Phospholipase digestion of bound cardiolipin reversibly inactivates bovine cytochrome bc1.

Phospholipids and tightly bound cardiolipin (CL) can be removed from Tween 20 solubilized bovine cytochrome bc(1) (EC 1.10.2.2) by digestion with Crotalus atrox phospholipase A(2). The resulting CL-free enzyme exhibits all the spectral properties of native cytochrome bc(1), but is completely inactive. Full electron transfer activity is restored by exogenous cardiolipin added in the presence of dioleoylphosphatidylcholine (DOPC) and dioleoylphosphatidylethanolamine (DOPE), but not by cardiolipin alone or by mixtures of phospholipids lacking cardiolipin. Acidic, nonmitochondrial phospholipids, e.g., monolysocardiolipin or phosphatidylglycerol, partially reactivate CL-free cytochrome bc(1) if they are added together with DOPC and DOPE. Phospholipid removal from the Tween 20 solubilized enzyme, including the tightly bound cardiolipin, does not perturb the environment of either cytochrome b(562) or b(566), nor does it cause the autoreduction of cytochrome c(1). Cardiolipin-free cytochrome bc(1) also binds antimycin and myxothiazol normally with the expected red shifts in b(562) and b(566), respectively. However, the CL-free enzyme is much less stable than the lipid-rich preparation, i.e., (1) many chromatographic methods perturb both cytochrome b(566)() (manifested by a hypsochromic effect, i.e., blue shift of 1.5-1.7 nm) and cytochrome c(1) (evidenced by autoreduction in the absence of reducing agents); (2) affinity chromatographic purification of the enzyme causes pronounced loss of subunits VII and XI (65-80% decrease) and less significant loss of subunits I, IV, V, and X (20-30% decrease); and (3) high detergent-to-protein ratios result in disassembly of the complex. We conclude that the major role of the phospholipids surrounding cytochrome bc(1), especially cardiolipin, is to stabilize the quaternary structure. In addition, bound cardiolipin has an additional functional role in that it is essential for enzyme activity.     

Contribution of peroxidized cardiolipin to inactivation of bovine heart cytochrome c oxidase

The lipid-soluble peroxides, tert-butyl hydroperoxide and peroxidized cardiolipin, each react with bovine cytochrome c oxidase and cause a loss of electron-transport activity. Coinciding with loss of activity is oxidation of Trp19 and Trp48 within subunits VIIc and IV, and partial dissociation of subunits VIa and VIIa. tert-Butyl hydroperoxide initiates these structural and functional changes of cytochrome c oxidase by three mechanisms: (1) radical generation at the binuclear center; (2) direct oxidation of Trp19 and Trp48; and (3) peroxidation of bound cardiolipin. All three mechanisms contribute to inactivation since blocking a single mechanism only partially prevents oxidative damage. The first mechanism is similar to that described for hydrogen peroxide [Biochemistry43:1003-1009; 2004], while the second and third mechanism are unique to organic hydroperoxides. Peroxidized cardiolipin inactivates cytochrome c oxidase in the absence of tert-butyl hydroperoxide and oxidizes the same tryptophans within the nuclear-encoded subunits. Peroxidized cardiolipin also inactivates cardiolipin-free cytochrome c oxidase rather than restoring full activity. Cardiolipin-free cytochrome c oxidase, although it does not contain cardiolipin, is still inactivated by tert-butyl hydroperoxide, indicating that the other oxidation products contribute to the inactivation of cytochrome c oxidase. We conclude that both peroxidized cardiolipin and tert-butyl hydroperoxide react with and triggers a cascade of structural alterations within cytochrome c oxidase. The summation of these events leads to cytochrome c oxidase inactivation.     

