Structural and functional organization of Complex I in the mitochondrial respiratory chain

Metabolic flux control analysis of NADH oxidation in bovine heart submitochondrial particles revealed high flux control coefficients for both Complex I and Complex III, suggesting that the two enzymes are functionally associated as a single enzyme, with channelling of the common substrate, Coenzyme Q. This is in contrast with the more accepted view of a mobile diffusable Coenzyme Q pool between these enzymes. Dilution with phospholipids of a mitochondrial fraction enriched in Complexes I and III, with consequent increased theoretical distance between complexes, determines adherence to pool behavior for Coenzyme Q, but only at dilution higher than 1:5 (protein:phospholipids), whereas, at lower phospholipid content, the turnover of NADH cytochrome c reductase is higher than expected by the pool equation.     

The interaction between mitochondrial NADH-ubiquinone oxidoreductase and ubiquinol-cytochrome c oxidoreductase. Restoration of ubiquinone-pool behaviour.

1. In the inner mitochondrial membrane, dehydrogenases and cytochromes appear to act independently of each other, and electron transport has been proposed to occur through a mobile pool of ubiquinone-10 molecules [Kröger & Klingenberg (1973) Eur. J. Biochem. 34, 358--368]. 2. Such behaviour can be restored to the interaction between purified Complex I and Complex III by addition of phospholipid and ubiquinone-10 to a concentrated mixture of the Complexes before dilution. 3. A model is proposed for the interaction of Complex I with Complex III in the natural membrane that emphasizes relative mobility of the Complexes rather than ubiquinone-10. Electron transfer occurs only through stoicheiometric Complex I-Complex III units, which, however, are formed and re-formed at rates higher than the rate of electron transfer.     

Mitochondrial Complex I: structure, function, and implications in neurodegeneration

Mitochondrial Complex I (NADH Coenzyme Q oxidoreductase) is the least understood of respiratory complexes. In this review we emphasize some novel findings on this enzyme that are of relevance to the pathogenesis of neurodegenerative diseases. Besides Coenzyme Q (CoQ), also oxygen may be an electron acceptor from the enzyme, with generation of superoxide radical in the mitochondrial matrix. The site of superoxide generation is debated: we present evidence based on the rational use of several inhibitors that the one-electron donor to oxygen is an iron-sulphur cluster, presumably N2. On this assumption we present a novel mechanism of electron transfer to the acceptor, CoQ. Strong evidence is accumulating that electron transfer from Complex I to Complex III via CoQ is not performed by operation of the CoQ pool but by direct channelling within a super-complex including Complex I, Complex III and bound CoQ. Besides structural evidence of a Complex I -Complex III aggregate obtained by native electrophoresis, we have obtained kinetic evidence based on metabolic flux analysis, demonstrating that Complexes I and III behave as an individual enzyme. Quantitative and qualitative changes of phospholipids, including peroxidation, may affect the supercomplex formation. Complex I is deeply involved in pathological changes, including neurodegeneration. Maternally inherited mutations in mitochondrial DNA genes encoding for Complex I subunits are at the basis of Leber's Hereditary Optic Neuropathy; a decrease of electron transfer in the complex, due to the mutations, is not sufficient per se to explain the clinical phenotype, and other factors including proton translocation and oxygen radical generation have been considered of importance. Complex I changes are also involved in more common neurological diseases of the adult and old ages. In this review we discuss Parkinson's disease, where the pathogenic involvement of Complex I is better understood; the accumulated evidence on the mode of action of Complex I inhibitors and their effect on oxygen radical generation is discussed in terms of the aetiology and pathogenesis of the disease.     

