Bioactivation of Benzylamine to Reactive ... - ACS Publications

1190

Chem. Res. Toxicol. 2002, 15, 1190-1207

Bioactivation of Benzylamine to Reactive Intermediates in Rodents: Formation of Glutathione, Glutamate, and Peptide Conjugates Abdul E. Mutlib,*,† Patricia Dickenson,‡ Shiang-Yuan Chen,‡ Robert J. Espina,‡ J. Scott Daniels,† and Liang-Shang Gan§ Drug Metabolism and Pharmacokinetics Section, Bristol-Myers Squibb Pharma Company, Wilmington, Delaware 19880 Received July 8, 2002

The in vivo and in vitro disposition of benzylamine was investigated in rats. Benzylamine was metabolized to only a small extent by rat liver subcellular fractions. In contrast, it was extensively metabolized in vivo in rats. In vivo studies performed with stable isotope-labeled benzylamine enabled rapid mass spectrometric identification of metabolites present in rat bile and urine. The major metabolite of benzylamine was the hippuric acid formed by glycine conjugation of benzoic acid. LC/MS analysis of bile and urine obtained from rats dosed with 1:1 equimolar mixture of either d0:d7- or d0:d2-benzylamine showed the presence of several glutathione adducts in addition to the hippuric acid metabolite. The presence of various glutathione adducts indicated that benzylamine was metabolized to a number of reactive intermediates. Various metabolic pathways, including those independent of P450, were found to produce these intermediates. A previously undocumented pathway included the formation of a new carbon-nitrogen bond that led to a potentially reactive intermediate, Ar-CH2-NH(CO)-X, capable of interacting with various nucleophiles. The origin of this reactive intermediate is postulated to occur via the formation of either a formamide or carbamic acid metabolites. Metabolites which were produced by the reaction of this intermediate, Ar-CH2-NH(CO)-X with nucleophiles included S-[benzylcarbamoyl] glutathione, N-acetyl-S-[benzylcarbamoyl]cysteine, S-[benzylcarbamoyl] cysteinylglycine, S-[benzylcarbamoyl] cysteinylglutamate, N-[benzylcarbamoyl] glutamate, and an oxidized glutathione adduct. Bioactivation of amines via this pathway has not been previously described. The oxidative deamination of benzylamine yielding the benzaldehyde was demonstrated to be a precursor to the hippuric acid metabolite and S-benzyl-L-glutathione. The formation of the S-benzyl-L-glutathione conjugate showed that a net displacement of amine from benzylamine had taken place with a subsequent addition of glutathione at the benzylic position. In addition to these novel pathways, a number of other glutathione-derived adducts formed as a result of epoxide formation was characterized. It was demonstrated that benzylamine was converted by rat P450 2A1 and 2E1 to benzamide that was rapidly metabolized to an epoxide. Mechanisms are proposed for the formation of various GSH adducts of benzylamine.

Introduction The ability to characterize unusual and minor metabolites has been greatly accelerated with the introduction of versatile analytical techniques such as liquid chromatography/mass spectrometry (LC/MS) and liquid chromatography/nuclear magnetic resonance (LC/NMR) spectroscopy. Characterization of such minor metabolites has demonstrated the existence of previously undocumented metabolic pathways, such as the formation of glutamic acid conjugates of acetaminophen glutathione-derived adducts (1). Recently we reported the characterization of unique glutamate conjugates of 1-[3-(aminomethyl) phenyl]-N-[3-fluoro-2′-(methylsulfonyl)-[1,1′-biphenyl]-4* To whom correspondence should be addressed. Telephone: (734) 622-2198. Fax: (734) 622-1459. E-mail: [email protected]. † Pharmacokinetics, Dynamics and Metabolism, Pfizer Inc., 2800 Plymouth Road, Ann Arbor, MI 48105. ‡ Bristol-Myers Squibb Pharma Company. § Department of Drug Metabolism and Pharmacokinetics, Millenium Pharmaceuticals Inc., Cambridge, MA 02139.

yl]-3-(trifluoromethyl)-1H-pyrazole-5-carboxamide (DPC 423) and its analogues, using various spectroscopic techniques (2, 3). The metabolism studies with DPC 423 and its structural analogues revealed that the benzylamine moiety was a metabolic soft spot that was being metabolized extensively to several products, including a number of novel metabolites. The benzylamine moiety was converted by phase I and phase II enzymes producing an aldehyde, a carboxylic acid, acyl glucuronides, a carbamyl glucuronide, oxime, and a glutathione adduct (2-5). The presence of these metabolites implicated bioactivation of these compounds. The interesting disposition pathways of DPC 423 (see Figure 1) prompted us to continue with a more detailed investigation into the rodent metabolism of benzylamine. Benzylamines differ from the more widely studied arylamines by virtue of a having a methylene bridge between the aromatic system and the amino moiety. Several reports have appeared in the literature describing the metabolism of arylamines. The reactivities as-

10.1021/tx020063q CCC: $22.00 © 2002 American Chemical Society Published on Web 08/17/2002

Bioactivation of Benzylamine

Chem. Res. Toxicol., Vol. 15, No. 9, 2002 1191

derived. Importantly, data from our DPC 423 work implicated the involvement of hydroxylamine species as a reactive precursor to the formation of oxime-derived GSH adducts in rats. Therefore, the in vivo metabolism of benzylhydroxylamine in rodents was also investigated in order to determine any possible links of this species to the bioactivation of benzylamine.

Materials and Methods

Figure 1. Structures of DPC 423 and d0-, d2-, d7-benzylamine.

sociated with the formation of hydroxylamines, nitroso, and other metabolic intermediates from arylamines have been very well documented (6-9). In contrast, the metabolism of benzylamines has largely been overlooked. A more comprehensive investigation into the metabolic disposition of benzylamine is therefore required. The literature reveals that several metabolism studies have been performed with benzylamine itself. However, most of these studies were limited to the description of oxidative deamination of benzylamine, mediated by various amine oxidases (10-15). The oxidative deamination of benzylamine by the amine oxidases affords benzaldehyde. The biotransformation of benzylamine to benzaldehyde has been attributed to mitochondrial monoamine oxidase (MAO)-B (10-12) and to semicarbazide-sensitive amine oxidases (SSAO) present in the plasma of various animal species (13-15). Benzylamine has been shown to be extensively deaminated to benzaldehyde by liver subcellular fractions from various species (11). However, due to the unavailability of LC/MS and LC/NMR techniques at that time, the structure elucidation of polar conjugates was not possible. We therefore re-visited the metabolism of benzylamine with an anticipation that it might be metabolized to similar metabolites as DPC 423 (2-5). One objective of our study was to thoroughly investigate the metabolic conversion of benzylamine in rodents. We therefore employed LC/MS and NMR to evaluate if reactive intermediates were formed during its biotransformation. Deuterium-labeled benzylamine was employed to improve the detection of metabolites in complex biological matrixes as well as to aid in structure elucidation. During the course of this study, synthetic standards were prepared and the LC/MS retention times and MS/MS1 spectra were obtained for comparison with the metabolites. We also attempted to isolate polar metabolites from biological samples for which no synthetic standards were available. NMR studies were performed with these samples to unambiguously assign the structures. The GSH adducts, in particular, were fully characterized in order to provide valuable insight into the possible identity of the reactive intermediates from which they were 1 Abbreviations: LC-ESI/MS, liquid chromatography electrospray ionization mass spectrometry; LC-APCI/MS, liquid chromatography atmospheric pressure chemical ionization mass spectrometry; 1H NMR, proton nuclear magnetic resonance; MS/MS, mass spectrometry/mass spectrometry; TOCSY, total correlated spectroscopy; HMBC, heteronuclear multiple bond correlation; HSQC, heteronuclear single quantum coherence.

Chemicals and Supplies. Benzylamine, benzaldehyde, synbenzaldehyde oxime, benzamide, benzohydroxamic acid, benzoic acid, benzyl alcohol, N-benzohydroxamic acid, N-benzylhydroxylamine, benzyl isocyanate, N-acetylcysteine, cysteinylglycine, cysteinylglutamate, hippuric acid, L-glutamic acid, GSH (reduced and oxidized), and various chemical reagents were purchased from Sigma-Aldrich Chemical Co. (St. Louis, MO). Benzyl-d7-amine hydrochloride was obtained from CDN Isotopes (Quebec, Canada). S-Benzyl-L-cysteine was purchased from Fluka Chemicals (Bassle, Switzerland). Bond Elut C18 cartridges (10 g/60 cm3) were obtained from Varian Sample Preparation Products (Harbor City, CA). All general solvents and reagents were of the highest grade available commercially. Synthesis of Metabolite Standards. Synthetic standards of potential metabolites of benzylamine were either obtained commercially or synthesized in-house. The purity of each standard was confirmed by LC/MS and NMR analysis to be greater than 95% as determined by the integration of peak signals. All of the standards showed the presence of a single peak [detected both by UV and total ion current (TIC)] during the LC/MS analysis. Synthesis of Benzyl-d2-amine Hydrochloride. To a suspension of LiAlD4 (4.2 g, 52 mmol) in THF (100 mL) was added a solution of benzonitrile (5.2 g, 50 mmol) in THF (20 mL) at 50 °C. The mixture was stirred at 50 °C for 4 h. After cooling in an ice bath, 6 mL of 1 N NaOH was added to the mixture slowly. The mixture was filtered, and the filtrate evaporated to dryness. The residue was dissolved in CH2Cl2 and purified using flash column chromatography (silica gel), eluting with a solution of CH2Cl2/methanol/NH4OH (80:15:5 v/v). The fraction corresponding to benzylamine was collected and evaporated to dryness. The residue was dissolved in 1 N HCl (20 mL), and the solution was lyophilized to yield the desired product d2-benzylamine hydrochloride (2 g, 18 mmol). 1H NMR in DMSO-d6 showed aromatic proton signals at δ7.2-7.5 (5H, m). LC-ESI/MS showed MH+ at m/z 110. Synthesis of S-Benzyl-L-glutathione. To a round-bottom flask containing 5 mL of methanol:water (1:1 v/v) were added reduced GSH (1.6 mmol) and benzyl bromide (1.5 mmol). Triethylamine (1.5 mmol) was added, and the reaction mixture stirred for 1 h at room temperature, after which the solvent was removed under a stream of nitrogen. The dried powder was dissolved in a methanol:water mixture (1:9 v/v) and chromatographed on a Bond Elut C18 cartridge that was previously preconditioned with methanol and water. After the samples eluted under gravity, the cartridge was washed with 30 mL of water followed by elution with 30 mL aliquots of different percentages of methanol in water. It was found that most of the GSH conjugate had eluted with 10-15% methanol in water. The solvents were removed, and the dried powder was subsequently analyzed by LC/MS and NMR. 1H NMR: δ 8.42 (1H, bs, cys NH), 8.10 (1H, bs, gly NH), 7.30 (4H, m, Ar-H), 7.20 (1H, m, Ar-H), 4.48 (1H, m, cys R), 3.75 (1H, s, Ar-CH2-S), 3.60 (2H, m, gly R), 3.35 (1H, m, glu R), 2.83 (1H, m, cys β), 2.62 (1H, m, cys β′), 2.35 (1H, m, glu γ), and 1.95 (1H, m, glu β′). LC-ESI/MS showed MH+ at m/z 398. Synthesis of S-[Benzylcarbamoyl]glutathione. GSH (1.7 mmol) was stirred in 5 mL of methanol in a 10 mL round-bottom flask. Benzyl isocyanate (1.5 mmol) was added to the flask and the reaction mixture stirred. To the reaction mixture was added 10 µL of concentrated sodium hydroxide (6 M) and the solution stirred for 30 min at room temperature. The solvents were

