sulfoxidation of cysteine and mercapturic acid conjugates of the sevoflurane degradation product...

Sulfoxidation of Cysteine and Mercapturic AcidConjugates of the Sevoflurane Degradation Product

Fluoromethyl-2,2-difluoro-1-(trifluoromethyl)vinyl Ether(Compound A)

T. Gul Altuntas,†,‡ Sang B. Park,‡ and Evan D. Kharasch*,‡,§

Department of Pharmaceutical Chemistry, Faculty of Pharmacy, Ankara University,Ankara, Turkey, and Departments of Anesthesiology and Medicinal Chemistry,

University of Washington, Seattle, Washington

Received December 6, 2003

The volatile anesthetic sevoflurane is degraded in anesthesia machines to the haloalkenefluoromethyl-2,2-difluoro-1-(trifluoromethyl)vinyl ether (FDVE), which can cause renal andhepatic toxicity in rats. FDVE is metabolized to S-[1,1-difluoro-2-fluoromethoxy-2-(trifluoro-methyl)ethyl]-L-cysteine (DFEC) and (E) and (Z)-S-[1-fluoro-2-fluoromethoxy-2-(trifluorom-ethyl)vinyl]-L-cysteine [(E,Z)-FFVC], which are N-acetylated to N-Ac-DFEC and (E,Z)-N-Ac-FFVC S-conjugates. Some haloalkene S-conjugates undergo sulfoxidation. This investigationtested the hypothesis that FDVE S-conjugates can also undergo sulfoxidation, by evaluatingsulfoxide formation by human and rat liver and kidney microsomes and expressed P450s andflavin monooxygenases. Rat, and at lower rates human, liver microsomes oxidized (Z)-N-Ac-FFVC and N-Ac-DFEC to the corresponding sulfoxides. Much lower rates of (Z)-N-Ac-FFVC,but not N-Ac-DFEC, sulfoxidation occurred with rat and human kidney microsomes. In humanliver microsomes, the P450 inhibitor 1-aminobenzotriazole completely inhibited S-oxidation,while heating to inactivate FMO decreased (Z)-N-Ac-FFVC and N-Ac-DFEC sulfoxidation only0 and 30%, respectively. Of the various cytochrome P450s examined, P450s 3A4 and 3A5 hadthe highest S-oxidase activity toward (Z)-N-Ac-FFVC; P450 3A4 was the predominant enzymeforming N-Ac-DFEC-SO. The P450 3A inhibitors troleandomycin and ketoconazole inhibited>95% of (Z)-N-Ac-FFVC sulfoxidation by P450 3A4 and 3A5 and 40-100% of (Z)-N-Ac-FFVCsulfoxidation by human liver microsomes and 15-85% of N-Ac-DFEC sulfoxidation by humanliver microsomes. Sulfoxidation of DFEC was also examined in human liver microsomes.Substantial amounts of sulfoxide were observed, even in the absence of NADPH or protein,while enzymatic formation was comparatively minimal. These results show that FDVES-conjugates undergo P450-catalyzed and nonenzymatic sulfoxidation and that enzymaticsulfoxidation of (Z)-N-Ac-FFVC and N-Ac-DFEC is catalyzed predominantly by P450 3A. Theextent of FDVE sulfoxidation in vivo and the toxicologic significance of FDVE sulfoxides remainunknown and merit further investigation.

Introduction

FDVE1 (referred to as “compound A” in the sevofluranelabeling) (Figure 1; 1) is the major degradation product

of sevoflurane formed via base-catalyzed dehydrofluori-nation by the carbon dioxide absorbents in anesthesiamachines (1, 2). FDVE is nephrotoxic when administeredto rats by inhalation or intraperitoneal injection (3-8).Several other chlorinated and fluorinated alkenes arenephrotoxic, and their nephrotoxocity is associated witha multistep pathway that includes hepatic glutathioneS-conjugate formation, enzymatic hydrolysis of the glu-tathione S-conjugates to cysteine S-conjugates, renal up-take of cysteine S-conjugates, and bioactivation by renalcysteine S-conjugate â-lyase to reactive species, whosereaction with cellular proteins is associated with celldamage and death (9-11).

FDVE undergoes enzymatic and nonenzymatic conju-gation with GSH to form several FDVE-GSH conjugates,subsequent conversion to the corresponding FDVE-cysteine and -mercapturic acid conjugates, and bioacti-vation of the cysteine conjugates by renal â-lyase. Suchconjugation and metabolism have been established inboth rats and humans. In rats, in vivo, FDVE undergoes

* To whom correspondence should be addressed. Tel: 206-543-4070.Fax: 206-685-3079. E-mail: [email protected].

† Ankara University.‡ Department of Anesthesiology, University of Washington.§ Department of Medicinal Chemistry, University of Washington.1 Abbreviations: FDVE, fluoromethyl-2,2-difluoro-1-(trifluorometh-

yl)vinyl ether; GSH, glutathione; GST, glutathione-S-transferase; GGT/DP, γ-glutamyltransferase/dipeptidase; FMO, flavin-containing mono-oxygenase; P450, cytochrome P450; DFEG, S-(1,1-difluoro-2-fluo-romethoxy-2-(trifluoromethyl)ethyl)glutathione; (E,Z)-FFVG, (E,Z)-S-(1-fluoro-2-fluoromethoxy-2-(trifluoromethyl)vinyl)glutathione; DFEC,S-(1,1-difluoro-2-fluoromethoxy-2-(trifluoromethyl)ethyl)-L-cysteine; (E,Z)-FFVC, (E,Z)-S-(1-fluoro-2-fluoromethoxy-2-(trifluoromethyl)vinyl)-L-cysteine; N-Ac-DFEC, N-acetyl-S-(1,1-difluoro-2-fluoromethoxy-2-(trifluoromethyl)ethyl)-L-cysteine; (E,Z)-N-Ac-FFVC, (E,Z)-N-acetyl-S-(1-fluoro-2-fluoromethoxy-2-(trifluoromethyl)vinyl)-L-cysteine; DFEC-SO, S-[1,1-difluoro-2-fluoromethoxy-2-(trifluoromethyl)ethyl]-L-cysteinesulfoxide; (Z)-N-Ac-FFVC-SO, (Z)-N-acetyl-S-[1-fluoro-2-fluoromethoxy-2-(trifluoromethyl)vinyl]-L-cysteine sulfoxide; N-Ac-DFEC-SO, N-acetyl-S-[1,1-difluoro-2-fluoromethoxy-2-(trifluoromethyl)ethyl]-L-cysteine sul-foxide.

435Chem. Res. Toxicol. 2004, 17, 435-445

10.1021/tx034254k CCC: $27.50 © 2004 American Chemical SocietyPublished on Web 02/27/2004

reaction with GSH to form (R)- and (S)-DFEG (Figure 1;2) and (E)- and (Z)-FFVG (Figure 1; 3), which undergocleavage to the corresponding cysteine S-conjugates(Figure 1; 4, 5) (12-14). In rats, N-acetylation forms themercapturates, (R)- and (S)-N-Ac-DFEC (Figure 1; 6) and(E)- and (Z)-N-Ac-FFVC (Figure 1; 7), which are excretedin urine, as identified by ionspray LC-MS/MS, 19F NMR,and selected ion mode GC-MS (12, 14, 15). The cysteineS-conjugates are also metabolized by rat renal â-lyasein vitro and in vivo to reactive intermediates, which maybind to cellular macromolecules or undergo hydrolysis to3,3,3-trifluoro-2-(fluoromethoxy)propanoic acid (Figure 1;8) (14-19). The latter has been identified in rat urineby 19F NMR and GC-MS, establishing â-lyase-catalyzedmetabolism of FDVE-cysteine conjugates in rats in vivo(14, 16).

In human subcellular fractions in vitro, includinghepatic and renal microsomes and cytosol and blood,FDVE also undergoes conjugation to form four GSHconjugates (20). FDVE-cysteine S-conjugates are me-tabolized in vitro to their corresponding mercapturates(6, 7) by human kidney cytosol, microsomes, and mito-chondria (17, 21), and mercapturates are also deacety-lated to the corresponding cyteine S-conjugates by humanrenal cytosol in vitro (21). FDVE mercapturates and3,3,3-trifluoro-2-(fluoromethoxy)propanoic acid have beenidentified by 19F NMR and GC-MS in the urine of patientsexposed to FDVE while undergoing sevoflurane anesthe-sia (15, 22). These results demonstrated that FDVEundergoes GSH conjugation, conversion to correspondingcysteine S-conjugates, and renal â-lyase-catalyzed me-tabolism in humans in vivo.

