Covalent Binding of a Reactive Metabolite Derived from Propranolol

Covalent Binding of a Reactive Metabolite Derived from. Propranolol and Its Active Metabolite. 4-Hydroxypropranolol to Hepatic Microsomal Proteins of ...
0 downloads 0 Views 248KB Size
Download PDF
Chem. Res. Toxicol. 1997, 10, 289-295

289

Covalent Binding of a Reactive Metabolite Derived from Propranolol and Its Active Metabolite 4-Hydroxypropranolol to Hepatic Microsomal Proteins of the Rat Shizuo Narimatsu,*,† Takayuki Arai,† Toshiyuki Watanabe,† Yasuhiro Masubuchi,† Toshiharu Horie,† Tokuji Suzuki,† Tsutomu Ishikawa,‡ Michio Tsutsui,§ Yoshito Kumagai,| and Arthur K. Cho⊥ Laboratories of Biopharmaceutics and Medicinal Organic Chemistry, Faculty of Pharmaceutical Sciences, Chiba University, 1-33 Yayoi-cho, Inage-ku, Chiba 263, Japan, Central Laboratory, Amersham K.K., 2082-1 Hiratsuka, Shiroi-machi, Inba-gun, Chiba 270-14, Japan, Institutes of Community Medicine, Tsukuba University, 1-1-1 Tennoudai, Tsukuba 305, Japan, and Department of Molecular and Medical Pharmacology, University of California, Los Angeles, School of Medicine, Los Angeles, California 90024 Received September 17, 1996X

Repeated administration of propranolol (PL) to rats causes the inhibition of cytochrome P4502D (P450-2D) enzyme. We recently found that 4-hydroxypropranolol (4-OH-PL) was biotransformed to 1,4-naphthoquinone (1,4-NQ) by superoxide (SO) anions in medium containing rat liver microsomes and NADPH and proposed that the binding of the quinone to P450-2D apoproteins might be one of mechanisms for the enzyme inhibition [Narimatsu et al. (1995) Chem. Res. Toxicol. 8, 721-728]. In this study, we have searched for possible sources of SO for the conversion of 4-OH-PL to 1,4-NQ in rat liver microsomes and determined the radioactivity covalently bound to microsomal proteins after incubation of radioactive PL and 4-OH-PL with rat liver microsomes. Elimination of 4-OH-PL from a mixture containing microsomes and NADPH was suppressed by carbon monoxide. Antibodies raised to P4502B1 and -3A2 partially, and antibody against NADPH-cytochrome P450 reductase (fp2) markedly suppressed the reaction. 1,4-NQ was formed concomitantly with 4-OH-PL elimination by a reconstituted preparation of fp2. Binding studies using naphthalene ring (NR)- and side chain (SC)-radiolabeled PL and 4-OH-PL showed that radioactivity covalently bound to microsomal proteins was much higher from 4-OH-PL than from PL for the NR-labeled compounds, but higher from PL than from 4-OH-PL for the SC-labeled compounds. These results suggest that the 4-OH-PL formed from PL by P450-2D enzyme is converted to 1,4-NQ with loss of the side chain, and the 1,4-NQ accounts for most of the radioactivity covalently bound to microsomal proteins, including the P450-2D enzymes. The SO for conversion of 4-OHPL to 1,4-NQ is supplied mainly by fp2 with some contribution by P450 enzymes.

Introduction Propranolol (PL)1 is widely used for the treatment of hypertension or arrhythmias as an adrenoceptor blocking agent. This drug is known to be a substrate of cytochrome P450-2D6 (P450-2D6) (1-3), a key enzyme in the debrisoquine/sparteine-type genetic polymorphism (4, 5). Clinically, the debrisoquine/sparteine-type genetic polymorphism has been observed in PL 4-hydroxylation (68). Therefore, it is generally thought that PL and other drugs that are also substrates of the P450-2D subfamily competitively inhibit the oxidative metabolism of each other, resulting in increases in the blood levels of PL and/ or the other drug(s) coadministered. * To whom correspondence should be addressed. † Laboratory of Biopharmaceutics, Chiba University. ‡ Laboratory of Medicinal Organic Chemistry, Chiba University. § Amersham K.K. | Tsukuba University. ⊥ University of California, Los Angeles. X Abstract published in Advance ACS Abstracts, February 15, 1997. 1 Abbreviations: PL, propranolol; x-OH-PL, x-hydroxypropranolol; 1,4-NQ, 1,4-naphthoquinone; G-6-P, glucose 6-phosphate; SO, superoxide; P450, cytochrome P450; fp2, NADPH-cytochrome P450 reductase; NR, naphthalene ring; SC, side chain; DLPC, dilauroylphosphatidylcholine; DA, Dark Agouti.

S0893-228x(96)00165-8 CCC: $14.00

On the other hand, it was reported that repeated administration of PL to humans resulted in the accumulation of the parent drug in the body (9, 10) suggesting that the hepatic drug-metabolizing enzymes may be inhibited by PL itself. Consistent with this notion, Schneck and Prichard (11) observed that pretreatment of rats with PL led to the inhibition of PL metabolism both in vivo and in vitro. Shaw et al. (12) reported that reactive metabolite(s) covalently bound to microsomal proteins after incubation of human liver microsomes and PL in the presence of an NADPHgenerating system. If PL were to inhibit P450-2D enzymes by some mechanism other than competitive inhibition, therapy involving the combination of PL and other drug(s) as substrates of the P450-2D subfamily could have major drug interaction consequences. In this context, we have examined the inhibitory effects of PL on the drug-metabolizing enzymes in liver microsomes from rats pretreated with PL, and found that covalent binding of a further metabolite of 4-hydroxypropranolol (4-OH-PL) might be responsible for the inhibition of cytochrome P450-2D enzymes (13-16). 4-OH-PL is a major metabolite of PL in the rat (17, 18), and is an active metabolite of PL, with β-blocking © 1997 American Chemical Society

