Possible involvement of multiple human ... - Semantic Scholar

1 downloads 0 Views 181KB Size Report
troleandomycin; 7, 8 bz : 7-8 benzoflavone; — : not performed. Results are expressed as percent inhibition. Liver. Propofol concentration. (M). TAO. (100 M). MID.
British Journal of Anaesthesia 1998; 80: 788–795

Possible involvement of multiple human cytochrome P450 isoforms in the liver metabolism of propofol J. GUITTON, T. BURONFOSSE, M. DESAGE, J.-P. FLINOIS, J.-P. PERDRIX, J.-L. BRAZIER, P. BEAUNE

Keywords: anaesthetics i.v. propofol; cytochrome P450; CYP2C9; interactions (drug)

most of the drug (73%) is eliminated as water-soluble conjugates via the kidneys on the first day.1 3 4 In human beings, the major metabolite is the glucuronic acid conjugate of propofol, which accounts for 53–73% of the total metabolites, depending principally on the administered dose of propofol.1 4 The importance of this metabolic pathway is also species-dependent.3 5 The other metabolites recovered in human urine are the two glucuronic acid (conjugation at the C1 or C4 position) and the sulphate conjugates (conjugation at the C4 position) of the ring-hydroxylated derivative of propofol, 2,6diisopropyl-1,4-quinol (1,4-quinol).1 4 These minor metabolites need a preliminary hydroxylation at the C4 position. Consequently, hydroxylation of the parent compound accounts for a maximum of 47% of the total dose of propofol.1 4 All these metabolic pathways reduce or suppress the activity of propofol. Glucuronidation at the 1 or 4 position or sulphation at the 4 position suppress the activity of the compound while the 1,4-quinol has about one-third the hypnotic activity of propofol.3 This important metabolic clearance to pharmacologically inactive derivatives, associated with a rapid initial distribution phase, limit the effect of propofol and contribute to the excellent recovery from anaesthesia observed in patients.6 The total body clearance of propofol is rapid, and as propofol is eliminated only by metabolism, the liver would be expected to be the organ predominantly responsible for the clearance of propofol,7 8 even if extrahepatic sites such as the kidneys8 or lungs9 10 are also involved in propofol glucuronidation. Previous investigations have shown that propofol decreased animal11 or human12 cytochrome P450 (CYP) activities in vitro. It has been shown that propofol interacts with the haem moiety of the haemoprotein in the CYP enzyme.13 The variable inhibition observed in CYP-mediated activities may be attributable to the differential binding of propofol to CYP isoforms.13 Even if hydroxylation of propofol on its phenol ring is not the major metabolic pathway,1 4 14 the involvement of the CYP-dependent mono-

Propofol (2,6-diisopropylphenol) is one of a series of substituted phenols with highly lipophilic properties resulting in a wide distribution.1 Propofol is used for its short-lasting anaesthetic action, which results from rapid decay of its plasma concentrations.1 2 In humans propofol is eliminated from the body only after being metabolized, because less than 0.3% of the dose is excreted from the body as the parent compound and

J. GUITTON*, PHARMD, M. DESAGE, PHD, J.-L. BRAZIER, PHD, LEACM-ISPB, Université Claude Bernard, Lyon. T. BURONFOSSE, PHD, UMTCX, INRA-DGER, Lyon. J.-P. FLINOIS, PHD, P. BEAUNE, PHD, Centre Universitaire des Saints-Pères, INSERM U 490, Paris. J.-P. PERDRIX, MD, Centre Hospitalier Lyon-Sud, Pierre Bénite. Accepted for publication: February 17, 1998. *Present address and address for correspondence: Université Claude Bernard, 8 avenue Rockefeller, 69373 Lyon Cédex 08, France.

Summary Previous studies of propofol (2,6-diisopropylphenol) pharmacology have shown that this widely used anaesthetic drug is extensively cleared from the body by conjugation of the parent molecule or its quinol metabolite. On the basis of potential influence of propofol on the metabolism of co-administered agents, many investigators have evaluated the effects of propofol on cytochrome P450 (CYP) activities. CYP isoforms involved in propofol metabolism are not defined. In this study, our objective was to elucidate further the CYP isoforms responsible for the hydroxylation of propofol. Using microsomes from 12 different human livers, we investigated CYP isoforms involved in propofol hydroxylase activity, using selective chemical inhibitors of CYP isoforms, correlation with immunoquantified specific CYP isoform content, immunoinhibition, and 11 functionally active human CYP isoforms expressed in a heterologous system (yeast and human B- lymphoblastoid cells). We found a low variability in the production of the hydroxylated metabolite of propofol, 2,6-diisopropyl-1,4-quinol. This activity was mediated by CYP and followed Michaelis—Menten kinetics with apparent KM and Vmax values of 18 ␮M (95% CI 15.1–20.1) and 2.6 nmol min1 mg1 (95% CI 2.45–2.68) respectively. Part of the propofol hydroxylase activity was mediated by CYP2C9 in human liver, especially at low substrate concentration. Moreover, propofol was likely to be metabolized by additional isoforms such as CYP2A6, 2C8, 2C18, 2C19 and 1A2, especially when substrate concentrations are high. This low specificity among CYP isoforms may contribute to the low interindividual variability (two-fold) and may contribute to the low level of metabolic drug interactions observed with propofol. (Br. J. Anaesth. 1998; 80: 788–795)

