- Email: [email protected]
Antagonistic action of uranyl nitrate on presynaptic neurotoxins from snake venoms
Nruroppharmacology Vol. 25, No. 1, pp. 95-101, Printed in Great Britain. Ail rights reserved
1986 Copyright
0
0028-3908/86 $3.00 + 0.00 1986 Pergamon Press Ltd
ANTAGONISTIC ACTION OF URANYL NITRATE ON PRESYNAPTIC NEUROTOXINS FROM SNAKE VENOMS S. Y. IAN-SHIAU and W. M. Fu Pharmacological Institute, College of Medicine, National Taiwan University, Taipei, Taiwan, Republic of China
(Accepted 12 March 1985) Sununary-Uranyl ion (UOi+) antagonized the neuromuscular blocking action and phospholipase A, activity of neurotoxins which act presynaptically [/I-bungarotoxin (fi-BuTX) and crotoxin] but did not affect the action of a-bungarotoxin and tetrodotoxin. On the basis of the kinetic analysis of the UO$+ and strontium ion (Sr2+) antagonism of muscle paralysis induced by j?-bungarotoxin, it was found that they inhibited both the binding of the toxin and the steps following binding that brought about the neuromuscular blocking action of ~-bungarotoxin. Uranyl ion was about 50 times more potent than Sr2+ in antagonizing ~-bungarotoxin. High Ca2+ (10 mM) abolished but low Ca’+ (0.25-1.25 mM) medium enhanced the antagonizing action of UOI’ and Sr2+. In low Ca2+ medium, UO:+ markedly potentiated the amplitude of the twitch, subsequent addition of /J-bungarotoxin produced three phases of effects on the twitches, e.g. an initial depression, followed by the second facilitation and finally a rapid depression of twitches; however, approx. 70 min after /I-bungarotoxin the small twitches reached a steady state which persisted for more than 350min. Therefore, it is evident that UO:+ is the most potent antagonist of /?-bungarotoxin so far tested. Key words: many1 nitrate, neurotoxins.
A variety of polypeptide neurotoxins are found in snake venoms. In addition to cr-type toxins, which act postsynaptically by binding to acetylcholine (ACh) receptors (Lee, 1972), another type of neurotoxin acts presynaptically to inhibit the evoked release of ACh from motor nerve terminals. These presynaptically acting toxins, such as ~-bungarotoxin (‘BuTX; from the Formosan krait), notexin (from the Australian tiger snake), taipoxin (from the Australian taipan) and crotoxin (from the South American rattlesnake) exhibit complex actions on cell membrane in addition to possessing phospholipase A, (PLAJ activity (Howard and Gundersen, 1980). The divalent cation, strontium ion (Sr’+), was found not only to inhibit phospholipase A2 activity but also to antagonize the presynaptic blockade caused by these toxins (Strong, Goerke, Oberg and Kelly, 1976; Chang, Su, Lee and Eaker, 1977). As described in previous papers, it was found that uranyl nitrate was very effective in antagonizing the in vivo toxicity of Bungurus multicinctus venom in chicks and mice (LinShiau, 1983; Lin-Shiau, Chen and Fu, 1983). In the present study, the antagonizing effect of many1 nitrate against the neuromuscular blockade induced by several neurotoxins was compared with that of Sr”+ in order to shed more light on the mechanism of action of uranyl nitrate. METHODS
Mouse phrenic nerve -diaphragm preparation
The mouse phrenic nerve-diaphragm preparation was isolated according to the method of Biilbring NP 2511-6
(1946). Modified Krebs’ solution, gassed with 95% O2 + 5% CO, at 37 + 0.5”C, was used throughout the experiment; the composition of the solution was as follows: 130.6mM NaCl, 4.8 mM KCl, 2.5 mM CaCl,, 1.2 mM MgSO,, 12.5 mM NaHCO, and il. 1 mM glucose. The phrenic nerve was stimulated with supramaximal rectangular pulses of 0.05 msec at a rate of 12/min. The contraction was recorded isometrically with a force-displacement transducer (Grass FT. 03) on a Grass Model 7 polygraph. Measurement neurotoxins
of
phospholipase
A,
activity
of
Purified egg lecithin was used as the substrate and the hydroIytic activity of neurotoxins was measured by the method described by Strong et al. (1976). Egg lecithin was emulsified with equimolar Na deoxycholate in the presence or absence of 0.05 mM EDTA and 10mM CaCl,. The hydrolytic activity of neurotoxins was determined by measuring the amount of 2mM NaOH required to titrate the fatty acid liberated from egg lecithin. The reaction was performed at 37°C and the titration endpoint was set at pH 7.4. Toxins a- and P-Bungarotoxins were isolated from the venom of Bungurus mukicinctus by the method described by Lee, Chang, Kau and Luh (1972). The homogeneity of the purified toxins was verified by disc gel electrophoresis (David, 1964). Crotoxin was a gift from Professor H. FraenkelConrat (University of California). Tetrodotoxin, 9.5
S. Y.
96
LIN-SHIAU
and W. M. Fu
(A)
o
to.4
22
t0
mM
pM
.
..I
p-
BuTX
-2_-----_ *.;,”
.
.
.
.
.
P-BuTX
22
PM
uo;+
.
.
l
.
10 20 +I
mM
to.22
40 pM
*,*.... 50
65
03
100
125
l
145 155
180
1
l
200220
P-BUTX
l
l
225
235
.
290
min
tw
Sr2+
Fig. 1. Effect of uranyl ion &JO;+) and strontium ion (Sr’+) j5bungarotoxin in mouse diaphragm. The phrenic nerve of mouse and the contraction of the diaphragm was recorded isometrically or in the presence of UOz+ (B) or Sr’+ (C), respectively. W denotes
purified three times by crystallization, from Sankyo Co. Ltd, Japan.
was obtained
Statistics Toxins were added to the organ bath with or without pretreatment of uranyl ion (nitrate salt) or strontium ion (chloride salt). The number of experiments in the control and tested groups are more than four as indicated in the results. The significance of difference between the control and treated groups was tested by Student’s r-test.
on neuromuscular blocking action of diaphragm was electrically stimulated in the absence of divalent cations (A) washout with modified Krebs’ solution.
