Molecular dissection of cholinergic binding sites: how do snakes escape the effect of their own toxins?

Comparative compositional and functional analyses of Bothrops moojeni specimens reveal several individual variations

Snake venoms are complex protein mixtures with different biological activities that can act in both their preys and human victims. Many of these proteins play a role in prey capture and in the digestive process of these animals. It is known that some snakes are resistant to the toxicity of their own venom by mechanisms not yet fully elucidated. However, it was observed in the Laboratory of Herpetology of Instituto Butantan that some Bothrops moojeni individuals injured by the same snake species showed mortalities caused by envenoming effects. This study analyzed the biochemical composition of 13 venom and plasma samples from Bothrops moojeni specimens to assess differences in their protein composition. Application of sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) showed distinct venom protein profiles, but very homogeneous plasma profiles. Western Blotting (WB) was performed with plasma samples, which were submitted to incubation with the respective venom. Some individuals showed an immunorecognized band zone around 25 kDa, indicating interaction between the same individual plasma and venom proteins. Crossed-WB assay using non-self-plasma and venom showed that this variability is due to venom protein composition instead of plasma composition. These venoms presented higher caseinolytic, collagenolytic and coagulant activities than the venoms without these regions recognized by WB. Mass spectrometry analyses performed on two individuals revealed that these individuals present, in addition to higher protein concentrations, other exclusive proteins in their composition. When these same two samples were tested in vivo, the results also showed higher lethality in these venoms, but lower hemorrhagic activity than in the venoms without these regions recognized by WB. In conclusion, some Bothrops moojeni specimens differ in venom composition, which may have implications in envenomation. Moreover, the high individual venom variability found in this species demonstrates the importance to work with individual analyses in studies involving intraspecific venom variability and venom evolution.

Differential transcript profile of inhibitors with potential anti-venom role in the liver of juvenile and adult Bothrops jararaca snake

Background

Snakes belonging to the Bothrops genus are vastly distributed in Central and South America and are responsible for most cases of reported snake bites in Latin America. The clinical manifestations of the envenomation caused by this genus are due to three major activities—proteolytic, hemorrhagic and coagulant—mediated by metalloproteinases, serine proteinases, phospholipases A₂ and other toxic compounds present in snake venom. Interestingly, it was observed that snakes are resistant to the toxic effects of its own and other snake’s venoms. This natural immunity may occur due the absence of toxin target or the presence of molecules in the snake plasma able to neutralize such toxins.

Methods

In order to identify anti-venom molecules, we construct a cDNA library from the liver of B. jararaca snakes. Moreover, we analyzed the expression profile of four molecules—the already known anti-hemorrhagic factor Bj46a, one gamma-phospholipase A₂ inhibitor, one inter-alpha inhibitor and one C1 plasma protease inhibitor—in the liver of juvenile and adult snakes by qPCR.

Results

The results revealed a 30-fold increase of gamma-phospholipase A₂ inhibitor and a minor increase of the inter-alpha inhibitor (5-fold) and of the C1 inhibitor (3-fold) in adults. However, the Bj46a factor seems to be equally transcribed in adults and juveniles.

Discussion

The results suggest the up-regulation of different inhibitors observed in the adult snakes might be a physiological adaptation to the recurrent contact with their own and even other snake’s venoms throughout its lifespan. This is the first comparative analysis of ontogenetic variation of expression profiles of plasmatic proteins with potential anti-venom activities of the venomous snake B. jararaca. Furthermore, the present data contributes to the understanding of the natural resistance described in these snakes.

[Relationships between venomous function and innate immune function]

