DOI
https://doi.org/10.25772/CQXM-W341
Defense Date
2008
Document Type
Dissertation
Degree Name
Doctor of Philosophy
Department
Medicinal Chemistry
First Advisor
Umesh R. Desai
Abstract
Glycosaminoglycans (GAGs) are complex biopolymers that play important roles in inflammation, coagulation, angiogenesis, cell adhesion and viral invasion by interacting with several different proteins.1,2 Structurally, GAGs are built up of several different sulfated disaccharide units.3 Specific GAG sequences that uniquely recognize their cognate proteins exist. Such specificity typically arises from the binding of unique sulfation patterns on the linear GAG chain to highly electropositive protein domains. Thus, these highly charged, sulfated biopolymers potentially represent a new class of therapeutics. Yet, the major stumbling block to the development to these agents is their extremely complicated and tedious chemical synthesis. We hypothesized that replacing the saccharide skeleton with an equivalent non-saccharide and readily synthesized organic skeleton would usher in an era of new, GAG-based therapeutics. This challenge has been addressed on two fronts, computational design and chemical synthesis, by focusing on the heparin pentasaccharide-antithrombin system that represents an exhaustively studied model GAG-protein system. With respect to chemical synthesis, a microwave-based synthetic procedure that can rapidly introduce multiple sulfate groups on a poly-hydroxyl substrate within minutes was developed.4 Using this method, the synthesis of a previously designed activator (IAS5), which otherwise proved to be problematic, was successfully completed. Biochemical screening of IAS5 and its analogs revealed that these molecules could activate antithrombin up to 30-fold in comparison to the 300-fold activation by the heparin pentasaccharide. In an effort to develop more potent antithrombin activators, a new method to predict high affinity GAG sequences for a given GAG-binding protein based on combinatorial virtual-library screening was developed.5 This combinatorial virtual-library screening method was applied to a library of 24,576 non-saccharide, sulfated molecules that were created using the structure of IAS5 as a template. Thirty seven‘hits’ that had common structural features were identified from this study. Interestingly, all these ‘hits’ bind to antithrombin similarly and orient the 4 negative charges identical to the corresponding groups in the heparin pentasaccharide. The synthesis of selected targets is currently in progress and several synthetic steps have already been optimized.
Rights
© The Author
Is Part Of
VCU University Archives
Is Part Of
VCU Theses and Dissertations
Date of Submission
August 2008