Abstract
RP 73401 is a potent inhibitor of cyclic nucleotide phosphodiesterase type IV. RP 73401 is metabolized by human liver microsomes almost exclusively by transhydroxylation of the cyclopentyl group to RPR 113406. Liquid chromatography/mass spectrometry/mass spectrometry analysis of plasma from patients given RP 73401 also revealed a molecular ion and fragmentation consistent with RPR 113406. Thus, the objective was to determine the oxidative enzyme(s) responsible for RP 73401 hydroxylation. Kinetic constants of RP 113406 formation ranged from 8 to 26 μM and 0.83 to 5.99 nmol/min/mg protein forKm and Vmax , respectively (n = 3). Enzyme activity varied 23-fold among 15 human liver microsome samples and correlated with CYP2A6-catalyzed coumarin hydroxylase (r2 = 0.85, P < .01) and CYP2B6-catalyzed 7-ethoxytrifluoromethylcoumarin O-deethylase (r2 = 0.82, P < .01) activities. Chemical inhibition studies showed a 63% decrease in RP 73401 hydroxylation by 500 μM orphenadrine. Coumarin (10 μM), however, did not inhibit RP 73401 hydroxylation. Also, anti-CYP2B1 IgG produced 85% inhibition of RP 73401 hydroxylation, but only a negligible decline in coumarin hydroxylase activity. Of the 10 expressed P450 forms studied, only CYP2B6 catalyzed RP 73401 hydroxylation. Finally, expressed CYP2B6 showed a high affinity (K m = 22.5 μM) for RP 73401 hydroxylation, similar to the human liver microsome studies.
The cyclic nucleotide PDE isozymes have been classified into seven families based on sequence homology and functional characteristics (Michaeliet al., 1993). The function and relative amount of each form is dependent on the tissue and cell type examined. Although the lack of specific PDE inhibitors has made it difficult to evaluate the role of specific isozymes in modulating tissue function, PDE IV inhibitors have been shown clearly to effectively reduce the activation of inflammatory cells. The hypothesis that this inflammation response is critical to the etiology of chronic arthritis creates a potential therapeutic application for PDE IV inhibitors. RP 73401 [3-cyclopentyloxy-N-(3,5-dichloro-4-pyridyl)-4-methoxybenzamide or Piclamilast] (fig. 1) is a potent inhibitor of cyclic nucleotide PDE IV. Inhibition of PDE IV by RP 73401 results in an increase in cAMP levels in vitro, which is presumably responsible for the anti-inflammatory effects observed for RP 73401 in vivo (Raeburn et al., 1994). With the advancement of RP 73401 in clinical trials, studies have been conducted to determine the metabolic fate of RP 73401 both in vivo andin vitro.
Pathways, interactions and variations of human drug metabolism can often be traced to the cytochrome P450 enzyme superfamily (Guengerich, 1995; Wrighton and Stevens, 1993). A review of the P450 literature has found that most drug metabolism reactions can be accounted for by CYP1A2, CYP2E1 and CYP2D6, in addition to the 2C and 3A subfamilies (Benet et al., 1996). In contrast to these major P450 forms, immunodetectable levels of CYP2B6 were found in only 24% of human liver samples by Mimura et al. (1993). CYP2B6 has been shown to contribute to the bioactivation of cyclophosphamide (Chang et al., 1993); however, clinical phenotyping studies to determine thein vivo role of CYP2B6 in drug metabolism have not been conducted. In fact, the lack of specific CYP2B6 probe substrates and inhibitors has made the accurate characterization of this enzyme difficult.
An understanding of the enzyme(s) involved in the metabolism of a new chemical entity is critical for predicting drug interactions and interindividual variability in metabolism and pharmacokinetics (Pecket al., 1993). Toward this goal, LC/MS analysis of plasma samples from patients treated with RP 73401 was used to identify that hydroxylation of the cyclopentyl functional group was a primary route of RP 73401 metabolism in vivo. In vitro metabolism studies with human liver microsomes were then conducted to identify the CYP450 enzyme(s) involved in the hydroxylation of RP 73401. Three complimentary approaches were used toward this objective: 1) correlation analysis of RP 73401 hydroxylase activity with marker P450 enzyme activities in a bank of human liver microsomes; 2) chemical and antibody inhibition of enzyme activity; and 3) analysis of RP 73401 hydroxylase activity by cDNA-expressed human P450 enzymes. The results show that RP 73401 is stereoselectively hydroxylated to RPR 113406 exclusively by CYP2B6.
