Received for publication May 23, 2008.
Revised August 12, 2008.
Accepted for publication August 13, 2008.

Drug binding to the inactivated state is necessary but not sufficient for high affinity binding to hERG channels

Abstract

Drug block of the human ether a-go-go related gene K⁺ channel (hERG) is the most common cause of acquired long QT syndrome, a disorder of cardiac repolarization that may result in ventricular tachycardia and sudden cardiac death. We investigated the open versus inactivated state dependence of drug block by utilizing hERG mutants N588K and N588E, that shift the voltage dependence of inactivation compared to wild-type but where the mutated residue is remote from the drug-binding pocket in the channel pore. Four high affinity drugs (cisapride, dofetilide, terfenadine, astemizole) demonstrated lower affinity for the inactivation-deficient N588K mutant hERG channel compared with N588E and wild type hERG. Three of four low affinity drugs (erythromycin, perhexiline, and quinidine) demonstrated no preference for N588E over N588K channels, while dl-sotalol was an example of a low-affinity state-dependent blocker. All five state-dependent blockers showed an even lower affinity for S620T mutant hERG (no inactivation) compared with N588K mutant hERG (greatly reduced inactivation). Computer modeling indicates that the reduced affinity for S620T compared to N588K and wild type channels can be explained by the relative kinetics of drug block and unblock compared to the kinetics of inactivation and recovery from inactivation. We were also able to calculate, for the first time, the relative affinities for the inactivated versus the open state, which for the drugs tested here ranged from 4-70 fold. Our results show that preferential binding to the inactivated state is necessary but not sufficient for high affinity binding to hERG channels.

Key words: Thermodynamic and kinetic processes and modeling, Func. analysis receptor/ion channel mutants