Structural basis for the subtype-selectivity of KCa2.2 channel activators

Structural basis for the subtype-selectivity of KCa2.2 channel activators | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Article Structural basis for the subtype-selectivity of K Ca 2.2 channel activators Miao Zhang, Young-Woo Nam, Alena Ramanishka, Yang Xu, Rose Marie Yasuda, and 4 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-6568445/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 08 Jan, 2026 Read the published version in Nature Communications → Version 1 posted You are reading this latest preprint version Abstract Small-conductance (K Ca 2.2) and intermediate-conductance (K Ca 3.1) Ca 2+ -activated K + channels are gated by a Ca 2+ -calmodulin dependent mechanism. NS309 potentiates the activity of both K Ca 2.2 and K Ca 3.1, while rimtuzalcap selectively activates K Ca 2.2. Rimtuzalcap has been used in clinical trials for the treatment of spinocerebellar ataxia and essential tremor. We report cryo-electron microscopy structures of K Ca 2.2 channels bound with NS309 and rimtuzalcap, in addition to K Ca 3.1 channels with NS309. The different conformations of calmodulin and the cytoplasmic HC helices in the two channels underlie the subtype-selectivity of rimtuzalcap for K Ca 2.2. Calmodulin’s N-lobes in the K Ca 2.2 structure are far apart and undergo conformational changes to accommodate either NS309 or rimtuzalcap. Calmodulin’s N-lobes in the K Ca 3.1 structure are closer to each other and are constrained by the HC helices of K Ca 3.1, which allows binding of NS309 but not of the bulkier rimtuzalcap. These structures provide a framework for structure-based drug design targeting K Ca 2.2 channels. Biological sciences/Biochemistry/Ion channels/Potassium channels Biological sciences/Structural biology/Electron microscopy/Cryoelectron microscopy Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Figure 7 Introduction Pharmacological activation of K + channels can dampen electrical signaling and may have therapeutic value for diverse diseases. Pharmacological agents that activate voltage-gated (K V ) 1 – 6 , inward-rectifier (K IR ) 7 , two-pore domain (K 2P ) 8 , and large conductance (K Ca 1.1) 9 K + channels widen the inner gate at the intracellular entrance to the channel pore, modulate the voltage-sensing domains, or regulate gating at the selectivity filter. The mechanism underlying the action of pharmacological activators of small-conductance Ca 2+ -activated K + (K Ca 2.1-K Ca 2.3, also called SK1-SK3) channels, an important sub-family of K + channels with distinctive biophysical and pharmacological properties, has not been defined. K Ca 2.x channels and the related K Ca 3.1 channel are encoded by the KCNN1-4 gene family. K Ca 2.x channels are critical modulators of neuronal and cardiac excitability, while KCa3.1 sustains Ca 2+ signaling through K + efflux-driven hyperpolarization in peripheral tissues, including erythrocytes, immune cells and vascular endothelium. 10 The shared Ca 2+ -calmodulin (CaM) dependent gating mechanism of these channels has been demonstrated by high resolution structure-determination using cryogenic electron microscopy (cryo-EM). 11 , 12 In the Ca 2+ -free state, CaM’s N-lobe is flexible and invisible in the cryo-EM structure, while CaM’s C-lobe interacts with the channel’s cytoplasmic HA/HB helices (previously called CaM binding domain). 12 During Ca 2+ -dependent activation, the calcified CaM N-lobe swings up and binds to the channel’s S4-S5 linker (primarily the S 45 A helix), causing the inner gate to open, allowing K + to flow through the channel pore. 11 , 12 Significant effort spanning decades has identified pharmacological activators of K Ca 2.x and K Ca 3.1 channels. These include compounds such as 1-EBIO, CyPPA, NS309, and GW-542573X. 10 Earlier crystallographic studies suggested that these activators bound to the interface between CaM’s N-lobe and the channel’s cytoplasmic HA/HB helices. 13 – 15 However, these studies were performed with truncated cytoplasmic HA/HB helices in complex with CaM and did not include the remainder of the channel. Their mechanism of activation therefore needs to be redefined within the context of the full-length structures. Here, we combined cryo-EM, site-directed mutagenesis, and electrophysiology to define the mechanism of action of rimtuzalcap, a CyPPA derivative that activates selectively K Ca 2.2/K Ca 2.3 channels, and NS309, a non-selective activator of both K Ca 2.x and K Ca 3.1 channels. Our studies show that both activators widen the inner gate of the K Ca 2.2 channel, while differences in their interactions with the Ca 2+ -CaM-dependent gating machinery underlie rimtuzalcap’s selectivity for K Ca 2.2/K Ca 2.3 channels. Results Differential interactions of CaM with the HC helices of K Ca 2.2 and K Ca 3.1 We compared our recently determined cryo-EM structure of Ca 2+ -bound apo_K Ca 2.2/CaM (resolution = 3.18 Å; referred to henceforth as apo_K Ca 2.2; Fig. 1a) 11 with the previously reported cryo-EM structure of Ca 2+ -bound apo_K Ca 3.1/CaM activation state II 12 (Protein Data Bank [PDB]: 6cno; referred to henceforth as apo_K Ca 3.1_II; Fig. 1b). The cytoplasmic HC helices are invisible in the apo_K Ca 2.2 structure possibly due to flexibility (Fig. 1a). In contrast, the C-terminal HC helices are well resolved and are seen between CaM molecules in the apo_K Ca 3.1_II structure (Fig. 1b). The distinct conformations of the HC helices in the two channels may be attributable to different conformations of CaM in the two structures. In apo_K Ca 2.2, CaM’s N-lobes are too far apart (Asn42 in opposite CaM molecules are ~35.0 Å apart) to stabilize the HC helices, which may contribute to their flexibility (Fig. 1c). In the apo_K Ca 3.1_II structure, in contrast, CaM’s N-lobes are sufficiently close (Asn42 in opposite CaM molecules are ~27.0 Å apart) to stabilize the HC helices, causing them to be visible (Fig. 1d). As will be seen below, these differences between apo_K Ca 2.2 and apo_K Ca 3.1_II contribute to the selectivity of rimtuzalcap for K Ca 2.2 over K Ca 3.1. The non-selective NS309 interacts similarly with K Ca 3.1/CaM and K Ca 2.2/CaM NS309 activates both K Ca 3.1 (EC 50 : ~74 nM) and K Ca 2.2 (EC 50 : ~1.7 mM) channels. 16,17 We determined structures of NS309 bound to K Ca 2.2/CaM and to K Ca 3.1/CaM. The cryo-EM map of the Ca 2+ - and NS309-bound K Ca 2.2/CaM complex (henceforth referred to as NS309_K Ca 2.2) was refined to a resolution of 2.71 Å (SupplementaryFig. 1,2,SupplementaryTable 1). The cryo-EM map of the Ca 2+ - and NS309-bound K Ca 3.1/CaM complex (henceforth referred to as NS309_K Ca 3.1) was refined to a resolution of 3.59 Å (SupplementaryFig. 3,4,SupplementaryTable 1). Cryo-EM densities for NS309 in each of the four CaM’s N-lobes are clearly visible in both NS309_K Ca 2.2 (Supplementary Fig. 5a,b) and NS309_K Ca 3.1 (Supplementary Fig. 6a,b). We successfully built models for NS309 into the cryo-EM densities in both structures at the interface between CaM’s N-lobe and the channels’ S 45 A helix (Fig. 2a,b). Since CaM’s N-lobe forms a significant portion of NS309’s binding pocket, we compared the conformations of CaM in the apo versus activator-bound structures. CaM molecules aligned well in the apo_K Ca 2.2 and NS309_K Ca 2.2 comparison (rmsd = 1.4 Å, Supplementary Fig. 7a,b), and even better in the apo_K Ca 3.1_II and NS309_K Ca 3.1 comparison (rmsd = 1.0 Å, Supplementary Fig. 7c,d). Binding of NS309 to the K Ca 2.2/CaM complex shortened the distance between CaM’s N-lobes from ~35.0 Å in apo_K Ca 2.2 (Fig. 1c) to ~31.0 Å in NS309_K Ca 2.2 (Fig. 2c). Binding of NS309 to the K Ca 3.1/CaM complex did not affect the distance between CaM’s N-lobes (Fig. 2d) compared to the apo_K Ca 3.1_II structure (Fig. 1d). The HC helices are visible in both apo_K Ca 3.1_II (Fig. 1d) and NS309_K Ca 3.1 (Fig. 2d), while they remain invisible in NS309_K Ca 2.2 (Fig. 2c). Based on these results, we conclude that the binding pocket for NS309 preexists in both K Ca 2.2 and K Ca 3.1 channels. We calculated the binding energy (van der Waals forces plus electrostatic interactions) between the four bound NS309 molecules and amino acid residues in the four subunits of the activator-bound structures of K Ca 2.2 and K Ca 3.1 using the Discovery Studio program. In the NS309_K Ca 2.2 structure, NS309 sits at the interface between CaM’s N-lobe and K Ca 2.2’s S 45 A helix, with Ser288 and Leu292 in K Ca 2.2’s S 45 A helix interacting with NS309, and a tetrad of hydrophobic residues (Phe19, Leu32, Met51, and Met71; FLMM N 18 ) in CaM’s N-lobe cradling the NS309 molecule (Fig. 3a). In the NS309_K Ca 3.1 structure, NS309 fits perfectly into a hydrophobic pocket between CaM’s N-lobe and K Ca 3.1’s S 45 A helix, with Ser181 and Leu185 in K Ca 3.1’s S 45 A helix interacting with NS309, and the same tetrad of hydrophobic residues (Phe19, Leu32, Met51, and Met71; FLMM N 18 ) in CaM’s N-lobe cradling the NS309 molecule (Fig. 3b). These results are consistent with earlier reports that mutations of Ser181 and Leu185 in K Ca 3.1 and mutations of Ser288 and Leu292 in K Ca 2.2 significantly reduced sensitivity to NS309. 17 The total binding energies were -27.6 ± 1.0 kCal/mol and -29.7 ± 1.1 kCal/mol for NS309_K Ca 2.2 and NS309_K Ca 3.1, respectively (P=0.03, n=4, unpaired two-tailed Student’s t -test, Fig. 3c). The van der Waals forces involved in NS309’s interactions with the two channels are comparable (NS309_K Ca 2.2: -24.9 ± 0.8 kCal/mol; NS309_K Ca 3.1: -23.7 ± 0.7 kCal/mol; P=0.06, n=4, unpaired two-tailed Student’s t -test). NS309’s stronger electrostatic interactions with K Ca 3.1 (-6.0 ± 0.7 kCal/mol) versus K Ca 2.2 (-2.7 ± 0.3 kCal/mol, P=0.0003, n=4, unpaired two-tailed Student’s t -test, Fig. 3c) account for NS309’s stronger total binding energy to K Ca 3.1 than K Ca 2.2. In both structures, CaM_Lys75 is the largest contributor to the electrostatic interactions with NS309 (Fig. 3a,b). Electrostatic interactions between CaM_Lys75 and NS309 are stronger in NS309_K Ca 3.1 (-4.9 ± 0.8 kCal/mol) than in NS309_K Ca 2.2 (-1.1 ± 0.5 kCal/mol, P=0.0002, n=4, unpaired two-tailed Student’s t -test). NS309’s (Fig. 3d) higher binding energy to K Ca 3.1 than K Ca 2.2 is consistent with NS309’s ~20-fold higher potency in activating K Ca 3.1 than K Ca 2.2. 16,17 In NS309_K Ca 2.2, the compact conformation of CaM is stabilized by salt bridges between CaM’s C-lobe (Glu83 and Glu 87 in helix V) and the K Ca 2.2 channel’s S 45 A helix (Lys294) and HB helix (Lys467) (Fig. 2a). CaM_Lys75 forms a salt bridge (~3.2 Å) with CaM_Glu83, which may weaken its hydrogen bond with NS309 (~4.3 Å, Fig. 2a). In NS309_K Ca 3.1, CaM’s more extended conformation is stabilized by salt bridges between CaM’s C-lobe (Glu83 and Glu87 in helix V) and K Ca 3.1’s HB helix (Arg355) (Fig. 2b). CaM_Lys75 is ~6.8 Å away from CaM_Glu83, while it is much closer to NS309 (~3.0 Å, Fig. 2b). In summary, both K Ca 2.2 and K Ca 3.1 can accommodate NS309 into pre-existing binding pockets at the interface between CaM’s N-lobe and the channels’ S 45 A helix, which may explain the non-selective activation of K Ca 2.x and K Ca 3.1 channels by NS309. 16,17 Structure of K Ca 2.2/CaM bound to the subtype-selective activator rimtuzalcap CyPPA, a positive allosteric modulator, potentiates the activity of K Ca 2.2 and K Ca 2.3, but is inactive on K Ca 3.1. 19 Rimtuzalcap is a CyPPA derivative that potentiates K Ca 2.2 at low micromolar concentrations (EC 50 : ~5.1 mM) and is inactive on K Ca 3.1. Rimtuzalcap was evaluated in a Phase-2 clinical trial as a treatment for essential tremor (ClinicalTrials.gov: NCT03688685), and a phase-2 clinical trial evaluating rimtuzalcap in patients with spinocerebellar ataxia was initiated (ClinicalTrials.gov: NCT03688685) and then withdrawn when Cadent Therapeutics was acquired by Novartis. We refined the cryo-EM map of the Ca 2+ - and rimtuzalcap-bound K Ca 2.2/CaM (henceforth referred to as rimtuzalcap_K Ca 2.2_I) to a resolution of 3.13 Å (SupplementaryFig. 8, 9, Supplementary Table 1). Cryo-EM densities for rimtuzalcap in each of the four CaM’s N-lobes are clearly visible in rimtuzalcap_K Ca 2.2_I (SupplementaryFig. 10a,b). We successfully built a model for rimtuzalcap into its cryo-EM density. Rimtuzalcap sits at the interface between CaM’s N-lobe and K Ca 2.2’s S 45 A/HA helices in each subunit of the tetrameric channel (Fig. 4a). Similar to the NS309_K Ca 2.2 structure (Fig. 2a), the rimtuzalcap_K Ca 2.2_I structure shows that the salt bridges between CaM helix V (Glu83 and Glu87), the S 45 A (Lys294), and the HB (Lys467) helices of K Ca 2.2 are present (Fig. 4a). Large differences are apparent in the alignment of CaM’s a-carbons (rmsd = 2.6 Å, Fig. 4b,c) in rimtuzalcap_K Ca 2.2_I versus apo_K Ca 2.2. Binding of rimtuzalcap expands CaM’s N-lobes, shortens the distances between the N-lobes in opposite subunits from ~35 Å in apo_K Ca 2.2 (Fig. 1c) to ~22 Å in rimtuzalcap_K Ca 2.2_I, and stabilizes K Ca 2.2’s cytoplasmic HC helices, rendering them visible (Fig. 4d). These observations suggest that a substantial conformational change is required to accommodate rimtuzalcap into its binding pocket in K Ca 2.2/CaM. In contrast, the conformational change required to fit NS309 into its binding pocket in K Ca 2.2 channels is minimal because comparison of the NS309_K Ca 2.2 versus apo_K Ca 2.2 structures shows CaM’s a-carbons in good alignment (rmsd = 1.4 Å, Supplementary Fig. 7a,b), small changes in the distances between opposite CaM’ N-lobes (~31 Å versus ~35 Å, Figs. 1c, 2c), and cytoplasmic HC helices that remain flexible and invisible (Figs. 1c, 2c). NS309 primarily interacts with K Ca 2.2’s S 45 A helix (Fig. 3), whereas the bulkier rimtuzalcap (Fig. 5a) forms contacts with K Ca 2.2’s HA helix in addition to the S 45 A helix (Fig. 5b). Rimtuzalcap fits perfectly into the hydrophobic pocket formed by residues in K Ca 2.2’s S 45 A helix (Ile289, Leu292, Asn293), K Ca 2.2’s HA helix in a neighboring subunit (His406, Phe410), and CaM’s N-lobe (Phe19, Leu32, Met36, Met51) (Fig. 5c). The binding energy between rimtuzalcap and its induced-fit binding pocket in K Ca 2.2 (-36.9 ± 1.3 kCal/mol) is stronger than NS309’s binding energy to its pre-existing binding pocket in K Ca 2.2 (-27.6 ± 1.0 kCal/mol). However, NS309 is a more potent activator of K Ca 2.2 (EC 50 : ~1.7 mM) than rimtuzalcap (EC 50 : ~5.1 mM) in electrophysiological assays. The discordance between binding energies and activation-potency of these two compounds may be due inherent variations in ligand-binding to pre-existing (NS309) versus induced-fit (rimtuzalcap) pockets 20 and to differences in ligand efficiency caused by differences in size of the two molecules (Figs. 3d, 5a). Unlike the K Ca 2.2 channel (Fig. 6a), the K Ca 3.1 channel is insensitive to rimtuzalcap (Fig. 6b). In the apo_K Ca 3.1_II structure, CaM’s N-lobes are close to each other (~27.0 Å apart, Fig. 1d) and the putative binding pocket in CaM’s N-lobe is constrained by the HC helices, which may prevent it from undergoing conformational changes required to accommodate rimtuzalcap. We therefore wondered if rearranging the salt bridges between K Ca 3.1’s HB helix and CaM’s helix V would release the putative binding pocket from the constraints of the HC helices and allow rimtuzalcap binding. To test this idea, we mutated Arg355 in K Ca 3.1’s HB helix to lysine to mimic the salt bridge-forming Lys467 in K Ca 2.2 (Fig. 4a). In support of our idea, the mutant K Ca 3.1_R355K channel was activated by rimtuzalcap with an EC 50 = 76.8 ± 21.3 mM (n=8; Fig. 6c,d). Additional structural determinants may underlie rimtuzalcap’s ~15-fold lower potency on K Ca 3.1_R355K compared to K Ca 2.2_WT (EC 50 : 5.1 ± 1.0 mM, n=5, P<0.0001, unpaired two-tailed Student’s t -test). Widening of the inner gate by NS309 and rimtuzalcap Rimtuzalcap and NS309 widened K Ca 2.2’s inner gate from ~12.6 Å in the apo-form (apo_K Ca 2.2) to ~13.5 Å and ~13.1 Å in the rimtuzalcap_K Ca 2.2_I and NS309_K Ca 2.2 structures, respectively (Fig. 7). In our previous cryo-EM study, the inhibitor AP14145 narrowed the inner gate to ~6.5 Å (PDB: 8v2h; referred to henceforth as AP14145_K Ca 2.2; Fig. 7). Thus, rimtuzalcap and NS309 both activate K Ca 2.2 by widening the inner gate. Differences in their interactions with the Ca 2+ -CaM-dependent gating machinery of K Ca 2.2 versus K Ca 3.1 likely underlie rimtuzalcap’s K Ca 2.x-selectivity and NS309’s non-selective activation of both K Ca 2.2 and K Ca 3.1. Activators of K Ca 2.2 have been termed positive allosteric modulators because they work only in the presence of Ca 2+ . 