Snake-venom phospholipase A2 neurotoxins. Potentiation of a single-chain neurotoxin by the chaperon subunit of a two-component neurotoxin

Animal Toxins: A Historical Outlook at the Institut Pasteur of Paris

Humans have faced poisonous animals since the most ancient times. It is recognized that certain animals, like specific plants, produce toxic substances that can be lethal, but that can also have therapeutic or psychoactive effects. The use of the term "venom", which initially designated a poison, remedy, or magic drug, is now confined to animal poisons delivered by biting. Following Louis Pasteur's work on pathogenic microorganisms, it was hypothesized that venoms could be related to bacterial toxins and that the process of pathogenicity attenuation could be applied to venoms for the prevention and treatment of envenomation. Cesaire Phisalix and Gabriel Bertrand from the National Museum of Natural History as well as Albert Calmette from the Institut Pasteur in Paris were pioneers in the development of antivenomous serotherapy. Gaston Ramon refined the process of venom attenuation for the immunization of horses using a formalin treatment method that was successful for diphtheria and tetanus toxins. This paved the way for the production of antivenomous sera at the Institut Pasteur, as well as for research on venom constituents and the characterization of their biological activities. The specific activities of certain venom components, such as those involved in blood coagulation or the regulation of chloride ion channels, raises the possibility of developing novel therapeutic drugs that could serve as anticoagulants or as a treatment for cystic fibrosis, for example. Scientists of the Institut Pasteur of Paris have significantly contributed to the study of snake venoms, a topic that is reported in this review.

Amplification of Snake Venom Toxicity by Endogenous Signaling Pathways

The active components of snake venoms encompass a complex and variable mixture of proteins that produce a diverse, but largely stereotypical, range of pharmacologic effects and toxicities. Venom protein diversity and host susceptibilities determine the relative contributions of five main pathologies: neuromuscular dysfunction, inflammation, coagulopathy, cell/organ injury, and disruption of homeostatic mechanisms of normal physiology. In this review, we describe how snakebite is not only a condition mediated directly by venom, but by the amplification of signals dysregulating inflammation, coagulation, neurotransmission, and cell survival. Although venom proteins are diverse, the majority of important pathologic events following envenoming follow from a small group of enzyme-like activities and the actions of small toxic peptides. This review focuses on two of the most important enzymatic activities: snake venom phospholipases (svPLA2) and snake venom metalloproteases (svMP). These two enzyme classes are adept at enabling venom to recruit homologous endogenous signaling systems with sufficient magnitude and duration to produce and amplify cell injury beyond what would be expected from the direct impact of a whole venom dose. This magnification produces many of the most acutely important consequences of envenoming as well as chronic sequelae. Snake venom PLA2s and MPs enzymes recruit prey analogs of similar activity. The transduction mechanisms that recruit endogenous responses include arachidonic acid, intracellular calcium, cytokines, bioactive peptides, and possibly dimerization of venom and prey protein homologs. Despite years of investigation, the precise mechanism of svPLA2-induced neuromuscular paralysis remains incomplete. Based on recent studies, paralysis results from a self-amplifying cycle of endogenous PLA2 activation, arachidonic acid, increases in intracellular Ca2+ and nicotinic receptor deactivation. When prolonged, synaptic suppression supports the degeneration of the synapse. Interaction between endothelium-damaging MPs, sPLA2s and hyaluronidases enhance venom spread, accentuating venom-induced neurotoxicity, inflammation, coagulopathy and tissue injury. Improving snakebite treatment requires new tools to understand direct and indirect effects of envenoming. Homologous PLA2 and MP activities in both venoms and prey/snakebite victim provide molecular targets for non-antibody, small molecule agents for dissecting mechanisms of venom toxicity. Importantly, these tools enable the separation of venom-specific and prey-specific pathological responses to venom.

