Interaction of tert-butyl hydroperoxide with hypochlorite induces peroxyl radicals. A chemiluminescence study

The interaction of hypochlorite (HOCl/OCl-) with tert-butyl hydroperoxide ((CH3)3COOH) was investigated by chemiluminescence. It was shown that the addition of HOCl/OCl- to (CH3)3COOH induces a fast chemiluminescent flash. The intensity of this flash increases with the increase in both HOCl/OCl- and (CH3)3COOH concentration. The chemiluminescence is quenched in a concentration-dependent manner in the presence of free radical spin traps N-tert-butyl nitrone and alpha-(4-pyridyl-1-oxyl)-N-tert-butyl nitrone. This fact proves that free radicals take part in the interaction of HOCl/OCl- and (CH3)3COOH. Hypochlorite yielded a very similar chemiluminescence spectrum in its reaction with (CH3)3COOH as Ce4+. It differed considerably from the spectrum in the system H2O2 and HOCl/OCl-. It is well known that the interaction of Ce4+ and (CH3)3COOH produces peroxyl radicals. These results confirm the hyothesis that the interaction of HOCl/OCl- and (CH3)3COOH is mediated by peroxyl radicals. Thus, organic hydroperoxides always present in unsaturated lipids can induce lipid peroxidation processes in the reaction with HOCl/OCl-.     

Quantitative 31P NMR detection of oxygen-centered and carbon-centered radical species

Quantitative 31P NMR spin trapping techniques can be used as effective tools for the detection and quantification of many free radical species. Free radicals react with a nitroxide phosphorus compound, 5-diisopropoxy-phosphoryl-5-methyl-1-pyrroline-N-oxide (DIPPMPO), to form stable radical adducts, which are suitably detected and accurately quantified using (31)P NMR in the presence of phosphorus containing internal standards. Initially, the 31P NMR signals for the radical adducts of oxygen-centered (*OH, O2*-) and carbon-centered (*CH3, *CH2OH, CH2*CH2OH) radicals were assigned. Subsequently, the quantitative reliability of the developed technique was demonstrated under a variety of experimental conditions. The 31P NMR chemical shifts for the hydroxyl and superoxide reaction adducts with DIPPMPO were found to be 25.3, 16.9, and 17.1 ppm (in phosphate buffer), respectively. The 31P NMR chemical shifts for *CH3, *CH2OH, *CH(OH)CH3, and *C(O)CH3 spin adducts were 23.1, 22.6, 27.3, and 30.2 ppm, respectively. Overall, this effort forms the foundations for a targeted understanding of the nature, identity, and mechanisms of radical activity in a variety of biomolecular processes.     

Psoralen-photosensitized formation of free lipid radicals at 77 degrees K

It is shown by ESR method that 8-methoxypsoralen (8-MOP) under low temperature UV-irradiation has no influence on free radical formation of saturated fatty acids (FA). Under the same conditions the quantum yield of free radical photogeneration for unsaturated FA (oleate and limolenate) after addition of 8-MOP increased 7.5-15 times. The ESR spectrum of free radicals generated after interaction of 8-MOP with polyunsaturated FA, may be presented as superposition of signals-CH-(CH = CH)2-, -CH2CHCH2- and -CH2CH2. The molecular mechanism of photosensitized reactions is discussed.     

Antioxidant properties of tea investigated by EPR spectroscopy

The antioxidant properties of green, black and mixed (fruit) tea samples of different origin were investigated by means of EPR spectroscopy. A six line EPR spectrum of solid tea samples indicates the presence of Mn(II) ions and it is superimposed with a sharp singlet line attributed to semiquinone radical species (Delta H(pp)=1 mT; g=2.0022). Antioxidant properties of aqueous tea extracts in H(2)O(2)/NaOH/dimethylsulfoxide system generating reactive radicals (*OH, O(2)*-), *CH(3)) were followed by spin trapping technique. In addition, antioxidant capacity of these samples was assessed using stable radicals 1,1-diphenyl-2-picrylhydrazyl (DPPH) and 4-hydroxy-2,2,6,6-tetramethyl-1-piperidinyloxy (TEMPOL). Typically, the highest antioxidant potential to terminate superoxide radicals was found in green teas, followed by black and fruity teas. The pro-oxidant activity of green teas evidenced by spin traps was promoted in samples with higher Mn(II) and ascorbic acid concentrations. Various sources of free radicals used in the antioxidant tests due to their specific action show different termination rates in the presence of the individual tea samples.     