19,21,22 Both NS309 and rimtuzalcap (Fig. 6) require a minimal concentration (~0.1-0.2 mM) of Ca 2+ to enhance the activity of K Ca 2.2 channels. This requirement for Ca 2+ is illustrated by comparison of two rimtuzalcap-bound K Ca 2.2 structures built using cryo-EM maps of two 3D classes from the same dataset (Supplementary Fig. 8, 9). The density for rimtuzalcap is visible in both rimtuzalcap_K Ca 2.2_I and rimtuzalcap_K Ca 2.2_II (Supplementary Fig. 10), whereas Ca 2+ ions are present only in CaM’s N-lobes of rimtuzalcap_K Ca 2.2_I (Supplementary Fig. 10a,b) and not rimtuzalcap_K Ca 2.2_II (Supplementary Fig. 10c-e). Rimtuzalcap_K Ca 2.2_II that lacks Ca 2+ in CaM’s N-lobes exhibits a closed inner gate (~6.9 Å), whereas rimtuzalcap_K Ca 2.2_I that contains Ca 2+ in CaM’s N-lobes (Supplementary Fig. 10f,g) exhibits a widened inner gate (~13.5 Å). Taken together, these results suggest that binding of NS309 and rimtuzalcap to the Ca 2+ -bound CaM N-lobe stabilizes its interaction with K Ca 2.2’s S 45 A helix and dilates the channel’s inner gate. Discussion Even though K Ca 2.2 and K Ca 3.1 channels share a Ca 2+ /CaM-dependent gating mechanism, CaM interacts differently with the cytoplasmic HC helices of K Ca 2.2 and K Ca 3.1. In apo_K Ca 2.2, the CaM N-lobes in opposite subunits are too far apart to stabilize K Ca 2.2’s HC helices (Fig. 1c). In apo_K Ca 3.1_II, in contrast, the CaM N-lobes in opposite subunits are close enough to stabilize K Ca 3.1’s HC helices (Fig. 1d). The distinct conformations of CaM in the two channel structures underlie the subtype-selectivity of pharmacological activators. Accommodation of NS309 into its binding pockets in K Ca 2.2 and K Ca 3.1 does not require a prominent conformational change within CaM as evidenced by the excellent alignment of CaM’s a-carbons between the apo- and NS309-bound K Ca 2.2 and K Ca 3.1 structures (Supplementary Fig. 7). This indicates that NS309 fits into pre-existing binding pockets in both NS309_K Ca 3.1 and NS309_K Ca 2.2. NS309’s stronger electrostatic interactions with CaM_Lys75 (Figs. 3a,b) accounts for its higher total binding energy to NS309_K Ca 3.1 versus NS309_K Ca 2.2 (Fig. 3c). This difference in binding strength likely underlies NS309’s ~20-fold higher potency in activating K Ca 3.1 than K Ca 2.2 channels. 16,17 In contrast, accommodation of rimtuzalcap into its binding pockets in K Ca 2.2 requires significant conformational changes within CaM as evidenced by large differences in the alignment of CaM’s a-carbons between the apo_K Ca 2.2 and rimtuzalcap_K Ca 2.2_I structures (Figs. 4b,c). These results suggest that the induced-fit binding pocket of rimtuzalcap does not exist in apo_K Ca 2.2. Binding of rimtuzalcap expands CaM’s N-lobes around rimtuzalcap and shortens the distances between opposite CaM’s N-lobes, resulting in rigidification of K Ca 2.2’s cytoplasmic HC helices (Fig. 4d). These activator-induced changes in rimtuzalcap_K Ca 2.2_I are reminiscent of the mechanism of action of AUT5, an activator of the voltage-gated K V 3.1 channel. Binding of AUT5 to K V 3.1 rearranges the turret region around the activator molecule, thereby inducing interactions between the turret and the transmembrane S4 segment in the voltage-sensor domain. 4 Previously, we reported that CyPPA, an analog of rimtuzalcap, exhibited subtype-selectivity for K Ca 2.2 over K Ca 3.1 due to the difference of a single residue in the HB helix (K Ca 2.2_K467 versus K Ca 3.1_R355). 23 However, our earlier docking studies were based on a homology model of K Ca 2.2 generated with K Ca 3.1 cryo-EM structures as templates. 12 Docking into this homology model placed CyPPA in a pre-existing binding pocket between CaM’s C-lobe and K Ca 2.2’s HA/HB helices 23 . This placement is most likely wrong because CyPPA and rimtuzalcap are structurally very similar (Fig. 5a), and we have shown above that rimtuzalcap fits into an induced-fit binding pocket that emerges as a result of conformational changes in CaM (Fig. 4). The K Ca 3.1 channel is insensitive to rimtuzalcap. Our structural, mutagenesis and electrophysiological studies suggest that this insensitivity to rimtuzalcap is because the putative activator binding pocket in CaM’s N-lobe of K Ca 3.1 is constrained by the HC helices and is prevented from undergoing the requisite conformational changes to accommodate rimtuzalcap. Replacing K Ca 3.1_R355 with lysine (corresponding to K Ca 2.2_K467) in K Ca 3.1’s HB helix frees the binding pocket from the constraints of the HC helices, and allows rimtuzalcap to activate the mutant K Ca 3.1_R355K channel (EC 50 : ~76.8 mM, Fig. 6). In conclusion, our structures provide a foundation for understanding the subtype-selectivity of the K Ca 2.2 activator rimtuzalcap and could enable structure-based drug design of more potent, subtype-selective activators targeting K Ca 2.2 channels. Subtype-selective K Ca 2.2 activators that avoid potential side effects associated with activation of peripheral K Ca 3.1 channels are critically needed to target K Ca 2.2 channels in the central nervous system for the treatment of spinocerebellar ataxia and essential tremor. Methods Protein expression and purification The rat K Ca 2.2/CaM protein complex was expressed and purified as described in our previous report. 11 Briefly, the cDNA of full-length rat K Ca 2.2 (accession no. NM_019314) with a C-terminus Strep-II tag was sub-cloned into pEG BacMam (a gift from Eric Gouaux; Addgene plasmid # 160451; http://n2t.net/addgene:160451; RRID:Addgene_160451). Un-tagged rat calmodulin (CaM) cDNA (accession no. BC063187) was also cloned into pEG BacMam. The amino acid sequence of the rat CaM is 100% identical to the human CaM. We expressed the K Ca 2.2/CaM protein complex in HEK293S GnTI− cells (ATCC) using a BacMam method 24 . The K Ca 2.2/CaM complex was purified using Strep-Tactin XT resin, followed by size exclusion chromatography column equilibrated by 20 mM Tris pH 8, 150 mM KCl, 2 mM CaCl 2 , and 0.01% lauryl maltose neopentyl glycol (LMNG, Anatrace). The peak fractions were collected and concentrated to ~3 mg/ml. The human K Ca 3.1/CaM protein complex was expressed and purified as described in previous reports. 12,25 Briefly, the cDNA of full-length human K Ca 3.1 (accession no. NM_002250.3) with a C-terminus Strep-II tag was sub-cloned into pEG BacMam. We expressed the K Ca 3.1/CaM protein complex in HEK293S GnTI− cells (ATCC) using a BacMam method 24 . The K Ca 3.1/CaM complex was purified using Strep-Tactin XT resin, followed by size exclusion chromatography column equilibrated by 20 mM Tris pH 8, 150 mM KCl, 2 mM CaCl 2 , and 0.007% glyco-diosgenin (GDN, Anatrace). The peak fractions were collected and concentrated to ~3 mg/ml. Cryo-EM Sample Preparation, Data Collection and Processing We performed cryo-EM data collection at the Stanford SLAC Cryo-EM Center (S 2 C 2 ) and Pacific Northwest Cryo-EM Center (PNCC). To determine the NS309-bound structures, saturating concentrations of NS309 were mixed with the K Ca 3.1/CaM or K Ca 2.2/CaM protein complexes 30 minutes before the grid preparation, respectively. To determine the rimtuzalcap-bound structure, saturating concentrations of rimtuzalcap were mixed with the K Ca 2.2/CaM protein complex 30 minutes before the grid preparation. Because K Ca 3.1 channels are not sensitive to rimtuzalcap, the K Ca 3.1/CaM protein complex was not mixed with rimtuzalcap. 3 μl of purified protein was applied to a glow-discharged Quantifoil R1.2/1.3 300 mesh Copper grid, at 4 °C and 100% humidity using Vitrobot Mark IV (ThermoFisher Scientific). The grid was then blotted for 3 seconds before being plunged into liquid ethane. Grids of NS309-bound K Ca 3.1/CaM were screened, and cryo-EM data were subsequently collected on a Titan Krios G3i (ThermoFisher Scientific) with a K3 detector (Gatan) and a BioQuantum energy filter. Dose fractionated movies were collected using EPU at the pixel size of 0.86 Å. A total cumulative dose of ∼50 electrons per Å 2 was used for recording the movies of 40 frames (1.25 electrons per Å 2 per frame) with a defocus range of –1.0 to –2.0 μm. Grids of rimtuzalcap-bound K Ca 2.2/CaM were screened, and cryo-EM data were subsequently collected on a Titan Krios G3i (ThermoFisher Scientific) with a K3 detector (Gatan) and a BioQuantum energy filter. Dose fractionated movies were collected using EPU at the pixel size of 0.86 Å. A total cumulative dose of ∼50 electrons per Å 2 was used for recording the movies of 40 frames (1.25 electrons per Å 2 per frame) with a defocus range of –1.3 to –2.3 μm. Grids of NS309-bound K Ca 2.2/CaM were screened, and cryo-EM data were subsequently collected on a Titan Krios G3i (ThermoFisher Scientific) with a Falcon4i detector (ThermoFisher Scientific) and a SelectrisX energy filter. Dose fractionated movies were collected using EPU at the pixel size of 0.73 Å. A total cumulative dose of ∼50 electrons per Å 2 was used for recording in the EER format with a defocus range of –0.6 to –2.2 μm. The statistics for data collections are summarized in Supplementary Tables 1. Cryo-EM image processing for NS309-bound K Ca 2.2/CaM was carried out using CryoSparc version 4.5.3 26 . After preprocessing of micrographs (motion correction and CTF estimation), ~2,000 particles were manually picked, followed by 2D classification to generate picking templates. Template-picked particles were cleaned up by multiple rounds of 2D classification. An ab initio reconstruction was performed, followed by heterogenous and non-uniform refinements to 2.72 Å resolution in the CryoSparc 26 program. The cryo-EM map with 2.72 Å resolution was improved by 3D classification followed by nonuniform refinement to 2.71 Å resolution (NS309_K Ca 2.2, Supplementary Fig. 1). The cryo-EM density for NS309 is clearly visible in the refined map of NS309_K Ca 2.2 (Fig. 2a and Supplementary Fig. 5). The NS309-bound K Ca 3.1/CaM dataset was similarly processed using the CryoSparc program 26 . An ab initio reconstruction was performed, followed by heterogenous and non-uniform refinements to 3.69 Å resolution in the CryoSparc 26 program. The cryo-EM map with 3.69 Å resolution was improved by 3D classification followed by nonuniform refinement to 3.59 Å resolution (NS309_K Ca 3.1, Supplementary Fig. 3). The cryo-EM density for NS309 is clearly visible in the refined map of NS309_K Ca 3.1 (Fig. 2b and Supplementary Fig. 6). For the rimtuzalcap-bound K Ca 2.2/CaM dataset, an ab initio reconstruction was performed, followed by heterogenous and non-uniform refinements to 2.95 Å resolution in the CryoSparc 26 program. To examine possible conformational heterogeneity, 3D classification was performed. Two classes were generated by 3D classification, which were further refined using non-uniform refinements to 3.13 Å resolution (rimtuzalcap_K Ca 2.2_I), and 2.96 Å resolution (rimtuzalcap_K Ca 2.2_II), which were used for model building (Supplementary Fig. 8). The cryo-EM density of rimtuzalcap is clearly visible at the interface between the CaM N-lobe and the S 45 A/HA helices of K Ca 2.2 in both rimtuzalcap_K Ca 2.2_I and rimtuzalcap_K Ca 2.2_II (Supplementary Fig. 10). The cryo-EM densities of two Ca 2+ ions are visible at the CaM N-lobes of rimtuzalcap_K Ca 2.2_I but not in rimtuzalcap_K Ca 2.2_II. In a portion of rimtuzalcap-bound K Ca 2.2 channels, the CaM N-lobe may lose Ca 2+ binding (rimtuzalcap_K Ca 2.2_II), which in turn closes the inner gate because Ca 2+ binding to the CaM N-lobe is required for the activation of K Ca 2.2 22 . The other portion of rimtuzalcap-bound K Ca 2.2 channels may retain both Ca 2+ and rimtuzalcap (rimtuzalcap_K Ca 2.2_I), which exhibits a wider inner gate (~13.5 Å) than the apo_K Ca 2.2 structure (~12.6 Å). Since Ca 2+ binding to the CaM N-lobe is required for the activation of K Ca 2.2 22 , rimtuzalcap_K Ca 2.2_I that contains both rimtuzalcap and Ca 2+ was used for further analysis of the interactions between rimtuzalcap and K Ca 2.2 channels. Resolutions were estimated using the gold standard criterion at the threshold of 0.143. The local resolution was calculated in CryoSparc. Model building Coordinates of the Ca 2+ -bound K Ca 2.2 (PDB: 8v2g; apo_K Ca 2.2) were used as the initial model for activator-bound K Ca 2.2 structures. Coordinates of the Ca 2+ -bound K Ca 3.1 activated state II (PDB: 6cno; apo_K Ca 3.1_II) were used as the initial model for activator-bound K Ca 3.1 structures. The initial model was manually docked into the cryo-EM density map and then adjusted in UCSF Chimera. Phenix.real_space_refine 27 was used to build and refine the model. Model building was achieved using phenix 27 and Coot 28 iteratively. Models for Ca 2+ , K + , and activators were built by visual inspection of the shape of the density in Coot 28 followed by refinement in phenix 27 . The statistics for model refinements are summarized in Supplementary Table 1. All structural graphics were generated using UCSF ChimeraX 29 . Patch-clamp electrophysiology Human Embryonic Kidney (HEK293) cells transiently transfected with the rat K Ca 2.2 or human K Ca 3.1 cDNAs were used for manual patch-clamp experiments. Site-directed mutagenesis of the K Ca 3.1_R355K was performed on the cDNAs through molecular cloning services (Genscript). The wildtype and mutant cDNAs, constructed in the pIRES2-AcGFP1 vector (Clontech),were transfected into HEK293 cells by a calcium–phosphate method. Inside-out K Ca 2.2 currents were recorded 1–2 days after transfec­tion, with an Axon200B amplifier (Molecular Devices) at room temperature. The resistance of the patch electrodes ranged from 2-3 MΩ. For inside-out recordings, the intracellular solution containing (in mM): 140 KCl, 10 Hepes (pH 7.2), 1 EGTA, 0.1 Dibromo-BAPTA, and 1 HEDTA was mixed with Ca 2+ to