Characterization of human cytochrome P450 enzymes involved in the metabolism of cyamemazine

e u r o p e a n j o u r n a l o f p h a r m a c e u t i c a l s c i e n c e s 3 2 ( 2 0 0 7 ) 357–366

available at www.sciencedirect.com

journal homepage: www.elsevier.com/locate/ejps

Characterization of human cytochrome P450 enzymes involved in the metabolism of cyamemazine Christophe Arbus a , Amine Benyamina b , Pierre-Michel Llorca c , Franck Bayl´e d , Norbert Bromet e , Fr´ed´eric Massiere f , Ricardo P. Garay g,∗ , Ahc`ene Hameg h a

CHU Toulouse, Toulouse, France ˆ INSERM U669, Hopital Universitaire Paul-Brousse, Villejuif, France c CHU Clermont-Ferrand, Clermont-Ferrand, France d INSERM EMI-E0117, Universit´e Paris V, Paris, France e Biotec Centre, Orl´eans, France f Biopredic, Rennes, France g EA2381, Universit´e Paris VII, Paris, France h Sanofi-Aventis France, Paris, France b

a r t i c l e

i n f o

a b s t r a c t

Article history:

Recombinant human liver microsomal enzymes of the cytochrome P450 family (CYP1A2,

Received 5 July 2007

CYP2A6, CYP3A4, CYP2B6, CYP2C8, CYP2C9, CYP2C19, CYP2D6, CYP2E1) were used to

Received in revised form

determine the metabolic fate of the antipsychotic anxiolytic agent cyamemazine. An

29 August 2007

LC/MS–MS tandem methodology was developed specifically for identifying the presence

Accepted 11 September 2007

of cyamemazine and its metabolites in reaction media. All P450 enzymes investigated,

Published on line 14 September 2007

with the exception of CYP2A6 and CYP2E1, degraded cyamemazine, albeit to a different extent, with CYP1A2, CYP2C8 and CYP2C19 being the most efficient (>80%). However, in

Keywords:

microsomes prepared from native human hepatocytes, only relatively specific competi-

Cyamemazine

tors (inhibitors and/or substrates) of CYP1A2, CYP2C8, CYP2C9 and CYP3A4 reduced notably

Antipsychotics

the degradation cyamemazine. The main routes of cyamemazine biotransformation are N-

Metabolism

mono-demethylation (CYP1A2, CYP3A4 and CYP2C8) and mono-oxidation (either S-oxidized

CYP

or hydroxylated derivatives which could not be discriminated because characterized by the same mass value) by CYP1A2 and CYP2C9. Secondary metabolic routes yields N,N-didemethylated and N-demethylated mono-oxidized products. Thus, under in vitro conditions, cyamemazine is extensively degraded by at least four distinct P450 enzymes, into two primary hydrophilic metabolites. These results suggest that cyamemazine detoxification process is unlikely to be significantly impaired by co-administration of therapeutic agents that are substrates of the CYP metabolic system. © 2007 Elsevier B.V. All rights reserved.

1.

Introduction

The phenothiazine derivative cyamemazine (Tercian® ) is the most widely used antipsychotic drug in France (Gury et al.,

∗

2006). Cyamemazine was primarily developed as an antipsychotic agent blocking central dopamine D2 receptors, but clinical experience and animal data showed that it possesses anxiolytic activity, particularly in schizophrenic and

´ ´ Corresponding author at: 46bis, rue du Marechal Gallieni, 91360 Villemoisson-sur-Orge, France. Tel.: +33 169048034; fax: +33 169048034. E-mail address: [email protected] (R.P. Garay). 0928-0987/$ – see front matter © 2007 Elsevier B.V. All rights reserved. doi:10.1016/j.ejps.2007.09.003

358

e u r o p e a n j o u r n a l o f p h a r m a c e u t i c a l s c i e n c e s 3 2 ( 2 0 0 7 ) 357–366

Table 1 – List of competitors and concentrations used effectively to block or saturate the enzymatic activity of the P450 enzymes in microsomes preparations from native human hepatocytes CYP enzyme

CYP1A2 CYP1A2 CYP2A6 CYP2C8 CYP2C9/2C8 CYP2C9 CYP2C19 CYP2D6 CYP2D6 CYP2E1 CYP3A4 CYP3A4

Enzyme reaction

De-ethylase De-ethylase 7-Hydroxylase 6␣-Hydroxylase Hydroxylase Hydroxylase 4-Hydroxylase O-Demethylase 6-Hydroxylase 6-Hydroxylase Oxidase Oxidase

