QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) Submit a response Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Cui, D. Articles by Prueksaritanont, T. Search for Related Content PubMed PubMed Citation Articles by Cui, D. Articles by Prueksaritanont, T.

IN VITRO AND IN VIVO METABOLISM OF A POTENT AND SELECTIVE INTEGRIN _vß₃ ANTAGONIST IN RATS, DOGS, AND MONKEYS

Drug Metabolism, Merck Research Laboratories, West Point, Pennsylvania (D.C., R.S., M.S., X.Y., J.A.Y., T.P.) and Rahway, New Jersey (M.A.W., M.P.B., B.H.A.)

(Received January 7, 2004; accepted May 18, 2004)

	Abstract

Top Abstract Materials and Methods Results Discussion References

 
Compound A (3-{2-oxo-3-[3-(5,6,7,8-tetrahydro-[1,8]naphthyrindin-2-yl)propyl]-imidazolidin-1-yl}-3(S)-(6-methoxy-pyridin-3-yl)-propionic acid), a potent and selective antagonist of integrin _vß₃ receptor, is under development for treatment of osteoporosis. This study describes metabolism and excretion of A in vivo in rats, dogs, and monkeys, and metabolism of A in vitro in primary hepatocytes from rats, dogs, monkeys, and humans. In all three animal species studied, A was primarily excreted as unchanged drug and, to a lesser degree, as phase I and phase II metabolites. Major biotransformation pathways of A included glucuronidation/glucosylation on the carboxylic group to form acyl-linked glucuronides/glucosides; and oxidation on the tetrahydronaphthyridine moiety to generate a carbinolamine and its further metabolized products. Minor pathways involved O-demethylation and hydroxylations on the alkyl chain. Only in rats, a glutathione adduct of A was also observed, and its formation is proposed to be via an iminium intermediate on the tetrahydronaphthyridine ring. Similar metabolic pathways were observed in the incubates of hepatocytes from the corresponding animals as well as from humans. CYP 3A and 2D subfamilies were capable of metabolizing A to its oxidative products. Overall, these in vitro and in vivo findings should provide useful insight on possible biotransformation pathways of A in humans.

View larger version (11K): [in this window] [in a new window]   FIG. 1. Structure of A. The position of carbon-14 label is indicated with .

Metabolism studies of drug candidates have become an integral part of the drug discovery and development process (Lin et al., 2003; Roberts, 2003). Results generated from these studies have been commonly used in designing chemicals that are metabolically more stable and often with better pharmacokinetic properties. Studies on the in vivo metabolic pathways of drug candidates in animals, together with in vitro metabolism information generated using animal and human liver preparations, often provide valuable foresight on metabolic pathways in humans. Detailed biotransformation studies of drug candidates in laboratory animals also provide guidance for selection of animal species used in safety assessment studies, to ensure that the selected species are exposed to all major drug metabolites formed in humans (Baillie et al., 2002).

Pharmacokinetics of A in rats, dogs, and monkeys have been previously reported (Prueksaritanont et al., 2004). After i.v. administration, A was shown to display a marked species difference in plasma clearance which ranged from 47 ml min^-1 kg^-1 in rats to 4 to 9 ml min^-1 kg^-1 in dogs and monkeys. Bioavailability of A was approximately 20% in rats and 70 to 80% for dogs and monkeys. In the current report, in vivo and in vitro metabolism of A was evaluated to study whether biotransformation of A contributed to the species difference in the pharmacokinetic parameters, and to support safety assessment studies in the clinical development program of A. The results reported in this article include the in vivo metabolism of A in rats, dogs, and monkeys, and in vitro experiments using primary hepatocytes from rat, dog, monkey, and human. Potential human biotransformation pathways of A are also proposed. In addition, incubations using cDNA-expressed human P450s were performed to elucidate the P450s involved in the oxidation of A. Finally, the mechanisms for the formation of a GSH adduct in rats and a lactam metabolite, a major metabolite observed in all species, were also studied.

	Materials and Methods

Top Abstract Materials and Methods Results Discussion References

 
General Chemicals. All reagents were purchased from Sigma-Aldrich (St. Louis, MO), except where indicated otherwise. Solvents used for HPLC analysis were of HPLC or analytical grade. Hepatocytes and cDNA-expressed P450s were prepared in house. Tru-Count (IN/US Systems, Tampa, FL) was used as the scintillation cocktail in on-line radioactivity detection. Ready Safe (Beckman Coulter, Fullerton, CA) was used for liquid scintillation counting. ¹⁴C-labeled A was synthesized with the radiolabel incorporated on the C-2 carbon of the imidazolidine ring (Fig. 1). It has a specific activity of 31.0 µCi/mg and a radiochemical purity >98% as determined by radio-HPLC. Authentic standards of A, metabolites M1 (O-desmethyl-A), M3c (7-hydroxytetrahydronaphthyridin-A), M6b (7-oxo-tetrahydronaphthyridin-A), M7 (naphthyridine derivative of A), and M8 (acyl glucuronide of A) were prepared as described previously (Hutchinson et al., 2003).

Animal Studies. All studies were conducted according to protocols approved by the Merck Animal Care and Use Committee. Male Sprague-Dawley rats, male beagle dogs, and male and female rhesus monkeys were used in the study. The animals were fasted 18 h before the dose administration and were fed 8 h after the dose. Water was available to the animals during the course of all studies. A cannular port system was surgically implanted on the animals, which allowed continuous collection or recirculation of bile. Specimens of bile, urine, and plasma samples were collected over dry ice from all animals at various time points. To minimize the degradation of the acyl-glucuronide metabolite, all samples collected were mixed with glacial acetic acid (5%) or 1 N HCl (3–4%) postcollection. The samples were kept frozen at -20°C until analysis. The intravenous (i.v.) dosing solution of ¹⁴C-A was prepared by dissolving ¹⁴C-labeled and cold A into phosphate-buffered saline solution. The final specific activities of the dosing solutions prepared for rats, dogs, and monkeys were approximately 31.0 µCi/mg, 4.2 µCi/mg, and 6.4 µCi/mg, respectively. Each animal received approximately 150 µCi of radioactivity.

Rats. A bolus dose of 13 mg/kg was administered i.v. to male rats (n = 3). Due to the higher clearance observed in rats than in dogs and monkeys (Prueksaritanont et al., 2004), a higher dose level was chosen for rats to ensure similar levels of exposure of A across species. Blood samples were taken via a cannulated jugular vein at 0, 30, 120, 240, and 420 min postdosing. Each blood sample (1 ml) was then mixed with heparin and 0.03 ml of 0.1 N HCl, and centrifuged to obtain plasma. Rat urine samples were collected at 0 to 8, 8 to 24, and 24 to 48 h postdosing. A 10-ml aliquot of urine samples from each time point was then mixed with 0.5 ml of glacial acetic acid before storage. Samples of rat bile were collected at 0 to 2, 2 to 4, 4 to 7, 7 to 24, and 24 to 48 h after administration. An aliquot from each bile sample (5 ml) was then mixed with 0.2 ml of 1 N HCl before storage. The acidified urine and bile samples were used for metabolite identification.

