An ‘Old World’ scorpion b-toxin that recognizes both insect and mammalian sodium channels A possible link towards diversification of b-toxins Dalia Gordon 1 , Nitza Ilan 1,6 , Noam Zilberberg 2 , Nicolas Gilles 3 , Daniel Urbach 1 , Lior Cohen 1 , Izhar Karbat 1 , Oren Froy 1 , Ariel Gaathon 4 , Roland G. Kallen 5 , Morris Benveniste 6 and Michael Gurevitz 1 1 Department of Plant Sciences, George S. Wise Faculty of Life Sciences, Tel-Aviv University, Israel; 2 Department of Life Sciences, Ben-Gurion University, Israel; 3 CEA, De ´ partment d’Inge ´ nie ´ rie et d’Etudes des Prote ´ ines, France; 4 Bletterman Research Laboratory for Macromolecules, The Hebrew University-Hadassah, Medical School Jerusalem, Israel; 5 Department of Biochemistry and Biophysics, University of Pennsylvania School of Medicine, Philadelphia, PA, USA; 6 Department of Physiology and Pharmacology, Sackler School of Medicine, Tel-Aviv University, Israel Scorpion toxins that affect sodium channel (NaCh) gating in excitable cells are divided into a- and b-classes. Whereas a-toxins have been found in scorpions throughout the world, anti-mammalian b-toxins have been assigned, thus far, to ÔNew WorldÕ scorpions while anti-insect selective b-toxins (depressant and excitatory) have been described only in the ÔOld WorldÕ. 2 This distribution suggested that diversification of b-toxins into distinct pharmacological groups occurred after the separation of the continents, 150 million years ago. We have characterized a unique toxin, Lqhb1, from the ÔOld WorldÕ scorpion, Leiurus quinquestriatus hebraeus,that resembles in sequence and activity both ÔNew WorldÕ b-toxins as well as ÔOld WorldÕ depressant toxins. Lqhb1 competes, with apparent high affinity, with anti-insect and anti-mammalian b-toxins for binding to cockroach and rat brain synaptosomes, respectively. Surprisingly, Lqhb1also competes with an anti-mammalian a-toxin on binding to rat brain NaChs. Analysis of Lqhb1 effects on rat brain and Drosophila Para NaChs expressed in Xenopus oocytes revealed a shift in the voltage-dependence of activation to more negative membrane potentials and a reduction in sodium peak currents in a manner typifying b-toxin activity. Moreover, Lqhb1 resembles b-toxins by having a weak effect on cardiac NaChs and a marked effect on rat brain and skeletal muscle NaChs 3 . These multifaceted features suggest that Lqhb1 may represent an ancestral b-toxin group in ÔOld WorldÕ scorpions that gave rise, after the separation of the continents, to depressant toxins in ÔOld WorldÕ scorpions and to various b-toxin subgroups in ÔNew WorldÕ scorpions. Keywords: scorpion toxins; sodium channel subtypes; toxin diversification. Scorpion Ôlong chainÕ toxins affecting voltage-gated sodium channels (NaCh) are polypeptides of 61–76 amino acids long that traditionally are divided between two major classes, a and b, according to their physiological effects on channel gating and their binding properties [1–3]. a-Toxins slow sodium channel inactivation upon binding to a homologous cluster of binding sites named receptor site-3, and are subdivided into distinct groups according to their potency for mammalian and insect receptors and their affinity for sodium channel subtypes [2,4–7]. a-Toxins predominate in the venom of Buthidae scorpions of the ÔOld WorldÕ (Africa and Asia), but some representatives have been also identified in ÔNew WorldÕ (America) scorpions [1]. b-Toxins shift the activation voltage of sodium channels to more negative membrane potentials upon binding to receptor site-4 [2,7,8], and vary greatly in their effects on various animals. Css2 and Css4 (from Centruroides suffusus suffusus) show specificity for mammals [1]; Cll1 (from C. limpidus limpidus), Cn5, and Cn11 (from C. noxius) are highly effective on crustaceans [9–12]; Ts7 and Tst1 (from Tityus serrulatus and T. stig- murus), and Tbs1 and Tbs2 (from T. bahiensis)are highly effective on both insects and mammals [1,12–16]. b-Toxins, active on mammals, have thus far been assigned to scorpions of the ÔNew WorldÕ (Tityus and Centruroides species), whereas depressant and excitatory b-toxins, which modify exclusively the activation of insect sodium channels, have been found strictly in ÔOld WorldÕ Buthoids [1,2,12,17,18] 4 . In addition to their effects on sodium channel gating, classification of toxins to either the a- or b-classrelieson competition binding assays against the ÔOld WorldÕ toxin, Aah2, from Androctonus australis hector,andtheÔNew WorldÕ toxin, Css2, from Centruroides suffusus suffusus, respectively [1,2]. Further distinction between toxins can be made using competition binding assays utilizing LqhaIT Correspondence to D. Gordon or M. Gurevitz, Department of Plant Sciences, George S. Wise Faculty of Life Sciences, Tel-Aviv University, Ramat-Aviv 69978, Tel-Aviv, Israel. 1 Fax: + 972 3 6406100, Tel.: + 972 3 6409844, E-mail: dgordon@post.tau.ac.il or mamgur@post.tau.ac.il Abbreviations:Lqhb1, Leiurus quinquestriatus hebraeus beta toxin 1; LqhIT2, anti-insect selective depressant toxin; LqhaIT, anti-insect a-toxin; Lqh2, anti-mammalian a-toxin; Css2, Css4, Centruroides suffusus suffusus b-toxins 2 and 4; Ts7, Tityus serrulatus toxin 7 (also called Ts1 and c-toxin); NaCh, sodium channel. (Received 27 March 2003, accepted 29 April 2003) Eur. J. Biochem. 