Document NEnr2BX3rE980EajZYJR2VxNg

TOXICOLOGY AND APPLIED pharmacology 130, 237-247 (1995) Chloroethylene Mixtures: Pharmacokinetic Modeling and in Vitro Metabolism of Vinyi Chloride, Trichloroethylene, and trans-1,2-Dichloroethylene in Rat1 Hugh A. Barton,*-2 Jason R. Creech,t C. Steven Godin,*-3 Gia M. Randall,* and Constance S. Seckel* *Toxicology Hazards Research Unit, Wrtght-Patterson Air Force Base. ManTech Environmental Technology Inc,. P O Box 31009. Dayton. Ohio 43437-0009: and f Toxicology Division. Armstrong Laboratory. Wnght-Patierson Air Force Base. Ohio 45433-6503 Received March 22. 1994; accepted September 27. 1994 Chloroethylene Mixtures: Pharmacokinetic Modeling and in Vitro Metabolism of Vinyl Chloride, Trichloroethylene, and trans- 1,2-Dichloroethylene in Rat. Barton, H. A,, Creech, J. R,, Godin, C, S,, Randall, G. M,, and Seckel, C, S. (1995). Toxicol. Appl. Pharmacol. 130, 237-247. Environmental and occupational exposures are typically to mixtures of chemicals, although most toxicity information is for individual compounds. Interactions between chemicals may involve pharmacokinetic and/or pharmacodynamic effects result ing in modulation of toxicity. Therefore, physiologically based pharmacokinetic modeling has been used to analyze data de scribing the metabolism of vinyl chloride (VC) and trichloroeth ylene (TCE) mixtures in rats. A single saturable pathway was modeled, representing cytochrome P450 2E1. This was par tially validated using preexposure to trans- 1.2-dichloroethylene (/DCE) which virtually eliminated in vivo metabolism of both VC and TCE at low concentrations. Microsomes from /DCEexposed animals showed inhibition of metabolism of P450 2E1 substrates (chlorzoxazone, p-nitrophenol. and TCE) and no ef fect on 7-ethoxycoumarin deethylation. Studies with liver mi crosomes from VC-exposed animals found that neither suicide inhibition nor induction occurred during 6-hr exposures to high concentrations. Therefore, these effects were not modeled. Mod eling of mixtures of VC and TCE was successful only using competitive inhibition, as might be predicted for cytochrome P450 2E1 substrates, and not uncompetitive or noncompetitive inhibition. These results were further confirmed by determining the depletion of glutathione due to VC metabolism. The valida tion of a detailed model for the inhibition kinetics of metabo lism of these two compounds permits better understanding of the implications of coexposures for toxicity. It is notable that competitive inhibition only becomes significant at relatively high concentrations (tens of ppm), while at typical low environ mental concentrations (ppb), absorption is perfusion limited and enzyme is in excess so that the chemicals will be metabo lized independently. ims Acnkmic pTM. inc. Interactions between chemicals affecting their toxicity may result from alterations in pharmacokinetics (absorp tion, distribution, metabolism, or elimination) and/or phar macodynamics (quantitative descriptions of the mecha nisms of toxicity). Pharmacokinetic modeling is currently more completely developed than modeling of toxicity pro cesses. Therefore, pharmacokineticallv important interac tions are more readily characterized quantitatively. Complex mixtures of organics and metals are typically present in contaminated surface and ground water (Germolec el ai. 1989). These mixtures arise from releases of the individual chemicals, environmental weathering including biodegradation, and formation ofby-products, for instance, during chlorination. All three of these processes create mix tures of chloroethylenes. Chloroethylenes are particularly common contaminants due to their widespread usage as industrial chemicals. Trichloroethylene (TCE)4 and tetrachloroethylene are known to undergo anaerobic dech lorination in field and laboratory studies (Freedman and Gossett, 1989). This process results in the formation of TCE, dichloroethylene isomers, vinyl chloride (VC), and ethylene, although the extent of formation ofeach species is site-specific and variable. Most animal studies of toxicity utilize individual pure chemicals while human exposures and consequently epide miological data mostly involve mixtures. The single-chcmical approach results from a scientific need to identify which chemicals in a mixture are of greatest concern. The prob lem of "reconstituting" mixtures is complicated by their 1 Presented as a poster at the 32nd Annual Meeting of the Society of Toxicology held in New Orleans, LA, March 14-18, 1993. 3 To whom correspondence should be addressed. 1 Present address: Cephalon. 145 Brandywine Parkway. West Chester. PA 19380. 4 Abbreviations used: VC, vinyl chloride; TCE, trichloroethylene; (DCE, Irani-1.2-dichloroethylene; PBPK, physiologically based pharmacoki netic: NPSH. nonprotein sulfhydryl; normalized to I kg body weight in PBPK modeling: P450. liver microsomal cytochrome P450. r 004I-008X/95 $6.00 Copyright <* 1995 by Academic Press, Inc. All rights of reproduction in any form reserved. CMA 022270 238 barton et al. infinite variation and the potential for complex interactions ranging from synergism and potentiation to antagonism (U.S. EPA. 1986). The difficulties, particularly in the area of pharmacodynamics, are also increased greatly when the concern is chronic toxicity, such as cancer. VC is a known human carcinogen which causes cancer, notably hepatoangiosarcomas, in every animal species tested (ATSDR, 1989). VC is a classic mutagenic chemical due to the formation of an electrophilic epoxide metabolite via metabolism by cytochrome P450 (P450), particularly P450 2E1 (Guengerich et al., 1991; Gwinner et al.. 1983: Guengerich and Strickland. 1977). Chloroethylene oxide can adduct DNA forming 7-(2-oxoethyl)guanine and three minor cyclic ethenoadducts (Fedtke et al., 1990). Chloroethylene oxide rearranges spontaneously to form chloroacetaldehyde (Fedtke et al.. 1990). Chloroacetaldehvde is responsible for the majority of adduction of pro teins. In vitro and in vivo studies have found P450 itself to be a target resulting in suicide inhibition (Ivanetich et al., 1977; Pessayre et at., 1979: Reynolds et al., 1975; Watanabe et al, 1976). GSH depletion has been demonstrated in vivo and linked to chloroacetaldehyde formation (Plugge and Safe, 1977). The extent of depletion is dose dependent. Metabolism of the GSH conjugates results in several stable metabolites identified in rat urine. TCE, in contrast, causes cancer in two animal species but the results are highly species dependent (Bruckner et al.. 1989) . Oral exposure resulted in hepatocellular carcinomas in B6C3F1 mice and some kidney tumors in male rats. Lung tumors were also seen in mice exposed by inhalation. P450 2E1 is the major enzyme responsible for the initial metabolism of TCE. although other P450s can also be in volved (Miller and Guengerich, 1983; Nakajima et al.. 1990) . Direct GSH conjugation, mediated by transferases, also occurs at very low rates (Bruckner et al., 1989). The extent of epoxide formation appears limited, apparently due to rearrangement occurring in the enzyme-active site (Miller and Guengerich. 