290 Chem. Res. Toxicol., Vol. 10, No. 3, 1997

Figure 1. Chemical structures of radiolabeled propranolols (PLs) and 4-hydroxypropranolols (4-OH-PLs) used. NR, naphthalene ring; SC, side chain.

activity of the same order as that of the parent drug (19). Recent studies in our laboratory have shown that 4-OHPL was converted to 1,4-naphthoquinone (1,4-NQ) by superoxide (SO) anions in a reaction system containing rat liver microsomes (20). We speculate that PL is oxidized to 4-OH-PL by P4502D enzymes, and that this metabolite is converted by SO to 1,4-NQ, which covalently binds to P450-2D apoprotein, resulting in the inhibition of the enzyme (20). In order to further support this hypothesis, we performed binding studies using radioactive PL and 4-OH-PL and microsomal protein of rat livers (Figure 1). Furthermore, we have examined possible sources of SO in rat liver microsomes using antibodies raised against P450 enzymes and NADPH-cytochrome P450 reductase (fp2). The results, reported here, suggest that most of the protein adducts derived from 4-OH-PL occurs via 1,4-NQ and the SO needed for the conversion of 4-OH-PL to 1,4-NQ is mainly supplied from fp2.

Materials and Methods Chemicals. PL HCl and dilauroylphosphatidylcholine (DLPC) were obtained from the Sigma Chemical Co. (St. Louis, MO); 3′-[14C]-PL (3.81 mCi/mmol) and 4-OH-PL HCl were from Sumitomo Chemical Ind. (Osaka, Japan); [14C]-1-naphthol (59 mCi/mmol) was from Amersham (Tokyo, Japan); [3H]propranolol HCl labeled at the 4-position (21 Ci/mmol) was from Amersham International (Amersham, U.K.); 1,4-NQ, epichlorhydrin, and isopropylamine were from Kanto Chemical Co. (Tokyo, Japan). Animals. Female Japanese white rabbits (2.5 kg body weight, 3 months old) and male and female Wistar rats (8 weeks old) were obtained from Takasugi Experimental Animals (Saitama, Japan); male Dark Agouti (DA) rats (8 weeks old) were from SLC (Shizuoka, Japan). Sodium phenobarbital (dissolved in saline, 80 mg/kg, daily for 3 days) or β-naphthoflavone (dissolved in corn oil, 80 mg/kg, daily for 3 days) was intraperitoneally administered to male Wistar rats. The animals were killed by decapitation 30 h after the final dose, and liver microsomal fractions were prepared by the method of Omura and Sato (21). Vehicle treated control animals were processed in parallel. Preparation of 3-[3H]-4-OH-PL. 3-[3H]-4-OH-PL was biochemically prepared as follows: 4-[3H]-PL (836 nmol, 14.92 µCi) was added to a reaction mixture consisting of 1.2 mmol of G-6P, 0.96 mmol of MgCl2, 250 U of G-6-P dehydrogenase, 0.6 mmol

