Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

Garlic is frequently added as a flavor-enhancing ingredient in food preparation, and is commonly regarded as an antidote in the practice of folk medicine in many cultures. A major constituent of garlic is S-allylcysteine sulfoxide (alliin), which is enzymatically converted by alliinase to allicin, an unstable component that can be further transformed to other garlic compounds, including diallyl sulfide (DAS). DAS also can be formed during cooking or after ingestion of garlic, and it is a component of garlic oil (Hayes et al., 1987). It has been estimated that ~30 to 100 µg of DAS is derived from 1 g of garlic (Sparnins et al., 1988).

The anticancer effects of garlic constituents such as DAS have been investigated in many studies. DAS has been reported to protect against carcinogenesis induced by chemicals, including benzo[a]pyrene (Sparnins et al., 1988), 4-(methylnitrosamino)-1-(3-pyridyl)-1-butanone (Hong et al., 1992), aflatoxin B₁ (Haber-Mignard et al., 1996), 7,12-dimethylbenz[a]anthracene (Wargovich and Goldberg, 1985; Hayes et al., 1987; Wargovich, 1987; Sumiyoshi and Wargovich, 1990), azoxymethane (Sohn et al., 1991), and several nitrosamines (Wargovich et al., 1988, 1992; Pereira 1995; Surh et al., 1995). The targets in which carcinogenicities are modified by DAS include tissues such as colon (Wargovich, 1987), esophagus (Wargovich et al., 1988), liver (Hayes et al., 1987), lung, and forestomach (Sparnins et al., 1988). Hence, DAS has inhibitory effects on carcinogenicities induced in a variety of tissues by a broad spectrum of chemicals.

It has been proposed that inhibition of metabolic activation may be linked to the protective activity of DAS against carcinogenicity induced by azoxymethane, 1,2-dimethylhydrazine, and N-nitrosodimethylamine (Yang et al., 1984). All are compounds that are metabolically activated by CYP2E1. On the basis of these findings, it was postulated that inhibition of CYP2E1-mediated activation of procarcinogens is a key event responsible for the anticarcinogenic activity of DAS (Brady et al., 1988; Wargovich et al., 1988; Kwak et al., 1994). Moreover, the absence of protection by DAS against N-methyl-N'-nitro-N-nitrosoguanidine, a compound that acts directly rather than through an intermediate, supported this concept (Wargovich and Goldberg, 1985). Enhanced detoxification also has been invoked as a mechanism for the anticarcinogenic activity of DAS. This proposal arose from findings showing that treatment with organosulfur compounds, including DAS, increased glutathione S-transferase activity and protected against benzo[a]pyrene neoplasia of the forestomach and lung (Sparnins et al., 1988). DAS treatment also has been reported to increase activity levels of glutathione peroxidase, glutathione reductase, UDP-glucuronosyltransferase, microsomal epoxide hydrolase, and glutathione S-transferases (Sumiyoshi and Wargovich, 1990; Maurya and Singh, 1992; Wargovich et al., 1992; Guyonnet et al., 1999). However, the dynamics of the factors contributing to the final outcome are poorly understood.

It has been reported that DAS is metabolized to diallyl sulfoxide (DASO) and subsequently to diallyl sulfone (DASO₂; Brady et al., 1991). Identification of this pathway for DAS metabolism was based on findings showing that both DASO and DASO₂ were detectable in extracts of liver, blood, and urine from rats treated with DAS (Brady et al., 1991). More recent studies confirmed that CYP2E1 catalyzes the oxidation of the sulfur atom of DAS to produce DASO and DASO₂ (Jin and Baillie, 1997). These garlic derivatives are all competitive inhibitors of CYP2E1. However, CYP2E1 inhibition by DASO₂ is more pronounced and is manifested more rapidly than under conditions in which either DAS or DASO₂ are used. The efficacy of DASO₂ as a CYP2E1 inhibitor has been proposed to be due to mechanism-based inactivation, and it is the final metabolic event involving DASO₂ that leads to CYP2E1 destruction and that is responsible for the chemoprotective effects of DAS (Jin and Baillie, 1997). However, the nature of the metabolite formed from DASO₂ that mediates CYP2E1 inactivation in a biological system has not been identified and characterized.