Gamma radiolysis as a tool to study lipoprotein oxidation mechanisms

Well-defined quantities of *OH, O2*-,HO2* or RO2*)radicals (reactive oxygen species) can be specifically produced by radiolysis of water or ethanol. Such radical species can initiate one-electron oxidation or one-electron reduction reactions on numerous biological systems. The oxidative hypothesis of atherosclerosis classically admits the involvement of the oxidation of low density lipoproteins (LDLs) but also of high density lipoproteins (HDLs) in the development of the atherosclerotic process. The initiation mechanisms of this oxidation are still incompletely defined, although free radicals are likely involved. Therefore, gamma-radiolysis appears as a method of choice for the in vitro study of the mechanisms of oxidation of LDLs and HDLs by oxygen-centred free radicals (*OH, O2*-,HO2* and RO2*). Radiolytically oxidized lipoproteins exhibited a very well defined oxidation status (radiation dose-dependent quantification of vitamin E, beta-carotene, lipid peroxidation, protein carbonylation ...). gamma-Radiolysis is a less drastic method than other oxidation procedures such as for example copper ions. Moreover, gamma-radiolysis is also especially suitable for studying the reducing properties of antioxidant compounds with regard to their scavenging capacity.     

Free radicals in physiology and pathology.

Free radicals are molecules with odd number of electrons and a high instability. Free radicals, which can occur in both organic (i.e., quinones) and inorganic molecules (i.e., O2-), are very reactive and their reactions are critical for the normal activity of a wide spectrum of biologic processes. They are also produced in the catalytic action of a variety of cellular enzymes and electron transport processes and are implicated in a number of physiologic and pathologic processes. Organisms can be exposed to free radicals in many ways other than through the processes of normal metabolism. Irradiation of organisms with electromagnetic radiation generates primary radicals (e-aq, OH., and H.), which can then undergo secondary reactions with dissolved O2 or with cellular solutes. In addition, a wide variety of environmental agents (drugs capable of redox cycling, and xenobiotics that can form free radical metabolites) including the aging process cause free radical damage to cells. This review deals with the reactions they can undergo and discusses the free radicals related to toxicology.     

Radical production from free and peptide-bound methionine sulfoxide oxidation by peroxynitrite and hydrogen peroxide/iron(II)

Methionine sulfoxide is a post-translational protein modification that has been receiving increasing attention in the literature. Here we used electron paramagnetic resonance spin trapping techniques to show that free and peptide-bound methionine sulfoxide is oxidized by hydrogen peroxide/iron(II)-EDTA and peroxynitrite through the intermediacy of the hydroxyl radical to produce both *CH3 and *CH2CH2CH radicals. The results indicate that methionine sulfoxide residues are important targets of reactive oxygen- and nitrogen-derived species in proteins. Since the produced protein-derived radicals can propagate oxidative damage, the results add a new antioxidant route for the action of the enzyme peptide methionine sulfoxide reductase.     

A comparative spin trapping study of hypobromous and hypochlorous acids interaction with tert-butyl hydroperoxide

We have demonstrated that hypochlorite (HOCI/OCl-) and hypobromite (HOBr/OBr-) can react with tert-butyl hydroperoxide with close rate constants (k(HOCl) = 10,8 M(-1) x s(1); k(HOBr) = 8,9 M(-1) x (s(-1)). By means of the spin trap 4-pyridyl-1-oxide-N-tert-butyl nitron we have found that both reactions proceed through decomposition of tert-butyl hydroperoxide and generation of tert-butyl peroxyl (OOC(CH3)3) and tert-butoxyl (OC(CH3)3) radicals, the ratio of their the concentrations being dependent on the concentration of tert-butyl hydroperoxide. Thus, hypobromite, similar to hypochlorite, is a precursor of free radicals produced in the reaction with organic hydroperoxides. This reaction can be of great importance in the intensification of free radical processes, namely, in lipid peroxidation at the stage of chain branching.     

Radical-induced oxidation of metformin.