270, 2663–2670 (2003) Ó FEBS 2003 doi:10.1046/j.1432-1033.2003.03643.x (from Leiurus quinquestriatus hebraeus)forthea-anti-insect and Lqh3 for the a-like groups [4,5]; Bj-xtrIT (from Buthotus judaicus [19]) and AahIT [17] for the anti-insect selective excitatory, and LqhIT2 for the depressant b-toxin groups [2,20]. The ability of excitatory and depressant anti-insect selective toxins to compete with ÔNew WorldÕ b-toxins, such as Ts7, suggests that common features may exist in their receptor sites on various insect and mammalian sodium channels [1,2,21,22]. As b-toxins that affect mammalian sodium channels have not been identified in the ÔOld WorldÕ, it was assumed that diversification of anti-mammalian b-toxins in the ÔNew WorldÕ, and excitatory and depressant toxins in the ÔOld WorldÕ occurred from an unknown ancestral progenitor after the separation of the continents 150 million years ago. Yet, an ÔOld WorldÕ scorpion toxin, AahIT4, with high affinity for the AahIT binding site on insect neuronal membranes, could also compete with a- (Aah2) and b- (Css2) anti-mammalian toxins for binding to rat brain synaptosomes [23]. As AahIT4 shares little sequence similarity with any of the known anti-mammalian scorpion toxins [1,12], and no information was available on its mode of action, it was considered a unique member of a new pharmacological group of neither a- nor b-type toxins [12,23]. More recently, two toxins have been purified from the Asian scorpion, Buthus martensii Karsch, BmK AS and BmK AS-1, which share 80 and 86% sequence identity with AahIT4 [24]. These toxins are weakly toxic to insects, are not toxic to mice and inhibit Na + currents in neurons of rat dorsal root ganglia [25]. Still, no pharmacological details that allow their classification to a- or b-toxins have been provided. Here we report the isolation and characterization of an ÔOld WorldÕ toxin, Lqhb1 that probably belongs to the same group as AahIT4 if sequence similarity and binding features are examined 5 .Lqhb1 competes with both a- and b-toxins for binding to rat brain synaptosomes and with excitatory anti-insect selective toxins in insect neuronal membranes. We show that Lqhb1 affects insect and mammalian NaCh subtypes in a manner that typifies ÔNew WorldÕ b-toxins. This suggests that b-toxins affecting mammalian NaChs have existed and are still present in ÔOld WorldÕ scorpions. The effects of Lqhb1 on various NaCh subtypes may suggest that this toxin represents an ancient group of b-toxins that gave rise to the anti-insect depressant toxins in the ÔOld WorldÕ andtotheb-toxins active on mammals in the ÔNew WorldÕ. Experimental procedures Biological material Venom from Leiurus quinquestriatus hebraeus was collected from scorpion stings to a parafilm membrane. Sarcophaga falculata (blowfly) larvae and Periplaneta americana (cock- roaches) were bred in the laboratory. Albino laboratory ICR mice were purchased from the Levenstein farm in Yokneam, Israel. As purified Lqhb1 was obtained in a limited amount, the toxin was also purchased from Latoxan (LTx-003; Valance, France) together with the anti-mam- malian a-toxin, Lqh2. Purification and analysis of Lqhb1 Toxin purification was carried out by conventional chro- matographic procedures described previously [19,20]. Briefly, crude venom (6 mg) from 30 scorpions was lyophilized, dissolved in 3 mL of 10 m M ammonium acetate pH 6.7, and subjected to two chromatographic steps: anion- exchange chromatography on a 0.8-mL DEAE-Sephadex column (Sigma, USA) equilibrated in 10 m M ammonium acetate pH 8.0. A linear gradient of 0.01–1.0 M ammonium acetate pH 6.7 at a flow rate of 0.5 mLÆmin )1 at room temperature was applied and 1.5 mL aliquots were collected and lyophilized. Several fractions eluted by 275–375 m M ammonium acetate that were toxic to blowfly larvae were combined and further purified via HPLC using an analytical C 18 column (250 · 10 mm; Vydac, USA). Sample was loaded in 0.1% trifluoroacetic acid in water (Buffer A) and eluted with a stepwise increasing gradient of 0.1% trifluoro- acetic acid in acetonitrile (Buffer B) at a flow rate of 1mLÆmin )1 . The protein eluted after 19.5 min and con- tained 163 lg pure polypeptide of 7463 Da (determined by Electrospray Mass Spectrometry, Technion, Haifa, Israel) and exerted depressant activity on blowfly larvae. Amino acid sequence analysis, carried out by automated Edman degradation using an Applied Biosystem (Foster City, CA, USA) gas-phase sequencer (470 A) connected to its corres- ponding PTH-analyzer (120 A) and data system (900 A) identified the first 24 N-terminal residues. This amino acid sequence was used to pull out the entire cDNA clone using Ôback-to-backÕ oligonucleotide primers in a PCR technique described by Zilberberg and Gurevitz [ 6 26]. Briefly, two degenerate oligonucleotides, designed according to the protein sequence (Primer 1: 5¢-ANACYTTPCANCC HGTNGC-3¢;Primer2:5¢-GGTGYGTNATHGAYGA YGC-3¢; N stands for either A, G, T or C; Y for C or T; PforAorG;HforA,T,orC;Fig.1A),wereusedas primers for PCR (MJ Research thermocycler, USA) with L. q. hebraeus cDNA library [27] as DNA template to amplify the entire Lqhb1-cDNA. Reaction conditions were: 30 cycles of 1 min at 94 °C, 1 min 50 °Cand1minat 72 °C. The PCR