1192

Chem. Res. Toxicol., Vol. 15, No. 9, 2002

removed under vacuum and the powder purified on a C18 cartridge as described above. The fraction containing the GSH adduct was dried and re-purified on a semipreparative column (Supelco, C18, 10 × 250 mm). The column was eluted with an isocratic mobile phase consisting of acetonitrile:0.05% TFA (1:5 v/v) delivered at 3.5 mL/min. The GSH conjugate eluting at tR ) 4.0 min was collected, dried and subsequently analyzed by LC/MS and NMR. 1H NMR: δ 7.31 (2H, m, Ar-H), 7.23 (3H, m, Ar-H), 4.30 (2H, m, Ar-CH2-NH-), 4.25 (1H, m, cys R), 3.50 (1H, m, glu R), 3.40 (2H, m, gly R), 3.35 (1H, m, cys β), 2.95 (1H, m, cys β′), 2.38 (1H, m, glu γ), 2.30 (1H, m, glu γ′), 2.00 (1H, m, glu β), and 1.90 (1H, m, glu β′). LC-ESI/MS showed MH+ at m/z 441. Synthesis of N-Acetyl-S-[benzylcarbamoyl]cysteine. NAcetylcysteine (1.7 mmol) was dissolved in 5 mL of methanol in a 10 mL round-bottom flask. Benzyl isocyanate (1.5 mmol) was added to the flask, and the reaction performed as described above. At the end of the reaction, the solvents were removed under vacuum, and the residue was purified on a C18 cartridge as described above. The N-acetylcysteine conjugate eluted with 40% methanol in water. The solvents were removed under vacuum, and the dried sample was analyzed by LC/MS and NMR without any further purification. 1H NMR: δ 8.70 (1H, m, Ar-CH2-NH-), 8.20 (1H, d, NH-CO(CH3), 7.32 (2H, m, ArH), 7.25 (3H, m, Ar-H), 4.32 (2H, m, Ar-CH2-NH-), 4.30 (1H, m, cys R), 3.35 (1H, m, cys β), 3.02 (1H, m, cys β′), and 1.85 (3H, s, CO(CH3). LC-ESI/MS showed MH+ at m/z 297. Synthesis of S-[Benzylcarbamoyl]cysteinylglycine. Cysteinylglycine (0.1 mmol) was dissolved in 5 mL of methanol in a 10 mL round-bottom flask. Benzyl isocyanate (0.1 mmol) was added to the flask, and the reaction performed as described above. At the end of the reaction, the solvents were removed under vacuum and the residue was purified on a C18 cartridge as described above. No further purification was required for this standard. 1H NMR showed δ at 8.75 (1H, m, Ar-CH2-NH-), 7.82 (1H, m, CO-NH-CH2-), 7.31 (2H, m, Ar-H), 7.23 (3H, m, Ar-H), 4.30 (2H, Ar-CH2-NH-), 4.30 (1H, m, cys R), 3.45 (2H, d, CHNH2) 3.30 (1H, m, cys β), 3.18 (2H, m, gly R), 2.90 (1H, m, cys β′). LC-ESI/MS showed MH+ at m/z 312. Synthesis of S-[Benzylcarbamoyl]cysteinylglutamate. Cys-Glu (0.8 µmol) was added to 1.1 M equiv of benzyl isocyanate in methanol, and the reaction performed at room temperature as described above. At the end of the reaction, the solvents were removed and the dried extract was purified on a semipreparative column (Supelco, 10 × 250 mm). The column was eluted with an isocratic mobile phase consisting of a mixture of acetonitrile and 0.05% TFA (3:7 v/v) delivered at 3.5 mL/min. The conjugate eluted at tR ) 5.0 min. The peak corresponding to this compound was collected from several injections, pooled, dried under vacuum, and analyzed by LC/MS and NMR. 1H NMR: δ 8.70 (1H, m, Ar-CH2-NH-), 8.35 (1H, d, CH-NH-CO-) 7.32 (2H, m, Ar-H), 7.23 (3H, m, Ar-H), 4.30 (2H, m, Ar-CH2NH-), 4.25 (1H, m, cys R), 3.60 (1H, m, glu R), 3.35 (1H, m, cys β), 3.00 (1H, m, cys β′), 2.30 (2H, m, glu γ), 2.00 (1H, m, glu β), and 1.90 (1H, m, glu β′). LC-ESI/MS showed MH+ at m/z 384. Synthesis of N-[Benzylcarbamoyl]glutamate. L-Glutamate (0.3 mmol) was reacted with benzyl isocyanate (0.3 mmol) in methanol as described above. At the end of the reaction, the solvents were removed under vacuum and the residue subsequently purified on a semipreparative column (Supelco, 10 × 250 mm). The column was eluted with an isocratic mobile phase consisting of a mixture of acetonitrile and 0.05% TFA (3:7 v/v) delivered at 3.5 mL/min. The glutamate conjugate eluted at tR ) 6.0 min. The peak corresponding to this conjugate was collected from several injections, pooled, dried under vacuum, and analyzed by LC/MS and NMR. 1H NMR: δ 7.28 (2H, m, Ar-H), 7.24 (3H, m, Ar-H), 4.20 (2H, s, Ar-CH2-NH-), 4.12 (1H, m, glu R), 2.30 (1H, m, glu γ), 2.25 (1H, m, glu γ′), 1.95 (1H, m, glu β), and 1.72 (1H, m, glu β′). LC-ESI/MS showed MH+ at m/z 281. Synthesis of Metabolite M11 (adduct of oxidized glutathione). Oxidized glutathione (GSSG, 0.2 mmol) was added

Mutlib et al. to a methanolic solution of benzyl isocyanate (0.2 mmol) and the reaction stirred for 30 min at room temperature. At the end of the reaction, the solvent was removed under a stream of nitrogen and the dried extract analyzed by LC/MS without any further purification. LC-ESI/MS showed MH+ at m/z 746. Due to the low yield of M11, further characterization by NMR was not performed. The structure of metabolite M11 was assigned tentatively. Liquid Chromatography/Mass Spectrometry. The metabolites were separated on a HPLC column (Phenomenex, Aqua C18, 4.6 × 150 mm, 5 µm) by a gradient solvent system consisting of two components, A (100% methanol with 0.1% formic acid) and B (ammonium formate, 10 mM, pH 3.6). The percentage of A was increased from 0 to 35 over 15 min with the solvent flow rate set at 1.0 mL/min. After maintaining the organic solvent at 35% for an additional 3 min, the proportion of A was ramped linearly to 75% over the next 7 min. After 25 min, the column was washed with 90% A for 5 min before reequilibrating with the initial mobile phase. To avoid contaminating the source of the mass spectrometer, the solvent was diverted to waste at 25 min. Aliquots of bile (20 µL) and urine samples (80 µL) were injected directly onto the HPLC column, and the eluent was introduced into the source of the mass spectrometer. Subsequent studies also utilized a smaller diameter column (Phenomenex, Aqua C18, 2.1 × 100 mm, 3 µm) using the same mobile phase eluted at a flow rate of 0.3 mL/ min. To detect the metabolites in the fractions from C18 cartridges and from semipreparative HPLC column, aliquots (20-50 µL) were introduced to the mass spectrometer using the flow injection analysis method. The mobile phase consisted of a mixture of A and B (1:1 v/v) delivered at a rate of 0.35 mL/min. LC/MS was carried out by coupling a Hewlett-Packard HPLC system (HP1100) to a Finnigan LCQ ion trap mass spectrometer. Liquid chromatography atmospheric pressure chemical ionization-mass spectrometry (LC-APCI/MS) or liquid chromatography electrospray ionization mass spectrometry (LC-ESI/ MS) was performed in the positive ion mode. The metabolites were detected by operating the mass spectrometer either in the full scan mode or by selected ion monitoring of the pseudomolecular ions of the conjugate. MS/MS spectra of fragment ions on the LCQ mass spectrometer were obtained with 25-30% relative collision energy. Selected ion monitoring (SIM) of metabolites present in bile and urine was performed, and the peak area response for each metabolite obtained. Known amounts of synthetic standards of metabolites were also injected onto the LC/MS, and the peak area response was obtained for each compound. An estimate of the levels of each metabolite was obtained by comparing the peak areas determined from SIM experiments. The volumes of urine and bile were taken into consideration when obtaining an estimate of the total amount of metabolite excreted. The ions monitored for M1, M2, M3, M5, M6, M8, M9, M10, M12, and M13 were at m/z 180, 445, 316, 441, 312, 297, 281, 384, 398, and 212, respectively. High-Field NMR. Spectra were obtained on a Bruker Avance 600 MHz NMR spectrometer equipped with 2.5 mm 1H/ 13C inverse conventional NMR probe. Suppression of the residual water and acetonitrile signals was carried out using the WET solvent suppression method in all the NMR experiments. Chemical shifts were referenced to DMSO at δ 2.49 ppm and to acetonitrile at δ 2.0 ppm. The structures of metabolites were determined from proton and carbon one-dimensional NMR as well as proton-proton total correlated spectroscopy (TOCSY), proton-carbon heteronuclear single-quantum coherence (HSQC), and long-range proton-carbon heteronuclear multiple-bond correlation (HMBC) two-dimensional NMR. In Vivo Studies. Male Sprague-Dawley rats with cannulated bile ducts (weighing between 250 and 350 g) were administered oral doses of benzylamine hydrochloride, Nbenzylhydroxylamine, benzamide or benzaldehyde at 300 mg/ kg, and urine and bile collected over ice. Benzylamine hydrochloride and N-benzylhydroxylamine were dissolved in normal

