Formation of protein tyrosine ortho-semiquinone radical and nitrotyrosine from cytochrome c-derived tyrosyl radical.

Oxidative alteration of mitochondrial cytochrome c (cyt c) has been linked to disease pathophysiology and is one of the causative factors for pro-apoptotic events. Hydrogen peroxide induces a short-lived cyt c-derived tyrosyl radical as detected by the electron spin resonance (ESR) spin-trapping technique. This investigation was undertaken to characterize the fate and consequences of the cyt c-derived tyrosyl radical. The direct ESR spectrum from the reaction of cyt c with H(2)O(2) revealed a single-line signal with a line width of approximately 10 G. The detected ESR signal could be prevented by pretreatment of cyt c with iodination, implying that the tyrosine residue of cyt c was involved. The ESR signal can be enhanced and stabilized by a divalent metal ion such as Zn(2+), indicating the formation of the protein tyrosine ortho-semiquinone radical (ToQ.). The production of cyt c-derived ToQ. is inhibited by the spin trap, 2-methyl-2-nitrosopropane (MNP), suggesting the participation of tyrosyl radical in the formation of the ortho-semiquinone radical. The endothelium relaxant factor nitric oxide is well known to mediate mitochondrial respiration and apoptosis. The consumption of NO by cyt c was enhanced by addition of H(2)O(2) as verified by inhibition electrochemical detection using an NO electrode. The rate of NO consumption in the system containing cyt c/NO/H(2)O(2) was decreased by the spin traps 5,5-dimethyl pyrroline N-oxide and MNP, suggesting NO trapping of the cyt c-derived tyrosyl radical. The above result was further confirmed by NO quenching of the ESR signal of the MNP adduct of cyt c tyrosyl radical. Immunoblotting analysis of cyt c after exposure to NO in the presence of H(2)O(2) revealed the formation of 3-nitrotyrosine. The addition of superoxide dismutase did not change the cyt c nitration, indicating that it is peroxynitrite-independent. The results of this study may provide useful information in understanding the interconnection among cyt c, H(2)O(2), NO, and apoptosis.

Oxidative stress in mitochondria is a normal physiological process during respiration (1); its major source is a variety of reactive oxygen species produced during oxidative phosphorylation by the electron transport chain (ETC) 1 (1,2). During normal mitochondrial respiration, electrons may be released during steps prior to oxidative phosphorylation and subsequently react with molecular oxygen to form superoxide (3), the first product in a chain of reactive oxygen species production. A decrease in the rate of mitochondrial phosphorylation (state 4 respiration) increases the production of superoxide in the early stages of the ETC.
Generally, superoxide produced by the ETC is converted to H 2 O 2 by mitochondrial manganese-superoxide dismutase. It has been estimated that ϳ2% of the oxygen uptake by mitochondria in state 4 respiration results in production of O 2 . and H 2 O 2 (4). Normally, the level of cellular H 2 O 2 can be kept relatively low through the action of mitochondrial glutathione peroxidase and cytosolic catalase. However, under certain conditions such as ischemia reperfusion injury or inflammation, the mitochondria can produce an excess of H 2 O 2 (5, 6) that can react with the metalloproteins in the ETC, e.g. heme and ironsulfur proteins, to form protein radicals. One of the most dominant proteins in mitochondria is cytochrome c (cyt c), the only soluble protein in the ETC. Cyt c has two physiological roles: 1) it mediates electron shuttling between ubiquinol-cytochrome c oxidoreductase (complex III) and cytochrome c oxidase (complex IV) during mitochondrial respiration, and 2) it serves as a factor to regulate preapoptotic events (7). Protein radical(s) induced by the reaction of cyt c with H 2 O 2 and other oxidants have been detected by the ESR spin-trapping technique (8,9). Barr et al. (8) identified a protein-derived tyrosyl radical formed in the reaction of cyt c with H 2 O 2 . Chen et al. (9) have demonstrated that site-specific oxidation at the sixth ligand of cyt c can be induced by hypochlorite, forming a modified cyt c with methionine sulfoxide with an enhanced peroxidase activity, which increases the formation of cyt c-derived tyrosyl radical by a factor of 2-3.
