Effect of changing the detergent bound to bovine cytochrome c oxidase upon its individual electron-transfer steps

The influence of the detergent environment upon individual electron-transfer rates of cytochrome c oxidase was investigated by stopped-flow spectrophotometry. The effects of three detergents were studied: lauryl maltoside, which supports a high turnover number (TN = 350 s-1), n-dodecyl octaethylene glycol monoether (C12E8), which supports an intermediate TN (150 s-1), and Triton X-100 in which oxidase is nearly inactive (TN = 2-3 s-1). Under limited turnover conditions (cytochrome c:cytochrome c oxidase ratio = 1:1 to 8:1), the rate of oxidation of cytochrome c was measured and compared with the fast reduction of cytochrome a and its relatively slow reoxidation. Two reducing equivalents of cytochrome c were rapidly oxidized in a burst phase; the remaining two to six equivalents were oxidized more slowly, concurrent with the reoxidation of cytochrome a; i.e., the percent reduced cytochrome a reflects the percent reduced cytochrome c. With the resting enzyme, the bimolecular reaction between reduced cytochrome c and cytochrome a was rapid, was insensitive to the detergent environment, and was not the rate-limiting step in the presence of any detergent. The rate of internal electron transfer from cytochrome a to cytochrome a3 in the resting enzyme was slow and only slightly affected by the detergent environment: 1.0-1.1 s-1 in Triton X-100, 5-7 s-1 in C12E8, and 5-12 s-1 in lauryl maltoside. With the pulsed enzyme, the intramolecular electron transfer between cytochrome a and cytochrome a3 increased 4-5-fold in the lauryl maltoside enzyme but did not increase in the Triton X-100 enzyme (intermediate values were obtained with the C12E8 enzyme). We conclude that cytochrome c oxidase acquires the pulsed conformation only in those detergents that support high TN's, e.g., lauryl maltoside and C12E8, but it is locked in the resting conformation in those detergents which result in low TN's, e.g., Triton X-100.     

Purification of yeast cytochrome c oxidase with a subunit composition resembling the mammalian enzyme.

Yeast cytochrome c oxidase has been isolated by ion exchange chromatography using lauryl maltoside (n-dodecyl beta-D-maltoside) as the solubilizing detergent. The enzyme prepared in this way has a heme aa3 concentration of 8-9 nmol/mg of protein and a turnover number in the range of 180-210 s-1 at pH 6.2 in 0.01% lauryl maltoside at 20 degrees C. Yeast cytochrome c oxidase prepared by any of several previously published methods which use Triton X-100 contains nine subunits. The enzyme isolated in lauryl maltoside contains these same nine different polypeptides and three others, including homologues of subunits VIa and VIb of the mammalian enzyme.     

Some magnetic properties of Pseudomonas cytochrome oxidase.

The magnetic properties of the haem groups of Pseudomonas cytochrome oxidase and its cyanide-bound derivatives were studied in both the oxidized and reduced states by means of m.c.d. (magnetic circular dichroism) at low temperatures. In addition, the oxidized forms of the enzyme were also investigated by e.p.r. (electron-paramagnetic-resonance) spectroscopy, and a parallel study, using both e.p.r. and m.c.d., was made on Pseudomonas cytochrome c-551 to aid spectral assignments. For ascorbate-reduced Pseudomonas cytochrome oxidase, the temperature-independence of those features in the m.c.d. spectrum corresponding to the haem c, and the temperature-dependence of those signals corresponding to the haem d1, showed the former to be low-spin and the latter to be high-spin (s = 2). However, addition of cyanide to the reduced enzyme gave a form of the protein that was completely low-spin. The e.p.r. and m.c.d. sectra of oxidized Pseudomonas cytochrome oxidase and its cyanide derivative were consistent with the haem c and d1 components being low-spin in both cases. Pseudomonas cytochrome c-551 was found to be low-spin in both its oxidized and reduced redox states.     

