Iron Catalyzes both Decomposition and Synthesis ofS-Nitrosothiols: Optical and Electron Paramagnetic Resonance Studies

NITRIC OXIDE: Biology and Chemistry Vol. 1, No. 3, June, pp. 191–203 (1997) Article No. NO970122

Iron Catalyzes both Decomposition and Synthesis of S-Nitrosothiols: Optical and Electron Paramagnetic Resonance Studies Anatoly F. Vanin,1 Irina V. Malenkova, and Vladimir A. Serezhenkov Institute of Chemical Physics, Russian Academy of Sciences, Kosygin Str. 4, Moscow 117977, Russia

Received September 30, 1996, and in revised form February 17, 1997

Formation of S-nitrosothiols was demonstrated in 1–50 mM aqueous solutions of cysteine or glutathione (cys-NO or GS-NO, respectively) upon contact of thiols with gaseous nitric oxide under a pressure of 50–600 mm Hg and anaerobic conditions. The yield of S-nitrosothiols was increased by mixing with NO plus air at a molar ratio [NO]/[O2 from air] of no less than 40. In this instance, the S-nitrosothiol formation was optimum at a NO pressure of 100–150 mm Hg. The addition of 0.25 mM o-phenanthroline, a selective Fe2/ chelator, to thiol solutions prior to the treatment with NO or NO / air completely blocked the formation of S-nitrosothiols. On the other hand, this process was potentiated by the addition of Fe2/ but not Cu2/ ions. These data indicated a crucial influence of Fe2/ on the process. The contact of o-phenanthroline with Snitrosothiols synthesized by a routine method (treatment of thiol solutions with the NO / NO2 mixture at pH õ1) did not induce their degradation at pH 3–10. Moreover, o-phenanthroline strikingly enhanced the cys-NO stability at neutral pH. Cysteine, glutathione, and desferal, a selective Fe3/ chelator, exerted a similar effect on cys-NO. The stabilizing effect of thiols on cys-NO was accompanied by the formation of dinitrosyl–iron complexes with thiol-containing ligands containing admixed (intrinsic) iron (1–2 mM). The addition of Fe2/ at a concentration higher than 10 mM abolished the stabilizing effect of thiols on cys-NO. Therefore iron can induce both degradation and synthesis of S-nitrosothiols. According to the proposed

1

To whom correspondence should be addressed. Fax: (095) 93821-56.

mechanisms such opposite effects of iron on S-nitrosothiols are determined by the ratio between S-nitrosothiols, thiols, iron, and NO in the reaction system. q 1997 Academic Press

Key Words: dinitrosyl–iron complexes; nitric oxide; S-nitrosothiols.

Investigators studying the biological functions of nitric oxide have displayed growing interest in Snitrosothiols (RSNO),2 which are detected in cells and tissues generating NO (1–3). Inclusion of NO in RSNO has been suggested to provide stabilization and transport of this highly reactive agent in the organism. In addition, S-nitrosylation of thiols sharply accelerates their oxidation as well as their reactivity in reactions of various attaching functional groups. This considerably affects many metabolic processes involving thiols (3). However, the mechanism of thiol S-nitrosylation in biosystems remains obscure. In pure chemical systems the process occurs in strongly acid media in the reaction of thiols with the nitrosonium ion (NO/) produced by higher nitrogen oxides, N2O4 and N2O3 (4–7). At neutral (‘‘physiological’’) pH values, this process is hampered 2

Abbreviations used: BSA, bovine serum albumin; cys-NO, Snitrosocysteine; DNIC, dinitrosyl–iron complex; EPR, electron paramagnetic resonance; GSH, reduced glutathione; GSNO, Snitrosoglutathione; L, anionic ligand in DNIC; NO/, nitrozonium ion; NO0, nitroxyl ion; RS0, ionized form of thiol; RSNO, S-nitrosothiol. 191

1089-8603/97 $25.00 Copyright q 1997 by Academic Press All rights of reproduction in any form reserved.

AID

NO 0122

/

am04$$$$81

06-27-97 16:13:04

noal

AP: NO

192

VANIN, MALENKOVA, AND SEREZHENKOV

by the high rate of hydrolysis of these oxides in aqueous solution. In effect, as analysis has shown, only glutathione at concentrations of 5 mM and higher could be S-nitrosylated in considerable amounts in the reaction with higher nitrogen oxides at the level of NO produced from L-arginine in animal cells and tissues (7). Thiol groups of proteins form virtually no S-nitrosothiols under these conditions (7). The situation is also complicated by the slow rate of NO autooxidation, which may allow the reaction of NO with other agents in cells or tissues to predominate (8). Therefore, only a small portion of intracellular NO is oxidized to generate the products which could nitrosylate thiols. There is another possible mechanism of thiol Snitrosylation in aqueous solutions at neutral pH, with transition metal ions catalyzing the conversion of neutral NO molecules into nitrosonium ions (7, 9–11). Such conversion may occur in the incorporation of NO into dinitrosyl–iron complexes (DNIC) with various anion ligands, including thiol-containing compounds (12–14). We have proposed a mechanism of NO to NO/ conversion in these complexes (15). Our study demonstrated that S-nitrosothiols were formed in a neutral pH aqueous solution in the absence of oxygen and with the participation of DNIC. It is known that iron ions also catalyze the decomposition of S-nitrosothiols (16, 17) accompanied by DNIC formation (17). Therefore, both the processes of formation and decomposition of S-nitrosothiols catalyzed by iron seem to be interrelated. Predomination of either process is apparently determined by the ratio of reagent (nitric oxide, iron ions, thiols, and S-nitrosothiols) concentrations. Using the methods of optical and EPR spectroscopy, we investigated the conditions under which iron ions catalyze primarily the synthesis or the decomposition of S-nitrosocysteine (cys-NO) and S-nitrosoglutathione (GSNO) in aqueous solutions at acid, neutral, or alkaline pH values.

periments. To remove the methemoglobin admixture from the hemoglobin preparation, the latter was treated with dithionite and passed through a column packed with Sephadex G-25. Gaseous NO was synthesized in the reaction of FeSO4 with NaNO2 in 0.1 M HCl with subsequent purification by the method of fractional lowtemperature sublimation in an evacuated system.

MATERIALS AND METHODS

Synthesis of S-nitrosothiols. S-Nitrosocysteine and S-nitrosoglutathione were synthesized in a Thunberg vial using several methods. In Method 1, 2 ml of 50 mM thiol solution in a mixture of 15 mM Hepes buffer and 0.1 M HCl (1:1) was treated with a mixture of NO / NO2 for 5 min. NO was administered to the Thunberg vial at a pressure of 100– 200 mm Hg. Then the 3-mm inlet of the vessel was opened to air for 1 s. This resulted in the advent of NO2 in the gas phase. When a thiol solution was shaken in this gas mixture, the solution quickly gained the pink coloration characteristic of S-nitrosothiol. After 5 min the NO / NO2 mixture was evacuated from the vessel for 10 min. In Method 2, 2 ml of 1–50 mM thiol solution in 15 mM Hepes buffer (pH 4–10) was treated with NO and 0.6 ml of air at atmospheric pressure. The other stages proceeded according to Method 1. In Method 3, 2 ml of 1–50 mM thiol solution in 15 mM Hepes buffer, pH 3–10 was treated with NO alone for 5 min. With Methods 2 and 3 of S-nitrosothiol synthesis, the NO pressure was varied within the range of 50–600 mm Hg. The concentration of synthesized S-nitrosothiols was evaluated using the optical method with the intensity of absorption band at 338 nm characteristic of these compounds. To test the effects of iron and copper ions on S-nitrosothiol synthesis, FeSO4 or CuSO4 solutions in distilled water were administered to the upper part of a Thunberg vial and poured together with the thiol solution in the NO or air / NO atmosphere. To remove the intrinsic or added iron from thiol solutions, a 0.1% solution of o-phenanthroline mixed with ethanol or dimethyl sulfoxide with water (1:10) was used. o-Phenanthroline was preliminarily dissolved in pure ethanol or dimethyl sulfoxide.

