Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

Hepatotoxicity is induced by exposure to 1,1-dichloroethylene (DCE), a chemical that is used extensively in industry as a monomeric intermediate in the manufacture of plastics, including flexible plastic films for food packaging (Reynolds et al., 1975; Forkert et al., 1986). Previous studies in mice have shown that the mechanism of DCE-induced hepatocellular damage involves cytochrome P-450 (CYP)-catalyzed metabolism of DCE to reactive intermediates that bind covalently to liver macromolecules (Okine and Gram, 1986; Forkert et al., 1987). The metabolites generated from DCE in rat liver microsomal incubations have been identified as 2,2-dichloroacetaldehyde, DCE-epoxide, and 2-chloroacetyl chloride (Liebler and Guengerich, 1983; Costa and Ivanetich, 1984; Liebler et al., 1985, 1988). The scheme of the proposed pathway of DCE metabolism is illustrated in Fig. 1. In our recent studies in mice, the major metabolic products formed in liver microsomal incubations containing GSH were the conjugates, 2-S-glutathionyl acetyl glutathione ([B]) and 2-S-glutathionyl acetate ([C]) (Dowsley et al., 1995). These are believed to arise from conjugation of GSH with the DCE-epoxide. The acetal of 2,2-dichloroacetaldehyde was also detected in our microsomal incubations, but S-(2,2-dichloro-1-hydroxy)ethyl glutathione ([A]), the GSH-conjugated product of 2,2-dichloroacetaldehyde, was not detectable. S-(2-Chloroacetyl)glutathione ([D]) and chloroacetic acid, the GSH-conjugated and hydrolysis products of 2-chloroacetyl chloride, respectively, were detected in our liver microsomal incubations but at minimal levels (Dowsley et al., 1995). Hence, neither 2,2-dichloroacetaldehyde nor 2-chloroacetyl chloride is likely to contribute to the depletion of GSH observed in vivo, a metabolic event that is associated with DCE-induced hepatotoxicity (Forkert and Moussa, 1991, 1993). The DCE-epoxide is the major metabolite formed in vitro and is conjugated efficiently with GSH, indicating that it is responsible for the in vivo depletion of GSH observed after DCE treatment (Forkert and Moussa, 1991, 1993). These findings strongly suggested that the epoxide is the most plausible candidate for mediating the hepatotoxic effects of DCE.

View larger version (12K): [in this window] [in a new window]   Fig. 1.   Proposed scheme of the pathway of DCE metabolism. [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.

Our previous studies have confirmed that CYP2E1 is a major P-450 isozyme catalyzing the metabolism of DCE to reactive intermediates including the epoxide in murine liver and lung (Lee and Forkert, 1994, 1995; Dowsley et al., 1995, 1996). In the lung, the quantities of the epoxide formed, as estimated from the levels of GSH conjugates produced, correlated with CYP2E1 levels and the extent of cellular damage (Forkert et al., 1996a,b). For example, levels of both CYP2E1 and the DCE-epoxide-derived conjugates [B] and [C] formed in vitro were significantly higher in the lungs of female than male mice. These metabolic parameters are linked to the enhanced severity of bronchiolar Clara cell injury manifested in vivo in the lungs of female versus male mice (Forkert et al., 1996a,b). Moreover, the site of damage coincided with the cellular location of CYP2E1 in the Clara cells (Forkert, 1995). More recently, we have shown that conjugates [B] and [C] were identified in vivo in cytosolic fractions isolated from the lungs of mice treated with DCE (Forkert, 1999). Importantly, conjugate [C], which is the major conjugate detected, was localized immunohistochemically within the target cells. These studies have provided evidence to support the premise that the epoxide is formed and conjugated in the target Clara cells. Interestingly, the metabolic disposition of DCE in the liver exhibits features similar to those in the lung. For example, CYP2E1 catalyzes the in vitro formation of the DCE-epoxide-derived GSH conjugates [B] and [C] in the liver (Lee and Forkert, 1994). In addition, the localization of CYP2E1 within the centrilobular areas of the hepatic lobule coincides with DCE-induced cytotoxicity affecting the centrilobular hepatocytes (Forkert et al., 1986, 1991). On the basis of these findings, we predicted that, when formed, the DCE-epoxide occupies a centrilobular location.

