A novel approach to dissect ions occupancy and dissociation constant of individualbinding sites within the K+-channel selectivity filter.

Abstract

K+-channels permeate Rb+ as well as they do K+, although with a lower single channel conductance. KcsA crystal structure displays four K+ bound to its selectivity filter (SF); however, when crystallized in Rb+, ions were only found at the binding sites 1, 3, and 4. This observation explains the lower single conductance of K+-channels in rubidium. We reported a KcsA mutant (G77A) that stabilized the SF in the 2-4 ion bound configuration with K+ ions coordinated in the 2nd and 4th binding sites. Crystallizing the KcsA-G77A structure in rubidium interestingly showed Rb+ ions bound to the 1st and 4th binding sites. Like in the wild-type channel, the 2nd binding site was uncapable of binding rubidium. Therefore, we used the G77A to measure the 1st binding site dissociation constant for Rb+ by Isothermal Titration Calorimetry (ITC). Previously, computational studies have suggested that: 1) the SF's 4th ion binding site can be occupied by K+ or Na+ and hence it should not contribute to the SF's K+ binding affinity and 2) the 2nd binding site is responsible for the K+ binding affinity displayed by K+-channels SF. In here, we are showing that in K+-channels the 2nd binding site is indeed responsible for K+ binding affinity while the 1st site is responsible for the Rb+ ion binding affinity, since the G77A mutant (a simplified structural model for the SF) showed a wild-type like Rb+ dissociation constant. Therefore, we propose that in K+-channels the 1st ion binding site is responsible for the high rubidium over sodium selectivity shown by KcsA and potentially other potassium channels. Our advances in applying this approach for the alkali metal ion series will be presented.