The metabolism of the piperazine-type phenothiazine neuroleptic perazine by the human cytochrome P-450 isoenzymes

European Neuropsychopharmacology 14 (2004) 199 – 208 www.elsevier.com/locate/euroneuro

The metabolism of the piperazine-type phenothiazine neuroleptic perazine by the human cytochrome P-450 isoenzymes Jacek Wo´jcikowski a,*, Lydiane Pichard-Garcia b, Patrick Maurel b, Wladyslawa A. Daniel a a

b

Polish Academy of Sciences, Institute of Pharmacology, Sme˛tna 12, 31-343, Cracow, Poland Institute National de la Sante´ et de la Recherche Me´dicale (INSERM U128), CNRS, 1919 Route de Mende, 34293 Montpellier, France Received 13 May 2003; received in revised form 3 July 2003; accepted 8 July 2003

Abstract Identification of cytochrome P-450 isoenzymes (CYPs) involved in perazine 5-sulphoxidation and N-demethylation was carried out using human liver microsomes and cDNA-expressed human CYPs (Supersomes). In human liver microsomes, the formation of perazine metabolites correlated significantly with the level of CYP1A2 and ethoxyrezorufin O-deethylase activity, as well as with the level of CYP3A4 and cyclosporin A oxidase activity. Moreover, the formation of N-desmethylperazine also correlated well with S-mephenytoin 4V-hydroxylase activity (CYP2C19). a-Naphthoflavone (a CYP1A2 inhibitor) and ketoconazole (a CYP3A4 inhibitor) significantly decreased the rate of perazine 5-sulphoxidation, while ticlopidine (a CYP2C19 inhibitor) strongly reduced the rate of perazine N-demethylation in human liver microsomes. The cDNA-expressed human CYPs generated different amounts of perazine metabolites, but the preference of CYP isoforms to catalyze perazine metabolism was as follows (pmol of product/pmol of CYP isoform/min): 1A1>2D6>2C19>1A2>2B6>2E1>2A6c 3A4>2C9 for 5-sulphoxidation and 2C19>2D6>1A1>1A2>2B6>3A4>2C9>2A6 for N-demethylation. In the light of the obtained results and regarding the contribution of each isoform to the total amount of CYP in human liver, it is concluded that CYP1A2 and CYP3A4 are the main isoenzymes catalyzing 5-sulphoxidation (32% and 30%, respectively), while CYP2C19 is the main isoform catalyzing perazine N-demethylation (68%). CYP2C9, CYP2E1 CYP2C19 and CYP2D6 are engaged to a lesser degree in 5sulphoxidation, while CYP1A2, CYP3A4 and CYP2D6 in perazine N-demethylation (6 – 10%, depending on the isoform). D 2003 Elsevier B.V./ECNP. All rights reserved. Keywords: Perazine; Metabolism; Human CYP; Microsomes; Supersomes; Specific inhibitors

1. Introduction Perazine belongs to a group of phenothiazine neuroleptics with the piperazine structure in a side chain. It is a moderate antagonist of dopaminergic D2 receptors and a weak antagonist of dopaminergic D1, adrenergic a1, serotonergic 5-HT2 and cholinergic muscarinic M1 receptors, hence it rarely produces side-effects in the central or Abbreviations: CHLRZ, chlorzoxazone 6-hydroxylation; COUM, coumarin 7-hydroxylation; CsA, cyclosporin A oxidation; CYP, cytochrome P-450; DDC, diethyldithiocarbamic acid; EROD, ethoxyresorufin O-deethylation; HPLC, high performance liquid chromatography; Km, the Michaelis constant; KET, ketoconazole; NAPH, a-naphthoflavone; NMEPH, S-mephenytoin N-demethylation; OH-MEPH, S-mephenytoin 4Vhydroxylation; QUIN, quinidine; SULF, sulfaphenazole; TICLOP, ticlopidine; TOLB, tolbutamide 4-methylhydroxylation; Vmax, maximum velocity of the reaction. * Corresponding author. Tel.: +48-12-6374022; fax: +48-12-6374500. E-mail address: [email protected] (J. Wo´jcikowski). 0924-977X/$ - see front matter D 2003 Elsevier B.V./ECNP. All rights reserved. doi:10.1016/S0924-977X(03)00105-6

autonomic nervous system. Moreover, unlike many other phenothiazine neuroleptics, perazine does not negatively influence mood, and some clinicians even attribute certain antidepressant properties to it. For this reason perazine is often used in geriatric patients and in a combination therapy with antidepressants (Nelson, 1993; Keck et al., 1994). In both man and rats, perazine is metabolized by sulphoxidation in the thiazine ring and by N-demethylation in the piperazine side chain (Fig. 1), as well as by aromatic hydroxylation in position 3, and by N-oxidation and degradation of the piperazine ring (Breyer, 1969, 1972; Kaning and Breyer, 1969). The main metabolites identified in human blood plasma or serum and urine are perazine 5sulphoxide, N-desmethylperazine, perazine N-oxide and 3hydroxyperazine (Breyer, 1969; Breyer and Villumsen, 1976; Kaning and Breyer, 1969; Rao, 1989). These metabolites are also formed in vitro in liver microsomes of the rat, rabbit, guinea-pig, pig and cat (Breyer, 1971; Daniel et al.,

200

J. Wo´jcikowski et al. / European Neuropsychopharmacology 14 (2004) 199–208

Fig. 1. Metabolic pathways of perazine.