This investigation was undertaken to address two primary objectives. The first objective was to establish that the DCE metabolites are formed in vivo in murine liver; this formation is determined by measuring the formation of the acetal, the hydrate of 2,2-dichloroacetaldehyde, and the epoxide-derived GSH conjugates [B] and [C] in liver cytosol and bile. These metabolites are detected under control conditions and after pretreatment with diallyl sulfone (DASO₂) and buthionine sulfoximine (BSO) to inhibit CYP2E1 and sulfhydryl levels, respectively. We have also investigated covalent binding of DCE in liver homogenates to estimate the amounts of DCE metabolites that are bound to liver macromolecules. The second objective was to determine the localization of the epoxide within the hepatic lobule. This was achieved by using immunohistochemical studies and a polyclonal antibody that recognizes conjugate [C] and proteins containing cysteinyl-SH groups (Forkert et al., 1997; Forkert, 1999).

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Chemicals and Reagents. Chemicals were obtained from suppliers as follows: DCE (>99% purity), phosphoric acid (85%, v/v), and GSH, Aldrich Chemical Co. (Montreal, Quebec, Canada); L-cysteine, glucose 6-phosphate, and glucose 6-phosphate dehydrogenase, Sigma Chemical Co. (St. Louis, MO); NADP⁺, BDH Chemical Co. (Toronto, Ontario, Canada); DASO₂, Parish Chemical Co. (Orem, UT). [¹⁴C]DCE (99% pure by gas liquid chromatography, specific activity 11.3 nCi/nmol) was obtained from Amersham Corp. (Arlington Heights, IL) and was diluted to 300 µCi/ml for our experiments. The DCE-epoxide-derived GSH conjugates [B] and [C] were synthesized as described previously (Dowsley et al., 1995) and were used as standards for metabolite identification. Other chemicals and reagents were purchased from standard suppliers.

Animal Treatment. Female CD-1 mice, weighing 20 to 25 g, were obtained from Charles River Canada (St. Constant, Quebec, Canada). The mice were kept in an animal facility maintained on a 12-h light/dark cycle, and were freely provided with food (Purina Rodent Chow) and water. The mice were housed for 7 days after arrival to acclimatize to laboratory conditions before being assigned to an experimental group. For in vivo detection of conjugates [B] and [C] in dose-response experiments, mice were treated with DCE (25, 75, 125, 175, and 225 mg/kg, 40 µCi [¹⁴C]DCE i.p.). The GSH conjugates were also determined in mice that were pretreated with BSO or DASO₂. Mice were treated with BSO (1.5 g/kg i.p.) that was dissolved in saline with the aid of 1 N NaOH, pH 8.5, and 4 h later were treated with DCE (125 mg/kg). This BSO dose has been used in previous studies (Drew and Miners, 1984). Mice were treated with DASO₂ (100 mg/kg p.o.) in water and then were treated with DCE (225 mg/kg, 40 µCi [¹⁴C]DCE) 4 h later. All the mice were sacrificed at 1 h after DCE treatment. The DCE metabolites were also determined in bile samples collected from mice treated with DCE (225 mg/kg, 40 µCi [¹⁴C]DCE). Metabolites in bile were also determined in mice pretreated with DASO₂ (100 mkg/kg p.o.); bile was collected from mice 4 h after DCE treatment. For the immunohistochemical studies, mice were treated with DCE (125, 275, and 225 mg/kg i.p.) alone or in conjunction with DASO₂ or BSO as described above, and were sacrificed at 1 h after DCE treatment. Control mice were treated with equivalent volumes of the appropriate vehicle and were sacrificed at time points corresponding to those in the experimental groups.

Isolation of Liver Cytosol. Livers of mice were excised, rinsed, blotted, and weighed. For isolation of cytosolic fractions, liver tissue from two mice that were treated with [¹⁴C]DCE were pooled and homogenized (1 ml of buffer per g) in cold phosphate-buffered KCl (140 mM KCl, 100 mM K₂HPO₄, 1.5 mM EDTA, pH 7.4). The homogenate was centrifuged at 9000g at 4°C for 40 min, after which the supernatant was obtained and centrifuged at 105,000g at 4°C for 40 min. Aliquots of the supernatant were dispensed into Eppendorf tubes, rapidly frozen in liquid nitrogen, and stored at 70°C. Protein concentrations were determined by the Bradford protein assay (1976).

