Chem. Res. Toxicol. 1994, 7, 170-176
170
Metabolism of the Chlorofluorocarbon Substitute 1,1-Dichloro-2,2,2-trifluoroethane by Rat and Human Liver Microsomes: The Role of Cytochrome P450 2E1 Gudrun Urban, Petra Speerschneider, and Wolfgang Dekant* Institut fiir Toxikologie, Universitat Wiirzburg, 97078 Wiirzburg, FRG Received May 24, 1993"
l,l-Dichloro-2,2,2-trifluoroethane (HCFC-123) has been developed as a substitute for ozonedepleting chlorofluorocarbons. The atmospheric lifetime of HCFC-123 is expected to be much shorter than those of chlorofluorocarbons; however, due to its lower stability and the presence of carbon-hydrogen bonds, metabolism of HCFC-123 in mammals and metabolism-dependent toxicity is likely. We compared the metabolism of HCFC-123 and its analogue halothane in rat and human liver microsomes. 19F-NMR studies showed that trifluoroacetic acid is a major metabolite of HCFC-123. Besides trifluoroacetic acid, chlorodifluoroacetic acid and inorganic fluoride were identified as products of the enzymatic oxidation of HCFC-123 in rat and human liver microsomes by l9F-NMR and mass spectrometry. These metabolites were not detected in incubations with halothane. HCFC-123 and halothane were transformed by liver microsomes from untreated rats a t low rates. Microsomesfrom ethanol- and pyridine-treated rats metabolized both HCFC-123 and halothane at much higher rates. These microsomes also exhibited high rates of p-nitrophenol oxidation. p-Nitrophenol is a model substrate mainly oxidized by P450 2E1 to p-nitrocatechol. Samples of human liver microsomes showed considerable differences in the extent of HCFC-123,p-nitrophenol oxidation, and chlorzoxazonehydroxylation. In human liver microsomes, rabbit anti-rat P450 2E1 IgG recognized a single protein band corresponding in apparent molecular weight to human P450 2E1. Immunoblot analysis revealed considerable heterogenity in the P450 2E1 protein content of the human liver samples. Trifluoroacetic acid formation from HCFC-123 and halothane and p-nitrocatechol formation from p-nitrophenol were significantly reduced by the P450 2E1 inhibitor diethyldithiocarbamate. p-Nitrophenol also inhibited halothane and HCFC-123 oxidation in both rat and human liver microsomes. Moreover, the rates of trifluoroacetic acid formation from HCFC-123 and halothane correlated well with the ability of rat and human liver microsomesto oxidizep-nitrophenoland chlorzoxazone and the amount of P450 2E1 protein in liver microsomes determined by sodium dodecyl sulfatepolyacrylamide gel electrophoresis and immunoblot analysis. These data indicate that cytochrome P450 2E1 plays a major role in the metabolism of HCFC-123 and halothane in vitro. Introduction Chlorofluorocarbons and their destruction in the stratosphere have been implicated in the depletion of stratospheric ozone. Depletion of stratospheric ozone may cause adverse effects in humans including increases in cataract formation and skin cancer (I). International agreements therefore have reduced the use of chlorofluorocarbons and prompted a search for chlorofluorocarbon replacements. Hydrochlorofluorocarbons and hydrofluorocarbons have many of the beneficial properties of chlorofluorocarbons but show little ozone-depleting potential (2). The decreased atmospheric half-lives of hydrochlorofluorocarbons and hydrofluorocarbons is due to the presence of C-H bonds, which make these compounds more prone to atmospheric degradation. Since manufacture and use of hydrochlorofluorocarbons and hydrofluorocarbons may be accompanied by human exposure, toxicity, metabolism, and pharmacokinetic studies of hydrochlorofluorocarbons and hydrofluorocarbons are warranted. Metabolism of hydrochlorofluorocarbons likely proceeds through a cytochrome P450 (P450)I-mediatedoxidation
* Correspondence should be addressed to this author at the Institut
fa Toxikologie, UniversitAt Wmburg, Versbacher Straase 9, 97078
Wihzburg, FRG. Phone: +49 0931 201 3449;Fax: +49 0931 201 3446. 0 Abstract published in Aduance ACS Abstracts, February 1, 1994.
of C-H bonds. This reaction has been demonstrated to occur in themetabolism of 1,l-dichloro-1-fluoroethaneand 1,1,1,2-tetrafluoroethane(3-8). HCFC-123 is a hydrochlorofluorocarbon presently under development(9).It is a structural analogue to the widely used anesthetic halothane (l-bromo-l-chloro-2,2,2-trifluoroethane). Halothane and HCFC-123 have been demonstrated to be metabolized in vivo to trifluoroacetic acid via a reactive acyl halide as a major metabolite (10-16). The P450 enzyme P450 2E1 has been implicated in the metabolism of halothane (12) and several hydrochlorofluorocarbons and hydrofluorocarbons (4-8). Constitutive expression of P450 2E1 in rat liver is low, but enzyme activity is induced by ethanol, pyridine, and many other xenobiotics. Diseases and lifestyle factors also influence the concentration of P450 2E1 in humans (I 7-23). The objective of this work was to compare the metabolism of HCFC-123 and halothane in rat and human liver subcellular fractions and to assess the extent of HCFC123 activation and the role of P450 2E1 in this process. Experimental Section Chemicals. HCFC-123,99.96% pure (impurities were 1,1,2trichloro-1,2,2-trifluoroethane, 0.03%; 1,2-dichloro-1,1,2-triAbbreviations: HCFC-123,l,l-dichloro-2,2,2-tifluoroethane; P460, cytochrome P450; DDTC, diethyldithiocarbamate.