2001, 2002). Moreover, perazine 5-sulphoxide and N-desmethylperazine are found in rat blood plasma and different tissues (Breyer, 1972; Daniel et al., 2000, 2001; Wo´jcikowski and Daniel, 2000). In vitro studies conducted by Daniel et al. (2002) on rat liver microsomes demonstrated that 5-sulphoxidation of the simplest and aliphatic-type phenothiazine neuroleptic promazine was catalyzed by CYP2D, while its N-demethylation was mediated by CYP2B and CYP2D. For comparison, our recent results (Wo´jcikowski et al., 2003) indicate that in human liver CYP1A2 and CYP3A4 are the main isoenzymes responsible for 5-sulphoxidation, while CYP1A2 and CYP2C19 are the key isoenzymes catalyzing promazine Ndemethylation. Metabolic studies with liver microsomes indicated that CYP3A was mainly involved in chlorpromazine 5-sulphoxidation in humans, while CYP2D in Ndemethylation in rats (Cashman et al., 1993; Valoti et al., 1998). Using human liver microsomes and cDNA-expressed human cytochrome P-450s (CYPs), Olesen and Linnet (2000) showed that N-dealkylation of the piperazine-type phenothiazine neuroleptic perphenazine was catalyzed mainly by CYP1A2, CYP3A4 and CYP2C19. According to Daniel et al. (2002), 5-sulphoxidation of another piperazine-type phenothiazine neuroleptic perazine in the rat was catalyzed by CYP2B and CYP2D, while N-demethylation was mediated by CYP1A2, CYP2B and CYP2D. Recent

studies by Sto¨rmer et al. (2000a), conducted on human liver microsomes and cDNA-expressed human CYPs, showed that CYP3A4 and CYP2C9 were the basic isoenzymes catalyzing perazine N-demethylation, while CYP1A2, CYP2D6 and CYP2C19 contributed to a lesser degree to that reaction. However, the latter authors did not investigate the 5-sulphoxidation process. Since the above data indicate some structure (side chain) and species differences in the CYP catalysis of phenothiazine neuroleptics, the aim of the present study was to concurrently investigate the contribution of human CYPs to perazine 5-sulphoxidation and N-demethylation using three complementary in vitro models. The results obtained in this study are compared with the findings of analogous experiments with other phenothiazine neuroleptics and are discussed in respect of the structure and species differences in the enzymatic catalysis of the metabolism of phenothiazines.

2. Materials and methods 2.1. Products Perazine (dimaleate) was obtained from Labor (Wroclaw, Poland). Perazine 5-sulphoxide and N-desmethylperazine

J. Wo´jcikowski et al. / European Neuropsychopharmacology 14 (2004) 199–208

were synthesized according to a previously described method (Daniel et al., 1998). Quinidine, a-naphtoflavone, sulfaphenazole, ketoconazole, diethyldithiocarbamic acid (DDC) and ticlopidine were purchased from Sigma (St. Louis, USA). NADPH came from Boehringer (Mannheim, Germany). Microsomes of patients MG43 and MG89 were purchased from BD Gentest (Woburn, MA, USA). Bovine serum albumin was obtained from Pierce Chemical (Rockford, USA). All the organic solvents with high performance liquid chromatography (HPLC) purity were supplied by Carlo Erba (Milan, Italy). 2.2. Human liver samples The use of human liver samples for scientific purposes was approved by the French National Ethics Committee. Human liver specimens were obtained from organ donors and patients undergoing hepatic lobectomies. The clinical characteristics of the donor and patients are presented in Table 1. 2.3. Liver microsomes Microsomes were prepared from liver samples by differential centrifugation, and were stored as described previously (Diaz et al., 1990). Protein concentration was determined by a bicinconinic acid method according to the protocol provided by the manufacturer (Pierce Chemical). The liver microsomes from patients FH61289, FT95 and FT100 were used to optimize the conditions of perazine metabolism. On the basis of the obtained results, perazine metabolism in liver microsomes was studied in respect of the linear dependence of product formation on the time and concentrations of protein and substrate. Microsomal protein, 500 Ag, was resuspended in 500 Al of 20 mM Tris/HCl buffer (pH = 7.4). To determine kinetic parameters of the enzyme, the perazine concentrations

Table 1 Clinical characteristics of patients and liver specimen donors Patient identification

Age (years)

Gender

Diagnosis

FT43 FT82 FT84 FT85 FT86 FT87 FT92 FT95 FT99 FT100 FT101 FH61289 HG43 HG89

61 68 43 24 58

female male male female male

30

male female male male female male female female

metastasis from colon cancer metastasis from colon cancer hepatocarcinoma in normal liver adenoma metastasis from colon cancer metastasis from colon cancer metastasis from colon cancer metastasis from colon cancer metastasis from colon cancer metastasis from colon cancer metastasis from colon cancer cerebral hemorrhage open head trauma intracranial hemorrhage