Narimatsu et al. of vitamin C, 160 mg of protein β-naphthoflavone-induced microsomes, and potassium phosphate buffer (pH 7.4, 18.5 mmol) in a final volume of 120 mL. The reaction was started by adding NADP (62.5 µmol) and continued at 37 °C for 75 min. The reaction was terminated by adding 62.5 mL of 1 N NaOH containing 5% NaHSO3 and 125 mL of 1 M carbonate buffer (pH 9.6), and the mixture extracted with ethyl acetate (200 mL × 3). The combined extracts were dried over anhydrous Na2SO4, and the organic layer was evaporated in vacuo. The residue was dissolved in 0.5 mL of methanol and subjected to preparative TLC using silica gel plates (20 × 20 cm, 0.25 mm thickness) and a solvent system of CHCl3-acetone-aqueous NH3-methanol (5:5:0.4:0.5 v/v). Development was performed in a glass tank placed at a dark chamber to avoid photodegradation of 4-OH-PL. A pale dark band corresponding to a spot of authentic 4-OH-PL (Rf 0.33) visualized with UV lamp on the plate was scraped off, and extracted with benzene-ethanoldiethylamine (5:4:0.2 v/v, 5 mL × 2). The pooled extracts were evaporated in vacuo, and the residue was dissolved in 1 mL of methanol. A portion of the sample was examined by TLC and HPLC. Because the sample contained some amounts of other PL metabolites such as 5-OH-PL and 7-OH-PL, it was further purified by HPLC. The methanolic solution was evaporated, and the residue was dissolved in 250 µL of an HPLC mobile phase (acetonitrile-methanol-water-acetic acid ) 20:20:60:1 v/v). Aliquots of 100 µL of the solution were injected into the HPLC and the peak corresponding to 4-OH-PL standard (tR ) 4.14 min) was collected. To the pooled eluate containing this fraction was added 1 N NaOH containing 5% NaHSO3 and 1 M carbonate buffer (pH 9.6), and 4-OH-PL was extracted with ethyl acetate in a similar manner as described above. The final 3-[3H]-4-OH-PL (66 nmol, 0.84 µCi) was obtained in the total yield of 7.9%. The purity of the final sample was over 96.3% in HPLC and TLC. Preparation of Naphthalene Ring (NR)-[14C]-PL. The total synthesis of [14C]-PL was accomplished by published procedures (22). Thus, [14C]-1-naphthol (19.4 µmol, 1.19 mCi) and cold 1-naphthol (1.73 mmol) were dissolved in 10 mL of N,Ndimethylformamide containing K2CO3 (2 mmol) and epichlorhydrin (5.10 mmol). The reaction mixture was stirred at 60 °C for 24 h and then extracted with diethylether (30 mL × 3). The organic solvent was evaporated, and the residue was dissolved in 1 mL of the solvent (n-hexane-ethyl acetatemethanol ) 8:1:0.5 v/v) followed by column chromatography. This manipulation gave [14C]-3-(1-naphthoxy)-1,2-epoxypropane (0.65 mCi, 0.944 mmol, yield 54.0%). To a portion of the product (0.13 mCi, 189 µmol) dissolved in toluene (2 mL) was added isopropylamine (14.1 mmol), and the mixture was stirred in a sealed vessel (20 mL) at 80 °C for 18 h. The mixture was evaporated in vacuo, and the residue was dissolved in ethyl acetate (3 mL) and extracted with 1 N HCl (10 mL x 3). To the combined 1 N HCl solution was added 15 mL of 5 N NaOH, and extracted with ethyl acetate (50 mL × 2). The organic layer was evaporated in vacuo, and the residue was dissolved in 10 mL of methanol. A small part of the solution was subjected to HPLC to confirm the preparation of PL, showing a single peak with a retention time of 3.48 min that coincided with that of authentic PL standard. Silicagel TLC also showed a single spot corresponding to PL standard. NR-[14C]-PL thus obtained was 159 µmol [109 µCi, yield 84.1% from 3-(1-naphthoxy)-1,2epoxypropane, 95.8% of purity in HPLC]. Preparation of NR-[14C]-4-OH-PL or Side Chain (SC)[14C]-4-OH-PL. NR-[14C]-4-OH-PL and SC-3′-[14C]-4-OH-PL were prepared from NR-[14C]-PL and SC 3′-[14C]-PL, respectively, by the same biochemical procedure described for the preparation of 3-[3H]-4-OH-PL. The total yield of NR-[14C]-4OH-PL from [14C]-1-naphthol was 5.4%, and that of SC-3′-[14C]4-OH-PL from SC-[14C]-PL was 11.2%. The purities of these compounds were confirmed to be at least above 95% by HPLC and TLC. Measurement of 4-OH-PL Elimination. The elimination of 4-OH-PL from the microsomal incubation mixture was determined by a previously published method (20). Briefly, a

Propranolol Metabolite Binds to Microsomal Protein typical incubation mixture (1.0 mL) contained 25 nmol of 4-OHPL, 1 mg of microsomal protein, 10 µmol of G-6-P, 2 units of G-6-P dehydrogenase, 8 µmol of MgCl2, and 100 µmol of potassium phosphate buffer (pH 7.4). After preincubation at 37 °C for 5 min, the reaction was started by adding NADPH and stopped 5 min later by adding 1 mL of 1 N NaOH including sodium bisulfite (25 mg) to avoid degradation of 4-OH-PL. The remaining 4-OH-PL was determined by the HPLC method reported previously (23). The effect of gas phases on 4-OH-PL elimination was studied with Warburg-type flasks (volume of 10 mL). Binding Studies. A typical reaction mixture (1.0 mL) contained the same ingredients as described above except for the radiolabeled substrate (4-OH-PL; 25 nmol, ca. 20 00030 000 dpm). After preincubation at 37 °C for 5 min, the reaction was started by adding NADPH and terminated 5 min later by addition of 5 mL of 10% aqueous TCA solution and vigorously mixing. After centrifugation at 2000g for 10 min, the pellet obtained was washed successively with each 4 mL × 2 of 7.5% TCA, 80% methanol, hot 80% methanol (70 °C), a mixture of methanol and diethyl ether (1:1 v/v), and 80% methanol. The thoroughly extracted pellet was solubilized in 0.5 mL of 1 N NaOH, then neutralized with 0.5 mL of 1 N HCl, mixed with 10 mL of scintillation medium, and the radioactivity was measured by an LS-1800 Beckman liquid scintillation counter. The scintillation medium used consisted of 1 volume of Triton X-100 and 2 volumes of toluene phosphor including 4 g of 2,5-diphenyloxazole and 100 mg of 1,4-bis[2-(4-methyl-5phenyloxazolyl)]benzene per 1000 mL of toluene. To confirm the stability of the prepared 3-[3H]-4-OH-PL, its known amounts were added to the reaction mixture containing heat-treated rat liver microsomes, being followed by extraction with ethyl acetate under alkaline conditions as described above. HPLC analysis and liquid scintilation counting revealed that 94-97% of the 3-[3H]-4-OH-PL added was recovered into the organic solvent. This stability of the tritium indicated that the tritium was located on the aromatic ring (probably at 3-position of the naphthalene ring). The stability of NR- and SC-[14C]-4-OHPLs was also confirmed in the same manner, showing satisfactory recoveries (93-97%). Other Procedures. Fp2 purified from liver microsomes of rats (24) and its polyclonal antibody [immunoglobulin G (IgG) fraction] were kindly supplied by Dr. M. Kitada, Chiba University Hospital. Antibodies (IgG fractions) against P450-2C11 (25) and -2D2 (26) were obtained by immunizing female Japanese white rabbits reported previously. Antibodies raised to P4502B1 and -3A2 were kindly supplied by Drs. S. Imaoka and Y. Funae, Osaka City University School of Medicine. Protein concentrations were measured by the method of Lowry et al. (27) using bovine serum albumin as standard. Statistical significance was calculated by the Student’s t-test.

