KCNE1 potassium channel (IKs)

Peptides 71 (2015) 77–83

Contents lists available at ScienceDirect

Peptides journal homepage: www.elsevier.com/locate/peptides

Engineering a peptide inhibitor towards the KCNQ1/KCNE1 potassium channel (IKs ) Youtian Hu a,1 , Jing Chen a,1 , Bin Wang a,1 , Weishan Yang a , Chuangeng Zhang a , Jun Hu a , Zili Xie a , Zhijian Cao a,c , Wenxin Li a,c , Yingliang Wu a,c,∗∗ , Zongyun Chen a,b,∗ a

State Key Laboratory of Virology, College of Life Sciences, Wuhan University, Wuhan 430072, China Department of Biochemistry and Molecular Biology, Institute of Basic Medical Sciences, Hubei University of Medicine, Hubei, China c Center for BioDrug Research, Wuhan University, Wuhan 430072, China b

a r t i c l e

i n f o

Article history: Received 10 February 2015 Received in revised form 5 July 2015 Accepted 7 July 2015 Available online 15 July 2015 Key words: KCNQ1/KCNE1 Peptide inhibitor Molecular design MT2 peptide MT2-2 peptide

a b s t r a c t The KCNQ1/KCNE1 channel (IKs ) plays important roles in the physiological and pathological process of heart, but no potent peptide acting on this channel has been reported. In this work, we found that the natural scorpion venom hardly inhibited KCNQ1/KCNE1 channel currents. Based on this observation, we attempted to use three natural scorpion toxins ChTX, BmKTX and OmTx2 with two different structural folds as templates to engineer potent peptide inhibitors towards the KCNQ1/KCNE1 channel. Pharmacological experiments showed that when we screen with 1 ␮M MT2 peptide, an analog derived from BmKTX toxin, KCNQ1/KCNE1 channel currents could be effectively inhibited. Concentration-dependent experiments showed that MT2 inhibited the KCNQ1/KCNE1 channel with an IC50 value of 4.6 ± 1.9 ␮M. The mutagenesis experiments indicated that MT2 peptide likely used Lys26 residue to interact with the KCNQ1/KCNE1 channel. With MT2 as a new template, we further designed a more potent MT2-2 peptide, which selectively inhibited the KCNQ1/KCNE1 channel with an IC50 of 1.51 ± 0.62 ␮M. Together, this work provided a much potent KCNQ1/KCNE1 channel peptide inhibitor so far, and highlighted the role of molecular strategy in developing potent peptide inhibitors for the natural toxin-insensitive orphan receptors. © 2015 Elsevier Inc. All rights reserved.

1. Introduction The KCNQ1/KCNE1 channel is a heart disease associated voltagegated potassium channel, which forms IKs currents [17]. The KCNQ1 channel is regulated by auxiliary protein KCNE1 and plays a crucial role in shaping the cardiac action potential as well as in controlling the water and salt homeostasis in several epithelial tissues. [18,26]. The KCNQ1 channel consists of four ␣-subunits, each containing six transmembrane ␣-helical segments, S1–S6, and a membrane-reentering P-loop, which are arranged circumferentially around a central pore as homotetramers [19,20,29]. Although many important progresses have been made in the physiological and pathological functions of the KCNQ1/KCNE1 channel, its struc-

∗ Corresponding author at: Department of Biochemistry and Molecular Biology, Institute of Basic Medical Sciences, Hubei University of Medicine, Hubei, China. ∗∗ Corresponding author at: State Key Laboratory of Virology, College of Life Sciences, Wuhan University, Wuhan, China. E-mail addresses: [email protected] (Y. Wu), [email protected] (Z. Chen). 1 These authors contributed equally to the work. http://dx.doi.org/10.1016/j.peptides.2015.07.002 0196-9781/© 2015 Elsevier Inc. All rights reserved.

