![]() |
|
|
Departments of Drug Metabolism (S.-E.W.H., B.J.D., G.A.D., Z.W., C.E.C.A.H., R.A., M.Z., B.W., R.B.F., T.A.B., S.-H.L.C.), Medicinal Chemistry (P.E.F., A.J.R.), and Comparative Medicine (J.R.S., P.K.C., W.P.F.), Merck Research Laboratories, Rahway, New Jersey
(Received June 13, 2003; accepted October 2, 2003)
| Abstract |
|---|
|
|
|---|
18% of the oral dose in rat bile. The instability of this glucuronide, coupled with its presence in bile but absence in feces, suggested the potential for enterohepatic circulation of aprepitant via this conjugate. In dogs, the glucuronide of [14C]aprepitant, together with four glucuronides derived from phase I metabolites, were present as major metabolites in the bile, accounting collectively for
14% of the radioactive dose over a 4- to 24-h period after i.v. dosing. Two very polar carboxylic acids, namely, 4-fluoro-
-hydroxybenzeneacetic acid and 4-fluoro-
-oxobenzeneacetic acid, were the predominant drug-related entities in rat and dog urine.
Aprepitant (MK-0869; 5-[[(2R,3S)-2-[(1R)-1-[3,5-bis(trifluoromethyl) phenyl]ethoxy]-3-(4-fluorophenyl)-4-morpholinyl]methyl]-1,2-dihydro-3H-1,2,4-triazol-3-one; Fig. 1), a potent and selective human NK1 receptor antagonist with a mean IC50 value of 0.12 nM, has been approved recently in the United States for clinical evaluation for the treatment of chemotherapy-induced nausea and vomiting (Navari et al., 1999
; Diemunsch and Grelot, 2000
; Hale et al., 2000
; Campos et al., 2001
; Cocquyt et al., 2001
). This article describes the absorption, metabolism, and elimination of [14C]aprepitant in adult male Sprague-Dawley (SD) rats and beagle dogs, after a single i.v. or p.o. dose of [14C]aprepitant.
|
Previously, several metabolites of aprepitant were identified in primary hepatocyte cultures from rats and humans (S. Huskey, R. I. Sanchez, G. Doss, B. Arison, B. J. Dean, J. Pang, K. Leung, B. Zhu, M. Braun, P. Finke, D. Luffer-Atlas, T. A. Baillie, and S. H. L. Chiu, unpublished data). As shown in Fig. 9, these metabolites included a des-triazolone derivative (M-1), an imine derivative of the morpholine ring (M-2), a lactam derivative (M-3), two epimers of a hydroxylactam derivative (M-4), a morpholine ring-opened keto acid (M-5), and the corresponding hydroxyacid (M-6). In the present investigation, we have extended the in vitro studies to examine the profiles of aprepitant metabolites present in the plasma, urine, bile, and feces of rats and dogs. It was found that, in addition to the above oxidative metabolites, glucuronides derived from aprepitant and its metabolites comprised the major metabolites in bile of rats and dogs. Most notably, two very polar novel metabolites, 4-fluoro-
-hydroxybenzeneacetic acid and 4-fluoro-
-oxobenzeneacetic acid, were isolated and identified from the urine of both rats and dogs.
|
| Materials and Methods |
|---|
|
|
|---|
-hydroxybenzeneacetic acid (M-10; Fig. 9) was purchased from Lancaster Synthesis Inc. (Windham, NH). ß-Glucuronidase from Escherichia coli was purchased from Sigma-Aldrich (St. Louis, MO). All chemicals were of the highest analytical purity available. Synthesis of the Triacetyl Methyl Ester Glucuronide of Aprepitant. To a mixture of aprepitant (200 mg, 0.37 mmol), the thioethyl sugar (205 mg, 0.56 mmol), and type 4A crushed molecular sieves (1 g) in 1,2-dichloroethane (6 ml), at 0°C were added N-iodosuccinimide (126 mg, 0.56 mmol) and trifluoromethane sulfonic acid (5.0 µl, 0.056 mmol). The mixture was stirred at room temperature for 18 h and then quenched by the addition of 0.25 M Na2S2O3, saturated aqueous NaHCO3, and ethyl acetate (20 ml each, 1:1, by volume). The organic layer was separated, followed by filtration through a 0.2-µm PTFE filter, and the filtrate was concentrated. The resulting residue was purified by column chromatography (40 g of SG 60 silica, 34-mm column diameter, 2.55.0% MeOH/CH2Cl2), affording a colorless glass. This glass was redissolved in acetonitrile (2 ml), filtered through a 0.2-µm PTFE filter, and purified by reverse phase HPLC (RXC-8 column, 20 x 250 mm, 60% CH3CN/40% H2O, 12 ml/min, 500-µl injection). The final product (8.2 mg) was isolated as the protected glucuronide (retention time = 28.6 min) as a colorless glass.
Synthesis of the Glucuronide of Aprepitant (M-7). Triethylamine (50 µl) was added to the triacetyl methyl ester glucuronide of aprepitant (2.0 mg in 450 µl of 50% aqueous MeOH) at room temperature. After stirring for 0.5 h, the reaction mixture was concentrated to yield the free glucuronide as a colorless glass.
