HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

J. Biol. Chem., Vol. 277, Issue 19, 17048-17056, May 10, 2002

This Article Abstract Full Text (PDF) All Versions of this Article: 277/19/17048    most recent M110797200v1 Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Lin, T.-K. Articles by Murphy, M. P. Search for Related Content PubMed PubMed Citation Articles by Lin, T.-K. Articles by Murphy, M. P. Social Bookmarking                      What's this?

Specific Modification of Mitochondrial Protein Thiols in Response to Oxidative Stress

A PROTEOMICS APPROACH^*

Tsu-Kung Lin^§, Gillian Hughes, Aleksandra Muratovska^¶, Frances H. Blaikie, Paul S. Brookes^**, Victor Darley-Usmar^**, Robin A. J. Smith, and Michael P. Murphy^§§

From the  Departments of Biochemistry and  Chemistry, University of Otago, Box 56, Dunedin, New Zealand, the ^§ Department of Neurology, Chang-Gung Memorial Hospital, Kaohsiung, Taiwan 833, the ^** Department of Pathology and Center for Free Radical Biology, University of Alabama, Birmingham, Alabama 35294, and  MRC-Dunn Human Nutrition Unit, MRC-Wellcome Trust Building, Hills Road, Cambridge CB2 2XY, United Kingdom

Received for publication, November 9, 2001, and in revised form, January 23, 2002

	ABSTRACT

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Mitochondria play a central role in redox-linked processes in the cell through mechanisms that are thought to involve modification of specific protein thiols, but this has proved difficult to assess. In particular, specific labeling and quantitation of mitochondrial protein cysteine residues have not been achieved due to the lack of reagents available that can be applied to the intact organelle or cell. To overcome these problems we have used a combination of mitochondrial proteomics and targeted labeling of mitochondrial thiols using a novel compound, (4-iodobutyl)triphenylphosphonium (IBTP). This lipophilic cation is accumulated by mitochondria and yields stable thioether adducts in a thiol-specific reaction. The selective uptake into mitochondria, due to the large membrane potential across the inner membrane, and the high pH of the matrix results in specific labeling of mitochondrial protein thiols by IBTP. Individual mitochondrial proteins that changed thiol redox state following oxidative stress could then be identified by their decreased reaction with IBTP and isolated by two-dimensional electrophoresis. We demonstrate the selectivity of IBTP labeling and use it to show that glutathione oxidation and exposure to an S-nitrosothiol or to peroxynitrite cause extensive redox changes to mitochondrial thiol proteins. In conjunction with blue native gel electrophoresis, we used IBTP labeling to demonstrate that thiols are exposed on the matrix faces of respiratory Complexes I, II, and IV. This novel approach enables measurement of the thiol redox state of individual mitochondrial proteins during oxidative stress and cell death. In addition the methodology has the potential to identify novel redox-dependent modulation of mitochondrial proteins.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Changes in the thiol redox state of mitochondrial proteins are significant in a number of cellular processes including the permeability transition, cell death due to calcium loading and oxidative stress, the response of cells to nitric oxide, tumor necrosis factor signaling, commitment to apoptosis, and in regulating respiratory chain function (1-9). However the detailed mechanisms and the proteins involved are uncertain. This is partly because of the technical challenges presented by determining thiol modifications of proteins in general and the difficulties inherent in mitochondrial proteomics. Potential protein thiol alterations include formation of mixed disulfides or internal disulfides from vicinal dithiols, S-nitrosation, and the formation of higher oxidation states (10-15). The differential reactivity of individual protein thiols and the range of lifetimes of altered redox states can act as signal sensors or transducers to influence mitochondrial function (13-17). Nitric oxide may be a particularly important regulator of mitochondrial protein thiols because it diffuses easily into mitochondria and partitions selectively into the lipid bilayer where it can modify otherwise inaccessible thiols (18-20). Modification of protein thiols by nitric oxide most likely occurs through its nitrosating derivatives peroxynitrite or ^·NO₂, and nitrosated protein thiols can go on to form mixed or internal disulfides (12). Finally, many reactive oxygen species oxidize thiols directly to thiyl radicals that can then form disulfides and higher thiol oxidation states (21). In all cases the extent of the redox changes to mitochondrial thiol proteins, and the proteins affected have not been determined at the molecular level.

To identify those mitochondrial thiol proteins that change redox state during oxidative damage or cell death, we first set out to label unmodified mitochondrial protein thiols selectively within cells. Changes in protein thiol redox state would then manifest themselves by decreased labeling. To achieve this, advantage was taken of the large membrane potential across the inner membrane that causes lipophilic cations to accumulate within mitochondria (22, 23), and the lipophilic cation (4-iodobutyl)triphenylphosphonium (IBTP)¹ was synthesized. It is predicted that IBTP will accumulate 5-10-fold in the cytoplasm driven by the plasma membrane potential and should then be further concentrated several hundred-fold within mitochondria (Fig. 1) (23). Inside mitochondria protein thiolates should displace the iodo functional group of IBTP to form stable phosphonium thioethers that can be identified by immunoblotting using antitriphenylphosphonium antiserum (Fig. 1). The pK_a values of protein thiols are typically 8-8.5; consequently thiols in mitochondria (pH ~8) are more reactive than in the cytosol (pH ~7.2), potentially further increasing the mitochondrial selectivity of IBTP labeling. These factors would enhance the selectivity of the modification of mitochondrial thiol proteins by IBTP, which may then be identified by two-dimensional immunoblotting. Modifications leading to changes in protein function that could be identified by this approach are formation of mixed disulfides, S-nitrosation, or oxidation. The approach taken here is to validate this approach in both cellular and organellar systems before the application to more complex physiological settings.

View larger version (19K): [in this window] [in a new window]   Fig. 1.   Selective labeling of mitochondrial thiol proteins. IBTP is taken up into the cytosol driven by the plasma membrane potential (plasma) and then further accumulated by mitochondria due to the mitochondrial membrane potential (mito). Inside mitochondria IBTP reacts with exposed protein thiols to form a stable thioether. Because the thiolate form of a protein cysteine residue is the species that reacts with IBTP, this reaction is faster at the higher pH of the mitochondrial matrix than in the cytosol.

