In the present study, BmK αIV, a novel modulator of sodium channels, was cloned from venomous glands of the Chinese scorpion (Buthus martensi Karsch) and expressed successfully in Escherichia coli. The BmK αIV gene is composed of two exons separated by a 503 bp intron. The mature polypeptide contains 66 amino acids. BmK αIV has potent toxicity in mice and cockroaches. Surface-plasmon-resonance analysis found that BmK αIV could bind to both rat cerebrocortical synaptosomes and cockroach neuronal membranes, and shared similar binding sites on sodium channels with classical AaH II (α-mammal neurotoxin from the scorpion Androctonus australis Hector), BmK AS (β-like neurotoxin), BmK IT2 (the depressant insect-selective neurotoxin) and BmK abT (transitional neurotoxin), but not with BmK I (α-like neurotoxin). Two-electrode voltage clamp recordings on rNav1.2 channels expressed in Xenopus laevis oocytes revealed that BmK αIV increased the peak amplitude and prolonged the inactivation phase of Na+ currents. The structural and pharmacological properties compared with those of other scorpion α-toxins suggests that BmK αIV represents a novel subgroup or functional hybrid of α-toxins and might be an evolutionary intermediate neurotoxin for α-toxins.
VGSCs (voltage-gated sodium channels) are critical for the initiation and propagation of action potential in excitable cells and are related to pain and a series of channel diseases such as epilepsy . They are composed of a pore-forming α-subunit associated with up to four auxiliary β-subunits [1,2]. To date, seven topologically distinct receptor sites for neurotoxins on the α-subunit of VGSCs have been identified, all of which are linked to specific effects on channel function. Numerous natural neurotoxins targeted to VGSCs by affecting either gating properties or ion permeation of VGSCs, have been used extensively for identification and characterization of the subtypes of sodium channels. Among them, scorpion neurotoxins are divided into α- and β-classes according to their mode of actions and binding properties to VGSCs [3,4].
Scorpion α-neurotoxins mainly slow the inactivation of VGSCs and consequently prolong the action potential upon binding to receptor site 3, and are classified into three major groups based on the preferential toxicity to mammals or insects and their differential binding properties. Classical α-mammal neurotoxins bind with high affinity to rat brain VGSCs and are highly toxic to mammals, while they are practically non-toxic to insects. α-Insect neurotoxins bind to insect VGSCs with high affinity and are very active in insects but less potent in mammals. α-Like neurotoxins, which could not bind to rat brain synaptosomes, are active in both mammal and insect nervous systems . Scorpion β-neurotoxins modify the voltage dependence of activation of sodium channels by binding to receptor site 4 of VGSCs. According to their pharmacological activities, the β-neurotoxins can be divided into several groups: β-mammal toxins; β-toxins active on both mammals and insects; excitatory insect-selective toxins; and depressant insect-selective toxins and β-like toxin [3,4].
The Chinese scorpion BmK (Buthus martensi Karsch), of the family Buthidae, is considered to possess a moderately active venom . Many kinds of long-chain scorpion neurotoxins modulating the pharmacological properties of VGSCs have been purified from BmK venom . Moreover, it has been demonstrated that the efficacy of BmK venom in treating pain and epilepsy depends on the different kinds of these modulators of sodium channels from scorpion BmK venom [7–11]. Therefore elucidation of the inherent functional and structural interaction between scorpion neurotoxins and sodium channels may provide a promising perspective for the future design of selective drugs. On the other hand, adaptive evolution of scorpion neurotoxins and ion channels have also been constructed on the basis of genomic organization, structure and pharmacology of toxin and ion channels, as well as scorpion species distribution. The difference of function and components in scorpion venom might reflect the divergence of scorpion species status and distribution [4,12,13].
In the present paper we report that BmK αIV, a novel sodium channel modulator, was cloned and expressed heterogenously, and its pharmacological and electrophysiological characteristics on sodium channels were investigated.
