Tyrosine Nitration by Peroxynitrite Formed from Nitric Oxide and Superoxide Generated by Xanthine Oxidase*

Peroxynitrite (ONOO−) is a potent nitrating and oxidizing agent that is formed by a rapid reaction of nitric oxide (NO) with superoxide anion (O⨪2). It appears to be involved in the pathophysiology of many inflammatory and neurodegenerative diseases. It has recently been reported (Pfeiffer, S., and Mayer, B. (1998) J. Biol. Chem. 273, 27280–27285) that ONOO− generated at neutral pH from NO and O⨪2 (NO/O⨪2) was substantially less efficient than preformed ONOO− at nitrating tyrosine. Here we re-evaluated tyrosine nitration by NO/O⨪2 with a shorter incubation period and a more sensitive electrochemical detection system. Appreciable amounts of nitrotyrosine were produced by ONOO− formed in situ (2.9 μm for 5 min; 10 nm/s) by NO/O⨪2 flux obtained from propylamine NONOate (CH3N[N(O)NO]−(CH2)3NH2 +CH3) and xanthine oxidase using pterin as a substrate in phosphate buffer (pH 7.0) containing 0.1 mm l-tyrosine. The yield of nitrotyrosine by this NO/O⨪2 flux was approximately 70% of that produced by the same flux of preformed ONOO−(2.9 μm/5 min). When hypoxanthine was used as a substrate, tyrosine nitration by NO/O⨪2 was largely eliminated because of the inhibitory effect of uric acid produced during the oxidation of hypoxanthine. Tyrosine nitration caused by NO/O⨪2was inhibited by the ONOO− scavenger ebselen and was enhanced 2-fold by NaHCO3, as would be expected, because CO2 promotes tyrosine nitration. The profile of nitrotyrosine and dityrosine formation produced by NO/O⨪2 flux (2.9 μm/5 min) was consistent with that produced by preformed ONOO−. Tyrosine nitration predominated compared with dityrosine formation caused by a low nanomolar flux of ONOO− at physiological concentrations of free tyrosine (<0.5 mm). In conclusion, our results show that NO generated with O⨪2 nitrates tyrosine with a reactivity and efficacy similar to those of chemically synthesized ONOO−, indicating that ONOO− can be a significant source of tyrosine nitration in physiological and pathological events in vivo.

consistent with that produced by preformed ONOO ؊ . Tyrosine nitration predominated compared with dityrosine formation caused by a low nanomolar flux of ONOO ؊ at physiological concentrations of free tyrosine (<0.5 mM). In conclusion, our results show that NO generated with O 2 . nitrates tyrosine with a reactivity and efficacy similar to those of chemically synthesized ONOO ؊ , indicating that ONOO ؊ can be a significant source of tyrosine nitration in physiological and pathological events in vivo.
Nitric oxide (NO) 1 is a simple inorganic radical exhibiting diverse physiological functions, including the regulation of neu-rotransmission and vascular tone (1). It is also closely linked to the pathogenesis of many inflammatory and degenerative disorders. Many of the physiological actions of NO are mediated by cyclic guanosine monophosphate (cGMP)-dependent pathways through the activation of soluble guanylate cyclase (1,2). Secondary oxidants derived from NO leading to S-nitrosylation and nitration of endogenous biomolecules can also be involved in various physiological and pathophysiological phenomena through cGMP-independent pathways (3)(4)(5)(6).
Although the mechanisms responsible for tyrosine nitration in vivo are not fully understood (5), peroxynitrite (ONOO Ϫ ), a reaction product of NO and superoxide anion radical (O 2 . ) (22), is likely to be a major contributor because of its considerable potential to nitrate tyrosine, as evidenced by in vitro experiments (23)(24)(25) (1 mM). Oxidation of hypoxanthine by xanthine oxidase would lead to an accumulation of uric acid in the reaction mixture. Uric acid is well known to be a potent ONOO Ϫ scavenger (12,13,28). This result suggests that loss of tyrosine nitration by NO/O 2 . may be due to scavenging of ONOO Ϫ by uric acid. 2 Because xanthine oxidase activity is inhibited by NO (29,30)  considering the effect of uric acid. By using a more sensitive and specific 12-electrode electrochemical assay for the HPLC, we could measure nitrotyrosine formation, using the same NO and O 2 . flux as that used by Pfeiffer and Mayer (27), over a 5-min period before substrate depletion and product formation became significant problems. The formation of nitrotyrosine by cogeneration of O 2 . and NO was found to be nearly equivalent to that obtained with preformed ONOO Ϫ .