Dogs. A bolus dose of 1 mg/kg was given to each dog (n = 2) via the cephalic vein. Blood samples were obtained from the jugular vein at 0, 0.5, 1, 2, 4, 6, and 24 h postdosing. Each blood sample (8 ml) was placed in a 10-ml draw Vacutainer tube containing heparin and 0.1 ml of 1 N HCl. Plasma was obtained through centrifugation and stored at -20°C. Dog urine samples were collected for the 0- to 8-, 8- to 24-, and 24- to 48-h time periods postdosing. Bile samples were obtained for the 0- to 2-, 2- to 4-, 4- to 8-, 8- to 24-, and 24- to 32-h time points. Aliquots of both dog urine and bile samples were fortified with acid the same way as described above.

Monkeys. A bolus dose of 1 mg/kg was given to each monkey (n = 2, one male and one female) via the cephalic vein. Blood samples were obtained from the cannulated iliac artery at 0, 0.5, 1, 2, 4, 6, and 24 h postdosing. Each blood sample (5 ml) was placed in a 7-ml draw Vacutainer tube containing heparin and 0.1 ml of 1 N HCl. Plasma was obtained through centrifugation and stored at -20°C. Monkey urine samples were collected at 0 to 3, 3 to 6, 6 to 24, 24 to 48, and 48 to 72 h after dosing. Bile samples were obtained for 0 to 3, 3 to 6, 6 to 24, 24 to 32, 32 to 48, 48 to 56, and 56 to 72 h. Aliquots of both monkey urine and bile samples were fortified with acid the same way as described above.

Hepatocyte Incubations. Viable hepatocytes from rats, dogs, monkeys, and humans were isolated using standard methods (Pang et al., 1995). Cells (5 million cells/ml) were suspended in Hanks' balanced salt solution with 10 mM HEPES (pH 7.4) and were incubated with A (10 and 100 µM) or metabolite M3c (7-OH-A, 50 µM) for 0 or 3 h. The incubations were terminated with the addition of 2 ml of acetonitrile acidified with 5% glacial acetic acid. The entire volume was vortex mixed and centrifuged, and the supernatant was dried under a stream of nitrogen gas (TurboVap; Zymark Corp., Hopkinton, MA). The residues were then reconstituted into 30% acetonitrile in water with 0.1% formic acid and analyzed by HPLC/MS.

Incubation with Recombinant P450s. Incubations with cDNA-expressed P450s were conducted in potassium phosphate buffer (100 mM, pH 7.4). Each incubation mixture (1-ml final volume) contained 100 pmol of P450 isozyme, 50 µM substrate, and 2 mM NADPH. The mixtures were incubated at 37°C for 1 h. Reactions were stopped with the addition of 2 ml of acetonitrile, vortex-mixed, and centrifuged. The supernatants were then transferred into new test tubes and dried under nitrogen, and the resultant residues were reconstituted into 30% acetonitrile in water and analyzed by HPLC/MS.

Preparation of Urine, Bile, and Plasma Samples for Metabolite Profiling. Aliquots of urine samples (1 ml) from rats, dogs, and monkeys were transferred to 1.5-ml microcentrifuge tubes and centrifuged at 14,000 rpm for 5 min. The supernatants were transferred onto autosampler vials for HPLC analysis. Aliquots of plasma (0- to 30-min) and bile (0- to 2- or 0- to 3-h) samples collected from each animal were fortified with 4 volumes of acetonitrile acidified with 1% formic acid to remove proteins. The mixture was vortex mixed and centrifuged, and the supernatants were transferred into new test tubes and evaporated to dryness in a centrifugal evaporator (SpeedVac; Thermo Savant, Holbrook, NY). The residues were reconstituted to the initial sample volume with 20% acetonitrile in water (acidified with 0.1% formic acid) and analyzed by HPLC-tandem MS (MS/MS)-radioactivity detection (described below). Recovery of radioactivity after protein precipitation was between 87 and 95% for plasma and bile samples from all three species.

Liquid Scintillation Counting (LSC). The amount of ¹⁴C radioactivity in liquid samples was measured using a liquid scintillation counter (LS 6000CE liquid scintillation spectrometer; Beckman Coulter). Aliquots (50- to 200-µl) of urine, plasma, or bile were mixed with scintillation cocktail (15 ml) and counted using LSC. The amount of radioactivity in urine and bile samples obtained at each time point was expressed as a percentage of the dose excreted in each sampling period.

HPLC-MS/MS Analysis. HPLC analysis was performed on a reverse phase HPLC system (Agilent 1050; Agilent Technologies, Palo Alto, CA). Separations were achieved at a constant flow rate of 1.0 ml/min on a Phenomenex (Torrance, CA) C18(2) column (4.6 mm x 25 cm, 5 µm) using a mobile phase consisting of 0.1% formic acid in water (solvent A) and 0.1% formic acid in acetonitrile (solvent B). The solvent gradient started with an isocratic stage of 10% B for 5 min followed by an increase to 15% B in 5 min. The gradient was then increased to 25% B in 20 min followed by another increase to 45% B in 5 min. The column was washed at 80% B for 3 min and equilibrated for 10 min at 10% B before the next injection.

The HPLC system described above was interfaced to a Finnigan TSQ 7000 tandem mass spectrometer. Mass spectral analyses were carried out using electrospray ionization (ESI) in the positive ion mode. The capillary temperature was 230°C and the ESI ionizing voltage was maintained at 5.0 kV for all analyses. MS/MS was based on collision-induced dissociation (CID) of ions entering the radio frequency-only octapole region where argon was used as the collision gas at a pressure of 1.7 mtorr. A cycle of three collision offset voltages (-25, -35, and -45 eV) was used in MS/MS analyses. In cases where the synthetic standards of metabolites were available, HPLC retention times and CID product-ion spectra of the metabolites were compared with those of the synthetic standards.

Accurate mass measurement of A was performed on a Micromass (Waters, Milford MA) quadrupole time-of-flight II mass spectrometer equipped with an orthogonal ESI source. The TOF analyzer was tuned and calibrated using a series of singly charged adduct ions generated from infusing 0.1% phosphoric acid into the ESI source. MS/MS spectrum of A was obtained using argon as the collision gas with the collision energy set at 25 eV. Accurate masses of the fragment ions were then corrected using the calculated mass of the precursor ion as an internal reference.

