Abstract
Benzyloxyresorufin-O-dealkylation (BROD) is usually used as a marker of cytochrome P450 (P450) 2B1 in rat. However, some reports show that CYP1A2 is also highly implicated. The purpose of the present study was to establish bupropion (BUP) hydroxylation, but not BROD, as a selective in vitro marker of CYP2B1 catalytic activity. IC50 for BROD and BUP hydroxylation were equivalent (40.8 ± 4.6 and 41.8 ± 3.4 μM, respectively) when using liver microsomes from β-naphthoflavone-pretreated rats in the presence of metyrapone (CYP2B1 inhibitor). When using the same microsomes in the presence of CYP1A1/2-selective inhibitor α-naphthoflavone, we found an IC50 of 2.5 × 10−3 ± 0.8 × 10−3 μM for BROD and >100 μM for BUP hydroxylation. These results suggest that CYP2B1 is similarly involved in both activities, whereas CYP1A2 is involved in BROD activity but not in BUP hydroxylation. BUP hydroxylation was assessed in microsomes from baculovirus-infected insect cells coexpressing NADPH-P450 oxidoreductase, and 14 rat P450s and kinetic parameters (Km and Vmax) were determined. BUP hydroxylation was predominantly catalyzed by CYP2B1 (75% of total hydroxybupropion formation), low activity was detected with CYP2E1 and CYP2C11 (10.9 and 8.7% of total hydroxybupropion, respectively), and activity was almost undetectable with the other P450 isoforms at saturating substrate concentrations (2500 μM), thereby validating the use of BUP as a diagnostic in vitro marker of CYP2B1 catalytic activity in rat.
Introduction
In the development of new chemical entities for use as medicines in humans, preclinical screening includes investigation of the metabolism rates and routes in the safety evaluation species to ensure that the metabolites produced in these species are consistent with those predicted to be produced in humans. Sprague-Dawley or Wistar rats are often used as the rodent species in these evaluations. Cytochromes P450 (P450s) are the family of heme-containing, drug-metabolizing enzymes that are involved in the biotransformation of xenobiotics, environmental contaminants, dietary components, and procarcinogens (Wrighton and Stevens, 1992; Gonzalez and Gelboin, 1994). Rat CYP2B1/2, mouse CYP2B9/10, and human CYP2B6 share approximately 80% nucleotide sequence identity (Lewis et al., 1999). Compared with several other P450 subfamilies, CYP2B enzymes (<1% of total P450) exhibit a relatively low degree of catalytic preservation across mammalian species (Kedzie et al., 1991; Schenkman and Griem, 1993). Three CYP2B isoenzymes, CYP2B1, -2B2, and -2B3, have been identified in rats (Desrochers et al., 1996), these enzymes being the main hepatic P450 isoform inducible by phenobarbital (PB) and other barbiturates. CYP2B1 and CYP2B2 are structurally related isoenzymes (sharing a 97% amino acid sequence similarity; Suwa et al., 1985) with very similar substrate specificities (Waxman, 1988). In previous studies, PB significantly induced mRNA expression of CYP2B1 (500-fold) as well as CYP3A1 (110-fold), whereas lower induction was observed with other isoforms such as CYP2C6 (4-fold) and CYP2E1 (3-fold), and no induction was observed for mRNA expression of CYP2C11 (Caron et al., 2005). CYP2B1 is generally much more catalytically active than CYP2B2. Both are expressed constitutively in the liver and in extrahepatic tissues such as small intestine and lungs (Lindell et al., 2003). CYP2B enzymes metabolize a diverse group of compounds, including pesticides (Dehal and Kupfer, 1994), chemotherapeutics such as cyclophosphamide (Chang et al., 1993), tobacco-specific nitrosamines (Stiborov á et al., 1996), and drugs of abuse such as cocaine (Poet et al., 1994), nicotine (Nakayama et al., 1993), and antidepressants (Lewis et al., 1999), including BUP.
