Pharmacological Reports Copyright © 2012 2012, 64, 15781583 by Institute of Pharmacology ISSN 1734-1140 Polish Academy of Sciences
Short communication
Autoinduction of the metabolism of phenothiazine neuroleptics in a primary culture of human hepatocytes
Jacek Wójcikowski1, Patrick Maurel2, W³adys³awa A. Daniel1
1Department of Pharmacokinetics and Drug Metabolism, Institute of Pharmacology, Polish Academy of Science, Smêtna 12, PL 31-343 Kraków, Poland
2INSERM, U632, Montpellier, France; Université Montpellier 1, UMR-S632, Montpellier, France; CHU Montpellier, Institut de Recherche en Biothérapie, Hôpital Saint Eloi, Montpellier, France
Correspondence: Jacek Wójcikowski e-mail: [email protected]
Abstract: Background: The metabolism of phenothiazine neuroleptics (promazine, perazine) in a primary culture of human hepatocytes after pretreatment of cells with those neuroleptics was studied. Methods: The hepatocytes were pretreated with 25 µM promazine or perazine for 96 h. Then, the cells were incubated for 2, 4, 6, 8 and 24 h in the presence of neuroleptics. At the indicated time points, concentrations of phenothiazines and their metabolites (5-sulfoxides and N-desmethyl derivatives) were measured in the culture medium using HPLC with UV detection. Results: Pretreatment of the primary culture of human hepatocytes with promazine or perazine resulted in accumulation of their me- tabolites in the culture medium. Such an effect was not observed in the case of control cultures (not pretreated with neuroleptics). Conclusion: The obtained results suggest that the tested phenothiazines may stimulate their own metabolism by inducing CYP1A2, CYP3A4 and CYP2C19 isoforms.
Key words: phenothiazine neuroleptics, metabolism, autoinduction, human cytochrome P450, hepatocytes
Introduction culture of human hepatocytes, which is a convenient model system to examine the oxidative metabolism of Cytochrome P450 (CYP) enzymes are members of drugs and xenobiotics [7, 11]. CYP1A2 and CYP3A4 the superfamily of heme-containing monooxygenases constitute the main pool of CYP protein in human that catalyze the metabolism of endogenous sub- liver (10–13% and 30–50%, respectively) [13, 22]. stances (e.g., steroid hormones, neurosteroids, mono- CYP1A2 can be induced by cigarette smoking and by aminergic neurotransmitters, arachidonic acid) and a number of polycyclic aromatic hydrocarbons such the majority of clinically important drugs including as TCDD and b-naphthoflavone, while CYP3A4 is psychotropics. It has been reported that several forms strongly inducible by a number of drugs including ri- of CYP (such as, e.g., CYP1A1/2, CYP2B6, fampicin, carbamazepine, barbiturates and cortico- CYP2C9/19, CYP3A4/5) are inducible in a primary steroids [9, 14, 15].
