Advertisement

Glutathione Depletion in PC12 Results in Selective Inhibition of Mitochondrial Complex I Activity

IMPLICATIONS FOR PARKINSON′S DISEASE*
      Oxidative stress appears to play an important role in degeneration of dopaminergic neurons of the substantia nigra (SN) associated with Parkinson's disease (PD). The SN of early PD patients have dramatically decreased levels of the thiol tripeptide glutathione (GSH). GSH plays multiple roles in the nervous system both as an antioxidant and a redox modulator. We have generated dopaminergic PC12 cell lines in which levels of GSH can be inducibly down-regulated via doxycycline induction of antisense messages against both the heavy and light subunits of γ-glutamyl-cysteine synthetase, the rate-limiting enzyme in glutathione synthesis. Down-regulation of glutamyl-cysteine synthetase results in reduction in mitochondrial GSH levels, increased oxidative stress, and decreased mitochondrial function. Interestingly, decreases in mitochondrial activities in GSH-depleted PC12 cells appears to be because of a selective inhibition of complex I activity as a result of thiol oxidation. These results suggest that the early observed GSH losses in the SN may be directly responsible for the noted decreases in complex I activity and the subsequent mitochondrial dysfunction, which ultimately leads to dopaminergic cell death associated with PD.
      PD
      Parkinson's disease
      SN
      substantia nigra
      ROS
      reactive oxygen species
      GSH
      glutathione
      GCS
      glutamyl-cysteine synthetase
      dox
      doxycycline
      GSSG
      glutathione disulfide
      DTT
      dithiothreitol
      rTta
      reverse transcriptional transactivator
      DCF
      2′,7′-dichlorofluorescein
      MTT
      5′,5-dithiobis(2-nitrobenzoic acid)
      Parkinson's disease (PD)1 involves degeneration of dopaminergic neurons of the substantia nigra (SN). Oxidative stress appears to play a major role in the death of these neurons. Dopaminergic neurons are believed to be especially prone to oxidative stress because of the potential for dopamine oxidation to occur either through auto-oxidation or via metabolism by the enzyme monoamine oxidase. Oxidation of dopamine produces reactive oxygen species (ROS) including hydrogen peroxide (H2O2). H2O2 can react with ferrous (Fe2+) iron to produce hydroxyl radicals (OH·), which can damage nearby proteins, nucleic acids, and membrane phospholipids (
      • Beal M.F.
      ). Iron levels are increased in the SN of PD patients along with elevations in various indices of oxidative damage (
      • Reiderer P.
      • Sofic E.
      • Rausch W.D.
      • Schmidt B.
      • Reynolds G.
      • Jellinger K.
      • Youdim M.B.H.
      ,
      • Dexter D.T.
      • Holley A.E.
      • Flitter W.D.
      • Slater T.F.
      • Wells F.R.
      • Daniel S.E.
      • Lees A.J.
      • Jenner P.
      • Marsden C.D.
      ,
      • Yoritaka A.
      • Hattori N.
      • Uchidak K.
      • Tanaka M.
      • Stadtman E.R.
      • Mizuno Y.
      ).
      PD is also characterized by decreases in SN levels of the thiol antioxidant glutathione (GSH). Although GSH is not the only antioxidant molecule reported to be altered in PD, the magnitude of GSH depletion appears to parallel disease severity and it is the earliest known indicator of oxidative stress in presymptomatic PD, preceding decreases in both mitochondrial complex I activity and dopamine levels (
      • Perry T.L.
      • Yong V.W.
      ,
      • Jenner P.
      ). Nigral neurons contain GSH, and levels are reduced in the PD brain (
      • Pearce R.K.
      • Owen A.
      • Daniel S.
      • Jenner P.
      • Marsden C.D.
      ).
      GSH is synthesized by a two-step reaction involving the enzymes γ-GCS and glutathione synthetase. γ-GCS is the rate-limiting enzyme in this process, and brain GSH appears to primarily arise through synthesis from its constituent amino acids via this enzyme (
      • Meister A.
      ). γ-GCS is a dimer composed of a heavy catalytic subunit and a light regulatory subunit (
      • Huang C.S.
      • Chang L.S.
      • Anderson M.E.
      • Meister A.
      ).
      GSH is synthesized in the cytosol and transported into the mitochondria via an energy-dependent transporter (
      • Meister A.
      ). The mitochondria contains GSH peroxidase and all other components necessary for detoxification of hydroperoxides but no catalase. Decreases in GSH availability in the brain therefore are believed to promote mitochondrial damage via increased ROS (
      • Jain A.
      • Martensson J.
      • Stole E.
      • Auld P.A.
      • Meister A.
      ). Mitochondrial dysfunction appears to play a role in the neurodegeneration associated with PD (
      • Beal M.F.
      ).
      We have created permanent PC12 cell lines in which GSH can be inducibly down-regulated by expressing antisense rat γ-GCS heavy and light subunit cDNAs in a reverse doxycycline (dox)-inducible system making decreases in GSH levels drug-dependent. These cells were used to explore the affects of lowering GSH levels on mitochondrial function in dopaminergic cells as a model for PD.

