A synthetic weak neurotoxin binds with low affinity to Torpedo and chicken a7 nicotinic acetylcholine receptors Siew Lay Poh 1,2 , Gilles Mourier 1 , Robert Thai 1 , Arunmozhiarasi Armugam 2 , Jordi Molgo ´ 3 , Denis Servent 1 , Kandiah Jeyaseelan 2 and Andre ´ Me ´ nez 1 1 CEA, Saclay, Gif-sur-Yvette, France; 2 National University of Singapore, Singapore; 3 UPR, CNRS, Gif-sur-Yvette, France Weak neurotoxins from snake venom are small proteins with five disulfide bonds, which have been shown to be poor binders of nicotinic acetylcholine receptors. We report on the cloning and sequencing of four cDNAs encoding weak neurotoxins from Naja sputatrix venom glands. The protein encoded by one of them, Wntx-5, has been synthesized by solid-phase synthesis and characterized. The physicochemi- cal properties of the synthetic toxin (sWntx-5) agree with those anticipated for the natural toxin. We show that this toxin interacts with relatively low affinity (K d ¼ 180 n M ) with the muscular-type acetylcholine receptor of the electric organ of T. marmorata, and with an even weaker affinity (90 l M ) with the neuronal a7 receptor of chicken. Electro- physiological recordings using isolated mouse hemidia- phragm and frog cutaneous pectoris nerve–muscle preparations revealed no blocking activity of sWntx-5 at l M concentrations. Our data confirm previous observations that natural weak neurotoxins from cobras have poor affinity for nicotinic acetylcholine receptors. Keywords: snake neurotoxins; nicotinic acetylcholine receptors. During the past three decades, the most ÔobviousÕ venom toxins have been uncovered either because they are present in large amounts and/or because they have been directly associated with the search for an important target. At present, two additional approaches may be considered to discover new toxin functions. One of them is a proteomic- type approach, which aims at isolating all components of the ÔtoxinomeÕ [1,2]. The second approach involves investi- gation of the vast number of venom components that have already been isolated, and sometimes chemically character- ized, but whose biological activity still remains mysterious. These functionally unknown components are often classi- fied as miscellaneous types of toxins, even though they usually belong to well-identified structural families [3]. This is the case of the so-called Ôweak neurotoxinsÕ (Wntxs) found in elapid snakes and isolated for the first time 26 years ago from the venom of Naja melanoleuca [4]. Since then, more such toxins have been isolated [5–19]. The Wntxs possess 62–68 amino acids and belong to the structural family of Ôthree-fingeredÕ folded toxins, which includes the cardiotoxins, muscarinic toxins, acetylcholin- esterase inhibitors and the a-neurotoxins that block mus- cular and/or neuronal nicotinic acetylcholine receptors (AChRs) [20–22]. The fold adopted by all these toxins is characterized by three adjacent loops rich in b-pleated sheet, tethered by four conserved disulphides. A fifth loop is sometimes observed in the second loop of the a/j-neuro- toxins and j-neurotoxins [22], where it specifically contri- butes to the binding of the toxins to the neuronal AChR [23–26]. Wntxs also possess a fifth disulfide bond, but this is located in the first loop [16,27,28]. Using Wntxs isolated from venom, it was shown that these molecules interact with AChRs but with low affinities [10,29]. Many efforts have been made to obtain pure Wntxs. However, it cannot be completely ruled out that their low activity may be due to the presence of minor but highly potent contaminants, as was previously observed in the case of j-bungarotoxin [30]. We identified four cDNAs enco- ding Wntxs in venom glands of the cobra N. sputatrix (previously known as Naja naja sputatrix [31]) and selec- ted one of them. Then, we synthesized the correspond- ing Wntx (Wntx-5) by chemical means, characterized its physicochemical properties and investigated its biological properties. We show that sWntx-5 is a weak binder of muscular-type AChR from Torpedo marmorata’s electric organ and an even weaker binder of the a7 neuronal-type receptor from chicken. Our data generally agree with a report published recently [29]. Moreover, the low AChR binding activity of Wntx-5 can be accounted for by the presence of a few residues that are also found in potent three-fingered snake neurotoxins [22,32,33]. MATERIALS AND METHODS Materials Bacterial strains used, JM109 [34] and EpicureanÒ coli SUREÒ cells, were from Stratagene (USA). Oligonucleo- tides were synthesized at the National University of Singa- pore. Molecular biology reagents were from Amersham International Inc. (UK), Promega, New England Biolabs, Correspondence to A. Me ´ nez, De ´ partement d’ Inge ´ nierie et d’Etudes des Prote ´ ines, CEA, Saclay, 91191 Gif-sur-Yvette Cedex, France. E-mail: andre.menez@cea.fr Abbreviations: Wntx, weak neurotoxin; AchR, nicotinic acetylcholine receptor; TCEP, tris(2-carboxyethyl)-phosphine hydrochloride; Bgtx, bungarotoxin; Ea, erabutoxin a. Note: The cDNA sequences reported in this paper have the GenBank accession numbers AF026891, AF026892, AF098923 and AF098923. (Received 18 February 2002, revised 17 June 2002, accepted 12 July 2002) Eur. J. Biochem. 