Chem. Res. Toxicol. 1999, 12, 831-839
831
Glutathione Conjugation of Electrophilic Metabolites of 1-Nitronaphthalene in Rat Tracheobronchial Airways and Liver: Identification by Mass Spectrometry and Proton Nuclear Magnetic Resonance Spectroscopy Katherine C. Watt,*,† Dexter M. Morin,† Mark J. Kurth,‡ Roger S. Mercer,§ Charles G. Plopper,| and Alan R. Buckpitt† Department of Molecular Biosciences, School of Veterinary Medicine, Department of Chemistry, Facility for Advanced Instrumentation, and Department of Anatomy, Physiology and Cell Biology, School of Veterinary Medicine, University of California at Davis, Davis, California 95616 Received February 8, 1999
1-Nitronaphthalene (1-NN) is a mutagenic nitroaromatic which has been detected in emissions from both heavy- and light-duty diesel engines, as well as in urban airborne particles. 1-NN is a cytochrome P450-bioactivated, nonciliated bronchiolar epithelial (Clara) cell cytotoxicant. These studies examined the metabolism of 1-NN to electrophilic metabolites which were trapped as glutathione conjugates in highly susceptible (lung) and less susceptible (liver) tissues of the rat. Significant depletion of reduced glutathione was observed at all levels of tracheobronchial airways of rats treated with 200 mg/kg 1-NN, ip. This observation of depleted glutathione was consistent with the HPLC radioactivity profiles demonstrating six glutathione conjugates isolated from liver and lung microsomal incubations with 1-NN, [3H]glutathione, and glutathione S-transferase. Mass spectrometry of all six metabolites in electrospray positive ion mode yielded an ion of m/z 497 (M + H), and daughter ions of m/z 479 (loss of water), m/z 306 (glutathione), and m/z 177 (loss of the nitro group and formation of hydroxy naphthalene thiolate ion), demonstrating the formation of hydroxy-dihydroglutathionyl derivatives presumably via intermediate epoxide(s). Proton nuclear magnetic resonance spectroscopy identified four different regioisomeric conjugates from lung and liver microsomal incubations: 1-nitro7-glutathionyl-8-hydroxy-7,8-dihydronaphthalene, 1-nitro-7-hydroxy-8-glutathionyl-7,8-dihydronaphthalene, 1-nitro-5-hydroxy-6-glutathionyl-5,6-dihydronaphthalene, and 1-nitro-5glutathionyl-6-hydroxy-5,6-dihydronaphthalene. HPLC radioactivity profiles demonstrated that major conjugates generated in the lung were derived from the C7,C8-epoxide, whereas the most prominent metabolites in the liver were derived from the C5,C6-epoxide.
Introduction Nitroaromatics are present in both the particulate and gas-phase fractions of the exhaust that is derived from the incomplete combustion of both gasoline and diesel fuel (1). 1-Nitronaphthalene (1-NN)1 is generated from naphthalene in the presence of N2O5 at room temperature * To whom correspondence should be addressed: Department of Molecular Biosciences, 1311 Haring Hall, School of Veterinary Medicine, University of California, One Shields Ave, Davis, CA 95616. Telephone: (530) 752-0793. Fax: (530) 752-4698. E-mail: kcwatt@ ucdavis.edu. † Department of Molecular Biosciences, School of Veterinary Medicine. ‡ Department of Chemistry. § Facility for Advanced Instrumentation. | Department of Anatomy, Physiology and Cell Biology, School of Veterinary Medicine. 1 Abbreviations: CDNB, 1-chloro-2,4-dinitrobenzene; conjugate 1, 1-nitro-7-hydroxy-8-glutathionyl-7,8-dihydronaphthalene; conjugates 3 and 4, 1-nitro-7-glutathionyl-8-hydroxy-7,8-dihydronaphthalene; conjugates 5 and 6, 1-nitro-5-hydroxy-6-glutathionyl-5,6-dihydronaphthalene; conjugate 7, 1-nitro-5-glutathionyl-6-hydroxy-5,6-dihydronaphthalene; COSY, correlation spectroscopy; dihydrodiol, 1-nitro-5,6dihydro-5,6-dihydroxynaphthalene; HEPA-filtered, high-efficiency particulate air-filtered; 1-NN, 1-nitronaphthalene; 2-NN, 2-nitronaphthalene; nitro-PAH, nitropolycyclic aromatic hydrocarbon; , extinction coefficient; λmax, wavelength at maximum absorbance.