Photolabeling of cardiolipin binding subunits within bovine heart cytochrome c oxidase

Subunits located near the cardiolipin binding sites of bovine heart cytochrome c oxidase (CcO) were identified by photolabeling with arylazido-cardiolipin analogues and detecting labeled subunits by reversed-phase HPLC and HPLC-electrospray ionization mass spectrometry. Two arylazido-containing cardiolipin analogues were synthesized: (1) 2-SAND-gly-CL with a nitrophenylazido group attached to the polar headgroup of cardiolipin (CL) via a linker containing a cleavable disulfide; (2) 2',2'-bis-(AzC12)-CL with two of the four fatty acid tails of cardiolipin replaced by 12-(N-4-azido-2-nitrophenyl) aminododecanoic acid. Both arylazido-CL derivatives were used to map the cardiolipin binding sites within two types of detergent-solubilized CcO: (1) intact 13-subunit CL-containing CcO (three to four molecules of endogenous CL remain bound per CcO monomer); (2) 11-subunit CL-free CcO (subunits VIa and VIb are missing because they dissociate during CL removal). Upon the basis of these photolabeling studies, we conclude that (1) subunits VIIa, VIIc, and possibly VIII are located near the two high-affinity cardiolipin binding sites, which are present in either form of CcO, and (2) subunit VIa is located adjacent to the lower affinity cardiolipin binding site, which is only present in the 13-subunit form of CcO. These data are consistent with the recent CcO crystal structure in which one cardiolipin is located near subunit VIIa and a second is located near subunit VIa (PDB ID code referenced in Tomitake, T. et al. (2003) Proc. Natl. Acad. Sci. U.S.A. 100, 15304-15309). However, we propose that a third cardiolipin is bound between subunits VIIa and VIIc near the entrance to the D-channel. Cardiolipin bound at this location could potentially function as a proton antenna to facilitate proton entry into the D-channel. If true, it would explain the CcO requirement of bound cardiolipin for full electron transport activity.     

Calcium-bound structure of bovine cytochrome c oxidase

The crystal structure of bovine cytochrome c oxidase (CcO) shows a sodium ion (Na⁺) bound to the surface of subunit I. Changes in the absorption spectrum of heme a caused by calcium ions (Ca²⁺) are detected as small red shifts, and inhibition of enzymatic activity under low turnover conditions is observed by addition of Ca²⁺ in a competitive manner with Na⁺. In this study, we determined the crystal structure of Ca²⁺-bound bovine CcO in the oxidized and reduced states at 1.7 Å resolution. Although Ca²⁺ and Na⁺ bound to the same site of oxidized and reduced CcO, they led to different coordination geometries. Replacement of Na⁺ with Ca²⁺ caused a small structural change in the loop segments near the heme a propionate and formyl groups, resulting in spectral changes in heme a. Redox-coupled structural changes observed in the Ca²⁺-bound form were the same as those previously observed in the Na⁺-bound form, suggesting that binding of Ca²⁺ does not severely affect enzymatic function, which depends on these structural changes. The relation between the Ca²⁺ binding and the inhibitory effect during slow turnover, as well as the possible role of bound Ca²⁺ are discussed.     

Activation of beef-heart cytochrome c oxidase by cardiolipin and analogues of cardiolipin

Beef-heart cytochrome c oxidase lacking endogenous lipids can be prepared by cholate-mediated exchange with dimyristoylphosphatidylcholine (Powell, G. L., Knowles, P. F. and Marsh, D. (1985) Biochim. Biophys. Acta 816, 191-194). These preparations retained practically no endogenous cardiolipin (less than 0.19 mol cardiolipin per mol of oxidase) but in Tween 80 they retained unaltered electron transport activity. Resupplementation of the dimyristoylphosphatidylcholine-substituted cytochrome oxidase with cardiolipin and cardiolipin analogues with different numbers of acyl chains or with a methylated headgroup enhanced the activity of the reconstituted enzyme to an extent dependent on the structure of the cardiolipin derivative. The Eadie-Hofstee plot showed biphasic kinetic behavior for all reconstituted preparations, even those completely lacking cardiolipin. This biphasic substrate dependence of the kinetics was simulated using the model of Brzezinski, P. and Malmstr?m, B. G. (Proc. Natl. Acad. Sci. USA 83 (1986) 4282-4286), which implicates two interconverting enzyme conformations in the proton transport step. The activation of cytochrome c oxidase by the cardiolipin analogues could be explained in terms of an electrostatic enhancement of the surface concentrations of both cytochrome c and protons, and a facilitated interconversion between the two enzyme conformations.     