ture and function relationship remains obviously unclear. One of important reasons is the lacking of peptide modulators towards the KCNQ1/KCNE1 channel, which makes this channel an orphan receptor till now. Over 400 million years, scorpion has evolved a vast array of ion channel acting-toxins derived from their venom [3,15,36]. Structurally, scorpion neurotoxins are divided into four structural folds: CS-␣/␤ fold [22], CS-␣/␣ fold [4,31], ICK (inhibitor cysteine knot) fold [24], and DDH (disulfide-directed ␤-hairpin) fold [30]. In the last 30 years, many potent scorpion toxins towards different functional receptors have been found. These toxins are widely used to probe the structural and functional relationships of toxin and receptors and develop specific molecular reagents and drugs, especially for different voltage-gated potassium channels, such as the Kv1.1 channel, the Kv1.2 channel, the Kv1.3 channel, and the Kv11.1 channel [33,35]. However, no potent KCNQ1/KCNE1 channel inhibitor has been found from scorpion venoms. In this work, we found that scorpion venom hardly inhibited the KCNQ1/KCNE1 channel currents although the KCNQ1 channel has similar structural fold to other scorpion toxin-sensitive potassium channels [5,10,23]. The molecular design was used and a

78

Y. Hu et al. / Peptides 71 (2015) 77–83

Fig. 1. The KCNQ1/KCNE1 channel was not sensitive towards nature scorpion venom from Buthus martensii Karsch. (A) Amino sequence of pore region of the KCNQ1 channle. (B) Modeled structure of possible toxin-peptide binding region of the KCNQ1/KCNE1 channel. (C) Characterization of KCNQ1/KCNE1 channel currents with XE991. (D) The KCNQ1/KCNE1 channel was not sensitive towards nature scorpion venom from Buthus martensii Karsch. (E) Kv1.3 channel was sensitive towards nature scorpion venom from Buthus martensii Karsch.

potent scorpion toxin analogue MT2-2 was designed with an IC50 of 1.51 ± 0.62 ␮M. To the best of our knowledge, MT2-2 was the most potent peptide inhibitor towards the KCNQ1/KCNE1 channel till now. Together, this work highlighted the role of molecular strategy in developing potent peptide modulators for the toxin-insensitive orphan receptors. 2. Materials and methods 2.1. Construction of toxin peptide expression vectors All designed peptide fragments were generated by overlapping PCR as we have described before [15]. The PCR products of toxin peptides were digested with BamHI and XhoI, and inserted into pGEX-4T-1 expression vector. QuikChangeH Site-Directed Mutagenesis Kit (Stratagene, U.S.A.) was used for generating the mutants based on the wild-type plasmid pGEX-4T-1-MT2. After confirmation by sequencing, the recombinant plasmids were transformed into E. coli Rosetta (DE3) cells for expression. 2.2. Purification and characterization of toxin peptides Peptides were expressed and characterized as previously described [7,14]. After being transformed into Escherichia coli Rosetta (DE3) cells, cells were cultured at 37 ◦ C in LB medium with ampicillin (100 ␮g/ml). When the cell density reached an OD of 0.6, 0.5 mM isopropyl ␤-D-1-thiogalactopyranoside (IPTG) was added to induce the expression at 28 ◦ C. Cells were harvested after 4 h and resuspended in 50 mM Tris–HCl, pH 8.0, 10 mM Na2 EDTA. Super-

natant from the bacterial cell lysate was loaded to a GST-binding binding column. The purified fusion protein was then desalted using centrifugal filtration (Millipore), and cleaved by enterokinase (Biowisdom) at 25 ◦ C for 16 h. Protein samples were then separated by HPLC on a C18 column (10×250 mm, 5 ␮m) (Elite-HPLC) using a linear gradient from 5% to 95% acetonitrile with 0.1% trifluoroacetic acid in 60 min, with detection at 230 nm. Peptides were eluted as major peaks and colleted. The molecular masses of the purified peptides were obtained by MALDI-TOF-MS (Voyager-DESTR, Applied Biosystems). The isolated toxin peptide was lyophilized and stored at −20 ◦ C. 2.3. Circular dichroism (CD) spectroscopy The secondary structures of toxin peptides and mutants were analyzed by circular dichroism (CD) spectroscopy [15]. All samples were dissolved in water at a concentration of 0.2 mg/ml. Spectra were recorded at 25 ◦ C from 250 to 190 nm with a scan rate of 50 nm/min, on a Jasco-810 spectropolarimeter (Jasco Analytical Instruments, Easton, MD, USA). The CD spetra were collected from averaging three scans after subtracting the blank spectrum of water. 2.4. Structure modeling and molecular dynamics (MD) simulation The structures of KCNQ1 and MT2 were modeled using KcsA (PDB code: 1BL8) and BmKTX (PDB code: 1BKT) as templates through the SWISSMODEL server as we have described before [11,19,27]. Using the modeled MT2 and KCNQ1 structures, ZDOCK

