Dynamical and temperature-dependent effects of lipid-protein interactions. Application of deuterium nuclear magnetic resonance and electron paramagnetic resonance spectroscopy to the same reconstitutions of cytochrome c oxidase.

2H NMR and EPR spectra have been obtained as a function of temperature and protein concentration from the same samples of beef heart cytochrome c oxidase reconstituted into 1-(16,16,16-trideuteriopalmitoyl)-2-palmitoleoyl-sn-glycero-3-phosphocholine. At all temperatures, the EPR spectra show the characteristic "bound" and "free" components, while the 2H NMR spectra show only a narrow distribution of orientational order parameters. At temperatures near the phase transition of the pure lipid, the dependence of the 2H NMR average orientational order on protein concentration fits a two-stage model in which the phospholipid molecular exchange rapidly between two states tentatively identified as sites either on or off the protein surface. From this model, the 2H NMR spectra yield a value of 0.18 mg of phospholipid per mg of protein as necessary to cover the surface of cytochrome c oxidase, which is the same value as derived from the EPR spectra at -20 degrees C. Both the 2H NMR and EPR spectra vary markedly with temperature. At temperatures well above the phase transition of the pure lipid, the average orientational parameters derived from the 2H NMR spectra are independent of protein concentration and are the same as for the lipid alone. Qualitatively, the EPR spectra show large apparent decreases in the average orientational order with increasing temperature. Analysis of 2H NMR relaxation rates indicates an additional motion in the presence of protein with a correlation time of 10(-6)-10(-7) s. If this new motion is associated with exchange between the two states, a minimum value of 10(6)-10(7) s-1 for the exchange rate is obtained, assuming that the lipids on the protein surface are much more motionally restricted than the rest of the lipid. Such an exchange rate is compatible with the observed differences in 2H NMR and EPR spectra. These results are consistent with short-lived, energetically weak interactions between cytochrome c oxidase and the phospholipids used in this study.