University of Southern California Dissertations and Theses

Studies of two naturally occurring compounds which effect release of acetylcholine from synaptosomes

(USC Thesis Other)

Studies of two naturally occurring compounds which effect release of acetylcholine from synaptosomes

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content STUDIES OF TWO NATURALLY OCCURRING COMPOUNDS WHICH EFFECT RELEASE OF ACETYLCHOLINE FROM SYNAPTOSOMES by Michael Leo Koenig A Dissertation Presented to the FACULTY OF THE GRADUATE SCHOOL UNIVERSITY OF SOUTHERN CALIFORNIA In Partial Fulfillment of the Requirements for the Degree DOCTOR OF PHILOSOPHY (Biology/Neurobiology) October 19 85 UNIVERSITY OF SOUTHERN CALIFORNIA THE GRADUATE SCHOOL UNIVERSITY PARK LOS ANGELES, CALIFORNIA 90089 This dissertation, written by Michael Leo Koenig under the direction of . . . . Dissertation Committee, and approved by all its members, has been presented to and accepted by The Graduate School, in partial fulfillment of re quirements for the degree of DO CTOR OF PHILO SO PHY K 7 f of Graduate Studies Date . . . . O c t o b e r ^ 2 9 , ^^ 1 9 8 5 DISSERTATION COMMITTEE Chairperson TABLE OF CONTENTS LIST OF TABLES .......... vii LIST OF FIGURES .................... viii ABSTRACT ................. xi I. INTRODUCTION .................. 1 Hypotheses to Account for the Release of Acetylcholine .............. 1 The Role of Calcium Ions in Release of Neurotransmitter ......... 13 The Use of Naturally Occurring Neurotoxins as Probes of ACh Release ....... 22 Toxins Which Inhibit Release.......... 23 Toxins Which Facilitate Release ........... 26 Natural History of lotrochota and Leptinotarsa . 36 1. lotrochota birotulata ........ 36 2. Leptinotarsa decemlineata ................. 37 Rationale for Research ............ 39 II. MATERIALS AND METHODS ....... 42 Animals ................ 42 Chemicals ...... 42 So lut ions ................... 43 Preparation of Rat Forebrain Synaptosomes ...... 44 Preparation of Synaptosomes from Squid Optic Lobes 48 Preparation of Synaptosomes from Electric Organ 48 3 Loading of Synaptosomes with [ H]Choline....... 49 q Loading of Synaptosomes with 2-[ H]Deoxyglucose. 50 11 Procedure Used for Assaying Release of Radioactivity .......................................... 51 Separation and Determination of [ H]Choline and [ H] Ace ty Ichol ine ................................ 52 Assays of Cytoplasmic Markers ........... 54 1. Lactate Dehydrogenase ...................... 54 2. Choline Acetyltransferase ............. 55 3. Adenosine Triphosphate .............. 55 Determination of Protein ............................ 56 Chromatographic Techniques .......................... 57 Concentration of Toxic Fractions ............... 57 Derivatization Procedures ........................... 58 1. Preparation of Fluorescein Derivatives . 58 12 5 2. Preparation of [ I] Derivatives ...... 59 Polyacrylamide Gel Electrophoresis ............... 61 1. Discontinuous Slab Gel Electrophoresis . 61 2. Two-Dimensional Gel Electrophoresis .... 62 Autoradiography ............ 63 Electrophysiology ............... 64 Receptor Binding Studies ............................ 65 III. RESULTS; Purification and Characterization of lotrochotin ......................................... 69 Activity of Exudate vs. Whole Tissue Homogenate. 69 Molecular Sieve Chromatography .................... 71 Anion Exchange Chromatography ...................... 74 Affinity Chromatography ............................. 75 ... I 111 High Performance Liquid Chromatography ...... 78 Sodium Dodecy1sulfate-Polyaerylamide Gel Electrophoresis ........................................ 81 Summary of the Purification of lOT ............ 84 Stability of lOT ........... 86 Time Course of lOT-Induced Release ............... 87 Effects of Varying the Concentration of lOT on Release ........... 90 Calcium Dependence of Release ...................... 93 Effect of lOT on Miniature End Plate Potentials at the Frog Neuromuscular Junction ............... 96 Release of Typical Cytoplasmic Markers ...... 97 3 Release of 2-[ H]Deoxyglucose .................. 102 Binding of lOT to Synaptosomal Membrane ..... 105 Comparison of the Effects of lotrochotin and Digitonin on Release of Cytoplasmic Markers .... 110 Effects of Synaptosomal Pre-Depolarization on lOT-lnduced Release .................................. Ill IV. RESULTS: Purification and Characterization of LPT-d .................................................. 117 Molecular Sieve Chromatography .................. 117 Anion Exchange Chromatography ........... 120 Cation Exchange Chromatography .................... 121 Chromatof ocusing ........... 126 Dye-Ligand Affinity Chromatography ............... 129 High Performance Liquid Chromatography ....... 133 SDS-Polyaerylamide Gel Electrophoresis .......... 134 Summary of the Purification of LPT-d ............. 137 IV Stability of LPT-d ....................... 140 Time Course of Release .................... 143 Release of Radioactivity at Varying Concentrations of LPT-d............................. 148 Calcium Dependence of Release .................... 151 Sodium Dependence of Release ............ 136 The Effects of Multivalent Metal Cations on LPT-d-Induced Release ................................ 160 The Effect of Verapamil on LPT-d-Stimulated Release ........... 164 Effects of Dihydropyridines on LPT-d-Stimulated Release .................................................. 166 Effect of Maitotoxin on LPT-d-Stimulated Release 173 Effects of LPT-d on the Calcium Channels of Other Tissues .......................................... 176 1. Paramecium caudatum ........................ 176 2. Squid Optic Lobe Synaptosomes ........... 17 8 3. Smooth Muscle «.............................. 179 4. Skeletal Muscle ............................. 182 5. Neuronal Clonal Cell Lines ............... 184 6. Isolated Diencephalon Cells .............. 185 7. Mouse and Bovine Brain Synaptosomes .... 187 Binding of LPT-d to Synaptosomal Membrane Receptors ............................................... 188 Effects of LPT-d on Phosphorylation of Synaptosomal Proteins ................................ 200 V. DISCUSSION; Purification and Characterization of l O T ....... 204 L . V VI. DISCUSSION: Purification and Characterization of LPT-d ...... 214 VII.BIBLIOGRAPHY ............................................. 227 VI LIST OF TABLES Table Page 1 RELEASE PROMOTING ACTIVITY OF lOTROCHOTA BIROTULATA EXTRACTS 7 0 2 SUMMARY OF lOT PURIFICATION.............. 85 3 SUMMARY OF LPT-d PURIFICATION......... 141 Vll LIST OF FIGURES F igure Page 1 Chromatography of lotrochotin on Sephadex G-50 ____________________________________ ____________ 72 2 Chromatography of lotrochotin on Cellex D ... 76 3 High Performance Liquid Chromatography of lotrochotin on SynchroPak AX-300 .............. 79 4 Sodium DodecyIsulfate-Polyacrylamide Gel Electrophoresis of lotrochotin .......... .. 82 5 Time Course of lotrochotin Induced Release of Radioactivity .................... ................. 88 6 Effect of Varying the Concentration of lotrochotin on Release of Radioactivity ..... 91 7 Effect of Varying the Extrasynaptosomal Concentration of Calcium on Release of Radio activity Induced by lotrochotin...... ........ 94 8 Effects of lotrochotin and Elevated on Release of Lactate Dehydrogenase from Synaptosomes ......................... 99 9 q Release of 2 -[ H]Deoxy-D-Glucose by lotrochotin .............................. 103 10 Binding of Native lotrochotin to Crude Synaptosomal Membrane ............................ 106 11 Binding of Native lotrochotin to Purified Synaptosomes ....................................... 108 12 Comparison of the Effects of lotrochotin and Digitonin on Release of Cytoplasmic Markers . 112 13 Effect of Prior Depolarization of Synapto somes on Release Induced by lotrochotin ..... 115 14 Chromatography of Leptinotarsin-d on Sephadex G-150 . ............................................. 118 15 Chromatography of Leptinotarsin-d on DEAE- Sephadex ............................. . ............. 122 viii 16 Chromatography of Leptinotarsin-d on CM- Sephadex............................................ 124 17 Chromâtofocusing of Leptinotarsin-d ...... 127 18 Chromatography of Leptinotarsin-d on Dye- Ligand Affinity Matrices ...... .................. 130 19 High Performance Liquid Chromatography of Leptinotarsin-d on SynchroPak AX-300 ........ 135 20 Sodium DodecyIsulfate-Po1yacrylamide Gel Electrophoresis of Leptinotarsin-d ........ 138 21 Stability of Leptinotarsin-d Purified Through the CM-Sephadex Step ................ 143 22 Time Course of the Release of Radioactivity Stimulated by Leptinotarsin-d ............... 146 23 Effect of Varying the Concentration of Leptinotarsin-d on the Release of Radioactivity .............................. ........ 149 24 Calcium Dependence of Leptinotarsin-d- Induced Release .............. ..................... 152 25 Effect of Calcium Chelation with EGTA on Synaptosomal Release of Radioactivity Stimulated by Leptino|atsin-d and by Depolarization with K ...................... 154 26 Effects of TTX and the Removal of Na^ Ions on Release Stimulated by Leptinotarsin-d .... 158 27 Effects of Multivalent Metal Cations on Leptinotarsin-d Promoted Release of Radio activity from Synaptosomes ............. 162 28 Effects of Verapamil on Release Stimulated by Leptinotarsin-d and by Depolarization with K ....................................................... 167 29 Effects of Nifedipine on Release Stimulated by Leptinotarsin-d ......... ..................... 171 30 Effect of Nitrendipine on Release Stimulated by Leptinotarsin-d ..................... 174 31 The Effect of Leptinotarsin-d on Release of ix Radioactivity from Synaptosomes Derived from Squid Optic Lobes ................... 180 32 Nonspecific Binding of lodinated Leptinotarsin-d ....... 190 33 Binding of lodinated Leptinotarsin-d to Trans-blotted Synaptosomal Membrane Proteins. 193 34 Binding of Native Leptinotarsin-d to Purified Synaptosomes ....... 196 35 The Time Course of Binding of Native Leptinotarsin-d to Purified Synaptosomes .... 198 36 The Effect of Leptinotarsin-d on Phosphoryla tion of Synaptosomal Proteins ............... 201 ABSTRACT Two naturally occurring compounds which effect the release of neurotransmit ter from synaptosomes have been purified to apparent homogeneity. lotrochotin (lOT) isolated from wound exudate of the Caribbean purple bleeder sponge promotes release in a manner that is 2 + independent of the extracellular Ca ion concentration, Leptinotarsin (LPT-d), a protein taken from hemolymph of the Colorado potato beetle, 2 + Leptinotarsa decemlineata. stimulates Ca -dependent release. lOT, purified by conventional chromatographic techniques, migrates as a single band on sodium dodecyIsufate polyacrylamide gel electrophoresis (SDS-PAGE) slab gels. The protein is slightly acidic and has a molecular weight of approximately 18 kD. 3 [ H]AcetyIcho1ine which has been introduced into synaptosomes as [ H]choline can be released by lOT. The toxin releasable pool of labelled neurotransmitter is not depleted by depolarization of the synaptosomes with high potassium, and therefore seems to be primarily extravesicular. That LOT has membrane permeabi1izing properties is suggested by the finding that small cytoplasmic molecules can be co-released with neurotransmitter. Adenosine triphosphate and exogenously xi toxin, whereas larger cytoplasmic marker enzymes are not- L'PT-d is a larger protein (molecular weight = 45 kB ) than TOT, and seems to effect primarily vesicular release 2 + by opening at least one type of presynaptic Ca channel* The facilitatorv effects of the toxin on synaptosomal release can be inhibited by inorganic 2 + Ca channel antagonists, but are not generally affected by organic antagonists thought to be most 2 + effective as inhibitors of slow Ca channels. There is some evidence that the organic antagonist nifedipine. if added after LPT-d. can close the toxin-activated channels in cultured rat diencephalon cells- If so- the channels affected may be homologous with those found in smooth muscle and heart. Studies of the effects of LPT-d 2 + on Ca channels from a variety of tissues including Paramecium- squid optic lobe- smooth and skeletal muscle, and neuronal clonal cell lines, have shown the toxin to be an activator specifically of a type of channel found in the mammalian central nervous system. XI 1 I INTRODUCTION Hypotheses t o Ac coun t for the Release of Acetylcholine The general process by which the neurotransmitter , acetylcholine (ACh) is released from presynaptic nerve ■ terminals has been known for some time and is widely accepted. Depolarization of the nerve terminal, resulting typically as a consequence of spike activity, causes calcium channels in the membrane to open briefly allowing I ' 2 + j Ca ions to enter. The suddenly increased 2 + I concentration of intraterminal Ca then effects a quantized release of ACh which can be measured, at I neuromuscular junctions, spontaneously as individual ! miniature end-plate potentials (mepps; Fa 11 and Katz, 1952 ) I or, with stimulation, as end plate potentials (epps) which I i are integral multiples of mepps. Presumably, quantized release is a general characterisitic of stimu 1us-evoked I secretion at chemical synapses including those of the ■ central nervous system (CNS). I In many, if not most neuronal tissues, there is also I evidence that release of acetylcholine can occur in a I nonquant al manner. This is particularly true when nerve I cells are not "firing" (i.e. under "resting" conditions j when as much as 90% of the ACh release may be attributable I simply to leakage (Kelly et al., 1983). Spontaneously ■ occurring nonquant al release may even be the predominant l i form of release under some circumstances such as when I target cells are relatively far from the presynaptic j terminals (Vizi et al ., 1 982) . Nonquant al release is characteristic of both developing neurons (Sun an d Poo, ^ 1983; Thesleff and Molgo, 1983), and those that are regenerating (Volkov and Poletaev, 1982; Volkov and Frosin, 1984). Although there is little argument that release of ACh, , like that of other neurotransmitters, occurs in this way, j the actual mechanism of release remains a subject of I considerable controversy. Two principal hypotheses have I I been advanced to account for experimental findings. The : i ' earlier, and more widely accepted, theory contends that ACh lis specifically packaged into and subsequently released j ! from small (approximately 50 nm dia.) vesicles contained ; i within the nerve terminal. A more recent argument suggests i jinstead, that release occurs primarily from the cytoplasmic; pool of ACh, and that the vesicles may serve as either 1 2 + [reservoirs of neurotransmitter or as Ca -sequestering organel1 es. I The vesicle hypothesis, as originally formulated by !de1 Castillo and Katz (1 957 ), was based on the need to find ,a way of accounting for the quant al nature of release at neuromuscular junctions (Fatt and Katz, 1952). With the ■discovery that large numbers of vesicles were present I .within nerve terminals (De Robertis and Bennett, 1954; ‘ Palade and Pal ay, 1 954), it was suggested that transmitter ^ I ! contained within the vesicles could be released by a process of exocytosis, a process known to be common in other secretory cells. Release of the contents of one vesicle, as the result of a "critical collision" with the terminal membrane (Katz, 1969), would account for the ! ' presence of a mepp; a simultaneous exocytotic release of j many vesicles, such as could be expected with appropriate , I stimulation, would explain the quantal nature of epps. I I I There are many experiments which support the vesicle I ! I j hypothesis of release. It does provide a satisfactory j explanation of the quantal nature of release, and isolated synaptic vesicles do contain very high concentrations of j I ; ! ACh relative to the cytoplasmic compartment (see Wagner et , I al . , 1 978). Furthermore, freeze-fracture studies of the . frog neuromuscular junction (NMJ) show that "dimples" and j I : "protuberances" are formed in the presynaptic membrane | I following nerve stimulation, and these "omega profiles" | have been correlated with exocytotic release (Heuser et j I al ., 1 974 in Ceccarelli and Hurlbut, 1 980). Additionally, ; ! I I antibodies directed against synaptic vesicle-specific ' antigens are transferred to frog nerve terminal membranes I ■ 3 ^ I after La -stimulation of quantal release (von Wedel, I et al ., 1981) . Finally, vesicular components other than ' ACh can be released simultaneously with the neurotransmitter (Zimmermann and Whittaker, 1974), and 3 J [ marker proteins present in the synaptic cleft are I incorporated into presumably recycling vesicles (Holtzman I et al., 1971; Ceccarelli and Hurlbut, 1975). There is also evidence that prolonged stimulation of neurons can result in a-significant decrease in the number . and density of synaptic vesicles with a concomitant | ' increase in the area of presynaptic membrane (Wickelgren et I al . , 1 985). Other experiments have shown that when ACh is ' prevented from entering vesicles (e.g. with AH518 3; Melega I and Howard, 19 8 4; Edwards et al., 1985), evoked release is j j ! , effectively blocked, and, when false transmitters incapable I I i of entering the vesicles are present, these usually are not ! released upon neuronal stimulation (Boksa and Collier, j 1980; Weiner and Collier, 1984). I j Since other secretory systems appear to effect release' I by exocytotically discharging their vesicular contents, it ; is attractive to think that release of neurotransmitter I I follows the same pattern. Indeed, Kelly et al. (1983) point out that "it is intrinsically unlikely” that neuronal I I release is based on a unique mechanism. It has also been j j suggested however, that the very rapid phasic release of I nerve terminals must be different from the less rapid, more ! ' sustained release characteristic, for example, of hormonal systems (Cooper et al., 1982). A great number of experimental findings argue against j the vesicle hypothesis. Whereas ACh is concentrated inside 4 r synaptic vesicles, it is also present in the cytoplasm, and! the latter pool represents at least one half of the total I ' neurotransmitter present in nerve terminals (Tauc, 1979; Jope, 1981; Dunant and Israel, 1985). Experiments with rat brain synaptosomes (Jope, 1981), Torpedo electric organ (Dunant et al., 1982), and the buccal ganglion of Aplysia calif o rni c a (Tauc and Baux, 1 985), have consistently shown ' that it is the newly-synthesized "free” ACh contained | within the cytoplasmic pool that is released i preferentially. Furthermore the turnover, or , ' replenishment, of ACh is more rapid in this compartment j ' i (Dunant et al., 1977). Some have suggested that the filling of vesicles with transmitter would be a I prohibitively slow process in the maintenance of the rapid | I ' firing rate characteristic of many neurons (Cooper et al., i ' 1 982) . A second argument against vesicular release is that ! there is not always a good correlation between the number i of vesicles remaining after neuronal stimulation and the ! degree of depletion of tissue ACh stores. In the Torpedo j electric organ, for example, Israel et al. (1978) found j that, after a prolonged afferent stimulation (10 Hz; 5-10 I min.) sufficient to reduce the electrical discharge by more j than 90%, the number of vesicles remained constant and the amount of vesicular ACh was unchanged. Similar findings have been reported by Ceccarelli et al., 1 978 in Israel et 5 ' al . , 1 979 ) for the frog cutaneous pectoris ne rve-mu s c 1 e j ' preparation. Although stimulation for four hours at two Hz ' exhausted transmission, no significant changes were found ; in the number of vesicles. That ACh release can occur even in the complete ' absence of vesicles has been demonstrated in many differentj preparations. Landmesser and Pilar (1972) showed that ' I chick ciliary ganglion cells were fully transmitting at : embryonic stage 33.5, even though no vesicles were present j j at that time. Release has also been characterized as emanating from cell bodies and dendrites (Tauc, 1982) as I ' well as from non-neuronal Schwann cells (Dennis and Miledi, ! j 1974). Furthermore two groups, almost simultaneously, I showed that ACh could be released from proteoliposomes I i ' 1 derived from synaptosomal membrane (Israel et al., 1983; Meyer and Cooper, 1983). Since these artificially developed synaptosomal "ghosts" contained no vesicles, the ! released ACh could only have come from cytoplasmic stores. Although the stimulation-induced appearance of dimples and pits in the nerve terminal membrane may well be attributable to vescular exocytosis, opponents of vescular j release argue that the electron microscopic figures may I just as well represent endocytotic events (Macintosh, 1978; Tauc, 1979). In that sense, the incorporation of j extracellular markers, cited as evidence for vesicle j recycling, may simply be reflective of endocytosis. If it I ' 6 I is assumed that the figures do represent exocytosis, the ' time of appearance does not correlate well with release. ; As determined by rapid freezing experiments (Heuser et al., ' 1979; Heuser and Reese, 1981), the pits most often appear , two to three msec after release should have occurred or in considerably lower numbers than had been predicted (Tauc, ; 1982; Kelly et al., 1983; Dunant and Israel, 1985). i ; Experiments evaluating the releasability of false i transmitters that have been incorporated into synaptic vesicles also argue against the sole involvement of I vesicles in the release process. O'Regan et al . ( 1 982) ' have shown, for example, that propionylcholine, although acetylated and taken up into the vesicles of Torpedo I ; electric organ as readily as is ACh, is not as releasable. : Experiments with other false transmitters have produced i results which also contradict the vesicle hypothesis. Large and Rang (1978a,b) showed that the release of j acetylmonoethylcholine, measured in the rat phrenic nerve-diaphragm preparation, occurred in a way that could | ! i [ not be explained if ACh were parcelled into vesicles as ! distinct and independent quanta (Israel et al ., 1979 ). j One of the arguments used in support of the vesicle i hypothesis has also been used against it. Whereas vesicular components other than ACh are often released with the transmitter during exocytosis, this is neither always the case (Kato et al., 1974; Michaelson, 1978) nor is the 7 I release necessarily stoichiometric as would be predicted. I I Smith (1 971 , 1 972 i n Tauc, 1 982 ) has even indicated that ; the release of soluble proteins from adrenergic vesicles (which are often referred to as model systems for exocytotic release), may be much lower than that expected based on transmitter content. j There are two principal alternatives to the classical , vesicular release hypothesis. The earlier postulate, first suggested by Birks (1 9 7 4) and elaborated upon by Mar chbanks ^ I (1978) implies that cytoplasmic ACh is released through a specific gate or channel. A more recent and more popular theory combines aspects of the vesicle and gate hypotheses ' in a "vesigate" proposal (Tauc, 1979). The second ' , alternative has also been called the operator hypothesis ; I I I (Dunant and Israel, 1979). I I The idea that ACh might be released through specific i I i channels in the presynaptic membrane would satisfy the j ‘ I requirement that there be a preferential release of j I [cytoplasmic ACh. According to this model, channels at rest] I would be closed except for random periodic openings which i ^ would be measured postsynaptica11y as mepps. With a 2 + ■ depolarization-induced opening of Ca channels, the 2 + subsequent rise in intracellular [Ca ]^ would j effect the simultaneous opening of many ACh channels, i i producing typical epps. Marchbanks (1975) has shown that I such a process would be able to release adequate amounts of I _ j ACh if the cytoplasmic ACh concentrations were in the range^ I of 5 - 20 mM. Most estimates indicate that it is (Tauc, 1979 ; Kelly _e_t aT . , 1 983). The gate hypothesis satisfies three requirements of I the release process. In accord with the findings of Fatt and Katz (1952), release would be quantized with only a | I specific number of ACh molecules passing through any one ^ I I , channel during its open time. Such release would also be I triggered by an increased intracellular calcium I concentration, and would be primarily, if not exclusively, ; I ! 1 from the extravesicular pool of transmitter molecules. I I ! There are however, at least two major objections to the proposal. i i First, the amount of ACh released by a simple gate I I mechanism should be directly proportional to the I I concentration of cytoplasmic ACh (Israel et al., 1979). As, i I < the level of ACh increases, mepp or quantum size should be j increased. Conversely, if the ACh level is decreased, ‘ there should be a commensurate decrease in the size of the ' released quanta. Lowering the level of cytoplasmic ACh by j injection of either acetylcholinesterase (Tauc and Baux, j 1982) or hemicholinium-3 (Macintosh, 1 978) decreases the I number of released quanta, but does not decrease the size , of individual quanta (Tauc and Baux, 1985). Similarly, the . introduction of ACh or its analog carbachol does not result i I in significant changes in the size of individual quanta _’j j released (Tauc and Baux, 1982). A second difficulty arises because ACh is a cation. Release is not electrogenic as would be expected if cytoplasmic ACh diffuses through a channel following its electrochemical gradient (Tauc, 1979). Furthermore, the amount of ACh released is not proportional to membrane potential, and this is not a property that is characteristic of known channels such as those responsible I for Na and K translocation, during the action i I potential (Israel et al., 1 979). A more viable alternative to the vesicle hypothesis is ■ the concept that specific, saturable "operators" exist j within the presynaptic membrane, and these are able to ^ J ' preferentially bind and release cytoplasmic ACh (Israel et I i al . , 1 979 ). The carriers would be closely apposed to the j , inner face or be integral parts of the presynaptic ; ! ■ membrane. To explain the constancy of quantal size, they , would be capable of binding ACh molecules to near | ■ ' saturation (Tauc, 1979). Loading would be from the j ' I : cytoplasmic pool, "but sufficiently separated from this j pool to release ACh in a non- or only slightly electrogenic I I manner" (Tauc and Baux, 1985), and the actual release event I would be initiated by an increased intracellular 1 concentration of calcium ions. Three possible models have been proposed by Israel et I al . ( 1 979). According to the "saturable gate" mechanism, I 101 r ACh is bound to membrane proteins which, upon increased j 2 + [Ca ]^, migrate within the terminal membrane to make contact with complementary proteins. The confluence of any two of these proteins forms a transmembrane channel •which would not necessarily effect electrogenic release if 4- Na were exchanged onto the negatively charged proteins j 1 (Israel et al., 19 79). Another idea is that a specific j I subset of synaptic vesicles exists which is always in close! ! proximity to the terminal membrane. The "operative I vesicle" hypothesis suggests that only these vesicles are | I involved in the release process which basically would occur! i I by exocytosis. This proposal has some support in the ! findings by Zimmerman and Whittaker (1977) that a discrete | I subpopulation of smaller, more dense vesicles (designated | I "VP«") does exist in association with the terminal ! 2 I membrane of Torpedo electric organ. The third model i ! I proposed by Israel's group, designated the "carrier" | \ hypothesis, is very similar to the ves i gate hypothesis of I I Tauc (1979). Both involve the binding of fixed amounts of I ACh to specific receptors which subsequently and very I j rapidly, effect release by diffusion through the membrane, I rotation, or a conformational change. I I Key to the credibility of these proposals is the i finding that specific intramembrane particles always seem I to undergo changes during neurotransmitter release whether I synaptic vesicles are present or not (Israel et al., 1982; 1 11 Israel and Manaranche, 1985). Either In purified and | i refilled synaptosomal sacs or in proteo1iposomes, there is i I consistently seen an increase in the number of large intramembrane particles (8 - 18 nm dia.) whether the preparations are stimulated with KCl, gramicidin, A23187, or Glycera convoiut a venom (Israel and Manaranche, 1985). , j These findings correlate well with the experiments of ; Tauc and Baux (1982; 1985) which suggest that the operators; have properties similar to those of postsynaptic ACh ' receptors (AChR). Injection of two traditional AChR I : blockers, curare and hexaméthonium, rapidly blocks , I transmission without affecting the size of quanta that are ; ' I i released (as might be expected if the inhibitors were I ' themselves released). Tauc and Baux (1985) suggest that I , I the intrasynaptic AChR are part of a vesigate; a saturable I structure functioning in an 'all-or-none' manner." ; The primary arguments against the operator or vesigate ! j j hypothesis include some of those which tend to support the > vesicular mechanism of release. Rearrangement of intramembrane particles or saturation of intrasynaptic AChR I cannot account for vesicular involvement as determined by ; I j the appearance of antibodies to synaptic vesicle antigens j in the presynaptic membrane (von Wedel et al., 1981) or by j the failure of false transmitters not taken up into I vesicles to be released (Weiner and Collier, 1984). 1 It may be that there is no single unifying mechanism I 12 of neurotransmitter release. Possibly One mechanism of 1 release predominates under some circumstances and not in others, or two mechanisms may be combined to give a single ' release process (Macintosh, 1978). It has been suggested that vesicular exocytosis may be the primary mechanism of release at synapses of the peripheral nervous system (Cooper et al., 1982) where rapid release of very high . concentrations of ACh is required (Vizi et al., 19 8 2). i I Such release might be augmented, at times, by release from ^ the extravesicular pool, which may itself represent the ' predominant form of release in the central nervous system ' (Cooper et al . , 1982 ) . The Role of Calcium Ions in Release of Neurotransmi tter Whether one accepts the vesicular or cytoplasmic hypothesis as more accurately describing the process of I release, there is no question that the release of neurotransmitter is ultimately triggered by an increased I 2 + ! intracellular concentration of Ca ions. In this I sense the release of neurotransmitter may be considered as a "specialized form of secretion" (Hopkins, 1979), since ' "in every secretory process in which the test has been made" (Llinas, 1982), secretion is ultimately effected by an increase in [Ca^"^]^. Silinsky (1 982a) has suggested that all secretory systems constitute a continuum, i 13 2 + based on Ca as the "universal provocateur," and that differences which do exist, arise principally with respect to the source of the calcium coupling a stimulus with secretion, The involvement of calcium ions in synaptic function was first recognized by the requirement that it be present in the extracellular fluid bathing the synapse (Dahl et a 1 « , 197 9). Locke, as early as 1894 (Foreman e_t a_l. , 2 + 1976), had determined that Ca was essential to proper function at the frog sartorius neuromuscular junction, but "there was some question as to whether the lack of calcium actually blocked release of transmitter" or merely rendered the postjunctional membrane less stable (Junge, 1981). Harvey and Macintosh resolved the question in 1940 when a study of perfused cat sympathetic ganglia revealed that 2 + omission of Ca ions from the perfusion medium resulted in cessation of measurable ACh release but did not affect the postganglionic cells' ability to respond to applied neurotransmit ter (Junge, 1981). In 1967 Katz and Miledi were able to show, by 2 + iontophoretic application of Ca to the frog neuromuscular junction, that external calcium ions were only effective in eliciting release of ACh if they were applied before or during depolarizing pulses (Katz and 2 + Miledi, 1967a). It was therefore determined that Ca ions must play some role in coupling depolarizing stimuli 14 with release. Furthermore the depolarization need not result from a presynaptic action potential (this could be blocked with tetrodotoxin), but there remained an absolute requirement for calcium in the bathing medium (Katz and Miledi, 1967b). In an elegant experiment to verify their "calcium hypothesis"for release, Katz and Miledi (1967b) injected the presynaptic terminal of the squid giant synapse with tetraethy1ammonium bromide. This effectively eliminated K^ delayed rectification so that the membrane potential could be maintained at large positive values following a depolarizing voltage pulse. The fact that release could be blocked, under these conditions, until after the stimulating pulse was turned off, was strong evidence that 2 + . an inward movement of Ca ions was required for 2 + transmitter release. When inward Ca movement was inhibited, so was release of neurotransmit ter as reflected in the fact that postsynaptic ptoentials were unchanged. In 1 976 Llinas ejt aj_. used the voltage clamp technique to confirm that release of transmitter was due specifically to the influx of calcium ions accompanying a depolarization of the nerve terminal. By pharmacologically blocking HI* • Na and K currents in the presynaptic terminal of the squid stellate ganglion and clamping the membrane potential at discrete "holding potentials," Llinas ejt a 1. (1976) were able to show that postsynaptic response was 15 linearly related to the calcium currrent flowing into the presynaptic terminal. If the presynaptic potential were held at a large positive value (approximately +130 mV), no current could be measured, and there was no postsynaptic response. Llinas e_t aJL. ( 1976) termed this potential the "suppression potential," and, as had Katz and Miledi earlier (1967b), equated it with , the equilibrium 2 + potential for Ca . At this membrane potential 2 + Ca ions cannot flow, and therefore release does not oc cur. Subsequent experiments have been performed correlating 2 + . Ca influx with release of neurotransmitter. Drapeau 3 and Blaustein (1983) showed that [ H] dopamine release from rat striatal synap t o some s previously loaded with the tritiated neurotransmit ter can be correlated with an uptake 45 of radioactive Ca. In this case calcium influx was found to occur in two distinct phases, termed fast and slow, and release could be correlated with both. Similar findings have been reported by Leslie e_t aJL. ( 1985). Endogenous dopamine released from mouse striatal synaptosomes closely paralleled the uptake of ^^Ca by both fast and slow processes. That release of neurotransmit ter is due to an increased intracellular concentration of calcium ions and 2 + not to Ca flux per se was first demonstrated by Miledi (1973). Using the mantle-ganglion preparation of 16 2 + 1 I squid (Loi1 go vulgarls), Miledi was able to inject Ca i ions directly into the presynaptic axon while ; simultaneously recording from the postsynaptic terminal. 