Introduction

Top Abstract Introduction Materials & Methods Results Discussion References

Liver is often a target of drug-induced toxicity, which is generally attributed to the following two causes. One is a direct toxicity in which reactive metabolites are formed and damage critical cell targets (Nelson and Pearson, 1990; Hinson and Roberts, 1992; Boelsterli, 1993). Another is an indirect toxicity in which reactive metabolites covalently bind to proteins, which then behave as neoantigens, and trigger an abnormal immunological response leading to the disease (Boelsterli, 1993; Pohl et al., 1988; Pirmohamed et al., 1996). An example of the latter is hepatitis induced by halothane (Pohl et al., 1989; Pohl, 1990) or tienilic acid (Homberg et al., 1984; Beaune et al., 1987). Antimicrosome autoantibodies commonly appear in the sera of the patients with the drug-induced hepatitis. The formation of reactive metabolites has been proposed as an initial step of the disease (Pohl, 1990; Beaune et al., 1987; Lecoeur et al., 1994). The step may be followed by covalent binding of the metabolites to the protein(s) generating the reactive metabolite(s) and/or other proteins, whereas it has not been always elucidated what the targets of the reactive metabolite are.

Dihydralazine, an antihypertensive drug, is known to induce immunoallergic hepatitis (Pariente et al., 1983; Nataf et al., 1986). The autoantibodies reacting with liver microsomes (anti-LM¹) found in the serum of the patients have been reported to be directed against cytochrome P450 1A2 (CYP1A2) (Bourdi et al., 1990). It is thus suggested that a reactive metabolite of dihydralazine covalently binds to the P450 protein and triggers an immunological response as a neoantigen. In practice, it was demonstrated that dihydralazine was activated into a chemically reactive metabolite that covalently binds to liver microsomal protein by CYP1A2 (Bourdi et al., 1994). However, it remains unknown whether the appearance of anti-LM resulted from covalent binding only to CYP1A2 because its selectivity as the target protein of dihydralazine-reactive metabolite has not been evaluated. In the present study, we investigated the selectivity of inactivation of P450 enzymes during the metabolism of dihydralazine to evaluate the P450 protein(s) as the specific target protein of the reactive metabolite.

	Materials and Methods

Top Abstract Introduction Materials & Methods Results Discussion References

Chemicals. Dihydralazine dihydrochloride and -naphthoflavone were purchased from Aldrich; hydralazine hydrochloride, sodium phenobarbital, and resorufin were from Tokyo Chemical Industry (Tokyo, Japan); phenacetin, 4-acetamidophenol, ethoxyresorufin, pentoxyresorufin, testosterone, and 2- and 16-hydroxytestosterones were from Sigma; 6-hydroxytestosterone was from Steraloids Inc. (Wilton, NH); glucose 6-phosphate (G-6-P), glucose 6-phosphate dehydrogenase (G-6-PDH), and NADPH were from Oriental Yeast Co., Ltd. (Tokyo, Japan); and reduced glutathione (GSH) was from Wako Pure Chemical (Osaka, Japan). All other chemicals and solvents used were of analytical grade.

Preparation of Liver Microsomes. Male Wistar rats (2 months old) were obtained from Takasugi Experimental Animals (Saitama, Japan). The animals were housed in an air-conditioned room (25°C) under a 12-hr light-dark cycle for 1 week prior to use. Food (commercially available pellet, Oriental Yeast Co., Ltd.) and water were given ad libitum. -Naphthoflavone (80 mg/kg in corn oil) or sodium phenobarbital (80 mg/kg in physiological saline) was given to the rats intraperitoneally for 4 days. The rats were killed along with untreated rats by decapitation 24 hr after the final doses, and liver microsomal fractions were prepared according to the method of Omura and Sato (1964). Protein concentrations were assayed by the method of Lowry et al. (1951).

