Volume 50, Issue 6 p. 573-580
Free Access

The xenobiotic inhibitor profile of cytochrome P4502C8

Chin-Eng Ong

Chin-Eng Ong

Department of Clinical Pharmacology, Flinders Medical Centre and Flinders University, Adelaide, Australia

Search for more papers by this author
Sally Coulter

Sally Coulter

Department of Clinical Pharmacology, Flinders Medical Centre and Flinders University, Adelaide, Australia

Search for more papers by this author
Donald J. Birkett

Donald J. Birkett

Department of Clinical Pharmacology, Flinders Medical Centre and Flinders University, Adelaide, Australia

Search for more papers by this author
C. Ramana Bhasker

C. Ramana Bhasker

Department of Clinical Pharmacology, Flinders Medical Centre and Flinders University, Adelaide, Australia

Search for more papers by this author
John O. Miners

John O. Miners

Department of Clinical Pharmacology, Flinders Medical Centre and Flinders University, Adelaide, Australia

Search for more papers by this author
First published: 09 October 2008
Citations: 73
Professor John Miners, Department of Clinical Pharmacology, Flinders Medical Centre, Bedford Park, SA 5042, Australia. Tel.: 61–8-82044131; Fax: 61 8-82045114; E-mail: [email protected]

Current address: International Medical University, Kuala Lumpur, Malaysia.

Abstract

Aims  To investigate inhibition of recombinant CYP2C8 by: (i) prototypic CYP isoform selective inhibitors (ii) imidazole/triazole antifungal agents (known inhibitors of CYP), and (iii) certain CYP3A substrates (given the apparent overlapping substrate specificity of CYP2C8 and CYP3A).

Methods  CYP2C8 and NADPH-cytochrome P450 oxidoreductase were coexpressed in Spodoptera frugiperda (Sf21) cells using the baculovirus expression system. CYP isoform selective inhibitors, imidazole/triazole antifungal agents and CYP3A substrates were screened for their inhibitory effects on CYP2C8-catalysed torsemide tolylmethylhydroxylation and, where appropriate, the kinetics of inhibition were characterized. The conversion of torsemide to its tolylmethylhydroxy metabolite was measured using an h.p.l.c. procedure.

Results  At concentrations of the CYP inhibitor ‘probes’ employed for isoform selectivity, only diethyldithiocarbamate and ketoconazole inhibited CYP2C8 by > 10%. Ketoconazole, at an added concentration of 10 µm, inhibited CYP2C8 by 89%. Another imidazole, clotrimazole, also potently inhibited CYP2C8. Ketoconazole and clotrimazole were both noncompetitive inhibitors of CYP2C8 with apparent Ki values of 2.5 µm. The CYP3A substrates amitriptyline, quinine, terfenadine and triazolam caused near complete inhibition (82–91% of control activity) of CYP2C8 at concentrations five-fold higher than the known CYP3A Km. Kinetic studies with selected CYP3A substrates demonstrated that most inhibited CYP2C8 noncompetitively. Apparent Ki values for midazolam, quinine, terfenadine and triazolam ranged from 5 to 25 µm.

Conclusions   Inhibition of CYP2C8 occurred at concentrations of ketoconazole and diethyldithiocarbamate normally employed for selective inhibition of CYP3A and CYP2E1, respectively. Some CYP3A substrates have the capacity to inhibit CYP2C8 activity and this may have implications for inhibitory drug interactions in vivo.

Introduction

The human cytochrome P450 2C subfamily comprises four members; CYP 2C8, 2C9, 2C18 and 2C19. Of these, CYP2C9 is known to catalyse the metabolism of numerous therapeutic agents (e.g. glipizide, losartan, phenytoin, tolbutamide, torsemide, S-warfarin, and many nonsteroidal anti-inflammatory drugs), and the importance of this enzyme in human drug elimination is now well established [1]. Similarly, CYP2C19 has the capacity to metabolize numerous drugs, including antidepressants, proton pump inhibitors, benzodiazepines, S-mephenytoin, and proguanil [2–4]. No selective substrates of CYP2C18 have been identified to date, and it was generally also thought that CYP2C8 had a limited role in drug and chemical metabolism.