We hypothesized that the metabolite formed from oxidative metabolism of DASO₂ is an epoxide (1,2-epoxypropyl-3,3'-sulfonyl-1'-propene; DASO₃). To test this hypothesis, we adopted an approach in which the epoxide was trapped with GSH and identified as GSH conjugates. The formation of the epoxide-derived GSH conjugates was determined in hepatic microsomal incubations. Generation of the epoxide was confirmed by measurement of DASO₃ derivatized with 4-(p-nitrobenzyl) pyridine (NBP). The role of CYP2E1 in epoxide production was investigated by identifying the GSH conjugates in a CYP2E1-expressed lymphoblastoid system and by determining the effects of DASO₂ on the CYP2E1 enzyme. These studies were undertaken with the anticipation that the findings would provide data for identifying the mechanism by which DASO₂ inactivates CYP2E1, and hence conferring protection from CYP2E1-selective substrates that are converted to carcinogenic metabolites.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Chemicals and Reagents. Chemicals were purchased from suppliers as follows: glucose 6-phosphate, glucose-6-phosphate dehydrogenase, NADPH (Sigma Chemical Co., St. Louis, MO); NBP, GSH, phosphoric acid (85%), acetone, and ethyl acetate (Aldrich Chemical Co., Montreal, Quebec, Canada); HPLC grade acetonitrile and HPLC grade methanol (EM Science Inc., Gibbstown, NJ); diethyl ether (BDH Inc., Toronto, Ontario, Canada); sodium acetate and 4Å molecular sieve (Mallinckrodt Inc., Paris, KY); [³H]GSH (specific activity 43.8 Ci/mol; DuPont Canada, NEN Ltd., Mississauga, Ontario, Canada); and rat CYP2E1-expressed human B-lymphoblastoid microsomes (Gentest Corp., Woburn, MA). DASO₂ was synthesized as described in Serra et al. (1990). 3-(S-Allyl-S-dioxomercapto)-1,2-epoxypropane (DASO₂ monoallylepoxide or DASO₃) was synthesized from DASO₂ by oxidation with dimethyldioxirane, prepared as described in Murray and Singh (1996). 4-Ethyl-3,5-diethoxycarbonyl-1,4-dihydro-2,6-dimethylpyridine was donated by Dr. G. S. Marks (Department of Pharmacology and Toxicology, Queen's University, Kingston, Ontario, Canada). The CYP2E1 monoclonal antibody (mAb) 1-98-1 for immunoblotting and the inhibitory CYP2E1 mAb 1-91-3 (Ko et al., 1987) were donated by Dr. S. S. Park (Laboratory of Comparative Carcinogenesis, National Cancer Institute, Frederick, MD). All other chemicals used were purchased from standard commercial suppliers.

Animal Treatment. Female CD-1 mice of 24 to 28 g b.wt. were purchased from Charles River (St. Constant, Quebec, Canada). Mice were maintained on a 12-h light/dark cycle and were provided freely with food (Mouse Diet 5015; PMI Nutrition International, Inc., Brentwood, MO) and water. The mice were acclimatized to laboratory conditions for at least 7 days before being entered into an experiment. Mice were sacrificed by cervical dislocation for preparation of liver microsomes. To determine the effects of DASO₂ on CYP2E1-dependent p-nitrophenol (PNP) hydroxylase activity in vivo, mice were treated with DASO₂ (25-200 mg/kg p.o.), and sacrificed 2 h later for preparation of liver microsomes.

Preparation of Microsomes. Livers from five mice were pooled and homogenized in four volumes of cold phosphate-buffered KCl (1.15% KCl, 100 mM K₂PO₄, and 1.5 mM EDTA, pH 7.4). Microsomes were then prepared by differential centrifugation according to procedures described in Forkert (1995). Microsomal pellets were resuspended in 100 mM phosphate buffer (100 mM K₂PO₄ and 1.5 mM EDTA, pH 6.8) and stored at 70°C. Protein concentrations were determined by the method of Lowry et al. (1951) with BSA as the standard.

Microsomal Incubations. Liver microsomes were resuspended in 100 mM K₂PO₄, pH 7.4, (pH adjusted with 70% phosphoric acid) at a protein concentration of 3.0 mg/ml. Reaction mixtures in a total volume of 2 ml consisted of 100 mM K₂PO₄ buffer, 1.5 mM EDTA, 5.0 mM MgCl₂, 3 mg/ml microsomal protein, an NADPH-generating system (7.5 mM glucose 6-phosphate, 4 U glucose-6-phosphate dehydrogenase, and 0.4 mM NADP⁺), and [³H]GSH (0.1 µCi/ml; 0.1 mM). The reaction mixtures were preincubated for 3 min at 37°C in a shaking water bath. The reaction was initiated by DASO₂ (0.25-1 mM), and the incubations were conducted for 90 min in the concentration-response studies. In the time course studies, the duration of incubation ranged from 0.5 to 3.0 h. After completion of the incubations, proteins in the samples were precipitated with perchloric acid (70%) and centrifugation.