Protocols for Preincubation of Liver Microsomes with Hydralazine Derivatives. Liver microsomes of male Wistar rats were preincubated with dihydralazine in the presence of NADPH to determine effects of its metabolites on microsomal monooxygenase activities. A 1-ml incubation mixture contained 0.5 mg of liver microsomal protein, 10 mM G-6-P, 2 units G-6-PDH, 5 mM MgCl₂, 0.1 mM EDTA, and various concentrations of dihydralazine in 0.15 M Tris-HCl buffer (pH 7.4). After temperature equilibration (37°C, 5 min), preincubation of microsomes with dihydralazine was started by adding NADPH (final 0.5 mM) and performed for various time periods. The subsequent incubation of the microsomes for the assay of enzymatic activities was started by the addition of each test substrate, phenacetin, ethoxyresorufin, pentoxyresorufin, or testosterone. Liver microsomes from -naphthoflavone-treated rats were assayed for phenacetin O-deethylase (POD) activity and ethoxyresorufin O-deethylase (EROD) activity assays; those from phenobarbital-treated rats were to POD activity and pentoxyresorufin O-depentylase activity (PROD) assays; those from untreated rats were to POD and testosterone oxidation assays. In some experiments, hydralazine was employed instead of dihydralazine. In the control experiments, the same reaction mixture in the absence of dihydralazine as above was preincubated for the corresponding time periods; or, the mixture including the corresponding concentrations of dihydralazine was not preincubated, and the incubation for the enzyme assay was started by adding NADPH.

Assay of Enzymatic Activities. POD (Masubuchi et al., 1994) and testosterone 2-, 16-, and 6-hydroxylase (Masubuchi et al., 1995) activities were determined according to the high pressure liquid chromatographic method previously described. EROD and PROD activities were assayed by the fluorometric method to determine the resorufin formation (Burke et al., 1985). All of the assays were performed under linear conditions of metabolite formation with regard to incubation time and protein concentration.

Data Analysis. Pseudo-first order kinetic constants for the enzyme inactivation (k_inact.) were calculated from the initial slopes of the linear regression lines of the semilogarithmic plots of the remaining enzyme activity against the preincubation time. Results were represented as means ± SE. Statistical significance was calculated by the Student's t test.

	Results

Top Abstract Introduction Materials & Methods Results Discussion References

Time-Dependent Decrease in POD Activity of Liver Microsomes. POD activity was determined with microsomes from -naphthoflavone-treated rats as an indicator of the activity for CYP1A2. Dihydralazine inhibited POD activity in a concentration-dependent manner (fig. 1A). Preincubation of the microsomes with dihydralazine in the presence of NADPH intensified the inhibitory effect of the compound, resulting in one-fifth less IC₅₀ values for the inhibition than that obtained without the preincubation (with preincubation, 28.8 ± 3.1 µM, mean ± SE of three determinations; without preincubation, >200 µM). The inhibitory effect was also found to be time-dependent, i.e. the enzymatic activity decreased exponentially vs. the preincubation time of the microsomes with dihydralazine in the presence of NADPH (fig. 1B), indicating inactivation of CYP1A2 during oxidative metabolism of dihydralazine. The pseudo-first order kinetic constant for the inactivation (k_inact.) thus obtained was 0.0729 ± 0.0124 min^-1, whereas that of control was 0.0100 ± 0.0030 min^-1.

View larger version (12K): [in this window] [in a new window]   Fig. 1.   Concentration- and time-dependent decrease in POD activity during preincubation of microsomes with dihydralazine. (A) Microsomes were not preincubated () or preincubated for 10 min with various concentrations of dihydralazine in the presence of NADPH (), followed by assay of POD activity. IC50 values obtained under the conditions were >200 µM and 28.8 ± 3.1 µM, respectively. Results are represented as per cent of the activity obtained without dihydralazine and are means ± SE (N = 3). (B) Microsomes were preincubated without () or with () dihydralazine (50 µM) in the presence of NADPH, followed by assay of POD activity. First-order inactivation constants by dihydralazine was 0.0729 ± 0.0124 min-1. Results are represented as per cent of the activity obtained without the preincubation and are means ± SE (N = 3).

View larger version (18K): [in this window] [in a new window]   Fig. 2.   Inhibition kinetics of POD activity by dihydralazine. (A) POD activities were determined in the absence or presence of various amounts of dihydralazine. (B) Microsomes were preincubated with various amounts of dihydralazine in the presence of NADPH for 10 min, followed by assay of POD activity. Results are expressed as Lineweaver-Burk plots and as typical results from three determinations.

