Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

Hepatocellular damage involving centrilobular hepatocytes is manifested in rodents after treatment with 1,1-dichloroethylene (DCE¹), a synthetic monomer used widely in the plastics industry and a prevalent environmental contaminant. Hepatocyte necrosis and thrombosis were observed in rats and mice exposed to DCE (Reynolds et al., 1975; Chieco et al., 1982; Forkert et al., 1986). Previous studies, using rat liver microsomal incubations, have identified the primary metabolites formed from cytochrome P450-dependent oxidation of DCE as the epoxide, 2,2-dichloroacetaldehyde, and 2-chloroacetyl chloride (Liebler and Guengerich, 1983; Costa and Invanetich, 1984; Liebler et al., 1985, 1988). Our studies in mice confirmed that these primary metabolites were also formed in liver microsomal incubations (Dowsley et al., 1995, 1996) (Fig. 1). The epoxide was the major metabolite generated, whereas 2-chloroacetyl chloride was detected only at minimal levels. Acetal, the hydrate of 2,2-dichloroacetaldehyde, was also detected. The secondary metabolites formed were the products of hydrolysis and/or conjugation of the DCE metabolites with glutathione (GSH). The major intermediates produced were derived from conjugation of the epoxide with GSH and were identified as 2-(S-glutathionyl)acetyl glutathione [B] and 2-S-glutathionyl acetate [C]. 2-Chloroacetic acid and S-(2-chloroacetyl)glutathione [D], the hydrolysis and GSH-conjugated products of 2-chloroacetyl chloride, respectively, were detected but were found at minimal levels in our microsomal incubations (Dowsley et al., 1995, 1996). S-(2,2-Dichloro-1-hydroxy)ethyl glutathione [A], the metabolite formed from conjugation of GSH with 2,2-dichloroacetaldehyde, was not detected. In subsequent studies, we have shown that conjugate [C] was the major DCE metabolite formed in vivo (Forkert, 1999a). These results confirmed that the epoxide is the major metabolite generated from DCE metabolism in murine liver.

View larger version (13K): [in this window] [in a new window]   Fig. 1.   Scheme of proposed pathway of DCE metabolism. The GSH conjugates formed from DCE metabolism are identified as follows: [A], S-(2,2-dichloro-1-hydroxy)ethyl glutathione; [B], 2-(S-glutathionyl)acetyl glutathione; [C], 2-S-glutathionyl acetate; [D], S-(2-chloroacetyl)glutathione.

Substantial data have accrued to demonstrate that hepatic CYP2E1 is a major P450 enzyme mediating the oxidative metabolism of DCE (Lee and Forkert, 1994). Evidence has been derived, in part, from experimental maneuvers designed to modulate formation of DCE metabolites, including the DCE epoxide. Findings from in vitro studies showed that the epoxide, as assessed by formation of conjugates [B] and [C], was generated readily from DCE in liver microsomal incubations and was augmented 3-fold in incubations containing liver microsomes from mice treated with 1% acetone for 8 days (Dowsley et al., 1995). This acetone treatment regimen has been shown to produce a 5- and 4-fold induction of the CYP2E1 protein and p-nitrophenol hydroxylase activity, respectively (Forkert et al., 1994). These data are consistent with results from in vivo studies, which showed that epoxide levels present in liver cytosol were markedly diminished in mice pretreated with the CYP2E1 inhibitor diallyl sulfone (Forkert et al., 1996b; Forkert, 1999a). The epoxide-derived GSH conjugates were also detected in bile isolated from gallbladders of mice treated with DCE, although the amounts were significantly reduced in mice pretreated with diallyl sulfone to inhibit the CYP2E1 enzyme.

