release of acetylcholine

16 downloads 0 Views 1MB Size Report
This research was supported by the Allan Hancock Foundation, the. National Science Foundation (NS 76-80657, 77-06782), and the Nelson. Research and ...
Proc. Natl. Acad. Sci. USA Vol. 77, No. 2, pp. 1219-1223, February 1980

Neurobiology

Leptinotarsin: A presynaptic neurotoxin that stimulates release of acetylcholine (neuromuscular junction/synaptosomes/miniature end plate potentials/calcium)

WILLIAM 0. MCCLURE*, BERNARD C. ABBOTT*, DANIEL E. BAXTER*, TING H. HSIAOt, LESLIE S. SATIN**, ALVIN SIGER*, AND JUN E. YOSHINO* *Section of Cellular Biology, University of Southern California, University Park, Los Angeles, California 90007; and tDepartment of Biology, Utah State University, Logan, Utah 84322

Communicated by James F. Bonner, November 14, 1979

ABSTRACT Leptinotarsin, a toxin found in the hemolymph of the beetle Leptinotarsa haidemani, can stimulate release of acetylcholine from synaptic termini. Leptinotarsin causes an increase in the frequency of miniature end plate potentials (mepps) of the rat phrenic nerve-diaphragm preparation. The increase in the frequency of mepps induced by leptinotarsin is biphasic: about 10% of the total mepps are released in an initial burst that lasts about 90 sec, after which the remaining mepps are released over a period of 10-20 min. Tetrodotoxin has no effect upon the release induced by leptinotarsin, but low-Ca2+ conditions abolish the first phase. The two phases of release may represent two presynaptic pools of acetylcholine, both of which can be released in quantized form. In a second study, rat brain synaptosomes were incubated with [3H]choline and were immobilized on Millipore filters. Leptinotarsin induced release of [3H]acetylcholine from this preparation, confirming the release seen by using neurophysiological methods. The ability of leptinotarsin to induce release from either intact nerve terminals or synaptosomes was abolished when the toxin was heated. The releasing activity of leptinotarsin from synaptosomes was also partially dependent upon the presence of Ca2+ in the perfusing solution. Release from synaptosomes followed first-order kinetics, and was not inhibited by commercial antibodies to black widow spider antigens. The data suggest that leptinotarsin acts as a presynaptic neurotoxin and may be of value as a mechanistic probe in understanding the storage and release of neurotransmitters.

The mechanisms mediating the release of neurotransmitter have not been resolved. Very little is known about the sequence of events after the entrance of Ca2+ into the presynaptic terminal and the eventual release of neurotransmitter. One approach to this problem has been the identification of toxins that interact with the presynaptic terminal to modify the process of release. By examining the mechanism of action of the toxin, as well as the mechanism of the step(s) upon which the toxin acts, a more detailed description of the process of release may be obtained. Extracts of black widow spider venom glands (BWGE) will stimulate a massive increase in the frequency of miniature end plate potentials (mepps) from the frog neuromuscular junction (1). When Ca2+ is removed from the perfusing medium, BWGE were still able to stimulate release, although at reduced rates, in a variety of preparations (1-4). These data suggest that the action of BWGE was not simply to depolarize the membrane, thereby allowing Ca2+ to enter and elicit release of acetylcholine (AcCho), but was a more specific interaction with the release mechanism. Purification (5, 6) indicates that the protein responsible for the releasing activity in BWGE could not be recovered in large amounts, and suggests that the lack The publication costs of this article were defrayed in part by page charge payment. This article must therefore be hereby marked "advertisement" in accordance with 18 U. S. C. §1734 solely to indicate this fact. 1219

of availability of pure protein may limit the usefulness of the toxin as a mechanistic probe. The present study was directed towards the search for a neurotoxin which was available in more abundant quantities. Hsiao and Fraenkel (7) described the properties of a toxin, leptinotarsin, which is present in the hemolymph of the Colorado potato beetle, Leptinotarsa decemlineata. The toxin is lethal to both insects and vertebrates. It is heat labile and is believed- to be a protein with a molecular weight of about 50,000. Similar toxic activities have also been demonstrated by Hsiao (8) in the hemolymph of other Leptinotarsa species. We have examined the toxin derived from one of these species, L. hakdnani. In this communication we report that leptinotarsin§ induces the release of AcCho from neuronal terminals that are characteristic of both the peripheral and central nervous system.