Materials and Methods
Chemicals.
RP 73401, RPR 113406 (trans-hydroxycyclopentyl isomer), RPR 112919 (cis-hydroxycyclopentyl isomer) and RPR 100510 (N-oxide metabolite) were obtained from RPR, Dagenham Research Center (Dagenham, UK). Glucose 6-phosphate, NADP+, β-NADPH, G6PDH, nifedipine, coumarin, 7-hydroxycoumarin, tolbutamide, quinidine, TAO, orphenadrine and sodium perchlorate were purchased from Sigma Chemical Co. (St. Louis, MO). S-mephenytoin,S-nirvanol, 4′-hydroxy-S-mephenytoin, 4′-hydroxytolbutamide, furafylline, sulfaphenazole and 6-hydroxychlorzoxazone were obtained from Ultrafine Chemicals (Manchester, UK). Dehydronifedipine was synthesized as described previously (Pfister, 1990). HEPES and acetonitrile were purchased from J.T. Baker (Phillipsburg, NJ). 7-EFC was purchased from Molecular Probes, Inc. (Eugene, OR) and 7-HFC was purchased from Enzyme Systems Products (Dublin, CA). Ammonium acetate and perchloric acid were purchased from Fisher Scientific (Fairlawn, NJ) and acetic acid was purchased from EM Science (Gibbstown, NJ). All other chemicals were purchased from standard vendors and were of the highest purity available.
Biological reagents.
Human liver samples were obtained through organ procurement agencies (Anatomic Gift Foundation, Woodbine, GA; International Institute for the Advancement of Medicine, Exton, PA; and the National Disease Research Interchange, Philadelphia, PA) in accordance with proper ethical procedures for consent. Liver microsomes were prepared according to the procedure of Wang et al.(1983). Microsomal protein concentrations (Lowry et al., 1951) and P450 content (Omura and Sato, 1964) were determined as described previously. Rabbit anti-rat CYP2B1 and CYP3A1 antibodies were purchased from Human Biologics (Phoenix, AZ). Microsomes prepared from human lymphoblastoid cells transfected with individual human P450 forms (1A1, 1A2, 2A6, 2B6, 2C8/OR, 2C9/Arg/OR, 2C19, 2D6, 2E1/OR and 3A4/OR) were purchased from Gentest Corp. (Woburn, MA).
LC/MS analysis of plasma samples.
Plasma samples for LC/MS/MS analysis were obtained after the administration of a single inhaled dose of 800 μg [14C]RP 73401 to healthy male volunteers. Structural confirmation was accomplished on a SCIEX API III (Toronto, Canada) fitted with a turbo ionspray interface. The structural information was obtained from tandem mass spectrometry (LC/MS/MS) analysis. The standards and samples were introduced by use of a gradient HPLC system with a 20-μl sample loop. Separation of RP 73401 and its metabolites was achieved with a narrowbore (150 × 2 mm) Hypersil phenyl column (Keystone Scientific, Inc., Bellefonte, PA).
RP 73401 metabolite profiling.
A gradient HPLC method was devised to separate the cis-hydroxycyclopentyl metabolite (RPR 112919) from the corresponding trans isomer (RPR 113406). For these experiments, RP 73401 was incubated with human liver microsomes (RPR-HL-08, 0.2 mg) or expressed CYP2B6 (0.5 mg) for 10 and 20 min, respectively, in the presence of an NADPH-regenerating system. The metabolites were separated with use of a Spherisorb ODS-2, 250 × 4.6 mm, 5-μm analytical column (Phase Separations, Norwalk, CT) with a ODS-2 5-cm guard column and a flow rate of 1 ml/min. Solvents A (80:20, 10 mM ammonium acetate, pH 4.0/acetonitrile) and B (55:45, 10 mM ammonium acetate, pH 4.0/acetonitrile) were mixed by the following gradient: 0 to 12 min, 100% A; 12 to 22 min, 100% A → 100% B; 35 to 40 min, 100% B → 100% A.