In view of the important role of CYP2E1 in DCE metabolism, we reasoned that differences in constitutive expression of this P450 should produce differential metabolic activation of DCE. This premise is substantiated by results from our previous studies, which showed that levels of lung CYP2E1 in female mice were about 50% higher than that in male mice, and these differing CYP2E1 levels correlated with magnitudes of formation of the DCE epoxide and 2,2-dichloroacetaldehyde (Forkert et al., 1996a). In recent studies, we have identified strain differences in CYP2E1 levels in the lungs of A/J, CD-1, and B6 mice (Forkert et al., 2001; Titis and Forkert, 2001). Strain A/J mice expressed significantly greater amounts of CYP2E1 than B6 mice, whereas intermediate levels were expressed in CD-1 mice. The differing CYP2E1 levels in these strains of mice were reflected in formation of varying amounts of the DCE epoxide (Forkert et al., 2001). Although CYP2E1 expression varied in the lungs of different strains of mice, this phenomenon has not been investigated in the liver, and it is not known whether differences in strain-related hepatic CYP2E1 levels are manifested. In this study, we have investigated CYP2E1 expression in the livers of A/J, CD-1, and B6 mice and correlated the potential differences with the extents to which DCE is metabolized in the livers of these murine strains. We evaluated DCE metabolism by measuring covalent binding of DCE to liver proteins and by determining formation in vivo of the DCE epoxide-derived GSH conjugates. We have also estimated hepatic GSH levels in control and DCE-treated mice to determine whether rates of conjugation of the epoxide with GSH differed in the three murine strains. We have additionally investigated the distribution of DCE protein adducts and determined the histopathological alterations resulting from DCE treatment.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Chemicals and Reagents. Chemicals and reagents were obtained from commercial suppliers as follows: DCE (>99% purity), phosphoric acid (85%, v/v), and GSH (Aldrich Chemical, Montreal, QC, Canada); Bio-Rad protein assay dye reagent concentrate (Bio-Rad, Hercules, CA); bovine serum albumin (BSA), glucose-6-phosphate, glucose-6-phosphate dehydrogenase, p-nitrophenol, 4-nitrocatechol, glutaraldehyde (50% aqueous), paraformaldehyde, hydrogen peroxide (30%, v/v), 3'3'-diaminobenzidine tetrahydrochloride, -glutamyl glutamate, NADP⁺, and NADPH (Sigma, St. Louis, MO); [¹⁴C]DCE (99% pure by gas liquid chromatography, specific activity 11.3 nCi/nmol; Amersham Pharmacia Biotech, Arlington Heights, IL); Universol scintillation fluid (ICN, Costa Mesa, CA); Spectrapor-3 dialysis tubing, 3500 molecular weight cut-off (Fisher Scientific, Nepean, ON, Canada); sodium pentobarbital (Somnotol; MTC Pharmaceuticals, Hamilton, ON, Canada); biotinylated goat anti-rabbit IgG, avidin-biotin blocking reagent (Vector Laboratories, Burlingame, CA); and streptavidin conjugated to horseradish peroxidase (Zymed Laboratories, Inc., South San Francisco, CA). Rat CYP2E1-expressed human B-lymphoblastoid microsomes (GENTEST, Woburn, MA) were used as CYP2E1 standards in the protein blots. The CYP2E1 monoclonal antibody (mAb 1-98-1) was donated by Dr. S. S. Park (Laboratory of Comparative Carcinogenesis, National Cancer Institute, Frederick, MD). The DCE epoxide was synthesized by oxidation of DCE with m-chloroperbenzoic acid and conjugated to GSH to form [C], using methods described previously (Dowsley et al., 1995). Conjugate [C] was used as a standard for metabolite identification, using HPLC analysis. The conjugate glycine-glutaraldehyde-BSA, used as a reagent to inhibit nonspecific reactions in the immunohistochemical experiments, was synthesized as described in our previous studies (Forkert et al., 1997).

Animal Treatment. Female A/J mice weighing 20 to 25 g were purchased from Jackson Laboratories (Bar Harbor, MA). Female C57BL/6 and CD-1 mice, also weighing 20 to 25 g, were purchased from Charles River Canada (St. Constant, QC, Canada). Mice were acclimated to laboratory conditions for 1 week before being entered into an experiment. They were maintained on a 12-h light/dark cycle, and provided free access to food (Mouse Diet 5015; PMI Nutrition International, Inc., Brentwood, MO) and drinking water.