1578 Pharmacological Reports, 2012, 64, 15781583 Autoinduction of phenothiazine metabolism Jacek Wójcikowski et al.
Our recent studies conducted on human liver micro- Biotransformation of neuroleptics in a primary somes and cDNA-expressed human CYP enzymes have culture of human hepatocytes shown that CYP1A2, CYP3A4 and CYP2C19 are the main isoforms involved in the metabolism of phenothi- The use of human liver samples for scientific pur- azine neuroleptics with different chemical structures [23, poses was approved by the French National Ethics 26, 27, 29]. The above-cited results were corroborated Committee. A human liver specimen was obtained by our experiments conducted on human hepatocytes. from patient FT176 (a 69-year-old female) who Treatment of the primary culture of human hepatocytes underwent liver lobectomy due to the hepatic metasta- with TCDD (a CYP1A1/2 inducer) and rifampicin sis of colon cancer. The tissue encompassing the tu- (mainly a CYP3A4 inducer) increased the rate of the mor was dissected by a surgeon and sent for anatomo- biotransformation of phenothiazines [27, 28]. pathological examination, whereas the remaining Some clinical studies suggested that phenothiazine healthy tissue was used for hepatocyte preparation. neuroleptics may induce human liver CYP, enhancing No information about the patient was available to us, the metabolism of co-administrated drugs. A signifi- apart from her age, sex and the cause of surgery. The cant reduction in antipyrine half-life in patients re- viral serological analysis of the tissue (hepatitis B and ceiving long-term chlorpromazine treatment was ob- C viruses and a human immunodeficiency virus) was served [10, 12]. Other studies showed that thioridaz- negative. ine visibly increased the oral clearance of quetiapine Primary cultures of human hepatocytes were pre- in psychiatric patients [19]. Moreover, Rivera-Calim- pared as described previously [6, 18]. Hepatocytes lim et al. [21] reported that plasma chlorpromazine with viability higher than 90%, as determined by levels declined with time in patients who were on trypan blue exclusion, were used in the experiment. fixed doses of that neuroleptic. They also suggested Ten million cells in 7 ml of a culture medium were that a possible cause of that phenomenon might have placed in 10-cm plastic Petri dishes pre-coated with been induction of CYP enzymes metabolizing chlor- collagen (Beckton-Dickinson, France). That medium promazine. consisted of a mixture of Ham F12 and Williams E The aim of the present study was to ascertain (1:1 v/v), supplemented as described previously [7]. whether the phenothiazine neuroleptics with different The culture medium was also supplemented with chemical structures may stimulate their own metabo- a 5% calf serum during the first 4 h after plating, to lism. We examined the metabolism of promazine (the favor the attachment of cells. Then, 4–6 h later, the simplest and an aliphatic-type phenothiazine neurolep- medium was exchanged for a serum-free medium and tic) and perazine (a piperazine-type phenothiazine neu- the culture was allowed to stabilize for another 24 h. roleptic) in a primary culture of human hepatocytes. The medium was exchanged every 24 h for a serum- and fenol red-free medium. The cultures were kept at 37°C in an atmosphere of a 95% air and a 5% CO2, at a ca. 100% humidity. Materials and Methods For the pretreatment of cells, promazine and pera- zine were diluted in DMSO and added to the culture Drugs and chemicals medium at a final concentration of 25 µM (the ex- pected concentration of neuroleptics in human liver Promazine was provided by Polfa (Jelenia Góra, after pharmacological doses of the tested drugs) [8, Poland), while perazine was obtained from Labor 24, 25]. Control cells were treated with DMSO (sol- (Wroc³aw, Poland). D-desmethylpromazine was do- vent) added to the culture medium. The concentration nated by Professor M.H. Bickel, the University of of DMSO in the culture medium (of both control and Bern, Switzerland. Promazine 5-sulfoxide, perazine neuroleptic-pretreated cultures) was 0.1%. Pretreat- 5-sulfoxide and N-desmethylperazine were synthe- ment lasted for 96 h (the time usually needed to ob- sized according to the previously described methods serve enzyme induction) [16, 17, 27, 28] and was re- [2]. Dimethyl sulfoxide (DMSO) was purchased from newed every 24 h when the culture medium was ex- Sigma (St. Louis, USA). All the organic solvents with changed. After 96 h, the culture medium (of both HPLC purity were supplied by Carlo Erba (Milan, control and neuroleptic-pretreated cultures) was ex- Italy). changed for a medium containing 25 µM promazine