Venomous function is investigated in relation to innate immune function in two cases selected from scorpion venom and serpent venom. In the first case, structural analysis of scorpion toxins and defensins reveals a close interrelation between both functions (toxic and innate immune system function). In the second case, structural and functional studies of natural inhibitors of toxic snake venom phospholipases A2 reveal homology with components of the innate immune system, leading to a similar conclusion. Although there is a clear functional distinction between neurotoxins, which act by targeting membrane ion channels, and the circulating defensins which protect the organism from pathogens, the scorpion short toxins and defensins share a common protein folding scaffold with a conserved cysteine-stabilized alpha-beta motif of three disulfide bridges linking a short alpha helix and an antiparallel beta sheet. Genomic analysis suggests that these proteins share a common ancestor (long venom toxins were separated from an early gene family which gave rise to separate short toxin and defensin families). Furthermore, a scorpion toxin has been experimentally synthetized from an insect defensin, and an antibacterial scorpion peptide, androctonin (whose structure is similar to that of a cone snail venom toxin), was shown to have a similar high affinity for the postsynaptic acetylcholine receptor of Torpedo sp. Natural inhibitors of phospholipase A2 found in the blood of snakes are associated with the resistance of venomous snakes to their own highly neurotoxic venom proteins. Three classes of phospholipases A2 inhibitors (PLI-α, PLI-β, PLI-γ) have been identified. These inhibitors display diverse structural motifs related to innate immune proteins including carbohydrate recognition domains (CRD), leucine rich repeat domains (found in Toll-like receptors) and three finger domains, which clearly differentiate them from components of the adaptive immune system. Thus, in structure, function and phylogeny, venomous function in both vertebrates and invertebrates are clearly interrelated with innate immune function.

Proteomic Analysis of the Ontogenetic Variability in Plasma Composition of Juvenile and Adult Bothrops jararaca Snakes

The ontogenetic variability in venom composition of some snake genera, including Bothrops, as well as the biological implications of such variability and the search of new molecules that can neutralize the toxic components of these venoms have been the subject of many studies. Thus, considering the resistance of Bothrops jararaca to the toxic action of its own venom and the ontogenetic variability in venom composition described in this species, a comparative study of the plasma composition of juvenile and adult B. jararaca snakes was performed through a proteomic approach based on 2D electrophoresis and mass spectrometry, which allowed the identification of proteins that might be present at different levels during ontogenetic development. Among the proteins identified by mass spectrometry, antihemorrhagic factor Bj46a was found only in adult plasma. Moreover, two spots identified as phospholipase A₂ inhibitors were significantly increased in juvenile plasma, which can be related to the higher catalytic PLA₂ activity shown by juvenile venom in comparison to that of adult snakes. This work shows the ontogenetic variability of B. jararaca plasma, and that these changes can be related to the ontogenetic variability described in its venom.

Snake alpha-neurotoxin binding site on the Egyptian cobra (Naja haje) nicotinic acetylcholine receptor Is conserved

Evolutionary success requires that animal venoms are targeted against phylogenetically conserved molecular structures of fundamental physiological processes. Species producing venoms must be resistant to their action. Venoms of Elapidae snakes (e.g., cobras, kraits) contain alpha-neurotoxins, represented by alpha-bungarotoxin (alpha-BTX) targeted against the nicotinic acetylcholine receptor (nAChR) of the neuromuscular junction. The model which presumes that cobras (Naja spp., Elapidae) have lost their binding site for conspecific alpha-neurotoxins because of the unique amino acid substitutions in their nAChR polypeptide backbone per se is incompatible with the evolutionary theory that (1) the molecular motifs forming the alpha-neurotoxin target site on the nAChR are fundamental for receptor structure and/or function, and (2) the alpha-neurotoxin target site is conserved among Chordata lineages. To test the hypothesis that the alpha-neurotoxin binding site is conserved in Elapidae snakes and to identify the mechanism of resistance against conspecific alpha-neurotoxins, we cloned the ligand binding domain of the Egyptian cobra (Naja haje) nAChR alpha subunit. When expressed as part of a functional Naja/mouse chimeric nAChR in Xenopus oocytes, this domain confers resistance against alpha-BTX but does not alter responses induced by the natural ligand acetylcholine. Further mutational analysis of the Naja/mouse nAChR demonstrated that an N-glycosylation signal in the ligand binding domain that is unique to N. haje is responsible for alpha-BTX resistance. However, when the N-glycosylation signal is eliminated, the nAChR containing the N. haje sequence is inhibited by alpha-BTX with a potency that is comparable to that in mammals. We conclude that the binding site for conspecific alpha-neurotoxin in Elapidae snakes is conserved in the nAChR ligand binding domain polypeptide backbone per se. This conclusion supports the hypothesis that animal toxins are targeted against evolutionarily conserved molecular motifs. Such conservation also calls for a revision of the present model of the alpha-BTX binding site. The approach described here can be used to identify the mechanism of resistance against conspecific venoms in other species and to characterize toxin-receptor coevolution.