Materials. L-Cysteine, reduced glutathione (GSH) (Sigma, U.S.A.), ferrous sulfate, cupric sulfate (Fluka, UK), o-phenanthroline (Chemapol, Czechia), desferrioxamine B (desferal) (Ciba, Switzerland), sodium dithionite (Merck, Germany), bovine serum albumin, and horse hemoglobin (Sigma, U.S.A.) were used in the ex-

S-Nitrosothiol degradation. Degradation of cysNO and GSNO synthesized by Method 1 was detected spectrophotometrically by the decrease in the absorbance at 338 nm on contact of the solutions with air. The 1 M NaOH solution was used to increase the pH of solutions. Cysteine, GSH, o-phenan-

Copyright q 1997 by Academic Press. All rights of reproduction in any form reserved.

AID

NO 0122

/

am04$$$$81

06-27-97 16:13:04

noal

AP: NO

DECOMPOSITION AND SYNTHESIS OF S-NITROSOTHIOLS

193

throline, or desferal was added to the initial S-nitrosothiol solutions and the pH was increased. Solutions of FeSO4 or CuSO4 in distilled water were added to the S-nitrosothiol solutions at neutral pH. Synthesis of dinitrosyl–iron complexes. DNICs with cysteine or GSH were obtained by treating the FeSO4 and cysteine (GSH) solutions with gaseous NO in 15 mM Hepes buffer (pH 7.4) at a molar ratio of 1:2 or 1:20 (dimeric diamagnetic DNIC 1:2 or monomeric paramagnetic DNIC 1:20 (18), respectively) at a NO pressure of 100–200 mm Hg. Synthesis was carried out in a Thunberg vial containing 100 ml of gas-phase volume and 5 ml of thiol solution. The FeSO4 solution was placed in the upper part of container and mixed with the thiol solution in the presence of NO. The DNIC with phosphate ligand was synthesized in a similar way. To form a DNIC with a thiol group in serum albumin or hemoglobin, DNIC with phosphate, cysteine, or GSH at 0.2–1.0 mM was added to the 1 mM solutions of these proteins. The concentration of synthesized paramagnetic DNIC was measured using the EPR method by the intensity of EPR signal from the complex and performing double integration. A frozen solution of stable nitroxyl radical 2,2*,6,6*,-tetramethylpiperidol-1-oxyl), a generous gift of Dr. A. B. Shapiro, was used as a standard paramagnetic sample. EPR and optical measurements. EPR spectra were recorded at 77K or 293K using a modified EPR radiospectrometer, either Radiopan (Poland) or an ESC-106 (Bruker, Germany) in the X-diapason, respectively. Optical spectra were recorded using both Uvikon-941 and Specord UV–VIS (Germany). RESULTS

Decomposition of Cys-NO and GSNO Catalyzed with Iron and Copper Ions In accordance with data of other authors (19, 20) the cys-NO synthesized by us using Method 1 decomposed rapidly when the pH was increased to 6–8. In acid (pH 4–5) or alkaline (pH 9–10) solutions, it remained unchanged for at least 10 min. Figure 1a illustrates the cys-NO instability at neutral pH by the kinetics of cys-NO monitored by the decrease in solution optical absorbance at 338 nm. When the pH of a 2 mM cysNO solution was rapidly increased from 0.8 to 7.5 , cys-NO decayed in less than 1 min. This process was sharply attenuated by the addition of an iron chelator,

FIG. 1. Kinetics of optic absorption at 338 nm of cys-NO (a–c) and GSNO (d) at neutral pH under different actions. (a) Curve 4, initial cys-NO solution (2 mM), (preparation 1); curves 1–3, addition of 0.25 mM o-phenanthroline, 1 mM cysteine, and 1 mM desferal, respectively, to preparation 1. (b) Sequential addition of 20 mM cysteine and 100 mM FeSO4 (curve 1) or 1 mM cysteine and 100 mM FeSO4 (curve 2) to preparation 1. (c) Sequential addition of 5 mM cysteine and 10 mM FeSO4 or 10 mM CuSO4 to the cys-NO solution (1.5 mM). (d) Curve 1, addition of 200 mM CuSO4 or FeSO4 to the initial GSNO (1.2 mM) solution (preparation 2); curves 2– 5, addition of 5 mM glutathione, 5 mM glutathione / 200 mM FeSO4 , 5 mM cysteine or 5 mM cysteine / 200 mM FeSO4 respectively to preparation 2.

desferal or o-phenanthroline, selectively binding Fe3/ and Fe2/, respectively. Cysteine or glutathione added in concentrations of 1 mM or higher exerted a similar effect (Fig. 1a). In this instance, the cys-NO solution displayed DNICs with thiol-containing ligands immediately after the addition of thiol as detected by a characteristic EPR signal (Fig. 2a). The signal intensity corresponded to the incorporation of 1–2 mM intrinsic iron into DNIC. Therefore the added thiols enhanced the stability of cys-NO in apparently the same way as the iron chelators: they transferred free intrinsic iron to the bound form, that is, DNIC. The stabilizing effect of cysteine on cys-NO was sharply attenuated by the addition of Fe2/ salt at a concentration exceeding 10 mM to the solution (Fig. 1b). In this process additional DNIC appeared and contained the added iron. The solution gained the green color characteristic of paramagnetic DNIC with thiol-containing ligands. Figure 2b shows an optical absorption spectrum of this solu-

Copyright q 1997 by Academic Press. All rights of reproduction in any form reserved.

AID

NO 0122

/

am04$$$$81

06-27-97 16:13:04

noal

AP: NO

194

VANIN, MALENKOVA, AND SEREZHENKOV

FIG. 2. (Left) Spectra of optical absorption (A) from 1 mM cysNO solution at pH 7.5 (a–c) after the addition of 5 mM cysteine (curve a); 1 mM cys-NO /5 mM cysteine / 200 mM FeSO4 (curve b); 2 mM cys-NO / 1 mM cysteine / 500 mM FeSO4 (curve c); and 1 mM cys-NO / 1 mM cysteine / 500 mM FeSO4 , pH 6.0 (curve d). (Right) EPR spectra from the above cys-NO solutions. The EPR spectra were recorded at 77K, microwave power 5 mW, and modulation amplitude 0.5 mT. To the right of the spectra, relative amplification of radiospectrometer.

tion. At a high concentration of added iron, which was only one to two times lower than the concentration of added cysteine, diamagnetic DNIC emerged with yellow to orange color of the solution and absorption bands at 290–310 and 360 nm (Fig. 2c) characteristic of this form of DNIC with cysteine (18). In this process, judging from the intensities of the bands, an overwhelming portion of the added iron (up to 80%) was incorporated into diamagnetic DNIC. The position of the band for DNIC with cysteine at 290–310 nm depended on pH: at pH 7.5 it localized to 290 nm, while at pH 6.0 it shifted to 310 nm (Fig. 2d). At pH 6–8, diamagnetic DNIC with glutathione displayed absorption bands at 310 and 360 nm. The formation of diamagnetic and paramagnetic DNICs started immediately after the addition of Fe2/ salt. Then the cys-NO level began to decline, sometimes in a sophisticated kinetic manner (Fig. 1b).

