensitivity and selectivity for detecting reactive metabolites listed in the MRM transition protocols. In the current study, strategy B was applied to the analysis of GSH adducts formed in the HLM incubations with clomipramine (Figure 6) and mefenamic acid (Figures 7 and 8) because the bioactivation of these two compounds by liver microsomes has not been previously reported. Clomipramine, a tricyclic antidepressant, is metabolized by P450 enzymes to desmethylclomipramine and 2- or 8-hydroxyclomipramine in HLM incubations (36). In this study, two GSH adducts of clomipramine (LM1 and LM2) were detected in the HLM incubation (Figure 6C), and it is proposed that LM1 and LM2 were formed from the addition of GSH to epoxide intermediates followed by a loss of water (Figure 6D). The bioactivation of imipramine that contains the same tricyclic


Figure 5. Detection of carbamazepine-GSH adducts formed in the HLM incubations by the MRM analysis with 88 transitions. (A) TIC of EPI; (B) extracted MRM chromatograms for m/z 560; (C) m/z 558; (D) m/z 576.


Figure 7. NL, PI, and MRM scan analyses of mefenamic acid-GSH adducts formed in the HLM incubation. (A) TIC of NL scanning of 129 Da; (B) TIC of negative PI scanning of m/z 272; (C) TIC of MRM with 44 transitions.

Figure 8. Product ion spectra of mefenamic acid-GSH adducts acquired by the MRM-EPI scan. (A) MM1; (B) MM2; (C) MM3.

Figure 6. NL and MRM analyses of clomipramine-GSH adducts formed in the HLM incubation. (A) TIC of NL scanning of 129 Da; (B) TIC of NL scanning of 307 Da; (C) TIC of MRM with 14 transitions; (D) product ion spectrum of LM1 acquired by MRM-EPI. The product ion spectrum of LM2 (data not shown) was identical to that of LM1. Tentative structures of LM1 and LM2 are presented in Figure 6C.

core structure may be also associated with an epoxide intermedaite (8, 37). Mefenamic acid is a diphenylamine nonsteroidal anti-inflammatory drug. Like its analogue diclofenac, mefenamic acid has been shown to be cytotoxic in rat hepatocytes (38). Previous reports have shown that mefenamic acid is metabolized

on the 3′-methyl group to form 3′-hydroxymethyl and 3′carboxyl derivatives in humans (39). In addition, acylglucuronide conjugates of mefenamic acid were observed in human urine (40). In the current study, three mefenamic acid GSH conjugates, MM1, MM2, and MM3, were detected by a combination of PI prescreening and MRM-EPI (Figure 7). It is proposed that MM1 was formed via a quinone imine intermediate, and MM2 and MM3 were formed via epoxide intermediates (Figure 9). These activation reactions represent new metabolic pathways that have not been previously reported for mefenamic acid (39, 40). The formation of quinone imines on the diphenylamine moiety of diclofenac has been suggested in several previous in vitro and in vivo metabolism studies (31, 41). Compared with the NL or PI scan analyses, the most attractive advantage of the MRM-based analysis is its superior sensitivity. The MRM-EPI method enabled the detection of a number of


Figure 9. Proposed CYP-mediated bioactivation pathways of mefenamic acid in HLM.

GSH adducts of the compounds tested, which were not detected by the NL scan on a triple quadrupole (Table 2) and Q-trap instrument (Figures 6 and 7). For example, in addition to the previously reported carbamazepine-GSH adducts CM2 and CM4 (15, 29), MRM analysis revealed new carbamazepine-GSH adducts CM3, CM5, CM6, and CM7 in the HLM incubations (Figure 5). The results provide direct evidence that at least three distinct reactive intermediates may be involved in the bioactivation of carbamazepine in humans in vitro. This includes the formation of CM1, CM2, and CM4 via CBZE (29), the formation of CM5, CM6, and CM7 via a different epoxide intermediate, possibly carbamazepine-1,2-epoxide, and the formation of CM3 via an epoxide derived from a stable monohydroxylated metabolite. The GSH adduct formation via carbamazepine-1,2-epoxide has been previously observed in rats and mice in vivo (42-44). To directly compare the sensitivities of the NL, PI, and MRM methods on the Q-trap instrument, the signal-to-noise ratios of selected GSH adduct peaks detected in the HLM incubations with clomipramine (Figure 6), mefenamic acid (Figure 7), and acetaminophen were determined. These data suggest that the MRM experiment was up to 90fold better than the NL scan and up to 10-fold better than the PI scan. The PI analysis showed better detection sensitivity for GSH adducts than the NL analysis, consistent with previously reported data (14). The MRM method also offered relatively high detection selectivity for GSH adducts, which is clearly illustrated in the analysis of GSH adducts formed in the HLM incubations with carbamazipine (Figure 5A), clomipramine (Figure 6C), and mefenamic acid (Figure 7C). The MRM and EPI ion chromatograms for these incubation samples displayed intense GSH adduct peaks with no or a few minor false positive peaks. In contrast, the NL or PI scanning method generated significant false positive peaks and relatively higher background (Figures 6A and B, and 7A and B). The use of stable isotope-labeled GSH (15-17), control sample comparison (without GSH or NADPH), and MS/ MS spectral interpretation can facilitate the recognition and elimination of false positives in the NL scan analysis on a triple quadrupole instrument. However, these approaches are either inapplicable to in vivo samples and/or require an additional LC/ MS/MS run to obtain product ion spectra. The higher selectivity of the MRM method and the capability to acquire dependent EPI spectra in the same LC/MS can be useful in the analysis of GSH adducts in bile because the matrix contains significant amounts of endogenous GSH conjugates, resulting in many false positive peaks in the NL analysis (14).

Zheng et al.

This study also demonstrates that the MRM-EPI method provides a means for the rapid and sensitive acquisition of MS/ MS spectra of GSH adducts. For example, high quality MS/ MS spectra of minor acetaminophen-GSH adducts, AM1 (Figure 4A) and AM4 (Figure 4D), were obtained in the MRM-EPI analysis even though their concentrations in the incubation were very low. In addition, the MS/MS spectral data collected by the Q-trap instrument using a collision energy spread across a preset range (45 ( 15 eV) showed rich product ions with no low mass cutoff, providing a great deal of useful diagnostic ion information. For example, the EPI spectrum of AM2 (3′GS-APAP, Figure 4B) displayed rich and equally distributed fragment ions, including those characteristic of GSH adducts (loss of 75, 129, or 307 Da) and lower mass ions (m/z 182, 166, 156, and 140) associated with the drug moiety. In contrast, the product ion spectrum of AM2 acquired by a linear ion trap instrument only showed a few large mass fragment ions (17). In order to generate these informative lower mass fragments, such as m/z 182 and 140, the acquisition of an MS3 spectrum was required. In addition to facilitating the structural elucidation of GSH adducts, MS/MS spectra generated from the MRMEPI experiment were useful for recognizing false positive peaks displayed in the MRM ion chromatograms. For example, the false positive peaks (retention times of 1.4, 2.0, and 3.1 min, Figure 3) from the acetaminophen incubation were immediately recognized because their EPI spectra did not show multiple characteristic fragments consistent with either the acetaminophen or GSH moiety (data not shown). In summary, the present study describes a new analytical approach using MRM as a survey scan to trigger the EPI scanning for rapid detection and characterization of GSH-trapped reactive metabolites on a triple quadrupole linear ion trap mass spectrometer (Figure 2). The MRM transitions were constructed on the basis of common and predicted metabolic activation reactions occurring in HLM. The results from this study demonstrate that the MRM-based approach has several advantages over the NL and PI scanning methods: (1) MRM provides higher sensitivity for the detection of GSH adducts, up to 10fold better than the PI scan and up to 179-fold better than the NL scan. (2) MRM is more selective than the PI or NL scan and provides fewer false positive peaks. (3) High quality MS/ MS spectral data obtained in the same LC/MS run are useful for the rapid structural elucidation of GSH adducts and the elimination of false positive peaks. The major limitation of the MRM-based approach is that it only detects the GSH adducts preset on an MRM transition protocol. To enable the detection of unexpected GSH adducts or reactive metabolites that are not listed in the generic MRM transition protocols, a strategy using the PI or NL scan as a prescreening method can be applicable. Overall, the MRM-EPI approach can be employed alone for the high throughput reactive metabolite screening of lead compounds, especially for compounds of the same chemotypes, or used in conjunction with PI or NL prescreening for the comprehensive analysis of predicted or unpredicted GSH adducts. The reactive metabolite screening capability of MRMEPI, together with metabolite characterization using NL, PI, and MS3 scans, makes the hybrid linear ion trap instrument a useful LC/MS platform for drug metabolism research in a drug discovery setting.

References (1) Uetrecht, J. P. (2000) Is it possible to more accurately predict which drug candidates will cause idiosyncratic drug reactions? Curr. Drug Metab. 1, 133-141.

GSH Adduct Screening by Q-Trap LC/MS (2) Nelson, S. D. (2001) In Biological ReactiVe Intermediates (Dansette, P. M., Delaforge, R. S. M., Gibson, G. G., Greim, H., Jollow, D. J., Monks, T. J., Sipes, I. G., Eds.) Vol. VI, Kluwer Academic/Plenum Publishers, New York. (3) Kalgutkar, A. S., Gardner, I., Obach, R. S., Shaffer, C. L., Callegari, E., Henne, K. R., Mutlib, A. E., Dalvie, D. K., Lee, J. S., Nakai, Y., O’Donnell, J. P., Boer, J., and Harriman, S. P. (2005) A comprehensive listing of bioactivation pathways of organic functional groups. Curr. Drug Metab. 6, 161-225. (4) Zhou, S., Chan, E., Duan, W., Huang, M., and Chen, Y. Z. (2005) Drug bioactivation, covalent binding to target proteins and toxicity relevance. Drug Metab. ReV. 37, 41-213. (5) Walgren, J. L., Mitchell, M. D., and Thompson, D. C. (2005) Role of metabolism in drug-induced idiosyncratic hepatotoxicity. Crit. ReV. Toxicol. 35, 325-361. (6) Ma, S., and Subramanian, R. (2006) Detecting and characterizing reactive metabolites by liquid chromatography/tandem mass spectrometry. J. Mass Spectrom. 41, 1121-1139. (7) Uetrecht, J. (2003) Screening for the potential of a drug candidate to cause idiosyncratic drug reactions. Drug DiscoVery Today 8, 832837. (8) Evans, D. C., Watt, A. P., Nicoll-Griffith, D. A., and Baillie, T. A. (2004) Drug-protein adducts: an industry perspective on minimizing the potential for drug bioactivation in drug discovery and development. Chem. Res. Toxicol. 17, 3-16. (9) Zhang, Z., Chen, Q., Li, Y., Doss, G. A., Dean, B. J., Ngui, J. S., Silva, Elipe, M., Kim, S., Wu, J. Y., Dininno, F., Hammond, M. L., Stearns, R. A., Evans, D. C., Baillie, T. A., and Tang, W. (2005) In vitro bioactivation of dihydrobenzoxathiin selective estrogen receptor modulators by cytochrome P450 3A4 in human liver microsomes: formation of reactive iminium and quinone type metabolites. Chem. Res. Toxicol. 18, 675-685. (10) Day, S. H., Mao, A., White, R., Schulz-Utermoehl, T., Miller, R., and Beconi, M. G. (2005) A semi-automated method for measuring the potential for protein covalent binding in drug discovery. J. Pharmacol. Toxicol. Methods 52, 278-285. (11) Samuel, K., Yin, W., Stearns, R. A., Tang, Y. S., Chaudhary, A. G., Jewell, J. P., Lanza, T., Jr., Lin, L. S., Hagmann, W. K., Evans, D. C., and Kumar, S. (2003) Addressing the metabolic activation potential of new leads in drug discovery: a case study using ion trap mass spectrometry and tritium labeling techniques. J. Mass Spectrom. 38, 211-221. (12) Baillie, T. A., Pearson, P. G., Rashed, M. S., and Howald, W. N. (1989) The use of mass spectrometry in the study of chemicallyreactive drug metabolites. Application of MS/MS and LC/MS to the analysis of glutathione- and related S-linked conjugates of Nmethylformamide. J. Pharm. Biomed. Anal. 7, 1351-1360. (13) Chen, W. G., Zhang, C., Avery, M. J., and Fouda, H. G. (2001) In Biological ReactiVe Intermediates (Dansette, P. M., Delaforge, R. S. M., Gibson, G. G., Greim, H., Jollow, D. J., Monks, T. J., Sipes, I. G., Eds.) Vol. VI, pp 521-524, Kluwer Academic/Plenum Publishers, New York. (14) Dieckhaus, C. M., Fernandez-Metzler, C. L., King, R., Krolikowski, P. H., and Baillie, T. A. (2005) Negative ion tandem mass spectrometry for the detection of glutathione conjugates. Chem. Res. Toxicol. 18, 630-638. (15) Yan, Z., and Caldwell, G. W. (2004) Stable-isotope trapping and highthroughput screenings of reactive metabolites using the isotope MS signature. Anal. Chem. 76, 6835-6847. (16) Yan, Z., Maher, N., Torres, R., Caldwell, G. W., and Huebert, N. (2005) Rapid detection and characterization of minor reactive metabolites using stable-isotope trapping in combination with tandem mass spectrometry. Rapid Commun. Mass Spectrom. 19, 33223330. (17) Mutlib, A., Lam, W., Atherton, J., Chen, H., Galatsis, P., and Stolle, W. (2005) Application of stable isotope labeled glutathione and rapid scanning mass spectrometers in detecting and characterizing reactive metabolites. Rapid Commun. Mass Spectrom. 19, 3482-3492. (18) Castro-Perez, J., Plumb, R., Liang, L., and Yang, E. (2005) A highthroughput liquid chromatography/tandem mass spectrometry method for screening glutathione conjugates using exact mass neutral loss acquisition. Rapid Commun. Mass Spectrom. 19, 798-804. (19) Soglia, J. R., Harriman, S. P., Zhao, S., Barberia, J., Cole, M. J., Boyd, J. G., and Contillo, L. G. (2004) The development of a higher throughput reactive intermediate screening assay incorporating microbore liquid chromatography-micro-electrospray ionization-tandem mass spectrometry and glutathione ethyl ester as an in vitro conjugating agent. J. Pharm. Biomed. Anal. 36, 105-116. (20) Hopfgartner, G., Varesio, E., Tschappat, V., Grivet, C., Bourgogne, E., and Leuthold, L. A. (2004) Triple quadrupole linear ion trap mass spectrometer for the analysis of small molecules and macromolecules. J. Mass Spectrom. 39, 845-855.

Chem. Res. Toxicol., Vol. 20, No. 5, 2007 765 (21) Hager, J. W., and Le, Blanc, J. C. (2003) High-performance liquid chromatography-tandem mass spectrometry with a new quadrupole/ linear ion trap instrument. J. Chromatogr., A 1020, 3-9. (22) King, R., and Fernandez-Metzler, C. (2006) The use of Qtrap technology in drug metabolism. Curr. Drug Metab. 7, 541-545. (23) Li, P., Wang, Y., Tang, Y., Fawcett, J. P., Cui, Y., and Gu, J. (2005) Determination of telmisartan in human plasma by liquid chromatography-tandem mass spectrometry. J. Chromatogr., B 828, 126-129. (24) Hopfgartner, G., Husser, C., and Zell, M. (2003) Rapid screening and characterization of drug metabolites using a new quadrupole-linear ion trap mass spectrometer. J. Mass Spectrom. 38, 138-150. (25) Hager, J. W., and Yves Le Blanc, J. C. (2003) Product ion scanning using a Q-q-Q linear ion trap (Q TRAP) mass spectrometer. Rapid Commun. Mass Spectrom. 17, 1056-1064. (26) Scholz, K., Dekant, W., Volkel, W., and Pahler, A. (2005) Rapid detection and identification of N-acetyl-L-cysteine thioethers using constant neutral loss and theoretical multiple reaction monitoring combined with enhanced product-ion scans on a linear ion trap mass spectrometer. J. Am. Soc. Mass Spectrom. 16, 1976-1984. (27) Shou, W. Z., Magis, L., Li, A. C., Naidong, W., and Bryant, M. S. (2005) A novel approach to perform metabolite screening during the quantitative LC-MS/MS analyses of in vitro metabolic stability samples using a hybrid triple-quadrupole linear ion trap mass spectrometer. J. Mass Spectrom. 40, 1347-1356. (28) Chen, W., Shockcor, J. P., Tonge, R., Hunter, A., Gartner, C., and Nelson, S. D. (1999) Protein and nonprotein cysteinyl thiol modification by N-acetyl-p-benzoquinone imine via a novel ipso adduct. Biochemistry 38, 8159-8166. (29) Bu, H. Z., Kang, P., Deese, A. J., Zhao, P., and Pool, W. F. (2005) Human in vitro glutathionyl and protein adducts of carbamazepine10,11-epoxide, a stable and pharmacologically active metabolite of carbamazepine. Drug. Metab. Dispos. 33, 1920-1924. (30) Tang, W., and Abbott, F. S. (1996) Characterization of thiol-conjugated metabolites of 2-propylpent-4-enoic acid (4-ene VPA), a toxic metabolite of valproic acid, by electrospray tandem mass spectrometry. J. Mass Spectrom. 31, 926-936. (31) Tang, W., Stearns, R. A., Bandiera, S. M., Zhang, Y., Raab, C., Braun, M. P., Dean, D. C., Pang, J., Leung, K. H., Doss, G. A., Strauss, J. R., Kwei, G. Y., Rushmore, T. H., Chiu, S. H., and Baillie, T. A. (1999) Studies on cytochrome P-450-mediated bioactivation of diclofenac in rats and in human hepatocytes: identification of glutathione conjugated metabolites. Drug Metab. Dispos. 27, 365-372. (32) Yu, L. J., Chen, Y., Deninno, M. P., O’Connell, T. N., and Hop, C. E. (2005) Identification of a novel glutathione adduct of diclofenac, 4′-hydroxy-2′-glutathion-deschloro-diclofenac, upon incubation with human liver microsomes. Drug Metab. Dispos. 33, 484-488. (33) Tang, W., Stearns, R. A., Wang, R. W., Chiu, S. H., and Baillie, T. A. (1999) Roles of human hepatic cytochrome P450s 2C9 and 3A4 in the metabolic activation of diclofenac. Chem. Res. Toxicol. 12, 192199. (34) Guengerich, F. P. (2001) Common and uncommon cytochrome P450 reactions related to metabolism and chemical toxicity. Chem. Res. Toxicol. 14, 611-650. (35) Uetrecht, J. (2003) In Drug Metabolizing Enzymes. Cytocgrine P450 and other Enzymes in Drug DiscoVery and DeVelopment. (Lee, J. S., Obach, R. S., and Fisher, M. B., Eds), pp 87-144, FontisMedia S.A. and Marcel Dekker, Inc., Lausanne, Switzerland. (36) Nielsen, K. K., Flinois, J. P., Beaune, P., and Brosen, K. (1996) The biotransformation of clomipramine in vitro, identification of the cytochrome P450s responsible for the separate metabolic pathways. J. Pharmacol. Exp. Ther. 277, 1659-1664. (37) Masubuchi, Y., Igarashi, S., Suzuki, T., Horie, T., and Narimatsu, S. (1996) Imipramine-induced inactivation of a cytochrome P450 2D enzyme in rat liver microsomes: in relation to covalent binding of its reactive intermediate. J. Pharmacol. Exp. Ther. 279, 724-731. (38) Masubuchi, Y., Saito, H., and Horie, T. (1998) Structural requirements for the hepatotoxicity of nonsteroidal anti-inflammatory drugs in isolated rat hepatocytes. J. Pharmacol. Exp. Ther. 287, 208-213. (39) Sato, J., Yamane, Y., Ito, K., and Bando, H. (1993) Structures of mefenamic acid metabolites from human urine. Biol. Pharm. Bull. 16, 811-812. (40) Sato, J., Kudo, N., Owada, E., Ito, K., Niida, Y., Umetsu, M., and Kikuta, T. (1997) Urinary excretion of mefenamic acid and its metabolites including their esterglucuronides in preterm infants undergoing mefenamic acid therapy. Biol. Pharm. Bull. 20, 443-445. (41) Tang, W., and Abbott, F. S. (1996) Bioactivation of a toxic metabolite of valproic acid, (E)-2-propyl-2,4-pentadienoic acid, via glucuronidation. LC/MS/MS characterization of the GSH-glucuronide diconjugates. Chem. Res. Toxicol. 9, 517-526. (42) Kalapos, M. P. (2002) Carbamazepine-provoked hepatotoxicity and possible aetiopathological role of glutathione in the events. Retrospective review of old data and call for new investigation. AdVerse Drug React. Toxicol. ReV. 21, 123-141.