A direct broad line ESR spectrum from cyt c has been assigned as a peroxynitrite-or H 2 O 2 -induced tyrosyl radical (8,10). Barr et al. (8) also detected a broad and featureless ESR spectrum with a line width of ϳ10 G from the reaction of cyt c with H 2 O 2 in the absence of spin trap. The direct ESR spectra of protein tyrosyl radicals have been reported in a number of other enzymatic or heme protein systems including prostaglandin H synthase (11), ribonucleotide reductase (12), photosystem II (13,14), cytochrome c oxidase (15), and H 2 O 2 -induced metmyoglobin (16). All of these have similar spectral characteristics with a line width (peak to trough) of more than 20 G, which is significantly broader than the parameter of cyt cderived protein radical. It has also been reported that the cyt c-derived tyrosyl radical can form dityrosine through radical-radical addition as evidenced by H 2 O 2 -induced cyt c dimerization (17).
These observations prompted us to re-evaluate the possible fate and consequences of H 2 O 2 -induced tyrosyl radical formation from cyt c. In addition to tyrosyl radical-initiated protein dimerization, we hypothesize that two potentially important reactions should be relevant in the mitochondrial and cellular milieu: 1) cyt c-derived tyrosyl radical (Scheme I, structure B), as detected by the ESR spin-trapping technique, is short-lived and subsequently oxidized to a long-lived radical with a narrower line width (compared with that of protein tyrosyl radical), forming a novel protein-derived tyrosine ortho-semiquinone radical (ToQ . ) (Scheme I, structure C). This orthosemiquinone anion radical can be stabilized by complexing with Zn 2ϩ in acidic medium (Scheme I, structure D). 2) The endothelium relaxant factor NO, which can mediate a number of mitochondrial functions (18), is capable of trapping the cyt c-derived tyrosyl radical at a favorable rate (1-2 ϫ 10 9 M Ϫ1 s Ϫ1 ), leading to the formation of 3-nitrotyrosine (Scheme I, structure E) (19). In the present work we have used a newly established technique combining metal divalent cation trapping with an approach based on the chemistry of heterobifunctional crosslinking reagent to characterize the cyt c-derived ToQ . . Cyt c nitration from NO trapping of the tyrosyl radical was identified by immunochemistry. This pathway of cyt c nitration is important in understanding the fundamental mechanism by which oxidants and NO induce mitochondrial damage and cellular apoptosis in the pathogenesis of disease.

EXPERIMENTAL PROCEDURES
Reagents-Diethylenetriaminepentaacetic acid (DTPA), horse heart cytochrome c, horseradish peroxidase (HRP), sodium iodide, dithiothre-itol, 5-bromo-4-chloro-3-indolyl phosphate, and nitro blue tetrazolium were purchased from Sigma and used as received. Zinc sulfate heptahydrate, 2-methyl-2-nitrosopropane (MNP), [3-(3,4-dihydroxylphenyl)-L-alanine] (L-DOPA), and sodium hypochlorite (4% chlorine available in solution) were purchased from Aldrich. The 5,5-dimethyl-1-pyrroline N-oxide (DMPO) spin trap (from Aldrich) was vacuum-sublimed twice and stored under nitrogen at Ϫ70°C until needed. Prepacked Sephadex G-25 (PD-10) size exclusion columns were purchased from Amersham Biosciences. Lysozyme (from hen egg white) and Pronase were obtained from Roche Applied Science. N-Succinimidyl iodoacetate was purchased from Pierce. The NO donor diethylamine nonoate (DEA/NO) was obtained from Cayman Chemical (Ann Arbor, MI). Perdeuterated MNP (MNP-d 9 ) was provided as a gift from Dr. Joy Joseph (Medical College of Wisconsin, Milwaukee, WI) Analytical Methods-Optical spectra were measured on either an SLM Aminco DW-2000 UV-visible or Cary 300 spectrophotometer. The concentration of cyt c was calculated from the difference spectrum after ascorbate reduction using a millimolar extinction coefficient of 18 (20). The concentration of HRP was determined by using an extinction coefficient of 102 mM Ϫ1 cm Ϫ1 at 402 nm. The protein concentrations of lysozyme and L-DOPA-conjugated lysozyme (L-DOPA-LYS) were determined by the Lowry assay with bovine serum albumin as the standard.
Iodonation of Tyrosine Residues of Cytochrome c-The reaction mixture contained 1-ml aliquots of cyt c (200 M) in 50 mM sodium phosphate buffer, pH 7.4, and NaI (final concentration of 40 mM). Iodination of tyrosine residues of cyt c was initiated by the addition of two Nchloro-benzenesulfonamide immobilized beads (Iodo Beads; Pierce). The reaction was allowed to proceed on a rapid thermomixer at 23°C for 15 min. To stop the reaction, the solution was removed from the vessel and passed through a prepacked Sephadex G-25 column to remove excess sodium iodide. The fraction containing iodonated cyt c was collected and measured with UV-visible spectrophotometry to determine its heme concentration, and the protein modification was confirmed with electrospray mass spectrometry (9).
