The unfolded state of a protein under physiological conditions has long been modeled as a random coil polymer with negligible self-interactions. Recent measurements by nuclear magnetic resonance (NMR) have revealed substantial residual structure and transient interactions in several proteins, contradicting the classic view of the unfolded state. However, the NMR data do not describe the global conformational dynamics of these highly disordered states with much specificity. This project seeks to marry these earlier results with complementary measurements of the rate of intramolecular diffusion on the same sequences. The PI uses the novel technique of quenching of the triplet state of tryptophan by cysteine, which is measured with an optical instrument with nanosecond resolution. Intramolecular diffusion coefficients can be extracted from these measured rates using a theory by Szabo, Schulten and Schulten which requires a probability distribution of equilibrium distances between the tryptophan and cysteine in the sequence. The PI and co-PI construct such distributions using a polymer model that exploits the structural information obtained in the prior NMR studies. These distributions are compared to existing distributions constructed by the Forman-Kay and Vendruscolo labs and also to molecular dynamics simulations that will be completed as part of this project.
The long-term goal is to develop sequence-specific distributions based on established polymer models, statistical structure predictors, hydrophobicity scales and measured diffusion coefficients that should lead to a general understanding of the dynamics of the unfolded state. This project encompasses aspects of structural biology, computation biology, optical spectroscopy and polymer physics. Therefore the students involved in this work will require cross-disciplinary training to accomplish the scientific goals. As part of a new graduate program in Quantitative Biology at MSU, the PI and co-PI are designing new lecture and laboratory courses to introduce physical science students to biology and biology students to quantitative methods. This project is jointly supported by the Molecular Biophysics Program in the Division of Molecular and Cellular Biosciences in the Directorate for Biological Sciences and the Division of Physics in the Mathematical and Physical Sciences Directorate.
This project investigated the dynamics of unfolded proteins to understand how reconfiguration of an unstructured protein chain affects folding and aggregation. The experimental technique, Trp-Cys contact quenching, is unique in its ability to give information about both the size of the chain and how fast it reconfigures. Early work showed that completely unstructured chains reconfigure very quickly, leading to predictions of the "speed limit" of protein folding of ~1 microsecond. However, work during this project has shown that real proteins before they fold can reconfigure more than ten times slower. Therefore the real speed limit for some proteins may be much slower because it is hard for the unfolded protein to reconfigure and find stable structures. Other work in this project has shown a correlation between the reconfiguration rate and the rate of aggregation between two different proteins and we have developed a model to explain this behavior. If a protein reconfigures much fast or much slower than the rate of two proteins colliding with each other, aggregation doesn't happen often. But those rates are similar, aggregation is likely. Finally, we have developed computational methods for predicting the size of an unfolded protein to help determine the reconfiguration rate more accurately. We expect these results to lead to more predictive models of protein folding and aggregation.