Animals. Male SD rats (
230300 g) were obtained from Charles River Laboratories, Inc. (Wilmington, MA), and male beagle dogs (
1014 kg) were purchased from Marshall Farms USA Inc. (North Rose, NY). The animals were housed under standard conditions, with a 12-h light/dark cycle, in the animal facilities of Merck Research Laboratories (Rahway, NJ). Animals were allowed access to water ad libitum and to commercial chow or dry dog food. They were fasted overnight before dosing, and for 4 h afterward. Animals were allowed unrestricted access to water during the fasting period. All experiments were carried out under the jurisdiction of the Merck Institutional Animal Care and Use Committee (Rahway, NJ).
Dose Preparation. For the total mass balance studies, the i.v. and p.o. doses of [14C]aprepitant were prepared in ethanol/propylene glycol/water (13:58:29, by volume). For the biliary excretion and metabolism studies, i.v. dose was prepared in a solution of ethanol/propylene glycol/water (13:58:29, by volume), and the oral dose was in a suspension of 0.5% aqueous methylcellulose containing 0.02% sodium laurylsulfate. The i.v. dosing solutions for dogs were passed through sterile 0.45-µm filters (Waters, Milford, MA) before dosing.
Total Mass Balance Studies of [14C]Aprepitant in Rats. Six adult male SD rats were housed separately in Nalgene metabolism cages. Rats (n = 3 per dosing route) received either a single i.v. dose of [14C]aprepitant via the lateral tail vein or a single p.o. dose of [14C]aprepitant by gavage at 2 mg/kg (specific activity 10 µCi/mg; 1.2 ml/kg). Urine and feces samples were collected into tared, separate plastic containers surrounded by dry ice, daily for 5 days. Samples were stored at -70°C until analyzed.
Biliary Excretion of [14C]Aprepitant in Rats. Six adult male SD rats had cannulas implanted into their bile ducts and duodena 1 day before dosing with [14C]aprepitant. The cannulas in each animal were connected to allow recirculation of bile overnight, before dosing. Three of the animals had cannulas implanted into the femoral vein for i.v. dosing. After recovery from surgery, all six rats were placed into separate metabolism cages and allowed unlimited access to an aqueous solution containing 5% dextrose, 0.9% NaCl, and 0.05% KCl.
The bile duct-cannulated rats (n = 3 for each dosing route) received either an i.v. dose of [14C]aprepitant via the femoral vein at 2 mg/kg (specific activity 29.83 µCi/mg; 1.2 ml/kg) or an oral dose of [14C]aprepitant by gavage at 5 mg/kg (specific activity 14.87 µCi/mg; 2 ml/kg). After dosing, the rats were infused with a solution of 10 mM sodium taurocholate, 0.9% NaCl, and 0.05% KCl (2 ml/h), via the duodenal catheter, for 48 h. Bile samples were collected into tared plastic containers surrounded by ice, predose and at 0- to 2-, 2- to 4-, 4- to 8-, 8- to 24-, and 24- to 48-h intervals postdose. Urine and feces samples were collected at room temperature, daily, for 2 days. Samples were stored at -70°C until analyzed.
Metabolism of [14C]Aprepitant in Rats. Each of 18 adult male SD rats received an oral dose of [14C]aprepitant, by gavage, at 100 mg/kg (specific activity 1 µCi/mg; 4 ml/kg). Three rats per time point were euthanized (CO2) at 2, 4, 6, 8, 18, and 24 h postdose, and blood (
10 ml) from each animal was collected into heparinized tubes. Plasma was prepared by centrifugation of the blood at 3000g for 10 min at 4°C.
Total Mass Balance Studies of [14C]Aprepitant in Dogs. Each of four adult male beagle dogs received either an i.v. dose of [14C]aprepitant via a cephalic vein at 1 mg/kg (specific activity 5 µCi/mg; 1 ml/kg) or a p.o. dose of [14C]aprepitant by gavage at 2 mg/kg (specific activity 5 µCi/mg; 2 ml/kg). Blood samples (1015 ml) were collected into heparinized tubes from all animals via the jugular vein at 2, 4, 6, 8, 24, 30, 48, 54, and 72 h postdose. Plasma was obtained by centrifugation of blood samples. Urine samples were collected in plastic containers surrounded by dry ice at 0- to 8-, 8- to 24-, and 24-h intervals up to 7 days postdose. Feces were collected at room temperature daily for 7 days. Samples were stored at -70°C until analyzed.
Biliary Excretion of [14C]Aprepitant in Dogs. Four male beagle dogs underwent bile duct cannulation surgery 2 weeks before the study. Blood samples were collected presurgery and at 9 days postsurgery for assessment of clinical chemistry and hematology parameters. Each dog received either an i.v. dose of [14C]aprepitant via the cephalic vein at 1 mg/kg (specific activity 10 µCi/mg; 1 ml/kg) or a p.o. dose of [14C]aprepitant by gavage at 2 mg/kg (specific activity 10 µCi/mg; 1 ml/kg). Bile samples were collected in plastic containers surrounded by dry ice, predose and at 0- to 4-, 4- to 8-, 8- to 12-, 12- to 24-, 24- to 48-, 48- to 72-, 72- to 96-, and 96- to 120-h intervals postdose. The bile samples were stored at -70°C until analyzed.