Preparation of Microsomes. Mice were sacrificed by cervical dislocation, and livers from six mice were pooled and homogenized in 4 volumes of cold phosphate-buffered KCl (100 mM K₂HPO₄, 1.15% KCl, and 1.15 mM EDTA, pH 7.4). Microsomes were prepared by differential centrifugation, as described previously (Forkert, 1995). Microsomal pellets were resuspended in 1 ml of the same buffer, and aliquots were frozen in liquid nitrogen and stored at 70°C. Protein concentrations were determined by the Bradford method (1976), using BSA as the standard.

Protein Immunoblotting. Liver microsomes were subjected to SDS-polyacrylamide gel electrophoresis, as described in our previous studies (Lee et al., 1998). Briefly, microsomal proteins and CYP2E1 standards were electrophoretically separated and transferred to a nitrocellulose membrane. The membrane was subsequently reacted with a CYP2E1 monoclonal antibody (mAb 1-98-1 (1:1000) (Park et al., 1986) diluted in Tween 20/Tris-buffered saline containing 1% gelatin. After thorough rinsing in the buffer to remove unreacted antibodies, the nitrocellulose membrane was incubated for 2 h with IgG conjugated to alkaline phosphatase (1:1000). The protein bands were visualized by reaction with a solution containing p-nitroblue tetrazolium chloride and 5-bromo-4-chloro-3-indolyl phosphate p-toluidine salt.

Covalent Binding of [¹⁴C]DCE. A/J, CD-1, and C57BL/6 mice were treated with [¹⁴C]DCE (20 µCi/kg, 125 mg/kg i.p.) and were sacrificed 1 h after treatment. Livers were frozen in liquid nitrogen and stored at 70°C until processed. Covalent binding of DCE metabolites to liver proteins was determined by equilibrium dialysis, as described previously (Forkert, 1999a). Tissues were homogenized in 5 volumes of cold 0.01 M sodium phosphate buffer, pH 7.0, containing 2% SDS. After addition of 5 volumes of 4% SDS, the samples were boiled for 15 min. Aliquots (1 ml) were transferred to dialysis tubing (molecular weight cut-off 3500 Da) and dialyzed overnight, with stirring, against 500 ml of 0.01 M sodium phosphate buffer, pH 7.0, containing 0.1% SDS. Subsequently, aliquots (250 µl) of the dialysate were combined with 1 ml of 1.0 N NaOH and reacted overnight for solubilization of proteins. Levels of radioactivity in the dialysate and buffer were determined, and the difference in the levels of radioactivity in the dialysate and buffer was considered as the quantity of covalently bound DCE in the sample. Protein concentrations were determined by the method of Lowry et al. (1951).

DCE Metabolites in Liver Cytosol. Levels of conjugate [C] were determined in liver cytosol 1 h after treatment with [¹⁴C]DCE (40 µCi/kg, 125 mg/kg i.p.). Livers were homogenized in cold phosphate-buffered KCl, pH 7.4, in a volume of 1 ml/g of liver tissue (n = 3 for A/J, CD-1, and C57BL/6 mice). Liver cytosolic fractions were isolated according to procedures used in our previous studies (Forkert, 1999a). Aliquots of cytosol were frozen in liquid nitrogen and stored at 70°C until metabolite determination. Protein concentrations were determined by the method of Bradford (1976). Proteins in the samples (250 µl) were precipitated with 70% (v/v) PCA and removed by centrifugation. The supernatant (100 µl) was subjected to reversed phase HPLC analysis using a C₁₈ column (5 µm, 4.6 × 250 mm, Microsorb-MV; Rainin Instruments, Woburn, MA). The mobile phase (0.2% phosphoric acid, pH 2.4) was set at 1 ml/min isocratic flow and monitored at 200 nm. Fractions of the column effluent (0.25 ml) were collected and radioactivity was determined by liquid scintillation spectroscopy. Metabolites were identified by retention times of the synthesized standards, and levels were estimated by summing the radioactivity associated with each peak and converting the data to nanomolar amounts using the specific activity of [¹⁴C]DCE.

