Neurotoxins from Snake Venom Mojave Toxin from Crotalus Scututa- Tus Is Also Structurally Similar to Crotoxin Anthony T

BIOLOGICALLY ACTIVE AGENTS IN NATURE 56

CHIMIA 52 (1998) NT. 1/2 (lnnu'T/FehruaT)

Chimia 52 (1998) 56-62 and also on the postsynaptic membrane © Neue Schweizerische Chemische Gesellschaft [7], in addition to the presynaptic binding ISSN 0009-4293 site. From a structural viewpoint, both subunits A and B are in the isoforms [8a- c]. Neurotoxins from Snake Venom Mojave toxin from Crotalus scututa- tus is also structurally similar to crotoxin Anthony T. Tu* and is composed of two subunits [9a, b]. There are considerable amino-acid sequence homologies between the two tox- Abstract. Found within snake venoms are a variety oftoxic and nontoxic proteins. The ins [lOa, b]. Mojave toxin inhibits calci- effects of snake venoms depend on all of the components of that venom. However, the um-channel dihydropyridine receptor bind- lethal effects are usually related to the most potent toxins found within the venom; the ing in rat brain [1 I]. two most toxic proteins found in snake venom are neuro- and cardiotoxins. This article Crotoxin's neurotoxic action is very reviews a variety of neurotoxins found in snake venom. The range of neurotoxins similar to t3-btx, but has some difference. present in snake venom is varied and includes: peripheral and noncentral, presynaptic, For instance, crotoxin and its subunit B postsynaptic, potassium-channel inhibiting, muscarinic transmission inhibiting, and have a postsynaptic effect, whereas t3-btx anticholinesterase types. has no such activity [7]. The acidic and basic subunit-types pre- Because a snake venom contains many turn causes the exocytosis of the nerve synaptic toxins are fairly common in neu- different proteins, some are highly toxic, transmitter. The exocytosis is made by the rotoxic snake venoms. For instance, such some are weakly toxic, and some are non- membrane fusion of the vesicle and the toxins have been isolated from the venoms toxic. Toxicity of a snake venom is due to nerve-terminal membranes. In vitro, it was of C. viridis conca/or [12] and C. durissus a combinational effect of all components demonstrated that t3-btx mediated the fu- collilineatus [13]. The amino-acid se- present. However, some snake venoms are sion of liposome. The result of in vitro quence of the basic subunit also has con- more toxic than others because they con- fusion may explain in vivo exocytosis [I]. siderable homology to other snake-venom tain more lethal components. In this re- This can be clearly seen by observing phospholipases A. view article, the most toxic neurotoxin is the change in the miniature end-plate po- The second type of presynaptic neuro- discussed. tential (MEPP). The MEPP is a very small toxin from a chemical viewpoint is that There are several types of neurotoxins potential, observed in the neuromuscular two polypeptide chains are connected by a and their structures; the site of action and junction, that is due to the natural leakage disulfide bond. The most typical toxin of the mechanism are not identical. Snake of acetylcholine from the vesicle. When t3- this type is f3-btx. It consists of two chains: neurotoxins are peripheral neurotoxins, toxin is applied, the MEPP frequently de- The A-chain has 120 amino acids, and the rather than centrally neurotoxic; apparent- creases first (5-10 min), then suddenly B-chain has 60 amino-acid residues. The ly, they do not pass through the blood- increases (for several hours). Finally, the amino-acid sequence of the A-chain is brain barrier. frequency decreases until it becomes zero. similar to the phospholipase A sequence There are several types of presynaptic and, in fact, the A-chain does possess toxins. They are usually structurally dis- phospholipase A activity. For presynaptic 1. Presynaptic Neurotoxins tinct among themselves. However, there activity, phospholipase A activity is es- is one common property, and that is the sential. For instance, when Ca2+ is re- The presynaptic neurotoxins are also possession of phospholipase A activity. placed with Sr2+, phospholi pase A acti vity called t3-toxins in contrast to postsynaptic The toxic phospholipase A is usually a and presynaptic activity are both lost. or a-toxins. When a t3-toxin is added to the basic protein. Tsai et at. [2] found that the Oxidation of methionine at the 6- and neuromuscular preparation, the muscle basic amino acid tended to cluster near the 8-positions lowered the toxicity without contraction starts without stimulation of surface region at the NHTterminal side in affecting antigenicity. Moreover, theNH2- the nerve axon. t3-Toxin usually does not basic phospholipase A. terminal region of the A-chain plays a affect the depolarization of the muscle One type of presynaptic toxin is com- crucial role in maintaining functional ac- itself or have a binding ability to the ace- posed oftwo subunits bound together. The tivity [14]. Heterodimeric f3-btx was ex- tylcholine receptor. It is thus clear that the basic subunit has phospholipase A activi- amined by X-ray diffraction [15]. f3-toxin somehow affects the presynaptic ty, whereas the acidic subunit has no en- The exact mechanism of t3-btx is not end of the nerve and initiates the release of zyme activity. Crotoxin is the first presyn- yet known. But it may be that phospholi- acetylcholine and then eventually stops aptic toxin isolated from snake venom. pase A creates a hole in the nerve-end the release. The role of the acidic subunit A is to membrane and Ca2+ flows to the cyto- At the end of the nerve terminal, there guide the toxin to a specific site; then the plasm. As a result, vesicles containing the are many vesicles containing acetylcho- basic subunit B functions as a presynaptic nerve transmitter acetylcholine discharge line. t3-Bungarotoxin (t3-btx) induces the toxin [3]. Each subunit alone is relatively it. The nucleotide-sequence-encoding t3- 2 influx ofCa + to the nerve terminal; this in nontoxic, but combined, the toxicity is btx Archain has been determined [16]. greatly enhanced [4]. The undissociated Only the B-subunit with phospholipase A2 *Correspondence: Prof. A.T. Tu crotoxin itself shows phospholipase A ac- activity had the ability to release acetyl- Department of Biochemistry and tivity, indicating the active site of subunit choline and the noncatalytic unit of A- Molecular Biology B is not masked by subunit A [5]. Besides subunit has no action [17]. and Colorado State University Yang Chang Fort Collins, Colorado 80523, USA neurotoxicity, subunit B also has hemolyt- [18] reported that phospholipase A2 activ- Tel.: +1970491 1591 ic activity. Subunit B attaches to many ity and presynaptic toxic activity can be Fax: +19704916313 parts of the erythrocyte membranes [6] separated. BIOLOGICALLY ACTIVE AGENTS IN NATURE 57