Hydroxylation of propofol by P450 isoforms oxygenases in this biotransformation could be the basis of metabolic interactions,12 15 or of a genetic polymorphism that could explain the particular metabolic profile observed in a patient who exhibited a high conjugation pathway of propofol and a slight ring-hydroxylation.4 The purpose of the present study was to characterize the isoform(s) of CYP metabolizing propofol. Identification of the major CYP isoform responsible for propofol hydroxylation in humans may aid a better understanding of the inhibitory effects of propofol on CYP-mediated activities.

Materials and methods DRUGS AND CHEMICALS

Propofol and 2,6-diisopropyl-1–4 quinol were obtained from Zeneca Pharma (Cergy, France). Midazolam was supplied by Roche (Neuilly sur Seine, France). Coumarin, sulphaphenazole, quinidine, diethyldithiocarbamate, troleandomycin, 7–8 benzoflavone, glucose-6-phosphate (G6P), glucose-6phosphate dehydrogenase (G6PDH), and nicotinamide adenine dinucleotide phosphate (NADP+) were purchased from Sigma (St Quentin Fallavier, France). All sodium dodecyl sulphate-polyacrylamide gel electrophoresis reagents were obtained from Bio-Rad (Paris, France). Other chemicals were of the highest grade commercially available. HUMAN LIVER MICROSOMES

Experiments with human tissue were approved by the local ethics committee. Fresh human liver samples were obtained from Caucasian patients who underwent partial hepatectomy. Morphologically normal fragments were collected on the material removed in excess and were prepared as reported previously.16 Less than 10 min elapsed from removal of liver to collecting and freezing a sample in liquid nitrogen. Patients with acute or chronic hepatitis and with cirrhosis were excluded from the study. In correlation studies, 12 different human liver samples (HL1 to HL12) were used. PREPARATION OF MICROSOMAL FRACTIONS

Microsomes from human livers were prepared by differential centrifugation as described previously.17 Microsomal fractions were aliquoted in small volumes and stored at 80⬚C in potassium phosphate buffer (200 mM, pH 7.4) containing 1mM EDTA and glycerol (20%, v/v) before use. Total CYP content was measured using the method developed by Omura and Sato.18 Microsomal proteins were assessed according to a modification of Lowry’s method using bovine serum albumin as standard.19

789 from human lymphoblastoïd cells) containing an NADPH-generating system composed of NADP+ (0.5 mM), G6P (5 mM), and G6PDH (1.6 unit), 0.5–0.7 mg of microsomal protein, and 10–50 ␮M propofol. The substrate was prepared daily in methanol and diluted in water (1:100, v/v) before being introduced into the reaction mixture (0.02% methanol final concentration). Reactions were performed at 37 ⬚C for 7–10 min; they were started by adding substrate and stopped by adding 50 ␮l sodium hydroxide (1 mol l1). Total propofol metabolism by liver microsomes was determined as the amount of 2,6-diisopropyl-1,4-quinol produced (1,4-quinol). Propofol and 2,6-diisopropyl-1,4-quinone (transformed under alkaline conditions from 1,4-quinol) in incubation mixtures were determined by using a gas chromatography-mass spectrometry assay used in our laboratory.20 Briefly, propofol, its metabolite, and thymol used as an internal standard were extracted in chloroform/ethyl acetate (70:30, v/v). After centrifugation, 1 of the organic layer was injected into the capillary column (HP5: 30 m0.25 mm i.d.0.25 m film thickness). Propofol, 2,6-diisopropyl-1,4quinone and thymol were detected by selected ion monitoring at mass/charge ratios of 163 (propofol), 149 (2,6-diisopropyl-1,4-quinone) and 135 (thymol). Identification of the metabolite, 1,4-quinol, was based on co-elution and comparison of the mass spectrum with the authentic standard. Precision and accuracy were below 10% (within-run) and below 16% (between-run) at low concentrations. The limit of detection of 1,4-quinol was 25 ng ml1. Coumarin 7-hydroxylase and 7 ethoxycoumarin O-deethylase activities at a substrate concentration of 100 ␮M were determined fluorometrically as described by Aitio and colleagues.21 Tolbutamide hydroxylase activity was determined on four different human microsomes. Microsomal incubations with 1 mM tolbutamide (methanol) and 2 mg protein were assessed in a final volume of 1 ml for 1 h at 37⬚C. Extraction and quantification of hydroxytolbutamide were assayed according to Relling and colleagues with minor modifications.22 The effect of substrate concentration on the rate of production of 1,4-quinol was evaluated with a concentration of propofol from 2 ␮M to 200 ␮M from three different microsomal fractions. All assays were performed in duplicate or triplicate.

