|
Advertisement | |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
J. Biol. Chem., Vol. 280, Issue 27, 25305-25312, July 8, 2005
Glutathione Redox State Regulates Mitochondrial Reactive Oxygen Production*![]() From the Department of Environmental Health and Center for Environmental Genetics, University of Cincinnati Medical Center, P. O. Box 670056, Cincinnati, Ohio 45267-0056
Received for publication, January 4, 2005 , and in revised form, May 2, 2005.
Oxidative stress induced by 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD; dioxin) is poorly understood. Following one dose of TCDD (5 µg/kg body weight), mitochondrial succinate-dependent production of superoxide and H2O2 in mouse liver doubled at 728 days, then subsided by day 56; concomitantly, levels of GSH and GSSG increased in both cytosol and mitochondria. Cytosol displayed a typical oxidative stress response, consisting of diminished GSH relative to GSSG, decreased potential to reduce protein-SSG mixed disulfide bonds (type 1 thiol redox switch) or protein-SS-protein disulfide bonds (type 2 thiol redox switch), and a +10 mV change in GSSG/2GSH reduction potential. In contrast, mitochondria showed a rise in reduction state, consisting of increased GSH relative to GSSG, increases in type 1 and type 2 thiol redox switches, and a 25 mV change in GSSG/2GSH reduction potential. Comparing Ahr(/) knock-out and wild-type mice, we found that TCDD-induced thiol changes in both cytosol and mitochondria were dependent on the aromatic hydrocarbon receptor (AHR). GSH was rapidly taken up by mitochondria and stimulated succinate-dependent H2O2 production. A linear dependence of H2O2 productionon thereduction potential for GSSG/2GSH exists between 150 and 300 mV. The TCDD-stimulated increase in succinate-dependent and thiol-stimulated production of reactive oxygen paralleled a four-fold increase in formamidopyrimidine DNA N-glycosylase (FPG)-sensitive cleavage sites in mitochondrial DNA, compared with a two-fold increase in nuclear DNA. These results suggest that TCDD produces an AHR-dependent oxidative stress in mitochondria, with concomitant mitochondrial DNA damage mediated, at least in part, by an increase in the mitochondrial thiol reduction state.
All living cells are involved in reduction-oxidation (redox)1 activities that are essential to cellular function. Many such processes, such as mitochondrial respiration, monooxygenase, and oxidase activities, chemically reduce molecular oxygen to form reactive oxygen. A resulting redox imbalance in favor of oxidation (oxidative stress) may result in a cellular oxidative stress response. The chemical reductive processes for the formation of reactive oxygen species (mostly superoxide, hydrogen peroxide, hydroxyl radical) are supported primarily by reduced pyridine nucleotides, such as NADH and NADPH. Thus, higher rates of reactive oxygen production may occur when the redox state of the cell becomes more negative, a process termed reductive stress that may originate in the cytosol (1) or in the mitochondrial compartment (2, 3).
It has previously been shown that the ubiquitous environmental contaminant 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) increases the succinate-dependent mitochondrial production of superoxide and hydrogen peroxide (4, 5) and generates a mitochondrial oxidative stress response (6). This finding may have etiological relevance for the variety of adverse health effects mediated by TCDD and a large class of structurally similar polyhalogenated aromatic hydrocarbons (PHAHs) that are hydrophobic and metabolically recalcitrant, resulting in biological persistence in the activation of aromatic hydrocarbon receptor (AHR)-mediated pathways (712). The AHR, a cytosolic ligand-activated transcription factor, modulates the expression of a diverse array of genes (1315) and is responsible for the majority of the biological and adverse effects of TCDD (1621). TCDD toxicity appears to be mediated, at least in part, by an oxidative stress response resulting from transcriptional activation and a rise in the production of reactive oxygen (21, 22). The response is characterized by increases in such biomarkers of oxidative stress as liver heme oxygenase and metallothionein (23), increases in lipid peroxidation and DNA damage (2426), and oxidative perturbations in glutathione homeostasis (27). To minimize adverse cellular effects resulting from constitutive or excessive exposure to reactive oxygen, most cells have an elaborate defense system that includes a broad spectrum of chemical and enzyme scavengers for the oxidizing radicals and electrophiles. We have previously shown that TCDD treatment of mice produces an increase in mitochondrial succinate-dependent reactive oxygen production; this increase was not the result of changes in the mitochondrial antioxidant enzymes SOD2 (Mn-superoxide dismutase) or GPX1 (glutathione peroxidase) (4, 5). TCDD also produces changes in the levels of mitochondrial GSH and GSSG (5). GSH is an integral oxidant scavenger, reacting as either a one-electron donor to radicals or a two-electron donor to electrophiles, and occurs in all mammalian cell types (28). In this study, we further evaluate the relationship between TCDD exposure, the GSSG/2GSH redox couple, and mitochondrial reactive oxygen production.
