Molecular Dynamics Simulation Study Of Protein-protein And Protein-DNA Complexes

The first molecular dynamics (MD) simulations of protein was reported by McCammon JA et al in1977. This opened a new era for "function from structure" research in biophysics. MD simulation provides direct information of biomolecule’s fluctuation behavior and uncovers important roles that dynamics played in biological processes such as protein-ligand binding, enzyme catalytic reaction and other important biomolecular functions. Furthermore, MD simulation provides an efficient approach for sampling structures needed for computing thermodynamic properties of proteins (such as reaction energy barrier of chemical process catalyzed by enzyme, ionization constant for ionizable amino acids in proteins and free energy landscape of protein folding).However, accuracy of these dynamic and thermodynamic properties derived from MD simulation depends very sensitively on the accuracy of the force field imported. This means that small inaccuracy in the potential energy surface may bring great conformation errors during MD simulations. But the classical force fields based on atom-atom pair interaction potential method were generally performed for theoretical research of biological macro molecules. This method omits electronic movement, therefore the electronic polarization effect is not considered in classical force field. This approximate treatment would bring some errors for calculating the electrostatic interaction. Furthermore, the electrostatic interaction is one of the most important part of non-bonded interactions.In order to accurately describe the electrostatic interactions of biological macromolecules, we developed a polarized protein-specific charge (PPC) and a polarized nucleic acid-specific charge (PNC) for accurate representation of electrostatic interaction in large biomolecules. The electronic density of large biomolecules is computed by combing a molecular fractionation with conjugate caps approach (MFCC) and Poisson-Boltzmann (PB) method. The computed electron density of biomolecules is utilized to derive PPC/PNC that represents polarized electrostatic state of biomolecules near the native structure. For the electronic density calculated by MFCC-PB method itself contains the polarization effect, thus the PPC/PNC charge fitted by the electronic density will automatically include the polarization effect.Glucokinase (GK) is a glycolic enzyme that catalyzes the phosphorylation of glucose to glucose-6-phosphate in the first step of glycolysis. Thus alteration in GK activity plays an important role in abnormal glycemia. In chapter3, we investigated the activation mechanism of GK by single point mutation. The explicit molecular dynamics simulations and implicit solvent binding free-energy calculations were investigated to understand the activation mechanism of GK M197V (Met197â†'Val) mutation. The root mean square fluctuation (RMSF) and dynamic cross-correlation matrices (DCCM) conformation analysis showed that GK M197V mutation resulted in a more stable active conformation. The binding free-energy analysis demonstrated that the GK M197V mutation increased its binding affinity with glucose. Our computed results show that M197V mutation can improve the stability of active conformation of GK, and its role is similar as the small activator.The lactose repressor (LacI) controls the expression of a set of genes involved in the lactose metabolism in the bacterium Escherichia coli. The DNA binding domain of LacI is made by the first62residues of the N-terminal, and this62-residue long amino terminal fragment of LacI can bind DNA in a specific and non-specific mode. In chapter4, MD simulations were performed to investigate the conformation and energy difference between the non-specific and specific complex. By comparing the hydrogen bond occupancy, we found that most hydrogen bonds for non-specific complex had a direct interaction with phosphate, and had little contribution to the direct base sequence recognition. Whereas the hydrogen bonds for specific complex with direct interaction with DNA bases are much more than those for non-specific complex. Our energy computed results show that the total binding free energy for specific complex is lower than that for non-specific complex. But the electrostatic interaction for non-specific complex is much lower than the corresponding value for specific complex. The van der Waals interaction is main driving force for switching from non-specific to specific complex. Our computed result is, in general, in accordance with the base recognition process predicted by the previous experimental study.In chapter5, we detailed investigate the interaction between the Tn916N-terminal domain (INT-DBD) and DNA. Unlike most reported DNA-binding domains that bind to the major groove using α-helix, INT-DBD recognizes the major groove using the face of a three-stranded β-sheet. Thus the information on this structurally novel protein-DNA complex is essential to gaining a deeper insight into the strategies used by protein for DNA recognition. Since the negatively charged DNA directly interacts with the positively charged residues (such as Arg and Lys) of INT-DBD, the electrostatic interaction is expected to play an important role in the dynamical stability of the protein-DNA binding complex. Therefore the combined use of quantum-based PPC for protein and PNC for DNA were employed to investigate the special DNA base recognition mechanism. Our study shows that the protein-DNA structure is stabilized by polarization and the calculated protein-DNA binding free energy is in good agreement with the experimental data. Furthermore, our study revealed a positive correlation between the measured binding energy difference in alanine mutation and the occupancy of the corresponding residue’s hydrogen bond. This correlation relation directly relates the contribution of a specific residue to protein-DNA binding energy to the strength of the hydrogen bond formed between the specific residue and DNA.It is well known, the hotspot residues between protein and protein have a very important role in the conformation stability. In chapter6, MD simulations of the wild type (WT) and three hotspot mutants (D51A, Y54A and Y55A of Im9) of the E9-Im9complexes were carried out to investigate specific interaction mechanisms of these three hotspot residues. The changes of binding energy between the WT and mutants of the complex were computed by the MM/PBSA method using a polarized force field and were in excellent agreement with experiment values. It verifies that these three residues were indeed hotspot residues of the binding complex. By contrast, calculation by using the standard (non-polarized) AMBER99SB force field produced binding energy changes from these mutations in opposite direction to the experimental observation. Dynamic hydrogen bond analysis showed that conformations sampled from MD simulation in the standard AMBER force field were distorted from the native state and they disrupted the inter-protein hydrogen bond network of the protein-protein complex. The current work further demonstrated that electrostatic polarization plays a critical role in modulating protein-protein binding.