NADH- and NADPH-dependent formation of superoxide anions by bovine heart submitochondrial particles and NADH–ubiquinone reductase preparation

1. Both NADH and NADPH supported the oxidation of adrenaline to adrenochrome in bovine heart submitochondrial particles. The reaction was completely inhibited in the presence of superoxide dismutase, suggesting that superoxide anions (O(2) (-)) are responsible for the oxidation. The optimal pH of the reaction with NADPH was at pH7.5, whereas that with NADH was at pH9.0. The reaction was inhibited by treatment of the preparation with p-hydroxymercuribenzoate and stimulated by treatment with rotenone. Antimycin A and cyanide stimulated the reaction to the same extent as rotenone. The NADPH-dependent reaction was inhibited by inorganic salts at high concentrations, whereas the NADH-dependent reaction was stimulated. 2. Production of O(2) (-) by NADH-ubiquinone reductase preparation (Complex I) with NADH or NADPH as an electron donor was assayed by measuring the formation of adrenochrome or the reduction of acetylated cytochrome c which does not react with the respiratory-chain components. p-Hydroxymercuribenzoate inhibited the reaction and rotenone stimulated the reaction. The effects of pH and inorganic salts at high concentrations on the NADH- and NADPH-dependent reactions of Complex I were essentially similar to those on the reactions of submitochondrial particles. 3. These findings suggest that a region between a mercurialsensitive site and the rotenone-sensitive site of the respiratory-chain NADH dehydrogenase is largely responsible for the NADH- and NADPH-dependent O(2) (-) production by the mitochondrial inner membranes.     

Production of reactive oxygen species by mitochondria: central role of complex III

The mitochondrial respiratory chain is a major source of reactive oxygen species (ROS) under pathological conditions including myocardial ischemia and reperfusion. Limitation of electron transport by the inhibitor rotenone immediately before ischemia decreases the production of ROS in cardiac myocytes and reduces damage to mitochondria. We asked if ROS generation by intact mitochondria during the oxidation of complex I substrates (glutamate, pyruvate/malate) occurred from complex I or III. ROS production by mitochondria of Sprague-Dawley rat hearts and corresponding submitochondrial particles was studied. ROS were measured as H2O2 using the amplex red assay. In mitochondria oxidizing complex I substrates, rotenone inhibition did not increase H2O2. Oxidation of complex I or II substrates in the presence of antimycin A markedly increased H2O2. Rotenone prevented antimycin A-induced H2O2 production in mitochondria with complex I substrates but not with complex II substrates. Catalase scavenged H2O2. In contrast to intact mitochondria, blockade of complex I with rotenone markedly increased H2O2 production from submitochondrial particles oxidizing the complex I substrate NADH. ROS are produced from complex I by the NADH dehydrogenase located in the matrix side of the inner membrane and are dissipated in mitochondria by matrix antioxidant defense. However, in submitochondrial particles devoid of antioxidant defense ROS from complex I are available for detection. In mitochondria, complex III is the principal site for ROS generation during the oxidation of complex I substrates, and rotenone protects by limiting electron flow into complex III.     

Generation of hydrogen peroxide by brain mitochondria: the effect of reoxygenation following postdecapitative ischemia

The hypothesis that mitochondria damaged during complete cerebral ischemia generate increased amounts of superoxide anion radical and hydrogen peroxide (H2O2) upon postischemic reoxygenation has been tested. In rat brain mitochondria, succinate supported H2O2 generation, whereas NADH-linked substrates, malate plus glutamate, did so only in the presence of respiratory chain inhibitors. Succinate-supported H2O2 generation was diminished by rotenone and the uncoupler carbonyl cyanide m-chlorphenylhydrazone and enhanced by antimycin A and increased oxygen tensions. When maximally reduced, the NADH dehydrogenase and the ubiquinone-cytochrome b regions of the electron transport chain are sources of H2O2. These studies suggest that a significant portion of H2O2 generation in brain mitochondria proceeds via the transfer of reducing equivalents from ubiquinone to the NADH dehydrogenase portion of the electron transport chain. Succinate-supported H2O2 generation by mitochondria isolated from rat brain exposed to 15 min of postdecapitative ischemia was 90% lower than that of control preparations. The effect of varying oxygen tensions on H2O2 generation by postischemic mitochondrial preparations was negligible compared with the increased H2O2 generation measured in control preparations. Comparison of the effects of respiratory chain inhibitors and oxygen tension on succinate-supported H2O2 generation suggests that the ability for reversed electron transfer is impaired during ischemia. These data do not support the hypothesis that mitochondrial free radical generation increases during postischemic reoxygenation.     