Supercomplex organization of the mitochondrial respiratory chain and the role of the Coenzyme Q pool: pathophysiological implications

In this review we examine early and recent evidence for an aggregated organization of the mitochondrial respiratory chain. Blue Native Electrophoresis suggests that in several types of mitochondria Complexes I, III and IV are aggregated as fixed supramolecular units having stoichiometric proportions of each individual complex. Kinetic evidence by flux control analysis agrees with this view, however the presence of Complex IV in bovine mitochondria cannot be demonstrated, presumably due to high levels of free Complex. Since most Coenzyme Q appears to be largely free in the lipid bilayer of the inner membrane, binding of Coenzyme Q molecules to the Complex I-III aggregate is forced by its dissociation equilibrium; furthermore free Coenzyme Q is required for succinate-supported respiration and reverse electron transfer. The advantage of the supercomplex organization is in a more efficient electron transfer by channelling of the redox intermediates and in the requirement of a supramolecular structure for the correct assembly of the individual complexes. Preliminary evidence suggests that dilution of the membrane proteins with extra phospholipids and lipid peroxidation may disrupt the supercomplex organization. This finding has pathophysiological implications, in view of the role of oxidative stress in the pathogenesis of many diseases.     

Structural and functional organization of the mitochondrial respiratory chain: A dynamic super-assembly

The structural organization of the mitochondrial oxidative phosphorylation (OXPHOS) system has received large attention in the past and most investigations led to the conclusion that the respiratory enzymatic complexes are randomly dispersed in the lipid bilayer of the inner membrane and functionally connected by fast diffusion of smaller redox components, Coenzyme Q and cytochrome c. More recent investigations by native gel electrophoresis, however, have shown the existence of supramolecular associations of the respiratory complexes, confirmed by electron microscopy analysis and single particle image processing. Flux control analysis has demonstrated that Complexes I and III in mammalian mitochondria and Complexes I, III, and IV in plant mitochondria kinetically behave as single units with control coefficients approaching unity for each single component, suggesting the existence of substrate channelling within the supercomplexes. The reasons why the presence of substrate channelling for Coenzyme Q and cytochrome c was overlooked in the past are analytically discussed. The review also discusses the forces and the conditions responsible for the formation of the supramolecular units. The function of the supercomplexes appears not to be restricted to kinetic advantages in electron transfer: we discuss evidence on their role in the stability and assembly of the individual complexes and in preventing excess oxygen radical formation. Finally, there is increasing evidence that disruption of the supercomplex organization leads to functional derangements responsible for pathological changes.     

Functional role of mitochondrial respiratory supercomplexes

Recent experimental evidence has replaced the random diffusion model of electron transfer with a model of supramolecular organisation based upon specific interactions between individual respiratory complexes. These supercomplexes were found to be functionally relevant by flux control analysis and to confer a kinetic advantage to NAD-linked respiration (channelling). However, the Coenzyme Q pool is still required for FAD-linked oxidations and for the proper equilibrium with Coenzyme Q bound in the supercomplex. Channelling in the cytochrome c region probably also occurs but does not seem to confer a particular kinetic advantage. The supramolecular association of individual complexes strongly depends on membrane lipid amount and composition and is affected by lipid peroxidation; it also seems to be modulated by membrane potential and protein phosphorylation. Additional properties of supercomplexes are stabilisation of Complex I, as evidenced by the destabilising effect on Complex I of mutations in either Complex III or IV, and prevention of excessive generation of reactive oxygen species. The dynamic character of the supercomplexes allows their involvement in metabolic adaptations and in control of cellular signalling pathways. This article is part of a Special Issue entitled: Dynamic and ultrastructure of bioenergetic membranes and their components.     

Mitochondrial Complex I: structural and functional aspects

This review examines two aspects of the structure and function of mitochondrial Complex I (NADH Coenzyme Q oxidoreductase) that have become matter of recent debate. The supramolecular organization of Complex I and its structural relation with the remainder of the respiratory chain are uncertain. Although the random diffusion model [C.R. Hackenbrock, B. Chazotte, S.S. Gupte, The random collision model and a critical assessment of diffusion and collision in mitochondrial electron transport, J. Bioenerg. Biomembranes 18 (1986) 331-368] has been widely accepted, recent evidence suggests the presence of supramolecular aggregates. In particular, evidence for a Complex I-Complex III supercomplex stems from both structural and kinetic studies. Electron transfer in the supercomplex may occur by electron channelling through bound Coenzyme Q in equilibrium with the pool in the membrane lipids. The amount and nature of the lipids modify the aggregation state and there is evidence that lipid peroxidation induces supercomplex disaggregation. Another important aspect in Complex I is its capacity to reduce oxygen with formation of superoxide anion. The site of escape of the single electron is debated and either FMN, iron-sulphur clusters, and ubisemiquinone have been suggested. The finding in our laboratory that two classes of hydrophobic inhibitors have opposite effects on superoxide production favours an iron-sulphur cluster (presumably N2) is the direct oxygen reductant. The implications in human pathology of better knowledge on these aspects of Complex I structure and function are briefly discussed.     