Cardiolipin-depleted bovine heart cytochrome c oxidase: binding stoichiometry and affinity for cardiolipin derivatives

Detergent-solubilized bovine heart cytochrome c oxidase requires 2 mol of tightly bound cardiolipin (CL) per mole of monomeric complex for functional activity. Four lines of evidence support this conclusion: (1) Phospholipid depletion shows that two tightly bound CL's must remain associated with cytochrome c oxidase in order to maintain full electron transport activity. (2) Removal of the two tightly bound CL's correlates with decreased activity that is restored by reassociation of 2 mol of exogenous CL. (3) CL-depleted cytochrome c oxidase has two high-affinity binding sites for 2-[14C]acetylcardiolipin (AcCL), Kd,app less than 0.1 microM, that are not present in enzyme containing endogenous CL. An additional 2-3 lower affinity AcCL binding sites, Kd,app = 4 microM, are present in the CL-depleted complex, but these sites are also present in enzyme containing endogenous CL. (4) CL, monolysocardiolipin (MLCL), and dilysocardiolipin (DLCL) compete for AcCL binding with approximately the same relative affinities as those measured by the restoration of electron transport activity (MLCL competes much better than DLCL). However, MLCL and DLCL are only 60% and 15% as effective as CL in restoring maximum activity when they are bound to the high-affinity sites. The binding specificity of CL, MLCL, DLCL, and some of their acylated derivatives indicates that the apolar tails are most important for binding, not the polar head group. The presence or absence of hydroxyl groups in CL, MLCL, or DLCL also has little effect upon binding affinities. Binding specificity clearly favors CL since phosphatidylglycerol, phosphatidic acid, and phosphatidylcholine each have very low affinity for the CL binding sites (Kd,app greater than 20 microM). We, therefore, conclude that restoration of activity to CL-depleted cytochrome c oxidase is highly specific and requires the reassociation of CL, or structurally similar compounds, with two high-affinity binding sites.     

Surface plasmon resonance studies of complex formation between cytochrome c and bovine cytochrome c oxidase incorporated into a supported planar lipid bilayer. I. Binding of cytochrome c to cardiolipin/phosphatidylcholine membranes in the absence of oxidase.

The mechanism of interaction between cytochrome c and a solid-supported planar phosphatidylcholine membrane containing varying amounts of cardiolipin (0-20 mol%) has been studied over a wide range of protein concentrations (0-450 microM) and ionic strength conditions (10-150 mM), by direct measurement of protein binding using surface plasmon resonance (SPR) spectroscopy. The results demonstrate that cytochrome c binds to such phospholipid membranes in two distinct phases characterized by very different (approximately one order of magnitude) affinity constants. The second phase is dependent upon the prior occurrence of the first binding process. Although the binding affinities for both modes of binding are highly sensitive to both the cardiolipin concentration and the ionic strength of the buffer solution, indicating that electrostatic forces are involved in these processes, binding cannot be reversed by salt addition or by dilution. Furthermore, the final saturation levels of adsorbed protein are independent of ionic strength and cardiolipin concentration. These observations suggest that binding involves more than a simple electrostatic interaction. Invariance in the shapes of the SPR spectra indicates that no major structural transitions occur in the proteolipid membrane due to cytochrome c binding, i.e., the bilayer character of the lipid phase appears to be preserved during these interactions. Based on these results, a model of the lipid membrane-cytochrome c interaction is proposed that involves varying degrees of protein unfolding and subsequent binding to the membrane interior via hydrophobic forces.     

