Dual receptor-sites reveal the structural basis for hyperactivation of sodium channels by poison-dart toxin batrachotoxin

Tonggu, Lige; Wisedchaisri, Goragot; Gamal El-Din, Tamer M.; Lenaeus, Michael J.; Logan, Matthew M.; Toma, Tatsuya; Du Bois, Justin; Zheng, Ning; Catterall, William A.

doi:10.1038/s41467-024-45958-w

Download PDF

Article
Open access
Published: 14 March 2024

Dual receptor-sites reveal the structural basis for hyperactivation of sodium channels by poison-dart toxin batrachotoxin

Nature Communications volume 15, Article number: 2306 (2024) Cite this article

1695 Accesses
21 Altmetric
Metrics details

Subjects

Abstract

The poison dart toxin batrachotoxin is exceptional for its high potency and toxicity, and for its multifaceted modification of the function of voltage-gated sodium channels. By using cryogenic electron microscopy, we identify two homologous, but nonidentical receptor sites that simultaneously bind two molecules of toxin, one at the interface between Domains I and IV, and the other at the interface between Domains III and IV of the cardiac sodium channel. Together, these two bound toxin molecules stabilize α/π helical conformation in the S6 segments that gate the pore, and one of the bound BTX-B molecules interacts with the crucial Lys1421 residue that is essential for sodium conductance and selectivity via an apparent water-bridged hydrogen bond. Overall, our structure provides insight into batrachotoxin’s potency, efficacy, and multifaceted functional effects on voltage-gated sodium channels via a dual receptor site mechanism.

Stepwise activation of a metabotropic glutamate receptor

Article 17 April 2024

Streptomyces umbrella toxin particles block hyphal growth of competing species

Article Open access 17 April 2024

Bitter taste receptor activation by cholesterol and an intracellular tastant

Article 10 April 2024

Introduction

Voltage-gated sodium channels (Na_Vs) initiate action potentials in nerve and muscle^1,2. Na_Vs are composed of four pseudosymmetric, homologous domains (DI-DIV) having six transmembrane segments each (S1–S6), which surround a central transmembrane pore^3,4. Segments S1–S4 form the voltage sensors (VS) in each domain. Segments S5 and S6, and the P loop between them, form the pore module (PM). Structural studies using X-ray crystallography revealed the overall transmembrane structure of ancestral bacterial sodium channels^5,6,7, and analyses by cryogenic electron microscopy (cryo-EM) have elucidated the structures of sodium channels from mammalian nerve^8,9, skeletal muscle¹⁰, and heart¹¹. In all of these structures, the central PM is composed of an extracellular vestibule, a narrow ion selectivity filter, a central cavity, and an activation gate formed by the intracellular ends of the pore-lining S6 segments. The four VS surround the central PM in a square array and modulate pore opening through the connecting S4-S5 linkers. Depolarized membrane potentials activate the VS from DI to DIII, which leads to pore opening and sodium influx. Within milliseconds, the VS in DIV triggers fast inactivation to close the pore using the fast inactivation gate located in the intracellular DIII-DIV linker.

Many highly toxic small molecules and peptides produced by a wide range of plant and animal species bind to five or more specific receptor sites on sodium channels and paralyze predators and prey^4,12,13. Batrachotoxin (BTX) is unique in that it is small, skin permeable, and highly potent, with a mean lethal dose in mice of 2 μg/kg¹⁴, the most potent of all sodium channel toxins. BTX is found predominantly in poison dart frogs in Central and South America^15,16,17,18 and in toxic birds in Papua New Guinea¹⁹. The Dendrobatid frogs use BTX for defense against attacks by predators^17,18, which provides a graphic illlustration of its speed and potency. The poison secreted from the skin of these tiny frogs has been used for centuries by indigenous tribes for hunting^16,17,18. A single Phyllobates terribilus frog can poison many darts, which are each sufficient to immobilize or kill prey.

Structural studies show that the Neurotoxin Receptor Site I, which binds the pore-blocking toxins tetrodotoxin, saxitoxin and μ-conotoxin, is located in the outer PM^9,10, and the receptor sites for Site III and Site IV polypeptide toxins from spiders and scorpions are located on the extracellular surfaces of the VS in Domains IV and II, respectively^9,20,21,22. In contrast, mutagenesis and structure/function studies indicate that Neurotoxin Receptor Site II containing the BTX binding site is located in the central cavity of the PM, spanning Domains I, III and IV, with contact residues from transmembrane segments DI-S6, DIII-S6 and DIV-S6^{23,24,25,26,27}. As a selective agonist, BTX causes a hyperpolarizing shift in the voltage-dependence of activation, inhibition of fast inactivation, reduced single channel conductance, and a striking ten-fold increase in calcium selectivity for Na_Vs²⁸. The negative shift in the voltage dependence of activation and the block of fast inactivation generate repetitive action potential firing and persistent depolarization, whereas the increase in calcium permeation further induces hyperactivity at synapses, hypercontraction of skeletal muscles, and life-threatening arrhythmias in the heart^29,30,31,32. All of this storm of gain-of-function effects requires persistent activation of only a few percent of Na_V channels, contributing greatly to the potency of the lethal effects of the toxin. Aconitine and veratridine from plants in the Aconitum and Veratrum genera, respectively, also act at Site II, but with lower affinity and potency than BTX^33,34,35. The structural and chemical basis for the high potency of BTX, and its strong effects on voltage-dependent activation, fast inactivation, ion conductance, and calcium selectivity has remained unclear.

Here, by taking advantage of new methods for total synthesis of BTX and derivatives^36,37, we report the cryo-EM structure of the cardiac sodium channel Na_V1.5 in complex with the BTX derivative batrachotoxinin-A 20-α-benzoate (BTX-B), functionally equivalent to BTX^{24,36,37,38,39} (Fig. 1a). Surprisingly, our results reveal dual toxin/receptor interaction sites at the Domain I/IV and Domain III/IV interfaces in the central cavity. These unexpected structural findings provide mechanistic insight into the high potency and efficacy as well as the multifaceted functional effects of BTX on Na_Vs.

**Fig. 1: Electrophysiological characterization and overall structure of rNa_V1.5c with BTX-B.**

Results

BTX-B effects on rNa_V1.5c

BTX is an ester of a 20-α-pyrrole carboxylic acid with the complex steroidal alkaloid Batrachotoxinin-A (BTX-A; Fig. 1a). Batrachotoxinin-A 20-α-benzoate (BTX-B; Fig. 1b) is an ester of benzoic acid and the same complex steroidal alkaloid backbone, in which the benzoate moiety replaces the 20-α-pyrrole ester in native BTX (Fig. 1b)³⁸. Importantly, BTX-B and BTX bind to the same receptor site and have equivalent potency and efficacy on nerve and muscle sodium channels^35,38. For structural studies, we used the construct rat Na_V1.5c with deletions in the DI-DII and DII-DIII intracellular linkers, and the C-terminus (rNa_V1.5c) in order to enhance expression and improve resolution in structural analysis⁴⁰. The rNa_V1.5c protein was expressed in HEK293 cells as reported previously (see Methods)^40,41. Patch-clamp electrophysiological recording showed that rNa_V1.5c is sensitive to BTX-B (Fig. 1c). At 10 μM concentration, BTX-B prolonged the peak current by nearly completely inhibiting fast inactivation (Fig. 1c). We used this potent effect of BTX-B as a quantitative metric of toxin activity. However, in contrast to its action on native sodium channels, BTX-B negatively shifted the foot of the activation curve but did not significantly shift the voltage-dependent activation of rNa_V1.5c measured at the midpoint relative to the control (Fig. 1d). This is likely because the rNa_V1.5c channel already has a strong negative shift in the voltage dependence of activation compared to wild type (WT)¹¹, which might occlude the full effect of BTX-B to further negatively shift the G/V curve.

To determine the structure of rNa_V1.5c in complex with BTX-B, rNa_V1.5c was purified in the presence of 600 nM BTX-B throughout (see Methods). Purified rNa_V1.5c was mixed with the α-scorpion toxin LqhIII, which prevents fast inactivation and increases the affinity for BTX¹², and the complex was further purified by size-exclusion chromatography (Supplementary Fig. 1a, b). We performed cryo-EM single particle analysis of the rNa_V1.5c/BTX-B sample frozen in vitreous ice from 7542 movie stacks collected on a 300-KeV Titan Krios microscope (Supplementary Figs. 1c, d and 2a, and Supplementary Table 1). A total of 86,763 particles were selected for refinement and 3D reconstruction, which yielded the final map at 3.3 Å overall resolution according to the Gold Standard Fourier Shell Correlation (FSC) criterion of 0.143 (Supplementary Fig. 2b). In the resulting structure, the transmembrane PM has a local resolution of ~3.0 Å, while the resolution of the intracellular N-terminal domain (NTD), VS, and the extracellular loops (ECL) is between 3.0–4.0 Å (Fig. 1e). In discussing individual amino acid residues, we use their numbers in the complete rNa_V1.5 amino acid sequence so that the deletions of intracellular domains in the Na_V1.5c construct do not alter the residue numbers.

