Cytochrome c Nitration by Peroxynitrite*

Peroxynitrite (ONOO 2 ), the product of superoxide (O 2 .) and nitric oxide ( z NO) reaction, inhibits mitochondrial respiration and can stimulate apoptosis. Cytochrome c , a mediator of these two aspects of mitochondrial function, thus represents an important potential target of ONOO 2 during conditions involving accelerated rates of oxygen radical and z NO generation. Horse heart cytochrome c 3 1 was nitrated by ONOO 2 , as indicated by spectral changes, Western blot analysis, and mass spectrometry. A dose-dependent loss of cytochrome c 3 1 695 nm absorption occurred, inferring that nitration of a critical heme-vicinal tyrosine (Tyr-67) promoted a conformational change, displacing the Met-80 heme ligand. Nitration was confirmed by cross-reactivity with a specific antibody against 3-nitrotyrosine and by increased molecular mass compatible with the addition of a nitro-(-NO 2 ) group. Mass analysis of tryptic digests indicated the preferential nitration phase column (5 m m; 4.6 3 250 mm; Jupiter; Phenomenex). Solvent A was 0.1% trifluoroacetic acid in ultra-pure water, and solvent B was acetonitrile. Peptides were eluted using an increasing linear gradient of solvent B from 0–45% in 60 min with a flow rate of 1 ml/min. The HPLC detector was set at 210, 275, and 365 nm. The new peptide peaks in the chromatogram of peroxynitrite-treated cytochrome c 3 1 , with peptides displaying absorbance maxima at 365 nm, were collected and subjected to molecular mass determination. chromatography-mass capillary and and (4 a in formic over the min. Percentages of native and modified cytochrome c were estimated from the integrated areas under the curves. The peptides from HPLC fractionation were concentrated in vacuo and by (VG Quattro with ESI and as

the inner mitochondrial membrane and participates in mitochondrial electron transport (1)(2)(3). The release of cytochrome c from mitochondria to the cytosol via a Bcl-2-inhibitable mechanism constitutes an early event in apoptosis and can occur before changes in mitochondrial membrane potential (4 -11). The mechanisms mediating cytochrome c release from mitochondria and how this event triggers apoptosis in the cytosol remain to be defined.
Cytochrome c is a small globular protein that contains a covalently bound heme located in an internal pocket formed by highly conserved amino acid residues (12)(13)(14). The porphyrin is covalently bound to Cys-14 and Cys-17, with the fifth and sixth coordination positions of the heme-Fe interacting with His-18 and Met-80, respectively. In addition to these four invariant amino acids, there are others that have been conserved during evolution, e.g. four tyrosine residues (Tyr-48, Tyr-67, Tyr-74, and Tyr-97) and nine lysine residues that give the protein its characteristically high isoelectric point, close to pH 10 (14).
Cytochrome c exists in high concentrations in mitochondria (ϳ400 M) (35) and may also become nitrated during conditions involving accelerated rates of oxygen radical and ⅐NO generation. We have previously reported the direct one-electron oxidation of the heme of cytochrome c 2ϩ by ONOO Ϫ (k ϭ 2 ϫ 10 5 M Ϫ1 s Ϫ1 ) (36), but modification of the protein moiety induced by ⅐ NO-derived reactive species has not been studied. In this regard, nitration of cytochrome c tyrosine residues can have a profound influence on protein structure and function because * This work was supported by National Institutes of Health Grants RO3-TW0099 (to R. R. and B. A. F.), HL54926 and AG13966 (to H. I.), and HL58418, HL58115, and HL51245 (to B. A. F.) and by grants from the Comisión Sectorial de Investigación Cientifica (Uruguay), International Centre for Genetic Engineering and Biotechnology (Italy), and Swedish Agency for Research Cooperation (Sweden) (to R. R.). The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
* two tyrosines (Tyr-67 and Tyr-48; Fig. 1) lie adjacent to the critical catalytic environment of the heme pocket. Indeed, early work indicated that exposure to excess amounts of the nitrating and oxidizing compound tetranitromethane (TNM) resulted in tyrosine nitration and was accompanied by changes in cytochrome c physicochemical properties and respiratory function (37).