DETERMINATIONS OF ENZYMATIC ACTIVITIES

Reaction rates were verified to be linear as a function of incubation time and protein concentrations under the conditions described below. In a final volume of 1 ml, a typical reaction mixture contained phosphate buffer (100 mM, pH 7.4 or Tris-HC1 buffer 50 mM pH 7.4 for incubations performed with microsomes

Figure 1 Eadie–Hofstee plot for the formation of the metabolite 2,6-diisopropyl-1,4-quinol after incubation of microsomes from human liver sample HL3. Values are the mean of three separate determinations.

790

British Journal of Anaesthesia

Figure 2 Relationships between microsomal CYP3A4, CYP2D6, CYP1A2, CYP2E1, CYP2C8, CYP2C9 immunoreactive content and the rate of formation of 2,6-diisopropyl-1,4-quinol by 12 different human liver microsome preparations. Each data point represents the mean of triplicate determinations from each human liver and is expressed in arbritrary units (percent of the fraction expressing the highest immunoreactivity). Statistical analysis was performed using a stepwise multiple regression.

INHIBITION STUDY

Effective chemical inhibitors were used as probes of specific CYP isoforms involved in the metabolism of propofol. Inhibitors were introduced into the incubation medium simultaneously with the substrate except for troleandomycin and orphenadrine; these were preincubated with microsomes and the NADPHgenerating system at 37⬚C for 20 min to produce sufficient reactive metabolites, which led to specific inhibition. The reaction was then continued with substrate for an additional 10 min. Inhibitors used as probes of specific CYP-mediated activities were diethyldithiocarbamate (100 ␮M; CYP2E1), sulphaphenazole (50 ␮M; CYP2C9), quinidine (10 ␮M; CYP2D6), midazolam (100 ␮M; CYP3A4), troleandomycine (100 ␮M; CYP3A4), 7–8 benzoflavone (10 ␮M; CY7P1A1), coumarin (100 ␮M; CYP2A6) and

orphenadrine (50 ␮M; CYP2B6), which are known to suppress more than 80% of the activities mediated by the CYP isoform given in parentheses.23 Clotrimazole (50 ␮M) and n-octylamine (1 mM) were also used for their low specificity in inhibiting CYP-mediated activities. Specific immunoinhibitions were also performed, using purified rabbit polyclonal immunoglobulins (IgG) directed against rat NADPHcytochrome P450 reductase. Rat liver NADPHcytochrome P450 reductase was purified using the method described by Yasukoshi and Masters.24 Immunological inhibition studies were assayed using 0.2 mg of microsomal protein preincubated with 10 mg of purified IgG for 30 min at 37⬚C. The NADPHcytochrome P450 reductase activity was evaluated according to the method described by Masters and colleagues.25 Purified IgG anti-rat NADPHcytochrome P450 reductase was shown to inhibit 80%

Hydroxylation of propofol by P450 isoforms of the cytochrome c reductase activity in human liver micosomes. Immunoinhibition against CYP2C9dependent activity was also performed using serum from primary immunized rabbits by CYP2C9 obtained from yeast expressing the corresponding cDNA. Experimental conditions for immunoinhibition were identical. For each inhibited activity, control experiments were conducted under the same conditions with the same amount of methanol needed for solubilization of the inhibitor (1% final concentration) or with serum obtained from standard rabbits. WESTERN BLOT ANALYSIS

Human hepatic microsomal fractions (2–100 ␮g protein) were subjected to immunoblotting as described previously,26 using rabbit immunoglobulins against the main human CYP isoforms (that is, CYP1A2, 2C8, 2C9, 2D6, 2E1 and 3A4). The intensity of the bands was determined by laser densitometry using a densitometer (Scan Jet 2C, Hewlett Packard) linked to a microcomputer running Scan Analysis software. Various concentrations of liver microsomes were loaded on each gel to establish a standard curve. The amount of each CYP isoform was expressed in arbitrary units and compared with the specific content of the corresponding CYP isoform of the human liver fraction exhibiting the highest activity.

