![[Feedback]](/icons/banner/feedbackACT.gif){kind=icon}
![[For Subscribers]](/icons/banner/subscriberACT.gif){kind=icon}
![[Archive]](/icons/banner/archiveACT.gif){kind=icon}
![[Search]](/icons/banner/searchACT.gif){kind=icon}
![[Contents]](/icons/banner/tocACT.gif){kind=icon}
|
|
|
|
|
||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Pharmaceutical Candidate Optimization (D.Z., N.R., H.Z., L.L., G.L.S., K.H.), Radiochemistry (S.-Y.C., S.J.B.), Discovery Chemistry (M.Q., P.Y.S.L.), Drug Safety Evaluation (L.M.P.), and Discovery Biology (R.M.K.), Bristol-Myers Squibb Research and Development, Princeton, New Jersey
(Received August 21, 2007; Accepted October 31, 2007)
 |
Abstract |
|---|
 Top Abstract Materials and MethodsResultsDiscussionReferences |
|---|
; Mann et al., 2003). Therefore, there is a significant medical need for safer and more effective orally active anticoagulants.
Factor Xa is a key serine protease in the coagulation cascade and has been shown to be a promising target enzyme for the treatment and prevention of arterial and venous thrombosis (Barrett et al., 2001; Samama, 2002; Walenga et al., 2003). Extensive preclinical and clinical studies show that inhibition of factor Xa is effective in both venous and arterial thrombosis (Pinto et al., 2001; Lam et al., 2003; Walenga et al., 2003). In animal models a better therapeutic index (antithrombotic efficacy versus antihemostatic effects) has been shown for direct factor Xa inhibitors compared with direct thrombin inhibitors (Leadley, 2001; Wong et al., 2002a,b, 2007).
Razaxaban (BMS-561389, DPC906), 1-(3'-aminobenzisoxazol-5'-yl)-3-trifluoromethyl-N-[2-fluoro-4-(2'-dimethyl aminomethylimidazol-1-yl)phenyl]-1H-pyrazole-5-carboxyamide, is a highly potent and selective factor Xa inhibitor. Razaxaban showed potent antithrombotic efficacy in animal models and in patients with deep vein thrombosis (Wong et al., 2007). The chemical structure of razaxaban contains an isoxazole ring (Fig. 1).
|
; Mannens et al., 1993; Mutlib et al., 1995; Yuan et al., 2002). The facile reductive cleavage of the N–O bond have been attributed to the greater electronegativity of the nitrogen atom adjacent to the oxygen in the isoxazole ring (Dalvie et al., 2002; Kalgutkar et al., 2003; Tschirret-Guth and Wood, 2003), a feature that is absent in the oxazole ring system that exclusively undergoes oxidative ring opening (Dalvie et al., 2002; Li et al., 2006). In this study, we report the reductive 1,2-benzisoxazole ring opening of razaxaban to form a stable benzamidine metabolite.
|
Materials and Methods |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
). All other organic solvents and the reagents were of high-performance liquid chromatography (HPLC) and reagent grade. Deepwell LumaPlate 96-well plates were purchased from PerkinElmer Life and Analytical Sciences, a automatic Environmental Speed Vac was obtained from Thermo Electron Corporation (Waltham, MA), an Eppendorf 5810 R centrifuge was acquired from Eppendorf Company (Hamburg, Germany), a Packard Microplate Scintillation and Luminescence Counter (model TopCount NXT) was purchased from PerkinElmer Life and Analytical Sciences (Meriden, CT), and a fraction collector was acquired from Gilson Medical Electronics (Middleton, WI). NMR spectra were recorded on a 300 MHz Biospin spectrometer (Bruker, Boston).
Synthesis of [14C]Razaxaban. Radiolabeled [14C]razaxaban was prepared as illustrated in Fig. 2. [14C]KCN was used to prepare Zn(14CN)2, which was reacted with an advanced aryl bromide to provide the labeled penultimate intermediate. The final product (radiochemical yield 43%) was isolated as its HCl salt after HPLC purification. The radiochemical purity of [14C]razaxaban was 98.8%, and its specific activity was 34.67 µCi/mg.
|
8.60 (1H, dd), 8.32 (1H, m), 7.90 (1H, t).
3-Bromo-4-fluoroaniline (2). To a solution of 1 (11.0 g, 50 mmol) in acetic acid (100 ml), iron powder (5.5 g, 98 mmol) was added in small portions. The mixture was stirred at room temperature for 2 h and then poured onto ice (200 g). The oily product was dissolved in dichloromethane and purified by column chromatography eluted with hexane/dichloromethane (1:1, v/v) to give the intermediate 2 (15 g, 80%). 1H NMR (CDCl3): 6.88 (1H, s), 6.83 (1H, m), 6.53 (1H, m).
