Toslrah Vol . 28, No . 9, pp. 1071-1076, 1990. Printed in Great Britain .
0041-0101/90 53.00+ .00 p) 1990 Pergamon Prcn pk
LACK OF THE BLOCKING EFFECT OF COBROTOXIN FROM NAJA NAJA ATRA VENOM ON NEUROMUSCULAR TRANSMISSION IN ISOLATED NERVE MUSCLE PREPARATIONS FROM POISONOUS AND NON-POISONOUS SNAKES YING-BING Llu" and KE Xu (KE Hsu)t Laboratory of Molecular Neurophatmacology, Shanghai Institute of Physiology, Chinese Academy of Sciences, Shanghai 200031, China (Acceptedforpublication 7 March 1990)
Y. B. LIU and K. Xu (K . Hsu) . Lack of the blocking effect of cobrotoxin from Naja raja atra venom on neuromuscular transmission in isolated nerve muscle preparations from poisonous and non-poisonous snakes . Toxicon 28, 1071-1076, 1990 .-Neuromuscular transmission in the Chinese cobra Naja naja atra was not affected by its own toxin (cobrotoxin) at a concentration of as high as 100 pM, while in the frog a concentration of less than 0.1 pM cobrotoxin depressed the amplitude of the end-plate potential to less than 10% of its original value within 30 min or directly blocked the nerve-evoked muscle action potential. The lack of effect of cobrotoxin was also observed in the nerve-muscle preparations from a pit viper (Agkistrodon blomho ~ti brevicaudus) and three species of non-poisonous snakes (Elaphe dione, Elaphe bimaculata and Elaphe rufodorsata) . On the other hand, the snake preparations were sensitive to the blocking effect of D-tubocurarine and less sensitive to that of atropine, indicating that the nicotinic acetylcholine receptor is responsible for transmission between the nerve and the skeletal muscle of the snake. We suggest that the snake nicotinic cholinergic receptor lacks the cobrotoxin binding-site . INTRODUCTION
SNAKE neurotoxins of the a-type, such as a-bungarotoxin (a-BuTX) and cobrotoxin bind selectively to acetylcholine receptors (AChRs) in the post functional membrane of vertebrate skeletal muscles. It has, however, been observed that snakes themselves are frequently resistant to the toxic action of snake venoms . The blood of certain snakes contains factors that inhibit or neutralize the lethal toxicity of the venoms (PHn.POT and SMITH, 1950 ; DEORAS and MHASALKAR, 1963 ; BONNETT and GUTTMAIV, 1971 ; OVADIA et al., 1976; OVADIA, 1978). Some authors have noted that the resistance of a snake to snake venoms could not be explained fully by humoral factors found in blood (OVADIA and 'Present address: Dept of Physiology, West China University of Medical Sciences, Chengdu 610041, China . tTo whom correspondence should be addressed . 1071
1072
Y.-H. LIU and KE XU (KE HSU)
KOCHVA, 1977 ; PEREZ et al., 1978 ; MIxTON and MINTON, 1981) . BURDEN et al. (1975) in their study of the phylogenetic differences of AChRs reported that miniature end-plate potentials (mEPPs) recorded from three species of non-poisonous snakes (Epicrates cenchris mauris, Elaphe obsoleta obsoleta, Thamnophis sauritus) were not influenced by aatra toxin, and two of them (Elaphe obsoleta obsoleta, Thamnophis sauritus) were not blocked by a-btmgarotoxin, indicating the involvement of a factor other than humoral. The existence of mEPPs does not always mean that the neuromuscular process is not influenced . For example, it has been reported that both the amplitude and frequency of mEPPs remains unchanged, when neuromuscular transmission is completely blocked by cobalt (SHAPOVALOV, 1962) or nickel ions (TANG and Hsu, 1964) . In this present manuscript the blocking action of cobrotoxin from the venom of cobra Naja naja atra and some other agents on neuromuscular transmission in this and several other species of snakes was studied. It was established that AChRs at the neuromuscular junctions of all the tested species of snakes were highly resistant to cobrotoxin and at the same time were sensitive to the effect of D-tubocurarine and less sensitive to atropine .