CHIMIA 52 (1998) Nr. 1/2 (Januar/Februnr)

The third type of presynaptic toxin is a tertiary complex of three polypeptide chains. Taipoxin from the venom of the Australian snake, taipan, has three subunits, a,p, and y, with an Mr of46 000. The number of amino-acid residues present in the subunits is 120, 120, and 135, respec- tively. The a-chain is basic and has phospholipase A activity. The fourth type is a quaternary complex of four polypeptide chains. Textilo- toxin isolated from Pseudonaja textilis consists of A, B, C, and D subunits. Sub- unit D consists of two identical covalently linked polypeptide chains [l9a-c]. The last type is a single polypeptide- chain presynaptic neurotoxin. An exam- ple of this is notexin from the venom of Notechis scutulatus scutulatus, consisting of 119 amino-acid residues with seven Fig. 1. The role of AChR in muscle depolarization disulfide bonds. It has an Mr of 13400. Notexin has isotoxins; they differ in only muscular junction at the postsynaptic site suddenly released from the vesicle at the one amino-acid residue among the two by combining with acetylcholine receptor end of the nerve. Acetylcholine moves isotoxins [20]. The three-dimensional (AChR). These neurotoxins act on the across the synaptic crevice and reaches the structure of notexin was determined by muscle side, rather than the nerve side. acetylcholine receptor in the muscle. The crystallography. The core of the protein is The so-labeled postsynaptic neurotoxins AChR is composed of five subunits, a, a, very similar to other phospholipases A. are really the toxins affecting the particu- 13, y, 8. When two molecules of acetylcho- The difference, however, exists mainly in lar site of the muscle and should not have line attach to the a-subunits, the AChR the area of residues 56-80 and 85-89 [21]. been designated neurotoxins. Actually, changes configuration and becomes an Not every presynaptic toxin is identi- postsynaptic neurotoxins bind to the ace- open ion channel, permitting certain ions cal in relation to the release of acetylcho- tylcholine receptor in the muscle that is to to pass through (Fig. 1). By this mecha- line from the presynaptic site. With ~btx, receive the neurotransmitter acetylcholine. nism, the depolarization wave reaches the there is an initial burst of acetylcholine, On the other hand, the attachment of ace- muscle. but eventually the release is stopped. Even tylcholine to the acetylcholine receptor is The structure of postsynaptic neuro- though toxins may behave like ~btx, the considered a part of the nerve-transmitter toxins is well studied. There are actually length of time for acetylcholine release is mechanism. From this functional view- two types of these neurotoxins (Fig. 2, A, different for each toxin. Some presynaptic point, it is not unreasonable to call them and C). One type has four disulfide bonds toxins do not release acetylcholine from postsynaptic neurotoxins because of their (called type I or ShOli-chain neurotoxins). the beginning and simply stop the release. activity. Therefore, the paralysis of the The short-chain neurotoxi n has one or two In such an event, the depolarization wave muscle by postsynaptic neurotoxin poi- amino acids in segment 8, whereas the never reaches the muscle, and the muscle soning is essentially due to the formation long-chain neurotoxins have a longer seg- is paralyzed. of an acetylcholine-receptor neurotoxin ment 8 (Fig. 2). Another difference is that A. halys pallas venom contains three complex. One should realize that usually a there is only one amino acid within seg- types of phospholipase A; they are acidic, snake venom contains multiple numbers ment 5 of the short-chain neurotoxin, neutral, and the basic. The neutral phos- of neurotoxins. Bungarus multicinctus whereas the long-chain neurotoxin has pholipase A2 has presynaptic toxic action venom is well known as the source of a- three amino-acid residues within the seg- and it was designated as agkistrodon toxin and ~btx, but the venom contains many ment (Fig. 2). [22]. Coral snake (Micrurus nigrocinctus other neurotoxins. For instance, toxin F, Both short- and long-chain neurotox- nigrocinctus) venom contains both pre- which also blocks neuronal nicotinic re- inshave the samebiological activity; name- and postsynaptic toxins, and the presynap- ceptors, has been isolated [25]. Venom ly, to bind to AChR, but there is some tic toxin is phospholipase A2 [23]. Re- from a similar snake, B. fasciatus, also difference in chemical properties. It was moval of the N-terminal octapeptide of the contains various neurotoxins [26a, b]. Sea well documented that the invariant tryp- phospholipase A2 subunit caused the dis- snake, Acalytophis peronii, venom also tophan residue in short-chain neurotoxin integration of the low-affinity Ca2+ bind- contains major and minor neurotoxins. is essential, because the chemical modifi- ing site. The neurotoxicity and the enzy- The only difference between the major cation of this residue caused the loss of matic activity of ~btx were lost [24]. and minor postsynaptic toxins is in the neurotoxicity. However, the modification 43rd residue. The major toxin at this posi- of a tryptophan residue in a-btx, which is tion contains glutamine, whereas the mi- a long-chain neurotoxin, did not apprecia- 2. Postsynaptic Toxins nor toxin contains glutamic acid [27a, b]. bly change the toxicity [28]. When a normal nerve impulse (depo- Most neurotoxins isolated from Aus- 2.1. Nicotinic Type larization wave) passes through the axon tralian Elapidae venoms were reported as Postsynaptic neurotoxins are common- and reaches the end of that axon, the calci- presynaptic neurotoxins, but a postsynap- ly found in the venoms of Hydrophiidae urn-ion concentration is increased and the tic one was isolated from Acanthophis and Elapidae. The toxins affect the neuro- neurotransmitter, acetylcholine (ACh), is antarcticus (Australian death adder) [29]. BIOLOGICALLY ACTIVE AGENTS IN NATURE S8

