![]() |
|
|
Department of Drug Metabolism, Merck Research Laboratories, West Point, Pennsylvania (B.M., R.S., M.L.S., C.T.); and Bristol-Myers Squibb, Princeton, New Jersey (A.D.R.)
(Received January 9, 2004; accepted February 13, 2004)
| Abstract |
|---|
|
|
|---|
-hydroxylation (r2
0.92; p < 0.0001) and amodiaquine N-de-ethylation (r2
0.91; p < 0.0001) in a bank of human liver microsomes (n = 15 organ donors). Finally, compound A hydroxylase activity conformed to Michaelis-Menten kinetics, and the Km (Michaelis constant) in human liver microsomes was similar to that of CYP2C8 (
10 µM). It is concluded that the hydroxylation of compound A is mainly catalyzed by CYP2C8, and thus the reaction can possibly serve as an alternative marker assay for CYP2C8 in human liver microsomes.
In recent years, there has been a growing interest in the involvement of CYP2C8 in drug metabolism. CYP2C8 has been reported to play a role in the biotransformation of paclitaxel, cerivastatin, rosiglitazone, troglitazone, verapamil, amodiaquine, amiodarone, and zopiclone (Baldwin et al., 1999
; Becquemont et al., 1999
; Tracy et al., 1999
; Yamazaki et al., 1999
; Ohyama et al., 2000
; Li et al., 2002
; Wang et al., 2002
). In the case of cerivastatin, a hydroxymethylglutaryl coenzyme A reductase inhibitor withdrawn from the market in 2001, patients were shown to have elevated cerivastatin levels in plasma when gemfibrozil was coadministered (Backman et al., 2002
). The inhibition of cerivastatin metabolism by gemfibrozil was, in part, mediated by CYP2C8 (Prueksaritanont et al., 2002
; Wang et al., 2002
). It is therefore of interest to prospectively evaluate drug candidates as inhibitors of CYP2C8. Toward this end, paclitaxel 6
-hydroxylation has been used as a marker assay for CYP2C8 (Rahman et al., 1994
), and, more recently, amodiaquine N-de-ethylation has been proposed (Li et al., 2002
).
In the present study, the oxidative metabolism of compound A, a potent endothelin ETA receptor antagonist (Okada et al., 2000
), was studied after incubation with human liver microsomes. Preliminary data indicated that the hydroxylation on the isopropyl moiety (Fig. 1) was catalyzed almost exclusively by CYP2C8. Therefore, we sought to determine the P450s involved in the reaction using inhibitory antibodies, chemical inhibitors, correlation analysis, and recombinant human P450 proteins.
|
| Materials and Methods |
|---|
|
|
|---|
-Hydroxypaclitaxel was purchased from BD Biosciences Discovery Labware (Bedford, MA). Human liver microsomes prepared from individual or pooled (n = 20) subjects were purchased from BD Biosciences Discovery Labware, Tissue Transformation Technologies (Edison, NJ), Xeno-Tech, LLC (Lenexa, KA), and In Vitro Technologies, Inc. (Baltimore, MD). Microsomes prepared from insect cells infected with baculovirus expressing human P450 isoforms (CYP1A2, 2A6, 2B6, 2C8, 2C9, 2C19, 2D6, 2E1, 3A4, and 3A5) were obtained in-house. Inhibitory mouse anti-human P450 antibodies selective for CYP2A6, CYP2C, CYP2D6, and CYP3A were prepared in-house (Mei et al., 1999Incubations with Human Liver Microsomes and Recombinant Human P450 Isoforms. Incubation mixtures (0.5 ml) contained 50 mM potassium phosphate (pH 7.4), 1 mM magnesium chloride, 0.1 mg/ml pooled human liver microsomal protein or 10 pmol/ml recombinant human P450, 1 mM NADPH, and 0.1 to 100 µM compound A dissolved in acetonitrile [1% (v/v) final concentration]. The reaction was started by the addition of compound A after a 2-min preincubation at 37°C in a shaking water bath. Acetonitrile (1 ml) was added after 3 min to terminate the reaction. A 50-µl aliquot of a 20 nM stock solution of verapamil was added to the reaction mixture as internal standard. After centrifugation at 2000g for 10 min, supernatant fractions were transferred to clean tubes and dried under nitrogen in a TurboVap evaporator (Zymark Corp., Hopkinton, MA). Dried samples were reconstituted in 0.25 ml of 5% (v/v) acetonitrile in water containing 0.1% (v/v) formic acid for LC-MS analysis. Under these conditions, the formation rate of all four metabolites of compound A was linear with respect to protein and P450 concentrations as well as time of incubation.
