Thermodynamic and EPR characteristics of two ferredoxin-type iron-sulfur centers in the succinate-ubiquinone reductase segment of the respiratory chain.

Two distinct ferredosin-type iron-sulfur centers (designated as Centers S-1 and S-2) are present in the soulble succinate dehydrogenase in approximately equivalent concentrations to that of bound flavin. Both Centers S-1 and S-2 exhibit electron paramagnetic resonance absorbance in the reduced state at the same magnetic field (gz = 2.03, gy = 1.93, and gx = 1.91) with similar line shape. Center S-2 is reducible only chemically with dithionite and remains oxidized under physiological conditions. Thus, its functional role is unknown; however, thermodynamic and EPR characterization of this iron-sulfur center has revealed important molecular events related to this dehydrogenase. The midpoint potentials of Centers S-1 and S-2 determined in the soluble succinate dehydrogenase preparations are -5 +/- 15 mV and -400 +/- 15 mV, respectively, while corresponding midpoint potentials determined in particulate preparations, such as succinate-cytochrome c reductase or succinate-ubiquinone reductase, are 0 +/- 15 mV and -260 +/- 15 mV. Reconstitution of soluble succinate dehydrogenase with the cytochrome b-c1 complex is accompanied by a reversion of the Center S-I midpoint from -400 +/- 15 mV to -250 +/- 15 mV with a concomitant restoration of antimycin A-sensitive succinate-cytochrome c reductase activity. There observations indicate that, during the reconstitution process, Center S-I is restored to its original molecular environment. In the reconstitutively active succinate dehydrogenase, the relaxation time of Center S-2 is much shorter than that of S-1, thus Center S-2 spectra are well discernible only below 20 K (at 1 milliwatt of power), while the resonance absorbance of Center S-1 is detectable at higher temperatures and readily saturates below 15 K. Over a wide temperature range the power saturation of Center S-1 resonance absorbance is relieved by Center S-2 in the paramagnetic state, and the Center S-2 central resonance absorbance is broadened by Center S-1 spins, due to a spin-spin interaction between these centers. These observations indicate an adjacent location of these centers in the enzyme molecule. In reconstitutively inactive enzymes, subtle modification of the enzyme structure appears to shift the temperature dependence of Center S-2 relaxation to the higher temperature. Thus the EPR signals of Center S-2 are also detectable at higher temperature. In this system a splitting of the central peak of the Center S-2 spectrum due to spin-spin interaction was observed at extremely low temperatures, while this was not observed in reconstitutively active enzymes or in paritculate preparations. This spin-spin interaction phenomena of inactive enzymes disappeared upon chemical reactivation with concomitant appearance of the reconstitutive activity. These observations provide a close correlation between the molecular integrity of the enzyme and its physiological function.     

Thermodynamic and EPR characteristics of a HiPIP-type iron-sulfur center in the succinate dehydrogenase of the respiratory chain.

In addition to the two species of ferredoxin-type iron-sulfur centers (Centers S-1 and S-2), a third iron-sulfur center (Center S-3), which is paramagnetic in the oxidezed state analogous to the bacterial high potential iron-sulfur protein, has bwen detected in the reconstitutively active soluble succinate dehydrogenase preparation. Midpoint potential (at pH 7.4) of Center S-3 determined in a particulate succinate-cytochrome c reductase is +60 +/- 15 mV. In soluble form, Center S-3 becomes extremely labile towards oxygen or ferricyanide plus phenazine methosulfate similar to reconstitutive activity of the dehydrogenase. Thus, even freshly prepared reconstitutively active enzyme preparations show EPR spectra of Center S-3 which correspond approximately to 0.5 eq per flavin; in particulate preparations this component was found in a 1:1 ratio to flavin. All reconstitutively inactive dehydrogenase preparations that Center S-3 is an innate constituent of succinate dehydrogenase and plays an important role in mediating electrons from the flavoprotein subunit to most probably ubiquinone and then to the cytochrome chain.     