1-(4'-Fluoro-3'-bromophenyl)-3-trifluoromethyl-5-furyl-1H-pyrole (3). To 2 (8 g, 42 mmol) in concentrated HCl (80 ml) at –5°C was added drop-wise a solution of sodium nitrite (5.8 g, 82 mmol) in water (10 ml) while maintaining the temperature at <4°C. The reaction mixture was stirred at 0°C for 30 min. Stannous chloride dihydrate (29.7 g, 126 mmol) dissolved in concentrated HCl (20 ml) at 0°C was added drop-wise to the reaction mixture with stirring. After 30 min a solution of 4,4,4-trifluoro-1-(2-furyl)-2,4-butanedione (8.65 g, 42 mmol) dissolved in acetic acid (50 ml) and methanol (100 ml) was added, and stirring was continued for 16 h at room temperature. The reaction mixture was diluted with water (100 ml) and extracted with dichloromethane (two 100-ml extractions). After solvent removal, the residue was purified by column chromatography eluted with hexane/ethyl acetate (4:1, v/v) giving the intermediate 3 (4 g, 10.7 mmol, 25%). 1H NMR (CDCl3): 7.7 (1H, dd), 7.42 (1H, d), 7.35(1H, m), 7.2 (1H, t), 6.88 (1H, s), 6.4 (1H, m), 6.16 (1H, d). MS: m/z 376 [M + H]+.
1-(4'-Fluoro-3'-bromophenyl)-3-trifluoromethyl-1H-pyrole-5-carboxylic acid (4). To a mixture of 3 (11.25 g, 30 mmol) in acetonitrile (20 ml), NaH2PO4 · H2O (20 g, 145 mmol) in water (20 ml) and phosphoric acid (1.2 ml) at –5°C was added sodium chlorite (31 g, 342 mmol) in water (60 ml) dropwise, maintaining the temperature under 0°C. The mixture was then stirred at ambient temperature for 4 h, acidified with concentrated HCl (3.3 ml), and extracted with t-butyl methyl ether (84 ml). After solvent removal, the residue was purified by flash chromatography eluted with dichloromethane-methanol-acetic acid (20:2:1, v/v/v) giving 4 (8 g, 22.6 mmol, 75%). 1H NMR (CDCl3): 7.7 (1H, dd), 7.4 (1H, m), 7.36 (1H, s), 7.22 (1H, d). MS: m/z 352 [M + H]+.
1-(3'-Bromo-4'-fluorophenyl)-3-trifluoromethyl-N-[2-fluoro-4-[(2'-pyrrolidin-1-yl-methyl)-imidazol-1-yl]phenyl]-1H-pyrazole-5-carboxyamide (5). To a suspension of 4 (3.53 g, 10 mmol) in dichloromethane (30 ml) was added oxalyl chloride (2 M in dichloromethane, 10 ml) and DMF (0.1 ml). The mixture was stirred at room temperature for 30 min and then heated at 40°C for 1 h. The mixture was evaporated to dryness. The residue was dissolved in dichloromethane (30 ml) and added dropwise to a solution of dimethylaminopyridine (1.2 g, 9.8 mmol) and BMS-564372 (2.4 g, 10 mmol) in dichloromethane (30 ml) at 0°C. The mixture was stirred at ambient temperature for 16 h. The solution was washed with water (10 ml), dried (Na2SO4), filtered, and concentrated. Flash column chromatography eluted with 5% methanol in dichloromethane on Celite produced 5 (4.3 g, 7.5 mmol, 75%) as a white solid. 1H NMR (CDCl3): 8.31 (1H, t), 8.10 (1H, s), 7.79 (1H, dd), 7.68 (1H, dd), 7.48 (1H, m), 7.36 (1H, d), 7.22 (1H, t), 7.17 (1H, s), 7.08 (2H, s), 3.40 (2H, s), 2.30 (6H, s). MS: m/z 569 and 571 [M + H]+.
1-[3'-Cyano[14C]-4'-fluorophenyl]-3-trifluoromethyl-N-[2-fluoro-4-(2'-dimethylaminomethyl)imidazol-1-yl]phenyl-1H-pyrarole-5-carboxyamide (6). To K14CN (100 mCi, 1.90 mmol) in 0.5 ml of water was added zinc chloride (116 mg, 0.853 mmol) in water (1 ml). The mixture was stirred for 10 min at room temperature. The mixture was lyophilized under high vacuum and dried to constant weight at 75°C for 3 h. [14C]Zinc cyanide was used directly for the subsequent cyanation. A round-bottomed flask was added with 5 (968 mg, 1.7 mmol), [14C]zinc cyanide (100 mCi), and tris(dibenzylideneacetone) dipalladium (85 mg), zinc powder (43 mg), [1,1'-bis(diphenylphosphine)ferrocene, 85 mg, and DMF (40 ml) under nitrogen. The mixture was heated at 120°C for 4 h, cooled to room temperature, and diluted with ethyl acetate (200 ml) and washed with water (three 50-ml washes). The ethyl acetate solution was separated, dried over Na2SO4, filtered, and evaporated to dryness. The residue was dissolved in dichloromethane. Flash chromatography eluted with 3% methanol in dichloromethane on Celite produced 765 mg (78 mCi, 1.48 mmol, 78%) of 6 (radiochemical purity 97%). 1H NMR (CDCl3): 8.3 (1H, t), 8.2 (1H, b), 7.8 (2H, m), 7.7 (1H, dd), 7.38 (2H, m), 7.2 (1H, s), 3.4 (2H, s), 2.3 (6H, s).
