Chemical and functional homology of myotoxin a from prairie rattlesnake venom and crotamine from South American rattlesnake venom.

Myonecrosis is a serious result of rattlesnake bite and constitutes a persistent clinical problem. In the current study we have isolated crotamine from the venom of Crotalus durissus terrificus to test its ability to cause structural damage to skeletal muscle, and to make direct chemical comparisons with Myotoxin a, a myotoxic polypeptide we recently isolated from prairie rattlesnake (Crotalus viridis viridis) venom. Disc gel electrophoresis, isoelectric focusing, circular dichroic spectroscopy, and amino acid analysis, all indicated a high degree of chemical similarity. Light microscope histology revealed that crotamine caused vacuolizationof skeletal muscle fibers, qualitatively the same as the vacuolization caused by Myotoxin a. The ability of these two basic snake venom polypeptides to cause structural damage to skeletal muscle fibers has significant implications toward more complete understanding of the cause of snake venom-induced myonecrosis.

Conformation of pardaxin, the toxin of the flatfish Pardachirus marmoratus.

Pardaxin is the toxic component isolated from the secretion of the Red Sea flatfish Pardachirus marmoratus. Pardaxin has attracted considerable attention recently because of its use as shark repellent. Conformational information regarding the peptide backbone, disulfide bonds, and tyrosine chromophores was obtained using Raman and circular dichroism (CD) spectroscopy. Analysis of the Raman spectra indicates that the toxin consists of 39% alpha-helix, 23% beta-structure, and 39% random coil. The occurrence of a band at 510--512 cm-1 indicates the conformational geometry of the disulfide C-C-S-S-C-C linkages to be gauche-gauche-gauche. From the lack of a peak in the 2500--2700 cm-1 region we concluded that all half-cystines are involved in disulfide linkages. The far-ultraviolet CD spectrum of pardaxin showed positive 196 and negative 208 and 222 nm bands. The best fit secondary structure based on the CD ellipticities was found to be: 23% alpha-helix, 21% beta-structure, and 56% random coil. Under highly acidic and alkaline conditions the far-ultraviolet pardaxin spectra showed extensive, reversible loss of its helical structure. Due to certain biological characteristics of pardaxin the possible influence of salt on pardaxin's conformation was examined. No evidence of conformation perturbation was detected in either the CD or Raman spectra of pardaxin.

The role of crotoxin subunits in tropical rattlesnake neurotoxic action.

The major toxin (crotoxin) of Crotalus durissus terrificus (neotropical rattlesnake) is known to be a reversible non-covalently associated complex consisting of an acidic and basic subunit. On separation biological activity is found only with the basic subunit, yet, although void of detectable biological activity, the acidic subunit is essential for the full neurotoxic activity of the complex. Recent evidence suggests that crotoxin A serves as a 'chaperone' to enhance the specificity of crotoxin B and, upon binding, crotoxin A is released to the medium. This study was designed to test this hypothesis. Dimethyl suberimidate, a bifunctional cross-linking agent, was used to irreversibly bind the two subunits. Disc electrophoresis, ion-exchange chromatography, molecular sieve chromatography, capillary isotachophoresis and isoelectric precipitation confirm the existence of an inter-subunit covalently cross-linked complex. The conversion of a dissociable complex to a non-dissociable complex abolished neurotoxicity. Although neurotoxicity was lost, phospholipase A2 (phosphatide 2-acyl-hydrolase, EC 3.1.1.4), which is found associated with many presynaptic neurotoxins, was unaffected. The data in this paper add credence to the 'chaperone' concept of crotoxin A and the importance of the reversible nature of the complex for full expression of neurotoxicity.

Studies of an acidic cardiotoxin isolated from the venom of Mojave rattlesnake (Crotalus scutulatus).

A major lethal protein was isolated from the venom of Mojave rattlesnake (Crotalus scutulatus) by successive purification in DEAE column chromatography and isoelectric focusing. This homogeneous and monomeric form of toxin is designated as "Mojave toxin". Unlike basic neurotoxins or cytotoxins isolated from venoms of cobras, kraits and sea snakes, the Mojave toxin is an acidic protein with an isoelectric point of 4.7. The toxin is also different from crotoxin (from Crotalus durissus terrificus) which consists of both acidic and basic components. The molecular weight determined by Sephadex G-75 column chromatography resulted in a value of about 22 000. A singel protein band with a molecular weight of about 12 000, was observed after sodium dodecyl sulfate gel electrophoresis of the reduced Mojave toxin. Isoelectric focusing gel in the presence of 8 M urea also showed a single protein band, suggesting that the toxin is composed of subunits. Unlike the neurotoxic nature of the basic proteins from the venoms of Elapidae and sea snakes (Hydrophiidae) and crotoxin, Mojave toxin is cardiotoxic rather than neurotoxic. It is very likely that venoms of all rattlesnakes from North and Central America contain Mojave toxin as the common toxin.

