Dynamics of Ubiquitin by Solid State Nuclear Magnetic Resonance

Recent advances in high field solid state NMR have led to great progress in the field of protein biophysics. Although not simple, chemical shift assignments and determination of structural restraints of small proteins have become a reasonable endeavor. However, in order to exploit the technique for wide reaching biophysical studies, it is necessary to develop techniques for studying the dynamics of these molecules. This thesis will address some of the complications that arise when performing spin relaxation experiments on these molecules in their crystalline form. Furthermore, in lieu of spin relaxation experiments, an alternative method for measuring dynamics of proteins in the solid state will be introduced. Currently, there are many techniques available to solution state NMR spectroscopists for studying the motion of large biomolecules. The most commonly used techniques belong to a class of experiments that is called spin relaxation experiments. However, when these same techniques are applied in the solid state, many complications arise that are not normally encountered in the solution state. The largest barrier to applying spin relaxation experiments to solid samples is spin diffusion. In short, spin diffusion works towards equilibrating the magnetization across the whole molecule. This is detrimental as it will mask the true internal dynamics of the molecule. It was determined that the rapid rotation of an spa functional group, such as a methyl group and to less of an extent an amino group, can act as an efficient relaxation sink for slower relaxing backbone sites. We shall show how spin diffusion among 13C spins in ubiquitin effectively masks the internal dynamics of this protein. In order to accurately extract dynamical information from spin relaxation experiments, it is necessary to separate the contribution from these spin diffusion effects. We address this issue through quantitative examination of 13C spin diffusion in a model compound: glycyl-alanyl-leucine. Because 13C relaxation data are complicated by spin diffusion effects, it was expected that 15N relaxation experiments would yield results that are a more direct reflection of internal dynamics. However, it was determined that there are other processes for which we have yet to account that are affecting these measurements. It has been surmised that these complications arise from interactions between the protein and the water molecules associated with the crystal structure of ubiquitin. In this light, an alternative method for measuring dynamics has been developed that is similar to the residual dipolar couplings (RDCs) experiment performed in the solution state. The technique involves measuring order parameters via the dipolar couplings. In effect one can measure dynamics by comparing measured dipolar couplings with a theoretically static dipolar coupling. This technique has been applied to microcrystalline ubiquitin and the results are in good agreement with RDC studies of ubiquitin performed in the solution.