Here the development of IBTP as a selective label of mitochondrial thiol proteins within cells and the intact organelle is reported under both normal conditions and with exposure to oxidants. A number of mitochondrial thiol proteins have been mapped as single spots by two-dimensional gel electrophoresis, highlighting the potential of this approach to assess changes in mitochondrial thiols in response to oxidative stress.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Chemicals-- IBTP was synthesized by reacting 1,4-diiodobutane (2.46 g; 7.9 mmol) with triphenylphosphine (0.423 g; 1.61 mmol) at 100 °C under argon in the dark for 1.5 h. The yellow solid was washed with diethyl ether (3 × 8 ml) and then dissolved in dichloromethane (5 ml) and precipitated with diethyl ether (3 × 50 ml); the solvents were decanted after each precipitation. Residual solvent was then removed under reduced pressure to give a pale yellow solid (0.518 g; 0.91 mmol; 57%) that was stored at 20 °C and protected from light. IBTP was readily soluble in ethanol giving an absorption spectrum with local maxima at 263, 268, and 275 nm and an extinction coefficient at 268 nm of 3000 M¹·cm¹, typical for alkyltriphenylphosphonium salts (24). For the elemental analysis,

¹H NMR in CDCl₃ 7.7-7.9 (15H, m, Ph₃CH₂CH₂CH₂CH₂I), 3.75-3.9 (2H, m, Ph₃CH₂CH₂CH₂CH₂I), 3.35 (2H, t, J = 6 Hz, Ph₃CH₂CH₂CH₂CH₂I), 2.28 (2H, quintet, J = 6.6 Hz, Ph₃CH₂CH₂CH₂CH₂I), 1.82 (2H, m, Ph₃CH₂CH₂CH₂CH₂I 1.7-2.3 (4H, m, Ph₃CH₂CH₂CH₂CH₂I) ppm. ³¹P NMR in CDCl₃ 25.2 ppm.

(4-Chlorobutyl)triphenylphosphonium (ClBTP) was synthesized by reacting 1-chloro-4-iodobutane (0.306 g, 1.4 mmol) with triphenylphosphine (0.133 g, 0.51 mmol) and stirring at 100 °C for 1.5 h under an argon atmosphere in the dark. The mixture was cooled to room temperature, and the glassy product washed with diethyl ether (3 × 5 ml). The residue was dissolved in dichloromethane (1 ml) and precipitated with diethyl ether (10 ml), and the solvents were removed by decantation. This precipitation process was repeated three more times, and the final solid material was dried under high vacuum at room temperature to give the product (0.159 g, 0.33 mmol, 65%), ¹H NMR (CDCl₃) 7.6-7.9 (15H, m, Ph₃CH₂CH₂CH₂CH₂Cl), 3.8-3.9 (2H, m, Ph₃CH₂CH₂CH₂CH₂Cl), 3.70 (2H, triplet, J = 6Hz, Ph₃CH₂CH₂CH₂CH₂Cl), 2.24 (2H, quintet, J = 6Hz, Ph₃CH₂CH₂CH₂CH₂Cl), 1.85 (2H, m, Ph₃CH₂CH₂CH₂CH₂Cl) ppm, ³¹P NMR (CDCl₃) 25.33 ppm. ClBTP was readily soluble in ethanol giving an absorption spectrum with local maxima at 263, 268, and 275 nm and an extinction coefficient at 268 nm of 3,000 M¹ cm¹, typical for alkyltriphenylphosphonium salts (24). (4-Bromobutyl)triphenylphosphonium (BrBTP) was from Aldrich.

Thiol Chemistry-- BSA (1.52 mM) in 10 mM potassium phosphate (pH 7.4) was reduced with 10 mM 2-mercaptoethanol followed by dialysis against the same buffer (25) and contained ~0.64 thiol/molecule as assessed by reaction with dithionitrobenzoic acid (DTNB; ₄₁₂ = 13,600 M¹ cm¹ (26)). To prepare IBTP-BSA for immunoblotting, reduced BSA (10 µM) was reacted with IBTP (50 µM) for 30 min at 37 °C, dissolved in SDS-PAGE loading buffer, and stored at 20 °C. Glutathione (0.5-2 mM) or reduced BSA (500 µM) was reacted with 2.5 mM IBTP at 37 °C in 80 mM sodium phosphate, pH 8.0 or 7.2. Samples were removed after various times, and free thiols were determined by reaction with DTNB as above (26).

Electrophoresis and Immunoblotting-- The thiolamine cross-linker m-maleimidobenzoic acid N-hydroxysuccinimide ester (0.75 mg) in 50 µl of Me₂SO was reacted with keyhole limpet hemocyanin (5 µg; KLH) in 0.5 ml of PBS, pH 6, at room temperature for 30 min. m-Maleimidobenzoic acid N-hydroxysuccinimide ester-KLH was purified on Sephadex G-25 and reacted with thiobutyltriphenylphosphonium (20 mg) (24) overnight at room temperature, and aliquots were stored at 20 °C. After taking samples of pre-immune serum, aliquots of TBTP-KLH were mixed with Freund's complete adjuvant (Sigma) by repeated passage through an 18-gauge syringe needle and injected into pairs of rabbits at multiple sites subcutaneously. This was followed by two booster injections of TBTP-KLH mixed with Freund's incomplete adjuvant (Sigma) and collection of the rabbit antiserum against triphenylphosphonium. For some experiments an IgG-enriched fraction was prepared using a protein A-agarose column (Affi-Gel; Bio-Rad).

SDS-PAGE gels (12.5% acrylamide) were run by standard procedures using a Bio-Rad Mini Protean system, transferred to 0.1 µm nitrocellulose using a Bio-Rad Mini Trans Blot system and blocked with 5% milk powder in TBS. Antibody binding was detected using 1:500-1:5,000 dilutions of antitriphenylphosphonium antiserum and 1:3,000-1:10,000 dilutions of secondary antibodies (Bio-Rad) against rabbit or mouse IgG conjugated to either alkaline phosphatase and visualized by chromogenic substrate or to horseradish peroxidase and visualized by enhanced chemiluminescence.

For two-dimensional electrophoresis, mitochondria or membrane/matrix fractions were loaded onto an Immobiline Drystrip IPG gel (Amersham Biosciences, pH range 3-10 or 4-7), and proteins were separated using the Amersham Biosciences IPGphor Isoelectric Focusing System following the manufacturer's recommendations. The strips were then incubated in 6 M urea, 2% SDS, 561 mM Tris-HCl, pH 8.8, 20% glycerol, 130 mM dithiothreitol for 15 min followed by incubation in 6 M urea, 2% SDS, 561 mM Tris-HCl, pH 8.8, 20% glycerol, 135 mM iodoacetamide for 15 min, and the proteins were separated in the second dimension by SDS-PAGE. Gels were either stained with Coomassie Blue or blotted onto 0.1 µm nitrocellulose using a Bio-Rad Semidry Blotting System and probed with antitriphenylphosphonium antiserum as above.