1983). Neither significant destruc tion of P450 nor GSH depletion have been reported for TCE. These results contrast strongly with those for VC but are similar to findings with tetrachloroethylene. The major products of TCE metabolism include the un stable product chloral hydrate and the chemically stable forms trichloroethanol and trichloroacetate. Dichloroacetate is a minor product. Oral exposure to chloral hydrate, trichloroacetate. or dichloroacetate causes B6C3F1 mouse liver tumors (Herren-Freund et al., 1987; Bull et al.. 1990; DeAngelo et al., 1991) demonstrating that formation of a genotoxic epoxide intermediate from TCE is not obligatory to its induction of cancer. The studies reported here utilize in vitro measurements of enzyme activities and cofactor levels to validate the structure of a physiologically based pharmacokinetic (PBPK) model. A model previously developed for mixtures of TCE and 1,1-dichloroethylene (vinyiidene chloride) was used with TCE and VC (Andersen et al.. 1987). Competi tive. noncompetitive, and uncompetitive inhibition of oxi dative P450 metabolism were all considered as possible in teractions between these chemicals. Differences in VC-induced GSH depletion assisted in better understanding of the results of coexposures to TCE and VC. Finally, the im plications of the results for risk assessment of mixtures were considered. EXPERIMENTAL PROCEDURES Animals. Male Sprague-Dawley rais were purchased from Charles River Breeding Laboratories (Raleigh. NC). Animals were quarantined for 2 weeks for verification of health status by pathological, fecal parisitological. and serum chemistry examinations. Water and food (Purina Formulab 5008) were provided ad libitum. On the day of experiments, body weights ranged from 250 to 400 g. A few experiments were repeated using animals outside this weight range. The animals used in this study were handled in accordance with the principles stated in the "Guide for the Care and Use of Laboratory Animals" prepared by the Committee on Care and Use of Laboratory Animal Resources (National Research Council. Department of Health and Human Services, National Institute of Health. Publication 86-23. 1985) and the Animal Welfare Act of 1966. as amended. Chemicals. Chemicals were obtained from manufacturers with the specified purity: VC (Fluka. New York. NY. 99.5%); /-1,2-dichloroethylene (tDCE) (Aldrich. Milwaukee. Wl. 98%) and TCE (Aldrich, 99.9%). Purities were confirmed by GC-MS. 6-Hvdroxychlorzoxazone was a gener ous gift ofDr. John Lipscomb (Toxicology Div.. WPAFB). All other chemi cals were purchased from standard sources. Vinyl chloride is a known human carcinogen and appropriate precautions should be taken to limit exposure of personnel to this gas. Trichloroethylene is a neurotoxin at high concentrations so exposure to the liquid or its vapor needs to be limited. Gas uptake exposures. Three male Sprague-Dawley rats were exposed to VC. TCE, /DCE. ora mixture using a closed recirculating system similar to that described by (Gargas et al., 1986). This method starts with a fixed amount ofchemical in the atmosphere. The concentration decreases as the chemical is absorbed and metabolized. Chamber concentrations were monitored using a gas-sampling valve connected to a gas chromatograph. Chromatography was performed on a 12 ft x J in, stainless steel column packed with 10% SE30 on 80/100 mesh Chromosorb WHP. The GC was equipped with a hydrogen flame ionization detector with the temperature set at 250C, injection temperature at 175C. constant oven temperature ofI20*C. and nitrogen carrier at 21 cc/min. Maximum exposure time was 6 hr. Partition coefficient determinations. A modification ofthe vial equili bration method was used to determine partition coefficients (Gearhart et al.. 1993). Animals were sacrificed by C02 asphyxiation. Blood was ob tained from the posterior vena cava. Liver, fat. muscle (slow), and kidney (rapid) were collected for analysis. Tissues (except blood) were homoge nized without diluent. Blood (I ml) was added or tissues (I g) were smeared on the side of a 21,9-ml vial. All samples were incubated for 3 hr with VC (114 ppm) or TCE (600 ppm) before determining headspace concentra tion. Chromatographic conditions were the same as those for gas uptake. Nonprotein sulfliydryl. Nonprotein sulfhydryl (NPSH) was measured as an estimate of reduced GSH levels on'livers following selected exposures (Ellman. 1959; Baker el al.. 1990). Animals were euthanized using C02. Livers were not perfused for NPSH analysis, but as much blood as possible was drained before mincing. Minced tissue was homogenized in 5 vol 5% sulfosalicylic acid. Supernatant (100 rxl) from a I500g spin (10 min) was CMA 022271 inhibition kinetics FOR VC/TCE mixtures 239 added to phosphatc/EDTA (80/0.8 mM final) and 5.5'-dithiobis-2-nitrobenzoicacid (0.15 mM final). Absorption at 412 nm was compared toGSH standard and quantitated as moles GSH per milligram wet weight liver. The NPSH data tnmol GSH/mg protein) were analyzed using a one-factorial analysis of variance with Tukey multiple comparisons (Rosner. 1990). The factor was exposure levels. Induction ofPiSOs and preparation of microtomes. Induction of rats was carried out by intraperitonea) injection of pyridine (200 mg/kg) for 4 days or by addition of 10% ethanol or 0.1% phenobarbital to drinking water for 7 days. Livers from control and exposed rats were removed and perfused with cold 0.15 M Tris-KCI(pH 7.4) with added heparin (2 U/ml). The liver was minced with 4 vol of Tns-KCl and homogenized with a Tri-R-Stir*R tissue homogenizer Model K4I (Tri R Instruments, Rock ville Centre. NY), The homogenate was centrifuged at SOOg for 10 min at 4C and the supernatant was recentrifuged at 9000g for 10 min. The result ing supernatant was further centrifuged at lOS.OOOy for 60 min. Micro somal pellets were suspended in 0.15 M Tris-KCl (pH 7.4) and then stored at -80C until use. Assays for enzymatic activity. In vitro metabolism was measured by the formation of 6-hydroxychlorzoxazone using a modification of the pro cedure of Peter et at. (1990). Samples were injected onto a Cl 8 reversedphase column (4.6 x 150 mm) using a 25% acetonitrile/75% phosphoric acid (0.5%) mobile phase and a 15-min run time, 7-Ethoxycoumarin Odeethylation was measured using the procedure of Greenlee et at. (1978) modified to use a plate reader (CytoFluor 2300. Millipore Inc.. Bedford. MA) with a fixed emission filter wavelength of 460 nm and excitation of 380 nm. TCE metabolism in rat microsomes was assessed by the disappearance of TCE from the headspace using a modification of the method of Nakajima et at. (1990). The protein content was 1-2 mg in 1 ml of 100 mM potassium phosphate (pH 7.4). Incubation time was 40 min at 37Caftera 20-min preincubation time with TCE and no regenerating system. The regenerating system contained 0.66 mM NADP. 13.7 mM glucose 6-phos phate. 