Exploring G-Protein Coupled Receptor Activation with Multiscale Molecular Simulations

This thesis is devoted to the study of the activation mechanism of G-protein coupled receptors (GPCRs), one of the largest and most diverse protein families in mammalian genomes, by means of different computer-based simulation techniques. Using force field based classical molecular dynamics (MD), the time evolution of two prototypical GPCRs, rhodopsin and β adrenergic receptors, has been followed for microseconds under different external conditions. Within this approach, the non-native modifications induced by the different engineering techniques used to crystallize most of the known GPCR structures have been identified and some of the mechanisms through which these receptors have managed to optimize their function through evolution have been described and quantified. In detail, this has led to suggest a possible binding mode of agonists to β adrenergic receptors and to identify crucial micro-switches during receptor activation, as well as to describe an asymmetric pathway in the rhodopsin dimer that suggests oligomerization in GPCRs as a biological strategy to enhance activation efficiency. The increased knowledge of GPCR activation mechanism obtained from the classical molecular dynamics simulations hinted to a crucial role of chemical reactions inside the binding pocket of the receptors during the activation cycle. These events have been studied in this thesis using a hybrid quantum mechanics/molecular mechanics (QM/MM) protocol based on Density Functional Theory. Using this approach, it has been possible to characterize the optical properties of the different intermediates along the activation pathway of rhodopsin and to quantitatively estimate the barrier for the proton transfer reaction that induces active state formation. Finally, a novel activation mechanism in diffusible ligands class A GPCRs that indicates proton transfer from the bound ligand to the receptor as a crucial step to reach the active conformation is proposed.