Preparation of L-DOPA-conjugated Lysozyme (L-DOPA-LYS)-Lysozyme (33 mg/ml, or 2.3 mM) was dissolved in 50 mM sodium borate buffer prebubbled with argon for 45 min, pH 8.3 (buffer A), and reduced with dithiothreitol (in excess by a factor of 2) at room temperature for 5 min. The reaction mixture was dialyzed against buffer A containing 5 mM EDTA overnight with one change of dialysis buffer. L-DOPA (5 mM) in 10 mM sodium phosphate buffer, pH 7.4, containing 150 mM of NaCl (phosphate-buffered saline) was incubated with an equal amount of N-succinimidyl iodoacetate (SIA, 5 mM in Me 2 SO) at room temperature for 30 min. The resultant L-DOPA-conjugated SIA was mixed with dithiothreitol-treated lysozyme at a molar ratio of L-DOPA/protein of ϳ1.2 and incubated at room temperature for 1 h. The reaction mixture was then dialyzed against 50 mM sodium phosphate buffer, pH 7.4, overnight with one change of buffer. The resultant L-DOPA-LYS was concentrated with centricon-3, and the concentration of protein was adjusted to 10 mg/ml prior to ESR experiments.
Electron Spin Resonance Experiments-ESR experiments were carried out on a Bruker EMX spectrometer operating at 9.8 GHz with 100-kHz modulation frequency at room temperature. The reaction mixture was transferred to a 10-mm quartz ESR flat cell, which was then positioned into the SHQ or HS cavity (Bruker Instrument, Billerica, MA). The sample was scanned using the parameters described in the figure legends. The concentration of the MNP adduct of cyt c radical was estimated by using the 4-hydroxyl-2,2,6,6-tetramethyl-1-piperidinyloxy radical standard (10, 50, and 100 M) in 50% glycerin. The calibration was performed by using a fingertip Dewar flask containing liquid nitrogen (77 K) and operated at the following instrumental parameters: microwave power, 20 mW; center field, 3374 G; field of sweep, 200 G; scanning time, 30.14 s; time constant, 81.92 ms; receiver gain, 1 ϫ 10 5 ; modulation amplitude, 1 G; and signal average, 10 scans. The spin quantitation of each spectrum was obtained by double integration.
Nitric Oxide Consumption Rate by Cytochrome c-The NO consumption rate was measured electrochemically at 37°C in an electrochemical vial using an APOLLO 4000 free radical analyzer and NO electrode (World Precision Institute, Sarasota, FL). The electrochemical detector continuously recorded the current through the working electrode, which is proportional to the NO concentration in the solution. The sensor was calibrated with known concentrations of NO, using NO equilibrated solutions as described in the literature (21,22).
Immunoblotting Analysis-The reaction mixture was mixed with NuPAGE TM LDS sample buffer at a ratio 3:1 (v/v), incubated at 37°C SCHEME I for 2 h, and immediately loaded onto a 4 -12% gradient Bis-Tris gel (Invitrogen). The samples were run on SDS-PAGE and were electrophoretically transferred to nitrocellulose membrane in 25 mM Biocine, 25 mM Bis-Tris, and 10% methanol. The membranes were blocked for 1 h at room temperature in Tris-buffered saline containing 0.1% Tween-20 (TTBS) with 2.5% bovine serum albumin. The blots were then incubated overnight with anti-nitrotyrosine polyclonal antibody (Upstate Biotechnology Inc.) in TTBS at room temperature, washed three times in TTBS, and incubated for 1 h with alkaline phosphataseconjugated anti-rabbit IgG in TTBS at room temperature. The blots were washed again, twice in TTBS, and twice in Tris-buffered saline and then visualized using a mixture of 5-bromo-4-chloro-3-indolyl phosphate and nitro blue tetrazolium.
Cell Culture and H 2 O 2 -induced Apoptosis as Analyzed with DNA Fragmentation and Mitochondrial Cyt c Release-Rat cardiac myoblasts (H9c2 cell line from ATCC, Manassas, VA) were grown and maintained in the Dulbecco's modified Eagle's medium (ATCC) supplemented with 10% fetal bovine serum and 1% penicillin/streptomycin antibiotics in 35-mm polystyrene tissue culture dishes at 37°C in the presence of 4.5% CO 2 . Confluent cells with Ͼ90% viability were used to conduct the experiments of apoptosis.