2 + Even though Ca had been replaced in the external medium by Mn^^ or La^"*", known calcium channel I antagonists, release could still be measured as distinct ' 2 + postsynaptic potentials whenever Ca was injected into ( the presynaptic fiber. "If the intensity of the calcium i pulse was increased, the postsynaptic potential occurred I I I earlier, and its rate of rise and amplitude increased" > I (Miledi, 1 973) . In 1975 Llinas and Nicholson used the ; Ca^sensitive photoprotein aequorin (isolated from the i I : luminous jellyfish Aequorea f orskalea) to confirm, by i another means, that changes in [Ca]. were ultimately ' I responsible for the release of neurotransmitters. If the ' j I presynaptic terminal of the squid giant synapse were filled; I 'With aequorin, light could be detected in the terminal I during normal synaptic transmission (i.e. in the presence j 1 of extracellular calcium). When a suppression potential was I ' applied to the presynaptic terminal, light emission ceased, I and so did release. j Mo re recently, investigators have made use of a large j I I ' number of specific calcium indicators to study how I alterations in the intracellular concentration of calcium I ions can affect release of neurotransmitters. j ___________________________________________________________________ . - . . -” j , Métallochromie indicators, particularly arsenazo III, have I I been used to investigate spatially nonuniform changes in the [Ca]^ of Li mulus ventral photoreceptors (Harary and Brown, 1984). In even more widespread use are the fluorescent indicators best represented by qu i n 2 (Tsien, I 1 980) . ' 1 The hydrophobic acetoxymethy1 ester of qui n 2 (qui n 2 | ! AM) readily permeates cellular plasma membranes and is | I attacked by endogenous esterases to yield an impermeant, I 2 4- I highly specific Ca probe sensitive to [Ca]^ in I the range of 10-2000 nM (Ashley et al ., 1984) . Quin2 has | been used to correlate increased [Ca]^^ with release of ' catecholamine from isolated chromaffin cells (Burgoyne and I Cheek, 1985) and from cultured rat phaeochromocytoma (PC12)‘ cells (Pozzan et al . , 1984). The fluorescent indicator has also been used in studies of Ca^dependent release j ! . 1 from synaptosomes. Ashley et al. (1984) showed that ACh release from guinea pig cerebral cortical synaptosomes I rises as does quin2 fluorescence. Similar studies have j been reported by Richards et al . (1 984) in studies of the ' depolarization-induced release of gamma-amino butyric acid j (GABA) from rat whole brain synaptosomes. Additional evidence that an increased intracellular Ca^"*" level is ultimately responsible for triggering ; i I L release has come from studies of selectively permeabili zed cells. Baker and Knight (1984) used high voltage I ' permeabi1ization to render bovine adrenal medullary ' chromaffin cells "leaky" and were then able to relate I release of catecholamines directly to the concentrations of 2+ 2 + , Ca in solutions bathing the cells. When the Ca concentration was increased from 10 to 1000 nM, release j (expressed as a percent of the total cellular > ' catecholamine) rose from less than 1% to very nearly 30% I I (Baker and Knight, 1978). ; Cells have also been permeabi1i z ed with detergents. ! Simultaneously in 1983, Dunn and Ho1z (1983) and Wilson and I Kirshner (1983) published results of studies investigating release of catechol amines from digitonin-permeabi1ized ; I chromaffin cells. Both groups found release of I neurotransmitter could be effected by increasing the I 24- ! Ca ion concentration in the bathing media. Half I maximal release was attained at calcium concentrations of -6 approximately 10 M. ; Similar findings have been reported by Smollen and | ! Stoehr (1985) for saponin-permeabilized human neutrophils. ! Release of lysozyme from these cells could be directly manipulated by altering the "free" calcium concentration in j the bathing media (Smollen and Stoehr, 1985). ' Another way to demonstrate the importance of I intracellular calcium in triggering release of 2 4- j neurotransmitter is to introduce Ca ions directly into presynaptic terminals via liposomes. Rahamimoff et 19 ; al. (1978) found that the fusion of liposomes containing 25 i 2 + I mM Ca with presynaptic motor nerve terminals in the frog resulted in markedly increased evoked and spontaneous release of ACh. Furthermore an effect, on release, of liposomal fusion alone was ruled out, because treatment of I nerve terminals with liposomes containing 116 mM KCl was j without effect. Similar studies have been performed by Cr o s1and et al . i 2 + I (1983). Liposomes loaded with 113 mM Ca were fused j I with rat brain synaptosomes, and the consequent increased 24- in t r a sy nap t o s oma 1 concentration of Ca was sufficient : to promote a preferential release of ACh even in the ' absence of extracellular calcium (Crosland et al ., 1983). ^ 2 4- I Precisely how intracellular Ca might actually I i effect quantized release remains a "mystery" (Silinsky, ! : 1 982a), and may involve a whole "series of largely unknown 1 processes" (Silinsky, 1982b; Nachshen and Drapeau, 1982). I I More likely, the triggering effect of calcium is a direct j I I one owing to "the rapidity and explosiveness" with which I j calcium itself effects release "in certain tissues" (Rubin, 1 1970) . In any event, the problem is compounded by the fact that no single unifying hypothesis for the release of | I neurotransmitters yet exists. ; ! ' Hypotheses to account for the mechanism by which | ! 24- I Ca ions could effect vesicular release seek to 2 + I explain how Ca might promote vesicular apposition to ! . 2 0. and fusion with the presynaptic membrane (Kelly et al.. 2 + 1979). One of the principal roles ascribed to Ca in this respect is a neutralization of the negative surface charges bounding the vesicular and terminal membranes (Douglas, 1968; Silinsky, 1985). This might be 2 + accomplished by an actual binding of Ca ions to specific receptors in one or both membranes and has been suggested by several investigators. Katz and Miledi (1967b) noted that calcium probably binds to specific intraterminal "critical sites;" Kelly ejt aJL. ( 1979) implied the existence of specific "effectors" to which Ca^* might bind, and similar ideas have been advanced by Llinas ejt a^l. (1981) and by Silinsky (1982b; 1985). Quite possibly the calcium binding protein is calmodulin. Calmodulin has been found in many, if not most, eukaryotic cells (Klee ejt a_l. , 1980), is highly concentrated in brain, and has been localized specifically to vesicle (DeLorenzo e_t jlI* » 1979; Rephaeli and Parsons, 1982 ; Walker et a1. . 1984) and nerve terminal membranes (Walker et a1.. 1984). Assuming calcium ions, alone or in conjunction with 2 + . . calmodulin or some other Ca binding protein, are able to screen electrostatic repulsive forces between vesicular and nerve terminal membranes, fusion must then be effected. 9 4- Fusion of membranes might be effected by Ca in at least two ways. Papahadjopoulos et a1. ( 1978) have shown 21 2 + that Ca ions themselves can destabilize some j liposomal membranes rendering them more susceptible to I 2 + fusion, and Campbell (1983) notes that Ca ions can activate potent fusogens which are present in most, if not all secretory cells. Hypotheses have not yet been advanced to explain 2 4- ^ specifically how Ca ions might effect quantized nonvesicular release. Whether release of neurotransmitter jwere to occur through specific "gates" or via "saturable 24- operators," intracellular free Ca is believed to be the trigger for these release events as well. The Use of Naturally Occurring Neurotoxins as Probes of ACh Release I Naturally occurring neuro toxins, by virtue of their high degree of specifity, have become valuable tools with Iwhich to study neuronal systems. Since the 1959 discovery I j that tetrodot oxin selectively blocks sodium currents, a great deal of interest has been given to the isolation and I characterization of other neurotoxins specific for ; different neuronal components. Of particular interest has been the elucidation of the mechanism(s) by which neurotransmitters are released from presynaptic cells. As noted earlier, controversy has centered principally over the question of whether exocytosis is the sole means by which transmitter substances are released, or whether other 22 release mechanisms are employed either independently or in ; concert with exocytotic events (Macintosh, 1978). The solution may well be found with the aid of presynaptical1 y active neurotoxins. I The usefulness of presynaptic neurotoxins as molecular ' probes of the release mechanism resides in the possibility that with the different toxins we may be able to: 1) I describe the actual mechanism - by activation or inhibition! of "key" steps in the pathway (Kelly et al., 1979), and 2) i I - I 1 I isolate and purify membrane components involved in the ; ! release process, thereby characterizing, in greater detail,' the "physiological structure" of presynaptic membranes ! I (Dolly et al., 19 8 2). The neurotoxins considered in this i j review all effect release of acetylcholine (ACh) by ' mechanisms that are fairly well understood. I Toxins Whi ch Inhibi t Release , I I ! The two bacterial toxins shown to have an effect on I ! 2 + !ACh release, do so by blocking Ca -triggered release (Wonnacott and Marchbanks, 1976). Furthermore, both I botulinum toxin (BoTx) and tetanus toxin are very i I specifically bound to receptors on only cholinergic I ' presynaptic membranes (Dolly et al., 1982), giving rise to speculation that their sites of action may be an ACh I release site which requires Ca I (Gunderson and Howard, 1978; Wonnacott, 1980; Dreyer et I , 1 984) . I 23 BoTx, isolated from Clostridium bo t ul i num, actually I \ I consists of eight anti genically distinct neurotoxins , all of which effectively block ACh release (Simpson, 1981; Dolly et al., 1982). The most toxic component. Type A (BoTx-A), has a molecular weight of approximately 150 kD . and is made up of two polypeptide chains linked by a disulfide bridge (Dolly et al ., 1 9 8 4; Ho ch et al ., 1985 ). ; I The heavier chain (approximately 100 kD) binds to specific j I receptors on motor nerve terminal membranes (Dolly et al . , ; 1 1984) and is capable of forming transmembrane channels in aj I ’ manner very similar to that by which the diphtheria toxin I heavy fragment acts as a "tunnel protein" (Ho ch et al . , j 19 8 5). The lighter chain (approximately 50 kD) is the ^ ! pharmacologically active component and is introduced into i I the axoplasm via channels formed by the heavy chains. ' ! BoTx-A does not appear to have any effect on terminal I I membrane integrity, choline uptake, or ACh synthesis 1 (Wonnacott and Marchbanks, 1976). Inhibition of ACh release has been attributed by some to a blockade of the vesicular ACh transport system (Edwards et al», 1985), but the ability of black and brown widow spider venoms to ; effect quantized release from BoTx-A-poisoned muscles j (Pump 1i n and delCasti1lo, 1 975), would argue against this hypothesis. The more generally accepted view is that BoTx-A inhibits ACh release by maintaining the intracellular calcium concentration at resting levels 24 1 insufficient to activate the release process (Thesleff, 1984; Dolly et al., 1984). In studies of the effects of BoTx-A on cultured mammalian spinal cord neurons, Dreyer et al . (1 984) suggest that the bacterial toxin specifically ! 2 4- re du ce s the likelihood of Ca -mediated fusion of vesicles with presynaptic membrane. Tetanus toxin has been purified from extracts of ■ I Clostridi um t e t ani and is remarkably similar to BoTx-A both 1 I I structurally and functionally. The toxin is virtually the j ; I t same size as BoTx-A and, like botulinum toxin, is composed j of two polypeptide fragments only one of which possesses j i any toxic activity (Bizzini, 1978; Howard, 1978; Matsuda I I and Yoneda, 1978; Boquet and Duflot, 1982; Lazarovici et j ; al., 1984). The principal effect of tetanus toxin is to block . release by interfering with the movement of vesicles to I active zones (Dreyer et a1 ., 1985), and it is tempting to j I I ' speculate that the inhibition may be due to j ADP-ribosylation of a key protein/enzyme involved in the vesicular apposition phase of release (Kelly et al., 1979). Such a mechanism of action would be analagous to that I described for diphtheria toxin (Metzler, 1977) and might be I ^ expected in view of the structural homology characteristic of the bacterial toxins. This concept is difficult to { : j I reconcile, however, with the data of Metezeau and Desban J i 1 (1982)i who found that increasing the extracellular 25 24- [ Ca concentration or addition of the depolarizing I toxins ATX II (from Anemoni a sulcata tentacles) or 2 4- I Androc tonus australis venom (both at physiological Ca concentrations) had a restorative effect on cholinergic transmission at tetanus-poisoned frog neuromuscular june t i ons . Neurotoxins of other than bacterial origin have also j been shown to effect an inhibition of ACh release, but | I I j these have not been as extensively studied. The spines of j I the long-spined sea urchin Pi adema anti 11 arum, for example,; I : : ! have been shown to contain a presynaptica11y active toxic \ I I factor which blocks release (McClure, pers. comm.), and ; there is some evidence that evoked release of ACh from I ; neuromuscular junctions can be inhibited by a toxin from i the Australian tick, Ixodes ho1ocyc1 us (Howard, 1978). i Still other toxins which have been shown to inhibit release I I also have a facilitatory effect and will be considered as stimulatory toxins. i Toxins Whi ch Facili tate Release I J The stimulatory neurotoxins enhance release of ACh and can be classified into at least four groups: 1) toxins Iwhi ch cause nerve terminals to be depolarized, 2) basic I : proteins with phospholipase activity, 3) toxins acting 2 4- directly on Ca channels, and 4) toxins effecting ! quantal release which is independent of extracellular 24- Ca .26 I The depolarizing neurotoxins facilitate ACh release by' I ! prolonging the amount of time Na channels remain open. As a result, nerve terminals exposed to the toxins remain depolarized for longer than normal periods of time, and 2 + Ca ions continue to move into the presynaptic 2+ I terminals through activated Ca channels. The I depolarizing toxins lend support to the classical vesicle 24- release hypothesis, since all are Ca -dependent and effect increases in both spontaneous and evoked release. Included in this group are several scorpion toxins, toxins I j derived from the tentacles of some sea anemones, and the I lipid soluble alkaloid batrachotoxin. Scorpion toxins can be further subdivided into two I groups based on their Na"*" channel binding properties j ; and the ways by which they effect prolonged depolarization., [ American scorpion toxins have been designated ^toxins and , bind to completely different sites than do the North [ : African o< scorpion toxins (Catterall, 1984; 1985). I ' The best-studied ^scorpion toxin is the iT'form of j tityustoxin (TsTx-Y) , a relatively small (M.W. 7 , 000) j ' I polypeptide isolated from venom of the Brazilian yellow ! I . ' I scorpion Ti tyus serrulatus. TsTx-o is composed of 62 amino! I acid residues and has a net positive charge at physiological pH (Alagon et al., 1978). The toxin is not j I specific for cholinergic terminals (Santos et al . , 1978 ; { I Freire Mai a et al ., 1 978a; Freire-Maia et al, 1978b), but 27 : effects a more general release consistent with an enhanced activation of sodium channels (Freire-Maia et al., 1978b; I Rathmayer et al., 1978; Catterall, 19 8 4). The specificity of TsTx for sodium channels is evidenced by the fact that toxin-generated release can be blocked completely by either, removal of extracellular sodium (Macedo and Gomez, 1982 ) or I 1 addition of the sodium channel blocker tetrodotoxin ■ (Abdul-Ghani et al., 1980; 1981; Macedo and Gomez, 1982; ; Rhoads et al., 19 8 3). I I Other less well characterized scorpion toxins include toxin II isolated from Centruroides suffusus (Wheeler et al., 1982) and toxins III and V from , Cent ruroi des sculp t ur a t u s (Barhanin et al ., 1984). All ' + share a common binding site on the Na channel as i I evidenced by a co mp etitive interaction at the TsTx-^ ' receptor (Barhanin et al., 1984). j The o( scorpion toxins are best represented by toxin V j i Lei urus quinquestriatus (Angel ides and Nutter, 1 983 ). ! These toxins binâ, in a voIt age-dependent manner, to sites ' + I on the Na channel which undergo conformational change . : during the process of channel activation (Catterall, 1984; 1985) . As a consequence, the bound AC-toxin causes inactivation to be slowed or completely blocked (Waxman and Ritchie, 1985). Other scorpion toxins in this class include those isolated from the venoms of Androctonus australis (Rathmayer et a1 ., 1978 ), Buthus occi t anus, and 28 He t er orne t r u s fuivi pe s (Tu, 1 977). ' Toxins derived from tentacles of the sea anemones Anemoni a sulcata and Anthopieura xanthogrammi ca appear to ' bind to the same receptor sites as do the PC-scorpion toxins (Frelin et al . , 1 9 8 4) , and they similarly block Na"^ channel inactivation (Catterall, 1984; 1985). I Batrachotoxins are among the most potent naturally ! ; occurring non-protein toxins, and have been used, for a i long time, as dart poisons by Central American Indians ! ! (Myers and Daly, 1983). All are derived from granular ' I I I glands in the skin of Phyllobatid frogs, and are : lipid-soluble alkaloids. Binding to the same sites as does’ 4- ! verat ri dine (Catterall, 1 985), these toxins cause Na channels to be persistently activated (Rosenberg et al . , 1 984; Ghiasuddin and Soderlund, 1 984; Hartshorne et al . , I 1985). The snake toxins are all basic proteins with 24- C a - dependent phospholipase A^ activity (Ha11iw ell et al . , 1982) and all are believed to facilitate the 24- release of ACh by transiently inducing Ca influx j (Spokes and Dolly, 1980) and/or liberating mitochondrial 24- j Ca (McClure, pers. comm.). Like the bacterial I I neurotoxins, many of the snake toxins are composed of two or more subunits, at least one of which is acidic and seems to function primarily in binding to specific protein I receptors in the presynaptic membrane (Strong et al . , 1 978 ; 29 Kelly et al., 1979; Rehm and Betz, 1982). The specificity ! rendered by the binding or "chaperone" subunit appears to be a requirement for neurotoxic activity in the snake venoms, since nontoxic phospholipases are not bound to nerve terminals and have no effect on the release of ACh (Lin-Shi au and Chen, 1 982 ) . yÔ-Bungaro toxin (yS-BuTx), isolated from the krait I j Bungarus muIticinctus, is the best characterized of the i I phospholipase neurotoxins. It is composed of two subunits ! I 'with molecular weights of 1 2, 000 and 9 , 000 d alt ons (Howard,' 19 7 8), and, as is characteristic of most of the snake neuro t oxins, generates a nonspecific release of j neurotransmitters (Rehm and Betz, 1982 ; Ha11iw el1 and I ! Dolly, 19 8 2). The facilated release is always preceded by I : ! i a transient decrease in the release of neurotransmitter and. followed by a complete blockage of cholinergic transmis sionj ; (Chang et al., 1973; Tobias, 1982). The triphasic i j ! I mechanism of action can be attributed to (Spokes and Dolly, | ; 1 980) : I 1. Binding of the "chaperone"subunit (decreased release) 2 4- I 2. Ca -dependent phospholipase activity and ' depolarization (increased release) 3. Energy depletion due to uncoupling of mitochondrial oxidative phosphorylation by free fatty acids or latent binding to _ 30j I ' ' 2 4- I transmitter release sites or Ca channels (blockage of release) All of the snake neurotoxins exhibit a muIti-phased course of action believed similar to that of yS-BuTx. They differ primarily in structure and potency. Taipoxin, for j I e X a m ple, isolated from the tai p an Oxynuranus scutellatus i I i : scutellat us, is about twice the size and ten times more j I lethal than any of the other presynaptically active toxins i ' (Howard, 1978). Notexin from the Australian tiger snake ' : I I Note chi s scutatus s cut atus (Halpert, 1 9 7 9 ; Gundersen and , Jenden, 1981) and caudoxin from the horned puff adder Bi t i s j caudali s are both relatively low molecular weight toxins (M.W. approx. 13,500), yet each exhibits the triphasic ! i course of action characteristic of the other phospho1ipas es. Finally, dendro t oxin from the Eastern ' green mamba D endro ap s i s angus t i c ep s and cr ot oxin from the ' South American rattlesnake Cro t alus durri s sus terrificus. ! I I are both structually similar to yô-BuTx (Harvey, 1 982 ; j (Howard, 1978) and may be binding to the same prejunctional , i ! ; sites in effecting the release of transmitter. I ' I At least three different neurotoxins have been shown , i 2 4- 'to bind specifically to Ca channels and may [ I eventual1 y represent the major part of a solution to the question of how transmitters are released in general and 'ACh released in particular. Omega toxin isolated from the I marine snail Conus geo graphus seems to selectively block I ________________________________ __ 2 + ■ Ca channels; toxins isolated from the dinof1age11 ate G a mb ierdiscus t oxi c us and two different species of the Colorado potato beetle. Le p tinot arsa sp., seem to selectively open them. Omega toxin is highly basic and composed of only I twenty amino acids, yet in a manner very similar to the ! ! snake phospholipases, it causes a transient increase in mepp frequency foilow ed by a total and irreversible block j of ep p s (Kerr and Yoshikami, 1 982) . Since the activity of omega toxin has been attributed to a specific inhibition of 24- ' Ca entry into nerve terminals during presynaptic : action potentials (Kerr and Yoshikami, 1982), it may be that the neurotoxicity of phospholipases is also due to 24- some kind of obstruction of Ca channels as has been suggested by Spokes and Dolly (1980). I ^ In contrast, the leptinotarsid toxins stimulate 1 24- release of ACh at least partially by promoting Ca , influx. The toxin isolated from hemolymph of L . ha1demani ^ (LPT-h) , for example, causes a bi ph a sic increase in mepp j frequency recorded from rat phrenic nerve-di aphragm I j preparations, and only the earlier, higher frequency phasej ! 2+ I ; Is abolished in low Ca media (McClure et al., 1980; | Crosland, 19 8 2). The two phased release process has given ^ I I ' rise to the suggestion that there may, in fact, be two | separate and distinct pools of ACh presynaptically both of , which are capable of liberating ACh in a qu ant al manner I I _________ 3 2j I (McClure et al., 1980). If this were true, it could be i ! that the toxins described earlier in this study are effecting release from only one pool, but it would also be possible that release from two pools occurs simultaneously ! in some cases, and that the experimental techniques did not permit differentiation of multiple release processes. derived from L . decemlineata (LPT-d) is ; similar in mode of action to LPT-h, although it appears to ; be a considerably smaller molecule than the latter (M.W. i I 40,000 vs 57,000 for LPT-h). Yoshino (1980) notes that the' 45 I toxin stimulates a dose-dependent uptake of Ca, and that this activity is partially inhibited by verapamil and , 2+ 2 + 1 Mg , both of which are inhibitors of inward Ca I currents. Release of neurotransmitters is nonspecific: ; Both of the leptinotarsid toxins stimulate release of i 3 3 [ H]-norepinephrine and [ H ]-GABA from synaptosomes as well as they do ACh. I The dinof 1 agel 1 ate G a mb ierdiscus t oxi eus is probably ! ! best known as being largely responsible for ciguatera food ! I i I j poisoning in tropical and subtropical regions of the world , i (Dickey et al., 1984; Legrand and Bagnis, 1984). One of ; I I the substances which may contribute to ciguatera poisoning ■ is maitotoxin (MTX), a nonproteinaceous water-soluble 24- comp ound which can specifically open Ca channels in I smooth muscle (Legrand and Bagnis, 1984; Miller et al., I 1985), cardiac muscle (Legrand and Bagnis, 1984; Dickey et j ____________________________________________ .. 33 , al ., 1 984), and some types of neuronal tissue (Freedman et al . , 1 984 ; Login et al », 1 985). Although the ability to effect release of ACh from intact neurons or synaptosomes has not yet been demos t ra t ed, MTX has been shown to stimulate release of catecholamines from both adrenal phaeochromocytoma (PCI 2) cells (Takahashi et al . , 1 982 ; Login et al., 1985) and neuronal clonal cell lines 2 + ' (Freedman et al ., 1 984) . Ca -dependent hormonal ! release can also be facilitated by maitotoxin and has been • described for GH^ clonal pituitary tumor cells (Login e t al . . 1 985 ) as well as for rat anterior pituitary cells i n vitro (Schettini et al ., 1 984 ). The final group of toxins to be considered are those I I 2 "i“ j that do not require the presence of extracellular Ca 1 to effect release of neurotransmitter. Included in this group of neurotoxins are 0(-1 atrotoxin, the principal toxic component of black and brown widow spider venoms, the venom ; of G1yc era convoiuta, and iotrochotin, the "purple blood" I of the marine purple bleeder sponge lotrochota bi rotula t a . 1 -La t r ot oxi n (c(-L T ) is the most toxic and first I characterized component of the venoms taken from venom ; glands of both black (Latrodectus mac t ans t redeci mgu 11 a tus) and brown (L . geomet ri eus) widow spiders (Tzeng and Siekevitz, 1978). The toxin binds to protein receptors in I the presynaptic membrane, induces fusion of synaptic I vesicles with the membrane resulting in a nonspecific 34 release of transmitters, and does not permit vesicle recycling (Tzeng et al ., 1 978 ; Pump 1i n and Reese, 1 977 ). Proponents of the vesicular release hypothesis point to the fact that o(-LT seems to bind preferentially to active zones I (Gorio and Mauro, 1979) and may therefore be activating the 2 4- same exocytotic release mechanism as Ca , but by different means (Tzeng et al., 1978). Indeed, the effects of oC-LT, like those of LPT-h, lend support to the concept ' 24- I that the Ca -dependent vesicular release process may 2 + ' be supplemented on occasion by a non-Ca -dependent process involving release from a cytoplasmic pool of , neurotransmitter (Henderson and McClure, 1978). i The venom of the polychaete annelid Glycera convo1uta I ; effects release in much the same way as does 0< - L T , except that no morphological changes in the presynaptic membrane | I are apparent following release (Manaranche et al., 1980; I I I Manaranche et al ., 1 982 ) . Like o(-LT, Glycera convo1u t a j venom stimulates a nonspecific release of neuro transmitters ' 2 4- j that has been described as both Ca -dependent 2 4- ^ (Manaranche et al., 1980) and Ca -independent (Morel . and Meunier, 1981). At least two studies have suggested that the venom is causing a release of non-vesi cular or ! "free" ACh [and that the release is quantized] (Israel and I I Lesbats, 1981; Morel and Meunier, 1981) . I Iotrochotin (TOT) has only recently been characterized I (Martin and McClure, 1982), but could prove very useful in i 3 5 elucidating the process of release. lOT (M.W. approx. ! 18,000) can stimulate the release of ACh from synaptosomes 24- i n the absence of extracellular Ca , and does not appear to cause a concomitant release of typical cytoplasmic markers. Furthermore, the lOT-releasable pool of ACh appears to be distinct from the pool releasable by depolarization with high K , and may therefore be ' p r i marily I Natural History of lotrochota and Leptinotarsa I I 1• lotrochota bi rotulata Sponges, because they are sessile, soft-bodied, and very often openly exposed (Green, 1977), have elaborated a diverse arsenal of chemical defenses. Some members of the I phylum Porifera produce substances which can be toxic to [ predators (Green, 1977; Sevcik and Barboza, 1983) often at relatively low concentrations (Bakus and Thun, 1979); : others are themselves distateful or release metabolites which can be noxious and odorous (Baslow, 1969). One representative of this latter group is To t r o cho t a j birotulata (Higgin). lotrochota birotulata is a Caribbean sponge found at i depths of 2 - 20 meters (Colin, 1978 ; Nei gel and Avise, ! 1983) usually on small coral heads or patch reefs. Dark purple in color, the sponge forms aggregations of branches 36 I typically one to four cm in diameter and up to 50 cm in length (Colin, 1978). Often the orange zooanthld : Parazoanthus swlftll Is found In close association. When squeezed or otherwise disturbed by fish, I . blrotulata emits a dark-purple, strong-smelling exudate which Is capable of arresting the attack (Green, 1977). A similar i defensive behavior Is characteristic of the Red Sea sponge I I batruncul1 a magnlflea (Spector et al ., 1 983 ). When j squeezed, that sponge "exudes a reddish fluid that causes I ' I fish to retreat from Its vicinity" (Spector et al., 1983). • lotrochota blrotulata Is often commonly referred to as the "purple bleeder sponge" (Baden and Corbett, 1979). ; Bakus and Thun (1979) have found aqueous and alcoholic: I extracts of the whole sponge to be "mildly toxic" to the ' marine sergeant major fish Abude f duf s axltills, but subsequent experiments using the (fresh water) goldfish . Cassarlus auratus failed to show evidence of toxicity. The i authors do note however that. In the latter case, fish | , appeared to be "stressed or weakened" after a 24 hour 1 I I j exposure to the sponge extracts. Similarly, Green (1977) ' I I I finds goldfish to be virtually unaffected by water, i I I I ethanol, chloroform, or acetone extracts of I . blrotulata ! j tissue, but notes that the exudate "Is avoided." , 2. LeptInotarsa de ceml1 neat a ^ I The Colorado potato beetle, Leptlnotarsa decemllneata, 37 is distinctively colored at all life stages. Larvae are , bright red-orange, and adults are best described as a dull yellow (Haedstrom, 1977), yet at no stage Is the beetle ! extensively preyed upon. Principal natural enemies of L . ■ decemllneata Include the two-spotted stink bug Perlllus bloculatus (Shagov, 1977; 1978; Tamakl and Butt, 1978), some carabld ground beetles (Sorokin, 1982), and. In some i I areas, the frog, Rana rldlbunda (Rudyak, 1977). Additionally, there Is some evidence that the beetles may serve as hosts for parasitic nematodes (Bozhkov and I Kajtazov, 1976; Grlgorovltch and Kravchenko, 1978) and mlcrosporldlans (Hostounsky and Welser, 1978) as well as some types of pathogenic fungi (Fargues, 1976), but "mortality from natural enemies rarely exceeds 15%" (Hare, 1983). It seems that L . decemlIneata has evolved an , effective chemical defense mechanism. i ; i I Ranging throughout North America, northern Europe, and; ■the USSR, Colorado potato beetles have been major economic j pests In this country since at least 1861 when they were ' I found feeding on cultivated potato plants (Solanum ' ; tub ero sum) (Hare, 1 983). The life cycle Is typical of jchrysomelld beetles. Eggs are deposited In late spring on the undersides of solanaceous foliage and hatch simultaneously to produce brightly colored larvae which j Ibegln to feed Immediately (Hare, 1983). Larval development | i I j takes about 10 - 20 days (Parker, 1970) and Is followed by j L _________________________ - -■ 38J I a pupation period In the soil of approximately the same duration (Hare, 1 983) . Adult L . decemlIneata then emerge from the soil to begin the cycle anew, usually occupying the nearest appropriate host plant (Hare, 1983). A protelnaceous toxin of molecular weight 40—60 kD can be Isolated from the beetles at all life stages (Including eggs; Hsalo and Fraenkel, 1969). Furthermore, the toxic principle appears to affect Invertebrates as well as vertebrates Including mammals (Hsalo and Fraenkel, 1969; i I Parker, 1970). Intraper1tonea1 Injection of the toxic protein into rats results, almost Immediately, In an ! , extreme paralysis of the hind extremities, and ultimately I In death, the timing of which Is dependent upon the dose i ' administered (Parker, 1970). ' Ra 11onale for Resear ch Questions remain regarding specific aspects of the i process by which neurotransmitters are released from presynaptlc nerve terminals. Although the vesicular release hypothesis has assumed an "aura of absolutism" ’ (Tauc, 1979), cases have been found In which release j appears to be primarily or, at least partially, I J nonves1cular, and even the Idea that release Is quantized I appears not to be universally true. Likewise, the specific role of calcium In promoting release Is unclear, since. In ' some tissues, depolarlzatlon-evoked release has been found ! 39 to occur in the absence of extracellular calcium. The objective of this research Is to contribute to a better understanding of the molecular mechanism by which neurotransmitters In general and ACh In particular are released from presynaptlc nerve endings, primarily those of the central nervous system. By better defining the process' I of release It might then be possible to Identify errors or , alterations In release which occur as a consequence of , disease and/or aging (see Peters on and Gibson, 1983; Les lie, I ! ! et al . , 1 985). An Investigation of the requirement for ^ calcium In release will also be made. j Naturally occurring neurotoxins have been utilized as I molecular probes of the release mechanism. Such toxins by | virtue of their high degree of specificity have proven to ! be Invaluable In studies of neuronal systems and may ' I provide a means by which to dissect and analyze the i , sequence of events which contribute to the release of ! i ; neurotransmitters. lot r ocho 11n (lOT), the wound exudate of j ' Caribbean purple bleeder sponges, effects release of ACh } I from rat brain synaptosomes in a manner that is Independent! I I ! of the extracellular calcium concentration and seems to j , preferentially Induce release from extraveslcular stores. | ' In contrast, the presynap11cally active toxins Isolated j 2+ ' I from hemolymph of Colorado potato beetles, require Ca | I In order to stimulate release and. In fact, seem to be I I I specifically opening presynaptlc calcium channels (Crosland' I 4 o i et al ., 1 984) . Specific alms for this research have been: 1. To purify and further characterize the neurotoxins. 2. To examine, in greater detail, the mechanisms by which the two toxins effect release of radioactivity from synaptosomes loaded with [ H]choline. 3. To Investigate the possibility that the toxins bind to specific receptors In the synaptosomal membrane, and. If so, to characterize the toxln-receptor Interaction 4. With respect to LPT-d, to find out If the toxln-stlmu 1a t e d release can be affected by a pre-exposure of synaptosomes to known organic and Inorganic calcium channel ant agonls t s . 5. With respect to LPT-d, to determine. In collaboration with other Investigators, If the toxin affects nonneuronal calcium channels. 