under laboratory conditions (N2O5 is a gas-phase reaction product of ozone and nitrogen dioxide) (2), and is a likely byproduct of gas-phase atmospheric reactions in the South Coast air basin of California (3). Parenteral administration of 1-NN produces lesions in the nonciliated bronchiolar epithelial (Clara) cells (46), and hence, this compound serves as a good model for studying the metabolism and toxicity of nitroaromatics in this cell type. Studies by Johnson and Cornish (7) indicate that naphthylamine is one of the primary metabolites isolated in urine after single ip injections of 1- or 2-NN in rats, indicating that nitronaphthalene is metabolized by reductive pathways in vivo. Subsequent work by Rasmussen (5) indicates that 1-NN is metabolized by P450 in both lung and liver microsomal incubations to electrophilic metabolites that are bound covalently to microsomal proteins. However, the failure of oxygen to inhibit the binding suggests that reductive metabolism has a minor role in the generation of reactive intermediates. Moreover, studies by Halladay et al. (8) indicated that 96 h after either an ip administration of 100 mg/kg or an iv administration of 10 mg/kg [14C]-1-NN in rats, the major urinary metabolite is N-acetyl-S-(hydroxydihydro-1-NN)-L-cysteine, suggesting the formation of an
10.1021/tx990023v CCC: $18.00 © 1999 American Chemical Society Published on Web 08/14/1999
832 Chem. Res. Toxicol., Vol. 12, No. 9, 1999
Watt et al.
Scheme 1. Possible Pathways of 1-NN Metabolism
intermediate epoxide. Hydroxylated, dihydroxylated, glucuronidated, and sulfated metabolites also were identified in the urine through LC/MS/MS, and the presence of glucuronide and sulfate metabolites was confirmed by enzymatic hydrolysis. The acute toxicity of 1-NN to the bronchiolar epithelium is decreased by pretreatment of rats with either pseudocumene or phosphorothionate, both of which inhibit pulmonary P450 activities (9). It has not been established whether toxicity is associated with the covalent binding of reactive metabolites generated by P450 metabolism and/or by reactive oxygen species that may have been generated by redox cycling to the nitrocation radical (Scheme 1). Previous studies with other pulmonary toxicants, including 4-ipomeanol, naphthalene, 3-methylindole, and dichloroethylene, in microsomal preparations from lung and liver indicate that GSH prevents the macromolecular covalent binding of the highly reactive metabolites through the formation of less reactive, water-soluble conjugates (10). This conjugation is believed to be an important modulator of the susceptibility to the pulmonary alkylation and cytotoxicity of 4-ipomeanol and naphthalene in vivo (11, 12). The time course and concentration dependence of GSH depletion correspond to the rapid formation of naphthalene oxide glutathione conjugates (13, 14). In vitro studies have shown striking differences in the stereochemistry and the rate of naphthalene metabolism to enantiomeric epoxides in the lung (target tissue) and the liver (nontarget tissue) of the mouse. Likewise, substantial differences in both the rates and stereochemistry of metabolites are observed between mouse lung and lungs of nonsusceptible species such as the rat (14). Accordingly, these studies were designed to utilize GSH to trap reactive metabolites generated from 1-NN in rat tracheobronchial airways and liver, to identify these conjugates using MS and proton NMR spectroscopy,
and to determine whether the nature of the reactive metabolites varied between target and nontarget tissues.
Experimental Procedures Caution: 1-NN (mutagen) is hazardous and should be handled with care. Wear proper protective clothing while handling this chemical. Experimental Animals. Male Sprague-Dawley rats (250300 g body weight) were purchased from Charles River Laboratories (Hollister, CA). Animals were housed over inert bedding in cages within HEPA-filtered laminar air flow cabinets. They were allowed free access to food and filtered, deionized water, and kept on a 12 h light/dark cycle in facilities at the University of California at Davis which are certified by the American Association for the Accreditation of Laboratory Animal Care. They were used no sooner than 7 days after receipt from the supplier. Chemicals and Reagents. 1-NN was purchased from Aldrich Chemical Co. (Milwaukee, WI) and was recrystallized from ethanol before use (mp 61.5 °C). GSH and GSSG were purchased from Fluka Chemical Corp. (Milwaukee, WI). Glutathione S-transferase was purified from mouse liver cytosol by affinity chromatography. Waymouth’s medium 752/1, containing 5 mM HEPES [4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid], was purchased from GIBCO Labs (Grand Island, NY). Lowgelling temperature agarose (Compatigel) was purchased from FMC BioProducts (Rockland, ME). All other chemicals were reagent grade or better. Radioactive Chemicals. [3H]-1-NN was prepared in the laboratory by the nitration of [2,3-3H]naphthalene (obtained by reductive dehalogenation of 2,3-dibromonaphthalene) (15). The product was purified by normal-phase chromatography using 100% hexane as the mobile phase. The radiochemical purity of the final product, tested by HPLC on a silica column, was shown to be greater than 95%. Impurities did not coelute with any of the metabolites and, accordingly, did not affect the analysis conducted during these studies. The final specific activity of [3H]1-NN for the in vitro metabolism studies was 2.4 mCi/mmol. [glycine-2-3H]-GSH (specific activity of 44 800 mCi/mmol) was