Apoptosis was induced by exposure of the cells to H 2 O 2 (200 M) and continued culturing for 0.5, 1, 2,4, and 8 h. The cells were washed and resuspended in phosphate-buffered saline and then harvested by centrifugation at 500 ϫ g for 10 min. The supernatants were removed, and 330 l of Sarkosyl detergent lysis buffer (50 mM Tris, 10 mM EDTA, 0.5% w/v sodium N-lauroyl sarcosine, pH 7.5) and 13-l of proteinase K (15.6 mg/ml) (Roche Applied Science) were added, vortexed, and allowed to digest overnight at 37°C. RNase A (0.3 mg/ml) was added, and the samples were incubated at 56°C for 1 h. The lysates were then extracted with phenol/chloroform/isoamyl alcohol (25:24:1), vortexed, and centrifuged at 16,000 ϫ g for 5 min. The DNA was precipitated by 3 M sodium acetate and 100% ethanol, washed with 70% ethanol, and then dried under the vacuum. The purified DNA (2 g dissolved in Tris/ EDTA buffer) was mixed with BlueJuice gel loading buffer (Invitrogen) and then subjected to gel electrophoresis (50 V for 2.5 h) using 1% agarose gels containing ethidium bromide. DNA fragmentation was visualized and examined with ultraviolet light.
The release of cyt c from mitochondria into the cytosol was examined by subfractionating the cell as described by Liu et al. (7). The prepared soluble cytosol fraction was subjected to Western blot analysis using mouse anti-cytochrome c monoclonal antibody (Upstate Biotechnology Inc.) and an ECL detection system (Amersham Biosciences). An equal amount of protein (85 g) was loaded prior to SDS-PAGE.

Direct ESR Spectra from the Reaction of Cyt c with Hydrogen
Peroxide-As previously demonstrated, a broad singlet ESR spectrum with a line width of ϳ10 G can be detected when cyt c solution is mixed with H 2 O 2 (8). We confirmed that the detected ESR signal was enhanced when native cyt c was replaced by HOCl-cyt c in the same reaction system ( Fig. 1) because of the enhanced peroxidase activity of HOCl-cyt c (9). Presumably, the concentration of any globin radical will increase with the peroxidase activity of the heme protein. The addition of heme blockers such as imidazole (100 mM) or cyanide (1 mM) completely inhibited formation of the detected ESR spectrum (data not shown), implying a heme-initiated reaction. The ESR signal was stable and even persisted after dialysis, suggesting a very stable, protein-derived free radical. Formation of the protein radical was prevented by prior iodination of cyt c, thereby indicating that tyrosine residues were involved in the radical formation ( Fig. 1).
In a parallel experiment, we conducted the same measurement at low temperature (77 K). At this temperature the ϳ10-G line width of the ESR spectrum from the reaction system of either cyt c/H 2 O 2 or HOCl-cyt c/H 2 O 2 was not further broadened (data not shown). This failure to broaden indicates a total immobilization of the free radical on the ESR time scale even at room temperature, consistent with a free radical rigidly attached to cyt c. H 2 O 2 induces a Short-lived Cyt c-derived Tyrosyl Radical as Verified by the ESR Spin-trapping Technique-Barr et al. (8) have reported that H 2 O 2 induces tyrosyl radical(s) derived from cyt c, as detected by ESR spin-trapping techniques. However, our direct ESR spectra indicated that the H 2 O 2 -induced protein radical is stable and long-lived. To test whether the cyt cderived tyrosyl radical as detected by MNP spin trapping is as stable as the above protein radical detected by direct ESR, we did the following experiments. First, we reacted cyt c with H 2 O 2 in the presence of perdeuterated MNP (MNP-d 9 ). The result was an immobilized nitroxide ( Fig. 2A). Nonspecific proteolytic digestion of the immobilized nitroxide by Pronase yielded an isotropic three-line spectrum with a N ϭ 15.6 G, indicative of a tyrosyl radical adduct (8,9,16,20,23) (Fig. 2B). To test the stability of the assigned tyrosyl radical, the spin trap MNP-d 9 was added to the reaction mixture after a 1-min incubation of cyt c with H 2 O 2 . The intensity of the immobilized nitroxide signal drastically decreased (Fig. 2C), suggesting that the cyt c-derived tyrosyl radical induced by peroxide is shortlived. We also observed an isotropic three-line spectrum with a nitrogen coupling constant of 17.1 G, which we assigned to di-tert-butylnitroxide, a decomposition product of MNP.
