Protein film voltammetry is used to probe the energetics of electron transfer and substrate binding at the active site of a respiratory flavoenzyme--the membrane-extrinsic catalytic domain of Escherichia coli fumarate reductase (FrdAB). The activity as a function of the electrochemical driving force is revealed in catalytic voltammograms, the shapes of which are interpreted using a Michaelis-Menten model that incorporates the potential dimension. Voltammetric experiments carried out at room temperature under turnover conditions reveal the reduction potentials of the FAD, the stability of the semiquinone, relevant protonation states, and pH-dependent succinate--enzyme binding constants for all three redox states of the FAD. Fast-scan experiments in the presence of substrate confirm the value of the two-electron reduction potential of the FAD and show that product release is not rate limiting. The sequence of binding and protonation events over the whole catalytic cycle is deduced. Importantly, comparisons are made with the electrocatalytic properties of SDH, the membrane-extrinsic catalytic domain of mitochondrial complex II.

Genetic defects affecting the mitochondrial respiratory chain are an important cause of neurological disease. Previously, we identified a family with complex II deficiency and late-onset neurodegenerative disease with progressive optic atrophy, ataxia, and myopathy. The affected family members are now shown to carry a C-to-T transition in one allele of the nuclear gene encoding the flavoprotein subunit of complex II. Mutation of the equivalent base in Escherichia coli generates an inactive enzyme unable to bind flavin adenine dinucleotide covalently. Compatible with these findings, our patients have an approximate 50% decrease in complex II and succinate dehydrogenase activity. These results suggest that genetic defects of nuclear-encoded subunits of the mitochondrial respiratory chain can result in late-onset neurodegenerative disease.

Complex II (succinate:quinone oxidoreductase) of aerobic respiratory chains oxidizes succinate to fumarate and passes the electrons directly into the quinone pool. It serves as the only direct link between activity in the citric acid cycle and electron transport in the membrane. Finer details of these reactions and interactions are but poorly understood. However, complex II has extremely similar structural and catalytic properties to quinol:fumarate oxidoreductases of anaerobic organisms, for which X‐ray structures have recently become available. These offer new insights into structure–function relationships of this class of flavoenzymes, including evidence favoring protein movement during catalysis.

Thyroid-associated ophthalmopathy is an autoimmune disorder of the extraocular muscles and orbital connective tissue, which is usually associated with Graves' hyperthyroidism. Well-studied markers of ophthalmopathy are eye muscle membrane antigens, reportedly of approximately 64-kDa molecular mass. One, originally identified only as the 64-kDa protein, has recently been shown to be the flavoprotein (Fp) subunit of mitochondrial succinate dehydrogenase, which has a correct molecular mass of 67 kDa. We have used purified beef heart Fp as antigen in an enzyme-linked immunosorbent assay for crossreactive human autoantibodies. Sera have been screened from patients with thyroid-associated ophthalmopathy classified according to activity and presence or not of eye muscle disease, and from those with Graves' hyperthyroidism without eye involvement. Also examined were serum samples taken periodically from 20 patients with Graves' hyperthyroidism during 24 months of treatment of their hyperthyroidism with antithyroid drugs. Four of these patients had ophthalmopathy at the onset, 12 developed ophthalmopathy, and 4 did not develop any eye signs during treatment. Anti-Fp subunit antibodies were detected in 73% of patients with active ophthalmopathy and evidence of eye muscle involvement but only in 25% if there was only congestive ophthalmopathy. These values were 0% and 11% for patients with chronic ophthalmopathy, with or without eye muscle dys-function, respectively. The antibodies were also detected in 14% of patients with Graves' hyperthyroidism without evident ophthalmopathy, 11% of patients with nonimmunologic thyroid disorders, 12% of type I diabetics, and 12% of age- and sex-matched normal subjects. Significantly, appearance of anti-Fp antibodies predicted the development of ophthalmopathy in 5 of the 6 patients with Graves' hyperthyroidism, who developed eye muscle dysfunction after treatment of the hyperthyroidism, and coincided with the onset of eye muscle signs in the other patient. Antibodies were not detected in any of 6 patients who developed congestive ophthalmopathy without evidence of eye muscle damage or in 4 patients who did not develop any eye signs. In conclusion we have shown a close relationship between eye muscle disease and serum antibodies against the Fp subunit of succinate dehydrogenase in patients with Graves' hyperthyroidism.