Radioactivity Quantification. The distribution of ¹⁴C-labeled metabolites of A in in vivo samples from rat, dog, and monkey was monitored using a flow-through radiochemical detector for HPLC (ß-RAM; IN/US Systems). HPLC effluent was split between the radiochemical detector and mass spectrometer at a ratio of 9 to 1. The ß-RAM was operated in the homogeneous liquid scintillation counting mode, and a 600-µl flow cell and a flow rate of 3 ml/min of scintillation cocktail were used for all the analyses. The percentage of the radiolabeled material for each chromatographic peak was obtained through integration of the peak, and the metabolites were quantified and expressed as percentage of the dose given to each animal. Recovery of radioactivity from the HPLC column was evaluated by comparing the amount of radioactivity injected (determined using LSC) and the amount of total radioactivity detected by ß-RAM.

Biosynthesis of Metabolites. Oxidative metabolites of A were generated using cDNA-expressed CYP2D6. The construction of baculovirus encoding the entire cDNA of CYP2D6 and P450-oxidoreductase (P450-OR) has been previously described (Rushmore et al., 2000). Sf21 insect cells were grown at 27°C in Grace's medium containing 10% fetal bovine serum to a density of 12 x 10⁶ cells/ml (200 ml in total) in 1-liter spinner flasks (Bellco Glass Inc., Vineland, NJ) with blades operating at 90 rpm. Cells were infected at approximately 1.0 multiplicity of infection of virus encoding the CYP2D6 and 0.1 multiplicity of infection of virus encoding P450-OR. Hemin medium (1 mg /ml) in the form of a hemin-albumin complex was added. After 48 or 60 h, P450 content was measured by the CO-difference spectrum. A was then added to the bioreactor at a final concentration of 100 µM and was incubated at 37°C for 6 h. At the end of the biosynthesis, the culture medium was recovered by centrifugation and stored at -70°C until further use.

Metabolite Isolation for NMR Analysis. Metabolite M6b was generated by incubating A with rat hepatocytes; and metabolites M3a, M3b, and M3c were produced using the bioreactor described above. Incubates were centrifuged and the supernatants were first purified by solid phase extraction (SPE; Varian Mega Bond Elut C18, 20 ml; Varian, Inc., Palo Alto, CA). Ten milliliters of the incubates were loaded onto each SPE column. After loading, the column was first washed with water (two times, 20 ml), and the metabolites were then eluted using acetonitrile/water (1:1; three times, 5 ml). The SPE fractions were then pooled and dried under nitrogen. The residues were reconstituted into 1 ml of 20% acetonitrile in water. Isolation of metabolites was achieved using a Micromass Mass Lynx HPLC-MS system, in which the chromatographic peaks were collected based on their molecular masses. The detailed procedures for metabolite isolation are described below.

Metabolites of A generated using rat hepatocytes or bioreactors were separated on a semipreparative HPLC system (Waters 600) using a Phenomenex Luna C18(2) preparation column (21.2 x 150 mm, 5 µm). The mobile phase, at a constant flow rate of 20 ml/min, consisted of 0.1% formic acid in water (solvent A) and 0.1% formic acid in acetonitrile (solvent B). The eluting gradient started with an isocratic condition of 15% B for 2 min followed by a linear increase to 20% B in 10 min. Solvent B was rapidly increased to 80% in the next minute, and the column was washed for 2 min at 80% B before returning to 15% B over 1 min. The system was equilibrated at 15% B for 10 min before the next injection. The isolated chromatographic peaks from multiple HPLC purifications were combined based on their retention times. The pooled fractions, with the exception of those containing M3c, were evaporated to dryness on a SpeedVac, and the resultant residues were used for NMR analysis. The pooled fractions containing M3c were concentrated to approximately 100 µl and then reconstituted to a total volume of 200 µl with 20% acetonitrile/80% H₂O for LC-NMR.

NMR and LC-NMR Analysis of Isolated Metabolites. NMR experiments were carried out on either a 400- or 500-MHz spectrometer (Inova; Varian, Inc.) fitted with a 3-mm inverse detection probe (MIDG-3 probe; Varian, Inc.). Samples of the isolated metabolites were dissolved in dimethyl sulfoxide-d₆, and NMR data sets were acquired at 25°C. All ¹H chemical shifts are reported on the scale (parts per million) downfield from tetramethyl silane using the dimethyl sulfoxide-d₆ lock signal for reference at 2.51 ppm.

Metabolite M3c was analyzed by LC-NMR. The LC-NMR setup consisted of a reverse phase HPLC (model 9012; Varian, Inc.) in-line with a photodiode array detector (model 9065; Varian, Inc.), followed by the NMR flow probe (IFC probe; Varian, Inc.) on a 500-MHz spectrometer. When the apex of an LC peak was detected by the photodiode array detector, the pump was stopped after a precalibrated delay time, resulting in the metabolite parked in the active volume (60 µl) of the NMR probe. The LC-UV conditions were as follows: UV monitor at 254 nm, C18 column (Supelco Discovery, 5 µm; 2.1 x 50 mm C18), 1 ml/min flow rate; and solvent gradient conditions: 0 to 5 min, 5 to 20% A; 5 to 7 min, 20 to 80% A; 7 to 8 min, 80% A; 8 to 9 min, 80% to 5% A, where mobile phase A is 100% CH₃CN and B is 100% D₂O. Under these conditions, A and M3c eluted at 5.6 min and 6.4 min, respectively. All NMR spectra were obtained under stopped flow conditions, and solvent-suppression techniques (Smallcombe et al., 1995) were used to suppress the CH₃CN resonance. A two-dimensional total correlation spectroscopy spectrum with an 80-ms mixing time (Summers et al., 1986) was also acquired.

	Results

Top Abstract Materials and Methods Results Discussion References

 
Excretion of ¹⁴C-Labeled A. After i.v. bolus administration of ¹⁴C-A to rats, the mean total recovery of radioactivity in urine and bile was 92% over a 48-h period. The majority of the radioactive dose was excreted in bile (80%), whereas only 12% was accounted for in urine (Table 1). In dogs, over the 32-h collection period, 62% of the radioactive dose was excreted in bile and 36% was in urine (Table 1). In monkeys, however, the majority of the radioactivity was recovered in urine (67%) and, to a lesser extent, in bile (27%; Table 1). In all three species, greater than 90% of the recovered dose was excreted in the first 24 h after administration of the drug (data not shown).

Metabolite Profiles in Urine, Bile, and Plasma. The representative HPLC-radioactivity profiles of metabolites in urine, bile, and plasma from rats, dogs, and monkeys that received [¹⁴C]-A are shown in Figs. 2 through 4, respectively. A small shift in HPLC retention times was observed in rat and dog bile samples (Figs. 2 and 3). The shift may be explained by the interactions of bile acids with A, which contains both a basic tetrahydronaphthyridine (pK_a 8) and a carboxylic acid (pK_a 4) moiety and may exist as zwitterions. In addition to unchanged parent drug, a total of 15 metabolites were identified in urine, bile, and plasma from the three species studied. The amounts of excreted parent drug and metabolites quantified as the percentage of the radioactive dose given to each animal are listed in Table 2. In all three species, unchanged A was the major component in both urine and bile. It accounted for 51 to 56% of the i.v. dose in rats and dogs (Table 2), and approximately 80% of the i.v. dose in monkeys (Table 2). A was also the major circulating component in plasma for all three species; less than 5% of the total radioactivity was attributed to metabolites (Figs. 2, 3, 4).