DCE Metabolites in Bile. Biliary DCE metabolites were estimated using procedures described in our previous studies (Forkert, 1999a). Bile samples were obtained from gallbladders of mice treated with [¹⁴C]DCE (40 µCi/kg, 125 mg/kg i.p.). Proteins in the bile samples were precipitated with 70% PCA and removed by centrifugation. Samples of the supernatants (5 µl) were diluted with 95 µl of mobile phase (0.2% phosphoric acid) and were then subjected to HPLC analysis for identification of DCE metabolites, as described previously.

Liver Glutathione Content. Liver GSH content was determined in control and DCE-treated (125 mg/kg DCE i.p.) mice using the method of Fariss and Reed (1987). Mice were sacrificed 1 h after DCE treatment. Liver tissue (50-100 mg) was pulverized in liquid nitrogen and then mixed with 1 ml of 10% (v/v) PCA containing 1 mM bathophenanthrolinedisulfonic acid. PCA-insoluble material was removed by centrifugation. Aliquots of the supernatant (0.5 ml) were combined with 50 µl of -glutamyl glutamate (1.5 mM in PCA; 0.3%, v/v), which was used as the internal standard. Iodoacetic acid (100 mM in 0.2 mM m-cresol purple, 50 µl) was then added, and the pH was adjusted to 8 to 9 with 0.48 ml of a solution containing 2 M KOH and 2.4 M KHCO₃. After a 10-min incubation in the dark, 1 ml of Sanger's reagent (1-fluoro-2,4-dinitrobenzene; 1%, v/v, in ethanol) was added, and the mixture was stored overnight at 4°C. After centrifugation, 100 µl of the supernatant was subjected to HPLC analysis using an aminopropyl silica ion-exchange column (5 µm, 4 × 250 mm; SGE International, PTY Ltd., Ringwood, Australia). Mobile phase A (80% methanol in H₂O) was maintained at a flow rate of 0.75 ml/min. Increasing concentrations of mobile phase B (0.5 M sodium acetate in 64% methanol) were added to elute the chromophore derivative. After injection of the derivatized sample (100 µl), the gradient of mobile phase B was adjusted from 0 to 95% over 10 min. The concentrations of the derivatized samples of GSH and -glutamyl glutamate were monitored and estimated at 365 nm.

Immunohistochemical Localization of DCE Protein Adducts. In the immunohistochemical experiments, we have used a polyclonal antibody that recognizes protein adducts comprising conjugates of DCE epoxide with GSH or cysteine residues on proteins (Forkert et al., 1997; Forkert, 1999b). For the sake of convenience and ease of reporting, we have designated the proteins detected by this antibody as DCE protein adducts. Immunohistochemical studies for detection and localization of the protein adducts were performed in liver tissues from all three strains of mice treated with DCE (125 mg/kg i.p.) or the vehicle. The immunohistochemical experiments were carried out as described in our previous studies (Forkert, 1999a). Briefly, tissues were fixed with 4% paraformaldehyde containing 0.2% glutaraldehyde in 0.1 M sodium cacodylate, pH 7.3. Immunohistochemical localization of the DCE protein adducts was performed with paraffin-embedded tissue sections using the avidin-biotin complex technique, as described (Forkert, 1999a). The distribution of the protein adducts was visualized by development in 3,3'-diaminobenzidine (0.05% in 0.01% hydrogen peroxide in phosphate-buffered saline). Tissue sections were then dehydrated, cleared, and mounted.

Histopathology. Hepatotoxicity in A/J, CD-1, and C57BL/6 mice was assessed at 24 h after DCE treatment (50, 75, 125, and 175 mg/kg i.p.). Preliminary studies confirmed that hepatocellular damage, as assessed by light microscopy, was not morphologically apparent at early time points. This finding is consistent with that reported in previous studies showing that hepatocyte necrosis was manifested 24 h after DCE exposure (Forkert et al., 1986). Liver tissue was prepared for histopathological evaluation as previously described (Forkert, 1999a), with minor modifications. Livers were fixed by vascular perfusion through the left ventricle with 4% paraformaldehyde in 0.1 M Sorensen's phosphate buffer (12.0 mM NaH₂PO₄, 69.0 mM Na₂HPO₄), pH 7.4. Tissues were processed and embedded in paraffin, using standard procedures. Liver sections (5 µm) were stained with hematoxylin and eosin.

