The applicant proposes to continue the development of REDOR methods using solid state NMR of lyophilized powders to describe ligand binding structures of large proteins. The goal is not the total structure, but rather structural elucidation of the ligand or ligands and the binding site. This approach uses: 1) REDOR methods for internuclear distances as great as 15 A; 2) uniform and specific isotope labeling schemes for three and four different kinds of nuclei; and 3) NMR probes tuned simultaneously to as many as six radiofrequencies. The approach provides key distance restraints for molecular modeling and in the last grant period has been successfully applied to complexes including: 1) 15N and 19F labeled enolpyruvyl shikimate-3-phosphate (EPSP) synthase complexed to 31P and 13C labeled substrates and inhibitors; 2) uniformly 15N labeled lumazine synthase complexed to 19F labeled inhibitors; and 3) natural abundance tubulin (as microtubules) complexed to specifically 15N, 13C, and 19F labeled taxol. A predictive model of the binding site has been produced in each instance that has led to biochemical and biophysical insights. The applicant now proposes to apply this approach to problems of biomedical interest including the characterization of the liganded binding sites of 1) cyclooxygenase-2, 2) 3-deoxy-D-manno-octulosonate 8-phosphate (KDO 8-P) synthase, 3) UDP-GlcNac enolpyruvyltransferase, 4) riboflavin synthase and 5) human factor Xa. As a part of this effort he will develop analytical method for extending REDOR experiments to multispin systems, including clusters of strongly coupled observed spins. These methods will be tested using the predictive models for binding sites of EPSP synthase and lumazine synthase developed during the past funding period.
Showing the most recent 10 out of 18 publications