Y. Hu et al. / Peptides 71 (2015) 77–83

program was used to generate the models for the MT2-KCNQ1 complex. Several conformations of MT2 modeled from BmKTX were used to improve the rigid performance of ZDOCK. The final MT2KCNQ1 complex structure was sufficiently equilibrated by 2 ns unrestrained molecular dynamic simulations using the SANDER module in the AMBER 11 programs to introduce enough flexibility for both the channel and toxin peptide.

2.5. Electrophysiological studies The cDNAs encoding mKv1.1 in pBSTA, hKv1.2 in pcDNA3/Hygro (+) and mKv1.3 in pSP64 (from Prof. Stephan Grissmer, University of Ulm, Ulm, Germany) were subcloned into the XhoI/BamH I sites of pIRES2-EGFP (Clontech). The constructs were verified by DNA sequencing (Sunbiotech, Wuhan, China). The cDNAs encoding KCNQ1 and mink subunit were cloned into the pcDNA3.1 vector (Invitrogen, Carlsbad, CA, USA). HEK293 cells were cultured in Dulbecco’s modified Eagle’s medium with 10% heat-inactivated fetal calf serum (Invitrogen) supplemented with ampicillin (100 units/ml) and streptomycin (100 ug/ml). Plasmids containing hKCNQ1 (mink), mKv1.1, hKv1.2, mKv1.3 or hIKCa channel was transiently transfected into HEK293 cells (China Center for Type Culture Collection, Wuhan, China) using SofastTM Transfection Reagent (Sunma Biotech, Xiamen, China). Channel currents were measured 1–3 days after transfection. For the KCNQ1/KCNE1 channel, Patch pipettes were filled with 95 mM K-gluconate, 30 mM KCl, 1.2 mM NaH2 PO4, 4.8 mM Na2 HPO4, 0.73 mM Ca-gluconate, 1 mM MgCl2 , 1 mM EGTA, 5 mM D-glucose and 3 mM ATP. The bath solution contained 145 mM Na-gluconate, 1.6 mM K2 HPO4, 0.4 mM KH2 PO4, 8 mM Ca-gluconate, 1 mM MgSO4 and 5 mM dglucose. Data acquisition were obtained with an EPC 10 patch clamp amplifier (HEKA Elektronik, Lambrecht, Germany), which was controlled by PULSE software (HEKA Elektronik) as previously described [13,34]. Data analyses were performed with IGOR software (WaveMetrics, Lake Oswego, OR, USA), and IC50 values were obtained by fitting a modified Hill equation to the data as shown in Equation 1, Itoxin /Icontrol = 1/(1 + ([toxin peptide]/IC50 )) (Eq. 1) where I is the stable current at −120 mV to the normalized data points obtained with at least five different toxin peptide concentrations. Results where indicated are shown as mean ± s.e.m. of at least three experiments.

3. Results 3.1. Unique insensitivity of the KCNQ1/KCNE1 channel towards scorpion venom Sequence analysis of turret and pore region residues of the KCNQ1/KCNE1 channel showed that there is a special Lys318 residue in the KCNQ1/KCNE1 channel pore region, which is usually replaced by an uncharged residue in the animal toxin-sensitive potassium channels, such as Met in the Kv1.3 channel (Fig. 1A and B). In addition, the previous studies showed that both basic scorpion toxins and acidic scorpion toxins were not potent the KCNQ1/KCNE1 channel inhibitors [5,23]. In order to investigate whether scorpion venom contains the KCNQ1/KCNE1 channel inhibitors, we further test the potency of scorpion venom towards the KCNQ1/KCNE1 channel. As shown in Figs. 1C and 1D, the KCNQ1/KCNE1 channel, expressed in HEK293 cells, were effectively blocked by typical small molecule XE991 blockers. However, their currents were less blocked by scorpion venom. Different from the KCNQ1/KCNE1 channel, Kv1.3 channel currents were significantly blocked by scorpion venom (Fig. 1E). These differential sensitivities of KCNQ1 and Kv1.3 channels towards the scorpion venom indi-

79

cated that screening KCNQ1peptide inhibitor from natural scorpion venom might be a huge challenge.

