Biphasic Regulation of Leukocyte Superoxide Generation by Nitric Oxide and Peroxynitrite*

Activation of the NADPH oxidase-derived oxidant burst of polymorphonuclear leukocytes (PMNs) is of critical importance in inflammatory disease. PMN-derived superoxide (O⨪2) can be scavenged by nitric oxide (NO⋅) with the formation of peroxynitrite (ONOO−); however, questions remain regarding the effects and mechanisms by which NO⋅ and ONOO− modulate the PMN oxidative burst. Therefore, we directly measured the dose-dependent effects of NO⋅and ONOO− on O⨪2 generation from human PMNs stimulated with phorbol 12-myristate 13-acetate using EPR spin trapping. Pretreatment with low physiological (μm) concentrations of NO⋅ from NO⋅ gas had no effect on PMN O⨪2 generation, whereas high levels (≥50 μm) exerted inhibition. With ONOO− pretreatment, however, a biphasic modulation of O⨪2 generation was seen with stimulation by μm levels, but inhibition at higher levels. With the NO⋅ donor NOR-1, which provides more sustained release of NO⋅ persisting at the time of O⨪2 generation, a similar biphasic modulation of O⨪2 generation was seen, and this was inhibited by ONOO− scavengers. The enhancement of O⨪2 generation by low concentrations of ONOO− or NOR-1 was associated with activation of the ERK MAPKs and was blocked by their inhibition. Thus, low physiological levels of NO⋅present following PMN activation are converted to ONOO−, which enhances O⨪2 generation through activation of the ERK MAPK pathway, whereas higher levels of NO⋅ or ONOO−feed back and inhibit O⨪2 generation. This biphasic concentration-dependent regulation of the PMN oxidant burst by NO⋅-derived ONOO− may be of critical importance in regulating the process of inflammation.

The influx and activation of polymorphonuclear leukocytes (PMNs) 1 are of critical importance in the process of tissue inflammation (1). PMNs are chemotaxed and then activated at sites of tissue injury with the generation of oxygen free radicals and the release of proteolytic enzymes and other toxic products (2). Activated human PMNs generate superoxide (O 2 . ) during their respiratory burst via the NADPH oxidase complex (3). It has been reported that high levels of nitric oxide (NO ⅐ ) can directly inhibit O 2 . generation from stimulated PMNs (4,5).
NO ⅐ can also react with O 2 . derived from PMNs or other sources to form the potent oxidant peroxynitrite (ONOO Ϫ ). However, questions remain regarding the effects of low physiological levels of NO ⅐ and how these or ONOO Ϫ modulate the NADPH oxidase function of and O 2 . generation from PMNs.
NO ⅐ can inhibit PMN-endothelial cell adhesion in the microcirculation and thus potentially exert anti-inflammatory effects (6). In contrast, it has been observed that increased NO ⅐ generation leads to cytotoxicity following sepsis, ischemia, and reperfusion (7) or anoxia and reoxygenation (8). It has been demonstrated that ONOO Ϫ is formed by the reaction of O 2 . and NO ⅐ in post-ischemic tissues and following the process of inflammation (9). It has also been shown that O 2 . plays a key role in NO ⅐ -induced toxicity, and previously proposed mechanisms for both O 2 . -and NO ⅐ -mediated tissue injury now include a role for their combined reaction product, ONOO Ϫ (10). The reaction between O 2 . and NO ⅐ is extremely fast, almost diffusion-limited in rate (2 ϫ 10 9 M Ϫ1 s Ϫ1 ) (11). ONOO Ϫ is a potent oxidant that can attack a wide variety of biological molecules and is produced in diverse inflammatory and pathological processes, including post-ischemic injury (7,9), septic shock (12), and atherosclerosis (13). ONOO Ϫ can directly oxidize sulfhydryl groups (14) and also reacts by both 1-and 2-electron oxidation reactions with a variety of biological target molecules (15). Notably, ONOO Ϫ -mediated nitration of tyrosine residues on a variety of proteins is associated with loss of enzyme or receptor function (16).