Native microsomes Competitor

Function

Phenacetin 400 ␮M Furafylline 10 ␮M Coumarin 30 ␮M Paclitaxel 100 ␮M Tolbutamide 1 mM Sulfaphenazole 5 ␮M S-mephenytoin 400 ␮M Dextromethorphan 200 ␮M Quinidine 2 ␮M Chlorzoxazone 500 ␮M Nifedipine 200 ␮M Ketoconazole 2 ␮M

Substrate Inhibitor Substrate Substrate Substrate Inhibitor Substrate Substrate Inhibitor Substrate Substrate Inhibitor

Recombinant enzymes Enzyme

activitya

Enzyme activitya

1.3

6.5

1.3

1.4

0.104

0.97

0.030 0.080

0.93 0.498

2.5 1.3

2.97 3.83

Recombinant enzyme activity is given for comparison. a

nmol/(min mg) microsomal protein.

depressed patients with suicidal tendencies (for review see Bourin et al., 2004). Cyamemazine is frequently used as anxiolytic drug in association with other antipsychotic compounds. However, little is known concerning potential drug interactions with cyamemazine metabolism. Indeed, fewer data are available on the pharmacokinetics and metabolism of cyamemazine. Thus, cyamemazine is known to be metabolized in the liver (half-life 11 h) in two main metabolites: demethylated and sulfoxide (Bourin et al., 2004). The aim of this study was to identify the human cytochromes P450 involved in the biotransformation of cyamemazine, in order to predict potential metabolic interactions with other drugs in clinical use, and to investigate the potential role of metabolites in the pharmacological activity. To fulfil these goals, a two-step strategy was followed: (i) cyamemazine was tested in vitro with recombinant (cDNA-expressed) human cytochrome P450 enzymes (CYP), in order to determine which isoforms are capable to metabolize the drug and (ii) the test compound was incubated with human liver microsomes and a number of CYP-specific competitors, in order to assess the importance of each enzyme in cyamemazine biotransformation. For identifying cyamemazine and its possible metabolites in biological reaction media, an LC/MS–MS tandem methodology was specifically developed.

2.

Materials and methods

2.1. Preparation and characterization of microsomes containing human recombinant P450 enzymes Recombinant human liver P450 enzymes (CYP1A2, CYP2A6, CYP3A4, CYPB6, CYP2C8, CYP2C9, CYP2C19, CYP2D6, CYP2E1) were expressed singly in BTI-TN-5B1-4 insect cell line infected with a baculovirus (Autographa Californica) expression system at Gentest Corporation (Woburn, MA, USA) from which they were purchased. The isoenzymes were delivered in frozen microsome fractions, added with P450 reductase SupersomesTM (1 nmol/ml) and before their use they were

tested for enzymatic activity with established specific substrates and inhibitors (Table 1).

2.2. Enzymatic reactions protocol using recombinant human liver P450 enzymes Enzymatic reactions were carried out in 50 mM Tris–HCl buffer (pH 7.4) at 37 ◦ C to which a microsome preparation endowed with single recombinant human P450 enzymes was added. SupersomesTM prepared from cells transfected with the vector lacking CYP genes were used for generating nonCYP-catalyzed control results. The reaction procedure consisted of 5-min incubation in 37 ◦ C buffer solution containing cyamemazine (20 or 200 ␮M) and the cofactor NADPH (1 mM) → addition of the ice-cold microsomal preparation (100 pmol/ml buffer of P450) followed by 60-min incubation at 37 ◦ C → addition of 0.5 ml/tube of ice-chilled acetonitrile to stop the enzymatic reaction → 2.5 × g centrifugation at 4 ◦ C for 10 min → collection of the buffer/acetonitrile supernatant in two 0.5 ml tubes and storage at −80 ◦ C before chemical analysis. Parallel control reaction media consisted of buffer solutions with: • No cofactor, no microsomes, no cyamemazine, not incubated (matrix control). • No cofactor, no microsomes and cyamemazine, not incubated (stability control). • No cofactor, no microsomes and cyamemazine, incubated for 60 min (possible NADPH-independent biotransformation control). • Cofactor, control SupersomesTM lacking any CYP enzyme, and cyamemazine, incubated for 60 min (no CYP-catalyzed biotransformation control).

2.3. Preparation and characterization of native human hepatocyte microsomes Microsomes from human hepatocytes used in this investigation were prepared from a pool of 29 liver tissues

e u r o p e a n j o u r n a l o f p h a r m a c e u t i c a l s c i e n c e s 3 2 ( 2 0 0 7 ) 357–366

(Biopredic International, Rennes, France) and characterized for protein concentration (Bradford, 1976), total CYP content (Omuraand and Sato, 1964) and phase I metabolic activity catalyzed by CYP1A2 deethylase (tested with phenacetin), CYP2A6 7-hydroxylase (coumarin), CYP2C9 hydroxylase (tolbutamide), CYP2C19 4-hydroxylase (Smephenytoin), CYP2D6 O-demethylase (dextromorphan), CYP2E1 6-hydroxylase (chlorzoxazone) and CYP3A4 oxidase (nifedipine) (Distlerath et al., 1985; Greenleeand and Poland, 1978; Guengerich et al., 1986; Dekant et al., 1995; Miners et al., 1988).