CHIMIA 52 (1998) Nr. 1/2 (JanuarlFchruar)

One interesting aspect from a structural viewpoint is that the two types of postsyn- A aptic neurotoxins are very similar to Elap- idae venom cardiotoxins (see Fig. 2, B). Cardiotoxins stop the heartbeat when they make contact with the heart. Cardiotoxins have four disulfide bonds and a very short segment 8.ln this manner, they are similar to short-chain neurotoxins. Although the similarity in disulfide bonds and the peptide backbone is remarkable for cardiotoxins and postsynaptic neurotoxins, there are considerable differences between them in amino-acid composition and sequences. Cardiotoxins do not bind to the AChR, whereas there is strong binding between the neurotoxins and the AChR. The hydrophilic index of cardiotoxins shows them to be quite hydrophobic molecules, whereas the neurotoxins are quite hydrophilic molecules. Cardiotoxins are more general toxins, affecting cell membranes, whereas neurotoxins are specific toxins, binding to acetylcholine receptors. Postsynaptic neurotoxins are composed c mainly of an anti parallel t3-sheet and a t3- turn structure, with only a small amount of a-helical structure [30a-f]. The toxin is comprised of three loops, A, B, and C (Fig. 3). The central loop is considered most important, and it is believed that this loop is attached to the acetylcholine-binding site of the AChR. The amino-acid sequences of over J00 postsynaptic neurotoxins have been determined by many investigators; therefore, Fig. 2. Examples of neurotoxins (A, C) and a cardiotoxin (B). A) Primary structure of lapemis toxin the sequence of all the toxins will not be short-chain postsynaptic neurotoxin. B) Cardiotoxin from Naja naja venom. C) Toxin B from Naja discussed. However, one shouJd be aware venom. of the incorrect sequence of a-btx, as originally reported earlier [31]. The correct primary structure of a-bungarotoxin was later established [32]. The original paper [31] reported the sequences ofIle- Pro-Ser(9-11), His-Pro (67-68), and Arg- Gin (71-72). However, these sequences are incorrect, and the correct sequences have now been established as Ser-Pro-Ile (9- J J), Pro-His (67-68), and Gln-Arg (7J- 72) by Ohta et aL. [32]. Snake venoms also contain nonneuro- toxic proteins with structures very similar to a postsynaptic neurotoxin. For instance, mambia is a platelet aggregation inhibitor isolated from the venom of Dendroaspis jamesonii. It has 59 amino-acid residues with four disulfide bonds and a high homology to postsynaptic neurotoxins [33]. Although postsynaptic neurotoxins are small polypeptides with an Mr of ca. 6800, they are antigenic. However, by conjugat-

ing neurotoxin to a protein with a higher Mp Fig. 3. Chemical structure of lapemis antigeneity can be further enhanced [34]. toxin showing three main loops A, B, Some antibodies to cobrotoxin bound andC to the toxin portion insensitive to con for- BIOLOGICALLY ACTIVE AGENTS IN NATURE S9

CHIMIA 52 (1998) Nr. 1/2 (la"uar/Fcbruar) mational change and some antibodies bound to the conformation-sensitive portion of neurotoxin [35]. Neurotoxins from several Elapidae and Hydrophidae venoms can be differentiated from the binding ability to monoclonal antibodies against Naja naja oxiana [36]. The AChR is a pentamer that is comprised offive subunits (two a, one each of {3, y, and 0), and two of them are identical (Fig. 2). The receptor is a ligand (acetylcholine)-gated channel protein, allowing Neurotoxin :> ions to pass through when activated. The - ligand, acetylcholine, attaches to the a- subunits. Since there are two a-subunits, the stoichiometry ofligand-receptor interaction is 2 mol of acetylcholine per receptor. Postsynaptic toxins attach to the same +. No Na Transport +. sites as acetylcholine; however, the AChR No Na Transport receptor fails to form a channel (Fig. 4). Each subunit is a glycoprotein; however, it is not yet clear just what role the polysaccharide, which is present in each subunit, plays. There are several types of Fig. 4. Attachment of a neurotoxin to the same site as that of acetylcholine causes the A ChR to fail polysaccharides in each subunit. One of informing an ion channel in the membrane them is shown here:

suggested that the central loop plays a M(al-2) - M(al-6)", dominant role in the toxin's ability to bind M(al-6) the receptor. M( al-3)"""- > M( 131-4) -NAc G (131-4)-NAcG Although the central loop is the most M(al-2) - M(al-2) - M(al-3) important one, the other two loops are also essential in order to express full neurotox- where M is mannose and NAcG is N- From studies of chemical modification ic activity. Lin eta!. [43] showed that Arg- acetylglucosamine [37]. of amino-acid residues studied by many 34, 37 at the central loop are essential for Neurotoxins actually recognize two investigators, it was shown that the ones neurotoxic activity. The modifications of fragments of a-subunit of AChR, 128- located in the central loop are essential for Lys 50, 56 and Arg 707, 72, Lys-71 at the 142 and 185-199 portions [38]. However, neurotoxicity. For instance, the Arg-31, C-terminal loop and Lys-l0 on the N- a different region of AChR a-subunit, Arg-34, Trp-27, Tyr-23, Lys-24, andLys- terminal loop cause the loss of neurotoxic- residues 173-204, was reported by Lentz 25 are known to be related to neurotoxicity. These results suggest that synergistic [39]. Although it is well recognized that a- ity. It is logical to assume that loop B is effect of three loops takes place in neuro- subunit is the place for toxin binding, it most likely to bind to the AChR. . toxic action. was reported that f3-subunit is also an im- To clarify this problem, synthetic pep- However, Lin and Chang [44] reported portant determinant in receptor localiza- tides identical with A, B, and C loops were that modification of arginine residues tion and sensitivity to neuronal bungaro- made, and their ability to bind to the ace- caused the decrease of lethality, but there toxin [40]. The binding of fragment 185- tylcholine receptor was studied [42]. was no effect of Arg of a-btx. 196to a-btx was studied by NMR, indicat- Peptide synthesis: ing that this portion is indeed the cholin- ergic ligand binding site [41]. A simple, nonradioactive, but sensitive, method was 10 20 30 40 50 60 MTCCNQQSSQPKTTTNCAESSCYKKTWSDHRGTRIERGCGCPVQKPGIKLECCHTNECNN Lapemis toxin developed by Nomoto eta/. [37], who used CCNQQWWQPKTTTNC Peptide A1 horseradish- peroxidase (HRP)-conjugat- YKKTWSDHRGTRIERG Peptide 82 Ls_s-1 CYKKTWSDHRGTRIERGC Peptide 81 ed neurotoxin. I CPQVKPGIKLEC Peptide C1 L s-s----1 L I The acetylcholine receptor and a neu- S- S--l rotoxin form a noncovalent bond-type com- EACDFGHIKLMNPQRSTVWY Nonsense peptide plex. The most important question is what portion of a neurotoxin is really involved in the receptor binding. Is it a particular Only the peptide identical with the Most AChR studies were done using residue, or are several residues involved? central loop B bound to the acetylcholine skeletal muscle or torpedo tissues. The Fig. 3 represents a two-dimensional struc- receptor, whereas the other peptides had acetylcholine-receptor concentration in the ture of a postsynaptic neurotoxin - lape- no detectable binding. The sulfide bond is brain is very small, but it is present. Re- mis toxin from Lapemis hardwickii - that essential for binding. When the central cently, the AChR in the brain has been is based on the X-ray diffraction study of loop peptide was reduced and alkylated, actively studied using snake postsynaptic another similar toxin. the binding ability was lost. This finding neurotoxins. Some of these are rather typ- BIOLOGICALLY ACTIVE AGENTS IN NATURE 60

CHIMIA 52 (1998) Nr. 1/2 Oo"uarlFebmor) ical neurotoxins that bind to both skeletal nals by an unknown mechanism. All j3- The active component was isolated muscles and the brain, and some of them toxins have phospholipase A2 activity, from N. naja sputatrix venom [49]. are specific to the brain AChR [45]. Since which is essential for the presynaptic neu- Both the crude venom and purified a brain a-subunit of AChR binds to a-btx, rotoxicity. Although more than one hun- muscarinicinhibitorsuppressed QNB (qui- there must be a similarity between the dred phospholipases A2 have been isolat- nuclidinyl benzilate )-receptor interaction toxin-binding site for the brain AChR and ed from snake venoms, only a few show in a dose-dependent manner. When the the muscle AChR [46a, b]. neurotoxic action. Unlike pres ynaptic tox- purified preparation was used, a concen- Neuronal membrane components that ins that possess phospholipase A2 activi- tration of muscarinic inhibitor as low as bind a-btx with high affinity can increase ty, all postsynaptic neurotoxins are non- 0.1 ~/ml caused sigificant inhibition of intracellular levels of free calcium, dem- enzymatic and bind irreversibly to nAChR, QNB binding. N. naja sputatrix muscarin- onstrating the components' function as resulting in the blockage of the nerve ic inhibitor was found to have much more nicotinic receptor [47]. Even the retina transmission. inhibitory potential than MTXs from D. contains the AChR that is sensitive to a- Although snake toxins, which associ- angusticeps which require 5.0 ~/ml for btx [48]. ate with the nAChR, have been extensive- the inhibitory action [50]. Iy studied and documented, very few stud- The molecular weight of purified mus- 2.2. Muscarinic Type ies on the interaction of snake toxins with carinic inhibitor was estimated to be 16 There are two types of acetylcholine the mAChR have been carried out. A few kDa. By SDS-PAGE, it was determined to receptors, those responsive to muscarine components which bind selectively to be 16 kDa without j3-mercaptoethanol and (mAChR) and those responsive to nico- mAChR were recently isolated from the to be 14 kDa with the reducing agent. tine (nAChR). The mAChR is present in venom of the green mamba, Dendroaspis Since the molecular weight estimated un- the brain, various smooth muscles, cardi- angusticeps. The toxins, designated as der reduced condition was roughly identi- ac muscle, and exocrine glands; it is clas- muscarinic toxins (MTXs), are small pep- cal with that under nonreduced condition, sified into five subtypes (m 1-5). tides consisting of 65 or 66 amino-acid N. naja sputatrix muscarinic inhibitor is a Many types of neurotoxins, which residues. The three-dimensional solution monomeric single-chain protein. By mass- block nerve transmission, have been iso- structure is very similar to those of short- spectrometric determination, the molecu- lated from the venoms of Elapidae (co- chain postsynaptic neurotoxins; the over- lar mass was confirmed to be 13633 Da. bras, mambas, and kraits), Hydrophiidae, all folding consists of three loops stabi- The sequence for the first 16 residues Crotalidae, and Viperidae. Among them, lized by the four disulfide bonds and form- was determined to be Asn-Leu-Tyr-Gln- the presynaptic and postsynaptic types of ing a two- and a three-stranded j3-sheet. Phe-Lys-Asn-Met-I1e-Gln-Cys- Thr- Val- neurotoxins are the most studied. The pre- Snake venoms that inhibit muscarinic Pro-Asn-Arg. Comparing the amino-acid synaptic type neurotoxins, which are of- AChR are actually quite common and are sequence with many snake-venom pro- ten called j3-toxins, ultimately inhibit the shown in Table 1. teins revealed that the primary structure of release of acetylcholine from nerve termi- the muscarinic inhibitor was very similar to those of phospholipases A2 from the genus Naja. The phospholipase A2 assay Table I, The Abilities of Snake Venoms to Inhibit pH lQNB Binding indicated that N. naja sputatrix muscarin-

