Abstract
The mechanism behind the observed inactivation of human P450 2B6 by phencyclidine (PCP) has been evaluated over the past 2 decades. The scope of the current investigation was to contribute to the fundamental knowledge of PCP oxidation and perhaps the mechanism behind P450 inactivation. To study the chemistry of PCP oxidation, we subjected PCP to the Fenton reagent. Under Fenton chemistry conditions, oxidation on all three PCP rings was observed by liquid chromatography/tandem mass spectrometry (LC-MS/MS). When PCP was incubated with the Fenton system in the presence of glutathione (GSH), three GSH-PCP conjugates were identified. Subsequent LC-MS/MS analysis of these conjugates revealed two species that had GSH attached to the cyclohexane ring of PCP and a third conjugate in which GSH was adducted to the piperidine ring. When PCP was incubated across a panel of P450 enzymes, several enzymes, including P450s 2D6 and 3A4, were able to catalyze the formation of the PCP iminium ion, whereas P450s 2B6 and 2C19 were exclusively able to hydroxylate secondary carbons on the cyclohexane ring of PCP. Subsequent mechanistic experiments revealed that only P450s 2B6 and 2C19 demonstrated loss of catalytic activity after preincubation with 10 μM PCP. Finally, investigation of P450 2B6 inactivation using structural analogs of PCP revealed that blocking the para-carbon atom on the cyclohexane ring of PCP from oxidation protected the P450 2B6 from inactivation, which suggests that a reactive intermediate generated during the hydroxylation of the cyclohexane ring may be linked to the mechanism of inactivation of P450 2B6 by PCP.
The biotransformation of phencyclidine, PCP (Fig. 1), by liver proteins was shown to result in the formation of reactive metabolites leading to P450 inactivation and the formation of covalent adducts with hepatic macromolecules more than 20 years ago (Castagnoli et al., 1997). Originally, covalent binding and enzyme inactivation was thought to occur via alkylation of protein-based nucleophiles by an iminium ion metabolite, M4 (Fig. 1), that arose from the P450-dependent two-electron α-carbon oxidation of PCP (Sayre et al., 1997). This hypothesis was supported by the formation of PCP iminium-cyanide adducts and the observation that the presence of cyanide in microsomal incubations protected against the formation of covalent adducts with proteins (Ward et al., 1982b; Hoag et al., 1984). However, subsequent studies demonstrated that the P450 inactivation by the PCP iminium ion (PCP-Im) required the presence of NADPH (Hoag et al., 1984; Osawa and Coon, 1989; Sayre et al., 1991), which inferred that metabolism of PCP beyond the iminium species was required for P450 inactivation.
Later, the PCP mechanism-based inactivation of P450 2B6, using purified, recombinant human P450 enzyme, was investigated in great detail (Jushchyshyn et al., 2003). Through a series of experiments, it was shown that P450 2B6 was inactivated via covalent modification of the P450 2B6 apoprotein. Moreover, inclusion of cyanide in the purified P450 2B6 incubation mixture did not protect the enzyme from inactivation. This observation supported the notion that perhaps something other than the PCP iminium ion was involved in P450 2B6 inactivation. Additional evidence toward this conclusion was obtained when coincubation of PCP with glutathione (GSH) did not protect P450 2B6 from inactivation. These observations led to speculation that perhaps the chemical species that leads to nonselective microsomal protein binding (e.g., PCP iminium species) is unique from a potential short-lived, active site-bound, reactive intermediate, which results in P450 2B6 inactivation.
To investigate the probability than an alternative mechanism may be responsible for PCP inactivation of P450 2B6, we used several lines of interrogation. First, Fenton chemistry, used for the production of hydroxyl radicals, was coupled with GSH trapping to mimic the oxidation of PCP and to identify reactive intermediates in the absence of enzyme. Second, reaction phenotyping experiments allowed for the identification of unique PCP metabolites formed by P450s 2B6 and 2C19, and an inhibition screen revealed that these were the only two enzymes inhibited by PCP. Finally, the time-dependent inhibition of P450 2B6 by PCP analogs was compared. The results of these studies suggest a mechanism for the inactivation of P450 2B6 by PCP that does not involve the initial formation of the PCP iminium ion.