Preparations
MATERIALS AND METHODS
Two species of poisonous snakes, Naja raja atra from family Elapidae and Agkistrodon blomhoffü brevicaudus (old name A . halys) from family Crotalidae, were collected in Anhui and Zhejiang Provinces, China respectively . Three species of non-poisonous snakes, Elaphe diane and Elaphe bimaculata and Elaphe rufodorsata from family Colubridae, were collected in the suburbs of Shanghai, where no N. n . atra was distributed. The snakes were kept in the laboratory for a few days and given access to water only . The obliquas externes abdominis nerve-muscle preparation was made from the snakes according to method of Ku~r IFR and Yostn~ (1975) . The preparations comprise only two to three layers of muscle fibers. Frogs (Rana tigerina regulars) were wllected in the suburbs of Shanghai . The cutaneous pectoris nerve-muscle preparation was made from the frog according to the method of Lt:ntvsxY et al. (1976) . Preparations were mounted in an acrylic chamber filled with the snake or frog solution . All experiments were performed at room temperature (18-20°C). Electrical recordings
A glass micrcelectrode filled with 3 M KCI (10-20 ML2 tip resistance) was inserted into an end-plate region of a muscle fiber for recording the end-plate potential (EPP) or into a muscle fiber for recording the intracellular action potential. The reference electrode was a piece of Ag-AgCI filament set on the bottom of the chamber. A suction electrode was used to stimulate the nerve trunk or the muscle fiber. The resting potential of both snake and frog muscle fibers was monitored by using PZ, DC Digital Voltage Meter (Shanghai Voltage Meter Factory) . The EPP and action potential were measured by using a micrcelectrode amplifier (MEZ-7101, Nihon-Kohden, Tokyo) and an oscilloscope (Tektronix 5112, Heaverton, Oregon). Solutions
Amphibian solution contained (mM) 117 .2 NaCI, 2 .5 KCI, 1.8 CaCI 2 .16 Na zHPO 0 .85 NaH,PO, (pH 7.1). The composition of the snake solution was as reported by KUFFI.ER and Yosttttuan (1975) (mM) : 158 NaCI, 2.15 KCI, 1.8 CaCI 1 .7 MgCI,. The solution was buffered with 1 mM phosphates (pH 7 .0). For EPP recording, the concentration of Mg'* was adjusted so that the amplitude of the EPPs was between 5-12 mV . The Mg'* concentrations used for the nerve-muscle preparation of each species of animals were (mM) : frog 12 .4, wbra 14.2, A . b . brevicaudus 9.7, E. thane and others 13 .8 . Toxin and chemical agents
Cobrotoxin was isolated and purified from the venom of the Chinese cobra Naja naja atra collected in Anhui Province, China. It was kindly provided by Z. Y. Zeou from Shanghai Institute of Biochemistry, Chinese Academy of Sciences . It is a short chain a-toxin. The mol. wt was 7000, as determined by electrophoresis on sodium dodecyl sulfate polyacrylamide gel. The N-terminal amino acid residue was leucine as reported by YANG
Cobrotoxin on Snake Neuromuscular Transmission
107 3
Sme FIG. l . EFFecr OF COHROTOXIN
THE END-PLATE P(71ENTIAL IN FROG (liana tigerina rugdlOSa) ANll SNAKE (Naja naja atra) . (a) End-plate potential recorded from frog cutaneous pectoris nerve-muscle preparation in the presence of 12 .4 mM Mg", and (b) end-plate potential recorded from cobra obliquas externus abdominis nerve-muscle preparation in the presence of 14.2 mM Mg" . Upper traces are controls, lower trace in 1(a) shows the record obtained 30 min after application of cobrotoxin at a concentration of 0.08 pM and that in l(b) obtained 2 hr after application of cobrotoxin at a concentration of 25 pM . ON
(1969) . D-Tubocurarine hydrochloride (DTC) was obtained from Charles Drug Ltd (London, England) . Atropine sulfate was from The Shanghai 10th Pharmaceutical Manufactory . et a!.
RESULTS
E,~ect of cobrotoxirr on end-plate potential The blocking potency of cobrotoxin on the EPP was tested in the cutaneous pectoris nerve-muscle preparation of the frog. At a concentration of 0.08-0.1 pM (n = 4), cobrotoxin reduced the amplitude of the EPP to less than 10% of its original value within 30 min (Fig. la). The blocking potency was consistent with that reported by LEE and CHANG (1966) . The effect of cobrotoxin on EPPs generated in cobra muscle fiber was examined at a cobrotoxin concentration range of 0.1-100 pM. In no case (n = 5) was the amplitude or the duration of the EPPs different from those observed in the absence of cobrotoxin . An example is shown in Fig. lb. For quantitatively collecting EPP data five functional regions in a nerve-muscle preparation ofcobra were chosen and 10 penetrations from each region were made before and after application for more than 2 hr of cobrotoxin at a concentration of 25-100 pM. The mean value of the EPP amplitudes obtained before and after application of cobrotoxin was calculated to be 8.0 t 1 .3 (mean ± S. E.) and 8.4 ±2.7 mV respectively . The difference between the two groups was not significant (Student's t-test, n = 8, P > 0.05). In none of the experiments in the frog and cobra preparations was there a change in the latent period between the stimulus artefact and start of the EPP (Fig. 1). We conclude that neither propagation of the prejunctional impulse nor synaptic delay were changed significantly . The resting membrane potential of cobra muscle fibers was also not affected by cobrotoxin at a high concentration (100 pM, 2 hr). The experimental results clearly showed that cobrotoxin had no effect on EPP generation of cobra neuromuscular junctions at concentration 1000 x that effective at blocking frog neuromuscular transmission . The blocking potency of cobrotoxin on the EPP of other poisonous (A . b. brevicautlus) and non-poisonous (E. dione, E. bimaculata and E. rufodorsata) snakes were examined. When the nerve-muscle preparations from these snakes were treated with cobrotoxin (25-50 pM) for more than 2 hr, the EPPs were also unaffected. The effect of cobrotoxin at higher concentration was not tested .