product was blunt-ended, phosphorylated, cloned into the SmaI site of pBluescript, and subjected to sequence analysis using Sequenase II (United States Biochemicals). The cloned gene was labeled with 32 Pand used as a probe to pull out by colony hybridization the original cDNA from the library (Fig. 1). Recombinant toxin production Bj-xtrIT and LqhIT2 were produced in Escherichia coli strain BL21 and reconstituted by in vitro folding as was described previously [19,28]. A synthetic gene was used to produce Css4 b-toxin (I. Karbat, D. Gordon & M. Gurevitz, unpublished observation) 7 following the procedure described for LqhIT2 [28]. Toxicity assays Five toxin concentrations were tested using four-day-old blowfly larvae ( 100 mg body weight). Ten larvae were injected intersegmentally at the rear side with each toxin concentration in three independent experiments. A positive 2664 D. Gordon et al. (Eur. J. Biochem. 270) Ó FEBS 2003 result was scored when a characteristic paralysis (transient immobilization and contraction replaced by gradually increasing flaccidity) was obtained and lasted at least 15 min. ED 50 values were calculated as was described previously [28]. Binding experiments Insect synaptosomes were prepared from whole heads of adult P. americana according to a previously described method [19]. Mammalian brain synaptosomes were pre- pared from adult albino Sprague–Dawley rats ( 300 g, laboratory bred), as was described previously 8 [29]. Mem- brane protein concentration was determined using a Bio- Rad Protein Assay, using bovine serum albumin (BSA) as a standard. Lqh2, LqhaIT, Bj-xtrIT and Css4 were radio- iodinated by iodogen (Pierce, Rockford, USA) using 5 lg toxin and 0.5 mCi carrier-free Na 125 I(Amersham,UK)and the monoiodotoxins were purified using an analytical Vydac RP-HPLC C 18 column, as was described previously [4,30]. The concentration of the radiolabeled toxin was determined according to the specific activity of the 125 I corresponding to 2500–3000 d.p.m.Æfmol )1 of monoiodotoxin, depending on the age of the radiotoxin and by estimation of its biological activity (usually 50–70%; see [30] for details). The compo- sition of the medium used in the binding assays was (in m M ): choline Cl, 130; CaCl 2 ,1.8;KCl,5;MgSO 4 ,0.8,Hepes50; Glucose 10, and 2 mgÆmL )1 BSA, pH 7. Wash buffer composition was (in m M ): choline Cl, 140; CaCl 2 ,1.8;KCl, 5.4; MgSO 4 ,0.8;Hepes,50;5mgÆmL )1 BSA, pH 7.5. Binding assays were performed as was described previously [29,30]. Nonspecific toxin binding was determined in the presence of a high concentration of unlabeled toxin, as specified in figure legends, and consisted typically of 5–15% of total binding. Equilibrium competition binding assays were performed using increasing concentrations of the unlabeled toxins in the presence of a constant low con- centration of [ 125 I]toxins, and analyzed by the computer program KALEIDAGRAPH (Synergy Software, Reading, PA, USA) using a nonlinear fit to the Hill equation (for IC 50 determination). The K i were calculated by the equation K i ¼ IC 50 /[1 + (L*/K d )], where L* is the concentration of hot toxin and K d is its dissociation constant. Each experiment was performed in duplicate samples and repeated multiple timesasindicated(n)foreachK i value. Bj-xtrIT excitatory toxin, a marker of receptor site-4 in insect sodium channels [19,21], was used in competition binding assays of LqhIT2 and Lqhb1 toxins to cockroach neuronal membranes. Sodium channel expression and two-electrode voltage-clamp assays using Xenopus oocytes cRNAs encoding rat skeletal muscle (Na v 1.4; rSkM1), rat brain IIa (rNa v 1.2a; rBIIA), human heart (hNa v 1.5; hH1) subtypes and insect Drosophila Para (DmNa v 1; gift from J. Warmke, Merck, New Jersey, USA) 9 sodium channel a-subunits, and the auxiliary human b1 and insect TipE subunits (gift from M. S. Williamson, IACR-Rothamsted, 10 UK), were transcribed in vitro usingT7RNApolymerase and the mMESSAGE mMACHINE TM system (Ambion, USA [31,32]); and were injected into Xenopus laevis oocytes as described by Shichor et al. [21]. One to four days after injection, currents were measured by two-electrode voltage clamp using a Gene Clamp 500 amplifier (Axon Instru- ments, Union City, CA, USA). Data were sampled at 10 kHz and filtered at 5 kHz. Data acquisition was con- trolled by a Macintosh PPC 7100/80 computer, equipped with ITC-16 analog/digital converter (Instrutech Corp., Fig. 1. Nucleotide sequence of the cDNA clone encoding Lqhb1 and its deduced amino acid sequence. The putative signal sequence is underlined and a polyadenylation signal appears in lowercase letters. The amino acid stretch used for design of back-to-back oligo- nucleotide primers is indicated by arrows 1 and 2 (see Experimental procedures for sequence). The technique used for cloning has been described previously [26]. Ó FEBS 2003 A novel ÔOld WorldÕ scorpion b-toxin (Eur. J. Biochem. 