Site of action of caudoxin, a neurotoxic phospholipase A2 from the horned puff adder (Bitis caudalis) venom

Differential coagulotoxic and neurotoxic venom activity from species of the arboreal viperid snake genus Bothriechis (palm-pitvipers)

The viperid snake genus Bothriechis consists of eleven species distributed among Central and South America, living across low and high-altitude habitats. Despite Bothriechis envenomations being prominent across the Central and South American region, the functional effects of Bothriechis venoms are poorly understood. Thus, the aim of this study was to investigate the coagulotoxic and neurotoxic activities of Bothriechis venoms to fill this knowledge gap. Coagulotoxic investigations revealed Bothriechis nigroviridis and B. schlegelii to have pseudo-procoagulant venom activity, forming weak clots that rapidly break down, thereby depleting fibrinogen levels and thus contributing to a net anticoagulant state. While one sample of B. lateralis also showed weaker pseudo-procoagulant activity, directly clotting fibrinogen, two samples of B. lateralis venom were anticoagulant through the inhibition of thrombin and factor Xa activity. Differential efficacy of PoliVal-ICP antivenom was also observed, with the pseudo-procoagulant effect of B. nigroviridis venom poorly neutralised, despite this same activity in the venom of B. schlegelii being effectively neutralised. Significant specificity of these fibrinogen cleaving toxins was also observed, with no activity upon model amphibian, avian, lizard or rodent plasma observed. However, upon avian plasma the venom of B. nigroviridis exerted a complete anticoagulant effect, in contrast to the pseudo-procoagulant effect seen on human plasma. Neurotoxic investigations revealed B. schlegelii to be unique among the genus in having potent binding to the orthosteric site of the alpha-1 postsynaptic nicotinic acetylcholine receptor (with B. lateralis having a weaker but still discernible effect). This represents the first identification of postsynaptic nAChR neurotoxic activity for Bothriechis. In conclusion this study identifies notable differential activity within the coagulotoxic and postsynaptic neurotoxic activity of Bothriechis venoms, supporting previous research, and highlights the need for further studies with respect to antivenom efficacy as well as coagulotoxin specificity for Bothriechis venoms.

Widespread and Differential Neurotoxicity in Venoms from the Bitis Genus of Viperid Snakes

Research into the neurotoxic activity of venoms from species within the snake family Viperidae is relatively neglected compared with snakes in the Elapidae family. Previous studies into venoms from the Bitis genus of vipers have identified the presence of presynaptic phospholipase A₂ neurotoxins in B. atropos and B. caudalis, as well as a postsynaptic phospholipase A₂ in B. arietans. Yet, no studies have investigated how widespread neurotoxicity is across the Bitis genus or if they exhibit prey selectivity of their neurotoxins. Utilising a biolayer interferometry assay, we were able to assess the binding of crude venom from 14 species of Bitis to the neuromuscular α-1 nAChR orthosteric site across a wide range of vertebrate taxa mimotopes. Postsynaptic binding was seen for venoms from B. arietans, B. armata, B. atropos, B. caudalis, B. cornuta, B. peringueyi and B. rubida. To further explore the types of neurotoxins present, venoms from the representatives B. armata, B. caudalis, B. cornuta and B. rubida were additionally tested in the chick biventer cervicis nerve muscle preparation, which showed presynaptic and postsynaptic activity for B. caudalis and only presynaptic neurotoxicity for B. cornuta and B. rubida, with myotoxicity also evident for some species. These results, combined with the biolayer interferometry results, indicate complex neurotoxicity exerted by Bitis species, which varies dramatically by lineage tested upon. Our data also further support the importance of sampling across geographical localities, as significant intraspecific variation of postsynaptic neurotoxicity was reported across the different localities.

