Antimicrobial and cytolytic peptides of venomous arthropods.

As a response to invading microorganisms, the innate immune system of arthropods has evolved a complex arrangement of constitutive and inducible antimicrobial peptides that immediately destroy a large variety of pathogens. At the same time, venomous arthropods have developed an additional offensive system in their venom glands to subdue their prey items. In this complex venom system, several enzymes, low-molecular-mass compounds, neurotoxins, antimicrobial and cytolytic peptides interact together, resulting in extremely rapid immobilization and/or killing of prey or aggressors. This review provides an overview of antimicrobial peptides identified in the hemolymph of venomous arthropods, and especially of cytolytic peptides in their venom. For these peptides a dual role is proposed: acting as antimicrobials as well as increasing the potency of the venom by influencing excitable cells.

CSTX-9, a toxic peptide from the spider Cupiennius salei: amino acid sequence, disulphide bridge pattern and comparison with other spider toxins containing the cystine knot structure.

CSTX-9 (68 residues, 7530.9 Da) is one of the most abundant toxic polypeptides in the venom of the wandering spider Cupiennius salei. The amino acid sequence was determined by Edman degradation using reduced and alkylated CSTX-9 and peptides generated by cleavages with endoproteinase Asp-N and trypsin, respectively. Sequence comparison with CSTX-1, the most abundant and the most toxic polypeptide in the crude spider venom, revealed a high degree of similarity (53% identity). By means of limited proteolysis with immobilised trypsin and RP-HPLC, the cystine-containing peptides of CSTX-9 were isolated and the disulphide bridges were assigned by amino acid analysis, Edman degradation and nanospray tandem mass spectrometry. The four disulphide bonds present in CSTX-9 are arranged in the following pattern: 1-4, 2-5, 3-8 and 6-7 (Cys6-Cys21, Cys13-Cys30, Cys20-Cys48, Cys32-Cys46). Sequence comparison of CSTX-1 with CSTX-9 clearly indicates the same disulphide bridge pattern, which is also found in other spider polypeptide toxins, e.g. agatoxins (omega-AGA-IVA, omega-AGA-IVB, mu-AGA-I and mu-AGA-VI) from Agelenopsis aperta, SNX-325 from Segestria florentina and curtatoxins (CT-I, CT-II and CT-III) from Hololena curta. CSTX-1/CSTX-9 belong to the family of ion channel toxins containing the inhibitor cystine knot structural motif. CSTX-9, lacking the lysine-rich C-terminal tail of CSTX-1, exhibits a ninefold lower toxicity to Drosophila melanogaster than CSTX-1. This is in accordance with previous observations of CSTX-2a and CSTX-2b, two truncated forms of CSTX-1 which, like CSTX-9, also lack the C-terminal lysine-rich tail.

A lysine rich C-terminal tail is directly involved in the toxicity of CSTX-1, a neurotoxic peptide from the venom of the spider Cupiennius salei.

CSTX-1 (74 amino acids, 8,352.62 Da) is a potent neurotoxin from the venom of Cupiennius salei. With the monoclonal antibody 9H3 against CSTX-1, we identified two similar peptides by Western blot analysis. These two peptides were purified by RP-HPLC: CSTX-2a (61 amino acids, 6865.75 Da) and CSTX-2b (60 amino acids, 6709.57 Da). Using ESI-MS analysis and sequencing we verified that CSTX-2a is a truncated version of CSTX-1. CSTX-2b differs from CSTX-2a by the absence of Arg61. Toxicity of CSTX-1, CSTX-2a, and CSTX-2b to Drosophila melanogaster showed that the absence of the last 13 amino acids of CSTX-1 results in a seven-fold activity loss. CSTX-2b, which lacks Arg61 is 190-fold less toxic. We conclude that the C-terminal part of CSTX-1, especially Arg61, is essential for the expression of toxicity. CSTX-1 is degraded to CSTX-2a and CSTX-2b by proteases that are released from venom gland cells by apocrine secretion.

Immunocytochemical localization and secretion process of the toxin CSTX-1 in the venom gland of the wandering spider Cupiennius salei (Araneae: Ctenidae).

Fluorescein and horseradish peroxidase-labeled monoclonal antibodies were used to localize the predominant toxic peptide CSTX-1 in the venom gland of the spider Cupiennius salei. There was no polarity of CSTX-1 expression in repleted glands, whereas the glands of previously milked spiders showed a decreasing immunofluorescent response from the distal to the proximal portion. Detailed investigation revealed a new structure in the venom-secreting epithelium, which is postulated to be an evolutionary adaptation to increasing gland volume. CSTX-1 was found to be synthesized and stored as a fully active toxin within complex units, composed of long interdigitating cells running perpendicular to the muscular sheath and extending into the central lumen of the gland. These venom-producing units were found in all sectors of the gland, including the transitional region between the main gland and the venom duct. The venom is liberated from the venom-producing units into the glandular lumen following the contraction of the surrounding muscle layer. Free nuclei or other cellular fragments, which would have provided evidence for a holocrine secretion process, were not found in the glandular lumen or in the crude venom obtained by electrical stimulation. The fine regulation of the spider's venom injection process is postulated to be the function of the bulbous ampulla, situated in the anterior third of the venom duct.