Preparation of Plasma Samples for LC-MS/MS Analysis. Equal volumes of plasma from rats or dogs (1.13.2 ml) were combined for each selected time point and proteins were precipitated with 4 volumes of isopropanol. After centrifugation at 3000g for 10 min, the organic layers were transferred to clean tubes and evaporated to dryness under N2. The residues were re-precipitated twice with 5 ml of isopropanol and resuspended in 200 or 400 µl of water/methanol (1:1, v/v). The suspensions were centrifuged at 3000g for 10 min (Eppendorf, model 5417C; Fisher Scientific Co., Pittsburgh, PA), and aliquots (20 µl) of the supernatants were analyzed by LC-MS/MS.
To monitor for the very polar metabolites, equal volumes of plasma (0.551 ml) were combined, extracted as described above, and subjected to HPLC analysis (see below, method A). The HPLC effluent fractions collected between 3 and 5 min were combined, evaporated to dryness under a stream of N2, solubilized with 400 µl of 1% aqueous trifluoroacetic acid (TFA), and analyzed by LC-MS/MS.
Preparation of Fecal Samples for LC-MS/MS Analysis. Weighed fecal samples from rats or dogs were homogenized with
3 volumes of deionized water. Each homogenate (
4 g) was extracted with 20 ml of methyl ethyl ketone (MEK) followed by a second extraction with 10 ml of MEK. The two organic extracts were combined and evaporated to dryness under N2. The resulting residue was resuspended in 5 ml of isopropanol, placed in a freezer at -20°C for 2 h, and centrifuged at 3000g for 10 min, and the organic layer was transferred to a clean glass tube and evaporated to dryness under N2. Each sample was resuspended in 120 µl of deionized water and 240 µl of acetonitrile, sonicated, vortex-mixed, and centrifuged at 3000g for 10 min. An aliquot (200 µl) of the resulting supernatant was analyzed by LC-MS/MS. The radioactive content of a duplicate aliquot (10 µl) was estimated by liquid scintillation counting to determine the efficiency of extraction.
Metabolite Isolation from Dog Feces. For the isolation of metabolites in dog feces, the aqueous fecal homogenate (
40 g from the 0- to 24-h sample) from dogs was extracted with MEK and isopropanol (see above). The dried extract was resuspended in 40 ml of water and 80 ml of acetonitrile, and extracted with 60 ml of n-hexane. After centrifugation at 3000g for 10 min, the aqueous acetonitrile layer was evaporated to dryness under N2 and resuspended in 300 µl of water and 600 µl of acetonitrile before HPLC analysis. The isolation and purification procedures were performed sequentially on three different columns (see below, HPLC analysis), and the resulting column fractions were subjected to LC-MS/MS and 1H NMR analyses.
Preparation of Urine Samples for LC-MS/MS Analysis. Combined urine samples from rats or dogs (30 ml) were concentrated under N2 to 2.3 ml, acidified by the addition of 2.3 ml of 30% TFA, and centrifuged at 3000g for 10 min, and aliquots (50 µl) of the resulting supernatants were analyzed by LC-MS/MS.
Preparation of Bile Samples for LC-MS/MS Analysis. For the analysis of biliary metabolites from rats or dogs, samples (3060 ml) were applied to a Varian Mega Bond Elut C18 cartridge (10 g) (Varian, Inc., Palo Alto, CA). The cartridges were washed with 30 ml of deionized water, and elution of the analytes of interest was accomplished using 10 ml of 90% aqueous methanol followed by 10 to 20 ml of methanol, and the methanol elution process was repeated four times. The combined methanol eluates were evaporated to dryness under N2, and the residues were redissolved in deionized water for subsequent LC-MS/MS analyses.
Chemical Hydrolysis of Bile Samples. Bile samples from rats were subjected to chemical hydrolysis to evaluate the stability of biliary metabolites. Three aliquots (125 µl) of bile were incubated at 37°C for 16 h, one with 0.1 N HCl (200 µl) and two with 0.1 N NaOH (200 µl). One of the latter incubations was subjected to HPLC analysis directly, and the other was treated further with 0.1 N HCl (400 µl) at 37°C for an additional 16 h and then analyzed by HPLC (see below, method A).
Enzymatic Hydrolysis of Bile Samples. HPLC separation of biliary metabolites from rats or dogs was performed using a Zorbax RX-C8 column (see below, HPLC analysis method A). The column effluent fractions collected from 9 to 14 min were combined, evaporated to dryness under N2, and reconstituted with 4 ml of 0.1 M phosphate buffer (pH 7.4). An aliquot (3 ml) was incubated in the presence of 0.5 ml of ß-glucuronidase (100 units) at 37°C for 16 to 24 h. The reaction was quenched with an equal volume of cold acetonitrile, and the suspension was centrifuged at 3000g for 10 min. The supernatant was transferred to a clean tube, concentrated to approximately 3 ml under N2, and an aliquot (200 µl) was analyzed by HPLC (see below, method A).
