Designing small, nonsugar activators of antithrombin using hydropathic interaction analyses.

Conformational activation of antithrombin is a critical mechanism for the inhibition of factor Xa, a proteinase of the blood coagulation cascade, and is typically achieved with heparin, a polyanionic polysaccharide clinically used for anticoagulation. Although numerous efforts have been directed toward the design of better activators, a fundamental tenet of these studies has been the assumed requirement of an oligo- or a polysaccharide backbone. We demonstrate here a concept that small nonsaccharidic nonpolymeric molecules may be rationally designed to interact with and activate antithrombin for enhanced inhibition of factor Xa. The rational design strategy is based on a study of complexes of natural and mutant antithrombins with heparin-based oligosaccharides using hydropathic interaction (HINT) technique, a quantitative computerized tool for analysis of molecular interactions. A linear correlation was observed between the free energy of binding for antithrombinminus signoligosaccharide complexes and the HINT score over a wide range of approximately 13 kcal/mol, indicating strong predictive capability of the HINT technique. Using this approach, a small, nonsugar, aromatic molecule, (minus sign)-epicatechin sulfate (ECS), was designed to mimic the nonreducing end trisaccharide unit DEF of the sequence specific heparin pentasaccharide DEFGH. HINT suggested a comparable antithrombin-binding geometry and interaction profile for ECS and trisaccharide DEF. Biochemical studies indicated that ECS binds antithrombin with equilibrium dissociation constants of 10.5 and 66 microM at pH 6.0, I 0.025, and pH 7.4, I 0.035, respectively, that compare favorably with 2 and 80 microM observed for the natural activator DEF. ECS accelerates the antithrombin inhibition of factor Xa nearly 8-fold demonstrating for the first time that conformational activation of antithrombin is feasible with appropriately designed small nonsugar organic molecules. The results present unique opportunities for de novo activator design based on this first-generation lead.