Results Effects of Gas Phases on PL Elimination. As reported previously (20), 60-80% of the 4-OH-PL added (25 µM) disappeared from a reaction medium containing control rat liver microsomes (1 mg of protein) in a 5 min incubation at 37 °C in the presence of NADPH. The reaction was performed under air. We then examined the effects of altered gas phases on 4-OH-PL elimination with nitrogen (N2, 100%) and with carbon monoxide (CO, CO:O2 ) 4:1 v/v). As shown in Figure 2, N2 (100%) suppressed 4-OH-PL elimination almost completely, whereas CO suppressed the activity by 65% (p < 0.05). Effects of P450 Inducers on 4-OH-PL Elimination. The effects of P450 inducers on the reaction was examined using an incubation time of 2 min, during which 45% of 4-OH-PL disappeared in control liver microsomes (Figure 3). When liver microsomal fractions from rats pretreated with phenobarbital (a P450-2B1/2 inducer) instead of control rats were used, the rate of 4-OH-PL

Chem. Res. Toxicol., Vol. 10, No. 3, 1997 291

Figure 2. Effects of gas phases of the reaction on the elimination of 4-OH-PL in the medium containing rat liver microsomes. The reaction medium (1.0 mL) consisted of 25 nmol of 4-OHPL, 1 mg of microsomes, 10 µmol of G-6-P, 8 µmol of MgCl2, 2 units of G-6-P dehydrogenase, and 100 µmol of potassium phosphate buffer (pH 7.4). Control experiment was performed under air. Each value represents the mean ( SD of 4 determinations using four different microsomal fractions. *,**Significantly different from control (p < 0.05 and p < 0.01, respectively).

Figure 3. Effects of P450 inducers on 4-OH-PL elimination in rat liver microsomes. Rats were pretreated intraperitoneally with sodium phenobarbital (in saline, 80 mg/kg) or β-naphthoflavone (in corn oil, 80 mg/kg) daily for 3 days, and killed 30 h after final dosing to prepare liver microsomes. Control was saline-treated animal. PB, phenobarbital-treated; β-NF, β-naphthoflavone-treated. Each value represents the mean ( SD of 4 determinations using four different microsomal fractions. **Significantly different from control (p < 0.01).

elimination was increased to 1.8-fold that of control, whereas β-naphthoflavone-induced microsomes did not show any significant change (Figure 3). Effects of Antibodies against P450 Enzymes in Control and Phenobarbital-Induced Rat Liver Microsomes. After preincubation with control microsomes, antibodies against P450-2C11 (data not shown) or -2D2 (Figure 4A) did not suppress 4-OH-PL elimination. On the other hand, antibodies raised against P450-2B1 and -3A2 significantly suppressed the activity in phenobarbital-induced microsomes, and the latter seemed slightly more potent than the former in the suppression (Figure 4B). Involvement of Fp2 in 4-OH-PL in Rat Liver Microsomes. Addition of antibody raised against fp2 to the incubation mixture substantially decreased the rate of 4-OH-PL (25 µM) elimination in control microsomes (Figure 4C). We then examined the ability of fp2 to eliminate 4-OH-PL in a reconstituted system containing

292 Chem. Res. Toxicol., Vol. 10, No. 3, 1997

Narimatsu et al.

Figure 4. Effects of antibodies raised to P450-2D2, -2B1, -3A2, and fp2 on 4-OH-PL elimination in liver microsomes from control and phenobarbital-pretreated rats. Liver microsomes from control rats (A and C) and phenobarbital-treated (80 mg/kg, ip, daily for 3 days) rats (B) were used. Immunoglobulin G fractions from control or immunized rabbits with rat P450 enzymes or fp2 were preincubated with the microsomal fractions at 25 °C for 30 min. To the medium were then added 4-OH-PL and an NADPH-generating system, and incubated at 37 °C for 5 min. Preimmune, preimmune IgG was added; anti-2D2, -2B1, -3A2, and fp2 IgG fractions raised to P450-2D2, -2B1, -3A2, and fp2 were added. Each value represents the mean ( SD of 4 determinations using four different microsomal fractions. *,**Significantly different from preimmune control (p < 0.05 and p < 0.01, respectively).

the reductase, DLPC, and NADPH. Under these conditions, the reductase converted 4-OH-PL to 1,4-NQ, and the activities for 4-OH-PL elimination and 1,4-NQ formation were calculated to be 129 and 71 nmol/(mg of protein‚5 min), respectively. Consumption of 3-[3H]-4-OH-PL and Binding of Radioactive Materials to Microsomal Proteins. Control microsomes (1 mg of protein) were incubated with 3-[3H]-4-OH-PL (25 µM) in the presence of an NADPHgenerating system at 37 °C for 5 and 10 min, and the amounts of the substrate remaining in the reaction mixture and radioactivities covalently bound to microsomal proteins were assessed. During the incubation, 73% (5 min) and 81% (10 min) of 3-[3H]-4-OH-PL disappeared from the medium, and 1.8% (5 min) and 3.3% (10 min) of the consumed substrate were found to covalently bind to microsomal proteins. Binding of Radioactive Materials Derived from Radiolabeled PLs and 4-OH-PLs to Microsomal Proteins. NR-[14C]-4-OH-PL or SC-[14C]-4-OH-PL (each 25 µM) was incubated with control microsomes and the NADPH-generating system for 5 min at 37 °C. Binding activities of radioactive materials to microsomal proteins are summarized in Table 1. The radioactivities derived from NR-and SC-labeled 4-OH-PLs were calculated to be 19.6% and 1.4%, respectively, of consumed substrates (p < 0.01). We then assessed binding of radioactivity from labeled PL to microsomal proteins. [14C]-PL or 4-[3H]PL was incubated with microsomes and the NADPHgenerating system, and covalently bound radioactivities were determined in a similar manner. The radioactivities derived from NR-[14C]-4-OH-PL were much higher than those from NR-[14C]-PL (Table 1). Similar results were obtained with 4-[3H]-PL and 3-[3H]-4-OH-PL. However, the bound radioactivity derived from SC-[14C]-PL was higher than that from SC-[14C]-4-OH-PL (Table 1). In addition, binding activities of radiolabeled 4-OH-PLs were significantly higher than those of the corresponding PLs in the control without NADPH (Table 1-B), although the activities of the radiolabeled 4-OH-PLs were much higher in the presence of NADPH than in the absence of NADPH. Comparison of Protein-Bound Radioactivities Derived from [14C]-4-OH-PL between Wistar and

