Binding of 2,6- and 2,7-dihydroxynaphthalene to wild-type and E-B13Q insulins: dynamic, equilibrium, and molecular modeling investigations.

The binding of phenolic ligands to the insulin hexamer occurs as a cooperative allosteric process. Investigations of the allosteric mechanism from this laboratory resulted in the postulation of a model consisting of a three-state conformational equilibrium and the derivation of a mathematical expression to describe the insulin system. The proposed mechanism involves allosteric transitions among two states of high symmetry, designated T3T3' (a low affinity state) and R3R3' (a high affinity state), and a third state of lower symmetry, designated T3oR3o (a state of mixed low and high affinities). To further characterize this mechanism, we present rapid kinetic fluorescence studies, equilibrium binding isotherms, and molecular modeling investigations for the Co(II)-substituted wild-type and E-B13Q mutant hexamers. These studies show that the measured on and off rates (kon and koff) for the binding of the allosteric ligands 2,6- and 2,7-dihydroxynaphthalene provide an independent measure of the dissociation constant for binding to the T3oR3o conformation (KRo). These constants are in agreement with the value obtained by computer fitting of the equilibrium binding isotherms to the quantitative allosteric mechanism. We analyze the structural differences between the T3oR3o and R6 phenolic binding sites and predict the structures of the T3oR3o-2,6-DHN and R6-2, 6-DHN complexes by 3-D molecular modeling. Assignment of H-bonding of the first hydroxyl group to CysA6 and CysA11 has been supported by stacking interactions analogous to phenol using 1H-NMR. H-bonding of the second hydroxyl group of 2,6-DHN to the GluB13 carboxylate side chains is predicted by molecular modeling and is supported by a reduction of affinity for Ca2+, which is postulated to bind to the GluB13 side chains.