Overall structure

The structure of rNa_V1.5c/BTX-B contains 1238 amino acids spanning residue 11 in the NTD to residue 1780 at the end of the DIV-S6 segment (Fig. 1e, f). The cryo-EM density for the NTD (residues 11–32 and 53–117) has a local resolution range of 3.5–4.0 Å (Fig. 1e) and resembles the N-terminal domain in both the predicted structure by AlphaFold⁴², and the conformation revealed in two recently reported structures of human Na_V1.7^43,44,45. The well-resolved density of the map in this region allowed unambiguous docking and incorporation of the NTD into the model of Na_V1.5 for the first time (Fig. 1f and Supplementary Fig. 3; Supplementary Information). The NTD appears to be stabilized by the intracellular helix of the DI-S6 segment that extends far beyond the membrane into the cytoplasm (Supplementary Fig. 3a, b) and by a hydrogen bond with the DI S4-S5 linker (Supplementary Fig. 3c). Since the NTD and the intracellular DI-S6 helix were not well resolved in any previously reported structures of Na_V1.5, we hypothesize that the binding of BTX-B may stabilize this region through allosteric interactions. As the BTX-B receptor site includes several residues in the upstream transmembrane region of the DI-S6 segment (see below), these interactions may help to rigidify the remaining cytoplasmic DI-S6 helix and in turn stabilize the conformation of the NTD.

Annular lipids/detergents are also well defined in the density map, and the final model includes putative cholesterol hemisuccinate (CHS), 1-palmitoyl-2-oleoyl-sn-glycero-3-phosphocholine (POPC), and glycodiosgenin (GDN) (Fig. 1f and Supplementary Fig. 4; Supplementary Information). CHS and GDN were included in the sample preparation, while POPC is a representative of major phospholipids co-purified from the plasma membrane. Seven N-linked glycosylation sites are located in the ECL connecting the S5 segments and P-loop segments in DI (Asn284, Asn319, Asn329) and DIII (Asn1367, Asn1376, Asn1382, Asn1390) (Fig. 1f).

The overall structure of rNa_V1.5c/BTX-B is similar to that of apo rNa_V1.5c (PDB: 3UZ3) and flecainide-bound rNa_V1.5c (PDB: 3UZ0), with a root mean square deviation (RMSD) for Cα atoms of ~1.2 Å and ~1.0 Å, respectively, indicating essentially no major conformational changes in the backbone of rNa_V1.5c upon BTX-B binding. The S4 segments in all VS are translocated outward, resembling the activated VS conformation (Fig. 1f). The fast inactivation gate and its IFM motif in the DIII-DIV linker are lodged in their receptor site near the junction of the DIII and DIV S4-S5 linkers to keep the pore closed (Fig. 1f, orange). Despite the addition of LqhIII in our sample preparation, the toxin is not visible in the density map compared to the structure of rNa_V1.5c/LqhIII complex (PDB: 7K18) with the DIV-VS trapped in an intermediate conformation²², possibly because the fully activated DIV-VS in rNa_V1.5c/BTX-B prevented LqhIII from binding with high affinity. Therefore, our rNa_V1.5/BTX-B structure is most likely captured in an inactivated state with the VS activated and the pore closed.

Dual BTX-B receptor sites

Because past mutagenesis and photolabeling studies have suggested that BTX binds in the PM near the S6 segments of Domains I, III, and IV^{23,24,25,26,27}, we examined our cryo-EM density map for strong density candidates corresponding to BTX-B in the areas that were supported by past functional studies. Two equally strong (~10σ peak) density components with a shape similar in size and form to BTX-B were observed simultaneously in the central cavity of the PM at the level of the fenestrations (Fig. 2a–d), where the local resolution of the map is ~3.0 Å. In order to distinguish these densities from endogenous lipids often found in the central cavity and the fenestrations, we compared our density map to cryo-EM maps of other related rNa_V1.5c structures, which employed similar methods for preparation and analysis but without BTX-B (Supplementary Fig. 5). The BTX-B densities are much bulkier and longer than the weaker, thinner, and shorter densities observed in other rNa_V1.5c structures without BTX-B, and the BTX-B densities resemble the BTX-B molecule in size and shape (Supplementary Fig. 5).

**Fig. 2: The dual receptor sites for BTX-B in rNa_V1.5c.**

Because BTX-B is a lipid-soluble toxin, we expected strong hydrophobic interactions of the toxin with lipids, which may result in contiguous densities connecting the bound toxin with lipids. Indeed such contiguous density is observed in our density map (Supplementary Fig. 4), but the phospholipid is easily distinguished from BTX-B by size and shape. In addition, the density level and connectivity for adjoining lipids varied in different refinement cycles, but the strong, bulky densities for BTX-B were consistent throughout all of the refinements. Based on these observations, we therefore concluded that the two cryo-EM densities observed in our map are bound BTX-B, with possibly some minor partial overlap of adjoining lipids, despite our best efforts to resolve these particles during image processing.

With the presence of BTX-B in our sample, the NTD is better resolved indicating some structural stabilization by BTX-B. Despite the use of LqhIII in a similar preparation as reported previously²², which partially deactivated the DIV-VS into a down state when a complex is formed, the DIV-VS is fully activated in the presence of BTX-B. There are also clear local conformational changes (see below) in the channel to support the binding of BTX-B that result in no structural clash.

One density specific for BTX-B is located partially in the DIII-DIV fenestration and on the intracellular side of the DIII-DIV selectivity filter (Fig. 2a, b). The components of PMIII and PMIV helices that bind BTX-B (cyan sticks) in this site are illustrated as dark yellow and dark red ribbons, respectively (Fig. 2b, c). The other density specific for BTX-B is located completely within the fenestration at the DI-DIV interface (Fig. 2a, b). BTX-B (bright green sticks) binds between the PMI and PMIV helices, as illustrated in dark blue and dark red ribbons, respectively, making strong interactions with the S6 segments in both domains (Fig. 2b, d). We designate these two BTX-B receptor sites as Site IIB₁ for the site in the DI/DIV interface and Site IIB₂ for the site in the DIII/DIV interface, reflecting their functional roles as BTX-binding motifs in Neurotoxin Receptor Site II.

For Site IIB₂, BTX-B lies horizontally parallel to the membrane plane with the benzoate moiety facing the center of the pore, while the BTX-A core is inserted into the fenestration between the S6 segments from DIII (dark yellow) and DIV (dark red; Fig. 2a, b). The mesh representing cryo-EM density fits the toxin structure closely (cyan sticks, Fig. 2c). For Site IIB₁, BTX-B also lies horizontally parallel to the membrane plane, but completely in the fenestration between DI and DIV, with the benzoate moiety pointing toward the pore (Fig. 2a, b). The mesh representing cryo-EM density fits closely with the structure of bound BTX-B (bright green sticks, Fig. 2d). Thus, these two toxin receptor sites lie in homologous subunit interfaces in pseudosymmetric positions, poised to modulate voltage-dependent gating in Domains I, III, and IV and ion selectivity through interactions with the nearby ion selectivity filter formed by the P loops.

Contact Amino Acid Residues in the Dual BTX-B Receptor Sites

As illustrated in Fig. 3a, the bound BTX-B at Site IIB₂ (cyan) nests in a cradle formed by hydrophobic residues from the DII P loop (green), segments DIII-S5 and DIII-S6 (Fig. 3a, yellow), and segment DIV-S6 (Fig. 3a, red; Supplementary Fig. 6a; and Supplementary Movie 1). As highlighted in yellow in Fig. 3a, the DIII P-loop amino acid residues L1415, A1418, T1419, F1420, and the DIII S6 residues G1459, S1460, T1463, L1464, and F1467 form the main binding site for the BTX-A core (Fig. 3a, yellow; Supplementary Fig. 6a; and Supplementary Movie 1, yellow sticks). Remarkably, K1421, which is part of the conserved DEKA locus in the DIII P-loop that forms the selectivity filter for sodium ions, interacts with the carbonyl oxygen of the benzoate moiety of BTX-B through an unexpected water-mediated hydrogen-bonding interaction (Fig. 3a, yellow). This carbonyl oxygen in BTX-B is also present in native BTX, highlighting a critical role of this functional group in altering ion selectivity by both compounds (Fig. 1a). The main chain oxygen of T1419 from the DIII P-loop and the side chain of S1712 from the DIV P-loop also coordinate this water molecule. Moreover, F1420 adjacent to K1421 forms an orthogonal (T-shape) π-π stacking with the aryl ring of the benzoate moiety of BTX-B that likely provides substantial binding energy (Fig. 3a, yellow). Finally, DIII contributes C1343 and W1347 from the S5 segment to the binding site (Fig. 3a, yellow, Supplementary Fig. 6a, and Supplementary Movie 1, yellow sticks). Thus, a total of 12 amino acid side chains from DIII work together to form the core of Site IIB₂ (Fig. 3a, yellow).

**Fig. 3: Close up views of the dual receptor sites.**

In addition to these extensive interactions with DIII, S1712 from the P-loop and I1758, I1759, and F1762 from the S6 segment in DIV all contribute to Site IIB₂ through interactions with the BTX-A core (Fig. 3a, red; Supplementary Fig. 6a; and Supplementary Movie 1, red sticks). The phenyl ring of F1762 makes a strong and well-elaborated hydrophobic interaction with the hydrophobic core of BTX-B, consistent with the strong negative impact on toxin binding and sodium channel activation for mutations at this site²⁴. On the other hand, C899 and G900 are the only residues from DII in the P-loop that interact with BTX-B bound in Site IIB₂ through its benzoate moiety using their main chain backbone atoms rather than their side chains (Fig. 3a, upper left quadrant, green; Supplementary Fig. 6a; and Supplementary Movie 1, lime green sticks). Together with the extensive amino acid contacts with DIII, these interactions with additional amino acid residues in DII and DIV make an exceptionally strong, specific, and three-dimensional Site IIB₂ for high-affinity binding of BTX-B.

