Two small-molecule activators share similar effector sites in the KCNQ1 channel pore but have distinct effects on voltage sensor movements

Abstract

ML277 and R-L3 are two small-molecule activators of KCNQ1, the pore-forming subunit of the slowly activating potassium channel I_Ks. KCNQ1 loss-of-function mutations prolong cardiac action potential duration and are associated with long QT syndrome, which predispose patients to lethal ventricular arrhythmia. ML277 and R-L3 enhance KCNQ1 current amplitude and slow deactivation. However, the presence of KCNE1, an auxiliary subunit of I_Ks channels, renders the channel insensitive to both activators. We found that ML277 effects are dependent on several residues in the KCNQ1 pore domain. Some of these residues are also necessary for R-L3 effects. These residues form a putative hydrophobic pocket located between two adjacent KCNQ1 subunits, where KCNE1 subunits are thought to dwell, thus providing an explanation for how KCNE1 renders the I_Ks channel insensitive to these activators. Our experiments showed that the effect of R-L3 on voltage sensor movement during channel deactivation was much more prominent than that of ML277. Simulations using a KCNQ1 kinetic model showed that the effects of ML277 and R-L3 could be reproduced through two different effects on channel gating: ML277 enhances KCNQ1 channel function through a pore-dependent and voltage sensor-independent mechanism, while R-L3 affects both channel pore and voltage sensor.

Keywords: KCNQ1; agonists; channel; long QT syndrome; voltage sensor.

FIGURE 1

KCNQ1 pore domain is the main drug site for ML277. Top panel) schematic structure of wt KCNQ1 (blue), WT KCNQ2 (red), and chimera channels C1 and C2. C1 consists of KCNQ1 S5-pore-S6 and KCNQ2 S1-4 and N, C-termini. C2 is mostly KCNQ1 except for the top of S5 and a small portion of the pore loop that was replaced by the KCNQ2 counterparts. Middle panel) representative current traces recorded at control condition (black) and with 10 uM ML277 (red). Bottom panel) summary of the effects of ML277 on current amplitude (grey bar) and deactivation kinetics as measured by t1/2 of tail current decay (black bar). The scale bar in the figure is a 1 s time bar.

FIGURE 2

Residues critical for ML277 binding. (A) Sequence alignment of KCNQ1 and KCNQ2 pore domains. The black underlined residues are located in the S5 and S6 transmembrane domains. The green underlined residues were excluded by chimera C2 and are unlikely drug sites for ML277. The residues in red boxes are not conserved between KCNQ1 and KCNQ2 and are potential sites critical for ML277 binding. (B) KCNQ1 residues L266, G272, V334, and F335 are responsible for ML277 effects. Current traces in the top row show the effect of ML277 on wt and the loss of effect on L266W, G272C, V334L, and F335I KCNQ1. The middle and lower rows show the effect of ML277 on other mutant channels. The black and red traces represent recording in control condition and with 10uM ML277, respectively. (C) Bar graph summarizing the effect of ML277 on current amplitude (grey bar) and deactivation kinetics (black bar) of all the mutants tested. The scale bar in panel B is a 1 s time bar.

FIGURE 3

Aromatic residues in a putative hydrophobic pocket are important for ML277 activity. (A) Y267, F335, and F340 are required for ML277 effect. Representative recordings of KCNQ1 Y267, F335, F339, and F340 mutated to non-aromatic residues (top panel) and aromatic residues (lower panels) before (black) and after (red) application of ML277. (B) Representative recordings of other aromatic residues in KCNQ1 mutated to non-aromatic residues before (black) and after (red) application of ML277. (C) Bar graph summarizing the effect of ML277 on current amplitude (grey bar) and deactivation kinetics (black bar) of all the mutants tested. The scale bar in the figure is a 1 s time bar.

FIGURE 4

R-L3 shares some common residues with ML277 for functional effects. Representative recordings of mutants of all eight residues necessary for ML277 effect before (black) and after (red) ML277 application (top panel). Representative recording of the same mutants before (black) and after (red) R-L3 application (lower panel). Common residues that abolish the effects of both ML277 and R-L3 are marked with a star (*). The scale bar in the figure is a 0.5 s time bar.

FIGURE 5

Voltage clamp fluorometry of the effects of ML277 and RL-3 on KCNQ1. (A–F) Representative current (top) and fluorescence (middle) recordings of KCNQ1 before and after (A,B) ML277 application or (D–E) R-L3 application. Voltage protocol shown in inset. (C,F) Summary of (C) ML277 and (F) R-L3 effects on the deactivation of the current (gray) and fluorescence (red). (G–J) G(V) (top) and F(V) (bottom) for KCNQ1 before (open symbols) and after (closed symbols) application of ML277 (G–H) or R-L3 (I–J).

FIGURE 6

Computational modeling the effects of ML277 and RL-3 on KCNQ1. A 12-state model with intermediate and activated states was used (A). ML277 was modeled as a pure modulator slowing pore opening with no effect on voltage sensing while RL-3 was modeled with only changes to deactivation of voltage sensing. Current, fluorescence, and time to half inactivation agree with experiments (B–G) and the voltage-dependent properties (H–K).

FIGURE 7

ML277 and R-L3 do not alter membrane expression. Scatter plots from Bungarotoxin (BTX₆₄₇) flow cytometry experiments showing the detection of BTX₆₄₇ (extracellular marker) and YFP (intracellular tag) where dots in the upper right quadrant represent channels that expressed and trafficked to the membrane. (A) Schematic diagram of KCNQ1 with BTX-Alexa 647 and YFP. For both RL-3 (B,C) and ML277 (D,E), there was no change in surface expression of channels between control condition (left panels) and after application of drug (right panels). (F) Shows the data as compressed onto the y-axis to further emphasize no significant change.