BUP is an aminoketone used as antidepressant and non-nicotine aid to smoking cessation (Stewart et al., 2001). In humans, BUP is metabolized to hydroxybupropion (HBUP), the pharmacologically active main metabolite (Fig. 1). Previous in vitro reaction phenotyping studies using cDNA-expressed recombinant P450s or human liver microsomes demonstrated that BUP is metabolized primarily by CYP2B6 to HBUP, with an apparent Km of 85 to 156 μM in recombinant expressed CYP2B6 and 89 to 130 μM in human liver microsomes; CYP2E1 and CYP3A4 have also been found to participate in the metabolism of BUP, but at significantly lower rates than CYP2B6 and only at extremely high BUP concentrations (Faucette et al., 2000, 2001; Hesse et al., 2000). In rats, it has been shown that after a single administration of BUP (40 mg/kg i.p. or 200 mg/kg oral), it is quickly metabolized to HBUP with Tmax of 0.68 h (Suckow et al., 1986; Welch et al., 1987); however, to our knowledge, the enzyme(s) involved in this metabolic pathway has not been identified. Previously investigated substrate probes of CYP2B1 activity include benzyloxyresorufin-O-dealkylation (BROD) and pentoxyresorufin-O-dealkylation in vitro (Burke et al., 1985, 1994; Lubet et al., 1985). However, it has been shown that the dealkylation of benzyloxyresorufin and pentoxyresorufin is, in addition to CYP2B1, catalyzed by multiple enzymes, most notably by CYP1A1/2, and to a lesser extent by CYP2C6/11/13, CYP2E1, and CYP3A1/2 (Kobayashi et al., 2002; Chovan et al., 2007). For BROD activity, CYP2B1 showed the highest activity (0.47 pmol · min−1 · pmol P450−1), whereas CYP1A2 was also active (0.27 pmol · min−1 · pmol P450−1) (Kobayashi et al., 2002). Therefore, alternative catalytic probes of CYP2B1 activity that are more selective would be very useful for in vitro assay. In a previous study (Richert et al., 2009), we found a poor correlation between BROD activity and CYP2B1 mRNA expression in rat hepatocytes after 24 and 72 h of culture. By contrast, BUP hydroxylation correlated well with CYP2B1 mRNA expression at both time points. The purpose of the present study was to establish BUP hydroxylation, but not BROD, as a selective in vitro marker of CYP2B1 catalytic activity by using liver microsomes from β-naphthoflavone (BNF)- and PB-pretreated rats, in the presence and absence of known P450 inhibitors and a panel of rat cDNA-expressed P450 enzymes.
Structure of bupropion and hydroxybupropion (major active metabolite).
Materials and Methods
Chemicals and Reagents.
BUP, bovine serum albumin (BSA), benzyloxyresorufin, resorufin, α-naphthoflavone (ANF), BNF, metyrapone, proadifen, NADPH, PB, dexamethasone (DEX), 3-methylcholantrene (3-MC), and fenofibrate (FEN) were purchased from Sigma-Aldrich (Saint Quentin Fallavier, France). HBUP and microsomes prepared from baculovirus-infected insect cells (Supersomes) were purchased from BD Gentest (Woburn, MA). NADPH-P450 oxidoreductase was coexpressed in all microsome preparations, and cytochrome b5 was expressed in microsomes containing cDNA-expressed CYP2A1, CYP2A2, CYP2B1, CYP2C6, CYP2C11, CYP2C12, CYP2C13, CYP2E1, CYP3A1, and CYP3A2. The P450 content of each preparation (picomoles of P450/mg protein) was spectrophotometrically determined by the supplier as follows: 286 (CYP1A1), 286 (CYP1A2), 196 (CYP2A1), 112 (CYP2A2), 77 (CYP2B1), 250 (CYP2C6), 385 (CYP2C11), 83 (CYP2C12), 143 (CYP2C13), 143 (CYP2D1), 50 (CYP2D2), 208 (CYP2E1), 270 (CYP3A1), and 71 (CYP3A2). Microsomes containing baculovirus vector served as controls only for experiments with cDNA-expressed enzymes. All other laboratory chemicals were obtained from commercial suppliers in the highest purity available.
Preparation of Rat Hepatocyte Microsomes.
This study has been performed in accordance with the Guide for the Care and Use of Laboratory Animals (Institute of Laboratory Animal Resources, 1996). Male Wistar rat hepatocytes were isolated as described previously (Richert et al., 2002). Rat hepatocytes (3.5 million cells) were seeded in 60-mm dishes in Human Hepatocyte Maintenance Medium (Primacyt, Schwerin, Germany) supplemented with 5% fetal calf serum, 50 mg/l gentamicin, 4 mg/l insulin, and 0.1 μM DEX. Cells were allowed to attach by incubating under a CO2/air (5:95%) humidified atmosphere maintained at 37°C. After 4 h, the culture medium was replaced with fresh, serum-free medium containing test compound. Test compounds were dissolved in serum-free Human Hepatocyte Maintenance Medium supplemented with 50 mg/l gentamicin, 1× Insulin-Transferrin-Selenium-A Supplement (Invitrogen, Cergy-Pontoise, France), and 0.1 μM DEX to give final concentrations of 5 μM 3-MC, 10 μM BNF, 10 μM DEX, 100 μM FEN, or 1000 μM PB. Control cultures were treated with the solvent, dimethyl sulfoxide (0.1% v/v final concentration). The medium containing test compound was replaced every 24 h. After 24 and 72 h of treatment, hepatocytes were harvested in homogenization buffer and were submitted to several differential centrifugations, as described previously (Richert et al., 2002). The final microsomal pellets were suspended in 0.25 M sucrose. All samples were stored at −80°C. The protein content was determined using the bicinchoninic acid protein determination kit (Sigma-Aldrich), and BSA was used as a standard. Hepatocyte microsomal enzyme activity determinations were performed by incubating hepatocyte microsomes with respective probe substrates: BROD (Burke et al., 1985) and BUP hydroxylation (Faucette et al., 2000).