concentration-dependent (Fig. 2). The concentrations of UO$+ and Sr2+ required for a 2-fold prolongation of the blocking time of /I-bungarotoxin were 0.2 and 10 mM, respectively (n = 4). Chang and Lee (1963) have shown that sufficient and irreversible binding of b-bungarotoxin takes place within 20 min and, therefore, the slow onset of the effect is not due to a slow binding of the toxin to
RESULTS
Effects of uranyl ion (lJO<+) and strontium ion (Sr”) on neuromuscular blocking action qf /I-bungarotoxin /I-Bungarotoxin (0.22 PM) completely and irreversibly blocked in 129.9 k 5.7 min (n = 7) the twitch responses to nerve stimulation (Fig. IA). UO<+ (0.4mM), potentiated the amplitude of the twitch, and upon subsequent application of /I-bungarotoxin, it markedly antagonized its action by prolonging the blocking time to more than 400 min (Fig. I B). Strontium ions (9 mM) prolonged the blocking time of /?-bungarotoxin to 199.0 k 14.0 min; small twitches (11% of control amplitude) appeared on subsequent washout; these twitches disappeared in 12.5 k I .5 min (Fig. IC). Antagonism of the neuromuscular blocking action of /I-bungarotoxin by UOf+ and Sr2+ was
+//’
005
’ 015
Cation
I
I
I
,
1
3
10
30
concentration
(mM)
Fig. 2. Concentration-dependent antagonism by uranyl ion (UOz+; 0) and strontium ion (Sr’+; a) on the neuromuscular blocking action of /I-bungarotoxin in mouse diaphragm. (0) Represents the time to neuromuscular block (N-M block) induced by 0.22nM /j’-bungarotoxin alone (n = 4).
Uranyl Table
I. Effect
of time course of application
nitrate
97
and neurotoxins
of uranyl ion (UO:*) and strontium ion (Sr”) (/j-BuTX) in mouse diaphragm
on their antagonism
against fl-bungarotoxin
Time to neuromuscular Incubation fl-BuTX
Time sequence of ion application
with
No washout Washout after 20 min incubation Washout after 20 min incubation
Added by the Added Added Added
Washout after 20 min incubation No washout No washout
simultaneously wth [GBuTX followed washout after 20min Incubation after washout of fl-BuTX simultaneously with /SBuTX 20min orior to I(-BuTX
uo:+
Sr’+
129.9 k 5.7 (7) 133.2 k 6.5 (4) 207.2 k 18.6 (4);
129.9 + 5.7 133.2 + 6.5 195.7 * II.1 (4)’
238.0 k 31.3 (6)* > 350 (3)* > 350 (4)*
274.0 k 26.2 (6)’ 31 1.3 k 7.6(3)* 314.0 k 3.8 (4)’
Final concentrations of /?-bungarotoxin, UO:+ and Sr’+ was 0.22 PM, 0.4 mM and I5 mM, respectively. respective control in the absence of divalent cation.
.E E
350 -
‘: 0 z I i z
250-
h 4-t-4 !-‘-_J
\ \\
‘\
‘\X\\
“p\\,,,
*\1’ -\
150
‘4.
5!
:
,o E G
block (min)
*P i 0.05 as compared
with the
antagonistic effect. When 0.4mM UO:+ was added 50min after /j’-bungarotoxin, there was no antagonistic action, but when the concentration was increased to 0.8 mM, it exerted again a significant antagonistic effect. Even when the twitch amplitude was reduced by b-bungarotoxin to about 209/o of control value, 0.8 mM concentration of IJ@+ still exerted antagonism against the neuromuscular blocking action of fi-bungarotoxin.
50 i!
0
’ 20
Time
after
’ 40
’ 60
B - BuTX
’ 80
’ 100
( rn~n1
Fig. 3. Effect of uranyl ion (UOz+), strontium ion (S?‘) on the neuromuscular blocking action of /I-bungarotoxin when they were applied at different time intervals after p-bungarotoxin in the mouse phrenic nerve-diaphragm preparation. UO:+ (0.4, 0 or 0.8 mM n ) Sr’+ (15 mM, A) was applied, respectively at a time interval indicated after B-bungarotoxin, and the time to complete neuromuscular block (N-M block) induced by /j’-bungarotoxin alone (0.22pM, 0) or fl-bungarotoxin in the presence of UO:+ or Sr*+ was recorded.
the target site, but must be due to slow changes that occur after binding. To test whether the antagonistic effect of UOs+ and Sr*+ was due to an inhibition of the binding of the toxin or to a retardation of the changes occurring after binding, UOi+ or Sr2+ was added simultaneously with /3-bungarotoxin for 20 min and then washed repeatedly with Krebs’ solution. Paralysis developed in these diaphragms at a slower rate than in those similarly exposed to /?-bungarotoxin alone without UO:+ or Sr2+ (Table l), indicating that less fl-bungarotoxin was bound. In another experiment the diaphragms were incubated with /3-bungarotoxin for 20min and then washed repeatedly with Krebs’ solution, UOs+ or Sr2+ was added after washout, these procedures also showed that UO: + and Sr2+ prolonged the development of neuromuscular blockade (Table Since 1). fi-bungarotoxin had ample time to bind to its site of action, it follows that the change of release mechanism occurring after binding with /?-bungarotoxin was also retarded. Both ions also revealed antagonistic action even if they were applied after longer periods of incubation with b-bungarotoxin (Fig. 3). The longer the incubation time, the weaker was the
Reversal by Ca’+ of UOg+ or Sr2+ antagonism /I-bungarotoxin
of
When /j’-bungarotoxin was applied in a low Ca*+ Krebs solution the neuromuscular blocking time of b-bungarotoxin was prolonged; the block by /l-bungarotoxin was not affected in Ca*+-rich (10 mM) Krebs solution. The antagonistic effect of Sr2+ against the block induced by /3-bungarotoxin was inversely proportional to the Ca*+ concentration in the medium. At concentrations of 3-10 mM, Ca*+ decreased and eventually abolished the antagonistic effect of Sr*+ (Fig. 4A). By contrast, reducing the concentration of Ca2+ (0252.5 mM) enhanced that of Sr2+. Increase of the concentration of Ca2+ also abolished the antagonistic effect of UOz+ on the block by fi-bungarotoxin (Fig. 4A). In low Ca*+ medium, UOi+ prolonged the time required for complete neuromuscular blockade (100% paralysis) by b-bungarotoxin to more than 350 min; however, based on the time required for 60% paralysis. low Ca*+ decreased the effect UOs+ (Figs 4B and 5). For instance, at a concentration of 0.7 mM Ca2+, UOi+ markedly potentiated the amplitude of the twitch; after 20min, /I-bungarotoxin was added and the three phases of the effects of p-bungarotoxin on the amplitude of the twitch were produced, e.g. the initial depression followed by the second increase and finally a rapid decrease (Fig. 5B). However, the decrease reached a steady state 70min after the application of b-bungarotoxin, and the small twitches remaining no longer decreased, which persisted for more than 350 min (Figs 4A and 5B). Effect of lJO$’ and Sr” activity of /I-bungarotoxin
on the phospholipase
A,
The phospholipase A, (PLA,) activity of /?-bungarotoxin was estimated to be 203.7 f 6.9 pmol fatty
98
S. Y. LIN-SHIAU and W. M. Fu 0 Control A uo2+04 2
0
(A)
Sr”9
mM mM
350 L-_L___~
r
7
50 t
OL
’
0.3
I
I
1
3
L
I
,
03
10
Coz+co”ce”trat~on
I
I
I
1
3
10
( mM I
Fig. 4. Effect of Ca2+ on the antagonistic action of uranyl ion (UO:+) or strontium ion (Sr2+) against the neuromuscular blocking action of fi-bungarotoxin (0.22pM) in the mouse diaphragm. Time to paralysis in (A) indicated 100% paralysis whereas that in (B) indicated 60% paralysis.
