Chem. Res. Toxicol. 1992,5, 720-725
720
Pentahaloethane-Based Chlorofluorocarbon Substitutes and Halothane: Correlation of in Vivo Hepatic Protein Trifluoroacetylation and Urinary Trifluoroacetic Acid Excretion with Calculated Enthalpies of Activation James W. Harris,+Jeffrey P. Jones,+Jackie L. Martin,$*$ Angela C. LaRosa,$ Michael J. Olson,ll Lance R. Pohl,t and M. W. Anders*?+ Department of Pharmacology and Environmental Health Sciences Center, University of Rochester School of Medicine, 601 Elmwood Avenue, Rochester, New York 14642, Laboratory of Chemical Pharmacology, National Heart, Lung, and Blood Institute, National Institutes of Health, Bethesda, Maryland 20892, Department of Anesthesiology and Critical Care Medicine, The Johns Hopkins Medical Institutions, Baltimore, Maryland 21205, and Biomedical Science Department, General Motors Research Laboratories, 30500 Mound Road, Warren, Michigan 48090 Received April 17, 1992
The hydrochlorofluorocarbons (HCFCs) 2,2-dichloro- l , l , 1-trifluoroethane (HCFC-123) and
2-chloro-1,1,1,2-tetrafluoroethane (HCFC-124) and the hydrofluorocarbon (HFC) pentafluoroethane (HFC-125) are being developed as substitutes for chlorofluorocarbons that deplete stratospheric ozone. The structural similarity of these HCFCs and HFCs to halothane, which is hepatotoxic under certain circumstances, indicates that the metabolism and cellular interactions of HCFCs and HFCs must be explored. In a previous study [Harris et al. (1991) Proc. Nutl. Acad. Sci. U.S.A. 88, 14071, similar patterns of trifluoroacetylated proteins (TFA-proteins) were detected by immunoblotting with anti-TFA-protein antibodies in livers of rats exposed to halothane or HCFC-123. The present study extends these results and demonstrates that in vivo TFA-protein formation resulting from a6-h exposure to a 17% atmosphere of these compounds follows the trend: halothane = HCFC-123 >> HCFC-124 > HFC-125. The calculated enthalpies of activation of halothane, HCFC-123, HCFC-124, and HFC-125 paralleled the observed rate of trifluoroacetic acid excretion in HCFC- or HFC-exposed rats. Exposure of rats to a range of HCFC-123 concentrations indicated that TFA-protein formation was saturated at an exposure concentration between 0.01 7% and 0.1 % HCFC-123. Deuteration of HCFC-123 decreased TFAprotein formation in vivo. Urinary trifluoroacetic acid excretion by treated rats correlated with the levels of TFA-proteins found after each of these treatments. No TFA-proteins were detected in hepatic fractions from rats given 1,1,1,2-tetrafluoroethane(HFC-l34a),which is not metabolized to a trifluoroacetyl halide. Because halothane hepatitis may be associated with an immune response directed against TFA-proteins in the livers of susceptible individuals, the quantity of TFA-proteins resulting from exposure to HCFCs or HFCs may be an important index for determining safe exposure limits for HCFCs or HFCs.
Introduction Hydrochlorofluorocarbons (HCFCs) and HFCs are being developed as substitutes for chlorofluorocarbons (CFCs) that deplete stratospheric ozone (1-3). HCFC123,HCFC-124,HFC-125, HCFC-225ca,and HCFC-225cb (seeTable I)are candidate CFC substitutes, and all contain a geminal dihalomethyl group (CHX2) and may be metabolized to an acyl halide. HCFC-123 and halothane are metabolized in vivo to a reactive intermediate, pre-
* Author to whom correspondence should be addressed. + University of Rochester School t National Institutes of Health.
of Medicine.
8 The Johns Hopkins Medical Institutions. 11 General Motors Research Laboratories.
Abbreviations: CFC, chlorofluorocarbon;HCFC, hydrochlorofluorocarbon; HFC, hydrofluorocarbon;halothane, 2-bromo-2-chloro-l,l,1HCFCtrifluoroethane; HCFC-123, 2,2-dichloro-l,l,l-trifluoroethane; HFC-125, pentafluoroethane; 124, 2-chloro-1,1,1,2-tetrafluoroethane; HCFC-225ca,3,3-dichloro-1,1,1,2,2HFC-l34a, 1,1,1,2-tetrafluoroethane; pentafluoropropane;HCFC-225cb,1,3-dichloro-l,1,2,2,3-pentafluoropropane; NMR, nuclear magnetic resonance spectroscopy; GC-MS, gas chromatogmphy-maw spectrometry; SDS-PAGE, sodium dodecyl sulfa+ polyacrylamide gel electrophoresis; TFA-proteins, Nf-(trifluoroacetyl)lysine-containing proteins.