Preparation of Rat Liver Microsomes.
This study has been carried out in accordance with Guide for the Care and Use of Laboratory Animals (Institute of Laboratory Animal Resources, 1996). Male Wistar rats (8 weeks old; purchased from Janvier, Saint Berthevin, France) were housed in metal cages with a 12-h light/dark cycle and fed ad libitum for 48 h. Rats were pretreated with the conventional P450 inducers; the dose and dosing periods used were as follows: intraperitoneal injections of BNF at 40 mg/kg in 0.8 ml of corn oil for 3 days, and PB at 80 mg/kg in 0.75 ml of corn oil for 3 days. The control rats received 0.8 ml of corn oil vehicle only by the daily intraperitoneal injections for the same duration. After the last treatment, rats were starved for 24 h before sacrifice to reduce the hepatic glycogen content. The rats were sacrificed, and the liver was immediately removed and homogenized in 50 mM Tris-HCl, 150 mM KCl, and 2 mM EDTA, pH 7.4. The homogenates were submitted to several differential centrifugations, as described previously (Richert et al., 2002). Microsomal samples were frozen at −80°C until analysis. The protein content was determined using the bicinchoninic acid protein determination kit (Sigma-Aldrich), and BSA was used as a standard.
Microsomal Assays.
Microsomal BROD was determined according to Burke et al. (1985). In brief, rat liver microsomes (RLM) (0.04 mg of protein) were incubated for 2 min at 37°C with benzyloxyresorufin (20.5 μM) in Tris buffer as substrate in a total volume of 0.1 ml. The reaction was initiated by adding NADPH (1 mM) and was stopped with ZnSO4 (87 mM) and Ba(OH)2 (79 mM). After centrifugation (800g, 5 min) to remove precipitated protein, the fluorescent metabolite resorufin was measured by spectrofluorometry (530 nm excitation and 580 nm emission). Calibration standards (1.25–50 pmol/ml) were prepared by adding known amounts of resorufin to microsomes and incubation buffer. Results were expressed as picomole of resorufin formed per minute per milligram microsomal proteins. Rates of HBUP formation were determined using insect cell-derived microsomes and RLM, according to Faucette et al. (2000). Preliminary experiments in RLM and insect cell-derived microsomes were conducted to identify microsomal protein amounts and incubation times resulting in linear rates of HBUP formation. Incubation mixtures consisted of 0.05 mg of RLM or 50 pmol of cDNA-expressed P450 enzyme, 25 to 2500 μM BUP, 62.5 mM potassium phosphate buffer (pH 7.4), 1 mM EDTA, 3 mM magnesium chloride, and 1 mM NADPH in a total volume of 0.25 ml. All BUP stock solutions were prepared in methanol. Reactions in RLM were initiated at 37°C by the addition of NADPH and stopped after 30 min with 125 μl of ice-cold acetonitrile. Incubation mixtures were centrifuged at 3000 rpm for 5 min. The supernatant (100 μl) was injected onto the high-performance liquid chromatography (HPLC) column. Similar procedures were followed for incubations with microsomes containing cDNA-expressed P450 enzymes, except that reactions were initiated by adding ice-cold microsomes rather than NADPH and then analyzed by liquid chromatography/tandem mass spectrometry (LC/MS/MS).
HPLC Analysis.
The HPLC system for detection of HBUP consisted of an Agilent 1100 liquid chromatography (Agilent Technologies, Waldbronn, Germany) connected to an Agilent Technologies model 1100 UV detector set at 210 nm. Peak of interest was separated on a 5-μm BDS Hypersil 15 × 0.46 cm C18 column (Thermo Scientific, Villebon-sur-Yvette, France). Mobile phases A (0.1% triethylamine) and B (100% acetonitrile) were pumped at a flow rate of 1 ml/min using a gradient ranging from 10 to 20% B at 0 to 7.5 min, 20 to 50% B at 7.5 to 8 min, 50% B at 8 to 12.5 min, and 10% B at 12.5 to 13 min. The column temperature was maintained at 35°C. HBUP peaks were integrated using an Agilent ChemStation system (Agilent Technologies). Retention time for HBUP was approximately 8.7 min. Calibration standards (120–2400 pmol/ml) were prepared by adding known amounts of HBUP to microsomes and reagent stock. HBUP concentrations were calculated from the peak area using least-squares linear regression, with weighting by the reciprocal of the squared standard concentrations. Interday coefficients of variation for calibration standards ranged from 12.5% for the lowest standard to 5% for the highest standard. The lower limit of detection was 120 pmol/ml. This concentration is lower than HBUP concentrations observed in rat plasma up to 4 h after 40 mg/kg i.p. dose of BUP (Suckow et al., 1986).