Radioactivity Measurements. For an estimate of total radioactivity, duplicate aliquots of urine (100200 µl), bile (100500 µl), and column effluent fractions (1020 µl) were analyzed in a liquid scintillation counter (Beckman Coulter, Fullerton, CA) following the addition of scintillation cocktail (Ultima-FLO M; PerkinElmer Life Sciences, Boston, MA). Duplicate or triplicate aliquots of plasma (0.2 ml) or fecal homogenate (
0.2 g) were pipetted into paper combustion cones and air-dried overnight before combustion. Sample combustions were performed in an oxidizer (model 307; PerkinElmer Life Sciences), followed by liquid scintillation counting. Quench corrections were made by the external standard method.
HPLC Analysis. A Shimadzu HPLC system (Shimadzu Scientific Instruments Inc., Columbia, MD) consisting of two pumps (LC-10AD), a controller (SCL-10A), an autosampler (SIL-10A), a radiomonitor (IN/US Systems Inc., Tampa, FL), and a fraction collector (model FC 204; Gilson, Inc., Middleton, WI) was used for all analyses.
Metabolite Isolation from Dog Feces. The isolation and purification of an unknown fecal metabolite was carried out using three consecutive chromatographic analyses. Three different columns: 1) Zorbax RX-C8 (5 µm; 4.6 x 250 mm; MacMod Analytical, Chadds Ford, PA), 2) Hypersil BDS-C18 (5 µm; 4.6 x 250 mm; MacMod Analytical), and 3) Zorbax SB-phenyl (5 µm; 4.6 x 250 mm; MacMod Analytical) were used for separation. The same mobile phase, consisting of solvent A (10 mM ammonium acetate in water) and solvent B (7.2 mM ammonium acetate in 7.2% methanol/92.8% acetonitrile), was used in all three separations. The flow rate was 1 ml/min. The linear gradients were from 35 to 80% B in 40 min for the Zorbax RX-C8 column (method A), 50 to 60% B in 30 min for the Hypersil BDS-C18 column, and 40 to 60% B in 30 min for the Zorbax SB-phenyl column. The retention times for the fractions containing the radioactive component of interest using HPLC conditions 1, 2, and 3 were
17,
7.9, and
13.1 min, respectively. The fractions containing the radioactive component were evaporated to dryness under N2 and subjected to LC-MS/MS and 1H NMR analyses.
HPLC Analysis of Urine Samples. The HPLC separation of urinary extracts from rats or dogs was performed using a Zorbax SB-Phenyl column (5 µm; 4.6 x 250 mm). The mobile phase consisted of solvent C (0.1% aqueous TFA) and solvent D (acetonitrile containing 0.1% TFA). Samples were injected onto the column via an autosampler. The flow rate was 1 ml/min, and elution was accomplished using a linear gradient from 5 to 80% D in 35 min, followed by 80 to 95% D in 7 min, and maintained at 95% D for another 2 min (method B). The column was connected to an on-line radiomonitor to obtain radiochromatograms.
LC-MS/MS Analysis. Analyses were performed using a Sciex API 300 (Applied Biosystems, Foster City, CA) upgraded to API 365 specifications or an API 3000 triple quadrupole tandem mass spectrometer utilizing the ionspray interface. The HPLC system interfaced with the mass spectrometer consisted of two PerkinElmer Series 200 micro LC pumps and a PerkinElmer Series 200 autosampler. Both positive and negative ion detection methods were employed, and the mass spectrometry conditions were optimized using metabolite M-1 in the positive ion mode. The MS/MS experiments were performed by collision-induced dissociation with N2 as the target gas, and full scan data were acquired. Multiple components were monitored in one LC-MS/MS experiment by separating the total acquisition time (41 min) in segments, during which MS/MS data were acquired for specific ions.
LC-MS/MS Analysis of Plasma, Bile, and Fecal Extracts. For plasma, bile, and fecal extracts from rats or dogs, chromatography was performed on a Zorbax RX-C8 (5 µm; 4.6 x 250 mm) column. A linear gradient from 35 to 80% B in 40 min was used, followed by a step gradient from 80 to 95% B in 0.1 min, and maintained at 95% B for 5 min (method A). The mobile phase consisted of solvent A (10 mM ammonium acetate in water) and solvent B (7.2 mM ammonium acetate in 7.2% methanol/92.8% acetonitrile). Samples (5200 µl) were injected onto the column via an autosampler. The flow rate was 1 ml/min, of which 0.04 ml/min was diverted into the mass spectrometer, while the remainder was collected in a fraction collector. An aliquot of the latter was subjected to liquid scintillation counting to determine radioactivity. Alternatively, the column was connected to an on-line radiomonitor to obtain radiochromatograms.
LC-MS/MS Analysis of Urine Extracts and Very Polar Metabolites from Plasma Samples. When the positive ion mode was utilized for LCMS/MS analysis, a Zorbax RX-C8 (5 µm; 4.6 x 250 mm) or a Zorbax SB-C8 (5 µm; 4.6 x 150 mm) column was utilized. The mobile phase consisted of solvent C (0.1% TFA in water) and solvent D (0.1% TFA in acetonitrile). Samples were injected onto the column via an autosampler and analytes were eluted using a linear gradient from 5 to 25% D in 15 min, followed by a step gradient from 25 to 90% D in 0.5 min, and maintained at 90% D for another 2 min. The flow rate was 1 ml/min. A switch valve was set up to divert the highly aqueous solvent to waste, and solvent B was introduced postcolumn and before the mass spectrometer at a flow rate of 0.6 ml/min during this 5-min period. Only 0.04 ml/min was directed into the mass spectrometer; the remainder was collected for liquid scintillation counting.
