Ion occupancy of the selectivity filter controls opening of a cytoplasmic gate in the K2P channel TALK-2

Neelsen, Lea C.; Riel, Elena B.; Rinné, Susanne; Schmid, Freya-Rebecca; Jürs, Björn C.; Bedoya, Mauricio; Langer, Jan P.; Eymsh, Bisher; Kiper, Aytug K.; Cordeiro, Sönke; Decher, Niels; Baukrowitz, Thomas; Schewe, Marcus

doi:10.1038/s41467-024-51812-w

Download PDF

Article
Open access
Published: 30 August 2024

Ion occupancy of the selectivity filter controls opening of a cytoplasmic gate in the K_2P channel TALK-2

Nature Communications volume 15, Article number: 7545 (2024) Cite this article

2666 Accesses
4 Citations
8 Altmetric
Metrics details

Subjects

Abstract

Two-pore ___domain K⁺ (K_2P) channel activity was previously thought to be controlled primarily via a selectivity filter (SF) gate. However, recent crystal structures of TASK-1 and TASK-2 revealed a lower gate at the cytoplasmic pore entrance. Here, we report functional evidence of such a lower gate in the K_2P channel K2P17.1 (TALK-2, TASK-4). We identified compounds (drugs and lipids) and mutations that opened the lower gate allowing the fast modification of pore cysteine residues. Surprisingly, stimuli that directly target the SF gate (i.e., pH_e., Rb⁺ permeation, membrane depolarization) also opened the cytoplasmic gate. Reciprocally, opening of the lower gate reduced the electric work to open the SF via voltage driven ion binding. Therefore, it appears that the SF is so rigidly locked into the TALK-2 protein structure that changes in ion occupancy can pry open a distant lower gate and, vice versa, opening of the lower gate concurrently promote SF gate opening. This concept might extent to other K⁺ channels that contain two gates (e.g., voltage-gated K⁺ channels) for which such a positive gate coupling has been suggested, but so far not directly demonstrated.

Structural basis for pH gating of the two-pore ___domain K⁺ channel TASK2

Article 30 September 2020

Mechanism of pore opening in the calcium-activated chloride channel TMEM16A

Article Open access 04 February 2021

Atomistic mechanism of coupling between cytosolic sensor ___domain and selectivity filter in TREK K2P channels

Article Open access 31 May 2024

Introduction

TWIK-related alkaline-pH-activated potassium (TALK-2, K_2P17.1, KCNK17) channels are members of the two-pore ___domain K⁺ (K_2P) channel family. They were also referred to as TWIK-related acid-sensitive potassium (TASK-4) channels when first identified in 2001¹. TALK-2 channels are expressed in various human cell types and organs (i.e., pancreas, aorta, brain, liver, placenta, and heart) of the human body^1,2,3. Despite their widespread distribution the functional role of these channels in biological processes has not been established yet, especially in contrast to other well-investigated acid-sensitive K_2P channels such as TASK-1^4,5,6,7,8,9. However, its malfunction, up- or down-regulation or genetic variants of TALK-2 K_2P channels have been associated with a number of cardiovascular diseases such as cardiac conduction disorders¹⁰, ischemic stroke, and cerebral hemorrhage^11,12,13,14 as well as arrhythmias including atrial fibrillation, idiopathic ventricular fibrillation and long QT syndrome^10,15,16. Furthermore, TALK-2 channels are highly and specifically expressed in the human pancreas and are considered as a risk factor for the pathogenesis of type 2 diabetes^17,18. TALK-2 channels are activated by alkaline extracellular pH (pH_e > 7.4) that is thought to occur by deprotonation of an extracellular lysine causing the opening of the SF gate^1,2,18,19. Furthermore, TALK-2 channel currents are enhanced by the production of nitric oxide radicals and reactive oxygen species¹⁸. Like most K_2P channels, TALK-2 channels are sensitive to changes in membrane voltage and permeating ion species²⁰. Recently, polyanionic lipids of the fatty acid metabolism (e.g. oleoyl-CoA) have been identified as natural TALK-2 channel ligands, increasing channel activity by more than 100-fold²¹. How exactly these stimuli regulate the opening and closing of TALK-2 K_2P channels is, with the exception of voltage and pH_e acting at the SF gate, so far unclear. TALK-2 channels are functional dimers and exhibit – as all other 14 K_2P channel family members - a characteristic topology of four transmembrane domains (TM1 to TM4) and two pore-forming domains (P1 and P2) within each channel subunit. The two P1 and P2 domains form the pseudo-tetrameric selectivity filter (SF) of the channel upon dimerization^22,23,24. For the last two decades, K_2P channels were thought to be gated at the SF and, thus, the various physicochemical stimuli acting on different regions of the channel (in particular on the cytoplasmic C-terminus) finally converge on the primary filter gate^{20,25,26,27,28,29}. Surprisingly, the recently resolved structures of TASK-1 and TASK-2, identified additional inner/cytoplasmic gates (further referred to as lower gates)^30,31. In TASK-1 this gate is formed by the crossing (therefore termed X-gate) of the late TM4 domains³¹, however, an activation mechanism for this gate is currently unknown. Based on the two cryo-EM structures of TASK-2 generated at pH 8.5 (open channel) and pH 6.5 (closed channel) the lower gate of TASK-2 is mainly formed by the interaction of two corresponding TM4 residues (K245, N243), hypothesized to function as a molecular barrier in the process of pH gating³⁰. In this study, we employ cysteine modification, alanine scanning mutagenesis, homology modeling and various pore blockers to identify and characterize an additional lower gate in TALK-2 channels. These approaches provide information on the status of the SF gate and lower gate, respectively, and reveal that the two gates are strongly positively-coupled. We show that the ion occupancy of the SF controls opening of the lower gate and estimate the electrical work required to open the lower gate. Our results establish a strong positive coupling of the two gates that can be envisioned as a concerted structural change involving both gates. Further, we demonstrate that the lower gate produces a state-dependent pharmacology that is unique in K_2P channels.

Results

Probing for a cytoplasmic constriction in TALK-2 channels with cysteine modification

The basal activity of TALK-2 channels in excised patches is low, but pharmacological compounds (e.g. 2-APB)³² or polyanionic lipids (e.g., oleoyl-CoA)²¹ can induce large TALK-2 channel currents (Fig. 1a, f). However, the binding sites of these compounds and the mechanisms through which they open the ion permeation pathway are currently unknown. To investigate the latter, we utilized a cysteine modification assay. Control experiments ensured that WT TALK-2 currents were insensitive to the application of the sulfhydryl reactive compound (2-(Trimethylammonium)ethyl) MethaneThioSulfonate (MTS-ET) regardless whether applied on the low-activity basal state (Supplementary Fig. 1a, Supplementary Table 1) or the high activity state induced with e.g., 2-APB or oleoyl-CoA (Supplementary Fig. 1b, d, f). To test for a cytoplasmic constriction in TALK-2, we introduced a cysteine at amino acid position 145 (L145C) in TM2, that corresponds to a residue in TREK-1 (G186C) located in the pore cavity below the SF and previously shown to result in a permeation block upon cysteine modification in TREK-1²⁹. TALK-2 L145C mutant channels showed low basal activity, and both 2-APB and oleoyl-CoA produced robust activation very similar to the WT (Fig. 1g and Supplementary Figs. 1c, e, g and 5a). Application of MTS-ET on the low activity basal state had no effect on the channel activity (Fig. 1d left panel). In marked contrast, application of MTS-ET on L145C TALK-2 currents activated by 2-APB or oleoyl-CoA resulted in complete and irreversible current inhibition, indicating the chemical modification of L145C (Fig. 1d middle and right panel, Supplementary Table 1). These findings indicate that access of MTS-ET to L145C is blocked in the low-activity state of the channel but possible upon activation. To gain a better structural understanding, we generated TALK-2 homology models based on the TASK-1 and TASK-2 structures that both show a lower permeation constriction (see “Methods” section)^30,31. In these models, the side chain of L145 points into the permeation pathway at a position between the SF and the lower gates (Fig. 1b, c and Supplementary Fig. 2a). Furthermore, we used these TALK-2 models to pick a residue for cysteine substitution (Q266) pointing into the cytosol directly below the lower constrictions. Application of MTS-ET to Q266C TALK-2 mutant channels caused a mono-exponential and irreversible current activation (Fig. 1e left panel). Importantly, the observed modification occurred at a similar rate in both the low- and high-activity states suggesting similar access to the cysteine under both conditions (Fig. 1e and Supplementary Table 2). The fact that MTS-ET modification activated Q266C TALK-2 mutant channels might indicate that the modification at this position (i.e., close to the putative lower gate) destabilizes the closed gate and thus results in channel activation. Our findings suggest the existence of a lower constriction blocking MTS-ET access at the level of the X-gate in TASK-1 that is opened by 2-APB or oleoyl-CoA in TALK-2 channels. Accordingly, we observed that current activation and the rate of L145C modification concurrently increased with the 2-APB concentration levelling off at a concentration of 2.0 mM 2-APB that caused maximal current activation (Fig. 1f–h and Supplementary Table 1).