METHODS Preparation of Synaptosomes. Immediately after decapitation, the forebrain of a male Sprague-Dawley rat of about 200 g body weight was removed and homogenized in 0.32 M sucrose (10 ml/g of forebrain). The homogenate was centrifuged (900 X g, 10 min), generating a supernatant that was decanted and further centrifuged (12,000 X g, 20 min). The resulting crude mitochondrial pellet (CMP) was washed twice with 0.32 M sucrose, but no further fractionation of the pellet was carried out. In this study the term "synaptosome" will refer to the CMP, and not to more purified preparations. Incubation of Synaptosomes with [3HJCholine. The CMP from one brain was resuspended, by being drawn several times through a pasteur pipette, in 20 ml of a physiological salt solution (PS) of the following composition: 142 mM NaCi, 4 mM KCI, 2 mM MgCl2, 2 mM CaC12, and 10 mM glucose in 25 mM Tris-HCl, pH 7.4. The salt solution used in "zero Ca2+" condition was modified by eliminating CaCl2 and increasing the concentration of NaCl to 148 mM. A solution of 10 gCi (1 Ci = 3.7 X 1010 becquerels) of [methyl-3H]choline chloride (New England Nuclear, 69.5 Ci/nmol) in ethanol was dried in a desiccator and taken up in 0.5 ml of PS. The resulting solution was then added to a vessel containing 2 ml of the synaptosomal suspension, after which the synaptosomes were incubated in a shaking water bath for 30 min at 37°C. Aliquots of the "loaded" synaptosomes were transferred to capped test tubes and maintained on ice until used. Our experience indicates that Abbreviations: AcCho, acetylcholine; BWGE, black widow gland extracts; CMP, crude mitochondrial pellet; mepp, miniature end plate potential; PS, physiological saline solution (chemical studies); TTX, tetrodotoxin. * Present address: Department of Biology, University of California, Los Angeles, CA 90024. § There are various forms of leptinotarsin (8); the toxin studied here is referred to as leptinotarsin-h in other papers.

1220

Proc. Natl. Acad. Sci. USA 77 (1980)

Neurobiology: McClure et al.

the degeneration of synaptosomes that is usually encountered can be greatly reduced at 40C, thereby allowing the use of the same batch for several assays over a period of 3-5 hr. Immediately prior to use, each aliquot was placed in a 370C bath for 3 min. Assay Procedure for Release of Radioactivity from Synaptosomes. Two hundred microliters of the rewarmed synaptosomal suspension were applied to a Millipore filter (AAWP 0025) held in a sampling manifold. The synaptosomes were washed repeatedly with 1.5-ml aliquots of PS to remove nonspecifically bound [3H]choline. Each wash was allowed to at 25-370C

remain in contact with the synaptosomes for 40

sec,

drawn

through the filter, and collected. Radioactivity in 1 ml of the filtrate was determined in a Searle Analytical Isocap 300 liquid scintillation spectrometer in a Triton X-114/xylene-based scintillation cocktail (9). Data were transferred on 8-channel punched paper tape to a Varian 620L minicomputer for analysis. Quenching was monitored by using the method of external standard ratios. Absolute efficiencies of counting ranged from 25% to 35%.