27 67 45 60 23 71

201

used ranged from 25 to 400 AM. To study maximum ability of individual patients to metabolise the neuroleptic, 750 AM perazine (3  Km) was used. For inhibition studies, 25 AM perazine was chosen as a therapeutic concentration in the liver, which did not saturate the enzyme. Perazine was incubated with liver microsomes and the specific CYP inhibitors: 2 AM a-naphthoflavone (a CYP1A2 inhibitor), 10 AM sulfaphenazole (a CYP2C9 inhibitor), 5 AM ticlopidine (a CYP2C19 inhibitor), 10 AM quinidine (a CYP2D6 inhibitor), 200 AM DDC (a CYP2A6 + CYP2E1 inhibitor) and 2 AM ketoconazole (a CYP3A4 inhibitor). After 3-min preincubation at 37 jC, the reaction was initiated by adding NADPH to a final concentration of 1 mM. After 15-min incubation, the reaction was stopped by adding 200 Al of methanol. Perazine and its metabolites were analysed by the HPLC method as described below. 2.4. Correlation analysis of the data The rates of perazine 5-sulphoxidation and N-demethylation were correlated with the rates of CYP-specific reactions (ethoxyresorufin O-deethylation, coumarin 7-hydroxylation, S-mephenytoin N-demethylation, tolbutamide 4-methylhydroxylation, S-mephenytoin 4V-hydroxylation, chlorzoxazone 6-hydroxylation, cyclosporin A oxidation) and the level of CYPs (CYP1A2, CYP2A6, CYP2B6, CYP2C9, CYP2D6, CYP2E1, CYP3A4) obtained by a western blot for each preparation of liver microsomes. Each pair of the data was compared by a simple linear regression analysis using the statistical programme Prism 2.01. Monooxygenase activities and the level of CYPs were determined as described previously (Daujat et al., 1987; Diaz et al., 1990; Maurice et al., 1992; Pichard et al., 1990). 2.5. cDNA-expressed human CYPs Microsomes from baculovirus-infected insect cells expressing CYP1A1, CYP1A2, CYP2A6, CYP2B6, CYP2C9, CYP2C19, CYP2D6, CYP2E1 and CYP3A4 (Supersomes) were obtained from Gentest. All the Supersomes were coexpressed with human P-450 reductase. CYP2E1 was additionally coexpressed with the human cytochrome b-5. Perazine metabolism was studied under experimental conditions similar to those described for liver microsomes, except for the fact that the concentration of perazine was 750 AM, the incubation time was 2 h, and the final concentration of CYPs was 100 pmol ml 1. Study into perazine metabolism in Supersomes was carried out at the neuroleptic concentration of 750 AM (3  Km) allowing to reach the velocity of reaction of about Vmax to show the maximum ability of cDNAexpressed enzyme to metabolize perazine. Perazine and its metabolites were analysed by the HPLC method described below.

J. Wo´jcikowski et al. / European Neuropsychopharmacology 14 (2004) 199–208

202

Table 2 Kinetic parameters of perazine 5-sulphoxidation and N-demethylation in human liver microsomes (Lineweaver – Burk analysis) Patients

FH61289 FT95 FT100 Mean F S.D.

Perazine 5-sulfoxidation

Perazine N-demethylation

Km [AM]

Vmax [nmol/mg protein/min]

Km [AM]

Vmax [nmol/mg protein/min]

291 229 297 272 F 38

0.62 0.58 0.28 0.49 F 0.18

259 194 242 232 F 34

0.36 0.34 0.39 0.36 F 0.02

2.6. Determination of perazine and its metabolites in the incubation medium Perazine and its metabolites were quantified using the previously described HPLC method (Daniel et al., 1998).

After incubation, the samples were centrifuged at 2000  g for 10 min. The water phase (pH = 12) containing perazine and its metabolites was extracted with hexane and dichloromethane (1:1, v/v). The residue obtained after evaporation of the microsomal extracts was dissolved in 100 Al of the mobile phase described below. An aliquot of 50 Al was injected into the HPLC system. The concentrations of perazine and its main metabolites (perazine 5-sulphoxide, N-desmethylperazine) were assayed using a Varian HPLC system with UV detection. The analytical column (Econosphere C18, 5 Am, 4.6  250 mm) was purchased from Alltech (Carnforth, England). The mobile phase consisted of an acetate buffer, pH = 3.4 (100 mmol of ammonium acetate, 20 mmol of citric acid and 1 ml of triethylamine in 1000 ml of the buffer adjusted to pH = 3.4 with an 85% phosphoric acid), and acetonitrile in a proportion of 50:50. Elution proceeded at an ambient temperature at a flow

Fig. 2. Formation of perazine 5-sulphoxide (A) and N-desmethylperazine (B) from perazine. Human liver microsomes of patient FT95 (1 mg of protein/ml) were incubated in a 20 mM Tris/HCl buffer (pH = 7.4) with perazine (50 – 400 AM) and NADPH (1 mM) for 15 min. The insets show Eadie – Hofstee plots.

J. Wo´jcikowski et al. / European Neuropsychopharmacology 14 (2004) 199–208

rate of 1.2 ml/min. The absorbance of perazine and its metabolites was measured at a wavelength of 265 nm. The compounds were eluted in the following order: perazine 5-sulphoxide (6.08 min), N-desmethylperazine (7.09 min), perazine (11.97 min). The sensitivity of the method allowed for the quantification of levels as low as 0.043 nmol of perazine 5-sulphoxihe, 0.013 nmol of Ndesmethylperazine and 0.04 nmol of perazine in 1 ml of the microsomal suspension.

3. Results 3.1. Kinetics of perazine metabolism in human liver microsomes The Lineweaver –Burk plots were linear (R2 = 0.936 – 0.995), which allowed for graphic estimation of the Km and Vmax values for perazine 5-sulphoxidation and Ndemethylation in three different human liver microsomes (Table 2). However, Fig. 2A and B shows the representative Eadie – Hofstee plots for perazine 5-sulphoxidation and N-demethylation in liver microsomes of patient FT95. These plots suggest that multiple enzymes are responsible for the biotransformation of perazine via 5-sulphoxidation and N-desmethylation. Similar results were obtained with microsomes of patients FT99 and FT100. Therefore, kinetic parameters shown in Table 2 are resultants of Km and Vmax values of individual enzymes contributing to the processes studied.