ChemicalsTCDD was purchased from Accustandard (New Haven, CT). GSH (G6529) and GSSG (G6654) were Sigma Ultra grade from Sigma-Aldrich Chemicals. All other chemicals and reagents were obtained from Sigma-Aldrich Chemicals as the highest available grades. Animals and TreatmentExperiments involving mice were performed according to the National Institutes of Health standards for care and use of experimental animals and the University of Cincinnati Institutional Animal Care and Use Committee (IACUC). Animals were group-housed, maintained on a 12-h light/dark cycle, and had access to standard rodent chow and water ad libitum. Male mice (812 weeks of age) of the C57BL/6J inbred strain were purchased from The Jackson Laboratories (Bar Harbor, ME). Ahr(/) knock-out mice were a generous gift from Frank J. Gonzalez (Bethesda, MD) and are maintained in our animal colony. Mice were administered a single dose of TCDD (5 µg/kg body weight) in corn oil by intraperitoneal injection; controls were given equivalent volumes of corn oil. At 1, 7, 28, or 56 days following treatment, the mice were killed by carbon dioxide asphyxiation, followed by cervical dislocation. The liver was excised and washed in ice-cold 0.9% NaCl. A 10% whole homogenate was prepared in 250 mM sucrose, 1 mM EDTA, and 1 mM EGTA, 0.1% defatted, and recrystallized bovine serum albumin, 10 mM HEPES, pH 7.2, using a motor-driven (1000 rpm) Potter-Elvejhem homogenizer. A mitochondrial fraction was prepared as described previously (5), and suspended in a potassium chloride respiratory buffer (KCl-RB), consisting of 140 mM KCl, 0.1 mM EDTA, 2.5 mM KH2PO4, 2.5 mM MgCl2, and 0.1% bovine serum albumin, in 5 mM HEPES (pH 7.4). Reactive Oxygen AssaysH2O2 was monitored in freshly prepared mitochondria as luminol (5-amino-2,3-dihydro-1,4-phthalazinedione) chemiluminescence (4, 5), using a Berthold Autolumat Plus luminometer. The reaction mixture consisted of 5 µM luminol, 2.5 units/ml horseradish peroxidase, 50 µg of mitochondrial protein, and KCl-RB, in a final volume of 1.0 ml; the reaction was initiated by the addition of 6 mM sodium succinate and monitored at 37 °C. Catalase abolished luminol luminescence in all cases. Superoxide production was monitored by chemiluminescence in the same manner as that for H2O2, except the probe was 20 µM lucigenin (bis-N-methylacridinium) (5). Luminol and lucigenin are highly specific for H2O2 and superoxide, respectively (29).
Redox Potential for GlutathioneFor the GSSG/2GSH
half-reaction, GSSG + 2H+
E is as shown in
Equation 2.
The type 2 switch regulates the thermodynamics of protein disulfide
formation (Equation 5).