41 MATERIALS AND METHODS Anima 1s Male Sprague—Dawley rats were obtained from the Simonsen Laboratories, Gilroy, CA. Frogs (Rana pipiens) were ordered from suppliers in Wisconsin. Fresh beef brains were purchased from the Beefco Meat Co., Los Angeles, CA. Electric rays (Ommata discopvee) were obtained from the Marinus Co., Westchester, CA, and live I ■squid ( Lo 1 igo opa 1 es c ens ) . freshly collected from the Catalina Channel, were obtained as gifts from Dr. George Augustine of the University of Southern California, Los Angele s. ChemicaIs [Methyl-^H] choline chloride (80 Ci/mmol) and o 2—[ H(G)]deoxy-D—glucose (5.0 Ci/mmol) were ordered from New England Nuclear, Boston, MA. 3 2 [ P]orthophosphate (1 mCi/ml) in dilute HCl solution was obtained from the Amersham Corp., Arlington Heights, ;IL, and carrier- free [^^^I]sodium iodide (17 C i/mg ) was received from the ICN Radiochemica1s Co., Irvine, CA. Chromât of o eus ing media and supplies, all Sephadex gel I filtration resins, DEAE-Sephadex (A50), CM-Sephadex (C50), ! and electrophoresis calibration kits were purchased from Pharmacia Fine Chemicals, Pis cataway, NJ. Cel lex D, Dowex I 42 f 1 j 50W-X8, Bio-Lytes, Coornas s le Brilliant Blue dye reagent, silver stain reagents, and all electrophoresis grade chemicals were obtained from Bio—Rad Laboratories, I Richmond, CA. Dye-matrix columns and all equipment used in ultrafiltration were ordered from the Amicon Corp., ^ Danvers, MA. lODO-GEN was obtained from the Pierce ■ Chemical Co., Rockford, IL. Choline kinase, eserine j ' sulfate, NADH, fluorescein isothiocyanate, verapamil and I I I firefly extract were purchased from the Sigma Chemical Co., I St. Loui s, MO. I I i Nitrendipine was a gift from Dr. Alexander Scriabine | of the Miles Institute of Preclini cal Pharmacology in New ; Haven, CT. Nifedipine was a gift from the Nelson Research j , I and Development Co. , Irvine, CA. Crude maitotoxin was | ! _ i ' generously provided by Dr. Donald M. Miller of the Southern i . . . . I Illinois University, Carbondale, IL. ' Basic laboratory chemicals and common biochemical j I , reagents were obtained from the Sigma Chemical Co., St. j I I Louis, MO. or from other reliable sources. So lut ions The physiological saline (PS) solution used in the routine preparation of synaptosomes from rat, mouse, and bovine brains consisted of 130 mM NaCl , 4 mM KCl, 10 mM ! glucose, 2 mM MgCl^ and 1 mM CaCl^ in a 20 mM j sodium phosphate buffer (pH 7.0). The buffer was prepared i 43 by mixing 20 mM solutions of the monobasic (NaH^PO^) and dibasic (NajHPO^) forms of sodium phosphate so as to achieve the desired pH. Titrations, when necessary, were performed using either 0.5 M NaOH or 0.5 M HCl. Modified PS solutions were prepared as follows. For 2 + 1 experiments involving inorganic Ca channel | antagonists or calcium concentrations greater than 2 mM, HEPES-NaOH buffer was used instead of sodium phosphate to j avoid precipitation of the metal phosphates. When j ! I : solutions containing no calcium were needed, CaCl_ was i I I substituted, in PS, by either MgCl^ or EGTA. I Synaptosomes were depolarized by exposure to PS solutions I I that typically contained 55 mM KCl (and to maintain ' isosmotic conditions only 79 mM NaCl). Zero Na^ I conditions were obtained by substituting choline chloride for NaCl on an equimolar basis. The solutions used in experiments not involving synaptosomes are described in the ^ appropriate sections. Preparation o f Ra t Forebrain Synap t o s ome s Whole forebrains of adult male Sprague-Dawley rats 1 (150-200 g) were used. Although synaptosomes prepared from mammalian brain contain many different families of nerve endings, the specificity of the high affinity choline I 1 uptake system and the virtually complete conversion of j ; incorporated choline into ACh (Raiteri et a 1. . 1984) allowed us reliably to measure release of ACh with this preparation. Furthermore, studies of the distribution of I ' cholinergic nerve terminals in whole rat brains ' consistently have shown that the highest concentrations of releasable ACh are located in the area rostral to the cerebellum (Leslie e_t a_l. , 1983), and, more specifically, within the regions that make up the corpus striatum : (Haycock and Meligeni, 1 977 ; Beani et a 1.. 1984; Ikarashi i e_t aj^. , 1985; Vickroy e_t a_l. , 1 985). Rats were killed by decapitation and their forebrains I rapidly removed and weighed. The brain tissue was then I homogenized in ice cold 0.32 M sucrose (10% w/v) using a : wide clearance (0.25 mm) Teflon and glass homogeniz er. ! Homogenates were centrifuged in a Sorvall Model RC-2B I centrifuge at 750 x g (10 min. ; 4° C) to yield a pellet , containing nuclear fragments and cellular debris (PI) and a supernatant (SI). The SI fraction was subsequently ' centrifuged at 10,000 x g (15 min.;4°C) and the resulting pellet (P2) washed once with the original volume of 0.32 M sucrose. The washed P2 fraction was reclaimed by I I centrifugation at 12,000 x g (20 min. ; 4®C) to produce a washed crude mitochondrial pellet which, after ! resuspension in PS, was used and will be referred to as the I j synaptosomal pool in most of the experiments assaying I release of ACh. __ f 1 Crude synaptosomal pools are commonly used in experiments concerned with release of neurotransmitters including ACh (Boksa and Collier, 1980; Gundersen and ' Jenden, 1981; Jope, 1981; Reese and Cooper, 1982; Rowell and Winkler, 1984; Weiner and Collier, 1984; Gomez and ^ Ribiero, 1 985 ), dopamine (Leslie e_t a 1 . . 1 985) and various amino acids (Rhoads e_t a 1. . 1 983 ; Arias e_t a_l. , 1984; i Wheeler, 1984). Washed P2 fractions are also routinely I used in studies of phosphorylation (Robinson and Dunkley, j 1983 ; Robinson e t a 1. . 1984; Whittemore e_t aJL. , 1984; Robinson and Dunkley, 1985) receptor binding (Harvey et , 1984; Janis aJL. , 1984; Seamon e_t a_l . , 1984; Wu e_t . a j : , » , 1 984; Greenberg e_t a_l. , 1 985 ; Maneckjee e_t a_l. , 1 985 ; I Meyer and Otero, 1 985) and calcium uptake (Rampe .e_t a 1. . I 1984; Leslie e_t al. , 1985). The crude synaptosomal preparation includes myelin andi ! I isolated mitochondria in addition to synaptosomes, and, for| some experiments, especially those concerned with toxin j binding and the calcium dependence of ACh release, it was ' : I I considered most feasible to use non-contarninated | 1 f ' synaptosomes. Mitochondria, for example, represent a | I I significant reservoir of intracellular calcium (Campbell, j 1983; Snelling and Nicholls, 1984) and have a | I 2+ i j well-developed Ca -transport system (Panfili et a 1.. j I 1983). Additionally, isolated mitochondria can contain as 'much as one half the choline acetyItransferase activity found in purified synaptosomes and can take up choline about one fifth as well as can synaptosomes (Weiner and Collier, 1984) . Purified synaptosomes were isolated essentially according to the procedure outlined by Gray and Whit taker ( 1962) as modified by Ray et a 1 . ( 197 8). The P2 fraction (prior to washing) was resuspended to a final volume of 10 ml in 0.32 M sucrose and carefully layered onto a stepwise sucrose gradient consisting of seven ml layers of 1.2, 1.0, j o . 8, and 0.6 M sucrose. Ultracentrifugation (Beckman L3-40) in a swinging bucket rotor (SW27) at 78,000 x g (120 min.;4°C) resulted in a separation of myelin (0.6 M layer) and mitochondria (sediment in 1.2 M layer) from synaptosomes (0.8 and 1.0 M sucrose fractions). Since the contamination of synaptosomes by broken membranes is greater in the 0.8 M fraction (Ray e_t jlL . , 1 978), the purified synaptosomes used in these studies were routinely , ' taken from the 1.0 M sucrose fraction and the 1.0 - 1.2 M , I I interface (approximately 10 ml total volume). The sucrose I I j concentration was then adjusted to 0.32 M by slow, dropwise ! addition of ice-cold distilled water with stirring, and the synaptosomes were reclaimed by centrifugation at 12,000 x g; (30 min., 4^C). The purified synaptosomes were then j : resuspended in PS to a concentration of 0.2 - 0.4 mg j i protein/ml, such that during the process of loading with | I I tritiated choline (see below), the amount of radioactive choline taken up was approximately 1.5 million cpm/mg ' synaptosomal protein. Préparât ion o f Svnap t o s ome s from Squid Optic Lobe s Because ce pha1opod central nervous system (CNS) nerve endings are predominately cholinergic and take up choline at rates much higher than do synaptosomes isolated from I mammalian brains (Dowda11 and Simon, 1973; Dowdall and I Whittaker, 1 973 ; Dowdall, 1974; Haghighat e_t a 1 . . 1984) , ' some experiments were performed with synaptosomes derived from the optic lobes of squid. The synaptosomes were I prepared essentially by the method of Pollard and Pappas , ( 1979) with modifications described by Pant e_t ( 1983). Four to six optic lobes (approximately 0.9 mg wet weight) I were homogenized in isotonic 1.0 M sucrose (10% w/v) and ; the homogenate centrifuged at 12,000 x g (1 hour ; 4°C). The floating pellicle containing the synaptosomes and about ; two ml of the isotonic sucrose were then drawn off by hand I pipette and mixed (1:5) with a "squid saline" solution (466j I :mM NaCl, 54 mM MgCl^, 10 mM KCl, 3 mM NaHCO^, and ; 10 mM HEPES - NaOH, pH 7.2). Preparation of Synaptosomes from Electric Organ I To consider the effects of toxins on cholinergic nerve' ! endings characteristic of the peripheral nervous system, I I synaptosomes were prepared from electric organs of the ! ___________ ________________ ... .48] I marine ray Ommata dis copyge by the procedure of Miljanich , _e_t ( 1982). Briefly, freshly excised electric organs were minced in a zero calcium buffer (pH 7.4) and homogenized in a Waring blender. After it was filtered through a stainless steel screen, the homogenate was centrifuged at 3 0,000 x g (15 min.; 4°C). The resulting pellet was re sus pende d in zero calcium buffer with a loose-fitting glass-TefIon homogenizer and carefully I I layered onto a continuous 3 - 20% linear Ficoll gradient. | After ultra-centrifugation for one hour at 80,000 x g, the ; i ' synaptosomes were isolated as a narrow band (first major 1 band below the buffer-gradient interface) and diluted four j ' fold with zero calcium buffer. The purified synaptosomes ! I I were then reclaimed by centrifugation at 30,000 x g (15 i I min. ;4^C), re suspended in a small volume (approximately '1 ml) of zero calcium buffer containing 1% BSA and stored at O^C until they could be used in experiments. I 3 I Load ing of Synap t o somes wi th [ H]Choiine I A solution of 50 uCi (50 ul) of [methyl— I^H]choline chloride in methanol was dried under a i stream of nitrogen at 4°C, taken up in 2.8 ml of PS and , I added to a solution of synaptosomes (1 g forebrain/8 ml PS)| 3 ' such that the final concentration of [ H]cho1ine chloride was approximately 60 nM. The two solutions were ! mixed by gentle swirling in a glass vessel which was i I ! _ . - 4 . 9 J subsequently sealed with parafilm and incubated in a shaking water bath at 3 7 ° C for 30 minutes. To remove unincorporated or adventitiously bound [ H]choline, the loaded synaptosomes were "washed" by centrifugation at 5 ,000 X g (5 min.; 4°C) followed by resuspension in the initial incubation volume of PS (10.8 ml). This procedure was repeated twice more with the final resuspension such that the synaptosomal protein concentration was typically one to two mg/ml. 3 I Loading of Synaptosomes with 2-[ HJDeoxyglucose To test for nonspecific release of small cytoplasmic ! molecules, some synaptosomes were loaded with I ' n n i 2-[ H(G)]-deoxy-D-glucose(2-[ H]DG) by a ; modification of the procedure defined by Diamond and Fishman (1973). In order to carry out these experiments, a modified PS was prepared containing 5 mM choline chloride i - ' ' as a substitute for glucose (Glucose competitively inhibits^ I I 3 ‘ ' the uptake of 2-[ H]DG; = 64 uM according to I Diamond and Fishman, 1973). Six ml of synaptosomes, ' suspended in the modified PS to a concentration of 4 - 5 ,mg Î protein/ml, were mixed with 200 ul of saline solution 3 containing 40 uCi of 2-[ H]DG. Synaptosomes were then ' 3 : incubated in the presence of 2-[ HjDG for 30 min. in a i I shaking water bath at 3 7 , and labelled 2-DG not taken up during that time was removed, as above, by three 50. I sequential centrifugation and re suspension cycles. Final resuspension was in normal (glucose-containing) PS. i Procedure Used for As saying Re lease o f Radioact ivity Loaded and washed synaptosomes, in 0.2 ml aliquots, were added to 1.0 ml samples in 1.5 ml plastic microfuge tubes, and the tubes were shaken briefly to insure mixing. The rack containing the capped microfuge tubes was then I j inverted (so that the solutions would be more adequately : mixed within the microfuge tubes), and placed in a water bath at 37°C. Incubation of the samples was generally 1 for 10 min. with gentle shaking. The time of incubation : was defined as the time between the first exposure to the 37°C water bath and the termination of incubation by immersion of the samples in an ice-water slush. Samples ; were then centrifuged in a Beckman Model 12 microfuge at 9,000 X g (5 min.; 4°C) and the supernatants (approximately 1.2 ml) were removed as indicative of the i amount of radioactivity released from the synaptosomes. Pellets resulting from the centrifugation were resuspended in 1.2 ml of distilled water and vigorously vortexed to j ly se the synaptosomes (Johnson and Whittaker, 1963). The i ! resulting homogeneous suspensions were centrifuged at 9,000 I I X g (5 min.; 4°C) to produce a 1.2 ml supernatant i containing cytosolic radioactivity (that had not been I released). Supernatants from the two centrifugations were 51 added (1:5) to scintillation cocktail developed by Anderson and McClure (1973) and counted (typically for one min.) on a Beckman Model LS7500 liquid scintillation counter. Release was quantified by defining a unit of release activity as follows: cpm released by sample unit = " ■ .......... ' ' ' ' ■■ ' - 1 cpm released by PS (control) 1 This calculation compensated for small differences in ‘ I choline uptake between the different synaptosomal i preparations. Lysate values, which represent maximum possible release, were consistently 3.5 to 4 activity I uni t s. ! 3 Senarat ion and Determination of [ R]Cho1ine and [ H]Acetvlcholine Since the ratio of choline to acetylcholine released by a given toxin is fairly constant (Suzkiw e_t ^1. , 1984), ; total release of tritium was often used as a measure of release, particularly in assays of activity following individual steps in the toxin purification schemes. In I some cases however, it was considered necessary to ! . . 3 I distinguish between released [ H]ACh and released ‘ 3 . [ E]choline. This was particularly true in experiments i evaluating the effects of 1 0 T when it became apparent that this toxin was not effecting release of ACh specifically and in studies of the effects of calcium channel agonists __________________________ -5.2 and antagonists on release stimulated by LPT-d. 3 3 Separation of [ H]choline and [ H]ACh was accomplished by a modification of the choline kinase assay described by Goldberg and MeCaman (1973). When a separation was planned, the PS was supplemented with 0.1 mM phySOStigmine (e ser ine sulfate; Sigma) to inhibit acetylcholinesterase. Supernatants resulting from the first centrifugation following synaptosomal incubation with toxin were divided ! into two 500 ul aliquots. To one aliquot was added 150 ul of a solution containing 0.075 units of choline kinase (Sigma), 62.5 mM MgCl^, 50 mM ATP, and 250 mM ,glycyIglycine buffer adjusted to pH 8.5 with 1 N NaOH. The I solutions were mixed by inversion and placed in a shaking ' : I ! water bath at 3 7 ° C for 30 minutes. Following I ' incubation, the mixtures were applied to 1.0 ml DOWEX j , 50W-X8 cation exchange mini-columns poured in disposable ! glass Pasteur pipettes, and the eluents, comprised largely I ^ I of [ H]phosphory1cho1ine, were collected. To insure j maximal separation of phosphory1choline from ACh bound to the columns, the columns were washed with 0.65 ml of i distilled water and the eluent collected from the wash was added to the first one. Five ml of scintillation cocktail j were then added to the combined eluent fractions (1.3 ml), I 3 • j and the amount of radioactivity released as [ H]choline I was determined. I _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ To calculate the total amount of released, the second 500 ul aliquot derived from the supernatant of the first post-incubation centrifugation was i diluted with 0.8 ml water (final volume = 1.3 ml) and 3 assayed in the scintillation counter. [ H]ACh released : could then be determined as the difference between the two counts after corrections for the recovery of standards. ! , I j Ass ays o f Cy toplasmic Marker s ' I 1. Lac tate Dehydrogenase ■ Lactate dehydrogenase (LDH) activity was determined by} monitoring the decrease in absorbance at 340 nM which is ' characteristic of NADH oxidation (Goldberg (1975). A 0.1 ! I ml sample was added to 2.0 ml of a solution containing 75 mM sodium phosphate buffer, pH 7.0 (2.0 ml), 1.2 ml NaOH (0.25 ml), and 0.4 ml PS. The solutions were mixed by . vortexing and the absorbance of the solution measured at | I i 340 nm with a Beckman Acta I I Spectrophotometer. Sodium j * pyruvate (3 mM) was then added to make the final volume 3.0 j ml, and, after vortexing, the absorbance ( = 3 40 nm) of the solution was again measured at time intervals of one, I ! two, and four minutes. The relative concentration of LDH 1 I released was determined for control (PS) and toxin-treated samples by comparing the differences between initial and j i final absorbance readings taken at two different time | i intervals. ! 54I 2. Choline Acetvltransferase i ' I Release of the enzyme choline acety1transferase (ChAT) I ' was assayed by a slight modification of the method devised by Fonnum (1975). Samples (lOul) to be tested were mixed with an assay solution consisting (in final concentration) of 100 uM [^H]a cetyICoA, 300 mM NaCl, 10 mM choline } , cl or ide, 1 mM EDTA, 0.1 mM e s e r ine sulfate, and 0.5 mg/ml , bovine serum albumin in 25 mM sodium phosphate buffer, pH 7.4. Following a 30 min. incubation at 3 7 ° C, the ^ I reactions were stopped by addition of ice-cold acetonitrile^ I containing sodium tetraphenylborate (5 mg/ml). Aliguots ! I (0.5 ml) of the incubated solutions were then added to 0.5 j ,ml of distilled water and 5 ml of a toluene-based j I scintillation cocktail containing 0.05% PPG and 0.02% : POPOP. The rate of [^H]ACh production, which is 1 proportional to the amount of ChAT present, was determined by counting the s ample s in a liquid scintillation counter (Bee km an LS7 500). , , 3. Ad eno sine Tr ipho s pha t e ! ' . . . I I Measurements of cytoplasmic adenosine triphosphate ’(ATP) released from the synaptosomes of both rat forebrain I and Torpedo electric organ were done under the supervision ! and in the laboratory of Dr. George Miljanich. The assay i procedure was varied to maximize sensitivity, but basically i conformed to the methods outlined by White (1978) with modifications by Schwneitzer and Kelly (pers. comm.). I______________________________ .5.5 j Firefly extracts (Sigma FLE50) were reconstituted in PS and passed over a column of Sephadex G-25 prepared in PS I containing 1% BSA. Eluted FLE50, minus arsenate and other j small molecules which have proven deleterious to synaptosomal stability, was mixed (100 ul) in a small glass: i test tube with 10 ul of synthetic luciferin (5 mg/ml) and | 3 90 ul of PS containing 1% BSA. Synaptosomes (20 - 50 ul ' for rat brain; 10 ul for electric organ) were then added to^ I I the mixture, the test tube placed into a luminometer | I (Biolog; Los Angeles, CA) and the background luminescence j I I was allowed to decay until a stable baseline had been | : . I attained. Synaptosomes were then stimulated by the rapid ^ . injection, through a light-tight septum, of the samples to ' I be tested, and release of ATP was recorded by monitoring | the light emitted as a result of the reaction of ATP with j ; the luciferin-luciferase mixture. Responses were ! j quantified by the subsequent addition of 10 pmol ATP ' external standards. Determination of Protein Protein concentrations were determined by the Bio-Rad j dye-binding assay which is based on the method of Bradford I (1976). Standards used were bovine serum albumin (in LPT-d ! assays) and bovine gamma globulin (in IDT assays). The ' standard Bio-Rad assay was used for protein concentrations between 0.2 - 1.4 mg/ml, and the more sensitive microassay — — — 56 I was used when protein concentrations ranged from 1 - 25 ug/ml. For protein concentrations less than 1 ug/ml (particularly characteristic of the later stages of LPT—d ; purification), samples to be assayed were first concentrated by ultrafiltration and then assayed under micro-assay conditions. In all cases the minimum acceptable correlation coefficient for the standard assay i i was set equal to 0.98. I Chromatographic Technioues ( I Both conventional and high performance liquid I I chromatographic techniques were used to purify the two toxins investigated in this thesis. Because the various , procedures employed differed depending on the properties of I the individual toxins, the actual methods used in the I ! different chromatographic separations will be described in the appropriate RESULTS sections. I I Concentration of Toxic Fractions i Ultrafiltration of samples was employed not only to increase the protein concentration at later purification j steps but also as a means of desalting and possibly further I purifying the samples. lOT was not concentrated by [ultrafiltration because dramatic losses of activity resulted. The procedures outlined are those used with partially purified LPT-d, most often after elution of the 57 I toxin from CM-Sephadex and dye matrix affinity columns, I Am icon ultrafilters (PMIO or PM30) were soaked in a ■ beaker of distilled water for one hour with at least three changes of the water during that time. The filters were then immersed in a solution of 0.1% BSA in 30 mM sodium ' phosphate buffer, pH 7.0 for 15 minutes, and centered within the appropriately-sized cell. Sample to be concentrated was then added to the cell and subjected to a i I pressure sufficient to generate an eluent flow rate of 0.5 \ ml/min. (approximately 10 psi). When it was necessary to minimize the salt concentration (e.g. prior to electrophoresis or in stability studies), ultrafiltered retentâtes were "washed" with one to two original volumes of 50 mM sodium phosphate buffer. [ Derivatization Procedures 1. Preparation of Fluorescein Perivatives Derivatization of toxin molecules with fluorescein isothiocyanate (FITC) was accomplished by several j modifications of the method outlined by Nairn ( 1 976 ). : Basically, FITC was dissolved in 50 mM sodium carbonate j ; buffer (pH 9.5) by crushing and stirring the particles with ; I I a glass rod for 5-10 minutes. The FITC solution (100 ug/ml) was then added slowly (in four 200 ul aliquots over two hrs), with stirring, to three ml of toxin (at least 25 ug/ml) adjusted to a pH greater than nine with 50 mM . _5_8 Na^CO^ . The reaction was allowed to proceed in , the dark at 4^C for 12 - 14 hours with constant mixing I on a multi-purpose rotator (Scientific Industries Co.). Unincorporated fluorescein was then removed by passing the solution, in 0.8 ml aliquots, over 5 ml minicolumns of Sephadex G-25 swollen in PS containing 0.1% BSA. To , accelerate the "desalting" procedure, columns containing ' the derivatized samples were centrifuged in a clinical I 1 centrifuge (see Paul son, 1974) and the fluorescence of the | I eluate (derivatized toxin) was verified by an increased , ' absorbance at 493.5 nm relative to PS - 0.1% BSA controls. ' I 12 5 . 2. Preparation of [ I]Derivatives I 1 Toxin samples were iodinated by direct (vs j ' conjugation) methods because direct iodinat ion of tyrosine I . . . . . ■ ' residues typically involves only a single reaction step and : tends to result in higher specific activity products j (Bolton, 1980). Two procedures were utilized. The I chi oramine-T method described by Van Eldik and Watters on I I (1981) was used in initial iodination attempts, whereas the milder lOUO-GEN technique developed by the Pierce Chemical 12 5 Co. was employed in subsequent [ I] derivatizations. Iodination by the chi oramine-T method was accomplished I by mixing 1 ml of toxin (15 - 20 ug) in a plastic microfuge I tube containing 20 ul (0.2 mCi) of carrier-free Na^^^I and 10 ul of chi oramine-T (25 mg/ml). The reaction was allowed to proceed for three to five minutes at room 59 temperature and then 20 ul of sodium me tabisulfite (25 mg/ml in 50 mM sodium phosphate buffer, pH 7*5) was added to reduce free iodine and any excess chi oramine-T. After 30 seconds, 200 ul of PS - 0.1% BSA was added, and the labelled protein was separated from unreacted iodine by rapid desalting as outlined above. Derivatization by the iodogen technique was based on } the procedure of Fraker and Speck (1978). To a 12x75 mm | I glass test tube that had been previously plated with ' I . . I IlODO-GEN, were added 1 ml of toxin (12 - 15 ug) and 5 ul 1 25 (100 uCi) of Na I. Incubation was allowed to proceed at room temperature for 10 minutes, and the samples were I mixed at intermittent intervals. After the incubation, a I 1.0 ml aliquot of the solution was transferred to a plastici ; I ' microfuge tube and allowed to stand an additional 5 minutes' (free of the lODO-GEN). Labelled protein was then \ I separated from free iodide, as above, by centrifugation , through Sephadex-G25 mini-columns ( 5000 rpm; 5 minutes). ' The presence of iodinated toxin was confirmed, after both j procedures, by counting aliquots of the mini-column eluates on a Beckman Biogamma counting spectrophotometer. Release I ^ _ I promoting activity of iodinated derivatives was evaluated I by the standard procedure used for assaying release of I I radioactivity from synaptosomes. 1 60 PoIvacry lamide Ge1 Electrophoresis } 1• Discontinous Slab Ge1 Electrophoresis Discontinuous polyacrylamide slab gels were prepared by the Laemmli (1970) technique. Running gels (9% for LPT-d; 13% for lOT) and stacking gels (3% in both cases) were prepared in a Bio-Rad Protein Dual 16 cm Slab Cell ’ fitted with spacers such that the gels were all 1.5 mm thick. I Samples to be electrophoresed were concentrated, if | I ' I necessary, by ultrafiltration as outlined above, or by : J ' J lyophi1ization. The samples were then diluted 1:5 with a , sample buffer consisting of 4.7 ml distilled water, 1.0 ml glycerol, 1.0 ml 10% (w/v) sodium dodecyl sulfate (SDS), 1 I 0.1 ml 2-mercaptoethanol, 0.2 ml bromophenol blue (0.05% ] w/v), and 1.0 ml Tris-HCl (0.5 M), pH 6.8. For computation: of protein molecular weights, low molecular weight i I standards (Pharmacia) including phosphoryla se b, BSA, I _ ' ovalbumin, carbonic anhydrase, soybean trypsin inhibitor, and 1 actalbumin, were typically run in the two outermost | I lanes. Occasionally individual proteins were also run as j I standards, particularly ovalbumin to verify the 43,000 I I dalton position in studies of LPT-d and cytochrome c I I (13,400 dal tons) in electrophoretic separations involving the smaller lOT molecules. All gels were electrophoresed at 40 V, constant I I j ' voltage, for 16 to 18 hours or until the bromophenol blue I 1 I t____________________ _ _______ I tracking dye had migrated to within approximately one cm from the bottom of the gel. Gels were stained either by the Coomassie blue method (crude toxin fractions, and solubilized membrane preparations) or, more often, by the more sensitive silver stain technique devised by Merril et a1. (1981). After they were stained, gels were either photographed directly (and then discarded) or dried using a ' Bio-Rad high capacity gel slab dryer. ! I 2. Two-Dimens ional Ge1 Electrophoresis ' Two-dimensional (2-D) gel electrophoresis of LPT-d was based on a modification of the protocol first described by O'Farrell ( 197 5). First dimension isoelectric focusing (lEF) gels containing ampholytes with a pH range of 3 to 10 (Bio-lytes 3/10 and 5/7) were poured in 1.5 mm (I.D.) by 150 mm glass tubes to form gels 10 cm in length. To these gels were added 50 ul aliquots of purified LPT-d . (concentrated to 25 ug/ml) mixed with 25 mg urea, and 10 ul of an lEF "concentrate" (0.2 ml Triton X-100, 0.1 ml 2-mercaptoethanol, 0.1 ml 10% (w/v)'SDS, 0.18 ml Bio-Lyte I I 5/7, and 0.02 ml Bio-Lyte 3/10). lEF was carried out at I 400 V constant voltage for 12 hours or until the current I remained stabilized for one hour. Gels were removed from j I the glass tubes by careful but rapid injection of ice-cold j 1 . ; j water into the space between the gel and the walls of the i I I tubes using a fine-tipped syringe. Freed gels were j ' equilibrated with a reducing buffer containing 0.01% ! i I ' ............. .‘ j bromopheno1 blue, and frozen at -20^C. The second dimensions were run exactly as described for the discontinuous slab gels (above) except that, in this case, there were no stacking gels. The separating gels were continuous, and thawed lEF gels were laid directly onto the upper surface^ Autorad iogranhv I Autoradiography was used to visualize the binding of I 125 ! [ I]LPT-d to synaptosomal membrane protein, to verify , the presence (and pur ity) of iodinated derivatives, and to look for changes in the patterns of 3 2 P-phosphorylation/dephosphorylation that might be due I to the presence of LPT-d. j Dried gels and nitrocellulose sheets were securely taped onto pre-expo sed film sheets to insure even exposure to film and then placed in a film holder with one sheet of 1 Kodak X-Omat AR film, and, in some cases, an intensifying screen. Exposures were typically for one to three days. ' ^ . . . 32 When the radioisotope being investigated was P, the length of time gels were kept in contact with film was I usually two to seven days* Following an appropriate ; exposure interval, the film was developed according to Kodak's manual film processing procedure (included with the film) . 63 ' Electrophvsiology Electrophy6iological studies were done in conduction I with and under the direction of Dr. Albert Herrera of the University of Southern California. The experiments were performed in Dr. Herrera’s laboratory using the sartorius : nerve-mus cle preparation of the frog Rana p ip iens. Musclesj : were pinned out to a length equivalent to 110% of the i resting length in a Sylgard chamber and bathed in a Ringer's solution (116 mM Ha Cl, 2 mM KCl, 1 mM MgCl^, * 1.8 mM CaCl_, and 1 mM NaHCO^, pH 7.26) at j J . Ô I15°C. I Ele ct rophy s iological recordings were made by conventional techniques. Microelectrodes filled with 3 M ! KCl were used to record miniature end plate potentials , I j (mepps), and (amplified) signals were measured both directly from the oscilloscope screen and by the analysis of film and strip chart recordings of control and I toxin-induced events. Since error in estimating mepp I frequency usually depends on the number of events counted j ! rather than the time period of measurement (Van der Kloot ■ and Latta, 1983), we typically recorded until at least 100 | j mepps had been registered. TOT used in these studies was | [exchanged into Ringer's solution by elution of the toxin I I through a 75 ml column of Sephadex G-25 previously I equilibrated in the Ringer's solution. ___ PA I Receptor B ind ing S tud ie s Binding of toxin to specific receptors can be ! demonstrated in at least two ways. Classical binding studies are based on the use of a labelled (active) ligand and depend on the ability to separate and measure the relative quantities of bound and free ligand after exposure I to the appropriate receptors (Wei land and Molinoff, 1981). ' j Alternately, if the ligand cannot be labelled, or can only ; i 1 I be labelled with a consequent loss of biological activity, ; I ' underivatized molecules can be used. In this case an as say I to determine the amount of "free" ligand remaining in I i solution after an appropriate exposure to receptors is ' j performed, and the extent of binding can be determined by ! subtraction of the concentration of free ligand from the total amount that was present initially (Hille, 1984). I Binding studies utilizing dériva tiz ed toxins were of | j ^ 1 ' three types, differing basically in the means by which free ! and bound toxin were separated. Separation of fluorescein derivatives could not be consistently and reliably j measured, so when these derivatives were used, binding of fluoresceinated toxin was assayed as an inhibition of the release stimulated by unlabel led toxin. Free and bound forms of the iodinated toxin were readily separable. Both ; rapid filtration and microcentrifugation techniques were I utilized. To assay the binding of fluoresceinated derivatives, 65 3 synaptosomes loaded with [ H]choline were pre-incubated for 10 minutes at 3 7 ^ C with a solution of de r iva t i z e d toxin previously shown not to effect release. At the same time an equal volume of synaptosomes was pre-incuba ted with normal PS to serve as a control of synaptosomal viability. Both pre-incubated synaptosomal suspensions were then exposed to varying concentrations of unlabelled, active LPT-d. The extent of binding of the fluoresceinated toxin was estimated based on the decrease in release elicited by I unlabelled toxin. Decreased activity was presumed to be duel I to the occupation of receptor sites by fluorescein derivatives. In many of the binding studies involving iodinated ! 1 t toxin, separation of the bound and free species was ! ' I < I effected by a rapid filtration of incubation mixtures [ 1 . through Whatman glass fiber filters (GF/B, 2.5 cm) ^ pre-soaked in PS - 0.1 BSA. When nonspecific binding to | : filters could not be eliminated by pretreatment of the filters, the derivatized samples were also pretreated. I i ' This typically involved pressure injection of the samples ; to be tested through pre-soaked circles of GF/B filters cut j to fit small (Icc) tuberculin syringes. The problem of nonspecific binding could also be I circumvented to some extent by using a microfuge tube assay I I as described by J over et al. ( 197 8) and by Harvey e t a 1. (19 84). Synaptosomes, after incubation with labelled toxin 66 ; for an appropriate period of time, were pelleted by microcentrifugation at 9,000xg for 5 minutes, washed quickly with PS, and recentrifuged. The amount of bound toxin could be calculated based on a count of activity in the pellets. The concentration of free toxin was estimated I based on the number of counts in the two supernatants. ! Binding studies conducted with unlabelled toxin were : ' similar to those described by Thieffry et a 1. (1984) in an i analysis of the binding of venom from G1vcera convoluta to ' receptors on Torpedo synaptosomes. Briefly, purified , synaptosomes were prepared as outlined above, diluted to a , concentration of approximately 200 ug/ml, and divided into . two equal portions. One pool was not loaded with ! 3 . I [ H]choline, but was further diluted, usually 100 fold, and then exposed to LPT-d either at different ' concentrations or at one concentration for varying amounts ' of time. Following the incubation period, synaptosomes I were sedimented by microcentrifugation at 9,000 x g (5 ' min.; 4®C) and the supernatants removed. Pellets were washed once in the original volume of PS and then again centrifuged to yield a pellet (containing bound toxin) and ; a supernatant which, when combined with the first ' supernatant, contained free, unbound LPT-d. The second pool of synaptosomes was labelled with 3 [ E ]choline and then exposed both to the combined supernatants and to the original concentrations of 67 (appropriately diluted) LPT-d used with the unloaded I synaptosomes* After the standard ten minute incubation ■ interval, release of radioactivity from the loaded , synaptosomes was measured according to the standard assay ' The difference between the amount of release elicited by toxin pre-exposed to unloaded synaptosomes and that characteristic of the same concentrations of LPT-d not ! pre-expo sed was taken as indicative of the amount of binding as a function of [LPT-d]. 