Introduction

Top Abstract Introduction Experimental Procedures Results and Discussion References

Although, the neuroleptic agent haloperidol [HP¹, 4-(4-chlorophenyl)-1-[4-(4-fluorophenyl)-4-oxobutyl]-4-piperidinol] is one of the most widely used antipsychotic drugs, a narrow therapeutic index and a large interindividual and interracial variability in pharmacokinetics results in the requirement of individualized HP dose optimization (Ulrich et al., 1998). The narrow therapeutic index is associated with the frequent occurrence of extrapyramidal side effects including acute dystonic reactions, akathisia, Parkinsonism, and, following chronic treatment, tardive dyskinesias (TD) that are slow to develop and often irreversible (Wright et al., 1998). The persistence of TD in many patients after discontinuation of HP therapy suggests that this condition may be related to a neuronal lesion induced by HP or a reactive metabolite(s) derived from it.

HP is extensively metabolized in the liver with only ~1% of the administered dose excreted in the urine (Forsman et al., 1977). Major biotransformation pathways of HP in humans have been extensively characterized (see Kudo and Ishizaki, 1999 for a review) and are summarized in Fig. 1. These include 1) glucuronidation of the 3⁰ alcohol moiety (Someya et al., 1992); 2) reduction of the carbonyl group by cytosolic carbonyl reductase, which leads to reduced HP (Eyles and Pond, 1992); 3) reverse oxidation of reduced HP back to HP (Pan et al., 1998); 4) N-dealkylation leading to the formation of 4-(4-chlorophenyl)-4-hydroxypiperidine (CPHP) (Fang et al., 2001); 5) dehydration of 3⁰ alcohol moiety to 4-(4-chlorophenyl)-1-[4-(4-fluorophenyl)-4-oxobutyl]-1,2,3,6-tetrahydropyridine (HPTP) (Subramanyam et al., 1991a; Van der Schyf et al., 1994); 6) oxidation of the piperidin-4-ol moiety in HP to the corresponding 4-(4-Chlorophenyl)-1-[4-(4-fluorophenyl)-4-oxobutyl]-pyridinium (HPP⁺) metabolite (Usuki et al., 1996); and 7) oxidation of the piperidin-4-ol moiety in reduced HP to the corresponding 4-(4-Chlorophenyl)-1-[4-(4-fluorophenyl)-4-hydroxybutyl]-pyridinium (RHPP⁺) metabolite (Eyles et al., 1997). In contrast to earlier studies (Llerena et al., 1992; Viala et al., 1996) that suggested the involvement of cytochrome P450 (P450) 2D6 in HP metabolism to CPHP, HPTP, and HPP⁺, recent reports have indicated that these pathways are almost exclusively catalyzed by CYP3A4 (Usuki et al., 1996; Fang et al., 2001).

View larger version (16K): [in this window] [in a new window]   Fig. 1.   Biotransformation pathways of haloperidol in humans.

In view of the neurotoxic properties of 1-methyl-4-phenylpyridinium (MPP⁺), the monoamine oxidase-B-generated metabolite of the Parkinsonian agent 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (Kalgutkar et al., 2001), the identification of MPP⁺-like HPP⁺ and RHPP⁺ metabolites in significant quantities in the urine (Subramanyam et al., 1991b), plasma (Avent et al., 1997), and post-mortem brain samples (Eyles et al., 1997) in schizophrenic patients treated with HP is of neurotoxicological importance especially in the pathogenesis of HP-induced TD. Additional support for this proposal is evident from the observations that HPP⁺ displays MPP⁺-like neurotoxicity in vivo as well as in vitro (Bloomquist et al., 1994).

Although glucuronidation and reduced HP formation constitute the major routes of HP clearance in humans, the wide interindividual variations in HP oral clearance have been mainly attributed to the "minor" P450-catalyzed biotransformation pathways (30% of overall HP metabolism) (Kudo and Ishizaki, 1999). These proposals are derived from in vitro observations on the lack of variations in HP glucuronidation or reduction activities in vitro and ~10- to 15-fold interindividual variation in CYP3A4-mediated HP metabolism. For instance, Usuki et al. found that the rate of formation of HPP⁺ varied ~10-fold among 14 liver samples (Usuki et al., 1996), similar to the variation reported for CPHP formation (Pan et al., 1998) and the back-oxidation of reduced HP to HP (Kudo and Odomi, 1998; Fang et al., 2001). In this context, the contribution of polymorphic CYP3A5 (Kuehl et al., 2001) in HP metabolism relative to CYP3A4 remains undetermined.

