ARTICLE pubs.acs.org/crt
Novel Mechanism for Dehalogenation and Glutathione Conjugation of Dihalogenated Anilines in Human Liver Microsomes: Evidence for ipso Glutathione Addition Chenghong Zhang, Jane R. Kenny, Hoa Le, Alan Deese,† Kevin A. Ford,‡ Luke K. Lightning,§ Peter W. Fan, James P. Driscoll, Jason S. Halladay, Cornelis E. C. A. Hop, and S. Cyrus Khojasteh* Drug Metabolism and Pharmacokinetics, and †Small Molecule Analytical Chemistry, ‡Safety Assessment Genentech, Inc., South San Francisco, California 94080, United States § Alquest Therapeutics, 100 St. Albans Road, Kensington, California 94708, United States
bS Supporting Information ABSTRACT: The objective of the present study was to investigate the influence of halogen position on the formation of reactive metabolites from dihalogenated anilines. Herein we report on a proposed mechanism for dehalogenation and glutathione (GSH) conjugation of a series of ortho-, meta-, and para-dihalogenated anilines observed in human liver microsomes. Of particular interest were conjugates formed in which one of the halogens on the aniline was replaced by GSH. We present evidence that a (4-iminocyclohexa-2,5-dienylidene)halogenium reactive intermediate (QX) was formed after oxidation, followed by ipso addition of GSH at the imine moiety. The ipso GSH thiol attacks at the ortho-carbon and eventually leads to a loss of a halogen and GSH replacement. The initial step of GSH addition at the ipso position is also supported by density functional theory, which suggests that the ipso carbon of the chloro, bromo, and iodo (but not fluoro) containing 2-fluoro-4haloanilines is the most positive carbon and that these molecules have the favorable highest occupied molecular orbital of the aniline and the lowest unoccupied orbital from GSH. The para-substituted halogen (chloro, bromo, or iodo but not fluoro) played a pivotal role in the formation of the QX, which required a delocalization of the positive charge on the para-halogen after oxidation. This mechanism was supported by structuremetabolism relationship analysis of a series of dihalogenated and monohalogenated aniline analogues.
1. INTRODUCTION Aniline and its halogenated derivatives are widely used as chemical intermediates during organic synthesis1,2 and have been the subject of many toxicological studies in recent years. The reported toxicities associated with exposure to these chemicals include methemoglobinemia,35 anemia,3 splenotoxicity,6 hepatotoxicity,7 and nephrotoxicity.710 From these findings, it appears that the type and severity of toxicity is based on the number and the sites of halogen substitutions on the aniline ring.5,11,12 It has been proposed that, in order for halogenated anilines to elicit toxicity, they must first undergo bioactivation. A wide range of oxidative and conjugative metabolic processes occur in the bioactivation of halogenated anilines, but initial oxidation by cytochrome P450 is considered to be the most important step in the formation of reactive metabolites that results in protein and/or DNA adducts.1315 The objective of the present study was to investigate the mechanism of a halogen atom displacement by glutathione (GSH) for a series of dihalogenated anilines using human liver r 2011 American Chemical Society
microsomes and GSH as a trapping agent. Herein, we present a proposed mechanism that suggests a (4-iminocyclohexa-2,5dienylidene)halogenium reactive intermediate (QX) is formed from anilines substituted with a halogen at the ortho-position. In addition, GSH reacts at the ipso position of the QX followed by an intramolecular cyclization and migration via reaction of its sulfhydryl group to the ortho-position. We did not observe evidence to support the formation of these reactive intermediates from meta- and para-dihalogenated anilines that were exposed to similar incubation conditions (Figure 1). The initial GSH addition to the ipso carbon is also supported by in silico calculations. On the basis of density functional theory calculations, the ipso position has the most positive electrostatic potential compared to that of the other carbons in the phenyl ring. Also the highest occupied molecular orbital (HOMO), and the lowest unoccupied orbital (LUMO) calculations are more Received: May 23, 2011 Published: September 09, 2011 1668
dx.doi.org/10.1021/tx2002228 | Chem. Res. Toxicol. 2011, 24, 1668–1677
Chemical Research in Toxicology
Figure 1. Comparison between 2,4-dihaloaniline and 3,4-dihaloaniline for the observation of dehalgenated GSH conjugation (M-X+SG). Y can be any halogen but fluorine, and X can be any halogen (iodine, bromine, chlorine, or fluorine).
favorable for the 2-fluoro-4-haloaniline reactions with GSH compared to those of the 3-fluoro-4-haloaniline. Our findings suggest that the type and position of the halogen in this series of dihalogenated anilines play a key role in the formation of the QX and the subsequent conjugate.
2. MATERIALS AND METHODS 2.1. Materials. 4-Bromo-2-fluoroaniline (2F4Br-A), 4-bromo-2iodoaniline (2I4Br-A), 4-chloro-2-fluoroaniline (2F4Cl-A), 4-chloro-2iodoaniline (2I4Cl-A), 3-fluoroaniline (3F-A), 4-fluoroaniline (4F-A), 2-fluoro-4-iodoaniline (2F4I-A), 3-fluoro-4-iodoaniline (3F4I-A), glutathione (GSH), 2-iodoaniline (2I-A), 3-iodoaniline (3I-A), reduced β-nicotinamide adenine dinucleotide phosphate (NADPH), and titanium(III) chloride solution (TiCl3; 12% in hydrochloric acid) were purchased from Sigma-Aldrich (St. Louis, MO). 4-Bromo-3-fluoroaniline (3F4BrA) was purchased from TCI America (Portland, OR). 2-Chloro-4iodoaniline (2Cl4I-A) was purchased from Ryan Scientific, Inc. (Mt. Pleasant, SC). 4-Chloro-3-fluoroaniline (3F4Cl-A) and 4-fluoro-2-iodoaniline (2I4F-A) were purchased from Oakwood Products, Inc. (West Columbia, SC). 2,4-Difluoroaniline (2F4F-A), 2-fluoroaniline (2F-A), and 4-iodoaniline (4I-A) were purchased from Acros Organics (Geel, Belgium). Stable labeled GSH (13C and 15N on glycine) was purchased from Cambridge Isotope Laboratories, Inc. (Andover, MA). Potassium phosphate buffer (pH 7.4) was prepared at Genentech, Inc. All other solvents were of high-performance liquid chromatography (HPLC) grade. Pooled human liver microsomes (lot HMMC-PL020) were purchased from CellzDirect (Durham, NC). 2.2. In Vitro Incubations and Sample Preparation. Human liver microsomes (1 mg/mL) in potassium phosphate buffer (100 mM; pH 7.4) were spiked with test compound (20 μM) and unlabeled and labeled GSH (1 mM; 50:50), with a final incubation volume of 100 μL and a solvent concentration of 0.2% DMSO. This mixture was preincubated for 5 min at 37 °C, and the reaction was initiated by the addition of NADPH (1 mM). The reaction was quenched after 60 min by the addition of acetonitrile (250 μL) containing formic acid (3%). After centrifugation at 2000g for 10 min, the supernatant was removed and evaporated to dryness on an Evaporex EVX-192 from Apricot Designs (Monrovia, CA) and reconstituted in a solution of methanol and water (1:1 v/v; 200 μL). In the case of three monofluoroanilines, we also incubated the substrate at 100 μM with the incubation volumes of 1 mL. 2.3. Sample Analysis. The samples were analyzed using an Accela UHPLC coupled to an LTQ mass spectrometer from Thermo Fisher Scientific (San Jose, CA). Two types of analytical columns were used: Hypersil Gold C18 Column (1.9 μm, 2.1 100 mm; Thermo Scientific) for 2F4I-A only using a 15 min LC gradient and Hypersil Gold C18 Column (2.1 50 mm) for all other compounds with a 8 min LC gradient.