acid released/mg b-bungarotoxin/min. Uranyl ions and Sr*+ inhibited the enzyme activity of p-bungarotoxin in a concentration-dependent manner (Fig. 6); 0.25 mM UO$+ and 10 mM of Sr*+, which prolonged the neuromuscular blocking time of /I-bungarotoxin approx. to 2-fold (Fig. 2) inhibited the phospholipase A, activity of /I-bungarotoxin by 63.5 t_ 4.9 and 65.2 + 1.9%, respectively. Effects
of UOi+ on the actions of other neurotoxins
Crotoxin (0.23 PM), a-bungarotoxin (0.3 PM) and tetrodotoxin (0.02 PM) blocked neuromuscular transmission in 101.9 * 8.8, 42.6 f 3.8 and 78.9 k 7.9 min (n = 4-lo), respectively. Uranyl ions antagonized the action of crotoxin in a concentration-dependent manner (Fig. 7; n = 4). However, UOi+ had no effect on the neuromuscular blocking action of r-bungarotoxin and tetrodotoxin (Fig. 7). In the presence of 2.5 mg/ml phosphatidylcholine, the phospholipase A, activity of crotoxin was estimated to be 41.3 f 3.9 pmol fatty acid released/mg crotoxin per min. Uranyl ions (0.5 mM) inhibited the enzyme activity of crotoxin by 31.0 + 3.0% (n = 5). DISCUSSION
It has been reported that strontium ion (Sr*+) not only inhibits the phospholipase A, activity but also antagonizes the presynaptic blockade caused by several neurotoxins isolated from snake venoms (Strong et al., 1976; Chang et al., 1977; Howard and Gundersen, 1980). In the present study, it was found that the inhibitory effect of uranyl ion (UOi+) on either the activity of phospholipase A, or the
neuromuscular blocking action of fi-bungarotoxin was much more potent than that of ST’+. So far as is known, UOi+ is the most potent inhibitor of P-bungarotoxin. It has been demonstrated that sufficient and irreversible binding of /3-bungarotoxin takes place within 20 min, and neuromuscular blockade developed more slowly (Chang and Lee, 1963). Uranyl ions and Sr2+ antagonized /I-bungarotoxin during both phases of the action of p-bungarotoxin: (1) when they were incubated with /I-bungarotoxin during the first 20min, followed by the washout; and (2) when they were added after 20-100 min of incubation with fi-bungarotoxin, followed by the washout of the toxin. These results suggest that UO:+ and Sr2+ not only inhibited the binding of /I-bungarotoxin but also inhibited the subsequent steps leading to the neuromuscular blockade by b-bungarotoxin. /?-Bungarotoxin has been reported to be able to penetrate the axon membrane and enter into the nerve terminals (Strong et al., 1976). On the other hand, its phospholipase A, activity was more marked when /I-bungarotoxin was injected intracellularly into the axon (Hinzer and Taut, 1977; Narahasi and Tobias, 1964). In the present study, it was found that UO:+ antagonized the action of b-bungarotoxin when it was applied at different incubating time intervals after P-bungarotoxin even when 80% paralysis was caused by P-bungarotoxin. Two possibilities seem to be indicated by these results. First, the effect of P-bungarotoxin may be due to its action on the outside of the membrane of nerve terminals; second, /I-bungarotoxin may increase the permeability of the presynaptic membrane so that UO:+ which normally is incapable of penetrating the membranes may enter
Uranyl
nitrate
and neurotoxins
S. Y. LIN-SHIAU and W. M. Fu
100 100
c 0
-
.
60-
2
4
; 60c
/
H 40E 8 ; 20a 0’
I
I
0 01
10
01
1
Concentrotlon
1mM 1
25
0) or strontium ion (Sr2+; A) on phospholipase A2 activity of /3-bungarotoxin. The specific activity of /?-bungarotoxin (control) was estimated to be 23.7 + 6.9 pmol/mg per min. Fig. 6. Inhibition
by uranyl ion (UO:+;
into nerve terminals and inhibit the actions of /?-bungarotoxin. Masukawa and Livengood (1982) reported that treatment with /I-bungarotoxin increased the permeability to cations such as Co2+, Mg2+, Mn2+ and Ni’+; that normally cannot enter the membranes, these cations greatly increased the miniature endplate potential (m.e.p.p.) frequency after treatment with /?-bungarotoxin. Uranyl ion is able, at very small concentrations, to change the surface potential of charged phospholipid membranes (McLaughline, Szabo and Eisenman, 1971); this indicates that UOi+ has high affinity to phospholipids. In previous studies, it was found that UOi+ could interact directly with phospholipids as it decreased the fluorescence intensity of the probe 8-anilino-1-naphthalene sulphonate (Lin-Shiau, Fu and Mao-Chen, 1984). It is proposed therefore, that UOi+ binds to phospholipid components of the nerve terminal membrane to form a metal ion-phospholipid complex. The formation of the
UO~+concentrar~on
Fig. 7. Doseeresponse
(mM)
curves of uranyl ion (UOz+) on the neuromuscular blocking action of neurotoxins (tetrodotoxin 0.02 PM, 0, 0; a-bungarotoxin 0.3 PM, 0, n ; crotoxin 0.23 PM, & A) in the mouse phrenic nerve diaphragm preparation. The open and closed symbols represent the time to neuromuscular block (N-M block) induced by these neurotoxins in the absence or in the presence of UO:+, respectively.
