Refsum disease is caused by mutations in the phytanoyl-CoA hydroxylase gene.

Refsum disease is an autosomal-recessively inherited disorder characterized clinically by a tetrad of abnormalities: retinitis pigmentosa, peripheral neuropathy, cerebellar ataxia and elevated protein levels in the cerebrospinal fluid (CSF) without an increase in the number of cells in the CSF. All patients exhibit accumulation of an unusual branched-chain fatty acid, phytanic acid (3,7,11,15-tetramethylhexadecanoic acid), in blood and tissues. Biochemically, the disease is caused by the deficiency of phytanoyl-CoA hydroxylase (PhyH), a peroxisomal protein catalyzing the first step in the alpha-oxidation of phytanic acid. We have purified PhyH from rat-liver peroxisomes and determined the N-terminal amino-acid sequence, as well as an additional internal amino-acid sequence obtained after Lys-C digestion of the purified protein. A search of the EST database with these partial amino-acid sequences led to the identification of the full-length human cDNA sequence encoding PhyH: the open reading frame encodes a 41.2-kD protein of 338 amino acids, which contains a cleavable peroxisomal targeting signal type 2 (PTS2). Sequence analysis of PHYH fibroblast cDNA from five patients with Refsum disease revealed distinct mutations, including a one-nucleotide deletion, a 111-nucleotide deletion and a point mutation. This analysis confirms our finding that Refsum disease is caused by a deficiency of PhyH.

Cloning and characterization of two guide RNA-binding proteins from mitochondria of Crithidia fasciculata: gBP27, a novel protein, and gBP29, the orthologue of Trypanosoma brucei gBP21.

In kinetoplastid protozoa, mitochondrial (mt) mRNAs are post-transcriptionally edited by insertion and deletion of uridylate residues, the information being provided by guide (g)RNAs. Currently popular mechanisms for the editing process envisage a series of consecutive 'cut-and-paste' reactions, carried out by a complex RNP machinery. Here we report on the purification, cloning and functional analysis of two gRNA-binding proteins of 28.8 (gBP29) and 26.8 kDa (gBP27) from mitochondria of the insect trypanosome Crithidia fasciculata. gBP29 and gBP27 proved to be similar, Arg + Ala-rich proteins, with pI values of approximately 10.0. gBP27 has no homology to known proteins, but gBP29 is the C.fasciculata orthologue of gBP21 from Trypanosoma brucei, a gRNA-binding protein that associates with active RNA editing complexes. As measured in UV cross-linking assays, His-tagged recombinant gBP29 and gBP27 bind to radiolabelled poly(U) and synthetic gRNAs, while competition experiments suggest a role for the gRNA 3'-(U)-tail in binding to these proteins. Immunoprecipitates of mt extracts generated with antibodies against gBP29 also contained gBP27 and vice versa. The immunoprecipitates further harbored a large proportion of the cellular content of four different gRNAs and of edited and pre-edited NADH dehydrogenase subunit 7 mRNAs, but only small amounts of mt rRNAs. In addition, the bulk of gBP29 and gBP27 co-eluted with gRNAs from gel filtration columns in the high molecular weight range. Together, these results suggest that the proteins are part of a large macromolecular complex(es). We infer that gBP29 and gBP27 are components of the C.fasciculata editing machinery that may interact with gRNAs.

The superoxide dismutase activity of myeloperoxidase; formation of compound III.