791 1,4-quinol was the only metabolite recovered. The mass spectrum of the metabolite peak was identical to that of authentic 1,4-quinol (data not shown). No other possible metabolite could be found in either the GC-MS system with or without silylation or in an HPLC system with UV detection. Furthermore, assays of propofol metabolism by measurement of the formation of 1,4-quinol or by measurement of propofol disappearance gave similar results. The interindividual variation in the rate of formation of the 1,4-quinol in the 12 different samples was no more than two-fold (0.66–1.30 nmol min1 mg1). Incubations performed without an NADPHgenerating system, or in the presence of clotrimazole, n-octylamine, or IgG anti-rat NADPH-cytochrome P450 reductase decreased the formation of 1,4quinol by more than 80%. The conversion of propofol was linear over the protein concentration range 0.5–3 mg ml1 and 0.2–1 mg ml1, for up to 20 min for 50 ␮M and 10 ␮M of propofol respectively. The propofol hydroxylase activity assayed with different concentrations of the substrate (2–200 ␮M), performed with microsomes from five human livers, exhibited Michaelis—Menten kinetics whose parameters were 18 ␮M (95% CI 15.1–20.1) and 2.6 nmol min1 mg1 (95% CI 2.45–2.68), for the apparent KM and Vmax values, respectively. An Eadie—Hofstee representation of these kinetics in HL3 microsomes is shown in fig. 1.

CDNA-EXPRESSED CYP ISOFORMS

Incubations with CYP expressed in recombinant yeast were also performed to assess the involvement of specific CYP isoforms in the production of 1,4-quinol. Transformed yeast were produced and microsomes were prepared as previously described.27 Incubations were performed in a final volume of 1 ml (28⬚C, 30 min) containing yeast expressing one of the following human CYP isoforms (CYP1A1, 1A2, 2C8, 2C9, 2C18, 2C19, 2D6, 2E1, 3A4 or 3A5: 200 pmol), cytochrome b5 (200 pmol), 10 and 50 ␮M propofol, in a NADPH-regenerating system containing 10 mM MgC12. Microsomes from human lymphoblast cells expressing, CYP2A6 (Gentest, Woburn, MA, USA) were also used under conditions recommended by the manufacturer (37⬚C, 30 min in Tris-HC1 buffer 50 mM, pH 7.4). The relationship between Western blot (or coumarin 7-hydroxylase activity) versus 1,4-quinol formation was calculated using a step-wise multiple regression performed in StatView software (Abaccus Concepts Inc., Berkeley, CA, USA), accepting P 0.05 as significant. Statistical significance in inhibition studies was assessed using Student’s t test with P 0.05. Kinetic parameters (propofol concentration from 2 to 200 ␮M) were obtained by non-linear least squares regression. The minimum values of the sum of the squared residuals were computed using NAG Fundation Library routine E04FDF (Numerical Algorithms Group Ltd, The Math Works Inc., Natick, MA, USA).

INVESTIGATIONS OF HUMAN LIVER CYP INVOLVED IN PROPOFOL HYDROXYLASE ACTIVITY

To characterize the involvement of liver-specific CYP in the metabolism of propofol, the rate of formation of 1,4-quinol by human microsomes was compared with their specific content of CYP isoforms (fig. 2). A stepwise multiple regression was performed to take into account the contribution of different CYP isoforms in propofol metabolism. Activities of all microsomal fractions obtained with a substrate concentration of 10 ␮M exhibited a significant correlation (r0.78, P0.0076) with the microsomal immunoquantified-content of CYP2C9. None of the other immunochemically-tested human CYP isoforms (that is, CYP1A2, 2C8, 2D6, 2E1, and 3A4)

Results PROPOFOL HYDROXYLASE ACTIVITY

Human liver microsomes exhibited propofol hydroxylase activity (0.89 (0.20) nmol min1 mg1) and

Figure 3 Correlation between the 7-OH coumarin hydroxylase activity and the formation of 2,6-diisopropyl-1,4-quinol by human liver microsomes. Each value is the mean of two separate assays.

792

British Journal of Anaesthesia Table 1 Formation of 2,6-diisopropyl-1,4-quinol by specific human cytochrome P450 (CYP) isoforms expressed in heterologous systems. Results represent the mean of duplicate or triplicate determinations. Incubations were performed with microsomes from transfected yeast or from human B-lymphoblastoid cells (*) incubated with 10 ␮M or 50 ␮M substrate concentration. NDnot determined Recombinant human CYP isoforms 1A1 1A2 2A6* 2C8 2C9 2C18 2C19 2D6 2E1 3A4 3A5

2,6-diisopropyl 1,4-quinol formation (nmol min1 nmol P4501) (Propofol 10 ␮M)

(Propofol 50 ␮M)

Vmax/Km (mg min1 nmol P4501)

0.89 0.12 0.90 0.26 1.24 0.25 0.18 0.01 0.01 0.01 0.01

3.96 1.84 2.99 1.10 5.65 1.36 0.82 0.12 0.01 0.04 0.01

ND 0.011 ND 0.016 0.118 ND ND ND ND 0.0007 ND

Table 2 Effect of selective cytochrome P450 (CYP) inhibitors on the metabolism of propofol. COUM coumarin; DDC diethyldithiocarbamate; MID  midazolam; ORPH  orphenadrine; QUIN  quinidine; SULF  sulfaphenazole; TAO  troleandomycin; 7, 8 bz  7-8 benzoflavone; —  not performed. Results are expressed as percent inhibition