MATERIALS AND METHODS
Escherichia coli strains DH5α and BL21(DE3) were used for plasmid amplification and protein expression respectively. pET-28a(+) was used as the vector for protein expression. Adult male Sprague–Dawley rats and Kunming mice were provided by the Shanghai Center for Experimental Animals, Shanghai, China. Male adult Periplaneta americana and Blattella germanica cockroaches were from the Shanghai Institute of Entomology (Chinese Academy of Sciences, Shanghai, People's Republic of China). Female Xenopus laevis were purchased from Chinese Academy of Sciences. BmK I (α-like neurotoxin), BmK abT (transitional neurotoxin), BmK AS (β-like neurotoxin) and BmK IT2 (depressant insect-selective neurotoxin) were purified according to procedures published previously [14–17]. AaH II (α-mammal neurotoxin from the scorpion Androctonus australis Hector) was generously given by Professor Hervé Rochat (Laboratoire de Biochimie, Faculté de Médecine Secteur Nord, Marseille, France) and Dr Marie-France Martin-Eauclaire (UMR 6560 CNRS, Université de la Mediterranée, Institut Jean Roche, Faculté de Médecine Nord, Marseille, France). The alkaloid veratridine was from Sigma. All other reagents used were of analytical grade. All animal protocols followed the National Institutes of Health Guide for the Care and Use of Laboratory Animals and were approved by the Committee of Laboratory Animals, Chinese Academy of Sciences.
Cloning of the BmK αIV gene
Total RNA was extracted using TRIzol® reagent (Invitrogen) from 2 g of venomous glands of BmK (collected in Henan Province, China). The BmK αIV gene was cloned by 5′- and 3′-RACE (rapid amplification of cDNA ends) using SMART™ RACE cDNA Amplification Kit (BD Biosciences Clontech). Primer 1 was designed as 5′-TTATGATACATTCCATCTTA-3′ according to a conserved fragment of the 3′ untranslated regions of two α-neurotoxins (BmK α1 and AaH II) [18,19]. The PCR product was amplified using primer 1 and the UPM (universal primer mix) for 5′-RACE. Primer 2 was 5′-CTTTCCCAGAAAATTCCGTAAAACGGTTC-3′ corresponding to the 5′ untranslated regions of BmK αIV determined by 5′-RACE. The PCR products were amplified using primer 2 and UPM primer for 3′-RACE. Genomic DNA was isolated from venomous glands of BmK using the Wizard Genomic DNA Purification Kit (Promega). The PCR products were amplified using primer 1 and primer 2 and ligated into pGEM-T vector (Promega), then transformed into E. coli DH5α. The plasmid was isolated and then subjected to DNA sequencing.
Construction of expression vector pET-BmK αIV
Two synthetic oligonucleotide primers were designed for amplifying the encoding region of BmK αIV. The forward primer was 5′-TGGAGTTAACATATGGTACGCGATGGTTATATTG-3′, corresponding to the cDNA sequence of the N-terminal residues (1–6) of BmK αIV and with the restriction site of NdeI is underlined. The reverse primer was 5′-GTGCACTGGATCCTTAACCGCCATTGCATCTTCC-3′, corresponding to the cDNA sequence of the C-terminal residues (61–66) of BmK αIV and with the BamHI restriction site underlined. Both primers were used in a PCR with cDNA and the product was digested with NdeI and BamHI. The digested product was subcloned into the corresponding NdeI and BamHI restriction sites so that the product was in-frame with the His6 tag of the pET-28a(+) vector (Novagen). This construct thus allowed the expression of a His6–BmK αIV fusion protein. When required, the fusion protein was cleaved with the protease thrombin (Amersham Biosciences) according to the manufacturer's protocol.
