Chem. Res. Toxicol. 1997, 10, 589-599
589
Metabolism of the Cytochrome P450 Mechanism-Based Inhibitor N-Benzyl-1-aminobenzotriazole to Products That Covalently Bind with Protein in Guinea Pig Liver and Lung Microsomes: Comparative Study with 1-Aminobenzotriazole Kimberley J. Woodcroft,† Chris D. Webb,† Ming Yao,† Alan C. Weedon,‡ and John R. Bend*,† Departments of Pharmacology and Toxicology and of Chemistry, University of Western Ontario, London, Ontario, Canada N6A 5C1 Received November 12, 1996X
The metabolism and covalent binding of radioactivity to microsomal protein of the cytochrome P450 (P450) mechanism-based inhibitors 1-amino-[14C]-2,3-benzotriazole ([14C]ABT) and two radiolabeled forms of N-benzyl-1-aminobenzotriazole (BBT), N-benzyl-1-amino-[14C]-2,3-benzotriazole ([14C]-2,3-BBT) and [14C]-N-7-benzyl-1-aminobenzotriazole ([14C]-7-BBT), were examined in hepatic or pulmonary microsomes from untreated and phenobarbital (PB)- or β-naphthoflavone (βNF)-induced guinea pigs. [14C]-2,3-BBT and [14C]-7-BBT were converted to multiple metabolites including ABT, benzotriazole, benzaldehyde, 2- or 3-hydroxy-BBT, and 4-hydroxy-BBT by hepatic microsomes, while [14C]ABT, whose primary metabolite was benzotriazole, underwent little biotransformation. Neither ABT nor BBT was extensively metabolized by pulmonary microsomes. Hepatic microsomes from βNF (vs PB)-treated guinea pigs metabolized [14C]ABT, [14C]-2,3-BBT, and [14C]-7-BBT more extensively. The degree of NADPH-dependent covalent binding of [14C]-2,3-BBT- or [14C]-7-BBT-derived radioactivity (1.0 nmol/mg of protein) was higher than that of [14C]ABT (0.3-0.8 nmol/mg of protein) in hepatic microsomes, especially those from PB-induced animals. Covalent binding per nmol of P450 in pulmonary microsomes was 3-4-fold higher with [14C]-2,3-BBT (2.9 nmol/nmol of P450) than with [14C]-7-BBT (1.0 nmol/nmol of P450), whereas in hepatic microsomes from PB- or βNFtreated animals the ratio of binding with the two forms of BBT was approximately 1:1. [14C]2,3-BBT- and [14C]-7-BBT-derived radioactivity was covalently bound to proteins that migrated in the molecular weight region corresponding to P450 on SDS-PAGE following incubation with NADPH. These data indicate that BBT is metabolized to at least two reactive compounds capable of covalent modification of protein and/or a single reactive product is formed which contains both the benzo ring (of benzotriazole) and the benzyl carbon atom (of the N-benzyl group); that P450 apoprotein modification may be an important mechanism of inactivation of pulmonary and hepatic P450 by BBT; and that hepatic microsomes from βNF-induced guinea pigs generate more metabolites that do not act as mechanism-based P450 inhibitors from BBT than do those from PB-induced animals.
Introduction The microsomal cytochrome P450 (P450)1 monooxygenase system is composed of multiple isozymes of the hemoprotein P450 and the flavoprotein NADPH-P450 reductase (1, 2). This system catalyzes the oxidation of a wide variety of lipophilic substrates, including both exogenous (drugs, pesticides, carcinogens) and endogenous (fatty acids, fatty vitamins, steroids) compounds (2, 3). The metabolites formed can have less, equal, or * Address all correspondence to this author. Phone: (519) 661-3312. Fax: (519) 661-3797. E-mail:
[email protected]. † Department of Pharmacology and Toxicology. ‡ Department of Chemistry. X Abstract published in Advance ACS Abstracts, May 1, 1997. 1 Abbreviations: P450, cytochrome P450; SDS-PAGE, sodium dodecyl sulfate-polyacrylamide gel electrophoresis; ABT, 1-aminobenzotriazole; BBT, N-benzyl-1-aminobenzotriazole; [14C]ABT, 1-amino[14C]-2,3-benzotriazole; [14C]-2,3-BBT, N-benzyl-1-amino-[14C]-2,3benzotriazole; [14C]-7-BBT, N-[14C]-7-benzyl-1-aminobenzotriazole; 2-OHBBT, 2-hydroxy-BBT; 3-OH-BBT, 3-hydroxy-BBT; 4-OH-BBT, 4-hydroxyBBT; PB, phenobarbital; βNF, β-naphthoflavone; GC/MS, gas chromatography/mass spectrometry; LPC, low-pressure chromatography; HPLC, high-performance liquid chromatography.