Bioactivation of Benzylamine saline while benazamide and benzaldehyde were prepared as a suspension in 0.5% methocel. These were administered orally at 10 mL/kg. In another study, groups of three rats were administered 1:1 equimolar mixtures of either d0:d7 or d0:d2 benzylamine hydrochloride (in normal saline) at 300 mg/kg and bile/urine collected. The rats were housed individually in suspended stainless steel wire-mesh cages equipped with an automatic watering system. The study room was environmentally controlled for temperature (72 ( 4 °F), relative humidity (40-70%), and light (a 12 h light/dark cycle). Rats had free access to water and were given a specific amount of certified Purina rodent chow each day. The samples were collected at 0-7 and 7-24 h time intervals and stored at -20 °C until analyzed. Study with Aminobenzotriazole (ABT). Three bile ductcannulated male Sprague-Dawley rats were orally administered ABT (100 mg/kg) for 3 days. On the second and third days, the rats were administered benzylamine hydrochloride (150 mg/ kg) 1 h after the ABT dose. The bile and urine samples were collected over ice and stored as described above. Microsomal, S9, and FMO Metabolism of Benzylamine. Benzylamine hydrochloride or a 1:1 w/w mixture of d0:d7 benzylamine hydrochloride or N-benzylhydroxylamine (all dissolved in normal saline) were incubated with rat liver microsomes (phenobarbital or dexamethasone induced or nonpretreated, 1 mg/mL) or S9 fraction (4 mg/mL) using the following protocol: NADPH (2 mM), substrate (10 or 100 µM), +/- GSH (3 mM), MgCl2 (3 mM), and 0.1 M phosphate buffer (pH 7.4) to a final volume of 1 mL. The mixtures were incubated for 1 h after which 2 mL of cold acetonitrile was added and the proteins precipitated. After centrifuging the samples at 3500g for 5 min, the supernatants were transferred to clean culture tubes and dried under a stream of nitrogen at 25 °C. The dried extracts were reconstituted in the HPLC mobile phase and analyzed by LC/MS as described earlier. Studies were also conducted with human FM01, FM03, and FM05 using the following protocol: NADPH (2 mM), substrate (10 or 100 µM), +/- GSH (3 mM), 0.5 mg of cDNA expressed human FMO1, FMO3 and FMO5, diethylenetriaminepentaacetic acid (1.2 mM), and 0.05 M phosphate buffer (pH 8.4) to a final volume of 1 mL. The incubations, extractions of samples, and LC/MS analysis of samples were done as described above. In Vitro Metabolism by cDNA Expressed Rat P450 Enzymes. To a buffered solution (either sodium phosphate or TRIS, 100 mM, pH 7.4) of d0:d7-, d0:d2-benzylamine or benzamide (20 µM), MgCl2 (3 mM), and rat P450 enzyme (50 pmol) was added NADPH (2 mM), and the solution was incubated at 37 °C for 45 min. The incubations were performed in duplicate with one set of samples fortified with GSH. The reactions were terminated by the addition of 2 mL of acetonitrile. The supernatant was dried and analyzed by LC/MS. The following is a list of rat cDNA-expressed enzymes employed in the study: 1A1, 1A2, 2A1, 2A2, 2B1, 2C6, 2C11, 2C12, 2D1, 2D2, 2E1, 3A1, and 3A2. Plasma Stability of Benzylamine in Rats. The in vitro plasma stability of benzylamine was determined by incubating the compound (50 µM) in rat plasma (1 mL) at 37 °C for 4 h. After precipitation of proteins with acetonitrile (4 mL), the supernatants were dried under nitrogen and the residue reconstituted in the HPLC mobile phase (200 µL) and analyzed for benzylamine metabolite(s). Isolation of Metabolites M2 and M3 from Rat Bile. Bile duct-cannulated male Sprague-Dawley rats were dosed with benzylamine and bile collected as described above. The bile samples were diluted 1:1 with distilled water and loaded onto C18 cartridges preconditioned with methanol and water. The samples were allowed to elute under gravity. The cartridges were subsequently washed with 30 mL of water followed by elution with 30 mL aliquots of different percentages of methanol in water. Each of the fractions was checked by LC/MS for the presence of metabolites. The fractions containing the metabolites were pooled and dried, and the residues reconstituted in (5:95

Chem. Res. Toxicol., Vol. 15, No. 9, 2002 1193 v/v) methanol:10 mM ammonium formate, pH 3.7, before further purification on a semipreparative HPLC column (Beckman C18, 10 × 250 mm). The mobile phase consisted of an isocratic solvent mixture of A (5% methanol in 10 mM ammonium formate, pH 3.7) and B (0.1% formic acid in methanol). Two peaks at approximate retention times of 5.4 and 4.8 min were collected, corresponding to metabolites M2 and M3, respectively. A Waters Fraction Collector II was used to collect the fractions from the HPLC. The fractions corresponding to each metabolite were dried under vacuum and subsequently analyzed by NMR spectroscopy.

Results In Vitro Studies. Rat liver S9 and microsomal fractions did not produce detectable levels of any of the benzylamine metabolites that were observed in vivo (Scheme 1). The human cDNA expressed FMO1, FMO3, or FMO5 did not metabolize benzylamine. The lack of in vitro benzylamine metabolism suggested a minor role for P450 and FMO in the metabolic disposition of benzylamine. It appears that the MAO activities responsible for the formation of benzaldehyde from benzylamine were either absent or present at low levels in the S9 fractions. Experiments performed with rat cDNA-expressed P450 enzymes, however, showed conversion of benzylamine by P450 2A1 and 2E1 to 3,4-dihydroxybenzamide (Scheme 2). The tentative structure of this metabolite was established by mass spectral analysis of the incubations employing deuterium labeled benzylamine (Figure 1). The metabolite showed MH+ at m/z 154, 154, and 157 from incubations performed with d0-, d2- and d7-benzylamine, respectively. This amounted to a net addition of 46 amu to the nonlabeled benzylamine. An addition of three oxygen atoms with the possibility of further oxidation of an alcohol functional group to the corresponding ketone appeared likely. The LC/MS analysis of extract in which CYP2A1 was incubated with d0:d7- or d0:d2benzylamine showed that four and two deuteriums were lost from the labeled compounds, respectively. The total loss of two deuteriums from d2-benzylamine in the d0: d2-benzylamine incubation implicates an oxidation had occurred at the benzylic position with the formation of benzamide. This would account for an addition of 14 amu to the mass of benzylamine. Further oxidation on the aromatic ring with an insertion of two oxygen atoms (resulting in a loss of two deuteriums from the aromatic ring) would account for the other 32 amu; hence, a net addition of 46 amu was observed. In vitro metabolism studies with benzamide (nonlabeled) incubated with either CYP 2E1 or 2A1 showed the formation of the same metabolite (MH+ at m/z 154), as was observed with benzylamine, providing evidence in support of benzamide as the likely precursor. Benzylamine was fairly stable in rat plasma. This is in contrast to the results we had obtained earlier with DPC 423 (Figure 1), a compound possessing a benzylamine moiety, which was susceptible to degradation by semicarbazide sensitive amine oxidase (SSAO) present in rat plasma (4). In the present study we observed that benzylamine was slowly metabolized to the glucoside conjugate (MH+ at m/z 270 and 277 for d0- and d7benzylamine, respectively) in the presence of rat plasma. Most of benzylamine had remained unchanged over the course of 4 h with the glucoside conjugates accounting for less than 5% conversion of the substrate. A similar conversion was observed from incubations with plasma from both dog and human.

1194


Mutlib et al.

Scheme 1. Proposed Metabolic Pathways of Benzylamine in Rats

Scheme 2. P450 2A1 and 2E1 Conversion of Benzylamine to Benzamide and 3,4-Dihydroxybenzamidea

a

The lack of specific GST in the cDNA expressed system leads to the catechol metabolite rather than to the GSH adduct, M2.

In Vivo Studies. Benzylamine was extensively metabolized by rats to a number of metabolites including hippuric acid and several GSH adducts. Little or no glucuronide or sulfate conjugates were detected. The

phase I metabolites such as the aldehyde and carboxylic acid (Scheme 1) were not detected in the bile or urine samples. The hippurate (M1) was the major metabolite detected in the bile and urine samples of rats dosed with



Figure 2. (A) LC/MS/MS total ion current (TIC); (B) UV trace; and (C) MS/MS spectrum of the protonated parent ion MH+ at m/z 180 of hippuric acid metabolite (glycine conjugate) present in urine of a rat dosed with 300 mg/kg of benzylamine hydrochloride.