0893-228x/94/2707-0170$04.50/00 1994 American Chemical Society
Metabolism of 1,l-Dichloro-2,2,2-trifluoroethane fluoroethane, 0.007 5% ; percentage based on flame ionization detector response), and halothane (stabilized with 0.01 % thymol) were supplied from Hoechst (Frankfurt, FRG). Trifluoroacetic acid, chlorodifluoroacetic acid, l,l,l-trichloroethanol, p-nitrophenol, and all other chemicals were purchased from Aldrich Chemical Co. (Steinheim, FRG) in the highest purity available. Enzymes and cofactors were obtained from Sigma Chemical Co. (Deisenhofen, FRG). Animals and Treatment. Male and female Sprague-Dawley rats (Zentralinstitut fur Versuchstierzucht, Hannover, FRG; 220250 g body weight) were used for all studies. To induce P450 2E1, animals were pretreated with ethanol or pyridine. For ethanolpretreatment, rats were administered 10% (w/v) ethanol in drinking water over 10 days (24).For pyridine pretreatment, rats received an ip injection of 100 mg/kg pyridine (dissolved in isotonic sodium chloride solution) once daily for 4 consecutive days (25). All animals were fasted 18 h before sacrifice and microsome preparation. Microsome Preparation. Human liver specimens were obtained from Keystone Skin Bank (Exton, PA) or through the Liver-bank in Kiel, FRG. All procedures were performed at 4 "C. To minced livers (rats) and tissue specimens (human) was added a %fold volume (w/v) of 0.154 M KCl, and tissues were homogenized. The homogenate was then centrifuged a t 10 000 rpm for 10 min and the resulting supernatant subjected to a centrifugation a t 40 000 rpm for 60 min to sediment microsomes. The microsomal pellet was resuspended in 0.01 M phosphate buffer (pH 7.4) and centrifuged again for 60 min at 40 000 rpm. The resulting pellet was suspended in 0.01 M phosphate buffer (pH 7.4),and aliquotaof the microsomal preparations were quickly frozen in liquid nitrogen and stored a t -70 "C. Protein concentrations were determined with a Bio-Rad protein assay kit with bovine serum albumine as standard. Quantitation of Trifluoroacetic Acid and Chlorodifluoroacetic Acid. Trifluoroacetic acid and chlorodifluoroacetic acid were determined by headspace GC of the trichloroethyl ester by GC/MS-selected ion monitoring. Incubation mixtures retaining the microsomal protein (500 pL) and 50 pL of aqueous pentafluoropropionicacid (1g/L as internal standard) were placed in a 5-mL vial, vortexed, and dried overnight in an evacuated desiccatorover PzO6. Trichloroethanol(100pL) and concentrated sulfuric acid (100 pL) were added to the resulting residue, which was then sealed with a Teflon-coated rubber septum and heated to 70 "C for at least 2 h. One hundred pL of the headspace was injected into the GC with a gas-tight syringe stored at 40 "C before sampling. Retention time of trichloroethyl trifluoroacetate was 0.8 min, retention time of trichloroethyl pentafluoropropionate was 1.1 min, and retention time of trichloroethyl chlorodifluoroacetate was 2.7 min. Mass fragments of m/z 69 and mlz 127 (for trichloroethyl trifluoroacetate), m/z 71 and mlz 177 (for trichloroethyl pentafluoropropionate), and mlz 85 and m/z 131(for trichloroethyl chlorodifluoroacetate)were measured with a dwell time of 150ms for quantitationby selected ion monitoring. Quantitation was based on the integration of m/z 127, m/z 177, and m/z 131of the ion chromatograms. Calibration curves were obtained with incubations containing boiled microsomes and trifluoroacetic acid concentrations between 1 and 50 nmol. Recovery of trifluoroacetic acid from standard solutions was always greater than 90%. Instrumental Analyses. Gas chromatography/mass spectrometry was performed with an HP 5890 GC coupled to an HP 5970 mass-selective detector operating in the electron impact mode (70 eV). For trifluoroacetic acid and chlorodifluoroacetic acid determination, a fused silica column coated with DB-1 (15 m X 0.32 mm i.d., film thickness0.25pm, J & W Scientific,Folsom, CA) was used for separation. Oven temperature was 40 OC, injector port temperature was 200 "C, and transfer line temperature was 280 "C. Splitless injection (purge time 0.1 min) was used. For determination of reductive metabolites, a DB-624 capillary column (25 m X 0.32 mm i.d., film thickness 1.8 pm, J & W Scientific) and an oven temperature of 30 "C, injector port temperature of 200 "C, and transfer line temperature of 280 "C were used. Split injection with a split ratio of 10:1 was used.
