The reaction of superoxide radical with N-acetylcysteine

Free Radical Biology & Medicine, Vol. 29, No. 8, pp. 775–782, 2000 Copyright © 2000 Elsevier Science Inc. Printed in the USA. All rights reserved 0891-5849/00/$–see front matter

PII S0891-5849(00)00380-4

Original Contribution THE REACTION OF SUPEROXIDE RADICAL WITH N-ACETYLCYSTEINE MUSTAPHA BENRAHMOUNE,* PATRICE THE´ ROND,†

and

ZOHREH ABEDINZADEH*

*Laboratoire de Chimie-Physique, Universite´ Paris V, Paris, France; and †Centre Hospitalie`re de Versailles, Hoˆpital Andre´ Mignot, Service de Biochimie-Pharmacologie-Toxicologie, Chesnay, France (Received 27 March 2000; Revised 20 June 2000; Accepted 7 July 2000)

Abstract—The interaction of superoxide radicals with N-acetylcysteine (RSH) in an aqueous solution of pH 7 using the technique of steady state radiolysis has been investigated in this paper. The radiolytic yield of the products (G value) of RSH consumption and disulfide of N-acetylcysteine (RSSR) formation has been determined. The G value of the products is not dependent on the concentration of RSH (at the plateau of dilution curve) or on the inverse of the square root of the dose rate (dose rate)⫺1/2, from which it is concluded that in this reaction there is no character of chain reaction. The disulfide of N-acetylcysteine is the only sulfur final product. Hydrogen peroxide is not a reaction product, and accordingly the reaction of O2•⫺ with RSH does not proceed via hydrogen atom abstraction from RSH. A reaction mechanism is proposed, and an overall rate constant of 68 M⫺1 s⫺1 has been estimated. © 2000 Elsevier Science Inc. Keywords—Radiolysis, Superoxide radical, N-Acetylcysteine, Free radicals

INTRODUCTION

of their reactions with the superoxide radicals. The reaction between superoxide radicals and thiols was initially investigated by Barton et al. in 1970 [10] and then by Al-Thannon et al. in 1974 [11]. Several investigations of the reaction of superoxide radicals with cysteine and glutathione have since been carried out [12–15], but there is still considerable difference between the rate constants reported. For example, for cysteine or glutathione, rate constants of ⬍15 or 22 M⫺1 s⫺1 have been given, respectively [12,13], whereas, for the same thiols, rate constants that are larger by several orders of magnitude have been reported [10,14 –15]. Moreover, in some studies, the low H-abstractive power of O2•⫺ radicals has been suggested. In many thiol/O2 systems using a radiolysis technique the chain reaction process has also been reported [11,13–14,16]. We were interested in the antioxidant effect of N-acetylcysteine, and we have already investigated its reaction with H2O2 [17,18] and with OH radicals [19]. In this paper we have studied the reaction of RSH with O2•⫺ using a steady state radiolysis technique. It will be seen that O2•⫺ does indeed react with RSH, but with a rate constant that is several orders of magnitude lower than those reported for glutathione [14,15] and in the same order of magnitude as the value obtained for glutathione [13], and also for dithiothreitol [20,21] and cysteine [12]. It will also be shown that its reaction with RSH is not

There is a lot of evidence indicating that many physiological and pathophysiological phenomena such as ageing, carcinogenesis, drug toxicity, inflammation, viral infections, myocardial infarction, neurodegenerative disease, and other diseases may be developed through the action of reactive oxygen species [1,2]. Some strategies have been developed to enhance antioxidant capacity, including reactive oxygen species scavenging and the increasing of intracellular antioxidative defense [3,4]. N-Acetylcysteine (RSH) is known to be a drug that counteracts oxidative stress and replenishes glutathione. This drug has been shown to have antioxidative activity in vitro [5] as well as in vivo [6 – 8]. Regarding the reactivity of N-acetylcysteine with O2•⫺, there is some doubt concerning this reaction. Some authors believe that O2•⫺ radicals do not react with RSH [9], while others suggest a rate constant of ⬍ 103 M⫺1 s⫺1 for its reaction [5]. Concerning the reactivity of O2•⫺ toward thiols in general, there is considerable uncertainty as to the nature Address correspondence to: Zohreh Abedinzadeh, Laboratoire de Chimie-Physique, UMR 8601 CNRS, Universite´ Paris V, 45 rue des Saints-Pe`res, 75270 Paris cedex 06 France; Tel: ⫹(33)1-42-86-22-65; Fax: ⫹(33)1-42-86-83-87; E-Mail: [email protected]. 775

776

M. BENRAHMOUNE et al.

achieved by hydrogen atom abstraction from thiol and that H2O2 is not involved in this oxidation. Moreover, in this reaction there is no character of chain reaction, and N-acetylcystine (RSSR) is the only sulfur end-product. MATERIALS AND METHODS