When the negative ion mode was employed for the LC-MS/MS analysis, a Zorbax RX-C8 (5 µm; 4.6 x 250 mm) or Zorbax SB-C8 (5 µm; 4.6 x 75 mm) column was used. The mobile phase consisted of solvent E (0.1% aqueous formic acid) and solvent F (0.1% aqueous formic acid in acetonitrile). Samples were injected onto the column via an autosampler, and the analytes were eluted using a linear gradient from 5 to 35% F in 5 min, followed by a step gradient from 35 to 90% F in 0.2 min, and maintained at 90% F for another 2 min. The flow rate was 1 ml/min. A switch valve was used to direct the highly aqueous eluate to waste for the first 2 min, during which time solvent B was introduced postcolumn. Only 0.04 ml/min was directed to the mass spectrometer; the remainder was collected for liquid scintillation counting.
NMR Analysis. Purified metabolites were analyzed by proton NMR on a 500 MHz Varian Inovd NMR instrument (Varian, Inc.) using deuterated methanol or acetonitrile as the solvent and a Nalorac 3-mm inverse detection probe. Chemical shifts are reported as ppm downfield from tetramethylsilane, and multiplicities are indicated as singlet (s), doublet (d), or multiplet (m).
| Results |
|---|
|
|
|---|
30% in urine and
55% in feces. After a single i.v. (2 mg/kg) or p.o. (5 mg/kg) dose of [14C]aprepitant to bile duct-cannulated rats, recovery of the radioactive dose in bile over 48 h was 42.3 ± 2.7% and 30.5 ± 10.1%, respectively (data not shown). Absorption of the radioactive oral dose was
43%, based on combined recovery of radioactivity in the bile (
31%) and urine (
12%) following a single p.o. (5 mg/kg) dose (data not shown).
|
Total Mass Balance and Biliary Excretion of [14C]Aprepitant in Dogs. After i.v. (1 mg/kg) and p.o. (2 mg/kg) administration of [14C]aprepitant to male beagle dogs, the average recovery of radioactivity in excreta over a 168-h period was
84% and
87%, respectively, with the majority of the radioactive dose (
5065%) recovered in 48 to 72 h, respectively (Table 2). The excretion of radioactivity was divided almost equally between the urine (
3543%) and the feces (
3945%). In bile duct-cannulated dogs given [14C]aprepitant, recovery of the radioactivity in bile over a 5-day period was
49% and
30% following i.v. (1 mg/kg) and p.o. (2 mg/kg) dosing, respectively (data not shown).
|
Metabolites of [14C]Aprepitant in Rat and Dog Plasma. The metabolism of aprepitant was evaluated after a single oral dose of [14C]aprepitant to rats (100 mg/kg) and dogs (2 mg/kg). Representative radioactivity profiles derived from plasma samples from rats and dogs at 8, 24, or 30 h postdose are shown in Fig. 2. Plasma metabolite profiles of aprepitant were qualitatively similar in rats and dogs at selected time points. [14C]Aprepitant was the major radioactive component in the plasma for up to 8 h in rats and dogs. The presence in plasma of all metabolites (M-1 to M-6) previously identified in hepatocytes (Huskey et al., 2003) was confirmed by LC-MS/MS analysis, except that M-2 was not detected in dog plasma (Fig. 2; Tables 3 and 4). In addition, the presence of very polar metabolites, designated M-10, M-11, and M-12, were confirmed by LC-MS/MS (Tables 3 and 4). Previously, M-10 and M-11 were identified from rat and dog urine (Dean et al., 2000
), and M-12 was determined to be an amino derivative of M-10, based on the LC-MS/MS analysis and comparison with authentic compound. Compounds M-5 and M-6 were the major metabolites in rat plasma at 24 h postdose, whereas M-6 and M-10 predominated in dog plasma at 30 h postdose.
|
|
|
Metabolites of [14C]Aprepitant in Rat Bile. A representative radiochromatogram derived from HPLC analysis of a rat bile sample (024 h after an oral dose) revealed two major radioactive components (913 and 2325 min, HPLC method A; Fig. 3, panel A), which accounted for
42% and
28% of the total radioactivity in the extract, respectively. The latter component was confirmed by LCMS/MS to be the parent compound, which accounted for
5% of the radioactive dose (data not shown).
|
The effluents from the 9- to 13-min fraction were combined (Fig. 3, panel B) and treated with ß-glucuronidase (Fig. 3, panel C). The radiochromatogram of the hydrolysate indicated that the major component (
54%) was [14C]aprepitant, suggesting that the major metabolite in rat bile was a glucuronide of aprepitant, designated as M-7, which accounted for
18% of the radioactive dose (data not shown). LC-MS/MS analysis also supported the assignment of M-7 as a glucuronide of aprepitant (Table 3). Similar findings were obtained in rats following i.v. administration of [14C]aprepitant when the unchanged parent drug and glucuronide M-7 accounted for
7% and
17% of the radioactive dose (data not shown), respectively.