Thyroid-associated ophthalmopathy (TAO) is a progressive eye disorder associated with thyroid autoimmunity, particularly Graves' hyperthyroidism, which is generally considered to have an autoimmune etiology. Eye muscle membrane proteins reportedly of 55 and 64 kDa are the best markers of the ophthalmopathy. The main focus of our recent studies has been to purify the pertinent proteins from porcine eye muscle membranes and characterize them. The 64-kDa protein is now shown from a partial sequence and by Western blotting using specific antibody probes to be the flavoprotein (Fp) subunit of succinate dehydrogenase and to have a correct molecular mass of 67 kDa. The protein was purified and cleaved with cyanogen bromide, and the N-terminal region of an immunoreactive partial peptide was determined. The 20-amino acid porcine sequence so obtained matched one within the Fp subunits of human and bovine succinate dehydrogenases in 20 and 18 of these positions, respectively. Succinate dehydrogenase is both a citric acid cycle enzyme and a component (complex II) of the mitochondrial respiratory chain. It is thus essential for aerobic energy production and is highly conserved. The mature human and bovine Fp subunits are 92% homologous and have a molecular mass of approximately 67 kDa, the same as our redetermined value for the 64-kDa marker protein. Sera from patients with TAO and from those with Graves' hyperthyroidism without evident ophthalmopathy highlighted the 64-kDa marker protein in crude porcine eye muscle membranes and the Fp subunit of highly purified bovine succinate dehydrogenase at the identical position on Western blots. Anti-beef Fp antibodies were detected in sera from 67% of patients with active TAO of more than 1-yr duration, in 30% with stable TAO of more than 3-yr duration, and in 30% of patients with Graves' hyperthyroidism without ophthalmopathy, but in only 7% of age- and sex-matched normal subjects. As succinate dehydrogenase is bound to the matrix (inside) surface of the mitochondrial inner membrane, it is unlikely to be accessible to circulating autoantibodies. We would postulate that eye muscle damage in ophthalmopathy is probably caused by cytotoxic antibodies or CD+ T lymphocytes targeting a cell membrane antigen, such as the thyroid and eye muscle shared protein G2s, and that presentation of succinate dehydrogenase is secondary. On the other hand, an autoantibody response to succinate dehydrogenase may be a good marker of immune-mediated damage to the eye muscle fiber and may support the idea that the extraocular muscles are targets of the autoimmune reactions of TAO.