[3HJAcCho Determination. For those samples that were to be analyzed for AcCho content, 0.1 mM eserine sulfate (Sigma) was added to the PS. One hundred microliters of a solution containing 3.0 mg of AcCho and 0.60 mg of choline was added to each ml of the sample, after which AcCho was precipitated by the addition of 100 ,l of auric chloride (150 mg/ml, Alfa). After the mixture was centrifuged for 30 sec in a Microfuge (Beckman Instruments, Fullerton, CA) the supernatant, which contained most of the choline, was removed for analysis. In order to measure the radioactivity in these supernatants, the gold color had to be removed. This was accomplished by adding finely divided silver powder (about 50 mg/ml), heating for 5 min at 60-80'C, and mixing vigorously on a Vortex mixer (10). After cooling, an aliquot of the colorless supernatant was diluted into the scintillation cocktail of Anderson and McClure (9) and its radioactivity was measured. The pellet was resuspended in 1.2 ml of H20 and the same decolorizing process was applied. In a typical separation 80% of the AcCho was precipitated, while 95% of the choline remained in the supernatant. Standards of [3H]choline and [3H]AcCho were carried through the procedure to allow correction for the incomplete separation. Protein Determination. The lyophilized Sephadex G-200 fraction (25-35 mg) prepared by Hsiao (8) was dissolved in 1 ml of 25 mM Tris-HCI, pH 7.4, and dialyzed overnight against two changes of 150 ml each of PS. The sample was reclaimed from the bag and was clarified by 15 sec of centrifugation in a Microfuge. The concentration of protein in the supernatant was determined spectrophotometrically, assuming an extinction coefficient of 1 ml/mg-cm at 280 nm. Preparation of Antivenin to BWGE. One vial (6000 units) of lyophilized commercial antivenin to black widow venom (Lyovac; Merck, Sharp, and Dohme) was diluted with 3 ml of PS. An aliquot of 450 units (one unit will neutralize approximately 0.1 mg of protein from BWGE) was added to a solution of leptinotarsin. The neutralization mixture was incubated for 20 min at room temperature prior to assay. Measurement of mepps. The composition of the physiological saline solutions used for all electrophysiological analyses differed slightly from that used for chemical studies. For electrical recordings, the physiological saline was made up of 145 mM NaCI, 6 mM KCI, 1 mM CaC12, 1.2 mM MgSO4, and 5 mM glucose in 5 mM Tris-HCl at pH 7.4, whereas the composition of Ca2+-free, high Mg2+, physiological saline was 127 mM NaCl, 6 mM KCI, 12 mM MgSO4, and 5 mM glucose in 5 mM Tris-HCI at pH 7.4. For convenience the Ca2+-free highMg2+ solution will be referred to simply as Ca2+-free medium.

For experiments in which tetrodotoxin (TTX) was employed, TTX was added both to the bathing medium and to the toxin solution to be administered. The final concentration of TTX was 1 ,ug/ml. Samples of leptinotarsin that were used for electrophysiological studies were prepared as follows. Fifty mg of the lyophilized Sephadex G-200 fraction (8) was dissolved in 0.2 ml of cold H20 and dialyzed at 40C first against 5 mM Tris-HCl buffer at pH 7.4 and then against the desired physiological saline solution. The dialysate was centrifuged for 5 min in the Microfuge to remove a small amount of precipitate that formed and was either stored at 4VC or used immediately. To observe mepps, small strips of muscle were dissected from a rat diaphragm, pinned to a Sylgard block, and transferred to an experimental chamber of Lucite of 4 ml capacity. Cells were impaled with conventional fiber-filled glass micropipette electrodes filled with 3 M KCI and of approximately 15 MQ resistance. The mepps were amplified with a high-impedance preamplifier (Stoelting model PAD-2), displayed on an oscilloscope (Tektronix model 5103) and on a chart recorder (Brush-Gould model 220), and stored on instrumentation tape (Lockheed model Store 4) for subsequent analysis. Instantaneous mepp frequencies were obtained by replaying the tape through a storage oscilloscope (Tektronix model 5111), storing selected segments, and visually counting the total number of events in each segment. RESULTS Release of AcCho in the Peripheral Nervous System. Leptinotarsin stimulates a biphasic increase in the frequency of mepps at the rat neuromuscular junction. The control frequency of mepps, observed immediately after penetration of a suitable endplate, was about 5 Hz (Fig. la). Leptinotarsin was applied to the preparation by adding a small drop of solution directly above the site of impalement. We elected to use this procedure, rather than perfusing with a solution containing the toxin, to conserve material. There was no immediate change in the frequency of mepps after leptinotarsin was introduced into the recording chamber. After a delay that varied from about 10 to 25 min, the frequency of mepps rose dramatically (Fig. lb) to a maximum (Fig. ic), which was maintained for only a few seconds before the frequency fell to a much lower value (Fig. ld), and release began a second, slower, phase. In the second phase the frequency of mepps again increased (Fig. le), but it never reached the value observed at the peak of the first phase. After a period of 10-15 min, the second phase also ceased, and the frequency of mepps fell to zero (Fig. if). The time course of release, measured from the time of initial application of toxin, varied between 20 and 40 min from one experiment to the next. We believe that much of the experimental variability is due to differences in the effective concentration of the toxin at the endplate involved. In support of this suggestion, the delay prior to the first phase was longer (about 45 min) in one case in which leptinotarsin was inadvertently added at some distance from the recording electrodes. It is also possible that the rate of change of concentration is important, so that applications close to the endplate shorten the overall time course due to the relatively rapid increase in the concentration of toxin that occurs while release is still taking place. Irrespective of the variable duration of release, the biphasic nature of the curve was observed in each of over a dozen experiments. In addition, the relative heights of the peaks of the two phases were nearly the same from one experiment to the next. The time course from an experiment of about average duration is presented in Fig. 2. To examine the effect of Ca2+ on release induced by leptinotarsin, a nerve-muscle preparation was bathed in a saline solution that lacked Ca2+. Lowering the