1074
Y.-B. LIU
and KE XU (KE HSU)
FIG. 2. EFFEC'T'S OF COBROTOXIN ON NEUROMUSCULAR TRANSMISSION . Intracellular action potential of a muscle fiber in the frog cutaneous pectoris nerve-muscle preparation evoked by nerve stimulation, and (b) intracellular action potential of a muscle fiber in cobra obliquas externus abdominis nerve-muscle preparation evoked by nerve stimulation. Upper traces are controls. Middle trace in 2(a) shows the record obtained 30 min after application of cobrotoxin at a concentration of 0.1 pM and that in 2(b) obtained 2 hr after application of cobrotoxin at a concentration of 25 uM . Lower traces are the records obtained by the direct stimulation of the muscle after the middle records were taken. (a)
Effect of cobrotoxin on the indirectly evoked muscle action potential
The lowest concentration of cobrotoxin which blocked the action potential of muscle fibers evoked by nerve stimulation in the frog was 0.1 pM (Fig. 2a). However, the indirectly evoked action potential of muscle fibers of all the five species of snakes was not affected, when their nerve-muscle preparation was treated with cobrotoxin for more than 2 hr at concentrations as high as 25 pM (Fig . 2b). Effect of n-tubocurarine (nTC) and atropine
Although the neuromuscular junction of skeletal muscles of the snake is cholinergic and YOSHIKAMI, 1975), our observations show that neuromuscular transmission is not influenced by cobrotoxin, a specific antagonist of the AChR. It was therefore necessary to test the reaction of the snake neuromuscular junction to the classical cholinergic antagonist, nTC. Transmission in nerve-muscle preparations isolated from all species tested (N. n. atra, A. b. brevicaudus and E. dione) were sensitive to nTC, an antagonist of the nicotinic AChR, although the concentration required in the snakes for blockade was about 10-fold higher than that for the frog. The lowest concentrations of nTC needed to reduce the amplitude of the EPP to less than 10% of its original value within 20 min in the frog and in the snakes were 0.9 pM (n = 5) and 14.0 pM (n = 7) respectively . On the other hand, it is well known that atropine, an antagonist of muscarinic AChR, also blocks the function of nicotinic AChRs, but at higher concentrations (WEINER, 1980). Neuromuscular transmission in snake and frog preparations were equally blocked by atropine within 30 min at a concentration of 140 pM. (KUFFLER
DISCUSSION
The blocking action of postsynaptically active a-neurotoxins has been demonstrated in nerve-muscle preparations isolated from all vertebrates including fish, amphibians, birds
Cobrotoxin on Snake Neuromuscular Transmission
1075
and mammals. The only exception to this general rule was doctunented by BURDEN et al. (1975), who demonstrated that mEPPs recorded from the end-plates of several species of snakes and lizards were resistant to the toxins . We have now shown that neuromuscular transmission in both poisonous and non-poisonous snake preparations was not blocked by cobrotoxin at a concentration as high as 100 pM, while the same toxin at a concentration of less than 0.1 pM was enough to block transmission completely within 30 min in frog preparations . The high resistance of the neuromuscular junction of the snakes certainly is not due to the poor access of the toxin into the synaptic cleft or to the involvement of blood humoral factors. indeed, the preparations are very thin, the connective tissue covering the thin muscle was carefully stripped off and the preparations were washed with the fresh snake solution several times for more than 2 hr before the experiments were performed. T'he neuromuscular transmission of the snakes is 10 times more sensitive to the blocking effect of DTC than to atropine, indicating that the related AChR is of the nicotinic type . Consequently, the binding site at the nicotinic AChR of the snake for ACh and DTC is probably similar to that of other vertebrates . Thus we infer that the nicotinic AChR at the postfunctional membrane of the snake muscle lacks the cobrotoxin binding-site . The nicotinic AChR is comprised of four different subunits assembled in a transmembrane pentamer azßy8. The protein carries two ACh binding sites at the level of the «subunits and contains the ion channel . Since the two a-subunits are not adjacent, the cooperative interactions between binding sites are thus indirect or allosteric . Morevoer, these two sites do not exhibit perfectly symmetrical properties and differ, in particular, in their kinetic parameters of interaction with the snake a-toxins (cf. GY-tANGEUX et al ., 1984). The snake nicotinic AChR is thus a useful preparation for studying the structure-function relationship of the nicotinic AChR-binding site . Animals are often resistant to the neurotoxins they produce or store within their body . Nerve fibers of the puffer fish and the newt are very resistant to tetrodotoxin which occurs in these animals' tissues (KAO and FUHRMAN, 1967). The sciatic nerve and sartorius muscle membranes of the frog Phyllobates aurotaenia are resistant to the depolarizing effects of its own poison, batrachotoxin, whereas the corresponding nerve and muscle membranes of Rana pipiers were quite sensitive to the poison (ALBUQuERQuE et al ., 1973). TERAKAWA et al. (1989) reported that the Na channel in the nerve membrane of the scorpion Buthus martensi Karsch is resistant to its own toxin (BmK I) . In the case of the snakes it is interesting that the nicotinic AChRs at the end-plates of not only the poisonous, but also the non-poisonous snakes are resistant to the a-toxins of the snakes . This suggests that in the evolution of snakes the development of nicotinic AChR's insensitive to a-neurotoxin appeared earlier than the a-neurotoxin molecules itself. Acknowledgement-We thank Ms W. P.