Chem. Res. Toxicol., Vol. 7, No. 2,1994 171 Spectra were recorded in the scan mode from m/z 40 to m/z 250. A double-beam spectrophotometer (Hitachi U-2000, Tokyo, Japan) with an accuracy of f0.002 mAU and a reproductibility of fO.OO1 mAU was used for p-nitrocatechol quantification in 1-cm plastic cuvettes. lgFNMR spectra were recorded with a Bruker AMX 500 NMR spectrometer with a 5-mm fluorine probe operating at 470 MHz. l9Fchemical shifts are referenced downfield to C6F6 in DzO ( 6 = 0 ppm) in a coaxial tube. 19F NMR spectra were recorded with a pulse length of 6 ps and an interpulse time of 1.5 s. A total of 15 000 data points with 5000-20 000 scans were recorded for Fourier transformation. Incubation Conditions. Hydroxylation ofp-nitrophenol was determined as described (15,26). The absorbance of p-nitrocatechol was measured spectrophotometrically a t 510 nm. Final protein concentrations were 0.5 and 1 mg/mL for pretreated Sprague-Dawley rats and human liver microsomes and 2 mg/mL for naive Sprague-Dawley rat microsomes. Formation rates of p-nitrocatechol are based on a molar extinction coefficient e = 15.4 mmol-l cm-l. Diethyldithiocarbamate, a selective inhibitor for P450 2E1 (27),was used a t a final concentration of 100 pM. For all these incubations, microsomes, NADPH generating system, and diethyldithiocarbamate were preincubated for 5 min a t 25 "C in the absence ofp-nitrophenol. Incubations with HCFC123 and halothane were performed in a total volume of 1.3 mL and contained microsomes, substrates or inhibitors, as noted, and an NADPH generating system (final concentrations: 9 mM glucose 6-phosphate, 1 mM NADP+, 0.2 unit/mL glucose-6phosphate dehydrogenase) in 0.1 M phosphate buffer with 1mM EDTA (pH 7.4). Microsomes (0.5mg/mLproteinfrompretreated Sprague-Dawley rats and human donors, 2 mg/mL protein for naive Sprague-Dawley rats) and the corresponding amount of buffer were placed in 38.7-mL Wheaton serum bottles. One microliter of HCFC-123 (4.32 pmol; for halothane, the equimolar amount was 0.46 pL) was added (at 4 "C) in sealed vials (Teflon septa) and preincubated for 10 min a t 37 "C to equilibrate the gas with the liquid phase. After addition of the NADPH generating system, the mixtures were incubated for 10 and 20 min, respectively, at 37 OC in a shaking water bath. The vials were submerged in water to ensure a constant temperature in the vials. The reactions were stopped by putting the vials on ice. Two aliquota of everysample (500pL) were used for trifluoroacetic acid determination, and every incubation was repeated four times. For the NMR analyses, the microsomal protein was removed by centrifugation at 40 000 rpm for 20 min. Incubations under anaerobic conditions were conducted in 3-mL reaction vials (Pierce, Rockford, IL) containing 5 mg/mL microsomal protein from EtOH-pretreated Sprague-Dawleyrata and an NADPH generating system in 0.1 M phosphate buffer with 1 mM EDTA (pH 7.4). Final incubation volume was 1.3 mL. The vials were capped with Teflon-coated rubber septa and stored on ice until the start of the incubation. The headspace of each vial was purged for at least 5 min with a mixture of argon and oxygen (95:5v/v) via a needle passed through a septum while a separate needle provided pressure relief. One microliter of HCFC-123 (or 0.46 pL of halothane) was added, and the mixtures were incubated for 25 min in a shaking water bath as described above. One hundred microliters of the headspace was analyzed by GC/MS for volatile metabolites. The 6-hydroxylation of chlorzoxazone (28)was determined in a incubation mixture containing 0.5 mM chlorzoxazone, microsomal protein, and NADPH generating system in 100 mM phosphate buffer (pH 7.4) in a final volume of 1.0 mL. Protein concentrations were 1mg/mL. Reactions were terminated after 20 min of incubation a t 37 "C by the addition of 50 pL of 43% phosphoric acid. Denaturated protein was removed by centrifugation. 6-Hydroxychlorzoxazone was determined in the supernatant by HPLC; 50 pL of the supernatant was injected. Separation was performed with a 125 X 4 mm steel column filled with Partisil ODs-I11 (5-pm particle size) by gradient elution: solvent A: water, acidified to pH 2 with trifluoroacetic acid; solvent B: methanol; linear gradient from 20% to 100% B in 20 min; flow rate 1 mL/min. 6-Hydroxychlorzoxazone was deter-
172 Chem. Res. Toxicol., Vol. 7, No. 2, 1994
Urban et al.
mined by monitoring the absorbance at 280 nm; the retention time of 6-hydroxychlorzoxazoneunder these conditions was 14.6 min and that of chlorzoxazone was 17.3 min. Quantitation of 6-hydroxychlorzoxazonewas accomplished by using a standard curve obtained with authentic 6-hydroxychlorzoxazone. Western Blot Analysis for P460 2El. Anti-rat P450 2E1 1gG (Amersham Buchler, Braunschweig, FRG) was used to quantitate P460 2E1 content in microsomes. The antibody shows no cross-reactivitywith P450 1A1, 2B1, and 4A1. The analysis was conducted by enhanced chemiluminescence as described in the technical note RPM 259 from Amersham Buchler. Quantitation of antibody binding to P450 2E1 in immunoblota in microsomal proteins was accomplished by scanning blots with a laser densitometer (Molecular Dynamics Image Quant). The peak heights obtained from scans of immunoblota from microsomes were expressed on a relative scale in which the value for subject H4 (least intense blot) was set as 1. Preliminary experiments in which multiple dilutions of human microsomes were immunoblotted for P450 2E1 showed protein loadingsof 20 pg per lane and produced immunoblota for which particle density was reproducible in the linear range.
A
I
I
I
I
I
I
I
I
120
100
80
60
40
20
0
I
-20
PPm
0
Results I
End Products of HCFC-123and Halothane Metabolism in Liver Microsomes. 19F-NMR analysis of incubation mixtures of rat and human liver microsomes with HCFC-123 and halothane without an NADPH regeneration system showed only small signals at 6 = 87.7 attributable to trifluoroacetic acid and at 6 = 87.5 and 6 = 85.1 attributable to HCFC-123 or halothane, respectively. In the presence of microsomes, NADPH, and halothane, the resonance of trifluoroacetic acid was greatly increased and two other small, overlapping resonances appeared. These resonances were identical to the resonances of trifluoroacetamides [e.g., N-(trifluoroacety1),but were 2-aminoethanol or Ne-(trifluoroacety1)-L-lysine] not related to specific structures (Figure 1A). Identical incubations with HCFC-123 contained a resonance identical in chemical shift to trifluoroacetic acid, but not the other small resonances seen with halothane, and, in addition, a small singlet at 6 = 43.1 assigned to inorganic fluoride and a larger signal at 6 = 100.1. These resonances were not formed in the absence of NADPH and when using heat-inactivated microsomes. The signal at 6 = 100.1 coincidenced in chemical shift with that of chlorodifluoroaceticacid. Mass spectrometry after chemical derivatization definitively identified chlorodifluoroacetic acid as a HCFC-123 metabolite (Figure 2). The mass spectra obtained after chemical derivatization of the samples were identical in all major fragments and their intensities to those obtained with trifluoroacetic acid and chlorodifluoroacetic acid when treated under identical conditions. Moreover, the retention times of both metabolites were identical to those of the synthetic references. Under conditions favoring reductive metabolism (5 % 02,95 % argon) of halothane (29)no HCFC-123metabolites were detected (Figure 3). The microsomes used, however, transformed halothane to l-chloro-2,2,2-trifluoroethane and l-chloro-2,2-difluoroethene(30) (Figure 3). The Role of P450 2El in the Oxidation of HCFC-123 and Halothane in Rats. HCFC-123 and halothane were incubated with microsomes obtained from rats after different pretreatments, and metabolism was assessed by quantitation of trifluoroacetic acid. Only at low concentrations of HCFC-123, the enzymatic formation of trifluoroacetic acid could be quantified. At higher concentrations of HCFC-123 in the gas phase a spontaneous formation of trifluoroacetic acid was observed, which was
' ,
I
I
I
I
I
I
120
100
80
60
40
20
0
I
-20
I
I
-40
-60
ppm
Figure 1. l9F-NMR spectra of metabolites produced from halothane (spectrum A) and l,l-dichloro-2,2,2-trifluoroethane (spectrum B) by human liver microsomes. Spectrum A trifluoroacetic acid, 6 = 87.7; halothane, 6 = 87.5. The resonance at 6 = 87.6 consisted of at least two unresolved signals. Spectrum B: trifluoroaceticacid,6 = 87.5; l,l-dichloro-2,2,2-trifluoroethane, 6 = 85.1; chlorodifluoroacetic acid, 6 = 100.1; metabolite not identified, 6 = 104.3; and inorganic fluoride, 6 = 43.1. External CeF6 was used as chemical shift reference, 6 = 0 ppm.