In these solutions the formation of DNIC could in principle occur by two mechanisms: either directly in the course of cys-NO molecule decomposition on their binding to Fe2/ or as a result of the binding of neutral NO molecules released from cys-NO with Fe2/ ions and cysteine, i.e., during the secondary reaction following the cys-NO decomposition. To elucidate this question we studied the effect of hemoglobin (0.25 mM) as a NO trap on the formation of DNIC upon contact of cys-NO (2 mM) with cysteine (10 mM) and Fe2/ (0.25 mM). Hemoglobin had virtually no influence on the formation of DNIC in these solutions. In the presence and in the absence of hemoglobin, the complex concentration was 0.20 { 0.05 and 0.20 { 0.05 mM (by three measurements), respectively. The EPR spectra of DNIC were recorded at room temperature 1 min after the addition of Fe2/ ions to the solutions. When gaseous NO instead of cys-NO was used as the NO source, hemoglobin (0.25 mM) sharply suppressed the formation of DNIC in cysteine (10 mM) and Fe2/ (0.25 mM) solutions. Two milliliters of the solutions was treated with gaseous NO in a 100-ml Thunberg vial. Gaseous NO was introduced into the vial in the quantity of 5 mmol (under a pressure of 7–8 mm Hg) (data not shown). The stabilizing effect of cysteine on cys-NO was abolished by the addition of Cu2/ at 10 mM and higher to the solution (Fig. 1c). It should be noted that in these experiments, the cys-NO decomposition was apparently due to Cu/. EPR analysis of the solutions showed that the EPR signal from Cu2/ disappeared when Cu2/ was added to the solution of cysteine, thus acting as a reducer of Cu2/. In a comparison of the kinetics of the 1.3 mM cys-NO decomposition initiated by 10 mM Fe2/ or Cu2/ in the presence of 10 mM cysteine, we did not observe any significant difference in the rate of 1.3 mM cys-NO decomposition induced by either ion. The half-time of cys-NO decomposition was 15–20 min in both instances. The time increased with the increase in cysteine concentration and decreased with the increase in copper or iron concentration to 50 mM. Furthermore, there were no significant differences in the kinetics of cys-NO decomposition initiated by copper or iron. The stabilizing effect of glutathione on cys-NO was also attenuated by the addition of Fe2/ (50–100 mM). In this instance however, the S-nitrosothiol decomposition induced by these ions was considerably slower: the half-time of S-nitrosothiol decomposition

Copyright q 1997 by Academic Press. All rights of reproduction in any form reserved.

AID

NO 0122

/

am04$$$$81

06-27-97 16:13:04

noal

AP: NO

DECOMPOSITION AND SYNTHESIS OF S-NITROSOTHIOLS

was up to 1 h at the same concentrations of metal ions. In an aqueous solution, GSNO remained completely stable for 1 h at pH fluctuating between 3 and 10. The addition of up to 200 mM Fe2/ or Cu 2/ to the 1 mM GSNO solution at neutral pH did not result in an appreciable decrease of GSNO within 15 min (Fig. 1d). In the presence of glutathione (5 mM), the GSNO content (1.2 mM) was reduced at neutral pH by no more than 10% within 60 min (Fig. 1d). The addition of 10–50 mM Fe2/ or Cu2/ together with glutathione (5 mM) decreased this level by no more than 20% within 1 h with no significant difference in the effects of iron and copper ions on the GSNO stability. Cu2/ ions initiated the GSNO degradation more efficiently than iron ions at higher concentrations (§200 mM). With the addition of 200 mM Cu2/ to the 1 mM GSNO solution at neutral pH in the presence of glutathione (5 mM), GSNO decayed within 1 min. In this process a compound, perhaps including copper, appeared and displayed an absorption band at 320 nm. In similar experiments with the addition of 200 mM Fe2/, the GSNO level decreased by only 30% in 1 h (Fig. 1d). In similar experiments the process of GSNO degradation sharply accelerated when cysteine was added instead of glutathione (Fig. 1d). In this process, as in the addition of glutathione, paramagnetic DNIC appeared. Thus Fe2/ and Cu2/ at concentrations not exceeding 50 mM catalyzed with equal efficiency the decomposition of cys-NO or GSNO in the presence of cysteine or glutathione in the solutions. This is in agreement with data of other investigators concerning the effects of these ions on the stability of S-nitroso-N-acetylpenicillamine (16) or GSNO (21). This fact appears to be most interesting in discussing the question of which of these metals is responsible for S-nitrosothiol decay. With regard to the stabilizing effect of ‘‘selective’’ chelators of copper or iron on S-nitrosothiols, for example, the effect of o-phenanthroline of desferal on cys-NO, we do not believe that this single fact is sufficient proof for the decisive role of intrinsic iron in the decay of the compound. These chelators also bind other metals. For instance, o-phenanthroline efficiently binds Cu/ (22). The same can be noted with respect to the number of works in which various chelators have been used for studying the role of copper ions in the decay of Snitrosothiols (16, 21, 23–25).

The Formation of Cys-NO and GSNO Catalyzed with Iron Ions We have earlier demonstrated that cys-NO and GSNO are formed in cysteine and glutathione aqueous solutions on contact with NO under anaerobic conditions at pH 7 (15). The yield of S-nitrosothiols, as calculated on a thiol basis, reached 30–40% (for 1–3 mM thiol solutions). Since the formation of Snitrosothiols was enhanced 1.5–2 times by the addition of Fe2/ salt (to 20 mM) and was completely suppressed by the selective iron chelator o-phenanthroline (250 mM), we concluded that iron ions, both intrinsic and added, play a catalytic role in this process. Iron binds to NO and forms DNIC, which is capable of producing nitrosonium ions. The latter induce the formation of S-nitrosothiols. In the present work we studied the effect of thiol solution pH on the synthesis of cys-NO and GSNO under the same conditions. As pH was decreased from 7 to 5, synthesis of these S-nitrosothiols was enhanced (3–4 times) to an equal extent and persisted at a pH down to and including 3 (Fig. 3). The addition of o-phenanthroline (250 mM) to these solutions before the treatment with NO completely blocked the formation of S-nitrosothiols. At alkaline pH (7.5 and more), cys-NO did not appear. In the NO treatment of the alkaline cysteine solution, a precipitation was observed as early as in 2–3 min, and was apparently due to the oxidation of this thiol to cystine slightly soluble in water (26, 27). o-Phenanthroline at a concentration of 250 mM blocked this oxidation. In the NO-treated glutathione solution, GSNO appeared even at pH values up to 10 in the same concentration independent of the pH value. The addition of a nitrosyl–iron complex with a phosphate or thiol ligand (20–180 mM) to the glutathione solution and the subsequent treatment of the solution with NO increased the synthesis of GSNO 1.5– 2 times independent of pH changes in the range of 7–10 (Fig. 3). The formation of cys-NO and GSNO in acid and neutral thiol solutions was sharply potentiated by the addition of 0.6 ml of air together with NO to the Thunberg vial. Moreover, in such treatment, cys-NO appeared even at alkaline pH (Fig. 3). Calculations show that the amount of air entering the Thunberg vial was adequate to produce no more than 15 mM of NO2 per 100 ml volume in the gas phase. The amount

Copyright q 1997 by Academic Press. All rights of reproduction in any form reserved.