EPR studies on two ferredoxin-type iron-sulfur centers in reconstitutively active, inactive, and reactivated soluble succinate dehydrogenases

Two distinct iron-sulfur centers, S-1 and S-2 are present in both reconstitutively active and inactive soluble succinate dehydrogenase preparations in approximately equivalent concentrations to that of bound flavin. The midpoint potentials at pH 7.4 of these centers are ?5 ± 15 mV and ?400 ± 15 mV, respectively. EPR characteristics of Center S-2, observed above 6° K, are not significantly different in the active and inactive dehydrogenases. At lower temperatures, however, major line shape modifications of Center S-2 spectra are observed in the reconstitutively inactive dehydrogenases, but neither in the active dehydrogenase nor in the particulate preparations. This phenomenon may reflect spin-spin interaction between Centers S-1 and S-2. Chemical reactivation of the reconstitutively inactive preparations abolishes this resonance modification and restores the normal line shape. This is a demonstration of another close correlation between a physical property and reconstitutive activity of succinate dehydrogenase.     

EPR studies on a Hipip type iron-sulfur center in the succinate dehydrogenase segment of the respiratory chain

In addition to the two species of ferredoxin-type iron-sulfur centers (Centers S-1 and S-2), a Hipip-type iron-sulfur center (Center S-3) has been detected in the reconstitutively active soluble succinate dehydrogenases. E_m7,4 determined in a particulate, antimycin A sensitive succinate-cytochrome $\text{c}$ reductase is +60 ± 15 mV. This center is extremely labile towards oxygen in a manner similar to the reconstitutive activity of the dehydrogenase. Even freshly prepared reconstitutively active enzyme shows a considerably diminished content of Center S-3 relative to flavin and displays a partly modified spectra. All reconstitutively inactive dehydrogenases give rise to a highly modified or no Center S-3 spectra at all. These observations indicate that Center S-3 is a constituent of succinate dehydrogenase and plays a role in the physiological function of the enzyme, $\text{i.e.}$ transferring electrons most probably to ubiquinone.     

Effects of ethanol and acetaldehyde on iron-sulfure centers in the mitochondrial respiratory chain.

Incubation of submitochondrial particles with relatively low concentrations of ethanol (20–100 mm) or acetaldehyde (1–10 mm) produces alterations in the electron paramagnetic resonance spectra of the iron-sulfur centers in the NADH dehydrogenase segments of the respiratory chain. The iron-sulfur centers in the NADH dehydrogenase region are most sensitive to both ethanol and acetaldehyde, in comparison to the iron-sulfur centers in succinate dehydrogenase and the cytochrome b-c region. Centers N-3, 4, N-5, 6 and N-1b are affected after relatively short incubation periods (3–30 min) while center N-2 shows considerable sensitivity over somewhat longer incubations (20–90 min). The most ethanol-sensitive center in the succinate dehydrogenase region of the respiratory chain is high potential iron-sulfur protein-type center S-3. Potentiometric analysis shows that these alterations are not due to simple changes in the redox state caused by addition of dissolved oxygen. Changes in the electron paramagnetic resonance spectra can be correlated with decreased rates of oxidation of NADH and, to a lesser extent, succinate in both ethanol- and acetaldehyde-treated submitochondrial particles.     

Characterization by electron paramagnetic resonance and studies on subunit location and assembly of the iron-sulfur clusters of Bacillus subtilis succinate dehydrogenase

Succinate dehydrogenase is a conserved membrane-bound enzyme consisting of two nonidentical subunits: a flavo iron-sulfur protein (Fp) subunit, containing a covalently bound flavin, and an iron-sulfur protein (Ip) subunit. Bacillus subtilis succinate dehydrogenase in wild type bacteria and 12 well characterized succinate dehydrogenase-defective mutants were examined by low temperature EPR spectroscopy to characterize the enzyme and study subunit location and biosynthesis of its iron-sulfur clusters. The wild type B. subtilis enzyme contains iron-sulfur clusters which are analogous to clusters S-1 and S-3 of bovine heart succinate dehydrogenase but with slightly different EPR characteristics. Spins from cluster S-2 were not detectable as in the case of the intact form of bovine heart succinate dehydrogenase. However, dithionite reduction of the B. subtilis enzyme greatly enhanced spin relaxation of the ferredoxin-type cluster S-1, indicating the presence of the cluster S-2. Iron-sulfur cluster S-1 was found to be assembled in soluble succinate dehydrogenase subunits in the cytoplasm, but only if full-length Fp polypeptides and relatively large fragments of Ip polypeptides were present. Cluster S-1 was not detected in mutants with soluble mutated Fp polypeptides or in a mutant totally lacking Ip subunit polypeptide. Iron-sulfur clusters S-1, S-2, and S-3 were assembled also when the covalently bound flavin in the Fp subunit was absent. Clusters S-1 and S-3 in the membrane-bound flavin-deficient succinate dehydrogenase were not reduced by succinate but could be reduced by electron transfer from NADH dehydrogenase via the menaquinone pool.     

