Introduction

Top Abstract Introduction Materials & Methods Results Discussion References

Tazofelone¹ has been shown to be highly effective in rat and rabbit models of inflammatory bowel disease and colitis. Previous in vivo disposition studies have indicated that the absorption of tazofelone in rats and dogs is approximately 20%. However, appreciable in vivo biotransformation to primarily a stereoisomeric mixture of sulfoxides results in a tazofelone bioavailability of only 2-3% in both species. Tazofelone is rapidly cleared from the plasma of rats and dogs with half-lives of 2 to 4 hr, respectively. Subsequent biotransformation studies with rat and dog hepatic microsomes have demonstrated that tazofelone is rapidly oxidized to the same stereoisomeric mixture of sulfoxides as observed in vivo (unpublished results).

Cytochrome P450 is a superfamily of heme containing enzymes that play a major role in oxidative metabolism of drugs, endogenous compounds, and xenobiotics (1). These enzymes exhibit broad and often overlapping substrate specificities (2). Moreover, they also show interspecies variation in their expression and catalytic activity (3). Therefore, identification of the cytochrome P450 isozyme responsible for the metabolism of a particular compound has many useful implications in the drug development process, namely, a) to predict drug-drug interactions (4), b) to predict variation in drug metabolism caused by genetic polymorphism of specific enzymes (3, 5), and c) to design new drugs with reduced metabolism and improved bioavailability (6). Recently, the availability of in vitro techniques to elucidate the metabolic profiles of drugs and identify the isozyme(s) responsible for a particular metabolite has improved the drug development process. Furthermore, the use of human liver tissue for these studies avoids the extrapolation of data across species and improves planning and interpretation of drug safety studies. The objective of the present study was to investigate the biotransformation of tazofelone by human liver microsomes and to identify the cytochrome P450 isozyme(s) responsible for the formation of the metabolites.

	Materials and Methods

Top Abstract Introduction Materials & Methods Results Discussion References

Chemicals. (R,S)-Tazofelone (99.4% pure) and (R,S)-¹⁴C-tazofelone (26.1 µCi/mg, >99% radiochemical purity) were synthesized at the Lilly Research Laboratories (Indianapolis, IN). Furafylline was obtained from Ultra-Fine Chemicals (Manchester, UK), triacetyloleandomycin was a gift from Pfizer (Groton, CT), quinidine sulfate, glucose-6-phosphate dehydrogenase, glucose-6-phosphate, and NADPH were purchased from Sigma Chemical Co. (St. Louis, MO). S-Mephenytoin was a gift from Dr. Grant Wilkinson, Vanderbilt University. Diethyldithiocarbamate and coumarin were purchased from Aldrich Chemical Co. (Milwaukee, WI).

Cytochrome P450. Human liver samples designated HL-A through HL-S were obtained from the liver transplant unit of the Medical College of Wisconsin under the protocol approved by the committee for the conduct of Human Research. Microsomes were prepared from these livers (7) and were characterized for their relative levels of cytochrome P450 and FMO3 by immunoquantification and for their form-selective catalytic activities (1, 8, 9). Microsomes prepared from human B-lymphoblastoid cell line AHH-1 expressing human P450 cDNAs (CYP1A1, CYP1A2, CYP2A6, CYP3A4, CYP2B6, CYP2C9, CYP2C19, CYP2D6, and CYP2E1) were purchased from Gentest Corp. (Woburn, MA) and used according to the supplier's instructions.

In Vitro Incubations. Preliminary in vitro microsomal incubations with tazofelone contained three major metabolite peaks in the HPLC chromatograms. Two of these peaks corresponded to two sulfoxide diastereoisomeric mixtures that were previously identified in vivo (unpublished results). To identify the third peak, incubations were carried out with 1 mg/ml human hepatic microsomal protein in pH 7.4 phosphate buffer, 50 µM tazofelone, 1 mM NADPH, and 0.5 U glucose-6-phosphate dehydrogenase, 1 mM glucose-6-phosphate, and 5 mM magnesium chloride in 35 ml for 1 hr. Incubations were stopped by the addition of an equal volume of acetonitrile and the precipitated protein was removed after centrifugation. The supernatant was evaporated to dryness under nitrogen and the residue was dissolved in acetonitrile:water (40:60) and subjected to HPLC analysis.