Snake envenomation and protective natural endogenous proteins: a mini review of the recent developments (1991-1997)

The properties of several factors--antihaemorrhagic, antineurotoxic, antimyotoxic--isolated from the blood serum or plasma of different animals are described with more emphasis placed on the structural differences and similarities among the factors of the snake (Trimeresurus flavoviridis) and mammals (Didelphis marsupials and Herpestes edwardsii). Classification of antihaemorrhagic factors of snake and mammals according to structural homologies, and their effectiveness in neutralizing venom haemorrhagic activities in comparison with that of commercial antivenoms are also reviewed. The antineurotoxic factors isolated so far from the sera of viperid (Vipera palestinae, Daboia r. siamensis), crotalid (Crotalus d. terrificus, T. flavoviridis, Agkistrodon b. siniticus) and elapid (Naja naja atra) snakes, as reviewed, are inhibitors of phospholipase A2, and the amino acid sequences, particularly of those inhibitors from the sera of crotalid snakes, do not share significant sequence homology even within the same family Crotalidae. The amino acid sequences of antineurotoxic factors of the snake (Crotalus d. terrificus) also are not homologous to those of the antihaemorrhagic factors from the blood of the snake (T. flavoviridis) or mammals (Didelphis virginiana, Herpestes edwardsii). The mechanism of action of antihaemorrhagic and antineurotoxic factors is briefly discussed as well as the possibility that crotalids and viperids might possess both of those endogenous neutralizing factors in their blood. Some recent findings on the antimyotoxic factors from the snake serum or plasma with inhibition properties against PLA2 activity and myotoxicity of venoms or toxins are also shortly reviewed.

The alpha-bungarotoxin binding site on the nicotinic acetylcholine receptor: analysis using a phage-epitope library

The nicotinic acetylcholine receptor (AcChoR) is a ligand-gated ion channel that is activated upon binding of acetylcholine. alpha-Neurotoxins, in particular alpha-bungarotoxin (alpha-BTX), bind specifically and with high affinity to the AcChoR and compete with binding of the natural ligand. We employed a 15-mer phage-display peptide library to select epitopes reacting with alpha-BTX. Phages bearing the motif YYXSSL as a consensus sequence were found to bind with high affinity to alpha-BTX. The library-derived peptide (MRYYESSLKSYPD) bears amino acid sequence similarities to a region of the alpha-subunit of the Torpedo muscle AcChoR, as well as of other muscle and neuronal AcChoRs that bind alpha-BTX. The library-derived peptide and the corresponding peptides containing residues 187-199 of the Torpedo AcChoR alpha-subunit (WVYYTCCPDTPYL), as well as peptides analogous to the above region in the neuronal AcChoR (e.g., human alpha7; ERFYECCKEPYPD) that binds alpha-BTX, inhibit the binding of alpha-BTX to the intact Torpedo AcChoR with IC50 values of 10(-6) M. A synthetic peptide from a neuronal AcChoR that does not bind alpha-BTX (e.g., human alpha2; ERKYECCKEPYPD) which differs by just one amino acid from the homologous peptide from the alpha-BTX-binding protein (alpha7)-i.e., Lys in alpha2 and Tyr in alpha7-does not inhibit the binding of alpha-BTX to Torpedo AcChoR. These results indicate the requirement for two adjacent aromatic amino acid residues for binding to alpha-BTX.

Localization of histidine residues relevant for the binding of alpha-bungarotoxin to the acetylcholine receptor alpha-subunit in V8-proteolytic fragments

Histidine residues have been shown to be critical for alpha-BgTx binding to the acetylcholine receptor (Lacorazza et al., 1992; Bouzat et al., 1993; Lacorazza et al., 1995). Receptor subunits from Discopyge tschudii were modified with diethylpyrocarbonate (DEP). DEP treatment produces a concentration-dependent decrease of [125I] alpha-BgTx binding to the alpha-subunit. The neurotoxin binding capacity was fully restored by adding the nucleophile hydroxylamine. By proteolytic mapping of the alpha-subunit with V8-protease, we determined that the binding capacity to the fragment alpha V8-19 decreased 80% by DEP treatment. In addition, the [125I] alpha-BgTx binding to the same fragment decreased by 70% when the subunits were reduced and affinity-alkylated. We report the N-terminal sequence of both subunits and V8-fragments (alpha V8-10, alpha V8-13, and alpha V8-18), which constitute a first contribution to the knowledge of the primary structure of the Discopyge tschudii receptor. We propose that the fragment alpha V8-19 contains one or more of the histidine residues involved in the alpha-BgTx binding and probably includes the Cys alpha 192-193 disulfide bond. Only two histidine residues are present in the extracellular sequence of Torpedo californica for such fragments: His alpha 186 and alpha 204.