Statistical Analysis. Data are expressed as mean ± S.D. Statistical analysis was performed by one- or two-way analysis of variance followed by Tukey's test to identify significant differences (p < 0.05) between experimental groups.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Protein Immunoblotting for CYP2E1. Protein immunoblotting of liver and lymphoblastoid microsomes for CYP2E1 revealed a single band of about 51 kDa (Fig. 2). This is similar to the apparent molecular mass of this P450 identified in liver microsomes in previous studies (Lee et al., 1998). Densitometric analysis of the CYP2E1 standards demonstrated linearity at the protein concentrations used (0.05, 0.01, 0.25, and 0.50 pmol). Densitometric analysis of the microsomal protein bands revealed that the amounts of CYP2E1 protein in liver microsomal samples from A/J mice were about 40% higher than in those from CD-1 and B6 mice (Fig. 2). No difference in CYP2E1 protein content between CD-1 and B6 mice was detected.

View larger version (14K): [in this window] [in a new window]   Fig. 2.   Protein immunoblotting for CYP2E1 in liver microsomes from A/J, CD-1, and B6 mice. Liver microsomal proteins were separated by SDS-polyacrylamide gel electrophoresis, transferred to a nitrocellulose membrane, and reacted with a CYP2E1 mAb (1-98-1) (top). Lanes were loaded with liver microsomal proteins (10 µg) from the three strains of untreated mice as follows: lane 2, A/J; lane 3, CD-1; and lane 4, B6. Lane 1 contained molecular weight standards. Lanes 5 to 8 were loaded with rat CYP2E1-expressed human B-lymphoblastoid microsomes containing 0.05, 0.01, 0.25, and 0.50 pmol of P450, respectively. Protein bands were subjected to densitometry, and the relative amounts of CYP2E1 protein were determined by reference to the standards (bottom). *p < 0.05 compared with levels in CD-1 and B6 mice (n = 3).

Covalent Binding. Covalent binding of [¹⁴C]DCE to liver proteins was detected in all strains of mice. Binding levels in A/J mice were about 30 and 60% higher than in CD-1 and B6 mice, respectively. Hence, magnitudes of binding in the three murine strains in decreasing order were the following: A/J > CD-1 > B6 (Fig. 3).

View larger version (15K): [in this window] [in a new window]   Fig. 3.   Covalent binding of [14C]DCE to liver proteins in A/J, CD-1, and B6 mice. Levels of covalent binding was determined in liver homogenates 1 h after treatment of mice with [14C]DCE (20 µCi, 125 mg/kg i.p.). Values are mean ± S.D. (n = 6 mice from each strain). *p < 0.001 compared with levels in CD-1 and B6 mice; **p < 0.001 compared with levels in B6 mice.

Formation of DCE Metabolites. In vivo formation of DCE metabolites was determined by HPLC analyses of liver cytosolic and bile samples isolated from mice treated 1 h previously with DCE. The DCE epoxide-derived conjugate [C] was readily detected in cytosolic samples from all three strains of mice. Representative radiochromatograms are shown in Fig. 4. Other DCE metabolites, including conjugate [B] and the acetal of 2,2-dichloroacetaldehyde identified in previous in vitro studies (Dowsley et al., 1995), were not detectable in the liver cytosolic samples analyzed in these in vivo experiments. Levels of [C] were significantly higher in A/J mice than in either CD-1 or B6 mice (Fig. 5). Although formation of [C] was slightly higher in CD-1 than in B6 mice, this difference was not statistically significant. HPLC analysis of bile samples isolated from the gallbladders of DCE-treated mice showed that [C] was present in amounts that were readily detectable (Fig. 5). Levels of [C] detected in bile samples from A/J mice were significantly higher than in CD-1 and B6 mice. However, a statistically significant difference between CD-1 and B6 mice was not found. Regression analysis of the relationship between amounts of [C] detected in the cytosol and bile demonstrated a highly positive correlation (r² = 0.954) between these two parameters in the three strains of mice studied.

