1,3-Dinitrobenzene Metabolism and GSH Depletion - Chemical

Department of Environmental Toxicology, University of California at Davis, Davis, California ... Chemical Research in Toxicology 2003 16 (12), 1547-15...
0 downloads 0 Views 76KB Size
Download PDF
Chem. Res. Toxicol. 2002, 15, 361-366

361

1,3-Dinitrobenzene Metabolism and GSH Depletion Ian T. Reeve,* John C. Voss,† and Marion G. Miller‡ Department of Environmental Toxicology, University of California at Davis, Davis, California 95616-8588 Received August 12, 2001

Previous work demonstrated that the mitochondrial fraction of rat seminiferous tubules is capable of metabolizing 1,3-dinitrobenzene, using NADPH as a cofactor. Moreover, 1,3dinitrobenzene treatment of rat tubules caused a decrease in mitochondrial GSH levels. In situ mitochondrial metabolism of 1,3-dinitrobenzene may have caused this depletion through the production of reactive oxygen intermediates, generating oxidative stress and/or one or more metabolites of 1,3-dinitrobenzene which reacted nonenzymatically with GSH. The goal of this study is to investigate which of these two potential mechanisms may have caused the observed GSH depletion. Liver microsomes, known to rapidly metabolize 1,3-dinitrobenzene, generated the superoxide anion radical when incubated with 1,3-dinitrobenzene and NADPH. However, with the seminiferous tubule mitochondria, no oxygen radicals were detected. Hence, the aforementioned GSH depletion is unlikely due to the production of reactive oxygen intermediates from in situ mitochondrial metabolism of 1,3-dinitrobenzene. To investigate the ability of 1,3-dinitrobenzene metabolites to deplete seminiferous tubule mitochondrial GSH, mitochondria were incubated with 1,3-dinitrobenzene and NADPH. Loss of GSH correlated with the appearance of the 1,3-dinitrobenzene metabolites, nitrophenylhydroxylamine and nitroaniline. Subsequent investigation demonstrated that the metabolites, nitrosonitrobenzene, known to react nonenzymatically with nonprotein sulfhydryls, and nitrophenylhydroxylamine both oxidized seminiferous tubule mitochondrial GSH. Further studies suggested that nitrophenylhydroxylamine could deplete GSH via a free radical mechanism. In aqueous solution, this metabolite was shown to exist in equilibrium with a radical form, thought to be the hydronitroxide radical. The addition of GSH eliminated the signal, implying that the radical reacted nonenzymatically with GSH. In conclusion, the data in this study suggest that the decrease in mitochondrial GSH observed in DNB-treated seminiferous tubules is due to the formation of NPHA and NNB and not reactive oxygen intermediates.

Introduction DNB (1,3-dinitrobenzene),1 a compound widely used in the production of pesticides and dyes (1), has been shown to induce testicular toxicity in the rat (2). It was utilized in this study as a model for investigating nitroaromaticinduced testicular toxicity. 2,4- and 2,6-dinitrobenzene (3), trinitrotoluene (4), and nitrobenzene (5) are other nitroaromatics, which also have been shown to induce male reproductive toxicity in laboratory animals. DNB targets the seminiferous tubules of the rat, producing Sertoli cell vacuolization that is followed by degeneration of the seminiferous tubule epithelium (6, 7). The testicular toxicity of DNB may be due to metabolic activation at the site of toxicity. In vitro studies have shown DNB to be metabolized in testicular cell cultures (8-10), isolated seminiferous tubules (11), and their * To whom correspondence should be addressed. Phone: (530) 7523164. Fax: (530) 752-3516. E-mail: [email protected]. † Department of Biological Chemistry, University of California at Davis School of Medicine, Davis, CA 95616. Phone: (530) 754-7583. Fax: (530) 752-3516. E-mail: [email protected]. ‡ Department of Environmental Toxicology, University of California at Davis, Davis, CA 95616. Phone: (530) 754-8982. Fax: (530) 7523394. E-mail: [email protected]. 1 Abbreviations: DNB, 1,3-dinitrobenzene; NPHA, nitrophenylhydroxylamine; NA, nitroaniline; NNB, nitrosonitrobenzene; SOD, superoxide dismutase; DEPMPO, 5-diethoxyphosphoryl-5-methyl-1pyrroline-N-oxide.

