ω-Grammotoxin-SIA inhibits voltage-gated Na⁺ channel currents

ω-Grammotoxin-SIA (GrTX-SIA) is a 36-amino acid peptide isolated from the venom of the Chilean rose tarantula (Grammostola rosea, G. spatulata, or G. cala) (Lampe et al., 1993) and is organized according to the ubiquitous inhibitory cystine-knot scaffold (Pallaghy et al., 1994; Takeuchi et al., 2002; Norton and McDonough, 2008; Dongol et al., 2019). As such, GrTX-SIA belongs to a large group of structurally related venom peptides that have been documented to target members of one or more ion channel families (Kalia et al., 2015). Notably, GrTX-SIA was originally identified as an inhibitor of P/Q- and N-type voltage-gated Ca²⁺ (Ca_V) channels because of its modulatory effects on voltage-dependent gating (Lampe et al., 1993; McDonough et al., 1997). Indeed, when tested on a range of animal tissues, GrTX-SIA prevented: (1) depolarization-induced Ca²⁺ influx into brain synaptosomes (Lampe et al., 1993), (2) electrically evoked Ca²⁺ influx in dorsal root ganglion neurons (Piser et al., 1994), (3) norepinephrine or aspartate release from brain slices (Keith et al., 1995), and (4) glutamate release from cultured brain synaptosomes and hippocampal neurons (Piser et al., 1995a, 1995b). Correspondingly, perfusion of GrTX-SIA into rat hippocampus led to a concentration-dependent reduction in glutamate and serotonin release with higher concentrations inducing convulsive-like behavior (Keith et al., 1995). In contrast to P/Q- and N-type, L-type Ca_V channels were reported to be insensitive to GrTX-SIA (Lampe et al., 1993; Piser et al., 1994).

About 5 years after its discovery, the amino acid sequence homology between GrTX-SIA and hanatoxin (Swartz and MacKinnon, 1995) as well as their similar mechanism of action, albeit on different ion channel families, prompted experiments in which GrTX-SIA was found to inhibit voltage-gated K⁺ (K_V) channels as well (Li-Smerin and Swartz, 1998). Moreover, this study also indicated that hanatoxin and GrTX-SIA could likely interact with similar regions in both Ca_V and K_V channel voltage-sensing domains (VSDs) to modify their gating process. During the following years, it became clear that these structural motifs are composed of the S3 and S4 transmembrane helices (Swartz and MacKinnon, 1997), and that they are virtually conserved in Ca_V and K_V channels (Bourinet et al., 2001; Jiang et al., 2003; Swartz, 2008; Bosmans and Swartz, 2010; Catterall, 2011; Bladen et al., 2014; Kuang et al., 2015; Van Theemsche et al., 2020; Yao et al., 2024). Notably, the architecture of VSDs in voltage-gated Na⁺ (Na_V) channels resemble those of Ca_V and K_V channels, suggesting possible interaction sites for these peptides (Jiang et al., 2003; Swartz, 2008; Chakrapani et al., 2008; Ahern et al., 2016; Clairfeuille et al., 2019; Catterall et al., 2020; Noreng et al., 2021; Li et al., 2024; Ngo et al., 2024). Indeed, hanatoxin and related K_V channel toxins can potently affect Na_V channel gating in a manner that is dependent on which VSD is primarily targeted and how this sensor is coupled to various aspects of Na_V channel gating (Bosmans et al., 2008, 2011; Bosmans and Swartz, 2010; Ahern et al., 2016). For example, toxins that target the voltage-sensing region in domains I–III (VSDI–III) influence opening of the channel, whereas toxins typically need to interact specifically with the voltage sensor in domain IV (VSDIV) to impede inactivation (Hodgkin and Huxley, 1952; Sheets et al., 1999; Chanda and Bezanilla, 2002; Chanda et al., 2004; Campos et al., 2008; Ahern et al., 2016).

Based on these observations of promiscuous toxin behavior, we postulated that GrTX-SIA could also bind to one or more Na_V channel VSDs to influence gating. To test our hypothesis, we expressed eight Na_V channel subtypes (Na_V1.1–Na_V1.8) in Xenopus laevis oocytes and measured toxin-induced effects using the two-electrode voltage-clamp technique. To outline a potential working mechanism, we combined molecular docking of GrTX-SIA onto Na_V1.6, the most susceptible subtype, with point mutagenesis and the use of K_V2.1/Na_V1.6 chimeric S3–S4 voltage-sensor constructs.

Lyophilized GrTX-SIA (CAS No.: 152617-90-8) and tetrodotoxin (CAS No.: 18660-81-6) were purchased from Alomone Labs and TOCRIS, respectively. Both toxins were dissolved in a recording solution supplemented with 0.1% BSA (bovine serum albumin, CAS No.: 9048-46-8) and stored at −20°C before use. Analytical grade reagents, enzymes, and antibiotics used in solutions were purchased from Sigma-Aldrich/Merck.

The cDNA sequences of Na_V1.1 (AF225985), Na_V1.2 (NP001035232), Na_V1.3 (AF225987), Na_V1.4 (NP000325), Na_V1.5 (AAI44622), Na_V1.6 (NP055006), Na_V1.7 (NP002968), Na_V1.8 (NP058943), and β1 (NP001028) were obtained from Origene and Genscript and confirmed by automated Sanger sequencing. The chimeric K_V2.1/Na_V1.6 constructs were previously described (Montnach et al., 2022). Mutations in human Na_V1.6 were inserted and verified by Genscript. Plasmids were linearized with an applicable restriction enzyme and RNA transcriptions were performed using the T7 polymerase or SP6 (for Na_V1.8) polymerases (mMessage mMachine kit; Life Technologies). The resulting cRNA of all constructs was dissolved in RNAse-free water and stored at −80°C prior to use. Xenopus frogs, obtained from Nasco and kept at the Department of Biomedical Molecular Biology (UGent), were handled in accordance with the UGent University Animal Care and Use Committee (protocol No.: ECD 23-02), Guide for the Care and Use of Laboratory Animals and the European Directive 86/609/EEC.

Na_V channels were expressed with β1 by microinjecting RNA in a 1:5 M ratio into defolliculated Xenopus oocytes as previously reported (Bosmans et al., 2008). The injected volume per oocyte was 50 nl with the following ion channel RNA concentrations: Na_V1.1 500 ng/μl, Na_V1.2 200 ng/μl, Na_V1.3 200 ng/μl, Na_V1.4 1,000 ng/μl, Na_V1.5 100 ng/μl, Na_V1.6 wild-type/mutants 500 ng/μl, Na_V1.7 1,000 ng/μl, Na_V1.8 1,000 ng/μl, and the K_V2.1/Na_V1.6 chimeras 200 ng/μl. After injection and before use, oocytes were incubated for 1–3 days at 17°C in Barth’s medium (in mM: 96 NaCl, 2 KCl, 5 HEPES (2-[4-(2-hydroxyethyl)piperazin-1-yl]methanesulfonic acid), 1 MgCl₂, 1.8 CaCl₂, and 50 μg/ml gentamycin, pH 7.6 with NaOH).

Following incubation, ionic currents were studied using the two-electrode voltage-clamp recording technique (model OC-725C; Warner Instruments) with a 150-μl recording chamber as previously described (Bosmans et al., 2008). Briefly, this established approach enables control of the transmembrane potential of cells and precise measurement of the ion flow across the cell membrane through voltage-gated ion channels as an electric current. Thus, it becomes possible to examine biophysical gating parameters (i.e., opening/closing) of voltage-gated ion channels in response to changes in the transmembrane potential and investigate the influence of GrTX-SIA on channel function. All data were filtered at 4 kHz (tunable active filter model nine hundred; Frequency Devices Inc.) and digitized at 20 kHz using pClamp 10 software (Molecular Devices). For Na_V channels, the external recording solution used was ND-100 (in mM: 100 NaCl, 5 HEPES, 1 MgCl₂, and 1.8 CaCl₂, pH 7.6 with NaOH). For K_V channel measurements, a 50K solution (in mM: 50 KCl, 50 NaCl, 0.3 CaCl₂, 1 MgCl₂, and 5 HEPES, pH 7.6 with NaOH) was used. Borosilicate microelectrode resistances (produced with a Narishige PC-10 vertical puller) were 0.5–1.0 MΩ when backfilled with 3 M KCl. All experiments were performed at room temperature (∼22°C).

Conductance–voltage relationships, channel availability, recovery from inactivation, entry into slow inactivation, and recovery from slow inactivation were obtained before and after application of specified GrTX-SIA concentrations directly into the 150 μl chamber. Data normalization was achieved by comparing toxin effects to the pretoxin condition within the same oocyte expressing a specific Na_V channel subtype. Typically, a 4-min incubation with depolarizations at 5-s intervals was used to monitor equilibration between GrTX-SIA and the target channel. For state-dependent studies, intervals varied from 1 s to 4 min. Stimulation protocols are shown in Fig. S1 A and described in the text and figure legends. For GrTX-SIA wash-off experiments, oocytes were continuously washed with a gravity-fed perfusion system (flow rate ∼1 ml/min) with ND-100 or 50K. Linear components of capacity and leak currents were identified and subtracted by inhibiting channels with 1 μM tetrodotoxin.

To find possible residues involved in GrTX-SIA binding to Na_V1.6, we used AlphaFold (version 2.3; DeepMind) (Jumper et al., 2021) to predict structures of GrTX-SIA in complex with either full-length human Na_V1.6 or the individual VSDII or IV. For each run, the top hits were further energy-minimized to remove potential geometrical errors. These predictions were performed using computational resources from the Digital Research Alliance of Canada. Predicted structures and their associated predicted local distance difference test (pLDDT) scores were analyzed using PyMol (Schrodinger), whereas positional alignment error (PAE). Scores were evaluated in ChimeraX (University of California, San Francisco) (Meng et al., 2023). Figures of the predicted complexes were made using PyMol (Schrodinger).

Voltage-dependence of slow inactivation was measured after a 10 s-pulse at multiple voltages followed by a depolarizing pulse to peak current voltage, which was normalized to maximum amplitude (I_max), and plotted versus conditioning potential. Recovery from fast inactivation was obtained by applying two depolarizing pulses with varying lengths of time in between to the voltage of peak current for each channel subtype; the second pulse was divided by the first pulse at that time point for normalization purposes. Single-exponential time constants (τ), which correspond to 64% decay (1/e) of the peak current, were calculated and indicated.

Between two sets of data, statistically significant differences (P < 0.05) were assessed using two-tailed Student’s t tests. To compare one time point between more than two groups, one-way ANOVA followed by Tukey’s post hoc analysis was used (P < 0.05). To compare multiple time points between two groups, two-way ANOVA followed by Bonferroni’s post hoc analysis was adopted (P < 0.05). Data are presented as mean ± SEM of n independent oocytes, as stated in the Results section.

Fig. S1 shows the electrophysiology protocols used and Na_V1.6 inactivation. Fig. S2 shows GrTX-SIA and Na_V1.6 slow inactivation. Fig. S3 shows GrTX-SIA interacts with VSDII and VSDIV in Na_V1.6. Table S1 shows the influence of GrTX-SIA on the gating properties of Na_V1.1–1.8. Table S2 shows concentration–response data. Table S3 shows the influence of GrTX-SIA on the gating properties of Na_V1.6 mutants.

We initially determined Na_V channel sensitivity for GrTX-SIA by applying 100 nM toxin to oocytes expressing Na_V1.1–Na_V1.8 and assessing conductance–voltage (G-V) and channel availability relationships. At 100 nM, GrTX-SIA significantly (P < 0.05) reduced currents of all tested Na_V channel subtypes, albeit to a varying degree (Fig. 1 and Table S1). Na_V1.8 and Na_V1.2 were least inhibited (12 ± 4% and 19 ± 5%, respectively), followed by Na_V1.3 (23 ± 9%), Na_V1.5 (27 ± 10%), Na_V1.7 (31 ± 6%), Na_V1.4 (32 ± 13%), and Na_V1.1 (54 ± 5%). The most substantial current reduction was observed with Na_V1.6 (57 ± 9%). Window currents were reduced by 6% (Na_V1.8), 22% (Na_V1.2), 34% (Na_V1.3), 39% (Na_V1.5), 4% (Na_V1.7), 29% (Na_V1.4), 65% (Na_V1.1), and 38% (Na_V1.6). At 100 nM, the V_1/2 deduced by a fit of the G-V relationship to Eq. 2 was significantly shifted to depolarized voltages for Na_V1.1 (∼3 mV), Na_V1.2 (∼2 mV), Na_V1.3 (∼2 mV), Na_V1.6 (∼3 mV), and Na_V1.8 (∼2 mV). The shift in V_1/2 for channel availability differed for Na_V1.1 (∼4 mV), Na_V1.6 (∼6 mV), and Na_V1.8 (∼2 mV). For all tested Na_V channel subtypes, no significant impact of GrTX-SIA was observed on recovery from the inactivation parameter (Fig. 1).

Of all tested channels, Na_V1.6 was most susceptible to GrTX-SIA. We therefore focused on the next set of experiments in this subtype. When incubating Na_V1.6 with 1 µM GrTX-SIA and depolarizing the membrane to −10 mV for 50 ms at 5 s intervals up to 4 min (Fig. 2), currents were reduced by 82 ± 7% (n = 6). At 1-s intervals, current inhibition was 65 ± 7% (n = 4). When the interval was increased to 30 s (n = 6), 1 min (n = 5), 2 min (n = 6), and 4 min (n = 5), inhibition decreased to 58 ± 5%, 47 ± 8%, 39 ± 5%, and 25 ± 3%, indicating that GrTX-SIA binding can occur, at least in part, when the channel resides in the closed state. The V_1/2 for both the G-V and channel availability curve moved significantly when adding 1 µM GrTX-SIA for 4 min (−22 ± 1 mV and −64 ± 2 mV, respectively) when compared with control recordings (−27 ± 1 mV and −59 ± 1 mV, respectively) (Fig. 2 and Table S2). No differences in curve slopes were noted. As with 100 nM (Fig. 1 and Fig. S1 B), 1 µM GrTX-SIA did not affect Na_V1.6 inactivation rate or recovery from fast inactivation (Fig. 2). A fit with the Hill equation of current inhibition measured at −10 mV versus different GrTX-SIA concentrations (Fig. 2 and Table S2) yielded an estimated apparent affinity value of 198 ± 78 nM. The Hill slope of 0.82 ± 0.21 suggested one toxin binding site, or perhaps multiple binding sites that are non- or negatively cooperative. To test whether GrTX-SIA interferes with channel slow inactivation, we held the membrane at a range of potentials for 10 s and monitored the voltage-dependence and entry into the slow inactivated state. We noted that 1 µM GrTX-SIA significantly shifted the V_1/2 to hyperpolarized potentials without affecting the slope of the voltage-dependence curve (control: −25 ± 2 mV versus toxin: −49 ±7 mV). Entry into the slow-inactivated state was unaffected (Fig. S2).

While conducting these experiments, toxin application appeared to induce a biphasic inhibition of Na_V1.6 currents (Fig. 3, A and B). Although we observed substantial variability between oocytes, the first phase happened rather quickly (average rate constant of 1.4 ± 0.1 min) and was responsible for the current reduction of ∼50% observed in the first 2 min after toxin application. At this time point, the V_1/2 of the G-V and channel availability curve was shifted from −26 ± 1 mV to −22 ± 1 mV and from −54 ± 1 mV to −60 ± 2 mV, respectively (Fig. 3 C). Although a clear distinction was challenging to consistently observe, a second, somewhat slower phase (average rate constant of 3.8 ± 0.7 min) followed and increased channel inhibition further at which point the V_1/2 value of the G-V curve no longer differed from control. Total current inhibition achieved after a 4-min application of 1 µM GrTX-SIA was slowly reversible and incomplete after 6 min washing with ND-100 solution (Fig. 3 B). The 4- and 6-min timescales for these experiments were chosen as a compromise between the onset of GrTX-SIA effect saturation, oocyte viability during prolonged trials, and variable start of partial Na_V1.6 current rundown observed in some oocyte batches. Combined, these data provided a motivation to check for multiple GrTX-SIA binding sites within Na_V1.6.

To explore whether GrTX-SIA interacts with one or more VSDs in Na_V1.6, we used K_V2.1/Na_V1.6 chimeric constructs in which known toxin-binding S3–S4 regions from each of the four voltage-sensor domains (VSDI–IV) of Na_V1.6 were transplanted into homotetrameric K_V2.1 channels (Montnach et al., 2022). Notably, a previous study on GrTX-SIA interactions with Na_V1.9 VSDs using this approach suggested potential binding to VSDIII and VSDIV (Bosmans et al., 2011). Additionally, it was demonstrated here that K_V2.1 channels are mildly inhibited by 500 nM GrTX-SIA. This inhibition was abolished when the S3–S4 region of VSDI or VSDII from Na_V1.9 was inserted, indicating that GrTX-SIA can also bind—with low affinity—to K_V2.1 VSDs (Bosmans et al., 2011). When applying 1 µM GrTX-SIA to the K_V2.1/Na_V1.6 chimeric constructs (Fig. 4), we observed that neither the VSDI nor the VSDIII construct was affected. In contrast, the toxin significantly inhibited K⁺ currents mediated by the VSDII and VSDIV chimeras indicating binding to the transferred Na_V1.6 regions. These measurements revealed distinct characteristics of GrTX-SIA interaction with VSDII and VSDIV. First, the VSDII chimera was inhibited by 67 ± 17% whereas the VSDIV construct showed a 92 ± 5% current reduction when measured at V_1/2 of the G-V curve after exposure to 1 µM toxin (Fig. 4). A fit with the Hill equation assuming occupancy of all four binding sites (Bosmans et al., 2008), yielded an apparent affinity value (Kd) of 3.1 µM for VSDII and 1.1 µM for VSDIV. Second, onset of inhibition was slower for VSDII (∼4 min) compared with the VSDIV chimera (<2 min). Third, effect reversibility was much faster for VSDII with complete recovery after 4 min whereas currents of the VSDIV chimera did not yet fully recover after 10 min of washing with 50K solution (Fig. 4 B and Fig. S3). These data suggest a preferential and tight binding to VSDIV with a secondary, lower affinity target in VSDII of Na_V1.6.

Next, we used AlphaFold-guided molecular modeling to predict interactions of GrTX-SIA with VSDII and VSDIV of Na_V1.6. Initial predictions with the full-length human Na_V1.6 sequence yielded nonsensical results, with GrTX-SIA either not forming contacts, or the binding site being in the toxin-inaccessible intracellular region. We therefore focused on predictions with the isolated VSDII and VSDIV regions. For VSDIV, the top hits all show GrTX-SIA bound to the extracellular region. Glu¹⁵⁵¹, a negatively charged residue in transmembrane segment 1 (S1), and Glu¹⁶⁰⁷ in S3 are expected to form salt bridge interactions with Lys²⁴ in GrTX-SIA (Fig. 5). For VSDII predictions, the top five hits were nonsensical, with GrTX-SIA again bound to the intracellular region. For lower-ranked hits, multiple models show GrTX-SIA interactions with the extracellular side of VSDII with residues Glu⁷⁸² and Glu⁸³⁸ being potentially involved in GrTX-SIA binding (Fig. 5). Since the associated PAE scores are modest (∼20) for the VSDIV data, and the potential for ambiguity in these models, we next mutated these residues in Na_V1.6 and experimentally reassessed GrTX-SIA binding.

Given the extensively documented binding of peptide toxins to the S3–S4 region in Na_V channel VSDs (Ahern et al., 2016) and the more pronounced binding of GrTX-SIA to the K_V2.1/Na_V1.6 VSDIV chimera, we first mutated Glu¹⁶⁰⁷ in S3 of VSDIV to a lysine (i.e., p.E1607K, VSDIV^mut). This mutation resulted in a depolarizing shift of the G-V and channel availability curve by ∼7 and ∼17 mV, respectively (Fig. 6 and Table S3). When applying 1 µM GrTX-SIA to this mutant, we noted a substantial reduction of inhibition from ∼80% with the wild-type channel to 32 ± 1% in Na_V1.6^p.E1607K, measured at 0 mV. Initial inhibition of current was rapid, perhaps because the p.E1607K substitution does not completely disrupt the VSDIV toxin binding site (Fig. 4). This was followed by a slower phase with an average rate constant of 6.8 ± 2.6 min (Fig. 6), likely reflecting toxin binding to VSDII (Fig. 4). No toxin-induced shift in the G-V relationship was detected, as in control conditions. Notably, the application of 1 µM GrTX-SIA to Na_V1.6^p.E1607K resulted in a reduced hyperpolarizing shift in channel availability (∼4 mV) compared with the wild-type channel (∼9 mV). Next, we mutated Glu⁷⁸² and Glu⁸³⁸ in Na_V1.6 VSDII to lysines (i.e., p.E782K and p.E838K, VSDII^mut). This double mutation resulted in a depolarizing shift of the G-V and channel availability curve by ∼12 and ∼3 mV, respectively (Fig. 6 and Table S3). When applying 1 µM GrTX-SIA to this mutant, we noted a large reduction of current inhibition from ∼80% with the wild-type channel to 26 ± 5% in Na_V1.6 VSDII^mut, measured at 0 mV. The time course of inhibition displayed one phase with an average rate constant of 2.3 ± 0.2 min (Fig. 6), presumably reflecting VSDIV binding. Application of 1 µM GrTX-SIA induced a significant depolarizing shift of the G-V curve by ∼3 mV whereas channel availability was unaffected (Table S3). It is worth noting here that the biphasic manner in which current inhibition in response to toxin application seemed to occur in the wild-type channel (Fig. 3) is partly reflected in the Na_V1.6 VSDII^mut (fast) and VSDIV^mut (slow) rate constants of binding as well as the onset of inhibition and wash-off times observed in the K_V2.1/Na_V1.6 VSDII and VSDIV chimeric constructs (Fig. 4).

Finally, we added the VSDII substitutions to Na_V1.6 p.E1607K (i.e., VSDII^mut–VSDIV^mut) and checked whether we could reduce GrTX-SIA efficacy even further. Inclusion of the three mutations resulted in a depolarizing shift of the G-V and channel availability curve by ∼18 and ∼17 mV, respectively, compared with the wild-type channel (Fig. 6 and Table S3). Application of 1 µM toxin brought about only a 14 ± 1% current inhibition after 4 min whereas wash-off was complete after ∼2 min. As in the wild-type channel, no toxin-induced shift in the G-V relationship was detected. In addition, the hyperpolarizing shift in channel availability, as seen in control and less pronounced in the Na_V1.6 p.E1607K channel, was no longer present. Combined, these data suggest that GrTX-SIA can inhibit Na_V1.6 currents by interacting with VSDII and VSDIV.

The data reported here show that GrTX-SIA can function as an inhibitory gating modifier of multiple Na_V channel subtypes, albeit with varying efficacy. When expressed in Xenopus oocytes, Na_V1.6 is the most susceptible of all tested subtypes (Na_V1.1–Na_V1.8) with ∼80% of the current inhibited in the presence of 1 μM (Figs. 1 and 2; and Table S1). Markedly, K_V2.1/Na_V1.6 chimeric constructs and molecular docking combined with targeted point mutations in Na_V1.6 suggested that GrTX-SIA achieves current inhibition through interaction with regions within VSDII and VSDIV, with the latter being the higher affinity binding site (Figs. 4 and 6). Among the four Na_V channel VSDs, VSDIV is unique because of its distinct role in inactivating the channel after it has opened (examples include Lacroix et al., 2013; Ahern et al., 2016; Bezanilla, 2018; Armstrong and Hollingworth, 2018; Cowgill and Chanda, 2021). As such, toxins that primarily target this region typically impede the inactivation process. However, the inactivation gate may also close before the channel reaches a conducting state (Armstrong, 2006). Mechanistic insights such as these provided a foundation for a model in which VSDIV movement occurs in two consecutive stages: (1) partial activation associated with channel opening after either VSDI–II or VSDIII activates, and (2) full activation of VSDIV after which the inactivation particle is free to obstruct the pore and prevent further ion conduction. Our results suggest that GrTX-SIA may affect the first VSDIV movement to decrease channel conductance whereas ligands such as the scorpion toxin AaHII and others prevent full VSDIV activation resulting in fast inactivation inhibition (Gilchrist and Bosmans, 2018; Clairfeuille et al., 2019; Jiang et al., 2021). It is worth noting that GrTX-SIA efficacy is further increased by an additional binding site within VSDII (Fig. 4), a region of the channel involved in pore opening (Ahern et al., 2016).

The results of this study combined with previous work on GrTX-SIA and similar toxins underscore the notion that these ligands can recognize and bind to a conserved three-dimensional module in Ca_V channels, Na_V channels, K_V channels, and possibly other ion channel families as well (examples include Siemens et al., 2006; Ono et al., 2011; Bohlen et al., 2010; Jensen et al., 2014; Kalia et al., 2015). Indeed, an extensive body of functional and structural evidence illustrates that voltage-gated ion channels follow a multidomain architecture whereby the Ca²⁺-, Na⁺-, or K⁺-selective pore domain has diverged to confer selectivity for a particular ion whereas the separate VSDs, which are largely conserved in their structure, are shared by all of them (examples include Long et al., 2005, 2007; Payandeh et al., 2011; de Lera Ruiz and Kraus, 2015; Wu et al., 2015; Tang et al., 2016; Pan et al., 2018; Shen et al., 2019; Clairfeuille et al., 2019; Noreng et al., 2021; Catterall, 2023). It is worth mentioning here that GrTX-SIA was part of a large-scale production effort of >600 cysteine-dense peptides that were screened for activity on a subset of ion channels using a commercial electrophysiology assay platform (Correnti et al., 2018). It was noted in the supplementary materials of this publication that GrTX-SIA could inhibit particular Ca_V, Na_V, K_V, and Transient Receptor Potential channel subtypes at 20 μM, a rather high concentration that may reflect the variability associated with high-throughput ion channel electrophysiology approaches. Alternatively, variations in toxin efficacy across different heterologous expression systems might be attributed to toxin–membrane interactions, particularly when testing peptides like GrTX-SIA that have a hydrophobic protrusion (Mihailescu et al., 2014; Takeuchi et al., 2002; Milescu et al., 2009). Combined with data reported here, this entails that although comprehensive gating-modifier peptides such as GrTX-SIA are superb tools to investigate ion channel function, their potential promiscuous nature requires a thorough characterization before robust conclusions can be drawn, in particular when used in complex systems that comprise multiple ion channel families.

Data are available in the article itself and its supplementary materials. Additional data are available from the corresponding author upon reasonable request via email.

Jeanne M. Nerbonne served as editor.

We thank the members of the Bosmans and Van Petegem laboratories for their helpful discussions.

This work was partly funded by a research grant from the Research Foundation—Flanders under G000220N. R.d.C. Collaço is funded by an FWO junior postdoctoral fellowship under application 12Z3922N.

Author contributions: R.d.C. Collaço: Conceptualization, Data curation, Formal analysis, Funding acquisition, Investigation, Methodology, Project administration, Resources, Software, Supervision, Validation, Visualization, Writing—original draft, Writing—review & editing, F. Van Petegem: Investigation, Writing—review & editing, F. Bosmans: Conceptualization, Data curation, Formal analysis, Funding acquisition, Investigation, Methodology, Project administration, Resources, Supervision, Validation, Visualization, Writing—original draft, Writing—review & editing.