View larger version (25K): [in this window] [in a new window]   FIG. 2. Representative HPLC-radiochromatograms of metabolites of A in bile (0–2 h), urine (0–7 h), and plasma (0–30 min) of rats after a 13 mg/kg i.v. dose of 14C-A.

View larger version (22K): [in this window] [in a new window]   FIG. 4. Representative HPLC-radiochromatograms of metabolites of A in bile (0–2 h), urine (0–8 h), and plasma (0–30 min) of monkeys after a 1 mg/kg i.v. dose of 14C-A.

View larger version (23K): [in this window] [in a new window]   FIG. 3. Representative HPLC-radiochromatograms of metabolites of A in bile (0–2 h), urine (0–8 h), and plasma (0–30 min) of dogs after a 1 mg/kg i.v. dose of 14C-A.

Identification of Metabolites. A list of all the metabolites of A detected in rats, dogs, and monkeys, together with the associated information used in the identification, are summarized in Table 3. Proposed structures of the detected metabolites are shown in Fig. 5.

View larger version (24K): [in this window] [in a new window]   FIG. 5. Proposed metabolic pathways of A in rats, dogs, and monkeys. Dotted arrows indicated tentative pathways.

The ESI CID product-ion spectrum of A is displayed in Fig. 6A. Based on comparison with the CID spectrum of the ¹⁴C-labeled A (not shown) and the high-resolution mass spectral information (Table 4), tentative pathways for the formation of the most informative fragment ions of A (MH⁺ = 440) under CID is proposed in Fig. 6B. Fragmentation of A in the gas phase appears to occur predominantly at the two linkages around the oxo-imidazolidine ring, which give rise to the formation of fragment ions at m/z 261, 175, 180, and 138 (Fig. 6B). Detection of these ions in the CID spectra of metabolites indicated either the unmodified propyl-tetrahydronaphthyridine moiety (m/z 261 and 175, Fig. 6B) or the unchanged methoxy-pyridinyl propionic acid (m/z 180 and 138, Fig. 6B). Mass shifts on these ions in the CID spectra of metabolites were used to characterize the structural modifications in metabolites.

View larger version (15K): [in this window] [in a new window]   FIG. 6. A, CID product-ion spectrum of A obtained on a triple quadrupole mass spectrometer. B, tentative structures of the most informative fragment ions of A (MH+ = 440) under CID.

Metabolite M1 was only detected in dogs. It had an MH⁺ ion at m/z 426, and MS/MS scans of m/z 426 generated fragment ions at m/z 261, 175, 166, and 124 (Table 3). M1 displayed an HPLC retention time and MS/MS patterns identical to those of the synthetic O-desmethyl-A. M1 was, therefore, identified as the O-desmethyl metabolite of A.

Metabolites M2a and M2b had HPLC retention times of 17 and 23 min, respectively. Both metabolites showed an MH⁺ ion at m/z 632, indicating that they were isomers. M2a was present in both rats and monkeys, whereas M2b was only detected in rats. CID product-ion spectra of m/z 632 from both metabolites contained major fragmentation ions at m/z 456, 277, 191, and 180 (Table 3). The neutral loss of 176 mass units (m/z 632 to m/z 456) pointed to the presence of a glucuronic acid moiety. Fragment ions at m/z 277 and 191 suggested that hydroxylation was on the tetrahydronaphthyridine part of the molecule. The fragment ion at m/z 180 indicated that the methoxypyridine moiety was unchanged. Therefore, M2a and M2b were tentatively identified as glucuronide conjugates of hydroxylated A. The exact positions of hydroxylation and glucuronidation were not characterized.

Metabolites M3a (in rat and dog), M3b (in rat, dog, and monkey), M3c (in rat and monkey), M3d (in rat), and M3e (in rat) all showed a protonated molecule at m/z 456, an addition of 16 Da to A, suggesting that these metabolites were isomers of monooxidized A. CID product-ion spectra of these metabolites all contained fragment ions at m/z 180 and 138 (Table 3), indicating that the methoxy-pyridine propionic acid portion was intact. Other major fragments for M3a, M3b, M3c, and M3d all included ions at m/z 259 and 173, suggesting an addition of the hydroxyl group on the propyl-tetrahydronaphthyridine moiety followed by loss of a water molecule. Considering the readiness of water loss upon CID in the collision cell (data not shown), the site of oxidation on these metabolites was proposed to be on one of the saturated carbons of the propyl-tetrahydronaphthyridine group. The major fragment ions of metabolite M3e (m/z 277 and 191; Table 3) did not show any loss of water during the CID process; therefore, hydroxylation of M3e was determined to be on the aromatic portion of the tetrahydronaphthyridine ring.

To pinpoint the site of hydroxylation, metabolites M3a, M3b, and M3c were isolated as described under Materials and Methods. Fragmentation patterns and HPLC retention times of the isolated metabolites from in vitro incubations were identical to those obtained from animals (data not shown). Proton chemical shifts for metabolites M3a, M3b, and M3c are summarized in Table 5. Comparing the ¹H one-dimensional spectra of A and M3a indicated that one of the methylenes in the naphthyridine ring in A had been hydroxylated. A new resonance consistent with a CHOH proton appeared at 4.62 ppm. When the 4.62-ppm peak is irradiated in a one-dimensional nuclear Overhauser effect difference experiment, an enhancement was seen for H4 (7.75 ppm), which established the OH group at C5 of the naphthyridine ring (labeling indicated in Fig. 1). The ¹H one-dimensional spectrum of M3b showed a close similarity with the M3a spectrum (H5 at 4.60 ppm) and was inferred to be a diastereomer, with hydroxylation at C5, of M3a.

View this table: [in this window] [in a new window]   TABLE 5 Identifiable 1H chemical shifts for A, M3a, M3b, M3c, and M6b All values are referenced to the CD2HCN resonance set at 1.95 ppm. M3c values are referenced to the O-Me peak, which was set to 3.87 ppm.