1-(3'-Amino-3'-[14C]benzisoxazol-5-yl)-3-trifluoromethyl-N-[2-fluoro-4-(2'-dimethylaminomethyl)imidazol-1-yl]phenyl-1H-pyrazole-5-carboxylamide (7). To a mixture of 6 (295 mg, 0.57 mmol, 30 mCi), acetohydroxamic acid (128 mg), and potassium carbonate (475 mg) were added DMF (3.6 ml) and water (0.5 ml). The mixture was stirred at room temperature for 4 h. The mixture was diluted with ethyl acetate (50 ml), washed with water (two 10-ml washes), and then evaporated to dryness. The residue was dissolved in methanol (20 ml) and purified by HPLC on a Zorbax C18, 21.2 x 250-mm column with a linear gradient of 0 min 20% B, 15 min 60% B, 20 min 95% B, 25 min 95% B, and 30 min 20% B and two solvents of A (A, water with 0.05% TFA; B, acetonitrile with 0.05% TFA). The product containing fractions were collected, combined, and freeze-dried. The residue was dissolved in ethyl acetate (20 ml), washed with 1 N KOH (two 5-ml washes), water (two 5-ml washes), dried with Na2SO4, filtered, and evaporated to dryness. The product (177 mg, 16.8 mCi, 56%) was dissolved in ethyl alcohol (20 ml). To this solution was added 1 N HCl (0.32 ml). The solution was evaporated to dryness to yield BMS-561389 (303 mg). Specific activity was determined gravimetrically to be 34.67 µCi/mg. The radiochemical purity was 98.8% as determined by HPLC. NMR confirms this radioactive material to be the structure of BMS-561389.
Synthesis of Metabolite Standards. M1: N-[4-(2-Dimethylaminomethyl)-1H-imidazol-1-yl-2-fluorophenyl]-1-(3-amidino-4-hydroxyphenyl)-3-trifluoromethyl-1H-pyrazole-5-carboxamide. Razaxaban (57 mg) was hydrogenated with 10% Pd/C (5 mg) and methanol (20 ml) at 35 psi for 30 min. The mixture was filtered through Celite, concentrated, and dried to give 33.5 mg of the desired product. 1H NMR (MeOH-d4): 7.94 (t, 1H), 7.81 (d, 1H), 7.64 (dd, 1H), 7.40 (d, 1H) 7.34 (d, 3H), 7.06 (s, 1H), 6.75 (d, 1H), 3.45 (s, 2H), 2.20 (s, 6H). MS: m/z 531.1 [M + H]+.
M4: N-[4-(2-Methylaminomethyl-1H-imidazol-1-yl)-2-fluorophenyl]-1-(3-amino-1,2-benzisoxazol-5-yl)-3-trifluoromethyl-1H-pyrazole-5-carboxamide, bis(trifluoroacetic acid) salt. M4 was synthesized as described previously (Quan et al., 2005).
M5: N-[4-(2-Methylaminomethyl-1H-imidazol-1-yl)-2-fluorophenyl]-1-(3-amidino-4-hydroxyphenyl)-3-trifluoromethyl-1H-pyrazole-5-carboxamide, bis(trifluoroacetic acid) salt. Metabolite M4 (100 mg) was hydrogenated with 10% Pd/C (10 mg) and methanol (20 ml) at 35 psi for 30 min. The mixture was filtered through Celite, concentrated, and dried to give 58 mg of the desired product as the bis(trifluoroacetic acid) salt. 1H NMR (MeOH-d4): 7.93 (t, 1H), 7.81 (d, 1H), 7.64 (dd, 1H), 7.40 (m, 3H) 7.30 (d, 1H), 7.28 (s, 1H), 7.12 (d, 1H), 4.29 (s, 2H), 2.74 (s, 3H), 2.01 (s, 2H). MS (ESI): m/z 517.3 [M+H]+.
M7: N-[4-(2-Hydroxymethyl-1H-imidazol-1-yl)-2-fluorophenyl]-1-(3-amino-1, 2-benzisoxazol-5-yl)-3-trifluoromethyl-1H-pyrazole-5-carboxamide, trifluoroacetic acid salt. M7 was synthesized as described previously (Quan et al., 2005).
M10. To a solution of razaxaban (5.06 g) in chloroform (40 ml) and methanol (120 ml) was added 35% H2O2 (20 ml) at room temperature. The reaction mixture was stirred at room temperature over 66 h. Water (180 ml) was then added to the reaction mixture and the resulting slurry was stirred for 30 min. The solid was collected by filtration and dried in vacuo with nitrogen purge at room temperature to a constant weight (3.97 g).
Animal Studies. A[14C]razaxaban dose suspension was prepared in 0.5% aqueous methylcellulose at 25 mg/ml as a free base and administered to rats and dogs by oral gavage. Mid-toxicological doses at 300 mg/kg for rats and 30 mg/kg for dogs were used in the metabolism studies.
Sprague-Dawley rats (approximately 250 g) were fasted for approximately 11 h before dose administration. [14C]Razaxaban was administered at 300 mg/kg (100 µCi/kg, 12 ml/kg) as a single oral dose by gavage. Each dosing syringe was weighed immediately before and after dosing, and the exact quantity of formulation administered was determined gravimetrically from the difference. The rats were euthanized by carbon dioxide asphyxiation over dry ice, which was performed in accordance with American Veterinary Medical Association guidelines. Blood was collected by cardiac puncture under CO2 anesthesia into tubes containing potassium EDTA at 1, 6, 12, and 24 h postdose from three rats per time point. The tubes of whole blood were placed on ice before plasma was prepared by centrifugation within 30 min of sample collection. Livers were also collected from these rats after blood collection. The livers were flash-frozen in liquid nitrogen and stored at –70 ± 10°C. For BDC rats, intraduodenal infusion of a bile salt replacement solution was at approximately 1 ml/h. The bile salt replacement solution was prepared by adding approximately 18.0 g of cholic acid per 1 liter of 0.9% sodium chloride solution followed by addition of sodium bicarbonate (approximately 1.3 g) to pH 7.4 (adjusted with hydrochloric acid), and sterile-filtering through 0.2-µm filters into sterile Viaflex bags. Bile was collected at 0 to 6, 6 to 24, and 24 to 48 h postdose into reservoirs that were surrounded by solid carbon dioxide. Urine was collected at 24-h intervals through 48 h postdose. The samples were collected into containers surrounded by solid carbon dioxide. Feces were collected at ambient temperature at 24-h intervals through 48 h postdose.
