KCNQ1 voltage-gated K+ channels are involved in a wide variety of fundamental physiological processes and exhibit the unique feature of being markedly inhibited by external K+. Despite the potential role of this regulatory mechanism in distinct physiological and pathological processes, its exact underpinnings are not well understood. In this study, using extensive mutagenesis, molecular dynamics simulations, and single-channel recordings, we delineate the molecular mechanism of KCNQ1 modulation by external K+. First, we demonstrate the involvement of the selectivity filter in the external K+ sensitivity of the channel. Then, we show that external K+ binds to the vacant outermost ion coordination site of the selectivity filter inducing a diminution in the unitary conductance of the channel. The larger reduction in the unitary conductance compared to whole-cell currents suggests an additional modulatory effect of external K+ on the channel. Further, we show that the external K+ sensitivity of the heteromeric KCNQ1/KCNE complexes depends on the type of associated KCNE subunits.

KCNQ1 K+ channels are expressed in several human excitable and epithelial tissues (Chouabe et al., 1997) where they play a key role in the regulation of cellular excitability and transepithelial ion transport (Jespersen et al., 2005; Abbott, 2014). Many mutations of the KCNQ1 gene are associated with diseases of the human heart (Wang et al., 1996; Duggal et al., 1998; Peroz et al., 2008; Hedley et al., 2009), inner ear (Chouabe et al., 1997), endocrine system (Torekov et al., 2014; Tommiska et al., 2017), and brain (Goldman et al., 2009). Despite the enormous therapeutic significance of these channels, our understanding of their subunit composition, biophysical and pharmacological features, as well as the mechanisms of their regulation remains incomplete.

KCNQ1 channels are multisubunit complexes composed of pore-forming α subunits arranged in a symmetrical fourfold structure in characteristic domain-swapped configuration (Sun and MacKinnon, 2020) and KCNE1–5 β subunits. The cellular Ca2+-sensor calmodulin is an integral constituent of native and heterologously expressed KCNQ1 channels (Ghosh et al., 2006; Sachyani et al., 2014; Sun and MacKinnon, 2020). KCNE proteins are thought to dock between voltage-sensor domains and the pore domain of α subunits (Sun and MacKinnon, 2020) whereas calmodulin molecules interact with the channel from the cytoplasmatic side of the cell membrane (Shamgar et al., 2006). Calmodulin is strictly required for proper channel assembly and trafficking (Ghosh et al., 2006; Shamgar et al., 2006). Phosphatidylinositol-4,5-bisphosphate (PIP2)-signaling lipid binds to KCNQ1 complexes as well (Sun and MacKinnon, 2020) and plays an essential role in channel gating (Loussouarn et al., 2003). The biophysical and pharmacology properties of KCNQ1 complexes vary quite considerably depending on the type (Barhanin et al., 1996; Jespersen et al., 2005) of associated KCNE protein and likely on their quantity as recent studies on KCNE1 subunit indicate (Wang et al., 2020). Intracellular second messengers such as free Ca2+ ions (Tohse, 1990; Shamgar et al., 2006) and adenosine nucleotides (Li et al., 2013; Kienitz and Vladimirova, 2015), as well as membranes lipids (Loussouarn et al., 2003; Zaydman et al., 2013; Liin et al., 2015b) and protein kinase A (Yazawa and Kameyama, 1990) exert a pronounced modulatory action on these channels.

A substantial inhibitory effect of external K+ (K+o) on KCNQ1 is a particularly intriguing feature of these channels (Larsen et al., 2011), reported for the first time nearly two decades ago (Yang and Sigworth, 1998). Such regulation may have a central role in the modulation of the endocochlear potential and K+ recycling in the inner ear, given the strong expression of these channels in strial marginal cells (MC; Knipper et al., 2006; Wang et al., 2015) and dark cells of the vestibular organ (Marcus and Shen, 1994). K+o dependency on the KCNQ1 channels can have a significant impact on the physiology of gastrointestinal organs where these channels are abundantly expressed (Chouabe et al., 1997; Liin et al., 2015a). For instance, KCNQ1/KCNE3 heteromeric channels are localized in the basolateral membranes of epithelial cells of the small intestine (Preston et al., 2010). The concentration of the luminal K+ in the small intestine increases several hours after consumption of food, which triggers passive absorption of K+ into the bloodstream through a paracellular pathway (Agarwal et al., 1994). During this process, the epithelial cells will be exposed to elevated [K+]o. Given the well-recognized stabilizing influence of external K+o on the conductive properties of most K+ channels, understanding the molecular mechanism of the inhibitory action of K+o on KCNQ1 has the potential to provide new insights into basic principles of K+ channel permeability and gating.

The exact molecular mechanism of how K+o modulates KCMQ1 channels is currently unclear. It has been proposed that the augmentation of channel’s fast inactivation underlies the inhibitory action of K+o (Larsen et al., 2011). However, a significant K+o-dependent inhibition of the non-inactivating KCNQ1/KCNE1 complex expressed in Xenopus laevis oocytes (Yang and Sigworth, 1998) and CHO cells (Wang et al., 2015) has also been reported. The negative charge of the E290 residue was proposed as an essential element of the K+o sensitivity (Wang et al., 2015), but the corresponding mechanism is not clear.

Here, we have explored the K+o dependency of KCNQ1 channels in more detail and provide a molecular basis underlying this phenomenon. We first show that the fast inactivation of the channel, as well as the negatively charged residues in the turret region of the pore, plays no principal role. An extensive mutational analysis and subsequent molecular dynamics (MD) simulations revealed the involvement of the selectivity filter (SF) in this process. MD simulations also demonstrate that the enhanced occupancy of the uppermost K+ binding site (S0) of the SF operating in a specific conducting mode triggers the modulation of the channel conductance. In accordance with these findings, a marked reduction in unitary conductance at elevated K+o conditions was observed in single-channel recordings. Furthermore, we show that K+o sensitivity of distinct KCNQ1 complexes varies considerably depending on the type of associated KCNE subunit.

Constructs and mutagenesis

Mutations in the human KCNQ1 gene, cloned into oocyte expression vector pGEM-He-Juell, were introduced by overlap PCR method as well as using the QuickChange kit (Stratagene). Each mutation was verified by DNA sequencing. mRNA synthesis was made with the T7 mMessage mMachine transcription kit (Ambion). mRNA was then analyzed in gel electrophoresis and quantified by RNA 6000 Nano kit (Agilent Technologies) or spectroscopic method. The EQQ construct was generated as previously described (Murray et al., 2016). It was subcloned into pGem He-Juell vector using HindIII and XbaI restriction sites. All new constructs were confirmed by sequencing.

Channel expression

Xenopus frogs (Nasco) were kept in the animal facility of the Institute of Molecular Biology National Academy of Sciences of the Republic of Armenia according to the guidelines of the local animal welfare authorities. Oocytes were removed surgically and defolliculated with 2–3 mg/ml collagenase (Roche) in Ca2+-free OR2 solution containing 82.5 mM NaCl, 2 mM KCl, 1 mM MgCl2, and 5 mM HEPES, pH 7.5. Stage IV or V oocytes were selected and injected with 2–50 ng cRNA on the next day. Injections were performed using an oocyte manual microinjection pipette (Drumond Scientific Company). Mixtures of KCNQ1/KCNE1, KCNQ1/KCNE2, and KCNQ1/KCNE3 cRNAs with different molar ratios of α to β were made prior to injection. Electrophysiological experiments were performed 2–7 d after injection. Oocytes were incubated in OR2 solution supplemented with 2 mM CaCl2, 5 mM sodium pyruvate, and 50 mg/ml gentamycin at 17°C conditions prior to electrophysiological recording.

For single-channel recordings, ltk-mouse fibroblast cells (LM) were cultured and plated for experiments as previously described (Murray et al., 2016). Cells were transfected using Lipofectamine2000 (Thermo Fisher Scientific) as per the manufacturer’s protocol. The EQQ constructs were transfected with GFP in a 2:0.7 ratio (in micrograms). F339A was GFP tagged and transfected alone at 1.5 μg/dish of cells.

Electrophysiological recordings

TEVC recordings were performed at room temperature using Tec-03X or Turbo TEC-05 amplifiers (NPI Electronics) linked to Patchmaster software (HEKA Electronics) using a Instrutec digitizer for data acquisition and monitoring. Some experiments were conducted in a setup based on Axon 500B amplifier (Molecular Devices) linked to Patchmaster software (HEKA Electronics) through ITC-16 Computer Interface (Port). Borosilicate glass capillaries (WPI) were pulled to fabricate pipettes in PB/10 puller (Narishige). Tips of the recording electrodes were prefilled with 1% agar-3M KCl solution and backfilled with 3M KCl to prevent KCl leakage into the oocytes. Pipette resistances were 0.1–0.3 MΩ. Leakage current subtraction was not used due to the voltage-independent gating component in KCNQ1 channels. Recordings were performed in ND96 solutions containing (in mM) 96 NaCl, 2 KCl, 1.8 CaCl2, 1 MgCl2, and 5 HEPES, pH 7.5. Different K+-containing solutions were obtained via substitution of NaCl with KCl. 0.2 mM K+ solution contained (in mM) 0.2 KCl, 95.8 NaCl, 1.8 CaCl2, 1 MgCl2, and 5 HEPES, pH 7.5; 100 mM Na+ solution (nominally K+ free) contained 100 NaCl, 1.8 CaCl2, 1 MgCl2, and 5 HEPES, pH 7.5. In the latter solution, 100 mM Na+ was substituted by 100 mM K+, Rb+, or NMDG+ to get the corresponding solutions. All chemicals were from Sigma unless otherwise stated. Oocytes were recorded in a chamber with a volume of 0.35 ml. A five-channel perfusion system was used to exchange the extracellular solution. Oocyte perfusion at 2.5 ml/min flow rate was kept constant in all experiments. Measurements of the oocytes for determination of the concentration dependency of KCNQ1 channels were performed according to the following procedure. Oocytes expressing wild type or mutant channels were placed from the incubation solution (OR2 supplemented with Ca2+ [2 mM], Na-pyruvate [5 mM], and gentamicin [50 mg/ml]) into the recording chamber perfused with 0.2 mM K+ containing ND96 solution. After clamping the oocytes at holding potential (usually −100 mV), continuous depolarization pulses with +60 mV amplitude and 120 s interpulse intervals were applied. KCNQ1 current amplitudes in this solution usually stabilized after four to six consequent pulses, after which the values of the next three points were taken for 0.2 mM K+o. Then, the perfusion was switched to 2 mM K+ containing solution and the next three points were recorded. The experiments were continued according to this scheme until the last three points corresponding to 100 mM K+o were recorded. Recordings that, for some reason, did not reach 100 mM K+ perfusion step were not analyzed. To estimate the unspecific currents, the water-injected oocytes from the parallel batch were recorded in an identical way. Recordings of these oocytes were done on the same day using the same solutions and by application of identical protocols and solution exchange schemes. Then, we averaged the unspecific currents values corresponding to three to four water-injected oocytes taken at specific cursor positions, which were identical to that of the analysis of wild type and mutant channels. Subsequently, these unspecific current values were subtracted from three data points for each K+o concentration, and the data was plotted against time. The concentration–effect relationships were determined via averaging three points of the given concentration and correcting them for theoretical values, if not otherwise specified. Data points were then fitted to the Hill equation using Igor Pro software (WaveMetrics).