68 I RESULTS: Purification and Characterization of lotrochotin Activity of Exudate vs Who 1e Tissue Homogenate Initial studies of iotrochotin (lOT) by Joseph Martin were done using whole tissue samples that had been collected during the 197 8 Alpha Helix Caribbean Expedition, frozen, and then returned to the laboratory for biochemical I I analyses. These samples were exhausted and had to be ! replaced. In 1982 Dr. J.S. Kittredge returned to the Caribbean, specifically off the coast of Belize, and was able to collect essentially pure I . birotulata exudate, free of any contaminating sponge tissue. Specific activities of the various tissue extracts and I spontaneously released exudate were calculated both for sponge tissue with associated zooanthid and without it. The results are summarized in Table 1. The presence or , absence of Parazoanthus swi f t i i did not seem to have any | I significant effect on activity of the extracts, although ! I I j _ I I the concentration of protein was, predictably, higher in ; ■ whole tissue homogenates that included the zooanthid j i ! species. Release activity attributed to X jl birotulata was , i determined not to be due to the presence of a commensal j I species. ! I The specific release activity of naturally-emitted j exudate was found to be over 20 fold greater than that of j I 6 9! >4 H CO H U < oü H X M <3 H <3 c=> H O od H b O U G <3 co O 4J W «r4 k C M S ü w M P m C O r-. CN O 'd- en en nO en en O m o 'd - P3 <3 H O P 3 o W G i T k . —4 i T k es o H ^ m O " es P C O b O .• • • • H P C G O P 4 ' m co i n o m CN O > 4 H H O <3 >4 H G > c o o o 'd - c r > c r > e s O O M 4 J e s e s m - 4 st rH e s I T » B H « H m - 4 r-l M U P 3 H < 3 P O S P P O 4 J 4 J P C 0 ) P P P P U 4 J *P ■ U 'P Ï0 P p P M C X •p p X C O 0) a • r 4 p a <3 b O .C a M • O O P •p ■u o P •p hQ •H G ■u p p G 4 J p M . 0 3 o w p 4 J p o p P 4 J P C t; a 4 J c ■ U o a • 4 J P 4 J C C O P •H o P P •H C 0 0 ) •P b O G N p a a G • O p P O a p p o a O c o X G 4 J œ X G e s co M o X P co a o X w •H a 1 — 1 O •H a r H H H •P r-l a H *o r— 4 x> U ■ i - t P p P p p P C Ü < • f i W Ü p U • r 4 p ü p w H P C > p 0 3 p a 1 — 1 p œ p H o •p œ ■ u o a œ 4 J X œ a c •H p C O a p •r4 P M w > M H a p a M H a O. G to co C l . G (0 C O 70 the whole tissue homogenate (Table 1), and for this reason, the exudate was utilized exclusively in all subsequent purification efforts. Mo 1ecular Sieve Chromatogranhv An aliquot (6 ml) of desalted exudate that had been stored frozen at -20°C was thawed slowly in a cold room (4°C), and then applied to a freshly poured Sephadex G-50 column (1.2x115 cm; 130 ml total bed volume) ! equilibrated in 50 mM sodium phosphate buffer, pH 7.0. . Sample application was typically followed by two 6 ml rinses to insure maximal incorporation of lOT into the resin bed. Elution, with 50 mM sodium phosphate buffer, : was effected by maintaining a constant pressure head of approximately 20 cm such that the flow rate was 10 ml/hr, and the eluate was collected in 1.5 ml fractions. In some cases the absorbance of the eluate at 280 nm was ' continuously monitored using an IS CO Model UA5 Ab sorbance I Monitor; in others, the protein concentration was : determined by the individual assay of alternate fractions. Two major peaks of protein, as reflected in the change I ! in absorbance at 280 nm, were resolved (Figure 1). The I i first, which eluted in the void volume, was pink to red in I ' color. A second, broader peak was characteristically golden-brown and emerged at approximately one complete ' column volume. I Figure 1. Chromatography of Iotrochotin on Sephadex i G-50. Desalted exudate (1-2 mg protein) from I. j birotulata was applied to a Sephadex G-50 column (1.2 x I 115 cm) and eluted at a flow rate of approximately 10 I ml/hr. Eluate was collected in 1.5 ml fractions. Fractions were assayed for releasing activity (- - -) and absorbance at 280 nm ( — — ). 72 '280 .3 .2 80 60 .1 40 20 0 70 10 30 50 40 ACTIVITY ICPM X 10*^1 [— ] FRACTION NO. L_. 73 All releasing activity was concentrated in a single I peak eluting at approximately 60 ml. This peak was coincident with a small, but definite increase in the concentration of protein which was approximately equidistant between the two larger protein peaks. Fractions with release promoting activity greater than 10,000 cpm above background (approximately 2.5 units/ml) I were pooled. I By comparison with the elution profile of a set of ! molecular weight standards run under identical conditions over the same column, the molecular weight of lOT was ! calculated as approximately 17 , 500 dal tons. An ion Exchange Chroma togranhv Experiments by Martin had indicated that lOT purified I by gel filtration had at least some negative charge at I 'neutral pH. The toxin bound to all three anion exchange , resins tested at pH 7, but not to several cation ' exchangers. Since there was some evidence that release I promoting activity could be recovered from anionic Cel lex D columns by elution with 0.2 M NaCl, it seemed prudent to I test these columns as a potential next step in the I purification process. I Cellex D (Bio-Rad) powder was swollen in 0.25 N NaOH, ; washed with the same concentration of HCl, and then rinsed in 0.25 N NaOH. After equilibration in 20 mM sodium 74 phosphate buffer (pH 7.0), a 50 ml column was prepared. * Twenty ml of the pooled, active G-50 fractions were diluted with distilled water to the same sodium phosphate ■ concentration (20 mM) and then applied to the column in 10 ml increments over a two hour period. After the sample had; completely entered the cellulose, the column was washed ^ ■ with 20 ml of buffer and then exposed to a linear gradient ' of NaCl increasing to 0.2 M over three column volumes. 1 i I Two distinct peaks of releasing activity were | recovered (Figure 2). One peak eluted during the wash and ; I apparently had only very slight affinity for the positively; charged matrix. The second, broader peak consistently ■ j eluted at a NaCl concentration of 0.1 M. Since the more I I j acidic pool had greater specific activity and a generally | I ' ; higher yield, fractions collected from that pool with i activity greater than 2000 cpm above background were I retained. These were either frozen immediately at j I -20°C or stored briefly (4 to 5 days) at 4°C for | I i I further experiments. | j i I I . . ■ ! Affini tv Chromatogranhv | I ! I Since dye-1igand chromatography, which is a separation! based on the affinity of some molecules for synthetic I textile dyes, has been used to purify a large number of 1 proteins, an attempt was made to purify lOT by this method.i I I ; Even if lOT itself did not bind, it was felt that Figure 2, Chromatography of Iotrochotin on Cellex D, The active fractions from the Sephadex G-5 0 column were pooled, diluted with distilled water to a final sodium phosphate concentration of 20 mM, and applied to a Cellex D column (bed volume = 50 ml). lOT was eluted with a linear salt gradient following a 20 ml wash. Each fraction collected was 1.5 ml in volume. Release promoting activity was determined for 50 ul aliquots taken from every other fraction and is expressed as cpm above background. 76 10 ^ ACTIVITY ICPM X10' I I 80 60 40 FRACTION NO 20 [NaCl] IMI 77 contaminants might have an affinity for one of the dye matrices and therefore be removed, effecting an increase in the specific activity of lOT. Aliquots of G-50 level lOT were passed over five different Am icon Dyematrix columns that had been washed ^ thoroughly with 8 M urea and subsequently equilibrated in ^ 50 mM sodium phosphate buffer (pH 7.0). The dyes tested ' were Blue A, Blue B, Red A, Green A, and Orange A. None of the five columns were able to bind lOT. All activity I eluted during the washes, and there was no change in the I specific activity of lOT recovered from any of the columns. High Performance Liquid Chromatography In an effort to accelerate the process of purification without sacrificing resolving power, high performance liquid chromatography (HPLC) was employed as a potential ! second step in the purification scheme. The HPLC system used was a programmable Varian Model 5020 gradient liquid ; chromatograph. Isolation of lOT on the basis of its mildly acidic I character was attempted using a SynchroPak AX-300 anion 1 exchange column (4.1 mm x 25 cm) equilibrated in 20 mM j sodium phosphate buffer, pH 7.0. G-50 purified lOT eluted ■ ^ j as at least two poorly resolved peaks within the wash phase; I of the program (Figure 3), before a gradient had been I i initiated. The specific activity of one ml fractions I ________________________ _ _ -7-8 Figure 3, High Performance Liquid Chromatography of Iotrochotin on SynchroPak AX-300. lOT (500 ul) purified by elution through Sephadex G-150 was eluted through a SynchroPak AX-300 anion exchange column (4.1 mm x 25 cm) which had been equilibrated with 20 mM sodium phosphate buffer, pH 7.0. Twelve minutes after the sample was injected (arrow), a linear salt gradient (0 to 0.2 M NaCl) was initiated, and 1 ml fractions of the eluate were collected over the entire range of the gradient. The chromatogram is a continuous plot of the absorbance of the eluate at X - 280 nm. Absorbance range = 0.02; 1 mV full scale. 79 <v CO n j 4 20 8 12 16 24 I Retention Time (mln) 80 representing these peaks was not increased, and there was no significant evidence to indicate that contaminating proteins had been removed. Sod ium Dodecvlsulfate-Polyacrylamide Gel Electrophoresis One criterion often used to evaluate the purity of a ; protein is the presence of a single well-defined band after I SDS-polyaerylamide gel electrophoresis (SDS-PAGE). I I Electrophoresis was used to monitor the purity of lOT at I all steps in the purification. A 13% gel stained initially with Coomassie blue ' (Figure 4) showed the whole tissue homogenate to be made up of several proteins at least eight of which formed recognizable bands. The most prominent band was a diffuse one centered at approximately the molecular weight of cytochrome C (13.4 kD). Other bands were less well defined! ; I I and of greater molecular weight. One particularly well I stained band was present at approximately 17 kD, and j another had a molecular weight very nearly that of trypsin j I I ' inhibitor (24 kD). . j I Naturally-emitted exudate was characterized by only - one band (17 kD) in this Coomassie blue-stained 1 'preparation. Amplification of the protein sensitivity by ' I staining the gel with silver revealed at least four ; additional bands that were well defined. Two were of ' _ 81 Figure 4. Sodium Dodecy1sulfate-Po1yacrylamide Gel Electrophoresis of Iotrochotin. Fractions of whole tissue homogenate, induced exudate, tissue homogenate, and naturally-emitted exudate were electrophoresed on a discontinuous slab gel. The separating gel was 13% aerylamide, and the stacking gel was 3% aerylamide. Electrophoresis was accomplished by applying a constant voltage (40 V) across the gel for 16 to 18 hrs or until the tracking dye (bromophenol blue) had migrated to within one cm of the bottom of the gel. Standards electrophoresed in the left lane were prepared by mixing reagent grade trypsin inhibitor and cytochrome c The gel was stained with Coomassie Brilliant Blue R-2 50 82 m c 0) O) o E o X d> 3 CO co p o « ■O 3 X LU TJ 0> O 3 TJ C fO c o o> o E o X co co 0) co TJ 3 X LU TJ m E LU CO hm 3 CO Z Trypsin Inhîb. i (24 kD) Cytochrome C (13.4 kD) 83 slightly greater molecular weight (but less than 24 kD), and two others were of such large size that they only migrated about 10 mm into the separating gel. G-50 purified lOT, run on several other 13% gels stained with silver, was characterized by at least three distinct protein bands ranging in molecular weight from 17 to 20 kD (gels not shown). Typically, fractions drawn from the peak of activity showed most intense staining in a protein of approximately 19 kD whereas those drawn from the leading and trailing edges were less concentrated in that particular protein. lOT purified by elution through Cellex D was consistently characterized by a single band corresponding to the 19 kD G-50 protein (not shown). This was true whether the separating gel contained either 9 or 13% aerylamide, and with staining times as long as 20 minutes. By the criterion of a single band on SDS-PAGE, Cellex D level lOT appeared to be pure. Summary o f the Pur if ication of IDT Since neither affinity chromatography, nor HPLC added significantly to the purification of lOT, these steps were omitted (Table 2). By beginning the purification sequence with naturally-emitted exudate as opposed to the whole tissue homogenate used by Martin, we were able to start with Sample 22 fold more concentrated in release promoting 84 r ' d o •H s s j- l o d o < u •H < u *4 S u •H 00 U (U d Q P4 s s o •H • d > ^ 1 — 4 •H X (U J-l ‘ M O >* <1 O O • • 0 0 1 — H m o m o o irv e s e s g • o M H o > , 00 .m l -w g U 44 •H M •H > CO P C 4 U •H 44 M < u 44 .m # od ou O d O c o <5 g P W H O d M •H < U 00 P C 4 44 a O O U >> PU od s S Ss T-^ D 44 a c o •H > d • i 4 44 • 44 •H es O d d < u 1 — 4 -û d d •H m U H d 8 44 O 00 u a d4 e u a * — 4 o > a o •H •U •H T3 d o u a o ir* e s vO o o o e s e s vO 0 0 e s vO i n • * o 1-4 O • o m m m I o i r » t - H f - 4 co X es o uo O o O uo O vO o> sf t - H co Q d 44 X d d T3 o r - 4 d UO r — 4 X 1 0 > M o o 85 [activity. Passage of the toxin over sequential columns of Sephadex G — 50 and Cellex D resulted in an overall 30.8 fold increase in specific activity over that of the exudate, but, only a 22% yield. Reversal of the order in which the chromatographic columns were used seemed to make little ! I difference in our effort to increase the recovery of j releasing activity. In fact, the yield was probably lower,| I since pooled active Cellex D fractions had to be desalted | j I(over another column) prior to their being eluted through i i I I Sephadex G-50. ! I j Stability of lOT i The spontaneously emitted exudate, which had been I received at room temperature in a solution of 0.1% sodium I I azide (in seawater), retained its stability when kept frozen at -20^C for at least three years. Furthermore, at protein concentrations greater than 200 ug/ml, the exudate could be stored at least one month at 4^C with no apparent loss of release-promoting activity. I G-50 level IDT was also remarkably stable. Pooled j active fractions could be frozen at -20^C or Ilyophilized and subsequently reconstituted with no I appreciable loss of activity.. lOT at this stage was less I I stable than exudate however, to long term storage at I ' 4°C. Activity measured after 30 days storage in a cold I j room was generally 30 to 40% less than it had been j 861 I_______ . ] immediately following elution from the G—50 columns. I Probably because the protein concentration was so low, i , unmodified Celiez D level lOT rapidly lost all activity 'when kept at 4^C. These samples could not be frozen I ^ , (-20 C) or lyophilized without a complete loss of activity, and seemed to be only partially stabilized by the addition of BSÂ to a final concentration of 0.1%. Typically, experiments conducted with IDT purified through ' this step were performed within one week of the I pur if ication. Time Course of IDT-Induced Release The effect of incubation time on lOT-s t imula ted release of radioactivity from synaptosomes was examined. ! For release at time periods of one to eight minutes, the I standard release assay was used, the time of incubation : being taken as the time between addition of synaptosomes to I the sample and immersion of incubates in an ice-water slush. I ; By 30 seconds, release induced by 40 nM lOT was ! essentially maximal (Figure 5) with only a very slight ! increase, possibly attributable to decreased synaptosomal I I viability, recorded at time intervals greater than one 1 jminute. Because decreasing the concentration of IDT did I not have any effect on the shape of the curve defining release as a function of time (McClure, pers. comm.), it 87 Figure 5. Time Course of lotrochotin Induced Release of Radioactivity. Time points represent the intervals between first exposure of loaded synaptosomes to lOT and the immersion of incubates in an ice-water slush. The lOT concentration was approximately 40 nM. 88 AC TIVITY (UNITS) 2.5 1 . 5 0 . 5 TIM E (MINI L . 89 I was considered unlikely that reliable estimates of initial , velocity could be determined using the standard release assay. A standard incubation time of 10 min. was therefore , used in all subsequent experiments. Effect s o f Varying the Concentration o f lOT on Release Release of radioactivity from synaptosomes incubated I for 10 minutes with different concentrations of lOT is ! I recorded in Figure 6. The data define a rectangular hyperbola. When redrawn as a Lineweaver-Burke double ! reciprocal plot (Figure 6, inset), the data fit a straight ' line (r = .981) and allow computation of R. _, the lOT I I i concentration at which release is half maximal, and ' ; the maximal amount of releasable I ! KQ ^, estimated by extrapolation of the line through i the X—axis, is approximately 19.2 nM, based on a molecular : weight of 18 kD for lOT. Maximum release activity ■ i6 estimated, by this method, to be 4.08 units or very nearly the amount of radioactivity typically ! released by lysis of (control) synaptosomes incubated in I ! PS. with respect to the proportion of total released 'radioactivity that is specifically [^H]ACh, experiments ! have consistently shown that slightly less than half of the I 3 released activity can be attributed to [ H]ACh. [McClure and Martin (1983) found that approximately 49% of i L 90 I Figure 6. Effect of Varying the Concentration of 1 I lotrochotin on Release of Radioactivity. Data have ! been plotted as release —promo ting activity vs. the I I concentration of lOT. The incubation time was 10 j I minutes. Data points are the average ± standard deviation of triplicate assays. Inset: Double reciprocal plot of the same data, r = 0.981. 9 1 ACTIVITY (UNITS) 3 1.0 2 as 1 .10 -.05 O 25 50 100 125 150 92 I subsequent experiments place the proportion I o o ! ([ H]ACH/[ H]total) at 45 to 54%. This ratio is i ; almost identical to that characteristic of release from synaptosomes incubated in control media and significantly less than the ratio of 65 to 70% characteristic of release ( induced by depolarization (with 50 mM K*). Ca1cium Dependence of Re lease i To test the effect of varying the extracellular I 2 + I calcium concentration [Ca ]^ on release induced by : . 3 . ; 40 nM lOT, synaptosomes were loaded with [ H]choline as 1 usual, but then washed in a modified PS containing no ; 2 + i [Ca ] and buffered with HEPES (pH 7.0). Prior to I o I I incubation with lOT, the synaptosomes were suspended in modified PS solutions which included (in final concentration) 0, 0.5, 1.0 and 4.0 mM Ca^*. I Compensation for differences in [Ca^*] were made I , 2+ + by varying the concentrations of Mg and Na such I that all solutions were isosmotic with the synaptosomes. I ! Figure 7 shows that the release stimulated by IDT is 2 + essentially independent of the extracellular Ca ion ' concentration. Martin and McClure (1982) have reported I that removal of Ca^* from the incubating medium (and I addition of 1 mM EGTA) decreases the effectiveness of lOT I by 28%, but that effect has not proven to be reproducible 93 r ' Figure 7. Effect of Varying the Extrasynaptosomal j Concentration of Calcium on Release of Radioactivity ! Induced by lotrochotin. Loaded synaptosomes were bathed I j in modified PS containing HEPES buffer and different concentrations of CaCl^. Isosmoticity was maintained by varying the concentrations of MgCl^ and NaCl. Data points represent the average ± standard deviation of quadruplicate assays. L 2 'L A C T IV IT Y (UN ITS) 1.0 2.0 [CaJglmM} 3.0 - I 4.0 L, 95 . In several experiments evaluating lOT-induced release in 2 + the presence or absence of [Ca ]^, lOT has I ' consistently effected release of radioactivity from : synaptosomes in a manner not requiring extracellular Ca^*^. Effect of lOT on Miniature End Plate Potentials 1 a t the Frog Neuromus cular June t ion Electrophysiological studies of transmitter release are far more sensitive to presynaptic events than the corresponding neurochemical studies of appropriately stimulated synaptosomes. An analysis of miniature endplate potential frequency and size, for example, can provide a great deal of information about the process of transmitter release at millisecond time resolution. Quantized release of ACh should be reflected in an increased mepp frequency, or, when mepps are small or absent, in an increased size of nerve s t imulus-ev oke d synaptic potentials (Hubbard e t a 1 . . 1969). Both frequency of mepps and size of epps were investigated at lOT-treated frog sartor iu s nerve-mus cle j une t ions. In three separate experiments, with concentrations of partially purified lOT as high as 50 x ^ there was never any significant change either in mepp frequency or amplitude. Muscle fibers were, however, rapidly j I depolarized on contact with lOT, the onset and extent of [ 96 j depolarization dependent on the concentration of lOT in the, ' bathing medium. At a concentration of approximately 3.5 : ug/ml (very roughly 5 x ^), G-50 level lOT caused ; damage to and effected depolarization of about one third of the muscle fibers examined after 10 minutes exposure to the toxin. With a 10 fold mo re concentrated solution of I ' exudate (exchanged into Ringer’s solution), all of the ^ 1 . . . . . I imuscle fibers were depolarized within four minutes of the j first contact with lOT, and all had sustained extensive I damage. ! i ^ ^ I Endplate potentials could not be reliably evaluated following exposure to lOT due to the massive fiber damage. Addition of d-1ubo curarine at concentrations sufficient to | I block ACh induced depolarization of muscle fibers had no I j effect on the lOT induced changes in membrane potential. j I i Î Re lease of Tvp i ca1 Cv t op la smi c Markers I i I ; To ascertain whether or not lOT effects release of i synaptosomal radioactivity by simply lysing the : ^ ! Isynaptosomes, the releasè of typical cytoplasmic markers as I I a result of exposure to lOT, was investigated. The I experiments were similar in purpose and design to those described by Fussle e t a 1. (1981) and Suttorp e t a 1. ] i (1985). I Lactate dehydrogenase is one of the most commonly used I markers of cellular (and synaptosomal) integrity. Presence! ,7: of LDH, at quantities greater than controls, in incubation media which included toxin, woud be considered good t evidence for synaptosomal disruption, but lOT, at concentrations ranging from 0.7 to 3.5 ug protein/ml, consistently caused less than 10% of the soluble LDH pool I to be released. | I j ' When the effects, on release of LDH, of lOT (40 nM) | I ' and 50 mM K are compared (Figure 8), it becomes j apparent that LOT effects no mor release than can be | I . ; j expected under depolarization conditions. Under the I I ! I conditions of the test, lOT released about 6.1% more LDH | than was released by controls. Exposure to 50mM ^ caused an approximate 5.7% increase in the quantity of LDH liberated relative to controls. ! ! I The enzyme choline acetyItransferase (ChAT) is anotheri ; good marker of cholinergic nerve terminals (More1 and I ' Meunier, 1981). If the effect of lOT is to produce | I transmembrane lesions rather than cause outright lysis of j the synaptosomes, ChAT molecules, because they are about j one half the size of LDH (68 vs 134 kD), might be expected I to provide some information as to the relative size of the j lesions. The smaller molecules, for example, might be able to pass through lOT-formed pores that were undetected in 1 studies with LDH. Results of two separate experiments i 1 however, have shown this not to be the case. lOT, at a concentration sufficient to effect maximal radioactivity 98 Figure 8. Effects of lotrochotin and Elevated on Release of Lactate Dehydrogenase From Synaptosomes. Release of lactate dehydrogenase (LDH) is plotted as the change in absorbance at 3 40 nm accompanying the enzyme induced oxidation of NADH. Reactions were typically allowed to proceed for five min. The concentration of lOT was 40 nM. Light bars represent the amounts of LDH released by exposure to PS, 50 mM , and lOT. Dark bars reflect lysate values for each of the three conditions. The values plotted are means of two experiments. 99 .200 O « < .100 m m # Ji Si i i t t i C o n trol SOmM K+ lO T 100 release, induces only a 5.3% increase in ChAT activity over that characteristic of controls. I I The nucleotide adenosine triphosphate (ATP) is found, ; in synaptosomes, both free within the cytoplasm and ! i contained inside synaptic vesicles (White, 1978). If the ' vesicular pool of ATP is released (by repeated I I I depolarization), then any subsequent release of the 1 _ I nucleotide can be used effectively as a gauge of release from the cytoplasmic pool. Furthermore, because the I i ! molecule is so small (M.W. = 0.5 kD), if the mode of action^ ! of IDT is to produce lesions within synaptosomal membrane s,I I ATP should be released. I Experiments with both rat forebrain and Torpedo , electric organ synaptosomes indicate that lOT does not cause release of a very large pool of ATP which is not , otherwise releasable by high induced depolarization I (data not shown). The ATP is released in the presence or ’ 2 + I absence of CA , and the release induced by lOT is , quanlitatively very similar to that recorded when the detergent Triton X-100 (at a final concentration of 0.1%) is added to the incubation medium. This suggests that lOT ; might have de tergent-1 ike properties. If that were true, I small molecules like ATP (and ACh) could be expected to pass relatively freely through what would be "leaky" synaptosomal membranes (Baker and Knight, 1984). 101 3 Re leas e o f 2-[ 1Deoxy-D-Glucose To confirm the hypothesis that lOT effects a nonspecific release of radioactivity by permeabi1izing the synaptosomal membrane, synaptosomes were loaded with 2-[^H]DG. Figure 9 shows that 2-[^ ^ ^ ^ 3 phosphorylated derivative 2-[ H]DG-6-phosphate) is released by partially purified 10T in a concentration dependent manner. Since 2-[ H]DG is not taken up by synaptic vesicles, its release must be occurring from the ■cytosol . 3 That significant release of 2-[ H]DG is not effected by depolarization with 50 mM (Figure 9, Inset) is considered further evidence that release of the tritiated glucose moiety is from the cytoplasmic pool. At a concentration of 50 mM, causes the synaptosomes to ! depolarize to an extent sufficient to release about 25% of 3 the [ H ]ACh releasable by lysis with distilled water. 3 The amount of 2-[ H]DG released under these conditions however, is only 5.6% of the lysate value, and only 8.2% above background. Other experiments have shown that lOT-stimulated i 3 release of 2-[ H]DG can occur even in the absence of I 2 + extracellular Ca . Therefore the process by which 10T (effects release of 2-[^H]DG is similar to that by which the toxin stimulates release of [^H]ACh. Both I 3 3 2-[ H]DG and [ H]ACh appear to be released from 10 2. 3 Figure 9. Release of 2-[ H ]Deoxy-D-Glucose by 3 lotrochotin. Synaptosomes loaded with 2-[ H ]deoxy —D- ! I glucose (2-DC) were exposed to varying concentrations of * lOT for the standard incubation time of 10 min. Release of the tritiated glucose moiety (expressed both as cpm above background and in terms of units of releasing I I activity) is plotted as a function of the concentration ' of lOT. At saturation, approximately 95% of the 2-DG releasable by lysis has been released. Inset: Release of j \ 2-DG induced by exposure of the loaded synaptosomes to 1 50 mM . Release is defined as a percentage of lysate values. Condition number 1 (light bar) represents actual release ; condition number 2 (dark bar) indicates the value predicted if 2-DG were released by 50 mM in the same way that [^E]ACh is. 103 2-DG RELEASE a 3 (CPM xIO"*) .5 [i Ot ] ( pg/m l 1 04 pools not significantly affected by depolarization with ' . + high K , and release in both cases is a 2 + Ca —independent phenomenon. B ind ing o f IOT t o Synaptosomal Memb rane To investigate the possibility that IOT might be exerting its effect by binding to synaptosomal membranes, release-promoting activity was measured before and after exposure of the toxin to a crude synaptosomal fraction (P2). The results, illustrated in Figure 10, suggest that t as much as 90.5% of the activity attributable to IOT (63 nM) is lost following a 10 minute pre-exposure of the toxin to membranes including those of presynaptic nerve terminals. There is also an indication, in this I preliminary experiment, that IOT is effectively bound irreversibly, since extensive washing (with vigorous yortaxing) fails to liberate a significantly greater amount of active toxin than was present following the initial exposure to the P2 fraction. I ! In an attempt to clarify the binding kinetics, release ! induced by native (underivatized) IOT was measured at six different toxin concentrations both before and after | exposure of the toxin to purified synaptosomes. The I I I results are illustrated in Figure 11a. IOT is completely i ! bound to synaptosomes (3.5 ug protein/ml) until the toxin I ' concentration exceeds 10 nM and the amount of binding, I 105 FigurelO. Binding of Native lotrochotin to Crude Synaptosomal Membranes. The binding of IOT to crude synaptosomal membranes was investigated by evaluating ; the release promoting activity of toxin previously I exposed to a crude synaptosomal fraction (P2). Decreased release activity following exposure to synaptosomes was I considered evidence of binding. Condition number 1 indicates background release characteristic of synaptosomal exposure to PS. Condition number 2 reflects the release promoted by 63 nM IOT, condition number 3 I I the release induced by the same concentration of IOT previously exposed to a P2 fraction, and condition 4 the ; release elicited by vigorously washing the synaptosomes to which IOT had been bound. 106 CO 3 - c 3 0> 2 CO (0 £ 9 C C 107 ‘ Figure 11. Binding of Native lotrochotin to Purified I Synaptosomes. (a) Release of radioactivity induced by IOT previously exposed to synaptosomes (open squares) and by toxin not previously exposed (closed squares) is plotted as a function of the concentration of IOT. Data points are the means of triplicate assays, (b) Replot of the data in (a) showing the extent of binding at each ■ toxin concentration as the amount of decrease in release i promoting activity. 108 9 OC 3 2 1 50 75 25 [IO T] (nM) 0 » « 9 ? OC 9 m ( D 9 U 9 O 2 1 0 50 75 25 [lO T ] (nM) 109 j which is reflected in the difference between the releasing I activities, continues to increase until the IOT ! concentration reaches approximately 35 nM (Figure 11b). At greater toxin concentrations, IOT binding sites appear to ! have become saturated. If the maximal decrease in release i 1 activity is estimated to be approximately two activity I units, the for binding can be calculated as 13.5 nM ! I which is in excellent agreement with the K_ (= I ^ K q ^) calculated for release (19.2 nM). Studies of the time course of binding show binding to be at least as rapid as the release-promoting event. At a I i concentration of 16.2 nM, all IOT was bound to a crude | I synaptosomal fraction (approx, 1.5 mg protein/ml) as early I i as one minute after the first exposure to synaptosomes. . All of the toxin remained bound over the full 12 minute j I duration of the experiment. It should also be noted that I interaction with the membranes did not simply inactivate the toxin (which like binding, would also be reflected in I decreased releasing activity). There was no significant decrease in IOT activity as a function of the time of exposure to synaptosomes even after 12 minutes. Comparison o f the Effects o f lo tro cho tin and Dig itonin on Re lease of Gytoplasmic Marker s Digitonin has been used to permeabilize adrenal medullary chromaffin cells (Dunn and Holz, 1983; Wilson and 110 Kirshner, 1983), and other detergents have proven valuable I because of their ability to render synaptosomes leaky (McGraw e t al». 1 980). To determine if IOT acts by a mechanism similar to that of digitonin, comparison studies I of the toxin- and detergent-stimulated release of a series of size-graded markers were done. The concentrations of both digitonin (Sigma) and IOT were adjusted such that, in final solution, each evoked release of radioactivity equialent to approximately 7 5% of the maximum (lysate) value. As is evident in Figure 12, digitonin effects release of the protein markers LDH and ChAT whereas IOT does not. Both digitonin and IOT are able to induce release of previously incorporated 2-[^H]DG to approximately the same extent as they are able to release 2 - [ ] ACh/cho1ine. Digitonin (75 ug/ml) induces release of up to 80% of the total recoverable labelled glucose both in the presence and absence of 2 + extrasynaptosomal Ca . Under identical conditions, 65 InM IOT releases approximately the same amounts. I Effects of Svnaptosoma1 Pre-Depolarization on lOT-Induced Re lease I If the radioactivity released by IOT were primarily I ; contained within synaptic vesicles, incubation of the toxinj I I with loaded synaptosomes that have been previously exposed j I Figure 12. Comparison of the Effects of lotrochotin and Digitonin on Release of Cytoplasmic Markers. The release of lactate dehydrogenase (LDH) and choline acetyltransferase (ChAT) as a result of synaptosomal exposure to three different conditions is presented. "C" refers to control conditions (PS); "I" indicates exposure to 40 nM IDT; "D" reflects treatment with digitonin (75 ug/ml). Values plotted are the means + standard deviation of triplicate assays. 112 I LDH CAT 5 r .3 1 “^340 50 40 30 DPM xio T2 20 10 1 1 3 : to depolarizing conditions should not result in the recovery of any additional radioactivity (exceeding control' [ ' levels). To test this hypothesis, crude synaptosomes were , 3 . loaded with [ H]choline as usual and then incubated in a saline solution containing 32.3 mM for 10 minutes (3 7 ° C) . Following the incubation, synaptosomes were j , exposed to normal saline, 65 nM lOT, or 55 mM , { I incubated an additional one to eight seconds and then I ' rapidly filtered over GF/B filters. i Figure 13 shows that release stimulated by lOT is , unaffected by a 10 minute pre-depolarization of synaptosomes. As was the case with normally loaded and ; washed synaptosomes, lOT effects a very rapid release of ! radioactivity, with evidence of saturation being attained ' as early as eight seconds after the first exposure of ' synaptosomes to the toxin. By comparison, a second incubation of the synaptosomes with PS containing elevated ; elicited no further release at all time intervals. 114 Figure 13, Effect of Prior Depolarization of Synaptosomes on Release Induced by lotrochotin. 3 Synaptosomes loaded with [ H ]cho1ine were exposed to depolarizing conditions (32.3 mM K^) for 10 min and then treated, for varying lengths of time, with lOT, 55 mM , or PS. Time intervals were defined by rapidly filtering (under vacuum) the treated synaptosomes through glass fiber (GF/B) filters at the times indicated. Release of radioactivity is plotted as total recovered counts without allowance for background (defined by the PS curve). 115 0 CO 1 s ■g X “ s 3 ? 