Here we report the nitration of mitochondrial cytochrome c by ONOO Ϫ , reveal the influence of this modification on cytochrome c function, and discuss the potential biological relevance of this reaction in the context of oxidant-mediated cell injury.
A rabbit polyclonal antibody against nitrotyrosine was raised with nitrated keyhole limpet hemocyanin and purified in our laboratory by affinity chromatography as described elsewhere (38). Peroxynitrite was synthesized, quantitated, and handled as described previously (39).
Cytochrome c Nitration-In some cases, nitration was performed by reaction of 40 mM tetranitromethane in 95% ethanol with cytochrome c 3ϩ (1 mM) in 0.1 M Tris-HCl and 0.1 M KCl (pH 8.0) at room temperature for 30 min. The final concentration of ethanol was always less than 6%. The reaction was terminated by passing the mixture through a Sephadex G-25 column in 0.1 M Tris-HCl and 0.1 M KCl, pH 8.0 (40).
During reaction with ONOO Ϫ , cytochrome c in 200 mM potassium phosphate and 100 M DTPA, pH 7.0, was subjected to bolus addition of ONOO Ϫ at 25°C. The pH was controlled after each addition and was always kept below pH 7.4. In some experiments, NaHCO 3 (25 mM) was added. Control experiments (reverse order addition experiments) in which ONOO Ϫ was first decomposed in the reaction buffer before the addition of cytochrome c were performed systematically to estimate the potential effects of the byproducts nitrite (NO 2 Ϫ ) and nitrate (NO 3 Ϫ ). Biochemical Analyses-Peroxidatic activity of native or ONOO Ϫtreated cytochrome c (0.6 M) in 100 mM potassium phosphate plus 100 M DTPA, pH 7.2, was assayed with 1.3 mM ABTS and 12 mM H 2 O 2 at 20°C. ABTS oxidation was followed at 420 nm, ⑀ ϭ 36 mM Ϫ1 cm Ϫ1 (41). Mitochondrial respiration was measured polarographically using a Cole-Palmer oxymeter fitted with a water-jacketed Clark-type electrode (YSI Model 5300) in a 1.0-ml reaction vessel. Oxygen consumption studies were performed in mitochondrial homogenization buffer at 37°C, pH 7.4, using 0.2-0.5 mg/ml mitochondria. Succinate (5 mM) was used to quantitate complex II-dependent respiration (21). All spectrophotometric measurements were performed in a Shimadzu UV-2401 PC spectrophotometer. Protein concentration was determined by the Bradford method (42).
Mitochondrial Preparation-Intact rat heart mitochondria were prepared by differential centrifugation as described previously (15,21). Mitochondrial pellets were resuspended in a minimal volume of the homogenization buffer containing 0.3 M sucrose, 5 mM potassium phosphate, 1 mM EGTA, 0.1% bovine serum albumin, pH 8.0, to 25-35 mg protein/ml and kept at 4°C until use (21). Respiratory control ratio for complex II-dependent respiration typically ranged between 3 and 5. For cytochrome c extraction, mitochondrial pellets were resuspended in 10 mM KCl and 2 mM Tris-HCl, pH 7.4, to a protein concentration of 2 mg/ml and then divided in two fractions, A and B. Samples were incubated for 10 min at 37°C and then centrifuged at 10,000 ϫ g at 4°C. The pellet of fraction A was resuspended in the same buffer, whereas pellet B was resuspended in 150 mM KCl and 2 mM Tris-HCl, pH 7.4, to extract cytochrome c (43). Fractions A and B were incubated and centrifuged twice using the same conditions, and supernatants were collected and measured spectrophotometrically after the addition of dithionite. The concentration of cytochrome c 2ϩ was determined by absorbance at 550 nm (⑀ ϭ 21 mM Ϫ1 cm Ϫ1 ) (43). Typically, 0.6 Ϯ 0.1 nmol cytochrome c/mg mitochondrial protein was extracted.