mitochondria.2 In previous studies, using seminiferous tubules predigested away to remove interstitial tissue, DNB treatment decreased mitochondrial GSH and cellular ATP levels relative to the vehicle control (12). In addition, other experiments have shown that the bulk of DNB metabolism, using NADPH as a cofactor, occurs in the mitochondrial fraction of this tissue.3 These data suggest that the observed mitochondrial GSH depletion is related to mitochondrial metabolism of DNB. The reductive pathway of nitroaromatic metabolism is well described in the literature and is illustrated for DNB (Scheme 1). The nitro-containing parent compound is first reduced via one electron to the nitroxyl anion radical. In the presence of oxygen, this metabolite delivers an electron to molecular oxygen creating the superoxide anion radical. This process oxidizes the nitroxyl anion radical back to the parent compound. In the absence of oxygen, the nitroxyl anion radical is further reduced by one electron to form NNB (nitrosonitrobenzene), which undergoes a two-electron reduction to NPHA (nitrophenylhydroxylamine), which accepts two electrons to form NA (nitroaniline) (13). 2 Reeve, I. T. and Miller, M. G. (2002) Chem. Res. Toxicol. 15, 352360. 3 Reeve, I. T. and Miller, M. G. (2002) Chem. Res. Toxicol. 15, 352360.

10.1021/tx0155552 CCC: $22.00 © 2002 American Chemical Society Published on Web 02/23/2002

362

Chem. Res. Toxicol., Vol. 15, No. 3, 2002

Scheme 1. Nitroreductive Metabolism of 1,3-Dinitrobenzene to Nitrophenylhydroxylamine and Nitroaniline

Reeve et al.

study, the seminiferous tubule mitochondria were incubated with DNB and NADPH under nitrogen atmospheric conditions, minimizing the putative redox-cycle, to determine if the generation of stable metabolites could deplete GSH. Subsequently, seminiferous tubule mitochondria were incubated with DNB or one of its metabolites, NPHA, NA, and NNB, in the absence of NADPH, to investigate the ability of these compounds to react nonenzymatically with GSH.

Experimental Procedures

Multiple points along this pathway may be involved in the observed GSH depletion. Enzymatic generation of the superoxide anion radical has been demonstrated using numerous nitroaromatic compounds (14). Through the sequential actions of SOD (superoxide dismutase) and glutathione peroxidase, generation of this radical may lead to GSH depletion via oxidation to GSSG (15). Other potential sources of oxidative stress in the nitroreductive pathway are NNB and NPHA. There are numerous studies depicting the ability of aryl-nitroso compounds such as nitrosobenzene (16), 4-nitrosotoluene (17), and 4-nitrosophenetole (18) to react with sulfhydryls. Work by Ellis et al. has demonstrated that NNB can react nonenzymatically with GSH to produce predominantly NPHA at physiological pH (19). In addition, Maples et al. discovered that phenylhydroxylamine, a compound very similar in structure to NPHA, in aqueous solution exists in equilibrium with the phenylhydronitroxide free radical. This species was shown to react with GSH to form phenylhydroxylamine and thiyl free radicals (20). The goal of this project was to gain insight into the mechanism of the observed decrease in mitochondrial GSH in DNB treated whole seminiferous tubules. To determine if enzymatic nitroreduction of DNB is capable of radical production, liver microsomes, due to their relatively high nitroreductase activity (21), were incubated with NADPH and DNB. These incubations were analyzed using ESR for the superoxide anion radical and its precursor, the nitroxyl anion radical, under aerobic and anaerobic (nitrogen) atmospheres, respectively. To investigate the presence of these radicals in the target tissue, the experiment was repeated with incubations containing seminiferous tubule mitochondria. In a second

