The principal objective of this project is a quantitative description of the physical chemistry that connects the amino acid sequence of staphylococcal nuclease to its three dimensional structure. The two approaches taken are NMR characterization of the structure and dynamics of partially folded conformations and computer simulation to estimate their thermodynamic properties. Previous studies of a fragment model of the denatured state have revealed, quite surprisingly, that it exhibits the same topology or low resolution structure as folded nuclease. To obtain a more detailed picture of the interactions that maintain this highly dynamic structure, advantage will be taken of the 15 to 50 fold increase in sensitivity for detecting HN-HN NOES that results from replacing all carbon-bound hydrogens with deuterium. In addition, residual dipolar couplings will be measured on partially oriented samples. An initial equilibrium folding pathway of nuclease will be extended to higher resolution using NMR experiments based on the TROSY-HSQC to follow the self organization of the peptide chain as a function of glycerol concentration. The specific chain-chain interactions responsible for this organization will be identified either directly through NOES or other structural parameters, or indirectly through correlated changes in NMR parameters sensitive to structure/dynamics produced by modifications in sequence. Recent studies of hydrogen exchange in four nuclease mutants have identified a role for the molten globule state in m-value effects --changes in sensitivity to denaturants. Analysis of additional m+ and m- mutants by hydrogen exchange, NMR parameters, and fluorescence will establish a quantitative relationship between m-values and changes in population/structure of this molten folding intermediate. To test this the hypothesis that the topology of the native state is determined in part by a high entropy of packing of secondary structural segments, Monte Carlo sampling methods are being used to estimate the density of low energy conformations near the true native structure and near grossly misfolded structures. For several small helical proteins, two independent simulation strategies demonstrate a higher density of conformations with the wild-topology. Future work will refine the computer model, address beta-strand containing proteins, and develop and test a strategy for predicting the low resolution structure of proteins from sequence plus secondary structure, through de novo construction of folds with maximal segment-packing entropy.
Showing the most recent 10 out of 29 publications