Role of ubiquinone in the mitochondrial generation of hydrogen peroxide.

Antimycin-inhibited bovine heart submitochondrial particles generate O2- and H2O2 with succinate as electron donor. H2O2 generation involves the action of the mitochondrial superoxide dismutase, in accordance with the McCord & Fridovich [(1969) j. biol. Chem. 244, 6049-6055] reaction mechanism. Removal of ubiquinone by acetone treatment decreases the ability of mitochondrial preparations to generate O2- and H2O2, whereas supplementation of the depleted membranes with ubiquinone enhances the peroxide-generating activity in the reconstituted membranes. Addition of superoxide dismutase to ubiquinone-reconstituted membranes is essential in order to obtain maximal rates of H2O2 generation since the acetone treatment of the membranes apparently inactivates (or removes) the mitochondrial superoxide dismutase. Parallel measurements of H2O2 production, succinate dehydrogenase and succinate-cytochrome c reductase activities show that peroxide generation by ubiquinone-supplemented membranes is a monotonous function of the reducible ubiquinone content, whereas the other two measured activities reach saturation at relatively low concentrations of reducible quinone. Alkaline treatment of submitochondrial particles causes a significant decrease in succinate dehydrogenase activity and succinate-dependent H2O2 production, which contrasts with the increase of peroxide production by the same particles with NADH as electron donor. Solubilized succinate dehydrogenase generates H2O2 at a much lower rate than the parent submitochondrial particles. It is postulated that ubisemiquinone (and ubiquinol) are chiefly responsible for the succinate-dependent peroxide production by the mitochondrial inner membrane.     

Reversible dissociation of flavin mononucleotide from the mammalian membrane-bound NADH: ubiquinone oxidoreductase (complex I)

Conditions for the reversible dissociation of flavin mononucleotide (FMN) from the membrane-bound mitochondrial NADH:ubiquinone oxidoreductase (complex I) are described. The catalytic activities of the enzyme, i.e. rotenone-insensitive NADH:hexaammineruthenium III reductase and rotenone-sensitive NADH:quinone reductase decline when bovine heart submitochondrial particles are incubated with NADH in the presence of rotenone or cyanide at alkaline pH. FMN protects and fully restores the NADH-induced inactivation whereas riboflavin and flavin adenine dinucleotide do not. The data show that the reduction of complex I significantly weakens the binding of FMN to protein thus resulting in its dissociation when the concentration of holoenzyme is comparable with K(d ( approximately 10(-8)M at pH 10.0).     

Evidence that the blockade of mitochondrial respiration by the neurotoxin 1-methyl-4-phenylpyridinium (MPP+) involves binding at the same site as the respiratory inhibitor, rotenone

It has been postulated that 1-methyl-4-phenylpyridinium (MPP+) blocks mitochondrial respiration by combining at the same site as rotenone, a potent inhibitor of NADH oxidation in mitochondria, known to act at the junction of NADH dehydrogenase and coenzyme Q (CoQ). The present experiments show that MPP+ and two of its analogs indeed act in a concentration dependent manner to prevent the binding of [14C]-rotenone to submitochondrial particles (ETP) and significantly decrease the inhibition of electron transport caused by rotenone. It therefore appears that MPP+ binds at the same site as rotenone or an adjacent site, supporting the hypothesis that its neurotoxic action is due to the inhibition of mitochondrial respiration.     

Redox cycling of anthracyclines by cardiac mitochondria. I. Anthracycline radical formation by NADH dehydrogenase

