Abstract
CP-93,393 [(7S,9aS)-1-(2-pyrimidin-2-yloctahydropyrido[1,2-a] pyrazin-7-ylmethyl)pyrrolidine-2,5-dione] is a new anxiolytic drug with highly selective serotonin 5-hydroxytryptamine 1A autoreceptor agonist, α2-adrenergic antagonist, and dopamine D2 agonist properties. The excretion, biotransformation, and pharmacokinetics of CP-93,393 were investigated in six healthy male volunteers after oral administration of a 5-mg dose of [14C]CP-93,393. The administered radioactivity was excreted predominantly in the urine. One week after administration of the dose, cumulative excretion amounted to 67.8 ± 2.5% in the urine and 22.0 ± 5.6% in the feces. In total, 89.8 ± 5.7% of the radioactive dose was recovered in urine and feces. Mean maximum plasma concentration values for unchanged CP-93,393 were 10.92 and 1.02 ng/ml for poor metabolizers (PMs) and extensive metabolizers (EMs) of dextromethorphan, respectively. AUC0–∞ values for unchanged CP-93,393 were also greater for PMs than for EMs, whereas the mean maximum plasma concentration and AUC0–∞ values for total radioactivity were similar for the two phenotypes. Less than 0.5% of the dose was excreted in urine as unchanged drug for both EMs and PMs, suggesting extensive metabolism of CP-93,393 in both phenotypes. Hydroxylation at the 5-position of the pyrimidine ring was identified as the main metabolic pathway. 5-Hydroxy-CP-93,393 (M-15) and its glucuronide and sulfate conjugates (M-7 and M-13, respectively) accounted for ∼51% of the administered dose in excreta of both PMs and EMs. Hydrolysis of the succinimide ring, in combination with 5-hydroxylation and/or conjugation or not, accounted for ∼9% of the dose. A novel metabolite, apparently resulting from oxidative degradation of the pyrimidine ring, was characterized as the amidine analog M-18. M-15 (47–62%), its sulfate conjugate (M-13, ∼9%), and the pyrimidine ring-cleaved product (M-18, 7–13%) were identified as the major circulating metabolites for both EMs and PMs. Therefore, CP-93,393 undergoes metabolism by three primary pathways, i.e. 1) aromatic hydroxylation followed by conjugation with glucuronic acid and sulfuric acid, 2) oxidative degradation of the pyrimidine ring, and 3) hydrolysis of the succinimide ring. The identified metabolites accounted for approximately 90, 91, and 92% of the total radioactivity present in urine, plasma, and feces, respectively. The major in vivo oxidative metabolites were also observed after in vitro incubations with human liver microsomes.
CP-93,393 [(7S,9aS)-1-(2-pyrimidin-2-yloctahydropyrido[1,2-a] pyrazin-7-ylmethyl)pyrrolidine-2,5-dione] (fig. 1) is a new anxiolytic drug currently undergoing clinical evaluation (Schmidt et al., 1995; Reynolds et al., 1995). It exhibits highly selective serotonin 5-hydroxytryptamine 1A autoreceptor agonist activity and potent antagonist activity at human α2-adrenergic receptors (Seymour et al., 1995; Schmidt et al., 1996). Preclinical and clinical pharmacological profiles of CP-93,393 suggest that it should be effective in the treatment of generalized anxiety disorder and major depressive disorder, without exhibiting the adverse side effects associated with benzodiazepines (Weisenburger et al., 1996;Rollema et al., 1996).
The metabolism and excretion of CP-93,393 in rats and monkeys after administration of a single oral dose have been reported recently (Prakash and Soliman, 1997; Prakash and Cui, 1997). These studies have shown that CP-93,393 is rapidly and completely absorbed in these species. In addition, extensive first-pass metabolism of CP-93,393 occurs after oral administration, resulting in low bioavailability (Baxter J et al., unpublished observations). CP-93,393 undergoes metabolism through oxidative pathways followed by conjugation, yielding metabolites that are eliminated primarily in bile in rats (Prakash and Soliman, 1997) and in urine in monkeys (Prakash and Cui, 1997). The main metabolic pathway involves hydroxylation at the 5-position of the pyrimidine ring in both rats and monkeys.In vitro metabolism studies using hepatic microsomes from rats, monkeys, and humans also suggest that CP-93,393 is mainly metabolized by hydroxylation at the pyrimidine ring (Luther et al., 1996).
Human pharmacokinetic studies using single and multiple dosing have shown considerable intersubject variability (Miceli J et al., unpublished observations). This variability is largely the result of genetically determined metabolic pathways (Balant and Gex-Fabry, 1994). Indeed, in vitro studies using CYP2D61-selective inhibitors and recombinant enzyme suggest that the metabolism of CP-93,393 is rapid and mainly (>80%) mediated by CYP2D6 (Baxter J and Soliman V, unpublished observations), an isoform that exhibits genetic polymorphism. The objectives of this study were to determine the excretion routes for CP-93,393 and its metabolites in humans, to evaluate the metabolic profiles in urine, feces, and plasma, and to compare the metabolites of humans with those of rats and monkeys. The large interindividual variability in systemic exposure to CP-93,393 in humans after oral dosing and the involvement of the CYP2D6 isoform in the metabolism of CP-93,393 prompted investigation of the routes of excretion and metabolism, in both EMs and PMs, of dextromethorphan after administration of a single oral dose of 5 mg containing ∼94.5 μCi of [14C]CP-93,393. The metabolites were characterized by LC/MS/MS and, where possible, the proposed structures were confirmed by comparisons of their HPLC retention times and MS spectra with those of synthetic standards and/or metabolites obtained from in vitro incubations.