Immunoinhibition Studies. Pooled human liver microsomes (0.05 mg protein) were preincubated on ice (1015 min) with the anti-P450 antibody preparation (2 µl) or control sera prepared from mouse ascites. The ratio of microsomal protein-to-antibody volume was chosen to give a maximal inhibitory effect. After the preincubation with antibody, the sample was diluted with potassium phosphate buffer and the other assay components as described above. Immunoinhibition was evaluated at two concentrations of compound A (2 and 50 µM).
Chemical Inhibition Experiments. Known P450 isoform-selective chemical inhibitors for CYP1A2 (furafylline), CYP2C9 (sulfaphenazole), CYP2C19 [(-)-N-3-benzyl-phenobarbital], CYP2D6 (quinidine), and CYP3A (troleandomycin) were used to perform inhibition experiments. Concentrations of furafylline (50 µM), sulfaphenazole (10 µM), (-)-N-3-benzyl-phenobarbital (5 µM), quinidine (5 µM), and troleandomycin (25 µM) were chosen to give maximal inhibitory effect, based on previously reported results (Newton et al., 1995
; Suzuki et al., 2002
). Stock solutions of chemical inhibitors were prepared in 50% acetonitrile/water (v/v). The final concentration of acetonitrile in the incubation mixture was 1.5% (v/v). Control incubations contained the same volume of acetonitrile but no inhibitor. In the experiments with furafylline and troleandomycin, inhibitors were preincubated with liver microsomes and NADPH for 30 min at 37°C before adding the substrate. All other inhibitors were coincubated with the substrate. Chemical inhibition experiments were performed at the substrate concentrations of 2 and 50 µM and were determined using pooled human liver microsomes.
LC-MS Analysis. Chromatographic separation of compound A and its metabolites was carried out on an Agilent 1100 LC system (Agilent Technologies, Palo Alto, CA) using a Phenomenex Synergi Hydro-RP column (2.0 x 150 mm, 4 µm; Phenomenex, Torrance, CA). Solvent A consisted of 25 mM ammonium formate, pH 7.0, and solvent B of acetonitrile, and was delivered at a constant flow rate of 0.25 ml/min. The initial mobile phase consisted of 30% of solvent B, which was linearly increased to 50% over 10 min, held for 3 min, then increased to 80% in another 5 min. The column was then equilibrated under initial conditions for 8 min.
Mass spectrometric analysis was performed on a Thermo Finnigan TSQ Quantum triple quadruple mass spectrometer using an electrospray ionization probe (Thermo Finnigan, San Jose, CA). Electrospray ionization was operated in a positive mode, with a spray voltage of 5.0 kV, a sheath gas flow of 40, an auxiliary gas flow of 18, and a capillary temperature of 350°C. To perform collision-induced dissociation, the mass isolation window, the collision energy, and the collision gas flow were set at 1 m/z unit, 34 V, and 1.5 mtorr, respectively. For quantitation, selected reaction monitoring experiments were performed to detect ion pairs at m/z 533/339 (compound A), 549/355 (hydroxy metabolite), and 455/165 (verapamil; internal standard).