sumably trifluoroacetyl chloride,that reacts with proteinbound lysine to yield similar patterns of trifluoroacetylated hepatic proteins, as determined by SDS-PAGE and immunoblotting with anti-TFA-protein serum (4). Exposure to halothane may cause a rare, fulminant hepatitis that is thought to result from an immunologically mediated hypersensitivity reaction directed against trifluoroacetylated proteins (5-8). Because the pathway for oxidative biotransformation of halothane, HCFC-123, HCFC-124, and HFC-125 may be similar, but may differ in the rate of metabolism, exposure to HCFCs and HFCs may result in increased risk for the development of hepatitis in susceptible individuals. The objectives of these studies were to measure the relative potential for hepatic protein trifluoroacetylation in halothane-, HCFC-123-, HCFC-124-, and HFC-125exposed rata and to determine whether urinary trifluoroacetic acid excretion and calculated enthalpies of activation could be used as an index of in vivo HCFC and HFC metabolism and protein trifluoroacetylation. Knowledge of in vivo rates of metabolism and TFA-protein formationas a result of HCFC or HFC exposure may prove 0 1992 American Chemical
Society
CFC Substitutes:
Hepatic Acylation Potentials
Chem. Res. Toxicol., Vol. 5, No.5,1992 721
Table I. HCFC and HFC Substitutes for Ozone-Depleting CFCs* amount of CFC chlorofluorocarbon used in U.S., (CFC) 1986 (metric tons) major uses CClsF (CFC-11) 90 000 flexible foam, rigid polyurethane foam, refrigeration, air-conditioning CCl2F2 (CFC-12) 143 000 rigid polyurethane and rigid nonpolyurethane foam, refrigeration, air-conditioning,aerosols, sterilization, food-freezing 79000 solvent CzClsF3 (CFC-113) CiClzFi (CFC-114) C;Clk5(CFC-115)
rigid nonurethane foam, refrigeration. air-conditioning rehigeration, air-conditioning
4 500 10 400
substitute (HCFC or HFC) CHClzCFs (HCFC-123) CH3CC12F (HCFC-14lb) CHzFCF3 (HFC-134a) CHClzCFzCFs (HCFC-225ca) CHClPCF2CClFz (HCFC-225cb) CHClFCFn (HCFC-124) CHF2CF3 (HFC-125)
For sources, see ref 28.
useful in establishingexposure limits for HCFCs and HFCs in the workplace and general environment.
Experimental Procedures Instrumental Analyses. Fluorine NMR spectra were obtained with a Bruker WP-270 instrument equipped with a dedicated 5-mm l9F probe and operating at 254.18 MHz for fluorine. The pulse width was 3 hs, the interpulse time was 1.7 s (spectral width = 5 kHz), and 16K data points were collected. Spectra were acquired at room temperature with sample spinning. Chemical shifts were referenced to external trifluoroacetamide in DzO (6 0 ppm). Urine trifluoroacetic acid concentrations were measured by leFNMR (9) with signal integrals referenced to a 5.0 mM aqueous solution of chlorodifluoroacetic acid (6 = 12.7 ppm) contained in asealed coaxialtube (WilmadGlass Co., Buena, NJ). Mass spectra were recorded with a Hewlett-Packard 5880A gas chromatograph equipped with a HP-1 (dimethylpolysiloxane stationary phase; 25-m X 0.2-mm X 0.5-pm film thickness) capillary column and coupled to a HP-5970 mass-selective detector (70 eV, electron impact). Chemicals. Halothane (HalocarbonLaboratories,N. Augusta, SC), HCFC-123 (DuPont Co., Wilmington, DE), HCFC-124 (DuPont Co., Wilmington, DE), HFC-125 (DuPont Co., Wilmington, DE), and HFC-134a (Allied-Signal,Morristown, NJ) were pure by 19F NMR and GC-MS. [2H]HCFC-123 (>99% 2H incorporation as determined by GC-MS) was prepared in 91% yield from HCFC-123 by the procedure reported by the preparation of PHIhalothane (10) and was pure by GC-MS. Animal Treatment. Male, Fischer 344 rats (200-300 g; Charles River, Wilmington, MA) were exposed for 6 h by inhalation to 1.1 f 0.1% halothane (n = 6), to 0.010 f 0.001% HCFC-123 (n = 6), to 0.11 f 0.004% HCFC-123 (n 3), to 1.1 0.1 % HCFC-123 (n= 6), to 1.0 0.1 % HCFC-124 (n = 6), to 0.97 f 0.1% HFC-125 (n = 6), or to 1%(nominal) HFC-134a (n = 3) for 6 h. All gas-phase concentrations are expressed on a volume/volume (v/v) basis. The inhalation chamber used has been described ( I I ) , except that CaClz was used to absorb water and consumed oxygen was replaced via an attached, oxygenfilled spirometer. Chamber halocarbon concentrations were measured periodically during exposure by GC-MS. PHIHCFC123 was quantitatively converted to HCFC-123 by C02 absorbanta (e.g., Baralyme, Ascarite) used in the inhalation chamber; therefore, in some experiments, rats were given 15 mmol/kg HCFC-123 (n = 4) or [2H]HCFC-123(n = 4), dissolved in corn oil, by intraperitoneal injection. Immediately after the end of the inhalation exposure or immediately after the intraperitoneal injection of HCFC-123 or [2HlHCFC-123,the rats were placed in metabolism cages; urine was collected for 12 h and was stored frozen until analyzed. After 12 h, the livers were removed from ether-anesthetized rats and homogenized in 0.1 M phosphate buffer (pH 7.0). The supernatant obtained by centrifuging the liver homogenate at loooOg for 20 min was c.entrifugedfor 60 min at 100000g, yielding the cytosolic fraction as the supernatant and the microsomal fraction as the pellet. The microsomal fraction was resuspended in fresh buffer and centrifuged again at lOOOOOg for 60 min.