View larger version (10K): [in this window] [in a new window]   Fig. 4.   Representative radiochromatograms of DCE epoxide detected in liver cytosol isolated from A/J (a), CD-1 (b), and B6 (c) mice. Mice were treated with [14C]DCE (40 µCi, 125 mg/kg i.p.), and cytosolic fractions were prepared 1 h later. Cytosolic proteins were precipitated with 70% PCA and removed by centrifugation. An aliquot of the supernatant (100 µl) was subjected to reversed phase HPLC analysis. Fractions (250 µl) of the column effluent were collected, and levels of radioactivity were determined. [C], 2-S-glutathionyl acetate.

View larger version (19K): [in this window] [in a new window]   Fig. 5.   Levels of DCE epoxide in liver cytosol and bile in A/J, CD-1, and B6 mice. Mice (n = 4 of each strain) were treated with [14C]DCE (40 µCi, 125 mg/kg i.p.) and were sacrificed at 1 and 4 h for preparation of liver cytosol and collection of bile from gallbladders, respectively. Aliquots (10 µl) of the bile were diluted to 100 µl with mobile phase (0.2% H3PO4), and subjected to HPLC analysis for identification of DCE metabolites and/or GSH conjugates. Data are expressed as mean ± S.D. [C], 2-S-glutathionyl acetate. ap < 0.001 compared with levels in CD-1 and B6 mice; bp < 0.05 compared with levels in CD-1 and B6 mice; cp < 0.001 compared with levels in B6 mice; dp < 0.05 compared with levels in liver cytosol.

Glutathione Content. An HPLC method was used to estimate GSH levels in the livers of the mice. The experiments yielded data showing that GSH levels were similar in all the strains of untreated mice (Fig. 6). Treatment with DCE produced significant reduction in GSH content in livers of all the mice, with magnitudes of diminution that were about 30 to 40% of controls in all three strains.

View larger version (18K): [in this window] [in a new window]   Fig. 6.   Levels of hepatic GSH in control and DCE-treated A/J, CD-1, and B6 mice. Measurements were determined 1 h after treatment of mice (n = 5) with DCE (125 mg/kg i.p.). Data are expressed as mean ± S.D. *p < 0.01 compared with levels in controls.

Immunohistochemical Detection of DCE Protein Adducts. Application of immunohistochemical procedures, using an antibody directed to DCE epoxide-cysteine protein adducts, confirmed the preferential localization of the DCE adducts within centrilobular hepatocytes of DCE-treated mice, a finding also reported in previous studies (Forkert, 1999a). Observations of liver sections from the three strains of DCE-treated mice revealed apparent differences in the amounts of staining obtained. Staining was substantial in the livers of A/J mice, was less pronounced in livers from CD-1 mice, and was lowest in B6 mice (Fig. 7). There was no visible staining in liver sections from untreated mice. This negative staining was also obtained in liver sections in which reactions were performed in the absence of the specific antibody.

View larger version (122K): [in this window] [in a new window]   Fig. 7.   Localization and distribution of DCE protein adducts in the livers of A/J (a and d), CD-1 (b), and B6 (c) mice. Mice were treated with the vehicle or with DCE (125 mg/kg i.p.) and were sacrificed 1 h later. Immunohistochemical staining was performed on paraffin-embedded tissue sections, using the avidin-biotin complex procedure and a polyclonal antibody that recognizes epoxide-cysteine adducts. Staining is localized preferentially in centrilobular hepatocytes (arrows) but is absent in tissue sections that were reacted in the absence of the specific antibody (d). Scale bar, 100 µm.

Histopathology. Histopathological evaluation was performed in liver tissue 24 h after treatment with 50, 75, 125, or 175 mg/kg DCE. Hepatocellular damage was not observed in all strains of mice treated with 50 or 75 mg/kg DCE, compared with liver structure of the controls. However, treatment with a dose of 125 mg/kg elicited hemorrhagic congestion and mild centrilobular necrosis in A/J mice (Fig. 7). Increase of the dose to 175 mg/kg produced severe hemorrhagic congestion and centrilobular necrosis in this strain. In CD-1 mice, treatment with 125 or 175 mg/kg DCE evoked hemorrhagic congestion and mild necrosis in some CD-1 mice, whereas others exhibited relatively normal liver structure. In contrast, no obvious liver damage was evident in C57BL/6 mice exposed to the two highest DCE doses. Hence, the three strains of mice treated with DCE sustained hepatocellular injury in decreasing order according to the following: A/J > CD-1 > C57BL/6.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