NMR Analysis. Proton NMR studies were carried out on metabolite I (hydroxy metabolite) isolated from a human liver microsomal incubate. Isolated metabolite was dissolved in 160 µl of CD3OD and transferred to a 3-mm NMR tube. All NMR data were acquired at 25°C on an Inova spectrometer operating at a nominal proton frequency of 500 MHz (Varian, Inc., Palo Alto, CA) equipped with a cryogenically cooled 5-mm HCN triple-resonance probe (Varian, Inc.). In a cryogenically cooled probe, the radio frequency detection coil is maintained at 25 K, thereby reducing the thermal noise contribution, which in turn raises the signal-to-noise ratio of an NMR peak. In our setup, for a given sample in a 3-mm tube, a signal-to-noise ratio gain of approximately 3.3 is seen from a cryogenically cooled probe compared with the one-dimensional spectrum acquired in a 3-mm conventional HCN triple-resonance probe.
Measurement of CYP2C8 Activities. Paclitaxel (30 µM) and amodiaquine (10 µM) were used as probe substrates for CYP2C8 in human liver microsomes. Assays for paclitaxel 6
-hydroxylation and amodiaquine N-de-ethylation were performed as described by Rahman et al. (1994
) and Li et al. (2002
), respectively. For each assay, the final substrate concentration was 5 times the reported Km value. The rates of compound A hydroxylation (2 and 50 µM) determined across a panel of 15 individual human liver microsomes were correlated with both paclitaxel 6
-hydroxylase and amodiaquine N-de-ethylase activities. In each case, the correlation coefficient (r2) was determined graphically using SigmaPlot (SPSS Inc., Chicago, IL).
Kinetic Analysis. The apparent Km and Vmax values were estimated by fitting the untransformed data to the Michaelis-Menten model, using SigmaPlot. The data were also analyzed by linear transformation (Eadie-Hofstee plot) to confirm a single Km system.
| Results and Discussion |
|---|
|
|
|---|
|
The kinetics of compound A isopropyl hydroxylation in pooled human liver microsomes were evaluated and conformed to the Michaelis-Menten model (Fig. 3A). These data suggested that one or more enzymes with similar Km values were responsible for the reaction. The apparent Km, Vmax, and Vmax/Km values were estimated to be 6.6 µM, 28 pmol/min/mg protein, and 4.3 µl/min/mg protein, respectively. Subsequent kinetic study showed that the apparent Km value determined in CYP2C8 (11.8 µM; Fig. 3B) was similar to that obtained in human liver microsomes. It was previously reported that compound A also underwent sugar conjugation in human liver microsomes (Tang et al., 2003
). The intrinsic clearance (Vmax/Km) of glucuronidation (42 µl/min/mg protein) and glucosidation (18 µl/min/mg protein) was higher than that of the hydroxylation, consistent with the observation that sugar conjugates were more abundant than the hydroxy metabolite (I) in human hepatocytes (data not shown).
|
The formation of metabolite I was monitored in a panel of 10 recombinant human P450 isoforms (CYPs 1A2, 2A6, 2B6, 2C8, 2C9, 2C19, 2D6, 2E1, 3A4, and 3A5). Compound A hydroxylation was only detected in CYP2C8-containing incubation mixtures. To confirm the observation that only CYP2C8 exhibited compound A isopropyl hydroxylase activity, pooled human liver microsomes were preincubated with inhibitory monoclonal antibody preparations raised against human P450 isoform(s). At a compound A concentration of 2 µM, preincubation with anti-CYP2C8 and anti-CYP2C monoclonal antibodies completely abolished the formation of metabolite I (Fig. 4). However, inhibitory antibodies selective for other P450 isoforms had no significant effect, indicating that metabolite I formation is catalyzed exclusively by CYP2C8. A similar inhibition study was conducted using previously characterized P450 isoform-selective inhibitors, furafylline (CYP1A2), sulfaphenazole (CYP2C9), (-)-N-3-benzyl-phenobarbital (CYP2C19), quinidine (CYP2D6), ketoconazole (CYP3A), and troleandomycin (CYP3A). As shown in Fig. 4, the majority of the inhibitors had a minimal effect on metabolite I formation. The 20 to 30% inhibition by ketoconazole and quinidine was consistent with, and possibly caused by, nonselective inhibition of CYP2C8 (Ong et al., 2000
; Sai et al., 2000
; Dierks et al., 2001
; Li et al., 2002
).