Materials and Methods
The eleven recombinant human cytochromes P450 (1A2, 2A6, 2B6, 2C8, 2C9, 2C19, 2D6, 2E1, 3A4, 3A5, and 4A11 SUPERSOMES), 3-cyano-7-ethyoxycoumarin (CEC), 7-methoxy-4-trifluoromethylcoumarin (MFC), 7-benzyloxy-trifluoromethylcoumarin (BFC), and 3-[2-(N,N diethyl-N-methylamino)ethyl]-7-methoxy-4-methylcoumarin (AMMC) were purchased from BD Biosciences (Woburn, MA). The 7-ethoxy-4-trifluoromethylcoumarin (EFC) was purchased from Molecular Probes (Eugene, OR). Glutathione, PCP, 1-(1-(phenyl-d5)-cyclohexyl)piperidine (d5-PCP), NADPH, and ferrous chloride were all purchased from Sigma (St. Louis, MO). Thirty-percent hydrogen peroxide was a JT Baker solvent (Phillipsburg, NJ). Authentic standards for PCP metabolites M1–M3 and M5 (Fig. 1) were obtained from within the Pfizer chemical archives (Pfizer Inc., Groton, CT). The PCP ketone derivatives, 4-phenyl-4-piperidinecyclohexanone and 1-(1-phenylcyclohexyl)piperidin-4-one, were also obtained from the Pfizer chemical archives.
Fenton Reactions. A typical Fenton reaction contained 100 μM PCP, 500 μM Fe(II)Cl2, 500 μM EDTA, and 3% H2O2 in water for a final volume of 150 μl. All reactions were incubated in borosilicate test tubes at room temperature. Two to 5 min after the addition of H2O2, the sample was diluted 3:1 with 60 mM GSH in water. Reactions performed with d5-PCP contained an equimolar concentration of nondeuterated PCP. Reaction samples were analyzed directly for PCP metabolites and GSH-PCP adducts by LC-ESI-MS/MS with no further work-up.
In Vitro Metabolism of PCP by Recombinant P450s. Each incubation mixture contained 100 pmol of recombinant cytochrome P450 with 50 μM PCP in 50 mM phosphate buffer, pH 7.4, in a final volume of 300 μl. PCP stock solutions were made up in dimethylsulfoxide. The final concentration of dimethylsulfoxide in each incubation mixture was less than 1% (v/v). The mixtures were preincubated at 37°C for 5 min, and reactions were initiated by the addition of 1 mM NADPH. The reaction mixtures were incubated for another 20 min at 37°C. Reaction samples were terminated by the addition of 3% trifluoroacetic acid in water. Samples were vortex-mixed and then centrifuged at 10,000g for 8 min to precipitate the protein. The supernatant was transferred to injector vials and was directly analyzed by LC-ESI-MS/MS as described below.
P450 Inhibition Assays. The P450 enzymes that catalyzed the metabolism of PCP (2B6, 2C9, 2C19, 2D6, and 3A4) were assayed for time-dependent inactivation. In brief, two primary reaction mixtures, a control sample and an inactivation sample containing 60 pmol of P450 and 10 μM PCP in 50 mM potassium phosphate buffer (pH 7.4), for a final volume of 90 μl were equilibrated at 37°C for 5 min. The inactivation sample received 1 mM NADPH to initiate the reaction, and the control sample received an equal volume of water (4.5 μl). At 0 and 20 min after the addition of NADPH, 10 pmol of P450 was transferred from the primary reaction to a secondary reaction containing 100 μM of the marker substrate and 1 mM NADPH in 50 mM potassium phosphate buffer (pH 7.4). Secondary reaction mixtures were incubated for 10 min at 37°C before being quenched with 100 μl of acetonitrile and vortex-mixed. The final volume of the secondary reaction was 250 μl. Immediately after the 0-min transfer, the +NADPH sample received 1 mM NADPH while the control sample received an equal volume of water. Marker substrate activity of the control sample at 0 min was assigned a value of 100% activity remaining. All experiments were conducted in triplicate.