Liver HL3 HL4 HL5 HL7 HL11

Propofol concentration (␮M)

TAO (100 ␮M)

MID (100 ␮M)

SULF (50 ␮M)

COUM (100 ␮M)

ORPH (50 ␮M)

QUIN (10 ␮M)

DDC (100 ␮M)

7,8 BZ (10 ␮M)

10 50 10 10 50 10 50 10

— — — — 9 — — —

— 15 — — 12 — 15 —

23 10 19 39 1 25 7 21

— 8 — — 29 — 25 —

— — — — 25 — — —

— — — — 0 — — —

— 17 — — 9 — 11 —

0 — — 13 8 5 — —

could be related to the specific activity of each microsome. Similarly, there was no significant correlation between the rate of formation of 1,4-quinol and the coumarin 7-hydroxylase activity (r0.42, fig. 3), a marker of CYP2A6-mediated activities, or total CYP content. When incubations were assessed with a higher concentration of propofol (50 ␮M), correlation between the rate of formation and CYP2C9 was not significant (r 0.51). Heterologous systems expressing 11 different CYP isoforms were used to assess the intrinsic ability of these isoforms to catalyse propofol. Results are shown in table 1. Microsomes from yeast expressing CYP2C9 exhibited a high activity for formation of 1,4-quinol. Lymphoblast cells expressing CYP2A6 were also reasonably active. Other cDNAs of CYP expressed in yeast (that is, CYP1A1, 2C8, 2C18, 2C19, or 1A2) produced the 1,4-quinol, at low concentration of substrate (10 ␮M). With an increase in the substrate concentration (50 ␮M), other isoforms hydroxylated propofol but the activities remained negligible. These concentration are commonly observed in plasma after a single bolus for induction of anaesthesia,2 28 although the actual intracellular concentration around the active sites is uncertain. Enzymatic efficency (Vmax/KM) of several isoforms was also determined (table 1), CYP2C9 exhibiting the highest rate. Selective chemical CYP inhibitors were also used in this study to determine the potential roles of CYP isoforms in the metabolism of propofol (table 2). A moderate inhibition (around 25%) was observed in

incubations performed, with a 10 ␮M concentration of propofol, with sulphaphenazole as a competitive inhibitor of CYP2C9 despite a broad inhibition of the tolbutamide hydroxylase activity under the same conditions. The percentage inhibition diminished when the concentration of the substrate was set at 50 ␮M. Similar inhibitions were found with coumarin (21%) or orphenadrine. No other agent tested significantly inhibited propofol hydroxylase activity. Immunoinhibition assessed with IgG antiCYP2C9, ranging from 25 to 75 ␮l of immune serum, exhibited a decrease in activity from 3.7% to 46.7% respectively (values are the mean of three assays performed separately in microsomes from three samples selected for their high response in immunoblot performed with anti-CYP2C9).

Discussion In human microsomal fractions, 1,4-quinol was the single hydroxylated metabolite of propofol recovered. In vivo studies showed the presence of sulphate or glucuronide conjugates of propofol or of its hydroxylated metabolite: 1,4-quinol.1 4 1,4-quinol is the only identified hydroxylated metabolite of propofol and could represent as much as 47% of the total metabolism of propofol over 1 day.1 4 Absence of an NADPH-generating system, or co-incubations with clotrimazole, n-octylamine or IgG anti-reductase from rat dramatically decrease the production of the 1,4-quinol indicating the activity of CYP in propofol metabolism. These results are complementary with