As illustrated in Fig. 3b, the bound BTX-B at Site IIB₁ (bright green) nests in a cradle formed by hydrophobic residues from the DI P-loop and segment DI-S6 (Fig. 3b, blue), and segments DIV-S5 and DIV-S6 (Fig. 3b, red; Supplementary Fig. 6b; and Supplementary Movie 2). As highlighted in blue in Fig. 3b, the DI amino acid residue Q372 in the P-loop and M395, I398, F399, S402, and F403 in the DI-S6 segment form a key part of Site IIB₁ (Fig. 3b, blue; Supplementary Fig. 6b; and Supplementary Movie 2, blue sticks) through strong interaction with bound BTX-B (Fig. 3b, bright green). In addition to these extensive interactions with DI, several residues in DIV contribute to the binding of BTX-B to Site IIB₁ as highlighted in red (Fig. 3b, red; Supplementary Fig. 6b; and Supplementary Movie 2, red sticks). These amino acid residues in DIV include L1659, F1660, G1663, L1666, F1667 and M1670 from the DIV-S5 segment; F1707, T1710, and T1711 from the P-loop; and V1765, M1768, and Y1769 from the DIV-S6 segment (Fig. 3b, red, Supplementary Fig. 6b, and Supplementary Movie 2, red sticks). Altogether, these amino acid residues surround the bound BTX-B (bright green sticks) and form a high-affinity three-dimensional Site IIB₁ for binding BTX-B.

The binding of BTX-B causes local induced-fit conformational changes in the rotamer of some key amino acid side chains. Notably, F399 and F403 from DI-S6, and F1667 from DIV-S5 adopt a “down” rotamer that allows for the binding of BTX-B in Site IIB₁ (Fig. 3b, blue: Supplementary Fig. 6b, and Supplementary Movie 2). F1467 from DIII-S6 also adopts a new rotamer to interact with BTX-B in Site IIB₂ (Fig. 3a, yellow; Supplementary Figs. 5d and 6a, and Supplementary Movie 1). In the absence of the toxin, these residues adopt different rotamers in other Na_V1.5 structures that would either clash or make no interaction with BTX-B^11,46. Further information on these rotamers in other Na_V1.5 structures is included in the legend to Fig. 3b. All amino acid residues forming the BTX-B receptor sites are conserved in human Na_V1.5 and many of them have clinical variants in which mutations are associated with cardiac arrhythmia (Supplementary Table 2).

Functional characterization of dual BTX receptor sites

Extensive photoaffinity labeling and mutagenesis studies in DI, DIII and DIV have shown that residues in these domains are critical for the binding and action of BTX-B^{23,24,25,26,27}. These previous mutagenesis studies identified several individual residues observed in Site IIB₂ in our structure as part of the receptor site for BTX (see next section). On the other hand, Site IIB₁ residues in DI were present in a peptide that was photoaffinity labeled by a photoreactive azido-BTX-B derivative²³ and the channels became BTX resistant when individual residues found in this photolabeled DI segment were mutated (see next section). However, Site IIB₁ contains some residues that have not been previously characterized by mutagenesis. To test whether the dual receptor sites observed in our cryo-EM structure of rNa_V1.5c are essential for the agonistic effect of BTX-B, we performed structure-guided mutagenesis and patch-clamp electrophysiological experiments with 10 μM BTX-B (Fig. 4 and Supplementary Figs. 8–11). A whole family of sodium current traces (−100 to 0 mV) were measured before and after bath application of BTX-B (Supplementary Figs. 8–11). At the depolarizing voltage that led to the biggest current amplitude, we measured the sodium current magnitude at 40 ms after the start of the depolarizing pulse and divided this value by the amplitude of the peak current in the same trace to obtain the inhibition ratio R_i as a quantitative metric for the BTX-B action (Supplementary Figs. 10-11 and Supplementary Table 3) (see Methods). For the WT channel, the R_i ratio was 0.78. As a positive control, we selected the DIV-S6 residue F1762 from Site IIB₂ for our initial mutagenesis study (Fig. 4a), as mutation of this residue has been shown to cause BTX resistance, and several other residues surrounding this site have already been characterized and validated for various Na_V subtypes^{24,26,36,47,48} (see next section). As expected, the F1762A showed a significant reduction of the functional effect of BTX-B by reducing the inhibition ratio R_i to 0.20 (p < 0.001) (Fig. 4c, d). The F1762A mutation almost completely abolished the effect of BTX-B as the channel became nearly totally BTX-B resistant, consistent with results from previous studies²⁴. As a negative control, we tested the DIV-S6 residue Y1769 (Fig. 4b), where mutations have been shown to have no effect on BTX function^24,26. Consistent with previous studies, Y1769A did not show any reduction in the ratio R_i from WT (Fig. 4c, d).

**Fig. 4: Functional characterization of the dual receptor sites for BTX-B.**

To evaluate Site IIB₁ further, residues F399 and F403 from the photolabeled segment in DI-S6, F1707 from DIV P-loop, and M1768 from DIV-S6 were selected (Fig. 4b). The F403A mutation caused a significant reduction of the BTX-B effect by reducing R_i to 0.23 (p < 0.001) (Fig. 4c, d). The M1768A and F1707A mutations gave significantly smaller R_i ratios of 0.55 (p < 0.05) and 0.61 (p < 0.05), respectively, compared to the value of 0.78 for the WT (Fig. 4 c, d), showing significant BTX-B resistance induced by these mutations. When we introduced the F1707A/M1768A double mutation to rNa_V1.5c, the R_i ratio further decreased to 0.32 (p < 0.001), indicating a synergistic effect on BTX-B function between these two residues (Fig. 4c, d). The R_i value for the F1707A/M1768A double mutations was similar to the R_i value of 0.20 for the F1762A mutation in Site IIB₂. In contrast, the F399A mutation did not show any effect on BTX-B function, as the calculated R_i ratio was 0.78, similar to the WT (Fig. 4c, d). It is possible that, due to its multiple rotamers, the bulky side chain of F399 may act as a gate to the fenestration rather than contributing directly to the BTX-B effect. Altogether, these results suggest that Site IIB₁ observed in our structure plays a significant role in the agonistic action of BTX-B.

Structure/function analysis of dual BTX receptor sites

Our dual BTX-B receptor sites are well supported by previous mutagenesis and photoaffinity labeling studies, as summarized with colored shading of the numerous interacting components of these two binding sites with bound BTX-B in Fig. 5 and Supplementary Table 4. For Site IIB₂, contact residues F1420 in DIII P-loop; S1460 and L1464 in DIII-S6 (Fig. 5a, red); and I1758, F1762, and V1766 in DIV-S6 (Fig. 5a, yellow) are all implicated in BTX binding and action by mutagenesis studies in other Na_V channels (Fig. 5a). F1236K and F1236R mutations in rNa_V1.4 (equivalent to F1420 in rNa_V1.5c) were also resistant to BTX⁴⁹ (Fig. 5a, yellow). S1276K and L1280K mutations in DIII-S6 of rNa_V1.4 (equivalent to S1460 and L1464 in rNa_V1.5c) caused the channel to become BTX resistant^25,27 (Fig. 5a, yellow)). The I1760A mutation in DIV-S6 in Na_V1.2 (I1758 in rNa_V1.5c) caused the channel to have significantly reduced sensitivity to BTX²⁴ (Fig. 5a, red). Mutations of residues equivalent to rNav1.5c F1762 in other channels (F1764A in Na_V1.2, F1710A and F1710I in Na_V1.3, and F1579K in rNa_V1.4) made them BTX resistant^24,26,36,48 (Fig. 5a, F1762, red), consistent with the F1762A mutation in our study (Fig. 3).

**Fig. 5: Mapping of mutagenesis studies for BTX-B on the structure of rNa_V1.5c.**

For Site IIB₁, the peptide segment N388 to Q429 in rat Na_V1.2 (corresponding to residues 377-418 in rNa_V1.5c) that included residues in DI-S6 (Fig. 5b, light blue) was found to be a binding site for BTX-B, as this segment of the channel was identified following peptide cleavage and antibody mapping to be specifically labeled by a photoreactive azido-derivative of BTX-B²³. Mutation of S401 in hNa_V1.5 (S402 in rNa_V1.5c) to R or K prevented BTX action (Fig. 5c, blue), indicating that this position is close enough to the BTX binding site to allow steric or charge-charge clashes with the substituted R or K side chains, as mutation of this residue to other amino acids had no effect on BTX action⁵⁰. The I433K, N434K, and L437K mutations in DI-S6 of rNa_V1.4 (V406, N407, and L410 in rNa_V1.5c) were BTX resistant while F430K (F403 in rNa_V1.5c) remained BTX sensitive^25,36 (Fig. 5c, blue). This is in contrast to the F403A mutation in our study that showed a significant BTX-B resistance (Fig. 4). For DIV-S6, BTX blocked cysteine modification in rNa_V1.4 with V1583C mutation (V1766 in rNav1.5c; Fig. 5a, c; red)⁴⁷. N1584K and N1584A mutations in rNa_V1.4 (Asn1767 in rNa_V1.5c) were BTX resistant²⁶ (Fig. 5c, red). Interestingly, mutation of the residues equivalent to Y1769 in other sodium channels (Y1771A in Na_V1.2, and Y1586K and Y1586A in rNa_V1.4) had no effect, and the channels remained fully BTX sensitive similar to the Y1769A mutation in our study^24,26 (Fig. 5c, red). Altogether, these structure/function results summarized in Fig. 5 and Supplementary Table 4 further define the functional roles of a peptide segment in the DI-S6 segment and individual amino acid residues in DI, DIII, and DIV in formation of BTX Site IIB₁ and Site IIB₂, providing strong support for the chemical interactions that we have observed at atomic resolution in our structure.