Caution: 1,3-Dinitrobenzene has been shown to produce methemoglobinemia, testicular toxicity and neurotoxicity in laboratory animals (1, 2). Reagents. DNB was purchased from Sigma Chemical Co.(St. Louis, MO). NA was purchased from Aldrich Chemical Co. (Milwaukee, WI). Both compounds were >98% pure as determined by HPLC. NPHA (>91% pure as determined by HPLC) was synthesized via reduction of DNB with palladium and phosphinite as described in Entwistle and Gilkerson (22). NNB (>98% pure as determined by HPLC) was donated by Dr. Mark Kurth (Department of Chemistry, UC Davis). The synthesis was carried out by initially reducing DNB with a rhodium-charcoal catalyst, in the presence of hydrazine, to NPHA, which was subsequently oxidized to NNB using FeCl3 (22). The spin-trap DEPMPO (5-diethoxyphosphoryl-5-methyl-1-pyrroline-N-oxide) was purchased from Oxis International, Inc. (Portland, OR). All other chemicals were obtained from Sigma or Bio-Rad (Hercules, CA) and were of the highest purity available. Structural Confirmation of NPHA. After synthesis, the dried NPHA product was solubilized in methanol. HPLC was then performed using an 8 mm i.d., 4 µm pore size uBondapak C-18 radial compression column from Waters Associates (Milford, MA). The mobile phase consisted of an isocratic solvent system of methanol and potassium phosphate buffer (50 mM, pH 6.8) at a ratio of 35:65, respectively. The flow rate was 2 mL/min. The product was detected using UV absorption (254 nm), fraction-collected, and extracted from the mobile phase using ethyl acetate. This extract was washed several times with water to remove any buffer salts and then evaporated under a stream of nitrogen. The dried product was solubilized in methanol, and negative ion MS was carried out using a triple quadrupole mass spectrometer. The m/z obtained for the product was 152.3, which, within the limit of instrument resolution, is the expected value for the molecular ion of NPHA. Animals. Male Sprague-Dawley rats (115-125 days old), were purchased from Charles-River (San Diego, CA). Animals were housed in a constant temperature and humidity environment (22 ( 2 °C and 50 ( 10%, respectively). A light/dark cycle of 12 h. was maintained. Food (Purina rat chow) and water were provided ad libitum. Preparation of Subcellular Fractions from Seminiferous Tubules and Liver. Rats were euthanized by carbon dioxide inhalation, and the testes were removed, detunicated, and quartered using a razor blade. To isolate the interstitiafree seminiferous tubules, the testes were digested using 1 mg/ mL Type IV collagenase (Sigma) for 30 min at 34 °C in a shaking water bath. The digestions were carried out in digestion medium (Dulbecco’s modified Eagle’s medium and nutrient mixture F12 [1:1] with 15 mM HEPES and 2.5 mM L-glutamine). After digestion, the tubules were washed several times with chilled isolation medium (0.25 M sucrose, 2 mM HEPES, pH 7.4). Washed, interstitia-free seminiferous tubules were homogenized using a glass/Teflon homogenizer. Mitochondria were isolated using the methods described previously by Meyers et al. (23). Liver microsomes were isolated as described by Lake (24). Electron Spin Resonance Studies. Incubations were carried out at room temperature in Tris (0.12 M) buffer, pH 7.4, and incubated for 1 min before analysis. ESR analysis was carried out at room temperature using a JEOL TE 300 spec-

1,3-Dinitrobenzene Metabolism and GSH Depletion trometer operating at 9.4 GHz and a modulation frequency of 100 kHz. For detection of the nitroxyl anion radical, the incubations were carrried out under anaerobic conditions: 20 µL of the incubation was pipetted into a one-inch section of gaspermeable Teflon tubing purchased from Norell, Inc. (Landisville, NJ). The tubing was folded on one end and placed in a modified quartz ESR tube (open on both ends), and the sample centered in the microwave cavity for analysis. One end of the tube was connected to a nitrogen tank. Nitrogen flowed continuously through the ESR tube and sample throughout the analysis. For the superoxide anion radical study, 5 µL of the incubation was placed in a quartz capillary tube, which was subsequently placed into a quartz ESR tube. The sample was then centered in the microwave cavity for analysis. For the NPHA generated radical species, the quartz capillary tube containing 5 µL of sample was analyzed using a loop gap resonator purchased from Medical Advances (Milwaukee, WI). ESR spectra were recorded using Origin (version 4.10) analytical software by Microcal (Northampton, MA). DNB Metabolite Formation vs GSH Depletion in Seminiferous Tubule Mitochondria. To investigate the influence of DNB metabolite formation on GSH, seminiferous tubule mitochondria were incubated in degassed and nitrogen sparged Tris (0.12 M) buffer, pH 7.4, under anaerobic conditions (nitrogen atmosphere) and at a protein concentration of 2.5 mg/ mL as determined using the BCA assay from Pierce (Rockford, IL). DNB (solubilized in acetone) and NADPH concentrations were 200 µM and 2 mM, respectively. The incubation concentration of acetone vehicle was 0.24% (v/v). All incubations were incubated for one minute before the addition of acetone vehicle or DNB. The incubation temperature was 32 °C. All incubations (vol ) 425 µL), were carried out for 60 min. with duplicate 100 µL sample aliquots taken at T ) 0 and 60 min. Each sample aliquot was quenched with 200 µL of chilled methanol for the metabolism study or 50 µL of a chilled aqueous solution containing 10% perchloric acid and 1 mM EDTA for GSH analysis. Influence of DNB, NNB, NPHA, and NA on Seminiferous Tubule GSH. To study the direct effects of DNB and the metabolites NNB, NPHA, and NA on seminiferous tubule mitochondrial GSH, incubations were carried out at room temperature and in the absence of NADPH. Seminiferous tubule mitochondria were incubated under aerobic conditions in Tris (0.12 M) buffer, pH 7.4, with acetone vehicle or 200 µM DNB, NPHA, NNB, or NA solubilized in acetone (0.24% v/v final concentration). Doses were delivered after a 1 min preincubation period. The incubation volume was 450 µL. Duplicate 100 µL aliquots were taken after T ) 0 and 20 min of incubation and the protein was precipitated as stated earlier. Analysis of GSH and GSSG. GSH and GSSG were derivatized via the methodology of Reed et al. (25). Briefly, the precipitated protein was pelleted via centrifugation at 10000g for 10 min at 4 °C. A total of 100 µL of the supernatant was removed and treated with 10 µL of 20 mg/mL of aqueous iodoacetic acid solution, 10 µL of 2.16 mM γ-L-glutamyl Lglutamic acid and 100 µL of 2 M KOH/2.3 M KHCO3, in that order. The sample was then mixed and incubated at room temperature for 30 min. Subsequently, 10 µL of 10% 2,4dinitrofluorobenzene was added (solution turned a golden yellow), and the mixture was incubated overnight in the dark at room temperature. After centrifugation of the precipitate at 10000g for 10 min at 4 °C, the supernatant was analyzed via HPLC. GSH and GSSG were resolved following the procedures of Reed et al. (25) except that the column was an Alltech Econosphere NH2 column with a pore size of 5 microns, a length of 150 mm and an internal diameter of 4.6 mm. The mobile phase flow rate was 1.0 mL/min. Analysis of DNB and Its Metabolites. The methanolic precipitate of each sample aliquot was centrifuged for 2 min at 2218g, and the levels of DNB and its metabolites in the supernatant were analyzed using the previously described HPLC and UV detection conditions. DNB (26) and to the only