Measurement of Sulfhydryl Content. Livers were homogenized in 0.1 M phosphate buffer, pH 7.4, and proteins in a 500-µl aliquot of homogenate were precipitated by addition of an equal volume of 4% sulfosalicylic acid. After thorough mixing and centrifugation, 500 µl of the supernatants were analyzed for concentration of acid-soluble sulfhydryls according to the method of Ellman (1959). Protein concentrations were determined by the method of Bradford (1976).

Synthesis and Purification of Conjugate [C]. Conjugate [C] was synthesized and purified according to methods described in our previous studies (Forkert et al., 1997), with slight modifications. Briefly, [C] was synthesized by combining GSH (307 mg) with chloroacetic acid (94 mg) in 10 ml of 100 mM potassium phosphate buffer, pH 7.4, yielding a final equimolar concentration of 100 mM. The solution was heated at 50°C for 2 h and then cooled overnight at room temperature. The solution was adjusted to pH 2.0 with 6 N HCl, and then extracted with diethyl ether (10 ml). Conjugate [C] was isolated by passage through a C-18 extraction cartridge and elution with methanol. The samples were dried down, reconstituted with water, and analyzed by HPLC with a reversed-phase C-18 column (5 µm, 4.6 × 250 mm; Phenomenex, Torrance, CA). The mobile phase was 0.1% trifluoroacetic acid in H₂O, and the flow rate was 1 ml/min. The eluent corresponding to the peak for [C] was collected from approximately 40 HPLC injections, pooled, concentrated in vacuo, and stored at 20°C. The identity and purity of the sample were confirmed by analysis by ¹H-NMR and electrospray mass spectroscopy, as described in our previous studies (Dowsley et al., 1995).

Metabolite Identification. Metabolites of DCE were analyzed and identified by using procedures described in our previous studies (Dowsley et al., 1995, 1996), with modification. Synthesized standards of the metabolites were identified by reversed-phase HPLC analysis with a C-18 column (5 µm, 4.6 × 250 mm, Microsorb-MV; Rainin Instruments Co., Inc., Woburn, MA). The isocratic mobile phase was 0.2% H₃PO₄ at a flow rate of 1.0 ml/min. The column effluent was monitored at 200 nm. Proteins in samples of cytosol (30 mg/ml) were precipitated with 70% perchloric acid and centrifuged for about 5 min; 100 µl of the supernatant was subjected to HPLC analysis. For detection of the DCE metabolites, 250-µl aliquots of the column effluent were collected every 15 s for each sample, and levels of radioactivity were determined by liquid scintillation spectroscopy. Levels of metabolites were estimated by summing the amounts of radioactivity associated with each peak and converting the data to nanomolar amounts, by using the specific activity of the [¹⁴C]DCE. The HPLC experiments were performed on a Beckman System Gold Programmable Solvent Module 126 HPLC with a Beckman System Gold Module 168 UV detector. UV spectra for all assays were determined with a Hewlett Packard model 8452 diode array UV spectrophotometer.

Covalent Binding of [¹⁴C]DCE. Covalent binding of [¹⁴C]DCE was measured in liver homogenates by using the technique of equilibrium dialysis as described previously (Moussa and Forkert, 1991). Briefly, livers were homogenized in 5 volumes of 0.01 M sodium phosphate buffer, pH 7.0, containing 2% SDS. An additional 5 volumes of 4% SDS in water were then added to the sample. The homogenates were then boiled for 15 min, after which they were cooled at room temperature. Aliquots (1.0 ml) of the boiled homogenate samples were dispensed into dialysis sacs (Spectrapor-3 dialysis tubing, 3500 molecular weight cutoff), and dialyzed against 500 ml of 0.01 M sodium phosphate buffer, pH 7.0, containing 0.1% SDS. After dialysis to equilibrium, aliquots (250 µl) were solubilized overnight in 2 ml of Solvable with gentle agitation. After addition of glacial acetic acid (300 µl) and aqueous scintillation fluid (Ecolite, 15 ml; ICN Biomedicals, Inc., Costa Mesa, CA) to each sample, radioactivity levels in the dialysate were determined. Levels of radioactivity in the buffer (250 µl) were also measured. The quantity of covalently bound DCE was estimated by calculating the difference between the amounts of radioactivity in the dialysate and the buffer.