Enzyme assays.
The following enzyme assays were used to monitor specific P450 forms as described (Heyn et al., 1996): phenacetin O-deethylation, CYP1A2; coumarin 7-hydroxylation, CYP2A6; 7-EFC deethylation and S-mephenytoin N-demethylation, CYP2B6; tolbutamide 4′-hydroxylation, CYP2C9;S-mephenytoin 4′-hydroxylation, CYP2C19; chlorzoxazone 6-hydroxylation, CYP2E1; and nifedipine oxidation, CYP3A4. Bufuralol 1′-hydroxylation was used to monitor CYP2D6 (Kronbach et al., 1987).
Incubations of RP 73401 with liver microsomes were conducted under the following conditions: 0.4 mg/ml microsomal protein, 100 mM potassium phosphate (pH 7.4), 1 mM NADP+, 4 U/ml G6PDH and 100 μM RP 73401 (total volume of 0.25 ml) were incubated for 3 min at 37°C in a shaking water bath. Reactions were started by the addition of 25 μl glucose 6-phosphate (10 mM final concentration). With the exception of the antibody inhibition experiments, the incubation time was 10 min. Reactions were terminated by the addition of 100 μl acetonitrile and the samples were centrifuged for 15 min at 5000 rpm. Samples were then transferred to HPLC vials before the injection of a 20-μl sample.
The kinetic parameters of RP 73401 hydroxylase activity were determined by incubating representative human liver microsome samples or expressed CYP2B6 with 1, 2, 4, 10, 20, 50, 100 and 200 μM RP 73401.Km and V maxvalues were calculated by the nonlinear regression program of GraphPad (GraphPad Software Inc., San Diego, CA) with use of unweighted raw data.
Chemical inhibition experiments used the following competitive inhibitors and concentrations; coumarin (CYP2A6, 10 μM), sulfaphenazole (CYP2C8/9, 100 μM), quinidine (CYP2D6, 0.5 μM) and 7-EFC (CYP2B6, 10 μM). Incubation mixtures containing furafylline (CYP1A2, 50 μM), TAO (CYP3A4, 25 μM), orphenadrine (CYP2B6, 500 μM) and diethyldithiocarbamate (CYP2E1, 40 μM) were preincubated with all components of the reaction except RP 73401 for 15 min at 37°C, and the reactions were started by the addition of substrate. For the RP 73401 hydroxylase and 7-EFC inhibition experiments with anti-CYP2B1 IgG, the incubation time was increased to 20 min. To conserve antibody, inhibition of 7-EFC O-deethylation by anti-CYP2B1 IgG used 0.125 mg microsomal protein. Inhibition of 7-EFC O-deethylation by anti-CYP1A1/2 used a ratio of 40 μl serum per mg of microsomal protein. This concentration has been shown to inhibit 60% of CYP1A2-catalyzed theophylline 3-demethylation (manufacturer’s data, Gentest Corp., Woburn, MA). IgG stock solutions were diluted in phosphate-buffered saline. Antibody, microsomal protein and all other incubation components except the substrate and glucose 6-phosphate were incubated with gentle shaking at room temperature, the samples were then returned to ice, the substrate was added and the incubations were conducted as described above. For the incubations with microsomes prepared from cells expressing individual P450 forms (Gentest), the protein concentration was 0.8 mg/ml and the incubation time was 20 min. RP 73401 microsomal incubations were analyzed with a Spherisorb ODS-2, 250 × 4.6 mm, 5 μm analytical column (Phase Separations, Norwalk, CT) with a ODS-2 5-cm guard column. The column was eluted at 1 ml/min with 55:45% 10 mM ammonium acetate, pH 4.0/acetonitrile, and peak detection was carried out at 268 nm. RPR 113406 and RP 73401 eluted at 3.8 and 15.3 min, respectively.
