Abstract
Vabicaserin is a potent 5-hydroxytryptamine2C agonist that is currently being developed for the treatment of the psychotic symptoms of schizophrenia. In this study, in vitro and in vivo metabolism of vabicaserin was evaluated in mice, rats, dogs, monkeys, and humans, and the structures of the metabolites were characterized by liquid chromatography/mass spectrometry and NMR spectroscopy. Vabicaserin underwent three major metabolic pathways in vitro: NADPH-dependent hydroxylation, NADPH-independent imine formation, and carbamoyl glucuronidation. After a single oral dose, vabicaserin was extensively metabolized in animals and humans, and its metabolites were mainly excreted via the urine in mice and rats. Along with the metabolites observed in vitro, secondary metabolism via oxidation and conjugation of the primary metabolites generated from the above-mentioned three pathways yielded a number of additional metabolites in vivo. Carbamoyl glucuronidation was the major metabolic pathway in humans but a minor pathway in rats. Although carbamoyl glucuronidation was a major metabolic pathway in mice, dogs, and monkeys, oxidative metabolism was also extensive in these species. Hydroxylation occurred in all species, although different regional selectivity was apparent. The imine pathway also appeared to be common to several species, because vabicaserin imine was observed in humans and hydroxyl imine metabolites were observed in mice, rats, and dogs. A nitrone metabolite of vabicaserin was observed in dogs and humans but not in other species. In conclusion, the major metabolic pathways for vabicaserin in humans and nonclinical safety species include carbamoyl glucuronidation, hydroxylation, formation of an imine, and a nitrone.
Introduction
Recent studies have demonstrated that 5-hydroxytryptamine (5-HT)2C agonists decrease levels of dopamine in the prefrontal cortex and nucleus accumbens, brain regions that are thought to mediate the critical effects of antipsychotic drugs (Millan et al., 1998; DiMatteo et al., 1999; DiGiovanni et al., 2000). In contrast, 5-HT2C agonists do not decrease dopamine levels in the striatum, the brain region most closely associated with extrapyramidal side effects (Millan et al., 1998; DiMatteo et al., 1999). In addition, a recent study demonstrated that 5-HT2C agonists decrease firing in the ventral tegmental area but not in the substantia nigra (DiMatteo et al., 1999; DiGiovanni et al., 2000). The differential effects of 5-HT2C agonists in the mesolimbic pathway relative to the nigrostriatal pathway suggest that 5-HT2C agonists have the potential to treat psychotic symptoms with lower liability for the extrapyramidal side effects associated with typical antipsychotics.
Vabicaserin is a potent 5-HT2C full agonist and demonstrates in vitro functional selectivity for 5-HT2C over 5-HT2A and 5-HT2B receptors (Dunlop et al., 2006). Vabicaserin is effective in several animal models that are predictive of antipsychotic activity, with an atypical antipsychotic profile (Marquis et al., 2006). Administration of vabicaserin decreases nucleus accumbens dopamine levels without affecting striatal dopamine, which is indicative of mesolimbic selectivity. This profile is consistent with potential efficacy in the treatment of the psychotic symptoms of schizophrenia with decreased liability for extrapyramidal side effects. In addition, chronic administration of vabicaserin significantly decreases the number of spontaneously active mesocorticolimbic dopamine neurons, without affecting nigrostriatal dopamine neurons, consistent with the effects of atypical antipsychotics.
Formation of a carbamoyl glucuronide (CG) was the major metabolic pathway in human liver microsomes with bicarbonate buffer and a CO2-enriched environment in the presence of both NADPH and uridine 5′-diphosphoglucuronic acid trisodium salt (UDPGA) (Tong et al., 2010). After a single oral dose to healthy human volunteers, the CG was the predominant metabolite in human plasma and urine, with average CG-to-vabicaserin concentration ratios ranging from 8 to 74 in plasma and 96 to 537 in urine (Tong et al., 2010). However, oxidative metabolism also appeared to be occurring in vitro in liver microsomes (Tong et al., 2010). Although, the in vivo formation of oxidative metabolites was evident from their presence in plasma and urine, the structures of those metabolites were previously not established due to limited availability of human biological samples (Tong et al., 2010). Therefore, the purpose of the current study was to generate and isolate sufficient amounts of these oxidative metabolites, elucidate the structures of the major vabicaserin metabolites by liquid chromatography/mass spectrometry (LC/MS) and NMR spectroscopy, determine mass balance in the safety species, and compare in vitro and in vivo metabolite profiles of vabicaserin in humans and nonclinical safety species.
Materials and Methods
Materials.