HPLC Analysis of Tazofelone Metabolites. Tazofelone, tazofelone sulfoxides, and the quinol metabolite were resolved on a Prodigy C₁₈ 250 × 4.6 mm column (Phenomenex). The HPLC system consisted of two Shimazdu LC-10AD pumps, a Kratos Spectraflow 783 UV variable detector, and a Waters Model 710B WISP autoinjector. An isocratic mobile phase composed of acetonitrile:water (60:40) was used at a flow rate of 1 ml/min. The column temperature was maintained at 40°C, and the eluate was monitored at 214 nm by UV detection. The fractions corresponding to the metabolite peaks were collected into scintillation vials with a Foxy fraction collector (Isco, Lincoln, NE), and UltimaGold scintillation liquid was added to the fractions and assayed for radioactivity by liquid scintillation counting (Beckmann LS7500). Quench correction was achieved by external standardization.

Identification of Quinol Metabolite. Purification of the quinol metabolite was achieved by a two-step procedure. The first involved the isocratic HPLC method described above. The eluate corresponding to the peak was repeatedly collected into a single vial and evaporated to dryness under nitrogen at 40°C and the residue dissolved in acetonitrile:water (20:80). The second step consisted of a linear gradient HPLC from 20:80 to 80:20 acetonitrile:water over 40 min. The peak corresponding to the quinol metabolite was collected and evaporated to dryness and subjected to spectral analysis.

Spectral Analysis. Proton nuclear magnetic resonance spectroscopy (¹H-NMR), HMQC, and homonuclear decoupling were performed on a Bruker 500 MHz FT-NMR using deuterated acetonitrile as solvent. A field desorption mass spectrum (FD-MS) was generated using a VG Model 70SE magnetic sector mass spectrometer and fast atom bombardment mass spectra (FAB-MS) and high resolution FAB mass spectra (HR-MS) were obtained on a VG ZAB-2SE mass spectrometer.

Kinetic and Statistical Analysis. Kinetic parameters for the formation of tazofelone metabolites were determined using k·cat enzyme kinetic software (BioMetallics, Inc., Princeton, NJ) and Eadie-Hofstee plots were visually inspected to assess whether one or two enzymes were involved. Estimates of the kinetic parameters from the k·cat program were used as initial estimates for nonlinear regression analysis using NONLIN (Statistical Consultants, Inc., Lexington, KY) software. For the formation of the total sulfoxides, the following one enzyme model for a saturated enzyme system (S> K_m) was used.

(1)

where Cl_int = V_max/K_m. To characterize the kinetics of the quinol metabolite formation, a two enzyme model was used. Under the experimental conditions, eq. 2 described the kinetics best with a high affinity site that was saturated and a low affinity site that was not saturated.

(2)

Statistical analysis was performed using JMP software (SAS Institute, Inc., Cary, NC).

	Results

Top Abstract Introduction Materials & Methods Results Discussion References

Identification of Metabolites. Incubation of tazofelone with human hepatic microsomes in the presence of NADPH and oxygen resulted in three major peaks (fig. 1). Two of these peaks were observed in plasma of rats and dogs and have been identified as a pair of diastereomeric sulfoxides which co-elute with authentic sulfoxide standards (unpublished results). Because of the epimerization of the tazofelone sulfoxide stereoisomers at pH values greater than 4, tazofelone sulfoxide was quantitated as the sum of the two HPLC peaks for all analysis. To elucidate the structure of the quinol metabolite, mass spectral and proton NMR studies were conducted on the material isolated and purified by HPLC from human hepatic microsomal incubations. The FD-MS contained a molecular ion peak (m/z = 337), indicating the addition of oxygen to tazofelone (MW = 321) (fig. 2). This was confirmed by FAB-MS which produced a molecular ion at m/z = 338 (MH⁺) (fig. 2). The FAB-MS also produced ions at m/z = 360, m/z = 320, m/z = 264, and m/z = 219 corresponding to the fragments [M+Na]⁺, [M-H₂O]⁺, [M-t-butyl]⁺, and [BHT tropylium cation], respectively (fig. 2). Also, the high resolution mass spectrum (HR-MS) indicated a molecular formula of C₁₈H₂₇NO₃S consistent with the proposed structure. The ¹H-NMR, HMQC, and homonuclear decoupling experiments indicated that the difference between tazofelone and the quinol metabolite was in the chemical shifts of the 5,6, 8/12, and phenol protons (table 1). The spin system corresponding to the five membered ring was present in both tazofelone and the metabolite which suggests that addition of oxygen was in the aromatic system. The six methyls of the t-butyl groups all have the same shift, and there was no new -CH₂-OH in the NMR spectrum indicating that the incorporation of oxygen was not on the t-butyl substituents. The mass spectral peaks M-57 and M-114 also confirmed the lack of substitution on t-butyl groups. From the variable temperature NMR data, decoupling experiments, and comparison of data with the parent molecule, the proposed structure of the quinol metabolite is shown in fig. 3.

