- Email: [email protected]
Solution structure of a short-chain insecticidal toxin LaIT1 from the venom of scorpion Liocheles australasiae
Biochemical and Biophysical Research Communications 411 (2011) 738–744
Contents lists available at ScienceDirect
Biochemical and Biophysical Research Communications journal homepage: www.elsevier.com/locate/ybbrc
Solution structure of a short-chain insecticidal toxin LaIT1 from the venom of scorpion Liocheles australasiae Shoichiro Horita a, Nobuto Matsushita b, Tomoyuki Kawachi b, Reed Ayabe b, Masahiro Miyashita b, Takuya Miyakawa a, Yoshiaki Nakagawa b, Koji Nagata a, Hisashi Miyagawa b, Masaru Tanokura a,⇑ a b
Department of Applied Biological Chemistry, Graduate School of Agricultural and Life Sciences, University of Tokyo, 1-1-1 Yayoi, Bunkyo-ku, Tokyo 113-8657, Japan Division of Applied Life Sciences, Graduate School of Agriculture, Kyoto University, Kyoto 606-8502, Japan
a r t i c l e
i n f o
Article history: Received 3 July 2011 Available online 18 July 2011 Keywords: Inhibitory cystine knot fold LaIT1 Nuclear magnetic resonance Scorpion toxin Solution structure
a b s t r a c t The solution structure of an insecticidal toxin LaIT1, a 36-residue peptide with a unique amino-acid sequence and two disulfide bonds, isolated from the venom of the scorpion Liocheles australasiae was determined by heteronuclear NMR spectroscopy. Structural similarity search showed that LaIT1 exhibits an inhibitory cystine knot (ICK)-like fold, which usually contains three or more disulfide bonds. Mutational analysis has revealed that two Arg residues of LaIT1, Arg13 and Arg15, play significant roles in insecticidal activity. Ó 2011 Elsevier Inc. All rights reserved.
1. Introduction Most toxins isolated from animal venoms are small peptides that bind to ion channels, such as sodium, potassium, calcium and chloride channels, with high affinity and specificity [1–5]. These toxins block or modulate gating mechanisms of ion channels to immobilize prey or predators. Scorpion venom contains a variety of peptides that are toxic to mammals, insects and crustaceans [6]. To date, more than 600 peptides have been identified from scorpion venoms. Scorpion toxins can be divided into two groups, a 6–8 kDa long-chain peptide group and a 3–5 kDa short-chain peptide group, based on the molecular size [7,8], and also classified into two groups, a cystine-stabilized a-helix/b-sheet (CSa/b) group and an inhibitor cystine knot (ICK) group, based on the backbone fold [9,10]. The CSa/b fold consists of a short a-helix linked to a double- or triple-stranded antiparallel b-sheet stabilized by three or four disulfide bonds. This fold is observed in many scorpion toxins acting on potassium and sodium channels [11,12]. The ICK fold, also referred to as the knottin fold, consists of a double- or triplestranded antiparallel b-sheet stabilized by three disulfide bonds, and forms a b-hairpin and a cystine knot [13,14]. In contrast to the CSa/b fold, the ICK fold is found in only a few scorpion toxins [10], although the ICK-fold peptides have been isolated from many species such as plants, marine cone snails and spiders.
LaIT1 is an insecticidal short-chain toxin isolated from the venom of the scorpion Liocheles australasiae [15]. This toxin consists of 36 amino-acid residues and two intramolecular disulfide bonds with a molecular mass of 4200 Da, and shares a sequence similarity to only one peptide OcyC10 (64% identical) isolated from the Brazilian scorpion venom [16]. Unlike other scorpion insecticidal toxins, LaIT1 and OcyC10 contain only two disulfide bonds. The unique amino-acid sequence and the lack of tertiary structural information prompted us to explore the three-dimensional structure of LaIT1 as well as to further characterize its insect toxicity. In this study, we have analyzed the solution structure by heteronuclear NMR spectroscopy using 15N-labeled and 13C, 15N-labeled LaIT1 that are produced in an Escherichia coli expression system. The NMR analysis has revealed that LaIT1 adopts an ICK-like fold although it contains only two disulfide bonds. We have also examined the species specificity in insect toxicity of LaIT1 and identified its amino-acid residues important for the insecticidal activity. We discuss the structure and function of LaIT1 by comparing the tertiary structures and the target ion channels with structurally similar scorpion and spider toxins.
2. Materials and methods 2.1. Expression and purification of
Abbreviations: ICK, inhibitory cystine knot; CSa/b, cystine-stabilized a-helix/bsheet. ⇑ Corresponding author. Fax: +81 3 5841 8023. E-mail address: [email protected] (M. Tanokura). 0006-291X/$ - see front matter Ó 2011 Elsevier Inc. All rights reserved. doi:10.1016/j.bbrc.2011.07.016
15
N- and
13
C,
15
N-labeled LaIT1
The double stranded DNA encoding LaIT1, ATGGATTTTCCGCT GAGCAAAGAATATGAAACCTGCGTGCGCCCGCGCAAATGTCAGCCGC CGCTGAAATGCAACAAAGCGCAGATTTGTGTGGACCCGAAAAAGGGC
S. Horita et al. / Biochemical and Biophysical Research Communications 411 (2011) 738–744
739
LaIT1 structures at intermediate and final stages in the structure calculation. Hydrogen-bond distance restraints were set as rHN–O = 1.8–2.5 and rN–O = 2.7–3.5 Å.
TGGtaataa, where taataa is the tandem stop codons, was synthesized by mixing the following four oligo nucleotides followed by thermal cycling: LaIT1-S1, cgcgcgcgCCATGGGATTTTCCGCTGAGCA AAGAATATGAAACCT; LaIT1-A2, GCGGCGGCTGACATTTGCGCGGGC GCACGCAGGTTTCATATTCTT; LaIT1-S3, AATGTCAGCCGCCGCTGAA ATGCAACAAAGCGCAGATTTGTGTGG; and LaIT1-A4, cgcGAATTC ttattaCCAGCCCTTTTTCGGGTCCACACAAATCTGCG. The NcoI and EcoRI sites (underlined in the above oligonucleotides) were used for the cloning of the DNA encoding LaIT1 to the expression vector pET-32a (Novagen). Non-labeled, 15N-labeled and 13C, 15N-labeled Trx-His6-ek-LaIT1, where ‘‘Trx’’ and ‘‘ek’’ represents the thioredoxin-tag and the enterokinase recognition site, respectively, were expressed in E. coli BL21(DE3) (Novagen) grown in LB medium, M9 minimal medium containing 15N-labeled ammonium chloride and in M9 minimal medium containing 13C-labeled glucose and 15N-labeled ammonium chloride, respectively, by adding a final concentration of 1 mM IPTG and cultivating at 37 °C for 3 h. Cells were harvested by centrifugation, resuspended in lysis buffer (50 mM Tris–HCl, pH 8.0, 300 mM NaCl, 10 mM imidazole), and disrupted by sonication. After centrifugation, Trx-His6-ek-LaIT1 was purified with a Ni Sepharose 6 Fast Flow (GE Healthcare) column. The Trx-His6-tag was removed by enterokinase (Invitrogen). LaIT1 was purified by cation exchange chromatography using a HiTrap SP column (GE Healthcare).
Inter-proton distance restraints were derived from peak intensities in the 15N-edited NOESY-HSQC and 13C-edited NOESY-HSQC spectra of 13C, 15N-labeled LaIT1 with a mixing time of 90 ms. The cross-peak intensities were translated into inter-proton distances based on the relationships, NOE / (distance)6 and the standard distance between Hai and Haj in the parallel b-sheet of 2.3 Å. Hydrogen-bond distance restraints were also added to the structure calculation as described above. Structure calculation was performed using CYANA [18]. A total of 200 conformers were annealed in 10,000 steps of torsion angle dynamics calculations, of which 10 conformers with the lowest values in the target function were used to represent the solution structure of LaIT1. The conformer with the lowest target function was used as the representative structure of LaIT1. The tertiary structure was visualized with the programs Molmol [19] and PyMol (http://pymol.sourceforge.net/). The representative structure with the lowest target function was submitted to the Dali server [20] to search for proteins structurally similar to LaIT1.
2.2. Determination of disulfide bond linkages in recombinant LaIT1
2.6. Chemical synthesis of LaIT1 and its analogs
To analyze the disulfide pattern of recombinant LaIT1, non-labeled recombinant LaIT1 was first digested with endoproteinase Lys-C (Wako Pure Chemical Industries, Osaka) in 25 mM Tris– HCl, pH 7.1, without reduction or alkylation of disulfide bonds. Then, the molecular masses of fragments were analyzed using an LCMS-IT-TOF mass spectrometer (Shimadzu, Kyoto) equipped with an electrospray ion source.