Materials and Methods
General Chemicals.
Commercially obtained chemicals and solvents were of HPLC or analytical grade. DL-Isocitric acid and isocitric dehydrogenase were obtained from Sigma Chemical Co. (St. Louis, MO). YMC basic columns were obtained from YMC (Wilmington, NC). Ecolite-(+) scintillation cocktail was obtained from ICN (Irvine, CA). Carbosorb and Permafluor V scintillation cocktails were purchased from Packard Instrument Co. (Downers Grove, IL). Triethylamine, methanesulfonyl chloride, succinimide, 2-methyl-2-thiopseudourea sulfate, and hexafluoroacetylacetone were obtained from Aldrich Chemical Co. (Milwaukee, WI). Silica gel, Woelm basic alumina, and TLC plates were purchased from Fisher Scientific (Springfield, NJ). Sephadex LH-20 resin was obtained from Pharmacia, Inc. (Columbus, OH).
Radiolabeled Drug and Reference Compounds.
[14C]CP-93,393 mono-HCl salt (labeled at the C2- and C3-positions of the succinimide ring), 5-hydroxy-CP-93,393, and CP-146,725 [(7S,9aS)-7-hydroxymethyl-2-(tert-butoxycarbonyl)octahydro-1H-pyrido[1,2-a]pyrazine] were synthesized at Pfizer Central Research (Groton, CT). [14C]CP-93,393 showed a specific activity of 6.22 mCi/mmol (18.9 μCi/mg) and a radiochemical purity of ≥98%, as determined by HPLC using an in-line radioactivity detector. NPMSA and 5-NHPMSA were prepared using published procedures (Shih and Rankin, 1989).
Synthesis of M-18 [(7S,9aS)-7-(2,5-Dioxopyrrolidin-1-ylmethyl)octahydropyrido[1,2-a]pyrazine-2-carboxamidine].
M-18 was synthesized in four steps starting from CP-146,725 (1, fig. 2) (Urban, 1998). In step I, methanesulfonyl chloride (2.24 ml, 29 mmol) was added to an ice bath-chilled solution of 1 (7.10 g, 26 mmol) and triethylamine (4.53 ml, 33 mmol) in methylene chloride (75 ml). The reaction was then stirred at ambient temperature for 3 hr, before quenching with 150 ml of saturated aqueous sodium bicarbonate. The resulting mixture was then extracted with three 100-ml portions of methylene chloride. The combined organic extracts were dried (with anhydrous sodium sulfate) and concentrated in vacuo to afford the mesylate derivative 2 as an amorphous solid (which was used in the next step without further purification) [TLC RF 0.8 (3% methanol/methylene chloride)].
Step II.
To a well-stirred suspension of sodium hydride (60% mineral oil dispersion, 2.60 g, 6.5 mmol) in anhydrous dimethylformamide (60 ml), succinimide (5.15 g, 5.2 mmol) was added in portions over several minutes. The mixture was then stirred for 10 min at ambient temperature and heated at 80–90°C for 1 hr. A solution of the mesylate2 (8.64 g, 26 mmol, prepared as described above) in anhydrous dimethylformamide (10 ml) was added all at once. The resulting mixture was heated at 90°C for 18 hr, cooled, and then partitioned between saturated aqueous sodium bicarbonate and ethyl acetate (100 ml of each). The separated aqueous phase was extracted with three 60-ml portions of ethyl acetate. The combined organic extracts were dried (with anhydrous sodium sulfate) and concentratedin vacuo to an oil (10.1 g). Flash chromatography (silica gel; 47–61-μm mesh; elution gradient, 30% ethyl acetate/hexanes to 100% ethyl acetate) afforded product 3 (4.44 g) as an amorphous solid [TLC RF 0.44 (70% ethyl acetate/hexanes)].
Step III.
A solution of product 3 (4.44 g) in trifluoroacetic acid (13.5 ml) was stirred for 4 hr at ambient temperature. Saturated aqueous sodium carbonate was added to elevate the pH to 9.1. The mixture was diluted with saturated sodium chloride solution and then extracted with five 75-ml portions of methylene chloride. The combined organic extracts were then dried (with anhydrous sodium sulfate) and concentrated in vacuo to afford product 4 as an amorphous foam (1.80 g), which was used in the next step without further purification.
Step IV.