S0893-228x(96)00185-3 CCC: $14.00
greater biological activity or toxicity than the parent compound, and consequently the microsomal P450 system is important in pharmacology, physiology, and toxicology. Evaluation of the role played by a specific P450 form in the metabolism of various substrates by single cell types in vivo is complicated by the differential distribution of a large number of P450 isozymes, which often have overlapping substrate selectivities (2, 3). One method of determining the relative catalytic contribution of various P450 isozymes in systems with intact cellular structure is the use of isozyme selective mechanism-based inhibitors (4-6). Several P450 subfamily and/or isozyme selective suicide inhibitors of P450 are known, including N-(2-p-nitrophenethyl)dichloroacetamide (7) and 1-(1propynyl)pyrene (8) for P450 1A; 9-ethynylphenanthrene (9), secobarbital (10), N-cyclopropylbenzylamine (11), phencyclidine (12), and N-(2-p-nitrophenyl)chloroacetamide (13) for P450 2B; 21,21-dichloropregnenolone for P450 2C5 (14); cimetidine for P450 2C11 (15); diallyl sulfone for P450 2E1 (16); spironolactone (17) and some © 1997 American Chemical Society
590 Chem. Res. Toxicol., Vol. 10, No. 5, 1997
thiosteroids (18) for P450 3A; gestodene for P450 3A4 (19); acetylenic fatty acids for P450 4A (20, 21); and 12hydroxy-16-heptadecynoic acid for P450 4A4 (21). 1-Aminobenzotriazole (ABT) is a mechanism-based inhibitor of P450 in hepatic (22, 23), pulmonary (24, 25), and renal (26, 27) microsomes in vitro, as well as in vivo (23, 26-29). ABT, a mechanism-based inhibitor with low isozyme selectivity, inactivates P450 via alkylation of the prosthetic heme moiety as demonstrated by the isolation of a N,N′-bridged phenylene-protoporphyrin IX adduct from livers of rats treated with ABT (22, 30). During P450 inactivation, ABT causes loss of the characteristic P450 absorption spectrum, in contrast to suicide inhibitors which cause covalent modification of the P450 apoprotein and can inhibit monooxygenase activity without concomitant loss of the P450 spectrum (9). Three N-aralkyl derivatives of ABT, N-benzyl-1-ABT (BBT), N-(R-methylbenzyl)-1-ABT (RMB), and N-(R-ethylbenzyl)1-ABT (R-EB), were synthesized in our laboratory and shown to be mechanism-based inhibitors of pulmonary P450 2B isozymes (25, 31). BBT and RMB have been characterized in detail, are potent, selective P450 2B inactivators in vitro (25, 29, 31, 32-34), and are also selective for pulmonary 2B in vivo (27, 29, 34, 35). A phenylene-protoporphyrin IX adduct, identical to that of ABT, was isolated from PB-treated rats administered high doses of BBT (400 mg/kg) in vivo (31), suggesting that N-debenzylation of BBT to ABT and P450-dependent oxidation of ABT to benzyne occurred under these experimental conditions. NADPH-dependent inhibition of P450 2B-catalyzed activity in guinea pig or rabbit hepatic and pulmonary microsomes by BBT and RMB occurs with little or no loss of P450 (25, 29, 33, 35), whereas the inhibition of activity closely parallels P450 loss with ABT in vitro and in vivo (25, 28, 29, 31). The inactivation of recombinant P450 2B4 and 2B5 by BBT and RMB in reconstituted monooxygenase systems demonstrates that other forms of P450 are not required for mechanism-based inhibition by these compounds (32). BBT and RMB also form P450 isozyme selective metabolic intermediate complexes with guinea pig hepatic but not pulmonary microsomes, indicating that P450 2B18, the guinea pig orthologue of P450 2B1 (1) present in lung and liver (34), is not involved in complex formation or inhibition of pulmonary P450 activity by these compounds (36). In concert, these data suggest that a primary mechanism of P450 inactivation by BBT and RMB may be modification of P450 protein by reactive metabolite(s). The objective of this study was to further examine potential mechanisms for the inactivation of P450 2B isozymes by BBT. To accomplish this, metabolism of BBT and covalent binding of BBT-derived radioactivity were studied in guinea pig hepatic and pulmonary microsomes with two radiolabeled forms of BBT synthesized for this purpose: [14C]-2,3-BBT, labeled in the 2,3positions of the benzotriazole ring, and [14C]-7-BBT, labeled at C-7 of the N-benzyl substituent.