Metformin (1,1-dimethylbiguanide) is an antihyperglycaemic drug used to normalize glucose concentrations in type 2 diabetes. Furthermore, antioxidant benefits have been reported in diabetic patients treated with metformin. This work was aimed at studying the scavenging capacity of this drug against reactive oxygen species (ROS) like *OH and (O2*-)-free radicals. ROS were produced by gamma radiolysis of water. The irradiated solutions of metformin were analyzed by UV/visible absorption spectrophotometry. It has been shown that hydroxyl free radicals react with metformin in a concentration-dependent way. The maximum scavenging activity was obtained for concentrations of metformin > or = 200 micromol.L(-1), under our experimental conditions. An estimated value of 10(7) L.mol(-1).s(-1) has been determined for the second order rate constant k(*OH + metformin). Superoxide free radicals and hydrogen peroxide do not initiate any oxidation on metformin in our in vitro experiments.     

Study of Plasma Induced Chemistry by DC Discharges in CO<Subscript>2</Subscript>/N<Subscript>2</Subscript>/H<Subscript>2</Subscript>O Mixtures Above a Water Surface

The chemistry induced by atmospheric pressure DC discharges above a water surface in CO(2)/N(2)/H(2)O mixtures was investigated. The gaseous mixtures studied represent a model prebiotic atmosphere of the Earth. The most remarkable changes in the chemical composition of the treated gas were the decomposition of CO(2) and the production of CO. The concentration of CO increased logarithmically with the increasing input energy density and an increasing initial concentration of CO(2) in the gas. The highest achieved concentration of CO was 4.0 +/- 0.6 vol. %. The production of CO was crucial for the synthesis of organic species, since reactions of CO with some reactive species generated in the plasma, e. g. H* or N* radicals, were probably the starting point in this synthesis. The presence of organic species (including the tentative identification of some amino acids) was demonstrated by the analysis of solid and liquid samples by high-performance liquid chromatography, infrared absorption spectroscopy and proton-transfer-reaction mass spectrometry. Formation of organic species in a completely inorganic CO(2)/N(2)/H(2)O atmosphere is a significant finding for the theory of the origins of life.     

Elevation of hydrogen peroxide after spinal cord injury detected by using the Fenton reaction.

To reveal whether reactive oxygen species (ROS) play a role after spinal cord injury, we developed a unique method for assaying hydrogen peroxide (H2O2) and determined the time course of its concentration changes following impact injury to the rat spinal cord. Microdialysis was used to sample H2O2 in the extracellular space and the dialysates were collected into a vial containing salicylate and ferrous chloride (FeCl2). H2O2 collected in the vial was converted to hydroxyl radicals (*OH) by FeCl2 catalysis. 2,3- and 2,5-dihydroxybenzoic acid produced by reaction of *OH with salicylate in the collecting vial were measured by HPLC and calibrated to H2O2 concentrations. The postinjury levels of H2O2 were significantly increased (p = 0.02) for over 11 h. FeCl2 administered through the dialysis fiber catalyzes H2O2 conversion in the cord to *OH. This *OH does not reach the collecting vial due to its extremely short lifetime (nanoseconds). The reduced H2O2 levels in the vials validate the measurement of H2O2. The relatively long-lasting formation of H2O2 and superoxide reported herein and previously suggests that ROS may be important in secondary spinal cord damage and that removal of ROS may be a realistic treatment strategy for reducing injury caused by free radicals.     

Reactions of desferrioxamine with peroxynitrite-derived carbonate and nitrogen dioxide radicals

The iron chelating agent desferrioxamine inhibits peroxynitrite-mediated oxidations and attenuates nitric oxide and oxygen radical-dependent oxidative damage both in vitro and in vivo. The mechanism of protection is independent of iron chelation and has remained elusive over the past decade. Herein, stopped-flow studies revealed that desferrioxamine does not react directly with peroxynitrite. However, addition of peroxynitrite to desferrioxamine in both the absence and the presence of physiological concentrations of CO2 and under excess nitrite led to the formation of a one-electron oxidation product, the desferrioxamine nitroxide radical, consistent with desferrioxamine reacting with the peroxynitrite-derived species carbonate (CO3*-) and nitrogen dioxide (*NO2) radicals. Desferrioxamine inhibited peroxynitrite-dependent free radical-mediated processes, including tyrosine dimerization and nitration, oxyhemoglobin oxidation in the presence of CO2, and peroxynitrite plus carbonate-dependent chemiluminescence. The direct two-electron oxidation of glutathione by peroxynitrite was unaffected by desferrioxamine. The reactions of desferrioxamine with CO3*- and *NO2 were unambiguously confirmed by pulse radiolysis studies, which yielded second-order rate constants of 1.7 x 10(9) and 7.6 x 10(6) M(-1) s(-1), respectively. Desferrioxamine also reacts with tyrosyl radicals with k = 6.3 x 10(6) M(-1) s(-1). However, radical/radical combination reactions between tyrosyl radicals or of tyrosyl radical with *NO2 outcompete the reaction with desferrioxamine and computer-assisted simulations indicate that the inhibition of tyrosine oxidation can be fully explained by scavenging of the peroxynitrite-derived radicals. The results shown herein provide an alternative mechanism to account for some of the biochemical and pharmacological actions of desferrioxamine via reactions with CO3*- and *NO2 radicals.     