LC/MS/MS Analysis.
The mass spectrophotometer used was Varian (Les Ulis, France) 320 MS triple quadrupole with pump (212-LC; Varian) and autosampler (ProStar 430; Varian). MS Workstation (Varian) was used for system control and chromatographic data acquisition. The injection volume of samples and standards was 20 μl. The analyses were separated on a Phenomenex Gemini C18 (50 × 2.0 mm, 5 μm; Phenomenex, Torrance, CA) reverse-phase column with Gemini security guard cartridge (4 × 2 mm), at room temperature. The flow rate was 300 μl/min. The mobile phases consisted of A: water, 0.1% formic acid; and B: acetonitrile, 0.1% formic acid. The linear gradient was as follows: 0% B from 0 to 1 min, 0 to 97% B from 1 to 4 min, 97% B from 4 to 5 min, and 97 to 0% B from 5 to 5.06 min. The run time was 5 min, and the equilibration time between injections was 5 min. The analyze peaks were detected by mass spectrometry. Auto-tuning was carried out for maximizing ion abundance followed by the identification of characteristic fragment ions using the generic parameters: collision-induced dissociation gas pressure, 2.2 mTorr; electrospray ionization (ESI) needle voltage, 6000 V; drying gas temperature, 350°C; nebulizer gas pressure, 40° psi; and drying gas pressure, 50 psi. Multiple reaction monitoring transitions were monitored in the positive mode as follows: m/z 238.1 → 138.9, ESI, capillary voltage 92 V, collision energy 25 eV; m/z 238.1 → 167.0, ESI, capillary voltage 92 V, collision energy 20 eV. The standard curve range was 10 nM (corresponding to lower limit of quantification) to 2400 nM (corresponding to upper limit of quantification) of HBUP in phosphate buffer. Interday coefficients of variation for calibration standards ranged from 4.15% for the lowest standard to 3.3% for the highest standard. The lower limit of detection was 5 pmol/ml. Activities were expressed as picomole of HBUP formed per minute per picomole of P450.
Statistical Analyses.
BROD activities in rat hepatocyte microsomes (RHM) were compared with BUP hydroxylase activities in the same microsomal samples using the linear regression program of GraphPad Prism (GraphPad Software Inc., San Diego, CA). p values for the r2 were determined from an F test. The acceptance limit for statistical significance was set at α = 0.05.
Inhibition of BROD and BUP Hydroxylase Activities by Selective Inhibitors.
The effects of ANF (CYP1A2 inhibitor), metyrapone (CYP2B1 inhibitor), and proadifen (CYP1A2/2B1 inhibitor) on BROD and BUP hydroxylase activities were evaluated in BNF-pretreated and PB-pretreated RLM. The inhibition experiments were conducted with 0.001 to 100 μM inhibitors at 500 μM BUP for BUP hydroxylation or 20.5 μM benzyloxyresorufin for BROD. This pilot experiment was performed to confirm the ability of metyrapone and proadifen to inhibit the BROD and BUP hydroxylation by using both PB-pretreated RLM and cDNA-expressed CYP2B1. The indicated amounts of inhibitor were preincubated on ice with 0.05 mg of RLM for 5 min before reactions were initiated by addition of reagent stock consisting of 500 μM BUP in 62.5 mM potassium phosphate buffer (pH 7.4) or 20.5 μM benzyloxyresorufin in Tris buffer, and 1 mM NADPH. Control incubations containing 62.5 mM potassium phosphate buffer or Tris buffer and 1% methanol without inhibitors were performed in parallel. Rates of HBUP or resorufin formation in the presence of inhibitors were expressed as the percentage of control activity. The IC50 values for inhibitors were determined by nonlinear regression analysis of the plot of the logarithm of inhibitor concentration versus percentage of remaining activity using GraphPad Prism. The enzyme activities in the presence of inhibitors were compared with the control incubation.
Enzyme Kinetic Analyses.
The kinetics of BUP hydroxylation for PB-induced RLM and for microsomes containing cDNA-expressed CYP2B1, CYP2C6, CYP2C11, CYP2E1, and CYP3A1 were examined over the concentration range of 25 to 2500 μM. Kinetic parameter estimates were selected by visual inspection of Michaelis-Menten and Lineweaver-Burk of experimental data. Weighted kinetic data were fit to one-component Michaelis-Menten models via iterative nonlinear regression analysis using initial parameter estimates or substrate inhibition model (GraphPad Software Inc.). The apparent Km and Vmax values of HBUP formation were estimated from the fitted data. Determination of appropriateness of fit to each kinetic model was accomplished by examination of the sum of squares of residuals, the size of the coefficients of variations, and S.E.s of the parameter estimates.
Results
Correlation between BUP Hydroxylase and BROD Activities in RHM.