203

3.2. Correlation study Thirteen different preparations of human liver microsomes were used to evaluate the interindividual variability of perazine biotransformation. The microsome preparations yielded different amounts of perazine 5sulphoxide and N-desmethylperazine (Fig. 3). As a rule, perazine 5-sulphoxide was produced in greater amounts compared to N-desmethylperazine. In two preparations (FT84 and FT87), N-desmethylperazine was produced in slightly greater amounts. Mean velocities of the reactions were 243 F 45 pmol/mg protein/min for 5sulphoxidation, and 143 F 88 pmol/mg protein/min for N-demethylation. Microsomes from the bank, used in this study, were tested for the levels of several CYPs (determined by immunoblotting) and for several monooxygenase activities (data not shown). The rate of formation of perazine metabolites was compared with CYP levels and monooxygenase activities. The results of such analyses are shown in Table 3, where the correlation coefficient (r) and the p value are quoted for each pair of data. The formation of perazine 5-sulphoxide correlated significantly with the level of CYP1A2 and with ethoxyrezorufin O-deethylase activity, as well as with the level of CYP3A4 and with cyclosporin A oxidase activity. The production of N-desmethylperazine showed a week correlation with the level of CYP1A2 and with ethoxyrezorufin O-deethylase activity, as well as with the level of CYP3A4 and with cyclosporin A oxidase activity. Moreover, the formation of N-desmethylperazine corre-

Fig. 3. Interindividual variability of perazine metabolism in human liver microsomes. Human liver microsomes (1 mg of protein/ml) were incubated in a 20 mM Tris/HCl buffer (pH = 7.4) with perazine (750 AM) and NADPH (1 mM) for 15 min.

J. Wo´jcikowski et al. / European Neuropsychopharmacology 14 (2004) 199–208

204

Table 3 A correlation (r and p values) between the rate of perazine 5-sulphoxidation and N-demethylation and the velocity of CYP specific reactions or the level of CYPs in human liver microsomes

Perazine 5-sulphoxide r p N-desmethylperazine r p

EROD 1A2

COUM

2A6

N-MEPH 2B6

0.688 0.009 0.566 0.044

0.492 0.088 0.016 0.957

0.418 0.155 0.136 0.657

0.234 0.442 0.423 0.150

0.667 0.013 0.570 0.043

0.269 0.374 0.384 0.195

TOLB 0.337 0.260 0.360 0.227

2C9

OH-MEPH

2D6

CHLRZ 2E1

CsA

3A4

0.108 0.724 0.284 0.347

0.087 0.776 0.577 0.039

0.101 0.744 0.323 0.281

0.502 0.080 0.324 0.280

0.640 0.018 0.528 0.049

0.794 0.0012 0.558 0.047

0.319 0.288 0.079 0.797

EROD—ethoxyresorufin O-deethylation, COUM—coumarin 7-hydroxylation, N-MEPH—S-mephenytoin N-demethylation, TOLB—tolbutamide 4methylhydroxylation, OH-MEPH—S-mephenytoin 4V-hydroxylation, CHLRZ—chlorzoxazone 6-hydroxylation, CsA—cyclosporin A oxidation. Each pair of data was compared by a simple linear regression analysis using the statistical programme Prism 2.01.

lated well with S-mephenytoin 4V-hydroxylase activity. No correlation was observed between the production of perazine metabolites and the levels or activities of other CYPs.

inhibitor) and quinidine (a CYP2D6 inhibitor) had no inhibitory effect on the rate of perazine 5-sulphoxidation and N-demethylation. 3.4. Study with cDNA-expressed human CYPs

3.3. Inhibition of perazine metabolism by specific CYP inhibitors a-Naphthoflavone (a CYP1A2 inhibitor) and ketoconazole (a CYP3A4 inhibitor) significantly reduced the rate of perazine 5-sulphoxidation up to 75 and 65% of the control value, respectively (Fig. 4). At the same time, ticlopidine (a CYP2C19 inhibitor) exerted a strong inhibitory effect (up to 48% of the control value) on the rate of perazine N-demethylation (Fig. 4). DDC (a CYP2A6 + CYP2E1 inhibitor), sulfaphenazol (a CYP2C9

The ability of cDNA-expressed human CYPs to metabolize perazine is shown in Fig. 5A and B. The preference of CYP isoforms for catalyzing perazine metabolism was as follows: 1A1>2D6>2C19>1A2>2B6>2E1>2A6 c 3A4> 2C9 for 5-sulphoxidation, and 2C19>2D6>1A1>1A2> 2B6>3A4>2C9>2A6 for N-demethylation. Perazine 5sulphoxide was generated in a ca. 1.5 – 5-fold greater amount compared to N-desmethylperazine by most of the CYPs studied. In contrast, CYP2C19 formed N-desmethylperazine in an amount ca. 6-fold greater than that of

Fig. 4. Effect of CYP-specific inhibitors on the rate of perazine 5-sulphoxidation and N-demethylation in pooled human liver microsomes (MG43 and MG89). Human liver microsomes were incubated with 25 AM perazine, in the absence (control) or presence of CYP-specific inhibitors: 2 AM a-naphthoflavone (NAPH), 200 AM diethyldithiocarbamic acid (DDC), 10 AM sulfaphenazole (SULF), 5 AM ticlopidine (TICLOP), 10 AM quinidine (QUIN), and 2 AM ketoconazole (KET). Absolute control values were 48 F 6 pmol of perazine 5-sulphoxide/mg protein/min and 46 F 3 pmol of N-desmethylperazine/mg protein/ min. Mean values F S.D. (n = 5) are presented. Statistical significance was assessed using Student’s t-test and indicated with **p < 0.01, ***p < 0.001. For further explanation, see Fig. 3.