ARTICLE
The liquid chromatograph (LC) solvent system consisted of mobile phase A (HPLC grade water containing 0.1% formic acid) and mobile phase B (acetonitrile containing 0.1% formic acid). For the metabolites of 2F4I-A, it required a 15 min LC gradient and a longer column to improve the chromatographic peak separation and detection. The LC gradient was a stepwise linear gradient of 5% B for 1 min, followed by an increase to 70% B in 9 min, then another increase to 95% B in 1 min, and kept at 95% B for 2 min, and then equilibrated for 2 min. The flow rate was 0.4 mL/min, and the total run time was 15 min. For all other compounds, the chromatographic separation was achieved by a stepwise linear gradient of 5% B for 0.5 min, followed by an increase to 70% B in 4.5 min, then another increase to 95% B in 0.5 min, and was kept at 95% B for 1 min, and then equilibrated for 1.5 min. The flow rate was 0.4 mL/min, and the total run time was 8 min. Samples were introduced at a flow rate of 300 nL/min into an LTQ mass spectrometer managed by TriVersa NanoMate ChipSoft software from Advion BioSystems Inc. (Ithaca, NY) with a spray voltage of 1.75 kV. The remaining flow was either fraction collected into 96-well plates for further structural identification or diverted to waste. Sample ionization was achieved by nanospray ionization in the positive ion mode (both full scan and MS/MS mode) to achieve optimum sensitivity for GSH conjugates. The ionization parameters were as follows: spray voltage of 2 kV, tube lens at 85 V, and capillary temperature of 150 °C. The maximum injection time was set to 300 ms with 3 μscans and averaging up to 100 scans. Full MS scans over a mass range of m/z 370 to 950 followed by MS2 and MS3 scans with a collision energy of 17 kV were used. MS2 was performed on the most intense peaks that had a mass difference of 3 amu (the delta mass of both high and low masses of the labeled and unlabeled GSH) with a relative ratio of 80100%. MS3 was triggered for the five most intense peaks that had a neutral loss of 129 amu after GSH fragmentation. 2.4. MC4 Isolation. 2F4I-A (40 μM) was incubated with human liver microsomes fortified with NADPH and GSH in a flask. The incubations were quenched by the addition of 3 volumes of cold acetonitrile. The quenched incubations were vortex-mixed for 5 min prior to centrifugation at 2690g for 10 min. The supernatants were transferred to individual glass tubes and dried down at room temperature with Labconco Centrivap DNA Concentrator (Kansas City, MS). The isolation was conducted using HPLC with a Synergy Hydro RP column (150 2.0 mm, 3.0 μm) from Phenomenex (Torrance, CA). The LC solvent system was the same as that described above. Mobile phase B was delivered initially at 5%, held for 1 min, and increased to 95% via a 25 min gradient, then decreased back to 5% at 26 min, at which time it was retained for 4 min at a flow rate of 400 μL/min with a total run time of 30 min. 2.5. NMR Methods. NMR was performed on a Bruker Avance 3, 600 MHz spectrometer equipped with a 5 mm, TCI, Z gradient CryoProbe. The sample was prepared by dissolving ∼50 μg into 0.2 mL of D2O (D, 100%, Cambridge Isotope) containing 0.05% v/v TSP as an internal chemical shift reference standard. The sample was transferred to a 3 mm NMR tube (Norell S-3-600-7), purged with nitrogen, and sealed. The sample temperature was maintained at 28 °C for all data collections, and spectra were acquired and processed using Bruker TopSpin software, version 3.0, and patch level 3. 1D 1H NMR spectra were acquired with 2000 acquisitions, 65,536 complex data points, and a spectral width of 12 ppm. An additional relaxation delay of 1.0 s was added between acquisitions to allow for T1 relaxation. The standard Bruker pulse sequences, zg30 and zggpw5, were used. 1D selective 1H/1H NOE data were collected using the standard Bruker pulse sequence selnogp. The NOE mixing time was set to 300 ms, and a selective 180° Gaussian shaped pulse, 80 ms in length, was used to selectively invert a resonance of interest; 20,000 acquisitions were acquired for each measurement. 1669
dx.doi.org/10.1021/tx2002228 |Chem. Res. Toxicol. 2011, 24, 1668–1677
Chemical Research in Toxicology
ARTICLE
Table 1. GSH Conjugates Formed and the Corresponding Major Fragment Ions Observed by LC-MS/MS after Incubation of Various 2-Fluoro-4-haloanilines with Human Liver Microsomes in the Presence of NADPH and GSHa
X Ic (2F4I-A)
Br (2F4Br-A) Cl (2F4Cl-A) F (2F4F-A)
modification
RT (min)
major fragment ions, m/z (fragment ions from +3 amu parentb)
[M + H]+ (m/z)
+SG (MC1)
6.6c
543
416 (419), 287 (290)
+SG+O (MC2a)
5.4
559
541 (544), 484 (484), 432 (435), 303 (306)
+SG+O (MC2b)
5.6
559
541 (544), 322 (325), 303 (306), 256 (259), 247 (247), 193 (196)
+SG+O-F (MC3a)
4.3
541
523 (526), 466 (466), 414 (417), 285 (288)
+SG+O-F (MC3b)
5.0
541
523 (526), 466 (466), 412 (415), 285 (288)
+SG-F (MC4) +SG+O-F
5.5 2.8
525 493
507 (510), 450 (450), 396 (399), 379 (382), 321 (321), 269 (272), 250 (250) 475 (478), 418 (418), 364 (367), 347 (350), 289 (289)
+SG-F
3.4
477
459 (462), 402 (402), 348 (351), 331 (334), 273 (273), 228 (228), 202 (202)
+SG-F
3.0
433
415 (418), 358 (358), 304 (307), 287 (290)
+SG+O-F-NH2
4.4
434
416 (419), 305 (308), 287 (290), 212 (212)
ND
a
2F4I-A = 2-fluoro-4-iodoaniline; 2F4Br-A = 4-bromo-2-fluoroaniline; 2F4Cl-A = 4-chloro-2-fluoroaniline; 2F4F-A = 2,4-difluoroaniline; [M + H]+ = protonated molecular ion; ND = not detected; RT = retention time. b Molecular ion of the conjugate formed from stable isotope-labeled GSH (γ-glutamylcysteinylglycine-13C2-15N) that has a 3 amu higher mass than the conjugate formed from naturally abundant GSH. c 2F4I-A was run under a 15 min LC gradient. All the other compounds were run under an 8 min LC gradient. 2D 1H/13C, multiplicity edited, HSQC data were collected using the standard Bruker pulse sequence, hsqcedetgpsisp2.3. The data were acquired with 128 acquisitions per t1 increment, 128 complex data points in t1, and 1,024 complex data points in t2. Composite pulse, GARP, 13C decoupling was used during the data acquisition period (0.09 s). The 13C spectral width was set to 180 ppm. 2D DQF 1H/1H COSY data were acquired with 128 acquisitions for each t1 increment (256 total) and 2,048 complex data points in t2. The t2 acquisition time was 0.179 s, and a 2 s T1 relaxation delay was employed. The standard Bruker pulse sequence, cosygpmfqf, was used. 2.6. In Silico Modeling. Geometries of the anilines utilized in this study were fully optimized by using density functional theory (DFT) with Becke’s three-parameter hybrid exchange function and the LeeYangParr correlation function (B3LYP) in combination with the 6-31+G(d) basis set (for dihaloanilines) or the Los Alamos National Laboratory 2-double-z (LANL2DZ) basis set (for the iodine-containing fluoroanilines). Vibrational frequency calculations were carried out to confirm the optimized stable molecular structures. All of the electronic structure calculations in the gas phase were performed by using the Gaussian 03 program.16