Table 1. Comparison of Radioactivity Covalently Bound to Microsomal Proteins after Incubation of Radiolabeled PLs and 4-OH-PLs with Rat Liver Microsomesa binding activity (nmol/(5 min‚mg of protein)) labeled compd

(A) NADPH-(+)

(B) NADPH-(-)

(A) - (B)

4-[3H]-PL 3-[3H]-4-OH-PL NR-[14C]-PL NR-[14C]-4-OH-PL SC-[14C]-PL SC-[14C]-4-OH-PL

0.220 ( 0.010 0.762 ( 0.095 0.695 ( 0.024 3.273 ( 0.501 0.193 ( 0.004 0.121 ( 0.050

0.011 ( 0.004 0.125 ( 0.020** 0.020 ( 0.005 0.391 ( 0.033** 0.013 ( 0.001 0.070 ( 0.045*

0.208 ( 0.003# 0.638 ( 0.013**,# 0.674 ( 0.027 2.873 ( 0.352** 0.180 ( 0.004# 0.051 ( 0.005**,#

a [3H]- or [14C]-labeled PLs and 4-OH-PLs (20 000-30 000 dpm) were incubated with microsomes (1 mg of protein) in the presence or absence of an NADPH-generating system at 37 °C for 5 min. Radioactivity covalently bound to microsomal proteins was determined as described in Materials and Methods. Each value represents the mean ( SD of 4 determinations using four different microsomal fractions. NADPH-(+), with NADPH; NADPH-(-), without NADPH. *,**Significantly different from corresponding PLs (p < 0.05 and p < 0.01, respectively). #Significantly different from corresponding naphthalene ring-radiolabeled compounds (NR-[14C]) (p < 0.01).

DA Rats. NR-[14C]-4-OH-PL and SC-[14C]-4-OH-PL were incubated with the NADPH-generating system and liver microsomes from male DA rats, used as a P450-2D enzyme-deficient animal model, and Wistar rats, used as a normal control. As shown in Figure 5, the radioactivities derived from NR-[14C]-4-OH-PL were significantly higher in Wistar rats than in DA rats. On the other hand, the radioactivities derived from SC-[14C]-4-OH-PL were much lower than those from NR-[14C]-4-OH-PL, and there was no statistical significance in the radioactivities between Wistar and DA rats.

Discussion We have recently reported that PL is oxidized mainly by P450-2D enzyme(s) to 4-OH-PL, which is then biotransformed to 1,4-NQ by SO in rat liver microsomes (20). The present study was conducted to identify the source of SO for this biotransformation in rat liver microsomes. It has been reported that the SO found in liver microsome mixtures is generated from P450 enzymes and fp2 (2831). In the first step of the present study, we examined the possible involvement of P450 and the reductase in the microsomal transformation from 4-OH-PL to 1,4-NQ using as an index the elimination of 4-OH-PL.

Propranolol Metabolite Binds to Microsomal Protein

Figure 5. Comparison of microsomal protein-bound radioactivities derived from [14C]-4-OH-PLs between Wistar and Dark Agouti rats. Radiolabeled 4-OH-PLs (25 µM) were incubated with liver microsomes (1 mg of protein) from male Wistar and DA rats and an NADPH-generating system. Radioactivities covalently bound to microsomal proteins were determined as described in Materials and Methods. Each value represents the mean ( SD (n ) 4). A, NR-[14C]-4-OH-PL; B, SC-[14C]-4-OHPL. *Significantly different from Wistar rats (p < 0.025).

In the studies of the contribution of P450, a reaction gas phase of CO/O2 (4:1) decreased the rate of 4-OH-PL elimination by 65%. Pretreatment of rats with phenobarbital increased 4-OH-PL elimination but β-naphthoflavone had no effect. Pretreatment of rats with phenobarbital is known to increase fp2 content as well as P450 content in liver micosomes (32). Since electrons are transferred to P450 as the terminal oxidase via fp2 in the microsomal electron transfer system (33), we cannot assess the participation of P450 enzymes from the results of the induction study. Although antibodies raised against P450-2C11 or -2D2 did not change the elimination rate in liver microsomes from untreated rats, antibodies against P450-2B1 and -3A2 decreased it slightly but significantly. It is known that not only P4502B1/2 but also P450-3A2 is induced in the rat by phenobarbital (34). These results suggest that some P450 enzymes such as the 2B and/or 3A subfamily may be possible sources of SO. In studies on the involvement of fp2, antibody to the reductase suppressed 4-OH-PL elimination by 44%. The addition of the antibody raised against the reductase could block the production of SO both by the reductase

