Evidence for Metabolism of Fluoromethyl 2, 2-Difluoro-1

The volatile anesthetic sevoflurane is degraded to fluoromethyl 2,2-difluoro-1-(trifluoro- methyl)vinyl ether (FDVE), a potent rat nephrotoxin. In rat...
11 downloads 0 Views 411KB Size
Download PDF
696

Chem. Res. Toxicol. 1996, 9, 696-702

Evidence for Metabolism of Fluoromethyl 2,2-Difluoro-1-(trifluoromethyl)vinyl Ether (Compound A), a Sevoflurane Degradation Product, by Cysteine Conjugate β-Lyase Douglas K. Spracklin and Evan D. Kharasch* Departments of Anesthesiology and Medicinal Chemistry, University of Washington, Seattle, Washington 98195 Received December 14, 1995X

The volatile anesthetic sevoflurane is degraded to fluoromethyl 2,2-difluoro-1-(trifluoromethyl)vinyl ether (FDVE), a potent rat nephrotoxin. In rats in vivo, FDVE undergoes glutathione conjugation and metabolism to cysteine conjugates, whose bioactivation by renal cysteine conjugate β-lyase has been implicated by the protective effects of (aminooxy)acetic acid, an inhibitor of cysteine conjugate β-lyase. We specifically tested the hypothesis that FDVE is metabolized via the β-lyase pathway to yield 3,3,3-trifluoro-2-(fluoromethoxy)propanoic acid. Urine of rats administered FDVE (0.3 mmol/kg) was extracted and derivatized with diazomethane. Headspace GC/MS analysis demonstrated a peak whose retention time and mass spectrum were identical to those of synthetic methyl 3,3,3-trifluoro-2-(fluoromethoxy)propanoate. Pretreatment of rats with (aminooxy)acetic acid significantly decreased the amount of 3,3,3-trifluoro-2-(fluoromethoxy)propanoic acid detected in the urine of FDVE-treated animals. The 19F NMR spectrum of urine from rats administered FDVE was consistent with the formation of 3,3,3-trifluoro-2-(fluoromethoxy)propanoic acid, but could not be differentiated from that of FDVE mercapturates, which are also excreted in urine. These results suggest that FDVE undergoes biotransformation via the β-lyase pathway and β-lyase-catalyzed metabolism may mediate the nephrotoxicity of this compound.

Introduction Sevoflurane (fluoromethyl 2,2,2-trifluoro-1-(trifluoromethyl)ethyl ether, 1) is a halogenated ether general anesthetic that is used in Japan and has recently gained approval for use in the U.S., South America, and Europe. Under certain conditions, strong bases in the soda lime (Ca(OH)2 + KOH + NaOH) or baralyme (Ca(OH)2 + Ba(OH)2 + KOH) carbon dioxide absorbent in the anesthesia machine may abstract the acidic proton of sevoflurane, yielding several chemical degradation products (1, 2). The quantitatively and toxicologically most significant degradation product has been identified as fluoromethyl 2,2-difluoro-1-(trifluoromethyl)vinyl ether (FDVE)1 (Figure 1, 2). FDVE is also referred to by the trivial name “Compound A” in the official product label. Additionally, FDVE may be a minor contaminant in sevoflurane (2). Human FDVE exposure during sevoflurane anesthesia has been well-described (3, 4). FDVE is nephrotoxic in rats (5), with clear biochemical and histological evidence of corticomedullary renal tubular cell necrosis after exposure to concentrations as low as 50-110 ppm (6-8). However, the mechanism of FDVE related renal toxicity has not been fully elucidated. FDVE bears two functional moieties that are potential sites for disparate routes of biotransformation. The * Address correspondence to: Evan D. Kharasch, M.D., Ph.D., Department of Anesthesiology, Box 356540, University of Washington, Seattle, WA 98195; phone: 206-543-2039; FAX: 206-685-3079; Email: [email protected]. X Abstract published in Advance ACS Abstracts, April 15, 1996. 1 Abbreviations: P450, cytochrome P450; FDVE, fluoromethyl 2,2difluoro-1-(trifluoromethyl)vinyl ether; Bn, benzyl; DMSO, dimethyl sulfoxide; PE, petroleum ether; CF3CO2H, trifluoroacetic acid; AOAA, (aminooxy)acetic acid; FAB/MS, fast atom bombardment mass spectrometry.

S0893-228x(95)00210-4 CCC: $12.00

Figure 1. Chemical degradation of sevoflurane to FDVE.

fluoromethyl carbon may undergo P450-catalyzed oxidation to yield inorganic and organic fluoride metabolites, similar to the metabolism of sevoflurane and other halogenated ether anesthetics (9, 10). Alternatively, FDVE may form glutathione adducts which are candidates for further metabolism to the corresponding cysteine conjugates, analogous to the metabolism of other haloalkene nephrotoxins (11). These cysteine conjugates represent potential substrates for renal cysteine conjugate β-lyase. Recent investigations have implicated both metabolic pathways in the metabolism and potential toxicity of FDVE (12-14). Human liver microsomes have been shown to metabolize FDVE to inorganic fluoride and the organic fluoride metabolite pentafluoroacetone in vitro (12). FDVE-glutathione adduct formation by rat liver microsomes in vitro as well as the hepatic biosynthesis of identical glutathione adducts and subsequent excretion of corresponding mercapturates in rats in vivo was recently demonstrated (13, 14). Metabolism of FDVEcysteine conjugates by renal cysteine conjugate β-lyase in vivo was recently implicated, based on the effects of the β-lyase inhibitor (aminooxy)acetic acid (13). However, the actual metabolism of FDVE-cysteine conjugates by renal β-lyase in vivo has not been demonstrated. The purpose of this investigation was to test the hypoth© 1996 American Chemical Society

Anesthetic Degradate Metabolism by β-Lyase

esis that FDVE is metabolized by β-lyase to the putative metabolite 3,3,3-trifluoro-2-(fluoromethoxy)propanoic acid.