269, 4247–4256 (2002) Ó FEBS 2002 doi:10.1046/j.1432-1033.2002.03113.x Novagen or Perkin Elmer (USA). Protected amino acid derivatives, resins, dicyclohexylcarbodiimide and N-hydro- xybenzotriazole were from Nova-Biochem (Meudon, France). Piperidine, N-methylpyrrolidone, dichloro- methane, methanol, trifluroacetic acid and ter-butyl- methyloxide were from SDS (Peypin, France). TCEP [tris (2-carboxyethyl)-phosphine hydrochloride] was from Pierce (Rockford, Illinois, USA, or Saint-Quentin-Fallavier, France). Oxidized and reduced glutathione (GSH and GSSH respectively) were from Sigma (St Louis, MO). Automated chain assembly was performed on a standard Applied Biosystems 431 peptide synthesiser. cDNA of the chimeric a7-V201–5HT 3 was kindly provided by J. P. Changeux (Pasteur Institute, Paris, France.) and 125 I-labelled bungaro- toxin (210–250 CiÆmmol )1 ) was from Amersham. RT-PCR and subcloning Total RNA prepared from the venom glands of N. sputatrix [35] was used in RT-PCR. Reverse transcription reactions were performed with 3 lgofRNAinatotalreaction volume of 10 lLof1·RT buffer (100 m M Tris/HCl, pH 8.4; 250 m M KCl; 12.5 m M MgCl 2 and 0.5 mgÆmL )1 BSA) containing 10 U of MuMLV reverse transcriptase, 0.5 m M of dNTP and 40 ng of antisense primer X191; 5¢-gCggCggAATTCTTTTTTTTTTTTTTTTTT-3¢.The reaction was carried out at 42 °C for 1 h, and terminated by heating at 95 °C for 5 min. The entire mixture was used in a 50-lL polymerase chain reaction. The full-length cDNA was cloned using two pairs of primers, which recognized conserved regions of genes encoding Wntxs. The first pair was X289 (5¢ TgTgCTACTTgCC CTggAA 3¢)andX191. The second pair was X133 (5¢ TCC AgAAAAgATCgCAA gATg 3¢) [35] and X300 (5¢ AgAgC CAAgCTTTTACT ATCggTT 3¢). The PCR products were fractionated using a low melting point agarose gel (1.2%). The DNA band was cut out and purified using freeze–thaw or centrifuged methods as described previously [36,37]. The amplified products were ligated to pT7 Blue(R) vector using procedures described by the supplier (Novagen, USA). The ligated products were transformed into E. coli, JM109 or SURE cells [34] by electroporation and selected on LB-ampicillin (50 lgÆmL )1 ) plates supplemented with 5-bromo-4-chloroindol-3-yl b- D - galactoside (X-gal) and isopropyl thio-b- D -pyranoside (IPTG). Putative recombinant plasmids were sequenced on both strands with M13/pUC forward and reverse universal primers using the dideoxy chain termination method [38] on an automated DNA sequencer (Model 373, Applied Biosystems, USA), using the manufacturer’s pro- tocol and reagents. Sequence analysis Searches for homologous proteins on GenBank databases (National Center for Biotechnology Information, USA) were performed using the BLAST program, with deduced amino acid sequences from the cDNAs [39]. Chemical synthesis of toxin The peptide was assembled by a stepwise solid-phase method with dicyclohexylcarbodiimide/hydroxybenzotriazole as coupling reagents and N-methyl pyrrolidone as a solvent. Fmoc-protected amino acids were used with t-butyl ester (Glu, Asp), t-butyl ether (Ser, Thr, Tyr), trityl (Cys, Asn, Gln), t-butylcarbonyl (Lys) and 2,2,5,7,8-pentamethyl- chromane-6-sulfonyl (Arg) [40]. Wntx-5 was assembled on Fmoc-Arg-(Pmc)-Wang resin (loading 0.52 mmolÆg )1 ) [41]. The synthesis was carried out using a version derived from the Applied Biosystem standard Fmoc 0.1 mmol small-scale program [42]. At the end of the synthesis, the peptide was cleaved from the resin and the protecting groups were removed from the amino acid side chains using a mixture of trifluoroacetic acid (90%), triisopropyl- silan (5%) and deionized water (5%). After 2 h incubation at room temperature with constant mixing, the mixture was filtered into cold t-butyl methyloxide (peroxide-free) and centrifuged at 2 960 g for 30 min. The peptide precipi- tates were washed three times and dried, dissolved in 10% acetic acid and lyophilized. The synthetic toxin was reduced with molar excess of TCEP under acidic condi- tions and purified by RP-HPLC using a Vydac C18 column (250 · 10 mm) with a gradient of 40–60% of 60% acetonitrile mixed with 0.1% trifluoroacetic acid in water. The flow rate was 3 mLÆmin )1 and the detection was monitored at 214 nm. Peptide purity was assessed using an analytical Vydac C18 column (250 · 0.46 mm) using the same elution conditions. Disulfide bond formation and protein purification The reduced synthetic peptide was oxidized in a refolding buffer (0.1 M sodium acetate, 1 M 4 GdnHCl and 1 m M EDTA, pH 7.8) containing GSH and GSSH in a molar ratio of 10 : 1. The reduced synthetic peptide was dissolved in 0.2 mL of 0.1% trifluoroacetic acid, and immediately diluted into oxidation buffer to a final concentration of 0.05 mgÆmL )1 . After incubation for up to 3 days at room temperature, the peptide was acidified with 30% trifluoro- acetic acid and purified on a Vydac C18 semipreparative column using the gradient employed to purify the reduced toxin form. The protein concentration was determined by means of spectrometry. Mass analysis, amino acid composition and sequence determination The masses of both the reduced and refolded peptides were determined using an ion spray mass spectra system, Micromass Platform II (Micromass, Altrincham, UK). For amino acid composition analysis, the peptide was hydrolysed in a sealed vial heated at 120 °C in the presence of 6 M HCl for 16 h. The hydrolysate was analysed using an Applied Biosystem Model 130A automatic analyser equipped with an online 420A derivatiser for the conversion of the free amino acid into phenyl thiocarbamoyl deriva- tives. The amino acid sequence of the peptide was determined using an applied Biosystems 477A protein sequencer. Circular dichroism CD spectra were recorded on a CD6 Jobin Yvon dichro- graph (Roussel Uclaf, France). Routinely, measurements were performed at 20 °C in 0.1 cm pathlength quartz cells 4248 S. L. Poh et al.(Eur. J. Biochem. 