[14C]Vabicaserin hydrochloride {(9aR,12aS)-4,5,6,7,9,9a, 10,11, 2,12a-decahydro-cyclopenta[c][1,4]diazepino[6,7,1-ij]quinoline} was synthesized by the Radiosynthesis Group, Chemical Development, Wyeth (now Pfizer, Pearl River, NY). The radiochemical purity of [14C]vabicaserin was 98.9% and the chemical purity was 99.9% by UV detection. The specific activity of the [14C]vabicaserin was 222.9 μCi/mg as a hydrochloride salt. Nonlabeled vabicaserin hydrochloride with a chemical purity of 98.6% was synthesized by Wyeth (Pearl River, NY). Vabicaserin CG (WAY-280107) was synthesized by the Chemical Development at Wyeth (Montreal, Quebec) and had a purity of 95.5%. The chemical structures of [14C]vabicaserin and its CG are shown in Fig. 1. Liver microsomes listed in Table 1 from CD-1 mice, Sprague-Dawley rats, beagle dogs, and cynomolgus monkeys were obtained from In Vitro Technologies (Baltimore, MD). Pooled human liver microsomes from 10 subjects of mixed sex were purchased from XenoTech, LLC (Lenexa, KS) (Table 1). Glucose 6-phosphate, glucose-6-phosphate dehydrogenase, NADP+, UDPGA, EDTA, 3-chloroperoxybenzoic acid (mCPBA; 77%), silica gel (230–400 mesh), and Fluka silica gel TLC plates were obtained from Sigma-Aldrich (St. Louis, MO). Acetonitrile, ethyl acetate, and methanol were high-performance liquid chromatography (HPLC) or American Chemical Society reagent grade and were purchased from EMD Chemicals (Gibbstown, NJ). Ultima Gold and Ultima Flo scintillation cocktails were purchased from PerkinElmer Life and Analytical Sciences (Waltham, MA). Dimethyl sulfoxide (DMSO)-d6 (2H, 99.96%) was obtained from Cambridge Isotope Laboratories (Andover, MA). Other chemicals of analytical grade and solvents of HPLC grade were obtained from EMD Chemicals (Gibbstown, NJ) or Mallinckrodt Baker Inc. (Phillipsburg, NJ).
Vabicaserin nitrone was synthesized for NMR spectroscopic analysis. At room temperature with stirring, a solution of mCPBA (216 mg, approximately 1 mmol) in acetonitrile (75 ml) was slowly added to a suspension of vabicaserin (230 mg, 1.0 mmol) in acetonitrile (100 ml) and kept stirring at room temperature for 1 h. Then, the reaction mixture was concentrated at room temperature using a rotary vacuum evaporator. Fractionation by silica gel flash chromatography was conducted with a gradient of ethyl acetate to ethyl acetate/methanol (6:1). The fraction containing vabicaserin nitrone was condensed and further purified on a silica gel TLC plate using a developing solvent system of ethyl acetate/methanol (6:1). The yellow band containing vabicaserin nitrone (Rf = 0.3) fluoresced under UV (365 nm) irradiation. This band was extracted using methanol and evaporated to dryness under a stream of nitrogen yielding vabicaserin nitrone (0.3 mg). LC/MS data for the synthetic material matched that obtained for P5 in human plasma (data not shown).
Incubation with Liver Microsomes.
In vitro incubations were conducted to isolate metabolites by HPLC for structure elucidation as well as to evaluate in vitro metabolite profiles in various species. In vitro incubations with liver microsomes of mouse, rat, dog, monkey, and human in the presence of NADPH and UDPGA were the same as described previously (Tong et al., 2010). In brief, [14C]vabicaserin (10 μM) was incubated for 20 min in a CO2-enriched environment with liver microsomes (0.5 mg/ml) in 0.5 ml of bicarbonate buffer (50 mM), pH 7.4, containing alamethicin (50 μg/mg protein) and magnesium chloride (10 mM), in the presence of an NADPH-regenerating system and UDPGA (2 mM). In addition, [14C]vabicaserin (10 μM) was also incubated for 20 min with liver microsomes (0.5 mg/ml) in 0.1 M phosphate buffer containing magnesium chloride (10 mM), pH 7.4, in the presence of the NADPH-regenerating system, to study the oxidative metabolism alone. The samples were prepared and analyzed as previously described (Tong et al., 2010) by HPLC with radioactivity detection for metabolite profiles and by LC/MS for metabolite identification. Vabicaserin metabolites were isolated for analysis by NMR spectroscopy.
Excretion Studies in Animals.
All animal housing and care were conducted in Association for Assessment and Accreditation of Laboratory Animal Care-accredited facilities. Animal care and use for excretion and metabolism studies were approved by the Wyeth Institutional Animal Care and Use Committee. Animal rooms were maintained on a 12-h light and dark cycle. Animals were provided food and water ad libitum.
Rats.