Since HP is widely used in the treatment of schizophrenia and other psychiatric disorders, HP is commonly coadministered with other antipsychotics and antidepressants. As a consequence, there are several documented reports on drug-drug interactions between HP and antidepressants including selective serotonin reuptake inhibitors (SSRIs) (Vandel et al., 1995; Barbhaiya et al., 1996; Avenoso et al., 1997). Biochemical mechanisms of these interactions were thought to involve the attenuation of CYP2D6-mediated HP metabolism by antidepressants (Nemeroff et al., 1996), based on the well established CYP2D6 inhibitory properties of antidepressants and earlier reports that suggested the involvement of CYP2D6 in HP metabolism. However, recent findings on the exclusive involvement of CYP3A4 and not CYP2D6 in HP metabolism and on the CYP3A4 inhibitory effects of antidepressants including fluoxetine (Schmider et al., 1995), fluvoxamine (Schmider et al., 1995), and nefazodone (von Moltke et al., 1999) on benzodiazepine metabolism suggest that attenuation of the CYP3A4/5-catalyzed metabolism of HP in vivo may also represent a viable mechanism for HP/antidepressant drug-drug interactions. Based on these proposals and our general interest in the bioactivation of cyclic tertiary amines, specific studies were designed to evaluate the relative contributions of CYP3A4 and the polymorphic CYP3A5 in HP bioactivation to the potentially neurotoxic HPP⁺ as a plausible explanation for the observed interindividual variability in HP metabolism. We also characterized the inhibitory effects of antidepressants including SSRIs on the CYP3A4/5-mediated bioactivation pathway.

	Experimental Procedures

Top Abstract Introduction Experimental Procedures Results and Discussion References

Materials. Haloperidol, buspirone, trazodone, fluoxetine, norfluoxetine, and ketoconazole were purchased from Sigma-Aldrich (St. Louis, MO). 4-(4-Chlorophenyl)-4-hydroxypiperidine, 1-(2-pyrimidyl)piperazine · 2HCl were purchased from Aldrich (Milwaukee, WI). 4-(4-Chlorophenyl)-1,2,3,6-tetrahydropyridine · HCl, 1-(3-chlorophenyl)piperazine · HCl, and fluvoxamine maleate were obtained from Across Organics (Pittsburgh, PA), Tocris Cookson Inc. (Ballwin, MO), and Avocado Research (Heysham, Lancs, UK), respectively. All other chemicals and reagents were obtained from Aldrich. ¹H NMR spectra in DMSO-d₆ or CD₃CN were recorded on a Varian Unity M-400 MHz spectrometer (Varian Medical Systems, Palo Alto, CA); chemical shifts are expressed in parts per million (ppm, ) relative to tetramethylsilane as internal standard. Spin multiplicities are given as d (doublet), dd (doublet of doublet), t (triplet), and m (multiplet). Baculovirus-expressed human CYP3A4 and CYP3A5 were purchased from PanVera Corp. (Madison, WI) and BD Gentest Corp. (Woburn, MA). Human liver microsomes were generated at Pfizer using liver tissue from 54 individual donors. HPTP · HCl and HPP⁺ were synthesized as described previously (Subramanyam et al., 1991a). The internal standard N-nonyl-4-phenylpyridinium (NPP⁺) iodide was synthesized via the quaternization of commercially available 4-phenylpyridine with 1-iodononane to afford the desired compound as a crystalline yellow solid in 57% yield. ¹H NMR (DMSO-d₆)  9.07-9.09 (d, 2 H, J = 6.8 Hz, ArH), 8.49-8.51 (d, 2H, J = 6.4 Hz, ArH), 8.04-8.06 (dd, 2H, J = 7.6 and 3.0 Hz, ArH), 7.61-7.63 (m, 3H, ArH), 4.52-4.56 (t, 2H, J = 7.2 Hz, CH₂), 1.90 (m, 2H, CH₂), 1.20-1.26 (m, 12 H, CH₂), 0.79-0.82 (t, J = 6 Hz, 3H, CH₃). LC-MS/MS analysis revealed a single peak (R_t = 15.94 min) with a molecular ion (M⁺) at 282 and a base fragment ion at 156 (loss of the nonyl side chain).