The formation of cyt c-derived immobilized nitroxide by MNP spin trapping allows us to quantify the amount of cyt c-derived tyrosyl radical as described under "Experimental Procedures." The concentration of peroxide-induced MNP adduct of cyt c is estimated to be ϳ24 M, which represents ϳ4.8% of cyt c used. However, replacement of cyt c with HOCl-treated cyt c resulted in the enhancement of detected immobilized nitroxide to ϳ60 M (ϳ12% of HOCl-cyt c used).  protein radical detected by direct ESR, it is necessary to evaluate the fate and consequence of the tyrosyl radical detected by spin trapping. The observation of a short-lived tyrosyl radical prompted us to hypothesize that it could be subsequently oxidized to ToQ . in the absence of spin trap. It has been observed in biological systems that the ortho-semiquinone anion radical can be generated by either enzymatic or nonenzymatic pathways from certain biologically active compounds such as catechol and its derivatives (24 -27).
Kalyanaraman et al. (24,25,27) have reported that orthosemiquinone radical can be stabilized by chelation through the use of diamagnetic divalent ions such as Zn 2ϩ under acidic pH, thereby enhancing the ESR signal effectively. We took this approach here, using Zn 2ϩ to test for cyt c-derived ToQ . generated by H 2 O 2 . As indicated in Fig. 3A, the direct ESR spectrum from the reaction system containing cyt c and H 2 O 2 at pH 7.4 was consistent with the published results by Barr et al. (8). When cyt c was reacted with 5 eq of H 2 O 2 in the presence of an excess of Zn 2ϩ (200 eq) at pH 5.5, the ESR spectrum revealed an enhanced signal with a peak-to-trough width of ϳ6 G (Fig.  3B). The spectrum of the cyt c/H 2 O 2 /Zn 2ϩ reaction system depended on the presence of both the protein (Fig. 3C) and peroxide (Fig. 3D), implying that Zn 2ϩ trapped an immobilized ortho-semiquinone radical within the cyt c protein (24,25,27).
When experiments including reaction systems of cyt c/H 2 O 2 and cyt c/H 2 O 2 /Zn 2ϩ were carried out at pH 5.5, ESR spectra from the initial and second scans were collected (Fig. 4). In the initial scans, an additional species with a broader line width was present (Fig. 4, A and C); this species subsequently decayed in the second scan to yield only ortho-semiquinone radical (Fig. 4B), which was stabilized and enhanced by Zn 2ϩ trapping (Fig. 4D). The broader direct ESR spectrum is consistent with the short-lived tyrosyl radical detected by spin trapping (Fig. 2).
Identification of Protein-derived ToQ . -The ToQ . can be generated by one-electron oxidation of L-DOPA using an enzymatic system consisting of HRP/H 2 O 2 (27). To provide further evi-dence that the cyt c system also generates an ortho-semiquinone radical in the presence of H 2 O 2 , we coupled L-DOPA to a non-heme protein and oxidized the L-DOPA-conjugated protein with HRP in the presence of H 2 O 2 . As the non-heme protein we chose lysozyme, which has a molecular mass similar to cyt c (14,300 Daltons versus 12,360 Daltons) and contains eight cysteinyl residues available for covalent cross-linking on the protein surface. The heterobifunctional cross-linker SIA was employed to conjugate the amino group of L-DOPA (formation of stable amide bond through condensation) and the sulfhydryl group of lysozyme (through carboxylmethylation), yielding L-DOPA-conjugated lysozyme (L-DOPA-LYS) (Fig. 5, inset).
The protein-derived ToQ . was generated by one-electron oxidation of the immobilized L-DOPA. At pH 7.4, the direct ESR spectrum from the reaction system of L-DOPA-LYS/HRP/H 2 O 2 took the form of a singlet with a broad line width of ϳ7 G (Fig.  5A). The radical was stable for up to 1 h without ESR signal decay (data not shown). The line width was ϳ3 G less than that (ϳ10 G) from the system of cyt c/H 2 O 2 . Two possible factors may contribute to the difference in line width: 1) The immobilized L-DOPA is not involved in peptide bond formation inside the protein but is covalently linked to the side chain of cysteinyl residues of lysozyme through the SIA with an arm 1.5 Å long, and 2) unlike tyrosine residues of cyt c, the conformation of protein (lysozyme here) may not restrict the molecular rotation of the L-DOPA. Both factors may lead to faster motion of the tyrosyl moiety and a narrower line width. The radical detected from the oxidation of a lysozyme was also much less stable and gradually decayed after the second scan (data not shown).
When the reaction was carried out in the L-DOPA-LYS/HRP/ H 2 O 2 system at pH 5.5, a weaker ESR signal with the same line width was obtained (Fig. 5B). The addition of Zn 2ϩ to the system resulted in an increase of the detected ESR signal (approximately two times the original), consistent with a tyrosine ortho-semiquinone radical (Fig. 5C).