Mitochondrial GSH UptakeFreshly prepared mitochondria, maintained at 0 °C, were suspended in 1.5-ml microcentrifuge tubes at 150 µg of protein/ml, in KCl-RB containing GSH at various concentrations. Using a syringe, 0.25 ml of a 27% sucrose cushion was layered under the solution in each tube. Tubes were placed in a metal temperature block set to 37 °C, such that the walls of the tube contacted the block. Under these conditions, the temperature in the tubes reached 37 °C in 2.5 min. At various times the tubes were centrifuged through the sucrose cushion at 14,000 x g for 10 min at 0 °C. The supernatant solution was aspirated, and the pellets were washed with fresh KCl-RB. After centrifugation at 14,000 x g for 10 min at 0 °C, supernatant solutions were aspirated, and 150 µl of 5% trichloroacetic acid was added to the mitochondrial pellet. After sonication, GSH was determined (32) and expressed as nmol/mg of protein. Mitochondrial Membrane PotentialThe mitochondrial inner membrane potential was quantified using the cationic lipophilic dye JC-1 (5,5',6,6'-tetrachloro-1,1',3,3'-tetraethylbenzimidazolylcarbocyanine iodide) (33). To 10 µg of mitochondrial protein in 1.0 ml of KCl-RB was added 0.05 nmol of JC-1. Fluorescence ratios were determined at 37 °C. Monomeric green fluorescent JC-1 accumulates in the mitochondria in accordance with the membrane potential, negative inside (34). Above a critical concentration, the red fluorescent aggregates of JC-1 form. Thus, membrane potential was quantified using our experimentally determined wavelength pairs, as the ratio of red (Ex = 488 nm; Em = 595620 nm) to green fluorescence (Ex = 488 nm; Em = 535510 nm). The fluorescence ratios were standardized to membrane potential by applying known concentration gradients of potassium across the membrane, in the presence of 2 µg of valinomycin/ml and applying the Nernst equation to calculate potassium diffusion potentials. This technique is applicable for generating membrane potassium diffusion potentials to about 50 mV, with the ratio of fluorescence ratio to membrane potential of about 2 per 100 mV. For Fig. 7, we assumed linearity to express the results in terms of membrane potential. DNA Strand BreakageTo estimate oxidative damage to mitochondrial and nuclear DNA, we employed an assay that takes advantage of the enzyme specificity of the bacterial DNA repair glycosylase formamidopyrimidine DNA N-glycosylase (FPG), which produces alkaline cleavage-sensitive apurinic sites at 8-hydroxyguanine and 8-hydroxyadenine, as well as the imidazole ring-opened purines (2,6-diamino-4-hydroxy-5-formamidopyrimidine and 4,6-diamino-5-formamidopyrimidine) (3539). Using this enzymic method, DNA is not prone to spurious oxidation, and estimates of oxidative DNA damage have lower background levels than those obtained using high performance liquid chromatography (HPLC) with electrochemical detection or mass spectrometry.
Our assay was based on previously reported techniques
(35,
40) but varied in many
respects. For FPG analysis of 8-OHdG, a 100-mg piece of liver was frozen in
liquid nitrogen, then thawed and homogenized in lysis buffer (6 mM
Na2HPO4, 1 mM KH2PO4,
137 mM NaCl, 3 mM KCl, 0.1% Triton X-100) at 0 °C.
DNA extraction employed the NaI method using the DNA extractor WB kit (Wako
BioProducts, Richmond, VA). In the absence or presence of FPG (100-fold
dilution), 3 µg of DNA was incubated with 20 µl of enzyme reaction
buffer (10 mM MgCl2, 1 mM dithiothreitol, 0.1
mg/ml bovine serum albumin, in 10 mM bis-Tris buffer, pH 7.0).
After 30 min at 37 °C, 37.5 µl of alkaline buffer (13 mM
sodium EDTA, 370 mM NaCl, and 72 mN NaOH, pH 12.3), were
added, and the DNA was allowed to unwind in the dark on ice for 30 min. The
solution was neutralized with 11.5 µl of 0.1 N HCl and
sonicated. A 100-µl aliquot was mixed with 870 µl of enzyme reaction
buffer plus 30 µl of 0.02 mg/ml Hoechst 33258. After 10 min, fluorescence
was determined at 360Ex/450Em. Standard fluorescence
curves for single-stranded DNA ( Other AssaysGSH and GSSG were determined fluorometrically using the o-phthalaldehyde procedure (32). Protein was measured by the bicinchoninic acid method (Pierce), according to details provided by the manufacturer. StatisticsStatistical differences between group mean values were determined by one-way analysis of variance, followed by the Student-Newman-Kuels test for pairwise comparison of means. Statistics were performed using SigmaStat Statistical Analysis software (SPSS Inc., Chicago, IL). Biohazard PrecautionTCDD is highly toxic and a likely human carcinogen. All personnel were instructed in safe handling procedures. Lab coats, gloves, and masks were worn at all times, and contaminated materials were collected separately for disposal by the Hazardous Waste Unit or by independent contractors. TCDD-treated mice were housed separately, and their carcasses were treated as contaminated biological materials.