We have reported recently that CYP2E1 levels varied in the lungs of A/J, CD-1, and B6 mice, and the polymorphic expression of this P450 was reflected in differences in the rates of metabolism of CYP2E1-selective substrates, including DCE (Forkert et al., 2001) and vinyl carbamate (Titis and Forkert, 2001). Maximal levels of the CYP2E1 enzyme were present in the lungs of A/J mice, whereas intermediate levels were found in CD-1 mice, with the lowest levels being manifested in B6 mice (Forkert et al., 2001). The variability in CYP2E1 expression in the lungs of these murine strains correlated with their relative rates of DCE metabolism as well as with the severities of lung cytotoxicity. In the case of the carcinogen vinyl carbamate, the levels of lung CYP2E1 in the three strains of mice correlated with magnitudes of formation of the DNA adducts 1,N⁶-ethenodeoxyadenosine and 3,N⁴-ethenodeoxycytidine (Titis and Forkert, 2001). Strain A/J mice with high lung CYP2E1 levels generated significantly more DNA adducts than did B6 mice with lower CYP2E1 levels and were more vulnerable to lung tumor formation than were B6 mice (Malkinson, 1991; Titis and Forkert, 2001). In view of these data showing that susceptibility to lung cytotoxicity and/or carcinogenicity is associated with the relative capacities for bioactivation, we have undertaken studies here to determine whether CYP2E1 expression is polymorphic in the livers of the three strains of mice and whether similar metabolic properties as found in the lung are manifested in the liver.

In this investigation, we have obtained data showing that constitutive expression of the CYP2E1 protein is variable in the three strains of mice examined, with amounts that were higher in the livers of A/J than in B6 and CD-1 mice (Fig. 2). These varying CYP2E1 levels coincided with quantities of covalent binding of DCE to liver proteins, such that binding levels in the three strains of mice were in the following descending order: A/J > CD-1 > B6 (Fig. 3). These findings suggested that CYP2E1 levels and, hence, the capacities for DCE bioactivation are associated with binding levels, which represents indirectly an index of reactive metabolite formation. In this regard, our previous studies have shown that increases in binding correlated with corresponding losses in GSH, indicating that magnitudes of DCE binding were dependent, in part, on the amounts of GSH available for conjugation reactions and hence detoxification (Forkert and Moussa, 1991; Moussa and Forkert, 1992). In this study, the levels of GSH detected in the three murine strains were not different from one another (Fig. 6). Moreover, treatment with DCE produced reduction in GSH levels that were comparable in the three strains. Hence, the differing levels of binding found in the three strains of mice were not due to differing rates of conjugation of the DCE epoxide with GSH, but are likely due to differences in magnitudes of bioactivation. This assertion is confirmed by our data showing that formation of the DCE epoxide, as estimated by levels of conjugate [C], was significantly higher in both liver cytosol and bile from A/J mice than in either CD-1 or B6 mice (Fig. 5). Furthermore, the relative amounts of [C] detected in the cytosolic fractions from the three murine strains correlated with levels detected in the bile (Fig. 5). However, levels of [C] detected in bile samples from A/J and CD-1 mice were higher than in the cytosol, but this difference was not apparent in B6 mice (Fig. 5), suggesting that epoxide formation was less pronounced in the latter strain. However, other metabolic processes such as differences in the rate of transport from hepatocytes to bile canaliculi may account for the differing levels of [C] identified in the bile of the various strains of mice. Nevertheless, our results suggested that the epoxide, when formed at sufficient levels, conjugated readily with GSH, and was transported efficiently from the biliary tract of the liver into the gallbladder. These data indicated that bioactivation capacities by CYP2E1 differed in the three strains of mice and were significantly higher in A/J than in either CD-1 or B6 mice. Interestingly, our previous studies showed that formation of the epoxide from DCE in human liver microsomal incubations correlated directly with the amounts of CYP2E1 expressed in the livers of individual subjects (Dowsley et al., 1999). Also of interest is the identification of genetic polymorphisms of human CYP2E1 that may cause up to a 10-fold variation in its transcriptional regulation (Hayashi et al., 1991). Moreover, induction of CYP2E1 in human liver as a result of ethanol exposure has been reported (Takahashi et al., 1993), thus supporting the assertion that chronic ethanol consumption leads to enhanced activation of CYP2E1-selective substrates. Hence, the extent of CYP2E1 expression is an important determinant of the potential of a tissue to sustain cytotoxicity from agents such as DCE that are bioactivated by this P450. Taken together, the variability in CYP2E1 levels observed in the strains of mice examined in this study suggested that they may be useful models for investigating metabolism of CYP2E1 substrates.