3.2. Engineering a KCNQ1/KCNE1 peptide inhibitor MT2 In view of the KCNQ1/KCNE1 channel insensitivity towards natural scorpion toxins, including classical basic peptides and novel acidic peptides [10,23], the computer aided molecular design was used to design KCNQ1 peptide inhibitors based on our previous experiences about scorpion toxin-potassium channel interactions [8–9,12,15]. By changing the distribution of amino acid residues of three representative natural scorpion toxins ChTX, BmKTX and OmTx2, we engineered three mutant toxin peptides MT1, MT2 and MT3 with the nearly neutral properties instead of basic or acidic features (Fig. 2A). MT1 and MT2 were designed from scorpion toxins ChTX and BmKTX with the CS␣/␤ structural folds, and MT3 was designed from an acidic scorpion toxin OmTx2 with the unique CS␣/␣ structural fold. Pharmacological experiments showed that all three natural scorpion toxins ChTX, BmKTX and OmTx2 were not KCNQ1channel inhibitors when we attempted to screen with 1 ␮M peptides, which was consistent with previous studies (Figs. 2B, C and D) [10]. Excitingly, the currents of KCNQ1 was effectively inhibited by 1 ␮M MT2, although the other two designed peptides MT1 and MT3 showed little effect on the KCNQ1/KCNE1 channel at the same concentration of 1 ␮M (Figs. 2E, F and G). The concentrationdependent experiments further showed that MT2 peptide inhibited the KCNQ1/KCNE1 channel with IC50 value of 4.6 ± 1.9 ␮M (Fig. 2H and I). Sequence analysis showed that MT2 peptide contained four basic residues (R11, K26, K31 and K37) and three acidic residues (E9, D19 and E23) with different distribution characteristics, suggesting that the distribution of charged residues might affect BmKTX potency towards the KCNQ1/KCNE1 channel.

3.3. Key residues of MT2 interacting with the KCNQ1/KCNE1 channel The peptide MT2 potency indicated that it could be a template for further improving its affinity. In order to design more potent KCNQ1/KCNE1 peptide inhibitors with MT2 as a new template, we first investigate its structure-function relationship through mutating three basic residues (Arg11, Lys26 and Lys37) into alanine residues, respectively. By the circular dichroism spectroscopy, all recombinant MT2 mutants were found to have similar secondary structures to peptide MT2 template, which suggested the conserved structures of toxin peptides with CS␣/␤ structural fold (Fig. 3A). Pharmacological experiments showed that Arg11 and Lys37 have weak effects on MT2 activities, and Lys26 was the most crucial to MT2 activity because its replacement had the greatest effect on the KCNQ1/KCNE1 channel (Fig. 3B). To further re-design potent KCNQ1 peptide inhibitors with MT2 peptide as the template, a structural model of MT2-KCNQ1 complex was obtained to dig the helpful information based on MT2-the KCNQ1 channel interaction through the combined computational approaches [15]. The structural analysis indicated that there were major polar interactions between four Lys318 residues of the KCNQ1 channel and some toxin residues including Glu9, Ser10, Phe24, Thr28, Asn29, His33 and Ser35 (Fig. 3C) within a contact distance of 5 Å. Based on these analyses, the binding interfaces of scorpion toxins with CS␣/␤ fold towards the KCNQ1 channel could bee divided into three major regions, one is the turn region just before ␣-Helix, the second is the turn region just after ␣-Helix, and the third is ␤-sheet region (Fig. 3D). These results suggested that Lys26 is a key residue of MT2 interacting with the KCNQ1/KCNE1 channel, and three associated regions might influence the interaction.