Materials and Methods Materials. [14C]-1-Aminobenzotriazole ([14C]ABT; 19.75 mCi/ mmol) was purchased from Pathfinder Laboratories Inc., St. Louis, MO, and purified to >97% chemical and radiochemical purity (analysis in system A as described below) before each experiment by low-pressure chromatography (LPC) on a silica
Woodcroft et al. gel 60 column (Lobar, 310 × 25 mm LiChroprep Si 60, 40-63 µm; Beckman) by elution with EtAc/CHCl3 (3:1). [14C]-7-Benzoic acid (32.3 mCi/mmol) was purchased from ICN, Mississauga, Canada. ABT and BBT were synthesized as previously described (31). NADPH was obtained from Sigma Chemical Co., St. Louis, MO, and β-naphthoflavone (βNF), benzaldehyde, benzotriazole, N,N-carbonyldiimidazole, and lithium aluminum hydride (LiAH4) were from Aldrich Chemical Co., Milwaukee, WI. Phenobarbital sodium (PB), sodium borohydride (NaBH4), benzoic acid, HPLC grade organic solvents, and all other chemicals (reagent grade or better) were purchased from BDH, Toronto, Canada. Synthesis of [14C]BBT. [14C]BBT was synthesized by minor modifications of the procedure used for BBT (31). The product was previously characterized in detail, including by exact mass spectrometry (31). Analysis of chemical and radiochemical purity of [14C]BBT was monitored using two HPLC systems. Normal phase isocratic HPLC on a Waters Resolve C18 Radial-Pak column, 5 µm, eluted with 1 mL/min C6H12/EtAc (6:1) gave a BBT retention time of 7.5 min (system A). Reverse phase gradient HPLC was performed in a Waters µBondapak C18 Radial-Pak column, 10 µm, using a solvent system, 5 mM KH2PO4 (pH 3.2)/CH3CN (95: 5, v/v) to KH2PO4/CH3CN (60:40, v/v), from 0 to 25 min in a hyperbolic gradient followed by a linear gradient to 100% CH3CN from 25 to 45 min at a flow rate of 1.5 mL/min, based on the work of Sidhu et al. (36). BBT had a retention time of 36 min in this system (B). The chemical and radiochemical purity of both forms of [14C]BBT used in these experiments was >97%. (A) N-Benzyl-1-amino-[14C]-2,3-benzotriazole ([14C]-2,3BBT). To a solution of [14C]-2,3-ABT (2 mg, 14.9 µmol, 300 µCi) in glacial acetic acid (0.5 mL) was added excess benzaldehyde (10 µL, 98 µmol), and the solution was stirred overnight under a nitrogen atmosphere at room temperature. The mixture was evaporated in vacuo to leave an oily residue which was dissolved in 7 mL of CH2Cl2/MeOH (95:5) and washed three times with 5 mL of 5% sodium bicarbonate and then with three 4 mL portions of water. The organic phase was dried over anhydrous Na2SO4, filtered, and evaporated in vacuo. The imine product in 10 mL of MeOH/CH2Cl2 (2:1) was reduced with NaBH4 (220 mg, 5.81 mmol). After stirring for 1 h at room temperature, the solvent was evaporated to leave a white solid, which was dissolved in 10 mL of CH2Cl2/MeOH (95:5), washed with three 150 mL portions of water, dried over anhydrous Na2SO4, and concentrated in vacuo. The crude solid was purified by LPC on a Lobar 310 × 25 mm LiChroprep Si 60 column by elution with EtAc/CHCl3 (3:1). The elution time from the column was approximately 40 min. The product had identical retention times with authentic BBT in systems A and B. The yield of [14C]-2,3-BBT, 19.75 mCi/mmol, was 36.7%. (B) N-[14C]-7-Benzyl-1-aminobenzotriazole ([14C]-7-BBT). To a solution of [14C]-7-benzoic acid (1.0 mg, 8.2 µmol, 250 µCi) and cold benzoic acid (9.0 mg, 73.8 µmol) in 4 mL of ether was added N,N-carbonyldiimidazole (14 mg, 86 µmol), and the solution was stirred at room temperature under a nitrogen atmosphere for 24 h. The solution was warmed gently under slight vacuum before being cooled to -20 °C. Reduction was accomplished by the dropwise addition of LiAH4 (2 mg, 52.6 µmol) in 4 mL of ether over 30 min with vigorous stirring under a nitrogen atmosphere. The ether was removed under vacuum, 2 mL of ice-cold MeOH was added to the vessel, and the solution was acidified to pH 4 with 10% H2SO4. MeOH was removed under a gentle stream of nitrogen; the residue was dissolved in 2 mL of glacial acetic acid and reacted with excess ABT (10.4 mg, 77.6 µmol) for 3 days under nitrogen at room temperature with constant stirring. Extraction of the Schiff’s base, reduction of this imine with NaBH4, and final product purification were completed as for [14C]BBT. The yield of purified product, 3.23 mCi/mmol, was 12.1%. Synthesis of N-(4-Hydroxybenzyl)-1-ABT (4-OH-BBT). ABT and 4-hydroxybenzaldehyde (4-HBA) were condensed, essentially as described for BBT (31). To a stirred solution of glacial acetic acid (10 mL) were added ABT (488 mg) and 4-HBA