Expression and purification of recombinant BmK αIV
The expression plasmid pET-28a(+)-BmK αIV was transformed into E. coli BL21(DE3) cells (Invitrogen). E. coli was grown at 37 °C overnight in Luria–Bertani broth containing 40 μg/ml kanamycin. Then isopropyl β-D-thiogalactopyranoside to a final concentration of 0.2 mM was added when the culture reached an attenuance (D600) of 0.4–0.6 and the cells were grown for a further 16 h at 22 °C. Cells from 1 litre of culture medium were harvested by centrifugation at 5000 g for 6 min, then resuspended in lysis buffer (30 mM Tris/HCl, 2.5 mM EDTA, pH 8.0, 0.5 mM PMSF and 10 μg/ml lysozyme), and sonicated on ice for 20 min (four bursts/min). The insoluble pellet obtained by centrifugation (12000 g, 15 min) was rinsed three times with washing buffer [25% (w/v) sucrose, 5 mM EDTA and 1% (v/v) Triton X-100 in PBS]. Then the fusion proteins in the pellets were purified by affinity chromatography on Ni2+-nitrilotriacetate–agarose beads (Qiagen) under the denaturing conditions according to the manufacturer's instructions. Protein renaturation was initiated by 50-fold dilution in 0.2 M ammonium acetate (pH 8.0). The dialysis buffer was changed every 24 h three times in total at 22 °C. After renaturation, the insoluble proteins were removed by centrifugation (12000 g, 15 min), while the soluble fusion proteins were concentrated by ultrafiltration membranes (3 kDa molecular-mass cut-off; Amicon) and purified by reverse-phase HPLC (Develosil; ODS-HG-5; 25 cm×0.5 cm; Nomura Chemical Co.). Mobile phases were solvent A [saturated HCl/water, 1:999, (v/v)] and solvent B [60% (v/v) acetonitrile], and elution was performed with two successive linear gradients: 0–15% of solvent B for 5 min, followed by 15–80% of solvent B for the next 25 min at a flow rate of 1 ml·min−1. Fractions were hand-collected by monitoring the effluent signal at 280 nm and dried in a vacuum centrifuge. Immunoblotting was carried out with anti-(His tag) antibody (Santa Cruz Biotechnology). The proteins were detected by ECL® (enhanced chemiluminescence; Amersham Biosciences). The fusion proteins were cleaved with 10 units of thrombin/mg of protein substrate (Amersham Biosciences) in buffer (50 mM Tris/HCl, pH 8.0, 150 mM NaCl and 2.5 mM CaCl2) for 16 h at 22 °C, and purified by reverse-phase HPLC. The molecular mass of the purified products was determined by a REFLEX III matrix-assisted laser-desorption ionization–time-of-flight mass spectrometer (Bruker-Franzen Analytik).
CD spectroscopy and N-terminal sequencing
CD spectroscopy of BmK αIV (0.2 mg/ml in 20 mM sodium phosphate buffer, pH 7.0) was carried out on a J-810 system (JASCO) at room temperature (21–23 °C). The spectrum was recorded from 190 to 250 nm at a scan rate of 100 nm/min with a time constant of 1 s on four occasions. N-terminal chemical protein sequencing was performed on PROCISE™492cLC Protein Sequencing System (Applied Biosystems).
Toxicity assays on mice were performed as described by Ji et al. . Kunming mice (20±2 g) were briefly anaesthetized with diethyl ether and injected i.c.v. (intracerebroventricularly) with five doses of BmK αIV (0.05, 0.15, 0.25, 0.35 and 0.5 μg in 2 μl of saline) or control (2 μl of saline). Eight mice were used for each dose. Then the mice were placed in a 40 cm×30 cm×50 cm transparent glass box for continuous monitoring of the symptoms of neurotoxicity during the first 1 h post-injection or until death. The LD50 for BmK αIV was estimated 24 h after injection. The cockroach was employed for the insect toxicity assay as described previously [20,15]. Different concentrations of BmK αIV (0.05, 0.1, 0.2, 0.5, 1.0 and 2.0 μg, dissolved in 2 μl of 0.15 M NaCl containing 1 mg/ml BSA) were injected into the abdominal segment of male B. germanica cockroaches (80–100 mg, n=8 for each dose). The control group received 2 μl of vehicle solution without BmK αIV (n=8). The toxicity was monitored for 48 h after injection of BmK αIV. LD50 values were calculated according to the method developed by Reed and Muench .
Neuronal synaptosomes preparation
Rat cerebrocortical synaptosomes were prepared from adult male Sprague–Dawley rats (250±10 g) according to the previous description . Insect neuronal membrane of nerve cord was prepared from male adult cockroach P. americana (700±50 mg) following the procedure described previously . Both types of synaptosomes were resuspended in loading buffer (140 mM choline chloride, 1.8 mM CaCl2, 5.4 mM KCl, 0.8 mM MgSO4, 10 mM D-glucose and 25 mM Hepes, pH 7.4) and used for the biosensor assay at once. A combination of proteinase inhibitors consisting of 50 μg/ml PMSF, 1 μM pepstatin A, 1 mM iodoacetamide and1 mM 1,10-phenanthroline was added to all buffers used in the procedure. The concentration of the synaptosome suspension was determined using a Bio-Rad protein assay with BSA as a standard.