LC-MS/MS analysis of M3c (CID fragment ion at m/z 173) indicated that hydroxylation was on the tetrahydronaphthyridine group. M3c was also analyzed by LC-NMR. A new resonance consistent with a CHOH proton was observed at 5.24 ppm in the M3c spectrum. The site of mono-hydroxylation was narrowed to one of the following positions: 5, 7, 3', or 1' (Fig. 1). Hydroxylation at C5 was ruled out since the resulting diastereomers (M3a and M3b) were both previously identified. Hydroxylation at carbon 3' or 1' was also ruled out based on the MS/MS data and the _H shift changes (Table 5) when compared with the spectrum of A. The total correlation spectroscopy spectrum of M3c showed two distinct connectivities: 5.24–2.75–1.70 ppm and 3.10–3.06–2.73–2.61–1.85 ppm, and they were assigned to CH (7), CH₂ (6), CH₂ (5), and CH₂ (1'), CH₂ (2'), CH₂ (3'), respectively. M3c was therefore identified as the 7-OH-A. Based on the proposed structure, a synthetic standard of M3c was later synthesized (Hutchinson et al., 2003), and the metabolite displayed MS and LC-NMR spectra identical to those of the synthetic M3c.

Metabolite M4 was only detected in rat bile and accounted for approximately 4% of the dose given to rats. M4 showed an MH⁺ ion at m/z 745 in full-scan MS analysis. A CID product-ion spectrum of m/z 745 displayed major fragment ions at m/z 438, 259, 180, and 173 (Table 3). The neutral loss of 307 Da from the molecular ion (m/z 745 to m/z 438) suggested the presence of a glutathione group. Other fragment ions were all 2 mass units lower than the corresponding fragments produced by the parent molecule, suggesting that, other than adding a glutathione group, no additional modifications had occurred on the parent molecule. Based also on the results from the in vitro metabolism studies on M3c (described below), M4 was tentatively identified as the 7-GS-A. Metabolite M4 appeared to be unstable in acidified rat bile, because it was not detected by LC/MS when left overnight at room temperature (data not shown).

M5 was detected only in dog bile. Full MS scans of M5 produced an MH⁺ ion of m/z 602. CID of m/z 602 showed a neutral loss of 162 mass units, and the other fragment ions were identical to that of A (Table 3). The addition of 162 Da to the parent drug suggested that M5 was a glucose conjugate of A.

Metabolites M6a and M6b both had an MH⁺ ion at m/z 454 in full MS scans, indicating that they were isomers. M6b accounted for 2% of the dose in rats and dogs and about 4% of the dose in monkeys. CID scans of m/z 454 from M6b showed major fragmentation ions at m/z 275, 189, 180, and 138 (Table 3). M6b displayed HPLC retention time, MS/MS spectrum, and NMR (_H values listed in Table 5) identical to those of the synthetic standard that had an oxo group at the C7 of the tetrahydronaphthyridine ring. Therefore, M6b was identified as 7-oxo-A. Metabolite M6a was detected in rats and dogs (<1% of the dose). It displayed MS/MS fragmentations identical to those of M6b, although a much shorter HPLC retention time was observed for M6a. M6a was tentatively identified as another oxo-derivative of A.

Metabolite M7 was detected in all three species (Figs. 2, 3, 4). Full MS scans of M7 produced a molecular ion of m/z 436, which was 4 mass units lower than that of A. CID scans of m/z 436 showed major fragmentation ions at m/z 257, 180, 171, and 138 (Table 3). M7 displayed HPLC retention time and MS/MS spectrum identical to those of the synthetic standard that had a naphthyridine, instead of the tetrahydronaphthyridine, ring; and therefore, M7 was identified as the naphthyridine derivative of A.

M8 accounted for 9, 19, and 2% of the dose in rats, dogs, and monkeys, respectively (Table 2; Figs. 2, 3, 4). It had a molecular ion of m/z 616 in full mass scans. CID scans of m/z 616 produced a base fragment ion at m/z 440 (Table 3), suggesting a loss of the neutral glucuronic acid moiety (neutral loss of 176 Da). Metabolite M8 showed the same MS/MS spectrum and HPLC retention time as the synthetic standard of the acyl-linked glucuronide of A. M8 is, therefore, identified as the acyl glucuronide of A. In bile samples, there were multiple closely eluting radioactive peaks, all displaying the same molecular ion at m/z 616. CID product-ion spectra of these peaks indicated the presence of a glucuronide moiety (data not shown), and they were tentatively identified as the acyl-migration products of M8.

Metabolite M9 was only detected in rats. It showed an MH⁺ ion at m/z 630 in full MS scans. CID product-ion scans generated major fragment ions at m/z 454, 275, 189, and 180 (Table 3). The neutral 176-Da loss (m/z 630 to m/z 454) indicated the presence of a glucuronic acid group. The remaining fragments suggested an unchanged methoxy-pyridine propionic acid and an addition of an oxo-group to the tetrahydronaphthyridine ring. Therefore, M9 was tentatively identified as the glucuronide of one of the oxo-derivative of A.

Hepatocyte Incubations. Representative HPLC-radiochromatograms of metabolites of A (100 µM) formed in incubations with rat, dog, monkey, and human hepatocytes are shown in Fig. 7. No significant difference was observed in metabolite profiles in all four species when a lower concentration of A (10 µM) was used (data not shown). A was more extensively metabolized in rat hepatocytes, and least in human hepatocytes. Based on HPLC retention times and MS/MS fragmentation patterns, major metabolites detected in rat, dog, and monkey hepatocytes were also found in vivo in the corresponding species. The metabolites identified in human hepatocytes included M3c, M6b, M7, and M8, and these metabolites were also observed in the other three species.

View larger version (24K): [in this window] [in a new window]   FIG. 7. Representative HPLC-radiochromatograms of metabolites of A in hepatocyte incubates from rat, dog, monkey, and human after incubating with 14C-A (100 µM) for 4 h at 37°C.

To study the metabolism of the carbinolamine metabolite M3c, its synthetic standard was incubated with freshly isolated rat hepatocytes. An extracted ion chromatogram of ions at m/z 436, 438, 454, 456, and 745 from MS full-scan analysis of the incubate is presented in Fig. 8A. Three metabolites, M4 (m/z 745), M7 (m/z 436), and M6b (m/z 454), all found in rat in vivo, were also detected in the rat hepatocyte incubation. A minor metabolite that has an MH⁺ ion of m/z 438, 2 mass units lower than that of the parent molecule, was also detected (Fig. 8A). Interpretation of the MS/MS spectrum of m/z 438 revealed that a desaturation (loss of 2 protons) had occurred on the 2-propyl tetrahydronaphthyridine moiety, and its proposed structure is shown in Fig. 8A.

View larger version (25K): [in this window] [in a new window]   FIG. 8. A, extracted ion chromatogram of ions at m/z 436, 438, 454, 456, and 745 from HPLC-ESI full mass scans for metabolites formed after incubating M3c (50 µM) with rat hepatocytes for 3 h at 37°C. B, extracted ion chromatogram of ions at m/z 426, 436, 440, 454, and 456 from HPLC-ESI full mass scans for metabolites formed after incubating A (100 µM) with cDNA-expressed human CYP2D6 for 1 h at 37°C. Identities of metabolites were proposed based on MS/MS analyses and comparison with similar metabolites identified in vivo.