Recently, it has been reported that mitogen-activated protein kinases (MAPKs) are important in the regulation of leukocyte function and the PMN oxidative burst (17)(18)(19). These kinases are also modulated by various oxidants; and based on these interactions, it has been suggested that O 2 . or secondary oxidants are important mediators of cellular signaling. Three groups of MAPK superfamilies have been identified in mammalian cells: the extracellular signal-regulated kinases 1 and 2 (ERK-1 and ERK-2), the c-Jun NH 2 -terminal kinase (also termed stress-activated protein kinase), and p38 MAPK (20). Two isoforms of ERK, ERK-1 (p44 MAPK ) and ERK-2 (p42 MAPK ), are expressed in PMNs (21), and they transduce signals elicited via several receptors that either have intrinsic tyrosine kinase activity or are associated with non-receptor tyrosine kinases (22). MAPK kinases (MEKs) are dual-specificity kinases that phosphorylate MAPKs such as ERK (23). MEKs have been detected in PMNs and have been shown to be critical for ERK activation (24,25). It has been reported that ERK may participate in activation of the PMN oxidative burst since one of the cytosolic components of the NADPH oxidase (p47 phox ) contains two serine residues within a sequence that is recognized by proline-directed kinases (26) such as ERK and since phosphorylation of p47 phox is involved in NADPH oxidase activation (27). Furthermore, recent studies have reported that MEK and ERK contribute to activation of the oxidative burst of human PMNs (18,19). However, it is unknown if there is a direct interaction between NO ⅐ or ONOO Ϫ and these pathways of activation. Therefore, in this study, we determined the concentrationdependent effects of NO ⅐ and ONOO Ϫ on activation of and O 2 .
NO ⅐ Gas Solution-NO ⅐ was scrubbed of higher nitrogen oxides by passage through a trap with solid NaOH pellets and a second trap with 1 M de-aerated (bubbled with argon for 30 min) NaOH solution. NO ⅐ gas-equilibrated solution was prepared using 500 ml of phosphatebuffered saline (pH 7.4) that was de-aerated by bubbling with argon for 30 min and then bubbled with scrubbed NO ⅐ for 30 min (28,29). To further verify the precise NO ⅐ concentration derived from NO ⅐ gasequilibrated solutions, electrochemical measurement of NO ⅐ concentrations were carried out at 25°C using a Model 832 detector with a Faraday cage and NO ⅐ electrode (CH Instruments, Inc., Cordova, IN).
ONOO Ϫ Preparation-The ONOO Ϫ was synthesized from acidified nitrite and hydrogen peroxide according to Beckman et al. (15). Alternatively, similarly prepared ONOO Ϫ was obtained from Alexis Corp. (San Diego, CA). The concentration of ONOO Ϫ was checked at the time of synthesis and again just prior to each experiment by absorbance measurements at 302 nm (extinction coefficient of 1670 M Ϫ1 cm Ϫ1 ), and only ONOO Ϫ that was Ͼ95% of the original concentration was used. To assure that pH did not change significantly upon addition of alkaline ONOO Ϫ to the reaction mixture, we also monitored the pH and limited the amount of alkaline ONOO Ϫ stock used.
Leukocyte Purification and Preparation-Human PMNs were isolated from freshly sampled venous blood of volunteers by Percoll density gradient centrifugation (30), which yielded PMNs with a purity of Ͼ95%. Each 10 ml of whole blood was mixed with 0.8 ml of 0.1 M EDTA and 25 ml of saline. The diluted blood was layered over 9 ml of Percoll at a specific density of 1.080 g/ml. After centrifugation at 400 ϫ g for 20 min at 20°C, the plasma, mononuclear cell, and Percoll layers were removed. Erythrocytes were lysed by addition of 18 ml of ice-cold water for 30 s, followed by 2 ml of 10ϫ PIPES buffer (25 mM PIPES, 110 mM NaCl, and 5 mM KCl, titrated to pH 7.4 with NaOH). Cells were pelleted at 4°C, the supernatant was decanted, and the procedure was repeated.
After the second hypotonic lysis, cells were washed twice with PAG buffer (PIPES buffer containing 0.003% human serum albumin and 0.1% glucose). Afterward, PMNs were counted by light microscopy on a hemocytometer. The final pellet was then suspended in PAG-CM buffer (PAG buffer with 1 mM CaCl 2 and 1 mM MgCl 2 ).