Specificity of hemorrhagic proteinase from Crotalus atrox (western diamondback rattlesnake) venom.

Hemorrhagic proteinase, HTb, isolated from Crotalus atrox (western diamondback rattlesnake) venom was studied for its specificity. HTb showed fibrinogenase activity, hydrolyzing the A alpha chain of fibrinogen first, followed by the cleavage of the B beta chain. HTb is different from thrombin and did not produce a fibrin clot. The degradation products of fibrinogen were found to be different, indicating that the cleavage sites in the A alpha and B beta chains are different from those of thrombin. N-Benzoyl-Phe-Val-Arg-p-nitroanilide was not hydrolyzed by HTb, although this substrate was hydrolyzed by thrombin and reptilase.

Protein sequencing by computer graphics.

A computer graphics system has been used to fit a putative sequence to an electron density map of a sea snake neurotoxic protein at 2.2 A resolution. The complete sequence of this small protein could be determined from the map with very little ambiguity. In two places probable errors in the published chemical sequence were detected. This is the first instance in which a complete three-dimensional structure was solved with the use of computer graphics alone, without the construction of a physical model.

Conformation of oxytocin studied by laser Raman spectroscopy.

The peptide backbone conformation and salient structural details of oxytocin were examined by laser Raman spectroscopy. Spectra were obtained in the solid phase, water, 2H2O, and dimethyl sulfoxide solutions. A distinct Amide I band was obtained at 1663 cm-1 for aqueous and deuterated samples and 1666 cm-1 for the solid sample. A relatively high frequency Amide III band at 1260 cm-1 was obtained. It is concluded that these Amide I and III bands arise from the "beta-turn"-like conformation of oxytocin. The tyrosine side chain, according to the I850 cm-1/I830 cm-1 intensity ratio, is exposed to the solvent. The S-S stretching vibration at 512 cm-1 indicates the conformation of C-C-S-S-C-C in the disulfide bridge of oxytocin in the ring is gauche-gauche-gauche.

Purification and characterization of a fibrinogenase from the venom of Western Diamondback rattlesnake (Crotalus atrox).

A fibrinogenolytic enzyme was isolated from the venom of Western Diamondback rattlesnake (Crotalus atrox) by a three-step procedure involving gel filtration and anion-exchange chromatography. The molecular weight was estimated as 22 900 by SDS-polyacrylamide gel electrophoresis. The isoelectric point was found to be pH 4.65. The enzyme rapidly destroyed the ability of bovine fibrinogen to form a clot on incubation with thrombin. Incubation of fibrinogen with the fibrinogenolytic enzyme for 5 min resulted in the disappearance of the beta-chain of fibrinogen and the appearance of lower molecular weight fragments. Thus the enzyme can be classified as a beta-fibrinogenase. However, on prolonged incubation of the fibrinogen there was also a partial digestion of the alpha-chain. The fibrinogenase showed no activity towards fibrin or casein or arginine esters. The fibrinogenolytic activity was inhibited by phenylmethanesulphonyl fluoride (PMSF) but was unaffected by EDTA.

Crystallization of the Ca2+-ATPase of skeletal muscle sarcoplasmic reticulum. Inhibition by myotoxin a.

Decavanadate produces extensive ordered arrays of Ca2+-ATPase molecules on sarcoplasmic reticulum (SR) vesicle surfaces [(1984) J. Bioenerg. Biomembranes 16, 491-505] and the basic unit of these crystalline structures seems to be a dimer of Ca2+-ATPase [(1983) J. Ultrastruct. Res. 24, 454-464; (1984) J. Mol. Biol. 174, 193-204]. Myotoxin a, isolated from the venom of the prairie rattlesnake Crotalus viridis viridis, is a muscle-degenerating polypeptide and its primary site of interaction is the SR membrane, where it uncouples CA2+-translocation from CA2+-dependent ATP hydrolysis [(1986) Arch. Biochem. Biophys. 246, 90-97]. The effect of myotoxin a on decavanadate-induced two-dimensional Ca2+-ATPase crystals of SR membranes has been investigated. The toxin inhibits the formation of two-dimensional SR-membrane crystals and disrupts previously formed crystals in a time- and concentration-dependent manner, which parallels the uncoupling of ATP hydrolysis from Ca2+ translocation. Two-dimensional crystalline arrays of the SR membrane have a typical diffraction pattern which, after myotoxin a treatment, displays a progressive loss of order. Decavanadate is an uncompetitive inhibitor of the Ca2+-ATPase enzyme-myotoxin a complex. The present results suggest that a Ca2+-ATPase dimer is required for coupling Ca2+ translocation to Ca2+-dependent ATP hydrolysis.