The reaction of superoxide anions with myeloperoxidase (donor: hydrogen-peroxide oxidoreductase, EC 1.11.1.7), which results in the formation of Compound III of myeloperoxidase, was investigated. It is shown that myeloperoxidase has a high affinity for superoxide anions because formation of Compound III was only partially inhibited by high concentrations of superoxide dismutase. Furthermore, when superoxide anions were generated in a mixture of both cytochrome c and myeloperoxidase in the absence of Cl-, only Compound III was formed and reduction of cytochrome c was not observed. In the presence of Cl-, Compound III was also formed and reduction of cytochrome c was inhibited. From the results described in this paper we conclude that Compound III is able to react with superoxide anions, probably resulting in formation of an intermediate (Compound I) which is catalytically active in the oxidation of Cl- to yield hypochlorous acid (HOCl). Because Compound III of myeloperoxidase is formed in phagocytosing neutrophils (Winterbourn, C.C., Garcia, R.C. and Segal, A.W. (1985) Biochem. J. 228, 583-592) we propose that, in vivo, myeloperoxidase also acts as a superoxide dismutase, and via formation of Compound I uses superoxide anions in the formation of HOCl.

The effect of D-penicillamine on human myeloperoxidase, a mechanism for the efficacy of the drug in rheumatoid arthritis.

We investigated the effect of D-penicillamine on the ability of myeloperoxidase, purified from human leukocytes, to catalyse the oxidation of chloride ions to hypochlorite (HOCl) in the presence of H2O2. It is shown that, due to the interaction of D-penicillamine with both myeloperoxidase itself and HOCl, the chlorinating activity of myeloperoxidase in the presence of H2O2 and chloride ions is prevented. A concentration of 100 microM D-penicillamine inhibits the chlorinating activity of myeloperoxidase completely, which Is due to the stabilization of Compound II, an inactive form of the enzyme. In addition, HOCl reacts directly with D-penicillamine. Analysis of the reaction products of D-penicillamine and HOCl showed that D-penicillamine was oxidized to penicillamine disulphide and penicillamine sulphinic acid, and eventually deaminated (indicated by the release of ammonia). Lower concentrations of D-penicillamine (10 microM) inhibited myeloperoxidase less, but still acted as effective scavengers of HOCl. In very low concentrations (1 microM), D-penicillamine did not scavenge HOCl effectively, but rather stimulated the chlorinating activity of myeloperoxidase. However, when instead of D-penicillamine a comparable amount of ascorbate was added, a similar but even larger stimulation was observed. Since the concentration of free D-penicillamine in serum from rheumatoid patients treated with this drug is about 20 microM (Saetre, R. and Rabenstein, D.L. (1978) Anal. Chem. 50, 276-280), the therapeutic effect of D-penicillamine may be due to the protection of tissues against the reactive HOCl released by activated granulocytes at inflammation sites.

The effects of pH and ionic strength on cytochrome c oxidase steady-state kinetics reveal a catalytic and a non-catalytic interaction domain for cytochrome c.

The influence of pH and ionic strength on the steady-state kinetics of purified bovine cytochrome c oxidase was studied by spectrophotometry. At low ionic strength, increasing the pH in the range between 5.4 and 8.6 resulted in a slight decrease in maximal turnover numbers of the high-affinity and the low-affinity reactions. The high-affinity Km was also found to decrease with increasing pH. The ionic-strength dependence of the steady-state kinetics of positively charged cytochrome c oxidase at pH 6.2 and that of negatively charged cytochrome c oxidase at pH 7.8 were similar; in both cases, high-affinity Km values and high-affinity and low-affinity TNmax values increased with ionic strength. The low-affinity Km was independent of both pH and ionic strength. Above I = 100 mM, no low-affinity reaction could be observed. A description of the electrostatic interactions between cytochrome c and cytochrome c oxidase, based on the overall monopoles and overall dipoles of the two proteins, could not explain our data. We propose that at I greater than or equal to 25 mM such an approximation cannot be used for electrostatic interactions between large proteins, since the assumption that all charges on the surfaces of the reacting proteins would contribute equally to the electrostatic interaction is not valid. A qualitative description of electrostatic interactions between the two cytochromes based on limited electrostatic interaction domains on the cytochrome c oxidase surface was found to be in good agreement with all our data and supports the model of Speck et al. (Speck, S.H., Dye, D. and Margoliash, E. (1984) Proc. Natl. Acad. Sci. USA 81, 347-351), who proposed one catalytic and one non-catalytic cytochrome c binding site. It is proposed that the allosteric effect of the cytochrome c at the non-catalytic site is of an electrostatic nature. At high ionic strength (occurring in vivo), this cytochrome c molecule would then no longer affect the catalytic site, resulting in the absence of the low-affinity reaction.