TCDD Effects on Reactive Oxygen Production and Mitochondrial and Cytosolic GSH and GSSGFollowing a single exposure to TCDD (5 µg/kg body weight), mitochondrial succinate-dependent production of both superoxide and hydrogen peroxide was unchanged at day 1, doubled by day 7, and remained elevated until day 28, and then subsided to near control levels by day 56 (Fig. 1). Concomitant with the increase in mitochondrial reactive oxygen production, the cytosolic levels of GSH were slightly increased, whereas GSSG levels more than doubled (Fig. 2, left panels); these relative changes in GSH and GSSG produced decreases in the potential to reduce protein-SSG mixed disulfide bonds (type 1 thiol redox switch) and protein-SS-protein disulfide bonds (type 2 thiol redox switch), as well as a +10 mV change in the GSSG/2GSH reduction potential. These results reflect a typical oxidative stress response for the cytosol following TCDD treatment. In sharp contrast to the changes observed in the cytosol, mitochondrial levels of GSH nearly doubled, whereas GSSG was only slightly increased (Fig. 2, right panels); this increase in GSH relative to GSSG resulted in an increase in reduction state, with increases in type 1 and type 2 thiol redox switches, accompanied by a negative shift in the GSSG/2GSH reduction potential by 25 mV. AHR-dependent Changes in TCDD-induced Cytosolic and Mitochondrial ThiolsWe have shown previously that TCDD-induced changes in mitochondrial levels of GSH and GSSG are dependent on the AHR (5). To extend these observations in the context of the present study, we examined hepatic glutathione levels in Ahr(/) knock-out mice and calculated parameters of thiol status in cytosol and in mitochondria (Fig. 3). TCDD-induced alterations in glutathione redox state were observed in wild-type but not Ahr(/) mice, indicating an AHR-dependent process.
GSH Uptake and Effects of GSH and GSSG Added in Vitro on H2O2 Production in Isolated MitochondriaTo determine the relationship between mitochondrial production of reactive oxygen, glutathione levels, and the redox state, we examined the direct effects of GSH and GSSG on mitochondrial H2O2 production. It was first important to determine whether externally added GSH can indeed enter the mitochondria. The rates of mitochondrial uptake or efflux of GSH are shown in Fig. 4. In the absence of external GSH, there is slow loss of internal GSH at 37 °C. Externally added GSH is rapidly taken up by mitochondria, with a slight lag as the reaction mixture warms. The addition of GSH in the range of 0.1 to 8 mM greatly stimulated succinate-dependent H2O2 production in a concentration-dependent fashion (Fig. 5, left panel). Conversely, GSSG, in the range of 0.05 to 0.4 mM, attenuated succinate-dependent H2O2 production at every GSH concentration. The data from the left panel of Fig. 5 were recalculated (Fig. 5, right panel) to show a linear dependence of H2O2 production on the reduction potential for GSSG/2GSH between 150 and 300 mV. Excluding exposure to strong oxidizing agents and electrophiles, this range represents the complete biological range of hepatic reduction potential for the GSSG/2GSH redox half-reaction.
Our previously published results showed that TCDD produces an increase in oxygen consumption without a change in energy coupling (4, 5). We might predict that the driving force for the increase in oxygen consumption could be a TCDD-mediated hyperpolarization of the liver mitochondrial membrane potential. This prediction was confirmed (Fig. 6). At 1 week following TCDD treatment, a succinate-dependent increase in the JC-1 aggregate fluorescence ratio (an estimate of membrane potential) approximately doubled. Mitochondrial versus Nuclear Genotoxicity Induced by TCDDTo determine the genotoxic relevance of TCDD-induced reactive oxygen production, we evaluated mitochondrial and nuclear oxidative DNA damage resulting from TCDD exposure. We used the enzyme FPG to generate single-strand breakage in DNA containing the major oxidative promutagenic DNA base modification 8-OHdG, followed by alkaline unwinding and quantification of single- and double-stranded DNA. At 1 week after TCDD treatment, the percentage of FPG-sensitive sites doubled in nuclear DNA, whereas the percentage increased 4-fold in mitochondrial DNA (Fig. 7).
Cellular Redox HomeostasisNormal cells maintain a specific redox homeostasis, such that important redox couples are maintained within physiologic ranges for oxidation and reduction. Such couples include the relatively oxidized NAD+/NADH, and the relatively reduced couples NADP+/NADPH and GSSG/2GSH (28, 41). When these redox couples are stimulated to move out of the physiologic range, biologically adverse consequences often result. For example, high reduction states for NADP+/NADPH activate the ubiquitous plasma membrane NADPH oxidase to generate superoxide, a process physiological for macrophage activation, but possibly toxicological in other cases (42, 43). Additionally, hypoxia and respiratory chain inhibitors can stimulate the production of reactive oxygen in mitochondria (2, 3). Such a stimulation of the generation of reactive oxygen by increasing the cellular reduction state has been termed reductive stress (44, 45).