However, recent evidence suggests that the contribution of CYP2C8 to drug elimination in humans has been underestimated. This enzyme appears to be primarily responsible for the metabolism of cerivastatin, paclitaxel, rosiglitazone and troglitazone [5–8]. Moreover, there appears to be a degree of overlapping substrate specificity between CYP2C8 and CYP3A4. For example, CYP2C8 contributes in part to the metabolism of the predominantly CYP3A4 substrates carbamazepine, verapamil and zopiclone, and, vice versa, CYP3A4 contributes in part to cervistatin and paclitaxel metabolism [6, 9–12]. CYP2C8 has also been implicated in the oxidation of retinoids and fatty acids, including all-trans-retinoic acid and arachidonic acid [13, 14].

Wide interindividual variability in clearance is a characteristic of drugs metabolized by CYP2C9 and CYP2C19, and genetic polymorphisms are associated with the impaired metabolism of substrates for both enzymes [1, 2]. Drug interactions are another important determinant of the altered clearance of CYP2C9 substrates [1], and the situation is presumably similar for other CYP2C isoforms. In vivo variability data are lacking for CYP2C8, but rates of CYP2C8 catalysed paclitaxel 6α-hydroxylation and rosiglitazone N-demethylation and p-hydroxylation have been reported to differ up to 38-fold in microsomes from panels of human livers [5, 11]. Polymorphism in the coding region of CYP2C8 has been reported recently [15], but the relative contribution of genetic and other factors to the variability in CYP2C8 activity remains unknown.

Despite increasing awareness of the apparent importance of CYP2C8 in the metabolism of xenobiotics and endogenous compounds, there have been no systematic studies of the inhibition profile of this enzyme. In particular, the effects of the prototypic CYP isoform-selective inhibitors, used widely to determine the contribution of individual isoforms to a metabolic pathway in human liver microsomes in reaction phenotyping [16], on CYP2C8 activity are incompletely characterized. Similarly, the potential for other drugs to inhibit CYP2C8-catalysed reactions has received little attention, and hence there is a poor understanding of potential inhibitory drug interactions involving CYP2C8. Here we describe studies which investigated inhibition of recombinant CYP2C8 by: (i) CYP isoform-selective inhibitors (ii) imidazole/triazole antifungal agents, and (iii) a number of CYP3A substrates. The imidazole/triazole antifungals were investigated because of their propensity to inhibit CYP-catalysed xenobiotic biotransformation, while CYP3A substrates were selected due to the apparently overlapping substrate specificity of this enzyme and CYP2C8.

Methods

Chemicals and reagents

Budesonide, coumarin (COUM), cyclosporin A, diethyldithiocarbamate (DDC), diethylstilbestrol (DES), diltiazem, glucose 6-phosphate, glucose 6-phosphate dehydrogenase, lignocaine, 4-methylumbelliferone (4 mU), midazolam, β-nicotinamide adenine dinucleotide phosphate (NADP), reduced β-nicotinamide adenine dinucleotide (NADH), paclitaxel, quinidine sulphate (QUIN), quinine sulphate, terfenadine, triazolam, and troleandomycin (TAO) were purchased from the Sigma Chemical Co (St Louis, MO, USA) and 6α-hydroxy-paclitaxel was purchased from the Gentest Corp (Woburn, MA, USA). Other chemicals were kindly donated by the following sources: bifonazole (BIF) and clotrimazole (CLO), Bayer Australia (Sydney, Australia); diazepam, Roche Products Pty Ltd (Sydney, Australia); econazole nitrate (ECO), Bristol Myers Squibb Pharmaceuticals (Melbourne, Australia); fluconazole (FLU), Pfizer Ltd (Sydney, Australia); furafylline (FUR), Dr R Gasser, Hoffman La Roche (Basel, Switzerland); itraconazole (ITRA), ketoconazole (KET) and miconazole nitrate (MIC), Janssen-Cilag Pty (Sydney, Australia); mephenytoin (MEPH), Sandoz Ltd (Basel, Switzerland); sulphaphenazole (SPZ) Ciba-Geigy Australia (Sydney, Australia); torsemide and tolyl methylhydroxytorsemide Boehringer Mannheim International (Mannheim, Germany). Reagents for the molecular biological procedures and expression of CYP2C8 in Sf21 cells were as described by Ong et al.[17]. All other chemicals and reagents were of analytical reagent grade.