Identification of DCE Metabolites in Bile. Bile samples were collected with a syringe from mice that had been treated with DCE (225 mg/kg). Samples from 10 mice were pooled and yielded a volume of 90 µl of bile. Proteins in aliquots (50 µl) of the bile were precipitated with 70% perchloric acid and centrifugation. The supernatant samples were frozen in liquid nitrogen and stored at 70°C until analyzed. The supernatants (5 µl) were diluted with 95 µl of mobile phase (0.2% H₃PO₄) and were then subjected to HPLC analysis for identification of DCE metabolites as described previously.

Immunohistochemical Detection of Conjugate [C]. Mice were anesthetized with sodium pentobarbital (120 mg/kg i.p.). The abdomen was exposed, a cannula was inserted into the pulmonary artery through the right ventricle, and the liver was flushed with saline to clear the tissue of blood. When the perfusate became clear and the liver was pale, perfusion was continued with 4% paraformaldehyde and 0.2% glutaraldehyde in 0.2 M sodium cacodylate, pH 7.3. Paraffin-embedded sections were subjected to immunohistochemical staining for conjugate [C] by application of the avidin-biotin complex technique. Liver sections were deparaffinized, cleared, hydrated, and rinsed in PBS. The tissue sections were incubated in 5% normal goat serum and 3% BSA. After rinsing in PBS, the sections were reacted for 60 min with an affinity-purified antibody for [C] (Forkert et al., 1997). The antibody was diluted in PBS containing 1% normal goat serum and 0.1% of the conjugate, glycine-glutaraldehyde-BSA, that was synthesized as described in our previous studies (Forkert et al., 1997), and has been shown to inhibit nonspecific antibody binding. The sections were rinsed to remove unbound antibodies and reacted for 10 min with a biotinylated goat anti-rabbit antibody. Adjacent liver sections were stained with a goat anti-rabbit liver microsomal CYP2E1 antibody and, subsequently, with a biotinylated horse anti-goat antibody. Endogenous peroxidase activity was blocked by incubating tissue sections for 30 min with 1% hydrogen peroxide in water. Sections were then reacted for 10 min with streptavidin conjugated to horseradish peroxidase, and the immunoperoxidase color reaction developed by incubation in PBS containing 0.05% 3,3'-diaminobenzidine and 0.01% hydrogen peroxide. The sections were then rinsed in running tap water for 5 min. After incubation for 5 min in 0.15 M sodium chloride containing 0.5% copper sulfate, the sections were dehydrated, cleared, and mounted. Controls for the specificity of the immunohistochemical reactions included incubations performed in the presence of a nonspecific antibody or incubations in the absence of the specific antibody.

Statistical Analysis. Data are expressed as mean ± S.D. Statistical analysis was performed by one-way ANOVA followed by the Student-Newman-Keuls test for pairwise multiple comparisons to identify significant differences between treatment groups. The level of significance was set at P < .05.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Identification of DCE Metabolites in Liver Cytosol. The formation of DCE metabolites was determined in cytosol isolated from the livers of mice treated in vivo with [¹⁴C]DCE. HPLC analysis showed that the DCE-epoxide-derived conjugates [B] and [C] were detectable in liver cytosol of mice treated in vivo with DCE (Fig. 2). However, the major product generated was [C], whereas [B] was detected at minimal levels. The acetal, the hydrate of 2,2-dichloroacetaldehyde, was not detectable in the cytosolic samples. However, formaldehyde and glycolic acid, the hydrolysis products of the DCE-epoxide, were detected and eluted between 2 and 4 min. Preliminary time course experiments showed that [C] was detected in the cytosol at 15 min (4.6 nmol/mg of protein) after DCE (125 mg/kg) treatment and peaked at 0.5 to 1.0 h (8.7 nmol/mg of protein). A decline in the level of [C] was observed at 2 h, and at 4 h the amount detected was minimal (0.3 nmol/mg of protein). Formation of [B] was detected at all the time points examined, but the quantities were extremely low compared with those identified for [C]. The results of dose-response experiments are summarized in Fig. 3 and showed that formation of [C] was concentration-dependent; the amounts detected were proportional to the concentrations of DCE administered (25-225 mg/kg). A similar dose-dependent profile was observed for [B], but the increases were minimal with the higher concentrations of DCE used. Hence, there was a dose-dependent response in the total levels of the epoxide-derived GSH conjugates formed. The regression line for the total amount of the epoxide-derived GSH conjugates formed overlapped that for [C], and this was due to the major contribution of [C] to the total pool (Fig. 3).