View larger version (21K): [in this window] [in a new window]   Fig. 3.   Inhibition of POD activity of various microsomes by dihydralazine. Microsomes were not preincubated (white bar) or preincubated for 10 min with dihydralazine (50 µM) in the presence of NADPH (stippled bar), followed by assay of POD activity. Results are represented as per cent of the activity obtained without dihydralazine and are means ± SE (N = 3). **, significantly different from "without preincubation" (p < 0.01).

Time-Dependent Decrease in EROD and PROD Activities of Liver Microsomes. Time-dependent effects of the preincubation of microsomes with dihydralazine and NADPH were also studied on EROD activity in microsomes of -naphthoflavone-treated rats and PROD activity in phenobarbital-treated rats, which are indicators for the activities of CYP1A1 and CYP2B1/2, respectively. The preincubation of microsomes with dihydralazine caused time-dependent decreases in these activities (fig. 4) but was less effective than that on POD activity in the microsomes of -naphthoflavone-treated rats. The k_inact. values for EROD and PROD activities thus obtained were 0.0172 ± 0.0004 and 0.0200 ± 0.0068 min^-1, respectively.

View larger version (14K): [in this window] [in a new window]   Fig. 4.   Time-dependent decrease in EROD and PROD activities during preincubation of microsomes with dihydralazine. Microsomes were preincubated without () or with () dihydralazine (50 µM) in the presence of NADPH, followed by assay of alkoxyresorufin O-dealkylase activities. First-order inactivation constants for EROD and PROD activities were 0.0172 ± 0.0004 and 0.0200 ± 0.0068 min-1, respectively. Results are represented as per cent of the activity obtained without the preincubation and are means ± SE (N = 3).

Time-Dependent Decrease in Testosterone Oxidation Activities of Liver Microsomes. Testosterone 2-, 16-, and 6-hydroxylase activities were determined with microsomes from untreated rats. Dihydralazine (50 µM) inhibited all of the activities measured here in a concentration-dependent manner, and the inhibitory effect was more potent on 6-hydroxylase activity than on 2- and 16-hydroxylase activities (fig. 5). The enzymatic activities decreased exponentially vs. the preincubation time of the microsomes with dihydralazine in the presence of NADPH, and the decreases were more pronounced in 2- and 16-hydroxylase activities than in 6-hydroxylase activity (fig. 6). The k_inact. for testosterone 2-, 16-, and 6-hydroxylase activities thus obtained was 0.0664 ± 0.0030, 0.0696 ± 0.0059, and 0.0347 ± 0.0055 min^-1, respectively.

View larger version (16K): [in this window] [in a new window]   Fig. 5.   Inhibition of testosterone oxidation by dihydralazine. Testosterone 2- (), 16- (), and 6- () hydroxylase activities were determined in the absence or presence of various amounts of dihydralazine. Results are represented as per cent of the activity obtained without dihydralazine and are means ± SE (N = 3).

View larger version (14K): [in this window] [in a new window]   Fig. 6.   Time-dependent decrease in testosterone oxidation activities during preincubation of microsomes with dihydralazine. Microsomes were preincubated without () or with () dihydralazine (50 µM) in the presence of NADPH, followed by assay of testosterone oxidation activities. First-order inactivation constants for testosterone 2-, 16-, and 6-hydroxylase activities were 0.0664 ± 0.0030, 0.0696 ± 0.0059, and 0.0347 ± 0.0055 min-1, respectively. Results are represented as per cent of the activity obtained without the preincubation and are means ± SE (N = 3).

Effect of GSH on Dihydralazine-Induced Decreases in Testosterone Oxidation Activities. Liver microsomes of untreated rats were preincubated with dihydralazine and NADPH in the presence or absence of GSH to determine its protective effect against the inhibition of testosterone oxidation activities by dihydralazine metabolites. Addition of GSH (5 mM) exhibited no significant effect on the decreases in testosterone 2-, 16-, and 6-hydroxylase activities induced by the preincubation of microsomes with dihydralazine (fig. 7).

View larger version (23K): [in this window] [in a new window]   Fig. 7.   Effect of GSH on decreases in testosterone oxidation activities during preincubation of microsomes with dihydralazine. Microsomes were preincubated for 10 min with dihydralazine (50 µM) and NADPH in the absence (white bar) or presence (stippled bar) of GSH (5 mM), followed by assay of testosterone 2-, 16-, and 6-hydroxylase activities. Results are represented as per cent of the activity obtained without dihydralazine and are means ± SE (N = 3).