benzylamine hydrochloride. The LC/MS total ion current (TIC) and LC/UV traces of urine samples (Figure 2) indicated that greater than 95% of the drug-related material was excreted as the hippuric acid. Likewise, the LC/MS and LC/UV traces of bile samples suggested that hippuric acid was the major component. The GSH adducts accounted for less than 20% of total dose excreted in bile as determined by selected ion monitoring studies. However, in the absence of a radiolabeled compound, it was difficult to obtain the absolute levels of each metabolite in the biological samples, most notably in the presence of endogenous components. Characterization of Metabolites in Benzylamine Dosed Rats. Metabolites M1-M4 showed a characteristic retention of five deuteriums as indicated by the presence of pairs of MH+ ions, separated by 5 amu, in the mass spectra of bile samples obtained from rats dosed with 1:1 w/w mixture of d0:d7-benzylamine. The mass spectra of metabolites from rats dosed with d0:d2-benzylamine showed a total loss of both deuteriums, and hence, MH+ of these compounds appeared as a single peak. Metabolite M1: This was the major metabolite found in rat bile and urine. Metabolite M1 produced MH+ at m/z 180 with the ammonium adduct [M + NH4]+ at m/z 197 in the mass spectrum. The structure of M1 was confirmed by comparing the LC/MS retention time and mass spectral data with those produced by hippuric acid. The formation of M1 as a metabolite of benzylamine was also confirmed by studies done with labeled compounds. The LC/MS spectrum of bile and urine from rats dosed with 1:1 mixture of d0:d7-benzylamine hydrochloride showed MH+ at m/z 180 and 185, respectively (Figure 2). The MS/MS spectrum of MH+ ion showed ions at m/z 105 and 162 corresponding to characteristic losses of glycine (-75 amu) and water, respectively. Metabolite M2: Metabolite M2 showed MH+ at m/z 445 and was found in rat bile. The mass spectrum showed a loss of water from the protonated parent ion (m/z 427)

as the base peak. The MS/MS spectrum of this metabolite showed ions at m/z 427, 370 (-glycine), 316 (-pyroglutamate), 306 (GSH), 298, 288, 281, and 177. Selected ion monitoring (SIM) studies suggested that this metabolite accounted for less than 2% of the administered dose. The structure of M2 was confirmed by isolating and characterizing by NMR experiments. The 1H NMR showed signals at δ 7.85 (1H, bs, gly NH), 7.70 (1H, bs, CONH2), 7.10 (1H, bs, CONH2), 6.80 (1H, d), 6.18 (1H, m), 6.05 (1H, m), 4.20 (1H, d, CH(OH)), 4.00 (1H, d, CHSCH2), 3.50 (1H, glu R, m), 3.35 (1H, m, cys R), 3.30 (2H, m, gly R), 2.72 (1H, m, cys β), and 2.60 (1H, m, cys β′), 2.35 (2H, m, glu γ), 2.00 (1H, m, glu β), and 1.90 (1H, m, glu β′). The 1H NMR of this metabolite was identical to that of M3 except for the presence of glutamate proton signals in the spectrum of GSH conjugate (M2). Metabolite M3: The cysteinylglycine conjugate derived from M2 produced MH+ at m/z 316. The SIM studies suggested that this metabolite accounted for less than 5% of the administered dose. This metabolite was isolated from rat bile, and its structure determined by NMR. The MS/MS spectrum of the protonated parent ion produced ions at m/z 298, 281, 177 (cys-gly), and 152. The 1H NMR (see Figure 3) showed signals at δ 7.85 (1H, bs, gly NH), 7.70 (1H, bs, CONH2), 7.10 (1H, bs, CONH2), 6.80 (1H, d), 6.18 (1H, m), 6.05 (1H, m), 4.20 (1H, d, CH(OH)), 4.00 (1H, d, CHSCH2), 3.35 (1H, m, cys R), 3.30 (2H, m, gly R), 2.72 (1H, m, cys β), and 2.60 (1H, m, cys β′). The structure of M3, shown in Figure 3 and Scheme 1, was determined by TOCSY and HMBC experiments. Metabolites M4: The cysteine conjugate formed from the catabolism of M2 produced MH+ at m/z 259. An estimate of M2 could not be obtained due to the unavailability of synthetic standard. This metabolite found in rat bile was not further characterized. Metabolites M5-M11 showed a characteristic retention of seven deuteriums as indicated by the presence of pairs of MH+ ions, separated by 7 amu, in the mass

1196


Mutlib et al.

Figure 3. 1H NMR of the cysteinylglycine conjugate (M3) isolated from bile of rats dosed with 300 mg/kg of benzylamine hydrochloride.

spectra of bile samples from rats dosed with 1:1 w/w mixture of d0:d7-benzylamine. Similarly these metabolites showed retention of both deuteriums as observed by the presence of pairs of MH+ ions, separated by 2 amu, in the spectra of samples from rats dosed with 1:1 w/w mixture of d0:d2-benzylamine. LC/MS/MS analyses of the metabolites suggested that the aminomethyl functional group was intact; however an increment of 28 amu to molecular mass of benzylamine, corresponding to carbon monoxide addition, suggested the possibility of an isocyanate intermediate. The existence of this additional CO group was confirmed by NMR studies with synthetic standards of metabolites produced by reacting benzyl isocyanate with the appropriate nucleophiles. Metabolite M5: This GSH conjugate was found in rat bile. The SIM studies suggested that this metabolite accounted for less than 7% of the administered dose. The protonated molecular ions for the nondeuterated and deuterated GSH adducts appeared at m/z 441 and 448, respectively. The MS/MS spectrum of m/z 441 showed ion fragments at m/z 366 (-glycine), 312 (-pyroglutamate), and 179. A similar mass spectrum of ions produced from loss of glycine and pyroglutamate was produced from MS/ MS of the deuterated analogue (MH+ at m/z 448). The LC/MS retention time and MS/MS fragmentation pattern of synthetic standard was obtained and found to be identical to those produced by the metabolite. The structure of the synthetic standard was confirmed by NMR experiments. It was important to demonstrate that during the chemical synthesis of this standard, isocyanate had reacted with the sulfhydryl of GSH forming a carbon-sulfur bond. The possibility of benzyl isocyanate reacting with the free amine of GSH (glutamate moiety)

leading to a carbon-nitrogen bond had to be ruled out. This was demonstrated by HMBC experiments whereby it was shown that the methylene of the cysteine moiety was coupled to the carbonyl group (data not shown). The methylene from the aminomethyl group of benzylamine was also coupled to the same carbonyl group. The structure of this metabolite was determined to be S[benzylcarbamoyl] glutathione (Scheme 1). Metabolite M6: The cysteinylglycine conjugates produced from M5 showed the protonated molecular ions at m/z 312 and 319, corresponding to nonlabeled and labeled metabolites, respectively. The SIM studies suggested that this metabolite accounted for less than 2% of the administered dose. The LC-MS/MS spectrum of this conjugate, found in rat bile, showed ions at m/z 179 (loss of Ar-CH2NHCO), 162, and 144. Further MS/MS studies with the fragment ion at m/z 179 showed the fragmentation pattern was identical to that produced by an authentic standard of cysteinylglycine. The LC/MS retention time and MS/MS fragmentation pattern of synthetic standard was obtained and found to be identical to those produced by the metabolite. The structure of M6, shown in Scheme 1, was confirmed as S-[benzylcarbamoyl]cysteinylglycine. Metabolite M7: The cysteine conjugates derived from M6 produced MH+ at m/z 255 and 262 for d0- and d7labeled compounds, respectively. An estimate of M7 levels in bile could not be obtained due to the unavailability of synthetic standard. No further characterization of this metabolite was done. Metabolite M8: This metabolite was detected in the urine of rats dosed with benzylamine. The SIM studies suggested that this metabolite accounted for less than 3% of the administered dose. The mass spectrum showed


Figure 4.

1H


NMR of the synthetic N-acetyl-S-[benzylcarbamoyl]cysteine, M8.

MH+ at m/z 297 with ion fragments at m/z 164 (Nacetylcysteine) and 122 (cysteine). The deuterium-labeled cysteine conjugate showed MH+ at m/z 304 as expected with the retention of all seven deuteriums. The LC/MS retention time and MS/MS fragmentation pattern of synthetic standard was obtained and found to be identical to those produced by the metabolite. The 1H NMR of the N-acetylcysteine conjugate is shown in Figure 4. The isocyanate linkage to the sulfhydryl of N-acetylcysteine was clearly demonstrated by HMBC experiments whereby it was shown that the methylene of the cysteine moiety was coupled to the carbonyl group (Figure 5). The methylene from the aminomethyl group of benzylamine was also found to be coupled to the same carbonyl group. The existence of this carbonyl group, added to the benzylamine during its metabolism was adequately demonstrated by these NMR experiments. The structure of M8 was confirmed as N-acetyl-S-[benzylcarbamoyl]cysteine. Metabolite M9: The glutamate conjugate of benzyl isocyanate was found in rat bile. The SIM studies suggested that this metabolite accounted for less than 1% of the administered dose. The metabolite produced MH+ at m/z 281 with the major fragment ion at m/z 148 (corresponding to the protonated glutamic acid). The urea linkage formed between the amine of the glutamic acid and the carbonyl group of the benzyl isocyanate was demonstrated by HMBC results (Figure 6). It was shown that the R carbon bearing the carboxylic acid of the

glutamate was coupled to the same carbonyl group as was the methylene of the benzylamine. The LC/MS retention time and MS/MS fragmentation pattern of synthetic standard was obtained and found to be identical to those produced by the metabolite. The structure of M9, shown in Scheme 1, was confirmed as N-[benzylcarbamoyl]glutamate. Metabolite M10: The cysteinylglutamate conjugate found in rat bile displayed a MH+ at m/z 384. The SIM studies suggested that this metabolite accounted for less than 2% of the administered dose. The MS/MS of pseudomolecular ion at m/z 384 produced major fragment ion at m/z 255 (-pyroglutamate). The LC/MS retention time and MS/MS fragmentation pattern of synthetic standard was obtained and found to be identical to those produced by the metabolite. The structure of M10, shown in Scheme 1, was confirmed as S-[benzylcarbamoyl]cysteinylglutamate. Metabolite M11: Rats dosed with benzylamine excreted a unique oxidized GSH conjugate of benzyl isocyanate. The SIM studies suggested that this metabolite accounted for less than 1% of the administered dose. The protonated molecular ions (MH+) were observed at m/z 746 and 753, for d0- and d7-labeled metabolites, respectively. The MS/MS spectra of these metabolites (MH+ at m/z 746 and 753) are shown in Figure 7. The ion fragments at m/z 639, 613, 595, 538, 510, 484, 466, 409, and 355 were found to be common in the mass spectra of both labeled and nonlabeled adduct. However, the MS/

1198


Mutlib et al.

Figure 5. HMBC spectrum of the N-acetylcysteine conjugate (M8) showing the correlation between the carbonyl group and the methylene groups (Ar-CH2-NH-CO-S-CH2-). The carbonyl group of the carboxylic acid of cysteine was found to be coupled to the cys R proton only.

Figure 6. HMBC spectrum of the glutamic acid conjugate (M9) showing the correlation between the carbonyl group and the methylene groups (Ar-CH2-NH-CO-NH-CH(COOH)-CH2-).