Single-channel recordings were acquired as previously published (Werry et al., 2013). Acquisitions were made using an Axopatch 200B amplifier, Digidata 1440A, and pClamp 10 software (Molecular Devices). Records were low-pass filtered at 2 kHz at acquisition using a 3 dB, four-pole Bessel filter, sampled at 10 kHz, and digitally filtered at 200 Hz before analysis. For single-channel recordings, the bath solution contained (in mM) 135 KCl, 1 MgCl2, 0.1 CaCl2, 10 dextrose, and 10 HEPES (pH 7.4 with KOH). The pipette solution contained (in mM) 6 NaCl, 129 MES, 1 MgCl2, 5 KCl, 1 CaCl2, and 10 HEPES (pH 7.4 with NaOH) or 97 KCl, 41.3 MES, 1 MgCl2, 1.9 NaCl, and 10 HEPES. Electrodes were pulled from thick-walled borosilicate glass (Sutter Instrument) using the linear multistage electrode puller (Sutter Instruments). Electrodes were coated with Sylgard (Dow Corning). After fire polishing, single-channel electrode resistances were between 40 and 60 MΩ. The n values refer to the number of individual cells recorded via TEVC or patch clamp methods.

Computational electrophysiology simulations

CompEL (computational electrophysiology) method (Kutzner et al., 2011) with dual membrane configuration implemented in GROMACS was used to investigate ion permeations through the pore region of the KCNQ1 channel. Permeation of ions was driven by a transmembrane voltage, which was generated by charge imbalance (2 Cl ions) between compartments. The open-state cryo-EM structure of KCNQ1 (PDB ID: 6V01; Sun and MacKinnon, 2020) with removed KCNE3 was used in simulations. To keep the intracellular gate in the open state, terminal amino acids of the S5 and S6 helices were restrained. The protein membrane system was assembled via CHARMM-GUI and embedded into a POPC lipid bilayer surrounded by water and K+/Na+/Cl ions. Systems were minimized and equilibrated gradually releasing position restraints ending with 10 ns of NTP ensemble equilibrations. N values in the text refer to the number of separate MD simulations.

Data analysis

Data were analyzed using Fitmaster (HEKA Electronics), Kaleidograph (Synergy Software), and Igor pro (Wavemetrics) software. Inactivation of wild type and mutant KCNQ1 channels was determined by fitting tail currents with two or three exponential functions with subsequent extrapolation of the fit to the beginning of the segment. Calculation of ratios was then performed as described in Fig. S1, adopted from early publication (Tristani-Firouzi and Sanguinetti, 1998). Voltage-dependent activation of T312C was determined from tail currents after their correction for inactivation. Data were then fitted to a Boltzmann function with offset as previously described (Ma et al., 2011). Concentration–response data were determined as follows: peak currents were determined from the entire depolarizing segment (+60 mV) for each external K+ concentration. Theoretical values were then subtracted assuming K+in = 108 mM (Weber, 1999). Theoretical reduction of the current amplitude and absolute single-channel permeation PK was calculated using Goldman–Hodgkin–Katz (GHK) flux equation of the form Ik = PK(VmF2/RT)[Ki − Koexp(−VmF/RT)/(1 − exp(−VmF/RT)], where PK is the absolute single-channel permeability; Vm is the membrane potential; Ki and Ko are K+ concentrations in inside and outside solutions, respectively; Vm is the membrane voltage; and F, R, and T (293 K) have their usual meanings. Assuming that channels are not modified by extracellular K+, the ratio of current reduction in two external K+ concentrations K+o(C1), K+o(C2), is given by the equation: Ic1/Ic2 = [K+in(C1) − K+o(C1)exp(−VmF/RT]/[K+in(C2) − K+o(C2)exp(−VmF/RT)], where K+o(C1), K+o(C2), K+in(C1), and K+in(C2) are the outside and inside K+ concentrations, accordingly (Larsen et al., 2011). Data corrected for theoretical values were then fitted with a Hill function of the form I(C) = (1 − Imax) + (Imax/(1 + [C/IC50]nH), where Imax is the maximal inhibition, IC50 is the concentration of half-maximal inhibition, nH is the Hill coefficient, and C is the concentration. Permeability ratios were calculated as PX/PK = exp(ΔErvZF/RT), where ΔErv is the mean difference of reversal potentials before and after exchange of extracellular cation, and Z, F, R, and T have their usual meanings. Exact P values are provided in Table S2.

To determine the conductance of EQQ, all-point histograms were generated from records containing the activity of a single channel at various voltages. Gaussian fits of the all-points histograms were made using Clampfit and the largest peak at each voltage was used to generate the conductance curves. Linear regression fitting and statistical analysis of the conductance curves were carried out using Prism 8 and their version of the analysis of covariance (ANCOVA; GraphPad Software).

Online supplemental material

Fig. S1 illustrates the methodology used to estimate the fractional fast inactivation of KCNQ1 channels. Fig. S2 shows voltage dependency as well as the time dependency of the fast inactivation for wild type and mutant homomeric channels. Fig. S3 presents the magnitude of inhibition of KCNQ1 channels exhibiting various degrees of fractional fast inactivation. Fig. S4 shows exemplary current traces of alanine mutants in the pore region measured at 0.2 and 100 mM external K+ conditions. Fig. S5 demonstrates sample current traces of the T312C selectivity-filter mutant and its conductance–voltage relationship. Fig. S6 schematically illustrates the molecular system for the simulation of permeation through the KCNQ1 pore. It also shows the average transmembrane voltage in such a system during simulations. Fig. S7 demonstrates the stability of the protein in this molecular system during the simulation both in symmetrical and asymmetric external K+ concentrations. Fig. S8 shows typical K+ occupancy substates of the selectivity filter, as well as the distribution of these substates in simulations at 5 and 150 mM external K+ conditions. Typical ion movements through the channel pore are illustrated in this figure as well. Fig. S9 shows the interchange of the conductive modes of the selectivity filter during 500 ns simulation at two different K+o concentrations. Fig. S10 demonstrates the existence of two main configurations for the aromatic rings of F339 and F340 residues and the statistical distribution among these configurations during simulations. Fig. S11 demonstrates the independence of the potentiating effect K+o on the F339A mutant from its slow inactivation phenotype. Videos 1 and 2 visualize the permeations occurring according to the canonical and spontaneous S0 modes, respectively. Video 3 shows the delay of forward K+ transmissions in the selectivity filter of the channel at high external K+ conditions. In Video 4, the flipping of aromatic rings of F339 and F340 residues is visualized.

K+o-sensitivity of KCNQ1 mutants with markedly reduced inactivation

We initially investigated the influence of different [K+]o on outward current mediated by homomeric KCNQ1. In agreement with previous studies (Yang and Sigworth, 1998; Larsen et al., 2011; Wang et al., 2015), we observed a significant reduction in KCNQ1-mediated current at +60 mV in the range of [K+]o between 0.2 and 100 mM (Fig. 1). Fitting a Hill equation to GHK-corrected data (Fig. 1 E) resulted in inhibition parameters (Table 1) comparable to what was reported in earlier studies in Xenopus oocytes and CHO cells (Larsen et al., 2011; Wang et al., 2015). Activated homomeric KCNQ1 channels undergo a fast inactivation process also known as hook inactivation (Sanguinetti et al., 1996; Pusch et al., 1998; Tristani-Firouzi and Sanguinetti, 1998). The name originates from the hook-like shaped tail current at repolarization potentials (−100 mV in Fig. 1 A), which reflects two simultaneously ongoing processes: (1) the recovery of channels from inactivation and (2) channel deactivation. We next tested the hypothesis of whether K+o sensitivity of KCNQ1 is associated with the fast inactivation by investigating the F351A mutant previously shown to exhibit considerable deficiency in inactivation (Hou et al., 2017) and L271A that is located in the S5 segment and markedly reduces the inactivation as well (unpublished data). As an initial step, we quantified inactivation in these mutants using a method (Tristani-Firouzi and Sanguinetti, 1998) described in detail in Fig. S1 depending on (1) the voltage of depolarization (Fig. 1, A and B) and (2) the duration of activating pulses (Fig. S2, C and D). The results show that the voltage- and time-dependent inactivation of L271A and F351A channels is strongly suppressed (Fig. 1, A and B; and Fig. S2 D). Analysis of K+o dependency in these mutants revealed that they are inhibited by K+o to an extent comparable with the wild type (Fig. 1, C–E; and Table 1). We also studied several other mutants showing different degrees of inactivation (Fig. S2, Fig. S3, and Table 1), including I274A, which exhibits enhanced inactivation, and performed correlation analysis represented in Fig. 1 F. Results clearly demonstrate that K+o-dependent inhibition of KCNQ1 is not correlated with the extent of fast inactivation.

