Introduction

Top Abstract Introduction Materials & Methods Results Discussion References

The microsomal P450¹ monooxygenase system is central in the metabolism of a wide variety of lipophilic compounds of both endogenous and exogenous origin. Numerous P450 isozymes have been identified and classified (based on amino acid sequence identity) into gene families and subfamilies (Nelson et al., 1996), many of which exhibit broad and overlapping substrate specificities, as well as tissue- and gender-selective patterns of regulation (Park et al., 1995; Prough et al., 1996). Consequently, highly selective biological and chemical probes are often required to assess the role of an individual P450 isozyme in a particular biotransformation reaction in intact cells. P450 inducers have been used in this regard; however, multiple isozymes are typically affected concomitantly by this treatment (Gonzalez et al., 1993). Inhibitory polyclonal and monoclonal antibodies raised against purified P450s have been successfully used to investigate the isozyme-selective metabolism of a number of substrates (Lubet et al., 1985; Nerurkar et al., 1993). However, this method generally requires broken-cell preparations and is not suitable for in vivo studies. A third approach is to design mechanism-based inactivators of P450. Mechanism-based inactivation provides a number of advantages for utility and selectivity over other methods, because 1) the inhibitor must be capable of specifically binding to the enzyme and serving as an acceptable substrate, 2) the inhibited enzyme is irreversibly inactivated and permanently removed from the catalytic pool, 3) mechanism-based inhibitors are generally effective in vivo and with intact cells, and 4) mechanism-based inhibitors can be designed for tissue-selective inhibition by taking advantage of unique transporter systems (Bend et al., 1985; Halpert, 1995; Halpert et al., 1994).

N-Aralkylated derivatives of ABT are potent, mechanism-based inhibitors of microsomal P450. These compounds, and in particular MB, exhibit pronounced isozyme (CYP2B) and tissue (lung) selectivity for P450 inactivation, both in vitro and in vivo (Knickle and Bend, 1992; Mathews and Bend, 1993; Woodcroft and Bend, 1990; Woodcroft et al., 1990). Consequently, these compounds have been used as sensitive biochemical probes of P450-dependent metabolism of important biological compounds (Knickle and Bend, 1994; Knickle et al., 1993). ABT derivatives can inhibit P450 by at least three mechanisms, i.e. covalent modification of the prosthetic heme group (Mathews and Bend, 1986), covalent modification of the apoprotein moiety (Woodcroft et al., 1997), and metabolic intermediate complexation (Sinal and Bend, 1995). Of these, protein modification appears to be the most important for CYP2B inactivation (Kent et al., 1997a; Woodcroft et al., 1997). These studies have also shown that the identity of the alkyl substituent present on the -carbon is an important determinant of the inhibition mechanism, isozyme/tissue selectivity, and inactivation kinetics exhibited by ABT derivatives. The goal of the present study was to examine the effect of the stereochemistry of the -carbon substituent on the isozyme selectivity and inactivation kinetics of MB in P450 inactivation. To this end, individual enantiomers of MB (fig. 1) were prepared and compared with respect to the kinetics of inactivation of CYP2B and CYP1A isoforms in guinea pig hepatic microsomes.

View larger version (12K): [in this window] [in a new window]   Fig. 1.   Structures of enantiomers of MB.

	Materials and Methods

Top Abstract Introduction Materials & Methods Results Discussion References

Chemicals. NADPH was purchased from Sigma Chemical Co. (St. Louis, MO); 7-ethoxyresorufin, 7-pentoxyresorufin, 7-methoxyresorufin, and resorufin from Molecular Probes (Eugene, OR); di-p-toluoyl-D-tartaric acid monohydrate and di-p-toluoyl-L-tartaric acid monohydrate from Aldrich Chemical Co. (Milwaukee, WI); and phenobarbital, dimethylsulfoxide, and all other chemicals (reagent grade or better) from BDH (Toronto, Canada).