Sodium acetate (5.88 g, 7.2 mmol) and 2-methyl-2-thiopseudourea sulfate (8.92 g, 32 mmol) were added to a well-stirred, ice bath-chilled solution of product 4 (1.80 g, 7.2 mmol) in isopropanol (110 ml). The reaction mixture was heated at 75°C for 6 hr and then stirred at ambient temperature for 18 hr. During the reaction period, the very volatile methyl mercaptan byproduct (with a stench) was effectively captured/quenched by means of a 5% sodium hypochlorite (Chlorox bleach) trap. After the reaction mixture had been thoroughly purged of methyl mercaptan (with a nitrogen stream), the solvent was removed in vacuo. The resulting solid residue was purified by gravity chromatography, first with a column packed with Woelm basic alumina (activity 1; elution with methanol/methylene chloride, 1:1, v/v) and then with a Sephadex LH-20 resin-packed column (elution with water). Evaporation of the solvent in vacuo affordedM-18 (460 mg, 22% yield) as an amorphous colorless solid [MS: m/z 294 (MH+);13C-NMR (Bruker AM300, CD3OD): δ 180.3, 157.8, 61.7, 58.3, 54.9, 51.4, 46.6, 41.3, 33.4, 29.1, 25.5, and 25.1].
Subjects and Dose Administration.
Six normal healthy male subjects (four subjects phenotyped as EMs and two subjects phenotyped as PMs, based on dextromethorphan metabolic ratios) between the ages of 18 and 45 years participated in the study. Their metabolic ratios were defined as the ratio of the recovery of dextromethorphan to that of dextrorphan in 0–8-hr urine samples after oral administration of 30 mg of dextromethorphan (Funck-Brentanoet al., 1992; Coutts, 1994). The metabolic ratios for the EMs were 0.040, 0.029, 0.007, and 0.007 and for the PMs were 2.971 and 2.242. The study protocol was approved by the Hennepin County Medical Center Medical Specialty Center (Minneapolis, MN) Institutional Review Board. After being informed of the purpose, design, and potential risks of the study, the volunteers gave written consent. They were confined to the Clinical Research Facility under continuous medical observation for 12 hr before dosing and until 168 hr after dosing. All subjects had fasted for at least 8 hr and were given a single 5-mg dose of [14C]CP-93,393 (∼94.5 μCi/subject). The dose was administered in an open fashion as a single oral dose in the morning, followed by a standard meal 4 hr later. The dosing solution was prepared by dissolving the radiolabeled material in deionized water. All subjects were required to refrain from lying down, eating, and drinking caffeinated beverages during the first 4 hr after drug administration.
Sample Collection.
After dosing, urine samples were collected for 7 days, at 0–4, 4–12, 12–24, 24–48, 48–96, 96–120, 120–144, and 144–168 hr after the dose. The total volumes of the urine samples were recorded after each collection. Urine samples were also collected from all subjects before dosing and were used as controls. Feces were collected as passed, from the time of dosing until at least 168 hr after dosing.
Blood sufficient to provide a minimum of 4 ml of plasma was collected, in evacuated heparinized tubes, from each subject at 0 (just before dosing), 0.5, 1, 1.5, 2, 4, 6, 8, 12, 18, 24, 48, 72, 96, 120, 144, and 168 hr after the dose. Within 1 hr after collection, the blood samples were centrifuged in a refrigerated centrifuge. Plasma was separated from whole blood, transferred into labeled plastic tubes, and immediately frozen. All biological samples were kept at −20°C until analyzed.
Determination of Radioactivity.
The radioactivity in urine and plasma was determined by liquid scintillation counting. Aliquots of plasma and urine (200–500 μl, in triplicate) for each sampling time were mixed with 5 ml of Ecolite-(+) scintillation cocktail and counted in a Packard 2500 TR liquid scintillation counter. Fecal samples were placed in Stomacher 400 bags and homogenized in equal amounts of water to a thick slurry, using a Stomacher 400 Lab Blender (Cooke Laboratory Products, Alexandria, VA). Aliquots (200–400 mg) of the homogenates were air-dried overnight and combusted using a United Packard model 306 oxidizer. The radioactivity in the combustion products was determined by trapping the liberated CO2 in Carbosorb and Permafluor V scintillation cocktails, followed by liquid scintillation counting. Combustion efficiency was determined by combustion of14C-labeled standards in an identical manner. The samples obtained at 0 hr after the dose were used as controls and counted to obtain background count rate.
The radioactivity in urine and feces at each sampling time was defined as the percentage of the dose excreted in the matrices at that sampling time. The amount of radioactivity in plasma was expressed as nanogram-equivalents of parent drug per milliliter and was calculated by using the specific activity of the dose administered.
Pharmacokinetic Analysis.
Plasma concentrations of unchanged CP-93,393 were determined with a standard HPLC/MS/MS assay (Avery and Fouda, 1996). Pharmacokinetic parameters were determined using noncompartmental methods (Rowland and Tozer, 1989). Cmax values for CP-93,393 or total radioactivity (parent drug equivalents) were estimated directly from the experimental data. Kel values were estimated using least-squares regression analysis of the plasma concentration-time data obtained during the terminal logarithmic-linear phase. Half-lives were calculated as 0.693/Kel. AUC0–t values (where t is the last time with a measurable concentration) were estimated using linear trapezoidal approximations. AUCt–∞values were estimated asCest/Kel, whereCest represents the estimated concentration at time t, based on the aforementioned regression analysis. AUC0–∞ values were estimated as the sum of AUC0–t and AUCt–∞ values.