Statistical analysis.
Marker P450 enzyme activity was correlated with RP 73401 hydroxylase activity in the corresponding liver microsome samples by the linear regression program of GraphPad. An F test was used to generate the P values for the correlation coefficients (r). A significant correlation was defined as having an r ≥ 0.70.
Results
RP 73401 metabolism in vivo.
Human plasma extracts from subjects treated with an 800-μg oral dose of [14C]RP 73401 were analyzed by LC/MS/MS. The fragmentation pattern observed for the cold RPR 113406 standard (fig.2A) was then compared with the spectrum obtained for a major metabolite in human plasma (fig. 2B). The plasma analyte gave a molecular ion of 397 compared with 381 for the parent drug, and a fragmentation pattern consistent with 3-hydroxylation of the cyclopentyl ring and the metabolite standard (RPR 113406). However, LC/MS/MS cannot distinguish the trans isomer (RPR 113406) from the cis isomer (RPR 112919). The pyridyl N-oxide metabolite was also identified in plasma by fragments ofm/z 151, 219 and 397, which matched those of the metabolite standard (RPR 100510, data not shown). This metabolite was produced only at very low levels from incubations of human liver microsomes with RP 73401 (fig. 3).
RP 73401 in vitro metabolite profile.
To generate a chromatographic profile of RP 73401 metabolites formed in vitro, human liver microsomes and microsomes prepared from cells expressing CYP2B6 were incubated with RP 73401 and analyzed by a gradient HPLC method. Figure 3 shows superimposed chromatograms from the incubation of RP 73401 with human liver microsomes (solid line) and a duplicate sample spiked with RPR 112919, RPR 113406 and RPR 100510 (dotted line). The predominant metabolite formed by human liver microsomes was RPR 113406, with the cis isomer (RPR 112919) constituting less than 3% of the amount of total hydroxycyclopentyl metabolite formed. For incubations with microsomes from cells expressing only CYP2B6, RPR 113406 was the only metabolite detected (data not shown). The involvement of FMO in RP 73401 N-oxidation was investigated by exploiting the documented instability of FMO after heating or in the absence of NADPH (Ring et al., 1996;Cashman et al., 1993). Despite the preincubation of human liver microsomes for 1 min at 50°C in the absence of NADPH, rates of RP 73401 N-oxidation were not inhibited significantly (data not shown).
Kinetics of RP 73401 hydroxylation in vitro.
After preliminary studies to determine the linearity of RPR 113406 formation with respect to the amount of microsomal protein and incubation time, the kinetic values of RP 73401 hydroxylation were determined for three human liver microsome samples. The Km andV max values for RPR 113406 formation ranged from 8 to 26 μM RP 73401 and from 0.83 to 5.99 nmol RPR 113406 formed/min/mg protein, respectively (table1). All samples produced typical Michaelis-Menten kinetics for a one-enzyme system.
Correlation and inhibition studies.
The rate of RPR 113406 formation was then determined in a bank of 15 characterized human liver microsome samples. Enzyme activity ranged from 0.25 to 5.85 nmol RPR 113406 formed/min/mg protein (fig. 4A) and correlated strongly with CYP2B6-marker 7-EFC O-deethylase (r 2 = 0.82, P < .01)(fig. 4B, table2). Because of preliminary reports from other laboratories questioning the reliability of 7-EFC O-deethylase activity due to interference from CYP1A2 (Madan et al., 1996), this enzyme activity was reassayed in the presence of a concentration of anti-CYP1A1/2 shown to inhibit approximately 60% of marker CYP1A2 enzyme activity. However, the resulting correlation coefficient (r 2 = 0.82) was unchanged in comparison with the correlation obtained in the absence of anti-CYP1A1/2 antibody (data not shown). Finally, RP 73401 hydroxylation was found to correlate with CYP2A6-catalyzed coumarin hydroxylase activity (r 2 = 0.85, P < .01) (Heyn et al., 1996). Because of the strong correlation of CYP2B6 and CYP2A6 marker activities (r 2= 0 .77, P < .01), other in vitro approaches were necessary to determine the relative involvement of CYP2A6 and CYP2B6 in RPR 113406 formation.
