The interconversion between monomeric and dimeric bovine heart cytochrome c oxidase

Monomers and dimers of bovine heart cytochrome c oxidase (EC 1.9.3.1.) were separated by gel filtration chromatography on Ultrogel AcA 34 or by sucrose gradient centrifugation. Factors influencing the interconversion of the two aggregation states of this enzyme were analyzed. At very low ionic strength, in the presence of dodecyl maltoside, monomers were the main species. Salts appeared to stabilize the dimeric form, divalent cations being more efficient than monovalent. High enzyme concentrations favoured the formation of dimers, also at low ionic strength. The type of detergent had a strong influence on the monomer-dimer interconversion; in Triton X-100 and dodecyl maltoside (at high ionic strength) cytochrome c oxidase was homogenously dispersed in its dimeric form, while in Tween-80 gel filtration showed only large particles eluting in the void volume. In cholate monomers and aggregates were observed but no dimers. The aggregation state had an influence on the steady state kinetics of the ferrocytochrome c oxidase activity. Monomers showed linear Eadie-Hofstee plots, whilst the dimeric and aggregated enzyme gave nonlinear Eadie-Hofstee plots. Ionic strength, enzyme concentration and type of detergent were affecting the enzyme's kinetics in a way consistent with the molecular form obtained by the gel filtration or sedimentation analysis. The data support a negative cooperative mechanism for the interaction of cytochrome c with the dimeric enzyme, as proposed earlier (K.A. Na?ecz et al., (1983) Biochem. Biophys. Res. Commun., 114, 822-828).     

Studies on the role of the oligomeric state and subunit III of cytochrome oxidase in proton translocation

Anion-exchange fast protein liquid chromatography in the presence of lauryldimethylamine N-oxide (LDAO) was introduced to separate cytochrome oxidase into different complexes that either did or did not contain subunit III. Both kinds of enzyme complex exhibited H+ translocation after reconstitution into phospholipid vesicles, but with a significantly (approx. 50-60%) reduced H+/e- ratio as compared with unchromatographed enzyme. The anion-exchange FPLC fractions of the enzyme (with or without subunit III) sedimented more slowly than the control enzyme upon sucrose gradient centrifugation in the presence of cholate and a high potassium phosphate concentration. When the control enzyme was subjected to the sucrose gradient centrifugation in the presence of LDAO or Triton X-100, instead of cholate, one band containing all subunits was observed, which sedimented slowly like the FPLC fractions. Transfer of this band to cholate medium, and reapplication on the sucrose gradient (with cholate), yielded both a slow- and a fast-migrating band after centrifugation. Enzyme complexes that sedimented slowly or rapidly in the sucrose gradients revealed longer and shorter elution times, respectively, in gel filtration FPLC. This suggests that these complexes corresponds to monomers and dimers of cytochrome oxidase. Solubilization of proteoliposomes and subsequent sucrose gradient centrifugation in cholate yielded one fast-migrating band for the untreated enzyme, but both a fast- and a slow-migrating band for the anion-exchange FPLC-treated enzyme, which was exclusively slow-migrating before reconstitution into liposomes. It is suggested that dimerisation of monomeric cytochrome oxidase may be favoured when the enzyme encounters a membranous milieu, and that the dimeric structure might be necessary for proton translocation.     

The aggregation state of bovine heart cytochrome c oxidase and its kinetics in monomeric and dimeric form