**Fig. 1: State-dependent modification of inner pore cysteine residues in TALK-2 K_2P channels.**

The lower constriction functions as a permeation gate

Our data strongly suggest a gated lower pore constriction site. However, whether this constriction is an actual permeation gate is still disputable, as 2-APB and oleoyl-CoA could have also opened the SF gate to cause activation. Furthermore, a constriction that blocks MTS-ET access may not necessarily block ion permeation. Our attempts to use silver ions (Ag⁺, which is much smaller than MTS-ET and similar in size to K⁺) to probe for a permeation gate were not conclusive as Ag⁺ also inhibited WT TALK-2 channels. Therefore, we addressed this issue with a different approach by taking advantage of the fact that Rb⁺ has a strong activating effect on the TALK-2 SF²⁰ and to do so, Rb⁺ needs to pass the lower constriction. And indeed, a stepwise increase in the 2-APB concentration resulted in a stepwise increase in Rb⁺ activation as if 2-APB activation had removed a constriction that prevented Rb⁺ to access the SF (Fig. 2a, c). Accordingly, the same effect on Rb⁺ activation was also observed when TALK-2 was activated by oleoyl-CoA (Supplementary Fig. 3a). As a control, we performed the same experiment with TREK-1 channels, since these K_2P channels lack a lower gate but also undergo activation upon 2-APB application (Fig. 1g). In TREK-1, Rb⁺ activation was the strongest in the absence of 2-APB activation, indicating that Rb⁺ ions had free access to the SF in the low activity state of the channel (Fig. 2b, c). Moreover, 2-APB activation progressively covered up Rb⁺ activation consistent with the concept that TREK-1 channels lack a lower gate and both activating stimuli (2-APB and Rb⁺) converge onto the SF gate (Fig. 2b). These results suggest that the lower constriction in TALK-2 channels functions as a permeation gate that must be open in addition to the SF gate to allow ion conduction (Fig. 2d).

Characterization of the lower gate by scanning mutagenesis and single-channel recordings

We performed systematic alanine scanning mutagenesis in the TM regions that contain the lower gates in TASK-1 and TASK-2 (i.e., TM2 and TM4) to identify mutations that would affect the stability of the lower gate in TALK-2 (Fig. 3a, b and Supplementary Fig. 4a). This approach identified four gain of function (g-o-f) mutations (V146A in TM2, W255A, L262A and L264A in TM4) that markedly increased TALK-2 basal currents 31.9 ± 3.4-fold (V146A), 4,5 ± 0.2-fold (W255A), 25.1 ± 2.1-fold (L262A), and 93.3 ± 9.8-fold (L264A) as assessed in two-electrode voltage-clamp (TEVC) experiments with oocytes (Fig. 3a). Mapping of the g-o-f residues on our TALK-2 homology models showed a clustering of the g-o-f residues in the region of the lower pore constrictions identified in TASK-1 and TASK-2 at the cytosolic pore entrance of the channels (Fig. 3c, Supplementary Fig. 4b). In the TASK-1-based TALK-2 homology model the two residues showing the strongest g-o-f phenotype (i.e., V146A and L264A) are in direct proximity and could form a permeation constriction (Fig. 3c). Further, the g-o-f mutant channel with the strongest effect (i.e., L264A) and WT TALK-2 channels were further analyzed with inside-out single channel recordings (Fig. 3d–h). As expected, the basal single channel activity of WT TALK-2 was very low with a relative channel-open probability (NP_O) of 3.9 ± 0.2 % and the L264A mutation resulted in a large increase to 46.7 ± 12.1 %, while the single-channel amplitude was not affected (Fig. 3d, e). The increase in NP_O appeared to primarily result from a destabilization of the closed state, as we observed a strong shortening of the short (~13-fold) and long closed times (~16-fold) (Fig. 3f), whereas the open times were comparatively little (~2-fold) affected (Fig. 3g). Overall, these changes result in a 29-fold increase of channel activity spent in a burst type mode with 767 ± 123 bursts/min for L264A channels compared to 27 ± 17 burst/min for WT channels (Fig. 3h). The observed shortening in closed times is consistent with the concept that the g-o-f mutations destabilize a permeation gate that now opens much more frequently. Further, if this destabilized permeation gate corresponds to the lower gate, we expect fast modification of the L145C mutation as indeed observed in L264A/L145C double mutant channels (Fig. 3i, j). Moreover, all four g-o-f mutants resulted in fast L145C modification (in the absence of ligand activation) (Fig. 3i–k) with rates that roughly correlated to the g-o-f effect (Fig. 3a) with L264A showing the fastest modification (Fig. 3i, k and Supplementary Table 2).

**Fig. 3: Functional characterization of the lower gate in TALK-2 K_2P channels.**

Gate coupling in TALK-2 channels

The existence of two activation gates in TALK-2 raises the question if they are coupled. To address this question, we tested a stimulus that directly affects the SF gate. Extracellular alkalinization is thought to open the SF gate in TALK-2 by deprotonation of a lysine residue (K242) located at the outer pore helix connected to the SF¹⁹ (Fig. 4b). Accordingly, raising the pH_e from 7.4 to 9.5 resulted in large TALK-2 currents (Fig. 4a and Supplementary Fig. 5a, b). Under this condition, we determined the MTS-ET accessibility of L145C and, surprisingly, observed a similar fast modification rate as seen with maximal (2.0 mM) 2-APB activation at pH_e 7.4 (Fig. 4c and Supplementary Table 1). However, raising the pH_e to 9.5 had little effect on the modification of Q266C consistent with the localization of this position at the cytoplasm-facing side of lower gate (Supplementary Fig. 5c and Supplementary Table 2). These results imply that opening the SF gate by high pH_e had also opened the lower gate in TALK-2.

**Fig. 4: Direct stimulation of the SF produces a state-dependent cysteine modification in the pore of TALK-2.**

Ion occupancy of the SF controls the opening of the lower gate

We have previously shown that the SF acts as an ion-flux voltage gate in many K_2P channels including TALK-2²⁰. According to this concept, the low basal activity of exclusively SF-gated K_2P channels – such as TREK-1 – results from an inactivated and ion-depleted filter. Upon depolarization, ions are forced into the SF by the transmembrane electric field to induce voltage-dependent channel activation. The electrical work necessary to open the SF is reflected in the conductance-voltage (G–V) curves obtained by plotting the relative open probability (tail current amplitudes) against the membrane pre-pulse voltage (Fig. 4d, e). Particularly strong ion activation is seen with intracellular Rb⁺ as this ion appears to stabilize the conductive state of the filter more efficiently than K⁺ (Fig. 4g, h and Supplementary Fig. 5d, e). By monitoring the modification of L145C mutant channels we tested whether opening of the SF with voltage and Rb⁺ would also affect the status of the lower gate (Fig. 4f, h, i). Indeed, activation of TALK-2 channels upon K⁺ by Rb⁺ replacement at a membrane potential of +40 mV allowed strong L145C MTS-ET modification (Fig. 4h and Supplementary Table 1). Thus, the mere change in SF ion occupancy upon exchanging K⁺ by Rb⁺ in the SF is sufficient to markedly increase the P_O of the lower gate allowing L145C modification (Fig. 4f). As expected, the L145C modification rate was strongly voltage-dependent and the increase in channel P_O reflected in the G–V curve mirrored the increase in modification rate and, thus, voltage-dependent opening of the SF gate concurrently opened also the lower gate (Fig. 4e, i).