A re-evaluation of some basic structural and functional properties of Pseudomonas cytochrome oxidase

Determinations of iron content and dry-weight measurements on samples of Pseudomonas cytochrome oxidase were coupled with sodium dodecyl sulphate/polyacrylamide-gel-electrophoresis studies of both the native protein and covalently cross-linked oligomers in order to estimate the enzyme's molecular weight and spectral absorption coefficients. A value of epsilon(ox.) (410)=282x10(3) litre.mol(-1).cm(-1) was calculated for a dimeric protein molecule having a total molecular weight of 122000 (based on iron analysis). Steady-state kinetic observations of the enzyme-catalysed oxidation of reduced azurin by nitrite indicated a marked increase in enzyme inactivation as the pH was raised from 5.7 to 7.2. Since NO, a product of the nitrite reductase activity of Pseudomonas cytochrome oxidase, is known to bind to the enzyme, a study was undertaken to try to assess the potential of NO as a product inhibitor. Investigations showed that samples of the oxidized protein at pH values 4, 5 and 6 bound NO to both haem c and d(1) components, but oxidized enzyme samples at pH7 and above formed their reduced ligand-bound forms when placed under an atmosphere of the gas. Ascorbate-reduced enzyme samples at pH4, 5, 6 and 7 were also found to bind NO at both haem components, although at pH7 the rate of haem c binding was very slow. At pH8 and 9 only the ferrohaem d(1) bound NO. Titration experiments on the reduced protein over the pH range 5-7, with nitrite as a precursor of NO, showed that the haem d(1) had a much higher affinity than the haem c: experiments at pH5.2 and 5.9 with NO-equilibrated solutions revealed the same pattern of behaviour with the oxidized enzyme.     

Lipid and subunit III depleted cytochrome c oxidase purified by horse cytochrome c affinity chromatography in lauryl maltoside

Cytochrome oxidase is purified from rat liver and beef heart by affinity chromatography on a matrix of horse cytochrome c-Sepharose 4B. The success of this procedure, which employs a matrix previously found ineffective with beef or yeast oxidase, is attributed to thorough dispersion of the enzyme with nonionic detergent and a low density of cross-linking between the lysine residues of cytochrome c and the cyanogen bromide activated Sepharose. Beef heart oxidase is purified in one step from mitochondrial membranes solubilized with lauryl maltoside, yielding an enzyme of purity comparable to that obtained on a yeast cytochrome c matrix [Azzi, A., Bill, K., & Broger, C. (1982) Proc. Natl. Acad. Sci. U.S.A. 79, 2447-2450]. Rat liver oxidase is prepared by hydroxyapatite and horse cytochrome c affinity chromatography in lauryl maltoside, yielding enzyme of high purity (12.5-13.5 nmol of heme a/mg of protein), high activity (TN = 270-400 s-1), and very low lipid content (1 mol of DPG and 1 mol of PI per mol of aa3). The activity of the enzyme is characterized by two kinetic phases, and electron transfer can be stimulated to maximal rates as high as 650 s-1 when supplemented with asolectin vesicles. The rat liver oxidase purified by this method does not contain the polypeptide designated as subunit III. Comparisons of the kinetic behavior of the enzyme in intact membranes, solubilized membranes, and the purified delipidated form reveal complex changes in kinetic parameters accompanying the changes in state and assay conditions, but do not support previous suggestions that subunit III is a critical factor in the binding of cytochrome c at the high-affinity site on oxidase or that cardiolipin is essential for the low-affinity interaction of cytochrome c. The purified rat liver oxidase retains the ability to exhibit respiratory control when reconstituted into phospholipid vesicles, providing definitive evidence that subunit III is not solely responsible for the ability of cytochrome oxidase to produce or respond to a membrane potential or proton gradient.     

The electron-transfer reaction between azurin and the cytochrome c oxidase from Pseudomonas aeruginosa.

A stopped-flow investigation of the electron-transfer reaction between oxidized azurin and reduced Pseudomonas aeruginosa cytochrome c-551 oxidase and between reduced azurin and oxidized Ps. aeruginosa cytochrome c-551 oxidase was performed. Electrons leave and enter the oxidase molecule via its haem c component, with the oxidation and reduction of the haem d1 occurring by internal electron transfer. The reaction mechanism in both directions is complex. In the direction of oxidase oxidation, two phases assigned on the basis of difference spectra to haem c proceed with rate constants of 3.2 X 10(5)M-1-S-1 and 2.0 X 10(4)M-1-S-1, whereas the haem d1 oxidation occurs at 0.35 +/- 0.1S-1. Addition of CO to the reduced enzyme profoundly modifies the rate of haem c oxidation, with the faster process tending towards a rate limit of 200S-1. Reduction of the oxidase was similarly complex, with a fast haem c phase tending to a rate limit of 120S-1, and a slower phase with a second-order rate of 1.5 X 10(4)M-1-S-1; the internal transfer rate in this direction was o.25 +/- 0.1S-1. These results have been applied to a kinetic model originally developed from temperature-jump studies.     