Chem. Res. Toxicol., Vol. 15, No. 3, 2002 363

Figure 1. (A) ESR spectrum produced by incubating DNB (1 mM) and NADPH (2 mM) with liver microsomes (2.5 mg of protein/mL) under nitrogen atmospheric conditions. (B) As in panel A, but with the addition of superoxide dismutase (25 units of activity/µL). (C) As in panel A, but with acetone vehicle in place of 1,3-DNB. (D) As in panel A, but with heat denatured liver microsomes. Instrument settings: microwave power, 1.02 mW; modulation width, 0.079 mT; gain, 100. metabolites detected, NPHA and NA (26), were identified and quantitated via cochromatography with authentic standards. Statistical Analysis of DNB Metabolite Formation and GSH Depletion. Differences in DNB (nmol/mg of protein) at the T0 timepoint vs those after 60 min of incubation were analyzed using the F-test followed by the Student’s t-test (27). Differences in seminiferous tubule mitochondrial GSH levels (% of vehicle control) at the T0 timepoint vs those after 60 min of incubation were analyzed using the F-test followed by the Student’s t-test (27). Statistical Analysis of GSH and GSSG Data. Differences between the concentrations of GSH and GSSG in seminiferous tubule mitochondria treated with vehicle vs those treated with DNB, NNB, NPHA, or NA were analyzed using the F-test followed by the Student’s t-test (27).

Results DNB-Generated Free Radicals. Under a nitrogen atmosphere, DNB treatment caused the production of a radical species which did not require the use of a spintrap for detection and was unaffected by the addition of SOD (superoxide dismutase) (Figure 1, panels A and B). Under aerobic conditions and in the presence of a spintrap, DNB treatment of liver microsomes produced a radical species, which was eliminated by the addition of SOD (Figure 2, panels A and B). Substituting the acetone vehicle for DNB or heat-denaturing the microsomal protein prior to ESR analysis eliminated the signal under anaerobic (Figure 1, panels C and D) and aerobic (Figure 2, panels C and D) conditions. No radicals could be detected in seminiferous tubule mitochondrial incubations treated with DNB. Effects of DNB Metabolism on Seminiferous Tubule Mitochondrial GSH. Under anaerobic conditions, at T ) 0 min, 93.5 ( 10.7 nmol of DNB/mg of protein was detected. Following 60 min. of incubation, this concentration decreased significantly (P < 0.05) to 56.1 ( 4.2 nmol of DNB/mg of protein. In addition, the metabolites, NPHA and NA, were detected at 25.4 ( 2.2 and 3.4 ( 0.4 nmol/mg of protein, respectively (Figure

364

Chem. Res. Toxicol., Vol. 15, No. 3, 2002

Reeve et al.