The effect of D-penicillamine on myeloperoxidase: formation of compound III and inhibition of the chlorinating activity.

The inhibitory effect of the anti-arthritic drug D-penicillamine on the formation of hypochlorite (HOCl) by myeloperoxidase from H2O2 and Cl- was investigated. When D-penicillamine was added to myeloperoxidase under turnover conditions, Compound III was formed, the superoxide derivative of the enzyme. Compound III was not formed when D-penicillamine was added in the presence of EDTA or in the absence of oxygen. However, when H2O2 was added to myeloperoxidase, D-penicillamine and EDTA, Compound III was formed. Therefore it is concluded that formation of Compound III is initiated by metal-catalysed oxidation of the thiol group of this anti-arthritic drug, resulting in formation of superoxide anions. Once Compound III is formed, a chain reaction is started via which the thiol groups of other D-penicillamine molecules are oxidized to disulphides. Concomitantly, Compound I of myeloperoxidase would be reduced to Compound II and superoxide anions would be generated from oxygen. This conclusion is supported by experiments which showed that formation of Compound III of myeloperoxidase by D-penicillamine depended on the chloride concentration. Thus, an enzyme intermediate which is active in chlorination (i.e. Compound I) participated in the generation of superoxide anions from the anti-arthritic drug. From the results described in this paper it is proposed that D-penicillamine may exert its therapeutic effect in the treatment of rheumatoid arthritis by scavenging HOCl and by converting myeloperoxidase to Compound III, which is inactive in the formation of HOCl.

The concept of high- and low-affinity reactions in bovine cytochrome c oxidase steady-state kinetics.

(1) Analysis of the data from steady-state kinetic studies shows that two reactions between cytochrome c and cytochrome c oxidase sufficed to describe the concave Eadie-Hofstee plots (Km congruent to 1.10(-8) M and Km congruent to 2.10(-5) M). It is not necessary to postulate a third reaction of Km congruent to 10(-6) M. (2) Change of temperature, type of detergent and type of cytochrome c affected both reactions to the same extent. The presence of only single catalytic cytochrome c interaction site on the oxidase could explain the kinetic data. (3) Our experiments support the notion that, at least under our conditions (pH 7.8, low-ionic strength), the dissociation of ferricytochrome c from cytochrome c oxidase is the rate-limiting step in the steady-state kinetics. (4) A series of models, proposed to describe the observed steady-state kinetics, is discussed.

The chlorinating activity of human myeloperoxidase: high initial activity at neutral pH value and activation by electron donors.