The properties of the lower gate resulted in a state-dependent TALK-2 pharmacology

In voltage-gated K⁺ (K_v) channels, the closing of the lower gate can be prevented by the binding of blockers such as tetra-pentyl-ammonium (TPenA) in the pore cavity and, thereby, causing a slowing of the deactivation kinetics which is known as the ‘foot in the door’ effect^33,34. To test for such a mechanism in TALK-2, we activated the channels with a voltage step to +100 mV in the presence of intracellular Rb⁺ followed by a repolarization step to -80 mV (to induce tail currents) with and without the K_2P channel pore blocker TPenA. Remarkably, we observed a tail current cross-over indicating that the deactivation time course is slowed by TPenA inhibition; i.e., with ~50 % TPenA block of the tail current amplitudes we observed a ~ 2.5-fold slowing of the deactivation kinetics (Fig. 5a). We hypothesize, that the two gates are strongly coupled and, therefore, opening the SF gate by depolarization should also open the lower gate, which then consequently would allow blocker (like TPenA) binding within the pore cavity (Fig. 5a cartoon). Upon repolarization, however, the bound TPenA blocker obstructs closure of the lower gate and thereby, hinders the SF from closure/inactivation (gate coupling). In accordance, when we tested this protocol on TREK-2 K_2P channels, known to lack a lower gate, TPenA inhibition had no effect on the tail current kinetics indicating that here the SF gate is able to close unhindered with the pore blocker bound (Fig. 5b). Further, TPenA inhibition had likewise little effect on tail current kinetics of 2-APB-activated TALK-2 channels indicating that 2-APB preferentially opens the lower gate in TALK-2 (Supplementary Fig. 6a).

**Fig. 5: Open channel blocker show state-dependent pore accessibility and slowing of deactivation kinetics in TALK-2.**

These findings suggest that TPenA is a state-dependent blocker in TALK-2 channels and, thereby, the apparent affinity for TPenA should strongly depend on the fraction of channels being open (Fig. 5h). Indeed, in the basal state (low P_O) hardly any TPenA block was observed (Fig. 5c), whereas 2-APB activation induced dose-dependent TPenA inhibition (Fig. 5d–f). Accordingly, the apparent affinity for TPenA inhibition increased dramatically with the degree of 2-APB activation (IC₅₀ 0.2 mM 2-APB = 1108 ± 411 µM, IC₅₀ 0.5 mM 2-APB = 225 ± 25 µM, IC₅₀ 1.0 mM 2-APB = 54 ± 10 µM) (Fig. 5e). We further explored this observation by testing several other compounds known to block K_2P channels (e.g., TASK-1) such as A1899, AVE0118 and S9947³⁵. Remarkably, for each blocker tested little inhibition was seen for unstimulated TALK-2 channels while strong inhibition was observed upon activation by 2-APB (Fig. 5g). Furthermore, the g-o-f mutation L264A resulted in TALK-2 channels with high sensitivity to inhibition for all tested blockers without 2-APB activation further indicating that the mutation promoted opening of the lower gate (Fig. 5f, g and Supplementary Fig. 6b, c). In conclusion, the existence of the lower gate in TALK-2 transformed the state-independent pore block as seen in TREK-2 into a state-dependent (foot in the door-like) block as typical for K_v channels.

Opening of the lower gate reduces the mechanical load coupled to the voltage-powered SF gate

The voltage-dependent activation mechanism of the SF gate in K_2P channels allows to estimate the electrical work (ΔG = zFΔV_1/2) required for pore opening by fitting the G–V curve to a Boltzmann equation. A leftward shift of the G–V curve (without a change in slope) indicates a reduction in the free energy between the closed and open state of the channel. The open state represents the situation of both gates being simultaneously open but voltage can only exert force on the SF gate that functions as voltage sensor. However, when the two gates are energetically coupled then any circumstance that opens the lower gate should also reduce the electrical work to open the SF gate (Figs. 6 and 7). We tested this concept using TPenA that we have shown to hinder lower gate closure. Indeed, the presence of 1 mM TPenA caused a 46.1 ± 4.9 mV leftward shift of the G–V curve, while in TREK-2 K_2P channels lacking a lower gate the G–V curve was not affected in the presence of TpenA (Fig. 6a, b, f and Supplementary Table 4). Likewise, 2-APB and LC-CoA that we have shown to open the lower gate also caused a leftward shift of the G–V curve (Fig. 6f and Supplementary Fig. 7a). Further, stabilizing the open state of SF directly by increasing pH_e also shifted the G–V curve leftwards, thus, deprotonation of the pH sensor reduces the free energy difference between the closed and open state of the SF (Fig. 6d, e). This implies that low pH and hyperpolarization induce a similar closed state of the SF. We further tested whether mutations that open the lower gate (L262A and L264A) would affect the extracellular pH sensitivity. Indeed, both mutations shifted the pH-current relationship towards more neutral pH suggesting that opening of the lower gate promoted the deprotonated pH sensor state indicating allosteric coupling of the lower gate to extracellular pH sensor (Fig. 6g and Supplementary Fig. 7b). We further explored this concept of allosteric coupling using the mutations identified in the functional alanine screen of TM2 and TM4. Indeed, all mutations that produced a g-o-f effect also caused a leftward shift of the G–V curve (Figs. 3a and 7a, b and Supplementary Table 3). Further, the V_1/2 shift was correlated to the increase in the L145C modification rate (Fig. 7c). Thus, the degree (i.e., frequency) of lower gate opening cause by the g-o-f mutations was correlated to the reduction in electric work to open the SF gate. Accordingly, the largest effect on the G–V curve (i.e., an 84.0 ± 5.3 mV shift) was seen for the L264A mutation that had the largest g-o-f effect, as well as caused the fastest modification of L145C and, thus, the highest P_O of the lower gate (Figs. 3a, k and 7a–c). Actually, this mutation promoted the open state of the lower gate so strongly that 2-APB produced only a minor further current increase (Supplementary Fig. 6d), the speed of L145C MTS-ET modification at +40 mV was not further increased by Rb⁺ (Supplementary Fig. 6e) and TPenA has no marked effect on the G–V curve (Fig. 6c). Therefore, we conclude that the ~84 mV shift in V_1/2 produced by the L264A mutant might roughly represent the mechanical load of the conformational change that the voltage-powered SF gate has to move for opening the lower gate.

**Fig. 6: Impact of ligand modulation on SF energetics in TALK-2 K_2P channels.**

**Fig. 7: Functional coupling of the SF and the lower gate in TALK-2 K_2P channels.**

Discussion

The present study on TALK-2 K_2P channels provides several lines of evidence suggesting the existence of a lower permeation gate at the cytoplasmic pore entrance in TALK-2. Firstly, access of cysteine modifying reagents to the inner pore cavity is blocked in closed TALK-2 channels, but possible in the presence of activating ligands, suggesting a pore entrance constriction. Furthermore, systematic alanine scanning mutagenesis identified several g-o-f mutations in distal parts of TM2 and TM4 that not only strongly activated TALK-2 but also removed the entrance constriction. Employing TALK-2 homology models based on the TASK-1 and TASK-2 structures revealed that the g-o-f mutations cluster in a region corresponding to the lower gates identified in these channels. Finally, using Rb⁺ as a probe we show that the identified lower constriction is actually a permeation gate.

Tight gate coupling defines the gating behavior in TALK-2 channels

In addition to the lower gate, TALK-2 channels are also operated by a SF gate similar to many other K_2P channels. We used three assays to dissect the status of the two gates and their coupling in the context of various gating stimuli: (i) the modification rate of L145C reports about the relative P_O of the lower gate, (ii) the G–V curves report on the electrical work required to open the SF filter gate and (iii) the current amplitudes report on the fraction of channels with both gates open (overall channel P_O). We found that the three parameters were strongly correlated in all g-o-f mutations suggesting a tight positive coupling of the two gates (Fig. 7d). This concept is also supported by a number of additional findings reported here. Opening the SF by high pH_e or voltage, likewise, opened the lower gate (i.e., increased L145C modification rate). Reciprocally, ligands that opened the lower gate (i.e., increased the rate of L145C modification) such as 2-APB or LC-CoA also positively shifted the G–V curve (opening of the SF gate). Particular striking was the observation that exchange of the permeating ion from K⁺ to Rb⁺ caused a large increase in the L145C modification rate indicating that the mere difference in SF ion occupancy is sufficient to induce a marked structural change at the pore entrance (i.e., opening the lower gate). In structural terms the strong positive gate coupling can be envisioned as rigid connection of the two gates within the protein structure and, thus, opening/closing of one gate also forces the other gate to open/close.