4> ' 0) (0 « OC lOT PS Tim e (sec) L6. j RESULTS: Purification and Characterization of LPT-d Molecular S ieve Chromatography Following the procedure outlined by Yoshino (1980), ; lyophilized hemolymph collected from fourth instar larvae of Lep t ino tarsa dec em1ine a ta was reconstituted in a 50 mM j sodium phosphate buffer, pH 7.0. Normally 150 mg of the lyophilizate was dissolved in three ml of cold (4°C) buffer, and all subsequent procedures were carried out in a i cold room. After being subjected to a gentle vortexing to insure maximal dissolution, the reddish-brown suspension ; was centrifuged ( 9000 x g ; 5 min; 4°C) to remove any undissolved material. The resulting supernatant was then applied to a 1.2 x 102 cm column of Sephadex G-150 previously swollen in 50 mM phosphate buffer. Fractions I (1.5 ml) were eluted at a constant rate of 12 ml/hr by | using a Buchler Polys taltic pump, and elution was continued I I i I until 80 fractions had been collected. i Typically the release-promoting activity eluted over a| I broad range approximately centered at fraction 45 (Figure I I 14). The elution pattern was very similar to that describedi I I I i by Yoshino. There was evidence that at least two peaks of i 1 I I activity were contained within the one large peak. ' , I I Subsequent gel filtration experiments, performed with I larger columns, occasionally effected a partial resolution I I of the two active pools. At this step however, all L Figure 14. Chromatography of Leptinotarsin-d on Sephadex G-150. Lyophilized hemolymph (150 mg) of L. decemlineata was reconstituted in three ml of cold 50 mM phosphate buffer, pH 7.0. Following clarification of the resulting suspension by centrifugation (9000 x g ; 5 min), the supernatant was applied to a Sephadex G-150 column (1.2 X 102 cm) and eluted at a constant flow rate of 12 ml/hr. 1.5 ml fractions were collected. Activity of the individual fractions is reflected as the number of units in a 10 ul aliquot. 1 1 8 « "c « (0 (0 ® ® OC 3 2 1 0 8 0 6 0 4 0 20 Fraction N um baf IJ.9J fractions with releasing activity greater than 100 units/ml' were generally pooled and either lyophilized, frozen, or further purified. In separate experiments, columns (1.5 x 115 cm) of Sephadex G-150 were calibrated by elution of a series of I protein standards. By comparing the elution volume of LPT-d] I with those of proteins eluted through the same column, the | ! : molecular weight of LPT-d was estimated to be approximately I I 4 5,000 dal tons. In three different experiments, active I I fractions consistently eluted in volumes typical of ; proteins in the 37 to 47 kD range. J Anion Exchange Chromatography ; I Yoshino (1980) had shown that LPT-d would bind to and | could be subsequently eluted from the anion exchanger j DEAE-Sephadex. For that reason pooled active G-150 ! ,fractions (30 to 40 ml) were usually next passed over an 8 I to 9 ml column of DEAE-Sephadex prepared in 50 mM sodium phosphate and washed to a pH of 7.0. | 1 . . . ' j A sample was applied in five ml aliquots over a period, I I I of about two hours, and then the column was washed with two i jcolumn volumes of buffer. Elution of activity was effected ! ’by exposure of the column to a linear salt (0-0.4M NaCl) ! I ; gradient of 100-120 ml. The flow rate was 10 to 15 ml/hr. j Activity was consistently recovered as a single well defined peak which eluted at a salt concentration of 0.1 M __________ _l_20j ; (Figure 15). Fractions (1.8 ml) with activity greater than I 20 units/ml were pooled. I Cation Exchange Chromatography In addition to having some affinity for anion exchange, resins, LPT-d also binds to negatively charged matrices, I and therefore presumably has some positively charged I domains. An attempt was made to purify G-150 level LPT-d by| I ^ I taking advantage of its ability to bind to columns preparedj with CM-Sephadex, but results were not significantly be tter j : than was found when DEAE chromatography followed the G-150 [ : step. For this reason chromatography on CM-Sephadex has : been routinely employed as the third step in the I purification scheme. Pooled, active fractions collected from the | I DEAE-Sephadex columns were diluted 1:1 with 50 mM phosphate j buffer and applied to an 18 ml column of CM-Sephadex (C50) . I (After the bed was washed with two column volumes of buffer , . to elute superficially bound protein, a linear (0 to 0.75 M ) NaCl gradient was applied at a rate comparable to that used with the DEAE-Sephadex column. The total volume of the Jgradient was approximately five column volumes. ^ Fractions (2.0 ml) with activity greater than eight I uni 15/ml typically formed a sharp peak which eluted at a salt concentration of approximately 0.2 M (Figure 16). 121 I Figure 15. Chromatography of Leptinotarsin-d on I DEAE-Sephadex. The active fractions from the Sephadex I G-150 column were pooled and applied directly to a j DEAE-Sephadex (A50) column (8-9 ml bed volume). LPT-d I ' was eluted with a linear salt gradient (- - -) of I I approximately 10 column volumes. Eluate was recovered in | 1.8 ml fractions, and the activity plotted indicates the ' amount of release elicited by 25 ul aliquots of each f ract ion. 1 2 2 I 0) c I 0) (0 CD ' .«> «> OC 3 2 1 sr ' 0 8 0 60 20 4 0 “t 0.4 Z 0) 0.2 O -* O Fraction Number 123 ; Figure 16. Chromatography of Leptinotar sin-d on j GM-Sephadex. Active fractions from the DEAE-Sephadex j column were pooled, diluted 1:1 with 50 mM phosphate buffer, and applied to a CM-Sephadex (C30) column (18 ' ml). Elution was effected by imposing a linear salt j gradient (- - -) of approximately five column volumes, i I and the eluate was collected in 2.0 ml fractions. Activity is indicated for 50 ul aliquots of the different fractions. 1 2 4 r ' ( 0 « « oc 2 1 0 80 6 0 4 0 20 Z ; .3 » ; O J 0 Fraction Number 125 Chromatofocusing Because LPT—d binds to both anion and cation exchange resins, it seemed to have both acidic and basic characteristics, a feature which, although not unique, might permit the toxin to be separated from contaminants i because of differences in their isoelectric points. To that! ; end a chromatofocusing column was set up to effect a i ' chromatographic separation of proteins based on differences I j in their ability to migrate through a pH gradient. I i A 17 ml column of the anion exchanger PBE 94 I (Pharmacia) was prepared and washed with imidazo1e-HC1 ] i ■ buffer until eluate fractions had a constant pH of 7.4. To | : test the system in such a way that releasing activity could i I ^ I ■ be readily measured, a three ml aliquot (approximately 47 5 i I I I units) of G-150 level LPT-d was applied to the column I I following an initial application of Polybuffer 74 j (Pharmacia) adjusted to pH 4.0. The sample was eluted with | ! ^ 1 Polybuffer until the pH of the eluate had decreased to pH I 1 LPT-d induced releasing activity seemed to be focused ' ! at pH 6.6 (Figure 17). The toxin was not so well resolved 'as had been hoped. There was a considerable amount of 'activity in the trailing edge of the peak over a pH range I of 6 to 5. Because recovery was generally poor (less than I 10%) and the specific activity was not appreciably I : increased, the use of chromatofocusing as a means of 1 26 'J r" Figure 17. Chromatofocusing of Leptinotarsin-d. A three ml aliquot (approximately 47 5 units) from the pooled, active Sephadex G-150 fractions was applied to a 17 ml column of the anion exchange resin PBE 9 4 (Pharmacia) which had previously been washed with imidazole-HCl buffer, pH 7.4. Elution of LPT-d with Polybuffer 74 (Pharmacia) adjusted to pH 4 resulted in the collection of fifty (2 ml) fractions. Eluate pH is indicated by the dashed line. 127 V» c 3 0) Vf (Q J) QC .2 .8 .4 0 5# 10 3 0 -I 9 8 m c 7 2. (D T3 Z -» 4 Fraction Number 1 28 i further purifying LPT-d was ruled out. Dve-Ligand Af f init v Chromatography Because the final step in the purification of LPT-h involved a specific binding of that toxin to agarose columns containing the dye Reactive Blue 2 (Crosland et a_l. , 1 984), it seemed prudent to investigate the : possibility that dye-ligand chromatography could be used in the purification of LPT-d. Screening of several dye-based I columns (Amicon) using LPT-d purified through the G-150 i step revealed that at least two had a high degree of specificity for the toxin. Matrex Gel Blue A effected a 13.5 fold purification of G-150 level LPT-d and had the I (best yield of all the columns tested. Toxin recovered from 'the Matrex Gel Orange A column was at least four times more pure than that eluted from the Blue A column, but its yield was considerably lower. Further tests were conducted with both of these columns as potential steps in the purification scheme. I To evaluate more carefully the effectiveness of Matrex Gel Blue A in resolving release-promoting activity, a two ! ml column was prepared in 50 mM phosphate buffer (pH 7.0). ! . ! I A 10 ml sample of G-150 purified LPT-d was then applied to ] I the column. The column was washed with 30 ml (15 column | I ^ ' ivclumes) of buffer and exposed to a 110 ml linear salt ' I gradient (0-1.0 M NaCl). The results (Figure 18a) show that; 129 ! Figure 18. Chromatography of Leptinotarsin-d on Dye-Ligand Affinity Matrices, (a) Blue A. A 10 ml sample of the active Sephadex G-150 pool was applied to a two ml column of the dye matrix Blue A and eluted with a 110 ml salt gradient (- - -). Activity values reflect the extent of release (units) elicited by 50 ul aliquots of the eluted fractions (2 ml each), (b) Orange A. Chromatography was effected under conditions identical to those described for the Blue A column. 130 a. 3 Blue A 2 1 0 30 60 40 20 1.0 - . 3 .4 -* 0 Fraction Number <0 4> O C 3 O range A 2 1 0 60 40 30 20 1.0 .3 .6 O 2 0 Fraction Number 1 3 . ; at least two distinct pools of releasing activity are present in the G-150 pool and that these can be separated based on their different affinities for the dye Blue A. The smaller of the two peaks elutes just before the salt gradient is initiated. The larger peak is more tightly bound, the major portion of activity not eluting until the NaCl concentration reaches 0.5 to 0.6 M. The presence of at least two active fractions in the G-150 pool was also confirmed by elution of the G-150 level i toxin through columns prepared with the Orange A dye matrix ^ (Figure 18b). In this case neither of the active toxin fractions was very tightly bound. The first peak did not , appear to have any affinity for the Orange A matrix, I eluting just after sample application, and the second . eluted just after the salt gradient was initiated (at approximately 0.1 M NaCl). In analyses of the abilities of these two columns to further purify CM-Sephadex level LPT-d, it was found that both produced only a single peak of releasing activity with affinity for the columns. Active fractions from the Blue A columns were approximately two fold richer in toxic activity than were the CM-Sephadex fractions. The elution | pattern from the Orange A columns was similar to that ' i ! ^ described for G-150 samples except that the smaller first ' peak recovered when G-150 level toxin had been applied was ' no longer evident. Although the specific activities of I 132 ! ------------. _ - - - - -j I LPT-d pools eluted from the two columns were similar, i ; because the yield from Blue A columns was almost twice that : from the Orange A columns, CM-Sephadex purified LPT-d was routinely purified over the Blue A columns. I I ; High Performance Liquid Chromatography | As had been the case with LOT, HPLC was considered as i ‘ a potential purification step because of the speed with > which protein separations can be effected and because i excellent resolution can be attained even over relatively short time periods. Both molecular sizing (TSK C3 000 SW) and anion exchange (SynchroPak AX-300) columns were used. j ' Elution of G-150 purified LPT-d over a TSK G300SW ^ I column resulted in a separation of two major protein peaks I I :(based on a continuous monitoring of the absorbance at 280 | nm) . The first peak to elute was a very broad one with | ' pronounced tailing. Fractions with the greatest protein I concentration eluted at approximately the same volume as I did BSA (MW = 68 kD). Trailing fractions eluted at volumes ' : corresponding to that of /5-lactoglobulin (MW = 18 kD). A ! second protein peak, about one twentieth the size the of the major peak, eluted at approximately 1.5 column volumes.' I Because resolution of the proteins with molecular weights ■ i I ! near that of LPT-d was very poor, no further attempts at i I purification were made with this type of HPLC column. I Experiments aimed at isolating LPT-d by passing the i 133 toxin over an HPLC column of SynchroPak AX-300 were much i I more successful. Two hundred ul of G-150 level LPT-d were applied to a 4.1 mm x 25 cm column of the anion exchanger and eluted with a linear gradient of 0-0.5 M NaCl. Several discrete protein peaks were discernible (Figure 19, top). A, very large protein pool consisting of at least seven different proteins, eluted before establishment of the ] gradient. Another four proteins eluted as distinct peaks | I I over a range of 0.3 to 0.5 M NaCl. I In order to determine which of the four proteins ' i . , : : exhibiting acidic characteristics was LPT-d, toxin purified : through the DEAE step was applied to the HPLC column and .eluted under the same conditions. As can be seen in Figure! I ■ .19 (bottom), the protein eluting at approximately 0.38 M NaCl is the most pronounced of the four, and fractions j collected across the entire gradient show all releasing | I activity to be contained within that one peak. i I SDS-Po1vacrVlamid e Ge1 Electrophoresis ! I SDS-PAGE was utilized in a routine analysis of toxin | i I purity at the various steps in the purification scheme. Samples were concentrated as necessary and then electrophoresed over a discontinuous slab gel. A series of I low molecular weight standards was always run in parallel I to give some indication not only of the effectiveness of I the separation but also of the molecular weights of the 134 I Figure 19. High Performance Liquid Chromatography of i I Leptinotarsin-d on SynchroPak AX-300. LPT-d at two I different steps in the purification scheme was I chromatographed over an HPLC column (4.1 mm x 25 cm) , of Synchropak AX-300. Top: Elution pattern characteristic of G-150 purified LPT-d (200 ul; j j approximately 50 ug). A linear (0 to 0.5 M) salt ^ gradient was initiated at the arrow (10 min. after I sample injection). Bottom: Elution profile of DEAE 1 purified LPT-d (600 ul; 14.5 ug). The arrow indicates the start of a linear salt gradient. Both chromatograms represent continuous recordings of the eluate absorbance at ^ = 280 nm. In both cases the absorbance range was 0.0 2; 1 mV full scale. 135 G -150 E c o Q O CM LPT-d <D U C «0 a o (O J Q < 10 20 30 Retention Tim e (min) DEAE E c o co CM « U C co LPT-d o M a < 10 20 30 Retention Time (min) 136 ! sample proteins. All gels were stained with the silver I stain technique developed originally by Merril et aI.(1981). As purification progressed, the number of proteins recognizable as discrete bands steadily decreased. At the Blue A level (in the sequence G-150 -> DEAE -> CM- Sephadex -> Blue A) LPT-d was consistently characterized by ' a single, well-defined band at approximately the molecular I I weight of ovalbumin (Figure 20). This was true even when I pooled active fractions ultrafiltered to 10 times the original concentration were subjected to electrophoresis I under conditions identical to those described above. To verify the purity of Blue A purified LPT-d, samples I ' of the concentrated active pool were subjected to 2-D j electrophoresis. Only one spot was apparent on the silver ! stained gel. Over a pH range of 4 to 7, the one protein , stained had an isoelectric point slightly less than 7.0. . This is in good agreement with the results of the ' chroma tofocusing experiment. The molecular weight was approximately 45 kD. j The sequence of steps utilized to purify LPT-d to apparent homogeneity was similar to that employed by [ Crosland (1984) to purify LPT-h. Passage of the hemolymph ; . Figure 20. Sodium DodecyIsulfate-Polyacrylamide Gel Electrophoresis of Leptinotarsin-d. An aliquot of the pooled, active fractions collected from the dye matrix gel Blue A was desalted and concentrated by ; ultrafiltration and then electrophoresed over a discontinuous slab gel. The separating gel contained 9% aerylamide; the stacking gel was 3% aerylamide. Electrophoresis (40 V, constant voltage) was allowed to : proceed for 16 hrs. The gel was stained with silver I ^ according to the procedure of Merril e_t al . (1981). I Molecular weight calibration is based on the ' simultaneous electrophoresis of protein standards ; purchased from Pharmacia (low molecular weight 1 electrophoresis calibration kit). 1 38 2 70 o> ® 50 iS 3 o 0) 3 0 L ^ LPT-d 139 DEAE-sephadex, CM-Sephadex, and Blue A resulted in an ■ overall purification of greater than 500 fold (Table 3). The purification scheme was about twice as effective as that used by Yoshino (1980) in purifying the toxin, but resulted in a considerably lower yield (4% in this case vs. 27%) . Chromâtofocusing and HPLC have not been included, I because neither technique contributed significantly to to I the purification of LPT-d without a substantial reduction I in yield. Likewise, variations in the order in which the different columns were used have not been reported, since no other sequence resulted in substantially better values either for purification or for the recovery of activity. Stability of LPT-d Activity assays of G-150 and DEAE purified LPT-d indicate that, at these steps in the purification scheme, the toxin is stable for at least 30 days when kept at 4°C in a darkened environment. The toxin can also be I frozen (-20°C) or lyophilized at these stages and later ; reconstituted to the original concentrations with virtually! I I no loss of activity. At purification steps beyond the level of anion I exchange, stability of the unmodified active fractions decreases. Pooled active CM-Sephadex fractions, kept in ; sodium phosphate buffer (pH 7) containing approximately 0.2 : 140j 1 o u O es m 0 ) > H • • • 0 ) m "d" o \ # — 4 sj- u *H f— 4 es t— < 60 M es m p Ph ss o M H < O PU 0 1 H P4 O >4 < s S > co w - i >, o -w • H > t— < ( U 4J • H O >4 < 6 co 4 - > • H a o O O CO " d " m o >, •H *J M-4 «H 60 3 c o m co es •rH > 4_) # — 4 sO o \ es es O «H •H • • • • • Ol 4J O, o co <3 a 3 O O es ro co o sO W m o 3 es es es •H • • • co O 6 O o \ # — 4 O O sj- es ( U O 60 es r-l M 3 # — 4 Xi P k es c o H 1 —4 B co O co -îl S O O co 1 —4 es es > 3 •3 O eu • 1 - 1 B 4J >> < • r 4 o O *3 o m M m < ü 3 B 1 —4 < 1 3 O < ü 1 M S « —1 O 33 O O o PQ 14 1 I M NaCl, lose activity at a rate of nearly 25% over a twenty ! day period (Figure 21). The half life of toxic activity at this stage of the purification is about 40 days, and it cannot be appreciably increased by exchange into different . buffers, or modification of the pH or salt content. Furthermore LPT-d eluted from CM 50 columns cannot be i frozen or lyophilized without a substantial loss of release I promoting activity. I j One modification of of the CM-Sephadex pool which did , I enhance the long term stability of toxic activity in these ' : I fractions was the addition of BSÂ. When added to a final I concentration of 0.01 or 0.1%, BSA substantially reduced i the rate of loss of activity. In the presence of 0.1% BSA, | i I I toxic activity could be maintained at very nearly the ! . . . ! initial levels for at least twenty days. Since BSA itself ; ; I did not have any effect on synaptosomal release of j I i radioactivity, it was routinely used to stabilize LPT-d not being further purified. i i At the Blue A level, LPT-d is very unstable, losing I all activity within seven days when kept in phosphate I buffer at 4°C. Stability could be enhanced at this step by the addition of BSA as described above, but best results were achieved when the pure toxin was made more j concentrated. Ultrafiltration of the Blue A pool, I j sufficient to effect a 15 to 20 fold increase in the I concentration of LPT-d, resulted in only moderate losses of i 142 i Figure 21. Stability of Leptinotarsin-d Purified Through | the CM-Sephadex Step. Aliquots (50—100 ul) of LPT-d | I collected from the CM-Sephadex columns were assayed for j release promoting activity at different time intervals ■ for up to 77 days. Activity has been normalized by ' I expressing it as units/ml. r = -0.996. | 143 50 r E « 30 c 3 > O < 10 0 6 0 4 0 8 0 20 Time (days) 144 activity measured at intervals over 20 days. ^ Time Course o f Re lea s e The time course of LPT-d-s timulated release of I ' radioactivity was evaluated using LPT-d purified through the CM-Sephadex level. Synaptosomes previously loaded with ' 3 . . [ Hjcholine were incubated for various times at ' 37^C with a concentration of LPT-d (.21 ug/ml) known to j induce release at approximately 7 5% of the maximum value. j The time of incubation was varied from 1 to 16 minutes, andî reactions were considered stopped by immersion of the j , samples in an ice-water slush. This was followed within a j ; few minutes by microcentrifugation ( 9000 x g ; 5 min; | I4°C). I : 1 As can be seen in Figure 22, the release induced by [ ■ ^ t ; LPT-d was rapid, reaching a plateau within 8-10 minutes. | This data is consistent with the findings of Connor (pers. ! I 2+1 comm.) in studies of the rate of LPT-d-stimulated Ca I influx into cultured rat diencephalon cells. LPT-d at this | I I I level of purity, and at approximately 150 ng/ml, promotes a. I very rapid uptake of calcium into the isolated,cel Is. The | I 2 + I influx of Ca becomes maximal approximately eight I minutes after the cells are first exposed to a solution | I : containing the toxin. To standardize all subsequent experiments j ; investigating the effects of LPT-d on the release of I 145 : Figure 22. Time Course of the Release of Radioactivity Stimulated by Leptinotarsin-d. Loaded synaptosomes were added to solutions containing LPT-d (200 ng/ml) and incubated for various times at 3 7 ° C. Incubations were stopped by immersion of the incubates in an ice-water slush. Each point represents the mean of three simultaneous assays. 146 » en C D » * 5 te 3 2 1 0 12 16 8 4 Time (min) L . from synaptosomes, the incubation time was set at 10 minutes. Assays of release therefore, were made following the attainment of equilibrium conditions. Kinetic ; analyses which should be based on initial velocities, will be performed using the data collected at equilibrium. Re 1ease of Radioactivity at Varying Concentrations of LPT-d The effect of varying the concentration of toxin on I I I the release of radioactivity from rat forebrain I : synaptosomes is shown in Figure 23. As had been characterized by Yoshino (1980), the LPT-d do se-rel ease I curve is sigmoidal reflecting some coopérâtivity between ‘ i toxin molecules in effecting release. When the data are redrawn in Hill coordinates (Figure 23, Inset) such that a best fit to the least squares regression is achieved, a straight line is obtained (r = I 0.978). The slope of the line, which is an estimate of the extent of cooperativity, is -1.7 in fairly good agreement with that computed by Yoshino (-2.0; 1980). The data suggest that two molecules of LPT-d are required to I initiate a release event. Similar results were obtained j j I with a less pure preparation of the toxin suggesting that I I contaminants, which may be present (at least at the DEAE I level), do not interfere with LPT-d-s timulated release j act iv ity. I 148 Figure 23. ^Effect of Varying the Concentration of Leptinotarsin-d on the Release of Radioactivity. Release stimulated by LPT-d is plotted as a function of toxin concentration. The standard (10 min) incubation time was used. Data points are the means of three assays. Inset: Hill plot of the same data. The maximal amount of release, oc, was calculated to be approximately 2.7 units. "i" represents the number of units released at a given concentration of LPT-d. The slope of the line is 1.7. r = -0.978. L _____ r « (O (0 0> 4> O C 3 2 1 Lofl Cancantratlon 0 50 10O 150 200 250 [LPT-d] (ng/mi; . 15 0. J ; Ca1c iurn Dependence o f Re 1e a s e [ Unlike that stimulated by lOT, the release evoked from ! synaptosomes by LPT-d is dependent on the presence of calcium in the bathing medium. When the concentration of CaCl^ is varied in synaptosomal bathing solutions, radioactivity released by LPT-d (200 ng/ml) is seen to I increase with increasing ca1cium concentration up to at I least 4 mM (Figure 24). The fact that some release is I evident even when calcium in the incubating medium has been 2 + completely replaced with Mg , suggests that micromolar quantities of calcium are still present in the extrasynaptosomal environment. Since the intracellular i “7 I calcium concentration is less than 10 M (Campbell, 1983), there could still be a substantial inward directed | calcium gradient under these conditions, and this might be ■ sufficient to initiate the release of neurotransmitter. ! In order to maintain the extracellular calcium ion , concentration near 0 mM, CaCl^ was omitted from, and 1 j ImM ethyleneglyCO1-bis-(B-aminoethy lether)-N,N '-tetra- I I ,acetic acid (EGTA) was added to all solutions with which I the synaptosomes were bathed after loading. As is evident | in Figure 25a, under these conditions, release stimulated | ‘ I \ by LPT-d (50 ng/ml) is completely blocked in the absence of ; I 2 + extracellular Ca . As the LPT-d concentration is increased however, some Figure 24. Calcium Dependence of Leptinotarsin-d-Induced Release. The effect of varying the extracellular concentration of calcium ions on release stimulated by LPT-d (200 ng/ml) is plotted. Loaded synaptosomes were incubated in a modified PS which was buffered with HEPES and contained different concentrations of CaCl Isosmotic conditions were maintained by varying the concentrations of MgCl^ and NaCl. Data points represent the average + standard deviation of quadruplicate assays. 2- I 152 0> ë ® o fie 2 1 0 1.0 2.0 3.0 4.0 153 Figure 25. Effect of Calcium Chelation with EGTA on Synaptosomal Release of Radioactivity Stimulated by Leptinotarsin-d and by Depolarization with . Loaded synaptosomes were washed either in a modified PS ; 2+ : containing 1 mM EGTA and no Ca (light bars) or in j I normal PS (dark bars). The washed synaptosomes were then j exposed to varying concentrations of LPT-d (a) or | I (b) and incubated for the standard 10 min period. Values i for release represent the means of triplicate assays , under each condition. 154 J a. 4 r C O 0> C O CD 4> 0) fiC 2 - 50 100 200 [LPT-d] (ng/ml) b. [K+](mM) 155 I EGTA-substituted medium, and the extent of release under I , calcium free conditions appears to be related to the toxin concentration. At 200 ng/ml, LPT-d can effect release in 2 + the absence of [Ca ]^ equivalent to almost one fourth (22.8%) of the release under normal calcium conditions. ^ I Results similar to these have been reported by Yoshinoj t ( 1980) and by Madeddu et a 1. (in press) for the related ; { toxin LPT-h. It may be that the absence of calcium renders ' the synaptosomal membrane less stable (see Szerb and ! I I 0'Regan, 1985), and, consequently, more susceptible to I toxin damage, particularly at relatively high toxin ! , concentrations. I Evidence that the complete removal of extracellular ; i calcium can have deleterious effects on synaptosomal membrane integrity is presented in Figure 25b. Release induced by depolarization of synaptosomes with solutions of; I increasing concentration, is not completely blocked by the chelation of calcium with EGTA, but instead remains ^ consistently almost one half that evoked when 1 mM calcium is present. ' I 1 Sodium Dependence of Release 1 ' j One mechanism by which LPT-d could effect a calcium I I dependent release of radioactivity from synaptosomes is by | , increasing Na^ influx. Depolarization resulting from an i _____________________________ ____ _________________ 1 56j • • + i , increased intracellular Na ion concentration would cause synaptosomal calcium channels to open and the ensuing influx of calcium would stimulate vesicular release. To test the possibility that LPT-d specifically opens Na channels in the synaptosomal membrane, the effect ; I of te trodo toxin (TTX) on LPT-d-s timulated release was ! I examined. Loaded synaptosomes were divided into two equal aliquots. One pool was washed in normal PS. The other pool i was washed in a saline solution containing TTX at a final I , I I concentration of 1.0 uM. Aliquots of the two pools were I ! then added, as appropriate, to solutions containing LPT-d ' I (210 ng/ml) with or without TTX, and incubated under the j I standard conditions. There was no significant difference ini I LPT-d-s timula ted release in the presence or absence of the i sodium channel blocking toxin (Figure 26a). These results | , are consistent with those described by Yoshino (1980) and by Crosland et a 1. (1984) and Madeddu et a 1. (in press) for : . I I the closely related toxin LPT-h. I ' Another means by which LPT-d could enhance Na^ ' + I influx without specifically opening Na channels is by j acting as a sodium ionophore. To investigate this j possibility, sodium ions were replaced in the bathing j medium by an equimolar substitution of choline ions. ■ Release evoked by LPT-d was largely unaffected by choline ; I substitution (Figure 26b). Similarly conducted experiments, using TRIS (Yoshino, 1980) and glucosammonium ions 137! j Figure 26. Effects of TTX and the Removal of Na^ ; Ions on Release Stimulated by Leptinotarsin-d. (a) I Loaded synaptosomes were exposed to 200 ng/ml LPT-d in the presence (light bar) or absence (dark bar) of 1 uM TTX. Release of radioactivity is expressed as a , percentage of the value obtained following synaptosomal lysis with distilled water, (b) Loaded synaptosomes were i exposed to 200 ng/ml LPT-d, and release of radioactivity induced by the toxin was evaluated in the presence (dark bar) or absence (light bar) of Na^ ions. The ; Na^-deficient solution was made by substituting I choline chloride for NaCl on an equimolar basis. Release I was determined following a four sec incubation period. 158 b. 1 (Q <0 > O 9 9 (0 OC 100 80 60 40 20 q> co « * OC LPT-d +TTX LPT-d -Na+ 159 (Crosland ejL al.. 1984) as substitutes for Na^, also ! showed no reduction in toxin stimulated release due to the i removal of extrasynaptosomal Na • The Effects of Multivalent Metal Cations on LPT-d-Induced Re lease I 2 + If LPT-d effects Ca -dependent release by opening synaptosomal calcium channels, release of radioactivity I should be reduced or blocked altogether in the presence of | i I !multivalent metal cations with ionic radii greater than \ ' 2+ : that of Ca . Nachshen (1984) and Drapeau and Nachshen | (19 84) have shown that several mult ivalent cations are | ^ able to compete with calcium for access to binding sites j I within the calcium channels of rat forebrain synaptosomes. j I . I ■As a consequence of the channel blocking activity of these | metal ions, calcium influx, and hence vesicular release, ' I can be markedly reduced. 2 + 1 It has already been reported that Mg , at a concentration of 18 mM, can block both K^- and | I LPT—d —induced release to approximately the same extent j (Yoshino, 1980). To investigate the effects of other known calcium channel blockers on toxin-evoked release, loaded Isynaptosomes were prepared in a modified PS solution I(containing 20 mM HEPES adjusted to pH 7.0) and then , i I I exposed to LPT-d or high and one of four mult ivalent j ! cations. | L_. _ _ _ _ At a concentration of 10 La^^ completely ! ! blocks release evoked by depolarization of the synaptosomes with 55mM and reduces the release promoted by LPT-d ; (75 ng/ml) by 98% (Figure 27a). A 10 fold lower i 2 + ! concentration of Cd is less effective in decreasing ' release but still inhibits both high K^- and ! , toxin-induced release to approximately the same extent ; Figure 27b). Release under conditions of elevated is I depressed by 36% whereas LPT-d in the presence of Cd^^ I I effects 35% less release. ; ' Less potent as inhibitors of synaptosomal calcium ' ! influx are the divalent cations Co^^ and Mn^^ I ; (Nachshen, 1984). Results of studies on the abilities of I 1 these ions to block LPT-d-stimulated release are presented I I ! in Figures 27c and 27d. At four times the concentration of ; i C d t h a t was used, Co^^ is able to inhibit the i ' . i I release of radioactivity from synaptosomes by almost 7 5% | 2 + I , ([LPT-d] = 62 ng/ml). Mn , on the other hand, at 10 | I 2 + times the concentration of Cd , actually stimulates i j the release induced by exposure of the synaptosomes to both I high and LPT-d. I The reason for the apparent stimulatory effect of 2 + ! Mn in these experiments is not clear. The cation has | ; been shown, in some cases, to be capable of passing through, I presynaptic calcium channels and to be virtually equivalentj ; to Ca^ ^ in its ability to induce release (Drapeau and j i . Figure 27, Effects of Multivalent Metal Cations on Leptinotarsin—d Promoted Release of Radioactivity from Synaptos ome s. (a) Effect of 1 mM La^ on release induced by LPT-d and by 55 mM , (b) Release in the presence of 100 uM Cd^*. (c) Effect on LPT-d stimulted release of 4 mM Co^*. Note that there are no data for induced release under these 2 + conditions, (d) Stimulatory effect of 1 mM Mn on release induced by both 55 mM and 75 ng/ml LPT-d, Release of radioactivity in all cases is plotted as percent of control values (release in the absence of the divalent metal cations). 1-6 2 100 r 100 r a. 50 C. 50 LPT-d LPT-d 100 100 r 50 m d. 50 h LPT-d LPT-d 1 63 ! I Nachshen, 1984). It is possible then that, under the I , ^ 2 + I conditions of this experiment, Mn ions which had entered the synaptosomes were able to effect release by 2+ . either substituting for Ca in the exocytotic process or by inducing release of Ca^* from intraterminal | stores (Drapeau and Nachshen, 1984). The fact that both depolarization- and LPT-d-induced I I I release are consistently depressed or stimulated to very i nearly the same extent, upon exposure of the synaptosomes ; ! to the various metal cations, suggests that LPT-d effects j ' release by the same mechanism as does depolarization. Both , release events are due, in part, to the opening of 2 + , presynaptic Ca channels. [ The Effect o f Veranami1 on LPT-d-Stimu1a ted Re lease j In addition to the inorganic metal cations, several | ; organic molecules have been shown to be potent calcium ! ; channel antagonists especially in heart (Sanguinetti and ( Kass, 1984) and smooth muscle (Murphy et a 1 . . 1 983). I j I Although the effects of these drugs are poorly manifested in nervous tissue (see Miller and Freedman, 1984 for a I I I I review), there is good evidence that some of them bind with, ! ' I high affinity to synaptic regions in rat brain (Gould et I a 1 . . 1 982 ; Greenberg e t a 1 .. 1984; 1985), and thus these j agents might be expected to interfere with release evoked by the opening of synaptosomal calcium channels. At least j : ! i 1641 three classes of organic calcium channel antagonists have been characterized: the phenylaIky lamines including verapamil and diltiazem; the dihydropyridines; and the diphenyla Iky lamines (Sped ding, 1 985 ). Yoshino (1980) has reported that verapamil (at a final concentration of 10 uM) is able to depress by 25% the release stimulated by LPT-d (23 0 ng/ml). Similar studies conducted with the related toxin LPT-h, however suggest that verapamil may actually potentiate the toxin effect (Madeddu e_t a 1. . in press). To resolve the question as to I what effect, if any, verapamil might have on synaptosomal calcium channels and/or LPT-d-stimulated release, experiments were undertaken to evaluate release, evoked ! ^ ^ either by 55 mM K or the toxin following an exposure 'of the synaptosomes to verapamil. Verapamil hydrochloride (Sigma) was dissolved in PS to -3 produce a stock solution of 2.5 x 10 M. Freshly prepared solutions were used in all experiments. Synaptosomes which had been previously loaded with 3 [ H ]cho1ine and washed were then added to plastic ; I minivials containing appropriately diluted solutions of j i i I verapamil, and incubated in a darkened room for 30 to 45 I 1 minutes. Following the incubation period, synaptosomes were Iwa shed once by centrifugation and re suspension (in iverapamil-containing PS), and then exposed to solutions containing the antagonist and either elevated or ______ 16 5J LPT-d. The results are presented in Figure 28. In accord with the findings of Madeddu e t a 1. (in press), verapamil (100 uM) does not inhibit LPT-d-induced release (Fig. 28a). In fact, there seems to be a slight potentiation of the toxin effect when the LPT-d concentration is 100 ng/ml, and this synergistic effect on release is even more pronounced at lower toxin concentrations (data not shown). 