Preparation and Characterization of MB Enantiomers. Diastereoisomeric di-p-toluoyl-tartrate salts of MB were prepared from racemic MB in the following manner. MB (250 mg, 0.66 mM) was mixed with di-p-toluoyl-D-tartaric acid monohydrate (305 mg, 0.75 mM), and the mixture was dissolved in a small volume (2 ml) of diethyl ether. After dissolution, the ether was removed by evaporation and the residue was dissolved (with warming) in a minimal volume of redistilled ethyl acetate. The solution (1.0 ml) was maintained in the cold at 4°C. Rosette crystals formed slowly. Three recrystallizations from warm ethyl acetate (0.6-0.8 ml) were conducted, yielding the D-tartrate salt (55 mg). Basification and extraction of the combined mother liquors yielded an MB residue (187.7 mg, 0.50 mM), which was mixed with di-p-toluoyl-L-tartaric acid (201 mg, 0.52 mM). This mixture was dissolved, as before, in diethyl ether (2 ml). The solvent was then removed by evaporation. The residue was dissolved in and recrystallized slowly three times from ethyl acetate (0.4-0.6 ml), yielding the L-tartrate salt (35 mg).

Animal Treatment and Preparation of Microsomes. Male Hartley guinea pigs (250-300 g) were treated ip with 80 mg/kg phenobarbital (2% in saline) for 4 days and were euthanized by asphyxiation with CO₂ 24 hr after the last injection. All animals were allowed free access to food (Purina guinea pig chow) and water throughout the treatment period. Hepatic microsomes were prepared by differential centrifugation, as previously described (Bend et al., 1972). Microsomal protein concentrations were determined by the method of Lowry et al. (1951), using bovine serum albumin as the standard. Microsomal P450 contents were determined from the dithionite difference absorption spectra of carbon monoxide-saturated microsomes ( = 100 mM¹·cm¹) (Estabrook et al., 1972). Microsomes were stored at 80°C until use.

Preparation of Inhibitor Solutions. The MB enantiomers were initially dissolved as stock solutions (0.1 M) in methanol. The concentrations of these solutions were verified by UV absorbance spectrophotometry ( = 280 nm), using standard curves. A racemic stock solution was produced by combing equal volumes of equimolar solutions of (+)-MB and ()-MB. Volumes of 100 µl were dispensed into 1-ml volumetric flasks, and the methanol was evaporated under a gentle stream of N₂. The remaining residue was made up to 1 ml with dimethylsulfoxide, to yield a final concentration of 0.01 M. Serial dilutions of this stock solution were used to obtain all of the inhibitor concentrations used.

Inhibition Assays. Primary inhibition incubation mixtures contained hepatic microsomal protein (3.75 mg) and 5 µl of the appropriate inhibitor dilution made up to a final volume of 975 µl with 0.1 M potassium phosphate buffer (pH 7.4) containing 1 mM EDTA. The samples were preincubated for 5 min in a shaking water bath maintained at 37°C. Reactions were started by the addition of NADPH (final concentration, 1 mM) and were incubated, with shaking, for periods of 15 sec to 2.5 min. For the determination of competitive inhibition (t = 0 min), identical inhibition incubations were performed in the absence of NADPH. Aliquots (15 µl) of the primary incubations were diluted 100-fold by transfer into prewarmed secondary enzyme incubation mixtures containing substrate (5 µM 7-pentoxyresorufin or 1.3 µM 7-ethoxyresorufin), 0.1 M potassium phosphate buffer (pH 7.4), 1 mM EDTA, and 1 mM NADPH. The secondary incubations were allowed to proceed for 3.5 min at 37°C before quenching with 2 volumes of ice-cold methanol. Precipitated protein was removed by centrifugation (Sorvall GLC-1 centrifuge, M rotor) for 5 min at 3000 rpm. The fluorescent product resorufin was measured with an excitation wavelength of 535 nm and an emission wavelength of 585 nm, using a Perkin-Elmer fluorescence spectrophotometer (model LS-5B). The rate of PROD or EROD activity was calculated based on a standard curve of fluorescence vs. resorufin concentration. In some experiments, the inhibition reaction was allowed to proceed for 30 min and then the reaction mixture was placed on ice. After this, the microsomes were washed by sedimentation of the inhibition mixtures at 412,160g for 15 min at 4°C (Beckman TL-100 ultracentrifuge; TLA 100.3 rotor), followed by resuspension in 0.1 M potassium phosphate buffer (pH 7.4) containing 1 mM EDTA. After determination of the protein concentration, 50 µg of these washed microsomes was assayed for PROD or EROD activity as described above. All assays were verified to proceed linearly with respect to time and protein concentration.