Negative charge in the turret region plays no principal role

The recently resolved atomic structure of Xenopus KCNQ1 revealed a unique negatively charged extracellular protein surface (Fig. 2 A) surrounding the pore entrance of the channel (Sun and MacKinnon, 2017). Calculation of the electrostatic potential based on recent cryo-EM structures of KCNQ1 channels (Sun and MacKinnon, 2017, 2020) shows less electronegativity in the human ortholog (Fig. 2 A). Nevertheless, negatively charged side chains forming the surface of human KCNQ1, hypothetically, could account for K+o sensitivity by attracting K+ from the external medium. Such a postulation is in accordance with an earlier report showing a reduction in K+o sensitivity by mutation of the E290 residue (Wang et al., 2015). To understand more about the impact of an external negative potential on the inhibitory effects of K+o, we first generated a mutant in which all amino acids involved in the formation of the electronegative surface (E290, S291, E295, and D317) were exchanged for alanine. KCNQ1-specific current could be recorded neither from Xenopus oocytes injected with the corresponding mRNA nor from transfected HEK 293T cells (data not shown). The aspartic acid at the 317 position is strongly conserved among K+ channels (Fig. 2 A) suggesting that this residue is unlikely to be responsible for K+o sensitivity of the channel. Thus, we next investigated the E290A/S291A/E295A triple mutant. Marked reduction of current expression hampered the analysis of the concentration dependency in this channel. Nevertheless, the comparison of current amplitudes at 0.2 mM and 100 mM K+o concentrations demonstrated that K+o strongly inhibits the channel, the extent of which was even slightly (∼8%) higher than that of the wild type (Fig. 2 B). We next assessed the K+o sensitivity of corresponding single alanine mutants as well. The extent of inhibition of S291A and E295A mutants by external K+ was very similar to that of the triple mutant when current amplitudes at 0.2 mM and 100 mM K+o are compared (Fig. 2, C–E). The E290A mutation induced about 34% reduction in K+o-dependent inhibition relative to wild type (Fig. 2, C–E; and Table 1), which was consistent with an earlier study conducted in CHO cells with E290C mutant (Wang et al., 2015). We next introduced a positive charge at the E290 position by exchanging glutamic acid for arginine. The resultant channel exhibited marginally (∼6%) increased sensitivity to external K+ (Fig. 2, C and F). Neutralization of the E290 side chain by E290Q mutation, which preserves the side chain volume, produced a channel with K+o sensitivity similar to E290R, additionally causing a small leftward shift (ΔIC50 = 3.64 ± 1.96 mM) in the concentration–response curve (Fig. 2, C and F; and Table 1). Proline substitution of the E290 position created a channel with K+o sensitivity very similar to wild type (Fig. 2, C and F; and Table 1). These results suggest that the negative charge at the E290 position is not necessary for K+o sensitivity. The side-chain volume at the E290 position seems to be positively correlated with the extent of K+o-induced inhibition (Fig. 2 G). Remarkably, the E290W mutant with the largest possible side chain volume did not produce functional channels (data not shown). Xenopus oocytes injected with mRNA encoding the D317A mutant produced small currents that were not suitable for analysis, and proline, glycine, or tryptophan substitutions at this position did not result in functional expression. Altogether, these results indicate that the electronegative pore surface of KCNQ1 is not the cause of the K+o sensitivity of the channel. A slightly increased K+o sensitivity in S291A, E295A, E290R, E290Q, and E290A/S291A/E295A mutants as well as its decrease in E290A mutant is consistent with the general picture of the broad mutagenesis study described below, indicating that statistically significant changes in K+o sensitivity can be induced by mutations located in various regions of the channel pore.

Alanine scanning mutagenesis of the pore domain

We next employed alanine scanning mutagenesis to map residues that are essential for the K+o sensitivity of KCNQ1. Since no considerable voltage-dependency is observed for the K+-induced KCNQ1 inhibitory process (Larsen et al., 2011), we studied the pore residues spanning the I268–V355 segment with a simplified approach that compares current at 0.2 and 100 mM K+o conditions (Fig. 3, A–C; and Fig. S4). This approach is reasonable due to our finding that IC50 values are not considerably changed for most of the mutants studied (Table 1). For more than half of the alanine mutants, only minor alterations were observed compared with the wild type (Fig. 3 C). 17 mutations induced statistically significant changes in K+o sensitivity to variable extents. Of those, V310A, F339A, and F340A exhibited a noticeable current potentiation by high K+o (Fig. 3, A–C; and Fig. S4). Mapping mutations that induced statistically significant changes onto the recent KCNQ1 structure revealed that the moderate changes are inducible by mutations located in different parts of the channel pore (Fig. 3 D). Stronger alterations in K+o sensitivity are induced by mutations of residues that are buried deeply in the hydrophobic part of the channel that is not accessible to external K+ (Sun and MacKinnon, 2020). Congregation of these residues below the SF (Fig. 3 D) pointed toward the potential involvement of the SF in the modulation process.

External permeant ions induce current modulation

One possible explanation for the involvement of the SF is that the enhanced binding of the K+o to its extracellularly located site may trigger functional modulations in SF leading to current inhibition. To ascertain that the inhibition is caused by enhanced binding of K+o to channel but not due to the reduction of [Na+]o necessary to keep the ionic strength of the external medium, we next studied the comparative effect of the external permeant (Rb+, K+) versus non-permeant (Na+, NMDG+) cations in representative mutant channels. We selected one channel in each mutant category—F351A, G269A, and F340A—that exhibited wild type-like, a weaker-, and a “reverse” K+o sensitivity, respectively, to conduct these experiments. If inhibition is triggered by the binding of permeant ions to SF, these experiments had the potential to provide insights into the mechanism of the inhibition process since K+ channels exhibit differential functional properties depending on whether K+ or Rb+ is bound to their SFs.

The inhibition of wild type KCNQ1 by 100 mM external K+ was mimicked by equimolar external Rb+ (Fig. 4, A and B). We also detected a minor alteration in the shape of the outward current trace (Fig. 4 A) and a tendency for current to increase during recordings in external Rb+ (Fig. 4 B). Such a trend was not observed at high K+o conditions even after prolonged K+o exposure (Fig. 3 B). We speculate that this effect could be related to the gradual elevation of intracellular Rb+ in oocytes due to large inward-holding currents. The continuously increasing intracellular Rb+ then would lead to a larger outward current due to higher Rb+ conductance of KCNQ1 compared with K+ (Pusch et al., 2000). Substitution of external Rb+ with the impermeable organic cation NMDG+ having a much larger ionic diameter (∼7.3 Å) than Na+ (1.9 Å) eliminated inward currents and re-established amplitudes observed at initial Na+ conditions (Fig. 4, A and B). Rb+o also mimicked the effect of K+o on the F351A and G269A mutants with subsequent restoration of current amplitudes observed at starting conditions when NMDG+ was applied (Fig. S4, A and B). The G269A mutant did not show a similar trend of increase in outward currents in the external 100 mM Rb+ condition despite the holding current in this mutant being comparable with the wild type (Fig. 4 A). This can be explained by a lower Rb+/K+ conduction ratio of the G269A mutant (see tails in Fig. 4 A). The potentiation of the F340A channel by high K+o (Fig. 3, A–C) was slightly augmented by application of external Rb+ (Fig. 4 C). Strikingly, in this case, the Rb+o substitution by NMDG+ did not restore the current amplitudes observed in the initial Na+o condition (Fig. 4 C). We did not study this phenomenon in further detail. However, experiments with the application of external NMDG+ after Na+o clearly demonstrated that NMDG+ itself has no influence on F340A (Fig. 4 D). A slight tendency toward increased inhibition in G269A mutant by Rb+o compared with K+o was not statistically significant (Fig. 4 B). In the F351A mutant, the inhibition of outward current was slightly more pronounced when Rb+o was applied. This prompted us to investigate the concentration dependency under external Rb+ conditions for this mutant. The F351A channel has a very small constitutively open fraction allowing for prolonged recordings at negative potentials (Fig. 4, E and F). The results show that external Rb+ is a more effective inhibitor of the outward current than K+o (Fig. 4 G and Table 1). Altogether, these results suggest that the modulation of KCNQ1 is induced by permeant ions applied at the extracellular side of the membrane. Observed alterations in the inhibitory effect of Rb+o compared with K+o most likely reflect the difference in the binding properties of these cations on the external modulatory site of the channel.

Loss of selectivity and external K+ sensitivity in the T312C mutant

To further investigate the involvement of the SF in the process of K+o dependency of KCNQ1, we searched for mutation(s) that might induce noticeable alterations in SF function. Although our previous study suggested that most of the mutations in the K+-selective motif or its proximity render the KCNQ1 channel nonfunctional (Ma et al., 2011), we found that the T312C mutant retained functional expression and the voltage-gating properties of KCNQ1 (Fig. S5). Membrane potentials near to 0 mV together with large inward tail currents in external K+-free solution indicated that the T312C mutation affects the ion selectivity of KCNQ1 (Fig. 5, A and B). Analysis of the relative permeability via estimation of reversal potential for external Na+ and K+ conditions revealed a dramatic drop of K+ selectivity in T312C (PNa+/PK+ = 0.77 ± 0.06, n = 9, ±SEM vs. 0.016 for Iks; Matsuura et al., 1987). In addition, we observed a significant increase in relative Rb+ permeability (PRb+/PK+ = 0.93 ± 0.02 in T312C, (n = 9) vs. 0.75 ± 0.01 in WT, n = 5, P = 0.0076, Student’s t test; ±SEM). More remarkably, no considerable inhibition of outward current in T312C was observed upon replacement of Na+o with equimolar K+o or Rb+o (Fig. 5, C and D). Hence, the T312C mutation induces a loss of channel selectivity—a feature reflecting functional modifications in the SF—which also eliminates K+o sensitivity of the channel. This confirmed our hypothesis that the K+o dependency of KCNQ1 is attributable to the SF of the channel.

Single-channel properties at high external K+

To explore the K+o dependency of KCNQ1 at the single-channel level, we next recorded EQQ tandem channels expressed in ltk-mouse fibroblast cells. EQQ was constructed by sequential linkage of one KCNE1 auxiliary subunit to two KCNQ1 pore-forming subunits (Fig. 6 A) via long flexible linkers (Murray et al., 2016). Expression of EQQ in heterologous systems produces KCNQ1/KCNE1 heteromeric channels with 4:2 stoichiometry (Fig. 6 A; Murray et al., 2016). The choice of EQQ was a compromise between the KCNQ1 homomers with high K+o dependency but with a single-channel conductance close to the limit of the resolution of the recordings system (Werry et al., 2013; Hou et al., 2017), and KCNQ1/KCNE1 with much larger conductance (3.2 pS; Werry et al., 2013; Murray et al., 2016) but having a lower K+o dependency compared with KCNQ1 homomers (see below). Our initial whole-cell oocyte recordings of EQQ revealed 26 ± 4% inhibition of current when K+o was increased from 5 to 97 mM (Fig. 6 A). Single-channel recordings at identical K+o conditions revealed 55 ± 5% reduction in single-channel amplitude at +60 mV (Fig. 6, B–D). Fitting the GHK flux equation to current–voltage data revealed a similar 57 ± 8% reduction in absolute permeability of the single channel (Fig. 6 D). Inhibition of single-channel current by K+o was nearly twice as large as on the corresponding whole-oocyte currents. This discrepancy could be attributable to the putative additional effect of external K+ on channel gating, flickering, or to the influence of the intracellular factor(s) different in oocytes versus ltk-mouse cells.

We next recorded a single F339A mutant channel in order to understand how external K+ produces potentiation in this mutant as well as those in V310A and F340A. The average single-channel amplitude of the F339A mutant at 5 mM K+o was larger (Fig. 6, E and F) than that of the wild type (Wang et al., 2020), which allowed us to analyze the K+o dependency for this mutant. The estimated single-channel amplitude of F339A at +60 mV was nearly unchanged in 97 mM K+o conditions compared with that of 5 mM K+o (Fig. 6, E and F). Calculations of absolute single-channel permeability via fitting data to the GHK flux equation resulted in values close to those predicted theoretically (Fig. 6 F). These results indicate that alanine substitution at this position eliminates the effect of K+o on the unitary conductance of the channel. The current potentiation of F339A mutant by high K+o observed in whole-oocyte recordings, therefore, is likely to be attributable to the influence of K+o on other channel properties.