Hydroxylation of propofol by P450 isoforms those previously reported by Chen and colleagues, who observed that propofol competed with carbon monoxide for a common binding site on the haem moiety.13 Propofol hydroxylation follows Michaelis— Menten kinetics and is consistent with the existence of a single active site or different catalytic sites with close affinities, because the existence of two or more CYP with similar affinities for propofol would not be distinguished in the kinetic studies. The KM value is near physiological plasma concentrations.2 29 The present study clearly demonstrates the involvement of CYP2C9 in the metabolism of propofol. This conclusion is supported by several lines of evidence. A significant correlation between immunochemically determined CYP and the hydroxylation of propofol was confirmed only for CYP2C9 at a concentration (10 ␮M) near the value of the Michaelis—Menten constant. Similarly, immunoinhibition and CYP2C9 expressed in yeast confirm the participation of this isoform in the production of the 1,4-quinol, despite a weak inhibition of the activity in fractions coincubated with sulphaphenazole, a chemical inhibitor of CYP2C9-catalysed activities.30 Although sulphaphenazole is an efficient inhibitor of reactions mediated by CYP2C9 with Ki values from 0.12 to 0.3 ␮M on tolbutamide hydroxylation,30 31 it was unable to inhibit more than 39% of the production of 1,4-quinol. Nevertheless, some studies suggest the possible involvement of isoforms other than CYP2C9 in the metabolism of propofol. First, the linear regression of the relationship between the immunochemical reactivity of CYP2C9 and the production of 1,4-quinol has a y-intercept significantly different from zero (46% of residual activity compared with the mean of the activity). Second, immunological and chemical inhibitions decrease propofol hydroxylase activity, which remains below 50% of the control activity. Finally, correlation between the propofol hydroxylase activity determined with a substrate concentration of 50 ␮M and immunochemical-CYP2C9 reactivity was not significant despite a two-fold increase in this activity (data not shown). These data suggest the participation of several isoforms, which should account for the remaining activity at therapeutic concentrations. Microsomes prepared from cells containing various CYP isoforms were used to support these arguments. Thus, microsomes from cells expressing CYP2A6 or 1A1 raised to a high propofol hydroxylase activity even at low substrate concentrations as well as several isoforms (that is, 2C8, 2C18, 2C19, 1A2) that exhibited a lower activity and a low enzymatic efficiency. These heterologous systems provide evidence on the intrinsic capacity of a specific isoform to catalyze the reaction32 33 and strongly suggest in vivo participation if the expression level in organs or tissues involved in the metabolic pathways of the drug, or the concentration of the drug around the active site, or both, are sufficient.34 Nevertheless, the involvement of these isoforms in metabolic pathways may be overestimated. Thus, when we focus on the effective involvement of these isoforms in microsomal fractions, all results are inconclusive. This may be the consequence of a low expression of these specific isoforms in the microsomal fractions (for example, CYP1A1 is usually expressed in liver at very low level, CYP2A6

793 accounts for about 4% of total CYP) or a narrow range of levels of some CYP isoforms in our liver bank (especially 2A6, 3A4). Many investigators have studied the possible interaction between propofol and other anaesthetic drugs such as alfentanil and derivatives12 13 15 35 or enflurane11 because these compounds are often associated in anaesthetic procedures. Previous studies in humans did not show any pharmacokinetic interactions between propofol and fentanyl36 or propofol and alfentanil.35 However, Janicki and colleagues showed a decrease in the in vitro oxidative metabolism of alfentanil and sufentanil by propofol.12 The metabolic transformation of these opioids is primarily mediated by CYP3A4.37 38 Furthermore, the present study shows a low participation of 3A4 even at high concentrations of propofol. Nevertheless, these data do not rule out a possible interaction between propofol and this isoform that might explain the interaction observed by Janicki and colleagues. In contrast, other authors showed, in rabbits, that pretreatment with cimetidine, an inhibitor of CYP3A4-mediated activities,39 did not affect the in vitro rate of propofol elimination.9 Our study also clearly shows that CYP2E1 is not involved in propofol metabolism at low substrate concentration as was previously observed in rats by Baker and colleagues.11 Nevertheless, some authors have observed an in vitro concentration-dependent inhibitory effect of propofol on CYP2E1-mediated activities in hamster and in human.13 15 These discrepancies may also be explained by the difference in the substrate concentration, because inhibition studies in human microsomes were performed with 0.5 and 1 mM, concentrations that are not clinically relevant.13 In the same way, earlier in vivo studies showed that the plasma concentrations of propofol were higher in patients pretreated with fentanyl.6 It seems that this kind of interaction may be related to an inhibition of the two compounds on the glucuronidation pathways8 rather than to a direct interaction on CYP. In vitro inhibition by propofol of the glucuronidation of propranolol has already been observed.40 41 Nevertheless, the rarity of the interactions occurring between clinically used drugs and propofol may be explained by other events. First, glucuronidation which is a major pathway of the metabolism of propofol may be catalysed by different isoforms of UDP-glucuronosyltransferases and the latter are known to exhibit a broad specificity towards a large number of compounds.8 42 43 Secondly, the hepatic extraction coefficient for propofol is very high. In conclusion, CYP2C9 is involved, by at least 50%, in the oxidative metabolism of propofol, particularly at low substrate concentration. Other isoforms, such as CYP2A6, 2C8, 2C18, 2C19 and 1A2, are also involved in this metabolism but the characterization of their relative activities remains difficult. It can be postulated that the increase in propofol concentration or the defect of CYP2C9 contributes to the participation of this CYPs, or promotes the glucuronidation, a pathway known to be of low specificity and high capacity.8 From the clinical point of view, the contribution of multiple isoforms in a metabolic pathway of a highly metabolized therapeutic agent might be desirable because its metabolism is not affected by polymorphism and is

794 less affected by metabolic interactions with other drugs, except in individuals expressing a high proportion of one of these isoforms.

Acknowledgements This work was made possible by to the Bioavenir Program supported by Rhône-Poulenc-Rorer, Roussel-Uclaf and the Ministère de la Recherche of the French government. The authors thank Dr Berny for correction of the manuscript, Drs Pignal and Ducerf for participation in the biological sample collection, and Dr Delignette-Müller for advice on statistical analysis.

British Journal of Anaesthesia

18. 19. 20. 21. 22.

