Abstract
We have performed systematic theoretical studies to elucidate the factors governing the binding affinity and specificity of lanthanide cations for protein binding sites. Specifically, we have addressed the following three questions: (1) what is the most thermodynamically preferable set of protein ligands for La3+ and Ca2+, (2) what is the most preferable Ca2+-binding site for La3+ to replace Ca2+, and (3) how would monodentate vs. bidentate carboxylate binding affect the substitution of Ca2+ for La3+ in EF-hand binding sites. To address these questions, we used density functional theory combined with continuum dielectric methods to compute the free energies for (1) successively replacing a metal-bound water molecule with a carboxylate or a carbonyl group in La3+ and Ca2+ complexes, and (2) replacing Ca2+ with La3+ in classical EF-hand binding sites with and without changing the original carboxylate-binding mode. The calculations reveal three key factors and the corresponding physical bases favoring the substitution of trivalent lanthanides for Ca2+ in EF-hand motifs. First, a solvent-shielded Ca-binding cavity facilitates lanthanides to replace Ca2+, as it enhances favorable metal-ligand interactions. Second, the more Asp/Glu there are in the Ca-binding pocket, the greater the affinity for lanthanides relative to Ca2+, as trivalent lanthanides can accept more negative charge from the carboxylates than divalent Ca2+. Third, the availability of both carboxylate oxygen atoms to bind without penalty to lanthanide cations also facilitates lanthanides to replace Ca2+, as a trivalent lanthanide cation prefers binding a carboxylate bidentately more than divalent Ca2+. Thus, we predict that La3+ can dislodge Ca2+ from carboxylate-rich Ca-binding sites in buried cavities, if it can bind at least one Asp/Glu bidentately. The findings of this work are in accord with available experimental data.