270) 2665 Port Washington, NY, USA), utilizing SYNAPSE (Synergistic Systems, Sweden). Capacitance transients and leak currents were removedbysubtracting ascaled control traceutilizing a P/6protocol[32].Bathsolution contained (in m M ):NaCl, 96; KCl, 2; MgCl 2 ,1;CaCl 2 ,1.8;Hepes,5;pH 7.85.Toxinswere diluted with bath solution containing 1 mgÆmL )1 BSA. OocyteswerewashedwithbathsolutionflowingfromaBPS- 8 perfusion system (ALA Scientific Instruments, Westbury, NY, USA) with a positive pressure of 4 psi. Approximately 1 mL of toxin-containing solution was perfused over the oocyte situated in a 200-lL chamber at room temperature. Results Purification and cloning of a functionally unique b-toxin from L. q. hebraeus Lqhb1 was purified from the venom of L. q. hebraeus by cation-exchange chromatography followed by RP-HPLC (see Experimental procedures for details). Upon injection of Lqhb1 to blowfly larvae, a short, transient contraction was observed followed by a dose-dependent, long-lasting flaccid paralysis (ED 50 ¼ 102ngper100mgbodyweight).These effects on blowfly larvae are typical of depressant toxins, such as LqhIT2 [20,28]. The amino acid sequence deduced from the cDNA nucleotide sequence reveals 73% sequence identity with AahIT4 [23], and 85 and 91% with BmK AS and BmK AS-1 [24], respectively (Fig. 2). The relative molecular mass of Lqhb1, determined by mass spectroscopy, is 7463 Da, which matches the calculated value of the amino acid sequence deduced from the cDNA nucleotide sequence. This suggests that Lqhb1 does not undergo post-trans- lational processing as has been shown for other toxins [27]. Binding of Lqhb1 to insect and rat brain NaCh The sequence similarity between Lqhb1 and AahIT4 [23] could suggest similar activity and therefore we analyzed the Fig. 2. Sequence alignment of toxin representatives of various pharmacological groups that affect sodium channels. The alignment relies on known structures, putative models [2] and conserved cysteine residues. Dashes indicate gaps. Asterisks (*) designate alpha-amidation of C-termini. Cysteine residues that are conserved in all scorpion toxins and form disulfide bonds (plane lines) are shaded by light grey, whereas cysteines involved in unique disulfide bonds (dashed lines) are shaded by dark grey. Aah, Androctonus australis hector;Bj,Buthotus judaicus;BmK,Buthus martensii Karsch; Cn, Centruroides noxius;Cll,Centruroides limpidus limpidus;CsE,Centruroides sculpturatus Ewing; Css, Centruroides suffusus suffusus;Lqh, Leiurus quinquestriatus hebraeus; Lqq, Leiurus quinquestriatus quinquestriatus;Ts,Tityus serrulatus;Tst,Tityus stigmurus;Tb,Tityus bahiensis [12]. 2666 D. Gordon et al. (Eur. J. Biochem. 270) Ó FEBS 2003 pharmacological profile of Lqhb1 in competition binding assays utilizing well established a- and b-toxin markers. Lqhb1 inhibited, though at high concentration, the binding of the classical anti-mammalian a-toxin, Lqh2 [5,29], to site-3 in rat brain synaptosomes (K i ¼ 1.85 l M , n ¼ 2; Fig. 3A). A significantly lower concentration of AahIT4 (60 n M ) inhibited the binding of Aah2 [23]. Lqhb1also competed with the anti-mammalian b-toxin, Css4, for binding to site-4 in rat brain synaptosomes with very high apparent affinity [K i ¼ 0.51 ± 0.27 n M , (mean ± SD), n ¼ 3; Fig. 3B]. In comparison, the apparent affinity of AahIT4 for the binding site of Css2 is much lower (IC 50 ¼ 40 n M [23]). Although Lqhb1 competed with both site-3 and site-4 toxins for binding to rat brain sodium channels, it did not inhibit the binding of the a-insect toxin, LqhaIT, to receptor site-3 on cockroach channels (not shown). Yet, Lqhb1 inhibited the binding of the excitatory toxin, Bj-xtrIT, to cockroach synaptosomes with high apparent affinity (K i ¼ 0.170 ± 0.017 n M , n ¼ 3; Fig. 3C). These results demonstrate the similarity in binding capabi- lities between Lqhb1 and AahIT4, although they vary quantitatively in the apparent affinity for receptor sites 3 and 4 in rat brain sodium channels. Effects of Lqhb1 on sodium channel subtypes The effects of Lqhb1 on sodium currents were examined on various mammalian and insect sodium channels expressed in Xenopus oocytes using two-electrode voltage-clamp. In the absence of toxin, a step-depolarization from a holding potential of )80 to )40 mV elicits almost no current in oocytes expressing the rNa v 1.2a channels (Fig. 4A). How- ever, in the presence of 2.5 l M Lqhb1, a significant peak current is observed (Fig. 4A) that indicates a shift of channel activation to more negative membrane potentials as readily observed in Fig. 4B. In contrast, upon depolarization to )20 mV, under which channel activation in the absence of toxin was near maximal, peak currents were inhibited 70 ± 10% (n ¼ 4) in the presence of Lqhb1 (Fig. 4A,B). The current–voltage (I–V) relationship delin- eated in Fig. 4B indicates two clear effects imposed by Lqhb1onrNa v 1.2a channels. The shift in the voltage- dependence of channel activation to more negative mem- brane potentials, and a marked decrease in the sodium peak-current amplitude typify the phenotypic change of sodium currents induced by scorpion b-toxins active on mammals, e.g. Ts7 and Css4 [8,14,33–35]. As ÔNew WorldÕ b-toxins that affect mammals show weak activity on cardiac sodium channels [34,36], the specificity of Lqhb1was further examined on rat skeletal muscle (rNa v 1.4) and human heart (hNa v 1.5) channels. The toxin effects on rNa v 1.4 were similar to those obtained with rNa v 1.2a, whereas no shift in channel activation and only little decrease in peak current were observed in hNa v 1.5 (Fig. 4C,D). These results indicate that Lqhb1 is similar to