Venomous animals like tarantulas or scorpions inject their prey with toxins to disable them. Some of these toxins work by altering the activity of proteins called ion channels, which are found within membranes in cells. These channels can allow potassium ions and/or other ions to pass through the membrane and have many important roles. For example, ion channels are involved in heart muscle contraction and allow information to travel between brain cells.

Researchers have used some of the toxins as tools to study how ion channel proteins operate. For example, a toxin produced by tarantulas called psalmotoxin binds to a type of ion channel called the acid-sensing ion channels (ASIC). Researchers have previously been able to visualize the three-dimensional structure of psalmotoxin attached to ASIC, which revealed how the toxin binds to and alters the activity of the ion channel.

Another tarantula toxin called guangxitoxin is very similar to psalmotoxin, but it binds to a different type of potassium ion channel. It is thought that guangxitoxin binds to a site on these ion channels that is deep within the membrane, but it is not clear how this works. Here, Gupta, Zamanian, Bae et al. compare some of the structural and chemical properties of the two toxins.

The experiments show that both toxins interact with the membrane to enable them to bind with their target ion channels. However, guangxitoxin moves deeper into the interior of the membrane. Also, Gupta, Zamanian, Bae et al.'s findings suggest that both toxins use a similar surface that curves inwards to clamp onto their target ion channels. This common structural feature will be useful for designing experiments to visualize the three-dimensional structure of guangxitoxin bound to a potassium ion channel.

Protein toxins from the venom of poisonous organisms target a wide variety of ion channel proteins, and they have been powerful tools for studying the physiology and molecular mechanisms of these crucial signaling proteins. For example, scorpion toxins that physically block the pore of K⁺ channels (MacKinnon and Miller, 1988) were initially used to identify pore-forming regions of the channel (MacKinnon and Miller, 1989), and to determine the stoichiometry of subunits (MacKinnon, 1991), inactivation gates (MacKinnon et al., 1993), as well as accessory subunits (Morin and Kobertz, 2008). Protein toxins can also interact with ion channels and allosterically modify the process by which the channel opens in response to the stimulus that activates the channel. For example, both tarantula and scorpion toxins can bind to voltage-sensing domains in voltage-activated ion channels and either retard or facilitate activation in response to changes in voltage (Rogers et al., 1996; Swartz and MacKinnon, 1997a, 1997b; Cestèle et al., 1998; Milescu et al., 2013). This second class of toxins have been valuable tools to explore the mechanism of voltage sensing (Phillips et al., 2005; Alabi et al., 2007; Campos et al., 2007, 2008) and the roles of individual voltage sensors in the mechanisms of activation and inactivation (Campos et al., 2007; Bosmans et al., 2008; Campos et al., 2008). Tarantula toxins have also been discovered that activate transient receptor potential (TRP) channels (Bohlen et al., 2010), and both tarantula toxin and snake toxins have been identified that activate acid-sensing ion channels (ASIC) (Chen et al., 2006; Salinas et al., 2006; Bohlen et al., 2011). For both the TRPV1 channel and ASICs, these toxins were used to stabilize specific conformations for structure determination by cryo-electron microscopy and X-ray diffraction (Baconguis and Gouaux, 2012; Dawson et al., 2012; Baconguis et al., 2014).

Tarantula toxins such as hanatoxin and GxTx-1E are thought to target voltage-activated potassium (Kv) channels by partitioning into the lipid membrane and binding to their voltage-sensing domains to modify how the channel opens in response to changes in membrane voltage (Swartz and MacKinnon, 1997a, 1997b; Li-Smerin and Swartz, 2000, 2001; Lee and MacKinnon, 2004; Wang et al., 2004; Phillips et al., 2005; Alabi et al., 2007; Milescu et al., 2007, 2009, 2013; Schmidt and MacKinnon, 2008; Bosmans et al., 2011a; Tilley et al., 2014) (Figure 1A, right). The strong interaction of these tarantula toxins with lipid membranes is mediated by a prominent amphipathic surface on the toxins (Figure 1F) (Takahashi et al., 2000; Lee et al., 2004, 2010), which facilitates their entry into the membrane (Lee and MacKinnon, 2004; Milescu et al., 2007, 2009; Jung et al., 2010; Mihailescu et al., 2014). Although many aspects of this inhibitory mechanism have been well-studied (Swartz and MacKinnon, 1997a, 1997b; Li-Smerin and Swartz, 1998, 2000; Lee et al., 2003; Wang et al., 2004; Phillips et al., 2005; Alabi et al., 2007; Milescu et al., 2007, 2009; Bosmans et al., 2011a; Milescu et al., 2013; Mihailescu et al., 2014; Tilley et al., 2014), no structures of complexes between tarantula toxins and Kv channels have been solved, limiting our understanding of how tarantula toxins dock onto voltage sensors and influence their conformational dynamics. However, a structurally related tarantula toxin named PcTx1 was recently crystalized in complex with ASIC1a (Baconguis and Gouaux, 2012; Dawson et al., 2012) (Figure 1A, left), revealing that the toxin clamps onto the thumb helix-5 (Figure 1G) and inserts an Arginine finger into the subunit interface to interact with residues involved in proton activation of the channel. PcTx1 modulates the gating properties of ASIC1 channels by enhancing the apparent affinity of these channels for protons, thereby promoting open and desensitized states of the channel (Escoubas et al., 2000, 2003; Chen et al., 2005; Salinas et al., 2006; Baconguis and Gouaux, 2012; Dawson et al., 2012), an allosteric mechanism related to that employed by tarantula toxins targeting Kv channels. However, in contrast to the present model for tarantula toxins binding to Kv channels within the membrane (Figure 1A, right), the structures of PcTx1 bound to ASIC show that the toxin binds to the protein in aqueous solution far from the transmembrane domains of the channel (Figure 1A left). Interestingly, the overall fold of PcTx1 and voltage sensor toxins are quite similar (Figure 1H) and PcTx1 has significant amphipathic character (Figure 1D), similar to that seen in the structure of voltage-sensor toxins (Figure 1E).

Figure 1

Download asset Open asset

The structural similarities of tarantula toxins targeting Kv and ASIC channels, when considered in the context of the unrelated structures of Kv and ASIC channels, as well as the distinct environments in which these toxins bind to their target ion channels, prompted us to undertake comparative studies of the membrane interaction properties and channel binding surfaces of the two classes of tarantula toxins. Does the amphipathic surface of PcTx1 enable the toxin to interact with membranes even though it binds to ASIC in solution? If so, how does the interaction of PcTx1 with membranes compare with that observed for toxins targeting voltage sensors? Do the two classes of structurally related toxins use distinct surfaces for binding to Kv channels within the membrane and to ASIC channels in solution? Our results demonstrate that both classes of tarantula toxins interact with membranes, but that PcTx1 interacts more superficially when compare to toxins targeting voltage sensors. In addition, our results suggest that PcTx1 and the voltage-sensor toxins use related surfaces for binding to helices within their target channels even though these events occur in very different physical environments.