Extraction of Metabolites from Biological Samples.
Urine samples obtained at 0–4, 4–12, and 12–24 hr after the dose were pooled. The pooled urine (∼7 ml) from each subject was centrifuged, and the supernatant was transferred to a clean tube and concentrated with a Speed Vac evaporator (Savant Instruments, Holbrook, NY). The residue was dissolved in 1 ml of acetonitrile/20 mM ammonium acetate (1:9), and an aliquot (100 μl) was injected into the HPLC system without further purification.
The homogenates from fecal samples obtained from the subjects at 0–24, 24–48, and 48–72 hr after the dose were pooled, and the pooled samples (∼4 g) were diluted with methanol (20 ml). The suspension was stirred overnight on a magnetic stirrer and centrifuged. The supernatants were combined, and small aliquots were counted. The recovery of radioactivity from the feces was 60–70%. The organic solvent was evaporated using a Speed Vac evaporator, and the residue was further purified by solid-phase extraction. The residue was resuspended in acetate buffer (pH 5.0) and was applied to a preconditioned C18 Bond Elut column (Waters Associate, Milford, MA). The column was washed with water, and the metabolites were eluted with methanol. The methanol solution was evaporated to dryness, and the residue was dissolved in 200 μl of mobile phase. An aliquot (80 μl) was analyzed by LC/MS.
Plasma (5 ml from the 0–24-hr pool and 0.5 ml from each time point) was diluted with 20 ml of acetonitrile, and the precipitated protein was removed by centrifugation. The pellet was washed with an additional 2 ml of acetonitrile, and the two supernatants were combined. The supernatant was concentrated with a Speed Vac concentrator, and the residue was reconstituted in 100 μl of acetonitrile/20 mM ammonium acetate (1:9). An aliquot (80 μl) was injected into the HPLC system.
Derivatization.
Aliquots (100 μl) of the concentrated urine samples were suspended in 500 μl of saturated sodium bicarbonate solution. Hexafluoroacetylacetone (50 μl) in toluene (500 μl) was then added, and the reaction mixture was heated at about 90°C for 2 hr (Malcolm and Marten, 1976). The solvents were evaporated with a Speed Vac evaporator, and the residue was reconstituted in mobile phase and analyzed by LC/MS.
In Vitro Incubations.
Human liver microsomes were prepared by differential centrifugation using published procedures (Tweedie and Burke, 1983). Protein concentrations were measured by the bicinchoninic acid assay, using crystalline bovine serum albumin as the standard (Smith et al., 1985), and CYP concentrations were determined by the method of Omura and Sato (1964). The incubation mixture (final volume, 1 ml) contained liver microsomes (0.5 μM CYP), 20 μM [14C]CP-93,393, 100 mM potassium phosphate buffer (pH 7.4), and an NADPH-generating system (9 mM MgCl2, 0.54 mM NADP+, 6.2 mM DL-isocitric acid, and 0.5 unit/ml isocitric dehydrogenase). The reactions were initiated by the addition of the NADPH-regenerating system, after a 5-min preincubation period. Reactions were incubated at 37°C for 30 min and were stopped by the addition of methanol (2 ml). The mixture was vortex-mixed and centrifuged. The methanol solution was evaporated and reconstituted for analysis by HPLC.
HPLC.
HPLC was conducted with a system consisting of a Rheodyne injector for manual injections, a LDC/Milton Roy (Riviera Beach, FL) constametric CM4100 gradient pump, a Waters (Milford, MA) Lambda-Max model 481 UV detector, a β-RAM (INUS, Tampa, FL), and a SP4200 computing integrator (LDC/Milton Roy, Riviera Beach, FL). Chromatography was performed on a YMC basic HPLC column (4.6 mm × 250 mm, 5 μm) with a binary mixture of 20 mM ammonium acetate (solvent A) and acetonitrile (solvent B). The mobile phase was initially composed of 95% solvent A/5% solvent B for 10 min. It was then linearly programmed to 60% solvent A/40% solvent B over 25 min. Chromatography was conducted under isocratic conditions for 10 min, and then the system was programmed back to the starting solvent mixture over 10 min. The system was allowed to equilibrate for approximately 10 min before the next injection was made.
Quantitative Assessment of Metabolite Excretion.
Quantification of the metabolites was carried out by measuring radioactivity in the individual HPLC-separated peaks using a β-RAM. The β-RAM provided an integrated printout in counts per minute and percentage of radiolabeled material, as well as peak representation. The β-RAM was operated in the homogeneous liquid scintillation counting mode, with addition of 4 ml/min Ecolite-(+) scintillation cocktail to the effluent after UV detection. The radiochromatograms of metabolites in plasma were generated by collecting fractions at 20-sec intervals and counting the fractions in a Packard 2200CA liquid scintillation counter. The retention times of the radioactive peaks, where possible, were compared with those of synthetic standards and/or metabolites obtained from rat or monkey urine (Prakash and Soliman, 1997; Prakash and Cui, 1997).