Synthesis of DASO₃-GSH Conjugate Standards. [³H]GSH (1.2 mM; 0.2 µCi/ml) was dissolved in H₂O (10 ml) and the pH of the solution was adjusted to 7.8 with 0.25 N NaOH. The DASO₃-GSH conjugates were synthesized by first dissolving an equimolar amount of DASO₃ in dry methanol (10 ml) and adding this solution immediately to the GSH solution. The DASO₃-GSH mixture was stirred at room temperature in the dark for 4 h. The resulting mixture was lyophilized in vacuo, redissolved in 2.0 ml of H₂O, and stored at 70°C. The products of this synthesis (100 µl) were analyzed with a reverse phase C18 column (5 µm, 4.6 × 250 mm; Phenomenex, Torrance, CA). The isocratic mobile phase was 5% aqueous methanol containing 0.06% trifluoroacetic acid (TFA) at a flow rate of 1.0 ml/min as described in Jin and Baillie (1997). The column effluent was monitored at 200 nm. To detect GSH-containing peaks, 0.25-ml aliquots of the column were collected and levels of radioactivity were determined by liquid scintillation spectroscopy. The relative size and position of the peaks were estimated from summation and transformation of radioactive counts. The DASO₃-GSH mixtures (0.5 ml) were subjected to semipreparative HPLC analysis with an Ultrasphere ODS column (5 µm, 10 × 250 mm; Beckman, Palo Alto, CA) and 5% aqueous methanol containing 0.06% TFA as the mobile phase at a flow rate of 5.0 ml/min. Peaks of interest were collected, lyophilized in vacuo, frozen at 70°C, and subjected to ¹H NMR analysis. The synthesized GSH conjugates were used as analytical standards.

HPLC Analysis of DASO₃-GSH Conjugates. The GSH conjugates formed in the liver microsomal incubations (60 µl) were analyzed with a reverse phase C18 column (5 µm, 4.6 × 250 mm; Phenomenex). The mobile phase consisted of solvent A (0.06% aqueous TFA) and solvent B (acetonitrile containing 0.06% TFA) at a constant flow rate of 1.0 ml/min. The gradient started at 100% solvent A, followed by a linear increase to solvent B in 30 min, then 5% B to 90% B. The column effluent was monitored at 200 nm. To detect GSH-containing peaks, 0.25-ml aliquots of the column were collected; the relative size of the peaks was determined by conversion of radioactivity levels to nanomolar or picomolar amounts by using the specific activity of [³H]GSH. The DASO₂-GSH incubation mixtures (0.25 ml) were subjected to semipreparative HPLC analysis with an Ultrasphere ODS column (5 µm, 10 × 250 mm; Beckman) and the previously described solvent A and solvent B gradient at a flow rate of 5.0 ml/min. The peak corresponding to the retention time of the DASO₃-GSH standard was collected, adjusted to pH 7.0 with ammonium hydroxide, lyophilized in vacuo, and stored at 70°C.

Formation of DASO₃ in Microsomal Incubations Levels of DASO₃ generated in the liver microsomal incubations were determined by formation of the NBP derivative (Imamura and Talcott, 1985). The reaction mixtures in the incubations contained 1.5 ml of 100 mM K₂PO₄ buffer, pH 7.4 (pH adjusted with 70% phosphoric acid), 1.5 mg of microsomal protein/ml, DASO₂, and an NADPH-generating system as described previously. It was important that the 100 mM K₂PO₄ buffer was adjusted with 70% phosphoric acid and did not contain EDTA, NaOH, or HCl because these components inhibited the NBP reaction. The reaction mixtures were preincubated for 3 min at 37°C, after which DASO₂ was added and the incubations continued. In the concentration-response experiments, DASO₂ was used at amounts ranging from 0.25 to 2 mM, and an incubation time of 30 min was used. In the time course experiments, 1.0 mM DASO₂ was used, and the duration of incubation ranged from 5 to 50 min. In the control samples, NADP⁺ and DASO₂ were omitted from the reaction mixtures.

PNP Hydroxylation. PNP activity was used as a catalytic marker for CYP2E1 and was determined as described in Forkert et al. (1996). Hydroxylation of PNP was determined in microsomes from untreated or DASO₂-treated mice or from microsomes that were incubated with DASO₂. Hydroxylase activity was estimated by the formation of 4-nitrocatechol determined spectrally at 546 nm.

Protein Immunoblotting Western blot analysis was carried out according to methods described in Forkert (1995), and was performed with samples of liver microsomes incubated with DASO₂ or liver microsomes from DASO₂-treated mice. Proteins were separated by SDS-polyacrylamide gel electrophoresis (PAGE) and stained with Coomassie Blue or transferred to a nitrocellulose membrane. The membrane was reacted overnight with the inhibitory CYP2E1 mAb 1-98-1 (1:1000). Protein bands recognized by the antibody were detected by incubation with goat anti-mouse IgG conjugated to alkaline phosphatase, and visualized by development in a solution containing p-nitroblue tetrazolium chloride and 5-bromo-4-chloro-3-indolylphosphate p-toluidine salt.

