A Multifrequency Electron Spin Resonance Study of T4 Lysozyme Dynamics Using the Slowly Relaxing Local Structure Model

Abstract

Electron spin resonance (ESR) spectra were obtained at 250 and 9 GHz for nitroxide-labeled mutants of the protein T4 lysozyme in aqueous solution over a range of temperatures from 2 to 37.5 °C. Two mutants labeled at sites 72 and 131 were studied and compared. The mutant sites are solvent exposed and free of tertiary interactions with other side chains, but the former is at the center of a 5 turn helix, whereas the latter site is on a small two and a half turn helix. The 250 GHz ESR spectra, because of their “fast time scale”, are rather insensitive to the slow overall tumbling motion of the protein. Thus, they are qualitatively different for the two mutants, implying that there are different local dynamics at the two sites. The 9 GHz spectra, which are significantly affected by the overall tumbling and are less sensitive to the internal dynamics, do not show such marked differences between the two sites. The 250 and 9 GHz spectra for each mutant and temperature were simultaneously fit to the slowly relaxing local structure (SRLS) model for slow-motional ESR, using newly developed software. The SRLS model explicitly accounts for the overall tumbling of the protein and the internal modes of motion, which include the motion of the nitroxide side chain (expected to be the same for both mutants) and backbone fluctuations. Very good simultaneous fits are obtained. Whereas two conformers (or spectral components) are typically detected at the lower temperatures, only a single component is observed at the higher temperatures. The significant differences in the high-frequency spectra for the two mutants are readily attributed mainly to a difference in their respective local ordering. That is, site 72 exhibits significantly greater local ordering than does site 131, which is expected from the greater rigidity of the larger helix on which the 72 site is located. The results of this multifrequency study are compared with a previous 9 GHz study. A description of the application of the SRLS model in such a multifrequency study is provided.