3. RESULTS Halogenated anilines were incubated with human liver microsomes (1 mg/mL) and fortified with NADPH (1 mM) and GSH (1 mM) for 60 min. The GSH was added at an equal molar ratio of natural isotopic abundance GSH mixed with stable isotopelabeled GSH with labeled glycine (γ-glutamylcysteinylglycine13 C2-15N) as described by Yan and Caldwell.17 The labeled GSH was used as a diagnostic tool for the presence of GSH conjugates formed during the incubations, in which the labeled glycine in GSH was 3 amu higher than that in natural glycine. Neutral loss of 129 amu in the positive ion mode of mass spectrometry was typically characterized two equally abundant molecular ions of
GSH conjugates with a 3 amu difference. To confirm GSH conjugation, the product ion scans of both the molecular ion and the one at 3 amu higher were acquired. In the following sections, we discuss the GSH conjugates that were formed from the various halogenated anilines. 3.1. 2-Fluoro-4-haloanilines. To evaluate both the oxidative metabolism and the formation of GSH conjugates of 2F4I-A, the compound was incubated with human liver microsomes in the presence and absence of NADPH and/or GSH. When the reaction was fortified with NADPH alone (no GSH), two oxidative metabolites were detected. One (M1) was an oxidative metabolite with the addition of an oxygen atom (M1; [M + H]+ = m/z 254), and another (M2) was an oxidative defluorinated metabolite ([M + H]+ = m/z 236) that represented oxidation at the benzene ring followed by loss of HF to form a phenol. The percentage remaining of 2F4I-A in the incubation fortified with NADPH and GSH for one hour was 40.2% (see Supporting Information (Figure S1)). There were no other dioxidative or multiple GSH conjugates detected. When the reaction was fortified with NADPH and GSH, six GSH conjugates plus the two oxidative metabolites were detected (Table 1; Figure S1, Supporting Information). These conjugates were formed in a time- and NADPH-dependent manner. 3.1.1. Characterizing MC1 to MC3. GSH conjugate MC1 had a molecular ion of m/z 543 (+SG; m/z 546 for the stably labeled GSH with equal abundance) and a retention time of 6.6 min (Table 1). The two major fragment ions were m/z 416 and 287 (Figure S2, Supporting Information). The fragment ion m/z 416 was formed as a result of the loss of the iodine radical, which is consistent with previous reports.18 Further neutral loss of a glutamic acid side chain (129 amu) from m/z 416 resulted in the formation of the fragment ion m/z 287. MC2a and MC2b were oxidative products plus GSH conjugation (+SG+O) with a 1670
dx.doi.org/10.1021/tx2002228 |Chem. Res. Toxicol. 2011, 24, 1668–1677
Chemical Research in Toxicology
Figure 2. Proposed fragmentation pattern of the GSH conjugate MC4 of 2-fluoro-4-iodoaniline (2F4I-A) formed after incubation in human liver microsomes supplemented with NADPH and GSH. The molecular ion of the conjugate formed from stable isotope-labeled GSH (γglutamylcysteinylglycine-13C2-15N) is in parentheses.
molecular ion of m/z 559 and a retention time of 5.4 and 5.6 min, respectively. The MC2a fragment ions were formed as a result of loss of water (m/z 541), glycine (m/z 484), iodine radical (m/z 432), or glutamic acid plus iodine radical (m/z 303). MC2b shared common fragments at m/z 541 and 303 plus additional fragment ions at m/z 322, 247, and 192. The fragment ion at m/z 322 was the oxidized GSH conjugate. This ion could further lose neutral glycine and glutamic acid to form m/z 247 and m/z 193, respectively. MC3a and MC3b had a molecular ion of m/z 541 with a retention time of 4.3 and 5.0 min, respectively. This was consistent with an oxidized defluorinated GSH conjugate (+SG +O-F). The fragment ions were formed as a result of the loss of a water molecule (m/z 523), glycine (m/z 466), iodine radical (m/ z 414), glutamic acid, and iodine radical (m/z 285) for both conjugates. One difference between fragment ions of MC3a and MC3b was the loss of iodine radical for MC3a to form m/z 414, and for MC3b, it was the loss of glutamic acid to form m/z 412. 3.1.2. Characterizing MC4. MC4 had a molecular ion of m/z 525 (+SG-F) with a retention time of 5.5 min. This ion was 288 amu higher than the parent compound, and it was indicative of the addition of one GSH molecule (307 amu) and loss of fluorine (19 amu). The product ion scan of MC4 at m/z 525 generated fragment ions at m/z 507, 450, 396, 379, 321, 269, and 250 (Figure 2). The fragment ions at m/z 507, 450, and 396 were formed by the neutral loss of water (18 amu), glycine (75 amu), and glutamic acid (129 amu) from this GSH conjugate. The fragment ions at m/z 379, 321, and 269 were formed by the loss of ammonia (17 amu), glycine (75 amu), and an iodine radical (127 amu) from m/z 396, respectively. The fragment ion at m/z 250 was formed by the loss of the majority of the GSH part of the molecule, but the thiol was retained on the aromatic aniline ring. The fragment ions for the stably labeled GSH conjugate of MC4 at m/z 528 confirmed the identity of the fragment ions at m/z 510, 450, 399, 382, 321, 272, and 250. NMR was used to further characterize MC4. MC4 was isolated by HPLC. In 1H NMR, three proton resonances were observed in the aromatic region (Figure 3 and Figure S7a, Supporting
ARTICLE
Figure 3. 1H-13C HSQC data shows that these protons are attached to carbons at 144.55 ppm, 138.36 ppm, and 117.36 ppm, respectively. The proton assignments are based on the coupling constants and the proton and carbon-13 chemical shift values. If the GSH moiety was attached to C-6, then the singlet proton would be attached to a carbon at ∼117 ppm instead of the observed value of 142.55 ppm.
