Most proteins spend the majority of their time in their folded native state. Adopting this conformation, however, involves passing through various partially folded forms, including transition states and potentially kinetic intermediates. Furthermore, even under conditions favoring the folded conformation, a protein will take excursions away from the native state, populating partially and fully unfolded conformations. All of these states together constitute the protein energy landscape, and exploration of this complete landscape is crucial for a complete understanding of protein folding behavior, as well as potentially function. This work discusses a variety of methods applied to various model proteins, elucidating novel details about the folding landscapes.
Kinetic investigations were conducted with both T4 lysozyme and E. coli ribonuclease H (RNase H). The lysozyme study revealed a "hidden" unfolding intermediate in addition to the previously characterized folding intermediate, resolving a long-standing discrepancy between kinetic and native-state hydrogen exchange experiments. The RNase H study, on the other hand, focused on the nature of the transition state. This protein is known to fold through an intermediate, which has been investigated by various methods, including kinetic hydrogen exchange and kinetic analysis; however, the post-intermediate transition state had not previously been thoroughly investigated. The results from the study in this thesis suggest that the protein traverses the rate-limiting transition state through a highly localized nucleation-condensation process involving part of the protein that is unfolded in the kinetic intermediate.
E. coli RNase H was further investigated using a novel technique called native-state thiol alkyl-proton exchange (NSSX), a method analogous to native-state hydrogen exchange that takes advantage of the unique reactivity of cysteine to monitor exchange at the side chain, rather than the amide position of the backbone that is the target of hydrogen exchange experiments. Initial studies indicated that the wild-type protein was not amenable for these studies, but introduction of a stabilizing mutation allowed for investigation of the folding landscape under native conditions, with probes exposed on the folded and unfolded sides of the rate-limiting barrier exchanging in different kinetic regimes. This kinetic partitioning allowed for identification and characterization of novel partially folded species on the native side of the barrier and revealed structural and kinetic data for probes that are only exposed on unfolded side of the barrier. Interestingly, some of the probes involved in the rate-limiting nucleation step, as identified in the kinetic analysis, are also shown to be structured in the transition state by the NSSX experiments.
These in vitro studies are complemented by in vivo translational misincorporation experiments with two pairs of homologous proteins. A library of mutant tRNAs was developed for NSSX to introduce cysteines in the place of a given amino acid during translation; however, it was found that E. coli RNase H was refractory to the misincorporation method. A highly similar protein from a thermophilic organism, Thermus thermophilus, on the other hand, shows robust misincorporation. Similarly, E. coli phosphoglycerate kinase (PGK) shows essentially no misincorporation, while yeast PGK misincorporates well. There is some in vitro evidence that those proteins that show significant misincorporate - T. thermophilus RNase H and yeast PGK - adopt partially folded conformations that are not accessible to their homologs. Therefore, it is plausible that misincorporation efficiency may report on the existence of partially folded forms in vivo; specifically, the absence of such conformations may result in degradation of the nascent chain on the ribosome, while adopting a protected conformation may allow for translation of the full-length misincorporated protein. While these results are preliminary and the hypothesis must be verified by further experiments, they provide an intriguing suggestion for a new in vivo probe of partially folded structure.