All chemicals were used without further purification. N-acetylcysteine (ultrapur), 5-5⬘-dithiobis-(2-nitrobenzoic acid), catalase, and superoxide dismutase were obtained from Sigma Chemical Co. (St. Louis, MO, USA). Sodium acid phosphate and sodium formate (ultrapur) were supplied by Prolabo. N-Acetycystine and the oxidized form of N-acetylcysteine (RSSR) were provided by Bachem and O2 (⬎99.9%) by CFPO (France). All solutions were prepared with triple-distilled water.

niques were used to detect the eventual formation of other sulfur-containing products (i.e., RSO2H). The concentration of RSH was determined by Ellman’s reagent (5-5⬘-dithiobis-(2-nitrobenzoic acid) [23]), which undergoes a thiolate disulfide exchange reaction to yield 2-nitromercaptobenzoic acid. This change was measured spectrophotometrically at 412 nm, with ⑀(412) ⫽ 13,600 M⫺1 cm⫺1. Hydrogen peroxide was measured spectrophotometrically at 410 nm in the form of its titanium sulfate complex [24], with ⑀(410) ⫽ 700 M⫺1 cm⫺1. G values of RSH consumption and G values of the oxidized form (RSSR) and H2O2 were obtained from the slopes of concentrations vs. dose plots (cf. Figs. 1 and 2). Superoxide dismutase and catalase, when used, were present in concentrations of 20 and 40 mg/liter, respectively.

␥-Radiolysis experiment ␥-Irradiations were performed with an IBL irradiator composed of 137Cs sources (6000 Ci) at a dose rate of between 1.3 and 22 Gy min⫺1. The dosimetry was performed by the Fricke method [22]. Irradiations were carried out in tubes that were cleaned in hot concentrated acid then rinsed thoroughly. All vessels used for irradiation were heated at 400°C for 4 h after washing. The solution was prepared just before irradiation and buffered at pH 7 in a 10⫺2 M phosphate medium and then divided. Part of it was irradiated with a 137Cs ␥ source at the desired dose rate, and the rest was kept as the reference. Immediately after irradiation, both the irradiated solution and the reference were acidified to block H2O2induced and autoxidation reactions. The oxidized form of N-acetylcysteine (RSSR) was determined by ion-pairing reverse phase high-performance liquid chromatography (HPLC) using a coulometricelectrochemical detector. Irradiated samples (50 ␮l) were monitored after separation on a C18 (150 ⫻ 4.6 mm) Nucleosil 5 ␮m column. Isocratic elution was performed employing the mobile phase: 10 mM sodium phosphate monobasic monohydrate adjusted to pH 2.7 with 85% phosphoric acid containing 0.05 mM of the ion-pairing reagent octyl sulfate and 2% acetonitrile (v/v). Separations were performed at room temperature at a flow rate of 1.3 ml min⫺1. RSSR was detected following HPLC with a coulometricelectrochemical detector (model 5100A; Environmental Sciences Associates, Bedford, MA, USA) equipped with a model 5010 dual analytical cell and a model 5020 guard cell. The applied electrode potentials for detector 1, detector 2, and guard cell working electrodes were set at ⫹0.4, ⫹1.3, and ⫹1.4 V, respectively. RSSR standards were prepared to produce a standard curve. HPLC, Raman, and infrared (IR) spectroscopy tech-

RESULTS AND DISCUSSION

The O2•⫺ radical is in equilibrium with its conjugated acid, HO2• (equilibrium (1), pKa ⫽ 4.8 [25]). Its disproportionation with HO2• is faster (reaction 2) than its self-disproportionation (reaction 3) [25]: HO 2 • ^ O 2 •⫺ ⫹ H ⫹

(1)

O 2 •⫺ ⫹ HO 2 • ⫹ H ⫹ 3 H 2 O 2 ⫹ O 2

(2) ⫺1 ⫺1

k 2 ⫽ 9.7 ⫻ 10 M s . 7

O 2 •⫺ ⫹ O 2 •⫺ ⫹ 2H ⫹ 3 H 2 O 2

k 3 ⬍ 0.3 M ⫺1 s ⫺1 (3)

Superoxyde radicals were produced by ␥-radiolysis. When the ionizing radiation interacts with water, OH radicals, solvated electrons, and H atoms are formed by reaction (4). H 2 O 3 • OH,eaq⫺, H•, H⫹, H2O2, H2

(4)

In O2-saturated solutions (the solubility of molecular oxygen [O2] ⫽ 1 mM) containing 0.1 M sodium formate, • OH and H• radicals can be transformed into CO2•⫺ radicals and then into O2•⫺ radicals (reactions 5, 6, and 7). Solvated electrons give rise directly to HO2•/O2•⫺ radicals (reaction 8): OH ⫹ HCOO ⫺ 3 H 2 O ⫹ CO 2 •⫺

(5)

H • ⫹ HCOO ⫺ 3 H 2 ⫹ ⫹ CO 2 •⫺

(6)

CO 2 •⫺ ⫹ O 2 3 CO 2 ⫹ O 2 •⫺

(7)

eaq ⫺ ⫹ O 2 3 O 2 •⫺

(8)

•

O2•⫺ oxidation of N-acetycysteine

777

Fig. 1. Consumed RSH concentration in oxygenated aqueous solution as a function of the radiation dose for different initial RSH concentrations. Phosphate buffer (0.01 M), sodium formate (0.1 M), pH 7, and I ⫽ 1.3 Gy.min⫺1.