Figure 4. GSH (light shaded area) and GSSG (dark shaded area) concentrations in seminiferous tubule mitochondrial incubations following 20 min of incubation with acetone vehicle or a 200 µM concentration of DNB, NNB, NPHA, or NA. Incubation GSH and GSSG concentrations found to be significantly different (P < 0.05) from those of the acetone vehicle control are indicated by an asterisk (*) or a tau (τ), respectively.

Figure 2. (A) ESR spectrum produced by incubating DNB (1 mM), DEPMPO (50 mM), and NADPH (2 mM) with liver microsomes (2.5 mg of protein/mL) under aerobic conditions. (B) As in panel A, but with the addition of superoxide dismutase (25 units of activity/µL). (C) As in panel A, but with acetone vehicle in place of DNB. (D) As in panel A, but with heat denatured liver microsomes. Instrument settings: microwave power, 1.02 mW, modulation width, 0.079 mT; gain, 100. Figure 5. (A) ESR spectrum produced by 1 mM NPHA in Tris (0.12 M) buffer, pH 7.4. (B) As in panel A, but with the addition of 1 mM GSH. Instrument settings: microwave power, 2.00 mW; modulation amplitude, 0.079 mT; gain, 500.

Figure 3. DNB metabolite formation vs GSH depletion in seminiferous tubule mitochondria. (A) Metabolic profile of preparation at T0 and after 60 min of incubation: DNB (light shaded area); NPHA (crosshatched area); NA (dark shaded area). The concentrations of DNB at the T0 incubation timepoint and after 60 min of incubation differed significantly (P < 0.05) as indicated by an asterisk(*). (B) GSH levels at T0 and T ) 60 min differed significantly (P < 0.05) as indicated by an asterisk (*).

3A). At T ) 0 min, the GSH levels within the DNB treated seminiferous tubule mitochondria were 102.2 ( 9.9% of the vehicle control. After 60 min of incubation with DNB under a nitrogen atmosphere, these levels decreased to 65.7 ( 3.3% of the vehicle control (Figure 3B). In the incubations treated with the vehicle control, concentrations of GSH after 60 min of incubation were 102 ( 8.2% of those at T ) 0 min.

GSH Depletion by DNB and Metabolites. NA had no effect on seminiferous tubule mitochondrial GSH. Relative to the vehicle control, NNB caused a decrease in GSH levels by 6.3 ( 0.9 nmol/mg of protein. This correlated with an increase in GSSG by 3.2 ( 0.4 nmol/ mg of protein. Treatment with NPHA produced a decrease in GSH by 6.2 ( 1.7 nmol/mg of protein and elevated GSSG levels by 3.8 ( 0.7 nmol/mg of protein. With the DNB treatment, GSH levels were decreased by 1.1 ( 1.2 nmol/mg of protein while GSSG levels increased by 0.4 ( 0.1 nmol/mg of protein. The differences in GSH and in GSSG concentrations between each of these treatments and the vehicle control were found to be significant (P < 0.05) (Figure 4). NPHA Radical Species. ESR analysis demonstrated that the NPHA in solution exists in equilibrium with a radical species. The addition of 1 mM GSH to the incubation eliminated the signal (Figure 5).

Discussion In studies carried out by Jacobsen and Miller (12), DNB treatment of isolated whole seminiferous tubules from which the interstitium had been removed by collagenase digestion, caused a decrease in mitochondrial GSH. In other investigations it was determined that mitochondrial metabolism in seminiferous tubules accounted for the bulk of DNB reductase activity.4 These data suggested that mitochondrial GSH depletion could be associated with mitochondrial metabolism of DNB. One possible mechanism of GSH depletion is the production of high levels of the superoxide anion radical. 4 Reeve, I. T. and Miller, M. G. (2002) Chem. Res. Toxicol. 15, 352360.