J. Wo´jcikowski et al. / European Neuropsychopharmacology 14 (2004) 199–208

205

4. Discussion

Fig. 5. Biotransformation of perazine via 5-sulphoxidation (A) and Ndemethylation (B) by the cDNA-expressed human CYPs (Supersomes). Perazine (750 AM) was incubated with Supersomes (100 pmol of CYP ml 1) and NADPH (1 mM) for 2 h.

perazine 5-sulphoxide. Both those metabolites were formed in similar amounts by CYP2D6. N-desmethylperazine was not formed by CYP2E1.

The results presented above indicate major contribution of CYP1A2 and CYP3A4 to perazine 5-sulphoxidation, as well as of CYP2C19 to its N-demethylation. CYP2C9, CYP2C19, CYP2D6 and CYP2E1 contribute to a lesser degree to 5-sulphoxidation, and so do CYP1A2, CYP3A4 and CYP2D6 to N-demethylation of perazine. This conclusion is based on the findings of the three experimental models used, the results of which are consistent. In the bank of human liver microsomes, the formation of perazine 5sulphoxide and N-desmethylperazine significantly correlated with the level of CYP1A2 and ethoxyresorufin Odeethylase activity, as well as with the level of CYP3A4, and cyclosporin A oxidase activity. Moreover, perazine Ndemethylation correlated positively with the activity of Smephenytoin 4V-hydroxylase (CYP2C19). The production of perazine 5-sulphoxide was significantly reduced by a-naphthoflavone (a CYP1A2 inhibitor) and ketoconazole (a CYP3A4 inhibitor), while the formation of N-desmethylperazine was strongly decreased by ticlopidine (a CYP2C19 inhibitor). Of the cDNA-expressed human CYPs studied, the highest rates of perazine metabolism were found for CYP1A (for 5-sulphoxidation), and for CYP2C19 and CYP2D6 (for 5-sulphoxidation and N-demethylation). Considering the relative amount of each isoform in the total CYP content in human liver, the role of CYP1A2, CYP3A4 and CYP2C19 seems to be predominant in perazine metabolism (Tables 4 and 5). The results obtained with cDNA-expressed human CYPs have shown that all the isoenzymes tested (except for CYP2E1) generate detectable, but different amounts of perazine 5-sulphoxide and N-desmethylperazine. These findings suggest a non-specific catalysis of perazine, which is consistent with the multienzyme Eadie – Hofstee plots derived from liver microsomes (Fig. 2A and B). However,

Table 4 Estimation of the contribution of CYP isoforms to perazine 5-sulphoxidation in liver microsomes on the basis of the rate of this reaction in Supersomes CYPs

Velocity in Supersomes [pmol perazine 5-sulphoxide/pmol CYP isoform/min]

Relative contribution of isoforms to the total CYP content in liver microsomes [fraction]*

Predicted velocity in liver microsomes [pmol perazine 5-sulphoxide/pmol of total CYP/min]**

Relative contribution of isoforms to perazine 5-sulphoxidation in liver microsomes [%]***

CYP1A1 (not constitutive) CYP1A2 CYP2A6 CYP2B6 CYP2C9 CYP2C19 CYP2D6 CYP2E1 CYP3A4

1.678 0.730 0.307 0.470 0.177 0.823 1.190 0.418 0.304

– 0.127(a) 0.040(a) 0.002(a) 0.156(b) 0.026(c) 0.015(a) 0.066(a) 0.288(a)

– 0.093 0.012 0.001 0.028 0.021 0.018 0.027 0.087

– 32.4 4.2 0.4 9.7 7.3 6.3 9.4 30.3

* Data according to Shimada et al., 1994(a); Maurel, 1998(b); Lasker et al., 1998(c). ** The predicted velocity in liver microsomes was calculated by multiplying the velocity in Supersomes by the relative contribution of an isoform to the total CYP content in liver microsomes. *** The relative contribution of CYPs to perazine 5-sulphoxidation was calculated as a percentage of the sum of the predicted velocities in liver microsomes.

206

J. Wo´jcikowski et al. / European Neuropsychopharmacology 14 (2004) 199–208

Table 5 Estimation of the contribution of CYP isoforms to perazine N-demethylation in liver microsomes on the basis of the rate of this reaction in Supersomes CYPs

Velocity in Supersomes [pmol N-desmethylperazine/ pmol CYP isoform/min]

Relative contribution of isoforms to the total CYP contents in liver microsomes [fraction]*

Predicted velocity in liver microsomes [pmol N-desmethylperazine/ pmol of total CYP/min]**

Relative contribution of isoforms to perazine N-demethylation in liver microsomes [%]***

CYP1A1 (not constitutive) CYP1A2 CYP2A6 CYP2B6 CYP2C9 CYP2C19 CYP2D6 CYP2E1 CYP3A4

0.326 0.135 0.023 0.093 0.042 4.826 1.083 n.d. 0.064

– 0.127(a) 0.040(a) 0.002(a) 0.156(b) 0.026(c) 0.015(a) 0.066(a) 0.288(a)

– 0.017 0.001 0.0002 0.006 0.125 0.016 – 0.018

– 9.3 0.5 0.1 3.3 68.3 8.7 – 9.8

n.d.—Not detected. * Data according to Shimada et al., 1994(a); Maurel, 1998(b); Lasker et al., 1998(c). ** The predicted velocity in liver microsomes was calculated by multiplying the velocity in Supersomes by the relative contribution of an isoform to the total CYP content in liver microsomes. *** The relative contribution of CYPs to perazine N-demethylation was calculated as a percentage of the sum of the predicted velocities in liver microsomes.