Regarding the reactivity of •OH, •H, eaq⫺, and CO2•⫺ radicals toward thiols, at sufficiently low RSH concentrations (⬍0.8 mM) such as have been used in the majority of these experiments, these radicals are essentially converted into O2•⫺ radicals via reactions 5 through 8 with G(O2•⫺) ⫽ 6.2 ⫻ 10⫺7 mol J⫺1. Consequently, in high RSH concentration there is a competition between the formation of O2•⫺ radicals and the reaction of •OH and eaq⫺ radicals with RSH. We have calculated, for instance, for RSH concentration ranging from 0.1 to 0.8 mM, that the rate of O2•⫺ radical formation change is between 100% and 85%. For higher concentrations of RSH, the rate of O2•⫺ radical formation decreases. The G values of RSH consumption, which are obtained from the slopes of concentrations vs. dose plots (Fig. 1), are reported in Fig. 3. As can be seen in this

Fig. 2. Hydrogen peroxide formation in oxygenated aqueous solution as a function of the radiation dose. Phosphate buffer (0.01 M), sodium formate (0.1 M), pH 7, and I ⫽ 1.3 Gy.min⫺1.

figure the initial G value of RSH consumption increases with the RSH concentration; then it reaches a pseudoplateau (G(⫺RSH) ⫽ 5.2 ⫻ 10⫺7 mol J⫺1) in the range of concentration from 0.7 to 1 mM. The G value increases again with RSH initial concentration up to (14.3 ⫾ 0.4) ⫻ 10⫺7 mol J⫺1 for initial concentration of [RSH] ⫽ 2.5 mM. At the first pseudo-plateau, at which O2•⫺ radicals are almost (more then 85%) the only radicals reacting with RSH, the G value of RSH consumption is in the same order of the G value of O2•⫺ formation (85% of G•OH ⫹ Geaq⫺ ⫹ GH• ⫽ 5.3 ⫻ 10⫺7mol J⫺1). This G value is consistent with our experimental data obtained in the first pseudo-plateau, whereas, in the second plateau (G(⫺RSH) ⫽ (14.3 ⫾ 0.4) ⫻ 10⫺7 mol J⫺1), the G value

Fig. 3. Dilution curve: evolution of the radiolytical yields (G values) of RSH consumption in oxygenated aqueous solutions as a function of RSH concentrations. Phosphate buffer (0.01 M), sodium formate (0.1 M), pH 7, and I ⫽ 1.3 Gy.min⫺1.

778

M. BENRAHMOUNE et al.

4.37 g.ml⫺1, O2: 0.049 g.ml⫺1)[27]. In irradiated solutions HS⫺ can be assessed by Ellman’s reagent in the same way as thiols [28]; this is why we do not observe, in the dilution curve (Fig. 3), the G value of the consumption of RSH due to the action of eaq⫺ at the first plateau (G(⫺RSH) ⫽ 5.3 ⫻ 10⫺7 mol J⫺1) instead of G(⫺RSH) ⫽ 6.2 ⫻ 10⫺7 mol J⫺1. R• radicals, in our experimental conditions, may react with O2 ([O2] ⫽ 1 mM) giving peroxyl radicals, rather than with RSH ([RSH] ⱕ 0.8 mM): R • ⫹ O 2 ^ ROO •

Fig. 4. G values of RSH consumption in ␥-irradiated RSH solution as a function of the inverse of the square root of dose rate, (dose rate)⫺1/2. Phosphate buffer (0.01 M), sodium formate (0.1 M), pH 7.

of RSH consumption is almost two and one-half times greater than the G value of radiolytic yield of O2•⫺ radicals. This high G value can be explained by a chain reaction. As shown in Fig. 4, the yield of RSH consumption is dependent on (dose-rate)⫺1/2 but only in high RSH concentration ([RSH] ⫽ 2 mM), whereas in low concentrations of RSH ([RSH] ⫽ 0.8 mM), the dependence of RSH consumption on (dose-rate)⫺1/2 is nonlinear. This means that there is a chain reaction, which obviously does not depend on the presence of O2•⫺, because SOD, which rapidly dismutates the O2•⫺ radicals, acts as a partial inhibitor. At the high thiol concentrations ([RSH] ⬎ 1 mM), reactions 5, 6, and 8 would come into competition with the reactions of •OH, H•, and eaq⫺ with RSH. Consequently, the impact of these latter reactions would be the decrease in the yield of O2•⫺ radical formation. Thus, we selected study conditions to minimize the reaction of •OH, H•, and eaq⫺ with RSH. On the other hand, we have observed that in the low concentrations of RSH there is no character of chain reaction. In keeping with these observations, the majority of the reported experiments were performed in the range of 0.1 to 0.8 mM of RSH concentration. As we have seen above, according to the competition ratio, O2•⫺ radicals are predominant (100% to 85%) for RSH concentrations ranging from 0.1 to 0.8 mM. In this range of RSH concentration, the reaction of eaq⫺ radicals may be considered to represent from 1% to 15% of the total G value of radical production. Solvated electrons react with RSH according to reaction 9 [26], eaq ⫺ ⫹ RSH 3 HS ⫺ ⫹ R • k ⫽ 5.6 ⫻ 10 9 M ⫺1 s ⫺1 (9) where HS⫺, the conjugated base of H2S, has a large solubility compared to molecular oxygen (H2S:

(10)

The formation of thiyl radicals by the rather slow reaction of unsubstituted peroxyl radicals (ROO•) with protonated thiols has not yet been unambiguously demonstrated [29,30]. In addition to the possible hydrogen transfer (reaction 11), peroxyl radicals may directly transfer an oxygen atom to yield alcohols and sulfinyl radicals according to: ROO • ⫹ RSH 3 ROOH ⫹ RS •

(11)

ROO • ⫹ RSH 3 RSO • ⫹ ROH

(12)

Product studies No sulfur-containing product other than the oxidized form of N-acetylcysteine, RSSR, has been detected, and we have observed systematically that G(RSHconsumption) ⫽ 2G(RSSR) (Fig. 5). However, it should be mentioned that the other products (i.e., RSO2H), if present, are in minor quantities that are not detectable by HPLC technique (signal-to-noise ratio:180 ng). Raman scattering and IR experiments were performed to detect the possible S—O bond (as in RSO2H) in irradiated samples ([RSH] ⫽ 0.8 mM). These experiments showed no characteristic band. Chain reactions in the radiolytic oxidation of thiols in the presence of oxygen are apparently common [13,16, 20,21]. To evaluate the reaction mechanism, and to investigate whether the O2•⫺ radical abstracts a hydrogen atom from RSH, the presence of hydrogen peroxide and its level were determined. H2O2 may have three origins; it could be a product of the reaction mechanism, a product of disproportionation of O2•⫺ radicals (reaction 3), or it may come from the molecular product of water radiolysis (reaction 4). This will give very important insight into the reaction mechanism.

Reaction of H2O2 with RSH In our pervious work [17,18] we showed that H2O2 oxidizes N-acetylcysteine, giving rise to the disulfide of

O2•⫺ oxidation of N-acetycysteine

779

Fig. 5. Dilution curve: evolution of the radiolytical yields (G values) of RSH consumption and disulfide production (2 ⫻ G values) in oxygenated aqueous solutions as a function of RSH concentrations. Phosphate buffer (0.01 M), sodium formate (0.1 M), pH 7, and I ⫽ 1.3 Gy.min⫺1.

N-acetylcysteine (RSSR). RSH reacts with H2O2 by forming a complex between RSH and H2O2 (reaction 13), the stoichiometry of the complex being RSH concentration-dependent. The complexation of these two species has been confirmed by infrared and Raman spectroscopy techniques [31]. The corresponding complex is then converted into the disulfide (reaction 14). (13) 2 RSH ⫹ H 2 O 2 ^ 关RSH. . .H 2 O 2 . . .RSH兴 k13 ⫽ 90 ⫾ 20 mol ⫺2 1 2 s ⫺1 k⫺13 ⫽ 0.9 ⫾ 0.1 s ⫺1 K ⫽ 110 ⫾ 20 mol ⫺2 1 2 关RSH. . .H 2 O 2 . . .RSH兴 3 RSSR ⫹ 2H 2 O k14 ⫽ 0.03 ⫾ 0.01 s ⫺1

(14)

Thus, at a pH of 7 and an RSH concentration of 0.8 mM, conditions under which the majority of the reported experiments have been carried out, the half-life of the reaction of hydrogen peroxide with RSH is 12 h. This is slow in the context of post-irradiation product analysis, and accordingly, the reaction must not be taken into account when assessing the radiolytic yield of oxidized form of RSH. To check a quantitative determination of H2O2, an O2-saturated thiol free formate solution was irradiated at pH 7. In this condition, solvated electrons give rise to O2•⫺ radicals (reaction 8), and OH and H radicals from the radiolysis of water react with formate ions yielding CO2•⫺ radicals. Then CO2•⫺ reacts rapidly with oxygen and forms carbon dioxide and O2•⫺ radicals with a total G value of 6.2 ⫻ 10⫺7 mol J⫺1. So, in this condition (thiol free solution) superoxide radicals disproportionate

according to reaction 2, giving H2O2. The observed value of G(H2O2) ⫽ 3.7 ⫻ 10⫺7 mol J⫺1 (Fig. 2) is in agreement with the expected value. It includes the molecular yield of H2O2 of about 0.7 ⫻ 10⫺7 mol J⫺1 from reaction 4 and the product of disproportionation of O2•⫺ radicals. It is hence concluded that H2O2 can be adequately determined at this condition.