For the PCP inactivation screen, the following P450 marker substrates were used: 7-EFC (P450 2B6), 7-MFC (P4502C9), CEC (P450 2C19), AMMC (P450 2D6), and BFC (P450 3A4). All samples (200-μl aliquots) were assayed directly for marker substrate turnover using a Safire florescence plate reader (Tecan, Durham, NC). The following excitation and emission wavelengths were used to detect the products of the marker substrates: for 7-EFC, 7-MFC, and BFC: λex = 409 nm and λem = 530 nm; for CEC: λex = 409 nm and λem = 460 nm; and for AMMC: λex = 390 nm and λem = 460 nm. For all samples, the bandwidth was 12 nm, and the gain was set to 60 (arbitrary units).
LC-ESI-MS/MS Analysis. PCP metabolites and GSH adducts were chromatographically separated on a Phenomenex (Torrance, CA) C-18 column (4.6 × 150 mm, 5 μm) that had been equilibrated with 90% solvent A (100% water, 0.1% FA) and 10% solvent B (100% CH3CN, 0.1% FA) at a flow rate of 0.35 ml/min. After the injection of 20 μl of sample, solvent B was increased linearly to 60% over 15 min. Metabolites and conjugates were detected using a Thermo-Finnigan DECA-XP (Woburn, MA) mass spectrometer set in the positive ion mode. The capillary voltage and temperature were 4.5 kV and 210°C, respectively. Sheath gas and auxiliary gas were 40 and 10 (arbitrary units), respectively. The tube lens offset was set at 5 V. Data-dependent scanning was switched on, and the collision energy was 27%.
Mechanism-Based Inactivation of P450 2B6 by PCP and PCP Ketone Analogs. Four primary P450 mixtures were equilibrated at 37°C for 5 min before reaction initiation with 1 mM NADPH. Three mixtures contained 7.5 pmol of P450 2B6, 10 μM PCP, 10 μM 4-phenyl-4-piperidinecyclohexanone, or 10 μM 1-(1-phenylcyclohexyl)piperidin-4-one in 50 mM potassium phosphate buffer (pH 7.4) resulting in a final volume of 70 μl. The PCP ketone analog stock solutions were in acetonitrile. The final concentration of acetonitrile in each primary reaction was less than 3% (v/v). The fourth reaction mixture, a control, contained 7.5 pmol of P450 in 50 mM potassium phosphate buffer (pH 7.4) and 2 μl of acetonitrile. After the mixtures were equilibrated, 10-μl aliquots (1 pmol of P450) were transferred to secondary reaction mixtures. The secondary reaction mixtures contained 100 μM 7-EFC and 1 mM NADPH in 50 mM potassium phosphate buffer (pH 7.4) for a final volume of 250 μl. These secondary reaction mixtures were incubated at 37°C for 10 min and were quenched with 150 μl of acetonitrile. Immediately after the initial transfer to the secondary reaction mixtures, the primary reaction mixtures were initiated with the addition of 1 mM NADPH. Aliquots were transferred from the primary to the secondary reaction mixtures at the time points indicated. The secondary reaction mixtures were assayed for the oxidation of 7-EFC to HFC as described above. 7-EFC O-deethylase activity of the control sample at 0 min was assigned a value of 100% activity remaining. All experiments were performed in triplicate.
Results
Oxidation of PCP by the Fenton Reagent. When PCP was subjected to chemical oxidation by the Fenton reagent, hydroxylated products (MH+ at m/z 260) were observed (data not shown). Of the products, three (M1–M3) were identical to the microsomal derived metabolites, since they possessed identical HPLC retention times and MS fragmentation with that of authentic PCP metabolite standards.