2.8 U/ml glucose-6-phosphate dehydrogenase, and 100 mM potas sium phosphate (pH 7.4). Experiments showed that the reaction was linear over this lime period and the NADPH regeneration was optimal. The TCE was added as a vapor. Headspace disappearance was measured by gas chro matographic analysis on a Hewlett-Packard 5880 or 5890 equipped w ith a flame ionization detector. A 10% SE 30 80/100 Chromasorb WHP 12 ft x j in. stainless steel column was used under the following conditions: oven temperature, 125C: detector temperature. 250C: injection temperature. 175C: carrier flow. 21 cc/min nitrogen. The amount of TCE metabolized was determined using the calculations of Sato and Nakajima (1979). Attempts to measure microsomal metabolism of VC in vitro by moni toring the disappearance ofparent from the vial headspace were unsuccess ful, whereas good results were obtained with TCE. This made it impossible to cany out experiments with VC/TCE mixtures to parallel the in vivo experiments. PBPK modeling. A modified version of a model for competitive, non competitive. and uncompetitive inhibition of enzyme metabolism b> two substrates was used (Andersen et at.. 1987). The model was run using SIMUSOIV (Dow Chemical Co. Midland. Ml). This model consists of four physiological compartments: liver, fat. rapidly perfused (e.g.. kidney), and slowly perfused (e.g.. muscle) tissues. Metabolism was modeled as occurring entirely in the liver. Parameter values other than chemical specific ones were: QK (alveolar ventilation. L hr'1 kg) = 14, Q^ (cardiac output. L hr"1 kg) - 14; tissue blood flows as percentage of cardiac output. Qic (liver) = 0.25. (fat) = 0.09, ftcfslow) = 0.15. Orc (rapid) - 0.51: tissue volumes as a fraction of body weight. I LC (liver) = 0.04, l'FC (fat) - 0.07, r^. (slow) = 0.75, IRC(rapid) = 0.05 (U.S. EPA. 1988: Gargas et at.. 1986). (?ec = (?cc " 14 has been widely used in PBPK models for several strains of rats (Andersen et at.. 1987 and U.S. EPA. 1988). Measurements of cardiac output in Sprague-Dawley rats have recently been published from which Qcc " 16 was calculated (Delp et at.. 1991). This small difference in estimates for (3,-^did not affect the modeling significantly. In this model, the Michaehs-Menten equation for metabolism of each compound was adapted as follows: L (I MAXI X 0'vli)/(^MI * Tx + CVL, X T2) 2. 7"| = 1 + CvuM-Mirt + (Gvl-I'/I^miji x AM22) 3. rs = 1 + CV.JKW + CVL:/A-,:: I maxi and AMi represent the apparent maximal rate and concentration producing half-maximal activity for chemical I. CVL, and Cvu are the venous concentrations of chemicals I and 2 in blood leaving the liver, the only metabolizing organ in the model. AM,12 is the apparent constant for inhibition of metabolism of chemical 1 by chemical 2. Similarly, A'M|2,. A'm2i- A',,ii, and A'M22 are apparent inhibition constants where the first number indicates the chemical being metabolized and the second number is the chemical acting as inhibitor. When the appropriate inhibitory constants are set to a large number (106). these equations reduce to the standard equations for the three types of inhibition. Values for competitive inhibition were A"M|,2 = A'MI, A'M|;| = AM2, and Am,2 = AM2, = A'Mn = A`M22 = 1.000,000. Values for noncompe titive inhibition were A'M112 = 0.65, A'Mi2i = 2.2. A'M,2 = 0.65. A'M2i = 2.2. and A"mu = AM22 = 1.000.000. Values for uncompetitive inhibition were A'mi2 ~ 3.0. Am2i = 3.0, and XMII2 - AMt21 = A'mm - Am22 c 1,000.000. RESULTS Estimating metabolic parametersfor vinyl chloride with a single saturable oxidative pathway. Modeling VC phar macokinetics requires chemical specific parameters de scribing its partitioning into tissues and its metabolism. At a minimum, a single saturable pathway representing P450 oxidation is required. More complex metabolic models could be required if it is necessary to describe modulation of P450 levels, multiple isoforms, or multiple metabolites. The initial hypothesis used was to model a single saturable pathway presumed to represent formation of chloroethy- lene oxide by P450. This would represent the obligatory step prior to rearrangement to chloroacetaldehyde. conjuga tion to GSH. or macromolecular binding. In vitro studies to support this choice of a single saturable pathway were then undertaken. VC partition coefficients were measured in vitro for blood, liver, fat, muscle, and kidney (Table 1). This PBPK model consists of four compartments requiring partition coefficients: liver, fat, slowly perfused (muscle), and rapidly perfused (kidney). These partitions are similar to previous measurements, although the blood/air is somewhat higher resulting in the tissue/blood partitions being generally lower (Clement International, 1990). Kinetic data for metabolism of VC were collected by measuring the disappearance of parent compound from a recirculating closed chamber system. Initial concentrations in the gas uptake studies ranged from 95 to 10,000 ppm (Fig. 1). Optimization of individual runs (n = 18) gave ap parent kinetic constants of V__ = 3.6 2.3 mg kg-1 hr-1 (58 //mol kg-1 hr-1) and Km = 0.08 0.09 mg/liter (1 pM). Optimization of all runs estimated = 3.0 mg kg-1 hr-1 CMA 022272 240 BARTON ET AL. table i Partition Coefficients for Vinyl Chloride and Trichloroethylene Vinyl Chloride Blood/air (FBI) Fal/blood (PF1) Muscle/blood (PSI) Liver/blood (PL 1) Kidney/blood (PRI) 2.4 0.5 10.0 3.0 0.4 0.2 0.7 0.3 0.7 0.4 n =1 n -5 n =4 n =5 n =6 Trichloroethylene Blood/air (PB2) Fat/blood (PF2) Muscle/blood (PS2) Liver/blood (PL2) Kidney/blood (PR2) 20.5+1.4 26 5 0.6 0.1 1.3 0.1 1.0 0.2 n=4 n =3 n =8 n =b n =5* 6 while Km was estimated at 0.01, which was the lower bound present for optimization of this parameter. Liver nonprotein sulfhydryl after VC exposure. Deple tion of NPSH, largely glutathione, occurs during exposure to VC (Bruckner et ai. 1989). Therefore, this biological effect of VC metabolism was used to validate some aspects of the model and demonstrate, by correlation, that altered VC metabolism was occurring as predicted by the model (Tables 2-4). Because these exposures are not done at con stant concentrations, the results are somewhat different (particularly at low concentrations) than those reported else where (Bruckner et al.. 1989) but the extent of depletion at high concentrations is very consistent. NPSH depletion increased with dose and time (6 hr ver sus 3 hr) resulting in a maximum of about 44% depletion during a 6-hr exposure starting at 5000 ppm VC (Tables 2 and 3). NPSH depletion was actually greater at 3 hr than at 6 hr, 18% versus 5%, with a starting concentration of ap proximately 