How is the lower gate coupled to the selectivity filter?

We speculate that movement of TM4 underlies lower gate opening as well as force transduction onto the SF gate to induce its opening in TALK-2. Thus, the TM4 might represent the rigid connection linking the two gates responsible for the tight gate coupling. This hypothesis appears appealing as a similar mechanism has been shown to couple the lower gate to the SF gate in MthK (prokaryotic K⁺ channel from Methanobacterium thermoautotrophicum), based on MD simulations performed with different opening diameters at the lower gate³⁶. Here an isoleucine (I84) at the end of TM2 (corresponding to TM4 in TALK-2) was shown to exert force on a threonine (T59) of the SF to promote the conductive SF state. Although members of the TREK/TRAAK subfamily lack a functional lower gate^28,29, the TM4 also moves during temperature-, mechano- or lipid-induced gating (known as the down- to up-state transition^37,38) and this movement is also thought to open the SF gate^25,26. Furthermore, also the lower X-gate in TASK-1 is formed by TM4 and, thus, envisioned to move during gating³¹. Therefore, movement of the TM4 segments coupled to the SF gate might be the general theme in K_2P channel gating (as well as other K⁺ channels). In some channels this movement may result in the formation of a lower gate (e.g., in TASK-1, TASK-2, and TALK-2) and in other channels not (e.g., in TREK-1/-2 and TRAAK), while ultimately the SF is affected in the activation process. How this coupling is realized in atomic detail in TALK-2 warrant future studies and in particular high resolution structural information of different states.

Physiological relevance of positive gate coupling in TALK-2

The existence of two gates that open and close in a strongly coupled (i.e., concurrently) fashion seams redundant as closure of one gate is sufficient to prevent permeation. However, in TALK-2 channels the two gates have evolved to respond to different stimuli with the SF gate sensing membrane voltage and extracellular pH whereas the lower gate responds to negatively-charged lipids of the cytoplasmic membrane leaflet. Thus, for the individual gate-specific stimuli to become effective strong gate coupling is mandatory as opening of one gate would be functionally silent if the other gate stays closed. Interestingly, this stays in contrast to the multi-sensory gating behavior of the related TREK/TRAAK K_2P channel subgroup, where all stimuli (including temperature, voltage, lipids, and pH) are converged directly to the SF representing their only gate.

Implications of the state-dependent pore access for TALK-2 gating and pharmacology

Functionally, the lower gate in TALK-2 resembles the helix-bundle crossing gate in voltage-gated K_v channels. In K_v channels voltage is thought to open the helix-bundle crossing gate via the voltage-dependent movement of segment 4 (S4)^39,40. By marked contrast, in K_2P channels, voltage opens the SF gate via a voltage-dependent ion binding step²⁰ that forces the filter in its conductive state. In TALK-2 K_2P channels this structural change in the SF appears also to force the lower gate open as apparent in voltage-dependent modification rate of L145C. Upon repolarization, the ion-flux inversion is thought to change the SF ion occupancy leading to its inactivation as visible in the fast reduction of the tail current amplitudes²⁰. Intriguingly, in the presence of the pore blocker TPenA the decay of the current is slowed resembling the ‘foot in the door’ effect seen in K_v channels^33,34. This suggests that blocker binding to the pore cavity prevented the lower gate from closing. Importantly, this also suggests that TPenA concurrently prevented the SF gate from closing as blocker unbinding from a channel with a non-conductive SF would be electrophysiological invisible and, thus, would not produce the apparent slowing of deactivation. In agreement, in TREK-2 channels (lacking the lower gate) TPenA had no effect on the tail current kinetics indicating that the SF gate can close unhindered with the blocker bound upon inversion of the electric field upon repolarization. What stops now the SF gate from closing with TPenA bound in TALK-2? In K_v channels blocker binding to the pore cavity prevents lower gate closure (the actual structural change is still unknown) suggesting that the pore cavity changes its structure (and possibly its size) in concert with the lower gate closure and this change is hindered (or delayed) by the blocker^41,42. Thus, we presume that a similar structural change is also occurring in TALK-2 and is also prevented by TPenA. Consequently, because the SF gate and the lower gate are tightly coupled, TPenA also disturbs the SF gate in the closing reaction. In further agreement, we found that TPenA caused a leftward shift of the G–V curve as expected for a blocker that only binds to the pore with both gates open. In single-channel recordings we estimated the basal activity (i.e., the NP_O) for WT TALK-2 to about 4 % and, thus, only a small fraction of channels would be sensitive to a pore blocker if only open channels can bind the compound (Fig. 5h). Indeed, we observed very little current inhibition when TPenA was applied to TALK-2 channels in absence of an activating stimulus, but strong inhibition was observed for activated TALK-2 channels. This behavior was also seen for various other small molecule pore blockers, thus, state-dependent pore inhibition appears to be a defining feature of the TALK-2 channel pharmacology as this was so far not seen in any other K_2P channel.

Gate coupling in other K⁺ channels: differences and similarities

The established gating cycles of K_v and KcsA (prokaryotic K⁺ channel from the soil bacterium Streptomyces lividans) channels suggest that the activation (lower) gate serves as the primary ion permeation barrier that needs to open to allow current flow, while in a following step the SF inactivates to terminate ion permeation^{39,40,43,44,45}. Thus, in these cases the two gates are coupled sequentially and, in a negative manner⁴⁵. However, it is currently unknown whether the SF is open (conductive) or closed (non-conductive) in the resting (not activated) state of K_v or KcsA channels because it is not possible to directly measure the conductivity of the SF in a closed channel. In TALK-2 K_2P channels the SF serves as a gate as well as the voltage sensor. The latter property provides information on stability of the conductive state of the filter that can be extracted from the V_1/2 value of the corresponding G–V curve. Here, we show that when the lower gate is mostly closed then it requires a large amount of electrical work to open the SF, i.e., a membrane depolarization to 72 ± 2 mV (V_1/2) for half maximal voltage activation. However, when the lower gate is mostly open as seen in the L264A TALK-2 mutant channel this work is strongly reduced (V_1/2 = −12 ± 2 mV) and, thus, might approximately reflect the work required to open the lower gate. Actually, even very negative potentials cannot close the SF gate to a large degree as seen by the pedestal for the relative P_O that levels off at ~0.4 in TALK-2 L264A channels (Fig. 7b). Notably, these results closely resemble the constitutively open phenotype in Shaker K⁺ channels seen with mutations at the assumed hydrophobic seal of the helix-bundle crossing gate. These mutations also result in strong positively shifted G–V curves and high pedestal P_Os even at very negative potentials⁴⁶. Interestingly, the two residues with the strongest g-o-f effect (V146A and L264A) in TALK-2 are also hydrophobic and in direct proximity (according to our TASK-1 based homology model) and, thus, might also form a hydrophobic seal⁴⁷ that, thereby, could represent the lower gate in TALK-2. Intriguingly, a recent MD simulation study on KcsA suggest that the SF might be non-conductive when the lower (activation) gate is closed as this also implies positive coupling of the two gates in this archetypical K⁺ channel⁴⁷. Here, using electrophysiological means, we have demonstrated directly that the SF gate in TALK-2 K_2P channels opens and closes in concerts with the lower gate. It will be interesting to see whether this concept also applies to other K⁺ channels possessing two gates such as members of the large family of voltage-gated K⁺ channels.

Methods

Molecular biology

In this study the coding sequences of human K_2P2.1 TREK-1 (GenBank accession number: NM_172042), human K_2P10.1 TREK-2 (NM_021161) and human K_2P17.1 TALK-2 (EU978944.1)/human K_2P17.1 TASK-4 (NM_031460.3) were used. For K⁺ channel constructs expressed in oocytes the respective K⁺ channel subtype coding sequences were subcloned into the oocyte expression vector pSGEM or the dual-purpose vector pFAW which can be used for HEK293 cell expression as well and verified by sequencing. All mutant channels were obtained by site-directed mutagenesis with custom oligonucleotides. Vector DNA was linearized with NheI or MluI and cRNA synthesized in vitro using the SP6 or T7 AmpliCap Max High Yield Message Maker Kit (Cellscript, USA) or HiScribe® T7 ARCA mRNA Kit (New England Biolabs) and stored at −20 °C (for frequent use) and −80 °C (for long term storage).