2.4. Enzymatic reaction protocol using native human hepatocyte microsomes. This procedure allows testing the effect on cyamemazine biotransformation, of selected specific inhibitors or substrates of P450 enzymes present in native human hepatocyte microsomes. Enzymatic reactions were carried out in 50 mM Tris–HCl buffer (pH 7.4) at 37 ◦ C. The concentrations of cyamemazine tested were 20 and 200 ␮M and the concentrations of established inhibitors and substrates were selected by taken into account the inhibitor Ki and the substrate Km values, respectively (Table 1) (Pelkonen et al., 1998; Clarke, 1998). Competitors were dissolved in DMSO and the final concentration of DMSO in any test tube was 0.5% (v/v). The reaction procedure consisted of: 5-min incubation of the microsome preparation (1 mg protein/ml) and a competitor in 37 ◦ C buffer solution → addition of cyamemazine (20 and 200 ␮M) and NADPH (1 mM) followed by adjustment of final incubation volume to 0.5 ml (DMSO, 0.5%) and 20-min incubation at 37 ◦ C → addition of 0.5 ml/tube of ice-chilled acetonitrile to stop the enzymatic reaction and 3 × g centrifugation at 4 ◦ C for 10 min → collection of the buffer/acetonitrile supernatant in two 0.5 ml tubes and storage at −80 ◦ C until chemical analysis. Control solutions run in a parallel manner were: • No cofactor, no microsomes, no competitor, no incubation (matrix control). • No cofactor, no microsomes, no competitor and cyamemazine, no incubation (stability control). • No cofactor, no microsomes, no competitor and cyamemazine, incubation for 20 min (stability control). • No cofactor, microsomes, no competitor, and cyamemazine, incubation for 20 min (NADPH-independent biotransformation control). • Cofactor, microsomes, no competitor and no cyamemazine, incubation for 20 min (analytical interference control). The following special procedure involving additional controls was used to study the time-dependent competition produced by furafylline: preincubation of microsomes (1 mg protein/ml) in the presence of NADPH (1 mM) at 37 ◦ C for 2 min → addition of furafylline (20 and 200 ␮M) and adjustment of the final incubation volume to 0.5 ml (DMSO, 0.5%) → incubation at 37 ◦ C for 20 min → addition of 0.5 ml/tube of ice-chilled acetonitrile and refrigerated (4 ◦ C) centrifugation at 3 × g for 10 min → collection of

359

buffer/acetonitrile supernatant in two tubes → storage of these tubes at −80 ◦ C until chemical analyses.

2.5. LC/MS–MS structural characterization of cyamemazine and its metabolites A HPLC method coupled to MS–MS detection was developed for determining the in vitro metabolic pathways of cyamemazine. The equipment used consisted of a LC pump (Perkin-Elmer, series 200), a LC autosampler (PerkinElmer, series 200), columns (Purospher RP 18 end capped 125 mm × 4.0 mm, Merck) and a MS–MS detector (PE Sciex API 300). The chromatographic conditions used were distilled water/acetonitrile (70/30, v/v) containing 0.1% formic acid (eluent; injected at 1.0 ml/min flow rate at room temperature) and ion spray positive ionisation for mass detection. Acetonitrile present in reaction media was evaporated and a 20-␮l sample of the remaining solution was directly injected by an automatic injector into the MS–MS spectrometer in which the detection parameters were set to: IS 5700, NC 0, TEM 0; OR 30, RNG 200, QO −10, IQ1 −11, ST −15, RO1 −11, IQ2 −29, RO2 −34, IQ3 −49, RO3 −39, DF −400, CEM 2100, NEB 14, CUR 8, CAD 1, QPE 0, POL 0, VCM 0 and IPE 0. Use of chlorpromazine as an external standard was attempted but this compound proved to be too poorly soluble in the Tris-buffer reaction medium adopted for this investigation. Unknown concentrations of cyamemazine expressed as percentages amounted to 85–115% of the initial 20 ␮M nominal concentration and were plotted to generate a standard calibration curve. Thus, experimental results within this range of values were considered to reflect fractions of cyamemazine not degraded while amounts lower than 85% were assumed to indicate quantities of cyamemazine biotransformed by the investigated CYP enzyme.

2.6.

Compounds

Cyamemazine acid tartrate (batch number 981921) was provided by Aventis Pharma (Antony, France) and was dissolved in distilled water. All other chemical agents used in this investigation were purchased from Sigma (St. Louis, USA) and other commercial sources.