AID

NO 0122

/

am04$$$$81

06-27-97 16:13:04

195

noal

AP: NO

196

VANIN, MALENKOVA, AND SEREZHENKOV

FIG. 3. The pH dependence of cys-NO (top) and GSNO (bottom) formation in 2 ml of 50 mM solutions of cysteine of glutathione on their contact with 1 mm NO (curves a) or with a mixture of 1 mm NO / 8 mM O2 in 0.6 ml of air(curves b). Curves ,–,: addition of 50 mM FeSO4 to glutathione solutions prior to the treatment with NO or NO / O2 , respectively. Data shown are expressed as means { SE of 5 individual estimations.

of NO was 0.4–5 mm in the NO pressure range of 50–600 mm Hg at a Thunberg volume of 100 ml, respectively. The administration of 0.6 ml of air together with NO to the Thunberg vial resulted in only a slight decrease in solution pH: in 2 ml of 15 mM Hepes buffer solution, the pH in the buffer range was reduced by no more than 0.5, while at low pH it was decreased by 1. This administration was sufficient to increase the S-nitrosothiol yield 4–8 times. In this process, as in the action of NO alone, the formation of cys-NO and GSNO was more efficient at low pH values (Fig. 3). There was no appreciable difference in the formation of these S-nitrosothiols. The S-nitrosothiol yield as converted to the initial thiol concentration decreased as the latter increased. As an example we refer to the result of evaluating the GSNO yield in 2 ml of glutathione solution (pH 8) treated with a NO / air mixture at a NO pressure

of 80 mm Hg: at 1, 5, 10, and 50 mM glutathione, the GSNO yield was 100, 80, 60, and 10% respectively. In solutions with high thiol concentrations, the S-nitrosothiol yield increased with repeated treatment of these solutions with the NO / air mixture. A similar result was by achieved by decreasing the volume of thiol solution in the flask. In a single treatment of 0.5 ml of 50 mM cysteine solution with the NO / air mixture at pH 5, the cys-NO yield was up to 80%. In the treatment of thiol solutions with the NO / air mixture (at a constant air amount of 0.6 ml), the yield of S-nitrosothiols was optimum at a NO pressure of 60–150 mm Hg. At higher pressures the yield of S-nitrosothiols decreased up to zero (at a NO pressure of 600–700 mm Hg). The addition of 20–200 mM Fe2/ to solutions of cysteine or glutathione with subsequent treatment with the NO / air mixture at pH lower than 5 slightly affected the formation of corresponding S-nitrosothiols: their yield increased by no more than 1.2 times. At pH above 7, the emerging cys-NO became unstable in the presence of added iron while the GSNO yield increased by 1.5 times (Fig. 3). The air-potentiated formation of S-nitrosothiols could be due to the appearance, along with DNIC, of another source of nitrosonium ion in thiol solutions. The source is N2O3 molecules, which are formed in the NO binding to NO2 (4–8). Therefore, S-nitrosylation of thiols could occur without the catalytic effect of iron ions by the mechanism (7, 8) mentioned in the introduction. This hypothesis, however, does not agree with the results of experiments in which o-phenanthroline (250 mM), a selective Fe2/ chelator, and sodium citrate (25 mM), which binds transition metals, were added to thiol solutions. When the solutions were treated with the NO / air mixture, S-nitrosothiols emerged in amounts not exceeding 0.1 mM, if at all (Fig. 4). At the same time, when the complexing agent was added to solutions of presynthesized S-nitrosothiols, the S-nitrosothiol content did not decrease. Therefore, ions of transition metals proved necessary for thiol S-nitrosylation by the NO / air mixture as has been demonstrated previously for a similar process initiated by the treatment of thiol solutions with NO alone (15). In both instances, suppression of this process with the selective Fe2/ chelator o-phenanthroline and potentiation by the addition of Fe2/ to thiol solutions suggest that Fe2/ is necessary as a catalyst

Copyright q 1997 by Academic Press. All rights of reproduction in any form reserved.

AID

NO 0122

/

am04$$$$81

06-27-97 16:13:04

noal

AP: NO

DECOMPOSITION AND SYNTHESIS OF S-NITROSOTHIOLS

FIG. 4. Spectra of optical absorption (A) from a 50 mM glutathione solution treated with a NO / air mixture alone and after addition of 250 mM o-phenanthroline to the solution (curves a and b, respectively). Experimental conditions: 2 ml of the glutathione solution in 15 mM Hepes buffer (pH 7.5) contacted with the NO (1 mM) / air (8 mM O2 administered with air) gas mixture for 5 min.

of thiol S-nitrosylation under these conditions. With regard to the possible influence of copper ions on this process, we failed to find a statistically significant increase in GSNO synthesis with the addition of a Cu2/ salt at a concentration of 20–100 mM prior to the treatment of glutathione solution with the NO / air mixture. In similar experiments, the addition of copper to cysteine solutions attenuated cys-NO synthesis. Thus, according to our data Fe2/ ions but not copper ions (Cu2/ or Cu/) were capable of initiating GSNO synthesis. This means that the inhibiting effect of o-phenanthroline on GSNO or cys-NO synthesis was due to binding of Fe2/ to this chelator. Formation of DNIC and S-Nitrosothiols in Bovine Serum Albumin and Hemoglobin The addition of 1 mM DNIC with phosphate to 1mM BSA at neutral pH resulted in the formation of DNIC with a thiol and apparently a histidine residue resulting from the transfer of Fe/(NO/)2groups from low-molecular-weight DNIC to the residues. The EPR spectrum recorded from the complex in BSA at

room temperature is shown in Fig. 5. The complex was described previously (28, 29). The content remained unchanged with NO treatment of the solution. However the complex disappeared when NO was mixed with air in the amount mentioned above. The disappearance was associated with a broad band at 340–360 nm in the optical absorption spectrum of the BSA solution (Fig. 5). This band has been assigned to S-nitrosothiol in BSA (30). In the treatment of DNIC-free BSA with the NO / air mixture, the BSA displayed a narrower band with a maximum at 340 nm and an amplitude five times lower than that in the previous preparation. The addition of 200 mM o-phenanthroline to the BSA solution prior to the treatment with NO / air mixture sharply attenuated S-nitrosylation of this protein. The transformation of protein-bound DNIC into S-nitrosothiol was reversible. The band of optical absorption at 340–360 nm disappeared from S-nitrosylated BSA with the addition of 50 mM cysteine / 1 mM Fe2/. Furthermore, in the EPR spectrum of the solution recorded at room temperature, a narrow signal appeared at g Å 2.03, which was characteristic of DNIC with cysteine (12) (Fig. 6a). The signal intensity was an order of magnitude higher in the BSA preparation that initially contained proteinbound DNIC. In addition to the low-molecular-

FIG. 5. (Left) EPR spectra from 1 mM solutions of bovine serum albumin (a) or horse hemoglobin (b) after addition of 1.0 mM DNIC with phosphate. The spectra were recorded at room temperature at a microwave power of 20 mW and modulation amplitude of 0.5 mT. (Right) Optical spectra of a and b preparations (curves c and d, respectively) after their treatment with the NO / air (1 ml of protein solution / 1 mM NO / 8 mM O2 oxygen in air) gas mixture. When the optical spectra were recorded, the reference cell was filled with 1 mM albumin or hemoglobin solutions pretreated with the same gas mixture but excluding the preliminary contact of proteins with DNIC.