1 Data from another experiment, depicted in Fig. 2 8b, indicate that verapamil may not have a direct effect on the i calcium channels of rat brain synaptosomes. Synaptosomes depolarized in a solution containing 50 uM verapamil are hot characterized by significantly lower values of release than are characteristic of controls. Data collected at lower concentrations suggest that verapamil may even enhance depolariz at io n-induced release under some c i r c um stances. Effects o f Dihydropyridines on LP T—d-S t imula ted j Re lease I i ! Because dihydropyridines bind to sites on heart and smooth muscle calcium channels that are distinct from those occupied by verapamil (Sped ding, 1 985), an examination was made of the effects of these drugs on toxin-induced release. Two dihydropyridines were tested. These were nitrendipine, because it has been shown to bind with high ! . 166 ; Figure 28. Effects of Verapamil on Release Stimulated by j Leptinotarsin-d and by Depolarization with K^.- (a) (Release stimulated by LPT-d (100 ng/ml) in the presence i and absence of 100 uM verapamil. Values represent means ; + standard deviations for triplicate assays, (b) Release I stimulated by 69 mM in the presence and absence t ! of 50 uM verapamil. Values represent means ± standard ■ deviations for duplicate determinations of release promoting activity. 1 67 a. b. <0 i 2 4> (O co d ) ® 1 OC a LPT-d +Vp + vp 168 . J affinity to rat brain membranes (Gould ejt aJL. , 1 982 ; i I Greenberg ^_t a_l . , 1 985) and nifedipine, because it is the I parent compound of BAY K 86 44, a calcium channel agonist with properties apparently very similar to those of LPT-d (Schramm e t a 1 . . 1 983). ‘ Since dihydropyridines are insoluble in aqueous media,! both nifedipine and nitrendipine were dissolved in 95% ! -2 . . . I I ethanol to form 10 M stock solutions. To eliminate I ‘ I the possibility of loss of activity, the stock solutions ; I I : )were freshly prepared no more than one to one and a half i hours before a planned experiment. Synaptosomes were loaded: 3 . ' with [ H]choline as usual, washed once in normal j saline, and then twice in dihydropyridine containing PS 1 : I solutions (ethanol concentration less than 0.5%). Following- the third wash, synaptosomes were incubated (with shaking) in the dihydropyridine-PS solutions for 30 to 45 minutes either at room temperature or at 3 7 ° C. After a final ! wash, aliquots of the pre-exposed synaptosomes were added , .to solutions containing one of the dihydropyridines and 1 ! jeither LPT-d or high K . The solutions were then ' incubated and assayed for release activity according to the| I [Standard protocol except that all procedures were carried I I out in a darkened environment. I Because preliminary experiments had shown that | I . I ,nifedipine has no effect at all on LPT-d-induced release ' except possibly at concentrations in excess of 10 ^ M, 1 169 ! the effects of the dihydropyridine at concentrations of 40—50 uM were evaluated for several different toxin concentrations. In three separate experiments, differing principally in synaptosomal pre-exposure time and temperature, there was consistently no effect on the j release of radioactivity elicited by LPT-d. Pre-exposure ofj synaptosomes to nifedipine for 30 minutes at room ' temperature produced the results shown in Figure 29. I j Variations of the pre-exposure conditions from 10 min at 1 37°C to 60 minutes at room temperature made essentially no difference. The results when synaptosomes were pre-exposed to ,nitrendipine were very similar to those described for ! nifedipine. Changes in the nitrendipine concentration from 10-100 uM had very little effect on release induced by LPT-d and did not alter release evoked by depolarization I(with 50 mM ). Turner and Goldin (1985) have shown that dihydropyridines can have an effect on synaptosomal fast ,calcium channels, but they suggest that this effect may be I 2 + I masked by Ca flux through slow channels. We were unsuccessful in several attempts to reproduce their results. Even on a short time scale (1 to 8 sec) and in a 1 zero Na^ medium (to eliminate complications due to I ‘ + 2 + Na -Ca exchange), nitrendipine had no effect on either high K^- or LPT-d-induced release. The results I 17 0 Figure 29. Effects of Nifedipine on Release Stimulated ! by Leptinotarsin-d. Release of radioactivity is plotted i as a function of toxin concentration. The dashed curve (with solid squares representing individual data points) ' represents LPT-d-stimulated release in the presence of I 40 uM nifedipine; the solid curve (with solid circles) I I is indicative of release stimulated by the toxin in the ! absence of nifedipine. < 0 c 3 (D tn CQ 0> 3 2 1 2 5 0 150 O 200 5 0 100 [LPT-d] (n g /m l) 1 72 f ^ 1 of an experiment evaluating the effect of 10 uM i I I nitrendipine on the release stimulated by LPT-d (150 ng/ml) over a short time course are presented in Figure 30. Effect of Maito toxin on LPT-d-S t imula t ed Re lease I I Because the organic antagonists could not be shown to I block synaptosomal calcium influx due to either ' ! depolarization or LPT-d, experiments were conducted to investigate the possibility that maitotoxin (MTX) might • I affect LPT-d-stimulated release. MTX has been shown to | I 2 + I facilitate Ca flux into, and hormone release from, } rat anterior pituitary cells (Schettini e_t aj^., 1984) and i , isolated GE^ pituitary tumor cells (Login et a 1.. 1985). MTX has also been demonstrated to have similar j agonistic effects on the calcium channels of PCI2h cells ,(Takahashi et al.. 1982) and cultured neurob1astoma-g1ioma , i : cells (Freedman et a 1.. 19 84), but no prior studies of the 2+ ' toxin’s effect on Ca -dependent release from * Isynaptosomes have been done. I Two different experiments were performed. One study 2 + evaluated the Ca -dependence of MTX promoted release i 'of radioactivity in order to determine whether or not ■ 2 + I Ca channels were being affected. A second experiment, | ^run in conjunction with the first, investigated the effects| I of MTX specifically upon LPT-d-induced release. I I I ' I \ I Loaded and washed synaptosomes were exposed to ! I ! Figure 3 0. Effect of Nitrendipine on Release Stimulated I ! by Leptinotarsin-d. Synaptosomal release of i I radioactivity (cpm above background) is graphed as a function of the time of exposure to LPT-d. The PS : bathing the synaptosomes contained no Na^ to + 2 + eliminate complications attributable to Na —Ca exchange (Turner and Goldin, 1985). Release in the presence of nitrendipine (100 uM) is indicated by the ; dashed line (open squares); release in the absence of nitrendipine is shown as a solid line (closed squares). 1 The LPT-d concentration was 150 ng/ml. Data points are I the means of quadruplicate assays. i I 1 74 ( 0 « oc 1.5 1.0 o— CO o X E a o 0.5 Time (sec) 1 75 f solutions of MTX diluted in either normal PS or 2 + . . Ca -deficient PS such that the final concentration of toxin was varied from 10-167 ng/ml To one of the MTX solutions made up in normal PS, LPT-d (100 ng/ml) was added. After a 15 minute incubation at 3 7 ^ C, the suspended synaptosomes were sedimented by microcentrifugation, and aliquots of the supernatant were assayed for release activity. ' MTX alone did not promote the release of radioactivity' ! from synaptosomes at any toxin concentration. Both in the j ! presence and absence of calcium, release measured when MTX was present did not differ from control (background) ; values. I ; Similarly, LPT-d-stimulated release remained j ’unaffected by the presence of MTX in the incubation medium.' i I Release induced by LPT-d in the absence of MTX was 0.92 + ; * 0.18 units. Release when the MTX concentration was 83 ng/mlI was 1.04 + 0.25 units. There did not appear to be any significant interactions between the two toxins. I Effects of LPT—d on the Calc ium Channe1s of Other : Tissues 1. Paramecium caudaturn I I Calcium channels of the ciliated protozoan Parame c ium ! have been well characterized electrophysiologica1ly (Saimi I and Rung, 1982 ; Ehrlich et a 1.. 1984). Furthermore, channel I I _ 176 ! " activity can be readily assayed behaviorally in these animals, since opening of the channels with a concomitant rise in the intracellular calcium concentration is ; reflected in the animal's swimming speed and direction (Campbell, 1983 ; Eille, 1984), Paramec ium typically swims 1 in an anterior direction as long as the intracellular calcium ion concentration is maintained at or below 10 ^M. On contact with a solid object, a mechanically I induced wave of depolarization opens ciliary calcium I I channels, and when the internal calcium ion concentration I has increased approximately 10 fold to 1 uM, forward motion slows to a stop. Any further increase in intracellular calcium levels results in ciliary reversal, and the Paramecium begins swimming backward. I Experiments were undertaken to see if LPT-d opens 'Parame c ium calcium channels. The assay protocol was based on that of Saimi et a 1 . ( 1983 ). Briefly, 10 to 15 P . caudatum (Ward Scientific) were transferred in a disposable I fine tip glass pipette to a small culture dish containing 4 ' + 2 + ImM K , 1 mM Ca , and 0.2 M NaCl in a HEPES-NaOH I buffer (pH 7.2). The transferred animals were contained within as small a drop as possible, and the drop was 2 usually overlaid with a small 15 mm glass cover slip to facilitate observation. Swimming behavior was observed for 5-10 minutes under a high power dissecting micros cope, and then LPT-d was added dropwise at one edge of the 177 I covers lip. Locomotion was continuously monitored for up to 30 minutes after exposure to toxin. LPT-d does not seem to affect the intracellular I calcium concentration of P . caudatum in any way. At toxin ' concentrations as high as 238 ng/ml, there were no changes 1 in the observed swimming behavior of the protozoans. In | 1 addition to demonstrating that LPT-d has no effect on I Paramecium calcium channels, the experiments (repeated twice) show that LPT-d is not itself a lytic agent. | Disruption of the outer membrane of Paramec ium would have I resulted, at the very least, in altered swimming behavior. ! 2. S Quid Op tic Lobe Synaptosomes Although the calcium channels of squid optic lobes I have not been as well characterized as those of Paramecium. i synaptosomes prepared from the cephalopod were examined to | i determine if Ca^^-dependent release could be elicited from them by LPT-d. Squid optic lobe synaptosomes are ! t primarily cholinergic (Dowdall, 1974; Haghighat je^ a 1.. I 1 I 1984). Furthermore, Pollard and Pappas ( 1 97 9) have j I I ! demonstrated that ATP can be released from optic lobe j i I ! synaptosomes in a calcium-dependent manner, and, more I ‘ I recently. Pant et a 1 . ( 1983) have shown that the degree of j ! 4-amino-pyridine stimulated protein I ! phosphorylation/dephosphorylation in these synaptosomes is ! a calcium-dependent phenomenon. i Synaptosomes prepared from the optic lobes of squid I ; were exposed to squid saline solutions containing either I LPT-d (250 ng/ml, final concentration) or 100 mM . ,Incubation and the assay of release activity were done according to the standard procedures. I As is evident in Figure 31, LPT-d exposure did not significantly increase the amount of release of radioactivity relative to controls (squid saline alone). That viable calcium channels were present however is I indicated by the fact that depolarization with elevated ' effected an almost two fold increase in release. Furthermore, the failure of LPT-d to induce release was not ! due to poor uptake of [ ] cho1ine, since lysate i I activity values were very similar to those for synaptosomes: prepared from rat forebrain. | I 3. Smoo th Mus cle j I I The dihydropyridine (DHP) calcium channel antagonists ‘ ibind specifically and with high affinity to receptors in I smooth muscle, and the correlation between binding and | pharmacological response is very nearly 1:1 (Gengo e_t a_l. , ' 1982; Miller and Freedman, 1984). In general the properties! 'of smooth muscle calcium channels appear to be almost i j identical to those of cardiac muscle calcium channels for I I which DHP activity was first characterized (Fleckenstein, j I 1 977 ) . Since DHP antagonists did not seem to have any J : 179] I Figure 31. The Effect of Leptinotarsin-d on Release of Radioactivity from Synaptosomes’ Derived from Squid Optic Lobes. Synaptosomes prepared from squid optic lobes were 3 . were loaded with [ H]choline as described in the text, and then treated in a number of different ways. Controls for the release assay were based on exposure of the synaptosomes to squid saline (SS). The synaptosomes were also treated with LPT-d (250 ng/ml), (100 mM), and distilled water. Release is expressed both in terms of total recovered counts and as units of activity. Values represent means + standard deviations of quadruplicate assays. 18 n r' CO 25 20 15 T O 4) ^ S ^ ® E ® a 10 t 3 <0 co CD ® C « 3 cc — SS LPT-d K+ Lysate 1 8 1 effect on synaptosomal calcium channels, experiments were performed to evaluate the possibility that LPT-d acts on I smooth muscle calcium channels, perhaps at a site bearing some homology with synaptosomal channels. This work was done in collaboration with Dr. Donald M. Miller of the Southern Illinois University according to procedures outlined in Dickey et al. ( 1984) and in Miller et a 1. 0984) . At concentrations ranging from 46 to 1150 ng/ml, LPT-d had no obvious effects on the calcium channels of guinea Î pig ileum. If the toxin were an activator of smooth muscle i calcium channels, contraction of the muscles should have followed toxin exposure. This did not happen in two I different experiments designed to evaluate potential agonistic effects of the toxin on ileal sphincter, longitudinal, and circular muscles. I I 4. Ske1et a 1 Mus cle The richest concentrations of DHP binding sites known to date are found in the T-tubular system of skeletal muscle (Miller and Freedman, 1 984; Schwartz e t a 1 . . 1 985). This system also has a high concentration of verapamil binding sites (Galizzi e_t a_l. , 1984), and, in fact, there I may be twice as many verapamil receptors as there are DHP binding sites (Goll e_t aj^. , 1984). Although the correlation between antagonist binding and functional channel blockade 182 r ' is much lower in skeletal muscle than in smooth or cardiac muscle, the rich source and ready availability of ^ antagonist receptors in skeletal muscle has resulted in , extensive e1ectrophysio1ogica1 and biochemical characterization of the calcium channels in this tissue. Because the discrepancy between DHP binding activity and physiological response is like that reported for neuronal calcium channels, experiments were conducted to determine iwhat effects LPT-d might have on skeletal muscle calcium I channels. The work was done by Dr. Mark Nelson at the I 'University of Miami, using calcium channels that had been incorporated into planar lipid bilayers. The procedure has been described (Nelson et a 1.. 1984). At concentrations ranging over a ten fold range from 100-1000 ng/ml, LPT-d did not appear to affect calcium 'channel conductance measured as current fluctuation when 'the bilayer was held at +50 to +100 mV (standard convention). It was not clear in these preliminry experiments however that LPT-d was completely without I jeffect. No analyses were made, for example, of possible :LPT-d-induced alterations in mean channel open or closed ‘times or in the number and type of functional calcium channels affected. Since the DHP effect on skeletal muscle is a complicated one, not involving a single site blocking mechanism (Nelson, pers. comm.), it may well be that subtle !LPT-d effects were simply not detected under the conditions I i 183 I of these experiments. 5. Neurona1 CIona 1 Cell Line s Experiments to evaluate the possible effects of LPT-d on calcium currents in cultured PC12 and neuroblastoma cell lines were performed. Calcium channel antagonists bind i with high affinity to these cells of neuronal origin, and there ia a reasonably good correlation between antagonist j binding and functional channel blockage (Rampe et a 1.. I 1984). Furthermore, LPT-h has been shown to effect I c a 1c i um-dependent dopamine release from phaeochromo cytoma . cells (Madeddu et a 1. . in press). Studies investigating the effects of LPT-d on release from neuroblastoma cell lines were performed by Dr. Sathapana Kongsamut at the University ' of Chicago following the procedures outlined by Freedman et ' a 1. (1984) and Kongsamut _e_t aJL. ( 1985). LPT-d did not seem to have any effect either on undifferentiated PC12 cells or on PCI 2 cells treated with nerve growth factor to induce differentiation. At I concentrations up to 400 ng/ml, the toxin did not stimulate I release of norepinephrine at a rate or in an amount I significantly greater than that observed for controls. Likewise, LPT-d was found to be ineffective as a calcium I J channel agonist in NG108-15 cells. j While we have no reason to doubt these results, it j should be mentioned that the experiments were performed I I 184 : only once and probably should be repeated. It is possible, I I although not likely, that the toxin used in these experiments had lost some activity in shipment. Assays of release activity were not performed immediately prior to the experiments. ; i 6 . Isolated Diencephalon Cells j I Since the only experimental system so far shown to to I ! be affected by LPT-d was made up of synaptosomes derived j 1 1 from rat forebrain, it seemed appropriate to consider 1 I possible effects of the toxin on intact cells of the j mammalian central nervous system (CNS). Such a study might eliminate the possibility that LPT-d-induced release was an! j artifactual event related to the peculiar properties of 1 , resealed synaptosomal membrane. Also, because changes in j intracellular calcium concentration can be measured | 2+ I directly with the Ca - sensitive dye FURA-2 (Connor, ' 1985 ; Grynkiewicz et al., 1985), an examination of LPT-d I S effects specifically on Ca^^ flux could be made. The i investigations were made by Dr. John Connor at the AT&T Bell Laboratories in Murray Hill, NJ. , LPT-d (50-100 ng/ml) stimulated a massive influx of j calcium into some, but not all of the embryonic rat I diencephalon cells that were examined. The effect was ! maximal after approximately eight minutes. With the use of Î antibodies to neuron-specific amylase, it was possible to ! l _ 8 j J show that while some of the unaffected cells were non-neuronal, others were clearly neuronal. This suggested that LPT-d acts either on a specific subpopulation of CNS neurons or on a specific subtype of calcium channel found only on certain neurons. These suspicions appear to have 1 been confirmed in subsequent experiments comparing the ' + I effects of depolarization (25 mM K ) and LPT-d on j 2 + Ca flux. Approximately 60% of the cells exposed to a medium containing elevated were found to have an ' 2 + i enhanced influx of Ca , whereas only about 40% of the ; I ; same cells responded similarly to LPT-d exposure. ' I I Furthermore, although there was some overlap between the i ' I ^ two sets of cells, neither was completely contained within ; 1 ! I the other. , ! If nifedipine (5-10 uM) was applied to cells ; I previously exposed to LPT-d, the intracellular calcium ! i : I concentration of the cells rapidly returned to normal (less| I —7 . . . . ■ than 10 M) levels within five min. after they were 1 j washed with the dihydropyridine. This implied that the i ' inward calcium currents facilitated by LPT-d were turned I I off by nifedipine. Furthermore, since all of the ! ■ . I LPT-d-affected cells were able to return to normal resting ! . Î 1 calcium levels, it seemed that LPT-d had neither , , indiscriminately damaged cell membranes nor interfered withj I normal calcium exchange mechanisms. | 1 ! 1_86J 7. Mou s e and Bovine Bra in S ynap t o s orne s 1 The effect of LPT-d on release of radioactivity from CNS synap t o s ome s of mammals other than the rat was examined. Because they are in the same order (Rodentia) as rats, mice were used in one study. Synaptosomes were also prepared from bovine brain because the much larger size of beef brains made possible the testing of LPT-d effect ivene s in different regions of the mammalian brain. Synaptosomes of both mouse and cow were prepared and loaded according to ! 1 the standard procedure used with rat brains. i LPT-d effected release of radioactivity from mouse whole brain synaptosomes in a manner almost identical to that when synaptosomes derived from rat forebrain were used. At a protein concentration of 200 ng/ml, the toxin 1 evoked release equivalent to 2.2 activity units from the mouse synaptosomes. This was very near the value of 2.4 units which is characteristic of the release induced by LPT^d, at the same concentration, from rat brain. I Furthermore, the LPT-d do se-release curve for mouse brain I synaptosome appeared to be sigmoidal, suggesting that, as I in rat brain, LPT-d receptors must be occupied by more than one toxin molecule in order to effect release. LPT-d-induced release from bovine caudate nucleus Isynaptosomes also exhibited sigmoidal kinetics, and was 1 : even more robust than was toxin evoked release from either I rat or mouse brain synaptosomes. When the concentration of I 187 ;LPT-d was 150 ng/ml, release of radioactivity equivalent to ,2.54 units was recorded from synaptosomes derived from the bovine caudate nucleus. LPT-d also stimulated release of radioactivity from synaptosomes derived from bovine frontal cortex, but the dose-release kinetics were not clearly resolved. At a I concentration of 150 ng/ml, the toxin induced release of only 1.45 units from the cortical synaptosomes. The [relatively poor release effects were probably due primarily 3 . to poor [ H ]choline uptake and to a significantly reduced synaptosomal viability, since the frontal cortex had been stored in ice cold 0.32 M sucrose for over four hours before synaptosomes were prepared. Kalman and Eajos ! I (1982) have reported that postmortem storage of brain ! I ■ tissue for intervals greater than two hours can result in a [ substantial reduction in the number of viable synaptosomes.' Binding of LPT-d to Synan to somal Membrane Receptors If LPT-d stimulates release of radioactivity from CNS nerve terminals by interacting directly with pr e synap 1 1 c calcium channels, the toxin might be a useful molecular I probe with which to isolate and characterize those channels. Experiments were therefore undertaken to determine whether or not LPT-d binds with reasonably high t I affinity to specific receptors within the synaptosomal I jmembrane. The experimental protocol has been described. [ 188 Preliminary experiments, designed to determine if the toxin would bind to crude synaptosomal membrane, were conducted. Native (underivatized) toxin (125 ng/ml), after exposure to synaptosomes (2 mg protein/ml) lost all release-promoting activity, whereas the same concentration I of toxin, not previously exposed to synaptosomes, stimulated release as expected. Since LPT-d exposed to ' I I membranes for up to 16 minutes retains full activity, it I I seems likely that the lack of activity in this case , after I I a 10 minute exposure, was due to binding and not merely to ; ' exposure to membranes. Further experiments were conducted ; to better define the binding kinetics using both labelled and unlabelled toxin. 12 5 [ I]LPT-d, which had been shown to be active, consistently bound to synaptosomes in a very nonspecific fashion (Figure 32). Whether free and bound derivatives were separated by rapid filtration over GF/B filters ^ pre-soaked in PS or in PS containing 1% BSA, there was no ' evidence of saturable binding even in the presence of a | I 10-20 fold excess of underivatized toxin (Figure 3 2a). ! Furthermore, the glass fiber filters themselves seemed to ; have a high affinity for [^^^I]LPT-d (Fig. 3 2b), even I after extensive pre— treatment with 1% BSA and excess native I i I toxin. In an effort to reduce the amount of nonspecific j binding seen with glass fiber filters, free and bound i ___ 189J ■ Figure 32. Nonspecific Binding of lodinated I ^ . r 12 5 ! Leptinotarsin-d. Binding of [ I]LPT-d (expressed i I as cpm bound) is plotted as a function of the : concentration (in ng/ml) of the derivative, (a) Binding to GF/B filters, r = 0.973. (b) Binding to GF/C filters , pr e-treat ed with PS containing 1% BSA. r = 0 .983 . (c) I I Binding to crude synaptosomal membrane, r = 0.999. (d) Binding to purified synaptosomes. r = 0 .987 . 1 90 a: 100 « b X g 50 CL O 0 25 50 C. 50 co O X E O. o 5 0 25 b. 50 co O X g 25 & 0 5 0 25 5.0 co O 0 25 50 1 9 1 [125ijLpT-_d were, in some cases, separated by J microcentrifugation. Even under these circumstances rl25 however, [ IjLPT-d was extensively bound to the sedimented synaptosomes. There was no evidence that a finite number of binding sites existed (Fig. 32c). Because' crude (P2) synaptosomes were used in this case, there was ai possibility that much of the nonspecific binding could be I attributed to indiscriminate adherence of the toxin to ; mitochondrial membrane or myelin fragments. Therefore ; ! I , additional experiments were conducted with purified t synaptosomes. The results, shown in Fig. 3 2d, indicate that' 12 5 [ I]LPT-d had not occupied all available binding sites at the concentration of synaptosomes used (0.3 mg/ml). ' That the binding of iodinated toxin to receptors in I the synaptosomal membrane is nonspecific was confirmed by transblot experiments (Figure 33). At least four , synaptosomal proteins transferred onto a nitrocellulose ■ 12 5 sheet could bind [ I . ^ , . ... LPT-d. Based on comparison with I the migration pattern of a set of protein standards, the I molecular weights of the four proteins were approximately I i245-250, 125, 100, and 50 kD. Conceivably, the largest of | I the four most intensely labelled proteins might represent a; calcium channel. Norman et a 1 . ( 1 983 ) have reported that nitrendipine receptors isolated from rat cortex have \ jmolecular weights of approximately 210 kD. Nonetheless, it j I 1.92J Figure 33. Binding of lodinated Leptinotarsin-d to Trans-b1otted Synaptosomal Membrane Proteins. The transb1otting technique described by Symington (1984) was used to transfer electrophoretically separated memrane proteins from an SDS-PAGE slab gel to a sheet of nitrocellulose. The transferred proteins on the nitrocellulose sheet were exposed to iodinated LPT-d for eight to twelve hours. After the exposure period, the sheets were washed carefully three times, and then subjected to autoradiography (see text). The molecular weight calibration scale was constructed based on the electrophoresis of a series of protein standards (Pharmacia). 193 Z9 2 o (D O c Ù) g ® ÔD - OPl ztz 1 94 ; is clear that the binding of iodinated toxin was not specific for calcium channels. Experiments with underivatized toxin further confirmed the lack of binding specificity. To show evidence of saturable binding, purified synaptosomes were diluted 100 fold to a protein concentration of 4 ug/ml. Exposure of ■ LPT-d at different concentrations, to the diluted I synaptosomes resulted in a loss of all activity at toxin j concentrations less than 0.2 ug/ml (Figure 34). Because not j ! all toxin was bound at concentrations greater than 0.2 ug/ml, binding sites were assumed to be fully occupied at that concentration, and B was estimated, under the max conditions of this assay, to be at least 5.5 nmoles/mg synaptosomal protein. The for binding, based on a calculation of the toxin concentration at one half saturation, is 100 ng/ml, in good agreement with the approximate for release (see Fig. 23). That LPT-d continues to bind to synaptosomal membrane ■ even after release has reached an equilibrium value, is I ; evidenced in Figure 35. At a toxin concentration of 200 ng/ml, the binding of LPT-d, reflected as a decrease in j ! release-promoting activity, steadily increases, approach ing j , ^ I ; a maximal value after 15 to 16 minutes. The same j j concentration of toxin not previously exposed to I I synaptosomal membrane effects near maximal release in about j : one half that time. ! I 195 Figure 34. Binding of Native Leptinotarsin-d to Purified Synaptosomes. Release of radioactivity by LPT-d previously exposed (- - -) to purified synaptosomes and by LPT-d not previously exposed (--- ) is plotted as a function of toxin concentration. Each data point represents the mean of three simultaneous assays. 1 9 6j 3 «> CR a JS o c 5 4 2 1 0 . 1 .3 .2 .4 .5 .5 ' [LPT-d] (ug/ml 197 Figure 35. The Time Course of Binding of Native Leptinotarsin-d to Purified Synaptosomes. Exposure of LPT-d to purified synaptosomes for varying lengths of time and subsequent treatment of loaded synaptosomes with the free LPT-d remaining after those times, results in the curve defined by the open squares. LPT-d not previously exposed to synaptosomes stimulates release, over time, as indicated by the solid squares. 1 98 2.5 2.0 1.5 v> c ■3 » m » % OC 0. 6 12 16 8 4 Time (min) 199 Effects of LPT-d on Phosphorylation o f Svnap to s oma1 Prote ins The possibility that LPT-d might indirectly open calcium channels by causing the phosphorylation of some specific pro te in(s) was examined. The procedure used was modified after that of Robinson and Dunkley (1983). Briefly, synaptosomes were incubated in the presence of 3 2 inorganic [ P ]orthophosphate for 45 minutes at 37^C and then exposed to either LPT-d, 55 mM , or normal PS for five seconds. The five second time interval was chosen because Robinson and Dunkley (1983; 1984) had found this time to be optimal for the detection of changes in the phosphorylation state of synaptosomal phosphoproteins. Following the exposure to toxin, synaptosomes were solubilized with an SDS-based "stop solution” and the proteins subjected to SDS-PAGE electrophoresis on either 9% or gradient (5-15%) gels. Autoradiography of the dried gels was performed as described in MATERIALS AND METHODS. After a five second exposure to LPT-d or elevated , there were no readily apparent differences in the number or concentration of phosphoproteins in the synaptosomes (Figure 36). It is possible that a phosphoprote in of molecular weight approximately equal to 30 kD was slightly more phosphorylated following exposure to LPT-d than was evident in the PS and high 200 Figure 36. The Effect of Leptinotarsin-d on Phosphorylation of Synaptosomal Proteins. Synaptosomes 3 2 were incubated with [ P ]orthophosphate for 45 min at 3 7 ^ C and then exposed to LPT-d (100 ng/ml), 55 mM , or PS. After the exposure period, synaptosomes were solubilized, and the proteins were subjected to electrophoresis (5-15% gradient slab gel) followed by autoradiography as described in the text. 20 1 PS LP T-d 94 68 O f o> ô ? 3 O 0> 43 30 20 202 Bolutions, but autoradiographs made after additional exposure times would be necessary to quantitate the effect. Since all phosphorylation reactions were stopped after five seconds, it was not possible to calculate changes in the rates of phosphorylation and/or dephosphorylation. 203 DISCUSSION: Purification and Characterization of IDT Previous work by Martin and McClure (1982) and by McClure et a1. (in press) has shown that IDT is able to stimulate release of radioactivity from rat forebrain 3 synaptosomes pre-loaded with either [ H]choline or the tritiated neurotransmit ters norepinephrine and gamma-amino-butyric acid (GABA). Furthermore, the release effected by lOT was found to be extremely rapid, the action being completed by the earliest time point measured, and lOT-induced release was not dependent on the presence of + 2 + either Na or Ca ions in the bathing medium. For these reasons it was postulated that the toxin might act primarily at a presynaptic site controlling release, a site distinct from the ion channels known to be involved in stimulus-seeretion coupling. lOT has now been purified to apparent homogeneity based on the criterion of a single band on silver-stained SDS-polyacrylamde slab gels. The purification was greatly facilated by beginning with exudate of the sponge rather than the whole tissue homogenates as had been used by Martin. Naturally-emitted exudate of the sponge was found to be at least twenty times more enriched in lOT than tissue homogena tes. Elution of the exudate through a column of Sephadex G-50 produced one peak of activity corresponding to a molecular weight of approximately 18 kD, 204 and this was further resolved into two active fractions by anion exchange chromatography. Pure lOT was obtained from those fractions which were more firmly bound to Cel lex D anion exhange columns and eluted only after displacement by 100 mM NaCl. Retention by the positively-charge matrix may indicate that lOT is acidic or at least that a negatively charged domain is present in the molecule at neutral pH. One of the most interesting aspects of the release elicited by lOT is that it can occur in the absence of 2 + . extracellular calcium. When Ca ions in the incubation medium are replaced, on an equimolar basis, with 2 + either Mg or the calcium chelator EC TA, the toxin-induced release of radioactivity from synaptosomes is unaf fe c t ed. 2 + Ca - independent release phenomena have been found in macrophages (Young e_t a_l,. , 1984), rat submandibular acini (McPherson and Dormer, 1984), PC12 cells (Mel do1e s i et a 1 . . 1 984; Pozzan e_t aj,. , 1984), brain slices ( Schof f elmeer and Mulder, 1983 ), and synaptosomes (Arias e_t a1.. 1984; Wheeler, 1984), and are usually attributed to a 2 + mobilization of the Ca sequestered in intraterminal stores (Kelly .âX*» 1 979 ; Adam-Viz i and Ligeti, 1984; Sil insky, 1 985). This may be the means by which LOT effects an apparently calcium-independent release, and it could be experimentally determined by measuring 2 + intrasynaptosomal Ca levels during and following ___________________________________ 205 synaptosomal exposure to the toxin. Studies evaluating 2 + changes in [Ca ]^ have been performed in synaptosomes using the fluorescent indicator quin2 (Ashley e t a 1. . 1984; Richards e_t aJL. , 1984). Similar types of experiments might prove valuable in further elucidating the process by which lOT effects release. It is also possible that lOT, like oV latrotoxin, can promote release in the absence of extracellular calcium 2 + without causing a redistribution of Ca from intracellular stores. Meldolesi's group has found that black widow spider toxin may be able to activate protein kinase C in PC 12 cells thereby stimulating dopamine release with no apparent change in the intracellular calcium concentration (Meldolesi e_t aJL. , 1984; Pozzan ejt a 1. . 