View larger version (11K): [in this window] [in a new window]   Fig. 2.   Chromatograms of DCE metabolites detected in liver cytosol. Mice were treated with DCE (40 µCi, 125 mg/kg) alone (A) or were pretreated with BSO (B) or DASO2 (C) before DCE treatment. Cytosolic fractions were isolated from liver tissue 1 h after DCE treatment. Proteins were precipitated with 70% perchloric acid, and 100 µl of the cytosol was subjected to reversed-phase HPLC analysis. Aliquots (250 µl) of the column effluent were collected, and levels of radioactivity were determined. [B], 2-(S-glutathionyl) acetyl glutathione; [C], 2-S-glutathionyl acetate.

View larger version (21K): [in this window] [in a new window]   Fig. 3.   Regression analysis of the relationship between the DCE dose administered to mice and the formation of conjugates [B] and [C] in liver cytosol. Cytosolic proteins were precipitated, and 100 µl of the supernatant was subjected to reversed-phase HPLC analysis. Aliquots of the column effluent were collected, and levels of radioactivity were determined. Data are expressed as mean ± S.D. Conjugate [B]: R2 = 0.91; conjugate [C]: R2 = 0.91; conjugates [B] + [C]: R2 = 0.93.

View larger version (28K): [in this window] [in a new window]   Fig. 4.   Measurements of conjugate [C] in cytosol and covalent binding of DCE to proteins in liver. Mice were treated with [14C]DCE (40 µCi, 125 mg/kg) or were pretreated with BSO or DASO2 before DCE treatment. Cytosolic fractions were prepared 1 h after DCE treatment. Cytosolic proteins were precipitated and samples of the supernatant were subjected to reversed-phase HPLC analysis. Aliquots of the column effluent were collected and levels of radioactivity were determined. Levels of [C] were estimated by summing the levels of radioactivity associated with the peak for [C] and converting the data to nanomolar amounts, by using the specific activity of the [14C]DCE. For measurement of covalent binding, mice were treated with [14C]DCE (40 µCi, 125 mg/kg), and liver homogenates were prepared 1 h after treatment. Binding to liver proteins was determined by using the method of equilibrium dialysis as described in Materials and Methods. aSignificantly different from levels of conjugate [C] in cytosol of mice treated with DCE. bSignificantly different from binding levels in mice treated with DCE. cSignificantly different from binding levels in mice treated with DCE and DCE/BSO.

Identification of DCE Metabolites in Bile. The acetal and conjugates [B] and [C] were all detected in bile collected from the gall bladders of mice treated with DCE (Fig. 5). As found in liver cytosol, [C] was the major conjugate found in bile, and was significantly higher (200%) than the level of [B] detected. The quantity of the DCE-epoxide formed, as estimated from the total level of [B] and [C], was significantly higher (850%) than the level of 2,2-dichloroacetaldehyde produced, as estimated from the level of the acetal (Fig. 6). All three metabolites were also detected in bile samples isolated from the gall bladders of mice pretreated with DASO₂ before treatment with DCE (Fig. 6). However, the amounts of [B] and [C] were significantly lower than those found in the bile of mice treated with DCE alone. As expected, DASO₂ pretreatment produced a significant reduction in the total level of [B] and [C] present, compared with the amount detected in mice treated only with DCE.