The concentration of ONOO Ϫ formed from NO/O 2 . was also estimated by the DCDHF oxidation assay according to the procedure reported by Crow (36). The reaction mixture, containing 0.1 mM DCDHF, 0.1 mM DTPA, 10 M P-NONOate, 50 M pterin, and 10 or 15 milliunits/ml xanthine oxidase in a total volume of 1.0 ml of 0.1 M sodium phosphate buffer (pH 7.0), was incubated at room temperature in the dark. After 5 min of incubation, oxidation of DCDHF was determined by measuring the increase in absorbance at 500 nm. A control reaction sample containing allopurinol (final concentration, 0.8 mM) was used to measure the background rate of DCDHF oxidation. The total amount of ONOO Ϫ formed from NO/O 2 . was estimated from a standard curve of DCDHF oxidation obtained with authentic ONOO Ϫ . Superoxide generation from xanthine oxidase was measured in a separate experiment by using the cytochrome c reduction assay as described previously (16). However, because O 2 . production could not be determined by the cytochrome c assay in the presence of NO, in the presence of an NO flux the amount of O 2 . produced was corrected by simultaneously quantifying the accumulation of isoxanthopterin. Specifically, when pterin was used as the substrate for xanthine oxidase, formation of isoxanthopterin in 0.1 M sodium phosphate buffer (pH 7.0) containing 0.1 mM DTPA, 10 M P-NONOate, and various concentrations of xanthine oxidase was measured fluorometrically, using an excitation wavelength of 345 nm and an emission wavelength of 390 nm (37). A linear correlation between isoxanthopterin formation and O 2 .
formation from xanthine oxidase was obtained in the absence of NO released from P-NONOate. In addition, isoxanthopterin formation from xanthine oxidase was not affected by NO released from P-NONOate under our experimental conditions. Therefore  (38)).
Effect of Uric Acid on Tyrosine Nitration-To examine the inhibitory effect of uric acid on tyrosine nitration, the reaction mixture of xanthine oxidase and hypoxanthine was incubated with or without uricase, which decomposes uric acid to give allantoin and CO 2 , and was then subjected to tyrosine nitration by authentic ONOO Ϫ . Hypoxanthine (50 M) was incubated with xanthine oxidase (10 milliunits/ml) in 0.1 M phosphate buffer containing 0.1 mM tyrosine and 0.1 mM DTPA at room temperature. After 5 min of incubation, allopurinol (final concentration, 0.2 mM) was added to the mixture to terminate the enzyme reaction. To the reaction mixture thus obtained was added uricase (final concentration, 420 milliunits/ml) or 0.1 M phosphate buffer (as a solvent control for uricase), and the mixture was further incubated at room temperature. After various incubation periods, aliquots of the reaction mixture were exposed to authentic ONOO Ϫ (bolus administration at final concentration, 3.0 M) at room temperature for 1 min, and nitrotyrosine was quantified by HPLC as described above. In addition, tyrosine nitration by both authentic ONOO Ϫ and an NO 2 -The electrochemical assay for nitrotyrosine used in this study could readily detect 2 nM nitrotyrosine in a 50-l sample (0.1 pmol), with linear results over the range 2 to 1000 nM (data not shown). With this sensitive assay, nitration was observed with as little as 50 nM preformed ONOO Ϫ added to 0.1 mM tyrosine in phosphate buffer (pH 7.0) with 25 mM NaHCO 3 (Fig. 1). Each result on the yield of nitrotyrosine, from 50 to 3000 nM preformed ONOO Ϫ , was equal to 13 Ϯ 2.3% (mean Ϯ S.D.) based on ONOO Ϫ . With no addition of NaHCO 3 to the reaction mixture, endogenous bicarbonate was 4.0 mM and CO 2 was 0.6 mM as determined using a CO 2  . produced with xanthine oxidase plus pterin in 5 min was one-fourth with that produced with xanthine oxidase plus hypoxanthine (Fig. 2B). Superoxide production by xanthine oxidase with either hypoxanthine or pterin as substrate remained linear for the 5-min incubation period (data not shown). These data indicate that no appreciable substrate depletion occurred under our experimental conditions. Moreover, the finding that the nitration yield and level of O 2 .
