Computational Modeling of Protein Folding under Cell-Like Conditions

Proteins are essential to supporting life in every living being. The three-dimensional shape of proteins is fundamental to their ability to function properly. The functional conformation is reached through a process known as protein folding. Most of the knowledge about protein folding was obtained from studies in dilute conditions; thus, it is necessary to understand how they fold in the cellular environment. Among the different interactions that a protein experiences inside the cell, there are two interactions: hydrodynamic interactions (HI) and macromolecular crowding. In this work, I used coarse-grained computational models of two-state proteins to study these interactions separately. First, I studied the impact of HI on the folding reaction of two proteins: chymotrypsin inhibitor 2 (CI2) and α-spectrin Src-homology 3 domain (SH3). I found that the dynamic effect of HI is temperature dependent relative to that in the absence of HI. This result implies the existence of a “crossover behavior” close to the folding temperature. I discovered that the acceleration due to HI for CI2 is greater than that for SH3, suggesting that the magnitude of the acceleration of a folding reaction is related to the protein topology. Furthermore, I investigated the factors that modulate the shape of the urea-unfolded ensemble of apoazurin in the presence of dextran 20, a synthetic crowder. Experiments discovered that dextran 20 increases the size of the unfolded apoazurin conformations. I demonstrated that crowder models with spherical and rod-like shapes using only steric repulsive interactions cannot explain this experimental finding. I found that the existence of attractive interactions between the protein and rod-like crowder models and the arrangement of these crowders in the vicinity of the protein could resemble the experimental observations.