Electrophysiological recordings in oocytes

Two-electrode voltage-clamp (TEVC) measurements

Electrophysiological studies were performed using the TEVC technique in oocytes. Ovarian lobes were obtained from frogs anesthetized with tricaine. Lobes were treated with collagenase (2 mg/ml, Worthington, type II) in OR2 solution containing (in mM): 82.5 NaCl, 2 KCl, 1 MgCl₂, 5 HEPES (pH 7.4 adjusted with (NaOH/HCl) for 2 h. Isolated oocytes were stored at 18 °C in ND96 recording solution (in mM): 96 NaCl, 2 KCl, 1.8 CaCl₂, 1 MgCl₂, 5 HEPES (pH 7.5 adjusted with NaOH/HCl) supplemented with Na-pyruvate (275 mg/l), theophylline (90 mg/l), and gentamicin (50 mg/l). Oocytes were injected with 50 nl of cRNA for WT or mutant TALK-2 and incubated for 2 days at 18 °C. Standard TEVC measurements were performed at room temperature (21 - 22 °C) with an Axoclamp 900 A amplifier, Digidata 1440 A, and pClamp10 software (Axon Instruments, Molecular Devices, LLC, USA). Microelectrodes were fabricated from glass pipettes, back-filled with 3 M KCl, and had a resistance of 0.2–1.0 MΩ.

Inside-out patch-clamp measurements

Oocytes were surgically removed from anesthetized adult females, treated with type II collagenase (Sigma-Aldrich/Merck, Germany) and manually defolliculated. 50 nl of a solution containing the K⁺ channel specific cRNA was injected into Dumont stage V - VI oocytes and subsequently incubated at 17 °C in a solution containing (mM): 54 NaCl, 30 KCl, 2.4 NaHCO₃, 0.82 MgSO₄ x 7 H₂O, 0.41 CaCl₂, 0.33 Ca(NO₃)₂ x 4 H₂O and 7.5 TRIS (pH 7.4 adjusted with NaOH/HCl) for 1–7 days before use. Electrophysiological recordings: Excised patch measurements in inside-out configuration under voltage-clamp conditions were performed at room temperature (22 – 24 °C). Patch pipettes were made from thick-walled borosilicate glass GB 200TF-8P (Science Products, Germany), had resistances of 0.2–0.5 MΩ (tip diameter of 10 - 25 µm) and filled with a pipette solution (in mM): 120 KCl, 10 HEPES and 3.6 CaCl₂ (pH 7.0 - 8.5 adjusted with KOH/HCl) or 120 KCl, 10 AMPSO and 3.6 CaCl₂ (pH 9.0 - 10.5 adjusted with KOH/HCl). Intracellular bath solutions and compounds were applied to the cytoplasmic side of excised patches for the various K⁺ channels via a gravity flow multi-barrel pipette system. Intracellular solution had the following composition (in mM): 120 KCl, 10 HEPES, 2 EGTA and 1 Pyrophosphate (pH adjusted with KOH/HCl). In other intracellular bath solutions, K⁺ was replaced by Rb⁺ (pH 7.4 adjusted with RbOH/HCl). Currents were recorded with an EPC10 amplifier (HEKA electronics, Germany) and sampled at 10 kHz or higher and filtered with 3 kHz (-3 dB) or higher as appropriate for sampling rate.

Inside-out single channel patch-clamp measurements

Single channel patch-clamp measurements in the inside-out configuration were performed under voltage-clamp conditions with oocytes. Briefly, the vitelline membranes of the oocytes were manually removed after shrinkage by adding mannitol to the bath solution. All experiments were conducted at room temperature (21 - 22 °C) 1–2 days after injection of 50 nl TALK-2 cRNA. Borosilicate glass capillaries GB 150TF-8P (Science Products, Germany) were pulled with a DMZ-Universal Puller (Zeitz Instruments, Germany) and had a resistance of 4–6 MΩ when filled with pipette solution containing (in mM): 120 KCl, 10 HEPES, 3.6 CaCl₂ (pH 8.5 adjusted with KOH/HCl). The bath solution had the following composition (in mM): 120 KCl, 10 HEPES, 2 EGTA, and 1 Pyrophosphate (pH 8.5 adjusted with KOH/HCl). Gap-free voltage pulses of −100 mV were continuously applied. Single channel currents were recorded with an Axopatch 200B amplifier, a Digidata 1550B A/D converter and pClamp10 software (Axon Instruments, Molecular Devices, LLC, USA) and were sampled at 15 kHz with the analog filter set to 5 kHz. Additionally, data was digitally filtered by 2 kHz (Lowpass, Bessel) with ClampFit10 before analysis. Data were analyzed with ClampFit10 and Origin 2016 (OriginLab Corporation, USA). The single channel search tool of the ClampFit10 software was used to identify and analyze the single channel events in a time frame of 60 s. A group of single channel events (minimum 3 events) has been classified as a burst, if the time between two single channel openings were lower than five times of the short-closed time (5 x t_C1).

Animals

The investigation conforms to the guide for the Care and Use of laboratory Animals (NIH Publication 85-23). For this study, sixty female Xenopus laevis animals were used to isolate oocytes. Experiments using Xenopus toads were approved by the local ethics commission.

Drugs, chemical compounds, and bioactive lipids

2-aminoethoxydiphenyl borate (2-APB) (Sigma-Aldrich/Merck, Germany), BL-1249, A1899 (Tocris Bioscience, Germany), AVE0118, S9947 (Axon Medchem, Germany) and oleoyl-CoA (LC-CoA 18:1) (Avanti Polar Lipids, USA) were prepared as stocks (1 - 100 mM) in DMSO, stored at −80 °C and diluted to the final concentration in the intracellular recording solution. Tetra-pentyl-ammonium chloride (TPenA) (Sigma-Aldrich/Merck, Germany) and (2-(Trimethylammonium)ethyl) MethaneThioSulfonate Chloride (MTS-ET) (Toronto Research Chemicals, USA) were directly dissolved to the desired concentration in the intracellular recording solution prior to each experiment. MTS-ET was used immediately after dilution for maximally 5 min.

Homology modeling - TALK-2 models based on TASK-1 or TASK-2

Human TASK-1 crystallographic structure (PDB ID: 6RV3, chains A, B)³¹ and mouse TASK-2 cryo-electron microscopy (Cryo-EM) structure (PDB ID: 6WLV, chains A, B)³⁰ were used as templates to build the TALK-2 models. First, the structures were prepared using ‘Protein preparation wizard’ module of the Schrödinger’s suite software (Protein Preparation Wizard; Epik, Schrödinger, LLC, New York, NY, 2018-4; Impact, Schrödinger, LLC, New York, NY, 2018-4; Prime, Schrödinger, LLC, New York, NY, 2018-4). Charges and parameters were assigned according to the force field OPLS-2005⁴⁸. The missing residues in TASK-1 (149 - 151 from chain A and 150 - 151 from chain B) were modeled using ‘crosslink protein’ tool from the Schrödinger suite. The sequence of TALK-2 (KCNK17) protein was obtained from the GenBank database⁴⁹. Alignments between the template protein (TASK-1 or TASK-2) sequences and the TALK-2 sequence were performed using the Smith-Waterman algorithm⁵⁰ with BLOSUM62⁵¹ matrix for scoring the alignment. The modeling was carried out using the BioLuminate⁵² module of the Schrödinger suite. The standard modeling protocol was used, which consists of a search for rotamers for non-conserved residues and loops to eliminate clashes and then an energetic minimization with the OPLS2005 force field in vacuum. The protonation states of the TALK-2 models were predicted at a pH of 7.0 with PROPKA⁵³, and another energetic minimization of the hydrogen atoms was performed using the conjugate gradient method⁵⁴ implemented in the Schrödinger suite. The sequence of the TALK-2 model based on 6RV3 extends from residue R15 to K282 and based on 6WLV from residue T22 to H278, respectively. It was verified with the procheck⁵⁵ software that 91.2 % and 93,5 % of the residues are in the most favored regions for the models based on 6RV3 and 6WLV, respectively.

