Optimizing the metal binding parameters of an EF-hand-like calcium chelation loop: coordinating side chains play a more important tuning role than chelation loop flexibility.

Abstract

In calcium signaling pathways regulated by the EF-hand Ca2+ binding motif, proper regulation requires that the equilibrium and kinetics of Ca2+ binding to the EF-hand chelation loop be precisely optimized for each physiological application. Studies of small-molecule organic chelators have shown that metal binding parameters can be tuned both by the nature of the coordinating ligands and by the structural framework to which these ligands are attached. By analogy, the present study tests the relative importance of (i) coordinating side chains and (ii) backbone torsion angle constraints to the tuning of an EF-hand-like Ca2+ chelation loop. A series of engineered chelation loops are generated by modifying Ca2+ binding site of the Escherichia coli galactose binding protein. The resulting loops, each containing an altered coordinating side chain or a Gly substitution, are compared with respect to their metal binding affinities, specificities, and dissociation kinetics. The Gly variants examined include substitutions which eliminate or introduce a Gly at each of the nine chelation loop positions. The results reveal that Gly is not tolerated at loop positions 1, 3, 5, or 8 or at the external coordinating position, where the removal of a key coordinating or hydrophobic side chain destabilizes the protein. In contrast, Gly residues at loop positions 2, 4, 6, and 7, none of which is required for side chain coordination, have little effect on Ca2+ affinity and the ability to discriminate between cations of different size and charge. Kinetic measurements show that some of these Gly residues measurably alter the rates of metal ion association and dissociation, but in each case the two rates are changed by approximately the same factor so that the effects on equilibrium are minor. Overall, Gly residues yield surprisingly small effects at loop positions 2, 4, 6, and 7, especially when compared to the larger equilibrium and kinetic effects observed for coordinating side chain substitutions. It follows that the conserved Gly at position 6 is not required for Ca2+ binding and that constraints on the backbone torsion angles at the non-coordinating side chain positions 2, 4, 6, and 7 play a relatively minor role in tuning metal binding parameters. Instead, specific coordinating side chains optimize the metal binding parameters of the GBP chelation loop for its protein context and biological application.