269) Ó FEBS 2002 (Hellma, Germany) under continuous nitrogen gas flow with a peptide concentration around 4.5 · 10 )5 M in deionized water. Spectra were recorded in the 186–260 nm wavelength range. Each spectrum represents the average of four spectra. Binding to acetylcholine receptors Binding assays were performed using 125 I-labelled a-bung- arotoxin (a-Bgtx, 210–250 CiÆmmol )1 , Amersham) as com- petitor. The AChR-rich membranes from the electric organ of T. marmorata were prepared as described previously [43]. The chimeric a7 receptors were obtained by expressing the chimeric cDNA (a7–5HT 3 ) in HEK cells [23]. In compet- itive experiments with AChR from T. marmorata we measured, at equilibrium, the effect of toxins on the 125 I-labelled a-Bgtx binding. Varying amounts of toxins were incubated with 3 n M of active sites of receptors and 5 n M of 125 I-labelled a-Bgtx for at least 4 h. With a7 receptors, the toxin was incubated at different concentra- tions with 250 lL of cells suspended in NaCl/P i for at least 30 min. Cell suspensions were filtered 6 min after addition of 5 n M 125 I-labelled a-Bgtx. The curves were fitted with the empirical Hill equation. K d values for Torpedo AChR were calculated according to Cheng and Prusoff [44]. For a7 receptors, protection constants (K p ) calculated by fitting the competition data using the Hill equation, correspond to K d values [45]. Electrophysiological recordings Electrophysiological recordings were carried out on both isolated mouse hemidiaphragm preparations (removed from adult female Swiss–Webster mice killed by dislocation of the cervical vertebrae followed by immediate exsanguin- ation), and from isolated cutaneous pectoris nerve-muscle preparations removed from double-pithed male frogs (Rana temporaria), as described previously [46]. Briefly, the motor nerve was stimulated with a suction microelec- trode adapted to the diameter of the nerve, with pulses of 0.05–0.1 ms duration and supra-maximal voltage (typically 3–8 V) supplied by a S-44 stimulator (Glass Instruments, West Warwick, USA) linked to a stimulus isolation unit. Membrane potentials and synaptic potentials were recor- ded, from endplate regions with intracellular microelec- trodes filled with 3 M KCl (8–18 MX resistance), using conventional techniques and an Axoclamp-2A system (Axon Instruments, Union City, CA, USA). Recordings were made continuously from the same endplate before and after application of toxins tested. Electrical signals after amplification were collected and digitized, at a sampling rate of 25 kHz, with the aid of a computer equipped with an analogue-to-digital interface board (DT2821, Data Trans- lation, Marlboro, USA). Endplate potentials and miniature endplate potentials were analysed individually for amplitude and time course. RESULTS Cloning and sequencing of cDNAs Thirty-three putative clones were obtained from a cDNA library prepared from venom glands of N. sputatrix,usinga conventional RT-PCR-based approach. The ORFs of these cDNAs encode a set of four novel proteins that were named Wntx-5, 6, 8 and 9 (Fig. 1). The putative leader sequences contain 21 amino residues and are typical of secreted proteins [47]. Only the isoform Wntx-5 showed variation in its signal peptide region due to a single first base substitution (Fig. 1). The calculated theoretical molecular masses of these basic Wntxs were 7504.5 Da, 7509.1 Da, 7508.1 Da and 7535 Da. The four derived amino acid sequences (Seq.1–4 in Fig. 2A) show high similarity. Wntx-6 possesses an aspartic acid at position 21 whilst other sequences have an asparagine, Wntx-5 has a lysine at position 29 whereas other sequences have a methionine, and Wntx-9 has an asparagine at position 65 whilst other sequences have a serine. They all exhibit high sequence similarity with other Wntxs (Fig. 2A) but they are clearly more similar to those from cobras than to those found in kraits, mamba and coral snake venom [4–19]. Comparative analysis of Wntx sequences Figure 2A shows a comparison of amino acid sequences of 26 putative Wntxs including those derived from cDNAs isolated from N. sputatrix. The high degree of identity of the sequences isolated from cobras is striking, both in terms of length and amino acid distribution. Those from kraits, mambas and coral snakes display more deviations and a smaller number of conserved residues (see for example the three Wntxs at the bottom of the group). Thus, 25 positions (indicated by open boxes in Fig. 2A) are strictly conserved among cobra Wntxs. These include 10 half-cystines and 15 additional residues. Using Fig. 2A numbering, these additional residues are Leu1, Pro7, Glu8, Gly22, Glu23, Phe27, Lys28, Tyr43, Gly46, Ala48, Thr50, Pro52, Thr66, Asp67 and Asn70. Sixteen additional positions of cobra Wntxs are occupied by highly conserved residues ( 80%, shaded boxes in Fig. 2). These residues include Thr2, Leu4, Phe/Tyr10, Asn21, Lys24, Lys/Arg29, Arg33, Arg42, Arg45, Lys55, Pro56, Arg/Lys57, Asp/ Glu58, Val61, Ser65 and Lys/Arg68. Therefore, cobras Wntxs form a highly homogeneous group of proteins, which share at least 56% sequence similarity (excluding their disulfide bonds). We noted that Wntx-5 has a particularly high degree (62–97%) of identity with other cobra Wntx sequences, making it a potential prototype of Wntxs from cobras. Therefore, we decided to synthesize Wntx-5 for the investigation of biological properties of cobra Wntxs. Synthesis and purification of synthetic Wntx-5 (sWntx-5) Wntx-5 was synthesized chemically using a modified version of the Fmoc/small-scale (0.1 mmol) programme developed by Applied Biosystems [42] using a preloaded Arg-(Pmc)- Wang resin as solid support [41]. After treatment with the trifluoroacetic acid cleavage mixture and lyophilization, the crude peptide was treated in acidic conditions with TCEP, a reducing agent, and was purified by reverse-phase HPLC on a C18 column. Figure 3A shows that the RP-HPLC profile of the crude peptide displayed three major peaks (a, b and c). Electrospray mass analyses revealed that peak a was a truncated form of sWntx-5 terminated at Pro33 (3749.6 Da), peak b contained peptides ranging from Ó FEBS 2002 Synthetic weak neurotoxin (Eur. J. Biochem. 