80

Y. Hu et al. / Peptides 71 (2015) 77–83

Fig. 2. Engineering a KCNQ1/KCNE1 sensitive peptide MT2 with scorpion BmKTX as a template. (A) Three designed peptides MT1, MT2 and MT3. (B–G) Pharmalogical activities of three natural scorpion toxins and three designed toxin peptides towards the KCNQ1/KCNE1 channel. (H) Inhibiting effects of designed toxin peptide MT2 on the KCNQ1/KCNE1 channel in transfected HEK-293 cells at a concentration of 3 ␮M. (I) Concentration-response curve for MT2 on the KCNQ1/KCNE1 channel in transfected HEK-293 cells. Each concentration was tested at least three times, and inhibiting rates are shown as mean ± SD. (C–D) Acidic and basic residue distributions of designed peptide MT2.

3.4. Engineering a KCNQ1/KCNE1 peptide inhibitor MT2-2 with MT2 as a template Based on previous analyses, two new mutant peptides were designed with MT2 as a template through changing the hydrophobic side chain in region1 and the charged residue distributions in region2 and region3 (Fig. 3E). Next, we produced the recombinant MT2-1 and MT2-2 peptides through the same methods as that of MT2 peptide. Pharmacological experiments showed that MT2-1 had similar inhibiting activity towards the KCNQ1/KCNE1 channel to that of MT2 template (Fig. 4A). Interestingly, MT2-2 showed better inhibiting affinity than that of MT2 peptide (Fig. 4B). MT2-2 peptide inhibited about 45% the KCNQ1/KCNE1 channel currents at a concentration of 1 ␮M, and the concentration-dependent experiments further showed that MT2-2 inhibited the KCNQ1/KCNE1

channel with an IC50 value of 1.5 ± 0.6 ␮M (Fig. 4C). Besides its better potency, MT2-2 peptide was also found to have good selectivity towards the KCNQ1/KCNE1 channel. As shown in Figs. 4D-4G, MT2-2 peptide had little effect on Kv1.1, Kv1.2, Kv1.3 and IKCa channel currents at a concentration of 1 ␮M. Together, these results provided the first engineered sensitive the KCNQ1/KCNE1 channel peptide inhibitor with a natural and insensitive scorpion toxin scaffold. 4. Discussion The KCNQ1/KCNE1 channel is closely linked to cardiac abnormalities, and has no previously identified potent peptide inhibitor [6,21]. Scorpion has evolved a vast array of potassium channel acting-toxins derived from their venom, which can interact with

Y. Hu et al. / Peptides 71 (2015) 77–83

81

Fig. 3. Molecular designs of two new MT2 analogues MT2-1 and MT2-2 based on MT2-KCNQ1 interaction mechanism. (A) Circular dichroism spectrum analyses of MT2 and its alanine-scanning-analogues MT2-R11A, MT2-K26A, and MT2-K37A. All data represent the mean of at least three experiments. (B) The conserved pore-blocking residue Lys26 of designed scorpion toxin mutant MT2 determined its interaction with the KCNQ1/KCNE1 channel. (C) Key residues of MT2 peptide interacting with the Lys318 residue of the KCNQ1/KCNE1 channel. (D) Three binding regions of MT2 peptide interacting with the Lys318 residue in the KCNQ1/KCNE1 channel. (E) Molecular design of two new peptides MT2-1 and MT2-2.

diverse potassium channel subtypes [25]. Our previous work identified a natural acidic scorpion toxin ImKTx104 that had weak activity towards the KCNQ1/KCNE1 channel with a Kd value of about 11.69 ␮M [10]. Here, a new scorpion toxin analogue MT22 with a conserved CS␣/␤ fold was designed, which inhibited the KCNQ1/KCNE1 channel with an IC50 value of 1.51 ± 0.62 ␮M. This work provided the first potent KCNQ1 peptide inhibitor and highlighted the potential of molecular design for natural toxin insensitive orphan receptors. At present, the molecular design of potent ligands towards orphan receptors still remains a huge challenge, and its development depends on understanding the detailed chemical and biological mechanisms by which ligand-receptor interactions occur

[1,32]. On the basis of known ligand-receptor recognition modes, the combined computational and experimental methods might be a new method to design some more potent and specific ligands using the known natural molecular ligands, as we and other teams have reported before [15]. Here, we attempted to design a peptide inhibitor towards orphan receptor the KCNQ1/KCNE1 channel based on the structural and functional analyses. In view of the insensitivities of both natural basic and acidic scorpion toxins towards the KCNQ1/KCNE1 channel, the peptides with weaker basic features were designed, and BmKTX analogue MT2 with an IC50 value of 4.6 ± 1.9 ␮M was obtained, suggesting that the strategy of the first round molecular design was partly successful. Subsequently, we further designed a more potent inhibitor