Oxalate Salt of Nefazodone. A 200-mg nefazodone · HCl tablet was neutralized with 1 N NaOH (pH ~ 9), and this aqueous solution was extracted with Et₂O (2 × 20 ml). The combined Et₂O extracts were washed with water (100 ml) and then treated dropwise with a solution of oxalic acid (1 equivalent based on the weight of nefazodone free base) in Et₂O to precipitate the crude oxalate salt that upon recrystallization from MeOH/Et₂O afforded a white crystalline solid. ¹H NMR (DMSO-d₆)  7.18-7.26 (m, 3H, ArH), 6.79-6.97 (m, 6H, ArH), 4.10-4.13 (t, 2 H, J = 5.2 Hz, CH₂), 3.91-3.94 (t, 2H, J = 5.2 Hz, CH₂), 3.66-3.69 (t, 2H, J = 6.8 Hz, CH₂), 3.29 (m, 4H, CH₂), 2.82-2.96 (m, 4 H, CH₂), 2.80-2.82 (m, 2H, CH₂), 2.60-2.65 (q, 2H, CH₂), 1.90-1.94 (m, 2H, CH₂), 1.14-1.18 (t, 3H, J = 6.8 Hz, CH₃).

Metabolite Identification Following Incubation of Haloperidol in Human Liver Microsomes and in the Presence of Recombinant CYP3A4 and CYP3A5. Haloperidol (50 µM) was incubated with human liver microsomes (P450 concentration = 0.25 µM), or human recombinant CYP3A4 (100 nM), or CYP3A5 (100 nM) in the presence of NADPH (1.3 mM). All incubations were carried out in 0.1 M potassium phosphate buffer (pH = 7.4) at 37°C. After preincubation at 37°C for 2 min, the reaction was initiated by adding NADPH and terminated after a 30 min incubation by adding cold acetonitrile (2:1 v/v). Samples were centrifuged at 3000g for 15 min, and the supernatant was dried under a steady nitrogen stream. The residue was reconstituted in 200 µl of mobile phase (10 mM ammonium formate, 0.1% formic acid/acetonitrile; 75:25) and analyzed by LC-MS/MS as outlined in previous protocols (Kuperman et al., 2001).

Determination of Kinetic Constants for Haloperidol Bioactivation to HPP⁺ in Human Liver Microsomes and Recombinant CYP3A4 and CYP3A5. Haloperidol (0-300 µM) was incubated with pooled human liver microsomes (P450 concentration = 50 nM), or baculovirus-expressed human recombinant CYP3A4 (10 nM) or CYP3A5 (50 nM), NADPH (1.3 mM) in 200 µl of 0.1 M phosphate buffer (pH 7.4), in triplicate. The reaction mixtures were prewarmed at 37°C for 2 min before adding NADPH, then incubated for 10 min. The reactions were terminated by the addition of 0.2 ml of acetonitrile containing NPP⁺ as an internal standard. Samples were centrifuged at 3000g for 15 min, and the supernatants were analyzed for HPP⁺ formation by LC-MS/MS.

Inhibition Studies. The effects of antidepressants and their major metabolites on the conversion of haloperidol to HPP⁺ were studied in human liver microsomes and in recombinant CYP3A4 and CYP3A5. In all experiments, nefazodone, buspirone, trazodone, fluoxetine, norfluoxetine, fluvoxamine, 1-(3-chlorophenyl)piperazine, 1-(2-pyrimidyl)piperazine, and ketoconazole were dissolved and diluted serially in DMSO. The final concentration of these compounds ranged from 0.1 to 200 µM and that of DMSO was less than 0.1% in 200 µl of reaction volume. Each inhibition study was performed in triplicate. Incubation mixtures (200 µl) contained haloperidol (30 or 200 µM) and human liver microsomes (P450 concentration = 50 nM), or human recombinant CYP3A4 (10 nM), or human recombinant CYP3A5 (50 nM) in 0.1 M phosphate buffer (pH 7.4). The reaction mixtures were prewarmed at 37°C for 2 min before adding NADPH (1.3 mM), then incubated for 10 min for the measurement of percentage remaining CYP3A activity. Reactions were stopped by the addition of 0.4 ml of acetonitrile containing NPP⁺ as an internal standard. Samples were centrifuged at 3000g for 15 min, and the supernatants were analyzed for HPP⁺ formation by LC-MS/MS.