ic inhibitor is an active phospholipase A2• [3HIQ B-bound Inhibitory rate nakc veil III [fmol] l'ill 3. Potassium-Channel-Binding Elapidae Neurotoxins BllI/garul ('m'm/ells I 71.3 BIll/guru 110 c/OIIIS 9 83.0 Dendm(l,lpis (mguslict'p 10 1.4 The potassium channel plays an im- Hemal hOlliS haemochuw . 15 65.5 portant role in the repolarization process aja ll«jn 13 71.0 in nerve transmission and is less well clja /lujcl philippi/lC!si.\' 87. known than the sodium channneis in the aja IllIja WlIlurc'IISIS 12 74.5 nerve. The K+ channel is composed of aja lIaja Ipllltlfri_\ 7 .7 membrane proteins and has six transmem- Ophiph(H:us /lOll/wll 2 15.9 brane helical regions. Both the NHr and COOH-terminal chains are located inside Hydrophiidac the membrane. The first snake toxin found WIJemi,\ hal"dll'ickii 2 30.1 to bind K+ is dendrotoxin. Dendrotoxin ullical/ll.a sl'mijalciata 25 and other K+-channel-inhibiting toxins are shown in Table 2. This toxin is a potent Crotalidae convulsant and facilitates transmitter re- A,~ki ImdtJ/l clClIlII 10 79.7 lease by inhibition of voltage-sensitive K+ DC/thral'S atm.\ 9 83.2 channels [51a-f]. CroW/II.1 odamolllC!/J3 30 22.9 Dendrotoxins are more suitable for Cmta/us (//1"0.1' 7 .4 study of the K+ channels than j3-btx, be- 82.3 rota/1/1 I'iridis I'Iridi. 10 cause they lack the intrinsic phospholi- Trim('n'l'u",,~ fla\'Ol'irid; 10 79.7 pase A activity [52]. Dendrotoxin induces repetitive firing in rat-visceral sensory neu- Saline 3 0.0 rons by inhibiting a slowly inactivating 'ajmc+~l tropine 3 100.0 outward K+ current [53]. BIOLOGICALL Y ACTIVE AGENTS IN NATURE 61 CHIMIA 52 (1998) Nr. 1/2 (JnounrlFcbruar)