The steady-state activity of myeloperoxidase in the chlorination of monochlorodimedone at neutral pH was investigated. Using a stopped-flow spectrophotometer we were able to show that the enzymic activity at pH 7.2 rapidly declined in time. During the first 50-100 ms after addition of H2O2 to the enzyme, a turnover number of about 320 s-1 per haem was observed. However, this activity decreased rapidly to a value of about 25s-1 after 1 s. This shows that in classical steady-state activity measurements, the real activity of the enzyme at neutral pH is grossly underestimated. By following the transient spectra of myeloperoxidase during turnover it was shown that the decrease in activity was probably caused by the formation of an enzymically inactive form of the enzyme, Compound II. As demonstrated before (Bolscher, B.G.J.M., Zoutberg, G.R., Cuperus, R.A. and Wever, R. (1984) Biochim. Biophys. Acta 784, 189-191) reductants such as ascorbic acid and ferrocyanide convert Compound II, which accumulates during turnover, into active myeloperoxidase. Activity measurements in the presence of ascorbic acid showed, indeed, that the moderate enzymic activity was higher than in the absence of ascorbic acid. With 5-aminosalicylic acid present, however, the myeloperoxidase activity remained at a much higher level, namely about 150 s-1 per haem during the time interval from 100 ms to 5 s after mixing. From combined stopped-flow/rapid-scan experiments during turnover it became clear that in the presence of 5-aminosalicylic acid the initially formed Compound II was rapidly converted back to native enzyme. Presteady-state experiments showed that 5-aminosalicylic acid reacted with Compound II with a K2 of 3.2 x 10(5) M-1.s-1, whereas for ascorbic acid a K2 of 1.5 x 10(4) M-1.s-1 was measured at pH 7.2. In the presence of 5-aminosalicylic acid during the time interval in which the myeloperoxidase activity remained constant, a Km for H2O2 at pH 7.2 was determined of about 30 microM at 200 mM chloride. In the absence of reductants the same value was found during the first 100 ms after addition of H2O2 to the enzyme. The physiological consequences of these findings are discussed.

A steady-state study on the formation of Compounds II and III of myeloperoxidase.

The reaction between native myeloperoxidase and hydrogen peroxide, yielding Compound II, was investigated using the stopped-flow technique. The pH dependence of the apparent second-order rate constant showed the existence of a protonatable group on the enzyme with a pKa of 4.9. This group is ascribed to the distal histidine imidazole, which must be deprotonated to enable the reaction of Compound I with hydrogen peroxidase to take place. The rate constant for the formation of Compound II by hydrogen peroxide was 3.5.10(4) M-1.s-1. During the reaction of myeloperoxidase with H2O2, rapid reduction of added cytochrome c was observed. This reduction was inhibitable by superoxide dismutase, and this demonstrates that superoxide anion radicals are generated. When potassium ferrocyanide was used as an electron donor to generate Compound II from Compound I, the pH dependence of the apparent second-order rate constant indicated involvement of a group with a pKa of 4.5. However, with ferrocyanide as an electron donor, protonation of the group was necessary to enable the reaction to take place. The rate constant for the generation of Compound II by ferrocyanide was 1.6.10(7) M-1.s-1. We also investigated the reaction of Compound II with hydrogen peroxide, yielding Compound III. Formation of Compound III (k = 50 M-1.s-1) proceeded via two different pathways, one of which was inhibitable by tetranitromethane. We further investigated the stability of Compound II and Compound III as a function of pH, ionic strength and enzyme concentration. The half-life values of both Compound II and Compound III were independent of the enzyme concentration and ionic strength. The half-life value of Compound III was pH-dependent, showing a decreasing stability with increasing pH, whereas the stability of Compound II was independent of pH over the range 3-11.

Separation, stability and kinetics of monomeric and dimeric bovine heart cytochrome c oxidase.

The stability of monomeric and dimeric bovine heart cytochrome c oxidase in laurylmaltoside-containing buffers of high ionic strength allowed separation of the two forms by gel-filtration high-performance liquid chromatography (HPLC). A solution of the dimeric oxidase could be diluted without monomerisation. Both monomeric and dimeric cytochrome c oxidase showed biphasic steady-state kinetics when assayed spectrophotometrically at low ionic strength. Thus, the biphasic kinetics did not result from negative cooperativity between the two adjacent cytochrome c binding sites of the monomers constituting the dimeric oxidase. On polyacrylamide gels in the presence of sodium dodecyl sulphate (SDS) a fraction of subunit III of the dimeric enzyme migrated as a dimer, a phenomenon not seen with the monomeric enzyme. This might suggest that in the dimeric oxidase subunit III lies on the contact surface between the protomers. If so, the presumably hydrophobic interaction between the two subunits III resisted dissociation by SDS to some extent. Addition of sufficient ascorbate and cytochrome c to the monomeric oxidase to allow a few turnovers induced slow dimerisation (on a time-scale of hours). This probably indicates that one of the transient forms arising upon reoxidation of the reduced enzyme is more easily converted to the dimeric state than the resting enzyme. Gel-filtration HPLC proved to be a useful step in small-scale purification of cytochrome c oxidase. In the presence of laurylmaltoside the monomeric oxidase eluted after the usual trace contaminants, the dimeric Complex III and the much larger Complex I. The procedure is fast and non-denaturing, although limited by the capacity of available columns.