In the present study we focused our investigations on PcTx1, a tarantula toxin for which X-ray structures are available in complex with ASIC1a (Baconguis and Gouaux, 2012; Dawson et al., 2012), and GxTx-1E, a tarantula toxin related to hanatoxin (Swartz and MacKinnon, 1995, 1997a, 1997b; Takahashi et al., 2000; Phillips et al., 2005; Alabi et al., 2007; Bosmans et al., 2008; Milescu et al., 2013; Tilley et al., 2014) that interacts with the S3b-S4 paddle motif within the S1–S4 voltage-sensing domain of a variety of Kv channels (Herrington et al., 2006; Milescu et al., 2009; Schmalhofer et al., 2009; Lee et al., 2010; Milescu et al., 2013). GxTx-1E inhibits opening of the Kv2.1 and engineered Shaker Kv channels with relatively high affinity (Herrington et al., 2006; Milescu et al., 2009; Schmalhofer et al., 2009; Lee et al., 2010; Milescu et al., 2013; Tilley et al., 2014), it can be produced using solid-phase peptide synthesis, efficiently folded in vitro (Lee et al., 2010; Tilley et al., 2014) (see ‘Materials and methods’) and its solution structure has been solved (Lee et al., 2010). Both toxins have a similar spacing of six Cys residues that participate in forming disulfide bonds (Figure 1B), and NMR solution structures reveal they adopt similar folds (Figure 1C,D,H), with backbone root mean square deviation (RMSD) over 20–22 core residues of 1.0–1.2 Å and over all residues of 2.9 Å. The two toxins have five other positions that are well-conserved, but otherwise their primary sequences vary considerably (Figure 1F). PcTx1 contains two solvent-exposed Trp residues, one in loop 1 and one in loop 4, whereas GxTx-1E contains three, two in loop 1 and one in loop 4. The amphipathic character of these toxins can be seen in surface renderings of the toxin structures when coloring residues using the Hessa-Heijne hydrophobicity scale (Hessa et al., 2005), although it is interesting that the clusters of solvent-exposed hydrophobic residues are on distinct surfaces of the two toxins (Figure 1D,E).

Interactions of tarantula toxins with lipid membranes

Previous studies have established that tarantula toxins targeting ion channel voltage sensors interact favorably with lipid membranes (Lee and MacKinnon, 2004; Suchyna et al., 2004; Jung et al., 2005; Phillips et al., 2005; Milescu et al., 2007, 2009; Swartz, 2007; Jung et al., 2010; Mihailescu et al., 2014). We conducted experiments to assess whether PcTx1 partitions into membranes similar to voltage sensor toxins. One common approach to measuring membrane partitioning is Trp fluorescence, where the emission spectra of Trp residues on the toxins are shifted to lower wavelengths (blue shifted) when the toxin moves from an aqueous environment to a more constraining and hydrophobic environment of the lipid membrane (Ladokhin et al., 2000). Pronounced blue shifts in Trp fluorescence are observed with GxTx-1E when lipid vesicles containing a 1:1 mix of POPC and POPG are added to aqueous solutions of the toxin (Figure 2A), consistent with a previous study (Milescu et al., 2009). Plotting of fluorescence intensity of GxTx-1E in the blue shifted region (320 nm) of the emission spectra as a function of lipid concentration and fitting of a partition function (Figure 2C) yields a mole fraction K_x of 4.6 ± 0.8 × 10⁶, indicating favorable interactions of the toxin with membranes. To investigate whether PcTx1 can interact with membranes, we performed similar experiments and observed detectable blue shifts in the emission spectra of the toxin upon the addition of lipid vesicles (Figure 2B). The blue shifts in the case of PcTx1 were very small (maximal shift of ∼ 2 nm), however, they varied as a function of lipid concentration and saturated at high lipid concentrations, suggestive of an interaction between PcTx1 and lipid membranes.

Figure 2

Download asset Open asset

Given the small blue shift observed for PcTx1, we further investigated its interactions with membranes by determining whether lipid vesicles could protect the toxin from quenching by acrylamide, a water soluble quencher of Trp fluorescence. In the case of GxTx-1E in aqueous solution, we observed that the Trp fluorescence of the toxin is quenched by the addition of acrylamide in a concentration-dependent manner (Figure 2E), and fitting of the quenching relationship with a linear function yields a slope (Stern–Volmer constant; K_sv) of 37.7 M⁻¹. Addition of lipid vesicles decreased the K_sv to 9.1 M⁻¹ (Figure 2E), indicating that membranes can shield the Trp residues in GxTx-1E from the soluble quencher. When these experiments were performed with PcTx1, we again observed strong quenching by acrylamide in aqueous solution (K_sv of 29.2 M⁻¹) and robust protection of quenching by the addition of lipid vesicles (K_sv of 11.2 M⁻¹; Figure 2F), confirming that PcTx1 can interact with membranes. Varying the concentration of lipid vesicles in the presence of a fixed concentration of acrylamide also resulted in robust and concentration-dependent protection of acrylamide quenching for both GxTx-1E (Figure 3A,C) and PcTx1 (Figure 3B,D). When the increase in maximal fluorescence intensity was plotted against the available lipid concentration and a partition function fit to the data, a K_dx value of (1.1 ± 0.2) × 10⁷ was obtained for GxTx-1E, in reasonable agreement with the mole fraction partition coefficient for the toxin determined using blue-shifts in Trp fluorescence. In the case of PcTx1, we obtained a value of (4.3 ± 0.6) × 10⁶, demonstrating that PcTx1 interacts strongly with membranes.

Figure 3

Download asset Open asset

The results thus far reveal that both classes of tarantula toxins interact quite favorably with lipid membranes. Although the structure of PcTx1 shows the presence of a clear hydrophobic cluster of residues on one face of the toxin, it is smaller than seen in structures of voltage-sensor toxins (Figure 1D,E). For example, the solvent accessible surface area for the hydrophobic surface of PcTx1 is 380 Ų, which compares with a value of 703 Ų for GxTx-1E. In addition, the hydrophobic surfaces on the two toxins are situated on distinct surfaces (Figure 1D,E). These differences between PcTx1 and GxTx-1E raise the possibility that the two classes of toxin might exhibit distinct depths of penetration into membranes. To estimate the depth to which these toxins penetrate lipid membranes, we compared the extent of quenching of Trp fluorescence by bromine atoms attached at different positions along lipid acyl chains (Ladokhin, 1997, 1999, 2014). In previous studies, the analysis of quenching profiles for lipids brominated at three positions along the acyl chain of POPC (from most superficial to deepest; 6,7-diBr, 9,10-diBr or 11,12-diBr) have shown the strongest quenching for bromines attached near the middle of the lipid tail (9,10-diBr), and analysis of these profiles suggested that Trp residues in hanatoxin, SGTx (a close relative of hanatoxin) and VSTx are positioned 8–9 Å from the center of the bilayer (Phillips et al., 2005; Swartz, 2007; Jung et al., 2010; Mihailescu et al., 2014). When we performed quenching experiments with these three brominated lipids and GxTx-1E, we observed strong quenching that is similar to what has been reported previously for other voltage-sensor toxins (Figure 3E) (Phillips et al., 2005; Swartz, 2007; Jung et al., 2010; Mihailescu et al., 2014), confirming an intimate interaction of the toxin with the hydrophobic core of the membrane. However, we observed only relatively small differences between the three brominated lipids, precluding further quantitative analysis of the profiles to obtain a specific depth of penetration. We suspect that this unique pattern of quenching by brominated lipids for GxTx-1E might be caused by the presence of three Trp residues in the toxin, each of which will likely be in distinct positions when the toxin interacts with membranes (see below). Interestingly, when we undertook similar experiments with PcTx1, we observed only very weak quenching by brominated lipids, again with only relatively small differences between the three lipids (Figure 3F). The measurable quenching observed for PcTx1 confirms that this toxin interacts with membranes, however, the much weaker quenching when compared to voltage-sensor toxins indicates that this toxin interacts considerably more superficially with membranes.