Alignments:

6RV3: MKRQNVRTLALIVCTFTYLLVGAAVFDALESEPELIERQRLELRQQELRARYN ~ LSQGGYEELERVVLRLKPHKAGVQ ~ ~ ~ ~ ~ ~ ~ ~ WRFAGSFYFAITVITTIGYGHAAPSTDGGKVFCMFYALLGIPLTLVMFQSLGERINTLVRYLLHRAKKGLGMRRADVSMANMVLIGFFSCISTLCIGAAAFSHYEHWTFFQAYYYCFITLTTIGFGDYVALQKDQALQTQPQYVAFSFVYILTGLTVIGAFLNLVVLRFMTMNAEDEKRDAENL

TALK-2 (TASK-4): RGCAVPSTVLLLLAYLAYLALGTGVFWTLEGRAAQDSSRSFQRDKWELLQNFTCLDRPALDSLIRDVVQAYKNGASLLSNTTSMGRWELVGSFFFSVSTITTIGYGNLSPNTMAARLFCIFFALVGIPLNLVVLNRLGHLMQQGVNHWASRLGGTWQDPDKARWLAGSGAL ~ LSGLLLFLLLPPLLFSHMEGWSYTEGFYFAFITLSTVGFGDYVIG ~ MNPSQRYPLWYKNMVSLWILFGMAWLALIIKLILSQLETPGRVCSCCHHSSK

6WLV: GPLLTSAIIFYLAIGAAIFEVLEEPHWKEAKKNYYTQKLHLLKEFPCLSQEGLDKILQVVSDAADQGVAITGNQT ~ FNNWNWPNAMIFAATVITTIGYGNVAPKTPAGRLFCVFYGLFGVPLCLTWISALGKFFGGRAKRLGQFLTRRGVSLRKAQITCTAIFIVWGVLVHLVIPPFVFMVTEEWNYIEGLYYSFITISTIGFGDFVAGVNPSANYHALYRYFVELWIYLGLAWLSLFVNWKVSMFVEVHKAIKKRR

TALK-2 (TASK-4): TVLLLLAYLAYLALGTGVFWTLEGRAAQDSSRSFQRDKWELLQNFTCLDRPALDSLIRDVVQAYKNGASLLSNTTSMGRWELVGSFFFSVSTITTIGYGNLSPNTMAARLFCIFFALVGIPLNLVVLNRLGHLMQQGVNHWASRLGGTWQDPDKARWLAGSGALLSGLLLFLLLPPLLFSHMEGWSYTEGFYFAFITLSTVGFGDYVIGMNPSQRYPLWYKNMVSLWILFGMAWLALIIKLILSQLETPGRVCSCCH

Cysteine mutations were introduced via PyMOL after the respective model was built.

Data acquisition and statistical analysis

Data analysis and statistics were done using Fitmaster (HEKA electronics, version: v2x73.5, Germany), Microsoft Excel 2021 (Microsoft Corporation, USA), and Igor Pro 9 software (WaveMetrics Inc., USA). Recorded currents were analyzed from stable membrane patches at a voltage defined in the respective figure legend or with a voltage protocol as indicated in the respective figure. The fold activation (fold change in (tail) current amplitude) of a ligand (drug or bioactive lipid) was calculated from Eq. 1:

$${{\rm{Fold\; activation}}}\left({{\rm{FA}}}\right)=\frac{{{{I}}}_{{{\rm{activated}}}}}{{{{I}}}_{{{\rm{basal}}}}}$$

(1)

where I_activated represents the stable current level in the presence of a given concentration of a respective ligand and I_basal the measured current before ligand application. Percentage inhibition or residual currents upon blocker application for a ligand (drug or bioactive lipid) was calculated from stable currents of excised membrane patches using the following Eqs. 2 and 3:

$$\%{{\rm{inhibition}}}=\left(1-\left(\frac{{{{I}}}_{{{\rm{inhibited}}}}}{{{{I}}}_{{{\rm{basal}}}}}\right)\right) * 100$$

(2)

$${{\rm{Residual\; current}}}=100-\%\,{{\rm{inhibition}}}$$

(3)

where I_inhibited refers to the stable current level recorded in the presence of a given concentration of a drug or bioactive lipid and I_basal to the measured current before ligand application. The macroscopic half-maximal concentration-inhibition relationship of a ligand was obtained using a Hill-fit for dose-response curves as depicted in Eq. 4:

$$\%\, {{\rm{inhibition}}}/{{\rm{activation}}}=\,\frac{{I}_{{base}}+\left({I}_{\max }-\,{I}_{{base}}\right)}{\left\{{1+\left[\frac{{x}_{1/2}}{x}\right]}^{{rate}}\right\}}$$

(4)

where base and max are the currents in the absence and presence of a respective ligand, x is the concentration of the ligand, x_1/2 is the ligand concentration at which the activatory or inhibitory effect is half-maximal, rate is the Hill coefficient. For analysis of activation and deactivation time constants (τ) as well as the time constants of MTS modification (τ) current traces were fitted with a mono-exponential equation as depicted in 5:

$${{y}_{0}+A}^{\left\{\frac{-\left(\right.x-{x}_{0}}{\tau }\right\}}$$

(5)

Conductance-voltage (G–V) relationships were determined from tail currents recorded at a holding potential (V_H) of −80 mV after 300 ms depolarizing steps as indicated in the respective figure legend. Data were analyzed with a single Boltzmann fit following Eq. 6:

$$G\left(V\right)=\frac{{I}_{\max }}{{1+e}^{\frac{\left(V-{V}_{1/2}\right)}{{slope}}}}+{I}_{{base}}$$

(6)

where V_1/2 represents the voltage of half-maximal activation, s is the slope factor and I_max and I_base represent the upper and lower asymptotes. Under appropriate experimental conditions, you can use slope to calculate the valence (charge) of the ion moving across the channel. Slope equals R * T/ z * F where R is the universal gas constant, T is temperature in K, F is the Faraday constant, and z is the valence.

Throughout the manuscript all values are represented as mean ± s.e.m. with n indicating the number of individual executed recordings of single patches or oocytes. Data from independent measurements (biological replicates) were normalized and fitted independently to facilitate averaging. Statistical significance between two groups (respective datasets) was validated using an unpaired, two-sided t-test. The results of statistical analyses are presented as follows: *P ≤ 0.05, **P ≤ 0.01 and ***P ≤ 0.001. The exact P values are indicated in the figures. Zero current levels were indicated using dotted lines in all figures. Image processing and figure design was done using Igor Pro 9 (64 bit) (WaveMetrics, Inc., USA), PyMOL 2.4.1 (Schrödinger, LLC), and Canvas X Draw (Version 20 Build 544) (ACD Systems, Canada).

Reporting summary

Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article.

Data availability

Data supporting the findings of this manuscript are available from the corresponding authors upon request. The source data underlying Figs. 1g, h, 2c, 3a, 3e–h, 3k, 4e, 4i, 5e–g, 6a–g, and 7a–c, and Supplementary Figs. 1c, e, g, 4a, 5a, 5d, 6c, d, and 7 are provided as Source data file. The accession codes for the PDB structures used to build the models are PDB ID: 6RV3 and 6WLV. Source data are provided with this paper.