600 ppm. This result likely is explained by the decrease in the total rate of VC metabolism that occurs as the exposure concentration decreased. Glutathione is com plexly regulated by the body in an attempt to maintain nor mal levels so resynthesis is expected to occur (D'Souza et a!.. 1988). Modulation of P450 levels potentially requiring model ing. Alterations in enzyme levels may occur through sev eral mechanisms including suicide inhibition, altered gene expression (induction or repression), and a variety of posttranscriptional alterations such as reduced turnover due to stabilization of protein. All three of these effects are known to occur for P450 2E1 (Yang et al., 1990) and other P450 isoforms. Enzymatic activities using several substrates were mea sured in liver microsomes from control and VC-exposed animals. The animals were exposed to a starting concentra tion of 10,000 ppm (1%) or 5000 ppm VC for a 6-hr period during which time chamber atmosphere was monitored. At these high (saturating) concentrations, the exposures more closely approximate constant concentration exposures. Chlorzoxazone (25 mm) hydroxylation was measured as an indicator of P450 2E1 activity (Table 5). No difference in activity was seen between animals treated with high VC concentrations and untreated controls. Microsomes from pyridine and ethanol-treated animals were used as positive controls and showed about a five-fold increase in activity as expected. Comparison of control and 10,000 ppm VCtreated animals at high chlorzoxazone concentrations (1.25 mM) also found no difference (data not shown). TCE metabolism was also measured in vitro (Table 6). Again, microsomes from animals exposed to high VC con centrations for 6 hr had identical activity compared to con trols. Good induction of activity was seen with the pyri dine-treated animals. Deethylation of 7-ethoxycoumarin was also apparently unaffected by 6-hr exposure to a high concentration of VC (Table 7). By contrast, a significant increase in activity was seen with phenobarbital induction. Together, these results show no significant effect of high VC treatment on selected P450 activities, suggesting that no effects would occur at low VC concentrations either. Thus, protein stabilization, enzyme induction, and suicide inhibi tion were not modeled. The possibility that any of these effects might occur during longer or repeated exposures FIG. 1. Metabolism of individual chemicals. (A) Vinyl chloride () chamber concentrations decrease as it is metabolized by rats. Solid lines represent model simulations of data. (B) Trichloroethylene (O) concentra tions measured in the chamber. Solid lines represent model simulations of data. CMA 022273 INHIBITION KINETICS FOR VC/TCE MIXTURES 241 TABLE 2 Nonprotein Sulfhydryl Content in Livers of Rats Exposed for 6 hr to VC, TCE, or Mixtures Exposures (6 hr) Untreated Vinyl chloride Trichloroethylene VC/TCE mixtures initial concentration (ppm) 200 600 1000 5000 600 1000 5000 5000:600 5000:5000 1000:1000 600:5000 35:1000 NPSH (nmol/mg) 5.5 0.6 5.8 0.3 5.2 0.5 4.1 0.6"-^ 3.0 0.5"-bc J 5.5 0.8 5.7 0.7 5.4 0.9 3.6 0.4" 4.4 0.4" 4.3 0.3" 5.2 0.3 5.9 0.4 NPSH depletion <%) _ 0 5 25 44 0 0 0 35 20 22 5 0 Replicaies:rats 30:6 15:3 14:3 45:9 24:6 15:3 15:3 15:3 15:3 30:6 29:6 15:3 15:3 * Significantly different from 0-ppm control (6 hr) at p < 0.01. b Significantly different from 200 ppm (6 hr) at p < 0.01. ' Significantly different from 600 ppm (6 hr) at p < 0.01. * Significantly different from 1000 ppm (6 hr) at p < 0.01. might need to be considered further for purposes of model ing chronic bioassay data. In vivo demonstration of P450 2E1 metabolism of VC. The involvement of P450 2E1 in VC metabolism in vivo was demonstrated using irons-1,2-dichloroethylene. /DCE has been suggested to be a suicide inhibitor based upon limited in vitro studies (Costa and Ivanetich, 1982) and modeling of gas uptake results (Gargas et al., 1990), There was no direct information available on the isoform specificity of fDCE metabolism, so studies were carried out with microsomes prepared from in vivo fDCE treated ani mals. In vitro chlorzoxazone hydroxylation was reduced 58% after 1.5- and 4.5-hr fDCE exposure (40 ppm initial concentration) (Table 5). Similarly, metabolism ofp-nitrophenol dropped approximately 50% following /DCE treat ment (data not shown). /DCE blocked the metabolism of TCE in vitro between 61 and 84% (Table 6). Deethylation of 7-ethoxycoumarin, in contrast, showed a small increase (Table 7) indicating that P450 2E1-catalyzed reactivity was reduced with some specificity. Animals were exposed to /DCE for 90 min (40 ppm start ing concentration) prior to the addition of VC. The VC was added directly into the chamber so both chemicals were present for the remainder of the exposure. Metabolism of VC (200 ppm starting concentration) for 3 hr was decreased approximately 85% (Fig. 2). Liver NPSH was unaffected by exposure to /DCE alone or followed by 200 ppm VC (Table 3). Microsomes prepared from these animals showed a 90% decrease for in vitro TCE metabolizing capability, in good agreement with the in vivo results. TABLE 3 Nonprotein Sulfhydryl in Rats Exposed to VC and /DCE Exposure Untreated Vinyl chloride (3 hr) /DCE (4.5 hr) /DCE:VC (1.5:3 hr) Initial concentration (ppm) _ 600 5000 40 40:200 40:5000 NPSH (nmol/mg) 6.0 1.0* 4.9 0.5" 4.0 0.9" 6.4 0.8 5.9 0,4 5.5 0.6 NPSH depletion (%) 18 34 0 0 0 * Note significantly different from 0 ppm control (6 hr) at p < 0.01. " Significantly different from 0 ppm control (3 hr) at p < 0.01. Replicatea:rats 43:17 16:6 4:2 6:3 10:5 16:8 CMA 022274 242 BARTON ET AL. table 4 Correlation of Modeled VC Metabolism with Measured NPSH Depletion TABLE 6 In Vitro Trichloroethylene Metabolism by Rat Liver Microsomes Starting concentration (ppm) NPSH depletion measured after 6 hr (%) Modeled VC metabolized Umol) Treatment Rate (nmol/min/mg protein) SD (% of control value) Replicates:rats 600 VC 1000 VC 5000 VC 5000 VC/600 TCE 5000 VC/5000 VC 5" 25 44 35 20 55 88 114 91 42 This value represents partial recovery since depletion after 3 hr was 18%. Unexposed controls 10000 ppm VC (6 hrr 5000 ppm VC (6 hrr 40 ppm (DCE (1.5 hrr 40 ppm (DCE (4.5 hrr 40 ppm (DCE (1.5 hr) and 200 ppm VC (3 hrr Pyridine-induced 0.70 0.20 0.74 0.22 0.72 0.18 0.21 0.07* (39%) 0.11 0.05* (16%) 0.07 0.02* (10%) 1.7 0.1* (243%) 62:12 8:3 15:5 9:3 7:3 12:3 5:3 * Nominal starting concentrations in closed-chamber experiments. * Statistically different from controls at p < 0.01. Estimating metabolic parameters for trichloroethyl ene. Partition coefficients were determined in vitro for blood, liver, fat, kidney, and muscle from male Sprague- Dawley rats (Table 1). These values were used in the PBPK model to fit gas uptake data for TCE and estimate an appar ent Km and (Fig. 1). Optimization of six individual gas uptake exposures (200-6500 ppm) estimated