Copyright q 1997 by Academic Press. All rights of reproduction in any form reserved.

AID

NO 0122

/

am04$$$$81

06-27-97 16:13:04

197

noal

AP: NO

198

VANIN, MALENKOVA, AND SEREZHENKOV

cient than on contact of this protein with DNIC with phosphate.

DISCUSSION

FIG. 6. EPR spectra from 1 mM solutions of bovine serum albumin and horse hemoglobin after the addition of 1 mM DNIC with phosphate to the solutions (a and d, respectively); a and d preparations treated with the NO / air gas mixture (as described in the legend to Fig. 5) (b and e, respectively); after addition of 50 mM cysteine / 1 mM FeSO4 to preparations b and e (c and f, respectively). Conditions of spectrum recording are the same as in Fig. 5. At right, relative amplification of radiospectrometer.

weight DNIC, a small amount of protein-bound DNIC appeared. The latter was characterized by a broader anisotropic EPR signal (Fig. 6c). The addition of a 1 mM solution of paramagnetic monomeric or diamagnetic dimeric DNIC with cysteine or GSH to 1 mM BSA solution resulted both in formation of protein-bound DNIC and in retention or appearance of the paramagnetic low-molecularweight form of DNIC. In subsequent treatments of these solutions with the NO / air mixture, both DNIC forms disappeared. In this process, the absorbance spectrum began to show a band at 340 nm indicative of the transformation of both proteinbound and low-molecular-weight DNIC into S-nitrosothiol. The same result was obtained in similar experiments on horse hemoglobin (Figs. 5 and 6). The specific nature of this protein was found to be the following. First, judging by the shape of EPR signal from hemoglobin-bound DNIC, the latter differed from low-molecular-weight DNIC in BSA in having higher (axial) symmetry. Second, the formation of proteinbound DNIC on contact of hemoglobin with low-molecular-weight thiol-containing DNIC was less effi-

This study demonstrates the ability of iron ions to catalyze both the degradation and the synthesis of S-nitrosothiols. As mentioned in the introduction, this controversial effect of iron can be due to the interrelation of these processes, as evidenced by the advent of DNIC with thiol-containing ligands characteristic of both processes. The advent of DNIC precedes the formation of S-nitrosothiols in the reaction of NO with thiols in the presence of Fe2/ and accompanies the degradation of S-nitrosothiols catalyzed by iron ions (Scheme 1). In both instances (Reactions I and II), the process of DNIC formation is irreversible (see below). The chemical equilibrium is characteristic of only DNIC and its constituents. It has been shown (31) that the constituents are represented by Fe2/, NO, thiols, and nitrosonium ion NO/. The latter forms a S-nitrosothiol with thiol (RSNO). Therefore, among these constituents are compounds and ions capable of producing DNIC not only in accordance with their chemical equilibrium with DNIC but also in Reactions I and II. In the framework of Scheme 1, synthesis or degradation of S-nitrosothiols is determined by the ratio between rates of Reactions I and II (V1 and V2) and V3 , the rate of recombination of DNIC constituents, which results in reconstruction of this complex. If V1 @ V3 , then Reaction I of DNIC synthesis would result in accumulation of RSNO because Fe2/ and NO released from DNIC are involved mainly in Reaction

Fe™1 1 2NO 1 2RS2 V⁄

N¤O RS2

[I]

NO1 V›

Fe1 V‹

RS2 RSSR

Fe™1 1 NO 1 RS2 1 NO1

NO1 RS©NO

V¤

Fe™1 1 2RSNO 1 2RS2 SCHEME 1

Copyright q 1997 by Academic Press. All rights of reproduction in any form reserved.

AID

NO 0122

/

am04$$$$81

06-27-97 16:13:04

noal

AP: NO

[II]

DECOMPOSITION AND SYNTHESIS OF S-NITROSOTHIOLS

RS2

NO Fe21

2

RS

RS2

NO1

2

2

RS

RS2

2

NO RS

2

NO

NO1 Fe1

Fe21 1 HNO

Fe21

e NO

NO1 RS2

RS

199

1 1/2 (H¤O 1 N¤O) 1

NO

H1 SCHEME 2

I rather than in the reaction of recombination with RSNO. This situation apparently takes place with the addition of excess NO to the system. If V2 @ V3 , then DNIC synthesis in Reaction II would result in degradation of RSNO and accumulation of NO in the system because Fe2/ and RSNO released from DNIC are involved mainly in Reaction II rather than in the reaction of recombination with NO. This situation takes place with the addition of excess RSNO to the system. Therefore, according to Scheme 1 RSNO leads to accumulation of NO and the corresponding disulfide characteristic of this process.

Mechanism of S-Nitrosothiol Synthesis Catalyzed with Iron Ions It follows from Scheme 1 that the synthesis of Snitrosothiols induced by Fe2/ and NO is determined by the ability of the DNIC with thiol-containing ligands formed in this system to produce thiol-nitrosylating NO/. The production mechanism may be as follows. The formation of DNIC with various anion ligands (L) in the reaction of NO with Fe2/ is attended by the release of nitrous oxide (N2O) in an amount equimolar to the amount of DNIC (32). The appearance of N2O signifies that NO is reduced to NO0 in the course of DNIC synthesis. Subsequent NO0 protonization results in the formation of nitroxyl (HNO). Two molecules of HNO recombine to release N2O and water (6). The reduction of NO to NO0 can occur also in the absence of anion ligands such as ascorbate, which could be responsible for this reduction. In this connection scheme 2 illustrates our proposed mechanism of NO reduction to NO0 in the synthesis of DNIC with thiol-containing ligands, which results in the transformation of the latter into the paramagnetic form studied in the present investigation. We assume that the binding of two NO molecules to Fe2/ ensures the electron

transfer from one of these ligands to another with subsequent protonization of the latter and with the release of the formed nitroxyl from DNIC. A NO molecule substitutes for the nitroxyl in the complex to form paramagnetic DNIC with the following structure: (RS0)2Fe/(NO/)2 . The process is irreversible because it is accompanied by generation of N2O from the nitroxyl molecules formed in this process. One of the nitroso groups released from the DNIC thus formed provides S-nitrosylation of thiols in accordance with Scheme 1. The possibility that this process may occur in binding of thiols to NO/ and in the DNIC composition cannot be excluded. As a result, in equilibrium degradation of the complex, it releases ‘‘ready-to-use’’ S-nitrosothiols. However, in this binding of thiols to DNIC, they can be also oxidized. It is known that thiols are oxidized by NO with the formation of N2O and disulfides (26, 27). The thiols supposedly reduce NO directly to NO0, which results in the formation of N2O. We observed oxidation of cysteine with NO at pH ú7.5 in our experiments. However, the oxidation was completely blocked by o-phenanthroline, which provided evidence for a catalytic role of iron in this process as well. Since iron ions completely incorporate into DNIC in the presence of thiols and NO, there are strong grounds for assuming that the thiol binding to DNIC provides their oxidation. A tentative mechanism of this process characteristic not only for cysteine but also for other thiols (RS0) is shown in Scheme 3. In accordance with Scheme 3 the cycle is completed by the regeneration of DNIC and formation of RS• and HNO, which produce disulfides and N2O, respectively. Therefore, the binding of RS0 to NO/ ligand in DNIC can result in either thiol oxidation or thiol nitrosylation. Realization of either process is defined by DNIC stability. At a high DNIC stability, thiol oxidation predominates while at a decreased DNIC stability (increased V4 , the rate of