Heme and P450 Contents Total cytochrome P450 and heme contents were determined after incubations with DASO₂. Microsomal P450 content was estimated from the sodium dithionite difference spectra of carbon monoxide-saturated microsomes, and heme content was determined with the pyridine-hemochromogen method (Omura and Sato, 1964; Estabrook et al., 1972). Incubations with ethyl 3,5-diethoxycarbonyl-1,4-dihydro-2,4,6-trimethylpyridine (DDC) (90 µM), a compound known to alkylate the heme moiety of P450 (Riddick et al., 1989), were performed as a positive control for heme degradation.

Instrumentation. HPLC experiments were performed on a Beckman System Gold Programmable Solvent Module 126 HPLC with a Beckman System Gold Module 168 UV detector. Spectral analyses for enzyme assays were performed on a Beckman Model DU 640B spectrophotometer. ¹H NMR spectra were obtained with a Bruker Avance spectrometer at 500 MHz.

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

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Synthesis and Characterization of DASO₃-GSH Conjugate Standards. HPLC analysis of the products of reaction of GSH with DASO₃ yielded a major peak at 20.5 min (Fig. 1A). The size of this peak was proportional to the amount of DASO₃ added to the initial synthesis and was not detectable in the absence of either GSH or DASO₃. The identities of the products contributing to this peak was confirmed to be the epoxide-derived GSH conjugates S-(1R,S- [[1-hydroxymethyl-2,3'-sulfonyl]-1'-propenyl]ethyl)glutathione (conjugates [A] and [B], diastereomers) and S-(1-[[2R,S-hydroxypropyl]-3,3'-sulfonyl]-1'-propenyl)glutathione (conjugates [C] and [D], diastereomers). The spectral data of the GSH conjugates were consistent with reported values (Jin and Baillie, 1997). ¹H NMR (D₂O) indicated approximately a 3:2 mixture of conjugates [C]/[D] to conjugates [A]/[B]. Assignments for conjugate [C] are as follows: : 1.97 (2H, m, Glu-, ¹), 2.55 (2H, m, Glu-, ¹), 2.84 (2H, m, CH₂SG), 2.95 (1H, m, Cys-), 3.14 (1H, m, Cys-¹), 3.48 (2H, m, CH(OH)CH₂SO₂), 3.80 (1H, m, Glu-CH), 3.92 (2H, s, Gly-CH₂), 4.08 (2H, m, CH₂-CH=CH₂), 4.39 (1H, m, CH(OH)-CH₂SG), 4.60 (1H, m, Cys-), 5.60 (2H, m, CH=CH₂), and 5.95 (1H, m, CH=CH₂). The presence of conjugates [A]/[B] was confirmed by a multiplet at : 3.42 (1H, m, S-CH-CH₂OH), which was shown by correlated spectroscopy to be coupled to multiplets at 3.75 (1H, m, CH-CH₂-OH), and 3.80 (1H, m, CH-CH₂-OH), and also to a multiplet at 3.62 (2H, m, S-CH-CH₂SO₂). The cysteinyl methine proton of conjugates [A]/[B] at  4.65 (1H, m, Cys-) also is partially resolved from that of conjugates [C]/[D] and shown by correlated spectroscopy to be coupled to multiplets at 3.05 (1H, m, Cys-) and 3.19 (1H, m, Cys-¹). All other resonances were unresolved from those of conjugates [C]/[D].

View larger version (19K): [in this window] [in a new window]   Fig. 1.   Chromatograms of DASO3-GSH conjugates subjected to reverse phase HPLC analysis. The standard (A) was synthesized by reacting DASO3 (1.2 mM) with [3H]GSH (1.2 mM; 0.2 µCi/ml). The epoxide-derived GSH conjugates were detected in liver microsomal incubations (B) containing 3.0 mg of microsomal protein/ml, DASO2 (0.5 mM), [3H]GSH (0.1 mM; 0.1 µCi/ml), and an NADPH-generating system in a final volume of 2 ml. The GSH-DASO3 conjugate also was identified in incubations containing CYP2E1-expressed human lymphoblastoid microsomes (1.5 pmol of cytochrome P450), [3H]GSH (0.5 mM; 0.5 µCi/ml), DASO2 (0.5 mM), and an NADPH-generating system in a final volume of 100 µl (C). The controls included incubations in which NADPH was omitted from the reaction mixtures (D). Procedures for the chemical synthesis and the incubations are detailed in Materials and Methods. Aliquots of the column effluent were collected, and levels of radioactivity were determined.

