The oxidation of ferrocytochrome c by Br−2, (SCN)−2, N3 and OH radicals studied by pulsed-electron and γ-ray radiolysis

The reactions of ferrocytochrome c with Br^?₂, (SCN)^?₂, N₃ and OH radicals were followed by measuring the change in the optical spectra of cytochrome c on γ-irradiation as well as the rate of change of absorbance upon pulse irradiation.Ferrocytochrome c is oxidized to ferricytochrome c by Br^?₂, (SCN)^?₂ or N₃ radical with an efficiency of about 100% through a second-order process in which no intermediates were observed. The rate constants in neutral solutions at I = 0.073 are 9.7 · 10⁸ M^?1 · s^?1, 7.9 · 10⁸ M^?1 · s^?1, 1.3 · 10⁹ M^?1 · s^?1 for the oxidation by Br^?₂, (SCN)^?₂ and N₃ radicals, respectively. The rate constants do not vary appreciably in alkaline solutions (pH 8.9). The ionic strength dependence was observed for the rate constants of the oxidation by Br^?₂ and (SCN)^?₂. Those rate constants estimated on the assumption that the radicals react only with the amino acid residues with the characteristic steric correction factors were less than one-tenth of the observed ones. These results suggest that the partially exposed region of the heme is the probable site of electron transfer from ferrocytochrome c to the radical.Hydroxyl radicals also oxidize ferrocytochrome c with a high rate constant (k > 1 · 10¹⁰ M^?1 · s^?1), but with a very small efficiency (5%).     

The role of interactions between O2, H2O2, .OH,e- and O2- in free radical damage to biological systems

The occurrence of the Haber-Weiss reaction and other interactions between free radicals has been investigated in the effects of mixtures of free radicals on the permeability of resealed erythrocyte ghosts and on the activity of membrane-bound glyceraldehyde-3-phosphate dehydrogenase. The following mixtures were found to induce damage greater than that which could be accounted for by the independent actions of the constituent free radicals: (i) · OH + H₂O₂, and (ii) · OH + H₂O₂ + O₂^?. In contrast, the following mixtures were found to induce less damage than that predicted on the basis of independent actions of constituent free radicals: (i) H₂O₂ + O₂^?, and (ii) oxidizing radicals ( · OH, H₂O₂) + reducing radicals (e^?, H · ). These results suggest a Haber-Weiss-like interaction between H₂O₂ and O₂^?and an interaction between H₂O₂ and · OH to produce a species more potent than either in causing increased permeability. The decrease in damage observed in the simultaneous presence of oxidizing and reducing radicals suggests an antagonistic effect by which each tends to moderate damage by the other. Inactivation of glyceraldehyde-3-phosphate dehydrogenase was found to be more sensitive to radiation than permeability by an order of magnitude, while permeability was more sensitive to the enhancement of damage by oxygen. Comparison of the effectiveness of free radical scavengers in inhibiting the increase in permeability caused by free radicals showed the following order of effectiveness, expressed in terms of percentage protection: formate (90%) > nitrogen (65%) > catalase (60%) > dismutase (32%), and with respect to enzymatic inactivation, nitrogen (100%) > formate (77%) > dismutase (48%) > catalase (44%). The relative rates observed anaerobically and aerobically in the presence and absence of the above scavengers suggest that (at least in the case of radiation damage to the membranes of erythrocyte ghost cells) the “oxygen effect” is due to the interaction of oxygen with e^? and H., producing O₂^? which aggravates damage under conditions which allow consequent Haber-Weiss-like reactions. The further increase in damage when oxygen concentration is raised yet higher is due to the interaction of oxygen with the sites of initial damage.     

Pulse-radiolytic investigations of catechols and catecholamines II. Reactions of Tiron with oxygen radical species

Transient spectra and kinetic data of Tiron (1,2-dihydroxybenzene-3,5-disulphonic acid) are reported, obtained after pulse-radiolytic oxidation by hydroxyl radicals (°OH), superoxide anions (O₂^?) or a combination of both oxygen radicals. The rate constant with °OH radicals was determined at 1.0·10⁹ M^?1·s^?1. Contrary to a previous report (Greenstock, C.L. and Miller, R.W. (1975) Biochim. Biophys. Acta 396, 11–16), the rate constant with O₂^? of 1.0·10⁷ M^?1·s^?1 is lower by one order of magnitude; also the semiquinone absorbs at 300 nm rather than at 400 nm. The ratio of the rate constants with °OH and O₂^? of 100 again demonstrates that any oxidation reaction by the latter radical is unspecific due to the more efficient reaction of °OH radicals, leading to the same products with catechol compounds.     