Km = 0.09 0.11 mg/liter (approximately 1 //M) and V__ = 11.4 4.2 mg kg"1 hr-1 (87 (trnol kg"1 hr-1). Liver nonprotein sulfhvdryl after TCE. Depletion of NPSH, largely glutathione, does not occur to any signifi cant extent following trichloroethylene metabolism. This result was confirmed by analysis of livers of rats exposed to high TCE concentrations (600 to 5000 ppm initial concen tration; Table 2). Effect of tDCE preexposure on TCE metabolism. Preexposure of rats to approximately 40 ppm /DCE largely inhibited trichloroethylene metabolism at initial concen trations, although there was some apparent concentration dependence. Exposure to an initial concentration of 200 ppm showed 95% inhibition (V__. = 0.5 mg kg"1 hr-1). Exposures starting at 500 and 2000 ppm showed 71 and 76% inhibition, respectively (L,^ = 2.6 and 2.1 mg kg-1 hr"1). These data support the involvement of P450 2E1 as the major form of P450 involved in TCE metabolism even at relatively high concentrations, although some involve ment of other isoforms, as suggested by in vitro data (Nakajima el al,, 1993), cannot be excluded. Gas uptake with mixtures of VC and TCE. Gas uptake exposures were performed using several different combina tions of VC and TCE. These initial results were used in the PBPK model to estimate the various inhibitory constants required for competitive, noncompetitive, and uncompeti tive inhibition. Using these parameters, the model was run to simulate potential results given different exposure con centrations. This gave estimates of concentrations that would likely distinguish between the three forms of inhibi tion. Metabolic parameters were estimated for 15 mixtures us ing the competitive inhibition model. Apparent kinetic constants for VC are estimated to be K,, = 0.10 0.03 TABLE 5 Chlorzoxazone (25 pm) Metabolism by Rat Liver Microsomes Treatment Rate (nmol/min/mg protein) SD (percentage of control value) Replicates: Rats Unexposed controls 10000 ppm VC 5000 ppm VC 40 ppm (DCE (1.5 hr 40 ppm (DCE (4.5 hr Pyridine-induced Ethanol-induced 0.060 0.018 0.054 0.016 0.055 0.012 0.025 0.012* (42%) 0.025 0.011* (42%) 0.37 0.08* (617%) 0.27 0.03* (450%) 29:10 20:3 12:5 6:3 6:3 6:3 3:3 TABLE 7 7-Ethoxycoumarin Deethylation by Rat Liver Microsomes Treatment Rate (nmol/min/mg protein) SD (% of control value) Replicates:rats Unexposed controls 10000 ppm VC (6 hrr 40 ppm (DCE (1.5 hrr 40 ppm (DCE (4.5 hrf Phcnobarbital-induced 1.3 0.4 1.1 0.2 1.7 0.1* (133%) 1.7 0.1* (132%) 5.1 0.5* (392%) 57:11 15:3 3:2 8:3 3:3 * Nominal starting concentrations in closed-chamber experiments. * Statistically different from controls at p < 0.01. * Nominal starting concentrations in closed-chamber experiments. * Statistically different from controls at p < 0.01. CMA 022275 INHIBITION KINETICS FOR VC/TCE MIXTURES 243. parameters for uncompetitive inhibition to simulate the mixture of 25:1000. However, these alternate parameters were unsuccessful at simulating the 700:1000 mixture (Fig. 4C). The mixtures of 85:1000 gave similar results. Thus, uncompetitive inhibition does not provide satisfactory sim ulation of all the mixtures using a single consistent set of parameters. Validation of mixtures modeling with NPSH measure ments. Livers of animals exposed to mixtures of TCE and VC were removed and analyzed for NPSH. As previously described, increasing VC resulted in lower NPSH level (i.e.. more depletion) (Table 2). Increasing the concentration of TCE (600 and 5000 ppm starting concentration) present simultaneously with 5000 ppm VC resulted in less NPSH depletion, 35 and 20% respectively, (Table 4) than in the absence of TCE, 44%. FIG, 2. Metabolism of VC in the presence and absence of /DCE. (A) The decrease in VC chamber concentration is much greater in the absence () than in the presence () of /DCE. Solid lines represent model simula tions. The dashed line is modeled VC concentrations with no metabolism = 0). (B) The /DCE exposure started 1.5 hr prior to the addition of VC. The effects of /DCE cannot be accounted for by competitive inhibi tion alone. mg/liter and KmiXt = 3.1 1.0 mg kg-1 hr-1. Constants for TCE are estimated to be Km = 0.21 0.08 mg/liter and Vmtx = 9.7 2.2 mg kg-1 hr-1. These values are consistent with those estimated for the individual chemicals. Results from a representative gas uptake exposure to a mixture and the simulated results are illustrated in Fig. 3. These data were used to distinguish between competitive, noncompetitive, and uncompetitive inhibitory kinetics. It should be noted that mixtures of high concentrations of both chemicals (e.g., 5000 ppm of both VC and TCE) were uninformative for distinguishing kinetic mechanisms. Figure 3 illustrates attempts to simulate the mixture of 700 ppm VC and 1000 ppm TCE with competitive, non competitive. and uncompetitive inhibition using parame ters that reasonably simulated several other data sets. Non competitive inhibition was easily excluded (Fig. 3B). Opti mization of each individual mixture for the inhibition constants of the noncompetitive model resulted in esti mates of and KM2, varying 100-fold. Thus, it was impossible to use a single set of parameters for noncompeti tive inhibition to consistently fit all mixtures data. Uncompetitive inhibition was more difficult to rule out. For many of the mixtures, a reasonable simulation of gas uptake data was obtained using a single set of constants describing uncompetitive inhibition (e.g.. Fig. 3C). Two mixtures (25:1000 and 85:1000) were predicted by the model as not being consistent with a single set of parame ters and the data were subsequently collected verifying this (Fig. 4A). As shown in Fig. 4B, it was possible to adjust the EQL. OC (B C4O) OO n E CB U 0 ( 2 3 4 507 Time (h) FIG. 3. A mixture with initial concentrations of 700 ppm VC () and 1000 ppm TCE (O). Data (O, ) and simulation (solid line). (A) Competi tive inhibition; (B) noncompetitive inhibition; (C) uncompetitive inhibi tion. Reasonable fits are obtained with competitive or uncompetitive inhi* bition but not noncompetitive using parameters listed under Methods. CHA 022276 244 BARTON ET AL. TCE is a more effective inhibitor of VC metabolism than vice versa. This is due to the greater blood-air partition coefficient for TCE which results in a higher concentration in blood at the same exposure concentration. At concentra tions below 30 ppm of each chemical there is little effect on the metabolwm of the other chemical. As the concentra tions of either chemical increases, inhibition of metabolism becomes increasingly significant (Fig. 5; Table 8). DISCUSSION FIG. 4. Demonstration that a single set of values of parameters for uncompetitive inhibition will not lit all the data. (A) Failure to model the mixture 25 ppm VC and 1000 ppm TCE using ATM1I = A(MI( = 3.0 (as in Fig. 3). (B) Modeling the mixture of 25:1000 using KUI1 = 3.0andA'M;, = 0.20 provides a