generation were inversely related for the two different xanthine oxidase substrates, suggested that inhibition by the hypoxanthine/xanthine oxidase system might account for the lack of tyrosine nitration by ONOO Ϫ . Carbon dioxide enhanced tyrosine nitration by both ONOO Ϫ and the NO/O 2 . system as reported previously (41,42). Addition , which was about 20% less than expected from the halflife of P-NONOate. After 25 min of incubation, however, the release of NO from P-NONOate (10 M) was approximately 18 M regardless of the presence of tyrosine (data not shown). The consumption of NO in 5 min after addition of 10 milliunits/ml xanthine oxidase to this system was 2.9 Ϯ 0.3 M (Fig. 3B), which reflected the total concentration of ONOO Ϫ formed in 5 min. Increasing the concentration of xanthine oxidase to 15 milliunits/ml raised the NO consumption to 4.0 Ϯ 0.1 M (Fig.  3C). Allopurinol blocked consumption of NO by the pterin/ xanthine oxidase system (Fig. 3D), and this system had no effect on PTIO in the absence of NO (data not shown).
To verify that the amount of ONOO Ϫ produced by the NO/ xanthine oxidase system was correctly estimated by our ESR assay, we measured ONOO Ϫ formation by using the DCDHF oxidation assay (36). The values determined by the DCDHF assay were consistent with those obtained by the ESR-based PTIO liposome assay: the amount of ONOO Ϫ formed in the reaction of 10 milliunits/ml xanthine oxidase plus 50 M pterin with 10 M P-NONOate for 5 min was estimated to be 3.0 Ϯ 0. and Authentic ONOO Ϫ -To determine the efficiency in nitration of the NO/O 2 . system, 2.9 M preformed alkaline ONOO Ϫ was continuously infused during a 5-min period into the same buffer system containing 0.1 mM tyrosine, with a yield of 171 Ϯ 20 nM nitrotyrosine. The NO/O 2 . system using pterin with 10 milliunits/ml xanthine oxidase yielded 124 Ϯ 5.3 nM nitrotyrosine, which was 73% of the yield with authentic ONOO Ϫ (Table I).
