Introduction

Top Abstract Introduction Experimental Procedures Results and Discussion References

In general, drugs are metabolized to more polar, hydrophilic entities, which can be excreted from the body more easily. Identification of metabolites may reveal metabolic liabilities in the compound under investigation. With this information, synthetic chemists can synthesize compounds that are more metabolically stable and, consequently, have a lower clearance rate and a longer half-life. Alternatively, in some cases, extensive metabolism has been used to create prodrugs, which may be better absorbed than the drug itself (e.g., fenofibrate). Sometimes, metabolites can possess pharmacological activity and explain pharmacodynamic observations. Finally, during toxicology studies, detailed knowledge about the metabolism is important, because, for proper assessment of the safety of a drug for human use, it must be shown that the animal species used for safety evaluation are exposed to the same metabolites as humans. Thus, details about the metabolism of a drug and the enzymes involved in the metabolism are a necessity throughout drug discovery and development (Borchardt et al., 1998).

LC-MS and LC-MS/MS have become popular analytical tools for drug metabolism studies, because they are sensitive techniques for obtaining the molecular mass, molecular formula, and (limited) structural information of metabolites (Korfmacher et al., 1997; Poon, 1997; Baillie and Pearson, 2000). A comparison of the HPLC retention times, as well as MS and MS/MS spectra of putative metabolites and authentic standards, may be sufficient to make more definitive assignments. However, in many cases, nuclear magnetic resonance (NMR) spectroscopy is required to obtain the exact structure of a metabolite. Here, we will describe the metabolism of [3R,5R,6S]-3-(2-cyclopropyloxy-5-trifluoromethoxyphenyl)-6-phenyl-1-oxa-7-azaspiro[4.5]decane (compound A hereafter) in rat plasma as well as in rat liver microsomes and rat hepatocytes. This compound is a Substance P (Neurokinin 1 receptor) antagonist, and the same applies to some of its metabolites. Substance P antagonists are proven in concept to have excellent potential for the treatment of major depression, and they allow superior and sustained protection from acute and delayed chemotherapy-induced emesis (Kramer et al., 1998; Navari et al., 1999; Tattersall et al., 2000).

	Experimental Procedures

Top Abstract Introduction Experimental Procedures Results and Discussion References

Materials. Compound A and authentic standards of the O-dealkylated metabolite (B), the hydroxylamine metabolite (C), the nitrones (D and E), the lactam metabolite (F), and the hydroxylated and O-dealkylated metabolite (G) were prepared at Merck Research Laboratories (98% purity), and their structures were confirmed by MS and NMR analysis; see Scheme 1 for the structures of compound A and its metabolites. ¹⁴C-labeled compound A was synthesized in the Radiolabeled Compound Synthesis facility (98.9% purity); the position of the ¹⁴C label is depicted in Scheme 1. The complex synthesis of [¹⁴C]A is described by Raab et al. (2001). All other chemicals were of analytical grade and solvents of HPLC grade.

View larger version (15K): [in this window] [in a new window]   Scheme 1.   Structures of compound A and its metabolites.

Rat Hepatocyte Incubations. Hepatocytes from male Sprague-Dawley rats weighing 200 to 400 g were isolated after collagenase digestion of the liver using a two-step perfusion procedure as described previously by Pang et al. (1997). Compound A was incubated at 10 µM with freshly prepared hepatocytes (1 million cells/ml) for 4 h. The incubation was terminated by the addition of an equal volume of acetonitrile followed by centrifugation. The supernatant was transferred to a PerkinElmer Series 200 autosampler (PerkinElmer Instruments, Norwalk, CT) for analysis by LC-MS.

In Vivo Rat Studies. Sprague-Dawley rats were dosed intravenously with 5 mg/kg of compound A; the vehicle was H₂O/C₂H₅OH/polyethylene glycol 400, 80:10:10 (v/v/v). Compound A was also administered orally at 10 mg/kg with 0.5% methyl cellulose as the vehicle. Blood was withdrawn at fixed time-points and plasma was obtained immediately by centrifugation. All animal experiments were approved by the institutional animal care and use committee. The samples were stored at 70°C until analysis. The samples were prepared for analysis by protein precipitation using 5 volumes of acetonitrile followed by centrifugation. The supernatant was transferred to a PerkinElmer Series 200 autosampler for analysis by LC-MS.

