Rotation of cytochrome oxidase in phospholipid vesicles. Investigations of interactions between cytochrome oxidases and between cytochrome oxidase and cytochrome bc1 complex.

Cytochrome oxidase was incorporated into lipid vesicles composed of phosphatidylethanolamine-phosphatidylcholine-cardiolipin. Large proteoliposomes of 1,000-15,000 A diameter were prepared by calcium-induced fusion of small vesicles. Rotational diffusion of cytochrome oxidase was measured by detecting the decay of the absorption anisotropy, r(t), after photolysis of the heme a3.CO complex by a vertically polarized laser flash. Because of the large size of the proteoliposomes, there was no contribution of vesicle-tumbling to r(t) over the experimental time range of 5 ms for samples in 60% sucrose. Analysis of r(t) curves was based on a "rotation-about-membrane normal" model. The measurements were used to investigate intermolecular interactions between cytochrome oxidases and between cytochrome oxidase and cytochrome bc1 complex co-reconstituted in the above lipid vesicles. In vesicles of a high lipid to protein ratio (congruent to 27), nearly all cytochrome oxidase molecules are rotating with an approximate rotational relaxation time, phi 1, on the order of 500 microseconds. In contrast, about 20% of cytochrome oxidase is immobile in vesicles with a relatively low lipid to protein ratio (congruent to 5), although phi 1 of the mobile population remains about 500 microseconds. In contrast, about 20% of cytochrome oxidase is immobile in vesicles with a relatively low lipid to protein ratio (congruent to 5), although phi 1 of the mobile population remains about 500 microseconds. The immobilized fraction is presumably due to nonspecific self-aggregation of cytochrome oxidase. The presence of cytochrome bc1 complex does not change r(t) curves significantly, either in the presence or absence of cytochrome c. Previously, we have observed the co-existence of mobile and immobile populations of cytochrome oxidase in bovine heart and rat heart mitochondria (Kawato, S., Sigel, E., Carafoli, E., and Cherry, R. J. (1980) J. Biol. Chem. 255, 5508-5510). The present results suggest that the immobile population of about one-half of cytochrome oxidase could be simply due to nonspecific protein aggregation resulting from the high concentration of enzymes in the inner mitochondrial membrane (lipid to protein ratio, less than or equal to 0.5). We also conclude that there is no specific interaction between cytochrome oxidase and cytochrome bc1 complex in the above large lipid vesicles. A lateral collision-controlled model for electron transfer from cytochrome bc1 complex to cytochrome oxidase through cytochrome c is discussed based on the above results.