Conductive modes of the KCNQ1 SF and the effect of external K+

To gain atomic-level insight into how K+o affects KCNQ1 current, we next employed a CompEl method (Fig. S6), recently used to explore the mechanism of ion permeation in prototypic K+ channels (Köpfer et al., 2014; Kopec et al., 2018). We constructed a similar molecular system with a double-membrane configuration using the atomic coordinates of the pore region together with S4-S5 helix (G245-K354) of the recently determined cryo-EM structure of KCNQ1 (PDB ID: 6V01; Sun and MacKinnon, 2020; see supplementary text at the end of the PDF for details), proposed to represent the open pore conformation. Judged by the RMSD values of the Cα atoms, the stability of the protein in the system (Fig. S7) was comparable with what has been found in early pore-only simulation studies (Bernèche and Roux, 2000; Linder et al., 2013; Bernsteiner et al., 2019). An imbalance of two Cl ions between solute compartments generated a transmembrane potential of ∼300 mV (Fig. S6 and Table S1), which was also in agreement with similar studies on Kv1.2 and KscA channels conducted with the same methodology (Kutzner et al., 2011). Then, MD simulations of 500 ns duration were run at 5 and 150 mM external K+ conditions (Fig. S8). Observation of numerous outward K+ permeation events at both K+o conditions (Fig. 7 A) confirmed that the cryo-EM structure of KCNQ1 with bound PIP2 is the open-conductive conformation of the channel (Sun and MacKinnon, 2020). Calculation of the outward current from the MD trajectories revealed ∼30% reduction of mean current amplitude at 150 mM K+o compared with 5 mM conditions (Fig. 7 B). Such inhibition of pore conductance by high K+ was qualitatively in agreement with the results obtained in single-channel recordings. Analysis of MD trajectories revealed two principally different K+ permeation modes of the SF illustrated schematically in Fig. 7 C and Fig. S8. Nearly half of all K+ permeation events at 5 mM K+o followed the mechanism described earlier for prototypic K+ channels (Jensen et al., 2010; Köpfer et al., 2014; Kopec et al., 2018). According to this mechanism, the potential-driven entering of intracellular K+ into the channel internal cavity and its subsequent binding to Scav site triggers a release of the K+ bound to the S0 site of the SF into the external medium (Fig. 7 C, Fig. S8, and Video 1). By contrast, nearly 40% of the observed K+ permeations at 5 mM K+o followed a slightly different mechanism that involved a spontaneous release of K+ from the S0 without any activity at the Scav site (Fig. 7 C, right loop; Fig. S8; and Video 2). We refer to these two K+ permeation cycles of the SF as canonical and spontaneous S0-release modes, respectively. An important difference between the two modes is the low probability of S0 site occupancy by K+ in spontaneous S0 mode. These permeation modes, along with others that comprise nearly 10% of permeation events in our simulations, randomly interchanged with each other during simulations (Fig. S9). In simulations with high external K+, we observed an increase in the frequency and duration of S0 site occupancy by K+o (Fig. 7 D and Fig. S8), which resulted in a diminution of permeation events occurring via the spontaneous S0 mechanism (Fig. 7 E). High K+o induced significant redistribution in the SF substates (Fig. S8) with an overall effect of delayed forward relocations of K+ ions in the SF. This is reflected in K+ density plot as a rise of the Scav peak (Fig. 7 F) and an increase of duration between two subsequent permeations of ions under high external K+ conditions (Fig. 7 A, red traces). These occur, for instance, when K+o temporarily binds to the vacant S0 site during S4-S3-S1 ion occupancy of the SF inducing a backward translocation of K+ ions to generate the Scav-S3-S2 substate (Video 3). Further delay in K+ outward translocations occurs when K+o binds to the S0 site of the SF in the Scav-S3-S2 occupancy configuration, resulting in Scav-S3-S2-S0 occupancy (Fig. S8 and Video 3). Thus, due to the temporary occupation of the S0 site by K+o, a delay in the outward translocations of K+ ion in the SF leads to a reduction of the channel conductance.

We next evaluated the dynamic properties of F339, F340, and V310 side chains to understand the K+o-induced potentiation in corresponding alanine mutants. Remarkably, the aromatic rings of both F339 and F340 adopt two discrete conformations stabilized by distinct interactions with defined residues of the same and the neighboring subunits (Fig. S10). Over the course of the MD simulations, we have observed a flipping of aromatic rings between these two stable configurations in an asymmetric manner (Fig. S10 and Video 4). The aromatic ring of F339 interacts with I268, G269, and T265 residues of the same subunit and L251′ of the neighboring subunit. Upon flipping, the pattern of interaction is changed to comprise L251′, V255′, and L347′ residues of the neighboring subunit (Fig. S10). The benzene ring of F340 is stabilized by V310, G269, and L273 residues of the same subunit (Fig. S10). It makes new hydrophobic and van der Waals contacts with the T311 residue of the SF and I337 as well as V334 residues of the neighboring subunits after flipping. Remarkably, the distribution of probabilities of C-Cα-Cβ-Cγ dihedral angles was changed in high external K+ (Fig. S10). These indicate that the F339–F340 region of the protein involved in the stabilization of the lower pore helix is very dynamic in KCNQ1 due to asymmetric flipping of the aromatic rings of different subunits. F339 and F340 residues correspond to the region in prototypic KcsA and Shaker K+ channels involved in the coupling of the intracellular gate to the SF. Particularly, residues I100–F103 in KcsA and V467–I470 in Shaker transmit the conformational changes induced by the wide opening of the gate to the SF leading to C-type inactivation. It has been shown that during this process the side-chain rotameric angle of I100 and F103 residues in KscA undergo significant redistribution (Li et al., 2018). Similar findings have been described for homologous residues (V476 and I470) in Shaker (Li et al., 2021). K+o sensitivity changes in F339A and F340A mutants observed in our study suggest that a similar interaction may exist in KCNQ1 that shapes the SF properties of the channel. In agreement with this notion, single-channel recordings of F339A have shown an increase in unitary conductance and almost no inhibitory action of K+o on this mutant. In addition, the F339A mutant seems to exhibit less flickery phenotype when compared with EQQ (Fig. 6). V310A, F339A, and F340A mutants exhibit slow inactivation phenotype (Fig. 3 A), raising the question of whether the modulatory action of K+o could be linked to the slow inactivation process in these mutants. We found that co-expression of KCNE1 eliminates the slow inactivation of F339A (Fig. S11). Remarkably, the potentiating effect of high K+o was retained in F339A/KCNE1 heteromeric channels. These results were similar to those obtained previously in R231A/F340W/KCNE1 leak channels (Panaghie et al., 2008). Based on these data, we propose that mutation-induced stabilization of the conductive conformation of the SF is the main cause of the K+o-sensitivity changes in F339A, F340A, and V310 mutants.

External K+ dependency of heteromeric KCNQ1 channels

KCNQ1 channel complexes in most human tissues are believed to be heteromeric, composed of pore-forming α subunits and KCNE1–5 β subunits. To evaluate the physiological significance of the above findings, we co-expressed different KCNE proteins with KCNQ1 and analyzed the K+o dependency of resultant heteromeric complexes. A marked suppression of K+o sensitivity of KCNQ1 was observed upon co-injection of KCNE1 subunit at 1:4 α to β mRNA molecular ratio (Fig. 8 and Table 1) to saturate KCNQ1 with KCNE1. An estimated ∼20% inhibition of current in the range of [K+]o from 0.2 to 100 mM is in contrast with the results of Larsen et al. (2011), who reported no effect of K+o on KCNQ1/KCNE1 heteromers. One possible reason for this discrepancy is the narrow range of [K+]o (1 and 10 mM) used in their study. We then analyzed EQQ tandem channels that correspond to the KCNQ1/KCNE1 heteromers with 4:2 stoichiometry. We revealed marked inhibition by K+o in this channel (Fig. 8, A–C). The dose–response curve of EQQ took an intermediate position between those of homomeric and saturated KCNQ1/KCNE1 heteromeric channels (Fig. 8 C and Table 1). Co-expression of KCNE2 with KCNQ1 in oocytes via mRNA injection at 1:4 α to β molecular ratio resulted in voltage-independent channels that were virtually insensitive to K+o (Fig. 8, A–C). We observed only 5–9% of inhibition when current amplitudes at 0.2 and 100 mM K+o conditions were compared. This was in disagreement with a previously reported ∼30% peak current difference comparing 1 and 10 mM K+o conditions in Xenopus oocyte expression system (Larsen et al., 2011). A possible reason for such a large discrepancy could be the low levels of KCNE2 co-association with KCNQ1, resulting in nonsaturated KCNQ1/KCNE2 heteromers in this latter study as the small voltage-independent current fraction and fast deactivation kinetics of tail currents indicate. For the KCNQ1/KCNE3 complex, we observed a maximal inhibition (Imax) comparable with that of the homomeric KCNQ1 channel and a statistically significant right-shift in the dose–response curve (Fig. 8 and Table 1). A small fractional voltage-dependent activation component visible in traces corresponding to 0.2 and 2 mM K+o pointed toward the possibility that KCNQ1 is not fully saturated by KCNE3 subunit despite the injection of mRNA at 1:1 α to β molecular ratio. Further fractional increase of mRNA encoding KCNE3 causes a significant deterioration of oocytes. Furthermore, the assessment of the dose response in the KCNQ1/KCNE3 leak channel was a difficult experimental goal in our study even at 1:1 molecular ratio since the long-lasting recordings of these channels became unstable during the time needed due to large inward holding currents. Therefore, in another set of experiments, we varied the α to β ratio and compared the current amplitudes at 0.2 and 100 mM K+o concentrations (Fig. 8, A and D). These experiments have clearly shown that an increase in KCNE3 ratio markedly changed the gating kinetics of the channel, but the inhibitory action of K+o was comparable. The extent of inhibition observed in these experiments with injected 1:1 molecular ratio is comparable with data derived from the dose–response analysis (Fig. 8 B), confirming the significant K+o sensitivity of KCNQ1/KCNE3 heteromers. Given that both KCNQ1/KCNE2 and KCNQ1/KCNE3 mediate similar potential-independent currents in the physiological range of voltages (Schroeder et al., 2000; Tinel et al., 2000), the observed difference in K+o sensitivity between these channels is remarkable. Overall, the results indicate that, along with homomeric KCNQ1, various KCNQ1/KCNE complexes are regulated by external K+. Our results and those of earlier studies (Larsen et al., 2011; Wang et al., 2015) show that the extent of the K+o-dependent regulation of KCNQ1/KCNE complexes may depend on the type and the quantity of the associated β subunits.