Experimental Procedures Chemicals. FDVE (99.9% by gas chromatography) was synthesized by Central Glass Co. (Ube City, Japan) and provided by Abbott Laboratories (Abbott Park, IL). FDVE was stored in amber vials and stabilized with 500 ppm butylated hydroxytoluene. Unless specified, all reagents were obtained from Aldrich Chemical Co. (Milwaukee, WI) or Fluka (Ronkonkoma, NY) and were the highest purity available. All solutions were prepared with high purity water (>18 MΩ‚cm) obtained from a Milli-Q UV Plus system (Millipore Corp., Bedford, MA). Synthesis of 3,3,3-Trifluoro-2-(fluoromethoxy)propanoic Acid. Under a N2 atmosphere, benzyl alcohol (5.00 g, 4.78 mL, 45.8 mmol) was dissolved in THF (200 mL) and the solution was cooled to 0 °C. Sodium hydride (1.92 g of a 60% dispersion in oil, 48.1 mmol) was added portionwise, and the resulting suspension was stirred at 0 °C for 25 min. FDVE (9.16 g, 6.11 mL, 50.8 mmol) was added in one portion, and the mixture was stirred at 0 °C. After 60 min, TLC analysis (silica gel, EtOAc/ PE, (1:13), visualization by I2 and UV light (λ ) 254 nm)) indicated the reaction was incomplete, so further FDVE was added (4.50 g, 3.00 mL, 25.0 mmol) and the mixture stirred at 0 °C for a further 40 min. The reaction was quenched with water, and the volume of the solution was reduced to approximately 100 mL by rotary evaporation. Saturated sodium chloride was added, and the aqueous phase was extracted with EtOAc. The organic phase was dried (Na2SO4) and concentrated to yield a mixture of products as a yellow oil. Preparative column chromatography (silica gel, 200-400 mesh) using a linear eluant gradient of EtOAc/PE (1:20 to 100:0) yielded a mixture of the alkene and alkane adducts (>95:5) as a yellow oil (3.91 g, 52%). Alkene 4: 1H NMR (CDCl3) δ: 7.35 (s, 5H, C6H5), 5.19 (d, J ) 52.8 Hz, 2H, FCH2O), 5.07 (s, 2H, CH2Ar); 19F NMR (CDCl ) δ: -3.3 (d, J ) 25.8 Hz, 3F, CF ), -35.1 (q, 3 3 J ) 23.8 Hz, 1F, CdC(F)), -90.1 (t, J ) 54.5 Hz, 1F, FCH2O); liquid phase GC/MS: tR)13.1 min; m/z 268 ([M]+), 91 ([C7H5]+). Alkane 5: NMR: below levels of detection; liquid phase GC/ MS: tR)16.2 min; m/z 288 ([M]+), 91 ([C7H5]+). A solution of the benzyl adducts (2.0 g, 7.2 mmol) in EtOAc (100 mL) was combined with Pd/C (10% Pd, 0.80 g, 40% w/w), and the mixture was stirred under an H2 atmosphere at 50 psi for 48 h. The mixture was filtered through Celite, and a solution of NaOH was added until the pH of the aqueous phase was 10. The layers were separated, and the aqueous phase was lyophilized to yield the crude sodium salt of 3,3,3-trifluoro-2-(fluoromethoxy)propanoic acid as a white solid (0.83 g) which was subsequently used without purification. 1H NMR (D2O) δ: 5.44 (dq, J ) 54.9, 2.8 Hz, 2H, FCH2O), 4.68 (q, J ) 7.7 Hz, 1H, CH); 19F NMR (D2O) δ: -11.25 (d, J ) 7.8 Hz, 3F, CF3), -90.1 (t, J ) 54.6 Hz, 1F, FCH2O); FAB/MS m/z: 175 ([M - H]+), 131 ([C3H2O2F2Na]+). A sample of the sodium salt of 3,3,3-trifluoro-2-(fluoromethoxy)propanoic acid was dissolved in water and the pH adjusted to 2 with HCl. The aqueous phase was extracted with EtOAc, and the combined extracts were dried (Na2SO4) and concentrated to yield 3,3,3-trifluoro-2-(fluoromethoxy)propanoic acid. 1H NMR (CDCl3) δ: 5.38 (d, J ) 54.1 Hz, 2H, FCH2O), 4.68 (q, J ) 6.6 Hz, 1H, CH); liquid phase GC/MS: tR ) 7.2 min; m/z 112 ([C3H3OF3]+, McLafferty rearrangement), 29 ([CHO]+, McLafferty rearrangement). Synthesis of Methyl 3,3,3-Trifluoro-2-(fluoromethoxy)propanoate. A sample of the sodium salt of 3,3,3-trifluoro-2(fluoromethoxy)propanoic acid was dissolved in water and the pH adjusted to 2 with HCl. The aqueous phase was extracted with EtOAc, and the combined extracts were dried (Na2SO4) and concentrated to yield 3,3,3-trifluoro-2-(fluoromethoxy)propanoic acid. An excess of diazomethane was added, and the solution was agitated for 5 min to yield methyl 3,3,3-trifluoro2-(fluoromethoxy)propanoate, as identified by headspace GC/ MS: tR ) 29.1 min; m/z 142 ([C4H5O2F3]+, McLafferty rear-