Beagle dogs were fasted for approximately 14.5 h before a single oral dose administration (20 mg/kg, 30 µCi/kg, 0.8 ml/kg) by gavage in a dosing procedure similar to that in rats. Blood samples were collected from two dogs (10 ml per time point) into tubes containing potassium EDTA at 1, 6, 12, and 24 h after the dose. Samples were collected from the left or right jugular vein within ±5% of the scheduled time (
24 h) or within 72 min of the scheduled time (>24 h). The blood samples were placed on ice before plasma was prepared by centrifugation within 30 min of blood collection. The bile salt replacement solution that was used for the BDC rats was administered to BDC dogs via the distal (flushing) catheter at a rate of approximately 23 ml/kg/day. The balloon within the bile duct cannula was inflated to the appropriate volume, and a right-angle Huber needle was inserted into the access port connected to the collection catheter. Bile was collected from two dogs at 0 to 6, 6 to 24, and 24 to 48 h postdose into reservoirs that were surrounded by solid carbon dioxide. After the final bile collection, the balloon within the bile duct cannula was deflated, and the normal bile flow pattern was reestablished. Urine was collected at 24-h intervals through 48 h postdose into containers surrounded by solid carbon dioxide. Feces were collected at ambient temperature at 24-h intervals through 48 h postdose. Water was added to each fecal sample (20%, w/w) before homogenization using a probe-type homogenizer.
HPLC Analysis. Sample analysis by HPLC was performed on a Shimadzu LC-10AT system equipped with a photodiode array UV detector from Shimadzu Scientific Instruments (Kyoto, Japan). A Xterra MS C18, 3.5 µ, 4.6 x 150-mm column (Waters, Milford, MA) was used. The mobile phase flow rate was 1 ml/min. The retention times of reference standards were confirmed by the UV detection. The HPLC solvent system was a gradient and consisted of two solvents: 100% of 0.01 M ammonium bicarbonate, pH 9.0 (A) and 100% acetonitrile (B). The step-linear gradient for elution was from 5 to 20% B in 5 min, from 20 to 40% B in 50 min, and then from 40 to 80% B in 5 min followed by a hold at 80% B for 3 min before returning to 5% B.
Preparation of Biological Samples for Analysis. A 1-ml portion of dog or rat urine or bile sample was loaded onto an Oasis (Waters) 3-ml solid-phase cartridge that was conditioned by one column of methanol and three columns of water. The sample was allowed to go through the column under gravity and then eluted with 1 ml of acetonitrile. The eluent was evaporated under nitrogen, reconstituted with 200 µl of acetonitrile-water mixture (50:50, v/v), and vortexed. After centrifugation at 2000g for 5 min, a 20-µl portion was injected for HPLC analysis.
Each pooled plasma sample was extracted by addition of 5 ml of methanol and 1 ml of acetonitrile to 1 ml of plasma, while the sample was mixed on a vortex mixer. After centrifugation of the methanol-acetonitrile-water mixture at 2000g for 10 min, the supernatant fraction was removed and saved, and the precipitate was resuspended and vortexed in acetonitrile/water (2:1, v/v). After centrifugation, the supernatant fraction was removed and combined with the first supernatant. The combined supernatant fraction was evaporated to dryness under nitrogen and reconstituted in 0.2 ml of acetonitrile and water mixture (50:50, v/v). After centrifugation at 2000g for 5 min, a 50-µl portion was injected for HPLC analysis.
Each pooled fecal homogenate sample (1 ml) was extracted by addition of 3 ml of acetonitrile, while the sample was mixed on a vortex mixer. After centrifugation of the acetonitrile-water mixture at 2000g for 10 min, the supernatant fraction was removed and saved, and the precipitate was resuspended and vortexed in acetonitrile-water (3:1, v/v). After centrifugation, the supernatant was removed and combined with the first supernatant. The second extraction was repeated. This combined supernatant was evaporated to dryness under nitrogen and reconstituted in 0.5 ml of an acetonitrile and water mixture (50:50, v/v). After centrifugation at 2000g for 5 min, a portion of 20 µl was injected for HPLC analysis.
Determination of Radioactivity and Radioactivity Profiles. Sample combustion was performed using a Packard Instruments model A0387 sample oxidizer. The resulting 14CO2 was trapped in Carbo-Sorb (PerkinElmer Life and Analytical Sciences, Meriden, CT) and mixed with Permafluor (PerkinElmer Life and Analytical Sciences) scintillation fluid, and the radioactivity was quantified by liquid scintillation counting. Samples were analyzed for radioactivity in a model LS 6500 liquid scintillation counter (Beckman Instruments, Fullerton, CA) for 10 min.