References 1. Simons PJ, Cockshott ID, Douglas EJ, Gordon EA, Hopkins K, Rowland M. Disposition in male volunteers of a subanaesthetic intravenous dose of an oil in water emulsion of 14C-propofol. Xenobiotica 1988; 18: 429–440. 2. Vree TR, Baars AM, De Grood PMRM. High-performance liquid chromatographic determination and preliminary pharmacokinetics of propofol and its metabolites in human plasma and urine. Journal of Chromatography B 1987; 417: 458–464. 3. Simons PJ, Cockshott ID, Glen JB, Gordon EA, Knott S, Ruane RJ. Disposition and pharmacology of propofol glucuronide administrered intravenously to animals. Xenobiotica 1992; 22: 1267–1273. 4. Sneyd JR, Simons PJ, Wright B. Use of proton nmr spectroscopy to measure propofol metabolites in the urine of the female caucasian patient. Xenobiotica 1994; 24: 1021–1028. 5. Simons PJ, Cockshott ID, Douglas EJ, Gordon EA, Knott S, Ruane RJ. Species differences in blood profiles, metabolism and excretion of 14C-propofol after intravenous dosing to rat, dog and rabbit. Xenobiotica 1991; 21: 1243–1256. 6. Cockshott ID, Douglas EJ, Prys-Roberts C, Turtle M, Coates DP. The pharmacokinetics of propofol during and after intravenous infusion in man. European Journal of Anaesthesiology 1990; 7: 265–275. 7. Cockshott ID, Douglas EJ, Plummer GF, Simons PJ. The pharmacokinetics of propofol in laboratory animals. Xenobiotica 1992; 22: 369–375. 8. Le Guellec C, Lacarelle B, Villard P-H, Point H, Catalin J, Durand A. Glucuronidation of propofol in microsomal fractions from various tissues and species including humans: Effect of different drugs. Anesthesia and Analgesia 1995; 81: 855–861. 9. Audibert G, Saunier CG, du Souich P. In vivo and in vitro effect of cimetidine, inflammation, and hypoxia on propofol kinetics. Drug Metabolism and Disposition 1993; 21: 7–12. 10. Raoof AA, Augustijns PF, Verbeeck RK. in vivo assessment of intestinal, hepatic, and pulmonary first pass metabolism of propofol in the rat. Pharmaceutical Research 1996; 13: 891–895. 11. Baker MT, Chadam MV, Ronnenberg WC. Inhibitory effects of propofol on cytochrome P450 activities in rat hepatic microsomes. Anesthesia and Analgesia 1993: 76: 817–821. 12. Janicki PK, James MFM, Erskine WAR. Propofol inhibits enzymatic degradation of alfentanil and sufentanil by isolated liver microsomes in vitro. British Journal of Anaesthesia 1992; 68: 311–312. 13. Chen TL, Ueng TH, Chen SH, Lee PH, Fan SZ, Liu CC. Human cytochrome P450- mono-oxygenases system is suppressed by propofol. British Journal of Anaesthesia 1995; 74: 558–562. 14. Kanto J, Gepts E. Pharmacokinetic implications for the clinical use of propofol. Clinical Pharmacokinetics 1989; 5: 308–326. 15. Chen TL, Wang MJ, Huang CH, Liu CC, Ueng TH. Difference between in vivo and in vitro effects of propofol on defluorination and metabolic activities of hamster hepatic cytochrome P450-dependent mono-oxygenases. British Journal of Anaesthesia 1995; 75: 462–466. 16. Kobayashi K, Chiba K, Tani M, Kuroiwa Y, Ishizaki T. Development and preliminary application of a high-performance liquid chromatographic assay for omeprazole metabolism in human liver microsomes. Journal of Pharmaceutical and Biomedical Analysis 1994; 12: 839–844. 17. Kremers P, Beaune P, Cresteil T, De Graeve J, Columelli S, Leroux JP, Gielen JE. Cytochrome P450 monooxygenase

23.

24.

25. 26.

27.

28. 29. 30.

31.

32. 33. 34. 35. 36. 37.

38.

39.