Effect of Hydralazine on POD and Testosterone Oxidation Activities. Addition of hydralazine, a metabolite of dihydralazine, significantly decreased POD activity. However, no additional effect of the preincubation of microsomes in the presence of NADPH was obtained with hydralazine (fig. 8).

View larger version (22K): [in this window] [in a new window]   Fig. 8.   Inhibition of POD activity by dihydralazine and hydralazine. Microsomes were not preincubated (white bar) or preincubated for 10 min with dihydralazine or hydralazine (50 µM) in the presence of NADPH (stippled bar), followed by assay of POD activity. Results are represented as per cent of the activity obtained without dihydralazine and hydralazine and are means ± SE (N = 3). **, significantly different from "without preincubation" (p < 0.01).

View larger version (23K): [in this window] [in a new window]   Fig. 9.   Inhibition of testosterone oxidation activities by dihydralazine and hydralazine. Microsomes were not preincubated (white bar) or preincubated for 10 min with dihydralazine (A) or hydralazine (B) (50 µM) in the presence of NADPH (stippled bar), followed by assay of testosterone oxidation activities. Results are represented as per cent of the activity obtained without dihydralazine and hydralazine and are means ± SE (N = 3). * and **, significantly different from "without preincubation" (p < 0.05, p < 0.01, respectively).

	Discussion

Top Abstract Introduction Materials & Methods Results Discussion References

Sera from patients with dihydralazine-induced hepatitis were shown to contain anti-liver microsomal antibodies (anti-LM) (Pariente et al., 1983; Nataf et al., 1986). The anti-LM antibodies were specific for the disease because none of the other antisera tested behaved in the same manner as anti-LM, even those from patients treated with dihydralazine and without hepatic disease. The sera recognized a single polypeptide in human liver microsomes as judged by immunoblotting, and the antigen was identified as CYP1A2 (Bourdi et al., 1990). The antibodies were found to react specifically with CYP1A2 but not with CYP1A1 expressed in yeast and bacteria (Bourdi et al., 1992). Incubation of microsomes from rat and human livers resulted in covalent binding of dihydralazine metabolites to the microsomes. Formation of these metabolites was shown to be mediated by the CYP1A isoenzyme(s) (Bourdi et al., 1994). These metabolites were also suggested to bind to CYP1A2, which produced them, as judged from migration of covalently bound radioactivity in sodium dodecyl sulfate-polyacrylamide gel electrophoresis (Bourdi et al., 1994). These results derived the following hypothesis: CYP1A2 metabolizes dihydralazine into reactive metabolites that bind to it, forming a neoantigen that triggers an immune response characterized by autoantibodies against CYP1A2.

The present study showed that preincubation of liver microsomes of -naphthoflavone-treated rats with dihydralazine in the presence of NADPH resulted in time-dependent loss of POD activity, an indicator of CYP1A2 activity (Sesardic et al., 1990), showing inactivation of CYP1A2 during the dihydralazine metabolism. It is consistent with the hypothesis that CYP1A2, which is involved in the metabolic activation of dihydralazine, becomes a target of the reactive metabolite thus formed, as described above. Therefore, the present in vitro protocol can be useful to evaluate the target macromolecule of reactive metabolites of chemicals. We thus studied the selectivity of the inactivation of the P450 enzymes. The inhibitory effects of dihydralazine metabolites were obtained on POD activity in microsomes of untreated rats, which was also shown to be mediated by CYP1A2 (Sesardic et al., 1990). On the other hand, no significant inhibitory effect was obtained on the activity in microsomes of phenobarbital-treated rats, which was shown to be mediated by CYP2B1/2 (Kahn et al., 1987), indicating no evidence for the inactivation of the enzymes. In addition, the preincubation with dihydralazine was less effective on EROD in microsomes of -naphthoflavone-treated rats and PROD in microsomes of phenobarbital-treated rats, a marker of the activity of CYP1A1 and CYP2B1, respectively (Burke et al., 1985). These apparently indicate selective inactivation of CYP1A2. However, dihydralazine inhibited the testosterone 2- and 16-hydroxylase activities, both of which were known to be catalyzed by CYP2C11 (Sonderfan et al., 1987; Imaoka et al., 1988), and also testosterone 6-hydroxylase activity, a marker of the P450 enzymes in the CYP3A subfamily (Sonderfan et al., 1987; Imaoka et al., 1988) in liver microsomes of untreated rats, indicating that dihydralazine is a substrate and/or an inhibitor of the CYP2C11 and CYP3A enzymes. In addition, preincubation of the microsomes with dihydralazine in the presence of NADPH caused time-dependent loss of the testosterone oxidation activities, demonstrating that CYP2C and CYP3A enzymes were inactivated during oxidative metabolism of dihydralazine. Moreover, the k_inact. value obtained from remaining activity vs. time profile clearly shows that CYP2C11 was inactivated to a similar extent of CYP1A2.