Blue native gel electrophoresis (BN-PAGE) was carried out on a 5-12% linear acrylamide gradient using either a mini gel or full sized gel format (27). Ferritin dimer (880 kDa) and catalase monomer and dimer (230/460 kDa) were used as molecular weight markers. The proteins were then transferred to nitrocellulose using a Bio-Rad Semidry Blotting System and probed with antitriphenylphosphonium antiserum as above or monoclonal antibodies against respiratory chain components. The following monoclonal antibodies were from Molecular Probes: Complex I (NADH-ubiquinone oxidoreductase) 39-kDa subunit, A-11140; Complex II (succinate-ubiquinol oxidoreductase) 70-kDa subunit, A-11142; and Complex IV (cytochrome c oxidase) subunit I, 57 kDa, A-6403. Separation in the second dimension after BN-PAGE was performed by excising a lane from a native gel, placing it on a Bio-Rad mini-gel electrophoresis glass plate, and denaturing the proteins by soaking in 1% SDS, 1% mercaptoethanol for 30 min. The electrophoresis cassette was then assembled and a three-layer (10, 8, and 5%) Tris-Tricine gel poured below the excised gel lane. Duplicate gels were run and either stained with Coomassie Blue or transferred to nitrocellulose and immunoblotted using antitriphenylphosphonium antiserum as described above.

Mitochondrial Incubations-- Rat liver mitochondria were prepared by homogenization followed by differential centrifugation in STE medium (250 mM sucrose, 5 mM Tris-HCl, 1 mM EGTA, pH 7.4) (28), and the protein concentration was determined by the biuret assay using BSA as a standard (29). Mitochondria were subfractionated into membrane and matrix-enriched fractions by pelleting followed by incubation on ice for 30 min with the detergent Lubrol (0.2 mg/mg protein) at 10 mg of protein/ml in KCl buffer followed by centrifugation (Airfuge at 24 pounds/square inch 100,000 × g for 10 min). The matrix-enriched supernatant fraction was retained and for some procedures concentrated by acetone precipitation (9 volumes of acetone at 20 °C for 30 min). The membrane fraction was washed in KCl buffer, and the protein concentration of both fractions was determined by the bicinchoninic acid assay using BSA as a standard (30).

To measure thiol content, mitochondria (1 mg of protein/ml) were incubated in KCl buffer (110 mM KCl, 10 mM Hepes, 1 mM EGTA, pH 7.2) containing 10 mM succinate and rotenone with various concentrations of IBTP at 30 °C for 15 min. To measure glutathione equivalents the mitochondria were pelleted, resuspended in 100 µl of 5-sulfosalicylic acid (5%), and glutathione equivalents quantitated by the recycling assay using glutathione as a standard (31). Total mitochondrial protein thiols were determined by precipitating protein with 10% trichloroacetic acid, resuspending the pellet in 2% SDS in 80 mM sodium phosphate, pH 8, and quantitating free thiols by reaction with DTNB (32). To measure free thiols exposed in non-denatured mitochondria, the mitochondria were pelleted, resuspended in 80 mM sodium phosphate, pH 8, and disrupted by freezing in dry ice/ethanol followed by thawing in a 30 °C water bath (3 times), and the thiol content was measured by reaction with DTNB. To measure membrane-associated free thiols in native mitochondria, a membrane fraction was isolated from the freeze-thawed mitochondria by centrifugation (13,000 × g for 10 min) followed by washing in KCl buffer.

To measure the effects of IBTP on respiration, mitochondria (2 mg of protein/ml) were incubated at 25 °C in a 3-ml oxygen electrode in 110 mM KCl, 10 mM Hepes, 1 mM EGTA, 1 mM potassium phosphate, pH 7.2, with either 10 mM succinate and rotenone or 5 mM each of glutamate and malate in the presence of various concentrations of IBTP. Coupled respiration was measured first, then phosphorylating respiration after addition of ADP (200 µM), and finally uncoupled respiration after addition of carbonylcyanide-p-trifluoromethoxyphenylhydrazone (FCCP; 200 nM). The membrane potential was measured by incubating mitochondria with 1 µM methyltriphenylphosphonium (TPMP) supplemented with [³H]TPMP (50 nCi·ml¹). After incubation the mitochondria were pelleted by centrifugation. The amount of TPMP in the pellet and supernatant was quantitated by scintillation counting, and the membrane potential was calculated, assuming a mitochondrial volume of 0.5 µl·mg protein¹ and that 60% of intramitochondrial TPMP was membrane-bound (33). Membrane potential data were the mean of triplicate determinations and were repeated on 2-3 independent mitochondrial preparations.

Mitoplasts were prepared from mitochondria by treatment with digitonin (0.125 mg of digitonin/mg of mitochondrial protein) on ice for 15 min followed by washing in STE supplemented with 0.1% fatty acid-free BSA (34). A fraction enriched for the outer membrane was prepared from the initial supernatant of the mitoplast preparation by centrifugation (144,000 × g for 20 min) followed by washing, and the supernatant from the first centrifugation was retained as the intermembrane space fraction (34).

Cell Incubations-- Human fibroblasts were grown in Dulbecco's modified Eagle's medium (DMEM) supplemented with 10% fetal calf serum (FCS) at 37 °C in a humidified atmosphere of 95% air, 5% CO₂. Protein was quantitated by the bicinchoninic acid assay using BSA as a standard (30). For immunocytochemistry human fibroblasts were plated onto 13-mm diameter glass coverslips overnight (35). Following incubation for 1 h at 37 °C with 1 µM IBTP ± 10 µM FCCP, cells were fixed with 4% paraformaldehyde in TBS for 30 min, washed with TBS, and incubated with 10% FCS, 0.1% Triton X-100/TBS (TBST) for 10 min. The IgG fraction of antitriphenylphosphonium antiserum (1:500) and anti-cytochrome oxidase (1:100, subunit I, mouse monoclonal, Molecular Probes) diluted in TBST were then added and incubated overnight at 4 °C. The IgG fraction of preimmune serum was used as a control. After washing with TBS (3 times for 5 min), the cells were incubated with appropriate fluorophore-conjugated secondary antibodies diluted in TBS (anti-rabbit IgG Oregon Green (1:100) or anti-mouse IgG Texas Red (1:400); Molecular Probes) for 15 min in the dark. Cells were then washed in TBS and mounted in diazabicyclooctane/polyvinyl alcohol medium on coverslips. Images were acquired using a Bio-Rad MRC 600 laser scanning confocal microscope using a Nikon Diaphot TMD inverted microscope and a 60×/NA 1.4 oil immersion Plan-Apochromat objective.