Lutein-chlorophyll-a energy transfer in detergent micelles

Absorption, fluorescence and fluorescence excitation spectra were determined for equimolar mixed micellar detergent solutions of lutein and chlorophyll-a in the concentration range from 9·10^?6 to 1.8·10^?4 M, with detergent (triton-X100) concentrations from 3·-10^?4 to 7·10^?3 M. In the range of detergent concentrations studied the pigments incorporated into the detergent micelles attained a high local concentration (0.1 to 0.01 M), reminiscent of pigment concentration within the chloroplast. A lutein → chlorophyll-a energy transfer with an efficiency of about 15% was found in these systems. In dilute (9·10^?6 M) pigment solution with concentrated (7·10^?3 M) detergent practically no transfer is observed. The extent of aggregation and the efficiency of transfer depend on the composition of the system. The aggregation of chlorophyll-a is partly inhibited by lutein molecules. It is shown that the energy transfer efficiency as function of distance follows anr ^?3 relationship,R ₀ being 22 å.     

Extraction and analysis of superoxide free radicals (·O−2) from whole mammalian liver

Extraction of whole lobes of normal rat liver with dimethyl sulphoxide (DMSO) under N₂ gives extracts which contain 5—10 μmol/l·O^?₂ (50-100 nmol·O^?₂ per 10 ml extract per 4 g liver; 1.25-2.50 nmol·O^?₂ per millilitre per gram liver). Evidence for ·O^?₂ in the extracts is given by: (1) electron spin resonance signals (ESR), (2) differential pulse polarography (DPP), (3) chemiluminescence (CL), and (4) nitroblue tetrazolium reduction (NBT). All tests yield results identical with those obtained with authentic ·O^?₂. Extraction of ·O^?₂ is enhanced by tetrabutyl ammonium ion, and is maximal at 1-3 min. These results raise the possibility that substantial amounts of ·O^?₂ are normally sequestered in protective membranous sites in vivo.     

Rates of reduced cytochrome c-ferricyanide binding and electron transfer

The oxidation of reduced cytochrome c by ferricyanide has been studied over a wide range of ferricyanide concentrations using a continuous-flow apparatus. The formation of a ferrocytochrome c-ferricyanide complex has been demonstrated and the binding and electron transfer processes separated to give both the oxidation electron transfer rate and the binding rate parameters. The electron transfer rate has been found to be 1.86 · 10³ s^?1 in H₂O buffer and 1.36 · 10³ s^?1 in ²H₂O demonstrating that a deuterium isotope effect of similar magnitude (R = 1.37) to that found in the cytochrome reactions in photosynthetic bacteria [18] is also found in the reaction studied here. The binding association rate parameters also show a similar deuterium isotope effect suggesting that water rotation may be involved in both the binding of ferricyanide to reduced cytochrome c and the subsequent oxidation electron transfer.     

The kinetics and specificity of electron transfer from cytochromes and copper proteins to P700

The rates of electron transfer to P700 from plastocyanin and cytochrome f have been compared with those from three other c-type cytochromes and azurin, a copper protein resembling plastocyanin. Three different disruptive techniques were used to expose P700; digitonin, Triton X-100 and sonication. The following rate constants were measured at 25 °C, pH 7.0, with digitonin-treated chloroplasts: plastocyanin, 8 · 10⁷ M^?1 · s^?1; red-algal cytochrome c-553, 1.9 · 10⁷ M^?1 · s^?1; Pseudomonas cytochrome c-551, 8 · 10⁶ M^?1 · s^?1; azurin, ? 3 · 10⁵ M^?1 · s^?1; cytochrome f, ? 2 · 10⁴ M^?1 · s^?1; mammalian cytochrome c, ? 2 · 10⁴ M^?1 · s^?1. For electron transfer from plastocyanin, the effects of ionic strength, pH and temperature were also studied, and saturation effects found in earlier work were avoided by a full consideration of the various secondary reactions and inclusion of superoxide dismutase. The relative rates are discussed in relation to photosynthetic electron transport.     