      EXPERIMENTAL PROCEDURES

       Materials

      Tissue culture supplies were obtained from Life Technologies, Inc. All other chemicals were obtained from Sigma unless otherwise stated.

       Creation of dox Inducible Anti-γ-GCS PC12 Cell Lines

      dox inducible anti-γ-GCS lines (anti-γ-GCS) were produced by transfecting PC12 cells with pUHD172–1neo regulatory plasmid using LipofectAMINE reagent® (Life Technologies, Inc.) (Fig. 1). Cells were grown for 15 days in medium containing 1 mg/ml geneticin (G418). Resulting drug-resistant colonies were transiently transfected with response plasmid containing lacZ (pBIG,CLONTECH) and treated with either 0 or 25 μg/ml dox followed by 5-bromo-4-chloro-3-indolyl β-d-galactopyranoside (X-gal) staining to identify rTta lines with low basal expression and high dox inducibility (data not shown). These lines were stably transfected with response plasmid (pBI,CLONTECH) containing antisense heavy and light rat γ-GCS cDNAs cloned via reverse transcriptase-polymerase chain reaction from adult rat brain tissue using primers generated from previously published sequences (
      • Huang C.S.
      • Chang L.S.
      • Anderson M.E.
      • Meister A.
      ,
      • Seelig G.F.
      • Meister A.
      ). Cells were grown in medium containing 1 mg/ml G418 and 200 μg/ml hygromycin for 15 days, and drug-resistant colonies were selected. Experiments were run on two to three separate anti-γ-GCS cell lines using cells containing rTta alone as a negative control.
      Figure thumbnail gr1
      Figure 1Schematic of the dox-inducible anti-γ-GCS system. The regulatory system used to generate dox inducible GSH depletion in PC12 consists of: 1) a “regulatory plasmid” containing a rTta protein composed of reverse tet repressor (rTet-R) fused to a herpes simplex virus (HSV) VP16 transcription activation domain, which is constitutively expressed from a cytomegalovirus (pCMV) immediate early promoter and 2) a “response plasmid” in which γ-GCS antisense heavy and antisense light subunit cDNAs were placed in separate multiple cloning sites on either side of a tet response element. The tet response element consists of seven tandem tet operator sequences (Tet-O) fused to two minimal cytomegalovirus promoters in opposite directions. The resulting rTta protein produced from the regulatory plasmid only binds and activates expression of the antisense γ-GCS light and heavy subunit mRNAs in the presence of tetracycline or its lipophilic derivatives, e.g. dox. Anti-γ-GCS cells containing both plasmids were selected on the basis of resistance to both neomycin and hygromycin; rTta control cells were selected on the basis of resistance to neomycin alone.

       Measurement of γ-GCS Heavy and Light Subunit Protein Levels by Western Blot Analysis

      Western blot analysis of cell extracts from anti-γ-GCS and rTta cells grown in the presence of increasing concentrations of dox was performed as described previously (
      • Liu R.M.
      • Gao L.
      • Choi J.
      • Forman H.J.
      ).

       γ-Glutamyl-cysteine Synthetase Activity

      Cell homogenates resuspended in 100 mm Tris-Cl, pH 8.0, were used to assay γ-GCS activity levels as described previously (
      • Seelig G.F.
      • Meister A.
      ). Assays were run in the absence of α-aminobutyrate as blank. γ-GCS values were normalized/protein using reagent from Bio-Rad.

       GSH and Glutathione Disulfide (GSSG) Levels in Whole Cells and Mitochondria

      Cellular GSH and GSSG levels were measured by the method of Griffith (
      • Griffith O.W.
      ). Total GSH and GSSG was measured after the addition of 5′-5′-dithiobis(2-nitrobenzoic acid), 3-carboxy-4-nitrophenyl disulfide at an absorbance of 412 nm (
      • Griffith O.W.
      ). GSSG was selectively measured after assaying samples in which GSH is masked by pretreatment with 2-vinylpyridine. The difference between the two values gives the GSH levels in the cells. For measurement of mitochondrial GSH and GSSG levels, high quality mitochondria were isolated from anti-γ-GCS and control cells as described previously (
      • Trounce I.A.
      • Kim Y.L.
      • Jun A.S.
      • Wallace D.C.
      ). All values were normalized/protein and/or citrate synthetase activity.

       Cell Viability and Growth

      Dying versus live cells were visualized by incubating cells for 5 min in 0.2% trypan blue in sterile phosphate-buffered saline at room temperature and noting the presence of any blue (nonviable) cells versusphase bright cells under the light microscope (n = 10 fields examined/cell type).