Pharmacological Reports, 2012, 64, 15781583 1579 or perazine. The cells were then incubated under stan- Results and Discussion dard culture conditions for 2, 4, 6, 8 and 24 h. At the indicated time points, 500 µl aliquots of the culture medium were collected, and 50 µl were injected After a 96-h pretreatment of the primary cultures of directly into the HPLC system. human hepatocytes with promazine or perazine, their metabolites (5-sulfoxides and N-desmethyl deriva- Determination of the concentration tives) were found to accumulate in the culture me- of promazine, perazine and their metabolites dium (Tabs. 1 and 2). Such an effect was not observed in the culture medium in the case of control cultures (not pretreated with neuroleptics). In the latter case, promazine and pera- The concentrations of promazine and its metabolites zine metabolites were not detected in the culture me- (promazine 5-sulfoxide and N-desmethylpromazine), dium. Approximately 100% of perazine metabolism and perazine and its metabolites (perazine 5-sulfoxide and 85% of promazine metabolism took place during and N-desmethylperazine) were assayed using a Varian the first 24 h in perazine- and promazine-treated cul- HPLC system with UV detection. The analytical col- tures, respectively (Tabs. 1 and 2). The accumulation umn (Econosphere C18, 5 µm, 4.6 × 250 mm) was of promazine 5-sulfoxide and perazine 5-sulfoxide in purchased from Alltech (Carnforth, England). The the culture medium was enhanced up until 24 h. On mobile phase consisted of an acetate buffer, pH = 3.4 the other hand, the concentration of N-desmethyl- (100 mmol of ammonium acetate, 20 mmol of citric promazine in the culture medium increased up until acid and 1 ml of triethylamine in 1,000 ml of the 6 h, and then decreased up until 24 h. N-desmethyl- buffer adjusted to pH = 3.4 with an 85% phosphoric perazine was detected in the culture medium after acid), and acetonitrile in a proportion of 50 : 50 (v/v). 24 h only (Tabs. 1 and 2). Elution proceeded at an ambient temperature at a flow The present study provides fresh evidence for an rate of 1.2 ml/min. The absorbance of promazine and its earlier clinical observation suggesting induction of metabolites was measured at a wavelength of 254 nm, CYP enzymes by phenothiazine neuroleptics. We pre- while the absorbance of perazine and its metabolites viously showed that CYP1A2 and CYP3A4 were in- was measured at a wavelength of 265 nm. The detection volved mainly in the 5-sulfoxidation of promazine limit for promazine and its metabolites was as low as and perazine. Moreover, promazine N-demethylation 0.01 nmol/sample. The detection limit for perazine was metabolized by CYP1A2 and CYP2C19, while and its metabolites was as low as 0.015 nmol/sample. CYP2C19 was the main enzyme catalyzing perazine
Tab. 1. The influence of a 96-h pretreatment with promazine on its own metabolism in a primary culture of human hepatocytes.
Time (h) Control Promazine
Promazine N-desmethyl Promazine Promazine N-desmethyl Promazine 5-sulfoxide (µM) promazine (µM) (µM) 5-sulfoxide (µM) promazine (µM) (µM)
2 nd nd 9.54 nd 1.52 14.01
4 nd nd 8.29 nd 1.75 13.76
6 nd nd 6.73 0.84 1.78 12.63
8 nd nd 4.40 0.90 1.43 9.46
24 nd nd 0.54 1.08 0.83 3.75
Human hepatocytes were kept in the primary culture and pretreated with 25 µM promazine (diluted in DMSO) for 96 h. Control cells were treated with DMSO. After 96 h, the culture medium (of both control and promazine-pretreated cultures) was exchanged for a medium contain- ing 25 µM of promazine. The cells were then incubated under standard culture conditions for 2, 4, 6, 8 and 24 h. The concentrations of promaz- ine and its metabolites (promazine 5-sulfoxide and N-desmethylpromazine formed from the neuroleptic) were measured in the culture medium (one culture dish per one time point); nd not detected
1580 Pharmacological Reports, 2012, 64, 15781583 Autoinduction of phenothiazine metabolism Jacek Wójcikowski et al.
Tab. 2. The influence of a 96-hour pretreatment with perazine on its own metabolism in a primary culture of human hepatocytes.