MS spectra also showed important differences between the ion fragments produced by these conjugates. Specific

ions at m/z 728, 671, 617, and 542 were only seen in the mass spectrum of the nonlabeled adduct (Figure 7).



Figure 7. LC/MS/MS of protonated parent ions of (A) nonlabeled (m/z 746) and (B) labeled (m/z 753) glutathione conjugate (M11) present in bile of rats dosed with d0:d7-benzylamine hydrochloride.

Similarly ions unique to the d7-labeled adduct were observed and included fragments at m/z 735, 678, 624, and 549 which were 7 amu higher than those produced by the nonlabeled adduct. Further MS/MS studies of the common fragments (found in spectra of both labeled and nonlabeled adducts) indicated losses of pyroglutamate (-129 amu) and glycine (-75 amu). The fragmentation patterns suggested an existence of a peptide capable of losing two glycine and one glutamate moiety. It was postulated that the adduct was formed by an interaction of reactive benzyl isocyanate (or its precursor) with oxidized GSH. The MS/MS studies showed a loss of only one glutamate, suggesting that the amine functional group of one of the glutamic acids in GSSG reacted with the benzyl isocyanate. A synthetic standard was prepared, and its LC/MS and MS/MS fragmentation pattern studied. The MS/MS of the MH+ (at m/z 746) of the synthetic standard matched that of the metabolite detected in the bile sample. However, due to the incomplete structural elucidation of the synthetic standard, the structure of M11 was tentatively assigned as the oxidized GSH conjugate of benzyl isocyanate. Metabolites M12-M13 showed a characteristic retention of six deuteriums as evidenced by the presence of pairs of MH+ ions, separated by 6 amu, in the mass spectra of bile samples obtained from rats dosed with 1:1 w/w mixture of d0:d7-benzylamine. The studies with 1:1 w/w mixture of d0:d2-benzylamine showed retention of only one deuterium in the MH+ of these two metabolites. Metabolite M12: This was a GSH conjugate with MH+ at m/z 398. The SIM studies suggested that this metabolite accounted for less than 5% of the administered dose. The MS/MS data revealed a characteristic loss of 129 amu (-pyroglutamate) to give the base peak at m/z 269. An ion at m/z 323 formed by loss of glycine was also

observed in the mass spectrum. The analysis of bile samples from rats dosed with d0- and d7-benzylamine showed MH+ ions at m/z 398 and 404, respectively. The LC/MS retention time and MS/MS fragmentation pattern of synthetic standard was obtained and found to be identical to those produced by the metabolite. The 1H NMR of the adduct M12 is shown in Figure 8. The structure of M12, shown in Scheme 1, was confirmed as S-benzyl-L-glutathione. Metabolite M13: This metabolite was formed via catabolism of the M12. The SIM studies suggested that this metabolite accounted for less than 3% of the administered dose. The mass spectrum of metabolite showed MH+ at m/z 212. The loss of one deuterium atom from the aminomethyl side chain was confirmed by observing MH+ at m/z 212, 213, and 218 from bile samples of rats dosed with d0-, d2-, and d7-benzylamine (see Scheme 1). The MS/MS spectrum of this conjugate showed ions at m/z 195 (base peak) and 91. The LC/MS retention time and MS/MS fragmentation pattern of synthetic standard was obtained and found to be identical to those produced by the metabolite. The structure of M13, shown in Scheme 1, was confirmed as S-benzyl-L-cysteine. Metabolites M14-M15: Metabolites M14 and M15 were confirmed as GSH conjugates with MH+ at m/z 473 and 455, respectively. The structures of these metabolites were tentatively based on MS/MS data from labeled and nonlabeled compounds. Synthetic standards were not available to confirm the structures of these metabolites. Metabolites M14 and M15, based on the SIM experiments done with other GSH conjugates, each appeared to account for less than 1% of the administered dose. Characterization of Metabolites in N-Hydroxylbenzylamine Dosed Rats. Rats dosed with N-hydroxylbenzylamine excreted a GSH conjugate, which produced

1200


Figure 8.

1H

Mutlib et al.

NMR of S-benzyl-L-glutathione, M12.

Scheme 3. Proposed Pathways Leading to the Formation of M5-M11 from Benzylaminea

a A reactive intermediate, Ar-CH -NH(CO)-X such as carbamyl glucuronide (pathway 1 with X as glucuronic acid) or benzyl isocyanate 2 (pathway 2) formed from a formamide (X ) H) are postulated to react with nucleophiles such as GSH, GSSG and glutamate to form the metabolites M5-M11.

MH+ at m/z 427. The SIM studies suggested that this metabolite accounted for less than 10% of the administered dose. The MS/MS spectrum of this conjugate showed ion fragments at m/z 352, 298, 265 (base peak), and 177. The formation of this conjugate has previously been postulated to occur via the hydroxylamine and the nitroso intermediates (Scheme 5) (5). The structure of this metabolite was confirmed by comparing the chromatographic and mass spectral behavior of the metabolite with an authentic standard that was previously synthesized (5). It was shown that the metabolite found in rat

bile was identical to the standard and hence was confirmed as having the structure as shown in Figure 9. In addition to the GSH adduct, rats excreted both M1 and benzylamine. These were the major metabolites accounting for greater than 90% of the administered dose as determined by SIM. None of the GSH adducts previously identified in the benzylamine-dosed rats were found in the bile of animals dosed with N-hydroxyl benzylamine. Characterization of Metabolites in Benzamide Dosed Rats. Rats dosed with benzamide produced large quantities of metabolites M2, M3, and M4. The LC/UV



Scheme 4. Proposed Pathway for the Formation of M2 from Benzylamine via Benzaldehyde Intermediate

Scheme 5. Oxidation of N-Benzylhydroxylamine to Reactive Intermediates and Subsequent Conjugations

profiles indicated that the GSH-derived adducts (M2, M3, and M4) were the major drug-related components in all of the bile samples. Trace quantities of benzamide were found in bile and urine samples. A representative LC/ UV profile of bile from a rat dosed with benzamide is shown in Figure 10. Interestingly, the hippuric acid (M1) was also found in the bile and urine samples. M1 was the major metabolite in urine of benzamide dosed rats. In addition to these metabolites, at least two glutamate conjugates, present as minor metabolites, were found. These were the glutamate adducts of M2 and M4, giving MH+ at m/z 574 and 388, respectively. These adducts were not characterized further. Characterization of Metabolites in Benzaldehyde Dosed Rats. Benzaldehyde dosed rats excreted hippuric M1 as the major metabolite (greater than 90% of the administered dose) in both bile and urine. However, it was found that metabolites M12 and M13 were also present in bile samples. Our SIM studies suggested that these metabolites accounted for less than 10% of the administered dose. The identities of M12 and M13 were confirmed by comparing the MS/MS data previously obtained from synthetic standards. A metabolite corresponding to the cysteinylglycine conjugate intermediate

derived from M12 was also detected in the bile. This metabolite, with MH+ at m/z 269, produced ion fragments at m/z 252, 166, and 149. The levels of this metabolite could not be estimated due to the unavailability of the synthetic standard.

Discussion We had previously shown that DPC 423 (Figure 1), a compound possessing a benzylamine moiety, was converted to several novel metabolites (3-5). Studies were performed to compare and contrast the metabolic disposition of benzylamine with DPC 423 and its analogues. The in vivo metabolism of benzylamine was investigated in rats, and the structures of metabolites were identified by LC/MS, high-field NMR, and by comparison with synthetic standards. Furthermore, the metabolism of primary metabolites (benzaldehyde and benzamide) of benzylamine was studied to confirm the proposed metabolic pathways (Scheme 1). Benzylamine was converted primarily to hippuric acid by rats, although a number of other minor metabolites were detected and characterized (Scheme 1). These included novel metabolites produced by pathways not previously described. Some of the

1202


Mutlib et al.

Figure 9. LC/MS/MS TIC (top) and MS/MS spectrum (middle) of the protonated parent ion MH+ at m/z 427 of the GSH conjugate of benzaldehyde oxime present in rat bile dosed with benzylhydroxylamine HCl. The UV trace of the sample is shown at the bottom [the asterisk (*) representing the GSH adduct in the bile].

Figure 10. LC/MS/UV profile of bile from a rat dosed with 300 mg/kg of benzamide. The UV chromatogram (A) shows the presence of M2, M3, and M4. The mass spectrum of the GSH adduct (M2) is shown in the bottom pane (B).

metabolites were formed through P450 oxidation while others were formed independent of P450. The metabolite profile of benzylamine in rats showed the presence of various GSH adducts which were formed via different metabolic pathways. In addition to the known metabolic pathways that led to bioactivation of benzylamines, at least two previously unreported metabolic routes were discovered during these studies. Benzylamine was metabolized to benzaldehyde via oxidative deamination. The aldehyde was rapidly converted to the carboxylic acid, which was subsequently conjugated with glycine and excreted as hippuric acid