Neurobiology:

McClure et al.

I FIG. 1. Oscilloscope records of mepps at the rat neuromuscular junction stimulated with leptinotarsin. Records were taken immediately after penetration of a muscle cell (control, a), during the rising limb of the first phase of release (b), at the peak of the first phase (c), in the valley between the two phases (d), at the peak of the second phase (e), and after the second phase was over (f). Calibration: 0.2 mV, 0.2 sec.

concentration of Ca2+ had no effect upon the control frequency of mepps, which was measured prior to the addition of leptinotarsin. When toxin was introduced into the bath, there was again a variable time period before an increase in the frequency of mepps was observed. In the case of Ca2+-free media, no brief intense first phase of release was seen. There was still a broad phase of massive release that occurred several minutes after the first increase in mepps was seen. Both the duration and time to peak frequency of the massive release were variable, but fit

Proc. Natl. Acad. Sci. USA 77 (1980)

1221

well with the range of values observed for the second phase of release in media containing Ca2+. The fastest releases seen in Ca2+-free media were never as fast, nor did they reach as high frequencies, as did the brief intense initial phase observed in normal media. A typical time course of release in a Ca2+_free medium is presented in Fig. 2. The actual experiment of Fig. 2 was selected to have approximately the same duration as the experiment in Ca2+-containing media. Because time courses obtained in both the presence and absence of Ca2+ are variable, the agreement of the two experiments presented in Fig. 2 does not require that the slower phase be independent of Ca2 . Nonetheless, it is our feeling that the release in Ca2+-free media is very similar to, and may be the same as, that observed in the slower phase in the presence of Ca2+. To clarify the role of propagated action potentials in the release stimulated by leptinotarsin, we examined the effect of pretreatment of the preparation with TTX. The biphasic character of the leptinotarsin-induced release is preserved in the presence of TTX (not shown), although the second phase may be abbreviated (duration less than 2 min). Because Hsiao (8) found that the fly-killing activity of leptinotarsin was abolished when the toxin was heated prior to testing, we examined the effect of heat on leptinotarsin's ability to stimulate release of mepps at the neuromuscular junction. A solution of leptinotarsin was heated at 60'C for 5 min and cooled before application to the preparation. The frequency of mepps was not noticeably altered by the heated preparation. The total number of mepps released by leptinotarsin was obtained by integrating the instantaneous frequency with respect to time. Although during release there was some loss of resting potential, which resulted in a corresponding decrease in the amplitude of the mepps themselves, the disappearance of the mepps was not due to their loss in the noise but rather to the complete failure of their occurrence. Counts of mepps for three experiments are given in Table 1. Although the first phase has the higher frequency of mepps, it contributes only about 10% to the total count. The data suggest that leptinotarsin stimulates the release of AcCho from two pools. The first of these is emptied quickly and contains 10% of the total quanta, whereas the second is depleted more slowly of the remaining 90%.