The identity of the remaining oxidative products reflected hydroxylation of the PCP phenyl ring. One oxidation product was consistent with that of the authentic standard for para-phenol PCP (M5). A second oxidative species contained a product ion spectrum for the CID of MH+m/z 260, which indicated that the phenyl-cyclohexane moiety of PCP was hydroxylated (Fig. 2A). However, additional LC/MS studies using d5-PCP yielded a peak with MH+ (m/z 264), indicating a loss of one deuterium from the PCP phenyl ring, and the MS/MS for the corresponding species further supported a phenol metabolite (Fig. 2B). Interestingly, for all Fenton oxidation reactions, the PCP iminium ion was not detected.
Fenton Reagent Generated PCP-GSH Adducts. To trap potential reactive intermediates of PCP formed via the Fenton reaction, 60 mM GSH was added to the reaction mixtures after the initiation of catalysis with H2O2. From these incubations, three PCP-GSH adducts having a MH+ of m/z 549 were observed. The product ion spectra for one conjugate is shown in Fig. 3A. The fragment ion at m/z 464 indicated that GSH was adducted to the phenyl-cyclohexane ring moiety. To reduce the ambiguity of the site of GSH adduction, a d5-PCP-GSH conjugate (MH+ of m/z 554) that coeluted with these products was inspected. The product ion spectrum for these species (Fig. 3B) exhibited no loss of deuterium, indicating that the GSH was adducted to the cyclohexane ring. The product ion spectrum for the other GSH-PCP species (Fig. 3C) supported that of a piperidine GSH adduct.
Metabolic Profiles of PCP by Recombinant Human P450s. Eleven recombinant human P450 enzymes were screened for the ability to metabolize PCP. Of the eleven P450 enzymes tested, P450s 1A2, 2A6, 2E1, and 4A11 were devoid of any detectable PCP metabolite formation. P450s 2B6 and 2C19 both catalyzed the oxidation of PCP to form metabolites M1 through M4, whereas P450s 2C9 and 2C8, albeit at a reduced level, were capable of forming only M3 and M4. Finally, P450s 2D6, 3A5, and 3A4 exclusively catalyzed the formation of M4, the PCP iminium ion. The information depicted in Fig. 4 details the chemical structure of PCP and summarizes the relative ability of each of the P450 enzymes to catalyze the formation of PCP metabolites. In all microsomal systems tested, M5 (PCP phenolic metabolite) was not detected.
Mechanism-Based Inactivation of P450s by PCP. Of the P450 isoforms that metabolized PCP, a select panel based upon isoform-specific metabolic profiles were assayed for potential time-dependent enzyme inactivation in the presence of 10 μM PCP over a 20-min preincubation period. Results of this study revealed that P450 2B6 lost approximately 40% catalytic activity over the 20-min preincubation period, whereas P450s 2C9, 2D6, and 3A4 exhibited no measurable no loss of activity in the presence of PCP (Fig. 5). Interestingly, P450 2C19, the only other P450 that possessed M1 and M2 oxidase activity, was inactivated approximately 20% by 10 μM PCP. Control samples that did not receive NADPH showed no significant loss of marker substrate activity over the 20-min incubation period (data not shown).
Mechanism-Based Inactivation of P450 2B6 by PCP Ketone Analogs. To determine whether carbon hydroxylation on the para position of the cyclohexane or piperidine ring was required for P450 2B6 inactivation, recombinant human P450 2B6 was incubated with 10 μM PCP or two PCP ketone analogs. For the PCP analogs, the ketone moieties were located on either the para-carbon of the cyclohexane ring or the para-carbon of the piperidine ring (Fig. 6A). Figure 6B shows that over a 30-min time period, the rates of P450 2B6 inactivation by PCP and the PCP piperidine ketone analog were similar. However, the PCP cyclohexane ketone analog showed a marked decreased rate of inactivation compared with PCP. A control mixture incubated with acetonitrile in the absence of NADPH did not lose activity over the 30-min assay. In addition, when either PCP ketone analog was incubated with P450 2B6, similar amounts of the corresponding hydroxyl metabolites were detected confirming that these compounds were able to enter the P450 2B6 active site and that they were substrates for this enzyme (data not shown).