269) 4249 7513.5 to 7530.5 Da and peak c contained a peptide with the calculated mass of the reduced form of sWntx-5 (7514.5 Da). This fraction corresponded to approximately 17% of the total crude mixture. The purity of the reduced sWntx-5 toxin was assessed on an analytical C18 column (Fig. 3B). Reduced sWntx-5 was oxidized using a redox buffer containing a mixture of GSH and GSSH in a peptide : GSH : GSSH molar ratio of 1 : 10 : 1 at pH 7.8. The resulting glutathione-mediated oxidation mixture was acidified and submitted to RP-HPLC, revealing that the oxidized sWntx-5 (Fig. 3) eluted as a major component (yield ¼ 10% of the reduced form), approximately 10 min before the reduced form (Fig. 3B). Amino acid composi- tion, N-terminal amino acid sequencing up to 75% of the total length of the protein and electrospray mass analyses (mass ¼ 7504.4 Da, which is virtually identical to the calculated value) confirmed the purity and identity of the sWntx-5. Circular dichroic spectrum of sWntx-5 As shown in Fig. 4, the far UV spectrum of the sWntx-5 displayed a positive band at 196 nm and a broad negative band at 222 nm, together with a slight shoulder around 210 nm. This pattern is highly reminiscent of the presence of b-structure in proteins [48,49]. This conclusion agrees with the previous structural studies made on the other weak neurotoxins bucandin [16,28] and WTX [27]. We compared this spectrum with that previously monitored for the natural homologue NNA2/NNAM2 that is present in Taiwan cobra venom, and which differs in sequence from that of sWntx-5 by only three substitutions [10]. Fig. 1. Nucleotide sequences of cDNA-encoding weak neurotoxins in Naja sputatrix. The 3¢ ends of primers used in RT-PCR are in bold and underlined. The regions coding for the putative signal peptides (CDS) and neurotoxins are shown. The encoded amino acids are indicated in capital letters below the second base of each codon. The nucleotides that vary among isoforms are indicated (+), the stop codon is shown (*) and the variant residues are in bold. 4250 S. L. Poh et al.(Eur. J. Biochem. 269) Ó FEBS 2002 Fig. 2. Alignment of neurotoxin sequences. (A) Sequence alignment of snake weak neurotoxins. Sequences 1–18 are cobra weak neurotoxins [4–12], while sequences 19–25 are from kraits, coral snakes and mambas [13–19]. Sequences 1–4 were determined in this study. All other sequences are taken from the databanks, GenBank, SWISS-PROT or TrEMBL. For simplicity, the accession numbers, species names and references are shown. The names of cobra species are those from the original papers that described the toxin sequences. From a taxonomy viewpoint, however, some species names have changed [31]. The numbering at the top includes all insertions. Residues that are strictly conserved and 80% conserved in the respective groups of weak neurotoxins are indicated with open and shaded boxes, respectively. (B) Optimized sequence alignment of Ea and Wntx-5. The 11 functional residues previously identified for Ea are indicated (+). Those that have a homologous residue in Wntx-5 are in bold boxes, whereas those specific to Ea are underlined. Ó FEBS 2002 Synthetic weak neurotoxin (Eur. J. Biochem. 269) 4251 A similar strong negative band around 220 nm is observed for both toxins. The slightly weaker band that is observed with NNA2 can be explained by the presence of a positive signal at 208 nm. This band might correspond to the shoulder observed at 210 nm for sWntx-5. Nevertheless, the common presence of a negative band of comparable intensity around 220 nm strongly suggests that the level of b-sheet content is comparable in both toxins. We also compared the spectrum of sWntx-5 with the spectra of toxin a from N. nigricollis [50,51] a short-chain neuro- toxin, and a-cobratoxin [52], a long-chain neurotoxin, which both possess highly similar three-fingered structures [53,54]. The overall pattern displayed by these two neurotoxins clearly agrees with the presence of b-sheet structure, with a positive band around 196–199 nm and a negative trough centred around 212–216 nm. The CD spectrum displayed by sWntx-5 is globally comparable, with some differences, however. In particular, its negative band is centred at a somewhat longer wavelength. However, this is not so surprising, since the minimum wavelength associated the n-p* transition of a peptide chromophore in b-sheet structure can be shifted to 223 nm [49]. Therefore, our data indicate that sWntx-5 adopts an overall structure rich in b-sheet. Probing biological activity of sWntx-5 The ability of sWntx-5 to bind to muscular-type and a7 neuronal-type AChRs was estimated from competition experiments using, respectively, T. mamorata and a chimer- ical version of chicken a7)5HT 3 receptor expressed in HEK cells [23], and 125 I-labelled a-Bgtx as a tracer. The deduced K d with Torpedo receptor was estimated to be 180 n M (Fig. 5, Table 1). With the a7 neuronal receptor, the affinity was much lower and we have not been able to complete the competition curve, due to a lack of material (Fig. 5). Nevertheless, from the available data we assumed that 90% of the binding of 125 I-labelled a-bungarotoxin should be inhibited by 10 )2 M sWntx-5, and we calculated a theoret- ical curve based on the limited number of available points. Hence, we estimated that sWntx-5 should inhibit the binding of 125 I-labelled a-Bgtx to a7–5HT 3 receptor with a K d value of approximately 10 )4 M (Table 1). All these binding data not only indicate that sWntx-5 is a weak binder of AChR from electric organ of T. marmorata but also that it is an even weaker binder of the chicken a7 neuronal-type AChR. It was previously shown that l M concentration of the weak neurotoxin NNA2 inhibits at least 50% of the ACh- induced contraction of nerve-muscle preparations from frog [10]. Since, Wntx-5 and NNA2 shows high sequence identity (three amino acid residues different, Fig. 2A), we investi- gated the ability of sWntx-5 to block neuromuscular transmission in both isolated frog cutaneous pectoris nerve-muscle and mouse hemidiaphragm preparations, using electrophysiological techniques. In the frog nerve– muscle preparation, sWntx-5 caused no blockage of neuro- muscular activity at concentrations up to 9 l M . In contrast, the control a-cobratoxin blocked both washed out and Fig. 3. RP-HPLC of (A) crude peptide giving 3 major peaks (a, b and c) representing the 3 major products present in the crude mixture, (B) purified reduced sWntx-5 and (C) refolded sWntx-5 present in the oxi- dation medium. A Vydac C18 column (0.46 · 25 cm) was used. Elution was performed with a profile of 40% of a solution of 6 60% acetonitrile and 0.1% trifluoroacetic acid in H 2 O for 15 min, followed by a gra- dient of 40–60% in 40 min, at 1 minÆmL )1 flow rate. Protein was monitored at 214 nm. Fig. 4. Far UV CD spectra of snake neurotoxins. The spectrum of sWntx-5 was monitored between 190 nm and 250 nm, in the presence of nitrogen. The cell path-length and temperature of measurement were 0.05 mm and 20 °C, respectively. Previously described venom- derived spectra of a short neurotoxin, toxin a [50], a long neurotoxin, a-cobratoxin [58], and NNA2, another weak neurotoxin from cobra venom [10], are also shown. 