1-Nitronaphthalene Glutathione Conjugates in Rats purchased from Dupont NEN Life Science Products (Boston, MA). The specified radiochemical purity was 97% and was verified by HPLC on a C18 column. The material was used without further purification. Each lot was opened and used immediately to lessen the possibility of oxidation to the disulfide. [3H]-GSH was diluted with unlabeled compound to achieve specific activities of 5-7.4 mCi/mmol for the in vitro studies. 1-NN Administration. 1-NN was dissolved in corn oil and administered ip at a dose of 200 mg/kg (2 mL of corn oil/kg of body weight) 2 h before the animals were euthanized. 1-NN was administered between 8:00 and 10:00 a.m. to avoid differences related to the diurnal variations in tissue GSH content. Control age-matched animals received the same volume of corn oil. Airway Microdissection. The procedure for obtaining defined specimens of the lung by blunt dissection has been described previously in detail (16). The animals were anesthetized with sodium pentobarbital, and the trachea was exposed and cannulated. The lungs were inflated with 1% agarose in Waymouth’s medium and cooled in Waymouth’s medium at 4 °C for 30 min. The airways were then stripped intact from the lobe by microdissection. The subcompartments that were isolated included trachea, lobar bronchi/major daughter intrapulmonary airways, minor daughter intrapulmonary airways, distal bronchioles, and parenchyma. GSH Determination. The amounts of GSH and GSSG in dissected airways were determined by HPLC with electrochemical detection by the method of Lakritz et al. (17). The lower limits of detection were 1 pmol for GSH and 2 pmol for GSSG. Microsome Preparation. Lungs were perfused with isotonic saline before removal from the animal. Dissected airways were prepared by microdissection, and lungs were homogenized in 3 volumes of pH 7.4 buffer consisting of 20 mM Tris, 150 mM KCl, 0.2 mM sodium EDTA, 0.5 mM dithiothreitol, and 15% glycerol at 4 °C with glass-glass or Teflon-glass homogenizers, respectively. Livers were removed quickly and homogenized with Teflon-glass homogenizers in 3 volumes of pH 7.4 buffer consisting of 20 mM Tris and 150 mM KCl at 4 °C. The homogenate was centrifuged at 11000g for 20 min, and microsomal pellets were recovered from the postmitochondrial supernatant by centrifugation at 105000g for 70 min. The final microsomal pellets were resuspended in 0.1 M sodium phosphate buffer (pH 7.4). Protein concentrations were determined by the method of Bradford (18), using bovine serum albumin as the standard. Incubations. Incubations were prepared on ice in a final volume of 300 µL of 0.1 M sodium phosphate buffer (pH 7.4). [3H]-1-NN incubations consisted of 300 µg of microsomal protein, 1 mM [3H]-1-NN, and the NADPH-generating system (consisting of 0.14 mM NADP, 3.8 mM glucose 6-phosphate, 0.1 unit of glucose-6-phosphate dehydrogenase, and 10 mM MgCl2). [3H]GSH incubations consisted of 300 µg of microsomal protein, 1 mM 1-NN, 0.1 mM [3H]-GSH, 10 units/mL CDNB (1 unit of activity is defined as the amount of enzyme catalyzing the formation of 1 µmol of CDNB glutathione conjugate per minute) (19), glutathione S-transferase (partially purified by affinity column chromatography), and the NADPH-generating system. After a 2 min preincubation period with the NADPH-generating system, 1-NN was added with or without GSH and glutathione S-transferase, and the incubation was allowed to proceed for 20 min at 37 °C. The reaction was terminated by adding 1 volume of methanol, and samples were stored at -20 °C overnight for protein to precipitate. All incubations were prepared in triplicate. Controls were prepared without the NADPH-generating system. For the generation of sufficient 1-NN glutathione conjugates for proton NMR spectroscopy, incubations containing a total of 2 g of liver microsomal protein with proportional quantities of 1-NN, GSH, glutathione Stransferase, and the NADPH-generating system were conducted in a final volume of 2 L. Sample Preparation and HPLC Analysis of 1-NN Metabolites and Glutathione Conjugates. Reactions were quenched with methanol, and the incubation mixture was