Formation of RSSR and H2O2 after radiolysis To properly determine the radiolytic yield of H2O2 and RSSR, we stopped the reaction of H2O2 with RSH at the end of irradiation by adding acid. In these conditions, after a dose of 111 Gy (dose rate 1.3 Gy min⫺1, [RSH]0 ⫽ 0.8 mM), 9 ␮M H2O2 was found, which corresponds to an apparent G value of (0.8 ⫾ 0.2) ⫻ 10⫺7 mol J⫺1 (results are means and ranges from 8 –10 experiments). This value is in agreement with our expectation; it comes from the so-called molecular yield of water radiolysis (reaction 4). The effect of H2O2 on RSH is relatively minor compared to the radiolytic effect. In different experiments, the G(⫺RSH) of the irradiated solutions containing different concentrations of RSH with catalase or without catalase were examined. The G(⫺RSH) values with catalase were moderately below those without (figure not shown). The concomitant yield G(RSSR) is systematically half the yield of consumption of RSH G(⫺RSH) according to the relation: G(⫺RSH) ⫽ 2G(RSSR). It is, therefore, concluded that O2•⫺ radicals oxidize stoichiometrically N-acetylcysteine in disulfide.

M. BENRAHMOUNE et al.

780

Kinetic scheme Because disulfide is the only oxidizing product and G(H2O2) comes only from molecular yield (reaction 4) and because the yield of H2O2 formation is considerably lower than the yield of RSSR, we can conclude that H abstraction reaction by O2•⫺ radicals from an S-H function of RSH (reaction 15) is not supported by our data. H⫹ O 2 •⫺ ⫹ RSH O ¡ H 2 O 2 ⫹ RS •

(15)

The following mechanism is proposed to explain the experimental results: RSH ⫹ O 2 •⫺ ^RS共OOH兲 •⫺

(16)

RS共OOH兲 •⫺ 3 RSO • ⫹ OH ⫺

(17)

H⫹ ¡ RSOH ⫹ O 2 RSO • ⫹ O 2 • ⫺ O

(18)

RSOH ⫹ RSH 3 RSSR ⫹ H 2 O

(19)

HO 2 • ⫹ O 2 •⫺ ⫹ H ⫹ 3 H 2 O 2 ⫹ O 2

(2)

This mechanism is based on the formation of a wellknown three-electron-bonded [32–35] radical intermediate, which breaks down into the sulfinyl radical (RSO•) and OH⫺ ions (reactions 16 and 17). These reactions have been suggested by Zhang et al. [20] for dithiothreitol, although no intermediates have been identified directly. Reaction 16 is in competition with reaction 2. RSO• is rather a moderate oxidizing radical, and an electron transfer reaction is more probable than an H abstraction reaction (reaction 20) from a thiol [20].

Fig. 6. Variation of the initial yields of consumption of RSH as a function of RSH. The curve is the best fit to experimental points using the equation given in the text.

reaction has been proposed by Zhang et al. [44] in the case of 2-mercaptoethanol. The disulfide results from reaction 19. We may attempt an estimation of an effective rate constant of the reaction of O2•⫺ with RSH. Leaving aside reaction-16, the initial consumption yield (G(⫺RSH)) is thus calculated using reactions 16 –19 and 2, and also the free radical steady state approximation:

G(⫺RSH) ⫽

2 k 16 共RSH兲 2 Ik2

冋

⫻ ⫺1 ⫹

冑

1⫹

2k2 G共O 2 •⫺ 兲I 2 k16 (RSH)2

册

(I)

This kind of behavior is exhibited by other moderately oxidizing radicals [36 –38]. Moreover, reaction 20 leads to the formation of a thiyl radical, which can be oxidized to a thiylperoxyl radical in the presence of oxygen (1 mM in our experimental condition)[39 – 43] and then to RSO2H according to reactions 21 and 22.

where I is the dose rate (1.3 Gy.min⫺1). The experimental points of Fig. 6 were fitted to this expression by a nonlinear regression procedure taking k16 as the adjustable parameter (Kaleidagraph software). The rate constant 2k2 was equal to 8.5 ⫻ 107 M ⫺1 s⫺1 [45]. Using this value, we calculated for the self-dismutation of the superoxide radical a global rate constant of 5.36 ⫻ 105 M⫺1 s⫺1 (2k2 ⫻ 10pK ⫺ pH) at pH 7. The fit given by the computed curve (Fig. 6) is obtained for:

RS • ⫹ O 2 ^ RSOO •

k16 ⫽ 68⫾6 M⫺1s⫺1

RSO • ⫹ RSH 3 RSOH ⫹ RS •

RSOO ⫹ RSH 3 RSO 2 H ⫹ RS •

(20)

(21) •

(II)

(22)

If this is the case we should find RSO2H in our endproduct analysis. This is not the case. From this we conclude that reaction 20 is of minor importance. An electron transfer from superoxide radicals to RSO• radicals seems plausible, e.g., reaction 18. This kind of

The very good agreement between the experimental points and the fit justifies the proposed mechanism. Indeed, the rate constant of O2•⫺ with RSH is relatively close to the value for dithiothreitol (35 M⫺1 s⫺1) [20], glutathione (22 M⫺1s⫺1) [13], and cysteine (⬍15 M⫺1 s⫺1) [12], and the oxidation mechanism is simpler. RSH