*
*
Protein Immunoblotting. Proteins were separated by SDSPAGE and were immunoblotted with anti-TFA-protein serum, as previously described (4). The amount of protein added to resolving gel lanes is indicated in the figure legends. Computational Methods. The enthalpies of activation of halothane, HCFC-123,HCFC-124,andHFC-125weredetermined by the method of Korzekwa et al. (12). All programs, nomenclature, and computationalmethods used are described in ref 12. This method predicts the enthalpy of activation from the groundstate energies of the reactant and product for a given reaction by the semiempirical method AM1 (13). The enthalpies of activation were calculated from the following regression line:
AHac, = 2.60 + 0.22(AHm) + 2.38(IP) where AHad,is the predicted enthalpy of activation, AHm is the AM1 enthalpy of reaction for hydrogen atom abstraction from a given halogenated hydrocarbon by p-nitrosophenoxy radical? and IP is the ionization energy of the halogenated hydrocarbon radical product.
Results Structure-Protein Trifluoroacetylation Experiments. Twelve hours after the end of 6-h inhalation exposuresto 1%halothane, HCFC-123, HCFC-124,HFC125, or HFC-l34a, subcellular fractions were prepared from the livers of exposed rats and examined by SDSPAGE and immunoblottingwith anti-TFA-protein serum. Similar patterns of TFA-proteins were detected in hepatic microsomal fractions (Figure lA, lanes 1and 2) and cytosolicfractions (Figure lB, lanes 1and 2) of rats exposed to halothane or HCFC-123,as previouslyreported (4).Immunoblots of microsomal fractions from rats exposed to HCFC-124 showed a pattern of TFA-proteins similar to that seen in rats exposed to halothane or HCFC-123, although the immunoreactivity of individuals bands was lower (Figure l A , lane 3). Only one, minimally reactive microsomal TFA-protein was detected in rats exposed to HFC-125 (Figure lA, lane 4). Two major cytosolic TFAproteins were detected in rata exposed to HCFC-124 and HFC-125 (Figure lB, lanes 3 and 4). The yield of these cytosolic TFA-proteins was similar in HCF-124- and HFC125-exposed rats, but was less than the intensity of the cytosolic TFA-proteins in halothane- or HCFC-123exposed rats (Figure lB, lanes 1and 2). The amount of protein was varied from lane to lane in the immunoblots presented in Figure 1 in order to ensure that the TFAproteins could be detected in spite of differing levels of immunostaining. Thus the difference between trifluoroacetylation of hepatic microsomal proteins resulting from HCFC-123 or HCFC-124 exposure is greater than the observed difference in immunochemical staining.
Harris et al.
722 Chem. Res. Toxicol., Vol. 5, No.5, 1992
B
A
B
A
200200-
i'
116-
97 -
116-
97 -
66-
66-
43 -
43
1
2
3
4
1
2
3
4
Figure 1. Immunochemical detection of TFA-proteins in hepatic microsomal (A) and cytosolic (B) fractions from rats 12 h after a 6-h inhalation exposure to 1?6 halothane (lane l),to 1% HCFC-123 (lane 2), to 1%HCFC-124 (lane 31, or to 1%HFC125 (lane 4). Markers a t the left indicate molecular mass in kilodaltons. Protein loading was 250 pg per lane (lanes A1 and A2), 600 pg per lane (lanes B1 and B2), and lo00 pg per lane (lanes A3, A4, B3, and B4).