Data Analysis. The apparent half-life for inactivation was calculated by linear regression analysis of the natural logarithm of residual enzyme activity with respect to inhibitor incubation time. Apparent k_inact values were derived from plots of t_1/2 vs. the reciprocal of the inhibitor concentration, by the method of Kitz and Wilson (1962). Data were analyzed by one-way analysis of variance, followed by the Tukey-Kramer multiple-comparisons test, using SuperANOVA for Macintosh software (version 1.11; Abacus Concepts, Berkeley, CA).

	Results

Top Abstract Introduction Materials & Methods Results Discussion References

Preparation and Identification of MB Enantiomers. Individual enantiomers of MB (fig. 1) at the single chiral center were obtained from racemic MB by separation of diastereomeric salts and purification of the regenerated enantiomers. The HPLC retention times, NMR chemical shifts, and mass spectral characteristics (data not shown) of the purified enantiomers were identical to those obtained for an authentic MB racemate (Mathews and Bend, 1986). The optical activity of the individual enantiomers was established through measurement of the optical rotatory dispersion, as shown in fig. 2. A positive Cotton effect (peak, 288 nm; trough, 260 nm) was clearly seen for one compound, whereas a negative Cotton effect (peak, 260 nm; trough, 288 nm) was observed for the other, indicating that the two compounds were indeed enantiomeric forms of the same compound. Therefore, the individual enantiomeric forms of MB are henceforth referred to as (+)- and ()-MB, with reference to their respective Cotton effects. It is also possible to tentatively assign an absolute configuration to each of the enantiomers, based on their Cotton effects and comparison with similar compounds reported in the literature. The benzene chromophore shows three well-defined absorption bands associated with * transitions. If the benzene ring is substituted with a chiral group, then these transitions become optically active and produce Cotton effects (Johnson et al., 1987). The B_2U transition at 240-270 nm, as observed by its Cotton effect, correlates with the absolute configuration of the chiral substituent attached to the benzene ring. (S)--Phenylethylamine exhibits a positive B_2U Cotton effect, and alkyl substitution on the amine group does not change the sign of this effect, because both (S)-N,N-dimethyl--phenylethylamine and its methyl iodide maintain a positive sign (Johnson et al., 1987). In addition, for -phenylethylamine a negative Cotton effect at 268 nm correlates with the R-configuration (Smith et al., 1968). Assuming that the benzotriazole substitution on the amino group of -phenylethylamine also does not change the sign of the Cotton effect, then the MB enantiomer with a positive Cotton effect [(+)-MB] is of the S-configuration and a negative Cotton effect [()-MB] correlates with the R-configuration.

View larger version (20K): [in this window] [in a new window]   Fig. 2.   Optical rotatory dispersion profiles of MB enantiomers. Individual optical rotatory dispersion spectra were obtained from a 0.1 mg/ml solution (in CH2Cl2) of each enantiomer, as detailed in Materials and Methods. The peak at 288 nm and the trough at 260 nm indicate a positive Cotton effect, whereas a negative Cotton effect is indicated by the peak at 260 nm and the trough at 288 nm.