Chem. Res. Toxicol., Vol. 9, No. 4, 1996 697 rangement), 131 ([M - CO2CH3]+), 112 ([C3H3OF3]+), 59 ([CO2CH3]+), 33 ([FCH2]+). Synthesis of FDVE Mercapturates: N-Acetyl-S-(1,1,3,3,3pentafluoro-2-(fluoromethoxy)propyl)-L-cysteine and NAcetyl-S-(1-fluoro-2-(fluoromethoxy)-2-(trifluoromethyl)vinyl)-L-cysteine. The mercapturates from the reaction of FDVE and N-acetyl-L-cysteine were prepared by the method of van Bladeren et al. (15). Briefly, to a solution of sodium methoxide (2.05 equiv) in methanol was added N-acetyl-Lcysteine (1.0 equiv), followed by FDVE (1.1 equiv). The products were isolated by evaporating the solvent of the reaction mixture and then resuspending the residue in chloroform and concentrated hydrochloric acid. The chloroform layer was separated and dried (Na2SO4). Evaporation of the solvent in vacuo yielded a mixture of the FDVE mercapturates as a yellow, gum-like residue. 1H NMR (acetone-d6) δ: 8.8 (br s, 2H, CO2H), 8.05 (d, J ) 6.5 Hz, 2H, C(O)NH), 5.75 (d, J ) 54.0 Hz, 4H, FCH2O), 5.30 (m, 1H, CH(CF3)), 5.0 (m, 2H, CH(CO2H)), 3.8-3.65 (m, 2H, SCH2), 3.6-3.45 (m, 2H, SCH2), 2.15 (s, 3H, NH(CH3)), 2.10 (s, 3H, NH(CH3)); 19F NMR (D2O) δ: -3.6 (dd, J ) 21.7, 5.4 Hz, 6F, CF3 alkene), -11.4 (s, 6F, CF3 alkane), -18.1 (dm, J ) 224.8 Hz, 1F, CF2, alkane), -18.5 (dm, J ) 224.8 Hz, 1F, CF2, alkane), -20.9 (dm, J ) 224.2 Hz, 1F, CF2, alkane), -21.3 (dm, J ) 224.2 Hz, 1F, CF2, alkane), -46.3 (q, J ) 22.4 Hz, 2F, CdC(F)), alkene), -89.16 (tm, J ) 51.9 Hz, 2F, FCH2O, alkene), -90.1 (t, J ) 51.3 Hz, 2F, FCH2O, alkane); FAB/MS m/z 344 ([M + H]+, alkane), 324 ([M + H]+, alkene). Preparative column chromatography (silica gel, 200-400 mesh) using a linear eluant gradient of CHCl3/MeOH/HOAc (7: 0.5:0.5 to 7:2:1), visualization by I2 and UV light (λ ) 254 nm) yielded two fractions, N-acetyl-S-(1-fluoro-2-(fluoromethoxy)-2(trifluoromethyl)vinyl)-L-cysteine followed by N-acetyl-S-(1,1,3,3,3pentafluoro-2-(fluoromethoxy)propyl)-L-cysteine. N-Acetyl-S-(1,1,3,3,3-pentafluoro-2-(fluoromethoxy)propyl)-L-cysteine: 1H NMR (acetone-d6) δ: 7.75 (d, J ) 8.2 Hz, 1H, C(O)NH), 5.6 (d, J ) 54.9 Hz, 2H, FCH2O), 5.2 (m, 1H, CH(CF3)), 4.8 (m, 1H, CH(CO2H)), 3.55 (dd, J ) 15.2, 5.0 Hz, 1H, CH2), 3.35 (dd, J ) 15.2, 6.6 Hz, 1H, CH2), 2.05 (s, 3H, NH(CH3)); 19F NMR (D2O) δ: -11.4 (s, 6F, CF3), -18.1 (dm, J ) 224.8 Hz, 1F, CF2), -18.6 (dm, J ) 224.8 Hz, 1F, CF2), -20.9 (dm, J ) 224.2 Hz, 1F, CF2), -21.4 (dm, J ) 224.2 Hz, 1F, CF2), -90.1 (t, J ) 51.9, 2F, FCH2O); FAB/MS m/z 344 ([M + H]+). No attempts were made to resolve the alkane diastereomers. N-Acetyl-S-(1-fluoro-2-(fluoromethoxy)-2-(trifluoromethyl)vinyl)-L-cysteine: 1H NMR (acetone-d6) δ: 7.8 (m, 1H, C(O)NH), 5.6 (d, J ) 54.3 Hz, 2H, FCH2O), 4.75 (m, 1H, CH(CO2H)), 3.6 (dd, J ) 15.0, 5.2 Hz, 1H, CH2), 3.35 (dd, J ) 15.0, 7.7 Hz, 1H, CH2), 2.0 (s, 3H, NH(CH3)); 19F NMR (D2O) δ: -3.36 (dd, J ) 21.4, 6.1 Hz, 3F, CF3), -46.2 (q, J ) 21.7 Hz, 1F, CdC(F)), -89.1 (tq, J ) 51.6, 6.1 Hz, 1F, FCH2O); FAB/MS m/z 324 ([M + H]+). No attempts were made to resolve the (E)- and (Z)-isomers. However, the large 4JF-F coupling constant observed between the vinyl fluoride and the trifluoromethyl group (J ) 21.7 Hz) suggests that the two groups are in a cis conformation (16) and that the N-acetyl-S-((Z)-(1-fluoro-2(fluoromethoxy)-2-(trifluoromethyl)vinyl))-L-cysteine isomer predominates. This is consistent with the previous observation that the predominant predecessor glutathione conjugate is also the (Z)-isomer (14). Analytical. For analysis of methyl 3,3,3-trifluoro-2-(fluoromethoxy)propanoate in rat urine, a sample of rat urine (1 mL) was acidified to pH 2 with HCl and extracted with EtOAc. The combined organic extracts were dried (Na2SO4) and concentrated by rotary evaporation. Diazomethane (0.5 mL) was added, and the solution was agitated for 5 min and then analyzed directly by headspace GC/MS. Headspace GC/MS analyses were performed on a 5890 Series II gas chromatograph (Hewlett-Packard, Wilmington, DE) with an HP 7694 headspace sampler interfaced to an HP 5971 mass selective detector, using a DB-VRX fused-silica capillary column (30 m × 0.32 mm × 1.8 µm film thickness) (J &W Scientific, Folsom, CA). The GC injector and detector temperatures were 150 and 250 °C, respectively, and the column head pressure was