Structural basis for agonistic actions of BTX-B

It has been suggested that S402 (DI-S6), N930 (DII-S6), S1460 (DIII-S6), and F1762 (DIV-S6) form the SNSF pore-facing ring in the inner cavity of sodium channels that is near to or forms a part of the receptor sites for BTX-B (Fig. 6a)⁵⁰. Each residue in the SNSF pore-facing ring is immediately adjacent to the flexible gating hinge G401, G929, G1459, and S1761 (GGGS) locus that is conserved among vertebrate sodium channels (Fig. 6a). Our structure shows that S402 directly forms an important interaction within Site IIB₁, whereas G1459, S1460 and F1762 are part of Site IIB₂ (Figs. 3 and 6a). On the contrary, the N930 side chain is 6.5 Å away from the benzoate moiety of BTX-B in Site IIB₂ (Fig. 6a). When the corresponding residues in rat Na_V1.4 (N784) and human Na_V1.5 (N927) were mutated to K or R, the mutant channel became resistant to BTX, possibly due to adverse effects of the positive charge introduced with K or R side chains, because mutation to other amino acids had no effect^51,52. Based on our analysis, the longer side chain of N927K and N927R mutations in human Na_V1.5 (also N784K and L788K in rNa_V1.4) would bring this residue as close as 4 Å to Site IIB₂. This would limit the space in the pore and create charge repulsion that may block local conformational changes required for pore opening and may present a barrier for hydrated sodium ion to pass, while other mutations with a short side chain can still provide enough space in the cavity.

**Fig. 6: Mechanism for the agonistic effects of BTX-B.**

The binding of BTX-B to both Site IIB₁ and Site IIB₂ appears to stabilize the π-helix conformation of the DI-S6 and DIII-S6 segments, and the α-helical conformation in DII-S6 and DIV-S6 segments near the flexible gating hinge GGGS locus (Fig. 6a). In DI-S6, the π-bulging appears around G401 (Fig. 6a) to accommodate the recruitment of F399 and F403 for the binding at Site IIB₁ (Fig. 3b). In DIII-S6, the π-bulging appears at G1459 (Fig. 6a), which provides the main chain oxygen to the binding of BTX-B in Site IIB₂ (Fig. 3a). On the other hand, the α-helix in DIV-S6 is stabilized by the binding of I1758, I1759, and F1762 to Site IIB₂ (Fig. 3a) adjacent to S1761 in the GGGS locus, and also by the binding of V1765, M1768 and Y1769 (Fig. 3b) near the conserved gating residue N1767. DII-S6 does not contribute to the binding of either BTX-B site and forms a continuous α-helix throughout the entire S6 segment. It is conceivable that this two-fold symmetric configuration of α-helix and π-helix in the four S6 segments allows BTX-B to stabilize an activated state of the channel without triggering fast inactivation.

Comparison to related structures of Na_V1.5 revealed that these π- and α-helix arrangements in the S6 segments of rNa_V1.5c/BTX-B are similar to those in some other structures. For DIV-S6, Site IIB₂ recruits F1762 (Fig. 6a, red) and nearby residues I1758 and I1759 (Fig. 6a, red arrows) for BTX-B binding and stabilizes the S6 conformation at the flexible gating hinge in the α-helical configuration. The α-helical form seen in this region is similar to the conformation found in other Na_V1.5 structures with the exception of the recently reported ranolazine-bound structure of rNa_V1.5c, which adopted the π-configuration instead⁵³. The π-helix in DIV-S6 has also been seen in a fraction of the single particles in the recently reanalyzed structures of human Na_V1.7 with Site I and IV toxins bound⁴³. The DIV-S6 π-to-α transition may be important for channel gating and can be modulated either directly or allosterically by drug and toxin ligands such as BTX-B to stabilize specific states of sodium channels.

We further examined whether the binding of BTX-B observed in our sodium channel structure captured in an inactivated state is compatible with the open-state conformation through which BTX-B asserts its agonistic effect. We compared our structure with the structure of the fast-inactivation-impaired rNa_V1.5c/QQQ (PDB: 7FBS), in which fast inactivation has been prevented by mutation of the essential IFM motif to QQQ (Fig. 6b, c)⁴⁰. A major difference between the two structures occurs at the C-terminus of the DIV-S6 segment, starting at the conserved gating residue N1767, where the S6 α-helix begins to bend significantly away from the pore axis and widen the pore at the activation gate in the open-state structure (Fig. 6b, pink vs. dark red). The DIV-S6 helix in the open state does not clash with BTX-B because the conformational change of the S6 segment mainly appears on the C-terminal side of the BTX-B receptor site (Fig. 6b). Our analysis revealed that both BTX-B sites are compatible with the open state channel and require only changes in the backbone of flexible gating hinge regions and in the rotamers of a few side chains in the open state to avoid clashes. Only F1467 in DIII-S6 in Site IIB₂ (Supplementary Fig. 7a, dark yellow vs. light yellow sticks) and F399 and F403 in DI-S6 in Site IIB₁ (Supplementary Fig. 7b, dark blue vs. light blue sticks), need to change their rotamers from “up” to “down” as seen in our structure to avoid clashing with BTX-B in the open state. The conformational change in the DIV-S6 segment in the open state shifted the amino acid register of residues on the C-terminal side of N1767 such that Y1769 displaces M1768 and can potentially form a hydrogen bond with BTX-B Site IIB₁ (Fig. 6c, pink) to stabilize the open state.

BTX-B Site IIB₂ overlaps with other Site II neurotoxins

BTX-B is thought to have similar or overlapping binding sites with the other classic Site II neurotoxins, aconitine and veratridine, as demonstrated by ion flux and ligand competition experiments^34,35. Currently only the structure of bulleyaconitine A (BLA), a well-characterized analog of aconitine, has been determined bound to human Na_V1.3 in an inactivated state (PDB: 7W77)⁵⁴. In this structure, a single molecule of BLA is located in the fenestration between DI-DII of the PM, and its binding stabilized the helical conformation of the S6 segments at the flexible gating hinge as π-α-π-α from DI to DIV, similar to the configuration observed in our rNa_V1.5c/BTX-B structure. The structure of BLA-bound Na_V1.3 can be superimposed onto the BTX-B bound rNa_V1.5c structure with ~1.0 Å r.m.s.d. (Fig. 7a). BTX-B bound at Site IIB₂ extends its benzoate moiety toward Domain II and occupies part of the binding site for BLA (Fig. 7b), including the main chain of C899 and G900 from the DII P-loop, and the side chain of F1420 in the DIII P-loop of rNa_V1.5c (Fig. 7c). This structural clash would create competitive binding interactions between BTX and aconitine, as observed in ion flux and ligand binding experiments^34,35. It is possible that this clash might be more prominent in the activated open state of Na_V1.5, which would further impair persistent activation of sodium channels by BTX. Recently, a mutagenesis and computational modeling study has suggested that veratridine binds to rNa_V1.4 at a receptor site similar to Site IIB₂⁵⁵. Our structural results reveal a potential mechanism for the competitive interactions of BTX with both of these structurally related Site II neurotoxins.

**Fig. 7: Overlapping receptor sites for Neurotoxin Receptor Site II.**

Discussion

Proposed mechanism of action of BTX

Binding of BTX favors activation of sodium channels by negatively shifting the voltage dependence of activation and preventing fast inactivation²⁸. BTX-activated sodium channels also have altered ion conductance and selectivity²⁸. The molecular and structural basis for these multifaceted toxin actions has been unknown. However, the locations of the dual BTX-B receptor sites at the interfaces of the DI-S6/DIV-S6 and DIII-S6/DIV-S6 segments suggest probable structural mechanisms for these complex actions.

Previous structural studies showed that the DIV-S6 segment of Na_V1.5 bends, twists, and moves 6.1 Å away from the axis of the pore in order to open the activation gate at the intracellular ends of the S6 segments⁴⁰. Shortly after opening of the pore, the fast inactivation gate folds into the PM, and the IFM inactivation particle binds in its receptor site formed by the intracellular end of the DIV-S6 segment and the S4-S5 linker adjacent to the activation gate, which pushes the activation gate into the closed conformation and induces fast inactivation^9,11,40. Based on this previous structural work, we propose that the unique BTX-induced conformation of the DIV-S6 segment has two important effects. First, it favors bending the S6 segment toward its activated position by modifying the GGGS gating hinge through direct interactions with Site IIB₁ and Site IIB₂ and by inducing the unique π-α-π-α conformation of the S6 helices in Domains I-IV (Fig. 6a). Second, these S6 movements prevent fast inactivation by greatly slowing the binding of the IFM motif to its receptor site. BTX-B binding at Site IIB₁ also appears to stabilize the interaction between the cytoplasmic end of DI-S6 and the NTD allosterically. Since the NTD interacts with the DI-VS and forms a hydrogen bond with the DI S4-S5 linker in the activated state (Supplementary Fig. 3b), this may further contribute to energetic stabilization of the activated state.

