KINETIC EVIDENCE FOR FOLDING AND UNFOLDING INTERMEDIATES IN STAPHYLOCOCCAL NUCLEASE

The complex kinetic behavior, commonly observed in protein folding stu dies suggests that a heterogeneous population of molecules exists in s olution and that a number of discrete steps are involved in the conver sion of unfolded molecules to the fully native form. A central issue i n protein folding is whether any of these kinetic events represent con formational steps important for efficient folding rather than side rea ctions caused by slow steps such as proline isomerization or misfoldin g of the polypeptide chain. In order to address this question, we used stopped-flow fluorescence techniques to characterize the kinetic mech anism of folding and unfolding for a Pro(-) variant of SNase in which all six proline residues were replaced by glycines or alanines. Compar ed to the wild-type protein, which exhibits a series of proline-depend ent slow folding phases, the folding kinetics of Pro(-) SNase were muc h simpler, which made quantitative kinetic analysis possible. Despite the absence of prolines or other complicating factors, the folding kin etics still contain several phases and exhibit a complex denaturant de pendence. The GuHCl dependence of the major observable folding phase a nd a distinct lag in the appearance of the native state provide clear evidence for an early folding intermediate. The fluorescence of Trp140 in the alpha-helical domain is insensitive to the formation of this e arly intermediate, which is consistent with a partially folded state w ith a stable beta-domain and a largely disordered alpha-helical region . A second intermediate is required to model the kinetics of unfolding for the Pro(-) variant, which shows evidence for a denaturant-induced change in the rate-limiting unfolding step. With the inclusion of the se two intermediates, we are able to completely model the major phase( s) in both folding and unfolding across a wide range of denaturant con centrations using a sequential four-state folding mechanism. In order to model the minor slow phase observed for the Pro(-) mutant, a six-st ate scheme containing a parallel pathway originating from a distinct u nfolded state was required. The properties of this alternate unfolded conformation are consistent with those expected due to the presence of a non-prolyl cis peptide bond. To test the kinetic model, we used sim ulations based on the six-state scheme and were able to completely rep roduce the folding kinetics for Pro(-) SNase across a range of denatur ant concentrations.