2.7.

Expression of results

Results are reported as means of two replication determinations expressed as percentage inhibitions of the nominal concentrations of cyamemazine added to the reaction media.

3.

Results

3.1. Characterization of recombinant and native P450 enzymes Nine batches of microsomes containing each a recombinant human liver CYP enzyme (CYP1A2, CYP2A6, CYP2B6, CYP2C8, CYP2C9, CYP2C19, CYP2D6, CYP2E1, and CYP3A4) expressed

360

e u r o p e a n j o u r n a l o f p h a r m a c e u t i c a l s c i e n c e s 3 2 ( 2 0 0 7 ) 357–366

in an insect cell line were used to determine cyamemazine biodegradation pathways. The enzymatic activity of the P450 enzymes was routinely determined with appropriate probe substrates and/or inhibitors (Table 1). It is worth to note that incubation conditions, such as the nature and the concen-

tration of salts or organic salt used as intermediary organic solvents may modify by a factor of 2–3 P450 activity values (Busby et al., 1999; Crespi, 1998). Cytochrome P450 content and CYP-dependent microsomal enzymatic capacity of human hepatocyte micro-

Fig. 1 – Mass spectrum analysis of cyamemazine. (Top) Positive ionization mode of cyamemazine. Cyamemazine appeared as a single ion species: [M + H]+ m/z: 324. (Middle) MS–MS conditions. Four distinct daughter ions appeared: 279, 237, 100 and 58. (Bottom) Proposed filiation scheme for daughter ions of cyamemazine.

e u r o p e a n j o u r n a l o f p h a r m a c e u t i c a l s c i e n c e s 3 2 ( 2 0 0 7 ) 357–366

Fig. 2 – LC/MS–MS assay of cyamemazine metabolites. (Top) Sample incubated with human recombinant CYP2C19 and NADPH (top trace: total ion current, TIC; medium traces: metabolites, XIC; bottom trace: parent drug, XIC). (Bottom) Theoretical biotransformation states of cyamemazine with the associated MS/MS transitions.

361

362

e u r o p e a n j o u r n a l o f p h a r m a c e u t i c a l s c i e n c e s 3 2 ( 2 0 0 7 ) 357–366

somes were in the range of values considered satisfactory for inclusion in the experimental procedures envisaged (Table 1). For instance, total P450 content amounted to a median value of 615 pmol/mg of microsomal protein (range of 180–1005 obtained in the laboratory for 83 different human hepatic microsome batches). Furthermore, the results of the enzymatic activity for recombinant enzymes is reported for comparison in the last column of Table 1.

3.2. LC/MS–MS structural characterization of cyamemazine and its metabolites Cyamemazine was detected as a peak with a 5.4 min retention time in LC traces. A standard linear concentration–response curve was obtained by adding 0.5, 10, 15, 20 ␮M of cyamemazine to the reaction medium containing NADPH cofactor. The more sensitive MS–MS transition was 324 > 100 and the more specific ones 324 > 279 or 324 > 237. In the positive ionization mode, cyamemazine appeared as a single ion species: [M + H]+ m/z at 324 mass (Fig. 1, top) whereas in MS–MS conditions, four distinct daughter ions (279, 237, 100 and 58 mass) were detected (Fig. 1, middle). The likely filiation scheme for these ions is depicted in Fig. 1 (bottom). Four transitions from mother to daughter ions (324 > 279, 324 > 237, 324 > 100, 324 > 58) were identified under MRM conditions and these reactions used for identifying cyamemazine in reaction media. Since cyamemazine showed high m/z daughter ions (>230), these ions represent transformations on the tricyclic moiety, while low m/z daughter ions (<100) underlines transformations on the side chain (Figs. 1 and 2). Thus, in MRM conditions, 25 transitions were followed for identifying possible metabolites of cyamemazine. Moreover, a possible oxidation at the C–N linkage was also searched. At the end of the incubation procedure, cyamemazine remained unchanged in reaction solutions lacking either liver microsomes or NADPH cofactor or containing liver microsomes not added with NADPH cofactor. However, in reaction media containing microsomes, NADPH and cyamemazine, the transitions detected revealed the following main metabolic reactions: N-demethylation (transition 310 > 237 or 310 > 86), mono-oxidation (hydroxylation or sulfur oxidation: transition 340 > 253 or 340 > 100), and N,N-di-demethylation (transition 296 > 237 or 296 > 72). A minor secondary pathway is mono-oxidation of N-demethylated cyamemazine (transition 326 > 253 or 326 > 86). Thus, the theoretical biotransformations occurring in the cyamemazine structure, inferred from MS/MS results and literature data on the metabolic degradation of chlorpromazine, a cyamemazine prototype, include N-dealkylation, oxidation to sulfone and monohydroxylation reactions on the benzene ring (Fig. 2). The position of the oxidation reaction on the phenothiazine tricyclic moiety of cyamemazine could not be identified because either mono-hydroxylation derivatives or sulfuroxidation of the thiazine moiety is characterized by the same “+16” parameter. However, a sulfonylation of the thiazine was suggested by the occurrence of a “+32” parameter. If these reactions occur in tandem in the same reaction tube, the resulting compounds can be easily differentiated by

HPLC. No oxidation of the cyamemazine cyano group was revealed.