|
The correlation between the hydroxylase activity of compound A and the activities of two marker substrates of CYP2C8 was determined using a panel of 15 human liver microsomes (Fig. 5). Compound A (2 µM) hydroxylation correlated well with paclitaxel 6
-hydroxylation (r2
0.92; p < 0.0001) and amodiaquine N-deethylation (r2
0.91; p < 0.0001).
|
Interestingly, the hydroxylation of compound A was nonlinear after a short (5 min) incubation, at relatively low microsomal protein content (0.1 mg/ml; data not shown), suggesting that significant product inhibition may occur during the biotransformation of compound A. It is well documented that N-dealkylation and methylenedioxyphenyl oxidation (and subsequent catechol formation) may result in the formation of a metabolite-P450 complex, which leads to quasi-reversible inhibition of P450 activity (Murray and Reidy, 1990
; Ma et al., 2000
; Chatterjee and Franklin, 2003
). Since CYP2C8, along with other P450s, catalyzes both N-dealkylation and methylenedioxyphenyl oxidation to form metabolites II and III, respectively (data not shown), product inhibition may result from metabolite-P450 complex formation through either or both mechanisms mentioned above.
Based on the in vitro data described, it is concluded that CYP2C8 is the major P450 isoform that catalyzes the formation of the hydroxy metabolite (I) in human liver microsomes. The kinetics of compound A hydroxylation performed using human liver microsomes and recombinant CYP2C8 conform to the Michaelis-Menten model, with apparent Km values of
10 µM. Results obtained from the inhibition studies and correlation studies performed at the substrate concentration of 2 µM were essentially the same as those determined at 50 µM, indicating CYP2C8 was the predominant P450 isoform involved in the isopropyl hydroxylation of compound A over that concentration range.
For nearly a decade, paclitaxel 6
-hydroxylation has been used as a marker for CYP2C8 (Rahman et al., 1994
). Recently, amodiaquine was proposed as an alternative CYP2C8 probe substrate because of its high affinity and high turnover rate (Li et al., 2002
). The hydroxylation of compound A correlated well with both paclitaxel 6
-hydroxylation and amodiaquine N-de-ethylation (r2 >0.90; Fig. 5), suggesting that the hydroxylation of compound A can be used as an alternative marker assay for CYP2C8 activity in human liver microsomes. The assay described herein requires a short incubation period (3 min) and a relatively low protein concentration (0.1 mg/ml). Also, the LC-MS/MS assay was more sensitive and less prone to interference by other compounds. Even though the limited availability of the compound A makes it less desirable, the current assay offers a quick and sensitive method to assess CYP2C8 activity in human liver microsomes.
| Acknowledgments |
|---|
| Footnotes |
|---|
Address correspondence to: Bennett Ma, Department of Drug Metabolism, WP 75A-203, Merck Research Laboratories, West Point, PA 19486. E-mail: bennett_ma{at}merck.com
| References |
|---|
|
|
|---|
-hydroxytaxol by human cytochrome P450 2C8. Cancer Res 54: 5543-5556.This article has been cited by other articles:
![]() |
B. Ma, S. L. Polsky-Fisher, S. Vickers, D. Cui, and A. D. Rodrigues Cytochrome P450 3A-Dependent Metabolism of a Potent and Selective {gamma}-Aminobutyric AcidA{alpha}2/3 Receptor Agonist in Vitro: Involvement of Cytochrome P450 3A5 Displaying Biphasic Kinetics Drug Metab. Dispos., August 1, 2007; 35(8): 1301 - 1307. [Abstract] [Full Text] [PDF] |
||||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||