698 Chem. Res. Toxicol., Vol. 9, No. 4, 1996 2.5 psi. For the headspace GC/MS analysis of methyl 3,3,3trifluoro-2-(fluoromethoxy)propanoate, the headspace sampler parameters were as follows: Agitation ) high; oven temperature ) 70 °C; loop temperature ) 80 °C; transfer line temperature ) 90 °C; sample equilibration time ) 0.1 min; vial pressurization time ) 0.1 min; loop fill time ) 0.25 min; loop equilibration time ) 0.15 min; injection time ) 0.25 min. The GC oven temperature was held at 30 °C for 5 min, increased at 2 °C/min to 185 °C, and held at this temperature for 15 min. Liquid phase GC/MS analyses were performed on a HP 5890 Series II gas chromatograph interfaced to an HP 5972 mass selective detector, using a DB-17 fused-silica capillary column (15 m × 0.32 mm × 0.5 µm film thickness) (J&W Scientific, Folsom, CA). The injector and detector temperatures of the GC were 190 and 280 °C, respectively, and the column head pressure was 3.0 psi. The oven temperature was held at 50 °C for 2 min, increased at 5 °C/min to 200 °C, and held for 7 min. Negative ion fast atom bombardment mass spectra (FAB/MS) were recorded on a VG 70SEQ Tandem Hybrid (EBqQ) MS/MS spectrometer (VG Analytical, Manchester, U.K.) using thioglycerol as the matrix. Positive ion FAB/MS utilized DMIX (thioglycerol, DMSO, CF3CO2H) as the matrix. Unless indicated otherwise, fluorine nuclear magnetic resonance spectra (19F NMR) were recorded in deuterated aqueous solutions on a Varian XL-300 spectrometer (Varian NMR Instruments, Palo Alto, CA) (282 MHz) equipped with a 5 mm 19F probe. Chemical shifts are reported relative to chlorodifluoroacetic acid (δ 0 ppm), referenced externally. Proton nuclear magnetic resonance spectra (1H NMR) were recorded on a Varian XL-300 spectrometer (300 MHz). These spectra were recorded in either deuterochloroform or deuterated aqueous solutions. Chemical shifts are referenced to tetramethylsilane (δ 0 ppm) for deuterochloroform solutions, or to residual solvent (HOD, δ 4.8 ppm) for aqueous solutions. Signal multiplicities, spin-spin coupling constants (where possible), integration ratios, and structural assignments are given in parentheses. Animals. This investigation was approved by the institutional Animal Care and Use Committee. Male Fischer 344 rats (220-240 g; Simonsen Laboratories, Gilroy, CA) were housed in individual metabolic cages, provided food and water ad libitum, and maintained on a 12 h light/dark cycle (6 a.m.-6 p.m.). Animals received FDVE (0.3 mmol/kg) in corn oil (4 mL/ kg) by intraperitoneal injection, 1 h after first receiving either 0.5 mmol/kg AOAA (Sigma, St. Louis, MO) (200 mM stock in saline adjusted to pH 6-6.5 with 0.1 M NaOH) or saline by intraperitoneal injection (2.5 mL/kg). Urine was collected on ice for 24 h and frozen at -20 °C for later analysis. Urine samples were filtered (Amicon, Inc., Beverly, MA) prior to analysis.

Results 3,3,3-Trifluoro-2-(fluoromethoxy)propanoic acid was synthesized in two steps from benzyl alcohol and FDVE (Figure 2). The synthetic route was devised to parallel the theoretical route of formation in vivo (Figure 6). The proposed mechanism of formation of acid 10 from oxy compounds 4 and 5 (via 6 and 7) is analogous to the in vivo formation of 10 from thio compounds 11 and 12 (via 13 and 14). Addition of benzyl alcohol to FDVE gave a mixture of products 4 and 5, although the alkene products were predominant (>95:5). No attempts were made to separate the alkene/alkane mixture as in the next step (hydrogenolysis of the benzyl group), both compounds would give the same product. Hydrogenolysis of these resulting benzyl adducts gave the postulated intermediates 6 and 7. Based on analogy to structurally related compounds (17), elimination of fluoride from 7, followed by hydrolysis of the postulated acyl fluoride 9, would give the desired acid 10. Elimination of fluoride

Spracklin and Kharasch

Figure 2. Synthetic route to 3,3,3-trifluoro-2-(fluoromethoxy)propanoic acid.