Metabolism of the Suicide Inhibitors ABT and BBT (444 mg). The reaction proceeded at room temperature (nitrogen atmosphere) with constant stirring for 48 h. Water (10 mL) was added and the Schiff’s base extracted in ether (3 × 20 mL). Unreacted aldehyde was removed with a sodium bisulfite wash. NMR spectroscopy (Varian Gemini 300 spectrometer at 300 MHz) of the imine intermediate in deuterated DMSO showed the expected signals: doublets at 6.98 and 7.96 ppm (phenyl hydrogens), a singlet at 9.48 ppm (imine hydrogen), multiplets at 7.52, 7.67, and 7.94 ppm, and a doublet at 8.05 ppm (benzotriazole ring hydrogens). To a stirred solution of crude imine (100 mg) in CH3OH/CH2Cl2 (2:1, v/v) was added NaBH4 (780 mg); the reaction was allowed to proceed at room temperature for 1 h. Product extraction was as for [14C]BBT. The yield was 78.2%. Synthesis of Trimethylsilyl (TMS) Derivatives of 4-OHBBT and Other Selected Metabolites of BBT. 4-OH-BBT was derivatized for GC/MS analysis by a modification of the method of Butts (38). Briefly, 10-100 µg of 4-OH-BBT (or other metabolite) were placed in a 1 mL reacti-vial, reconstituted in 100 µL of dried pyridine, and reacted with 100 µL of BSTFA [bis(TMS)trifluoroacetamide] for 3 h at 60 °C in a capped vial. The TMS derivative was taken to dryness under nitrogen and dissolved in dried pyridine at a concentration of 1 mg/mL. GC/MS Analysis of 4-OH-BBT-TMS and TMS Derivatives of BBT Metabolites. 4-OH-BBT-TMS and two metabolites were analyzed by GC/MS under chemical ionizing conditions using isobutane as the ionizing agent. GC separation was accomplished on a Varian 3400 gas chromatograph using a DB5 30m capillary column (J&W Scientific) run under the following conditions: injector temperature, 230 °C; transline temperature, 200-230 °C; temperature profile, 3 min at 65 °C, 15 °C/min to 300 °C, and then held at 300 °C (system C, GC). The retention time of 4-OH-BBT-TMS was 19.0 min. A Finnigan MAT Model 8230 mass spectrometer was used to record the mass spectrum. Animal Treatment. Male Hartley guinea pigs (300-350 g) were used; some were treated intraperitoneally with 80 mg/kg PB (2% in saline) or 80 mg/kg βNF (2% in corn oil) daily for 4 days. Animals were sacrificed 24 h following the last injection by asphyxiation with CO2. All animals were allowed free access to food and water throughout the treatment period. In Vitro Incubation of Hepatic and Pulmonary Microsomes with 14C-Labeled Suicide Inhibitors. Liver and lung microsomes were prepared as previously described (33). Incubation mixtures (2 mL) contained hepatic or pulmonary microsomal protein (14-16 mg) and [14C]ABT, [14C]-2,3-BBT, or [14C]-7-BBT (100 µM, 0.225 µCi) in 0.1 M potassium phosphate buffer, pH 7.4. Reactions were started by the addition of 1 mM NADPH (no NADPH in controls). [14C]ABT, [14C]-2,3BBT, or [14C]-7-BBT was dissolved in EtAc and ABT or BBT was dissolved in MeOH and then added to the incubation vessel, and the solvents were removed under a gentle stream of nitrogen at room temperature. The residue was dissolved in 10 µL of DMSO prior to addition of other components. Incubations were performed in duplicate. After incubation (45 min at 37 °C), the mixtures were immediately centrifuged for 15 min (1-4 °C) at 412160g (Beckman TL-100 ultracentrifuge, TLA 100.3 rotor) to sediment the microsomes. The resulting supernatant fraction was saved for extraction of metabolites. The microsomal pellets were washed by resuspension and resedimentation to remove excess inhibitor. This supernatant was stored at -80 °C. The resulting microsomal pellets (M) were resuspended in 0.1 M potassium phosphate buffer (pH 7.4) to a final concentration of 3-5 mg of protein/mL; the 14C content of all fractions was determined by liquid scintillation spectroscopy (LSS) of small aliquots. Extraction of Microsomal Supernatant (A) and Microsomal Pellet (B) Fractions. (A) The microsomal supernatant was extracted repeatedly with 2 mL of EtAc at neutral pH until no more radioactivity was extracted into the organic phase (6-10 extractions). The aqueous phase was adjusted to pH 10 with 1 N NaOH and re-extracted with EtAc (as above). The aqueous phase was then adjusted to pH 2 with 1 N HCl and re-extracted with EtAC.