Mechanism-Based Inactivation of Hepatic PROD Activity. In these studies, PROD (Lubet et al., 1985) was used as a marker for CYP2B18, the guinea pig orthologue of CYP2B1 (Nelson et al., 1996), which is the isozyme primarily responsible for this catalytic activity in liver microsomes prepared from phenobarbital-treated rats. The time course for inactivation of this activity by the individual MB enantiomer preparations is shown in fig. 3. All three preparations produced a rapid time- and NADPH-dependent loss of PROD activity, consistent with mechanism-based inactivation by these compounds. The highest concentrations of these inhibitors caused an extensive (90%) loss of total PROD activity within 3 min of incubation. The rates of inactivation were found to increase in a concentration-dependent manner, indicating a pseudo-first order rate process (table 1). At all concentrations studied, with the exception of 1 µM, the time required for 50% loss of the initial PROD activity (t_1/2) was significantly shorter for ()-MB, compared with (+)-MB. However, statistically significant differences in the rates of ()-MB and (±)-MB inactivation of PROD activity were not observed.

View larger version (25K): [in this window] [in a new window]   Fig. 3.   Time course of NADPH-dependent, mechanism-based inactivation of PROD activity, in hepatic microsomes from phenobarbital-treated guinea pigs, by MB. At the indicated times, aliquots of microsomes incubated in the presence of ()-MB, (+)-MB, or (±)-MB and 1 mM NADPH were transferred to secondary incubation mixtures and assayed for PROD activity as described in Materials and Methods. Each data point represents the mean of experiments performed in duplicate, using microsomes prepared from four individual livers. The control PROD activities for the experiments with ()-MB, (+)-MB, and (±)-MB were 247.4 ± 19.7, 262.6 ± 27.9, and 260.0 ± 19.7 pmol/min/nmol of P450, respectively.

Mechanism-Based Inactivation of Hepatic EROD Activity. In these experiments, EROD activity was used as a marker of CYP1A-dependent metabolism in liver microsomes from phenobarbital-treated guinea pigs (Burke et al., 1985). As with PROD inactivation, all three of the MB preparations caused time- and NADPH-dependent, mechanism-based inactivation of hepatic EROD activity (fig. 4). However, in this case, the maximal amount of inactivation that could be achieved within 3 min was approximately 50%. Furthermore, the rates for EROD inactivation were less than those observed for PROD (table 2). For example, at a concentration of 2.5 µM, the t_1/2 for EROD inhibition was 3.5-7.3-fold longer than for PROD inactivation, depending on the MB preparation used. Although the rates of EROD inactivation increased in a concentration-dependent manner, there were no significant differences among the t_1/2 values for any of the compounds at any of the concentrations used in this study.

View larger version (23K): [in this window] [in a new window]   Fig. 4.   Time course of NADPH-dependent, mechanism-based inactivation of EROD activity, in hepatic microsomes from phenobarbital-treated guinea pigs, by MB. At the indicated times, aliquots of microsomes incubated in the presence of ()-MB, (+)-MB, or (±)-MB and 1 mM NADPH were transferred to secondary incubation mixtures and assayed for EROD activity as described in Materials and Methods. Each data point represents the mean of experiments performed in duplicate, using microsomes prepared from four individual livers. The control EROD activities for the experiments with ()-MB, (+)-MB, and (±)-MB were 81.2 ± 12.7, 76.0 ± 6.9, and 74.0 ± 13.4 pmol/min/nmol of P450, respectively.

Inactivation Rate Constants. The maximal k_inact value and the inhibitor concentration required for the half-maximal rate of inactivation (K_i) were determined for each of the MB preparations (table 3), using the t_1/2 values obtained (tables 1 and 2). This was accomplished by the use of Kitz-Wilson plots (Kitz and Wilson, 1962) of the inactivation t_1/2 vs. the reciprocal of the inhibitor concentration (fig. 5). As indicated by the positive value of the point of intersection with the ordinate, inactivation of CYP2B or CYP1A isozymes was a saturable process with respect to inhibitor concentration. The k_inact value for PROD inactivation by ()-MB (0.49 ± 0.06 min¹) was significantly greater than that for (+)-MB (0.35 ± 0.03 min¹). In contrast, the K_i value for ()-MB (2.4 ± 0.7 µM) was not significantly greater than that for (+)-MB (2.7 ± 0.5 µM). No significant differences were found when the individual enantiomers were compared with the racemate, and no significant differences were found among the k_inact and K_i values for EROD inactivation by the MB compounds. However, the k_inact values for PROD inactivation were, on average, approximately 2.4-fold larger than those for EROD inactivation. In contrast, the K_i values were generally similar for EROD inactivation and PROD inactivation.