For immunogold electron microscopy, human fibroblasts in 24-well tissue culture plates were incubated with 200 µl of DMEM/FCS with 1 µM IBTP ± 10 µM FCCP for 4 h at 37 °C, then harvested with trypsin, pelleted, and washed in PBS. The pellet was warmed to 37 °C, and 200 µl of fixative (4% formaldehyde, 0.5% glutaraldehyde in 0.2 M cacodylate, pH 7.2, 6 mM CaCl₂) was added and replaced with fresh fixative after 30 s. After incubation at 37 °C for 1 h, the cell pellet was washed with 200 µl of 0.15 M cacodylate, pH 7.2, 3 mM CaCl₂ (3 × 10 min at room temperature), warmed to 45 °C, mixed with 2% agar in 0.1 M cacodylate, pH 7.2, and pelleted. After rinsing in PBS (3 times for 10 min), the pellets were dehydrated in ethanol, embedded in Lowicryl white (SPI Supplies) and polymerized under UV light at 0 °C. Ultrathin sections were mounted on nickel grids that were floated on drops of 0.1 M glycine in PBS for 10 min, then 1% BSA in PBS for 15 min, followed by antitriphenylphosphonium antiserum overnight at 4 °C (1:10 in 1% BSA, 1% Tween 20 in PBS). After washing in PBS (3 times for 5 min) and 1% BSA/PBS, sections were incubated with colloidal gold particles (10 nm) linked to either goat anti-rabbit IgG (1:5 in 1% Tween 20/PBS, Sigma) or extravidin (1:10 in 1% Tween 20/PBS, Sigma) for 1 h, followed by washing and fixation (2% glutaraldehyde for 15 min). The sections were then stained with osmium tetroxide/uranyl acetate/lead citrate and examined in a CM100 transmission electron microscope.

Fibroblasts were harvested by trypsin treatment, incubated at 5 × 10⁶ cells·ml¹ in DMEM without FCS at 37 °C for 10 min, and then IBTP (5 µM) was added, and 10 min later the cells were pelleted by centrifugation (3,000 × g for 5 min) and washed in PBS. For whole cell samples the pellet was suspended directly in 100 µl of loading buffer. For subfractionation cells were resuspended in 500 µl of STE, homogenized using a 5-ml Dounce homogenizer, and nuclei and unbroken cells were then pelleted (3,000 × g for 5 min). This pellet was washed in 500 µl of STE, and the combined supernatants were centrifuged (13,000 × g for 2 min) to give a mitochondria-enriched pellet and a cytosol-enriched supernatant. The mitochondrial pellet was washed in STE, and the supernatant was acetone-precipitated. Assays of the mitochondrial enzyme citrate synthase and of the cytosolic enzyme lactate dehydrogenase confirmed that this fractionation procedure was effective, with less than 5% lactate dehydrogenase activity in the mitochondrial fraction that contained about 40% of the total citrate synthase activity (36).

Jurkat cells were grown in RPMI supplemented with 10% FCS at 37 °C in a humidified atmosphere of 95% air, 5% CO₂. Cells (5 × 10⁶/ml) were incubated in RPMI, 1% FCS at 37 °C in medium supplemented with 10 mM Hepes. After incubation cells were pelleted by centrifugation (3,000 × g for 5 min) and washed in PBS, and the pellet was suspended directly in loading buffer. Bovine aortic endothelial cells were harvested from descending thoracic aortas and maintained in DMEM containing 1 g·l¹ glucose and 10% FCS at 37 °C in a humidified atmosphere of 95% air, 5% CO₂. Cells used in this study were between passages 5 and 10. For IBTP treatment 5 × 10⁵ cells were seeded in each well of 6-well plates in the growth medium and used when confluent. After incubation with IBTP the medium was removed, and the cells were suspended in loading buffer using a cell scraper. The toxicity of IBTP was estimated from the amount of lactate dehydrogenase released after 24 h of incubation with human osteosarcoma 143B cells (37).

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Mitochondrial Localization of IBTP within Cells-- Human fibroblasts were incubated with IBTP (1 µM) for 1 h, and the intracellular localization of IBTP was then determined by immunocytochemistry using laser scanning confocal microscopy (Fig. 2A). IBTP binding co-localized with the mitochondrial enzyme cytochrome oxidase, and there was no IBTP binding outside mitochondria (Fig. 2A, upper panel). The uptake and binding of IBTP were prevented by dissipating the mitochondrial membrane potential with the uncoupler FCCP (Fig. 2A, lower panel). Both these results are consistent with but not conclusive for a mitochondrial localization of IBTP. As a further test for mitochondrial localization, immunogold electron microscopy was also employed in fibroblasts incubated with IBTP. The diffuse higher density structure characteristic of mitochondria imaged in cells using this technique is evident, with the black dots indicating the binding of the gold-labeled antibody surrounding these structures again consistent with a mitochondrial localization of IBTP within cells (Fig. 2B). The specificity of the antitriphenylphosphonium serum was confirmed by its detection of IBTP-labeled BSA on an immunoblot and by the selective blocking of this reaction by preincubation of the antiserum with TPMP (Fig. 3A). IBTP only showed signs of cytoxicity at concentrations in excess of 25-50 µM (data not shown).

View larger version (69K): [in this window] [in a new window]   Fig. 2.   IBTP binding to mitochondrial thiol proteins within cells. A, laser scanning confocal fluorescence microscopy. Human fibroblasts were incubated with IBTP (1 µM) for 1 h, fixed, and probed with antitriphenylphosphonium antiserum and a monoclonal antibody against cytochrome oxidase. Fluorescent secondary antibodies were used to visualize IBTP binding (green), cytochrome oxidase (red), and their co-localization (yellow) (+). Scale bar, 20 µm. This experiment was repeated in the presence of FCCP (10 µM) (). Cells incubated with IBTP and then probed with preimmune serum and a fluorescent secondary antibody did not show any fluorescence. B, location of IBTP labeling determined by immunogold electron microscopy. Human fibroblasts were incubated with IBTP as for A and then fixed and incubated with antitriphenylphosphonium antiserum and with a secondary antibody linked to gold particles, prepared for electron microscopy and visualized (arrow, mitochondrion). There was no immunoreactivity in the cytoplasm or in the mitochondria of cells incubated with IBTP in the presence of FCCP (data not shown). Scale bar, 100 nm.