Stability of an Aprepitant Glucuronide Metabolite, M-7. Based on HPLC analysis, the glucuronide of aprepitant was acid-labile since it was found to revert completely to the parent compound upon mild acid treatment (0.1 N HCl at 37°C for 16 h). However, this conjugate was stable under mildly basic conditions (0.1 N NaOH at 37°C for 16 h) but was hydrolyzed completely to the parent drug upon further treatment with 0.1 N HCl (data not shown). Based on the structure of aprepitant, four regioisomeric structures of glucuronide are possible.
1H NMR Analysis of an Authentic Glucuronide of Aprepitant. An authentic glucuronide of aprepitant was synthesized chemically in an attempt to distinguish between the four possible regioisomeric structures. NMR studies were carried out on a synthetic protected (as triacetyl methyl ester) glucuronide intermediate of aprepitant due to the instability of glucuronide of aprepitant under experimental conditions (Table 5). The possibility that conjugation occurred at the nitrogen of the morpholine ring was considered unlikely since there were no significant downfield shifts in the adjacent protons (Fig. 4, upper panel). In a nuclear Overhauser effect (NOE) experiment, the anomeric proton of the glucuronide was irradiated and a strong NOE was observed within the glucuronic acid moiety, but none was detectable at the exocyclic methylene protons (Fig. 4, lower panel). The results likely exclude structures in which glucuronic acid is attached to either of the nitrogen atoms adjacent to the junction carbon of the triazolone ring. Attempts to distinguish between the two remaining structures were unsuccessful. Collectively, data obtained to date (NMR and LC-MS/MS) do not differentiate between an O- and an N-glucuronide on the triazolone ring.
|
|
After removing the protecting groups with triethylamine, the synthetic glucuronide of aprepitant was subjected to Fourier-transform infrared spectroscopic analysis. The spectrum showed that the isolate was an adduct of the triethylamine salt of a carboxylate glucuronide as exhibited by the carbonyl stretch at 1606 cm-1. This is indicative of an amine salt of a carboxylate. The lack of an additional carbonyl stretch at
1700 cm-1, as was present in the unsubstituted parent, and the characteristic of an imide carbonyl, suggests that the glucuronide adduct exists as the O-substituted regioisomer (data not shown). However, the possibility of an N-substituted regioisomer with the resultant carbonyl in the tautomeric enolic form still exists.
Comparison of the Major Biliary Metabolite, M-7, and the Authentic Glucuronide of Aprepitant. The synthetic glucuronide of aprepitant and bile samples were compared in terms of chemical stability and chromatographic properties. Under acidic and basic conditions, the synthetic glucuronide of aprepitant demonstrated similar instability in acid and stability in base, as was observed for the biliary metabolite M-7 (data not shown). In addition, the synthetic glucuronide of aprepitant and M-7 exhibited similar chromatographic properties under two different conditions (method A, Zorbax RX-C8 column, and Hypersil BDS-C18 column). When the two compounds were mixed prior to HPLC, they coeluted as indicated by UV absorbance and radioactivity measurements. Collectively, the data support the contention that M-7 is a glucuronide of aprepitant, but the site of attachment of the sugar moiety remains unknown.
In addition to the glucuronide of aprepitant, which proved to be the major metabolite in rat bile, metabolites M-1 to M-6, M-8 (see below), M-10, and M-11 were confirmed by LC-MS/MS analysis to be minor metabolites of aprepitant (Tables 3 and 4).
Metabolites of [14C]Aprepitant in Dog Bile. A representative radiochromatogram of dog bile (424 h after i.v. dose) revealed two major radioactive components (914, 2728 min, HPLC method A; Fig. 5, panel A), which accounted for
54% and
13% of the total radioactivity in the extract. The latter component was confirmed by LC-MS/MS to be parent compound, which accounted for
3% of the radioactive i.v. dose.
|
The effluents from the 9- to 14-min fractions were combined and treated with ß-glucuronidase. The radiochromatogram of the resulting hydrolysate (Fig. 5, panel B) indicated the presence of aprepitant, as well as several metabolites (M-1, M-4, M-8, and M-9). The findings suggested that the major components in dog bile were glucuronides derived from aprepitant and its metabolites, which accounted collectively for
14% of the radioactive i.v. dose.
Moreover, all of the metabolites observed in plasma also were present in bile, together with M-2. Interestingly, M-9 per se was not detected in dog bile, although the glucuronide conjugate of this metabolite was present (Tables 3 and 4).