A novel pathway of haloalkene S-conjugates bioacti-vation and toxification, involving rat, rabbit, and/orhuman hepatic microsomal sulfoxidation of cysteine andmercapturic acid conjugates of dichloropropene, hexachlo-robutadiene, trichloroethene, and tetrachloroethene hasbeen identified (23-28). In general, S-conjugates sulfoxi-dation may be mediated by P450 or flavin monooxyge-nases. For example, sulfoxidation of S-allyl-L-cysteine andS-benzyl-L-cysteine, and to a lesser extent S-(1,2-dichlo-rovinyl)-L-cysteine and S-(1,2,2-trichlorovinyl)-L-cysteine,was catalyzed by flavin monooxygenases (29, 30). Incontrast, N-acetyl-S-(pentachlorobutadienyl)-L-cysteine,N-acetyl-S-(1,2,2-trichlorovinyl)-L-cysteine, N-acetyl-S-(1,2-dichlorovinyl)-L-cysteine, and N-acetyl-S-(2,2-dichlo-rovinyl)-L-cysteine sulfoxidation were catalyzed mainlyby P450 (25-27). Sulfoxidation of haloalkyl cysteineS-conjugates can constitute a toxification pathway, whichis independent of â-lyase-mediated bioactivation (24-28).

It is unknown, however, whether FDVE-cysteine andFDVE-mercapturic acid conjugates can undergo sulfoxi-dation. Therefore, the objective of this investigation wasto test the hypothesis that FDVE-cysteine and -mer-capturic acid S-conjugates undergo metabolism in humanand rat liver and kidney microsomes to novel sulfoxidemetabolites, to determine whether species and tissuedifferences exist in this oxidative process and to elucidatethe enzyme(s) involved.

Experimental Procedures

Materials. NADPH, troleandomycin, ketoconazole, meth-imazole, and 1-aminobenzotriazole were obtained from Sigma-

Figure 1. Pathways of FDVE metabolism in rats and humans. (1) FDVE; (2) DFEG; (3) (E) and (Z)-FFVG; (4) DFEC; (5) (E) and(Z)-FFVC; (6) N-Ac-DFEC; (7) (E) and (Z)-N-Ac-FFVC; (8) 3,3,3-trifluoro-2-(fluoromethoxy)propanoic acid; and (9) 3,3,3-trifluorolacticacid.

436 Chem. Res. Toxicol., Vol. 17, No. 3, 2004 Altuntas et al.

Aldrich Co. (St. Louis, MO). DFEC, (Z)-N-Ac-FFVC, and N-Ac-DFEC were synthesized as previously described (16). Humanliver and kidney tissues medically unsuitable for transplantwere obtained from the University of Washington Human LiverBank and the National Disease Research Interchange, respec-tively. Microsomes were prepared from thawed specimens asdescribed previously (31) and stored at -80 °C until required.Microsomal protein concentrations were measured by themethod of Lowry et al. (32) with bovine serum albumin as thestandard. Animal experiments were approved by the Universityof Washington Animal Use Committee in accordance with theAmerican Association for Accreditation of Laboratory AnimalCare guidelines. Male Fisher 344 rats (220-240 g, Madison, WI)were treated with 0.1% phenobarbitol in the drinking water for10 days. Liver and kidney microsomes were prepared thefollowing day. Microsomes from baculovirus-transfected insectcells (Supersomes) expressing human P450 1A1, 1A2, 2A6, 2B6,2C9, 2C19, 2D6, 2E1, 3A4, and 3A5, rat P450 3A1 and 3A2,and human FMO1, FMO3, and FMO5 were obtained from BDGentest Co. (Woburn, MA). P450 3A4 and P450 3A5 were alsopurchased from PanVera Co. (Madison, WI). All other reagentswere obtained from commercial suppliers and used withoutfurther purification.

Synthesis of DFEC-SO. DFEC (100 mg) was stirred withhydrogen peroxide (30%, 0.037 mL) in 3 mL of trifluoroaceticacid at 4 °C for 1 h and then at 25 °C. The progress of thereaction was monitored by HPLC. After the starting materialwas completely consumed, the solvent was removed in vacuoand the product was precipitated by the addition of diethyl etherto give 82 mg of white solid. 1H NMR spectra were recorded ona Varian XL-400 spectrometer, and chemical shifts were refer-enced to methanol (δ 3.3 for -CH3). 1H NMR (CD3OD): δ5.55(d of d, 4/3H, J ) 53 Hz, -FCH2O), 5.52 (d of d, 2/3H, J ) 53Hz, -FCH2O), 5.47 (m, 1H, -F2C-CH-CF3), 4.32 (m, 1H, RH),3.75 (m, 1H, OS-CH-C), 3.50 (m, 1H, OS-CH-C). LC-ESI:318 [M + H]+, 340 [M + Na]+.

Synthesis of (Z)-N-Ac-FFVC-SO. (Z)-N-Ac-FFVC (20 mg)was dissolved in 1 mL of trifluoroacetic acid, and then, hydrogenperoxide (30%, 0.008 mL) was added at 4 °C. The mixture wasstirred for 1 h at 4 °C and then at 25 °C. The progress of thereaction was monitored by HPLC. After the starting materialwas completely consumed, the solvent was removed in vacuo.The residue was precipitated in diethyl ether and purified byHPLC to give 11 mg of product. 1H NMR (CD3OD): δ 5.61 (d ofd, 1H, J ) 52 Hz, FCH2O), 5.50 (d of d, 1H, J ) 39 Hz, FCH2O),4.81 (m, 1H, RH), 3.79 (m, 1H, OS-CH-C), 3.34-3.62 (m, 1H,OS-CH-C), 1.99 and 2.01 (2 singlets, 3Hs, -CH3). LC-ESI: 340[M + H]+, 362 [M + Na]+, 130 (-CH2CHNH(Ac)COOH).

Synthesis of N-Ac-DFEC-SO. N-Ac-DFEC (15 mg) wasdissolved in 1 mL of trifluoroacetic acid, hydrogen peroxide (30%,0.001 mL) was added at 4 °C, and the mixture was stirred. Theprogress of the reaction was monitored by HPLC. After thestarting material was completely consumed, the solvent wasremoved in vacuo. The residue was precipitated in diethyl etherand purified by HPLC to give 12 mg of product. 1H NMR(CD3OD): δ 5.35-5.62 (m, 3Hs, J ) 52 Hz, FCH2O and -F2C-CH-CF3), 4.82-4.95 (m, 1H, RH), 3.57-3.75 (m, 1H, OS-CH-C), 3.32-3.49 (m, 1H, OS-CH-C), 2.01-2.03 (m, 3Hs, -CH3).LC-ESI: 360 [M + H]+, 382 [M + Na]+, 130 (-CH2CHNH(Ac)-COOH).

Enzyme Systems Involved in FDVE-Mercapturic Acidand -Cysteine Sulfoxidation. Incubation Conditions.Preliminary experiments showed that liver microsomal sulfoxideformation was linear for up to 60 min, 0.1-2 mM (Z)-N-Ac-FFVC, and 1-4 mg/mL protein. Routine incubations (0.25 mL)contained microsomes (4.0 mg/mL protein) and NADPH (2 mM)in 0.1 M potassium phosphate buffer (pH 7.4) at 37 °C, andreactions were started by the addition of 2 mM substrate.Control reactions lacking NADPH, protein, or substrate wererun in parallel. Reactions were terminated after 30 min with20% perchloric acid and placed on ice, vortexed, and centrifugedfor 15 min at 3000 rpm to remove precipitated proteins.

Supernatants (5 µL) were analyzed directly by LC-MS. Assayswith expressed human P450s (1A1, 1A2, 2A6, 2B6, 2C9, 2C19,2D6, 2E1, 3A4, and 3A5), rat P450 3A1 and 3A2 and FMOs(FMO1, FMO3, and FMO5), and (Z)-N-Ac-FFVC and N-Ac-DFEC were done in the same manner as with microsomes,except 10 pmol/mL P450 or 200 µg/mL FMO was used insteadof microsomes. To discriminate between P450 enzymes andFMOs, microsomes were heated for 5 min at 45 °C in theabsence of NADPH to inactivate FMO and placed on ice. AfterNADPH was added and there was a brief preincubation period(5 min, 37 °C), substrates were added and incubations wereperformed as described above.