PcTx1 sensitivity is not readily transferable to voltage-sensor toxin binding sites

The X-ray structure of PcTx1 bound to ASIC shows that the toxin interacts primarily with residues on one face of thumb helix-5, with an interface comprised of both hydrophobic and polar residue interactions (Figure 1G) (Baconguis and Gouaux, 2012; Dawson et al., 2012). PcTx1 also contains an Arg finger motif composed of R26, R27 and R28 that is required for activation of ASIC through interactions with channel residues situated in a crevice between subunits. In the case of voltage-sensor toxins such as hanatoxin and GxTx-1E, the S3b helix within the voltage-sensing domains of Kv channels contains the most influential determinants of toxin binding, and these also involve a combination of hydrophobic and polar residue interactions (Swartz and MacKinnon, 1997b; Li-Smerin and Swartz, 2000, 2001; Alabi et al., 2007; Milescu et al., 2009). Given the common helical targets of PcTx1 and voltage-sensor toxins, and the finding that PcTx1 can interact with membranes, we wondered whether it might be possible to transfer sensitivity to PcTx1 into a Kv channel by exchanging the S3b helix with helix-5 from ASIC. We did not attempt to transfer channel residues interacting with the arginine finger motif because these residues and helix-5 are noncontiguous in the ASIC sequence. PcTx1 has nM affinity for ASIC1a (Chen et al., 2006; Dawson et al., 2012), but even at very high concentrations (5 µM) has no effect on the Kv2.1 channel (Figure 4, Figure 4—figure supplement 1B). Previous chimera studies have shown that the voltage-sensor paddle, a helix-turn-helix motif comprised of S3b and S4 helices, can be transplanted between different channels that contain S1–S4 domains without disrupting voltage sensor function (Alabi et al., 2007; Bosmans et al., 2008, 2011b; Kalia and Swartz, 2013), making us optimistic that helix-5 transfer might be possible.

Figure 4 with 1 supplement see all

Download asset Open asset

We chose to make chimeras between ASIC1a and Kv2.1 by transferring helix-5 of ASIC1a into the Kv2.1, a channel that is inhibited by hanatoxin and GxTx-1E (Swartz and MacKinnon, 1997a; Lee et al., 2003, 2010; Phillips et al., 2005; Milescu et al., 2009; Tilley et al., 2014) and has been a successful recipient of transplanted peptide toxin binding sites (Alabi et al., 2007; Bosmans et al., 2008, 2011b; Kalia and Swartz, 2013). We generated a total of five chimeras where nine residues of helix-5 replaced nine residues in the S3b helix in five different frames starting at the N-terminal end of the S3b helix (Figure 4—figure supplement 1A). All five chimeras gave rise to robust voltage-activated K⁺ currents and displayed voltage-activation relations that were shifted to more depolarized voltages when compared to the wild-type Kv2.1 channel (Figure 4—figure supplement 1C–G), consistent with disruption of the interactions between S3b and S4 helices that have been shown to stabilize voltage sensors in an activated state (Xu et al., 2013). Although these relatively radical chimeras were all functional, extracellular application of high concentrations of PcTx1 had no discernible effect on the activity of any of them (Figure 4—figure supplement 1C–G). We also constructed models of PcTx1 bound to the five chimeras by transplanting PcTx1 bound to helix-5 from the complex structure (Baconguis and Gouaux, 2012) into the X-ray structure of the Kv1.2/2.1 paddle chimera (Long et al., 2007) (Figure 4—figure supplement 1H–Q). As expected, several of the models predict steric clashes between PcTx1 and neighboring helices within the S1–S4 domain, whereas in others the toxin could be readily accommodated when bound to helix-5. Even in this latter category of models, however, PcTx1 would need to partition relatively deeply into the membrane in order to bind to helix-5 in the chimeras, a possibility that seems unlikely given the relatively superficial interaction of the toxin we inferred with brominated lipids.

GxTx-1E and PcTx1 use overlapping surfaces for binding to helices

To further compare how PcTx1 and GxTx-1E interact with helices in ASIC and Kv channels, we alanine-scanned GxTx-1E to identify residues that are most critical for interaction with the Kv2.1 channel. We generated Ala mutants for 29 out of 36 residues of GxTx-1E (excluding one native Ala and 6 Cys residues) using solid phase chemical synthesis, and successfully folded all but one in mg quantities (see ‘Materials and methods’). To determine the apparent affinity of toxin mutants for the resting state of Kv2.1, we obtained conductance (G)-voltage (V) relationships from negative holding voltages in the absence and presence of different concentrations of each mutant (Figure 4A,B) and estimated fractional occupancy of the channel from the fractional inhibition for relatively weak depolarizations (−20 to 0 mV) where toxin bound channels do not open, as previously described (Swartz and MacKinnon, 1997a, 1997b; Li-Smerin and Swartz, 2000; Phillips et al., 2005; Milescu et al., 2009). We obtained estimates of the apparent affinity for each mutant, and observed a wide range of different phenotypes, from small changes in affinity to almost 400-fold weakening of toxin affinity for Kv2.1 (Figure 4D; Table 1). When perturbation energies (∆∆G values) for each mutant were mapped onto the NMR structure of GxTx-1E, a well-defined surface of the toxin was identified where mutations having dramatic effects on the apparent affinity of the toxin cluster together (Figure 5A). This surface contains many aromatic residues (F7, W8, W9, Y22 and W28), one aliphatic hydrophobic residue (L30) and several polar residues (K10, K15, S25, K27 and N32). The mutations producing the most dramatic perturbations on this surface were for mutations of aromatic residues, whereas those for the polar residues were relatively modest (Figure 5A; Table 1).

Table 1

Toxin	Kd (nM)	Kdmut/Kdwt	∆∆G (kcal mol−1)
Gxtx-1E(Nle)	224 ± 25	1.0	–
E1A	414 ± 38	1.8	0.36
G2A	297 ± 50	1.3	0.17
E3A	125 ± 62	0.6	−0.35
G5A	13,652 ± 320	60.9	2.44
G6A	379 ± 51	1.7	0.31
F7A	89,033 ± 9413	397.4	3.55
W8A	9641 ± 504	43.0	2.23
W9A	5837 ± 1850	26.1	1.93
K10A	984 ± 90	4.4	0.88
G12A	21,755 ± 1086	97.1	2.71
S13A	545 ± 56	2.4	0.53
G14A	461 ± 89	2.1	0.43
K15A	823 ± 127	3.7	0.77
P16A	22,967 ± 2812	102.5	2.75
P20A	652 ± 73	2.9	0.63
K21A	440 ± 82	2.0	0.40
Y22A	35,337 ± 7464	157.7	3.00
V23A	516 ± 60	2.3	0.50
S25A	1169 ± 64	5.2	0.98
P26A	813 ± 99	3.6	0.76
K27A	763 ± 49	3.4	0.73
W28A	44,298 ± 2471	197.7	3.14
L30A	2474 ± 278	11.0	1.42
N32A	1228 ± 166	5.5	1.01
F33A	491 ± 18	2.2	0.47
P34A	348 ± 60	1.6	0.26
Nle35A	306 ± 19.8	1.4	0.19
P36A	250 ± 22	1.1	0.07

Figure 5 with 3 supplements see all

Download asset Open asset

Because voltage-sensor toxins bind to Kv channels within the membrane, it is possible that some of the mutations we have identified influence the apparent affinity of the toxin by weakening the interaction of the toxin with membranes rather than exclusively weakening the protein–protein interaction. To verify that the surface we identified is involved in binding to Kv2.1, we investigated membrane interactions for each of the three Trp residues (W8, W9 and W28), positions where mutations have some of the largest perturbation energies. Even though the apparent affinities of these mutants for Kv2.1 ranged from 26 to almost 200-fold lower (Figure 4D; Table 1), we obtained K_x values that were within a factor of two for the wild-type toxin (Figure 5—figure supplement 1). We also measured quenching of the three Trp mutants by brominated lipids and found that each exhibited robust quenching (Figure 5—figure supplement 2). It is notable that the blue-shifts observed with membrane partitioning vary for the three Trp mutants, as does the extent of quenching by brominated lipids (Figure 5—figure supplement 2), suggesting that the three Trp residues reside in non-equivalent environments within the membrane. Taken together, these collective data with GxTx-1E mutants strongly support the notion that the active surface identified by mutagenesis (Figure 5A) is involved in binding to the voltage-sensing domain. To compare this surface of GxTx-1E with the binding surface of PcTx1, we calculated the change in solvent accessible surface area (ΔSASA) of PcTx1 upon binding to ASIC1a, and compared it to the binding surface identified in GxTx-1E (Figure 5A–C). The comparison reveals an extensive overlap between the active surfaces of the two toxins responsible for binding to ASIC1a and Kv2.1 channels, indicating that the two classes of toxins use related surfaces for interacting with helices within structurally unrelated ion channel proteins.