1,3-Dinitrobenzene Metabolism and GSH Depletion

Chem. Res. Toxicol., Vol. 15, No. 3, 2002 365

This reactive oxygen intermediate, created by the nitroxyl anion radical and oxygen (Scheme 1), may indirectly deplete GSH via SOD and glutathione peroxidase. SOD catalyzes the formation of hydrogen peroxide from the superoxide anion radical and water. Glutathione peroxidase then reduces the hydrogen peroxide to water and oxygen at the expense of GSH, which is concomitantly oxidized to GSSG (15). Another possible mechanism responsible for the observed loss of GSH, is the generation of one or more DNB metabolites, which are capable of nonenzymatic reaction with GSH. The data obtained in this study suggest that this latter mechanism may be involved in the observed thiol depletion. The present ESR studies utilizing liver microsomes demonstrated that enzymatic nitroreduction of DNB, in the presence of oxygen, generates the superoxide anion radical (Figure 2). Incubation of 2.5 mg/mL liver microsomal protein with 2 mM NADPH, 1 mM DNB, and 50 mM DEPMPO under aerobic conditions produced a paramagnetic species. The identity of this radical was investigated using the three control incubations. The first control was identical to the above incubation but with the addition of SOD (25 units of activity/µL), which eliminated the ESR signal. Since SOD is specific for the superoxide anion radical, these data indicated that the signal was due to the production of this radical species. The second control substituted the acetone vehicle for DNB. This incubation revealed no ESR signal, indicating that DNB is required to produce the radical. The final control utilized heat-denatured microsomal protein. This incubation also lacked signal, demonstrating that enzymatic activity is required to produce the radical. Under nitrogen atmospheric conditions, 2.5 mg/mL liver microsomes, 2 mM NADPH, and 1 mM DNB generated a radical species characteristic of the nitroxyl anion radical (Figure 1). The inclusion of SOD in the incubation had no effect on the signal. This control plus the anaerobic atmosphere of the incubation ruled out the possibility of the radical being the superoxide anion radical. In addition, the use of a spin-trap was not required. This is typical of the nitroxyl anion radical, which, under anaerobic conditions, is stable enough to accumulate and be detected without the use of a spintrap (28). The lack of signal in the vehicle and denatured protein controls demonstrate that both DNB and enzymatic activity were required to produce this radical. Although evidence for the presence of both the superoxide anion and nitroxyl anion radicals was found in liver microsomal incubations, neither radical was detected in seminiferous tubule mitochondrial incubations. In a previous study that monitored DNB metabolite formation in seminiferous tubule mitochondrial incubations,5 seminiferous tubule mitochondria at 2.5 mg of protein/mL were incubated for 30 min with 2 mM NADPH and 200 µM DNB at 32 °C under aerobic vs nitrogen atmospheric conditions. The metabolites, NPHA and NA, were generated under anaerobic conditions but in the presence of oxygen no metabolites were detected. This inhibition of metabolite formation by oxygen suggests the occurrence of redox-cycling with formation of the superoxide anion radical. Hence, the absence of signal in the ESR studies may be due to the relatively low nitroreductase activity of the seminiferous tubule mitochondria and the limit of

detection of the ESR instrument. It is possible that the nitroxyl anion and superoxide anion radicals were being generated but were at too low a concentration to be detected. A second study investigated the ability of DNB and its metabolites to deplete seminiferous tubule mitochondrial GSH. In the first experiment, DNB was incubated with seminiferous tubule mitochondria and NADPH under nitrogen atmospheric conditions to eliminate redoxcycling and promote the generation of DNB metabolites. The disappearance of DNB correlated with the generation of NPHA and NA and the depletion of GSH (Figure 3). To identify the DNB metabolite(s) capable of directly interacting with GSH, DNB, NNB, NPHA, or NA were incubated with seminiferous tubule mitochondria at room temperature and in the absence of NADPH, minimizing enzymatic activity. This experiment identified NPHA and NNB, as relatively potent thiol-depleting agents, decreasing GSH levels and generating GSSG (Figure 4). NNB, at pH 7.4, has been demonstrated to react directly with GSH resulting in the formation of NPHA (19). In this study, ESR experiments demonstrated that NPHA might oxidize GSH via a radical mechanism. In aqueous solution, NPHA was shown to exist in equilibrium with a radical form, which was eliminated by the addition of GSH (Figure 5). Hence, the mechanism of thiol depletion of NPHA may be analogous to that of phenylhydroxylamine, which was shown to deplete GSH via the phenylhydronitroxide radical (20). The data obtained in this study suggest that the observed decrease in mitochondrial GSH in DNB treated seminiferous tubules is the result of NNB and NPHA generation in the mitochondria. Although liver microsomal studies have shown that enzymatic nitroreduction of DNB is capable of generating the superoxide anion radical, it does not appear to be generated at a toxicologically significant level within the seminiferous tubule mitochondria. However, if enzymatic reduction of DNB does generate a redox-cycle in the seminiferous tubule mitochondria, it may be indirectly decreasing GSH levels through cofactor depletion. In this scenario, the DNB reductase(s) involved in this futile cycling process compete with GSSG reductase for NADPH. If this competition is substantial enough, the antioxidant capacity of the mitochondria could be compromised. Evidence of NADPH depletion was seen in previous metabolism studies using DNB.6 In addition, studies utilizing other nitroaromatics, nitrofurantoin and nifurtimox, have demonstrated that these compounds, through depletion of NADPH, may perturb redox metabolism within the liver, causing the observed increase in GSSG excreted into the bile (29, 30). Other experiments indicated that nitrofurantoin treatment increased GSSG concentration through inhibition of GSSG reductase (29). This study provides data to support the hypothesis that nitroreduction of DNB within the seminiferous tubules may be causing testicular toxicity by compromising mitochondrial antioxidant capacity. Under aerobic conditions, redox-cycling occurs, depleting NADPH. With less NADPH available, the activity of GSSG reductase is decreased, lowering the levels of mitochondrial GSH. Once the oxygen surrounding the site of metabolism is reduced, NNB is generated which reacts with GSH to