-
1
2
3
4
2
3
4
B
A
As an additionalcontrolexperimentto test the selectivity of the serum used for immunoblotting, hepatic microsomal and cytosolicfractions from rats exposed to 1% HFC134a were analyzed. Although lo00 pg of protein from these fractions was loaded in each lane, immunoreactive bands were not detected (data not shown). HCFC-123: Exposure Concentration-Response Experiments. Rats were exposed to 0.01 % ,0.1 % ,or 1% HCFC-123 for 6 h. Identical patterns of TFA-proteins were detected in microsomal (Figure 2A) and cytosolic (Figure 2B) fractions in the different treatment groups. The intensity and patterns of immunostainingwere similar in rats exposed to 0.1 % or 1%HFCF-123 (lanes 3 and 4, respectively), but were lower in rats exposed to 0.01% HCFC-123 (lane 2). Protein fractions from animals exposed to 1%halothane served as positive controls (Figure 2A and 2B, lane 1). HCFC-123: Effect of Deuteration. Previous studies showedthat deuterated halothane was metabolized in vivo to a lesser extent than was halothane (10,14). The formation of microsomal (Figure3A) and cytosolic (Figure 3B) TFA-proteins was compared in rats given 15mmol/ kg body weight HCFC-123 or [2H]HCFC-123. Hepatic protein trifluoroacetylation was decreased in both cytosolic and microsomal fractions in rats given [2H]HCFC123 (lanes 2 and 4) compared with rats given HCFC-123 (lanes 1and 3). Control experiments were conducted to determine whether [2H]HCFC-123 was converted to HCFC-123 in vivo. [2H]HCFC-123,but not HCFC-123, was detected by single-ion monitoring GC-MS analysis of the exhaled air of rats at 10,30, and 60 min after giving 15 mmol/kg [2H]HCFC-123 intraperitoneally (data not shown). Urinary Trifluoroacetic Acid Excretion. Urinary trifluoroacetic acid excretion in rats exposed to halothane, HCFC-123, HCFC-124,and HFC-125was quantified from 0 to 12 h after the end of the inhalation exposure (Table
1
Figure 2. Immunochemical detection of "FA-proteins in hepatic microsomal (A) and cytosolic (B) fractions from rata 12 h after a 6-h inhalation exposure to 196 halothane (lane l),to 0.01 % HCFC-123 (lane 2), to 0.1 % HCFC-123 (lane 3), or to 1% HCFC123 (lane 4). Markers a t the left indicate molecular mass in kilodaltons. Protein loading was 150 pg of microsomal protein per lane and 600 pg of cytosolic protein per lane.
200-
-
.-. -
. -.
--
116-
**
97 66
b-w
-
43 -
1
2
3
4
Figure 3. Immunochemical detection of "FA-proteins in hepatic microsomal (A) and cytosolic (B) fractions from rata 12 h after giving 15mmol/kg HCFC-123(lanes 1and 3) or [2H]HCFC123 (lanes 2 and 4) by intraperitoneal injection. Markers a t the left indicate molecular mass in kilodaltons. Protein loading was 30 pg of microsomal protein per lane and 225 pg of cytosolic protein per lane.
11). Urinary trifluoroacetic acid excretion in HCFCexposed rats followed this order: HCFC-123 = halothane > HCFC-124 > HFC-125. Trifluoroacetic acid excretion was also quantified in rats exposed to 0.01 % ,0.01 % ,or 1.0 % HCFC-123(Table III). Trifluoroaceticacid excretion was similar in rats exposed to 0.1 % and 1.0% HCFC-123, but was much lower in rats exposed to 0.01 % HCFC-123. Urinary trifluoroacetic acid excretion was also compared in rats given [2H]HCFC-123or HCFC-123 by intraperitoneal injection. Trifluoroacetic acid excretion was decreased by 42% in rats given [2H]HCFC-123 [24.9 5.0 pmol(l2 h)-l (kg body weight)-', n = 41 compared with
*
Chem. Res. Toricol., Vol. 5, No. 5, 1992 723
CFC Substitutes: Hepatic Acylation Potentials Table 11. Urinary Trifluoroacetic Acid Excretion in Rats Exposed to Halothane, HCFC-123, HCFC-124, or HFC-125 trifluoroacetic acid excretionb halocarbona [pmol(12 h)-l (kg rat)-'] halothane 65.1 f 16.4 82.9 f 18.9 HCFC-123 HCFC-124 15.6 f 2.1 1.65 f 1.74 HFC-125
5 w
5
I
3 5-
30: 2 25-
Y
Rata were exposed to a 1.0% halocarbon atmosphere (v/v) for 6 h, and urine was collectedand analyzed as described in Experimental Procedures. b Values are reported as means f standard error; n = 3-6. a
Table 111. Urinary Trifluoroacetic Acid Excretion in Rats Exposed to a Range of HCFC-123 Concentrations exposure trifluoroacetic acid excretionb concentrationa [pmol(12 h)-1 (kg rat)-'] 1.0% HCFC-123 82.9 f 18.9 0.1% HCFC-123 87.2 f 21.0 0.01% HCFC-123 13.8 f 2.4
5.01
d
20-
2
15-
3 'c
10-
05-
coot
\ '
250
'
'
'
255
260
'
'
265
'
'
270
'
1
275
P r e d i c t e d Enthalpy o f Activation (kcal)
Figure 4. Comparison of the predicted enthalpy of activation of halothane, HCFC-123, HCFC-124, and HFC-125 withobserved urinary trifluoroacetic acid excretion in HCFC- and HFC-exposed rats. The trifluoroacetic acid excretion rates and the enthalpies of activation are from Tables I1 and IV, respectively.
0 Rats were exposed to the indicated HCFC-123 concentrations (viv) for 6 h, and urine was collected and analyzed as described in Experimental Procedures. Valuesare reported as means f standard error; n = 3-6.