O2•⫺ oxidation of N-acetycysteine

seems unable to generate a chain reaction with O2•⫺ unlike dithiothreitol [20]. Recently, Dikalov et al. [46] queried a number of earlier rate constant values on the basis of assays used. They reported the rate constants of 1–5 ⫻ 105 M⫺1 s⫺1 for the reaction of superoxide radicals with several thiols using electron paramagnetic resonance. With these constants, the reaction of O2•⫺ with cellular glutathione (concentration as high as 10⫺2 M) would be its main biological sink and much more important than its reaction with SOD. We consider that rate constants of the order of 105 M⫺1 s⫺1 for the superoxide reactions are likely to be too high. More recently, Winterbourn et al. [47] compared the reactivity of O2•⫺ radicals (generated from xantine oxidase and hypoxanthine) and hydrogen peroxide with different thiols. They reported that the relative reactivities of the different thiols with both oxidants were inversely related to the pK of the thiol group, in such a way that at pH 7.4, N-acetylcysteine (pK ⫽ 9.5) was weakly reactive (without giving the precise rate constant value). The rate constants were estimated to be in the 30 –1000 M⫺1 s⫺1 range. Although, according to Eqn I, the G value of RSH consumption is dose-rate dependent, it should be noted that this kind of behavior is observed whenever a first order radical reaction is in competition with a second order reaction with the same radical such as reactions 24 and 25 [48,49]: R • ⫹ subst. 3 product

(24)

R• ⫹ R• 3 R2

(25)

Moreover, it is shown that RSH reacts with the superoxide radical, but it cannot be regarded as a very efficient superoxide scavenger. If we consider that chemically N-acetylcysteine is similar to glutathione, the rate constant of reaction of both molecules with superoxide may be of the same order of magnitude. The cellular glutathione concentration may reach a value as high as 0.01 M [50], but the reaction of superoxide with glutathione would be much less significant than its reaction with SOD. CONCLUSION

There has been considerable uncertainty about the nature of the reaction of thiols with superoxide radicals using the radiolysis technique, especially concerning: (i) the chain reaction; (ii) the hydrogen atom abstraction from thiol; (iii) the nature of the end-product(s) of this reaction. In the present work, by studying the radiationinduced oxidation of RSH for different initial concentra-

781

tions (0.1–2.5 mM) and the effect of the inverse of the square root of the dose rate, (dose rate)⫺1/2, we have shown that the reaction of O2•⫺ radicals with N-acetylcysteine does not proceed via a chain reaction. The reaction rate constants of O2•⫺ with N-acetylcysteine was found to be 68 ⫾ 6 M⫺1 s⫺1. Hydrogen peroxide is not a reaction product, and we have concluded that the reaction of O2•⫺ with RSH does not proceed by hydrogen transfer and the formation of a thiyl radical. The latter could, in the presence of O2, give rise to the production of to RSOO• and then to RSO2H in the presence of RSH. This is not in agreement with our data. Moreover, by HPLC and electrochemical detection, IR, and Raman spectroscopy techniques, we have shown that the Nacetylcystine is the only sulfur end-product. However, it should be mentioned that the other products, if present, are in minor quantities that are not detectable by our techniques. Acknowledgements — The authors cordially thank Dr. D. Averbeck and Dr. E. Moustacchi from the Institut Curie of Paris for offering them access to ␥-irradiation facilities. They also thank Dr. M. Picquart for IR and Raman spectroscopic investigations.

REFERENCES [1] Halliwell, B.; Gutteridge, J. M. Free radicals in biology and medicine (2nd ed.). Oxford: Clarendon Press; 1989. [2] Davies, K. J. A., ed. Oxidative damage and repair, chemical, biological and medical aspects. Oxford: Pergamon Press; 1991. [3] Meister, A. Strategies for increasing cellular glutathione. In: Packer, L.; Cadenas, E., eds. Biothiols in health and disease. New York: Marcel Dekker; 1995:165–188. [4] Bast, A.; Haenen, G. R. M. M. ␣-Lipoic acid and N-acetylcysteine: new concepts for old drugs. In: Packer, L.; Cadenas, E., eds. Biothiols in health and disease. New York: Marcel Dekker; 1995:409 – 426. [5] Aruoma, O. I.; Halliwell, B.; Hoey, B. M.; Butler, J. The antioxidant action of N-acetylcysteine: its reaction with hydrogen peroxide, hydroxyl radical, superoxide and hypochlorous acid. Free Radic. Biol. Med. 6:593–597; 1989. [6] Molde´us, P.; Cotgreave, I. A.; Berggren, M. Lung protection by a thiol-containing antioxidant, N-acetylcysteine. Respiration 50(Suppl. 1):31– 42; 1986. [7] Staal, F. J. T.; Ela, S. W.; Roederer, M.; Anderson, M. T.; Herzenberg, L. A.; Herzenberg, L. A. Glutathione deficiency and human immunodeficiency virus infection. Lancet 339:909 –912; 1992. [8] Look, M. P.; Rockstroh, J. K.; Rao, G. S.; Barton, S.; Lemock, H.; Kaiser, R.; Kupfer, B.; Sudhop, T.; Spengler, U.; Sauerbruch, T. Sodium selenite and N-acetylcysteine in antiretroviral-naive HIVI-infected patients: a randomized, controlled pilot study. Eur. J. Clin. Investig. 28:389 –397; 1998. [9] Felix, K.; Pairet, M.; Zimmermann, R. The antioxidative activity of the mucoregulatory agents: ambroxol, bromhexine and Nacetyl-L-cysteine. A pulse radiolysis study. Life Sci. 59:1141– 1147; 1996. [10] Barton, J. P.; Packer, J. E. The radiolysis of oxygenated cysteine solutions at neutral pH. The role of RSSR⫺ and O2⫺. Int. J. Radiat. Phys. Chem. 2:159 –166; 1970. [11] Al-Thannon, A. A.; Barton, J. P.; Packer, J. E.; Sims, R. J.; Trumbore, C. N.; Winchester, R. V. The radiolysis of aqueous solutions of cysteine in the presence of oxygen. Int. J. Radiat. Phys. Chem. 6:233–248; 1974.