Four male Sprague-Dawley rats, obtained from Charles River Laboratories (Wilmington, MA), were used in this study. The animal body weights ranged from 337 to 348 g on the day of dosing. The dose vehicle contained 2% Tween 80 and 0.5% methylcellulose in water. Appropriate amounts of [14C]vabicaserin and nonlabeled vabicaserin were dissolved in the dose vehicle to give a solution. The final concentration of vabicaserin was approximately 2 mg/ml. All animals (n = 4) received a targeted single oral dose of 5 mg/kg (80.7 μCi/kg) via intragastric gavage at a volume of 2.5 ml/kg. Urine and feces were collected for 120 h from all animals after administration of [14C]vabicaserin. Urine samples were collected at ambient temperature at intervals of 0 to 8, 8 to 24, 24 to 48, 48 to 72, 72 to 96, and 96 to 120 h after dosing. Fecal samples were collected at ambient temperature at intervals of 0 to 24, 24 to 48, 48 to 72, 72 to 96, and 96 to 120 h after dosing. Cage rinses were collected at intervals of 0 to 24, 24 to 72, and 72 to 120 h by rinsing each cage with approximately 100 ml of ethanol (30%) in water.
Bile study in rats.
Four male rats weighing 387 to 411 g and four female rats weighing 291 to 325 g at the time of dosing were purchased from Charles River Laboratories, and bile-duct cannulation was performed in-house. The dose vehicle was the same as in the excretion study. On the day of dosing, [14C]vabicaserin and nonlabeled vabicaserin were dissolved in the vehicle to a final concentration of approximately 1 mg/ml. Nonfasted rats were given a single 5 mg/kg (323 μCi/kg) target dose of vabicaserin at a volume of 5.0 ml/kg via intragastric gavage. Animals were provided standard rat chow and water ad libitum and were kept in metabolism cages individually.
Bile was collected into tubes on dry ice at 0 to 4, 4 to 8, 8 to 24, and 24 to 48-h intervals after dosing. Feces and urine were collected into containers on dry ice at 0 to 24 and 24 to 48-h intervals. All biological samples and aliquots from the before and after dosing formulations were stored at approximately −70°C until analyzed.
Dogs.
Four male beagle dogs, weighing 7.6 to 9.8 kg at the time of dosing, were from an in-house colony. [14C]Vabicaserin hydrochloride and nonlabeled vabicaserin hydrochloride were dissolved in methanol and then evaporated under a nitrogen stream to dryness. Capsules (number 2) were filled with accurate amounts (approximately 130 mg) of the mixed drug substance according to animal weights to give a dosage of 15 mg/kg (39 μCi/kg). The filled gelatin capsules were then enteric-coated manually. Each dog was given one enteric-coated capsule containing [14C]vabicaserin as a hydrochloride salt. Animals were fed 2 h before dosing and provided Purina dog chow and water ad libitum and housed individually in metabolic cages.
Urine samples were collected at intervals of 0 to 8, 8 to 24, and 24 to 48 h into tubes on dry ice and at intervals of 48 to 72, 72 to 96, 96 to 120, 120 to 144, and 144 to 168 h after dosing at room temperature. Fecal samples were collected at ambient temperature at intervals of 0 to 8, 8 to 24, 24 to 48, 48 to 72, 72 to 96, 96 to 120, 120 to 144, and 144 to 168 h after dosing. Cage rinses were collected daily by rinsing each cage with approximately 250 to 1100 ml of ethanol (30%) in water.
In Vivo Metabolism.
For metabolism studies in male mice, rats, and dogs, radiolabeled doses were used. Male CD-1 mice and Sprague-Dawley rats were purchased from Charles River Laboratories. The dose vehicle for mice and rats contained 2% (w/w) Tween 80 and 0.5% methylcellulose in water. Twenty (five/time point) nonfasted male mice weighing 28 to 34 g at the time of dosing were given a single 50 mg/kg (∼300 μCi/kg) dose of vabicaserin at a volume of 20 ml/kg via intragastric gavage. Mice were kept in metabolic cages in groups of five. Twelve (three/time point) nonfasted male rats weighing approximately 320 to 350 g at the time of dosing were given a single 5 mg/kg (∼300 μCi/kg) dose of vabicaserin at a volume of 2.5 ml/kg via intragastric gavage. Rats were kept individually in metabolism cages. Dog plasma, urine, and feces for metabolite profiling were collected from the same animals used in the excretion study described earlier.
Four male cynomolgus monkeys, weighing approximately 5 to 10 kg at the time of dosing, were from an in-house colony. Nonfasted monkeys were given a single 25 mg/kg dose of nonradiolabeled vabicaserin at a volume of 2 ml/kg via intragastric gavage. The vehicle was the same as used in mice and rats. Animals were housed individually in metabolic cages.