It has been known that GSH can react only with the reactive metabolites which can diffuse from the active site of the enzyme that generate the metabolites. In the present study, addition of GSH did not prevent dihydralazine-induced and NADPH-dependent inactivation of the CYP2C11 or CYP3A enzyme. Thus, the inhibition of the CYP2C11 and CYP3A enzymes was demonstrated not to be a result of an unstable inhibitory metabolite formed by CYP1A2 but to the covalent binding of reactive metabolites in a mechanism-based manner. It was thus suggested that the CYP2C and CYP3A enzymes as well as CYP1A2 metabolized dihydralazine into a chemically reactive metabolite and led to the target of the reactive metabolite.

The autoantibodies reacting with the CYP2C enzyme have been observed in sera from patients with tienilic acid-induced hepatitis called anti-LKM2 (liver-kidney microsomes) (Beaune et al., 1987), and those reacting with the CYP3A enzyme have been observed with anticonvulsants (Riley et al., 1993). The present results suggest that the reactive metabolite of dihydralazine binds to the CYP2C and -3A enzymes as well as CYP1A2 and inactivates them. However, in humans, the autoantibody that recognizes the CYP2C or CYP3A enzyme has not been found in the sera from patients with dihydralazine-induced hepatitis. The reason is not presently understood, but if a CYP2C enzyme(s) really plays a role in development of dihydralazine-induced autoimmune hepatitis, the autoantibodies directed to other P450 enzymes than CYP1A2, such as CYP2C9, could be found in sera from patients with dihydralazine-induced hepatitis in future clinical trials. However, there may be a possibility that the covalently bound product, i.e. the CYP1A2-reactive metabolite complex, behaves as an antigen, but the product with other P450 enzymes does not. In any event, the selectivity of covalent binding and/or antigenicity of P450 enzymes may play a role in drug-induced autoimmune hepatitis. However, it remains to be clarified whether the autoantibodies are causative of the hepatitis or are only a marker of the disease.

The preincubation of the microsomes with hydralazine, a metabolite of dihydralazine, in the presence of NADPH did not affect either POD or testosterone oxidation activities. These data indicate that dihydralazine but not hydralazine is a precursor of the reactive intermediate(s) that binds to the P450 enzymes and inactivates them and suggest that the reactive species is formed during transformation of dihydralazine into hydralazine by CYP1A2, CYP2C, and CYP3A enzymes (fig. 10). Bourdi et al. (1994) speculated that a radical intermediate from dihydralazine to hydralazine could be a candidate to adduct with CYP1A2. The present results described here are conceivable with this speculation.

View larger version (10K): [in this window] [in a new window]   Fig. 10.   Postulated pathway of dihydralazine-induced inactivation of the P450 enzymes in rat liver microsomes.

The present results may provide other clinical importance than the drug-induced hepatitis, i.e. the inactivation of the P450 enzymes might result in elevation of the blood concentration of the drugs metabolized by CYP1A2, CYP2C, or CYP3A enzyme, which are co- and/or postadministrated with dihydralazine, suggesting an importance in view of such drug-drug interactions. On the other hand, as dihydralazine is reported to induce CYP1A2 (Bourdi et al., 1992), the induction also should be taken into account for the drug-drug interactions.

In conclusion, the present study demonstrated that dihydralazine was metabolically activated not only by CYP1A2 but also by CYP2C and CYP3A enzymes, and the chemically reactive metabolite inactivated the enzymes themselves, probably by the covalent binding to the enzymes. Further studies with radiolabeled dihydralazine are required to confirm the formation of the covalent adducts.