Chem. Res. Toxicol., Vol. 10, No. 3, 1997 293

and P450 enzymes. Thus, the 44% reduction could reflect the total suppression of SO production by both the reductase and P450. To further assess the role of the reductase, we reconstituted a SO producing system with purified fp2 and DLPC. The system efficiently converted 4-OH-PL to 1,4-NQ. From these results, it seems likely that fp2 and P450 enzyme(s) in rat liver microsomes are the sources of SO invoolved in the conversion of 4-OHPL to 1,4-NQ. In the second phase of the study, we performed binding studies using radiolabeled 4-OH-PLs to assess how much of the reactive metabolite(s) derived from 4-OH-PL covalently binds to microsomal proteins. We first prepared 3-[3H]-4-OH-PL from 4-[3H]-PL, which is the only radiolabeled PL commercially available. We chose a biochemical method to prepare 3-[3H]-4-OH-PL: by incubation of 4-[3H]-PL with β-naphthoflavone-induced microsomes and the NADPH-generating system. The 3-[3H]-4-OH-PL formed via the NIH shift was purified by TLC and HPLC. The percentage of the NIH shift occurring was estimated to be 71%. The binding experiments using 4-[3H]-PL and 3-[3H]4-OH-PL showed that radioactivity derived from 4-OHPL was significantly higher than that from PL. A similar result was obtained in the binding study using NR-[14C]PL and NR-[14C]-4-OH-PL. From the results, it appears that a reactive metabolite that covalently binds to microsomal proteins is formed more easily from 4-OHPL than from PL. Conversely, radioactivity derived from SC-[14C]-4-OH-PL was lower than that from SC-[14C]-PL with a statistical significance. In addition, radioactivity derived from SC-[14C]-PL was about one fourth that from NR-[14C]-PL. The results indicate that about three fourths of the covalently bound radioactive materials derived from 4-OH-PL lost their side chains during oxidative metabolism and covalent binding. In the binding studies, we ran negative control experiments without NADPH in parallel. Interestingly, binding activities of radiolabeled 4-OH-PLs were significantly higher than those of the corresponding radiolabeled parent compounds in the absence of NADPH, although the activities of the radiolabeled 4-OH-PLs were much higher in the presence of NADPH than in the absence of NADPH. As a possibility, endogenous NADPH remaining in microsomal fractions, even if in a small amount, may give rise to the results.

Figure 6. Possible mechanisms for superoxide mediated formation of 1,4-NQ from 4-OH-PL.

294 Chem. Res. Toxicol., Vol. 10, No. 3, 1997

Figure 7. Possible mechanisms for the release of tritium from 4-[3H]-PL and 3-[3H]-4-OH-PL during the incubation of these compounds with rat liver microsomes.

The conversion of 4-OH-PL to NQ can be mediated by SO in the absence of P450, indicating that SO alone can effect the reaction as well as P450. There are two possible mechanisms by which SO can effect the conversion of 4-OH-PL to NQ. As SO is a weak base (pKa 4.5) (35) and a weak oxidant, it could act by ionizing an acidic proton or by removal of a hydrogen atom. However, 4-OH-PL is extracted into ethyl acetate from strongly alkaline solutions without conversion to 1,4-NQ (see Materials and Methods) so the ionization pathway is unlikely. Thus, the conversion could occur by the pathway shown in Figure 6, in which SO is abstracting a hydrogen atom. First, SO could abstract a hydrogen atom from 4-OHPL (A) to generate the semiquinone ether shown (B). This ether could then undergo an electron abstraction or transfer to SO together with hydroxide attack at the ether carbon to generate the 1,4-NQ and the amino diol (E) shown. In this manner, 1,4-NQ would be generated by a sequential two electron oxidation mediated by 2 equivalents of SO which would be reduced to two equivalents of hydrogen peroxide. There are two characteristics of the reaction, first that hydrogen peroxide is generated, i.e., the formation of hydrogen peroxide would increase with addition of 4-OH-PL to the reaction mixture. Second, that the side chain alcohol would result from oxidative cleavage of the ether link so that the aldehyde that might be expected from a P450 type dealkylation would not be present. To further confirm this speculation, we are performing further experiments for identification and determination of a side chain fragment.

Narimatsu et al.

As described above our results indicate that radioactivity derived from [3H]-labeled PL and 4-OH-PL was much lower than that from NR-[14C]-labeled PL and 4-OH-PL. We speculate possible causes for the difference as shown in Figure 7. 4-[3H]-PL was biotransformed to 4-OH-PL by P450-2D enzymes via 3,4-epoxy-PL. About 70% of tritium at the 4-position migrated to the 3-position by the NIH shift, and about 30% of the radioisotope was lost during the reaction. The 3-[3H]-4-OH-PL thus formed is converted to 3-[3H]-1,4-NQ by a mechanism(s) such as that shown in Figure 6, and the quinone is attacked typically by a nucleophile such as the SH group of cysteine at the 3- or 4-position. There are two pathways for thiol addition to [3H]-1,4NQ (Figure 7). In path a, addition takes place at the C-T carbon (1) followed by tautomerization to the 1,4addition product (3). The second path, b, has the addition occuring at the C-H carbon (2), and the tritium is retained. If the addition was random, i.e., the rates of a and b would be the same, the mixture would have the tritiated and protonated product in equivalent amounts. Then, the ratio of [14C]-1,4-NQ- to [3H]-1,4-NQ-based covalent product would be 2:1. However, the observed ratio is actually 5:1, suggesting that the nature of the addition is more complex, i.e., that either reaction path a is favored or tritium is lost by another reaction. As shown in Figure 5, the equivalents of bound radioactivity derived from SC-[14C]-PL were one fourth those of NR-[14C]-PL, suggesting that either the side chain is retained in some of the bound material or that the side chain itself is covalently bound. One possibility is that covalent bonds may be formed directly from epoxides of PL. It should be noted that the radioactivity derived from NR-[14C]-4-OH-PL was significantly lower in male DA rats which exhibit reduced P450-2D activity compared to Wistar rats (36-38). The radioactivity derived from SC-[14C]-4-OH-PL was only 2% that from NR-[14C]-4-OHPL in control microsomes. Furthermore, no difference was observed in the bound radioactivity from SC-[14C]4-OH-PL between Wistar and DA rats (see Figure 5). These results suggest that most of 4-OH-PL formed from PL by P450-2D enzymes was converted to 1,4-NQ, and P450-2D enzymes may play some role in the protein binding of the quinone. The results obtained from the present study indicate that 1,4-NQ and/or epoxy metabolites of PL or 4-OH-PL are possible candidates for reactive metabolites responsible for the protein binding. The metabolite that causes P450-2D enzyme inhibition during chronic PL treatment remains to be identified and experiments addressing this question are in progress.