K+o exerts a stabilizing influence on the conductive conformation (Doi et al., 1996; Cohen et al., 2008; Sandoz et al., 2009; Massaeli et al., 2010; Edvinsson et al., 2011) in most K+ channels, a typical example of which is the attenuation of C-type inactivation in a large number of K+ channels (Pardo et al., 1992; López-Barneo et al., 1993; Kiss and Korn, 1998; Zhou et al., 2001; Cuello et al., 2010). K+o is obligatory for the conductive SF conformation of HERG channels and their membrane availability (Massaeli et al., 2010). Although the mechanism and structural particularities of C-type inactivation of K+ channels are not completely understood (Liu et al., 2015; Armstrong and Hollingworth, 2018), the bulk of data suggests that it involves conformational changes in the external mouth of the channel pore (Kurata and Fedida, 2006; Armstrong and Hoshi, 2014; Pau et al., 2017; Tan et al., 2022). Importantly, at the single channel level, C-type inactivation is the decline of the channel’s open probability over time (Yang et al., 1997; Cuello et al., 2010; Pau et al., 2017), implying that external K+ shifts the open–closed equilibrium of the pore toward open states without any influence on unitary conductance. In notable contrast, here, we report a marked reduction of unitary conductance of KCNQ1 under high external K+ conditions.

Alanine mutation in three residues G348, F350, and K354 that are located in the gate region markedly reduced the inhibitory effect of K+o (Fig. 3 C), pointing toward the possible involvement of the cytoplasmic gate in the inhibition process. Following lines of evidence, yet, suggest that the K+ sensitivity of KCNQ1 is largely independent of the conformational state of the gate: (1) constitutively open KCNQ1/KCNE3 heteromeric channels displays K+o sensitivity comparable to wild type homomers (Fig. 8); (2) ∼30% inhibition of current amplitude observed in MD simulations with the intracellular gate “clamped” at open conformation (Fig. 7 B) is comparable with that of homomeric KCNQ1 (∼28%) when 5 and 100 mM concentrations are compared (Fig. 1 E); (3) G348A mutation induces a shift in conductance–voltage relationship toward hyperpolarized potentials (ΔV1/2 G348A ∼ −10 mV), whereas F350A and K354A exhibit positive GV shifts (ΔV1/2 F350A ∼ +20 mV, ΔV1/2 K354A ∼ +15 mV; Ma et al., 2011). The K+o sensitivity of all these mutants is changed in the same direction (Fig. 3 C). Conversely, marked gating perturbations observed in alanine mutants of S349 and L353 residues (ΔV1/2 S349A ∼ +20 mV, ΔV1/2 K353A ∼ −20 mV; Ma et al., 2011) do not match with unchanged K+o sensitivity of these channels (Fig. 3 C). The side chains of S349 and L353 form the cytoplasmatic seal according to available cryo-EM structures (Sun and MacKinnon, 2017, 2020), confirming the notion that no correlation between steady-state gating of the channel and its K+o dependency exists. How could G348A, F350A, and K354A mutations change the K+o sensitivity of the channel? The cytoplasmatic gate of many K+ channels is allosterically coupled to the SF (Ader et al., 2009; Li et al., 2018; Kopec et al., 2019; Li et al., 2021). Amino acids I100–F103 in KcsA and V467–I470 in Shaker prototypic channels play an important role in such coupling (Li et al., 2018; Li et al., 2021). The analogous I337–F340 region in KCNQ1 having a high mutational impact on K+o sensitivity may be involved in a similar coupling process that transmits conformational changes due to G348A, F350A, and K354A mutations to the SF influencing its function.

K+o differentially affects the function of the SF in KCNQ1 versus KcsA channels causing a reduction of conductance in one case and a slowing of the inactivation in the other. One possible explanation could be the asymmetricity of interactions in I337–F340 region of the KCNQ1 channel. I100–F103–T74 interactions in KcsA that play an essential role in the gate–SF coupling of the channel are symmetrical. Analogous interactions of homologous residues I337–T311–F340 in KCNQ1 are asymmetric. They take place in one (S1) subunit at a much higher probability (Fig. S10) at the dihedral angle of F340 side chain χ = 20–80o. In the other three subunits, the G269–V310–F340 interactions prevail at dihedral angle χ = 120–180o. Furthermore, F339 residue in KCNQ1 is substituted with serine 102 in KcsA and threonine 469 in Shaker. The large side-chain volume at this position with highly dynamic properties (Fig. S10) might be another contributing factor. Overall, these data indicate that the allosteric coupling of the KCNQ1 inner gate to the selectivity could be highly dynamic in KCNQ1. Certain mutations in the gate region such as G348A, F350A, and K354A may strengthen coupling, leading to changes in the SF properties.

The results of MD simulations suggest that the flipping of F339 and F340 aromatic rings most likely destabilizes the conductive dynamics of the KCNQ1 SF, which may be further deteriorated by the binding and unbinding of external K+ at high K+o conditions. Single-channel analysis shows that the removal of the aromatic ring by F339A mutation increased the unitary conductance and almost completely eliminated the effect of K+o on unitary conductance (Fig. 6). A large mutational effect observed by alanine substitution of V310 is likely to be due to modification of the V310–F340 interactions (Fig. 3 D and Fig. S10).

In our study, current amplitudes calculated from MD trajectories are about 8-fold higher than the average single-channel amplitudes of EQQ and, at least, 20-fold higher than corresponding single-channel currents of KCNQ1 homomers. The factors contributing to this discrepancy may include, but are not limited to, applied supraphysiological voltages, imperfections in force field parameters, truncation of the channel to pore-only structures, and movement restrictions applied to keep the channels in the open state. Many aspects of these limitations of MD methods are covered in recent reviews (Flood et al., 2019; Mironenko et al., 2021). Despite these limitations, MD simulations translate the modulatory effect of the external K+ on unitary conductance reasonably well and provided atomic insights into the mechanism underlying this phenomenon.

Co-expression of KCNE1–3 subunits with KCNQ1 at 1:1 and higher β to α mRNA ratios almost completely eliminated the fast inactivation of KCNQ1. Nevertheless, we observed a differential influence of these β subunits on the K+o dependency of KCNQ1. This confirmed our findings of alanine screening experiments (Fig. 2) on the heteromeric channel level. How does the co-expression of KCNE subunits change the K+o sensitivity of KCNQ1? Early studies have shown that the extracellular region of KCNE1 interacts with external residues of S6 segments of the KCNQ1 pore (Xu et al., 2008; Chung et al., 2009). Whether these interactions contribute to the increased single-channel conductance in KCNQ1/KCNE1 heteromers (Sesti and Goldstein, 1998; Murray et al., 2016) remains unknown. The involvement of the SF in this process is also unclear. We consider it likely that KCNE1 stabilizes the conductive dynamics of the SF leading to an increase in unitary conductance and the reduction of the K+o sensitivity of the channel. KCNE2 accessory subunits affect both the gating (Tinel et al., 2000) and the plasma membrane targeting of KCNQ1 (Roura-Ferrer et al., 2010). Almost complete elimination of the K+o sensitivity of KCNQ1 by KCNE2 coexpression confirms the action of KCNE2 on the SF proposed previously based on the measurements of Rb+/K+ permeability ratio in KCNQ1/KCNE2 heteromers (Wang et al., 2012). Co-assembly of KCNE3 with KCNQ1 shifts the dose–response curve of the channel toward higher K+o concentrations (Fig. 8) indicating either a decrease of K+o affinity to S0 or a lower efficacy of the inhibition following the K+o binding to S0. A recent KCNQ1/KCNE3 structure (Sun and MacKinnon, 2020) together with earlier crosslinking electrophysiological investigations suggests that the pattern of interaction of the extracellular region of KCNE3 with KCNQ1 differs from that of KCNE1. The differential effect of K+o on heteromeric KCNQ1/KCNE3 compared with that of KCNQ1/KCNE1 channels, therefore, could be attributable to distinct conformational differences in the upper pore region of these two channels. Comparison of available atomic resolution structures of KCNQ1 and KCNQ1/KCNE3 channels reveals no discernable structural difference(s) in the SF (Sun and MacKinnon, 2020), suggesting that the shift of the dose–response curve is likely attributable rather to the dynamic properties of the SF.

Physiological significance

In mammals, KCNQ1 and KCNE1 proteins are expressed in the apical surfaces of MCs of the stria vascularis (Knipper et al., 2006) and the dark cells of vestibular organs (Nicolas et al., 2001). In both cell types, the extracellular side of the channels is exposed to endolymph, the K+ concentration of which changes from the low millimolar range during embryonic development (Anniko and Wersäll, 1979) to 150 mM in adults (Hibino and Kurachi, 2006). The extent of reported K+o-induced Iks current reduction in mouse MCs (Wang et al., 2015) is comparable with what we have observed for homomeric KCNQ1 and heteromeric EQQ channels (Fig. 8). This suggests a KCNQ1:KCNE1 stoichiometry lower or equal to 4:2 in these MCs. An elegant study by Kurachi and coworkers has shown that the K+ diffusion through the apical membrane of MCs mediated by Iks channels contributes to the endocochlear potential (EP; Nin et al., 2008). The voltage-dependent regulation of Iks in these cells will be partially or completely lost since these channels experience membrane potentials close to +10 mV (Offner et al., 1987) or much higher (+80 mV) according to a more recent study (Nin et al., 2008). K+o sensitivity of KCNQ1/KCNE1 under these conditions will be a powerful mechanism for regulation of K+ flow toward endolymph serving as an important feedback regulatory tool for maintenance of the high endolymph [K+]. Such regulation can be particularly pronounced during the period from prehearing to hearing onset when [K+] in the endolymph gradually increases. Since K+ diffusion through the apical membrane of MCs contributes to the generation of the endocochlear potential (Nin et al., 2008), the K+o-sensitivity of KCNQ1 is essential for hearing and balance.

The stoichiometric ratio for KCNQ1/KCNE in human cardiomyocytes is currently uncertain. It is becoming increasingly clear that it can vary between 4:1 and 4:4 depending on the availability of KCNE1 subunits as evidenced in heterologous expression studies (Nakajo et al., 2010; Murray et al., 2016). Our estimation of K+o-induced Iks reduction for various heteromers under the severe hypokalemic condition with reported serum [K+] as low as 1.2 mM (Kratz et al., 2004; Garcia et al., 2008) is <15% compared with normal serum K+ (3.5–5.5 mM). At first glance, such a small modulation of Iks implies no large effect on cardiac excitability. Nevertheless, one must consider that hypokalemia significantly reduces Ikr current in cardiomyocytes mediated by HERG channels (Massaeli et al., 2010), suggesting that even a minor reduction of Iks under these circumstances may have a significant impact. At present, the cellular expression, localization, and subunit composition of KCNQ1 channels in other organs, such as in stomach, kidney, pancreas, thyroid, and testis are far from known, leaving the full physiological significance of the described phenomenon as an open question for future studies.

