Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

An increased production of nitric oxide (NO) for prolonged periods of time in many pathological states is known to contribute to oxidative damage of critical cellular macromolecules, including proteins, which often show elevated levels of 3-nitrotyrosine (Abe et al., 1997; Kooy et al., 1997; Moriel and Abdalla, 1997; Petruzzelli et al., 1997). This process is thought to involve the reaction of NO with superoxide radical (O₂) that generates peroxynitrite (ONOO), a key oxidant and nitrating molecule. The amount of ONOO depends on the competition for O₂ between superoxide dismutase (SOD) and NO. Because the NO/O₂ reaction occurs at a rate of 2 × 10¹⁰ M¹ s¹ (Koppenol, 1998), it is thought that NO effectively competes with SOD for the scavenging of O₂.

Platelets play important roles in hemostasis, thrombosis, and inflammation. The initial work by Mondoro et al. (1997) and our studies have indicated that platelets are very sensitive to low amounts of ONOO. At a concentration range of 10 to 50 µM, ONOO increases the levels of platelet protein-bound 3-nitrotyrosine by 4- to 100-fold (Jiang and Balazy, 1998). Several reports have demonstrated that ONOO inhibits (Moro et al., 1994; Yin et al., 1995; Mondoro et al., 1997) or stimulates platelet function (Moro et al., 1994); however, the processes through which ONOO-dependent protein tyrosine nitration modulates platelet structure and function remain to be elucidated. In this study, we focused on the effect of ONOO on platelet eicosanoid formation because the generation of tyrosyl radicals within prostaglandin synthase protein [cyclooxygenase (COX)] is a critical initial step in the formation of thromboxane (TX) A₂, a potent lipid mediator of platelet activation and vasoconstriction. The interaction of COX with compounds that modify the critical tyrosines is known to impair prostaglandin biosynthesis (Shimokawa et al., 1990). It has been demonstrated that NO can stimulate prostaglandin formation in vivo (Salvemini et al., 1993); however, because conflicting reports have been published regarding the ability of NO to activate purified COX (Salvemini et al., 1993; Tsai et al., 1994), it has been hypothesized that ONOO could be an important regulator of prostaglandin biosynthesis (Landino et al., 1996). The effects of ONOO on COX protein nitration and prostaglandin formation in platelets have not been investigated. Consequently, the present study focuses on examining the effects of ONOO on eicosanoid formation by human platelets and purified COX enzymes.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Chemicals and Reagents. Arachidonic acid, 4,4'-diisothiocyanatostilbene-2,2'-disulfonic acid (DIDS), and ()-epigallocatechin gallate were purchased from Sigma Chemical Co. (St. Louis, MO). COX-1 (from ram seminal vesicles), COX-2 (from sheep placenta), and polyclonal anti-COX-1 and anti-COX-2 antibodies were purchased from Cayman Chemical Co. (Ann Arbor, MI). Polyclonal anti-nitrotyrosine antibody and nitrotyrosine-containing protein standards were purchased from Upstate Biologicals (Lake Placid, NY). 2',7'-Dichlorodihydrofluorescein diacetate (DCF-DA) was purchased from Molecular Probes (Eugene, OR). [¹³C₆]Tyrosine (isotopic purity, >99%) was obtained from Isotec Corp. (Miamisburg, OH). [¹³C₆]-3-Nitrotyrosine was prepared as described previously (Jiang and Balazy, 1998). Pentafluorobenzyl bromide and N,O-bis(trimethylsilyl)trifluoroacetamide were obtained from Aldrich Chemical Co. (Milwaukee, WI). [1-¹⁴C]Arachidonic acid was obtained from New England Nuclear (Boston, MA). All other reagents were of the highest research grade quality. ONOO was prepared from sodium nitrite and H₂O₂ in a quenched-flow reactor as described previously (Balazy, 1994). H₂O₂ and oxygen were removed from the ONOO solution by the addition of granular manganese oxide followed by bubbling with a gentle stream of pure nitrogen for 30 min. The ONOO solution was maintained at 20°C, and the concentrated top yellow layer was carefully thawed and transferred to a clean glass tube. The concentration of ONOO was determined spectrophotometrically by the measurement of absorbance at 302 nm in 1 N NaOH ( = 1670 M¹ cm¹) with a Hewlett-Packard 8452A diode array spectrophotometer. The stock solution of ONOO typically contained 190 to 322 mM ONOO. Dilutions in 0.1 N NaOH were made immediately before use to achieve the desired concentrations. A micro pH meter was used to eliminate the possibility of pH increase after the addition of an aliquot of ONOO solution. The concentration of H₂O₂ in ONOO stock preparation was ~1% and did not interfere with platelets. Control experiments were carried out with ONOO decomposed in phosphate buffer at 37°C for 15 min. This allowed us to compare the effect of other compounds in the ONOO solution.