References

Decher, N. et al. Characterization of TASK-4, a novel member of the pH-sensitive, two-pore ___domain potassium channel family. FEBS Lett. 492, 84–89 (2001).
Article CAS PubMed Google Scholar
Girard, C. et al. Genomic and functional characteristics of novel human pancreatic 2P ___domain K(+) channels. Biochem. Biophys. Res. Commun. 282, 249–256 (2001).
Article CAS PubMed Google Scholar
Schmidt, C. et al. Upregulation of K(2P)3.1 K+ current causes action potential shortening in patients with chronic atrial fibrillation. Circulation 132, 82–92 (2015).
Article CAS PubMed Google Scholar
Trapp, S., Aller, M. I., Wisden, W. & Gourine, A. V. A role for TASK-1 (KCNK3) channels in the chemosensory control of breathing. J. Neurosci. 28, 8844–8850 (2008).
Article CAS PubMed PubMed Central Google Scholar
Leithner, K. et al. TASK-1 regulates apoptosis and proliferation in a subset of non-small cell lung cancers. PLoS ONE 11, e0157453 (2016).
Article PubMed PubMed Central Google Scholar
Sirois, J. E., Lei, Q., Talley, E. M., Lynch, C. 3rd & Bayliss, D. A. The TASK-1 two-pore ___domain K+ channel is a molecular substrate for neuronal effects of inhalation anesthetics. J. Neurosci. 20, 6347–6354 (2000).
Article CAS PubMed PubMed Central Google Scholar
Dadi, P. K., Vierra, N. C. & Jacobson, D. A. Pancreatic beta-cell-specific ablation of TASK-1 channels augments glucose-stimulated calcium entry and insulin secretion, improving glucose tolerance. Endocrinology 155, 3757–3768 (2014).
Article PubMed PubMed Central Google Scholar
Dadi, P. K., Luo, B., Vierra, N. C. & Jacobson, D. A. TASK-1 potassium channels limit pancreatic alpha-cell calcium influx and glucagon secretion. Mol. Endocrinol. 29, 777–787 (2015).
Article CAS PubMed PubMed Central Google Scholar
Marionneau, C. et al. Distinct cellular and molecular mechanisms underlie functional remodeling of repolarizing K+ currents with left ventricular hypertrophy. Circ. Res. 102, 1406–1415 (2008).
Article CAS PubMed PubMed Central Google Scholar
Friedrich, C. et al. Gain-of-function mutation in TASK-4 channels and severe cardiac conduction disorder. EMBO Mol. Med. 6, 937–951 (2014).
Article CAS PubMed PubMed Central Google Scholar
Ma, Q. et al. The rs10947803 SNP of KCNK17 is associated with cerebral hemorrhage but not ischemic stroke in a Chinese population. Neurosci. Lett. 539, 82–85 (2013).
Article CAS PubMed Google Scholar
He, L. et al. Association of variants in KCNK17 gene with ischemic stroke and cerebral hemorrhage in a Chinese population. J. Stroke Cerebrovasc. Dis. 23, 2322–2327 (2014).
Article PubMed Google Scholar
Domingues-Montanari, S. et al. KCNK17 genetic variants in ischemic stroke. Atherosclerosis 208, 203–209 (2010).
Article CAS PubMed Google Scholar
Ding, H. et al. Confirmation of genomewide association signals in Chinese Han population reveals risk loci for ischemic stroke. Stroke 41, 177–180 (2010).
Article PubMed Google Scholar
Schmidt, C. et al. Inverse remodelling of K(2P)3.1 K+ channel expression and action potential duration in left ventricular dysfunction and atrial fibrillation: implications for patient-specific antiarrhythmic drug therapy. Eur. Heart J. 38, 1764–1774 (2017).
CAS PubMed Google Scholar
Chai, S. et al. Physiological genomics identifies genetic modifiers of long QT syndrome type 2 severity. J. Clin. Invest. 128, 1043–1056 (2018).
Article PubMed PubMed Central Google Scholar
Varshney, A. et al. Genetic regulatory signatures underlying islet gene expression and type 2 diabetes. Proc. Natl Acad. Sci. USA 114, 2301–2306 (2017).
Article ADS CAS PubMed PubMed Central Google Scholar
Duprat, F., Girard, C., Jarretou, G. & Lazdunski, M. Pancreatic two P ___domain K+ channels TALK-1 and TALK-2 are activated by nitric oxide and reactive oxygen species. J. Physiol. 562, 235–244 (2005).
Article CAS PubMed Google Scholar
Niemeyer, M. I. et al. Neutralization of a single arginine residue gates open a two-pore ___domain, alkali-activated K+ channel. Proc. Natl Acad. Sci. USA 104, 666–671 (2007).
Article ADS CAS PubMed Google Scholar
Schewe, M. et al. A non-canonical voltage-sensing mechanism controls gating in K2P K(+) channels. Cell 164, 937–949 (2016).
Article CAS PubMed PubMed Central Google Scholar
Riel, E. B. et al. The versatile regulation of K2P channels by polyanionic lipids of the phosphoinositide and fatty acid metabolism. J. Gen. Physiol. 154, e202112989 (2022).
Lesage, F. et al. Dimerization of TWIK-1 K+ channel subunits via a disulfide bridge. EMBO J. 15, 6400–6407 (1996).
Article CAS PubMed PubMed Central Google Scholar
Goldstein, S. A., Bockenhauer, D., O’Kelly, I. & Zilberberg, N. Potassium leak channels and the KCNK family of two-P-___domain subunits. Nat. Rev. Neurosci. 2, 175–184 (2001).
Article CAS PubMed Google Scholar
Feliciangeli, S., Chatelain, F. C., Bichet, D. & Lesage, F. The family of K channels: salient structural and functional properties. J. Physiol. https://doi.org/10.1113/jphysiol.2014.287268 (2014).
Bagriantsev, S. N., Peyronnet, R., Clark, K. A., Honore, E. & Minor, D. L. Jr. Multiple modalities converge on a common gate to control K2P channel function. EMBO J. 30, 3594–3606 (2011).
Article CAS PubMed PubMed Central Google Scholar
Bagriantsev, S. N., Clark, K. A. & Minor, D. L. Jr. Metabolic and thermal stimuli control K(2P)2.1 (TREK-1) through modular sensory and gating domains. EMBO J. 31, 3297–3308 (2012).
Article CAS PubMed PubMed Central Google Scholar
Cohen, A., Ben-Abu, Y. & Zilberberg, N. Gating the pore of potassium leak channels. Eur. Biophys. J. 39, 61–73 (2009).
Article CAS PubMed Google Scholar
Piechotta, P. L. et al. The pore structure and gating mechanism of K2P channels. EMBO J. 30, 3607–3619 (2011).
Article CAS PubMed PubMed Central Google Scholar
Rapedius, M. et al. State-independent intracellular access of quaternary ammonium blockers to the pore of TREK-1. Channels 6, 473–478 (2012).
Article CAS PubMed PubMed Central Google Scholar
Li, B., Rietmeijer, R. A. & Brohawn, S. G. Structural basis for pH gating of the two-pore ___domain K(+) channel TASK2. Nature 586, 457–462 (2020).
Article ADS CAS PubMed PubMed Central Google Scholar
Rodstrom, K. E. J. et al. A lower X-gate in TASK channels traps inhibitors within the vestibule. Nature 582, 443–447 (2020).
Article ADS CAS PubMed Google Scholar
Schewe, M. et al. A pharmacological master key mechanism that unlocks the selectivity filter gate in K(+) channels. Science 363, 875–880 (2019).
Article ADS CAS PubMed PubMed Central Google Scholar
Choi, K. L., Mossman, C., Aube, J. & Yellen, G. The internal quaternary ammonium receptor site of Shaker potassium channels. Neuron 10, 533–541 (1993).
Article CAS PubMed Google Scholar
Armstrong, C. M. Inactivation of the potassium conductance and related phenomena caused by quaternary ammonium ion injection in squid axons. J. Gen. Physiol. 54, 553–575 (1969).
Article CAS PubMed PubMed Central Google Scholar
Kiper, A. K. et al. Kv1.5 blockers preferentially inhibit TASK-1 channels: TASK-1 as a target against atrial fibrillation and obstructive sleep apnea? Pflug. Arch.: Eur. J. Physiol. 467, 1081–1090 (2015).
Article CAS Google Scholar
Kopec, W., Rothberg, B. S. & de Groot, B. L. Molecular mechanism of a potassium channel gating through activation gate-selectivity filter coupling. Nat. Commun. 10, 5366 (2019).
Article ADS PubMed PubMed Central Google Scholar
Dong, Y. Y. et al. K2P channel gating mechanisms revealed by structures of TREK-2 and a complex with Prozac. Science 347, 1256–1259 (2015).
Article ADS CAS PubMed PubMed Central Google Scholar
McClenaghan, C. et al. Polymodal activation of the TREK-2 K2P channel produces structurally distinct open states. J. Gen. Physiol. 147, 497–505 (2016).
Article CAS PubMed PubMed Central Google Scholar
Grizel, A. V., Glukhov, G. S. & Sokolova, O. S. Mechanisms of activation of voltage-gated potassium channels. Acta Nat. 6, 10–26 (2014).
Article CAS Google Scholar
Kim, D. M. & Nimigean, C. M. Voltage-gated potassium channels: a structural examination of selectivity and gating. Cold Spring Harb. Perspect. Biol. https://doi.org/10.1101/cshperspect.a029231 (2016).
Holmgren, M., Smith, P. L. & Yellen, G. Trapping of organic blockers by closing of voltage-dependent K+ channels: evidence for a trap door mechanism of activation gating. J. Gen. Physiol. 109, 527–535 (1997).
Article CAS PubMed PubMed Central Google Scholar
Yellen, G. The moving parts of voltage-gated ion channels. Q. Rev. Biophys. 31, 239–295 (1998).
Article CAS PubMed Google Scholar
Cuello, L. G. et al. Structural basis for the coupling between activation and inactivation gates in K(+) channels. Nature 466, 272–275 (2010).
Article ADS CAS PubMed PubMed Central Google Scholar
Cuello, L. G., Cortes, D. M. & Perozo, E. The gating cycle of a K(+) channel at atomic resolution. Elife https://doi.org/10.7554/eLife.28032 (2017).
Panyi, G. & Deutsch, C. Probing the cavity of the slow inactivated conformation of shaker potassium channels. J. Gen. Physiol. 129, 403–418 (2007).
Article CAS PubMed PubMed Central Google Scholar
Sukhareva, M., Hackos, D. H. & Swartz, K. J. Constitutive activation of the Shaker Kv channel. J. Gen. Physiol. 122, 541–556 (2003).
Article CAS PubMed PubMed Central Google Scholar
Kitaguchi, T., Sukhareva, M. & Swartz, K. J. Stabilizing the closed S6 gate in the Shaker Kv channel through modification of a hydrophobic seal. J. Gen. Physiol. 124, 319–332 (2004).
Article CAS PubMed PubMed Central Google Scholar
Kaminski, G. A., Friesner, R. A., Tirado-Rives, J. & Jorgensen, W. L. Evaluation and reparametrization of the OPLS-AA force field for proteins via comparison with accurate quantum chemical calculations on peptides. J. Phys. Chem. B 105, 6474–6487 (2001).
Article CAS Google Scholar
Clark, K., Karsch-Mizrachi, I., Lipman, D. J., Ostell, J. & Sayers, E. W. GenBank. Nucleic Acids Res 44, D67–D72 (2016).
Article CAS PubMed Google Scholar
Smith, T. F. & Waterman, M. S. Identification of common molecular subsequences. J. Mol. Biol. 147, 195–197 (1981).
Article CAS PubMed Google Scholar
Henikoff, S. & Henikoff, J. G. Amino acid substitution matrices from protein blocks. Proc. Natl Acad. Sci. USA 89, 10915–10919 (1992).
Article ADS CAS PubMed PubMed Central Google Scholar
Zhu, K. et al. Antibody structure determination using a combination of homology modeling, energy-based refinement, and loop prediction. Proteins 82, 1646–1655 (2014).
Article CAS PubMed PubMed Central Google Scholar
Olsson, M. H., Sondergaard, C. R., Rostkowski, M. & Jensen, J. H. PROPKA3: consistent treatment of internal and surface residues in empirical pKa predictions. J. Chem. Theory Comput. 7, 525–537 (2011).
Article CAS PubMed Google Scholar
Hestenes, M. R. & Stiefel, E. Methods of conjugate gradients for solving linear systems. J. Res. Nat. Bur. Stand 49, 409–436 (1952).
Article MathSciNet Google Scholar
Laskowski, R. A., Macarthur, M. W., Moss, D. S. & Thornton, J. M. Procheck - a program to check the stereochemical quality of protein structures. J. Appl. Crystallogr. 26, 283–291 (1993).
Article ADS CAS Google Scholar