It has been reported that proteins containing some of the post-translational modified tyrosine side chains (such as DOPA) exhibit unusual silver staining properties (28). DOPA can effectively reduce silver ions to silver metals in alkaline development conditions of the silver staining procedures because of 3,4-dihydroxylphenol. Therefore, dopa-proteins significantly enhance the silver reduction in SDS-polyacrylamide gel.
Cyt c-derived ToQ . can be reduced to yield DOPA-cyt c, which is expected to increase silver reduction in polyacrylamide gels. To enhance the yield of cyt c-derived ToQ . (Fig. 1), native cyt c was pretreated with a low concentration of HOCl to enhance its peroxidase activity prior to H 2 O 2 induction. Excess hydrogen peroxide in the reaction mixture was removed by catalase, and cyt c was reduced with dithiothreitol prior to SDS-PAGE and silver staining. As indicated in Fig. 6, significant enhancement of silver staining in the polyacrylamide gel was detected for the H 2 O 2 -treated cyt c (Fig. 6, lane A), suggesting that the formation of ToQ . by H 2 O 2 contributes the above signal enhancement of silver staining. The signal enhancement was substantially weakened by the addition of spin trap into the reaction mixture (Fig. 6, lane B), further supporting the evidence of peroxide-induced ToQ . generation.
Nitric Oxide Trapping of Tyrosyl Radical Leads to Cyt c Nitration as Detected by Immunochemistry-The endothelium relaxant factor NO has been shown to react with a tyrosyl radical prosthetic group on protein during enzyme turnover of ribonucleotide reductase (29,30), photosystem II (31), and prostaglandin H synthases I and II (32). NO can quench tyrosyl radical to form a tyrosyl radical-NO complex, which then undergoes rearrangement to generate a series of reaction intermediates that eventually lead to tyrosine nitration. The mech-  A and B) and their corresponding spectra of spin stabilization with Zn 2؉ at pH 5.5 (C and D). A, cyt c (0.5 mM) in 50 mM acetate-acetic acid buffer at pH 5.5 containing 1 mM DTPA was mixed with H 2 O 2 (2.5 mM). The spectrum was from the initial scan. B, same as A, except spectrum was from the second scan. C, same as A, except Zn 2ϩ was included in the system prior to the addition of H 2 O 2 . D, same as spectrum C, except the spectrum was from the second scan. ESR parameters are identical to those for Fig. 2, except that receiver gain was 1.00 ϫ 10 5 . M), NO was consumed, as detected electrochemically (Fig. 7, solid curve of trace b). The addition of H 2 O 2 (100 M) to the system increased NO consumption (Fig. 7, solid curve of trace b), implying the activity of NO oxidase for mitochondrial cyt c as seen in the myeloperoxidase (34), whereas the inclusion of the spin trap DMPO or MNP inhibited it (Fig. 7, dotted curve of  trace c). The quenching of cyt c-derived tyrosyl radical by NO was further verified by ESR spin-trapping experiments. Including the NO donor diethylamine nonoate (DEA/NO) in a solution containing cyt c, H 2 O 2 , and MNP significantly reduced the ESR intensity of the tyrosyl radical in a dose-dependent manner (Fig. 8A).
The latter reaction mixtures were recovered from the ESR flat cell and subjected to SDS-PAGE; the protein band was then transferred to the nitrocellulose membrane and immunoblotted with anti-3-nitrotyrosine antibody. The signal intensity of the Western blot was proportional to the amount of DEA/NO (Fig.  8B), suggesting that cyt c nitration is the consequence of the reaction of NO with the cyt c-derived tyrosyl radical.
To further clarify the NO-dependent nitration of cyt c, we conducted the experiment under the same conditions in the absence of spin trap. No 3-nitrotyrosine was formed without H 2 O 2 (Fig. 9), indicating the essential role of the tyrosyl radical. Including superoxide dismutase in the system did not affect cyt c nitration, suggesting that the reaction is independent of peroxynitrite. Finally, replacement of NO with nitrite did not cause significant nitration, thus ruling out the involvement of the nitric dioxide radical derived from the one-electron oxidation of nitrite. It is worth noting that the addition of cyanide (up to 200 eq) inhibited the cyt c nitration, indicating the essential role of heme in the nitration process (data not shown).