3.3. Cyamemazine biotransformation by pure recombinant CYP enzymes Cyamemazine did not undergo any chemical change at the end of 60-min incubation in reaction tubes lacking either microsomes or the cofactor NADPH. However, in reaction media containing microsomes endowed with seven of the nine P450 enzymes studied, cyamemazine concentration declined below 85% of the initial nominal concentration, thus indicating a significant biodegradation of cyamemazine. These enzymes can be grouped in strong (>80%: CYP1A2, CYP2C8 and CYP2C19) and moderate (∼25–50%: CYP2B6, CYP2C9, CYP2D6 and CYP3A4) cyamemazine metabolizers. CYP2A6 and CYP2E1 did not degrade cyamemazine (Fig. 3). CYP1A2 transformed approximately 82–85% of the 20 ␮M concentration of cyamemazine incubated into four metabolites with the sulfoxide and N-demethyl species being in the majority. CYPB6 and CYP2C8 catalyzed predominantly the formation of the N-demethyl derivative and, to a minor extent, the sulfoxide and the N,N-di-demethyl compounds. CYP2C9 and CYP2D6 favoured the formation of both mono-oxidized derivatives and, to a minor extent, the N-demethyl sulfoxide. CYP3A4 catalyzed the formation of the four metabolites and primarily the N-mono-demethyl species.

Fig. 3 – Amounts of cyamemazine (%: mean of two replicate measures) transformed by recombinant human liver CYP enzymes incubated in the presence of 20 ␮M cyamemazine and NADPH cofactor for 60 min. In the absence of the cofactor no biotransformation of cyamemazine took place.

e u r o p e a n j o u r n a l o f p h a r m a c e u t i c a l s c i e n c e s 3 2 ( 2 0 0 7 ) 357–366

363

3.4. Cyamemazine biotransformation by CYP enzymes in microsomes from native hepatocytes Cyamemazine at 20 and 200 ␮M was incubated for 20 min with microsomes from pooled human livers in the presence of CYP enzyme-selective substrates or inhibitors. Then, reaction media were analyzed for their content into two main primary metabolites, namely, the N-demethyl (Fig. 4) and the sulfoxide (or hydroxylated, Fig. 5) derivatives of cyamemazine. Both metabolites were quantified as peak heights relative to the peak generated by the parent drug. As commonly done in inhibition studies with CYP enzymes, inhibitions below 10–15% were not considered of biochemical significance. In microsomes from native human hepatocytes (Table 1 details the nature of the competitor and the concentration used) phenacetin inhibited by ∼18–24% cyamemazine (20 ␮M) biotransformation into the N-demethylated metabolite whereas paclitaxel, tolbutamide, ketoconazole and chlorzoxazone reduced moderately (∼37–51%) this process (Fig. 4). However, for paclitaxel the inhibition dropped from 37% to 23% when incubated with a 10-fold higher concentration of cyamemazine (200 ␮M). Nifedipine prevented entirely cyamemazine demethylation by liver microsomes (Fig. 4). As described for the formation of the demethylated derivative, the concentration of cyamemazine (20 and 200 ␮M) studied, played no role in the extent of inhibition of monooxidized metabolite formation in the presence of specific

Fig. 5 – Inhibition of the formation of the S-oxidated metabolite by microsomes prepared from native human hepatocytes incubated with cyamemazine (20 and 200 ␮M) and established competitors of CYP enzymes (see Table 1).

competitors if one excludes the effects of sulfaphenazole which increased by 20–30% in the reaction media containing 200, instead of 20 ␮M, of cyamemazine (Fig. 5). Coumarin, paclitaxel, S-mephenytoin, dextromorphan, quinidine and ketoconazole had no effect on the formation of the putative sulfoxide metabolite whereas tolbutamide and sulfaphenazole were weak inhibitors (∼26–28%) with the latter agent lacking significant effects when added to reaction media containing a low concentration of cyamemazine (Fig. 5). Phenacetin and furafylline inhibited moderately (∼37–47%) and nifedipine fully the formation of sulfoxide or hydroxylated metabolites of cyamemazine (Fig. 5).