A comparative survey of several bacterial aa3-type cytochrome c oxidases

The aa3-type cytochrome c oxidases purified from Nitrobacter agilis, Thiobacillus novellus, Nitrosomonas europaea, and Pseudomonas AM 1 were compared. They have haem a and copper atom as the prosthertic groups and show alpha and gamma absorption peaks at around 600 and 440 nm, respectively. Each oxidase molecule is composed of two kinds of subunits. The N. agilis oxidase has 2 moles of haem a and 2 atoms of copper in the minimal structural unit composed of one molecule each of the two kinds of subunits, while the T. novellus enzyme seems to contain one molecule of the haem and one atom of the metal in the unit. The N. europaea oxidase shows very low affinity for carbon monoxide. Each oxidase reacts rapidly with some eukaryotic cytochromes c as well as with its native cytochrome c. The cytochrome c oxidase activity of the N. agilis oxidase is 50% inhibited by 1 microM KCN, while 50% inhibition of the activity requires 100 microM KCN in the case of the N. europaea enzyme.     

Single electron reduction of 'slow' and 'fast' cytochrome-c oxidase.

Evidence is presented that single electron reduction is sufficient for rapid electron transfer (k greater than 20 s-1 at pH 8.0 in 0.43 M potassium EDTA) between haem a/CuA and the binuclear centre in 'fast' oxidase, whereas in 'slow' oxidase intramolecular electron transfer is slow even when both CuA and haem a are reduced (k congruent to 0.01 s-1). However, while a single electron can equilibrate rapidly between CuA, haem a and CuB in 'fast' oxidase, it seems that equilibration with haem a3 is relatively slow (k congruent to 2 s-1). Electron transfer between cytochrome c and CuA/haem a is similar for both types of enzyme (k congruent to 2.4 x 10(5) M-1.s-1).     

Characterization of the partially reduced cyanide-inhibited derivative of cytochrome c oxidase by optical, electron-paramagnetic-resonance and magnetic-circular-dichroism spectroscopy.

Optical. e.p.r. and near-infrared low-temperature m.c.d. (magnetic-circular-dichroism) spectroscopy were used to characterize the partially reduced cyanide-inhibited derivative of cytochrome c oxidase produced by anaerobic reductive titration with dithionite. The reductions of cytochrome a3+ and Cu2+a were followed by observation of the e.p.r. signals at g = 3.03, 2.21 and 1.5 and at g = 2.18, 2.03 and 1.99. As reduction proceeds new e.p.r. signals (g = 3.58 and 1.56) appear that quantify to give one haem per enzyme unit when a small excess of dithionite has been titrated in. The e.p.r. signal of the Cu2+a titrates in parallel with the disappearance of the band and 820nm in the optical absorption spectrum. The near-infrared m.c.d. spectrum shows the presence of the low-spin ferric haem, a3+, in the oxidized state of the enzyme, as a well-resolved positive peak at 1650nm. As reduction proceeds this band is replaced by one at 1550nm due to haem a3+(3)--CN in the partially reduced state. Hence as haem a3+(3)--CN becomes e.p.r.-detectable it also shows a near-infrared m.c.d. spectrum characteristic of a low-spin ferric haem. It is concluded that the partially reduced state of cyanide-inhibited cytochrome c oxidase contains a2+ . Cu+a . a3+(3)--CN . Cu+a3.     

Ca2+-dependent and independent mitochondrial damage in HepG2 cells that overexpress CYP2E1

The effect of detergents on electron and proton transfer in bovine cytochrome c oxidase was studied using steady-state and transient-state methods. Cytochrome c oxidase in lauryl maltoside has high maximal turnover (TN(max)=400 s(-1)), whereas activity is low (TN(max)=10 s(-1)) in Triton X-100. Single turnover studies of intramolecular electron transfer show similar rates in either detergent. Transient proton uptake experiments show the oxidase in lauryl maltoside consumes 1.8+/-0.3 H(+)/aa(3) during either partial reduction of the oxidase or reaction of fully reduced enzyme with O(2). However, the oxidase in Triton X-100 consumes 2.6+/-0.4 H(+)/aa(3) during partial reduction and 1.0+/-0.2 H(+)/aa(3) in the O(2) reaction. Absorption spectra recorded during turnover show that the enzyme undergoes activation in lauryl maltoside, but does not activate in Triton X-100. We propose that cytochrome c oxidase in different detergents allows access to different sites of protonation, which in turn influences steady-state activity.     

