Computational studies of proteins

A variety of biological systems have been studied using computational methods to aid understanding of their physiological function. These systems include: (i) Cdc42-an important signalling protein, (ii) glutamate receptors-the main excitatory receptors in the central nervous system and (iii) proteins involved in electron and hydrogen transfer-processes essential for life. The three-dimensional solution structure of Cdc42 bound to its effector, PBD46, has been solved from NMR-derived restraints. Structurally, the extensive binding surface between PBD46 and Cdc42 can account for both the high affinity of the complex and the inhibition of GTP hydrolysis by PBD46. Homology models of glutamate receptors have provided valuable insight into (i) ligand binding/selectivity and (ii) channel activity. The model of the ligand binding domain of the kainate receptor GFKAR, consistent with experimental data, suggests that both the residues in the binding site and at the interface between the 2 lobes are important for kainate binding. A single hydrogen bond may be responsible for explaining ligand selectivity for the NMDA receptors NMDA-R1 (binds glycine) and NMDA-R2C (binds glutamate). Models of the transmembrane domain suggest that unfavourable electrostatics is the likely reason why homomeric NMDA-R2C forms a non-functional cation channel; making the electrostatics more favourable could potentially transform this channel into a functional one. Calculated intrinsic electron transfer rates on models of trimethylamine dehydrogenase bound to electron acceptors suggest the possibility of different electron transfer pathways depending on the redox acceptor, these on the whole reflect experimental observations. Electron transfer calculations on models of the complex of human electron transferring flavoprotein and its redox partner medium chain acyl-CoA dehydrogenase, suggest that protein dynamics enhance electronic coupling in this electron transfer system, consistent with experimental data. QM/MM studies, consistent with experimental data have shown substantial hydrogen tunnelling for the hydrogen abstraction step by methylamine dehydrogenase.