Centrilobular hepatocytes are preferential targets of DCE-induced toxicity and are also the sites in which CYP2E1 is located (Forkert et al., 1991). These findings led us to postulate that the DCE epoxide is formed within the same cellular sites. We have also postulated that it is at the centrilobular sites that protein adducts are formed with the DCE epoxide. To address these postulates, we have used an immunohistochemical approach to determine the intralobular locations where the DCE protein adducts are produced, and to identify potential differences in adduct formation in the livers of A/J, CD-1, and B6 mice treated with DCE. Using an antibody that recognizes the DCE epoxide-cysteine protein adducts, results from our immunohistochemical studies showed that the adducts were localized preferentially in the centrilobular region of the liver lobule (Fig. 7). Our results also showed that immunoreactivity for the protein adducts was variable in the three strains of mice, with the highest amounts being exhibited in the livers of A/J mice (Fig. 7a). The staining in CD-1 mice was lower than that in A/J mice (Fig. 7b), whereas staining in liver sections from B6 mice was barely visible (Fig. 7c). The differing amounts of staining in the livers of the three strains of mice indicated that formation of the DCE protein adducts was higher in A/J mice, compared with those produced in either CD-1 and B6 mice. The staining data are in agreement with both the relative quantities of the DCE epoxide generated in the murine strains (Fig. 5) and with the amounts of covalent binding to liver proteins (Fig. 3). These findings are consistent with the premise that bioactivation of DCE occurs to a greater extent in A/J than in CD-1 or B6 mice, and importantly that this metabolic event occurs primarily within the centrilobular hepatocytes.

Previous studies have shown that covalent binding of DCE to liver proteins was incremental with augmented doses, and magnitudes of DCE binding correlated with severities of hepatotoxicity (Forkert and Moussa, 1991). Here, we have shown in three different strains of mice that levels of DCE binding coincided with the extents of formation of the epoxide and protein adducts, as assessed by levels of the DCE epoxide-derived GSH conjugate found in liver cytosol and bile (Fig. 5) as well as by the results of our immunohistochemical staining experiments (Fig. 7). These parameters were all augmented in A/J mice, compared with those in CD-1 or A/J mice (Figs. 3-5 and 7). These findings correlated with the hepatotoxic effects induced by DCE, which were manifested to a greater extent in A/J than in CD-1 and B6 mice. Centrilobular necrosis and hemorrhagic congestion were evident after treatment of A/J mice with a single dose of DCE (175 mg/kg) (Fig. 8). However, the centrilobular damage produced in CD-1 and B6 mice treated with the same DCE dose was mild or negligible. In conclusion, the findings from these studies indicated that the strains of mice used here are useful models for investigating the metabolism of toxicants that are CYP2E1-selective substrates and for determining the outcomes of exposures to these agents.

View larger version (78K): [in this window] [in a new window]   Fig. 8.   Liver sections from DCE-treated A/J (a) and B6 (b) mice. Mice were treated with DCE (175 mg/kg i.p.) and were sacrificed 24 h later. Livers were fixed by vascular perfusion with 4% paraformaldehyde in 0.1 M Sorensen's phosphate buffer, pH 7.4, and embedded in paraffin. Tissue sections were stained with hematoxylin and eosin. Hemorrhagic congestion (arrowhead) and necrosis of centrilobular hepatocytes (arrow) are manifested after DCE treatment. Scale bar, 50 µm.