Orientation of the haems of the ubiquinol oxidase:O2 reductase, cytochrome bo of Escherichia coli.

The orientation of the two haems of the Escherichia coli ubiquinol oxidase:O2 reductase, cytochrome bo, has been determined by electron paramagnetic resonance studies on oriented multilayer preparations of cytoplasmic membrane fragments. The enzyme contains a low-spin b-like haem and a high-spin b-like haem, designated cytochromes b and o respectively. Both haems are oriented with their planes perpendicular to the membrane plane, further extending the catalogue of structural and functional similarities between this enzyme and the mammalian cytochrome c oxidase, cytochrome aa3.     

The reactions of Pseudomonas cytochrome c-551 oxidase with potassium cyanide.

The binding of cyanide to both oxidized and ascorbate-reduced forms of Pseudomonas cytochrome c-551 oxidase was investigated. Spectral studies on the oxidized enzyme and its apoprotein showed that the ligand can bind to both the c and d, haem components of the molecule, and kinetic observations indicated that both chromophores reacted, under a variety of conditions, with very similar rates. Cyanide combination velocities were dependent on ligand concentration, and increasing the pH also accelerated the reaction; the second-order rate constant was estimated as approx. 0.2M-1 . s-1 at pH 7.0. The binding of cyanide to the protein was observed to have a considerable influence on reduction of the enzyme by ascorbate. Spectral and kinetic observations have revealed that the species haem d13+-cyanide and any unbound haem c may react relatively rapidly with the reductant, but the behaviour of cyanide-bound haem c indicates that it may not be reduced without prior dissociation of the ligand, which occurs relatively slowly. The reaction of reduced Pseudomonas cytochrome oxidase with cyanide is radically different from that of the oxidized protein. In this case the ligand only binds to the haem d1 component and reacts much more rapidly. Stopped-flow kinetic measurements showed the binding to be biphasic in form. Both the rates of these processes were dependent on cyanide concentration, with the fast phase having a second-order rate constant of 9.3 X 10(5) M-1 . s-1 and the slow phase one of 2.3 X 10(5) M-1 . s-1. The relative proportions of the two phases also showed a dependency on cyanide concentration, the slower phase increasing as the cyanide concentration decreased. Computer simulations indicate that a reaction scheme originally proposed for the reaction of the enzyme with CO is capable of providing a reasonable explanation of the experimental results. Static-titration data of the reduced enzyme with with cyanide indicated that the binding was non-stoicheiometric, the ligand-binding curve being sigmoidal in shape. A Hill plot of the results yielded a Hill coefficient of 2.6.     

Cyanide binding to bovine heart cytochrome c oxidase depleted of subunit III by treatment with lauryl maltoside

Subunit III was removed from beef heart cytochrome oxidase by incubation of the isolated enzyme at 25 degrees C for 24 h in lauryl maltoside buffer at a detergent to protein ratio of 10:1 (w:w). During the course of the incubation, the reaction of the enzyme with cyanide was followed by spectrophotometry in the Soret region. The starting material binds cyanide in a multiexponential process with 70% of the reaction occurring during the slow phase of the reaction at an observed rate of 3.85 X 10(-5) S-1 with 1 mM KCN. More of the enzyme binds cyanide during the fast phase of the reaction at an observed rate of 3.8 X 10(-3) S-1 as subunit III is removed by lauryl maltoside. After 24 h of incubation in lauryl maltoside, the enzyme reacts with cyanide completely in a rapid, single exponential process. When the protein from such an incubation is recovered by cytochrome c affinity chromatography and analyzed for its subunit content, subunit III is absent. The position of the Soret maximum of the oxidized enzyme shifts from its maximum at 418 nm in the starting material to 422 nm in the subunit III-depleted enzyme. The subunit III-depleted enzyme binds cyanide completely in a simple bimolecular reaction with a rate constant of 3.8 M-1 S-1. We discuss this result in terms of the possible structural and functional roles for subunit III in the cytochrome oxidase complex.     

Nitric oxide and cytochrome oxidase: substrate,inhibitor or effector?