-
I
0.5
I
1.o
I
1.5 Time (min)
I
I
2.0
2.5
Figure 2. Gas chromatogramand mass spectra of a derivatized incubation mixture of l,l-dichloro-2,2,2-trifluoroethaneand human liver microsomes. Trichloroethyl trifluoroacetate, t R = 0.85 min; trichloroethyl chlorodifluoroacetate, t R = 2.8 min. dependent on the concentration of HCFC-123, but not dependent on active microsomes, and occurred also in the presence of bovine serum albumin (Figure 4). In experiments on the enzymatic metabolism of HCFC-123, conditions were selected in which enzymatic oxidation of HCFC-123 was significantly greater than spontaneous oxidation (Table 1). Since p-nitrophenol oxidation was not observed in microsomes of kidney and lung of untreated and ethanol-treated rats (data not shown),all experiments on the metabolism of halothane and HCFC-123 were performed in liver microsomes only. Microsomes from native rats formed only low amounts of trifluoroacetic acid; these rates were, however, significantly different from the rates of spontaneous oxidation of HCFC-123 (Table 1). The formation of trifluoroacetate was dependent on time, microsomal protein concentration, and NADPH at low
Metabolism of 1,l-Dichloro-2,2,2-trifluoroethane
Chem. Res. Toxicol., Vol. 7, No. 2, 1994 173
Table 1. Metabolism of HCFC-123 and Halothane by Rat Hepatic Microsomes. rate of trifluoroacetic acid formation [nmol.mg1.(20 min)-l] rata and treatment HCFC-123 halothane Sprague-Dawley rat (-NADPH) 1.1f 0.2 1.9 f 0.2 Sprague-Dawley rat Sprague-Dawley rat, EtOH-pretreated Sprague-Dawley rat, EtOH-pre. + 100 pM DDTC Sprague-Dawley rat, pyridine-pretreated Sprague-Dawley rat, pyridine-pre. 100 pM DDTC
4.1 f 2.3 25.8 f 4.1 8.3 k 0.5 31.3 f 6.6 21.3 f 2.8
+
3.1 f 1.3 27.7 f 7.1 13.8 f 3.9 36.9 f 2.1 24.2 f 3.1
0 Incubations with HCFC-123and halothane were performed in a t otalvolume of 1.3 mL and contained microsomes, substrates or inhibitors as noted and an NADPH generating system. Reactions were performed in sealed flasks. DDTC = diethyldithiocarbamate.
F H
$ C m
U C
3
a I
0.5
I
I
1.o
1.5
I
2.0
2.5
Time (min)
Figure 3. Gas chromatography/mass spectrometry of the headspace of incubations of halothane (traceI) and 1,l-dichloro2,2,2-trifluoroethane (trace 11) under reduced oxygen concentrations and mass spectra of volatile metabolites. Halothane incubations resulted in l-chloro-2,2-difluoroethene (B) and l-chloro-2,2,2-trifluoroethane(C). In l,l-dichloro-2,2,2-trifluoroethaneincubations, no volatile metabolites were detected. Other peaks in the spectra shown are COz (A), halothane (D), and l,l-dichloro-2,2,2-trifluoroethane(E). Incubations were performed as described in the Experimental Section.