Microsomal Incubations. Preliminary studies showed that formation of the DASO₃-GSH conjugates in the microsomal incubations was most efficient at 37°C. On the basis of this finding, all subsequent incubations were performed at this temperature. HPLC analysis of the products of the microsomal incubations produced a major peak with a retention time (20.5 min) corresponding to the DASO₃-GSH standard (Fig. 1, A and B). This peak was not detectable when DASO₂ or the NADPH-generating system was omitted from the incubation mixtures (Fig. 1D). The relative size of the 20.5-min peak, as assessed by summation of radioactive counts, was used as an estimate of the conjugation of DASO₃ to GSH. Our results showed that incubation with amounts of DASO₂ ranging from 0.25 to 1.0 mM produced concentration-dependent increases in GSH conjugation that reached a plateau between 0.75 and 1 mM DASO₂ (Fig. 2A). The response was saturable, and concentrations of DASO₂ that were >0.75 mM produced little increase in the levels of GSH conjugates formed. Results from the time course experiments showed that increasing amounts of the DASO₃-GSH conjugates continued to be produced up to 2 h, and reached a plateau thereafter (Fig. 2B). Addition of liver glutathione S-transferase enzymes (2.0-10.0 µM) to the microsomal incubations produced only a minor increase (<10%) in the rate of conjugation of DASO₃ to GSH (data not shown).

View larger version (17K): [in this window] [in a new window]   Fig. 2.   Concentration- and time-dependent formation of DASO3-GSH conjugates from DASO2 in liver microsomal incubations. In the incubations, the reaction mixtures contained 3.0 mg of microsomal protein/ml, an NADPH-generating system, [3H]GSH (0.1 mM; 0.1 µCi/ml), and DASO2 in a final incubation volume of 2 ml. In the concentration-response studies (A), the amounts of DASO2 used in the incubations ranged from 0.25 to 1.0 mM; an incubation time of 1.5 h was used. In the time course experiments (B), the reaction was performed with 0.5 mM DASO2, and the incubation time ranged from 0.5 to 3.0 h. After precipitation of proteins, the supernates (100 µl) were subjected to reverse phase HPLC analysis. The values represent the relative size of the major peak at 20.5 min measured by the summation of the radioactive counts and conversion using the specific activity of GSH. Data are expressed as the mean ± S.D. of triplicate determinations performed on three different microsomal preparations.

Formation of DASO₃ in Liver Microsomal Incubations The formation of DASO₃ in liver microsomal incubations was estimated by measuring the formation of the NBP derivative. Microsomes were incubated in the presence of an NADPH-generating system with concentrations of DASO₂ ranging from 0.25 to 2 mM. Formation of the NBP derivative was concentration-dependent and was proportional to the amounts of DASO₂ used in the incubations. Saturation was attained at a concentration of 1.0 mM DASO₂ (Fig. 3A). Increase of the DASO₂ concentration to 2 mM produced no further increase in the levels of DASO₃ formed, as assessed by levels of the NBP derivative. Time course experiments revealed that the formation of DASO₃ from DASO₂ (1 mM) increased from 5 to 30 min, and gradually declined thereafter (Fig. 3B). Moreover, ~50% of DASO₃ was generated within the first 10 min of the microsomal incubations containing DASO₂ and an NADPH-generating system, whereas the remaining quantities of DASO₃ were generated over a period of 20 min (Fig. 3B). Thus, DASO₃ was readily produced from DASO₂ in the microsomal incubations in a time- and dose-dependent manner, and was not detectable in microsomal incubations in which NADPH or DASO₂ was omitted.

View larger version (16K): [in this window] [in a new window]   Fig. 3.   Concentration- and time-dependent formation of DASO3 from DASO2 in hepatic microsomal incubations. The amounts of DASO3 formed were estimated from production of the NBP derivative and related to a reference standard curve constructed from the synthesized DASO3. The reaction mixtures in the incubations contained 2.25 mg of microsomal protein, an NADPH-generating system, and DASO2. In the concentration-response experiments (A), the amounts of DASO2 used in the incubations ranged from 0.25 to 2.0 mM; an incubation time of 30 min was used. In the time course studies (B), DASO3 formation from DASO2 (1 mM) was determined at incubation times ranging from 5 to 50 min. Formation of the alkylated NBP derivative was determined as described in Materials and Methods. Data are expressed as the mean ± S.D. of triplicate determinations performed on three different microsomal preparations.

PNP Hydroxylation. In vitro and in vivo studies were performed to evaluate the inhibitory effects of DASO₂ on CYP2E1-dependent PNP hydroxylase activity. In the in vitro studies, incubations of liver microsomes with DASO₂ in the presence of an NADPH-generating system produced inhibition of PNP hydroxylase activity. The inhibitory effects were concentration-dependent between 0.25 to 1.0 mM DASO₂, with decreases that were proportional to the amounts of DASO₂ used in the incubations (Fig. 4A). Hydroxylase activity was not inhibited in microsomes incubated with DASO₂ in the absence of the NADPH-generating system. In the in vivo studies, PNP hydroxylase activity was determined in liver microsomes prepared from control and DASO₂-treated mice. In preliminary time course studies, mice were treated with 100 mg/kg DASO₂ and PNP hydroxylase activity was determined at 1 to 24 h after treatment. Hydroxylase activity was maximally depressed 2 h after DASO₂ treatment but returned to control levels after 4 h. Subsequent measurements of hydroxylase activity were determined at 2 h after DASO₂ exposure. Treatment with doses of DASO₂ ranging from 25 to 200 mg/kg produced marked decreases in enzyme activity that were maximal from the 25- to 100-mg/kg doses. Treatment of mice with 100 mg/kg DASO₂ produced a reduction in enzyme activity that was ~70% of the control level. Treatment with higher doses of DASO₂ (150 and 200 mg/kg) evoked decreases in PNP hydroxylase activity that were not as pronounced as those found at the lower DASO₂ doses. Nevertheless, residual hydroxylase activity at the 200-mg/kg dose was only ~12% of the control levels. Thus, exposure to DASO₂ produced marked decreases in the levels of CYP2E1-associated PNP hydroxylase activity.