On the repair of oxidative damage to apoferritin: a model study with the flavonoids quercetin and rutin in aerated and deaerated solutions

Abstract

Ferritin (Ft) impairment through ^?O₂^–, H₂O₂, and ^?OH production occurs in the cases of ketoses, diabetes mellitus, acute intermittent porphyria and tyrosinemia. In addition to ^?Trp and TyrO^? radical production, ferrous iron liberation and Ft synthesis stimulation, site-specific oxidation reactions are induced leading to toxic iron accumulation in organs with high Ft content, for example, liver and brain. To elucidate the potential pathways to Ft recovery, repair of oxidative damage to horse spleen apoferritin (apoFt) and Ft by quercetin (QH) or rutin (RH) was studied in the presence and absence of oxygen. ^?Trp and TyrO^? radicals were produced in pulse radiolysis through apoFt oxidation by ^?Br₂^– radicals. QH and RH bind to apoFt on eight sites with binding constants of ?80,000 and ?32,000 M^?1, respectively. In deaerated solutions, a repair of apoFt radicals is observed involving both bound and free flavonoids. This repair occurs by a fast intra- and a slow inter-molecular electron transfer from bound and free flavonoids, respectively. With QH, the rate constants are 10⁴ s^–1 and 3.5 × 10⁷ M^–1 s^–1 for the intra- and intermolecular repair reactions, respectively. Oxygen does not interfere with repair of apoFt or Ft by bound QH but inhibits 90% of Ft repair by RH. These results taken together indicate that flavonoid antioxidants may help alleviate Ft impairment in diseases involving an oxidative stress.     

Scavenging of reactive oxygen species by N‐substituted indole‐2 and 3‐carboxamides

Indole‐2 and 3‐carboxamides (IDs) are proposed to be selective cyclooxygenase inhibitors. Since cyclooxygenase‐1 may be involved in reactive oxygen species (ROS) production, we hypothesize that these indole derivatives have antioxidative properties. We have employed chemiluminescence (CL) and electron spin resonance (ESR) spin trapping to examine this hypothesis. We report here the results of a study of reactivity of 10 selected indole derivatives towards ROS. The following generators of ROS were applied: potassium superoxide (KO₂) as a source of superoxide radicals (O₂^·?), the Fenton reaction (Co‐EDTA/H₂O₂) for hydroxyl radicals (HO^·), and a mixture of alkaline aqueous H₂O₂ and acetonitrile for singlet oxygen (¹O₂). Hydroxyl radicals were detected as 5,5‐dimethyl‐1‐pyrroline‐N‐oxide (DMPO) spin adduct, whereas 2,2,6,6‐tetramethyl‐piperidine (TEMP) was used as a detector of ¹O₂. Using the Fenton reaction, 0.5 mmol/L IDs were found to inhibit DMPO‐?H radical formation in the range 7–37%. Furthermore the tested compounds containing the thiazolyl group also inhibited the ¹O₂‐dependent TEMPO radical, generated in the acetonitrile + H₂O₂ system. About 20% inhibition was obtained in the presence of 0.5 mmol/L IDs. 1 mmol/L IDs caused an approximately 13–70% decrease in the CL sum from the O₂^·? generating system (1 mmol/L). The aim of this paper is to evaluate these indole derivatives as antioxidants and their abilities to scavenge ROS. Copyright © 2004 John Wiley & Sons, Ltd.     

Interactions of superoxide anion with enzyme radicals: Kinetics of reaction with lysozyme tryptophan radicals and corresponding effects on tyrosine electron transfer

The kinetics of O^·-₂ reaction with semi-oxidized tryptophan radicals in lysozyme, Trp^·(Lyz) have been investigated at various pHs and conformational states by pulse radiolysis. The Trp^·(Lyz) radicals were formed by Br^·-₂ oxidation of the 3–4 exposed Trp residues in the protein. At pH lower than 6.2, the apparent bimolecular rate is about 2 × 10⁸M^-1s^-1; but drops to 8 × 10⁷M^-1s^-1 or less above pH 6.3 and in CTAC micelles. Similarly, the apparent bimolecular rate constant for the intermolecular Trp^·(Lyz) + Trp^·(Lyz) recombination reaction is about (4-7 × 10⁶M^-1s^-1) at/or below pH 6.2 then drops to 1.3-1.6 × 10⁶M^-1s^-1 at higher pH or in micelles. This behavior suggests important conformational and/or microenvironmental rearrangement with pH, leading to less accessible semioxidized Trp^· residues upon Br^·-₂ reaction. The kinetics of Trp^·(Lyz) with ascorbate, a reducing species rather larger than O^·-₂ have been measured for comparison. The well-established long range intramolecular electron transfer from Tyr residues to Trp radicals-leading to the repair of the semi-oxidized Trp^·(Lyz) and formation of the tyrosyl phenoxyl radical is inhibited by the Trp^·(Lyz)+O^·-₂ reaction, as is most of the Trp^·(Lyz)+Trp^·(Lyz) reaction. However, the kinetic behavior of Trp^·(Lyz) suggests that not all oxidized Trp residues are involved in the intermolecular recombination or reaction with O^·-₂. As the kinetics are found to be quite pH sensitive, this study demonstrates the effect of the protein conformation on O^·-₂ reactivity. To our knowledge, this is the first report on the kinetics of a protein-O^·-₂ reaction not involving the detection of change in the redox state of a prosthetic group to probe the reactivity of the superoxide anion.     