good fit. (C) The parameter values in (B) do not provide a good simulation for 700 ppm VC and 1000 ppm TCE. Modeling competitive inhibition during constant concen tration exposures. Human exposures of interest for risk assessment may approximate constant concentration expo sures. Occupational exposures are typically of 8-hr duration and environmental exposure is 24 hr. Therefore, exposures of rats to mixtures of VC and TCE were simulated for 8and 24-hr periods. It should be noted that near steady state is achieved rapidly at low concentrations for both chemi cals. Within a half hour, the venous concentration of VC is near steady state and TCE is about 75% steady state. The metabolism of VC and TCE over 8 and 24 hr is approxi mately a linear function of time, due to the rapid approach to steady state. Therefore, only the 24-hr results are pre sented here (Table 8; Fig. 5). Modeling was begun with the assumption that a single saturable pathway representing P450 2E1 in the liver com partment might be adequate to describe initial oxidative metabolism of VC and TCE. However, suicide inhibition via formation of a heme adduct by VC or related vinyl halogens has previously been demonstrated (Watanabe el al., 1976; Ortiz de Montellano et al. 1982). P450 2E1 is also frequently "induced" by exposure to its substrates, al though the mechanisms may not be transcriptional (Yang et al., 1990). Finally, limited in vitro data suggested that VC is metabolized by several P450 isoforms (Reynolds et al, 1975) while extensive studies show this to be true for TCE (Miller and Guengerich, 1983). Modeling mixtures requires having good models for the individual chemicals, so these aspects were investigated further. Suicide inhibition of P450 by VC has been demonstrated both in vitro and in vivo at high concentrations (about 5% or 50,000 ppm) using untreated and phenobarbital-induced animals (Ivanetich et al., 1977; Reynolds et al., 1975). To determine if this effect needed to be modeled, metabolism of several P450 substrates was measured in liver microsomes from rats exposed to several thousand ppm of VC. No differences were found between control and VC-treated rats in metabolism of chlorzoxazone, TCE, or 7-ethoxycouraarin, Because suicide inhibition is a stochastic occur rence, these data indicate that it occurs too rarely to be apparent in vivo at concentrations used in chronic studies (^10,000 ppm) with VC. No evidence exists for such effects with TCE (Pessayre et al., 1979). Therefore, suicide inhibi tion was not modeled for either compound. These studies also provide no evidence for increased P450 activity due to exposure to VC or TCE for 6 hr at high concentrations. P450 2E1 activity is increased by both tran scriptional and post-transcriptional regulation including protein stabilization by substrate (Yang et al, 1990). No such effects were apparent using in vitro assays and microsomes from VC-exposed animals. The in vivo data for both VC and TCE were consistent with a constant V___ Whether changes might occur during chronic exposures such as during cancer studies has not been determined. The issue of modeling P450 isoforms was considered. This issue is important for extrapolation from high to low CMA 022277 INHIBITION KINETICS FOR VC/TCE MIXTURES TABLE 8 Estimated Values Obtained by Simulating Constant Concentration Exposures to VC/TCE Mixtures for Rats VC (ppm) TCE (ppm) Inhibition (%)' ''AMl b (jimo!) A*/ (iimol) Cf (,iM) Cvt/ (jiM) Cv,' UiM) 0 10 0 0 61 0 0.2 0 1 10 0 3 61 0.01 0.2~- " 0.05 10 10 0 25 60 0.1 0.3 0.5 100 10 5 231 58 1.7 0.5 5.5 300 10 21 423 48 15 2 22 1.000 10 59 476 25 82 5 89 3.000 10 83 495 11 277 7 285 10.000 10 94 502 4 966 8 978 Percentage inhibition of TCE metabolism due to increasing VC. 4 rlM. amount metabolized of chemical 1 (VC) and chemical 2 (TCE). f CVL, venous concentration leaving liver, ' Cv, venous blood concentration. 245 Cv/ (M1 1.2 1.2 1.2 1.4 2.7 5.3 7.1 8.0 doses and across species using PBPK models as is required for cancer risk assessment. Substrates for P450 are frequently metabolized by multi ple isoforms, often in a concentration-dependent manner. Descriptions of this complexity might be required, for in stance, ifthe metabolite profile changed with concentration or across species. Such modeling would require estimates of the kinetic parameters (e.g., K^, Km) and concentration for each isoform. Alternatively, it may be satisfactory to simply model an "averaged" P450 enzyme where the appar ent Vnuu and K,, describe the metabolism without describ ing the activity of each isoform. Based upon existing literature one would expect P450 2E1 to metabolism both TCE and VC (Guengerich et at., 1991; Miller and Guengerich, 1983), However, monoclonal FIG. 5. Prediction of inhibition of metabolism VC or TCE during a 24-hr exposure simulated for constant concentration exposures of rats to mixtures of VC and TCE. Indicated points represent modeling output with lines connecting them to approximate the shape ofthe curve. Inhibition by TCEofVC metabolism (VCat (O) 1 or 10ppm;(O) 100ppm)and metabo lism and inhibition by VC ofTCE (TCE at (A) I or 10 ppm; (0) 100 ppm). TCE is a more effective inhibitor of VC metabolism than vice versa. antibodies to P450 2E1 only inhibited 60% of the trichlor oethylene metabolism in control microsomes (Nakajima et ai, 1993). P450 2B1 also metabolizes TCE, but is present at very low levels in untreated rats so its involvement would be expected to be minimal. TCE is also metabolized by the male rat-specific isoform P450 2C11 in vitro. /DCE exposure at moderate concentrations was a highly effective inhibitor of the metabolism of the other chlori nated ethylenes. Quantitatively, the inhibition of TCE me tabolism was similar in vitro and in vivo. In vitro assays of enzyme activity with microsomes from the /DCE-treated rats clearly show large decreases in other activities asso ciated with P450 2E1 (chlorzoxazone, p-nitrophenol). Es sentially no change was seen in 7-ethoxycoumarin activity by comparison. These results show that P450 2E1 activity accounts for much of the TCE and VC metabolism in vivo with un treated animals, except perhaps at high concentrations. /DCE was somewhat less effective at inhibiting the metabo lism of TCE at 500 and 2000 ppm than at 200 ppm and lower (71-76% versus 95%). The involvement of a single isoform may simplify the difficulties of interspecies extrapo lation. Measurements of P450 2E1 activity using readily available in vitro systems could permit scaling of metabolic terms or direct use of the values determined in vitro. If the values for Km are similar across species, one may be able to use the relative activity of P450 2EI across species deter mined with any substrate. This would be particularly use ful with humans where in vivo estimates are unavailable for VC. Coexposure to VC and TCE are successfully described as. competitive