Platelet Isolation and Aggregometry. Blood by phlebotomy was obtained from healthy volunteers who had not ingested any drugs known to interfere with platelet function for at least 15 days. Fresh platelet concentrates from healthy donors were obtained from Hudson Valley Blood Service (Elmsford, NY). Washed platelets were prepared as described previously (Balazy, 1991). Briefly, 10 ml of platelet concentrate was subjected to 1100g centrifugation for 15 min at 21°C, and the pellet was resuspended in 5 ml of buffer A [140 mM NaCl, 2 mM KCl, 14 mM NaHCO₃, 5.5 mM glucose, 1 mM MgCl₂, 5 mM HEPES buffer, 0.2% BSA, 0.2 µM prostaglandin (PG)E₁, and 1.5 µg/ml apyrase], and the pH was adjusted to 6.4. Platelets were centrifuged again at 1100g for 15 min at 21°C, and the pellet was resuspended in buffer B (same as buffer A but without BSA, PGE₁, and apyrase, pH 7.4) to adjust the platelet concentration to 1 × 10⁶ cells/µl. Platelets were diluted 2-fold before incubations with ONOO and radiolabeled arachidonic acid. Aggregometry (Born's method) was performed at 37°C in siliconized glass cuvettes with constant stirring at 900 rpm in a Payton aggregometer as described previously (Balazy, 1991). Platelets (1.6 × 10⁸ cells/ml) were diluted to the final concentration of 2.2 × 10⁷ cells/ml and stimulated with arachidonic acid (final concentration, 68 µM). Platelets were counted using a Coulter P540 (Hialeah, FL) electronic cell counter.

HPLC Analyses. Samples were analyzed using a 1050 HPLC system (Hewlett-Packard, Palo Alto, CA) equipped with a quaternary pump, a variable-wavelength UV detector, and a radiodetector. Arachidonic acid metabolites were analyzed on a C18 Ultrasphere column (250 × 4.6 mm; Beckman Instruments, Fullerton, CA), which was eluted with a gradient of acetonitrile in water (pH 4, adjusted with acetic acid at a flow of 1 ml/min). The initial concentration of acetonitrile was 25%, and it gradually increased to 100% (v/v) within 20 min. The effluent from the column was collected in 1-ml fractions using a Gilson FC203B fraction collector or analyzed with a radioactivity monitor (Packard, Meriden, CT) using an Ecolite scintillation fluid (ICN, Costa Mesa, CA).

Detection of 3-Nitrotyrosine in Platelets and COX Enzymes. 3-Nitrotyrosine content was measured quantitatively with a gas chromatography/mass spectrometry (GC/MS) assay as described previously (Jiang and Balazy, 1998). Briefly, proteins were hydrolyzed with 6 N HCl for 12 to 18 h at 120°C. Before hydrolysis, an internal standard, [¹³C₆]-3-nitrotyrosine (5 ng), was added. Hydrolysates were applied to SepPak cartridges (Waters) and eluted with 25% methanol/water. The eluates were purified on a C18 HPLC column (250 × 4.6 mm), and fractions containing 3-nitrotyrosine were derivatized with pentafluorobenzyl (PFB) bromide at 70°C for 40 min. The derivatives were purified by HPLC, and fractions containing 3-nitrotyrosine-PFB were finally analyzed by GC/MS as described previously (Jiang and Balazy, 1998). The ratio of ions m/z 585 (corresponding to endogenous 3-nitrotyrosine) and m/z 591 (internal standard) were measured, and the amount of 3-nitrotyrosine was calculated from the standard curve.

Immunoblotting of 3-Nitrotyrosine. ONOO (final concentration, 1-100 µM) was sampled with a gas-tight syringe and added as a bolus injection into a COX solution in 50 mM phosphate buffer or into suspension of platelets (1.6 × 10⁸/ml) with stirring at room temperature for 3 min. Platelet lysates were mixed with Laemmli's reagent under reducing conditions. SDS-polyacrylamide gel electrophoresis was performed using a 6% separating gel. The proteins were electrophoretically transferred to nitrocellulose membranes (0.2-µm pore size, Hybond-C extra; Amersham, Arlington Heights, IL) and detected using a chemiluminescence assay. After overnight saturation at 4°C in Tris-saline buffer containing 0.1% Tween 20, membranes were incubated for 12 h at room temperature with each of the two monoclonal antibodies at a concentration of 50 µg/ml (0.1 ml/cm²). After several washings, blots were incubated with the secondary antibody conjugated with horseradish peroxidase at 1:2000 dilution. After washing, immunoreactive signals were revealed using chemiluminescence reagent (ECL Kit; Amersham) and visualized by exposure to Hyperfilm ECL (Amersham) for 7 s. Signals were quantified by optical densitometry.

GC/MS Analyses. PFB esters of eicosanoids were prepared with pentafluorobenzyl bromide and diisopropylethylamine. Additionally, the hydroxyl groups were protected by trimethylsilyl groups through derivatization with N,O-bis(trimethylsilyl)trifluoroacetamide. Platelet eicosanoids were analyzed using a Hewlett-Packard 5980A gas chromatograph/mass spectrometer via the conditions described previously (Balazy, 1991). Samples were analyzed via electron capture chemical ionization using a DB-1 GC column and a temperature program of 150-300°C at a rate of 25°C/min. Samples containing PFB esters of arachidonic acid isomers were additionally analyzed with a tandem mass spectrometer (TSQ 5000; Finnigan) using GC inlet and the negative ion chemical ionization. Anion at m/z 303 was collisionally activated using argon gas at 1.5 torr.

Measurement of ONOO Permeability into Platelets. Platelet suspension (1 ml) was incubated with DCF-DA (50 µM) in buffer B for 30 min. Platelets were rinsed several times and resuspended in 1 ml of buffer B. In separate experiments, aliquots of DCF-DA-treated platelets were incubated with 300 µM DIDS for 30 min at room temperature. The pellets were rinsed, resuspended in 1 ml of buffer B, and used for fluorescence measurement. The platelets were diluted in the working solution (90 mM NaCl, 5 mM KCl, 50 mM sodium phosphate buffer at pH 7.4) so that the final concentration of platelets was 5.3 × 10⁶ cells/ml. Fluorescence measurements were conducted on a Shimadzu RF-5000 fluorospectrophotometer using excitation and emission wavelength at 488 and 515 nm, respectively, with a 10-nm band split. The cells were placed in plastic cuvettes with zero background absorbance (Fisher, Springfield, NJ) and were continuously stirred with a magnetic microbar, and the temperature of the cuvette was maintained at 37°C.

Statistical Analyses. Data were analyzed by the Student's t test or ANOVA with a Tukey's modified t test to determine statistical significance. Results are reported as mean ± S.E., with n equal to the number of blood donors or separate determinations. Statistical significance was assumed for a value of P < .05.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Inhibition of Arachidonate-Induced Platelet Aggregation by ONOO. The stimulation of platelets with ONOO (100-190 µM) produced a moderate aggregatory effect, which did not develop into a full response such as that observed with arachidonic acid (Fig. 1). At concentrations below 100 µM, ONOO failed to produce detectable platelet aggregations. However, the treatment of platelets with 4 to 50 µM ONOO for 1 min inhibited arachidonate-induced aggregations in a dose-dependent manner with an IC₅₀ value of 5.8 ± 1.2 µM (Table 1). Decomposed ONOO was without the effect on arachidonate-induced aggregations (Table 1). Preincubation of platelets with ()-epigallocatechin gallate (2 µM for 10 min at 37°C), a tea polyphenol and inhibitor of tyrosine nitration (Fiala et al., 1996), reduced the inhibitory effect of ONOO on arachidonate-induced aggregations (Table 1) by 78 to 82% (n = 5, P < .05). Two compounds that induce protein nitration, tetranitromethane (TNM; 10 µM) and NO₂BF₄ (10 µM), caused full inhibition of arachidonate-induced platelet aggregations (not shown).

View larger version (9K): [in this window] [in a new window]   Fig. 1.   ONOO inhibits platelet aggregations induced by arachidonic acid (AA). A, ONOO alone (190 µM) produced a proaggregatory response that abolished the effect of arachidonic acid (68 µM). Decomposed ONOO was without effect. B, effect of ONOO (7.4 µM) on platelet aggregation induced by arachidonic acid (68 µM).

View this table: [in this window] [in a new window]   TABLE 1 Effect of ONOO on aggregation of human platelets induced by arachidonic acid Values are mean ± S.E. (n = 5).

Modulation of Platelet Arachidonic Acid Metabolism by ONOO. Exposure of platelets to ONOO for 1 min, followed by incubation with 0.05 µCi of arachidonic acid (final concentration, 1 µM), inhibited the formation of radiolabeled TXB₂ and 12-hydroxyheptadecatrienoic acid (12-HHT) in a dose-dependent manner (Fig. 2). The IC₅₀ value for TXB₂ inhibition was ~150 µM because the number of platelets in this experiment was 23-fold higher than that in the aggregation experiment. More cells were needed to obtain radiochromatograms. 12-Lipoxygenase appeared to be less sensitive to ONOO than COX (Figs. 2 and 3). At a dose that fully inhibited COX, 12-hydroxyeicosatetraenoic acid (12-HETE) levels were decreased by ~50%. In experiments in which ONOO was added to platelets simultaneously with arachidonic acid, platelet eicosanoid formation was increased (28-50%, n = 5, P < .05) (Fig. 2). Thus, the effect of ONOO on arachidonate-stimulated platelet eicosanoid formation depended on the temporal availability of the substrate for COX.

View larger version (12K): [in this window] [in a new window]   Fig. 2.   The effect of ONOO on formation of platelet eicosanoids. Inhibition of COX metabolites, TXA2 (measured as TXB2), and 12-HHT was observed when ONOO was added before (), but not when it was added simultaneously with (), arachidonic acid (n = 5, *P < .05,  versus ). Note that the number of platelets per milliliter used in this experiment was 23-fold more than that in the aggregation experiments.

View larger version (22K): [in this window] [in a new window]   Fig. 3.   Radiochromatograms showing the effect of ONOO (3 × 520 µM) and TNM (1 µl/ml) on 14C-labeled arachidonic acid (AA; 1 µM) metabolism in human platelets (7 × 108 cells/ml). Control incubations show generation of COX metabolites TXB2 and 12-HHT and a lipoxygenase metabolite, 12-HETE. A, iso-PGF2. B, EpHETrE. C, trans-arachidonic acids.

Identification of Arachidonate Metabolites. We noticed that increasing doses of ONOO not only inhibited platelet eicosanoid generation but also caused the formation of additional radiolabeled metabolites (labeled A-C, Fig. 3). To establish their structures, platelets were pretreated with ONOO (3 × 520 µM added to 7 × 10⁹ cells in 10 ml of Krebs' buffer at pH 7.4) for 1 min and then incubated (37°C, 30 min) with ¹⁴C-labeled arachidonic acid (final concentration, 100 µM). Lipids were extracted with ethyl acetate and purified on a reversed phase HPLC column with monitoring of radioactivity (Fig. 3). The treatment of platelets with TNM (1 µl/10 ml) produced a similar pattern of arachidonate metabolism (Fig. 3). Metabolite A (eluting at 10-11 min, Fig. 3), analyzed by GC/MS as a PFB, trimethylsilyl derivative, produced a mass spectrum showing an abundant ion at m/z 569 (Fig. 4A). Additional fragment ions at m/z 479, 389, and 299 appeared to correspond to consecutive losses of (CH₃)₃SiOH from ion m/z 569 (Fig. 4A). Less abundant ions at m/z 407 and 317 most likely originated from consecutive (CH₃)₃SiOH losses from ion at m/z 497 [m/z 569  (CH₃)₂Si==CH₂]. This spectrum suggested that metabolite A was a prostanoid, having three hydroxyl groups and a molecular weight of 354, which is consistent with the structure of PGF₂ (Fig. 4A). A major component eluted between the two PGF₂ standards that were used (Fig. 4A), whereas the less abundant component had a retention time nearly identical with that of 8-iso-PGF₂. Additional experiments revealed that indomethacin did not inhibit the formation of these PGF₂ isomers, nor were they produced by nitrated COX. Thus, these prostanoids appeared to originate from a direct attack of free radicals on arachidonic acid. Metabolite B produced a mass spectrum (Fig. 4B) containing an abundant ion at m/z 407 and fragment ions at m/z 389 (loss of H₂O) and m/z 317 [loss of (CH₃)₃SiOH]. This fragmentation pattern was consistent with a structure of an epoxyhydroxyeicosatrienoic acid (EpHETrE). Metabolite C eluted as a peak of radioactivity at 25.5 to 26 min (Fig. 3) and thus was less polar than arachidonic acid. GC/MS/MS analyses revealed several chromatographic peaks with similar mass spectra containing a prominent ion at m/z 303 (Fig. 4C). These mass spectra were also similar to the mass spectrum of an arachidonate PFB ester. Metabolite C was similar to the product of the NO₂/arachidonic acid reaction (Jiang et al., 1999) and was likely to be a mixture of trans-isomers of arachidonic acid. Additional experiments revealed that a direct reaction of ONOO with arachidonic acid at pH 7.4 also produced compounds of similar properties as metabolite C. Fractions eluting at 20 to 22 min (Fig. 3) produced mass spectra containing an ion at m/z 319 and displayed similar properties as epoxyeicosatrienoic acids (Balazy, 1994).

View larger version (19K): [in this window] [in a new window]   Fig. 4.   Identification of metabolites A, B, and C produced by exposure of platelets to ONOO. A, mass spectrum of metabolite A and a chromatogram (inset) show ion m/z 569 and a comparison of retention time with two PGF2 standards. B, mass spectrum of metabolite B, epoxyhydroxyeicosatrienoic acid. C, GC/MS/MS analysis of metabolite C as a PFB ester shows multiple isomers of trans-arachidonic acids (left). The mass spectrum (right) of isomer 7 was obtained by collisional activation of ion m/z 303. The retention time of arachidonic acid standard is indicated with an arrow.

Nitration and Inhibition of COX-1 and COX-2 by ONOO. Figure 5 shows that the exposure of platelets, COX-1, and COX-2 to ONOO generated COX proteins that were immunoreactive to 3-nitrotyrosine antibody. Also, ONOO dose dependently induced formation of additional three or four bands that were not immunoreactive to COX antibody but contained 3-nitrotyrosine (Fig. 5). Because COX preparations were free of other proteins, as verified by a direct staining with the Coomassie blue, the data in Fig. 5 suggested that ONOO caused dose-dependent nitration of COX tyrosine residues as well as disruption of COX proteins into fragment proteins of lower molecular weight that also contained 3-nitrotyrosine. COX-2 was found to be more sensitive to nitration by ONOO than COX-1. Increased nitration of COX-2 could be seen with 1 to 5 µM ONOO, whereas COX-1 nitration was detectable with >10 µM ONOO. Further evidence that ONOO reacts with COX and generates 3-nitrotyrosine residues was obtained by a direct analysis of COX protein hydrolysates with GC/MS (Table 2). The levels of 3-nitrotyrosine in COX proteins increased 3- to 8-fold after exposure to 50 to 100 µM ONOO. Although immunoblotting of COX samples showed a linear increase of 3-nitrotyrosine levels with increasing ONOO dose (Fig. 5), the GC/MS analyses of COX-1 treated with 100 µM ONOO showed less of 3-nitrotyrosine than may be expected from an examination of the immunoblot at this concentration level (Table 2). It is possible that ONOO caused additional changes in COX-1 protein, such as formation of dinitrotyrosines, which are detectable by immunoblotting but not by GC/MS. Amino acid sequence data indicate that both COX-1 and COX-2 contain 27 tyrosine residues (GenBank protein sequence). A dose of 50 µM ONOO generated an average of two 3-nitrotyrosine residues per COX molecule. Exposure of COX enzymes to ONOO followed by incubation with ¹⁴C-labeled arachidonic acid inhibited PGH₂ formation (Table 3). Thus, the inhibitory effect of ONOO on COX appeared to correlate with 3-nitrotyrosine levels in COX protein. The metabolic profile of COX premixed with increasing doses of ONOO and then added to arachidonic acid showed a decrease in a PGH₂ peak without the formation of other substances. However, simultaneous incubations of COX with ONOO and arachidonic acid increased the PGH₂ levels (Table 4) and produced 6 to 8% of EpHETrE. Additional experiments revealed that 50 µM NO induced a comparable nitration of COX-1 as 10 µM ONOO, whereas COX-2 was more sensitive to ONOO- than NO-induced nitration.

View larger version (41K): [in this window] [in a new window]   Fig. 5.   Detection of nitrotyrosine by Western analysis in the human platelet COX, COX-1, and COX-2 after exposure to ONOO. A, COX-1 and platelet immunoblots using the polyclonal COX-1 antibody. Also shown is a COX-2 immunoblot using the polyclonal COX-2 antibody (1:1000). B, nitrotyrosine immunoblot of the same blot after stripping of the first antibody using the polyclonal nitrotyrosine antibody (1:1500). Molecular mass markers were used to estimate approximate sizes (kDa) as shown on the left (c, standard nitrated COX-1).

View this table: [in this window] [in a new window]   TABLE 2 Effect of ONOO on 3-nitrotyrosine content in COX-1 and COX-2

View this table: [in this window] [in a new window]   TABLE 3 Effect of 1-min preincubation of COX-1 or COX-2 with ONOO on PGH2 formation COX-1 (45.14 U, 0.92 µg of protein) and COX-2 (23.35 U, 3.36 µg of protein) per ml were incubated with ONOO at pH 7.8 for 5 min then mixed with 10 µM 14C-labeled arachidonic acid, and incubated for 30 min. PGH2 was rapidly extracted with cold ethyl acetate and analyzed by a reversed-phase HPLC. The amount of PGH2 was calculated from the specific activity of arachidonic acid (54.6 mCi/mmol) and the area of PGH2 peak of radioactivity. Values are in ng/µg of protein and are averages of three measurements ± S.E.M.

View this table: [in this window] [in a new window]   TABLE 4 Effect of simultaneous incubation of COX-1 or COX-2 with ONOO and arachidonic acid on PGH2 formation COX-1 (45.14 U, 0.92 µg of protein) and COX-2 (23.35 U, 3.36 µg of protein) per ml were incubated with ONOO at pH 7.8 for 5 min then mixed with 10 µM 14C-labeled arachidonic acid, and incubated for 30 min. PGH2 was rapidly extracted with cold ethyl acetate and analyzed by a reversed-phase HPLC. The amount of PGH2 was calculated from the specific activity of arachidonic acid (54.6 mCi/mmol) and the area of PGH2 peak of radioactivity. Values are in ng/µg of protein and are averages of three measurements ± S.E.M.

Diffusion of ONOO across Platelet Membrane. Permeability of ONOO through the platelet membrane was studied with a fluorescence assay (Possel et al., 1997). The membrane-permeable nonfluorescent DCF-DA remained in the cytoplasm of the resting platelets as a nonfluorescent free acid, 2',7'-dihydrodichlorofluorescein (DCFH), after deesterification by intracellular esterases. Detection of fluorescence from DCF-DA-labeled platelets (Fig. 6) indicated that ONOO crossed the platelet membrane to reach intracellular sites where DCFH was accumulated. The generation of fluorescence by platelets was specific and exceptionally sensitive to ONOO. Bolus injection of hydrogen peroxide (100 µM), sodium nitroprusside (100 µM), or potassium superoxide (100 µM) did not increase fluorescence in platelets, whereas cumulative increases of fluorescence can be observed with 50 nM ONOO (Fig. 6A). DIDS slowed the diffusion of ONOO into the platelet cytosol (Fig. 6B). The results indicate that DIDS, in agreement with its known effect on erythrocytes (Denicola et al., 1998), was blocking the transport of ONOO through the ion channel, possibly platelet HCO₃/Cl anion transporter. We also examined the effect of NO₂, a decomposition product of ONOO, on platelet fluorescence. A solution of NO₂ in helium, delivered at concentration of 47 µM/min, produced a rapid onset of fluorescence of the similar shape as ONOO (not shown).

View larger version (29K): [in this window] [in a new window]   Fig. 6.   Diffusion of ONOO into human platelets (top). DCF-DA (50 µM)-labeled platelets (3 ml, 4 × 107 cells/ml) were analyzed in a fluorescence spectrophotometer with stirring at 37°C. ONOO was delivered in 0.1 N NaOH solution (1-3 µl) to the bottom of the cuvette, outside the light beam, via a microsyringe equipped with a 25-cm-long flexible thin needle (0.25 mm i.d.). Cells were irradiated with ex of 488 nm, and fluorescence emission was recorded at em of 515 nm. Decomposed ONOO showed no effect. Diffusion of ONOO across platelet membrane was slowed by DIDS (300 µM), an inhibitor of HCO3/Cl exchanger (bottom).

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

Platelets have been long known to play important roles in hemostasis, thrombosis, and inflammation and to embody multiple mechanisms that modulate their function. More recent studies have described the potent effects of reactive oxygen and nitrogen radicals on platelet adhesion, aggregation, and structure (Salvemini et al., 1991; Iuliano et al., 1994; Jiang et al., 1999). Because NO effectively competes with SOD for the scavenging of O₂, it has been hypothesized that an association between NO and O₂ leads to formation of ONOO, which is thought to be a key oxidant molecule in many pathological processes (Wolin, 1996). Interestingly, much attention has been focused recently on the regulation of superoxide levels by NO and the risk of cardiovascular disease (Rubbo et al., 1994; Kooy et al., 1997; Leeuwenburgh et al., 1997). ONOO could also play a role by affecting platelets, which are implicated in the development of such a disease. It has also been suggested that an association between NO and O₂, which generates radicals of OH and NO₂ reactivity via decomposition of ONOO, may lead to lipid peroxidation and, ultimately, atherosclerosis (Radi et al., 1991; Balazy, 1994; Rubbo et al., 1994). We used an in vitro model to investigate conditions under which ONOO could modulate arachidonic acid metabolism in platelets. We hypothesized that ONOO would nitrate platelet COX, which would subsequently inhibit platelet activation. Because platelets generate both O₂ (Leo et al., 1977; Marcus et al., 1997) and NO (Malinski et al., 1993; Mehta et al., 1995), the formation of ONOO likely occurs in resting and stimulated platelets. Detection of specific fluorescence has provided indirect evidence that platelets of diabetic patients could generate ONOO (Tannous et al., 1999). Under normal conditions, the amount of platelet SOD is ~5 µM (1 fg/platelet) (Marcus et al., 1977), and this sufficiently deactivates all superoxide to the dismutation products, oxygen and H₂O₂. However, SOD competes for O₂ with NO:

The rate constant of O₂ with SOD (2.4 × 10⁹ M¹ s¹) (Koppenol, 1998) multiplied by SOD concentration gives a rate of 1.2 × 10⁴ s¹ for O₂ disappearance in platelets. Because basal levels of NO in platelets are ~10 nM (Zhou et al., 1995), this concentration multiplied by the rate constant for the reaction of NO with O₂ (2 × 10¹⁰ M¹ s¹) gives the rate for O₂ disappearance of 2 × 10² s¹, which is smaller than the value 1.2 × 10⁴ s¹. Activated platelets can produce 1 µM NO (Malinski et al., 1993; Zhou et al., 1995). This concentration multiplied by the rate constant yields the rate of ONOO formation of 2 × 10⁴ s¹. This rate is 67% higher than the rate of O₂ disappearance via the reaction with SOD. Thus, it is likely that in activated platelets, a significant amount of O₂ will react with NO to form ONOO rather than with SOD. The maximal rate of ONOO formation in platelets could reach 0.5 × 10⁷ M s¹. This value is similar to the rate of O₂ dismutation to H₂O₂ (Finazzi-Agro et al., 1982) and NO production by platelet NOS (Malinski et al., 1993; Zhou et al., 1995). Thus, a steady-state production of ONOO in stimulated platelets within 15 min could reach 45 µM (15 × 60 s × 0.5 × 10⁷ M s¹).

Our results further extend these observations and show that ONOO diffuses into human platelets and modulates arachidonic acid metabolism via three mechanisms. First, nitration of platelet proteins inhibited platelet aggregation stimulated by arachidonic acid. Detection of 3-nitrotyrosine-containing COX by Western blotting and GC/MS confirmed that increased nitration of COX tyrosine residues correlated with inhibitory effect of ONOO. Additionally, the treatment of platelets with epigallocatechin gallate appeared to provide efficient protection against ONOO-induced nitration of platelet COX. Experiments with two other compounds, TNM and NO₂BF₄, further supported our observation that the nitration of critical tyrosine residues of COX decreases formation of proaggregatory prostanoids in platelets. Inhibition of TXA₂ formation may potentially involve interaction of ONOO with TXA₂ synthase. However, the work of Zou et al. (1997) has demonstrated that ONOO is not likely to react with TXA₂ synthase to inhibit TXA₂ formation. Our experiments also show that COX-2 is more sensitive to nitration by ONOO than COX-1. Although it has been suggested that COX-2 in platelets may play a role in aspirin resistance (Weber et al., 1999), our results also indicate that platelet COX-2 activity may be regulated by ONOO. Modification of tyrosine residues by nitration substantially increases acidity of the tyrosine hydroxyls (Jiang and Balazy, 1998; Koppenol, 1998). Consequently, nitrated proteins are negatively charged and are unlikely to form tyrosine radicals at 3-nitrotyrosine residues. Because formation of tyrosyl radicals in COX protein at Tyr385 is thought to be the initial step in the prostaglandin biosynthesis (Shimokawa et al., 1990), nitration of these critical COX tyrosine residues is likely to impair formation of tyrosyl radicals and eventually to inhibit enzymatic activity. Nitration could also alter the conformation of COX and its ability to recognize and bind to the substrate. Platelets contain high levels of CO₂ (~1 mM) and thiols (~5 mM), which will react with ONOO. Although the reaction of glutathione with ONOO produces a relatively stable S-nitroglutathione (Balazy et al., 1998), the product of the ONOO/CO₂ reaction, carboxyperoxynitrite, is highly unstable and has been suggested to be the ultimate protein nitration compound (Koppenol, 1998).

The use of high concentrations in vivo has questioned the significance of ONOO as a biologically relevant oxidant mediator. However, because the half-life of ONOO is ~1 s in aqueous buffers, the understanding of ONOO toxicity requires a comparison of not only the concentration but also the time of exposure with the species involved (e.g., a product with units of time × concentration). The net exposure to ONOO can be obtained from the pseudo first-order kinetics of ONOO decomposition as [ONOO ]₀/k₁ (k₁ = 0.64 s¹) (Koppenol, 1998). The exposure to bolus 300 µM ONOO is equivalent to the exposure to a steady-state concentration of 1 µM ONOO for only 8 min. We observed that bolus addition of ONOO caused nitration of COX proteins at 1 to 10 µM, nitration of whole platelet COX at 50 µM, and inhibition of platelet aggregation at 4 to 50 µM. These concentrations would be equivalent to the exposure of 4 to 25 nM steady-state concentrations of ONOO within 8 min. Although direct measurements of ONOO by platelets have not yet been reported, it appears that platelets can generate ONOO at concentrations that may affect prostaglandin biosynthesis.

Second, ONOO stimulated prostaglandin biosynthesis when it was added simultaneously with arachidonic acid. In this situation, ONOO was likely to contribute to peroxide tone required for eicosanoid signaling. As a hydroperoxide substrate for the COX peroxidase activity, ONOO has the potential to increase prostaglandin formation. Landino et al. (1996) have shown that ONOO is a better substrate for the peroxidase of both COX-1 and COX-2 and exhibited higher V_max values than either H₂O₂ or 15-HpETE. Because the rate of ONOO-induced tyrosine nitration is ~2 × 10⁵ M¹ s¹, which is ~10- to 100-fold lower than the reaction of ONOO with iron of hemoproteins (Koppenol, 1998), it is unlikely that ONOO causes nitration of COX in the presence of arachidonic acid.

Finally, a new aspect of ONOO interaction with platelets is the observation of unique eicosanoids that appeared to originate from free radical processes. Although it has been suggested that ONOO-mediated oxidation of DCFH into a fluorescent form may be due to a direct reaction of peroxynitrous acid (a protonated form of ONOO at physiological pH) with DCFH (Possel et al., 1997), our experiments also indicate that NO₂ could contribute to increases in platelet fluorescence. Additionally, the similarity between the arachidonate metabolic profile in platelets exposed to either ONOO (the present study) or NO₂ (Jiang et al., 1999) suggests that NO₂ is a key substance that targeted arachidonic acid. Our experiments also suggest that ONOO-derived NO₂ remained in cells for periods of time sufficiently long to react with added arachidonic acid. Prütz et al. (1985) has shown that the generation of arachidonyl radicals by NO₂ is faster than disproportionation of NO₂ to nitrite and nitrate. We have previously observed that NO₂ reacts with arachidonic acid via two major mechanisms. One leads to the formation of hydroperoxides and, eventually, isoprostaglandins, and the other leads to trans-arachidonic acids (Jiang et al., 1999). The cis-trans isomerization of double bonds is a process characteristic for NO₂. Iso-PGF₂ and trans-arachidonic acids are likely to contribute to the complexity of biological effects induced by ONOO because both are biologically active in platelets (Morrow et al., 1992; Berdeaux et al., 1996). Another metabolite that we observed in ONOO-treated platelets, epoxyhydroxyeicosatrienoic acid, was likely to originate from 12-hydroperoxyeicosatetraenoic acid (12-HpETE), a platelet lipoxygenase product, under inhibition of platelet selenium-dependent glutathione peroxidase (GSH-Px). ONOO is known to inactivate GSH-Px via oxidation of the ionized selenol of the selenocysteine residue in the enzyme's active site (Padmaja et al., 1998). Thus, inactivation of platelet GSH-Px, a major enzyme involved in the conversion of 12-HpETE into 12-HETE, would elevate the levels of 12-HpETE. Consequently, in the absence of active peroxidase, a scission of the O---O bond could transform 12-HpETE into EpHETrE (Blee et al., 1993).

In conclusion, our work indicates that ONOO is a reactive and invasive molecule that targets COX and arachidonic acid and could play the role of a hormone-like mediator that modulates eicosanoid biosynthesis. Because the generation of NO and O₂ and, therefore, ONOO may occur independently of processes that lead to prostaglandin formation, the timing of these events is critical to predict the net effect of ONOO.