HPLC-MS/MS Assay for HPP⁺ Quantitation. HPP⁺ formation was monitored on a Sciex API model 2000 or 3000 LC-MS/MS triple quadrupole mass spectrometer (PerkinElmerSciex Instruments, Boston, MA). Standard curves containing HPP⁺ in 0.2 ml of control matrix (liver microsomes plus buffer, without cofactors) and 0.2 ml of acetonitrile containing NPP⁺ as internal standard were constructed to estimate HPP⁺ concentrations in incubation mixtures. Analytes were chromatographically separated using a Hewlett Packard Series 1100 HPLC system (Hewlett Packard, Palo Alto, CA). An autosampler was programmed to inject 20 µl on a Phenomenex Primesphere 5 µ C₁₈-HC 30 × 2.0 mm column (Phenomenex, Torrance, CA) using a mobile phase consisting of 10 mM ammonium acetate buffer-acetonitrile (60:40 v/v) containing 0.2% (v/v) triethylamine and 0.1% (v/v) acetic acid at a flow rate varying from 1 to 1.5 ml/min. Ionization was conducted in the positive ion mode at the ionspray interface temperature of 400°C, using nitrogen for nebulizing and heating gas. The ion spray voltage was 5.0 kV, and the orifice voltage was optimized at 30 eV. HPP⁺ and NPP⁺ were analyzed using multiple reaction monitoring at mass ranges m/z 354  165 and m/z 282  156, respectively. For HPP⁺, this reaction followed the protonated parent mass M⁺ = 354 to its corresponding collision-induced dissociated fragment at m/z 165, which corresponded to the 4-fluorophenyl-4-oxobutyl side chain.

Data Analysis. Enzyme kinetic analyses were performed by nonlinear regression of substrate concentration (S)  velocity (v) data using the Michaelis-Menten Equation:
Model selection was based on examination of Eadie-Hofstee-transformed plots.

	Results and Discussion

Top Abstract Introduction Experimental Procedures Results and Discussion References

Metabolism of Haloperidol in Human Liver Microsomes and Recombinant CYP3A4 and CYP3A5. Although the biotransformation pathways of HP in liver microsomes and recombinant CYP3A4 have been extensively studied, routine analysis of HP (50 µM) per human liver microsomal incubation mixtures in the present study revealed the presence of two predominant metabolites eluting at R_t = 12.3 and 13.4 min. The identity of the major metabolite (M⁺ = 354; R_t = 13.4 min) was established as HPP⁺ based on comparison of its HPLC retention time and mass spectral properties with the synthetic standard. The LC-MS/MS spectrum of the second more polar metabolite (MH⁺ = 392; R_t = 12.3 min) displayed a protonated parent ion (MH⁺) at 392, i.e., 16 mass units higher than those observed for HP consistent with a monohydroxylated-HP metabolite (M1). The collision-induced dissociation (CID) spectrum of MH⁺ 392 from M1 showed fragment ions at m/z 181 (100%) and 139 (70%) [i.e., 16 mass units higher than the fragment ions A and B observed in the corresponding CID spectrum of HP (Fig. 2)]. These data suggested that M1 is a fluorophenyl ring-hydroxylated derivative of HP, in which case the CID fragment ions at m/z 181 and 139 may be assigned to fragment ion structures C and D. Although the formation of M1 has not been disclosed in prior HP biotransformation literature, its detection in our studies is consistent with published evidence by Oida and coworkers on the detection of O-sulfate and -glucuronide conjugates derived from phase II conjugation of a fluorophenyl ring-hydroxylated metabolite of HP in human urine (Oida et al., 1989). Likewise, Van der Schyf and coworkers have also detected the 4-fluorophenyl ring-hydroxylated metabolite of HPP⁺ in mouse urine and brain tissue extracts following administration of HP or its dehydrated HPTP metabolite (Van der Schyf et al., 1994). The positional assignment of hydroxylation on the 4-fluorophenyl ring in M1, however, remains to be determined.

View larger version (11K): [in this window] [in a new window]   Fig. 2.   Collision-induced dissociation fragment ions for HP and the hydroxylated-4-fluorophenyl metabolite of HP.