Metabolites of [14C]Aprepitant in Rat and Dog Feces. Combined fecal extracts from three rats or two dogs (024 h postdose; extraction recovery >70%) were analyzed by HPLC and LC-MS/MS. As shown in Fig. 6, the radioactivity profiles of the fecal extract from rats and dogs revealed a major radioactive component, which was confirmed to be unchanged [14C]aprepitant, as well as several other radioactive components (Tables 3 and 4). These metabolites, including M-1 to M-6, observed previously in rat or dog plasma, also were detected in the fecal samples. In addition, M-8 was found in dog feces, and its identification by 1H NMR is described below. Although low levels of radioactive components were present in the very polar region of the chromatograms, the identities of the responsible components remain to be determined.
|
Structural Identification of Metabolite M-8. The 1H NMR spectrum of metabolite M-8, isolated from dog fecal extracts, showed that all nonexchangeable protons of aprepitant were present, including those of the methylene bridge. Based on the chemical shift of the exocyclic methylene group (Table 6), the metabolite was proposed to be a triazolone ring-opened carboxylic acid, designated as M-8 (Fig. 7). Addition of deuterated ammonia caused an upfield shift of the methylene bridge signals consistent with the attachment of the carboxylic acid group (Fig. 7, lower panel). LC-MS/MS analysis revealed that M-8 had a molecular weight of 495, suggesting the presence of an odd number of nitrogens in its structure (Table 3). Aprepitant (molecular weight 534) contains four nitrogens, one on the morpholine ring and three on the triazolone ring. Therefore, M-8 was generated by the loss of one or three nitrogens from the parent compound. The presence in the positive ion spectrum of a fragment ion at m/z 179, indicative of an intact morpholine ring, suggested that M-8 was a product derived from opening of the triazolone ring. Subsequently, an authentic sample was obtained by synthesis, and provided LC-MS/MS and NMR spectra identical to those of M-8.
|
|
Metabolites of [14C]Aprepitant in Rat and Dog Urine. A representative HPLC radiochromatogram derived from rat urine extracts is depicted in Fig. 8. Qualitatively similar profiles were obtained from urine extracts from dogs (data not shown). These chromatograms revealed that no parent compound was excreted unchanged into urine, which contained the very polar metabolites M-10, M-11, M-13, and M-14 (Tables 3 and 4). M-11 and M-12 were identified as 4-fluoro-
-hydroxybenzeneacetic acid and 4-fluoro-
-oxobenzeneacetic acid, respectively, and M-13 and M-14 were triazolone-containing carboxylic acid and the corresponding alcohol. Metabolites M-10 and M-11 were the most predominant metabolites eliminated by the urinary route in rats and dogs.
|
Figure 9 and Table 4 summarize the metabolites detected in plasma, bile, urine, and feces from rats and dogs following i.v. or p.o. administration of [14C]aprepitant.
| Discussion |
|---|
|
|
|---|
In addition to the N-dealkylation pathway, glucuronidation was found in the present study to be a quantitatively important pathway in the metabolism and excretion of aprepitant in rats and dogs. Thus, a glucuronide conjugate of aprepitant, M-7, was the major biliary metabolite in rats (Fig. 3). In addition to M-7, several glucuronides derived from the oxidative metabolites, M-1, M-4, M-8, and M-9, were present in the bile of dogs (Fig. 5). To our knowledge, M-7 represents a novel glucuronide with glucuronic acid attached to the triazolone ring, which contains four possible sites (three nitrogens and one carbonyl oxygen) for attachment of the sugar residue (Fig. 4). The instability of M-7 necessitated the structural analysis of a synthetic glucuronide containing protecting groups to stabilize the molecule. Based on proton NMR analyses, the site of glucuronic acid conjugation was narrowed down to either the carbonyl oxygen or the nitrogen distant from the carbon to which the methylene bridge was attached, but the site of attachment could not be defined unambiguously. After NMR analysis, the protecting groups were removed from this synthetic glucuronide, and the resulting conjugate was compared with M-7 for both chemical stability and chromatographic properties. Metabolite M-7 shared the same chemical and chromatographic properties with the synthetic glucuronide, although the possibility that the two species were regioisomers cannot be excluded completely. Furthermore, the instability of M-7, its abundance in rat and dog bile, and its absence in rat and dog feces suggest that it may play a role in enterohepatic recycling of the parent drug. Finally, it should be noted that whereas M-7 was detected at low levels when aprepitant was incubated with rat hepatocytes (S. Huskey, R. I. Sanchez, G. Doss, B. Arison, B. J. Dean, J. Pang, K. Leung, B. Zhu, M. Braun, P. Finke, D. Luffer-Atlas, T. A. Baillie, and S. H. L. Chiu, unpublished data), it was not detected when aprepitant was incubated with rat liver microsomes fortified with UDP-glucuronic acid and detergent (0.02% Lubrol PX). Low turnover of aprepitant to M-7 in the in vitro system may be attributed to the low intrinsic reactivity of aprepitant and/or the instability of M-7.
Another metabolic event worthy of discussion is related to the phase I metabolism of the triazolone ring of aprepitant. Based on the 1H NMR and LC-MS/MS analyses, M-8 was confirmed to be a carboxylic acid, derived from the opening of the triazolone ring. Presumably, metabolism of aprepitant can proceed by way of two hydrolytic steps and a spontaneous decarboxylation, leading to the formation of an amide derivative, which can be hydrolyzed further to the carboxylic acid (Fig. 10). This carboxylic acid metabolite was detected in plasma from dogs and feces from rats and dogs, as well as in bile from dogs upon treatment with ß-glucuronidase, suggesting that the glucuronide of M-8 was present in dog bile. The presence of M-8 and its glucuronide in rat bile was confirmed by LC-MS/MS, albeit at low concentrations; however, their presence in rat plasma was not studied because the information on M-8 was not available at the time of analysis. Similar to the situation with M-7 described above, this conjugate of M-8 was not detected in incubations with rat liver microsomes. However, its presence in rat hepatocyte cultures was confirmed by LC-MS/MS (S. Huskey, R. I. Sanchez, G. Doss, B. Arison, B. J. Dean, J. Pang, K. Leung, B. Zhu, M. Braun, P. Finke, D. Luffer-Atlas, T. A. Baillie, and S. H. L. Chiu, unpublished data), consistent with the in vivo findings. The stability of the acylglucuronide of M-8 in plasma was not studied due to the unavailability of authentic acylglucuronide.