Information); a broad singlet at 7.73 ppm, a doublet at 7.50 ppm (8.64 Hz splitting), and a doublet at 6.70 ppm (8.64 Hz splitting). The 1H13C HSQC data demonstrated that these protons were attached to carbons at 144.55 ppm, 138.36 ppm, and 117.36 ppm, respectively. The proton assignments were based on the coupling constants and the proton and carbon-13 chemical shift values. If the GSH moiety was attached to C-6 (meta position instead of ortho), then the singlet proton would be attached to a carbon at ∼117 ppm instead of the observed value of 142.55 ppm. Additionally, a NOE was observed between H-6 and the methylene protons from GSH next to the thiol (C-8 in Figure S7b, Supporting Information). Thus, for the MC4 GSH conjugate, the location of the thiol attached to the phenyl ring was at the ortho position, where the fluorine atom was originally located. In order to understand if an N-oxide was formed or if oxidation occurred on the phenyl ring, the samples were treated with TiCl3, but there were no changes observed in the abundance of conjugates except for MC2b, confirming the absence of N-oxide metabolites. The formation of GSH conjugates from other 2-fluoro-4haloaniline analogues was also examined using human liver microsomes in the presence of NADPH and GSH (Table 1). In cases where an iodine was replaced with either bromo or chloro analogues, defluorinated GSH conjugates were observed, similarly to the conjugates formed from 2F4I-A. Metabolism of 2F4Br-A led to the formation of two GSH conjugates. One GSH conjugate had a molecular ion of m/z 493 and a retention time of 2.8 min that corresponded to the addition of GSH and oxygen, as well as a loss of fluorine (+SG+O-F). The other GSH conjugate had a molecular ion of m/z 477 and a retention time of 3.4 min that corresponded to the defluorinated GSH conjugate (+SG-F). Two GSH conjugates were formed from 2F4Cl-A as well. One conjugate had a molecular ion of m/z 433 and a retention 1671
dx.doi.org/10.1021/tx2002228 |Chem. Res. Toxicol. 2011, 24, 1668–1677
Chemical Research in Toxicology
ARTICLE
Table 2. Presented Are Energies for the Unoccupied Molecular Orbital (LUMO) and Electrostatic Potential Energy for the Fluoro-haloaniline Seriesa electrostatic potential energy (eV) compd
LUMO energy (kcal/mol)
C1
C2
C3
C4
C5
C6
2-Fluoro-4-haloaniline Series 2F4F-A
4.78
0.148
0.275
0.423
0.363
0.335
0.172
2F4Cl-A 2F4Br-A
6.85 6.82
0.242 0.254
0.175 0.175
0.191 0.125
0.010 0.144
0.096 0.013
0.278 0.292
2F4I-A
20.08
0.327
0.087
0.008
0.285
0.108
0.377
3F4F-A
101.62
0.362
0.288
0.224
0.174
0.231
0.397
3F4Cl-A
103.35
0.452
0.387
0.012
0.184
0.448
0.498
3F4Br-A
106.31
0.662
0.381
0.127
0.457
0.643
0.645
3F4I-A
18.88
0.529
0.484
0.225
0.487
0.633
0.581
4F2I-A
14.57
0.692
0.045
0.168
0.155
0.315
3-Fluoro-4-haloaniline Series
4-Fluoro-2-iodoaniline 0.451
a
The most positively charged carbon is highlighted in bold. Note that the electrostatic potential (ESP) energy for the GSH thiol was calculated as 0.323 eV, and the GSH LUMO energy was 17.57 kcal/mol.
time of 3.0 min that corresponded to the defluorinated GSH conjugate (+SG-F). The other conjugate had a molecular ion of m/z 434 with a retention time of 4.4 min that corresponded to the addition of GSH and oxygen plus defluorination and deamination (+SG+O-F-NH2). This GSH conjugate was not detected with any other dihalogenated aniline. No detectable levels of GSH conjugates were observed for 2F4F-A. 3.1.3. In Silico Analysis of the 2-Fluoro-4-haloanilines. The electrostatic potential energy19 is the most positive for the ipso carbon compared to that of the other 5 carbon atoms in the phenyl ring (Table 2). For all of the 2-fluoro-4-haloaniline analogues with chloro, bromo, and iodo containing anilines, the electrostatic potential (ESP) ranges between 0.242 and 0.327 cV. In the case of 2F4F-A, the positive carbon is no longer the ipso position, but it is the C4 carbon (Table 2). This is also consistent with not observing the defluorinated GSH conjugate (M-F+GS) for this analog. In addition, based on the shape and the HOMO and LUMO energies of the 2-fluoro-4-haloanilines and GSH, the conjugation reaction of the thiol of GSH is predicted to occur via the interaction of its HOMO with the LUMO of the fluoroanilines. The LUMO of the 2F4F-A is approximately 2 kcal/mol higher than that of the 2F4Cl-A and 2F4Br-A, and approximately 15 kcal/mol higher than that of the 2F4I-A. Consequently, the conjugation of the thiol group to the ipso carbon of 2F4F-A is energetically less favorable than that for the other 2-fluoro-4haloanilines (Table 2). 3.2. Monohalogenated Anilines (2-, 3-, or 4-Flouroanilines and 2-, 3-, or 4-Iodoanilines). In order to further understand the mechanism that led to the formation of MC4 from 2F4I-A, three monofluoroaniline analogues and monoiodoaniline analogues were investigated. Monofluorinated anilines (2-fluoro-, 3-fluoro-, and 4-fluoroaniline) did not form any detectable levels of GSH conjugates under regular conditions, as described. When samples were concentrated 50-fold, this afforded the detection of several GSH conjugates. However, all defluoronated GSH conjugates contained the addition of oxygen (+O+SG-F). However, monoiodonated anilines (2-iodo-, 3-iodo-, and 4-iodoaniline)