For quantification of radioactivity, the HPLC effluent was collected in 0.25-min intervals using a Gilson model 204 fraction collector. The plates were dried in an Automatic Environmental Speed Vac and counted for radioactivity for 10 min using a Packard TopCount NXT microplate scintillation and luminescence counter. Radiochromatograms were reconstructed from the TopCount data using Microsoft Excel software.
Biotransformation profiles were prepared by plotting the counts per minute values against time after injection. For each injection, the average counts per minute value from a baseline section of 2 to 3 min in the chromatogram was subtracted from the counts per minute value of each fraction. Radioactivity peaks in the biotransformation profiles were reported as a percentage of the total radioactivity collected during the entire HPLC run. Because of the very low level of radioactivity, 12- and 24-h plasma samples of rats and dogs were not profiled.
LC/MS Analysis. The extracts of pooled plasma, bile, urine, and fecal samples were analyzed by LC/MS using a Finnigan LCQ Deca XP ion-trap mass spectrometer (ThermoFinnigan, San Jose, CA) and a Micromass Q-Tof Ultima mass spectrometer (Waters, Boston, MA). The samples were analyzed in the positive ESI mode. The mobile phase flow rate was 1 ml/min. The mobile phase was split before entering the mass spectrometer. The gradient used for LC/MS analysis was the same as that for the HPLC analysis described above. Approximately 30% of the HPLC eluent was directed to the mass spectrometer through a divert valve set to divert the flow from 0 to 5 min. From 5 min until the end of the HPLC run, the effluent flow was directed to the mass spectrometer. The nitrogen gas flow rate, spray current, and voltages were adjusted to give maximum sensitivity for razaxaban. The capillary temperature used for analysis was 230°C for the LCQ mass spectrometer. Exact mass measurements were obtained on a Micromass Q-Tof mass spectrometer that was equipped with a Lock-Spray and an ESI source. The desolvation temperature used for Q-Tof analysis was 300°C. The m/z 556.2771 of an infused 20 ng/µl leucine enkephalin solution was used as the lock mass. The Q-Tof was tuned to 18,000 resolution at half-peak height using an insulin tuning solution and was calibrated up to 1500 Da using a polyalanine calibration solution. The experimentally obtained masses were compared with their respective calculated values with an error of <5 mDa.
|
Rat Liver Subcellular Fractionation and in Vitro Incubations. The subcellular fractionation from livers of untreated male Sprague-Dawley rats was performed using a literature procedure with minor modifications (Lecureux and Wattenberg, 1994). Briefly, approximately 1 g of the fresh rat liver tissue was immediately placed in 6 ml of ice-cold isotonic buffer containing 0.25 M sucrose in 10 mM Tris buffer at pH 7.2. After homogenization of the tissue using a Teflon homogenizer, the samples were subjected to differential centrifugation steps at 4°C. The homogenate was first centrifuged at 600g for 10 min, and cell debris containing nuclei, lysosomes, and tissue fragments were collected in the pellet. The supernatant was then centrifuged at 10,000g for 10 min to precipitate mitochondria. Finally, the supernatant was centrifuged at 100,000g for 60 min to separate the microsomes from the cytosol fraction. All pellet fractions were immediately placed on ice, and 3 ml of the isotonic buffer was added to suspend the pellets. Protein concentration was determined using a BCA protein assay kit (Pierce, Rockford, IL). The protein concentrations were 11.4, 8.35, 10.9, 4.85, and 6.39 mg/ml for homogenate, tissue debris, mitochondria, microsomes, and cytosol, respectively.
Incubation mixtures with rat liver fractions in a final volume of 1 ml contained 50 µl of 1 M phosphate (pH 7.4), 5 µl of 9.6 mM [14C]razaxaban, 0.1 ml of liver fractions (except 0.4 ml of cytosol), 0.1 ml of 1 mg/ml NADPH or NADH, and water. The protein concentrations were 1.1, 0.84, 1.1, 0.49, and 2.5 mg/ml for the homogenate, tissue debris, mitochondria, microsomes, and cytosol fractions, respectively. The reactions were initiated by adding NADPH or NADH.
Similar incubations were also performed with alcohol dehydrogenase and aldehyde dehydrogenase at a final protein concentration of 1 mg/ml replacing rat liver fractions. The control incubations were done without the addition of cofactors. The incubations were for 30 min at 37°C. Acetonitrile (2 ml) was added to the incubation mixture to terminate the reactions. The mixtures were centrifuged at 3000g for 10 min, and the supernatant were dried under a nitrogen stream before reconstituting in 0.2 ml of methanol-water (50:50, v/v). Aliquots of 50 µl were analyzed by HPLC and LC/MS.
|
|
Results |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
|
Metabolite Identification. Metabolites were characterized by LC/MS/MS analysis, in some cases with comparison to synthetic standards. Metabolites were identified from plasma, urine, bile, and feces of rats and dogs. Table 2 shows HPLC retention times, major fragmentation patterns, and chemical formula derived from accurate mass measurements. These m/z values are for C-12-containing metabolite species. The typical fragmentation for razaxaban and its metabolites was cleavage of the dimethylamine moiety (M-45). The glucuronidation sites for several glucuronide metabolites were not determined and glucuronidation could occur on multiple sites in the razaxaban molecule.
|
Parent. The parent compound had a molecular ion [M + H]+ at m/z 529 and a major fragment ion at m/z 484 (Table 2). The chromatographic peak for the parent compound matched that of the synthetic standard.