782

M. BENRAHMOUNE et al.

[12] Bielski, B. H. J.; Shive, G. G. Reaction rates of superoxide radicals with the essential amino acids. In: Oxygen free radicals and tissue damage. CIBA Found. Symp. 65:43–56; 1979. [13] Sekaki, A. Etude par radiolyse de la reactivite´ du radical superoxide vis-a`-vis du glutathion. The`se de Doctorat, Universite´ ParisSud; 1984. [14] Asada, K.; Kanematsu, S. Reactivity of thiols with superoxide radical. Agric. Biol. Chem. 40:1891–1892; 1976. [15] Winterbourn, C. C.; Metodiewa, D. The reaction of superoxide with reduced glutathione. Arch. Biochem. Biophy. 314:284 –290; 1994. [16] Lal, M. Radiation induced oxidation of sulphydryl molecules in aqueous solutions. A comprehensive review. Radiat. Phys. Chem. 43:595– 611; 1994. [17] Arroub, J.; Berge`s, J.; Abedinzadeh, Z.; Langlet, J.; Garde`sAlbert, M. On N-acetylcysteine. Part I. Experimental and theoretical approaches of the N-acetylcysteine/H2O2 complexation. Can. J. Chem. 72:2093–2101; 1994. [18] Abedinzadeh, Z.; Arroub, J.; Garde`s-Albert, M. On N-acetylcysteine. Part II. Oxidation of N-acetylcysteine by hydrogen peroxide: kinetic study of the overall process. Can. J. Chem. 72:2102– 2107; 1994. [19] Charif, A.; Abedinzadeh, Z.; Garde`s-Albert, M.; Ferradini, C. Action des radicaux •OH ou N3• sur la N-acetylcysteine. J. Chim. Phys. 90:907–916; 1993. [20] Zhang, N.; Schchmann, H. P.; von Sonntag, C. The reaction of superoxide radical anion with dithiothreitol: a chain process. J. Phys. Chem. 95:4718 – 4722; 1991. [21] Lal, M.; Rao, R.; Fang, X.; Schchmann, H. P.; von Sonntag, C. Radical-induced oxidation of dithiothreitol in acidic oxygenated aqueous solution: a chain reaction. J. Am. Chem. Soc. 119:5735– 5739; 1997. [22] Frick, H.; Hart, E. J. In: Attix, F. H.; Roesch, W. C., eds. Radiation dosimetry. New York: Academic Press; 1966:167–239. [23] Ellman, G. L. A colorimetric method for determining low concentrations of mercaptans. Arch. Biochem. Biophys. 82:70 –77; 1959. [24] Eisenberg, G. M. Indian Eng. Chem. Anal. Ed. 15:327–328; 1943. [25] Bielski, B. H. J.; Cabelli, D. E.; Arudi, R. L. Reactivity of HO2/O2⫺ radicals in aqueous solution. J. Phys. Chem. Ref. Data 14:1041–1100; 1985. [26] Hoffman, M. Z.; Hayon, E. Pluse radiolysis study of sulfhydryl compounds in aqueous solution. J. Phys. Chem. 77:990 –996; 1973. [27] Weast, R. C., ed. Handbook of chemistry and physics. Boca Raton, FL: CRC Press; 1980. [28] Akhlaq, M. S.; von Sonntag, C. Free radical-induced elimination of H2S from dithiothreitol: a chain reaction. J. Am. Chem. 108: 3542–3544; 1986. ¨ nal, A. Lifetime of peroxyl [29] Schulte-Frohlinde, D.; Behrens, G.; O radicals of poly(U), poly(A) and single and double-stranded DNA and their rate of reaction with thiols. Int. J. Radiat. Biol. 50:103– 110; 1986. [30] Swarts, S. G.; Becker, D.; DeBolt, S.; Sevilla, M. D. An electron spin resonance investigation of the structure and formation of sulfinyl radicals: reaction of peroxyl radicals with thiols. J. Phys. Chem. 93:155–161; 1989. [31] Picquart, M.; Abedinzadeh, Z.; Grajcar, L.; Baron, M. H. Spectroscopic study of N-acetylcysteine and N-acetylcystine/hydrogen peroxide complexation. Chem. Phys. 228:279 –291; 1998. [32] Asmus, K. D. Stabilization of oxidized sulfur centers in organic sulfides. Radical cations and odd-electron sulfur-sulfur bonds. Acc. Chem. Res. 12:436 – 442; 1979. [33] Go¨bl, M.; Bonifacic, M.; Asmus, K. D. Substituted effects on the stability of three-electron bonded radicals and radical ions from organic sulfur compounds. J. Am. Chem. Soc. 106:5984 –5988; 1984.