The mechanism-based P450 3A inhibitor troleandomycin (10and 100 µM) and the nonselective P450 inhibitor 1-aminoben-zotriazole (0.5 mM) (33) were preincubated with human livermicrosomes in potassium phosphate buffer (pH 7.4) and NADPH(2 mM) for 15 min at 37 °C, and then, substrates (2 mM) wereadded to start the 30 min reaction. Competitive inhibitors keto-conazole (1 and 5 µM) and methimazole (1 mM) were coincu-bated with substrate and human liver microsomes in potassiumphosphate buffer (pH 7.4) at 37 °C for 5 min, and the reactionwas initiated by the addition of NADPH. After 30 min, thereaction was terminated as above. The effect of troleandomycin(30 and 100 µM) and ketoconazole (1 and 5 µM) on the formationof (Z)-N-Ac-FFVC-SO was also determined with cDNA expressedP450 3A4 and P450 3A5 isoforms (10 pmol/mL). All inhibitorswere diluted in methanol (final methanol concentration 1%).

To generate samples for LC-MS/MS analysis, mixtures (5 mL)containing 0.1 M potassium phosphate buffer (pH 7.4), phe-nobarbital-induced rat liver microsomes (4 mg/mL), and (Z)-N-Ac-FFVC or N-Ac-DFEC (2 mM) were preincubated at 37 °Cfor 5 min prior to adding 2 mM NADPH. After 60 min, incu-bations were quenched with the addition of 878 µL of 20%perchloric acid and placed on ice. The incubations were thencentrifuged for 15 min and extracted with diethyl ether (3 × 5mL, samples were centrifuged each time prior to removal of theorganic layer). The combined organic layers were evaporatedto dryness at 40 °C using a TurboVap LV evaporator (Zymark,Hopkinton, MA). The resulting residue was kept at -20 °C untilanalysis. At the time of analysis, the samples were reconstitutedwith 20% acetonitrile in water.

Analytical Methods. 1. LC-MS. The LC-MS system (1100Series MSD, Agilent Technologies, Palo Alto, CA) consisted ofa binary solvent delivery system, autosampler, Supelcosil LC-18-DB C18 reverse phase HPLC column (150 mm × 3 mm × 3µm) (Supelco Co., Bellefonte, PA), and quadrupole mass spec-trometer equipped with an electrospray interface and operatedin the positive ionization mode. The mass spectrometer interfacewas maintained at 325 °C, with a nitrogen nebulization pressureof 25 psi and a flow of 10 L/min. The gradient mobile phase(0.5 mL/min) was water (0.05% TFA):acetonitrile (0.05% TFA)(90:10) for 1 min, increased to 33% acetonitrile over 6 min andheld for 2 min, and then increased to 35% acetonitrile over 0.5min and held for 2.5 min. The column was briefly washed with90% acetonitrile and reequilibrated back to 10% acetonitrile.Using the above conditions, the retention time of (Z)-N-Ac-FFVC-SO was 7.5 min, the two diastereomeric peaks for N-Ac-DFEC-SO eluted at 8.2 and 8.5 min, and DFEC-SO eluted asthree peaks at 5.3, 5.5, and 5.9 min, which were presumed torepresent diastereomers and quantified together. (Z)-N-Ac-FFVC-SO, N-Ac-DFEC-SO, and DFEC-SO were quantified byselected ion monitoring ([M + H]+ m/z 340, 360, and 318,respectively) using standard curves (r2 > 0.99) of peak area vsconcentration generated using synthetic sulfoxide standards.Limits of quantification were 20, 20, and 8 ng/mL for the (Z)-N-Ac-FFVC-SO, N-Ac-DFEC-SO, and DFEC-SO, respectively.Unless otherwise indicated, both N-Ac-DFEC-SO diastereomerswere quantified together and formation rates were reported asthe sum.

2. LC-MS/MS. Accurate mass verification of the syntheticstandards and metabolically generated (Z)-N-Ac-FFVC-SO andN-Ac-DFEC-SO was done using a quadrupole time-of-flight

Sulfoxidation of Haloalkene S-Conjugates Chem. Res. Toxicol., Vol. 17, No. 3, 2004 437

tandem hybrid mass spectrometer (QTOF, Waters-Micromass,Manchester, U.K.) equipped with the CapLC system (Waters,Milford, MA). The stream select module was configured withan OPTI-PAK Symmetry 300 C18 trap column (Waters) con-nected in series with a nanoscale analytical column (75 µm i.d.× 15 cm, packed with 5 µm Jupiter C18 particles (Phenomenex,Torrance, CA). The samples (5 µL) were injected onto the trapcolumn at 10 µL/min, desalted, and back-flushed to the analyti-cal column at 0.5 µL/min using gradient elution. The gradientstarted at 5% B for 5 min and then went from 5 to 90% B in 5min, followed by 90% B for 35 min (A ) 5% acetonitrile and0.1% formic acid in water; B ) 95% acetonitrile with 0.1% formicacid in water). The (Z)-N-Ac-FFVC-SO eluted at 17.2 min, andthe N-Ac-DFEC-SO eluted in two peaks at 13.4 and 15.8 min.The QTOF parameters were set as follows: the electrospraypotential was set to 3.5 kV, the cone voltage was set to 40 V,the extraction cone was set to 2 V, and the source temperaturewas set to 80 °C. The instrument was operated in the MS/MSmode with the quadrupole isolation width set to include onlythe monoisotopic peak of each compound (the low mass resolu-tion and high mass resolution parameters were set to 15resulting in a mass window of 1 m/z centered about the [M +H]+ ion). The TOF scan range was from m/z 10-500, and thecollision energy was set to 5 eV. The instrument was tuned toobtain a resolving power of 6000 for the corresponding massrange.

Results

Formation of Sulfoxide Metabolites in Vitro.When the mercapturates (Z)-N-Ac-FFVC and N-Ac-DFEC were incubated with human liver and kidneymicrosomes and NADPH, the corresponding sulfoxides,(Z)-N-Ac-FFVC-SO and N-Ac-DFEC-SO, were detected(Figure 2). The identity of the sulfoxides resulting frommetabolism of (Z)-N-Ac-FFVC and N-Ac-DFEC was veri-fied by consistent retention times, mass spectral char-acteristics, and the mass accuracy of the protonatedmonoisotopic molecular ion when compared to those ofthe synthetic compounds using electrospray MS/MS(Figures 3 and 4, respectively). N-Ac-DFEC-SO eluted astwo peaks (Figure 2), whose mass spectra were identical(Figure 4), presumed to represent two diastereomers.Further structural identification of the diastereomers wasnot pursued.

Incubation of DFEC with liver microsomes formedDFEC-SO. The identity of this metabolite was confirmedusing electrospray LC-MS by comparison with the reten-tion time and spectrum of the synthetic compound(Figure 5).

Reaction components required for sulfoxidation of (Z)-N-Ac-FFVC, N-Ac-DFEC, and DFEC were evaluatedusing human liver microsomes (Table 1). Sulfoxidationof both (Z)-N-Ac-FFVC and N-Ac-DFEC was negligiblein the absence of NADPH or microsomal protein. Incontrast, sulfoxidation of the DFEC S-conjugate at 37 °Cproceeded equally well in the absence or presence ofmicrosomal protein, and substantial amounts of DFEC-SO were formed without NADPH. Minimal DFEC sul-foxidation was observed at room temperature. Theseresults suggested that (Z)-N-Ac-FFVC-SO and N-Ac-DFEC-SO formation was enzymatic, while DFEC sul-foxidation occurred nonenzymatically.

Species and Tissue Differences in the Formationof (Z)-N-Ac-FFVC-SO and N-Ac-DFEC-SO. Formationof (Z)-N-Ac-FFVC-SO and N-Ac-DFEC-SO was comparedin human and rat liver and kidney microsomes (Figure6). Both (Z)-N-Ac-FFVC-SO and N-Ac-DFEC-SO wereformed by human liver microsomes, and considerablevariability was observed between the three livers. Both(Z)-N-Ac-FFVC-SO and N-Ac-DFEC-SO were also formedby rat liver microsomes, although the rates were sub-stantially greater (2-30-fold) than with human livermicrosomes (Figure 6A). Phenobarbital induction ap-proximately doubled rat liver microsomal formation ofboth (Z)-N-Ac-FFVC-SO and N-Ac-DFEC-SO (Table 1).

Human kidney microsomes catalyzed the formation of(Z)-N-Ac-FFVC-SO. Rat kidney microsomes also cata-lyzed the sulfoxidation of (Z)-N-Ac-FFVC, at rates com-parable to human liver microsomes. In contrast to humanand rat livers, there was no difference in human and ratkidney (Z)-N-Ac-FFVC-SO formation. Phenobarbital in-duction had little effect on rat kidney microsomal (Z)-N-Ac-FFVC sulfoxidation. Any formation of N-Ac-DFEC-SO by either human or rat kidney microsomes, if at allpresent, was below the limit of quantification.