from 6 would give thioketene 8, which would give the desired acid upon hydrolysis. However, vinyl alcohol 6 may also tautomerize to 9. The present results do not permit identification of the proportion of 10 derived from each pathway. The identity of the product 10 was confirmed by NMR (1H and 19F) and GC/MS, as well as by direct insertion FAB/MS of the sodium salt. Identification of 3,3,3-trifluoro-2-(fluoromethoxy)propanoic acid in the urine of rats administered intraperitoneal FDVE was accomplished by extraction, derivatization with diazomethane, and GC/MS analysis. GC analysis of an extracted, methylated urine sample (Figure 3A) showed a peak at 29.1 min which was not present in the urine of control animals receiving only corn oil (Figure 3B). The mass spectrum of this peak is shown in Figure 3C. The spectrum of synthetic methyl 3,3,3trifluoro-2-(fluoromethoxy)propanoate is provided in Figure 3D. The retention time for this synthetic compound was 29.1 min, identical to that of the derivatized urine metabolite. Diagnostic ions and the fragments from which they arise for both the methylated urine metabolite and synthetic methyl 3,3,3-trifluoro-2-(fluoromethoxy)propanoate were m/z 142 ([C4H5O2F3]+, McLafferty rearrangement), 131 ([M - CO2CH3]+), 112 ([C3H3OF3]+), 69 ([CF3]+), 59 ([CO2CH3]+), and 33 ([FCH2]+). Based on the identical retention times and similarity of the mass spectra, the material excreted in rat urine is identified as 3,3,3-trifluoro-2-(fluoromethoxy)propanoic acid.

Anesthetic Degradate Metabolism by β-Lyase

Chem. Res. Toxicol., Vol. 9, No. 4, 1996 699

Figure 4. GC/MS analysis of rat urine. (A) Selected ion mode chromatograph of a derivatized extract of urine from a rat receiving 0.2 mmol/kg FDVE. A 1 mL sample from the total urine collected after 24 h (35 mL) was derivatized using diazomethane as described in the Experimental Procedures. (B) Selected ion mode chromatograph of a derivatized extract of urine from a rat receiving 0.2 mmol/kg FDVE after pretreatment with AOAA (0.5 mmol/kg). A 1 mL sample from the total urine collected after 24 h (12 mL) was derivatized using diazomethane.

coupling constant of the CF3 group (7.8 and 6.1 Hz) for the synthetic material and the metabolite. Signal broadening in the spectrum of the metabolite in urine compared to the synthetic material is consistent with previous 19F NMR studies and has been ascribed to interaction with proteins (18). High urine protein concentrations are also known to occur in rats treated with FDVE (13). Figure 3. GC/MS analysis of rat urine. (A) Total ion chromatograph of a derivatized extract of urine from a rat receiving 0.3 mmol/kg FDVE. (B) Total ion chromatograph of a derivatized extract of urine from a control rat receiving corn oil. (C) Mass spectrum of the peak at 29.1 min in (A) (arrow). (D) Mass spectrum of synthetic methyl 3,3,3-trifluoro-2-(fluoromethoxy)propanoate.

Pretreatment of rats with AOAA, a competitive inhibitor of renal cysteine conjugate β-lyase, prior to FDVE injection, resulted in a significant decrease in the amount of methyl 3,3,3-trifluoro-2-(fluoromethoxy)propanoate detected in urine after derivatization with diazomethane. The selected ion mode GC/MS chromatograms of three ions characteristic of methyl 3,3,3-trifluoro-2-(fluoromethoxy)propanoate are shown in Figure 4. The amount of methyl 3,3,3-trifluoro-2-(fluoromethoxy)propanoate detected in saline pretreated animals (Figure 4A) was severalfold greater than that detected in animals pretreated with AOAA (Figure 4B), thus further suggesting a role for cysteine conjugate β-lyase in mediating the metabolism of FDVE. The 19F NMR spectrum of urine from rats given FDVE is shown in Figure 5A. The 19F resonances are observed at δ -11.5 (d, J ) 6.1 Hz), -14.5 (s), -18.2 (dm, J ) 224.1 Hz), -21.4 (dm, J ) 224.1 Hz), and -90.1 ppm (t, J ) 54.9 Hz). The resonance at -58 ppm is attributable to inorganic fluoride. This identification was confirmed in experiments involving addition of known amounts of NaF (not shown). No 19F resonances were in the urine of control animals receiving only corn oil vehicle (not shown). The 19F NMR spectrum of synthetic 3,3,3trifluoro-2-(fluoromethoxy)propanoic acid is shown in Figure 5B. Similar to the spectrum from urine, resonances are also observed at δ -11.25 ppm (d, J ) 7.8 Hz, CF3) and δ -90.1 ppm (t, J ) 54.6 Hz, FCH2O). There was also similarity of the 2JH-F coupling constant of the FCH2O group (54.6 and 54.9 Hz) and the 3JH-F