(M1). The hippurate was the major metabolite excreted in the bile and urine of rats dosed with benzylamine, benzaldehyde, or benzamide. Studies with benzaldehyde confirmed that it was being formed from benzylamine as a precursor to the carboxylic acid. Dosing rats with an equimolar mixture of d0- and d7-benzylamine confirmed the oxidative deamination of benzylamine. The LC-APCI/ MS spectrum showed that the d0- and d7-benzylamine produced protonated (MH+) ions at m/z 180 and 185 of nonlabeled and labeled hippuric acid metabolites, respectively. However, the relative intensities of these ions suggested a significant intramolecular deuterium isotope


effect in vivo. The deuterium isotope effect in forming a carboxylic acid from a benzylamine has been previously demonstrated (4). The use of deuterium-labeled benzylamine also provided evidence that the hippuric acid was indeed a metabolite and not entirely an endogenous product. Analysis of control bile samples showed that hippuric acid was excreted in the bile and urine of rats as an endogenous product, albeit at a very low level. Hippuric acids are normal constituents in urine of mammals, and its elevated levels in urine of animals are sometimes used as a biomarker of tissue damage (16, 17). On dosing the animals with benzylamine, benzaldehyde or benzamide, the levels of hippuric acid in urine increased by greater than 1000-fold. The deuterium-labeled benzylamine demonstrated that the hippuric acid was indeed a metabolite, and its high levels in the urine/bile were not due to an injury to a tissue. The in vivo studies performed with the primary metabolites, benzaldehyde (formed by MAO), and benzamide (formed by P450) showed that both of these compounds were capable of forming the hippuric acid. The aldehydes produced by oxidative deamination of xenobiotics are generally metabolized either to alcohols by aldehyde reductase or alcohol dehydrogenase or to acids by aldehyde dehydrogenase or aldehyde oxidase. In this instance it appears that rats preferentially convert the aldehyde to carboxylic acid. The aldehyde and carboxylic acid metabolic intermediates were not detected in the bile/urine samples probably due to the rapid conjugation of the benzoic acid with glycine. The mitochondrial metabolism of benzylamine to the benzaldehyde, mediated by amine oxidases, has been previously described (10-15). Benzaldehydes are converted to benzoic acid by the aldehyde dehydrogenase present in the hepatic mitochondria. The benzoic acid has been found to be conjugated with glycine in the mitochondria where all the cofactorssATP, CoA, and glycinesexist (18). The extremely low turnover of benzylamine by rat microsomes suggests that P450 may not play a significant role in converting it to the benzoic acid. This was supported by coadministering benzylamine with ABT, a potent P450 inhibitor (19, 20), to rats. The analysis of bile/urine from these rats showed that the levels of M1 were essentially unchanged while the formation of other metabolites such as M2 and M12 was reduced but not totally abolished. This is consistent with the previous observations that benzylamine is a good substrate for MAO present in mitochondria (10-15). The structures of the GSH adducts suggest that benzylamine is metabolized to several chemically reactive intermediates capable of interacting with a nucleophile such as GSH. The formation of M2 appears to be via an initial oxidation of benzylamine to benzamide, which is subsequently converted to an epoxide intermediate. The formation of amide from amines has been very well documented, especially for cyclic compounds (21, 22). The formation of cotinine from nicotine is a typical example of such a metabolic pathway (21). Nicotine is R-hydroxylated in the pyrrolidine ring to give a carbinolamine that is oxidized further to the major metabolite, cotinine. However, the formation of amides from alkylamines is usually precluded by the rapid hydrolysis of the unstable carbinolamine that is formed during the P450-catalyzed N-dealkylation. This usually leads to an N-dealkylated product and an aldehyde instead of a further oxidized


product. On the other hand, hydroxylation at the benzylic position of benzylamine leads to a product capable of being further oxidized. The remaining proton at the benzylic position becomes very acidic (after the initial hydroxylation) and is amenable to an abstraction by P450. An abstraction of the benzylic proton (one electron abstraction) would produce a radical that is resonancestabilized over the aromatic ring. A subsequent P450catalyzed electron abstraction would yield the corresponding benzamide. The existence of M2 is indirect evidence for the existence of a putative epoxide intermediate (Scheme 1). The epoxide intermediate is cleaved by GSH to form a product that yields a characteristic NMR pattern for a trisubstituted hexadiene ring (see Figure 3 for 1H NMR of cysteinylglycine conjugate, M3). The identification of M2 and M3 was achieved by NMR analysis of the isolated metabolites. Rats that were dosed with 1:1 w/w mixture of d0:d2-benzylamine produced M3 that showed MH+ at m/z 316 only; i.e., both the deuteriums were lost. Studies with 1:1 w/w mixture of d0:d7benzylamine showed loss of two deuteriums and retention of five labels, presumably on the aromatic ring of M3 (see Figure 11). This study clearly demonstrated that oxidation occurred at the labeled aminomethyl side chain. Additional in vivo studies with benzamide confirmed the hypothesis that benzylamine was converted to benzamide prior to the formation of M2. Metabolites M3 (cysteinylglycine) and M4 (cysteine) were found as catabolites of M2. In addition to these metabolites, glutamate conjugates of M2 and M4 were also found in rat bile. This is consistent with our previous observations that glutathione-derived adducts are capable of being conjugated with glutamic acid via γ-glutamyltranspeptidase (1, 3). The results from the in vivo studies were consistent with the data obtained from in vitro studies. It was demonstrated that benzylamine was metabolized to the benzamide and 3,4-dihydroxybenzamide by rat CYP2A1 and 2E1 enzymes (data not shown). The presence of 3,4dihydroxybenzamide from incubations with either benzylamine or benzamide suggests that CYP2E1 and 2A1 are responsible for both of the metabolic steps (see Scheme 2). It is postulated that the benzamide is converted to an epoxide that reacts with GSH in the presence of a specific GST to form M2. In contrast, in the absence of GST, as is the case with cDNA expressed microsomal system, this expoxide is further catabolized to the catechol, 3,4-dihydroxybenzamide. The structures of M5-M11 show an unusual addition of nucleophiles such as GSH and glutamate to benzylamine. A prerequisite modification of benzylamine may be required prior to the addition of these nucleophiles. The structures of the metabolites suggest that benzylamine is converted to a reactive metabolite via a novel metabolic pathway. It is postulated that the reactive intermediate is either the benzyl isocyanate or an unknown precursor, Ar-CH2-NH(CO)-X (see Scheme 3), capable of undergoing a nucleophilic displacement. Interaction of benzyl isocyanate or Ar-CH2-NH(CO)-X with GSH, glutamic acid, cysteinyl glutamate, or oxidized GSH leads to the formation of M5, M9, M10, and M11, respectively. Further catabolism of M5 via the mercapturic acid pathway produces metabolites M6, M7, and M8. The formation of GSH adducts from isocyanates and isothiocyanates have been described in the literature (23-26). The partial positive charge residing on the carbonyl of isocyanates makes it susceptible to nucleo-

1204


Mutlib et al.

Figure 11. LC/MS spectra of cysteinylglycine conjugate (showing the MH+) present in bile samples of rats dosed with 300 mg/kg of (1:1 w/w) mixture of d0:d7-benzylamine hydrochloride (A) and d0:d2-benzylamine hydrochloride (B).

philic attack (e.g., GSH). However, the mechanism for the formation of the benzyl isocyanate from benzylamine has not been fully elucidated. In vitro studies with various rat P450 enzymes (fortified with GSH) suggested that M5 could not be produced by any of the isoenzymes. This was expected since P450 would not add an element of CO to the benzylamine. The possible pathways leading to the formation of these unusual conjugates (M5-M11) are shown in Scheme 3. These pathways suggest a formation of new carbon-nitrogen bond in benzylamine. The formation of new carbon-nitrogen bonds via Nmethylation, N-acetylation, and N-glucuronidation has been described previously (27-29). All of these metabolic processes are non-P450 dependent and are considered phase II metabolism. However, none of these metabolic pathways are expected to yield an intermediate that could ultimately lead to the formation of metabolites M5-M11. It is postulated that a formamide of benzylamine could be formed where X ) H (in Scheme 3). The existence of formamides as metabolites of a number of amines has been described previously (30-34). The formamide metabolites of these compounds have been attributed to both enzymatic formylation mechanisms involving enzymes such as kynurenine formamidase (35) in the case of some arylamines or have been demonstrated to arise from alpha carbon oxidation in the case of aminopyrine (36) and recipavrin (34). Kynurenine formamidase is capable of transferring the formyl group (from N-formyl-L-kynurenine) to aromatic amines such as aniline (35). Usually formylation catalyzed by this enzyme is regarded as a detoxification pathway, especially for aromatic amines. However, in this case the formation of formamide and its subsequent metabolism could have led to a more reactive isocyanate intermediate as shown previously (25, 37, 38). The toxicities of formamides have been attributed to the formation of these reactive isocyante intermediates via P450-catalyzed reactions (38-41). Sequential one-electron oxidations of methyformamide by P450 2E1 have been suggested to lead to the reactive methyl isocyanate metabolite (40). The

administration of ABT (P450 inhibitor) to rats dosed with benzylamine perhaps did not produce a total elimination of P450 activity, hence the conversion of formamide to isocyanate was not totally blocked. It is also proposed that the amine nitrogen can react with an endogenous compound to form an acyl conjugate (such as the carbamyl glucuronide), which has the potential to interact with other endogenous components or decompose to an isocyanate (Scheme 3). The formation of carbamic acid from amines is a metabolic pathway that leads to the creation of a new carbon-nitrogen bond. Due to its inherent instability, the existence of carbamic acid metabolites is usually demonstrated by characterizing their glucuronide conjugates, often referred to as carbamyl glucuronides. The formation of carbamic acids from amines trapped as carbamyl O-β-D-glucuronides has been described previously (42-44). The pathway leading to carbamic acid intermediates from amines is believed be due to the nonenzymatic reaction of the amine with dissolved carbon dioxide (44). Spontaneous formation of carbamates has been reported for a number of amino acids (45) and other amines (46). Compared to our previously studied benzylamine, DPC 423 (4, 5), which formed the carbamyl glucuronide as the major metabolite, benzylamine itself did not excrete a similar carbamyl glucuronide conjugate. One could postulate that if the carbamyl glucuronide of benzylamine was formed, it could have acted as the reactive intermediate Ar-CH2NH(CO)-X, with glucuronic acid as a good leaving group (X in Scheme 3). The high reactivity of this carbamyl glucuronide could explain its absence in the biological fluids of rats dosed with benzylamine. Furthermore, since the formation of carbamyl glucuronide is not dependent on P450 enzymes, administration of ABT would not have affected the levels of metabolites M5-M11. Indeed, rats dosed with ABT showed only a slight decrease in the levels of M5-M11. Experiments will be conducted in the future to determine if the carbamyl glucuronide or the formamide of benzylamine are potential precursors to metabolites M5-M11.