WANG
for her technical support and assistance with the manuscript . REFERENCES
E. X., W~xrnctc, J. F., $AIVSONE, F. M. and DULY, J. (1973) The pharmacology of batrachotoxin . V. A comparative study of membrane properties and the effect of batrachotoxin on sartorius muscles of the frogs Phyllobates aurotaenia and Rana pipiens. J. Pltarmacol. exp. Ther . 184, 315-329. BoxxE-rr, D. E. and GUTTMAN, S. I. (1971) Inhibition of moccasin (Agkistrodon piscivorus) venom proteolytic activity by the serum of the Florida king snake (Lampropeltis getulus.floridana). Toxicon 9, 41725. Btrxn~v, S. J., HARTZELL, H. C. and Y06HIKAMI, D. (1975) Acetylcholine receptors at neuromuscular synapses : phylogenetic differences detected by snake neurotoxins . Proc . natn . Acad. Sci. U.S .A . 72, 3245-3249.
A~euQtlExQUE,
1076
Y .-B . LIU and KE XU (KE HSU)
C~uNOeux, J.-P., DEVILLIeR3-TFiIERY, A . and CHEMOUILLI, P . (1984) Acetylcholine receptor : an allosteric protein . Science, N .Y . 225, 1335-1345 . Deox~s, P . J, and Ma~t.tun, V. B . (1963) Antivenin activity of some snake sera . Toxicon 1, 89--90 . Kwo, C. Y . and FUHRMAN, F . A . (1967) Differentiation of the actions of tetrodotoxin and saxitoxin. Toxicon 5, 25-34. Ktri~.eß, S . W . and Yosfntu~, D . (1975) The distribution of acetylcholine sensitivity at the post-synaptic membrane of vertebrate skeletal twitch muscles: iontophoretic mapping in the micron range . J. Physiol., Lond. 244, 703-730. Lee, C . Y . and C~uxc, C . C . (1966) Modes of actions of purified toxins from elapid venoms on neuromuscular transmission. Mem . Inst . Butantan 33, 555-572 . LsrnvsKV, M . S ., F .cK, K . H . and McM~utv, U . J. (1976) Precision of reinnervation of original postsynaptic sites in frog muscle after a nerve crush . J. Neurocytol. 5, 691-718 . Mtrnvtv, S . A. and Mn~rrort, M . R . (1981) Toxicity of some Australian snake venoms for potential prey species of reptiles and amphibians . Toxicon 19, 749-755 . Oveou, M. (1978) Purification and characterization of an antihemorrhagic factor from the serum of the snake Vipera palaestinae. Toxicon 16, 661-b72 . OVADIA, M . and Kocxvn, E . (1977) Neutralization of Viperidae and Elapidae snake venoms by sera of difl'erent animals . Toxicon 15, 541-547 . Ovnuu, M ., Mow, B . and KOCHV~, E . (1976) Factors in the blood serum of Vipera palaestinae neutralizing toxic fractions of its venom . In : Animal, Plant and Microbial Toxins, Vol .l, pp . 137-142 (OH3AKA, A ., HAYASHI, K . and SAWAI, Y ., Ed .) . New York: Plenum Press . Pexez, J. C ., Hews, W. C. and HATGN, C. H . (1978) Resistance of woodrats (Neotoma micropus) to Crotalus atrox venom . Toxicon 16, 198-200 . PwLror, V . B . and SMITH, R . G . (1950) Neutralization of pit viper venom by king snake serum . Proc . Soc. exp . Biol. 74, 521-523 . SHAPOVALOV, A . I . (1962) The inRuence of some substances inhibiting neuromuscular transmission on the endplate micropotentials . Cytology 4, 66973 (in Russian) . TeN, T. P. and Hsu, K. (1964) The effect of nickel on neuromuscular transmission . Acta Physiol. Sin . 27, 5-10 (in Chinese with an abstract in English) . TERAKAWA, 5 ., KIMURA, Y ., Hsu, K . and JI, Y. H. (1989) Lack of effect of a neurotoxin from the swrpion Buthus martensi Karsch on nerve fibers of this scorpion. Toxicon 27, 569-578 . Welxex, N . (1980) Atropine, scopolamine and related antimuscarinic drugs . In: The Pharmacological basis of therapeutics (6th edn), pp . 120-137 (GOODMAN, L. S . and GILMAN, A ., Ed .) . New York : The Macmillan Publishing Company . YANG, C . C ., CSiANO, C . C ., HAYASHI, K . and SuzuKl, T . (1969) Amino acid composition and end group analysis of cobrotoxin . Toxicon 7, 437 .