       Measurement of Oxidative Stress via DCF Fluorescence in Whole Cells and Mitochondria

      Cells or isolated mitochondria were washed with phosphate-buffered saline and resuspended in 20 μm 2′-7′-dichlorofluorescein diacetate (Molecular Probes,) for 30 min, 37 °C. Cells were centrifuged at 12,000 ×g in a microfuge, and pellets were resuspended in 50 μm digitonin for 20 min, room temperature. The addition of digitonin alone did not contribute to DCF fluorescence. Following centrifugation, fluorescence was monitored in the supernatant in a Turner fluorimeter at an excitation wavelength of 488 nm and emission wavelength of 525 nm. DCF values were normalized/protein.

       Cellular ATP Levels

      Cells were harvested in a cold 1:1 mixture of 12% trichloroacetic acid and 0.2 msodium citrate, pH 7.0, and centrifuged, and the supernatant was used for assay of total cellular ATP levels by the method of Mattsonet al. (
      • Mattson M.P.
      • Zhang Y.
      • Bose S.
      ). ATP values were normalized/protein.

       Pyruvate-dependent 5′,5-Dithiobis(2-nitrobenzoic acid) (MTT) Reduction in Isolated Mitochondria

      MTT reduction in isolated mitochondria was measured using the method of Berridge and Tan (
      • Berridge M.V.
      • Tan A.S.
      ). MTT values were normalized/protein and reported as percent control (
      • Cohen G.
      • Rafay F.
      • Natasa K.
      ).

       Mitochondrial Respiration

      Oxygen consumption was monitored in isolated mitochondria using a biological oxygen monitor (YSI Model 5300, YSI Inc.). Rate of oxygen consumption was measured after the addition of 7 mm pyruvate/malate and 125 nmol of ADP as described previously (
      • Cohen G.
      • Rafay F.
      • Natasa K.
      ,
      • Ku H.H.
      • Sohal R.S.
      ).

       Activities of Mitochondrial Complexes

      Measurement of all mitochondrial enzyme complex activities were done according to the method of Trounce et al. (
      • Trounce I.A.
      • Kim Y.L.
      • Jun A.S.
      • Wallace D.C.
      ). Isolated mitochondria were preincubated in either the presence or absence of the thiol reductant DTT (4 mm, 35 min) prior to assay.

       Immunoprecipitation

      Immunoprecipitation was performed on isolated mitochondrial fractions under nondenaturing conditions with an antibody generated against the ND1 subunit of NADH dehydrogenase using the IMMUNOcatcher kit from CytoSignal Research Products (Irvine, CA) according the manufacturer's instructions. The ND1 polyclonal antibody was raised against a synthetic peptide representing amino acid residues 33–43 of the ND1 subunit of complex I (
      • Pettus E.H.
      • Cottrell B.
      • Wallace D.
      • Greenamyre J.T.
      ). Western blot analysis was performed as described previously to confirm the presence of immunoprecipitated protein.

       Measurement of Reduced Protein Sulfhydryl Residues

      Amounts of reduced sulfhydryl groups were measured in both whole mitochondrial and immunoprecipitated NADH dehydrogenase subunit-containing fractions by the method of Habeeb (
      • Habeeb A.F.S.A.
      ). Samples treated with 50 mm N-ethylmaleimide were used as a blank.

       Statistical Analysis

      Biochemical data are given as mean ± SD, and significance testing was performed using ANOVA (analysis of variance).