Correlations between BUP hydroxylase and BROD activities were examined in RHM isolated from various rat hepatocyte cultures, controls (n = 5), or treated with CYP2B1 inducer (PB, n = 3), CYP3A1 inducer (DEX, n = 5), CYP4A inducer (FEN, n = 4), or CYP1A2 inducers (3-MC or BNF, n = 5) (Fig. 2). The substrates were used in excess, i.e., at a concentration of 500 μM BUP and 20.5 μM benzyloxyresorufin. When including 3-MC- and BNF-treated groups, BUP hydroxylase activity did not correlate well with BROD activity (r2 = 0.49, p = 0.014; Fig. 2, ■). A notable difference between BROD and BUP hydroxylase activities was with respect to CYP1A2 inducer (3-MC and BNF) treatment. Both compounds induced BROD activity but did not induce BUP hydroxylase activity. In contrast, BUP hydroxylase activity from controls, DEX-, FEN-, and PB-treated group (4.17–53.72 pmol · min−1 · mg protein−1) significantly correlated with BROD (0.15–27.87 pmol · min−1 · mg protein−1; r2 = 0.81, p < 0.01; Fig. 2, ♦). This suggests that BROD was not selective for CYP2B1 activity, because neither 3-MC nor BNF would be expected to induce CYP2B1.
Correlation between BUP hydroxylase and BROD activities in microsomes from rat hepatocytes, control, and treated with CYP2B, CYP3A1, and CYP4A inducers (n = 17; ♦) or with CYP1A2 inducers (n = 5; ■). BUP hydroxylase activity was determined by incubating 500 μM BUP, 0.05 mg of microsomal protein, 62.5 mM KH2PO4 (pH 7.4), 1 mM EDTA, 3 mM MgCl2, and 1 mM NADPH for 30 min. HBUP formation was quantitated by HPLC and normalized for microsomal protein and incubation time. BROD activity was determined by incubating with 20.5 μM benzyloxyresorufin. Resorufin formation was quantified by spectrophotometry. The correlation of the two activities was evaluated by linear regression analysis. Statistical significance was assessed by an F test.
Inhibition Experiments.
The pilot experiment was performed to confirm the ability of proadifen and metyrapone to inhibit the BUP hydroxylation and BROD by using cDNA-expressed CYP2B1. Proadifen and metyrapone effectively inhibited cDNA-expressed CYP2B1-dependent BUP hydroxylase activity with an IC50 of 14.9 ± 3.7 and 7.52 ± 0.54 μM, respectively (Table 1). These two inhibitors also effectively inhibited cDNA-expressed CYP2B1-dependent BROD activity with an IC50 of 38.1 ± 1.3 and 1.85 ± 1.11 μM, respectively (Table 1). To determine the potential substrate selectivity of BUP for CYP2B1 but not CYP1A2 involved in these reactions, liver microsomes from BNF- and PB-pretreated rats were used. BROD activity and BUP hydroxylation were measured in these microsomes preincubated or not with three known P450 inhibitors in the presence of substrate before the addition of NADPH, which initiates enzyme activity (Table 1). When incubated with PB-induced RLM, proadifen (CYP1A2/2B1 inhibitor) and metyrapone (CYP2B1 inhibitor) caused a great reduction in both BROD and BUP hydroxylation, the IC50 being 1.21 ± 0.49 and 3.89 ± 1.54 μM for BUP hydroxylation and 0.688 ± 0.161 and 0.806 ± 0.663 μM for BROD, respectively, as seen in Table 1. By contrast, in BNF-induced RLM, the IC50s for BROD by ANF (CYP1A2 inhibitor) and proadifen were much lower (2.5 × 10−3 ± 0.8 × 10−3 μM) than that of BUP hydroxylation (>100 μM), whereas the IC50s for BROD activity and BUP hydroxylation were equivalent (40.8 ± 4.6 and 41.8 ± 3.4 μM, respectively) in the presence of metyrapone (CYP2B1 inhibitor). Taken together, these results suggest that CYP2B1 is similarly involved in both activities, whereas CYP1A1/2 is involved in BROD activity, but not in BUP hydroxylation.
Inhibition of BROD and BUP hydroxylase activities in BNF- and PB-pretreated RLM and cDNA-expressed CYP2B1 by ANF, proadifen, and metyrapone
BUP hydroxylase activity was determined by incubating 500 μM BUP, 0.05 mg of microsomal protein, and 1 mM NADPH for 30 min. HBUP formation was quantitated by HPLC. BROD activity was determined by incubating with 20.5 μM benzyloxyresorufin and 1 mM NADPH for 2 min. Resorufin formation was quantitated by spectrofluorometry. IC50 values are the mean ± S.E.
Evaluation of BUP Hydroxylation by Individual cDNA-expressed Rat P450s.