View larger version (41K): [in this window] [in a new window]   Fig. 3.   Immunoblots of mitochondrial thiol proteins labeled by IBTP within cells. A, immunodetection of protein-bound IBTP. Reduced BSA (10 mg/ml) was incubated for 30 min in KCl buffer, pH 8, at 30 °C, and then IBTP (250 µM) was added, and samples were removed at various times (5-60 min) and precipitated by acetone. BSA (1 µg) was then resolved by SDS-PAGE, transferred to nitrocellulose, probed with antitriphenylphosphonium antiserum that had been preincubated at 37 °C for 60 min, and visualized using a secondary antibody conjugated to alkaline phosphatase: +TPMP, BSA reacted with IBTP for 60 min and probed with antitriphenylphosphonium antiserum that had been preincubated with TPMP (100 µM) at 37 °C for 60 min. B, labeling of mitochondrial proteins within cells. Human fibroblasts were preincubated for 10 min, incubated for a further 15 min + IBTP (5 µM), and then pelleted and fractionated into mitochondria (mit) and cytosolic (cyt) fractions. IBTP-labeled proteins were separated by SDS-PAGE, detected by immunoblotting, and compared with intact cells (cell). Each lane was loaded with protein from 5 × 105 cells or a cell subfraction isolated from 2.5 to 5 × 105 cells. Increasing both the amount of protein in the cytosolic sample and the exposure time led to a pattern of labeling that was similar to that of the mitochondrial fraction (data not shown), presumably due to mitochondrial proteins released during cell homogenization. C, effect of uncoupling on IBTP labeling in Jurkat cells. Jurkat cells were preincubated + FCCP (200 nM) for 10 min, incubated for a further 15 min + IBTP (5 µM), and then pelleted. IBTP-labeled proteins were separated by SDS-PAGE and detected by immunoblotting. Each lane was loaded with protein from 106 cells. D, effect of uncoupling on IBTP labeling in bovine aortic endothelial cells. Cells were grown to confluence in 6-well culture dishes and then incubated with the indicated IBTP concentrations + FCCP for 1 h, and then IBTP-labeled proteins were separated by SDS-PAGE and detected by immunoblotting.

Those proteins labeled by incubating human fibroblasts with IBTP were separated by electrophoresis, transferred to nitrocellulose, and visualized by immunoblotting (Fig. 3B). There was extensive labeling of mitochondrial proteins within cells by IBTP (Fig. 3B, cell). That this labeling was due to the specific reaction of IBTP with mitochondrial proteins was confirmed by subfractionation of labeled cells into mitochondria- and cytosol-enriched fractions (Fig. 3B). This showed an identical pattern of protein IBTP labeling in the total cell lysate and in the mitochondrial fraction (Fig. 3B). The labeling of mitochondrial proteins within cells by IBTP was independent of cell type, as incubation of Jurkat (Fig. 3C) or bovine aortic endothelial cells (Fig. 3D) with IBTP also led to uncoupler-sensitive protein labeling. Therefore, IBTP reacts selectively with mitochondrial thiol proteins within cells and consequently can be used to probe their redox state. In the following series of studies the IBTP labeling of proteins in intact mitochondria was examined in more detail together with a proteomic analysis of the modifications involved.

Location of IBTP-reactive Protein Thiols within Mitochondria-- Isolated mitochondria incubated with IBTP showed extensive labeling of mitochondrial protein thiols (Fig. 4A). Labeling was prevented by abolishing the membrane potential with the uncoupler FCCP (Fig. 4A), as would be expected if only thiols in the matrix and on the internal face of the inner membrane reacted with IBTP. This was confirmed by separating IBTP-treated mitochondria into mitoplast, outer membrane, and intermembrane space fractions; only the mitoplast fraction contained IBTP-labeled proteins, and the pattern of protein labeling was identical to that of intact mitochondria (data not shown). When IBTP-treated mitochondria were separated into membrane and matrix fractions, both fractions contained significant numbers of protein thiols, with more in the matrix fraction (Fig. 4B). The cohorts of thiol proteins labeled in the two fractions were clearly distinct, with several bands present in only the matrix or the membrane fraction (Fig. 4B). Therefore IBTP is accumulated into the mitochondria driven by the membrane potential and labels matrix and matrix-facing protein thiols which can then be detected by immunoblotting.

View larger version (30K): [in this window] [in a new window]   Fig. 4.   Reaction of IBTP with thiol proteins in isolated mitochondria. A, IBTP labeling of whole mitochondria. Mitochondria (1 mg of protein/ml) were incubated at 25 °C in KCl medium at pH 7.2 with 10 mM succinate, 8 µg/ml rotenone for 2 min, and then IBTP (10 µM) was added, and 5 min later the mitochondria were pelleted by centrifugation. Mitochondrial protein (25 µg) was separated by SDS-PAGE, transferred to nitrocellulose, and probed with antitriphenylphosphonium antiserum. Where indicated FCCP (200 nM) was present throughout the incubation, or NEM (100 µM) was added followed 10 min later by IBTP. The mitochondrial membrane potentials () were determined after 15 min of incubation. +TPMP, IBTP-treated mitochondria were probed with antitriphenylphosphonium antiserum that had been preincubated with TPMP (100 µM) at 37 °C for 60 min. Addition of FCCP prior to pelleting IBTP-treated mitochondria did not change the pattern or intensity of protein labeling (data not shown). B, subfractionation of mitochondria. Mitochondria (1 mg of protein/ml) were incubated in KCl medium at 30 °C supplemented with glutamate/malate (5 mM of each) for 2 min, and then IBTP (10 µM) was added, and 5 min later mitochondria were pelleted by centrifugation. Protein (25 µg) from whole mitochondria, and from membrane and matrix fractions prepared by Lubrol-treatment, were separated by SDS-PAGE, transferred to nitrocellulose, and probed with antitriphenylphosphonium antiserum.

Selectivity of IBTP Reaction with Mitochondrial Protein Thiols-- The thiol alkylating reagent NEM prevented the reaction of IBTP with mitochondrial thiols at a concentration that did not disrupt the membrane potential (Fig. 4A); therefore, its action was solely due to blocking mitochondrial thiols and not to preventing IBTP uptake. NEM also blocked the reaction of IBTP with BSA (data not shown) further indicating that the reaction of IBTP was predominantly with cysteine residues. The antiserum used was specific for IBTP-labeled proteins as its binding was blocked by preincubation with the simple triphenylphosphonium cation TPMP, and mitochondrial proteins that had not been treated with IBTP were not detected (Fig. 4A). The selectivity of protein thiol labeling could be modulated by replacing the iodo functional group of IBTP with the less reactive bromo or chloro groups, and these compounds labeled mitochondrial thiols with the following reactivity order: IBTP > BrBTP > ClBTP (data not shown). To assess the potential of the higher intra-mitochondrial pH to enhance selective labeling of mitochondria, rates of reaction of IBTP with the free thiol of glutathione and with a protein thiol (BSA) were measured. The rates were found to be relatively slow (36 ± 7 × 10³ M¹ s¹ for glutathione and 53 ± 5 × 10³ M¹ s¹ for BSA at 37 °C, pH 8) and were highly pH-sensitive, being 50-70% slower at pH 7.2 compared with pH 8. These data are consistent with a reaction occurring through nucleophilic displacement of iodide by a thiolate and supporting the role of matrix pH in enhancing the mitochondrial selectivity of IBTP labeling.