The monomeric and dimeric forms of bovine cytochrome c oxidase (EC 1.9.3.1) were obtained from gel filtration chromatography on Ultrogel AcA 34 and analyzed. Both species contained all 12-13 subunits described for this enzyme. In the dimer 320 molecules [3H]dodecyl-beta-D-maltoside were bound per heme aa3 and in the monomer 360 molecules per heme aa3. The monomers contained 10 mol of tightly bound phospholipid/mol heme aa3 and the dimers 14. Sedimentation coefficients of 15.5-18 S for the dimer and 9.6 S for the monomer were calculated from sucrose density centrifugation analysis and analytical centrifugation. By the laser beam light-scattering technique a Stokes radius of 70 A for the dimeric detergent-lipid-protein complex was measured. From those parameters and the densitometric determined partial specific volumes of the detergent and the enzyme, the molecular weights of 400,000 for the protein moiety of the dimer and 170,000-200,000 for the monomer were calculated. Under very low ionic strength conditions the monomer/dimer equilibrium was found to be dependent on the protein concentration. At low enzyme concentrations (10(-9) M) monomers were predominant, whereas at concentrations above 5 X 10(-6) M the amounts of dimers and higher aggregates were more represented. The cytochrome c oxidase activity, measured spectrophotometrically and analyzed by Eadie-Hofstee plot, was biphasic as a function of cytochrome c concentration for the dimeric enzyme. Pure monomers gave monophasic kinetics. The data, fitting with a homotropic negative cooperative mechanism for the dimer of cytochrome c oxidase, are discussed and compared with other described mechanisms.     

Biochemical and biophysical properties of purified phospholipid vesicles containing bovine heart cytochrome c oxidase

Liposomes containing bovine heart cytochrome c oxidase (COV) prepared by the cholate dialysis technique were purified from those devoid of the enzyme using discontinuous sucrose density ultra centrifugation to eliminate interference in proton-pumping assays. This technique was also used to purify liposomes containing cytochrome c oxidase depleted in subunit III (COV-III), a COX enzyme preparation with altered subunit structure, to assess if the technique could be applied to COX enzymes in which structural and functional changes have occurred. Upon discontinuous sucrose density ultra gradient ultracentrifugation, either COV or COV-III were separated into two bands. Liposomes devoid of enzyme sedimented into the 12% sucrose layer, whereas enzyme-containing liposomes (pCOV or pCOV-III) were found in the 13% sucrose layer. The yield of both pCOV or pCOV-III was greater than 60% (based on heme aa(3) content), suggesting a similar distribution of cytochrome c oxidase (COX) and subunit III-depleted enzyme (COX-III) in the purified liposomes. The number of COX or COX-III molecules per phospholipid vesicle in purified fractions was estimated to be two. Removal of subunit III (M(r)=29,918) from COX resulted in a 30% decrease in electron transfer activity (either in COV-III or pCOV-III) when compared with COV and pCOV, respectively. Both pCOV and pCOV-III exhibited low endogenous proton permeability, as assessed by possessing high respiratory control ratios (14 and greater) and by having similar valinomycin concentration dependencies for stimulation of electron transfer activity in the presence of saturating amounts of CCCP. COV-III and pCOV-III exhibited a 39-44% decrease in proton-pumping activity when compared with COV and pCOV. These results showed that the separation of COX containing liposomes from those lacking enzyme by sucrose density gradient centrifugation can be used to characterize the biophysical properties of these liposomes.     

Spectroscopic and functional properties of subunit III-depleted cytochrome oxidase

Beef heart cytochrome c oxidase has been depleted of subunit III by treatment with chymotrypsin. The removal of subunit III has been evaluated by sodium dodecyl sulfate-polyacrylamide gel electrophoresis and gel fluorography of preparations of the oxidase labeled with [14C]dicyclohexylcarbodiimide prior to proteolysis. Removal of subunit III resulted in a perturbation of the visible spectrum of reduced cytochrome oxidase. Subunit III-depleted oxidase is spectroscopically very similar to the oxidase from Paracoccus denitrificans. When reconstituted into liposomes, the depleted enzyme still pumped protons in response to a pulse of reduced cytochrome c. The H+/e- stoichiometry averaged 0.5. Redox-linked proton translocation could be observed only when respiratory control ratios were higher than 3 and the reductant pulse was of a magnitude that allowed for no more than 5 turnovers of the oxidase.     