Nuclear magnetic resonance (NMR) analysis of ligand receptor interactions: the cholinergic system--a model

Elucidation of the molecular mechanisms that govern ligand-receptor recognition is essential to the rational design of specific pharmacological reagents. Whereas often the receptor and its binding site are the target of investigation, study of the ligand in its free and bound state can also reveal important information regarding this recognition process. Nuclear magnetic resonance (NMR) spectroscopy can be extremely useful for such studies. In this review, we discuss the attributes of NMR in the study of ligand receptor interactions. The cholinergic receptor and its binding to the neurotransmitter, acetylcholine, and cholinergic antagonists serve as a model system, illustrating the power of ligand analysis by NMR. The results discussed prove that the region of residues alpha 180-205 of the nicotinic acetylcholine receptor are an essential component of the cholinergic binding site and that ligand binding involves a positively charged hydrophobic motif.

Two subsites in the binding domain of the acetylcholine receptor: an aromatic subsite and a proline subsite

The ligand binding site of the nicotinic acetylcholine receptor (AcChoR) is localized in the alpha-subunit within a domain containing the tandem Cys-192 and -193. By analyzing the binding-site region of AcChoR from animal species that are resistant to alpha-neurotoxins, we have previously shown that four residues in this region, at positions 187, 189, 194, and 197, differ between animals sensitive (e.g., mouse) and resistant (e.g., mongoose and snake) to alpha-bungarotoxin (alpha-BTX). In the present study, we performed site-directed mutagenesis on a fragment of the mongoose AcChoR alpha-subunit (residues 122-205) and exchanged residues 187, 189, 194, and 197, either alone or in combination, with those present in the mouse alpha-subunit sequence. Only the mongoose fragment in which all four residues were mutated to the mouse ones exhibited alpha-BTX binding similar to that of the mouse fragment. The mongoose double mutation in which Leu-194 and His-197 were replaced with proline residues, which are present at these positions in the mouse AcChoR and in all other toxin binders, bound alpha-BTX to approximately 60% of the level of binding exhibited by the mouse fragment. In addition, replacement of either Pro-194 or -197 in the mouse fragment with serine and histidine, respectively, markedly decreased alpha-BTX binding. All other mutations resulted in no or just a small increase in alpha-BTX binding. These results have led us to propose two subsites in the binding domain for alpha-BTX: the proline subsite, which includes Pro-194 and -197 and is critical for alpha-BTX binding, and the aromatic subsite, which includes amino acid residues 187 and 189 and determines the extent of alpha-BTX binding.

Molecular determinants conferring alpha-toxin resistance in recombinant DNA-derived acetylcholine receptors

Sequences of the alpha-subunits of the nicotinic acetylcholine receptor from the snake and mongoose contain several differences in the region between amino acids 183 and 200. Receptors from both of these species reveal resistance to the snake alpha-toxins presumably arising as a protective evolutionary mechanism. Sequence differences include the added glycosylation signals at residue 187 in the mongoose and at residues 189 and 111 in snake. Although previous observations with peptides and fusion proteins either synthesized chemically or in a bacterial expression system indicate that certain amino acid residues may contribute to the resistance, our findings with the intact receptor in an eukaryotic expression system indicate the major role for glycosylation. In this study, we show that addition of glycosylation signals gives rise to virtually complete glycosylation at the added sites, although heterogeneity of oligosaccharide processing is evident. By analysis of combinations of mutants, we document that glycosylation exerts the predominant influence on alpha-toxin binding. Substitutions at other residues are largely without influence as single mutations but appear to decrease affinity further in multiple mutants, particularly where the receptor is glycosylated at the 187 and 189 positions. Glycosylation exerts a major influence on the dissociation as well as the association rates of the alpha-toxin-receptor complex, suggesting that the decrease for alpha-toxin affinity is not simply a consequence of restricted diffusional access, rather glycosylation affects the conformation and stability of the bound complex.

