Research ArticleNEUROSCIENCE

Target switch of centipede toxins for antagonistic switch

See allHide authors and affiliations

Science Advances  07 Aug 2020:
Vol. 6, no. 32, eabb5734
DOI: 10.1126/sciadv.abb5734


Animal venoms are powerful, highly evolved chemical weapons for defense and predation. While venoms are used mainly to lethally antagonize heterospecifics (individuals of a different species), nonlethal envenomation of conspecifics (individuals of the same species) is occasionally observed. Both the venom and target specifications underlying these two forms of envenomation are still poorly understood. Here, we show a target-switching mechanism in centipede (Scolopendra subspinipes) venom. On the basis of this mechanism, a major toxin component [Ssm Spooky Toxin (SsTx)] in centipede venom inhibits the Shal channel in conspecifics but not in heterospecifics to cause short-term, recoverable, and nonlethal envenomation. This same toxin causes fatal heterospecific envenomation, for example, by switching its target to the Shaker channels in heterospecifics without inhibiting the Shaker channel of conspecific S. subspinipes individuals. These findings suggest that venom components exhibit intricate coevolution with their targets in both heterospecifics and conspecifics, which enables a single toxin to develop graded intraspecific and interspecific antagonistic interactions.


Venomous animals have a specialized venom system as an evolutionary innovation that contributes to their survival and prosperity. Animal venom is a mixture of gene-encoded peptide toxins that facilitate predation (1, 2), defense against predators (35), and intraspecific competition (6). The predation and defensive toxins target receptors in heterospecifics, whereas the toxins used for intraspecific competition likely target receptors in conspecifics. So far, few venoms seem to have a prominent function in intraspecific competitive interactions (7); most identified receptors in venomous animals are insensitive to their venom components (810). In addition, they are thought to inject the same combination of venom components to antagonize both heterospecifics and conspecifics. Because of the high metabolic costs and regional specialization of these highly evolved venoms, we hypothesized that a unique toxin-based mechanism is required for efficient intraspecific competition.

With very few exceptions, centipedes are typically venomous animals with specialized forcipules evolved from the first pair of legs, which may represent the oldest living terrestrial venomous lineage (1113). These venom-derived components enable the use of biochemical rather than physical means for predation, interspecific defense, and intraspecific competition. To the best of our knowledge, a centipede may directly inject venom into individuals of the same species for intraspecific competition as a warning signal without involving other defensive strategies such as aposematism (1416) or chelae (17). Therefore, we used centipede (Scolopendra subspinipes) as the representative model species to understand the molecular basis for intraspecific competition.

In this study, we determined that the Shal channel of S. subspinipes serves as the main target for intraspecific competition or deterrence. Peptide neurotoxins block the specialized Shal channel to induce neuronal hyperexcitation and vascular constriction, which further cause a nonlethal and short-term paralysis within 10 min. In addition, most receptors are resistant to centipede venom using mutations to repel toxin components, thus avoiding lethality among conspecifics.


In this work, we observed that the centipede (S. subspinipes) deterred conspecifics by injecting venom into their body (Fig. 1A and movie S1). To simulate the intraspecific envenomation, we injected 10 μl of centipede crude venom into the third segment of a centipede’s body (one bite from a centipede likely injects 10 to 30 μl of crude venom, ~400 mg of protein product/1 ml of crude venom). The venom injection induced a short-term immobilization in the centipede, which completely recovered within 10 min (Fig. 1B). This observation prompted us to isolate the dorsal unpaired median (DUM) neurons of the centipede (Fig. 1C) and tested the sensitivity of these neurons with the centipede venom. The currents of voltage-gated calcium, sodium, and potassium were observed from centipede DUM neurons, which were blocked by verapamil, Ni+, tetrodotoxin (TTX), and tetraethylammonium (TEA), respectively (Fig. 1, D to F). In the presence of crude venom (1 mg/ml), the voltage-gated calcium and sodium currents were completely intact (Fig. 1, D and E). However, the voltage-gated potassium (KV) currents evoked from DUM neurons were partially inhibited (Fig. 1F), implying that the centipede venom components target a manifested subtype of centipede KV channels.

Fig. 1 Short-term immobilization is related to KV channel subtypes.

(A) Photograph of the S. subspinipes self-envenomation. These centipedes inject venom to each other during intraspecific interaction. Photo credit: Y.W., Northeast Forestry University. (B) Movement distance recorded per minute following injection of 10 μl of crude venom or saline (n = 5 centipedes for each condition). Red arrow, crude venom application. (C) Images of the S. subspinipes and the isolated DUM neuron. (D to F) Whole-cell DUM calcium (D), sodium (E), and potassium (F) currents challenged by crude venom (1 mg/ml), 20 μM verapamil, 25 μM Ni+, 1 μM TTX, and 100 mM TEA, respectively. (G) Phylogenetic tree of centipede KV channel subtypes. (H and I) Voltage-evoked whole-cell currents (H) and conductance-voltage relationships (I) of centipede KV channel subtypes.