In the present study we have used beef heart submitochondrial preparations (BH-SMP) to demonstrate that a component of mitochondrial Complex I, probably the NADH dehydrogenase flavin, is the mitochondrial site of anthracycline reduction. During forward electron transport, the anthracyclines doxorubicin (Adriamycin) and daunorubicin acted as one-electron acceptors for BH-SMP (i.e. were reduced to semiquinone radical species) only when NADH was used as substrate; succinate and ascorbate were without effect. Inhibitor experiments (rotenone, amytal, piericidin A) indicated that the anthracycline reduction site lies on the substrate side of ubiquinone. Doxorubicin and daunorubicin semiquinone radicals were readily detected by ESR spectroscopy. Doxorubicin and daunorubicin semiquinone radicals (g congruent to 2.004, signal width congruent to 4.5 G) reacted avidly with molecular oxygen, presumably to produce O2-, to complete the redox cycle. The identification of Complex I as the site of anthracycline reduction was confirmed by studies of ATP-energized reverse electron transport using succinate or ascorbate as substrates, in the presence of antimycin A or KCN respiratory blocks. Doxorubicin and daunorubicin inhibited the reduction of NAD+ to NADH during reverse electron transport. Furthermore, during reverse electron transport in the absence of added NAD+, doxorubicin and daunorubicin addition caused oxygen consumption due to reduction of molecular oxygen (to O2-) by the anthracycline semiquinone radicals. With succinate as electron source both thenoyltrifluoroacetone (an inhibitor of Complex II) and rotenone blocked oxygen consumption, but with ascorbate as electron source only rotenone was an effective inhibitor. NADH oxidation by doxorubicin during BH-SMP forward electron transport had a KM of 99 microM and a Vmax of 30 nmol X min-1 X mg-1 (at pH 7.4 and 23 degrees C); values for daunorubicin were 71 microM and 37 nmol X min-1 X mg-1. Oxygen consumption at pH 7.2 and 37 degrees C exhibited KM values of 65 microM for doxorubicin and 47 microM for daunorubicin, and Vmax values of 116 nmol X min-1 X mg-1 for doxorubicin and 114 nmol X min-1 X mg-1 for daunorubicin. In marked contrast with these results, 5-iminodaunodrubicin (a new anthracycline with diminished cardiotoxic potential) exhibited little or no tendency to undergo reduction, or to redox cycle with BH-SMP. Redox cycling of anthracyclines by mitochondrial NADH dehydrogenase is shown, in the accompanying paper (Doroshow, J. H., and Davies, K. J. A. (1986) J. Biol. Chem. 261, 3068-3074), to generate O2-, H2O2, and OH which may underlie the cardiotoxicity of these antitumor agents.     

The mechanism of oxidation of reduced nicotinamide dinucleotide phosphate by submitochondrial particles from beef heart.

1. Oxidation of NADPH by various acceptors catalyzed by submitochondrial particles and a partially purified NADH dehydrogenase from beef heart was investigated. Submitochondrial particles devoid of nicotinamide nucleotide transhydrogenase activity catalyze an oxidation of NADPH by oxygen. The partially purified NADH dehydrogenase prepared from these particles catalyzes an oxidation of NADPH by acetylpyridine-NAD. In both cases the rates of oxidation are about two orders of magnitude lower than those obtained with NADH as electron donor. 2. The kinetic characteristics of the NADPH oxidase reaction and reduction of acetylpyridine-NAD by NADPH are similar with regard to pH dependences and affinities for NADPH, indicating that both reactions involve the same binding site for NADPH. The binding of NADPH to this site appears to be rate limiting for the overall reactions. 3. At redox equilibrium NADPH and NADH reduce FMN and iron-sulphur center 1 of NADH dehydrogenase to the same extents. The rate of reduction of FMN by NADPH is at least two orders of magnitude lower than with NADH. 4. It is concluded that NADPH is a substrate of NADH dehydrogenase and that the nicotinamide nucleotide is oxidized by submitochondrial particles via the NADH--binding site of the enzyme.     

2-Phenyl-beta-lapachone can affect mitochondrial function by redox cycling mediated oxidation

2-Phenyl-beta-lapachone (3,4-dihydro-2-methyl-2-phenyl-2H-naphtho[1,2b]pyran-5,6-dione) (2PBL) is a o-naphthoquinone synthesized as a possible antitumoral agent. The addition of micromolar concentrations of 2PBL to rat liver mitochondria (in the presence of malate-glutamate or succinate, as respiratory substrates): (1) stimulated O(2) consumption in state 4 and inhibited O(2) consumption in state 3, thus decreasing respiratory control index (RCI); and (2) collapsed the mitochondrial membrane potential. The addition of 2PBL to rat liver submitochondrial particles: (1) stimulated NADH oxidation in the presence of rotenone, antimycin, myxothiazol or cyanide; (2) stimulated (.-)O(2)(-) production in the presence of NADH and antimycin; and (3) led to 2PBL semiquinone radical production. Control studies carried out with two p-naphthoquinones, menadione and atovaquone, did not produced equivalent effects. These findings support the hypothesis that 2PBL, undergoes redox cycling and affects mitochondrial function. The 2PBL effect is complex, involving inhibition of electron transfer, uncoupling of oxidative phosphorylation, collapse of mitochondrial membrane potential and (.-)O(2)(-) production by redox cycling. The mitochondrion could be a target organelle for 2PBL cytotoxicity.     