View larger version (15K): [in this window] [in a new window]   Fig. 5.   Kitz-Wilson plot of the half-life for mechanism-based inactivation of PROD or EROD activity as a function of the reciprocal of the inhibitor concentration. The apparent kinetic constants for P450 inactivation by ()-MB, (+)-MB, or (±)-MB were derived from the intercepts of the straight lines with the x and y axes (kinact = 0.693/y intercept and Ki = 1/x intercept). Each data point represents the mean of experiments performed in duplicate, using microsomes prepared from four individual livers.

Maximal Extent of Inactivation. To determine whether the differences in the rates of P450 inactivation between the MB enantiomers had any effect on the maximal extent of inhibition, we performed extended inhibition reactions that were 30 min in length. After a 30-min incubation, the inhibited microsomes were washed by sedimentation and resuspension in fresh buffer, to minimize the effect of competitive inhibition resulting from excess inhibitor. As shown in fig. 6, no significant differences in the extent of PROD inactivation were observed among any of the MB preparations at any of the concentrations used. A similar result occurred with EROD inactivation; however, a significant difference between 10 µM (+)-MB and ()-MB or (±)-MB was found. Comparatively little loss (approximately 20%) of spectrally measurable P450 occurred with the highest concentration of the MB preparations, in spite of the substantial loss of PROD and, to a much lesser extent, EROD activity. No significant differences were found in the loss of P450 caused by equimolar (+)-MB, ()-MB, or (±)-MB.

View larger version (38K): [in this window] [in a new window]   Fig. 6.   Maximal extent of mechanism-based inactivation of hepatic PROD and EROD activities and P450 content by individual MB enantiomers or the racemate. Guinea pig hepatic microsomes were incubated for 30 min in the presence of NADPH and ()-MB, (+)-MB, or (±)-MB at the indicated concentrations. The inhibited microsomes were then washed by sedimentation and resuspension and were assayed for PROD, EROD, and P450 content as described in Materials and Methods. Each bar represents the mean ± SD of experiments performed in duplicate, using microsomes prepared from four individual livers. *, significantly different from (+)-MB, p < 0.05.

	Discussion

Top Abstract Introduction Materials & Methods Results Discussion References

N-Aralkylated derivatives of ABT are well-established isozyme (CYP2B)- and tissue (lung)-selective inhibitors of guinea pig and rabbit P450, both in vivo and in vitro (Knickle and Bend, 1992; Mathews and Bend, 1993; Woodcroft and Bend, 1990; Woodcroft et al., 1990). The length of the alkyl substituent present at the -carbon position is an important determinant of the actions of these compounds. For example, MB and N-(-ethylbenzyl)-1-aminobenzotriazole are both substantially more potent and selective than BBT for mechanism-based inactivation of hepatic and pulmonary CYP2B isoforms in vitro (Mathews and Bend, 1986; Woodcroft et al., 1990). Similarly, MB is a more potent and selective inactivator of pulmonary CYP2B in vivo than is BBT (Knickle et al., 1994; Mathews and Bend, 1993). In contrast, metabolic intermediate complexation of P450, an inhibition mechanism shown not to be relevant for CYP2B inactivation, exhibits a 3-fold higher K_m value for MB than for BBT (Sinal and Bend, 1995). Although the effects of different-sized -substituents have been documented, the effect of stereochemistry at this chiral center position has not. To this end, individual enantiomers of MB were prepared and positively identified using established chromatographic and spectroscopic methods. Based on these data and information obtained from the literature, a tentative assignment of absolute configuration was also made for each of the enantiomers. A kinetic study of these compounds was undertaken to compare the enantiomers in terms of inactivation rates and isozyme selectivity. Previous studies of these ABT derivatives with respect to inactivation kinetics have helped to explain the P450 isozyme selectivity of this class of compounds, as well as the relative potency and selectivity of MB vs. BBT. For example, inactivation of rabbit CYP2B4 and CYP2B5 is much more rapid for MB (k_inact = 0.68 and 0.55 min¹, respectively) than for BBT (0.29 and 0.18 min¹, respectively), consistent with the greater isozyme selectivity of the former compound (Grimm et al., 1995). Similarly, inactivation of guinea pig pulmonary CYP2B-dependent PROD activity by 25 nM MB is much more rapid (t_1/2 = 0.9 min) than with equimolar BBT (t_1/2 = 32 min) (Sinal and Bend, 1996). The maximal k_inact reflects both the rate of conversion of inhibitor to reactive intermediates and the partition ratio (a measure of the efficiency of generating reactive intermediates, compared with stable products) (Halpert, 1995; Rando, 1984). Therefore, k_inact provides a measure of isozyme selectivity as well as a means of comparing the inactivation of a specific enzyme activity, under identical reaction conditions, by different compounds.