We also compared the yields of nitrotyrosine produced by ONOO Ϫ obtained in this study with those of Pfeiffer and Mayer under similar reaction conditions (pH 7.4; tyrosine concentration, 1.0 mM). In general, the yield of tyrosine nitration occurring in the reaction mixture at pH 7.4 was lower than that at pH 7.0. The nitration yield by NO/O 2 . fluxes produced by our system of P-NONOate/pterin plus xanthine oxidase was much greater than that measured by Pfeiffer and Mayer with spermine NONOate/hypoxanthine plus xanthine oxidase. However, ONOO Ϫ generated in situ was as effective at nitrating tyrosine as preformed ONOO Ϫ even with a higher concentration of tyrosine (1.0 mM) ( Table I). The lower yield of nitrotyrosine produced by authentic ONOO Ϫ at pH 7.4 obtained in this study compared with that obtained by Pfeiffer and Mayer may be due to the different ONOO Ϫ concentrations used. In this study, 2.3 M ONOO Ϫ was used. With such a low concentration of ONOO Ϫ , the nitrotyrosine yield was largely affected by tyrosine concentration (cf. Fig. 4). The yields of nitrotyrosine produced by NO/O 2 . flux and preformed ONOO Ϫ were examined with different concentrations of tyrosine at pH 7.0. The yield of nitrotyrosine produced by a low flux of ONOO Ϫ (2.9 M for 5 min, 10 nM/s) decreased at higher tyrosine concentrations; tyrosine nitration was most effective when the tyrosine concentrations were 0.1 mM or below (Fig. 4). van der Vliet et al. (25) previously reported similar results. In contrast, dityrosine formation was elevated in a tyrosine concentration-dependent manner and was inversely related to tyrosine nitration. Very similar profiles of nitrotyrosine and dityrosine production were obtained with ONOO Ϫ formed in situ by NO/O 2 . flux (P-NONOate/pterin plus xanthine oxidase) and preformed ONOO Ϫ . The yield of nitrotyrosine was much higher than that of dityrosine when tyrosine was used at 0.1 mM or below (Fig. 4), indicating that nitrotyrosine is a major product under such reaction conditions, with a low flux of ONOO Ϫ . Similar profiles of tyrosine modifications produced by a low flux of ONOO Ϫ were also observed with different concentrations of tyrosine at pH 7.4 (data not shown).

Effect of O 2 . and ONOO Ϫ Scavengers on Tyrosine Nitration-
The selenium-containing ONOO Ϫ scavenger ebselen (100 M) (43) inhibited nitration caused by both ONOO Ϫ and the NO/O 2 . system (Fig. 5). To better characterize the limited tyrosine nitration observed with hypoxanthine used as a substrate for xanthine oxidase (Fig. 2), we examined the inhibitory effects of uric acid on tyrosine nitration. Tyrosine nitration caused by 3.0 M ONOO Ϫ was completely inhibited when added to the reaction mixture of 10 milliunits/ml xanthine oxidase plus 50 M hypoxanthine that was preincubated for 5 min (Fig. 6). Addition of uricase restored nitration by ONOO Ϫ in a time-dependent fashion, showing that uric acid was a major inhibitory product of tyrosine nitration formed when hypoxanthine was the substrate.
Nitration-inhibitory Effect of Uric Acid with ONOO Ϫ and with the NO 2 Ϫ /H 2 O 2 /Myeloperoxidase System-We examined the effect of uric acid and other scavengers on tyrosine nitration catalyzed by myeloperoxidase from NO 2 Ϫ and H 2 O 2 (Fig. 7), a significant nitrating system proposed by Eiserich et al. (39). Allopurinol, ebselen, SOD, and NaHCO 3 did not affect tyrosine nitration catalyzed by myeloperoxidase, whereas uric acid strongly inhibited tyrosine nitration by myeloperoxidase.  We then compared the efficacy of uric acid for suppression of tyrosine nitration produced by both ONOO Ϫ and the NO 2 Ϫ / H 2 O 2 /myeloperoxidase system (Fig. 8). The IC 50  (generated via xanthine oxidase) produced significant amounts of nitrotyrosine.  addition, both NO and ONOO Ϫ have inhibitory effects on xanthine oxidase activity (29,30).
To minimize such interference, we used a low concentration of P-NONOate ( Although we observed a linear increase in ONOO Ϫ formation with increasing xanthine oxidase concentration in the P-NONOate/pterin/xanthine oxidase system, we found a bellshaped pattern of nitrotyrosine formation that peaked at similar fluxes of NO and O 2 . , as shown in Fig. 2A (44). It was recently reported by Pfeiffer et al. that dityrosine formation competes with tyrosine nitration with low concentrations of ONOO Ϫ , such that dityrosine could be a major product of tyrosine modification by ONOO Ϫ (46). However, our present results showed that, although tyrosine nitration is attenuated at tyrosine concentrations above 0.5 mM, very effective nitration occurs even with a low flux of ONOO Ϫ (10 nM/s), compared with dityrosine formation, particularly at low tyrosine concentrations (0.1 mM or below) (Fig. 4). The normal range of free L-tyrosine in the human plasma described in the literature is 40 -130 M (47), suggesting that tyrosine nitration will be the dominant tyrosine modification caused by ONOO Ϫ at physiological concentrations of free tyrosine or when tyrosine bioavailability is somehow limited.