Crina M. Nimigean served as editor.

We gratefully acknowledge Prof. Olaf Pongs and Dr. Ruben Stepanyan for reading and commenting on the manuscript. We also acknowledge Yvonne Pechmann (ZMNH, University of Hamburg) and Dr. Emely Thompson (University of British Columbia) for technical assistance. Mr. Johny Petrosyan helped in the initial phase of the work.

This work was supported by grants from the Volkswagen Foundation (grant # AZ 86659, AZ 92111, and AZ 9A908 to V. Vardanyan) and by the Natural Sciences and Engineering Research Council of Canada (grant # RGPIN-2016-05422), the Canadian Institutes of Health Research (#PJT-156181), and the Heart and Stroke Foundation of Canada (#G17-0018392) grants to D. Fedida. V. Vardanyan is supported also by the Armenian National Science and Education Fund (ANSEF grant # Molbiol 4057 and NS-hubio-2509). E. Sargsyan received support from the Swedish Institute (#22729/2017).

The authors declare no competing financial interests.

Author contributions: All authors contributed to the conception and design of the study, analysis of data, interpretation of data, and writing the manuscript. A. Abrahamyan, J. Eldstrom, N. Karagulyan, L. Mkrtchyan, T. Karapetyan, and E. Sargsyan carried out mutagenesis and electrophysiological experiments. H. Sahakyan performed MD simulations.