obtain the desired free Ca 2+ concentrations, calculated using the MaxChelator software. The extracellular solution contained (in mM): 140 KCl, 10 Hepes (pH 7.4), 1 MgSO 4 . NS309 (6,7-dichloro-1 H -indole-2,3-dione 3-oxime) was purchased from Alomone labs. NS309 dilutions were prepared freshly in extracellular solution from 20 mM stock solutions in DMSO. Rimtuzalcap (also called CAD-1883, N -(4,4-difluorocyclohexyl)-2-(3-methylpyrazol-1-yl)-6-morpholin-4-ylpyrimidin-4-amine) was purchased from MedChemExpress. Rimtuzalcap dilutions were prepared freshly in extracellular solution from 100 mM stock solutions in DMSO. pClamp 10.5 (Molecular Devices) was used for data acquisi­tion and analysis. To characterize the responses of the K Ca 3.1_R355K mutants to rimtuzalcap, inside-out patch recordings were performed. Seals (> 1 GΩ) were formed before the inside-out patch configuration was obtained. The intracel­lular face was exposed to a series of activator concentrations at a fixed Ca 2+ concentration (0.15 mM Ca 2+ ). Currents were recorded by repetitive 1s voltage ramps from − 100 mV to 100 mV from a holding potential of 0 mV. One minute after switching of bath solutions, ten sweeps with a 1 s interval were recorded. To construct the concentration-dependent activation of channel activities, the current amplitudes at − 90 mV in response to various concentrations of activatorswere normalized to that obtained at 10 mM of Ca 2+ . The normalized currents were plotted as a function of the concentrations of the activators. EC 50 values and Hill coefficients were determined by fitting the data points to a standard concentration–response curve. Data analysis was performed using pClamp 10.5 (Molecular Devices) in a blinded fashion. Concentration-response curves were analyzed in GraphPad Prism 10 (GraphPad Software Inc.). All data are shown as mean ± SD unless otherwise indicated. One-way ANOVA and Tukey’s post hoc tests were used for data comparison of three or more groups. The unpaired two-tailed Student’s t -test was used for data comparison if there were only two groups. Figures were made using GraphPad Prism 10 (GraphPad Software Inc.). Calculation of binding energy The binding energy between NS309 and rimtuzalcap in their binding pockets were calculated using the cryo-EM structures and the Discovery Studio program (Dassault Systemes Biovia LLC). Briefly, the channel complex structures were subjected to energy minimization using Smart Minimizer algorithm (200 steps) and Generalized Born (GB) Implicit Solvent and Membrane model using the CHARMm forcefield. Interaction energies between the drugs and channels were calculated using an Implicit Distance-Dependent Dielectrics (Dielectric Constant=2.0) solvent model. Residues within 6.5 Å from the drugs were selected for interaction energy and energy decomposition calculations. Declarations Data availability The atomic coordinates have been deposited in the Protein Data Bank (PDB) under accession codes 9O7S [https://doi.org/10.2210/pdb9O7S/pdb] (NS309_K Ca 2.2); 9OA8 [https://doi.org/10.2210/pdb9OA8/pdb] (NS309_K Ca 3.1); 9O85 [https://doi.org/10.2210/pdb9O85/pdb] (rimtuzalcap_K Ca 2.2_I); and 9O93 [https://doi.org/10.2210/pdb9O93/pdb] (rimtuzalcap_K Ca 2.2_II). The cryo-EM maps have been deposited in the Electron Microscopy Data Bank (EMDB) under accession codes EMD-70207 [https://www.ebi.ac.uk/pdbe/entry/emdb/EMD-70207] (NS309_K Ca 2.2); EMDB-70275 [https://www.ebi.ac.uk/pdbe/entry/emdb/EMDB-70275] (NS309_K Ca 3.1); EMD-70217 [https://www.ebi.ac.uk/pdbe/entry/emdb/EMD-70217] (rimtuzalcap_K Ca 2.2_I); and EMD-70240 [https://www.ebi.ac.uk/pdbe/entry/emdb/EMD-70240] (rimtuzalcap_K Ca 2.2_II). The following previously published PDB codes are referred to: 8v2g [https://doi.org/10.2210/pdb8v2g/pdb]; 8v2h [https://doi.org/10.2210/pdb8v2h/pdb]; 6cnn [https://doi.org/10.2210/pdb6cnn/pdb]; 6cno [https://doi.org/10.2210/pdb6cno/pdb]. All other data supporting the findings of this study are available within the paper and its Supplementary information files. Acknowledgments M.Z. was supported by National Institutes of Health (Grant 4R33NS101182-03 and Grant 1R15NS130420), American Heart Association (Grant 23AIREA1039423), Chapman University Institutional FGRSC Grant, and Office of the Assistant Secretary of Defense for Health Affairs through the Alcohol and Substance Use Research Program under Award No.: W81XWH-22-2-0081. Y.W.N was supported by American Heart Association (24CDA1260237). Some of this work was performed at the Stanford-SLAC Cryo-EM Center (S 2 C 2 ), which is supported by the National Institutes of Health Common Fund Transformative High-Resolution Cryo-Electron Microscopy program (U24 GM129541). The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health. The authors would also like to thank the following S 2 C 2 personnel for their invaluable support and assistance: Patrick Mitchell, Alexandre Cassago, Nathan D. Burrows, Yan Liu, Grace Nye, Chensong Zhang, and Patrick J. Pascual. A portion of this research was supported by NIH grant R24GM154185 and performed at the Pacific Northwest Center for Cryo-EM (PNCC) with assistance from Vamsee Rayaprolu. Competing interests The authors declare no competing interests. Author contributions M.Z. conceptualized the project. Y.W.N., A.R., and M.Z. undertook protein expression and purification. Y.W.N., A.R., Y.X., R.M.H.Y., D.I., and M.Z. undertook cryo-EM studies. A.R. undertook electrophysiology studies. M.C. performed computational studies. M.Z., K.G.C, and H.W. wrote the manuscript. All authors contributed to the manuscript and the figures. References Li, T. et al. Structural Basis for the Modulation of Human KCNQ4 by Small-Molecule Drugs. Mol Cell 81 , 25-37 e24 (2021). https://doi.org:10.1016/j.molcel.2020.10.037 Willegems, K. et al. Structural and electrophysiological basis for the modulation of KCNQ1 channel currents by ML277. Nat Commun 13 , 3760 (2022). https://doi.org:10.1038/s41467-022-31526-7 Zhang, S. et al. A small-molecule activation mechanism that directly opens the KCNQ2 channel. Nat Chem Biol 20 , 847-856 (2024). https://doi.org:10.1038/s41589-023-01515-y Liang, Q. et al. The binding and mechanism of a positive allosteric modulator of Kv3 channels. Nat Commun 15 , 2533 (2024). https://doi.org:10.1038/s41467-024-46813-8 Chen, Y. T. et al. Identification, structural, and biophysical characterization of a positive modulator of human Kv3.1 channels. Proc Natl Acad Sci U S A 120 , e2220029120 (2023). https://doi.org:10.1073/pnas.2220029120 Botte, M. et al. Apo and ligand-bound high resolution Cryo-EM structures of the human Kv3.1 channel reveal a novel binding site for positive modulators. PNAS Nexus 1 , pgac083 (2022). https://doi.org:10.1093/pnasnexus/pgac083 Cui, M. et al. A novel small-molecule selective activator of homomeric GIRK4 channels. The Journal of biological chemistry 298 , 102009 (2022). https://doi.org:10.1016/j.jbc.2022.102009 Lolicato, M. et al. K(2P) channel C-type gating involves asymmetric selectivity filter order-disorder transitions. Sci Adv 6 (2020). https://doi.org:10.1126/sciadv.abc9174 Zhang, G. et al. An allosteric modulator activates BK channels by perturbing coupling between Ca(2+) binding and pore opening. Nat Commun 13 , 6784 (2022). https://doi.org:10.1038/s41467-022-34359-6 Brown, B. M., Shim, H., Christophersen, P. & Wulff, H. Pharmacology of Small- and Intermediate-Conductance Calcium-Activated Potassium Channels. Annu Rev Pharmacol Toxicol 60 , 219-240 (2020). https://doi.org:10.1146/annurev-pharmtox-010919-023420 Nam, Y. W. et al. Cryo-EM structures of the small-conductance Ca(2+)-activated K(Ca)2.2 channel. Nat Commun 16 , 3690 (2025). https://doi.org:10.1038/s41467-025-59061-1 Lee, C. H. & MacKinnon, R. Activation mechanism of a human SK-calmodulin channel complex elucidated by cryo-EM structures. Science 360 , 508-513 (2018). https://doi.org:10.1126/science.aas9466 Cho, L. T. et al. An Intracellular Allosteric Modulator Binding Pocket in SK2 Ion Channels Is Shared by Multiple Chemotypes. Structure 26 , 533-544 e533 (2018). https://doi.org:10.1016/j.str.2018.02.017 Zhang, M., Pascal, J. M. & Zhang, J. F. Unstructured to structured transition of an intrinsically disordered protein peptide in coupling Ca(2)(+)-sensing and SK channel activation. Proc Natl Acad Sci U S A 110 , 4828-4833 (2013). https://doi.org:10.1073/pnas.1220253110 Zhang, M., Pascal, J. M., Schumann, M., Armen, R. S. & Zhang, J. F. Identification of the functional binding pocket for compounds targeting small-conductance Ca(2)(+)-activated potassium channels. Nat Commun 3 , 1021 (2012). https://doi.org:10.1038/ncomms2017 Strobaek, D. et al. Activation of human IK and SK Ca2+ -activated K+ channels by NS309 (6,7-dichloro-1H-indole-2,3-dione 3-oxime). Biochimica et biophysica acta 1665 , 1-5 (2004). https://doi.org:10.1016/j.bbamem.2004.07.006 Nasburg, J. A. et al. 6,7-Dichloro-1H-indole-2,3-dione-3-oxime functions as a superagonist for the intermediate-conductance Ca(2+)-activated K(+) channel K(Ca)3.1. Molecular pharmacology 107 , 100018 (2025). https://doi.org:10.1016/j.molpha.2025.100018 Ataman, Z. A., Gakhar, L., Sorensen, B. R., Hell, J. W. & Shea, M. A. The NMDA Receptor NR1 C1 Region Bound to Calmodulin: Structural Insights into Functional Differences between Homologous Domains. Structure 15 , 1603-1617 (2007). Hougaard, C. et al. Selective positive modulation of the SK3 and SK2 subtypes of small conductance Ca2+-activated K+ channels. British journal of pharmacology 151 , 655-665 (2007). https://doi.org:10.1038/sj.bjp.0707281 Weikl, T. R. & von Deuster, C. Selected-fit versus induced-fit protein binding: kinetic differences and mutational analysis. Proteins 75 , 104-110 (2009). https://doi.org:10.1002/prot.22223 Shim, H. et al. The Trials and Tribulations of Structure Assisted Design of K(Ca) Channel Activators. Front Pharmacol 10 , 972 (2019). https://doi.org:10.3389/fphar.2019.00972 Li, W., Halling, D. B., Hall, A. W. & Aldrich, R. W. EF hands at the N-lobe of calmodulin are required for both SK channel gating and stable SK-calmodulin interaction. J Gen Physiol 134 , 281-293 (2009). https://doi.org:10.1085/jgp.200910295 Nam, Y. W. et al. Subtype-selective positive modulation of K(Ca) 2 channels depends on the HA/HB helices. British journal of pharmacology 179 , 460-472 (2022). https://doi.org:10.1111/bph.15676 Goehring, A. et al. Screening and large-scale expression of membrane proteins in mammalian cells for structural studies. Nat Protoc 9 , 2574-2585 (2014). https://doi.org:10.1038/nprot.2014.173 Ong, S. T. et al. Design and structural basis of selective 1,4-dihydropyridine inhibitors of the calcium-activated potassium channel K(Ca)3.1. Proc Natl Acad Sci U S A 122 , e2425494122 (2025). https://doi.org:10.1073/pnas.2425494122 Punjani, A., Rubinstein, J. L., Fleet, D. J. & Brubaker, M. A. cryoSPARC: algorithms for rapid unsupervised cryo-EM structure determination. Nature methods 14 , 290-296 (2017). https://doi.org:10.1038/nmeth.4169 Afonine, P. V. et al. Real-space refinement in PHENIX for cryo-EM and crystallography. Acta Crystallogr D Struct Biol 74 , 531-544 (2018). https://doi.org:10.1107/S2059798318006551 Emsley, P., Lohkamp, B., Scott, W. G. & Cowtan, K. Features and development of Coot. Acta Crystallogr D Biol Crystallogr 66 , 486-501 (2010). https://doi.org:10.1107/S0907444910007493 Goddard, T. D. et al. UCSF ChimeraX: Meeting modern challenges in visualization and analysis. Protein Sci 27 , 14-25 (2018). https://doi.org:10.1002/pro.3235 Additional Declarations There is NO Competing Interest. Supplementary Files Validationreports.pdf Four PDB Validation reports RimSupplementary.pdf Supplementary Info Cite Share Download PDF Status: Published Journal Publication published 08 Jan, 2026 Read the published version in Nature Communications → Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-6568445","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":457173113,"identity":"2cda5e58-71f6-4be7-8dc1-9dbb0f4ce1db","order_by":0,"name":"Miao Zhang","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA0klEQVRIiWNgGAWjYDACCRBRwZAA4zM2EKflDMlaGNtI0SI/u/nZw6/z7PL4Z3cnfi5gsJHdcICAFsY5x8yNZbclF0vcObtZegZDmjFBLcwSCWbSktsOJDbcyN0gzcNwOJGgFjaJ9G/SknMOJM6/kbv5Nw/Df8JaeCRyzCQ/NhxI3HAjdxvQlgOEtUhI5JRJMxxLTtx45+w2ax6DZOOZhLTIz0jfJvmjxi5x3u3ezbd5Kuxk+whpAQFmHrB9IMKACOUgwPgDrmUUjIJRMApGARYAAHRCRVIFSwzqAAAAAElFTkSuQmCC","orcid":"https://orcid.org/0000-0002-3099-155X","institution":"Chapman University","correspondingAuthor":true,"prefix":"","firstName":"Miao","middleName":"","lastName":"Zhang","suffix":""},{"id":457173114,"identity":"f7fe9939-efc0-408d-b3e2-4a4bc548e8ab","order_by":1,"name":"Young-Woo Nam","email":"","orcid":"https://orcid.org/0000-0003-2960-9752","institution":"Chapman University","correspondingAuthor":false,"prefix":"","firstName":"Young-Woo","middleName":"","lastName":"Nam","suffix":""},{"id":457173115,"identity":"95b8276a-8ec0-4356-913b-e7eb5f540dd1","order_by":2,"name":"Alena Ramanishka","email":"","orcid":"https://orcid.org/0009-0004-5827-8627","institution":"Chapman University","correspondingAuthor":false,"prefix":"","firstName":"Alena","middleName":"","lastName":"Ramanishka","suffix":""},{"id":457173116,"identity":"1e39b274-e6a1-4775-a877-195b897628d6","order_by":3,"name":"Yang Xu","email":"","orcid":"","institution":"SLAC National Accelerator Laboratory,","correspondingAuthor":false,"prefix":"","firstName":"Yang","middleName":"","lastName":"Xu","suffix":""},{"id":457173117,"identity":"7a9e64fb-9660-4199-9eb0-75f6198c19ce","order_by":4,"name":"Rose Marie Yasuda","email":"","orcid":"https://orcid.org/0000-0002-7382-3169","institution":"Oregon Health \u0026 Science University","correspondingAuthor":false,"prefix":"","firstName":"Rose","middleName":"Marie","lastName":"Yasuda","suffix":""},{"id":457173118,"identity":"0e93cd08-f6c6-4eb8-8e21-2854c6b7b028","order_by":5,"name":"Dohyun Im","email":"","orcid":"https://orcid.org/0000-0002-6939-7718","institution":"Kyoto University","correspondingAuthor":false,"prefix":"","firstName":"Dohyun","middleName":"","lastName":"Im","suffix":""},{"id":457173119,"identity":"a9f85abd-fede-4245-8133-5fad9d95e067","order_by":6,"name":"Meng Cui","email":"","orcid":"https://orcid.org/0000-0002-3895-135X","institution":"Northeastern University","correspondingAuthor":false,"prefix":"","firstName":"Meng","middleName":"","lastName":"Cui","suffix":""},{"id":457173120,"identity":"6ce04512-b7cb-4fa8-8078-9af6adb83fd8","order_by":7,"name":"George Chandy","email":"","orcid":"","institution":"Nanyang Technological University","correspondingAuthor":false,"prefix":"","firstName":"George","middleName":"","lastName":"Chandy","suffix":""},{"id":457173121,"identity":"127b1790-ae38-47de-b086-7c5799cd4a42","order_by":8,"name":"Heike Wulff","email":"","orcid":"https://orcid.org/0000-0003-4437-5763","institution":"University of California, Davis","correspondingAuthor":false,"prefix":"","firstName":"Heike","middleName":"","lastName":"Wulff","suffix":""}],"badges":[],"createdAt":"2025-05-01 01:35:10","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-6568445/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-6568445/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1038/s41467-025-67232-3","type":"published","date":"2026-01-08T05:00:00+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":82868826,"identity":"b461f444-1b58-4c55-a3dd-9361fbebe2c5","added_by":"auto","created_at":"2025-05-16 08:24:42","extension":"jpg","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":305224,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eDifferent conformations of CaM in apo_K\u003c/strong\u003e\u003csub\u003e\u003cstrong\u003eCa\u003c/strong\u003e\u003c/sub\u003e\u003cstrong\u003e2.2 and apo_K\u003c/strong\u003e\u003csub\u003e\u003cstrong\u003eCa\u003c/strong\u003e\u003c/sub\u003e\u003cstrong\u003e3.1_II. a\u003c/strong\u003e Side view of apo_K\u003csub\u003eCa\u003c/sub\u003e2.2 (K\u003csub\u003eCa\u003c/sub\u003e2.2: blue/CaM: gray) shows channel-CaM interactions. The cytoplasmic HC helices are invisible. \u003cstrong\u003eb\u003c/strong\u003e Side view of apo_K\u003csub\u003eCa\u003c/sub\u003e3.1_II (K\u003csub\u003eCa\u003c/sub\u003e3.1: yellow/CaM: pink, PDB 6cno) shows channel-CaM interactions. The C-terminal HC helices are visible. \u003cstrong\u003ec \u003c/strong\u003eIntracellular view of apo_K\u003csub\u003eCa\u003c/sub\u003e2.2 (K\u003csub\u003eCa\u003c/sub\u003e2.2: blue cartoon/CaM: grey surface). The CaM N-lobes are positioned far apart, and the HC helices are not visible probably due to flexibility. \u003cstrong\u003ed\u003c/strong\u003e Intracellular view of apo_K\u003csub\u003eCa\u003c/sub\u003e3.1_II (K\u003csub\u003eCa\u003c/sub\u003e3.1: yellow cartoon/CaM: pink surface). The CaM N-lobes are positioned close to each other, which stabilize the HC helices in the center.