Studies of the mitochondria from Eimeria tenella and inhibition of the electron transport by quinolone coccidiostats.

Intact but fragile mitochondria were isolated from unsporulated oocysts of Eimeria tenella. The mitochondria respired in response to succinate, malate plus pyruvate, and L-ascorbate at rates of 1.00, 0.40, and 0.25 mu1 O2/min/mg protein, respectively. Spectrophotometric analyses of the cytochromes in mitochondria and whole oocysts revealed b-type and o-type cytochromes, at roughly similar levels, but no cytochrome c could be detected. The mitochondrial respiration was inhibited by cyanide, azide, carbon monoxide, antimycin A, and 2-heptyl-4-hydroxyquinoline-N-oxide, but was relatively resistant to rotenone and amytal. The quinolone coccidiostats buquinolate, amquinate, methyl benzoquate, and decoquinate were identified as very powerful inhibitiors of succinate and malate plus pyruvate supported respiration in E. tenella mitochondria. None of these four drugs exhibited any inhibitory effect on chicken liver mitochondria. Only 3 pmol of the quinolones per mg mitochondrial protein was needed to achieve 50% inhibition. The inhibition could not be reversed by coenzymes Q6 or Q10. Since the quinolones did not affect L-ascorbate-supported respiration or the activities of submitochondrial succinate dehydrogenase and NADH dehydrogenase, the site of action of the quinolone coccidiostats was tentatively identified as probably near cytochrome b in E. tenella mitochondria. Mitochondria isolated from an E. tenella amquinate-resistant mutant were much less susceptible to quinolone coccidiostats; 50% inhibition was attained by 300 pmol of the drugs/mg mitochondrial protein. The results suggest that the mechanisms of action of quinolone coccidiostats is by inhibiting the cytochrome-mediated electron transport in the mitochondria of coccidia. 2-Hydroxynaphthoquinone coccidiostats were identified as inhibitors of mitochondrial respiration of both E. tenella and chicken liver. They inhibited submitochondrial succinate dehydrogenase and NADH dehydrogenase of E. tenella, and remained equally active against the mitochondrial function of E. tenella amquinolate-resistant mutant.     

The interaction of rubroskyrin, a modified bis-anthraquinone pigment fromPenicillium islandicum Sopp, with respiratory chain of liver mitochondria

Summary  Rubroskyrin, a modified bisanthraquinone pigment from an yellow rice moldPenicillium islandicum Sopp, was examined for its redox-interaction with the mitochondrial respiratory chain by using rat liver submitochondrial particles (SMP) and was compared with luteoskyrin and rugulosin. Rubroskyrin showed a redox interaction with the NAD-linked respiratory chain of SMP, promoting NADH oxidase in the presence of rotenone, a specific inhibitor to coupling site I of the respiratory chain. Rubroskyrin-mediated NADH oxidase was not inhibited by antimycin A and cyanide, inhibitors to coupling sites II and III, respectively, indicating a generation of an electron transport shunt from a rotenone-insensitive site of NADH dehydrogenase (complex I) to dissolved oxygen. An electrontransport shunt to cytochromec oxidase from complex I was also observed in the experiment with cytochromec and antimycin A. Rubroskyrin did not interact with succinate-linked respiratory chain. Such enzymatic redox response which generates electron transport shunt was not detected for luteoskyrin and rugulosin in the present study.     

Relation of superoxide generation and lipid peroxidation to the inhibition of NADH-Q oxidoreductase by rotenone, piericidin A, and MPP+.