Using the major subtypes of KV channels in Drosophila as guides (18), we cloned the full-length complementary DNA (cDNA) sequences of these channel orthologs in centipede from the cDNA library of S. subspinipes DUM neurons (Fig. 1G and table S1). For these KV channels, a homotetramer forms the functional channel complex, with each subunit composes of six putative transmembrane segments (Fig. 1G). Whole-cell recordings showed that centipede’s Shal, Shaker, Shab, Slowpoke, and Eag channels expressed in human embryonic kidney 293 (HEK293) cells exhibited sensitivity to changes in membrane potential (Fig. 1, H and I). We found that these functional KV channels exhibited high expression levels in DUM neurons but were hardly detected in muscles (fig. S1). In addition, we found that Shal and Shab channels were also expressed in the heart tube, suggesting a crucial role of these channels in the centipede’s circulatory system (fig. S1). Therefore, we attempted to figure out which of these KV channel subtypes was inhibited by crude venom, as seen in the DUM neurons of centipede (Fig. 1F). These KV channels were sequentially challenged with crude venom. As shown in Fig. 2A, currents from the centipede’s Shaker, Shab, Slowpoke, and Eag channels were intact in the presence of crude venom (1 mg/ml). The same concentration of crude venom potently inhibited currents from centipede’s Shal channel (Fig. 2, A and B). Compared with a relatively weak inhibition found in DUM KV currents (Fig. 1F), we were aware that the Shal-specific property of crude venom resulted in such a difference in inhibitory effect. To identify the key component that targets the Shal channel, Shal-expressing HEK293 cells were challenged sequentially by purified centipede neurotoxins. SsTx (ssm Spooky Toxin) (1, 19), one of the most abundant neurotoxins in centipede venom (0.01 to 0.02 mg of SsTx/10 μl of crude venom), showed a robust inhibition activity on the centipede’s Shal channel (Fig. 2C and fig. S2). cDNA cloning and BLAST (Basic Local Alignment Search Tool) search from our database (2, 20) revealed a sequence diversity of SsTx family in the venom glands of centipede (fig. S3A). As the conspecific target, we first mapped the conserved region [NAB (N-terminal A and B box) domain, six transmembrane helices, and one pore-forming region] on the amino acid sequence of the centipede’s Shal channel (fig. S3B) (21). Consistently, our computational modeling (based on the structure of Shaker family K+ channel KV1.2-2.1 paddle chimera channel) indicated that each subunit consists of six transmembrane helices and the S5-S6 region forms the central pore of the centipede Shal (fig. S3C) (22). These results allowed us to further investigate the SsTx-Shal interaction and its physiological significance.

Fig. 2 SsTx inhibits centipede Shal channel.

(A) Comparison of venom sensitivity (1 mg/ml) of centipede KV channel subtypes, including Shaker, Shal, Shab, Slowpoke, and Eag channels. n = 5 for each bar. *P < 0.05. Photo credit: Y.W., Northeast Forestry University. (B) Representative whole-cell recoding of Shal currents challenged by crude venom (1 mg/ml). Before application of the crude venom, the cells were perfused with bath solution for 30 s. (C) Representative inhibitory effect of Shal in the presence of 1 μM SsTx (top). Overlapped absorbance peaks of venom components (gray) and purified SsTx (red) by a C18 reversed-phase high-performance liquid chromatography (RP-HPLC) column (bottom). The protein fractions were labeled by circles in red (active) or gray (inactive) when they were subsequently tested on the Shal currents. The number of protein fractions is indicated. The effect of these fractions (1 mg/ml) on the centipede’s Shal channel is shown in the Supplementary Materials.