The objective of this study was to compare membrane interactions and channel binding surfaces of the voltage-sensor toxin GxTx-1E and the ASIC toxin PcTx1. The recent X-ray structures of complexes between PcTx1 and ASIC1a indicates that the tarantula toxin binds to the channel in aqueous solution (Baconguis and Gouaux, 2012; Dawson et al., 2012), whereas a growing body of work supports that idea that voltage-sensor toxins bind to voltage sensors within the lipid bilayer (Lee and MacKinnon, 2004; Wang et al., 2004; Phillips et al., 2005; Alabi et al., 2007; Milescu et al., 2007, 2009, 2013; Schmidt and MacKinnon, 2008; Bosmans et al., 2011a). The results of our experiments demonstrate that both classes of toxins interact with membranes (Figures 2, 3), a finding that is consistent with the amphipathic character evident in the solution structures of these toxins (Figure 1). However, our results also show that the interaction of PcTx1 with membranes is different than what has been observed for voltage-sensor toxins. Fluorescence spectroscopy results show that Trp residues on PcTx1 exhibit only very small blue shifts when interacting with membranes (Figure 2), and that the Trp fluorescence of the toxin is only weakly quenched by bromine atoms on the acyl chains of lipids (Figure 3). In contrast, voltage-sensor toxins typically exhibit strong blue shifts in Trp fluorescence and this is robustly quenched by brominated lipids (Figures 2, 3). Although these properties have been consistently observed for other voltage-sensor toxins (Jung et al., 2005; Phillips et al., 2005; Milescu et al., 2007, 2009; Swartz, 2007; Jung et al., 2010; Mihailescu et al., 2014), the stark differences observed here for PcTx1 demonstrate that tarantula toxins of similar structure do not necessarily have similar membrane interacting properties. The membrane interactions of voltage-sensor toxins are unique in how far they can penetrate the lipid bilayer, further supporting the idea that this class of toxins interacts with voltage sensors within the lipid bilayer.

In addition to providing a valuable comparison with voltage-sensor toxins, the interaction between PcTx1 and membranes may be interesting in its own right. Although a complex structure of PcTx1 and ASIC shows that the PcTx1 binding site is located about 45 Å above from the lipid bilayer, superficial adsorption of the toxin to the membrane would be predicted to enhance the rate of complex formation due to a reduction in dimensionality of diffusion (Axelrod and Wang, 1994). Interestingly, membrane interactions have also been reported for charybdotoxin (Ben-Tal et al., 1997), a strongly cationic scorpion toxin that blocks the external pore of some potassium channels (Miller, 1995; Banerjee et al., 2013). The interaction of charybdotoxin with membranes is heavily influenced by electrostatic interactions, suggesting that the toxin interacts relatively superficially, similar to what we observed with PcTx1. The superficial nature of the interaction between either PcTx1 or charybdotoxin and lipid membranes could be important because it would allow these toxins to departition rapidly so that it can bind to its receptor above the membrane, or possibly to explore the protein surface as an extension of the membrane surface. The voltage-sensor toxins, in contrast, must bind to helices embedded within the membrane, and thus must venture deeper into the lipid bilayer.

The inhibitor cystine knot (ICK) fold is found in many toxins and is thought to provide a structurally stable scaffold for presentation of surfaces to bind to receptors and ion channel proteins (Pallaghy et al., 1994; Norton and Pallaghy, 1998), and in principle many unique surfaces on these toxins could be used for this purpose. However, our results show that GxTx-1E uses a surface to bind to the S3b helix within voltage sensors that is similar to that which PcTx1 uses to bind to helix-5 in ASIC (Figure 5), even though the former binds within a membrane environment and the latter in aqueous solution. The underlying rationale for using common surfaces is likely related to the fact that both GxTx-1E and PcTx1 bind directly to α-helices. The X-ray structure of PcTx1 in complex with ASIC shows that the toxin clamps onto a solvent-exposed lateral surface of helix-5 (Figure 1) (Baconguis and Gouaux, 2012; Dawson et al., 2012), using a somewhat concave surface of the toxin formed by residues in loop 1 and loop 3, including both hydrophobic (W7, W24, F30, and P35) and polar residues (K8, K25, R26, S29 and T37). The equivalent surface of GxTx-1E is also formed by residues in loop 1 and loop 3 (Figure 5), which contain a similar variety of hydrophobic (F7, W8, W9, Y22, W28 and L30) and polar residues (K10, K15, S25, K27 and N32) where mutations weaken the apparent affinity of the toxin. This surface is also similar to that identified on SGTx1 (Wang et al., 2004; Milescu et al., 2007), a tarantula toxin related to hanatoxin and GxTx-1E that inhibits the Kv2.1 channel. Although the apparent affinity of the most critical mutants on SGTx1 could not be accurately determined because the affinity of the toxin is quite low (Wang et al., 2004), residues having disproportionately large effects on the apparent affinity of the toxin compared to the strength of membrane partitioning cluster together on a surface that is similar to what we identified here for GxTx-1E (Figure 5—figure supplement 3) (Milescu et al., 2007). Given the similar binding faces for α-helices on 3 peptide toxins that share little commonality besides an tri-cystine scaffold, we suggest that this face of the ICK fold forms a shape complement (Jones, 2012) for α-helices that may be exploited by many other peptide ligands.

To further explore the structure of GxTx-1E bound to voltage-sensors, we constructed models of GxTx-1E bound to the X-ray structure of the activated/open Kv1.2/Kv2.1 paddle chimera by superimposing GxTx-1E with PcTx1 bound to ASIC and then transposing the toxin-helix interactions into the paddle chimera. We constructed five models where the toxin-helix interaction occurs in different registers in the S3b helix, similar to what we did in making the helix-5/paddle chimera models, and identified three models where the toxin could be accommodated without steric clashes with the channel (Figure 6, Figure 6—figure supplements 1, 2). Although all three models position critical residues on GxTx-1E where they can interact with the S3b helix, our preferred model (Figure 6; Figure 6—figure supplements 1E, 2E) positions the largest number of critical residues in GxTx-1E where they can interact with residues in the S3b helix. In addition, in this model the toxin interacts with many of the most critical residues within the S3b helix that have been identified using mutagenesis (Figure 6) (Swartz and MacKinnon, 1997b; Li-Smerin and Swartz, 2000, 2001; Milescu et al., 2009), with a placement of the toxin within the membrane that is compatible with the depth of partitioning inferred using spectroscopic and neutron diffraction approaches (Phillips et al., 2005; Jung et al., 2010; Mihailescu et al., 2014). Voltage-sensor toxins like GxTx-1E bind tighter to and stabilize the resting/closed state of Kv channels, however, they can remain bound when the voltage sensors activate and the channel opens (Phillips et al., 2005; Tilley et al., 2014). Thus, although our preferred model of GxTx-1E bound to the activated/open Kv1.2/Kv2.1 paddle chimera represents the toxin bound to a lower affinity activated state of the Kv channel, it nicely illustrates how GxTx-1E could interact with the S3b helix in ways that closely resemble what is seen in the X-ray structure of PcTx1 bound to ASIC. In the future it will be interesting to refine models of the complex by attempting to form bridges between toxin and channel, and ultimately solving structures of these fascinating complexes.

Figure 6 with 2 supplements see all

Download asset Open asset

1. 1. Baconguis I
   2. Gouaux E

   (2012) Crystal structure of acid-sensing ion channel in complex with psalmotoxin 1 at low pH

   Publicly available at RCSB Protein Data Bank (Accession No. 4FZ0).

   http://www.rcsb.org/pdb/explore/explore.do?structureId=4FZ0
2. 1. Long SB
   2. Tao X
   3. Campbell EB
   4. MacKinnon R

   (2007) Shaker family voltage dependent potassium channel (kv1.2-kv2.1 paddle chimera channel) in association with beta subunit

   Publicly available at RCSB Protein Data Bank (Accession No. 2R9R).

   http://www.rcsb.org/pdb/explore/explore.do?structureId=2R9R
3. 1. Saez NJ
   2. Mobli M
   3. Bieri M
   4. Chassagnon IR
   5. Malde AK
   6. Gamsjaeger R
   7. Mark AE
   8. Gooley PR
   9. Rash LD
   10. King GF

   (2011) High-resolution solution structure of the ASIC1a blocker PcTX1

   Publicly available at RCSB Protein Data Bank (Accession No. 2KNI).

   http://www.rcsb.org/pdb/explore/explore.do?structureId=2KNI
4. 1. Lee S
   2. Milescu M
   3. Jung HH
   4. Lee JY
   5. Bae CH
   6. Lee CW
   7. Kim HH
   8. Swartz KJ
   9. Kim JI

   (2010) Solution structure of GxTX-1E

   Publicly available at RCSB Protein Data Bank (Accession No. 2WH9).

   http://www.rcsb.org/pdb/explore/explore.do?structureId=2WH9

1. 1. Alabi AA
   2. Bahamonde MI
   3. Jung HJ
   4. Kim JI
   5. Swartz KJ

   (2007) Portability of paddle motif function and pharmacology in voltage sensors

   Nature 450:370–375.

   https://doi.org/10.1038/nature06266

   • Google Scholar
2. 1. Axelrod D
   2. Wang MD

   (1994) Reduction-of-dimensionality kinetics at reaction-limited cell surface receptors

   Biophysical Journal 66:588–600.

   https://doi.org/10.1016/S0006-3495(94)80834-3

   • Google Scholar
3. 1. Baconguis I
   2. Bohlen CJ
   3. Goehring A
   4. Julius D
   5. Gouaux E

   (2014) X-ray structure of acid-sensing ion channel 1-snake toxin complex reveals open state of a Na(+)-selective channel

   Cell 156:717–729.

   https://doi.org/10.1016/j.cell.2014.01.011

   • Google Scholar
4. 1. Baconguis I
   2. Gouaux E

   (2012) Structural plasticity and dynamic selectivity of acid-sensing ion channel-spider toxin complexes

   Nature 489:400–405.

   https://doi.org/10.1038/nature11375

   • Google Scholar
5. 1. Banerjee A
   2. Lee A
   3. Campbell E
   4. Mackinnon R

   (2013) Structure of a pore-blocking toxin in complex with a eukaryotic voltage-dependent K(+) channel

   eLife 2:e00594.

   https://doi.org/10.7554/eLife.00594

   • Google Scholar
6. 1. Ben-Tal N
   2. Honig B
   3. Miller C
   4. McLaughlin S

   (1997) Electrostatic binding of proteins to membranes. Theoretical predictions and experimental results with charybdotoxin and phospholipid vesicles

   Biophysical Journal 73:1717–1727.

   https://doi.org/10.1016/S0006-3495(97)78203-1

   • Google Scholar
7. 1. Bohlen CJ
   2. Chesler AT
   3. Sharif-Naeini R
   4. Medzihradszky KF
   5. Zhou S
   6. King D
   7. Sánchez EE
   8. Burlingame AL
   9. Basbaum AI
   10. Julius D

   (2011) A heteromeric Texas coral snake toxin targets acid-sensing ion channels to produce pain

   Nature 479:410–414.

   https://doi.org/10.1038/nature10607

   • Google Scholar
8. 1. Bohlen CJ
   2. Priel A
   3. Zhou S
   4. King D
   5. Siemens J
   6. Julius D

   (2010) A bivalent tarantula toxin activates the capsaicin receptor, TRPV1, by targeting the outer pore domain

   Cell 141:834–845.

   https://doi.org/10.1016/j.cell.2010.03.052

   • Google Scholar
9. 1. Bosmans F
   2. Martin-Eauclaire MF
   3. Swartz KJ

   (2008) Deconstructing voltage sensor function and pharmacology in sodium channels

   Nature 456:202–208.

   https://doi.org/10.1038/nature07473

   • Google Scholar
10. 1. Bosmans F
    2. Milescu M
    3. Swartz KJ

    (2011a) Palmitoylation influences the function and pharmacology of sodium channels

    Proceedings of the National Academy of Sciences of USA 108:20213–20218.

    https://doi.org/10.1073/pnas.1108497108

    • Google Scholar
11. 1. Bosmans F
    2. Puopolo M
    3. Martin-Eauclaire MF
    4. Bean BP
    5. Swartz KJ

    (2011b) Functional properties and toxin pharmacology of a dorsal root ganglion sodium channel viewed through its voltage sensors

    The Journal of General Physiology 138:59–72.

    https://doi.org/10.1085/jgp.201110614

    • Google Scholar
12. 1. Bradley P
    2. Misura KM
    3. Baker D

    (2005) Toward high-resolution de novo structure prediction for small proteins

    Science 309:1868–1871.

    https://doi.org/10.1126/science.1113801

    • Google Scholar
13. 1. Campos FV
    2. Chanda B
    3. Beirão PS
    4. Bezanilla F

    (2007) beta-Scorpion toxin modifies gating transitions in all four voltage sensors of the sodium channel

    The Journal of General Physiology 130:257–268.

    https://doi.org/10.1085/jgp.200609719

    • Google Scholar
14. 1. Campos FV
    2. Chanda B
    3. Beirão PS
    4. Bezanilla F

    (2008) Alpha-scorpion toxin impairs a conformational change that leads to fast inactivation of muscle sodium channels

    The Journal of General Physiology 132:251–263.

    https://doi.org/10.1085/jgp.200809995

    • Google Scholar
15. 1. Cestèle S
    2. Qu Y
    3. Rogers JC
    4. Rochat H
    5. Scheuer T
    6. Catterall WA

    (1998) Voltage sensor-trapping: enhanced activation of sodium channels by beta- scorpion toxin bound to the S3-S4 loop in domain II

    Neuron 21:919–931.

    https://doi.org/10.1016/S0896-6273(00)80606-6

    • Google Scholar
16. 1. Chen X
    2. Kalbacher H
    3. Gründer S

    (2005) The tarantula toxin psalmotoxin 1 inhibits acid-sensing ion channel (ASIC) 1a by increasing its apparent H+ affinity

    The Journal of General Physiology 126:71–79.

    https://doi.org/10.1085/jgp.200509303

    • Google Scholar
17. 1. Chen X
    2. Kalbacher H
    3. Gründer S

    (2006) Interaction of acid-sensing ion channel (ASIC) 1 with the tarantula toxin psalmotoxin 1 is state dependent

    The Journal of General Physiology 127:267–276.

    https://doi.org/10.1085/jgp.200509409

    • Google Scholar
18. 1. Conway P
    2. Tyka MD
    3. DiMaio F
    4. Konerding DE
    5. Baker D

    (2014) Relaxation of backbone bond geometry improves protein energy landscape modeling

    Protein Science 23:47–55.

    https://doi.org/10.1002/pro.2389

    • Google Scholar
19. 1. Dawson RJ
    2. Benz J
    3. Stohler P
    4. Tetaz T
    5. Joseph C
    6. Huber S
    7. Schmid G
    8. Hügin D
    9. Pflimlin P
    10. Trube G
    11. Rudolph MG
    12. Hennig M
    13. Ruf A

    (2012) Structure of the acid-sensing ion channel 1 in complex with the gating modifier Psalmotoxin 1

    Nature Communications 3:936.

    https://doi.org/10.1038/ncomms1917

    • Google Scholar
20. 1. Escoubas P
    2. Bernard C
    3. Lambeau G
    4. Lazdunski M
    5. Darbon H

    (2003) Recombinant production and solution structure of PcTx1, the specific peptide inhibitor of ASIC1a proton-gated cation channels

    Protein Science 12:1332–1343.

    https://doi.org/10.1110/ps.0307003

    • Google Scholar
21. 1. Escoubas P
    2. De Weille JR
    3. Lecoq A
    4. Diochot S
    5. Waldmann R
    6. Champigny G
    7. Moinier D
    8. Ménez A
    9. Lazdunski M

    (2000) Isolation of a tarantula toxin specific for a class of proton-gated Na+ channels

    The Journal of Biological Chemistry 275:25116–25121.

    https://doi.org/10.1074/jbc.M003643200

    • Google Scholar
22. 1. Gill SC
    2. von Hippel PH

    (1989) Calculation of protein extinction coefficients from amino acid sequence data

    Analytical Biochemistry 182:319–326.

    https://doi.org/10.1016/0003-2697(89)90602-7

    • Google Scholar
23. 1. Gray JJ
    2. Moughon S
    3. Wang C
    4. Schueler-Furman O
    5. Kuhlman B
    6. Rohl CA
    7. Baker D

    (2003) Protein-protein docking with simultaneous optimization of rigid-body displacement and side-chain conformations

    Journal of Molecular Biology 331:281–299.

    https://doi.org/10.1016/S0022-2836(03)00670-3

    • Google Scholar
24. 1. Herrington J
    2. Zhou YP
    3. Bugianesi RM
    4. Dulski PM
    5. Feng Y
    6. Warren VA
    7. Smith MM
    8. Kohler MG
    9. Garsky VM
    10. Sanchez M
    11. Wagner M
    12. Raphaelli K
    13. Banerjee P
    14. Ahaghotu C
    15. Wunderler D
    16. Priest BT
    17. Mehl JT
    18. Garcia ML
    19. McManus OB
    20. Kaczorowski GJ
    21. Slaughter RS

    (2006) Blockers of the delayed-rectifier potassium current in pancreatic beta-cells enhance glucose-dependent insulin secretion

    Diabetes 55:1034–1042.

    https://doi.org/10.2337/diabetes.55.04.06.db05-0788

    • Google Scholar
25. 1. Hessa T
    2. Kim H
    3. Bihlmaier K
    4. Lundin C
    5. Boekel J
    6. Andersson H
    7. Nilsson I
    8. White SH
    9. von Heijne G

    (2005) Recognition of transmembrane helices by the endoplasmic reticulum translocon

    Nature 433:377–381.

    https://doi.org/10.1038/nature03216

    • Google Scholar
26. 1. Jones S

    (2012) Computational and structural characterisation of protein associations

    Advances in Experimental Medicine and Biology 747:42–54.

    https://doi.org/10.1007/978-1-4614-3229-6_3

    • Google Scholar
27. 1. Jung HH
    2. Jung HJ
    3. Milescu M
    4. Lee CW
    5. Lee S
    6. Lee JY
    7. Eu YJ
    8. Kim HH
    9. Swartz KJ
    10. Kim JI

    (2010) Structure and orientation of a voltage-sensor toxin in lipid membranes

    Biophysical Journal 99:638–646.

    https://doi.org/10.1016/j.bpj.2010.04.061

    • Google Scholar
28. 1. Jung HJ
    2. Lee JY
    3. Kim SH
    4. Eu YJ
    5. Shin SY
    6. Milescu M
    7. Swartz KJ
    8. Kim JI

    (2005) Solution structure and lipid membrane partitioning of VSTx1, an inhibitor of the KvAP potassium channel

    Biochemistry 44:6015–6023.

    https://doi.org/10.1021/bi0477034

    • Google Scholar
29. 1. Kalia J
    2. Swartz KJ

    (2013) Exploring structure-function relationships between TRP and Kv channels

    Scientific Reports 3:1523.

    https://doi.org/10.1038/srep01523

    • Google Scholar
30. 1. Kortemme T
    2. Baker D

    (2002) A simple physical model for binding energy hot spots in protein-protein complexes

    Proceedings of the National Academy of Sciences of USA 99:14116–14121.

    https://doi.org/10.1073/pnas.202485799

    • Google Scholar
31. 1. Kortemme T
    2. Kim DE
    3. Baker D

    (2004) Computational alanine scanning of protein-protein interfaces

    Science's STKE 2004:pl2.

    https://doi.org/10.1126/stke.2192004pl2

    • Google Scholar
32. 1. Ladokhin AS

    (1997) Distribution analysis of depth-dependent fluorescence quenching in membranes: a practical guide

    Methods in Enzymology 278:462–473.

    https://doi.org/10.1016/S0076-6879(97)78024-8

    • Google Scholar
33. 1. Ladokhin AS

    (1999) Analysis of protein and peptide penetration into membranes by depth-dependent fluorescence quenching: theoretical considerations

    Biophysical Journal 76:946–955.

    https://doi.org/10.1016/S0006-3495(99)77258-9

    • Google Scholar
34. 1. Ladokhin AS

    (2014) Measuring membrane penetration with depth-dependent fluorescence quenching: distribution analysis is coming of age

    Biochimica Et Biophysica Acta 1838:2289–2295.

    https://doi.org/10.1016/j.bbamem.2014.02.019

    • Google Scholar
35. 1. Ladokhin AS
    2. Jayasinghe S
    3. White SH

    (2000) How to measure and analyze tryptophan fluorescence in membranes properly, and why bother?

    Analytical Biochemistry 285:235–245.

    https://doi.org/10.1006/abio.2000.4773

    • Google Scholar
36. 1. Lee CW
    2. Kim S
    3. Roh SH
    4. Endoh H
    5. Kodera Y
    6. Maeda T
    7. Kohno T
    8. Wang JM
    9. Swartz KJ
    10. Kim JI

    (2004) Solution structure and functional characterization of SGTx1, a modifier of Kv2.1 channel gating

    Biochemistry 43:890–897.

    https://doi.org/10.1021/bi0353373

    • Google Scholar
37. 1. Lee HC
    2. Wang JM
    3. Swartz KJ

    (2003) Interaction between extracellular Hanatoxin and the resting conformation of the voltage-sensor paddle in Kv channels

    Neuron 40:527–536.

    https://doi.org/10.1016/S0896-6273(03)00636-6

    • Google Scholar
38. 1. Lee S
    2. Milescu M
    3. Jung HH
    4. Lee JY
    5. Bae CH
    6. Lee CW
    7. Kim HH
    8. Swartz KJ
    9. Kim JI

    (2010) Solution structure of GxTX-1E, a high-affinity tarantula toxin interacting with voltage sensors in Kv2.1 potassium channels

    Biochemistry 49:5134–5142.

    https://doi.org/10.1021/bi100246u

    • Google Scholar
39. 1. Lee SY
    2. MacKinnon R

    (2004) A membrane-access mechanism of ion channel inhibition by voltage sensor toxins from spider venom

    Nature 430:232–235.

    https://doi.org/10.1038/nature02632

    • Google Scholar
40. 1. Li-Smerin Y
    2. Swartz KJ

    (1998) Gating modifier toxins reveal a conserved structural motif in voltage- gated Ca2+ and K+ channels

    Proceedings of the National Academy of Sciences of USA 95:8585–8589.

    https://doi.org/10.1073/pnas.95.15.8585

    • Google Scholar
41. 1. Li-Smerin Y
    2. Swartz KJ

    (2000) Localization and molecular determinants of the Hanatoxin receptors on the voltage-sensing domain of a K+ channel

    The Journal of General Physiology 115:673–684.

    https://doi.org/10.1085/jgp.115.6.673

    • Google Scholar
42. 1. Li-Smerin Y
    2. Swartz KJ

    (2001) Helical structure of the COOH terminus of S3 and its contribution to the gating modifier toxin receptor in voltage-gated ion channels

    The Journal of General Physiology 117:205–218.

    https://doi.org/10.1085/jgp.117.3.205

    • Google Scholar
43. 1. Long SB
    2. Tao X
    3. Campbell EB
    4. MacKinnon R

    (2007) Atomic structure of a voltage-dependent K+ channel in a lipid membrane-like environment

    Nature 450:376–382.

    https://doi.org/10.1038/nature06265

    • Google Scholar
44. 1. MacKinnon R

    (1991) Determination of the subunit stoichiometry of a voltage-activated potassium channel

    Nature 350:232–235.

    https://doi.org/10.1038/350232a0

    • Google Scholar
45. 1. MacKinnon R
    2. Aldrich RW
    3. Lee AW

    (1993) Functional stoichiometry of Shaker potassium channel inactivation

    Science 262:757–759.

    https://doi.org/10.1126/science.7694359

    • Google Scholar
46. 1. MacKinnon R
    2. Miller C

    (1988) Mechanism of charybdotoxin block of the high-conductance, Ca2+- activated K+ channel

    The Journal of General Physiology 91:335–349.

    https://doi.org/10.1085/jgp.91.3.335

    • Google Scholar
47. 1. MacKinnon R
    2. Miller C

    (1989) Mutant potassium channels with altered binding of charybdotoxin, a pore- blocking peptide inhibitor

    Science 245:1382–1385.

    https://doi.org/10.1126/science.2476850

    • Google Scholar
48. 1. Mihailescu M
    2. Krepkiy D
    3. Milescu M
    4. Gawrisch K
    5. Swartz KJ
    6. White S

    (2014) Structural interactions of a voltage sensor toxin with lipid membranes

    Proceedings of the National Academy of Sciences of USA 111:E5463–E5470.

    https://doi.org/10.1073/pnas.1415324111

    • Google Scholar
49. 1. Milescu M
    2. Bosmans F
    3. Lee S
    4. Alabi AA
    5. Kim JI
    6. Swartz KJ

    (2009) Interactions between lipids and voltage sensor paddles detected with tarantula toxins

    Nature Structural & Molecular Biology 16:1080–1085.

    https://doi.org/10.1038/nsmb.1679

    • Google Scholar
50. 1. Milescu M
    2. Lee HC
    3. Bae CH
    4. Kim JI
    5. Swartz KJ

    (2013) Opening the shaker K+ channel with hanatoxin

    The Journal of General Physiology 141:203–216.

    https://doi.org/10.1085/jgp.201210914

    • Google Scholar
51. 1. Milescu M
    2. Vobecky J
    3. Roh SH
    4. Kim SH
    5. Jung HJ
    6. Kim JI
    7. Swartz KJ

    (2007) Tarantula toxins interact with voltage sensors within lipid membranes

    The Journal of General Physiology 130:497–511.

    https://doi.org/10.1085/jgp.200709869

    • Google Scholar
52. 1. Miller C

    (1995) The charybdotoxin family of K+ channel-blocking peptides

    Neuron 15:5–10.

    https://doi.org/10.1016/0896-6273(95)90057-8

    • Google Scholar
53. 1. Morin TJ
    2. Kobertz WR

    (2008) Counting membrane-embedded KCNE beta-subunits in functioning K+ channel complexes

    Proceedings of the National Academy of Sciences of USA 105:1478–1482.

    https://doi.org/10.1073/pnas.0710366105

    • Google Scholar
54. 1. Norton RS
    2. Pallaghy PK

    (1998) The cystine knot structure of ion channel toxins and related polypeptides

    Toxicon 36:1573–1583.

    https://doi.org/10.1016/S0041-0101(98)00149-4

    • Google Scholar
55. 1. Pallaghy PK
    2. Nielsen KJ
    3. Craik DJ
    4. Norton RS

    (1994) A common structural motif incorporating a cystine knot and a triple-stranded beta-sheet in toxic and inhibitory polypeptides

    Protein Science 3:1833–1839.

    https://doi.org/10.1002/pro.5560031022

    • Google Scholar
56. 1. Phillips LR
    2. Milescu M
    3. Li-Smerin Y
    4. Mindell JA
    5. Kim JI
    6. Swartz KJ

    (2005) Voltage-sensor activation with a tarantula toxin as cargo

    Nature 436:857–860.

    https://doi.org/10.1038/nature03873

    • Google Scholar
57. 1. Posokhov YO
    2. Gottlieb PA
    3. Ladokhin AS

    (2007) Quenching-enhanced fluorescence titration protocol for accurate determination of free energy of membrane binding

    Analytical Biochemistry 362:290–292.

    https://doi.org/10.1016/j.ab.2006.11.022

    • Google Scholar
58. 1. Raman S
    2. Vernon R
    3. Thompson J
    4. Tyka M
    5. Sadreyev R
    6. Pei J
    7. Kim D
    8. Kellogg E
    9. Dimaio F
    10. Lange O
    11. Kinch L
    12. Sheffler W
    13. Kim BH
    14. Das R
    15. Grishin NV
    16. Baker D

    (2009) Structure prediction for CASP8 with all-atom refinement using Rosetta

    Proteins 77:89–99.

    https://doi.org/10.1002/prot.22540

    • Google Scholar
59. 1. Rogers JC
    2. Qu Y
    3. Tanada TN
    4. Scheuer T
    5. Catterall WA

    (1996) Molecular determinants of high affinity binding of alpha-scorpion toxin and sea anemone toxin in the S3-S4 extracellular loop in domain IV of the Na+ channel alpha subunit

    The Journal of Biological Chemistry 271:15950–15962.

    https://doi.org/10.1074/jbc.271.27.15950

    • Google Scholar
60. 1. Rohl CA
    2. Strauss CE
    3. Misura KM
    4. Baker D

    (2004) Protein structure prediction using Rosetta

    Methods in Enzymology 383:66–93.

    https://doi.org/10.1016/S0076-6879(04)83004-0

    • Google Scholar
61. 1. Salinas M
    2. Rash LD
    3. Baron A
    4. Lambeau G
    5. Escoubas P
    6. Lazdunski M

    (2006) The receptor site of the spider toxin PcTx1 on the proton-gated cation channel ASIC1a

    The Journal of Physiology 570:339–354.

    https://doi.org/10.1113/jphysiol.2005.095810

    • Google Scholar
62. 1. Schmalhofer WA
    2. Ratliff KS
    3. Weinglass A
    4. Kaczorowski GJ
    5. Garcia ML
    6. Herrington J

    (2009) A KV2.1 gating modifier binding assay suitable for high throughput screening

    Channels 3:437–447.

    https://doi.org/10.4161/chan.3.6.10201

    • Google Scholar
63. 1. Schmidt D
    2. MacKinnon R

    (2008) Voltage-dependent K+ channel gating and voltage sensor toxin sensitivity depend on the mechanical state of the lipid membrane

    Proceedings of the National Academy of Sciences of USA 105:19276–19281.

    https://doi.org/10.1073/pnas.0810187105

    • Google Scholar
64. 1. Suchyna TM
    2. Tape SE
    3. Koeppe RE II
    4. Andersen OS
    5. Sachs F
    6. Gottlieb PA

    (2004) Bilayer-dependent inhibition of mechanosensitive channels by neuroactive peptide enantiomers

    Nature 430:235–240.

    https://doi.org/10.1038/nature02743

    • Google Scholar
65. 1. Swartz KJ

    (2007) Tarantula toxins interacting with voltage sensors in potassium channels

    Toxicon 49:213–230.

    https://doi.org/10.1016/j.toxicon.2006.09.024

    • Google Scholar
66. 1. Swartz KJ
    2. MacKinnon R

    (1995) An inhibitor of the Kv2.1 potassium channel isolated from the venom of a Chilean tarantula

    Neuron 15:941–949.

    https://doi.org/10.1016/0896-6273(95)90184-1

    • Google Scholar
67. 1. Swartz KJ
    2. MacKinnon R

    (1997a) Hanatoxin modifies the gating of a voltage-dependent K+ channel through multiple binding sites

    Neuron 18:665–673.

    https://doi.org/10.1016/S0896-6273(00)80306-2

    • Google Scholar
68. 1. Swartz KJ
    2. MacKinnon R

    (1997b) Mapping the receptor site for hanatoxin, a gating modifier of voltage- dependent K+ channels

    Neuron 18:675–682.

    https://doi.org/10.1016/S0896-6273(00)80307-4

    • Google Scholar
69. 1. Takahashi H
    2. Kim JI
    3. Min HJ
    4. Sato K
    5. Swartz KJ
    6. Shimada I

    (2000) Solution structure of hanatoxin1, a gating modifier of voltage-dependent K(+) channels: common surface features of gating modifier toxins

    Journal of Molecular Biology 297:771–780.

    https://doi.org/10.1006/jmbi.2000.3609

    • Google Scholar
70. 1. Tilley DC
    2. Eum KS
    3. Fletcher-Taylor S
    4. Austin DC
    5. Dupré C
    6. Patrón LA
    7. Garcia RL
    8. Lam K
    9. Yarov-Yarovoy V
    10. Cohen BE
    11. Sack JT

    (2014) Chemoselective tarantula toxins report voltage activation of wild-type ion channels in live cells

    Proceedings of the National Academy of Sciences of USA 111:E4789–E4796.

    https://doi.org/10.1073/pnas.1406876111

    • Google Scholar
71. 1. Tyka MD
    2. Keedy DA
    3. André I
    4. Dimaio F
    5. Song Y
    6. Richardson DC
    7. Richardson JS
    8. Baker D

    (2011) Alternate states of proteins revealed by detailed energy landscape mapping

    Journal of Molecular Biology 405:607–618.

    https://doi.org/10.1016/j.jmb.2010.11.008

    • Google Scholar
72. 1. Wang C
    2. Bradley P
    3. Baker D

    (2007) Protein-protein docking with backbone flexibility

    Journal of Molecular Biology 373:503–519.

    https://doi.org/10.1016/j.jmb.2007.07.050

    • Google Scholar
73. 1. Wang JM
    2. Roh SH
    3. Kim S
    4. Lee CW
    5. Kim JI
    6. Swartz KJ

    (2004) Molecular surface of tarantula toxins interacting with voltage sensors in K(v) channels

    The Journal of General Physiology 123:455–467.

    https://doi.org/10.1085/jgp.200309005

    • Google Scholar
74. 1. Xu Y
    2. Ramu Y
    3. Shin HG
    4. Yamakaze J
    5. Lu Z

    (2013) Energetic role of the paddle motif in voltage gating of Shaker K(+) channels

    Nature Structural & Molecular Biology 20:574–581.

    https://doi.org/10.1038/nsmb.2535

    • Google Scholar

Author details

1. Kanchan Gupta

   1. Molecular Physiology and Biophysics Section, Porter Neuroscience Research Center, National Institutes of Health, Bethesda, United States
   2. National Institute of Neurological Disorders and Stroke, National Institutes of Health, Bethesda, United States

   Contribution

   KG, Conception and design, Acquisition of data, Analysis and interpretation of data, Drafting or revising the article, Contributed unpublished essential data or reagents

   Contributed equally with

   Maryam Zamanian and Chanhyung Bae

   Competing interests

   The authors declare that no competing interests exist.

2. Maryam Zamanian

   1. Molecular Physiology and Biophysics Section, Porter Neuroscience Research Center, National Institutes of Health, Bethesda, United States
   2. National Institute of Neurological Disorders and Stroke, National Institutes of Health, Bethesda, United States

   Contribution

   MZ, Conception and design, Acquisition of data, Analysis and interpretation of data, Drafting or revising the article, Contributed unpublished essential data or reagents

   Contributed equally with

   Kanchan Gupta and Chanhyung Bae

   Competing interests

   The authors declare that no competing interests exist.

3. Chanhyung Bae

   1. Molecular Physiology and Biophysics Section, Porter Neuroscience Research Center, National Institutes of Health, Bethesda, United States
   2. National Institute of Neurological Disorders and Stroke, National Institutes of Health, Bethesda, United States
   3. Department of Life Science, Gwangju Institute of Science and Technology, Gwangju, Republic of Korea

   Contribution

   CB, Conception and design, Acquisition of data, Analysis and interpretation of data, Drafting or revising the article, Contributed unpublished essential data or reagents

   Contributed equally with

   Kanchan Gupta and Maryam Zamanian

   Competing interests

   The authors declare that no competing interests exist.

4. Mirela Milescu

   1. Molecular Physiology and Biophysics Section, Porter Neuroscience Research Center, National Institutes of Health, Bethesda, United States
   2. National Institute of Neurological Disorders and Stroke, National Institutes of Health, Bethesda, United States
   3. Biology Division, University of Missouri, Columbia, United States

   Contribution

   MM, Conception and design, Acquisition of data, Analysis and interpretation of data

   Competing interests

   The authors declare that no competing interests exist.

5. Dmitriy Krepkiy

   1. Molecular Physiology and Biophysics Section, Porter Neuroscience Research Center, National Institutes of Health, Bethesda, United States
   2. National Institute of Neurological Disorders and Stroke, National Institutes of Health, Bethesda, United States

   Contribution

   DK, Conception and design, Acquisition of data, Analysis and interpretation of data

   Competing interests

   The authors declare that no competing interests exist.

6. Drew C Tilley

   Department of Physiology and Membrane Biology, University of California, Davis, Davis, United States

   Contribution

   DCT, Conception and design, Acquisition of data, Analysis and interpretation of data, Drafting or revising the article

   Competing interests

   The authors declare that no competing interests exist.

7. Jon T Sack

   Department of Physiology and Membrane Biology, University of California, Davis, Davis, United States

   Contribution

   JTS, Conception and design, Analysis and interpretation of data, Drafting or revising the article

   Competing interests

   The authors declare that no competing interests exist.

8. Vladimir Yarov-Yarovoy

   Department of Physiology and Membrane Biology, University of California, Davis, Davis, United States

   Contribution

   VY-Y, Conception and design, Acquisition of data, Analysis and interpretation of data, Drafting or revising the article

   Competing interests

   The authors declare that no competing interests exist.

9. Jae Il Kim

   Department of Life Science, Gwangju Institute of Science and Technology, Gwangju, Republic of Korea

   Contribution

   JIK, Conception and design, Drafting or revising the article, Contributed unpublished essential data or reagents

   Competing interests

   The authors declare that no competing interests exist.

10. Kenton J Swartz

    1. Molecular Physiology and Biophysics Section, Porter Neuroscience Research Center, National Institutes of Health, Bethesda, United States
    2. National Institute of Neurological Disorders and Stroke, National Institutes of Health, Bethesda, United States

    Contribution

    KJS, Conception and design, Acquisition of data, Analysis and interpretation of data, Drafting or revising the article, Contributed unpublished essential data or reagents

    For correspondence

    swartzk@ninds.nih.gov

    Competing interests

    The authors declare that no competing interests exist.

• 2,660

  views

• 577

  downloads

• 36

  citations

Views, downloads and citations are aggregated across all versions of this paper published by eLife.