      RESULTS

      Generation of PC12 cells with inducibly reduced GSH levels. Using a reverse doxycycline inducible system, we have generated PC12 cell lines in which activity levels of γ-GCS, the rate-limiting enzyme in GSH synthesis, can be inducibly down-regulated via addition of dox (Fig. 1). Western blot analysis of anti-γ-GCS cells demonstrated that increasing concentrations of dox resulted in a dose-dependent decrease in levels of both heavy and light subunit γ-GCS protein in these cells (Fig.2, A and B). No corresponding decrease in either heavy or light γ-GCS subunit protein was seen in rTta control cells treated with dox at the same concentrations (data not shown). Levels of γ-GCS activity were also found to decrease in antisense γ-GCS containing cell lines in a dose-dependent manner following treatment with dox (Fig.3 A). In contrast, control rTta cells showed no significant changes in γ-GCS activity. dox-dependent decreases in γ-GCS levels were found to vary proportionally with time up to at least 72 h following initial drug induction while no change was observed in γ-GCS enzyme levels in the rTta control cells (Fig. 3 B). Cells containing only rTta plasmid showed no significant changes in γ-GCS levels for cultures incubated for the same time periods.
      Figure thumbnail gr2
      Figure 2Western blot analysis of γ-GCS heavy and light subunit protein levels in dox-treated anti-γ-GCS cell lines. A, heavy subunit protein levels after growth of cells in dox concentrations of 0, 20, 40, 60, and 80 μg/ml for 24 h.B, light subunit protein levels after growth of cells in dox concentrations of 0, 20, 40, 60, and 80 μg/ml for 24 h. In contrast, rTta cells showed no change in γ-GCS protein levels following dox addition (data not shown).
      Figure thumbnail gr3
      Figure 3Measurement of GCS activity and cellular and mitochondrial glutathione levels in dox-treated anti-γ-GCS cells. A, dox concentration curve of γ-GCS activity in anti-γ-GCSversus control rTta cells. Cells were treated with 0, 20, 40, and 60 μg/ml dox for 24 h (100% value is equivalent to 2.5 ± 0.3 nmol/min/mg of protein). B, time curve of γ-GCS activity in antisense γ-GCS versus rTta control cells. Cells were treated with 25 μg/ml dox and assayed for γ-GCS activity after 24, 48, and 72 h (100% value, 2.5 ± 0.2 nmol/min/mg of protein). C, dox concentration curve of GSSG and GSH levels in anti-γ-GCS versus control rTta cells. Cells were treated for 24 h in 0, 20, 40, and 60 μg/ml dox (100% GSH value, 20.0 ± 2.0 nmol/mg protein, 100% GSSG value, 0.8 ± 0.04 nmol/mg protein). *, p < 0.01versus control at 0 μg/ml. D, mitochondrial glutathione levels in anti-γ-GCS cells treated for 24 h in 0versus 25 μg/ml dox (100% GSH value, 3.2 ± 0.4 nmol/mg protein, 100% GSSG value, 0.12 ± 0.03 nmol/mg protein). *, p < 0.01 versus control at 0 μg/ml.
      The dependence of the levels of total cellular glutathione (GSH and its oxidized form, GSSG) on dox concentration were seen to parallel γ-GCS activities, decreasing with decreasing enzyme activity (Fig.3 C). Treatment with 25 μg/ml dox for 24 h caused a decrease in GSH levels of ∼50% in the antisense γ-GCS cell lines, a similar decrease to that observed in the SN of early PD brains (
      • Reiderer P.
      • Sofic E.
      • Rausch W.D.
      • Schmidt B.
      • Reynolds G.
      • Jellinger K.
      • Youdim M.B.H.
      ,
      • Perry T.L.
      • Yong V.W.
      ,
      • Perry T.L.
      • Godin D.V.
      • Hansen S.
      ,
      • Bannon M.J.
      • Goedert M.
      • Wolff O.H.
      ,
      • Sian J.
      • Dexter D.T.
      • Lees A.
      • Daniel S.
      • Agid Y.
      • Javoy-Agid F.
      • Jenner P.
      • Marsden C.D.
      ). This dosage was used for all subsequent studies unless otherwise noted.
      Acute decreases in GSH levels had no effect on cell viability or growth as measured by trypan blue staining (data not shown); only rare blue cells were noted in either the rTta or antisense γ-GCS cells following a 24-h treatment with 25 μg/ml dox.

       Reduced GSH Levels Result in Increased Oxidative Stress

      As GSH is an important redox regulator, cellular levels of ROS and related species were estimated to examine the effects of reduced GSH levels on production of oxidative stress (Fig. 4). Lowering GSH levels was found to cause a significant increase in both cellular and mitochondrial ROS values in the anti-γ-GCS cell lines. rTta lines showed no change (data not shown).
      Figure thumbnail gr4
      Figure 4Effects of glutathione depletion in PC12 on cellular and mitochondrial ROS levels. Estimation of whole cell and mitochondrial ROS by DCF fluorescence after treatment of anti-γ-GCS cells with 25 μg/ml dox for 24 h. rTta cells show no change in ROS levels following dox addition (data not shown). *,p < 0.05 versus control at 0 μg/ml. Digitonin alone had no effect on DCF fluorescence levels.

       Reduced GSH Levels Result in Decreases in Mitochondrial Function because of Specific Inhibition of Complex I via Thiol Oxidation