Thioctic Acid and Dihydrolipoic Acid are Novel Antioxidants Which Interact With Reactive Oxygen Species

Thioctic acid (TA) and its reduced form dihydrolipoic acid (DHLA) have recently gained somc recognition as useful biological antioxidants. In particular, the ability of DHLA to inhibit lipid peroxidation has been reported. In the present study, the effects of TA and DHLA on reactive oxygen species (ROS) generated in the aqueous phase have been investigated. Xanthine plus xanthine oxidase-generated superoxide radicals (O₂), detected by electron spin resonance spectroscopy (ESR) using DMPO as a spin trap. were eliminated by DHLA but not by TA. The sulhydryl content of DHLA, measured using Ellman's reagent decreased subsequent to the incubation with xanthine plus xanthine oxidase confirming the interaction between DHLA and O₂^-. An increase of hydrogen peroxide concentration accompanied the reaction between DHLA and O₂x, suggesting the reduction of O₂^- by DHLA. Competition of O₂^- with epinephrine allowed us to estimate a second order kinetic constant of the reaction between O₂^- and DHLA, which was found to be a 3.3 × 10⁵ M^-1 s^-1. On the other hand, the DMPO signal of hydroxyl radicals (HO ·) generated by Fenton's reagent were eliminated by both TA and DHLA. Inhibition of the Fenton reaction by TA was confirmed by a chemiluminescence measurement using luminol as a probe for HO ·. There was no electron transfer from Fe²⁺ to TA or from DHLA to Fe^{3 +} detected by measuring the Fe²⁺ -phenanthroline complex. DHLA did not potentiate the DMPO signal of HO · indicating no prooxidant activity of DHLA. These results suggest that both TA and DHLA possess antioxidant properties. In particular. DHLA is very effective as shown by its dual capability by eliminating both O₂^-; and HO ·.     

The effect of antioxidants on the formation of free radicals and primary products of the peroxidase reaction

Antioxidants suppress the formation of radicals in peroxidase processes. The chemiluminescence kinetic curves in the horseradish peroxidase/luminol/H₂O₂ system have been compared with those obtained from the mathematical model of the reaction. It was shown that the effect of trolox, ascorbate, and mexidol is a result of the reaction of the luminol radical with the inhibitor molecule (rate constants, 1.0·10¹⁰, 9.0·10⁹, and 2.3·10⁵ mol^–1 min^–1, respectively). The antiradical action of quercetin has been described by eight reactions that were based on the assumption of two reaction centers in the molecule, each reacting with two radicals. The hypothesis that the antioxidant molecule captures the enzyme intermediate radicals in peroxidase cycle, rather than the radical of the reaction product was not confirmed because the calculated curves differed from the experimental point positions. Apparently, the formation of radicals in the presence of peroxidases in living cells and the subsequent events, such as apoptosis, may be prevented not only by the inhibition of an enzyme but also by antioxidants that capture free-radical reaction products     

Fast repair of purine deoxynucleotide radical cations by rutin and quercetin

Repair effects of rutin and quercetin on purine deoxynucleotide radical cations were studied using pulse radiolysis technique. On electron pulse irradiation of N2 saturated deoxynucleotide aqueous solution containing 20 mmol/L K2S2O8, 200 mmol/L f-BuOH and rutin or quercetin, the transient absorption spectra of the deoxynucleotide radical cations decayed quickly. At the same time, the spectra of flavonoid phenoxyl radicals formed within several dozen microseconds. The results indicated that deoxynucleotide radical cations can be repaired by flavonoids. The rate constants of the repair reactions were 3.8 × 108-4.4×108 mol-1 · L · s-1 and 1.3×108-1.8×108 mol-1 · L · s-1 for dAMP and dGMP radical cations, respectively.     