Overall, FDVE mercapturate sulfoxidation was sub-stantially greater in hepatic as compared with renalmicrosomes. FDVE mercapturate sulfoxidation was greaterin rat than human livers.

Enzymes Catalyzing (Z)-N-Ac-FFVC, N-Ac-DFEC,and DFEC Sulfoxidation. Thermal inactivation of heatlabile FMO while leaving P450 activities unaffected,accomplished by incubating microsomes for 5 min at 45°C in the absence of the NADPH, is an excellent methodto discriminate between the participation of FMO andP450 enzymes in microsomal reactions (34). Heat inac-tivation of human liver microsomes resulted in no reduc-tion in sulfoxide formation from (Z)-N-Ac-FFVC and onlya 30% reduction in sulfoxide formation from N-Ac-DFEC(Table 2). The nonselective P450 inhibitor 1-aminoben-zotriazole essentially prevented (Z)-N-Ac-FFVC-SO andN-Ac-DFEC-SO formation. The FMO alternate substrateinhibitor methimazole decreased (Z)-N-Ac-FFVC andN-Ac-DFEC sulfoxidation 60-80%. In contrast to (Z)-N-Ac-FFVC and N-Ac-DFEC sulfoxidation, 1-aminobenzo-triazole and methimazole decreased the formation ofDFEC-SO by only 25 and 15%, respectively (Table 2).Together, these results suggested a predominant role forP450 enzymes in (Z)-N-Ac-FFVC-SO and N-Ac-DFEC-

Figure 2. Electrospray LC-MS chromatograms obtained byselected ion monitoring. (A) Synthetic standard of (Z)-N-Ac-FFVC-SO, monitored at m/z 340. (B) Synthetic standard of N-Ac-DFEC-SO, monitored at m/z 360. Peaks 1 and 2 eluted at 8.2and 8.5 min, respectively. The area ratio of peak 1 to peak 2 inthe synthetic standard was 0.3, and it was 1.4-2.3 in enzymatic-ally generated samples.


SO formation and confirmed the nonenymatic nature ofDFEC sulfoxidation.

(Z)-N-Ac-FFVC and N-Ac-DFEC SulfoxidationCatalyzed by Human and Rat Expressed 450s andFMOs. Because the use of P450 and FMO inhibitorssuggested that P450 enzymes mediate the majority of (Z)-N-Ac-FFVC and N-Ac-DFEC sulfoxidation, additional

experiments were directed toward identifying the P450enzyme(s) involved. Therefore, 10 different expressedhuman P450 isoforms (1A1, 1A2, 2A6, 2B6, 2C9, 2C19,2D6, 2E1, 3A4, and 3A5) and three human FMO isoforms(FMO1, FMO3, and FMO5) were evaluated (Figure 7).P450 3A4 catalyzed the greatest (Z)-N-Ac-FFVC sulfoxi-dation, which was markedly enhanced by coexpressed

Figure 3. Electrospray LC-MS/MS spectra of (Z)-N-Ac-FFVC-SO obtained with a QTOF mass spectrometer. (A) Synthetic standardof (Z)-N-Ac-FFVC-SO (mass accuracy for the measured protonated monoisotopic molecular ion, m/z 340, was 21 ppm). (B) Extractedincubation of phenobarbital-induced rat liver microsomes with (Z)-N-Ac-FFVC as substrate (mass accuracy for the measured protonatedmonoisotopic molecular ion was 9 ppm).

Figure 4. Electrospray LC-MS/MS spectra of N-Ac-DFEC-SO obtained with a QTOF mass spectrometer. (A) Synthetic standard ofN-Ac-DFEC-SO (mass accuracy for the measured protonated monoisotopic molecular ion, m/z 360, was 17 ppm). (B) Spectrum of thefirst chromatographic peak of N-Ac-DFEC-SO formed by incubation of phenobarbital-induced rat liver microsomes with N-Ac-DFECas substrate (mass accuracy for the measured protonated monoisotopic molecular ion was 33 ppm). (C) Spectrum obtained from thesecond chromatographic peak of N-Ac-DFEC-SO formed by incubation of phenobarbital-induced rat liver microsomes with N-Ac-DFEC as substrate (mass accuracy for the measured protonated monoisotopic molecular ion was 31 ppm). Additional backgroundions are visible due to the low peak intensity.


cytochrome b5. P450 3A5 also catalyzed (Z)-N-Ac-FFVCsulfoxidation, at a rate approximately half that of P4503A4. P450 3A5 with coexpressed b5 was not availablefor direct comparison to P450 3A4 with coexpressed b5.P450 3A enzymes (without coexpressed b5) obtained from

a second source (PanVera) confirmed that P450s 3A5, andmore so 3A4, had significant (Z)-N-Ac-FFVC sulfoxida-tion activity. The other P450 isoforms catalyzed com-paratively minimal (Z)-N-Ac-FFVC sulfoxidation, andFMOs were relatively inactive. N-Ac-DFEC sulfoxidationby expressed enzymes was substantially less than thatof (Z)-N-Ac-FFVC. Essentially, only human P450s 3A4and 3A5 catalyzed the sulfoxidation of N-Ac-DFEC. Noor negligible (Z)-N-Ac-FFVC and N-Ac-DFEC sulfoxida-tion was detected in incubations without NADPH.

Comparison of the P450 3A-catalyzed sulfoxidation of(Z)-N-Ac-FFVC and N-Ac-DFEC with rat and humanenzymes is shown in Figure 8. With coexpressed b5, P4503A4-catalyzed (Z)-N-Ac-FFVC sulfoxidation was 6-foldgreater than that of N-Ac-DFEC. Without b5, N-Ac-DFEC sulfoxidation by P450s 3A4 and 3A5 was substan-tially less than that of (Z)-N-Ac-FFVC (using enzymefrom PanVera) or undetectable (using enzyme from BDGentest). Overall rates of sulfoxidation were greater withPanVera as compared with BD Gentest enzyme. Ratesof (Z)-N-Ac-FFVC sulfoxidation by human P450 3A4 werecomparable to those by rat P450s 3A1 and 3A2. Incontrast, whereas human P450s 3A4 and 3A5 catalyzednegligible N-Ac-DFEC sulfoxidation, rat P450s 3A1 and3A2 formed N-Ac-DFEC-SO at rates equal to (Z)-N-Ac-FFVC. Thus, there are apparent species differences inP450 3A-catalyzed sulfoxidation of N-Ac-DFEC but not(Z)-N-Ac-FFVC.

Stereochemical aspects of N-Ac-DFEC sulfoxidationwere evaluated using liver microsomes and expressedP450s. N-Ac-DFEC-SO eluted as two diastereomericpeaks with identical mass spectra (Figure 2). The peak1:peak 2 area ratio was 0.3 in synthetic standards and>1 in enzymatically generated samples (containing suf-ficient amounts for quantification) (Table 3). The peak1:2 area ratio was 2.0-2.5 in human liver microsomesand expressed P450 3A4 and 1.3-1.7 in rat liver mi-crosomes and expressed P450 3A. Thus, there is anadditional species difference in FDVE conjugates sul-foxidation.

Inhibition of FDVE Mercapturates Sulfoxidationby Troleandomycin and Ketoconazole. (Z)-N-Ac-FFVC sulfoxide formation by expressed P450s 3A4 and3A5 was completely inhibited by the mechanism-basedP450 3A inhibitor troleandomycin (Figure 9). The com-petitive inhibitor ketoconazole reduced P450 3A4- andP450 3A5-catalyzed (Z)-N-Ac-FFVC-SO formation 70-95% and 35-95%, respectively. Human liver microsomal(Z)-N-Ac-FFVC sulfoxidation was inhibited 85-100% bytroleandomycin and 80-90% by ketoconazole at thehighest inhibitor concentrations (Figure 10A). Microso-mal N-Ac-DFEC sulfoxidation was generally inhibited70-80% by troleandomycin and ketoconazole (Figure10B), although the liver with the lowest uninhibited rate(human liver no. 140) showed little inhibition by trole-andomycin.

Discussion

The first objective of this investigation was to test thehypothesis that FDVE-cysteine and -mercapturic acidS-conjugates undergo metabolism in human and rat liverand kidney microsomes to sulfoxide metabolites. Theresults show that liver and kidney microsomes from ratsand humans are capable of oxidizing the mercapturic acidconjugates (Z)-N-Ac-FFVC and N-Ac-DFEC to their cor-

Figure 5. Electrospray LC-MS analysis of DFEC-SO. (A)Chromatogram obtained by selected ion monitoring of m/z 318and (B) spectrum from a synthetic standard of DFEC-SO.

Table 1. Effect of Phenobarbital Induction on FDVEMercapturatic Acid Sulfoxidation by Rat Microsomesa

sulfoxide formation (pmol/mg/min)

(Z)-N-Ac-FFVC-SO N-Ac-DFEC-SO

liver (control). 32.6 ( 0.7 56.6 ( 6.8liver (phenobarbital

pretreated)122.4 ( 4.5 256.7 ( 8.0

kidney (control) 0.66 ( 0.15 NDkidney (phenobarbital

pretreated)1.20 ( 0.44 ND

a Incubations contained 2 mM substrate, 2 mM NADPH, andmicrosomes (4 mg). Results are the mean ( SD (n ) 3). ND, notdetectable or below the limit of quantification.

Figure 6. Rates of sulfoxide formation from (Z)-N-Ac-FFVCand N-Ac-DFEC by (A) human (HLM) and rat (RLM) livermicrosomes and (B) human (HKM) and rat (RKM) kidneymicrosomes. Rat tissues were from phenobarbital-induced (PB)and control (C) animals. Incubations contained 2 mM substrate,2 mM NADPH, and microsomes (4 mg). Results are the mean( SD (n ) 3). No sulfoxide formation from N-Ac-DFEC wasobserved with kidney microsomes from either humans or rats.


responding sulfoxides. In addition, the cysteine conjugateDFEC underwent facile nonenzymatic autoxidation to therespective sulfoxide. Enzymatic and nonenzymatic sul-foxidation of FDVE S-conjugates represent a novelbiotransformation pathway of FDVE, which has notpreviously been described. This is in addition to FDVE-cysteine conjugates metabolism by renal â-lyase andN-acetylation and cleavage of FDVE mercapturates byacylases (12, 14, 17, 18, 21, 22, 35). Accordingly, themetabolic scheme for FDVE can be revised to incorporatethese new routes of biotransformation (Figure 11).

(Z)-N-Ac-FFVC and N-Ac-DFEC sulfoxidation waspredominantly hepatic. (Z)-N-Ac-FFVC-SO formationwas approximately 10-fold greater (2-17 vs 0.2-1.8pmol/min/mg, respectively) in human liver as comparedwith human kidney microsomes. In uninduced and phe-nobarbital-induced rats, (Z)-N-Ac-FFVC-SO formationwas approximately 50- and 100-fold greater, respectively,in liver as compared with kidney microsomes. N-Ac-DFEC-SO formation was detected only in human andrat liver but not renal microsomes. Similar organ differ-

Table 2. Effect of Various FMO and P450 Inhibitors on the Sulfoxidation of FDVE Cysteine and Mercapturic AcidConjugates by Human Liver Microsomesa

sulfoxide formation (pmol/mg/min)

incubation conditions (Z)-N-Ac-FFVC-SO N-Ac-DFEC-SO DFEC-SO

complete system 31.8 ( 0.01 31.0 ( 0.9 20.5 ( 0.1NADPH ND ND 11.8 ( 1.4protein ND 0.39 ( 0.48 20.1 ( 0.6heat inactivation 30.8 ( 0.3 22.0 ( 1.6 15.6 ( 1.1+ methimazole (0.1 mM) 12.9 ( 0.2 5.8 ( 0.3 17.4 ( 0.1+ 1-aminobenzotriazole (0.5 mM) ND 0.49 ( 0.13 15.0 ( 0.02

a Incubations were carried out with substrate (2 mM), microsomes from human liver no. 158 (1 mg of protein), inhibitors, and NADPH(2 mM). Results are the mean ( SD of two determinations. ND, not detectable.

Figure 7. Rates of sulfoxide formation from (Z)-N-Ac-FFVCand N-Ac-DFEC by expressed P450 (CYP) and FMO isoforms.Incubations were carried out with substrate (2 mM), NADPH(2 mM), P450 (10 pmol/mL), or FMO supersomes (200 µgprotein/mL) for 30 min at 37 °C. Results are the mean ( SD (n) 3). Results for FMO supersomes are pmol/mg/min. Absoluteformation rates (pmol/min) with FMOs were comparable to orless than those with non-P4503A isoforms. Asterisks denoteenzymes from PanVera Co; all others were from BD Gentest.

Figure 8. Sulfoxidation of (Z)-N-Ac-FFVC and N-Ac-DFEC byexpressed human P450s 3A4 and 3A5 and rat P450s 3A1 and3A2. Incubations contained substrate (2 mM), NADPH (2 mM),and P450 (10 pmol/mL) (30 min, 37 °C). Results are the mean( SD (n ) 3). Asterisks denote enzymes from PanVera Co; allothers were from BD Gentest.

Table 3. Stereoselectivity in N-Ac-DFEC SulfoxidesFormation by Microsomes and Expressed P450 3Aa

species enzyme preparation peak 1/peak 2a

human liver microsomes (no. 140) 2.0 ( 0.1liver microsomes (no. 158) 2.0 ( 0.03liver microsomes (no. 167) 2.3 ( 0.2P4503A4+b5b 2.3 ( 0.05P4503A4c 2.5 ( 0.2

rat liver microsomes (control) 1.6 ( 0.01liver microsomes (phenobarbital

pretreated)1.7 ( 0.02

P4503A1b 1.4 ( 0.09P4503A2b 1.3 ( 0.02

a Incubations were carried out with N-Ac-DFEC-SO (2 mM),microsomes (4 mg/mL protein) or expressed P450 3A (10 pmol/mL), and NADPH (2 mM). Results are the mean ( SD of threedeterminations. a Results are the ratio of integrated areas for thediastereomers identified as peaks 1 and 2, shown in Figure 2.b Human and rat expressed P450 3A was obtained from BDGentest. c Human expressed P450 3A was obtained from PanVera.

Figure 9. Effect of P450 3A inhibitors on (Z)-N-Ac-FFVCsulfoxidation by expressed P450s 3A4 and 3A5. Results areexpressed as activity remaining relative to controls (withoutinhibitor). Incubations contained substrate (2 mM), P450 (10pmol/mL), NADPH (2 mM), and inhibitors [30 and 100 µMtroleandomycin (TAO); 1 and 5 µM ketoconazole]. Results arethe mean ( SD (n ) 3).


ences in N-acetyl-S-(1,2-dichlorovinyl)-L-cysteine sulfoxi-dation were also observed, whereby sulfoxidation byhuman kidney microsomes was not detected (30). Sul-foxidation of cysteine and mercapturic acid conjugatesof cis- and trans-1,3-dichloropropene was not detected inrat kidney microsomes, whereas pig liver was active inthe sulfoxidation of both cysteine S-conjugates and N-acetyl cysteine S-conjugates (23). Thus, FDVE mercap-turates, like others, undergo sulfoxidation primarily inliver.

The second objective of this investigation was to iden-tify the enzyme(s) involved in FDVE S-conjugate sulfoxi-dation. DFEC sulfoxidation was deemed predominantlynonenzymatic, since it was not dependent on microsomalprotein or NADPH and minimally affected by P450 andFMO inhibitors. Human liver microsomal sulfoxidationof (Z)-N-Ac-FFVC and N-Ac-DFEC was minimally af-fected by heat inactivation of FMO, and expressed humanFMOs 1, 3, and 5 formed negligible amounts of (Z)-N-Ac-FFVC-SO and N-Ac-DFEC-SO. These results suggestminimal involvement of FMO in FDVE mercapturatessulfoxidation. At variance with this conclusion, however,was the inhibitory effect of methimazole on FDVE mer-capturates sulfoxidation, although methimazole can af-fect P450 activity (36). Human liver microsomal sulfoxi-dation of (Z)-N-Ac-FFVC and N-Ac-DFEC was essentiallyprevented by the nonselective P450 inhibitor 1-amino-benzotriazole and substantially decreased by the P4503A inhibitors troleandomycin and ketoconazole, and ex-pressed P450 3A4 (and to a lesser extent P450 3A5)formed the greatest amounts of (Z)-N-Ac-FFVC-SO amongthe various P450 isoforms examined. These results sug-gest that P450 3A isoforms are the predominant catalysts

of human liver microsomal FDVE mercapturates sulfoxi-dation.

Rat liver microsomal (Z)-N-Ac-FFVC and N-Ac-DFECsulfoxidation was significantly enhanced by phenobar-bital induction, and expressed rat P450s 3A1 and 3A2catalyzed substantial amounts of (Z)-N-Ac-FFVC-SO andN-Ac-DFEC-SO formation. Phenobarbital induces ratP450s 2A, 2B, and 2C, in addition to P450s 3A1 and 3A2(37). However, the more selective P4503A1/2 inducerdexamethasone also increased FDVE mercapturates sul-foxidation (results not shown). These results suggest thatP450 3A isoforms are also the predominant catalysts ofrat liver microsomal FDVE mercapturates sulfoxidation.