Tryptic Digestion and Peptide Mapping-Cytochrome c 3ϩ samples (control and samples treated with peroxynitrite) were exhaustively dialyzed against 100 mM ammonium bicarbonate, pH 8.0, and cleaved with sequencing grade-modified trypsin in a 1:100 ratio (w/w) at 37°C for 16 h. The peptides were analyzed by a Hewlett Packard HPLC system with a diode array detector using an octadecyl silica gel reverse- phase column (5 m; 4.6 ϫ 250 mm; Jupiter; Phenomenex). Solvent A was 0.1% trifluoroacetic acid in ultra-pure water, and solvent B was acetonitrile. Peptides were eluted using an increasing linear gradient of solvent B from 0 -45% in 60 min with a flow rate of 1 ml/min. The HPLC detector was set at 210, 275, and 365 nm. The new peptide peaks in the chromatogram of peroxynitrite-treated cytochrome c 3ϩ , with peptides displaying absorbance maxima at 365 nm, were collected and subjected to molecular mass determination.
Mass Spectrometry Studies-Electrospray ionization mass spectrometry was performed on a PE-Sciex (Concord, Ontario, Canada) API-III triple-quadrupole mass spectrometer equipped with an atmospheric pressure ion source. Positive ion mass spectra were acquired for capillary liquid chromatography-mass spectroscopy. For the capillary liquid chromatography-mass spectroscopy analyses, the effluent from a 300 m i.d. ϫ 15 cm capillary Vydac C18 column (LC-Packings, San Francisco, CA) was introduced directly into the ionization needle of the mass spectrometer. The protein was dissolved to a final concentration of 0.5 mg/ml and loaded in and eluted with 0.1% formic acid (4 min), followed by a linear gradient of 0 -80% acetonitrile in 0.1% formic acid over the next 10 min. Percentages of native and modified cytochrome c were estimated from the integrated areas under the curves.
The peptides collected from HPLC fractionation were concentrated in vacuo and analyzed by electrospray mass spectrometry (VG Quattro Quadrapole mass spectrometer with ESI source and Masslynx software) as described previously (44).

Spectroscopic Analysis of Cytochrome c Nitration-Cyto-
chrome c 3ϩ (200 M) was exposed to successive additions of 3 mM ONOO Ϫ , and spectral changes in the visible region were observed. Peroxynitrite caused a dose-dependent disappearance of the characteristic 695 nm absorbance of native cytochrome c 3ϩ at pH 7.0, indicating loss of Met-80-heme iron interactions ( Fig. 2; Ref. 14). This spectral change may be due to the nitration of Tyr-67, which promotes the displacement of Met-80 from its coordination bond at the Fe atom (45,46). Because yields of phenolic nitration by ONOO Ϫ usually increase in the presence of CO 2 through the intermediate formation of nitroso-peroxocarboxylate (ONO 2 CO 2 Ϫ ) (47-49), a comparative study of the spectral changes of cytochrome c mediated by ONOO Ϫ , with and without CO 2 and TNM exposure, was performed (Fig. 3). Peroxynitrite at a cumulative dose of 18 mM caused a similar decrease in the 695 nm band with or without CO 2 , whereas TNM (40 mM) led to more pronounced but comparable spectral changes (Fig. 3A). We next analyzed the 605-615 nm absorbance of cytochrome c, in which TNM reaction with cytochrome c reveals a characteristic shoulder (45). In this region, the reaction of ONOO Ϫ yielded an effect similar to that of TNM in both the presence and absence of CO 2 (Fig. 3B). Finally, a blue shift of 2-3 nm in the Soret band was observed after all three nitrating reactions (Fig. 3C). A moderate loss of Soret absorption was observed, especially in the case of TNM, indicating that at high concentrations (40 mM), there may have been some opening of the heme. In contrast, similar concentrations of H 2 O 2 led to total heme loss (41,50).