LaIT1 and its analogs were synthesized by the Fmoc-based solid-phase peptide synthesis method using Wang-PEG resin (Watanabe Chemical Industries, Hiroshima). The C-terminal amino acid was attached to the resin by the symmetrical anhydride method catalyzed by 4-dimethylaminopyridine (DMAP). Symmetrical anhydrides of the Fmoc-protected amino acids were prepared in a reaction with 0.5 eq. of a diisopropylcarbodiimide in anhydrous dichloromethane. The symmetrical anhydrides (5 eq.) and DMAP (0.1 eq.) in anhydrous N,N-dimethylformamide (DMF) were added to the resin and incubated for 1 h at room temperature. After the resin was washed with DMF, the Fmoc group of the amino acid attached to the resin was deprotected with 20% piperidine in DMF. Fmoc-protected amino acids (3 eq.), O-(benzotriazol-1-yl)N,N,N0 ,N0 -tetramethyluronium hexafluorophosphate (HBTU, 3 eq.), and 1-hydroxybenzotriazole (3 eq.) in DMF were added to the resin. After N,N-diisopropylethylamine (6 eq.) was added to the resins, the reaction mixture was incubated for 1.5 h at room temperature. This deprotection/coupling cycle was repeated until the coupling of all residues was finished. Removal of side-chain protecting groups and cleavage of the peptide from the resin was carried out with trifluoroacetic acid (TFA)/triisopropylsilane/ water/ethandithiol [94/2/2.5/2.5(v/v)] for 2 h at room temperature. After the resin was filtered off, cold diethyl ether was added to the filtrate to precipitate the cleaved peptides. The precipitated peptides were washed with cold diethyl ether twice and dried in vacuo. The crude linear peptides were then oxidized to form disulfide linkages in the glutathione redox buffer (1 mM reduced glutathione and 0.1 mM oxidized glutathione in 0.2 M Tris–HCl, pH 8.0). After incubation for 24–48 h at room temperature, the solution was concentrated. The folded peptides were purified by RP-HPLC using a linear gradient of CH3CN in H2O containing 0.1% TFA, and lyophilized. LC/MS measurements were carried out on an LCMSIT-TOF mass spectrometer equipped with an electrospray ion source to confirm the molecular mass of the desired products.
2.3. NMR spectroscopy The solvent of the purified 13C, 15N-labeled LaIT1 was changed to 20 mM MES-NaOH, pH 6.0, and 50 mM NaCl in 90% H2O/10% D2O (v/v) and the peptide was concentrated to a final concentration of 1.5 mM by ultrafiltration using a Vivaspin-20 (MWCO 3000, Sartorius Stedim Biotech). All NMR spectra were measured at 25 °C on a Unity Inova 600-MHz NMR spectroscopy (Varian) equipped with a triple-resonance probe. The following NMR data were acquired for 13C, 15N-labeled LaIT1: 1H–15N HSQC, HNCO, HN(CO)CA, CBCA(CO)NH, HNCACB, HBHA(CO)NH, HCCH-COSY, HCCH-TOCSY, 15N-edited NOESY-HSQC (mixing time, 90 and 175 ms), and 13C-edited NOESY-HSQC (mixing time, 90 and 175 ms). In addition, 1H–15N HSQC and 15N{1H}-NOE (1H saturation time, 3 s) were acquired for 15N-labeled LaIT1. All the NMR data were processed, visualized and analyzed with the programs NMRPipe [17], NMRDraw [17] and Sparky (http://www.cgl.ucsf.edu/ home/sparky/), respectively. 2.4. Hydrogen bond restraints Hydrogen bonds were detected by the H ? D exchange experiment. 1H–15N HSQC spectrum of 1.0 mM 15N-labeled LaIT1 dissolved in 20 mM MES-NaOH, pH 6.0, and 50 mM NaCl in 90% H2O/10% D2O (v/v) were acquired. Then, the solvent was changed to 20 mM MES-NaOH, pH 6.0, and 50 mM NaCl in D2O by ultrafiltration using a Vivaspin-20 (MWCO 3000). 1H–15N HSQC spectra were acquired at 12, 24, and 36 h after the solvent exchange and the peak intensity of each residue in these 1H–15N HSQC spectra was analyzed to detect slowly exchanging amide protons. Then, their hydrogen-bonded oxygen atoms were deduced from the
2.5. Structure calculation and structural similarity search
2.7. Insect toxicity test Insect toxicity was tested using crickets (Acheta domestica, 50 mg body weight), cockroaches (Periplaneta americana,
740
S. Horita et al. / Biochemical and Biophysical Research Communications 411 (2011) 738–744
100 mg body weight) and common cutworms (Spodoptera litura, 50 mg body weight) by injection of 1–2 ll of LaIT1 and its analogs dissolved in distilled water into their abdominal region. Distilled water was injected as a control. Paralysis and death were monitored for up to 48 h. For each measurement, 6–12 animals were used. The doses required to induce paralysis (ED50) or death (LD50) in half of the test animals were determined from dose–response curves using a statistical software PRISM (GraphPad Software).
was proved to be identical to native LaIT1 in structure and activity (see Section 3.3), was used as the reference. By using heteronuclear 3D NMR data, all the observable backbone atoms (15N, 1HN, 13Ca, 1 a H , and 13C0 ) except N atoms in proline residues and C0 atoms in residues just before prolines, and almost all observable side-chain atoms (1H, 13C, 15N) were assigned. The 1H–15N HSQC spectrum of LaIT1 is shown in Fig. 1A. Two hydrogen bonds (Lys22-HNVal30O and Val30-HNLys22-O) were assigned based on the low H ? D exchange rates of amide protons and the LaIT1 structures at intermediate and final stages of structure calculations (Table S1).
3. Results and discussion 3.2. Tertiary structure 3.1. NMR structural analysis 13
C, 15N-labeled Trx-His6-ek-LaIT1 was expressed in E. coli, purified by immobilized metal affinity column, tag-removed by enterokinase, and further purified by cation exchange chromatography. The correct fold of recombinant LaIT1 was verified by the correct linkage pattern of the two intramolecular disulfide bonds (Supplementary Fig. S1) and the same spectral pattern of 2D NOESY between synthetic LaIT1 and recombinant non-labeled LaIT1 (Fig. S2). Since the amount of native LaIT1 obtained from the venom of L. australasiae is very limited, the synthetic LaIT1, which
The three-dimensional structure of LaIT1 was calculated based on 299 NOE-derived distance restraints, two disulfide bond restraints (12 distance restraints) and two hydrogen bond restraints (four distance restraints) (Biological Magnetic Resonance Bank (BMRB) code: 17681). A total of 200 structures were calculated, and the 10 structures with the lowest target functions were selected as an ensemble representing the solution structure of LaIT1 (Fig. 1B; Protein Data Bank (PDB) code: 2LDS). The restraints used and the structural statistics for the final structure are summarized in Table 1. The LaIT1 molecule contains a two-stranded antiparallel
Fig. 1. (A) 1H–15N HSQC spectrum of LaIT1. The capital letters indicate the assignments for the backbone 1H–15N groups, while the lower case letters indicate the assignments for the side chain 1H–15N groups. (B) Stereoview of the superposition of the 10 NMR structures of LaIT1. The disulfide bonds, Cys11–Cys23 and Cys17–Cys29, are shown in gray.
S. Horita et al. / Biochemical and Biophysical Research Communications 411 (2011) 738–744 Table 1 Statistics for the 10 NMR structures of LaIT1. Restraints Distance restraints derived from NOEs All Intraresidual (|ij| = 0) Sequential (|ij| = 1) Medium-range (2 6 |ij| 6 5) Long-range (|ij| > 6) Hydrogen bond restraintsa Disulfide bond restraintsb
a b c d
299 57 112 54 76 4 12
Ramachandran plot
No. of residues
Percentage (%)
Most and additionally favored Generously allowed Disallowed
239 36 10
83.6 12.9 3.6
RMS deviations of atomic coordinates Backbone atoms (Å)c,d Non-hydrogen atoms (Å)d
0.54 ± 0.15 1.24 ± 0.09
Restraint violations r.m.s. distance restraint violations (Å)
0.003 ± 0.0002
Target function CYANA target function (Å2)
1.1 ± 0.1
Two hydrogen bonds, each with two distance restraints. Two disulfide bonds, each with six distance restraints. N, Ca and C0 of each residue. For well-defined regions (residue numbers 6–30).