Blood samples were collected from mice and rats at sacrifice by cardiac puncture at 2, 4, 8, and 24 h after dosing. Blood samples of approximately 3 ml from the jugular vein of dogs and from the femoral vein of monkeys were collected at 2, 4, 8, and 24 h after dosing. Potassium EDTA was used as the anticoagulant, and plasma was immediately harvested from the blood by centrifugation at 4°C. Urine samples were collected from animals at 0 to 8 and 8 to 24-h intervals for all species. Fecal samples were collected at 0 to 24 h after dosing. Urine and feces were also collected from dogs 24 to 48 h after dosing. All biological specimens were stored at approximately −70°C until analysis.
Dosing and sample collections from humans were conducted at a single investigational site (Methodist Hospital, Philadelphia, PA). Oral doses of vabicaserin capsules of 500 mg were administered to three healthy male subjects under fasting conditions. Plasma samples were collected within 2 h before test article administration (predose) and at 6, 12, and 24 h after dose for metabolite profiling. Urine specimens were collected at intervals of 0 to 4, 4 to 12, and 12 to 24 h. Samples were stored at approximately −70°C until analysis.
For human studies, the protocol and the informed consent forms were reviewed and approved by the study site institutional review board (IRB). Subsequent amendments to the protocols and/or any revisions to the informed consent forms were also reviewed and approved by the IRB. This study was conducted in accordance with ethical principles that have origins in the Declaration of Helsinki and in any amendments that were in place when the study was conducted. Written informed consent was obtained from all subjects before their enrollment.
Radioactivity Determination for Mouse, Rat, and Dog Studies.
Plasma (10 μl) and urine (100 μl) aliquots were analyzed for radioactivity concentrations. Each urine sample was mixed and triplicate aliquots (100 or 200 μl) were analyzed for radioactivity concentrations. Before analysis, cage rinses were weighed and mixed. Triplicate aliquots (100 or 200 μl) were weighed and analyzed to determine the concentration of radioactivity. Radioactivity in plasma, urine, and cage rinses were determined with a Tri-Carb model 3100 TR/LL liquid scintillation counter (LSC) (PerkinElmer Life and Analytical Sciences) using 5 ml of Ultima Gold as the scintillation fluid. Fecal samples were weighed and homogenized in water at volume-to-weight ratios of approximately 5:1, and homogenates (0.1–0.3 g) were placed on Combusto-cones with Combusto-pads and combusted. A model 307 Tri-Carb Sample Oxidizer, equipped with an Oximate-80 Robotic Automatic Sampler (PerkinElmer Life and Analytical Sciences), was used for combustion. The liberated 14CO2 was trapped with Carbo-Sorb E carbon dioxide absorber, mixed with PermaFluor E+ liquid scintillation cocktail, and counted in a Tri-Carb model 3100 TR/LL LSC (PerkinElmer Life and Analytical Sciences). The oxidation efficiency of the oxidizer was 97.7%.
Sample Preparation for Metabolite Profiling and Metabolite Identification.
Plasma samples were pooled by mixing an equal volume from each animal for each time point and processed as described previously (Tong et al., 2010). In brief, aliquots of pooled plasma were mixed with two volumes of methanol, placed on ice for approximately 5 min, and then centrifuged. The supernatant was transferred to a clean tube and evaporated at 22°C under nitrogen in a TurboVap LV (Caliper Life Sciences) to a volume of approximately 0.3 ml. The concentrated extracts were centrifuged, the supernatant volumes were measured, and duplicate 10-μl aliquots were analyzed by LSC for extraction efficiency. Extraction efficiency was determined by comparing the total radioactivity in the extracts versus that in the samples before extraction. An average of >80% of the plasma radioactivity was recovered in the extracts. The 24-h plasma samples from mice, rats, and dogs were not analyzed due to low radioactivity concentration. Urine samples were pooled in a proportional manner to give 0 to 24-h samples for mouse, rat, and dog. Pooled urine samples were analyzed without further preparation.
Fecal homogenates were pooled proportionally to their weight to give 0 to 24-h samples for mouse, rat, and dog. Aliquots of 1 g of the pooled fecal homogenate were mixed with 2 ml of methanol, placed on ice for approximately 10 min, and centrifuged. The supernatant was transferred to a clean tube. The residue was extracted three times with 2 ml of a water/methanol (3:7) mixture. The supernatants from each sample were combined, evaporated to approximately 1 ml, and centrifuged. Extraction efficiency was determined by analyzing aliquots of 10 μl of the supernatant, by comparing the total radioactivity in the extracts versus the total radioactivity in the samples before extraction. An average of >70% of the fecal radioactivity was extracted. Monkey and human fecal samples were not analyzed. Aliquots of urine and extracts of plasma and fecal samples were analyzed by HPLC with radioactivity detection described below for metabolite profiles and by LC/MS for metabolite characterization.
HPLC.
