In the present study, we show that venom of the ant spider Lachesana tarabaevi is unique in terms of molecular composition and toxicity. Whereas venom of most spiders studied is rich in disulfide-containing neurotoxic peptides, L. tarabaevi relies on the production of linear (no disulfide bridges) cytolytic polypeptides. We performed full-scale peptidomic examination of L. tarabaevi venom supported by cDNA library analysis. As a result, we identified several dozen components, and a majority (∼80% of total venom protein) exhibited membrane-active properties. In total, 33 membrane-interacting polypeptides (length of 18–79 amino acid residues) comprise five major groups: repetitive polypeptide elements (Rpe), latarcins (Ltc), met-lysines (MLys), cyto-insectotoxins (CIT) and latartoxins (LtTx). Rpe are short (18 residues) amphiphilic molecules that are encoded by the same genes as antimicrobial peptides Ltc 4a and 4b. Isolation of Rpe confirms the validity of the iPQM (inverted processing quadruplet motif) proposed to mark the cleavage sites in spider toxin precursors that are processed into several mature chains. MLys (51 residues) present ‘idealized’ amphiphilicity when modelled in a helical wheel projection with sharply demarcated sectors of hydrophobic, cationic and anionic residues. Four families of CIT (61–79 residues) are the primary weapon of the spider, accounting for its venom toxicity. Toxins from the CIT 1 and 2 families have a modular structure consisting of two shorter Ltc-like peptides. We demonstrate that in CIT 1a, these two parts act in synergy when they are covalently linked. This finding supports the assumption that CIT have evolved through the joining of two shorter membrane-active peptides into one larger molecule.

INTRODUCTION

Lachesana tarabaevi Zonstein and Ovtchinnikov, 1999 (Figure 1A) is an araneomorph spider from the family Zodariidae (ant spiders). It inhabits Central Asia where it hunts insects, and is known to produce very potent insecticidal venom (one of the most potent among 100 spider species from our collection). Previous work has shown that a major part of L. tarabaevi venom consists of peptides (molecular mass of ∼2–8 kDa; fraction II from size-exclusion chromatography, Figure 1B, inset), which is not uncommon among spiders. In general, peptide toxins from spiders form two groups: (i) neurotoxins that target different neuroreceptors and usually contain three or four disulfide bridges; and (ii) cytolytic peptides with membrane-active properties, most of which are linear, i.e. do not have disulfide bridges [1,2]. In most venoms studied to date, disulfide-rich neurotoxins play a central role, whereas cytolytic peptides appear to have an auxiliary function. Quite surprisingly, as we discuss in the present paper, L. tarabaevi ‘chose’ an opposite strategy: cytolytic membrane-active peptides are predominant in its venom.

Lachesana tarabaevi venom separation

Figure 1
Lachesana tarabaevi venom separation

(A) Male and female specimens of L. tarabaevi, photographs kindly provided by Sergei Zonstein. (B) RP-HPLC separation of L. tarabaevi venom peptide fraction obtained from size-exclusion chromatography. Coloured numbers indicate each individual component identified in the venom and correspond to the ‘#’ column in Table 1. The colour scheme relates components to different classes: blue for Rpe, green for Ltc, yellow for MLys, red for CIT and magenta for LtTx. Inset, size-exclusion chromatography of 1.3 mg of crude venom from both male and female individuals (Fauna Labs; [11]). The peptide fraction II is indicated by grey shading.

Figure 1
Lachesana tarabaevi venom separation

(A) Male and female specimens of L. tarabaevi, photographs kindly provided by Sergei Zonstein. (B) RP-HPLC separation of L. tarabaevi venom peptide fraction obtained from size-exclusion chromatography. Coloured numbers indicate each individual component identified in the venom and correspond to the ‘#’ column in Table 1. The colour scheme relates components to different classes: blue for Rpe, green for Ltc, yellow for MLys, red for CIT and magenta for LtTx. Inset, size-exclusion chromatography of 1.3 mg of crude venom from both male and female individuals (Fauna Labs; [11]). The peptide fraction II is indicated by grey shading.

Three groups of cytolytic polypeptides have been reported from L. tarabaevi venom so far, whereas typical knottin neurotoxins lacking membrane activity are still unknown. First, short antimicrobial peptides latarcins (Ltc) were extracted [3,4]. They show antimicrobial effects against both Gram-positive and Gram-negative bacteria, as well as antifungal and haemolytic activities [5]. When injected into flesh fly larvae, Ltc provoke paralysis and appearance of necrotic spots. All of the toxic effects are due to a cytolytic mechanism of action involving membrane rupture. As most other linear antimicrobial peptides, Ltc are positively charged at pH 7, and disordered in aqueous solution, but assume α-helical conformation in membrane-mimicking environments. Pronounced hydrophobic and positively charged surface clusters are typical of the Ltc α-helices [6,7].

Cyto-insectotoxins (CIT) discovered in L. tarabaevi venom are long linear polypeptides with marked cytolytic activity and toxicity towards insects [8]. CIT 1a, a major constituent of the venom, exhibits broad antimicrobial and, in contrast with Ltc, potent insecticidal effects. Cytolytic activity against eukaryotic cells was also reported. CIT 1a is a relatively long (69 residues) linear polypeptide forming amphipathic α-helices in membrane-mimicking environment. The most striking feature of CIT is their primary structure: they appear to be built of two Ltc-like modules, each of which represents a ‘common’ linear membrane-active peptide. The two structural modules are linked by a short sequence strongly resembling a mutated PQM (processing quadruplet motif). PQM (XXXR, where at least one X is glutamate, and R is arginine) is found in sequences of toxin precursor from spider venom glands [9]. This motif demarcates the N-terminus of mature toxin from the preceding propeptide. Cleavage by an as yet unknown processing enzyme occurs at the compulsory arginine residue of the motif, and this arginine is found to be mutated in CIT. Descent from a complex precursor by mutation of the processing site is a probable mechanism of CIT evolutionary origin.

Along with CIT, latartoxins (LtTx), insecticidal polypeptides found in L. tarabaevi venom, can be referred to as two-domain toxins [10]. Apart from an N-terminal disulfide-rich ‘neurotoxin’ domain, LtTx possess a C-terminal linear domain with antimicrobial, cytolytic and membrane-binding activities. The C-terminal domain possesses typical structural features of cytolytic peptides: it assumes an amphipathic α-helical conformation in membrane-mimicking environment. A membrane-dependent mechanism of LtTx action was suggested ascribing the role of an anchoring device to the C-terminal membrane-active domain.

In the present paper, we report detailed peptidomic characterization of L. tarabaevi venom. We show that, unlike other spider venoms studied, it is enriched in membrane-active toxins. In total, we report full amino acid sequences of 33 membrane-active polypeptides and corresponding precursor proteins deduced from cDNA library analysis. We report the discovery of two new groups of membrane-active polypeptides, i.e. met-lysines (MLys) and repetitive polypeptide elements (Rpe). We also found three new families of CIT, which exhibit limited or no sequence similarity to the CIT 1 family, but share structural and functional characteristics. CIT are concluded to represent the major toxin fraction of the venom. We went on to show that CIT activity is due to functional synergy between the two modules of these peculiar membrane-active toxins.

EXPERIMENTAL

Venom separation

Freeze-dried spider venoms (from both male and female individuals) were purchased from Fauna Labs. All stages of venom separation were performed according to our conventional protocol [11]. A 10 mg amount of L. tarabaevi crude venom was dissolved in 150 μl of solution containing 10% acetonitrile in 0.1% aqueous TFA (trifluoroacetic acid). The sample was subjected to fractionation using size-exclusion chromatography on a TSK 2000SW column (7.5 mm×600 mm, 12.5 nm pore size, 10 μm particle size; Toyo Soda Manufacturing). The eluent contained 10% acetonitrile in 0.1% aqueous TFA, and the flow rate was 0.5 ml/min. Effluent absorbance was monitored at 210 nm. Fractions were collected manually in 15-ml tubes and stored at 4°C.