Rat Liver Microsomal Incubations. Compound A was incubated at 25 µM with hepatic microsomes from dexamethasone-treated rats (2 mg of protein/ml) for 2 h. Rat liver microsomes were isolated by differential centrifugation (Raucy and Lasker, 1991) following oral dosing with dexamethasone for 4 days at 200 mg/kg. The incubation was terminated by the addition of an equal volume of acetonitrile, followed by centrifugation of the sample. A metabolite (J) was isolated by semipreparative HPLC on a Zorbax RX-C₈ column (9.4 × 250 mm, 5 µm; Agilent Technologies Inc., Wilmington, DE). A variety of gradient elution profiles, with various proportions of 10 mM ammonium acetate buffer, acetonitrile, and methanol (each with or without 0.1% trifluoroacetic acid), were used to resolve the metabolite from endogenous biological materials. Appropriate fractions, containing the metabolite of interest, were collected, dried under a gentle stream of nitrogen, and subjected to ¹H-NMR analysis.

LC/MS Analysis. The metabolic profile was determined by reversed-phase HPLC analysis on a Zorbax RX-C₁₈ column (4.6 × 250 mm) using two PerkinElmer series 200 micro LC pumps. Reservoir A contained H₂0 + 10 mM ammonium acetate + 0.1% trifluoroacetic acid, and reservoir B contained acetonitrile/methanol, 50:50 (v/v) + 0.1% trifluoroacetic acid. For the first 2 min, the organic content remained at 30%, followed by a linear gradient from 30% B to 60% B in 8 min, a linear gradient from 60% B to 70% B in 15 min, a linear gradient from 70% B to 80% B in 8 min and, finally, the organic content was kept at 80% B for 4 min. Additional experiments were performed without trifluoroacetic acid in the mobile phase. The flow rate was 1.0 ml/min of which 0.04 ml/min was directed into the mass spectrometer, and the rest was diverted to waste or was collected in 1-min fractions for liquid scintillation counting (Beckman LS6500; Beckman Coulter, Inc., Fullerton, CA). Mass spectral analysis was performed with a PE-Sciex API 3000 triple quadrupole mass spectrometer (PE-Sciex, Concord, ON, Canada) using an ionspray interface operating in the positive or negative ion mode. The MS and MS/MS conditions were optimized using compound A. Characterization of metabolites was achieved using collisional-induced dissociation to obtain product ion mass spectra. The collision energy was 45 eV, and the collision gas setting was 4.

NMR Analysis. Compound A and the purified metabolite J were analyzed by ¹H NMR. The spectra were acquired in CD₃CN (0.13 ml) at 25°C (298 K) in 3 mm NMR tubes and acquired at 400 MHz on a Varian Unity 400 spectrometer using a Nalorac 3 mm indirect detection gradient probe (Varian Inc., Palo Alto, CA). The data processing was done on the spectrometer. Chemical shifts are reported on the  scale (parts per million) assigning the residual proton solvent peak to 1.93 ppm.

	Results and Discussion

Top Abstract Introduction Experimental Procedures Results and Discussion References

Product Ion Mass Spectra of the Authentic Standards. The product ion spectra of the parent compound (A and [¹⁴C]A), the O-dealkylated metabolite (B), the hydroxylamine metabolite (C), the nitrones (D and E), the lactam metabolite (F), and the hydroxylated O-dealkylated metabolite (G) are summarized in Table 1. Comparison of the product ion spectrum of compound A with that of ¹⁴C-labeled compound A indicates that the fragments at m/z 231, 215, 203 (relatively weak signal), 191, and 175 are associated with the trifluoromethoxy phenyl moiety, whereas the fragments at m/z 184, 172, 159, 131, 91, and 56 are associated with the phenyl piperidine moiety. Additional information is provided by the product ion mass spectrum of compound A acquired with the Q-TOF II mass spectrometer (Micromass Inc., Beverly, MA). The mass accuracy of the fragment ions achieved by the Q-TOF II facilitates interpretation of product ion spectra (Hop et al., 2001; Qin and Frech, 2001; Tiller et al., 2001). Table 2 summarizes the molecular formula assignments for the major fragment ions of [A + H]⁺. Tentative structural assignments for the most informative fragmentation pathways of the pseudo molecular ion of compound A are summarized in Scheme 2. The fragmentation pattern is complex, and additional labeling experiments are required for unambiguous structural assignment of all fragment ions. Nevertheless, comparison of the product ion spectra of the metabolites B to G with that of the parent compound A indicates that the product ion spectra are compatible with the assigned structures.