Dendrotoxin (DTX) has an Mr of7000 Table 2. Potassium-Channel-Binding Neurotoxins [54] and strongly binds to synaptic plasma membranes of rat or chick brain [55]. The Venom amc Jdcnli al wilh receptor has high Mr of 405 000-465 000 [56].Rhem and Lazdunski [57] also isolat- Dendroa,~pis (jllgu.l·ticeps Dendrolo in. nS~C3 ed the K+-channel proteins that bind to ew\o in DTX I. By using neuraminidase and gly- ew\ in copeptidase, K+-channel proteins that bind 1.1Ie, to DTX, f3-btx, and MCD (most cell de- D. poly/epis po/vlepis T in 1 granuting peptide) were reduced to 65000 Da. This indicates that a peptide core of Bungari.\ lIIulricinClLI.I' ~Bungarol in the K+-channel protein that binds to the toxins is ca. 65000 Da [58]. f3-Bungaro- toxin, normally considered to be a presyn- F7 isolated by Viljoen and Botes [66]. Table 3. Antiacetylcholinesterase- Type Neuro- aptic neurotoxin affecting the nerve end- Similarly, toxins C and D from D. polyle- toxins ing, is also a K+-channel blocker [59a, b]. pis polylepis venom are also related to There are considerable sequence homolo- acetylcholinesterase-type neurotoxin [67a, Venom Toxin gies between l3-bungarotoxin and den- b]. Although anticholinesterase neurotox- Dendrowpis aJIgu.\t;C('rS 7, asciculin droaspis-venom toxins. The K+-channel ins are structurally similar to postsynaptic inhibitory action of l3-btx is independent type neurotoxins and cardiotoxins, they D. polyl ph polylepis of its phospholipase A activity [60]. Alva- differ immunologically. rez and Garcia-Sancho [61], using crude The crystalline structure of fasciculin venoms of Notechis scutulatus, Oxyura- 2 indicates that the toxin is structurally nus scutulatus, and Vipera russelli, found related to both cardiotoxins and a-neuro- that they did inhibit K+channels. The first toxins. Fasciculin Iwas also examined by two venoms are known to contain potent X-ray crystallography [68]. Received: November 28, J997 presynaptic toxins. One should, however, The toxin binds to acetylcholinesterase notice that Alvarez and Garcia-Sancho and renders acetylcholine unhydrolyzed. [I] S. Rufini, J.2. Pedersen, A. Desideri, P. used K+ channels of red cells, whereas This causes continuous excitement of the Luly, Biochemistry 1990, 29, 9644. most other studies were done on synapto- muscle. The inhibition of acetylcholinest- [2] l.H. Tsai, H.C. Liu, T. Chang, Biochim. somes. Anderson and Harvey [62] used erase is seen not only in vitro, but also in Biophys. Acta 1987, 916, 94. vivo. For instance, 80% of the acetylcho- [3] KA Hendon, A.T. Tu, Biochim. Biophys. other tissues, such as diaphragm and the Acta 1979, 578, 243. nerve-muscle preparation, and observed linesterase activity in the locus coeruleus [4] A. Trivedi, 1.1. Kaiser, M. Tanaka, LL the same inhibition as in synaptosomes was inhibited by the injection of fasciculin Simpson, J. Pharmco/. Exp. Ther. 1989, studied by many other workers. 2 in rats [69]. The inhibition of the enzyme 25],490. by fasciculin is long-lasting, and a 74% [5] F. Radvanyi, C. Bon, FEBS Lett. 1989,247, inhibition five days after injection was 28. 4. Antiacetylcholinesterase observed [70]. By inhibiting acetylcho- [6] T.W. Jeng, KA. Hendon, H. Fraenkel-Con- rat, Proc. Natl. Acad. Sci. U.S.A. 1978, 75, Neurotoxins linesterase, fasciculin increased the am- 600. plitude and time course of the end-plate [7] C. Bon, J.P. Changeaux, T.W. Jeng, H. The fourth type of neurotoxin is the potential [71]. Fasciculin also increased Fraenkel-Conrat, Eur. J. Biochem. 1979, one that binds to acetylcholinesterase [63a, the amplitude of the miniature end-plate 99,471. b]. When acetylcholinesterase is not func- potential [64b]. Acetylcholinesterase en- [8] a) S.D. Aird, I.I. Kaiser, KV. Lewis, W.G. tioning, acetylcholine (after binding to the veloped in an artificialliposome can also Kruggel, Biochemistry 1985, 24, 7054; b) acetylcholine receptor) cannot be hydro- bind to fasciculin [72]. Because of the S.D. Aird, 1.I. Kaiser, R.V. Lewis, W.G. lyzed; consequently, normal nerve trans- inhibition of acetylcholinesterase, dendro- Kruggel, Arch. Biochem. Biophys. 1986, 249, 296; c) G. Faure, J.L Guillame, L. mission is impaired. Acetylcholinesterase toxins or other facilitatory toxins enhance Camoin, B. Saliou, C. Bon, Biochemistry action of D. Angusticeps venom was first the release of acetylcholine. Thus, den- 1991, 30, 8074. reported by Rodriguez-Ithurralde et al. drotoxins and fasciculins have synergistic [9] a) AL. Bieber, T. Tu, AT. Tu. Biochim. [64]. action that enhances the lethality. Fasci- Biophys. Acta 1975, 400, 178; b) R.L Cate, Fasciculin forms a complex with ace- culin 2 has no presynaptic action on trans- AL Bieber,Arch. Biochem. Biophys. 1978, tylcholinesterase, and the toxin in the com- mitter release or on postsynaptic receptor- 189,397. plex is very similar to the isolated toxin blocking action; the main action is on [10] a) S.D. Aird, B.L. Steadman. C.R. Mid- daugh, I.I. Kaiser, Biochim. Biophys. Acta structure [65]. anticholinesterase [73]. There was no sig- 1989,997,21 I; b) S.D. Aird, J.R. Yates, Antiacetylcholinesterase-type neuro- nificant change in dopamine or serotonin P.A. Martino, J. Shabanowitz, D.E Hunt, toxins have so far only been isolated from concentration in rats after fasciculin 2 I.I. Kaiser, ibid. ]990, 1040,217. African mambas (Denndroaspis). The injection [74]. [11] J.1. Valdes, KG. Thompson, V.L. Wolff, names of the snake venoms from which D.E. Menking, E.D. Rael, J.P. Chambers, anticholinesterase-type toxin was isolated This work was supported by NIH MERIT Neurotoxico!. Terata!. 1989, 11, 129. are shown in Table 3. Award 5R37GM15591. [12] S.D. Aird, 1.1. Kaiser, Toxicoll 1985, 23, 361. Anticholinesterase-type neurotoxin has [13] B.W. Lennon, 1.1.Kaiser, Compo Biochem. 57-60 amino acids in a single polypeptide Physio!. 1990, 97b, 695. chain, cross-linked by three disulfide [14] LS. Chang, c.c. Yang, J. Protein Chem. bonds. Fasciculin 2 is identical with toxin 1988,7,713. BIOLOGICALLY ACTIVE AGENTS IN NATURE 62