Low temperature electron paramagnetic resonance studies on two iron-sulfur centers in cardiac succinate dehydrogenase

At temperatures below 20°K, EPR signals from a new iron-sulfur center (designated here as Center S-2 or (Fe-S)_S-2) in addition to the classical “g = 1.94 signal” (designated as Center S-1 or (Fe-S)_S-1) were detected in purified, soluble succinate dehydrogenase, particulate succinate ubiquinone reductase (Complex II) and particulate succinate cytochrome $\text{c}$ reductase from bovine heart. The measured half-reduction potential (E_m7.4) of Center S-1 was 0 ± 10 mV, while E_m7.4 of Center S-2 was ?260 ± 15 mV in the membrane bound preparations. Upon solubilization of succinate dehydrogenase, the EPR behavior of Center S-2 became extremely labile similar to the characteristics of the reconstitutive activity of succinate dehydrogenase toward the rest of the respiratory chain.     

EPR studies of higher plant mitochondria. II. Center S-3 of succinate dehydrogenase and its relation to alternative respiratory oxidations.

1. An electron paramagnetic resonance study of the high potential iron sulfur (HiPIP-type) Center S-3 of higher plant mitochondria is described. This center is the major HiPIP-type center associated with plant mitochondria and it displays physical properties which are similar to its mammalian counterpart. It has a pH-independent midpoint potential of +65 +/- 10 mV between pH 6.0 and 8.5. 2. The behavior of Center S-3 in a variety of steady-state conditions suggests that it is of physiological significance in electron transport. Furthermore, it can be shown that the alternative oxidase, which is present in many higher plant mitochondria, tends to keep this center oxidized in the presence of succinate and cyanide. This indicates that the alternative oxidation site is on the electron-donating side of the Center S-3. 3. Salicylhydroxamic acid, an inhibitor of the alternative pathway, does not affect the midpoint potential, signal size or shape, or temperature and power saturation profiles of Center S-3, suggesting that direct autoxidation of this center cannot account for alternative oxidase activity. This is further confirmed by the finding that the presence of succinate dehydrogenase is not necessary for alternative oxidase activity with NADH as respiratory substrate in submitochondrial particles.     

Electron paramagnetic resonance studies of mammalian succinate dehydrogenase. Detection of the tetranuclear cluster S2

Electron paramagnetic resonance studies of Complex II from the mitochondrial respiratory chain and soluble preparations of succinate dehydrogenase have, for the first time, identified a signal arising from a [4Fe-4S]1+ cluster, S2, in dithionite-reduced samples. Redox titrations, monitored by electron paramagnetic resonance spectroscopy demonstrate that this signal appears at the same midpoint potential as the enhancement of the spin relaxation properties of the [2Fe-2S]1+ center, S1, in both Complex II and reconstitutively active soluble enzyme. The results complement recent magnetic circular dichroism studies of succinate dehydrogenase (Johnson, M. K., Morningstar, J. E., Bennett, D. E., Ackrell, B. A. C., and Kearney, E. B. (1985) J. Biol. Chem. 260, 7368-7378) which assigned cluster S2 as a [4Fe-4S]2+,1+ center and provide evidence for spin interaction between the paramagnetic reduced forms of centers S1 and S2.     

EPR signals of NADH: Q oxidoreductase. Shape and intensity.

(1) The EPR spectrum of Center 1 of NADH dehydrogenase in isolated Complex I or submitochondrial particles from beef heart consists of two overlapping nearly axial signals of the same intensity. They are defined as Center 1a (gll = 0.021, gl = 1.938) and Center 1b (gll = 2.021, gl = 1.928). (2) The line shape of the EPR spectrum of the Center 3+4 can be interpreted as an overlap of two rhombic signals of the same intensity. We define Center 3 by the g-values: gz=2.103, gy = 1.93-1.94, gx=1.884, and Center 4 by the values gz=2.04, gy=1.92-1.93, gx=1.863. (3) Direct quantitation of the individuals signals as well as computer stimulation suggests that the amount of the Centers 1a and 1b is only 25% of that of the other individuals centers and FMN. As EPR spectra of beef-heart submitochondrial particles at 10-20 K are nearly identical to those of Complex I, the same relative concentrations of the Fe-S centers are also present in the particles. (4) The signals either observed by us in EPR spectra of Complex I and submitochondrial particles at 4.2 K and high microwave powers can now be explained without assuming more than 5 paramagnetic centers in NADH dehydrogenase.     