766 Chem. Res. Toxicol., Vol. 20, No. 5, 2007 (43) Madden, S., Maggs, J. L., and Park, B. K. (1996) Bioactivation of carbamazepine in the rat in vivo. Evidence for the formation of reactive arene oxide(s). Drug Metab. Dispos. 24, 469-479. (44) Amore, B. M., Kalhorn, T. F., Skiles, G. L., Hunter, A. P., Bennett, G. D., Finnell, R. H., Nelson, S. D., and Slattery, J. T. (1997) Characterization of carbamazepine metabolism in a mouse model of carbamazepine teratogenicity. Drug Metab. Dispos. 25, 953-962. (45) Khojasteh-Bakht, S. C., Chen, W., Koenigs, L. L., Peter, R. M., and Nelson, S. D. (1999) Metabolism of (R)-(+)-pulegone and (R)-

Zheng et al. (+)-menthofuran by human liver cytochrome P-450s: evidence for formation of a furan epoxide. Drug Metab. Dispos. 27, 574580. (46) Gan, J., Harper, T. W., Hsueh, M. M., Qu, Q., and Humphreys, W. G. (2005) Dansyl glutathione as a trapping agent for the quantitative estimation and identification of reactive metabolites. Chem. Res. Toxicol. 18, 896-903.

TX600277Y

Screening and Identification of GSH-Trapped Reactive Metabolites

Recommend Documents