1984). Although no specific inhibitors of protein kinase C are available yet (Pozzan et al.. 1984), it should be possible to see if known activators of the enzyme stimulate 2 + Ca -independent release in a manner analogous to that stimulated by lOT. The time course of IOT—promo ted release, in good agreement with Martin and McClure (1982), is extremely rapid, maximal release typically being attained within 30 seconds of the time synaptosomes were first exposed to toxin. Interestingly, as the lOT concentration is increased, the amount of radioactivity released is also increased even though the shape of the curve defining 206 release as a function of time remains essentially unchanged. This would be expected for a simple titration of releasing sites with toxin. The data suggest that there may be a very large number of synaptosomal receptors capable of interacting with lOT. If release of radioactivity were determined by the number of release sites occupied by lOT, any one concentration of toxin might appear to evoke a maximal release. Saturating values would ultimately depend on the number of lOT binding sites available. As the number of molecules that were bound increased, so would the apparent maximal release. Release of radioactivity determined at 10 minutes as a funtion of IDT concentration defines a rectangular hyperbola. Maximal release measured under these conditions appears to be very nearly equal to the amount of release obtained by hypotonically ly sing the synaptosomes, and therefore should represent the maximum possible release. K release is estimated to be approximately 19.2 nM (Fig. 6). Q ^ , the concentration of lOT stimulating half maximal Only about 50% of the radioactivity released by TOT is 3 3 [ E]ACh, the balance being [ H]choline. Although there is always some "background leakage” of 3 . [ H ]choline from synaptosomes loaded with the tritiated molecule (Ashley _e_t a_l. , 1984), the amount released by lOT is similar to that released by hypotonic lysis. This suggests that lOT is not effecting a preferential release 207 of neurotransmitter but is somehow causing the synaptosomes to become leaky. That the toxin is not itself causing synaptosomal lysis was demonstrated by the fact that neither LDH nor ChAT were releasable, in substantial quantities, by lOT. At concentrations sufficient to release 7 5% or more of the radioactivity incorporated into synaptosomes, lOT induced release of the enzymes at levels consistenly less than 10% above those for controls. McClure and Martin (1983) have similarly reported finding no appreciable effects of lOT on synaptosomal integrity (based on LDH retention). Because electrophysiological studies of the LOT effect on frog neuromuscular junctions showed that the toxin could depolarize muscle fibers, the possibility that lOT might be opening small lesions in the membrane was investigated (see Fus sie et a 1 . . 1981 and Zalman and Wisnieski, 1984). The fact that cytoplasmic ATP is releasable by lOT suggests that at least part of the toxin-induced radioactivity release is due to a permeabilization of the synaptosomal membr ane. That lOT is capable of creating small pores in synaptosonal membrane is also indicated by the toxin's ability to release 2-[^H]DG/2-[^H]DG-6-P from synaptosomes previously loaded with the tritiated metabolites. Release of the glucose derivatives could very nearly be described by the same kinetic parameters which 208 o o characterized the release of [ H]ACh/[ H ]choline. V for 2“ [^H]DG release is 3.75 units (vs. 4.08 max units for [^H]Ch/ACh); in this case, is 13.9nM (vs. 19.2nM). It is therefore conceivable, and even probable, that 10T effects release, in both cases, by the same mechanism. It is not clear why Martin and McClure (1982) saw no evidence of toxin-induced synaptosomal depolarization, since the opening of small pores in the synaptosmal membrane should have resulted in a change in membrane potential. The dye used in the studies of Martin and McClure was sensitive to changes in synaptosomal potential, because depolarization with was readily detectable (McClure, pers. comm.). Although a number of experimental factors may have contributed to the dye’s apparent insensitivity to potential changes induced by lOT, we feel that the current studies, involving direct measurement of membrane potential with intracellular electrodes, provide a more valid estimate of the action of lOT. There is certainly no question that the muscle fibers exposed to toxin were depolarized. Many marine organisms produce defensive secretions which have detergent-1 ike properties (Halstead, 1981). Perhaps most notable are starfish of the genus As ter ias. The echinoderms synthesize large amounts of saponin, a detergent which renders cholesterol-rich membranes 209 permeable to small molecules such as ATP (McGraw e t a 1 .. 1980). The toxin produced by I . birotulata does not seem to permeabiliz e membranes in the same way as do detergents lowever. Exposure of synaptosomes to the detergent digitonin which has been used successfully to permeabi1ize adrenal medullary chromaffin cells (Dunn and Eolz, 1983; ]Nilson and Kirshner, 1 983), results in release of the larger molecules LDE and ChAT. These enzymes are not releasable by lOT, suggesting that either the detergent is opening larger pores than the sponge toxin or that synaptosomal lysis has resulted from the digitonin exposure. In any event lOT does not cause radioactivity to be released from synaptosomes in the same way that digitonin does. The data are most consistent with the hypothesis that lOT effects release of primarily the extravesicular pool of neurotransmitter by causing synaptosomes to become leaky. The contention that release is principally from cytoplasmic stores is based on three lines of evidence. First, the radioactivity released by lOT is about half [ HjACh and 3 half [ E]choline. This ratio is, similar to that released by hypotonic lysis and is characterisitic of the cytoplasmic compartment. Second, even after exhaustion of vesicular ATP stores by repeated synaptosomal depolarization, lOT can effect a dose-dependent release of the nucleotide. Third, a 10 minute predepolarization of 210 synaptosomes with 32 mM does not affect lOT-induce d release of radioactivity, whereas a subsequent depolarization of the synaptosomes with fails to elicit release. It would be interesting to test the effect of a vesicular transport blocker such as AH5183 (Melega and Howard, 1984; Edwards e_t a 1. . 1 985 ; Toth and Suszkiw, 1 985) on the release of radioactivity from lOT-treated synaptosomes. If lOT is effecting release principally from extraves icular pools, AH 5183 should not affect the toxin— s t imula ted release. Vesicular release, on the other hand, should be drastically curtailed in synaptosomes treated with the transport inhibitor (see, for example, Toth and Suszkiw, 1985). The actual mechanism by which 10T permeabi1izes membranes may be similar to that described for the proteinaceous Staphylococcus aureus OC-toxin (Fus s1e et a 1. . 1981). The water soluble molecules are only a little larger than lOT (34 vs. 18 kD), and are believed to form amphiphilic oligomeric complexes within membrane lipid bilayers. The oligomers form "thick-walled cylinders” bounding aqueous channels, and permit the free passage of molecules smaller than myoglobin (Fus s1e e t a 1 .. 1 985). If IDT similarly forms oligomeric complexes within synaptosomal membranes, the very rapid time courses of apparent "binding” and release could be explained. 211 Furthermore release through transmembrane pores made up of lOT would require neither Ca^^ nor Na^, and would be principally, at least initially, from nonvesicular pooIs. Other mechanisms by which lOT might effect 2 + . Ca - independent release also exist. It is possible, as noted earlier, that the sponge toxin mobilizes 2 + intracellular stores of Ca . In this case, release would still be triggered by an increased intracellular 2 + . 2 + concentration of Ca ions, but extracellular Ca would not be required. At least two different processes exist by which Ca^^^-independent release can be effected. Both enhanced phosphatidylinositoi turnover (Campbell, 1983; Silinsky, 1985) and activation of 2 + Ca -phospholipid-dependent protein kinase (protein kinase C ) (Pozzan e t a_l. , 1984) have been found to be capable of initiating release in the absence of extracellular Ca^^. Neither of these possibilities has been investigated yet, as potential means by which lOT might effect release of neurotransmit ters from synap t o s ome s. The evidence that has been accumulated suggests that lOT acts as a permeabilizing agent in effecting synaptosomal release of radioactivity. The sponge toxin might be of some use in studies evaluating changes in nonvesicular neurotransmit ter stores under various 212 release-promoting conditions. Used in conjunction with specific vesicular transport blockers like AH5183, lOT may be a tool with which to determine the role and relative contribution of nonvesicular release in the secretion of neurotransmit ters. 213 DISCUSSION: Purification and Characterization of LPT-d Leptinotars in was initially described as a relatively large molecular weight toxin found in hemolymph of the Colorado potato beetle Lent ino tarsa decemlineata (Hsiao and Fraenkel, 1969). The toxin has since heen purified in excess of 200 fold by conventional chromatographic techniques and was found to contain at least one molecule capable of promoting the release of ACh from synaptosomes O I in a Ca -dependent manner (Yoshino, 1980). Designated thejS-form of lep t ino tar s in (j^-LPT-d ) because it has no fly-killing activity (a characteristic of the oC-form) , the molecule was believed to effect neurotransmitter release by specifically opening presynaptic calcium channels (Yoshino, 1980) . Hagiwara and Byerly (1981) have established criteria which must he satisfied in order to verify the existence of calcium currents in a given preparation. Since release of neurotransmit ters is a consequence of inward-directed calcium currents, it should be possible to use alterations 2 + in Ca -dependent release as an indirect measure of the effect of ^-LPT-d on presynaptic calcium channels. If the criteria of Hagiwara and Byerly (1981) are restated in terms of what would happen- to the release of neurotransmit ter under various conditions known to affect Ca^^ flux, it is evident that ^-LPT-d acts, at least 214 partially, as an agonist of synaptosomal calcium channels. As shown by Yoshino (1980), the toxin effects release of neurotransmitter: 1. in a calcium-dependent manner. 2. when external Na^ is replaced with a large monovaient cation incapable of passing through Na channels. 3. in the presence of the Na channel blocker tetrodotoxin (T TX). 4. in parallel with the uptake of 2 + 2 + 5. when Ba or Sr are substituted for Ca . 2 + 6. except when blocked by less than lOmM Co , La , Cd , Mn^'*’, or Ni . Most of these findings have been confirmed. Further characterization indicates that^-LPT-d acts most effectively, and possibly even exclusively, on calcium channels of the mammalian central nervous system. The toxin has now been purified to apparent homogeneity by an extension of the procedure described originally by Yoshino (1980). Substitution of an affinity dye-ligand column (Blue A) for the hydroxylapatite column used as a last step by Yoshino resulted in the electrophoretic isolation of a single band on silver-stained SDS-polyacrylamide slab gels. The molecular weight of the protein is very nearly 45 kD in excellent agreement with the value of 40 kD reported by Yoshino (1980), and the toxin is at least 2.5 times more pure. 215 Homogeneity has been verified by a second criterion: the presence of a single spot following two-dimensional gel electrophoresis. Although only acidic proteins were separated during the isoelectric focusing portion of the two dimensional electrophoretic separation, the presence of contaminating basic proteins is not considered likely, since the toxin had previously been subjected to cation exchange chromatography at pH 7.0. Proteins with more positive charge than ^ —LPT-d should have been more firmly bound to the CM-Sephadex columns than was the toxin, and thus effectively separated from it. The isoelectric point of pure ^-LPT-d is very near 6.6 as determined by 2-D gel electrophoresis. This value is identical to that found for less pure toxin subjected to chromatofocusing (Fig. 17). In many ways the pure toxin is similar to the less pure form described earlier by Yoshino. Both forms of -LPT-d are similar in the manner by which they effect a 3 preferential release of [ H]ACh from rat forebrain synaptosomes. Furthermore the release facilitating properties of the two toxins show a similar time course, 2 + concentration dependence, and Ca -dependence. Additionally, both pure^-LPT-d and the less pure form of the toxin are similarly affected by inorganic calcium channel antagonists. Release of radioactivity promoted by pure ^-LPT-d does 216 not become maximal until 8 to 10 minutes following exposure of synaptosomes to the toxin. This time course is not as rapid as that described for the related toxin LPT-h (Crosland, 1 982 ; Madeddu et a1. , in press), but is reasonably well correlated with the time course found earlier by Yoshino (1980) using less pure toxin. Experiments conducted by John Connor on LPT-d induced 2 + Ca flux in cultured rat diencephalon cells confirm that the toxin is maximally effective within about eight minutes, there being no substantial increase in 2 + [Ca after that time. It should be noted that any comparison of time courses is complicated by the fact that the time course is a function of toxin concentration. In the cases cited above, the concentrations of toxin are approximately the same (i.e. sufficient to cause release of radioactivity equivalent to very nearly 7 5% of lysate values). When release of radioactivity from synaptosomes is plotted as a function of the concentration of ^-LPT-d (Figure 23), a sigmoid curve is obtained. This evidence of cooperativity is very similar to that described for the less pure toxin (Yoshino, 1980). The Hill coefficient for the more pure toxin is 1.7 (Fig. 23, Inset), whereas that defined earlier is approximately 2.0. From the data collected it is not possible to determine how |3-LPT-d cooperatively stimulates release of 217 ACh, It may be that two molecules of the toxin must bind to a specific receptor in order to effect release. It is also 2 + possible that, since Ca -dependent release at the neuromuscular junction has been shown to be a multi-order function of the extracellular calcium ion concentration (Dodge and Rahamimoff, 1 967 ; Baker, 1 97 5 ; Rahamimoff e t a 1 . . 1 975), the sigmoidal release curve seen in this case is 6 imply a consequence of the binding or sequestration of 2 + ^Ca ions after they have entered the synaptosomes. In any event, the apparent cooperativity of ^-LPT-d in effecting release is another indication that the toxins of related leptinotarsid species are not identical. Release stimulated by j[?-LPT-h is not characterized by cooperative kinetics (Crosland et a1. . 1984). Pure jS-LPT-d effects release primarily of 3 3 [ H]ACh, the relatively small amount of [ H ]cho1ine which is also released being most likely attributable to background leakage (see Ashley et aJL. , 1984). At every concentration of toxin that was studied, at least 70% of 3 the released radioactivity was in the form of [ H]ACh. In this respect, ^-LPT-d is similar to the toxin from L . haldemani. Seventy-four percent of the radioactivity released by ^-LPT-h is represented by the tritiated neuro transmitter (Crosland e_t a 1 . . 1984). In accord with the hypothesis that ^-LPT-d stimulates a preferential release of ACh by opening presynaptic 218 calcium channels, toxin-induced release was found to be an 2 + essentially Ca -dependent process. As the 2 + extracellular Ca concentration is increased from 1 to 4 mM, the amount of radioactivity released also increases (Fig. 24). At ajp-LPT-d concentration of 200 ng/ml, there 2 + is some release even when Ca has been replaced by 2 + Mg , but this is probably due at least partially to 2 + . the presence of some residual Ca in the bathing medium (see Brennan and Cantr ill, 1 980). Even at external 2 + concentrations as low as 0.1 uM, Ca can still effect neurotransmit ter release if an appropriate stimulus causes calcium channels to be opened (Cambell, 1983; Szerb and O'Regan, 1985). 2 + When 1 mM EGTA is substituted for Ca , release is completely blocked at ^-LPT-d concentrations as high as 50 2 + • ng/ml, but a Ca -independent component of release is evident at still higher toxin concentrations (Fig. 25). Similar findings, have been reported by Yoshino (1980) for the less pure toxin of L . decemlineata and by Madeddu et a1 . (in press) for LPT-h. It may be that, like mai to toxin (another presumed Ca^^ channel agonist), j^-LPT-d, at high concentrations, interacts with the presynaptic membrane in a nonspecific way. It has been noted that at concentrations above 300 ng/ml, MTX promotes "very large 2 + increases in Ca uptake, [and this effect] could not be blocked by organic and inorganic calcium channel 219 blockers" (Freedman et a 1.. 1984). It is also possible that removal of Ca^^ from the synaptosomal bathing medium renders the synaptosomal membranes inherently less stable. Schmalzing (1985) has 2 + reported that a reduction of external Ca to 2 x 10 ^ M causes rat cortical synaptosomes to become more permeable to Na^ ions in a way that cannot be blocked by TTX. It may be that membrane-perturbing effects which occur at very low calcium concentrations are exacerbated by exposure of synaptosomes to relatively high concentrations of toxin. That release induced by ^5 —LPT-d is due at least in part to an interaction of the toxin with synaptosomal calcium channels is confirmed by the effects of inorganic calcium channel blockers on toxin stimulated release. La^^, Cd^*^, and Co^^ all block LPT-d-evoked release in the presence of 1 mM Ca^^ (Fig. 27). Furthermore the extent to which the cations are able to inhibit release can be directly related to the order of 2 + potency with which they reduce Ca influx into synaptosomes (Nachshen, 1984; Drapeau and Nachshen, 1984). La^^ is the most effective inhibitor of toxin-induced 2 + 2 + release, followed by Cd and Co 2 + Mn , which is often considered a calcium channel blocker, actually enhances^-LPT-d-stimulated release. This does not necessarily indicate that the toxin is acting' 220 at sites other than presynaptic calcium channels. Drapeau and Nachshen (1984) have found that, under some 2 + circumstances, Mn may be equally as effective at 2 + triggering synaptosomal release as is Ca . Apparently 2 ^ • • Mn ions as well as other typically impermeant cations are able to traverse some types of calcium channels (Hagiwara, 1981) probably due to "tiny structural differences" in the channel pores (Hille,1984). In any 2 + event, Mn supports release induced by depolarization as well as it does release promoted by the toxin, therefore presumably the same mechanism of release is involved in both cases. Pure jS-LPT-d differs from the less pure form described by Yoshino (1980) in only one significant respect. The organic calcium channel antagonist verapamil does not seem to affect release stimulated by the pure toxin even at verapamil concentrations as high as 100 uM. With the less pure form of ^-LPT-d, Yoshino found evidence that verapamil could have a small but statistically significant effect on toxin-induced release. The reason for the discrepancy is not cleai;. It may be that the more purified toxin lacks a contaminant which was responsible for the effects observed by Yoshino. In any event, the lack of effect of verapamil on release induced by purified ^-LPT-d is more consistent with the findings of most other investigators who have studied effects of the ph enylaIky lamine on synaptosomal 221 re lease of Madeddu e t a 1. (in press) have reported that 100 uM verapamil actually potentiates the effect of LPT-h on release of dopamine from rat forebrain synaptosomes, and the authors attribute this to an "unspecific side effect" of the drug. Other investigators find verapamil to have no effect on synaptosomal release of neurotransmitter when the release is induced by depolarization with high (see Miller and Freedman, 1984). The effects of other organic calcium channel antagonists on ^(3-LPT-d-s t imula ted release have also been examined. Nitrendipine, which has been shown to bind with high affinity to specific receptors in rat brain (Gould e_t a 1 . . 1 982; Greenberg e_t a_l. , 1984; 1985), does not inhibit either ^-LPT-d- or depolarization-induced release of 3 [ H]ACh from synaptosomes. Likewise, the related dihydropyridine nifedipine is without effect on synaptosomal release of radioactivity, even at concentrations of this drug as high as 50 uM. These findings suggest that the synaptosomal calcium channels opened by ^^LPT - d and high have properties which make them distinct from the calcium channels of other tissues. This is not very surprising, since there appears to be great diversity in the different types of calcium channels which are present not only in different cells (Kazazoglou et a 1., 1985) but even within the same - ^ membranes (Hagiwara and Byerly, 1981; 1983 ). Ca^^ channels differ with respect to activation rate (Nowvckv e t al.. 1985), inactivation kinetics (Armstrong and Matteson, 1985), voltage dependence, ion selectivity (Hille, 1984), and the effectiveness of inorganic and organic inhibitors. Two types of calcium channel have been reported to exist in rat pituitary GH^ cells (Armstrong and Matteson, 1 985) and in Paramecium cilia (Ehrlich et a 1 .. 1984). Likewise there appear to be at least two different types of calcium channel in synaptosomes derived from both rat (Nachshen, 1984) and mouse (Leslie e_t a 1 . . 1 985) forebrains. One type which is sometimes referred to as the fast, or low threshold, channel is activated at very negative membrane potentials, and is phasic, remaining open on a millisecond time scale (Nachshen, 1984; Mil1er, 1985). Slow channels, on the other hand, are typically activated only at potentials substantially more depolarized than the resting condition and show very little tendency to inactivate (Miller, 1985). Slow channels seem to be particularly sensitive to blockade by organic antagonists (Miller, 1985). The fact that organic antagonists do not seem to have significant effects on ^-LPT-d-stimulated release suggests that the toxin should not be very effective in opening slow calcium channels, and this has been demonstrated in tissues known to be enriched in this type of channel. The toxin 223 does not appear to open calcium channels derived from smooth or skeletal muscle or from clonal neuronal cell lines. Madeddu et a 1.(in press) have reported that the L . 2 + haIdemani toxin can activate the Ca channels of PC 12 cells, but this was evident only at ten fold higher concentrations of the toxin than are required to open synaptosomal calcium channels. It may be that the toxin, at these elevated concentrations, is exerting nonspecific effects as have been described for mai t o toxin (Freedman e t al. . 1 984). The only tissues consistently responsive to the toxin derived from L . decem1ineata have been those of the mammalian central nervous system. There are some preliminary indications that j3“ LPT~d may be able to activate calcium channels of Ommata electric organ, and the related toxin LPT-h is active on mammalian phrenic nerve-hemidiaphragm preparations, but j3-LPT-d seems to act most effectively on calcium channels of the mammalian central nervous system. Release of neurotransmit ter can be elicited from synaptosomes prepared from either rat or mouse brain as well as from isolated rat diencephalon cells. j&-LPT-d is also able to stimulate the release 3 [ H]ACh from bovine brain synaptosomes derived from of either frontal cortex or caudate nucleus. Because the toxin-induced release of radioactivity from synaptosomes takes place over a period of 8 to 10 224 minutes, it is not possible to determine whether fast- or slow-type channels are being primarily affected by -LPT-d. Release of neuro transmit ter has been correlated with calcium entry through both fast and slow channels (Drapeau and Blaustein, 1983 ; Leslie et a 1 .. 1 985), and more recent findings suggest that at least as many as three different types of calcium channel may be present in some neuronal tissues (Fox et al. .1 985). The most tenable conclusion as to the mechanism by which j0-LPT-d effects a preferential release of [^H]ACh from synaptosomes is that the toxin opens at least one and perhaps more than one type of presynaptic calcium channel. Furthermore, the only types of calcium channels which seem to be affected by the toxin are restricted to neurons of the mammalian central nervous system. Unfortunately, because j^-LPT-d binds very nonspecifically to a number of different presynaptic membrane receptors, it may not prove to be as useful a tool for isolating neuronal calcium channels as, for example, verapamil has been in the characterization of calcium channels from skeletal muscle. Nonetheless, ^-LPT-d could still prove to be quite valuable as a means of confirming that calcium channels of the type found in the central nervous system are present and viable in a given experimental preparation. For examp1e, the presence of functional ca1cium 225 channels in proteoliposomes prepared from rat forebrain could be confirmed by exposure of the artificial membranes to ^-LPT-d. If the proteoliposomes had been constructed in a solution containing oxalic acid, those containing functional toxin-sensitive channels could easily be isolated, because they would contain a calcium oxalate precipitate following j^-LPT-d exposure. Isolation of the toxin-sensitive channels could be accomplished by solubilizing both heavy and light proteoliposomes and electrophoretically separating the integral membrane proteins. Heavier proteoliposomes should be characterized by the presence of one or more proteins (corresponding to functional calcium channels) that would not be found in the proteoliposomes devoid of precipitate. The potential usefulness of ^-LPT-d as a molecular probe then, resides in its ability to specifically open calcium channels found in mammalian brains. Although other molecules have been found to bind with specificity and high affinity to receptors in rat brain which are presumably calcium channels, none of these have been shown to have physiological effects except under dramatically altered conditions or at concentrations much higher than those at which specific binding occurs. 226 BIBLIOGRAPHY Abdul-Ghani A.-S., Coutinho-Netto J., and Bradford H.F. K 19 80) In vivo release of acetylcholine evoked by brachial plexus stimulation and tityustoxin. Biochem. Pharmacol. 29. 2179-2182. Abdul-Ghani A.-S., Boyar M.M., Coutinho-Netto J., Bradford H.F. , Berrie C.P., Hulme E.G., and Birdsall N.J.M, (1981) Effect of tityustoxin and sensory stimulation on muscarinic cholinergic receptors in vivo. Biochem. Pharmacol. 3 0 . 2713—2714. Adam-Viz i V. and Ligeti E. ( 1985) Release of acetylcholine from rat brain synaptosomes by various agents in the absence of external calcium ions. J. Physiol. 353. 505-521. Alagon A.C., Possoni L.D., and Erickson B.W. (1978) Isolation and characterization of several proteins from Tityus s erru1a tu s scorpion venom, pp. 609-618 in (Rosenberg P., ed) Toxins : Animal, Plant, and Microbial. Pergamon Press, New York. Angel ides K.J. and Nutter T.J. ( 1983 ) Preparation and characterization of fluorescent scorpion toxins from Le iurus Quinouestriatus ouinquestriatus as probes of the sodium channel of excitable cells. J. Biol. Chem. 258. 11948-11957. Arias €., Sitges M . , and Tapia R. (1984) Stimulation of [ H ]2T-aminobuty r i c acid release by calcium chelators in synaptosomes. J. Neuro ch em. 4 2 . 1507-1514. Armstrong C.M. and Mat teson D.R. ( 1985) Two distinct populations of calcium channels in a clonal cell line of pituitary cells. Science 227. 65-67. Ashley R.E., Brammer, M.J., and Marchbanks R. (1984) Measurement of intrasynaptosomal free calcium by using the fluorescent indicator quin—2. Biochem. J. 219 . 149-158. Baba A. and Cooper J.R. (1980) The action of black widow spider venom on cholinergic mechanisms in synaptosomes. J. Neur o chem. 34 . 1369 — 1379 . Baden D .G. and Corbett M.D. (1979) Peroxidases produced by the marine sponge lotrochota birotulata. Comp. Biochem. Physiol. 6 4 B , 279-283 . 227 Baker P.F, (1975) Excitation-sécrétion coupling. Recent Adv. Physiol. 9_, 51-86. Baker P.P. and Knight D .E . (1978) Ca1cium-dependent exocytosis in bovine adrenal medullary cells with leaky plasma membranes. Nature 2 7 6 . 620—622. Baker P.F. and Knight D .E . (1984) Calcium control of exocytosis in bovine adrenal medullary cells. Trends Neurosci. 7_, 120-126 . Bakus G.J. and Thun M.A. (1979) Bioassays on the toxicity of Caribbean sponges, pp. 417-422. in (C. Levi and N. Boury-Esnau1t, eds) Biologie des Spongi are s. Collogues internationaux du CNRS No. 291. Edition du CNRS, Paris. Baraban J.M., Snyder S.H., Alger B.E. (1985) Protein kinase C regulates ionic conductance in hippocampal pyramidal neurons; E lectrophysiological effects of phorbol esters. Proc. Natl. Acad. Sci. 8 2 . 2538-2542. Barhanin J., Pauron D ., Lombet A . , Hanke W . , Boheim G., and Lazdunski M. (1984) New scorpion toxins with a very high affinity for Na channels. Biochemical characterization and use for the purification of Na channels. J. Physiol. (Paris) 7 9 . 304-308. Bas1ow M.H., ( 1963) Marine Pharmacology. The Williams and Wilkins Co., Baltimore. 286pp. Beani L. , Bianchi C. , S ini s calch i A., Sivilotti L. , Tanganelli S., and Veratti E. (1984) Different approaches to study acetylcholine release: endogenous ACh versus tritium efflux. Naunyn-Schmiedeberg's Arch. Pharmacol. 3 28. 119-126. Bergey G.K., Bigalke E ., and Nelson P.G. (1982) A presynaptic locus of action for tetanus toxin : quantal analysis of effects on monosynaptic transmission in mouse spinal cord neurons in culture. Soc. Neurosci. Abstr. 5 2 2. Birks R.I. (1974) The relationship of transmitter release and storage to fine structure in sympathetic ganglion. J. Neurocytol. 2» 133-160. 228 Bizzini B. (1978) Properties of the binding to gangliosides and synaptosomes of a fragment of the tetanus toxin molecule, pp. 961-96 9 Jja (Rosenberg P., ed) Toxins: Animal, Plant, and Microbial. Pergamon Press, New York. Boksa P. and Collier B. (1980) Spontaneous evoked release of acetylcholine and a cholinergic false transmitter from brain slices: comparison to true and false transmitter in subcellular stores. Neurosci. 5., 1517-1532. Bolton A.E. (1980) Radioiodination Techniques. Amer sham Corp., Arlington Heights, XL. 77pp. Boquet P. and Duflot E . (1982) Tetanus toxin fragment forms channels in lipid vesicles at low pH. Proc. Natl. Acad. Sci. 7_9 , 7614-7618. Bozhkov D. and Kajtazov A. (1976) Study of the nematodes infesting the Colorado beetle Leptinotarsa decemlineata in Bulgaria. Biosis Previews Abstr. 63057220. Bradford M. (1976) A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of pro tein-dye binding. Anal. Biochem. 7 2 . 248-254. Brennan M.J.W. and Cantri11 R.C. ( 1980) Presynaptic membrane protein phosphorylation modulates the release of GABA from preloaded synaptosomes. J. Neur o chem . 3_5 , 506-508. Burgoyne R .D . and Cheek T.R. (1985) Is the transient nature of the secretory responce of chromaffin cells due to inactivation of calcium channels? FEBS Lett. 182 . 115-118. Campbell A.K. (1983) Intracellular Calcium. Its Universal Role as Regulator. John Wiley and Sons, New York. 556pp. Catterall W.A. (1984) The molecular basis of neuronal excitability. Science 223. 653-661. Catterall W.A. (1985) Studies of voltage-sensitive sodium channels using ligand binding methods, pp. 103-121 in (Yamamura H.I., Enna S.J., and Kuhar M.J., eds) Neur o tr ansmi t ter Receptor Binding. Raven Press, New York. 229 Ceccarelli B. and Hurlbut W.P. (1975) The effects of prolonged repetitive stimulation in h em i ch o1 ini um on the frog neuromuscular junction. J. Physiol. 247 . 163-188. Ceccarelli B. and Hurlbut W.P. (1980) Vesicle hypothesis of the release of quanta of acetylcholine. Physiol. Rev. 6 0 . 396-441 . Ceccarelli B., Hurlbut W.P., De Camil1i P., and Me1do 1e s i J. ( 197 8) The effect of extracellular calcium on the topological organization of the plasmalemma in two secretory systems. Ann. N.Y. Acad. Sci. 307. 653-655. Chang C.C., Chen T.F., and Lee C.Y. (1973) Studies of the presynaptic effect of -bungaro toxn on neuromuscular transmission. J. Pharmacol. Exp. Ther. 184. 339-345. Cohen P. (1982) The role of protein phosphorylation in neuronal and hormonal control of cellular activity. Nature 2 9 6 . 613-620. Colin P.L. (1978) Caribbean Reef Invertebrates and Plants. TFH Publications, Inc., Neptune City, NJ. 512pp. 2 + Connor J.A. (1985) Ca measurements using the fluoresecent indicators quin2 and fura-2 combined with digital imaging in mammalian CNS cells. Soc. Neurosci. Abstr. 11 (part 1), p. 176. (ABSTR.). Cooper J.R., Bloom F.E., and Roth R.H. (1982) The Biochemical Basis of Neuropharmacology. Oxford University Press, New York. 367pp. Crosland R.D. (1982) Purification and Characterization of LPT-h, a Protein that Opens Presynaptic Calcium Channels. PhD Thesis, University of So. Calif., Los Angel es. 13 9pp. Crosland R.D., Martin J.V., and McClure W.O. (1983) Effect of liposomes containing various divalent cations on the release of acetylcholine from synaptosomes. J. Neur o ch em. 40 . 681-687 . Crosland R.D., Hsaio T.H., and McClure W.O. (1984) Purification and characterization of leptinotarsin-h, an activator of presynaptic 230 calcium channels. Biochem. 23. 734-741. Curtis D.R. (1971) Tetanus toxin as a neur oph arma CO logical tool, pp. 263-281 in (Simpson L.L., ed) Neuropoisons: Their Pathophysiological Actions, Vol. 1. Plenum Press, New York. Dahl G., Eckerdt R., and Gratzl M. (1979) Models for exocytotic membrane fusion, pp. 349-368 in Symp. Soc. Exp. Biol. 3 3 . Secretory Mechanisms. Cambridge Univ. Press. New York. Del Castillo J. and Katz B. (1957) La base "quantale" de la transmission neuromusculaire, pp. 245-25 8 in Microphysiologie Comparée des Elements Excitables. Paris, Cente Natl. Rech. Sci. Colloq. Int. Vol. 67. DeLorenzo R.J. (1980) Role of calmodulin in neurotransmitter release and synaptic function. Ann. New York Acad. Sci. 3 56. 92-109. DeLorenzo R.J., Freedman S.D., Yohe W.B., and Maurer S.C. (1979) Stimulation of Ca ^-dependent neurotransmitter release and presynaptic nerve terminal protein phosphorylation by calmodulin and a calmodulin-1 ike protein isolated from synaptic vesicles. Proc. Natl. Acad. Sci. 7 6 . 1838-1842 . Dennis M.J. and Miledi R. ( 1974) Electrically induced release of acetylcholine from denervated Schwann cells. J. Physiol. 2 3 7 . 431-452. De Robertis E.D.P. and Bennett H.S. (1954) Submicro SCOpic vesicular component in the synapse. Fed. Proc. H , 35. (ABSTR.). . Diamond I. and Fishman R.A. (1973) High-affinity transport and phosphorylation of 2-deoxy-D-glucose in synaptosomes. J. Neurochem. 20. 1533-1542. Dickey R.W., Miller D.M., and Tindall D.R. (1984) Extraction of a water-soluble toxin from a dinoflagellate, Gambierd is eus toxicus. ACS Symposium Series, No. 262., 257-269. Dodge F.A. and Rahamimoff R. (1967) Co-operative action of calcium ions in transmitter rlease at the neuro muscular junction. J. Physiol. 193. 419-43 2. Dolly J.O., Williams R.S., Black J.D., Tse C.K., 231 Hambleton and Mailing J. (1982) Localization of sites for 1-label led bo tu1inum neuro toxin at murine neuromuscular junction and its binding to rat brain synaptosomes. Toxicon 20. 141-148. Dolly J.O., Halliwell J.V., Black J.D., Williams R.S., Pelchen-Mat thews A., Breeze A.L., Mehraban P., Othman I.B., and Black A.R. (1984) Botulinum neurotoxin and dendrotoxin as probes for studies on neurotransmit ter release. J. Physiol. (Paris) 7 9 . 280-303. Douglas W.W. (1968) Stimulus-secretion coupling: the concept and clues from chromaffin and other cells. Br. J. Pharmacol. 3j4, 451-474. Dowdall M.J. (1974) Biochemical properties of synaptosomes from,the optic lobe of squid (Lo1igo pealii ) . Biochem. Soc. Trans. 2_, 270-274. Dowdall M.J. and Simon E.J. (1973) Comparative studies on synaptosomes: uptake of [N-Me- H ]cho1ine by synaptosomes from squid optic lobes. J. Neurochem. 21. 969-982. Dowdall M.J. and Whittaker V.P. (1973) Comparative studies in synaptosome formation: the preparation of synaptosomes from the head ganglion of the squid, Lo 1 igo pealii. J. Neur ochem. 20 . 