The binding assays were performed using surface plasmon resonance using BIAcore 3000 biosensor systems (Amersham Biosciences) at room temperature. BmK αIV (40 μl of 40 μg/ml in 10 mM acetate buffer, pH 5.0) was immobilized on the surface of the carboxymethylated dextran chip (CM5) using the Amine Coupling Kit according to the manufacturer's instructions (Amersham Biosciences). The control channel on the same sensor chip was processed according to the protocol described above but without peptide couple. Rat cerebrocortical synaptosomes and cockroach nerve-cord neuronal membranes dissolved in the loading buffer were passed over the chip surface. The procedure was carried out in a running buffer of Hepes-buffered saline (150 mM NaCl, 3.5 mM EDTA, 0.05% BIAcore surfactant P-20 and 10 mM Hepes, pH 7.4) at a flow rate of 5 μl/min. The regeneration of the chip surface was accomplished by the injection of 10 mM NaOH. The results were fitted and analysed using BIAevaluation 3.1 software.
Two-electrode voltage clamp recording in X. laevis oocytes
cRNAs encoding rNav1.2 (rat brain II sodium channel α-subunit; kindly given by Professor Alan L. Goldin, Department of Microbiology and Molecular Genetics, University of California Irvine, Irvine, CA, U.S.A.) were transcribed in vitro with T7 RNA polymerase mMESSAGE mMACHINE® transcription kit (Ambion). Oocytes were removed from adult female X. laevis and incubated with 1 mg/ml type IA collagenase (Sigma) in an OR2 medium (96 mM NaCl, 2 mM KCl, 1 mM MgCl2 and 5 mM Hepes, pH 7.5) for about 3 h. After washing, the isolated oocytes were transferred into ND-96 medium (96 mM NaCl, 2 mM KCl, 1.8 mM CaCl2, 1 mM MgCl2 and 5 mM Hepes, pH 7.5) supplemented with 0.1 mg/ml gentamicin, and the healthy oocytes in stage V–VI were chosen to be injected (1 ng/oocyte) with rNav1.2 cRNAs. Both injected and non-injected oocytes were further incubated for 40 h at 20 °C in ND-96 medium.
Na+ currents were recorded using two-electrode voltage clamps with the TURBO TEC-03X, INT-20 and Cellwork E 5.5 software (NPI Electronic, Tamm, Germany) at room temperature. The sensitivity of current output was 1 V/μA, and the filter was chosen at 2 kHz. Na+ current amplitudes were controlled between 5000 and 20000 nA. Data acquisition and analysis were performed using the Cellworks Reader 3.6 (NPI Electronic) and Origin 7.0 (OriginLab Corp., Northampton, MA, U.S.A.).
Sequence analysis and molecular modelling
Amino acid sequences of representative α-toxins were obtained from the Swiss-Prot database (http://www.expasy.org/sprot) and aligned using ClustalX 1.81 (http://bips.u-strasbg.fr/en/Documentation/ClustalX/). A bootstrap consensus tree was generated using the PHYLIP package (http://www.genebee.msu.su/services/phtree_reduced.html).
The three-dimensional homology modelling of BmK αIV was performed by the program MODELLER  running on an SGI O2 workstation. The three-dimensional structure of AaH II experimentally resolved at 1.3 Å (1 Å=0.1 nm; Protein Data Bank accession no. 1PTX) was used as a template, it having close sequence identity with BmK αIV (68.2%). The models were screened with the PROCHECK program , from which the best model was selected. The electrostatic calculations were done by using the GRASP software  running on an SGI O2 Workstation.
Results are presented as means±S.E.M. and submitted to one-way ANOVA. The Dunnet test was employed to compare the control group with the different treatment groups. P<0.05 was considered to be significant.