complex may alter the conformation of the structure of the membrane so that the binding of /?-bungarotoxin is inhibited. Furthermore, the complex formed is not, a suitable substrate for P-bungarotoxin, thus leading to inhibition of phospholipase A, activity and then inhibition of the neuromuscular blocking action of P-bungarotoxin. Calcium ions are essential for the release of transmitter (Bennett, Florin and Hall, 1975). Calcium ions bind to the membrane phospholipids and neutralize the negative charges on the membrane, enhancing release of transmitter by facilitating the fusion of synaptic vesicles with the nerve terminal membrane (Portis, Newon, Pangborn and Papahadjopaulos, 1979). In the present study, it was found that high levels of Ca2+ effectively reversed while low Ca2+ (0.25-l mM) augmented the antagonistic action of UOi+ and Sr2+. It is suspected that Ca2+ exerts the inhibitory action on the antagonism by UO:+ and Sr*+ possibly by competing for binding to membrane phospholipids. In a previous experiment, it was demonstrated that UOi+ potentiated the amplitude of the twitch by the release of ACh from motor nerve terminals (LinShiau and Fu, 1979). This potentiating property of UOi+ was enhanced in a low Ca2+ medium. It is interesting that in a low Ca2+ medium, plus UO:+, the addition of /?-bungarotoxin exhibited three phases of effects, e.g. initial depression of the twitches, followed by increased amplitude of the twitches and finally a rapid decrease of the amplitude of the twitch which reached a steady state approx. 70 min after fl-bungarotoxin and persisted for more than 350 min. Spokes and Dolly (1980) claimed that the first depression was due to the specific binding of fl-bungarotoxin to the neural membrane; the second facilitation may be due to the increased release of acetylcholine and the final neuromuscular blockade is due to the inhibition of the release of ACh. The findings of the present study indicated that UO:+ enhanced the second facilitatory phase, possibly owing to increased release of ACh by UOi’ (LinShiau and Fu, 1979). Crotoxin is a presynaptically active neurotoxin containing phospholipase A, activity (Breithaput, Riibsamen and Habermann, 1971; Hendon and Fraenkel-Conrat, 1971). That UOi+ inhibits both the activity of phospholipase A, and the neuromuscular blocking action of this toxin further supports the effective antagonism by UOi+ on the presynaptic neurotoxins. Uranyl ions do not affect the actions of a-bungarotoxin and tetrodotoxin, indicating that the action of UO:+ is fairly specific. It is apparent that the mode of action of the presynaptic neurotoxin is more complex than the postsynaptic neurotoxin which simply occupies ACh receptors of the endplate. In order to shed some light on the action of presynaptic neurotoxin on the release mechanism of ACh, a potent antagonist will certainly be helpful. In this study, it has been shown that UOz+
Uranyl nitrate and neurotoxins antagonized the action of preferentially j?-bungarotoxin and crotoxin but did not affect that of a-bungarotoxin and tetrodotoxin.
REFERENCES Bennett M. R., Florin T. and Hall R. (1975) The effect of calcium ion on the binominal statistical parameters which control acetylcholine release at synapses in striated muscle. .I. Physiol., Land. 241: 429446. Breithaput H., Riibsamen K. and Habermann E. (1971) In vitro and in viva interactions between phosphohpase A and novel potentiator isolated from so-called crotoxin. Naunyn-Schmiedebergs Arch. Pharmac. 269: 403404. Bulbring E:. (1946) Observations on the isolated phrenic nerve diaphragm preparation of the rat. Br. J. Pharmac. 1: 38-61. Chang C. C. and Lee C. Y. (1963) Isolation of neurotoxins from the venom of Bungarus multicinctus and their modes of neuromuscular blocking action. Archs int. Pharmacodyn. Ther. 144: 241-257. Chang C. C., Su M. J., Lee J. D. and Eaker D. (1977) Effects A and the of Sr*+ and Mg’+ on the phosphohpase neuromuscular blocking actions of presynaptic and taipoxin. Naunyn crotoxin fi-bungarotoxin, Schmiedebergs Arch. Pharmac. 299: 1555161. David B. J. (1964) Disc electrophoresis II. Method and application to human serum proteins. Ann. N.Y. Acad. Sci. 121: 404. Hendon R. A. and Fraenkel-Conrat H. (1971) Biological roles of the two components of crotoxin. Proc. nafn. Acad. Sci. U.S.A. 68: 156&1563. Hinzen D. H. and Taut L. (1977) Membrane properties of aplysia neurons intracellularly injected with phosphohpase A and C. J. Physiol., Lond. 268: 21-34. Howard B. D. and Gundersen C. B. (1980) Effects and mechanisms of polypeptide neurotoxins that act presynaptically. A. Reu. Pharmac. Toxic. 20: 37-336. Lee C. Y. (1972) Chemistry and pharmacology of poly-
101
peptide toxins in snake venoms. A. Rev. Pharmac. Toxic. 12: 265-286. Lee C. Y., Chang S. L., Kau S. T. and Luh S. Y. (1972) Chromatographic separation of the venom of Bungarus multicinctus and characterization of its components. J. Chromat. 72: 71-81. Lin-Shiau S. Y. (1983) Selective antagonism by uranyl nitrate against Formosan Krait venom. Part I. J. Formosan med. Ass. 82: 640-643. Lin-Shiau S. Y. and Fu W. M. (1979) Effects of uranyl ions on neuromuscular transmission of chick biventer cervicis muscle. Archs int. Pharmacodyn. Ther. 241: 332-343. Lin-Shiau S. Y., Chen C. C. and Fu W. M. (1983) Selective antagonism by uranyl nitrate against Formosan Krait venom. Part II. J. Formosan med. Ass. 82: 1099-1103. Lin-Shiau S. Y., Fu W. M. and Mao-Chen C. T. (1984) Selective antagonism by uranyl nitrate against Formosan Krait venom. Part III. Mechanism of action of uranyl nitrate. J. Formosan med. Ass. 83: 21-26. Masukawa L. M. and Livengood D. R. (1982) Alterations in spontaneous transmitter release by divalent cations after treatment of the neuromuscular junction with p-bungarotoxin. Cell. molec. Neurobiol. 2: 277-290. McLaughlin S. G. A., Szabo G. and Eisenman G. (1971) Divalent ions and the surface potential of charged phospholipid membrane. J. gen. Physiol. 58: 667-687. Narahashi T. and Tobias J. M. (1964) Properties of axon membrane as affected by cobra venom, digitonin and protease. Am. J. Physiol. 207: 1441-1446. Portis A., Newon C., Pangborn W. and Papahadjopauios D. (1979) Studies on the mechanism of membrane fusion: Evidence for an intermembrane Ca2+-phosphohpid complex, synergism with Mg2+ and inhibition by spectrin. Biochemistry 18: 780-790. Spokes J. W. and Dolly J. 0. (1980) Complete purification of /I-bungarotoxin, characterization of its action and that of tityustoxin on synaptosomal accumulation and release of acetylcholine. Biochim. biophys. Acta 596: 81-93. Strong P. N., Goerke J., Oberg S. G. and Kelly R. B. (1976) /I-Bungarotoxin, a presynaptic toxin with enzymatic activity. Proc. natn. Acad. Sci. U.S.A. 73: 175-182.
