A MULTISCALE COMPUTATIONAL APPROACH TO STUDY RNASE A CATALYSIS Written under the direction of
OF THE DISSERTATION A multiscale computational approach to study RNase A catalysis Enzyme catalysis is an extremely important and complex process that is fundamental to biology. Experiments provide a wealth of valuable information about the function of enzymes; however, this information requires the use of computational models to establish a meaningful interpretation that can be used to guide design. Multiscale computational models, which integrate a hierarchy of theoretical methods to address
... omplex biomolecular problems that span large spatial and temporal ranges, afford powerful tools to provide a detailed molecular level interpretation of a wide range of experimental data from which a consensus view of catalytic mechanism may emerge. In this dissertation, I detail my efforts to develop and apply multiscale methods to study the mechanisms of RNA backbone cleavage catalyzed by Ribonuclease A, an important archetype enzyme system, and the factors that regulate its activity. In the first phase of this research, I use molecular dynamics simulations to characterize the structure and dynamics of the active enzyme in solution at different stages along the reaction path. In this work, I demonstrate that the crystallographic structure represents an inactive, catalytically non-relevant state, and make predictions that a conformational change involving the flipping of the side chain of a conserved histidine residue (His12) is required to adopt a catalytically competent conformation. In the second phase of this research, I apply "constant pH molecular dynamics simulations" (CpHMD) to characterize the conditional probability of finding key active site residues in a protonation state that supports general acid-base catalysis. This allowed the prediction of pK a shifts for His12, His119 and Lys41, and, for the first time, activity-pH profiles ii for an enzyme system that can be compared directly with those measured in kinetic experiments. In the third phase of this research, I use combined quantum mechanical/molecular mechanical methods to study the catalytic chemical steps of transphosphorylation. Results of this work predict a free energy landscape for the reaction, from which the minimum free energy pathway that connects reactants and products allows a detailed molecular-level picture of mechanism. In the fourth phase of this research, I extend the CpHMD method to nucleic acid systems, to benchmark the method for the study of ribozymes that catalyze the same reaction as RNase A. iii Acknowledgments I would first like to thank my advisor, Dr. Darrin M. York, for his guidance and encouragement over the last five years.