4.

Fig. 4 – Inhibition of the formation of N-demethylated cyamemazine in microsomes prepared from native human hepatocytes incubated with 20 and 200 ␮M cyamemazine and established competitors of CYP enzymes (see Table 1).

Discussion

The main result from this investigation is that cyamemazine can function as a substrate of at least five CYP enzymes as depicted in Fig. 6. Cytochrome P450 (CYP) enzymes are essentially located in hepatic microsomes and mediate phase I xenobiotic metabolism degrading drugs into chemically functionalized derivatives which are eliminated from the body as such or as hydrophilic conjugated (e.g. glucuronide or sulphate) products (phase II metabolism). Hence, the elucidation of the role of CYP enzymes in the metabolic fate of cyamemazine allows not only to identify possible metabolites endowed of biological activity but also to predict possible undesirable metabolic drug–drug

364

e u r o p e a n j o u r n a l o f p h a r m a c e u t i c a l s c i e n c e s 3 2 ( 2 0 0 7 ) 357–366

Fig. 6 – CYP enzymes mediated metabolic transformation of cyamemazine.

interaction that may occur in patients exposed to concurrently administered therapeutic agents sharing CYP isoenzyme catalyzed degradation processes. The strategy used in this investigation to phenotype human liver CYP enzymes intervening in cyamemazine biotransformation consisted in firstly determining which of these enzymes prepared in a pure form by recombinant and expression procedures biotransformed cyamemazine. Then, cyamemazine was incubated with native human liver microsomes in the presence of relatively specific competitors of selected P450 enzymes in order to assess the importance of each enzyme in cyamemazine biotransformation. CYP1A2 has been demonstrated to play a major role in the N-demethylation and S-oxidation (or hydroxylation) degradation processes of 5% of drugs used in human therapeutics (Rendicand and Di Carlo, 1997) which, like cyamemazine, are characterized by rigid, flat and hydrophobic polycyclic aromatic hydrocarbons or aromatic amines. This enzyme degraded extensively (82–85%) cyamemazine into four distinct metabolites. However, in native microsomes from human hepatocytes in which the metabolic capacity of CYP1A2 was either inhibited (furafylline) or saturated (phenacetin), there was only a reduction (12–18% and 37–46%, respectively) of N-demethylated and S-oxidized cyamemazine formation confirming that the native human liver CYP1A2 biodegrades cyamemazine. CYP2A6, a N-demethylating enzyme for 2% of drugs (Rendicand and Di Carlo, 1997), does not appear to catalyze cyamemazine biotransformation to any significant extent since only traces of expected cyamemazine metabolites were detected in reaction media containing recombinant CYP2A6 or native human liver microsomes exposed to coumarin, a specific substrate of CYP2A6. CYP2B6 is an inducible rather than a constitutive enzyme in human liver (Rendicand and Di Carlo, 1997). Recombinant CYP2B6 biotransformed cyamemazine moderately (32–40%) into a N-demethylated derivative. However, the possible role of this metabolic pathway in native microsomes from human liver could not be further assessed due to the lack of established specific CYP2B6 competitors.

CYP2C8 has been recently shown to play a major role in the biotransformation of the anticancer agent paclitaxel and determine also the metabolic fate of 1–5% of available drugs (Cresteil et al., 1994). This enzyme degraded 82–86% of incubated cyamemazine into four metabolites of which the N-demethylated species predominated. The formation of the N-demethylated, but not the S-oxidized, derivative was substantially reduced (23–37%) by preincubating native human liver microsomes with paclitaxel. CYP2C9, a close homologue of CYP2C8, intervenes in the biotransformation of approximately 10% of pharmaceuticals, particularly non-steroidal antiinflammatory and antidiabetic agents (Rendicand and Di Carlo, 1997). CYP2C9 biotransformed 44–48% of incubated cyamemazine into N-demethylated and sulfoxide derivatives. Tolbutamide, a CYP2C9 substrate, inhibited the demethylation process by 42%, whereas the CYP2C9 inhibitor sulfaphenazole did it by only 10%. Additionally, tolbutamide was more potent (28%) than sulfaphenazole (17%) to curb the S-oxidation reaction. The differential effects obtained by using a substrate and an inhibitor of CYP2C9 may be satisfactorily accounted for if one considers that tolbutamide, which, in this experiment, was used in a relatively high concentration (1 mM) for saturating CYP2C9 enzymatic capacity, is also a concomitant substrate of CYP2C8 enzyme. Furthermore, the role of CYP2C9 in cyamemazine biotransformation may be limited to the formation of the S-oxide species which is quantitatively expressed by the results obtained with sulfaphenazole. CYP2C19 is weakly expressed in the human liver but is an inducible metabolizing enzyme which intervenes in the degradation of approximately 2% of drugs and, in particular, antipsychotics and proton pump inhibitors (Rendicand and Di Carlo, 1997). CYP2C19 almost completely biotransformed cyamemazine into N-demethylated and S-oxidized derivatives. However, S-mephenytoin, a CYP219 substrate, did not reduce the degradation process of cyamemazine by native human liver microsome preparations. The minor expression and role of CYP2C19 in human liver may provide a satisfactory explanation for the different findings using recombinant and native CYP2C19. One could also raise the possibility that this enzyme catalyze the formation of a metabolite distinct from N-demethylated and S-oxidized cyamemazine. Further studies are necessary to clarify this issue since CYP2C19 is a polymorphic enzyme showing variable interindividual expression and thus could be the object of a potential site of drug–drug interaction. CYP2D6 intervenes in the metabolism of approximately 30% of human pharmaceuticals of large prescription (antiarrhythmics, antihypertensives, antidepressant and neuroleptics). In recent years, this enzyme received substantial attention since it undergoes polymorphic expression (Rendicand and Di Carlo, 1997). CYP2D6 biotransformed cyamemazine to a limited extent (33–36%) into N-demethylated and S-oxide metabolites. However, in microsomes from human hepatocytes the biotransformation of cyamemazine was not impaired by dextromorphan or quinidine, that are, respectively, a substrate and an inhibitor of CYP2D6. Thus, CYP2D6 is unlikely to intervene in the metabolic degradation of cyamemazine under in vivo conditions.