b-sheet (b1: Lys22–Cys23 and b2: Cys29–Val30). The two intramolecular disulfide bonds were formed at Cys11–Cys23 and Cys17–Cys29. Fig. S3 shows the r.m.s. deviation (RMSD) of Ca atom for each
741
residue in the superposed 10 conformers and 15N{1H}-NOE for each residue of LaIT1. The pairwise RMSDs for backbone and all nonhydrogen atoms are 0.54 ± 0.09 and 1.24 ± 0.15 Å, respectively, among the 10 structures when the less well-defined terminal residues (residue numbers 1–5 and 31–36) are neglected. A structural similarity search using the Dali server [20] showed that LaIT1 is structurally similar to peptides that adopt the inhibitor cystine knot (ICK)-fold or the knottin fold (Fig. 2A–C). The ICK fold is structurally characterized by the presence of a b-hairpin and a cystine knot, in which a ring formed by two disulfide bonds and their connecting backbone segments are threaded by the third disulfide bond. However, LaIT1 contains only two disulfide bonds (CysII–CysV and CysIII–CysVI) and lacks the third disulfide bond (CysI–CysIV) where the roman numbers represent the order of cysteine residues in the amino-acid sequence of the ICK-fold peptides. Due to the lack of this CysI–CysIV disulfide bond, the LaIT1 molecule does not contain a cystine knot. Thus, we call its fold an ICK-like fold. Although the third disulfide bond is conserved in all the known ICK-fold peptides (except LaIT1), it is shown that the two other disulfide bonds (CysII–CysV and CysIII–CysVI) are sufficient to stabilize most of the native structure [21,22]. Moreover, it has been reported that the disulfide bond CysI–CysIV is formed at the last step after the formation of the other two disulfide bonds [23]. These observations suggest that not only LaIT1 but also ICKfold peptides does not require the disulfide CysI–CysIV to adopt the ICK(-like) fold. Since the disulfide bond CysI–CysIV is absent in LaIT1, the N-terminal region (residues 1–3) of LaIT1 is relatively flexible as indicated by high RMSDs and 15N{1H}-NOE analysis (Fig. S3). However, the following region consisting of residues 5–10 is well
Fig. 2. (A–C) The backbone folds and disulfide bonds of LaIT1 (A), J-ACTX-Hv1c (B) and psalmotoxin 1 (C). The residues forming disulfide bonds are shown as yellow balls and sticks. b strands are shown as blue arrows. In LaIT1, two functionally important Arg residues (Arg13 and Arg15) and hydrophobic residues around the two Arg residues are shown in cyan and green, respectively. Lys16 is shown in purple. (D) Sequence alignments of LaIT1, J-ACTX-Hv1c, psalmotoxin 1, imperatoxin A and maurocalcin.
742
S. Horita et al. / Biochemical and Biophysical Research Communications 411 (2011) 738–744
defined even without stabilization by the disulfide bond. This is probably because of the presence of a turn structure stabilized by hydrogen bonds (e.g., between Ser5 and Glu9) and hydrophobic interactions (e.g., between Val4 and Gln18, and Val4 and Pro19). Since these intramolecular interactions are not strong enough to tightly pack the N-terminal region onto the central b-sheet structure, the third disulfide bond may be introduced to stabilize the conformation of N-terminal region in the typical ICK-fold peptides. 3.3. Proposed active site of LaIT1 In many insecticidal toxins having the ICK fold, basic residues play critical roles in the interaction with the target molecule [4,24,25]. In the case of LaIT1, there are eight basic residues in its sequence, and particularly three residues (Arg13, Arg15 and Lys16) among them are closely located at the loop formed between Cys11(II) and Cys17(III). In this study, each or all of these basic residues were substituted with alanine(s) to examine their significance in the insecticidal activity. The wild-type LaIT1 and its alanine-substituted analogs were synthesized by the standard Fmoc solid-phase method followed by oxidization in the redox buffer to form the two intramolecular disulfide bonds. As shown in Fig. S4, no disulfide isomers were observed when wild-type LaIT1 was synthesized. The oxidized peptide was eluted at the same retention time on RP-HPLC as native LaIT1, gave the identical molecular mass to native LaIT1 as measured by LC/MS, and possessed comparable insecticidal activity to native LaIT1, demonstrating that native and synthetic LaIT1 are identical in primary structure and equivalent in activity. The CD spectrum of each analog showed similar pattern to that of the wild-type LaIT1 (Fig. S5), indicating that the pattern of disulfide linkages in each analog is identical to that in the wildtype LaIT1 and that the overall secondary structure of each analog is similar to that of the wild-type LaIT1. As shown in Table 2, when all of Arg13, Arg15 and Lys16 were simultaneously substituted with alanines, no activity was detected even at a dose of 800 lg/g body weight. When each of these basic residues was substituted with alanine, the individual substitution of Arg13 and Arg15 resulted in a 4.2- and 12-fold reduction in insecticidal activity, respectively. On the other hand, no significant reduction in insecticidal activity was observed when Lys16 was substituted with alanine. These results demonstrate the two Arg residues, particularly Arg15, play important roles in the insecticidal activity of LaIT1, probably by directly interacting with the target molecule. 3.4. Structural similarity of LaIT1 to spider ICK-fold toxins LaIT1 is structurally similar to two spider toxins, J-ACTXs-Hv1c from the venom of the funnel-web spider ([26]; PDB code: 1DL0) and psalmotoxin 1 from the venom of the South American tarantula ([27]; PDB code: 1LMM), respectively (Fig. 2; Z score = 2.0 and RMSD = 2.5 Å for 32 Ca pairs for both pairs of molecules). LaIT1, consisting of 36 residues, has two intramolecular disulfide
Table 2 Insecticidal activity of LaIT1 and its alanine-mutated analogs. Analog b
LaIT1 R13A, R15A, K16A R13A R15A K16A a
LD50 (lg/g body weight)a
Fold reduction
22 >800 92 260 30
1.0 >730 4.2 12 1.4
Lethal toxicity against crickets. LD50 values against common cutworms and cockroaches were 52 and >400 lg/g body weight, respectively. LaIT1 induced irreversible paralysis against cockroaches (ED50 = 120 lg/g body weight). b
bonds (Cys11–Cys23 and Cys17–Cys29), whereas J-ACTXs-Hv1c, consisting of 37 residues, has four intramolecular disulfide bonds (Cys3–Cys17, Cys10–Cys22, Cys13–Cys14 and Cys16–Cys32), and psalmotoxin 1, consisting of 40 residues, has three intramolecular disulfide bonds (Cys3–Cys18, Cys10–Cys23 and Cys17–Cys33). All of LaIT1, J-ACTXs-Hv1c and psalmotoxin 1 possess two b strands and adopt the ICK(-like) fold, despite the lack of amino-acid sequence similarity between LaIT1 and J-ACTXs-Hv1c/psalmotoxin 1 (Fig. 2D). Although the backbone folds of J-ACTXs-Hv1c and psalmotoxin 1 are very similar, these spider toxins have different target molecules. J-ACTXs-Hv1c blocks invertebrate-specific calcium-activated potassium channel [28], whereas psalmotoxin 1 blocks acid-sensing sodium channels [27]. The molecular surfaces of LaIT1, J-ACTXs-Hv1c and psalmotoxin 1 are shown in Fig. 3A–C, respectively. The functionally important residues in J-ACTXsHv1c are located on a single face of the toxin; Arg8, Pro9, Cys13, Cys14 and Tyr31 are considered to form the hot spot to interact with its target molecule [25]. In the case of LaIT1, the two arginine residues, Arg13 and Arg15, were shown to contribute to the insecticidal toxicity (Table 2), probably by interacting with its target molecule(s). These basic residues are surrounded by hydrophobic residues (Phe2, Pro3, Val12 and Pro14), in a similar manner to Arg8 of J-ACTXs-Hv1c. The functionally important patch of LaIT1 containing the two arginine residues looks similar to the hot spot of J-ACTXs-Hv1c and the corresponding patch of psalmotoxin 1. As mentioned above, ICK-fold peptides exhibit a variety of biological activities, even though they have similar hot spot structures. Thus, it is difficult to know what type of ion channel LaIT1 would block, from the comparison of the functional patch structures. 3.5. Structural comparison with the scorpion ICK-fold toxins LaIT1 is the third example of scorpion toxin to adopt an ICK-like fold. The other scorpion toxins with the ICK fold are imperatoxin A (Fig. 3D; [29]; PDB code: 1IE6) and maurocalcin (Fig. 3E; [30]; PDB code: 1C6W), both of which contain three intramolecular disulfide bonds and adopt the canonical ICK folds (Fig. 2D), and are potent effectors of ryanodyne-sensitive calcium channel from skeletal muscles. The basic residues, which form a large basic patch on the imperatoxin A molecule (Fig. 3D), are important for its calcium channel inhibition [29]. The similar basic patch on the maurocalcin molecule (Fig. 3E) would be also important for its inhibitory activity. In contrast, the LaIT1 molecule does not have such a large basic patch (Fig. 3A). Thus, LaIT1 would not inhibit the ryanodynesensitive calcium channel. 3.6. Evaluation of insect toxicity of LaIT1 Our previous study demonstrated that native LaIT1 showed significant insecticidal activity against crickets, but was inactive against mice [15]. In this study, the insecticidal activity of LaIT1 was further evaluated against cockroaches (Blattodea) and common cutworms (Lepidoptera) as well as crickets (Orthoptera) to examine the species specificity in the toxicity of LaIT1. The synthetic LaIT1 showed the lethal effect on crickets and common cutworms with LD50s of 22 and 52 lg/g body weight, respectively. In contrast, no lethal effect was induced against cockroaches even after injection of 400 lg/g body weight, and only irreversible paralysis was observed 48 h after injection with an EC50 of 120 lg/g body weight. It is known that the long-chain scorpion toxins that act on insect sodium channels, such as LqhaIT and AaHIT1, have extremely high toxicity against cockroaches, whereas their toxicity is relatively weak against lepidopteran insects [1]. The different specificity of activity between LaIT1 and sodium channel-targeting toxins may reflect differences in molecular targets. Although further
S. Horita et al. / Biochemical and Biophysical Research Communications 411 (2011) 738–744
743
Fig. 3. (A–E) Molecular surfaces of LaIT1 (A), J-ACTX-Hv1c (B), psalmotoxin 1 (C), imperatoxin A (D) and maurocalcin (E). Functionally important basic and hydrophobic residues colored in blue and green, respectively, and shown with labels. Acidic, basic and hydrophobic residues are shown in pink, light blue and light green, respectively.