LC/MS.
LC/MS was conducted with a Perkin-Elmer SCIEX (Toronto, Ontario, Canada) API III Plus triple-stage quadrupole instrument equipped with an ion-spray interface. The effluent from the HPLC column was split, and about 50 μl/min was introduced into the atmospheric ionization source. The remaining effluent was directed into the flow cell of the β-RAM. The β-RAM response was recorded in real time by the mass spectrometer, which provided simultaneous detection of radioactivity and MS data. The delay in response between the two detectors was about 0.2 min, with the mass spectrometer response being recorded first. The ion-spray interface was operated at 6000 V, and the mass spectrometer was operated in the positive ion mode. CAD studies were performed using argon gas at a collision energy of 25 eV and a collision gas thickness of 2.5 × 1014molecules/cm2. Data were processed with a Quadra 950 Macintosh computer operating with Mac-Spec software (Perkin-Elmer SCIEX).
Results
Mass Balance.
As shown in table 1, after a single oral dose of [14C]CP-93,393 to human subjects, the administered radioactivity was eliminated predominantly in the urine. Cumulative excretion during the 168-hr collection period amounted to 67.8 ± 2.5% in the urine and 22.0 ± 5.6% in the feces. The total recovery in urine and feces averaged 88.1% for EMs and 93.1% for PMs. Of the total radioactivity recovered in urine, about 71% was excreted in the first 24 hr after the dose. Because feces were collected after natural defecation, the fecal elimination of radioactivity was delayed, compared with that in urine. The major portion of radioactivity in feces appeared during the 24–72-hr time period after the dose, for both EMs and PMs. No discernible difference in radioactivity excretion patterns was observed between the EMs and PMs.
Pharmacokinetics. CP-93,393.
Mean plasma concentration-time curves for CP-93,393 in EMs and PMs are shown in fig. 3. Calculated pharmacokinetic parameters for the unchanged drug are presented in table 2. Cmaxvalues for parent drug were attained at 0.5 hr in all EMs and at 4 and 6 hr for the PMs. On average, Cmax values were 10-fold greater in PM subjects, compared with EMs, with respective mean values of 10.9 and 1.0 ng/ml. A greater difference in overall systemic exposure was observed between PMs and EMs based on an AUC0–∞ comparison. Mean AUC0–∞ values were approximately 40-fold greater in PMs, compared with EMs, with respective values of 87.3 and 2.4 ng·hr/ml. Among the four EM subjects,Cmax values varied about 3-fold; AUC0–∞ values varied about 7-fold. These parameters varied <3-fold for the PM subjects. As expected, terminal-phase half-lives were, on average, approximately 4-fold longer in PMs, with PM and EM values of 4.1 and 0.9 hr, respectively.
Total Radioactivity.
Mean plasma concentration-time curves for total radioactivity in EMs and PMs are shown in fig. 4. Calculated pharmacokinetic parameters for total radioactivity are presented in table 2. Relative to parent drug, the concentrations of total radioactivity peaked slightly later in EMs and at similar times in PMs, with mean values of 1.9 and 6 hr, respectively. The difference in exposure to total radioactivity between PMs and EMs was significantly less, compared with parent drug, with respective meanCmax values of 39.2 and 52.7 ng-eq/ml and mean AUC0–∞ values of 1046.4 and 778.5 ng-eq·hr/ml. Elimination of total radioactivity was similar between PMs and EMs, with mean values of 19.2 and 19.6 hr, respectively.
Metabolite Profiles in Biological Samples. Urine.
Representative HPLC chromatograms, recorded with an in-line radioactivity detector, for urine from two human subjects (one PM and one EM) after oral administration of [14C]CP-93,393 are shown in fig.5. The percentages of metabolites excreted in urine are presented in table3 for both EMs and PMs. There were no qualitative differences in the urinary metabolic profiles of EMs and PMs. Less than 0.5% of the dose was present as unchanged CP-93, 393 in the urine. Five metabolites, M-6, M-7,M-13, M-15, and M-18, were the main urinary metabolites for both EMs and PMs. Small amounts of other two metabolites, M-4 and M-9, were also detected.
Feces.
As in urine, there were no significant qualitative differences in the profiles of metabolites between PM and EM subjects (data not shown). Most of the radioactivity in feces was from metabolites, and no parent drug was detected. Three metabolites, M-6, M-15, and M-18, were identified in feces of both EMs and PMs. The percentages of fecal metabolites are presented in table 3.
Plasma.
There were only minor qualitative differences in the metabolic profiles of PMs and EMs. However, there was a notable difference in the relative amounts of unchanged drug in the plasma of EMs and PMs. Unchanged CP-93,393 accounted for 15–18% of total circulating radioactivity in the PMs but <3% in EMs (table 4). Six metabolites, M-7, M-9, M-13,M-15, M-18, and M-19, along with CP-93,393, were identified in plasma.
Identification of Metabolites. Metabolite M-4 .