We found that SsTx inhibited currents from the centipede’s Shal channel through a pore-blocking mechanism since the binding affinity exhibited discernable sensitivity to ion concentration (Fig. 3, A and B). Among residues located in the outer pore region, the negatively charged glutamate on site 351 (centipede Shal number) is distinct from the other species, including those that are potential prey of centipedes (fig. S3D). We therefore focused on site 351 using mutagenesis and found that the charge property on this site largely contributed to the binding affinity of SsTx (Fig. 3, C and D). Negatively charged residues located at site 351 provided SsTx a high affinity, yielding half-maximum inhibitory concentration (IC50) values ranging from 0.1 to 0.3 μM, while noncharged or positively charged residues markedly decreased binding affinity, which exhibited much larger (over 10 μM) IC50 values (Fig. 3, C and D). Furthermore, we used thermodynamic mutant cycle analysis (23, 24) to further investigate the specific interaction between SsTx and centipede’s Shal. The calculated coupling energy between K17 on SsTx and E351 on Shal is 1.9 ± 0.1 kT (Fig. 3, E and F), which is much larger than the 1.5-kT threshold for a direct interaction, suggesting the existence of a specific interaction between these two charged residues. As illustrated in Fig. 3 (G to I), the experimental observations are in agreement with results from molecular docking, indicating that a salt bridge is formed between E351 on Shal and K17 on SsTx. Unlike the mechanism found in SsTx-KCNQ4 (potassium voltage-gated channel subfamily KQT member 4) complex (1), K17 rather than R12 or K13 on SsTx provides the positive charge for the SsTx-Shal interaction. In addition, only centipede Shal responded to SsTx and the crude venom at low concentrations, compared to its orthologs from other species (fig. S4A and table S2). These results suggest that centipede’s Shal channel acts as a function-specialized receptor. Collectively, our mechanistic investigation reveals that such a negatively charged residue at site 351 serves as an anchor for the toxin-channel interaction (fig. S4B and table S3), which enables the centipede’s Shal channel to increase sensitivity of SsTx with more than 10-fold.

Fig. 3 Enhanced SsTx-Shal interaction in centipede.

(A) Representative inhibition of 300 nM SsTx in different K+ concentrations recorded from centipede Shal-expressing HEK293 cells. (B) IC50 of SsTx on Shal channel recorded from pipette solution with different K+ concentrations. *P < 0.05 (C) Concentration-response relationships of centipede Shal and channel mutant (E351A) fitted to a Hill equation (n = 5 per data point). (D) Comparison of IC50 values of wild-type (WT) Shal and its single-point mutants. n.s., not significant. *P < 0.05 (E) Analysis of the pairwise coupling between the SsTx K17 and Shal E351. Data points were fitted to a Hill equation. (F) Comparison of coupling energy between K17 of SsTx and E351 of centipede Shal (n = 3 to 5) with the 1.5-kT threshold for direct interaction indicated by a dashed line. (G) Distribution of E351 mapped to the centipede Shal model from top view. (H) Molecular docking of SsTx with the electrostatic potential distribution shown in color (scale bar) to centipede Shal (shown in gray). Two of the four copies of the toxins are shown in this modeled complex. (I) The salt bridge formed between SsTx K17 and Shal E351 was indicated in SsTx-Shal structural model. *P < 0.05.

The unique SsTx-Shal interaction triggered our great interest to investigate the physiological significance of the Shal channel in centipede. On the basis of the quantitative polymerase chain reaction (PCR) results (fig. S1), we used the Shal antibody to further confirm protein expression in the DUM neurons, heart tube, intestine, trachea, and muscle. Consistently, Shal protein is abundant in the DUM neurons and heart tube of centipede (Fig. 4A). Furthermore, the Shal proteins were detected in the intestine but not in the trachea and muscles (Fig. 4A and fig. S1). We next used immunohistochemical analysis to determine the localization of Shal channel in the DUM neurons and heart tube. Shal channel proteins are located in the cell membrane of DUM neurons and vascular cells (Fig. 4B). In the presence of 1 μM SsTx, inhibition of Shal channel caused enhanced excitation of DUM neurons with a much higher spiking frequency of 6.4 ± 0.8 Hz, compared with the basal level frequency of 2.2 ± 0.5 Hz in bath solution (Fig. 4, C and D). In addition, we found that the constriction force of isolated centipede heart tube is related to the function of Shal channel. As illustrated in Fig. 4E, similar to phenylephrine, 1 μM SsTx induced a strong vasoconstriction on the acutely isolated heart tube of centipede. These results suggest that Shal is a neuronal and vascular channel in S. subspinipes, which participates in the toxin-channel interaction for both DUM superexcitation and vasoconstriction. Therefore, the short-term recovery from immobilization (Fig. 1B) is likely caused by a reduced SsTx concentration surrounding the bite site after venom injection, which leads to the dissociation of SsTx at low concentrations (Fig. 3C).

Fig. 4 Target switching mechanism of centipede toxins.