Discussion
The α-carbon oxidation of cyclic tertiary amines by cytochromes P450 to yield toxic products is a well studied metabolic pathway (Castagnoli et al., 1997). For years, researchers have postulated that P450-mediated α-carbon oxidation of the PCP piperidine ring to form the corresponding iminium ion was responsible for the observed to P450 inactivation and covalent modification of microsomal macromolecules (Osawa et al., 1995). This belief was based upon early experimental evidence that demonstrated that cyanide, which is known to trap the PCP-Im, could attenuate the metabolism-dependent covalent binding of PCP to hepatic microsomal proteins (Ward et al., 1982b). However, subsequent studies with synthesized PCP-Im revealed that the PCP iminium intermediate was not directly involved in P450 inactivation and that further oxidation of the PCP iminium species was required for P450 inactivation (Castagnoli et al., 1997). As a consequence, a PCP dihydropyridinium species was proposed as the ultimate reactive species responsible for P450 inactivation (Castagnoli et al., 1997). However, when the mechanism-based inactivation of P450 2B6 was investigated using purified, recombinant enzyme, cyanide failed to protect the enzyme from inactivation, which suggests that PCP-Im was not involved in the inactivation of P450 2B6 (Jushchyshyn et al., 2003). Intrigued by the potential that an alternative explanation for P450 2B6 inactivation by PCP was possible, we sought out to reexamine the mechanism that accounts for phencyclidine inactivation of P450 2B6.
Trapping with nucleophiles (such as GSH and N-acetyl cysteine) is a commonly used technique for the identification of reactive metabolites formed as a result of P450 oxidation (Chen et al., 2002; Tang et al., 2005; Shebley et al., 2006). In many cases, the generated adduct is useful for identifying the site and chemical mechanism of a reactive intermediate (Chen et al., 2002). In the case of P450 2B6 and PCP, trapping experiments with GSH did not protect the enzyme from inactivation (Jushchyshyn et al., 2003). However, these experiments did reveal that the stoichiometry of PCP binding to inactivated P450 2B6 was reduced from 5:1 to 1:1 in the presence of GSH (Jushchyshyn et al., 2003). This result provides the basis for the current hypothesis that PCP electrophiles released from the P450 2B6 active site and trapped by GSH did not represent the ultimate reactive species responsible for P450 2B6 inactivation. In this light, although GSH may be able to trap reactive intermediates of PCP that escape from the active site, it would seem that an in vitro GSH-trapping study would provide negligible information regarding the identity of the true P450-inactivating species. To this end, we used a chemical, P450 biomimetic system that might serve to emulate the P450-catalyzed oxidative metabolism of PCP and generate reactive intermediates that may reflect species that exist only in the enzyme active site.
The Fenton reagent has been previously used to mimic the oxidative biotransformation of various P450 substrates and other molecules (Groves and McClusky, 1976; Zbaida et al., 1988; Heur et al., 1989; Zbaida and Kariv, 1990; Maldotti et al., 1996; Reis et al., 2003). Oxidation of organic compounds by Fenton's reagent proceeds via an initial one-electron reduction of hydrogen peroxide to yield free hydroxyl radicals and subsequent hydrogen abstraction from the substrate. The resulting carbon-centered radical is then rapidly oxidized by Fe(III) (Wardman and Candeias, 1996). Inclusion of spin-trapping agents such as 5,5-dimethyl-1-pyrroline N-oxide in Fenton reactions has allowed for the unequivocal identification of secondary carbon-centered radical intermediates (O'Neill et al., 2000; Reis et al., 2003, 2004). Therefore, it was reasoned that the Fenton reagent would be a useful tool in elucidating the identities of highly reactive substrate intermediates that may not otherwise been detected in conventional microsomal experiments.