The full-scan mass spectrum of M-4 revealed a protonated molecular ion at m/z 380, 50 amu higher than the parent drug, indicating the addition of two oxygen atoms and a molecule of water. The CAD product-ion spectrum (MS/MS) of the ion atm/z 380 showed high-abundance ions at m/z 138 [[OH(C4H2N2)NHCH2CH2]+], 152 [[OH(C4H2N2)NHCH2CH2CH2]+], 247 [(MH−NH2COCH2CHOHCOOH)+], 264 [(MH-COCH2CHOHCOOH+H)+], and 306 [[MH−·C(OH)COOH]+]. The MS/MS spectrum and the HPLC retention time for M-4 were the same as those for M-4 obtained from rat and monkey urine (Prakash and Soliman, 1997; Prakash and Cui, 1997). M-4 was thus identified as (7S,9aS)-1-[2-(5-hydroxypyrimidin-2-yl)octahydropyrido[1,2-a]pyrazin-7-ylmethyl]-2-hydroxysuccinamic acid.
Metabolite M-6.
The full-scan mass spectrum of M-6 revealed a protonated molecular ion at m/z 364, 34 amu greater than the parent drug, indicating the addition of an oxygen atom and a molecule of water. The CAD product-ion spectrum of the ion at m/z 364 showed prominent and significant ions at m/z 138 [[OH(C4H2N2) NHCH2CH2]+] and 152 [[OH(C4H2N2)NHCH2CH2CH2]+], suggesting hydroxylation on the pyrimidine ring. The other ion, atm/z 247 [(MH−NH2COCH2CH2COOH)+], indicated that the addition of water had occurred at the succinimide ring. The MS/MS spectrum and the HPLC retention time for M-6were the same as those for synthetic 5-NHPMSA. M-6 was thus identified as 5-NHPMSA.
Metabolite M-7.
The full-scan mass spectrum of M-7 revealed a protonated molecular ion at m/z 522, 176 amu greater than hydroxylated CP-93,393, indicating that this metabolite was a conjugate of the oxidative metabolite. The CAD product-ion spectrum of the ion atm/z 522 showed an abundant ion at m/z 346,i.e. loss of 176 amu (dehydroglucuronic acid) from the protonated molecular ion, suggesting that the metabolite was a glucuronide conjugate. The MS/MS spectrum and the HPLC retention time for M-7 were the same as those for M-7 obtained from rat and monkey urine (Prakash and Soliman, 1997; Prakash and Cui, 1997). Based on these data, M-7 was identified as the glucuronide conjugate of 5-hydroxy-CP-93,393.
Metabolite M-18.
The full-scan mass spectrum of M-18 revealed a protonated molecular ion at m/z 294, lower than that of the parent drug, suggesting that this metabolite was a cleaved product. The protonated molecular ion was 42 amu greater than despyrimidinyl-CP-93,393, which suggested the addition of an acetyl or carboxamidino group to despyrimidinyl-CP-93,393. The CAD product-ion spectrum of M-18 showed ions at m/z 252, 235, 209, and 195 (fig. 6A). The prominent ions at m/z 195 [[MH−N(C⋕NHNH2) CH2CH2CH2]+], 209 [[MH−N(C⋕NHNH2)CH2CH2]+], and 235 [[MH−NH2(C⋕NHNH2)]+] suggested that the succinimide portion of the molecule was unsubstituted. Treatment of the urine samples with hexafluoroacetylacetone resulted in the disappearance of the peak coinciding with metabolite M-18 and the appearance of a new peak, which had an HPLC retention time of 38.8 min. The mass spectrum displayed a protonated molecular ion at m/z 466, 172 amu greater than the underivatized compound, indicating addition of a hexafluoroacetylacetone moiety and subsequent loss of two molecules of water. The MS/MS spectrum and the HPLC retention time forM-18 were the same as those for the synthetic amidine (fig.6B). On the basis of these data, M-18 was identified as 7-(2,5-dioxopyrrolidin-1-ylmethyl)octahydropyrido[1,2-a]pyrazine-2-carboxamidine.
Metabolite M-9.
The full-scan mass spectrum of M-9 revealed a protonated molecular ion at m/z 348, 18 amu greater than the parent drug, suggesting that the molecule had undergone hydrolysis. The MS/MS spectrum and the HPLC retention time for M-9 were the same as those for synthetic NPMSA. M-9 was thus identified as NPMSA.
Metabolite M-13.
The full-scan mass spectrum of M-13 revealed a protonated molecular ion at m/z 426, 80 amu greater than hydroxylated CP-93,393, indicating that the metabolite was a conjugate. The CAD product-ion spectrum of the ion at m/z 426 showed an abundant ion at m/z 346, i.e. loss of 80 amu (SO3) from the protonated molecular ion, suggesting that the metabolite was a sulfate conjugate. The MS/MS spectrum and the HPLC retention time for M-13 were the same as those for M-13 obtained from rat and monkey urine (Prakash and Soliman, 1997; Prakash and Cui, 1997). Based on these data, M-13 was identified as the sulfate conjugate of 5-hydroxy-CP-93,393.
Metabolite M-15.