82

Y. Hu et al. / Peptides 71 (2015) 77–83

Fig. 4. Pharmacological activities of two designed toxin peptides MT2-1 and MT2-2. (A) Inhibiting effects of MT2-1 on the KCNQ1/KCNE1 channel at a concentration of 1 ␮M. (B) Inhibiting effects of MT2-2 on the KCNQ1/KCNE1 channel at a concentration of 1 ␮M. (C) Concentration-response curve for MT2-2 on the KCNQ1/KCNE1 channel. Each concentration was tested at least three times, and inhibiting rates are shown as mean ± SD. (D) Blocking effects of MT2-2 on mKv1.1 currents. (E) Blocking effects of MT2-2 on hKv1.2 currents. (F) Blocking effects of MT2-2 on mKv1.3 currents. (G) Blocking effects of MT2-2 on hIKCa currents.

of MT2-2 peptide (Figs. 3 and 4). Sequence analysis showed that MT2-2 contains much fewer basic and acidic residues than those of MT2 peptide, suggesting that the distribution of charged residues might play an important role in peptide-the KCNQ1/KCNE1 channel interactions, but the detailed residue(s) in MT2-2 to determine its potency and selectivity were still unclear before further structure and function studies were done [8,15,16]. For potassium channel receptors, the two domains responsible for specific peptide ligand recognition are the turret and the filter region [28]. Lys318 residue in the KCNQ1/KCNE1 channel, adjacent to the conserved channel selectivity filter motif “GYGD”, might confer the insensitivity of natural scorpion toxins towards the KCNQ1/KCNE1 channel. Among the known scorpion toxinsensitive potassium channels, there are no basic residues near the selective filter [25]. Our primary experiments indicated that MT2 peptide inhibitor involved the classical basic residue Lys26 as the pore-blocking residue, which blocked the channel pore in many scorpion toxins [2,15]. The conserved residue Lys26 presumably also blocked the channel pore of the KCNQ1/KCNE1 channel pore, while the other designed residues might interact with the Lys318 residue near the channel selectivity filter and contributed the affinity of MT2 or MT2-2 peptides towards the KCNQ1/KCNE1 channel. Take together, natural scorpion venom provide many candidate peptides for potential peptide drug leads targeting diverse receptors. However, for natural venom insensitive “orphan receptors”, it still remains a significant challenge. Our primary success with the design of MT2 and MT2-2 peptide inhibitors suggests that

molecular design based on the charged residue distribution and the model of the peptide-potassium channel complex would accelerate the development of diagnostic and therapeutic agents for orphan receptors using natural venom peptides as scaffolds. Conflict of interest The authors declare no conflict of interest. Acknowledgments This work was supported by grants from the National Natural Sciences Foundation of China (Nos. 31470812, 31170789 and 31200557). References [1] D.G. Baden, K.S. Rein, R.E. Gawley, G. Jeglitsch, D.J. Adams, Is the A-ring lactone of brevetoxin PbTx-3 required for sodium channel orphan receptor binding and activity, Nat Toxins. 2 (1994) 212–221. [2] A. Banerjee, A. Lee, E. Campbell, R. Mackinnon, Structure of a pore-blocking toxin in complex with a eukaryotic voltage-dependent K+ channel, eLife 2 (2013) e00594. [3] C. Beeton, H. Wulff, N.E. Standifer, P. Azam, K.M. Mullen, M.W. Pennington, et al., Kv1.3 channels are a therapeutic target for T cell-mediated autoimmune diseases, Proc. Natl. Acad. Sci. U.S.A. 103 (2006) 17414–17419. [4] B. Chagot, C. Pimentel, L. Dai, J. Pil, J. Tytgat, T. Nakajima, et al., An unusual fold for potassium channel blockers: NMR structure of three toxins from the scorpion Opisthacanthus madagascariensis, Biochem. J. 388 (2005) 263–271.