\u0026nbsp;\u003c/p\u003e","description":"","filename":"1.jpg","url":"https://assets-eu.researchsquare.com/files/rs-6568445/v1/13342c9c6788f9116ba75c69.jpg"},{"id":82868828,"identity":"8a2cc781-9426-425d-976c-327489d2b3b0","added_by":"auto","created_at":"2025-05-16 08:24:42","extension":"jpg","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":222388,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eConformation of CaM in activator-bound NS309_K\u003c/strong\u003e\u003csub\u003e\u003cstrong\u003eCa\u003c/strong\u003e\u003c/sub\u003e\u003cstrong\u003e2.2 and NS309_K\u003c/strong\u003e\u003csub\u003e\u003cstrong\u003eCa\u003c/strong\u003e\u003c/sub\u003e\u003cstrong\u003e3.1 structures.\u003c/strong\u003e \u003cstrong\u003ea\u003c/strong\u003e Cryo-EM densities for Ca\u003csup\u003e2+\u003c/sup\u003e and NS309 bound to one CaM molecule in NS309_K\u003csub\u003eCa\u003c/sub\u003e2.2 are shown as blue and green mesh contoured at s = 6, respectively. CaM_Lys75 interacts with both CaM_Glu83 and NS309. \u003cstrong\u003eb\u003c/strong\u003e Cryo-EM densities for Ca\u003csup\u003e2+\u003c/sup\u003e and NS309 bound to one CaM molecule in NS309_K\u003csub\u003eCa\u003c/sub\u003e3.1\u003cstrong\u003e \u003c/strong\u003eare shown as magenta mesh contoured at s = 6. CaM_Lys75 interacts with NS309 primarily. \u0026nbsp;\u003cstrong\u003ec \u003c/strong\u003eIntracellular view of NS309_K\u003csub\u003eCa\u003c/sub\u003e2.2 (K\u003csub\u003eCa\u003c/sub\u003e2.2: purple cartoon/CaM: light blue surface). The CaM N-lobes are positioned far apart, and the HC helices are not visible probably due to flexibility.\u0026nbsp; \u003cstrong\u003ed\u003c/strong\u003e Intracellular view of NS309_K\u003csub\u003eCa\u003c/sub\u003e3.1 (K\u003csub\u003eCa\u003c/sub\u003e3.1: salmon cartoon/CaM: cyan surface). The CaM N-lobes are positioned close to each other, which stabilize the HC helices in the center.\u0026nbsp;\u003c/p\u003e","description":"","filename":"2.jpg","url":"https://assets-eu.researchsquare.com/files/rs-6568445/v1/35ba9792a77d9ce956bc21aa.jpg"},{"id":82868827,"identity":"ed5a8475-ccf6-4744-bc6f-f6bbb19a0c26","added_by":"auto","created_at":"2025-05-16 08:24:42","extension":"jpg","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":90987,"visible":true,"origin":"","legend":"\u003cp\u003eCalculated binding energy between NS309 and the binding pockets in NS309_K\u003csub\u003eCa\u003c/sub\u003e2.2 and NS309_K\u003csub\u003eCa\u003c/sub\u003e3.1. a Binding energy between NS309 and amino acid residues in CaM and the S\u003csub\u003e45\u003c/sub\u003eA helix of K\u003csub\u003eCa\u003c/sub\u003e2.2. b Binding energy between NS309 and amino acid residues in CaM and the S\u003csub\u003e45\u003c/sub\u003eA helix of K\u003csub\u003eCa\u003c/sub\u003e3.1. The total binding energy between NS309 and the four subunits of the activator-bound structure include van der Waals forces (VDW, black) and electrostatic interactions (Electrostatic, red). c Total binding energy of NS309 to binding pockets in NS309_K\u003csub\u003eCa\u003c/sub\u003e2.2 and NS309_K\u003csub\u003eCa\u003c/sub\u003e3.1, including van der Waals forces (VDW, black) and electrostatic interactions (Electrostatic, red). \u0026nbsp;d Chemical structure of NS309.\u003c/p\u003e","description":"","filename":"3.jpg","url":"https://assets-eu.researchsquare.com/files/rs-6568445/v1/e9ab60147de75d773e96bb2e.jpg"},{"id":82868831,"identity":"aca37e9d-2da8-4e5b-a109-70974326453d","added_by":"auto","created_at":"2025-05-16 08:24:42","extension":"jpg","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":204885,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eConformation of CaM in the activator-bound rimtuzalcap_K\u003c/strong\u003e\u003csub\u003e\u003cstrong\u003eCa\u003c/strong\u003e\u003c/sub\u003e\u003cstrong\u003e2.2_I structure.\u003c/strong\u003e \u003cstrong\u003ea\u003c/strong\u003e Cryo-EM densities for Ca\u003csup\u003e2+\u003c/sup\u003e and rimtuzalcap bound to one CaM molecule are shown as blue and magenta mesh, respectively. The densities are contoured at s = 6.\u003cstrong\u003e \u003c/strong\u003eSalt bridges between helix V of CaM (Glu83 and Glu87, yellow) and the S\u003csub\u003e45\u003c/sub\u003eA (Lys294) and HB (Lys467) helices of K\u003csub\u003eCa\u003c/sub\u003e2.2 (green) are present in the rimtuzalcap_K\u003csub\u003eCa\u003c/sub\u003e2.2_I structure. \u003cstrong\u003eb\u003c/strong\u003e One Ca\u003csup\u003e2+\u003c/sup\u003e-bound CaM molecule of rimtuzalcap_K\u003csub\u003eCa\u003c/sub\u003e2.2_I (CaM: yellow/Ca\u003csup\u003e2+\u003c/sup\u003e: orange) superimposed on a Ca\u003csup\u003e2+\u003c/sup\u003e-bound CaM molecule of apo_K\u003csub\u003eCa\u003c/sub\u003e2.2 (CaM: gray/Ca\u003csup\u003e2+\u003c/sup\u003e: black) with the hydrophobic surfaces at the N- and C-lobes facing away. \u003cstrong\u003ec\u003c/strong\u003e One Ca\u003csup\u003e2+\u003c/sup\u003e-bound CaM molecule of rimtuzalcap_K\u003csub\u003eCa\u003c/sub\u003e2.2_I superimposed on a Ca\u003csup\u003e2+\u003c/sup\u003e-bound CaM molecule of apo_K\u003csub\u003eCa\u003c/sub\u003e2.2 with the hydrophobic surfaces at the N- and C-lobes facing the viewer. \u003cstrong\u003e\u0026nbsp;d \u003c/strong\u003eIntracellular view of rimtuzalcap_K\u003csub\u003eCa\u003c/sub\u003e2.2_I (K\u003csub\u003eCa\u003c/sub\u003e2.2: green cartoon/CaM: yellow surface). Rimtuzalcap shortens the distances between the CaM N-lobes, which stabilize and make the HC helices visible.\u003c/p\u003e","description":"","filename":"4.jpg","url":"https://assets-eu.researchsquare.com/files/rs-6568445/v1/ce50874c16ccc345ab51d5cc.jpg"},{"id":82869322,"identity":"24e11921-b6de-49a0-bf2a-280fec929c6a","added_by":"auto","created_at":"2025-05-16 08:32:42","extension":"jpg","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":128835,"visible":true,"origin":"","legend":"\u003cp\u003eCalculated binding energy between rimtuzalcap and the binding pocket in Rimtuzalcap_K\u003csub\u003eCa\u003c/sub\u003e2.2_I. a Chemical structures of rimtuzalcap and CyPPA. b The binding pocket of rimtuzalcap in rimtuzalcap_K\u003csub\u003eCa\u003c/sub\u003e2.2_I (K\u003csub\u003eCa\u003c/sub\u003e2.2: green/CaM: yellow) superimposed onto the binding pocket of NS309 in NS309_K\u003csub\u003eCa\u003c/sub\u003e2.2 (K\u003csub\u003eCa\u003c/sub\u003e2.2: purple/CaM: light blue). Rimtuzalcap forms contacts with both the S\u003csub\u003e45\u003c/sub\u003eA and HA helices, while NS309 primarily interacts with the S\u003csub\u003e45\u003c/sub\u003eA helix. c Binding energy between rimtuzalcap and amino acid residues in CaM and the S\u003csub\u003e45\u003c/sub\u003eA helix of K\u003csub\u003eCa\u003c/sub\u003e2.2. The total binding energy between rimtuzalcap and the four subunits of the activator-bound structure include van der Waals forces (VDW, black) and electrostatic interactions (Electrostatic, red).\u003c/p\u003e","description":"","filename":"5.jpg","url":"https://assets-eu.researchsquare.com/files/rs-6568445/v1/c1579970f3c212cf7e4705dc.jpg"},{"id":82869323,"identity":"d03e6f29-75e1-4aba-bbea-d0f6f0a47ae4","added_by":"auto","created_at":"2025-05-16 08:32:43","extension":"jpg","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":122447,"visible":true,"origin":"","legend":"\u003cp\u003eThe sensitivity of K\u003csub\u003eCa\u003c/sub\u003e2.2 and K\u003csub\u003eCa\u003c/sub\u003e3.1 channels to rimtuzalcap. a Representative current traces of concentration-dependent activation of the K\u003csub\u003eCa\u003c/sub\u003e2.2_WT channel by rimtuzalcap in inside-out patch clamp recordings in the presence of 0.15 mM Ca\u003csup\u003e2+\u003c/sup\u003e. b Limited responses to rimtuzalcap of the K\u003csub\u003eCa\u003c/sub\u003e3.1_WT channel in inside-out patch clamp recordings in the presence of 0.15 mM Ca\u003csup\u003e2+\u003c/sup\u003e. c Representative current traces of concentration-dependent activation of the K\u003csub\u003eCa\u003c/sub\u003e3.1_R355K mutant channel by rimtuzalcap in inside-out patch clamp recordings in the presence of 0.15 mM Ca\u003csup\u003e2+\u003c/sup\u003e. d Responses of K\u003csub\u003eCa\u003c/sub\u003e3.1_WT and K\u003csub\u003eCa\u003c/sub\u003e3.1_R355K channels versus K\u003csub\u003eCa\u003c/sub\u003e2.2_WT channels to rimtuzalcap. Rimtuzalcap activates K\u003csub\u003eCa\u003c/sub\u003e3.1_R355K (EC\u003csub\u003e50\u003c/sub\u003e: 76.8 ± 21.3 mM, n=8) with ~15-fold lower potency than K\u003csub\u003eCa\u003c/sub\u003e2.2_WT (EC\u003csub\u003e50\u003c/sub\u003e: 5.1 ± 1.0 mM, n=5, P\u0026lt;0.0001, unpaired two-tailed Student’s \u003cem\u003et\u003c/em\u003e-test). The responses were normalized by the maximal currents induced by 10 μM Ca\u003csup\u003e2+\u003c/sup\u003e.\u003c/p\u003e","description":"","filename":"6.jpg","url":"https://assets-eu.researchsquare.com/files/rs-6568445/v1/e5feb9c7587b6f6323ad8333.jpg"},{"id":82868840,"identity":"1aaab91e-abc9-402d-8ba4-e6bbcad33e56","added_by":"auto","created_at":"2025-05-16 08:24:43","extension":"jpg","order_by":7,"title":"Figure 7","display":"","copyAsset":false,"role":"figure","size":211021,"visible":true,"origin":"","legend":"\u003cp\u003eInner gate of K\u003csub\u003eCa\u003c/sub\u003e2.2 channels regulated by pharmacological agents. In side views, dimensions of the inner gate are measured as distances between Val391 in the transmembrane S6 helices of opposite K\u003csub\u003eCa\u003c/sub\u003e2.2 subunits in a apo_K\u003csub\u003eCa\u003c/sub\u003e2.2, b NS309_K\u003csub\u003eCa\u003c/sub\u003e2.2, c rimtuzalcap_K\u003csub\u003eCa\u003c/sub\u003e2.2_I, and d AP14145_K\u003csub\u003eCa\u003c/sub\u003e2.2 structures. Two opposite channel subunits are shown for clarity.\u003c/p\u003e","description":"","filename":"7.jpg","url":"https://assets-eu.researchsquare.com/files/rs-6568445/v1/2217baa42d00ec689d487c30.jpg"},{"id":100296087,"identity":"a7c98ecc-056e-4d0a-a20b-2d316e877f87","added_by":"auto","created_at":"2026-01-15 08:10:42","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":2311058,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-6568445/v1/3a002059-3bdd-4dc7-a736-04380038b7d8.pdf"},{"id":82868837,"identity":"50db6c46-7e56-48fc-9a11-2444556bb12b","added_by":"auto","created_at":"2025-05-16 08:24:43","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":5412321,"visible":true,"origin":"","legend":"\u003cp\u003eFour PDB Validation reports\u003c/p\u003e","description":"","filename":"Validationreports.pdf","url":"https://assets-eu.researchsquare.com/files/rs-6568445/v1/2c7efbfdc3643db0b3884626.pdf"},{"id":82868851,"identity":"bf3564b0-658e-4ded-b4fc-6c1af77258b1","added_by":"auto","created_at":"2025-05-16 08:24:43","extension":"pdf","order_by":2,"title":"","display":"","copyAsset":false,"role":"supplement","size":8171906,"visible":true,"origin":"","legend":"Supplementary Info","description":"","filename":"RimSupplementary.pdf","url":"https://assets-eu.researchsquare.com/files/rs-6568445/v1/dfffe8e73db3e8ea7d5b812a.pdf"}],"financialInterests":"There is \u003cb\u003eNO\u003c/b\u003e Competing Interest.","formattedTitle":"\u003cp\u003eStructural basis for the subtype-selectivity of K\u003csub\u003eCa\u003c/sub\u003e2.2 channel activators\u003c/p\u003e","fulltext":[{"header":"Introduction","content":"\u003cp\u003ePharmacological activation of K\u003csup\u003e+\u003c/sup\u003e channels can dampen electrical signaling and may have therapeutic value for diverse diseases. Pharmacological agents that activate voltage-gated (K\u003csub\u003eV\u003c/sub\u003e)\u003csup\u003e\u003cspan additionalcitationids=\"CR2 CR3 CR4 CR5\" citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e\u003c/sup\u003e, inward-rectifier (K\u003csub\u003eIR\u003c/sub\u003e)\u003csup\u003e\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e\u003c/sup\u003e, two-pore domain (K\u003csub\u003e2P\u003c/sub\u003e)\u003csup\u003e\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e\u003c/sup\u003e, and large conductance (K\u003csub\u003eCa\u003c/sub\u003e1.1)\u003csup\u003e\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e\u003c/sup\u003e K\u003csup\u003e+\u003c/sup\u003e channels widen the inner gate at the intracellular entrance to the channel pore, modulate the voltage-sensing domains, or regulate gating at the selectivity filter. The mechanism underlying the action of pharmacological activators of small-conductance Ca\u003csup\u003e2+\u003c/sup\u003e-activated K\u003csup\u003e+\u003c/sup\u003e (K\u003csub\u003eCa\u003c/sub\u003e2.1-K\u003csub\u003eCa\u003c/sub\u003e2.3, also called SK1-SK3) channels, an important sub-family of K\u003csup\u003e+\u003c/sup\u003e channels with distinctive biophysical and pharmacological properties, has not been defined.\u003c/p\u003e \u003cp\u003eK\u003csub\u003eCa\u003c/sub\u003e2.x channels and the related K\u003csub\u003eCa\u003c/sub\u003e3.1 channel are encoded by the \u003cem\u003eKCNN1-4\u003c/em\u003e gene family. K\u003csub\u003eCa\u003c/sub\u003e2.x channels are critical modulators of neuronal and cardiac excitability, while KCa3.1 sustains Ca\u003csup\u003e2+\u003c/sup\u003e signaling through K\u0026thinsp;+\u0026thinsp;efflux-driven hyperpolarization in peripheral tissues, including erythrocytes, immune cells and vascular endothelium.