The active polypeptide-containing fraction was subjected to further separation by RP (reversed-phase)-HPLC on a Vydac 218TP54 C18 column (4.6 mm×250 mm, 30 nm pore size, 5 μm particle size; Separations Group) using a 90-min linear gradient of acetonitrile concentration (0–60%) in 0.1% TFA at a flow rate of 1 ml/min. Absorbance was monitored at 210 and 280 nm. Fractions were collected manually in 1.7-ml tubes and stored at 4°C. Additional rounds of RP-HPLC were performed on the same Vydac column using a 90-min linear gradient of solvent containing 50% acetonitrile and 20% propan-2-ol in 0.1% TFA. Fractions containing individual compounds were collected manually in 1.7-ml tubes and stored in eluent at 4°C or freeze-dried for further experiments. Several rounds of crude venom separation were performed to obtain sufficient quantities of polypeptides required for structural and functional investigations.

Mass spectrometry

Molecular mass measurements were carried out using MALDI on an Ultraflex TOF-TOF (Bruker Daltonik) spectrometer as described previously [8]. Measurements were carried out using 10–100 pmol of polypeptides. 2,5-Dihydroxybenzoic acid (Sigma–Aldrich) was used as a matrix. Measurements were performed in the reflector mode, which enabled isotopic resolution and a mass-accuracy error not exceeding 30 p.p.m. for peptides with a maximum m/z of 3000. For larger molecules, measurements were performed in the linear mode, and an average mass was estimated with the accuracy error not exceeding 100 p.p.m. Mass spectra were analysed with the Data Analysis 4.3 and Data Analysis Viewer 4.3 software (Bruker).

Protein sequencing and selective proteolysis

The N-terminal amino acid sequences of purified polypeptides (100 pmol–2 nmol) were determined by automated Edman degradation using a Procise Model 492 (Applied Biosystems) or a PPSQ-33A (Shimadzu Biotech) protein sequencer according to protocols recommended by the manufacturers. Sequences of polypeptides with length over 40 amino acid residues were determined using selective proteolysis as specified below and Edman degradation of the fragments. Our conventional approach to polypeptide sequencing [11] was followed.

Cleavage of N-terminal pyroglutamate residues

Pyroglutamate residues were removed from the N-terminally blocked peptide Ltc 6a by pyroglutamate aminopeptidase (Sigma–Aldrich) following the published guidelines [11]. Ltc 6a (1 nmol) was dissolved in 50 μl of solution containing 50 mM Na2HPO4/NaH2PO4 (pH 7), 10 mM DTT and 1 mM EDTA. Pyroglutamate aminopeptidase (1 m-unit) dissolved in 10 μl of the same solution was added, and the mixture was incubated at 75°C for 2 h. The product was isolated by RP-HPLC on a Luna C18 column (3.0 mm×150 mm, 30 nm pore size, 5 μm particle size; Phenomenex) using a 60-min linear gradient of acetonitrile concentration (0–60%) in 0.1% TFA at a flow rate of 0.5 ml/min. Effluent absorbance was monitored at 210 nm.

Endoproteinase Glu-C digestion

Digestion of 1 nmol of pure polypeptide was performed with 0.2 μg of endoproteinase Glu-C (Sigma–Aldrich) in 20 μl of 50 mM ammonium bicarbonate buffer (pH 8.0). Probes were incubated for 4 h at 37°C. Peptide fragments were separated by RP-HPLC on a Luna C18 column as described above.

CNBr cleavage

Pure polypeptides (1–5 nmol) were dissolved in 20 μl of 80% aqueous TFA, and 1 μl of 5 M CNBr in acetonitrile (Sigma–Aldrich) was added. The samples were incubated for 24 h at room temperature in the dark. The reaction was terminated by diluting the mixture with 1 ml of cool distilled water, and the samples were dried using a vacuum evaporator. Reaction products were separated on a Luna C18 column as described above.

cDNA library construction, sequencing and analysis

All steps were carried out in collaboration with DuPont Agriculture and Nutrition as described previously [9]. Briefly, venom glands of L. tarabaevi were treated with liquid nitrogen, and total RNA (1.5 μg) was prepared by tissue homogenization using a mortar and pestle, followed by cell lysis in the presence of TRIzol® (Life Technologies). PolyA(+) RNA was purified on an oligo(dT)-cellulose affinity column using the mRNA Purification Kit (GE Healthcare) according to the manufacturer's protocol. The first-strand cDNA synthesis using Superscript II (Life Technologies) and subsequent second-strand synthesis, linker addition and directional cloning into the EcoRI and XhoI sites of pBlueScript SK (Stratagene) were performed in accordance with the Stratagene cDNA kit instructions. cDNA was purified using a cDNA column (Life Technologies). Sequencing of cDNA library clones was accomplished using the ABI PRISM BigDye Terminator Cycle Sequencing Ready Reaction Kit with AmpliTaq DNA polymerase, FS (PerkinElmer) and analysed on an ABI Prism 373 DNA Sequencer (Applied Biosystems).

The sequences obtained were translated in silico using the EditSeq program (DNASTAR) and compared with nucleotide and protein sequences in GenBank® and UniProt databases using a conventional BLAST algorithm. The Winstar suite (DNASTAR) was used for advanced sequence comparison and alignment. The SignalP 4.1 online server (http://www.cbs.dtu.dk/services/SignalP/) was used to predict signal peptide sequences.

Peptide synthesis

The fragments of CIT 1a (CIT 1a-N, CIT 1a-C and CIT 1a-M) were synthesized on a peptide synthesizer Syro I (MultiSynTech) using Fmoc chemistry and HBTU [2-(1H-benzotriazol-1-yl)-1,1,3,3-tetramethyluronium hexafluorophosphate]/HOBt (1-hydroxybenzotriazole) activation following a procedure described in [8]. Crude peptides were purified by semi-preparative RP-HPLC on a Luna C18 column (10 mm×250 mm, 10 nm pore size, 10 μm particle size; Phenomenex) using a 60-min linear gradient of acetonitrile concentration (0–60%) in 0.1% TFA at a flow rate of 2 ml/min. Final peptide purity was estimated to be >95% on the basis of analytical RP-HPLC analysis on a Vydac 218TP54 C18 column; effluent absorbance was monitored at 210 nm. Peptide sequence fidelity was confirmed by MALDI-MS.

Preparation of lipid vesicles

GUVs (giant unilamellar vesicles) (1000 nm diameter) and LUVs (large unilamellar vesicles) (100 nm diameter) were prepared as described in [10]. Vesicles were formed from DOPC (1,2-dioleoyl-sn-glycero-3-phosphocholine) and a mixture of DOPE (1,2-dioleoyl-sn-glycero-3-phosphoethanolamine) and DOPG [1,2-dioleoyl-sn-glycero-3-phospho-rac-(1-glycerol)] (7:3, w/w), all lipids were from Avanti Polar Lipids. DOPC (30 mg) or a mixture of DOPE/DOPG (21 mg/9 mg) was dissolved in 3 ml of chloroform/methanol (2:1, v/v) solution in a small round-bottomed flask. Lipid films were formed using a vacuum evaporator (Ika Works) and freeze-drying for 12 h. The samples were then suspended in aqueous solution containing 110 mM NaCl and 50 mM Na2HPO4/NaH2PO4 (pH 7.5). The resulting lipid concentration was 5 mM. Samples were subjected to quick ultrasonication and extruded through a Whatman 1000 nm-pore-size polycarbonate membrane (30 times at 25°C) using a Mini-Extruder (Avanti Polar Lipids) to obtain GUVs. For CD measurements of CIT1a-N and CIT 1a-C (see below), further extrusion through a Whatman 100 nm-pore-size polycarbonate membrane was performed to obtain LUVs.

CD spectra and secondary structure analysis

CD spectra were measured between 190 and 250 nm at 20°C on a Jasco J-810 spectropolarimeter. Polypeptides were dissolved in water, 50% (v/v) TFE (trifluoroethanol), 20 mM SDS and 5 mM 100 nm DOPC or DOPG/DOPE liposome (LUV) suspensions; polypeptide concentrations were 1 mg/ml. Data were analysed using the CONTINLL program as described previously [3]. Secondary-structure prediction was performed using the JPred 4 server (http://www.compbio.dundee.ac.uk/jpred4/index.html).