|
The urinary metabolites of aprepitant consisted of very polar low molecular weight entities (M-10 to M-14; Fig. 8). Notable common features among these metabolites were the absence of the bis-trifluoromethylphenyl moiety and the loss of an intact morpholine ring. The presence of M-13 and M-14, which do retain the intact triazolone ring, suggests that O-dealkylation of [14C]aprepitant triggers the metabolic pathway that leads to the formation of this group of very polar metabolites, M-10 to M-14. As reported previously, at least two very polar metabolites were shown to be formed when [14C]aprepitant, M-1, or M-2, was incubated in the presence of Sf21 cells expressing human CYP3A4, CYP1A2,,or CYP2C19 (S. Huskey, R. I. Sanchez, G. Doss, B. Arison, B. J. Dean, J. Pang, K. Leung, B. Zhu, M. Braun, P. Finke, D. Luffer-Atlas, T. A. Baillie, and S. H. L. Chiu, unpublished data). The identification of these very polar metabolites currently is under investigation in our laboratories.
Aprepitant is a chiral molecule containing three asymmetric carbons. Therefore, stereoisomers may exist among the metabolites as a result of metabolic events occurring at, or adjacent to, these chiral centers. Since the biotransformations leading to six of the metabolites occurred either at a nonchiral part of the molecule (M-1, M-3, M-7, M-8) or by the elimination of one of the three original chiral centers (M-2, M-5), the chirality at the other chiral centers of aprepitant should not be affected and stereoisomers of these metabolites are not expected to be formed. The interconversion of M-4 to its C-3 hydroxy stereoisomer can occur chemically under neutral conditions, presumably through reversible opening of the lactam ring, as described previously (S. Huskey, R. I. Sanchez, G. Doss, B. Arison, B. J. Dean, J. Pang, K. Leung, B. Zhu, M. Braun, P. Finke, D. Luffer-Atlas, T. A. Baillie, and S. H. L. Chiu, unpublished data). M-6 was shown to be formed as a 10:1 mixture of its two hydroxy isomers (the absolute stereochemical assignments were not determined) when incubated with M-1 in rat hepatocytes, presumably via enzymatic reduction of the ketone metabolite (M-5) (S. Huskey, R. I. Sanchez, G. Doss, B. Arison, B. J. Dean, J. Pang, K. Leung, B. Zhu, M. Braun, P. Finke, D. Luffer-Atlas, T. A. Baillie, and S. H. L. Chiu, unpublished data). Since the incubation of either of the pure isomers in hepatocytes afforded the same isomer mixture, in vivo interconversion of this hydroxy via enzymatic oxidation to M-5 and subsequent reduction might also be possible (data not shown). No attempts were made, however, to study the rate of interconversion and the relative stability of all possible stereoisomers in the biological matrices examined.
| Acknowledgments |
|---|
| Footnotes |
|---|
-[(R)-[(1R)-[3,5-bis(trifluoromethyl) phenyl]ethoxy](2-hydroxyethoxy) methyl]-4-fluorobenzenemethanol; M-10, 4-fluoro-
-hydroxybenzeneacetic acid; M-11, 4-fluoro-
-oxobenzeneacetic acid; M-12, 4-fluorophenylglycine; M-13, (
S)-4-fluoro-
-[[(2,5-dihydro-5-oxo-1H-1,2,4-triazol-3-yl)methyl] amino]benzeneacetic-carboxy acid; M-14, 5-[[(1S)-1-(4-fluorophenyl)-2-hydroxyethyl-2-amino]methyl]-1,2-dihydro-3H-1,2,4-triazol-3-one; HPLC, high-performance liquid chromatography; LC-MS/MS, liquid chromatography-tandem mass spectrometry; TFA, trifluoroacetic acid; MEK, methyl ethyl ketone; NOE, nuclear Overhauser effect. Address correspondence to: Dr. Su-Er W. Huskey, Dept. of Drug Metabolism, Merck Research Laboratories, P.O. Box 2000, Rahway, NJ 07065. E-mail: su_huskey{at}merck.com
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
R. I. Sanchez, R. W. Wang, D. J. Newton, R. Bakhtiar, P. Lu, S.-H. L. Chiu, D. C. Evans, and S.-E. W. Huskey CYTOCHROME P450 3A4 IS THE MAJOR ENZYME INVOLVED IN THE METABOLISM OF THE SUBSTANCE P RECEPTOR ANTAGONIST APREPITANT Drug Metab. Dispos., November 1, 2004; 32(11): 1287 - 1292. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||