formed a conjugate at m/z 541 that corresponded to the addition of GSH and oxygen (+SG+O) (Table 3). 3.3. 3-Fluoro-4-haloanilines. Three major GSH conjugates were formed from 3F4I-A, under the same conditions as those described above. However, no conjugate corresponding to the addition of GSH and loss of fluorine (M-X+SG) was detected. The observed GSH conjugates had molecular ions at m/z 541 (+SG+O-F), 543 (+SG), and 559 (+SG+O) (Table 4). The major fragment ions were consistent with the addition of GSH followed by a neutral loss of 129 amu (glutamic acid), 75 amu (glycine) or a combination of these ions. Unlike 2F4I-A, all of the defluorinated GSH conjugates formed by 3-fluoro-4-haloanilines included the addition of oxygen atoms. The in silico modeling was consistent with this observation (Table 2). Our modeling showed that the most positively charged carbon was the ipso position, similar to 2-fluoro-4haloaniline, except for 3F4I-A where the most positive carbon was the C5 position. The LUMO energy for these molecules were relatively higher compared to that of 2-fluoro-4-haloaniline analogues (on average 104 vs 6 kcal/mol), except for 3F4I-A with a LUMO energy of 19 kcal/mol. 3.4. Other 2,4-Dihaloanilines. When 2I4F-A was examined in human liver microsomes fortified with NADPH and GSH, four GSH conjugates were observed (Table 5). These conjugates were the result of the addition of oxygen and GSH, and loss or retention of fluorine. As with the 3-fluoro-4-haloanilines, 2I4F-A did not form any GSH defluorinated conjugates that did not include the addition of oxygen atoms. Two GSH conjugates with molecular ions at m/z 559 (+SG+O) and retention times of 2.6 and 3.0 min were the result of addition of GSH and oxygen. Another pair of GSH conjugates with molecular ions at m/z 541 (+SG+O-F) and retention times of 1.5 and 2.8 min were the result of addition of GSH and oxygen as well as the loss of fluorine. Our in silico predictions agreed with these findings by showing that the most positively charged carbon in the phenyl ring of 2I4F-A is the ipso carbon, but the LUMO energy is relatively high (E = 14.57 kcal/mol) compared to that of the other iodo-fluoroanilines (i.e., 3F4I-A and 2F4I-A), making deflourination of this molecule energetically unfavorable. 1672
dx.doi.org/10.1021/tx2002228 |Chem. Res. Toxicol. 2011, 24, 1668–1677
Chemical Research in Toxicology
ARTICLE
Table 3. GSH Conjugates Formed and the Corresponding Major Fragment Ions Observed by LC-MS/MS after Incubation of Monohalogenated Fluoro- and Iodoanilines with Human Liver Microsomes in the Presence of NADPH and GSHa RT (min)
major fragment ions, m/z (fragment ions from +3 amu parentb)
[M + H]+ (m/z)
compound
modification
2F-Ac
+SG+O-F
1.2
415
397 (400), 340 (340), 308 (311), 290 (293), 286 (289)
3F-Ac
+SG+O +SG+O
2.7 1.5
433 433
304 (307) 415 (418), 304 (307), 287 (290), 158 (158)
2.4
415 (418), 358 (358), 304 (307), 287 (290), 158 (158)
3.4 4F-Ac
415 (418), 358 (358), 304 (307), 287 (290), 179 (182), 158 (158)
+SG+O-F
1.2
415
+SG+O
1.3
433
3.3 2I-A 3I-A
+SG+O
1.4
+SG+O
2.5 2.8 2.3
+SG+O
358 (358), 308 (311), 304 (307), 287 (290), 158 (158) 415 (418), 358 (358), 304 (307), 287 (290), 229 (229)
541
523 (526), 466 (466), 414 (417), 285 (288)
541
523 (526), 466 (466), 412 (415), 285 (288) 523 (526), 466 (466), 412 (414), 285 (288)
2.7 2.8 4I-A
397 (400), 340 (340), 308 (311), 290 (293), 286 (289), 269 (272), 140 (140)
2.4
523 (526), 466 (466), 412 (415) 541
2.8
523 (526), 466 (466), 414 (417), 285 (288) 523 (526), 466 (466), 412 (415), 285 (288)
a
2F-A = 2-fluoroaniline; 3F-A = 3-fluoroaniline; 4F-A = 4-fluoroaniline; 2I-A = 2-iodoaniline; 3I-A = 3-iodoaniline; 4I-A = 4-iodoaniline; [M + H]+ = protonated molecular ion; ND = not detected; RT = retention time. b Molecular ion of the conjugate formed from stable isotope-labeled GSH (γglutamylcysteinylglycine-13C2-15N) that has a 3 amu higher mass than the conjugate formed from naturally abundant GSH. c GSH conjugates have been detected when the incubation concentration was increased from 20 μM to 100 μM and the incubation volume was increased from 100 μL to 1,000 μL (totally concentrated by 50-fold) with a longer gradient.
Table 4. GSH Conjugates Formed and Their Corresponding Major Fragment Ions Observed by LC-MS/MS after Incubation of Various 4-Halogenated 3-Fluoroanilines with Human Liver Microsomes in the Presence of NADPH and GSHa
X
modification
RT (min)
[M + H]+ (m/z)
major fragment ions, m/z (fragment ions from +3 amu parentb)
Cl (3F4Cl-A)
+SG+O
2.7
467
449 (452), 431 (434), 392 (392), 338 (341), 302 (305)
Br (3F4Br-A)
+SG+O-F
2.1
493
475 (478), 418 (418), 413 (416), 364 (367), 284 (287), 181 (181)
+SG+O
2.8
511
493 (496), 436 (436), 431 (434), 413 (416), 411 (414), 382 (385), 302 (305)
+SG+O-F +SG
2.3 2.5
541 543
523 (526), 414 (417), 285 (288) 525 (528), 416 (419), 287 (290)
+SG+O
3.0
559
541 (544), 432 (435), 303 (306)
I (3F4I-A)
a
3F4I-A = 3-fluoro-4-iodoaniline; 3F4Br-A = 4-bromo-3-fluoroaniline; 3F4Cl-A = chloro-3-fluoroaniline; [M + H]+ = protonated molecular ion; RT = retention time. b Molecular ion of the conjugate formed from stable isotope-labeled GSH (γ-glutamylcysteinylglycine-13C2-15N) that has a 3 amu higher mass than the conjugate formed from naturally abundant GSH.
Several GSH conjugates were observed with 2I4Cl-A and 2I4Br-A. In addition to the conjugates formed by the addition of oxygen and GSH, as well as loss or retention of a halide, both compounds formed GSH conjugates that resulted from the addition of GSH and loss of halide at the 2 position without the addition of oxygen. The three conjugates of 2I4Cl-A had molecular ions at m/z 541 (+SG+O-Cl), 433 (+SG-I), and 449 (+SG+O-I). Four GSH conjugates were formed from 2I4Br-A. Two of the conjugates had molecular ions at m/z 541 (+SG+O-Br), and the other two were at m/z 477 (+SG-I) and 493 (+SG+O-I). Switching the positions of chlorine and iodine from 2I4Cl-A to 2Cl4I-A not only leads to the formation of a conjugate with
a molecular ion at m/z 541 (+SG+O-Cl) but also to the formation of a conjugate with a molecular ion at m/z 525 (+SG-Cl). This second conjugate was the result of addition of GSH and loss of chlorine at the 2 position without addition of oxygen atoms.