Chem. Res. Toxicol., Vol. 12, No. 9, 1999 833 centrifuged at 13000g for 30 min at 4 °C to remove the protein. The remaining supernatant was evaporated under vacuum to approximately 50 µL. Samples were stored at -80 °C until they were analyzed. Samples were chromatographed on a Phase Sep C18 reversed-phase column (25 cm × 4.6 mm i.d.; 5 µm particle). The eluates were monitored by UV absorbance at 256 nm. For samples incubated with [3H]-1-NN, a mobile phase of 1% acetic acid in water and acetonitrile was run at a flow rate of 1.0 mL/ min with a linear increase from 5 to 16% acetonitrile over the course of 60 min. For samples incubated with [3H]-GSH, a mobile phase of 0.06% triethylamine phosphate in water (pH 3.1) and acetonitrile was run at a flow rate of 1.0 mL/min with a linear increase from 5 to 16% acetonitrile over the course of 60 min. Triethylamine phosphate was used as a stronger pairing agent for the 1-NN glutathione conjugates to enhance baseline separation on column. The column eluate was collected directly into scintillation vials at 0.5 min intervals. Safety-Solve liquid scintillation fluid (Research Products International Corp., Mt. Prospect, IL) was added, and the samples were counted for either 1 or 5 min each. Quench corrections were made by external standardization. Complete radiochromatographic profiles were obtained. Radioactive peaks corresponding to either the metabolites of 1-NN or glutathione conjugates were summed, and the appropriate background counts were subtracted. For preparative chromatography, either a Rainin Dynamax C18 column (25 cm × 22.5 mm i.d.; 8 µm particle) or a Phenomenex Sphereclone C18 reversed-phase column (25 cm × 21.2 mm i.d.; 5 µm particle) was used. The different peaks were pooled and lyophilized for MS and proton NMR spectroscopy. MS. 1-NN metabolites were analyzed on a Finnigan LCQ (Finnigan Corp., San Jose, CA) ion trap mass spectrometer with a 2000 amu mass range and MS/MS capability using 50:50 v/v acetonitrile/water with 1% acetic acid as the mobile phase. An ABI 140B solvent delivery system was used to deliver the mobile phase at a rate of 100 µL/min. A direct flow injection of 5 µL was used. The instrument was operated in the MS/MS daughter ion mode. The parent ion was isolated with a 2.0 amu window, and a collision energy of 20%. Spectra were obtained in positive ion electrospray mode using a source voltage of 4250 V and a capillary temperature of 220 °C, and scanned over the range of m/z 50-550. The collected data were summed using the acquisition mode in the LCQ software version 1.2. Mass calibration was performed using the standard Finnigan calibration solution containing caffeine, MRFA (a tetrapeptide, Met-Arg-Phe-Ala), and Ultramark as references. Proton NMR Spectroscopy. Spectra were recorded in either deuterated methanol (CD3OD) or deuterium oxide (D2O) (99.99% deuterium) after the samples had been exchanged by evaporation under nitrogen in CD3OD or D2O. Spectra were obtained on either a GE NMR Omega 500 MHz or an Avance 600 MHz instrument equipped with a microprobe. The onedimensional NMR spectra were obtained using an acquisition time of 2.73 s, a block size of 16 384 time domain data points, and a spectral width of 6006 Hz. COSY spectra were obtained using an acquisition time of 0.17 s, a block size of 1024 time domain data points, and 256 blocks. δ values are reported in parts per million. UV Absorbance Analysis. After eluting from the HPLC column in the mobile phase, each of the 1-NN metabolites and glutathione conjugates was collected, evaporated under vacuum to dryness, and redissolved in 50 µL of 50:50 v/v acetonitrile/ water. Each sample was scanned from a wavelength of 700 to 200 nm using a Beckman DU 70 spectrophotometer. The peak pick mode was selected for detecting the wavelength with the highest absorbance. The concentration of each conjugate was then determined according to the specific activity. The was calculated according to the Beer-Lambert law, A ) bc (where A is the absorbance, is the extinction coefficient in M-1 cm-1, b is the path length in centimeters, and c is the molar concentration). Statistical Methods. GSH levels in dissected airways were expressed as picomoles per microgram of protein (mean ( one
834 Chem. Res. Toxicol., Vol. 12, No. 9, 1999
Figure 1. GSH levels in rat airway subcompartments 2 h after an ip dose of 200 mg/kg 1-NN. The GSH content was reported as the mean ( SD obtained from eight rats. An asterisk indicates a significant difference from control using ANOVA followed by the Student-Newman-Keuls test and a two-tailed t test (p < 0.05). SD). Comparisons between control and 1-NN-treated animals were performed by one-way ANOVA using the Student-Newman-Keuls test and a two-tailed t test to identify significant differences (20). A p value of