0041-0101/90 53.00+ .00 p) 1990 Pergamon Prcn pk
LACK OF THE BLOCKING EFFECT OF COBROTOXIN FROM NAJA NAJA ATRA VENOM ON NEUROMUSCULAR TRANSMISSION IN ISOLATED NERVE MUSCLE PREPARATIONS FROM POISONOUS AND NON-POISONOUS SNAKES YING-BING Llu" and KE Xu (KE Hsu)t Laboratory of Molecular Neurophatmacology, Shanghai Institute of Physiology, Chinese Academy of Sciences, Shanghai 200031, China (Acceptedforpublication 7 March 1990)
Y. B. LIU and K. Xu (K . Hsu) . Lack of the blocking effect of cobrotoxin from Naja raja atra venom on neuromuscular transmission in isolated nerve muscle preparations from poisonous and non-poisonous snakes . Toxicon 28, 1071-1076, 1990 .-Neuromuscular transmission in the Chinese cobra Naja naja atra was not affected by its own toxin (cobrotoxin) at a concentration of as high as 100 pM, while in the frog a concentration of less than 0.1 pM cobrotoxin depressed the amplitude of the end-plate potential to less than 10% of its original value within 30 min or directly blocked the nerve-evoked muscle action potential. The lack of effect of cobrotoxin was also observed in the nerve-muscle preparations from a pit viper (Agkistrodon blomho ~ti brevicaudus) and three species of non-poisonous snakes (Elaphe dione, Elaphe bimaculata and Elaphe rufodorsata) . On the other hand, the snake preparations were sensitive to the blocking effect of D-tubocurarine and less sensitive to that of atropine, indicating that the nicotinic acetylcholine receptor is responsible for transmission between the nerve and the skeletal muscle of the snake. We suggest that the snake nicotinic cholinergic receptor lacks the cobrotoxin binding-site . INTRODUCTION
SNAKE neurotoxins of the a-type, such as a-bungarotoxin (a-BuTX) and cobrotoxin bind selectively to acetylcholine receptors (AChRs) in the post functional membrane of vertebrate skeletal muscles. It has, however, been observed that snakes themselves are frequently resistant to the toxic action of snake venoms . The blood of certain snakes contains factors that inhibit or neutralize the lethal toxicity of the venoms (PHn.POT and SMITH, 1950 ; DEORAS and MHASALKAR, 1963 ; BONNETT and GUTTMAIV, 1971 ; OVADIA et al., 1976; OVADIA, 1978). Some authors have noted that the resistance of a snake to snake venoms could not be explained fully by humoral factors found in blood (OVADIA and 'Present address: Dept of Physiology, West China University of Medical Sciences, Chengdu 610041, China . tTo whom correspondence should be addressed . 1071
1072
Y.-H. LIU and KE XU (KE HSU)
KOCHVA, 1977 ; PEREZ et al., 1978 ; MIxTON and MINTON, 1981) . BURDEN et al. (1975) in their study of the phylogenetic differences of AChRs reported that miniature end-plate potentials (mEPPs) recorded from three species of non-poisonous snakes (Epicrates cenchris mauris, Elaphe obsoleta obsoleta, Thamnophis sauritus) were not influenced by aatra toxin, and two of them (Elaphe obsoleta obsoleta, Thamnophis sauritus) were not blocked by a-btmgarotoxin, indicating the involvement of a factor other than humoral. The existence of mEPPs does not always mean that the neuromuscular process is not influenced . For example, it has been reported that both the amplitude and frequency of mEPPs remains unchanged, when neuromuscular transmission is completely blocked by cobalt (SHAPOVALOV, 1962) or nickel ions (TANG and Hsu, 1964) . In this present manuscript the blocking action of cobrotoxin from the venom of cobra Naja naja atra and some other agents on neuromuscular transmission in this and several other species of snakes was studied. It was established that AChRs at the neuromuscular junctions of all the tested species of snakes were highly resistant to cobrotoxin and at the same time were sensitive to the effect of D-tubocurarine and less sensitive to atropine .