View larger version (17K): [in this window] [in a new window]   Fig. 1.   HPLC chromatogram of tazofelone incubated with human hepatic microsomes.

View larger version (15K): [in this window] [in a new window]   Fig. 2.   FD-MS and FAB-MS of the purified quinol metabolite of tazofelone isolated from human hepatic microsomes.

View larger version (15K): [in this window] [in a new window]   Fig. 3.   Metabolic pathway of tazofelone in human microsomes.

Kinetics of Formation. Kinetic analysis of total sulfoxide and quinol metabolite formation was conducted in human liver microsomes at concentrations ranging from 3.9 µM to 1500 µM. Nonlinear regression modeling was done on the data to fit Michaelis-Menten kinetics. Eadie-Hofstee plots (not shown) indicated that sulfoxidation was catalyzed by a single enzyme while formation of the quinol metabolite was catalyzed by a two enzyme system. Kinetic parameters for sulfoxide and quinol metabolite formation are presented in table 2. As shown in the table, the apparent K_m and V_max values for sulfoxide formation were 12.4 µM and 0.27 nmol/min/mg protein, respectively, which resulted in an intrinsic clearance of 0.022 nmol/min/mg protein. The kinetics of quinol metabolite formation followed a two enzyme model. The apparent K_m and V_max values for the high affinity enzyme were 7.5 µM and 0.17 nmol/min/mg protein, respectively. The intrinsic clearance by the high affinity enzyme was 0.023 ml/min/mg protein and by the low affinity enzyme was 0.002 ml/min/mg protein.

Identification of Cytochrome P450 Isozyme. The rates of formation of total sulfoxide and quinol metabolites were determined using microsomes from a bank of 18 human livers. These 18 human livers have been previously characterized for their isozyme-selective catalytic activities, and the immunoquantifiable levels of CYP1A2, CYP2A6, CYP2C8/2C9, CYP2D6, CYP2E1, and CYP3A were characterized for 14 livers (3, 8, 9, 17, 18). The history of livers, the range of relative isozyme-specific catalytic activities, and the range of relative levels of specific cytochrome P450 isozymes in the bank of microsomes prepared from these livers have been published (18). The range of activities for the formation of sulfoxide and quinol metabolites followed the general trend noticed with the range of activities for the formation of midazolam 1-hydroxylation. The medical history indicates exposure to phenobarbital in liver samples HL-E, HL-I, and HL-O. The immunoquantified CYP3A concentrations and the CYP3A isozyme-specific midazolam 1-hydroxylation were the highest in these livers, and correspondingly high formation rates were also noticed for sulfoxide and quinol metabolites (table 3). Of the human liver samples tested, no correlation was observed with sex, age, tobacco, or alcohol consumption. The rates of formation of total sulfoxide and quinol metabolite were correlated with isozyme-selective catalytic activities and immunoquantitated concentrations of cytochrome P450 (table 3). As shown in table 4, the rate of formation of total sulfoxide correlated significantly with the rate of midazolam 1-hydroxylation (r² = 0.80) and with the immunoquantifiable concentration of CYP3A (r² = 0.85). Furthermore, the rate of formation of the quinol metabolite was also significantly correlated with the immunoquantifiable concentrations of CYP3A (r² = 0.93) and with the rate of midazolam 1-hydroxylation (r² = 0.95).

View larger version (32K): [in this window] [in a new window]   Fig. 4.   Effect of isozyme-specific inhibitors on the formation of the total sulfoxide (A) and the quinol (B) metabolites of tazofelone. Values represent the mean of duplicate measurements determined as described in Materials and Methods using a 20 µM tazofelone concentration and expressed as a percentage of duplicate control measurements. The values in parenthesis indicate the µM concentrations of inhibitors used in the experiment.

View larger version (16K): [in this window] [in a new window]   Fig. 5.   Formation of the total sulfoxide (A) and quinol (B) metabolites of tazofelone by cDNA expressed isozymes present in microsomes from human B lymphoblastoid cell line AHH-1. Values represent the mean of duplicate rates of formation measurements for total sulfoxide (A) and quinol (B) determined as described in Materials and Methods using 20 µM tazofelone. Controls are microsomes prepared from the B lymphoblastoid cell line expressing no detectable cytochrome P450 concentrations.