Chem. Res. Toxicol., Vol. 10, No. 5, 1997 591 Table 1. [14C]ABT and Metabolites (nmol) in Ethyl Acetate Extracts of Incubation Mixtures of [14C]ABT, 1 mM NADPH, and Hepatic or Pulmonary Microsomes from Untreated or PB- or βNF-Induced Guinea Pigsa microsome source
benzotriazole (ABT-M1)
metabolite (ABT-M2)
ABT
liver PB liver βNF liver lung
8.2 12.4 16.5 4.5
1 nmol from [14C]-2,3-BBT by hepatic microsomes (+NADPH) from βNF-induced guinea pigs (Figures 1A and 2A, Table 2). These metabolites include M1, approximate retention time (tR) of 15 min; M2, 16 min; M3, 19 min; M4, 22 min; M5, 24 min (major); M7, 28 min; M9, 31 min (major); M10, 36.5 min; and M12, 41 min; the retention time of BBT was 35.5 min. Six significant metabolites were produced by liver microsomes of PB-treated guinea pigs (Figures 1B and 2B, Table 1): M1, M3 (major), M5 (major), M8, M9 (major), and M12 (major). Three metabolites, M2, M4, and M7, were formed only in liver microsomes from βNFinduced animals and one, M8, only by microsomes from PB-induced animals. M1 and M3 were identified as [14C]ABT and [14C]bentotriazole, respectively, by coelution with authentic standards in systems B and D. M9 was identified as 4-OH-BBT and M5 as 2-OH-BBT or 3-OHBBT by GC/MS analysis (see below). The remainder could not be identified due to the limited amount of purified product available. The metabolic profile of [14C]-2,3-BBT differed considerably between hepatic microsomes from PB- and βNFinduced animals (Table 2). Of relevance to mechanismbased inactivation of P450 2B isozymes, ABT, benzo-
Figure 2. Representative elution profile of radioactivity obtained by HPLC analysis of ethyl acetate extracts of supernatant and microsomes from incubation mixtures of hepatic microsomes from βNF- (A) or PB- (B) induced guinea pigs with [14C]-2,3BBT (100 µM, 0.225 µCi) and 1 mM NADPH.
triazole, and M12 were formed in higher amounts in PBtreated (4 vs 1, 12 vs 2, and 12 vs 2 nmol, respectively) than in βNF-induced animals, while the reverse was true for 2- or 3-OH-BBT (M5) and 4-OH-BBT (M9), 38 vs 13 and 23 vs 7 nmol, respectively. Two other very lipophilic metabolites (in addition to M12), M11 and M13 (tR 38 and 43 min), were produced in trace quantities (1 nmol) radiolabeled metabolites by microsomes from βNF liver (Figures 5A and 6A, Table 2). These included M2, 2- or 3-OH-BBT (major), M6 (ben-
Metabolism of the Suicide Inhibitors ABT and BBT
Chem. Res. Toxicol., Vol. 10, No. 5, 1997 593
Table 2. [14C]-2,3-BBT or [14C]-7-BBT and Metabolites (nmol) in Ethyl Acetate Extracts of Incubation Mixtures of [14C]-2,3-BBT or [14C]-7-BBT, 1 mM NAPDH, and Hepatic Microsomes from PB- or βNF-Induced Guinea Pigs or Pulmonary Microsomes from Untreated Guinea Pigsa HPLC peak and metabolite no. 2,3-BBT 2,3-BBT 2,3-BBT 7-BBT 7-BBT 7-BBT
microsome source
M1
M2
M3
M4
M5
M6
M7
M8
M9
M10
M11
M12
M13
BBT
PB liver βNF liver lung PB liver βNF liver lung
3.9 1.4