The identification of selective chemical inhibitors of human P450 forms has provided a useful tool in defining the role of individual cytochrome P450s in metabolic pathways (Newton et al., 1995). RP 73401 was therefore incubated in the presence and absence of P450 form-selective inhibitors to determine the effect on RPR 113406 formation by human liver microsomes. As shown in figure5, RP 73401 hydroxylation was inhibited by the CYP2B6 inhibitor orphenadrine (63%) and the CYP2B6-marker substrate 7-EFC (20%). In addition, a Ki of 14.8 μM was determined for the inhibition of RP 73401 hydroxylase activity by 7-EFC (data not shown). None of the other compounds produced more than 6% inhibition at concentrations previously shown to produce significant inhibition of marker P450 activities (Newton et al., 1995). Also, the CYP2A6 substrate coumarin clearly did not act as a competitive inhibitor of RP 73401 hydroxylase activity.
Previous studies have shown that a rabbit polyclonal antibody against CYP2B1 produced 62% inhibition of human liver microsomal 7-EFC O-deethylase activity at a concentration of 5 mg IgG/mg microsomal protein (Heyn et al., 1996). In a similar experiment, 5 mg anti-CYP2B1 IgG/mg microsomal protein produced 85% inhibition of human liver microsomal RP 73401 hydroxylase activity (fig.6). In contrast, 5 mg anti-CYP2B1/mg protein produced only 10% inhibition of CYP2A6-catalyzed coumarin 7-hydroxylase activity, which suggests little or no involvement of CYP2A6 in RP 73401 hydroxylation (data not shown).
RP 73401 metabolism by expressed P450 forms.
Microsomal fractions prepared from cells transformed with individual cDNAs for P450 forms 1A1, 1A2, 2A6, 2B6, 2D6, 2E1, 2C8, 2C9, 2C19 and 3A4 were assayed for RP 73401 hydroxylase activity. Expressed CYP2B6 was found to hydroxylate RP 73401 at a rate of 0.31 nmol RPR 113406 formed/min/mg protein (3.6 nmol/min/nmol CYP2B6); however, none of the remaining nine P450 forms metabolized RP 73401 above background levels (0.01 nmol/min/mg). Kinetic experiments with microsomes prepared from cells expressing CYP2B6 produced a Km of 22.5 μM, consistent with the human liver microsome kinetic studies. These results support the aforementioned data showing that RP 73401 is a CYP2B6-specific substrate.
Discussion
This report documents the cyclopentyl hydroxylation of RP 73401 by human liver CYP2B6 in vitro. In addition, we have reported that this reaction occurs in vivo, as evidenced by the mass fragmentation pattern obtained from plasma samples of volunteers treated with RP 73401. Based on co-chromatography with the metabolite standards, the metabolite profiles obtained from incubations with human liver microsomes and expressed CYP2B6 showed that the hydroxylation of RP 73401 was stereoselective for the trans isomer, RPR 113406. Despite efforts to alter the human liver microsome incubation conditions to favor FMO enzyme activity and thus potentially increase the rate of formation of the N-oxide metabolite (RPR 10510), cyclopentyl hydroxylation was essentially the only metabolic pathway observed in all human liver samples. In fact, the intrinsic clearance (V max/Km ) of 229 μl/min/mg for sample RPR-HL-12 illustrates the high metabolic rate possible in human liver microsomes. The high efficiency of CYP2B6 for catalyzing RP 73401 hydroxylation may explain why CYP2B6 enzyme activity was detected in all human liver microsome samples (n = 15), whereas immunodetectable levels of CYP2B6 were found in only 24% of the samples studied by Mimura et al. (1993).