CYP2C8 expression and human liver microsomes

CYP2C8 and NADPH-cytochrome P450 oxidoreductase (OxR) were coexpressed in Spodoptera frugiperda (Sf21) cells using the baculovirus expression system, as described previously [17]. The baculovirus dual expression plasmid pAcUW31 was used to insert CYP2C8 and OxR cDNAs downstream of the polyhedrin and p10 promoters, respectively. Microsomes derived from Sf21 cells infected with selected dual gene clones were pooled for the kinetic studies described here. The CYP spectral content and OxR activity of microsomes were 79 pmol CYP mg−1 and 600 nmol cytochrome c reduced min−1 mg−1, respectively.

Microsomes from four human livers (from the Department of Clinical Pharmacology of Flinders Medical Centre liver ‘bank’) were used for the characterization of paclitaxel 6α-hydroxylation (see below). Approval of the Clinical Investigation Committee of Flinders Medical Centre was obtained for the use of human liver tissue in xenobiotic metabolism studies.

Enzyme assays

Torsemide hydroxylation was determined by the procedure of Miners et al.[18]. Briefly, incubation mixtures, in a total volume of 1 ml, contained Sf21 microsomes (0.3 mg), NADPH generating system (1 m m NADP, 10 m m glucose 6-phosphate, 2 IU glucose 6-phosphate dehydrogenase, 5 m m MgCl2), torsemide (see ‘Kinetic and inhibition experiments’ for concentrations) and phosphate buffer (0.1 m, pH 7.4). Reactions were initiated by the addition of NADPH generating system and carried out at 37 °C for 30 min. Incubations were terminated by the addition of perchloric acid (0.01 ml, 11.6 m) and cooling on ice. After addition of the assay internal standard (4 mU, 4 nmol), methylhydroxytorsemide was extracted from the supernatant fraction (saturated with ammonium sulphate; 1.5 g) with dichloromethane-iso-propanol (85 : 15; 2 × 4 ml). The extract was analysed by h.p.l.c. as described previously [18]. Unknown concentrations of metabolite were determined by comparison of hydroxytorsemide with internal standard (4 mU) peak height ratios with those of a standard curve. Under the conditions employed, rates of torsemide methylhydroxylation were linear with respect to both microsomal protein concentration and incubation time, and assay within-day imprecision was  < 4% at substrate concentrations of 10 and 50 µm.

Paclitaxel 6α-hydroxylation was measured by a modification of the method of Sonnichsen et al.[11]. A standard 1 ml incubation contained human liver microsomal or Sf21 microsomal protein (1 mg), NADPH generating system (as above) and paclitaxel (see ‘Kinetic and inhibition experiments’ for concentrations) in phosphate buffer (0.1 m, pH 7.4). Reactions were initiated by the addition of NADPH generating system and were carried out at 37 °C for 10 min. Rates of 6α-hydroxypaclitaxel formation were shown to be linear for incubation times to 10 min and for microsomal protein concentrations to 1.5 mg ml−1. The reaction was terminated by rapid cooling on ice. After addition of the internal standard (DES, 0.5 µg), the incubation mixture was extracted with diethylether (2 × 4 ml). The combined ether extracts were evaporated under a stream of dry nitrogen. Extraction efficiencies for metabolite and internal standard were 85 ± 4% and 83 ± 2%, respectively. The residue was reconstituted in the mobile phase (0.15 ml) and injected on to the h.p.l.c. The chromatograph (Beckman System Gold Programmable Solvent Module, Beckman Instruments, Fullerton, CA, USA) was fitted with an Ultrasphere ODS column (25 cm × 4.6 mm id, 5 µm particle size; Beckman Instruments) which was eluted with water-methanol (1 : 2) for 14 min following injection, and then with acetonitrile for 1 min before reverting to the original aqueous methanol phase for another 5 min. The mobile phase flow rate was 1.2 ml min−1 and absorbance was monitored at 230 nm. Under these conditions, retention times for DES, 6α-hydroxypaclitaxel and paclitaxel were 8.4, 10.8, and 13 min, respectively. Concentrations of 6α-hydroxy-paclitaxel in incubations were determined by comparison of peak height ratios with those of a standard curve. Overall assay within-day imprecision, assessed by measuring 6α-hydroxypaclitaxel formation in 10 separate incubations of the same batch of human liver microsomes, was 10.0% and 7.0% for substrate concentrations of 2 µm and 10 µm, respectively.