Endogenously produced nitric oxide (NO) controls oxygen consumption by inhibiting cytochrome c oxidase, the terminal electron acceptor of the mitochondrial electron transport chain. The oxygen-binding site of the enzyme is an iron/copper (haem a3/CuB) binuclear centre. At high substrate (ferrocytochrome c) concentrations, NO binds reversibly to the reduced iron in competition with oxygen. At low substrate concentrations, NO binds to the oxidized copper. Inhibition at the haem iron site is relieved by dissociation of the NO from the reduced iron. Inhibition at the copper site is relieved by oxidation of the bound NO and subsequent dissociation of nitrite from the enzyme. Therefore, NO can be a substrate, inhibitor or effector of cytochrome oxidase, depending on cellular conditions.     

Detection, characterization, and quenching of the intrinsic fluorescence of bovine heart cytochrome c oxidase

The intrinsic fluorescence of lauryl maltoside solubilized bovine heart cytochrome c oxidase has been determined to arise from tryptophan residues of the oxidase complex. The magnitude of the fluorescence is approximately 34% of that from n-acetyltryptophanamide (NATA). This level of fluorescence is consistent with an average heme to tryptophan distance of 30 A. The majority of the fluorescent tryptophan residues are in a hydrophobic environment as indicated by the fluorescence emission maximum at 328 nm and the differing effectiveness of the quenching agents: Cs+, I-, and acrylamide. Cesium was ineffective up to a concentration of 0.7 M, whereas quenching by the other surface quenching agent, iodide, was complex. Below 0.2 M, KI was ineffective whereas between 0.2 and 0.7 M 15% of the tryptophan fluorescence was found to be accessible to iodide. This pattern indicates that protein structural changes were induced by iodide and may be related to the chaotropic character of KI. Acrylamide was moderately effective as a quenching agent of the oxidase fluorescence with a Stern-Volmer constant of 2 M-1 compared with acrylamide quenching of NATA and the water-soluble enzyme aldolase having Stern-Volmer constants of 12 M-1 and 0.3 M-1, respectively. There was no effect of cytochrome c on the tryptophan emission intensity from cytochrome c oxidase under conditions where the two proteins form a tight, 1:1 complex, implying that the tryptophan residues near the cytochrome c binding site are already quenched by energy transfer to the homes of the oxidase. The lauryl maltoside concentration used to solubilize the enzyme did not affect the fluorescence of NATA.(ABSTRACT TRUNCATED AT 250 WORDS)     

Electron-paramagnetic-resonance studies of structure and function of the two-haem enzymes Pseudomonas cytochrome c peroxidase and beef heart cytochrome c oxidase

Beef heart cytochrome c oxidase contains two cytochromes, a and a3, and Pseudomonas aeruginosa cytochrome c peroxidase has one high- and one low-potential c haem, cHP and cLP. The parallelism in co-ordination and spin states between cytochrome a and haem cHP on the one hand and between cytochrome a3 and haem cLP on the other is illustrated. The two latter haems become accessible to cyanide, when the former are reduced. Such reduction also leads to an activation of the enzymes. Mechanisms are presented in which ferryl forms of cytochromes a3 and haem cLP take part. The enzymes reach an oxidation state, formally the same as resting enzyme, but with different properties.     

The photoreactivity of the copper-NO complexes in cytochrome c oxidase and in other copper-containing proteins

The complexes of NO with CuB of cytochrome c oxidase in which cytochrome a3 may or may not be ligated to cyanide or fluoride are photodissociable. NO does not appear to react with CuB in complexes of cytochrome c oxidase in which sulphide or mercaptans are ligated to the haem iron of cytochrome a3. A comparison is made between the photoreactivity of the complexes of NO with cytochrome c oxidase and those with ceruloplasmin, ascorbate oxidase, and haemocyanin. It is shown that the photoreactivity of CuB 2+.NO in cytochrome c oxidase is not unique for this enzyme, but may also be observed in the complexes of NO with type-1 copper-containing enzymes. This would suggest that the ligation of CuB in cytochrome c oxidase shows some similarity to type-1 copper in blue oxidases.     

Cytochrome c/cytochrome c oxidase interaction. Direct structural evidence for conformational changes during enzyme turnover.