Myasthenia gravis is an organ-specific autoimmune disorder generally thought to be caused by an antibody-mediated attack against the skeletal muscle nicotinic acetylcholine (Ach) receptor (AchR) at the neuromuscular junction. Extraocular muscle weakness and double vision are present in about 90% of patients with myasthenia gravis and are the predominant complaints in about 20% of patients, when the condition is called ocular myasthenia gravis (OMG). While serum antibodies against the AchR are detected in most patients with generalized myasthenia gravis (GMG), they are not found in about one-third of patients with the ocular variety, and epidemiological, clinical, and serological studies suggest that OMG and GMG are two separate diseases. Both forms of myasthenia gravis are sometimes associated with thyroid autoimmunity or thyroid-associated ophthalmopathy (TAO). We have therefore tested the sera of patients with GMG and OMG by Western blotting for antibodies against porcine eye muscle membrane proteins in general, and by enzyme-linked immunosorbent assays (ELISA) specifically for reaction with two skeletal muscle antigens which are prominent marker antigens for TAO, namely, the calcium-binding protein calsequestrin and the so-called "64-kDa protein." The 64-kDa protein has recently been identified as the flavoprotein subunit of mitochondrial succinate dehydrogenase. Patients with ophthalmopathy and myasthenia were excluded. Nine of the patients had associated Graves' hyperthyroidism without evident ophthalmopathy and one had Hashimoto's thyroiditis. Antibodies against porcine eye muscle membrane antigens of M(r) 15-110 kDa were detected in patients with GMG or OMG, one or more antibodies being detected in 100% of patients with GMG and in 88% of those with OMG. The most frequently found antibodies were those targeting eye muscle membrane proteins of 15, 67, and 110 kDa. Antibodies reactive with purified calsequestrin (63 kDa) were detected in 21% of patients with OMG but in no patient with GMG. Antibodies recognizing purified succinate dehydrogenase (67 kDa) were found in 42% of patients with OMG, in 100% (5 of 5) of patients with GMG, and in 48% of all patients with myasthenia gravis not associated with Graves' hyperthyroidism. There was no close correlation between any eye muscle-reactive antibody and antibodies against the AchR in either group of myasthenic patients. The findings support the notion that immunoreactivity against skeletal muscle proteins other than the AchR may play a role in the development of the muscle weakness in AchR antibody-negative patients with OMG and GMG, although it is unlikely that any of the antibodies demonstrated in this study are directly implicated. Similarly, while the demonstration of antibodies reactive with eye muscle antigens associated with TAO in patients with OMG raises the possibility that the link between the ocular lesions of myasthenia gravis and Graves' disease may be autoimmunity against a common antigen(s), it is more likely that both disorders are mediated by cytotoxic T cells recognizing another cell membrane antigen, such as the novel thyroid and eye muscle shared protein G2s, and that serum antibodies reactive with succinate dehydrogenase Fp subunit and calsequestrin are markers of an immune-mediated eye muscle reaction.

Protein-Film voltammetry reveals the global effects of H/D isotopic substitutions, both in organic substrates and solvent, on bidirectional catalytic electron transport in a mitochondrial respiratory enzyme, succinate dehydrogenase. The voltammetry is controlled by the enzyme kinetics and therefore provides a direct display of the characteristic relationships between turnover rates (current) and driving force (potential). This enables simultaneous measurement of relative electron-transport rates for oxidative and reductive directions alongside thermodynamic data (reduction potentials) relating to substrates and active sites. Measured over a range of pH and pD, the relationships between effective catalytic potentials of the enzyme and formal reduction potentials for substrate and active-site yield a ''potential-domain'' description of the catalytic energetics, which complements and extends the picture obtained by conventional kinetic methods. For the organic substrates, a marked decrease is observed in the rate of oxidation of perdeuteriosuccinate compared to-succinate, but there is no change upon deuterating fumarate. Can changing the solvent from H2O to D2O there is a significant decrease in catalytic activity, particularly in the direction of fumarate reduction. Simultaneously, the characteristic potentials of the enzyme (active-site FAD and effective catalytic potentials) are raised but to a lesser extent than is the reduction potential of the substrate. The catalytic energetics are thus altered. For the first time with a complex redox enzyme, the results enable H/D substitution effects to be rationalized in terms of the changes introduced in overall driving force and enzyme reduction potentials, as well as the effects of intrinsic kinetic processes.

Succinate dehydrogenase (SDH), the membrane-extrinsic component of Complex II, adsorbs at a pyrolytic graphite edge electrode and catalyzes interconversion of succinate and fumarate depending on the electrochemical potential that is applied. The catalytic activity is measured over a continuous potential range, leading to a quantitative description of the interlinked energetics and kinetics of catalyzed electron transport, including the degree to which. the enzyme is intrinsically tuned, at a particular pH, to function either in the direction of succinate oxidation or fumarate reduction. It is revealed that under reversible conditions (i.e. near the reduction potential of the fumarate/ succinate couple) and at the physiological temperature of 38 degrees C, SDH is biased to catalyze fumarate reduction (reversal of the tricarboxylic acid cycle) at pH values below 7.7. Subtle effects which gate electron transport are detected. First, the sharp drop in catalytic activity observed as the potential is made more negative is an intrinsic property that is associated with two-electron/two-proton reduction of the FAD, and second, binding and release of the competitive inhibitor/regulator oxalacetate is observed as the enzyme is cycled between FAD(ox) (tight binding) and FAD(red) (weaker binding) states. It is thereby demonstrated how the electron-transport characteristics of a complex redox enzyme, integrating both kinetic and thermodynamic information, can be derived from voltammetric experiments.