in liver microsomes or in vivo the amount of the metabolite formed by an individual isoform depends on both the catalytic activity and the contribution of an isoform to the total CYP content in the liver. Therefore we attempted to roughly estimate the contribution of the CYP isoforms studied to the 5-sulphoxidation and N-demethylation of perazine in liver microsomes on the basis of the rates of those reactions in Supersomes and the relative contribution of the isoforms to the total CYP content in the liver (Tables 4 and 5). Our calculations indicate that CYP1A2 and CYP3A4 are the main isoenzymes responsible for 5sulphoxidation, while CYP2C19 chiefly catalyzes the Ndemethylation of perazine in the liver. Moreover, of the other isoforms studied, CYP2C9, CYP2E1, CYP2C19 and CYP2D6 can moderately contribute to 5-sulphoxidation, while CYP3A4, CYP1A2 and CYP2D6 to the N-demethylation of perazine. The calculated data are in line with other authors’ results showing that the CYP3A subfamily is involved mainly in the 5-sulphoxidation of chlorpromazine (Cashman et al., 1993). These results are also consistent with our recent data showing the main contribution of CYP1A2 and CYP3A4 to the 5-sulphoxidation of the simplest phenothiazine neuroleptic promazine (Wo´jcikowski et al., 2003). The 5-sulphoxidation of both promazine and perazine was mediated to a similar extent by CYP1A2 and CYP3A4 (between 30% and 39% for each drug and the isoform). However, some inter-drug differences were observed in the catalysis of N-demethylation process. The contribution of CYP1A2 and CYP2C19 to promazine N-demethylation was similar (35% and 32%, respectively); in the case of perazine, CYP2C19 was the main isoenzyme catalyzing that reaction (68%). Of the other isoenzymes tested, CYP2C9 and CYP3A4 also contributed to promazine N-demethylation (9% and 12%, respectively), while CYP3A4, CYP1A2 and CYP2D6

participated in the N-demethylation of perazine (10%, 9% and 9%, respectively). Moreover, the results obtained in Supersomes at low (10 AM) and at high (300 AM) concentrations of promazine were similar and indicated that only the role of CYPs contributing to a lesser degree to the promazine metabolism might slightly increase at a higher concentration of the neuroleptic (Wo´jcikowski et al., 2003). Accordingly, in the present experiment the results obtained at a therapeutic concentration (inhibition study) are consistent with those conducted at a higher concentration of perazine (metabolism of perazine in Supersomes and in correlation study showing maximum capacity of the studied isoforms). However, our data (Tables 4 and 5) are not consistent with those obtained by Sto¨rmer et al. (2000a), which indicate that CYP3A4 and CYP2C9 are the main isoforms catalyzing the N-demethylation of perazine, while CYP1A2 and CYP2C19 play a minor role in this process. The observed discrepancies may stem from different methods of calculation used. In the study by Sto¨rmer et al. (2000a), the relative contribution of CYPs to perazine N-demethylation in human liver was calculated using a relative activity factor approach (RAF), where the rate of perazine Ndemethylation by a cDNA-expressed isoform was multiplied by RAF (RAF = Vmax of the specific reaction in liver microsomes/V max of the specific reaction in cDNAexpressed human CYP). Moreover, the latter authors did not report about the agents co-expressed in Supersomes, while the presence and concentration of NADPH oxidoreductase and cytochrome b-5 may influence the velocity of the reaction in cDNA-expressed human CYPs (Roy et al., 1999; Venkatakrishnan et al., 2000). Theoretically, the RAF approach should show more properly the relative contribution of CYPs to drug metabolism, since it takes into consideration not only the relative amount of each isoform

J. Wo´jcikowski et al. / European Neuropsychopharmacology 14 (2004) 199–208

in human liver, but also the conditions of CYP catalysis in Supersomes and human microsomes. However, according to Venkatakrishnan et al. (2000) and Sto¨rmer et al. (2000b), the value of RAF depends on the activity of an isoenzyme in human samples of liver microsomes and on the CYPspecific reaction used for determining this factor. Hence, the two methods of calculation, i.e. ours based on CYP isoform abudance in the liver and that by Sto¨rmer et al. (2000a) using the RAF approach, may yield inconsistent results. However, our results are consistent by those obtained by Olesen and Linnet (2000) which show that CYP1A2, CYP3A4 and CYPC19 are the main isoforms catalyzing N-dealkylation of another piperazine-type phenothiazine neuroleptic perphenazine. Our results also differ from those obtained in rats (Daniel et al., 2002), which show that the 5-sulphoxidation of perazine is catalyzed by CYP2D and CYP2B, while its N-demethylation is mediated by CYP2D, CYP2B and CYP1A2. Such species differences were also observed in our previous study with promazine (Wo´jcikowski et al., 2003). The observed interspecies differences regarding the participation of CYPs in the catalysis of the metabolism of phenothiazines may stem from diverse relative contribution of individual isoenzymes to the total pool of CYP (Shimada et al., 1994; Shimojo et al., 1993), and from different catalytic competence of CYP1A2 and CYP2D in the species (Boobis et al., 1990; Kobayashi et al., 1989; Sesardic et al., 1990; Steiner et al., 1988; Zhi-Guang et al., 1988). In conclusion, the results of present study show that (1) CYP1A2 and CYP3A4 are the main isoenzymes responsible for 5-sulphoxidation, while CYP2C19 is the main one that catalyzes perazine N-demethylation in human liver; (2) CYP2C19 contribution to the catalysis of Ndemethylation process is considerably higher for perazine (68%) than for promazine (32%); and (3) CYP isoform contribution to the metabolism of perazine in humans differs from that in rats. The results of the present study may be of great practical value, since perazine is administered to patients for months or even years—and very often in combination with antidepressants, antimanic or antianxiety drugs—to treat severe, complex or ‘‘treatmentresistant’’ psychiatric disorders. Such situations permit of pharmacokinetic interactions, since the above-mentioned psychotropics involve the same enzymes for their biotransformation. It is also important to note that the metabolism of perazine may be dependent on the known CYP2C19 polymorphism occurring at the highest rate in Oriental populations (15 – 23%). The expression of CYP3A4 may also vary between patients (up to 60 times) because of the polymophisms of CYP3A4 and PXR nuclear receptor genes (Hustert et al., 2001; Shimada et al., 1994). Our results also provide further evidence that the experimental data on CYP catalysis, obtained with animals, have to be approached with care when referred to clinical conditions.