Preparations
MATERIALS AND METHODS
Two species of poisonous snakes, Naja raja atra from family Elapidae and Agkistrodon blomhoffü brevicaudus (old name A . halys) from family Crotalidae, were collected in Anhui and Zhejiang Provinces, China respectively . Three species of non-poisonous snakes, Elaphe diane and Elaphe bimaculata and Elaphe rufodorsata from family Colubridae, were collected in the suburbs of Shanghai, where no N. n . atra was distributed. The snakes were kept in the laboratory for a few days and given access to water only . The obliquas externes abdominis nerve-muscle preparation was made from the snakes according to method of Ku~r IFR and Yostn~ (1975) . The preparations comprise only two to three layers of muscle fibers. Frogs (Rana tigerina regulars) were wllected in the suburbs of Shanghai . The cutaneous pectoris nerve-muscle preparation was made from the frog according to the method of Lt:ntvsxY et al. (1976) . Preparations were mounted in an acrylic chamber filled with the snake or frog solution . All experiments were performed at room temperature (18-20°C). Electrical recordings
A glass micrcelectrode filled with 3 M KCI (10-20 ML2 tip resistance) was inserted into an end-plate region of a muscle fiber for recording the end-plate potential (EPP) or into a muscle fiber for recording the intracellular action potential. The reference electrode was a piece of Ag-AgCI filament set on the bottom of the chamber. A suction electrode was used to stimulate the nerve trunk or the muscle fiber. The resting potential of both snake and frog muscle fibers was monitored by using PZ, DC Digital Voltage Meter (Shanghai Voltage Meter Factory) . The EPP and action potential were measured by using a micrcelectrode amplifier (MEZ-7101, Nihon-Kohden, Tokyo) and an oscilloscope (Tektronix 5112, Heaverton, Oregon). Solutions
Amphibian solution contained (mM) 117 .2 NaCI, 2 .5 KCI, 1.8 CaCI 2 .16 Na zHPO 0 .85 NaH,PO, (pH 7.1). The composition of the snake solution was as reported by KUFFI.ER and Yosttttuan (1975) (mM) : 158 NaCI, 2.15 KCI, 1.8 CaCI 1 .7 MgCI,. The solution was buffered with 1 mM phosphates (pH 7 .0). For EPP recording, the concentration of Mg'* was adjusted so that the amplitude of the EPPs was between 5-12 mV . The Mg'* concentrations used for the nerve-muscle preparation of each species of animals were (mM) : frog 12 .4, wbra 14.2, A . b . brevicaudus 9.7, E. thane and others 13 .8 . Toxin and chemical agents
Cobrotoxin was isolated and purified from the venom of the Chinese cobra Naja naja atra collected in Anhui Province, China. It was kindly provided by Z. Y. Zeou from Shanghai Institute of Biochemistry, Chinese Academy of Sciences . It is a short chain a-toxin. The mol. wt was 7000, as determined by electrophoresis on sodium dodecyl sulfate polyacrylamide gel. The N-terminal amino acid residue was leucine as reported by YANG
Cobrotoxin on Snake Neuromuscular Transmission
107 3
Sme FIG. l . EFFecr OF COHROTOXIN
THE END-PLATE P(71ENTIAL IN FROG (liana tigerina rugdlOSa) ANll SNAKE (Naja naja atra) . (a) End-plate potential recorded from frog cutaneous pectoris nerve-muscle preparation in the presence of 12 .4 mM Mg", and (b) end-plate potential recorded from cobra obliquas externus abdominis nerve-muscle preparation in the presence of 14.2 mM Mg" . Upper traces are controls, lower trace in 1(a) shows the record obtained 30 min after application of cobrotoxin at a concentration of 0.08 pM and that in l(b) obtained 2 hr after application of cobrotoxin at a concentration of 25 pM . ON
(1969) . D-Tubocurarine hydrochloride (DTC) was obtained from Charles Drug Ltd (London, England) . Atropine sulfate was from The Shanghai 10th Pharmaceutical Manufactory . et a!.
RESULTS
E,~ect of cobrotoxirr on end-plate potential The blocking potency of cobrotoxin on the EPP was tested in the cutaneous pectoris nerve-muscle preparation of the frog. At a concentration of 0.08-0.1 pM (n = 4), cobrotoxin reduced the amplitude of the EPP to less than 10% of its original value within 30 min (Fig. la). The blocking potency was consistent with that reported by LEE and CHANG (1966) . The effect of cobrotoxin on EPPs generated in cobra muscle fiber was examined at a cobrotoxin concentration range of 0.1-100 pM. In no case (n = 5) was the amplitude or the duration of the EPPs different from those observed in the absence of cobrotoxin . An example is shown in Fig. lb. For quantitatively collecting EPP data five functional regions in a nerve-muscle preparation ofcobra were chosen and 10 penetrations from each region were made before and after application for more than 2 hr of cobrotoxin at a concentration of 25-100 pM. The mean value of the EPP amplitudes obtained before and after application of cobrotoxin was calculated to be 8.0 t 1 .3 (mean ± S. E.) and 8.4 ±2.7 mV respectively . The difference between the two groups was not significant (Student's t-test, n = 8, P > 0.05). In none of the experiments in the frog and cobra preparations was there a change in the latent period between the stimulus artefact and start of the EPP (Fig. 1). We conclude that neither propagation of the prejunctional impulse nor synaptic delay were changed significantly . The resting membrane potential of cobra muscle fibers was also not affected by cobrotoxin at a high concentration (100 pM, 2 hr). The experimental results clearly showed that cobrotoxin had no effect on EPP generation of cobra neuromuscular junctions at concentration 1000 x that effective at blocking frog neuromuscular transmission . The blocking potency of cobrotoxin on the EPP of other poisonous (A . b. brevicautlus) and non-poisonous (E. dione, E. bimaculata and E. rufodorsata) snakes were examined. When the nerve-muscle preparations from these snakes were treated with cobrotoxin (25-50 pM) for more than 2 hr, the EPPs were also unaffected. The effect of cobrotoxin at higher concentration was not tested .