Acknowledgment. We are very grateful to Dr. Y. Funae, Osaka City University School of Medicine, and Dr. M. Kitada, Chiba University Hospital, for their kind gifts of antibodies raised to rat P450-2B1 and -3A2, and those raised to rat fp2, respectively.

References (1) Otton, S. V., Inaba, T., and Kalow, W. (1984) Competitive inhibition of sparteine oxidation in human liver by β-adrenoceptor antagonists and other cardiovascular drugs. Life Sci. 34, 73-80. (2) Distlerath, L. M., and Guengerich, F. P. (1984) Characterization of a human liver cytochrome P-450 involved in the oxidation of debrisoquine and other drugs by using antibodies raised to the analogues rat enzymes. Proc. Natl. Acad. Sci. U.S.A. 81, 73487352.

Propranolol Metabolite Binds to Microsomal Protein (3) Otton, S. V., Gillam, E. M. J., Lennard, M. S., Tucker, G. T., and Woods, H. F. (1990) Prorpanolol oxidation by human liver microsomes - the use of cumene hydroperoxide to probe isoenzyme specificity and regio- and stereoselectivity. Br. J. Clin. Pharmacol. 30, 751-760. (4) Eichelbaum, E., and Gross, A. S. (1990) The genetic polymorphism of debrisoquine/sparteine metabolism 9 clinical aspects. Pharmacol. Ther. 46, 377-394. (5) Gonzalez, F. J., and Meyer, U. A. (1991) Molecular genetics of the debrisoquine/sparteine polymorphism. Clin. Pharmacol. Ther. 50, 233-238. (6) Lennard, M. S., Jackson, P. R., Freestone, S., Tucker, G. T., Ramsay, L. E., and Woods, H. F. (1984) The relationship between debrisoquine oxidation phenotype and pharmacokinetics and pharmacodynamics of propranolol. Br. J. Clin. Pharmacol. 17, 679-685. (7) Raghuram, T. C., Koshakji, P. K., Wilkinson, G. R., and Wood, A. J. J. (1984) Polymorphic ability to metabolize propranolol alters 4-hydroxypropranolol levels but not bata blockade. Clin. Pharmacol. Ther. 36, 51-56. (8) Ward, S. A., Walle, T., Walle, U. K., Wilkinson, G. R., and Branch, R. A. (1989) Propranolol’s metabolism is determined by both mephenytoin and debrisoquine hydroxylase activities. Clin. Pharmacol. Ther. 45, 72-79. (9) Evans, G. H., and Shand, D. G. (1973) Disposition of propranolol. V. Drug accumulation and steady-state concentrations during chronic oral administration in man. Clin. Pharmacol. Ther. 4, 487-493. (10) Straka, R. J., Lalonde, R. L., Pieper, J. A., Bottorff, M. B., and Mirvis, D. M. (1987) Nonlinear pharmacokinetics of unbound propranolol after oral administration. J. Pharm. Sci. 76, 521524. (11) Schneck, D. W., and Prichard, J. F. (1981) The inhibitory effect of propranolol pretreatment on its own metabolism in the rats. J. Pharmacol. Exp. Ther. 218, 575-581. (12) Shaw, L., Lennard, M. S., Tucker, G. T., Bax, N. D. S., and Woods, H. F. (1987) Irreversible binding and metabolism of propranolol by human liver microsomessRelationship to polymorphic oxidation. Biochem. Pharmacol. 36, 2283-2288. (13) Masubuchi, Y., Fujita, S., Chiba, M., Kagimoto, N., Umeda, S., and Suzuki, T. (1991) Impairment of debrisoquine 4-hydroxylase and related monooxygenase activities in the rat following treatment with propranolol. Biochem. Pharmacol. 41, 861-865. (14) Masubuchi, Y., Narimatsu, S., and Suzuki, T. (1992) Activation of propranolol and irreversible binding to rat liver microsomes: strain differences and effects of inhibitors. Biochem. Pharmacol. 43, 635-637. (15) Masubuchi, Y., Suzuki, K., Fujita, S., and Suzuki, T. (1992) A possible mechanism of the impairment of hepatic microsomal monooxygenase activities after multiple administration of propranolol in rats. Biochem. Pharmacol. 43, 757-762. (16) Masubuchi, Y., Narimatsu, S., Hosokawa, S., and Suzuki, T. (1994) Role of the CYP2D subfamily in metabolism-dependent covalent binding of propranolol to liver microsomal protein in rats. Biochem. Pharmacol. 48, 1891-1898. (17) Walle, T., Oatis, J. E., Walle, U. K., and Knapp, D. R. (1982) New ring-hydroxylated metabolites of propranolol. Species differences and stereospecific 7-hydroxylation. Drug Metab. Dispos. 10, 122127. (18) Bargar, E. M., Walle, U. K., Bai, S. A., and Walle, T. (1983) Quantitative metabolic fate of propranolol in the dog, rat, and hamster using radiotracer, high performance liquid chromatography and gas chromatography-mass spectrometry techniques. Drug Metab. Dispos. 11, 266-272. (19) Fitzgerald, D. J., and O’Donnell, S. R. (1971) Pharmacology of 4-hydroxypropranolol. Br. J. Pharmacol. 43, 222-235. (20) Narimatsu, S., Watanabe, T., Masubuchi, Y., Horie, T., Kumagai, Y., Cho, A. K., Imaoka, S., Funae, Y., Ishikawa, T., and Suzuki,