	Discussion

Top Abstract Introduction Materials & Methods Results Discussion References

In vivo biotransformation studies have demonstrated that tazofelone undergoes oxidation forming a stereoisomeric mixture of sulfoxides (unpublished results). The present in vitro study with human liver microsomes has demonstrated that tazofelone undergoes biotransformation to a mixture of stereoisomeric sulfoxides and a hydroxy metabolite. The hydroxy metabolite of tazofelone has been isolated and shown by mass spectral and proton NMR analysis to be hydroxylated at the para-position of the t-butyl phenol moiety to the quinol form of tazofelone. The biotransformation of the butylated hydroxytoluene (BHT) moiety of tazofelone to the quinol form is consistent with existing literature. Thompson et al. have shown that BHT undergoes cytochrome P450 mediated oxidation forming a quinol along with nine other metabolites in rat liver microsomes (20). Recently, Ohe et al. have shown that when p-cresol was incubated with rat liver microsomes, a p-toluquinol metabolite was formed (21). As with the quinol metabolite of BHT, the quinol metabolite of tazofelone has not been identified in vivo.

Kinetic analysis of the formation of the metabolites in human liver microsomes indicates that tazofelone sulfoxidation is catalyzed by a single enzyme system while tazofelone hydroxylation to the quinol metabolite is catalyzed by a two-enzyme system. The mean peak plasma concentrations of tazofelone achieved in rats after oral administration of a single dose of 30 mg/kg was about 1 µM (unpublished results). It is reasonable to assume that only the high affinity enzyme plays a major role in the metabolism of tazofelone. Hence, a 20 µM concentration of tazofelone was used to identify the isozyme(s) responsible for the metabolism of tazofelone. Identification of the specific isozyme involved in the formation of the sulfoxide and quinol metabolites was achieved by using correlation analysis, isozyme specific inhibitors, and cDNA expressed P450 isozymes. The correlation analysis was accomplished by comparing the rates of sulfoxidation and hydroxylation of tazofelone with the isozyme-specific catalytic activity and immunoquantified isozyme-specific P450 concentrations. A significant correlation (p<0.01) was seen between the formation of both the sulfoxide and quinol metabolites of tazofelone and immunoquantified concentrations of CYP3A. A similar correlation (p<0.01) was also noted between the formation of both these metabolites and midazolam 1-hydroxylation, a CYP3A mediated activity. The correlation analysis strongly suggests the involvement of CYP3A in the metabolism of tazofelone in vitro, and this evidence is supported by the inhibition studies that have been carried out using selective isozyme inhibitors. Triacetyloleandomycin almost completely inhibited the formation of the sulfoxide and quinol metabolites at both 10 and 50 µM concentrations. Since TAO is a known mechanism-based inhibitor of CYP3A mediated metabolism, and TAO significantly inhibited the formation of the sulfoxide and quinol metabolites, this corroborates the earlier evidence obtained in the correlation analysis. Further evidence was obtained using commercially available cDNA expressed cytochrome P450 isozymes. All the cDNA expressed enzymes formed sulfoxide and quinol to some extent, but CYP3A4 formed the largest amount of both metabolites. This clearly supports the earlier evidence that CYP3A is involved in the metabolism of tazofelone.

Flavin monooxygenases are known to mediate the sulfoxidation of drugs and other xenobiotics (19). The human liver flavin-containing monooxygenase (FMO3) is known to mediate the S-oxidation of drugs such as cimetidine to their S-oxide metabolites (22, 23). Data presented herein strongly indicate that sulfoxidation is a cytochrome P450 mediated process and that FMO3 may not have any significant role. Moreover, the lack of correlation between the rate of formation of total sulfoxides and immunoquantified concentrations of FMO3 supports the hypothesis that tazofelone metabolism is a cytochrome P450 mediated process.

In summary, a monohydroxylated quinol metabolite of tazofelone has been identified and characterized in vitro. It was demonstrated that tazofelone undergoes oxidative metabolism forming stereoisomeric sulfoxides and a quinol metabolite. The correlation analysis, inhibition studies, and cDNA expressed isozyme specific microsomal studies implicate CYP3A4 as the isozyme responsible for the formation of both tazofelone sulfoxide and quinol metabolites.