Nitric oxide inhibits peroxidase activity of cytochrome c.cardiolipin complex and blocks cardiolipin oxidation

The increased production of NO during the early stages of apoptosis indicates its potential involvement in the regulation of programmed cell death through yet to be identified mechanisms. Recently, an important role for catalytically competent peroxidase form of pentacoordinate cytochrome c (cyt c) in a complex with a mitochondria-specific phospholipid, cardiolipin (CL), has been demonstrated during execution of the apoptotic program. Because the cyt c.CL complex acts as CL oxygenase and selectively oxidizes CL in apoptotic cells in a reaction dependent on the generation of protein-derived (tyrosyl) radicals, we hypothesized that binding and nitrosylation of cyt c regulates CL oxidation. Here we demonstrate by low temperature electron paramagnetic resonance spectroscopy that CL facilitated interactions of ferro- and ferri-states of cyt c with NO and NO(-), respectively, to yield a mixture of penta- and hexa-coordinate nitrosylated cyt c. In the nitrosylated cyt c.CL complex, NO chemically reacted with H(2)O(2)-activated peroxidase intermediates resulting in their reduction. A dose-dependent quenching of H(2)O(2)-induced protein-derived radicals by NO donors was shown using direct electron paramagnetic resonance measurements as well as immuno-spin trapping with antibodies against protein 5,5-dimethyl-1-pyrroline N-oxide-nitrone adducts. In the presence of NO donors, H(2)O(2)-induced oligomeric forms of cyt c positively stained for 3-nitrotyrosine confirming the reactivity of NO toward tyrosyl radicals of cyt c. Interaction of NO with the cyt c.CL complex inhibited its peroxidase activity with three different substrates: CL, etoposide, and 3,3'-diaminobenzidine. Given the importance of CL oxidation in apoptosis, mass spectrometry analysis was utilized to assess the effects of NO on oxidation of 1,1'2,2'-tertalinoleoyl cardiolipin. NO effectively inhibited 1,1'2,2'-tertalinoleoyl cardiolipin oxidation catalyzed by the peroxidase activity of cyt c. Thus, NO can act as a regulator of peroxidase activity of cyt c.CL complexes.     

ATP binding to bovine heart cytochrome c oxidase. A photoaffinity labelling study.

ATP influences the kinetic properties of cytochrome c oxidase. A photoactivatable radioactive ATP analogue was used to localize the nucleotide-binding site on the bovine heart enzyme. Subunits IV and VIII were specifically labelled, suggesting that these two nuclear-coded polypeptides may play a regulatory role on the oxidase functions.     

Functional equivalence of monomeric (shark) and dimeric (bovine) cytochrome c oxidase

Cytochrome c oxidase isolated from hammerhead shark red muscle is monomeric in relation to the dimeric form of isolated bovine cytochrome c oxidase but in other ways bears a close resemblance to the enzyme isolated from mammalian tissue [1, 2]. Comparative studies of shark and bovine cytochrome c oxidase were extended to address the degree of functional similarity between the monomeric (shark) and dimeric (bovine) enzymes in the kinetics of peroxide binding and in the extent to which the catalytic action of the enzymes in vesicles can establish a proton gradient. Although the kinetics of peroxide binding and the proton pumping processes are complex, the dimeric and monomeric forms are quite similar with respect to these functional attributes. The kinetic heterogeneity of the process of peroxide binding is expressed in the shark enzyme as well as in the bovine enzyme, and both types of enzymes in vesicles can generate transmembrane proton gradients. On this basis we conclude that the dimeric state of isolated cytochrome c oxidase from mammalian sources is not essential for its function in vitro.     

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.     

X-ray structure and the reaction mechanism of bovine heart cytochrome c oxidase

X-ray structure of bovine heart cytochrome c oxidase in the fully oxidized state shows a peroxide bridging between Fe2+ and Cu2+ in the O2 reduction site. The bond distances for Fe-O and Cu-O are 2.52 and 2.16 A, respectively. The structure is consistent with antiferromagnetic coupling between the two metals, which has long been known and to recent redox titration results [J. Biol. Chem. 274 (1999) 33403]. The trigonal planer coordination of Cu1+ in the O2 reduction site is consistent with the very weak interaction between Cu1+ and O2 bound at Fe2+ revealed by time-resolved resonance Raman investigations. One of the three histidine imidazoles coordinated to the Cu ion in the O2 reduction site fixes a tyrosine phenol group near the O2 reduction site with the direct covalent link between the two groups. The structure suggests that the phenol group is the site for donating protons to the bound O2. Redox-coupled conformational change in an extramembrane loop indicates that an aspartate (Asp51) in the loop apart from the O2 reduction site is the site for proton pumping.     