View larger version (14K): [in this window] [in a new window]   Fig. 5.   Chromatograms of DCE metabolites in bile isolated from gall bladders of mice treated with [14C]DCE (40 µCi, 225 mg/kg) (A) or pretreated with DASO2 (B) before treatment with DCE. Samples of bile were diluted with mobile phase (0.2% phosphoric acid, pH 2.0, 1:20) and analyzed by reversed-phase HPLC for identification of DCE metabolites. [B], 2-(S-glutathionyl) acetyl glutathione; [C], 2-S-glutathionyl acetate.

View larger version (34K): [in this window] [in a new window]   Fig. 6.   Levels of DCE metabolites in bile isolated from gall bladders of mice treated with [14C]DCE (40 µCi, 225 mg/kg) or pretreated with DASO2 before treatment with DCE. aSignficantly different from levels of the acetal within the same experimental group. bSignificantly different from levels of conjugate [B] within the same experimental group. cSignificantly different from levels of the same metabolite in mice treated with DCE alone.

Covalent Binding of [¹⁴C]DCE in Liver Homogenates. In vivo treatment of mice with [¹⁴C]DCE produced substantial levels of covalent binding to proteins in liver homogenates (Fig. 4). A significant increase in the amounts of binding were detected in liver homogenates prepared from mice that were pretreated with BSO, and this increased binding coincided with the reduction in sulfhydryl levels observed (Fig. 4). In contrast, pretreatment with DASO₂ resulted in significantly reduced levels of binding. Thus, binding levels were modified either by decreasing sulfhydryl levels or by inhibiting the extent to which DCE was metabolically activated.

Immunohistochemical Localization of Conjugate [C]. Immunohistochemical studies were performed on liver tissues from mice that were treated with DCE. These studies were also performed in liver tissues from mice that were treated with BSO or DASO₂ before treatment with DCE. The mice were fasted overnight to deplete the liver of glycogen. This maneuver was undertaken to prevent the washing out of areas of cytoplasm containing glycogen deposits, thus producing optimal morphological preservation for immunohistochemical staining. Our preliminary studies confirmed that the fasting regimen used in our experiments did not decrease the level of the GSH conjugates formed and it did not alter the level of CYP2E1-dependent catalytic activity.

View larger version (169K): [in this window] [in a new window]   Fig. 7.   Hepatic distribution of conjugate [C] in mice treated with DCE. Immunohistochemical staining was performed with a rabbit antibody directed to the chemically synthesized conjugate [C] and the avidin-biotin complex procedure. Immunoreactivity in representative liver sections from mice treated with 125 mg/kg DCE (a) or vehicle (b) ,or from mice pretreated with BSO (c) or DASO2 (d). cv, central vein; pv, periportal vein.

View larger version (94K): [in this window] [in a new window]   Fig. 8.   Hepatic distribution of conjugate [C] and CYP2E1 in adjacent liver sections from mice treated with DCE. Immunohistochemical staining was performed with antibodies specific for conjugate [C] or for liver microsomal CYP2E1 and the avidin-biotin complex procedure. Immunoreactivity in liver sections from mice treated with 225 mg/kg DCE (a) or from mice that were treated with DASO2 and subsequently treated with DCE (225 mg/kg) (b). cv, central vein; pv, portal vein.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