The addition of NADH to submitochondrial particles inhibited by agents which interrupt electron transport from NADH-Q oxidoreductase (Complex I) to Q10 (rotenone, piericidin A, and MPP+) results in superoxide formation and lipid peroxidation. A study of the quantitative relations now shows that oxyradical formation does not appear to be the direct result of the inhibition. Although tetraphenyl boron (TPB) greatly enhances the inhibition by MPP+, it has no effect on O2. formation or lipid peroxidation. When submitochondrial particles completely inhibited by rotenone or piericidin A are treated with bovine serum albumin to remove spuriously bound inhibitor molecules without affecting those bound at the specific inhibition site, NADH-Q activity remains inhibited and lipid peroxidation occurs but superoxide formation ceases. Thus oxyradical formation may be the result of the binding of inhibitors at sites in the membrane other than those related to the inhibition of electron transport.     

Lipid peroxidation and the reduction of ADP-Fe3+ chelate by NADH-ubiquinone reductase preparation from bovine heart mitochondria.

The NADH-ubiquinone reductase preparation (Complex I) of bovine hart mitochondria catalysed in the presence of reduced coenzymes and ADP-Fe3+ the lipid peroxidation of liposomes prepared from mitochondrial lipids. The apparent Km values for the coenzymes and the optimal pH of the reactions agreed well with those of the lipid peroxidation of the submitochondrial particles treated with rotenone. On assay of the reduction of ADP-Fe3+ chelate by the reduction of cytochrome c in the presence of superoxide dismutase and antimycin A or by the oxidation of reduced coenzymes, the reactions were not affected by rotenone but were inhibited by thiol-group inhibitors. The properties of the ADP-Fe3+ reductase activity were highly consistent with those of the lipid-peroxidation reaction. These observations suggest that electrons from reduced coenzymes are transferred to ADP-Fe3+ chelate from a component between a mercurial-sensitive site and the rotenone-sensitive one of the NADH dehydrogenase and that the reduction of ADP-Fe3+ chelate by the NADH dehydrogenase is an essential step in the lipid peroxidation.     

Quantitative determination of the rate of superoxide radical formation in mitochondrial membranes by electron paramagnetic resonance

The reaction of hydroxylamine (1-hydroxy-2,2,6,6-tetramethyl-4-oxopiperidine) with O2-. resulting in the stable nitroxyl radical formation recorded by ESR-technique was applied to estimate quantitatively the rate of O2-. superoxide radical generation (VO2.-) by submitochondrial particles (SMP) of liver (of mice and rats). The VO2.- dependence on concentrations of NADH, succinate and protein of SMP was established. The method allows detecting VO2.- greater than 0.05 nmol.min-1.ml-1. It has been shown that in the NADH-dependent site of the chain VO2.- is 3-4 times that in the succinate-dependent site. In the presence of rhotenone and antimycin A VO2.- increases by 30-35%, while cyanide retards VO2.- by about 30%. The data comparison with regard to VO2.- and O2 absorption rate polarographically determined has indicated that about 2% of the absorbed O2 is consumed to form O2-.     

Redox cycling of anthracyclines by cardiac mitochondria. II. Formation of superoxide anion, hydrogen peroxide, and hydroxyl radical

In the accompanying paper (Davies, K. J. A., and Doroshow, J. A. (1986) J. Biol. Chem. 261, 3060-3067), we have demonstrated that anthracycline antibiotics are reduced to the semiquinone form at Complex I of the mitochondrial electron transport chain. In the experiments presented in this study we examined the effects of doxorubicin (Adriamycin), daunorubicin, and related quinonoid anticancer agents on superoxide, hydrogen peroxide, and hydroxyl radical production by preparations of beef heart submitochondrial particles. Superoxide anion formation was stimulated from (mean +/- S.E.) 1.6 +/- 0.2 to 69.6 +/- 2.7 or 32.1 +/- 1.5 nmol X min-1 X mg-1 by the addition of 90 microM doxorubicin or daunorubicin, respectively. However, the anthracycline 5-iminodaunorubicin, in which an imine group has been substituted in the C ring quinone moiety, did not increase superoxide production over control levels. In the presence of rotenone, initial rates of oxygen consumption and superoxide formation were identical under comparable experimental conditions. Furthermore, H2O2 production increased from undetectable control levels to 2.2 +/- 0.3 nmol X min-1 X mg-1 after treatment of submitochondrial particles with doxorubicin (200 microM). The hydroxyl radical, or a related chemical oxidant, was also detected after the addition of an anthracycline to this system by both ESR spectroscopy using the spin trap 5,5-dimethylpyrroline-N-oxide and by gas chromatographic quantitation of CH4 produced from dimethyl sulfoxide. Hydroxyl radical production, which was iron-dependent in this system, occurred in a nonlinear fashion with an initial lag phase due to a requirement for H2O2 accumulation. We also found that two quinonoid anti-cancer agents which produce less cardiotoxicity than the anthracyclines, mitomycin C, and mitoxantrone, stimulated significantly less or no hydroxyl radical production by submitochondrial particles. These experiments suggest that injury to cardiac mitochondria which is produced by anthracycline antibiotics may result from the generation of the hydroxyl radical during anthracycline metabolism by NADH dehydrogenase.     