\u003csup\u003e\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e\u003c/sup\u003e The shared Ca\u003csup\u003e2+\u003c/sup\u003e-calmodulin (CaM) dependent gating mechanism of these channels has been demonstrated by high resolution structure-determination using cryogenic electron microscopy (cryo-EM). \u003csup\u003e\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e,\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e\u003c/sup\u003e In the Ca\u003csup\u003e2+\u003c/sup\u003e-free state, CaM\u0026rsquo;s N-lobe is flexible and invisible in the cryo-EM structure, while CaM\u0026rsquo;s C-lobe interacts with the channel\u0026rsquo;s cytoplasmic HA/HB helices (previously called CaM binding domain). \u003csup\u003e\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e\u003c/sup\u003e During Ca\u003csup\u003e2+\u003c/sup\u003e-dependent activation, the calcified CaM N-lobe swings up and binds to the channel\u0026rsquo;s S4-S5 linker (primarily the S\u003csub\u003e45\u003c/sub\u003eA helix), causing the inner gate to open, allowing K\u003csup\u003e+\u003c/sup\u003e to flow through the channel pore. \u003csup\u003e\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e,\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e\u003c/sup\u003e\u003c/p\u003e \u003cp\u003eSignificant effort spanning decades has identified pharmacological activators of K\u003csub\u003eCa\u003c/sub\u003e2.x and K\u003csub\u003eCa\u003c/sub\u003e3.1 channels. These include compounds such as 1-EBIO, CyPPA, NS309, and GW-542573X. \u003csup\u003e\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e\u003c/sup\u003e Earlier crystallographic studies suggested that these activators bound to the interface between CaM\u0026rsquo;s N-lobe and the channel\u0026rsquo;s cytoplasmic HA/HB helices.\u003csup\u003e\u003cspan additionalcitationids=\"CR14\" citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e\u003c/sup\u003e However, these studies were performed with truncated cytoplasmic HA/HB helices in complex with CaM and did not include the remainder of the channel. Their mechanism of activation therefore needs to be redefined within the context of the full-length structures. Here, we combined cryo-EM, site-directed mutagenesis, and electrophysiology to define the mechanism of action of rimtuzalcap, a CyPPA derivative that activates selectively K\u003csub\u003eCa\u003c/sub\u003e2.2/K\u003csub\u003eCa\u003c/sub\u003e2.3 channels, and NS309, a non-selective activator of both K\u003csub\u003eCa\u003c/sub\u003e2.x and K\u003csub\u003eCa\u003c/sub\u003e3.1 channels. Our studies show that both activators widen the inner gate of the K\u003csub\u003eCa\u003c/sub\u003e2.2 channel, while differences in their interactions with the Ca\u003csup\u003e2+\u003c/sup\u003e-CaM-dependent gating machinery underlie rimtuzalcap\u0026rsquo;s selectivity for K\u003csub\u003eCa\u003c/sub\u003e2.2/K\u003csub\u003eCa\u003c/sub\u003e2.3 channels.\u003c/p\u003e"},{"header":"Results","content":"\u003cp\u003e\u003cstrong\u003eDifferential interactions of CaM with the HC helices of K\u003csub\u003eCa\u003c/sub\u003e2.2 and K\u003csub\u003eCa\u003c/sub\u003e3.1\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWe compared our recently determined cryo-EM structure of Ca\u003csup\u003e2+\u003c/sup\u003e-bound apo_K\u003csub\u003eCa\u003c/sub\u003e2.2/CaM (resolution = 3.18 Å; referred to henceforth as apo_K\u003csub\u003eCa\u003c/sub\u003e2.2; Fig. 1a) \u003csup\u003e11\u003c/sup\u003e with the previously reported cryo-EM structure of Ca\u003csup\u003e2+\u003c/sup\u003e-bound apo_K\u003csub\u003eCa\u003c/sub\u003e3.1/CaM activation state II \u003csup\u003e12\u003c/sup\u003e (Protein Data Bank [PDB]: 6cno; referred to henceforth as apo_K\u003csub\u003eCa\u003c/sub\u003e3.1_II; Fig. 1b). The cytoplasmic HC helices are invisible in the apo_K\u003csub\u003eCa\u003c/sub\u003e2.2 structure possibly due to flexibility (Fig. 1a). In contrast, the C-terminal HC helices are well resolved and are seen between CaM molecules in the apo_K\u003csub\u003eCa\u003c/sub\u003e3.1_II structure (Fig. 1b).\u003c/p\u003e\n\u003cp\u003eThe distinct conformations of the HC helices in the two channels may be attributable to different conformations of CaM in the two structures. In apo_K\u003csub\u003eCa\u003c/sub\u003e2.2, CaM’s N-lobes are too far apart (Asn42 in opposite CaM molecules are ~35.0 Å apart) to stabilize the HC helices, which may contribute to their flexibility (Fig. 1c). In the apo_K\u003csub\u003eCa\u003c/sub\u003e3.1_II structure, in contrast, CaM’s N-lobes are sufficiently close (Asn42 in opposite CaM molecules are ~27.0 Å apart) to stabilize the HC helices, causing them to be visible (Fig. 1d). As will be seen below, these differences between apo_K\u003csub\u003eCa\u003c/sub\u003e2.2 and apo_K\u003csub\u003eCa\u003c/sub\u003e3.1_II contribute to the selectivity of rimtuzalcap for K\u003csub\u003eCa\u003c/sub\u003e2.2 over K\u003csub\u003eCa\u003c/sub\u003e3.1.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eThe non-selective NS309 interacts similarly with K\u003csub\u003eCa\u003c/sub\u003e3.1/CaM\u003c/strong\u003e \u003cstrong\u003eand K\u003csub\u003eCa\u003c/sub\u003e2.2/CaM\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNS309 activates both K\u003csub\u003eCa\u003c/sub\u003e3.1 (EC\u003csub\u003e50\u003c/sub\u003e: ~74 nM) and K\u003csub\u003eCa\u003c/sub\u003e2.2 (EC\u003csub\u003e50\u003c/sub\u003e: ~1.7\u0026nbsp;mM) channels. \u003csup\u003e16,17\u003c/sup\u003e We determined structures of NS309 bound to K\u003csub\u003eCa\u003c/sub\u003e2.2/CaM and to K\u003csub\u003eCa\u003c/sub\u003e3.1/CaM. The cryo-EM map of the Ca\u003csup\u003e2+\u003c/sup\u003e- and NS309-bound K\u003csub\u003eCa\u003c/sub\u003e2.2/CaM complex (henceforth referred to as NS309_K\u003csub\u003eCa\u003c/sub\u003e2.2) was refined to a resolution of 2.71 Å (SupplementaryFig. 1,2,SupplementaryTable 1). The cryo-EM map of the Ca\u003csup\u003e2+\u003c/sup\u003e- and NS309-bound K\u003csub\u003eCa\u003c/sub\u003e3.1/CaM complex (henceforth referred to as NS309_K\u003csub\u003eCa\u003c/sub\u003e3.1) was refined to a resolution of 3.59 Å (SupplementaryFig. 3,4,SupplementaryTable 1). Cryo-EM densities for NS309 in each of the four CaM’s N-lobes are clearly visible in both NS309_K\u003csub\u003eCa\u003c/sub\u003e2.2 (Supplementary Fig. 5a,b) and NS309_K\u003csub\u003eCa\u003c/sub\u003e3.1 (Supplementary Fig. 6a,b). \u0026nbsp;We successfully built models for NS309 into the cryo-EM densities in both structures at the interface between CaM’s N-lobe and the channels’ S\u003csub\u003e45\u003c/sub\u003eA helix (Fig. 2a,b). Since CaM’s N-lobe forms a significant portion of NS309’s binding pocket, we compared the conformations of CaM in the apo versus activator-bound structures. CaM molecules aligned well in the apo_K\u003csub\u003eCa\u003c/sub\u003e2.2 and NS309_K\u003csub\u003eCa\u003c/sub\u003e2.2 comparison (rmsd = 1.4 Å, Supplementary Fig. 7a,b), and even better in the apo_K\u003csub\u003eCa\u003c/sub\u003e3.1_II and NS309_K\u003csub\u003eCa\u003c/sub\u003e3.1 comparison (rmsd = 1.0 Å, Supplementary Fig. 7c,d). Binding of NS309 to the K\u003csub\u003eCa\u003c/sub\u003e2.2/CaM complex shortened the distance between CaM’s N-lobes from ~35.0 Å in apo_K\u003csub\u003eCa\u003c/sub\u003e2.2 (Fig. 1c) to ~31.0 Å in NS309_K\u003csub\u003eCa\u003c/sub\u003e2.2 (Fig. 2c). Binding of NS309 to the K\u003csub\u003eCa\u003c/sub\u003e3.1/CaM complex did not affect the distance between CaM’s N-lobes (Fig. 2d) compared to the apo_K\u003csub\u003eCa\u003c/sub\u003e3.1_II structure (Fig. 1d). The HC helices are visible in both apo_K\u003csub\u003eCa\u003c/sub\u003e3.1_II (Fig. 1d) and NS309_K\u003csub\u003eCa\u003c/sub\u003e3.1 (Fig. 2d), while they remain invisible in NS309_K\u003csub\u003eCa\u003c/sub\u003e2.2 (Fig. 2c). \u0026nbsp;Based on these results, we conclude that the binding pocket for NS309 preexists in both K\u003csub\u003eCa\u003c/sub\u003e2.2 and K\u003csub\u003eCa\u003c/sub\u003e3.1 channels.\u003c/p\u003e\n\u003cp\u003eWe calculated the binding energy (van der Waals forces plus electrostatic interactions) between the four bound NS309 molecules and amino acid residues in the four subunits of the activator-bound structures of K\u003csub\u003eCa\u003c/sub\u003e2.2 and K\u003csub\u003eCa\u003c/sub\u003e3.1 using the Discovery Studio program. In the NS309_K\u003csub\u003eCa\u003c/sub\u003e2.2 structure, NS309 sits at the interface between CaM’s N-lobe and K\u003csub\u003eCa\u003c/sub\u003e2.2’s S\u003csub\u003e45\u003c/sub\u003eA helix, \u0026nbsp;with Ser288 and Leu292 in K\u003csub\u003eCa\u003c/sub\u003e2.2’s S\u003csub\u003e45\u003c/sub\u003eA helix interacting with NS309, and a tetrad of hydrophobic residues (Phe19, Leu32, Met51, and Met71; FLMM\u003csub\u003eN\u003c/sub\u003e \u003csup\u003e18\u003c/sup\u003e) in CaM’s N-lobe cradling the NS309 molecule (Fig. 3a). In the NS309_K\u003csub\u003eCa\u003c/sub\u003e3.1 structure, NS309 fits perfectly into a hydrophobic pocket between CaM’s N-lobe and K\u003csub\u003eCa\u003c/sub\u003e3.1’s S\u003csub\u003e45\u003c/sub\u003eA helix, with Ser181 and Leu185 in K\u003csub\u003eCa\u003c/sub\u003e3.1’s S\u003csub\u003e45\u003c/sub\u003eA helix interacting with NS309, and the same tetrad of hydrophobic residues (Phe19, Leu32, Met51, and Met71; FLMM\u003csub\u003eN\u003c/sub\u003e \u003csup\u003e18\u003c/sup\u003e) in CaM’s N-lobe cradling the NS309 molecule (Fig. 3b). These results are consistent with earlier reports that mutations of Ser181 and Leu185 in K\u003csub\u003eCa\u003c/sub\u003e3.1 and mutations of Ser288 and Leu292 in K\u003csub\u003eCa\u003c/sub\u003e2.2 significantly reduced sensitivity to NS309.\u003csup\u003e17\u003c/sup\u003e\u003c/p\u003e\n\u003cp\u003eThe total binding energies were -27.6\u0026nbsp;±\u0026nbsp;1.0 kCal/mol and -29.7\u0026nbsp;±\u0026nbsp;1.1 kCal/mol for NS309_K\u003csub\u003eCa\u003c/sub\u003e2.2 and NS309_K\u003csub\u003eCa\u003c/sub\u003e3.1, respectively (P=0.03, n=4, unpaired two-tailed Student’s \u003cem\u003et\u003c/em\u003e-test, Fig. 3c). The van der Waals forces involved in NS309’s interactions with the two channels are comparable (NS309_K\u003csub\u003eCa\u003c/sub\u003e2.2: -24.9\u0026nbsp;±\u0026nbsp;0.8 kCal/mol; NS309_K\u003csub\u003eCa\u003c/sub\u003e3.1: -23.7\u0026nbsp;±\u0026nbsp;0.7 kCal/mol; P=0.06, n=4, unpaired two-tailed Student’s \u003cem\u003et\u003c/em\u003e-test). NS309’s stronger electrostatic interactions with K\u003csub\u003eCa\u003c/sub\u003e3.1 (-6.0\u0026nbsp;±\u0026nbsp;0.7 kCal/mol) versus K\u003csub\u003eCa\u003c/sub\u003e2.2 (-2.7\u0026nbsp;±\u0026nbsp;0.3 kCal/mol, P=0.0003, n=4, unpaired two-tailed Student’s \u003cem\u003et\u003c/em\u003e-test, Fig. 3c) account for NS309’s stronger total binding energy to K\u003csub\u003eCa\u003c/sub\u003e3.1 than K\u003csub\u003eCa\u003c/sub\u003e2.2. In both structures, CaM_Lys75 is the largest contributor to the electrostatic interactions with NS309 (Fig. 3a,b). Electrostatic interactions between CaM_Lys75 and NS309 are stronger in NS309_K\u003csub\u003eCa\u003c/sub\u003e3.1 (-4.9\u0026nbsp;±\u0026nbsp;0.8 kCal/mol) than in NS309_K\u003csub\u003eCa\u003c/sub\u003e2.2 (-1.1\u0026nbsp;±\u0026nbsp;0.5 kCal/mol, P=0.0002, n=4, unpaired two-tailed Student’s \u003cem\u003et\u003c/em\u003e-test). NS309’s (Fig. 3d) higher binding energy to K\u003csub\u003eCa\u003c/sub\u003e3.1 than K\u003csub\u003eCa\u003c/sub\u003e2.2 is consistent with NS309’s ~20-fold higher potency in activating K\u003csub\u003eCa\u003c/sub\u003e3.1 than K\u003csub\u003eCa\u003c/sub\u003e2.2. \u003csup\u003e16,17\u003c/sup\u003e\u003c/p\u003e\n\u003cp\u003eIn NS309_K\u003csub\u003eCa\u003c/sub\u003e2.2, the compact conformation of CaM is stabilized by salt bridges between CaM’s C-lobe (Glu83 and Glu 87 in helix V) and the K\u003csub\u003eCa\u003c/sub\u003e2.2 channel’s S\u003csub\u003e45\u003c/sub\u003eA helix (Lys294) and HB helix (Lys467) (Fig. 2a). CaM_Lys75 forms a salt bridge (~3.2 Å) with CaM_Glu83, which may weaken its hydrogen bond with NS309 (~4.3 Å, Fig. 2a). \u0026nbsp;In NS309_K\u003csub\u003eCa\u003c/sub\u003e3.1, CaM’s more extended conformation is stabilized by salt bridges between CaM’s C-lobe (Glu83 and Glu87 in helix V) and K\u003csub\u003eCa\u003c/sub\u003e3.1’s HB helix (Arg355) (Fig. 2b). CaM_Lys75 is ~6.8 Å away from CaM_Glu83, while it is much closer to NS309 (~3.0 Å, Fig. 2b). In summary, both K\u003csub\u003eCa\u003c/sub\u003e2.2 and K\u003csub\u003eCa\u003c/sub\u003e3.1 can accommodate NS309 into pre-existing binding pockets at the interface between CaM’s N-lobe and the channels’ S\u003csub\u003e45\u003c/sub\u003eA helix, which may explain the non-selective activation of K\u003csub\u003eCa\u003c/sub\u003e2.x and K\u003csub\u003eCa\u003c/sub\u003e3.1 channels by NS309. \u003csup\u003e16,17\u003c/sup\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eStructure of K\u003csub\u003eCa\u003c/sub\u003e2.2/CaM\u003c/strong\u003e \u003cstrong\u003ebound to the subtype-selective activator rimtuzalcap\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eCyPPA, a positive allosteric modulator, potentiates the activity of K\u003csub\u003eCa\u003c/sub\u003e2.2 and K\u003csub\u003eCa\u003c/sub\u003e2.3, but is inactive on K\u003csub\u003eCa\u003c/sub\u003e3.1. \u003csup\u003e19\u003c/sup\u003e Rimtuzalcap\u0026nbsp;is a CyPPA derivative that potentiates K\u003csub\u003eCa\u003c/sub\u003e2.2 at low micromolar concentrations (EC\u003csub\u003e50\u003c/sub\u003e: ~5.1\u0026nbsp;mM) and is inactive on K\u003csub\u003eCa\u003c/sub\u003e3.1. \u0026nbsp;Rimtuzalcap\u0026nbsp;was evaluated in a Phase-2 clinical trial as a treatment for essential tremor (ClinicalTrials.gov: NCT03688685), and a phase-2 clinical trial evaluating rimtuzalcap in patients with spinocerebellar ataxia was initiated (ClinicalTrials.gov: NCT03688685) and then withdrawn when Cadent Therapeutics was acquired by Novartis.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eWe refined the cryo-EM map of the Ca\u003csup\u003e2+\u003c/sup\u003e- and rimtuzalcap-bound K\u003csub\u003eCa\u003c/sub\u003e2.2/CaM (henceforth referred to as rimtuzalcap_K\u003csub\u003eCa\u003c/sub\u003e2.2_I) to a resolution of 3.13 Å (SupplementaryFig. 8, 9, Supplementary Table 1). Cryo-EM densities for rimtuzalcap in each of the four CaM’s N-lobes are clearly visible in rimtuzalcap_K\u003csub\u003eCa\u003c/sub\u003e2.2_I (SupplementaryFig. 10a,b). We successfully built a model for rimtuzalcap into its cryo-EM density. Rimtuzalcap sits at the interface between CaM’s N-lobe and K\u003csub\u003eCa\u003c/sub\u003e2.2’s S\u003csub\u003e45\u003c/sub\u003eA/HA helices in each subunit of the tetrameric channel (Fig. 4a). Similar to the NS309_K\u003csub\u003eCa\u003c/sub\u003e2.2 structure (Fig. 2a), the rimtuzalcap_K\u003csub\u003eCa\u003c/sub\u003e2.2_I structure shows that the salt bridges between CaM helix V (Glu83 and Glu87), the S\u003csub\u003e45\u003c/sub\u003eA (Lys294), and the HB (Lys467) helices of K\u003csub\u003eCa\u003c/sub\u003e2.2 are present (Fig. 4a).