Defects of the mitochondrial respiratory chain are increasingly being recognized as an important cause of neurological disease in humans. In many of these patients, the biochemical defect results from an abnormality of the mitochondrial genome. Respiratory chain defects involving complex II, which is entirely encoded by the nuclear genome, are comparatively rare. We report the clinical and biochemical findings in 2 elderly sisters who presented with late-onset neurodegenerative disease. In both patients, a partial deficiency of complex II (approximately 50% of control values) was shown to be present in mitochondria from muscle and platelets. The enzyme defect was not expressed in cultured skin fibroblasts or immortalized lymphocytes. There was an overexpression of the 70-kd flavoprotein subunit in muscle mitochondria from both patients, although we showed that this subunit is present in normal amounts in mitochondrial membranes. Our studies highlight the diversity of the clinical presentation of respiratory chain disease and that complex II deficiency should enter the differential diagnosis of certain patients with late-onset neurodegenerative disease.

The structural and catalytic properties of beef heart succinate dehydrogenase (succinate-ubiquinone oxidoreductase, complex II) and Escherichia coli fumarate reductase are remarkably similar. One exception is that whereas electron exchange between the mammalian enzyme and its quinone pool is inhibited by thenoyltrifluoroacetone and carboxanilides, the enzyme from E. coli is not sensitive to these inhibitors. The lack of good inhibitors has seriously hampered the elucidation of the mechanism of quinone oxidation/reduction in the E. coli enzyme. We have previously reported (Tan, A. K., Ramsay, R. R., Singer, T. P., and Miyoshi, H. (1993) J. Biol. Chem. 268, 19328-19333) that 2-alkyl-4,6-dinitrophenols inhibit mammalian complexes I, II, and III, but with different potencies and kinetic characteristics. Based on these studies we have selected a series of 2-alkyl-4,6-dinitrophenols which proved to be very effective noncompetitive inhibitors of mammalian complex II, particularly when acting in the direction of quinone reduction, the physiological event. These compounds turned out to be even more potent inhibitors of E. coli fumarate reductase, particularly when acting in the direction of quinol oxidation, again, the physiological event. Kinetic analysis revealed that with both enzymes 2 inhibitor binding sites seem to be involved in the oxidation of succinate by quinone, but one seems to be functioning when fumarate is reduced by external quinol. Since the E. coli enzyme can be modified by site-directed mutagenesis, these studies were extended to four mutants of fumarate reductase, impaired by single amino acid substitutions at either of the putative quinone binding sites (QA or QB) of the enzyme. The results were analyzed in terms of the model of these dual sites of quinone binding in fumarate reductase, as well as the nature of the substituent in the 2-position of the dinitrophenol inhibitors.

We report the full-length cDNA sequence for the flavoprotein subunit of human heart succinate dehydrogenase (succinate: (acceptor) oxidoreductase EC 1.3.99.1). Identical sequence was obtained for part of the cDNA of the human placental flavoprotein, in contrast to a previously published sequence. The human sequence, like the bovine one, contains a cysteine triplet and at the active site there is an additional cysteine when compared with yeast or prokaryotes.