Two aspects of expressed P450 3A-catalyzed FDVEmercapturates sulfoxidation are notable. First, (Z)-N-Ac-FFVC-SO was formed by P450 3A5, although at ratesapproximately half that of P450 3A4. This is similar topreviously reported differences in the metabolic capacitiesof P450s 3A4 and 3A5 (38). Because P450 3A5 ispolymorphically expressed (39), there may be pharma-cogenetic differences in FDVE mercapturates sulfoxida-tion. The existence of these differences, and any phar-macokinetic or toxicologic significance, remain unknown.In contrast, P450 3A5 formed negligible amounts of N-Ac-DFEC-SO. Second, P450 3A-catalyzed sulfoxidation dif-fered markedly, depending on the enzyme source, with5-10-fold greater sulfoxide formation with P450s 3A4and 3A5 obtained from PanVera as compared with BDGentest. Greater activity with PanVera enzymes resultedin detection of N-Ac-DFEC-SO, which was not observedat meaningful rates with P450s from BD Gentest. Dif-ferences in turnover may be due to coexpression of rabbitrather than human P450 reductase in PanVera ascompared with BD Gentest P4503As.

The role of P4503A1/2 and P4503A4/5 in FDVE mer-capturates sulfoxidation was consistent with previousobservations with other haloalkyl mercapturates. N-Acetyl-S-(pentachlorobutadienyl)-L-cysteine, N-acetyl-S-(1,2,2-trichlorovinyl)-L-cysteine, N-acetyl-S-(1,2-dichlo-rovinyl)-L-cysteine, and N-acetyl-S-(2,2-dichlorovinyl)-L-cysteine sulfoxidation were greater in liver microsomesfrom phenobarbital- and dexamethasone-induced rats(26, 27). N-Acetyl-S-(pentachlorobutadienyl)-L-cysteinesulfoxidation was catalyzed predominantly by humanliver microsomal and expressed P450s 3A4 and 3A5 (25).Sulfoxides formation from N-acetyl-S-(pentachlorobuta-dienyl)-L-cysteine, N-acetyl-S-(1,2,2-trichlorovinyl)-L-cys-teine, N-acetyl-S-(1,2-dichlorovinyl)-L-cysteine, and N-ace-tyl-S-(2,2-dichlorovinyl)-L-cysteine in rat liver microsomeswas catalyzed mainly by P4503A1/2 (26, 27). Thus,P4503A isoforms are, in general, the major enzymesresponsible for haloalkyl mercapturates sulfoxidation.

The relative contribution of FMOs toward microsomalcysteine S-conjugate S-oxidation clearly depends on theconjugate structure. Generally, nucleophilic sulfur atomsare oxidized preferentially by FMO, whereas nonnucleo-philic sulfur atoms are preferentially oxidized by P450(29, 40). Cysteine conjugates with more nucleophilicsulfur atoms, S-allyl-L-cysteine and S-benzyl-L-cysteine,were much better human kidney and liver and rabbitliver microsomal FMO substrates than those with lessnucleophilic sulfur atoms [S-(1,2-dichlorovinyl)-L-cysteineand S-(1,2,2-trichlorovinyl)-L-cysteine] (29, 30). This islikely attributed to the sulfur of allyl and benzyl com-pounds being more nucleophilic than that of vinyl com-pounds and the tendency for FMOs to oxidize strong

Figure 10. Effect of P4503A inhibitors on human liver mi-crosomal sulfoxidation of (A) (Z)-N-Ac-FFVC and (B) N-Ac-DFEC. Results are expressed as activity remaining relative tocontrols (without inhibitor). Incubations contained substrate (2mM), microsomes (4 mg), NADPH (2 mM), and inhibitors.Results are the mean ( SD (n ) 3).


nucleophiles (40). Lipophilicity may also affect haloalkeneS-conjugates sulfoxidation by FMO. S-Benzyl-L-cysteineis relatively lipophilic, with a nuclephilic sulfur atom, andhas been shown to be a selective substrate for FMO (41).(Z)-N-Ac-FFVC-SO has a vinylic sulfur atom as well asstrong electron-withdrawing fluorine atoms, which makethe sulfur atom much less nucleophilic than that ofS-allyl-L-cysteine, S-benzyl-L-cysteine, S-(1,2-dichlorovi-nyl)-L-cysteine, and S-(1,2,2-trichlorovinyl)-L-cysteine.(Z)-N-Ac-FFVC and N-Ac-DFEC are less lipophilic thenS-benzyl-L-cysteine, rendering them theoretically lesssusceptible to FMO sulfoxidation. This may partly ex-plain the lack of FMO activity toward (Z)-N-Ac-FFVC-SO and N-Ac-DFEC-SO formation.

The third objective of this investigation was to deter-mine whether species differences exist in FDVE S-con-jugate sulfoxidation. Although FDVE is nephrotoxic inrats (3-8), sevoflurane (the parent drug), under condi-tions in which patients are exposed to FDVE, has beenused extensively in patients without evidence of neph-rotoxicity (42-46), although nephrotoxicity in healthyvolunteers has been reported (47, 48) but not substanti-ated (49, 50). The mechanism(s) for this species differencein FDVE nephrotoxicity remains incompletely elucidated.The present results show that overall, FDVE mercaptu-rates sulfoxidation was greater in rat than humantissues. Specifically, formation of both (Z)-N-Ac-FFVC-SO and N-Ac-DFEC-SO was substantially (2-30-fold)greater in rat than human liver microsomes. In addition,N-Ac-DFEC-SO formation by rat P450s 3A1 and 3A2 wassubstantially greater than by human P450 3A4 and 3A5.Last, although the absolute configurations of the two

N-Ac-DFEC-SO diastereomers are unknown, there wasa species difference in their relative formation by bothliver microsomes and expressed P4503As. Thus, speciesdifferences in FDVE S-conjugates sulfoxidation mightexplain, in part, apparent differences in susceptibility toFDVE nephrotoxicity.

There are other known interspecies differences inFDVE metabolism and/or toxification in rats vs humans,including (i) greater rates of FDVE-GSH conjugate for-mation in hepatic microsomes and cytosol (20), (ii) greaterâ-lyase-catalyzed metabolism of FDVE-cysteine conju-gates in vitro in rat kidneys (17), (iii) greater excretionof N-Ac-DFEC relative to (E,Z)-N-Ac-FFVC in rats (18,22), (iv) greater excretion of 3,3,3-trifluoro-2-fluorometh-oxypropanoic acid (reflecting â-lyase-catalyzed FDVEcysteine conjugates metabolism) in vivo in rats (18, 22),(v) greater ratio of 3,3,3-trifluoro-2-fluoromethoxypro-panoic acid (toxification) to mercapturates (detoxication)in urine in rats (18, 22), and (vi) relative resistance ofhuman proximal tubular cells to the cytotoxic effects ofFDVE-cysteine S-conjugates (51).

The toxicologic significance of FDVE S-conjugatessulfoxidation remains unknown. Sulfoxidation representsan alternative route of metabolism for DFEC (alternativeto â-lyase-catalyzed toxification) and an alternative todeacetylation back to potentially toxic cysteine conjugatesfor the nontoxic mercapturates N-Ac-DFEC and (E,Z)-N-Ac-FFVC. Recent experiments showed that DFEC-SOand (Z)-N-Ac-FFVC-SO were more toxic in a humanproximal tubular cell line in culture than the correspond-ing parent cysteine and mercapturic acid FDVE conju-gates, and DFEC-SO was more toxic than (Z)-N-Ac-

Figure 11. Revised pathway of FDVE metabolism in rats and humans. Nomenclature is the same as in Figure 1. Additionalcompounds are (10) DFEC-SO; (11) (E) and (Z)-FFVC-SO; (12) N-Ac-DFEC-SO; and (13) (E)- and (Z)-N-Ac-FFVC-SO).


FFVC-SO (51). Nevertheless, it is not known whetherFDVE conjugates sulfoxidation occurs in vivo, either inrats or in humans, or whether FDVE conjugates sulfox-ides are nephrotoxic in vivo. The significance of the newlyidentified sulfoxidation pathway will depend on therelative toxicities of the various alternative routes ofmetabolism and merits further investigation.