View larger version (17K): [in this window] [in a new window]   Fig. 4.   Concentration- and dose-dependent effects of DASO2 on PNP hydroxylase activity. In the in vitro experiments, liver microsomal incubations were performed with various concentrations of DASO2 (A). Reaction mixtures contained 6.0 mg of microsomal protein, an NADPH-generating system, and DASO2 (0.25-1.0 mM). In the in vivo studies, mice were treated with DASO2 (25 to 200 mg/kg p.o.), and PNP hydroxylase activity was determined in liver microsomes 2 h after treatment. PNP hydroxylase activity was estimated by measuring the formation of 4-nitrocatechol from PNP as described in Materials and Methods. Data are expressed as mean ± S.D. of triplicate determinations for each DASO2 concentration or dose, and was performed with three separate microsomal preparations.

Protein Immunoblotting. Protein immunoblots were prepared with liver microsomes that were incubated with DASO₂ in the presence of an NADPH-generating system as described previously. Staining with Coomassie Brilliant Blue of proteins separated by SDS-PAGE confirmed the integrity of the protein bands (Fig. 5A). The CYP2E1 antibody detected a single protein band of 51 kDa (Fig. 5, B and C) and this molecular mass was similar to that of immunodetectable CYP2E1 obtained previously (Lee and Forkert, 1994). There was a loss of immunoreactivity in the protein blots prepared from liver microsomes incubated with DASO₂, compared with CYP2E1 content in the microsomes from untreated mice (Fig. 5C). Immunoreactivity also was decreased in the protein bands prepared from liver microsomes of mice treated in vivo with DASO₂ (Fig. 5B). The loss of immunodetectable CYP2E1 in the microsomal samples was dose-dependent (25-200 mg/kg), with signal band reduction being most pronounced at the highest dose of DASO₂ administered to the mice.

View larger version (40K): [in this window] [in a new window]   Fig. 5.   Protein immunoblotting for CYP2E1 in liver microsomes exposed to DASO2 in vivo (B) or in vitro (C). In the in vivo experiments, mice were treated with DASO2 and were sacrificed 2 h later for isolation of microsomes. In the in vitro experiments, microsomes were incubated with DASO2 in the presence of an NADPH-generating system. Microsomal proteins were separated by SDS-PAGE, stained with Coomassie Blue (A) or transferred to a nitrocellulose membrane and reacted with a CYP2E1 mAb. In A and B, lanes were loaded as follows: lane 1, molecular mass standards; lane 2, microsomes from untreated mice; and lanes 3 to 7, microsomes from mice treated with 25, 50, 100, 150, and 200 mg/kg DASO2, respectively. In C, lanes were loaded as follows: lane 1, molecular mass standards; lane 2, microsomes from control incubations; lanes 3 to 6, microsomes incubated with DASO2 at concentrations of 0.001, 0.01, 0.1, and 1.0 mM, respectively; and lane 7, buffer. All lanes in the immunoblots were loaded with 3.0 µg of microsomal protein.

Cytochrome P450 and Heme Contents. Levels of cytochrome P450 and heme were determined in liver microsomes incubated with DASO₂ in the presence or absence of an NADPH-generating system. The results are summarized in Fig. 6. Controls consisted of microsomes that were incubated in the absence of an NADPH-generating system or DASO₂. Control levels of P450 (0.41 ± 0.01 nmol/mg of protein) were similar to those described in previous studies (Lee and Forkert, 1994). Incubation of liver microsomes with DASO₂ concentrations of 0.25 and 0.50 mM produced no alterations in P450 levels. However, incubation with 0.75 and 1 mM DASO₂ evoked decreases of ~20 and 30% in P450 levels, respectively. Loss of P450 was not observed in incubations in which NADPH was omitted from the reaction mixtures, over the same concentration range of DASO₂.

