Evaluation of functional and binding assays in cells expressing either recombinant or endogenous hERG channel

Introduction: The hERG (human ether-a-go-go related gene) potassium channel is required for normal cardiac repolarization, is susceptible to inhibition by a wide variety of compounds, and its blockage can lead to cardiac QT interval prolongation and life threatening arrhythmias. The present report examines the ability of hERG binding and functional assays to identify compounds with potential cardiovascular liabilities at the earliest stages of drug discovery. Methods: Competitive binding assays were developed using 3H-dofetilide and membranes from HEK293EBNA cells stably expressing recombinant hERG (HEK293-hERG) and IMR-32 cells expressing hERG endogenously. hERG functional assays were also developed using membrane potential indicator dye and rubidium efflux. The ability of these assays to identify compounds with potential adverse cardiac effects was examined using drugs with known cardiac effects ranging from those with no known adverse effects to drugs that were withdrawn from the market due to increased risk of sudden death associated with Torsades de Points. Results: Binding assays using HEK293-hERG membranes and 3H-dofetilide were robust (Z′ = 0.69 ± 0.015, mean ± S.E.M.), highly reproducible (test–retest slope = 1.04, r2 = 0.98), and correlated well with IC50 values obtained by patch clamp (slope = 0.98, r2 = 0.89). Binding assays using IMR-32 membranes were less sensitive (Z′ = 0.4 ± 0.03, mean ± S.E.M., false negative rate = 0.4) but still correlated well with patch clamp data (slope = 1.06, r2 = 0.83). The hERG membrane potential assay could detect potent hERG inhibitors (defined by hERG patch clamp IC50 < 0.1 μM) using HEK293-hERG cells, but were prone to generate false-negative results with less potent inhibitors (false negative rate = 0.5). Finally, the rubidium efflux assay gave highly reproducible results (Z′ = 0.80 ± 0.02, mean ± S.E.M.) that correlated with patch clamp IC50 values (slope = 0.87, r2 = 0.73). Discussion: The hERG binding and rubidium efflux assays are robust, predictive of patch clamp results, and can be used at the earliest stages of drug discovery.