Generation of oxygen free radicals in thyroid cells and inhibition of thyroid peroxidase

We examined whether superoxide (O(2)(-)) is produced as a precursor of hydrogen peroxide (H(2)O(2)) in cultured thyroid cells using the cytochrome c method and the electron paramagnetic resonance (EPR) method. No O(2)(-) or its related radicals was detected in thyroid cells under the physiological condition. The presence of quinone, 2,3-dimethoxy-l-naphthoquinone (DMNQ), or 2-methyl-1, 4-naphthoquinone (menadione), in the medium produced O(2)(-) and hydroxyl radicals (OH*); the amount of H(2)O(2) generation was also increased. Incubation of follicles with DMNQ or menadione inhibited iodine organification (a step of thyroid hormone formation) and its catalytic enzyme, thyroid peroxidase (TPO). This inhibition should be caused by reactive oxygen species because the two quinones, particularly DMNQ, exert their effect through the generation of reactive oxygen species. It is speculated that the site-specific inactivation of TPO might have occurred at the heme-linked histidine residue of the TPO molecule, a critical amino acid for enzyme activity because OH* (vicious free radicals) can be formed at the iron-linked amino acid. TPO mRNA level and electrophoretic mobility of TPO were not inhibited by quinones. Our study suggests that thyroid H(2)O(2) is produced by divalent reduction of oxygen without O(2)(-) generation. If thyroid cells happen to be exposed to significant amount of reactive oxygen species, TPO and subsequent thyroid hormone formation are inhibited.     

ESR studies on reaction of saccharide with the free radicals generated from the xanthine oxidase/hypoxanthine system containing iron

The free radicals generated from the iron containing system of xanthine oxidase and hypoxanthine (Fe-XO/HX) were directly detected by using spin trapping. It was found that not only superoxide anion (O(2)*-) and hydroxyl radical (OH*), but also alkyl or alkoxyl radicals (R*) were formed when saccharides such as glucose, fructose and sucrose were added into the Fe-XO/HX system. The generated amount of R* was dependent on the kind and concentration of saccharides added into the Fe-XO/HX system and no R* were detected in the absence of saccharides, indicating that there is an interaction between the saccharide molecules and the free radicals generated from the Fe-XO/HX system and saccharide molecules are essential for generating R* in the Fe-XO/HX system. It is expected that the toxicity of R* would be greater than of hydrophilic O(2)*- and OH* because they are liposoluble and their lives are longer and the active sites of biomolecules are closely related with lipophilic phase, thus they can damage cells more seriously than O(2)*- and OH*. The R* generated from the saccharide containing Fe-XO/HX can be effectively scavenged by selenium containing abzyme (Se-abzyme), indicating Se-abzyme is a promising antioxidant.     

Proton-motive-force-driven formation of CO from CO2 and H2 in methanogenic bacteria

Cell suspensions of methanogenic bacteria (Methanosarcina barkeri, Methanospirillum hungatei, Methano-brevibacter arboriphilus, and Methanobacterium thermoautotrophicum) were found to form CO from CO2 and H2 according to the reaction: CO2 + H2----CO + H2O; delta G0 = +20 kJ/mol. Up to 15,000 ppm CO in the gas phase were reached which is significantly higher than the equilibrium concentration calculated from delta G0 (95 ppm under the experimental conditions). This indicated that CO2 reduction with H2 to CO is energy-driven and indeed the cells only generated CO when forming CH4. The coupling of the two reactions was studied in more detail with acetate-grown cells of M. barkeri using methanogenic substrates. The effects of the protonophore tetrachlorosalicylanilide (TCS) and of the proton-translocating ATPase inhibitor N,N'-dicyclohexylcarbodiimide (cHxN)2C were determined. TCS completely inhibited CO formation from CO2 and H2 without affecting methanogenesis from CH3OH and H2. In the presence of the protonophore the proton motive force delta p and the intracellular ATP concentration were very low. (cHxN)2C, which partially inhibited methanogenesis from CH3OH and H2, had no effect on CO2 reduction to CO. In the presence of (cHxN)2C delta p was high and the intracellular ATP content was low. These findings suggest that the endergonic formation of CO from CO2 and H2 is coupled to the exergonic formation of CH4 from CH3OH and H2 via the proton motive force and not via ATP. CO formation was not stimulated by the addition of sodium ions.     