Spectral properties of myeloperoxidase compounds II and III.

In order to resolve the confusion about the spectral properties of myeloperoxidase Compound II and Compound III (myeloperoxidase is donor:hydrogen-peroxide oxidoreductase, EC 1.11.1.7), the absorbance spectra in the visible and ultraviolet regions were measured under conditions where either Compound II or Compound III was present. Peak positions, isosbestic points and absorption coefficients are presented. We conclude that in most studies on Compound II or Compound III, mixtures of these compounds had been present. Our data indicate that the relative contributions of Compound II and Compound III in a sample can be determined from the absorbance ratio A625nm/A456nm. The optical absorbance spectrum of myeloperoxidase compound III was not affected by pH (pH 3-8). The absorbance spectrum of Compound II, however, was dependent on pH. The absorbance spectrum of Compound II at high pH is described.

Human cytochrome c oxidase isoenzymes from heart and skeletal muscle; purification and properties.

Human cytochrome c oxidase was isolated in an active form from heart and from skeletal muscle by a fast, small-scale isolation method. The procedure involves differential solubilisation of the oxidase from mitochondrial fragments by laurylmaltoside and KCl, followed by size-exclusion high-performance liquid chromatography. Polyacrylamide gel electrophoresis in the presence of sodium dodecyl sulphate showed differences between the subunit VI region of cytochrome c oxidases from human heart and skeletal muscle, suggesting different isoenzyme forms in the two organs. This finding might be of importance in explaining mitochondrial myopathy which shows a deficiency of cytochrome c oxidase in skeletal muscle only. In SDS polyacrylamide gel electrophoresis most human cytochrome c oxidase subunits migrated differently from their bovine counterparts. However, the position of subunits III and IV was the same in the human and in the bovine enzymes. The much higher mobility of human cytochrome c oxidase subunit II is explained by a greater hydrophobicity of this polypeptide than of that of the subunit II of the bovine enzyme.

FSBA modifies both alpha- and beta-subunits of F1 specifically and can be bound together with AXP at the same alpha-subunit.

Binding of 1 mole 5'-fluorosulfonylbenzoyladenosine (FSBA) per mol F1 induces about 50% inhibition of ATPase activity and 80% inhibition of ITPase activity. The binding of additional ligand results in a further inhibition of both activities. Maximally 5 mol/mol F1, causing complete inhibition of activity, can be bound. Using radioactive FSBA more label is found on alpha-subunits than on beta-subunits under the usual buffer conditions. The modified amino acids are alpha-Tyr300, alpha-Tyr244 and beta-Tyr368. Binding of FSBA, at least up to 3 mol/mol F1, does not result in loss of bound ADP, whether the starting enzyme contains 2, 3 or 4 bound nucleotides. Added adenine nucleotides compete with FSBA only for binding that results in modification of beta-subunits, shifting the alpha/beta ratio of bound label to higher values. It is concluded that the alpha-subunits contain two hydrophobic pockets for the binding of nucleoside moieties, with a different orientation relative to the P-loop. One pocket contains alpha-Tyr244 and alpha-Tyr300, the other beta-Tyr368. Since, however, in the binding of adenine nucleotide di- or triphosphates the P-loop is involved, only one of these ligands can bind per subunit. The previously not understood binding characteristics of several substrate analogues have now become interpretable on the assumption that also the structurally homologous beta-subunits contain 2 pockets where nucleoside moieties can bind. The kinetic effects of FSBA binding indicate that the first FSBA binds at the regulatory site that has a high affinity for ADP and pyrophosphate. Binding of pyrophosphate at this high-affinity regulatory site increases the Vmax of the enzyme, while binding at a second regulatory site, a low-affinity site, increases the rate of binding of FSBA with a factor of about 3. Binding of bicarbonate at this latter site is responsible for the disappearance of the apparent negative cooperativity of the ATPase activity.