Our structure suggests that BTX-B may cause a reduction of the single channel conductance by binding to Site IIB₂ and physically reducing the space in the central cavity of the pore, thereby reducing the rate of sodium influx. Moreover, BTX-B binding at Site IIB₂ also makes an apparent water-mediated hydrogen-bonding interaction with K1421 in the DEKA locus of the selectivity filter (Fig. 3a). Because K1421 is essential for sodium selectivity⁵⁶, toxin binding in this position likely modifies the interaction between K1421 and permeant ions, leading to further alterations in ion conductance and to physiologically important changes in ion selectivity. Site-directed mutations of K1421 greatly increase calcium permeation through Na_V channels⁵⁶. Therefore, the water-bridged hydrogen bond network formed by BTX-B and K1421 may be the source of the pathogenic calcium permeation through Na_V channels induced by BTX. Together, these BTX actions would generate a modified open state with fast inactivation blocked, reduced sodium conductance, and greatly increased calcium selectivity.

Although this model for toxin action is consistent with the position of the BTX receptor sites, the intracellular ends of the S6 segments are interacting closely with each other and the conformation of the activation gate in our structure is tightly closed. Why has binding of BTX-B not opened the pore? Pore-opening requires movement of the intracellular ends of the S5 and S6 segments laterally, away from the axis of the pore. This movement is opposed by the force of the lateral pressure of the lipid bilayer, which is known to alter ion channel gating⁵⁷. In a voltage-gated sodium channel, the S5 and S6 segments interact with this lateral force of the lipid bilayer directly and must overcome it to open the activation gate. We speculate that the lipid/detergent environment of our purified preparation of rNa_V1.5c exerts a larger lateral pressure on the S6 segments than a native lipid bilayer and prevents the BTX-induced conformation of the activation gate and release of the IFM motif. Further structural studies with intact Na_V1.5 and more native lipid environment in nanodiscs or other protein/lipid preparations may allow these conformational movements to proceed and thereby trap the BTX-modified open state of the sodium channel for structure determination.

Dual receptor sites for BTX

Our results reveal surprisingly that there are two receptor sites for BTX, here designated Site IIB₁ and Site IIB₂. Structure/function studies described above support the conclusion that both of these sites contribute to the functional effects of the bound toxin (Figs. 2–5). BTX has much higher affinity and efficacy in persistent activation of sodium channels than the other Site II toxins aconitine and veratridine^33,34, even though these toxins compete with each other for binding and action on sodium channels³⁵. BTX is also far more toxic than aconitine and veratridine¹⁵. The dual binding sites for BTX-B observed in our structure may help to explain the strong agonistic effect of BTX.

Although such a model of dual receptor sites binding two small molecules of a toxin to a single molecular target is uncommon, the bivalent doubleknot toxin from the Earth Tiger tarantula has two peptide toxin moieties tethered in the single toxin molecule. It interacts with two identical target sites in the homotetrameric TRPV1 channel and irreversibly activates the channel⁵⁸. The doubleknot structure of the toxin is required for high-affinity binding and irreversible activation of TRPV1⁵⁸. These results provide clear precedent that binding of two toxin molecules can generate high-affinity binding and irreversible activation of an ion channel⁵⁹. Another example is the Peruvian green velvet tarantula Protoxin-II, which binds to voltage sensors of Na_V1.7 in Domains II and IV, but no evidence for functional effects of toxin molecules bound at these two sites was presented⁹. Our studies show that binding of two small-molecule toxins to separate but homologous sites on a pseudotetrameric sodium channel can give both high affinity binding and irreversible activation and therefore support the proposal that these dual receptor sites underlie the exceptionally potent and multi-faceted effects of the toxin.

Past experiments have shown that BTX binding and action have a Hill coefficient near 1.0 (e.g. ref. ³³), which suggests that there is a single binding site. However, if two binding sites have similar affinity and efficacy of toxin action, a Hill Coefficient near 1.0 may be expected. Moreover, the Hill Coefficient is defined for a process at equilibrium, whereas BTX action is irreversible on the time scale of biochemical and electrophysiological experiments in cellular preparations. Thus, these prior studies do not provide substantial evidence against dual BTX receptor sites.

Fenestrations and BTX access

Fenestrations lead from the phospholipid bilayer into the central cavity of bacterial and mammalian sodium channels^5,9,10,11,60. Their conservation over billions of years of evolution suggests an important biological function, which remains unknown. However, it is clear that fenestrations in the bacterial sodium channel Na_VAb provide a pathway for pore-blocking local anesthetic and antiarrhythmic drugs to reach their receptor sites in the central cavity in the resting state⁶¹. In light of those results, it is intriguing that BTX-B binds within the fenestrations at Sites IIB₁ and IIB₂. We speculate that the fenestrations may provide an access pathway for BTX to reach these two subsites of Neurotoxin Receptor Site II in the resting state of sodium channels. It is conceivable that Site IIB₁ is an intermediate binding site along the fenestration for BTX en route into Site IIB₂ where BTX asserts its agonistic effect. However, our mutagenesis results for Site IIB₁ argue that this is not the case. Mutations of large amino acid side chains to Ala would have widened the fenestration to enhance access of ligands to the central cavity⁶¹. However, these mutations appeared to significantly diminish the functional effect of BTX-B instead of enhancing it, indicating that these side chains are important for the BTX-B function through significant contributions to toxin binding or to conformational stabilization.

Multiple neurotoxin receptor sites on sodium channels

Our studies further define the multi-site model for toxin action on sodium channels at the atomic level^{12,13,20,21,22,62}. The pore blockers tetrodotoxin, saxitoxin, and μ-conotoxins bind in overlapping positions in Neurotoxin Receptor Site I in the external vestibule and physically block the pore⁹. The polypeptide scorpion and sea anemone toxins bind to Neurotoxin Receptor Site III on the extracellular side of the VS in Domain IV and block functional coupling of VS activation to fast inactivation²². Similarly, the polypeptide toxins from spider venom bind to Neurotoxin Receptor Site IV and trap the VS in Domain II in its resting conformation, completely preventing the outward movement of the VS gating charges and the activation of the sodium channel^20,21. BTX and other lipid-soluble toxins act differently, but ligand binding and ion flux studies led to the model that these toxins bind to a common site, designated Neurotoxin Receptor Site II, which enhances activation of sodium channels by an allosteric mechanism^12,34. In contrast to expectation, the structural results presented here and in Li et al. ⁵⁴ show that Site II toxins do not bind to a single binding site. Instead, BLA binds to a site at the interface of PMI and PMII (Fig. 6)⁵⁴, whereas we report here that BTX-B binds to two analogous sites at the interfaces between PMI and PMIV and between PMIII and PMIV. BTX Site IIB₂ overlaps partially with the BLA binding site on DII (Fig. 7). This structural clash provides a molecular mechanism for competitive interactions between these toxins, even though their binding sites are closely spaced but not fully overlapping. In light of these results, we propose that the aconitine binding site be named Neurotoxin Site IIA, as it was first identified at the structural level, and that the BTX site be named Neurotoxin Receptor Site IIB. As we show here, Neurotoxin Receptor Site IIB is composed of two separate binding motifs, which we propose to name Site IIB₁ and Site IIB_2. These dual receptor sites provide a unique mechanism to enhance the potency and efficacy of BTX and greatly increase its toxicity to both predators and prey. In addition to revealing the underlying mechanism for the potent effects of BTX, our results provide a high-resolution template for structure-based design of sodium channel antagonist drugs that might occupy one of these sites, reduce hyperexcitability, and function as sodium channel inhibitors that would be useful in treatment of pain, cardiac arrhythmia, or epilepsy.

Methods

Microbe strains

E. coli GC10 (Genesee Scientific, catalog no. 42-661) was cultured at 37 °C in LB medium supplemented with 100 μg/mL of ampicillin for plasmid DNA extraction. E. coli DH10Bac (ThermoFisher Scientific, catalog no. 10361-012) was cultured at 37 °C in LB medium supplemented with 50 μg/mL kanamycin sulfate, 7 μg/mL gentamicin and 10 μg/mL tetracycline for bacmid production.

Cell lines

Sf9 (Spodoptera frugiperda) insect cells (ThermoFisher Scientific, catalog no. B82501) were maintained in Grace’s Insect Medium and supplemented with 10% FBS and penicillin/streptomycin at 27 °C and passaged at 80–95% confluence for baculovirus production. HEK293S GnTI⁻ (Homo sapiens) mammalian cells (American Type Culture Collection, catalog no. CRL-3022) were maintained and infected on cell culture plates in Dulbecco’s Modified Eagle Medium (DMEM) supplemented with 10% FBS and glutamine/penicillin/streptomycin at 37 °C and 5% CO₂ for electrophysiology and protein expression. The cell lines were not authenticated or tested for Mycoplasma contamination.

Mutagenesis

Site-directed mutagenesis of rNa_V1.5c mutants were generated by polymerase chain reaction (PCR) using Q5 High-Fidelity DNA polymerase (NEB) according to the manufacturer’s protocol with pEG-BacMam rNa_V1.5c-eGFP-Flag plasmid^11,41 as a template. The 5’-phosphorylated oligonucleotides used as mutagenesis primers are listed in Supplementary Data. The resulting PCR products were treated with DpnI (NEB) at 37 °C overnight, followed by a PCR clean up (Omega Bio-tek). The purified DNAs were circularized using T4 DNA ligase (NEB) and transformed into E. coli GC10 chemically competent cells (Genesee Scientific). Plasmid DNAs were isolated from the cultures and mutagenesis results were confirmed by DNA sequencing. Baculoviruses were prepared using the Bac-to-Bac protocol from the plasmids according to the manufacturer with Sf9 insect cells (ThermoFisher Scientific).

Electrophysiological recordings

All experiments were performed at room temperature (21–24 °C) as described previously¹¹. rNa_V1.5c and its mutants were overexpressed in HEK293S GnTI^- cells using baculoviruses. Unless otherwise stated, HEK293S GnTI⁻ cells were held at −140 mV and 50-ms pulses were applied in 10 mV increments from −140 mV to +10 mV. A P/−4 holding leak potential was set to −140 mV. Extracellular solution contained in mM: 140 NaCl, 2 CaCl₂, 2 MgCl₂, 10 HEPES, pH 7.4. Intracellular solution contained: 35 NaCl, 105 CsF, 10 EGTA, 10 HEPES, pH 7.4. Glass electrodes had a resistance 1.5–2 MΩ. The pulse protocol was repeated every 5 min to assure that the effect of the 10 μM BTX-B had reached a plateau. Currents resulting from applied pulses were filtered at 5 kHz with a low-pass Bessel filter, and then digitized at 20 kHz. Data were acquired using an Axopatch 200B amplifier (Molecular Devices). Voltage commands were generated using Pulse 8.5 software (HEKA, Germany), and ITC18 analog-to-digital interface (Instrutech, Port Washington, NY).

Analysis of electrophysiological data

Original records of examples of our electrophysiological analyses are presented in Supplementary Figs. 8–11. Current–voltage (I–V) relationships were recorded in response to voltage steps ranging from −140 to +10 mV in 10-mV increments from a holding potential of −140 mV. Conductance-voltage (G − V) curves were calculated from the corresponding (I–V) curves. Data were analyzed using Igor Pro 6.37 (WaveMetrics). G − V curves were fit with a Boltzmann Equation 1/(1+exp(V_1/2 − V_p)/k) in which V_p is the stimulus potential, V_1/2 is the half-activation voltage, and k is the slope factor. The data are presented as mean and standard error of the mean (SEM). Statistical significance was evaluated with Student’s t-test.

To quantify the effect of BTX-B, we selected the pulse potential at 0 mV which gave maximum sodium current and complete steady state inactivation for all the mutants and the WT. We measured the sodium current amplitude at 40 ms after the start of the depolarizing pulse and divided this value by the amplitude of the peak current to obtain the inhibition ratio R_i as a metric for BTX-B action. Because BTX-B slows the rate of inactivation, R_i is large (∼0.8) for WT with a strong toxin effect, whereas R_i is much smaller (∼0.2) for mutations that effectively block the toxin effect.

Expression and purification of rNa_v1.5_C

Detailed expression of rNa_V1.5 _C was described in our previous study^11,41. Briefly, rNa_V1.5c was overexpressed in HEK293S GnTI^- cells using baculovirus generated from pEG-BacMam rNa_V1.5c-eGFP-Flag plasmid^11,41. For purification, 12 L of infected cells were harvested and washed with Phosphate Buffer Saline (PBS) (Gibco) and 10% glycerol. Washed cell pellet was collected by centrifugation at 1000 × g for 20 min and stored at −80 °C. The cell pellet was thawed and resuspended in 160 ml of PBS, supplemented with Roche cOmplete™ protease inhibitor cocktail (Sigma) and 600 nM BTX-B and lysed with Dounce homogenizer. The protein was extracted by mixing 160 mL of the cell lysate with 25 ml of stock solution containing n-dodecyl-β-D-maltopyranoside (DDM, Anatrace) and cholesteryl hemisuccinate Tris salt (CHS, Anatrace) at the final concentration of 1% (w/v) DDM and 0.2% (w/v) CHS with agitation for 2 h at 4 °C. The supernatant was separated by ultra-centrifugation at 100,000 × g for 30 min, then incubated with 5 mL of anti-Flag M2 affinity gel (Sigma) and agitated for 1 h at 4 °C. The resin was washed with 10 column volumes of Buffer A (25 mM HEPES pH = 7.4, 150 mM NaCl, 10% glycerol) supplemented with 0.06% glycol-diosgenin (GDN, Anatrace) and 600 nM BTX-B. The rNa_V1.5c protein was eluted with buffer A supplemented with 0.06% GDN, 600 nM BTX-B and 500 μg/mL Flag peptide (Bio Basic). Purified protein was subsequently loaded onto a Superose-6 column (GE Healthcare) pre-equilibrated in Buffer A, 0.06% GDN, and 600 nM BTX-B. Peak fractions were concentrated to ~1 mg/ml at 1 ml and mixed with 50 μg LqhIII (Latoxan Laboratory) and additional 12 nmol BTX-B. The mixture was then re-loaded to Superose-6 column pre-equilibrated with buffer containing 25 mM imidazole pH = 6.0, 150 mM NaCl, 0.006% GDN, and 600 nM BTX-B. Finally, peak fractions were concentrated to 60 μl at 5 mg/ml. Protein concentrations were estimated using 1 A₂₈₀ absorbance unit = 1 mg/ml on a NanoDrop spectrophotometer.

Cryo-EM grid preparation and data collection

Three microliters of purified sample were applied to glow-discharged holey grids (QuantiFoil, 300 mesh, R1.2/1.3), and blotted for 2.5–4 s at 100% humidity and 22 °C before being plunge-frozen in liquid ethane cooled by liquid nitrogen using an FEI Mark IV Vitrobot. All data were acquired using a Titan Krios transmission electron microscope operated at 300 kV equipped with a Gatan K3 direct detector, and Gatan Quantum GIF energy filter with a slit width of 20 eV. A total of 7,542 movie stacks were automatically collected using SerialEM⁶³ at a nominal magnification of 105,000x with a pixel size of 0.41275 Å (super-resolution mode). Defocus range was set between −0.5 and −2 μm. Each stack was exposed for 2.04 s with 60 frames with a total dose of 60 e⁻/Å².

Cryo-EM data processing

CryoEM data were processed in CryoSPARC™ software systems version 3.3. The movie stacks were patch motion corrected, Fourier-cropped, and dose weighted, yielding a pixel size of 0.8255 Å. The contrast transfer function parameters for each motion-corrected image were estimated using patch CTF estimation in CryoSPARC. 7,063 micrographs with CTF fits better than 6 Å were used for particle picking. Approximately ~1.69 M particles were automatically picked using Blob Picker. After several rounds of 2D classification, ~558 K good particles were selected and subjected to 3D classification using the low-pass filtered cryoEM map of apo rNa_V1.5c¹¹ as the initial model. Heterogeneous refinement and non-uniform refinement protocols were employed for several iterations of 3D classification. After combining particles from the best 3D classes and removing duplicated particles, 86,763 particles were subjected to global and per-particle CTF refinement followed by non-uniform refinement and map sharpening. Global resolution estimation of the cryo-EM density map was based on the Fourier Shell Correlation criterion of 0.143. Local resolution was estimated in CryoSPARC. A diagram of data processing is presented in Supplementary Fig. 2.

Model building and refinement

The structure of apo rNa_V1.5c α-subunit (PDB code: 6UZ3) was fitted into the cryo-EM density map of rNa_V1.5c in Chimera⁶⁴. The NTD from the structure of human Na_V1.7 (PDB: 7XM9)⁴⁴, which is similar to the NTD structure predicted by AlphaFold for rNa_V1.5, was initially docked into the density map corresponding to the NTD region. The model was manually rebuilt in COOT⁶⁵ and subsequently refined using real-space refinement in Phenix⁶⁶. The model vs. map FSC curve was calculated in Phenix. Statistics for cryo-EM data collection and model refinement are summarized in Supplementary Table 1.

Reporting summary

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

Data availability

The data that support this study are available from the corresponding authors upon request. The cryo-EM map has been deposited in the Electron Microscopy Data Bank (EMDB) under accession code EMD-41071 (the rNa_v1.5c/BTX-B complex). The atomic coordinates have been deposited in the Protein Data Bank (PDB) under accession code 8T6L (the rNa_v1.5c/BTX-B complex).

Additional structures used in this study:

3UZ0; 3UZ3; 6LQA; 7FBS; 7K18; 7W77; 7XM9. The source data underlying Figs. 1d and 4d, and Supplementary Fig. 1 are provided as a Source Data file. Requests for resources and reagents should be directed to and will be fulfilled by the Lead Contact, William A. Catterall (wcatt@uw.edu). All unique/stable reagents generated in this study are available from the Lead Contact with a completed Materials Transfer Agreement. Source data are provided with this paper.

References

Hodgkin, A. L. & Huxley, A. F. A quantitative description of membrane current and its application to conduction and excitation in nerve. J. Physiol. 117, 500–544 (1952).
Article CAS PubMed PubMed Central Google Scholar
Adrian, R. H., Chandler, W. K. & Hodgkin, A. L. Voltage clamp experiments in striated muscle fibres. J. Physiol. 208, 607–644 (1970).
Article CAS PubMed PubMed Central Google Scholar
Catterall, W. A. From ionic currents to molecular mechanisms: The structure and function of voltage-gated sodium channels. Neuron 26, 13–25 (2000).
Article CAS PubMed Google Scholar
Ahern, C. A., Payandeh, J., Bosmans, F. & Chanda, B. The hitchhiker’s guide to the voltage-gated sodium channel galaxy. J. Gen. Physiol. 147, 1–24 (2016).
Article CAS PubMed PubMed Central Google Scholar
Payandeh, J., Scheuer, T., Zheng, N. & Catterall, W. A. The crystal structure of a voltage-gated sodium channel. Nature 475, 353–358 (2011).
Article CAS PubMed PubMed Central Google Scholar
Payandeh, J., Gamal El-Din, T. M., Scheuer, T., Zheng, N. & Catterall, W. A. Crystal structure of a voltage-gated sodium channel in two potentially inactivated states. Nature 486, 135–139 (2012).
Article ADS CAS PubMed PubMed Central Google Scholar
Zhang, X. et al. Crystal structure of an orthologue of the NaChBac voltage-gated sodium channel. Nature 486, 130–134 (2012).
Article ADS CAS PubMed PubMed Central Google Scholar
Pan, X. et al. Molecular basis for pore blockade of human Na⁺ channel Na_V1.2 by the μ-conotoxin KIIIA. Science 363, 1309–1313 (2019).
Article ADS CAS PubMed Google Scholar
Shen, H., Liu, D., Wu, K., Lei, J. & Yan, N. Structures of human Na_V1.7 channel in complex with auxiliary subunits and animal toxins. Science 363, 1303–1308 (2019).
Article ADS CAS PubMed Google Scholar
Pan, X. et al. Structure of the human voltage-gated sodium channel Nav1.4 in complex with beta1. Science 362. https://doi.org/10.1126/science.aau2486 (2018).
Jiang, D. et al. Structure of the cardiac sodium channel. Cell 180, 122–134.e110 (2020).
Article CAS PubMed Google Scholar
Catterall, W. A. Neurotoxins that act on voltage-sensitive sodium channels in excitable membranes. Annu. Rev. Pharmacol. Toxicol. 20, 15–43 (1980).
Article CAS PubMed Google Scholar
Catterall, W. A. et al. Voltage-gated ion channels and gating modifier toxins. Toxicon 49, 124–141 (2007).
Article CAS PubMed Google Scholar
Tokuyama, T., Daly, J. & Witkop, B. The structure of batrachotoxin, a steroidal alkaloid from the Colombian arrow poison frog, Phyllobates aurotaenia, and partial synthesis of batrachotoxin and its analogs and homologs. J. Am. Chem. Soc. 91, 3931–3938 (1969).
Article CAS PubMed Google Scholar
Daly, J. W., Witkop, B., Bommer, P. & Biemann, K. Batrachotoxin. The active principle of the Colombian arrow poison frog, Phyllobates bicolor. J. Am. Chem. Soc. 87, 124–126 (1965).
Article CAS PubMed Google Scholar
Albuquerque, E. X., Daly, J. W. & Witcop, B. Batrachotoxin: chemistry and pharmacology. Science 172, 995–1002 (1971).
Article ADS CAS PubMed Google Scholar
Myers, C. W. & Daly, J. W. Dart-poison frogs. Sci. Am. 248, 120–133 (1983).
Article CAS PubMed Google Scholar
Myers, C. W. & Daly, J. W. Tropical poison frogs. Science 262, 1193 (1993).
Article ADS CAS PubMed Google Scholar
Dumbacher, J. P., Beehler, B. M., Spande, T. F., Garraffo, H. M. & Daly, J. W. Homobatrachotoxin in the genus Pitohui: chemical defense in birds? Science 258, 799–801 (1992).
Article ADS CAS PubMed Google Scholar
Xu, H. et al. Structural basis of Na_V1.7 inhibition by a gating-modifier spider toxin. Cell. https://doi.org/10.1016/j.cell.2018.12.018 (2019).
Wisedchaisri, G. et al. Structural basis for high-affinity trapping of the Na_V1.7 channel in its resting state by a tarantula toxin. Mol. Cell 81, 38–48.e34 (2021).
Article CAS PubMed Google Scholar
Jiang, D. et al. Structural basis for voltage-sensor trapping of the cardiac sodium channel by a deathstalker scorpion toxin. Nat. Commun. 12, 128 (2021).
Article ADS CAS PubMed PubMed Central Google Scholar
Trainer, V. L., Brown, G. B. & Catterall, W. A. Site of covalent labeling by a photoreactive batrachotoxin derivative near transmembrane segment IS6 of the sodium channel a subunit. J. Biol. Chem. 271, 11261–11267 (1996).
Article CAS PubMed Google Scholar
Linford, N. J., Cantrell, A. R., Qu, Y., Scheuer, T. & Catterall, W. A. Interaction of batrachotoxin with the local anesthetic receptor site in transmembrane segment IVS6 of the voltage-gated sodium channel. Proc. Natl Acad. Sci. USA 95, 13947–13952 (1998).
Article ADS CAS PubMed PubMed Central Google Scholar
Wang, S. Y. & Wang, G. K. Point mutations in segment I-S6 render voltage-gated Na+ channels resistant to batrachotoxin. Proc. Natl Acad. Sci. USA 95, 2653–2658 (1998).
Article ADS CAS PubMed PubMed Central Google Scholar
Wang, S. Y. & Wang, G. K. Batrachotoxin-resistant Na+ channels derived from point mutations in transmembrane segment D4-S6. Biophys. J. 76, 3141–3149 (1999).
Article ADS CAS PubMed PubMed Central Google Scholar
Wang, S. Y., Nau, C. & Wang, G. K. Residues in Na(+) channel D3-S6 segment modulate both batrachotoxin and local anesthetic affinities. Biophys. J. 79, 1379–1387 (2000).
Article ADS CAS PubMed PubMed Central Google Scholar
Khodorov, B. I. Batrachotoxin as a tool to study voltage-sensitive sodium channels of excitable membranes. Proc. Biophys. Mol. Biol. 45, 57–148 (1985).
Article CAS Google Scholar
Warnick, J. E., Albuquerque, E. X. & Sansone, F. M. The pharmacology of batrachotoxin. I. Effects on the contractile mechanism and on neuromuscular transmission of mammalian skeletal muscle. J. Pharm. Exp. Ther. 176, 497–510 (1971).
CAS Google Scholar
Albuquerque, E. X., Warnick, J. E. & Sansone, F. M. The pharmacology of batrachotoxin. II. Effect on electrical properties of the mammalian nerve and skeletal muscle membranes. J. Pharm. Exp. Ther. 176, 511–528 (1971).
CAS Google Scholar
Hogan, P. M. & Albuquerque, E. X. The pharmacology of batrachotoxin. 3. Effect on the heart Purkinje fibers. J. Pharm. Exp. Ther. 176, 529–537 (1971).
CAS Google Scholar
Jansson, S. E., Albuquerque, E. X. & Daly, J. The pharmacology of batrachotoxin. VI. Effects on the mammalian motor nerve terminal. J. Pharm. Exp. Ther. 189, 525–537 (1974).
CAS Google Scholar
Catterall, W. A. Activation of the action potential Na+ ionophore of cultured neuroblastoma cells by veratridine and batrachotoxin. J. Biol. Chem. 250, 4053–4059 (1975).
Article CAS PubMed Google Scholar
Catterall, W. A. Activation of the action potential Na+ ionophore by neurotoxins An allosteric model. J. Biol. Chem. 252, 8669–8676 (1977).
Article CAS PubMed Google Scholar
Catterall, W. A., Morrow, C. S., Daly, J. W. & Brown, G. B. Binding of batrachotoxinin A 20-alpha-benzoate to a receptor site associated with sodium channels in synaptic nerve ending particles. J. Biol. Chem. 256, 8922–8927 (1981).
Article CAS PubMed Google Scholar
Logan, M. M., Toma, T., Thomas-Tran, R. & Du Bois, J. Asymmetric synthesis of batrachotoxin: Enantiomeric toxins show functional divergence against NaV. Science 354, 865–869 (2016).
Article ADS CAS PubMed Google Scholar
MacKenzie, T. M. G., Abderemane-Ali, F., Garrison, C. E., Minor, D. L. & Bois, J. D. Differential effects of modified batrachotoxins on voltage-gated sodium channel fast and slow inactivation. Cell Chem. Biol. 29, 615–624.e615 (2022).
Article CAS PubMed Google Scholar
Brown, G. B. et al. 20-alpha-benzoate: A new radioligand for voltage-sensitive sodium channels. Cell. Mol. Neurobiol. 1, 19–40 (1981).
Article PubMed Google Scholar
Brown, G. B. 3H-batrachotoxinin-A benzoate binding to voltage-sensitive sodium channels: inhibition by the channel blockers tetrodotoxin and saxitoxin. J. Neurosci. 6, 2065–2070 (1986).
Article Google Scholar
Jiang, D. et al. Open-state structure and pore gating mechanism of the cardiac sodium channel. Cell 184, 5151–5162.e5111 (2021).
Article CAS PubMed PubMed Central Google Scholar
Jiang, D., Gamal El-Din, T., Zheng, N. & Catterall, W. A. Expression and purification of the cardiac sodium channel Na. Methods Enzymol. 653, 89–101 (2021).
Article CAS PubMed Google Scholar
Jumper, J. et al. Highly accurate protein structure prediction with AlphaFold. Nature 596, 583–589 (2021).
Article ADS CAS PubMed PubMed Central Google Scholar
Huang, G. et al. High-resolution structures of human Na_V1.7 reveal gating modulation through α-π helical transition of S6_IV. Cell Rep. 39, 110735 (2022).
Article CAS PubMed Google Scholar
Zhang, J. et al. Structural basis for Na_V1.7 inhibition by pore blockers. Nat. Struct. Mol. Biol. 29, 1208–1216 (2022).
Article CAS PubMed Google Scholar
Jiang, D., Zhang, J. & Xia, Z. Structural Advances in Voltage-Gated Sodium Channels. Front Pharm. 13, 908867 (2022).
Article CAS Google Scholar
Li, Z. et al. Structural basis for pore blockade of the human cardiac sodium channel Na_V1.5 by antiarrhythmic drug quinidine. Angew. Chem. Int Ed. Engl. 60, 11474–11480 (2021).
Article CAS PubMed Google Scholar
Vedantham, V. & Cannon, S. C. Rapid and slow voltage-dependent conformational changes in segment IVS6 of voltage-gated Na(+) channels. Biophys. J. 78, 2943–2958 (2000).
Article CAS PubMed PubMed Central Google Scholar
Li, H. L., Hadid, D. & Ragsdale, D. S. The batrachotoxin receptor on the voltage-gated sodium channel is guarded by the channel activation gate. Mol. Pharm. 61, 905–912 (2002).
Article CAS Google Scholar
Wang, S. Y., Mitchell, J., Tikhonov, D. B., Zhorov, B. S. & Wang, G. K. How batrachotoxin modifies the sodium channel permeation pathway: computer modeling and site-directed mutagenesis. Mol. Pharm. 69, 788–795 (2006).
Article CAS Google Scholar
Wang, S. Y., Tikhonov, D. B., Zhorov, B. S., Mitchell, J. & Wang, G. K. Serine-401 as a batrachotoxin- and local anesthetic-sensing residue in the human cardiac Na+ channel. Pflug. Arch. 454, 277–287 (2007).
Article CAS Google Scholar
Wang, S. Y., Tikhonov, D. B., Mitchell, J., Zhorov, B. S. & Wang, G. K. Irreversible block of cardiac mutant Na⁺ channels by batrachotoxin. Channels 1, 179–188 (2007).
Article PubMed Google Scholar
Wang, S. Y., Barile, M. & Wang, G. K. Disparate role of Na(+) channel D2-S6 residues in batrachotoxin and local anesthetic action. Mol. Pharm. 59, 1100–1107 (2001).
Article CAS Google Scholar
Lenaeus, M. L., Gamal El-Din, T. M., Tonggu, L., Zheng, N. & Catterall, W. A. Structural basis for inhibition of the cardiac sodium channels by the atypical antiarrhythmic drug ranolazine. Nature Cardiovascular Research https://doi.org/10.1038/s44161-023-00271-5 (2023).
Li, X. et al. Structural basis for modulation of human Na_V1.3 by clinical drug and selective antagonist. Nat. Commun. 13, 1286 (2022).
Article ADS CAS PubMed PubMed Central Google Scholar
Craig, R. A., Garrison, C. E., Nguyen, P. T., Yarov-Yarovoy, V. & Du Bois, J. Veratridine: a janus-faced modulator of voltage-gated sodium ion channels. ACS Chem. Neurosci. 11, 418–426 (2020).
Article CAS PubMed Google Scholar
Favre, I., Moczydlowski, E. & Schild, L. On the structural basis for ionic selectivity among Na⁺, K⁺, and Ca²⁺ in the voltage-gated sodium channel. Biophys. J. 71, 3110–3125 (1996).
Article ADS CAS PubMed PubMed Central Google Scholar
Pliotas, C. & Naismith, J. H. Spectator no more, the role of the membrane in regulating ion channel function. Curr. Opin. Struct. Biol. 45, 59–66 (2017).
Article CAS PubMed Google Scholar
Bohlen, C. J. et al. A bivalent tarantula toxin activates the capsaicin receptor, TRPV1, by targeting the outer pore domain. Cell 141, 834–845 (2010).
Article CAS PubMed PubMed Central Google Scholar
Bae, C. et al. Structural insights into the mechanism of activation of the TRPV1 channel by a membrane-bound tarantula toxin. Elife 5. https://doi.org/10.7554/eLife.11273 (2016).
Montini, G., Booker, J., Sula, A. & Wallace, B. A. Comparisons of voltage-gated sodium channel structures with open and closed gates and implications for state-dependent drug design. Biochem Soc. Trans. 46, 1567–1575 (2018).
Article CAS PubMed Google Scholar
Gamal El-Din, T. M., Lenaeus, M. J., Zheng, N. & Catterall, W. A. Fenestrations control resting-state block of a voltage-gated sodium channel. Proc. Natl Acad. Sci. USA 115, 13111–13116 (2018).
Article ADS CAS PubMed PubMed Central Google Scholar
Shen, H. et al. Structural basis for the modulation of voltage-gated sodium channels by animal toxins. Science 362. https://doi.org/10.1126/science.aau2596 (2018).
Mastronarde, D. N. Automated electron microscope tomography using robust prediction of specimen movements. J. Struct. Biol. 152, 36–51 (2005).
Article PubMed Google Scholar
Pettersen, E. F. et al. UCSF Chimera-a visualization system for exploratory research and analysis. J. Comput Chem. 25, 1605–1612 (2004).
Article CAS PubMed Google Scholar
Emsley, P., Lohkamp, B., Scott, W. G. & Cowtan, K. Features and development of Coot. Acta Crystallogr D. Biol. Crystallogr 66, 486–501 (2010).
Article ADS CAS PubMed PubMed Central Google Scholar
Adams, P. D. et al. PHENIX: a comprehensive Python-based system for macromolecular structure solution. Acta Crystallogr D. Biol. Crystallogr 66, 213–221 (2010).
Article ADS CAS PubMed PubMed Central Google Scholar