An EPR analysis of cyanide-resistant mitochondria isolated from the mutant poky strain of Neurospora crassa.

An analysis of the paramagnetic components present in mitochondria isolated from the poky mutant of Neurospora crassa is described. The study was undertaken with a view to shedding light on the nature of the cyanide- and antimycin A-resistant alternative terminal oxidase which is present in these preparations. Of the ferredoxin-type iron-sulfure centers, only Centers S-1 and S-2 of succinate dehydrogenase could be detected in significant quantities. Paramagnetic centers attributable to Site I were virtually absent. In the oxidized state, at least two 'high potential iron sulfur' centers could be distinguished and these were attributed to Center S-3 of succinate dehydrogenase and a second component analogous to that found in mammalian systems. Much of the Center S-3 signal was in a highly distorted state which was apparently dependent upon the presence of an accompanying free radical species. At lower field positions, a succinate-reducible signal peaking around g = 3.15 was found. This signal is caused by a low spin heme species, presumably the cytochrome c which is the only major cytochrome in these mitochondria. At even lower field positions, signals attributable to iron in a field of low symmetry at g = 4.3 and multiple high spin heme species around g = 6, could be distinguished. The effects of salicylhydroxamic acid, an inhibitor of the alternative oxidase, were tested on these components. Effects could be seen on at least one high spin heme component and also partially upon the distorted Center S-3 signal converting part of it to a signal indistinguishable from center S-3. Some increase in the g = 4.3 iron signal was also noted. No effects of the inhibitor on the ferredoxin-type centers were detected.     

Intramitochondrial positions of ubiquinone and iron-sulphur centres determined by dipolar interactions with paramagnetic ions.

E.p.r. (electron-paramagnetic-resonance) spectra of ubisemiquinone (QH) organic radicals and all of the known iron-sulphur centres were studied in normal and 'nickle-plated' pigeon heart mitochondria, submitochondrial particles and submitochondrial particles from which succinate dehydrogenase had been removed. Incubation of pigeon heart mitochondria, submitochondrial particles or succinate dehydrogenase-depleted submitochondrial particles with substrate in the presence of pure O2 results in the accumulation of Q-H. In mitochondria, the e.p.r. spectrum of Q-H is characterized by in-homogeneous line broadening. A heterogeneous population of semiquinones appears to be partly responsible for these effects in mitochondria. Additon of Ni(II) to mitochondria renders saturation of the Q-H resonance more difficult. On the other hand, the resonance in either submitochondrial particles or succinate dehydrogenase-depleted particles is narrower than the same spectrum in mitochondria, and saturates like a homogeneous line. The presence of Ni(II) in either of these preparations, further, has no effect on either the A-H spectrum or the saturation curve. Therefore QH appears to be situated on the exterior surface of the mitochondrion. Likewise, the e.p.r. spectra and saturation curves of iron-sulphur centre N-2 exhibit characteristics of inhomogeneous line broadening, not only in intact mitochondria but also in both submitochondrial particles and succinate dehydrogenase-depleted particles. Because of the small pool size of centre N-2, this effect is likely to arise from a spin interaction with some other component in the membrane. Ni(II) has no effect on the saturation in centre N-2 in mitochondria or submitochondrial particles, and only a marginal effect in the succinate dehydrogenase-depleted preparation. These results are indeterminate with respect to the position of centre N-2 in the membrane; but suggest that its distance from the succinate dehydrogenase binding site is on the order of 1 nm. All of the other ferredoxin-type iron-sulphur centres in both preparations were not affected by paramagnetic ions. Homogeneous e.p.r. spectra and saturation curves are observed for both of the HiPIP-type (high-potential iron-sulphur protein-type) iron-sulphur centres in mitochondrial centres S-3 and bc-3. Addition of No(II) to intact mitochondria results in a dipolar interaction with centre bc-3. No effect was observed on centre S-3 in either preparation. A comprehensive model is presented for the structure of the respiratory electron-transport system in mitochondria, based on e.p.r. relaxation studies in the present and the preceding paper. There is no direct evidence for transmembrane electron flow through any of the known energy-coupling sites in mitochondria, so that direct hydrogen atom transfer across the membrane (as a combination of H+ translocation coupled to electron flow) does not occur...     

