Applications Of Molecular Dynamics Simulations In Biological Macromolecules

In recent years, with the development of computational science, moleculardynamics simulation has been widely applicated in biology, chemistry, physics andmaterial. Molecular dynamics simulation can provide detail information of the systemmotions as a function of time, as well as the information of thermodynamic property.Both of them are hardly to be gained from the experiments. To date, the investigation inlife science has gone deep into molecular level. Molecular dynamics simulation iscompletely under the control of scientists, so that by altering specific contributions theirrole in determining a property at atomic level can be known. Thus, this method isplaying an increasingly important role in the study of protein folding/unfolding,molecular recognition, ion transport, and enzyme catalyzed reaction mechanism. In thisdissertation, we investigated several important biological macromolecules by means ofmolecular dynamics simulation method, mainly included the following parts:1. Exploring the Molecular Basis of dsRNA Recognition by Mss116p UsingMolecular Dynamics Simulations and Free-Energy CalculationsDEAD-box proteins are the largest family of helicase that are important in nearlyall aspects of RNA metabolism. However, it is unclear how these proteins recognize andbind RNA. Here, we present a detailed analysis of the related DEAD-box proteinMss116p-RNA interaction, using molecular dynamics simulations with MM-GBSAcalculations. The energetic analysis indicates that the two strands of double strandsRNA (dsRNA) are recognized asymmetrically by Mss116p. The strand1of dsRNAprovides the main binding affinity. Meanwhile, the nonpolar interaction provides themain driving force for the binding process. Although the contribution of polarinteraction is small, it is vital in stabilizing the protein RNA interaction. Comparedwith the wild type Mss116p, two studied mutants Q412A and D441A have obviouslyreduced binding free energies with dsRNA because of the decreasing of polarinteraction. Three important residues Lys409, Arg415and Arg438lose their bindingaffinity significantly in mutants. In conclusion, these results complement previousexperiments to advance comprehensive understanding of Mss116p-dsRNA interaction. The results also would provide support for the application of similar approaches to theunderstanding of other DEAD-box protein-RNA complexes.2. Exploring the mechanism how Marburg virus VP35recognizes and bindsdsRNA by molecular dynamics simulations and free energy calculationsFiloviruses often cause terrible infectious disease which has not been successfullydealt with pharmacologically. All filoviruses encode a unique protein termed VP35which can mask doubled-stranded RNA to deactivate interferon. The interface ofVP35-dsRNA would be a feasible target for structure-based antiviral agent design. Toexplore the essence of VP35-dsRNA interaction, molecular dynamics simulationcombined with MM-GBSA calculations were performed on Marburg virusVP35-dsRNA complex and several mutational complexes. The energetic analysisindicates that nonpolar interactions provide the main driving force for the bindingprocess. Although the intermolecular electrostatic interactions play important roles inVP35-dsRNA interaction, the whole polar interactions are unfavorable for bindingwhich result in a low binding affinity. Compared with wild type VP35, the studiedmutants F228A, R271A and K298A have obviously reduced binding free energies withdsRNA reflecting in the reduction of polar or nonpolar interactions. The results alsoindicate that the loss of binding affinity for one dsRNA strand would abolish the totalbinding affinity. Three important residues Arg271, Arg294and Lys298which makesthe largest contribution for binding in VP35lose their binding affinity significantly inmutants. The uncovering of VP35-dsRNA recognition mechanism will provide someinsights for development of antiviral drug.3. Molecular dynamic investigations of BioH protein substrate specificity forbiotin synthesisPimeloyl-ACP methyl ester, which is long known to be a biotin precursor, is thephysiological substrate of BioH. Azelayl methyl esters conjugated to ACP is also indeedaccepted by BioH with very low rate of hydrolysis. To date, the substrate specificity forBioH and the molecular origin for the experimentally observed rate changes ofhydrolysis by the chain elongation to C9species have remained elusive. To this end, wehave investigated chain elongation effects on the structures by using the fully atomisticmolecular dynamics simulations combined with binding free energy calculations. Theresults indicate that the substrate specificity is determined by BioH together with ACP.The added two methylenes would increase the structural flexibility by protein motionsat the interface of ACP and BioH, instead of making steric clashes with the side chains of the BioH hydrophobic cavity. On the other hand, the slower hydrolysis of azelaylsubstrate is suggested to be associated with the loose of contacts between BioH andACP, and with the lost electrostatic interactions of two ionic/hydrogen bondingnetworks at the interface of the two proteins. The present study provides importantinsights into the structure-function relationships of the complex of BioH withMe-pimeloyl-ACP, which could contribute to further understanding about themechanism of the biotin synthetic pathway, including the catalytic role of BioH.4. Mutation and low pH effect on the stability as well as unfolding kinetics oftransthyretin dimerTransthyretin (TTR) dissociation and aggregation appear to cause several amyloiddiseases. TTR dimer is an important intermediate that is hard to be observed from thebiological experiments. To date, the molecular origin and the structural motifs for TTRdimer dissociation, as well as the unfolding process have not been rationalized at atomicresolution. To this end, we have investigated the effect of low pH and mutation L55P onstability as well as the unfolding pathway of TTR dimer using constant pH moleculardynamics simulations. The result shows that acidic environment results in loose TTRdimer structure. Mutation L55P causes the disruption of strand D and makes theCE-loop very flexible. In acidic conditions, dimeric L55P mutant exhibits notableconformation changes and an evident trend to separate. Our work shows that themovements of strand C and the loops nearby are the beginning of unfolding process. Inaddition, hydrogen bond network at the interface of the two monomers plays a part instabilizing TTR dimer. The dynamic investigation on TTR dimer provides importantinsights into the structure-function relationships of TTR, and rationalizes the structuralorigin for the tendency of unfolding and changes of structure that occur uponintroduction of mutation and pH along the TTR dimer dissociation and unfoldingprocess.