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eLarge differences are apparent in the alignment of CaM’s\u0026nbsp;a-carbons (rmsd = 2.6 Å, Fig. 4b,c) in rimtuzalcap_K\u003csub\u003eCa\u003c/sub\u003e2.2_I versus apo_K\u003csub\u003eCa\u003c/sub\u003e2.2. Binding of rimtuzalcap expands CaM’s N-lobes, shortens the distances between the N-lobes in opposite subunits from ~35 Å in apo_K\u003csub\u003eCa\u003c/sub\u003e2.2 (Fig. 1c) to ~22 Å in rimtuzalcap_K\u003csub\u003eCa\u003c/sub\u003e2.2_I, and stabilizes K\u003csub\u003eCa\u003c/sub\u003e2.2’s cytoplasmic HC helices, rendering them visible (Fig. 4d). These observations suggest that a substantial conformational change is required to accommodate rimtuzalcap into its binding pocket in K\u003csub\u003eCa\u003c/sub\u003e2.2/CaM. In contrast, the conformational change required to fit NS309 into its binding pocket in K\u003csub\u003eCa\u003c/sub\u003e2.2 channels is minimal because comparison of the NS309_K\u003csub\u003eCa\u003c/sub\u003e2.2 versus apo_K\u003csub\u003eCa\u003c/sub\u003e2.2 structures shows CaM’s\u0026nbsp;a-carbons in good alignment (rmsd = 1.4 Å, Supplementary Fig. 7a,b), small changes in the distances between opposite CaM’ N-lobes (~31 Å versus ~35 Å, Figs. 1c, 2c), and cytoplasmic HC helices that remain flexible and invisible (Figs. 1c, 2c).\u003c/p\u003e\n\u003cp\u003eNS309 primarily interacts with K\u003csub\u003eCa\u003c/sub\u003e2.2’s S\u003csub\u003e45\u003c/sub\u003eA helix (Fig. 3), whereas the bulkier rimtuzalcap (Fig. 5a) forms contacts with K\u003csub\u003eCa\u003c/sub\u003e2.2’s HA helix in addition to the S\u003csub\u003e45\u003c/sub\u003eA helix (Fig. 5b). Rimtuzalcap fits perfectly into the hydrophobic pocket formed by residues in K\u003csub\u003eCa\u003c/sub\u003e2.2’s S\u003csub\u003e45\u003c/sub\u003eA helix (Ile289, Leu292, Asn293), K\u003csub\u003eCa\u003c/sub\u003e2.2’s HA helix in a neighboring subunit (His406, Phe410), and CaM’s N-lobe (Phe19, Leu32, Met36, Met51) (Fig. 5c). The binding energy between rimtuzalcap and its induced-fit binding pocket in K\u003csub\u003eCa\u003c/sub\u003e2.2 (-36.9\u0026nbsp;±\u0026nbsp;1.3 kCal/mol) is stronger than NS309’s binding energy to its pre-existing binding pocket in K\u003csub\u003eCa\u003c/sub\u003e2.2 (-27.6\u0026nbsp;±\u0026nbsp;1.0 kCal/mol). However, NS309 is a more potent activator of K\u003csub\u003eCa\u003c/sub\u003e2.2 (EC\u003csub\u003e50\u003c/sub\u003e: ~1.7\u0026nbsp;mM) than rimtuzalcap (EC\u003csub\u003e50\u003c/sub\u003e: ~5.1\u0026nbsp;mM) in electrophysiological assays. The discordance between binding energies and activation-potency of these two compounds may be due inherent variations in ligand-binding to pre-existing (NS309) versus induced-fit (rimtuzalcap) pockets \u003csup\u003e20\u003c/sup\u003e and to differences in ligand efficiency caused by differences in size of the two molecules (Figs. 3d, 5a).\u003c/p\u003e\n\u003cp\u003eUnlike the K\u003csub\u003eCa\u003c/sub\u003e2.2 channel (Fig. 6a), the K\u003csub\u003eCa\u003c/sub\u003e3.1 channel is insensitive to rimtuzalcap (Fig. 6b). In the apo_K\u003csub\u003eCa\u003c/sub\u003e3.1_II structure, CaM’s N-lobes are close to each other (~27.0 Å apart, Fig. 1d) and the putative binding pocket in CaM’s N-lobe is constrained by the HC helices, which may prevent it from undergoing conformational changes required to accommodate rimtuzalcap. We therefore wondered if rearranging the salt bridges between K\u003csub\u003eCa\u003c/sub\u003e3.1’s HB helix and CaM’s helix V would release the putative binding pocket from the constraints of the HC helices and allow rimtuzalcap binding. To test this idea, we mutated Arg355 in K\u003csub\u003eCa\u003c/sub\u003e3.1’s HB helix to lysine to mimic the salt bridge-forming Lys467 in K\u003csub\u003eCa\u003c/sub\u003e2.2 (Fig. 4a). In support of our idea, the mutant K\u003csub\u003eCa\u003c/sub\u003e3.1_R355K channel was activated by rimtuzalcap with an EC\u003csub\u003e50\u003c/sub\u003e = 76.8\u0026nbsp;±\u0026nbsp;21.3\u0026nbsp;mM (n=8; Fig. 6c,d). Additional structural determinants may underlie rimtuzalcap’s ~15-fold lower potency on K\u003csub\u003eCa\u003c/sub\u003e3.1_R355K compared to K\u003csub\u003eCa\u003c/sub\u003e2.2_WT (EC\u003csub\u003e50\u003c/sub\u003e: 5.1\u0026nbsp;±\u0026nbsp;1.0\u0026nbsp;mM, n=5, P\u0026lt;0.0001, unpaired two-tailed Student’s \u003cem\u003et\u003c/em\u003e-test).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eWidening of the inner gate by NS309 and rimtuzalcap\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eRimtuzalcap and NS309 widened K\u003csub\u003eCa\u003c/sub\u003e2.2’s inner gate from ~12.6 Å in the apo-form (apo_K\u003csub\u003eCa\u003c/sub\u003e2.2) to ~13.5 Å and ~13.1 Å in the rimtuzalcap_K\u003csub\u003eCa\u003c/sub\u003e2.2_I and NS309_K\u003csub\u003eCa\u003c/sub\u003e2.2 structures, respectively (Fig. 7). In our previous cryo-EM study, the inhibitor AP14145 narrowed the inner gate to ~6.5 Å (PDB: 8v2h; referred to henceforth as AP14145_K\u003csub\u003eCa\u003c/sub\u003e2.2; Fig. 7). Thus, rimtuzalcap and NS309 both activate K\u003csub\u003eCa\u003c/sub\u003e2.2 by widening the inner gate.\u0026nbsp;Differences in their\u0026nbsp;interactions with the Ca\u003csup\u003e2+\u003c/sup\u003e-CaM-dependent gating machinery of K\u003csub\u003eCa\u003c/sub\u003e2.2 versus K\u003csub\u003eCa\u003c/sub\u003e3.1 likely underlie rimtuzalcap’s K\u003csub\u003eCa\u003c/sub\u003e2.x-selectivity and NS309’s non-selective activation of both K\u003csub\u003eCa\u003c/sub\u003e2.2 and K\u003csub\u003eCa\u003c/sub\u003e3.1.\u003c/p\u003e\n\u003cp\u003eActivators of K\u003csub\u003eCa\u003c/sub\u003e2.2\u0026nbsp;have been termed positive allosteric modulators because they work only in the presence of Ca\u003csup\u003e2+\u003c/sup\u003e. \u003csup\u003e19,21,22\u003c/sup\u003e Both\u0026nbsp;NS309 and rimtuzalcap (Fig. 6) require a minimal concentration (~0.1-0.2 mM) of Ca\u003csup\u003e2+\u003c/sup\u003e to enhance the activity of K\u003csub\u003eCa\u003c/sub\u003e2.2 channels. This requirement for\u0026nbsp;Ca\u003csup\u003e2+\u003c/sup\u003e is illustrated by comparison of two rimtuzalcap-bound K\u003csub\u003eCa\u003c/sub\u003e2.2 structures built using cryo-EM maps of two 3D classes from the same dataset (Supplementary Fig. 8, 9). The density for rimtuzalcap is visible in both rimtuzalcap_K\u003csub\u003eCa\u003c/sub\u003e2.2_I and rimtuzalcap_K\u003csub\u003eCa\u003c/sub\u003e2.2_II (Supplementary Fig. 10), whereas Ca\u003csup\u003e2+\u003c/sup\u003e ions are present only in CaM’s N-lobes of rimtuzalcap_K\u003csub\u003eCa\u003c/sub\u003e2.2_I (Supplementary Fig. 10a,b) and not rimtuzalcap_K\u003csub\u003eCa\u003c/sub\u003e2.2_II (Supplementary Fig. 10c-e). Rimtuzalcap_K\u003csub\u003eCa\u003c/sub\u003e2.2_II that lacks Ca\u003csup\u003e2+\u003c/sup\u003e in CaM’s N-lobes exhibits a closed inner gate (~6.9 Å), whereas rimtuzalcap_K\u003csub\u003eCa\u003c/sub\u003e2.2_I that contains Ca\u003csup\u003e2+\u003c/sup\u003e in CaM’s N-lobes (Supplementary Fig. 10f,g) exhibits a widened inner gate (~13.5 Å). Taken together, these results suggest that binding of\u0026nbsp;NS309 and rimtuzalcap to the Ca\u003csup\u003e2+\u003c/sup\u003e-bound CaM N-lobe stabilizes its interaction with K\u003csub\u003eCa\u003c/sub\u003e2.2’s S\u003csub\u003e45\u003c/sub\u003eA helix and dilates the channel’s inner gate.\u003c/p\u003e"},{"header":"Discussion","content":"\u003cp\u003eEven though K\u003csub\u003eCa\u003c/sub\u003e2.2 and K\u003csub\u003eCa\u003c/sub\u003e3.1 channels share a Ca\u003csup\u003e2+\u003c/sup\u003e/CaM-dependent gating mechanism, CaM interacts differently with the cytoplasmic HC helices of\u0026nbsp;K\u003csub\u003eCa\u003c/sub\u003e2.2 and K\u003csub\u003eCa\u003c/sub\u003e3.1. In apo_K\u003csub\u003eCa\u003c/sub\u003e2.2, the CaM N-lobes in opposite subunits are too far apart to stabilize K\u003csub\u003eCa\u003c/sub\u003e2.2’s HC helices (Fig. 1c). In apo_K\u003csub\u003eCa\u003c/sub\u003e3.1_II, in contrast, the CaM N-lobes in opposite subunits are close enough to stabilize K\u003csub\u003eCa\u003c/sub\u003e3.1’s HC helices (Fig. 1d). The distinct conformations of CaM in the two channel structures underlie the subtype-selectivity of pharmacological activators.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eAccommodation of NS309 into its binding pockets in K\u003csub\u003eCa\u003c/sub\u003e2.2 and K\u003csub\u003eCa\u003c/sub\u003e3.1 does not require a prominent conformational change within CaM as evidenced by the excellent alignment of CaM’s\u0026nbsp;a-carbons between the apo- and NS309-bound K\u003csub\u003eCa\u003c/sub\u003e2.2 and K\u003csub\u003eCa\u003c/sub\u003e3.1 structures (Supplementary Fig. 7). This indicates that NS309 fits into pre-existing binding pockets in both NS309_K\u003csub\u003eCa\u003c/sub\u003e3.1 and NS309_K\u003csub\u003eCa\u003c/sub\u003e2.2. \u0026nbsp;NS309’s stronger electrostatic interactions with CaM_Lys75 (Figs. 3a,b) accounts for its higher total binding energy to NS309_K\u003csub\u003eCa\u003c/sub\u003e3.1 versus NS309_K\u003csub\u003eCa\u003c/sub\u003e2.2 (Fig. 3c). This difference in binding strength likely underlies NS309’s ~20-fold higher potency in activating K\u003csub\u003eCa\u003c/sub\u003e3.1 than K\u003csub\u003eCa\u003c/sub\u003e2.2 channels. \u003csup\u003e16,17\u003c/sup\u003e\u003c/p\u003e\n\u003cp\u003eIn contrast, accommodation of rimtuzalcap into its binding pockets in K\u003csub\u003eCa\u003c/sub\u003e2.2 requires significant conformational changes within CaM as evidenced by large differences in the alignment of CaM’s\u0026nbsp;a-carbons between the apo_K\u003csub\u003eCa\u003c/sub\u003e2.2 and rimtuzalcap_K\u003csub\u003eCa\u003c/sub\u003e2.2_I structures (Figs. 4b,c). These results suggest that the induced-fit binding pocket of rimtuzalcap does not exist in apo_K\u003csub\u003eCa\u003c/sub\u003e2.2. Binding of rimtuzalcap expands CaM’s N-lobes around rimtuzalcap and shortens the distances between opposite CaM’s N-lobes, resulting in rigidification of K\u003csub\u003eCa\u003c/sub\u003e2.2’s cytoplasmic HC helices (Fig. 4d). These activator-induced changes in rimtuzalcap_K\u003csub\u003eCa\u003c/sub\u003e2.2_I are reminiscent of the mechanism of action of AUT5, an activator of the voltage-gated K\u003csub\u003eV\u003c/sub\u003e3.1 channel. Binding of AUT5 to K\u003csub\u003eV\u003c/sub\u003e3.1 rearranges the turret region around the activator molecule, thereby inducing interactions between the turret and the transmembrane S4 segment in the voltage-sensor domain. \u003csup\u003e4\u003c/sup\u003e\u003c/p\u003e\n\u003cp\u003ePreviously, we reported that CyPPA, an analog of rimtuzalcap, exhibited subtype-selectivity for K\u003csub\u003eCa\u003c/sub\u003e2.2 over K\u003csub\u003eCa\u003c/sub\u003e3.1 due to the difference of a single residue in the HB helix (K\u003csub\u003eCa\u003c/sub\u003e2.2_K467 versus K\u003csub\u003eCa\u003c/sub\u003e3.1_R355). \u003csup\u003e23\u003c/sup\u003e However, our earlier docking studies were based on a homology model of K\u003csub\u003eCa\u003c/sub\u003e2.2 generated with K\u003csub\u003eCa\u003c/sub\u003e3.1 cryo-EM structures as templates. \u003csup\u003e12\u003c/sup\u003e Docking into this homology model placed CyPPA in a pre-existing binding pocket between CaM’s C-lobe and K\u003csub\u003eCa\u003c/sub\u003e2.2’s HA/HB helices \u003csup\u003e23\u003c/sup\u003e. This placement is most likely wrong because CyPPA and rimtuzalcap are structurally very similar (Fig. 5a), and we have shown above that rimtuzalcap fits into an induced-fit binding pocket that emerges as a result of conformational changes in CaM (Fig. 4).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eThe K\u003csub\u003eCa\u003c/sub\u003e3.1 channel is insensitive to rimtuzalcap. Our structural, mutagenesis and electrophysiological studies suggest that this insensitivity to rimtuzalcap is because the putative activator binding pocket in CaM’s N-lobe of K\u003csub\u003eCa\u003c/sub\u003e3.1 is constrained by the HC helices and is prevented from undergoing the requisite conformational changes to accommodate rimtuzalcap. Replacing K\u003csub\u003eCa\u003c/sub\u003e3.1_R355 with lysine (corresponding to K\u003csub\u003eCa\u003c/sub\u003e2.2_K467) in K\u003csub\u003eCa\u003c/sub\u003e3.1’s HB helix frees the binding pocket from the constraints of the HC helices, and allows rimtuzalcap to activate the mutant K\u003csub\u003eCa\u003c/sub\u003e3.1_R355K channel (EC\u003csub\u003e50\u003c/sub\u003e: ~76.8\u0026nbsp;mM, Fig. 6).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eIn conclusion, our structures provide a foundation for understanding the subtype-selectivity of the K\u003csub\u003eCa\u003c/sub\u003e2.2 activator\u0026nbsp;rimtuzalcap\u0026nbsp;and could enable structure-based drug design of more potent, subtype-selective activators targeting K\u003csub\u003eCa\u003c/sub\u003e2.2 channels. Subtype-selective K\u003csub\u003eCa\u003c/sub\u003e2.2 activators that avoid potential side effects associated with activation of peripheral K\u003csub\u003eCa\u003c/sub\u003e3.1 channels are critically needed to target K\u003csub\u003eCa\u003c/sub\u003e2.2 channels in the central nervous system for the treatment of spinocerebellar ataxia and essential tremor.\u003c/p\u003e"},{"header":"Methods","content":"\u003cp\u003e\u003cstrong\u003eProtein expression\u0026nbsp;\u003c/strong\u003e\u003cstrong\u003eand purification\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe rat K\u003csub\u003eCa\u003c/sub\u003e2.2/CaM protein complex was expressed and purified as described in our previous report. \u003csup\u003e11\u003c/sup\u003e Briefly, the cDNA of full-length rat K\u003csub\u003eCa\u003c/sub\u003e2.2 (accession no. NM_019314) with a C-terminus Strep-II tag was sub-cloned into pEG BacMam (a gift from Eric Gouaux; Addgene plasmid # 160451; http://n2t.net/addgene:160451; RRID:Addgene_160451). Un-tagged \u003cem\u003erat\u003c/em\u003e calmodulin (CaM) cDNA (accession no. BC063187) was also cloned into pEG BacMam. The amino acid sequence of the \u003cem\u003erat\u003c/em\u003e CaM is 100% identical to the \u003cem\u003ehuman\u003c/em\u003e CaM.\u0026nbsp;We expressed the\u0026nbsp;K\u003csub\u003eCa\u003c/sub\u003e2.2/CaM protein complex\u0026nbsp;in HEK293S GnTI− cells (ATCC) using a BacMam method\u0026nbsp;\u003csup\u003e24\u003c/sup\u003e. The K\u003csub\u003eCa\u003c/sub\u003e2.2/CaM complex was purified using Strep-Tactin XT resin, followed by size exclusion chromatography column equilibrated by 20 mM Tris pH 8, 150 mM KCl, 2 mM CaCl\u003csub\u003e2\u003c/sub\u003e, and 0.01% lauryl maltose neopentyl glycol (LMNG, Anatrace).\u0026nbsp;The peak fractions were collected and concentrated to ~3 mg/ml.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eThe human K\u003csub\u003eCa\u003c/sub\u003e3.1/CaM protein complex was expressed and purified as described in previous reports.\u0026nbsp;\u003csup\u003e12,25\u003c/sup\u003e Briefly, the cDNA of full-length human K\u003csub\u003eCa\u003c/sub\u003e3.1 (accession no. NM_002250.3) with a C-terminus Strep-II tag was sub-cloned into pEG BacMam.\u0026nbsp;We expressed the K\u003csub\u003eCa\u003c/sub\u003e3.1/CaM protein complex in HEK293S GnTI− cells (ATCC) using a BacMam method \u003csup\u003e24\u003c/sup\u003e. The K\u003csub\u003eCa\u003c/sub\u003e3.1/CaM complex was purified using Strep-Tactin XT resin, followed by size exclusion chromatography column equilibrated by 20 mM Tris pH 8, 150 mM KCl, 2 mM CaCl\u003csub\u003e2\u003c/sub\u003e, and 0.007%\u0026nbsp;glyco-diosgenin (GDN, Anatrace). The peak fractions were collected and concentrated to ~3 mg/ml.