Chem. Res. Toxicol., Vol. 10, No. 3, 1997 295

(21)

(22) (23)

(24)

(25)

(26)

(27)

(28)

(29)

(30)

(31)

(32)

(33)

(34)

(35) (36)

(37)

(38)

T. (1995) Characterization of a chemically reactive propranolol metabolite that binds to microsomal proteins of rat liver. Chem. Res. Toxicol. 8, 721-728. Omura, T., and Sato, R. (1964) The carbon monoxide-binding pigment of liver microsomes. I. Evidence for its hemoprotein nature. J. Biol. Chem. 239, 2370-2378. Crowther, A. F., and Smith, L. H. (1967) U.S. Patent 3337628. Masubuchi, Y., Hososkawa, S., Horie, T., Suzuki, T., Ohmori, S., Kitada, M., and Narimatsu, S. (1994) Cytochrome P450 isozymes involved in propranolol metabolism in human liver microsomes. The role of CYP2D6 as ring-hydroxylase and CYP1A2 as Ndesisopropylase. Drug Metab. Dispos. 22, 909-915. Yasukochi, Y., and Masters, B. S. S. (1976) Some properties of detergent-solubilized NADPH-cytochrome c (cytochrome P-450) reductase purified by biospecific affinity chromatography. J. Biol. Chem. 251, 5337-5344. Fujita, S., Morimoto, R., Chiba, M., Kitani, K., and Suzuki, T. (1989) Evaluation of the involvement of a male specific cytochrome P-450 isozyme in senescence-associated decline of hepatic drug metabolism in male rats. Biochem. Pharmacol. 38, 3925-3931. Suzuki, T., Narimatsu, S., Fujita, S., Masubuchi, Y., Umeda, S., Imaoka, S., and Funae, Y. (1992) Purification and characterization of a cytochrome P-450 isozyme catalyzing bunitrolol 4-hydroxylation in liver microsomes of male rats. Drug Metab. Dispos. 20, 367-373. Lowry, O. H., Rosebrough, N. J., Farr, A. L., and Randall, R. J. (1951) Protein measurement with the Folin phenol reagent. J. Biol. Chem. 193, 265-275. Strobel, H. W., and Coon, M. J. (1971) Effects of superoxide genaration and dismutation on hydroxylation reactions catalyzed by liver microsomal cytochrome P-450. J. Biol. Chem. 246, 78277829. Lai, C. S. T. A., Grover, T. A., and Piette, L. H. (1979) Hydroxyl radical production in a purified NADPH-cytochrome c (P-450) reductase system. Arch. Biochem. Biophys. 193, 373-378. Winston, G. W., and Cederbaum, A. I. (1983) NADPH-dependent production of oxy radicals by purified components of the rat liver mixed function oxidase system. J. Biol. Chem. 258, 1508-1513. Rasgba-Step, J., and Cederbaum, A. I. (1993) Generation of reactive oxygen intermediates by human liver microsomes in the present of NADPH and NADH. Mol. Pharmacol. 45, 150-157. Imaoka, S., Terano, Y., and Funae, Y. (1989) Expression of four phenobarbital-inducible cytochrome P-450s in liver, kidney, and lung of rats. J. Biochem. 105, 939-945. Coon, M. J., Ding, X., Pernecky, S. J., and Vaz, A. D. N. (1992) Cytochrome P450: progress and prediction. FASEB J. 6, 669673. Seidegard, J., Depierre, J. W., Morgenstern, R., Pilotti, A., and Ernster, L. (1981) Induction of drug-metabolizing systems and related enzymes with metabolites and structural analogues of stilbene. Biochim. Biophys. Acta 672, 65-78. Sawyer, D. T., and Valentine, J. S. (1981) How super is superoxide? Acc. Chem. Res. 14, 393-400. Suzuki, T., Fujita, S., Narimatsu, S., Masubuchi, Y., and Umeda, S. (1991) Impairment of bunitrolol 4-hydroxylase activity in liver microsomes of Dark Agouti rats. Biochem. Pharmacol. 42, 22412244. Masubuchi, Y., Umeda, S., Chiba, M., Fujita, S., and Suzuki, T. (1991) Selective 3-hydroxylation deficiency of lidocaine and its metabolite in Dark Agouti rats. Biochem. Pharmacol. 42, 693695. Masubuchi, Y., Kagimoto, N., Narimatsu, S., Fujita, S., and Suzuki, T. (1993) Regioselective contribution of the cytochrome P-450 2D subfamily to propranolol metabolism in rat liver microsomes. Drug Metab. Dispos. 21, 1012-1016.

TX960165E