Copyright q 1997 by Academic Press. All rights of reproduction in any form reserved.

AID

NO 0122

/

am04$$$$81

06-27-97 16:13:04

noal

AP: NO

200

VANIN, MALENKOVA, AND SEREZHENKOV

RS2

NO1

RS2

Fe1 1 2RS2 RS2

NO1

2

1

RS

Fe RS2

Fe1

2

2

RS2 1

RS

NO 1

Fe RS2

RS2

RS2

NO2

NO1 Fe21 1 2HNO 1 RSSR

RS2

NO

1

RS

NO 1

Fe RS2

NO1 Fe21 1 HNO 1 RSSR

NO

2

1 2RS NO2

RS2

NO1 Fe1 1 RS 1 RS2

NO1

e 1 RS NO

RS2

e 1 RS2

RS2

NO RS 1

NO1RS2

H1

1 HNO 1 RSSR

RS2

NO1

NO Fe1 1 N¤O 1 RSSR 1 H¤O

H1

RS2

NO1

SCHEME 3

DNIC decomposition, according to Scheme 1) S-nitrosothiols are formed. We believe that such a decrease took place in our experiments when DNIC was treated with NO / air with the formation of cys-NO at both neutral and alkaline pH values. The destabilizing influence of NO / air mixture on DNIC was most striking in experiments with BSA and hemoglobin. The treatment of these DNIC-containing proteins with the NO / air mixture resulted in the disappearance of DNIC and the S-nitrosylation of these proteins. This transformation was reversed by treatment of preparations with thiols and Fe2/: the DNIC bound primarily to low-molecularweight ligands reappeared. It should be noted that tryptophan residues may be nitrosylated along with thiol residues in BSA (33). Nitrosylated tryptophan residues display optic absorbance in the same range as S-nitrosothiols. Therefore, we cannot exclude the possibility that tryptophan residues in BSA and hemoglobin were nitrosylated also in our experiments. The mechanism of the destabilizing effect of NO / air mixture on DNIC remains unclear. First, the question arises of which compounds generated in the system are responsible for this effect. The compounds may be nitrogen oxides formed in NO oxidation with air oxygen or peroxynitrite (ONOO0) produced by the reaction of NO with superoxide anion (O2●0) in the thiol solution (6). Superoxide anions can emerge in the solution during the oxygen–thiol interaction catalyzed by intrinsic ions of transition metals such as iron. Second, how these compounds incorporating into DNIC can reduce the DNIC stability is unclear. The questions require additional investigation. The DNIC with thiol ligands is formed at pH ú7. At lower pH, the DNIC synthesis is sharply attenuated (12, 31). Nevertheless, we also observed the S-

nitrosylation of thiols catalyzed by Fe2/ in acid solutions, with S-nitrosylation more efficient than at high pH values. It seems that at pH õ7, DNIC with other ligands, for example, water, could serve as a NO/ source. The stability of these complexes is considerably lower than that of DNIC with thiol-containing ligands (12). This is why the formation of DNIC with weak ligands was not detectable with the EPR method. The higher efficiency of thiol Snitrosylation in this pH range could be due to the enhanced stability of S-nitrosothiols in these solutions. Another source of the NO/ nitrosylating thiols in acid solutions could be high-spin mononitrosyl– iron complexes (electron spin S Å 32), the formation of which is optimum at pH 3–4 (34, 35). Mechanisms of S-Nitrosothiol Degradation According to current data (3, 5, 6, 36–40), decomposition of S-nitrosothiols can result in the release of neutral NO molecules and the appearance of corresponding disulfides. The latter can emerge as a result of recombination of the thiol radicals (RS•) formed in homolytical decomposition of S-nitrosothiols. Disulfides can also result from oxygen oxidation of reduced disulfide (RSSR)0 which is formed in the association of thyil radical and thiol (41). To explain the destabilizing effect of iron ions on cys-NO and other S-nitrosothiols one could have proposed a mechanism similar to that proposed elsewhere (25, 42) to explain the effect of copper ions on these compounds: Fe2/ / RSNO r RS0 / NO / Fe3/ Fe3/ / RS0 r RS• / Fe2/ Overall reaction: RSNO r RS• / NO

Copyright q 1997 by Academic Press. All rights of reproduction in any form reserved.

AID

NO 0122

/

am04$$$$81

06-27-97 16:13:04

noal

AP: NO

DECOMPOSITION AND SYNTHESIS OF S-NITROSOTHIOLS

NO1RS2

L Fe21 L

L

NO1 Fe1 1 RSı 1 RS2

Fe21

e NO1RS2

NO1… RSı L

L

NO … RS2 L

NO1

SCHEME 4

In the presence of thiols, neutral NO molecules released in the decomposition react with Fe2/ and thereby result in the formation of DNIC observed in the experiment. The blocking effect of o-phenanthroline and desferal, selective chelators of Fe2/ and Fe3/, on cys-NO decomposition is due to the chelator binding of both ions. However, the mechanism is inconsistent with the results of experiments which demonstrated no effect of hemoglobin, an efficient trap of neutral NO molecules, on the formation of DNIC in the reaction of cys-NO with Fe2/ in the presence of cysteine. In this connection we propose another mechanism of the destabilizing effect of Fe2/ ions, according to which DNIC is formed directly in the process of the reaction of RSNO decomposition itself as shown in Scheme 4. It is suggested that Fe2/ ions can bind two RSNO molecules each in the presence of both thiol and nonthiol (L) ligands. Such coordination ensures electron transfer through molecular orbitals of the complex from one S-nitrosothiol molecule to another with the formation of an oxidized and a reduced molecule of Snitrosothiols; i.e., their mutual oxidation reduction occurs. S-Nitrosothiol species emerging in this process are unstable (37) and they degrade to release NO, NO/, RS•, and RS0. The release of thyil radical in the process makes the latter irreversible because these radicals recombine with the formation of corresponding disulfide. The NO and NO/ released from RSNO remain bound to iron, which results in the formation of DNIC. The formed DNIC, which contains anion ligands of a nonthiol nature, is unstable (13). According to Scheme 1 the DNIC degradation results in the release of Fe2/ to catalyze again the degradation of Snitrosothiols according to Scheme 4, which results in the rapid degradation of cys-NO. Replacing the L with thiol-containing ligands leads to stabilization of DNIC. As a result Fe2/ ions contained in the solution as a micromolar admixture remain in DNIC. This prevents their destructive effect on cys-NO. The