Metabolite M1. M1 had a molecular ion [M + H]+ at m/z 531 and a major fragment ion at m/z 486, 2 mass units higher than that of the parent compound (Table 2). Accurate mass measurement of M1 gave a molecular ion at m/z 531.1896 and a derived formula of C24H23N8O2F4 (1.6 mDa from the calculated value), suggesting that M1 resulted from reduction of razaxaban. M1 had the same HPLC retention time and fragmentation pattern as the synthetic standard. M1 was identified as dihydrorazaxaban.
Metabolite M2. M2 had a molecular ion [M + H]+ at m/z 707. LC/MS/MS analysis showed a product ion at m/z 531 that was the same as M1 (Table 2). The loss of 176 from the protonated molecular ion at m/z 707 suggested that M2 was a glucuronide of M1. Accurate mass measurement of M2 gave a molecular ion of 707.2181 and a derived formula of C30H31N8O8F4 (–2.0 mDa from the calculated value), confirming that M2 is a glucuronide conjugate of dihydrorazaxaban. The position of the glucuronic acid was not determined.
|
Metabolite M4. M4 had a molecular ion [M + H]+ at m/z 515 and a major fragment ion at m/z 484, which was consistent with demethylated razaxaban (loss of 14 mass units). Accurate mass measurement of M4 gave a molecular ion of 515.1572 and a derived formula of C23H19N8O2F4 (+0.5 mDa from the calculated value), suggesting that M4 resulted from N-demethylation of razaxaban. This metabolite had the same HPLC retention time and fragmentation pattern as the synthetic standard. M4 and was identified as N-demethyl razaxaban.
Metabolite M5. M5 had a molecular ion [M + H]+ at m/z 517 and a major fragment ion at m/z 486, which was consistent with a reductive isoxazole ring-opening product of N-demethyl razaxaban (M4). This metabolite had the same HPLC retention time and fragmentation pattern as the synthetic standard. M5 was thus identified as N-demethyl dihydrorazaxaban.
Metabolite M7. M7 had a molecular ion [M + H]+ at m/z 502 and a major fragment at m/z 486, suggesting a deaminated product. Accurate mass measurement of M7 gave a molecular ion of 515.1572 and a derived formula of C22H16N7O3F4 (+2.9 mDa from the calculated value), supporting the structure proposed for M7. This metabolite had the same HPLC retention time and fragmentation pattern as the synthetic standard. M7 was presumably formed by deamination followed by reduction of razaxaban.
Metabolite M10. M10 had a molecular ion [M + H]+ at m/z 545 and a major fragment at m/z 484, which was consistent with an added 16 mass unit (oxygen) being cleaved off during the fragmentation, and the oxygenation site was not on the isoxazole side of the molecule. Accurate mass measurement of M10 gave a molecular ion of 545.1671 and a derived formula of C24H21N8O3F4 (–0.2 mDa from the calculated value). These results suggested that M10 was an N-oxide of the parent. This metabolite had the same HPLC retention time and fragmentation pattern as the synthetic standard. M10 was identified as razaxaban N-oxide.
Metabolite M11. M11 had a molecular ion [M + H]+ at m/z 721. LC/MS/MS analysis showed a product ion at m/z 545. The loss of 176 from the protonated molecular ion at m/z 721 suggests that M11 was a glucuronide of M10. Accurate mass measurement of M11 gave a molecular ion of 721.1988 (Table 2) and a derived formula of C30H29N8O9F4 (–0.6 mDa from the calculated value), confirming that M11 was a glucuronide of M10. The position of the glucuronic acid was not determined.
|
Metabolite M13. M13 had a molecular ion [M + H]+ at m/z 547 and a major fragment ion at m/z 486, consistent with the reductive ring opening of M10. Accurate mass measurement of M13 gave a molecular ion of 547.1834 and a derived formula of C24H23N8O3F4 (0.5 mDa from the calculated value), confirming that M13 was the reductive ring opening of M10. Metabolite M13 was identified as dihydrorazaxaban N-oxide.
Metabolite M14. M14 had a molecular ion [M + H]+ at m/z 723. LC/MS/MS analysis showed a product ion at m/z 547, indicating the loss of 176 from the protonated molecular ion of m/z 723. Accurate mass measurement of M14 gave a molecular ion of 723.2148 and a derived formula of C30H31N8O9F4 (–0.2 mDa from the calculated value). These results suggested that M14 was a glucuronide of M13, dihydrorazaxaban N-oxide glucuronide.
Metabolite M15. M15 had a molecular ion [M + H]+ at m/z 503 and a major fragment ion at m/z 486, suggesting di-demethylation of M1. Accurate mass measurement of M15 gave a molecular ion of 503.1581 and a derived formula of C22H19N8O2F4 (1.6 mDa from the calculated value), confirming that M15 was N,N-des-dimethyl dihydrorazaxaban.
Biotransformation Profile in Rats. [14C]Razaxaban accounted for approximately 76% of the radioactivity at 1 h and 69% at 6 h in rat plasma. At 1 h, metabolites M1, M4, M7, and M12 accounted for 1, 1,5, and 8% of the rat plasma radioactivity, respectively (Fig. 3). These metabolites were not observed in 6-h rat plasma samples. In the 6-h plasma, there was an unidentified radioactive peak in the void volume (data not shown). This metabolite could result from further glucuronidation of a glucuronide metabolite.
