Date of Award
Doctor of Philosophy (PhD)
Jeffery G. Saven
Molecular design and modeling can provide stringent assessment of our understanding of the structure and function of proteins. Due to the subtleness of the interactions that largely stabilize proteins, computational methods have been particularly valuable in establishing practical, formal and physically grounded protocols to study the structure and function of these biomolecules. Especifically, computational protein design seeks to identify sequences that fold into a desired structure and have specific structural and functional properties using computational methodologies. Among current techniques, an entropy-based formalism that efficiently determines the number and composition of sequences satisfying a predefined set of constraints seems particularly promising and powerful. Complementary to this methodology are the well-established molecular dynamics simulation techniques that have been extensively used to study structure, function and dynamics of biologically relevant systems. Herein different studies of systems using computational techniques to address particular molecular problems are described. Efforts to redesign membrane proteins to generate water-soluble variants were applied to a widely studied pentameric ligand-gated ion channel, the nicotinic acetylchoilne receptor (nAChR). NMR structures and binding studies demostrated the robustness and applicability of the computational design approach. Toward the creation of water-soluble variants of a G protein–coupled receptor (GPCR), comparative modeling and docking calculations were used to investigate the structure of the human μ opioid receptor and presented in light of previous mutagenesis studies of structure and agonist-induced activation. Candidate peptides for possible therapeutic agents were computationally analyzed. Peptide design, loop modeling and MD simulations were applied to investigate the stromal cell-derived factor-1&a; (SDF-1&a;). SDF-1&a; displays promising therapeutic benefits to treat blood-supply related heart disease and elicit growth of microvasculature. Simplified analogs of SDF-1&a; exhibit enhanced therapeutic properties in cell-based assays. MD simulations provide insights about the molecular features of this enhancement. One simplified peptide offers a potentially clinically translatable neovasculogenic therapy. Lastly, MD simulations were utilized to analyze a molecule with hindered internal rotors, a tribenzylamine hemicryptophane. The molecule was characterized by different experimental and computational techniques. The structural and dynamic features of the hemicryptophane molecule make it an attractive starting point for controlling internal rotation of aromatic rings within molecular systems.
Perez Aguilar, Jose Manuel, "Computational Protein Design and Molecular Dynamics Simulations: A Study of Membrane Proteins, Small Peptides and Molecular Systems" (2012). Publicly Accessible Penn Dissertations. 480.