Time (h) Control Perazine
Perazine N-desmethyl Perazine Perazine N-desmethyl Perazine 5-sulfoxide (µM) perazine (µM) (µM) 5-sulfoxide (µM) perazine (µM) (µM)
2 nd nd 6.46 nd nd 7.88
4 nd nd 5.30 1.15 nd 6.45
6 nd nd 3.97 1.90 nd 4.93
8 nd nd 2.20 3.11 nd 3.77
24 nd nd nd 5.14 3.34 nd
Human hepatocytes were kept in the primary culture and pretreated with 25 µM perazine (diluted in DMSO) for 96 h. Control cells were treated with DMSO. After 96 h, the culture medium (of both control and perazine-pretreated cultures) was exchanged for a medium containing 25 µM of perazine. The cells were then incubated under standard culture conditions for 2, 4, 6, 8 and 24 h. The concentrations of perazine and its me- tabolites (perazine 5-sulfoxide and N-desmethylperazine formed from the neuroleptic) were measured in the culture medium (one culture dish per one time point); nd not detected
N-demethylation [27-29]. In the light of the present over, it has been reported that chlorpromazine reduces results, it seems feasible that promazine and perazine antipyrine half-life in patients undergoing long-term may accelerate their own metabolism by inducing the chlorpromazine therapy, while thioridazine (a piperi- CYP isoforms mentioned above. dine-type phenothiazine neuroleptic) visibly enhances The obtained results are in line with our earlier data the oral clearance of quetiapine in psychiatric patients showing that treatment of the primary culture of human [10, 12, 19]. Since antipyrine and quetiapine are me- hepatocytes with rifampicin (an inducer of CYP3A4 and, tabolized chiefly by CYP1A2 and CYP3A4, respec- to a lesser degree, of CYP2C19 and CYP2B6) and tively [20], the changes observed in the pharmacoki- TCDD (a CYP1A1/2 inducer) increased formation of netics of these drugs may result from the induction of promazine 5-sulfoxide, perazine 5-sulfoxide and N-des- CYP1A2 and CYP3A4 by co-administered phenothi- methylpromazine [27, 28]. The formation of N-des- azine neuroleptics. methylperazine was also observed in rifampicin-treated Interestingly, in the case of control hepatocytes, the cultures, but not in TCDD-treated ones. However, ri- concentrations of promazine and perazine measured in fampicin induced formation of N-desmethylperazine in the culture medium after 2 h were lower than those hepatocytes after 24 h only [28]. In the present study, measured in neuroleptic-treated hepatocytes. The latter a certain amount of N-desmethylperazine, formed by effect may spring from the fact that after a 96-h treat- a perazine-treated culture, was also detected after 24 h ment of cells with neuroleptics, promazine and pera- only, which seems to suggest that this neuroleptic is zine (as well as their metabolites) may still be present a weak inducer of CYP2C19. in hepatocytes. According to Daniel and Wójcikowski The present results are also consistent with the [3–5], distribution of psychotropic drugs that are basic clinical data suggesting that chlorpromazine, an lipohilic compounds (such as promazine, perazine and aliphatic-type phenothiazine neuroleptic, may induce their N-desmethyl derivatives) depends mainly on CYP [21]. Our recent studies have shown that phospholipid binding and lysosomal trapping. CYP1A2 is the only CYP isoform that catalyzes Furthermore, during the first 24 h, phenothiazines chlorpromazine N-demethylation and the main iso- show a decline in their concentration, which is faster form responsible for chlorpromazine 5-sulfoxidation. in control than in neuroleptic-treated hepatocytes. CYP3A4 contributes to a lesser degree to the 5-sulf- Thus, inhibition of another route of promazine and oxidation of chlorpromazine [23]. Therefore, it seems perazine metabolism (e.g., aromatic hydroxylation) in possible that chlorpromazine may accelerate its me- neuroleptic-treated hepatocytes seems likely. It has tabolism by inducing CYP1A2 and CYP3A4. More- been reported that phenothiazine neuroleptics may