1986 Copyright
0
0028-3908/86 $3.00 + 0.00 1986 Pergamon Press Ltd
ANTAGONISTIC ACTION OF URANYL NITRATE ON PRESYNAPTIC NEUROTOXINS FROM SNAKE VENOMS S. Y. IAN-SHIAU and W. M. Fu Pharmacological Institute, College of Medicine, National Taiwan University, Taipei, Taiwan, Republic of China
(Accepted 12 March 1985) Sununary-Uranyl ion (UOi+) antagonized the neuromuscular blocking action and phospholipase A, activity of neurotoxins which act presynaptically [/I-bungarotoxin (fi-BuTX) and crotoxin] but did not affect the action of a-bungarotoxin and tetrodotoxin. On the basis of the kinetic analysis of the UO$+ and strontium ion (Sr2+) antagonism of muscle paralysis induced by j?-bungarotoxin, it was found that they inhibited both the binding of the toxin and the steps following binding that brought about the neuromuscular blocking action of ~-bungarotoxin. Uranyl ion was about 50 times more potent than Sr2+ in antagonizing ~-bungarotoxin. High Ca2+ (10 mM) abolished but low Ca’+ (0.25-1.25 mM) medium enhanced the antagonizing action of UOI’ and Sr2+. In low Ca2+ medium, UO:+ markedly potentiated the amplitude of the twitch, subsequent addition of /J-bungarotoxin produced three phases of effects on the twitches, e.g. an initial depression, followed by the second facilitation and finally a rapid depression of twitches; however, approx. 70 min after /I-bungarotoxin the small twitches reached a steady state which persisted for more than 350min. Therefore, it is evident that UO:+ is the most potent antagonist of /?-bungarotoxin so far tested. Key words: many1 nitrate, neurotoxins.
A variety of polypeptide neurotoxins are found in snake venoms. In addition to cr-type toxins, which act postsynaptically by binding to acetylcholine (ACh) receptors (Lee, 1972), another type of neurotoxin acts presynaptically to inhibit the evoked release of ACh from motor nerve terminals. These presynaptically acting toxins, such as ~-bungarotoxin (‘BuTX; from the Formosan krait), notexin (from the Australian tiger snake), taipoxin (from the Australian taipan) and crotoxin (from the South American rattlesnake) exhibit complex actions on cell membrane in addition to possessing phospholipase A, (PLAJ activity (Howard and Gundersen, 1980). The divalent cation, strontium ion (Sr’+), was found not only to inhibit phospholipase A2 activity but also to antagonize the presynaptic blockade caused by these toxins (Strong, Goerke, Oberg and Kelly, 1976; Chang, Su, Lee and Eaker, 1977). As described in previous papers, it was found that uranyl nitrate was very effective in antagonizing the in vivo toxicity of Bungurus multicinctus venom in chicks and mice (LinShiau, 1983; Lin-Shiau, Chen and Fu, 1983). In the present study, the antagonizing effect of many1 nitrate against the neuromuscular blockade induced by several neurotoxins was compared with that of Sr”+ in order to shed more light on the mechanism of action of uranyl nitrate. METHODS
Mouse phrenic nerve -diaphragm preparation
The mouse phrenic nerve-diaphragm preparation was isolated according to the method of Biilbring NP 2511-6
(1946). Modified Krebs’ solution, gassed with 95% O2 + 5% CO, at 37 + 0.5”C, was used throughout the experiment; the composition of the solution was as follows: 130.6mM NaCl, 4.8 mM KCl, 2.5 mM CaCl,, 1.2 mM MgSO,, 12.5 mM NaHCO, and il. 1 mM glucose. The phrenic nerve was stimulated with supramaximal rectangular pulses of 0.05 msec at a rate of 12/min. The contraction was recorded isometrically with a force-displacement transducer (Grass FT. 03) on a Grass Model 7 polygraph. Measurement neurotoxins
of
phospholipase
A,
activity
of
Purified egg lecithin was used as the substrate and the hydroIytic activity of neurotoxins was measured by the method described by Strong et al. (1976). Egg lecithin was emulsified with equimolar Na deoxycholate in the presence or absence of 0.05 mM EDTA and 10mM CaCl,. The hydrolytic activity of neurotoxins was determined by measuring the amount of 2mM NaOH required to titrate the fatty acid liberated from egg lecithin. The reaction was performed at 37°C and the titration endpoint was set at pH 7.4. Toxins a- and P-Bungarotoxins were isolated from the venom of Bungurus mukicinctus by the method described by Lee, Chang, Kau and Luh (1972). The homogeneity of the purified toxins was verified by disc gel electrophoresis (David, 1964). Crotoxin was a gift from Professor H. FraenkelConrat (University of California). Tetrodotoxin, 9.5
S. Y.
96
LIN-SHIAU
and W. M. Fu
(A)
o
to.4
22
t0
mM
pM
.
..I
p-
BuTX
-2_-----_ *.;,”
.
.
.
.
.
P-BuTX
22
PM
uo;+
.
.
l
.
10 20 +I
mM
to.22
40 pM
*,*.... 50
65
03
100
125
l
145 155
180
1
l
200220
P-BUTX
l
l
225
235
.
290
min
tw
Sr2+
Fig. 1. Effect of uranyl ion &JO;+) and strontium ion (Sr’+) j5bungarotoxin in mouse diaphragm. The phrenic nerve of mouse and the contraction of the diaphragm was recorded isometrically or in the presence of UOz+ (B) or Sr’+ (C), respectively. W denotes
purified three times by crystallization, from Sankyo Co. Ltd, Japan.
was obtained
Statistics Toxins were added to the organ bath with or without pretreatment of uranyl ion (nitrate salt) or strontium ion (chloride salt). The number of experiments in the control and tested groups are more than four as indicated in the results. The significance of difference between the control and treated groups was tested by Student’s r-test.
on neuromuscular blocking action of diaphragm was electrically stimulated in the absence of divalent cations (A) washout with modified Krebs’ solution.