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCryo-EM Sample Preparation, Data Collection and Processing\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWe performed cryo-EM data collection at the Stanford SLAC Cryo-EM Center (S\u003csup\u003e2\u003c/sup\u003eC\u003csup\u003e2\u003c/sup\u003e) and Pacific Northwest Cryo-EM Center (PNCC). To determine the NS309-bound structures, saturating concentrations of NS309 were mixed with the K\u003csub\u003eCa\u003c/sub\u003e3.1/CaM or K\u003csub\u003eCa\u003c/sub\u003e2.2/CaM protein complexes 30 minutes before the grid preparation, respectively. To determine the rimtuzalcap-bound structure, saturating concentrations of rimtuzalcap were mixed with the K\u003csub\u003eCa\u003c/sub\u003e2.2/CaM protein complex 30 minutes before the grid preparation. Because K\u003csub\u003eCa\u003c/sub\u003e3.1 channels are not sensitive to rimtuzalcap, the K\u003csub\u003eCa\u003c/sub\u003e3.1/CaM protein complex was not mixed with rimtuzalcap.\u003c/p\u003e\n\u003cp\u003e3 μl of purified protein was applied to a glow-discharged Quantifoil R1.2/1.3 300 mesh Copper grid, at 4 °C and 100% humidity using Vitrobot Mark IV (ThermoFisher Scientific). The grid was then blotted for 3 seconds before being plunged into liquid ethane. Grids of NS309-bound K\u003csub\u003eCa\u003c/sub\u003e3.1/CaM were screened, and cryo-EM data were subsequently collected on a Titan Krios G3i (ThermoFisher Scientific) with a K3 detector (Gatan) and a BioQuantum energy filter. Dose fractionated movies were collected using EPU at the pixel size of 0.86 Å. A total cumulative dose of\u0026nbsp;∼50 electrons per Å\u003csup\u003e2\u003c/sup\u003e was used for recording the movies of 40 frames (1.25 electrons per Å\u003csup\u003e2\u003c/sup\u003e per frame) with a defocus range of –1.0 to –2.0\u0026nbsp;μm.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eGrids of rimtuzalcap-bound K\u003csub\u003eCa\u003c/sub\u003e2.2/CaM were screened, and cryo-EM data were subsequently collected on a Titan Krios G3i (ThermoFisher Scientific) with a K3 detector (Gatan) and a BioQuantum energy filter. Dose fractionated movies were collected using EPU at the pixel size of 0.86 Å. A total cumulative dose of\u0026nbsp;∼50 electrons per Å\u003csup\u003e2\u003c/sup\u003e was used for recording the movies of 40 frames (1.25 electrons per Å\u003csup\u003e2\u003c/sup\u003e per frame) with a defocus range of –1.3 to –2.3\u0026nbsp;μm.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eGrids of NS309-bound K\u003csub\u003eCa\u003c/sub\u003e2.2/CaM were screened, and cryo-EM data were subsequently collected on a Titan Krios G3i (ThermoFisher Scientific) with a Falcon4i detector (ThermoFisher Scientific) and a SelectrisX energy filter. Dose fractionated movies were collected using EPU at the pixel size of 0.73 Å. A total cumulative dose of\u0026nbsp;∼50 electrons per Å\u003csup\u003e2\u003c/sup\u003e was used for recording in the EER format with a defocus range of –0.6 to –2.2\u0026nbsp;μm. The statistics for data collections are summarized in Supplementary Tables 1.\u003c/p\u003e\n\u003cp\u003eCryo-EM image processing for NS309-bound K\u003csub\u003eCa\u003c/sub\u003e2.2/CaM was carried out using CryoSparc version 4.5.3 \u003csup\u003e26\u003c/sup\u003e. After preprocessing of micrographs (motion correction and CTF estimation), ~2,000 particles were manually picked, followed by 2D classification to generate picking templates. Template-picked particles were cleaned up by multiple rounds of 2D classification. An \u003cem\u003eab initio\u003c/em\u003e reconstruction was performed, followed by heterogenous and non-uniform refinements to 2.72 Å resolution in the CryoSparc \u003csup\u003e26\u003c/sup\u003e program. The cryo-EM map with 2.72 Å resolution was improved by 3D classification followed by nonuniform refinement to 2.71 Å resolution (NS309_K\u003csub\u003eCa\u003c/sub\u003e2.2,\u0026nbsp;Supplementary Fig. 1).\u0026nbsp;The cryo-EM density for NS309 is clearly visible in the refined map of NS309_K\u003csub\u003eCa\u003c/sub\u003e2.2 (Fig. 2a and Supplementary Fig. 5).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eThe NS309-bound K\u003csub\u003eCa\u003c/sub\u003e3.1/CaM dataset was similarly processed using the CryoSparc program\u003csup\u003e26\u003c/sup\u003e. An \u003cem\u003eab initio\u003c/em\u003e reconstruction was performed, followed by heterogenous and non-uniform refinements to 3.69 Å resolution in the CryoSparc \u003csup\u003e26\u003c/sup\u003e program. The cryo-EM map with 3.69 Å resolution was improved by 3D classification followed by nonuniform refinement to 3.59 Å resolution (NS309_K\u003csub\u003eCa\u003c/sub\u003e3.1,\u0026nbsp;Supplementary Fig. 3).\u0026nbsp;The cryo-EM density for NS309 is clearly visible in the refined map of NS309_K\u003csub\u003eCa\u003c/sub\u003e3.1 (Fig. 2b and Supplementary Fig. 6).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eFor the rimtuzalcap-bound K\u003csub\u003eCa\u003c/sub\u003e2.2/CaM dataset, an \u003cem\u003eab initio\u003c/em\u003e reconstruction was performed, followed by heterogenous and non-uniform refinements to 2.95 Å resolution in the CryoSparc \u003csup\u003e26\u003c/sup\u003e program. To examine possible conformational heterogeneity, 3D classification was performed. Two classes were generated by 3D classification, which were further refined using non-uniform refinements to 3.13 Å resolution (rimtuzalcap_K\u003csub\u003eCa\u003c/sub\u003e2.2_I), and 2.96 Å resolution (rimtuzalcap_K\u003csub\u003eCa\u003c/sub\u003e2.2_II), which were used for model building\u0026nbsp;(Supplementary Fig. 8).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eThe cryo-EM density of rimtuzalcap is clearly visible at the interface between the CaM N-lobe and the S\u003csub\u003e45\u003c/sub\u003eA/HA helices of K\u003csub\u003eCa\u003c/sub\u003e2.2 in both rimtuzalcap_K\u003csub\u003eCa\u003c/sub\u003e2.2_I and rimtuzalcap_K\u003csub\u003eCa\u003c/sub\u003e2.2_II (Supplementary Fig. 10). The cryo-EM densities of two Ca\u003csup\u003e2+\u003c/sup\u003e ions are visible at the CaM N-lobes of rimtuzalcap_K\u003csub\u003eCa\u003c/sub\u003e2.2_I but not in rimtuzalcap_K\u003csub\u003eCa\u003c/sub\u003e2.2_II. In a portion of\u0026nbsp;rimtuzalcap-bound K\u003csub\u003eCa\u003c/sub\u003e2.2\u0026nbsp;channels, the CaM N-lobe may lose Ca\u003csup\u003e2+\u0026nbsp;\u003c/sup\u003ebinding (rimtuzalcap_K\u003csub\u003eCa\u003c/sub\u003e2.2_II), which in turn closes the inner gate because Ca\u003csup\u003e2+\u003c/sup\u003e binding to the CaM N-lobe is required for the activation of K\u003csub\u003eCa\u003c/sub\u003e2.2 \u003csup\u003e22\u003c/sup\u003e. The other portion of\u0026nbsp;rimtuzalcap-bound K\u003csub\u003eCa\u003c/sub\u003e2.2\u0026nbsp;channels may retain both Ca\u003csup\u003e2+\u003c/sup\u003e and rimtuzalcap (rimtuzalcap_K\u003csub\u003eCa\u003c/sub\u003e2.2_I), which exhibits a wider inner gate (~13.5 Å) than the\u0026nbsp;apo_K\u003csub\u003eCa\u003c/sub\u003e2.2 structure (~12.6 Å). Since Ca\u003csup\u003e2+\u003c/sup\u003e binding to the CaM N-lobe is required for the activation of K\u003csub\u003eCa\u003c/sub\u003e2.2\u0026nbsp;\u003csup\u003e22\u003c/sup\u003e, rimtuzalcap_K\u003csub\u003eCa\u003c/sub\u003e2.2_I that contains both rimtuzalcap and Ca\u003csup\u003e2+\u003c/sup\u003e was used for further analysis of the interactions between rimtuzalcap and K\u003csub\u003eCa\u003c/sub\u003e2.2 channels.\u003c/p\u003e\n\u003cp\u003eResolutions were estimated using the gold standard criterion at the threshold of 0.143. The local resolution was calculated in CryoSparc.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eModel building\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eCoordinates of the Ca\u003csup\u003e2+\u003c/sup\u003e-bound K\u003csub\u003eCa\u003c/sub\u003e2.2 (PDB: 8v2g; apo_K\u003csub\u003eCa\u003c/sub\u003e2.2) were used as the initial model for activator-bound K\u003csub\u003eCa\u003c/sub\u003e2.2 structures. Coordinates of the Ca\u003csup\u003e2+\u003c/sup\u003e-bound K\u003csub\u003eCa\u003c/sub\u003e3.1 activated state II (PDB: 6cno; apo_K\u003csub\u003eCa\u003c/sub\u003e3.1_II) were used as the initial model for activator-bound K\u003csub\u003eCa\u003c/sub\u003e3.1 structures. The initial model was manually docked into the cryo-EM density map and then adjusted in UCSF Chimera. Phenix.real_space_refine \u003csup\u003e27\u003c/sup\u003e was used to build and refine the model. Model building was achieved using phenix \u003csup\u003e27\u003c/sup\u003e and Coot \u003csup\u003e28\u003c/sup\u003e iteratively. Models for Ca\u003csup\u003e2+\u003c/sup\u003e, K\u003csup\u003e+\u003c/sup\u003e, and activators were built by visual inspection of the shape of the density in Coot \u003csup\u003e28\u003c/sup\u003e followed by refinement in phenix \u003csup\u003e27\u003c/sup\u003e. The statistics for model refinements are summarized in Supplementary Table 1. All structural graphics were generated using UCSF ChimeraX \u003csup\u003e29\u003c/sup\u003e.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ePatch-clamp electrophysiology\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eHuman Embryonic Kidney (HEK293) cells transiently transfected with the rat K\u003csub\u003eCa\u003c/sub\u003e2.2 or human K\u003csub\u003eCa\u003c/sub\u003e3.1 cDNAs were used for manual patch-clamp experiments. Site-directed mutagenesis of the K\u003csub\u003eCa\u003c/sub\u003e3.1_R355K was performed on the cDNAs through molecular cloning services (Genscript). The wildtype and mutant cDNAs,\u0026nbsp;constructed in the pIRES2-AcGFP1 vector (Clontech),were transfected into HEK293 cells by a calcium–phosphate method. Inside-out K\u003csub\u003eCa\u003c/sub\u003e2.2 currents were recorded 1–2 days after transfec­tion, with an Axon200B amplifier (Molecular Devices) at room temperature.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eThe resistance of the patch electrodes ranged from 2-3 MΩ. For inside-out recordings, the intracellular solution containing (in mM): 140 KCl, 10 Hepes (pH 7.2), 1 EGTA, 0.1 Dibromo-BAPTA, and 1 HEDTA was mixed with Ca\u003csup\u003e2+\u003c/sup\u003e to obtain the desired free Ca\u003csup\u003e2+\u003c/sup\u003e concentrations, calculated using the MaxChelator software. The extracellular solution contained (in mM): 140 KCl, 10 Hepes (pH 7.4), 1 MgSO\u003csub\u003e4\u003c/sub\u003e.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eNS309 (6,7-dichloro-1\u003cem\u003eH\u003c/em\u003e-indole-2,3-dione 3-oxime) was purchased from Alomone labs. NS309 dilutions were prepared freshly in extracellular solution from 20 mM stock solutions in DMSO.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eRimtuzalcap (also called CAD-1883, \u003cem\u003eN\u003c/em\u003e-(4,4-difluorocyclohexyl)-2-(3-methylpyrazol-1-yl)-6-morpholin-4-ylpyrimidin-4-amine) was purchased from MedChemExpress. Rimtuzalcap dilutions were prepared freshly in extracellular solution from 100 mM stock solutions in DMSO.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003epClamp 10.5 (Molecular Devices) was used for data acquisi­tion and analysis.\u0026nbsp;To characterize the responses of\u0026nbsp;the K\u003csub\u003eCa\u003c/sub\u003e3.1_R355K mutants to rimtuzalcap,\u0026nbsp;inside-out patch recordings were performed.\u0026nbsp;Seals (\u0026gt; 1 GΩ) were formed before the inside-out patch configuration was obtained.\u0026nbsp;The intracel­lular face was exposed to a series of activator concentrations at a fixed Ca\u003csup\u003e2+\u003c/sup\u003e concentration (0.15\u0026nbsp;mM Ca\u003csup\u003e2+\u003c/sup\u003e). Currents were recorded by repetitive 1s voltage ramps from − 100 mV to 100 mV from a holding potential of 0 mV. One minute after switching of bath solutions, ten sweeps with a 1 s interval were recorded. To construct the concentration-dependent activation of channel activities, the current amplitudes at − 90 mV in response to various concentrations of activatorswere normalized to that obtained at 10\u0026nbsp;mM\u0026nbsp;of Ca\u003csup\u003e2+\u003c/sup\u003e. The normalized currents were plotted as a function of the concentrations of the activators. EC\u003csub\u003e50\u003c/sub\u003e values and Hill coefficients were determined by fitting the data points to a standard concentration–response curve.\u003c/p\u003e\n\u003cp\u003eData analysis was performed using pClamp 10.5 (Molecular Devices) in a blinded fashion. Concentration-response curves were analyzed in GraphPad Prism 10 (GraphPad Software Inc.). All data are shown as mean ± SD unless otherwise indicated. One-way ANOVA and Tukey’s post hoc tests were used for data comparison of three or more groups. The unpaired two-tailed Student’s \u003cem\u003et\u003c/em\u003e-test was used for data comparison if there were only two groups.\u0026nbsp;Figures were made using GraphPad Prism 10 (GraphPad Software Inc.).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCalculation of binding energy\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe binding energy between NS309 and rimtuzalcap in their binding pockets were calculated using the cryo-EM structures and the Discovery Studio program (Dassault Systemes Biovia LLC). Briefly, the channel complex structures were subjected to energy minimization using Smart Minimizer algorithm (200 steps) and Generalized Born (GB) Implicit Solvent and Membrane model using the CHARMm forcefield. Interaction energies between the drugs and channels were calculated using an Implicit Distance-Dependent Dielectrics (Dielectric Constant=2.0) solvent model. Residues within 6.5 Å from the drugs were selected for interaction energy and energy decomposition calculations.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eData availability\u003c/strong\u003e\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eThe atomic coordinates have been deposited in the Protein Data Bank (PDB) under accession codes 9O7S [https://doi.org/10.2210/pdb9O7S/pdb] (NS309_K\u003csub\u003eCa\u003c/sub\u003e2.2); 9OA8 [https://doi.org/10.2210/pdb9OA8/pdb] (NS309_K\u003csub\u003eCa\u003c/sub\u003e3.1); 9O85 [https://doi.org/10.2210/pdb9O85/pdb] (rimtuzalcap_K\u003csub\u003eCa\u003c/sub\u003e2.2_I); and 9O93 [https://doi.org/10.2210/pdb9O93/pdb] (rimtuzalcap_K\u003csub\u003eCa\u003c/sub\u003e2.2_II). The cryo-EM maps have been deposited in the Electron Microscopy Data Bank (EMDB) under accession codes EMD-70207 [https://www.ebi.ac.uk/pdbe/entry/emdb/EMD-70207] (NS309_K\u003csub\u003eCa\u003c/sub\u003e2.2); EMDB-70275 [https://www.ebi.ac.uk/pdbe/entry/emdb/EMDB-70275] (NS309_K\u003csub\u003eCa\u003c/sub\u003e3.1); EMD-70217 [https://www.ebi.ac.uk/pdbe/entry/emdb/EMD-70217] (rimtuzalcap_K\u003csub\u003eCa\u003c/sub\u003e2.2_I); and EMD-70240 [https://www.ebi.ac.uk/pdbe/entry/emdb/EMD-70240] (rimtuzalcap_K\u003csub\u003eCa\u003c/sub\u003e2.2_II). The following previously published PDB codes are referred to: 8v2g [https://doi.org/10.2210/pdb8v2g/pdb]; 8v2h [https://doi.org/10.2210/pdb8v2h/pdb]; 6cnn [https://doi.org/10.2210/pdb6cnn/pdb]; 6cno [https://doi.org/10.2210/pdb6cno/pdb]. All other data supporting the findings of this study are available within the paper and its Supplementary information files.