5 Reeve, I. T. and Miller, M. G. (2002) Chem. Res. Toxicol. 15, 352360.

6 Reeve, I. T. and Miller, M. G. (2002) Chem. Res. Toxicol. 15, 352360.

366

Chem. Res. Toxicol., Vol. 15, No. 3, 2002

form NPHA and GSSG. The radical species, which exists in equilibrium with NPHA, also oxidizes mitochondrial GSH to generate GSSG. Therefore, there may be a sequence of events which act to deplete the GSH levels within the mitochondria. This mechanism could be involved in the observed decrease in mitochondrial GSH levels in DNB treated rat seminiferous tubules (12) and in the testicular toxicity of DNB (6, 7).

Acknowledgment. The authors’ research reported here was supported by the NIEHS Center for Environmental Health Sciences Grant ES05707 (P01) and the NIH Grant ES05701 (R01).

References (1) Hartter, D. R. (1985) The use and importance of nitroaromatic chemicals in the chemical industry. In Toxicity of Nitroaromatic Compounds (Rickert, D. E., Ed.) pp 1-13, Hemisphere Publishing Corporation, Washington, DC. (2) Cody, T. E., Witherup, S., Hastings, L, Stemmer, K., Christian, and R. T. (1981) 1,3-Dinitrobenzene: toxic effects in vivo and in vitro. J. Toxicol. Environ. Health. 7, 829-847. (3) Rickert, D. E., Butterworth: B. E., and Popp, J. A. (1984) Dinitrotoluene: acute toxicity, oncogenicity, genotoxicity and metabolism. CRC Crit. Rev. Toxicol. 13, 217-234. (4) Levine, B. S., Furedi, E. M, Gordon, D. E., Lish, P. M., and Barkley, J. J. (1984) Subchronic toxicity of trinitrotoluene in Fischer 344 rats. Toxicology 32, 253-265. (5) Bond, J. A., Chism, J. P, Rickert, D. E., and Popp, J. A. (1981) Induction of hepatic and testicular lesions in Fischer-344 rats by single oral doses of nitrobenzene. Fundam. Appl. Toxicol. 1, 389394. (6) Hess, R. A., Linder, R. E., Strader, L. F., and Perreault, S. D. (1988) Acute effects and long-term sequelae of 1,3-dinitrobenzene on male reproduction in the rat II. Quantitative and qualitative histopathology of the testes. J. Androl. 9, 327-342. (7) Blackburn, D. M., Gray, A. J., Lloyd, S. C., Sheard, C. M., and Foster, P. M. D. (1988) A comparison of the effects of the three isomers of dinitrobenzene on the testes in the rat. Toxicol. Appl. Pharmacol. 92, 54-64. (8) Foster, P. M. D., Sheard, C. M., and Lloyd, S. C. (1986) 1,3Dinitrobenzene: a Sertoli cell toxicant? In Molecular and cellular endrocrinology of the testes (Stefanini, M., Conti, M., Geremia, R., and Ziparo, E., Eds.) pp 281-288, Elsevier, Amsterdam. (9) Foster, P. M. D, Lloyd, S. C., and Prout, M. S. (1987) Toxicity and metabolism of 1,3-dinitrobenzene in rat testicular cell cultures. Toxicol. in Vitro 1, 31-37. (10) Brown, C. D., and Miller, M. G. (1991) Effect of culture age on 1,3-dinitrobenzene metabolism and indicators of cellular toxicity in rat testicular cells. Toxicol. in Vitro 5, 269-275. (11) Jacobson, C. F., and Miller, M. G. (1997) 1,3-Dinitrobenzene metabolism and toxicity in seminiferous tubules isolated from rats of different ages. Toxicology 123, 15-26. (12) Jacobson, C. F., and Miller, M. G. (1998) Species difference in 1,3-dinitrobenzene testicular toxicity: in vitro correlation with glutathione status. Reprod. Toxicol. 12, 49-56. (13) Mason, R. P., and Josephy, P. D. (1985) Free radical mechanism of nitroreductase. In Toxicity of Nitroaromatic Compounds (Rickert, D. E., Ed.) pp 121-140, Hemisphere Publishing Corporation, Washington, DC.