Y. Hu et al. / Peptides 71 (2015) 77–83 [5] H. Chen, L.A. Kim, S. Rajan, S. Xu, S.A. Goldstein, Charybdotoxin binding in the I(Ks) pore demonstrates two MinK subunits in each channel complex, Neuron 40 (2003) 15–23. [6] Y.H. Chen, S.J. Xu, S. Bendahhou, X.L. Wang, Y. Wang, W.Y. Xu, et al., KCNQ1 gain-of-function mutation in familial atrial fibrillation, Sci. N.Y. 299 (2003) 251–254. [7] Z. Chen, S. Han, Z. Cao, Y. Wu, R. Zhuo, W. Li, Fusion expression and purification of four disulfide-rich peptides reveals enterokinase secondary cleavage sites in animal toxins, Peptides 39 (2013) 145–151. [8] Z. Chen, Y. Hu, J. Hu, W. Yang, J.M. Sabatier, M. De Waard, et al., Unusual binding mode of scorpion toxin BmKTX onto potassium channels relies on its distribution of acidic residues, Biochem. Biophys. Res. Commun. 447 (2014) 70–76. [9] Z.Y. Chen, Y.T. Hu, W.S. Yang, Y.W. He, J. Feng, B. Wang, et al., Hg1, novel peptide inhibitor specific for Kv1.3 channels from first scorpion Kunitz-type potassium channel toxin family, J. Biol. Chem. 287 (2012) 13813–13821. [10] Z.Y. Chen, D.Y. Zeng, Y.T. Hu, Y.W. He, N. Pan, J.P. Ding, et al., Structural and functional diversity of acidic scorpion potassium channel toxins, PLoS One 7 (2012) e35154. [11] D.A. Doyle, J. Morais Cabral, R.A. Pfuetzner, A. Kuo, J.M. Gulbis, S.L. Cohen, et al., The structure of the potassium channel: molecular basis of K+ conduction and selectivity, Science 280 (1998) 69–77. [12] J. Feng, Y. Hu, H. Yi, S. Yin, S. Han, J. Hu, et al., Two conserved arginine residues from the SK3 potassium channel outer vestibule control selectivity of recognition by scorpion toxins, J. Biol. Chem. 288 (2013) 12544–12553. [13] B.G. Fry, K. Roelants, D.E. Champagne, H. Scheib, J.D. Tyndall, G.F. King, et al., The toxicogenomic multiverse: convergent recruitment of proteins into animal venoms, Ann. Rev. Genom. Hum. Genet. 10 (2009) 483–511. [14] G. Gan, H. Yi, M. Chen, L. Sun, W. Li, Y. Wu, et al., Structural basis for toxin resistance of beta4-associated calcium-activated potassium (BK) channels, J. Biol. Chem. 283 (2008) 24177–24184. [15] S. Han, H. Yi, S.J. Yin, Z.Y. Chen, H. Liu, Z.J. Cao, et al., Structural basis of a potent peptide inhibitor designed for Kv1.3 channel, a therapeutic target of autoimmune disease, J. Biol. Chem. 283 (2008) 19058–19065. [16] S. Han, S. Yin, H. Yi, S. Mouhat, S. Qiu, Z. Cao, et al., Protein–protein recognition control by modulating electrostatic interactions, J. Proteome Res. 9 (2010) 3118–3125. [17] S.C. Harmer, J.S. Mohal, A.A. Royal, W.J. McKenna, P.D. Lambiase, A. Tinker, Cellular mechanisms underlying the increased disease severity seen for patients with long QT syndrome caused by compound mutations in KCNQ1, Biochem. J. 462 (2014) 133–142. [18] D. Heitzmann, V. Koren, M. Wagner, C. Sterner, M. Reichold, I. Tegtmeier, et al., KCNE beta subunits determine pH sensitivity of KCNQ1 potassium channels, Cell. Physiol. Biochem. Int. J. Exp. Cell. Physiol. Biochem. Pharmacol. 19 (2007) 21–32. [19] R.J. Howard, K.A. Clark, J.M. Holton, D.L. Minor Jr., Structural insight into KCNQ (Kv7) channel assembly and channelopathy, Neuron 53 (2007) 663–675. [20] S.B. Long, E.B. Campbell, R. Mackinnon, Voltage sensor of Kv1.2: structural basis of electromechanical coupling, Science 309 (2005) 903–908.