other b-toxins in action and specificity to mammalian sodium channel subtypes. Lqhb1 is similar to the anti-insect depressant toxin, LqhIT2, in its toxic symptoms induced in blowfly larvae, and the ability to compete for the binding site of the excitatory toxin, Bj-xtrIT, in insect NaChs [19,22]. There- fore, the electrophysiological effects of Lqhb1 and LqhIT2 were compared on an insect NaCh. Both toxins revealed typical b-toxin effects on oocytes expressing the Drosophila Para NaCh and TipE. The peak sodium current, elicited by depolarization to )10 mV, decreased 45 ± 13% and 66 ± 5% (n ¼ 3) in the presence of 2.5 l M Lqhb1and LqhIT2, respectively. In addition, the I–V curves obtained indicate that the appearance of the sodium current is shifted to more negative potentials in the presence of either toxin (Fig. 5A). Similar effects were observed when LqhIT2 was applied on an isolated cockroach axon [18]. However, in contrast to Lqhb1 (Fig. 4B), LqhIT2 in concentrations as high as 50 l M did not affect the rNa v 1.4 channel (Fig. 5B), Fig. 3. Binding of Lqhb1 to receptor sites 3 and 4 in rat brain and cockroach sodium channels. Competition of Lqhb1 with the site-3 a-toxin, Lqh2 (A), and with the site-4 b-toxin, Css4 (B), for binding to rat brain NaChs. Rat brain synaptosomes (65 lgproteinÆmL )1 ) were incubated 30 min at 22 °C with 110 p M [ 125 I]Lqh2 or 120 p M [ 125 I]Css4 and increasing concentrations of the indicated toxins. Nonspecific binding, determined in the presence of 200 n M Lqh2 or 1 l M Css4, respectively, was subtracted. (C) Competition of Lqhb1with[ 125 I]Bj-xtrIT for binding to cockroach NaCh. Cockroach neuronal membranes (16 lgÆmL )1 ) were incubated for 60 min at 22 °Cwith180p M [ 125 I]Bj-xtrIT and increasing concentrations of the indicated toxins. Nonspecific binding, determined in the presence of 1 l M Bj-xtrIT, was subtracted. The amount of bound [ 125 I]toxin is provided as the percentage of maximal specific binding without competitor. The competition curves were analyzed using the nonlinear fit of the Hill equation (withaHillcoefficientof1)fordeterminingoftheIC 50 values (see Experimental procedures). The data points are means of two to three measurements from representative experiments, of which the following K i values were obtained (in n M ): Lqh2, 2.1; Lqhb1, 1900 (in A). Css4, 1.1; Lqhb1, 0.64 (in B). Bj-xtrIT, 0.35; Lqhb1, 0.2 (in C). Ó FEBS 2003 A novel ÔOld WorldÕ scorpion b-toxin (Eur. J. Biochem. 270) 2667 nor the rNa v 1.2a and hNa v 1.5 mammalian channels (not shown), indicating high specificity for insect sodium channels. Discussion Comparison of toxins found in ÔOld WorldÕ vs. ÔNew WorldÕ scorpions may provide a hint about their diversification, but first thorough characterization of their pharmacological properties and genetic relations is needed. Scorpions of the family Buthidae originated approximately 350 MA from the Carboniferous scorpions, Neoscorpiones [37], and were physically divided 150 MA upon the partition of the Upper Jurassic Brazilo-Ethiopian continent (Africa–South Amer- ica). Scorpion a-toxins, that affect NaCh inactivation and resemble ÔOld WorldÕ classical a-toxins in their amino acid sequences, have been described in the ÔNew WorldÕ,e.g. CsE V, Ts IV, and TsTX V [1,12], which suggests that they existed before the separation of the continents. As b-toxins active on mammals have not been found thus far in scorpions of the ÔOld WorldÕ, it could be assumed that they have developed in Tityus and Centruroides ÔNew WorldÕ scorpions after the separation of the continents [12]. Yet, scorpion polypeptides that resemble ÔNew WorldÕ b-toxins have been reported in the ÔOld WorldÕ and include the nontoxic polypeptide, AahSTR1 [38], the glycosylated toxin, Aah6 [39], and the anti-insect selective excitatory and depressant toxin groups [2,12,17,20]. In addition, polypep- tides with some b-toxin properties have been found in the venom of the old world scorpions, Leiurus quinquestriatus hebraeus [40] and Buthus martensii Karsch [25]. Another peculiar toxin that competes for both a and b-toxin receptor sites in rat brain synaptosomes, AahIT4, seems to be related to b-toxins because it was recognized by antibodies raised against the b-toxin, Css2, but not by antibodies against the a-toxin, Aah2, or the excitatory toxin, AahIT [23]. The pharmacological properties of Lqhb1 describe for the first time a typical b-toxinintheÔOld WorldÕ. The high similarity in sequence of Lqhb1, Bmk AS, BmK AS-1, and AahIT4 suggests that these toxins may belong to a unique group of ÔOld WorldÕ b-toxins (Fig. 2). Lqhb1 resembles AahIT4 in its high apparent affinity for receptor site-4 in insect sodium channels (Fig. 3 [23]). Yet, whereas AahIT4 has moderate affinity for both receptor sites 3 and 4 on rat brain synaptosomes [23], Lqhb1 competes with Lqh2 on binding to receptor site-3 only at high concentrations, and binds receptor site-4 in mammalian sodium channels with a very high apparent affinity (Fig. 3). Although both toxins seem to belong to one pharmacological group, Lqhb1exerts typical b-toxin binding properties to a greater extent than AahIT4. Lqhb1 and AahIT4 vary also in their effect on blowfly larvae as AahIT4 induces contraction [23] and Lqhb1 induces flaccid paralysis. The sequence of