EPR Measurements-All EPR measurements were performed using a Bruker ER 300 EPR spectrometer operating at X-band with a TM 110 cavity.  (15); and after 1 min, no detectable ONOO Ϫ remains. To achieve sustained release of NO ⅐ continuing during the PMN oxidative burst so as to trigger intrinsic ONOO Ϫ generation, the NO ⅐ donor NOR-1 (0 -100 M) was used, and PMNs were immediately stimulated with PMA after addition of NOR-1.
Western Blot Analysis of Phosphorylation of ERKs in PMNs-PMNs in PAG-CM buffer were subjected to the same experimental conditions described above; reactions were stopped by addition of ice-cold PAG buffer; and the PMNs were rapidly pelleted. Cell pellets were immediately lysed in lysis buffer (50 mM Tris HCl (pH 7.5), 5 mM EDTA, 10 mM EGTA, 5 mM dithiothreitol, 1% Nonidet P-40, 1 mM phenylmethylsulfonyl fluoride, 20 mg/ml leupeptin, 100 g/ml aprotinin, 10 mM benzamidine, 50 mM NaF, 5 mM Na 4 P 2 O 7 , and 1 mM Na 3 VO 4 ). Extracts from an equal number of PMNs (10 6 cells/lane) were diluted with an equal volume of 2ϫ loading buffer (0.125 M Tris-HCl (pH 6.8), 4% SDS, 0.005% bromphenol blue, 20% glycerol, and 5% 2-mercaptoethanol) and subjected to 4 -20% Tris/glycine gradient gel electrophoresis (Novex, San Diego, CA). After electrophoresis, gels were then transferred to pure nitrocellulose membranes (Schleicher & Schü ll) with TransBlot (Novex). After transfer, membranes were immersed in TBST (50 mM Tris (pH 7.5), 0.15 M NaCl, and 0.05% Tween 20) containing 5% nonfat dried skim milk overnight to block nonspecific binding. Membranes were then washed three times (5 min each) with TBST. Immunoreactive proteins were detected using anti-phospho-ERK-1 and -ERK-2 antibodies (0.5 g/ml) in TBST containing 1% skim milk. After 4 h of incubation, membranes were washed with TBST and then incubated with peroxidase-conjugated anti-mouse Ig antibody for 1 h. After five 10-min washes, membrane-bound anti-mouse Ig antibody was visualized with ECL Western blotting detection reagents (Pierce), and Hyper ECL luminescence detection film (Amersham Pharmacia Biotech). The enhanced chemiluminescence film images were converted to digital format with a URL digital camera, and the images were analyzed with National Institutes of Health Image (32). Although the comparisons were made on the basis of equal numbers of cells, the protein contents of these samples were also found to be equal.
Statistical Analysis-All the experiments were performed in triplicate and repeated a minimum of three times. Results are expressed as means Ϯ S.E. Statistical analysis was performed using Student's t test or one-way analysis of variance. A p value Ͻ0.05 was considered to be statistically significant. were performed to determine the dose-dependent effects of PMA and to identify a concentration that would provide significant but less than half-maximal activation. As reported previously (31), after addition of PMA (1-100 ng/ml) to PMNs, we observed primarily a characteristic DEPMPO-OOH adduct spectrum with hyperfine splitting giving rise to 12 resolved peaks (Fig. 1A, spectra b-d). The O 2 . adduct DEPMPO-OOH comprised 92% of the total intensity, and the DEPMPO-OH adduct 8%. The DEPMPO-OOH signals increased as a function of the PMA concentration (Fig. 1). PMA is known to be a strong activator of protein kinase C. As shown in Fig. 1B We examined the effects of NO ⅐ pre-exposure on subsequent PMN function. PMNs were exposed to given initial concentrations of NO ⅐ prepared using NO ⅐ gas, and the concentration of NO ⅐ in solution was monitored by NO ⅐ electrode. After 10 min, when the NO ⅐ concentration decayed to near zero, PMA was added to trigger O 2 . generation from the PMNs. After addition of PMA (1 ng/ml), a large signal of DEPMPO-OOH and a small signal of DEPMPO-OH was observed ( Fig. 