83

[21] S.O. Marx, J. Kurokawa, S. Reiken, H. Motoike, J. D’Armiento, A.R. Marks, et al., Requirement of a macromolecular signaling complex for beta adrenergic receptor modulation of the KCNQ1-KCNE1 potassium channel, Sci. N.Y. 295 (2002) 496–499. [22] W. Massefski Jr., A.G. Redfield, D.R. Hare, C. Miller, Molecular structure of charybdotoxin: retraction, Science 252 (1991) 631. [23] T.J. Morin, W.R. Kobertz, A derivatized scorpion toxin reveals the functional output of heteromeric KCNQ1-KCNE K+ channel complexes, ACS Chem. Biol. 2 (2007) 469–473. [24] A. Mosbah, R. Kharrat, Z. Fajloun, J.G. Renisio, E. Blanc, J.M. Sabatier, et al., A new fold in the scorpion toxin family, associated with an activity on a ryanodine-sensitive calcium channel, Proteins 40 (2000) 436–442. [25] S. Mouhat, N. Andreotti, B. Jouirou, J.M. Sabatier, Animal toxins acting on voltage-gated potassium channels, Curr. Pharm. Des. 14 (2008) 2503–2518. [26] D. Peroz, N. Rodriguez, F. Choveau, I. Baro, J. Merot, G. Loussouarn, Kv7. 1 (KCNQ1) properties and channelopathies, J. Physiol. 586 (2008) 1785–1789. [27] J.G. Renisio, R. Romi-Lebrun, E. Blanc, O. Bornet, T. Nakajima, H. Darbon, Solution structure of BmKTX, a K+ blocker toxin from the Chinese scorpion Buthus Martensi, Proteins 38 (2000) 70–78. [28] R.C. Rodriguez de la Vega, E. Merino, B. Becerril, L.D. Possani, Novel interactions between K+ channels and scorpion toxins, Trends Pharmacol. Sci. 24 (2003) 222–227. [29] J.A. Smith, C.G. Vanoye, A.L. George Jr., J. Meiler, C.R. Sanders, Structural models for the KCNQ1 voltage-gated potassium channel, Biochemistry 46 (2007) 14141–14152. [30] J.J. Smith, J.M. Hill, M.J. Little, G.M. Nicholson, G.F. King, P.F. Alewood, Unique scorpion toxin with a putative ancestral fold provides insight into evolution of the inhibitor cystine knot motif, Proc. Natl. Acad. Sci. U.S.A. 108 (2011) 10478–10483. [31] K.N. Srinivasan, V. Sivaraja, I. Huys, T. Sasaki, B. Cheng, T.K. Kumar, et al., kappa-Hefutoxin1, a novel toxin from the scorpion Heterometrus fulvipes with unique structure and function. Importance of the functional diad in potassium channel selectivity, J. Biol. Chem. 277 (2002) 30040–30047. [32] W.J. Wang, Y. Wang, H.Z. Chen, Y.Z. Xing, F.W. Li, Q. Zhang, et al., Orphan nuclear receptor TR3 acts in autophagic cell death via mitochondrial signaling pathway, Nat. Chem. Biol. 10 (2014) 133–140. [33] H. Yi, Z. Cao, S. Yin, C. Dai, Y. Wu, W. Li, Interaction simulation of hERG K+ channel with its specific BeKm-1 peptide: insights into the selectivity of molecular recognition, J. Proteome Res. 6 (2007) 611–620. [34] S.J. Yin, L. Jiang, H. Yi, S. Han, D.W. Yang, M.L. Liu, et al., Different residues in channel turret determining the selectivity of ADWX-1 inhibitor peptide between Kv1.1 and Kv1.3 channels, J. Proteome Res. (2008) 4890–4897. [35] M. Zhang, Y.V. Korolkova, J. Liu, M. Jiang, E.V. Grishin, G.N. Tseng, BeKm-1 is a HERG-specific toxin that shares the structure with ChTx but the mechanism of action with ErgTx1, Biophys. J. 84 (2003) 3022–3036. [36] S. Zhu, S. Peigneur, B. Gao, L. Luo, D. Jin, Y. Zhao, et al., Molecular diversity and functional evolution of scorpion potassium channel toxins, Mol. Cell Proteomics 10 (2010), M110 002832.