A panel of 14 cDNA-expressed enzymes (Supersomes) was screened for BUP hydroxylase activity at 500 and 2500 μM BUP (Fig. 3). These substrate concentrations were selected to ensure saturation of any high Km isozyme capable of catalyzing BUP hydroxylation. cDNA-expressed CYP2B1 exhibited the highest percentage of total BUP hydroxylation at 2500 μM BUP (75%), compared to 63% at 500 μM BUP. Two other cDNA-expressed P450s also contributed to BUP hydroxylation but at lower rates: CYP2C11 participated for 23 and 8.7% of total HBUP, at 500 and 2500 μM BUP, respectively, and CYP2E1 participated for 1.8 and 10.9% of total HBUP, respectively, at 500 and 2500 μM BUP. Rates of BUP hydroxylation by CYP1A1, CYP1A2, CYP2A1, CYP2A2, CYP2C6, CYP2C12, CYP2C13, CYP2D1, CYP2D2, CYP3A1, and CYP3A2 were less than 10% of total HBUP formation. HBUP formation in control microsomes was undetectable.
Rates of HBUP formation were assessed in microsomes from baculovirus-infected insect cells (Supersomes) that expressed cDNA for NADPH-P450 oxidoreductase and CYP1A1, CYP1A2, CYP2A1, CYP2A2, CYP2B1, CYP2C6, CYP2C11, CYP2C12, CYP2C13, CYP2D1, CYP2D2, CYP2E1, CYP3A1, or CYP3A2. HBUP formation was quantitated by LC/MS/MS and normalized for picomoles of P450 and incubation time. BUP concentration, P450 amount, and incubation time were 500 and 2500 μM, 50 pmol, and 30 min, respectively. Control incubations were also conducted with insect cell-derived microsomes transfected with baculovirus only. Data are expressed as percentage of total HBUP formation.
Kinetic Analyses of BUP Hydroxylation.
Kinetic parameters of BUP hydroxylation were first estimated by fitting kinetic data with PB-pretreated RLM (Fig. 4A) and by fitting kinetic data with cDNA-expressed CYP2B1 (Fig. 4B) to the single enzyme Michaelis-Menten equation. The apparent Km and Vmax for BUP hydroxylation in the PB-pretreated RLM with single enzyme kinetics were 158.5 ± 23.1 μM and 1697 ± 81 pmol · min−1 · mg protein−1, respectively (Table 2), and the apparent Km and Vmax for BUP hydroxylation by cDNA-expressed CYP2B1 were 152.5 ± 11.5 μM and 181.2 ± 5.5 pmol · min−1 · pmol P450−1, respectively (Table 2). Kinetic parameters of BUP hydroxylation were also estimated by fitting kinetic data with cDNA-expressed CYP2E1, -2C6, -2C11, and -3A1. The apparent Km and Vmax for BUP hydroxylation by cDNA-expressed CYP2E1 were 914 ± 305 μM and 0.143 ± 0.020 pmol · min−1 · pmol P450−1, respectively (Fig. 5A). Substrate inhibition for BUP hydroxylation was observed with cDNA-expressed CYP2C6 (Fig. 5B), -2C11 (Fig. 5C), and -3A1 (Fig. 5D) for BUP concentrations over 500 μM. Using the substrate inhibition model (Lin et al., 2001), the Km values for BUP hydroxylation by cDNA-expressed CYP2C6, -2C11, and -3A1 were 67.8 ± 11.8, 190 ± 52.5, and 453 ± 81.7 μM, respectively. Vmax for BUP hydroxylation by cDNA-expressed CYP2C6, -2C11, and -3A1 were 0.219 ± 0.016, 1.65 ± 0.30, and 0.376 ± 0.044 pmol · min−1 · pmol P450−1, respectively, i.e., more than 100 times lower than that of cDNA-expressed CYP2B1.
Michaelis-Menten plots of HBUP after the incubation of BUP (concentration range 25–2500 μM) with PB-induced RLM (A) or microsomes containing cDNA-expressed CYP2B1 (B); each value is a mean ± S.E.M. of triplicate determinations.
Kinetic parameters of BUP hydroxylation in PB-treated RLM and in microsome containing cDNA-expressed CYP2B1
Rates of HBUP formation were determined in triplicate at 25 to 2500 μM BUP in RLM and in CYP2B1-containing insect cell-derived microsomes. HBUP formation was quantitated by LC/MS/MS. Weighted concentration-rate data were fit to a one-component Michaelis-Menten model by nonlinear regression using the GraphPad program. Kinetic parameters were estimated from the fitted data.
Michaelis-Menten plot of HBUP after the incubation of BUP (concentration range, 25–2500 μM) with microsomes containing cDNA-expressed CYP2E1 (A) and substrate-inhibition plot of HBUP after the incubation of BUP (concentration range, 25–2500 μM) with microsomes containing cDNA-expressed CYP2C6 (B), microsomes containing cDNA-expressed CYP2C11 (C), and microsomes containing cDNA-expressed CYP3A1 (D); each value is a mean ± S.E.M. of triplicate determinations.