      GSH is known to be synthesized in the cytoplasm and to enter the mitochondria via an energy-dependent transport-mediated process (
      • Meister A.
      ). Decreases in cellular GSH in the anti-γ-GCS cells following dox induction resulted in a significant decrease in both mitochondrial GSH and GSSG levels (Fig.3 D).
      The effect of decreased mitochondrial GSH levels on mitochondrial performance was quantified by measurement of cellular ATP levels (Fig.5 A) as well as mitochondrial pyruvate-dependent MTT reduction (Fig. 5 B) and oxygen consumption (Fig. 5, C and D). All were found to be significantly reduced after treatment with dox. These results suggest that decreasing GSH levels have a profound affect on mitochondrial function.
      Figure thumbnail gr5
      Figure 5Assays of mitochondrial function following GSH depletion in PC12. A, cellular ATP levels (100% value, 15.0 ± 1.0 nmol/min/mg); B,pyruvate-dependent MTT reduction assay in mitochondria isolated from anti-γ-GCS cells treated with 0 versus 25 μg/ml dox for 24 h. *, p < 0.01versus control and **, p < 0.001versus control at 0 μg/ml dox. C, state 3 and 4 respiration rates in mitochondria isolated from anti-γ-GCS cells following treatment with 0 versus 25 μg/ml dox for 24 h; rTta cells showed no change in oxygen consumption rate following dox addition (data not shown). D, state 3 respiration rates quantified from graph C (100% value, 15.0 ± 2.0 ng of atom 02/min/mg of mitochondrial protein). *, p< 0.01 versus anti-γ-GCS at 0 μg/ml dox.
      PD appears to involve selective decreases in mitochondrial complex I activity (
      • Beal M.F.
      ,
      • Jenner P.
      ,
      • Hillered L.
      • Chan P.H
      ,
      • Haas R.H.
      • Nasirian F.
      • Nakano K.
      • Ward D.
      • Pay M.
      • Hill R.
      • Shults C.W.
      ). MTT reduction and state 3 respiration assayed using the complex I substrate pyruvate were both found to be decreased upon GSH depletion of PC12. To see whether the effects of GSH depletion were specific to complex I or more general, we compared the activities of complex I versus complexes II-III and IV in isolated mitochondria from dox-treated versus untreated cells. Lowering of GSH levels in these cells resulted in a significant decrease in complex I activity, but interestingly no significant losses were seen in either complex II-III or IV (Fig.6). The addition of the thiol reducing agent DTT was found to restore complex I activity to levels comparable with those found in corresponding controls, indicating that the inhibition of enzyme activity is likely because of oxidation of sulfhydryl groups within the enzyme complex. DTT had no affect on either complex I activity alone or on II-III and IV activities with or without dox addition. Levels of reduced protein sulfhydryl residues were decreased in a dox-dependent manner in mitochondria isolated from GSH-depleted cells suggesting that decreases in cellular GSH levels results in increased thiol oxidation of mitochondrial proteins (Fig.7 A).
      Figure thumbnail gr6
      Figure 6Selective inhibition of complex I (NADH dehydrogenase) activity following GSH depletion in PC12 can be reversed by treatment with the thiol reducing agent, DTT. Measurement of complex I (100% value, 120.0 ± 4.0 nmol/min/mg of mitochondrial protein), complex II-III (100% value, 130.0 ± 3.0 nmol/min/mg of protein), and complex IV (100% value, 500 nmol/min/mg protein) activities in anti-γ-GCS cells after treatment with 0versus 25 μg/ml dox for 24 h. All mitochondrial enzyme activities were measured at 25 μg/ml dox in the absence and presence of 4 mm DTT. *, p < 0.01versus control at 0 μg/ml. DTT has no effect on complex activities at 0 μg/ml dox (data not shown).
      Figure thumbnail gr7
      Figure 7Measurement of reduced sulfhydryl groups before and after ND1 immunoprecipitation. A, levels of reduced sulfhydryl residues in mitochondria isolated from anti-γ-GCS cells after treatment with 0, 25, and 50 μg/ml dox for 24 h (100% value, 90 nmol of SH/mg of protein). B,Western analysis of NADH dehydrogenase immunoprecipitates isolated from mitochondria of anti-γ-GCS cells after growth in 0 versus25 μg/ml dox for 24 h; band is the 33-kDa ND1 subunit.C, levels of reduced sulfhydryl residues in immunoprecipitates isolated from mitochondria of anti-γ-GCS cells after growth in 0 versus 25 μg/ml dox for 24 h.*,p < 0.01 versus control at 0 μg/ml.
      Nondenaturing immunoprecipitation performed using antibodies against NADH dehydrogenase (ND1 subunit), the active enzymatic component of complex I, followed by subsequent measurement of reduced protein sulfhydryls in the immunoprecipitated fractions revealed a significant decrease in their levels in this protein following GSH depletion (Fig.7 C). This decrease in reduced thiol groups was in the absence of any GSH depletion-mediated loss in immunoprecipitated protein levels (Fig. 7 B).