VOI
Ye HCFC-123
Figure 4. Spontaneous,nonenzymatic oxidation of 1,l-dichloro2,2,2-trifluoroethaneto trifluoroaceticacid at various 1,l-dichloro2,2,2-trifluoroethaneconcentrations. Comparison of trifluoroacetic acid formationin metabolicallycompetentliver microsomes (e), boiled liver microsomes (A), and albumin (m) as model proteins. Incubations were performed containing2 mg of proteid mL as described in the Experimental Section. HCFC-123 concentrations (data not shown). No significant differences were seen in the extent of HCFC-123, halothane, and p-nitrophenol oxidation between microsomes from male and female rats (data not shown). The rates of trifluoroacetic acid formation were increased approximately 10-fold by pretreatment of rata with the P450 2E1 inducers, ethanol and pyridine (Table 1). Microsomes from pyridine-induced rata showed a slightly increased formation of trifluoroacetic acid from HCFC-
123 when compared with microsomes form ethanolpretreated rata. Very similar rates of trifluoroacetic acid formation were observed using halothane as substrate when compared to HCFC-123 (Table 1). Inhibition of P450 2E1 with diethyldithiocarbamate resulted in a more than 60 $7, decrease in the rates of trifluoroacetic acid formation from both halothane and HCFC-123. Trifluoroacetic acid formation was also inhibited competively byp-nitrophenol (data not shown). The rates of trifluoroacetic acid formation in treated and nontreated rat liver microsomes correlated well with the ability of the microsomes to oxidize p-nitrophenol to the corresponding catechol for ethanol pretreatment, but not for pyridine pretreatment (Table 2). Pyridine pretreatment did not cause a large difference (20-25% increase in reaction rates) in the extent of HCFC-123 and halothane metabolism when compared to ethanol pretreatment. In contrast, pyridine pretreatment caused almost 400 5% higher rates for p-nitrophenol oxidation when compared with ethanol pretreatment. When a rabbit antirat P450 2E1 IgG was used to quantitate the relative amounts of P450 2E1 protein in rat liver microsomes after the different pretreatments (Figure 51, pyridine treatment resulted in a 10-fold increase in P450 2E1 concentration relative to untreated rats. The extent of p-nitrophenol oxidation with rat liver microsomes after different treatments correlated very well (r = 0.99) with the blot intensities. Only a weak correlation was obtained between antibody staining and the extent of HCFC-123 oxidation (r = 0.63). The Role of P450 2El in HCFC-123 Metabolism in Human Liver Microsomes. Microsomal Cytochrome P-4502E1 Content. Rabbit anti-rat P450 2E1 IgG recognized a single protein band in rat and human liver microsomes that corresponded in apparent molecular weight to purified human P450 2E1 (27). The Western blot analysis revealed considerable heterogeneity in the microsomal P450 2E1 protein content in the seven human liver specimens examined (Figure 6). Donors were, to our knowledge, not treated with medications known to induce P450 2E1 and, according to the information available, did not consume excessive amounts of alcohol or sulfur from diabetes. Due to low amounts of sample available, complete data on the metabolic capacity of human liver microsomes for the substrates used here could only be obtained with samples from five of the seven individuals. HCFC-123 was biotransformed to trifluoroacetic acid in widely differing rates in human liver microsomes (Table 3). Moreover, different rates of p-nitrophenol oxidation and chlorzoxazone 6-hydroxylation were observed. Chlorzoxazone 6-hydroxylation is exclusively catalyzed by human P450 2E1(28). The extent of HCFC-123 oxidation observed in different samples of human liver microsomes correlated well with the ability of the microsomes to oxidize chlorzoxazone (r = 0.87) and p-nitrophenol (r = 0.83) and
Urban et al.
174 Chem. Res. Toxicol., Vol. 7, No.2, 1994
Table 2. Metaboliem of pNitropheno1 by Rat Hepatic Microsomes. p-nitrocatechol [nmol.mg-1420 min)-l] rata and treatment antibody stainingintensity no DDTC in the presence of 300 p M DDTC Sprague-Dawleyrat 1 2.0 f 0.2 1.1f 0.1 2 7.4 f 0.4 2.2 f 0.4 Sprague-Dawleyrat, EtOH-pretreated Sprague-Dawleyrat, pyridine-pretreated 10 31.4 f 1.3 7.8 f 0.4 0 Hydroxylationofp-nitrophenol(PNP)wah determinedasdescribed. The a b m r h c e of p-nitrocatecho1was measured spectrophotometridy at 510 nm. Final protein concentrations were 0.5 and 1mg/mL for pretreated Sprague-Dawleyrata and 2 mg/mL for naive Sprague-Dawley rat microsomes. Formation rates of p-nitrocatechol are based on a molar extinction coefficient of 15.4 mmol-1 cm-'. DDTC = diethyldithiocarbamate. Immunoblottingwas performed after SDS-PAGE, and the relative amounts of P450 2E1were quantified by enhanced chemiluminescence. 2 pg of microsomal protein was applied to each lane of the gel.
subject H1 H4 H5 H6 H7
Table 3. Metabolism of pNitropheno1. Chlorzoxazone, and HCFC-123 by Human Liver Microsomes. p-nitrocatechol chlorzoxazone trifluoroacetic gender/age from p-nitrophenol 6-hydroxylation re1 amt of acid from HCFC-123 (Yr) [nmol=ma1*(20 min)-l] [nmol=mg1-(20min)-I] P450 2E1 protein [nmol*mg1*(20min)-l] m/24 15.4 f 0.5 99.2 f 7.6 2.4 14.7 f 4.9 6.4 f 1.3 43.7 f 3.2 1.0 5.2 f 0.2 f/53 152.3 f 8.5 7.7 46.4 f 8.3 26.0 f 1.9 f/55 4.8 15.8 f 2.6 16.6 f 0.4 98 f 7.4 f/62 57.7 f 6.5 1.8 7.4 f 2.3 12.1 f 0.3 m/63
0 Incubations with HCFC-123were performed in a total volume of 1.3 mL and contained microsomes and substrates or inhibitors as noted. Values are X f SD of four incubations. All human liver samples were seronegativefor HIV and hepatitis. For procedures, see Experimental Section. Relative amounts of P450 2E1 proteins were determined by Western blotting. 20 pg of microsomal protein was applied to each lane of the gel for quantitation of P450 2E1 protein concentrations.
Daltons 58100
58100
39800
39800
29000
29000 14300
14300
1 2 3 4 Figure 5. Immunostaining (Western blotting) of rat liver microsomal proteins with a rabbit anti-rat P450 2E1 antibody. Lane 1,protein standard; lane 2,liver microsomesfrom pyridinepretreated rats; lane 3, liver microsomes from untreated rats, lane 4, liver microsomes from ethanol-pretreated rats. All lanes contain 2 pg of microsomal protein.
with the amount of P450 2E1 indicated by Western blots (r = 0.88) (Figure 7). Antibody blot intensity also correlated well with the extent of p-nitrophenol oxidation (r = 0.92). Low concentrations of the P450 2E1 inhibitor diethyldithiocarbamate again inhibited HCFC-123 oxidation by more than 65 % in two samples tested (data not shown).