Fast repair of purine deoxynucleotide radical cations by rutin and quercetin

Repair effects of rutin and quercetin on purine deoxynucleotide radical cations were studied using pulse radiolysis technique. On electron pulse irradiation of N₂ saturated deoxynucleotide aqueous solution containing 20 mmol/L K₂S₂O₈, 200 mmol/Lt-BuOH and rutin or quercetin, the transient absorption spectra of the deoxynucleotide radical cations decayed quickly. At the same time, the spectra of flavonoid phenoxyl radicals formed within several dozen microseconds. The results indicated that deoxynucleotide radical cations can be repaired by flavonoids. The rate constants of the repair reactions were 3.8 ×10⁸-4.4 ×10⁸ mol⁻¹ · L · s⁻¹ and 1.3×10⁸-1.8×10⁸ mol⁻¹ · L · s⁻¹ for dAMP and dGMP radical cations, respectively.     

Production of reactive oxygen species in decoupled, Ca2+-depleted PSII and their use in assigning a function to chloride on both sides of PSII

Extraction of Ca²⁺ from the oxygen-evolving complex of photosystem II (PSII) in the absence of a chelator inhibits O₂ evolution without significant inhibition of the light-dependent reduction of the exogenous electron acceptor, 2,6-dichlorophenolindophenol (DCPIP) on the reducing side of PSII. The phenomenon is known as “the decoupling effect” (Semin et al. Photosynth Res 98:235–249, 2008). Extraction of Cl^? from Ca²⁺-depleted membranes (PSII[–Ca]) suppresses the reduction of DCPIP. In the current study we investigated the nature of the oxidized substrate and the nature of the product(s) of the substrate oxidation. After elimination of all other possible donors, water was identified as the substrate. Generation of reactive oxygen species HO, H₂O₂, and O ₂ ^·? , as possible products of water oxidation in PSII(–Ca) membranes was examined. During the investigation of O ₂ ^·? production in PSII(–Ca) samples, we found that (i) O ₂ ^·? is formed on the acceptor side of PSII due to the reduction of O₂; (ii) depletion of Cl^? does not inhibit water oxidation, but (iii) Cl^? depletion does decrease the efficiency of the reduction of exogenous electron acceptors. In the absence of Cl^? under aerobic conditions, electron transport is diverted from reducing exogenous acceptors to reducing O₂, thereby increasing the rate of O ₂ ^·? generation. From these observations we conclude that the product of water oxidation is H₂O₂ and that Cl^? anions are not involved in the oxidation of water to H₂O₂ in decoupled PSII(–Ca) membranes. These results also indicate that Cl^? anions are not directly involved in water oxidation by the Mn cluster in the native PSII membranes, but possibly provide access for H₂O molecules to the Mn₄CaO₅ cluster and/or facilitate the release of H⁺ ions into the lumenal space.     

Scavenging of hydroxyl,methoxy, and nitrogen dioxide free radicals by some methylated isoflavones

Free radicals can be scavenged from biological systems by genistein, daidzein, and their methyl derivatives through hydrogen atom transfer (HAT), single-electron transfer (SET), and sequential proton-loss electron-transfer (SPLET) mechanisms. Reactions between these derivatives and the free radicals OH^., OCH₃^., and NO₂^. via the HAT mechanism in the gas phase were studied using the transition state theory within the framework of DFT. Solvation of all the species and complexes involved in the HAT reactions in aqueous media was treated by performing single point energy calculations using the polarizable continuum model (PCM). The SET and SPLET mechanisms for the above reactions were also considered by applying the Marcus theory of electron transfer, and were found to be quite sensitive to geometry and solvation. Therefore, the geometries of all the species involved in the SET and SPLET mechanisms were fully optimized in aqueous media. The calculated barrier energies and rate constants of the HAT-based scavenging reactions showed that the OH group of the B ring in genistein, daidzein, and their methyl derivatives plays a major role in the scavenging of free radicals, and the role of this OH group in the HAT-based free-radical scavenging decreases in the following order: OH^.?>?OCH₃^. > NO₂^.. The SPLET mechanism was found to be an important mechanism in these free-radical scavenging reactions, whereas the SET mechanism was not important in this context.     