(A) Expression level of centipede Shal protein in the intestine, trachea, DUM neuron, heart tube, and muscle. (B) Representative photomicrographs of DUM or heart tube sections stained with hematoxylin and eosin (left) or Shal channel antibody (right). Scale bars, 200 μm (for the DUM) and 50 μm (for the heart tube). (C) Representative response of DUM neuron after application of 40-pA depolarizing current injected into the cell when challenged with 1 μM SsTx. (D) Comparison of spiking frequency in the presence of bath solution (n = 4) and 1 μM SsTx (n = 5). *P < 0.05 (E) Representative vascular contractility of centipede heart tube contraction when challenged with 1 μM SsTx and 10 μM phenylephrine (PE). (F) Representative whole-cell recordings of centipede Shaker (left) and single-point channel mutant (right) in the presence of crude venom (1 mg/ml) or 1 μM SsTx. (G) SsTx concentration-response relationships of centipede Shaker and channel mutants. Data points were fitted to a Hill equation. (H) Site 399 (centipede Shaker number) is highlighted in the amino acid sequence alignment of species-specific Shaker channels. (I) Comparison of inhibitory effects of 10 μM SsTx on centipede Shaker and its orthologs. *P < 0.05 (J) Schematic diagram summarizing the target-switching mechanism in centipede survival strategy of both intraspecific and interspecific interaction. Target specifications (Shal, Shaker, TRPV1, and KCNQ channels) underlying the envenomation of conspecifics and heterospecifics are shown. Representative point mutations of species-specific Shal and Shaker channels are differentiated in conspecifics and heterospecifics. *P < 0.05.

Consistent with our observations, following crude venom injection, we found that application of SsTx induced paralysis in centipede. Furthermore, we observed that SsTx injection caused short-term immobilization (movie S2) and recoverable toxic effect, which was similar to crude venom injection (Fig. 1B). This type of nonlethal and short-term warning signal seems to be suitable for intraspecific deterrence or competition purposes, given that intraspecific aggression is likely frequent but not lethal. Consistently, Shal is known to be expressed in insect crawling motoneurons for mediating A-type–like potassium currents (25, 26). Together, these results again support the central role of the centipede’s Shal channel in antagonistic interaction among conspecifics but not in predation and predator deterrence.

Although centipede’s Shal channel evolved to serve as the main target for intraspecific interaction or deterrence, we subscribe to the idea that most receptors of venomous animals are resistant to their own neurotoxins, as preying on conspecifics offers little evolutionally benefit for the survival of the species. The crude venom is invalid on most of centipede receptors (Fig. 1, D to F). We provided an example in this study to suggest an unusual tolerance mechanism, which allows the other centipede’s receptors to avoid toxin binding. Centipede’s Shaker channel exhibits resistance to SsTx and crude venom (Fig. 4F). We observed that the arginine residue at site 399 bestows centipede Shaker with resistance to the toxin. A single mutation replacing R399 with a negatively charged or noncharged residue made centipede’s Shaker sensitive to SsTx (Fig. 4G). Negatively charged and noncharged substitutions exhibited similar sensitivity to SsTx, suggesting that R399 in centipede’s Shaker channel provides a repulsion force. Without such a positively charged residue located at the homologous site, Shaker channels of those potential preys were all sensitive to SsTx (Fig. 4, H and I). On the basis of these findings, we reason that centipede’s Shaker channel is one of the receptors in conspecifics with good tolerance to neurotoxins. Besides the Shaker channel, it is understandable that most receptors of centipede evolved to be resistant against their own venom components, which is necessary to narrow down the range of receptors involved in intraspecific interaction.


Venomous animals use a venom system to carry out predation, defense, and intraspecific competition, implying that neurotoxins are equipped to achieve toxin-receptor interactions and to deal with different species-species interactions. Centipede is an excellent model to understand the molecular basis of venom use for survival strategies, due to its diverse neurotoxins and sophisticated interspecific and intraspecific interactions during the coevolutionary process. In combination with our previous studies (1, 2, 5, 20, 2729), we are able to partially draw the mechanistic insights for several important survival strategies of centipede (Fig. 4J). Specifically, (i) centipede’s Shal channel evolved to establish a unique toxin-target interaction, which serves as the mechanism for eliciting nonlethal warning signal to conspecifics; (ii) SsTx-induced immobilization is likely due to the inhibition of centipede’s Shal channel in motoneurons, which leads to an enhanced excitation in these neurons; (iii) most of physiological receptors, such as Shaker, TRPV1 (transient receptor potential vanilloid 1), and KCNQ channels of centipede, are expected to coevolve with the venom system to resist against their own venom components; and (iv) multiple receptors of heterospecifics including Shaker, KCNQ, and TRPV1 channels are efficiently modulated by centipede toxins, which enables centipedes to carry out either predatory envenomation or defensive strategy against predators.