A Waters model 2695 HPLC system (Waters Corporation, Milford, MA) with a built-in autosampler was used for metabolite isolation and analysis for metabolite profiles. Separations were accomplished on a Luna C18(2) column (150 × 2.0 mm, 5 μm) (Phenomenex, Torrance, CA) coupled with a guard (4 × 2 mm) cartridge. The sample chamber of the autosampler was maintained at 4°C, whereas the column was at an ambient temperature of approximately 20°C. Radioactivity chromatograms were recorded as described above. The mobile phase consisted of 10 mM ammonium acetate (pH 4.5) (A) and methanol (B) and was delivered at 0.2 ml/min. Table 2 lists the linear gradients used for each study.
The HPLC eluent was collected at 20-s intervals into 96-well LumaPlates (PerkinElmer Life and Analytical Sciences), due to low radioactivity concentrations in mouse and dog plasma samples. The plates were dried overnight in an oven at 40°C and analyzed by the TopCount NXT radiometric microplate reader. For other radioactive animal samples and in vitro incubations, a Flo-One β Model A525 radioactivity detector with a 250 μl flow cell was used for data acquisition. The flow rate of Ultima Flow M scintillation fluid was 1 ml/min, providing a mixing ratio of scintillation cocktail to mobile phase of 5:1.
LC/MS.
For metabolite characterization, the HPLC conditions were the same as described above, except that Agilent model 1100 (Agilent Technologies, Palo Alto, CA), Waters 2695, and Acquity UPLC (Waters) liquid chromatography systems were used. UV spectra were recorded with diode array UV detectors for all analyses. Most of the LC/MS analyses were conducted with mobile phase Gradient C (Table 2). Mass spectral data for vabicaserin and its metabolites were obtained with LCQ ion trap (Thermo Fisher Scientific, Waltham, MA), quadrupole time-of-flight (Waters), and triple quadrupole (Waters) mass spectrometers. Each mass spectrometer was equipped with an electrospray ionization source and operated in the positive ionization mode. Settings for each mass spectrometer were optimized to provide a structurally relevant range of product ions from MS/MS and MSn experiments. Product ion spectra recorded with the triple quadrupole mass spectrometer using a collision energy setting of 22 eV were used as the reference spectra to confirm metabolite identifications. MassLynx (versions 3.5, 4.0, and 4.1; Waters) and Xcalibur (version 1.3; Thermo Fisher Scientific) software were used for collection and analysis of LC/MS data.
NMR Spectroscopy.
Vabicaserin metabolites P1, P2, P3, P4, P5, P6, and CG were isolated, concentrated, and repurified using the same HPLC conditions as above except that 0.1% formic acid in water was used as mobile phase A and 0.1% formic acid in methanol was used as mobile phase B. Vabicaserin metabolites P7, P9, and P10 were isolated from rat urine by HPLC with an ammonium acetate mobile phase followed by HPLC with a trifluoroacetic acid mobile phase. Each isolated metabolite was evaporated to dryness under a nitrogen stream. Vabicaserin metabolites P1, P2, P3, P4, P5, and P6 were dissolved in 500 μl of DMSO-d6 and transferred into a 5-mm NMR tube. NMR spectroscopic data were recorded at 30°C on a 600 MHz Bruker Avance III spectrometer with a 5-mm CPTCI CryoProbe (Bruker BioSpin Corporation, Billerica, MA). Proton, heteronuclear single quantum correlation (HSQC), and heteronuclear multiple-bond correlation (HMBC), experiments were performed. All chemical shifts were referenced to the DMSO signal at 1H δ 2.49 ppm and 13C δ 39.5 ppm. The proton spectra were acquired with 32,768 data points over a 6602 Hz spectral window using a 30° pulse and standard pulse sequence. Data were Fourier transformed with a 0.3 Hz line broadening window function. The two-dimensional (2D) 1H-13C HSQC spectra were acquired with 1024 data points in F2 and 256 increments in F1, with 32 scans per increment, using a phase-sensitive pulse sequence. The 2D 1H-13C HMBC spectra were acquired with 4096 data points in F2 and 256 increments in F1, with 64 scans per increment. Vabicaserin metabolites P7, P9, and P10 were dissolved in 150 μl of DMSO-d6 and transferred into a 3-mm NMR tube. One-dimensional proton NMR and 2D NMR (COSY, ROESY) data were collected on an Inova 500 MHz NMR spectrometer (Varian, Palo Alto, CA) equipped with a Nalorac 5-mm z-gradient indirect detection probe (Varian).
Results
Metabolite Characterization.
Mass spectral data for vabicaserin and its metabolites are summarized in Table 3. Based on their molecular ions and fragmentations in comparison to vabicaserin, the proposed structures of metabolites are presented in Fig. 1. Vabicaserin metabolites were isolated by HPLC from in vitro incubations (P1, P2, P3, P4, P5, P6, and CG) and from rat urine (P7, P9, and P10) for analysis by liquid chromatography/tandem mass spectrometry and NMR spectroscopy, and the structures of the aforementioned metabolites were confirmed by NMR spectroscopic data from 1H and 13C (1H-1H COSY, 1H-13C HSQC, and 1H-13C HMBC) (Table 4). HPLC retention time and mass spectral data for CG matched those for the synthetic carbamoyl glucuronide of vabicaserin.