Discussion
Over the past several decades, CYP2B enzymes have served as prototypical models for investigation of the mechanism by which drugs and environmental contaminants activate gene expression. CYP2B enzymes are also very versatile catalysts with a broad range of substrates including drugs, environmental pollutants, and steroids (Kedzie et al., 1991). CYP2B is the main hepatic P450 isoform inducible by PB and other barbiturates in experimental animals, CYP2B-mediated biotransformations being extensively studied both in small rodents and in rabbit (Nims and Lubet, 1996). In humans, BUP hydroxylation has been shown to be a selective marker of CYP2B6 (Faucette et al., 2000), but to our knowledge, a fully selective probe for CYP2B1 catalytic activity in rodent, which would facilitate further examination of the role of this enzyme in xenobiotic metabolism, has not been described. Although BROD is routinely used for evaluating CYP2B1 activity (Burke et al., 1985, 1994; Lubet et al., 1985), it can also be partly related to CYP1A2 activity (Kobayashi et al., 2002; Chovan et al., 2007). In a previous study (Richert et al., 2009), we found a good correlation between BUP hydroxylation and CYP2B1 mRNA expression in rat hepatocyte cultures. The results of the present study support the use of BUP at saturating concentrations as a selective in vitro probe substrate for the determination of CYP2B1 catalytic activity.
Previous reports described that BROD reaction clearly involved multiple enzymes: according to Kobayashi et al. (2002), CYP2B1 accounted for 60% and CYP1A2 for 35%; and according to Chovan et al. (2007), the respective involvements of CYP2B1 and CYP1A2 were 32 and 34%. In the present study, the apparent Km for BUP hydroxylation in PB-pretreated RLM (158.5 ± 23.1 μM) was equivalent to that of cDNA-expressed CYP2B1 (152.5 ± 11.5 μM). Although this observation alone cannot support the conclusion that a single enzyme is involved in BUP hydroxylation in RLM, these results are in accordance with data obtained with human CYP2B6 for which it has been reported that BUP is metabolized primarily by CYP2B6 to HBUP with an apparent Km of 107.5 ± 20.5 μM (Ekins et al., 1999) or of 155.8 ± 18.2 μM (Faucette et al., 2000) in human cDNA-expressed CYP2B6 and an apparent Km of 130.2 ± 22.0 μM in human liver microsomes (Faucette et al., 2000).
The results from inhibition experiments by using IC50 values as a measure of the efficacy of inhibition of microsomal CYP1A2 and CYP2B1 activities further suggest the selectivity of CYP2B1 for BUP hydroxylation but not BROD. Proadifen and metyrapone effectively inhibited cDNA-expressed CYP2B1-dependent BUP hydroxylase activity with an IC50 of 14.9 ± 3.7 and 7.52 ± 0.54 μM, respectively. IC50s for BROD and BUP hydroxylation were equivalent (40.8 ± 4.6 and 41.8 ± 3.4 μM, respectively) when using liver microsomes from BNF-pretreated rats in the presence of metyrapone, a CYP2B1-selective inhibitor. However, when using liver microsomes from rats pretreated with BNF, in the presence of ANF, a CYP1A1/2-selective inhibitor, we found an IC50 of 2.5 × 10−3 ± 0.8 × 10−3 μM for BROD and >100 μM for BUP hydroxylation. These results suggest that CYP2B1 is similarly involved in both activities, whereas CYP1A1/2 is involved in BROD activity, but not in BUP hydroxylation. This is further supported by the good correlation of BUP hydroxylation with BROD activity (r2 = 0.81, p < 0.01; Fig. 2) when considering control, DEX-, FEN-, and PB-treated rat hepatocytes cultures but poor correlation when including rat hepatocytes treated with CYP1A2 inducer (3-MC and BNF) (r2 = 0.49, p = 0.014). In CYP1A2-induced rat hepatocytes, BROD activity was increased, whereas BUP hydroxylase activity was unchanged, clearly demonstrating that BROD is not selective for CYP2B1 activity, because neither 3-MC nor BNF would be expected to induce CYP2B1. Our data confirm the principle role of CYP2B1 in catalyzing HBUP formation, as suggested by our finding of a good correlation between CYP2B1 expression and BUP hydroxylation but not BROD activity (Richert et al., 2009).