4252 S. L. Poh et al.(Eur. J. Biochem. 269) Ó FEBS 2002 nonwashed out preparations at 0.2 l M in 2 min. Phrenic nerve stimulation of isolated mouse hemidiaphragms, previously treated with formamide (to uncouple excita- tion-contraction coupling), elicited action potentials at junctional areas without contraction triggered by endplate potentials. Addition of 6 l M sWntx-5 to the standard solution did not block neuromuscular transmission, even after 30-min incubation. In contrast, the potent a-cobra- toxin used as a control, on both washed out and nonwashed out preparations, blocked neuromuscular transmission by 60% at 0.6 l M and completely at 1.6 l M . Therefore, sWntx- 5atl M concentrations does not seem able to block muscle AChRs from frogs and mice. DISCUSSION A weak neurotoxin is currently defined as a protein isolated from elapid venom that possesses about 65 residues including 10 cysteines, eight of which can be readily aligned with those of the well-known three-fingered toxins [21,22,55]. That Wntxs also adopt this fold has been confirmed recently with the resolution of the X-ray and NMR structures of the Wntx called bucandin and WTX [16,27,28]. When we started this work, 22 amino acid sequences of Wntxs were known and it was clear to us that this family of proteins could be divided into two categories. The first one includes the cobra Wntxs whereas the second category involves mostly those from kraits, mambas (Dendroaspis jamesoni) and coral snakes (Micru- rus corallinus). We confirmed the homogeneous character of the subgroup of cobra Wntxs by introducing four new sequences (Wntx-5, Wntx-6, Wntx-7 and Wntx-9) derived from cDNAs isolated from venom glands of Naja sputatrix. This subgroup is highly homogenous, with few insertions or deletions and about 56% of the residues other than the half- cystines, that are strictly or highly conserved. In view of such a high degree of sequence similarities, we anticipate that all toxins from this subgroup may exert a highly similar biological function. This may be in contrast to the Wntxs from the second subgroup, which display many deviations and few conserved residues (besides the conserved half- cystines). During the past few years, a number of studies have been attempted to identify the biological function of Wntxs. Recent reports have shown that Wntxs from cobra venom are low-affinity blockers of muscular and a7 neuronal AChRs [10–12,29,56]. However, these results were deduced from experiments done with venom-derived toxins. There- fore, despite many efforts to obtain highly purified toxins, it Fig. 5. Inhibition of binding of 125 I-labelled a-bungarotoxin to (A) nicotinic acetylcholine receptor from T. marmorata and (B) chick chimeric a7 receptor (a7-V201–5HT 3 ) expressed in HEK cells by varying amounts of toxin a (N. nigricollis), a-cobratoxin (N. kaouthia) and sWntx-5. The continuous lines correspond to theoretical dose– responses fitted through the data points using the nonlinear Hill equation. Table 1. Summary of the effects of weak neurotoxins on various types of AchRs in competitive binding experiments. Data for sWntx-5 were from this study, while those of WTX have been previously reported [12,29]. ND, not determined. Ligands Types of AchRs K d ( M )IC 50 ( M ) Muscular-type AChR sWntx-5 T. marmorata 1.8 · 10 )7 1.8 · 10 )5 WTX T. californica 9.0 · 10 )8 2.2 · 10 )6 a7-neuronal AChR sWntx-5 Chick chimeric a7-V201–5HT3 (HEKcells) ± 9.0 · 10 )5a ± 9.0 · 10 )5a WTX GST-Rat a7 (1–208) fusion protein ND 4.3 · 10 )6 a Estimated K d due to lack of points at high concentrations of ligands. Ó FEBS 2002 Synthetic weak neurotoxin (Eur. J. Biochem. 269) 4253 could not be totally excluded that these low activities may have resulted from contamination by a potent neurotoxin. For example, the poorly reproducible activity of venom- derived j-bungarotoxin toward muscular AChRs, which was contaminated by a potent a-neurotoxin [30]. We therefore decided to produce an artificial Wntx and to study its activity on AChRs. In this paper, we have described the chemical synthesis of a cobra Wntx and the activity of this synthetic toxin on muscular and a7AChRs. We synthesized Wntx-5 because its amino acid sequence shares between 62% and 97% identity with other toxins from the cobra subgroup, and so it appeared to us as a potential prototype of this subgroup. Chemical synthesis of proteins of the size of Wntx-5 is now feasible, even if they possess a high density of disulfide bonds, as shown in a previous study with long and short neurotoxins [42]. Similarly, Wntx-5 has been synthesized successfully using an Fmoc-based chemical approach and the resulting synthetic toxin, named sWntx- 5, was obtained with a final yield of approximately 10% of the reduced form. Mass spectrometry and amino acid analyses indicated that the oxidized peptide had the expected chemical characteristics of the natural toxin. Also, amino acid sequencing of the first 49 residues confirmed that the sequence of sWntx-5 was identical to that expected. Since no native toxin was available, it was not possible to compare the chromatographic behaviour of sWntx-5 with that of the wild-type toxin. However, inspection of the far-UV CD spectrum of sWntx-5 recorded between 205 nm and 250 nm strongly confirms that it adopts a structure rich in b-sheet. We have not identified the pairings of the cysteines of sWntx-5. However, we assumed that they correspond to the expected ones because it has been shown repeatedly that the presence of the conserved disulphides of all three- fingered toxins is