Limited Depletion of Mitochondrial Thiols by IBTP-- Ideally a reagent for indicating thiol redox status would modify only a small pool of available protein thiol and have little effect on oxidative phosphorylation. Indeed, the relatively slow reaction between IBTP and thiols should label a small proportion of mitochondrial thiols without disrupting mitochondrial function by thiol depletion. This was supported by increased IBTP protein labeling at higher IBTP concentrations or longer incubation times (data not shown). Incubation with IBTP up to 10 µM did not deplete the mitochondrial glutathione pool to a significant extent, and even incubation with 10 µM IBTP for 30 min at 37 °C depleted mitochondrial glutathione by less than 15% (data not shown) The effects of IBTP concentrations up to 25 µM on total mitochondrial protein thiols and on those free thiols exposed on non-denatured proteins were also not significant. We conclude that under conditions of low IBTP concentration, labeling of protein thiols occurs without significantly depleting overall mitochondrial thiol content.

Effect of IBTP on Mitochondrial Function-- The effects of IBTP on phosphorylating or uncoupled respiration by isolated mitochondria were negligible below 50 µM (data not shown). There were small increases in coupled respiration at 25 µM IBTP, presumably due to slightly increased proton leak in the presence of the lipophilic cations (data not shown). Therefore, mitochondria can be incubated with up to 10-25 µM IBTP without significant effects on oxidative phosphorylation as would be expected if a small subset of the available thiols in functionally important proteins was being modified by the reagent. Consequently in subsequent experiments to label mitochondrial thiol proteins, we incubated mitochondria with 5-25 µM IBTP for up to 15 min to assess responses under conditions where a decrease in mitochondrial thiol content was not detected and at the threshold of the minor effects on coupling.

Resolution of IBTP-reactive Mitochondrial Thiol Proteins by Two-dimensional Electrophoresis-- To resolve mitochondrial thiol proteins, IBTP-labeled mitochondrial samples were separated by two-dimensional electrophoresis on paired gels, one of which was Coomassie-stained, whereas the other was probed with antitriphenylphosphonium antiserum (Fig. 5). In some cases the IEF separation was carried out over a narrower pH range to facilitate resolution of IBTP-labeled proteins (Fig. 5, B and D). Markedly different patterns of IBTP labeling were seen when comparing whole mitochondria with the membrane-enriched fraction (Fig. 5, A and C) or whole mitochondria with the matrix fraction (Fig. 5, B and D).

View larger version (90K): [in this window] [in a new window]   Fig. 5.   Resolution of IBTP-labeled mitochondrial proteins by two-dimensional electrophoresis. Mitochondria, matrix, or membrane fractions (200 µg of protein) from IBTP-treated mitochondria were separated by two-dimensional gel electrophoresis. Identical gels were either Coomassie-stained or transferred to nitrocellulose and probed with antitriphenylphosphonium antiserum. A, mitochondria, IEF pH 3-10. B, mitochondria, IEF pH 4-7. C, membrane, IEF pH 3-10. D, matrix, IEF pH 4-7. Control immunoblots of mitochondria that had not been reacted with IBTP gave no immunoreactivity (data not shown). Circles show spots present in both Coomassie- and IBTP-stained gels, and squares show spots present only in the Coomassie-stained gels.

Labeling of cysteine residues with IBTP may have a small effect on the pI of the protein measured by IEF because the introduction of the cationic phosphorium label will cause a slight basic shift in the pI. Hence in comparing the immunoblots and Coomassie-stained gels, identical samples of IBTP-treated proteins were analyzed (Fig. 5). As expected, only a proportion of mitochondrial proteins visible on Coomassie-stained gels also contained IBTP-reactive thiols, and some examples of mitochondrial proteins that did not contain IBTP-reactive thiols are indicated (squares, Fig. 5, A-D). Many proteins containing IBTP-reactive thiol groups could be identified on both Coomassie-stained gels and on immunoblots, and some examples are indicated (circles, Fig. 5, A-D). We conclude that mitochondria contain many matrix and inner membrane proteins with reactive thiols exposed to the matrix and that these proteins can be isolated as single spots by two-dimensional electrophoresis.

IBTP Labeling of Respiratory Complexes Identified by Blue Native Gel Electrophoresis-- A drawback with the use of conventional two-dimensional electrophoresis is that many hydrophobic mitochondrial inner membrane proteins cannot be separated because they precipitate in the isoelectric focusing gel (38). This fraction contains many polypeptides of interest in the respiratory complexes that are thought to contain thiols susceptible to modification by reactive oxygen and nitrogen species. An alternative approach is to use blue native gel electrophoresis (BN-PAGE) in which respiratory complexes are separated in an intact and native form (27). Mitochondria were incubated with IBTP, and respiratory complexes were then separated by BN-PAGE, transferred to nitrocellulose, and probed with antitriphenylphosphonium antiserum (Fig. 6). Prominent among the IBTP-labeled protein complexes was one the size of Complex I (~900 kDa), and its identity was confirmed by probing a parallel sample with a Complex I-specific antibody (Fig. 6A). Other protein complexes were labeled by IBTP, but their intensity was less than for Complex I as indicated by the overexposed Complex I band (Fig. 6B). The molecular weights of two of these IBTP-reactive bands were similar to Complex II (succinate dehydrogenase; 130 kDa) and Complex IV (cytochrome oxidase; 200 kDa), and these identities were confirmed by probing parallel samples with specific antibodies (Fig. 6, C and D).

View larger version (51K): [in this window] [in a new window]   Fig. 6.   Reaction of IBTP with oxidative phosphorylation complexes. Mitochondria (1 mg of protein/ml) were incubated for 5 min at 30 °C in KCl medium supplemented with 10 mM glutamate/malate and IBTP (25 µM). The mitochondria were pelleted, and 250 µg of protein was separated by blue native PAGE, transferred to nitrocellulose, and probed with antitriphenylphosphonium antiserum, or with monoclonal antibodies against respiratory complexes. Where indicated FCCP (200 nM) was present or IBTP was omitted. A, Complex I. Parallel blots were probed with antiserum against triphenylphosphonium and with a monoclonal antibody against Complex I. B, respiratory complexes. The blot was probed with antitriphenylphosphonium antiserum and was overexposed to visualize further IBTP-labeled bands in addition to Complex I (overexposed band). C, cytochrome oxidase. Parallel blots were probed with antiserum against triphenylphosphonium and with a monoclonal antibody against cytochrome oxidase (COX). D, succinate dehydrogenase. Parallel blots were probed with antiserum against triphenylphosphonium and with a monoclonal antibody against succinate dehydrogenase (SDH). Data are typical of results repeated on several gels for each sample. E, two-dimensional separation of mitochondrial proteins. Mitochondrial proteins were separated by BN-PAGE in the first dimension followed by SDS-PAGE in the second dimension. Identical gels were then either stained with Coomassie Blue or probed with antitriphenylphosphonium antiserum. The approximate locations of mitochondrial oxidative phosphorylation complexes after BN-PAGE are indicated.