Three established approaches were used to identify the human P450 form(s) responsible for RP 73401 hydroxylation: 1) correlation of RP 73401 hydroxylase activity with marker P450 enzyme activities in a bank of human liver microsomes; 2) inhibition of enzyme activity by P450-selective inhibitors and antibodies; and 3) measurement of RP 73401 hydroxylation by expressed P450 forms. However, the scarcity of published data on CYP2B6 characterization required some additional studies and warrants some discussion. First, the correlation analysis produced a significant coefficient for both CYP2A6-catalyzed coumarin 7-hydroxylation and CYP2B6-catalyzed 7-EFC O-deethylation. However, the fact that coumarin did not inhibit RP 73401 hydroxylation and that expressed CYP2A6 did not metabolize RP 73401 refutes the involvement of CYP2A6 in RP 73401 hydroxylation. Based on previous data from our laboratory (Heyn et al., 1996) and others (Forresteret al., 1992), the potential co-regulation of CYPs 2A6 and 2B6 is a reasonable explanation for the apparent inconsistency of the correlation analysis with other in vitro approaches. Also, a recent abstract has questioned the reliability of 7-EFC O-deethylation as a marker of CYP2B6 enzyme activity because of interference from CYP1A2 (Madan et al., 1996). Therefore, we repeated our measurement of 7-EFC O-deethylase activity for 15 human liver microsome samples in the presence of anti-CYP1A1/2 antibody to eliminate any potential 1A2 involvement. The resultant correlation with RP 73401 hydroxylase activity in the same bank of microsome samples was unchanged. Because detailed comparative kinetic information on 7-EFC O-deethylation by CYP1A2 and CYP2B6 was not available from Madanet al. (1996), it is possible that, at a substrate concentration of 10 μM, 7-EFC O-deethylase activity accurately measured CYP2B6 enzyme activity in our study.
An examination of the specificity of several chemicals to inhibit multiple human P450s has been described (Newton et al., 1995); however, CYP2B6 enzyme activity was not monitored. Our laboratory recently evaluated the specificity of orphenadrine as a CYP2B6 inhibitor and found that it was generally nonselective (Guoet al., 1997). Therefore, although the extent of inhibition of human liver microsomal RP 73401 hydroxylation by orphenadrine is similar to that observed for (S)-mephenytoin N-demethylation and 7-EFC O-deethylation, chemical inhibition studies should be viewed as supportive rather than definitive evidence of CYP2B6 involvement in RP 73401 hydroxylation. Inhibitors of CYP2A6, -2D6, -2E1, -2C9 and -3A4 clearly had no effect on RP 73401 hydroxylation.
The demonstration of RP 73401 hydroxylase activity by expressed CYP2B6 and the close agreement in affinity (Km ) between human liver microsomes and the expressed form provide the most direct evidence for the metabolism of RP 73401 by CYP2B6. In addition, because the kinetics of RP 73401 hydroxylation fitted a one-enzyme model and RPR 113406 was not formed by any of the other nine expressed P450 forms tested, we conclude that CYP2B6 is the sole catalyst of RP 73401 hydroxylation by human liver microsomes. Future studies will investigate the use of this unique enzyme activity to study species differences in CYP2B-catalyzed reactions and the interindividual variability and inducibility of CYP2B6.
Acknowledgments
The authors acknowledge the help of Paul Cox and Andrew Ratcliffe for the synthesis and characterization of the RP 73401 metabolite standards and the experimental assistance of Dr. Heleen Heyn.
Footnotes
-
Send reprint requests to: Jeffrey C. Stevens, Ph.D., Department of Drug Metabolism and Pharmacokinetics, Rhône-Poulenc Rorer, Mail Stop NW12, 500 Arcola Road, Collegeville, PA 19426.
- Abbreviations:
- PDE
- phosphodiesterase
- 7-EFC
- 7-ethoxytrifluoromethylcoumarin
- 7-HFC
- 7-hydroxytrifluoromethyl coumarin
- G6PDH
- glucose-6-phosphate dehydrogenase
- TAO
- troleandomycin
- LC
- liquid chromatography
- MS
- mass spectrometry
- HPLC
- high-performance liquid chromatography
- HEPES
- 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid
- FMO
- flavin monooxygenase
- Received January 6, 1997.
- Accepted May 19, 1997.
- The American Society for Pharmacology and Experimental Therapeutics