indispensable for their fold to be acquired [22]. Wntxs isolated from cobra venom have been described as poor blockers of muscular-type AChRs [10–12,29,56,]. Thus, using preparations of AChR from Torpedo califor- nica, a weak neurotoxin from Naja kaouthia (WTX) was found to inhibit binding of radioactive a-bungarotoxin with apparent K d values around 90 n M [12,29]. In close agreement with this observation, sWntx-5 inhibits binding of radioactive a-bungarotoxin to AChRs from T. marmo- rata,withaK d value of 180 n M . That these results agree so well confirms the view that a Wntx from cobra venom can bind with moderate affinity to muscular type AChRs, at least in vitro. Though acting as a binder of muscular-type AChR, the Wntx from N. kaouthia was nontoxic to rodents, even when high doses (2 mgÆkg )1 ) were adminis- tered by intravenous injection. Due to a lack of material, we have not tested the toxic activity in vivo of sWntx-5. Instead, we investigated its ability to block neuromuscular transmission in both isolated mouse hemidiaphragm and isolated frog cutaneous pectoris muscle, using electrophys- iological techniques. We found that 6 l M of sWntx-5 failed to block neuromuscular transmission in mouse phrenic nerve hemidiaphragm muscle. Previously, it was reported that NNA2, a weak neurotoxin from the Formosan cobra, inhibits ACh-induced contraction of frog muscle prepara- tions, with IC 50 concentrations around 1–4 l M [10]. In contrast, 3–9 l M of sWntx-5 caused no effect on a stimulated frog cutaneous pectoris nerve muscle toxin preparation. The toxin also had no effect on the more sensitive miniature endplate potentials. Therefore, although sWntx-5 and NNA2 share a high degree of sequence identity (Fig. 2), they behave differently in the frog cutaneous pectoris nerve–muscle experiments. This situ- ation could be due to one or more of the three mutations that differentiate the two toxins, or to differences in the experimental protocols, such as, for example, the use of different frog species that may discriminate between neurotoxins [60]. It has also been shown that the Wntx from N. kaouthia is an antagonist of human and rat a7 AChRs [29]. In vitro binding experiments and electrophys- iological assays showed that this toxin has a low affinity (IC 50 ) in the range of 4–15 l M for these receptors. On the basis of competition binding experiments with a chimerical version of chicken a7–5HT 3 receptor, we found that sWntx-5 has an affinity (IC 50 )closeto90l M for this receptor. This is 6–22 times lower than that observed for WTX from N. kaouthia. Considering that the two toxins display 11 residue differences and that the competition systems used (human and rat on one hand, and chicken on the other) are not identical in the two studies, the two toxins appear to behave as comparable weak antagonists of neuronal a7 receptors. Do cobra Wntxs and the potent a-neurotoxins bind to muscular AChRs using similar determinants? To address this question, the sequence of sWntx-5 was optimally aligned with that of erabutoxin a (Ea), a short chain and potent neurotoxin from sea snake that possesses 11 functionally important residues [32,33] (Fig. 2B). Five of these amino acids (shown in bold) are observed at homologous positions in Wntx-5. These are Lys29 (homologous to Lys27 in Ea), Phe36 (Phe32), Arg39 (Arg33), Arg42 (Ile36), and Lys52 (Lys47). Note that mutation of Ile36 into an Arg increases the affinity of Ea for the muscular receptor by 7-fold [33] and that an arginine is found in Wntx-5 at this location. Therefore, if we assume that these common residues have a comparable binding function in both toxins, sWntx-5 appears to lack six of the 11 functional residues of Ea, which may explain its low potency to muscular AChRs. In agreement with our observation that sWntx-5 binds with a very low affinity to the neuronal a7 receptor, we found only two residues (Phe36 and Arg39) whose positions could be aligned with those identified to be critical for this particular binding in a-cobratoxin. Another intriguing question concerns the significance of a 180-n M affinity for a toxin isolated from a venom, which also possesses toxins acting on the same target with much higher affinities (with K d s varying from n M to p M ). It has been shown that despite their low affinities, some weak neurotoxins can be slow-dissociating proteins [17,56]. This might also be the case for sWntx-5. What is the role of the disulfide bond that is uniquely present in the first loop of the weak neurotoxins? Previously, it was demonstrated that the additional disulfide that is present in the second loop of the long neurotoxins is specifically involved in the capacity of these toxins to interact with a7 neuronal receptors [23,24,26,57]. We suggest therefore that the disulfide bond that is found in the first loop of Wntxs may be associated with a binding to a specific tissue target, which however, remains to be identified. 4254 S. L. Poh et al.(Eur. J. Biochem. 269) Ó FEBS 2002 ACKNOWLEDGEMENTS This work was supported by research grants from CEA and National University of Singapore (RP 960324). S. L. Poh is a research scholar of NUS and received scholarships from NUS (Singapore), ARET (France) and EGIDE (France). REFERENCES 1. Sto ¨ cklin, R., Mebs, D., Boulain, J.C., Panchaud, P.A., Virelizier, H. & Gillard–Factor, C. (2000) Identification of snake species by toxin mass fingerprinting of their venoms. Methods Mol Biol. (2000) 146, 317–335. 2. Sherman, N., Shannon, J., Gallagher, P., Dragulev, B., Kamiguti, A.S., Theakston, R.D.G., Bland, L. & Fox, J.W. (2000) Discovery Science in toxinology: the genomic/proteomic interface in venom research. 13th World Congress on Animal, Plant and Microbial Toxins,Paris. 3. Dufton, M.J. & Hider, R.C. (1983) Conformational properties of the neurotoxins and cytotoxins isolated from Elapid snake venoms. CRC Crit Rev. Biochem. 14, 113–171. 4. Carlsson, F.H.H. (1975) Snake venom toxins: the primary struc- ture of protein S 4 C 11 . A neurotoxin homologue from the venom of forest cobra (Naja melanoleuca). Biochim. Biophys. Acta 400, 310– 321. 