In summary, sulfoxidation of the mercapturates (Z)-N-Ac-FFVC and N-Ac-DFEC is a newly revealed biotrans-formation pathway in the GSH-dependent metabolismof FDVE. DFEC sulfoxidation also occurs but via autoxi-dation. P4503A4/5 and P4503A1/2 are the major enzymesresponsible for (Z)-N-Ac-FFVC and N-Ac-DFEC sulfoxi-dation in human and rat liver microsomes. FDVE mer-capturates sulfoxidation was greater in rat as comparedwith human liver microsomes. This could contribute tospecies differences in FDVE nephrotoxicity.

Acknowledgment. This investigation was supportedby NIH Grants R01DK53765 and P30ES07033. We thankPam Sheffels for her outstanding experimental contribu-tions and Catalin Doneanu from the University ofWashington Department of Medicinal Chemistry MassSpectrometry Center for QTOF analysis.

References

(1) Hanaki, C., Fujii, K., Morio, M., and Tashima, T. (1987) Decom-position of sevoflurane by soda lime. Hiroshima J. Med. Sci. 36,61-67.

(2) Frink, E. J., Jr., Malan, T. P., Morgan, S. E., Brown, E. A.,Malcomson, M., Gandolfi, A. J., and Brown, B. R., Jr. (1992)Quantification of the degradation products of sevoflurane in twoCO2 absorbents during low-flow anesthesia in surgical patients.Anesthesiology 77, 1064-1069.

(3) Morio, M., Fujii, K., Satoh, N., Imai, M., Kawakami, U., Mizuno,T., Kawai, Y., Ogasawara, Y., Tamura, T., Negishi, A., Kumagai,Y., and Kawai, T. (1992) Reaction of sevoflurane and its degrada-tion products with soda lime. Toxicity of the byproducts. Anes-thesiology 77, 1155-1164.

(4) Gonsowski, C. T., Laster, M. J., Eger, E. I., II, Ferrell, L. D., andKerschmann, R. L. (1994) Toxicity of compound A in rats. Effectof increasing duration of administration. Anesthesiology 80, 566-573.

(5) Gonsowski, C. T., Laster, M. J., Eger, E. I., II, Ferrell, L. D., andKerschmann, R. L. (1994) Toxicity of compound A in rats. Effectof a 3-hour administration. Anesthesiology 80, 556-565.

(6) Keller, K. A., Callan, C., Prokocimer, P., Delgado-Herrera, M. S.,Friedman, M. B., Hoffman, G. M., Wooding, W. L., Cusick, P. K.,and Krasula, R. W. (1995) Inhalation toxicology study of ahaloalkene degradant of sevoflurane, Compound A (PIFE), inSprague-Dawley rats. Anesthesiology 83, 1220-1232.

(7) Kharasch, E. D., Thorning, D. T., Garton, K., Hankins, D. C., andKilty, C. G. (1997) Role of renal cysteine conjugate â-lyase in themechanism of compound A nephrotoxicity in rats. Anesthesiology86, 160-171.

(8) Kharasch, E. D., Hoffman, G. M., Thorning, D., Hankins, D. C.,and Kilty, C. G. (1998) Role of the renal cysteine conjugate â-lyasepathway in inhaled compound A nephrotoxicity in rats. Anesthe-siology 88, 1624-1633.

(9) Dekant, W., Martens, G., Vamvakas, S., Metzler, M., and Hen-schler, D. (1987) Bioactivation of tetrachloroethylene. Role ofglutathione S-transferase-catalyzed conjugation versus cyto-chrome P-450-dependent phospholipid alkylation. Drug Metab.Dispos. 15, 702-709.

(10) Dekant, W., Urban, G., Gorsmann, C., and Anders, M. W. (1991)Thioketene formation from R-haloalkenyl 2-nitrophenyl disul-fides: Models for biological reactive intermediates of cytotoxicS-conjugates. J. Am. Chem. Soc. 113, 5120-5122.

(11) Pahler, A., Parker, J., and Dekant, W. (1999) Dose-dependentprotein adduct formation in kidney, liver, and blood of rats andin human blood after perchloroethene inhalation. Toxicol. Sci. 48,5-13.

(12) Jin, L., Baillie, T. A., Davis, M. R., and Kharasch, E. D. (1995)Nephrotoxicity of sevoflurane compound A [fluoromethyl-2,2-difluoro-1-(trifluoromethyl)vinyl ether) in rats: Evidence forglutathione and cysteine conjugate formation and the role of renal

cysteine conjugate â-lyase. Biochem. Biophys. Res Commun. 210,498-506.

(13) Jin, L., Davis, M. R., Kharasch, E. D., Doss, G. A., and Baillie, T.A. (1996) Identification in rat bile of glutathione conjugates offluoromethyl 2,2-difluoro-1-(trifluoromethyl)vinyl ether, a neph-rotoxic degradate of the anesthetic agent sevoflurane. Chem. Res.Toxicol. 9, 555-561.

(14) Spracklin, D., and Kharasch, E. D. (1996) Evidence for themetabolism of fluoromethyl-2,2-difluoro-1-(trifluoromethyl)vinylether (Compound A), by cysteine conjugate â-lyase in rats. Chem.Res. Toxicol. 9, 696-702.

(15) Iyer, R. A., Frink, E. J., Jr., Ebert, T. J., and Anders, M. W. (1998)Cysteine conjugate â-lyase-dependent metabolism of compoundA (2-[fluoromethoxy]-1,1,3,3,3-pentafluoro-1-propene) in humansubjects anesthetized with sevoflurane and in rats given com-pound A. Anesthesiology 88, 611-618.

(16) Iyer, R. A., and Anders, M. W. (1997) Cysteine conjugate â-lyase-dependent biotransformation of the cysteine S-conjugates of thesevoflurane degradation product 2-(fluoromethoxy)-1,1,3,3,3-pen-tafluoro-1-propene (compound A). Chem. Res. Toxicol. 10, 811-819.

(17) Iyer, R. A., and Anders, M. W. (1996) Cysteine conjugate â-lyase-dependent biotransformation of the cysteine S-conjugates of thesevoflurane degradation product compound A in human, nonhu-man primate, and rat kidney cytosol and mitochondria. Anesthe-siology 85, 1454-1461.

(18) Kharasch, E. D., Jubert, C., Spracklin, D., and Hoffmann, G.(1999) Dose-dependent metabolism of fluoromethyl-2,2-difluoro-1-(trifluoromethyl)vinyl ether (Compound A), an anesthetic deg-radation product, to mercapturic acids and 3,3,3-trifluoro-2-fluoromethoxypropanoic acid in rats. Toxicol. Appl. Pharmacol.160, 49-59.

(19) Tong, Z., and Anders, M. W. (2002) Reactive intermediate for-mation from the 2-(fluoromethoxy)-1,1,3,3,3-pentafluoro-1-pro-pene (compound A)-derived cysteine S-conjugate S-[2-(fluo-romethoxy)-1,1,3,3,3-pentafluoropropyl]-L-cysteine in pyridoxalmodel systems. Chem. Res. Toxicol. 15, 623-628.

(20) Altuntas, T. G., and Kharasch, E. D. (2001) Glutathione S-conjugation of the sevoflurane degradation product, fluoromethyl-2,2-difluoro-1-(trifluoromethyl)vinyl ether (Compound A) in hu-man liver, kidney, and blood in vitro. Toxicol. Appl. Pharmacol.177, 85-93.

(21) Altuntas, T. G., and Kharasch, E. D. (2002) Biotransformation ofL-cysteine S-conjugates and N-acetyl-L-cysteine S- conjugates ofthe sevoflurane degradation product fluoromethyl-2,2-difluoro-1-(trifluoromethyl)vinyl ether (Compound A) in human kidney invitro: Interindividual variability in N-acetylation, N-deacetyla-tion, and â-lyase-catalyzed metabolism. Drug Metab. Dispos. 30,148-154.

(22) Kharasch, E. D., and Jubert, C. (1999) Compound A uptake andmetabolism to mercapturic acids and 3,3,3-trifluoro-2-fluorometh-oxypropanoic acid during low-flow sevoflurane anesthesia. Biom-arkers for exposure, risk assessment, and interspecies compari-son. Anesthesiology 91, 1267-1278.

(23) Park, S. B., Osterloh, J. D., Vamvakas, S., Hashmi, M., Anders,M. W., and Cashman, J. R. (1992) Flavin-containing monooxy-genase-dependent stereoselective S-oxygenation and cytotoxicityof cysteine S-conjugates and mercapturates. Chem. Res. Toxicol.5, 193-201.