Table IV. Predicted AM1 Thermodynamic Parameters for Hydrogen Atom Abstraction from Halothane, HCFC-123, HCFC-124, and HFC-125 halocarbon HCFC-123 halothane HCFC-124 HFC-125
hHf(sub)a
"d)
-175.42 -161.35 -214.05 -264.9
-150.41 -137.48 -188.55 -237.06
M, -8.98 -9.82 -8.19 -5.87
IPb 10.39 10.54 10.51 10.87
5
c
E
c
matt. 25.42 25.52 25.82 27.18
a Enthalpies are expressed in kilocalories per mole (kcal/mol) and use the nomenclature of Korzekwa et al. (12). Ionization potentials were calculated according to Koopman's theorem and are expressed in electronvolts (eV).
those given HCFC-123 i42.5 f 9.3 pmol(12 h)-l (kg body weight)-', n = 41. Enthalpies of Activation of Halothane, HCFC-123, HCFC-124, and HFC-125. A predictive model for hydrogen atom abstractions catalyzed by the cytochromes P-450 was used to calculate the enthalpies of activation of halothane, HCFC-123, HCFC-124, and HFC-125 (Table IV). The results indicate that the rate of hydrogen atom abstraction from halothane is comparable to that from HCFC-123 and is approximately 2-fold faster than that from HCFC-124 and approximately 20-fold faster than that from HFC-125. The predicted enthalpies of activation of HCFCs can be compared with trifluoroacetic acid excretion in HCFC-exposed rats: trifluoroacetic acid excretion in HCFC-123 and halothane-exposed rats was comparable and was 4- to 5-fold greater than in HCFC124-exposedrats and 40- and 50-foldgreater than in HFC125-exposed rats (Table 11). A plot of the predicted enthalpy of activation versus the natural logarithm of the observed trifluoroacetic acid excretion showed a strong correlation (r2 = 0.955) (Figure 4).
Discussion Immunochemical evidence for hepatic protein trifluoroacetylation was found in rats exposed to 1% halothane, HCFC-123, HCFC-124, or HFC-125. With HCFC-123, TFA-protein formation was maximal at exposure concentrations greater than 0.1%, indicating that the metabolic processes leading to TFA-protein formation were saturated at this exposure concentration. The relative
4
I F
F
5 -
+Nu:. -X-i,z
F
Y 0r
U
6
Figure 5. Proposed bioactivation mechanism for halothane (1; X1 = Br, Xz = Cl), HCFC-123 (1; X1 = Xz = Cl), HCFC-124 (1; X1 = C1, Xz = F), and HFC-125 (1; X1 = Xz = F). 2,l,l-dihalo2,2,2-trifluoroethyl radical; 3, l,l-dihalo-2,2,2-trifluoroethanol; 4, trifluoroacetyl halide; 5, trifluoroacetic acid; 6, TFA-protein (Nu: = Nf-amino group of lysine); P-450, cytochrome P-450.
magnitude of TFA-protein formation was a function of halocarbon structure. The rank order of TFA-protein formation estimated from the immunoblotswas halothane = HCFC-123 >> HCFC-124 > HFC-125. Urinary trifluoroacetic acid excretion in 2,2-dihalo-l,l,l-trifluoroethaneexposed rats followed the same order. The observed rank order of TFA-protein formation and trifluoroacetic acid excretion can be attributed to different rates of biotransformation of halothane, HCFCs, and HFCs and to differences in the pharmacokinetic behavior of the halocarbons. The reaction scheme shown in Figure 5 is based on previous studies on halothane and HCFC biotransformation (4,15). Halothane, HCFCs, and HFCs (Figure 5, 1) undergo a cytochrome P-450-catalyzed hydrogen atom abstraction to yield the intermediate l,l-dihalo-2,2,2-trifluoroethyl radical (Figure 5, 2); oxygen rebound would give the geminal halohydrin (Figure 5, 3). Loss of HX from geminal halohydrin 3 would give trifluoroacetyl chloride or fluoride (Figure 5,4), which may undergo hydrolysis to give trifluoroacetic acid (Figure 5, 5) or may react with nucleophilic sites in proteins to give TFAproteins (Figure 5, 6). The correlation between the theoretically predicted enthalpy of activation and the observed trifluoroacetic acid
724 Chem. Res. Toxicol., Vol. 5, No. 5, 1992
Harris et al.