activities in human and rat liver microsomes. European Journal of Biochemistry 1981; 118: 599–606. Omura T, Sato R. The carbon monoxide binding pigment of liver microsomes. Evidences for its hematoprotein nature. Journal of Biological Chemistry 1964; 239: 2370–2378. Hartree EF. Determination of protein: a modification of the Lowry method that gives a linear photometric response. Analytical Biochemistry 1972; 48: 422–427. Guitton J, Buronfosse T, Sanchez M, Désage M. Quantitation of propofol metabolite, 2,6-diisopropyl-1,4-quinol, by gas chromatography-mass. Analytical Letters 1997; 30: 1369–1378. Aitio A. A simple and sensitive assay of 7-ethoxycoumarin O-deethylation. Analytical Biochemistry 1978; 85: 488–491. Relling MV, Aoyama T, Gonzalez FJ, Meyer UA. Tolbutamide and mephenytoin hydroxylation by human cytochrome P450s in the CYP2C subfamily. Journal of Pharmacology and Experimental Therapeutics 1990; 252: 442–447. Newton DJ, Wang RW, Lu AYH. Cytochrome P450 inhibitors. Evaluation of specificities in the in vitro metabolism of therapeutic agents by human liver microsomes. Drug Metabolism and Disposition 1995; 23: 154–158. Yasukoshi Y, Masters BSS. Some properties of a detergent solubilized NADPH-cytochrome c (cytochrome P-450) reductase purified by biospecific affinity chromatography. Journal of Biological Chemistry 1976; 251: 5337–5344. Masters BSS, Williams, JR. CH, Kamin H. The preparation and properties of microsomal TPNH-cytochrome c reductase from pig liver. Methods in Enzymology 1967; 10: 565–573. De Waziers P, Cugnenc PH, Yang CS, Leroux JP, Beaune PH. Cytochrome P450 isoenzymes, epoxide hydrolase and glutathione transferases in rat and human hepatic and extrahepatic tissues. Journal of Pharmacology and Experimental Therapeutics 1990; 253: 387–394. Lopez-Garcia MP, Dansette PM, Valadon P, Amar C, Beaune PH, Guengerich FP, Mansuy D. Human-liver cytochromes P-450 expressed in yeast as tools for reactive-metabolite formation studies: Oxidative activation of tienilic acid by cytochromes P-450 2C9 and 2C10. European Journal of Biochemistry 1993; 213: 223–232. Yu H-Y, Liau J-K. Quantitation of propofol in plasma by capillary gas chromatography. Journal of Chromatography B 1993; 615: 77–81. Bryson HM, Fulton BR, Faulds D. Propofol — an update of its use in anaesthesia and conscientious sedation. Drugs 1995; 50: 513–559. Miners JO, Smith KJ, Robson RA, McManus ME, Veronese ME, Birkett DJ. Tolbutamide hydroxylation by human liver microsomes: Kinetic characterisation and relationship to other cytochrome P-450 dependent xenobiotic oxidations. Biochemical Pharmacology 1988; 37: 1137–1144. Bourrié M, Meunier V, Berger Y, Fabre G. Cytochrome P450 isoform inhibitors as a tool for the investigation of metabolic reactions catalyzed by human liver microsomes. Journal of Pharmacology and Experimental Therapeutics 1996; 277: 321–332. Gonzalez FJ, Crespi CL, Gelboin HV. cDNA-expressed human cytochrome P450s: A new age of molecular toxicology and human risk assessment. Mutation Research 1991; 247: 113–127. Gonzalez FJ. Human cytochromes P450. Problems and prospects. Trends in Pharmacological Sciences 1992; 13: 346–352. Kato R, Yamazoe Y. The importance of substrate concentration in determining cytochromes P450 therapeutically relevant in vivo. Pharmacogenetics 1994; 4: 359–362. Gepts E, Jonckheer K, Maes V, Sonck W, Camu F. Disposition kinetics of propofol during alfentanil anaesthesia. Anaesthesia 1988; 43: 8–13. Gill SS, Wright EM, Reilly CS. Pharmacokinetic interaction of propofol and fentanyl: Single bolus injection study. British Journal of Anaesthesia 1990; 65: 760–765. Tateishi T, Krivoruk Y, Ueng Y-F, Wood AJJ, Guengerich FP, Wood M. Identification of human liver cytochrome P-450 3A4 as the enzyme responsible for fentanyl and sufentanil N-dealkylation. Anesthesia and Analgesia 1996; 82: 167–172. Kharasch ED, Thummel KE. Human alfentanil metabolism by cytochrome P450 3A3/4: An explanation for the interindividual variability in alfentanil clearance? Anesthesia and Analgesia 1993; 76: 1033–1039. Knodell RG, Browne DG, Gwozdz GP, Brian WR, Guengerich FP. Differential inhibition of individual human liver cytochromes P-450 by cimetidine. Gastroenterology 1991; 101: 1680–1691.

Hydroxylation of propofol by P450 isoforms 40. Whalley PM, Tucker GT, Reilly CS. Effect of propofol on propranolol metabolism by human liver microsomes. British Journal of Anaesthesia 1993; 71: 306P. 41. Whalley PM, Tucker GT, Reilly CS. The effect of propofol on propranolol metabolism by rat and dog liver microsomes. British Journal of Clinical Pharmacology 1992; 34: 163P.

795 42. Ebner T, Burchell B Substrate specificities of two stably expressed human liver UDP-glucuronosyltransferases of the UGT1 gene family. Drug Metabolism and Disposition 1993; 21: 50–55. 43. Burchell B, Brierley CH, Rance D. Specificity of human UDP-glucuronosyltransferases and xenobiotic glucuronidation. Life Science 1995; 57: 1819–1831.