The large mitochondrial complexes initially isolated by BN-PAGE could be resolved as individual polypeptides by carrying out SDS-PAGE in the second dimension (Fig. 6E). This was done for identical samples from IBTP-labeled mitochondria, and the resulting gels were then either stained with Coomassie Blue or transferred to nitrocellulose and probed with antitriphenylphosphonium antiserum (Fig. 6E). A number of IBTP-labeled polypeptides were seen, including polypeptides from all the protein complexes detected by immunoblotting of the BN-PAGE. In addition many more IBTP-labeled polypeptides were evident after the second dimension SDS-PAGE than would have been predicted from immunoblots of the BN-PAGE. This increased sensitivity following second dimension SDS-PAGE is probably due to removal of the high background found for BN-PAGE immunoblots that limit sensitivity and to greater binding of the antiserum to the denatured proteins.

Loss of Mitochondrial Protein Thiols in Response to Oxidative Stress-- To determine how protein thiols change redox state in response to oxidative stress, we measured the effect of oxidants on IBTP binding to mitochondrial protein thiols (Fig. 7). Addition of diamide, a thiol-selective oxidant, led to a significant decrease in labeling of mitochondrial proteins by IBTP (Fig. 7A). To eliminate the possibility that decreased IBTP labeling merely reflected less IBTP uptake due to disruption of the membrane potential, the effect of diamide on membrane potential was shown to be negligible (Fig. 7A, inset panel). Addition of tert-butyl hydroperoxide (tBHP) also decreased labeling of mitochondrial thiols by IBTP without disrupting the membrane potential (Fig. 7B). This occurs because tBHP is a glutathione peroxidase substrate that oxidizes mitochondrial glutathione to glutathione disulfide that then forms protein mixed disulfides (39), and also as a consequence of oxidative stress induced by the peroxide itself.

View larger version (60K): [in this window] [in a new window]   Fig. 7.   Effect of oxidative stress on IBTP labeling of mitochondrial protein thiols. Mitochondria (1 mg of protein/ml) were incubated at 30 °C in KCl medium with 10 mM succinate and 8 µg/ml rotenone. After oxidant treatment IBTP (25 µM) was added, and then mitochondria were pelleted by centrifugation; protein (25 µg) was separated by SDS-PAGE, transferred to nitrocellulose, and probed with antitriphenylphosphonium antiserum. In parallel experiments the effect of the oxidant on membrane potential was determined, and the values are shown below the immunoblots. A, diamide. After incubation with diamide for 5 min, IBTP was added, and mitochondria were pelleted 5 min later. The membrane potential was measured after 5 min of incubation with diamide. B, tBHP. After incubation with tBHP for 5 min, IBTP was added, and mitochondria were pelleted 5 min later. The membrane potential was measured after incubation with tBHP for 5 min. C, peroxynitrite. Mitochondria were exposed to peroxynitrite for 15 s, and then IBTP was added, and 3 min later the mitochondria were pelleted and analyzed. In parallel the membrane potential was determined after incubation with peroxynitrite for 5 min. The decrease in IBTP binding was reproducibly maximal at 500 µM and lower at 1 mM for reasons that are unclear. D, SNAP. Mitochondria were incubated with SNAP for 10 min, and then IBTP was added, and 5 min later mitochondria were pelleted. The membrane potential was determined after incubation with SNAP for 5 min. The increase in IBTP binding for some protein bands at 50 µM SNAP (arrows) was reproducible. E, effect of diamide on IBTP labeling of mitochondrial protein thiols. Mitochondria (200 µg of protein) were incubated as described in A above, + 500 µM diamide, separated by two-dimensional gel electrophoresis, transferred to nitrocellulose, and probed with antitriphenylphosphonium antiserum (IEF, pH 4-7). Arrows indicate proteins with diamide-sensitive IBTP labeling.

To investigate the interaction of reactive nitrogen species with mitochondrial thiols, we studied the effects of peroxynitrite and SNAP on IBTP labeling of mitochondrial proteins (Fig. 7, C and D). Peroxynitrite contributes to oxidative stress in vivo and is more reactive than its precursors, superoxide and nitric oxide (40). The S-nitrosothiol SNAP is a model for the endogenous low molecular weight S-nitrosothiols capable of mediating S-nitrosation of protein thiols in vivo during nitric oxide signaling and nitrosative stress. Peroxynitrite concentrations sufficient to induce the mitochondrial permeability transition (41) blocked many mitochondrial protein thiols without affecting the membrane potential, but the thiol depletion was less than that caused by diamide or tBHP (Fig. 7C). Incubation of mitochondria with SNAP decreased IBTP binding to mitochondrial thiol proteins without affecting the membrane potential, but this effect was also less than that found for diamide or tBHP (Fig. 7D). Therefore, loss of IBTP labeling can be used to measure redox changes in mitochondrial thiols in response to oxidative stress. From these studies we were also able to conclude that oxidation of the mitochondrial glutathione pool or the formation of ROS and RNS has a dramatic effect on the redox state of mitochondrial thiol proteins.

Treatment with diamide, a reagent that reacts with both protein and low molecular weight thiols, decreased the overall intensity of protein labeling by IBTP (Fig. 7A). This could be due to a generalized decrease in the IBTP reactivity of all thiol proteins, to the selective loss of particular protein thiols, or to a combination of both processes. To investigate this, we compared the proteins labeled by IBTP in diamide-treated and control mitochondria using two-dimensional immunoblots (Fig. 7E). Immunoblots of diamide-treated samples were developed until their intensities were similar to controls to facilitate comparison of the patterns of thiol protein labeling. The pattern of IBTP labeling seen on two-dimensional immunoblots was largely unchanged by treatment with diamide, suggesting that most mitochondrial protein thiols are susceptible to oxidative stress to a similar degree with this reagent. Therefore the decrease in IBTP labeling caused by diamide seen in Fig. 7A is due to the formation of mixed disulfides by most exposed protein thiols. However, some protein thiols were particularly susceptible to alteration by oxidative stress as IBTP labeling was lost following treatment with diamide (arrows, Fig. 7E). A further point of interest was that some mitochondrial thiol proteins increased their IBTP reactivity on treatment with diamide, tBHP, SNAP, and NEM (arrows, Fig. 7, A, B, and D, and Fig. 4A). Therefore, in response to oxidative stress most mitochondrial protein thiols are affected to a similar extent; in addition there is a subset of protein thiols that are particularly susceptible to changes in redox state. In all cases changes in the reactivity of protein thiols with IBTP can be used to identify those proteins that change their thiol redox state in response to oxidative stress.