An EPR analysis of cyanide-resistant mitochondria isolated from the mutant poky strain of Neurospora crassa

An analysis of the paramagnetic components present in mitochondria isolated from the poky mutant of Neurospora crassa is described. The study was undertaken with a view to shedding light on the nature of the cyanide- and antimycin A-resistant alternative terminal oxidase which is present in these preparations.

Of the ferredoxin-type iron-sulfur centers, only Centers S-1 and S-2 of succinate dehydrogenase could be detected in significant quantities. Paramagnetic centers attributable to Site I were virtually absent. In the oxidized state, at least two ‘high potential iron sulfur’ centers could be distinguished and these were attributed to Center S-3 of succinate dehydrogenase and a second component analogous to that found in mammalian systems. Much of the Center S-3 signal was in a highly distorted state which was apparently dependent upon the presence of an accompanying free radical species. At lower field positions, a succinate-reducible signal peaking around g = 3.15 was found. This signal is caused by a low spin heme species, presumably the cytochrome c which is the only major cytochrome in these mitochondria. At even lower field positions, signals attributable to iron in a field of low symmetry at g = 4.3 and multiple high spin heme species around g = 6, could be distinguished.

The effects of salicylhydroxamic acid, an inhibitor of the alternative oxidase, were tested on these components. Effects could be seen on at least one high spin heme component and also partially upon the distorted Center S-3 signal converting part of it to a signal indistinguishable from Center S-3. Some increase in the g = 4.3 iron signal was also noted. No effects of the inhibitor on the ferredoxin-type centers were detected.

These results are interpreted with respect to the nature and location of the alternative oxidase and with respect to possible models for the nature of the alternative oxygen-consuming component.     

EPR characterization of soluble fragments of succinate dehydrogenase from mutant strains of Bacillus subtilis

Succinate dehydrogenase is a membrane-bound metallo-flavo-enzyme containing a bi- (S-1), a tri- (S-3) and a tetranuclear (S-2) iron-sulfur cluster. The catalytic portion of the enzyme contains two distinct subunits designated Fp and Ip. Using concentrated extracts from mutant strains of Bacillus subtilis it was demonstrated, by using low temperature EPR, that cluster S-2 can be assembled in a soluble succinate dehydrogenase. In a mutant with a truncated Ip subunit which lacks 7 of the 11 conserved cysteine residues, cluster S-1 lacked the spin relaxation properties attributable to an adjacent cluster S-2. These data are consistent with a model where one or more cysteine residues from the middle set of 4 conserved cysteines in the Ip subunit are ligands to the tetranuclear cluster.     

Determination of the exchange integral in binuclear iron-sulfur clusters in proteins of varying complexity.

The coupling constants J between the iron atoms in ferredoxin type iron-sulfur proteins containing binuclear clusters were evaluated by two parallel methods. The temperature dependence of the EPR linewidths and integrated abosrption intensities are both related to the energy of the first excited state. The values of J obtained were: center S-1 in succinate dehydrogenase, 90 cm-1; Rieske's iron-sulfur center, 65 cm-1; adrenodoxin, 270 cm-1. The behavior of iron-sulfur center N-1a in NADH:UQ reductase was also examined; its similarity to that of center S-1 indicates that center N-1a is also a binuclear iron-sulfur center, with J = 90 cm-1. Greater rhombic distortion present in the EPR spectrum of a binuclear cluster was associated with smaller values of J.     

The succinate dehydrogenase of Escherichia coli. Immunochemical resolution and biophysical characterization of a 4-subunit enzyme complex

Using EPR spectroscopy to monitor the integrity of the enzyme, conditions have been established which allow specific immunoprecipitation of the succinate dehydrogenase complex of Escherichia coli. The enzyme complex precipitated from Lubrol PX-solubilized membranes by monospecific antiserum in the presence of a cocktail of protease inhibitors contains four polypeptides of apparent MrS 71,000, 26,000, 17,000, and 15,000. The 71-kDa flavopeptide is readily susceptible to proteolysis, and the enzyme complex shows unusual facile dissociation. Spectroscopic measurements indicate the presence of a [2Fe-2S] cluster (Center 1), a [3Fe-xS] cluster (Center 3), and a b-type cytochrome. In addition, a change in relaxation of Center 1 at low potentials is indicative of Center 2. Midpoint redox potentials of Centers 1-3 for both the membrane-bound and detergent-solubilized enzyme were estimated to be +10 mV, -175 mV, and +65 mV, respectively.     