In Vitro Metabolite Profiles.
Radiochromatographic profiles of [14C]vabicaserin (10 μM) after incubation with human liver microsomes in the absence and presence of the NADPH-regenerating system, and in the presence of both NADPH and UDPGA, are depicted in Fig. 2. When [14C]vabicaserin was incubated with human liver microsomes in phosphate buffer in the presence of NADPH, five metabolites (P1–P5) were detected. Formation of P6 required microsomes but was not NADPH-dependent. Metabolites P1, P2, P3, and P4 were characterized by LC/MS and NMR spectroscopy as hydroxyl vabicaserin, P5 as vabicaserin nitrone, and P6 as vabicaserin imine. When [14C]vabicaserin was incubated with human liver microsomes in bicarbonate buffer and a CO2-enriched environment in the presence of both NADPH and UDPGA, a new product (CG) was detected with higher abundance than any of the other aforementioned metabolites. CG was identified as vabicaserin carbamoyl glucuronide by LC/MS analysis and confirmed by comparison with the synthetic carbamoyl glucuronide.
Radiochromatographic profiles of [14C]vabicaserin after incubation at a concentration of 10 μM with mouse, rat, dog, and monkey liver microsomes in the presence of both NADPH and UDPGA are shown in Fig. 3. Although quantitative differences among the various species were apparent, the in vitro metabolite profiles of vabicaserin were qualitatively similar across the species (Figs. 2 and 3), and all human metabolites were detected in one or more animal species. Monkey liver microsomes generated similar profiles to those of humans, except that P2 was detected in relatively higher intensity, whereas P6 was observed in relatively lower intensity than in human liver microsomes. However, because the purpose of the in vitro study was to generate metabolites for structural elucidation and for qualitative profiling, the in vitro incubation conditions were not optimized for each species to provide in vitro and in vivo extrapolation.
Excretion.
Mice.
Excretion of radioactivity was determined in the samples collected up to 24 h after oral administration of [14C]vabicaserin (50 mg/kg) to male CD-1 mice in the metabolism study. Urine was the major route of excretion, with 59.6% of the radioactive dose recovered within 24 h after the dose. Fecal elimination in the first 24 h after the dose accounted for 13.6% of the dosed radioactivity.
Rats.
The excretion of radioactivity in male Sprague-Dawley rats was evaluated after a single oral (gavage) dosage of 5 mg/kg [14C]vabicaserin. The total recovery of radioactivity (feces, urine, and cage rinse) over a 5-day (120 h) period was 93.8%. The major route of excretion of radioactivity was the urine (64.3%); fecal excretion was relatively minor (28.0%). The rate of excretion of radioactivity was rapid, with 87.4% of the oral dose recovered in urine and feces (including cage rinse) in the first 24 h after dosing.
Biliary excretion in rats.
Excretion of radioactivity was also determined in bile-duct cannulated rats after a single oral dose of [14C]vabicaserin (5 mg/kg). Urine was the major excretion route for vabicaserin-related radioactivity after a single oral dose, whereas biliary excretion was relatively minor. In the first 48 h after dosing, an average of 60% of the radioactive dose was excreted in the urine, whereas an average of 17% of the administered radioactivity was excreted in the bile. Excretion of radioactivity in rats was rapid, with the majority of the radioactivity recovered in the first 24 h after dosing. These results were consistent with the finding in the mass balance study, where 64% of the administered radioactivity was recovered in urine.
Dogs.
The excretion of radioactivity in male beagle dogs was evaluated after a single oral (enteric-coated capsule) dose of [14C]vabicaserin (15 mg/kg). The total recovery of radioactivity (feces, urine, and cage rinse) over a period of 7 days (168 h) was 97.3%. Large individual variations in daily recovery were observed. The major route of excretion of radioactivity was the feces (58.5%); urinary excretion was relatively minor (32.7%). The rate of excretion of radioactivity was rapid, with 84.2% of the oral dose recovered in urine and feces (including cage rinse) in the first 48 h after dosing, which may have been due to the fact that enteric coating affected the release of vabicaserin.
In Vivo Metabolite Profiles.
Representative radiochromatographic profiles of plasma and urine samples of mice, rats, and dogs administered a single oral dose of [14C]vabicaserin are depicted in Figs. 4 to 6. Vabicaserin was extensively metabolized in all three species, with the parent drug representing less than 21% of the total plasma radioactivity. Estimated concentrations of vabicaserin and its metabolites observed in plasma of mice, rats, and dogs after oral administration of [14C]vabicaserin are summarized in Table 5. Percentages of vabicaserin and its metabolites in urine of various species as determined by their chromatographic distribution are summarized in Table 6.
