A new approach for studying the equilibrium folding pathway in proteins has been developed in this laboratory. The strategy is to construct with the computer a large ensemble of conformational states, then derive the most probable population distribution; i.e. the distribution of states that best accounts for a wide array of experimental observables. The most probable distribution is determined by evaluating the Boltzmann factors for each state using an empirical structural parametrization of the energetics. It is assumed that the computer generated ensemble mimics the actual protein ensemble in solution if it is able to predict the observed experimental behavior of the protein. This approach has been applied to several proteins for which native state hydrogen exchange protection factors as well as thermodynamic data are available. The general agreement between predicted and experimental values indicates that the computer generated distribution of states approximates well the ensemble of conformations existing in solution. Examination of the derived most probable distribution of partially folded states permits identification of the most prevalent structural features of equilibrium folding intermediates. the rank order in which different elements become structured as a function of the degree of folding, the probabilities that individual residues are found in a native-like conformation at different stages of the folding process, the apparent folding constants per residue, etc. Together these statistical descriptors provide a quantitative definition of the equilibrium folding pathway.
Showing the most recent 10 out of 86 publications