View larger version (16K): [in this window] [in a new window]   Scheme 2.   Tentative structural assignments for the most informative fragmentation pathways observed for protonated compound A upon collisional activation.

Rat Hepatocytes. Compound A was incubated for 4 h with freshly prepared rat hepatocytes. The radiochromatographic profile obtained with trifluoroacetic acid in the mobile phase is presented in Fig. 1. The turnover was about 30% and, therefore, the most abundant signal (at a retention time of about 23.8 min) is the parent compound, A. Three metabolites are visible in the radiochromatogram at retention times of 27.0, 29.8, and 32.8 min. By comparison of the product ion spectra of these metabolites with those of authentic standards, these metabolites were identified as the two nitrones (D and E) and the lactam (F), respectively. In addition, three additional minor metabolites were observed at retention times of 10.0, 15.2, and 32.8 min. The MS spectrum of the metabolite eluting at 10.0 min indicated that the molecular mass of this metabolite is 585 Da. Because of the low abundance of this metabolite, the only significant signal observed in its product ion spectrum is m/z 410, which corresponds to the loss of 176 Da. These data suggest that this metabolite (H) is a glucuronide of the hydroxylated and O-dealkylated metabolite (G, retention time = 13.0 min) or one of its isomers. The product ion spectrum of the metabolite at 15.2 min indicates that this was the O-dealkylated metabolite (B). By mass spectrometry, a molecular mass of 463 Da was obtained for the metabolite (I) eluting at 32.8 min and signals at m/z 175, 191, 203, and 215 in the product ion spectrum imply that 30 Da has been added to the phenyl-piperidine moiety of the molecule (see below for more details).

View larger version (12K): [in this window] [in a new window]   Fig. 1.   Radiochromatogram of compound A incubated with rat hepatocytes for 4 h. The liquid chromatography mobile phase contained trifluoroacetic acid.

Rat Plasma. Sprague-Dawley rats received a 10 mg/kg oral dose or a 5 mg/kg intravenous dose of [¹⁴C]compound A diluted with unlabeled compound A. The radiochromatogram with trifluoroacetic acid in the mobile phase for the 4 h postdosing i.v. plasma sample is presented in Fig. 2, and it is quite similar to the radiochromatogram of the 6 h postdosing p.o. plasma sample (not shown). The elimination half-life after intravenous dosing at 2 mg/kg was 6.2 ± 0.8 h. These radiochromatograms were remarkably different from those obtained with rat hepatocytes. Clearly, compound A was extensively metabolized, and the major component in the plasma was a new metabolite (J) eluting later (at 35.9 min) than all metabolites observed in rat hepatocytes. The mass spectrum of metabolite J contained signals at m/z 447, 465, 482, and 487, which correspond to [M + H  H₂O]⁺, [M + H]⁺, [M + NH₄]⁺, and [M + Na]⁺ ions. Thus, the molecular mass of metabolite J is 464 Da. Since the molecular mass is an even number, this metabolite must have lost (or gained) a nitrogen atom.

View larger version (13K): [in this window] [in a new window]   Fig. 2.   Radiochromatogram of rat plasma following intravenous dosing of compound A at 10 mg/kg. The liquid chromatography mobile phase contained trifluoroacetic acid.

View larger version (13K): [in this window] [in a new window]   Fig. 3.   Radiochromatogram of rat plasma following intravenous dosing of compound A at 10 mg/kg. The liquid chromatography mobile phase did not contain trifluoroacetic acid.

View larger version (9K): [in this window] [in a new window]   Scheme 3.   Keto acid structure.

View larger version (11K): [in this window] [in a new window]   Scheme 4.   Mechanism associated with formation of metabolite J via oxidative de-amination.