Metabolism by cDNA-Expressed P450s. Oxidation of A was evaluated in vitro using a series of cDNA-expressed human P450s. An extracted ion chromatogram of ions at m/z 426, 436, 440, 454, and 456 from full MS scan analysis of the CYP2D6 incubation is presented in Fig. 8B. Both CYP3A4 and CYP2D6 were able to metabolize A to several of its oxidative metabolites (including M3a, M3b, M3c, and M6b; Fig. 8B), whereas other P450s (CYP1A2, 2A6, 2B6, 2C8, 2C9, and 2C19) showed little activity. Due to limited metabolism of A in human liver microsomes, additional studies were not conducted to further assess the relative contribution of CYP3A and CYP2D6 to the overall oxidative metabolism of A.

	Discussion

Top Abstract Materials and Methods Results Discussion References

 
Metabolite profiles of urine, bile, and plasma obtained after i.v. administration of ¹⁴C-A (Figs. 2, 3, 4) revealed that A was not extensively metabolized; unchanged drug accounted for 51%, 56%, and 76% of the administered dose in rats, dogs, and monkeys, respectively. These results indicated that direct biliary and urinary excretion of parent drug was the major route of elimination for A in animals. A was also shown to be the major circulating component in plasma from all three species. Biotransformation is therefore unlikely to contribute to the observed species difference in plasma clearance in rats, dogs, and monkeys (Prueksaritanont et al., 2004). In general, qualitative in vivo metabolic profiles of A in rats, dogs, and monkeys were similar. Notable exceptions included metabolites M4 (in rat only) and M5 (in dog only). The metabolite profiles of A observed in vivo also were qualitatively similar to those observed in hepatocytes from the corresponding species, suggesting that results obtained from human hepatocyte incubations in this study may provide a valuable insight on biotransformation pathways of A in humans. In this regard, although the turnover rate of A was slower in human than in animal hepatocytes, all metabolites observed in hepatocytes from humans also were detected in those from rats, dogs, and monkeys.

In the present study, HPLC-MS/MS was used as the main tool for structural identification of metabolites of A. NMR, HPLC-NMR, and chemical synthetic standards were also utilized in the detailed characterization of metabolite structures. A total of 15 metabolites were identified in all the species studied. Since A was administered to animals as a pure enantiomer with S-configuration, hydroxylation on one of the prochiral carbons may produce diastereomers which could then be separated on an achiral HPLC system as different metabolites (Herraez-Hernandez et al., 2002). However, due to the low extent of metabolism and the small amount of metabolites produced, detailed characterizations on the stereochemistry of the metabolites were not attempted. The plausible biotransformation pathways of A in rats, dogs, and monkeys are summarized in Fig. 5. Formation of M8, an acyl-linked glucuronide of A, accounted for 10 to 20% of the dose excreted in rats and dogs, and about 2% of the dose in monkeys (Table 2). Glucuronidation has been shown to be a common pathway for many drugs that contain a carboxylic acid group (Spahn-Langguth and Benet, 1992). The acyl-linked glucuronides have been shown to undergo acyl-migration and are more susceptible to chemical hydrolysis than the O-linked glucuronides (Dutton, 1978). Since both the acyl-migration and the hydrolysis reactions are enhanced under basic conditions (Faed, 1984), all of our samples were acidified during collection and preparation processes to minimize the degradation of the metabolites. However, a small degree of acyl-migration was still observed (Figs. 2, 3, 4; the acyl-glucuronide and its migration products were all designated as M8). Under our experimental conditions, chemical hydrolysis of the acyl-glucuronide back to the parent molecule has been shown to be minimal (data not shown). These results suggested that the detection of large amounts of unchanged A in both urine and bile samples was likely due to the direct excretion of the parent molecule, rather than an ex vivo degradation of the acyl-linked glucuronide.

A relatively stable carbinolamine on the tetrahydronapthyridine moiety, M3c, was detected and represented approximately 5 to 8% of the dose in rats and monkeys (M3c was not detected by radiochemical detection in dogs; Table 2). It was also detected as a circulating metabolite in rat plasma. Two possible mechanisms for the formation of carbinolamines have been proposed. One suggests a single-electron transfer mechanism that involves the initial formation of a nitrogen radical cation followed by deprotonation and a stepwise electron reorganization (Guengerich, 2001). Other studies, such as the recent publication on cyclopropylamine by Shaffer et al. (2002), provide evidence for the mechanism that involves an initial hydrogen atom abstraction from the -carbon followed by a direct hydroxyl group rebound. In the current study, we did not investigate the mechanism for the formation of metabolite M3c, although in vitro experiments using cDNA-expressed human P450s indicated that M3c formation could be mediated by both CYP3A and CYP2D subfamilies.

In general, carbinolamines are unstable and rapidly undergo nonenzymatic dissociations to give rise to the final amine and ketone/aldehyde products in N-dealkylation reactions (Rose and Castagnoli, 1983). Despite the facile decomposition of carbinolamines, their detection and isolation as drug metabolites have been reported in the literature (Ebner et al., 1991; Walles et al., 2002), suggesting that some carbinolamines are relatively stable. Electron-withdrawing groups attached at the -carbon have been shown to increase the stability of carbinolamine metabolites (Upthagrove and Nelson, 2001). Carbinolamines formed on nitrogen-containing heterocyclic rings, similar to M3c, are also shown to be more stable than their aliphatic counterparts (Vickers and Polsky, 2000). In our studies, metabolite M3c was also found to be relatively stable in acidified bile and plasma samples (pH 2) but was subject to degradation during the drying process post isolation. For this reason, LC-NMR proved to be a valuable tool in the identification of M3c.

Consistent with the finding that M3c is a relatively stable carbinolamine, metabolic products of M3c were also detected in both in vivo and in vitro systems. When a synthetic standard of M3c was incubated with rat hepatocytes, major metabolites identified by mass spectrometry were M6b (a lactam), M7 (a naphthyridine derivative of A), and M4 (a glutathione adduct of A). A minor metabolite, with a loss of water from the 2-propyl tetrahydronaphthyridine moiety, also was detected (Fig. 8A). A mechanism for the formation of these products through the carbinolamine metabolite is proposed in Fig. 9. Further oxidation of carbinolamines by P450 to form stable lactam metabolites has been reported (Tanaka et al., 1995; Walker et al., 1999). In the current study, formation of M6b by expressed CYP 3A and 2D provided additional evidence for the involvement of P450s in the generation of lactams. Another possible mechanism for the formation of lactam metabolite is an aldehyde oxidase-catalyzed oxidation of the iminium intermediates, which may exist through equilibrium with the carbinolamine (Fig. 9). Examples of the involvement of aldehyde oxidase in the formation of lactams have been shown in the oxidation/detoxification of 3-hydroxy-3-methylindolenine, an iminium metabolite of 3-methylindole (Skordos et al., 1998), and in the oxidation of phenothiazine-containing antipsychotic drugs (Lin et al., 1996). In our studies, the lactam metabolite, M6b, was detected as one of the major metabolites in rats, dogs, and monkeys (Table 2). Because dogs have been shown to lack aldehyde oxidase activity (Beedham et al., 1987), formation of M6b in dogs is most likely to be catalyzed by P450s. However, the possible involvement of aldehyde oxidase in the generation of M6b in rats, monkeys, and humans cannot be excluded.