1074
Y.-B. LIU
and KE XU (KE HSU)
FIG. 2. EFFEC'T'S OF COBROTOXIN ON NEUROMUSCULAR TRANSMISSION . Intracellular action potential of a muscle fiber in the frog cutaneous pectoris nerve-muscle preparation evoked by nerve stimulation, and (b) intracellular action potential of a muscle fiber in cobra obliquas externus abdominis nerve-muscle preparation evoked by nerve stimulation. Upper traces are controls. Middle trace in 2(a) shows the record obtained 30 min after application of cobrotoxin at a concentration of 0.1 pM and that in 2(b) obtained 2 hr after application of cobrotoxin at a concentration of 25 uM . Lower traces are the records obtained by the direct stimulation of the muscle after the middle records were taken. (a)
Effect of cobrotoxin on the indirectly evoked muscle action potential
The lowest concentration of cobrotoxin which blocked the action potential of muscle fibers evoked by nerve stimulation in the frog was 0.1 pM (Fig. 2a). However, the indirectly evoked action potential of muscle fibers of all the five species of snakes was not affected, when their nerve-muscle preparation was treated with cobrotoxin for more than 2 hr at concentrations as high as 25 pM (Fig . 2b). Effect of n-tubocurarine (nTC) and atropine
Although the neuromuscular junction of skeletal muscles of the snake is cholinergic and YOSHIKAMI, 1975), our observations show that neuromuscular transmission is not influenced by cobrotoxin, a specific antagonist of the AChR. It was therefore necessary to test the reaction of the snake neuromuscular junction to the classical cholinergic antagonist, nTC. Transmission in nerve-muscle preparations isolated from all species tested (N. n. atra, A. b. brevicaudus and E. dione) were sensitive to nTC, an antagonist of the nicotinic AChR, although the concentration required in the snakes for blockade was about 10-fold higher than that for the frog. The lowest concentrations of nTC needed to reduce the amplitude of the EPP to less than 10% of its original value within 20 min in the frog and in the snakes were 0.9 pM (n = 5) and 14.0 pM (n = 7) respectively . On the other hand, it is well known that atropine, an antagonist of muscarinic AChR, also blocks the function of nicotinic AChRs, but at higher concentrations (WEINER, 1980). Neuromuscular transmission in snake and frog preparations were equally blocked by atropine within 30 min at a concentration of 140 pM. (KUFFLER
DISCUSSION
The blocking action of postsynaptically active a-neurotoxins has been demonstrated in nerve-muscle preparations isolated from all vertebrates including fish, amphibians, birds
Cobrotoxin on Snake Neuromuscular Transmission
1075
and mammals. The only exception to this general rule was doctunented by BURDEN et al. (1975), who demonstrated that mEPPs recorded from the end-plates of several species of snakes and lizards were resistant to the toxins . We have now shown that neuromuscular transmission in both poisonous and non-poisonous snake preparations was not blocked by cobrotoxin at a concentration as high as 100 pM, while the same toxin at a concentration of less than 0.1 pM was enough to block transmission completely within 30 min in frog preparations . The high resistance of the neuromuscular junction of the snakes certainly is not due to the poor access of the toxin into the synaptic cleft or to the involvement of blood humoral factors. indeed, the preparations are very thin, the connective tissue covering the thin muscle was carefully stripped off and the preparations were washed with the fresh snake solution several times for more than 2 hr before the experiments were performed. T'he neuromuscular transmission of the snakes is 10 times more sensitive to the blocking effect of DTC than to atropine, indicating that the related AChR is of the nicotinic type . Consequently, the binding site at the nicotinic AChR of the snake for ACh and DTC is probably similar to that of other vertebrates . Thus we infer that the nicotinic AChR at the postfunctional membrane of the snake muscle lacks the cobrotoxin binding-site . The nicotinic AChR is comprised of four different subunits assembled in a transmembrane pentamer azßy8. The protein carries two ACh binding sites at the level of the «subunits and contains the ion channel . Since the two a-subunits are not adjacent, the cooperative interactions between binding sites are thus indirect or allosteric . Morevoer, these two sites do not exhibit perfectly symmetrical properties and differ, in particular, in their kinetic parameters of interaction with the snake a-toxins (cf. GY-tANGEUX et al ., 1984). The snake nicotinic AChR is thus a useful preparation for studying the structure-function relationship of the nicotinic AChR-binding site . Animals are often resistant to the neurotoxins they produce or store within their body . Nerve fibers of the puffer fish and the newt are very resistant to tetrodotoxin which occurs in these animals' tissues (KAO and FUHRMAN, 1967). The sciatic nerve and sartorius muscle membranes of the frog Phyllobates aurotaenia are resistant to the depolarizing effects of its own poison, batrachotoxin, whereas the corresponding nerve and muscle membranes of Rana pipiers were quite sensitive to the poison (ALBUQuERQuE et al ., 1973). TERAKAWA et al. (1989) reported that the Na channel in the nerve membrane of the scorpion Buthus martensi Karsch is resistant to its own toxin (BmK I) . In the case of the snakes it is interesting that the nicotinic AChRs at the end-plates of not only the poisonous, but also the non-poisonous snakes are resistant to the a-toxins of the snakes . This suggests that in the evolution of snakes the development of nicotinic AChR's insensitive to a-neurotoxin appeared earlier than the a-neurotoxin molecules itself. Acknowledgement-We thank Ms W. P.