The reactivity of benzyl isocyanate or its precursor was also demonstrated by isolating and characterizing other peptide adducts such as the glutamate (M9) and oxidized GSH adducts (M11). The formation of isocyanate-derived peptide adducts was interesting since it showed that the reactive intermediate was not only capable of interacting with sulfhydryl but also reacted with other nucleophiles such as the amino groups of the glutamate and GSSG. Adducts of GSSG are seldom, if ever, described for compounds that form reactive intermediates. Typically, the GSH adduct, formed by an interaction of sulfhydryl of GSH with an electrophile, is identified. An interaction of amino group of GSSG with an electrophilic metabolite such as the benzyl isocyanate has not been previously described. LC/MS/MS capability of an ion trap instrument as well as the availability of selectively labeled benzylamine facilitated the identification of this conjugate. The LC/MS/MS spectra of labeled and nonlabeled benzylamine-GSSG adduct with MH+ at m/z 746 and 753, respectively, are shown in Figure 7. MS/MS studies were done to demonstrate that the benzylamine moiety was linked through a urea bond to the glutamate amino functional group of the oxidized GSH. A synthetic standard of this hexapeptide adduct was made, and LC/MS/ MS studies were performed to show that it was identical to the biological product. The postulated presence of a reactive isocyanate or its precursor Ar-CH2-NH(CO)-X is indeed a novel pathway for metabolism of a nitrogen containing compound. A review of the literature revealed that such a metabolic reaction has not been described before. Importantly, the reactivity of benzylamine via formation of isocyanate or an isocyanate-derived metabolic intermediate may possibly extend to other amines. Toxicities associated with the bioactivation of amines may be attributed in part to existence of such reactive intermediates. With the application of versatile analytical techniques, it is now possible to identify some of these elusive metabolites that were once difficult to characterize. The formation of M12-M13 showed a net displacement of an amino group by GSH. A nucleophilic displacement of halogen, such as bromide, with GSH has been previously reported for haloalkanes (47-49). Halogens are good leaving groups and their displacement by a nucleophile such as GSH is expected to occur with or without enzyme catalysis. However, a direct displacement of amino functional group with GSH is unprecedented and is not expected to occur without any bioactivation of the amine. Studies done with deuteriumlabeled benzylamine showed that one of the deuteriums on the aminomethyl side chain was lost on formation of M12. This implied that at least one electron oxidation must have taken place prior to the formation of M12. The metabolic conversion of benzylamine to benzaldehyde was proposed and a study was subsequently conducted in vivo to demonstrate that benzaldehyde was the precursor to M12. A mechanism is proposed whereby the aldehyde is further reduced to an alcohol, which is subsequently converted by phase II enzymes to a good leaving group. Displacement of this leaving group with GSH would produce M12 (Scheme 4). Metabolite M13 is a cysteine adduct derived from M12 via the mercapturic acid pathway. The identities of M12 and M13 were confirmed by comparisons with synthetic standards. In addition to these metabolites, a number of other GSH adducts were also detected and characterized in bile of rats dosed with


benzylamine. These included metabolites such as M14 and M15 whose structures were tentatively assigned based on mass spectral data. Because these studies were performed at appreciable doses of benzylamine, it is difficult to predict the relative contribution of each pathway at trace doses of the compound. It is quite possible that, at the high doses of benzylamine, metabolic switching toward the formation of hippuric acid predominated. Alternatively, the monoamine oxidases responsible for the deamination to benzaldehyde are present in abundant quantities in vivo and are capable of easily metabolizing benzylamine at all dose levels. Further studies with low doses of benzylamine were not performed. However, such experiments may elucidate the relative contributions of other metabolic pathways (as compared to hippuric acid pathway) at low doses of benzylamine. The metabolic disposition of benzylamine was also investigated to provide mechanistic insight into the biotransformation of the previously described benzylamine, DPC 423. Although there were clear similarities observed in the metabolites generated from the respective compound, e.g., oxidative deamination to the aldehyde, we found there to be several diverging pathways from that observed previously from DPC 423 metabolism. Previous work in our laboratory resulted in a proposed scheme for DPC 423 involving the oxidation of the benzylamine moiety to a hydroxylamine intermediate, which, after further oxidation to the nitroso species, tautomerizes to its corresponding oxime. The oxime of DPC 423 was consequently conjugated by GSH (4), the mechanism of which is not fully elucidated. Although the current data support the involvement of a similar oxidation pathway for the metabolism of the aminomethyl side chain of N-benzylhydroxylamine (see Scheme 5), the metabolism of benzylamine diverges, mediated predominantly by MAO, to yield the benzaldehyde intermediate. Subsequent oxidation of benzaldehyde and conjugation by glycine leads to the formation of the hippurate M1 and represents a divergent pathway from that observed with DPC 423. Importantly, one can also imagine a scheme in which the oxime of N-benzylhydroxylamine is hydrolyzed and conjugated to form the hippurate M1 (Scheme 5). However, it appears that benzylamine does not form the N-hydroxylbenzylamine as readily as DPC 423. It is noteworthy that N-hydroxylbenzylamine is disposed of in a manner similar to DPC 423, i.e., further oxidation to a nitroso and forms the GSH conjugate of the oxime. The GSH conjugation of the putative benzaldehyde metabolite, leading to M12 and M13, represents the second divergent metabolic pathway between benzylamine and DPC 423. While there was no apparent intermediacy of an amide implicated in the disposition of DPC 423, such an intermediate appears to represent the third divergent pathway for benzylamine from DPC 423, leading to the GSH-derived metabolites M2, M3, and M4. The formation of the benzamide as a metabolite of benzylamine is perhaps due to the stability conferred on a P450-generated radical intermediate capable of undergoing a second electron abstraction. Likewise, the involvement of the proposed reactive intermediate, Ar-CH2NH(CO)-X, which leads to the formation of the conjugates M5, M9, and M10, represents another major distinguishing fate during the metabolism of benzylamine. The determinants that orchestrate the final outcomes of benzylamine and DPC 423 metabolism seem to be

1206


dictated by the initial oxidation step, i.e., MAO versus P450 oxidation, respectively. However, the obvious structural differences between DPC 423 and benzylamine may alone influence metabolic fates due to substrate specifications of the various phase I and II enzymes involved in disposition of these compounds. Overall, benzylamine was metabolized principally to hippuric acid and to several GSH adducts, the structures of which led us to postulate the existence of reactive intermediates formed via novel metabolic pathways. The characterization of metabolites M5-M11 was achieved by synthesizing appropriate standards and making mass spectral and chromatographic comparisons with metabolites. This provided conclusive evidence for the existence of reactive metabolic intermediates such as the benzyl isocyanate. The existence of these reactive intermediates from benzylamine is novel, and further studies are needed to define the metabolic pathways/enzymes responsible for their formations. The conversion of an amine to such reactive metabolites has been overlooked in the past perhaps due to either difficulty in characterizing such polar metabolites or low levels of metabolites produced via this route. The formation of metabolite M12 was also unique in that the amino functional group was displaced by a GSH moiety, though the in vivo data suggests that several metabolic steps were involved including the formation of benzaldehyde as a precursor. The other GSH-derived adducts (M2, M3, M14, and M15) that were characterized included those which had epoxides as reactive metabolic intermediates. By the use of very sensitive analytical methods, we were able to demonstrate the existence of multiple metabolic pathways, which produced unusual metabolites. Characterization of some of the minor metabolites provided an insight into the possible bioactivation pathways for benzylamine. It is essential to elucidate the structures of all metabolites of a compound so that an appreciation of various metabolic pathways involved in its disposition (detoxification and bioactivation) can be realized.

References (1) Mutlib, A. E., Shockcor, J., Espina, R., Graciani, N., Du, A., and Gan, L. (2000) Disposition of glutathione conjugates in rats by a novel glutamic acid pathway: Characterization of unique peptide conjugates by liquid chromatography/mass spectrometry and liquid chromatography/NMR. J. Pharmacol. Exp. Ther. 294, 735745. (2) Mutlib, A. E., Shockcor, J., Chen, S., Espina, R., Graciani, N., Prakash, S., and Gan, L. (2000) Metabolism of benzylamines to reactive metabolites. Characterization of unusual glutamate, glutathione, glucuronide and sulfamate conjugates, Sixth international symposium on biological reactive intermediates, Paris, July 7-12. (3) Mutlib, A. E., Shockcor, J., Chen, S., Espina, R., Lin, J., Graciani, N., Prakash, S., and Gan, L. (2001) Formation of unusual glutamate conjugates of 1-[3-(aminomethyl)phenyl]-N-[3-fluoro2′-(methylsulfonyl)-[1,1′-biphenyl]-4-yl]-3-(trifluoromethyl)-1Hpyrazole-5-carboxamide (DPC 423) and its analogues. The role of γ-glutamyltranspeptidase in the biotransformation of benzylamines. Drug Metab. Dispos. 29, 1296-1306. (4) Mutlib, A. E., Chen, S., Espina, R., Shockcor, J., Prakash, S., and Gan, L. (2002) P450 mediated metabolism of 1-[3-(aminomethyl)phenyl]-N-[3-fluoro-2′-(methylsulfonyl)-[1,1′-biphenyl]-4-yl]-3-(trifluoromethyl)-1H-pyrazole-5-carboxamide (DPC 423) and its analogues to aldoximes. Characterization of novel glutathione conjugates of postulated intermediates derived from aldoximes. Chem. Res. Toxicol. 15, 63-75. (5) Mutlib, A. E., Shockcor, J., Chen, S. Y., Espina, R., Pinto, D. J., Orwatt, M., Prakash, S., and Gan, L. (2002) Disposition of 1-[3(aminomethyl)phenyl]-N-[3-fluoro-2′-(methylsulfonyl)-[1,1′-biphenyl]-4-yl]-3-(trifluoromethyl)-1H-pyrazole-5-carboxamide (DPC 423)

Mutlib et al.