Lqhb1 resembles those of ÔNew WorldÕ b-toxins (41–50% identity; Fig. 2). The pharmacological features of Lqhb1 are mostly similar to those of the b-toxin, Ts7, with high affinity binding for insect and mammalian NaChs (Figs 3–5 [1,2,13]), and preference for mammalian brain and skeletal muscle NaCh subtypes (Fig. 4 [34]). Notably, the sequence, binding, and electrophysiological properties of Lqhb1 show substantial resemblance to those of ÔOld WorldÕ anti-insect selective depressant toxins, such Fig. 5. Comparison of the effects of Lqhb1 and LqhIT2 on insect sodium channels. Curves I–V were obtained from two-electrode voltage-clamp experiments using Xenopus oocytes that coexpress the Drosophila Para a-subunit with the auxiliary insect b-subunit TipE (A), or the rNa v 1.4 a-subunit with b1 (B). Peak amplitudes obtained in the absence of toxin are designated by filled circles; and peak amplitudes measured from traces elicited in the presence of toxin are represented by open circles. Lqhb1 or LqhIT2 (2.5 l M ) enabled Para channel sodium cur- rents to be elicited at more hyperpolarized potentials, and decreased the peak inward currents at potentials > )20mV.Incontrast,50l M LqhIT2 had no effect on currents mediated by rNa v 1.4. Fig. 4. Effects of Lqhb1 on mammalian sodium channel subtypes. Current–voltage (I–V) curves were obtained with two-electrode voltage-clamp experiments using Xenopus oocytes that coexpress rat brain, rNa v 1.2a (A, B), rat skeletal muscle, rNa v 1.4 (C), and human heart, hNa v 1.5 (D) channel a-subunits together with the mammalian auxiliary subunit b1. In A, currents elicited during a depolarizing pulse to )40 mV (upper traces) or )20 mV (lower traces) in the absence and presence of 2.5 l M Lqhb1, are shown. Note that in the presence of Lqhb1, the current appears during the )40 mV pulse, in contrast to the control trace. (B–D) Representative I–V curves are shown for experi- ments in which the different channels were expressed in Xenopus oocytes in the absence of toxin (control; d) and in the presence of 2.5 l M Lqhb1(s). Appearance of current occurs at more hyper- polarized potentials in the presence of toxin for oocytes expressing rNa v 1.2a or rNa v 1.4, but not rNa v 1.5 channels. 2668 D. Gordon et al. (Eur. J. Biochem. 270) Ó FEBS 2003 as LqhIT2 (44–50% identity; Figs 2,3 and 5 [18,20]). The similarity between Lqhb1 and LqhIT2 is further exemplified in the effect on the Para NaCh expressed in Xenopus oocytes. The two toxins still differ substantially in that only Lqhb1 affects mammalian NaChs (Fig. 5). These features may suggest that the group of toxins represented by Lqhb1 and AahIT4 evolved into the anti-insect selective depressant toxins in the ÔOld WorldÕ,andintob-toxins presently found in ÔNew WorldÕ scorpions, after the separation of the continents. It seems that diversification of the b-toxins in the ÔNew WorldÕ proceeded toward those with affinity for mammals (e.g. Css2 and Css4 [1,12,36]), crustaceans (e.g. Cn5, Cn11, and Cll1 [9–12]) and a group that acquired a-like activity while maintaining the structural features of b-toxins (CsEv1–3 [2,12,14]) 11 . Tityus b-toxins, such as Ts7, Tst1, and Tb1 are highly active on mammals and insects [16], and thus seem to preserve ancient properties of Lqhb1 in the ÔNew WorldÕ. Although all known Ôlong chainÕ scorpion toxins share a similar structural core (a-helix packed against three anti- parallel b-strands) [2,12,41], and genomic organization [41,42], the identity of the ancestor polypeptide, from which they had diverged is a riddle, which is further accentuated due to difficulties to establish reliable phylogenetic relations between the available toxin sequences (M. Gurevitz & D. Gordon, unpublished observations) 12 . Nonetheless, the recent description of a Ôlong chainÕ polypeptide with only three disulfide bonds, birtoxin, found in the venom of the South African scorpion Parabuthus transvaalicus [43,44], together with the features of Lqhb1 may enable speculation on a putative route for toxin diversification. Birtoxin is toxic to mice, inhibits Na + currents in dissociated fish retinal cells in a manner resembling the effect of b-toxins, and shows substantial sequence similarity to b-toxins from various ÔNew WorldÕ Centruroides species [43,44]. These features may suggest that birtoxin represents an ancient group of toxins that could have evolved into the excitatory anti-insect selective toxins by acquiring a structurally distinctive forth disulfide bond (Fig. 2), or, alternatively, into the group represented by Lqhb1 by acquiring the forth conserved disulfide bond found in all but the excitatory toxins. The Lqhb1 group may have evolved after the separation of the continents to depressant toxins in the ÔOld WorldÕ,andto b-toxins in the ÔNew WorldÕ. Since the forth disulfide bond in a and most b-toxins is spatially conserved [2,12,41,45], it may be hypothesized that toxins with three disulfide bonds, such as birtoxin, were the progenitors of a-toxins as well. Acknowledgements This research was supported in part by the United States–Israel Binational Agricultural Research and Development grant IS-3259–01 (D. G. and M. G.); by the Israeli Science Foundation, grants 508/00 (D. G.) and 733/01 (M. G.); and by an EU grant QLK3-CT-2000– 00204 (D. G. and M. G.). References 1. Martin-Eauclaire, M F. & Couraud, F. (1995) Scorpion neuro- toxins: effects and mechanisms. In Handbook of Neurotoxicology (Chang, L.W. & Dyer, R.S., eds), pp. 683–716. Marcel Dekker, New York. 2. Gordon, D., Savarin, P., Gurevitz, M. & Zinn-Justin, S. (1998) Functional anatomy of scorpion toxins affecting sodium channels. J. Toxicol. Toxin Rev. 17, 131–159. 