concentration-dependent (Fig. 2). The concentrations of UO$+ and Sr2+ required for a 2-fold prolongation of the blocking time of /I-bungarotoxin were 0.2 and 10 mM, respectively (n = 4). Chang and Lee (1963) have shown that sufficient and irreversible binding of b-bungarotoxin takes place within 20 min and, therefore, the slow onset of the effect is not due to a slow binding of the toxin to
RESULTS
Effects of uranyl ion (lJO<+) and strontium ion (Sr”) on neuromuscular blocking action qf /I-bungarotoxin /I-Bungarotoxin (0.22 PM) completely and irreversibly blocked in 129.9 k 5.7 min (n = 7) the twitch responses to nerve stimulation (Fig. IA). UO<+ (0.4mM), potentiated the amplitude of the twitch, and upon subsequent application of /I-bungarotoxin, it markedly antagonized its action by prolonging the blocking time to more than 400 min (Fig. I B). Strontium ions (9 mM) prolonged the blocking time of /?-bungarotoxin to 199.0 k 14.0 min; small twitches (11% of control amplitude) appeared on subsequent washout; these twitches disappeared in 12.5 k I .5 min (Fig. IC). Antagonism of the neuromuscular blocking action of /I-bungarotoxin by UOf+ and Sr2+ was
+//’
005
’ 015
Cation
I
I
I
,
1
3
10
30
concentration
(mM)
Fig. 2. Concentration-dependent antagonism by uranyl ion (UOz+; 0) and strontium ion (Sr’+; a) on the neuromuscular blocking action of /I-bungarotoxin in mouse diaphragm. (0) Represents the time to neuromuscular block (N-M block) induced by 0.22nM /j’-bungarotoxin alone (n = 4).
Uranyl Table
I. Effect
of time course of application
nitrate
97
and neurotoxins
of uranyl ion (UO:*) and strontium ion (Sr”) (/j-BuTX) in mouse diaphragm
on their antagonism
against fl-bungarotoxin
Time to neuromuscular Incubation fl-BuTX
Time sequence of ion application
with
No washout Washout after 20 min incubation Washout after 20 min incubation
Added by the Added Added Added
Washout after 20 min incubation No washout No washout
simultaneously wth [GBuTX followed washout after 20min Incubation after washout of fl-BuTX simultaneously with /SBuTX 20min orior to I(-BuTX
uo:+
Sr’+
129.9 k 5.7 (7) 133.2 k 6.5 (4) 207.2 k 18.6 (4);
129.9 + 5.7 133.2 + 6.5 195.7 * II.1 (4)’
238.0 k 31.3 (6)* > 350 (3)* > 350 (4)*
274.0 k 26.2 (6)’ 31 1.3 k 7.6(3)* 314.0 k 3.8 (4)’
Final concentrations of /?-bungarotoxin, UO:+ and Sr’+ was 0.22 PM, 0.4 mM and I5 mM, respectively. respective control in the absence of divalent cation.
.E E
350 -
‘: 0 z I i z
250-
h 4-t-4 !-‘-_J
\ \\
‘\
‘\X\\
“p\\,,,
*\1’ -\
150
‘4.
5!
:
,o E G
block (min)
*P i 0.05 as compared
with the
antagonistic effect. When 0.4mM UO:+ was added 50min after /j’-bungarotoxin, there was no antagonistic action, but when the concentration was increased to 0.8 mM, it exerted again a significant antagonistic effect. Even when the twitch amplitude was reduced by b-bungarotoxin to about 209/o of control value, 0.8 mM concentration of IJ@+ still exerted antagonism against the neuromuscular blocking action of fi-bungarotoxin.
50 i!
0
’ 20
Time
after
’ 40
’ 60
B - BuTX
’ 80
’ 100
( rn~n1
Fig. 3. Effect of uranyl ion (UOz+), strontium ion (S?‘) on the neuromuscular blocking action of /I-bungarotoxin when they were applied at different time intervals after p-bungarotoxin in the mouse phrenic nerve-diaphragm preparation. UO:+ (0.4, 0 or 0.8 mM n ) Sr’+ (15 mM, A) was applied, respectively at a time interval indicated after B-bungarotoxin, and the time to complete neuromuscular block (N-M block) induced by /j’-bungarotoxin alone (0.22pM, 0) or fl-bungarotoxin in the presence of UO:+ or Sr*+ was recorded.
the target site, but must be due to slow changes that occur after binding. To test whether the antagonistic effect of UOs+ and Sr*+ was due to an inhibition of the binding of the toxin or to a retardation of the changes occurring after binding, UOi+ or Sr2+ was added simultaneously with /3-bungarotoxin for 20 min and then washed repeatedly with Krebs’ solution. Paralysis developed in these diaphragms at a slower rate than in those similarly exposed to /?-bungarotoxin alone without UO:+ or Sr2+ (Table l), indicating that less fl-bungarotoxin was bound. In another experiment the diaphragms were incubated with /3-bungarotoxin for 20min and then washed repeatedly with Krebs’ solution, UOs+ or Sr2+ was added after washout, these procedures also showed that UO: + and Sr2+ prolonged the development of neuromuscular blockade (Table Since 1). fi-bungarotoxin had ample time to bind to its site of action, it follows that the change of release mechanism occurring after binding with /?-bungarotoxin was also retarded. Both ions also revealed antagonistic action even if they were applied after longer periods of incubation with b-bungarotoxin (Fig. 3). The longer the incubation time, the weaker was the
Reversal by Ca’+ of UOg+ or Sr2+ antagonism /I-bungarotoxin
of
When /j’-bungarotoxin was applied in a low Ca*+ Krebs solution the neuromuscular blocking time of b-bungarotoxin was prolonged; the block by /l-bungarotoxin was not affected in Ca*+-rich (10 mM) Krebs solution. The antagonistic effect of Sr2+ against the block induced by /3-bungarotoxin was inversely proportional to the Ca*+ concentration in the medium. At concentrations of 3-10 mM, Ca*+ decreased and eventually abolished the antagonistic effect of Sr*+ (Fig. 4A). By contrast, reducing the concentration of Ca2+ (0252.5 mM) enhanced that of Sr2+. Increase of the concentration of Ca2+ also abolished the antagonistic effect of UOz+ on the block by fi-bungarotoxin (Fig. 4A). In low Ca*+ medium, UOi+ prolonged the time required for complete neuromuscular blockade (100% paralysis) by b-bungarotoxin to more than 350 min; however, based on the time required for 60% paralysis. low Ca*+ decreased the effect UOs+ (Figs 4B and 5). For instance, at a concentration of 0.7 mM Ca2+, UOi+ markedly potentiated the amplitude of the twitch; after 20min, /I-bungarotoxin was added and the three phases of the effects of p-bungarotoxin on the amplitude of the twitch were produced, e.g. the initial depression followed by the second increase and finally a rapid decrease (Fig. 5B). However, the decrease reached a steady state 70min after the application of b-bungarotoxin, and the small twitches remaining no longer decreased, which persisted for more than 350 min (Figs 4A and 5B). Effect of lJO$’ and Sr” activity of /I-bungarotoxin
on the phospholipase
A,
The phospholipase A, (PLA,) activity of /?-bungarotoxin was estimated to be 203.7 f 6.9 pmol fatty
98
S. Y. LIN-SHIAU and W. M. Fu 0 Control A uo2+04 2
0
(A)
Sr”9
mM mM
350 L-_L___~
r
7
50 t
OL
’
0.3
I
I
1
3
L
I
,
03
10
Coz+co”ce”trat~on
I
I
I
1
3
10
( mM I
Fig. 4. Effect of Ca2+ on the antagonistic action of uranyl ion (UO:+) or strontium ion (Sr2+) against the neuromuscular blocking action of fi-bungarotoxin (0.22pM) in the mouse diaphragm. Time to paralysis in (A) indicated 100% paralysis whereas that in (B) indicated 60% paralysis.