5. Joubert, F.J. (1975) The purification and amino acid sequence of toxin CM-13b from Naja haje annulifera (Egyptian cobra) venom. Hoppe Seylers Z. Physiol. Chem. 356, 1901–1908. 6. Joubert, F.J. & Talijaard, N. (1978) Naja haje haje (Egyptian cobra) venom. Some properties and the complete primary struc- ture of three toxins (CM-2, CM-11 and CM-12). Eur. J. Biochem. 90, 359–367. 7. Joubert, F.J. & Talijaard, N. (1980) Snake venoms. The amino acid sequences of two Melanoleuca-type toxins. Hoppe Seylers Z. Physiol. Chem. 361, 425–436. 8. Shafqat, J., Siddiqi, A.R., Zaidi, Z.H. & Jornvall, H. (1991) Extensive multiplicity of the miscellaneous type of neurotoxins from the venom of the cobra Naja naja naja and structural char- acterization of major components. FEBS Lett. 284, 70–72. 9. Qian, Y.C., Fan, C.Y., Gong, Y. & Yang, S L. (1998) cDNA sequence analysis and expression of four long neurotoxin homo- logues from Naja naja atra. Biochim. Biophys. Acta 1443, 233–238. 10. Chang, L., Lin, S., Wang, J., Hu, W.P., Wu, B. & Huang, H. (2000) Structure-function studies on Taiwan cobra long neuro- toxin homolog. Biochim. Biophys. Acta 1480, 293–301. 11. Lin, S.R., Huang, H.B., Wu, B.N. & Chang, L.S. (1998) Char- acterization and cloning of long neurotoxin homolog from Naja naja atra. Biochem. Mol. Biol. Int. 46, 1211–1217. 12. Utkin, Y.N., Kukhtina, V.V., Maslennikov, I.V., Eletsky, A.V., Starkov, V.G., Weise, C., Franke, P., Hucho, F. & Tsetlin, V.I. (2001) First tryptophan-containing weak neurotoxin from cobra venom. Toxicon 39, 921–927. 13. Qian, Y C., Fan, C Y., Gong, Y. & Yang, S L. (1998) cDNA cloning and sequence analysis of six neurotoxin-like proteins from Chinese continental banded krait. Biochem. Mol. Biol. Int. 46, 821–828. 14. Chang, L S. & Lin, J. (1997) cDNA sequence of a novel neuro- toxin homolog from Taiwan banded krait. Biochem. Mol. Biol. Int. 43, 347–354. 15. Aird, S.D., Womble, G.C., Yates, J.R. & Griffin, P.R. (1999) Pri- mary structure of c-bungarotoxin, a new postsynaptic neurotoxin from venom of Bungarus multicinctus. Toxicon 37, 609–625. 16. Khun,P.,Deacon,A.M.,Comoso,S.,Rajaseger,G.,Kini,R.M., Uson, I. & Kolatkar, P.R. (2000) The atomic resolution structure of bucandin, a novel toxin isolated from the Malayan krait, determined by direct methods. Acta Crystallogr. D Biol. Crystal- logr. 56, 1401–1407. 17. Nirthanan, S., Charpantier, E., Gopalakrishnakone, P., Gwee, M.C., Khoo, H.E., Cheah, L.S., Bertrand, D. & Kini, R.M. (2002) Candoxin, a novel toxin from Bungarus candidus is a reversible antagonist of muscle (abcd) but a poorly reversible antagonist of neuronal alpha 7 nicotinic acetylcholine receptors. J. Biol. Chem. 277, 17811–17820. 18. Ho, P.L., Soares, M.B., Yamane, T. & Raw, I. (1995) Reverse biology applied to Micrurus corallinus, a South American coral snake. J. Toxicol. Toxin Rev. 14 (3), 309–326. 19. Joubert, F.J. & Taljaard, N. (1979) Complete primary structure of toxin S 6 C 4 from Dendroaspis jamesoni kaimosae (Jameson’s mamba). S. Afr. J. Chem. 32, 151–155. 20. Me ´ nez, A. (1993) Les structures des toxins des animaux venimeux. Pour Sci. 190, 34–40. 21. Ohno, M., Me ´ nez,R.,Ogawa,T.,Danse,J.M.,Shimohigashi,Y., Fromen, C., Ducancel, F., Zinn-Justin, S., Du Le, M.H., Boulain, J C., Tamiya, T. & Me ´ nez, A. (1998) Molecular evolution of snake toxins: is the functional diversity of snake toxins associated with a mechanism of accelerated evolution? Prog. Nucleic Acid Res. Mol. Biol. 59, 307–364. 22. Servent, D. & Me ´ nez, A. (2001) Snake toxins that interact with nicotinic acetylcholine receptors. In Neurotoxicological Handbook Vol. I (Massaro, E.J., ed.), Humana Press, Totowa, NJ. 23. Servent, D., Winckler-Dietrich, V., Hu, H.Y., Kessler, P., Drevet, P., Bertrand, D. & Me ´ nez, A. (1997) Only snake curaremimetic toxins with a fifth disulfide bond have high affinity for the neu- ronal a 7 nicotinic receptor. J. Biol. Chem. 272, 24279–24286. 24. Servent. D., Thanh, H.L., Antil, S., Bertrand, D., Corringer, P.J., Changeux,J.P.&Me ´ nez, A. (1998) Functional determinants by which snake and cone snail toxins block the alpha 7 neuronal nicotinic acetylcholine receptors. J. Physiol. Paris 92, 107–111. 25. Grant, G.A., Luetje, C.W., Summers, R. & Xu, X.L. (1998) Dif- ferential roles for disulfide bonds in the structural integrity and biological activity of j-Bungarotoxin, a neuronal nicotinic acet- ylcholine receptor antagonist. Biochemistry 37, 12166–12171. 26. Antil-Delbeke, S., Gaillard, C., Tamiya, T., Corringer, P J., Changeux, J.P., Servent, D. & Me ´ nez, A. (2000) Molecular determinants by which a long chain toxin from snake venom interacts with the neuronal alpha 7-nicotinic acetylcholine receptor. J. Biol. Chem. 275, 29594–29601. 27. Eletskii, A.V., Maslennikov, I.V., Kukhtina, V.V., Utkin IuN., Tsetlin,V.I.&Arsen’ev,A.S.(2001)Structureandconformational heterogeneity of the weak toxin from the cobra Naja kaouthia venom. Bioorg Khim 27, 89–101. 28. Torres, A.M., Kini, R.M., Selvanayagam, N. & Kuchel, P.W. (2001) NMR structure of bucandin, a neurotoxin from the venom of the Malayan krait (Bungarus candidus). Biochem. J. 360, 539– 548. 29. Utkin, Y.N., Kukhtina, V.V., Kryukova, E.V., Chiodini, F., Bertrand,D.,Methfessel,C.&Tsetlin,V.I.(2001)ÔWeak toxinÕ from Naja kaouthia is a nontoxic antagonist of alpha 7 and muscle-type nicotinic acetylcholine receptors. J. Biol. Chem 276, 15810–11815. 30. Fiordalisi, J.J., Al-Rabiee, R., Chiappinelli, V.A. & Grant, G.A. (1994) Affinity of native j-bungarotoxin and site directed mutants for the muscle nicotinic acetylcholine receptor. Biochemistry 33, 12963–12967. 31. Wu ¨ ster, W. (1996) Taxonomic changes and toxinology: systematic revisions of the Asiatic cobras (Naja naja species complex). Toxicon 34, 399–406. 32. Pillet, L., Tre ´ meau, O., Ducancel, F., Drevet, P., Zinn-Justin, S., Pinkasfeld, S., Boulain, J C. & Me ´ nez, A. (1993) Genetic engineering of snake toxins. Role of invariant residues in the structural and functional properties of a curaremimetic toxin, as probed by site-directed mutagensis. J. Biol. Chem. 268, 909–916. 