      DISCUSSION

      Early depletions in nigral GSH levels observed in the Parkinsonian brain are not explainable by increased oxidation of GSH to GSSG. GSH losses have been suggested to be because of increased activity of the enzyme γ-glutamyltranspeptidase resulting in increased removal of both GSH and GSSG from cells, although this has yet to be definitely proven (
      • Sian J.
      • Dexter D.T.
      • Lees A.J.
      • Daniel S.
      • Jenner P.
      • Marsden C.D.
      ). To explore the effects of a depletion in GSH on dopaminergic cells like those in the SN, we constructed a model in which levels of the rate-limiting enzyme in glutathione synthesis, γGCS, are inducibly depleted. Although GCS activity levels do not appear to be specifically impaired in sporadic cases of PD (
      • Sian J.
      • Dexter D.T.
      • Lees A.J.
      • Daniel S.
      • Jenner P.
      • Marsden C.D.
      ), the net effect of our genetic manipulation mimics that which is seen in the Parkinsonian brain, i.e. a decrease in GSH levels without corresponding increases in GSSG levels. Using this model we have demonstrated that lowering GSH levels in PC12 appears to elicit a selective inhibition of mitochondrial complex I activity leading to decreased mitochondrial function. Decreased mitochondrial activities following glutathione depletion in PC12 includes a loss in cellular ATP levels. Although acute depletion of GSH appeared to have no effect on overall cell viability or growth after 24 h, ATP is required for various cellular activities including the synthesis of GSH itself and therefore prolonged decreases in ATP levels would be expected to eventually result in decreased cell viability such as seen in PD.
      Selective reductions in GSH levels, which precede losses in mitochondrial complex I activity, have been reported to occur not only in Parkinson's disease but also in toxin models associated with it such as 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (
      • Perry T.L.
      • Godin D.V.
      • Hansen S.
      ,
      • Hallman H.
      • Lange J.
      • Olson L.
      • Strömberg I.
      • Jonsson G.
      ,
      • Hung H.-C.
      • Lee E.H.Y.
      ,
      • Sriram K.
      • Shankar S.K.
      • Boyd M.R.
      • Ravindranath V.
      ). Whether the inhibition of complex I activity and subsequent decreases in ATP following 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine administration can be totally accounted for by decreases in GSH levels is unclear as complex I also appears to be inhibited by direct interaction with 1-methyl-4-phenyl pyridium (MPP+) formed during monoamine oxidase-B-mediated oxidation of 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (
      • Jha N.
      • Andersen J.K.
      ). However decreasing GSH levels by treatment with the pharmacological agent buthionine sulfoxamine has been shown to potentiate the neurodegenerative effects of 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine in the SN and to alone cause sublethal damage to the nigrostriatal system (
      • Wüllner U.
      • Löschmann P.A.
      • Schulz J.B.
      • Schmid A.
      • Dringen R.
      • Eblen F.
      • Turski L.
      • Klockgether T.
      ,
      • Andersen J.K.
      • Mo J.-Q.
      • Koek L.L.
      • Hom D.G.
      • McNeill T.H.
      ).
      Our studies suggest that complex I activity is especially susceptible to decreases in GSH levels. Previous studies have shown that complex I activity in bovine heart submitochondrial particles is particularly effected by oxidative stress and that ROS generated by the hypoxanthine-xanthine system selectively affects utilization of oxygen by NAD-linked substrates in rat brain mitochondria (
      • Haas R.H.
      • Nasirian F.
      • Nakano K.
      • Ward D.
      • Pay M.
      • Hill R.
      • Shults C.W.
      ,
      • Zhang Y.
      • Marcillat O.
      • Giulivi C.
      • Ernster L.
      • Davies K.J.
      ). Complex I is in fact considered to be one of the most severely affected by age-related increases in oxidative stress (
      • Lenaz G.
      • Bovina C.
      • Castelluccio C.
      • Fato R.
      • Formiggini G.
      • Genova M.L.
      • Marchetti M.
      • Pich M.M.
      • Pallotti F.
      • Parenti-Castelli G.
      • Biagini G.
      ). In synaptic mitochondria, complex I exerts a major control over oxidative phosphorylation such that at 25% inhibition, energy metabolism is disturbed resulting in decreased ATP synthesis; complex III and IV inhibition in the range of 70–80% is required to exert similar effects (
