Discovery of a novel activator of KCNQ1-KCNE1 K channel complexes

Abstract

KCNQ1 voltage-gated K(+) channels (Kv7.1) associate with the family of five KCNE peptides to form complexes with diverse gating properties and pharmacological sensitivities. The varied gating properties of the different KCNQ1-KCNE complexes enables the same K(+) channel to function in both excitable and non excitable tissues. Small molecule activators would be valuable tools for dissecting the gating mechanisms of KCNQ1-KCNE complexes; however, there are very few known activators of KCNQ1 channels and most are ineffective on the physiologically relevant KCNQ1-KCNE complexes. Here we show that a simple boronic acid, phenylboronic acid (PBA), activates KCNQ1/KCNE1 complexes co-expressed in Xenopus oocytes at millimolar concentrations. PBA shifts the voltage sensitivity of KCNQ1 channel complexes to favor the open state at negative potentials. Analysis of different-sized charge carriers revealed that PBA also targets the permeation pathway of KCNQ1 channels. Activation by the boronic acid moiety has some specificity for the Kv7 family members (KCNQ1, KCNQ2/3, and KCNQ4) since PBA does not activate Shaker or hERG channels. Furthermore, the commercial availability of numerous PBA derivatives provides a large class of compounds to investigate the gating mechanisms of KCNQ1-KCNE complexes.

Figure 1. Modulation of Q1/E1 channels by boronates.

(A) Time course of Q1/E1 current measured in ND96 at +40 mV at the end of a 2 s pulse. The current was normalized before compound application. Borax reversibly inhibits while methylboronic acid has little to no effect on channel current. Phenylboronic acid (PBA) initially inhibits current and then slowly potentiates. Benzoic acid reversibly reduces Q1/E1 channel current whereas benzyl alcohol reversibly activates the channel complex. (B) Families of currents recorded before and during treatment with PBA or benzyl alcohol. Currents were elicited by 4 s step test potentials from −80 to +60 mV in 10 mV increments from a holding potential of −80 mV followed by a tail pulse to −30 mV. Dashed line indicates zero current. Scale bars represent 1 µA and 0.5 s. (C) Voltage-activation curves for Q1/E1 calculated from tail current analysis. Solid curves represent Boltzmann fits to the data. Data are presented as the mean±SEM (n = 10).

Figure 2. PBA activates Q1/E3 complexes.

(A) Time course of Q1/E3 current measured in ND96 at +40 mV at the end of a 2 s pulse. The current was normalized before PBA application. PBA initially inhibits then slowly potentiates channel current. (B) Families of currents recorded in high external potassium (50 mM) before and during treatment with PBA. Currents were elicited by 2 s test potentials from −100 to +60 mV in 20 mV increments. The holding and tail potentials were −80 mV. Dashed line indicates zero current. Scale bars represent 1 µA and 0.5 s. (C) Voltage-activation curves for Q1/E3 calculated from tail current analysis. Solid curves represent Boltzmann fits to the data. Data are presented as the mean±SEM (n = 10).

Figure 3. PBA activates all tested members of the KCNQ family.

Left panel: Time courses of current recorded in ND96 at +40 mV at the end of a 2 s pulse. The current was normalized before PBA application. PBA initially inhibits and then slowly potentiates Q1 and Q2/Q3 current. PBA only activates Q4 channels. Middle panel: Families of currents recorded before and during treatment with PBA. Currents were elicited by 4 s test potentials from −100 to +60 mV in 10 mV increments from a holding potential of −80 mV followed by a tail pulse to −30 mV. Dashed line indicates zero current. Scale bars represent 1 µA and 0.5 s. Right panel: Voltage-activation curves calculated from tail current analysis. Solid curves represent Boltzmann fits to the data. Data are presented as the mean±SEM (n = 4–6).

Figure 4. PBA inhibits other Kv channels.

(A) Time course of Shaker (inactivation removed) current measured in ND96 at +40 mV, 100 ms pulse when 10 mM PBA was applied through the bath solution. (B) Families of Shaker currents recorded before and during treatment with PBA. Currents were elicited by 100 ms step test potentials from −100 to +60 mV in 10 mV increments from a holding potential of −80 mV. Dashed line indicates zero current. Scale bars represent 1 µA and 0.1 s. (C) Time course of hERG current measured in ND96 at 0 mV, 2 s pulse when 10 mM PBA was applied through the bath solution. (D) Families of hERG currents recorded before and during treatment with PBA. Currents were elicited by 2 s step test potentials from −100 to +60 mV in 10 mV increments from a holding potential of −80 mV. Dashed line indicates zero current. Scale bars represent 1 µA and 0.5 s.

Figure 5. Activation of Q1 by PBA is dependent on the external charge carrier ion.

Current traces were recorded in (A) 50 mM K⁺, (B) Rb⁺, or (C) Cs⁺. Left panel: Representative overlaid traces elicited by a +40 mV test and −80 mV tail pulse before and after the onset of PBA potentiation. Inset: Normalized tail currents comparing the deactivation kinetics before and after the onset of PBA potentiation. Tick marks represent 200 ms. Right panel: Time course of current recorded during the PBA potentiation phase. Outward current measured at the end of a 2 s pulse to +40 mV; maximal inward current measured during the −80 mV tail pulse. Time zero is the amount of current after initial inhibition but before potentiation by PBA. Data are represented as the mean±SEM (n = 5–10).