Pharmacological Reports, 2012, 64, 15781583 1581 yield a great number of reactive phenothiazine me- posure to neuroleptics in vivo – possible involvement of tabolites, such as radical cations, which are able to in- different mechanisms. Eur Neuropsychopharmacol, 2005, 15, 103–110. hibit CYP2D6, the latter catalyzing the hydroxylation 2. Daniel WA, Syrek M, Haduch A, Wójcikowski J: of phenothiazines [1, 30]. Pharmacokinetics of phenothiazine neuroleptics after In summary, our preliminary results concerning the coadministration of carbamazepine. Pol J Pharmacol, primary culture of human hepatocytes suggest that 1998, 50, 431–442. 3. phenothiazine neuroleptics may stimulate their own Daniel WA, Wójcikowski J: Contribution of lysosomal trapping to the total tissue uptake of psychotropic drugs. metabolism by inducing CYP1A2, CYP3A4 and Pharmacol Toxicol, 1997, 80, 62–68. CYP2C19. By inducing CYP isoforms (e.g., during 4. Daniel WA, Wójcikowski J: Interactions between pro- a long-term neuroleptic therapy of psychiatric pa- mazine and antidepressants at the level of cellular distri- tients), phenothiazines may increase their elimination bution. Pharmacol Toxicol, 1997, 81, 259–264. 5. Daniel WA, Wójcikowski J: Lysosomal trapping as an im- and thus attenuate the desired pharmacological effect. portant mechanism involved in the cellular distribution of Moreover, phenothiazine neuroleptics are adminis- perazine and in pharmacokinetic interaction with antide- tered for months or even years, also to patients treated pressants. Eur Neuropsychopharmacol, 1999, 9, 483–491. simultaneously with other clinically important drugs 6. Diaz D, Fabre I, Daujat M, Saint-Aubert B, Bories P, that are substrates of CYP1A2 (e.g., caffeine, theo- Michel H, Maurel P: Omeprazole is an aryl hydrocarbon-like inducer of human hepatic cytochrome phylline, phenacetin, imipramine, propranolol, clozap- P450. Gastroenterology, 1990, 99, 737–747. ine, melatonin), CYP3A4 (e.g., carbamazepine, cy- 7. Ferrini JB, Ourlin JC, Pichard L, Fabre G, Maurel P: closporin A, calcium channel antagonists, macrolide Human hepatocyte culture. In: Cytochrome P450 Proto- antibiotics, benzodiazepines) or CYP2C19 (e.g., tri- cole. Methods in Molecular Biology. Eds. Phillips IR, cyclic antidepressants, S-mephenytoin, omeprazole). Shephard E, Humana Press Inc., Totowa, NJ, 1998, 341–352. They may enhance metabolism of the co-administered 8. Fichtl B, Nieciecki A, Walter K: Tissue binding versus drugs, leading to the diminution of their pharmacol- plasma binding of drugs: general principles and pharma- ogical effect. On the other hand, hepatic CYP3A4 in- cokinetic consequences. Adv Drug Res, 1991, 20, 117–166. duction by phenothiazine neuroleptics may alter me- 9. Graham MJ, Lake BG: Induction of drug metabolism: tabolism of endogenous substrates (e.g., steroid hor- Species differences and toxicological relevance. Toxicol- ogy, 2008, 254, 184–191. mones), while the induction of CYP1A2 may increase 10. Harman AW, Frewin DB, Priestly BG: Comparative the metabolic transformation of heterocyclic aromatic enzyme-inducing effects of chlorpromazine and fluphe- amines into reactive intermediates, resulting in toxic- nazine therapies in psychotic patients. Psychopharmacol- ity and cancer. Hence, the induction of CYP enzymes ogy, 1980, 69, 35–37. 11. Hewitt NJ, de Kanter R, LeCluyse E: Induction of drug by phenothiazines may be of physiological, pharma- metabolizing enzymes: a survey of in vitro methodolo- cological and toxicological importance. However, fur- gies and interpretations used in the pharmaceutical in- ther metabolic and molecular studies are scheduled to dustry – do they comply with FDA recommendation? corroborate our present results. Moreover, since the Chem Biol Interact, 2007, 168, 51–65. 12. neuroleptics studied are basic lipophilic drugs, their Loga S, Curry S, Lader M: Interactions of orphenadrine and phenobarbitone with chlorpromazine: plasma con- concentrations will be analyzed simultaneously in an centrations and effects in man. Br J Clin Pharmacol, incubation medium and in hepatocytes. 1975, 2, 197–208. 13. Maurel P: Drug metabolizing enzymes. Curr Opin Crit Care, 1998, 4, 213–218. 14. Acknowledgments: Monostory K, Dvorak Z: Steroid regulation of drug- This study was supported by the Poste Jaune fellowship, granted metabolizing cytochrome P450. Curr Drug Metab, 2011, to Jacek Wójcikowski by the INSERM, Paris, and by the INSERM 12, 154–172. U632, Montpellier, France; support also came from the statutory 15. Pelkonen O, Turpeinen M, Hakkola J, Honkakoski P, funds of the Institute of Pharmacology, Polish Academy of Hukkanen J, Raunio H: Inhibition and induction of hu- Sciences, Kraków, Poland. man cytochrome P450 enzymes: current status. Arch Toxicol, 2008, 82, 667–715. 16. Pichard L., Fabre I, Daujat M, Domergue J, Joyeux H, Maurel P: Effect of corticosteroids on the expression of References: cytochromes P450 and on cyclosporine A oxidase activ- ity in primary cultures of human hepatocytes. Mol Phar- macol, 1992, 41, 1047–1055. 1. Daniel WA, Haduch A, Wójcikowski J: Inhibition of rat 17. Pichard-Garcia L, Hyland R, Baulieu J, Fabre J-M, liver CYP2D in vitro and after 1-day and long-term ex- Milton A, Maurel P: Human hepatocytes in primary cul-