The urine of rats treated with FDVE is also known to contain FDVE-mercapturates, as shown previously by liquid chromatography-mass spectrometry (13). The 19F NMR spectra of synthetic N-acetyl-S-(1,1,3,3,3-pentafluoro2-(fluoromethoxy)propyl)-L-cysteine and N-acetyl-S-(1fluoro-2-(fluoromethoxy)-2-(trifluoromethyl)vinyl)-L-cysteine are shown in Figure 5, panels D and E, respectively. Resonances for the alkyl mercapturate are observed at δ -11.4 (CF3), -18.1, -18.6, -20.9, -21.4 (CF2), and -90.1 ppm (FCH2O). Resonances for the alkenyl mercapturate are observed at δ -3.36 (CF3), -46.2 (CdC(F)), and -89.1 ppm (FCH2O). Comparison of the urine spectrum to that of the synthetic compounds suggests that 3,3,3-trifluoro-2-(fluoromethoxy)propanoic acid is present in urine. Nevertheless, due to the spectral similarity of the acid and the alkyl mercapturate, unique identification of the acid is not possible. Indeed, the urine spectrum suggests that N-acetyl-S-(1,1,3,3,3-pentafluoro2-(fluoromethoxy)propyl)-L-cysteine is present, while resonances for the N-acetyl-S-(1-fluoro-2-(fluoromethoxy)-2(trifluoromethyl)vinyl)-L-cysteine mercapturate are not observed. Furthermore, it appears that one diastereomer of N-acetyl-S-(1,1,3,3,3-pentafluoro-2-(fluoromethoxy)propyl)-L-cysteine predominates. Specifically, the spectrum of synthetic N-acetyl-S-(1,1,3,3,3-pentafluoro-2-(fluoromethoxy)propyl)-L-cysteine contains a CF2 signal centered at δ -20 ppm comprised of two AB quartets, consistent with the presence of two diastereomers, while the spectrum of urine contains a CF2 signal centered at δ -20 ppm comprised of only a single AB quartet, suggesting the predominance of one diastereomer. The in vivo predominance of one diastereomer of the alkyl FDVE-mercapturate is consistent with the previous observation that, in vivo, one diastereomer of the alkyl FDVE-glutathione conjugate is formed preferentially (14).

700 Chem. Res. Toxicol., Vol. 9, No. 4, 1996

Spracklin and Kharasch

Figure 6. Proposed mechanism of β-lyase-mediated biotransformation of FDVE-cysteine conjugates to 3,3,3-trifluoro-2(fluoromethoxy)propanoic acid.

Figure 5. 19F NMR analysis of rat urine. (A) Spectrum of urine from a rat receiving 0.3 mmol/kg FDVE. 19F resonances were not observed in the urine of control animals receiving only corn oil vehicle. (B) Spectrum of synthetic 3,3,3-trifluoro-2-(fluoromethoxy)propanoic acid. (C) Spectrum of a mixture of synthetic FDVE mercapturates: N-acetyl-S-(1,1,3,3,3-pentafluoro2-(fluoromethoxy)propyl)-L-cysteine and N-acetyl-S-(1-fluoro-2(fluoromethoxy)-2-(trifluoromethyl)vinyl)-L-cysteine. (D) Spectrum of synthetic N-acetyl-S-(1,1,3,3,3-pentafluoro-2-(fluoromethoxy)propyl)-L-cysteine after purification by column chromatography (see Experimental Procedures). (E) Spectrum of synthetic N-acetyl-S-(1-fluoro-2-(fluoromethoxy)-2-(trifluoromethyl)vinyl)L-cysteine after purification by column chromatography (see Experimental Procedures). Samples were dissolved in D2O, and chemical shifts are expressed in parts per million relative to external chlorodifluoroacetic acid (δ 0 ppm).

Discussion Previous investigations in rats in vivo have demonstrated the metabolism of FDVE to glutathione and cysteine conjugates (13, 14). Four Compound A-glutathione conjugates were isolated in rat bile, including two fluoroalkanes and two fluoroalkenes (14). The fluoroalkanes were diastereomers of S-[1,1-difluoro-2-(fluoromethoxy)-2-(trifluoromethyl)ethyl]glutathione, and the fluoroalkenes were (E)- and (Z)-S-[1-fluoro-2-(fluoromethoxy)-2-(trifluoromethyl)vinyl]glutathione. Experiments with rat liver subcellular fractions showed that all four conjugates were formed predominantly by mi-

crosomal, rather than cytosolic, glutathione transferases (14). Evidence for glutathione conjugate cleavage by γ-glutamyltranspeptidase and dipeptidases to the corresponding cysteine conjugates was provided by the appearance of the corresponding mercapturates in rat urine (13). In the present investigation, headspace GC/MS analysis showed that 3,3,3-trifluoro-2-(fluoromethoxy)propanoic acid was excreted in the urine of rats administered FDVE (0.3 mmol/kg). Furthermore, inhibition of renal cysteine conjugate β-lyase with AOAA substantially decreased the amount of 3,3,3-trifluoro-2-(fluoromethoxy)propanoic acid excreted by FDVE-treated animals. The 19F NMR spectrum of the urine of FDVE-treated animals was consistent with the formation of 3,3,3-trifluoro-2(fluoromethoxy)propanoic acid, although it could not be differentiated from that of the FDVE mercapturates which are also known to be excreted in urine (13). AOAA protection against haloalkane-cysteine conjugate-induced toxicity has been demonstrated previously and been interpreted to indicate a mechanistic role for β-lyase (19). The present observations suggest a role for β-lyase in the metabolism of FDVE to 3,3,3-trifluoro-2-(fluoromethoxy)propanoic acid. Based on analogy with structurally related compounds (17), the postulated mechanism for FDVE-cysteine conjugate metabolism to 3,3,3-trifluoro-2-(fluoromethoxy)propanoic acid is shown in Figure 6. The in vivo formation of 3,3,3-trifluoro-2-(fluoromethoxy)propanoic acid presumably occurs through the metabolism of FDVEcysteine conjugates by renal cysteine conjugate β-lyase