Venomous Landmines: Clinical Implications of Extreme Coagulotoxic Diversification and Differential Neutralization by Antivenom of Venoms within the Viperid Snake Genus Bitis

The genus Bitis comprises 17 snake species that inhabit Africa and the Arabian Peninsula. They are responsible for a significant proportion of snakebites in the region. The venoms of the two independent lineages of giant Bitis (B. arietans and again in the common ancestor of the clade consisting of B. gabonica, B. nasicornis, B. parviocula and B. rhinoceros) induce an array of debilitating effects including anticoagulation, hemorrhagic shock and cytotoxicity, whilst the dwarf species B. atropos is known to have strong neurotoxic effects. However, the venom effects of the other species within the genus have not been explored in detail. A series of coagulation assays were implemented to assess the coagulotoxic venom effects of fourteen species within the genus. This study identified procoagulant venom as the ancestral condition, retained only by the basal dwarf species B. worthingtoni, suggesting anticoagulant venom is a derived trait within the Bitis genus and has been secondarily amplified on at least four occasions. A wide range of anticoagulant mechanisms were identified, such as coagulant and destructive activities upon fibrinogen in both giant and dwarf Bitis and the action of inhibiting the prothrombinase complex, which is present in a clade of dwarf Bitis. Antivenom studies revealed that while the procoagulant effects of B. worthingtoni were poorly neutralized, and thus a cause for concern, the differential mechanisms of anticoagulation in other species were all well neutralized. Thus, this study concludes there is a wide range of coagulotoxic mechanisms which have evolved within the Bitis genus and that clinical management strategies are limited for the procoagulant effects of B. worthingtoni, but that anticoagulant effects of other species are readily treated by the South African polyvalent antivenom. These results therefore have direct, real-work implications for the treatment of envenomed patients.

Berg adder (Bitis atropos) envenoming: an analysis of 14 cases

INTRODUCTION

The berg adder (Bitis atropos) is a little-studied, lesser-known viperid snake found in southern Africa and there is limited information available regarding the manifestations of envenoming.

MATERIALS AND METHODS

This observational series of 14 cases documents features of berg adder envenoming over a period of 16 years (1987-2003). Clinical features of envenomed patients: All 14 patients developed local cytotoxic effects. Thirteen patients developed systemic effects manifesting and documented in varying degrees. These include (1) prominent vomiting (2) disturbances in cranial nerve function (anosmia and altered taste, an ophthalmological triad of ptosis, mydriasis and visual disturbances including loss of accommodation, and dysphagia) (3) a global decrease in motor power where mechanical ventilation was often required for respiratory failure and (4) hyponatraemia (lowest value recorded 111 mmol/L), sometimes with associated convulsions.

DISCUSSION

The full range of polypeptides present in berg adder venom is yet to be characterised. However, two closely related phospholipases A₂ (PLA₂; PLA₂-1 and PLA₂-2) have been purified from the venom of B. atropos and clinical evidence suggests that a natriuretic peptide is also possibly present. Envenoming results in distinctive, sometimes life-threatening, manifestations.

Professor Chen-Yuan Lee, MD (1915-2001), pharmacologist: snake venom research at the Institute of Pharmacology, National Taiwan University

Professor Chen-Yuan Lee was born in Tainan, Taiwan. In 1940, he joined the staff of the Institute of Pharmacology of the university, now named National Taiwan University. Dr Lee began a study of Daboia russelli formosensis venom under the direction of Professor Tsungming Tu who, in the 1930s, initiated the pharmacological studies of Formosan snake venoms carried out at the Institute. Under Professor Lee's direction, the Institute became known internationally for its work on the isolation, composition and characterization of the pharmacological properties of neurotoxins isolated from Formosan elapid venoms. Sophisticated use of the latest techniques revealed the mode of action of postsynaptic -bungarotoxin and presynaptic -bungarotoxin from Bungarus multicinctus venom, postsynaptic cobrotoxin, cytotoxic cardiotoxin and phospholipase A2 from Naja naja atra venom. Through work undertaken with colleagues at the Institute and in foreign countries, Professor Lee made an important contribution to our understanding of the mode of action of snake neurotoxins, and to their use in the elucidation of neuromuscular transmission. In the past decade, C.-Y. Lee was a prominent campaigner for social and political justice in Taiwan.