Previous studies demonstrated that the major metabolites generated from DCE in vitro in murine liver are the acetal, the hydrate of 2,2-dichloroacetaldehyde, and the DCE-epoxide-derived GSH conjugates [B] and [C] (Dowsley et al., 1995). An objective of this investigation was to determine whether the same DCE metabolites are also formed in vivo in the liver. Our results showed that the major metabolite identified in liver cytosol after treatment in vivo with DCE was the epoxide-derived conjugate [C] (Figs. 2 and 3). Conjugate [B] was detected, albeit at low levels, but the acetal, which was identified in vitro in previous studies (Dowsley et al., 1995), was not detectable in the cytosolic fractions. However, all three of the DCE metabolites found in the microsomal incubations in previous experiments were present in bile isolated from gall bladders of mice treated in vivo with DCE (Fig. 6). Conjugate [C] was found at the highest level, whereas considerably lower amounts of [B] and the acetal were detected (Figs. 5 and 6). Moreover, the metabolites identified in the bile paralleled those seen in the microsomal incubations. These findings indicated that the metabolites generated in vivo are identical with those produced in vitro. This conclusion is based on identification of the DCE metabolites present in the bile and the cytosol. The low levels of [B] and the nondetectability of acetal in the cytosol may be ascribed to the difference in the length of time after DCE treatment that the metabolites were analyzed. The metabolites in the cytosol were identified 1 h after treatment, a time point that coincided with peak metabolite production. In contrast, the metabolites in bile were analyzed 4 h after treatment to allow the bile products to accumulate. It should be noted that the amount of [C] in the cytosol detected at 4 h was extremely low (0.31 ± 0.07 nmol/mg of protein), whereas [B] and the acetal were not detectable at this time point. These results suggest that the metabolites were transported efficiently from the liver, and their accumulation in the bile facilitated the analysis of metabolic products formed in vivo. The findings further suggest that the presence of metabolites in the bile represented a sensitive indicator of metabolic events occurring in vivo. However, it should be noted that it is expected that GSH conjugates are excreted preferentially through the bile because of their high molecular weight (Klassen and Watkins, 1984).

We have also investigated the formation of DCE metabolites under conditions in which GSH and CYP2E1 were inhibited by pretreatment with BSO and DASO₂, respectively. Our results showed that formation of conjugate [C] from DCE in the cytosol was significantly inhibited by both pretreatment regimens (Figs. 2 and 4). The magnitudes of inhibition induced by BSO and DASO₂ were similar, indicating that significantly decreased production of [C] could be efficiently achieved by either lowering sulfhydryl levels or inhibiting DCE activation. As expected, decreased formation of [C] occurred in conjunction with significantly increased levels of covalent binding of DCE to liver proteins after BSO, compared with the amounts found in livers treated with DCE alone (Fig. 4). In contrast, binding levels were significantly decreased in the livers of mice pretreated with DASO₂ (Fig. 4), thus confirming that metabolic activation of DCE by CYP2E1 was inhibited. These findings indicated that quantitation of the level of GSH conjugates formed from DCE together with binding levels represented a good estimate of the production of the epoxide.

An additional objective of our studies was to determine the intralobular distribution of the DCE-epoxide. We have used in immunohistochemical studies a polyclonal antibody that was directed against conjugate [C] (Forkert et al., 1997). This antibody was developed by immunization with a hapten that was synthesized by reacting the DCE-epoxide with the -SH group of the cysteine of GSH. Based on this reaction, the presumption may be made that, in addition to its specificity for recognizing [C], our antibody could be used as a probe to identify tissue protein sites of binding of the electrophilic epoxide to cysteinyl thiols, which are major nucleophilic centers on proteins. Indeed, our recent studies demonstrated by protein immunoblotting that our antibody recognizes an antigen consisting of an epoxide-cysteine conjugate that is coupled to BSA (Forkert, 1999). These findings supported the view that this antibody detects conjugated products of the epoxide with GSH as well as the sites of binding of the epoxide to liver proteins and, presumably, to -SH groups. Hence, the location of immunohistochemical staining should correspond with the intracellular sites in which the epoxide resided. The results of our immunohistochemical experiments revealed staining that was proportional to the dose of DCE administered and was concentrated in the centrilobular areas of the hepatic lobules; periportal areas lacked staining at all DCE doses administered (Fig. 7, a and c; Fig. 8a). This staining pattern corresponded to the intralobular distribution of CYP2E1, where immunohistochemical staining for this P-450 was concentrated in the centrilobular hepatocytes and was absent in the periportal hepatocytes (Fig. 8b). Within the hepatocyte, staining was found in the cytoplasm, the intracellular sites of GSH, and the endoplasmic reticulum, where P-450 enzymes including CYP2E1 reside. These observations suggested that the DCE-epoxide binds to proteins at the site of formation, which is not a surprising finding due to the high reactivity of the epoxide (Dowsley et al., 1995). This is also the site where the hepatotoxic effects of DCE are manifested (Forkert et al., 1986), and it supported the premise that binding of the epoxide to liver proteins mediates the centrilobular necrosis elicited by DCE exposure.