\u0026nbsp;\u003c/p\u003e\u003cp\u003e\u003cstrong\u003eAcknowledgments\u003c/strong\u003e\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eM.Z. was supported by National Institutes of Health (Grant 4R33NS101182-03 and Grant 1R15NS130420), American Heart Association (Grant 23AIREA1039423), Chapman University Institutional FGRSC Grant, and Office of the Assistant Secretary of Defense for Health Affairs through the Alcohol and Substance Use Research Program under Award No.: W81XWH-22-2-0081. Y.W.N was supported by American Heart Association (24CDA1260237). Some of this work was performed at the Stanford-SLAC Cryo-EM Center (S\u003csup\u003e2\u003c/sup\u003eC\u003csup\u003e2\u003c/sup\u003e), which is supported by the National Institutes of Health Common Fund Transformative High-Resolution Cryo-Electron Microscopy program (U24 GM129541). The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health. \u0026nbsp; The authors would also like to thank the following S\u003csup\u003e2\u003c/sup\u003eC\u003csup\u003e2\u003c/sup\u003e personnel for their invaluable support and assistance: Patrick Mitchell, Alexandre Cassago, Nathan D. Burrows, Yan Liu, Grace Nye, Chensong Zhang, and Patrick J. Pascual. A portion of this research was supported by NIH grant R24GM154185 and performed at the Pacific Northwest Center for Cryo-EM (PNCC) with assistance from Vamsee Rayaprolu.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCompeting interests\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe authors declare no competing interests.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthor contributions\u003c/strong\u003e\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eM.Z. conceptualized the project. Y.W.N., A.R., and M.Z. undertook protein expression and purification. Y.W.N., A.R., Y.X., R.M.H.Y., D.I., and M.Z. undertook cryo-EM studies. A.R. undertook electrophysiology studies. M.C. performed computational studies. M.Z., K.G.C, and H.W. wrote the manuscript. All authors contributed to the manuscript and the figures.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eLi, T.\u003cem\u003e et al.\u003c/em\u003e Structural Basis for the Modulation of Human KCNQ4 by Small-Molecule Drugs. \u003cem\u003eMol Cell\u003c/em\u003e \u003cstrong\u003e81\u003c/strong\u003e, 25-37 e24 (2021). https://doi.org:10.1016/j.molcel.2020.10.037\u003c/li\u003e\n\u003cli\u003eWillegems, K.\u003cem\u003e et al.\u003c/em\u003e Structural and electrophysiological basis for the modulation of KCNQ1 channel currents by ML277. \u003cem\u003eNat Commun\u003c/em\u003e \u003cstrong\u003e13\u003c/strong\u003e, 3760 (2022). https://doi.org:10.1038/s41467-022-31526-7\u003c/li\u003e\n\u003cli\u003eZhang, S.\u003cem\u003e et al.\u003c/em\u003e A small-molecule activation mechanism that directly opens the KCNQ2 channel. \u003cem\u003eNat Chem Biol\u003c/em\u003e \u003cstrong\u003e20\u003c/strong\u003e, 847-856 (2024). https://doi.org:10.1038/s41589-023-01515-y\u003c/li\u003e\n\u003cli\u003eLiang, Q.\u003cem\u003e et al.\u003c/em\u003e The binding and mechanism of a positive allosteric modulator of Kv3 channels. \u003cem\u003eNat Commun\u003c/em\u003e \u003cstrong\u003e15\u003c/strong\u003e, 2533 (2024). https://doi.org:10.1038/s41467-024-46813-8\u003c/li\u003e\n\u003cli\u003eChen, Y. T.\u003cem\u003e et al.\u003c/em\u003e Identification, structural, and biophysical characterization of a positive modulator of human Kv3.1 channels. \u003cem\u003eProc Natl Acad Sci U S A\u003c/em\u003e \u003cstrong\u003e120\u003c/strong\u003e, e2220029120 (2023). https://doi.org:10.1073/pnas.2220029120\u003c/li\u003e\n\u003cli\u003eBotte, M.\u003cem\u003e et al.\u003c/em\u003e Apo and ligand-bound high resolution Cryo-EM structures of the human Kv3.1 channel reveal a novel binding site for positive modulators. \u003cem\u003ePNAS Nexus\u003c/em\u003e \u003cstrong\u003e1\u003c/strong\u003e, pgac083 (2022). https://doi.org:10.1093/pnasnexus/pgac083\u003c/li\u003e\n\u003cli\u003eCui, M.\u003cem\u003e et al.\u003c/em\u003e A novel small-molecule selective activator of homomeric GIRK4 channels. \u003cem\u003eThe Journal of biological chemistry\u003c/em\u003e \u003cstrong\u003e298\u003c/strong\u003e, 102009 (2022). https://doi.org:10.1016/j.jbc.2022.102009\u003c/li\u003e\n\u003cli\u003eLolicato, M.\u003cem\u003e et al.\u003c/em\u003e K(2P) channel C-type gating involves asymmetric selectivity filter order-disorder transitions. \u003cem\u003eSci Adv\u003c/em\u003e \u003cstrong\u003e6\u003c/strong\u003e (2020). https://doi.org:10.1126/sciadv.abc9174\u003c/li\u003e\n\u003cli\u003eZhang, G.\u003cem\u003e et al.\u003c/em\u003e An allosteric modulator activates BK channels by perturbing coupling between Ca(2+) binding and pore opening. \u003cem\u003eNat Commun\u003c/em\u003e \u003cstrong\u003e13\u003c/strong\u003e, 6784 (2022). https://doi.org:10.1038/s41467-022-34359-6\u003c/li\u003e\n\u003cli\u003eBrown, B. M., Shim, H., Christophersen, P. \u0026amp; Wulff, H. Pharmacology of Small- and Intermediate-Conductance Calcium-Activated Potassium Channels. \u003cem\u003eAnnu Rev Pharmacol Toxicol\u003c/em\u003e \u003cstrong\u003e60\u003c/strong\u003e, 219-240 (2020). https://doi.org:10.1146/annurev-pharmtox-010919-023420\u003c/li\u003e\n\u003cli\u003eNam, Y. W.\u003cem\u003e et al.\u003c/em\u003e Cryo-EM structures of the small-conductance Ca(2+)-activated K(Ca)2.2 channel. \u003cem\u003eNat Commun\u003c/em\u003e \u003cstrong\u003e16\u003c/strong\u003e, 3690 (2025). https://doi.org:10.1038/s41467-025-59061-1\u003c/li\u003e\n\u003cli\u003eLee, C. H. \u0026amp; MacKinnon, R. Activation mechanism of a human SK-calmodulin channel complex elucidated by cryo-EM structures. \u003cem\u003eScience\u003c/em\u003e \u003cstrong\u003e360\u003c/strong\u003e, 508-513 (2018). https://doi.org:10.1126/science.aas9466\u003c/li\u003e\n\u003cli\u003eCho, L. T.\u003cem\u003e et al.\u003c/em\u003e An Intracellular Allosteric Modulator Binding Pocket in SK2 Ion Channels Is Shared by Multiple Chemotypes. \u003cem\u003eStructure\u003c/em\u003e \u003cstrong\u003e26\u003c/strong\u003e, 533-544 e533 (2018). https://doi.org:10.1016/j.str.2018.02.017\u003c/li\u003e\n\u003cli\u003eZhang, M., Pascal, J. M. \u0026amp; Zhang, J. F. Unstructured to structured transition of an intrinsically disordered protein peptide in coupling Ca(2)(+)-sensing and SK channel activation. \u003cem\u003eProc Natl Acad Sci U S A\u003c/em\u003e \u003cstrong\u003e110\u003c/strong\u003e, 4828-4833 (2013). https://doi.org:10.1073/pnas.1220253110\u003c/li\u003e\n\u003cli\u003eZhang, M., Pascal, J. M., Schumann, M., Armen, R. S. \u0026amp; Zhang, J. F. Identification of the functional binding pocket for compounds targeting small-conductance Ca(2)(+)-activated potassium channels. \u003cem\u003eNat Commun\u003c/em\u003e \u003cstrong\u003e3\u003c/strong\u003e, 1021 (2012). https://doi.org:10.1038/ncomms2017\u003c/li\u003e\n\u003cli\u003eStrobaek, D.\u003cem\u003e et al.\u003c/em\u003e Activation of human IK and SK Ca2+ -activated K+ channels by NS309 (6,7-dichloro-1H-indole-2,3-dione 3-oxime). \u003cem\u003eBiochimica et biophysica acta\u003c/em\u003e \u003cstrong\u003e1665\u003c/strong\u003e, 1-5 (2004). https://doi.org:10.1016/j.bbamem.2004.07.006\u003c/li\u003e\n\u003cli\u003eNasburg, J. A.\u003cem\u003e et al.\u003c/em\u003e 6,7-Dichloro-1H-indole-2,3-dione-3-oxime functions as a superagonist for the intermediate-conductance Ca(2+)-activated K(+) channel K(Ca)3.1. \u003cem\u003eMolecular pharmacology\u003c/em\u003e \u003cstrong\u003e107\u003c/strong\u003e, 100018 (2025). https://doi.org:10.1016/j.molpha.2025.100018\u003c/li\u003e\n\u003cli\u003eAtaman, Z. A., Gakhar, L., Sorensen, B. R., Hell, J. W. \u0026amp; Shea, M. A. The NMDA Receptor NR1 C1 Region Bound to Calmodulin: Structural Insights into Functional Differences between Homologous Domains. \u003cem\u003eStructure\u003c/em\u003e \u003cstrong\u003e15\u003c/strong\u003e, 1603-1617 (2007). \u003c/li\u003e\n\u003cli\u003eHougaard, C.\u003cem\u003e et al.\u003c/em\u003e Selective positive modulation of the SK3 and SK2 subtypes of small conductance Ca2+-activated K+ channels. \u003cem\u003eBritish journal of pharmacology\u003c/em\u003e \u003cstrong\u003e151\u003c/strong\u003e, 655-665 (2007). https://doi.org:10.1038/sj.bjp.0707281\u003c/li\u003e\n\u003cli\u003eWeikl, T. R. \u0026amp; von Deuster, C. Selected-fit versus induced-fit protein binding: kinetic differences and mutational analysis. \u003cem\u003eProteins\u003c/em\u003e \u003cstrong\u003e75\u003c/strong\u003e, 104-110 (2009). https://doi.org:10.1002/prot.22223\u003c/li\u003e\n\u003cli\u003eShim, H.\u003cem\u003e et al.\u003c/em\u003e The Trials and Tribulations of Structure Assisted Design of K(Ca) Channel Activators. \u003cem\u003eFront Pharmacol\u003c/em\u003e \u003cstrong\u003e10\u003c/strong\u003e, 972 (2019). https://doi.org:10.3389/fphar.2019.00972\u003c/li\u003e\n\u003cli\u003eLi, W., Halling, D. B., Hall, A. W. \u0026amp; Aldrich, R. W. EF hands at the N-lobe of calmodulin are required for both SK channel gating and stable SK-calmodulin interaction. \u003cem\u003eJ Gen Physiol\u003c/em\u003e \u003cstrong\u003e134\u003c/strong\u003e, 281-293 (2009). https://doi.org:10.1085/jgp.200910295\u003c/li\u003e\n\u003cli\u003eNam, Y. W.\u003cem\u003e et al.\u003c/em\u003e Subtype-selective positive modulation of K(Ca) 2 channels depends on the HA/HB helices. \u003cem\u003eBritish journal of pharmacology\u003c/em\u003e \u003cstrong\u003e179\u003c/strong\u003e, 460-472 (2022). https://doi.org:10.1111/bph.15676\u003c/li\u003e\n\u003cli\u003eGoehring, A.\u003cem\u003e et al.\u003c/em\u003e Screening and large-scale expression of membrane proteins in mammalian cells for structural studies. \u003cem\u003eNat Protoc\u003c/em\u003e \u003cstrong\u003e9\u003c/strong\u003e, 2574-2585 (2014). https://doi.org:10.1038/nprot.2014.173\u003c/li\u003e\n\u003cli\u003eOng, S. T.\u003cem\u003e et al.\u003c/em\u003e Design and structural basis of selective 1,4-dihydropyridine inhibitors of the calcium-activated potassium channel K(Ca)3.1. \u003cem\u003eProc Natl Acad Sci U S A\u003c/em\u003e \u003cstrong\u003e122\u003c/strong\u003e, e2425494122 (2025). https://doi.org:10.1073/pnas.2425494122\u003c/li\u003e\n\u003cli\u003ePunjani, A., Rubinstein, J. L., Fleet, D. J. \u0026amp; Brubaker, M. A. cryoSPARC: algorithms for rapid unsupervised cryo-EM structure determination. \u003cem\u003eNature methods\u003c/em\u003e \u003cstrong\u003e14\u003c/strong\u003e, 290-296 (2017). https://doi.org:10.1038/nmeth.4169\u003c/li\u003e\n\u003cli\u003eAfonine, P. V.\u003cem\u003e et al.\u003c/em\u003e Real-space refinement in PHENIX for cryo-EM and crystallography. \u003cem\u003eActa Crystallogr D Struct Biol\u003c/em\u003e \u003cstrong\u003e74\u003c/strong\u003e, 531-544 (2018). https://doi.org:10.1107/S2059798318006551\u003c/li\u003e\n\u003cli\u003eEmsley, P., Lohkamp, B., Scott, W. G. \u0026amp; Cowtan, K. Features and development of Coot. \u003cem\u003eActa Crystallogr D Biol Crystallogr\u003c/em\u003e \u003cstrong\u003e66\u003c/strong\u003e, 486-501 (2010). https://doi.org:10.1107/S0907444910007493\u003c/li\u003e\n\u003cli\u003eGoddard, T. D.\u003cem\u003e et al.\u003c/em\u003e UCSF ChimeraX: Meeting modern challenges in visualization and analysis. \u003cem\u003eProtein Sci\u003c/em\u003e\u003cstrong\u003e27\u003c/strong\u003e, 14-25 (2018). https://doi.org:10.1002/pro.3235\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":true,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"nature-portfolio","isNatureJournal":true,"hasQc":false,"allowDirectSubmit":false,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"","title":"Nature Portfolio","twitterHandle":"","acdcEnabled":false,"dfaEnabled":false,"editorialSystem":"ejp","reportingPortfolio":"","inReviewEnabled":true,"inReviewRevisionsEnabled":false},"keywords":"","lastPublishedDoi":"10.21203/rs.3.rs-6568445/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-6568445/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eSmall-conductance (K\u003csub\u003eCa\u003c/sub\u003e2.2) and intermediate-conductance (K\u003csub\u003eCa\u003c/sub\u003e3.1) Ca\u003csup\u003e2+\u003c/sup\u003e-activated K\u003csup\u003e+\u003c/sup\u003e channels are gated by a Ca\u003csup\u003e2+\u003c/sup\u003e-calmodulin dependent mechanism. NS309 potentiates the activity of both K\u003csub\u003eCa\u003c/sub\u003e2.2 and K\u003csub\u003eCa\u003c/sub\u003e3.1, while rimtuzalcap selectively activates K\u003csub\u003eCa\u003c/sub\u003e2.2. Rimtuzalcap has been used in clinical trials for the treatment of spinocerebellar ataxia and essential tremor. We report cryo-electron microscopy structures of K\u003csub\u003eCa\u003c/sub\u003e2.2 channels bound with NS309 and rimtuzalcap, in addition to K\u003csub\u003eCa\u003c/sub\u003e3.1 channels with NS309. The different conformations of calmodulin and the cytoplasmic HC helices in the two channels underlie the subtype-selectivity of rimtuzalcap for K\u003csub\u003eCa\u003c/sub\u003e2.2. Calmodulin\u0026rsquo;s N-lobes in the K\u003csub\u003eCa\u003c/sub\u003e2.2 structure are far apart and undergo conformational changes to accommodate either NS309 or rimtuzalcap. Calmodulin\u0026rsquo;s N-lobes in the K\u003csub\u003eCa\u003c/sub\u003e3.1 structure are closer to each other and are constrained by the HC helices of K\u003csub\u003eCa\u003c/sub\u003e3.1, which allows binding of NS309 but not of the bulkier rimtuzalcap. These structures provide a framework for structure-based drug design targeting K\u003csub\u003eCa\u003c/sub\u003e2.2 channels.\u003c/p\u003e","manuscriptTitle":"Structural basis for the subtype-selectivity of KCa2.2 channel activators","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-05-16 08:24:38","doi":"10.21203/rs.3.rs-6568445/v1","editorialEvents":[],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"nature-communications","isNatureJournal":true,"hasQc":false,"allowDirectSubmit":false,"externalIdentity":"NCOMMS","sideBox":"Learn more about [Nature Communications](http://www.nature.com/ncomms/)","snPcode":"","submissionUrl":"https://mts-ncomms.nature.com/","title":"Nature Communications","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"ejp","reportingPortfolio":"Nature Communications","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"bf8cb515-f7c1-45a5-a308-694c6a1234ad","owner":[],"postedDate":"May 16th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[{"id":48621186,"name":"Biological sciences/Biochemistry/Ion channels/Potassium channels"},{"id":48621187,"name":"Biological sciences/Structural biology/Electron microscopy/Cryoelectron microscopy"}],"tags":[],"updatedAt":"2026-01-15T08:10:34+00:00","versionOfRecord":{"articleIdentity":"rs-6568445","link":"https://doi.org/10.1038/s41467-025-67232-3","journal":{"identity":"nature-communications","isVorOnly":false,"title":"Nature Communications"},"publishedOn":"2026-01-08 05:00:00","publishedOnDateReadable":"January 8th, 2026"},"versionCreatedAt":"2025-05-16 08:24:38","video":"","vorDoi":"10.1038/s41467-025-67232-3","vorDoiUrl":"https://doi.org/10.1038/s41467-025-67232-3","workflowStages":[]},"version":"v1","identity":"rs-6568445","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-6568445","identity":"rs-6568445","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}