(6)

(7)

(8)

(9)

(10) (11)

(12)

(13)

(14)

(15)

(16)

(17)

(18)

(19)

(20)

(21)

(22) (23)

(24)

(25)

by novel metabolic pathways. Characterization of unusual metabolites by liquid chromatography/mass spectrometry and NMR. Chem. Res. Toxicol. 15, 48-62 Miller, J. A., Cramer, J. W., and Miller, E. C. (1960) The N- and ring-hydroxylation of 2-acetylaminofluorene during carcinogenesis in the rat. Cancer Res. 20, 950-962. Beland, F. A., and Kadlubar, F. F. (1990) Metabolic activation and DNA adducts of aromatic amines and nitroaromatic hydrocarbons. In Chemical Carcinogenesis and Mutagenesis I, Handbook of Experimental Pharmacology (Cooper, C. S., and Grover, P. L.. Eds.) Vol. 94/1, pp 267-325, Springer-Verlag, Berlin. Kadlubar, F. F., and Hammons, G. J. (1987) The role of cytochrome P-450 in the metabolism of chemical carcinogens. In Mammalian Cytochromes P-450 (Guengerich F. P., Ed) Vol. 2, pp 81-130, CRC Press, Boca Raton, FL. Josephy, P. D., Evans, D. H., Parikh, A., and Guengerich, F. P. (1998) Metabolic activation of aromatic amine mutagens by simultaneous expression of human cytochrome P450 1A2, NADPHcytochrome P450 reductase, and N-acetyltransferase in Escherichia coli. Chem. Res. Toxicol. 11, 70-74. Mantle, T. J., and Tipton, K. F. (1978) Monoamine oxidase A and B: a useful concept? Biochem. Pharamcol. 27, 97-101. Belanger, P. M., and Lalande, M. (1980) The cellular distribution of the metabolic deamination of benzylamine. Life Sci. 26, 20372044. Parkinson, D., Lyles, G. A., Browne, B. J., and Callingham, B. A. (1980) Some factors influencing the metabolism of benzylamine by type A and B monoamine oxidase in rat heart and liver. J. Pharm. Pharmacol. 32, 844-850. Yu, P. H., and Davis, B. A. (1988) Stereospecific deamination of benzylamine catalyzed by different amine oxidases. Int. J. Biochem. 20, 1197-1201. Alton, G., Taher, T. H., Beever, R. J., and Palcic, M. M. (1995) Stereochemistry of benzylamine oxidation by copper amine oxidases. Arch. Biochem. Biophys. 316, 353-361. Lyles, G. A. (1996) Mammalian plasma and tissue-bound semicarbazide-sensitive amine oxidases: biochemical, pharmacological and toxicological aspects. Int. J. Biochem. Cell. Biol. 28, 259274. Holmes, E., Bonner, F. W., Sweatman, B. C., Lindon, J. C., Beddell, C. R., Rahr, E., and Nicholson, J. K. (1992) Nuclear magnetic resonance spectroscopy and pattern recognition analysis of biochemical processes associated with the progression of and recovery from nephrotoxic lesions in the rat induced by mercury (II) chloride and 2-bromoethanamine. Mol. Pharmacol. 42, 922930. Gavaghan, C. L., Nicholson, J. K., Connor, S. C., Wilson, I. D., Wright, B., and Holmes, E. (2001) Directly coupled highperformance liquid chromatography and nuclear magnetic resonance spectroscopic with chemometric studies on metabolic variation in Sprague-Dawley rats. Anal. Biochem. 291, 245-252 Krahenbuhl, L., Reichen, J., Talos, C., and Krahenbuhl, S. (1997) Benzoic acid metabolism reflects hepatic mitochondrial function of rats with long-term extrahepatic cholestasis. Hepatology 25, 278-283. Ortiz de Montellano, P. R., and Reich, N. O. (1986) Inhibition of cytochrome P450 enzymes, Cytochrome P-450 (Ortiz de Montellano P. R., Ed.) p 273, Plenum Press, New York. Meschter, C. L., Mico, B. A., Mortillo, M., Feldman, D., Garland, W. A., Riley, J. A., and Kaufman, L. S. (1994) A 13-week toxicologic and pathologic evaluation of prolonged cytochromes P450 inhibition by 1-aminobenzotriazole in male rats. Fundam. Appl. Toxicol. 22, 369-375. Williams, D. E., Shigenaga, M. K., and Castagnoli, N., Jr. (1990) The role of cytochromes P-450 and flavin-containing monooxygenase in the metabolism of (S)-nicotine by rabbit lung. Drug Metab Dispos. 18, 418-428. Rose, J., and Castagnoli, N., Jr. (1983) The metabolism of tertiary amines. Med. Res. Rev. 3, 73-88. Brusewitz, G., Cameron, B. D., Chasseud, L. F., Gorler, K., Hawkins, D. R., Koch, H., and Mennicke, W. H. (1977) The metabolism of benzyl isothiocyanate and its cysteine conjugate. Biochem. J. 162, 99-107. Pearson, P. G., Slatter, J. G., Rashed, M. S., Han, D. H., Grillo, M. P., and Baillie, T. A. (1990) S-(N-methylcarbamoyl)glutathione: a reactive S-linked metabolite of methyl isocyanate. Biochem. Biophys. Res. Commun. 166, 245-250. Mutlib, A. E., Talaat, R. E., Slatter, J. G., and Abbott, F. S. (1990) Formation and reversibility of S-linked conjugates of N-(1-methyl3,3-diphenylpropyl) isocyanate, an in vivo metabolite of N-(1methyl-3,3-diphenylpropyl)formamide, in rats. Drug Metab. Dispos. 18, 1038-1045.

Bioactivation of Benzylamine (26) Slatter, J. G., Rashed, M. S., Pearson, P. G., Han, D. H., and Baillie, T. A. (1991) Biotransformation of methyl isocyanate in the rat. Evidence for glutathione conjugation as a major pathway of metabolism and implications for isocyanate-mediated toxicities. Chem Res. Toxicol. 4, 157-161. (27) Cundy, K. C., Sato, M., and Crooks, P. A. (1985) Stereospecific in vivo N-methylation of nicotine in the guinea pig. Drug Metab. Dispos. 13, 175-185. (28) Evans, D. A. P. (1992) N-Acetyltransferase. In Pharmacogenetics of Drug Metabolism (Kalow, W., Ed.) pp 95-178, Pergamon, New York. (29) Clarke, D. J., and Burchell, B. (1994) The uridine diphosphate glucuronosyltransferase multigene family: Function and regulation. Handbook Exp. Pharmacol. 112, 3-43. (30) Iguchi, S., Goromaru, T., and Noda, A. (1975) A new metabolite of aminopyrine in man, 4-formylaminoantipyrine. Chem. Pharm. Bull. Jpn. 23, 932-934. (31) Gothoskar, S. V., Benjamin, T., Roller, P. P., and Weisberger, E. K. (1979) N-Formylation of an aromatic amine as a metabolic pathway. Xenobiotica 9, 533-537. (32) Tang, B. K., Grant, D. M., and Kalow, W. (1983) Isolation and identification of 5-acetylamino-6-formylamino-3-methyluracil as a major metabolite of caffeine in man. Drug. Metab. Dispos. 11, 218-220. (33) Koppel, C., and Tenczer, J. (1985) A revision of the metabolic disposition of amantidine. Biomed. Mass. Spectrom. 12, 499-501. (34) Slatter, J. G., Abbott, F. S., Burton, R. (1990) Identification of the biliary metabolites of (()-3-methylamino-1,1-diphenylbutane HCL (Recipavrin) in rats. Xenobiotica 20, 999-1024. (35) Santi, R. S. S., and Hopsu-Havu, V. K. (1968) Transformylation of carcinogenic aromatic amines by kynurenine formamidase: a detxification mechanism. Biochem. Pharmacol. 17, 1110-1113. (36) Noda, A., Tsubone, N., Mihara, M., Goromaru, T., and Iguchi, S. (1976) Enzymatic demethylation and oxidation of aminopyrine and 4-monomethylaminoantipyrine. Chem. Pharm. Bull. Jpn. 24, 3229-3231. (37) Kestell, P., Threadgill, M. D., Gescher, A., Gledhill, A. P., Shaw, A. J., and Farmer, P. B. (1987) An investigation of the relationship between hepatotoxicity and the metabolism on N-alkylformamides. J. Pharmacol Exp. Ther. 240, 265-270. (38) Cross, H., Dayal, R., Hyland, R., and Gescher, A. (1990) Nalkylformamides are metabolized to N-alkylcarbamoylating species by hepatic microsomes from rodents and humans. Chem Res. Toxicol. 3, 357-362.

Chem. Res. Toxicol., Vol. 15, No. 9, 2002 1207 (39) Pearson, P. G., Gescher, A., and Harpur, S. (1987) Hepatotoxicity of N-methylformamide in mice-I. Relationship to glutathione status. Biochem. Pharmacol. 36, 381-390. (40) Hyland, R., Gesher, A., Thummel, K., Sciller, C., Jheeta, P., Mynett, K., Smith, A. W., and Mraz, J. (1991) Metabolic oxidation and toxification of N-methylformamide catalyzed by the cytochrome P450 isoenzyme CYP2E1. Mol. Pharmacol. 41, 259-266. (41) Mraz, J., Jheeta, P., Gescher, A., Hyland, R., Thummel, K., and Threadgill, M. D. (1993) Investigation of the mechanistic basis of N,N-dimethylformamaide toxicity. Metabolism of N,N-dimethylformamide and its deuterated isotopomers by cytochrome P450 2E1. Chem Res. Toxicol. 6, 197-207. (42) Straub, K., Davis, M., and Hwang, B. (1988) Benzazepine metabolism revisited. Evidence for the formation of novel amine conjugates. Drug Metab. Dispos. 16, 359-366. (43) Tremaine, L. M., Stroh, J. G., and Ronfeld, R. A. (1989) Characterization of a carbamic acid ester glucuronide of the secondary amine sertraline. Drug Metab. Dispos. 17, 58-63. (44) Delbressine, L. P. C., Funke, C. W., van Tilborg, M., and Kaspersen (1990) On the formation of carbamate glucuronides. Xenobiotica 20, 133-134. (45) Mirrow, J. S., Keim, P., and Gurd, F. R. N. (1974) Carbon dioxide adducts of certain amino acids, peptides and sperm whale myoglobin studied by carbon 13 and proton nuclear magnetic resonance. J. Biol. Chem. 249, 7484-7494. (46) Greenway, W., and Whatley, F. R. (1987) Formation of N-carboxy compounds during trimethylsilanization of amino acids in the presence of carbon dioxide. J. Chromatogr. 409, 383-389. (47) Van Bladeren, P. J., Van der Gen, A., Breimer, D. D., and Mohn, G. R. (1979) Stereoselective activation of vicinal dihalogen compounds to mutagens by glutathione conjugation. Biochem. Pharmacol. 28, 2521-2524. (48) Guengerich, F. P., Peterson, L. A., Cmarik, J. L., Koga, N., and Inskeep, P. B. (1987) Activation of dihaloalkanes by glutathione conjugation and formation of DNA adducts. Environ. Health Perspect. 76, 15-18. (49) Marchand, D. H., and Reed, D. J. (1989) Identification of the reactive glutathione conjugates S-(2-chloroethyl)glutathione in the bile of 1-bromo-2-chloroethane-treated rats by HPLC and precolumn derivatization with o-phthalaldehyde. Chem Res. Toxicol. 2, 449-454.

TX020063Q

Bioactivation of Benzylamine to Reactive ... - ACS Publications

Recommend Documents