In mice, vabicaserin, CG, and hydroxyl vabicaserin (P2 and P3) were the major radioactive components in plasma (Fig. 4). Vabicaserin represented no more than 1.2% of the administered radioactive dose in urine. Hydroxyl vabicaserin (P2 and P3), hydroxyl vabicaserin glucuronide (P9), didehydrohydroxyl vabicaserin glucuronide (P19), and CG were the major metabolites in mouse urine. Unchanged vabicaserin represented 18% of fecal radioactivity in samples 0 to 8 h after the dose and 11% for 8 to 24 h after the dose. Hydroxyl vabicaserin (P2 and P3), keto vabicaserin (P7), hydroxyl vabicaserin imine sulfate (P8), hydroxyl vabicaserin sulfate (P10), and hydroxyl vabicaserin imine (P14) were identified in feces. CG was not observed in mice feces, likely because of cleavage in the gastrointestinal tract.
In rats, vabicaserin was extensively metabolized to predominantly oxidative metabolites. Unchanged vabicaserin represented ≤20% of the total radioactivity in plasma and <1% of the administered radioactive dose in urine. Hydroxyl vabicaserin (P1, P2, and P3) and keto vabicaserin (P7) were the major drug-related components in rat plasma (Fig. 5). The hydroxyl metabolites (P1 and P3), keto vabicaserin (P7), and the glucuronide (P9) were the major metabolites in urine. CG was not detected in rat plasma or urine. Metabolites P3, P8, P10, and P11 and only trace amounts of vabicaserin were detected in rat feces.
In the biliary excretion and metabolite profile study in rats, vabicaserin represented an average of <3% of radioactivity in bile and 3 to 4% of radioactivity in urine. Vabicaserin and its metabolites were excreted in bile almost solely as conjugates. Biliary metabolite profiles did not change significantly over time. The major biliary metabolites included CG, hydroxyl vabicaserin imine sulfate (P8), hydroxyl vabicaserin glucuronides (P9 and P24), N-acetyl hydroxyl vabicaserin (P12), and hydroxyl vabicaserin carbamoyl glucuronide (P21). CG, which was not detected in plasma or urine, was the most abundant metabolite in bile, representing 22 to 30% of the radioactivity in bile. Urinary metabolites included hydroxyl vabicaserin (P1, P2, P3, and P4), keto vabicaserin (P7), hydroxyl vabicaserin imine sulfate (P8), hydroxyl vabicaserin sulfate (P10), hydroxyl vabicaserin glucuronide (P9), and N-acetyl hydroxyl vabicaserin (P12).
In dogs, vabicaserin was extensively metabolized after administration of an enteric-coated capsule containing [14C]vabicaserin. Oxidative metabolism was the major metabolic pathway, although formation of CG was also observed. Vabicaserin represented <21% of the radioactivity in plasma and no more than 0.5% of the administered radioactive dose in urine, similar to mice and rats. In dog plasma, vabicaserin, CG, P1, P2, P3, as well as hydroxyl vabicaserin imine (P13) were the major drug-related components. Metabolites observed in dog plasma were also detected in dog urine. A hydroxyl vabicaserin sulfate (P17) and a diazepinyl vabicaserin carboxylic acid (P18), which was not detected in plasma, were observed in urine. Hydroxyl vabicaserin metabolites (P2, P3, and P20), a keto vabicaserin (P7), and a hydroxyl vabicaserin imine (M15) were detected in fecal extracts. Some metabolites of vabicaserin differed in dogs compared with those in rats. Sulfate metabolites were less abundant in dogs, and CG was detected in dogs but not in rats.
Representative summed mass chromatograms of plasma and urine samples obtained from monkeys and humans after oral administration vabicaserin are shown in Figs. 7 and 8. Nonlabeled material was administered to monkeys and humans, and oxidative metabolites were not quantified in these samples. In monkeys, vabicaserin was extensively metabolized via both phase I and II metabolism. In monkey plasma, along with the metabolites observed in the in vitro incubation with monkey liver microsomes, additional oxidative metabolites (P20 and P25) were present (Fig. 7).
After oral (capsule) administration of vabicaserin to humans, vabicaserin was extensively metabolized, primarily to the CG. Vabicaserin nitrone (P5) was observed in humans, as seen in dog plasma (Figs. 6 and 8). In general, urinary metabolite profiles were similar to plasma profiles (Figs. 7 and 8). A number of metabolites, which were not detected in in vitro incubations, were observed in plasma and/or urine of humans and animal species examined. These were the secondary metabolite formed through oxidative metabolism and conjugation of the primary metabolites of hydroxylation, imine formation, or carbamoyl glucuronidation.