This study demonstrates a stereoselective difference in mechanism-based inactivation of P450 by enantiomers of MB. The shorter t_1/2 values at all concentrations studied and the larger k_inact value clearly indicate that ()-MB is a more rapid inactivator of CYP2B-dependent PROD, compared with (+)-MB. In contrast, there were no differences in the inactivation of CYP1A-dependent EROD activity by enantiomers of MB. Consistent with previous data (Sinal and Bend, 1996), the faster rate of PROD vs. EROD inactivation indicates that the selectivity of MB for CYP2B(s) is the result, at least in part, of a higher rate of inactivation. The much larger degree of competitive inhibition of PROD vs. EROD activity, as determined by incubation with the inhibitors in the absence of NADPH, also indicates that selectivity for CYP2B(s) is the result, in large part, of greater binding affinity, in spite of the similar K_i values determined for mechanism-based inactivation. It is important to note that the PROD and EROD assays were performed after 100-fold dilution of the inhibition incubations and, therefore, the actual concentrations of the inhibitors in the assay mixtures were approximately 1% of those shown in the figures and tables. Thus, the actual affinity of MB for binding to CYP2B and CYP1A isozymes is somewhat underrepresented. Surprisingly, (+)-MB caused a greater degree of competitive inhibition at the three highest concentrations used but showed a similar K_i value for PROD inhibition, compared with ()-MB. This indicates that, although affinity is important for metabolic activation of MB, oxidation is most likely the rate-limiting step for mechanism-based inhibition.

The experiments involving 30-min incubations of the inhibitors with guinea pig hepatic microsomes and NADPH were performed to determine the maximal extent of inactivation by MB enantiomers. Although these data further demonstrate the selectivity of MB for CYP2B vs. CYP1A inhibition, there were no differences between the enantiomers. The similarity of PROD and EROD inactivation, as well as the loss of spectrally measurable P450, indicates that 30 min is sufficient time to overcome the difference in initial inactivation rates exhibited by the two MB enantiomers. Furthermore, these data indicate that, although significant differences in the kinetic parameters for CYP2B inactivation by MB do exist, it seems unlikely that any differences exist for the reaction mechanism. Taken together, these data indicate that the stereochemical orientation of the -substituent of MB has significant effects on binding affinity and oxidation rates but does not affect the reaction mechanism that ultimately results in mechanism-based inhibition of P450.