View larger version (13K): [in this window] [in a new window]   FIG. 9. Proposed mechanism for the biotransformation of a carbinolamine metabolite (M3c) and formation of a GSH adduct of A. [CYP], P450; [AO], aldehyde oxidase; [O], oxidation.

Glutathione conjugation is usually considered a detoxification pathway for electrophilic and chemically reactive drug metabolites (Armstrong, 1997). Structural characterization of GSH adducts can provide information on the identity of the reactive intermediates. In this study, a glutathione conjugate of A was detected as a biliary metabolite in rats, but not in other species. The same GSH adduct was also formed when the carbinolamine metabolite M3c was incubated with rat hepatocytes. Position of the GSH attachment (at carbon-7) was tentatively determined based on MS/MS analysis and also results from M3c metabolism in rat hepatocytes. In the CID product-ion spectrum, we did not observe the characteristic fragmentation patterns of GSH conjugates (Baillie and Davis, 1993). Instead, a direct neutral loss of the glutathione moiety (307 Da) from the GSH adduct was observed (Table 3). This can be explained by the cleavage of the carbon-sulfur bond with hydrogen migration and charge retention to form a relatively stable iminium ion and a neutral GSH. The identification of the GSH conjugate of A also provides evidence for the possible presence of a reactive iminium intermediate (Fig. 9). GSH adduct formation through iminium intermediates has been reported for several drugs bearing nitrogen-containing heterocyclic moieties (Streeper et al., 1997; Iverson et al., 2002). Consistent with these findings, we also observed that the GSH conjugate of A is unstable. Degradation of the GSH adduct may involve the formation of an iminium ion and its degradation products (Fig. 9). Since the initial amount of the GSH adduct detected was very small, the concomitant increase of the products resulting from it could not be quantified. Hence, the exact pathway for the decomposition of the GSH metabolite requires further investigation.

Formation of a naphthyridine metabolite of A (M7) was detected in all three species studied. It was also formed in incubations of the carbinolamine metabolite M3c with rat hepatocytes. As proposed in Fig. 9, the mechanism for the formation of M7 might involve an imine metabolite (detected in the rat hepatocyte incubation with M3c). Dehydrogenation of the imine to form the aromatized A could be either an enzyme-mediated oxidation reaction or an autoxidation process as described in the metabolism of compounds containing a tetrahydropyridinyl moiety (Inoue et al., 1999).

In summary, direct biliary and urinary excretion of unchanged A represented the major elimination pathway in rats, dogs, and monkeys. Biotransformation of A to both phase I and II metabolites was found to be a relatively minor route of elimination in all three species. Major metabolic pathways of A in animals included glucuronidation (and glucosylation in dogs) on the carboxylic group to form acyl-linked glucuronides (and glucosides), and oxidation on the tetrahydronaphthyridine moiety to generate a carbinolamine and its further metabolized products. In the three animal species studied, the metabolite profiles obtained in vivo were represented well by those observed in hepatocytes. Based on these results and the findings that the metabolite profiles of A in both animal and human hepatocytes were qualitatively similar, it may be anticipated that in humans, A may also undergo minimal metabolism to metabolites similar to those observed in animals.

	Acknowledgments

 
We thank Dr. Kelem Kassahun for helpful discussions, Dr. Dennis Dean and Yolunda Jakubowski for the synthesis and purification of the ¹⁴C-labeled A, and Dr. John H. Hutchinson and Wasyl Halczenko for providing synthetic standards for metabolites of A. We also thank Kristie Strong-Basalyga for conducting the hepatocyte incubations and Polly Deluna, Yuan Meng, Kim Michael, and Janice Brunner for help in conducting the animal studies.

	Footnotes

 
ABBREVIATIONS: compound A, 3-{2-oxo-3-[3-(5,6,7,8-tetrahydro-[1,8]naphthyrindin-2-yl)propyl]-imidazolidin-1-yl}-3(S)-(6-methoxypyridin-3-yl)propionic acid; CID, collision-induced dissociation; GSH, glutathione; HPLC, high-performance liquid chromatography; LC, liquid chromatography; MS/MS, tandem mass spectrometry; LSC, liquid scintillation counting; NMR, nuclear magnetic resonance spectroscopy; P450, cytochrome P450; ESI, electrospray ionization; SPE, solid phase extraction.

Address correspondence to: Dr. Donghui Cui, WP75A-203, Department of Drug Metabolism, Merck & Co., Inc., West Point, PA 19486. E-mail: dan_cui{at}merck.com

	References

Top Abstract Materials and Methods Results Discussion References

 


Armstrong RN (1997) Structure, catalytic mechanism and evolution of the glutathione transferases. Chem Res Toxicol 10: 2-18.[CrossRef][Medline]

Baillie TA, Cayen MN, Fouda F, Gerson RJ, Gree JD, Grossman SJ, Klunk LJ, Leblanc B, Perkins DG, and Shipley LA (2002) Drug metabolites in safety testing. Toxicol Appl Pharmacol 182: 188-196.[CrossRef][Medline]

Baillie TA and Davis MR (1993) Mass spectrometry in the analysis of glutathione conjugates. Biol Mass Spectrom 22: 319-325.[CrossRef][Medline]

Beedham C, Bruce SE, Critchley DJ, al Tayib Y, and Rance DJ (1987) Species variation in hepatic aldehyde oxidase activity. Eur J Drug Metab Pharmacokinet 12: 307-310.[Medline]

Dutton GJ (1978) Developmental aspects of drug conjugation, with special reference to glucuronidation. Annu Rev Pharmacol Toxicol 18: 17-35.[CrossRef][Medline]

Ebner T, Meese CO, and Eichelbaum M (1991) Regioselectivity and stereoselectivity of the metabolism of the chiral quinolizidine alkaloids sparteine and pachycarpine in the rat. Xenobiotica 21: 847-857.[Medline]

Faed EM (1984) Properties of acyl glucuronides: implications for studies of the pharmacokinetics and metabolism of acidic drugs. Drug Metab Rev 15: 1213-1249.[Medline]

Guengerich FP (2001) Common and uncommon cytochrome P450 reactions related to metabolism and chemical toxicity. Chem Res Toxicol 14: 611-650.[Medline]