Mechanism of action of sparsomycin in protein synthesis.

Before CI isomerizes to C*I, we detect a competitive phase of inhibition (Ki = k5/k4 = 0.05 microM) which eventually, by increasing the concentration of I, becomes linear mixed noncompetitive and involves C*I in place of CI. The equilibration of C and I according to reaction 2 is much slower than the equilibration between C and S in reaction 1 (time-dependent inhibition). The inactivation plots obey reaction 2 and allow us to estimate k6 as equal to 2.2 min-1. The isomerized C*I, free of excess I, can be studied as a mixture with complex C. From the kinetics of the regeneration of C from C*I, in the presence of puromycin, we can estimate k7 to be between 0.22 min-1 and 0.06 min-1. Although the isomerized C*I survives after adsorption on cellulose nitrate filter disks, it does not survive after gel chromatography on a Sepharose CL-4B column but is converted quantitatively to complex C containing D of unchanged reactivity. This result does not support the proposed [Flynn, G. A., & Ash, R. J., (1990) Biochem. Biophys. Res. Commun. 166, 673-680] chemical reaction between D and I toward new products. The isomerized C*I can be obtained not only from the already-made complex C but also de novo from D, R, and M. In the latter case, the reactions which lead to C are represented by the following hypothetical scheme: D + R + M in equilibrium with DRM or C (binding reaction). When C*I is formed de novo, this reaction is coupled to reaction 2 and the ultimate product is a mixture of C and C*I.(ABSTRACT TRUNCATED AT 250 WORDS)     

Reactions of peroxyl radicals with Fe(H(2)O)(6)(2+)

The reactions of RO(2)* radicals with Fe(H(2)O)(6)(2+) were studied, R[double bond]H; CH(3); CH(2)COOH; CH(2)CN; CH(2)C(CH(3))(2)OH; CH(2)OH; CHCl(2)/CCl(3). All these processes involve the following reactions: Fe(H(2)O)(6)(2+)+RO(2)*<==>(H(2)O)(5)Fe(III)[bond]OOR(2+) K(1) approximately 250 M(-1); (H(2)O)(5)Fe(III)[bond]OOR(2+)+H(3)O(+)/H(2)O-->Fe(H(2)O)(6)(3+)+ROOH+H(2)O/OH(-); (H(2)O)(5)Fe(III)[bond]OOR(2+)+2Fe(H(2)O)(6)(2+)-->3Fe(H(2)O)(6)(3+)+ROH; 2 RO(2)*-->Products; RO(2)*+(H(2)O)(5)Fe(III)[bond]OOR(2+)-->Fe(H(2)O)(6)(2+)+products. The values of k(1) and k(3) [reaction is clearly not an elementary reaction] approach the ligand exchange rate of Fe(H(2)O)(6)(2+), i.e. these reactions follow an inner sphere mechanism and the rate determining step is the ligand exchange step. The rate of reaction is several orders of magnitude faster than that of the Fenton reaction. Surprisingly enough the K(1) values are nearly independent of the redox potential of the radical and are considerably higher than calculated from the relevant redox potentials. These results indicate that the ROO(-) ligands considerably stabilise the Fe(III) complex, this stabilisation is smaller for radicals with electron withdrawing groups which raise the redox potential of the radical but decrease the basicity of the ROO(-) ligands, two effects which seem to nearly cancel each other. Finally, the results clearly indicate that reaction (5) is relatively fast and affects the nature of the final products. The contribution of these reactions to oxidation processes involving 'Fenton-like' processes is discussed.     