Abbott
,
G.W.
2014
.
Biology of the KCNQ1 potassium channel
.
New J. Sci.
2014
:
26
.
Ader
,
C.
,
R.
Schneider
,
S.
Hornig
,
P.
Velisetty
,
V.
Vardanyan
,
K.
Giller
,
I.
Ohmert
,
S.
Becker
,
O.
Pongs
, and
M.
Baldus
.
2009
.
Coupling of activation and inactivation gate in a K+-channel: Potassium and ligand sensitivity
.
EMBO J.
28
:
2825
2834
.
Agarwal
,
R.
,
R.
Afzalpurkar
, and
J.S.
Fordtran
.
1994
.
Pathophysiology of potassium absorption and secretion by the human intestine
.
Gastroenterology
.
107
:
548
571
.
Anniko
,
M.
, and
J.
Wersäll
.
1979
.
The inner ear and the in vitro system
.
Arch. Otorhinolaryngol.
224
:
25
30
.
Armstrong
,
C.M.
, and
S.
Hollingworth
.
2018
.
A perspective on Na and K channel inactivation
.
J. Gen. Physiol.
150
:
7
18
.
Armstrong
,
C.M.
, and
T.
Hoshi
.
2014
.
K⁺ channel gating: C-Type inactivation is enhanced by calcium or lanthanum outside
.
J. Gen. Physiol.
144
:
221
230
.
Barhanin
,
J.
,
F.
Lesage
,
E.
Guillemare
,
M.
Fink
,
M.
Lazdunski
, and
G.
Romey
.
1996
.
KVLQT1 and IsK (minK) proteins associate to form the IKs cardiac potassium current
.
Nature
.
384
:
78
80
.
Bernèche
,
S.
, and
B.
Roux
.
2000
.
Molecular dynamics of the KcsA K+ channel in a bilayer membrane
.
Biophys. J.
78
:
2900
2917
.
Bernsteiner
,
H.
,
E.M.
Zangerl-Plessl
,
X.
Chen
, and
A.
Stary-Weinzinger
.
2019
.
Conduction through a narrow inward-rectifier K+ channel pore
.
J. Gen. Physiol.
151
:
1231
1246
.
Chouabe
,
C.
,
N.
Neyroud
,
P.
Guicheney
,
M.
Lazdunski
,
G.
Romey
, and
J.
Barhanin
.
1997
.
Properties of KvLQT1 K+ channel mutations in Romano-Ward and Jervell and Lange-Nielsen inherited cardiac arrhythmias
.
EMBO J.
16
:
5472
5479
.
Chung
,
D.Y.
,
P.J.
Chan
,
J.R.
Bankston
,
L.
Yang
,
G.
Liu
,
S.O.
Marx
,
A.
Karlin
, and
R.S.
Kass
.
2009
.
Location of KCNE1 relative to KCNQ1 in the IKS potassium channel by disulfide cross-linking of substituted cysteines
.
Proc. Natl. Acad. Sci. USA.
106
:
743
748
.
Cohen
,
A.
,
Y.
Ben-Abu
,
S.
Hen
, and
N.
Zilberberg
.
2008
.
A novel mechanism for human K2P2.1 channel gating. Facilitation of C-type gating by protonation of extracellular histidine residues
.
J. Biol. Chem.
283
:
19448
19455
.
Cuello
,
L.G.
,
V.
Jogini
,
D.M.
Cortes
, and
E.
Perozo
.
2010
.
Structural mechanism of C-type inactivation in K+ channels
.
Nature
.
466
:
203
208
.
Darden
,
T.
,
D.
York
, and
L.
Pedersen
.
1993
.
Particle mesh Ewald: An N⋅log(N) method for Ewald sums in large systems
.
J. Chem. Phys.
98
:
10089
10092
.
Doi
,
T.
,
B.
Fakler
,
J.H.
Schultz
,
U.
Schulte
,
U.
Brändle
,
S.
Weidemann
,
H.P.
Zenner
,
F.
Lang
, and
J.P.
Ruppersberg
.
1996
.
Extracellular K+ and intracellular pH allosterically regulate renal Kir1.1 channels
.
J. Biol. Chem.
271
:
17261
17266
.
Duggal
,
P.
,
M.R.
Vesely
,
D.
Wattanasirichaigoon
,
J.
Villafane
,
V.
Kaushik
, and
A.H.
Beggs
.
1998
.
Mutation of the gene for IsK associated with both Jervell and Lange-Nielsen and Romano-Ward forms of long-QT syndrome
.
Circulation
.
97
:
142
146
.
Edvinsson
,
J.M.
,
A.J.
Shah
, and
L.G.
Palmer
.
2011
.
Potassium-dependent activation of Kir4.2 K⁺ channels
.
J. Physiol.
589
:
5949
5963
.
Flood
,
E.
,
C.
Boiteux
,
B.
Lev
,
I.
Vorobyov
, and
T.W.
Allen
.
2019
.
Atomistic simulations of membrane ion channel conduction, gating, and modulation
.
Chem. Rev.
119
:
7737
7832
.
Garcia
,
E.
,
N.
Nakhleh
,
D.
Simmons
, and
C.
Ramsay
.
2008
.
Profound hypokalemia: Unusual presentation and management in a 12-year-old boy
.
Pediatr. Emerg. Care
.
24
:
157
160
.
Ghosh
,
S.
,
D.A.
Nunziato
, and
G.S.
Pitt
.
2006
.
KCNQ1 assembly and function is blocked by long-QT syndrome mutations that disrupt interaction with calmodulin
.
Circ. Res.
98
:
1048
1054
.
Goldman
,
A.M.
,
E.
Glasscock
,
J.
Yoo
,
T.T.
Chen
,
T.L.
Klassen
, and
J.L.
Noebels
.
2009
.
Arrhythmia in heart and brain: KCNQ1 mutations link epilepsy and sudden unexplained death
.
Sci. Transl. Med.
1
:
2ra6
.
Hedley
,
P.L.
,
P.
Jørgensen
,
S.
Schlamowitz
,
R.
Wangari
,
J.
Moolman-Smook
,
P.A.
Brink
,
J.K.
Kanters
,
V.A.
Corfield
, and
M.
Christiansen
.
2009
.
The genetic basis of long QT and short QT syndromes: A mutation update
.
Hum. Mutat.
30
:
1486
1511
.
Hess
,
B.
,
H.
Bekker
,
H.J.
Berendsen
, and
J.G.
Faaije
.
1997
.
LINCS: A linear constraint solver for molecular simulations
.
J. Comput. Chem.
18
:
1463
1472
.
Hess
,
B.
,
C.
Kutzner
,
D.
van der Spoel
, and
E.
Lindahl
.
2008
.
GROMACS 4: Algorithms for highly efficient, load-balanced, and scalable molecular simulation
.
J. Chem. Theory Comput.
4
:
435
447
.
Hibino
,
H.
, and
Y.
Kurachi
.
2006
.
Molecular and physiological bases of the K+ circulation in the mammalian inner ear
.
Physiology (Bethesda)
.
21
:
336
345
.
Hopkins
,
C.W.
,
S.
Le Grand
,
R.C.
Walker
, and
A.E.
Roitberg
.
2015
.
Long-time-step molecular dynamics through hydrogen mass repartitioning
.
J. Chem. Theory Comput.
11
:
1864
1874
.
Hou
,
P.
,
J.
Eldstrom
,
J.
Shi
,
L.
Zhong
,
K.
McFarland
,
Y.
Gao
,
D.
Fedida
, and
J.
Cui
.
2017
.
Inactivation of KCNQ1 potassium channels reveals dynamic coupling between voltage sensing and pore opening
.
Nat. Commun.
8
:
1730
.
Hoover
,
W.G.
1985
.
Canonical dynamics: Equilibrium phase-space distributions
.
Phys. Rev. A Gen. Phys.
31
:
1695
1697
.
Jensen
,
M.Ø.
,
D.W.
Borhani
,
K.
Lindorff-Larsen
,
P.
Maragakis
,
V.
Jogini
,
M.P.
Eastwood
,
R.O.
Dror
, and
D.E.
Shaw
.
2010
.
Principles of conduction and hydrophobic gating in K+ channels
.
Proc. Natl. Acad. Sci. USA.
107
:
5833
5838
.
Jespersen
,
T.
,
M.
Grunnet
, and
S.P.
Olesen
.
2005
.
The KCNQ1 potassium channel: From gene to physiological function
.
Physiology (Bethesda)
.
20
:
408
416
.
Jo
,
S.
,
T.
Kim
,
V.G.
Iyer
, and
W.
Im
.
2008
.
CHARMM-GUI: a web-based graphical user interface for CHARMM
.
J. Comput. Chem.
29
:
1859
1865
.
Jo
,
S.
,
J.B.
Lim
,
J.B.
Klauda
, and
W.
Im
.
2009
.
CHARMM-GUI Membrane Builder for mixed bilayers and its application to yeast membranes
.
Biophys. J.
97
:
50
58
.
Jorgensen
,
W.L.
,
J.
Chandrasekhar
,
J.D.
Madura
,
R.
Impey
, and
M.
Klein
.
1983
.
Comparison of simple potential functions for simulating liquid water
.
J. Chem. Phys.
79
:
926
935
.
Joung
,
I.S.
and
T.E.
Cheatham
III
.
2008
.
Determination of alkali and halide monovalent ion parameters for use in explicitly solvated biomolecular simulations
.
J. Phys. Chem. B
.
112
:
9020
9041
.
Kienitz
,
M.C.
, and
D.
Vladimirova
.
2015
.
Synergistic modulation of KCNQ1/KCNE1 K+ channels (IKs) by phosphatidylinositol 4,5-bisphosphate (PIP2) and [ATP]i
.
Cell. Signal.
27
:
1457
1468
.
Kiss
,
L.
, and
S.J.
Korn
.
1998
.
Modulation of C-type inactivation by K+ at the potassium channel selectivity filter
.
Biophys. J.
74
:
1840
1849
.
Knipper
,
M.
,
C.
Claussen
,
L.
Rüttiger
,
U.
Zimmermann
,
R.
Lüllmann-Rauch
,
E.L.
Eskelinen
,
J.
Schröder
,
M.
Schwake
, and
P.
Saftig
.
2006
.
Deafness in LIMP2-deficient mice due to early loss of the potassium channel KCNQ1/KCNE1 in marginal cells of the stria vascularis
.
J. Physiol.
576
:
73
86
.
Kopec
,
W.
,
D.A.
Köpfer
,
O.N.
Vickery
,
A.S.
Bondarenko
,
T.L.C.
Jansen
,
B.L.
de Groot
, and
U.
Zachariae
.
2018
.
Direct knock-on of desolvated ions governs strict ion selectivity in K+ channels
.
Nat. Chem.
10
:
813
820
.
Kopec
,
W.
,
B.S.
Rothberg
, and
B.L.
de Groot
.
2019
.
Molecular mechanism of a potassium channel gating through activation gate-selectivity filter coupling
.
Nat. Commun.
10
:
5366
.
Köpfer
,
D.A.
,
C.
Song
,
T.
Gruene
,
G.M.
Sheldrick
,
U.
Zachariae
, and
B.L.
de Groot
.
2014
.
Ion permeation in K⁺ channels occurs by direct Coulomb knock-on
.
Science
.
346
:
352
355
.
Kratz
,
A.
,
M.
Ferraro
,
P.M.
Sluss
, and
K.B.
Lewandrowski
.
2004
.
Case records of the Massachusetts General Hospital. Weekly clinicopathological exercises. Laboratory reference values
.
N. Engl. J. Med.
351
:
1548
1563
.
Kurata
,
H.T.
, and
D.
Fedida
.
2006
.
A structural interpretation of voltage-gated potassium channel inactivation
.
Prog. Biophys. Mol. Biol.
92
:
185
208
.
Kutzner
,
C.
,
H.
Grubmüller
,
B.L.
de Groot
, and
U.
Zachariae
.
2011
.
Computational electrophysiology: The molecular dynamics of ion channel permeation and selectivity in atomistic detail
.
Biophys. J.
101
:
809
817
.
Kutzner
,
C.
,
D.A.
Köpfer
,
J.P.
Machtens
,
B.L.
de Groot
,
C.
Song
, and
U.
Zachariae
.
2016
.
Insights into the function of ion channels by computational electrophysiology simulations
.
Biochim. Biophys. Acta
.
1858
(
7 Pt B
):
1741
1752
.
Larsen
,
A.P.
,
A.B.
Steffensen
,
M.
Grunnet
, and
S.P.
Olesen
.
2011
.
Extracellular potassium inhibits Kv7.1 potassium channels by stabilizing an inactivated state
.
Biophys. J.
101
:
818
827
.
Li
,
J.
,
J.
Ostmeyer
,
L.G.
Cuello
,
E.
Perozo
, and
B.
Roux
.
2018
.
Rapid constriction of the selectivity filter underlies C-type inactivation in the KcsA potassium channel
.
J. Gen. Physiol.
150
:
1408
1420
.
Li
,
J.
,
R.
Shen
,
A.
Rohaim
,
R.
Mendoza Uriarte
,
M.
Fajer
,
E.
Perozo
, and
B.
Roux
.
2021
.
Computational study of non-conductive selectivity filter conformations and C-type inactivation in a voltage-dependent potassium channel
.
J. Gen. Physiol.
153
:
153
.
Li
,
Y.
,
J.
Gao
,
Z.
Lu
,
K.
McFarland
,
J.
Shi
,
K.
Bock
,
I.S.
Cohen
, and
J.
Cui
.
2013
.
Intracellular ATP binding is required to activate the slowly activating K+ channel IKs
.
Proc. Natl. Acad. Sci. USA.
110
:
18922
18927
.
Liin
,
S.I.
,
R.
Barro-Soria
, and
H.P.
Larsson
.
2015a
.
The KCNQ1 channel - remarkable flexibility in gating allows for functional versatility
.
J. Physiol.
593
:
2605
2615
.
Liin
,
S.I.
,
M.
Silverå Ejneby
,
R.
Barro-Soria
,
M.A.
Skarsfeldt
,
J.E.
Larsson
,
F.
Starck Härlin
,
T.
Parkkari
,
B.H.
Bentzen
,
N.
Schmitt
,
H.P.
Larsson
, and
F.
Elinder
.
2015b
.
Polyunsaturated fatty acid analogs act antiarrhythmically on the cardiac IKs channel
.
Proc. Natl. Acad. Sci. USA.
112
:
5714
5719
.
Linder
,
T.
,
B.L.
de Groot
, and
A.
Stary-Weinzinger
.
2013
.
Probing the energy landscape of activation gating of the bacterial potassium channel KcsA
.
PLOS Comput. Biol.
9
:e1003058.
Liu
,
S.
,
P.J.
Focke
,
K.
Matulef
,
X.
Bian
,
P.
Moënne-Loccoz
,
F.I.
Valiyaveetil
, and
S.W.
Lockless
.
2015
.
Ion-binding properties of a K+ channel selectivity filter in different conformations
.
Proc. Natl. Acad. Sci. USA.
112
:
15096
15100
.
López-Barneo
,
J.
,
T.
Hoshi
,
S.H.
Heinemann
, and
R.W.
Aldrich
.
1993
.
Effects of external cations and mutations in the pore region on C-type inactivation of Shaker potassium channels