[34] von Sonntag, C. The chemical basis of radiation biology. London: Taylor and Francis; 1987. [35] Asmus, K. D.; Go¨bl, M.; Hiller, K. O.; Mahling, S.; Mo¨nig, J. S䡠䡠䡠N and S䡠䡠䡠O three-electron bonded radicals and radical cation in aqueous solution. J. Chem. Soc. Perkin Trans. 2:641– 646; 1985. [36] O’Neill, P. Pulse radiolytic study of the interaction of thiols and ascorbate with OH adducts of dGMP and dG: implication for DNA repair processes. Radiat. Res. 96:198 –210; 1983. [37] O’Neill, P.; Champan, P. W. Potential repair of free radical adducts of dGMP and dG by a series of reductants: a pulse radiolysis study. Int. J. Radiat. Biol. 47:71– 80; 1985. [38] Tamba, M.; O’Neill, P. Redox reactions of thiol free radicals with the antioxidants ascorbate and chlorpromazine: role in radioprotection. J. Chem. Soc. Perkin Trans. 2:1681–1685; 1991.. [39] Jayson, G. G.; Stirling, D. A.; Swallow, A. J. Pulse and Xradiolysis of 2-mercaptoethanol in aqueous solution. Int. J. Radiat. Biol. 19:143–156; 1971. [40] Tamba, M.; Simone, G.; Quintilliani, M. Interactions of thiyl free radicals with oxygen: a pulse radiolysis study. Int. J. Radiat. Biol. 50:595– 600; 1986. [41] Chatgilialoglu, Ch.; Guerra, M. Alkanethiylperoxl radicals. In: Chatgilialoglu, Ch.; Asmus, K.-D., eds. Sulfur-centered reactive intermediates in chemistry and biology. New York: Plenum Press; 1990:31–36. [42] Sevilla, M. D.; Becker, D.; Yan, M. Structure and reactivity of peroxyl and sulphoxyl radicals from measurement of oxygen-17 hyperfine coupling: relationship with Taft substituent parameters. J. Chem. Soc. Faraday Trans. 86:3279 –3286; 1990. [43] Sevilla, M. D.; Becker, D.; Yan, M. The formation and structure of the sulfoxyl radicals RSO•, RSOO•, RSO2•, and RSO2OO• from the reaction of cysteine, glutathione and penicillamine thiyl radicals with molecular oxygen. Int. J. Radiat. Biol. 57:65– 81; 1990. [44] Zhang, X.; Zhang, N.; Schuchmann, H. P.; von Sonntag, C. Pulse radiolysis of 2-mercaptoethanol in oxygenated aqueous solution. Generation and reactions of the thiylperoxyl radical. J. Phys. Chem. 98:6541– 6547; 1994. [45] Bielski, B. H. J.; Allen, A. O. Mechanism of the disproportionation of superoxide radicals. J. Phys. Chem. 81:1048 –1050; 1977. [46] Dikalov, S.; Khramtsov, V.; Zimmer, G. Determination of rate constants of the reactions of thiols with superoxide radical by electron paramagnetic resonance: critical remarks on spectrophotometric approaches. Arch. Biochem. Biophys. 326:207–218; 1996. [47] Winterbourn, C. C.; Metodiewa, D. Reactivity of biologically important thiol compounds with superoxide and hydrogen peroxide. Free Radic. Biol. Med. 27:322–328; 1999. [48] Burns, W. G.; Barker, R. Radiation chemistry, Vol. 3. New York: Pergamon Press; 1965. [49] Mankhetkorn, S.; Abedinzadeh, Z.; Houe´e-Levin, Ch. Antioxidant action of sodium diethylthiocarbamate: reaction with hydrogen peroxide and superoxide radical. Free Radic. Biol. Med. 17:517–527; 1994. [50] Reed, D. J. Glutathione: toxicological implications. Annu. Rev. Pharmacol. Toxicol. 30:603– 631; 1990. ABBREVIATIONS

RSH—N-acetylcysteine RSSR—N-acetylcystine O2•⫺—superoxide radical SOD—superoxide dismutase Ci—Curie Gy—Gray HPLC— high performance liquid chromatography