inhibition using a fixed single saturable path way for oxidative metabolism. A wide range of mixtures was used in gas uptake studies in order to distinguish be tween kinetic inhibition models. Competitive inhibition consistently simulated the data using a single set ofparame CMA 022278 246 BARTON ET AL. ters. Furthermore, the kinetic parameter values calculated from the competitive inhibition model are essentially iden tical to those obtained from studies with individual chemi cals. By contrast, noncompetitive inhibition would fit very few data sets with a single set of parameter values. Uncom petitive inhibition tended to more readily simulate multiple data sets except for several key mixtures that were only fit with widely varying parameter values. The results of the mixtures simulations were further vali dated using glutathione depletion. This endpoint was cho sen because it was dependent upon metabolism of VC. Glu tathione conjugation is an important pathway for detoxifi cation of VC metabolites. This is apparently only an acute effect since GSH levels return to control values as exposures continue over a period of days (Du et at.. 1979). The poten tial implications of this aspect for modeling VC carcinoge nicity remain to be explored. Finally, the results of this study are of interest from the perspective of risk assessment for mixtures. Exposure of humans to chlorinated ethylene mixtures may occur occu pationally. generally by inhalation. Environmental expo sure may be oral or by inhalation. Exposure to water may involve both oral and inhalation pathways, the latter due to water uses such as washing or showering. The inhalation studies reported here may readily be extrapolated to the oral route using the PBPK model. Although one might expect P450 2E1 substrates to be competitive inhibitors in mixtures, such effects are highly concentration dependent. The inhibition effects observed here only occur at relatively high concentrations, as has also been reported for competitive inhibition between toluene and xylene (Tardif et at.. 1993). This arises from several factors. The perfusion limitation at low concentrations is one factor. Metabolism of the inhibitor, the parent com pound, at low concentrations diminishes its effectiveness. The degree of inhibition is not simply predictable from the metabolism equation due to the dependence of the steady-state venous concentration upon the rate of metabo lism and vice versa. Competitive inhibition may be de scribed as resulting from an alteration of the K,,, for the substrate (e.g., r = (1 1 + I/K,) + 5]: v is velocity. 5 is substrate concentration, and / is inhibitor concentra tion). Thus, as the concentration of inhibitor increases the effective A'm would increase and metabolism would de crease. However, as metabolism decreases, the steady-state venous concentration of the substrate increases. These two effects counterbalance up to a point. For instance, the predicted steady-state concentration of VC in venous rat blood at an exposure concentration of 10 ppm is 0.5 /jm. However, as metabolism decreases to zero, the venous concentration increases to 1 hm. Similarly. TCE (10 ppm) is predicted to give a venous concentration of 1.2 um increasing to 8 in the absence of metabolism (see Table 8). This effect loses importance as the inhibitor ve nous concentration increases dramatically (orders of mag nitude) as exposure concentration increases similarly. These discussions are all based upon experimental data and simulations for rats. However, similar results are ob tained for simulations of humans. Simulations were run scaling the rat model to a 70-kg human using values in the literature (Chen and Blancato. 1989; Fisher and Allen. 1993). Although the values for VC. particularly, must be considered limited estimates due to a lack of much data, the results for the mixture were essentially the same. That is. no inhibition was seen for either chemical at concentrations below approximately 10 ppm. Occupational exposures to these chlorinated hydrocar bons may be as high as tens of ppm, but environmental exposures are more typically low ppb levels. These latter are clearly in a range where the metabolism becomes perfusion limited and enzyme is saturating, so no inhibition would be predicted to occur. U nder these conditions the pharmacoki netics of the compounds would be independent. This is consistent with the standard U.S. EPA approach of assum ing additivity. However, it remains to be shown that the pharmacodynamics of the two chemicals are also indepen dent. Toxicity, and thus, chemical risk assessment, ulti mately depends upon both pharmacokinetics and pharma codynamics. ACKNOWLEDGMENTS We acknowledge the expen assistance of Carlyle Flemming for the sta tistical analyses and Joanne Drerup for her laboratory work. This work was performed under Depanment of the Air Force Contract F336I5-90-C0532 with funding provided by the Strategic Environmental Research and Development Program. AL/OE-TR-1995-OOOl. REFERENCES ATSDR (Agency for Toxic Substances and Disease Registry!! 1989). Toxi cological Profile for l invl Chloride. ATSDR/TP-88/25. Andersen. M. E. (1981). A physiologically based toxicokinetic description of the metabolism of inhaled gases and vapors: Analysis at steady slate. Toxicol Appl. Pharmacol. 60, 509-526. Andersen, M. E.. Gargas, M. L.. Clewell. H. J..and Severyn. K. M. (1987), Quantitative evaluation of the metabolic interactions between trichlor oethylene and l.l-dichloroethylene in vivo using gas uptake methods. Toxicol Appl. Pharmacol. 89, 149-157. Baker. M. A.. Cemiglia, G. J,, and Zaman. A. (1990). Microtiter plate assay for the measurement of glutathione and glutathione disulfide in large numbers of biological samples. Anal. Biochem. 190, 360-365. Bruckner. J. V,. Davis, B. D.. and Blancato, J. N. (1989). Metabolism, toxicity, and carcinogenicity oftrichloroethylene. Crii. Rev. Toxicol. 20, 31-50. Bull, R. J., Sanchez. I. M., Nelson, M. A.. Larson, J. L., and Lansing, A. J. (1990). Liver tumor induction in B6C3F1 mice by dichloroacetate and trichloroacetate. Toxicology 63, 341-359. Chen, C. W.. and Blancato. J. N, (1989). Incorporation ofbiologica) infor mation in cancer risk assessment: Example-vinyl chloride. Cell Biol Toxicol 5, 417-444. CHA 022279 INHIBITION KINETICS FOR VC/TCE MIXTURES 247 Clement International (1990). Development and I atidaiion ofMethods for Applying Pharmacokinetic Data in Risk Assessment I'invl Chloride. V0I. 5. AAMRL-TR-90-072. Air Force Medical Research Laboratory Wnght-Patterson Air Force Base. OH. Costa. A. K., and Ivanetich. K. M. (1982). The 1,2-dichloroethvlenes; Their metabolism by hepatic cytochrome P-450 in vitro Biochem. Phar- nMJFW 31. 