Molecular structure and mechanism of action of the crotoxin inhibitor from Crotalus durissus terrificus serum

An antivenom protein has been identified in the blood of the snake Crotalus durissus terrificus and proved to act by specifically neutralizing crotoxin, the main lethal component of rattlesnake venoms. The aim of this study was to purify the crotoxin inhibitor from Crotalus serum (CICS), and to analyze its mechanism of action. CICS has been purified from blood serum of the Crotalus snake by gel filtration on Sephadex G-200, ion-exchange chromatography on DEAE-Sephacel, and FPLC gel filtration on a Superose 12 column. It is an oligomeric glycoprotein of 130 kDa, made by the non-covalent association of 23-25-kDa subunits. Two different subunit peptides were identified by SDS/PAGE, however, their N-terminal sequences are identical. They are characterized by the absence of methionine residues and a high content of acidic, hydrophobic and cysteine residues. The neutralizing effect of purified CICS towards the neurotoxic effects of crotoxin has been demonstrated in vivo by lethality assays. CICS binds to the phospholipase subunit CB of crotoxin, but not to the acidic chaperon subunit CA; it efficiently inhibits the phospholipase activity of crotoxin and its isolated CB subunit and evokes the dissociation of the crotoxin complex. The molecular mechanism of the interaction between CICS and crotoxin seems to be very similar to that of crotoxin with its acceptor. It is, therefore, tempting to suggest that CICS acts physiologically as a false crotoxin acceptor that would retain the toxin in the vascular system, thus preventing its action on the neuromuscular system.

The nicotinic acetylcholine receptor: structure and autoimmune pathology

The nicotinic acetylcholine receptors (AChR) are presently the best-characterized neurotransmitter receptors. They are pentamers of homologous or identical subunits, symmetrically arranged to form a transmembrane cation channel. The AChR subunits form a family of homologous proteins, derived from a common ancestor. An autoimmune response to muscle AChR causes the disease myasthenia gravis. This review summarizes recent developments in the understanding of the AChR structure and its molecular recognition by the immune system in myasthenia.

Effects of mutations of Torpedo acetylcholine receptor alpha 1 subunit residues 184-200 on alpha-bungarotoxin binding in a recombinant fusion protein

Residues between positions 184 and 200 of the Torpedo acetylcholine receptor alpha 1 subunit were changed by oligonucleotide-directed mutagenesis in a recombinant fusion protein containing residues 166-211. Amino acids were substituted with residues present in the snake alpha subunit, with an alanine, or with a functionally dissimilar residue. The competitive antagonist alpha-bungarotoxin bound to the fusion protein with high apparent affinity (IC50 = 3.2 x 10(-8) M), and binding was competed by agonists and antagonists. Mutation of His-186, Tyr-189, Tyr-190, Cys-192, Cys-193, Pro-194, and Asp-195 greatly reduced or abolished alpha-bungarotoxin binding, while mutation of Tyr-198 reduced binding, indicating these residues play an important role in binding either through functional interaction with neurotoxin residues or by stabilizing the conformation of the binding site. Molecular modeling of acetylcholine receptor residues 184-200 and knowledge of both neurotoxin and receptor residues essential for binding allow analysis of possible structure-function relationships of the interaction of alpha-bungarotoxin with this region of the receptor.

Inhibition of metalloproteinases in Bothrops asper venom by endogenous peptides

Bothrops asper venom contains a variety of degradative enzymes, including metal-ion dependent proteinases as well as low molecular weight peptides. Two of these peptides, pyroglutamate-glutamine-tryptophan (pEQW) and pyroglutamate-asparagine-tryptophan are present in crude venom at concentrations of about 4.5 and 1 mM, respectively. Proteinase fractions from B. asper are inhibited from digesting oxidized insulin B-chain in vitro by both of these tripeptides with an IC50 for pEQW of approximately 0.3 mM. Digestion of purified myotoxin MIII from B. asper venom is also inhibited in vitro by pEQW, suggesting that similar inhibition of proteinase activities probably occurs in the venom gland. Inhibitory peptides present in venom allow snakes to be protected from their own toxic proteinases and inhibit hydrolysis of venom proteins during storage in the venom gland. Upon dilution, such as when venom is injected into prey, peptide inhibitors dissociate from the proteinase and allow their activation. A simple procedure for isolation of these inhibitory peptides is described.