Enzyme Kinetics for the Reaction Sequence HP  HPP⁺ in Human Liver Microsomes and in Recombinant CYP3A4 and CYP3A5. The contribution of CYP3A5 (relative to CYP3A4) toward HP bioactivation to HPP⁺ was assessed in human liver microsomes and in a heterologously expressed enzyme system, especially since this isozyme represents ~50% of the total hepatic CYP3A content in individuals polymorphically expressing it (Kuehl et al., 2001). HPP⁺ formation in human liver microsomes and recombinant CYP3A4 and CYP3A5 followed Michaelis-Menten kinetics (Figs. 3 and 4). The K_m values for HPP⁺ formation in human liver microsomes, recombinant CYP3A4, and CYP3A5 were 24.4 ± 8.9 µM, 18.3 ± 4.9 µM, and 200.2 ± 47.6 µM, respectively, whereas V_max values in liver microsomes, CYP3A4, and CYP3A5 were 157.6 ± 13.2 pmol/min/mg of protein, 10.4 ± 0.6 pmol/min/pmol P450, and 5.16 ± 0.6 pmol/min/pmol P450, respectively. These K_m values are several orders of magnitude higher than the typical therapeutic plasma concentrations of HP (5-25 ng/ml or 13-66 nM total plasma concentration that represents an unbound plasma concentration range of 1-5 nM) (Javaid et al., 1996) suggesting that hepatic metabolism via this pathway is not expected to be saturated in vivo. Examination of the enzyme kinetic parameters for CYP3A4 and 3A5 indicates that the CYP3A4 K_m is an order of magnitude lower than the CYP3A5 K_m, and the CYP3A4 V_max is twice that estimated for CYP3A5, resulting in an intrinsic clearance for CYP3A4 that is 22 times greater than that calculated for CYP3A5. The significantly lower intrinsic clearance of CYP3A5 for this metabolic pathway relative to CYP3A4 suggests that CYP3A5 may only play a minor role in HPP⁺ formation in human liver microsomes, with CYP3A4 being the major contributor. This is consistent with the similarity in K_m values for this metabolic pathway determined in human liver microsomal and recombinant CYP3A4 incubations and suggests that polymorphic differences in the CYP3A5 gene may not be important contributors to the interindividual and interracial differences in HP clearance as well as its bioactivation to the potentially neurotoxic pyridinium species. In this context, the effect of polymorphic CYP3A4 expression and its subsequent effects on HP pharmacokinetics remain to be determined. For instance, recent reports have described three genetic variants of the CYP3A4 gene including CYP3A4*1B, CYP3A4*2, and CYP3A4*3. The allelic frequency for the CYP3A4*1B allele, which contains an A(290)G substitution in the promoter region of CYP3A4, ranges from 0% in Chinese and Japanese Americans to >54% in African Americans; American and European Caucasians were reported to have an allelic frequency of ~4-5% (Sata et al., 2000; Lamba et al., 2002).

View larger version (8K): [in this window] [in a new window]   Fig. 3.   Kinetic analysis of the metabolism of HP to HPP+ in cDNA-expressed CYP3A4 (left panel) and CYP3A5 (right panel). Symbols are experimental data points. Lines are fitted functions described by a Michaelis-Menten equation.

View larger version (12K): [in this window] [in a new window]   Fig. 4.   Kinetic analysis of the metabolism of HP to HPP+ in human liver microsomes. Symbols are experimental data points. Lines are fitted functions described by a Michaelis-Menten equation.

Effect of Antidepressants on the CYP3A-Catalyzed HP Bioactivation. The ability of antidepressants and their major metabolites to inhibit HP bioactivation was investigated in human liver microsomes and recombinant CYP3A4 and CYP3A5. HP concentrations in these incubations reflected its approximate K_m (30 µM for human liver microsomes and recombinant CYP3A4 and 200 µM for recombinant CYP3A5). The use of a single substrate concentration around K_m was rationalized on the basis of the relationship between the IC₅₀ and K_i as described by the Cheng-Prusoff equation (Craig, 1993). Thus at K_m, the IC₅₀ value is equal to K_i for competitive inhibition as has been previously documented for the inhibition of CYP3A4 activity by antidepressants (von Moltke et al., 1999). Under these conditions, ketoconazole inhibited the reaction sequence HP  HPP⁺ in human liver microsomes and recombinant CYP3A4 and CYP3A5 with IC₅₀ values of 0.02, 0.02, and 0.19 µM, respectively (Table 1).

View larger version (13K): [in this window] [in a new window]   Fig. 5.   Inhibition of HP bioactivation to HPP+ by nefazodone in human liver microsomes (closed circles), cDNA-expressed CYP3A4 (open circles), and CYP3A5 (open squares). Each value represents an average from two separate experiments.