In native cytochrome c 3ϩ , the 695 nm band has a pK a of 9.5 because of the conformational transition between states III and IV, which involves the substitution of the Met-80 from the heme, presumably by Lys-79 (14). After the TNM reaction, the pK a of the transition decreases to 6.5 because the nitration of Tyr-67 causes pK a lowering (45) and affects the liganding tendency of Lys-79 to the heme iron. The lost 695 nm band in cytochrome c after ONOO Ϫ treatment at neutral pH was partially recovered at an acidic pH (i.e. pH 5) but not at an alkaline pH (i.e. pH 8.8; data not shown), resembling the effect of TNM and further suggesting the nitration of Tyr-67. Electrophoretic Analysis of Native and Nitrated Cytochrome c-The addition of ONOO Ϫ to cytochrome c results in protein nitration, as revealed by reactivity with anti-nitrotyrosine antibody. Extents of nitration by ONOO Ϫ were dose-dependent. Importantly, the oxidized form of cytochrome c (cytochrome c 3ϩ ) was more extensively nitrated than the reduced form (cytochrome c 2ϩ ), supporting a preferential reaction between reduced heme and ONOO Ϫ (36) (Fig. 4). At greater concentrations of ONOO Ϫ , a second nitrated band of approximately 24 kDa appears, compatible with the formation of dimerized cytochrome c, similar to hemoglobin nitration by ONOO Ϫ (51). Tetranitromethane-treated cytochrome c also had cross-reactivity with the anti-nitrotyrosine antibody (data not shown).
Native gel electrophoresis shows that ONOO Ϫ causes a dosedependent appearance of up three species, displaying decreased migration toward the cathode (Fig. 5), whereas native cytochrome c progressively diminished in staining intensity. The electrophoretic properties of these novel cytochrome c species infer a decreased isoelectric point compatible with the nitration of tyrosine residues and a consequent lowering of pK a (52), as observed previously during ONOO Ϫ -mediated nitration of bovine CuZn-SOD (53). Similar products were observed when ONOO Ϫ was added to cytochrome c in the presence of CO 2 . High concentrations of TNM (40 mM) yielded a band pattern similar to that observed with ONOO Ϫ , plus an additional fifth band close to the anode (Fig. 5, lane 9).
Mass Spectrometry Studies and HPLC Analysis of Tryptic Digests-Exposure of cytochrome c 3ϩ (200 M) to 0.5 mM ONOO Ϫ resulted in a 46-Da increase in the mass (12,356 to 12,402 Da; Fig. 6B) of ϳ 20% of the parent protein. This is consistent with the addition of one nitro group (45 Da) per molecule of cytochrome c. At 2 mM ONOO Ϫ , there was a further increase in the 12,402 Da species (from 20% to 28%), plus the appearance of a new species (8%) of increased molecular mass of 91 Da (12,447 Da) with respect to the parent protein, indicating the addition of two nitro groups to a small fraction of the protein (Fig. 6C). At a cumulative dose of 18 mM ONOO Ϫ , the 12,447 Da (i.e. containing two 3-nitrotyrosine residues) species was predominant, and other molecular mass species were detected as well, in particular, (a) a component in the region of 12,491 Da compatible with the addition of a third nitro group, and (b) a component of 12,415 compatible with the addition of one nitro group and one oxygen (data not shown). Similarly, cytochrome c 3ϩ treatment with 40 mM TNM resulted in the formation of three main species, consistent with mononitrated, dinitrated, and, to a lesser extent, trinitrated forms of the protein (data not shown).