Formate dehydrogenase from Methanobacterium formicicum. Electron paramagnetic resonance spectroscopy of the molybdenum and iron-sulfur centers

Formate dehydrogenase from Methanobacterium formicicum was examined by electron paramagnetic resonance spectroscopy. Although oxidized enzyme yielded no EPR signals over the temperature range 8-200 K, dithionite reduction resulted in generation of two paramagnetic components. The first, a nearly isotropic signal visible at temperatures below 200 K with g1 = 2.018, g2 = 2.003, and g3 = 1.994, exhibited nuclear hyperfine interaction with two equivalent protons (A1 = 0.45, A2 = 0.6, and A3 = 0.55 milliTeslas). EPR spectra of partially reduced 95Mo-enriched formate dehydrogenase exhibited additional 3-4 milliTeslas splittings, due to spin interaction with the 95Mo nucleus. Thus, this signal is due to a Mo center. This is the first reported example of a Mo center with gav greater than 2.0 in a biological system. The second species, a rhombic signal visible below 40 K with g values of g1 = 2.0465, g2 = 1.9482, and g3 = 1.9111 showed no hyperfine coupling and was assigned to reduced Fe/S. Both paramagnetic species could be detected in samples of M. formicicum whole cells anaerobically reduced with sodium formate. The Mo(V) signal was altered following addition of cyanide (g1 = 1.996, g2 = 1.988, and g3 = 1.980). Growth of bacteria in the presence of 1 mM WO4(2-) resulted in abolition of the Mo(V) EPR signal and formate dehydrogenase activity. Em, 7.7 was -330 mV for Mo(VI)/Mo(V) and -470 mV for Mo(V)/Mo(IV).     

Active site structure of SoxB-type cytochrome bo3 oxidase from thermophilic Bacillus

Two-subunit SoxB-type cytochrome c oxidase in Bacillus stearothermophilus was over-produced, purified, and examined for its active site structures by electron paramagnetic resonance (EPR) and resonance Raman (RR) spectroscopies. This is cytochrome bo3 oxidase containing heme B at the low-spin heme site and heme O at the high-spin heme site of the binuclear center. EPR spectra of the enzyme in the oxidized form indicated that structures of the high-spin heme O and the low-spin heme B were similar to those of SoxM-type oxidases based on the signals at g=6.1, and g=3.04. However, the EPR signals from the CuA center and the integer spin system at the binuclear center showed slight differences. RR spectra of the oxidized form showed that heme O was in a 6-coordinated high-spin (nu3 = 1472 cm(-1)), and heme B was in a 6-coordinated low-spin (nu3 = 1500 cm(-1)) state. The Fe2+-His stretching mode was observed at 211 cm(-1), indicating that the Fe2+-His bond strength is not so much different from those of SoxM-type oxidases. On the contrary, both the Fe2+-CO stretching and Fe2+-C-O bending modes differed distinctly from those of SoxM-type enzymes, suggesting some differences in the coordination geometry and the protein structure in the proximity of bound CO in cytochrome bo3 from those of SoxM-type enzymes.     

A model of the copper centres of nitrous oxide reductase (Pseudomonas stutzeri). Evidence from optical, EPR and MCD spectroscopy.

Nitrous oxide reductase (N2OR), Pseudomonas stutzeri, catalyses the 2 electron reduction of nitrous oxide to di-nitrogen. The enzyme has 2 identical subunits (Mr approximately 70,000) of known amino acid sequence and contains approximately 4 Cu ions per subunit. By measurement of the optical absorption, electron paramagnetic resonance (EPR) and low-temperature magnetic circular dichroism (MCD) spectra of the oxidised state, a semi-reduced form and the fully reduced state of the enzyme it is shown that the enzyme contains 2 distinct copper centres of which one is assigned to an electron-transfer function, centre A, and the other to a catalytic site, centre Z. The latter is a binuclear copper centre with at least 1 cysteine ligand and cycles between oxidation levels Cu(II)/Cu(II) and Cu(II)/Cu(I) in the absence of substrate or inhibitors. The state Cu(II)/Cu(I) is enzymatically inactive. The MCD spectra provide evidence for a second form of centre Z, which may be enzymatically active, in the oxidised state of the enzyme. Centre A is structurally similar to that of CuA in bovine and bacterial cytochrome c oxidase and also contains copper ligated by cysteine. This centre may also be a binuclear copper complex.