EPR studies at 20° K on the mitochondrial respiratory chain

EPR spectrometry at 20°K of oxidized phosphorylating submitochondrial particles has revealed new paramagnetic species, with lines at g = 2.014 (centre) and 1.990 (trough), respectively. The reduction by NADH of the iron-sulphur centre 2 (N.R. Orme-Johnson, W.H. Orme-Johnson, R.E. Hansen, H. Beinert and Y. Hatefi, Proc. Second International Symp. on Oxidases and Related Oxidation-Reduction Systems, Memphis, Tennessee, 1971, in the press) of NADH dehydrogenase, with lines at g = 2.052 and 1.922, is unaffected by rotenone. Succinate also partially reduces this species in phosphorylating sub-mitochondrial particles. An additional species with lines at g = 2.027 (top) and 1.886 is also seen in reduced particles.     

The kinetics of NADH oxidation by complex I of isolated plant mitochondria

The oxidation of matrix NADH in the presence and absence of rotenone was investigated in submitochondrial particles prepared from purified beetroot ( Beta vulgaris L.) mitochondria. The submitochondrial particles oxidised NADH using oxygen and artificial electron acceptors such as ferricyanide (FeCN) and short-chain analogues of ubiquinone(UQ)-10, although the NADH-FeCN reductase activity was not inhibited by rotenone. NADH-oxygen reductase activity in the presence and absence of rotenone displayed different affinities for NADH (145 ± 37 and 24 ± 9 μ M , respectively). However, in the presence of 0.15 m M UQ-1 where any contribution from non-specific sites of UQ-reduction was minimal, the rotenone-insensitive oxygen uptake was stimulated dramatically and the K_m(NADH) decreased from 167 ± 55 μ M to 11 ± 1 μ M ; a value close to that determined for the total oxygen uptake which itself was virtually unaffected by the addition of UO-1 [K_m(NADH) of 13 ± 3 μ M ).
The similar affinity of NADH-oxygen reductase for NADH when UQ-1 was present in both the presence and absence of rotenone, suggested that there may be only one NADH binding site involved in the two activities. A quantitative two-stage model for Complex I is postulated with one NADH binding site and two sites of UQ-reduction (one of which is insensitive to rotenone) with a common intermediate 'P' whose level of reduction can influence the NADH binding site. The poor affinity that rotenone-insensitive NADH-oxygen reductase activity displayed for NADH results from a limitation on the interaction of its UQ-reduction site with UQ-10 in the membrane; possibly from a low concentration of UQ-10 around this site or from steric hindrance restricting the access of UQ-10 to this reduction site.     

Generation of superoxide by the mitochondrial Complex I

Superoxide production by inside-out coupled bovine heart submitochondrial particles, respiring with succinate or NADH, was measured. The succinate-supported production was inhibited by rotenone and uncouplers, showing that most part of superoxide produced during succinate oxidation is originated from univalent oxygen reduction by Complex I. The rate of the superoxide (O2*-)) production during respiration at a high concentration of NADH (1 mM) was significantly lower than that with succinate. Moreover, the succinate-supported O2*- production was significantly decreased in the presence of 1 mM NADH. The titration curves, i.e., initial rates of superoxide production versus NADH concentration, were bell-shaped with the maximal rate (at 50 microM NADH) approaching that seen with succinate. Both NAD+ and acetyl-NAD+ inhibited the succinate-supported reaction with apparent Ki's close to their Km's in the Complex I-catalyzed succinate-dependent energy-linked NAD+ reduction (reverse electron transfer) and NADH:acetyl-NAD+ transhydrogenase reaction, respectively. We conclude that: (i) under the artificial experimental conditions the major part of superoxide produced by the respiratory chain is formed by some redox component of Complex I (most likely FMN in its reduced or free radical form); (ii) two different binding sites for NADH (F-site) and NAD+ (R-site) in Complex I provide accessibility of the substrates-nucleotides to the enzyme red-ox component(s); F-site operates as an entry for NADH oxidation, whereas R-site operates in the reverse electron transfer and univalent oxygen reduction; (iii) it is unlikely that under the physiological conditions (high concentrations of NADH and NAD+) Complex I is responsible for the mitochondrial superoxide generation. We propose that the specific NAD(P)H:oxygen superoxide (hydrogen peroxide) producing oxidoreductase(s) poised in equilibrium with NAD(P)H/NAD(P)+ couple should exist in the mitochondrial matrix, if mitochondria are, indeed, participate in ROS-controlled processes under physiologically relevant conditions.     