mode of action studies are required, it is possible that LaIT1 acts on insect potassium or calcium channels rather than sodium channels, as shown for many spider toxins including J-ACTXs-Hv1c. While this manuscript was in reviewing process, the isolation, amino-acid sequence and solution structure of U1-LITX-Lw1a, which was isolated from the venom of the scorpion Liocheles waigiensis, was reported online [31]. U1-LITX-Lw1a possesses 92% sequence identity to LaIT1, and the solution structure of U1-LITXLw1a is similar to that of LaIT1 with an RMSD of 2.0 Å. However, there is no description on amino-acid residues important for the insecticidal activity of U1-LITX-Lw1a in that report [31]. Acknowledgments This work was supported in part by the Targeted Proteins Research Program (TPRP) of the Ministry of Education, Culture, Sports, Science and Technology of Japan, and Grants-in-Aid for Scientific Research from the Ministry of Education, Culture, Sports, Science and Technology of Japan. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at doi:10.1016/j.bbrc.2011.07.016. References [1] M.E. De Lima, S.G. Figueiredo, A.M. Pimenta, et al., Peptides of arachnid venoms with insecticidal activity targeting sodium channels, Comp. Biochem. Physiol. C Toxicol. Pharmacol. 146 (2007) 264–279. [2] P. Escoubas, L. Rash, Tarantulas: eight-legged pharmacists and combinatorial chemists, Toxicon 43 (2004) 555–574. [3] S. Mouhat, N. Andreotti, B. Jouirou, et al., Animal toxins acting on voltage-gated potassium channels, Curr. Pharm. Design 14 (2008) 2503–2518. [4] H.W. Tedford, B.L. Sollod, F. Maggio, et al., Australian funnel-web spiders: master insecticide chemists, Toxicon 43 (2004) 601–618. [5] W.A. Catterall, S. Cestele, V. Yarov-Yarovoy, et al., Toxicon 49 (2007) 124–141.
[6] E. Zlotkin, Scorpion venoms, in: L.I. Gilbert, K. Iatrou, S.S. Gill (Eds.), Comprehensive Molecular Insect Science, Elsevier B.V., Oxford, 2005, pp. 173–220. [7] L.D. Possani, E. Merino, M. Corona, et al., Peptides and genes coding for scorpion toxins that affect ion-channels, Biochimie 82 (2000) 861–868. [8] C. Goudet, C.W. Chi, J. Tytgat, An overview of toxins and genes from the venom of the Asian scorpion Buthus martensi Karsch, Toxicon 40 (2002) 1239–1258. [9] A. Menez, Functional architectures of animal toxins: a clue to drug design?, Toxicon 36 (1998) 1557–1572 [10] S.Y. Zhu, H. Darbon, K. Dyason, et al., Evolutionary origin of inhibitor cystine knot peptides, FASEB J. 17 (2003) 1765–1767. [11] R.C. Rodriguez de la Vega, L.D. Possani, Overview of scorpion toxins specific for Na+ channels and related peptides: biodiversity structure–function relationships and evolution, Toxicon 46 (2005) 831–844. [12] R.C. Rodriguez de la Vega, L.D. Possani, Current views on scorpion toxins specific for K+-channels, Toxicon 43 (2004) 865–875. [13] L. Chiche, A. Heitz, J.C. Gelly, et al., Squash inhibitors: from structural motifs to macrocyclic knottins, Curr. Protein Pept. Sci. 5 (2004) 341–349. [14] D.J. Craik, N.L. Daly, C. Waine, The cystine knot motif in toxins and implications for drug design, Toxicon 39 (2001) 43–60. [15] N. Matsushita, M. Miyashita, A. Sakai, et al., Purification and characterization of a novel short-chain insecticidal toxin with two disulfide bridges from the venom of the scorpion L. australasiae, Toxicon 50 (2007) 861–867. [16] E.C.N. Silva, T.S. Camargos, A.Q. Maranhao, et al., Cloning and characterization of cDNA sequences encoding for new venom peptides of the Brazilian scorpion Opisthacanthus cayaporum, Toxicon 54 (2009) 252–261. [17] F. Delaglio, S. Grzesiek, G.W. Vuister, et al., NMRpipe – a multidimensional spectral processing system based on unix pipes, J. Biomol. NMR 6 (1995) 277– 293. [18] J. Jee, P. Guntert, Influence of the completeness of chemical shift assignments on NMR structures obtained with automated NOE assignment, J. Struct. Funct. Genomics 4 (2003) 179–189. [19] R. Koradi, M. Billeter, K. Wuthrich, MOLMOL: a program for display and analysis of macromolecular structures, J. Mol. Graphics 14 (1996) 51–55. [20] L. Holm, P. Rosenstrom, Dali server: conservation mapping in 3D, Nucleic Acids Res. 38 (2010) W545–W549. [21] A. Heitz, D. Le-Nguyen, B. Castro, et al., Conformational study of a native monodisulfide bridge analogue of EETI II, Lett. Pept. Sci. 4 (1997) 245–249. [22] A. Heitz, D. Le-Nguyen, L. Chiche, Min-21 and Min-23, the smallest peptides that fold like a cystine-stabilized beta-sheet motif: design, solution structure, and thermal stability, Biochemistry 38 (1999) 10615–10625. [23] D. Le-Nguyen, A. Heitz, L. Chiche, et al., Characterization and 2D NMR-study of the stable [9–21, 15–27] 2 disulfide intermediate in the folding of the three disulfide trypsin-inhibitor EETI-II, Protein Sci. 2 (1993) 165–174. [24] I. Karbat, F. Frolow, O. Froy, et al., Molecular basis of the high insecticidal potency of scorpion alpha-toxins, J. Biol. Chem. 279 (2004) 31679–31686.
744
S. Horita et al. / Biochemical and Biophysical Research Communications 411 (2011) 738–744
[25] F. Maggio, G.F. King, Scanning mutagenesis of a Janus-faced atracotoxin reveals a bipartite surface patch that is essential for neurotoxic function, J. Biol. Chem. 277 (2002) 22806–22813. [26] X.H. Wang, M. Connor, R. Smith, et al., Discovery and characterization of a family of insecticidal neurotoxins with a rare vicinal disulfide bridge, Nat. Struct. Biol. 7 (2000) 505–513. [27] P. Escoubas, C. Bernard, G. Lambeau, et al., Recombinant production and solution structure of PcTx1, the specific peptide inhibitor of ASIC1a protongated cation channels, Protein Sci. 12 (2003) 1332–1343. [28] S.J. Gunning, F. Maggio, M.J. Windley, et al., The Janus-faced atracotoxins are specific blockers of invertebrate KCa channels, FEBS J. 275 (2008) 4045–4059.
[29] C.W. Lee, E.H. Lee, K. Takeuchi, et al., Molecular basis of the high-affinity activation of type 1 ryanodine receptors by imperatoxin A, Biochem. J. 377 (2004) 385–394. [30] A. Mosbah, R. Kharrat, Z. Fajloun, et al., A new fold in the scorpion toxin family associated with an activity on a ryanodine-sensitive calcium channel, Proteins 40 (2000) 436–442. [31] J.J. Smith, J.M. Hill, M.J. Little, et al., Unique scorpion toxin with a putative ancestral fold provides insight into evolution of the inhibitor cystine knot motif, Proc. Natl. Acad. Sci. USA 108 (2011) 10478–10483.
Contents lists available at ScienceDirect
Biochemical and Biophysical Research Communications journal homepage: www.elsevier.com/locate/ybbrc
Solution structure of a short-chain insecticidal toxin LaIT1 from the venom of scorpion Liocheles australasiae Shoichiro Horita a, Nobuto Matsushita b, Tomoyuki Kawachi b, Reed Ayabe b, Masahiro Miyashita b, Takuya Miyakawa a, Yoshiaki Nakagawa b, Koji Nagata a, Hisashi Miyagawa b, Masaru Tanokura a,⇑ a b
Department of Applied Biological Chemistry, Graduate School of Agricultural and Life Sciences, University of Tokyo, 1-1-1 Yayoi, Bunkyo-ku, Tokyo 113-8657, Japan Division of Applied Life Sciences, Graduate School of Agriculture, Kyoto University, Kyoto 606-8502, Japan
a r t i c l e
i n f o
Article history: Received 3 July 2011 Available online 18 July 2011 Keywords: Inhibitory cystine knot fold LaIT1 Nuclear magnetic resonance Scorpion toxin Solution structure
a b s t r a c t The solution structure of an insecticidal toxin LaIT1, a 36-residue peptide with a unique amino-acid sequence and two disulfide bonds, isolated from the venom of the scorpion Liocheles australasiae was determined by heteronuclear NMR spectroscopy. Structural similarity search showed that LaIT1 exhibits an inhibitory cystine knot (ICK)-like fold, which usually contains three or more disulfide bonds. Mutational analysis has revealed that two Arg residues of LaIT1, Arg13 and Arg15, play significant roles in insecticidal activity. Ó 2011 Elsevier Inc. All rights reserved.
1. Introduction Most toxins isolated from animal venoms are small peptides that bind to ion channels, such as sodium, potassium, calcium and chloride channels, with high affinity and specificity [1–5]. These toxins block or modulate gating mechanisms of ion channels to immobilize prey or predators. Scorpion venom contains a variety of peptides that are toxic to mammals, insects and crustaceans [6]. To date, more than 600 peptides have been identified from scorpion venoms. Scorpion toxins can be divided into two groups, a 6–8 kDa long-chain peptide group and a 3–5 kDa short-chain peptide group, based on the molecular size [7,8], and also classified into two groups, a cystine-stabilized a-helix/b-sheet (CSa/b) group and an inhibitor cystine knot (ICK) group, based on the backbone fold [9,10]. The CSa/b fold consists of a short a-helix linked to a double- or triple-stranded antiparallel b-sheet stabilized by three or four disulfide bonds. This fold is observed in many scorpion toxins acting on potassium and sodium channels [11,12]. The ICK fold, also referred to as the knottin fold, consists of a double- or triplestranded antiparallel b-sheet stabilized by three disulfide bonds, and forms a b-hairpin and a cystine knot [13,14]. In contrast to the CSa/b fold, the ICK fold is found in only a few scorpion toxins [10], although the ICK-fold peptides have been isolated from many species such as plants, marine cone snails and spiders.
LaIT1 is an insecticidal short-chain toxin isolated from the venom of the scorpion Liocheles australasiae [15]. This toxin consists of 36 amino-acid residues and two intramolecular disulfide bonds with a molecular mass of 4200 Da, and shares a sequence similarity to only one peptide OcyC10 (64% identical) isolated from the Brazilian scorpion venom [16]. Unlike other scorpion insecticidal toxins, LaIT1 and OcyC10 contain only two disulfide bonds. The unique amino-acid sequence and the lack of tertiary structural information prompted us to explore the three-dimensional structure of LaIT1 as well as to further characterize its insect toxicity. In this study, we have analyzed the solution structure by heteronuclear NMR spectroscopy using 15N-labeled and 13C, 15N-labeled LaIT1 that are produced in an Escherichia coli expression system. The NMR analysis has revealed that LaIT1 adopts an ICK-like fold although it contains only two disulfide bonds. We have also examined the species specificity in insect toxicity of LaIT1 and identified its amino-acid residues important for the insecticidal activity. We discuss the structure and function of LaIT1 by comparing the tertiary structures and the target ion channels with structurally similar scorpion and spider toxins.
2. Materials and methods 2.1. Expression and purification of
Abbreviations: ICK, inhibitory cystine knot; CSa/b, cystine-stabilized a-helix/bsheet. ⇑ Corresponding author. Fax: +81 3 5841 8023. E-mail address: [email protected] (M. Tanokura). 0006-291X/$ - see front matter Ó 2011 Elsevier Inc. All rights reserved. doi:10.1016/j.bbrc.2011.07.016
15
N- and
13
C,
15
N-labeled LaIT1
The double stranded DNA encoding LaIT1, ATGGATTTTCCGCT GAGCAAAGAATATGAAACCTGCGTGCGCCCGCGCAAATGTCAGCCGC CGCTGAAATGCAACAAAGCGCAGATTTGTGTGGACCCGAAAAAGGGC
S. Horita et al. / Biochemical and Biophysical Research Communications 411 (2011) 738–744
739
LaIT1 structures at intermediate and final stages in the structure calculation. Hydrogen-bond distance restraints were set as rHN–O = 1.8–2.5 and rN–O = 2.7–3.5 Å.
TGGtaataa, where taataa is the tandem stop codons, was synthesized by mixing the following four oligo nucleotides followed by thermal cycling: LaIT1-S1, cgcgcgcgCCATGGGATTTTCCGCTGAGCA AAGAATATGAAACCT; LaIT1-A2, GCGGCGGCTGACATTTGCGCGGGC GCACGCAGGTTTCATATTCTT; LaIT1-S3, AATGTCAGCCGCCGCTGAA ATGCAACAAAGCGCAGATTTGTGTGG; and LaIT1-A4, cgcGAATTC ttattaCCAGCCCTTTTTCGGGTCCACACAAATCTGCG. The NcoI and EcoRI sites (underlined in the above oligonucleotides) were used for the cloning of the DNA encoding LaIT1 to the expression vector pET-32a (Novagen). Non-labeled, 15N-labeled and 13C, 15N-labeled Trx-His6-ek-LaIT1, where ‘‘Trx’’ and ‘‘ek’’ represents the thioredoxin-tag and the enterokinase recognition site, respectively, were expressed in E. coli BL21(DE3) (Novagen) grown in LB medium, M9 minimal medium containing 15N-labeled ammonium chloride and in M9 minimal medium containing 13C-labeled glucose and 15N-labeled ammonium chloride, respectively, by adding a final concentration of 1 mM IPTG and cultivating at 37 °C for 3 h. Cells were harvested by centrifugation, resuspended in lysis buffer (50 mM Tris–HCl, pH 8.0, 300 mM NaCl, 10 mM imidazole), and disrupted by sonication. After centrifugation, Trx-His6-ek-LaIT1 was purified with a Ni Sepharose 6 Fast Flow (GE Healthcare) column. The Trx-His6-tag was removed by enterokinase (Invitrogen). LaIT1 was purified by cation exchange chromatography using a HiTrap SP column (GE Healthcare).
Inter-proton distance restraints were derived from peak intensities in the 15N-edited NOESY-HSQC and 13C-edited NOESY-HSQC spectra of 13C, 15N-labeled LaIT1 with a mixing time of 90 ms. The cross-peak intensities were translated into inter-proton distances based on the relationships, NOE / (distance)6 and the standard distance between Hai and Haj in the parallel b-sheet of 2.3 Å. Hydrogen-bond distance restraints were also added to the structure calculation as described above. Structure calculation was performed using CYANA [18]. A total of 200 conformers were annealed in 10,000 steps of torsion angle dynamics calculations, of which 10 conformers with the lowest values in the target function were used to represent the solution structure of LaIT1. The conformer with the lowest target function was used as the representative structure of LaIT1. The tertiary structure was visualized with the programs Molmol [19] and PyMol (http://pymol.sourceforge.net/). The representative structure with the lowest target function was submitted to the Dali server [20] to search for proteins structurally similar to LaIT1.
2.2. Determination of disulfide bond linkages in recombinant LaIT1
2.6. Chemical synthesis of LaIT1 and its analogs
To analyze the disulfide pattern of recombinant LaIT1, non-labeled recombinant LaIT1 was first digested with endoproteinase Lys-C (Wako Pure Chemical Industries, Osaka) in 25 mM Tris– HCl, pH 7.1, without reduction or alkylation of disulfide bonds. Then, the molecular masses of fragments were analyzed using an LCMS-IT-TOF mass spectrometer (Shimadzu, Kyoto) equipped with an electrospray ion source.
LaIT1 and its analogs were synthesized by the Fmoc-based solid-phase peptide synthesis method using Wang-PEG resin (Watanabe Chemical Industries, Hiroshima). The C-terminal amino acid was attached to the resin by the symmetrical anhydride method catalyzed by 4-dimethylaminopyridine (DMAP). Symmetrical anhydrides of the Fmoc-protected amino acids were prepared in a reaction with 0.5 eq. of a diisopropylcarbodiimide in anhydrous dichloromethane. The symmetrical anhydrides (5 eq.) and DMAP (0.1 eq.) in anhydrous N,N-dimethylformamide (DMF) were added to the resin and incubated for 1 h at room temperature. After the resin was washed with DMF, the Fmoc group of the amino acid attached to the resin was deprotected with 20% piperidine in DMF. Fmoc-protected amino acids (3 eq.), O-(benzotriazol-1-yl)N,N,N0 ,N0 -tetramethyluronium hexafluorophosphate (HBTU, 3 eq.), and 1-hydroxybenzotriazole (3 eq.) in DMF were added to the resin. After N,N-diisopropylethylamine (6 eq.) was added to the resins, the reaction mixture was incubated for 1.5 h at room temperature. This deprotection/coupling cycle was repeated until the coupling of all residues was finished. Removal of side-chain protecting groups and cleavage of the peptide from the resin was carried out with trifluoroacetic acid (TFA)/triisopropylsilane/ water/ethandithiol [94/2/2.5/2.5(v/v)] for 2 h at room temperature. After the resin was filtered off, cold diethyl ether was added to the filtrate to precipitate the cleaved peptides. The precipitated peptides were washed with cold diethyl ether twice and dried in vacuo. The crude linear peptides were then oxidized to form disulfide linkages in the glutathione redox buffer (1 mM reduced glutathione and 0.1 mM oxidized glutathione in 0.2 M Tris–HCl, pH 8.0). After incubation for 24–48 h at room temperature, the solution was concentrated. The folded peptides were purified by RP-HPLC using a linear gradient of CH3CN in H2O containing 0.1% TFA, and lyophilized. LC/MS measurements were carried out on an LCMSIT-TOF mass spectrometer equipped with an electrospray ion source to confirm the molecular mass of the desired products.
2.3. NMR spectroscopy The solvent of the purified 13C, 15N-labeled LaIT1 was changed to 20 mM MES-NaOH, pH 6.0, and 50 mM NaCl in 90% H2O/10% D2O (v/v) and the peptide was concentrated to a final concentration of 1.5 mM by ultrafiltration using a Vivaspin-20 (MWCO 3000, Sartorius Stedim Biotech). All NMR spectra were measured at 25 °C on a Unity Inova 600-MHz NMR spectroscopy (Varian) equipped with a triple-resonance probe. The following NMR data were acquired for 13C, 15N-labeled LaIT1: 1H–15N HSQC, HNCO, HN(CO)CA, CBCA(CO)NH, HNCACB, HBHA(CO)NH, HCCH-COSY, HCCH-TOCSY, 15N-edited NOESY-HSQC (mixing time, 90 and 175 ms), and 13C-edited NOESY-HSQC (mixing time, 90 and 175 ms). In addition, 1H–15N HSQC and 15N{1H}-NOE (1H saturation time, 3 s) were acquired for 15N-labeled LaIT1. All the NMR data were processed, visualized and analyzed with the programs NMRPipe [17], NMRDraw [17] and Sparky (http://www.cgl.ucsf.edu/ home/sparky/), respectively. 2.4. Hydrogen bond restraints Hydrogen bonds were detected by the H ? D exchange experiment. 1H–15N HSQC spectrum of 1.0 mM 15N-labeled LaIT1 dissolved in 20 mM MES-NaOH, pH 6.0, and 50 mM NaCl in 90% H2O/10% D2O (v/v) were acquired. Then, the solvent was changed to 20 mM MES-NaOH, pH 6.0, and 50 mM NaCl in D2O by ultrafiltration using a Vivaspin-20 (MWCO 3000). 1H–15N HSQC spectra were acquired at 12, 24, and 36 h after the solvent exchange and the peak intensity of each residue in these 1H–15N HSQC spectra was analyzed to detect slowly exchanging amide protons. Then, their hydrogen-bonded oxygen atoms were deduced from the
2.5. Structure calculation and structural similarity search
2.7. Insect toxicity test Insect toxicity was tested using crickets (Acheta domestica, 50 mg body weight), cockroaches (Periplaneta americana,
740
S. Horita et al. / Biochemical and Biophysical Research Communications 411 (2011) 738–744
100 mg body weight) and common cutworms (Spodoptera litura, 50 mg body weight) by injection of 1–2 ll of LaIT1 and its analogs dissolved in distilled water into their abdominal region. Distilled water was injected as a control. Paralysis and death were monitored for up to 48 h. For each measurement, 6–12 animals were used. The doses required to induce paralysis (ED50) or death (LD50) in half of the test animals were determined from dose–response curves using a statistical software PRISM (GraphPad Software).
was proved to be identical to native LaIT1 in structure and activity (see Section 3.3), was used as the reference. By using heteronuclear 3D NMR data, all the observable backbone atoms (15N, 1HN, 13Ca, 1 a H , and 13C0 ) except N atoms in proline residues and C0 atoms in residues just before prolines, and almost all observable side-chain atoms (1H, 13C, 15N) were assigned. The 1H–15N HSQC spectrum of LaIT1 is shown in Fig. 1A. Two hydrogen bonds (Lys22-HNVal30O and Val30-HNLys22-O) were assigned based on the low H ? D exchange rates of amide protons and the LaIT1 structures at intermediate and final stages of structure calculations (Table S1).
3. Results and discussion 3.2. Tertiary structure 3.1. NMR structural analysis 13
C, 15N-labeled Trx-His6-ek-LaIT1 was expressed in E. coli, purified by immobilized metal affinity column, tag-removed by enterokinase, and further purified by cation exchange chromatography. The correct fold of recombinant LaIT1 was verified by the correct linkage pattern of the two intramolecular disulfide bonds (Supplementary Fig. S1) and the same spectral pattern of 2D NOESY between synthetic LaIT1 and recombinant non-labeled LaIT1 (Fig. S2). Since the amount of native LaIT1 obtained from the venom of L. australasiae is very limited, the synthetic LaIT1, which
The three-dimensional structure of LaIT1 was calculated based on 299 NOE-derived distance restraints, two disulfide bond restraints (12 distance restraints) and two hydrogen bond restraints (four distance restraints) (Biological Magnetic Resonance Bank (BMRB) code: 17681). A total of 200 structures were calculated, and the 10 structures with the lowest target functions were selected as an ensemble representing the solution structure of LaIT1 (Fig. 1B; Protein Data Bank (PDB) code: 2LDS). The restraints used and the structural statistics for the final structure are summarized in Table 1. The LaIT1 molecule contains a two-stranded antiparallel
Fig. 1. (A) 1H–15N HSQC spectrum of LaIT1. The capital letters indicate the assignments for the backbone 1H–15N groups, while the lower case letters indicate the assignments for the side chain 1H–15N groups. (B) Stereoview of the superposition of the 10 NMR structures of LaIT1. The disulfide bonds, Cys11–Cys23 and Cys17–Cys29, are shown in gray.
S. Horita et al. / Biochemical and Biophysical Research Communications 411 (2011) 738–744 Table 1 Statistics for the 10 NMR structures of LaIT1. Restraints Distance restraints derived from NOEs All Intraresidual (|ij| = 0) Sequential (|ij| = 1) Medium-range (2 6 |ij| 6 5) Long-range (|ij| > 6) Hydrogen bond restraintsa Disulfide bond restraintsb
a b c d
299 57 112 54 76 4 12
Ramachandran plot
No. of residues
Percentage (%)
Most and additionally favored Generously allowed Disallowed
239 36 10
83.6 12.9 3.6
RMS deviations of atomic coordinates Backbone atoms (Å)c,d Non-hydrogen atoms (Å)d
0.54 ± 0.15 1.24 ± 0.09
Restraint violations r.m.s. distance restraint violations (Å)
0.003 ± 0.0002
Target function CYANA target function (Å2)
1.1 ± 0.1
Two hydrogen bonds, each with two distance restraints. Two disulfide bonds, each with six distance restraints. N, Ca and C0 of each residue. For well-defined regions (residue numbers 6–30).
b-sheet (b1: Lys22–Cys23 and b2: Cys29–Val30). The two intramolecular disulfide bonds were formed at Cys11–Cys23 and Cys17–Cys29. Fig. S3 shows the r.m.s. deviation (RMSD) of Ca atom for each
741
residue in the superposed 10 conformers and 15N{1H}-NOE for each residue of LaIT1. The pairwise RMSDs for backbone and all nonhydrogen atoms are 0.54 ± 0.09 and 1.24 ± 0.15 Å, respectively, among the 10 structures when the less well-defined terminal residues (residue numbers 1–5 and 31–36) are neglected. A structural similarity search using the Dali server [20] showed that LaIT1 is structurally similar to peptides that adopt the inhibitor cystine knot (ICK)-fold or the knottin fold (Fig. 2A–C). The ICK fold is structurally characterized by the presence of a b-hairpin and a cystine knot, in which a ring formed by two disulfide bonds and their connecting backbone segments are threaded by the third disulfide bond. However, LaIT1 contains only two disulfide bonds (CysII–CysV and CysIII–CysVI) and lacks the third disulfide bond (CysI–CysIV) where the roman numbers represent the order of cysteine residues in the amino-acid sequence of the ICK-fold peptides. Due to the lack of this CysI–CysIV disulfide bond, the LaIT1 molecule does not contain a cystine knot. Thus, we call its fold an ICK-like fold. Although the third disulfide bond is conserved in all the known ICK-fold peptides (except LaIT1), it is shown that the two other disulfide bonds (CysII–CysV and CysIII–CysVI) are sufficient to stabilize most of the native structure [21,22]. Moreover, it has been reported that the disulfide bond CysI–CysIV is formed at the last step after the formation of the other two disulfide bonds [23]. These observations suggest that not only LaIT1 but also ICKfold peptides does not require the disulfide CysI–CysIV to adopt the ICK(-like) fold. Since the disulfide bond CysI–CysIV is absent in LaIT1, the N-terminal region (residues 1–3) of LaIT1 is relatively flexible as indicated by high RMSDs and 15N{1H}-NOE analysis (Fig. S3). However, the following region consisting of residues 5–10 is well
Fig. 2. (A–C) The backbone folds and disulfide bonds of LaIT1 (A), J-ACTX-Hv1c (B) and psalmotoxin 1 (C). The residues forming disulfide bonds are shown as yellow balls and sticks. b strands are shown as blue arrows. In LaIT1, two functionally important Arg residues (Arg13 and Arg15) and hydrophobic residues around the two Arg residues are shown in cyan and green, respectively. Lys16 is shown in purple. (D) Sequence alignments of LaIT1, J-ACTX-Hv1c, psalmotoxin 1, imperatoxin A and maurocalcin.
742
S. Horita et al. / Biochemical and Biophysical Research Communications 411 (2011) 738–744
defined even without stabilization by the disulfide bond. This is probably because of the presence of a turn structure stabilized by hydrogen bonds (e.g., between Ser5 and Glu9) and hydrophobic interactions (e.g., between Val4 and Gln18, and Val4 and Pro19). Since these intramolecular interactions are not strong enough to tightly pack the N-terminal region onto the central b-sheet structure, the third disulfide bond may be introduced to stabilize the conformation of N-terminal region in the typical ICK-fold peptides. 3.3. Proposed active site of LaIT1 In many insecticidal toxins having the ICK fold, basic residues play critical roles in the interaction with the target molecule [4,24,25]. In the case of LaIT1, there are eight basic residues in its sequence, and particularly three residues (Arg13, Arg15 and Lys16) among them are closely located at the loop formed between Cys11(II) and Cys17(III). In this study, each or all of these basic residues were substituted with alanine(s) to examine their significance in the insecticidal activity. The wild-type LaIT1 and its alanine-substituted analogs were synthesized by the standard Fmoc solid-phase method followed by oxidization in the redox buffer to form the two intramolecular disulfide bonds. As shown in Fig. S4, no disulfide isomers were observed when wild-type LaIT1 was synthesized. The oxidized peptide was eluted at the same retention time on RP-HPLC as native LaIT1, gave the identical molecular mass to native LaIT1 as measured by LC/MS, and possessed comparable insecticidal activity to native LaIT1, demonstrating that native and synthetic LaIT1 are identical in primary structure and equivalent in activity. The CD spectrum of each analog showed similar pattern to that of the wild-type LaIT1 (Fig. S5), indicating that the pattern of disulfide linkages in each analog is identical to that in the wildtype LaIT1 and that the overall secondary structure of each analog is similar to that of the wild-type LaIT1. As shown in Table 2, when all of Arg13, Arg15 and Lys16 were simultaneously substituted with alanines, no activity was detected even at a dose of 800 lg/g body weight. When each of these basic residues was substituted with alanine, the individual substitution of Arg13 and Arg15 resulted in a 4.2- and 12-fold reduction in insecticidal activity, respectively. On the other hand, no significant reduction in insecticidal activity was observed when Lys16 was substituted with alanine. These results demonstrate the two Arg residues, particularly Arg15, play important roles in the insecticidal activity of LaIT1, probably by directly interacting with the target molecule. 3.4. Structural similarity of LaIT1 to spider ICK-fold toxins LaIT1 is structurally similar to two spider toxins, J-ACTXs-Hv1c from the venom of the funnel-web spider ([26]; PDB code: 1DL0) and psalmotoxin 1 from the venom of the South American tarantula ([27]; PDB code: 1LMM), respectively (Fig. 2; Z score = 2.0 and RMSD = 2.5 Å for 32 Ca pairs for both pairs of molecules). LaIT1, consisting of 36 residues, has two intramolecular disulfide
Table 2 Insecticidal activity of LaIT1 and its alanine-mutated analogs. Analog b
LaIT1 R13A, R15A, K16A R13A R15A K16A a
LD50 (lg/g body weight)a
Fold reduction
22 >800 92 260 30
1.0 >730 4.2 12 1.4
Lethal toxicity against crickets. LD50 values against common cutworms and cockroaches were 52 and >400 lg/g body weight, respectively. LaIT1 induced irreversible paralysis against cockroaches (ED50 = 120 lg/g body weight). b
bonds (Cys11–Cys23 and Cys17–Cys29), whereas J-ACTXs-Hv1c, consisting of 37 residues, has four intramolecular disulfide bonds (Cys3–Cys17, Cys10–Cys22, Cys13–Cys14 and Cys16–Cys32), and psalmotoxin 1, consisting of 40 residues, has three intramolecular disulfide bonds (Cys3–Cys18, Cys10–Cys23 and Cys17–Cys33). All of LaIT1, J-ACTXs-Hv1c and psalmotoxin 1 possess two b strands and adopt the ICK(-like) fold, despite the lack of amino-acid sequence similarity between LaIT1 and J-ACTXs-Hv1c/psalmotoxin 1 (Fig. 2D). Although the backbone folds of J-ACTXs-Hv1c and psalmotoxin 1 are very similar, these spider toxins have different target molecules. J-ACTXs-Hv1c blocks invertebrate-specific calcium-activated potassium channel [28], whereas psalmotoxin 1 blocks acid-sensing sodium channels [27]. The molecular surfaces of LaIT1, J-ACTXs-Hv1c and psalmotoxin 1 are shown in Fig. 3A–C, respectively. The functionally important residues in J-ACTXsHv1c are located on a single face of the toxin; Arg8, Pro9, Cys13, Cys14 and Tyr31 are considered to form the hot spot to interact with its target molecule [25]. In the case of LaIT1, the two arginine residues, Arg13 and Arg15, were shown to contribute to the insecticidal toxicity (Table 2), probably by interacting with its target molecule(s). These basic residues are surrounded by hydrophobic residues (Phe2, Pro3, Val12 and Pro14), in a similar manner to Arg8 of J-ACTXs-Hv1c. The functionally important patch of LaIT1 containing the two arginine residues looks similar to the hot spot of J-ACTXs-Hv1c and the corresponding patch of psalmotoxin 1. As mentioned above, ICK-fold peptides exhibit a variety of biological activities, even though they have similar hot spot structures. Thus, it is difficult to know what type of ion channel LaIT1 would block, from the comparison of the functional patch structures. 3.5. Structural comparison with the scorpion ICK-fold toxins LaIT1 is the third example of scorpion toxin to adopt an ICK-like fold. The other scorpion toxins with the ICK fold are imperatoxin A (Fig. 3D; [29]; PDB code: 1IE6) and maurocalcin (Fig. 3E; [30]; PDB code: 1C6W), both of which contain three intramolecular disulfide bonds and adopt the canonical ICK folds (Fig. 2D), and are potent effectors of ryanodyne-sensitive calcium channel from skeletal muscles. The basic residues, which form a large basic patch on the imperatoxin A molecule (Fig. 3D), are important for its calcium channel inhibition [29]. The similar basic patch on the maurocalcin molecule (Fig. 3E) would be also important for its inhibitory activity. In contrast, the LaIT1 molecule does not have such a large basic patch (Fig. 3A). Thus, LaIT1 would not inhibit the ryanodynesensitive calcium channel. 3.6. Evaluation of insect toxicity of LaIT1 Our previous study demonstrated that native LaIT1 showed significant insecticidal activity against crickets, but was inactive against mice [15]. In this study, the insecticidal activity of LaIT1 was further evaluated against cockroaches (Blattodea) and common cutworms (Lepidoptera) as well as crickets (Orthoptera) to examine the species specificity in the toxicity of LaIT1. The synthetic LaIT1 showed the lethal effect on crickets and common cutworms with LD50s of 22 and 52 lg/g body weight, respectively. In contrast, no lethal effect was induced against cockroaches even after injection of 400 lg/g body weight, and only irreversible paralysis was observed 48 h after injection with an EC50 of 120 lg/g body weight. It is known that the long-chain scorpion toxins that act on insect sodium channels, such as LqhaIT and AaHIT1, have extremely high toxicity against cockroaches, whereas their toxicity is relatively weak against lepidopteran insects [1]. The different specificity of activity between LaIT1 and sodium channel-targeting toxins may reflect differences in molecular targets. Although further
S. Horita et al. / Biochemical and Biophysical Research Communications 411 (2011) 738–744
743
Fig. 3. (A–E) Molecular surfaces of LaIT1 (A), J-ACTX-Hv1c (B), psalmotoxin 1 (C), imperatoxin A (D) and maurocalcin (E). Functionally important basic and hydrophobic residues colored in blue and green, respectively, and shown with labels. Acidic, basic and hydrophobic residues are shown in pink, light blue and light green, respectively.
mode of action studies are required, it is possible that LaIT1 acts on insect potassium or calcium channels rather than sodium channels, as shown for many spider toxins including J-ACTXs-Hv1c. While this manuscript was in reviewing process, the isolation, amino-acid sequence and solution structure of U1-LITX-Lw1a, which was isolated from the venom of the scorpion Liocheles waigiensis, was reported online [31]. U1-LITX-Lw1a possesses 92% sequence identity to LaIT1, and the solution structure of U1-LITXLw1a is similar to that of LaIT1 with an RMSD of 2.0 Å. However, there is no description on amino-acid residues important for the insecticidal activity of U1-LITX-Lw1a in that report [31]. Acknowledgments This work was supported in part by the Targeted Proteins Research Program (TPRP) of the Ministry of Education, Culture, Sports, Science and Technology of Japan, and Grants-in-Aid for Scientific Research from the Ministry of Education, Culture, Sports, Science and Technology of Japan. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at doi:10.1016/j.bbrc.2011.07.016. References [1] M.E. De Lima, S.G. Figueiredo, A.M. Pimenta, et al., Peptides of arachnid venoms with insecticidal activity targeting sodium channels, Comp. Biochem. Physiol. C Toxicol. Pharmacol. 146 (2007) 264–279. [2] P. Escoubas, L. Rash, Tarantulas: eight-legged pharmacists and combinatorial chemists, Toxicon 43 (2004) 555–574. [3] S. Mouhat, N. Andreotti, B. Jouirou, et al., Animal toxins acting on voltage-gated potassium channels, Curr. Pharm. Design 14 (2008) 2503–2518. [4] H.W. Tedford, B.L. Sollod, F. Maggio, et al., Australian funnel-web spiders: master insecticide chemists, Toxicon 43 (2004) 601–618. [5] W.A. Catterall, S. Cestele, V. Yarov-Yarovoy, et al., Toxicon 49 (2007) 124–141.
[6] E. Zlotkin, Scorpion venoms, in: L.I. Gilbert, K. Iatrou, S.S. Gill (Eds.), Comprehensive Molecular Insect Science, Elsevier B.V., Oxford, 2005, pp. 173–220. [7] L.D. Possani, E. Merino, M. Corona, et al., Peptides and genes coding for scorpion toxins that affect ion-channels, Biochimie 82 (2000) 861–868. [8] C. Goudet, C.W. Chi, J. Tytgat, An overview of toxins and genes from the venom of the Asian scorpion Buthus martensi Karsch, Toxicon 40 (2002) 1239–1258. [9] A. Menez, Functional architectures of animal toxins: a clue to drug design?, Toxicon 36 (1998) 1557–1572 [10] S.Y. Zhu, H. Darbon, K. Dyason, et al., Evolutionary origin of inhibitor cystine knot peptides, FASEB J. 17 (2003) 1765–1767. [11] R.C. Rodriguez de la Vega, L.D. Possani, Overview of scorpion toxins specific for Na+ channels and related peptides: biodiversity structure–function relationships and evolution, Toxicon 46 (2005) 831–844. [12] R.C. Rodriguez de la Vega, L.D. Possani, Current views on scorpion toxins specific for K+-channels, Toxicon 43 (2004) 865–875. [13] L. Chiche, A. Heitz, J.C. Gelly, et al., Squash inhibitors: from structural motifs to macrocyclic knottins, Curr. Protein Pept. Sci. 5 (2004) 341–349. [14] D.J. Craik, N.L. Daly, C. Waine, The cystine knot motif in toxins and implications for drug design, Toxicon 39 (2001) 43–60. [15] N. Matsushita, M. Miyashita, A. Sakai, et al., Purification and characterization of a novel short-chain insecticidal toxin with two disulfide bridges from the venom of the scorpion L. australasiae, Toxicon 50 (2007) 861–867. [16] E.C.N. Silva, T.S. Camargos, A.Q. Maranhao, et al., Cloning and characterization of cDNA sequences encoding for new venom peptides of the Brazilian scorpion Opisthacanthus cayaporum, Toxicon 54 (2009) 252–261. [17] F. Delaglio, S. Grzesiek, G.W. Vuister, et al., NMRpipe – a multidimensional spectral processing system based on unix pipes, J. Biomol. NMR 6 (1995) 277– 293. [18] J. Jee, P. Guntert, Influence of the completeness of chemical shift assignments on NMR structures obtained with automated NOE assignment, J. Struct. Funct. Genomics 4 (2003) 179–189. [19] R. Koradi, M. Billeter, K. Wuthrich, MOLMOL: a program for display and analysis of macromolecular structures, J. Mol. Graphics 14 (1996) 51–55. [20] L. Holm, P. Rosenstrom, Dali server: conservation mapping in 3D, Nucleic Acids Res. 38 (2010) W545–W549. [21] A. Heitz, D. Le-Nguyen, B. Castro, et al., Conformational study of a native monodisulfide bridge analogue of EETI II, Lett. Pept. Sci. 4 (1997) 245–249. [22] A. Heitz, D. Le-Nguyen, L. Chiche, Min-21 and Min-23, the smallest peptides that fold like a cystine-stabilized beta-sheet motif: design, solution structure, and thermal stability, Biochemistry 38 (1999) 10615–10625. [23] D. Le-Nguyen, A. Heitz, L. Chiche, et al., Characterization and 2D NMR-study of the stable [9–21, 15–27] 2 disulfide intermediate in the folding of the three disulfide trypsin-inhibitor EETI-II, Protein Sci. 2 (1993) 165–174. [24] I. Karbat, F. Frolow, O. Froy, et al., Molecular basis of the high insecticidal potency of scorpion alpha-toxins, J. Biol. Chem. 279 (2004) 31679–31686.
744
S. Horita et al. / Biochemical and Biophysical Research Communications 411 (2011) 738–744
[25] F. Maggio, G.F. King, Scanning mutagenesis of a Janus-faced atracotoxin reveals a bipartite surface patch that is essential for neurotoxic function, J. Biol. Chem. 277 (2002) 22806–22813. [26] X.H. Wang, M. Connor, R. Smith, et al., Discovery and characterization of a family of insecticidal neurotoxins with a rare vicinal disulfide bridge, Nat. Struct. Biol. 7 (2000) 505–513. [27] P. Escoubas, C. Bernard, G. Lambeau, et al., Recombinant production and solution structure of PcTx1, the specific peptide inhibitor of ASIC1a protongated cation channels, Protein Sci. 12 (2003) 1332–1343. [28] S.J. Gunning, F. Maggio, M.J. Windley, et al., The Janus-faced atracotoxins are specific blockers of invertebrate KCa channels, FEBS J. 275 (2008) 4045–4059.
[29] C.W. Lee, E.H. Lee, K. Takeuchi, et al., Molecular basis of the high-affinity activation of type 1 ryanodine receptors by imperatoxin A, Biochem. J. 377 (2004) 385–394. [30] A. Mosbah, R. Kharrat, Z. Fajloun, et al., A new fold in the scorpion toxin family associated with an activity on a ryanodine-sensitive calcium channel, Proteins 40 (2000) 436–442. [31] J.J. Smith, J.M. Hill, M.J. Little, et al., Unique scorpion toxin with a putative ancestral fold provides insight into evolution of the inhibitor cystine knot motif, Proc. Natl. Acad. Sci. USA 108 (2011) 10478–10483.