lifetime of these concentrations of DNIC does not exceed 1 min at 1–5 mM cysteine in the solution (18). Therefore, the constant occurrence of DNIC in the cys-NO solution in the presence of cysteine should be considered a result of a steady-state process, i.e., the DNIC destruction and synthesis. What is the explanation for the cys-NO degradation in the presence of free cysteine or glutathione with the addition of Fe2/ at concentrations of 10 mM and higher to the solution? Apparently, the addition of iron increases the stationary Fe2/ level according to the chemical equilibrium in the DNIC } DNIC constituent system. This increase must potentiate the destabilizing effect of iron on cys-NO. The mechanism underlying the high stability of GSNO remains obscure. Possibly, the stability is determined by the inability for steric reasons of this Snitrosothiol to incorporate into complexes with Fe2/ with the participation of nonthiol ligands, which could provide mutual oxidation reduction and thereby degradation of GSNO in accordance with Scheme 4. Significantly, the weak degradation of GSNO is initiated by free glutathione; i.e., in this instance, the thiol exerts an effect on GSNO opposite to its effect on the cys-NO stability. We connect this effect of glutathione with its stabilizing action as a ligand in complexes of Fe2/ with GSNO to provide GSNO degradation in these complexes. The effect of iron ions on degradation and synthesis of S-nitrosothiols demonstrated in the present study in no way casts any doubt on the important role of copper ions at least in degradation of S-nitrosothiols. The mechanism of their catalytic effect on this process is currently under extensive study (16, 21, 23–25, 42); the effect has also been demonstrated in the present work. However, we failed to find a significant difference between the efficiency of the iron and copper action in all S-nitrosothiols studied at ion concentrations of 10–50 mM, which is in agreement with data of other authors (16, 21). This is why in the presence of environmental copper and iron at micromolar (contaminant) concentrations, the predominating effect of iron or copper on S-nitrosothiol decomposition will obviously be defined by their relative amount. Given that the content of iron is two orders of magnitude higher than that of copper in animal cells and tissues (43), it is clear that iron would play the major role in both degradation and synthesis of S-nitrosothiols. On this basis, we can

Copyright q 1997 by Academic Press. All rights of reproduction in any form reserved.

AID

NO 0122

/

am04$$$$81

06-27-97 16:13:04

201

noal

AP: NO

202

VANIN, MALENKOVA, AND SEREZHENKOV

propose the following model of the stabilization and transport of NO and the release of free NO in cells and tissues. Initially, the NO produced by NO-synthase incorporates into DNIC with various low-molecular-weight anion ligands, in particular with such thiol-containing compounds as cysteine or glutathione. Then low-molecular-weight DNIC together with free NO molecules moves away from NO-synthase to regions with a higher oxygen content, where part of the free NO is oxidized by oxygen or forms peroxynitrite in the reaction with superoxide anions. The emerging nitrogen oxides or peroxynitrites destabilize DNIC. As a result the NO/ groups released from DNIC initiate the formation of S-nitrosothiols. This process can occur both inside and outside cells. Since the intracellular content of low-molecular-weight thiols does not exceed 10 mM and judging from our data, DNIC must completely transform into S-nitrosothiols inside and especially outside cells. Therefore, the S-nitrosothiols serve as the major transport form of NO which carries NO from cell to cell. Subsequently, entering a zone with a high content of nonheme iron and thiols, S-nitrosothiols (by Scheme 4) initiate the formation of DNIC, which degrades to release NO. Experiments on cultured cells show that the formation of DNIC takes place upon contact of animal cells with S-nitrosothiols (28, 44). ACKNOWLEDGMENTS This work was supported by the Russian Foundation for Basic Researches (Grant 96-04-48066) and PECO (Contract ERBCIPDCT940250).

REFERENCES 1. Stamler, J. S., Jaraki, O., Osborne, J., Simon, D. I., Keaney, J., Vita, J., Singel, D., Valery, R., and Loscalzo, J. (1992). Nitric oxide circulates in mammalian plasma primarily as an S-nitroso adducts of serum albumin. Proc. Natl. Acad. Sci. USA 89, 7674–7677. 2. Clancy, R. M., Levartovsky, D., Leszcynska-Piziak, J., Yegudin, J., and Abramson, S. B. (1994). Nitric oxide reacts with intracellular glutathione and activates the hexose monophosphate shunt in human neutrophils: Evidence for S-nitrosoglutathione as a bioactive intermediary. Proc. Natl. Acad. Sci. USA 91, 3680–3684.

5. Williams, D. L. H. (1988). Nitrosation, pp. 1-35, Cambridge Univ. Press, Cambridge, UK. 6. Butler, A. R., Flitney, F. W., and Williams, D. L. H. (1995). NO, nitrozonium ions, nitrosothiols and iron–nitrosyls in biology: A chemist’s perspective. Trends Physiol. Sci. 16, 18– 23. 7. Kharitonov, V. G., Sundquist, A. R., and Sharma, V. S. (1995). Kinetics of nitrosation of thiols by nitric oxide in the presence of oxygen. J. Biol. Chem. 270, 28158–28164. 8. Wink, D. A., Dardyshire, J. F., Nims, R. W., Saavedra, J. E., and Ford, P. C. (1994). Reaction of the bioregulatory agent nitric oxide in oxygenated aqueous media: Determination of the kinetics for oxidation, and nitrosation by intermediates generated in the NO/O2 reaction. Chem. Res. Toxicol. 6, 23– 27. 9. Gwost, D., and Caulton, K. G. (1973). Reductive nitrosation of group YIIIb compounds. Inorg. Chem. 12, 2095–2099. 10. Wilkinson, G. (1974). The long search for stable transition metal alkyls. Science 185, 109–115. 11. Wayland, B. B., and Olson, L. W. (1974). Spectroscopic studies and bonding model for nitric oxide complexes of iron porphyrin. J. Am. Chem. Soc. 96, 6037–6041. 12. McDonald, C. C., Phillips, W., and Mower, H. F. (1965). An electron spin resonance study of some complexes of iron, nitric oxide and anionic ligands. J. Am. Chem. Soc. 87, 3319– 3326. 13. Burbaev, D. Sh., Vanin, A. F., and Blumenfeld, L. A. (1971). Electronic and spatial structures of paramagnetic dinitrosyl ferrous complexes. Z. Strukt. Khim. 12, 252–258. [Russian] 14. Bryar, T. R., and Eaton, D. R. (1991). Electronic configuration and structure of paramagnetic iron dinitrosyl complexes. Can. J. Chem. 70, 1917–1926. 15. Vanin, A. F., and Malenkova, I. V. (1996). Iron is a catalyst of cysteine and glutathione S-nitrosation on contact with nitric oxide in aqueous solutions at neutral pH. Biochemistry (Moscow) 61, 374–379. 16. McAninly, J., Williams, D. L. H., Askew, S. C., Butler, A. R., and Russel, C. (1993). Metal ion catalysis of nitrosothiol (RSNO) decomposition. J. Chem. Soc. Chem. Commun. 1758– 1759. 17. Vanin, A. F. (1995). Role of iron ions and cysteine in formation and decomposition of S-nitrosocysteine and S-nitrosoglutathione. Biochemistry (Moscow) 60, 441–447. 18. Vanin, A. F. (1995). On the stability of the dinitrosyl–iron complex, a candidate for the endothelium-derived relaxing factor. Biochemistry (Moscow) 60, 225–229. 19. Mathews, W. R., and Kerr, S. W. (1993). Biological activity of S-nitrosothiols: The role of nitric oxide. J. Pharmacol. Exp. Ther. 267, 1529–1537.