On the sidedness of the ubiquinone redox cycle. Kinetic studies in mitochondrial membranes

The inhibition of NADH oxidation but not of succinate oxidation by the low ubiquinone homologs UQ-2 and UQ-3 is not due to a lower rate of reduction of ubiquinone by NADH dehydrogenase: experiments in submitochondrial particles and in pentane-extracted mitochondria show that UQ-3 is reduced at similar rates using either NADH or succinate as substrates. The fact that reduced UQ-3 cannot be reoxidized when reduced by NADH but can be reoxidized when reduced by succinate may be explained by a compartmentation of ubiquinone.Using reduced ubiquinones as substrates of ubiquinol oxidase activity in intact mitochondria and in submitochondrial particles we found that ubiquinol-3 is oxidized at higher rates in submitochondrial particles than in mitochondria. The initial rates of ubiquinol oxidation increased with increasing lengths of isoprenoid side chains in mitochondria, but decreased in submitochondrial particles. These findings suggest that the site of oxidation of reduced ubiquinone is on the matrix side of the membrane; reduced ubiquinones may reach their oxidation site in mitochondria only crossing the lipid bilayer: the rate of diffusion of ubiquinol-3 is presumably lower than that of ubiquinol-7 due to the differences in hydrophobicity of the two quinones.     

Spectroscopic studies of flavoproteins and non-haem iron proteins of submitochondrial particles of Torulopsis utilis modified by iron- and sulphate-limited growth in continuous culture

1. A spectroscopic resolution has been made of the components contributing to the ;iron-flavoprotein' trough extending from 450 to 520nm in the reduced-minus-oxidized difference spectrum of submitochondrial particles of Torulopsis utilis. 2. Seven components were identified other than cytochrome b, ubiquinone and succinate dehydrogenase. On the basis of the effects of iron- and sulphate-limited growth of cells on their subsequently derived electron-transport particles, and also by consideration of analytical measurements of the concentration of FMN, FAD, non-haem iron and acid-labile sulphide in the electron-transport particles in relation to the magnitude of the spectroscopic changes, it was possible to identify five of these components as follows: species 1a, the flavin of NADH dehydrogenase ferroflavoprotein; species 1b, the iron-sulphur component of NADH dehydrogenase ferroflavoprotein; species 1', the flavin of an NADPH dehydrogenase; species 2, an iron-sulphur or ferroflavoprotein component; species 3, the flavin of l-3-glycerophosphate dehydrogenase. Two additional components were a fluorescent flavoprotein, probably lipoamide dehydrogenase, and a b-type cytochrome reducible by NADH or NADPH but not reoxidizable by the respiratory chain. 3. Species 1b and 2 were undetectable in electron-transport particles from iron- or sulphate-limited cells, but could be recovered in vivo under non-growing conditions. 4. The recovery in vivo of species 2 but not species 1b was inhibited by cycloheximide. 5. The recovery of species 1b correlates with the recovery of site 1 conservation. 6. The recovery of species 1b with species 2 correlates with the recovery of piericidin A sensitivity. 7. Evidence is presented for an NADPH dehydrogenase distinct from NADH dehydrogenase. The oxidation of NADH and NADPH by the respiratory chain is sensitive to piericidin A, and an iron-sulphur protein common to both pathways (species 2) is suggested as the piericidin A-sensitive component. 8. The approximate E'(0) (pH7.0) values of species 1 (a and b, low potential) and species 2 (high potential) indicate that site 1 energy conservation occurs between the levels of species 1 (a and b) and species 2.