SOLID STATE NMR STUDIES OF CONFORMATIONAL DYNAMICS

Ligand-gated conformational change of proteins is often important for rational drug design and mechanistic enzymology and has been demonstrated to occur in a large number of the enzymes for which we have structural data with and without ligand. How and why these changes occur is a totally open and fundamental question about protein function. We wish to study the classic case of an enzymatically critical loop motion, the 'flexible loop' in Triose Phosphate Isomerase, using broadline deuterium NMR of a unique tryptophan located in the center of the loop. When substrate analogs bind it closes over the active site, completing the protein contacts to the substrate and sealing the active site from solvent. In the open conformation there appears to be considerable disorder in the loop. The rate and energetics of this motion from the 'open' to 'closed' conformation will be measured as a function of the ligand -- for example whether the ligand is a substrate, a transition state or an intermediate analog. We propose to use site-directed mutagenesis in the loop, especially the hinges to identify features of the protein that pre-dispose it to mobility. We can prepare catalytically active samples for solid state NMR and have demonstrated the technical feasibility of the NMR measurement. Preliminary NMR results offer a different picture of the loop motion than has been assumed previously. The ligand does not trigger motion -- the extent of motion is comparable in the presence or the absence of substrate and is sufficiently fast so as to not limit the overall rate of the enzyme either during binding or for release. However the binding of a transition state analog compound appears to substantially slow the loop motion. Upon binding, the substrate reportedly experiences a conformational distortion that can direct the reactivity stereoelectronically. More definite evidence for this distortion will be obtained using recently developed dipolar coupling methods with doubly 13C- labeled substrate at low temperature. Preliminary results on a transition state analog compound and demonstrate the feasibility. The 13C SSNMR measurements also confirmed for the first time the ionization state of the TSA compound and the strong hydrogen bond between the TSA and the protein. Determination of the conformation of the bound substrate has not been possible by X-ray crystallography due a dynamic equilibrium mixture of several species. We also propose to determine the ionization states of the active site aspartic acids in HIV protease using the chemical shift anisotropy of 4- 13C aspartic acid. These groups are central for drug design and their ionization state is totally unknown. The motion of the protease flap will also be probed by deuterium broadline NMR of a tryptophan as we have done for TIM.