4. DISCUSSION Halogenated anilines are known to cause toxicity.7,11,20,21 The mechanism of toxicity for many of these compounds involves bioactivation by cytochrome P450.22 This bioactivation results in the formation of reactive metabolites that, in many cases, can be trapped by GSH. The three well-known reactive intermediates 1673
dx.doi.org/10.1021/tx2002228 |Chem. Res. Toxicol. 2011, 24, 1668–1677
Chemical Research in Toxicology
ARTICLE
Table 5. GSH Conjugates Formed and Their Corresponding Major Fragment Ions Observed by LC/MS/MS after Incubation of Various 2-Iodoaniline or 2-Chloroanilines with Human Liver Microsomes in the Presence of NADPH and GSHa
Y I
I
Cl
Br
(2I4Br-A)
Cl (2Cl4I-A)
+SG+O
2.6
559
541 (544), 484 (484), 430 (433)
+SG+O
3.0
559
484 (484), 430 (433), 303 (306), 283 (286)
+SG+O-F
1.5
541
523 (526), 466 (466), 412 (415), 394 (397), 285 (288), 266 (266)
+SG+O-F +SG+O-Cl
2.8 2.8
541 541
523 (526), 466 (466), 414 (417), 285 (288) 523 (526), 412 (415), 266 (266)
+SG-I
2.9
433
415 (418), 304 (307), 287 (290), 286 (289)
+SG+O-I
2.5
449
431 (434), 320 (323)
+SG+O-Br
1.5
541
523 (526), 466 (466), 412 (415), 395 (398), 285 (288)
+SG+O-Br
2.8
541
523 (526), 412 (415), 266 (266)
+SG-I
2.9
477
459 (462), 402 (402), 348 (351), 331 (334), 228 (228)
+SG+O-I
2.6
493
475 (478), 418 (418), 364 (367), 347 (350)
+SG+O-Cl +SG-Cl
2.8 3.1
541 525
523 (526), 412 (415) 507 (510), 396 (399)
F
(2I4Cl-A) I
[M + H]+ (m/z)
modification
(2I4F-A)
I
major fragment ions m/z (fragment ions from +3 amu parentb)
RT (min)
X
a 2Cl4I-A = 2-chloro-4-iodoaniline; 2I4Br-A = 4-bromo-2-iodoaniline; 2I4Cl-A = 4-chloro-2-iodoaniline; 2I4F-A = 2-fluoro-4-iodoaniline; [M + H]+ = protonated molecular ion; RT = retention time. b Molecular ion of the conjugate formed from stable isotope-labeled GSH (γglutamylcysteinylglycine-13C2-15N) that has a 3 amu higher mass than the conjugate formed from naturally abundant GSH.
that can result in the formation of covalent binding or formation of GSH conjugates for anilines are N-hydroxylamines, epoxides, and QX.23 For the series of dihalogenated anilines in this article, we discovered an intriguing trend in the formation of a dehalogenated GSH conjugate (M-X+SG) in human liver microsomes. We observed that, in general, this conjugate formed only when one halogen was in the ortho-position, and the other was in the para-position of the aniline. Our findings indicated that the orthohalogen could be any halogen, yet the para-halogen could not be a fluorine atom. We propose that the reactive intermediate formed from the ortho- and para-dihalogenated anilines tested was QX with a positive charge retained on the halogen in the para-position. 2F4I-A was used as a model to determine the overall metabolism of this series and all the possible GSH conjugates that can form. Four types of GSH conjugates were formed when 2F4I-A was incubated in human liver microsomes supplemented with NADPH and GSH. Five of these conjugates were the expected conjugates (MC1, MC2a, MC2b, MC3a, and MC3b), and one conjugate was formed as a result of defluorination plus GSH conjugation (MC4 or M-X+SG) (Figure S1, Supporting Information). The MC1-MC3 GSH conjugates were formed by GSH with or without the addition of oxygen and defluorination, but in the case of MC4, defluorination and GSH conjugation occurred without the addition of oxygen. This finding is significant because 3F4I-A does not form such a conjugate. To further evaluate this phenomenon, we conducted a structuremetabolism relationship analysis with an expanded list of halogenated anilines and analyzed the samples for the corresponding M-X+SG conjugates. Interestingly, we observed a consistent pattern in which only
anilines that were ortho- and para-halogenated (except for para fluorine) formed this type of dehalogenated GSH conjugate (Figure 1). We propose a mechanism that is consistent with the observations that involve the formation of QX followed by ipso substitution of GSH at the ortho-position through rearrangement that leads to the formation of M-X+SG. To examine which reactive intermediate was responsible for the formation of this conjugate, we considered the formation of N-oxide (N-hydroxylamine) and epoxide intermediates. To test for the formation of N-hydroxylamines, the samples were treated with TiCl3, but the amount of GSH conjugates was not reduced, suggesting that the GSH conjugates were not N-hydroxylamines. The only exception was MC2b. Also, if the reactive intermediate was formed via the formation of a hydroxylamine, we would observe a hydroxylamine conjugate consistent with M-X+SG. The absence of this conjugate suggests that N-hydroxylamine does not contribute to the formation of M-X+SG. If an epoxide intermediate was involved, the epoxide could form anywhere on the benzene ring where hydrogen atoms reside. In Figure 4, we depict two epoxide intermediates Ia and Ib (Mechanism I). These epoxides could react with a GSH molecule (a1 and b1 pathways) or a water molecule (a2 and b2 pathways) to open the epoxide ring. The GSH conjugates (a1 and a2 pathways) would be formed with or without phenol or halogens depending on the intermediate and the neutral loss of water or HX. The products of these GSH reactions are shown as A1, A2, B1, and B2. In the case of a water molecule opening the epoxide ring, a phenol is formed (A3, A4, B3, and B4). These phenol metabolites could be further oxidized and react with GSH to form the AB oxidative GSH conjugate that may retain the halogen. These 1674
dx.doi.org/10.1021/tx2002228 |Chem. Res. Toxicol. 2011, 24, 1668–1677
Chemical Research in Toxicology
ARTICLE
Figure 4. Proposed mechanisms for reactive metabolite formation from 2,4-dihaloaniline in human liver microsomes supplemented with NADPH and GSH. In these mechanisms, the epoxides (Ia and Ib) that are formed can react with either water or GSH to form A1A4 or B1B4. A3, A4, B3, and B4 can be further metabolized to form the AB conjugates.
Figure 5. Proposed mechanisms for reactive metabolite formation from 2,4-dihaloaniline in human liver microsomes supplemented with NADPH and GSH. In this mechanism, a quinone-type intermediate is formed and can react at the ortho and meta positions (pathway c) or at the ipso position (pathway d).