33. Tre ´ meau, O., Lemaire, C., Drevet, P., Pinkasfeld, S., Ducancel, F., Boulain, J C. & Me ´ nez, A. (1995) Genetic engineering of snake Ó FEBS 2002 Synthetic weak neurotoxin (Eur. J. Biochem. 269) 4255 toxins, the functional site of erabutoxin a as delineated by site- directed mutagensis, includes variant residues. J. Biol. Chem. 268, 9362–9369. 34. Yanish-Perron, C., Vieire, J. & Messing, J. (1985) Improved M13 phage cloning vectors and host strains: nucleotide sequences of M13mp18 and pUC19 vectors. Gene 33, 103–199. 35. Afifiyan, F., Armugam, A., Tan, C.H., Gopalakrishnakone, P. & Jeyaseelan, K. (1999) Postsynaptic alpha-neurotoxin gene of the spitting cobra, Naja naja sputatrix: structure, organization, and phylogenetic analysis. Genome Res. 9, 259–366. 36. Sambrook, J., Fritsch, E.F. & Maniatis, T. (1989) Molecular cloning: A Laboratory Manual, 2nd edn. Cold Spring Harbor Laboratory Press. 37. Weichenhan, D. (1991) Fast recovery of DNA from agarose gel by centrifugation through blotting paper. Trends Genet 7, 109. 38. Sanger, F., Nicklen, S. & Coulson, A.R. (1977) DNA sequencing with chain-terminating inhibitors. Proc. Natl Acad. Sci. USA 74, 5463–5467. 39. Altschul, S.F., Gish, W., Miller, W., Myers, E.W. & Lipman, D.J. (1990) Basic local alignment search tool. J. Mol. Biol. 215,403– 410. 40. Riniker, B., Flo ¨ rsheimer, A., Fretz, H., Sieber, P. & Kamber, B. (1993) A general strategy for the synthesis of large peptides: 5 the combined solid-phase and solution approach. Tetrahedron 49, 9307–9320. 41. Wang, S S. (1973) p-Alkoxybenzyl alcohol resin and p-alkoxy benzyloxycarbonylhydrazide resin for solid phase synthesis of protected peptide fragments. J. Am. Chem. Soc. 94, 1328. 42. Mourier, G., Servent, D., Zinn-Justin, S. & Me ´ nez, A. (2000) Chemical engineering of a three-fingered toxin with anti-a7 neuronal acetylcholine receptor activity. Protein Eng. 13,217– 225. 43. Saitoh, T., Oswald, R., Wennogle, L.P. & Changeux, J.P. (1980) Conditions for the selective labelling of the 66 000 dalton chain of the acetylcholine receptor by the covalent non-competitive blocker 5-azido-[ 3 H]trimethisoquin. FEBS Lett. 116, 30–36. 44. Cheng, Y.C. & Prusoff, W.H. (1973) Relationship between the inhibition constant (K i ) and the concentration of inhibitor, which causes 50 per cent inhibition of an enzymatic reaction. Biochem. Pharmacol. 22, 3099–3108. 45. Weber, M. & Changeux, J P. (1974) Binding of Naja nigricolis [ 3 H]a-toxin to membrane fragments from Electrophorus and Torpedo electric organs. Mol. Pharmacol. 10, 15–34. 46. Favreau,P.,Krimm,I.,LeGall,F.,Bobenrieth,M.J.,Lamthanh, H.,Bouet,F.,Servent,D.,Molgo,J.,Me ´ nez, A., Letourneux, Y. & Lancelin, J.M. (1999) Biochemical characterization and nuclear magnetic resonance structure of novel alpha-conotoxins isolated from the venom of Conus consors. Biochemistry 38, 6317– 6326. 47. Blobel, G. & Dobberstein, B. (1975) Transfer of proteins across membrane. I-Presence of proteolytically processed and unprocessed nascent immunoglobulin light chains on membrane- bound ribosomes of murine myeloma. J. Cell Biol. 67, 852–862. 48. Woody, R.W. (1995) Circular Dichroism. Methods Enzymol. 246, 34–71. 49. Sreerama, N., Manning, M.C., Powers, M.E., Zhang, J.X., Goldenberg,D.P.&Woody,R.W.(1999)Tyrosine,phenylala- nine, and disulfide contributions to the circular dichroism of proteins: circular dichroism spectra of wild-type and mutant bovine pancreatic trypsin inhibitor. Biochemistry 38, 10814–10822. 50. Me ´ nez, A., Bouet, F., Tamiya, N. & Fromageot, P. (1976) Conformational changes in two neurotoxic proteins from snake venoms. Biochim. Biophys. Acta 26, 121–132. 51. Me ´ nez, A., Langlet, G., Tamiya, N. & Fromageot, P. (1978) Conformation of snake toxic polypeptides studied by a method of prediction and circular dichroism. Biochimie 60, 505–516. 52. Hider,R.C.,Drake,A.F.&Tamiya,N.(1988)Ananalysisofthe 225–230-nm CD band of elapid toxins. Biopolymers 27, 113–122. 53. Zinn-Justin, S., Roumestand, C., Gilquin, B., Bontems, F., Me ´ nez, A. & Toma, F. (1992) Three-dimensional solution struc- ture of a curaremimetic toxin from Naja nigricollis venom: a proton NMR and molecular modeling study. Biochemistry 31, 11335–11347. 54. Le Goas, R., LaPlante, S.R., Mikou, A., Delsuc, M.A., Guittet, E., Robin, M., Charpentier, I. & Lallemand, J.Y. (1992) Alpha- cobratoxin: proton NMR assignments and solution structure. Biochemistry 31, 4867–4875. 55. Me ´ nez, A. (1998) Functional architectures of animal toxins: a clue to drug design? Toxicon 36, 1557–1572. 56. Vulfius, C.A., Krasts, I.V., Utkin, Y.N. & Tsetlin, V.I. (2001) Nicotinic receptors in Lymnea stagnalis neurons are blocked by alpha-neurotoxins from cobra venoms. Neurosci. Lett. 309,189– 192. 57. Servent, D., Antil-Delbeke, S., Gaillard, C., Corringer, P J., Changeux, J.P. & Me ´ nez, A. (2000) Molecular characterization of the specificity of interactions of various neurotoxins on two dis- tinct nicotinic acetylcholine receptors. Eur. J. Pharmacol. 393, 197–204. 58. Antil, S., Servent, D. & Me ´ nez, A. (1999) Variability among the sites by which curaremimetic toxins bind to torpedo acetylcholine receptor, as revealed by identification of the functional residues of alpha-cobratoxin. J. Biol. Chem. 274, 34851–34858. 59. Sato, S. & Tamiya, N. (1971) The amino acid sequence of erabutoxins, neurotoxic proteins of sea-snake (Laticauda semi- fasciata)venom.Biochem. J. 122, 453–461. 60. Chang, C.C. (1979) The action of snake venom on nerve and muscle. In Snake Venoms, Handbook of Experimental Pharma- cology (Lee, C.Y., ed.), pp. 309–376. 4256 S. L. Poh et al.(Eur. J. Biochem. 269) Ó FEBS 2002 . two toxins appear to behave as comparable weak antagonists of neuronal a7 receptors. Do cobra Wntxs and the potent a- neurotoxins bind to muscular AChRs. A synthetic weak neurotoxin binds with low affinity to Torpedo and chicken a7 nicotinic acetylcholine receptors Siew Lay Poh 1,2 , Gilles