Spin-down assay

The spin-down assay was used to investigate the membranotropic activity of L. tarabaevi venom components. The crude venom of Agelena orientalis was chosen as a negative control. Venom from either L. tarabaevi or A. orientalis (1 mg) was dissolved in 90 μl of 110 mM NaCl and 50 mM Na2HPO4/NaH2PO4 (pH 7.5). Each sample was then divided into three parts (30 μl each). Then, 120 μl of the same phosphate buffer was added to one part, and 120 μl of DOPG/DOPE or DOPC GUV suspension (5 mM lipid concentration in the same phosphate buffer) were added to the other two parts. The resulting mixtures (150 μl each) were incubated for 1 h at 37°C with constant stirring and centrifuged (45 min, 16000 g). To determine toxin quantities remaining in the aqueous phase, all probes were assessed by RP-HPLC on a Jupiter C5 column (4.6 mm×250 mm, 30 nm pore size, 5 μm particle size; Phenomenex).

Insecticidal assays

The novel components were tested on flesh fly (Sarcophaga carnaria) larvae according to our standard protocol [8]. Samples were dissolved in physiological saline (140 mM NaCl, 5 mM KCl, 5 mM CaCl2, 1 mM MgCl2, 4 mM NaHCO3 and 5 mM Hepes, pH 7.2), and a volume of up to 2 μl was injected into the fourth segment of larva abdomen. Groups of five individuals were used for every dose assayed. Control animals received pure saline. Paralytic and lethal effects were observed for 24 h after injection.

Insecticidal effect of CIT 1a, CIT 1a-N and CIT 1a-C was tested on Drosophila melanogaster as described by Escoubas et al. [12]. Briefly, adult D. melanogaster flies were placed at 4°C and then transferred to a bed of ice. Glass micropipettes were prepared from 3.5-inch (1 inch=2.54 cm) replacement glass capillaries (0.044 inch outer diameter, 0.02 inch inner diameter; Drummond Scientific) on a P-1000 Micropipette Puller (Sutter Instrument). Injections were performed using a Nanoject II Auto-Nanoliter Injector (Drummond Scientific). Flies were injected with 50 nl of solution containing the polypeptides tested dissolved in physiological saline; control flies received pure saline. Groups of 20 individuals were used for a given concentration. Lethal effects were observed for 24 h. Median lethal doses (LD50) were calculated using the probit analysis. The relative standard errors of the obtained mean values did not exceed 5%.

Antimicrobial assay

All novel components, CIT 1a-N, CIT 1a-C and CIT 1a-M, were tested against Gram-positive and Gram-negative bacteria (Bacillus subtilis VKM B-501 and Escherichia coli DH5α) following the previously described modification of the microtitre broth dilution method [11]. Briefly, Gram-positive (B. subtilis VKM B-501) or Gram-negative (E. coli DH5α) bacteria were cultured in low-salt LB medium. The 2-fold microtitre broth dilution assay was carried out in 96-well sterile plates in a final volume of 100 μl. Mid-exponential-phase cultures were diluted to a final concentration of 105 colony-forming units/ml. Pure polypeptides were dissolved in 10 μl of water and added to 90 μl of the bacterial dilution. The samples, a non-treated control and a sterility control were tested in triplicate. The microtitre plates were incubated for 24 h at 37°C, and growth inhibition was determined by measuring absorbance at 620 nm. MICs (minimum inhibitory concentrations) are expressed as the lowest concentration of polypeptide that caused 100% bacterial growth inhibition.

RESULTS

Membrane activity of L. tarabaevi venom

Crude venom of the ant spider L. tarabaevi is known to possess marked antimicrobial, cytolytic and insecticidal activities, all of which have been related to the presence of membrane-active polypeptides [35,8,10]. We therefore decided to perform comprehensive profiling of membrane-active components in this venom. Incubation of crude venom with DOPC or DOPE/DOPG GUV suspension followed by vesicle spin-down led to a depletion of membrane-active components from solution. RP-HPLC analysis of the supernatant revealed that most constituents of L. tarabaevi venom, accounting for as much as ∼80% of crude dry venom, bound to anionic (DOPE/DOPG) GUVs (Figure 2, left column). The corresponding peaks disappeared from the RP-HPLC profile after incubation with lipid vesicles and spin-down. Moreover, components eluting from a C5 column at acetonitrile concentration of above ∼30% and accounting for ∼30% of crude venom also interacted with zwitterionic (DOPC) GUVs. Thus most L. tarabaevi venom components are membrane-active. Of note, the same procedure with crude venom of the funnel-web spider A. orientalis (family Agelenidae) did not result in any HPLC profile change for either GUV suspension (Figure 2, right column).

Spin-down assay

Figure 2
Spin-down assay

Venoms of L. tarabaevi (left panel) and A. orientalis (right panel) were incubated with phosphate buffer (upper row) and zwitterionic GUVs (middle row) or anionic GUVs (lower row) suspensions. Shown are RP-HPLC separation profiles of aqueous solutions remaining after centrifugation of the samples. In the case of GUV suspensions, membrane-active polypeptides were depleted from the solutions. β/δ-Aga, the elution interval for β/δ-agatoxins, major constituents of A. orientalis venom [38] (all profiles were obtained in the present study).

Figure 2
Spin-down assay

Venoms of L. tarabaevi (left panel) and A. orientalis (right panel) were incubated with phosphate buffer (upper row) and zwitterionic GUVs (middle row) or anionic GUVs (lower row) suspensions. Shown are RP-HPLC separation profiles of aqueous solutions remaining after centrifugation of the samples. In the case of GUV suspensions, membrane-active polypeptides were depleted from the solutions. β/δ-Aga, the elution interval for β/δ-agatoxins, major constituents of A. orientalis venom [38] (all profiles were obtained in the present study).

Venom separation and purification of membrane-active components

Crude venom was first fractionated using size-exclusion chromatography (Figure 1B, inset), and the active polypeptide-containing fraction was subjected to further separation by RP-HPLC (Figure 1B). MS analysis and comparison of chromatographic profiles in Figure 2 permitted us to detect all active components described previously: Ltc, CIT and LtTx (Figure 1B and Table 1). Moreover, 13 novel membrane-active polypeptides were identified in L. tarabaevi venom. These components were put through additional steps of RP-HPLC for purification and, as a result, all molecules were obtained in an individual state, and their masses were measured by MALDI-MS (Supplementary Table S1). We found four ‘small’ peptides (Rpe, ∼2 kDa) eluting relatively early at ∼30% acetonitrile, three ‘small’ peptides (Ltc, ∼3–4 kDa) and four ‘large’ polypeptides (CIT, ∼7–9 kDa) eluting at ∼30–40% acetonitrile, and two ‘medium-sized’ polypeptides (MLys, ∼6 kDa) eluting late at ∼40–50% acetonitrile. Sequence analysis resulted in identification of as many as seven new families of membrane-active polypeptides.

Table 1
Sequences of membrane-active polypeptides from L. tarabaevi venom

Names of polypeptides sequenced here for the first time are in bold. Residues differing in polypeptides from the same family are shaded. A break is introduced into LtTx-2c sequence for the sake of comparison. Z, pyroglutamic acid residue; -NH2, C-terminal amidation. Measured average molecular masses are presented in the rightmost column.

Names of polypeptides sequenced here for the first time are in bold. Residues differing in polypeptides from the same family are shaded. A break is introduced into LtTx-2c sequence for the sake of comparison. Z, pyroglutamic acid residue; -NH2, C-terminal amidation. Measured average molecular masses are presented in the rightmost column.
Names of polypeptides sequenced here for the first time are in bold. Residues differing in polypeptides from the same family are shaded. A break is introduced into LtTx-2c sequence for the sake of comparison. Z, pyroglutamic acid residue; -NH2, C-terminal amidation. Measured average molecular masses are presented in the rightmost column.