In our hands, the Fenton reagent did not oxidize the piperidine α-carbon to form the PCP iminium ion; however, it was able to catalyze the secondary carbon hydroxylation of PCP to form microsomal-like products M1, M2, and M3. In addition, hydroxylation of the phenyl ring to form M5 and a second unknown phenol metabolite also seemed to be significant routes of PCP oxidation by the Fenton system. Previously, Masumoto et al. (1989) also observed formation of M3 and PCP phenol products when they exposed PCP to the Fenton reagent. The fact that PCP phenolic products were formed with the Fenton system and not produced in microsomal incubations is not considered a major discrepancy for our purposes. In the Fenton oxidation system, there are no enzymatic constraints and free hydroxyl radicals can add readily to unsaturated, aromatic moieties so it was expected that phenolic products would be formed in the Fenton-mediated PCP oxidation. However, for PCP oxidation by P450 2B6, chemical reactivity and active site topography govern the site of enzymatic oxidation. Homology modeling and quantitative structure-activity relationship investigations have shown that hydrophobic moieties on the substrate are critical for substrate binding and orientation in the P450 2B6 active site (Holsztynska and Domino, 1985; Lewis et al., 1999; Wang and Halpert, 2002). In particular, the aromatic rings of several model substrates appeared to form π-stacking associations, with a phenylalanine residue located within the P450 2B6 active site.
When GSH was included in the Fenton reaction mixture, three GSH-PCP adducts were observed having an MH+ of m/z 549, which equates to a species composed of one PCP molecule plus one GSH molecule. Two conjugates had product ion spectra that indicated the GSH was adducted to the phenyl-cyclohexane ring fragment. The MS/MS of the corresponding d5-PCP-GSH adducts narrowed down the location of the GSH molecule to the cyclohexane ring moiety. The third GSH-PCP conjugate seemed to be a species in which GSH was attached to the piperidine ring. Because the Fenton reagent system was incapable of causing the formation of PCP-Im, this adduct must have been the result of the piperidine ring hydroxylation pathway. The same would be true for the GSH adduct on the cyclohexane ring. This result suggested that in an oxidative system, the initial formation of the PCP-Im is not required for PCP to be converted to reactive species. As described previously, the hydroxylation of aliphatic compounds by the Fenton reagent proceeds via hydrogen-abstraction, resulting in a short-lived carbon centered radical or cation intermediate. Because the GSH-PCP conjugates did not show desaturation or addition of oxygen, we postulate that GSH trapped PCP carbon-centered radical/cation intermediates.
To link metabolism to P450 2B6 inactivation, the oxidative metabolism of PCP was evaluated across a panel of eleven recombinant human P450s. From this survey, a total of seven P450 isoforms (in particular, P450 3A4) were found to form the PCP-Im (M4), which substantiates earlier findings that this metabolite represents a significant pathway for human PCP hepatic metabolism (Ward et al., 1982a,b; Holsztynska and Domino, 1985). Moreover, P450s 2D6 and 3A4, although able to produce PCP-Im, did not exhibit inactivation when incubated with 10 μM PCP. Interestingly, only P450s 2B6 and 2C19 were able to hydroxylate secondary carbon atoms on PCP to form metabolites M1, M2, and M3. In addition, these were the only two P450s that demonstrated time-dependent inactivation when incubated with 10 μM PCP for 20 min. These two observations when viewed in aggregate provide a compelling rationale to support the notion that PCP secondary carbon hydroxylation pathways, rather than formation of the PCP-Im, could be the mechanistic basis for P450 2B6 inactivation.