      • Davey G.P.
      • Peuchen S.
      • Clark J.B.
      ).
      GSH is known to protect proteins from oxidation by conjugating with oxidized thiol groups to form protein-SS-GSH mixed disulfides, which can then be re-reduced to protein and GSH by glutathione reductase, thioredoxin, or protein disulfide isomerase (
      • Ravindranath V.
      • Reed D.J.
      ,
      • Jung C.H.
      • Thomas J.A.
      ). GSH is the major cellular component involved in maintaining protein sulfhydryl groups in their reduced state and much emphasis has been placed lately on its role in redox regulation as a mechanism for controlling activities of various thiol-dependent enzymes including those involved in metabolic regulation (for review, see Refs.
      • Ziegler D.M.
      and
      • Del Corso A.
      • Cappiello M.
      • Mura A.
      ). Furthermore, a recent study by Sriram et al. (
      • Sriram K.
      • Shankar S.K.
      • Boyd M.R.
      • Ravindranath V.
      ) demonstrated that administration of the excitotoxic compoundl-β-N-oxalylamino-l-alaninein vitro and in vivo resulted in depletion of GSH leading to inactivation of mitochondrial complex I activity via thiol oxidation. This appears to be the cause for the characteristic mitochondrial dysfunction and subsequent corticospinal neurodegeneration mediated by this neurotoxin; both could be prevented by treatment with antioxidant thiol agents. Previous experiments by Cohen et al. (
      • Cohen G.
      • Rafay F.
      • Natasa K.
      ) suggest that oxidation of protein sulfhydryl residues can result in inhibition of mitochondrial electron transport when the complex I substrate pyruvate is used as the electron donor. Activity of NADH dehydrogenase, the enzymatic component of complex I, appears to be thiol-regulated (
      • Shivakumar B.R.
      • Kolluri S.V.
      • Ravindranath V.
      ,
      • Balijepalli S.
      • Boyd M.R.
      • Ravindranath V.
      ). Our data demonstrate that not only is the decrease in NADH dehydrogenase activity elicited by GSH depletion of dopamine-containing cells restored by treatment with the thiol-reducing agent DTT but also that lowering levels of GSH results in the biochemical oxidation of protein sulfhydryl groups contained within this enzyme. Taken together, these data suggest that GSH depletion may lead to oxidation of protein sulfhydryl residues in the enzyme important for its function resulting in profound effects on subsequent mitochondrial performance. Total glutathione depletion (GSH+GSSG) cannot cause oxidation of a sulfhydryl itself, but lowering glutathione levels can cause a rise in steady-state levels of H202 (as we see in our cells), which in turn can result in oxidation of vicinal dithiols to a dithiol. The reversibility in the loss of complex I activity with DTT suggests that this is likely a mixed disulfide or intramolecular disulfide formed from vicinal thiols rather than the formation of a sulfinic (−S02H) or sulfonic (S03H) acid. Mixed disulfides can be formed through a sulfenic (−SOH) intermediate produced by reaction of H202 with a thiolate (-S-), followed by reaction with GSH. Intramolecular disulfides can be formed by the same route where the vicinal thiol displaces the GS- or by direct reaction of the vicinal thiol with an −SOH intermediate.
      Based on our data, we propose that the early depletion of GSH in dopaminergic neurons of the SN in the PD brain may be responsible for selective inhibition of complex I activity and concomitant loss of mitochondrial function, which have been associated with the neurodegeneration characteristic of the disease. A recent pilot study examining the effects of GSH administration in a small group of untreated PD patients report that daily intravenous delivery of the tripeptide for the period of a month resulted in a significant improvement in disability (
      • Sechi G.
      • Deledda M.G.
      • Bua G.
      • Satta W.G.
      • Deina G.A.
      • Pes G.M.
      • Rosati G.
      ). Whether such treatment is effective in actually altering brain levels of GSH and having lasting effects that can act to retard the progress of the disease is unclear; however, our data suggest that maintaining thiol homeostasis may be critical for protecting dopaminergic neurons of the SN against neurodegeneration and that thiol reductants may be therapeutic in this disorder.