Release of AcCho in the Central Nervous System. The previous data define an action of leptinotarsin at peripheral synapses. To extend the action of the toxin to the central nervous system, we examined the ability of leptinotarsin to stimulate release of AcCho from synaptosomes. The properties of central synapses, as reflected by synaptosomes, may vary significantly from those of the neuromuscular junction. Synaptosomes were prepared from rat brain, labeled with [3H]choline, immobilized on Millipore filters, and washed. A solution of leptinotarsin was allowed to cover the bed for 2 min, drawn through the filter, and collected. Two more washes, each 40 sec in duration, were Table 1. Numbers of mepps released from terminals of the phrenic nerve after treatment with leptinotarsin Total mepps Fraction count in first peak Exp.

Time,

sec

FIG. 2. Time course of the release of AcCho from the rat neuromuscular junction stimulated with leptinotarsin. Zero time is arbitrarily taken at the onset of release. Experiments were conducted in control saline solution (0) and in a saline lacking Ca2+ (0).

1 2 3

117,000 135,000 120,000

0.095 0.094 0.119

Mean + SD 124,000 ± 9600 0.103 ± 0.014 Individual counts are given for the total mepps in both phases of three experiments.

1222

Neurobiology: McClure et al.

Proc. Natl. Acad. Sci. USA 77 (1980) Table 3. Effect of temperature and Ca2+ on leptinotarsininduced release from synaptosomes Index of release Condition

0.31 ± 0.02 Control 0.21 ± 0.01 Zero Ca2+ 0.00 Heated* All three experiments contained leptinotarsin at an initial concentration of 10 Ag/ml. Values represent the mean ± SEM of three samples. * 100'C, 10 min.

1

2 3 4

5 6 7 8 9 1U F iltrate

FIG. 3. Release of radioactivity from synaptosomes labeled with

[3H]choline, immobilized on Millipore filters, and stimulated with

leptinotarsin. Solutions containing the toxin at 10 ,g/ml were added at the arrow. Release was followed in two successive washes (solid line), and was compared to controls lacking leptinotarsin and conducted in parallel (broken line). After the 10th filtration, the synaptosomes were lysed with distilled water (hatched bar).

collected individually to determine if release continued after an initial exposure to the toxin. Finally, the synaptosomes were lysed with water, and the lysate was collected. A profile of the levels of radioactivity in the collected fractions from a typical experiment is presented in Fig. 3. The amount of radioactivity fell with each wash, and, in controls, dropped to low levels in washes after the eighth. When the eighth wash was replaced by a solution containing leptinotarsin, an increase in the levels of radioactivity was seen. Release induced by the toxin continued for a few minutes, as indicated by the succeeding washes, which also had levels of radioactivity in excess of controls. In this preparation, the released radioactivity could represent choline, AcCho, or a mixture of both. To determine whether leptinotarsin caused a specific release of either of these two compounds, experiments were carried out in the presence of an anticholinesterase, and the released radioactivity was separated into AcCho-rich and choline-rich fractions by precipitation as the auricchloride complex (Table 2). Leptinotarsin stimulated release of material that was rich in AcCho. Several characteristics of the leptinotarsin-induced release of radioactivity from synaptosomes were examined. The results of these experiments are expressed as an index of release (i), which was calculated by summing the cpm in filtrates 8, 9, and