Characterisation of antibacterial activity of peptides isolated from the venom of the spider Cupiennius salei (Araneae: Ctenidae).

The characterisation of the antimicrobial activity of five antibacterial peptides, isolated from the venom of the neotropical wandering spider Cupiennius salei is reported here. The peptides have a molecular mass, determined by electrospray ionisation-mass spectrometry, between 3-4 kDa. Minimal inhibitory concentrations against five different bacteria species were determined by a liquid growth inhibition assay. All five peptides showed minimal inhibitory concentrations that are comparable to those of other known antibacterial peptides, like insect defensins and cecropins, found in the last years in a large diversity of animals. The peptides are supposed to lyse the cells by formation of either distinct channels or pores, but their mode of action is not yet revealed.

Effects of size, motility and paralysation time of prey on the quantity of venom injected by the hunting spider Cupiennius salei.

Previous experimental studies have shown that neotropical wandering spiders (Cupiennius salei) inject more venom when attacking larger crickets. It has been postulated that this is a consequence of predator-prey interactions during envenomation, which increase in intensity with the size of a given prey species. The present study was designed to test this hypothesis using anaesthetized crickets of different sizes that were moved artificially. Cupiennius salei was found (1) to inject more venom the greater the intensity of the struggling movement of the crickets (prey size kept constant); (2) to inject more venom the longer the duration of the struggling movement of the crickets (prey size and intensity of movement kept constant); and (3) to inject equal amounts into crickets of different size (duration and intensity of movement kept constant). These results indicate that C. salei alters the amount of venom it releases according to the size and motility of its prey. Venom expenditure depends mainly on the extent of the interactions with the prey during the envenomation process, whereas prey size is of minor significance. The regulation of venom injection in concert with behavioural adaptations in response to various types of prey minimizes the energetic cost of venom production, thus increasing the profitability of a given prey item.

Quantifying the venom dose of the spider Cupiennius salei using monoclonal antibodies.

The variation in venom dose with prey size of the neotropical wandering spider Cupieinnius salei was examined experimentally. Monoclonal antibodies were raised against the venom toxins of C. salei. Mab 9H3, recognizing the main toxin CSTX-1, was used to quantify the venom by enzyme-linked immunosorbent assay (ELISA). Crickets (Achta domesticus) in four size classes were randomly offered to sixteen mature female spiders at 14d intervals. The prey items were removed from spiders five minutes after the initial bite and subsequently homogenized for ELISA measurements. The quantity of venom expended was significantly related to the size of prey, ranging from 0.15 microl for the smallest (100 110 mg) to 1.53 microl for the largest (600-660 mg) crickets. Adaptations to prey size were also reflected in capturing behavior. None of the smallest, but almost 50% of the largest crickets were wrapped in silk following the spiders bite. Some other behavioral features may reduce the energetic costs of venom production. In 22% of the smallest crickets no venom was detectable, with the majority showing mechanical damage as a result of fang contact. This indicates. that C. salei does not rely exclusively on its venom when feeding on small prey. Some other aspects such as the site of the bite and the speed of paralyzation and their consequences associated with the amount of venom expended are discussed.

Quantity and quality of venom released by a spider (Cupiennius salei, Ctenidae).

The amount of venom injected by the spider Cupiennius salei depended on the efficiency of the mechanical defence of the prey species. Spiders were milked for the first venom (i.e. the first microlitre of venom emitted) versus remaining venom, and for venom regenerated from emptied glands. HPLC gel filtration and IEF electrophoresis showed that the protein content of the first venom was only half as compared to that of the remaining venom, and that this was due to a dilution of all proteins. Venom regeneration came in two speeds. The amount of venom was regenerated more rapidly than the protein concentration. Newly regenerated venom as compared to older venom was characterized by a lower concentration of all proteins and by a higher total concentration of free amino acids, whereas histamine and taurine did not follow this trend. K+ concentration and pH remained similar during venom regeneration. Crickets injected with the venoms showed less acute symptoms when the protein concentration was lower, namely with the first venom and with newly regenerated venom. Consequently, a spider which modulates the quantity of venom injected into a prey also directly changes the venom quality. The ecological consequences of this are discussed. This paper also discusses which region of a gland (ampulla, extracellular and intracellular parts of the glandular sac) is involved in the changes of the venom quality.

Purification of toxic peptides and the amino acid sequence of CSTX-1 from the multicomponent venom of Cupiennius salei (Araneae:Ctenidae).

The venom of the wandering spider Cupiennius salei was analysed biochemically by gel filtration, cation exchange chromatography, RP-HPLC, IEF, SDS-PAGE and TLC-electrophoresis. The native venom contains high levels of Na+, K+, Ca2+, histamine and taurine. It shows considerable activity of hyaluronidase, but not proteolytic activity. Thirteen peptides (CSTX-1 to CSTX-13) with an apparent mol. wt between 2.6 and 12.5 kDa causing differently strong toxic, effects were purified. Toxicity data of the crude venom (insects and mouse) are given and compared with the toxicity of CSTX-1, which causes most of the crude venom's toxicity. CSTX-1 has a mol. wt of 8352.6 and its amino acid sequence of 74 amino acids is given.

Isolation and characterization of human heart cytochrome c oxidase.

Cytochrome c oxidase was isolated from human hearts and separated by SDS gel electrophoresis. The identity of polypeptide bands with known subunits was demonstrated by immunoblotting with monospecific antisera to rat liver cytochrome c oxidase subunits. The polarographically determined kinetics of cytochrome c oxidation were similar to those reported for the bovine heart enzyme.

Orientation of rat liver cytochrome c oxidase subunits investigated with subunit-specific antisera.

The orientation of rat liver cytochrome c oxidase subunits in the inner mitochondrial membrane was investigated with monospecific antisera against subunit II and nine nuclear-coded subunits. Mitoplasts were incubated with the antisera and the amount of bound antibodies was determined either directly with fluorescein-conjugated protein A or indirectly by back-titration of unbound antibodies with a nitrocellulose immunoassay. All subunits were found oriented to the cytosolic side, except subunits VIb and VIIc which did not react with their corresponding antisera. Antisera against subunits I, III and Vb were not available.

Isolation and properties of cytochrome c oxidase from rat liver and quantification of immunological differences between isozymes from various rat tissues with subunit-specific antisera.

Cytochrome c oxidase was isolated from rat liver either by affinity chromatography on cytochrome-c--Sepharose 4B or by chromatography on DEAE-Sepharose. Dodecyl sulfate gel electrophoresis of both preparations showed the same subunit pattern consisting of 13 different polypeptides. Kinetic analysis of the two preparations gave a higher Vmax for the enzyme isolated by chromatography on DEAE-Sephacel. Specific antisera were raised in rabbits against nine of the ten nuclear endoded subunits. A monospecific reaction of each antiserum with its corresponding subunit was obtained by Western blot analysis, thus excluding artificial bands in the gel electrophoretic pattern of the isolated enzyme due to proteolysis, aggregation or conformational modification of subunits. With an antiserum against rat liver holocytochrome c oxidase a different reactivity was found by Western blot analysis for subunits VIa and VIII between isolated cytochrome c oxidases from pig liver or kidney and heart or skeletal muscle. For a quantitative analysis of immunological differences a nitrocellulose enzyme-linked immunosorbent assay was developed. Monospecific antisera against 12 of the 13 subunits of rat liver cytochrome c oxidase were titrated with increasing amounts of total mitochondrial proteins from different rat tissues dissolved in dodecyl sulfate and dotted on nitrocellulose. The absorbance of a soluble dye developed by the second peroxidase-conjugated antibody was measured. From the data the following conclusions were obtained: (a) The mitochondrial encoded catalytic subunits I-III of cytochrome c oxidase are probably identical in all rat tissues. (b) All nine investigated nuclear encoded subunits of cytochrome c oxidase showed immunological differences between two or more tissues. Large immunological differences were found between liver, kidney or brain and heart or skeletal muscle. Minor but significant differences were observed for some subunits between heart and skeletal muscle and between liver, kidney and brain. (c) Between corresponding nuclear encoded subunits of cytochrome c oxidase from fetal and adult tissues of liver, heart and skeletal muscle apparent immunological differences were observed. The data could explain cases of fatal infantile myopathy due to cytochrome c oxidase deficiency.

Immunological identification of four different polypeptides in 'subunit VII' of mammalian cytochrome c oxidase.

Rat liver cytochrome c oxidase was separated by SDS-gel electrophoresis into 13 polypeptide bands. Monospecific antisera against the isolated polypeptides VIIa, VIIb and VIIc were raised in rabbits. Cytochrome c oxidase was blotted on nitrocellulose and incubated with the antisera. The antisera reacted only with their corresponding polypeptides, indicating no immunological relationship between polypeptides VIIa, VIIb and VIIc. The data also exclude that these polypeptides are proteolytic breakdown products of larger subunits.