3. Gordon, D. (1997) Sodium channels as targets for neurotoxins: mode of action and interaction of neurotoxins with receptor sites on sodium channels. In Toxins and Signal Transduction (Lazarowici, P. & Gutman, Y., eds), pp. 119–149. Harwood, Amsterdam. 4. Gordon, D., Martin-Eauclaire, M.F., Ceste ` le, S., Kopeyan, C., Carlier, E., Ben Khalifa, R., Pelhate, M. & Rochat, H. (1996) Scorpion toxins affecting sodium current inactivation bind to distinct homologous receptor sites on rat brain and insect sodium channels. J. Biol. Chem. 271, 8034–8045. 5. Gilles, N., Blanchet, C., Shichor, I., Zaninetti, M., Lotan, I., Bertrand,D.&Gordon,D.(1999)Ascorpiona-like toxin that is active in insects and mammals reveals an unexpected specificity of distribution of sodium channel subtypes in rat brain neurons. J. Neurosci. 19, 8730–8739. 6. Gilles, N., Chen, H., Wilson, H., Le Gall, F., Montoya, G., Heinemann, S.H. & Gordon, D. (2000) Scorpion a and a-like toxins differentially interact with sodium channels in mammalian CNS and periphery. Eur. J. Neurosci. 12, 2823–2832. 7. Catterall, W.A. (1992) Cellular and molecular biology of voltage- gated sodium channels. Physiol. Rev. 72, S15–S48. 8. Strichartz, G., Rando, T. & Wang, G.K. (1987) An integrated view of the molecular toxicology of sodium channel gating in excitable cells. Annu. Rev. Neurosci. 10, 237–267. 9. Lebreton, F., Delepierre, M., Ramirez, A.N., Balderas, C. & Possani, L.D. (1994) Primary and NMR three-dimensional structure determination of a novel Crustacean toxin from the venom of the scorpion Centruroides limpidus limpidus Karsch. Biochemistry 33, 11135–11149. 10. Garcia, C., Becerril, B., Selisko, B., Delepierre, M. & Possani, L.D. (1997) Isolation, characterization and comparison of a novel Crustacean toxin with a mammalian toxin from the venom of the scorpion Centruroides noxius Hoffmann. Comp. Biochem. Physiol. 116B, 315–322. 11. Ramirez-Dominguez, M.E., Olamendi-Portugal, T., Garcia, U., Garcia,C.,Arechiga,H.&Possani,L.D.(2002)Cn11,thefirst example of a scorpion toxin that is a true blocker of Na + currents in crayfish neurons. J. Exp. Biol. 205, 869–876. 12. Possani, L.D., Becerril, B., Delepierre, M. & Tytgat, J. (1999) Scorpion toxins specific for Na + -channels. Eur. J. Biochem. 264, 287–300. 13. Barhanin, J., Giglio, J.R., Leopold, P., Schmid, A., Sampaio, S.V. & Lazdunski, M. (1982) Tityus serrulatus venom contains two classes of toxins. J. Biol. Chem. 257, 12553–12558. 14. Meves, H., Simard, J.M. & Watt, D.D. (1986) Interactions of scorpion toxin with the sodium channel. Annu. NY Acad. Sci. 479, 113–131. 15. Becerril, B., Corona, M., Coronas, F.I.V., Zamudio, F., Calderon- Aranda, E.S., Fletcher, P.L., Martin, B.M. & Possani, L.D. (1996) Toxic peptides and genes encoding toxin c of the Brazilian scorpions Tityus bahiensis and Tityus stigmurus. Biochem. J. 313, 753–760. 16. Becerril, B., Marangoni, S. & Possani, L.D. (1997) Toxins and genes isolated from scorpions of the genus Tityus. Toxicon 35, 821–835. 17. Zlotkin, E. (1983) Insect selective toxins derived from scorpion venoms: An approach to insect neuropharmacology. Insect Biochem. 13, 219–236. 18. Ben Khalifa, R., Stankiewicz, M., Lapied, B., Turkov, M., Zilberberg, N., Gurevitz, M. & Pelhate, M. (1997) Refined electrophysiological analysis suggests that a depressant toxin is a sodium channel opener rather than a blocker. Life Sci. 61, 819–830. Ó FEBS 2003 A novel ÔOld WorldÕ scorpion b-toxin (Eur. J. Biochem. 270) 2669 19. Froy, O., Zilberberg, N., Gordon, D., Turkov, M., Gilles, N., Stankiewicz, M., Pelhate, M., Loret, E., Oren, D.A., Shaanan, B. & Gurevitz, M. (1999) The putative bioactive surface of insect- selective scorpion excitatory neurotoxins. J. Biol. Chem. 274, 5769–5776. 20. Zlotkin. E., Gurevitz. M., Fowler. E., Moyer, M. & Adams, M.E. (1993) Depressant insect selective neurotoxins from scorpion venom: chemistry, action and gene cloning. Arch. Insect Biochem. Physiol. 22, 55–73. 21. Shichor, I., Zlotkin, E., Ilan, N., Chikashvili, D., Stuhmer, W., Gordon,D.&Lotan,I.(2002)Domain2ofDrosophila Para voltage-gated sodium channel confers insect properties to a rat brain channel. J. Neurosci. 22, 4364–4371. 22. Gordon, D., Moskowitz, H., Eitan, M., Warner, C., Catterall, W.A. & Zlotkin, E. (1992) Localization of receptor sites for insect- selective toxins on sodium channels by site-directed antibodies. Biochemistry 31, 7622–7628. 23. Loret, E.P., Martin-Euclaire, M.F., Mansuelle, P., Sampieri, F., Granier, C. & Rochat, H. (1991) An anti-insect toxin purified from the scorpion Androctonus australis hector also acts on the a and b sites of the mammalian sodium channel: sequence and circular dichroism study. Biochemistry 30, 633–640. 24. Ji, Y.H., Li, Y.J., Zhang, J.W., Song, B.L., Yamaki, T., Mochizuki, T., Hoshino, M. & Yanaihara, N. (1999) Covalent structure of BmK AS and BmK AS-1, two novel bioactive poly- peptides purified from Chinese scorpion Buthus martensii Karsch. Toxicon 37, 519–536. 