Hartman GD and Duggan ME (2000) Alpha (v) beta(3) integrin antagonists as inhibitors of bone resorption. Expert Opin Investig Drugs 9: 1281-1291.[CrossRef][Medline]

Herraez-Hernandez R, Campins-Falco P, and Verdu-Andres J (2002) Strategies for the enantiomeric determination of amphetamine and related compounds by liquid chromatography. J Biochem Biophys Methods 54: 147-167.[Medline]

Hutchinson JH, Halczenko W, Brashear KM, Breslin MJ, Coleman PJ, Duong le T, Fernandez-Metzler C, Gentile MA, Fisher JE, Hartman GD, et al. (2003) Nonpeptide _vß₃ antagonists. 8. In vitro and in vivo evaluation of a potent _vß₃ antagonist for the prevention and treatment of osteoporosis. J Med Chem 46: 4790-4798.[Medline]

Inoue H, Castagnoli K, Van Der Schyf C, Mabic S, Igarashi K, and Castagnoli N Jr (1999) Species-dependent differences in monoamine oxidase A and B-catalyzed oxidation of various C4 substituted 1-methyl-4phenyl-1,2,3,6-tetrahydropyridinyl derivatives. J Pharmacol Exp Ther 291: 856-864.[Abstract/Free Full Text]

Iverson S, Zahid N, and Uetrecht JP (2002) Predicting drug-induced agranulocytosis: characterizing neutrophil-generated metabolites of a model compound, DMP 406 and assessing the relevance of an in vitro apoptosis assay for identifying drugs that may cause agranulocytosis. Chem Biol Interact 142: 175-199.[Medline]

Lin G, Chu KW, Damani LA, and Hawes EM (1996) Identification of lactams as in vitro metabolites of piperidine-type phenothiazine antipsychotic drugs. J Pharm Biomed Anal 14: 727-738.[Medline]

Lin J, Sahakian DC, de-Morais SM, Xu JJ, Polzer RJ, and Winter SM (2003) The role of absorption, distribution, metabolism, excretion and toxicity in drug discovery. Curr Top Med Chem 3: 1125-1154.[CrossRef][Medline]

Pang KS, Baker F, Simard A, Schwab AJ, and Goresjy CA (1995) Sulfation of acetaminophen by the perfused rat liver: the effect of red blood cell carriage. Hepatology 22: 267-282.[CrossRef][Medline]

Prueksaritanont T, Fernandez-Metzler C. Meng Y, Barrish A, Halczenko W, Rodan S, Hutchinson J, Duggan M, and Lin JH (2004) Disposition of a novel and potent _vß₃ antagonist in animals and extrapolation to man. Xenobiotica 34: 103-115.[CrossRef][Medline]

Roberts SA (2003) Drug metabolism and pharmacokinetics in drug discovery. Curr Opin Drug Discov Dev 6: 66-80.[Medline]

Rodan SB, Duong LT, Duggan ME, Fisher JE, Gentile MA, Fernandez-Metzler C, Halczenko W, Hartman G, Hutchinson JH, Leu C-T, et al. (2002) A potent antagonist of _vß₃ integrin is a powerful inhibitor of bone resorption in vivo. Bone Miner Res 17: S198.

Rose J and Castagnoli N Jr (1983) The metabolism of tertiary amines. Med Res Rev 3: 73-88.[CrossRef][Medline]

Rushmore TH, Reider PJ, Slaughter D, Assang C, and Shou M (2000) Bioreactor systems in drug metabolism: synthesis of cytochrome P450-generated metabolites. Metab Eng 2: 115-125.[CrossRef][Medline]

Shaffer CL, Harriman S, Koen YM, and Hanzlik RP (2002) Formation of cyclopropane during cytochrome P450-catalyzed N-dealkylation of a cyclopropylamine. J Am Chem Soc 124: 8268-8274.[CrossRef][Medline]

Skordos KW, Skiles GL, Laycock JD, Lanza DL, and Yost GS (1998) Evidence supporting the formation of 2,3-epoxy-3-methylindoline: a reactive intermediate of the pnueumotoxin 3-methylindole. Chem Res Toxicol 11: 741-749.[CrossRef][Medline]

Smallcombe SH, Patt SL, and Keifer PA (1995) WET solvent suppression and its applications to LC-NMR and high resolution NMR spectroscopy. J Magn Reson A 117: 295-303.[CrossRef]

Spahn-Langguth H and Benet LZ (1992) Acyl glucuronides revisited: is the glucuronidation process a toxification as well as a detoxification mechanism? Drug Metab Rev 24: 5-47.[Medline]

Streeper RT, Pearson PG, Zhao Z, Mizsak SA, Sanders PE, Wienkers LC, and Vrbanac JJ (1997) In vitro metabolic transformation of 2,4-dipyrrolidinylpyrimidine: a chemical probe for P450-mediated oxidation of tirilazad mesylate. Xenobiotica 27: 1131-1145.[Medline]

Summers MF, Marzilli LG, and Bax A (1986) Complete ¹H and ¹³C assignments of coenzyme B₁₂ through the use of new two-dimensional NMR experiments. J Am Chem Soc 108: 4258-4294.

Tanaka M, Ishikawa F, and Hakusui H (1995) Isolation and identification of seven metabolites of a water-soluble platelet aggregation inhibitor in rat urine. Xenobiotica 25: 1247-1257.[Medline]

Upthagrove AL and Nelson WL (2001) Carbinolamines, imines and oxazolidines from fluorinated propranolol analogs. ¹⁹F NMR and mass spectral characterization and evidence for formation as intermediates in cytochrome P450-catalyzed N-dealkylation. Drug Metab Dispos 20: 1114-1122.

Vickers S and Polsky SL (2000) The biotransformation of nitrogen containing xenobiotics to lactams. Curr Drug Metab 1: 357-389.[CrossRef][Medline]

Walker DK, Ackland MJ, James GC, Muirhead GJ, Rance DJ, Wastfall P, and Wright PA (1999) Pharmacokinetics and metabolism of sildenafil in mouse, rat, rabbit, dog and man. Xenobiotica 29: 297-310.[CrossRef][Medline]

Walles M, Thum T, Levsen K, and Borlak (2002) Verapamil: new insight into the molecular mechanism of drug oxidation in the human heart. J Chromatogr A 970: 117-130.[CrossRef][Medline]

Wilder RL (2002) Integrin alpha V beta 3 as a target for treatment of rheumatoid arthritis and related rheumatic diseases. Ann Rheum Dis 61 (Suppl 2): ii96-99.[Abstract/Free Full Text]

This Article Abstract Full Text (PDF) Submit a response Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Cui, D. Articles by Prueksaritanont, T. Search for Related Content PubMed PubMed Citation Articles by Cui, D. Articles by Prueksaritanont, T.