The goal of this proposal is a molecular level description of protein unfolding, and by extension folding, using realistic molecular dynamics (MD) simulations in solution. Both the general and sequence-specific rules of unfolding will be pursued. The general rules will be investigated by making use of a large database of protein unfolding trajectories that already exist in the lab. In addition, new trajectories will be added. So far, this database contains over 2300 simulations of more than 300 proteins. This repository represents the largest collection of protein simulations and protein structures in the world. The simulations were designed so that representatives of all proteins folds will eventually be investigated, working from the most to least populated folds. The current set represents approximately 80% of all known protein structures. We have already developed a novel relational/multidimensional database to house these data.
Specific Aim 1 of this proposal seeks to determine the general rules of protein unfolding by mining this database. In addition, multiple representatives of highly populated folds are being investigated to determine sequence-specific effects. Our hypothesis is that all-atom molecular dynamics simulations of isolated proteins in solution can provide continuous and realistic protein unfolding pathways and that the general rules for unfolding and folding can be determined once a large number of protein folds have been simulated. While most relatives within a fold family fold by the same mechanism based on experimental studies, there are some exceptions. Consequently, sequence-specific effects will be determined by investigating multiple members of three common fold families with different architectures. In addition, the assumption that the thermal unfolding of an isolated protein in water is generally valid for the many conditions in which a protein may find itself is being tested through `test tube'simulations in which multiple copies of a protein are simulated together in a large box of water to determine how intermolecular interactions perturb the processes of folding and unfolding. Finally, the effect of aqueous organic solvents on the process is being investigated. This proposal is essentially hypothesis-driven discovery science.
Protein folding remains one of the most important unsolved problems in molecular biology, and it represents an important missing link necessary for full utilization of the information becoming available from the mapping of genomic sequences. Characterization of the unfolding process is equally important, both from the perspective of fully understanding a fundamental biochemical phenomenon and for the light shed on the folding process. An understanding of protein folding/unfolding also has important implications for all biological processes, including protein degradation, protein translocation, aging, and many human diseases, including amyloid diseases.
Showing the most recent 10 out of 55 publications