Ca2+ release induced by myotoxin alpha, a radio-labellable probe having novel Ca2+ release properties in sarcoplasmic reticulum.

1. Myotoxin alpha (MYTX), a polypeptide toxin purified from the venom of prairie rattlesnakes (Crotalus viridis viridis) induced Ca2+ release from the heavy fraction (HSR) but not the light fraction of skeletal sarcoplasmic reticulum at concentrations higher than 1 microM, followed by spontaneous Ca2+ reuptake by measuring extravesicular Ca2+ concentrations using the Ca2+ electrode. 2. The rate of 45Ca2+ release from HSR vesicles was markedly accelerated by MYTX in a concentration-dependent manner in the range of concentrations between 30 nM and 10 microM, indicating the most potent Ca2+ releaser in HSR. 3. The Ca2+ dependency of MYTX-induced 45Ca2+ release has a bell-shaped profile but it was quite different from that of caffeine, an inducer of Ca(2+)-induced Ca2+ release. 4. 45Ca2+ release induced by MYTX was remarkable in the range of pCa between 8 and 3, whereas that by caffeine was prominent in the range of pCa, i.e., between 7 and 5.5. 5. MYTX-induced 45Ca2+ release consists of both early and late components. The early component caused by MYTX at low concentrations (30-300 nM) completed within 20 s, while the late component induced by it at higher concentrations (> 0.3 microM) was maintained for at least 1 min. 6. Both the components were almost completely inhibited by inhibitors of Ca2+ such as Mg2+, ruthenium red and spermine. 7. 45Ca2+ release induced by caffeine or beta,gamma-methyleneadenosine 5'-triphosphate (AMP-PCP) was completely inhibited by high concentrations of procaine. Procaine abolished the early component but not the late one, suggesting that at least the early component is mediated through Ca(2+)-induced Ca2+ release channels. 8. On the basis of these results, the character of Ca2+ release induced by MYTX was quite different from that caused by caffeine or AMP-PCP, suggesting that MYTX induces Ca2+ release having novel properties in HSR. MYTX is the first polypeptide Ca2+ inducer and has become a useful pharmacological tool for clarifying the mechanism of Ca2+ release from skeletal muscle SR.

Sites of action of Mojave toxin isolated from the venom of the Mojave rattlesnake.

1 Mojave toxin isolated from the venom of the Mojave rattlesnake (Crotalus scutulatus scutulatus) produced an irreversible blockade of the contractile response of the mouse hemidiaphragm to stimulation of the phrenic nerve in vitro, at concentrations of 0.16 to 20 mug/ml; the response to direct stimulation was not affected over a testing period of several hours.2 Mojave toxin (1 to 4 mug/g) was injected into the tail vein of mice and the intoxicated hemidiaphragm preparation was removed either for testing the contractile response or for intracellular recording.3 In fully intoxicated hemidiaphragms the contractile response to indirect stimulation was either small and transient or absent, whilst the response to direct stimulation was well maintained.4 Intracellular recording showed that resting membrane potentials of the muscle fibres were within the normal range. Endplates were difficult to locate but miniature endplate potentials (m.e.p.ps) were recorded at sites at which neurally evoked responses either could not be detected or did not exceed 2 mV which corresponds to transmitter release of a few quanta only.5 The mean frequency of m.e.p.ps at fully intoxicated endplates was not significantly different from controls but potassium depolarization produced only a small increase in m.e.p.p. frequency relative to the control response. A 50 Hz tetanus had no effect on m.e.p.p. frequency.6 When a sub-lethal dose (3 mug) of Mojave toxin was injected into one hindlimb of mice and the tissues examined at 72 h, there was histological evidence of myonecrosis.7 The isolated perfused heart of the rat was exposed to recycled Mojave toxin (50 and 100 mug/ml) but showed no change in rate or force of ventricular contraction.8 Post-mortem examination of intoxicated mice showed a frequent incidence of localized areas of interstitial and intra-alveolar haemorrhage in the lungs. Other organs including skin and muscle were not affected.9 Mojave toxin showed antigenic similarities to crotoxin, the lethal neurotoxin in the venom of the South American rattlesnake, as determined by the ability of antiserum raised against crotoxin to neutralize Mojave toxin.10 With systemic Mojave intoxication of rapid onset, the cause of death was respiratory paralysis. However, the toxin acts at multiple sites at differing rates of action. With a slower rate of intoxication, impaired respiration may act synergistically with cardiovascular changes to produce circulatory failure. The desirability of using an antivenin with a high titre against Mojave toxin is indicated.