Download references

Acknowledgements

We thank the members of our laboratories for their technical support and helpful comments on the manuscript. FONDECYT−ANID (grant number 3210774) supported M.B. The Deutsche Forschungsgemeinschaft (DFG, German Research Foundation) supported N.D. (DE1482-9/1) and T.B. (BA 1793/6-2)/ M.S. (SCHE 2112/1-2) as part of the Research Unit FOR2518, DynIon.

Funding

Open Access funding enabled and organized by Projekt DEAL.

Author information

These authors contributed equally: Lea C. Neelsen, Elena B. Riel, Susanne Rinné.
These authors jointly supervised this work: Niels Decher, Thomas Baukrowitz, Marcus Schewe.

Authors and Affiliations

Institute of Physiology, Christian-Albrechts University of Kiel, Kiel, Germany
Lea C. Neelsen, Elena B. Riel, Björn C. Jürs, Jan P. Langer, Bisher Eymsh, Sönke Cordeiro, Thomas Baukrowitz & Marcus Schewe
Department of Anesthesiology, Weill Cornell Medical College, New York, NY, USA
Elena B. Riel & Freya-Rebecca Schmid
Institute of Physiology and Pathophysiology, Philipps-University of Marburg, Marburg, Germany
Susanne Rinné, Aytug K. Kiper & Niels Decher
MSH Medical School Hamburg, University of Applied Sciences and Medical University, Hamburg, Germany
Björn C. Jürs
Centro de Investigación de Estudios Avanzados del Maule (CIEAM), Vicerrectoría de Investigación y Postgrado, Universidad Católica del Maule, Talca, Chile
Mauricio Bedoya
Laboratorio de Bioinformática y Química Computacional (LBQC), Departamento de Medicina Traslacional, Facultad de Medicina, Universidad Católica del Maule, Talca, Chile
Mauricio Bedoya

Authors

Lea C. Neelsen
View author publications
Search author on:PubMed Google Scholar
Elena B. Riel
View author publications
Search author on:PubMed Google Scholar
Susanne Rinné
View author publications
Search author on:PubMed Google Scholar
Freya-Rebecca Schmid
View author publications
Search author on:PubMed Google Scholar
Björn C. Jürs
View author publications
Search author on:PubMed Google Scholar
Mauricio Bedoya
View author publications
Search author on:PubMed Google Scholar
Jan P. Langer
View author publications
Search author on:PubMed Google Scholar
Bisher Eymsh
View author publications
Search author on:PubMed Google Scholar
Aytug K. Kiper
View author publications
Search author on:PubMed Google Scholar
Sönke Cordeiro
View author publications
Search author on:PubMed Google Scholar
Niels Decher
View author publications
Search author on:PubMed Google Scholar
Thomas Baukrowitz
View author publications
Search author on:PubMed Google Scholar
Marcus Schewe
View author publications
Search author on:PubMed Google Scholar

Contributions

N.D., T.B., and M.S. conceived and supervised the project; L.C.N., E.B.R., B.C.J., J.L., B.E., and M.S. performed patch-clamp experiments and analyzed the data. S.R. and N.D. planned and generated mutant constructs for two-electrode voltage-clamp experiments. F.-R.S. and S.R. performed, and A.K.K. and S.R. analyzed TEVC experiments. N.D. planned, S.R. performed, and A.K.K. analyzed single-channel recordings. S.C. planned mutant constructs for patch-clamp experiments. M. B. generated the TALK-2 homology models. M.S. prepared figures; T.B. and M.S. wrote the article with contributions from all authors.

Corresponding authors

Correspondence to Niels Decher, Thomas Baukrowitz or Marcus Schewe.

Ethics declarations

Competing interests

The authors declare no competing interests.

Peer review

Peer review information

Nature Communications thanks the anonymous reviewers for their contribution to the peer review of this work. A peer review file is available.

Additional information

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary information

Supplementary Information

Peer Review File

Reporting Summary

Source data

Source Data

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.

Reprints and permissions

About this article

Cite this article

Neelsen, L.C., Riel, E.B., Rinné, S. et al. Ion occupancy of the selectivity filter controls opening of a cytoplasmic gate in the K_2P channel TALK-2. Nat Commun 15, 7545 (2024). https://doi.org/10.1038/s41467-024-51812-w

Download citation

Received: 23 November 2023
Accepted: 16 August 2024
Published: 30 August 2024
DOI: https://doi.org/10.1038/s41467-024-51812-w

This article is cited by

Gating mechanism of the two-pore-___domain potassium channel THIK1
- Xiangyun Fang
- Haichao Jin
- Baobin Li
Nature Structural & Molecular Biology (2025)