Mitochondrial Complex I defects in aging

According to the 'mitochondrial theory of aging' it is expected that the activity of NADH Coenzyme Q reductase (Complex I) would be most severely affected among mitochondrial enzymes, since mitochondrial DNA encodes for 7 subunits of this enzyme. Being these subunits the site of binding of the acceptor substrate (Coenzyme Q) and of most inhibitors of the enzyme, it is also expected that subtle kinetic changes of quinone affinity and enzyme inhibition could develop in aging before an overall loss of activity would be observed.The overall activity of Complex I was decreased in several tissues from aged rats, nevertheless it was found that direct assay of Complex I using artificial quinone acceptors may underevaluate the enzyme activity. The most acceptable results could be obtained by applying the 'pool equation' to calculate Complex I activity from aerobic NADH oxidation; using this method it was found that the decrease in Complex I activity in mitochondria from old animals was greater than the activity calculated by direct assay of NADH Coenzyme Q reductase.A decrease of NADH oxidation and its rotenone sensitivity was observed in nonsynaptic mitochondria, but not in synaptic 'light' and 'heavy' mitochondria of brain cortex from aged rats.In a study of Complex I activity in human platelet membranes we found that the enzyme activity was unchanged but the titre for half-inhibition by rotenone was significantly increased in aged individuals and proposed this change as a suitable biomarker of aging and age-related diseases. (Mol Cell Biochem 174: 329–333, 1997)     

The role of Coenzyme Q in mitochondrial electron transport

In mitochondria, most Coenzyme Q is free in the lipid bilayer; the question as to whether tightly bound, non-exchangeable Coenzyme Q molecules exist in mitochondrial complexes is still an open question.We review the mechanism of inter-complex electron transfer mediated by ubiquinone and discuss the kinetic consequences of the supramolecular organization of the respiratory complexes (randomly dispersed vs. super-complexes) in terms of Coenzyme Q pool behavior vs. metabolic channeling, respectively, both in physiological and in some pathological conditions. As an example of intra-complex electron transfer, we discuss in particular Complex I, a topic that is still under active investigation.     

Structural organization of the mitochondrial respiratory chain

Two models exist of the mitochondrial respiratory chain: the model of a random organization of the individual respiratory enzyme complexes and that of a super-complex assembly formed by stable association between the individual complexes. Recently Sch?gger, using digitonin solubilization and Blue Native PAGE produced new evidence of preferential associations, in particular a Complex I monomer with a Complex III dimer, and suggested a model of the respiratory chain (the respirasome) based on direct electron channelling between complexes. Discrimination between the two models is amenable to kinetic testing using flux control analysis. Experimental evidence obtained in beef heart SMP, according to the extension of the Metabolic Control Theory for pathways with metabolic channelling, showed that enzyme associations involving Complex I and Complex III take place in the respiratory chain while Complex IV seems to be randomly distributed, with cytochrome c behaving as a mobile component. Flux control analysis at anyone of the respiratory complexes involved in aerobic succinate oxidation indicated that Complex II and III are not functionally associated in a stable supercomplex. A critical appraisal of the solid-state model of the mitochondrial respiratory chain requires its reconciliation with previous biophysical and kinetic evidence that CoQ behaves as a homogeneous diffusible pool between all reducing enzyme and all oxidizing enzymes: the hypothesis can be advanced that both models (CoQ pool and supercomplexes) are true, by postulating that supercomplexes physiologically exist in equilibrium with isolated complexes depending on metabolic conditions of the cell.     