The reactivity of thiol groups in bovine heart cytochrome c oxidase towards 5,5'-dithiobis(2-nitrobenzoic acid).

Bovine heart cytochrome c oxidase consists of 12 stoicheiometric polypeptide chains of at least 11 different types. The enzyme contains 14--16 cysteine residues; the distribution of nearly all cysteine residues over the subunits has been established. In native cytochrome c oxidase two thiol groups reacted rapidly and stoicheiometrically with 5,5'-dithiobis(2-nitrobenzoic acid) (DTNB). These thiol groups are located in subunits I and III, respectively. This implies that subunit I is not fully buried in the hydrophobic core of the enzyme. After dissociation of the enzyme by sodium dodecyl sulphate more thiol groups became available to DTNB, in addition to those in subunits I and III, at least one in subunit II, two in fraction V/VI and one to two in the smallest subunit fraction. It is shown that separation of the subunits of cytochrome c oxidase by gel permeation chromatography in the presence of sodium dodecyl sulphate depends on the pH of the elution medium. The elution volume of subunits I, III and VII is dependent on pH, that of the others independent.

Subunit IV of human cytochrome c oxidase, polymorphism and a putative isoform.

As part of our study of isoenzyme forms of human cytochrome c oxidase, we purified subunit IV from human heart and skeletal muscle with reversed-phase HPLC and determined the N-terminal amino acid sequences and the electrophoretic mobility. The N-terminus of human heart subunit IV proved to be ragged with 30% of the protein lacking the first three residues. Also a Tyr/Phe polymorphism was observed at residue 16. No differences in N-terminal sequence and electrophoretic mobility were observed between subunit IV of cytochrome c oxidase from human heart and skeletal muscle. Therefore, our results suggest that identical subunits IV are present in cytochrome c oxidase from human heart and skeletal muscle. A putative isoform of subunit IV with a blocked N-terminus was purified from human heart cytochrome c oxidase, which proved to have a different retention time on a reversed-phase column and also a slightly higher electrophoretic mobility on an SDS-polyacrylamide gel compared to the native subunit IV. We could not demonstrate the existence of isoforms of subunit IV in human skeletal muscle.

Bovine cytochrome c oxidases, purified from heart, skeletal muscle, liver and kidney, differ in the small subunits but show the same reaction kinetics with cytochrome c.

(1) Polyacrylamide gel electrophoresis in the presence of sodium dodecyl sulphate of purified cytochrome c oxidase preparations revealed that bovine kidney, skeletal muscle and heart contain different cytochrome c oxidase isoenzymes, which show differences in mobility of the subunits encoded by the nuclear genome. No differences in subunit pattern were observed between the oxidase preparations isolated from kidney and liver. (2) The kinetics of the steady-state reactions between bovine ferrocytochrome c and the four types of bovine cytochrome c oxidase preparation were compared under conditions of both high- and low-ionic strength. Also the pre-steady-state kinetics were studied. Only minor differences were observed in the electron-transfer activity of the isoenzymes. Thus, our experiments do not support the notion that the subunits encoded by the nuclear genome act as modulators conferring different activities to the isoenzymes of cytochrome c oxidase. (3) The cytochrome c oxidase preparation from bovine skeletal muscle was found to consist mainly of dimers, whereas the enzymes isolated from bovine kidney, liver and heart were monomeric.

The effect of detergents on bovine cytochrome c oxidase: a kinetic approach.