Surface plasmon resonance studies of complex formation between cytochrome c and bovine cytochrome c oxidase incorporated into a supported planar lipid bilayer. II. Binding of cytochrome c to oxidase-containing cardiolipin/phosphatidylcholine membranes.

Complex formation between horse heart cytochrome c (cyt c) and bovine cytochrome c oxidase (cco) incorporated into a supported planar egg phosphatidylcholine membrane containing varying amounts of cardiolipin (CL) (0-20 mol%) has been studied under low (10 mM) and medium (160 mM) ionic strength conditions by surface plasmon resonance (SPR) spectroscopy. Both specific and nonspecific modes of cyt c binding are observed. The dissociation constant of the specific interaction between cyt c and cco increases from approximately 6.5 microM at low ionic strength to 18 microM at medium ionic strength, whereas the final saturation level of bound protein is independent of salt concentration and corresponds to approximately 53% of the total cco molecules present in the membrane. This suggests a 1:1 binding stoichiometry between the two proteins. The nonspecific binding component is governed by electrostatic interactions between cyt c and the membrane lipids and results in a partially ionic strength-reversible protein-membrane association. Thus, hydrophobic interactions between cyt c and the membrane, which are the predominant mode of binding in the absence of cco, are greatly suppressed. Both the amount of nonspecifically bound protein and the binding affinity can be varied over a broad range by changing the ionic strength and the extent of CL incorporation into the membrane. Under conditions approximating the physiological state in the mitochondrion (i.e., 20 mol% CL and medium ionic strength), 1-1.5 cyt c molecules are bound to the lipid phase per molecule of cco, with a dissociation constant of 0.1 microM. The possible physiological significance of these observations is discussed.     

Ionic strength dependence of the kinetics of electron transfer from bovine mitochondrial cytochrome c to bovine cytochrome c oxidase.

The effect of ionic strength on the one-electron reduction of oxidized bovine cytochrome c oxidase by reduced bovine cytochrome c has been studied by using flavin semiquinone reductants generated in situ by laser flash photolysis. In the absence of cytochrome c, direct reduction of the heme a prosthetic group of the oxidase by the one-electron reductant 5-deazariboflavin semiquinone occurred slowly, despite a driving force of approximately +1 V. This is consistent with a sterically inaccessible heme a center. This reduction process was independent of ionic strength from 10 to 100 mM. Addition of cytochrome c resulted in a marked increase in the amount of reduced oxidase generated per laser flash. Reduction of the oxidase at the heme a site was monophasic, whereas oxidation of cytochrome c was multiphasic, the fastest phase corresponding in rate constant to the reduction of the heme a. During the fast kinetic phase, 2 equiv of cytochrome c was oxidized per heme a reduced. We presume that the second equivalent was used to reduce the Cua center, although this was not directly measured. The first-order rate-limiting process which controls electron transfer to the heme a showed a marked ionic strength effect, with a maximum rate constant occurring at mu = 110 mM (1470 s-1), whereas the rate constant obtained at mu = 10 mM was 630 s-1 and at mu = 510 mM was 45 s-1. There was no effect of "pulsing" the enzyme on this rate-limiting one-electron transfer process. These results suggest that there are structural differences in the complex(es) formed between mitochondrial cytochrome c and cytochrome c oxidase at very low and more physiologically relevant ionic strengths, which lead to differences in electron-transfer rate constants.     

Lipid-substituted cytochrome oxidase: no absolute requirement of cardiolipin for activity.