WANG
for her technical support and assistance with the manuscript . REFERENCES
E. X., W~xrnctc, J. F., $AIVSONE, F. M. and DULY, J. (1973) The pharmacology of batrachotoxin . V. A comparative study of membrane properties and the effect of batrachotoxin on sartorius muscles of the frogs Phyllobates aurotaenia and Rana pipiens. J. Pltarmacol. exp. Ther . 184, 315-329. BoxxE-rr, D. E. and GUTTMAN, S. I. (1971) Inhibition of moccasin (Agkistrodon piscivorus) venom proteolytic activity by the serum of the Florida king snake (Lampropeltis getulus.floridana). Toxicon 9, 41725. Btrxn~v, S. J., HARTZELL, H. C. and Y06HIKAMI, D. (1975) Acetylcholine receptors at neuromuscular synapses : phylogenetic differences detected by snake neurotoxins . Proc . natn . Acad. Sci. U.S .A . 72, 3245-3249.
A~euQtlExQUE,
1076
Y .-B . LIU and KE XU (KE HSU)
C~uNOeux, J.-P., DEVILLIeR3-TFiIERY, A . and CHEMOUILLI, P . (1984) Acetylcholine receptor : an allosteric protein . Science, N .Y . 225, 1335-1345 . Deox~s, P . J, and Ma~t.tun, V. B . (1963) Antivenin activity of some snake sera . Toxicon 1, 89--90 . Kwo, C. Y . and FUHRMAN, F . A . (1967) Differentiation of the actions of tetrodotoxin and saxitoxin. Toxicon 5, 25-34. Ktri~.eß, S . W . and Yosfntu~, D . (1975) The distribution of acetylcholine sensitivity at the post-synaptic membrane of vertebrate skeletal twitch muscles: iontophoretic mapping in the micron range . J. Physiol., Lond. 244, 703-730. Lee, C . Y . and C~uxc, C . C . (1966) Modes of actions of purified toxins from elapid venoms on neuromuscular transmission. Mem . Inst . Butantan 33, 555-572 . LsrnvsKV, M . S ., F .cK, K . H . and McM~utv, U . J. (1976) Precision of reinnervation of original postsynaptic sites in frog muscle after a nerve crush . J. Neurocytol. 5, 691-718 . Mtrnvtv, S . A. and Mn~rrort, M . R . (1981) Toxicity of some Australian snake venoms for potential prey species of reptiles and amphibians . Toxicon 19, 749-755 . Oveou, M. (1978) Purification and characterization of an antihemorrhagic factor from the serum of the snake Vipera palaestinae. Toxicon 16, 661-b72 . OVADIA, M . and Kocxvn, E . (1977) Neutralization of Viperidae and Elapidae snake venoms by sera of difl'erent animals . Toxicon 15, 541-547 . Ovnuu, M ., Mow, B . and KOCHV~, E . (1976) Factors in the blood serum of Vipera palaestinae neutralizing toxic fractions of its venom . In : Animal, Plant and Microbial Toxins, Vol .l, pp . 137-142 (OH3AKA, A ., HAYASHI, K . and SAWAI, Y ., Ed .) . New York: Plenum Press . Pexez, J. C ., Hews, W. C. and HATGN, C. H . (1978) Resistance of woodrats (Neotoma micropus) to Crotalus atrox venom . Toxicon 16, 198-200 . PwLror, V . B . and SMITH, R . G . (1950) Neutralization of pit viper venom by king snake serum . Proc . Soc. exp . Biol. 74, 521-523 . SHAPOVALOV, A . I . (1962) The inRuence of some substances inhibiting neuromuscular transmission on the endplate micropotentials . Cytology 4, 66973 (in Russian) . TeN, T. P. and Hsu, K. (1964) The effect of nickel on neuromuscular transmission . Acta Physiol. Sin . 27, 5-10 (in Chinese with an abstract in English) . TERAKAWA, 5 ., KIMURA, Y ., Hsu, K . and JI, Y. H. (1989) Lack of effect of a neurotoxin from the swrpion Buthus martensi Karsch on nerve fibers of this scorpion. Toxicon 27, 569-578 . Welxex, N . (1980) Atropine, scopolamine and related antimuscarinic drugs . In: The Pharmacological basis of therapeutics (6th edn), pp . 120-137 (GOODMAN, L. S . and GILMAN, A ., Ed .) . New York : The Macmillan Publishing Company . YANG, C . C ., CSiANO, C . C ., HAYASHI, K . and SuzuKl, T . (1969) Amino acid composition and end group analysis of cobrotoxin . Toxicon 7, 437 .