Effect of diethylenetriaminepentaacetic acid and procaine on hemorrhage induced by rattlesnake venom.

Nineteen compounds and seven combinations of compounds were tested for their ability to neutralize the hemorrhagic activity of Crotalus atrox venom in vitro and in vivo. Two compounds and four combinations were effective in reducing hemorrhage in an in vivo test in which the venom was injected before injection of compounds. DTPA plus procaine HCl was the most effective combination and reduced hemorrhage when injected 1, 5, or 15 minutes after injection of venom. DTPA-procaine reduced hemorrhage induced by injection of C. atrox venom into dogs as well as mice. DTPA in combination with procaine did not reduce myonecrosis or lethality resulting from injection of venom into mice, but it could be used in conjunction with antivenin to treat local tissue damage resulting from rattlesnake venom poisoning.

Raman studies on bradykinin and a cyclic bradykinin.

Laser Raman spectra of bradykinin in water, deuterium oxide, and the solid phase were recorded. From the spectra it was concluded that bradykinin conformation is comprised of ordered and unordered structure. The ordered structure appears to be some form of reverse turn. Furthermore, it seems that there is an enhancement of the turn structure in the solid phase. A cyclic cystine containing analog of bradykinin was also examined with Raman spectroscopy. The cyclic bradykinin analog gives a Raman spectrum very similar to that of the linear bradykinin and therefore must share similar conformational forms with bradykinin. The restrictive Cys-Cys disulfide in the cyclic bradykinin must serve to maintain a conformation acceptable to bradykinin receptors since the cyclic peptide exhibits biological activity.

Calsequestrin is a major binding protein of myotoxin alpha and an endogenous Ca2+ releaser in sarcoplasmic reticulum.

Myotoxin alpha from prairie rattlesnakes induced Ca2+ release from the heavy fraction of sarcoplasmic reticulum at submicromolar concentrations. 125I-Labeled myotoxin alpha (125I-myotoxin alpha) specifically bound to the heavy fraction. Fractionation of the solubilized heavy fraction with a spermine-agarose column gave purified calsequestrin as a major binding protein of 125I-myotoxin alpha. We have first indicated that calsequestrin is a target protein for Ca(2+)-releasing action of myotoxin alpha. Calsequestrin probably plays a key role in physiological Ca2+ release from sarcoplasmic reticulum.

Mechanism of the anticoagulant, Cerastase F-4, isolated from Cerastes cerastes (Egyptian sand viper) venom.

An anticoagulant enzyme, Cerastase F-4, from the venom of Cerastes cerastes was purified to homogeneity and was characterized (1). In the present report the mode of its fibrinogenolytic and fibrinolytic actions, and its effects on some other blood coagulation factors are described. Cerastes F-4 was shown to readily hydrolyze the alpha A chain of fibrinogen followed by the hydrolysis of the beta B chain. The gamma-chain was relatively resistant to hydrolysis. It also degrades the three chains of fibrin at different rates. The degradation products of the two substrates shown on SDS-polyacrylamide gel were quite different from those produced by plasmin, indicating different sites of cleavage by the enzyme. Using specific chromogenic substrates, Cerastase F-4 seems not to show thrombin-like, plasmin-like, kallikrein-like, antithrombin, or antiplasmin actions. Also, it does not activate prothrombin or plasminogen but degrades both of them slowly. It is concluded that the anticoagulation property of the purified enzyme, Cerastase F-4, is due to its destruction of fibrinogen.

Isolation and characterization of an anticoagulant proteinase, cerastase F-4, from Cerastes cerastes (Egyptian sand viper) venom.