2A, spectrum a). We observed that low concentrations of NO ⅐ (1 M) had no effects on the magnitude or kinetics of DEPMPO-OOH formation (Fig. 2, A, spectrum b; and B), whereas with high concentrations of NO ⅐ (100 M), the DEPMPO-OOH signal was significantly decreased ( Fig. 2A, spectrum c; and B and C). The effects of pre-exposure of PMNs to ONOO Ϫ on subsequent O 2 . generation were also studied. It was observed that low ONOO Ϫ concentrations (1 M) markedly increased O 2 . generation following PMA stimulation, with higher levels of DEPMPO-OOH spin adduct seen (Fig. 3, A, spectra a and b; and B). In contrast, pretreatment with high levels of ONOO Ϫ (100 M) markedly inhibited subsequent PMA-stimulated O 2 . generation (Fig. 3, A, (Fig. 4, A, spectrum b; and B), in a manner similar to that seen with extrinsic ONOO Ϫ . With addition of NOR-1 (100 M), the DEPMPO-OOH signal was decreased by Ͼ5-fold (Fig. 4, A, spectrum c;

Regulation of Leukocyte O 2 . Generation by ONOO Ϫ
and PMA-activated PMNs, NO ⅐ electrode measurements were performed. In non-activated PMNs, NO ⅐ solution concentrations initially increased over the first 5 min after NOR-1 addition and then achieved a plateau, which was followed by a gradual decrease over the next 10 min (Fig. 5). A similar time course was seen with 1, 10, or 100 M NOR-1 (Fig. 5, A-C). In activated PMNs, the observed NO ⅐ concentrations were markedly decreased, with complete scavenging seen with 1 M NOR-1, a prominent decrease with 10 M NOR-1, and only a partial decrease with 100 M NOR-  (Fig. 6B).
Effects of ONOO Ϫ on Phosphorylation of ERKs in PMAstimulated PMNs-Recently, it has been shown that MEK and ERK MAPKs can contribute to activation of the oxidative burst in human PMNs stimulated with a variety of activators (18,19). Furthermore, it has also been reported that NO ⅐ -derived species and ONOO Ϫ can activate the phosphorylation of ERKs (33,34). To understand the mechanism by which extrinsic or intrinsic PMN-generated ONOO Ϫ enhances PMN O 2 . generation, we examined the relationship between ERK activation and O 2 . generation. In preliminary experiments, we verified that PMA (1-100 ng/ml) induced the phosphorylation of ERK-1 and ERK-2 in a PMA concentration-dependent manner. The phosphorylation of ERK-1 and ERK-2 was maximized above 10 ng/ml PMA in a manner similar to that seen for O 2 . generation in the EPR measurements (Fig. 1), suggesting that O 2 . generation from PMA-stimulated PMNs could be related to activation of ERK-1/2. As shown in Fig. 7, the phosphorylation of ERK-1/2 did occur in PMA (1 ng/ml)-stimulated PMNs. In these experiments, phosphorylated and non-phosphorylated ERK-2 proteins were used as references. Treatment either with ONOO Ϫ (1 M) or NOR-1 (1 M) as performed in the EPR experiments, in which enhancement of O 2 . generation was seen, also enhanced the phosphorylation of ERK-1/2 (Fig. 7). These data suggest that activation of ERK-    (23,35). The inhibitor PD98059 blocks the activation and phosphorylation of MEK via an allosteric mechanism that is not associated with inhibition of ATP binding (36). The potency and specificity of PD98059 have been reported in various biological systems (18,19,32,36). As reported previously, PD98059 (100 M) totally blocks ERK phosphorylation in PMNs (18). PD98059 (100 M) itself inhibited O 2 . generation from PMA-stimulated PMNs (Fig. 8A), indicating that the MEK/ERK pathway could be of importance in the respiratory burst of PMA-stimulated PMNs. The enhancement of O 2 . generation induced by both ONOO Ϫ (1 M) and NOR-1 (1 M) was completely inhibited by PD98059 (10 or 100 M) (Fig. 8, B and C). These results suggest that activation of the MEK/ERK pathway is of critical importance in the mechanism by which low concentrations of ONOO Ϫ cause enhanced PMN O 2 . generation.