excretion in HCFC- or HFC-exposed animals (Figure 4) indicates that the rate of radical formation mediated by cytochromes P-450 is rate limiting in the production of both trifluoroacetic acid and TFA-protein formation. This conclusion is supported not only by the results of the deuterium isotope studies but also by the predicted rates of each step in Figure 5. The first step in HCFC or HFC biotransformation is hydrogen atom abstraction to yield a carbon-based radical (Figure 5, 1 2). The hydrogen atom abstraction step is predicted to occur with an enthalpy of activation of over 25 kcal/mol (Table IV). This step is followed by a rapid recombination of the ironcoordinated hydroxyradical wtih the carbon-based radical (Figure 5, 2 3) (16). The third step is elimination of hydrogen halide from the geminal halohydrin to give the trifluoroacetyl halide (Figure 5, 3 4). No data on the rate of halide elimination from geminal halohydrins are apparently available in the literature. Chemical intuition indicates that elimination of halide from a geminal halohydrin would have a lower enthalpy of activation than abstraction of a hydrogen atom from a primary carbon atom (Figure 5, 1 2). If, however, the conservative assumption is made that the intrinsic barrier to reaction is the same as for hydrogen atom abstraction and that the reaction is exothermic, the actual enthalpy of activation for hydrogen halide loss from halohydrin 3 would be lower than that for the relatively symmetrical hydrogen atom abstraction on the basis of the predictions of the Hammond postulate (17). Although hydrogen atom abstraction is the rate-limiting step in halohydrin formation, the chemistry of the halohydrins may explain why halothane and HCFC-123 yield more TFA-protein adducts in microsomes than in cytosol whereas HCFC-124 and HFC-125 yield comparable levels of TFA-protein adducts in both subcellular fractions. HCFC-124 and HFC-125 are expected to yield geminal fluorohydrins, whereas halothane and HCFC-123 should form geminal chlorohydrins. The fluorohydrins would be expected to be more stable and more water soluble than the chlorohydrins and may, therefore, diffuse from the site of formation in the microsomes to react with cytosolic proteins. The less stable and more lipophilic chlorohydrins may tend to react with microsomal proteins. The final step in HCFC or HFC biotransformation is a branched pathway that can lead either to trifluoroacetic acid production (Figure 5, 4 5) or to covalent modification of proteins (Figure 5, 4 6). The rate of hydrolysis of the trifluoroacetyl halide might be expected to be the closest in rate to the hydrogen atom abstraction step. Although data on the rate of hydrolysis of trifluoroacetyl halides are apparently not available, hydrolysis of phenyl chloroformate in water occurs with a enthalpy of activation of 14 kcal/mol and an entropy of activation of 5 eu (18). Furthermore, substitution of CC13 for CH3 in acetyl chloride increases the rate of hydrolysis by lo4 (19, 20). These data lead to a predicted energy of activation of less than 14 kcal/mol for the hydrolysis of trifluoroacetyl chloride. The enthalpy of activation for hydrogen atom abstraction may be overestimated by the semiempirical calculations,but it is likely that it is 14kcal/mol or greater.3 Thus, the theoretical and experimental data support the
hypothesis that the cytochrome P-450-catalyzed formation of halohydrins from halothane, HCFC-123, HCFC-124, and HFC-125 is the rate-limiting step in the production of the trifluoroacetic acid and TFA-protein adducts from parent compound. HCFCs and HFCs may yield trifluoroacetyl chlorides or fluorides (Figure 5,4),which may undergo nucleophilic displacement reactions, including hydrolysis and acylation of amines (19, 20). The ratio of hydrolysis to acylation as well as the ratio of acylation of microsomal to cytosolic proteins may be influenced by the halogen substituent in the acyl halide. Although trifluoroacetyl fluoride formed as a metabolite of HCFC-124 and HFC125 would be more stable than trifluoroacetyl chloride (201, the greater formation of TFA-proteins in cytosolic and microsomal fractions of halothane- and HCFC-123exposed rats compared with HCFC-124- and HFC-125exposed rats may reflect the higher rate of trifluoroacetyl chloride formation rather than the lower reactivity of trifluoroacetyl fluoride formed from HCFC-124 and HFC125. Deuteration of HCFC-123 decreased TFA-protein formation and trifluoroacetic acid excretion in vivo. Previous work showed that substrate deuteration decreased halothane-induced TFA-protein formation (7)and halothane metabolism in vivo (10,141. The effect of deuteration of HCFC-123 on in vivo metabolism (a decrease of ca. 40%) is less than, but consistent with, the effect of deuteration on in vivo halothane metabolism (a decrease of ca. 5070%) (10). No evidence for the in vivo conversion of L2H]HCFC-123 to HCFC-123, which would decrease the observed effect of deuteration on metabolism, was found. Numerous kinetic studies on deuterium isotope effects have shown that the hydrogen atom abstraction step is partially rate limiting for most reactions catalyzed by cytochromes P-450 (21,221. This raises the question of how a theoretical method for hydrogen atom abstraction can predict the relative rates of reaction for cytochrome P450-catalyzedreactions. The answer is most likely because the active oxygen species in cytochromes P-450 cannot only abstract a hydrogen atom from the substrate but can also be further reduced by two electrons to water. This branched pathway can unmask the effect of substituents on the rate of product formation (21). As the substituent becomes more difficult to oxidize, more of the active oxygen species is reduced to water and less reacts with the hydrocarbon ~ u b s t r a t e .It~ is interesting to note that the observed isotope effects for halothane are larger than those for HCFC-123, which is consistent with a larger amount of water being formed from the poorer substrate halothane. The increased water formation would be expected to increase the magnitude of the observed isotope effect for the slower reacting substrate (22). Cytochrome P-450 2E1 catalyzes the hydroxylation of HCFC-123,5 HCFC-124 (151,HFC-134a (23),and 1,2dichloro-1,l-difluoroethane(HCFC-132b) (9). Thus cytochrome P-450 2E1, which shows selectivity for lowmolecular-weight compounds (24), may be of primary importance in the oxidative metabolism of HCFCs and HFCs in animals and man. In support of this contention, treatment of rats with isoniazid, an inducer of cytochrome
Although it is difficult to estimate accurately the effect of entropy on the activation barrier for hydrogen atom abstraction, it is most likely negative. If it is negative, this would increase further the free energy of activation for hydrogen atom abstraction.