CHIMIA 52 (1998) Nr. 1/2 (Januar/Fcbruar)

[]5] P.O. Kwong, W.A. Hendrickson, P.B. Sig- [36] B.G. Stiles, F.W. Sexton, S.B. Guest, M.A. [61] 1.Alvarez,J. Garcia-Sancho, Biochim. Bio- ler, J. BioI. Chem. 1989,264, ]9349. Olson, D.C. Hack, Biochem. 1. 1994, 303, phys. Acta 1989, 890, 134. [16] J.-M. Danse, J.-M. Gamier, Nucleic Acids ]63. [62] AJ. Anderson, A.L. Harvey, Br. J. Phar- Res. 1990, 18, 1050. [37] H. Nomoto, Y. Nagaki, M. Ishikawa, H. macal. 1988,93,215. [17] E. Delot, c. Bon, J. Neurochem. 1992, 58, Shoji, K. Hayashi, J. Nat. Toxicol. 1992, 1, [63] a) D. Rodriguez-Ithurralde, R. Silveira, F. 311. 33. Dajas, Braz. J. Med. Sci. 1981, 14,394; b) [18] C.C. Yang, L.S. Chang, Biochem. J. 1991, [38] M.H. Fulachier, G. Mourier, J. Cotton, D. C. Cervenansky, F. Dajas, A.L. Harvey, E. 280,739. Servent, A Menez, FEBS Lett. 1994,338, Karlsson, in 'Snake Venoms', Ed. A.L. [19] a) M.I. Tyler, D. Barnett, P. Nicholson, I. 331. Harvey, Pergamon Press, New York, 1991, Spence, M.E.H. Howden, Biochim. Bio- [39] T.L. Lentz, Biochemistry 1995, 34, 1316. p.303-321. phys. Acta 1987, 915,210; b) J.A. Pearson, [40] S.V. Wheeler, S.D. Jane, K.M.L. Cross, [64] D. Rodriguez-Ithurralde, R. Silveira, L. M.I. Tyler, K.V. Retson, M.E. Howden, J.E. Chad, R.C. Foreman, J. Neurochem. Barbeito, F. Dajas, Neurochem. Int. 1983, ibid. 1991, 1077, ]47; c) J.A. Pearson, M.I. 1994,63,1891. 5,267. Tyler, K.V. Retson, M.E. Howden, ibid. [4]] V.I. Basus, G. Song, E. Hawrot, Biochem- [65] M. Harel, GJ. Kleywegt, R.B.C. Ravelli, I. 1993, 1161,223. istry 1993, 32, ]2290. Silman, J.L. Sussman, Structure (London) [20] S. Chwetzoff, P. Mollier, F. Bouet, E.G. [42] R.A. Miller, A.T. Tu, Arch. Biochem. Bio- 1995,3, 1355. Rowan, A.L. Harvey, A. Menez, FEBS phys.1991,291,69. [66] c.c. Viljoen, D.P. Botes, J. BioI. Chell!. Lett. 1990, 261, 226. [43] S.R. Lin, S.H. Chi, L.S. Chang, K.W. Kuo, 1973,248,4915. [21 ] B. Westerlund, P. Nordlund, U. Uhlin, D. C.C. Chang, J. Prot. Chem. 1996, 15,95. [67] a) F.J. Joubert, OJ. Strydom, Eur. J. Bio- Eaker, H. Ek]und, FEBS Lett. 1992,301, [44] S.R. Lin, c.c. Chang, Biochim. Biophys. chem. 1978,87,191; b) E.D. Karlsson, P. 159. Acta 1992, 1159, 255. Mbugua, D. Rodriguez-lthurralde, Phar- [22] K. Kondo, J. Zhang, K. Xu, H. Kagamiya- [45] F. Zamudio, K.M. Wolf, B.M. Martin, L.D. macol. Ther. 1984, 30, 259. rna, J. Biochem. 1989,105, 196. Possani, U.A. Chiappinelli, Biochemistry [68] R. Menez, A. Ducruix, J. Mol. Bioi. 1990, [23] A. Alape-Giron, B. Stiles, J. Schmidt, M. 1996,35,79]0. 216,233. Giron-Corates, M. Thelestam, H. Jornvall, [46] a) K.E. McLane, X. Wu, B.M. Conti-Tron- [69] V. Abo, L. Viera, R. Silveira, F. Dajas, T. Bergman, FEBS Lett. 1996,380,29. conti,J. BioI. Chem. 1990,265,9816; b) A. Neurosci. Lett. 1989, 98, 253. [24] S.T. Chu, Y.H. Chen, Biochem. J. 1991, Scheidler, P. Kaulen, G. BrUning, J. Erber, [70] J. Quillfeldt, S. Raskovsky, C. Dalmaz, M. 278,481. Brain Res. 1990, 534, 332. Dias, C. Huang, C.A. Netto, F. Schneider, [25] R.H. Loring, D. Andrews, W. Lane, R.E. [47] Z.W. Zhang, S. Vijayaraghavan, D.K. Berg, I. Izquierdo, J.H. Medina, R. Silveira, Phar- Zigmond, Brain Res. 1986, 385, 30. Neuron 1994, 12, 167. macal. Biochem. Behav. 1990, 37, 439. [26] a) C.-S. Liu, P.W. Hsiao, c.-S. Chang, [48] D.E. Hamassaki-Britto, A. Brzozowska- [71] c.Y. Lee, M.C. Tsai, M.-L. Tsaur, W.