It has been reported that N-methyl-beta-carbolinium analogues of the neurotoxic N-methyl-4-phenylpyridinium cation (MPP+) inhibit NADH-linked mitochondrial oxidations, as well as mitochondrial respiration on succinate nearly to the same extent [Fields, Albores, Neafsey and Collins (1992) Arch. Biochem. Biophys. 294, 539-544]. Those authors further claimed that MPP+ itself also blocks respiration through succinate dehydrogenase, in addition to its well-known effect on NADH dehydrogenase (Complex I), and concluded that both effects may contribute to the development of Parkinsonian symptoms. Since N-methyl-beta-carboliniums are thought to be endogenous metabolites, these findings, if verified, would have important implications on the etiology of idiopathic Parkinsonism. We have re-examined these observations, using mitochondria after full activation of succinate dehydrogenase, as well as submitochondrial particles, in which complexities due to membrane transport are not present. We report the following observations. (1) N-Methyl-beta-carboliniums inhibit mitochondrial respiration on NAD+-linked substrates in a time-dependent manner, and the inhibition is potentiated by the presence of tetraphenylboron anion (TPB-), as expected for positively charged compounds. (2) Unlike MPP+ itself, however, these compounds are uncouplers at higher concentrations, so that the effects seen in State 3 cannot be assigned exclusively to inhibition of NADH oxidation. (3) The effects on succinate oxidation in mitochondria, in which the full activity of the enzyme is expressed, are 1-1.5 orders of magnitude lower than on respiration via Complex I and are thus unlikely to contribute significantly to the neurotoxicity. (4) The effect of MPP+ on mitochondrial respiration via succinate dehydrogenase is trivial, in accord with previous reports from several laboratories, but contradicting the findings of Fields et al. (cited above). (5) In submitochondrial particles the inhibition of NADH oxidation (via the complete respiratory chain) has been confirmed, but it differs markedly from the action of MPP+ in two respects. First, the enhancement by TPB- is very small; secondly, the inhibition of NADH oxidation measured using ubiquinone (Q) analogues is far lower, suggesting that Complex I is not the only target. (6) In submitochondrial particles the inhibition of succinate oxidation by either O2 or Q analogues is incomplete, trivial or absent. (7) We thus conclude that we find no basis for assigning any potential biological effect of N-methyl-beta-carboliniums to the blockade of succinate oxidation.

Escherichia coli fumarate reductase (FRD) is a four-subunit enzyme that catalyzes the terminal step in anaerobic respiration to fumarate. The hydrophobic FrdC and FrdD subunits anchor the FrdA and FrdB catalytic subunits to the inner surface of the cytoplasmic membrane and are required for the enzyme to interact with quinones. Thirty-five single-site mutations were constructed in the FrdC and FrdD polypeptides by site-directed mutagenesis. Each mutant enzyme was characterized for its ability to catalyze quinone oxidation and reduction and to support growth of E. coli DW35 (delta frdABCD sdhC::kan) under selective conditions requiring functional enzyme. Replacement of FrdCE29 with Asp, Leu, Lys, or Phe had a deleterious effect both on quinol oxidase and quinone reductase activities. Substitution of FrdCH82 with Arg, Leu, Tyr, or Glu also decreased menaquinol oxidase activity, but had variable effects on the reverse reaction, the reduction of ubiquinone. Data are presented to support the hypothesis that the positive charge at FrdCH82 is required for stabilization of the quinone radical intermediate and the negative charge at FrdCE29 for deprotonation of menaquinol. Other critical amino acids identified in FrdC included Ala-32, Phe-38, Trp-86, Phe-87, and in FrdD residues Phe-57, Gln-59, Ser-60, and His-80. The established roles of such residues in the QA and QB sites of the photosynthetic reaction center would suggest a similar type of structure operative in the FRD complex. In such a model, Glu-29, Ala-32, His-82, Trp-86 of FrdC and His-80 of FrdD are considered participants in a QB-type site, and FrdD Phe-57, Gln-59, and Ser-60 components in an apolar QA-type site.