Differential effects of presynaptic phospholipase A2 neurotoxins on Torpedo synaptosomes

The effects of several phospholipase A2 neurotoxins from snake venoms were examined on purely cholinergic synaptosomes from Torpedo electric organ. The noncatalytic component A of crotoxin had no effect, whereas its phospholipase component B, used alone or complexed to component A, elicited a rapid and dose-dependent acetylcholine (ACh) release and a depolarization of the preparation. Subsequent ACh release evoked by high K+ levels or calcium ionophore was identical to the control after the action of component A but reduced after the action of crotoxin or of component B. These effects were not observed when the phospholipase A2 activity of the toxin was blocked either by replacing Ca2+ by Ba2+ (respectively, activator and inhibitor of phospholipase A2) or by alkylation of component B with p-bromophenacyl bromide. beta-Bungarotoxin, another very potent phospholipase A2 neurotoxin, induced release of little ACh, did not affect ionophore-evoked ACh release, but significantly reduced depolarization-induced ACh release. The single-chain phospholipase A2 neurotoxin agkistrodotoxin behaved like crotoxin component B. A nonneurotoxic phospholipase A2 from mammalian pancrease induced release of an amount of ACh similar to that released by crotoxin but did not affect the evoked responses. The obvious differences in effect of the various neurotoxins suggest that they exert their specific actions on the excitation-secretion coupling process at different sites or by different mechanisms.

Myotoxic components of snake venoms: their biochemical and biological activities

Necrosis of skeletal muscle is produced by two types of snake venom components: single chain peptides consisting of 42-44 amino acid residues and phospholipases A2 representing either single chain proteins or existing as complexes of several enzyme subunits or combined with other nonenzymatic proteins. Vacuolation, lysis and necrosis of skeletal muscle cells are the major pathological effects of these myotoxins. Although the exact mode of action of these toxins is not clear, interactions with the plasma membrane leading to permeability changes for ions and to their complete destruction is evident. The high specificities of some venom phospholipases A2 for skeletal muscle cells suggest a specific binding to certain membrane receptors; however, an enzymatic action on membranes may also be involved.

Primary structure of ammodytoxin C further reveals the toxic site of ammodytoxin

The sequence of ammodytoxin C, a presynaptically toxic, basic phospholipase A2 of Vipera ammodytes ammodytes venom was determined. The toxin differs only in two amino acid residues from the most toxic isotoxin ammodytoxin A and is 18-times less lethal. Ammodytoxin B which is 30-times less lethal than ammodytoxin A differs from it only in three amino acid residues. From the three-dimensional model of ammodytoxin A, it can be seen that mutated regions of ammodytoxin B and ammodytoxin C are on the surface, and relatively distant from each other. The observed decrease in toxicity of ammodytoxin C could be a consequence of changed charge in position 128 where a Lys is exchanged for Glu. The resulting change in electrostatic properties of the molecule which influences the orientation of the molecule during the approach to the charged nerve-terminal membrane might be responsible for the observed decrease in toxicity.

Binding of divalent and trivalent cations with crotoxin and with its phospholipase and its non-catalytic subunits: effects on enzymatic activity and on the interaction of phospholipase component with phospholipids

We have studied the interaction of divalent and trivalent with a potent phospholipase A(2) neurotoxin, crotoxin, from Crotalus durissus terrificus venom. The pharmacological action of crotoxin requires dissociation of its catalytic subunit (component B) and of its non-enzymatic chaperone subunit (component A), then the binding of the phospholipase subunit to target sites on cellular membranes and finally phospholipid hydrolysis. In this report, we show that the phospholipase A(2) activity of crotoxin and of component B required Ca2+ and that other divalent cations (Sr2+, Cd2+ and Ba2+) and trivalent lanthanide ions are inhibitors. The lowest phospholipase A(2) activity was observed in the presence of Ba2+, which proved to be a competitive inhibitor of Ca2+. The binding of divalent cations and trivalent lanthanide ions to crotoxin and to its subunits has been examined by equilibrium dialysis and by spectrofluorimetric methods. We found that crotoxin binds two divalent cations per mole with different affinities; the site presenting the highest affinity (K(d) in the mM range) in involved in the activation (or inhibition) of the phospholipase A(2) activity and must therefore be located on component B, the other site (K(d) higher than 10 mM) is probably localized on component A and does not play any role in the catalytic activity of crotoxin. We also observed that crotoxin component B binds to vesicular and micellar phospholipids, even in the absence of divalent cations. The affinity of this interaction either does not change or else increases by an order of magnitude in the presence of divalent cations.