The reverse-phase HPLC elution profile of peptides generated by tryptic digestion of cytochrome c after treatment with 18 mM ONOO Ϫ (Fig. 7B) was compared with native cytochrome c (Fig. 7A). The same chromatographic profile was obtained with cytochrome c 3ϩ after treatment with TNM (data not shown). To identify the site(s) of nitration, tryptic peptides of ONOO Ϫ -treated cytochrome c having absorbance at 365 nm were collected and subjected to mass spectrometric analysis. The peptide eluting at 32 min (the first asterisk in Fig. 7B)  showed the same peak at 42.4 min and the same mass of the peptide but did not include a peptide eluting at 32 min. Thus, at lower ONOO Ϫ concentrations, only nitration of Tyr-67 is observed, whereas at higher concentrations of ONOO Ϫ , Tyr-67 and Tyr-48 are nitrated. Although mass spectrometric analysis of cytochrome c exposed to 40 mM TNM or 18 mM ONOO Ϫ showed an increase in mass suggestive of nitration of three tyrosine residues, a third nitrated tyrosine was not apparent after tryptic digestion and HPLC-mass spectrometry analysis. Sokolovsky et al. (37) reported nitration of Tyr-67 and Tyr-48 after exposure to a 60-fold excess of TNM. They also reported a small amount of nitrated residue Tyr-74 (37). It is possible that the third nitrated tyrosine (Tyr-74) was not detected after tryptic digestion and HPLC-mass spectrometry analysis because of the low yield of nitration of this residue and/or the small size of the tryptic fragment (723.4 Da) where Tyr-74 is located.
Changes in Cytochrome c Redox and Catalytic Properties-Cytochrome c 3ϩ induces ABTS oxidation in the presence of H 2 O 2 , with this catalytic activity of cytochrome c increased by preincubation with H 2 O 2 (41). Peroxynitrite-treated cytochrome c 3ϩ was also a better catalyst of ABTS oxidation than native cytochrome c (Fig. 8), increasing the rate of ABTS oxidation in a dose-dependent manner. The activation of this peroxidase activity of cytochrome c is related to nitration. In the absence of electron donors such as ABTS, cytochrome c 3ϩ pretreated with ONOO Ϫ and TNM becomes more sensitive to H 2 O 2 -mediated inactivation, as evidenced by the facile loss of the Soret absorption band (Fig. 8, inset). This supports the concept that nitration of Tyr-67 promotes a displacement of the sixth ligand position of the heme, favoring the interaction with H 2 O 2 and oxidative degradation of the heme via formation of oxo-iron complexes (i.e. ferryl iron) (41). Finally, nitrated cytochrome c 3ϩ was resistant to reduction by ascorbate (Fig. 9).
Mitochondrial Respiration after Nitration of Cytochrome c-Intact rat heart mitochondrial respiration was studied in the presence of native and ONOO Ϫ -nitrated cytochrome c 3ϩ . State 4 mitochondrial respiration was inhibited 70% upon cytochrome c extraction (Fig. 10, A and B), affirming the essential role of cytochrome c in electron transport (2,3). Supplementation of cytochrome c-depleted mitochondria with native cytochrome c 3ϩ restored oxygen consumption rates to control values (Fig. 10B). Re-supplementation of mitochondria with ONOO Ϫ -treated cytochrome c restored less than 50% respiration rates (Fig. 10C). It is important to note that the treatment for nitration used here (six bolus additions of 1 mM ONOO Ϫ to 200 M cytochrome c 3ϩ ) led to nitration of Tyr-67 (Figs. 6 and 7), but also left a significant fraction of native cytochrome c 3ϩ as well (Figs. 5 and 6), the latter being the main one, responsible for the partial recovery of respiration (Fig. 10C). Similar results were obtained with cytochrome c nitrated by TNM exposure (data not shown), in line with previous results (45). DISCUSSION Peroxynitrite is capable of nitrating cytochrome c 3ϩ , as demonstrated by Western blot analysis and mass spectrometry studies. The nitrating capability of ONOO Ϫ was comparable to that of TNM. Cytochrome c 3ϩ reaction with ONOO Ϫ led to changes in spectral properties, most notably the disappearance of the absorption band at 695 nm, indicating that Met-80 ceases to coordinate with the Fe atom. This could be due to nitration of the nearby tyrosine residue Tyr-67 (45), but additional or alternative mechanisms such as Met-80 oxidation or modifica- tions of Trp-59 could not be a priori ruled out. However, the appearance of the absorption band in the 610 nm region after treatment with ONOO Ϫ , characteristic of TNM-nitrated cyto-chrome c (45), and the data obtained with mass spectral analysis of tryptic digests (Fig. 7B) indicate that the pattern of nitration by ONOO Ϫ resembles that of TNM and unambiguously supports the idea that Tyr-67 is being nitrated.