Cytosolic versus Mitochondrial GSHCells maintain at least two distinct regulated pools of GSH, cytosolic and mitochondrial (46). Because GSH is synthesized in the cytosol by the sequential action of glutamate-cysteine ligase and glutathione synthetase, its mitochondrial presence requires inner membrane transport. Two mitochondrial electroneutral antiport carrier proteins have been shown to have the capacity to transport GSH, the dicarboxylate carrier protein and the 2-oxoglutarate carrier protein (47, 48). In the present study, externally added GSH is readily taken up by mitochondria, despite the 8
mM GSH present in the mitochondrial matrix. It therefore appears
that GSH is taken up against a concentration gradient. Because the membrane
potential under these conditions is low or absent, and the transport is
electrically neutral, it would appear that the energy for GSH uptake comes
from the cumulative concentration gradient of dicarboxylic acids and other
substrates for the GSH uptake carriers, which are located in the mitochondrial
matrix. Such a mechanism involving GSH competition with other carrier ligands
for carrier proteins would also explain why mitochondrial GSH does not escape
from mitochondria, even though efflux may be energetically favorable.
It is important to maintain the physiological levels of mitochondrial GSH.
Depletion of mitochondrial GSH by electrophilic chemicals, or by inhibiting
mitochondrial GSH transport, leads to respiratory toxicity, with a decrease in
ATP production, as well as increased production of reactive oxygen and an
augmented oxidative stress response
(4951).
In Ehrlich ascites tumor cells, depletion of mitochondrial GSH renders tumor
cells highly susceptible to TNF AHR-dependent Reductive Stress in Mitochondria by TCDDIn a recent report (5), we investigated the involvement of the AHR, CYP1A1, and CYP1A2 in the oxidative stress response elicited by TCDD. Because of the increase in mitochondrial GSSG levels and the increase in mitochondrial production of reactive oxygen, we concluded that TCDD produces a mitochondrial thiol oxidative stress response. In the present study, we further evaluated these findings plus new data, estimating the mitochondrial pH at 7.8, and mitochondrial volume at about 1 µl/mg of protein (55). Our analysis indicates that TCDD actually elicited a more reducing, rather than oxidizing, mitochondrial thiol status. These effects of TCDD on oxidative stress and glutathione redox status are not seen in the Ahr(/) mouse, but are seen in the Cyp1a1(/) or Cyp1a2(/) mouse (current data and Ref. 5). There is the apparent anomaly that TCDD produces a more reducing mitochondrial thiol environment, while stimulating the respiration-linked generation of reactive oxygen (4, 5). In fact, however, such a situation is quite reasonable. We have previously described a TCDD-mediated decrease in the activity of cytochrome oxidase. Diminished cytochrome oxidase activity could result in effects on mitochondrial electron transport similar to those described for respiration-limiting hypoxic conditions and for chemically induced simulation of mitochondrial hypoxia (2, 3). In those studies, hypoxia or cyanide were used to generate a mitochondrial reductive stress response, leading to the production of H2O2. The mechanism suggested was that the reducing environment facilitates enhanced redox cycling or autoxidation of Q, leading to increased rates for the production of superoxide (56). TCDD-induced DNA DamageTCDD is well known to generate increased amounts of reactive oxygen, both in the endoplasmic reticulum (29, 57) and in the mitochondria (4, 5). Therefore, it is logical to suppose that TCDD should generate nuclear DNA damage, leading to mutations; this proposal has been examined in the liver of Big BlueTM lacI transgenic rats of both sexes (58). The weekly dose of 2 µg/kg body weight for 6 weeks should have been sufficient to generate AHR-mediated gene induction, an oxidative stress response, and genotoxicity (25, 59, 60). However, TCDD did not produce any increase in the numbers of reversion mutants, nor in the mutational spectra of either sex (58). A possible explanation for the lack of TCDD-induced mutagenicity of nuclear DNA is that mitochondrial GSH may have a dual role with respect to production and utilization of H2O2. In other words, although the production of mitochondrial H2O2 may be stimulated by increased levels of GSH, GSH also acts in conjunction with mitochondrial GPX1 to scavenge H2O2 (Fig. 8). As we have shown previously, the TCDD-stimulated increase in mitochondrial production of succinate-dependent reactive oxygen cannot be explained by a decrease in intramitochondrial GPX1, because GPX1 activity actually increases in liver mitochondrial from TCDD-treated mice (4, 5). In practice, the amount of H2O2 detected by the standard methods employing extramitochondrial fluorescent and luminescent probes represents an underestimate of the amount produced. In this scenario, intramitochondrial H2O2-scavenging by GPX1 would result in TCDD actually eliciting a much greater production of H2O2 than reported (4, 5). This would explain, at least in part, the minimal nuclear genotoxicity, and lack of nuclear mutagenicity (58, 61), reported to date. This also explains our present results, showing a greater extent of mitochondrial FPG-sensitive DNA damage sites, relative to the extent observed for nuclear DNA (Fig. 7).