Genomic organization of BmK αIV
A full-length sequence of BmK αIV genomic DNA (GenBank® accession no. DQ058408) was cloned and found to be 896 bp in length. As shown in Figure 1, it consisted of two exons disrupted by a single intron of 503 bp. The intron was arranged between the first two bases of the glycine codon (the glycine residue is situated four residues upstream from the start of the mature peptide). The 5′ and 3′ splice sites were GTAAGATTTA and CTGACTACAG respectively. The intron of BmK αIV was AT-rich (62.0%). The initiation codon (ATG) was at the +70 position, relative to the transcription start site, and the stop codon (TAA) was at the +828 position. The poly(A) signal (AATAAA) was located 47 bp downstream from the stop codon. The open-reading-frame region encoded BmK αIV precursor polypeptide with 85 amino acid residues which consisted of signal peptide (19 residues) and mature peptide named BmK αIV (66 residues).
Nucleotide sequence of genomic DNA encoding BmK αIV
Expression and purification of recombinant BmK αIV
His6–BmK αIV was found predominantly in inclusion bodies when expressed in E. coli. The identification of His6–BmK αIV was processed by immunochemical recognition by an anti-(His tag) antibody (Figure 2B). The recombinant BmK αIV was gained by thrombin cleavage from His6–BmK αIV detected by Coomassie Blue staining of tricine/polyacrylamide gel (Figure 2A). The molecular mass (7733.59 Da) of BmK αIV determined by MS (Figure 2C) was consistent with the calculated value of the amino acid sequence deduced from the cDNA nucleotide sequence and four extra amino acids resulting from the thrombin cleavage site and the restriction site using for cloning (7734.84 Da). The largest yield of bioactive products was about 2 mg/l of culture medium. The recombinant BmK αIV, hereafter just called ‘BmK αIV’, was used for the following experiments. The N-terminal sequence of BmK αIV was (Gly-Ser-His-Met)-Val-Arg-Asp-Gly-Tyr-Ile-Ala-Asp determined by Edman degradation (Figure 1), in which the first two residues (Gly-Ser) was the rest of the thrombin recognition site, and the following two (His-Met) were encoded by restriction-enzyme-site sequences. The secondary structure of the recombinant BmK αIV, determined by CD spectroscopy, showed it to be folded as a well-structured neurotoxin (Figure 2D).
Identification of BmK αIV
Toxicity assay of BmK αIV
Excitatory neurotoxicity in mice, such as spasms, convulsions, rapid movements and death, appeared within 10–20 min after injection of BmK αIV. The LD50 of BmK αIV by i.c.v injection to mice was estimated to be about 208 ng/20 g of body weight. The maximal dose and LD50 of BmK αIV in cockroaches were determined to be 2.0 and 0.47 μg/100 mg of cockroach body weight respectively at 6 h post-injection. During the 6 h post-injection, the symptoms of cockroach paralysis appeared within 15 min after injection of BmK αIV. At the lower dose of BmK αIV (0.05, 0.1 or 0.2 μg), some of cockroaches were able to recover from paralysis, whereas at the higher dose of 0.5 or 1.0 μg, there was little recovery. The mortality was increased gradually within 12–48 h of the application of BmK αIV at 0.5 or 1.0 μg dose. No toxic responses or death occurred in the control group of mice or cockroaches.
Rat-cerebrocortical-synaptosome- and cockroach-neuronal-membrane-binding properties of BmK αIV
BmK αIV could bind to rat cerebrocortical synaptosomes and cockroach neuronal membranes in a concentration-dependent manner, but could not be dissociated from its binding site completely after elution with running buffer. The binding of BmK αIV to rat cerebrocortical synaptosomes was increased progressively during the association phase of 240 s. The dissociation happened rapidly at first and then became slower later (Figure 3A). However, BmK αIV bound quickly to cockroach neuronal membranes and then reached saturation gradually. The dissociation of BmK αIV from cockroach neuronal membranes was a slow process (Figure 3B). The kinetic dissociation constant (koff) was calculated to be (1.71±0.13)×10−3 s−1 to rat cerebrocortical synaptosomes and (4.16±0.15)×10−3 s−1 to cockroach neuronal membranes during 260–380 s respectively. The other kinetic parameters could not be calculated, since it was difficult to determine the molarity of target protein in synaptosomes with any certainty.
Representative sensorgrams showing the kinetics of BmK αIV binding to rat cerebrocortical synaptosomes (A) and cockroach neuronal membranes (B)
The results of the competitive assay of BmK αIV binding to rat cerebrocortical synaptosomes and cockroach neuronal membranes is shown in Figures 4(A) and Figure 4(B) respectively. The binding of BmK αIV could be inhibited by BmK αIV itself and potently by AaH II, even at quite low concentration (10−10 M). The inhibition of AaH II to BmK αIV binding on rat cerebrocortical synaptosomes (70.1±2.0%) was stronger than that on cockroach neuronal membranes (10.9±1.8%). Veratridine (10−4 M) could partially inhibit the binding of BmK αIV to rat cerebrocortical synaptosome (73.4±2.8%), but not to cockroach neuronal membranes. Furthermore, the binding of BmK αIV to rat cerebrocortical synaptosome and cockroach neuronal membranes could be completely blocked by BmK abT and BmK AS. The competitive effect of BmK IT2 on the binding of BmK αIV to rat cerebrocortical synatosomes was about 81.2±2.0%, whereas to the cockroach neuronal membranes it was complete. However, BmK I, an α-like neurotoxin, was unexpectedly found not to inhibit BmK αIV binding to either rat cerebrocortical synaptosomes or cockroach neuronal membranes. Additionally, the binding of BmK αIV to rat cerebrocortical synaptosomes was voltage-dependent, but this was not the case for binding of the toxin to cockroach neuronal membranes, as shown in Figures 5(A) and Figure 5(B) respectively.
Competitive experiments between BmK αIV and neurotoxins
Voltage-dependent binding of BmK αIV to rat cerebrocortical synaptosomes (A) and cockroach neuronal membranes (B)
Electrophysiological analysis of the effect of BmK αIV on rNav1.2
Xenopus oocytes with rNav1.2 cRNA injection were held at −80 mV and depolarized by step pulses ranging from −80 mV to +60 mV for 600 ms in 10 mV increments. During BmK αIV application, the transient Na+ currents were increased in a dose-dependent manner, and this effect could not be diminished completely by washing (Figure 6A). Moreover, the dynamic course of inactivation was slowed by BmK αIV at 100 nM, and the current component of ultra-slow inactivation was enhanced significantly (Figure 6C). The degree of ultra-slow inactivation was measured with the amplitudes of steady-state activated Na+ currents, which were determined as the averages during the last 10 ms of the pulse steps. As a result, the time constant of dynamic inactivation was increased after 100 nM BmK αIV application (Figure 6D). The steady-state Na+ currents were increased by about 70% (P=0.015; Figure 6E) with 100 nM BmK αIV. Neither the voltage-dependence of activation (Figure 6B) nor that of the steady-state inactivation (results not shown) was affected significantly by BmK αIV.
Electrophysiological effect of BmK αIV on rNav1.2 channel expressed in Xenopus oocytes
Alignment of BmK αIV with known scorpion α-neurotoxins
In Figure 7, the sequence comparison of BmK αIV with the partially known scorpion toxin sequences showed that BmK αIV shared high sequence similarity with bukatoxin (84.8%, α-toxin from BmK) and classical α-mammal neurotoxins, such as Lqq V (Leiurus quinquestriatus quinquestriatus V)/Amm V (Androctonus mauretanicus mauretanicus V) (75.8%), AaH II/Lqq III/Lqh αIT (L. quinquestriatus hebraeus αIT) (68.2%) and Lqh II (66.7%). A phylogenetic tree was constructed using a number of representative toxins from each pharmacological group. As shown in Figure 8, BmK αIV was preferentially close to the α-mammal subgroup. An energy-minimized model of BmK αIV was generated as shown in Figure 9. Analysis by PROCHECK program showed that the Ramachandran scores of BmK αIV model are close to those of the AaH II template. The best model showed 94.3% residues in the most favoured regions and 100% in the most favoured or additional allowed regions (Figure 9B). The analysis showed that four clusters of charged amino acids were distributed on the external region of the AaH II and BmK αIV molecular surface (ellipses traced with a broken pink line in Figure 9): region 1, Asp8, Asp9, Val10 and His64; region 2, Arg18, Glu24, Glu25 and Lys28; region 3, Lys58 and Arg62; region 4, Lys2, Asp53, and Arg56. However, at the relative position of BmK αIV, the residues were Asp8, Asp9 and Lys10 (region 1); Arg18, Asp24, Asp25 and Lys28 (region 2); Arg62 (region 3); Arg2 Asp53 and Lys54 (region 4) respectively.