acid released/mg b-bungarotoxin/min. Uranyl ions and Sr*+ inhibited the enzyme activity of p-bungarotoxin in a concentration-dependent manner (Fig. 6); 0.25 mM UO$+ and 10 mM of Sr*+, which prolonged the neuromuscular blocking time of /I-bungarotoxin approx. to 2-fold (Fig. 2) inhibited the phospholipase A, activity of /I-bungarotoxin by 63.5 t_ 4.9 and 65.2 + 1.9%, respectively. Effects
of UOi+ on the actions of other neurotoxins
Crotoxin (0.23 PM), a-bungarotoxin (0.3 PM) and tetrodotoxin (0.02 PM) blocked neuromuscular transmission in 101.9 * 8.8, 42.6 f 3.8 and 78.9 k 7.9 min (n = 4-lo), respectively. Uranyl ions antagonized the action of crotoxin in a concentration-dependent manner (Fig. 7; n = 4). However, UOi+ had no effect on the neuromuscular blocking action of r-bungarotoxin and tetrodotoxin (Fig. 7). In the presence of 2.5 mg/ml phosphatidylcholine, the phospholipase A, activity of crotoxin was estimated to be 41.3 f 3.9 pmol fatty acid released/mg crotoxin per min. Uranyl ions (0.5 mM) inhibited the enzyme activity of crotoxin by 31.0 + 3.0% (n = 5). DISCUSSION
It has been reported that strontium ion (Sr*+) not only inhibits the phospholipase A, activity but also antagonizes the presynaptic blockade caused by several neurotoxins isolated from snake venoms (Strong et al., 1976; Chang et al., 1977; Howard and Gundersen, 1980). In the present study, it was found that the inhibitory effect of uranyl ion (UOi+) on either the activity of phospholipase A, or the
neuromuscular blocking action of fi-bungarotoxin was much more potent than that of ST’+. So far as is known, UOi+ is the most potent inhibitor of P-bungarotoxin. It has been demonstrated that sufficient and irreversible binding of /3-bungarotoxin takes place within 20 min, and neuromuscular blockade developed more slowly (Chang and Lee, 1963). Uranyl ions and Sr2+ antagonized /I-bungarotoxin during both phases of the action of p-bungarotoxin: (1) when they were incubated with /I-bungarotoxin during the first 20min, followed by the washout; and (2) when they were added after 20-100 min of incubation with fi-bungarotoxin, followed by the washout of the toxin. These results suggest that UO:+ and Sr2+ not only inhibited the binding of /I-bungarotoxin but also inhibited the subsequent steps leading to the neuromuscular blockade by b-bungarotoxin. /?-Bungarotoxin has been reported to be able to penetrate the axon membrane and enter into the nerve terminals (Strong et al., 1976). On the other hand, its phospholipase A, activity was more marked when /I-bungarotoxin was injected intracellularly into the axon (Hinzer and Taut, 1977; Narahasi and Tobias, 1964). In the present study, it was found that UO:+ antagonized the action of b-bungarotoxin when it was applied at different incubating time intervals after P-bungarotoxin even when 80% paralysis was caused by P-bungarotoxin. Two possibilities seem to be indicated by these results. First, the effect of P-bungarotoxin may be due to its action on the outside of the membrane of nerve terminals; second, /I-bungarotoxin may increase the permeability of the presynaptic membrane so that UO:+ which normally is incapable of penetrating the membranes may enter
Uranyl
nitrate
and neurotoxins
S. Y. LIN-SHIAU and W. M. Fu
100 100
c 0
-
.
60-
2
4
; 60c
/
H 40E 8 ; 20a 0’
I
I
0 01
10
01
1
Concentrotlon
1mM 1
25
0) or strontium ion (Sr2+; A) on phospholipase A2 activity of /3-bungarotoxin. The specific activity of /?-bungarotoxin (control) was estimated to be 23.7 + 6.9 pmol/mg per min. Fig. 6. Inhibition
by uranyl ion (UO:+;
into nerve terminals and inhibit the actions of /?-bungarotoxin. Masukawa and Livengood (1982) reported that treatment with /I-bungarotoxin increased the permeability to cations such as Co2+, Mg2+, Mn2+ and Ni’+; that normally cannot enter the membranes, these cations greatly increased the miniature endplate potential (m.e.p.p.) frequency after treatment with /?-bungarotoxin. Uranyl ion is able, at very small concentrations, to change the surface potential of charged phospholipid membranes (McLaughline, Szabo and Eisenman, 1971); this indicates that UOi+ has high affinity to phospholipids. In previous studies, it was found that UOi+ could interact directly with phospholipids as it decreased the fluorescence intensity of the probe 8-anilino-1-naphthalene sulphonate (Lin-Shiau, Fu and Mao-Chen, 1984). It is proposed therefore, that UOi+ binds to phospholipid components of the nerve terminal membrane to form a metal ion-phospholipid complex. The formation of the
UO~+concentrar~on
Fig. 7. Doseeresponse
(mM)
curves of uranyl ion (UOz+) on the neuromuscular blocking action of neurotoxins (tetrodotoxin 0.02 PM, 0, 0; a-bungarotoxin 0.3 PM, 0, n ; crotoxin 0.23 PM, & A) in the mouse phrenic nerve diaphragm preparation. The open and closed symbols represent the time to neuromuscular block (N-M block) induced by these neurotoxins in the absence or in the presence of UO:+, respectively.