Kinetic and inhibitor experiments

The kinetics of paclitaxel 6α-hydroxylation by CYP2C8 and human liver microsomes were determined using seven or eight substrate concentrations in the range 2–40 µm. Similarly, the kinetics of CYP2C8-catalysed torsemide tolylmethylhydroxylation were determined for eight substrate concentrations in the range 25–400 µm. Prototypic CYP isoform-selective probes, azole/triazole antifungal agents and CYP3A substrates screened for inhibition of CYP2C8 activity, and their concentrations, are given in Figure 2, Figure 3 and Table 3, respectively. Inhibition of CYP2C8-catalysed torsemide hydroxylation by each of the compounds was investigated at a torsemide concentration of 170 µm, which corresponds to the apparent Km for this reaction (see Results). Incubations without substrate were performed for each potential inhibitor to exclude the possibility of assay interference by the compound or its metabolites. Individual compounds were added to incubations as solutions in either water (for salts) or DMSO (0.5% v/v). Control incubations contained an equivalent volume of solvent, although it was shown that DMSO (0.5%, v/v) had a negligible effect on CYP2C8 activity. The mechanism-based inhibitors diethyldithiocarbamate, furafylline and troleandomycin were preincubated with microsomes and NADPH generating system for 10 min prior to the addition of substrate to incubations. Where marked inhibition of CYP2C8-catalysed torsemide hydroxylation was observed, kinetic studies were undertaken to determine the mechanism of inhibition and to calculate the apparent inhibition constant (Ki). Torsemide hydroxylase activity was measured at four or five inhibitor concentrations at each of three substrate concentrations.

Details are in the caption following the image

Inhibition of CYP2C8 catalysed torsemide tolylmethylhydroxylation by prototypic CYP isoform selective probes. The torsemide concentration was 170 µm and concentrations of inhibitors are shown beside the abbreviation for each compound. Inhibition data represent the mean of duplicate measurements. Abbreviations: FUR, furafylline; COUM, coumarin; SPZ, sulphaphenazole; MEPH, mephenytoin; QUIN, quinidine; DDC, diethyldithiocarbamate; TAO, troleandomycin; KET, ketoconazole.

Details are in the caption following the image

Inhibition of CYP2C8 catalysed torsemide tolylmethylhydroxylation by azole and triazole antifungal agents. The torsemide concentration was 170 µm. Each azole/triazole was screened at added concentrations of 10 and 100 µm. Inhibition data represent the mean of duplicate measurements. Abbreviations; FLU, fluconazole; ITRA, itraconazole; BIF, bifonazole; MIC, miconazole; ECO, econazole; CLO, clotrimazole.

Analysis of results

All data points represent the mean of duplicate estimations. Kinetic and inhibition data were model-fitted using MK Model, an extended least squares modelling program.

Results

The kinetics of paclitaxel 6α-hydroxylation by human liver microsomes and Sf21 microsomes were compared in order to validate Sf21-expressed CYP2C8 as a model for the human liver enzyme ( Figure 1a and Figure 1b). Apparent Km and Vmax values for the human liver microsomal reaction (n = 4 livers) ranged from 7 to 19 µm (mean 12.2 µm) and 40–349 pmol min−1mg−1 (mean 142 pmol min−1 mg−1), respectively. The apparent Km for paclitaxel 6α-hydroxylation by Sf21-expressed CYP2C8 was 6 µm, and the Vmax was 234 pmol min−1 mg−1.

Details are in the caption following the image

Eadie-Hofstee plots for paclitaxel 6α-hydroxylation by microsomes from a representative human liver (panel a) and Sf21-expressed CYP2C8 (panel b), and for torsemide tolylmethylhydroxylation by Sf21-expressed CYP2C8 (panel c). Points are experimentally derived values (means of duplicate measurements at each concentration) while the solid lines show computer-derived curves of best fit.

Sf21-expressed CYP2C8 was also shown to convert torsemide to its tolylmethylhydroxy metabolite ( Figure 1c), with apparent Km and Vmax values of 170 µm and 35 pmol min−1 mg−1, respectively. Although torsemide is more efficiently hydroxylated by wild-type CYP2C9 than by CYP2C8, as evidenced by a 12-fold higher intrinsic clearance [19], torsemide is nevertheless a convenient substrate for expressed CYP2C8. Total chromatography time for the hydroxytorsemide assay is shorter than that for the 6α-hydroxypaclitaxel assay, and chromatography associated with the former is more reproducible. Thus, subsequent studies of the inhibition of expressed-CYP2C8 activity measured effects on the conversion of torsemide to hydroxytorsemide.