We investigated the interaction between cytochrome c oxidase and its substrate cytochrome c by catalyzing the covalent linkage of the two proteins to yield 1 : 1 covalent enzyme-substrate complexes under conditions of low ionic strength. In addition to the 'traditional' oxidized complex formed between oxidized cytochrome c and the oxidized enzyme we prepared complexes under steady-state reducing conditions. Whereas for the 'oxidized' complex cytochrome c became bound exclusively to subunit II of the enzyme, for the 'steady-state' complex cytochrome c became bound to subunit II and two low molecular mass subunits, most likely VIb and IV. For both complexes we investigated: (a) the ability of the covalently bound cytochrome c to relay electrons into the enzyme, and (b) the ability of the covalently bound enzyme to catalyze the oxidation of unbound (exogenous) ferrocytochrome c. Steady-state spectral analysis (400-630 nm) combined with stopped-flow studies, confirmed that the bound cytochrome c mediated the efficient transfer of electrons from the reducing agent ascorbate to the enzyme. In the case of the latter, the half life for the ascorbate reduction of the bound cytochrome c and that for the subsequent transfer of electrons to haem a were both < 5 ms. In contrast the covalent complexes, when reduced, were found to be totally unreactive towards oxidized cytochrome c oxidase confirming that the previously observed reduction of haem a within the complexes occurred via intramolecular rather than intermolecular electron transfer. Additionally, stopped-flow analysis at 550 nm showed that haem a within both covalent complexes catalyzed the oxidation of exogenous ferrocytochrome c: The second order rate constant for the traditional complex was 0.55x10(6) m(-1) x s(-1) while that for the steady-state was 0.27x10(6) m(-1) x s(-1). These values were approximately 25-50% of those observed for 1 : 1 electrostatic complexes of similar concentrations. These results combined with those of the ascorbate and the electrophoresis studies suggest that electrons are able to enter cytochrome c oxidase via two independent pathways. We propose that during enzyme turnover the enzyme cycles between two conformers, one with a substrate binding site at subunit II and the other along the interface of subunits II, IV and VIb. Structural analysis suggests that Glu112, Glu113, Glu114 and Asp125 of subunit IV and Glu40, Glu54, Glu78, Asp35, Asp49, Asp73 and Asp74 of subunit VIb are residues that might possibly be involved.     

Characterisation of a near infra-red absorption band of the Escherichia coli quinol oxidase, cytochrome o, which is attributable to the high-spin ferrous haem of the binuclear site.

The bacterial quinol oxidase, cytochrome o, is an enzyme which is highly analogous to the better known cytochrome c oxidase, cytochrome aa3, but with the important difference that it lacks the near infra-red absorbing pigment CuA. In this article we report an absorption band in the near IR spectrum of cytochrome o with a maximal absorption at 758 nm, and which is attributable to the ferrous high-spin haem. The 758 nm band has an extinction coefficient of 0.2-0.3 mM-1.cm-1 at 758-800 nm. This region in cytochrome aa3 is dominated by the CuA absorption. The 758 nm absorption is lost on addition of CO or cyanide to the reduced enzyme. The carbon monoxide compound of cytochrome o also has absorbance bands in the near infra-red, and these may be attributable to a low-spin ferrous haem compound.     

A study of the magnetic properties of haem a3 in cytochrome c oxidase by using magnetic-circular-dichroism spectroscopy.

M.c.d. (magnetic-circular-dichroism) spectroscopy was used to study the magnetization properties of the haem centres in cytochrome c oxidase with magnetic fields of between 0 and 5.3 T over the temperature range 1.5--200 K. The oxidized, oxidized cyanide and partially reduced cyanide forms of the enzyme were studied. In the oxidized state only cytochrome a3+ is detectable by m.c.d. spectroscopy, and its magnetization characteristics show it to be a low-spin ferric haem. In the partially reduced cyanide form of the enzyme cytochrome a is in the diamagnetic low-spin ferrous form, whereas cytochrome a3--CN is e.p.r.-detectable and gives an m.c.d.-magnetization curve typical of a low-spin ferric haem. In the oxidized cyanide form of the enzyme both cytochrome a and cytochrome a3--CN are detectable by m.c.d. spectroscopy, although only cytochrome a gives an e.p.r. signal. The magnetization characteristics of haem a3--CN show clearly that its ground state is an electronic doublet and that another state, probably a spin singlet, lies greater than 10 cm-1 above this. These features are well accounted for by an electronic state of spin S = 1 with a predominantly axial distortion, which leaves the doublet, Ms = +/- 1, as the ground state and the component Ms = 0 as the excited state. This state would not give an e.p.r. signal. Such an electronic state could arise either from a ferromagnetic coupling between haem a3+(3)-CN and the cupric ion, Cua3, or form a haem in the Fe(IV) state.