e u r o p e a n j o u r n a l o f p h a r m a c e u t i c a l s c i e n c e s 3 2 ( 2 0 0 7 ) 357–366

CYP2E1, which metabolizes ∼2% of small molecules and solvents, when studied with the recombinant form, did not degrade cyamemazine. Surprisingly, chlorzoxazone, a substrate of this enzyme, reduced by approximately 50% the formation of both the N-demethylated and S-oxidized cyamemazine. This result is probably due to the high concentration (500 ␮M) of chlorzoxazone used which may have functioned as a substrate of CYP2E1 as well as other P450 enzymes intervening in the metabolic transformation of cyamemazine. CYP3A4 operates the biotransformation of approximately 50% of clinically used therapeutic agents and has been often found to be responsible for metabolic drug–drug interactions. CYP3A4 biotransformed 50% of cyamemazine into the N-demethylated metabolite. In microsomes from human hepatocytes, ketoconazole, a specific inhibitor of this enzyme, significantly reduced (24–43%) the N-demethylation, but not the S-oxidation, reaction. Thus, CPY3A4 mediates exclusively the N-demethylation reaction of cyamemazine. However, nifedipine, a substrate of CYP3A4, prevented entirely the biotransformation of cyamemazine. This result suggests that nifedipine, in the concentration used in these experiments, functioned as a saturating metabolic substrate of other P450 enzymes degrading cyamemazine. Antipsychotic agents belong to chemical and pharmacological classes with recognized proclivity to adversely affect ion channels shaping the action potential of cardiac myocytes and trigger torsades de pointes (Cavero and Crumb, 2005). However, cyamemazine has been associated with only a single case of cardiac arrhythmia since 1974, date of its introduction in the French therapeutic market, with more than 5.7 millions days of treatment (Gury et al., 2000). This clinical finding was supported by a recent study (Crumb et al., 2006) showing that therapeutical concentrations of cyamemazine do not inhibit hERG current and, in contrast to terfenadine, cyamemazine does not delay cardiac repolarization in the anesthetized guinea pig. The excellent cardiac tolerance of cyamemazine can be explained by the fact that if associated pharmaceuticals share one or more of the P450 enzymes operating the biodegradation of cyamemazine, they are unlikely to increase the plasma concentration of cyamemazine and thus cause possible adverse effects linked to supratherapeutic concentration of this antipsychotic drug. This hypothesis deserves confirmation by further studies testing the effect of antipsychotic drugs (including cyamemazine) on cyamemazine degradation by the specific substrate-based CYP enzymes. In conclusion, the main message from this investigation is that cyamemazine can function as a substrate of at least five CYP enzymes (Fig. 6). This finding has important clinical implications for those patients who, to be efficaciously treated, require polypharmacy interventions, i.e., associated pharmaceuticals are unlikely to increase plasma cyamemazine concentrations. The relative amount of cyamemazine biotransformed by each individual CYP enzyme depends on the dose of cyamemazine used and remains to be determined. This could be achieved by measuring the intrinsic clearance of cyamemazine by each concerned CYP coenzyme.