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

The importance of redox changes to mitochondrial thiol proteins during oxidative stress and cell death is well recognized, but direct measurement of the modification of different mitochondrial protein thiols by oxidative stress has not been undertaken. By using the novel compound IBTP, specific labeling of mitochondrial protein thiols was demonstrated in both intact cells and the isolated organelle. From the pattern of IBTP labeling on immunoblots, it was clear that mitochondria contain a number of relatively abundant proteins with IBTP-reactive thiols. Most of these proteins were in the matrix, but a significant proportion of membrane proteins also contained reactive thiols on their matrix-facing surfaces. In future studies, IBTP labeling will enable individual mitochondrial thiol proteins to be isolated and identified by spot excision from the gel followed by N-terminal sequencing or in-gel trypsin digestion in conjunction with mass spectrometry. The mass shift due to the butyltriphenylphosphonium adduct (+316.1381 for the singly charged peptide) will also facilitate identification by mass spectrometry of tryptic peptides containing active thiol residues.

Changes in thiol redox state prevented IBTP labeling of protein thiols, facilitating their isolation by electrophoresis and immunoblotting. The reactivity of the majority of mitochondrial thiol proteins with IBTP decreased on oxidizing the glutathione pool with diamide or tBHP (39). As expected, exposure to the S-nitrosothiol SNAP or to peroxynitrite did not have as extensive effects on the redox state of mitochondrial thiol proteins as direct oxidation of the glutathione pool. These changes are likely to occur due to S-nitrosation of proteins or as a secondary consequence of oxidation or S-nitrosation of mitochondrial glutathione. Very little is known about intraorganelle signaling in the mitochondrion. By analogy with established pathways in the cytosol, one can propose that thiol modification of key metabolic or signaling molecules occurs within the organelle due to the formation of protein mixed disulfides or S-nitrosated species. Interestingly, oxidant production in the cytosol could coordinate responses in the mitochondrion through thiol modification, in addition to activating signaling pathways in the cytosol. This is supported by the finding that SNAP modified intramitochondrial protein thiols when added to intact mitochondria. The protein-glutathione mixed disulfides formed will be regenerated after oxidative stress by thioredoxin/thioredoxin reductase, glutaredoxin/glutaredoxin reductase, or by re-equilibration with glutathione. It is anticipated that reductive processes involving GSH will determine the persistence and stability of modified thiol proteins. Indeed in a metabolic regulatory role GSH is likely to be critical in controlling the duration of a signal in an analogous fashion to a phosphatase.

Analysis of the mitochondrial inner membrane fraction, which contains key components in electron transport and apoptosis, was particularly interesting. Using IBTP labeling in conjunction with BN-PAGE enabled us to confirm the presence of free thiol residues on the matrix surface of respiratory Complexes I, II, and IV. Whereas the presence of reactive thiols on these complexes is known, their locations and roles are uncertain; IBTP labeling will help address these points (1, 42-45). Using a combination of proteomics approaches, we have uncovered a number of candidate protein thiols that are particularly susceptible to thiol modification, and these are now being investigated as potential regulators or redox sensors in the response of mitochondria to oxidative stress.

Interestingly, some protein thiols increased their reactivity with IBTP on exposure to diamide, SNAP, or NEM. Whereas the cause of this unexpected finding is uncertain, there are a number of possibilities. Oxidative stress may cause conformational changes that render some protein thiols more reactive, either by increasing their exposure to the matrix or by decreasing their pK_a values by moving them relative to charged residues. Alternatively, the increased thiol reactivity may follow alteration to an iron-sulfur center that exposes a cysteine that had been blocked previously by an iron atom.

To conclude, we have developed a procedure to label mitochondrial thiol proteins within living cells that enables us to follow changes in their redox state. By using this procedure we have shown that there are a number of active thiols on the surface of mitochondrial matrix and membrane proteins that change their redox state in response to both glutathione oxidation and exposure to S-nitrosating agents or peroxynitrite. Among these are individual protein thiols that are particularly responsive to oxidative stress, either decreasing or increasing their IBTP binding. The development of a selective mitochondrial thiol reagent enables the technology of proteomics to be focused on this important subset of mitochondrial proteins, and will help characterize those mitochondrial thiol proteins that change redox state during oxidative stress, apoptosis, and aging.

	FOOTNOTES

^* This work was supported by grants from the Health Research Council of New Zealand (to M. P. M. and R. A. J. S.), the Marsden Fund, administered by the Royal Society of New Zealand (to M. P. M. and R. A. J. S.), and by National Institutes of Health Grant HL58031 (to V. D. U.).The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

^¶ Health Research Council of New Zealand scholar.

^§§ To whom correspondence should be addressed: MRC-Dunn Human Nutrition Unit, MRC-Wellcome Trust Bldg., Hills Rd., Cambridge CB2 2XY, UK. Tel.: 44-1223-252900; Fax: 44-1223-252905; E-mail: mpm@MRC-Dunn.cam.ac.uk.

Published, JBC Papers in Press, February 22, 2002, DOI 10.1074/jbc.M110797200

	ABBREVIATIONS

The abbreviations used are: IBTP, (4-iodobutyl)triphenylphosphonium; BN-PAGE, blue native PAGE; BrBTP, (4- bromobutyl)triphenylphosphonium; ClBTP, (4-chlorobutyl)triphenylphosphonium; DTNB, dithionitrobenzoic acid; FCCP, carbonyl cyanide p-trifluoromethoxyphenylhydrazone; IEF, isoelectric focusing; NEM, N-ethylmaleimide; SNAP, S-nitroso-N-acetylpenicillamine; tBHP, tert-butyl hydroperoxide; TPMP, methyltriphenylphosphonium; Tricine, N-[2-hydroxy-1,1-bis(hydroxymethyl)ethyl]glycine; PBS, phosphate-buffered saline; TBS, Tris-buffered saline; DMEM, Dulbecco's modified Eagle's medium; KLH, keyhole limpet hemocyanin; FCS, fetal calf serum; BSA, bovine serum albumin.

	REFERENCES

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES