J Biol Chem, Vol. 274, Issue 35, 24664-24670, August 27, 1999

Reaction of Peroxynitrite with Reduced Nicotinamide Nucleotides, the Formation of Hydrogen Peroxide^*

Michael Kirsch and Herbert de Groot

From the Institut für Physiologische Chemie, Universitätsklinikum, Hufelandstrasse 55, D-45122 Essen, Germany

	ABSTRACT

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

NAD(P)H acts as a two-electron reductant in physiological, enzyme-controlled processes. Under nonenzymatic conditions, a couple of one-electron oxidants easily oxidize NADH to the NAD^· radical. This radical reduces molecular oxygen to the superoxide radical (O₂) at a near to the diffusion-controlled rate, thereby subsequently forming hydrogen peroxide (H₂O₂). Because peroxynitrite can act as a one-electron oxidant, the reaction of NAD(P)H with both authentic peroxynitrite and the nitric oxide (^·NO) and O₂ releasing compound 3-morpholinosydnonimine N-ethylcarbamide (SIN-1) was studied. Authentic peroxynitrite oxidized NADH with an efficiency of ~25 and 8% in the absence and presence of bicarbonate/carbon dioxide (HCO₃/CO₂), respectively. NADH reacted 5-100 times faster with peroxynitrite than do the known peroxynitrite scavengers glutathione, cysteine, and tryptophan. Furthermore, NADH was found to be highly effective in suppressing peroxynitrite-mediated nitration reactions even in the presence of HCO₃/CO₂. Reaction of NADH with authentic peroxynitrite resulted in the formation of NAD⁺ and O₂ and, thus, of H₂O₂ with yields of about 3 and 10% relative to the added amounts of peroxynitrite and NADH, respectively. Peroxynitrite generated in situ from SIN-1 gave virtually the same results; however, two remarkable exceptions were recognized. First, the efficiency of NADH oxidation increased to 60-90% regardless of the presence of HCO₃/CO₂, along with an increase of H₂O₂ formation to about 23 and 35% relative to the amounts of added SIN-1 and NADH. Second, and more interesting, the peroxynitrite scavenger glutathione (GSH) was needed in a 75-fold surplus to inhibit the SIN-1-dependent oxidation of NADH half-maximal in the presence of HCO₃/CO₂. Similar results were obtained with NADPH. Hence, peroxynitrite or radicals derived from it (such as, e.g. the bicarbonate radical or nitrogen dioxide) indeed oxidize NADH, leading to the formation of NAD⁺ and, via O₂, of H₂O₂. When peroxynitrite is generated in situ in the presence of HCO₃/CO₂, i.e. under conditions mimicking the in vivo situation, NAD(P)H effectively competes with other known scavengers of peroxynitrite.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Oxoperoxonitrate(1) (ONOO) can be formed in vivo from the diffusion-controlled reaction (k = 3.9-6.7 × 10⁹ M¹ s¹) between superoxide (O₂) and nitric oxide (nitrogen monoxide, ^·NO) (1, 2). This anion and its conjugated acid (hydrogen oxoperoxonitrate(1), ONOOH), collectively often referred to as peroxynitrite, have been suggested to be decisively involved as cytotoxic agents in several pathophysiological processes like, e.g. atherosclerosis (3) and stroke (4).

The pathological activity of peroxynitrite presumably is based on its capability to oxidize protein and nonprotein sulfhydryls (5), membrane phospholipids (6), low density lipoproteins (3), and to nitrate tyrosine (7). Under in vivo conditions, low amounts of peroxynitrite seem to be formed continuously. This process can be mimicked in experimental systems with the O₂ and ^·NO-releasing compound SIN-1.¹ Peroxynitrite generated in situ from SIN-1 has been shown to attack many biological targets (e.g. low density lipoproteins (3)) in nearly the same manner as bolus addition of authentic peroxynitrite.

We have recently reported (8) that both authentic and in situ generated peroxynitrite oxidizes HEPES and related tertiary amines in a one-electron step with initial formation of an amine radical cation. Subsequent -deprotonation generates an -aminoalkyl radical that rapidly reduces molecular oxygen to O₂, from which H₂O₂ is further produced. This reaction appears to be so effective that, in the presence of HEPES, SIN-1-induced cytotoxicity to L929 cells was mainly conveyed by H₂O₂ but not by ONOO/ONOOH (9).

As HEPES and related tertiary amines are widely used to maintain the pH in biological systems, numerous studies of the (patho)physiological effects of peroxynitrite must be expected to be obscured by this mechanism, that is H₂O₂-driven pathways instead of peroxynitrite-mediated effects may have been observed. Consequently, the question arises whether the reaction mechanism elucidated for the reaction of peroxynitrite with HEPES also applies to biologically relevant tertiary amines. One very likely candidate for such an in vivo relevant tertiary amine is NADH as indicated by the following findings: (i) Kobayashi et al. (10) have found that peroxynitrite reacts with NADH with a second-order rate constant of 4 × 10³ M¹ s¹ at pH 7.4, and Ewing et al. (11) observed a high efficiency of SIN-1 to oxidize NADH to NAD⁺ in the absence of HCO₃/CO₂. However, both groups did not provide any further information on the resulting products and on the underlying mechanism. (ii) NADH is very sensitive toward one-electron oxidants (12). (iii) The Br₂ radical, which has a standard reduction potential close to that of peroxynitrous acid [E°(Br₂/2 Br) = 1.62 V (13), E°(ONOOH/NO₂^·_(aq)) = 1.6 V (14)], has been reported to attack NADH via one-electron transfer (Reaction 1) (15). Subsequent deprotonation leads to an alkyl radical (NAD^·) that rapidly (k = 2 × 10⁹ M¹ s¹) reduces oxygen to superoxide radicals (15) from which H₂O₂ is formed (Reactions 2-4). These findings led us to assume that substitution of Br₂ by peroxynitrite should also result in O₂/H₂O₂ formation via the same mechanism. The present study was carried out to verify this hypothesis. Interestingly, the experiments with authentic peroxynitrite and peroxynitrite generated in situ from SIN-1 led to surprising different results.

	EXPERIMENTAL PROCEDURES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Materials-- Catalase from beef liver (EC 1.11.1.6), peroxidase from horseradish (EC 1.11.1.7), lactate dehydrogenase from hog muscle (EC 1.1.1. 27), alcohol dehydrogenase from yeast (EC 1.1.1.1), Cu,Zn-superoxide dismutase from bovine erythrocytes (EC 1.15.1.1), NADH, NAD⁺, and NADPH were obtained from Roche Molecular Biochemicals (Mannheim, Germany). Manganese superoxide dismutase from Escherichia coli, manganese dioxide, H₂O₂, DTPA, and ADP-ribose were from Sigma (Deisenhofen, Germany). Commercially available mixtures of oxygen 5.0 and nitrogen 5.0 (20.5% O₂, 79.5% N₂, "synthetic air") and commercially available mixtures of oxygen 5.0 and nitrogen 5.0 and carbon dioxide 4.6 (20.5% O₂, 74.5% N₂, 5% CO₂) were purchased from Messer-Griessheim (Oberhausen, Germany). SIN-1 and its decomposition product, SIN-1C, were generously provided by Dr. K. Schönafinger (Hoechst Marion Roussel, Frankfurt/Main, Germany). Oxoperoxonitrate(1) (0.73 M) was prepared by isoamylnitrite-induced nitrosation of hydrogen peroxide (0.12 mol isoamylnitrite, 100 ml H₂O₂ (1 M) plus DTPA (2 mM)) and purified (e.g. solvent extraction, removal of excess H₂O₂, N₂-purging) as described by Uppu and Pryor (16) and stored at 79 °C. All other chemicals were of the highest purity commercially available.

Solutions-- Care was taken to exclude possible contamination by bicarbonate/carbon dioxide. Double-distilled water was bubbled (2 l/min) with synthetic air at room temperature for 20 min. This water was used for synthesis of oxoperoxonitrate(1), NaOH (0.01-0.5 N) and for all other solutions. Potassium phosphate buffer (50 mM) containing DTPA (0.1 mM) was prepared freshly each day. The pH was adjusted to 7.5 at 37 °C, and the solution was again bubbled (2 l/min) with synthetic air or with the CO₂ mixture for 20 min. In the case of bubbling with the CO₂ mixture, the pH had to be readjusted to 7.5. SIN-1 solutions were prepared as 100× stock solutions at 4 °C in 50 mM KH₂PO₄ and used within 15 min.

Experimental Conditions-- SIN-1 was added to 1 ml of phosphate buffer and incubated in 12-well cell culture plates (volume of each well was 7 ml, Falcon, Heidelberg, Germany). For the detection of H₂O₂ with the catalase assay, SIN-1 was added to 10 ml of buffer and incubated in tissue culture dishes (75 ml, Falcon, Heidelberg, Germany). Under HCO₃/CO₂-free conditions, these plates/dishes were placed in an air-tight vessel (10 l). During the first 15 min of each experiment, these vessels were flushed (5 l/min) with synthetic air in a warming incubator (Heraeus, Hanau, Germany). In the presence of HCO₃/CO₂, the plates/dishes were placed in an incubator for cell culture (37 °C, humidified atmosphere of 95% authentic air and 5% CO₂, Labotect, Göttingen, Germany). The experiments with authentic peroxynitrite (2 µl of 0.35-0.035 M ONOO in 0.5 N NaOH was added to 1 ml of reaction solution) were performed in reaction tubes (1.4 ml, Eppendorf, Hamburg, Germany) by using the drop-tube Vortex mixer technique as described previously (8). Under HCO₃/CO₂-free conditions, the experiments with authentic peroxynitrite were performed in a glove-bag (Roth, Karlsruhe, Germany) under synthetic air.

Determination of H₂O₂ and O₂-- Hydrogen peroxide was quantified by various techniques. In peroxidase assays, horseradish peroxidase-catalyzed formation of a colored or a fluorescent product was measured. In these assays, residual NADH (0-200 µM) was removed first by its enzymatic conversion to NAD⁺ (10 units/ml lactate dehydrogenase, 0-200 µM pyruvate, 15-min incubation at 37 °C). Relatively high amounts of H₂O₂, i.e. 5-100 µM, were detected with 4-aminoantipyrine and 3,5-dichloro-2-hydroxybenzenesulfonic acid as substrates. The quinoneimine dye formed from these substrates was measured spectrophotometrically at 546 nm (17) (4-aminoantipyrine peroxidase assay). Concentrations of H₂O₂ in the range of 1-30 µM were analyzed with p-hydroxyphenylacetic acid (p-HPA) as the peroxidase substrate as described by Gonzalez-Flecha (18) (p-HPA peroxidase assay). The fluorescence of the dimer of p-HPA was recorded at excitation 322 nm and emission 410 nm on a spectrophotofluorometer (Amico-Bowman, Silver Spring, Maryland, USA). Alternatively, H₂O₂ was quantified by the amount of O₂ released upon addition of catalase (1000 units/ml) (catalase assay). O₂ was determined polarographically with a Clark-type oxygen electrode (Eschweiler, Kiel, Germany).

Superoxide radicals were determined by using the modified ferricytochrome c reduction technique of McCord and Fridovich (19). Various concentrations of peroxynitrite (50-2000 µM) were vortexed to the reaction solution in the presence of NADH (500 µM) and cytochrome c³⁺ (20 µM) or cytochrome c³⁺ plus SOD (625 nM, i.e. 100 units/ml). NADH was added in surplus amounts to prevent reaction of (residual) peroxynitrite with SOD and cytochrome c²⁺ formed. The resulting mixture was stored for 2 min at 37 °C. Cytochrome c²⁺ formation was determined by reading its absorbance at 550 nm (₅₅₀ = 21000 M¹ cm¹) (20). The difference in cytochrome c reduction in the presence and absence of SOD was used to calculate the amount of trapped O₂.

Determination of SIN-1 and SIN-1C-- SIN-1 and SIN-1C were quantified by capillary zone electrophoresis on a Beckman P/ACE 5000 apparatus as described previously (8).

Determination of NAD⁺, NADH, and NADPH-- NAD⁺ was determined by capillary zone electrophoresis under the following conditions: fused silica capillary (50-cm effective length, 75-µm internal diameter), hydrodynamic injection for 5 s, 30 °C temperature, 15 kV voltage, reversed polarity, UV detection at 254 nm. 50 mM sodium phosphate, 1 mM EDTA, 100 mM dodecyltrimethylammonium bromide (pH 7.0) was used as electrolyte system. To each sample, 0.25 mM 4-aminobenzamide was added as internal standard. Alternatively, NAD⁺ was determined by an enzymatic method using yeast alcohol dehydrogenase, ethanol, and semicarbazide (alcohol dehydrogenase assay) as described by Klingenberg (21). After treatment with authentic peroxynitrite or with SIN-1, residual NAD(P)H was quantified by reading its fluorescence with excitation at 339 nm and emission at 460 nm (21). Standard calibration curves were prepared from known amounts of NAD(P)H. Additionally, the oxidation of NAD(P)H was also followed photometrically at 340 nm using ₃₄₀ = 6200 M¹ cm¹ (21). Both methods gave identical results; therefore, only one parameter, decrease of fluorescence, will be shown.

Determination of Peroxynitrite-driven Nitration Reactions-- Peroxynitrite (1 mM)-dependent nitrations of p-HPA, tyrosine, and tryptophan (each 1 mM) were employed. After vortexing, the samples were allowed to stand for 2 min. In the case of tryptophan, the absorbance of 6-nitrotryptophan (22) was detected photometrically at 400 nm (_M = 5200 M¹ cm¹). With p-HPA and tyrosine as targets, 0.5-1 N NaOH was added (4:1 v/v, final pH 11-11.5). The formed products, i.e. 3-nitro-4-hydroxyphenylacetic acid (3-NO₂-4-HPA) or 3-nitrotyrosine, were determined by reading the absorbance at 430 nm (_M = 4400 M¹ cm¹) (16) and at 428 nm (_M = 4200 M¹ cm¹) (23), respectively.

	RESULTS

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Authentic Peroxynitrite

NAD(P)H as a Target for Peroxynitrite-- In the absence of HCO₃/CO₂, peroxynitrite (0-800 µM) oxidized NADH (200 µM) with an efficiency of ~25%, whereas in the presence HCO₃/CO₂ (20 mM, 5%) only ~8% of the added amount of peroxynitrite was able to oxidize NADH (Fig. 1).

View larger version (26K): [in this window] [in a new window]   Fig. 1.   Peroxynitrite-dependent NADH oxidation. Peroxynitrite (0-800 µM) was added by vortexing to 200 µM NADH in 50 mM potassium phosphate buffer (0.1 mM DTPA, 37 °C, pH 7.5) in the absence and presence of HCO3/CO2 (20 mM, 5%). After vortexing, the samples were stored for 5 min in a water bath maintained at 37 °C. Residual NADH was quantified by reading its fluorescence with excitation at 339 nm and emission at 460 nm. Each value represents the mean ± S.D. of three experiments performed in duplicate.

Because HCO₃/CO₂ is believed to act as peroxynitrite scavenger, the influence of other known peroxynitrite scavengers, namely cysteine, glutathione, methionine, and tryptophan, on peroxynitrite-dependent consumption of NADH was studied (Table I). Rather high concentrations of these scavengers, i.e. [scavenger] >5 × [NADH], were needed to inhibit peroxynitrite-dependent consumption of NADH half-maximally. Interestingly, the HO^· radical scavenger Me₂SO (up to 50 mM), which convincingly has been shown to scavenge HO^· radicals derived from peroxynitrite (25), did not exhibit any protective effect on peroxynitrite (0.25 mM)-mediated oxidation of NADH (200 µM, data not shown). In further experiments, the chemical reactivity of NADPH toward peroxynitrite was compared with that of NADH. Virtually the same results were obtained. Thus, after addition of 0.25 mM peroxynitrite to 200 µM NADPH in the absence of HCO₃/CO₂ 133.8 ± 4.7 µM, residual NADPH was detected, and the concentration of GSH necessary to protect NADPH half-maximal was also 1.1 ± 0.1 mM.

View this table: [in this window] [in a new window]   Table I Influence of scavengers on peroxynitrite-mediated consumption of NADH Peroxynitrite (0.25 mM) was vortexed to NADH (0.2 mM) in the presence of various scavengers (0-10 mM) in 50 mM potassium phosphate buffer (0.1 mM DTPA, 37 °C, pH 7.5) solutions. To find the scavenger concentration necessary to inhibit peroxynitrite-mediated oxidation of NADH half-maximal, concentrations of scavengers were increased stepwise ([scavenger] 250 µM) from 0 to 2.5 mM (Cys, GSH) or to 5 mM (Met). In the case of tryptophan, concentration was increased by 1 mM steps to reach a final concentration of 10 mM. After vortexing, the samples were stored for 5 min in a water bath maintained at 37 °C. NADH was quantified by reading the fluorescence with excitation at 339 nm and emission at 460 nm. Data are means ± S.D. of three experiments performed in duplicate.

Based on the data of Table I, we suspected that NADH inhibited peroxynitrite-mediated nitration reactions (Table II). Interactions of peroxynitrite (1 mM) with the target compounds p-HPA, tyrosine, and tryptophan (each 1 mM) in the absence and in the presence of HCO₃/CO₂ yielded amounts of nitrated products that were virtually identical to those reported previously (8, 22, 26). The addition of NADH very strongly decreased the peroxynitrite-mediated nitration of the selected compounds under both conditions. Concentrations of NADH in the range of 100-200 µM reduced the formation of nitroaromatics by 50% in the HCO₃/CO₂-free situation, in the presence of HCO₃/CO₂ the concentrations of NADH had to be increased to 480-580 µM to inhibit the formation of the nitrated products half-maximally.

View this table: [in this window] [in a new window]   Table II Effect of NADH on the peroxynitrite-induced nitration of aromatic compounds Peroxynitrite (1 mM) was vortexed to para-hydroxyphenylacetic acid (p-HPA), tyrosine, and tryptophan (each 1 mM) in the absence and presence of HCO3/CO2 (20 mM, 5%), NADH (0-1000 µM) in 50 mM potassium phosphate buffer (0.1 mM DTPA, 37 °C, pH 7.5). After a 5-min incubation at 37 °C, the formation of nitrated products (3-(NO2)-4-HPA, 3-nitrotyrosine, 5-nitrotryptophan) were determined by reading the absorbance at maximum (400-430 nm). Each value represents the mean ± S.D. of four experiments performed in duplicate.

Formation of NAD⁺-- The reaction of peroxynitrite with NADH yielded exclusively NAD⁺ at concentrations of peroxynitrite lower or comparable with those of NADH, irrespective of the presence of HCO₃/CO₂ (capillary zone electrophoresis and alcohol dehydrogenase assay, data not shown). The situation somehow changed when excess concentrations of peroxynitrite were employed. Whereas in the presence of HCO₃/CO₂, NADH was still exclusively oxidized to NAD⁺ (up to 4 mM peroxynitrite, 200 µM NADH), in the absence of HCO₃/CO₂, only 167 ± 7 µM NAD⁺ were formed upon addition of 2 mM peroxynitrite to 200 µM NADH (alcohol dehydrogenase assay). Now ADP-ribose was detected as additional reaction product (capillary zone electrophoresis, data not shown).

Formation of O₂-- The most commonly used assay for O₂ comprises the reduction of cytochrome c³⁺ in the absence and presence of SOD (19). Cytochrome c²⁺ and SOD, however, react rapidly with peroxynitrite at reaction rate constants of 2.3 × 10⁵ M¹ s¹ (27) and about 1 × 10⁵ M¹ s¹ (28), respectively. Thus, the use of both proteins to detect O₂ in the presence of peroxynitrite would yield ambiguous results. On the other hand, as peroxynitrite reacts with NADH at a rate constant of k  1 × 10⁴ M¹ s¹ (Table I), the reaction of (residual) peroxynitrite with both cytochrome c²⁺ (formed from reaction of cytochrome c³⁺ (20 µM) with O₂) and SOD (625 nM, i.e. 100 units/ml) should occur to only a very limited degree in the presence of 500 µM NADH. Formation of O₂ was undetectable in the absence of peroxynitrite (Fig. 2). After addition of 50 µM peroxynitrite, the formation of about 10 µM O₂ was detected. Formation of O₂ further slightly increased with increasing concentration of peroxynitrite to a maximum value of 12.1 ± 0.2 µM O₂ at 200 µM peroxynitrite. Thus, attack of peroxynitrite on NADH indeed generates O₂. At higher peroxynitrite concentrations, formation of superoxide apparently decreased until, at 2 mM peroxynitrite, formation of O₂ could not longer be detected, presumably because of the reaction of excess peroxynitrite with cytochrome c²⁺.

View larger version (16K): [in this window] [in a new window]   Fig. 2.   Peroxynitrite-dependent O2 production. Peroxynitrite (0-2 mM) was added by vortexing to 500 µM NADH, 20 µM cytochrome c3+ in 50 mM potassium phosphate buffer (0.1 mM DTPA, 37 °C, pH 7.5) in the absence and presence of 625 nM SOD (100 units/ml). After vortexing, the samples were stored for 2 min in a water bath maintained at 37 °C. Cytochrome c2+ was detected by reading the absorbance at 550 nm. For each peroxynitrite concentration, the difference A550absence of SOD  A550presence of SOD was used to calculate the trapped amounts of O2. Each value represents the mean ± S.D. of four experiments performed in duplicate.

Formation of H₂O₂-- In the absence of NADH, only low amounts of H₂O₂ (0-3.4 ± 0.1 µM) (p-HPA peroxidase assay) were found upon addition of ONOO to mere phosphate buffer (Fig. 3), reflecting the detectable hydrogen peroxide base level of 0.29 mol % of our peroxynitrite stock solution rather than H₂O₂ formation from peroxynitrite. In the presence of NADH, H₂O₂ was produced, and its concentration increased linearly with the concentration of peroxynitrite. The yield of H₂O₂ accounted for ~2.7% of the amount of added peroxynitrite and about 10.8% of the consumed NADH. At peroxynitrite concentrations higher than 700 µM, i.e. when all of the NADH was consumed (see Fig. 1), formation of H₂O₂ levelled off. Noticeably, the stepwise addition of ONOO (five × 140 µM each) to the 200 µM NADH doubled the yield of H₂O₂ to 39.8 ± 1.3 µM (4-aminoantipyrine peroxidase assay), corresponding to ~5.7% of the accumulated amount of peroxynitrite and ~19.9% of the consumed NADH.

View larger version (24K): [in this window] [in a new window]   Fig. 3.   Peroxynitrite-dependent H2O2 production. Peroxynitrite (0-800 µM) was added by vortexing to 200 µM NADH in 50 mM potassium phosphate buffer (0.1 mM DTPA, 37 °C, pH 7.5). After vortexing, the samples were stored for 5 min in a water bath maintained at 37 °C. H2O2 was quantified using the p-HPA peroxidase assay. The values in the presence of NADH were corrected for the corresponding amount of H2O2 derived from peroxynitrite. Each value represents the mean ± S.D. of three experiments performed in duplicate.

In the presence of HCO₃/CO₂ (20 mM, 5%), only 13.1 ± 1.0 µM H₂O₂ (p-HPA peroxidase assay) was formed upon addition of peroxynitrite (700 µM) to NADH (200 µM). This result is in line with the decreased consumption of NADH in the presence of HCO₃/CO₂. However, the yield of H₂O₂ now accounted for ~21% relative to the consumed NADH, i.e. twice as much as in the absence of HCO₃/CO₂.

Peroxynitrite from SIN-1

Guided by the fact that a stepwise addition of small amounts of authentic peroxynitrite were much more effective in the generation of H₂O₂ from NADH than a single bolus addition, experiments with peroxynitrite continuously generated from the ^·NO/O₂-releasing compound SIN-1 were performed. The experiments were carried out as endpoint determinations after 4 h of incubation at 37 °C. After that time, SIN-1 (400 µM or lower) was completely degraded, as indicated by capillary zone electrophoresis (data not shown).

NAD(P)H as a Target for Peroxynitrite-- In the absence of SIN-1, about 95% (85%) of the initial NADH (NADPH) concentration (200 µM) could still be detected after 4 h of incubation, regardless of the presence of HCO₃/CO₂ (20 mM, 5%) (Fig. 4). In the absence of HCO₃/CO₂, SIN-1 oxidized NADH with an efficiency decreasing from 90 to 60% with increasing SIN-1 concentration. Surprisingly, and in sharp contrast to the experiments performed with authentic peroxynitrite in the presence of HCO₃/CO₂, a slightly higher efficiency of SIN-1 to degrade NADH was observed. The nonlinear concentration dependences (Fig. 4) implied that the decomposition product of SIN-1, SIN-1C, might interfere with the NADH oxidation. SIN-1C (0-400 µM), however, did not have any effect on NADH oxidation both in the absence and presence of HCO₃/CO₂ (data not shown). SIN-1 (0.2 mM) also effectively oxidized NADPH (0.2 mM). In the absence and presence of HCO₃/CO₂, only 56.2 ± 1.2 µM and 35.9 ± 1.8 µM residual NADPH, respectively, were found after 4 h of incubation. Thus, peroxynitrite in situ generated from SIN-1 was remarkably more effective than authentic peroxynitrite added as a bolus. This is especially true in the presence of HCO₃/CO₂.

View larger version (23K): [in this window] [in a new window]   Fig. 4.   SIN-1-dependent oxidation of NADH. SIN-1 (0-400 µM) was incubated for 4 h with 200 µM NADH in 50 mM potassium phosphate buffer (pH 7.5, 0.1 mM DTPA, 37 °C) in the absence and presence of HCO3/CO2 (20 mM, 5%). Residual NADH was quantified by reading its fluorescence with excitation at 339 nm and emission at 460 nm. These values were corrected for autoxidation of NADH. Each value represents the mean ± S.D. of three experiments performed in duplicate.

To compare the reactivity of peroxynitrite generated from SIN-1 with that of authentic peroxynitrite, the effects of peroxynitrite scavengers on SIN-1-mediated NADH oxidation were analyzed in the absence of HCO₃/CO₂ (Table III). In line with the experiments performed with authentic peroxynitrite (Table I), concentrations of scavengers severalfold higher than those of NADH were necessary to inhibit the SIN-1-mediated oxidation of NADH half-maximally. The apparent reaction rate constants (~1 × 10⁴ M¹ s¹, Table III) are in good agreement with the values found for authentic peroxynitrite (Table I). Likewise, NADPH (200 µM) oxidation by SIN-1 (0.15 mM) was inhibited half-maximally at 1.1 ± 0.1 mM GSH.

View this table: [in this window] [in a new window]   Table III Effects of peroxynitrite scavengers on SIN-1-mediated oxidation of NADH SIN-1 (0.15 mM) was mixed to NADH (0.2 mM) in the presence of various scavengers (0-10 mM) in 50 mM potassium phosphate buffer (0.1 mM DTPA, 37 °C, pH 7.5). To find the scavenger concentration necessary to inhibit SIN-1-mediated oxidation of NADH half-maximal, the concentration of the scavengers was increased stepwise ([scavenger], 125 µM) from 0 to 1.25 mM (Cys, GSH, thiourea). In the case of both methionine and tryptophan, concentration was increased by 1 mM steps to reach a final concentration of 10 mM. After the addition of SIN-1, the samples were stored for 4 h in a warming incubator. NADH was quantified by reading the fluorescence with excitation at 339 nm and emission at 460 nm. Data are means ± S.D. of three experiments performed in duplicate.

Because the peroxynitrite scavenger GSH is a major antioxidant in vivo, the inhibitory effect of GSH (0-25 mM) on SIN-1 (0.15 mM)-dependent oxidation of NAD(P)H (0.2 mM) was also studied in the presence of HCO₃/CO₂ (Fig. 5, A and B). Under that condition, roughly 65 µM residual NAD(P)H could be detected after 4 h of incubation in the absence of GSH. Between 0.4 and 5 mM GSH, SIN-1-mediated oxidation of NAD(P)H was somewhat reduced. The residual amount of NAD(P)H increased to values in the range of 80-100 µM and was found to be largely independent of the concentration of GSH. A further increase of the GSH concentration led to an approximate exponential increase of the residual concentration of NAD(P)H. A limiting value of 200 µM of residual NAD(P)H was extrapolated at ~36 mM GSH. The effect of GSH on SIN-1-dependent oxidation of NAD(P)H was half-maximal at 15.6 ± 0.1 mM GSH (for NADH) and at 14.8 ± 0.1 mM GSH (for NADPH), respectively. As deduced from the data in Table III, GSH protected NAD(P)H effectively in the absence of HCO₃/CO₂ (Fig. 5, A and B). Noticeably, now the "plateau" at 80-100 µM NAD(P)H in the 0.4-5 mM GSH region could not be detected, and full protection of NAD(P)H oxidation was already achieved at ~10 mM GSH. Thus, the addition of HCO₃/CO₂ strongly diminished the inhibitory effect of GSH on SIN-1-mediated oxidation of NAD(P)H. Cysteine inhibited SIN-1 (0.15 mM)-mediated oxidation of NAD(P)H (0.2 mM) in the presence of HCO₃/CO₂ more effectively, with half-maximal concentrations of 2.0 ± 0.1 mM (for NADH) and 3.0 ± 0.1 mM (for NADPH), respectively. These experiments with thiols strongly suggest that an oxidant(s) other than peroxynitrite itself is mainly responsible for the oxidation of NAD(P)H in the presence of HCO₃/CO₂.

View larger version (16K): [in this window] [in a new window]   Fig. 5.   SIN-1-dependent oxidation of NAD(P)H, effect of GSH. SIN-1 (150 µM) was incubated for 4 h with various concentrations of GSH in 50 mM potassium phosphate buffer (pH 7.5, 0.1 mM DTPA, 37 °C) in the absence and presence of HCO3/CO2 (20 mM, 5%). A, 200 µM NADH as additive; B, 200 µM NADPH as additive. Residual NAD(P)H was quantified by reading its fluorescence with excitation at 339 nm and emission at 460 nm. These values were corrected for autoxidation of NAD(P)H. Each value represents the mean ± S.D. of three experiments performed in duplicate.

To demonstrate that the NADH-oxidizing species indeed derives from peroxynitrite, Cu, Zn-, and Mn-superoxide dismutases (100 units/ml each) were tested for their abilities to suppress SIN-1 (150 µM)-mediated oxidation of NADH (200 µM). In the absence and in the presence of HCO₃/CO₂, SIN-1-dependent oxidation of NADH was inhibited by SOD by about 50 and 35%, respectively, virtually independent of the type of SOD applied (three experiments performed in duplicate).

Formation of NAD⁺-- In line with the results reported for authentic peroxynitrite, SIN-1-mediated oxidation of NADH yielded exclusively NAD⁺ (alcohol dehydrogenase assay, data not shown), provided that the concentration of SIN-1 was less than twice the concentration of NADH.

Formation of H₂O₂-- In the absence of NADH, 400 µM SIN-1 yielded only 4.1 ± 0.2 µM H₂O₂ (p-HPA peroxidase assay), i.e. 2.1 mol % if a 2:1 stoichiometry of SIN-1/H₂O₂ is assumed, irrespective of the presence of HCO₃/CO₂. However, addition of SIN-1 to NADH (200 µM) resulted in the formation of amazingly high amounts of H₂O₂ (Fig. 6). In the absence of HCO₃/CO₂, the maximal amount of H₂O₂ (about 20 µM) observed in the experiments with authentic peroxynitrite (see above) was already obtained by application of 50 µM SIN-1 (4-aminoantipyrine peroxidase assay). Formation of H₂O₂ increased continuously with the concentration of SIN-1 to reach a plateau value of ~60 µM at about 300 µM SIN-1, indicating complete consumption of NADH. In the presence of HCO₃/CO₂ (20 mM, 5%), 50 µM SIN-1 produced nearly 10 µM H₂O₂, and generation of H₂O₂ increased approximately linearly with the concentration of SIN-1 to reach a plateau value of about 70 µM H₂O₂. In further experiments, decomposition of SIN-1 (300 µM) in the presence NADPH (200 µM) instead of NADH yielded identical amounts of H₂O₂ after 4 h of incubation, that is, 60.2 ± 3.4 µM H₂O₂ were formed in the absence of HCO₃/CO₂ and 69.6 ± 3.7 µM H₂O₂ in its presence (catalase assay). Thus, SIN-1 dependent production of H₂O₂ from NAD(P)H is highly effective, corresponding to about 30 and 35% of the amount of decomposed NAD(P)H in the absence and in the presence of HCO₃/CO₂, respectively.

View larger version (23K): [in this window] [in a new window]   Fig. 6.   SIN-1-dependent H2O2 production. SIN-1 (0-400 µM) was incubated for 4 h with 200 µM NADH in 50 mM potassium phosphate buffer (pH 7.5, 0.1 mM DTPA, 37 °C) in the absence and presence of HCO3/CO2 (20 mM, 5%). After the incubation, H2O2 was quantified using the 4-aminoantipyrine peroxidase assay. Each value represents the mean ± S.D. of three experiments performed in duplicate.

	DISCUSSION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Oxidizing Species-- Various oxidants and radicals can be formed from peroxynitrite at physiological pH values (Fig. 7). Under HCO₃/CO₂-free conditions, peroxynitrous acid (ONOOH) has been reported to directly attack a variety of substrates such as tryptophan (22). Alternatively, it may dissociate with a yield of roughly 10% (25) into the highly reactive free radicals HO^· and NO₂^· (29). In the presence of HCO₃/CO₂, however, ONOO rapidly reacts with CO₂ (30) to give the free radicals CO₃ (31) and NO₂^· (32) via the putative intermediate nitrosoperoxycarbonate (ONOOCO₂).

View larger version (11K): [in this window] [in a new window]   Fig. 7.   Proposed mechanism of hydrogen peroxide formation from NAD(P)H and peroxynitrite. SIN-1 releases ·NO/O2 which combine to ONOO. In the absence of HCO3/CO2, peroxynitrite acts via peroxynitrous acid (ONOOH) and/or via HO· and NO2· derived from ONOOH. The present paper supported the view that ONOOH mainly oxidized NADH, but an involvement of HO· in the oxidation process cannot be ruled out with certainty. In the presence of HCO3/CO2, peroxynitrite-derived radicals, especially CO3, seemed to oxidize NADH. After the attack of each oxidant (ONOOH, CO3, "NO2", "HO·") the NAD· radical should be formed, so that O2 is subsequently generated from reaction with O2. Dismutation of O2 finally leads to H2O2.

Because of the facts that (i) NADH is readily oxidized by one-electron oxidants (12) to the NAD^· radical (15), (ii) ONOOH can act as an one-electron oxidant (33), and (iii) HO^·, NO₂^· and CO₃ radicals are also one-electron oxidants, there can be little doubt that reaction of peroxynitrite with NADH will initially produce the NAD^· radical. This radical readily reduces O₂ to O₂, and both products of this reaction, namely NAD⁺ and O₂, were indeed found from reaction of authentic peroxynitrite with NADH. As O₂ dismutates spontaneously to H₂O₂, the latter was likewise detected after reaction of NAD(P)H with both authentic peroxynitrite and peroxynitrite generated in situ from SIN-1. As expected, both reduced nicotinamides, NADH and NADPH, react in virtually the same manner. Because most of the above free radicals can react with each other, the yields of the various recombination products cannot be estimated. For example, NO₂^· reacts at pH 7.5 about 2 × 10⁵ times faster with O₂ than O₂ reacts with itself (34, 35). Therefore, H₂O₂ is unlikely to be formed in roughly stoichiometric amounts.

In the absence of HCO₃/CO₂, ONOOH appears to oxidize NADH directly. This follows from the fact that the HO^· radical scavenger Me₂SO was unable to protect NADH against the attack of peroxynitrite and also because reaction of HO^· with NADH should lead to significant (~40%) formation of products other than NAD⁺ (12). In the presence of HCO₃/CO₂, peroxynitrite scavengers but not superoxide dismutases were largely ineffective in protecting NADH against the attack of SIN-1. Hence, ONOOH appears not to be the major oxidant in such a situation. The experiments performed with GSH and cysteine as additive indicate that the CO₃ radical is the presumed attacking species. Compared with cysteine, a 7.6-fold higher concentration of GSH was necessary to inhibit SIN-1-mediated oxidation of NADH half-maximal, in good agreement with the fact that the CO₃ radical reacts about 8.7 times faster with cysteine than with GSH (36). The reactivity of CO₃ on NADH can be estimated by making use of the fact that the rate constants of the reaction of NADH with a variety of radicals correlate with the reduction potentials of the oxidizing radicals.² Thus, CO₃ is expected to react with NADH with a rate constant of 7 × 10⁸ M¹ s¹. Because of its lower reduction potential, NO₂^· should react about seven-fold slower with NADH_.

Authentic Versus in Situ-generated Peroxynitrite-- In the absence of HCO₃/CO₂, the steady-state concentration of ONOOH generated in situ from SIN-1 should be much less than upon bolus addition of authentic peroxynitrite, as the release of ^·NO and O₂ from SIN-1 is a slow process at pH 7.5 (t_1/2 = 40 min (9)). Because the relative concentration of NAD(P)H remains high under these conditions, the oxidation of NAD(P)H is favored and the attack of either ONOOH and HO^· on ONOO is disfavored. Consequently, the relative yield of NADH oxidation by SIN-1 is increased about three-fold compared with the situation where authentic peroxynitrite is employed.

In the presence of HCO₃/CO₂, in situ generated peroxynitrite from SIN-1 was about 10-fold more effective in oxidizing NADH than authentic peroxynitrite. This surprisingly different capability is hard to explain with the current knowledge about the chemical characteristics of authentic and in situ generated peroxynitrite.

Putative (Patho)Physiological Significance-- GSH is considered to be a major scavenger of intracellularly operating peroxynitrite (5), and, indeed, in the absence of HCO₃/CO₂ GSH prevented effectively the (physiologically uncommon) one-electron oxidation of NAD(P)H by in situ generated peroxynitrite. In the presence of HCO₃/CO₂, however, peroxynitrite predominantly reacts with CO₂ (39) to yield CO₃ and NO₂^· (see above). GSH and other thiols readily react with NO₂^· (40, 41), but GSH reacts about 100-fold slower with the CO₃ radical (36). Therefore, the GSH/NAD(P)H ratio must be high to protect NAD(P)H effectively against the attack of peroxynitrite-derived CO₃ radicals. For instance, the total amount of GSH in rat liver typically lies in the range of 5-10 µmol/g wet weight (e.g. Ref. 42). To the contrary, the total amount of NAD(P)H in rat liver cytosol was found in the range of 90 nmol/g wet weight (43), whereas the bulk of NAD(P)H is located in the mitochondria with a total amount up to ~0.4 µmol/g wet weight (43), i.e. ~20 times lower than the total amount of GSH. On the other hand, as GSH in a 75 × surplus was necessary to inhibit the SIN-1-mediated oxidation of NAD(P)H half-maximal in the presence of HCO₃/CO₂, GSH should be rather ineffective in inhibiting the attack of in vivo generated peroxynitrite on NAD(P)H. This implies that the reaction products O₂/H₂O₂ may be the dominating factor decisively contributing to the (patho)physiological effects ascribed originally attributed to peroxynitrite. In line with this notion, H₂O₂ and nitrotyrosine (the latter is believed to be a marker for peroxynitrite) have been reported to be elevated in patients with adult respiratory distress syndrome (44, 45). Hydrogen peroxide induces the activation of NF-B (46, 47), and this transcription factor has been implicated in atherosclerosis (48) and rheumatoid arthritis (49), i.e. in diseases where formation of nitrotyrosine has also been detected (50, 51). Furthermore, both peroxides can induce DNA single strand breaks (52-55), thus activating poly(ADP-ribose) synthetase (52, 56). Accordingly, authentic peroxynitrite-dependent DNA cleavage and subsequent apoptosis in HL-60 cells in phosphate-buffered saline has been reported to be mediated mainly by hydrogen peroxide (57).

	ACKNOWLEDGEMENTS

We are indebted to Professor Lehnig for access in his CIDNP experiments. We thank Dr. H.-G. Korth for a series of clarifying discussions pertaining to the nature of the attacking species and for useful comments on this manuscript. The present investigation would have been impossible without the technical assistance of E. Heimeshoff and A. Wensing.

	FOOTNOTES

^* The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

To whom correspondence should be addressed. Tel.: +201/723-4101; Fax: +201/723-5943.

² Correlation: log k (NADH + radical) = 42.64 × (1  exp(3.5628 × E°))  33.58; r² = 0.98. Calculated from nine different radicals: 1, [Fe(CN)₆]^3-, log k = 0.5, E° = 0.45 V (37); 2, promazine radical cation, log k = 5.7, E° = 0.71 V (37); 3, HO₂^·, log k = 5.26 (38), E° = 0.75 V (13); 4, chlorpromazine radical cation, log k = 6.9, E° = 0.78 V (37); 5, m-benzosemiquinone, log k = 6.9, E° = 0.81 V (37); 6, promethazine radical cation, log k = 7.4, E° = 0.86 V (37); 7, J₂, log k = 7.7 (15), E° = 1.03 V (13); 8, (SCN)₂, log k = 8.67 (15), E° = 1.32 V (13); 9, Br₂, log k = 8.95 (15), E° = 1.62 V (13). Radicals of interest: CO₃, E° = 1.5 V; NO₂^·_, E° = 1.04 V (13).

	ABBREVIATIONS

The abbreviations used are: SIN-1, 3-morpholinosydnonimine N-ethylcarbamide; SOD, superoxide dismutase; p-HPA, 4-hydroxyphenylacetic acid; 3-NO₂-4-HPA, 3-nitro-4-hydroxyphenylacetic acid; DTPA, diethylenetriaminepentaacetic acid; GSH, glutathione.

	REFERENCES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES