Naunyn-Schmiedeberg's Arch Pharmacol (1987) 335:1 - 7 Naunyn-Schmiedeberg's

Archivesof Pharmacology © Springer-Verlag 1987

Differential effects of various secretagogues on quantal transmitter release from mouse motor nerve terminals treated with botulinum A and tetanus toxin

F. Dreyer, F. Rosenberg, C. Becket, H. Bigalke, and R. Penner Rudolf-Buchheim-Institut f/Jr Pharmakologie, Frankfurter Strasse 107, D-6300 Giel3en, Federal Republic of Germany

Summary. 1. Electrophysiological and electron microscopic techniques were used to investigate the actions of potassium depolarization, black widow spider venom (BWSV), Ca 2 +- ionophore A 23187 and hyperosmotic solution on mouse hemidiaphragms poisoned in vitro with botulinum A toxin (BoTx) and tetanus toxin (TeTx). These neurotoxins reduced the frequency of miniature endplate potentials (m.e.p.ps) from 5/s of the control to 2/min and 21/rain, respectively.

2. High potassium (25 mmol/1) increased the m.e.p.p.- frequency at BoTx- and TeTx-poisoned endplates to 30/rain and 50/s, respectively. The ultrastructure of endplates showed no obvious changes.

3. BWSV (0.04 glands/ml) was just as effective in pro- moting transmitter release from BoTx-treated endplates as in control preparations. Electron micrographs revealed depletion of vesicles as well as swollen and disrupted mitochondria. When preparations were pretreated with TeTx, BWSV only moderately increased transmitter release and no alterations of the ultrastructure could be observed.

4. At TeTx- or BoTx-poisoned endplates Ca2+-iono - phore A 23187 usually produced an extreme reduction of m.e.p.p.-frequency (0.005/s), sometimes preceded by a short burst-like release. The ultrastructure of these endplates was not obviously affected.

5. Application of hyperosmotic solution to BoTx- or TeTx-poisoned preparations further reduced the already low m.e.p.p.-frequency.

6. These results further demonstrate that TeTx and BoTx act at different sites in the transmitter releasing pro- cess.

Key words: Tetanus toxin - Botulinum A toxin - Minia- ture endplate potential - Motor endplate - Black widow spider venom - Ca2+-ionophore - Hyperosmotic solu- tion - Potassium depolarization

Introduction

Botulinum A toxin (BoTx) is known to reduce nerve-evoked and spontaneous quantal transmitter release (for reviews Gundersen 1980; Simpson 1981; Mellanby 1984). Tetanus toxin (TeTx), another clostridial neurotoxin has a similar

Send offprint requests to F. Dreyer at the above address

depressing action (for reviews Mellanby and Green 1981; Wellh6ner 1982). Recent work, however, has indicated that the two toxins may act at different sites in the chain of events leading to transmitter release (Dreyer and Schmitt 1981, 1983).

Various agents and procedures like high concentration of extracellular potassium, black widow spider venom (BWSV), Ca 2 +-ionophore A 23187 and hyperosmotic solu- tion induce a strong increase of the frequency of miniature endplate potentials (m.e.p.ps) at nerve terminals. Only few studies have assessed the interactions of these releasers with TeTx (Duchen and Tonge 1973; Bevan and Wendon 1984) and investigations with BoTx are difficult to compare with each other, because the data were obtained from different species and muscles which may respond differently to the toxins. Further complications may arise from the use of local toxin injection in vivo, with the drawback that the degree of neuromuscular block achieved remains uncertain. In addition, experiments were carried out at different temperatures and with different types of botulinum toxin.

For the better understanding of the mechanisms of ac- tion of the clostridial toxins it was of interest to compare the actions of the so-called releasers on TeTx or BoTx treated nerve endings under identical experimental conditions using in vitro poisoned hemidiaphragm preparations of the mouse.

Parts of the results have been presented on the Spring Meeting of the German Pharmacological Society (Dreyer and Rosenberg 1983) and on the Gif-Lectures in Neurobiology 1983 (Dreyer et al. 1984).

Methods

The experiments were performed on isolated phrenic nerve- hemidiaphragm preparations of mice. The Krebs-Ringer so- lution had the following composition (mmol/1): NaC1 115, KC1 5, CaC12 2.5, MgSO4 1, Na2HPO4 1, NaHCO3 25, glucose 11, gassed with 95% 02 and 5% CO2 (pH 7.4). Recordings of miniature endplate potentials (m.e.p.ps) were performed at 37°C, unless otherwise stated, using con- ventional electrophysiological techniques. Frequencies of m.e.p.ps exceeding 300/s could not accurately be counted and therefore were designated in the diagrams as 'more than 300/s'.

The procedure of intoxication has been described elsewhere (Dreyer and Schmitt 1981; 1983). In the present

2

Table 1. Effect of potassium depolarization, black widow spider venom (BWSV), Ca 2 ÷-ionophoreA 23187 and hypertonic sucrose solution on frequency of m.e.p.ps at motor endplates intoxicated with BoTx or TeTx

Treatment Unpoisoned BoTx TeTx

Control 5.2 4- 2.7/s (12/10) 2.2 4- 0.9/min (9/9) 21 _ 8/min (10/10) KC1 (25 mmol/1) > 300/s (20/3) 29 4- 11/min (7/3) 47 4- 35/s (11/3)

BWSV (0.04 glands/ml) Ca 2 +-ionophore (20 gmol/1) Sucrose solution (150 mmol/1)

> 300/s (11/4) > 300/s (5/5) 170 4- 90/s (6/3)

> 300/s (11/6) 0/min- 250/s (16/6) 0.9 4- 0.8/min (9/3)

13 _+ 15/s (14/6) 0/min- 150/s (12/6) 5.9 4- 5.2/rain (10/3)

The numbers in parentheses denote the number of endplates and muscles examined, respectively. The results are expressed as mean 4- SD. M.e.p.p.-frequency higher than 300/s could not be measured accurately. Only those experiments in which the development of drug effects could be traced for at least 20 min at a single endplate are included. In some preparations the m.e.p.p.-recording was continued in other endplates of the same muscle

>300 • - - • - - • - • - - - 0 - - - • - - - • - - - • - - - • - - - • - - - • - - - 0 - - - •

10~

×

gL 50 • ×,. i

ck

x E ~ . / BoTx

o ~(Se,-=-,4go-'t&-o-o-o-o-o-o-o-o-o-o-o-o-o-o-o-o-o-o-o I I I I E I 0 10 20 30 40 rnin

Fig. 1. Effect of high potassium on the m.e.p.p.-frequency at control (0), BoTx- (©) and TeTx- (x) poisoned endplates. At time zero the muscle preparations were superfused with a saline containing 25 mmol/l KC1

>300~- ~ o - o - o - o

~ T I / / B o T x 400 - I ""°*//!

i 5 0 - I II × / ? I I / /~ \ × ~eTx # / × \ * '

- , a / - I E 2,,-,,_, ",x" - -

0 - & , d & - , , - ~ ffCX:,,~'-'-'~'~-A-,S-A~-'~ ~-'a I I I I I I 0 10 20 30 4.0 min

Fig. 2. Effect of black widow spider venom on the m.e.p.p.- frequency at control (0), BoTx- (©) and TeTx- ( x, A) poisoned endplates. At zero time BWSV was added directly into the bath solution (0.04 glands/ml)

experiments toxin concentrations of 5 gg/ml for TeTx and 25 ng/ml for BoTx type A were used, which ensured complete abolition of evoked quantal release at 0.1 Hz nerve stimulation within 90 min. The frequency of spontaneous m.e.p.ps was reduced from about 300/rain to 20/rain and 2/min, respectively (Table 1). The degree of poisoning may be regarded as supramaximal, since both ten-fold lower or higher toxin concentrations were equally effective at in- hinting transmitter release, eventually. They only affected time course of paralysis.

Tetanus toxin was kindly provided by Dr. Bizzini (In- stitut Pasteur, Paris, France). Its LDs0 (mice subcutaneous in the neck region) was about 3 ng/kg. Botulinum A toxin (LDs0 8 ng/kg) was a gift from Dr. Schantz (Food Research Institute, Madison, WI, USA). Black widow spider venom (BWSV) was a crude extract from the venom glands of the black widow spider, Latrodectus mactans. Living spiders were kindly supplied by Dr. M. Ziegler who breeds them. For extraction of the venom the cephalothoraces were thawed, the glands pulled out and homogenized in 0.9% NaC1. The homogenate was centrifuged and the supernatant was kept frozen at - 2 0 ° C. The final concentration in the bath solution was 0.4 glands/ml. The Ca2+-ionophore A23187 (Sigma) was dissolved in dimethylsulfoxide (DMSO) and used in a final concentration of 20 gmol/1. Nerve endings were depolarized by increasing the KC1 in the Krebs-Ringer solution to 25 mmol/1. The osmolarity was maintained by decreasing the Na÷-concentration. The

hyperosmotic solution was obtained by adding sucrose to the Krebs-Ringer solution to a concentration of 150 mmol/1.

For electron microscopy the poisoned diaphragms were incubated for I h with the substances to be tested. Sub- sequently, they were prefixed by replacing the bath solution with cold glutaraldehyde (2.5 %) in 0.1 tool/1 phosphate buf- fer (pH 7.2). After 15 min small pieces were cut out from the endplate regions of the diaphragms and fixed at least for 15 h in glutaraldehyde (2.5%) at 4°C. The specimens were post-fixed in OsO4 (1%) for 2 h in the dark at room tempera- ture. Following dehydration the specimens were embedded in Epon 812. Ultrathin sections were cut with a diamond knife in a Reichert-ultramicrotome. They were double- stained with uranyl acetate and lead citrate and examined with a Siemens EM 101 electron microscope.

Results

Effect of K+-depolarization

Superfusion of an unpoisoned preparation with high K + solution (25retool/l) usually increased the m.e.p.p.- frequency within a few minutes to values above 300/s and remained at this high level for more than I h (Fig. 1, Table 1). At BoTx-poisoned endplates the same treatment produced an increase of the m.e.p.p.-frequency to only 30/ min. At TeTx-poisoned nerve terminals high K + solution caused, after a delay of 10 min, a transient increase in m.e.p.p.-frequency to values in the range of 100/s. It declined

Fig. 3 Effect of BWSV on the nerve terminal ultrastructure of (A) control, (B) BoTx- and (C) TeTx-poisoned motor- endplates. Hemidiaphragm preparations were treated with the toxin (0.04 glands/ml) for 1 h, Pr, presynaptic membrane; Pro, postsynaptic membrane; Sc, synaptic cleft; jr, junctional folds; Nt, nerve ter- minal; sv, synaptic vesicles; M, mitochondria; S, Schwann cell; cf, collagen fibres. Scale bars: I gm

A T e t o x B B W S V 3 '

. . . . . . . . ii ii iii i i i tll i I I

C B W S V 5 ' D B W S V 8 '

I I I l l I I III . . . . .

_ _ L I I I I I I I II I

IIIII 1 |1 I I I

E I ~ W S V 14 ' 2 m V | L._ 25 ms

i' 2 m V ~

Fig. 4 A - E . Effect of BWSV on spontaneous transmitter release of a TeTx-poisoned M. triangularis sterni preparation. (A) m.e.p.p.- frequency after intoxication with TeTx (superimposed traces showing 15 s of continuous recording). (B) m.e.p.p.-frequency 3 min after application of BWSV (0.04 glands/ml). (C) episodic bursts of m.e.p.ps. (D) m.e.p.p.-frequency 8 min after application of BWSV (superimposed traces showing 2 min of continuous recording). (E) 'anomalous m.e.p.p.' (note the different time scale)

within 10 min to values around 50/s and remained at that level for more than 1 h (Fig. 1).

The treatment with high potassium for 1 h did not pro- duce significant alterations of the ultrastructural appearance of nerve terminals in toxin-treated motor endplates.

Effect of black widow spider venom

The venom of black widow spiders (BWSV) is known to cause a massive release of transmitter from nerve terminals. Morphological data reveal depletion of synaptic vesicles and other intraterminal organelles (Longenecker et al. 1970). In agreement with previous reports (Cull-Candy et al. 1976; Kao et al. 1976) the enhancement in quantal release was also observed in BoTx-treated endplates. The onset of the release was, however, delayed and the duration prolonged compared to the control (Fig. 2).

Surprisingly, BWSV was much less potent at TeTx- poisoned endplates (Table 1). In 50% of the examined endplates the m.e.p.p.-frequency increased only moderately, i.e. by a factor of about 2 (Fig. 2). In other endplates episodic bursts of m.e.p.ps lasting for 5 - 1 0 s occurred after a delay of about 10 min. Sometimes the m.e.p.p.-frequency within these burst-like discharges increased to about 100/s in- terrupted by periods of much lower values (Fig. 2).

The ultrastructure of the nerve terminals reflected the different action of BWSV in BoTx- and TeTx-poisoned prep- arations. The control nerve terminals as well as those treated with BoTx were almost depleted of synaptic vesicles with alterations and damages of mitochondria and plasma mem- brane (Figs. 3 A, B). The treatment of TeTx-poisoned endplates with BWSV for 1 h caused no such alterations (Fig. 3 C).

The action of BWSV on TeTx-poisoned endplates was also investigated on the M. triangularis sterni of the mouse (McArdle et al. 1981). The procedure of intoxication was

>300 - BoTx--

1001- 212

C (9

5C- I

cJ~

E 0

I 0

/ --control

TeYx

10 20 min

- •

TeTx-- 1 ontrol

I

BoTx

x ~ P

; 10 Fig. 5. Typical examples of the effect of Ca2+-ionophore A 23187 (20 gmol/1) on the m.e.p.p.-frequency at control (O), BoTx- (0) and TeTx- (x) poisoned endplates. Note that the effect of the Ca z +- ionophore on toxin-treated nerve endings varies considerably

similar to that of the hemidiaphragm. However, the experi- ments were conducted at room temperature (20-23°C). The frequency ofm.e.p.ps of a control preparation decreased from 6/s to a level of 25/min after the treatment with TeTx (Fig. 4A). Subsequent application of BWSV (0.04 glands/ ml) led to a 4 0 - 50-fold increase of transmitter release within 2 - 3 min and lasted for about 4 - 5 min (Fig. 4 B). In two out of five experiments episodic bursts could be observed (Fig. 4C). After 7 - 8 min the release invariably decreased to values far below the control level (0.5/min, Fig. 4D). Within the subsequent observation period of 60 min 2 - 5 'anomalous m.e.p.ps' of variable amplitude ( 2 - 1 2 mV) and extremely long duration (100-200ms) were observed (Fig. 4E) in four out of five experiments. Such signals were never observed under any other experimental conditions.

The decline to such an extremely low frequency of m.e.p.ps as well as the occurrence of 'anomalous m.e.p.ps' was not observed in the hemidiaphragm preparation. These differences do not arise from differences in temperature at which the experiments had been performed, since the effects observed on the M. triangularis sterni were not recorded in hemidiaphragms kept at room temperature. The differences between the two preparations may arise from genuine differences between endplates of the two preparations.

Recently, specific receptors for e-latrotoxin, the major component of BWSV, have been identified in the nerve terminal plasma membrane (Valtorta et al. 1984). Therefore, the question arose if TeTx interacts with these receptors, thus preventing the binding and the action of BWSV. Pre- viously, we have shown that the action of TeTx on transmitter release involves three sequential steps: 1. binding to the presynaptic membrane, 2. translocation into or through the presynaptic membrane, and 3. paralysis, i.e. blockade of quantal release (Schmitt et al. 1981). Since these steps can be separated by appropriate experimental conditions (cf. Schmitt et al. 1981), we were able to test the effect of BWSV on TeTx-treated endplates after the binding, the transloca- tion or the paralytic step had been completed. The results of these experiments showed that BWSV was still very effective on nerve endings in which the action of TeTx either had been stopped after the binding step or after the binding and translocation step. The ultrastructure of these endplates showed comparable damages as control endplates treated with BWSV (cf. Fig. 3 A). Only when the paralytic step was completed, BWSV lost its marked effect on transmitter re-

Fig. 6 Effect of C a 2 +-ionophore A 23187 (20 p.mol/ 1) on the ultrastructure of (A) control, (B) BoTx- and (C) TeTx-poisoned endplates after 1 h treat- ment. For abbreviations see Fig. 3. Scale bars: I ~tm

lease and no longer caused substantial changes in the ultrastructure of the nerve terminals.

Ca 2 +-ionophore A 23187

In agreement with previous studies (Cull-Candy et al. 1976; Kao et al. 1976), CaZ+-ionophore A23187 (20 ~tmol/1) consistently caused a massive quantal release from control nerve terminals (Fig. 5, Table 1). At BoTx- and TeTx- poisoned preparations the same treatment exhibited con- siderable variations between different endplates. In 70% of the terminals treated with either toxin, the ionophore pro- duced a decline in m.e.p.p.-frequency to extremely low levels (0.005/s). In other endplates there was a sudden burst-like release with frequencies in the range of 150-300/s lasting only about 1 - 2 min. The bursts were followed by a total disappearance of m.e.p.ps. Similar results were obtained when the external CaZ+-concentration was raised to 10 retool/1 in the presence of the Ca z +-ionophore.

The decline of m.e.p.p.-frequency to zero cannot be due to the depletion of synaptic vesicles since the nerve terminals of BoTx- (n = 5) or TeTx-treated (n = 5) preparations were full of vesicles as shown by electron microscopical examina- tions (Fig. 6 B, C). Only few mitochondria were swollen. In unpoisoned endplates (n=4) treatment with Ca 2+- ionophore led to significant alterations of the general appearance of the nerve terminals (Fig. 6A). No vesicles were left, the mitochondria were swollen and partly des- troyed.

Hyperosmotic solution

Hyperosmotic solutions are known to increase the rate of spontaneous m.e.p.ps (Fatt and Katz 1952). Adding 150 mmol/1 sucrose to the bath medium increased the m.e.p.p.-frequency about 30-fold within 2 to 10rain (Table 1). This increase was followed by a slow decline to values around 50/s, in some endplates to values of l/s and lower within 1 h. Application of hyperosmotic solution to BoTx- or TeTx-poisoned endplates did not increase, but instead further reduced the already low m.e.p.p.-frequency of toxin-treated endplates by a factor of 2 to 3 (Table 1).

Discussion

One important property of BoTx- or TeTx-poisoned nerve terminals is the striking reduction in the frequency of spontaneous quantal release. In BoTx-poisoned endplates potassium depolarization (25 mmol/1) increased the rate of spontaneous quantal release from 2/min to about 30/min, a value very close to the rate of nerve-evoked quantal release during 50 Hz nerve stimulation. In the latter case the release probability of quanta is less than 1%, thus yielding about 30 quanta/min (Dreyer and Schmitt 1981, 1983). At the ex- tensor digitorum longus muscle of rats, Thesleff et al. (1983) observed that nerve terminals poisoned with BoTx A were completely insensitive to high K+-depolarization. At the same muscle, Kim et al. (1984) found a slight increase in m.e.p.p.-frequency, working with partially paralyzed muscle preparations. In our experiments with TeTx-poisoned endplates, high potassium raised the m.e.p.p.-frequency 140- fold to about 50 quanta/s. Again, this value is comparable with the number of quanta of about 20/s which can be

recorded in TeTx-poisoned endplates during 50 Hz nerve stimulation. This observation is in contrast to the report of Duchen and Tonge (1973), who found that high potassium failed to increase the frequency of m.e.p.ps in TeTx-treated slow (soleus) and fast (extensor digitorum longus) skeletal muscles of the mouse, although repetitive nerve stimulation was effective in this respect.

BWSV is known to induce a massive quantal release of acetylcholine with marked depletion of synaptic vesicles at normal (Longenecker et al. 1970) and BoTx-treated neuromuscular junctions (Cull-Candy et al. 1976; Kao et al. 1976; Hirokawa and Heuser 1981). One current hypothesis of BWSV action is that c~-latrotoxin, the major component of the venom, induces uptake of Ca 2 + and other divalent cations across the presynaptic membrane by activating a 'non-conventional' cation channel of low ionic specificity (Meldolesi et al. 1983). Interestingly, we found that BWSV was much less effective to stimulate quantal release at the TeTx-poisoned nerve terminals compared to normal or BoTx-treated ones (Table 1) and no alteration of ultrastructure of nerve endings could be observed.

A similar inhibition of BWSV activity has been found at neuromuscular junctions that had been pretreated with concanavalin A (Con A) (Rubin et al. 1978) or with fi- bungarotoxin (fl-BuTx) (Tzeng and Tian 1984). In these experiments it could not be excluded that Con A or fi-BuTx may affect the binding of BWSV to the nerve terminal. In the present study, the use of appropriate experimental conditions allowed us to exclude that c~-latrotoxin and tetanus toxin compete for the same binding sites at the presynaptic membrane.

It has been proposed that BWSV affects the relationship between the plasma membrane and cytoplasmic elements like microfilaments or microtubuli (Rubin et al. 1978) or that BWSV activates a contractile network thereby stimulating exocytosis (Tzeng and Siekevitz 1979). In contrast TeTx may inactivate this contractile network resulting in the known asynchroneous release of quanta during nerve stimulation (Dreyer and Schmitt 1981, 1983).

It is generally believed that changes in the intracellular concentration of calcium provide the control for quantal exocytosis. We have studied the effect of ionophore A 23187 which increases the permeability of cell membranes to calcium (Foreman et al. 1973). At normal endplates the ionophore elicited the expected drastic release of transmitter. At neurotoxin-treated nerve terminals its effect varied be- tween different endplates with a strong decrease of m.e.p.p.- frequency or a brief acceleration followed by a dis- appearance of m.e.p.ps. The electron micrographs revealed nerve endings unaltered in their ultrastructure. The experi- ments with the Ca 2 +-ionophore suggest that the amount of calcium entering the nerve endings may not be sufficient to stimulate quantal release at both BoTx- or TeTx-poisoned endplates. The reason for the disappearance of m.e.p.ps is still unkown. That BoTx-treated preparations do not re- spond to Ca 2 +-ionophores has already been shown by Cull- Candy et al. (1976) and Kao et al. (1976). Raising the extra- cellular Ca 2 +-concentration to 10 mmol/1 did not increase the effect of the ionophore although this has been reported for BoTx-poisoned preparations (Cull-Candy et al. 1976). One reason that we failed to observe such an effect could be that in our experiments the transmitter release is blocked more completely because of the in vitro poisoning proce- dure.

The mechanism by which the osmotic effect increases quantal release is not fully understood but probably due to the mobilization of Ca 2 + from intracellular stores like the mitochondria (Shimoni et al. 1977). In TeTx- or BoTx- poisoned endplates the already low rate of m.e.p.ps further decreased by a factor of 2 to 3. For BoTx this is in agreement with previous results of Thesleffet al. (1983), while for TeTx no change in m.e.p.p.-frequency has been reported (Bevan and Wendon 1984).

Excessive Ca 2 +-influx through Ca 2 + channels has been suggested to be responsible for swollen and/or disrupted mitochondria (Yates and Yates 1968; Hirokawa and Heuser 1981). F rom our experiments it is evident that the potent releasers which all are suggested to act by increasing the intraterminal Ca 2 +-level did not affect the ultrastructure of mitochondria in BoTx- or TeTx-treated nerve terminals with the exception of BWSV on BoTx-poisoned endplates. Since the neurotoxins do not block the Ca 2+ influx induced by nerve stimulation (Gundersen et al. 1982; Dreyer et al. 1983) and since it is unlikely that BoTx or TeTx block the Ca 2+ influx through the Ca 2 +-ionophore, one may postulate that this Ca 2 +-uptake is not sufficient to damage mitochondria.

Presumably, an additional Ca 2 +-uptake which may occur during the massive quantal release and the accompanied endocytotic activity leads to swollen and/or disrupted mitochondria.

The presented results further support the hypothesis that TeTx and BoTx act at different steps of the transmitter releasing process.

Acknowledgements. We wish to thank Miss H. Will for her excellent technical assistance. This work was supported by the Deutsche Forschungsgemeinschaft, SFB 47.

References

Bevan S, Wendon LMB (1984) A study of the action of tetanus toxin at rat soleus neuromuscular junctions. J Physiol (Lond) 348:1-17

Cull-Candy SG, Lundh H, Thesleff S (1976) Effects of botulinum toxin on neuromuscular transmission in the rat. J Physiol (Lond) 260:177-203

Dreyer F, Schmitt A (1981) Different effects of botulinum A toxin and tetanus toxin on the transmitter releasing process at the mammalian neuromuscular junction. Neurosci Lett 26:307- 311

Dreyer F, Schmitt A (1983) Transmitter release in tetanus and botulinum A toxin-poisoned mammalian endplates and its dependence on nerve stimulation and temperature. Pfl/igers Arch 399 : 228 - 234

Dreyer F, Rosenberg F (1983) Botulinum and tetanus toxin- poisoned motor nerve endings respond differently to agents which normally increase the rate of spontaneous transmitter release. Naunyn-Schmiedeberg's Arch Pharmacol 322:R 67

Dreyer F, Mallart A, Brigant JL (1983) Botulinum A toxin and tetanus toxin do not affect presynaptic membrane currents in mammalian motor nerve endings. Brain Res 270:373-375

Dreyer F, Becker C, Bigalke H, Funk J, Penner R, Rosenberg F, Ziegler M (1984) Action of botulinum A toxin and tetanus toxin on synaptic transmission: J Physiol (Paris) 79:252--258

Duchen LW, Tonge DA (1973) The effects of tetanus toxin on neuromuscular transmission and on the morphology of motor end-plates in slow and fast skeletal muscle of the mouse. J Physiol (Lond) 228 : 157 - 172

Fatt P, Katz B (1952) Spontaneous subthreshold activity at motor nerve endings. J Physiol (Lond) 117 : 109 - 128

Foreman JC, Mongar JL, Gomperts BD (1973) Calcium ionophores and movement of calcium ions following the physiological stimulus to secretory process. Nature 245: 249 - 251

Gundersen CB (1980) The effects of botulinum toxin on the syn- thesis, storage and release of acetylcholine. Progr Neurobiol 14:99-119

Gundersen CB, Katz B, Miledy R (1982) The antagonism between botulinum toxin and calcium in motor nerve terminals. Proc Roy Soc B 216:369-376

Hirokawa N, Heuser JE (1981) Structural evidence that botulinum toxin blocks neuromuscular transmission by impairing the calcium influx that normally accompanies nerve depolarization. J Cell Biol 88:160-171

Kao I, Drachman DB, Price DL (1976) Botulinum toxin: Mecha- nism of presynaptic blockade. Science 193:1256-1258

Kim YI, Lomo T, Lupa MT, Thesleff S (1984) Miniature endplate potentials in rat skeletal muscle poisoned with botulinum toxin. J Physiol (Lond) 356:587-599

Longenecker HE, Hurlbut WP, Mauro A, Clark AW (1970) Effects of black widow spider venom on the frog neuromuscular junc- tion. Nature 225:701 - 705

McArdle J, Angaut-Petit D, Mallart A, Bournaud R, Faille L, Brigant JL (1981) Advantages of the triangularis sterni muscle of the mouse for investigations of synaptic phenomena. J Neurosci Meth 4:109-115

Meldolesi J, Madeddu L, Torda M, Gatti G, Niutta E (1983) The effect ofe-latrotoxin on the neurosecretory PC12 cell line: Stud- ies on toxin binding and stimulation of transmitter release. Neuroscience 10: 997-1009

Mellanby J (1984) Comparative activities of tetanus and botulinum toxins. J Neurosci 11 : 29 - 34

Mellanby J, Green J (1981) Commentary. How does tetanus toxin act? J Neurosci 6:281 - 300

Rubin LL, Gorio A, Mauro A (1978) Effect of concanavalin A on black widow spider venom activity at the neuromuscular junction: Implications for mechanisms of venom action. Brain Res 143:107-124

Schmitt A, Dreyer F, John Chr (1981) At least three sequential steps are involved in the tetanus toxin-induced block of neuromuscular transmission. Naunyn-Schmiedeberg's Arch Pharmacol 317: 326- 330

Shimoni Y, Alnaes E, Rahaminoff R (1977) Is hyperosmotic neurosecretion from motor nerve endings a calcium-dependent process? Nature 267:170-172

Simpson LL (1981) The origin, structure and pharmacological activ- ity of botulinum toxin. Pharmacol Rev 33:155 - 188

Thesleff S, Molgo J, Lundh H (1983) Botulinum toxin and 4- aminoquinoline induce a similar abnormal type of spontaneous quantal transmitter release at the rat neuromuscular junction. Brain Res 264:89-97

Tzeng MC, Siekevitz P (1979) The binding interaction between c~- latrotoxin from black widow spider venom and a dog cerebral cortex synaptosomal membrane preparation. J Neurochem 33 : 263 - 274

Tzeng MC, Tian BS (1984) Beta-bungarotoxin antagonizes the effect of alpha-latrotoxin from black-widow-spider venom on the neuromuscular junction. J Neurobiol 15 : 157-160

Valtorta F, Madeddu L, Meldolesi J, Ceccarelli B (1984) Specific localization of the c~-latrotoxin receptor in the nerve terminal plasma membrane. J Cell Biol 99:124-132

Wellh6ner HH (1982) Tetanus neurotoxin. Rev Physiol Biochem Pharmacol 93 : 1 - 68

Yates RD, Yates JC (1968) The occurrence of intramitochondrial granules in nerve cells. Z Zellforsch Mikrosk Anat 92: 388- 393

Received March, 1986/Accepted Septenber 9, 1986