Effects of antioxidants on X-ray- or hyperthermia-induced apoptosis in human lymphoma U937 cells

Hydroxyl radicals (.OH) and superoxide anion radicals (O2.-) are known to play cardinal roles in cell killing and various types of cell damage. In order to elucidate the mechanism of the involvement of both free radicals on apoptosis, the correlation between anti-apoptotic effects and free radical scavenging abilities of anti-oxidants was studied. As an indicator of anti-apoptotic effects, C1/2 (antioxidant concentration to inhibit DNA fragmentation by 50%) was evaluated in human lymphoma cell line U937 cells 6 hr after X-ray (10 Gy) or hyperthermia (44 degrees C, 30 min) treatment. Rate constants of the reactions between antioxidants and .OH or O2.- were calculated as the scavenging ability of the antioxidants with graded concentration estimated by EPR spectroscopy. No apparent correlation between C1/2 obtained in apoptosis induced by X-rays or hyperthermia and the rate constants of antioxidants for .OH or O2.- was observed. On the other hand, the partition coefficients in 1-octanol/water of the antioxidants, an indicator of hydrophobicity, revealed a correlation with the C1/2 of the agents with hyperthermia, but not with X-ray irradiation. These results indicate that the prevention of apoptosis by an antioxidant is not simply associated with its scavenging ability for .OH or O2.-. The hydrophobicity of the antioxidant, among other possible factors, is involved in the inhibition of hyperthermia- induced apoptosis.     

Inhibitory effect of resveratrol on free radical generation in blood platelets

Resveratrol (3,4',5-trihydroxystilbene), a compound found in many plants, has been shown to prevent coronary heart diseases and to exert a variety of antiinflammatory and anticancerogenic effects. It is effective in lowering the level of serum lipids and in inhibiting platelet aggregation. We evaluated the effect of trans-resveratrol on the production of free radicals in pig blood platelets and showed that resveratrol inhibited the production of different reactive oxygen species (O2*-, H2O2, singlet oxygen and organic radicals) measured by the luminol-dependent chemiluminescence in resting platelets (P < 0.05). Resveratrol inhibited also the generation of radicals in platelets activated by thrombin (P < 0.05). Treatment of platelets with resveratrol at concentrations of 6.25 and 12.5 microg/ml caused a statistically insignificant increase in the production of O2*- in these cells, as measured by reduction of cytochrome c; however, at higher doses (25, 50 and 100 microg/ml) resveratrol distinctly reduced the generation of O2*- in platelets (P < 0.05). We suggest that free radicals play an important role in the reduced reactivity of blood platelets induced by resveratrol.     

Mechanism of peroxidase actions for salicylic acid-induced generation of active oxygen species and an increase in cytosolic calcium in tobacco cell suspension culture

Extracellularly secreted peroxidases in cell suspension culture of tobacco (Nicotiana tabacum L. cv. Bright Yellow-2, cell line BY-2) catalyse the salicylic acid (SA)-dependent formation of active oxygen species (AOS) which, in turn, triggers an increase in cytosolic Ca2+ concentration. Addition of horseradish peroxidase (HRP) to tobacco cell suspension culture enhanced the SA-induced increase in cytosolic Ca2+ concentration, suggesting that HRP enhanced the production of AOS. The mechanism of peroxidase-catalysed generation of AOS in SA signalling was investigated with chemiluminescence sensitive to AOS and electron spin resonance (ESR) spectroscopy, using the cell suspension culture of tobacco, and HRP as a model system of peroxidase reaction. The results showed that SA induced the peroxidase inhibitor-sensitive production of superoxide and H2O2 in tobacco suspension culture, but no production of hydroxy radicals was detected. Similar results were obtained using HRP. It was also observed that SA suppressed the H2O2-dependent formation of hydroxy radicals in vitro. The results suggest that SA protect the cells from highly reactive hydroxy radicals, while producing the less reactive superoxide and H2O2 through peroxidase-catalysed reaction, as the intermediate signals. The formation of superoxide was followed by that of H2O2, suggesting that superoxide was converted to H2O2. In addition, it was observed that superoxide dismutase-insensitive ESR signal of monodehydroascorbate radical was induced by SA both in the tobacco suspension culture and HRP reaction mixture, suggesting that SA free radicals, highly reactive against ascorbate, were formed by peroxidase-catalysed reactions. The formation of SA free radicals may lead to subsequent monovalent reduction of O2 to superoxide.