      Acknowledgments

      We thank Dr. Herman Bujard, University of Heidelberg, for the gift of regulatory plasmid, Drs. Slobodanka Vukosavic and Serge Prezeborski, Columbia University, for immunoprecipitation protocols and advise, and Dr. Tom McNeill, University of Southern California, for use of his oxygen monitor.

      REFERENCES

        • Beal M.F.
        Neurobiology. 1992; 31: 119-130
        • Reiderer P.
        • Sofic E.
        • Rausch W.D.
        • Schmidt B.
        • Reynolds G.
        • Jellinger K.
        • Youdim M.B.H.
        J. Neurochem. 1989; 52: 515-520
        • Dexter D.T.
        • Holley A.E.
        • Flitter W.D.
        • Slater T.F.
        • Wells F.R.
        • Daniel S.E.
        • Lees A.J.
        • Jenner P.
        • Marsden C.D.
        Movement Disorders. 1994; 9: 92-97
        • Yoritaka A.
        • Hattori N.
        • Uchidak K.
        • Tanaka M.
        • Stadtman E.R.
        • Mizuno Y.
        Proc. Natl. Acad. Sci. U. S. A. 1996; 93: 2696-2701
        • Perry T.L.
        • Yong V.W.
        Neurosci. Lett. 1986; 67: 269-274
        • Jenner P.
        Acta Neurol. Scand. 1993; 87: 1913-1917
        • Pearce R.K.
        • Owen A.
        • Daniel S.
        • Jenner P.
        • Marsden C.D.
        J. Neural Transm. 1998; 104: 661-677
        • Meister A.
        J. Biol. Chem. 1988; 263: 17205-17208
        • Huang C.S.
        • Chang L.S.
        • Anderson M.E.
        • Meister A.
        J. Biol. Chem. 1993; 268: 19675-19680
        • Jain A.
        • Martensson J.
        • Stole E.
        • Auld P.A.
        • Meister A.
        Proc. Natl. Acad. Sci. U. S. A. 1991; 88: 1913-1917
        • Seelig G.F.
        • Meister A.
        Methods Enzymol. 1985; 113: 379-390
        • Liu R.M.
        • Gao L.
        • Choi J.
        • Forman H.J.
        Am. J. Physiol. 1998; 275: L861-L869
        • Griffith O.W.
        Anal. Biochem. 1980; 106: 207-212
        • Trounce I.A.
        • Kim Y.L.
        • Jun A.S.
        • Wallace D.C.
        Methods Enzymol. 1996; 264: 484-509
        • Mattson M.P.
        • Zhang Y.
        • Bose S.
        Exp. Neurol. 1993; 121: 1-13
        • Berridge M.V.
        • Tan A.S.
        Arch. Biochem. Biophys. 1993; 303: 474-482
        • Cohen G.
        • Rafay F.
        • Natasa K.
        Proc. Natl. Acad. Sci. U. S. A. 1997; 94: 4890-4894
        • Ku H.H.
        • Sohal R.S.
        Mech. Ageing Dev. 1993; 72: 67-76
        • Pettus E.H.
        • Cottrell B.
        • Wallace D.
        • Greenamyre J.T.
        Soc. Neurosci. Abstr. 1997; 23: 1640
        • Habeeb A.F.S.A.
        Methods Enzymol. 1972; 25: 457-464
        • Perry T.L.
        • Godin D.V.
        • Hansen S.
        Neurosci. Lett. 1982; 33: 305-310
        • Bannon M.J.
        • Goedert M.
        • Wolff O.H.
        Biochem. Pharmacol. 1984; 33: 2697-2698
        • Sian J.
        • Dexter D.T.
        • Lees A.
        • Daniel S.
        • Agid Y.
        • Javoy-Agid F.
        • Jenner P.
        • Marsden C.D.
        Ann. Neurol. 1994; 36: 348-355
        • Hillered L.
        • Chan P.H
        J. Neurosci. Res. 1988; 19: 94-100
        • Haas R.H.
        • Nasirian F.
        • Nakano K.
        • Ward D.
        • Pay M.
        • Hill R.
        • Shults C.W.
        Ann. Neurol. 1995; 37: 714-722
        • Sian J.
        • Dexter D.T.
        • Lees A.J.
        • Daniel S.
        • Jenner P.
        • Marsden C.D.
        Ann. Neurol. 1994; 36: 356-361
        • Hallman H.
        • Lange J.
        • Olson L.
        • Strömberg I.
        • Jonsson G.
        J. Neurochem. 1985; 44: 117-127
        • Hung H.-C.
        • Lee E.H.Y.
        Free Radic. Mol. Biol. 1997; 24: 76-84
        • Sriram K.
        • Shankar S.K.
        • Boyd M.R.
        • Ravindranath V.
        J. Neurosci. 1998; 18: 10287-10296
        • Jha N.
        • Andersen J.K.
        Pandalai S.G. Recent Research Developments in Neurochemistry. Research Signpost, Trivandrum, India1999: 99-108
        • Wüllner U.
        • Löschmann P.A.
        • Schulz J.B.
        • Schmid A.
        • Dringen R.
        • Eblen F.
        • Turski L.
        • Klockgether T.
        Neuroreport. 1996; 7: 921-923
        • Andersen J.K.
        • Mo J.-Q.
        • Koek L.L.
        • Hom D.G.
        • McNeill T.H.
        J. Neurochem. 1996; 67: 2164-2171
        • Zhang Y.
        • Marcillat O.
        • Giulivi C.
        • Ernster L.
        • Davies K.J.
        J. Biol. Chem. 1990; 265: 16330-16336
        • Lenaz G.
        • Bovina C.
        • Castelluccio C.
        • Fato R.
        • Formiggini G.
        • Genova M.L.
        • Marchetti M.
        • Pich M.M.
        • Pallotti F.
        • Parenti-Castelli G.
        • Biagini G.
        Mol. Cell. Biochem. 1997; 174: 329-333
        • Davey G.P.
        • Peuchen S.
        • Clark J.B.
        J. Biol. Chem. 1998; 273: 12753-12757
        • Ravindranath V.
        • Reed D.J.
        Biochem. Biophys. Res. Comm. 1990; 169: 1075-1079
        • Jung C.H.
        • Thomas J.A.
        Arch. Biochem. Biophys. 1996; 335: 61-72
        • Ziegler D.M.
        Ann. Rev. Biochem. 1985; 54: 305-329
        • Del Corso A.
        • Cappiello M.
        • Mura A.
        Int. J. Biochem. 1994; 26: 745-750
        • Shivakumar B.R.
        • Kolluri S.V.
        • Ravindranath V.
        J. Pharmacol. Exp. Ther. 1995; 274: 1167-1173
        • Balijepalli S.
        • Boyd M.R.
        • Ravindranath V.
        Neuropharmacology. 1999; 38: 567-577
        • Sechi G.
        • Deledda M.G.
        • Bua G.
        • Satta W.G.
        • Deina G.A.
        • Pes G.M.
        • Rosati G.
        Prog. Neuropsychopharmacol. Biol. Psychiatry. 1996; 20: 1159-1170