View larger version (21K): [in this window] [in a new window]   Fig. 6.   Effects of DASO2 concentrations on cytochrome P450 and heme levels in liver microsomes. Cytochrome P450 and heme contents were determined in microsomes incubated with DASO2, using the pyridine-hemochromogen method as described in Materials and Methods. Data are expressed as the mean ± S.D. of triplicate determinations performed on three different microsomal preparations. The data were analyzed by two-way ANOVA and by the Student-Newman-Keuls test, P < .05. Control levels of P450 and heme were 0.41 ± 0.01 and 1.39 ± 0.08 nmol/mg of protein, respectively. a, significantly different from P450 levels at the same DASO2 concentrations; b, significantly different from P450 levels in the control and in incubations containing 0.25 or 0.50 mM DASO2; and c, significantly different from heme levels in the control and in incubations containing 0.25 mM DASO2.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

Previous studies have reported that DAS is converted to DASO, which is subsequently metabolized to DASO₂ (Brady et al., 1991). All three compounds are competitive inhibitors of CYP2E1, and inhibited PNP hydroxylase activity in incubations with liver microsomes from acetone-treated rats. However, the inhibitory effects on PNP hydroxylase activity were more pronounced with DASO₂ than with either DAS or DASO, and occurred in a reaction that was time-dependent, required NADPH, and was saturable (Brady et al., 1991). These findings suggested that DASO₂ is a mechanism-based inhibitor and inactivated CYP2E1 in a process involving the formation of a metabolite. More recent studies confirmed that DASO₂ was generated from DAS through DASO, and it was hypothesized that it is the final metabolic event involving DASO₂ that is mainly responsible for the destruction of CYP2E1 and that mediates the chemoprotective effects seen in vivo with DAS (Jin and Baillie, 1997).

An objective of this study was to investigate the mechanism responsible for DASO₂-mediated inactivation of CYP2E1 and to identify the reactive intermediate that is formed from DASO₂ in liver microsomal incubations. We predicted that CYP2E1-mediated oxidation of DASO₂ is likely to occur at either one or both of the terminal double bonds of DASO₂ to yield the monoallylepoxide (DASO₃) or the diallylepoxide (DASO₄) or possibly a mixture of these compounds. Data from our preliminary experiments indicated that DASO₃ was the major product formed from oxidation of DASO₂, whereas DASO₄ was formed at minimal amounts. As a result of these findings, it was anticipated that DASO₃ is likely the metabolite responsible for inactivation of CYP2E1. Our experimental approach was to use [³H]GSH to trap the short-lived epoxide and to determine its formation as a GSH conjugate with HPLC analyses. The GSH conjugates identified from reaction of the chemically synthesized DASO₃ with GSH were conjugates [A]/[B] and conjugates [C]/[D] (Fig. 7). These pairs of diastereomers were regioisomers with identical molecular masses, similar structures, similar ¹H NMR spectra, and eluted from the HPLC column at the same time. These synthesized DASO₃-GSH conjugates were used as analytical standards to identify the epoxide-derived GSH conjugates generated from DASO₂ in the liver microsomal incubations. The formation of the conjugates was contingent on the presence of NADPH (Fig. 1), and was time- and concentration-dependent (Fig. 2). Conjugate formation was maximal at a concentration of 0.75 to 1.0 mM DASO₂ (Fig. 2A), and proceeded steadily over a period of 2.0 h in the microsomal incubations (Fig. 2B). Conjugation of DASO₃ with GSH appears to be catalyzed nonenzymatically inasmuch as addition of glutathione S-transferases to the microsomal incubations produced only a small increase in the amounts of conjugates produced (data not shown). These findings indicated that the reaction was P450 mediated and supported the premise that an epoxide is formed and conjugated with GSH. The amounts of DASO₃ generated in the microsomal incubations also were determined by formation of the NBP-derivatized product. The level of derivatized DASO₃ detected was maximal at 30 min (18.3 ± 0.20 nmol/mg of protein/min); however, ~50% of levels of the derivative was produced within the first 10 min of the initial reaction (Fig. 3B). Peak formation of the derivatized DASO₃ in the microsomal incubations was manifested at a DASO₂ concentration of 1.0 mM DASO₂ (Fig. 3A), and this is similar to the amount of DASO₂ required to produce saturation in terms of formation of the epoxide-derived GSH conjugates (Fig. 2A). These findings suggested that the DASO₃-GSH conjugates are formed readily, are relatively stable, and are detectable up to 2 to 3 h after the initial reaction of the microsomes with DASO₂. In contrast and as expected, the epoxide is a highly reactive species and is likely to bind to cellular macromolecules or undergo rapid hydrolysis.

View larger version (14K): [in this window] [in a new window]   Fig. 7.   Proposed scheme of metabolism of DASO2.