10 and dividing by the sum of the cpm in filtrates 8, 9, 10, and 11 (see Fig. 3). By including the radioactivity released by lysis, we were able partially to account for variations from one preparation of synaptosomes to the next. Heating the solution of leptinotarsin for 10 min in a bath maintained at 1000C abolished releasing activity (Table 3). To determine the Ca2+ dependence of the release induced by the toxin, Ca2+ was removed from the PS, from the washes of the synaptosomal bed, and from the solution containing the toxin. Releasing activity appears to be partially Ca2+ dependent (Table 3). The effect of varying concentration of toxin was also examined. The release appears to saturate at a concentration of leptinotarsin of 40 .tg/ml, which generates an index of release of 0.60 (not

shown). Time Course of Release of Radioactivity from Synaptosomes. In order to follow the time course of release, a slightly modified experiment was carried out. Synaptosomes were prepared and loaded with [3H]choline as described in Methods. The synaptosomal suspension was then recollected as a pellet, washed twice with PS to remove adventitious [3H]choline, resuspended, and warmed for 3 min at 370C. At this time enough toxin was added to bring the final concentration of leptinotarsin in the suspension to 40 ,ug/ml. At designated time intervals volumes of 200 ,ul of the synaptosomal suspension were removed, diluted into a solution of 1.5 ml of PS previously placed on a Millipore filter, and immediately drawn through the filter. Radioactivity released due to leptinotarsin rose monotonically toward a plateau (Fig. 4). The time points of Fig. 4 represent the moment the sample was withdrawn from the reaction vessel. There was, however, a time lag between removal of the sample and collection of the filtrate. We estimate

Table 2. AcCho and choline released from synaptosomes by leptinotarsin AcCho, Choline, AcCho, % of total cpm Condition cpm cpm 41 1085 796 Leptinotarsin 26 666 224 PS alone Leptinotarsin - PS 58 419 572 Values result from assay of the eighth filtrate (see Fig. 3). The concentration of leptinotarsin was 20 ug/ml. n = 2. Temperature = 250C.

Time, sec

FIG. 4. Time course of the release of radioactivity from synaptosomes previously incubated with [3H]choline and stimulated with leptinotarsin at 40 ,g/ml.

Proc. Natl. Acad. Sc. USA 77 (1980)

Neurobiology: McClure et al. that this lag was about 5 sec, which would not seriously change the results. The data of Fig. 4 yield a straight line when cast in appropriate semilogarithmic coordinates, indicating that release of total radioactivity follows first-order kinetics. Immunological Studies. Because both leptinotarsin and BWGE were able to cause an increase in the frequency of mepps, the ability of antibodies to black widow antigens to crossreact with leptinotarsin and to abolish the releasing activity was examined. Leptinotarsin was mixed with commercial antivenin, incubated at room temperature to allow reaction to take place, and assayed using the synaptosomal preparation. This treatment did not impair the ability of leptinotarsin to stimulate release from synaptosomes (data not shown).

DISCUSSION The present study has characterized the toxin leptinotarsin, whose action is similar to, but not identical with, that of BWGE. When applied to the neuromuscular junction of a rat, leptinotarsin stimulates an increase in the frequency of mepps. Unlike BWGE, which causes an initial massive outpouring followed by a slow decline, leptinotarsin stimulates an increase in frequency that follows a biphasic pattern. The biphasic release, although variable in its details (see Fig. 2), has a basically reproducible pattern which suggests that it is controlled by some threshold event which, once set in motion, elicits the full-blown epoch of release. Although it is possible that the biphasic release is due to two toxic components in the leptinotarsin, each with its own kinetic characteristics, it is also possible that the biphasic curve reflects separate components of the nerve terminal that are brought into sequential activity by a single toxic component of leptinotarsin. It is possible that the calcium-sensitive highfrequency first phase might correspond to release from an easily mobilizable pool of the transmitter, whereas the second, broader, calcium-insensitive phase could represent release from a larger, less mobile, pool. In addition to the toxin(s) that stimulates release of AcCho, these preparations of leptinotarsin probably contain an activity that prevents release of transmitter. This conclusion rests upon the fact that the frequency of mepps falls eventually to zero, rather than becoming stabilized at a higher level that would be maintained indefinitely if control frequencies were reestablished after treatment with toxin. Whether the toxin that causes complete cessation of mepps is related to the toxin(s) that stimulates release or is due to a separate molecule is not yet known. The physiological data suggest that leptinotarsin is not identical to previously described presynaptic neurotoxins. Other agents that stimulate release, such as f3-bungarotoxin (11) and crotoxin (12), rarely stimulate frequencies of mepps over 100 Hz, and do not demonstrate a biphasic release. Additionally, the ability of either f3-bungarotoxin or crotoxin to stimulate release was abolished when Ca2+ was removed from the perfusing medium. Other studies (13, 14) suggest that the Ca2+dependent release of these two toxins may be explained by phospholipase A activity. Extracts of the venom glands of the black widow do stimulate high frequencies of mepps at the neuromuscular junction, but they also fail to generate biphasic release in a single experiment. BWGE may show two quantized