Reeve et al. (14) Sealy, R. C., Swartz, H. M., and Olive, P. L. (1978) Electron spin resonance-spin trapping. Detection of superoxide formation during aerobic microsomal reduction of nitro-compounds. Biochem. Biophys. Res. Commun. 82, 680-684. (15) Alvares, A. P., and Pratt, W. B. (1990) Pathways of Drug Metabolism. In Principles of Drug Action: The Basis of Pharmacology (Pratt, W. B., and Taylor, P., Eds.) 3rd. ed., p 372, Churchill Livingstone, New York. (16) Eyer, P. (1979) Reactions of nitrosobenzene with reduced glutathione. Chem.-Biol. Interact. 24, 227-239. (17) Dolle, B., Topner, W., and Neumann, H. G. (1980) Reactions of arylnitroso compounds with mercaptans. Xenobiotica 10, 527536. (18) Mulder, G. J., Kadlubar, F. F., Mays, J. B., and Hinson, J. A. (1984) Reaction of mutagenic phenacetin metabolites with glutathione and DNB. Mol. Pharmacol. 26, 342-347. (19) Ellis, M. K., Hill, S., and Foster, P. M. D. (1992) Reactions of nitrosonitrobenzenes with biological thiols: identification and reactivity of glutathion-s-yl conjugates. Chem.-Biol. Interact. 82, 151-163. (20) Maples, K. R., Eyer, P., and Mason, R. P. (1989) Aniline-, phenylhydroxylamine-, nitrosobenzene-, and nitrobenzene-induced hemoglobin thiyl free radical formation in vivo and in vitro. Mol. Pharmacol. 37, 311-318. (21) Harada, N., and Omura, T. (1980) Participation of cytochrome P-450 in the reduction of nitro compounds by rat liver microsomes. J. Biochem. 87, 1539-1554. (22) Entwistle, I. D., and Gilkerson, T. (1978) Rapid catalytic transfer reduction of aromatic nitro compounds to hydroxylamines. Tetrahedron 34, 213-215. (23) Meyers, L. L., Beierschmitt, W. P., Khairallah, E. A., and Cohen, S. D. (1988) Acetaminophe-induced inhibition of hepatic mitochondrial respiration in mice. Toxicol. Appl. Pharmacol. 93, 378387. (24) Lake, B. G. (1987) Preparation and characterisation of microsomal fractions for studies on xenobiotic metabolism. In Biochemical Toxicology: A Practical Approach (Snell, K., and Mullock, B., Eds.) pp 183-187, IRL Press: Washington, DC. (25) Reed, D. J., Babson, J. R., Beatty, P. W., Brodie, A. E., Ellis, W. W., and Potter, D. W. (1980) High-performance liquid chromatography analysis of nanomole levels of glutathione, glutathione disulfide and related thiols and disulfides. Anal. Biochem. 106, 55-62. (26) McEuen, S., and Miller, M. G. (1991) Metabolism and pharmocokinetics of 1,3-dinitrobenzene in the rat and the hamster. Drug Metab. Dispos. 19, 661-666. (27) Gad, S., and Weil, C. S. (1991) Hypothesis testing: univariate parametric tests. Statistics and Experimental Design for Toxicologists, 3rd ed., pp 70-92, CRC Press, Ann Arbor. (28) Ayscough, P. B., Elliot, A. J., and Salmon, G. A. (1978) In situ radiolysis electron spin resonance study of the radical-anions of substituted nitroimidazoles and nitroaromatic compounds. J. Chem. Soc., Faraday Trans. I 74, 511-518. (29) Dubin, M., Moreno, S. N. J., Martino, E. E., Docampo, R., and Stoppani, A. O. M. (1984) Increased biliary secretion and loss of hepatic glutathione in rat liver after nifurtimox treatment. Biochem. Pharmacol. 32, 483-487. (30) Beatrice, M. C., Stiers, D. L., and Pfeiffer, D. R. (1984) The role of glutathione in the retention of Ca2+ by liver mitochondria. J. Biol. Chem. 259, 1279-1287.

TX0155552