Anesthetic Degradate Metabolism by β-Lyase

to R-fluoroenethiol 13 and/or R-fluorothiol 14. Elimination of fluoride from 13 and 14 would give 15 and 16, respectively. Subsequent hydrolysis of either intermediate would yield 3,3,3-trifluoro-2-(fluoromethoxy)propanoic acid. Alternatively, 13 may tautomerize to 16, rather than proceed via thioketene 15. This pathway is analogous to the β-lyase-catalyzed metabolism of tetrafluoroethyl-L-cysteine to difluoroacetic acid (20). Further investigation is required to identify the relative rates of metabolism of the four FDVE-cysteine conjugates to 3,3,3-trifluoro-2-(fluoromethoxy)propanoic acid. The toxicity of 3,3,3-trifluoro-2-(fluoromethoxy)propanoic acid is unknown; however, the thioketene and/or thioacyl fluoride intermediates 15 and 16, from which it is presumably formed, are potential nephrotoxins. Covalent binding of highly reactive thioketene or thioacyl fluoride intermediates to cellular macromolecules has been proposed as a critical event in the mechanism of β-lyase-mediated fluoroalkene and fluoroalkane-cysteine conjugate nephrotoxicity (21-27). By analogy with the structurally similar compounds tetrafluoroethene and chlorotrifluoroethene, the thioketene or thioacyl fluoride intermediates derived from β-lyase-catalyzed metabolism of FDVE-cysteine conjugates may bind to tissue proteins. Indeed, covalent binding of FDVE to corticomedullary cells in rat kidneys was detected by immunohistochemistry using an antibody against trifluoroacetyl groups bound to proteins (28). Mitochondrial dysfunction is a characteristic feature of renal mitochondrial β-lyasedependent bioactivation of fluoroalkene-derived cysteine conjugates to thioketene or thioacyl fluoride intermediates (29). Mitochondrial injury, assessed by electron microscopy, was the earliest ultrastructural feature of renal tubular cell toxicity after FDVE administration in rats.2 Until recently, the metabolic disposition and mechanism of nephrotoxicity of FDVE in rats was unknown. There is now evidence that FDVE undergoes metabolism and toxification in rats by the renal β-lyase pathway. FDVE undergoes spontaneous and glutathione transferase-catalyzed conjugation to four glutathione conjugates, which are cleaved to the corresponding cysteine conjugates (13, 14). The present investigation provides evidence for the biotransformation of FDVE-cysteine conjugates to putative thioketene or thioacyl fluoride intermediates which give rise to 3,3,3-trifluoro-2-(fluoromethoxy)propanoic acid. Renal β-lyase is a logical catalyst for this reaction. The β-lyase inhibitor (aminooxy)acetic acid significantly inhibited the diuresis and increased urinary excretion of glucose, protein, and R-glutathione S-transferase caused by FDVE in rats (13). This observation implicated the β-lyase pathway in FDVE toxification, is consistent with FDVE-cysteine conjugate metabolism by β-lyase, and supports the mechanism proposed above.

Acknowledgment. Supported by grants from the National Institutes of Health (R01 GM48712), Abbott Laboratories, and a Faculty Development Award in Clinical Pharmacology from the Pharmaceutical Manufacturers Association Foundation. We thank Sid Nelson for his careful review of the manuscript.

References (1) Wallin, R. F., Regan, B. M., Napoli, M. D., and Stern, I. J. (1975) 2

E. D. Kharasch and D. Thorning, unpublished observations.

Chem. Res. Toxicol., Vol. 9, No. 4, 1996 701

(2) (3) (4)

(5)

(6)

(7) (8)

(9) (10)

(11)

(12)

(13)

(14)

(15) (16) (17)

(18)

(19) (20)

(21) (22)

