Molecular Simulations On The Orientation And Conformation Of Protein Adsorbed On Surfaces

It is of great importance to understand, predict and ultimately control the orientation and conformation of proteins at interfaces for protein-surface related applications. Computer molecular simulation is especially suitable for providing detailed information of processes of complicated molecules at the molecular or atomic level. In this dissertation, the orientation and conformation of adsorbed proteins at interfaces were studied by molecular simulations. The mechanism and possibility of controlling protein behavior at interfaces by applying external electric fields were explored. Also, the mechanism of surface resistance to protein adsorption was explained.1. In molecular simulations of protein adsorption, it is necessary to accurately predict the orientation of the adsorbed protein before the subsequent conformational change to be studied. Currently, few works could accurately and fast determine the orientation of the adsorbed protein on surfaces. Therefore, we develop a parallel tempering Monte Carlo (PTMC) algorithm in temperature space to obtain the global-energy-minimum orientation and the orientation distribution in a single simulation. We then apply the PTMC algorithm in the simulations of lysozyme adsorption on charged surfaces at different surface charge densities (SCD) and ionic strengths (IS). Simulation results show that when attractive electrostatic interactions dominate, lysozyme prefers to be adsorbed on negatively charged surfaces with"side-on"orientation. For this orientation, the active site of lysozyme faces sideways, the binding sites locate around residue Lys1, Lys33 and Lys116; while residue Lys13, Lys96 and Lys97 on the opposite side expose to solution. With the increasing contribution from van der Waals interactions,"back-on"orientation becomes the preferred one. On positively charged surfaces, lysozyme is not adsorbed when IS is small due to the strong repulsive electrostatic interactions. However, as IS increases, the screening effect from ions in solution becomes more pronounced, and lysozyme could be adsorbed on positively charged surfaces with"end-on"or"back-on"orientations. The simulation results agree well with experimental results and verify the reliability of PTMC algorithm in probing protein orientation on surfaces.2. We extend the temperature PTMC to Hamiltonian replica exchange Monte Carlo (REMC) algorithm in the SCD and IS space. Then we use the algorithm to simulate the orientation of adsorbed glucose oxidase (GOx) which has wide applications in biotechnology. The findings show that when attractive electrostatic interactions dominate, GOx tends to be adsorbed on positively charged surfaces with the"standing"orientation; whereas van der Waals interactions become dominant, the number of possible orientations increases and GOx is adsorbed mainly with the"front-lying"orientation, in which the active site of the enzyme becomes inaccessible. When adsorbed on negatively charged surfaces,"back-lying"orientation is the dominant one. The adsorbed orientation is not significantly affected by IS due to the nonuniform charge distribution over GOx surface and the local highly positively charged region of GOx. In general, when driven by electrostatic interactions, GOx tends to be adsorbed with ordered orientations, either"standing"or"back-lying"which are favorable to retain its bioactivity. However, when taking the interaction energy and kinetics into account, GOx adsorption on positively charged surfaces is more preferred for biotechnological applications.3. With the preferred orientations of protein adsorbed on surfaces by REMC simulations, we are able to further study protein conformation on surfaces by molecular dynamics (MD) simulations. Considering the convenience and flexibility of controlling protein behavior via external factors and the sensitive response of protein to an electric field (EF), we investigate the lysozyme adsorption on a negatively charged SAM by all-atom MD simulations. Results suggest that lysozyme adsorption is facilitated by applied high positive EF and inhibited by high negative EF. However, the promotion or inhibition effect by EF does not monotonically increase with the strength of EF, possibly resulted by the competitive adsorption of cations with lysozyme on the negatively charged surface. Binding site analysis reveals that the positively charged residues Arg and Lys are mainly responsible for lysozyme adsorption on the negatively charged surface. The binding residues interact with the functional groups of the SAM through hydrogen bonding. In the presence of EFs, lysozyme tends to be adsorbed with the"side-on"orientation. Though the detailed binding residues may vary for these orientations, residues Lys1, Lys33 and Lys116 are found to be on the binding side close to the surface. Conformation analysis suggests that too high field strength may induce severe structural deformation of lysozyme; while lower strength EFs influence the structure deformation differently. Certain field strengths are observed to have a stabilizing effect on the protein conformation possibly due to the rearrangement of the position of the local atomic charges of protein to couple its dipole with the applied external EF, which may help immobilize internal flexible structure and thus restrict the occurrence of specific conformational changes.4. It is helpful to understand the mechanism of surface resistance to protein adsorption in experimentally synthesizing anti-fouling materials. Since there is still no systematic explanation on the mechanism of sulfobetaine materials, we perform all-atom MD simulations to study the interactions between peptide Neuromedin-B (NMB) and the S(CH2)10N+(CH3)2CH2CH(OH)CH2SO3- (SBT-) SAM. We also simulate the adsorption of NMB on the S(CH2)10OH (OH-) and S(CH2)10CH3 (CH3-) SAMs for comparison. The surface and water force-distance profiles show that the surface resistance to peptide adsorption is mainly generated by the hydrated water molecules adjacent to surfaces; but surfaces themselves may also set an energy barrier for the approaching peptide. For the SBT-SAM, the surface first exerts a relatively high repulsive force and then a rather week attractive force on the approaching peptide; meanwhile the hydrated water molecules act strong repulsive force on the peptide. Therefore, the SBT-SAM exhibits good protein-resistant property. For the OH-SAM and CH3-SAM, surfaces show low or little energy barrier but strong affinity for the peptide; and the hydrated water molecules only act repulsive force within much narrower range and with lower intensity compared with the case for the SBT-SAM. The analyses of the peptide, surface and water structure and dynamics properties indicate that possible factors contributing to surface resistance include the number of hydrogen bonds formed adjacent to SAMs, mobility of water molecules near surfaces, surface packing density, chain flexibility of SAMs. The SBT-SAM has a much larger number of hydrogen bonds formed between the hydrated water molecules and the functional groups of the SAM, which greatly lowers the mobility of the water molecules near the surface. This tightly bound water layer effectively reduces the direct contact between surface and peptide. Furthermore, the SBT-SAM also has high flexibility and low surface packing density, which allow water molecules to penetrate into the surface to form tightly bound hydrogen bond networks and therefore reduce the affinity between surface and peptide. The results show that the protein-resistant ability of the SAMs are in the decreasing order of SBT-SAM > OH-SAM > CH3-SAM.