-W. M.C. Tzeng, T.B. Lo, J. Biochem. (Tokyo) Prechtl, HJ. Karten, J.M. Lindstrom, Visu- Line, F.H.H. Carlsson, F.I. Joubert, J. Phar- 1989,259,153; b) C.-S. Liu,J.-P. Chen, C.- 01 Neurosci. 1994, 11, 63. macal. Exp. Ther. 1985, 233, 491. S. Chang, T.B. Lo, J. Biochem. 1989,105, [49] S. Miyoshi, A.T. Tu, Arch. Biochem. Bio- [72] G. Puu, M. Koch, Biochem. Pharmacal. 93. phys. 1996,328,17. 1990, 40, 2209. [27] a) N. Mori, A.T. Tu, Arch. Biochem. Bio- [50J A. Adem, A. Asblom, G. Johansson, P.M. [73] AJ. Anderson, A.L. Harvey, P.M. Mbu- phys. 1988,260, 10; b) N. Mori, A.T. Tu, Mbugua, E. Karlsson, Biochim. Biophys. gua, Neurosci. Lett. 1985,54, 123. Bioi. Chem. Hoppe-Seyler 1988, 369, 521. Acta 1988, 968, 340. [74] B. Bolioli, M.E. Castello, D. Jerusalinsky, [28] c.c. Chang, Y. Kawata, F. Sakiyama, K. [51] a) U. Weller, U. Bernhardt, D. Siemen, F. M. Rubinstein, 1. Medina, F. Dajas, Brain Hayashi, Eur. J. Biochem. 1990, 193,567. Dreyer, W. Vogel, E. Habermann, Naun- Res. 1989, 504, I. [29J D.D. Sheumack, I. Spence, M.l. Tyler, yn-Schmiedeberg's Arch. Pharmacal. 1985, M.E.H. Howden, Camp. Biochem.Physiol. 330, 77; b) R.Penner, M. Peterson, F.K. 1990, 95b, 45. Pierau, F. Dreyer, Pfluegers Arch. 1986, [30] a) N. Yu, T. Lin, AT. Tu, J. BioI. Chem. 407, 365; c) A.L. Harvey, E. Karlsson, 1975,250, 1782; b) A.T. Tu, in 'Marine Naunyn-Schmiedeberg's Arch. Pharmacal. Toxins: Origin, Structure and Molecular 1980,312, I; d) A.R. Black, C.M. Done- Pharmacology', Eds. S. Hall and G. Strin- gan, B.J. Denny, J.O. Dolly, Biochemistry chartz, ACS Symp. Ser., 1990, Vol. 418, p. 1988,27,6814; e) E. Benoit, J.M. Dubois, 336; c) C. Betzel, G. Lange, G.-P. Pal, K.S. Brain Res. 1986,377,374; f) A.L. Harvey, Wilson, A. Maelicke, W. Saenger, J. BioI. Gen. Pharmacol. 1997,28,7. Chem. 1991,266,2]530; d) R. Le Goas, [52J E. Moczydlowski, K. Lucchesi, A. Ravin- S.R. La Plante, M.-A. Delsac, E. Guittet, dran, J. Membr. BioI. 1988,105,95. M. Robin, l. Charpentier, J.- Y. Lallemand, [53J C.E. Stansfeld, SJ. Marsh, J.V. Halliwell, Biochemistry 1992, 31,4867; e) C. Yu, C.- D.A. Brown, Neurosci. Lett. 1986,64,299. S. Lee, L.-c. Chuang, Y.-R. Shei, c.Y. [54] A.E. Busch, M.P. Kavanaugh, P.B. Os- Wang, Eur. J. Biochem. 1990,193,789; f) borne, R.A. North, J.P. Adelman, Mol. A.T. Tu, B.H. Jo, N. Yu, Int. 1. Pept. Pharmacol. 1991, 40, 572. Protein Res. 1976,8, 337. [55] c.G. Benishin, R. G. Sorensen, W.E. Brown, [31] D. Mebs, K. Narita, S. Iwanaga, Y. Same- B.K. Krueger, M.P. Blaustein, Mol. Phar- jima, c.Y. Lee, Biochem. Biophys. Res. macol. 1988, 34, 152. Commun. 1971, 44, 711. [56] AR. Black, A.L. Breeze, I.E. Othman, J.O. [32] M. Ohta, K. Ohta, H.Nishitani, K. Hayashi, Dolly, Biochem. J. 1986,237,397. FEBS Lett. 1987,222,79. [57] H. Rhem, M. Lazdunski, Proc. Natl. Acad. [33] R.S. McDowell, M.S. Dennis, A. Louie, M. Sci. U.S.A. 1988,85,4919. Shuster, M.G. Mulkerrin, R.A. Lazarus, [58] H. Rhem, FEBS Lett. 1989, 247, 28. Biochemistry 1992, 31,4766. [59] a) M. Peterson, R. Penner, F.K. Pierau, F. [34] P. Sunthornandh, P. Matangkasombut, K. Dreyer, Neurosci. Lett. 1986, 68, 141; b) Ratanabanangkoon, Mol. Immunol. 1992, R.R. Schmidt, H. Betz, H. Rehm, Biochem- 29,50 I. istry 1988, 27, 963. [35J L.S. Chang, K.W. Kuo, J. Lin, S.R. Lin, [60] E.G. Rowan, A.L. Harvey, Br. J. Pharma- c.c. Chang, 1. Biochem. 1995, 117, 863. col. 1988, 94, 839.