Characterization of electron-transfer and proton-translocation activities in bovine heart mitochondrial cytochrome c oxidase deficient in subunit III

The electron-transfer and proton-translocation activities of cytochrome c oxidase deficient in subunit III (Mr 29 884) prepared by native gel electrophoresis [Ludwig, B., Downer, N. W., & Capaldi, R. A. (1979) Biochemistry 18, 1401-1407] have been investigated. This preparation has been depleted of 82-87% of its subunit III content as quantitated by Coomassie Brilliant Blue staining intensity on sodium dodecyl sulfate-polyacrylamide gel electrophoresis and [14C]dicyclohexylcarbodiimide labeling. The maximum rate of electron transfer of the subunit III deficient enzyme at pH 6.5 is 383 s-1, 78% of control enzyme. Neither the high-affinity site (Km = 10(-8) M) nor the low-affinity site (Km = 10(-6) M) of the cytochrome c kinetic interaction with cytochrome c oxidase is affected by the removal of subunit III. Subunit III deficient cytochrome c oxidase retains the ability to bind cytochrome c in both the high- and low-affinity sites as determined in direct thermodynamic binding experiments. Liposomes containing this preparation exhibit a respiratory control ratio [Hinkle, P. C., Kim, J. J., & Racker, E. (1972) J. Biol. Chem. 247, 1338-1341] of 3.9, while liposomes containing control enzyme exhibit a ratio of 4.3, suggesting that they have a similar proton permeability. Vectorial proton translocation initiated by the addition of ferrocytochrome c in liposomes containing subunit III deficient enzyme is decreased by 64% compared to those containing control enzyme. When the proton-translocated to electron-transferred ratio is measured in these phospholipid vesicles at constant enzyme turnover, removal of subunit III from the enzyme decreases the ratio from 0.52 to 0.21, a 60% decrease.(ABSTRACT TRUNCATED AT 250 WORDS)     

Effect of subunit III removal on control of cytochrome c oxidase activity by pH

Studies were undertaken to assess the postulated involvement of subunit III in the proton-linked functions of cytochrome c oxidase. The effect of pH on the steady-state kinetic [corrected] parameters of subunit III containing and subunit III depleted cytochrome oxidase was determined by using beef heart and rat liver enzymes reconstituted into phospholipid vesicles. The TNmax and Km values for the III-containing enzyme increase with decreasing pH in a manner quantitatively similar to that reported by Thornstrom et al. [(1984) Chem. Scr. 24, 230-235], giving three apparent pKa values of less than 5.0, 6.2, and 7.8. The maximal activities of the subunit III depleted enzymes (beef heart and rat liver) show a similar dependence on pH, but the Km values are consistently higher than those of the III-containing enzyme, an effect that is accentuated at low pH. The pH dependence of TNmax/Km for both forms of the enzyme (+/- subunit III) indicates that protonation of a group with an apparent pKa of 5.7 lowers the affinity for substrate (cytochrome c) independently of a continued increase in maximal velocity. N,N'-Dicyclohexylcarbodiimide (DCCD) decreases the pH responsiveness of the electron-transfer activity to the same extent in both III-containing and III-depleted enzymes, indicating that this effect is mediated by a peptide other than subunit III. Control of intramolecular electron transfer by a transmembrane pH gradient (or alkaline intravesicular pH) is shown to occur in cytochrome oxidase vesicles with cytochrome c as the electron donor, in agreement with results of Moroney et al. [(1984) Biochemistry 23, 4991-4997] using hexaammineruthenium(II) as the reductant.(ABSTRACT TRUNCATED AT 250 WORDS)     

Subunit III of cytochrome c oxidase is expressed in Paracoccus denitrificans

Polyclonal antibodies have been obtained against a synthetic dodecapeptide identical to the aminoacid sequence 120-131 DSPIKDGVWPPE (inferred from its DNA sequence) of Paracoccus denitrificans cytochrome c oxidase subunit III. The antibodies had a titer higher than 1:10000 when tested against the antigen. These antibodies have been used to produce immunological evidence that, despite the fact that subunit III is not isolated with cytochrome c oxidase, it exists in Paracoccus denitrificans lysates. The antibodies did not show reactivity with bovine heart cytochrome c oxidase either by ELISA or immunoblotting. It was also shown that the antibodies react with a single polypeptide present in Paracoccus denitrificans cell lysates, having an apparent molecular weight close to that of subunit III of bovine heart oxidase.     