Neutralization of lethal potency and inhibition of enzymatic activity of a phospholipase A2 neurotoxin, crotoxin, by non-precipitating antibodies (Fab)

Rabbit antibodies were prepared against both purified catalytic (component-B) and purified non-catalytic (component-A) subunits of crotoxin, the major phospholipase A2 neurotoxin from the South American rattlesnake. They cross-react with crotoxin-like toxins from the venom of several Crotalus species as well as with single-chain phospholipase A2 neurotoxins from Crotalid and Viperid venoms (agkistrodontoxin and ammodytoxin A) but not from Elapid venoms (notexin). Immunological cross-reactions of anti-component-A and anti-component-B sera with crotoxin and with its isolated components A and B showed that component-A exposes determinants of low immunogenicity which are present on component-B, whereas the major antigenic determinants of component-B are not present on component-A. Anti-component-B antibodies, but not anti-component-A antibodies, neutralize the lethal potency of crotoxin and inhibit its enzymatic activity. Furthermore, non-precipitating anti-component-B Fab fragments were as potent as antibodies, indicating that crotoxin neutralization results from the binding of the antibodies to the catalytic subunit, rather than the formation of an immunoprecipitate.

Toxicity domain in presynaptically toxic phospholipase A2 of snake venom

About 42 complete amino-acid sequences of phospholipases A2 (phosphatidylcholine 2-acylhydrolase, EC 3.1.1.4) are known, including those of 13 presynaptically toxic enzymes, but the structural features responsible for the neurotoxicity and distinguishing the toxins from the non-neurotoxic enzymes are far from being clear. In this study, we examined the charged-residue distributions and hydrophobic characteristics based on the sequence data and the predicted tertiary structure and proposed a possible toxicity domain. We found that the presynaptically toxic enzymes have three or four more basic amino-acid residues than the non-neurotoxic enzymes at positions 59, 60, 65, 70-73 and 97 or 98. These residues appear to cluster near the surface region at the N-terminal side. The cationic nature of this basic cluster in the toxin is enhanced by the alpha-amino group of the N-terminus and the dipole moment of helices 96-110 and 1-10. Moreover, these toxic-site residues are usually associated with hydrophobic regions at 1-7, 64-81 and 97-109.

Studies on the subunit structure of textilotoxin, a potent neurotoxin from the venom of the Australian common brown snake (Pseudonaja textilis)

Textilotoxin is a presynaptic neurotoxin from the venom of the Australian common brown snake, Pseudonaja textilis. It has the highest lethality and is structurally the most complex of any known snake venom neurotoxin. It was resolved into its five non-covalently linked subunits in a single step by reverse-phase HPLC. Two of the subunits were identical. The N-terminal amino-acid sequence and amino-acid composition of each subunit were determined. Subunit A was the only one found to possess phospholipase A activity. Separation of textilotoxin into its subunits was reversible and reformed textilotoxin had the same Mr and lethality in mice as the native toxin. Experiments with various unnatural combinations of subunits have led to interesting variations in lethality and Mr of the resulting complexes.

The role of enzyme activity and charge properties on the presynaptic neurotoxicity and the contracture-inducing activity of snake venom phospholipases A2