The Involvement of H 2 O 2 -induced Protein Radical in Apoptosis of Cardiac Myoblast-
The biological significance of cyt c-derived protein radicals has been speculated to be a role in the preapoptotic events via a sequential process of radical transfer, phospholipid peroxidation, and membrane permeability transition (35). It is important to provide experimental evidence to support the above hypothesis. To address this issue, rat heart myoblast was cultured to 90% confluent. H 2 O 2 (0 -200 M) was added to induce apoptosis, and the cells were cultured for 4 h. The nuclear DNA was isolated by phenol/isoamyl alcohol extraction and then analyzed by agarose gel electrophoresis. As shown in Fig. 10A, nuclear DNA fragmentation was detected for the H 2 O 2 -treated cardiac myoblast. Specifically, we have observed that the fraction of induced DNA fragmentation from intact nuclear DNA is peroxide dosage-dependent up to 200 M. To see whether cyt c release was accompanied by the detection of DNA fragmentation, the cytosol of H 2 O 2treated myoblast was prepared by differential centrifugation and then subjected to SDS-PAGE and Western blot analysis using anti-cyt c monoclonal antibody. As indicated in Fig. 10B, release of cyt c to cytosol was detected in the H 2 O 2 -treated myoblast. To further address the role of cyt c-derived radical in this event, the H 2 O 2 -treated cell was cultured with DMPO (40 mM) to quench cyt c-derived radical. As a result of spin trapping, the cyt c release to cytosol can be efficiently inhibited (Fig.  10C), implicating the involvement of cyt c-derived radical. In a parallel experiment, we added ascorbate (500 M) to the culture, which will prevent cyt c-derived radical formation (36). In any case, as a result of the antioxidant effect, the cyt c release was prevented (Fig. 10C). Both results suggest the contributory role of cyt c-derived protein radical in the event of apoptosis. DISCUSSION In the current investigation, we have provided direct evidence for the formation of ToQ . derived from protein. With a combination of Zn 2ϩ spin stabilization and the oxidation of the L-DOPA-conjugated lysozyme model compound, we have characterized the cyt c-derived ToQ . generated from H 2 O 2 . The use of silver staining under alkaline conditions to detect an en-hanced and MNP-sensitive signal derived from DOPA-cyt c provides additional evidence for the formation of cyt c-bound ToQ . induced by H 2 O 2 (Fig. 6). Our spin-trapping study showed that the ToQ . is likely derived from the oxidation of a shortlived tyrosyl radical, presumably via H 2 O 2 , because the ESR spectral intensity of the cyt c ortho-semiquinone radical was enhanced proportional to the amount of H 2 O 2 (pH 7.4, room temperature; data not shown).
The formation of the short-lived tyrosyl radical is due to electron transfer from vicinal tyrosine residue(s) to the porphyrin cation radical (analogous to HRP compound I) (8,20,23). The tyrosyl radical can be further oxidized to the orthosemiquinone radical by incorporating one oxygen into the C 3 position of the tyrosine phenoxyl group, as verified by Zn 2ϩ spin stabilization (Scheme I). This result is also consistent with MNP trapping at the C 3 position in the carbon-centered radical (37).
According to Kalyanaraman et al. (27), the possible toxicity of ortho-semiquinone radical arises from its potential (or that of the corresponding quinone) in a biological milieu to bind covalently to protein, enzyme, or other endogenous constituents, producing secondary free radicals and products. In fact, a similar singlet ESR signal with a line width of 10 G has been detected from the peroxynitrite-treated macrophages (37).
Oxidative alteration of cyt c has been linked to proapoptotic events. This hypothesis gains support from the experimental evidence in which exposure of cells to an appropriate concentration of H 2 O 2 is capable of causing cyt c release and triggering subsequent nuclear DNA fragmentation (Fig. 10). The participation of cyt c-derived radical intermediate during apoptosis is plausible because both cyt c release and DNA fragmentation as induced by peroxide can be partially inhibited by spin trap and radical scavenger (ascorbate in this work). In agreement with this study, available evidence reported in the literature includes: 1) H 2 O 2 has been reported to induce apoptosis by causing the loss of the membrane potential (⌬⌿), which results from the damage of the membrane structure by free radical generation (38), and 2) cyt c release from mitochondria can be inhibited by H 2 O 2 scavengers (39). It has been shown that antioxidants and antioxidant enzymes can prevent H 2 O 2 -induced apoptosis through the attenuation of lipid peroxidation (40,41). 3) A peroxidase compound I-type species of cyt c (produced from the reaction of cyt c with H 2 O 2 ) has been proposed as the major player for the redox changes (e.g., thiol oxidation to induce the permeability transition pore) that occur during apoptosis (42). However, the possibility of inhibiting thiol oxidation should not be omitted by either an antioxidant or a spin trap thereby preventing the permeability transition pore from opening.