Reduction of fumarate by soluble beef heart succinate dehydrogenase has been shown previously by voltammetry to become increasingly retarded as the potential is lowered below a threshold potential of -80 mV at pH 7.5. The behaviour resembles that of a tunnel diode, an electronic device exhibiting the property of negative resistance. The enzyme thus acts to oppose fumarate reduction under conditions of high thermodynamic driving force. We now provide independent evidence for this phenomenon from spectrophotometric kinetic assays. With reduced benzylviologen as electron donor, we have studied the reduction of fumarate catalysed by various enzymes classified either as succinate dehydrogenases or fumarate reductases. For succinate dehydrogenases, the rate increases as the concentration of reduced dye (driving force) decreases during the reaction. In contrast, authentic fumarate reductases of anaerobic cells (and 'succinate dehydrogenase' from Bacillus subtilis) neither exhibit the electrochemical effect nor deviate from simple kinetic behaviour in the cuvette assay. The 'tunnel-diode' effect may thus represent an evolutionary adaptation to aerobic metabolism.

Studies have demonstrated that accumulation of mitochondrial tocopheroxyl radical, the primary oxidation product of α-tocopherol, accompanies rapid consumption of tocopherol. Enzyme-linked electron flow lowers both the steady-state concentration of the radical and the consumption of tocopherol. Reduction of tocopheroxyl radical by a mitochondrial electron carrier(s) seems a likely mechanism of tocopherol recycling. Succinate-ubiquinone reductase (complex II) was incorporated into liposomes in the presence of tocopherol and ubiquinone-10. After inducing formation of tocopheroxyl radical, it was possible to show that reduced ubiquinone prevents radical accumulation and tocopherol consumption. There was no evidence of direct reduction of tocopheroxyl radical by succinate-reduced complex II. These reactions were also measured using ubiquinone-1 and α-C-6-chromanol (2,5,7,8-tetramethyl-2-(4′-methylpentyl)-6-chromanol) which are less hydrophobic analogues of ubiquinone-10 and α-tocopherol. Mitochondrial membranes were made deficient in ubiquinone but sufficient in α-tocopherol and were reconstituted with added quinone. With these membranes it was shown that mitochondrial enzymelinked reduction of ubiquinone protects α-tocopherol from consumption, and there is a requirement for ubiquinone. This complements the observations made in liposomes and we propose that reduced mitochondrial ubiquinones have a role in α-tocopherol protection, presumably through efficient reduction of the tocopheroxyl radical.

Two new methods have been devised for measuring fumarate reduction by beef heart succinate-ubiquinone oxidoreductase with quinols as electron donors. In one assay the quinone is maintained in the reduced state by coupling fumarate reduction with the DT-diaphorase reaction, in the other assay by the presence of excess dithionite. The advantages of these methods are discussed, along with preliminary characterization of the quinol-protein interaction.

The cDNA sequence of the flavoprotein subunit of bovine heart succinate dehydrogenase is reported. This is the first complete eukaryotic sequence of the flavoprotein subunit to be characterized, and it encodes a 665-amino acid protein that consists of a presequence and a 621-residue mature protein. The deduced bovine sequence shows homology to the corresponding peptides of prokaryotic succinate dehydrogenase and the related fumarate reductases; in particular, there is good overall homology (48%) to the flavoprotein subunit of Escherichia coli succinate dehydrogenase. The conserved sequences comprising the active site and those involved in FAD binding are also found in the bovine protein. The active site of the bovine polypeptide contains a cysteine that confers sensitivity of the enzyme to sulfhydryl reagents; this cysteine is only present in some sequences and thus provides a discriminatory biochemical marker. A putative flavoprotein subunit of human placental succinate dehydrogenase (partial sequence) that lacks this critical cysteine (Malcovati, M., Marchetti, T., Zanelli, T., and Tenchini, M. L. (1991) in Flavins and Flavoproteins 1990 (Curti, B., Ronchi, S., and Zanetti, G., eds) pp. 727-730, Walter de Gruyter & Co., Berlin) has only 16% homology to the bovine heart flavoprotein subunit. However, we show that the enzyme from human placenta is as sensitive to N-ethylmaleimide as that from bovine tissues. In addition, a transcript in human placenta and muscle hybridizes to the bovine heart flavoprotein cDNA and is the same size as that in bovine tissues.