complex may alter the conformation of the structure of the membrane so that the binding of /?-bungarotoxin is inhibited. Furthermore, the complex formed is not, a suitable substrate for P-bungarotoxin, thus leading to inhibition of phospholipase A, activity and then inhibition of the neuromuscular blocking action of P-bungarotoxin. Calcium ions are essential for the release of transmitter (Bennett, Florin and Hall, 1975). Calcium ions bind to the membrane phospholipids and neutralize the negative charges on the membrane, enhancing release of transmitter by facilitating the fusion of synaptic vesicles with the nerve terminal membrane (Portis, Newon, Pangborn and Papahadjopaulos, 1979). In the present study, it was found that high levels of Ca2+ effectively reversed while low Ca2+ (0.25-l mM) augmented the antagonistic action of UOi+ and Sr2+. It is suspected that Ca2+ exerts the inhibitory action on the antagonism by UO:+ and Sr*+ possibly by competing for binding to membrane phospholipids. In a previous experiment, it was demonstrated that UOi+ potentiated the amplitude of the twitch by the release of ACh from motor nerve terminals (LinShiau and Fu, 1979). This potentiating property of UOi+ was enhanced in a low Ca2+ medium. It is interesting that in a low Ca2+ medium, plus UO:+, the addition of /?-bungarotoxin exhibited three phases of effects, e.g. initial depression of the twitches, followed by increased amplitude of the twitches and finally a rapid decrease of the amplitude of the twitch which reached a steady state approx. 70 min after fl-bungarotoxin and persisted for more than 350 min. Spokes and Dolly (1980) claimed that the first depression was due to the specific binding of fl-bungarotoxin to the neural membrane; the second facilitation may be due to the increased release of acetylcholine and the final neuromuscular blockade is due to the inhibition of the release of ACh. The findings of the present study indicated that UO:+ enhanced the second facilitatory phase, possibly owing to increased release of ACh by UOi’ (LinShiau and Fu, 1979). Crotoxin is a presynaptically active neurotoxin containing phospholipase A, activity (Breithaput, Riibsamen and Habermann, 1971; Hendon and Fraenkel-Conrat, 1971). That UOi+ inhibits both the activity of phospholipase A, and the neuromuscular blocking action of this toxin further supports the effective antagonism by UOi+ on the presynaptic neurotoxins. Uranyl ions do not affect the actions of a-bungarotoxin and tetrodotoxin, indicating that the action of UO:+ is fairly specific. It is apparent that the mode of action of the presynaptic neurotoxin is more complex than the postsynaptic neurotoxin which simply occupies ACh receptors of the endplate. In order to shed some light on the action of presynaptic neurotoxin on the release mechanism of ACh, a potent antagonist will certainly be helpful. In this study, it has been shown that UOz+
Uranyl nitrate and neurotoxins antagonized the action of preferentially j?-bungarotoxin and crotoxin but did not affect that of a-bungarotoxin and tetrodotoxin.
REFERENCES Bennett M. R., Florin T. and Hall R. (1975) The effect of calcium ion on the binominal statistical parameters which control acetylcholine release at synapses in striated muscle. .I. Physiol., Land. 241: 429446. Breithaput H., Riibsamen K. and Habermann E. (1971) In vitro and in viva interactions between phosphohpase A and novel potentiator isolated from so-called crotoxin. Naunyn-Schmiedebergs Arch. Pharmac. 269: 403404. Bulbring E:. (1946) Observations on the isolated phrenic nerve diaphragm preparation of the rat. Br. J. Pharmac. 1: 38-61. Chang C. C. and Lee C. Y. (1963) Isolation of neurotoxins from the venom of Bungarus multicinctus and their modes of neuromuscular blocking action. Archs int. Pharmacodyn. Ther. 144: 241-257. Chang C. C., Su M. J., Lee J. D. and Eaker D. (1977) Effects A and the of Sr*+ and Mg’+ on the phosphohpase neuromuscular blocking actions of presynaptic and taipoxin. Naunyn crotoxin fi-bungarotoxin, Schmiedebergs Arch. Pharmac. 299: 1555161. David B. J. (1964) Disc electrophoresis II. Method and application to human serum proteins. Ann. N.Y. Acad. Sci. 121: 404. Hendon R. A. and Fraenkel-Conrat H. (1971) Biological roles of the two components of crotoxin. Proc. nafn. Acad. Sci. U.S.A. 68: 156&1563. Hinzen D. H. and Taut L. (1977) Membrane properties of aplysia neurons intracellularly injected with phosphohpase A and C. J. Physiol., Lond. 268: 21-34. Howard B. D. and Gundersen C. B. (1980) Effects and mechanisms of polypeptide neurotoxins that act presynaptically. A. Reu. Pharmac. Toxic. 20: 37-336. Lee C. Y. (1972) Chemistry and pharmacology of poly-
101
peptide toxins in snake venoms. A. Rev. Pharmac. Toxic. 12: 265-286. Lee C. Y., Chang S. L., Kau S. T. and Luh S. Y. (1972) Chromatographic separation of the venom of Bungarus multicinctus and characterization of its components. J. Chromat. 72: 71-81. Lin-Shiau S. Y. (1983) Selective antagonism by uranyl nitrate against Formosan Krait venom. Part I. J. Formosan med. Ass. 82: 640-643. Lin-Shiau S. Y. and Fu W. M. (1979) Effects of uranyl ions on neuromuscular transmission of chick biventer cervicis muscle. Archs int. Pharmacodyn. Ther. 241: 332-343. Lin-Shiau S. Y., Chen C. C. and Fu W. M. (1983) Selective antagonism by uranyl nitrate against Formosan Krait venom. Part II. J. Formosan med. Ass. 82: 1099-1103. Lin-Shiau S. Y., Fu W. M. and Mao-Chen C. T. (1984) Selective antagonism by uranyl nitrate against Formosan Krait venom. Part III. Mechanism of action of uranyl nitrate. J. Formosan med. Ass. 83: 21-26. Masukawa L. M. and Livengood D. R. (1982) Alterations in spontaneous transmitter release by divalent cations after treatment of the neuromuscular junction with p-bungarotoxin. Cell. molec. Neurobiol. 2: 277-290. McLaughlin S. G. A., Szabo G. and Eisenman G. (1971) Divalent ions and the surface potential of charged phospholipid membrane. J. gen. Physiol. 58: 667-687. Narahashi T. and Tobias J. M. (1964) Properties of axon membrane as affected by cobra venom, digitonin and protease. Am. J. Physiol. 207: 1441-1446. Portis A., Newon C., Pangborn W. and Papahadjopauios D. (1979) Studies on the mechanism of membrane fusion: Evidence for an intermembrane Ca2+-phosphohpid complex, synergism with Mg2+ and inhibition by spectrin. Biochemistry 18: 780-790. Spokes J. W. and Dolly J. 0. (1980) Complete purification of /I-bungarotoxin, characterization of its action and that of tityustoxin on synaptosomal accumulation and release of acetylcholine. Biochim. biophys. Acta 596: 81-93. Strong P. N., Goerke J., Oberg S. G. and Kelly R. B. (1976) /I-Bungarotoxin, a presynaptic toxin with enzymatic activity. Proc. natn. Acad. Sci. U.S.A. 73: 175-182.