Regulation of electron transfer by the quinone pool

Strong evidence for a random collisional mechanism for ubiquinone-mediated electron transfer is provided by the characteristic kinetic properties of respiratory chains originally explored by Kröger, A., and Klingenberg, M. (1973),Eur. J. Biochem. 34, 313–323. A kinetic model which leads to this so-called simple Q-pool behavior has been described and we use this in reviewing evidence that electron transfer is diffusion-controlled as well as diffusion-coupled. We also consider mechanisms by which the kinetics of electron transfer might deviate from simple Q-pool behavior and how these might be implicated in the regulation of electron transport.     

Modes of coenzyme Q function in electron transfer

Summary In the mitochondrial respiratory chain, coenzyme Q acts in different ways. A diffusable coenzyme Q pool as a common substrate-like intermediate links the low-potential complexes with complex III. Its diffusion in the lipids is not rate-limiting for electron transfer, but its content is not saturating for maximal rate of NADH oxidation. Protein-bound coenzyme Q is involved in energy conservation, and may be part of enzyme supercomplexes, as in succinate cytochromec reductase. The reason for lack of kinetic saturation of the respiratory chain by quinone concentration is in the low extent of solubility of monomeric coenzyme Q in the membrane lipids. Assays of respiratory enzymes are performed using water soluble coenzyme Q homologs and analogs; several problems exist in using oxidized quinones as acceptors of coenzyme Q reductases. In particular, for complex I no acceptor appears to favorably substitute the endogenous quinone. In addition, quinone reduction sites in complex III compete with the sites in the dehydrogenases, particularly when using duroquinone. The different extent by which these sites operate when different donor substrates (NADH, succinate, glycerol-3-phosphate) are used is best explained by different exposure of the quinone acceptor sites in the dehydrogenases.     

The interaction between mitochondrial NADH-ubiquinone oxidoreductase and ubiquinol-cytochrome c oxidoreductase. Evidence for stoicheiometric association.

1. The NADH-ubiquinone oxidoreductase complex (Complex I) and the ubiquinol-cytochrome c oxidoreductase complex (Complex III) combine in a 1:1 molar ratio to give NADH-cytochrome c oxidoreductase (Complex I-Complex III). 2. Experiments on the inhibition of the NADH-cytochrome c oxidoreductase activity of mixtures of Complexes I and III by rotenone and antimycin indicate that electron transfer between a unit of Complex I-Complex III and extra molecules of Complexes I or III does not contribute to the overall rate of cytochrome c reduction. 3. The reduction by NADH of the cytochrome b of mixtures of Complexes I and III is biphasic. The extents of the fast and slow phases of reduction are determined by the proportion of the total Complex III specifically associated with Complex I. 4. Activation-energy measurements suggest that the structural features of the Complex I-Complex III unit promote oxidoreduction of endogenous ubiquinone-10.     

Photolabile ubiquinone analogues for identification and characterization of quinone binding sites in proteins

Quinones are essential components in most cell and organelle bioenergetic processes both for direct electron and/or proton transfer reactions but also as means to regulate various bioenergetic processes by sensing cell redox states. To understand how quinones interact with proteins, it is important to have tools for identifying and characterizing quinone binding sites. In this work three different photo-reactive azidoquinones were synthesized, two of which are novel compounds, and the methods of synthesis was improved. The reactivity of the azidoquinones was first tested with model peptides, and the adducts formed were analyzed by mass spectrometry. The added mass detected was that of the respective azidoquinone minus N₂. Subsequently, the biological activity of the three azidoquinones was assessed, using three enzyme systems of different complexity, and the ability of the compounds to inactivate the enzymes upon illumination with long wavelength UV light was investigated. The soluble flavodoxin-like protein WrbA could only use two of the azidoquinones as substrates, whereas respiratory chain Complexes I and II could utilize all three compounds as electron acceptors. Complex II, purified in detergent, was very sensitive to illumination also in the absence of azidoquinones, making the ‘therapeutic window’ in that enzyme rather narrow. In membrane bound Complex I, only two of the compounds inactivated the enzyme, whereas illumination in the presence of the third compound left enzyme activity essentially unchanged. Since unspecific labeling should be equally effective for all the compounds, this demonstrates that the observed inactivation is indeed caused by specific labeling.     

