hERG 1b is critical for human cardiac repolarization

Abstract

The human ether-à-go-go-related gene (hERG; or KCNH2) encodes the voltage-gated potassium channel underlying IKr, a repolarizing current in the heart. Mutations in KCNH2 or pharmacological agents that reduce IKr slow action potential (AP) repolarization and can trigger cardiac arrhythmias associated with long QT syndrome. Two channel-forming subunits encoded by KCNH2 (hERG 1a and 1b) are expressed in cardiac tissue. In heterologous expression systems, these subunits avidly coassemble and exhibit biophysical and pharmacological properties distinct from those of homomeric hERG 1a channels. Despite these findings, adoption of hERG 1a/1b heteromeric channels as a model for cardiac IKr has been hampered by the lack of evidence for a direct functional role for the 1b subunit in native tissue. In this study, we measured IKr and APs at physiological temperature in cardiomyocytes derived from human induced pluripotent stem cells (iPSC-CMs). We found that specific knockdown of the 1b subunit using shRNA caused reductions in 1b mRNA, 1b protein levels, and IKr magnitude by roughly one-half. AP duration was increased and AP variability was enhanced relative to controls. Early afterdepolarizations, considered cellular substrates for arrhythmia, were also observed in cells with reduced 1b expression. Similar behavior was elicited when channels were effectively converted from heteromers to 1a homomers by expressing a fragment corresponding to the 1a-specific N-terminal Per-Arnt-Sim domain, which is omitted from hERG 1b by alternate transcription. These findings establish that hERG 1b is critical for normal repolarization and that loss of 1b is proarrhythmic in human cardiac cells.

Keywords: KCNH2; arrhythmia; early afterdepolarization; hERG; long QT syndrome.

Fig. 1.

shRNA reduces hERG 1b expression levels. iPSC-CMs immunostained for hERG 1b (magenta) and actin (green; A) under mock-transfected control conditions or (B) after transfection with 1b shRNA. (C) Sample linear fluorescence intensity profiles displaying hERG 1b fluorescence in control (blue) and shRNA-transfected (orange) iPSC-CMs. [Scale bar: 10 arbitrary units (AUs) × 5 μm.) (D) Plot of average fluorescence intensity in control (336.3 ± 24.5 AU/μm; n = 20) and shRNA-transfected (153.7 ± 11.8 AU/μm; n = 20; P < 0.001) cells. (E) Plot of hERG 1a and 1b mRNA levels amplified with qRT-PCR in 1b shRNA-transfected cells normalized to those in untransfected cells (n = 10; P < 0.0001). Error bars are ± SEM. *Statistical significance relative to controls.

Fig. 2.

hERG 1b modulation alters native I_Kr. (A) Sample E-4031–sensitive traces representing endogenous I_Kr from (Top) control, (Middle) shRNA-transfected, and (Bottom) PAS-infected iPSC-CMs. Pulses from −40 to 20 mV in 10-mV steps are shown. The final 3 s of the 6-s tail pulse are omitted for clarity. (Scale bar: 0.5 pA/pF × 1 s.) (B) Peak tail current density measured at −40 mV, plotted as a function of prepulse potential, and fitted with a Boltzmann function for control (black squares), shRNA-transfected (blue circles), and PAS-infected (green triangles). (C) Normalized samples of (Upper) −40-mV tail current traces of native I_Kr and (Lower) hERG currents overexpressed in iPSC-CMs. (Scale bar: 1 s.) (D) Weighted time constants of deactivation (τ_W) for native I_Kr (control, shRNA, and PAS) compared with overexpressed 1a and 1a/1b currents in iPSC-CMs. *Statistical significance at P < 0.05.

Fig. 3.

hERG 1b modulation prolongs iPSC-CM APD. (A) Typical AP recorded from iPSC-CM transfected with nontargeting shRNA control. Sample traces showing EADs from (B) mock-transfected control, (C) 1b shRNA-transfected, and (D) PAS-infected iPSC-CMs. The orange trace in B is from the same recording after perfusion of 12 nM E-4031. (E) APD₉₀ for control, shRNA-, and PAS-transduced iPSC-CMs during 1-Hz pacing. (F) Fractional occurrence of EADs in control, shRNA-, and PAS-transduced iPSC-CMs. All data are reported as means ± SEMs. n = 5–10. (Scale bar: 40 mV × 400 ms.) *Statistical significance at P < 0.05.

Fig. 4.

hERG 1b modulation increases APD variability. (A) Sample APs depicting beat-to-beat variability at 1-Hz pacing in nontargeting shRNA-transfected (Non), 1b shRNA-transfected, mock-transfected, and PAS-infected iPSC-CMs; 50 sequential APs are shown for each. (B) Average variance from 50 paced APs for non (35 ± 9 ms²; n = 5), shRNA (3,441 ± 1,171 ms²; n = 10), mock (220 ± 93 ms²; n = 6), and PAS (1,780 ± 463 ms²; n = 7) iPSC-CMs. (C) Histograms depicting the distribution of the normalized APD₉₀ for non and mock control (gray), shRNA (green), and PAS (blue) iPSC-CMs. (Scale bar: 20 mV × 200 ms.) *Statistical significance at P < 0.05.