Ionic-strength-dependence of the oxidation of native and pyridoxal 5''-phosphate-modified cytochromes c by cytochrome c oxidase.

The ionic-strength-dependences of the rate constants (log k plotted versus square root of 1) for oxidation of native and pyridoxal 5'-phosphate-modified cytochromes c by three different preparations of cytochrome c oxidase have complex non-linear character, which may be explained on the basis of present knowledge of the structure of the oxidase and the monomer-dimer equilibrium of the enzyme. The wave-type curve (with a minimum and a maximum) for oxidation of native cytochrome c by purified cytochrome c oxidase depleted of phospholipids may reflect consecutively inhibition of oxidase monomers (initial descending part), competition between this inhibition and dimer formation, resulting in increased activity (second part with positive slope), and finally inhibition of oxidase dimers (last descending part of the curve). The dependence of oxidation of native cytochrome c by cytochrome c oxidase reconstituted into phospholipid vesicles is a curve with a maximum, without the initial descending part described above. This may reflect the lack of pure monomers in the vesicles, where equilibrium is shifted to dimers even at low ionic strength. Subunit-III-depleted cytochrome c oxidase does not exhibit the maximum seen with the other two enzyme preparations. This may mean that removal of subunit III hinders dimer formation. The charge interactions of each of the cytochromes c (native or modified) with the three cytochrome c oxidase preparations are similar, as judged by the similar slopes of the linear dependences at I values above the optimal one. This shows that subunit III and the phospholipid membrane do not seem to be involved in the specific charge interaction of cytochrome c oxidase with cytochrome c.     

Biophysical and biochemical characterization of reconstituted and purified Rhodobacter sphaeroides cytochrome c oxidase in phospholipid vesicles sheds insight into its functional oligomeric structure

Discontinuous sucrose gradient ultracentrifugation was used to separate liposomes containing Rhodobacter sphaeroides cytochrome c oxidase (pCOV) from liposomes devoid of the enzyme, and the biophysical and biochemical properties of pCOV were compared to unpurified liposomes containing cytochrome c oxidase (COV). Isolated and purified R. sphaeroides cytochrome c oxidase (COX) was reconstituted into asolectin phospholipid vesicles by cholate dialysis, and this preparation was purified further on a discontinuous sucrose gradient to isolate only those vesicles which contained the enzyme (pCOV). After centrifugation at 300,000g for 22h, 80% of the enzyme recovered was in a single band. The number of COX molecules per pCOV liposome was estimated by measuring the visible absorbance spectrum of cytochrome c oxidase (for heme aa(3)) and inorganic phosphate concentration (for phospholipid). The number of COX molecules incorporated per pCOV was estimated to be approximately one (0.72+/-0.19-1.09+/-0.28). The pCOV exhibited similar physical properties as COV; respiratory control ratios (indicators of endogenous proton permeability) and maximum enzymatic turnover number at pH 7.4 were comparable (6.0+/-1.3 and 535+/-130s(-1)). Furthermore, proton pumping activities of the pCOV were at least 70% of COV, indicating that discontinuous sucrose gradient centrifugation is a useful technique for functional experiments in R. sphaeroides cytochrome c oxidase. Our results suggest that the monomeric form of R. sphaeroides COX when reconstituted into a phospholipid bilayer is completely functionally active in its ability to perform electron transfer and proton pumping activities of the enzyme.     

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.     

Studies on the oligomeric state of isolated cytochrome oxidase using cross-linking reagents

(1) Sucrose gradient centrifugation of cytochrome oxidase in the presence of Triton X-100 gave one slowly sedimenting green band. After cross-linking with dithiobis(succinimidylpropionate) (DSP), two green bands were observed, one sedimenting like the control and the other one more rapidly. Only the slowly sedimenting band was observed if the cross-linker was cleaved by dithiothreitol before centrifugation. (2) The rapidly sedimenting band in the Triton-containing sucrose gradient is probably the internally cross-linked dimer of cytochrome oxidase; the one sedimenting slowly is the monomeric enzyme. (3) Cross-linking with DSP after monomerization yields a small fraction of internally cross-linked dimers in addition to the internally cross-linked monomers. Under similar conditions, but using the shorter cross-linker disuccinimidyl tartarate (DST), no dimers are detected. (4) Both DSP and DST cross-link the dimeric enzyme so that it could no longer be monomerized by centrifugation in Triton, unless the cross-link is cleaved. (5) Polypeptide analysis using two-dimensional gel electrophoresis of cross-linked dimers and monomers suggest that subunit VIb is involved in intermonomeric cross-linking of dimeric enzyme by DSP.     