Cytochrome c-mediated electron transfer between ubiquinol-cytochrome c reductase and cytochrome c oxidase. Kinetic evidence for a mobile cytochrome c pool.

Ubiquinol oxidase has been reconstituted from ubiquinol-cytochrome c reductase (Complex III), cytochrome c and cytochrome c oxidase (Complex IV). The steady-state level of reduction of cytochrome c by ubiquinol-2 varies with the molar ratios of the complexes and with the presence of antimycin in a way that can be quantitatively accounted for by a model in which cytochrome c acts as a freely diffusible pool on the membrane. This model was based on that of Kröger & Klingenberg [(1973) Eur. J. Biochem. 34, 358-368] for ubiquinone-pool behaviour. Further confirmation of the pool model was provided by analysis of ubiquinol oxidase activity as a function of the molar ratio of the complexes and prediction of the degree of inhibition by antimycin.     

The mitochondrial respiratory chain is partially organized in a supercomplex assembly: kinetic evidence using flux control analysis

The model of the respiratory chain in which the enzyme complexes are independently embedded in the lipid bilayer of the inner mitochondrial membrane and connected by randomly diffusing coenzyme Q and cytochrome c is mostly favored. However, multicomplex units can be isolated from mammalian mitochondria, suggesting a model based on direct electron channeling between complexes. Kinetic testing using metabolic flux control analysis can discriminate between the two models: the former model implies that each enzyme may be rate-controlling to a different extent, whereas in the latter, the whole metabolic pathway would behave as a single supercomplex and inhibition of any one of its components would elicit the same flux control. In particular, in the absence of other components of the oxidative phosphorylation apparatus (i.e. ATP synthase, membrane potential, carriers), the existence of a supercomplex would elicit a flux control coefficient near unity for each respiratory complex, and the sum of all coefficients would be well above unity. Using bovine heart mitochondria and submitochondrial particles devoid of substrate permeability barriers, we investigated the flux control coefficients of the complexes involved in aerobic NADH oxidation (I, III, IV) and in succinate oxidation (II, III, IV). Both Complexes I and III were found to be highly rate-controlling over NADH oxidation, a strong kinetic evidence suggesting the existence of functionally relevant association between the two complexes, whereas Complex IV appears randomly distributed. Moreover, we show that Complex II is fully rate-limiting for succinate oxidation, clearly indicating the absence of substrate channeling toward Complexes III and IV.     

Association of ferrochelatase with Complex I in bovine heart mitochondria

The location of ferrochelatase in bovine heart mitochondria has been studied. When the mitochondria were fractionated into Complexes I, II and III, ferrochelatase activity was only found in Complex I. Complex I also showed heme synthesis from ferric ion in the presence of NADH as an electron donor. Immunoblot experiments confirmed the presence of ferrochelatase in Complex I, but not in Complexes II or III. Some phospholipids, including phosphatidylserine and cardiolipin, stimulated NADH-dependent heme synthesis from ferric ion. When purified ferrochelatase was incubated with the low molecular weight form of NADH dehydrogenase prepared from Complex I, heme synthesis from ferric ion occurred by the addition of NADH. FMN markedly elevated the synthesis. These results indicate that ferrous ion is produced by NADH oxidation in Complex I and is then utilized for heme synthesis by ferrochelatase.     