It is also potentially important that uric acid inhibits tyrosine nitration not only by ONOO Ϫ (12, 13, 28) but also by the NO 2 Ϫ /H 2 O 2 /myeloperoxidase system (39,40). Uric acid may thus play an important role in modulating tyrosine nitration and the detrimental actions of ONOO Ϫ as well in vivo. However, because uric acid is readily oxidized by ONOO Ϫ , resulting in formation of oxidized metabolites such as allantoin, parabanic acid, urazole, and oxonic acid (48), excessive and prolonged production of ONOO Ϫ may decrease the level of uric acid in the microenvironment of the site of ONOO Ϫ production, resulting in enhanced susceptibility of free and protein-bound tyrosine to nitration by ONOO Ϫ as well as by the NO 2 Ϫ /H 2 O 2 / myeloperoxidase system. Uric acid is an end product of purine and nitrogen catabolism in avian and human species; in other animals uric acid is further metabolized to allantoin by uricase. The normal concentration of uric acid in plasma has been reported to be 20 -40 M for mice (13) and 200 -300 M for humans (13,48). In the presence of such high concentrations of uric acid, tyrosine nitration will be largely inhibited. However, we also found that a physiologically relevant concentration of bicarbonate (25 mM) significantly attenuated the inhibitory effect of uric acid on tyrosine nitration caused by ONOO Ϫ . Even in the presence of 50 M uric acid, ONOO Ϫ (3 M) produces a significant amount of nitrotyrosine (41 Ϯ 2.0 nM). This result suggests that in vivo tyrosine nitration by ONOO Ϫ is largely due to formation of the more potent nitrating agent ONOOCO 2 Ϫ . It has been long known that uric acid is an important endogenous component of the defense system against free radical tissue injuries because of its antioxidant activity (49). Recently, Hooper et al. (12) and Bagasra et al. (13) reported that uric acid treatment had a remarkable effect on experimental allergic encephalomyelitis (an animal model of MS), in which ONOO Ϫ is suggested to be involved in pathogenesis. They also found a significantly lower incidence of MS in hyperuricemic patients (50). Skinner et al. documented that a reaction product of ONOO Ϫ and uric acid could act as a nitrovasodilator (48). Therefore, interaction of uric acid with ONOO Ϫ would be an important factor in NO-linked pathogenesis of various diseases.
It has been reported that nitration of protein-bound and free tyrosine may modulate cellular functions through inactivation of enzymes such as mitochondrial SOD (18), disruption of assembly of neurofilament L (11,24), effects on tyrosine phosphorylation-mediated signal transduction (20), and dysfunction of microtubule formation (21). A recent intriguing finding of Kamisaki et al. (51) shows the presence of nitrotyrosine denitrase activity, which repairs protein nitration, in rat tissues, suggesting that a tyrosine nitration-denitration pathway participates in NO-or ONOO Ϫ -dependent signal transduction, similar to phosphorylation-dephosphorylation systems. These findings indicate that nitrotyrosine formation has great importance not only as a biomarker of reactive nitrogen-mediated tissue injury but also as a means to gain insight into molecular mechanisms of NO-related physiological and pathophysiological phenomena.
In conclusion, our data unambiguously verify that ONOO Ϫ generated from NO and O 2 . can nitrate tyrosine with similar reactivity and efficacy as chemically synthesized ONOO Ϫ . Thus, ONOO Ϫ formed by the reaction of NO with O 2 . may be one of the key intermediates responsible for tyrosine nitration in vivo. Moreover, although uric acid may serve a critical function by modulating tissue injury caused by reactive nitrogen species, it is also possible that ONOO Ϫ and NO 2 Ϫ /H 2 O 2 /myeloperoxidase enhance their tissue-damaging potential by destroying uric acid in vivo.