207

Acknowledgements This study was supported by the Poste Jaune fellowship, granted to Jacek Wo´jcikowski by the INSERM, Paris, and by the INSERM U128, Montpellier, France; support also came from the statuatory funds of the Institute of Pharmacology, Polish Academy of Sciences, Krako´w, Poland. References Boobis, A.R., Sesardic, D., Murray, B.P., Edwards, R.J., Singleton, A.M., Rich, K.J., Murray, S., de la Torre, R., Segura, J., Pelkonen, O., Kobayashi, S., Zhi-Guang, T., Davies, D.S., 1990. Species variation in the response of the cytochrome P-450 dependent monooxygenase system to inducers and inhibitors. Xenobiotica 11, 1139 – 1161. Breyer, U., 1969. Urinary metabolites of 10-[3V-(4W-methyl-piperazinyl)propyl]-phenothiazine (perazine) in psychiatric patients. Isolation, identification and determination of metabolites. Biochem. Pharmacol. 18, 777 – 788. Breyer, U., 1971. Metabolism of the phenothiazine drug perazine by liver and lung microsomes from various species. Biochem. Pharmacol. 20, 3341 – 3351. Breyer, U., 1972. Accumulation and elimination of a novel metabolite during chronic administration of the phenothiazine drug perazine to rats. Biochem. Pharmacol. 21, 1419 – 1429. Breyer, U., Villumsen, K., 1976. Measurement of plasma levels of tricyclic psychoactive drugs and their metabolites by UV reflectance photometry of thin layer chromatograms. Eur. J. Clin. Pharmacol. 9, 457 – 465. Cashman, J.R., Yang, Z., Yang, L., Wrighton, S., 1993. Stereo- and regioselective n- and s-oxidation of tertiaryn amines and sulfides in the presence of adult human liver microsomes. Drug Metab. Dispos. 21, 492 – 501. Daniel, W.A., Syrek, M., Haduch, A., Wo´jcikowski, J., 1998. Pharmacokinetics of phenothiazine neuroleptics after coadministration of carbamazepine. Pol. J. Pharmacol. 50, 431 – 442. Daniel, W.A., Syrek, M., Haduch, A., Wo´jcikowski, J., 2000. Different effects of amitriptyline and imipramine on the pharmacokinetics and metabolism of perazine in rats. J. Pharm. Pharmacol. 52, 1473 – 1481. Daniel, W.A., Syrek, M., Haduch, A., Wo´jcikowski, J., 2001. The effect of selective serotonin reuptake inhibitors (SSRIs) on the pharmacokinetics and metabolism of perazine in the rat. J. Pharm. Pharmacol. 53, 449 – 461. Daniel, W.A., Syrek, M., Haduch, A., 2002. Contribution of cytochrome P450 isoenzymes to the metabolism of phenothiazine neuroleptics. Eur. Neuropsychopharmacol. 12, 371 – 377. Daujat, M., Pichard, L., Dalet, C., Larroque, C., Bonfils, C., Pompon, D., Li, D., Guzelian, P.S., Maurel, P., 1987. Expression of five forms of microsomal cytochrome P-450 in primary cultures of rabbit hepatocytes treated with various classes of inducers. Biochem. Pharmacol. 36, 3597 – 3606. Diaz, D., Fabre, I., Daujat, M., Saint-Aubert, B., Bories, P., Michel, H., Maurel, P., 1990. Omeprazole is an aryl hydrocarbon-like inducer of human hepatic cytochrome P450. Gastroenterology 99, 737 – 747. Hustert, E., Zibat, A., Prescan-Siedel, E., Eiselt, R., Mueller, R., Fub, C., Brehm, I., Brinkmann, U., Eichelbaum, M., Wojnowski, L., Burk, O., 2001. Natural protein variants of pregnane X receptor with altered transactivation activity toward CYP3A4. Drug Metab. Dispos. 29, 1454 – 1459. Kaning, K., Breyer, U., 1969. Urinary metabolites of 10-[3V-(4W-methylpiperazinyl)-propyl]-phenothiazine (perazine) in psychiatric patients. Individual metabolite patterns and changes in the course of treatment. Psychopharmacologia 14, 211 – 220.