2093-2102. * D'Souza, R. W,, Francis. W, R.. and Andersen. M. E. (1988). Physiological model for tissue glutathione depletion and increased resvnthesis after ethylene dichloride exposure. J. Pharmacol Exp Ther. 245, 563-568. DeAngelo. A. B,, Daniel. F. B.. Stober, J. A., and Olson. G. R. (1991). The carcinogenicity of dichloroacetic acid in the male B6C3F1 mouse. Fundam. Appl. Toxicol. 16, 337-347. Delp. M. E., Manning, R. O., Bruckner, J. V,, and Armstrong. R. B. (1991). Distribution ofcardiac output during diurnal changes of activity in rats. Am. J. Physiol. 261. H1487-H1493. Du. J. T.. Sandoz, J. P,, Tseng. M. T,. and Tamburro. C. H. (1979). Bio chemical alterations in livers of rats exposed to vinyl chloride. J. Toxi col. Environ. Health 5, 1119-1132. Ellman. G. L. (1959). Tissue sulfhydrvl groups. Arch. Biochem. Biophys 82. 70-77. Fedtke. N., Boucheron, J. A.. Walker. V. E., and Swenberg. J. A. (1990). Vinyl chloride-induced DNA adducts. II. Formation and persistence of 7*(2'-oxoethvl)guanine and A^-eihenoguanine in rat tissue DNA. Car cinogenesis 11, 1287-1292. Fisher. J. W,, and Allen. B. C. (1993). Evaluating the risk of liver cancer in humans exposed to trichloroethylene using physiological models. Risk Anal. 13, 87-95. Freedman, D. L.. and Gossett. J. M. (1989). Biological reductive dechlor ination of tetrachloroethylene and trichloroethylene to ethylene under methanogenic conditions. Appl. Environ. Microbiol. 55, 2144-2151. Gargas. M. L.. Andersen. M. E.. and Clewell. H. J. (1986). A physiologi cally based simulation approach for determining metabolic constants from gas uptake data. Toxicol Appl. Pharmacol 86, 341-352. Gargas. M. L., Clewell. H. J.. and Andersen. M. E. (1990). Gas uptake inhalation techniques and the rates of metabolism of chloromethanes. chloroethanes. and chloroethvlenes in the rat. Inhal Toxicol 2, 295- 319. ' Gearhart. J. M., Mahle, D. A.. Greene. R. J.. Seckel. C. S,, Hemming. C. D.. Fisher. J. W.. and Clewell. H. J. (1993). Variability of physiologi cally based pharmacokinetic (PBPK) model parameters and their effects on PBPK model predictions in a risk assessment for perchloroethylene (PCE). Toxicol. Lett. 68. 131-144. Germolec. D. R., Vang. R. S. H.. Ackermann. M. F.. Rosenthal. G. J.. Boorman. G. A.. Blair. P,, and Luster. M. 1. (1989). Toxicology studies ofachemical mixture or25 groundwater contaminants. 11. Immunosup pression in B6C3FI Mice. Fundam. Appl. Toxicol. 13. 377-387. Greenlee. W. F.. and Poland, A. (1978). An improved assay of 7-ethoxycoumarin O-deethvlase activity: Induction of hepatic enzyme activity in C57BL/6J and DBA/2J mice by phenobarbital. 3-methylcholanthrcne. and 2.3.7.8-tetrachlorodibcnzo-p-dioxin. J. Pharmacol Exp Tlier. 205, 596-605. Guengerich. F, P.. Kim. D-H., and Jwasaki. M. (1991). Role of human cytochrome P-450 HE I in the oxidation of many low molecular weight cancer suspects. Chem. Res. Toxicol. 4, 168-179. Guengerich. F. P., and Strickland. T. W. (1977). Metabolism of vinyl chlo ride: Destruction of the heme of highly purified liver microsomal cy tochrome P-450 by a metabolite. Alolec. Pharmacol. 13, 993-1004. Gwinner. L. M., Laib. R. J.. Filser. J. G., and Bolt. H. M. (1983). Evidence of chloroethvlene oxide being the reactive metabolite of vinyl chloride towards DNA: Comparative studies with 2.2'-dichloro-diethy leiher. Car cinogenesis 4, 1483-1486. Herren-Freund. S. L.. Pereira. M. A.. Khourv. M. D.. and Olson. G. (1987). The carcinogenicity of trichloroethylene and its metabolites, tri chloroacetic acid and dichloroacetic acid, in mouse liver. Toxicol. Appl Pharmacol. 90, 183-189. Ivanetich, K. M.. Aronson. I., and Katz. I. D. (197.7). The interaction of vinyl chloride with rat hepatic microsomal cytochrome P-450 in uiro. Biochem. Biophys Res. Commun. 74, 1411-1418. Miller. R. E.. and Guengerich. F. P. (1983). Metabolism of trichloroethy lene in isolated hepaiocvies. microsomes. and reconstituted enzyme sys tems containing cytochrome P450. Cancer Res. 43. 1145-1152. Nakajima, T., Wang, R-S.. Elovaara. E.. Park. S. S.. Gelboin. H. 3'.. and Vainio. H. (1992). A comparative study on the contribution of cy tochrome P450 isozymes to metabolism of benzene, toluene, and trich loroethylene in rat liver. .Brdc/iem Pharmacol 43, 251-257. Nakajima, T., Wang. R-S.. Elovaara. E.. Park. S. S.. Gelboin. H. V.. and Vainio. H. (1993). Cytochrome P450-related differences between rats and mice in the metabolism of benzene, toluene and trichloroethylene in liver microsomes. Biochem. Pharmacol 45, 1079-1085. Nakajima. T.. Wang. R-S.. Murayama, N., and Sato, A. (1990). Three forms of trichloroethylene-metabolizing enzymes in rat liver induced by ethanol, phenobarbital. and 3-methylcholanthrene. Toxicol. Appl. Phar macol. 102, 546-552. Ortiz de Montellano. P. R.. Kunze. K. L.. Beilan. H. S.. and Wheeler. C. (1982). Destruction of cytochrome P-450 by vinyl fluoride, fluroxene, and acetylene: Evidence for a radical intermediate in olefin oxidation. Biochemistry 21. 1331-1339. Pessayre. D.. Allemand. H.. Wandscheer. J. C.. Descatoire, V., Artigou. J-Y., and Benhamou. J.-P. (1979). Inhibition, activation, destruction, and induction of drug-metabolizing enzymes by trichloroethylene. Tox icol. Appl. Pharmacol. 49, 355-363. Pessayre, D.. Wandscheer. J. C.. Descatoire, V.. Artigou. J. Y..and Benha mou. J. P. (1979). Formation and inactivation of a chemically reactive metabolite of vinyl chloride. Toxicol. Appl. Pharmacol. 49, 505-515. Peter. R., Bocker, R.. Beaune. P. H.. Iwasaki. M.. Guengerich. F. P.. and Yang. C. S. (1990). Hydroxy lation of chlorzoxazone as a specific probe for human liver cytochrome P-45011E1. Cliem. Res. Toxicol. 3, 566573. Plugge. H.. and Safe. S. (1977). Vinyl chloride metabolism--A review. Cliemospliere 6, 309-325. Reynolds. E. S.. Moslen. M. T., Szabo. S., and Jaeger, R. J. (1975). Vinyl chloride-induced deactivation of cytochrome P-450 and other compo nents of the liver mixed function oxidase system: An in vivo study. Res. Commun Chem Pathol. Pharmacol 12, 685-694. Rosner. B. (1990). Fundamentals ol Biostaiistics. PWS-Kent. Boston. Sato. A., and Nakajima. T. (1979). A vial-equilibration method to evaluate the drug-metabolizing enzyme activity for volatile hydrocarbons. Toxi col Appl. Pharmacol 47,41-46. Tardif. R.. Lapare. S.. Krishnan. K.. and Brodeur. J. (1993). Physiologi cally based modeling of the toxicokinetic interaction between toluene and m-xylene in the rat. Toxicol. Appl. Pharmacol. 120, 266-273. U.S. EPA (1986). Guidelinesfor the Health Risk Assessment ofChemical Mixtures. 51 FR 34014. U.S. EPA (1988). Reference Physiological Parameters in Pharmacokinetic Modeling. EPA/600/6-88/004. Watanabe. P. G.. Hefner, R. E.. and Gehring. P. J. (1976). Vinyl chlorideinduced depression of hepatic non-protein sulfhydrvl content and ef fects on bromosulphalein (BSP) clearance in rats. Toxicology 6, 1-8. Yang. C. S.. Yoo. J-S. H., Ishizaki. H.. and Hong. J. (1990). Cytochrome P450IIEI: Roles in nitrosamine metabolism and mechanisms of regula tion. Drug. Mctah Rev. 22, 147-159. CMA 022280