Mechanism by Which TCDD Induces Thiol-regulated Mitochondrial Reactive Oxygen ProductionIn the 1960s, it was demonstrated that iron, copper, and calcium were contaminants of GSH and GSSG that could cause high amplitude mitochondrial swelling (6264). These ions are now known to affect mitochondrial respiration and function by generating free radicals, or by uncoupling and generating permeability transition. The concentrations of ions that caused mitochondrial swelling was about 0.3 µM Cu2+,3 µM Fe2+, or 10 µM Ca2+(62). At that time 1 mM commercial GSH or 3 mM commercial GSSG contained 0.10.48 µM Cu2+, 0.110.82 µM Fe2+, and 0.233.8 µM Ca2+ (64). At 1 mM, the commercial GSH or GSSG utilized for the current studies contains <5 nM Cu2+, <5 nM Fe2+, and< 10 nM Ca2+. These concentrations of iron, copper, and calcium are at least an order of magnitude lower than those shown to produce high amplitude swelling. Therefore, we need not consider further the possibility that the GSH and GSSG effects reported in the current study could be attributable to ionic impurities. Mitochondria generate reactive oxygen as superoxide, resulting from one-electron uncoupling of respiratory chain substrate oxidation from oxygen reduction (65). Superoxide disproportionates to hydrogen peroxide and oxygen via superoxide dismutase (mitochondrial Mn-dependent SOD2 or cytosolic Cu-Zn-dependent SOD1). The production of respiratory-linked superoxide depends on redox cycling within complexes 1 and 3 (66). Therefore, we considered the possibility that glutathionylation (formation of protein cysteinyl mixed disulfides with glutathione) of complex 1 may be responsible for glutathione-stimulated reactive oxygen production. In this scenario, complex 1 generates superoxide following protein glutathionylation and intracomplex inhibition of electron transport, leading to redox cycling of iron-sulfur or flavoprotein centers of complex 1 (67, 68). We believe that this model is untenable, because any block in the respiratory chain that would increase reduction state and redox cycling of respiratory components would be expected to decrease the rate of oxygen consumption. To the contrary, TCDD increases mitochondrial oxygen consumption (5). Any model that seeks to explain the mechanism of thiol-regulated mitochondrial reactive oxygen production by TCDD would need to be consistent with the previous observation (5) that TCDD increases mitochondrial respiration, but does not change the respiratory control ratio. We have developed a working model that might explain these observations (Fig. 8 and Table I), based upon the relationship between membrane potential and mitochondrial production of reactive oxygen (69). In our model, superoxide generation depends on the thiol reduction state through the mitochondrial permeability transition pore (MPTP), a cyclosporin A-sensitive Ca2+-dependent nonspecific channel of about 1500 Da (70). The MPTP is composed of an inner-outer mitochondrial membrane junctional complex (71), containing the electrogenic inner membrane adenine nucleotide translocase (72, 73), a voltage-dependent anion channel of the outer membrane (74) and dissociable modifier proteins such as cyclophilin-D, hexokinase, and creatine kinase (71, 73). The MPTP has two voltage-sensitive sites, one of which is in equilibrium with the GSSG/GSH pool and inhibited both by N-ethylmaleimide (thiol-alkylating agent) and monobromobimane (thiol-reacting fluorescent probe) (75, 76). This pore is gated by a critical dithiol in the voltage-sensing region, and opened following GSH oxidation to GSSG, a process that is inhibited by the reducing thiol dithiothreitol (77). Opening the pore results in a Ca2+-dependent decrease in the mitochondrial inner membrane potential (76). The MPTP has a second site that is in redox equilibrium with the pyridine nucleotide pool. An increase in [(NAD + NADP)/(NADH + NADPH)], where GSH remains fully reduced, opens the pore (75, 78). This site is blocked by N-ethylmaleimide but not by monobromobimane, suggesting that its critical thiol is buried within the inner mitochondrial membrane. In general, therefore, the MPTP is closed under reducing conditions and open under oxidizing conditions. However, the MPTP can also rapidly and transiently alternate between the open and closed states, a process termed flicker (7982). We propose that under the normal thiol redox state, the pore is in flicker mode, maintaining the membrane potential at normal levels. When mitochondrial GSH becomes oxidized, permeability transition occurs, the membrane potential collapses, and apoptosis is initiated. Following TCDD treatment, the elevated redox state of glutathione inhibits MPTP flicker, resulting in an elevated (more negative) membrane potential. Increased production of superoxide follows from the increase in redox cycling of ubiquinone within complex 3 (8385) or following reverse electron transport from succinate to NAD+ with the production of reactive oxygen within complex 1 (79, 86, 87).
