Molecular modeling of a putative antagonist binding site on helix III of the beta-adrenoceptor.

In recent biochemical studies it was demonstrated that residue Asp113 of the beta-adrenoceptor (beta-AR) is an indispensable amino acid for the binding of beta-AR antagonists. Earlier fluorescence studies showed that a tryptophan-rich region of the beta-AR is involved in the binding of propranolol, the prototype beta-AR antagonist. Bearing these two biochemical findings in mind, we explored the beta-AR part containing Asp113, for an energetically favorable antagonist binding site. This was done by performing molecular docking studies with the antagonist propranolol and a specific beta-AR peptide which included, besides Asp113, two possibly relevant tryptophan residues. In the docking calculations, the propranolol molecule was allowed to vary all its internal torsional angles. The receptor peptide was kept in an alpha-helix conformation, while side chains relevant to ligand binding were flexible to enable optimal adaptations to the ligand's binding conformation. By means of force-field calculations the total energy was minimized, consisting of the intramolecular energies of both ligand and receptor peptide, and the intermolecular energy. We found an antagonist binding site, consisting of amino acids Asp113 and Trp109, which enabled energetically favorable interactions with the receptor-binding groups of propranolol. According to these results, binding involves three main interaction points: (i) a reinforced ionic bond; (ii) a hydrogen bond; and (iii) a hydrophobic/charge transfer interaction. The deduced binding site shows a difference in affinity between the levo- and dextrorotatory isomers of propranolol caused by a difference in ability to form a hydrogen bond, which is in conformity with the experimentally observed stereoselectivity. Moreover, it also provides an explanation for the beta 1-selectivity of p-phenyl substituted phenoxypropanolamines like betaxolol. The p-phenyl substituent of betaxolol was shown to be sterically hindered upon binding to the beta 2-AR peptide, whereas this hindrance is very likely to be much less with the beta 1-AR peptide. Finally, the proposed antagonist binding site is discussed in the light of some recent biochemical findings and theories.