(24) Lash, L. H., Sausen, P. J., Duescher, R. J., Cooley, A. J., andElfarra, A. A. (1994) Roles of cysteine conjugate â-lyase andS-oxidase in nephrotoxicity: Studies with S-(1,2-dichlorovinyl)-L-cysteine and S-(1,2-dichlorovinyl)-L-cysteine sulfoxide. J. Phar-macol. Exp. Ther. 269, 374-383.

(25) Werner, M., Guo, Z., Birner, G., Dekant, W., and Guengerich, F.P. (1995) The sulfoxidation of the hexachlorobutadiene metaboliteN-acetyl-S-(1,2,3,4,4-pentachlorobutadienyl)-L-cysteine is cata-lyzed by human cytochrome P450 3A enzymes. Chem. Res.Toxicol. 8, 917-923.

(26) Werner, M., Birner, G., and Dekant, W. (1995) The role ofcytochrome P4503A1/2 in the sex-specific sulfoxidation of thehexachlorobutadiene metabolite, N-acetyl-S-(pentachlorobutadi-enyl)-L-cysteine in rats. Drug Metab. Dispos. 23, 861-868.

(27) Werner, M., Birner, G., and Dekant, W. (1996) Sulfoxidation ofmercapturic acids derived from tri- and tetrachloroethene bycytochromes P450 3A: A bioactivation reaction in addition todeacetylation and cysteine conjugate â-lyase mediated cleavage.Chem. Res. Toxicol. 9, 41-49.

(28) Birner, G., Werner, M., Rosner, E., Mehler, C., and Dekant, W.(1998) Biotransformation, excretion, and nephrotoxicity of thehexachlorobutadiene metabolite (E)-N-acetyl-S-(1,2,3,4,4-pen-tachlorobutadienyl)-L-cysteine sulfoxide. Chem. Res. Toxicol. 11,750-757.


(29) Ripp, S. L., Overby, L. H., Philpot, R. M., and Elfarra, A. A. (1997)Oxidation of cysteine S-conjugates by rabbit liver microsomes andcDNA-expressed flavin-containing mono-oxygenases: studies withS-(1,2-dichlorovinyl)-L-cysteine, S-(1,2,2-trichlorovinyl)-L-cysteine,S-allyl-L-cysteine, and S-benzyl-L-cysteine. Mol. Pharmacol. 51,507-515.

(30) Krause, R. J., Lash, L. H., and Elfarra, A. A. (2003) Human kidneyflavin-containing monooxygenases and their potential roles incysteine S-conjugate metabolism and nephrotoxicity. J. Pharma-col. Exp. Ther. 304, 185-191.

(31) Kharasch, E. D., and Thummel, K. E. (1993) Identification ofcytochrome P450 2E1 as the predominant enzyme catalyzinghuman liver microsomal defluorination of sevoflurane, isofluraneand methoxyflurane. Anesthesiology 79, 795-807.

(32) Lowry, O. H., Rosebrough, N. J., Farr, A. L., and Randall, R. J.(1951) Protein measurement with the Folin phenol reagent. J.Biol. Chem. 193, 265-275.

(33) Mugford, C. A., Mortillo, M., Mico, B. A., and Tarloff, J. B. (1992)1-aminobenzotriazole-induced destruction of hepatic and renalcytochromes P450 in male Sprague-Dawley rats. Fundam. Appl.Toxicol. 19, 43-49.

(34) Lemoine, A., Johann, M., and Cresteil, T. (1990) Evidence for thepresence of distinct flavin-containing monooxygenases in humantissues. Arch. Biochem. Biophys. 276, 336.

(35) Uttamsingh, V., Iyer, R. A., Baggs, R. B., and Anders, M. W. (1998)Fate and toxicity of 2-(fluoromethoxy)-1,1,3,3,3-pentafluoro-1-propene (compound A)-derived mercapturates in male, Fischer344 rats. Anesthesiology 89, 1174-1183.

(36) Kedderis, G. L., and Rickert, D. E. (1985) Loss of rat livermicrosomal cytochrome P-450 during methimazole metabolism.Role of flavin-containing monooxygenase. Drug. Metab. Dispos.13, 58-61.

(37) Waxman, D. J., and Azaroff, L. (1992) Phenobarbital Inductionof cytochrome-P-450 gene expression. Biochem. J. 281, 577-592.

(38) Williams, J. A., Ring, B. J., Cantrell, V. E., Jones, D. R., Eckstein,J., Ruterbories, K., Hamman, M. A., Hall, S. D., and Wrighton,S. A. (2002) Comparative metabolic capabilities of CYP3A4,CYP3A5, and CYP3A7. Drug Metab. Dispos. 30, 883-891.

(39) Kuehl, P., Zhang, J., Lin, Y., Lamba, J., Assem, M., Schuetz, J.,Watkins, P. B., Daly, A., Wrighton, S. A., Hall, S. D., Maurel, P.,Relling, M., Brimer, C., Yasuda, K., Venkataramanan, R., Strom,S., Thummel, K., Boguski, M. S., and Schuetz, E. (2001) Sequencediversity in CYP3A promoters and characterization of the geneticbasis of polymorphic CYP3A5 expression. Nat. Genet. 27, 383-391.

(40) Damani, L. A., and Houdi, A. A. (1988) Cytochrome P-450 andFAD-monooxygenase mediated S- and N-oxygenations. DrugMetab. Drug Interact. 6, 235-244.

(41) Sausen, P. J., Duescher, R. J., and Elfarra, A. A. (1993) Furthercharacterization and purification of the flavin-dependent S-benzyl-L-cysteine S-oxidase activities of rat liver and kidneymicrosomes. Mol. Pharmacol. 43, 388-396.

(42) Ebert, T. J., and Arain, S. R. (2000) Renal responses to low-flowdesflurane, sevoflurane, and propofol in patients. Anesthesiology93, 1401-1406.

(43) Obata, R., Bito, H., Ohmura, M., Moriwaki, G., Ikeuchi, Y., Katoh,T., and Sato, S. (2000) The effects of prolonged low-flow sevoflu-rane anesthesia on renal and hepatic function. Anesth. Analg. 91,1262-1268.

(44) Kharasch, E. D., Frink, E. J., Jr., Artru, A., Michalowski, P.,Rooke, G. A., and Nogami, W. (2001) Long-duration low-flowsevoflurane and isoflurane effects on postoperative renal andhepatic function. Anesth. Analg. 93, 1511-1520.

(45) Conzen, P. F., Kharasch, E. D., Czerner, S. F., Artru, A. A.,Reichle, F. M., Michalowski, P., Rooke, G. A., Weiss, B. M., andEbert, T. J. (2002) Low-flow sevoflurane compared with low-flowisoflurane anesthesia in patients with stable renal insufficiency.Anesthesiology 97, 578-584.

(46) Kharasch, E. D., Conzen, P. F., Michalowski, P., Weiss, B. M.,Rooke, G. A., Artru, A. A., Ebert, T. J., Czerner, S. F., and Reichle,F. M. (2003) Safety of low-flow sevoflurane anesthesia in patients.Anesthesiology 99, 752-754.

(47) Eger, E. I., II, Koblin, D. D., Bowland, T., Ionescu, P., Laster, M.J., Fang, Z., Gong, D., Sonner, J., and Weiskopf, R. B. (1997)Nephrotoxicity of sevoflurane versus desflurane anesthesia involunteers. Anesth. Analg. 84, 160-168.

(48) Eger, E. I., II, Gong, D., Koblin, D. D., Bowland, T., Ionescu, P.,Laster, M. J., and Weiskopf, R. B. (1997) Dose-related biochemicalmarkers of renal injury after sevoflurane versus desfluraneanesthesia in volunteers. Anesth. Analg. 85, 1154-1163.

(49) Ebert, T. J., Frink, E. J., Jr., and Kharasch, E. D. (1998) Absenceof biochemical evidence for renal and hepatic dysfunction follow-ing 8 h of 1.25 MAC sevoflurane anesthesia in volunteers.Anesthesiology 88, 601-610.

(50) Ebert, T. J., Messana, L. D., Uhrich, T. D., and Staacke, T. S.(1998) Absence of renal and hepatic toxicity after 4 h of 1.25minimum alveolar concentration sevoflurane anesthesia in vol-unteers. Anesth. Analg. 86, 662-667.

(51) Altuntas, T. G., Zager, R. A., and Kharasch, E. D. (2003)Cytotoxicity of S-conjugates of the sevoflurane degradationproduct fluoromethyl-2,2-difluoro-1-(trifluoromethyl)vinyl ether(compound A) in a human proximal tubular cell line. Toxicol ApplPharmacol 193, 55-65.

TX034254K


sulfoxidation of cysteine and mercapturic acid conjugates of the sevoflurane degradation product...

Documents