KCNQs: Ligand- and Voltage-Gated Potassium Channels

Abstract

Voltage-gated potassium (Kv) channels in the KCNQ (Kv7) family are essential features of a broad range of excitable and non-excitable cell types and are found in organisms ranging from Hydra vulgaris to Homo sapiens. Although they are firmly in the superfamily of S4 domain-bearing voltage-sensing ion channels, KCNQ channels are highly sensitive to a range of endogenous and exogenous small molecules that act directly on the pore, the voltage-sensing domain, or the interface between the two. The focus of this review is regulation of KCNQs by direct binding of neurotransmitters and metabolites from both animals and plants and the role of the latter in the effects of plants consumed for food and as traditional folk medicines. The conceptual question arises: Are KCNQs voltage-gated channels that are also sensitive to ligands or ligand-gated channels that are also sensitive to voltage?

Keywords: GABA; KCNE; KCNQ2; KCNQ3; KCNQ5; epilepsy; herbal medicine; hypertension.

FIGURE 1

KCNQ channel architecture. (A) KCNQ1–KCNQ3 chimeric structural model based on cryo-EM structure of Xenopus KCNQ1 (Sun and MacKinnon, 2017). (B) Schematic showing membrane topology; domain coloring as above. VSD, voltage sensing domain; PH, pore helix. (C) Section through KCNQ1–KCNE complex showing main features. (D) Cartoon of voltage gating depicting the S4–S6 portions of a Kv α subunit monomer and the types of conformational shifts that may occur upon membrane depolarization.

FIGURE 2

KCNQ residues influential in effects of PIP₂, GABA, and metabolites. (A) KCNQ1–KCNQ3 chimeric structural model showing residues important for effects of PIP₂ (yellow) and GABA and metabolites (red). PH, pore helix. (B) Close-up of residues in A with human KCNQ3 residue numbering.

FIGURE 3

Predicted structure of KCNQ1-KCNE1 complex. KCNQ1–KCNE1 structural model showing predicted position of KCNE1 (blue between two KCNQ1 subunits (sand and salmon) and close to S5 (yellow) and R243 (red) or one subunit. Plotted from model coordinates made available in Kang et al. (2008).

FIGURE 4

Structures and calculated electrostatic surface potentials of KCNQ activators. Structures and calculated electrostatic surface potentials of KCNQ activators as indicated; red = negative, blue = positive extremes. Plotted using Jmol.

FIGURE 5

Conservation of the KCNQ5 S5 Trp. (A) Sequence alignment of the GABA-binding S5 Trp-containing portion of KCNQ5 or the closest ortholog in various organisms. Adapted with permission from Manville et al., 2018. (B) Evolutionary relationships of organisms showing representation of KCNQ channels in major animal clades. Red, organisms contain multiple KCNQs, including one or more with the GABA-binding S5 Trp; blue, one KCNQ, lacking the GABA-binding S5 Trp; gray, no KCNQ. From Manville et al. (2018) used with permission.

FIGURE 6

Predicted KCNQ GABA binding site and accessibility. (A) SwissDock prediction of GABA binding pose in KCNQ3. VSD, voltage sending domain. (B) Extracellular view of KCNQ3-W265 (red) showing possibility access site from outside the cell. Purple, S5; blue, VSD, pink, S4, S5 linker; green pore helix. From Manville et al. (2018) used with permission. (C) Similar view as in panel B, but in the Kv1.2–Kv2.1 “paddle-chimera” structure (Long et al., 2007) showing S5 accessibility (*) despite lipid presence (lipids shown in gray/red). Adapted from Manville et al. (2018) with permission.