Prototypic CYP-isoform selective probes were assessed for their inhibitory effects on CYP2C8-catalysed torsemide hydroxylation. Compounds screened included furafylline (CYP1A), coumarin (CYP2A6), sulphaphenazole (CYP2C9), mephenytoin (CYP2C19), quinidine (CYP2D6), diethyldithiocarbamate (CYP2E1), and ketoconazole and troleandomycin (CYP3A). Two concentrations of each inhibitor were investigated ( Figure 2). The lower corresponded to the concentration normally employed to confer isoform selectivity [16] while the higher concentration was 5-to 10-fold greater. Only diethyldithiocarbamate and ketoconazole inhibited CYP2C8 activity by > 10% at the lower ‘selective’ probe concentration ( Figure 2). At the higher concentration, quinidine, diethyldithiocarbamate and ketoconazole caused > 20% inhibition of CYP2C8. Indeed, at an added concentration of 10 µm, ketoconazole almost abolished CYP2C8 activity. Ketoconazole inhibited CYP2C8 noncompetitively with an apparent Ki of 2.5 µm ( Figure 4).

Details are in the caption following the image

Dixon plots for inhibition of CYP2C8 catalysed torsemide tolylmethylhydroxylation by ketoconazole, clotrimazole, terfenadine, midazolam, triazolam and quinine. Concentrations of substrate (torsemide) are shown in the left hand panel of each plot. Each point represents the mean of duplicate measurements.

Clotrimazole, another azole antifungal agent, inhibited CYP2C8 catalysed torsemide hydroxylation by > 50% at an added concentration of 10 µm ( Figure 3). However, at this concentration the azoles econazole, miconazole and bifonazole and the triazoles fluconazole and itraconazole had a negligible effect on CYP2C8 activity ( Figure 3). Even at a 10-fold higher concentration, no inhibition was apparent for bifonazole, fluconazole and itraconazole. Like ketoconazole, clotrimazole was a noncompetitive inhibitor of CYP2C8 with an apparent Ki of 2.5 µm ( Figure 4).

Inhibitory effects of the CYP3A substrates amiodarone, amitriptyline, budesonide, cyclosporin A, diazepam, diltiazem, lignocaine, midazolam, quinine, terfenadine and triazolam on CYP2C8 activity are summarized in Table 1. Two inhibitor concentrations were examined; the lower corresponded to the reported approximate CYP3A4 Km or Ki for each compound while the higher was five times this value. Amiodarone, amitriptyline, quinine, terfenadine and triazolam caused near complete inhibition of CYP2C8 activity at the higher added concentration. Midazolam, quinine, terfenadine and triazolam were the only compounds exhibiting IC50 values below 25 µm (data not shown), and the kinetics of CYP2C8 inhibition by these compounds were subsequently investigated. Midazolam, terfenadine and triazolam were noncompetitive inhibitors with apparent Ki values ranging from 5 to 25 µm, while quinine was a competitive inhibitor with an apparent Ki of 11 µm ( Table 1 and Figure 4).

Table 1. Inhibition of CYP2C8–catalysed torsemide tolylmethylhydroxylation by CYP3A substrates a.
Drug (reported Km, reference) Inhibitor concentrations (µM) %t b inhibition K i (µM) Mechanism of inhibition
Midazolam 3 2 18 noncompetitive
3.6 µm[29] 15 40
Quinine 80 80 11 competitive
80 µm[30] 400 90
Terfenadine 10 68 5 noncompetitive
10 µm[31] 50 82
Triazolam 80 79 25 noncompetitive
74 µm[32] 400 91
Amiodarone c 300 55
310 µm[33]
Amitriptyline 200 62
214 µm[34] 1000 89
Budesonide 10 5
11 µm[35] 50 58
Cyclosporin A 10 7
6 µm[36] 50 14
Diazepam 200 6
190 µm[37] 1000 59
Diltiazem 50 4
51 µm[38] 250 31
Lignocaine 1600 1
1622 µm[39] 8000 2
  • a Torsemide (substrate) concentration was 170 µ m, equivalent to the Km. b Inhibition data are mean values from duplicate determinations. c Higher amiodarone concentrations not investigated due to limited solubility.

