Modeling carbohydrates and their interactions with proteins

Carbohydrates are ubiquitously expressed at cell surfaces and involved in various cellular interactions. Understanding the shapes and dynamic properties of carbohydrates advances our knowledge of the fundamental properties that determine the specificity and affinity of their interactions with other biomolecules that are ultimately responsible for their biological functions. In this thesis, modeling methods were developed and applied to a number of particularly challenging aspects of carbohydrates and carbohydrate-protein complexes. Firstly, a new set of force field parameters was developed for modeling the highly plastic structures of monosaccharides containing five-membered rings (furanoses) that are constituents of DNA and RNA. Oligoand polysaccharides of furanoses are also important components of bacterial and fungal pathogen surfaces. This work illustrated that it is no longer necessary to make assumptions about ring conformational preferences in order to interpret experimental NMR data for furanoses; instead they may be determined directly and objectively from molecular dynamics (MD) simulation. Secondly, many oligoand polysaccharides decorated with chemical modifications, such as sulfation. Sulfated carbohydrates include the well-known example of heparin, as well as may others, including a sulfated fucan isolated from sea urchin Lytechinus variegatus. The impact of sulfation on oligosaccharide shape has received relatively study, and is the focus of a combined NMR and MD analysis here. That research led to the conclusion that the 3D orientation of residues in an oligosaccharide could be influenced significantly by stabilizing hydrogen bonds between sulfate and hydroxyl groups, and by destabilizing electrostatic repulsions between the anionic sulfate groups. Lastly, the binding of a sulfated glycopeptide (PSGL-1) and its analogs to P-Selectin was examined by MD simulation, with the goal of quantifying the contributions made to binding by each of the component amino acids and monosaccharides. The results provide a basis for the rational design of inhibitors for diseaserelated interactions associated with P-selectin. INDEX WORDS: Carbohydrate, Force field, Molecular dynamics simulation, 3D structure, Furanoses, Ring conformation, Anomeric effect, Pseudorotation, GLYCAM, AMBER, NMR, Scalar J-coupling constants, Glycan-protein interaction, MM/GBSA, MM/PBSA, Binding free energy, Sulfate fucan, Hydrogen bond, P-Selectin, Drug design. MODELING CARBOHYDRATES AND THEIR INTERACTIONS WITH PROTEINS