A Single Mutation Unlocks Cascading Exaptations in the Origin of a Potent Pitviper Neurotoxin

Evolutionary innovations and complex phenotypes seemingly require an improbable amount of genetic change to evolve. Rattlesnakes display two dramatically different venom phenotypes. Type I venoms are hemorrhagic with low systemic toxicity and high expression of tissue-destroying snake venom metalloproteinases. Type II venoms are highly neurotoxic and lack snake venom metalloproteinase expression and associated hemorrhagic activity. This dichotomy hinges on Mojave toxin (MTx), a phospholipase A2 (PLA2) based β-neurotoxin expressed in Type II venoms. MTx is comprised of a nontoxic acidic subunit that undergoes extensive proteolytic processing and allosterically regulates activity of a neurotoxic basic subunit. Evolution of the acidic subunit presents an evolutionary challenge because the need for high expression of a nontoxic venom component and the proteolytic machinery required for processing suggests genetic changes of seemingly little immediate benefit to fitness. We showed that MTx evolved through a cascading series of exaptations unlocked by a single nucleotide change. The evolution of one new cleavage site in the acidic subunit unmasked buried cleavage sites already present in ancestral PLA2s, enabling proteolytic processing. Snake venom serine proteases, already present in the venom to disrupt prey hemostasis, possess the requisite specificities for MTx acidic subunit proteolysis. The dimerization interface between MTx subunits evolved by exploiting a latent, but masked, hydrophobic interaction between ancestral PLA2s. The evolution of MTx through exaptation of existing functional and structural features suggests complex phenotypes that depend on evolutionary innovations can arise from minimal genetic change enabled by prior evolution.

Biophysical studies suggest a new structural arrangement of crotoxin and provide insights into its toxic mechanism

Crotoxin (CTX) is the main neurotoxin found in Crotalus durissus rattlesnake venoms being composed by a nontoxic and non-enzymatic component (CA) and a toxic phospholipase A2 (CB). Previous crystallographic structures of CTX and CB provided relevant insights: (i) CTX structure showed a 1:1 molecular ratio between CA and CB, presenting three tryptophan residues in the CA/CB interface and one exposed to solvent; (ii) CB structure displayed a tetrameric conformation. This study aims to provide further information on the CTX mechanism of action by several biophysical methods. Our data show that isolated CB can in fact form tetramers in solution; however, these tetramers can be dissociated by CA titration. Furthermore, CTX exhibits a strong reduction in fluorescence intensity and lifetime compared with isolated CA and CB, suggesting that all tryptophan residues in CTX may be hidden by the CA/CB interface. By companying spectroscopy fluorescence and SAXS data, we obtained a new structural model for the CTX heterodimer in which all tryptophans are located in the interface, and the N-terminal region of CB is largely exposed to the solvent. Based on this model, we propose a toxic mechanism of action for CTX, involving the interaction of N-terminal region of CB with the target before CA dissociation.

Structural and functional characterization of complex formation between two Kunitz-type serine protease inhibitors from Russell's Viper venom

Snake venom Kunitz-type serine protease inhibitors (KSPIs) exhibit various biological functions including anticoagulant activity. This study elucidates the occurrence and subunit stoichiometry of a putative complex formed between two KSPIs (Rusvikunin and Rusvikunin-II) purified from the native Rusvikunin complex of Pakistan Russell's Viper (Daboia russelii russelii) venom (RVV). The protein components of the Rusvikunin complex were identified by LC-MS/MS analysis. The non-covalent interaction between two major components of the complex (Rusvikunin and Rusvikunin-II) at 1:2 stoichiometric ratio to form a stable complex was demonstrated by biophysical techniques such as spectrofluorometric, classical gel-filtration, equilibrium gel-filtration, circular dichroism (CD), dynamic light scattering (DLS), RP-HPLC and SDS-PAGE analyses. CD measurement showed that interaction between Rusvikunin and Rusvikunin-II did not change their overall secondary structure; however, the protein complex exhibited enhanced hydrodynamic diameter and anticoagulant activity as compared to the individual components of the complex. This study may lay the foundation for understanding the basis of protein complexes in snake venoms and their role in pathophysiology of snakebite.

Synergistically acting PLA₂: peptide hemorrhagic complex from Daboia russelii venom

Snake venoms are complex mixture of enzymatic and non-enzymatic proteins. Non-covalent protein-protein interaction leads to protein complexes, which bring about enhanced pharmacological injuries by their synergistic action. Here we report identification and characterization of a new Daboia russelii hemorrhagic complex I (DR-HC-I) containing phospholipase A₂ (PLA₂) and non-enzymatic peptide. DR-HC-I was isolated from the venom of D. russelii by CM-Shepadex-C25 and gel permeation chromatography. Individual components were purified and identified by RP-HPL chromatography, mass spectrometry and N-terminal amino acid sequencing. DR-HC-I complex was lethal to mice with the LD₅₀ dose of 0.7 mg/kg body weight with hemorrhagic and neurotoxic properties. DR-HC-I complex consists of non-hemorrhagic PLA₂ and neurotoxic non-enzymatic peptide. The non-enzymatic peptide quenched the intrinsic fluorescence of PLA₂ in a dose dependent manner, signifying the synergistic interaction between two proteins. PLA₂ and peptide toxin in a 5:2 M ratio induced skin hemorrhage in mice with MHD 20 μg. However, addition of ANS (1-Anilino-8-naphthalene sulfonate) to DR-HC-I complex inhibited skin hemorrhagic effect and also synergic interaction. But there was no impact on PLA₂ due to this synergistic interaction, and indirect hemolytic or plasma re-calcification activity. However, the synergistic interaction of PLA₂ and non-enzymatic peptide contributes to the enhanced venom-induced hemorrhage and toxicity of Daboia russellii venom.

Enzymatic toxins from snake venom: structural characterization and mechanism of catalysis

Snake venoms are cocktails of enzymes and non-enzymatic proteins used for both the immobilization and digestion of prey. The most common snake venom enzymes include acetylcholinesterases, l-amino acid oxidases, serine proteinases, metalloproteinases and phospholipases A(2) . Higher catalytic efficiency, thermal stability and resistance to proteolysis make these enzymes attractive models for biochemists, enzymologists and structural biologists. Here, we review the structures of these enzymes and describe their structure-based mechanisms of catalysis and inhibition. Some of the enzymes exist as protein complexes in the venom. Thus we also discuss the functional role of non-enzymatic subunits and the pharmacological effects of such protein complexes. The structures of inhibitor-enzyme complexes provide ideal platforms for the design of potent inhibitors which are useful in the development of prototypes and lead compounds with potential therapeutic applications.

Molecular evolution and structure-function relationships of crotoxin-like and asparagine-6-containing phospholipases A2 in pit viper venoms

Some myotoxic or neurotoxic PLA2s (phospholipases A2) from pit viper venoms contain characteristic N6 substitutions. Our survey of the venoms of more than ten pit viper genera revealed that N6-PLA2s exist only in limited Asian pit vipers of two genera, Protobothrops and Gloydius, and exist as either monomers or the basic subunits of heterodimers in some New World pit vipers. For the newly identified N6-PLA2s, the neuromuscular blocking activities were assayed with the chick biventer cervicis neuromuscular tissue, whereas the increased serum creatine kinase level assessed their myotoxicities. The purified N6-PLA2s from Protobothrops mangshanensis and Gloydius intermedius saxatilis were found to be presynaptic neurotoxins. In contrast, all N6-PLA2s from the venoms of Sistrurus miliarius strackeri, S. m. barbouri, Crotalus viridis viridis, C. lepidus lepidus, Cerrophidion godmani and Bothreichis schlegelii were myotoxins without neurotoxicity even in the presence of crotoxin A. Crotoxin-like complexes were for the first time purified from the venoms of Sitrurus catenatus tergeminus, C. mitchelli mitchelli, C. horridus atricaudatus, C. basiliscus and C. durissus cumanensis. The cDNAs encoding six novel N6-PLA2s and subunits of the crotoxin-like complex from S. c. tergeminus were cloned and fully sequenced. Phylogeny analysis showed that two structural subtypes of N6-PLA2s with either F24 or S24 substitution have been evolved in parallel, possibly descended respectively from species related to present-day Protobothrops and Gloydius. Calmodulin binds all the N6-PLA2s but crotoxin A may inhibit its binding to crotoxin B and to other neurotoxic N6-PLA2s. Structure-activity relationships at various regions of the PLA2 molecules were extensively discussed.

Inflammatory oedema induced by phospholipases A2 isolated from Crotalus durissus sp. in the rat dorsal skin: a role for mast cells and sensory C-fibers

The ability of the phospholipases A(2) (PLA(2)s) from Crotalus durissus cascavella, Crotalus durissus collilineatus and Crotalus durissus terrificus venoms and crotapotin to increase the vascular permeability in the rat skin as well as the contribution of both mast cells and sensory C-fibers have been investigated in this study. Vascular permeability was measured as the plasma extravascular accumulation at skin sites of intravenously injected 125I-human serum albumin. Intradermal injection of crotalic PLA(2)s (0.05-0.5 microg/site) in the rat skin resulted in dose-dependent increase in plasma extravascular whereas crotapotin (1 microg/site) failed to affect this response. Co-injection of crotapotin (1 microg/site) did not modify the increased vascular permeability induced by the PLA(2)s (0.05-0.5 microg/site). Previous treatment (30 min) of the animals with cyproheptadine (2 mg/kg, i.p.) markedly reduced PLA(2) (0.5 microg/site)-induced oedema. In rats treated neonatally with capsaicin to deplete neuropeptides, the plasma extravasation induced by all PLA(2)s (0.5 microg/site) was also significantly reduced. Similarly, the tachykinin NK(1) receptor antagonist SR140333 (1nmol/site) significantly reduced the PLA(2)-induced oedema. In addition, the combination of SR140333 with cyproheptadine further reduced the increased plasma extravasation by PLA(2) from C. d. cascavella venom, but not by PLA(2) from C. d. terrificus and C. d. collilineatus venoms. Our results suggest that increase in skin vascular permeability by crotalic PLA(2)s is mediated by activation of sensory C-fibers culminating in the release of substance P, as well as by activation of mast cells which in turn release amines such as histamine and serotonin.

Combining phage display and molecular modeling to map the epitope of a neutralizing antitoxin antibody

Crotoxin is a potent presynaptic neurotoxin from the venom of the rattlesnake Crotalus durissus terrificus. It is composed of the noncovalent and synergistic association of a weakly toxic phospholipase A2, CB, and a nontoxic three-chain subunit, CA, which increases the lethal potency of CB. The A-56.36 mAb is able to dissociate the crotoxin complex by binding to the CA subunit, thereby neutralizing its toxicity. Because A-56.36 and CB show sequence homology and both compete for binding to CA, we postulated that A-56.36 and CB had overlapping binding sites on CA. By screening random phage-displayed libraries with the mAb, phagotopes bearing the (D/S)GY(A/G) or AAXI consensus motifs were selected. They all bound A-56.36 in ELISA and competed with CA for mAb binding, although with different reactivities. When mice were immunized with the selected clones, polyclonal sera reacting with CA were induced. Interestingly, the raised antibodies retained the crotoxin-dissociating effect of A-56.36, suggesting that the selected peptides may be used to produce neutralizing antibodies. By combining these data with the molecular modeling of CA, it appeared that the functional epitope of A-56.36 on CA was conformational, one subregion being discontinuous and corresponding to the first family of peptides, the other subregion being continuous and composed of amino acids of the second family. Phage-displayed peptides corresponding to fragments of the two identified regions on CA reacted with A-56.36 and with CB. Our data support the hypothesis that A-56.36 and CB interact with common regions of CA, and highlight residues which are likely to be critical for CA-CB complex formation.

Biochemical characterization of two crotamine isoforms isolated by a single step RP-HPLC from Crotalus durissus terrificus (South American rattlesnake) venom and their action on insulin secretion by pancreatic islets

Crotamine, a neurotoxin present in the venom of the South American rattlesnake Crotalus durrisus terrificus exists as several polymorphic variants, as demonstrated by recombinant DNA technology (Smith and Schmidt, Toxicon 28 (1990) 575-585). We have isolated native crotamine by chromatography on Sephadex G75, and have purified two crotamine isoforms (F2 and F3) by a single step of RP-HPLC. Native crotamine and RP-HPLC fractions F2 and F3 produced skeletal muscle spasms and spastic paralysis in mice. At low glucose concentrations (2.8-5.6 mmol/l), none of the crotamines altered the insulin secretion by rat isolated islets. In the presence of 16.7 mmol glucose/l, F2 (5 microg/ml), but not F3, increased insulin secretion two-fold, whereas native crotamine (1.5, 5 and 16.5 microg/ml) potentiated the secretion dose-dependently. The increase in insulin secretion induced by F2 fraction (5 microg/ml) was similar to that obtained with 16.5 microg of native crotamine/ml. These results indicate that the mode of action of the F2 and F3 isoforms in beta-cells is different from that in muscle cells. This difference may be related to the binding affinity of each isoform for the Na(+) channels located in the beta-cell membrane. Crotamine isoforms may be valuable tools for studying the involvement of Na(+) channels in the mechanism of insulin secretion.

Effect of crotapotin and heparin on the rat paw oedema induced by different secretory phospholipases A2

The effects of crotapotin (a non-toxic and non-enzymatic acid polypeptide naturally complexed with phospholipase A2) and heparin on rat paw edema induced by different secretory phospholipases A2 (sPLA2) have been investigated. The ability of crotapotin to affect the enzymatic activity of the sPLA2(s) have also been evaluated. Secretory PLA2(s) obtained from both snake (Naja naja, Naja mocambique mocambique, Crotalus adamanteus and Crotalus durissus terrificus) and bee (Apis mellifera) venoms as well as that from bovine pancreas were used in this study. Rat paw oedema was induced by a single subplantar injection of the sPLA2s (5-30 microg/paw) in absence and presence of either crotapotin (10-100 microg/paw) or heparin (50 U/paw). Paw volume was measured using a hydroplethysmometer. Phospholipase A2 from Naja naja, Naja mocambique mocambique, Apis mellifera venoms and the basic component of Crotalus durissus terrificus venom all induced dose-dependent rat paw oedema whereas those from Crotalus adamanteus venom and bovine pancreas were ineffective. Paw oedema induced by PLA2(s) from both Naja naja and Apis mellifera venoms was significantly (P < 0.05) inhibited by crotapotin (0.1-100 microg/site) whereas the Naja mocambique mocambique venom PLA2-induced oedema was significantly potentiated (P < 0.05) by this polypeptide (40 microg/site). On the other hand, heparin (50 U/paw) had no effect on the paw oedema induced by PLA2 from Naja naja and Apis mellifera venoms but significantly inhibited the Naja mocambique mocambique venom PLA2-induced oedema. The measurement of the in vitro phospholipasic activity revealed that crotapotin inhibited by 60-70% the enzymatic activities of PLA2(s) from Crotalus adamanteus, Naja mocambique mocambique, Apis mellifera venoms and bovine pancreas. Our results suggest that despite the great homology between the various types of sPLA2 they interact with crotapotin on cell surfaces in different ways leading to either inhibition or potentiation of the paw oedema by a mechanism unrelated to their enzymatic activities. Since heparin reduced paw oedema induced by PLA2 from Naja mocambique mocambique venom it is likely that this sPLA2 is similar to the novel heparin-sensitive PLA2 found in mast cells.

Molecular mimicry between a monoclonal antibody and one subunit of crotoxin, a heterodimeric phospholipase A2 neurotoxin

Crotoxin is a heterodimeric phospholipase A2 neurotoxin formed by the non-covalent association of an acidic and non-toxic subunit, CA, and a basic and weakly toxic phospholipase A2, CB. The two subunits behave in a synergistic manner. CA enhances the lethal potency of CB by increasing its selectivity of action. The mAb A-56.36, directed against the non-toxic subunit CA, was previously shown to neutralize crotoxin toxicity by dissociating the crotoxin complex. In the present report, a polypeptide sequence similarity was observed between some CDRs of mAb A-56.36 and two regions of CB (pos. 60-80 and 95-110). Phage displayed peptides corresponding to VH2 and VH3 of mAb A-56.36 and to their homologous sequences in CB bind CA to different extents. This observation shows that mAb A-56.36 interacts with a region of CA involved in its interaction with CB, therefore mimicking the binding of CB to CA. A similar approach was used to determine the regions of ammodytoxin A and of agkistrodotoxin, two phospholipase A2 neurotoxins similar to CB, which are involved in the formation of heterocomplexes with CA. The analysis of these data contributes to the determination of stretches of amino acids which could constitute the paratope of mAb A-56.36, as well as the region of association of CB with CA in crotoxin.

Ammodytin L, an inactive phospholipase A2 homologue with myotoxicity in mice, binds to the presynaptic acceptor of the beta-neurotoxic ammodytoxin C in Torpedo: an indication for a phospholipase A2 activity-independent mechanism of action of beta-neurotoxins in fish?

A Ser48 phospholipase A2-homologue, ammodytin L, which is myotoxic in mammals and devoid of any phospholipase A2 activity, completely inhibits the specific binding of the neurotoxic phospholipase A2, ammodytoxin C, to fish presynaptic membranes from Torpedo marmorata electric organ. In cross-linking experiments, 125I-ammodytin L labels the same membrane proteins as 125I-ammodytoxin C (70, 38.5-57.4 and 19.7 kDa). The formation of these adducts is completely prevented by the presence of ammodytoxin C but not of a non-toxic phospholipase A2, ammodytin I2. A chimeric phospholipase A2, constructed by associating the N-terminal half of ammodytoxin to the C-terminal half of ammodytin L, possesses a low, but significant phospholipase A2 activity, however it is not toxic to mice, probably due to abolition of the specific neuronal acceptor binding in mammals. Nevertheless, the chimeric phospholipase A2 is able to interact with the ammodytoxin acceptor in Torpedo marmorata electric organ. The existence of neuronal acceptors for ammodytin L and for the chimeric phospholipase A2 suggests that they may act as neurotoxins in fish. As ammodytin L does not possess any enzymatic activity it, therefore, appears to be an excellent tool to investigate the mechanism of action of beta-neurotoxins independently of their phospholipase A2 activity.

Biologically active human anti-crotoxin scFv isolated from a semi-synthetic phage library

BACKGROUND

The display of repertoires of antibody fragments on the surface of filamentous bacteriophages offers a new way of making antibodies with predefined binding specificities.

OBJECTIVES

Here we explored the use of this technology to find human antibodies with biological properties. Phage-scFv specific for crotoxin, the main toxic component of the venom of the South-American rattlesnake Crotalus durissus terrificus, were isolated from a 'single pot' repertoire of more than 10(8) clones made in vitro from human V gene segments [1]. The crotoxin molecule is composed of two noncovalently linked subunits: a basic and weakly toxic phospholipase A2 (PLA2) called component B (CB) and an acidic, nonenzymatic and nontoxic subunit called component A (CA). CA is able to increase the toxicity as well as the specificity of action of CB simultaneously reducing its enzymatic activity.

STUDY DESIGN

Two clones were isolated (4-21 and 5-3-1) which are specific of the basic subunit CB, but of a moderate affinity (about 10(-7) M). Clones 4-21 and 5-3-1 have different amino acid sequences and different effects on CB properties suggesting that they are raised against different CB epitopes. Purely cholinergic synaptosomes isolated from Torpedo electric organs provide a suitable model to study the presynaptic effects of crotoxin. In this model, CB was shown to induce a larger acetylcholine release than crotoxin.

RESULTS

A dose-dependent increase of acetylcholine release was observed when crotoxin was incubated with increasing amounts of phage-scFv 4-21. This clone was also shown to increase the enzymatic activity of crotoxin. These observations suggest that phage-scFv might dissociate the complex CA-CB. It could be therefore a neutralizing antibody since CB is much less toxic than crotoxin. This shows that 'single pot' libraries are capable of providing not only immunochemical reagents of high specificity but also biological reagents of high quality. The use of this library appears to open new possibilities for immune passive therapy.

Neurotoxic phospholipases A2 ammodytoxin and crotoxin bind to distinct high-affinity protein acceptors in Torpedo marmorata electric organ

We studied the binding of radioiodinated ammodytoxin C, a monomeric phospholipase A2 neurotoxin from Vipera ammodytes, and of radioiodinated crotoxin, a dimeric phospholipase A2 neurotoxin from Crotalus durissus terrificus, to presynaptic membranes from the electric organ of Torpedo marmorata. In both cases, two different families of specific binding sites were identified and characterized. The high-affinity binding sites for both toxins have been shown to be proteins. The low-affinity binding sites were not affected by proteinases or heat, suggesting the involvement of certain lipid structures in this type of binding. By affinity-labeling, [125I]ammodytoxin C was shown to be associated predominantly with membrane proteins of apparent molecular masses of 70,000 and 20,000 Da and to a lesser extent with several proteins of apparent molecular masses ranging between 39,000 and 57,000 Da. [125I]crotoxin, on the other hand bound primarily to a 48,000 Da membrane protein. All phospholipases A2 tested, except beta-bungarotoxin, inhibited the low-affinity specific binding of ammodytoxin C, whereas only neurotoxic phospholipases A2 prevented the high-affinity binding and the cross-linking of ammodytoxin C and crotoxin. The inhibition profiles of high-affinity binding for [125I]crotoxin and for [125I]ammodytoxin C were quite different. Ammodytoxin C and crotoxin did not inhibit each other on their respective high-affinity binding sites. These observations indicate that at least high-affinity binding sites of these two toxins are different. In contrast with crotoxin, the isolated basic subunit CB of crotoxin was able to completely inhibit the high-affinity binding of [125I]ammodytoxin C. Therefore, the acidic subunit CA of crotoxin does not simply act as a chaperone for CB subunit, but it also confers a distinct binding specificity to the crotoxin.

Re-examination of crotoxin-membrane interactions

The interaction of crotoxin with synaptic membranes from Torpedo marmorata has been re-examined, using radioiodinated toxin. In competition experiments, the 'saturable binding' is usually calculated by subtracting the non-saturable binding, determined in the presence of an excess of unlabelled crotoxin, from total binding. Paradoxically, we observed a notable increase of the 'saturable binding' of 125I-crotoxin, defined in this manner in the presence of a high concentration of certain competitors (e.g. crotoxin subunit B, agkistrodotoxin, ammodytoxin and ammodytin I2). This potentiation effect was analysed by competition and cross-linking experiments. The dissociation of the basic crotoxin subunit CB, which differs markedly from crotoxin in its membrane-binding characteristics, was found to be the main reason for the observed phenomenon. 125I-CB could be released from 125I-crotoxin by exchange with a molecule having sufficient affinity towards CA (e.g. CB or agkistrodotoxin) or it could be dissociated from the specific crotoxin membrane-binding site(s) by a competitor molecule. Our results, therefore, suggest a reinterpretation of previous 125I-crotoxin binding studies.

Synaptosomal binding of 125I-labelled daboiatoxin, a new PLA2 neurotoxin from the venom of Daboia russelli siamensis

Daboiatoxin (DbTx), the PLA2 neurotoxin from Daboia russelli siamensis venom, was shown to bind specifically and saturably to rat cerebrocortical synaptosomes and synaptic membrane fragments. Two families of binding sites were detected by equilibrium binding analysis in the presence and absence of Ca2+. Scatchard analysis of biphasic plateaus revealed Kdl 5 nM and Bmax1, 6 pmoles/mg protein, and Kd2 80 nM and Bmax2 20 pmoles/mg protein, respectively, for the high- and low-affinity binding sites. The binding of 125I-DbTx to synaptosomes did not show marked dependence on Ca2+, Mg2+, Co2+ and Sr2+. Native DbTx was the only strong competitor to 125I-DbTx synaptosomal binding (IC50 12.5 nM, KI 5.5 nM). Two other crotalid PLA2 neurotoxins, crotoxin CB and mojave toxin basic subunit, and nontoxic C. Atrox PLA2 enzyme, were relatively weaker inhibitors, while two viperid PLA2 neurotoxins, ammodytoxin A and VRV PL V, were very weak inhibitors. Crotoxin CA was a poor inhibitor even at microM concentrations, whereas no inhibitory effect at all was observed with crotoxin CACB, ammodytoxin C, VRV PL VIIIa, taipoxin, beta-bungarotoxin, or with PLA2 enzymes from N. naja venom, E. schistosa venom, bee venom and porcine pancreas. All other pharmacologically active ligands examined (epinephrine, norepinephrine, histamine, choline, dopamine, serotonin, GABA, naloxone, WB-4101, atropine, hexamethonium and alpha-bun-garotoxin) also failed to interfere with 125I-DbTx binding. As those competitors that showed partial inhibition were effective only at microM concentration range compared to the Kd (5 nM) of 125I-DbTx synaptosomal binding, DbTx could well recognize a different neuronal binding site. Rabbit anti-DbTx polyclonal antisera completely blocked the specific binding. When a range of Ca2+ and K+ channels modulators were examined, Ca2+ channel blockers (omega-conotoxins GVIA and MVIIC, taicatoxin, calciseptine and nitrendiprene) did not affect the binding even at high concentrations, while charybdotoxin was the only K+ channel effector that could partially displace 125I-DbTx synaptosomal binding amongst the K+ channel blockers tested (apamin, dendrotoxin-I, iberiotoxin, MCD-peptide, 4-aminopyridine and tetraethylammonium), suggesting that neither K+ nor Ca2+ channels are associated with DbTx binding sites.

Molecular cloning and characterization of a neurotoxic phospholipase A2 from the venom of Taiwan habu (Trimeresurus mucrosquamatus)

Using gel-filtration chromatography and reverse-phase (RP) HPLC we have purified a presynaptic neurotoxin (designated as trimucrotoxin) from the crude venom of Taiwan habu (Trimeresurus mucrosquamatus). Its complete primary structure was solved by an automated N-terminal sequencing and cDNA sequencing method. The enzyme inhibited the twitch of the chick biventer cervicis muscle at 0.1-1 micrograms/ml and showed lethality in mice (LD50 = 1.2 micrograms/g, when given intravenously). Trimucrotoxin exists mainly as a homodimer of 14 kDa subunits as shown by a gel-filtration experiment, and dissociates into monomers during SDS/PAGE in the absence of Ca2+. However, most of trimucrotoxin migrated as slowly as a trimer during nondenaturing SDS/PAGE in the presence of Ca2+ or Sr2+. Its amino acid sequence identity to crotoxin B and agkistrodotoxin is about 75%, and its cDNA sequence is 82% identical to that of crotoxin B. Rabbit antiserum against trimucrotoxin also cross-reacted with the other crotalid neurotoxic phospholipases A2. Furthermore, the purified acidic subunit of crotoxin potentiated the neurotoxicity of trimucrotoxin. A comparison of the sequences of these crotalid neurotoxins revealed some common features of the possible neurotoxic sites, including residues 6, 11, 76-81 and 119-125.

Inhibition of carrageenin-induced rat paw oedema by crotapotin, a polypeptide complexed with phospholipase A2

1. The effect of purified crotapotin, a non-toxic non-enzymatic chaperon protein normally complexed to a phospholipase A2 (PLA2) in South America rattlesnake venom, was studied in the acute inflammatory response induced by carrageenin (1 mg/paw), compound 48/80 (3 micrograms/paw) and 5-hydroxytryptamine (5-HT) (3 micrograms/paw) in the rat hind-paw. The effects of crotapotin on platelet aggregation, mast cell degranulation and eicosanoid release from guinea-pig isolated lung were also investigated. 2. Subplantar co-injection of crotapotin (1 and 10 micrograms/paw) with carrageenin or injection of crotapotin (10 micrograms/paw) into the contralateral paw significantly inhibited the carrageenin-induced oedema. This inhibition was also observed when crotapotin (10-30 micrograms/paw) was administered either intraperitoneally or orally. Subplantar injection of heated crotapotin (15 min at 60 degrees C) failed to inhibit carrageenin-induced oedema. Subplantar injection of crotapotin (10 micrograms/paw) also significantly inhibited the rat paw oedema induced by compound 48/80, but it did not affect 5-HT-induced oedema. 3. In adrenalectomized animals, subplantar injection of crotapotin markedly inhibited the oedema induced by carrageenin. The inhibitory effect of crotapotin was also observed in rats depleted of histamine and 5-HT stores. 4. Crotapotin (30 micrograms/paw) had no effect on either the histamine release induced by compound 48/80 in vitro or on the platelet aggregation induced by both arachidonic acid (1 nM) and platelet activating factor (1 microM) in human platelet-rich plasma. The platelet aggregation and thromboxane B2 (TXB2) release induced by thrombin (100 mu ml-1) in washed human platelets were also not affected by crotapotin. In addition, crotapotin (10 microg/paw) did not affect the release of 6-oxo-prostaglandin Fla, and TXB2 induced by ovalbumin in sensitized guinea-pig isolated lungs.5. Our results indicate that the anti-inflammatory activity of crotapotin is not due to endogenous corticosteroid release or inhibition of cyclo-oxygenase activity. It is possible that crotapotin may interact with extracellular PLA2 generated during the inflammatory process thereby reducing its hydrolytic activity.