Despite major roles played by heparin and heparan sulfate in growth and morphology, coagulation, angiogenesis, immune response and viral infection, their interaction with nearly all proteins, except for antithrombin, remains poorly defined. The events of mid-2008 in which a contaminant present in pharmaceutical grade heparin led to 81 deaths in the US demonstrate further the critical importance of understanding heparin recognition of proteins. A major reason for the ill-defined structure-activity relationships of heparin/heparan sulfate is their phenomenal structural diversity. Both heparin and heparan sulfate are complex, highly anionic polysaccharides containing millions of structures, of which only select few sequences are expected to preferentially recognize a target protein. Identification of these structures is important not only for designing molecules that may modulate the role of heparin/heparan sulfate, but also for understanding fundamental biochemical principles that define these interactions, e.g., specificity of recognition, the mechanism and order of assembly of complexes, etc. Despite the major advances in biophysical techniques, no technology is available to study the millions of diverse structures presented by heparin/heparan sulfate. In this context, computational docking approaches represent a powerful means of deducing both fundamental biochemical principles of interactions as well as identifying key sequence(s) or 'needle(s) in a haystack'. We developed a robust computational approach, called the combinatorial virtual library screening (CVLS) approach, that predicted 'high affinity and high specificity'heparin/heparan sulfate sequences binding to antithrombin from a library of nearly 7,000 hexasaccharides of varying levels of sulfation. CVLS represents a major advance in understanding heparin/heparan sulfate interactions with proteins and possesses major capabilities of deciphering and unraveling the diversity of heparin/heparan sulfate interactions. To further develop our CVLS technology in understanding biochemical principles and designing H/HS mimetics, we propose to: 1. Further develop the CVLS technology for studying H/HS binding to proteins. The working hypothesis in this aim is that application of CVLS to i) serpins (antithrombin and heparin cofactor II);ii) proteases (factors IIa, Xa and IXa of the coagulation cascade);and iii) viral glycoproteins (gD of HSV-1) will enhance fundamental mechanistic understanding the interactions and identify 'specific'interactions that may be targeted for modulation. 2. Design a combinatorial virtual library of aromatic, non-sugar H/HS mimetics to identify potential modulators of H/HS - antithrombin interaction. The working hypothesis is that the CVLS technology should be applicable to any class of molecule, and not just sulfated H/HS oligosaccharides. We have previously discovered that the bicyclic-unicyclic scaffold potently recognizes antithrombin. A library of these sulfated organic structures will be built in a combinatorial manner and assessed using the dual-filter approach for identification of advanced H/HS mimetics. We hypothesize that our strategy can be generalized and may be exploited for other H/HS - protein systems.
Despite major roles played by heparin and heparan sulfate in growth and morphology, coagulation, angiogenesis, immune response and viral infection, their interaction with nearly all proteins, except for antithrombin, remains poorly defined. The events of mid-2008 in which a contaminant present in pharmaceutical grade heparin led to 81 deaths in the US demonstrate further the critical importance of understanding heparin recognition of proteins. The proposed research on developing a robust technology for understanding how heparin and heparan sulfate bind to proteins aims to improve upon current pharmaceutical agents as well as design better drugs.
Showing the most recent 10 out of 14 publications