Homology between bacterial DNA and bovine mitochondrial DNA encoding cytochrome c oxidase subunit III

A segment of mitochondrial DNA encoding the bovine cytochrome c oxidase subunit III gene was isolated and inserted into an Escherichia coli plasmid vector. A 556 base pair fragment of the insert DNA representing about 70% of the 3'-end of the subunit III gene was used to search for homology with bacterial DNA from strains that contain heme aa3-type cytochrome c oxidases. Bacillus subtilis, Thermus thermophilus, and PS3 DNAs all showed strong hybridization to the probe, whereas Paracoccus denitrificans and Rhodopseudomonas sphaeroides DNAs showed only weak hybridization to the probe, even under low stringency conditions.     

An active cytochrome c oxidase depleted of subunit III prepared by covalent chromatography on yeast cytochrome c

A covalent chromatography technique is described for the preparation of an active cytochrome c oxidase from bovine heart devoid of subunit III. Yeast cytochrome c is immobilized on a Sepharose 4B gel, its cysteine 107 activated and reacted with the oxidase. Elution with Triton X-100 releases an oxidase devoid of subunit III, which is recovered after elution with β-mercaptoethanol.     

A gene in Paracoccus for subunit III of cytochrome oxidase

The region of Paracoccus denitrificans chromosome where the genes coding for cytochrome oxidase (cytochrome aa3) subunits are located has been cloned. DNA sequencing revealed an open reading frame that codes for a protein homologous to the subunit III of the eukaryotic, mitochondrial enzyme. This subunit is absent from the isolated Paracoccus oxidase. It now seems that it is part of the native enzyme in the bacterial cytoplasmic membrane. This may explain the observed discrepancies in the function of the isolated enzyme.     

C-terminal truncation and histidine-tagging of cytochrome c oxidase subunit II reveals the native processing site, shows involvement of the C-terminus in cytochrome c binding, and improves the assay for proton pumping

To enable metal affinity purification of cytochrome c oxidase reconstituted into phospholipid vesicles, a histidine-tag was engineered onto the C-terminal end of the Rhodobacter sphaeroides cytochrome c oxidase subunit II. Characterization of the natively processed wildtype oxidase and artificially processed forms (truncated with and without a his-tag) reveals Km values for cytochrome c that are 6-14-fold higher for the truncated and his-tagged forms than for the wildtype. This lowered ability to bind cytochrome c indicates a previously undetected role for the C-terminus in cytochrome c binding and is mimicked by reduced affinity for an FPLC anion exchange column. The elution profiles and kinetics indicate that the removal of 16 amino acids from the C-terminus, predicted from the known processing site of the Paracoccus denitrificans oxidase, does not produce the same enzyme as the native processing reaction. MALDI-TOF MS data show the true C-terminus of subunit II is at serine 290, three amino acids longer than expected. When the his-tagged form is reconstituted into lipid vesicles and further purified by metal affinity chromatography, significant improvement is observed in proton pumping analysis by the stopped-flow method. The improved kinetic results are attributed to a homogeneous, correctly oriented vesicle population with higher activity and less buffering from extraneous lipids.     