Discussion

It has been demonstrated here that concentrations of ketoconazole employed to ‘selectively’ inhibit microsomal CYP3A activity also inhibit CYP2C8. A ketoconazole concentration of 10 µm, not infrequently used to demonstrate CYP3A involvement in a metabolic pathway in vitro, essentially abolished CYP2C8 activity. Apart from ketoconazole, the putative CYP2E1 selective inhibitor diethyldithiocarbamate significantly inhibited CYP2C8 at the lower concentration investigated (50 µm). Quinidine at a concentration of 25 µm, which is10-fold higher than the concentration normally employed for selective inhibition of CYP2D6, reduced CYP2C8 activity by over 30%. However, even at concentrations 5-to 10-fold higher than those used to confer isoform selectivity, furafylline, coumarin, sulphaphenazole, and mephenytoin inhibited CYP2C8 by < 20%.

These data are consistent with previous reports which suggest that, in addition to CYP2E1, diethyldithiocarbamate (125–200 µm) may also inhibit CYP 2A6, 2B6, 2C9, and 2C19 [20, 21] and human liver microsomal paclitaxel 6α-hydroxylation [22]. Harris et al.[22] further demonstrated that ketoconazole, quinidine (> 20 µm) and gestodene (another putative selective CYP3A inhibitor) inhibited human liver microsomal paclitaxel 6α-hydroxylation, but coumarin, mephenytoin and sulphaphenazole had no effect on this pathway. α-Naphthoflavone, used occasionally for ‘selective’ inhibition of CYP1A, is also known to inhibit CYP2C8 [20, 22].

These observations confirm the importance of careful selection of inhibitor concentration when determining the contribution of individual CYP isoforms to a metabolic pathway in human liver microsomes. As noted above, a ketoconazole concentration of 10 µm essentially abolished CYP2C8 activity. Reported apparent Ki values for ketoconazole inhibition of CYP3A4 and CYP3A5 range from 0.02 to 0.11 µm[23, 24], and it would be expected that concentrations of this compound below 1 µm would be necessary to confer CYP3A selectivity.

In addition to ketoconazole, other azole and triazole antifungal agents are known to inhibit CYP-catalysed xenobiotic biotransformation in humans, sometimes in an isoform-selective manner. For example, fluconazole selectively inhibits CYP2C9 [1]. Of the compounds from this class screened here, only clotrimazole and ketoconazole were potent inhibitors of CYP2C8. Both were noncompetitive inhibitors, which is characteristic of azoles and appears to arise from the ability of a nitrogen lone pair of electrons to ligand to the haem [25]. Despite being a potent inhibitor of CYP2C8, clotrimazole is of no value as an isoform-selective probe since it is also known to inhibit potently CYP 2C9, 2E1 and 3A activities [26–28].

As noted previously, there appears to be a degree of overlapping substrate specificity between CYP3A and CYP2C8. Amiodarone, amitriptyline, quinine, terfenadine, and triazolam, which are all metabolized completely or in part by CYP3A, caused > 50% inhibition of CYP2C8 activity at concentrations corresponding to their CYP3A Km values, suggesting similar affinities for both enzymes. While an inhibitor of CYP2D6, quinidine is a substrate for CYP3A and, as shown here, this compound also significantly inhibited CYP2C8 at an added concentration of 25 µm. Apparent Ki values for midazolam, quinine, terfenadine and triazolam ranged from 5 to 25 µm, which is in excess of plasma concentrations observed for the respective compounds during normal therapeutic use. Nevertheless, these results highlight the possibility that some CYP3A substrates may interact with CYP2C8-catalysed xenobiotic metabolism in vivo. The CYP3A substrates nifedipine, felodipine and testosterone have been shown previously to inhibit human liver microsomal paclitaxel 6α-hydroxylation, whereas erythromycin was without effect [22]. Interestingly, both CYP2C8 and CYP3A4 can accommodate substrates of widely varying molecular mass, although the predominant mechanism of inhibition observed here (noncompetitive) is not consistent with the compounds investigated acting as alternate substrates. The lesser or absent inhibition of CYP2C8 by budesonide, cyclosporin A, diazepam, diltiazem and lignocaine indicates that certain, as yet unidentified, structural characteristics are essential for the interaction of CYP3A substrates with CYP2C8.

Acknowledgments

This work was supported by a grant from the National Health and Medical Research Council of Australia. C-E Ong was a recipient of a Flinders University International Postgraduate Research Scholarship.