(1) Investigation of the relationship between the detergent concentration and steady-state and pre-steady-state kinetics of cytochrome c oxidase proved to be a valid approach in the study of protein-detergent interaction. (2) Laurylmaltoside, sodium cholate and Triton X-100 influenced the kinetics of cytochrome c oxidase cooperatively at detergent concentrations near their critical micelle concentration. This mode of interaction reflects disaggregation of the oxidase as a result of cooperative binding of the detergent. (3) Addition of increasing concentrations of Tween-80 to the aggregated enzyme caused a more gradual decrease in aggregation of the oxidase, which did not result in a change in activity of the enzyme. This suggests that aggregation of cytochrome c oxidase occurs in a highly regular manner in which no catalytic sites are shielded off. (4) Oxidase aggregates present at detergent concentrations below the critical micelle concentration of laurylmaltoside and Triton X-100 showed considerable activity. Their kinetics were equal to those of the oxidase in Tween-80, suggesting that the protein molecules are aligned in a similar way in all oligomers. Aggregates present in low concentrations of sodium cholate showed turnover rates that were twice as low as those observed with other aggregates. (5) Solubilisation of the oxidase by sodium cholate or Triton X-100 resulted in almost complete inhibition of enzymic activity, whereas the association rate of ferrocytochrome c was almost equal to that found for monomeric oxidase in laurylmaltoside. These results are in agreement with a mixed-type inhibition.

Isoforms of cytochrome c oxidase in tissues and cell lines of the mouse.

The subunit pattern of immunopurified cytochrome c oxidase from cultured mouse cells and mature tissues of the mouse was investigated by electrophoretic analysis. In mature tissues two forms of cytochrome c oxidase could clearly be identified on the basis of differences in morbidity or staining intensity of subunits VIa and VIII. One form was present in muscle and heart, and the other in liver, kidney and spleen. In lung both forms were found. In the thymus, subunit VIII showed the characteristics of subunit VIII found in muscle and heart, whereas subunit VIa resembled subunit VIa found in liver. This suggest the existence of a third cytochrome c oxidase isoform. The subunits of cytochrome c oxidase from cultured cell lines showed no differences between the various cell lines and resembled those of mature mouse liver tissue. The cytochrome c oxidase isoform from cultured proliferating cells might therefore be the same as the one found in liver. Alternatively, it might represent either a normally occurring fetal isoform, or a form specific for poorly differentiated cultured cells.

The isolation of bovine-heart cytochrome c oxidase subunits. Dependence on phospholipid and cholate-content.

The polypeptide chains of bovine-heart cytochrome c oxidase were preparatively isolated by a simple large-scale procedure based on gel permeation chromatography in the presence of sodium dodecyl sulphate. The resolution of the subunits as a function of the cholate and phospholipid content of the preparation was investigated. Cholate, and to a lesser extent, phospholipids interfere with the separation of the subunits; however, they do not prevent dissociation of the enzyme by SDS. Bovine-heart cytochrome c oxidase consists of six major subunits (estimated molecular weights in thousands: 40, 25, 20, 14, 12 and 10). In addition, the enzyme preparation contains at least five minor constituents, present in less than stoichiometric amounts. The first two of the three large subunits, all of which are hydrophobic, have amino-terminal N-formylmethionine. Subunit III, however, has a free methionine N-terminus.

Human heart cytochrome c oxidase subunit VIII. Purification and determination of the complete amino acid sequence.

Subunit VIII was purified from a preparation of the human heart cytochrome c oxidase and its complete amino acid sequence was determined. The sequence proved to be much more related to that of the bovine liver oxidase subunit VIII than to that found in bovine heart. Our finding of a 'liver-type' subunit VIII in the human heart enzyme suggests that either there are no isoforms of human subunit VIII or the human oxidase does not show the type of tissue specificity that has been reported for the oxidase in other mammals.