Discussion OurresultsdemonstratethatbothhalothaneandHCFC123 are biotransformed to trifluoroacetic acid. Trifluoroacetic acid has long been known as a major halothane metabolite (10). The l9F-NMR spectra of halothane incubations showed two additional resonances attributed to amides likely formed by the reaction of a metabolically formed trifluoroacetylating agent with amines. Both proteins and lipids have been identified as targets for the covalent binding of trifluoroacetyl chloride, the reactive metabolite formed from halothane by P450-dependent oxidation (10, 31). The identification of trifluoroacetic acid as a major microsomal metaboliteof the hydrochlorofluorocarbon demonstrates that an identical reaction occurs with HCFC-123. Trifluoroacetic acid is formed by
1 2 3 4 5 6 7 8 9 Figure 6. Immunostaining (Western blotting) of human liver microsomal proteins with a rabbit anti-rat P450 2E1 antibody. 20 pg of microsomal protein was used with human liver microsomes. Lane l,protein stand=& lane 2,rat liver microsomes, ethanol-pretreated, 2.5 pg of protein; lanes 3-9; human liver microsomes from subjects 1-7. Due to low amounts of sample available, microsomes from subjects 2 and 3 were not tested for 6-chlorzoxazone hydroxylation.
the oxidation of the carbon-hydrogen bond in HCFC-123 followed by the loss of hydrochloric acid .togive trifluoroacetyl chloride. This reactive acylating agent may bind to proteins. Indeed, studies on the in vivo metabolism and covalent binding of HCFC-123 using immunoblotting with a hapten-specific anti-trifluoroacetic acid protein serum and l9F-NMRspectra identified Nc-trifluoroacetylated lysine residues in proteins. Trifluoroacetic acid was also identified as a major urinary metabolite of HCFC123 in rats (14, 32, 33). However, our microsomal incubations also showed differences in the structures of metabolites formed from halothane and HCFC-123. 19FNMR and mass spectrometry identified chlorodifluoroacetic acid as a metabolite formed from HCFC-123, but not from halothane. The mechanism for the enzymatic formation of chlorodifluoroacetate is unclear. Chlorodifluoroacetic acid may be formed from HCFC-123 by a P45O-mediated oxidation of the potential metabolite or decomposition product l,l-dichloro-2,2-difluoroethene. Oxidation of this olefin by P450 should proceed under rearrangement and finally give chlorodifluoroacetate. Indeed, chlorodifluoroacetatewas formed as a metabolite
Chem. Res. Toxicol., Vol. 7, No. 2, 1994 175
Metabolism of 1,l -Dichloro-2,2,2-trifluoroethane
F
i
? .HF
+(. CI
>=cJ12L
F
F
0
10
20
30
40
50
I 10
20
30
40
50
Trlfluoroacetlcacid from HCFC.123 (nmol/mgl20 min)
"
I
0
i o
ao
30
40
O
"
Figure 8. Metabolic oxidation of l,l-dichloro-2,2,2-trifluoroethaneby microsomal cytochromeP450to trifluoroacetylchloride ahd a possible mechanism of chlorodifluoroaceticacid formation from l,l-dichloro-2,2,2-trifluoroethane.
Trlfluoroawtlc acid from HCFC-123 (nmoUmg/PO min)
04 0
V
so
Trlfluoroacetlcacld from HCFC-123 (nmoi/mg/PO mln)
Figure 7. Correlation of HCFC-123 oxidation and P450 2E1 protein content, chlorzoxazone 6-hydrotylation, and p-nitrophenol oxidation in human liver microsomes. of l,l-dichloro-2,2-difluoroethene (34). 1,l,-Dichloro-2,2difluoroethene could be formed by a base-catalyzed hydrogen fluoride elimination from HCFC-123 or by cytochrome P450,which may catalyze the transformation of aliphatic hydrocarbons to olefins (36-37) (Figure 8). The proton in HCFC-123 is more acidic than the proton in halothane, thus facilitating hydrogen fluoride elimination in HCFC-123. This may explain the differences seen in the products from halothane and HCFC-123. The enzyme P450 2E1 has bees implicated in the oxidation of many low molecular weight xenobiotics (27). Our results confirm that this enzyme plays a major role in the oxidation of halothane at low gas phase concentrations and demonstrate that this enzyme is also responsible for the oxidation of HCFC-123 under the conditions used.
Rat liver microsomes prepared from rats treated with the P450 2E1 inducers ethanol and pyridine had significantly higher rates of trifluoroacetatefrom both halothane and HCFC-123. The increase in chlorofluoroalkane oxidation was paralelled by an increase in the oxidation of p-nitrophenol and by increased P450 2E1 protein concentrations. P450 2E1 is a high-affinity, low-capacity enzyme system for the oxidation of haloalkanes and has been shown to be the primary enzyme-oxidizinghalothane at low concentrations (12). As observed with other P450 2E1 substrates, a t higher substrate concentrations, enzymes P450 2B could participate in the oxidatioh. In our experiments, the participation of P4502B1/2 could not be determined due to the nonenzymatic hydrolysis of HCFC123 at higher concentrations in the gas phase. However, P450 2B seems only to be active at high substrate concentrations (12,38,3!3);kinetic investigationswith the structurally similar 1,1,1,2-tetrafluoroethanedid not indicate a major contribution to oxidation by this isoform (5). In human liver microsomes, HCFC-123 oxidation was significantly higher than in uninduced rats and correlated well with pnitrophenol oxidation and chlorzoxazone hydroxylation. Chlorzoxazone is a specific probe for human cytochrome P450 2E1. Moreover, HCFC-123 oxidation could be inhibited by low concentrations of diethyldithiocarbamate. Also, staining with an anti-rat P450 2E1 IgG showed considerable differences in the amount of P450 2E1 protein present in the human liver samples, which also correlated with the extent of HCFC123 oxidation. Large differences in the concentration of P450 2E1 have been described in other studies in human liver and are likely related to diet and life style factors or certain diseases. P4502E1 seems to be a major cytochrome P46O in human liver (40). Further support for the involvement of P450 2E1 in the oxidation of HCFC-123 is provided by in uiuo gas uptake studies. Diallyl sulfide, a suicide inhibitor of P450 2E1 (411, almost completely inhibited the metabolism-dependent uptake of HCFC. 123and prevented its biotransformation to trifluoroacetic acid in rats (33). In summary, our results support a major role for P450 2E1 oxidation in rat and human HCFC-123 metabolism and demonstrate a much higher capacity of human liver to bioactivate HCFC-123 to toxic metabolites. These observations indicate that the rat may not be a suitable model to study HCFC-123 metabolism. The differences in HCFC-123 oxidation observed and its correlation to the P4502E1 content suggested that individualsexpressing different cytochrome P450 2E1 concentrations in liver react differently to adverse effects of HCFC-123 based on metabolic activation.