25. Goudet, C., Chi, C W. & Tytgat, J. (2002) An overview of toxins and genes from the venom of the Asian scorpion Buthus martensii Karsch. Toxicon 40, 1239–1258. 26. Zilberberg, N. & Gurevitz, M. (1993) Rapid isolation of full length cDNA clones by Ôinverse PCRÕ: Purification of a scorpion cDNA family encoding a-neurotoxins. Anal. Biochem. 209, 203–205. 27. Zilberberg, N., Zlotkin, E. & Gurevitz, M. (1992) Molecular analysis of cDNA and transcript encoding the depressant insect neurotoxin of the scorpion. Leiurus quinquestriatus hebraeus. Insect Biochem. Molec. Biol. 22, 199–203. 28. Turkov, M., Rashi, S., Zilberberg, N., Gordon, D., Ben Khalifa, R.,Stankiewicz,M.,Pelhate,M.&Gurevitz,M.(1997)Invitro folding and functional analysis of an anti-insect selective scorpion depressant neurotoxin produced in Escherichia coli. Prot. Express. Purific. 9, 123–131. 29. Gilles, N., Leipold, E., Chen, H., Heinemann, S.H. & Gordon, D. (2001) Effect of depolarization on binding kinetics of scorpion a-toxin highlights conformational changes of rat brain sodium channels. Biochemistry 40, 14576–14584. 30. Gilles, N., Krimm, I., Bouet, F., Froy, D., Gurevitz, M., Lancelin, J.M. & Gordon, D. (2000) Structural implications on the inter- action of scorpion a-like toxins with the sodium channel receptor site inferred from toxin iodination and pH-dependent binding. J. Neurochem. 75, 1735–1745. 31. Gershon, E., Weigl, L., Lotan, I., Schreibmayer, W. & Dascal, N. (1992) Protein kinase A reduces voltage-dependent Na + current in Xenopus oocytes. J. Neurosci. 12, 3743–3752. 32. Wallner, M., Weigl, L., Meera, P. & Lotan, I. (1993) Modulation of the skeletal muscle sodium channel a-subunit by the b 1 -subunit. FEBS Lett. 336, 535–539. 33. Vijverberg, H.P., Pauron, D. & Lazdunski, M. (1984) The effect of Tityus serrulatus scorpion toxin c on sodium channel in neuro- blastoma cells. Pflugers Arch. 40, 297–303. 34. Marcotte,P.,Chen,L Q.,Kallen,R.G.&Chahine,M.(1997) Effects of Tityus serrulatus scorpion toxin c on voltage-gated Na + channels. Circ. Res. 80, 363–369. 35. Ceste ` le, S. & Catterall, W.A. (2000) Molecular mechanisms of neurotoxin action on voltage-gated sodium channels. Biochimie 82, 883–892. 36. Cestele, S., Qu, Y., Rogers, J.C., Rochat, H. & Catterall, W.A. (1998) Voltage sensor-trapping: enhanced activation of sodium channels by beta-scorpion toxin bound to the S3–S4 loop in domain II. Neuron 21, 919–931. 37. Byalynitskii-Birula, A.A. (1917) Arthrogastric arachnids of Caucasia.PartI.InScorpions (Nasonov, N.V., ed.). [English translation from Russian by Israel Program for Scientific Trans- lations, Jerusalem (1964)]. 38. Blanc, E., Hassani, U., Meunier, S., Mansuelle, P., Sampieri, F., Rochat, H. & Darbon, H. (1997) 1 H-NMR-derived sec- ondary structure and overall fold of a natural anatoxin from the scorpion Androctonus australis hector. Eur. J. Biochem. 247, 1181–1126. 39. Hassani, O., Loew, D., Van Dorsselaer, A., Papandreou, M.J., Sorokine,O.,Rochat,H.,Sampieri,F.&Mansuelle,P.(1999) Aah VI, a novel, N-glycosylated anti-insect toxin from Androc- tonus australis hector scorpion venom: isolation, characteri- sation, and glycan structure determination. FEBS Lett. 443, 175–180. 40. Rack, M., Richter, D. & Rubly, N. (1987) Purification and characterization of a b-toxinfromthevenomoftheAfrican scorpion Leiurus quinquestriatus. FEBS Lett. 214, 163–166. 41. Gurevitz, M., Gordon, D., Ben-Natan, S., Turkov, M. & Froy, O. (2001) Diversification of neurotoxins by C-tail ÔwigglingÕ –a scorpion recipe for survival. FASEB J. 15, 1201–1205. 42. Froy, O., Sagiv, T., Poreh, M., Urbach, D., Zilberberg, N. & Gurevitz, M. (1999) Dynamic diversification from a putative common ancestor of scorpion toxins affecting sodium, potassium, and chloride channels. J. Mol. Evol. 48, 187–196. 43. Inceoglu, B., Lango, J., Wu, J., Hawkina, P., Southern, J. & Hammock, B.D. (2001) Isolation and characterization of a novel type of neurotoxic polypeptide from the venom of the South African scorpion Parabuthus transvaalicus (Buthidae). Eur. J. Biochem. 268, 5407–5413. 44. Inceoglu,B.,Hayashida,Y.,Longo,J.,Ishida,A.T.&Hammock, B.D. (2002) A single charged surface residue modifies the activity of ikitoxin, a beta-type Na + channel toxin from Parabuthus transvaalicus. Eur. J. Biochem. 269, 5369–5376. 45. Fontecilla-Camps, J.C. (1989) Three-dimensional model of the insect-directed scorpion toxin from Androctonus australis hector and its implication for the evolution of scorpion toxins in general. J. Mol. Evol. 29, 63–67. 2670 D. Gordon et al. (Eur. J. Biochem. 270) Ó FEBS 2003 . An ‘Old World’ scorpion b-toxin that recognizes both insect and mammalian sodium channels A possible link towards diversification of b-toxins Dalia Gordon 1 , Nitza Ilan 1,6 , Noam Zilberberg 2 ,. various insect and mammalian sodium channels [1,2,21,22]. As b-toxins that affect mammalian sodium channels have not been identified in the ÔOld WorldÕ, it was assumed that diversification of anti -mammalian b-toxins. quinquestriatus hebraeus was collected from scorpion stings to a parafilm membrane. Sarcophaga falculata (blowfly) larvae and Periplaneta americana (cock- roaches) were bred in the laboratory. Albino laboratory ICR