Employment of molecular dynamics simulations in the investigation of biological systems.

• Main Author: Yip, Yew Mun.
• Other Authors: School of Physical and Mathematical Sciences
• Format: Theses and Dissertations
• Language: English
• Published: 2013
• Subjects: DRNTU::Science::Chemistry::Physical chemistry::Molecular structure and bonding
• Online Access: http://hdl.handle.net/10356/54761
• Institution: Nanyang Technological University

Diseases and viruses are prevalent in the world today. While many of the disease pathways have been unearthed, many are still trying to locate the key proteins that have the greatest impact on inhibiting the disease. Upon discovering these proteins, potential inhibitors would then be synthesized and tested in vitro before they make their way to in vivo testing. Such a process is time-consuming, especially if there is no structural information of the inhibitor-binding site to direct the design of the inhibitor. Therefore, molecular modeling becomes an important tool in eliciting such information. Besides constructing a structural model of the protein, molecular dynamics simulations are also required to illustrate the dynamic nature of the protein so as to simulate the actual conformational changes that may have occurred in vivo. This thesis presents the construction of the HK2-VDAC1 complex (Project 1) found in cancer cells ever since the Warburg effect was discovered. This complex is important in aiding the understanding of how cancer cells obtain a source of cellular energy for cell proliferation. Hexokinase (HK) also plays an important role by binding to the Voltage Dependent Anion Channel 1 (VDAC1) and prevents apoptosis. Therefore, eliciting the structural model of the complex will allow the understanding of how various inhibitors such as 3-bromopyruvate (3-BP) and methyl jasmonate (MJ) disrupt this association and cause apoptosis. In addition, the VDAC1 N-terminus helix has been found to regulate the flow of ATP, which serves as an important source of cellular energy, from within the mitochondria to the cytosol. Therefore, with the proposed model, we established an understanding of how HK2 prevents VDAC1 from down-regulating the ATP flow in cancer cells and maintain a constant source of cellular energy for cell proliferation. Project 2 targets the knowledge that antibodies HK20 and D5 bind to HIV-1 gp41, thereby inhibiting membrane fusion that facilitates viral entry. However, the binding mode is not dynamic since it is simply based on the X-ray crystal structures which are analogous to a snapshot during the crystallization process. Therefore, in order to illustrate the dynamic binding picture, molecular dynamics simulation was carried out. The binding free energies calculated from the simulations are found to correlate with the trend of the experimental valuesfor both antibodies. Alanine scanning mutagenesis at the protein–protein interface reveals that the highest contributors to binding for the antibodies are aromatic and hydrophobic residues belonging to the VH and VL regions. In addition, various residues of HK20 and D5 bind to the gp41 hydrophobic binding pocket, an important region targeted by many other fusion inhibitors. Hydrogen bonds with high-occupancy were also identified at the periphery of gp41 hydrophobic pocket via hydrogen bonding analysis. Since the interface residues identified are turn residues, further work may involve turn mimics in the design of potential inhibitors. Pre-orientation by the hydrogen bonds to poise this particular turn towards the binding pocket may also be a point worth pursuing. In Project 3, the thermodynamic quantities of Trp-cage were investigated using replica exchange molecular dynamics (REMD) simulations by the incorporation of a newly developed polarized protein-specific charge (PPC). The native conformation of Trp-cage was eventually attained via a 3-step PPC update in REMD simulation. After obtaining the native conformation of Trp-cage, new PPC was derived from the obtained conformation and another REMD simulation was performed to explore the thermodynamic stability of Trp-cage. The theoretical melting temperature T m of 325 K obtained from the incorporation of PPC was found to be in close agreement with the experimental Tm of 315 K. This indicates that PPC was not only able to fold the protein correctly, but it also gives a Tm that correlates with the experimental value. This study illustrates the importance of electrostatic polarization in protein folding.