176 Chem. Res. Toxicol., VoE. 7, No. 2, 1994
Acknowledgment. This work was supported by the Program for Alternative Fluorocarbon Toxicity Testing and the Doktor-Robert-Pfleger-Stiftung (mass spectrometry equipment). The authors wish to thank Dr. R. Kerssebaum, Bruker Analytische Messtechnik GmbH (Rheinstetten, FRG), and Dr. M. Horn (Universitit Wiirzburg) for recording the NMR spectra, and Hannelore Popa-Henning for assistance in the preparation of the manuscript.
References (1) . , McFarland. M.. and Kave. J. (1992)Chlorofluorocarbonsand ozone.
Photochem: Photobiol.-56, 911-929. (2) Fisher. D. A.. Hales. C. H.. Filkin, D. L., KO,M. K. W., Sze, N. D., Connell, P. S.,Wuebbles, D. J., Isaksen, I. S. A., and Stordal, F. (1990). Model calculations of the relative effects of CFCs and their replacements on stratospheric ozone. Nature 344, 508-512. (3) Harris, J. W., and Anders, M. W. (1991). Metabolism of the hydrochlorofluorocarbon 1,2-dichloro-l,l-difluoroethane.Chem. Res. Toxicol. 4, 180-186. (4) Olson, M. J., Johnson, J. T., OGara, J. F., and Surbrook, S. 2. J. (1991). Metabolism in vivo and in vitro of the refrigerant substitute 1,1,1,2-tetrafluoro-2-chloroethane.Drug Metab. Dispos. 19, 10041011. (5) Olson, M. J., Kim, S. G., Reidy, C. A,, Johnson, J. T., and Novak, R. F. (1991). Oxidation of 1,1,1,2-tetrafluoro6thanein rat liver microsomes is catalyzed primerily by cytochrome P-450IIEl. Drug. Metab. Dispos. 19, 298-303. (6) Olson, M. J., Reidy, C. A,, and Jobson, J. T. (1990). DefluQrination of 1,1,1,2-tetrafluoroethane(R-134a) by rat hepatocytes. Bioehem. Biophys. Res. Comntun. 166,1390-1397. (7) Olson, M. J., and Surbrook, S.E. J. (1991). Deflilorination of the CFC-substitute 1,1,1,2-tetrafluoroethane:Comparison in human, rat and rabbit hepatic microsomes. Toxicol. Lett. 59,89-99. (8)Surbrook, S. E. J., and Olson, M. J. (1992). Dominant role of cytochrome P-450 2E1 in human hepatic microsomal olidation of the CFC-substitute 1,1,1,2-tetrduoroethane.Drug Metab. Dispos. 20,518-524. (9) Malley, L. A., Carakostas, M. C., Hansen, J. F., Trochimowicz, H. J., and Rusch, G. M. (1991). Chronic toxicity of hydrochlorofluorocarbon HCFC-123. Toxicologist 11, 103. (10) Cohen,E. N.,Trudell, J. R.,Edmunds, H. N., and Watson, E. (1975). Urinary metabolites of halothane in man. Anesthesiology 43,392401. (11) Gandolfi, A. J., White, R. D., Sipes, 1. G., and Pohl, L. R. (1980). Bioactivation and covalent binding of halothane in vitro. Studies with [SHI- and [W]-halothane. J.Pharmacol. Exp. Ther. 214,721725. (12) Gruenke, L. D., Konopka, K., Koop, D. R., and Waskell, L. A. (1988). Characterization of halothane oxidation by hepatic microsomes and purified cytochromes P-450 using a gas chromatografic mass spectrometric assay. J. Pharmacol. Exp. Ther. 246,454-459. (13) Harris, J. W., and Anders, M. W. (1991). In vivo metabolism of the hydrochlorofluorocarbon1,I-dichloro-1-fluoroethane (HCFC-141b). Biochem. Pharmacol. 41, R13-Rl6. (14) Lind, R. C., Gandolfi, A. J., and Hall, P. d. 1. M. (1989). The role of oxidative biotransformation of halothane in the guinea pig model of halothane-associated hepatotoxicity. Anesthesiology 70, 649653. (15) Harris, J. W., Pohl, L. R., Martin, J. L., and Anders, M. W. (1991). Tissue acylation by the chlorofluorocarbonsubstitute l,l,-dichloro2,2,2-trifluoroethane (HCFC-123). Proc. Natl. Acad. Sci. U.S.A. 88,1407-1410. (16) Sipes, I. G., Gandolfi, A. J., Pohl, L. R., Krishna, G., and Brown, B. R. (1980). Comparison of the biotransformationand hepatotoxicity of halothane and deuterated halothane. J. Pharmacol. Erp. Ther. 214,716-720. (17) Albano, E., Tomasi, A,, Oersson, J. O., Ingelman-Sundberg, M., and Dianzani, M. U. (1991). Role of ethanol-inducible cytochrome P450 (P450IIE1) in catalyzing the free radical activation of aliphatic alcohols. Biochem. Pharmacol. 41,1895-1902. (18) Kim, S. G., and Novak, R. F. (1990). Induction of rat hepatic P45OIIE1 (CYP 2E1) by pyridine: evidence for a role of protein synthesis in the absence of transcriptional activation. Biochem. Biophys. Res. Commun. 166,1072-1079. (19) Koop, D. R. (1992). Oxidative and reductive metabolism by cytochrome P450 2E1. FASEB J. 6, 724-730.