Changes in the electrophoretic mobility of cytochrome c on native gels (Fig. 5) correlated well with increased molecular mass (Fig. 6), indicating that new cytochrome c species displaying reduced migration toward the cathode corresponded to nitrated cytochrome c 3ϩ adducts (containing one, two, and possible three 3-nitrotyrosine moieties) displaying a lower isoelectric point than native cytochrome c 3ϩ . Indeed, whereas the dissociation of the phenolate in tyrosine has a pK of ϳ10.07, this value decreases to ϳ7.5 in free 3-nitrotyrosine (52) and to ϳ7.0 in nitrated Tyr-67 of cytochrome c 3ϩ (54). Peroxynitrite reaction with cytochrome c 3ϩ affected its redox properties and its respiratory function. There was a decrease in the reducibility of cytochrome c by ascorbate, with this functional modification potentially explaining the inability to recover basal respiration of cytochrome c 3ϩ -depleted mitochondria after reconstitution with ONOO Ϫ -treated cytochrome c 3ϩ . Also, an increase in the peroxidatic activity of cytochrome c 3ϩ was observed. This could be relevant because peroxidases have been shown to also catalyze nitration reactions in the presence of H 2 O 2 and NO 2 Ϫ (55). The biological relevance of these findings remains to be defined. First, cytochrome c 3ϩ nitration yields by ONOO Ϫ were not high because 0.5 mM ONOO Ϫ nitrated ϳ20% of 200 M cytochrome c 3ϩ (i.e. ϳ40 M; Fig. 6B), implying ϳ8% yield. Secondly, nitration of cytochrome c 2ϩ was attenuated due to preferential oxidation of the heme (36). Interestingly, cytochrome c nitration occurred in the presence of CO 2 , which is abundant in mitochondria and a catalyst of the intermediate formation of the nitrating species ONO 2 CO 2 Ϫ . However, the possibility that cytochrome c 3ϩ nitration would significantly impair mitochondrial respiration is remote in light of the high amounts of ONOO Ϫ required and the recognition that other more oxidant-sensitive electron transport components would be inactivated at lower ONOO Ϫ concentrations (15,21). On the other hand, small quantities of nitrated cytochrome c 3ϩ could mediate signaling reactions and/or serve as a footprint during ⅐NOand ONOO Ϫ -mediated injury and apoptotic processes, which involve cytochrome c 3ϩ release to the cytosol. Indeed, despite the various potential intramitochondrial protein targets susceptible of nitration, current work reveals that biologically relevant concentrations of ONOO Ϫ cause cytochrome c nitration 2 and release (11) in intact mitochondria. Moreover, nitrated cytochrome c may be released to the cytosol in cells exposed to ⅐NO (56), although the correlation between apoptotic cell death and cytochrome c nitration requires further investigation.
In various pathophysiological situations, mitochondrial ONOO Ϫ formation occurs readily due to the increased formation of O 2 . by the respiratory chain and the presence of enhanced NO synthesis in extramitochondrial or possibly intramitochondrial sites (57)(58)(59). Thus, cytochrome c 3ϩ could be nitrated by mitochondrial but also cytosolic or extracellularderived ONOO Ϫ diffusing to the intermembrane space. In summary, we report that peroxynitrite can mediate the nitration and functional modification of cytochrome c, a critical component of respiratory and apoptotic signaling reactions.
Acknowledgments-We thank Dr. Evgueni Daikhin and Marion Kirk for assistance with liquid chromatography-mass spectroscopy, Dr. Alfonso Cayota for advice in Western blot studies, Gerardo Ferrer-Sueta for help with software, and Paola Hodara for her contribution to the artwork. Respiration of control mitochondria (A) and cytochrome c-depleted mitochondria followed by supplementation with 0.7 mmol/mg native (B) or ONOO Ϫ -treated cytochrome c (C), is shown. The reaction vessel contained 1.6 ml of final reaction volume.