Two toxic phospholipases A2, caudoxin from Bitis caudalis venom and the basic phospholipase A2 from Naja nigricollis venom, were selected for a study of the contribution of enzyme activity and charge properties to the presynaptic neurotoxicity and the contracture-inducing activity of the enzyme. Reducing catalytic activity by lowering the incubation temperature or by substitution of Sr2+ for Ca2+ in the bathing medium causes a significant prolongation of the time to neuromuscular block by caudoxin, but does not change the contracture-inducing activity of Naja nigricollis basic phospholipase A2. In the chick muscle pretreated with heparin, the latter enzyme fails to evoke contracture, whereas the presynaptic neuromuscular blocking action of caudoxin remains little affected. Histidine-modified Naja nigricollis enzyme, which has lost 95% of its enzyme activity, retains considerable ability to induce contracture and to abolish indirectly evoked contraction, while lysine-methylated enzyme, which loses only 10% of its enzyme activity, almost completely loses its ability to evoke muscle contracture. It is concluded that enzyme activity is important for the presynaptic effects of caudoxin, but not for the contracture-inducing action of the Naja nigricollis enzyme. In the latter enzyme certain easily methylated lysine residues are critical for its ability to induce muscle contracture.

Structure-function relationships of phospholipases. I: Prediction of presynaptic neurotoxicity

The hydropathic indexes of 24 phospholipases have been calculated from their amino acid sequences. The presynaptic neurotoxic potential of venom phospholipases can be predicted by the use of hydropathy profiles. The presynaptically acting phospholipases have a distinct hydrophobic region around the residues 80-110 which is probably involved in the interaction with the presynaptic membranes. This region is present as a separate helix in the tertiary structure of phospholipases. Such a hydrophobic region is absent in non-neurotoxic phospholipases.

Mode of neuromuscular blocking action of ceruleotoxin

Ceruleotoxin is an acidic toxin protein from a Bungarus venom which has recently been clarified as B. fasciatus venom. The toxin at a low concentration (10(-6) g/ml) abolished the twitch response of the indirectly stimulated biventer cervicis muscle of the chick, without affecting the response of acetylcholine or carbamylcholine. At a higher concentration (10(-5) g/ml), the toxin, in addition to inhibition of the twitch response, caused contracture in the chick biventer cervicis muscle and reduced the response to acetylcholine or carbamylcholine in both the innervated chick cervical and denervated rat diaphragm muscles. Electrophysiological studies on the rat phrenic nerve--diaphragm preparation showed that ceruleotoxin at a low concentration caused an initial inhibition followed by recovery of the quantal content of endplate potentials and then a second phase of inhibition leading to complete failure. At a higher concentration, the toxin gradually reduced the resting membrane potential of the diaphragm muscle to a level lower than 50 mV within 2 hr. In addition, enzyme assay showed that the toxin possessed phospholipase A activity comparable to that of other basic phospholipases A2 of snake venom origin. It is concluded that ceruleotoxin is a phospholipase A2 with presynaptic and myotropic actions on vertebrate nerve--muscle system similar to those of notexin from Notechis scutatus scutatus venom.

Mode of neuromuscular blocking action of a toxic phospholipase A2 from Pseudocerastes fieldi (Field's horned viper) snake venom

The effects of a toxic phospholipase A2 (Fr.Cb) isolated from the venom of Pseudocerastes fieldi were studied on the chick biventer cervicis muscle and the mouse phrenic nerve--diaphragm preparations. In the chick muscle, Fr.Cb (10 micrograms/ml) caused complete neuromuscular blockade without producing contracture or affecting the response of the muscle to acetylcholine. In the mouse diaphragm, Fr.Cb blocked the indirectly elicited contraction without affecting that evoked directly. In a low calcium medium (0.5 mM), Fr.Cb produced a triphasic change of the indirectly elicited contractions. The frequency of miniature endplate potentials (m.e.p.p.s) in the mouse diaphragm was first increased 3--4 fold 40 min after toxin (10 micrograms/ml) application, then gradually decreased, while the amplitude of m.e.p.p.s. was not decreased, even after the evoked release of transmitter had failed. Giant m.e.p.p.s were frequently observed. The quantal content first increased and then decreased gradually. The resting membrane potential and the compound phrenic nerve action potential were not significantly affected by the toxin at 10 micrograms/ml after 2 hr of incubation. The motor nerve terminals in the Fr.Cb intoxicated mouse diaphragm showed swelling and vacuolization of both synaptic vesicles and mitochondria. It is concluded that the toxin produces a neuromuscular blockade by acting selectively on the presynaptic site.