365

Acknowledgement We are indebted to Icilio Cavero (Lucca, Italia) for helping with the manuscript.

references

Bourin, M., Dailly, E., Hascoet, M., 2004. Preclinical and clinical pharmacology of cyamemazine: anxiolytic effects and prevention of alcohol and benzodiazepine withdrawal syndrome. CNS Drug Rev. 10, 219–229. Bradford, M.A., 1976. Rapid and sensitive method for the quantification of microgram quantities of protein utilizing the principle of protein–dye binding. Anal. Biochem. 22, 248–256. Busby, W.F.J., Ackermann, J.M., Crespi, C.L., 1999. Effect of methanol, ethanol, dimethyl sulfoxide, and acetonitrile on in vitro activities of cDNA-expressed human cytochromes P-450. Drug Metab. Dispos. 27, 246–249. Caveroand, I., Crumb, W., 2005. ICH S7B draft guideline on the non-clinical strategy for testing delayed cardiac repolarisation risk of drugs: a critical analysis. Expert Opin. Drug Saf. 4, 509–530. Clarke, S.E., 1998. In vitro assessment of human cytochrome P450. Xenobiotica 12. Crespi, C.L., 1998. Effect of salt concentration on the activity of liver microsomal and cDNA-expressed human cytochromes P450. In: Proceedings of 12th International Symposium on micrososmes and drug oxidation (ISMDO), Montpellier, France. Cresteil, T., Monarrat, B., Alvinerie, P., Treluyer, J.M., Viera, I., Wright, M., 1994. Taxol metabolism by human liver microsomes: identification f cytochrome P450 isoenzymes involved in its biotransformation. Cancer Res. 54, 386–392. Crumb, W., Llorca, P.-M., Lancon, C., Thomas, G.P., Garay, R.P., Hameg, A., 2006. Effects of cyamemazine on hERG, INa, ICa, Ito, Isus and IK1 channel currents, and on the QTc interval in guinea pigs. Eur. J. Pharmacol. 532, 270–278. Dekant, W., Frischmann, C., Speerschneider, P., 1995. Sex, organ and species specific bioactivation of chloromethane by cytochrome P450 2E1. Xenobiotica 25, 1259–1265. Distlerath, L., Reilly, P.E., Martin, M.V., Davis, G.G., Wilkinson, G.R., Guengerich, F., 1985. Purification and characterization of the human liver cytochromes P-450 involved in debrisoquine 4-hydroxylation and phenacetin O-deethylation, two prototypes for genetic polymorphism in oxidative drug metabolism. J. Biol. Chem. 260, 9057–9065. Greenleeand, W., Poland, A., 1978. An improved assay of 7-ethoxycoumarin O-deethylase activity: induction of hepatic enzyme activity in C57BL/6J and DBA/2J mice by phenobarbital, 3-methylcholanthrene and 2,3,7,8-tetrachlordibeno-p-dioxin. J. Pharmacol. Exp. Ther. 205, 596–605. Guengerich, P.P., Martin, M.V., Beaune, P.H., Kremers, P., Wolf, T., Waxman, D.J., 1986. Characterization of rat and human liver microsomal cytochrome P-450 forms involved in nifedipine oxidation, a prototype for genetic polymorphism in oxidative drug metabolism. J. Biol. Chem. 261, 5051–5060. ´ Gury, C., Canceil, O., Iaria, P., 2000. Antipsychotiques et securit e´ ´ actuelles sur les allongements de cardio-vasculaire: donnees l’intervalle QT et le risque d’arythmies ventriculaires. ´ L’Encephale 26, 62–72. Gury, C., Fabre, C., Hameg, A., Simo, E., Surugue, J., Garay, R.P., Welcomme, N., 2006. Prescription d’antipsychotiques en

366

e u r o p e a n j o u r n a l o f p h a r m a c e u t i c a l s c i e n c e s 3 2 ( 2 0 0 7 ) 357–366

´ ´ l’Information Psychiatrique 82, milieu hospitalier specialis e. 503–509. Miners, J.O., Smith, K.J., Robson, R.A., Mac Manus, M.E., Veronese, M.E., Birkett, D.J., 1988. Tolbutamide hydroxylation by human liver microsomes. Biochem. Pharmacol. 37, 1137–1144. Omuraand, T., Sato, R., 1964. The carbon monoxide-binding pigment of liver microsomes. J. Biol. Chem. 239, 2370–2378.

¨ ¨ J., Taavitssainen, P., Rautio, A., Raunio, H., Pelkonen, O., Maenp a¨ a, 1998. Inhibition and induction of human cytochrome P450 (CYP) enzymes. Xenobiotica 12, 1203–1253. Rendicand, S., Di Carlo, F.G., 1997. Human cytochrome P450 enzymes: a status report summarizing their reactions, substrates, inducers and inhibitors. Drug Metab. Rev. 29, 413–580.