Discussion
The present study indicated that vabicaserin undergoes NADPH-dependent oxidative metabolism to form hydroxyl metabolites (P1–P4), NADPH-independent formation of an imine (P6), as well as UDP-glucuronosyltransferase-catalyzed formation of a CG in a CO2-enriched environment in human liver preparations (Fig. 9). Vabicaserin nitrone (P5) was generated directly from vabicaserin or from P6 in human liver microsomes in the presence of NADPH.
Radioactivity was eliminated rapidly in mice (73.2% in 24 h), rats (87.4% in 24 h), and dogs (84.2% in 48 h) after oral administration of [14C]vabicaserin. Urine was the major route of elimination of the dosed radioactivity after oral administration of [14C]vabicaserin to mice and rats, whereas fecal elimination was the major route of excretion in dogs. Because nonradiolabeled vabicaserin was administered to monkeys and humans, excretion of total drug-derived materials could not be determined. However, quantitative data obtained by LC/MS analysis indicated that less than 1% of the oral dose was eliminated as unchanged drug and >50% of the administered dose was excreted as CG in urine of humans (Tong et al., 2010).
Vabicaserin was extensively metabolized in all species, and species differences were observed after a single oral dose to humans and animals. Although a number of additional metabolites were observed, in vivo metabolism of vabicaserin generally followed the same metabolic pathways observed in vitro: hydroxylation, imine formation, and carbamoyl glucuronidation. Further metabolism of the primary metabolites through oxidative metabolism and conjugation yielded additional metabolites. As previously discussed (Tong et al., 2010), carbamoyl glucuronidation was the major metabolic pathway in humans, but a minor metabolic pathway in rats. Carbamoyl glucuronidation was also a major metabolic pathway in mice, dogs, and monkeys. However, oxidative metabolism was also extensive in these species. Hydroxylation reactions occurred in all species, although different regional selectivity seemed to exist. Although vabicaserin imine was observed only in humans, secondary metabolites of the imine were observed in all species except monkey, suggesting that this metabolic pathway also exists in mice, rats, and dogs. Vabicaserin nitrone (P5) was observed in dogs and humans but not in the other species studied.
Vabicaserin is composed of a fused four-ring system, including benzene, diazepane, piperidine, and cyclopentane rings. Few metabolites were observed as a result of oxidation on the benzene ring or at positions of 6, 7, 8, and 9 (Fig. 1), even though these positions were considered to be an electron-rich region. Instead, oxidation occurred either on the cyclopentane ring with the benzene and diazepane rings serving as the metabolic enzyme binding region to generate P1 to P4, or on the diazepane ring with the other side of the molecule as the binding region for P450 or other oxidation enzymes to form P5 and P6. Identification of enzymes responsible for generating these metabolites is under investigation; because such information would be critical to determine potential clinical drug-drug interactions.
In conclusion, vabicaserin was extensively metabolized in mice, rats, dogs, monkeys, and humans after oral administration. Urinary excretion was the major route of elimination of the dosed radioactivity after oral administration of [14C]vabicaserin to mice and rats, whereas fecal elimination was the major route of excretion in dogs. The structures of the metabolites isolated and identified by LC/MS and NMR spectroscopic analysis from in vitro incubation of vabicaserin with human liver microsomes indicated that vabicaserin metabolism involved three major metabolic pathways: NADPH-dependent hydroxylation, NADPH-independent formation of vabicaserin imine, and carbamoyl glucuronidation.
Acknowledgments.
We thank Hongshan Li, Michael Carbonaro, C. Paul Wang, Abdul Mutlib, Theresa Hultin, and Rasmy Talaat for contributions to these studies.
Footnotes
Article, publication date, and citation information can be found at http://dmd.aspetjournals.org.
doi:10.1124/dmd.110.033670.
-
ABBREVIATIONS:
- 5-HT
- 5-hydroxytryptamine
- CG
- carbamoyl glucuronide
- UDPGA
- uridine 5′-diphosphoglucuronic acid trisodium salt
- LC/MS
- liquid chromatography/mass spectrometry
- mCPBA
- 3-chloroperoxybenzoic acid
- HPLC
- high-performance liquid chromatography
- DMSO
- dimethyl sulfoxide
- LSC
- liquid scintillation counter
- HSQC
- heteronuclear single quantum correlation
- HMBC
- heteronuclear multiple-bond correlation
- 2D
- two-dimensional
- WAY-280107
- 1-O-[(9aR,12aS)-6,7,9,9a,1011,12,12a-octahydrocyclopenta[c][1,4]diazepino[6,7,1-ij]quinolin-5(4H)-ylcarbonyl]-d-glucopyranuronic acid.
- Received March 31, 2010.
- Accepted August 25, 2010.
- Copyright © 2010 by The American Society for Pharmacology and Experimental Therapeutics