3. Stamler, J. S. (1994). Redox signalling: Nitrosylation and related target interactions of nitric oxide. Cell 78, 931–936.

20. Meyer, D. J., Kramer, H., Ozer, N., Coles, B., and Ketterer, B. (1994). Kinetics and equilibria of S-nitrosothiol exchange between glutathione, cysteine, penicillamines, and serum albumin. FEBS Lett. 345, 177–180.

4. Oae, S., and Shinhama, K. (1983). Organic thionitrites and related substances. Org. Prep. Proc. Int. 15, 165–198.

21. Gorren, A. C. F., Schrammel, A., Schmidt, K., and Mayer, B. (1996). Decomposition of S-nitrosoglutathione in the presence

Copyright q 1997 by Academic Press. All rights of reproduction in any form reserved.

AID

NO 0122

/

am04$$$$81

06-27-97 16:13:04

noal

AP: NO

DECOMPOSITION AND SYNTHESIS OF S-NITROSOTHIOLS

203

of copper ions and glutathione. Arch. Biochem. Biophys. 330, 219–228.

reduction. The dinitrosyl complex. Inorg. Chem. 21, 1978– 1985.

22. Sammes, P. G., and Yahioglu, G. (1994). 1,10-Phenantroline: A versatile ligands. Chem. Soc. Rev. 327–334.

33. Zhang, Y.-Y., Xu, A.-M., Nomen, M., Walsh, M., Keaney, J. F., and Loscalzo, J. (1996). Nitrosation of tryptophan residue(s) in serum albumin and model dipeptides. J. Biol. Chem. 271, 14271–14279.

23. Tulett, J. M., Hodson, H. F., Gescher, A., Shuker, D. E. G., and Moncada, S. (1995). Stability of S-nitrosothiols related to glutathione. Endothelium 3(Suppl.), S261. 24. Askew, S. C., Barnett, D. J., McAnninly, J., and Williams, D. L. H. (1995). Catalysis by Cu2/ of nitric oxide release from S-nitrosothiols (RSNO). J. Chem. Soc. Perkin Trans. 2, 741– 745. 25. Dicks, A. P., Swift, H. R., Williams, D. L. H., Butler, A. R., AlSadoni, H. H., and Brian, G. C. (1996). Identification of Cu/ as the effective reagent in nitric oxide formation from S-nitrosothiols (RSNO). J. Chem. Soc. Perkin Trans. 2, 481–487. 26. Pryor, W. A., Church, D. F., Govindan, C. K., and Crank, G. (1982). Oxidation of thiols by nitric oxide and nitrogen dioxide: Synthetic utility and toxicological implications. J. Org. Chem. 47, 156–159. 27. DeMaster, E. G., Quast, B. J., Redfern, B., and Nagasawa, H. T. (1995). Reaction of nitric oxide with the free sulfhydryl group of human serum albumin yields a sulfenic acid and nitrous oxide. Biochemistry 34, 11494–11499. 28. Vanin, A. F., Malenkova, I. V., Mordvintcev, P. I., and Mulsch, A. (1992). Dinitrosyl iron complexes with thiol-containing ligands and their reversible conversion into nitrosothiols. Biokhimiya 58, 1094–1103. [Russian] 29. Boese, M., Mordvintcev, P. I., Vanin, A. F., Busse, R., and Mulsch, A. (1995). S-nitrosation of serum albumin by dinitrosyl–iron complex. J. Biol. Chem. 270, 29244–29249. 30. Stamler, J. S., Simon, D. I., Osborne, J. A., Mullins, M. E., Jaraki, O., Michel, T., Singel, D., and Loscalzo, J. (1992). S-nitrosylation of proteins with nitric oxide: Synthesis and characterization of biologically active compounds. Proc. Natl. Acad. Sci. USA 89, 444–448. 31. Vanin, A. F., Stukan, R. A., and Manukhina, E. B. (1996). Physical properties of dinitrosyl iron complexes with thiolcontaining ligands with their relation to vasodilator activity. Biochim. Biophys. Acta 1295, 5–12. 32. Pearsall, K. A., and Bonner, F. T. (1982). Aqueous nitrosyl iron(11) chemistry. 2. Kinetics and mechanism of nitric oxide

34. Vanin, A. F., and Aliev, D. I. (1983). High-spin nitrosyl iron complexes in animal tissues. Stud. Biophys. 93, 63–68. 35. Farrar, J. A., Grinter, R., Pountney, D. L., and Thomson, A. J. (1993). Optical and magnetic properties of iron(II)–nitrosyl complexes in model compounds. J. Chem. Soc. Dalton Trans. 2703–2709. 36. Myers, P. R., Minor, R. L., Guerra, R., Bates, F. N., and Harrison, D. (1990). Vasorelaxant properties of the endotheliumderived relaxing factor more closely resemble S-nitrosocysteine than nitric oxide. Nature 345, 161–163. 37. Rubanyi, G. M., Johns, A., Wilcox, D., Bates, F. N., and Harrison, D. (1991). Evidence that a S-nitrosothiol, but not nitric oxide, may be identical with endothelium-derived relaxing factor. J. Cardiovasc. Pharmacol. 17(Suppl. 3), S41–S45. 38. Radomski. M. W., Rees, D. D., Dutra, A., and Moncada, S. (1992). S-Nitrosoglutathione inhibits platelet activation in vitro and in vivo. Br. J. Pharmacol. 107, 745–749. 39. Stamler, J. S., Singel, D. J., and Loscalzo, J. (1992). Biochemistry of nitric oxide and its redox-activated forms. Science 258, 1898–1902. 40. Arnell, D. R., and Stamler, J. S. (1995). NO/, NO and NO0 donation by S-nitrosothiols: Implications for regulation of physiological functions by S-nitrosylation and acceleration of disulfide formation. Arch. Biochem. Biophys. 318, 279–285. 41. Harman, L. S., Mottley, C., and Mason, R. P. (1984). Free radical metabolites of L-cysteine oxidation. J. Biol. Chem. 269, 5606–5611. 42. Singh, R. J., Hogg, N., Joseph, J., and Kalyanaraman, B. (1996). Mechanism of nitric oxide release from S-nitrosothiols. J. Biol. Chem. 271, 18596–18603. 43. Forth, W., and Rummel, W. (1973). Iron absorption. Physiol. Rev. 53, 724–732. 44. Roy, B., Lepoivre, M., Henry Y., and Fontecare, M. (1995). Inhibition of ribonucleotide reductase by nitric oxide derived from thionitrites: Reversible modifications of both subunits. Biochemistry 34, 5411–5420.

Copyright q 1997 by Academic Press. All rights of reproduction in any form reserved.

AID

NO 0122

/

am04$$$$81

06-27-97 16:13:04

noal

AP: NO