The role of CYP2E1 in the metabolism of DASO₂ has been investigated in the current studies by using measurements of catalytic activity, protein immunoblotting, and a CYP2E1 expression system. Levels of CYP2E1-dependent PNP hydroxylase activity were decreased in incubations of liver microsomes with DASO₂, with alterations that were concentration-dependent (Fig. 4A). The loss of PNP hydroxylase activity also was detected in mice treated in vivo with DASO₂, and this occurred in a dose-dependent manner (Fig. 4B). The reduced levels of catalytic activity corresponded with decreased content of immunodetectable CYP2E1 (Fig. 5). These findings strongly implicated CYP2E1 in the metabolism of DASO₂. To establish a role for CYP2E1 in catalyzing the oxidation of DASO₂ to DASO₃, incubations with DASO₂ were performed in a system containing CYP2E1-expressed lymphoblastoid microsomes. HPLC analysis of products of the microsomal incubations revealed a major peak eluting at a time that is similar to the metabolism-dependent peak observed in incubations containing liver microsomes. ¹H NMR analysis of the column eluent containing this peak revealed the presence of the GSH-DASO₃ conjugates that were identified in the liver microsomal incubations (Fig. 1). These results supported the premise that the CYP2E1-mediated generation of DASO₃ from DASO₂ is responsible for the inactivation of CYP2E1.

We have observed decreased levels of spectrally detectable P450 and heme after incubation of liver microsomes with DASO₂ (Fig. 6). However, the loss of heme was more pronounced than that sustained by P450 in incubations with the same DASO₂ concentrations, and was manifested at a lower DASO₂ concentration than for P450. Relevant in this context is a comparative analysis of the relative loss of P450 content versus the level of CYP2E1 inactivation. The CYP2E1 enzyme comprises only a small fraction of the total P450 pool, and therefore the marked decreases in PNP hydroxylase activity and P450 and heme contents detected in this study (Figs. 4 and 6) cannot all be attributed to alterations of the CYP2E1 enzyme. Although PNP hydroxylation is regarded as an index of CYP2E1-dependent catalytic activity (Koop, 1986), PNP also can be a substrate for other P450 enzymes, including CYP2F2 (Shultz et al., 1999). Our results therefore suggested that DASO₂ may mediate alkylation of other P450 isozymes that have not to date been identified. The marked loss of heme content may be a result of not only P450 destruction but also effects of DASO₂ on other hemeproteins such as cytochrome b₅.

Our findings raise the question as to whether the inactivation of CYP2E1 is due to a primary effect of the DASO₂ metabolite on the heme or apoprotein of P450. There are at least three possible mechanisms for P450 inactivation. These include covalent binding of the reactive metabolite with the apoprotein moiety of CYP2E1 (Halpert and Neal, 1981; Halpert et al., 1983) or irreversible binding directly to the prosthetic heme group of the P450 enzyme (Ortiz de Montellano and Correia, 1983), leading to destabilization or degradation of the apoprotein. In a third mechanism, a reactive metabolite inactivates the heme moiety of P450 but produces subsequent fragmentation of the heme into reactive metabolites that bind irreversibly to the apoprotein (Davies et al., 1986a; Correia et al., 1987). Ethyl DDC, the suicide substrate of P450 used in this study as a positive control for DASO₂-mediated heme destruction, is thought to work via such a mechanism (Davies et al., 1986b). In this case, activation of the prosthetic heme group leads to the release of reactive fragments that preferentially alkylate the P450 apoprotein (Davies et al., 1986b; Riddick et al., 1989). The significantly greater effect of DASO₂ on heme versus P450 suggested that the reactive intermediate formed from DASO₂ may initially affect the heme component and/or that other hemoproteins on the microsomal membrane also are affected. However, the precise mechanism for CYP2E1 destruction remains to be established, and it has yet to be clarified whether DASO₃ affects first the heme and subsequently the apoprotein or whether both moieties are inactivated independently.

An assumption has been made herein that the epoxide formed from DASO₂ (DASO₃) is responsible for alkylation of CYP2E1. Although the formation of DASO₃ and DASO₃-GSH conjugates is coincidental with CYP2E1 inhibition, the epoxide may not be the reactive species responsible directly for P450 alkylation. This concept emanated from studies with 2-allyl-2-isopropylacetamide, a compound that causes P450 destruction by alkylating the prosthetic heme group (Ortiz de Montellano et al., 1979). The findings from these studies indicated that neither the epoxide nor secondary metabolites derived from the epoxide mediates heme alkylation, and the P450 loss is thought to be due to the action of a precursor cationic intermediate species. This mechanism of P450 destruction as a result of heme alkylation appears to be the case also for methyl 2-isopropyl-4-pentenoate, the methyl ester analog of 2-allyl-2-isopropylacetamide as well as for the therapeutic agent novonol (Ortiz de Montellano et al., 1979, 1984). In the context of DASO₂, it is possible that DASO₃ mediates CYP2E1 alkylation; however, the precise species acting on this P450 leading to its inactivation is not known and requires further investigation.

In summary, our results have provided data to support the assertion that the reactive intermediate generated from DASO₂ is DASO₃, and that this metabolite may be responsible for CYP2E1 inactivation. These data are also consistent with the premise that CYP2E1 inactivation results in inhibition of metabolic activation, leading to the chemoprotective effects reported for CYP2E1-dependent substrates that are metabolized to carcinogenic metabolites.