1582 Pharmacological Reports, 2012, 64, 15781583 Autoinduction of phenothiazine metabolism Jacek Wójcikowski et al.
ture predict lack of cytochrome P-450 induction by elet- 25. Wójcikowski J, Daniel WA: Thioridazine-fluoxetine in- riptan in vivo. Drug Metab Dispos, 2000, 28, 51–57. teraction at the level of the distribution process in vivo. 18. Pichard L., Raulet E, Fabre G, Ferrini JB, Ourlin J-C, Pol J Pharmacol, 2002, 54, 647–654. Maurel P: Human hepatocyte culture. Methods Mol Biol, 26. Wójcikowski J, Maurel P, Daniel WA: Characterization 2006, 320, 283–293. of human cytochrome P450 enzymes involved in the me- 19. Potkin SG, Thyrum PT, Alva G, Bera R, Yeh C, Arvanitis tabolism of the piperidine-type phenothiazine neuroleptic LA: The safety and pharmacokinetics of quetiapine when thioridazine. Drug Metab Dispos, 2006, 3, 471–476. coadministered with haloperidol, risperidone, or thioridaz- 27. Wójcikowski J, Pichard-Garcia L, Maurel P, Daniel WA: ine. J Clin Psychopharmacol, 2002, 22, 121–130. Contribution of human cytochrome P-450 isoforms to 20. Prior TI, Baker GB: Interactions between the cyto- the metabolism of the simplest phenothiazine neuroleptic chrome P450 system and the second-generation antipsy- promazine. Br J Pharmacol, 2003, 138, 1465–1474. chotics. J Psychiatry Neurosci, 2003, 28, 99–112. 28. Wójcikowski J, Pichard-Garcia L, Maurel P, Daniel WA: 21. Rivera-Calimlim L, Griesbach PH, Perlmutter R: Plasma Effects of cytochrome P-450 inducers on the perazine chlorpromazine concentrations in children with behav- metabolism in a primary culture of human hepatocytes. ioral disorders and mental illness. Clin Pharmacol Ther, Pol J Pharmacol, 2003, 55, 655–658. 1979, 26, 114–121. 29. Wójcikowski J, Pichard-Garcia L, Maurel P, Daniel WA: 22. Shimada T, Yamazaki H, Mimura M, Inui Y, Guengerich The metabolism of the piperazine-type phenothiazine FP: Interindividual variations in human liver cytochrome neuroleptic perazine by the human cytochrome P-450 P-450 enzymes involved in the oxidation of drugs, car- isoenzymes. Eur Neuropsychopharmacol, 2004, 14, cinogenes and toxic chemicals: studies with liver micro- 199–208. somes of 30 Japanese and 30 Caucasians. J Pharmacol 30. Yoshii K, Kobayashi K, Tsumuji M, Tani M, Shimada N, Exp Ther, 1994, 270, 414–423. Chiba K: Identification of human cytochrome P450 iso- 23. Wójcikowski J, Boksa J, Daniel WA: Main contribution forms involved in the 7-hydroxylation of chlorpromazine of the cytochrome P450 isoenzyme 1A2 (CYP1A2) to by human liver microsomes. Life Sci, 2000, 67, N-demethylation and 5-sulfoxidation of the phenothiazine 175–184. neuroleptic chlorpromazine in human liver – A compari- son with other phenothiazines. Biochem Pharmacol, 2010, 80, 1252–1259. 24. Wójcikowski J, Daniel WA: Distribution interactions be- tween perazine and antidepressant drugs. In vivo studies. Received: September 25, 2012; in the revised form: November Pol J Pharmacol, 2000, 52, 449–457. 12, 2012; accepted: December 10, 2012.
Pharmacological Reports, 2012, 64, 15781583 1583