Alignment of the amino acid sequence of BmK αIV with the partially known sequences of α- and β-scorpion toxins
Unrooted phylogenetic tree of scorpion α-toxins
Comparison of the electrostatic potentials of AaH II (A) with BmK αIV (B)
The modulatory effects of BmK αIV on rNav1.2
The gene encoding a novel neurotoxic polypeptide named BmK αIV was cloned from a genomic DNA library prepared from BmK venom glands, and the mature toxin was successfully heterologously expressed in E. coli. Electrophysiological results showed that both the inactivation and the transient peak amplitude of Na+ currents of rNav1.2 expressed in Xenopus oocytes could be modulated by BmK αIV (Figure 6), suggesting that rNav1.2 is one of the preferred targets for BmK αIV. Owing to the lack of effect of BmK αIV on the voltage-dependence of activation of rNav1.2 channel (Figure 6B), the binding of BmK αIV to rNav1.2 channels might depend on the conformational changes just before the channel opening induced by depolarization .
The receptor site of BmK αIV on sodium channels
The major differences among scorpion neurotoxins came from their toxicity and binding properties . The binding of BmK αIV not only to rat cerebrocortical synaptosomes, but also to cockroach neuronal membranes (Figure 3), was distinct from the binding properties of the other known scorpion α-toxins belonging to α-mammal, α-insect or α-like toxins [3,4]. Moreover, the toxicity of BmK αIV to mammals or insects was much weaker when compared with high selective toxicity of the representative classical α-mammal toxins to mice, α-insect and α-like toxins to cockroach. These results thus suggest that BmK αIV might represent a novel functional subgroup of scorpion α-toxins, even if phylogenetic analysis (Figure 8) showed that BmK αIV was preferentially closer to the α-mammal toxins.
The receptor site of BmK αIV was explored by using a series of sodium-channel-targeted neurotoxins belonging to different groups (Figure 4). The binding of BmK αIV to rat cerebrocortical synaptosomes could be potently competed for by AaH II, the most active classical α-mammal toxin , and veratridine, activators acting on receptor site 2 of sodium channels , but not by BmK I, an α-like toxin [30,31], suggesting that the receptor site of BmK αIV on rat brain cerebrocortical sodium channels was similar, but incompletely uniform with site 3 of α-mammal toxin and independent of that of α-like toxin. Moreover, the affinity and dynamic selectivity of BmK αIV to sodium channels were distinct. However, the absent competition of AaH II, veratridine and BmK I on BmK αIV binding to the cockroach neronal membrane indicated that the receptor site of BmK αIV on insect sodium channels might be similar to, but not identical with, that on rat brain-type sodium channels and also unrelated to that of α-like toxin.
Despite their lower overall sequence identity, the binding of BmK αIV to rat cerebrocortical and cockroach neuronal membranes could also be competitively inhibited by BmK AS (β-like neurotoxin) [4,32], BmK IT2 (depressant insect-selective neurotoxin) [33,34] and BmK abT (a transitional member between α- and β-type toxins) , indicating that the receptor site of BmK αIV might be shared, at least partially, with the putative receptor site 3 of β-like neurotoxins, one of the two non-interacting binding sites of depressant insect neurotoxin and that of transitional neurotoxins. These results thus indicate that the receptor site for BmK αIV on sodium channels is distinct from that of known scorpion neurotoxins.
Structural properties of BmK αIV
The possible structural basis for the distinct binding properties of BmK αIV was pursued. Biochemical experiments and site-directed mutagenesis suggested that the interaction surface of scorpion and sodium channels contained at least two distinct domains [36,37]. Core domain (residues 17, 18, 38 and 44) appeared in α-neurotoxins as well as BmK αIV, which was responsible for recognition of receptor site 3 on sodium channels. NC domain (residues 8–10 and 56–64) was responsible for high affinity and the specificity toward sodium channel subtypes. In this domain, residues Asp8-Asp9-Lys10 in BmK αIV were similar to those of AaH II (Asp8-Asp9-Val10), whereas it differed in Lqh αIT (Lys8-Asn9-Tyr10) and BmK I (Lys8-Pro9-His10) in (Figures 7 and 9). Interestingly, the segment (position 56–64) of BmK αIV was similar to that of Lqh αIT, a representative α-insect neurotoxin , indicating that this region might be involved in BmK αIV binding to insect sodium channel subtypes. However, residue Phe17 in Lqh αIT, which might interact directly with cockroach sodium-channel receptor site, was replaced by glycine in most classical α-mammal toxins, as well as in BmK αIV. In addition, the extra sequence Asn64-Gly65-Gly66 of BmK αIV was rarely seen in other known scorpion neurotoxins. Thus the diversities of functional surface formed by the above regions might play important determinants for subtle binding and toxicity profiles of BmK αIV, and indicated that BmK αIV may be a distinct structural and functional hybrid of α-toxins.
Gene organization of BmK αIV and its evolutionary implications
It was found that the intron length of BmK αIV (503 bp) was much longer than that of other known scorpion α-toxins such as AaH I′ (α-toxin, 425 bp) and BmK I (α-like toxin, 408 bp) [38,39]. The intron sequence of BmK αIV shared a lower sequence similarity with that of AaH I′ (50.5%) and BmK I (49.6%). Also, the content of A+T in BmK αIV (62.0%) was different in AaH I′ (74.82%) and BmK I (75%) respectively. Intron size has been found to be associated with protein evolution and expression level in the fruitfly Drosophila . It was inferred that the intron might have emerged and evolved separately before the divergence of α- and β-toxins, and the larger intron for BmK αIV may be correlated with its apparent low level of transcription [12,40]. No identical peptide corresponding to the amino acid sequence and molecular mass of BmK αIV was purified from BmK crude venom so far . The much longer intron of BmK αIV provided a possible hint for its lower expression abundance in BmK venom. The longer intron, lower expression and lower toxicity of BmK αIV might reflect the positive selectivity and adaptive evolution process of scorpion BmK species under environmental pressure. In combination with the pharmacological and structural properties, these findings suggest that BmK αIV might be an evolutionary intermediate neurotoxin for α-toxins.
In summary, the analysis of genomic organization, electrophysiological characteristics, binding activity and molecular modelling revealed that BmK αIV represents a novel subgroup or structural and functional hybrid of α-toxins, and might be an evolutionary intermediate neurotoxin in scorpion α-toxin family. Thus BmK αIV is valuable as a new finding in the family of scorpion toxins for studying the adaptive evolution of scorpion species, structure–function divergence and diversity of scorpion neurotoxic polypeptides. Likewise, BmK αIV may be another useful tool to approach physiological and pharmacological dynamic regulation of sodium-channel targets.
This study was supported by the National Basic Research Program of China (2006CB500801), partially by grants from National Nature Sciences Foundation of China (30270428 and 30370446). We thank Professor Herré Rochat and Dr Marie-France Martin-Eauclaire for kindly providing AaH II, and Professor Ai-Hua Liang (Key Laboratory of Chemical Biology and Molecular Engineering of Ministry of Education, Institute of Biotechnology, Shanxi University, 030006 Taiyuan, China) for her fruitful discussions.
II, α-mammal neurotoxin from Androctonus australis Hector (Sahara scorpion)
Androctonus mauretanicus mauretanicus (black scorpion)
Buthus martensi Karsch (Chinese scorpion)
abT, transitional neurotoxin
AS, β-like neurotoxin
I, α-like neurotoxin
IT2, depressant insect-selective neurotoxin
Leiurus quinquestriatus hebraeus (Hebraei scorpion)
Leiurus quinquestriatus quinquestriatus (death-stalker scorpion)
rapid amplification of cDNA ends
rat brain II sodium channel α-subunit
voltage-gated sodium channel
universal primer mix
The nucleotide sequence data reported have been deposited in the DDBJ, EMBL, GenBank® and GSDB Nucleotide Sequence Databases under the accession number DQ058408.