mechanisms explain most of the reported conjugates such as MC1-MC3 (Table 1). However, they do not explain the formation of M-X+SG because once the ring is epoxidated, the final product could not be the result of a loss of both a water molecule and an HX. Also, in the proposed mechanisms, the distinguishing differences between ortho- and meta-substituted halogens cannot be accounted for. In the formation of QX, the GSH conjugate could be formed via 1,4-addition (Michael-type reaction; Figure 5). One potential mechanism is for GSH to react directly at the ortho-position (pathway c). This would result in a tetrahedral intermediate (not shown), followed by loss of HX to form C2 (M-X+SG). However, a direct reaction of GSH can occur at any position to result in the formation of C1 (ortho- and meta-positions), so this mechanism does not explain why the X+SG reaction does
not occur when the halogen is at the meta-position. The mechanism we propose (Mechanism II) is one in which GSH attacks the QX at the ipso position (pathway d), followed by migration of GSH through an episulfonium ion intermediate to the ortho-position and loss of HX.24 This mechanism accounts for the formation of the unusual conjugate (M-X+SG) when a halogen is in the ortho-position. What distinguishes the two mechanisms is that dehalogenation via pathway d can take place only when a halogen atom is at the ortho-position and not at the meta-position, which is consistent with the results reported here. This trend is present even when the steric factors are considered with 2I4Cl-A to 2Cl4I-A. In both cases, the replacement of the halogen occurs at the ortho-position. In silico methodologies also support this mechanism. The most positive ESP for 2-fluoro-4-haloaniline (except for 2F4F-A) is the 1675
dx.doi.org/10.1021/tx2002228 |Chem. Res. Toxicol. 2011, 24, 1668–1677
Chemical Research in Toxicology ipso carbon. This fact plus the favorable LUMO energy for this series, makes the GSH thiol nucleophilic reaction at the ipso position a reasonable initial step. This in contrast to the 3-fluoro-4-haloaniline, where the ipso position is still the most positive ESP, but the LUMO energy is no longer compatible for 3F4F-A, 3F4Cl-A, and 3F4Br-A, with the LUMO energy range between 101 to 106 kcal/mol. In addition, for the 3-fluoro-4haloaniline series, if the GSH attacks the ipso position there is no halogen in the neighboring ortho position for the dehalogenation reaction. In the case of 3F4I-A, the most positive carbon is C5 with reasonable LUMO energy (19 kcal/mol). Interestingly in the latter case, this is the only analogue that forms just the addition of GSH (+SG at [M + H]+ = m/z 543), which is consistent with what is predicted from in silico methodologies. Another aspect in the mechanism (Mechanism II, pathway d) for the formation of the QX is delocalization of the positive charge on the para-halogen after oxidation. In other words, in order for the QX to form, stabilization of the cationic sigmacomplex is required by the para-substituted halogen. This hypothesis is consistent with the observation that 2X4F-A does not form the M-X+SG conjugate because the chemical stability of the QX is dependent on the ability of the parahalogen to stabilize the intermediate by resonance,25 and in this case fluorine, the most electronegative atom among the halogens, would strongly destabilize the cationic sigma-complex; therefore, the QX does not form. Several reports are consistent with the mechanism we report here. For example, the ultimate toxic metabolite of acetaminophen is N-acetyl-p-benzoquinone imine (NAPQI); however, because of the short half-life of NAPQI, only the thioether conjugates are observed in most cases.26,27 When ortho-dimethyl acetaminophen was examined, its toxicity was diminished compared to that of meta-dimethyl acetaminophen. The researchers speculated that ipso addition of GSH to the ortho-substituted compound did not form a stable conjugate, and therefore, it reverted back to the parent compound,28 Another reported dehalogenation with a GSH conjugate that is consistent with our proposed mechanism is the novel GSH conjugate of diclofenac that forms in human liver microsomes.29,30 The authors proposed that GSH reacts at the quinone imine of the carbon that contains a chlorine atom, similar to our observation that only dehalogenated GSH conjugates are formed via this mechanism. The last example is the GSH conjugation of 2-fluoro-4-hydroxyaniline that resulted in the loss of a fluoro group at the orthoposition and displacement by GSH.31 In conclusion, our results suggest that displacement of the halogen with GSH in dihalogenated anilines takes place only when one of the halogens is at the ortho-position, and the other halogen is at the para-position (the para-halogen can be any halogen except fluorine). On the basis of our observations and those reported in the literature, we propose that reactive intermediate QX was formed by initial oxidation, GSH reacted at the ipso position, followed by rearrangement of GSH to the orthoposition through an episulfonium intermediate and loss of HX to form the final monohalogenated conjugate (Mechanism II, pathway d).
’ ASSOCIATED CONTENT
bS
Supporting Information. Characterization of metabolites and in silico calculations. This material is available free of charge via the Internet at http://pubs.acs.org.
ARTICLE
’ AUTHOR INFORMATION Corresponding Author
*Tel: 650-225-6094. Fax: 650-467-3487. E-mail: khojasteh.cyrus@ gene.com.
’ ACKNOWLEDGMENT We thank Ronitte Libedinsky for her editorial help. ’ ABBREVIATIONS COSY, homonuclear correlation spectroscopy; 2F4I-A, 2-fluoro4-iodoaniline (note that the halogenated anilines are abbreviated by using the halogen position first plus halogen followed by a dash A for aniline); ESP, electrostatic potential; HOMO, highest occupied molecular orbital; HSQC, heteronuclear single-quantum correlation spectroscopy; LUMO, lowest unoccupied orbital; NOE, nuclear Overhauser effect; QX, (4-iminocyclohexa2,5-dienylidene)halogenium reactive intermediate. ’ REFERENCES (1) Hartwig, J. F., Shekhar, S., Shen, Q. Barrios-Landeros, F. (2007) Chemistry of Anilines, John Wiley & Sons, Inc., West Sussex, England. (2) Anonymous. (1993) 2,6-Dimethylaniline (2,6-xylidine): Occupational Exposures of Hairdressers and Barbers and Personal Use of Hair Colourants; Some Hair Dyes, Cosmetic Colourants, Industrial Dyestuffs and Aromatic Amines, IARC Monographs on the Evaluation of Carcinogenic Risks to Humans, pp 323335, World Health Organization: International Agency for Research on Cancer (3) Chhabra, R. S., Thompson, M., Elwell, M. R., and Gerken, D. K. (1990) Toxicity of p-chloroaniline in rats and mice. Food Chem. Toxicol. 28 (10), 717–722. (4) Guilhermino, L., Soares, A. M., Carvalho, A. P., and Lopes, M. C. (1998) Acute effects of 3,4-dichloroaniline on blood of male Wistar rats. Chemosphere 37 (4), 619–632. (5) McLean, S., Starmer, G. A., and Thomas, J. (1969) Methaemoglobin formation by aromatic amines. J. Pharm. Pharmacol. 