FIGURE 7

Molecular basis for cilantro anticonvulsant effects. (A) Images of cilantro (left) and culantro (right). Required photo permissions: https://creativecommons.org/licenses/by/2.0/; https://commons.wikimedia.org/wiki/File:Culantro_(Eryngium_foetidum)_6.jpg. (B) Left, structures and Jmol surface electrostatic potential plots of three structurally similar fatty aldehydes; right, their corresponding effects on KCNQ2/3 G/V relationships (quantified from tail currents). Adapted from Manville and Abbott (2019a) with permission. (C) SwissDock prediction of E-2-dodecenal binding poses in KCNQ2. Blue, VSD; sand, S5; pink, S4, S5 linker.

FIGURE 8

Molecular basis for Mallotus oppostitifolius anticonvulsant effects. Images from or adapted with permission from De Silva et al. (2018) and Manville and Abbott (2018). (A) Image of Malloutus oppositifolius. (B) Structures and Jmol surface electrostatic potential plots of three KCNQ-activating compounds from Malloutus oppositifolius. (C) Predicted binding sites of CPT1 and MTX within KCNQ1. VSD, voltage sensor. (D) Upper, predicted overlap of CPT1 and MTX binding sites that would prevent both occupying a single binding pocket in KCNQ1. Lower, Effects of CPT1 and MTX (100 μM) alone or together quantified as KCNQ1 current fold-change versus voltage, showing lack of synergy or summation. (E) Predicted binding site of isovaleric acid (IVA) within KCNQ2/3. Red, KCNQ2–W236, or KCNQ3–W265. (F) Upper, predicted neighboring binding poses of IVA and MTX binding sites that could allow both to occupy a single binding pocket in KCNQ2 or 3. Lower, Effects of IVA and MTX alone or together quantified as KCNQ2/3 current fold-change versus voltage, showing synergistic effects.

FIGURE 9

KCNQ5 activation by botanical hypotensive folk medicines. Images from or adapted with permission from Manville et al. (2019). (A) Approximate timeline of human use of some of the hypotensive plants discovered to activate KCNQ5. YA, years ago. Required photo permissions: https://creativecommons.org/licenses/by-sa/3.0/; https://en.wikipedia.org/wiki/Chamomile#/media/File:Kamomillasaunio_(Matricaria_recutita).JPG; https://creativecommons.org/publicdomain/zero/1.0/Fennel seed; https://en.wikipedia.org/wiki/Fennel#/media/File:Fennel_seed.jpg; https://creativecommons.org/licenses/by-sa/3.0/Sophora flavescens; https://commons.wikimedia.org/wiki/File:Sophora_flavescens.jpg; https://en.wikipedia.org/wiki/Homo_erectus#/media/File:Homo_Georgicus_IMG_2921.JPG; https://creativecommons.org/licenses/by/2.5/; https://en.wikipedia.org/wiki/Neanderthal#/media/File:Homo_sapiens_neanderthalensis.jpg. (B) Dried Sophora flavescens root slices. (C) Chemical structures of S. flavescens compounds showing activity for KCNQ5; lower left, surface electrostatic potential plot of aloperine, the most KCNQ5-active compound. (D) Predicted binding site of aloperine within KCNQ5.

FIGURE 10

Models of GABA and PIP₂ signaling in KCNQ and KCNQ2/3–SMIT1 complexes. (A) Cartoon showing lack of communication between the KCNQ VSD and pore in the absence of PIP₂. (B) Cartoon showing voltage-dependent activation of KCNQs in the presence of PIP2. (C) Cartoon showing hypothetical GABA-induced KCNQ activation in the absence of PIP₂. (D) Cartoon showing hypothetical GABA-induced KCNQ activation in the presence of PIP₂. (E) Cartoon of KCNQ2/3–SMIT1 interactions showing the functional consequences of physical interaction and modulation by small molecules. DAG, diacylglycerol; G6P, glucose-6-phosphate; IMPase, inositol monophosphatase; IP, inositol phosphate; inositol polyphosphatase 1-phosphatase; MI, myo-inositol; PI4K, phosphatidylinositol 4-kinase; PI5K, phosphatidylinositol-5 kinase; PLC, phospholipase C. From Manville and Abbott (2020b) with permission.