The full-scan mass spectrum of M-15 revealed a protonated molecular ion at m/z 346, 16 amu greater than the parent drug, suggesting that the metabolite was a mono-oxidation product of CP-93,393. The CAD product-ion spectrum of M-15showed ions atm/z 138 [[OH(C4H2N2)NHCH2CH2]+], 152 [[OH(C4H2N2) NHCH2CH2CH2]+], 209 [[MH−OH(C4H2N2)NCH2CH2]+], and 235 [[MH−OH(C4H2N2)NH2]+].M-15 had the same HPLC retention time as synthetic 5-hydroxy-CP-93,393 and showed an identical CAD spectrum.M-15 was thus identified as 5-hydroxy-CP-93,393.
Metabolite M-19 .
M-19 was detected only in plasma. It showed a protonated molecular ion at m/z 540. The MS/MS spectrum ofm/z 540 showed a fragment ion at m/z 364,i.e. loss of 176 amu (a characteristic loss for glucuronides) from the protonated molecular ion, indicating thatM-19 was a glucuronic acid conjugate. The fragment ion atm/z 364, 34 amu greater than the parent drug, was indicative of the addition of an oxygen atom and a molecule of water to the parent molecule. The prominent fragment ion at m/z 138 indicated that the oxidation had occurred at the pyrimidine ring. Based on these data, M-19 was tentatively identified as the glucuronide conjugate of 5-NHPMSA (M-6).
In Vitro Metabolites.
CP-93,393 (M-17) and a total of four metabolites (M-6, M-9, M-15, and M-18) were identified in the incubation mixture of CP-93,393 with human liver microsomes. The structures of metabolites were elucidated by LC/MS/MS analysis and confirmed by comparisons of HPLC retention times with those of synthetic standards. Thus, all major oxidative metabolites formed in vivo, including the pyrimidine-cleaved productM-18, were also observed in vitro.
Discussion
In the present study, the metabolism and excretion of CP-93,393 were investigated after oral administration of a single 5-mg dose of [14C]CP-93,393 to six normal healthy male volunteers (four subjects phenotyped as EMs and two subjects phenotyped as PMs, based on their dextromethorphan metabolic ratios) (Funck-Brentano et al., 1992; Coutts, 1994). Excretion of the dose was essentially complete (∼88%) within 96 hr of dose administration, with the majority of the dose being excreted within 24 hr of dosing. There was no discernible difference in the overall dose excretion patterns between EMs and PMs. Because the drug was not administered via the iv route, accurate assessment of the extent of absorption in humans was not possible. However, excretion of the major portion of the administered radioactive dose in urine (69%) and the absence of unchanged drug in feces supported the hypothesis that, as in rats and monkeys, the extent of absorption was high in humans (Prakash and Soliman, 1997; Prakash and Cui, 1997).
No significant difference was observed between EMs and PMs in the exposure to total radioactivity, suggesting that the amount of CP-93,393 absorbed did not differ among these subjects. However, significantly (6-fold) higher CP-93,393 concentrations were observed in PM subjects. This is indicative of diminished first-pass metabolism in these subjects and supports in vitro observations that CP-93,393 metabolism is strongly dependent on CYP2D6. The apparent diminished ability to metabolize CP-93,393 in PMs also led to a longer half-life (4.1 vs. 0.9 hr) in these subjects and a prolongation of the time of first occurrence ofCmax for CP-93,393, because this parameter is a function of the rates of both absorption and elimination processes.
Extensive biotransformation of CP-93,393 occurs in both EMs and PMs, as suggested by AUC0–∞ values for unchanged drug corresponding to <8% of the corresponding AUC0–∞ values for total radioactivity. Furthermore, only a small percentage of the administered dose (<0.5%) was excreted as unchanged CP-93,393 in urine. The major portion of the administered radioactivity recovered in urine was excreted as conjugates of the hydroxylated metabolites. There were no significant qualitative or quantitative differences in the profiles of urinary and fecal metabolites in EMs and PMs. In addition to CP-93,393, a total of seven metabolites in urine, three metabolites in feces, and six metabolites in plasma were identified by ion-spray LC/MS/MS (Prakashet al. 1997). A plausible scheme for the biotransformation pathways of CP-93,393 in humans is shown in fig.7.
A major pathway for metabolism of CP-93,393 observed in rats (Prakash and Soliman, 1997) and monkeys (Prakash and Cui, 1997) involved oxidation at the 5-position of the pyrimidine ring to form 5-hydroxy-CP-93,393 (M-15). This pathway was also found to be the major pathway for metabolism in humans. MetaboliteM-15, along with its glucuronide and sulfate conjugates (M-7 and M-13, respectively) accounted for ∼51% of the dose in both PMs and EMs. Hydroxylation at the pyrimidine ring was previously observed for other drugs containing a pyrimidinyl-piperazine ring structure (Jajoo et al., 1989a,b). The other metabolic pathways of CP-93,393 were the result of hydrolysis of the succinimide ring (M-4, M-6, andM-9) and oxidative degradation of the pyrimidine ring (M-18). Hydrolysis of the succinimide ring was observed for both monkeys and rats, but oxidative degradation of the pyrimidine ring was observed only in monkeys (Prakash and Cui, 1997). Hydroxylation of the succinimide ring, one of the major metabolic pathways in rats, was not detected in humans (Prakash and Soliman, 1997).