It is well known that NO produced from endothelium cells has the capacity to modulate mitochondrial functions such as regulation of metabolism, respiration, and mitochondrial biogenesis (18,43). The interaction of NO with mitochondrial ferricytochrome c has been reported to form a ferricytochrome c-NO complex with a rate constant of 1 ϫ 10 3 M Ϫ1 s Ϫ1 and a dissociation constant of 22 M (44). Therefore, cyt c has been suggested as a reversible sink for excess NO in mitochondria.
Our experimental results here provide new insight regarding the irreversible consumption of NO by cyt c, leading to the nitration of cyt c during oxidative stress. This nitration is peroxynitrite-independent and derived from the interaction of H 2 O 2 -induced protein tyrosyl radical with NO. Under normal physiological conditions, mitochondria are an important intracellular source of H 2 O 2 (produced from state 4 respiration). However, under certain pathophysiological conditions such as inflammation and ischemia reperfusion injury, intracellular H 2 O 2 production can be greatly enhanced, and we have shown that the presence of hydrogen peroxide greatly enhances NO Spectrum a, H 2 O 2 (2.5 mM) was added to initiate the reaction prior to ESR measurement. Spectrum b, same as spectrum a, but DEA/NO (5 mM) was added prior to H 2 O 2 initiation. Spectrum c, same as spectrum a, but DEA/NO (10 mM) was added prior to H 2 O 2 initiation. B, sample from A was recovered from the ESR flat cell and diluted 10 times with argon-saturated phosphate buffer. Excess H 2 O 2 was removed by incubation with catalase (666 units/ml) at room temperature for 10 min. Protein (10 g) was subjected to SDS-PAGE and Western blot as described under "Experimental Procedures." consumption by cyt c (Fig. 7). Similar physiological phenomena have also been observed in the consumption of NO by leukocyte-derived myeloperoxidase (34) and by the aortic walls of blood vessels. 2 Protein-derived tyrosyl radical thus can contribute to either physiological or pathophysiological regulation of NO metabolism in endothelial cells.
It has been documented that NO could have both cytotoxic and cytoprotective effects in the vasculature (45). Elevated levels of NO in response to stimulation have been reported to cause cell injury (45). Indeed, the nitration of cyt c has been shown to impair its ability to support state 3 respiration in the mitochondrial inner membrane (10). As implied from the results of this work, we can logically expect that cyt c nitration can be synergistically enhanced when the levels of both H 2 O 2 and NO are elevated under certain pathophysiological conditions such as ischemia reperfusion injury.
The nitration of protein tyrosine residues has been used as a fingerprint for in vivo production of reactive nitrogen species (46,47). There is a growing body of evidence that protein nitration is intimately correlated with the pathogenesis of neurodegeneration and cardiovascular disease. Specifically, cyt c nitration has been detected during chronic allograft nephropathy (48) and renal ischemia/reperfusion (49). Peroxynitrite has been implicated as a causative factor for the above events. However, we suggest that the pathway independent of peroxynitrite proposed in this work is potentially contributory as well.
The anti-apoptotic effect of NO has been reported in endothelial cells in which multiple factors are involved (45). The protection by NO against cell death caused by H 2 O 2 stimulus is interesting. Brookes et al. (50) have shown that NO effectively inhibits the permeability transition and subsequent cyt c release from the mitochondria. Indirect up-regulation of Bcl-2 by NO has been suggested to participate in this process (51). However, we observed that nitration can even be detected at low concentrations of cyt c (NO/cyt c, ϳ0.1-0.5; data not shown), suggesting that NO can efficiently scavenge cyt cderived tyrosyl radical and subsequent ToQ . as induced by H 2 O 2 . Therefore, our results here provide a possible linkage among H 2 O 2 , NO, cyt c, permeability transition, and apoptosis through the role of cyt c-derived tyrosyl radical.
In conclusion, the present studies clarify the molecular mechanism of oxidative stress-induced alterations of cyt c caused by H 2 O 2 . The mechanism provides a useful concept for understanding the fundamental question of how cyt c-derived protein radical formation is related to the cytotoxicity role of H 2 O 2 and perhaps the cytoprotective role of NO in apoptosis resulting from phospholipid peroxidation, cyt c leakage, and caspase activation. Recognition of this peroxynitrite-independent pathway for cyt c nitration is important in understanding the fundamental basis by which oxidants and nitric oxide modulate mitochondrial function in a variety of diseases associated with inflammation and oxidant toxicity.