if this is the case (J. P. Jones, unpublished work). J. W. Harris, M. J. Olson, and M. W. Anders, unpublished obser-
-
-
-
-
--
4
Studies are underway on the stoichiometry of the reaction to confirm
vations.
CFC Substitutes: Hepatic Acylation Potentials P-450 2E1, increases TFA-protein formation in rats given halothane (25). The data presented on the relative trifluoroacetylation of hepatic proteins by the CFC substitutes HCFC-123, HCFC-124, and HFC-125 demonstrate that increased fluorine substitution on dihalomethyl groups (CHXd decreases the metabolism of these compounds in vivo. A similar relationship exists among the inhalation anesthetics halothane, enflurane, and isoflurane, where approximately 20%, 2%, and 0.2%,respectively, of an anesthetic dose undergoes oxidative metabolism in humans (26)and where a strong correlation between rate of metabolism and antiTFA immunoreactive protein adduct formation exists for these anesthetics (26). Furthermore, the incidence of anesthetic-induced hepatic toxicity seen with halothane, enflurane, or isoflurane parallels the extent to which they are metabolized (27). Laboratory and clinical evidence for cross-sensitization between the various fluorocarbon inhalation anesthetics has been reported (26,27). On the basis of studies of these anesthetics, a correlation may be presumed to exist between the relative hepatic trifluoroacetylation potentials of halothane, HCFC-123, HCFC124, and HFC-125. Thus, a propensity for immunemediated hepatic toxicity due to repeated exposure to HCFCs or HFCs or to fluorocarbon-based anesthetic agents may exist. Thus HCFC-124 and HFC-125 may pose less risk than HCFC-123 because TFA-protein formation by the former compounds is lower. Because the minimum level of TFA-protein formation required to induce an immune response in sensitized individuals is not known and because no animal model for halothane hepatitis exists, the assessment of risk of immune-mediated hepatitis for workers exposed to pentahaloethane-based HCFCs or HFCs is not currenty possible.
Acknowledgment. This work was supported in part by National Institute of Environmental Health Sciences Grants ES05407 (M.W.A. and J.P.J.) and ES07026 (J.W.H.). We thank Ms. Sandra E. Morgan for preparing the manuscript. References (1) Molina, M. J., and Rowland, F. S. (1974) Stratospheric sink for chlorofluoromethanes: Chlorine atom-catalyzed destruction of ozone. Nature 249,810-812. (2) Anders, M. W. (1991) Metabolism and toxicity of hydrochlorofluorocarbons: current knowledge and needs for the future. Enuiron. Health Perspect. 96, 185-191. (3) Manzer, L. E. (1990) The CFC-ozone issue: Progress on the development of alternatives to CFCs. Science 249, 31-35. (4) Harris, J. W., Pohl, L. R., Martin, J. L., and Anders, M. W. (1991) Tissue acylation by the chlorofluorocarbon substitute 2,2-dichlorol,l,l-trifluoroethane. Proc. Natl. Acad. Sci. U.S.A.88,1407-1410. (5) Vergani, D., Mieli-Vergani, G., Alberti, A,, Neuberger, J., Eddleston, A. L. W. F., Davis, M., and Williams, R. (1980) Antibodies to the surface of halothane-altered rabbit hepatocytes in patients with severe halothane-associated hepatitis. N . Engl. J . Med. 303,66-71. ( 6 ) Kenna, J. G., Neuberger, J., and Williams, R. (1987) Identification by immunoblotting of three halothane-induced liver microsomal polypeptide antigens recognized by antibodies in sera from patients with halothane-associated hepatitis. J.Pharmacol. Exp. Ther.242, 733-740.