DISCUSSION
The generation of oxygen radicals from activated PMNs is of critical importance in the pathophysiology of a variety of inflammatory diseases and in the tissue injury that occurs following ischemia and reperfusion (1-3, 37, 38). More recently, it has been demonstrated in the setting of inflammation that increased NO ⅐ generation occurs (7,8), and we have also demonstrated that there is markedly increased NO ⅐ generation and accumulation in ischemic and post-ischemic tissues (39 generation from stimulated PMNs that was exerted by extrinsic ONOO Ϫ (Fig. 3) or ONOO Ϫ formed by the reaction of PMNderived O 2 . and NO ⅐ (Fig. 4). Whereas low M levels of ONOO Ϫ stimulated PMN O 2 . generation, high levels above 50 M caused inhibition. With NO ⅐ , however, no effect was seen with low  ONOO Ϫ , resulting in a low level flux of more sustained ONOO Ϫ production (Fig. 5). Furthermore, these effects of NOR-1 were completely blocked by the ONOO Ϫ scavenger urate or by the ONOO Ϫ decomposition catalyst FeTMPS (Fig. 6A) ation, whereas with continued and increased ONOO Ϫ generation, this oxidative inflammatory response would be inhibited. This feedback process could be important in regulating the oxidative inflammatory response, with initial sensitization as required in anti-pathogen-directed immune response followed by regulatory inhibition to prevent excessive oxidative cellular injury.
A number of researchers have reported that activation of PMNs leads to a prominent activation of the MEK/ERK pathway (18,21,46,47). Furthermore, it has been shown that NO ⅐ -derived species and ONOO Ϫ can elicit phosphorylation of ERKs (33,34). We have demonstrated that low concentrations (1 M) of ONOO Ϫ or NOR-1 elicited activation of ERKs (Fig. 7). We observed that inhibition with the MEK/ERK-specific, cell-permeant, pharmacological inhibitor PD98059 completely inhibited the ONOO Ϫ -or NOR-1-induced increase in O 2 . generation from stimulated PMNs (Fig. 8). These findings demonstrate that the MEK/ERK pathway is activated by low concentrations of ONOO Ϫ or NO ⅐ , which reacts to form ONOO Ϫ , and indicate that activation of this MAPK pathway is required for this ONOO Ϫ -mediated modulation of the PMN oxidative burst. In addition, we directly observed that ERK inhibition with PD98059 partially inhibited O 2 . generation from activated PMNs (Fig. 8). These findings are in agreement with a recent report (18) in which it was observed that this inhibitor also partially blocked the PMN activation and respiratory burst induced by activators including N-formyl-methionyl-leucyl-phenylalanine and zymosan. However, both results show that the effects of PD98059 are only partial inhibition of the respiratory burst, suggesting the possibility of the involvement of other signaling pathways. It has been reported that the respiratory burst can be triggered by a variety of second messengers, including calcium, agonists of protein kinase C, arachidonate, and other lipid metabolites and agents that favor phosphotyrosine accumulation (48 -51). Recent reports have shown that another MAPK family member, p38 MAPK , can also provide input to regulate activation of the oxidative burst (52,53). Furthermore, it has also been shown that both ERK MAPKs and p38 MAPK can phosphorylate p47 phox (54 -56), which is a critical step in NADPH oxidase activation (26,27). It has been proposed that   4). B, PMNs were simultaneously treated with NOR-1 (1 M) and activated with PMA (1 ng/ml) to trigger intrinsic PMN-derived ONOO Ϫ formation. After 10 min, reactions were stopped by addition of ice-cold PAG buffer, and the cells were centrifuged. The cell pellets were lysed, and the lysates were subjected to Western blot analysis as described under "Experimental Procedures." phosphorylation of this cytosolic component by MAPKs is important for activation of NADPH oxidase since there are specific serine residues that are not phosphorylated by protein kinase C and specifically recognized by MAPKs (56). In the future, additional studies will be required to evaluate the relationship between the p38 MAPK and ERK MAPK pathways and the regulation of O 2 . generation from stimulated PMNs. In