Radioactivity distribution in rat urine, bile, and feces after oral administration is shown in Table 3. Metabolites M1, M2, M4, and M12 each represented <0.5% of dose in urine of intact rats (Fig. 4). In urine (0–48 h) of BDC rats (Fig. 5), the major metabolites were M1 and M4, which accounted for 3.3 and 1.3% of dose, respectively. M2 and M5 represented <1% of dose. The parent compound was a major component in rat urine.
|
In the 0- to 48-h bile from BDC rats, the parent compound accounted for 2% of the dose. M1 recovered in the bile (18% of dose) represented more than half of the biliary excretion (34% of dose). Other metabolites identified in the bile were M2, M4, M5, M10, and M12, each accounting for <4% of the dose. The extraction recovery of the rat feces homogenate ranged from 90 to 102%. From BDC rats, the 0- to 48-h fecal samples contained 18% of the dose. The parent compound accounted for 9.6% of the dose. M1 was the major metabolite and accounted for 4.3% of the dose (Fig. 5). Metabolites M4 and M7 accounted for <1% of the dose. From intact rats, the 0- to 48-h feces samples contained 88% of the dose. The parent compound accounted for 7.0% of the dose. M1 was the major component, accounting for >64% of the dose (Fig. 4). Other metabolites identified were M5, M7, and M12, each accounting for 1 to 7% of the dose. Parent and identified metabolites together accounted for >85% of radioactivity in urine, bile, and feces of rats.
|
Radioactivity distribution in dog urine, bile, and feces after oral administration is shown in Table 3. The dog urine samples (0–48 h) contained 13.6 and 5% of the dose from BDC and intact animals. The parent compound accounted for 2.0 to 2.8% of the dose. The eight metabolites M1, M2, M3, M4, M5, M7, M13, and M14 each accounted for 0.1 to 2.3% of dose, among which M1 and M4 were prominent metabolites (Figs. 6 and 7). The extraction recovery of the feces homogenate ranged from 91 to 106% in dogs. From intact dogs, the pooled fecal samples (0–48 h) contained 75.5% of the dose. The parent compound was the major component, which accounted for 46.8% of the dose. M1 was the major metabolite identified in the fecal sample (0–48 h) (Fig. 6). From BDC dogs, the pooled fecal samples (0–48 h) contained 25% of the dose. The parent compound accounted for 7.5% of the dose. Among eight prominent metabolites (Fig. 7) in BDC dog feces, M1 was a major metabolite (10.3% of dose). Metabolites M2, M7, M10, and M11 were <1% of dose.
The 0- to 48-h bile samples from BDC dogs contained 44% of the dose. The parent compound accounted for 6.6% of the dose. The metabolites identified in the bile were M1, M2, M3, M4, M5, M7, M10, M11, M12, and M13, among which M1, M4, M10, and M11 were prominent metabolites, each representing >3.1% of dose (Fig. 7). The parent compound and identified metabolites together accounted for >88% of radioactivity in urine, bile, and feces of dogs. The N-oxide metabolite M10 was a significant metabolite in dog bile (5.5% of dose) but not in the feces.
Metabolite Profile in Rat Liver Tissue. The metabolite profiles of [14C]razaxaban in rat liver tissues at different time points are shown in Fig. 8. Metabolites M1 and M5 were the major metabolites in the liver homogenates at 1, 6, and 12 h after a single oral administration of [14C]razaxaban. In addition to M1 and M5, minor metabolites M10 and M12 were also detected but no parent compound was found.
|
|
|
|
Discussion |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
; Mannens et al., 1993; Mutlib et al., 1995; Yuan et al., 2002). However, the imine intermediates themselves have never been isolated because of their rapid hydrolysis in aqueous media to ketones. The antipsychotic ziprasidone, containing a 1,2-benzisothiazole ring structure, also undergoes similar reductive metabolism, resulting in the cleavage of an N–S bond to form an amidine intermediate. This amidine intermediate was not detectedin vivo in preclinical species due either to instability or rapid metabolism to a methylthioether by methylation with S-thiomethyl transferases (Prakash et al., 1997a,b, 2000). In this study, we reported that the 1,2-benzisoxazole ring structure of razaxaban undergoes similar reductive metabolism to form a stable benzamidine metabolite both in rats and dogs and inhepatocyte incubations. [14C]Razaxaban was extensively metabolized in BDC rats and dogs, primarily by reductive isoxazole ring opening. The metabolites were primarily excreted into bile. Among the 12 metabolites identified, M1, M2 (glucuronide of M1), M5 (N-demethyl M1), M12 (N,N-di-desmethyl deamination of M1), M13 (N-oxide of M1), M14 (glucuronide of M13), and M15 (N,N-di-des-methyl M1) all had a benzamidine structure resulting from isoxazole ring reduction. Other metabolites were minor, including M3 (glucuronide of parent), M4 (N-demethyl M1), M7 (deamination followed by reduction of the parent), M9 (glucuronide of M4), M10 (N-oxide of parent), and M11 (glucuronide of M10). In addition, several glucuronide metabolites were observed in the bile samples from BDC rats and dogs but not in the feces from the intact rats and dogs, suggesting that they were hydrolyzed by gut microflora. Similarly, razaxaban N-oxide was a prominent metabolite in dog bile but not feces. The N-oxide was presumably reduced back to the parent before being excreted in feces. There seemed to be more N-oxide metabolites formed in dogs than in rats in both in vivo and in hepatocyte incubations.