This working model accounts for the thiol redox state-dependent increase in reactive oxygen production following TCDD exposure. We are aware that changes in thiol redox state may secondarily shift the NAD+/NADH and NADP+/NADPH redox potential toward more reduced states, possibly through glutathione reductase and the energy-linked transhydrogenase. Thus, the increase in glutathione redox state may supply NADH as substrate to increase the reduction state of complex 1 components, which may generate redox cycling within complex 1 and result in additional reactive oxygen production. Regardless of intermediary events in the mechanism by which the glutathione redox state regulates the production of mitochondrial reactive oxygen, the linear relationship between succinate-dependent reactive oxygen production, and the reduction potential of the GSSG/2GSH redox couple, suggests that this thiol redox potential is the primary driving force for the generation of reactive oxygen. Our calculations show that the magnitude of change of the thiol redox potential is sufficient to explain the increase in reactive oxygen following TCDD treatment. According to Fig. 3, the reduction potential for mitochondrial GSH, following TCDD treatment, decreases from 260 to 285 mV. The corresponding H2O2 released from mitochondria at these reduction potentials can be estimated, using Fig. 5 (right), as 1.1 x 106 and 1.4 x 106, respectively. The difference of 0.3 x 106 luminescence units corresponds to 2.5 nmol of
H2O2/min/mg protein
(4), a value similar to the
succinate-stimulated increase in mouse liver mitochondrial
H2O2 production following TCDD treatment
(5).
In summary, we have shown that the TCDD-induced stimulation of mitochondrial succinate-dependent production of superoxide and hydrogen peroxide is accompanied, and most likely exacerbated, by an increase in the thiol reduction state of mitochondrial GSH. These TCDD-induced changes in hepatic GSH, GSSH and reactive oxygen production are somehow controlled by the AHR. The mechanisms responsible for the thiol reduction state regulation of mitochondrial respiration and production of reactive oxygen are under investigation, according to our proposed models. Despite the increased production of reactive oxygen in mitochondria from TCDD-treated mice, release of H2O2 to the cytosol may be limited, because of intramitochondrial scavenging of reactive oxygen by Mn-SOD and by GPX1. Thus, it is reasonable that mitochondrial DNA suffers a disproportionate amount of damage, relative to nuclear DNA damage, as we have observed. It is therefore important to determine whether increased mitochondrial DNA damage or other forms of mitochondrial damage are directly associated with the pathogenic effects of TCDD.
* This work was supported in part by National Institutes of Health Grants R01 ES10133 (to H. G. S.), R01 ES08147 (to D. W. N), RO1 ES12463 (to T. P. D.), and P30 ES06096 (to T. P. D., D. W. N., and H. G. S.). The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
1 The abbreviations used are: redox, reduction-oxidation; AHR, aromatic
hydrocarbon receptor; GPX1, glutathione peroxidase-1; KCl-RB, potassium
chloride respiratory buffer; TCDD,
2,3,7,8-tetrachlorodibenzo-p-dioxin; FPG, formamidopyrimidine DNA
N-glycosylase; MPTP, mitochondrial permeability transition pore.
We thank our colleagues for a careful reading of this manuscript. Greg G. Oakley generously supplied the X174 Virion and RFII DNA.
This article has been cited by other articles:
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
Advertisement | |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||