Binding and unbinding of potent melatonin receptor ligands: Mechanistic simulations and experimental evidence

Abstract

The labeled ligand commonly employed in competition binding studies for melatonin receptor ligands, 2-[¹²⁵I]iodomelatonin, showed slow dissociation with different half-lives at the two receptor subtypes. This may affect the operational measures of affinity constants, which at short incubation times could not be obtained in equilibrium conditions, and structure–activity relationships, as the K_i values of tested ligands could depend on either interaction at the binding site or the dissociation path. To address these issues, the kinetic and saturation binding parameters of 2-[¹²⁵I]iodomelatonin as well as the competition constants for a series of representative ligands were measured at a short (2 h) and a long (20 h) incubation time. Concurrently, we simulated by molecular modeling the dissociation path of 2-iodomelatonin from MT₁ and MT₂ receptors and investigated the role of interactions at the binding site on the stereoselectivity observed for the enantiomers of the subtype-selective ligand UCM1014. We found that equilibrium conditions for 2-[¹²⁵I]iodomelatonin binding can be reached only with long incubation times, particularly for the MT₂ receptor subtype, for which a time of 20 h approximates this condition. On the other hand, measured K_i values for a set of ligands including agonists, antagonists, nonselective, and subtype-selective compounds were not significantly affected by the length of incubation, suggesting that structure–activity relationships based on data collected at shorter time reflect different interactions at the binding site. Molecular modeling simulations evidenced that the slower dissociation of 2-iodomelatonin from the MT₂ receptor can be related to the restricted mobility of a gatekeeper tyrosine along a lipophilic path from the binding site to the membrane bilayer. The enantiomers of the potent, MT₂-selective agonist UCM1014 were separately synthesized and tested. Molecular dynamics simulations of the receptor–ligand complexes provided an explanation for their stereoselectivity as due to the preference shown by the eutomer at the binding site for the most abundant axial conformation adopted by the ligand in solution. These results suggest that, despite the slow-binding kinetics occurring for the labeled ligand, affinity measures at shorter incubation times give robust results consistent with known structure–activity relationships and with interactions taken at the receptor binding site.