M1 could be formed by enzymes in the liver or other organs or presystemically by intestinal microflora. There are several lines of evidence that support the formation of M1 in the liver including 1) the presence of M1 in bile and urine of BDC dogs after oral administration of razaxaban, 2) the detection of M1 as the major metabolite in rat liver samples collected at 1, 6, and 12 h after a single oral dose of razaxaban, and 3) the formation of M1 as the major metabolite in incubations in hepatocytes of rats and dogs. The observation of M1 in feces of BDC rats and dogs suggested that M1 was also formed in fecal contents, presumably through reduction by gut microflora or was directly excreted into the intestines.
The enzymes responsible for reductive metabolism of other isoxazole ring structures have been described in the literature. The reductive metabolism of zonisamide can be catalyzed by several enzymes, especially CYP3A4 (Nakasa et al., 1993) and aldehyde oxidase (Tatsumi and Ishigai, 1987; Sugihara et al., 1996), as well as intestinal bacteria (Kitamura et al., 1997). The reduction of the isoxazole ring of risperidone was mainly due to gut microflora (Mannens et al., 1993). The N–O bonds in 3-(indole-1-yl)-1,2-benzisoxazoles were reported to be reduced followed by N-dearylation via CYP3A4 in rat liver microsomes under anaerobic conditions (Tschirret-Guth and Wood, 2003). The reduction of the N–S bond in ziprasidone appeared to be catalyzed by aldehyde oxidase as menadione, a specific aldehyde oxidase inhibitor, abolished >70% of the formation of dihydroziprasidone (Miao et al., 2005). The 1,2-isoxazole ring-opening reaction of razaxaban was not catalyzed by equine liver alcohol dehydrogenase nor by yeast aldehyde dehydrogenase. Our preliminary results suggested that a NADH-dependent cytosolic enzyme catalyzes the isoxazole reduction of razaxaban in rat liver, although the exact enzyme responsible for this reaction is not known. The enzyme activity observed in the cell debris could be due to a membrane-bound activity but is more likely due to contamination by the cytosolic enzymes.
Tschirret-Guth and Wood (2003) studied the isoxazole ring-substitution effects on the reduction of the N–O bond of 1,2-benzisoxazole. Addition of electro-withdrawing groups (Cl– or CH3SO –2) at the 6-position of the 1,2-benzisoxazole ring causes an increase in the rate of reduction. The proposed mechanism suggested that initial electron transfer to the 1,2-isoxazole ring would yield a radical anion with nitrogen bearing the negative charge and a benzylic radical. Homolysis at the N–O bond would yield a phenoxy radical and the imine. Reductive ring opening of the 1,2-isoxazole heterocycles appeared to require a specific substitution pattern. The 3-methyl-5-alkyl-substituted 1,2-isoxazole ABT-418 (Martin et al., 1997) and 5-methylisoxazole 3-carboxamides D2624 and AG7088 (Rodrigues et al., 1995; Zhang et al., 2001) and 3,5-dimethylisoxazole-4-carboxamide (3-methylleflunomide) underwent extensive methyl hydroxylation, and no reductive ring opening was reported for these systems. However, 5-methylisoxazole-4-carboxamide (leflunomide) underwent extensive reductive cleavage of the N–O bond (Kalgutkar et al., 2003). It is not know how the amino substitute in the benzisoxazole of razaxaban affected the N–O bond reduction compared with the reduction of other isoxazole-containing compounds.
In summary, [14C]razaxaban was extensively metabolized in rats and dogs. Isoxazole reduction forming a stable benzamidine (M1) was the major metabolic clearance pathway for razaxaban. This reduction reaction was catalyzed by NADH-dependent cytosolic enzyme(s) in the liver and possibly by intestinal microflora.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
ABBREVIATIONS: razaxaban, BMS-561389, DPC906, 1-(3'-aminobenzisoxazol-5'-yl)-3-trifluoromethyl-N-[2-fluoro-4-(2'-dimethyl aminomethylimidazol-1-yl)phenyl]-1H-pyrazole-5-carboxyamide; HPLC, high performance liquid chromatography; NMR; nuclear magnetic resonance; MS, mass spectrometry; DMF, dimethylformamide; TFA, trifluoroacetic acid; ESI, electrospray ionization; BDC, bile duct cannulated; LC, liquid chromatography; MS/MS, tandem mass spectrometry; ABT-418, (S)-3-methyl-5-(1-methyl-2-pyrrolidinyl)isoxazole; D2624, N-(2,6-dimethylphenyl)-5-methyl-3-isoxazolecarboxamide; AG7088, trans-(4S,2'R,5'S,3'''S)-4-{2'-4-(4-fluorobenzyl)-6'-methyl-5'-[(5''-methylisoxazole-3'-carbonylamino]-4-oxoheptanoylamino}-5-(2'''-oxopyrrolidin-3'''-yl)pent-2-enoic acid ethyl ester.
1 Current affiliation: Amgen, Department of Pharmacokinetics and Drug Metabolism, One Amgen Center Drive, Thousand Oaks, CA 91320. 
Address correspondence to: Dr. Donglu Zhang, Pharmaceutical Candidate Optimization, Bristol-Myers Squibb Research and Development, P.O. Box 4000, Princeton, NJ 08543. E-mail: donglu.zhang{at}bms.com
|
References |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
-cyanoenol metabolite A771726: Mechanistic similarities with the cytochrome P450-catalyzed dehydration of aldoximes. Drug Metab Dispos 31: 1240–1250.
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
 |
 |
 |
|
 |
 |
 |