The mechanism of P450-mediated aliphatic oxidation involves initial H-abstraction from the alkane carbon to form a carbon-centered radical. This is followed by radical rebound resulting in the formation and release of the alcohol metabolite (Atkinson and Ingold, 1993; Shaik et al., 2002). The carbon-centered radical is very short-lived and is rapidly hydroxylated and released from the active site as a metabolite much faster than the radical itself can escape from the hydrophobic enzyme active site (Heur et al., 1989; Bowry et al., 1990; Rota et al., 1997). This scenario reflects the metabolism of PCP by P450 2B6 and accounts for the partition ratio for P450 2B6 inactivation by PCP as being 45 (Jushchyshyn et al., 2003), which suggests that the majority of PCP secondary carbons are oxidized and released from the active site as hydroxylated metabolites (M1 and M2). However, a small fraction of the PCP carbon-centered radicals, as with the Fenton chemistry, may have the opportunity to undergo single-electron transfer oxidation to form a cation that could readily react with proximal nucleophilic residues at or near the active site of the enzyme, which results in covalent binding to the apoprotein and ultimately P450 2B6 inactivation (Fig. 7).
To further investigate the possibility that pathways leading to either piperidine ring and cyclohexane ring hydroxylation were responsible for P450 2B6 inactivation, we examined the time-dependent loss of P450 2B6 activity in the presence of two PCP ketone analogs. For the PCP analogs, the ketone moieties were located at either the para-carbon on the cyclohexane ring or the para-carbon on the piperidine ring, which essentially blocked these sites from P450-mediated activation. The cyclohexanone PCP analog showed a marked decrease in the rate of P450 2B6 inactivation compared with native PCP, whereas the piperidine ketone analog exhibited a similar rate of inactivation as PCP. Restated, blocking the para-carbon of the cyclohexane ring from oxidation protected P450 2B6 from inactivation and suggests that the pathway leading to C-hydroxylation at this site is involved in the inactivation of P450 2B6 as depicted in Fig. 7. A similar hypothesis was proposed for the inactivation of P450 1A by tacrine (Peng et al., 2004). In this study, the authors postulated that a carbon-centered radical or cation intermediate lies on the pathway to secondary carbon hydroxylation and may be responsible for P450 1A inactivation.
Based upon existing PCP data in the literature and three new lines of experimental evidence, 1) GSH-trapping experiments using the Fenton system showed that both the cyclohexane and piperidine rings could be oxidized to yield reactive intermediates; 2) reaction phenotyping showed that the P450s (e.g., 2B6 and 2C19) that were inhibited by PCP uniquely catalyzed the formation of secondary carbon hydroxyl metabolites and that only P450s 2B6 and 2C19 were susceptible to time-dependent inactivation; and 3) blocking the para-carbon site on the PCP cyclohexane ring preserves P450 2B6 catalytic activity compared with PCP. In summary, we propose that pathways leading to secondary carbon hydroxylation on the PCP cyclohexane ring are responsible in human P450 2B6 inactivation.
Acknowledgments
We thank Dan Rock, J. Matthew Hutzler, and Barclay Shilliday (Pfizer, St. Louis, MO) for helpful discussions.
Footnotes
-
Article, publication date, and citation information can be found at http://dmd.aspetjournals.org.
-
doi:10.1124/dmd.106.010579.
-
ABBREVIATIONS: PCP, phencyclidine; P450, cytochrome P450; PCP-Im, phencyclidine iminiun ion; GSH, glutathione; LC, liquid chromatography; ESI, electrospray ionization; MS/MS, tandem mass spectrometry; CEC, 3-cyano-7-ethoxycoumarin; MFC, 7-methoxy-4-trifluoromethylcoumarin; BFC, 7-benzyloxy-trifluoromethylcoumarin; AMMC, 3-[2-(N, N-diethyl-N-methylamino)ethyl]-7-methoxy-4-methylcoumarin; 7-EFC, 7-ethoxy-4-trifluoromethylcoumarin; d5-PCP, 1-(1-(phenyl-d5)-cyclohexyl)piperidine; MS, mass spectrometry; CID, collision-induced dissociation.
- Received April 14, 2006.
- Accepted June 12, 2006.
- The American Society for Pharmacology and Experimental Therapeutics