Immunohistochemical studies were also performed in liver tissues from mice that were treated with BSO or DASO₂ before treatment with DCE (Fig. 7, c and d). Immunostaining was highly localized in the centrilobular hepatocytes, as was found in liver sections from mice that were treated with only DCE (Fig. 7a). The decreased formation of conjugate [C] and increased DCE binding that occurred in conjunction with BSO pretreatment yielded staining levels that were increased compared with those observed in the livers of mice treated with DCE alone (Figs. 7c and 8a). These findings supported the assumption that our antibody detects protein adducts, the formation of which are presumably increased in view of the enhanced DCE binding found after BSO pretreatment (Fig. 4). Although BSO pretreatment evoked a small increase in immunostaining, there was a marked decrease in staining seen in the livers of mice subjected to the DASO₂ pretreatment regimen (Fig. 7d). This decreased staining coincided with the inhibitory effects of DASO₂ on the magnitudes of formation of [C] and levels of DCE binding to liver proteins (Fig. 4). Taken together, our results indicated that the cellular and intracellular sites in which immunohistochemical staining is localized represented the locations in which the DCE-epoxide is found. Furthermore, the coincidence of the distribution of the epoxide and CYP2E1 in the centrilobular areas of the hepatic lobules suggests that the epoxide is formed and bound to liver proteins at the site of formation within the centrilobular hepatocytes. These centrilobular hepatocytes may also be the location of conjugation of the epoxide to GSH. This assertion is based on the presumption that inclusion of glutaraldehyde in the tissue fixative cross-links cellular proteins to an extent that the GSH conjugates formed from DCE are retained within the cytoplasmic matrix of hepatocytes in liver sections subjected to immunohistochemical procedures. The lack of immunostaining in the periportal regions that are deficient in CYP2E1 suggests that formation of the epoxide was not a metabolic event occurring in the periportal hepatocytes.

An immunochemical approach has also been used in previous studies and in particular with the analgesic drug acetaminophen, which causes hepatic centrilobular necrosis when administered in high doses to both humans and experimental animals (Hinson, 1980; Roberts et al., 1987). N-Acetyl-p-benzoquinone imine has been identified as the reactive species responsible for mediating the hepatoxicity (Miner and Kissinger, 1979; Dahlin et al., 1984; Harvison et al., 1988). By using a polyclonal antibody that was developed against 3-(cystein-S-yl) acetaminophen, a major protein adduct formed between N-acetyl-p-benzoquinone imine and cysteine sulfhydryl groups on proteins, immunohistochemical studies showed that the reactive metabolite and protein adducts were localized in the centrilobular zones of the hepatic lobules (Roberts et al., 1991). This preferential distribution in the centrilobular hepatocytes is consistent with the centrilobular location of the P-450 enzymes, CYP2E1 and CYP1A1, that are implicated in the oxidative metabolism of acetaminophen (Raucy et al., 1989). These findings suggested that, as in the case of DCE described herein, formation of both the reactive metabolite and protein adducts occurs within the target areas. This phenomenon is more clearly demonstrated in the lung where DCE induces cytotoxicity selectively in the Clara cell (Forkert and Reynolds, 1982). Application of a highly specific immunohistochemical method revealed that the DCE-epoxide is formed in situ in the target Clara cells (Forkert, 1999). Reducing the formation of the epoxide within the Clara cells by inhibiting lung CYP2E1, a P-450 that activates DCE, protected this cell type from damage (Forkert et al., 1996b). These findings underscore the usefulness of the immunochemical approach for elucidating the mechanisms mediating cytotoxic responses at the cellular level.

In summary, our results demonstrated that the DCE metabolites, 2,2-dichloroacetaldehyde and epoxide, identified initially in a liver microsomal incubation system, are also formed in vivo. They further demonstrated that the target centrilobular hepatocytes are the sites of formation and alkylation of proteins by the DCE-epoxide, and supported the identification of this electrophilic intermediate as the ultimate toxic species.