Effect of Macromolecular Crowding on Diffusive Processes

Detta är en avhandling från Uppsala : Uppsala University

Sammanfattning: Macromolecular crowding are innate to cellular environment. Understanding their effect on cellular components and processes is essential. This is often neglected in dilute experimental setup both in vitro and in silico.In this thesis I have dealt with challenges in biomolecular simulations at two levels of modeling, Brownian Dynamics (BD) and Molecular Dynamics (MD).Conventional BD simulations become inefficient since most of the computational time is spent propagating the particles towards each other before any reaction takes place. Event-driven algorithms have proven to be several orders of magnitude faster than conventional BD algorithms. However, the presence of diffusion-limited reactions in biochemical networks lead to multiple rebindings in case of a reversible reaction which deteriorates the efficiency of these types of algorithms. In this thesis, I modeled a reversible reaction coupled with diffusion in order to incorporate multiple rebindings. I implemented a Green's Function Reaction Dynamics (GFRD) algorithm by using the analytical solution of the reversible reaction diffusion equation. I show that the algorithm performance is independent of the number of rebindings.Nevertheless, the gain in computational power still deteriorates when it comes to the simulation of crowded systems. However, given the effects of macromolecular crowding on diffusion coefficient and kinetic parameters are known, one can implicitly incorporate the effect of crowding into coarse-grain algorithms by choosing right parameters. Therefore, understanding the effect of crowding at atomistic resolution would be beneficial.I studied the effect of high concentration of macromolecules on diffusive properties at atomistic level with MD simulations. The findings emphasize the effect of chemical interactions at atomistic level on mobility of macromolecules.Simulating macromolecules in high concentration raised challenges for atomistic physical models. Current force fields lead to aggregation of proteins at high concentration. I probed scenarios based on weakening and strengthening protein-protein and protein-water interactions, respectively. Furthermore, I built a cytoplasmic model at atomistic level based on the data available on Escherichia coli cytoplasm. This model was simulated in time and space by MD simulation package, GROMACS. Through this model, it is possible to study structural and dynamical properties under cellular like environment at physiological concentration.