Differential effects of mitochondrial Complex I inhibitors on production of reactive oxygen species

We have investigated the production of reactive oxygen species (ROS) by Complex I in isolated open bovine heart submitochondrial membrane fragments during forward electron transfer in presence of NADH, by means of the probe 2′,7′-Dichlorodihydrofluorescein diacetate. ROS production by Complex I is strictly related to its inhibited state. Our results indicate that different Complex I inhibitors can be grouped into two classes: Class A inhibitors (Rotenone, Piericidin A and Rolliniastatin 1 and 2) increase ROS production; Class B inhibitors (Stigmatellin, Mucidin, Capsaicin and Coenzyme Q₂) prevent ROS production also in the presence of Class A inhibitors. Addition of the hydrophilic Coenzyme Q₁ as an electron acceptor potentiates the effect of Rotenone-like inhibitors in increasing ROS production, but has no effect in the presence of Stigmatellin-like inhibitors; the effect is not shared by more hydrophobic quinones such as decyl-ubiquinone. This behaviour relates the prooxidant CoQ₁ activity to a hydrophilic electron escape site. Moreover the two classes of Complex I inhibitors have an opposite effect on the increase of NADH-DCIP reduction induced by short chain quinones: only Class B inhibitors allow this increase, indicating the presence of a Rotenone-sensitive but Stigmatellin-insensitive semiquinone species in the active site of the enzyme. The presence of this semiquinone was also suggested by preliminary EPR data. The results suggest that electron transfer from the iron-sulphur clusters (N2) to Coenzyme Q occurs in two steps gated by two different conformations, the former being sensitive to Rotenone and the latter to Stigmatellin.     

The effects of lipid fluidity on the rotational diffusion of complex I and complex III in reconstituted NADH-cytochrome c oxidoreductase

NADH-ubiquinone oxidoreductase (Complex I) can be recombined with ubiquinol-cytochrome c oxidoreductase (Complex III) to reconstitute NADH-cytochrome c oxidoreductase. Two modes of interaction have been found. In one, the Complexes interact stoichiometrically in one to one molar ratios to give a binary Complex I-III unit. In the other, the kinetics of NADH-cytochrome c oxidoreductase are characteristic of 'Q-pool' behaviour seen in intact mitochondria and submitochondrial particles in which the Complexes need not interact directly but can do so via a pool of mobile ubiquinone. Stoichiometric behaviour is found when only boundary layer or annular lipid is present or the lipid is in the gel phase. The lipid is immobile on the ESR time scale and protein rotational diffusion, measured by saturation transfer ESR, is very slow. Q-pool behaviour is found when mobile extra-annular lipid phase is also present. Protein rotational diffusion is rapid and characteristic of a fully disaggregated state. We have also used freeze-fracture electron microscopy of reconstituted NADH-cytochrome c oxidoreductase to monitor protein aggregation and lateral phase separation of lipids and proteins under various conditions. We discuss our findings in relation to models for lateral interactions between respiratory chain enzymes.     

Altered redox status of coenzyme Q9 reflects mitochondrial electron transport chain deficiencies in Caenorhabditis elegans

Mitochondrial disorders are often associated with primary or secondary CoQ10 decrease. In clinical practice, Coenzyme Q10 (CoQ10) levels are measured to diagnose deficiencies and to direct and monitor supplemental therapy. CoQ10 is reduced by complex I or II and oxidized by complex III in the mitochondrial respiratory chain. Therefore, the ratio between the reduced (ubiquinol) and oxidized (ubiquinone) CoQ10 may provide clinically significant information in patients with mitochondrial electron transport chain (ETC) defects. Here, we exploit mutants of Caenorhabditis elegans (C. elegans) with defined defects of the ETC to demonstrate an altered redox ratio in Coenzyme Q9 (CoQ9), the native quinone in these organisms. The percentage of reduced CoQ9 is decreased in complex I (gas-1) and complex II (mev-1) deficient animals, consistent with the diminished activity of these complexes that normally reduce CoQ9. As anticipated, reduced CoQ9 is increased in the complex III deficient mutant (isp-1), since the oxidase activity of the complex is severely defective. These data provide proof of principle of our hypothesis that an altered redox status of CoQ may be present in respiratory complex deficiencies. The assessment of CoQ10 redox status in patients with mitochondrial disorders may be a simple and useful tool to uncover and monitor specific respiratory complex defects.