Polypeptide sequencing

Analysis of 13 novel polypeptides (Rpe 1a-d; Ltc 2b, 6a and 7; MLys 1a and 1b; and CIT 2a, 2c, 3 and 4) was conducted by direct N-terminal Edman degradation. The small peptides comprising fewer than 40 amino acid residues (Rpe and Ltc) were sequenced completely, but only partial amino acid sequences were determined for medium-sized (MLys) and large (CIT) polypeptides (Figure 3). One small peptide (Ltc 6a) appeared to be N-terminally blocked by a pyroglutamate residue, so pyroglutamate aminopeptidase was used to cleave this residue and enable Edman sequencing.

Determination of novel polypeptide primary structures

Figure 3
Determination of novel polypeptide primary structures

Schemes are shown for one representative of Rpe, Ltc, MLys and CIT. See Supplementary Figure S1 for all other polypeptides. Amino acid residues targeted during specific polypeptide chain fragmentation are shown in bold. Unbroken arrows above the sequences indicate sequencing by N-terminal Edman degradation. Fragments released through cleavage at specific bonds are indicated with broken lines below.

Figure 3
Determination of novel polypeptide primary structures

Schemes are shown for one representative of Rpe, Ltc, MLys and CIT. See Supplementary Figure S1 for all other polypeptides. Amino acid residues targeted during specific polypeptide chain fragmentation are shown in bold. Unbroken arrows above the sequences indicate sequencing by N-terminal Edman degradation. Fragments released through cleavage at specific bonds are indicated with broken lines below.

Comparison of N-terminal sequences with data obtained from cDNA library (see below) allowed us to establish the full-length amino acid sequences for MLys and CIT. Sequences deduced from transcriptomic data were verified by a combination of selective proteolysis at glutamate (CIT 3) or methionine (MLys 1a and 1b; CIT 2a, 2c and 4) residues with further RP-HPLC separation and MS analysis (Supplementary Figure S1). CIT 3 was cleaved by Glu-C into the following fragments: Lys1–Glu37 (measured average molecular mass 4076.4 Da) that overlaps with the initial N-terminal Edman sequence, Met38–Glu48 (monoisotopic mass 1254.6), Met49–Glu64 (1820.0 Da), and Ala65–Ser79 (1696.0 Da). For MLys 1a and 1b, we identified the following CNBr fragments (measured monoisotopic masses are listed, note that methionine residues are converted to homoserine lactone residues): Ala8–Met18 (1239.8 Da) and Gly26–Met32 (785.4 Da) that overlap with the Edman sequence, and Leu33–Met47 (1722.0 Da) in MLys 1a and Glu38–Met44 (841.5 Da) in MLys 1b. For CIT 2a, the following fragments were produced: Ser1–Met13 (1531.8 Da), Arg18–Met25 (997.6 Da), Ala26–Met43 (1898.0 Da) and Ser44–Glu69 (3040.7 Da). In CIT 2c, we observed the same fragments as in CIT 2a, except for the C-terminal Ser44–Glu69; two fragments were detected instead: Ser44–Met50 (828.5 Da), and His51–Glu70 (2329.2 Da). Finally, CIT 4 gave the following CNBr fragments: Ser7–Met26 (2315.4 Da), which overlapped completely with the Edman sequence, and Lys32–Phe61 (measured average molecular mass 3427.2 Da). Figure 3 shows the sequencing scheme for one representative of Rpe, Ltc, MLys and CIT; and Supplementary Figure S1 contains schemes for all other polypeptides. The resulting full sequences are grouped in Table 1, and measured and calculated average molecular masses for each polypeptide are shown in Supplementary Table S1.

Seven molecules (Rpe 1a-d, MLys 1a and 1b and CIT 4) were found to be post-translationally modified and carry C-terminal amidation, which was inferred from two observations: (i) a 1 Da discrepancy between the measured and calculated molecular masses, and (ii) the presence of an additional C-terminal glycine residue in the structure of precursor proteins (see below). In the case of Rpe, the accuracy of mass measurements was sufficient to claim the presence of the modification (experimental error did not exceed 0.1 Da). CIT 4 was cleaved with CNBr, and the mass of its C-terminal fragment (Lys32–Phe61) differed from the calculated value by 1 Da (experimental error did not exceed 0.4 Da). In the case of MLys, the experimental error was 0.6 Da, but presence of C-terminal amidation was substantiated by the precursor structures inferred from cDNA sequences.

Polypeptide sequence analysis

All of the new polypeptides exhibit certain common properties such as lack of disulfide bonds, positive charge at pH 7, elevated helix-forming propensity and amphiphilicity in α-helical conformation with clusters of hydrophobic or hydrophilic residues. Helix-forming propensity of synthetic Ltc 6a and 7 was probed previously by CD [3] and is assumed for the other new polypeptides on the basis of sequence similarity with related Ltc and CIT, and secondary-structure prediction using the JPred 4 server. These properties explain well the observed membrane activity of the components, and are also common to a wide group of so-called α-helical antimicrobial peptides found throughout Nature [1315], and in arthropod venom in particular [1,16].

The four smallest (∼2 kDa) relatively early-eluting peptides were named repetitive polypeptide elements (Rpe 1a–1d; see discussion below). All Rpe contain 18 residues, are moderately basic (carry a charge ranging from +3 to +5 at pH 7), and are C-terminally amidated. Rpe are very closely related peptides and share 78–94% identical residues.

Three new small peptides (∼3–4 kDa) eluting at ∼30–40% acetonitrile are Ltc 2b, 6a and 7. These peptides have been predicted previously from L. tarabaevi venom gland cDNA [3], but have not been purified from venom. Ltc 2b contains 26 residues and differs by just one C-terminal residue from Ltc 2a (asparagine rather than histidine), which is probably the best-studied Ltc with high non-specific cytolytic activity [3,5,6,1722]. We assume that Ltc 2b has very similar properties to those of Ltc 2a. Ltc 6a is N-terminally blocked by pyroglutamate, and Edman sequencing was performed after selective removal of that residue. As reported previously, Ltc 6a (33 residues) and 7 (34 residues) are ‘atypical’ Ltc, and, although, just like many antimicrobial peptides from spider venom, they form amphipathic α-helices, present overall positive charge at pH 7 (+5 for Ltc 6a and +2 for Ltc 7) and disrupt artificial membranes, no antimicrobial or cytolytic activity was observed [3].

The four large polypeptides (∼7–9 kDa) were allocated to CIT based on their structural features and activity (CIT 2a, 2c, 3 and 4; see discussion below). They form three new families of CIT (CIT 1 has been described previously [8]), enlarging substantially the known repertoire of these peculiar molecules. CIT 2 is related to CIT 1 (55% identical residues in CIT 1a and CIT 2a). Similarly to CIT 1, CIT 2 can be split into two modules connected by a short linker sequence E34EVQ37 that resembles a mutated PQM. CIT are comparatively long polypeptides (from 61 residues in CIT 4 to 79 in CIT 3), approximately twice the length of an ‘average’ antimicrobial peptide [the average length of 2692 peptides in the APD database (http://aps.unmc.edu/AP/statistic/statistic.php) is 32.8 residues]. They are also twice or even more as charged (from +7 in CIT 2c to +21 in CIT 3a at neutral pH) and the average net charge of all peptides in the APD database is 3.2. CIT may be called ‘giant’ antimicrobial, cytolytic or membrane-active polypeptides; they are among the longest known disulfide-free α-helical antimicrobial polypeptides, the closest contender from spider venom being oxyopinin 1 from Oxyopes takobius with 49 residues [23].

Finally, the two late-eluting medium-sized (∼6 kDa) polypeptides were termed met-lysines (MLys 1a and 1b) due to the high content of methionine (14–18%) and lysine (28%) residues. Both molecules contain 51 amino acid residues, are C-terminally amidated, and differ by just two residues (Ile37 and Thr44 in MLys 1a are replaced by methionine in MLys 1b). MLys exhibit an elevated content of negatively charged (especially glutamate) residues (20%) and therefore carry a moderate positive charge (+5 at pH 7). MLys present the highest helix-forming propensity, and corresponding predicted α-helices are almost ideally amphiphilic (Supplementary Figure S2). Not only are hydrophobic and hydrophilic residues segregated, but so are positively and negatively charged residues: cationic residues flank the hydrophobic cluster, whereas anionic residues locate to exactly the opposite side of the helical cylinder. This assumed idealized amphiphilicity explains well the very late elution of MLys in RP-HPLC (Figure 1B).

No significant similarity is found between Rpe, Ltc, CIT and MLys and other known protein sequences. Some additional peculiarities of these membrane-active polypeptides include marked preference for lysine over arginine, preference for glycine or other small residues at the N-terminus, and introduction of helix-breaking glycine or proline residues into the middle of the sequences.

Venom gland cDNA database analysis

The EST database from L. tarabaevi venom glands was translated in silico and searched for the chemically established polypeptide sequences. We were able to deduce full amino acid sequences for the medium-sized MLys and long CIT, and precursor protein structures for all membrane-active polypeptides. The deduced precursor proteins show a typical ‘prepropeptide’ organization. The N-terminal signal peptide sequences were predicted using the SignalP 4.1 online server. Propeptides were defined as fragments flanked by signal peptide at the N-terminus and PQM at the C-terminus. As reported previously, among L. tarabaevi toxin precursors there are not only ‘simple’ precursors containing single mature chain but also ‘binary’ and ‘complex’ precursors containing two or more mature chains separated by two or more PQM [3] (Figure 4). Ltc, CIT 1 and LtTx precursor sequences have been published previously [3,8,10]. In the present paper, we report precursor structures for the novel membrane-active polypeptides (Supplementary Figure S3).

Graphical representation of latarcin precursors

Figure 4
Graphical representation of latarcin precursors

Three functional parts are recognized: signal peptide, propeptide and mature chain. In the case of binary and complex precursors, several mature chains are shown. PQM and iPQM are indicated.

Figure 4
Graphical representation of latarcin precursors

Three functional parts are recognized: signal peptide, propeptide and mature chain. In the case of binary and complex precursors, several mature chains are shown. PQM and iPQM are indicated.

All CIT and MLys precursors (named pCIT and pMLys), as well as pLtc 2b and pLtc 7, are conventional simple precursors. Ltc 6a is processed from two binary precursors pLtc 6a-1 and 6a-2 and corresponds to their C-terminal parts. We were not able to detect other mature peptides (Ltc 6b or 6c) predicted from these precursors, in L. tarabaevi venom. Rpe were found encoded by the same genes that code for Ltc 4a and 4b. The corresponding precursor proteins pLtc 4a-1 and 4a-2 (Q1ELU5) and pLtc 4b (Q1ELU4) are complex precursors and are processed into mature peptides of two kinds: Ltc, which are the C-terminal parts of the precursors, and Rpe, which come in tandem repeats (hence their name) between the N-terminal preprosequences and Ltc. We find Rpe 1a and three copies of Rpe 1b in pLtc 4a-1, Rpe 1a and four copies of Rpe 1b in pLtc 4a-2, and Rpe 1d and three copies of Rpe 1c in pLtc 4b. Importantly, just as the N-termini of mature peptides are specified by PQM, their C-termini are specified by a different kind of motif called iPQM (inverted processing quadruplet motif): RXXX, where at least one X is glutamate, and R is arginine [24]. As in PQM, cleavage by an as yet unknown processing enzyme occurs at the compulsory arginine residue of the motif.

As already mentioned above, all C-terminally amidated mature polypeptides contain an additional glycine residue in an unprocessed form. The immature form of Ltc 6a has an N-terminal glutamine residue, which is cyclized to a pyroglutamate.

Biological activity of Rpe, MLys and CIT

For insecticidal activity, Rpe, CIT and MLys were tested on flesh fly larvae. CIT provoked lethal effects with LD50 in the range 40–75 μg/g (Table 2). Consistent with their cytolytic mode of action, CIT caused appearance of necrotic spots and liquid flow out of the injection opening. Neither Rpe nor MLys showed any activity up to the dose of 600 μg/g. In the antibacterial assay, only CIT showed high activity against both Gram-positive (B. subtilis) and Gram-negative (E. coli) bacteria with MIC values between 0.63 and 10 μM (Table 2). For MLys and Rpe, no antimicrobial effect was observed up to a concentration of 20 μM.

Table 2
Insecticidal and antimicrobial activities of purified polypeptides

Presented are values of median lethal doses (LD50) of polypeptides injected into flesh fly larvae, and MICs of polypeptides against selected bacterial strains. N/T, not tested

 Rpe 1a Rpe 1b Rpe 1c MLys 1a MLys 1b CIT 2a CIT 2c CIT 3 CIT 4 
Insecticidal activity (LD50, μg/g) 
S. carnaria N/T >600 N/T >600 >600 40 40 66 75 
Antimicrobial activity (MIC, μM) 
B. subtilis B-501 >20 >20 >20 >20 >20 1.25 1.25 0.63 2.5 
E. coli DH5α >20 >20 >20 >20 >20 2.50 2.50 2.50 10 
 Rpe 1a Rpe 1b Rpe 1c MLys 1a MLys 1b CIT 2a CIT 2c CIT 3 CIT 4 
Insecticidal activity (LD50, μg/g) 
S. carnaria N/T >600 N/T >600 >600 40 40 66 75 
Antimicrobial activity (MIC, μM) 
B. subtilis B-501 >20 >20 >20 >20 >20 1.25 1.25 0.63 2.5 
E. coli DH5α >20 >20 >20 >20 >20 2.50 2.50 2.50 10 

Structural and functional analysis of CIT 1a modules

Using solid-phase peptide synthesis, we synthesized three derivatives of CIT 1a that correspond to the N-terminal module (CIT 1a-N, residues 1–34), C-terminal module (CIT 1a-C, residues 39–69) and middle part (CIT 1a-M, residues 21–52) of CIT 1a. CIT 1a-M included the E35EAQ38 linker sequence that was supposed to result from a point mutation in a PQM processing site (Figure 5).

Sequences of CIT 1a and its derivatives obtained by chemical synthesis

Figure 5
Sequences of CIT 1a and its derivatives obtained by chemical synthesis

CIT 1a-N, CIT 1a-C and CIT 1a-M peptides are indicated by rectangles.

Figure 5
Sequences of CIT 1a and its derivatives obtained by chemical synthesis

CIT 1a-N, CIT 1a-C and CIT 1a-M peptides are indicated by rectangles.

Bioinformatics analysis indicated that CIT1a-N and CIT 1a-C are prone to form α-helices with hydrophobic clusters on their surfaces (Figure 6B). According to CD spectra (Figure 6A), both CIT 1a-N and CIT 1a-C are disordered in aqueous solution, whereas the α-helical conformation is dominating in membrane-mimicking environments such as DOPC and DOPG/DOPE LUV suspensions, or 50% TFE (Supplementary Table S2). Both derivatives are positively charged at neutral pH: CIT 1a-N and CIT 1a-C carry a charge of +9 and +7 respectively. Thus separated modules behave similarly to ‘usual’ short α-helical antimicrobial peptides.

CIT 1a-N and CIT 1a-C secondary structure

Figure 6
CIT 1a-N and CIT 1a-C secondary structure

(A) CD spectra of the peptides in different environments: 1, 50% TFE; 2, DOPG/DOPE liposome suspension; 3, DOPC liposome suspension; 4, water. (B) Helical wheel projections for CIT 1a-N and CIT 1a-C. Amino acid sequences of the peptides are shown above the wheels. Hydrophobic residues are black, positively charged residues are shaded grey and all other residues are white.

Figure 6
CIT 1a-N and CIT 1a-C secondary structure

(A) CD spectra of the peptides in different environments: 1, 50% TFE; 2, DOPG/DOPE liposome suspension; 3, DOPC liposome suspension; 4, water. (B) Helical wheel projections for CIT 1a-N and CIT 1a-C. Amino acid sequences of the peptides are shown above the wheels. Hydrophobic residues are black, positively charged residues are shaded grey and all other residues are white.

Antimicrobial activity of the three derivatives was tested on Gram-positive and Gram-negative bacteria in comparison with the full-length CIT 1a (Table 3). MIC values of CIT 1a-N are significantly lower than MIC values of CIT 1a-C, whereas MIC values of full-length CIT 1a are lower than MIC of both derivatives and their mixture. In insecticidal tests on D. melanogaster, both CIT 1a-N and CIT 1a-C showed rather low activity with LD50 values of ∼100–150 μg/g compared with ∼10 μg/g for full-length CIT 1a (Table 3). CIT 1a-M was inactive in both tests.

Table 3
Insecticidal and antimicrobial activities of CIT 1a derivatives

Presented are values of median lethal doses (LD50) of polypeptides injected into Drosophila flies, and MICs of polypeptides against selected bacterial strains.

 CIT 1a CIT 1a-N CIT 1a-C CIT 1a-N+CIT 1a-C CIT 1a-M 
Insecticidal activity (LD50, μg/g) 
D. melanogaster 12.3 96.3 144.6 224.0 >250 
Antimicrobial activity (MIC, μM) 
B. subtilis B-501 0.9 1.3–2.5 20–40 1.3–2.5 >100 
E. coli DH5α 0.6 5–10 40–80 2.5–5 >100 
 CIT 1a CIT 1a-N CIT 1a-C CIT 1a-N+CIT 1a-C CIT 1a-M 
Insecticidal activity (LD50, μg/g) 
D. melanogaster 12.3 96.3 144.6 224.0 >250 
Antimicrobial activity (MIC, μM) 
B. subtilis B-501 0.9 1.3–2.5 20–40 1.3–2.5 >100 
E. coli DH5α 0.6 5–10 40–80 2.5–5 >100 

DISCUSSION

Lachesana venom is a library of membrane-active toxins

Spider venoms are complex mixtures of dozens of diverse compounds such as ions, biogenic amines, polyamines, disulfide-rich neurotoxins, linear peptides and enzymes [1,2]. Similarly to many other venomous animals (snakes, sea anemones, cone snails and scorpions), spiders are famous for producing so-called combinatorial libraries of toxins, i.e. groups of structurally related toxins sharing a common scaffold but differing in radicals. Most spiders produce neurotoxins acting on neuroreceptors. In this case, the dominant molecular scaffold is the ICK (inhibitor cystine knot) [25]. Less often, spider venoms demonstrate cytolytic effects caused by linear membrane-active peptides [23,26] or enzymes such as sphingomyelinase [27].

Our experiments have shown that most L. tarabaevi venom components present membrane activity (Figure 2). As we went on to demonstrate, L. tarabaevi venom is composed of a wide variety of membrane-active polypeptides. Its major components underlying toxicity act through cytolysis. Cytolytic components were also found in the venom of some other spiders. Several species studied belong to the families Lycosidae (in particular, the wolf spiders Lycosa carolinensis [28] and L. singoriensis [29]), Oxyopidae (the lynx spider O. takobius [23,30]) and Ctenidae (the wandering spider Cupiennius salei [26]). However, for each spider in this case, a more abundant and/or functionally important fraction of the venom is represented by cysteine-containing neurotoxins with the ICK structural motif. ICK-containing toxins have been noted in L. tarabaevi venom (LtTx), but they are not the dominant part of the venom. An interesting feature of these molecules is that their ICK core is equipped with a C-terminal membrane-active region. Moreover, this C-terminal region corresponds to a ‘common’ α-helical antimicrobial peptide [10].

It seems that L. tarabaevi largely relies on the membrane-active amphipathic α-helical scaffold; and yet again an impressive variety of toxins with this common structural feature is noted: at least 33 individual molecules are grouped into 15 families (Table 1). These characteristics make L. tarabaevi venom unique among spider venoms and even among venoms with cytolytic properties.

Unique venom components

Five groups of membrane-active components were identified in L. tarabaevi venom (Ltc, CIT, LtTx, Rpe and MLys; Table 1) and three of them (CIT, Rpe and MLys) appear to be ‘the first in class’, i.e. these three groups of components are unique in terms of structure. Presence of short linear peptides such as Ltc is common among spiders [23,26,28] and other arthropods [3133]; and molecules with the ‘ICK+membrane-active C-terminus’ architecture similar to LtTx have been noted in several spiders including C. salei [34] and Cheiracanthium punctorium [35].

The function of two unique families of membrane-active polypeptides (short Rpe and medium-sized MLys) is still unknown. Rpe show an ability to bind to negatively charged DOPE/DOPG GUVs mimicking bacterial membranes; however, they fail to demonstrate antibacterial effects. Similarly, MLys are able to bind to both GUV types, but do not show any effect in either test. One possibility is that we did not find a proper target for these molecules. From the idealized structure of MLys (Supplementary Figure S2), one may suppose that they bind most effectively to zwitterionic membranes. Another possibility is that Rpe and MLys may act synergistically with other venom components to enhance their activities.

Eight CIT belonging to the CIT 1 family have been reported previously [8], and in the present paper we describe three novel CIT families (CIT 2–4) showing low, if any, sequence similarity to each other. However, they possess common structural features being long linear positively charged amphipathic polypeptides. Long linear CIT account for ∼30% of the active polypeptide fraction of L. tarabaevi venom (Figure 1B) and confirm their role as major components underlying the overall insect toxicity and antimicrobial properties of the venom. Even though both activities have been noted for LtTx as well [10], their effectiveness is lower than that of CIT.

CIT are two-domain self-enhancing cytolytic toxins

We demonstrated synergism-based activity of CIT using the major component CIT 1a (∼5% of total polypeptide content) as an example. CIT 1 and 2 are assumed to have evolved through mutation of a processing motif that once separated two short α-helical membrane-active peptides in a binary precursor. As a result, those two peptides became modules or domains of giant linear CIT. This mutation is beneficial, since CIT exhibit unusually high insecticidal activity (LD50 ∼10 μg/g) comparable with that of neurotoxins [25,36] and high wide-spectrum cytolytic activity [8].

We show in the present study that, indeed, the two modules of CIT 1a (Figure 5) are very much like short α-helical membrane-active peptides: they are cationic linear peptides with membrane activity and high helix-forming propensity (Figure 6A and Supplementary Table S2). Both modules acting synergistically are required for the effects of CIT 1a. Separated domains show very weak insecticidal activity (Table 3), comparable with that of short α-helical membrane-active peptides from spider venom such as Ltc (LD50 ∼100–150 μg/g) [8,26]. Antibacterial activity of the separated domains is significantly lower than that of the full-length toxin (Table 3). CIT 1a-N is noticeably more active than CIT 1a-C, which is perhaps due to its greater amphiphilicity. Activity of an equimolar mixture is on the same level as the N-terminal domain activity. CIT 1a-M, a central fragment of CIT 1a, is devoid of activity. This testifies in favour of CIT 1a-N and CIT 1a-C being regarded as functional modules and points to a lack of specific function for the linker sequence separating the domains.

Thus the mode of action of CIT 1a is based on co-operation between the two covalently linked modules. Since CIT 1a-N shows higher activity, we hypothesize that the N-terminal domain is responsible for cytolytic activity of CIT 1a, whereas the C-terminal domain plays a role of an enhancer. Similar synergistic mode of action was shown for CsTx-1 neurotoxin from C. salei with the ‘ICK+membrane-active C-terminus’ architecture. In this case, the C-terminal linear domain acts as an enhancer for the N-terminal ICK domain [34].

Unique biosynthetic apparatus

Comparison of protein sequences obtained from transcriptomic analysis of L. tarabaevi venom glands with the mature polypeptides purified from the venom permitted us to establish full toxin precursor structures. We define three kinds of precursors (Figure 4): (i) simple ‘prepropeptide’ precursors containing signal peptide, propeptide and mature chain; (ii) binary precursors that include all parts from simple precursors and an additional mature chain; and (iii) complex precursors containing four or even five mature chains. Most L. tarabaevi toxins are processed from simple precursors, characteristic of a vast majority of known spider toxins [1]. Ltc 6a is processed from binary precursors and Ltc 4a/4b and Rpe are produced from complex precursors (Supplementary Figure S3).

C-termini of Rpe are due to precursor cleavage at iPQM and isolation of Rpe from venom validates this processing motif. iPQM is also involved in biosynthesis of two-chain ω-agatoxins IA and IB from Agelenopsis aperta: both chains are encoded by one gene and synthesized from one precursor, which is cut into two pieces precisely at PQM and iPQM [37]. We believe that further instances of iPQM will be found as more data accumulate. We also note that it is the invariable arginine that forms the peptide bond cleaved during processing. Perhaps this is the reason for the observed preference for lysine over arginine in mature toxins: restriction of arginine to processing motifs augments the specificity of limited proteolysis.

To our knowledge, L. tarabaevi is currently the only spider known to produce such a variety of toxin precursors. We are convinced, however, that both binary and complex types of precursor are more widespread. Occurrence of iPQM in ω-agatoxin precursors points to the conservation of the corresponding processing enzymes, hence other examples of several toxins encoded by single gene are expected.

AUTHOR CONTRIBUTION

Alexey Kuzmenkov carried out or took part in most experiments, analysed the data and wrote the paper. Maria Sachkova performed experiments involving CIT 1a and wrote the paper. Sergey Kovalchuk conducted peptide synthesis. Eugene Grishin supervised the work. Alexander Vassilevski designed the work, performed venom separation and protein sequencing, analysed the data and wrote the paper.

We thank the DuPont Agriculture and Nutrition staff (especially Maureen Dolan, Will Krespan, Bill McCutchen and Rafi Herrmann) for cDNA library construction and sequencing. We are grateful to Lucia Kuhn-Nentwig (Institute of Ecology and Evolution, University of Bern, Bern, Switzerland) for her kind help in performing insecticidal experiments on D. melanogaster and to Kseniya Kudryashova (Shemyakin-Ovchinnikov Institute) for her help with CD. We also thank Sergei Zonstein (currently at the Zoology Department, Tel Aviv University, Tel Aviv, Israel) for sharing photographs of L. tarabaevi and Daniil Osipov (Moscow Zoo, Moscow, Russia) for his professional account of spider biology.

FUNDING

This work was supported by the Russian Science Foundation [grant number 14-14-01180]. A.I.K. was also supported by the Russian Foundation for Basic Research [grant number 14-04-32091] to perform spin-down and biological tests.

Abbreviations

     
  • CIT

    cyto-insectotoxin(s)

  •  
  • DOPC

    1,2-dioleoyl-sn-glycero-3-phosphocholine

  •  
  • DOPE

    1,2-dioleoyl-sn-glycero-3-phosphoethanolamine

  •  
  • DOPG

    1,2-dioleoyl-sn-glycero-3-phospho-rac-(1-glycerol)

  •  
  • GUV

    giant unilamellar vesicle

  •  
  • ICK

    inhibitor cystine knot

  •  
  • iPQM

    inverted processing quadruplet motif

  •  
  • Ltc

    latarcin(s)

  •  
  • LtTx

    latartoxin(s)

  •  
  • LUV

    large unilamellar vesicle

  •  
  • MIC

    minimum inhibitory concentration

  •  
  • MLys

    met-lysine(s)

  •  
  • PQM

    processing quadruplet motif

  •  
  • RP

    reversed-phase

  •  
  • Rpe

    repetitive polypeptide element(s)

  •  
  • TFA

    trifluoroacetic acid

  •  
  • TFE

    trifluoroethanol

References

References
1
Vassilevski
A.A.
Kozlov
S.A.
Grishin
E.V.
Molecular diversity of spider venom
Biochemistry (Mosc.)
2009
, vol. 
74
 (pg. 
1505
-
1534
)
[PubMed]
2
Kuhn-Nentwig
L.
Stöcklin
R.
Nentwig
W.
Venom composition and strategies in spiders: is everything possible?
Adv. Insect Physiol.
2011
, vol. 
40
 (pg. 
1
-
86
)
3
Kozlov
S.A.
Vassilevski
A.A.
Feofanov
A.V.
Surovoy
A.Y.
Karpunin
D.V.
Grishin
E.V.
Latarcins, antimicrobial and cytolytic peptides from the venom of the spider Lachesana tarabaevi (Zodariidae) that exemplify biomolecular diversity
J. Biol. Chem.
2006
, vol. 
281
 (pg. 
20983
-
20992
)
[PubMed]
4
Dubovskii
P.V.
Vassilevski
A.A.
Kozlov
S.A.
Feofanov
A.V.
Grishin
E.V.
Efremov
R.G.
Latarcins: versatile spider venom peptides
Cell. Mol. Life Sci.
2015
, vol. 
72
 (pg. 
4501
-
4522
)
[PubMed]
5
Vorontsova
O.V.
Egorova
N.S.
Arseniev
A.S.
Feofanov
A.V.
Haemolytic and cytotoxic action of latarcin Ltc2a
Biochimie
2010
, vol. 
93
 (pg. 
227
-
241
)
[PubMed]
6
Dubovskii
P.V.
Volynsky
P.E.
Polyansky
A.A.
Chupin
V.V.
Efremov
R.G.
Arseniev
A.S.
Spatial structure and activity mechanism of a novel spider antimicrobial peptide
Biochemistry
2006
, vol. 
45
 (pg. 
10759
-
10767
)
[PubMed]
7
Dubovskii
P.V.
Volynsky
P.E.
Polyansky
A.A.
Karpunin
D.V.
Chupin
V.V.
Efremov
R.G.
Arseniev
A.S.
Three-dimensional structure/hydrophobicity of latarcins specifies their mode of membrane activity
Biochemistry
2008
, vol. 
47
 (pg. 
3525
-
3533
)
[PubMed]
8
Vassilevski
A.A.
Kozlov
S.A.
Samsonova
O.V.
Egorova
N.S.
Karpunin
D.V.
Pluzhnikov
K.A.
Feofanov
A.V.
Grishin
E.V.
Cyto-insectotoxins, a novel class of cytolytic and insecticidal peptides from spider venom
Biochem. J.
2008
, vol. 
411
 (pg. 
687
-
696
)
[PubMed]
9
Kozlov
S.
Malyavka
A.
McCutchen
B.
Lu
A.
Schepers
E.
Herrmann
R.
Grishin
E.
A novel strategy for the identification of toxinlike structures in spider venom
Proteins
2005
, vol. 
59
 (pg. 
131
-
140
)
[PubMed]
10
Kuzmenkov
A.I.
Fedorova
I.M.
Vassilevski
A.A.
Grishin
E.V.
Cysteine-rich toxins from Lachesana tarabaevi spider venom with amphiphilic C-terminal segments
Biochim. Biophys. Acta
2013
, vol. 
1828
 (pg. 
724
-
731
)
[PubMed]
11
Vassilevski
A.A.
Kozlov
S.A.
Egorov
T.A.
Grishin
E.V.
Purification and characterization of biologically active peptides from spider venoms
Methods Mol. Biol.
2010
, vol. 
615
 (pg. 
87
-
100
)
[PubMed]
12
Escoubas
P.
Palma
M.F.
Nakajima
T.
A microinjection technique using Drosophila melanogaster for bioassay-guided isolation of neurotoxins in arthropod venoms
Toxicon
1995
, vol. 
33
 (pg. 
1549
-
1555
)
[PubMed]
13
Shai
Y.
Mechanism of the binding, insertion and destabilization of phospholipid bilayer membranes by α-helical antimicrobial and cell non-selective membrane-lytic peptides
Biochim. Biophys. Acta
1999
, vol. 
1462
 (pg. 
55
-
70
)
[PubMed]
14
Dennison
S.R.
Wallace
J.
Harris
F.
Phoenix
D.A.
Amphiphilic α-helical antimicrobial peptides and their structure/function relationships
Protein Pept. Lett.
2005
, vol. 
12
 (pg. 
31
-
39
)
[PubMed]
15
Polyansky
A.A.
Chugunov
A.O.
Vassilevski
A.A.
Grishin
E.V.
Efremov
R.G.
Recent advances in computational modeling of α-helical membrane-active peptides
Curr. Protein Pept. Sci.
2012
, vol. 
13
 (pg. 
644
-
657
)
[PubMed]
16
Kuhn-Nentwig
L.
Antimicrobial and cytolytic peptides of venomous arthropods
Cell. Mol. Life Sci.
2003
, vol. 
60
 (pg. 
2651
-
2668
)
[PubMed]
17
Polyansky
A.A.
Volynsky
P.E.
Efremov
R.G.
Computer simulations of membrane-lytic peptides: perspectives in drug design
J. Bioinform. Comput. Biol.
2007
, vol. 
5
 (pg. 
611
-
626
)
[PubMed]
18
Shlyapnikov
Y.M.
Andreev
Y.A.
Kozlov
S.A.
Vassilevski
A.A.
Grishin
E.V.
Bacterial production of latarcin 2a, a potent antimicrobial peptide from spider venom
Protein Expr. Purif.
2008
, vol. 
60
 (pg. 
89
-
95
)
[PubMed]
19
Polyansky
A.A.
Vassilevski
A.A.
Volynsky
P.E.
Vorontsova
O.V.
Samsonova
O.V.
Egorova
N.S.
Krylov
N.A.
Feofanov
A.V.
Arseniev
A.S.
Grishin
E.V.
Efremov
R.G.
N-terminal amphipathic helix as a trigger of hemolytic activity in antimicrobial peptides: a case study in latarcins
FEBS Lett.
2009
, vol. 
583
 (pg. 
2425
-
2428
)
[PubMed]
20
Idiong
G.
Won
A.
Ruscito
A.
Leung
B.O.
Hitchcock
A.P.
Ianoul
A.
Investigating the effect of a single glycine to alanine substitution on interactions of antimicrobial peptide latarcin 2a with a lipid membrane
Eur. Biophys. J.
2011
, vol. 
40
 (pg. 
1087
-
1100
)
[PubMed]
21
Won
A.
Ruscito
A.
Ianoul
A.
Imaging the membrane lytic activity of bioactive peptide latarcin 2a
Biochim. Biophys. Acta
2012
, vol. 
1818
 (pg. 
3072
-
3080
)
[PubMed]
22
Kuznetsov
A.S.
Dubovskii
P.V.
Vorontsova
O.V.
Feofanov
A.V.
Efremov
R.G.
Interaction of linear cationic peptides with phospholipid membranes and polymers of sialic acid
Biochemistry (Mosc.)
2014
, vol. 
79
 (pg. 
459
-
468
)
[PubMed]
23
Corzo
G.
Villegas
E.
Gomez-Lagunas
F
Possani
L.D.
Belokoneva
O.S.
Nakajima
T.
Oxyopinins, large amphipathic peptides isolated from the venom of the wolf spider Oxyopes kitabensis with cytolytic properties and positive insecticidal cooperativity with spider neurotoxins
J. Biol. Chem.
2002
, vol. 
277
 (pg. 
23627
-
23637
)
[PubMed]
24
Kozlov
S.A.
Grishin
E.V.
The universal algorithm of maturation for secretory and excretory protein precursors
Toxicon
2007
, vol. 
49
 (pg. 
721
-
726
)
[PubMed]
25
King
G.F.
Hardy
M.C.
Spider-venom peptides: structure, pharmacology, and potential for control of insect pests
Annu. Rev. Entomol.
2013
, vol. 
58
 (pg. 
475
-
496
)
[PubMed]
26
Kuhn-Nentwig
L.
Muller
J.
Schaller
J.
Walz
A.
Dathe
M.
Nentwig
W.
Cupiennin 1, a new family of highly basic antimicrobial peptides in the venom of the spider Cupiennius salei (Ctenidae)
J. Biol. Chem.
2002
, vol. 
277
 (pg. 
11208
-
11216
)
[PubMed]
27
Gremski
L.H.
Trevisan-Silva
D.
Ferrer
V.P.
Matsubara
F.H.
Meissner
G.O.
Wille
A.C.
Vuitika
L.
Dias-Lopes
C.
Ullah
A.
de Moraes
F.R.
, et al. 
Recent advances in the understanding of brown spider venoms: from the biology of spiders to the molecular mechanisms of toxins
Toxicon
2014
, vol. 
83
 (pg. 
91
-
120
)
[PubMed]
28
Yan
L.
Adams
M.E.
Lycotoxins, antimicrobial peptides from venom of the wolf spider Lycosa carolinensis
J. Biol. Chem.
1998
, vol. 
273
 (pg. 
2059
-
2066
)
[PubMed]
29
Budnik
B.A.
Olsen
J.V.
Egorov
T.A.
Anisimova
V.E.
Galkina
T.G.
Musolyamov
A.K.
Grishin
E.V.
Zubarev
R.A.
De novo sequencing of antimicrobial peptides isolated from the venom glands of the wolf spider Lycosa singoriensis
J. Mass Spectrom.
2004
, vol. 
39
 (pg. 
193
-
201
)
[PubMed]
30
Dubovskii
P.V.
Vassilevski
A.A.
Samsonova
O.V.
Egorova
N.S.
Kozlov
S.A.
Feofanov
A.V.
Arseniev
A.S.
Grishin
E.V.
Novel lynx spider toxin shares common molecular architecture with defense peptides from frog skin
FEBS J.
2011
, vol. 
278
 (pg. 
4382
-
4393
)
[PubMed]
31
Raghuraman
H.
Chattopadhyay
A.
Melittin: a membrane-active peptide with diverse functions
Biosci. Rep.
2007
, vol. 
27
 (pg. 
189
-
223
)
[PubMed]
32
Harrison
P.L.
Abdel-Rahman
M.A.
Miller
K.
Strong
P.N.
Antimicrobial peptides from scorpion venoms
Toxicon
2014
, vol. 
88
 (pg. 
115
-
137
)
[PubMed]
33
Orivel
J.
Redeker
V.
Le Caer
J.P.
Krier
F.
Revol-Junelles
A.M.
Longeon
A.
Chaffotte
A.
Dejean
A.
Rossier
J.
Ponericins, new antibacterial and insecticidal peptides from the venom of the ant Pachycondyla goeldii
J. Biol. Chem.
2011
, vol. 
276
 (pg. 
17823
-
17829
)
34
Kuhn-Nentwig
L.
Fedorova
I.M.
Luscher
B.P.
Kopp
L.S.
Trachsel
C.
Schaller
J.
Vu
X.L.
Seebeck
T.
Streitberger
K.
Nentwig
W.
, et al. 
A venom-derived neurotoxin, CsTx-1, from the spider Cupiennius salei exhibits cytolytic activities
J. Biol. Chem.
2012
, vol. 
287
 (pg. 
25640
-
25649
)
[PubMed]
35
Vassilevski
A.A.
Fedorova
I.M.
Maleeva
E.E.
Korolkova
Y.V.
Efimova
S.S.
Samsonova
O.V.
Schagina
L.V.
Feofanov
A.V.
Magazanik
L.G.
Grishin
E.V.
Novel class of spider toxin: active principle from the yellow sac spider Cheiracanthium punctorium venom is a unique two-domain polypeptide
J. Biol. Chem.
2010
, vol. 
285
 (pg. 
32293
-
32302
)
[PubMed]
36
Kuhn-Nentwig
L.
Schaller
J.
Nentwig
W.
Biochemistry, toxicology and ecology of the venom of the spider Cupiennius salei (Ctenidae)
Toxicon
2004
, vol. 
43
 (pg. 
543
-
553
)
[PubMed]
37
Santos
A.D.
Imperial
J.S.
Chaudhary
T.
Beavis
R.C.
Chait
B.T.
Hunsperger
J.P.
Olivera
B.M.
Adams
M.E.
Hillyard
D.R.
Heterodimeric structure of the spider toxin ω-agatoxin IA revealed by precursor analysis and mass spectrometry
J. Biol. Chem.
1992
, vol. 
267
 (pg. 
20701
-
20705
)
[PubMed]
38
Billen
B.
Vassilevski
A.
Nikolsky
A.
Debaveye
S.
Tytgat
J.
Grishin
E.
Unique bell-shaped voltage-dependent modulation of Na+ channel gating by novel insect-selective toxins from the spider Agelena orientalis
J. Biol. Chem.
2010
, vol. 
285
 (pg. 
18545
-
18554
)
[PubMed]