Since the early 1990s, the use of cDNA-expressed P450s has assisted, in addition to liver microsomes, in the evaluation of metabolic specificity of probe substrates and the identification of the P450 enzymes involved in the metabolism of xenobiotics in human and rat liver microsomes. We show that among a panel of 14 rat cDNA-expressed P450 isozymes (CYP1A1, CYP1A2, CYP2A1, CYP2A2, CYP2B1, CYP2C6, CYP2C11, CYP2C12, CYP2C13, CYP2D1, CYP2D2, CYP2E1, CYP3A1, and CYP3A2), the rate of BUP hydroxylation at high concentrations of BUP (500 and 2500 μM) was highest with CYP2B1. When using BUP at 2500 μM to ensure saturation of any high Km isozyme capable of catalyzing BUP hydroxylation, two other P450 isozymes, i.e., CYP2C11 and CYP2E1, were also found to be catalytically competent (8.7 and 10.9% of total HBUP formation, respectively, compared to 75% for CYP2B1). Previous in vitro studies dealing with the involvement of human P450 using various BUP concentrations reached similar conclusions (Chen et al., 2010); in addition, Faucette et al. (2000, 2001) and Hesse et al. (2000) both reported that, although CYP2B6 was the major isoform (71%) involved in BUP hydroxylation, CYP2E1 and CYP3A4 can also be involved in BUP hydroxylation (24 and 2%, respectively).
Kinetic parameters of BUP hydroxylation were estimated by fitting kinetic data with cDNA-expressed CYP2C6, -2C11, -2E1, and -3A1. Kinetic data for cDNA-expressed CYP2E1 (Km of 914 ± 305 μM and Vmax of 0.143 ± 0.02 pmol · min−1 · pmol P450−1) and for cDNA-expressed CYP3A1 (Km of 453 ± 51 μM and Vmax of 0.376 ± 0.044 pmol · min−1 · pmol P450−1) reveal that these isoforms are not involved in BUP hydroxylation. By contrast, Km values for BUP hydroxylation with cDNA-expressed CYP2C6 (67.8 ± 11.8 μM) and CY2C11 (190 ± 52.5 μM) were close to that of CYP2B1 (152.5 ± 11.5 μM) and suggest that at low concentrations of BUP, the CYP2C isoforms can be involved in BUP metabolism. Because plasmatic concentrations of BUP in rat have been shown to be 4 μM (Suckow et al., 1986), BUP cannot be used in vivo as a marker for CYP2B1 activity. However, the apparent Vmax values for CYP2C isoforms were very low (0.22 ± 0.02 pmol · min−1 · pmol P450−1 for CYP2C6 and 1.65 ± 0.30 pmol · min−1 · pmol P450−1 for CYP2C11) compared to Vmax for BUP hydroxylation with cDNA-expressed CYP2B1 (181.2 ± 5.5 pmol · min−1 · pmol P450−1). In addition, substrate inhibition of the CYP2C isoforms occurs from BUP concentrations over 500 μM, whereas no such effect was found with CYP2B1. As a consequence, the involvement of the CYP2C isoforms in BUP hydroxylation at saturating concentrations of BUP is negligible compared to that of CYP2B1.
Taken together, the data suggest that BUP is a selective probe substrate for CYP2B1, characterized by a relatively high Km value (158.5 μM) that maintains P450 selectivity due to the absence of other significantly contributing P450 isozymes, especially at saturating BUP concentrations. Other selective substrates have high Km values such as the caffeine for human CYP1A2 (Km ∼ 1.2 mM) (Hickman et al., 1998), tolbutamide for human CYP2C9 (Km ∼ 200 μM) (Hickman et al., 1998), and testosterone for rat CYP3A1 (Km ∼ 150 μM) (Cooper et al., 1993).
In conclusion, this study validates BUP hydroxylation as an in vitro diagnostic marker for CYP2B1 catalytic activity when assayed at >500 μM BUP. This will now allow the assessment of the contribution of CYP2B1 to the metabolism of the given drug.
Authorship Contributions
Participated in research design: Pekthong, Martin, and Richert.
Conducted experiments: Pekthong and Desbans.
Performed data analysis: Pekthong, Martin, and Richert.
Wrote or contributed to the writing of the manuscript: Pekthong, Martin, and Richert.
Acknowledgment
We thank A. Bonet for technical assistance with the rat induction study.
Footnotes
This work was funded by KaLy-Cell [Grant Université de Franche-Comté-KaLy-Cell, number I233].
Article, publication date, and citation information can be found at http://dmd.aspetjournals.org.
-
ABBREVIATIONS:
- P450
- cytochrome P450
- ANF
- α-naphthoflavone
- BNF
- β-naphthoflavone
- BROD
- benzyloxyresorufin-O-dealkylase
- BUP
- bupropion
- HBUP
- hydroxybupropion
- PB
- phenobarbital
- RHM
- rat hepatocyte microsomes
- RLM
- rat liver microsomes
- BSA
- bovine serum albumin
- DEX
- dexamethasone
- 3-MC
- 3-methylcholantrene
- FEN
- fenofibrate
- LC/MS/MS
- liquid chromatography/tandem mass spectrometry
- HPLC
- high-performance liquid chromatography
- ESI
- electrospray ionization.
- Received June 24, 2011.
- Accepted September 30, 2011.
- Copyright © 2012 by The American Society for Pharmacology and Experimental Therapeutics
References
In this issue
BUPROPION HYDROXYLATION A MARKER OF RAT CYP2B1 ACTIVITY