Oxidative degradation of the pyrimidine ring resulted in the formation of metabolite M-18. Similar splitting of an imidazolidine ring (Koch et al., 1989; Fujimaki et al., 1989) and, recently, cleavage of a pyridine ring have also been reported (Chang et al., 1997). The structure of M-18 was assigned by the interpretation of its mass spectrum and by chemical derivatization. Treatment of urine samples with hexafluoroacetylacetone yielded a cyclic product, suggesting the presence of an amidine moiety (Malcolm and Marten, 1976). To further confirm the structure ofM-18, a synthetic standard of M-18 was prepared (fig. 2). M-18 showed chromatographic and spectral properties similar to those of the synthetic standard. These data supported the hypothesis that M-18 was formed by oxidative degradation of the pyrimidine ring. The mechanism for the formation ofM-18 is not known at this time.
The metabolic profiles of plasma samples from both EMs and PMs showed only minor qualitative differences. However, the amount of circulating unchanged drug was ∼6 times greater in the PMs than in the EMs, suggesting slower metabolism of CP-93,393 in PMs. 5-Hydroxy-CP-93,393 (M-15, 47–62%), its sulfate conjugate (M-13, ∼9%), and despyrimidinylamidine (M-18, 7–13%) were identified as major circulating metabolites in both EMs and PMs. The glucuronide conjugate (M-7, 4.9%) was present only in PMs. However, a smaller amount of metabolite M-19, the hydrolysis product of M-7, was present in the plasma of both EMs and PMs. The other metabolite, M-9, the hydrolysis product of CP-93,393, was also present in smaller amounts in both EMs and PMs.
Preliminary in vitro studies have suggested that CP-93,393 is mainly (>80%) metabolized by CYP2D6 (Baxter J and Soliman V, unpublished observations), a major drug-metabolizing enzyme that exhibits genetic polymorphism (Meyer and Zanger, 1997). Approximately 7–10% of the Caucasian population shows an inherited deficiency in this enzyme resulting from the presence of one or several mutant alleles at the CYP2D6 gene locus (Hanioka et al., 1990). These subjects are characterized by the PM phenotype. Compared with normal (or EM) subjects, PM subjects demonstrate markedly greater AUC values for parent drugs that are metabolized by CYP2D6 and therefore require lower doses to achieve therapeutic effects (Kivisto and Kroemer, 1997). In this study, we observed higher plasma concentrations and AUC values for unchanged CP-93,393 in the PMs, compared with the EM subjects, thus providing additional evidence that CP-93,393 is a substrate for CYP2D6. Because the majority of the excreted and circulating radioactivity was attributable to metabolites and the metabolic profiles were similar for the EMs and the PMs, the involvement of other CYP isoforms in the metabolism of CP-93,393 could not be ruled out. Further studies will be required to address this issue.
In summary, CP-93,393 is well absorbed in humans after oral administration, and the radioactive dose is eliminated mainly by renal excretion. The compound undergoes extensive metabolism in both EMs and PMs, by three oxidative metabolic pathways, i.e.hydroxylation at the pyrimidine ring, hydrolysis of the succinimide ring, and oxidative degradation of the pyrimidine ring. Although similarities existed in the nature of metabolites being formed in the EMs and PMs, there were significant differences in the plasma concentrations of the unchanged drug. Therefore, it can be concluded that the metabolism of CP-93,393 cosegregates with theO-demethylation of dextromethorphan, a probe substrate for monitoring CYP2D6 activity.
Acknowledgments
We thank Drs. Keith McCarthy and Kathleen Zandi for providing radiolabeled CP-93,393, Dr. Charles Halstenson, Hennepin County Medical Center Medical Specialty Center (Minneapolis, MN), for conducting the dosing and sample collection, Beth Obach, Victor Soliman, and Karen Coffman for technical assistance, and Drs. Hassan Fouda and Robert Ronfeld for valuable suggestions.
Footnotes
-
Send reprint requests to: Chandra Prakash, Ph.D., Department of Drug Metabolism, Central Research Division, Pfizer Inc., Groton, CT 06340.
- Abbreviations used are::
- CYP
- cytochrome P450
- EM
- extensive metabolizer
- PM
- poor metabolizer
- NPMSA
- (7S,9aS)-1-(2-pyrimidin-2-yloctahydropyrido[1,2-a]pyrazin-7-ylmethyl)succinamic acid
- 5-NHPMSA
- (7S,9aS)-1-[2-(5-hydroxypyrimidin-2-yl)octahydropyrido[1,2-a]pyrazin-7-ylmethyl]succinamic acid
- β-RAM
- radioactivity monitor
- CAD
- collisionally activated dissociation
- Cmax
- maximum plasma concentration
- Kel
- terminal-phase rate constant
- Received October 15, 1997.
- Accepted January 12, 1998.
- The American Society for Pharmacology and Experimental Therapeutics