Experimental determination of the redox potential of the superoxide radical O 2

The redox potential of ^?O₂^? was determined based on the dependence of the electron transfer reaction from ^?O₂^? upon the known redox potential of various acceptors A (including a range of quinones, dyes and ferricyanide). The efficiencies and the rates of these electron transfer processes were determined, using the technique of pulse radiolysis, by monitoring the formation kinetics of the semiquinone radical anions at the appropriate wavelength. From the percentage efficiency versus E^o′ plot, an E^o′ value of + 0.15 ± 0.01 V at pH 7.0 and 22°C, or E^o = + 0.57 ± 0.01 V, for the ${}^{\mathrm{?}}\text{O}{}_2{}^{\mathrm{?}}\text{O}{}_2$ couple was obtained. The rate k(^?O₂^? + A → O₂ + ^?A^?) = 9.8 × 10⁸ M^?1 sec^?1 where A = p-benzoquinone. The E^o value for the ^?HO₂ radical is > 1.0 V. It was also found that hydroquinone can quantitatively reduce ^?O₂^? to H₂O₂, k(^?O₂^? + QH₂ → QH^? + HO₂^?) = 1.6 ± 0.1 × 10⁷ M^?1 sec^?1 at pH 7.0 and 22°C.     

Biphasic regulation of superoxide radical levels in Mn-depleted and photoactivated photosystem II

In the past decades, it has become clear that superoxide radical (O₂ ^·?) can be generated from photosystem II (PSII) during photosynthesis. Depending on the extent of its accumulation, O₂ ^·? plays an important role in plant physiology and pathology. The photoinhibition/repair cycle is a typical process in PSII which is mainly responsible for the survival of plants under the photoinihibition condition. It is therefore of significant importance to determine O₂ ^·? production in this cycle, and then explore how O₂ ^·? is controlled by PSII within a normal physiological level. With this in mind, we herein investigate the variation of the O₂ ^·? levels in PSII under Mn-depleted and photoactivated conditions mimicking the photoinhibition/repair cycle in vitro. The effect of intrinsic SOD-like component on the O₂ ^·? levels was also studied. Results show that PSII has the ability to regulate the O₂ ^·? levels in these two processes by simultaneously modulating the O₂ ^·? generation activity and intrinsic SOD-like activity. This finding could shed new lights on the photoprotective property of PSII against O₂ ^·? and other reactive oxygen species.     

Chemiluminescence of lucigenin by electrogenerated superoxide ions in aqueous solutions

The chemiluminescence reaction of lucigenin (Luc²⁺?2NO₃^?, N,N′‐dimethyl‐9,9′‐biacridinium dinitrate) at gold electrodes in dioxygen‐saturated alkaline aqueous solutions (pH 10) was investigated in detail by the use of electrochemical emission spectroscopy. We noted that both O₂ and Luc²⁺ are reduced on a gold electrode in aqueous solution of pH 10 in almost the same potential region. From this fact, we expected chemiluminescence based on a radical–radical coupling reaction of superoxide ion (O₂·^?) and one‐electron reduced form of Luc²⁺ (Luc·⁺, a radical cation). Chemiluminescence was actually observed in the potential range where O₂ and Luc²⁺ were simultaneously reduced at the electrodes. The effects were examined upon addition of enzymes, i.e. superoxide dismutase (SOD) and catalase, into the solution and the substitution of heavy water (D₂O) for light water (H₂O) as a solvent on the chemiluminescence. In the presence of native and active SOD, chemiluminescence was completely absent. On the other hand, chemiluminescence was observed, unchanged in the presence of either denatured and inert SOD or catalase. In addition, the amount of chemiluminescence in D₂O solution was about three times greater than that in H₂O solution. These results, together with cyclic voltammetric results, suggest that O₂·^? participates directly in the chemiluminescence but H₂O₂ does not, and the chemiluminescence results from the coupling reaction between O₂·^? and Luc^·+ under the present experimental conditions. These chemically unstable species, O₂·^? and Luc·⁺, are produced during the simultaneous electroreduction of O₂ and Luc²⁺. The coupling reaction between those radical species would lead to the formation of a dioxetane‐type intermediate and, finally, to chemiluminescence. The chemiluminescence reaction mechanism is discussed. Copyright © 2002 John Wiley & Sons, Ltd.