1223

releasable pools of AcCho, however, because frog muscle poisoned with botulinum toxin can still be stimulated by the extract to release AcCho in a quantized fashion (15). The relationship, if any, between the pools detected with BWGE and those observed after treatment with leptinotarsin remains to be established. In order to confirm at a central synapse the release of AcCho observed at the neuromuscular junction, synaptosomes immobilized on Millipore filters were examined. Prior to immobilization, synaptosomes were incubated with [3H]choline to label the intracellular pools of choline and AcCho. When a solution of leptinotarsin was applied to this preparation, radioactive material was released. Fractionation of the released radioactivity indicated that 58% of the material was [3H]AcCho. These values agree well with those observed when release was stimulated with high concentrations of K+. The ability of leptinotarsin to stimulate release from synaptosomes was dependent upon the concentration of leptinotarsin, followed firstorder kinetics, and was partially dependent upon the presence of Ca2 . Additionally, releasing activity was abolished by heating, which is in agreement with data of Hsiao (8) and which supports preliminary fractionation data in suggesting that this toxin is a protein. Leptinotarsin is a neurotoxin that appears specifically to stimulate release of AcCho both from synaptosomes and at the rat neuromuscular junction. Preliminary purification studies, using release from synaptosomes as an assay, suggest also that leptinotarsin will be available in much greater quantities than is the toxin from the black widow. Further neurochemical and physiological studies will be facilitated when leptinotarsin is purified to homogeneity. This research was supported by the Allan Hancock Foundation, the National Science Foundation (NS 76-80657, 77-06782), and the Nelson Research and Development Corporation. 1. Longenecker, H. E., Jr., Hurlbut, W. P., Mauro, A. & Clark, A. W. (1970) Nature (London) 225, 701-703. 2. Paggi, P. & Toschi, G. (1972) Life Sci. 11, 413-417. 3. Pumplin, D. W. & McClure, W. 0. (1977) J. Pharmacol. Exp.

Ther. 201,312-319. 4. Baba, A., Sen, I. & Cooper, J. R. (1977) Life Sci. 20, 833-842. 5. Pumplin, D. W. (1973) Dissertation (Univ. of Illinois, Urbana,

IL). 6. Frontali, N., Ceccarelli, B., Gorio, A., Mauro, A., Siekevitz, P., Tzeng, M. & Hurlbut, W. P. (1976) J. Cell. Biol. 68,462-479. 7. Hsiao, T. H. & Fraenkel, G. (1969) Toxicon 7, 119-130. 8. Hsiao, T. H. (1978) in Toxins: Animal, Plant and Microbial, ed. Rosenberg, P. (Pergamon, Elmsford, NY), pp. 675-688. 9. Anderson, L. E. & McClure, W. O: (1973) Anal. Biochem. 51, 173-179. 10. Collier, B. & Lang, C. (1969) Can. J. Physiol. Pharmacol. 47, 119-126. 11. Chang, C. C., Chen, T. F. & Lee, C. Y. (1973) J. Pharmacol. Exp. Ther. 184, 339-345. 12. Chang, C. C. & Lee, J. D. (1977) Naunyn-Schmiedebergs Arch. Pharmakol. 296, 159-168. 13. Sen, I. & Cooper, J. R. (1978) J. Neurochem. 30, 1369-1375. 14. Hendon, R. A. & Fraenkel-Conrat, H. (1971) Proc. Natl. Acad. Sci. USA 68,1560-1563. 15. Pumplin, D. W. & del Castillo, J. (1975) Life Sci. 17, 137-142.