21 (7), 441–450. (6) Khan, M. F., Wu, X., Boor, P. J., and Ansari, G. A. (1999) Oxidative modification of lipids and proteins in aniline-induced splenic toxicity. Toxicol. Sci. 48 (1), 134–140. (7) Valentovic, M. A., Ball, J. G., Anestis, D. K., Beers, K. W., Madan, E., Hubbard, J. L., and Rankin, G. O. (1992) Acute renal and hepatic toxicity of 2-haloanilines in Fischer 344 rats. Toxicology 75 (2), 121–131. (8) Rankin, G. O., Yang, D. J., Cressey-Veneziano, K., Casto, S., Wang, R. T., and Brown, P. I. (1986) In vivo and in vitro nephrotoxicity of aniline and its monochlorophenyl derivatives in the Fischer 344 rat. Toxicology 38 (3), 269–283. (9) Rankin, G. O., Yang, D. J., Teets, V. J., Lo, H. H., and Brown, P. I. (1986) 3,5-Dichloroaniline-induced nephrotoxicity in the SpragueDawley rat. Toxicol. Lett. 30 (2), 173–179. (10) Hong, S. K., Anestis, D. K., Henderson, T. T., and Rankin, G. O. (2000) Haloaniline-induced in vitro nephrotoxicity: effects of 4-haloanilines and 3,5-dihaloanilines. Toxicol. Lett. 114 (13), 125–133. (11) Cnubben, N. H., Soffers, E. M., Peters, M. A., Vervoort, J., and Rietjens, I. M. (1996) Influence of the halogen-substituent pattern of fluoronitrobenzenes on their biotransformation and capacity to induce methemoglobinemia. Toxicol. Appl. Pharmacol. 139 (1), 71–83. (12) Nendza, M., and Seydel, J. K. (1987) Quantitative structuretoxicity relationships and multivariate data analysis for ecotoxic chemicals in different biotest systems. Chemosphere 17 (8), 1575–1584. (13) Sabbioni, G. (1992) Hemoglobin binding of monocyclic aromatic amines: molecular dosimetry and quantitative structure activity relationships for the N-oxidation. Chem.-Biol. Interact. 81 (12), 91–117. (14) Eyer, P., and Hell, W. (1983) The metabolism of 3-benzoylpyridine. Xenobiotica 13 (11), 649–659. 1676
dx.doi.org/10.1021/tx2002228 |Chem. Res. Toxicol. 2011, 24, 1668–1677
Chemical Research in Toxicology (15) Rietjens, I. M., Tyrakowska, B., Veeger, C., and Vervoort, J. (1990) Reaction pathways for biodehalogenation of fluorinated anilines. Eur. J. Biochem. 194 (3), 945–954. (16) Frisch, M. J., Trucks, G. W., Schlegel, H. B., Scuseria, G. E., Robb, M. A., Cheeseman, J. R., Montgomery, J. A., Vreven, T., Kudin, N., Burant, J. C., Millam, J. M., Iyengar, S. S., Tomasi, J., Barone, V., Mennucci, B., Cossi, M., Scalmani, G., Rega, N., Petersson, G. A., Nakatsuji, H., Hada, M., Ehara, M., Toyota, K., Fukuda, R., Hasegawa, J., Ishida, M., Nakajima, T., Honda, Y., Kitao, O., Nakai, H., Klene, M., Li, X., Knox, J. E., Hratchian, H. P., Cross, J. B., Adamo, C., Jaramillo, J., Gomperts, R., Stratmann, R. E., Yazyev, O., Austin, A. J., Cammi, R., Pomelli, C., Ochterski, J. W., Ayala, P. Y., Morokuma, K., Voth, G. A., Salvador, P., Dannenberg, J. J., Zakrzewski, V. G., Dapprich, S., Daniels, A. D., Strain, M. C., Farkas, O., Malick, D. K., Rabuck, A. D., Raghavachari, K., Foresman, B., Ortiz, J. V., Cui, Q., Baboul, A. G., Clifford, S., Cioslowski, J., Stefanov, B. B., Liu, G., Liashenko, A., Piskorz, P., Komaromi, I., Martin, R. L., Fox, D. J., Keith, T., Al-Laham, M. A., Peng, C. Y., Nanayakkara, A., Challacombe, M., Gill, P. M. W., Johnson, B., Chen, W., Wong, M. W., Gonzalez, C., and Pople, J. A. (2004) Gaussian03, revision C.02, Gaussian Inc., Wallgford, CT. (17) Yan, Z., and Caldwell, G. W. (2004) Stable-isotope trapping and high-throughput screenings of reactive metabolites using the isotope MS signature. Anal. Chem. 76 (23), 6835–6847. (18) Duckett, C. J., Wilson, I. D., Douce, D. S., Walker, H. J., AbouShakra, F. R., Lindon, J. C., and Nicholson, J. K. (2007) Metabolism of 2-fluoro-4-iodoaniline in earthworm Eisenia veneta using (19)F-NMR spectroscopy, HPLC-MS, and HPLC-ICPMS ((127)I). Xenobiotica 37 (12), 1378–1393. (19) Chen, H., Shockcor, J., Chen, W., Espina, R., Gan, L. S., and Mutlib, A. E. (2002) Delineating novel metabolic pathways of DPC 963, a non-nucleoside reverse transcriptase inhibitor, in rats. Characterization of glutathione conjugates of postulated oxirene and benzoquinone imine intermediates by LC/MS and LC/NMR. Chem. Res. Toxicol. 15 (3), 388–399. (20) Cnubben, N. H., Vervoort, J., Boersma, M. G., and Rietjens, I. M. (1995) The effect of varying halogen substituent patterns on the cytochrome P450 catalysed dehalogenation of 4-halogenated anilines to 4-aminophenol metabolites. Biochem. Pharmacol. 49 (9), 1235–1248. (21) Cnubben, N. H., van den Berg, C. L., and Rietjens, I. M. (1996) Relationships between the regioselectivity of the hydroxylation of C4substituted 2-fluoroaniline derivatives and their toxic endpoints. Toxicol. Appl. Pharmacol. 141 (2), 403–415. (22) Rietjens, I. M., Cnubben, N. H., van Haandel, M., Tyrakowska, B., Soffers, A. E., and Vervoort, J. (1995) Different metabolic pathways of 2,5-difluoronitrobenzene and 2,5-difluoroaminobenzene compared to molecular orbital substrate characteristics. Chem.-Biol. Interact. 94 (1), 49–72. (23) Rietjens, I. M., den Besten, C., Hanzlik, R. P., and van Bladeren, P. J. (1997) Cytochrome P450-catalyzed oxidation of halobenzene derivatives. Chem. Res. Toxicol. 10 (6), 629–635. (24) Fan, P. W., Zhang, F., and Bolton, J. L. (2000) 4-Hydroxylated metabolites of the antiestrogens tamoxifen and toremifene are metabolized to unusually stable quinone methides. Chem. Res. Toxicol. 13 (1), 45–52. (25) Carey, F. A., and Sundberg, R. J. (1990) Advanced Organic Chemistry. Part A: Structure and Mechanisms, 3rd ed., Plenum Press, New York. (26) Gillette, J. R., Nelson, S. D., Mulder, G. J., Jollow, D. J., Mitchell, J. R., Pohl, L. R., and Hinson, J. A. (1981) Formation of chemically reactive metabolites of phenacetin and acetaminophen. Adv. Exp. Med. Biol. 136 (Pt B), 931–950. (27) Dahlin, D. C., and Nelson, S. D. (1982) Synthesis, decomposition kinetics, and preliminary toxicological studies of pure N-acetyl-pbenzoquinone imine, a proposed toxic metabolite of acetaminophen. J. Med. Chem. 25 (8), 885–886. (28) Rundgren, M., Porubek, D. J., Harvison, P. J., Cotgreave, I. A., Moldeus, P., and Nelson, S. D. (1988) Comparative cytotoxic effects of N-acetyl-p-benzoquinone imine and two dimethylated analogues. Mol. Pharmacol. 34 (4), 566–572.
ARTICLE
(29) Yan, Z., Li, J., Huebert, N., Caldwell, G. W., Du, Y., and Zhong, H. (2005) Detection of a novel reactive metabolite of diclofenac: evidence for CYP2C9-mediated bioactivation via arene oxides. Drug Metab. Dispos. 33 (6), 706–713. (30) Yu, L. J., Chen, Y., Deninno, M. P., O’Connell, T. N., and Hop, C. E. (2005) Identification of a novel glutathione adduct of diclofenac, 40 -hydroxy-20 -glutathion-deschloro-diclofenac, upon incubation with human liver microsomes. Drug Metab. Dispos. 33 (4), 484–488. (31) Rietjens, I. M., Tyrakowska, B., Veeger, C., and Vervoort, J. (1990) Reaction pathways for biodehalogenation of fluorinated anilines. Eur. J. Biochem. 194 (3), 945–954.
1677
dx.doi.org/10.1021/tx2002228 |Chem. Res. Toxicol. 2011, 24, 1668–1677