Download references

Acknowledgements

We thank Edelmar Navaluna (Department of Pharmacology, University of Washington) for technical support in baculovirus preparations, Dr. Jin Li (Department of Pharmacology, University of Washington) for technical support in molecular biology and for editorial support, and Catherine Garrison and Anne Wampler (Department of Chemistry, Stanford University) for technical support in chemical synthesis. This research was supported by National Institutes of Health Research Grants K08 HL145630 (M.J.L.), R35 NS111573 (W.A.C.), R01 HL112808 (W.A.C. and N.Z.), R01 GM117263 (J.D.), and by the Howard Hughes Medical Institute (N.Z.). A portion of this research was supported by National Institutes of Health Research Grant U24GM129547 and performed at the Pacific Northwest Center for Cryo-EM at Oregon Health & Science University, accessed through EMSL (grid.436923.9), a DOE Office of Science User Facility sponsored by the Office of Biological & Environmental Research.

Author information

Matthew M. Logan
Present address: Vividion Therapeutics, Inc., 5820 Nancy Ridge Dr., San Diego, CA, 92121, USA
Tatsuya Toma
Present address: PRISM BioLab Co., Ltd., 2-26-1 Muraokahigashi, Fujisawa-shi, Kanagawa, 251-8555, Japan
These authors contributed equally: Lige Tonggu, Goragot Wisedchaisri, Tamer M. Gamal El-Din.

Authors and Affiliations

Department of Pharmacology, University of Washington, Seattle, WA, 98195, USA
Lige Tonggu, Goragot Wisedchaisri, Tamer M. Gamal El-Din, Ning Zheng & William A. Catterall
Division of General Internal Medicine, Department of Medicine, University of Washington, Seattle, WA, 98195, USA
Michael J. Lenaeus
Department of Chemistry, Stanford University, Stanford, CA, 94305, USA
Matthew M. Logan, Tatsuya Toma & Justin Du Bois
Howard Hughes Medical Institute, University of Washington, Seattle, WA, 98195, USA
Ning Zheng

Authors

Lige Tonggu
View author publications
You can also search for this author in PubMed Google Scholar
Goragot Wisedchaisri
View author publications
You can also search for this author in PubMed Google Scholar
Tamer M. Gamal El-Din
View author publications
You can also search for this author in PubMed Google Scholar
Michael J. Lenaeus
View author publications
You can also search for this author in PubMed Google Scholar
Matthew M. Logan
View author publications
You can also search for this author in PubMed Google Scholar
Tatsuya Toma
View author publications
You can also search for this author in PubMed Google Scholar
Justin Du Bois
View author publications
You can also search for this author in PubMed Google Scholar
Ning Zheng
View author publications
You can also search for this author in PubMed Google Scholar
William A. Catterall
View author publications
You can also search for this author in PubMed Google Scholar

Contributions

L.T., G.W., T.M.G., and W.A.C. conceived and designed experiments. L.T. purified protein and performed cryo-EM experiments with input from G.W. and M.J.L. L.T., G.W., and M.J.L. performed cryo-EM data analysis, atomic model building, and mutagenesis. T.M.G. performed electrophysiology experiments. M.M.L., T.T., and J.D. synthesized and supplied BTX-B. G.W., T.M.G., and W.A.C. analyzed the data and wrote the manuscript with input from L.T., M.J.L., J.D. and N.Z. All authors contributed to reviewing and revising the paper.

Corresponding authors

Correspondence to Ning Zheng or William A. Catterall.

Ethics declarations

Competing interests

J.D. is a cofounder and holds equity shares in SiteOne Therapeutics, Inc., a start-up company interested in developing subtype-selective modulators of Na_Vs. The other authors declare no competing interests.

Peer review

Peer review information

Nature Communications thanks Tobias Raisch and the other, anonymous, reviewer(s) for their contribution to the peer review of this work. A peer review file is available.

Additional information

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

Supplementary information

Supplementary Information

Peer Review File

Description of Additional Supplementary Files

Supplementary Movie 1

Supplementary Movie 2

Supplementary Data 1

Reporting Summary

Source data

Source Data

Rights and permissions

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

Reprints and permissions

About this article

Cite this article

Tonggu, L., Wisedchaisri, G., Gamal El-Din, T.M. et al. Dual receptor-sites reveal the structural basis for hyperactivation of sodium channels by poison-dart toxin batrachotoxin. Nat Commun 15, 2306 (2024). https://doi.org/10.1038/s41467-024-45958-w

Download citation

Received: 05 October 2023
Accepted: 08 February 2024
Published: 14 March 2024
DOI: https://doi.org/10.1038/s41467-024-45958-w

Comments

By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.