The endogenous lipid of yeast cytochrome oxidase has been replaced by dimyristoyl phosphatidylcholine. Thin layer chromatography of the total lipid extract from the substituted enzyme revealed phosphatidylcholine only and no cardiolipin. Gas-liquid chromatography showed that >99% of the lipid chains derived from the substituted lipid, and that cardiolipin must be <0.03 mole/mole enzyme. The activity of the lipid-substituted enzyme was 10% of the original activity and increased to 47% by addition of dimyristoyl phosphatidylcholine. Thus there is no absolute requirement of cardiolipin for oxidative activity.     

pH-induced alterations in the structure and activity of cytochrome c oxidase

The effects of pH on the activity and structure of beef heart cytochrome c oxidase have been studied in the pH range 5.0-7.6. (i) A group with pK of approximately 5.45 has been readily detected in the pH vs. activity curve. This group must be deprotonated to achieve maximal activity. (ii) A group with a similar pK (5.45) has been detected and contributes to the spectral character of the reduced oxidase. Over the range pH 5.0-7.6 no other acid-sensitive group contributes to the spectrum of the reduced oxidase. (iii) The oxidized oxidase shows at least three acid-sensitive groups contributing to the spectrum. One occurs in the pH 7 range and another in the pH 5.6 range; below pH 5.2 additional pH-sensitive groups are apparent. Accurate estimation of the pK's of the groups responsible for the spectral changes in the oxidized oxidase has not been possible. (iv) The spectrum of the "oxygenated" (428 nm) conformer of the oxidized protein is invariant over the range ph 5.5-7. (v) The changes occurring in the spectrum of the purified oxidase also occur in the protein contained in phospholipid vesicles. (vi) The data are discussed in terms of the mechanism by which the oxidase, during its in situ catalytic cycle, may give rise to the primary events in energy coupling.     

Association of spin-labelled cardiolipin with dimyristoylphosphatidylcholine-substituted bovine heart cytochrome c oxidase. A generalized specificity increase rather than highly specific binding sites

The endogeneous lipid of bovine heart cytochrome c oxidase has been replaced by dimyristoylphosphatidylcholine using cholate-mediated exchange. The lipid-substituted preparation contained less than 1 mole cardiolipin per mole enzyme and possessed full oxidative activity. The association of spin-labelled cardiolipin with such lipid-substituted cytochrome oxidase preparations has been assayed using ESR spectroscopy. An average relative association constant 5.4-times that for phosphatidylcholine is obtained for cardiolipin. Measurements on preparations with increasing contents of unlabelled cardiolipin, introduced during lipid exchange, reveal that this selectivity corresponds to a generalized increase in specificity for all lipid association sites on the protein.     

Labeling of bovine heart cytochrome c oxidase with analogues of phospholipids. Synthesis and reactivity of a new cardiolipin benzaldehyde probe

The syntheses of two new radioactive probes derived from cardiolipin and phosphatidylcholine are reported. These probes are derivatives of natural lipids and contain an amine-specific benzaldehyde in the head-group region. This functional group allows a choice of timing of the reaction (e.g., after equilibration and detergent removal) because an irreversible covalent bond is formed only upon the addition of reducing agent. These probes, as well as a benzaldehyde analogue of phosphatidic acid, and a water-soluble benzaldehyde reagent were covalently attached to bovine heart cytochrome c oxidase. After reconstitution into vesicles, the lipid-benzaldehyde probes selectively incorporated into the smaller polypeptides of the enzyme, while the remaining subunits (I-IV) exhibited little incorporation of label. The accessibility of amine groups labeled under the conditions used here was independent of the structural and charge differences between the benzaldehyde probes. This suggests that all three lipid probes react with polypeptides of the cytochrome c oxidase complex at general contact sites for membrane phospholipids. A water-soluble benzaldehyde reagent predominantly labeled subunits IV, Va, and Vb and polypeptides of VII-VIII. A comparison of these results facilitates a more refined view of the disposition of polypeptides of cytochrome c oxidase in respect to the lipid and aqueous phases.