All of the N-aralkylated derivatives of ABT were designed to have molecular features that structurally mimic benzphetamine, a substrate for CYP2B-dependent N-demethylation (Serabjit-Singh et al., 1983). Common structural features include an N-benzyl group, an aromatic region of similar size, an amino group in the desired region of oxidation, and an -carbon for substitution. Consistent with the stereoselectivity for mechanism-based inactivation of CYP2B(s) by MB demonstrated in this study, benzphetamine also exhibits stereoselectivity for CYP2B-dependent metabolism. For example, the V_max for amphetamine formation via -carbon oxidation is 3-fold greater for (+)-benzphetamine than for ()-benzphetamine, whereas the K_m values are similar (Beckett and Gibson, 1978). Given that a difference was found in the rates of inactivation for MB enantiomers, but not the K_i values, -carbon oxidation may be an important determinant of the rate of CYP2B inactivation.

Studies with radiolabeled BBT have shown that this ABT derivative is metabolized by guinea pig hepatic microsomes (Woodcroft et al., 1997) or purified rat CYP2B1 (Kent et al., 1997a) to benzotriazole, benzaldehyde, ABT, and a previously unidentified metabolite designated as metabolite 27 (fig. 7), which is generated at a relatively high rate (12 nmol/nmol P450/min). Recently, metabolite 27 has been putatively identified as a dimeric product of BBT oxidation (Kent et al., 1997b). Interestingly, this is the only metabolite generated by a mutant form of rat CYP2B1 that contains a glycine-478 to alanine substitution and is not inactivated by BBT (Kent et al., 1997b). Benzotriazole, benzaldehyde, and ABT are believed to result from oxidation at the 1-amino nitrogen of BBT (Woodcroft et al., 1997), whereas the BBT dimer product is thought to arise from -carbon oxidation (Kent et al., 1997b). Molecular modeling studies, in combination with experimental observations, indicate that mutation of glycine-478 to alanine results in steric hindrance that favors oxidation of BBT at the -carbon, rather than the 1-amino nitrogen (Kent et al., 1997b). Taken together, these data indicate that 1-amino nitrogen oxidation results in generation of a reactive intermediate capable of inhibiting rat CYP2B1 in a mechanism-based manner, whereas -carbon oxidation results in metabolism of BBT to a noninhibitory dimer. By analogy, the stereoselectivity of guinea pig CYP2B inactivation shown in this study may indicate that, compared with ()-MB, (+)-MB is oriented in the P450 active site in a manner more favorable for -carbon oxidation, leading to more frequent metabolism to noninhibitory products, such as the dimer observed for BBT. This difference in the apparent partition ratio is consistent with both the faster rate of inactivation of guinea pig hepatic CYP2B isozymes by ()-MB, compared with (+)-MB, and the equivalent extent of inactivation after 30 min. Presumably, guinea pig hepatic CYP1A isozymes do not catalyze -carbon oxidation of MB or are not sensitive to the stereochemistry of the -substituent.

View larger version (9K): [in this window] [in a new window]   Fig. 7.   Simplified scheme outlining the potential outcomes of BBT metabolism initiated at the amino nitrogen (*) or C7-carbon (**). Parent compound and metabolites shown are as follows: A, BBT; B, benzotriazole; C, ABT; D, benzaldehyde; E, BBT dimer; F, molecular nitrogen. Ph is used to differentiate the benzene ring of the N-benzyl group from the benzene ring of the benzotriazole moiety of BBT.

In summary, mechanism-based inactivation of CYP2B-dependent PROD activity in hepatic microsomes prepared from phenobarbital-treated guinea pigs exhibits stereoselectivity for ()-MB, with respect to inactivation rate. In contrast, (+)-MB has a higher binding affinity for CYP2B, based on the amount of competitive inhibition of PROD activity, although this was not reflected by a difference in K_i values. Mechanism-based inactivation of CYP1A-dependent EROD activity by MB is not stereoselective with respect to either inactivation rate or binding affinity. After extended incubations in the presence of inhibitor and NADPH, stereoselectivity for maximal PROD or EROD inactivation or P450 loss was absent. We conclude that mechanism-based inhibition of guinea pig hepatic CYP2B but not CYP1A isozymes by MB occurs in a stereoselective manner, most likely as a result of a difference in the balance between metabolic activation and deactivation for the MB enantiomers.