Urban et al. (20) Koop, D. R., andTiemey, D. J. (1990). Multiple mechanisms in the regulation of ethanol-inducible cytochrome P450IIEl. Bioassays 12,429-435. (21) Yang,C. S., Brady, J. F., and Hong, J.-Y. (1992). Dietary effects on cytochromes P450, xenobiotic metabolism and toxicity. FASEB J. 6,737-744. (22) Yoo, J. S., Ning, S. M., Pantuck, C. B., and Yang, C. S. (1991). Regulation of hepatic microsomal cytochrome P450llE1 level by dietary lipids and carbohydrates in rata. J. Nutr. 121,959-965. (23) Yun, Y.-P., Cassaza, J. P., Sohn, D. H., Veech, R. L., and Song, B. 3. (1992). Pretranslationalactivationof cytochromeP450IIE during ketosis induced by a high fat diet. Mol. Pharmacol. 41,474479. (24) Lieber, C. S.,Lasker, J. M., DeCarli, L. M., and Wojtowicz,T. (1988). Role of acetone, dietary fat and total energy intake in induction of hepatic microsomal ethanol oxidizing system. J.Pharmacol. Exp. Ther. 247, 791-795. (25) Kim, S. G., Williams, D. E., Schuetz, E. G., Guzelian, P. S., and Novak, R. F. (1988). Pyridine induction of cytochrome P-450 in the rat: role of P-450j (alcohol-inducibleform) in pyridine N-oxidation. J. Pharmacol. Exp. Ther. 246, 1175-1182. (26) Koop, D. R. (1986). Hydroxylation of p-nitrophenol by rabbit ethanol-induciblecytochromeP-450 isoenzyme3a. Mol. Pharmucol. 29,399-404. (27) Guengerich,F.P.,Kim,D.-H.,andIwasai,M. (1991). Roleofhuman cytochromeP-450IIEl in the oxidation of many low molecularweight cancer suspects. Chem. Res. poxicol. 4,168-179. (28) Peter, R., Bhker, R., Beaune, P. H., Iwasaki, M., Guengerich,F. P., and Yang, C. S.(1990). Hyroxylation of chlorzoxazone as a specific probe for human liver cytochrome P-450IIE1. Chem. Res. Toxicol. 3,566-573. (29) Baker, M. T., Nelson, R. M., and Van Dyke, R. A. (1983). The release of inorganic fluoride from halothane and halothane metabolities by cytochrome P-450,hemin, and hemoglobin. Drug. Metab. Dispos. 11, 308-311. (30) Trudell, J. R., Bbterling, B., and Trevor, A. J. (1982). Reductive metabolism of halothane by human and rabbit cytochrome P-450. Mol. Pharmacol. 21, 710-717. (31) Mtiiler, R., Stier, A. (1982). Modification of liver microsomal lipids by halothane metabolites; a multi nuclear NMR spectroscopicstudy. Naunyn-Schmiedeberg’s Arch. Pharmacol. 321,234-237. (32) Harris, J. W., Jones, J. P., Martin, J. L., LaRosa, A. C., Olson, M. J., Pohl., L. R., and Anders, M. W. (1992). Pentahaloethane-based chlorofluorocarbonsubstitutes and halothane: correlation of in vivo hepatic protein trifluoroacetylation and urinary trifluoroacetic acid excretion with calculated enthalpies of activation. Chem. Res. Toxicol. 5, 720-725. (33) Loizou, G. D., Urban, G., Dekant, W., and Anders, M. W. (1993). Gas-uptake pharmacokinetics of 2,2-dichloro-l,l,l-trifluoroethane (HCFC-123). Drug Metab. Dispos. (in press). (34) Commandeur, J. N. M.,Oostendorp,R.A. J., Schoofs,P.R.,Xu,B., and Vermeulen,N. P. E. (1987). Nephrotoxicity and Hepatotoxicity of l.l.-dichloro-2.2.-difluoroethvlene in the rat. Biochem. Pharmacii. 36, 4229-4237. (35) Gueneerich. F. P., and Kim. D.-H. (1991). Enzvmatic oxidation of ethylcarbaxkate &vinyl carbamate and ita roleas an intermediate in the formation of 1,”-ethenoadenosine. Chem. Res. Toxicol. 4, 413-421. (36) Kassahun,K.,andBaillie,T.A.(1993). CytochromeP-450-mediated dehydrogenation of 2-n-propyl-2(E)-pentenoic acid, a pharmacologically-activemetabolite of valproic acid, in rat liver microsomal preparations. Drug. Metab. Dispos. 21, 242-248. (37) Korzekawa, K. R., Jones, J. P.,and Gillem, J. R. (1990). Theoretical studies on cytochrome P-450 mediated hydroxylation. A predictive model for hydrogen atomabstractiom. J.Am. Chem. SOC.112,70427046. (38) Nakajima,T., Wang, R.-S.,Murayama, N., andSato,A. (1990).Three forms of trichloroethylene-metabolizingenzymes in rat liver induced by ethanol, phenobarbital, and 3-methylcholanthrene. Toxicol. Appl. Pharmacol. 102,546-552. (39) Nakajima, T., Wang, R. S., Elovaara, E., Park, S.S.,Gelboin, H. V., and Vainio, H. (1992). A comparative study on the contribution of cytochrome P450 isoenzymes to metabolism of benzene, toluene and trichloroethylene in rat liver. Biochem. Pharmacol. 43, 251257. (40) Thummel, K. E., Kharasch, E. D., Podoll, T., and Kunze, K. (1993). Human liver microsomal enflurane deflurination catalyzed by cytochrome P-450 2E1. Drug Metab. Dispos. 21, 350-357. (41) Brady, J. F., Wang, M.-H., Hong, J.-Y., Xiao, F., Li, Y., Yoo, J.-S. H., Ning, S.M., Lee, M.-J., Fukuto, J. M., Gapac, J. M., and Yang, C. S.(1991). Modulation of rat hepatic microsomal monooxygenase enzymes and cytotoxicity by diallyl sulfide. Toxicol. Appl. Pharmacol. 108, 342-354.