208

J. Wo´jcikowski et al. / European Neuropsychopharmacology 14 (2004) 199–208

Keck, P., McElroy, S.L., Strakowski, S.M., West, S.A., 1994. Pharmacologic treatment of schizoaffective disorder. Psychopharmacology 114, 529 – 538. Kobayashi, S., Murray, S., Watson, D., Sesardic, D., Davies, D.S., Boobis, A.R., 1989. The specificity of inhibition of debrisoquine 4-hydroxylase activity by quinidine and quinine in the rat is the inverse than in man. Biochem. Pharmacol. 38, 2795 – 2799. Lasker, J.M., Wester, M.R., Aramsombatdee, E., Raucy, J.L., 1998. Characterization of CYP2C19 and CYP2C9 from human liver: respective roles in microsomal tolbutamide, S-mephenytoin, and omeprazole hydroxylations. Arch. Biochem. Biophys. 353, 16 – 28. Maurel, P., 1998. Drug metabolizing enzymes. Curr. Opin. Crit. Care 4, 213 – 218. Maurice, M., Pichard, L., Daujat, M., Fabre, I., Joyeux, H., Domergue, J., Maurel, P., 1992. Effects of imidazole derivatives on cytochromes P450 from human hepatocytes in primary culture. FASEB J. 6, 752 – 758. Nelson, J.C., 1993. Combined treatment strategies in psychiatry. J. Clin. Psychiatry 54 (Suppl. 9), 42 – 49. Olesen, O.V., Linnet, K., 2000. Identification of the human cytochrome P450 isoforms mediating in vitro N-dealkylation of perphenazine. Br. J. Clin. Pharmacol. 50, 563 – 571. Pichard, L., Fabre, I., Fabre, G., Domergue, J., Saint-Aubert, B., Mourad, G., Maurel, P., 1990. Cyclosporin A drug interaction. Screening for inducers and inhibitors of cytochrome P450 (cyclosporin oxidase) in primary culture of human hepatocytes and in liver microsomes. Drug Metab. Dispos. 18, 595 – 606. Rao, M.L., 1989. Monitoring of serum bioactivity levels of perazine and its metabolites by radioreceptor assay. Pharmacopsychiatry 22, 104 – 107. Roy, P., Yu, L.J., Crespi, C.J., Waxmann, D.J., 1999. Development of a substrate-activity based approach to identify the major human liver P450 catalysts of cyclophosphamide and ifosfamide activation based on cDNA-expressed activities and liver microsomal P-450 profiles. Drug Metab. Dispos. 27, 655 – 666. Sesardic, D., Boobis, A.R., Murray, B.P., Murray, S., Segura, J., de la Torre, R., Davies, D.S., 1990. Furafylline is a potent and selective inhibitor of cytochrome P4501A2 in man. Br. J. Pharmacol. 29, 651 – 663. Shimada, T., Yamazaki, H., Mimura, M., Inui, Y., Guengerich, F.P., 1994.

Interindividual variations in human liver cytochrome P-450 enzymes involved in the oxidation of drugs, carcinogenes and toxic chemicals: studies with liver microsomes of 30 Japanese and 30 Caucasians. J. Pharmacol. Exp. Ther. 270, 414 – 423. Shimojo, N., Ishizaki, T., Imaoka, S., Funae, Y., Fuji, S., Okuda, K., 1993. Changes in amounts of cytochrome P450 isoenzymes and levels of catalytic activities in hepatic and renal microsomes of rats with streptozotocin-induced diabetes. Biochem. Pharmacol. 46, 621 – 627. Steiner, E., Dumond, E., Spina, E., Dahlqvist, R., 1988. Inhibition of desipramine 2-hydroxylation by quinidine and quinine. Clin. Pharmacol. Ther. 43, 577 – 581. Sto¨rmer, E., Bro¨ckmo¨ller, J., Roots, I., Schmieder, J., 2000a. Cytochrome P-450 enzymes and FMO3 contribute to the disposition of the antipsychotic drug perazine in vitro. Psychopharmacology 151, 312 – 320. Sto¨rmer, E., von Moltke, L.L., Greenblatt, D.J., 2000b. Scaling drug biotransformation data from cDNA-expressed cytochrome P-450 to human liver: a comparison of relative activity factors and human liver abundance in studies of mirtazapine metabolism. J. Pharmacol. Exp. Ther. 295, 793 – 801. Valoti, M., Frosini, M., Palmi, M., de Matteis, F., Sgaragli, G., 1998. Ndealkylation of chlorimipramine and chlorpromazine by rat liver microsomal cytochrome P450 isoenzymes. J. Pharm. Pharmacol. 50, 1005 – 1011. Venkatakrishnan, K., von Moltke, L.L., Court, M.H., Harmatz, J.S., Crespi, C.L., Greenblatt, D.J., 2000. Comparison between cytochrome P450 (CYP) and relative activity approaches to scaling from cDNA-expressed CYPs to human liver microsomes: ratios of accessory proteins and sources of discrepancies between the approaches. Drug Metab. Dispos. 28, 1493 – 1504. Wo´jcikowski, J., Daniel, W.A., 2000. Distribution interactions between perazine and antidepressant drugs. In vivo studies. Pol. J. Pharmacol. 52, 449 – 457. Wo´jcikowski, J., Pichard-Garcia, L., Maurel, P., Daniel, W.A., 2003. Contribution of human cytochrome P-450 isoforms to the metabolism of the simplest phenothiazine neuroleptic promazine. Br. J. Pharmacol. 138, 1465 – 1474. Zhi-Guang, T., Murray, B., Sesardic, D., Kobayashi, S., Murray, S., Davies, D.S., Boobis, A.R., 1988. Quinidine discriminates between rat and human debrisoquine 4-hydroxylase. Br. J. Clin. Pharmacol. 25, 674.