Sevoflurane: A new inhalational anesthetic agent. Anesth. Analg. 54, 758-766. Hanaki, C., Fujii, K., Morio, M., and Tashima, T. (1987) Decomposition of sevoflurane by soda lime. Hiroshima J. Med. Sci. 36, 61-67. Bito, H., and Ikeda, K. (1996) Renal and hepatic function in surgical patients after low-flow sevoflurane or isoflurane anesthesia. Anesth. Analg. 82, 173-176. 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 two CO2 absorbants during low-flow anesthesia in surgical patients. Anesthesiology 77, 1064-1069. 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 degradation products with soda lime. Toxicity of the byproducts. Anesthesiology 77, 1155-1164. Gonsowski, C. T., Laster, M. J., Eger, E. I., II, Ferrell, L. D., and Kerschmann, R. L. (1994) Toxicity of compound A in rats. Effect of increasing duration of administration. Anesthesiology 80, 566573. Gonsowski, C. T., Laster, M. J., Eger, E. I., II, Ferrell, L. D., and Kerschmann, R. L. (1994) Toxicity of compound A in rats. Effect of a 3-hour administration. Anesthesiology 80, 556-565. 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 a haloalkene degradant of sevoflurane, Compound A (PIFE), in Sprague-Dawley rats. Anesthesiology 83, 1220-1232. Thummel, K. E., Kharasch, E. D., Podoll, T., and Kunze, K. (1993) Human liver microsomal enflurane defluorination catalyzed by cytochrome P-450 2E1. Drug Metab. Dispos. 21, 350-357. Kharasch, E. D., and Thummel, K. E. (1993) Identification of cytochrome P450 2E1 as the predominant enzyme catalyzing human liver microsomal defluorination of sevoflurane, isoflurane and methoxyflurane. Anesthesiology 79, 795-807. Dekant, W., Vamvakas, S., and Anders, M. W. (1994) Formation and fate of nephrotoxic and cytotoxic glutathione-S-conjugates: Cysteine conjugate β-lyase pathway. Adv. Pharmacol. 27, 115162. Kharasch, E. D., and Hankins, D. C. P450-dependent and nonenzymatic human liver microsomal defluorination of fluoromethyl-2,2-difluoro-1-(trifluoromethyl)vinyl ether (Compound A), a sevoflurane degradation product. Drug Metab. Dispos., in press. Jin, L., Baillie, T. A., Davis, M. R., and Kharasch, E. D. (1995) Nephrotoxicity of sevoflurane compound A [fluoromethyl-2,2difluoro-1-(trifluoromethyl)vinyl ether] in rats: Evidence for glutathione and cysteine conjugate formation and the role of renal cysteine conjugate β-lyase. Biochem. Biophys. Res. Commun. 210, 498-506. Jin, L., Davis, M. R., Kharasch, E. D., Doss, G. A., and Baillie, T. A. (1996) Identification in rat bile of glutathione conjugates of fluoromethyl 2,2-difluoro-1-(trifluoromethyl)vinyl ether, a nephrotoxic degradate of the anesthetic agent sevoflurane. Chem. Res. Toxicol. 9, 555-561. van Bladeren, P. J., Buys, W., Breimer, D. D., and van der Gen, A. (1980) The synthesis of mercapturic acids and their esters. Eur. J. Med. Chem. 15, 495-497. Emsley, J. W., Phillips, L., and Wary, V. (1977) in Fluorine Coupling Constants (Emsley, J. W., Feeney, J., and Sutcliffe, L. H., Eds.) Pergamon Press, New York. Commandeur, J. N. M., Stijntjes, G. L., and Vermeulen, N. P. E. (1995) Enzymes and transport systems involved in the formation and disposition of glutathione-S-conjugates. Pharmacol. Rev. 47, 271-330. Chang, Y.-C., Scott, R. D., and Graves, R. D. (1986) Function of pyridoxal 5′-phosphate in glycogen phosphorylase: 19F NMR and kinetic studies of phosphorylase reconstituted with 6-fluoropyridoxal and 6-fluoropyridoxal phosphate. Biochemistry 25, 19321939. Elfarra, A. A., Jakobson, I., and Anders, M. W. (1986) Mechanism of S-(1,2-dichlorovinyl)glutathione-induced nephrotoxicity. Biochem. Pharmacol. 35, 283-288. Commandeur, J. N. M., Brakenhoff, J. P. G., de Kanter, F. J. J., and Vermeulen, N. P. E. (1988) Nephrotoxicity of mercapturic acids of three structurally related 2,2-difluoroethylenes in the rat. Indications for different bioactivation mechanisms. Biochem. Pharmacol. 37, 4495-4504. Hayden, P. J., and Stevens, J. L. (1990) Cysteine conjugate toxicity, metabolism, and binding to macromolecules in isolated rat kidney mitochondria. Mol. Pharmacol. 37, 468-476. Hayden, P. J., Ichimura, T., McCann, D. J., Pohl, L. R., and Stevens, J. L. (1991) Detection of cysteine conjugate metabolite adduct formation with specific mitochondrial proteins using

702 Chem. Res. Toxicol., Vol. 9, No. 4, 1996

(23)

(24)

(25)

(26)

antibodies raised against halothane metabolite adducts. J. Biol. Chem. 266, 18415-18418. Hayden, P. J., Yang, Y., Ward, A. J. I., Dulik, D. M., McCann, D. J., and Stevens, J. L. (1991) Formation of difluorothionoacetylprotein adducts by S-(1,1,2,2-tetrafluoroethyl)-L-cysteine metabolites: Nucleophilic catalysis of stable lysyl adduct formation by histidine and tyrosine. Biochemistry 30, 5935-5943. Hargus, S. J., and Anders, M. W. (1991) Immunochemical detection of covalently modified kidney proteins in S-(1,1,2,2tetrafluoroethyl)-L-cysteine-treated rats. Biochem. Pharmacol. 42, R17-R20. Harris, J. W., Dekant, W., and Anders, M. W. (1992) In vivo detection and characterization of protein adducts resulting from bioactivation of haloethane cysteine S-conjugates by 19F NMR: Chlorotrifluorethene and tetrafluoroethene. Chem. Res. Toxicol. 5, 34-41. Bruschi, S. A., West, K. A., Crabb, J. W., Gupta, R. S., and Stevens, J. L. (1993) Mitochondrial HSP60 (P1 protein) and a

Spracklin and Kharasch HSP70-like protein (mortalin) are major targets for modification during S-(1,1,2,2-tetrafluoroethyl)-L-cysteine-induced nephrotoxicity. J. Biol. Chem. 268, 23157-23161. (27) Fisher, M. B., Hayden, P. J., Bruschi, S. A., Dulik, D. M., Yang, Y., Ward, A. J. I., and Stevens, J. L. (1993) Formation, characterization, and immunoreactivity of lysine thioamide adducts from fluorinated nephrotoxic cysteine conjugates in vitro and in vivo. Chem. Res. Toxicol. 6, 223-230. (28) Martin, J. L., Dodge, M. L., and Pohl, L. R. (1994) Immunochemical detection of covalently modified kidney proteins in rats treated with the sevoflurane degradation product, Compound A (sevoolefin). Anesthesiology 81, A436. (29) Anders, M. W. (1995) Mitochondrial bioactivation of cysteine S-conjugates and 4-thiaalkanoates: Implications for mitochondrial dysfunction and mitochondrial diseases. Biochim. Biophys. Acta 1271, 51-57.

TX9502103