.
Receptors Channels
.
1
:
61
71
.
Loussouarn
,
G.
,
K.H.
Park
,
C.
Bellocq
,
I.
Baró
,
F.
Charpentier
, and
D.
Escande
.
2003
.
Phosphatidylinositol-4,5-bisphosphate, PIP2, controls KCNQ1/KCNE1 voltage-gated potassium channels: A functional homology between voltage-gated and inward rectifier K+ channels
.
EMBO J.
22
:
5412
5421
.
Ma
,
L.J.
,
I.
Ohmert
, and
V.
Vardanyan
.
2011
.
Allosteric features of KCNQ1 gating revealed by alanine scanning mutagenesis
.
Biophys. J.
100
:
885
894
.
Maier
,
J.A.
,
C.
Martinez
,
K.
Kasavajhala
,
L.
Wickstrom
,
K.E.
Hauser
, and
C.
Simmerling
.
2015
.
ff14SB: Improving the accuracy of protein side chain and backbone parameters from ff99SB
.
J. Chem. Theory Comput.
11
:
3696
3713
.
Marcus
,
D.C.
, and
Z.
Shen
.
1994
.
Slowly activating voltage-dependent K+ conductance is apical pathway for K+ secretion in vestibular dark cells
.
Am. J. Physiol.
267
:
C857
C864
.
Massaeli
,
H.
,
J.
Guo
,
J.
Xu
, and
S.
Zhang
.
2010
.
Extracellular K+ is a prerequisite for the function and plasma membrane stability of HERG channels
.
Circ. Res.
106
:
1072
1082
.
Matsuura
,
H.
,
T.
Ehara
, and
Y.
Imoto
.
1987
.
An analysis of the delayed outward current in single ventricular cells of the guinea-pig
.
Pflugers Arch.
410
:
596
603
.
Mironenko
,
A.
,
U.
Zachariae
,
B.L.
de Groot
, and
W.
Kopec
.
2021
.
The persistent question of potassium channel permeation mechanisms
.
J. Mol. Biol.
433
:
167002
.
Murray
,
C.I.
,
M.
Westhoff
,
J.
Eldstrom
,
E.
Thompson
,
R.
Emes
, and
D.
Fedida
.
2016
.
Unnatural amino acid photo-crosslinking of the IKs channel complex demonstrates a KCNE1:KCNQ1 stoichiometry of up to 4:4
.
eLife
.
5
:
5
.
Nakajo
,
K.
,
M.H.
Ulbrich
,
Y.
Kubo
, and
E.Y.
Isacoff
.
2010
.
Stoichiometry of the KCNQ1 - KCNE1 ion channel complex
.
Proc. Natl. Acad. Sci. USA.
107
:
18862
18867
.
Nicolas
,
M.
,
D.
Demêmes
,
A.
Martin
,
S.
Kupershmidt
, and
J.
Barhanin
.
2001
.
KCNQ1/KCNE1 potassium channels in mammalian vestibular dark cells
.
Hear. Res.
153
:
132
145
.
Nin
,
F.
,
H.
Hibino
,
K.
Doi
,
T.
Suzuki
,
Y.
Hisa
, and
Y.
Kurachi
.
2008
.
The endocochlear potential depends on two K+ diffusion potentials and an electrical barrier in the stria vascularis of the inner ear
.
Proc. Natl. Acad. Sci. USA.
105
:
1751
1756
.
Nosé
,
S.
1984
.
A molecular dynamics method for simulations in the canonical ensemble
.
Mol. Phys.
52
:
255
268
.
Offner
,
F.F.
,
P.
Dallos
, and
M.A.
Cheatham
.
1987
.
Positive endocochlear potential: Mechanism of production by marginal cells of stria vascularis
.
Hear. Res.
29
:
117
124
.
Panaghie
,
G.
,
K.
Purtell
,
K.K.
Tai
, and
G.W.
Abbott
.
2008
.
Voltage-dependent C-type inactivation in a constitutively open K+ channel
.
Biophys. J.
95
:
2759
2778
.
Pardo
,
L.A.
,
S.H.
Heinemann
,
H.
Terlau
,
U.
Ludewig
,
C.
Lorra
,
O.
Pongs
, and
W.
Stühmer
.
1992
.
Extracellular K+ specifically modulates a rat brain K+ channel
.
Proc. Natl. Acad. Sci. USA.
89
:
2466
2470
.
Partrinello
,
M.
and
A.
Rahman
.
1981
.
Polymorphic transitions in single crystals: A new molecular dynamics method
.
J. Appl. Phys.
52
:
7182
7190
.
Pau
,
V.
,
Y.
Zhou
,
Y.
Ramu
,
Y.
Xu
, and
Z.
Lu
.
2017
.
Crystal structure of an inactivated mutant mammalian voltage-gated K+ channel
.
Nat. Struct. Mol. Biol.
24
:
857
865
.
Peroz
,
D.
,
N.
Rodriguez
,
F.
Choveau
,
I.
Baró
,
J.
Mérot
, and
G.
Loussouarn
.
2008
.
Kv7.1 (KCNQ1) properties and channelopathies
.
J. Physiol.
586
:
1785
1789
.
Preston
,
P.
,
L.
Wartosch
,
D.
Günzel
,
M.
Fromm
,
P.
Kongsuphol
,
J.
Ousingsawat
,
K.
Kunzelmann
,
J.
Barhanin
,
R.
Warth
, and
T.J.
Jentsch
.
2010
.
Disruption of the K+ channel β-subunit KCNE3 reveals an important role in intestinal and tracheal Cl transport
.
J. Biol. Chem.
285
:
7165
7175
.
Pusch
,
M.
,
L.
Bertorello
, and
F.
Conti
.
2000
.
Gating and flickery block differentially affected by rubidium in homomeric KCNQ1 and heteromeric KCNQ1/KCNE1 potassium channels
.
Biophys. J.
78
:
211
226
.
Pusch
,
M.
,
R.
Magrassi
,
B.
Wollnik
, and
F.
Conti
.
1998
.
Activation and inactivation of homomeric KvLQT1 potassium channels
.
Biophys. J.
75
:
785
792
.
Roura-Ferrer
,
M.
,
L.
Solé
,
A.
Oliveras
,
R.
Dahan
,
J.
Bielanska
,
A.
Villarroel
,
N.
Comes
, and
A.
Felipe
.
2010
.
Impact of KCNE subunits on KCNQ1 (Kv7.1) channel membrane surface targeting
.
J. Cell. Physiol.
225
:
692
700
.
Sachyani
,
D.
,
M.
Dvir
,
R.
Strulovich
,
G.
Tria
,
W.
Tobelaim
,
A.
Peretz
,
O.
Pongs
,
D.
Svergun
,
B.
Attali
, and
J.A.
Hirsch
.
2014
.
Structural basis of a Kv7.1 potassium channel gating module: Studies of the intracellular c-terminal domain in complex with calmodulin
.
Structure
.
22
:
1582
1594
.
Sandoz
,
G.
,
D.
Douguet
,
F.
Chatelain
,
M.
Lazdunski
, and
F.
Lesage
.
2009
.
Extracellular acidification exerts opposite actions on TREK1 and TREK2 potassium channels via a single conserved histidine residue
.
Proc. Natl. Acad. Sci. USA.
106
:
14628
14633
.
Sanguinetti
,
M.C.
,
M.E.
Curran
,
A.
Zou
,
J.
Shen
,
P.S.
Spector
,
D.L.
Atkinson
, and
M.T.
Keating
.
1996
.
Coassembly of KVLQT1 and minK (IsK) proteins to form cardiac IKs potassium channel
.
Nature
.
384
:
80
83
.
Schroeder
,
B.C.
,
S.
Waldegger
,
S.
Fehr
,
M.
Bleich
,
R.
Warth
,
R.
Greger
, and
T.J.
Jentsch
.
2000
.
A constitutively open potassium channel formed by KCNQ1 and KCNE3
.
Nature
.
403
:
196
199
.
Sesti
,
F.
, and
S.A.
Goldstein
.
1998
.
Single-channel characteristics of wild-type IKs channels and channels formed with two minK mutants that cause long QT syndrome
.
J. Gen. Physiol.
112
:
651
663
.
Shamgar
,
L.
,
L.
Ma
,
N.
Schmitt
,
Y.
Haitin
,
A.
Peretz
,
R.
Wiener
,
J.
Hirsch
,
O.
Pongs
, and
B.
Attali
.
2006
.
Calmodulin is essential for cardiac IKS channel gating and assembly: Impaired function in long-QT mutations
.
Circ. Res.
98
:
1055
1063
.
Sun
,
J.
, and
R.
MacKinnon
.
2017
.
Cryo-EM structure of a KCNQ1/CaM complex reveals insights into congenital long QT syndrome
.
Cell
.
169
:
1042
1050.e9
.
Sun
,
J.
, and
R.
MacKinnon
.
2020
.
Structural basis of human KCNQ1 modulation and gating
.
Cell
.
180
:
340
347.e9
.
Tan
,
X.F.
,
C.
Bae
,
R.
Stix
,
A.I.
Fernández-Mariño
,
K.
Huffer
,
T.H.
Chang
,
J.
Jiang
,
J.D.
Faraldo-Gómez
, and
K.J.
Swartz
.
2022
.
Structure of the Shaker Kv channel and mechanism of slow C-type inactivation
.
Sci. Adv.
8
:eabm7814.
Tinel
,
N.
,
S.
Diochot
,
M.
Borsotto
,
M.
Lazdunski
, and
J.
Barhanin
.
2000
.
KCNE2 confers background current characteristics to the cardiac KCNQ1 potassium channel
.
EMBO J.
19
:
6326
6330
.
Tohse
,
N.
1990
.
Calcium-sensitive delayed rectifier potassium current in Guinea pig ventricular cells
.
Am. J. Physiol.
258
:
H1200
H1207
.
Tommiska
,
J.
,
J.
Känsäkoski
,
L.
Skibsbye
,
K.
Vaaralahti
,
X.
Liu
,
E.J.
Lodge
,
C.
Tang
,
L.
Yuan
,
R.
Fagerholm
,
J.K.
Kanters
, et al
.
2017
.
Two missense mutations in KCNQ1 cause pituitary hormone deficiency and maternally inherited gingival fibromatosis
.
Nat. Commun.
8
:
1289
.
Torekov
,
S.S.
,
E.
Iepsen
,
M.
Christiansen
,
A.
Linneberg
,
O.
Pedersen
,
J.J.
Holst
,
J.K.
Kanters
, and
T.
Hansen
.
2014
.
KCNQ1 long QT syndrome patients have hyperinsulinemia and symptomatic hypoglycemia
.
Diabetes
.
63
:
1315
1325
.
Tristani-Firouzi
,
M.
, and
M.C.
Sanguinetti
.
1998
.
Voltage-dependent inactivation of the human K+ channel KvLQT1 is eliminated by association with minimal K+ channel (minK) subunits
.
J. Physiol.
510
:
37
45
.
Van Der Spoel
,
D.
,
E.
Lindahl
,
B.
Hess
,
G.
Groenhof
,
A.E.
Mark
, and
H.J.
Berendsen
.
2005
.
GROMACS: fast, flexible, and free
.
J. Comput. Chem.
26
:
1701
1718
.
Verlet
,
L.
1967
.
Computer “experiments” on classical fluids. I. Thermodynamical properties of Lennard-Jones molecules
.
Phys. Rev.
159
:
98
103
.
Wang
,
Q.
,
M.E.
Curran
,
I.
Splawski
,
T.C.
Burn
,
J.M.
Millholland
,
T.J.
VanRaay
,
J.
Shen
,
K.W.
Timothy
,
G.M.
Vincent
,
T.
de Jager
, et al
.
1996
.
Positional cloning of a novel potassium channel gene: KVLQT1 mutations cause cardiac arrhythmias
.
Nat. Genet.
12
:
17
23
.
Wang
,
W.
,
M.C.
Flores
,
C.R.
Sihn
,
H.J.
Kim
,
Y.
Zhang
,
K.J.
Doyle
,
N.
Chiamvimonvat
,
X.D.
Zhang
, and
E.N.
Yamoah
.
2015
.
Identification of a key residue in Kv7.1 potassium channel essential for sensing external potassium ions
.
J. Gen. Physiol.
145
:
201
212
.
Wang
,
Y.
,
J.
Eldstrom
, and
D.
Fedida
.
2020
.
Gating and regulation of KCNQ1 and KCNQ1 + KCNE1 channel complexes
.
Front. Physiol.
11
:
504
.
Wang
,
Y.
,
M.
Zhang
,
Y.
Xu
,
M.
Jiang
,
D.P.
Zankov
,
M.
Cui
, and
G.N.
Tseng
.
2012
.
Probing the structural basis for differential KCNQ1 modulation by KCNE1 and KCNE2
.
J. Gen. Physiol.
140
:
653
669
.
Weber
,
W.
1999
.
Ion currents of Xenopus laevis oocytes: State of the art
.
Biochim. Biophys. Acta
.
1421
:
213
233
.
Werry
,
D.
,
J.
Eldstrom
,
Z.
Wang
, and
D.
Fedida
.
2013
.
Single-channel basis for the slow activation of the repolarizing cardiac potassium current, IKs
.
Proc. Natl. Acad. Sci. USA.
110
:
E996
E1005
.
Xu
,
X.
,
M.
Jiang
,
K.L.
Hsu
,
M.
Zhang
, and
G.N.
Tseng
.
2008
.
KCNQ1 and KCNE1 in the IKs channel complex make state-dependent contacts in their extracellular domains
.
J. Gen. Physiol.
131
:
589
603
.
Yang
,
Y.
, and
F.J.
Sigworth
.
1998
.
Single-channel properties of IKs potassium channels
.
J. Gen. Physiol.
112
:
665
678
.
Yang
,
Y.
,
Y.
Yan
, and
F.J.
Sigworth
.
1997
.
How does the W434F mutation block current in Shaker potassium channels?
.
J. Gen. Physiol.
109
:
779
789
.
Yazawa
,
K.
, and
M.
Kameyama
.
1990
.
Mechanism of receptor-mediated modulation of the delayed outward potassium current in Guinea-pig ventricular myocytes
.
J. Physiol.
421
:
135
150
.
Zaydman
,
M.A.
,
J.R.
Silva
,
K.
Delaloye
,
Y.
Li
,
H.
Liang
,
H.P.
Larsson
,
J.
Shi
, and
J.
Cui
.
2013
.
Kv7.1 ion channels require a lipid to couple voltage sensing to pore opening
.
Proc. Natl. Acad. Sci. USA.
110
:
13180
13185
.
Zhou
,
Y.
,
J.H.
Morais-Cabral
,
A.
Kaufman
, and
R.
MacKinnon
.
2001
.
Chemistry of ion coordination and hydration revealed by a K+ channel-Fab complex at 2.0 A resolution
.
Nature
.
414
:
43
48
.

Author notes

H. Sahakyan’s present address is National Center for Biotechnology Information, National Library of Medicine, National Institutes for Health, Bethesda, MD, USA.

This article is available under a Creative Commons License (Attribution 4.0 International, as described at https://creativecommons.org/licenses/by/4.0/).