Chem. Res. Toxicol., Vol. 5, No. 5, 1992 725 (7) Kenna, J., Satoh, H., Christ, D., and Pohl, L. (1988) Metabolic basis for a drug hypersensitivity: Antibodies in sera from patients with halothane hepatitis recognize liver neoantigens that contain the trifluoroacetyl group drived from halothane. J. Pharmacol. Exp. Ther. 245, 1103-1109. (8) Pohl, L. R., Satoh, H., Christ, D. D., and Kenna, J. G. (1988) The immunologic and metabolic basis of drug hypersensitivities. Annu. Reu. Pharmacol. Toxicol. 28, 367-387. (9) Harris, J. W., and Anders, M. W. (1991) Metabolism of the hydrochlorofluorocarbon l,2-dichloro-l,l-difluoroethane. Chem. Res. Toxicol. 4, 180-186. (10) Sipes, I. G., Gandolfi, A. J., Pohl, L. R., Krishna, G., and Brown, B. R. (1980) Comparison of the biotransformation and hepatotoxicity of halothane and deuterated halothane. J. Pharmacol. Exp. Ther. 214, 716-720. (11) Shen, E. S.,Garry, V. F., and Anders, M. W. (1982) Effect of hypoxia on carbon tetrachloride hepatotoxicity. Biochem. Pharmacol. 31, 3787-3793. (12) Korzekwa, K. R., Jones, J. P., and Gillette, J. R. (1990) Theoretical studies on cytochrome P-450 mediated hydroxylation: a predictive model for hydrogen atom abstractiom. J.Am. Chem. SOC. 112,70427046. (13) Dewar, M. J. S., Zoebisch, E. G., Healy, E. F., and Stewart, J. J. P. (1985)AM1: anew general purpose quantum mechanical molecular model. J.Am. Chem. SOC. 107, 3902-3909. (14) McCarty, L. P., Malek, R. S., and Larsen, E. R. (1979) The effects of deuteration on the metabolism of halogenated anesthetics in the rat. Anesthesiology 51, 106-110. (15) Olson, M. J., Johnson, J. T.,OGara, J. F., and Surbrook, S. E. (1991) Metabolism in vivo and in uitro of the refrigerant substitute 1,1,1,2tetrafluoro-2-chlorethane.Drug. Metab. Dispos. 19, 1004-1011. (16) Shapiro, S., Piper, J. U., and Caspi, E. (1982) Steric course of hydroxylation at primary carbon atoms. Biosynthesis of 1-octanol from (1R)- and (1S)-[1-3H,2H,1H;1-14Cloctane by rat liver microsomes. J. Am. Chem. SOC. 104, 2301-2305. (17) Jones, J. P., Rettie, A. E., and Trager, W. F. (199) Intrinsic isotope effects suggest that the reaction coordinate symmetry for cytochrome P-450 catalyzed hydroxylation of octane is isozyme independent. J. Med. Chem. 33, 1242-1246. (18) Queen, A. (1967) Kinetics of the hydrolysis of acyl chlorides in pure water. Can. J. Chem. 45, 1619-1629. (19) Ugi, I., and Beck, F. (1961) Solvolyse von Carbonsaurederivaten. I. Reaktion von Carbonsaurehalogeniden mit Wasser und Aminen. Chem. Ber. 94, 1839-1850. (20) Talbott, R. J. E. (1972) The hydrolysis of carboxylicacidderivatives. In Comprehemiue ChemicalKinetics (Bamford, C . H., and Tipper, C. F. H., Eds) Vol. 10, pp 209-293, Elsevier, Amsterdam. (21) Atkins, W., H., and Sligar, S. G. (1987) Metabolic switching in cytochrome P-450,: deuterium isotope effects on the regiospecificity and the monooxygenaseioxidase ratio. J. Am. Chem. SOC. 109, 3754-3760. (22) Jones, J. P., Korzekwa, K. R., Rettie, A. E., and Trager, W. F. (1986) Isotopically sensitive branching and its effect on the observed intramolecular isotope effects in cytochrome P-450 catalyzed reactions:a new method for the estimation of intrinsic isotope effects. J . Am. Chem. SOC.108, 7074-7078. (23) Olson, M. J., Kim, S. G., Reidy, C. A., Johnson, J. T., and Novak, R. F. (1991) Oxidation of 1,1,1,2-tetrafluoroethane in rat liver microsomes is catalyzed primarily by cytochrome P-450IIE1. Drug. Metab. Dispos. 19, 298-303. (24) Guengerich,F.P.,Kim,D.H.,andIwasaki,M.(1991)Roleofhuman cytochrome P-450IIE1 in the oxidation of many low molecular weight cancer suspects. Chem. Res. Toxicol. 4, 168-179. (25) Kenna, J. G., Martin, J. L., Satoh, H., and Pohl, L. R. (1990) Factors affecting the expression of trifluoroacetylated liver microsomal protein neoantigens in rats treated with halothane. Drug. Metab. Dispos. 18, 788-793. (26) Christ, D. D., Satoh,H., Kenna, J. G., andPoh1,L. R. (1988)Potential metabolic basis for enflurane hepatitis and the apparent crosssensitization between enflurane and halothane. Drug. Metab. D ~ s ~ o16, s . 135-140. (27) Christ, D. D., Kenna, J. G., Kammerer, W., Satoh, H., and Pohl, L. R. (1988) Enflurane metabolism produces covalently bound liver adducts recognized by antibodies from patients with halothane hepatitis. Anesthesiology 69, 833-838. (28) (a) Dagani, R. (1990) Chem. Eng. News 68 (36, September 3), 8. (b) Zurer, P. S. (1989) Chem. Eng.News 67 (30, July 24), 7-13. (c) Fed. Regist. (1988) 53, 30599.