Amino acid residues within the sequence region alpha 55-74 of Torpedo nicotinic acetylcholine receptor interacting with antibodies to the main immunogenic region and with snake alpha-neurotoxins

The sequence region 55-74 of the alpha-subunit of the acetylcholine receptor (AChR) from Torpedo californica electroplax comprises the amino-terminal end of a sequence segment--residues alpha 67-76--forming the main immunogenic region (MIR), which is most frequently recognized by anti-AChR autoantibodies in myasthenia gravis. The synthetic sequence alpha 55-74 of Torpedo AChR binds alpha-bungarotoxin (alpha BTX), suggesting that amino acid residues within this sequence region may contribute to formation of an alpha BTX binding site. Using single-residue substituted synthetic analogues of the sequence alpha 55-74 of Torpedo AChR, in which each residue was sequentially substituted by either glycine or alanine, we sought identification of the amino acids involved in interaction with alpha-neurotoxins and with three different anti-MIR monoclonal antibodies (mAbs 6, 22, and 198). Substitution of Arg55, Arg57, Trp60, Arg64, Leu65, Arg66, Trp67, or Asn68 strongly inhibited alpha-toxin binding, whereas substitutions of Ile61, Val63, Pro69, Ala70, Asp71, or Tyr72 had marginal effects. Substitutions within the region alpha 68-72 significantly diminished binding of anti-MIR mAbs, although residue preferences differed among mAbs. Further, substituting Trp60 substantially reduced binding of mAb 198, and moderately affected binding of mAb 6, and substitution of Asp62 slightly but consistently affected binding of mAbs 6 and 22.

Citrate is an endogenous inhibitor of snake venom enzymes by metal-ion chelation

Citrate levels in selected snake venoms were determined by an enzymatic assay coupled to NADP+ reduction. Citrate concentrations in different viper venoms (n = 5) varied from 95 to 150 mM, in crotalids (n = 3) from 63 to 142 mM, and in elapids (n = 4) from 17 to 163 mM. In Bothrops asper venom Ca(2+)-ion concentrations varied from 2.5 to 3.6 mM, suggesting that the high relative citrate levels may serve to chelate endogenous divalent metal cations, thereby inactivating divalent cation requiring enzymes. Control experiments with B. asper phospholipase A2 MIII in the presence of 2.5 mM Ca2+, showed that the enzyme is completely inhibited by 20 mM citrate. Crotalus adamanteus 5'-nucleotidase and phosphodiesterase are also inhibited 100 and 75%, respectively, by 100 mM citrate. By forming complexes with divalent metal ions, citrate markedly reduces the activities of selected enzymes in snake venoms. Secretion of high concentrations of citrate may represent an important mechanism by which snakes protect themselves against the toxic effects of their own venoms.

Substitution of Torpedo acetylcholine receptor alpha 1-subunit residues with snake alpha 1- and rat nerve alpha 3-subunit residues in recombinant fusion proteins: effect on alpha-bungarotoxin binding

A fusion protein consisting of the TrpE protein and residues 166-211 of the Torpedo acetylcholine receptor alpha 1 subunit was produced in Escherichia coli using a pATH10 expression vector. Residues in the Torpedo sequence were changed by means of oligonucleotide-directed mutagenesis to residues present in snake alpha 1 subunit and rat nerve alpha 3 subunit which do not bind alpha-bungarotoxin. The fusion protein of the Torpedo sequence bound 125I-alpha-bungarotoxin with high affinity (IC50 = 2.5 x 10(-8) M from competition with unlabeled toxin, KD = 2.3 x 10(-8) M from equilibrium saturation binding data). Mutation of three Torpedo residues to snake residues, W184F, K185W, and W187S, had no effect on binding. Conversion of two additional Torpedo residues to snake, T191S and P194L, reduced alpha-bungarotoxin binding to undetectable levels. The P194L mutation alone abolished toxin binding. Mutation of three Torpedo alpha 1 residues to neuronal alpha 3-subunit residues, W187E, Y189K, and T191N, also abolished detectable alpha-bungarotoxin binding. Conversion of Try-189 to Asn which is present in the snake sequence (Y189N) abolished toxin binding. It is concluded that in the sequence of the alpha subunit of Torpedo encompassing Cys-192 and Cys-193, Try-189 and Pro-194 are important determinants of alpha-bungarotoxin binding. Tyr-189 may interact directly with cationic groups or participate in aromatic-aromatic interactions while Pro-194 may be necessary to maintain a conformation conductive to neurotoxin binding.