An anticoagulant protease, Cerastase F-4, was isolated from the venom of Cerastes cerastes (Egyptian sand viper) by a combination of gel filtration, ion-exchange chromatography, and HPLC. Homogeneity of the purified anticoagulant was established by discontinuous polyacrylamide disc gel electrophoresis and by isotachophoresis. The anticoagulant enzyme is a single polypeptide chain without subunits having a molecular weight of 22,500. It consists of 28% aspartic acid residues and only 7% are basic amino acids. This agrees well with the fact that the anticoagulant is an acidic protein with an isoelectric point of 5.2. The anticoagulant is a proteolytic enzyme which hydrolyzes casein, fibrinogen and fibrin. The enzyme's optimum activity occurs around 55 degrees C. The anticoagulant showed no phospholipase A activity, low lethal activity, low hemorrhagic and capillary permeability activity, and no myotoxic activity.

Thrombolysis with a snake venom protease in a rat model of venous thrombosis.

A fibrin(ogen)olytic protease isolated from the venom of Crotalus atrox (the western diamondback rattlesnake) was tested for thrombolytic activity. The protease, called atroxase, solubilized fibrin when tested on fibrin plates and hydrolyzed fibrinogen rendering it incoagulable with a specific fibrinogenolytic activity of 42 mg fibrinogen/min/mg protein. Atroxase was unable to activate plasminogen. In vivo, fibrinolytic activity was tested on artificial thrombi induced in the posterior vena cava of Sprague-Dawley rats. Thrombolysis was then characterized by angiographic techniques over a period of three hours. Intravenous administration of the protease, at a dosage of 6.0 mg/kg, resulted in thrombolysis within one hour followed by recanalization of the originally occluded vein within two hours. Fibrinogenolytic activity resulted in a 60% decrease in the rat's plasma fibrinogen level. Histological examination of kidney, liver, heart and lung tissue showed no necrosis nor hemorrhage. These results are the first step in evaluating the thrombolytic potential of anticoagulant proteases within C. atrox venom using laboratory animals.

Biochemical characterization of lebetase, a direct-acting fibrinolytic enzyme from Vipera lebetina snake venom.

Lebetase, the fibrinolytic enzyme isolated from Vipera lebetina (Levantine viper) snake venom is a metalloenzyme that contains one mole of Zn2+ and one mole of Ca2+ per mole of protein. Lebetase is inhibited by dithiothreitol, suggesting that disulfide bonds are necessary for holding the structure. Vipera lebetina venom contains several isoforms of lebetase in the interval of pI 4.6-5.4. Two lebetase fractions I (pI of the main component 5.0) and II (pI of the main component 5.3) degrade fibrin and fibrinogen by hydrolysis of the alpha and beta chains. The molecular weights of the cleavage products produced by the two different lebetase fractions are identical. The metal ions, Cd2+, Cu2+, Co2+, inhibit fibrinolytic and caseinolytic activity of lebetase I and II. Using mass spectrometry we characterized differences in molecular masses of lebetase I and II (22719 Da and 22912 Da). Vipera lebetina venom from a single snake contains mainly one form of lebetase. Lebetase I is more stable at low pH than lebetase II. The lebetases I and II inhibit platelet aggregation induced by ADP in a dose-dependent manner.

Binding of myotoxin a to cultured muscle cells.

The binding of radiolabeled myotoxin a to various cultured cell lines was evaluated. One rat skeletal muscle-derived cell line, L8, bound substantially more myotoxin a than did all all other cell lines examined. Several biophysical parameters of myotoxin a-L8 binding were determined. Binding was saturable with a moderate binding affinity. Scatchard analysis and Hill plots indicated a single class of binding sites. The binding was reversible, as demonstrated by chase experiments. Radiolabeled myotoxin a bound to the cell surface at a site inaccessible to the general protease, pronase. Specificity and biological relevance of the binding was suggested by competition with unlabeled toxin and various peptides derived from the toxin. Biologically active peptides, corresponding to the N- and C-terminal sequence of myotoxin a, competed with radiolabeled toxin for L8 binding. It was concluded that the L8 system is a suitable cell model to study myotoxin a mechanism of action.

Cross-reactivities of polyclonal antibodies against lebetase, fibrinolytic enzyme of Levantine viper (Vipera lebetina) venom.

Antibodies to two fractions of fibrinolytic enzyme in Vipera lebetina venom were produced by immunizing mice with chromatographically purified lebetase probes. The antibodies were isolated from mice blood by protein A affinity chromatography. The antibodies to both fractions reacted only with the fibrinolytic enzyme in V. lebetina venom as demonstrated by western immunoblotting. Immunodot assays, ELISA and western immunoblotting revealed that 11 snake venoms including species of Viperidae and Crotalidae but not Elapidae cross-react with lebetase antibodies to varying degrees. The molecular weights of cross-reacting components were detected and compared with the earlier published data.