More generally, to improve the biological efficiency of envenomation, certain toxins have been found to exhibit bifunctional or even multifunctional activities (30, 31). In the present case, SsTx targets several receptors in conspecifics or heterospecifics including Shal, Shaker, KV1.3 (19), and KCNQ (1) to exert different ecological functions. Although centipedes and other species have orthologous receptor families, such as Shal and Shaker, SsTx antagonistically targets these receptors to attribute for either conspecific or heterospecific interplays. These ecological functions and antagonistic modes appear to be elegantly tuned by the coevolution between these receptors and venom components, providing the possibility for switched targets of centipede toxin as reported here. Venomous individuals take full advantage of optimized chemical weapons to interact with other participants in their niches by identifying self and non-self. Therefore, as a consequence of coevolution between venom components and receptors, our study reveals that the target-switching mechanism is used against different opponents by venomous animals to exhibit a flexible and low-cost antagonistic strategy.


Study approval

All animal experiments were performed in accordance with recommendations in the Guide for the Care and Use of Laboratory Animals of Kunming Institute of Zoology, Chinese Academy of Sciences. Experimental protocols using animals in this study were approved by the Institutional Animal Care And Use Committees at Kunming Institute of Zoology, Chinese Academy of Sciences (approval ID: SMKX2017008).

cDNA library synthesis and full-length cDNA cloning

Total RNA was extracted from the DUM neurons or venom gland of centipedes (S. subspinipes) (both sexes, n = 6) using TRIzol (Invitrogen). cDNA library was constructed using a SMART PCR cDNA synthesis kit (Clontech). The 3′ SMART CDS Primer II A (5′-A AGCAGTGGTATCAACGCAGAGTACT(30)N-1 N-3′, where N = A, C, G, or T; N-1 = A, G, or C) available in the kit was used to synthesize the first strand. The second strand was synthesized using Advantage Polymerase (Clontech) by 5′ PCR Primer II A (5′-AAGCAGTGGTATCAACGCAGAGT-3′). To clone the full-length cDNA sequence of the KV channels from the centipede DUM cDNA library, RACE (rapid amplification of cDNA ends) primers were designed according to the conserved amino acid sequences. PCR was performed using Advantage Polymerase (Clontech), and the PCR products were cloned into pGEM-T Easy Vector (Promega, Madison, WI). DNA sequencing was performed on an ABI PRISM 377 DNA sequencer (Applied Biosystems).

Toxins purification

Crude venoms were collected manually by stimulating the venom glands in the first pair forceps of centipedes (both sexes, n = 200) using a 3-V alternating current. One milliliter of venom was mixed with 10 μl of proteinase inhibitor cocktail. Following collection, venoms were stored at −80°C until further use. SsTx was purified from the crude venom using a combination of a Sephadex G-50 gel filtration column, fast protein liquid chromatography, and reversed-phase high-performance liquid chromatography (RP-HPLC). The purity and molecular weight of toxins were analyzed using a matrix-assisted laser desorption/ionization–time-of-flight mass spectrometry (MALDI-TOF MS; Bruker Daltonik GmBH). Primary amino acid sequences were obtained by automated Edman degradation using a Shimadzu protein sequencer (PPSQ-31A, Shimadzu).

Molecular biology

The cDNA sequences of centipede Shaker, Shab, Shaw, Shal, Slowpoke, and Eag channels were synthesized by Sangon Biotech (Shanghai, China) and subcloned into the pcDNA3.1 vector. All channel point mutants were constructed using the QuikChange II XL Site-Directed Mutagenesis Kit (Agilent Technologies) following the manufacturer’s instructions. These channel mutants were sequenced to confirm that appropriate constructs were made.

Cell preparation and transfection

HEK293 cells were purchased from Kunming Cell Bank, Kunming Institute of Zoology, Chinese Academy of Sciences (CRL-3216, American Type Culture Collection). Cells were cultured in Dulbecco’s modified Eagle’s medium supplemented with 10% fetal bovine serum and 1% penicillin/streptomycin and incubated at 37°C in 5% CO2. Cells were transiently transfected with DNA mixture [channel constructs and a green fluorescent protein (GFP) reporter plasmid] using Lipofectamine 3000 transfection reagent (Life Technologies, USA) following the manufacturer’s instructions. Cells with GFP fluorescence were selected for patch-clamp recordings 1 day after the transfection.

Electrophysiological recordings of centipede KV currents

All electrophysiological experiments in this study were performed at room temperature using an EPC 10 amplifier with the Patchmaster software (HEKA Elektronik, Lambrecht, Germany). For whole-cell recordings, voltage errors were minimized using 80% series resistance compensation, and the capacitance artifact was canceled using the computer-controlled circuitry of the patch-clamp amplifier. Current was sampled at 10 kHz and filtered at 2.9 kHz. Patch pipettes were pulled from borosilicate glass and fire-polished to a resistance of 3 to 6 megohms. To evoke centipede KV currents (Shal, Shaker, Eag, and Shab for example), a holding potential of −80 mV was used, from which a testing pulse to 0 mV was applied. To evoke centipede Slowpoke currents, the testing pulse of over 220 mV was applied. The conductance-voltage curve was determined from currents in response to a series of voltage steps starting from a deep hyperpolarized voltage. The concentration-response relationship determined using a solution exchanger (RSC-200, Biological Science Instruments) with eight separate tubes to deliver different concentrations of SsTx. The tube number sent by the solution exchanger was fed into an analog input port of the EPC 10 patch-clamp amplifier, and the current was recorded simultaneously. The stable current amplitude at different concentrations was recorded. To record the centipede KV channels, the pipette solution contained 160 mM KCl, 5 mM MgCl2, and 5 mM Hepes; pH was adjusted to 7.2 with KOH. Bath solution contained 160 mM NaCl, 2.5 mM KCl, 2 mM CaCl2, 1 mM MgCl2, 10 mM glucose, and 10 mM Hepes; pH was adjusted to 7.2 with NaOH.

Isolation and patch-clamp recording of centipede DUM neurons

The S. subspinipes DUM neurons were acutely dissociated and maintained in a short-term primary culture according to the procedures below. Briefly, the ganglia of the centipede were dissected and placed in sterile Ca2+-free Hanks’ balanced salt solution (HBSS; Sigma-Aldrich, USA). The ganglia were then washed twice and incubated for 20 min in HBSS containing type I collagenase (0.5 mg/ml) and trypsin (0.5 mg/ml) at 28°C. The cells were centrifuged at 900 rpm for 5 min and subsequently washed three times with SF-900 medium (Gibco, Life Technologies, USA) containing 5% fetal bovine serum and then dispersed mechanically by pipetting. The single-cell suspension was then allowed to adhere to glass coverslips for 3 hours at 28°C. Large tear-shaped DUM neurons were selected for patch clamp recordings.

To record sodium currents, pipette solution contained 20 mM NaCl, 135 mM CsF, 1 mM MgCl2, 5 mM EGTA, 10 mM glucose, and 10 mM Hepes; pH was adjusted to 7.4 with CsOH. The external solution contained 130 mM NaCl, 5 mM CsCl, 20 mM TEA-Cl, 1.8 mM CaCl2, 5 mM 4-aminopyridine, 0.01 mM verapamil-HCl, 0.1 mM NiCl2, 1 mM CdCl2, and 10 mM Hepes; pH was adjusted to 7.4 with NaOH. A holding potential of −80 mV was used, from which a testing pulse of −10 mV was applied.

To record potassium currents, pipette solution contained 120 mM KF, 20 N-methyl-d-glucamine, 10 mM Hepes, 11 mM EGTA, 2 mM adenosine 5′-triphosphate (ATP)–Mg, and 0.5 mM Li2-guanosine 5′-triphosphate; pH was adjusted to 7.4 with KOH. The bath solution contained 130 mM choline chloride, 5 mM KOH, 10 mM Hepes, 12 mM glucose, 2 mM MgCl2, and 2 mM CaCl2; pH was adjusted to 7.4 with KOH. A holding potential of −80 mV was used, from which a testing pulse of 10 mV was applied.

To record calcium currents, pipette solution contained 140 mM CsCl, 2 mM MgC12, 2 mM ATP-Mg, 10 mM Hepes, and 1.5 mM EGTA; pH was adjusted to 7.4 with CsOH. The bath solution contained 150 mM choline chloride, 2 mM MgC12, 10 mM BaC12, 10 mM Hepes, 25 mM TEA, and 0.15 mM TTX; pH was adjusted to 7.4 with KOH. A holding potential of −80 mV was used, from which a testing pulse of 0 mV was applied.

To record action potentials from isolated DUM neurons, patch pipettes were filled with internal solution containing 160 mM potassium aspartate, 10 mM KF, 1 mM ATP-Mg, 0.5 mM CaCl2, 15 mM NaCl, 1 mM MgCl2, 10 mM EGTA, and 10 mM Hepes; pH was adjusted to 7.2 with KOH. The bath solution contained 200 mM NaCl, 3.1 mM KCl, 5 mM CaCl2, 4 mM MgCl2, and 10 mM Hepes; pH was adjusted to 7.2 with NaOH. Action potentials were recorded under current clamp conditions and evoked by injecting 40-pA depolarizing current of 200-ms duration into the DUM neuron cell bodies. Between recordings (interval, 10 s), the cells were held at −80 mV.

Peptides synthesis, refolding, and purification

An automatic peptide synthesizer (PerSeptive Biosystems) with a 9-fluorenyl methoxycarbonyl/tert-butyl strategy and HOBt/TBTU/NMM (1-hydroxybenzotriazole/O-(benzotriazol-1-yl)-N,N,N’,N’-tetramethyluronium tetrafluoroborate/N-methylmorpholine) coupling method to synthesize linear SsTx was used. Crude peptidic products were purified by RP-HPLC. Once the purity of linear SsTx was determined to be >95% by MALDI-TOF MS and HPLC, the peak was pooled and lyophilized. The linear reduced peptide was dissolved in 0.1 M tris-HCl and 0.1 M sodium chloride buffer (pH 8.0) at a final concentration of 30 μM glutathione containing 5 mM reduced glutathione and 0.5 mM oxidized glutathione at 28°C for 24 hours. Oxidized and folded peptides were fractionated by analytical C18 RP-HPLC using a linear acetonitrile gradient, and the purity was detected by MALDI-TOF MS.

Quantitative real-time PCR analysis

Total RNA was isolated from centipede (S. subspinipes) ganglion using TRIzol reagent (Invitrogen). cDNA was reverse-transcribed from 1 μg of RNA using Moloney murine leukemia virus reverse transcriptase (Promega). Quantitative real-time PCR was used to explore potassium channel (Shaker, Shab, Shal, Slowpoke, and Eag,) tissue distribution in centipede, and it was performed on the Bio-Rad CFX-96 Touch Real-Time Detection System.

Construction of centipede Shal model

The Shal channel model was constructed by membrane-symmetry-loop modeling using the Rosetta molecular modeling suite version 2016.20 (32). The cryo–electron microscopy structure of a mammalian voltage-dependent Shaker family K+ channel Kv1.2-2.1 paddle chimera channel [Protein Data Bank (PDB) ID: 6EBK] was used as the template (22), the S5-S6 linker was modeled de novo with the kinematic closure (KIC) loop modeling protocol (33, 34). Briefly, around 10,000 models were generated each round; the top 10 lowest-energy models were selected as the inputs for next round of loop modeling. After several rounds of KIC loop modeling, the top 10 models converged well. The lowest energy model was lastly selected as the Shal channel model.

Molecular docking

SsTx (PDB ID: 5X0S) structure was based on our previous nuclear magnetic resonance structure (1). RosettaDock application from Rosetta program suite version 3.4 was used to dock SsTx to Shal channel models. Models of the transmembrane domains of Shal were first relaxed in a membrane environment using the Rosetta-Membrane application. From the results of double mutation cycle experiments, the distances between E351 and K17 were constrained to move within a 4-Å-diameter sphere. After docking, the top 1000 models with the lowest total energy score were first selected. They were further scored with the binding energy between the toxin and the channel. The top 10 models with the lowest binding energy were identified as the candidates. The model with the lowest binding energy among the largest cluster of the top 10 models was used as the final SsTx/Shal docking model.


Centipede tissues sample were separated by 12% SDS–polyacrylamide gel electrophoresis, followed by electrotransfer onto polyvinylidene fluoride (PVDF) membranes (Millipore). The PVDF membranes were incubated overnight with a primary antibody at 4°C. The primary antibody diluted with 5% bovine serum albumin (BSA; 4240GR100, BioFroxx) was dissolved in tris-buffered saline containing 0.1% Tween 20 [tris base (2.42 g/liter), NaCl (8 g/liter), and 0.1% Tween 20 (pH 7.6)] and then exposed to chemiluminescent reagents, and images were captured using an ImageQuant LAS 4000 instrument (GE Healthcare). The immunoblotting was performed using the following antibodies: rabbit anti-KV4.1 antibody (1:1000; NBP1-81336, Novus) and mouse β-actin antibody (1:1000; 60008-1-Ig, ProteinTech), which were used as the primary antibodies, and horseradish peroxidase–labeled anti-rabbit (1:5000; #7074, Cell Signaling Technology) or anti-mouse (1:5000; #7076, Cell Signaling Technology) antibodies were used as secondary antibodies.

Histological analysis

As previously described (35), following fixation by 10% formalin and dehydration by an increasing concentration of alcohol, centipede tissues were embedded in paraffin and sectioned to a thickness of 5 μm using a histocut (RM2235, Leica, Germany). Sections of centipede tissues were deparaffinized and rehydrated for hematoxylin and eosin staining.

Similarly, sections of centipede tissues were also deparaffinized and rehydrated for immunohistochemistry (IHC) analysis. For IHC analysis, sections were incubated with a rabbit anti-KV4.1 antibody (1:50; NBP1-81336, Novus) and 2% BSA at 37°C for 1 hour. The epitope recognized by the antibody is between amino acids 56 and 155 (N terminus). After washing in phosphate-buffered saline, the sections were exposed to horseradish peroxidase–labeled anti-rabbit immunoglobulin G (Thermo Fisher Scientific, USA) for 1 hour at room temperature. Immunoreactivity was visualized by incubation with 0.05% 3,3′-diaminobenzidine tetrahydrochloride. Stained sections were observed by light microscopy (X81, Olympus, Japan).

Centipede heart tube isolation and contractility study

Six S. subspinipes (both sexes, 20 g) were adopted for the vascular contractility study. The heart tube was quickly isolated and placed in Krebs solution containing 120 mM NaCl, 1.2 mM CaCl2, 1.2 mM MgSO4, 5 mM KCl, 1 mM NaH2PO4, 25 mM NaHCO3, 12 mM glucose, and 10 mM Hepes (pH 7.4 at 28°C), aerated with 95% O2 and 5% CO2. The heart tube segments were cut into rings (3 mm in length) and mounted on two stainless-steel wires passing through the heart tube. Heart tube isometric tension recording was carried out using amyograph system (Techman) connected to a computer. Each heart tube device was placed individually in a 10-ml tissue bath.

Video recording

The centipedes (S. subspinipes, male) with similar body weights and length were picked for further study. To observe the intraspecific interaction of centipede, two centipedes from the same group were put into a 3-liter conical flask, and intraspecific interaction was recorded by a digital camera. To observe the reaction of SsTx or crude venom injection, the centipedes were put into a glass box (20 cm by 20 cm by 20 cm) and allowed to adapt to the environment for 15 min. SsTx or crude venom was injected into the third segment of centipede’s body. The behavioral response of the centipede was recorded for 30 min after SsTx or crude venom injection. The time course for recovery was 10 min after crude venom injection.

Data analysis

Data from whole-cell recordings were analyzed by Igor Pro (WaveMetrics) and Prism (GraphPad). The results are expressed as means ± SEMs. Student’s t test was applied to examine statistical significance. Statistical significance was accepted at a level of P < 0.05.

Conductance (G)–voltage (V) curves were fitted to a single Boltzmann function, where G/Gmax is the normalized conductance, z is the equivalent gating charge, Vhalf is the half-activation voltage, F is the Faraday’s constant, R is the gas constant, and T is the temperature in kelvinGGmax=11+ezFRT(VVhalf)(1)

IC50 values were derived from fitting a Hill equation to the concentration-response relationship. Changes in IC50 by point mutation may be caused by perturbation of toxin binding. Where Ix represents the difference between the steady-state Shal current and the leaking current in the presence of concentration [x], Imax represents the difference between the maximal current amplitude and the leaking current. IC50 is the concentration for the half-maximal effectIxImax=1[x]nIC50n+[x]n(2)

To perform double-mutant cycle analysis, IC50 values of the four Shal-SsTx combinations [wild-type (WT) channel, WT toxin: IC50_1; channel mutant, WT toxin: IC50_2; WT channel, toxin mutant: IC50_3; channel mutant, toxin mutant: IC50_4] were determined separately. The coupling energy between residues was determined as lnΩ (36)lnΩ=ln(IC50_1IC50_4IC50_2IC50_3)(3)


Supplementary material for this article is available at

This is an open-access article distributed under the terms of the Creative Commons Attribution-NonCommercial license, which permits use, distribution, and reproduction in any medium, so long as the resultant use is not for commercial advantage and provided the original work is properly cited.


Acknowledgments: We thank J. Zheng for discussions. Funding: This work was supported by funding from the National Science Foundation of China (331372208), Chinese Academy of Sciences (XDB31020303 and QYZDJ-SSW-SMC012), and Yunnan Province (2015HA023) to R.L. and from the National Science Foundation of China (31640071 and 31770835), Chinese Academy of Sciences (Youth Innovation Promotion Association and “Light of West China” Program), and Yunnan Province (2017FB037, 2018FA003, and 2019FI005) to S.Y. Author contributions: S.Y., Y.W., and L.W. conducted the majority of experiments including mutagenesis, patch-clamp recordings, cDNA cloning, and animal assays. X.L. and H.Z. constructed the centipede Shal model. X.L. and A.L. performed the molecular docking. S.Y., L.W., P.K., and R.L. prepared the manuscript. Y.W. and A.L. participated in data analysis. R.L. and S.Y. conceived and supervised the project. Competing interests: The authors declare that they have no competing interests. Data and materials availability: The full-length cDNA sequences of centipede channels used in this study have been deposited in GenBank database. Sequence data from this article can be found in the GenBank data library under the accession numbers shown in table S1. All data needed to evaluate the conclusions in the paper are present in the paper and/or the Supplementary Materials. The plasmids of centipede’s KV channels can be provided by R.L. or S.Y. pending scientific review and a completed material transfer agreement. Requests for the plasmids should be submitted to Kunming Institute of Zoology, Chinese Academy of Sciences, China.

Stay Connected to Science Advances

Navigate This Article