Stopped-flow studies of cytochrome oxidase reconstituted into liposomes: proton pumping and control of activity

The transient kinetics of proton pumping and the electron transfer properties of cytochrome oxidase inserted into small unilamellar vesicles have been investigated by stopped-flow spectrophotometry. In the presence of valinomycin, proton pumping and cytochrome c oxidation by cytochrome oxidase are synchronous up to rate constants of approximately 9 sec-1. Moreover, the enzyme depleted of subunit III ("three-less oxidase") was also shown to pump protons, although with a significantly smaller stoichiometry. Thus, subunit III is not the only (or even the main) proton channel, although it may be involved in the regulation of activity. The kinetics of cytochrome c oxidation by COV in the absence and in the presence of ionophores have been investigated. Analysis of the time course of the process in the transient and steady state phases indicates that the onset of control by the electrochemical gradient follows the transfer of four electrons, i.e., one complete turnover of the oxidase. Two possible alternative interpretations for the control of the turnover phase are presented and discussed.     

Electrophoretically monodisperse cytochrome c oxidases

A discontinuous gradient polyacrylamide gel electrophoresis under nondenaturing conditions has been used to demonstrate monodispersity of procaryotic and eucaryotic cytochrome c oxidase preparations. Alkaline treated bovine enzyme which contains nine subunits as analysed by subsequent discontinuous SDS-polyacrylamide gel electrophoresis is a monodisperse dimer in 0.1% Triton X-100 and a monomer in 0.1% dodecyl maltoside. The Mr-values corrected for bound detergent are 286,000 in Triton X-100 and 152,000 in dodecyl maltoside respectively. The two-subunit bacterial cytochrome c oxidase of Paracoccus denitrificans is proved to be a monomer with a corrected Mr of 76,000 in both nonionic detergents Triton X-100 and dodecyl maltoside.     

Copper and manganese electron spin resonance studies of cytochrome c oxidase from Paracoccus denitrificans

The two-subunit cytochrome c oxidase from Paracoccus denitrificans contains two heme a groups and two copper atoms. However, when the enzyme is isolated from cells grown on a commonly employed medium, its electron paramagnetic resonance (EPR) spectrum reveals not only a Cu(II) powder pattern, but also a hyperfine pattern from tightly bound Mn(II). The pure Mn(II) spectrum is observed at -40 degrees C; the pure Cu(II) spectrum can be seen with cytochrome c oxidase from P. denitrificans cells that had been grown in a Mn(II)-depleted medium. This Cu(II) spectrum is very similar to that of cytochrome c oxidase from yeast or bovine heart. Manganese is apparently not an essential component of P. denitrificans cytochrome c oxidase since it is present in substoichometric amounts relative to copper or heme a and since the manganese-free enzyme retains essentially full activity in oxidizing ferrocytochrome c. However, the manganese is not removed by EDTA and its EPR spectrum responds to the oxidation state of the oxidase. In contrast, manganese added to the yeast oxidase or to the manganese-free P. denitrificans enzyme can be removed by EDTA and does not respond to the oxidation state of the enzyme. This suggests that the manganese normally associated with P. denitrificans cytochrome c oxidase is incorporated into one or more internal sites during the biogenesis of the enzyme.     

Intraliposomal nucleotides change the kinetics of reconstituted cytochrome c oxidase from bovine heart but not from Paracoccus denitrificans

Isolated cytochrome c oxidases of P. denitrificans and bovine heart were reconstituted in liposomes and the kinetics of cytochrome c oxidation were measured in the presence and absence of nucleotides either inside or outside of proteoliposomes, and after photolabelling with 8-azido-ATP. Intraliposomal ATP increases and ADP decreases the kinetics of ferrocytochrome c oxidation of the bovine but not of the Paracoccus enzyme. Extra-liposomal ATP and ADP increase the Km for cytochrome c of both enzymes, but ATP acts at lower concentrations than ADP. The increase of the Km for cytochrome c is obtained in coupled as well as in uncoupled proteoliposomes. Photolabelling with 8-azido-ATP of the reconstituted Paracoccus enzyme also increases the Km for cytochrome c which is completely prevented if ATP but not if ADP is present during illumination as was found with reconstituted cytochrome c oxidase from bovine heart. The data suggest a specific interaction of ATP and ADP with nuclear-coded subunits of bovine heart cytochrome c oxidase from the matrix side, because the effects are not found with the Paracoccus enzyme, which lacks these subunits.