921-93 5. Drapeau ^ . and Blaustein M.P. (1983) Initial release of I H ]dopamine frOm rat striatal synaptosomes: correlation with calcium entry. J. Neurosci. 2, 703-713. Drapeau P. and Nachshen D.A. (1984) Manganese fluxes and maganese-dependent neurotransmit ter release in presynaptic nerve endings isolated from rat brain. J. Physiol. 348. 493-510. Dreyer F., Becker C., Bigalke E . , Funk J . , Penner R., Rosenberg F ., and Ziegler M. (1984) Action of botulinum A toxin and tetanus toxin on synaptic transmission. J. Physiol. (Paris) 7 9 . 252-258. Dunant Y . and Israel M. (1979) When the vesicular hypothesis is no longer the vesicular hypothesis. Trends Neurosci. 7^» 130-132. Dunant Y . and Israel M. (1985) The release of acetylcholine. Sci. Amer. 25 2 . 58-66. 232 Dunant Y., Jones G .J. , and Scha11er-Clostre F. ( 1 982) Acetylcholine release in the electric organ of Torpedo. J. Physiol. (Paris) 7^, 357-365. Dunant Y., Israel M . , Lesbats B., and Manaranche R. (1977) Oscillation of acetylcholine during nerve activity in the Torpedo electric organ. Brain Res. 125. 123-140. Dunn L.A. and Eolz R.W. (1983) Catecholamine secretion from digitonin-treated adrenal medullary chromaffin cells. J. Biol. Chem. 258 . 4989-4993 . Edwards C., Dolezal V., Tucek S., Zemkova, E. and Vyskocil F . (1985) Is an acetylcholine transport system responsible for nonquantal release of acetylcholine at the rodent myoneuronal junction? Proc. Natl. Acad. Sci. 82 . 3 514-3 518. Ehrlich B.E., Finke1 stein A., Forte M. , and Rung C. (1984) Voltage-dependent calcium channels from Paramecium cilia incorporated into planar lipid bilayers. Science 2 2 5 . 427—428. Fargues J. (1976) Specificity of pathogenic fungi Imperfecti hvphomvcetes to larvae of Coleoptera 6 ca raba eidae and chrysomelidae. Biosis Previews Abstr. 63038867. Fa 11 P. and Katz B. ( 1952) Spontaneous subthreshold activity at motor nerve endings. J. Physiol. 117. 109-128. Fleckenstein A. (1977) Specific pharmacology of calcium in myocardium, cardiac pacemakers, and vascular smooth muscle. Ann. Rev. Pharmacol. Toxicol. 17. 149-166. Fonnum F . ( 197 5) A rapid radio chemical method for the determination of choline acetyItransferase. J. Neurochem. 24. 407—409. Foreman J.C., Garland L.G., and Mongar J.L. (1976) The role of calcium in secretory processes: model studies in mast cells, pp. 193-218 in Symp. Soc. Exp. Biol, 3 0 . Cambridge Univ. Press, New York. Fox A.P., Nowycky M.C., and Tsien R.W. (1985) Kinetic and pharmacological properties distinguish N-type calcium channels from other types of calcium channel 233 in chick sensory neurons. Soc. Neurosci. Abstr. 11 (Pt. 2), 792. (ABSTR.), Fraker P.J. and Speck J.C. (1978) Protein and cell membrane iodinations with a sparingly soluble chi oroamide, 1,3,4,6-tetrachloro-3a,6a-diphenylgly- coluril. Biochem. Biophys. Res. Comm. 80. 849-857 . Freedman S.B., Dawson G. , Villareal M.L., and Miller R.J. (1984a) Identification and characterization of voltage-sensitive calcium channels in neuronal clonal cell lines. J. Neurosci. 1453-1467. Freedman, S.B., Miller R.J., Miller D.M., and Tindall D.R. (1984b) Interactions of ma i to toxin with voltage-sensitive calcium channels in cultured neuronal cells. Proc. Natl. Acad. Sci. 81, 4582-4585. F r e ir e-Mai a L. , Armeida H.O. , and Lima E.G. ( 197 8a) Mechanism of the pulmonary edema induced by tityustoxin in the rat. p. 6 94 in (Rosenberg P., ed) Toxins: Animal, Plant, and Microbial. Pergamon Press, New York (ABSTR.). Freire-Maia L. , Futuro-Neto H.A., and Cunha-Melo J.R. (1978b) Mechanism of action of tityustoxin on isolated smooth muscles, p.695 in (Rosenberg P., ed)Toxins : Animal, Plant, and Microbial. Permagon Press, New York (ABSTR.). Frelin C., Vigne P., Schweitz E., and Lazdunski M. ( 1984) The interaction of sea anemone and^s corpion neurotoxins with tetrodotoxin-resistant Na channels in rat myoblasts. Mol. Pharmacol. 26. 70-74. Fus s1e R. , Bhakdi S., Sziegoleit A., Tranumjensen J. , Kranz T., and Wellensiek B.-J. (1981) On the mechanism of membrane damage by Staphylococcus aureus ol-toxin. J. Cell Biol. 91 . 83-94. Galizzi J.-P., Posset M . , and Lazdunsk^^M. (1984) Properties of receptors for the Ca -channel blocker verapamil in transverse-tubule membranes of skeletal muscle. Eur. J. Biochem. 144. 211-215. Gengo P.J., Luchowski E., Rampe D.E., Rutledge A., Triggle A.M., Triggle D.J., and Janis R.A. (1983) Chemical and pharmacological approaches to the 234 definition and quantitation of calcium channels. Cold Spring Harbor Symposia on Quant. Biol. 4 8 . 279-285. Ghiasuddin S.M. and Soderlund D.M. (1984) Mouse brain synaptosomal sodium channels: activation by aconitine, batrachotoxin, and veratridine, and inhibition by tetrodotoxin. Comp. Biochem. Physiol. 77C . 267-27 1 . Glenney J.R. and Weber K. (1980) Calmodulin-binding proteins of the microfilaments present in isolated brush borders and microvilli of intestinal epithelial cells. J. Biol. Chem. 2 5 5 . 10551-10554. Goldberg A.M. and McCaman R.E. (1973) The determination of picomole amounts of acetylcholine in mammalian brain. J. N eur o ch em. 20. 1-8. Goldberg E. (1975) Lactate dehydrogenase-X from mouse testes and spermatozoa, pp. 318-323 in (Wood W.P., ed) Methods in Enzymology, Vol. XL I. Acad. Press, New York. Go 11 A., Ferry D.R., and Glossmann H. (19B4^^Target size analysis and molecular properties of Ca clnnels labeled with [ H ]verapamil. Eur. J. Biochem. 141. 177-186. Gomez M.V. and Ribeiro A.M. (1985) The effects of calmodulin inhibition on the release of acetylcholine and protein phosphorylation induced by t i t T V, ; toxin. Trans. Amer. Soc. Neur o ch em . 1 6 . 164 IABSTR.). Gorio A. and Mauro A. (1979) Reversibility and mode of action of black widow spider venom on the vertebrate neuromuscular junction. J. Gen. Physiol. 73. 245-263. Gould-R.J., Murphy K.M.M., and Snyder S.E. (1982) [ E]Nitredipine-labeled calcium channels discriminate inorganic calcium agonists and antagonists. Proc. Natl. Acad. Sci. 7 9 . 3656-3660. Gould R.J., Murphy K.M., and Snyder S.E. (1985) Autoradiographic localization of calcium channel antogonist receptors in rat brain with [ E ]nitredipine. Brain Res. 3 3 0 . 217-223. 23 5 Gray E.G. and Whittaker V.P. (1962) The isolation of nerve endings from brain: an electron microscopic study of cell fragments derived by homogenization and centrifugation. J. Ana t. 7 9-87 . Green G. (1977) Ecology of toxicity in marine sponges. Mar. Biol. 40 . 207-215. Greenberg D .A . , Cooper E.C., and Carpenter C.L. (1984) Calcium channel 'agonist' BAY K 8644 inhibits calcium antagonist binding to brain and PCI 2 membranes. Brain Res. 3 0 5 . 365-368. Greenberg D.A., Cooper E.C., and Carpenter C.L. (1985) Reversible dihydropyridine isothiocyanate binding to brain calcium channels. J. Neur o ch em. 44. 319-321 . Grigorovich I.V. and Kravchenko O.N. (1978) Large scale infestation of the Colorado beetle Lep t ino t ar s a dec emlinea ta by parasitic nematodes from the genus Hexamermis and their effect on the vitality of the pest. Biosis Previews Ahstr. 67035454. Grynkiewicz G., Poen^e M. , and Tsien R.Y. (1985) A new generation of Ca * indicators with greatly improved fluorescence properties. J. Biol. Chem. 260. 3440-3450. Gunderson C.B. and Howard B.D. (1978) The effects of botulinum toxin on acetylcholine metabolism in mouse brain slices and synaptosomes. J. Neurochem. 31. 1005-1013. Gunderson C.B. and Jenden D.J. (1981) Notexin preferentially inhibits the release of newly synthesized acetylcholine from rat brain synaptosomal fractions. J. Neurosci. 1_> 1113-1116. Ea eds trom R. ( 197 7 ) The Beetles of America. A.S. Barnes and Co., New York. 488pp. Eaghighat N ., Cohen R.S., and Pappas G .D . (1984) Fine structure of squid (Lo1igo pealei) optic lobe synapses. Neurosci. 13. 527-546. Hagiwara S. and Byerly L. (1981) Calcium channel. Ann. Rev. Neurosci. j4, 6 9-125 . Hagiwara S. and Byerly L. (1983) The calcium channel. 236 Trends Neurosci. 189-193 . Halliwell J.V. and Dolly J.O. (1982) Electrophysiological analysis of the presynaptic action of 5-bnngarotoxin in the central nervous system. Toxicon 2j0, 121-127 . Hal 1iwe11 J.V., Tse C.K., Spokes J.W., 0thman I. , and Dolly J.O. (1982) Biochemical and electrophysiological demonstrations of the actions of -bungarotoxin on synapses in brain. J. Neurochem. 3 9 . 543-550 . Halpert J. (1979) Structure and function of neuro- and myotoxic phospholipases: modification with ethoxy formic anhydride of Notechis 11-5 from the venom of the Australian tiger snake Notechis scutatus s cu ta tus. pp. 45-62 in (Ceccarelli B. and Clementi F., eds) Advances in Cy toph arma cology, Vol.3. Raven Press, New York. Halstead B.W. (191) Current status of marine biotoxicology : an overview. Clin. Toxicol. 18 . 1-24. Earary E. and Brown J.E. (1984) Spatially nonuniform changes in intracellular calcium ion concentrations. Science 224. 292-294. Eare J.D. ( 1983 ) Seasonal variation in plant-in se ct associations: utilization of So 1anum du 1camara by Lep t ino t ar s a decemlineata. Ecology 6 4 . 345-361 . Eartshorne R.P., Keller B.Ü., Talvenheimo J.A., Cat teral1 W.A. , and Montai M. ( 1985) Functional reconstitution of the purified brain sodium channel in planar lipid bilayers. Proc. Natl. Acad. Sci. 82. 240-244. Earvey A.L. (1982) Common site of action of dendrotoxin-like faciltatory neuro toxins and - bungarotoxin. pp. 699-700 (Rosenberg P., ed) Toxins: Animal, Plant, and Microbial. Pergamon Press, New York. (ABSTR.). Earvey A.L., Anderson A.J., and Karlsson E. (1984) Facilitation of transmitter release by neurotoxins from snake venoms. J. Physiol. (Paris) 7 9 . 222-227 . Haycock J.W. and Meligeni J.A. ( 1 977 ) Neurotransmit ter 237 accumulation and calcium-dependent release from different regions of rat brain. Life Sci. 21 . 1837-1844. Henderson R.C, and McClure W.O. (1978) Effects of extracts of black widow venom glands on pool sizes of choline and acetylcholine in rat brain synaptosomes. pp. 699-700 Jja (Rosenberg P., ed) Toxins: Animal, plant and Microbial. Pergamon Press, New York. (ABSTR.). Heuser J.E. and Reese T.S. (1981) Structural changes after transmitter release at the frog neuromuscular junction. J. Cell Biol. 88 . 564-580 . Heuser J.E., Reese T.S., Dennis M.J., Jan Y ., Jan L., and Evans L. ( 197 9) Synaptic vesicle exocy tosis captured by quick-freezing and correlated with quanta1 transmitter release. J. Cell. Biol. 81 . 275-300. Hille B. (1984) Ionic Channels of Excitable Membranes. Sinauer Assoc., Inc., Sunderland, Mass. 426 pp. Ho ch D.H., Romero-Mira M. , Ehrlich B.E., Finkle stein A., DasGupta B.R., and S imp son L.L. ( 1985) Channels formed by botulinum, tetanus, and diphtheria toxins in planar lipid bilayers : Relevance to translocation of proteins across membranes. Proc. Natl. Acad. Sci. 82 . 1692-1696 . Holtzman E., Freeman A.R., and Kashner L.A. (1971) Stimulation-dependent alterations in peroxidase uptake at lobster neuromuscular junctions. Science 173. 733-736. Hopkins C.R. (1979) The secretory process in outline, pp. 1-8 Ln Symp. Soc. Exp. Biol., 33. Secretory Mechanisms. Cambridge Univ. Press., New York. Eos tounsky Z. and Weiser J. ( 1 97 8) Pleis tophora gros sa new species PIeistophoridae microsporidia parasite of chrysomelid beetles in Yugoslavia. Biosis Previews Abstr. 67016277 . Howard B. (1978) Effects of polypeptide neurotoxins on acetylcholine storage and release, pp. 56 5-585 _in_ (Jenden D.J., ed) Cholinergic Mechanisms and Psychopharmacology. Plenum Press, New York. Esaio T.E. and Fraenkel G. (1969) Properties of 238 1eptinotars in, a toxic protein from the Colorado potato beetle. Toxicon %, 119-130. Hubbard J.I., Llinas R., and Quastel D.M.J. (1969) Electrophysiological Analysis of Synaptic Transmission. Edward Arnold Pub 1. , London. 372pp. Ikarashi Y., Sa Sahara T., and Maruyama Y . ( 1985) Determination of choline and acetylcholine levels in rat brain regions by liquid chromatpgraphy with electrochemical detection. J. Chroma t ogr. 3 2 2 . 191-199,. Israel M. and Lesbats B. (1981) Continuous determination by a chemiluminescent method of acetylcholine release and compartmentation in Torpedo electric organ synaptos om es. J. Neurochem. 37. 147 5-1483. Israel M. and Manaranche R. (1985) Cholinergic transmission, is it the result of the release of cytosolic acetylcholine? pp. 229-250 in (Ro thman S.S. and Ho J.J.L., eds) Nonvesicular Transport. John Wiley and Sons, New York. Israel M. , Dunant Y. , and Manaranche R. ( 197 9) The present status of the vesicle hypothesis. Progr. Neurobiol. 13. 237-275. Israel M. , Lesbats B. , and Manaranche R. ( 1978) Quantitative description of acetylcholine release and fluctuations in nerve terminals of Torpedo electric organ submitted to stimulation. Pflugers Arch. 377. 117-118. Israel M. , Lesbats B. , Manaranche R. , Morel N. , Gulik-Krzywicki T., and Dedieu J.C. (1982) Rearrangement of intramembrane particles as a possible mechanism for the release of acetylcholine. J. Physiol. (Paris) 7 8 . 348-356 . Israel M. , Lesbats B. , Manaranche R. , and Morel N. ( 1983) Acetylcholine release from proteolipos ome s equipped with synaptosomal membrane constituents. Biochim. Biophys. Acta 7 28. 438-448. Janis R.A., Rampe D ., Sarmiento J.G., and Triggle D.J. (1984) Specific binding of a calcium channel activator, [ H ]Bay K 86 44, to membranes from cardiac muscle and brain. Biochem. Biophys. Res. Comm. 121. 317-323. 23 9 Johnson M.K. and Whittaker V.P. (1963) Lactate dehydrogenase as a cytoplasmic marker in brain. Biochem. J. 88. 404-40 9. Jope R.S. (1981) Acetylcholine turnover and compartmentation in rat brain synaptosomes. J. Neurochem. 36. 1712—1721. Jover E. , Martin-Moutot N. , Cour aud F. , and Ro chat H. (1978) Scorpion toxin: specific binding to rat brain synaptosomes. Biochem. Biophys. Res. Comm. 8 5 . 37 7-3 82. Junge D . (1981) Nerve and Muscle Excitation. Sinauer Assoc., Inc., Sunderland, Mass. pp. 159-165. Kalman M. and Eajos F . (1982) Comparative studies of fresh and post mortem isolated rat cortical 6ynap t os ome s : morphology, K transport and respiration. J. Neurochem. 3 9 . 17 40-1742. Kato A.C., Katz H.S., and Collier B. (1974) Absence of adenine nucleotide release from autonomic ganglion. Nature 249. 576-577. Katz B. (1969) The Release of Neural Transmitter Substances. The Sherrington Lectures, Vol. X. Liverpool Univ. Press, Liverpool. 60 pp. Katz B. and Miledi R. ( 1967a) The timing of calcium action during neuromuscular transmission. J. Physiol. 189. 535-544. Katz B. and Miledi R. ( 1967b) A study of synaptic transmission in the absence of nerve impulses. J. Physiol. 19 2 . 407-436. Kazazoglou T., Schackmann R.W., Fosset M., and Shapiro B.M. (1985) Calcium channel antagonists inhibit the acrosome reaction and bind to plasma membranes of sea urchin sperm. Proc. Natl. Acad. Sci. 8 2 . 1 46 0—1 46 4 . Kelly R.B. , Deutsch J.W. , Carlson S.S. , and Wagner J.A . (1979) Biochemistry of neurotransmitter release. Ann. Rev. Neurosci. 2_, 399-446 . Kelly R.B., Miljanich G.P., and Pfeffer S. (1983) Presynaptic mechanisms of neuromuscular transmission, pp. 43-104 in (Albuquerque E.X. and Eldefrawi A.T., eds) Myasthenia Gravis. Chapman and 240 Ha11, New York. Kerr L.M. and Yoshikami D . (1982) A toxic peptide from the venom of a marine snail blocks calcium evoked-release of transmitter. Soc. Neurosci. Abstr. Vol. ^ p. 497, (ABSTR.). Klee C.B., Crouch T.H., and Richman P.G. (1980) Calmodulin. Ann. Rev. Biochem. 49. 489-515. Kongsamut S., Freedman S.B., Simon B.E., and Miller R.J. (1985) Interaction of steroidal alkaloid toxins with calcium channels in neuronal cell lines. Life Sci. 36. 1493-1501, Kostyuk P.G. (1985) Metabolic dependence of ionic channel function studied in the nerve cell membrane. Neurophysiol. (Engl, transi.) 16. 221-229. Laemmli U.K. (1970) Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227. 680-685. Landmesser L. and Pilar G. (1972) The onset and development of transmission in the chick ciliary ganglion. J. Physiol. 222 . 691-713 . Large W.A. and Rang H.P. (1978a) Factors affecting the rate of incorporation of a false transmitter into mammalian motor nerve terminals. J. Physiol. 2 8 5 . 1-24. Large W.A. and Rang H.P. (1978b) Variability of transmitter quanta released during incorporation of a false transmitter into cholinergic nerve terminals. J. Physiol. 2 8 5 . 25-34. Lazarovici P., Tayot J.L., and Yavin E. (1984) Affinity chromatographic purification and characterization of two iodinated tetanus toxin fractions exhibiting different binding properties. Toxicon 22. 401-413 . Legrand A.M. and Bagnis R. (1984a) Effects of highly purified maitotoxin extracted from dinoflagellate Gambierdiscus toxicus on action potential of isolated rat heart. J. Mol. Cell Cardiol. 16 . 663-666 . Legrand A.M. and Bagnis R. (1984b) Effects of ciguatoxin and maitotoxin on isolated rat atria and 241 rabbit duodenum. Toxi con 22. 47 1-47 5 . Lemos J.R., Novak-Hofer I., and Levitan I.B. (1984) Synaptic stimulation alters protein phosphorylation in vivo in a single Ap1vs ia neuron. Proc. Natl. Acad. Sci. 3233-3237. Leslie S.W., %arr E., and Chandler L^^. ^1983) Comparison of voltage-dependent Ca uptake rates by synaptosomes isolated from rat brain regions. J. Neurochem. 4 1 . 1602-1605. Leslie S.W., Woodward J.J., and Wilcox R.E. (1985) Correlation of rates of calcium entry and endogenous dopamine release in mouse striatal synaptosomes. Brain Res. 3 2 5 . 99-105. Lin-Shiau S.-Y. and Chen C.-C. (1982) Effects of j5-bungarotoxin and phospholipase A from Naia na 1 a atra snake venom on ATPase activities of synaptic membranes from rat cerebral cortex. Toxicon 20 . 409-417 . Llinas R.R. (1982) Calcium in synaptic transmission. Sci. Amer. 247. 56-65. Llinas R. and Nicholson C. (1975) Calcium role in depolarization-secretion coupling: an aequorin study in squid giant synapse. Proc Natl. Acad. Sci. 7 2 . 187-190. Llinas R., Steinberg I.Z., and Walton K. (1976) Presynaptic calcium currents and their relation to synaptic transmission: Voltage clamp study in squid giant synapse and theoretical model for the calcium gate. Proc. Natl. Acad. Sci. 73 . 2918-2922 . Llinas R., Steinberg I.Z., and Walton K. (1981) Relationship between presynaptic calcium current and postsynaptic potential in squid giant synapse. Biophys. J. 22» 323-351. Macedo T.M. and Gomez M.V. (1982) Effects of ti tyustoxin (TsTx) from scorpion venom on the release and synthesis of acetylcholine in brain slices. Toxicon 2_0_, 601-606 . Macintosh F.C. (1978) The present status of the vesicle hypothesis, pp. 297-319 Ln (Jenden D.J., ed) Cholinergic Mechanisms and Psychopharmacology Plenum Press, New York. 242 Madeddu L., Saito I,, Esaio T.H,, and Meldolesi J, (in press, a) Leptino toxin-h action in synaptosomes, neurosecretory cells and artificial membranes. II. Stimulation of neurotransmitter release. J. Neurochem. Madeddu L ., Pozzan T., Robe11o M. , Rolandi R. , Hsaio T., and Meldolesi J. (in press, b) Leptinotoxin-h action in synaptosomes; neurosecretory cells and artificial membranes. I. Stimulation of ion fluxes. J. Neurochem. Manaranche R. , Thief fry M. , and Israel M. ( 1980) Effect of the venom of Glycera convoluta on the spontaneous quantal release of transmitter. J. Cell Biol. 8 5 . 446-458. Manaranche R. , Thief fry M. , Israel M. , and Gulik-Rrzywicki T. (1982) Presynaptic action of G1vcera convoluta venom. Toxicon 20. 118-119. (ABSTR.). Mane ckj e e R . , Zukin R.S., Archer S., Michael J., and Osei-Gyimah P. (1985) Purification and characterization of the -opiate receptor from rat brain using affinity chromatography. Proc. Natl. Acad. Sci. 8_2 , 594-598 . Marchbanks R.M. (1975) The subcellular origin of the acetylcholine released at synapses. Int. J. Biochem. 6_, 303-312. Marchbanks R.M. (1978) The vesicular hypothesis questioned. Trends Neurosci. 1_, 83-84. Martin J.V. and McClure W.O. ( 1982) Iotrochotin, a novel factor from a Caribbean sponge, increases release of neurotransmitters. Soc. Neurosci. Abstr. 8, 516. Matsuda M. and Yoneda M. ( 197 8) Structure-function relationship of tetanus neurotoxin molecule: analyses with complementary polypeptide fragments and a subfragment, pp. 977-989 in (Rosenberg P., ed) Toxins: Animal, Plant, and Microbial. Pergamon Press, New York. McClure W.O. and Martin J.V. (1983) On the probability of finding shark repellants in marine organisms, pp. 75-90 in (Zahuranec B.H., ed) Shark Repellants 243 from the Sea. AÂÂS Selected Symposia v. 83. Westview Press, Boulder CO. McClure W.O., Abbott B.C., Baxter D.E., Hsiao T.H., Satin L.S., Siger A., and Yoshino J.E. (1980) Leptinotarsin: a presynaptic neurotoxin that stimulates release of acetylcholine. Proc. Natl. Acad. Sci. 7%, 1219-1223. McClure W.O., Baxter D.E., Brusca R., Crosland R.D., Hsaio T.H., Koenig M.L., Martin J.V., and Yoshino J.E. (in press) Studies of two novel presynaptic neuro toxins. _in (Hanin I., ed) Dynamics of Cholinergic Function. Plenum Press, New York. McGeer E.G. and McGeer P.L. (1981) Neurotoxins as tools in neurobiology. Int. Rev. Neurobiol. 22 . 173-204. McGraw C.F., Somlyo A.V. and Blaustein M.P. (1980) Localization of calcium in presynaptic nerve terminals. J. Cell Biol. 8 5 . 228-241. McPherson M.A. and Dormer R.L. (1984) Mucin release and calcium fluxes in isolated rat submandibular acini. Biochem. J. 224. 473-481. Meldolesi J., Huttner W.B., Tsjen R.Y., and Pozzan T. (1984) Free cytoplasmic Ca and neurotransmitter release: Studies on PC12 cells and synaptosomes exposed to - latrotoxin. Proc. Natl. Acad. Sci. 81_, 620-624. Melega W.P. and Howard B.D. (1984) Biochemical evidence that vesicles are the source of the acetylcholine released from stimulated PC12 cells. Proc. Natl. Acad. Sci. 81 .653 5-6538. Merril C.R., Goldman D., Sedman S.A., and Ebert M.H. (1981) Ultrasensitive stain for proteins in polyacrylamide gels shows regional variation in cerebrospinal fluid proteins. Science 211. 1437-1438. Metezeau P. and Desban M. (1982) Botulinum toxin type A: kinetics of calcium dependent paralysis of the neuromuscular junction and antagonism by drugs and animal toxins. Toxicon 20 , 649-654. Metz 1er D.E. ( 1977) Biochemistry. The Chemical Reactions of Living Cells. Academic Press, New York. 2 44 1129pp. Meyer E.M. and Cooper J.R. (1982) High affinity choline transport in proteolipos om e s derived from rat cortical synaptosomes. Science 217. 843-845. Meyer E.M. and Otero D.E. (1985) Pharmacological and ionic characterization of the muscarinic receptors modulating [ ]a cetyIcholine release from rat cortical synaptosomes. J. Neurosci. _5.» 1202-1207 . Michael son D.M. ( 197 8) Is presynaptic acetylcholine release accomplished by the secretion of the synaptic vesicle contents? FEES Lett. 89. 51-53. Middlemiss D.N. and Spedding M. (1985) A functional correlate for the dihydropyridine binding site in rat brain. Nature 314. 94-96. Miledi R. (1973) Transmitter release induced by injection of calcium ions into nerve terminals. Proc. R. Soc. Lond. B. 18 3 . 421-425. Miljanich G.P., Brasier A.R., and Kelly R.B. (1982) Partial purification of presynaptic plasma membrane by immuno a dsorption. J. Cell Biol. 9 4 . 88-96 . Miller D.M., Dickey R.W., and Tindall D.R. (1984) Lipid-extracted toxins from a dinoflagellate, Gambierdis eus toxicus . ACS Symposium Series No. 262, 241-255. Miller D.M., Tindall D.R., and Hassan F. (1985) Effects of mai totoxin on guinea pig ileum. Fed. Proc. 44 . 1642. (ABSTR.). Miller R.J. (1985) How many types of calcium channels exist in neurones? Trends Neurosci. 8_, 45-47 . Miller R.J. and Freedman S.'B. (1984) Minireview. Are dihydropyridine binding sites voltage-dependent • calcium channels? Life Sci. 34. 1205-1221. Morel N. and Meunier F.-M. (1981) Simultaneous release of acetylcholine and ATP from stimulated cholinergic synaptosomes. J. Neurochem. 3 6 . 1766-1773 . Murphy K.M.M., Gould R.J., Largent B.L., and Snyder S.H. A unitary mechanism of calcium channel antagonist drug action. Proc. Natl. Acad. Sci. 80. 860-864. 245 Myers C.W. and Daly J.W, (1983) Dart-poison frogs. Sci. Amer. 248. 120-133. Nachshen D.A. (1984) Selectivity of the Ca binding site in synaptosome Ca channels. J. Gen. Physiol. 83 . 941-967. Nachshen D.A. (1985) The early time course of po ta s s i um-s t imula ted calcium uptake in presynaptic nerve terminals isolated from rat brain. J. Physiol. 3 6 1 . 251-268. Nachshen D.A. and Drapeau P. (1982) A buffering model for calcium-dependent neurotransmitter release. Biophys. J, 3^, 205-208. Nairn R.C. (1976) Fluorescent Protein Tracing (4 ed.) Churchill Livingstone, New York. 648 pp. Neigel J.E. and Avise J.C. (1983) Histocompatibility bioassays of population structure in marine sponges: clonal structure in Verone ia long i s s ima and lotrochota birotulata. J. Heredity 74. 134-140. Nelson M.T., French R.J., and Krueger B.K. (1984) Voltage-dependent calcium channels from brain incorporated into planar lipid bilayers. Nature 308. 77-80. Norman R.I. , Borsotto M. , Fosset M. , Lazdunski M. , and E1lory J.C. ( 1983 ) Determination of^the molecular size of the nitredipine-sensitive Ca channel by radiation inactivation. Biochem. Biophys. Res. Comm. 111. 87 8 — 883 . Nowycky M.C., Fox A.P., and Tsien R.W. (1985) Long-opening mode of gating of neuronal calcium channels and its promotion by the dihydropyridine calcium agonist BAY K 86 44. Proc. Natl. Acad. Sci. 82, 2178-2182. O'Fallon J^V. ,^Brosemer R.W., and Harding J.W. (1981) The Na , K -ATPase: a plausible trigger for voltage-independent release of cytoplasmic neurotransmitters. J. Neurochem. 3 6 . 369-378 . 0'Farrel1 P.H. (1975) High resolution two-dimensional electrophoresis of proteins. J. Biol. Chem. 2 50. 4007-4021. 246 0'Regan S., Collier B,, and Israel M, (1982) Studies on presynaptic cholinergic mechanisms using analogues of choline and acetate. J. Physiol. (Paris) 78 . 454-460. Palade G.E. and Palay S.L. ( 1954) Electron micros cope observations of interneuronal and neuromuscular synapses. Anat. Rec. 118. 335-336. Panfili E., Sandri G., Liut G., Stancher B., and Sottocasa G. (1983) An improved technique for the preparation of the calcium-binding protein involved in the mitochondrial calcium uniport. pp. 347-364 in Calcium Binding Proteins 1983. Elsevier Science Pub1., New York. Pant B.C. , Gallant P.E. , Cohen R. , Neary J.T. , and Gainer E. (1983) Calcium-dependent 4-aminopyridine stimulation of protein phosphorylation in squid optic lobe synaptosomes. Cell Mol. Neurobiol. 3_, 223-238. Papahadjopoulos D ., Portis A., and Pangborn W. (1978) Calc ium-induced lipid phase transitions and membrane fusion. Ann. New York Acad. Sci. 308. 50-63. Parker R. (1970) Comparative Studies of Some Biochemical Components in the Eemolymph of Two Species of Lent ino t ar sa. PhD Thesis. Kent State Univ., 16 9pp. Paulson J.C. (1974) Detection and Quantitation of Drug-induced Inhibition of Rapid Axoplasmic Transport: Microtubules and Axoplasmic Transport. Ph.D. Thesis. Univ. of Illinois, Urbana-Champa ign. 20 8pp. Peterson C. and Gibson G.E. (1983) Aging and 3,4-d iaminopyr id ine alter synaptosomal calcium uptake. J. Biol. Chem. 258. 11482-11486. Pollard H.B. and Pappas G.D. (1979) Veratridine-activated release of adenosine-5'-triphosphate from synaptosomes: evidence for calcium dependence and blockade by tetrodotoxin. Biochem. Biophys. Res. Comm. 88 . 1315-1321 . Pozzan T. , Gat ti G . , Dozio.N. , Vicentini L.M. , and Meldolesi J.(1984) Ca ^-dependent and independent release of neurotransmit ters from PC12 247 cells: a role for protein kinase C activation? J. Cell Biol. 11, 628-638. Pump1 in D.W. and Del Castillo J. ( 197 5) Release of packets of acetylcholine and synaptic vesicles elicited by brown widow spider venom in frog motor nerve endings poisoned by botulinum toxin. Life Sci. 12. 137-142. Pump1 in D.W. and Reese T.S. ( 1977) Action of brown widow spider venom and botulinum toxin on the frog neuromuscular junction examined with the f reeze-f ractur e technique. J. Physiol. 22.1* 443-457. Rahamimoff R. , Erulkar S.D., Aina e s E. » Meiri E., and Rahamimoff E. (197 5) Modulation of transmitter release by calcium ions and nerve impulses. Cold Spring Earbor Symp. Quant. Biol. 4Q.* 107-116. Rahamimoff R.* Meiri E ., Erulkar S.D., and Barenholz Y . (1978) Changes in transmitter release induced by ion-containing liposomes. Proc. Natl. Acad. Sci. 21. 5214-5216. Ra it er i M. , Marchi M. , and Maura G. ( 1984) Release of cat e cholamine s. serotonin, and acetylcholine from isolated brain tissue, pp. 431-462 is (Lajtha A., ed) handbook of Neurochemis try (2 ed.). Vol. 6. Plenum Press, New York. Rampe D . , Janis R.A., and Triggle ( 1984) BAY K 86 44, a 1,4-dihyrdopyridine Ca channel activator: dissociation of binding and functional effects in brain synaptosomes. J. Neurochem. 43 « 1688-1692. Rathmayer W. , Ruhland M . , Tintpulver M. , Walther C. and Zlotkin E . ( 197 8) The effect of toxins derived from the venom of the scorpion An dr o_c tonus austr a 1 is Eector on neuromuscular transmission, pp. 629-637 in (Rosenberg P., ed) Toxins : Animal, Plant, and Microbial. Pergamon Press, New York. Ray R., Morrow C.S., and Catterall W.A. (1978) Binding of scorpion toxin to receptor sites associated with voltage-sensitive sodium channels in synaptic nerve ending particles. J. Biol. Chem. 2 5 3 . 7307-7313. Reese J.E. and Cooper J.R. (1982) Modulation of the 248 adenosine and adenosine 5 '-triphosphate. J. Pharmacol. Exp. Ther. 2 23. 612-616. Rehm H.A. and Betz H. (1982) Specific cytotoxicity and binding of -bungarotoxin in chick CNS. Soc. Neurosci. Abstr. Vol. 8. p. 794. (ABSTR.). Rephaeli A. and Parg^ons ^ S . N . ( 1982) Calmodulin and stimulation of Ca transport and protein phosphorylation in cholinergic synaptic vesicles. Proc. Natl. Acad. Sci. 7 9 . 5783-5787. Rhoads D.E., Osborn L.D., Peterson N.A., and Raghupa thy E. ( 1983 ) Release of neur o transmi t ter amino acids from synaptosomes: Enhancement of calcium- independent efflux by oleic and arachidonic acids. J. Neurochem. 41 . 531-537. Ribeiro J.A. and Sa-Almeida A.M. (1984) Excitatory effects of dibutyryl cyclic AMP, noradrenaline, and theophylline on Ca uptake by synaptosomes from rat brain. Arch. Int. Pharmacodyn. 27 0 . 215-231. Richards C.D., Metcalfe J.C., Smith G.A., and Hesketh T.R. ( 1984) Changes in free-calcium levels and pH in synaptosomes during transmitter release. Biochim. Biophys. Acta. 803. 215-220. Robinson P.J. and Dunkley P.R. (1983) Depolarisation- dependent protein phosphorylation in rat cortical synaptosomes: Factors determining the magnitude of the response. J. Neurochem. 41. 909-918. Robinson P.J. and Dunkley P.R. (1985) Depolarisation- dependent protein phosphorylation and dephosporylation in rat cortical synaptosomes is modulated by calcium. J. Neurochem. 44. 338-348. Robinson P.J., Jarvie P.E., and Dunkley P.R. (1984) Depolarisation-dependent protein phosphorylation in rat cortical synaptosomes is inhibited by fluphenazine at a step after calcium entry. J. Neurochem. 43. 659-667. Rosenberg R.L., Tomiko S.A., and Agnew W.S. (1984) Reconstitution of neurotoxin-modulated ion transport by the voltage regulated sodium channel isolated from the electroplax of Electronhorus electricus . Proc. Natl. Acad. Sci. 8J^, 123 9-1243 . Rowell P.P. and Winkler D.L. (1984) Nicotinic 249 2 stimulation of [ H ]a cety1cho1ine release from mouse cerebral cortical synaptosomes. J. Neurochem. 43 . 1 593-1598 . Rubin R.P. (1970) The role of calcium in the release of neurotransmit ter substances and hormones. Pharmacol. Rev. 2 2 . 389-428. Rudyak I . I . ( 1977 ) Eating of Colorado beetle imagoes Lent ino tars a decemlineata by frogs Rana ridibunda. Biosis Previews Abstr. 78031703 . Saimi Y. and Rung C. (1982) Are ions involved in the gating of calcium channels? Science 218. 153-156. Saimi Y ., Hinrichsen R.D., Forte M. , gnd Rung C. (1983) Mutant analysis shows that the Ca -induced K current shuts off one type of excitation in Paramec ium. Proc. Natl. Acad. Sci. 8 0 . 5112-5116 . Sanguinetti M.C. and Kass R.S. (1984) Regulation of cardiac calcium channel current and contracti1e activity by the dihydropyridine BAY K 86 44 is voltage-dependent. J. Mol. Cell. Cardiol. 16. 667-670. Santos R.R., Pimenta A.F., Coutinho-Netto J., and Diniz C.R. (1978) Tityustoxin interaction with neuromuscular tissues as revealed by immunofluorescence, p. 6 96 rn(Rosenberg P.,ed) Toxins: Animal, Plant, and Microbial. Pergamon Press, New York. (ABSTR.). Schettini G . , Koike K. , Login I . S., Judd A.M., Cronin M.J., Yasumoto T., and MacLeod R. (19 84) Maitotoxin stimulates hormonal release and calcium flux in rat anterior pituitary cells in_ vitro. Am. J. Physiol. 247. E520-E525. Schof|elmeer A.N.M. and Mulder A.H. (1983) [ E]Noradrenaline release from brain slices induced by an increase in the intracellular sodium concentration: role of intracellular calcium stores. J. Neurochem. 40. 615-621. Shramm M., Thomas G ., Towart R., and Franckowiak G. (1983) Novel dihydropyridine with positive inotropic action through activation of Ca channels. Nature 303. 535-537. 250 Schwarz L.M., McCleskey E.W*, and Aimers W. (1985) Dihydropyridine receptors in muscle are voltage-dependent but most are not functional calcium channels. Nature 314. 747-750. Seamon K.B., Vaillancourt R., Edwards M . , and Daly J.W. (1984) Binding of [ H]forskolin to rat brain membranes. Proc. Natl. Acad. Sci. 8_2, 5081-5085 . Sevcik C. and Barboza C.A. (1983) The presynaptic effect of fractions isolated from the sponge Tedania ignis . Toxicon 2_2, 1 91-200 . Shagov E.M. (1977) Temperature preferences of the two-spotted stink bug, parasite of the Colorado potato beetle. Biosis Previews Abstr. 65071099. Shagov E.M. (1978) Photoperiodic reaction and variability in the two— spotted stink bug Peri1lus bioculatus . Biosis Previews Abstr. 66034201 . S il insky E.M, ( 1982a) Recent approaches to secretion and S timulus-secretion coupling. Fed. Proc. 41 . 2169-2171 . Silinsky E.M. (1982b) Properties of calcium receptors that initiate depolarization-secretion coupling. Fed. Proc. 42, 2172-2180. Silinsky E.M. (1985) The biophysical pharmacology of calcium-dependent acetylcholine secretion. Pharmacol. Rev. 3 7 . 81-132. Simpson L.L. (1981) The origin, structure, and pharmacological activity of botulinum toxin. Pharmacol. Rev. 3 3 . 1 55-188. Smolien J.E. and Stoehr S.J. (1985) Micromolar concentrations of free calcium provoke secretion of lysozyme from human neutrophils permeabi1ized with saponin. J. Immunol. 13 4 . 1859-1865. Snelling R. and Nicholls D. (1984) The calmodulin antagonists, trifluoperazine and R24571, depolarize the mitochondria within guinea pig cerebral cortical synaptosomes. J. Neurochem. 1552-1557. Sorokin N.S. (1982) Ground beetles Coleoptera carabidae as natural enemies of the Colorado beetle Lentino tar s a decemlineata. Biosis Previews Abstr. 24008054. 251 Spector I., Shochet N.R., Kashman Y., and Groweiss A. (1983) Latrunculins: novel marine toxins that disrupt microfilament organization in cultured cells. Science 219. 493-49 5. Spedding M. (1985) Calcium antagonist subgroups. Trends Pharmacol. Sci. 6_, 109-114. Spokes J.W. and Dolly J.O. (1980) Complete purification of -bungarotoxin. Characterization of its action and that of tityus toxin on synaptosomal accumulation and release of acetylcholine. Biochim. Biophys. Acta 596 . 81-93 . Strong P.N., Heuser J.E., Oberg S.G., and Kelly R.B. ( 1 97 8) j3”bungarotoxin : phospholipase activity is responsible for presynaptic action, p.436 Ln (Rosenberg P., ed) Toxins: Animal, Plant, and Microbial. Pergamon Press, New York. (ABSTR.). Sun Y.-A. and Poo M.-M. (1983) Non-quantal release of acetylcholine at embryonic neuromuscular synapses in culture. Soc. Neurosci. Abstr. 9., 688. Suszkiw J., Toth G ., Murawsky M . , and Cooper G.P. (1984) Effects of Pb 2 ^and Cd on acetylcholine release and Ca movements in synaptosomes and sub ce 1lular fractions from rat brain and Torpedo electric organ. Brain Res. 3 23. 31-46. Sut t orp N . , Seeger W. , Uhl J. , Lutz F. , and Roka L. ( 1 985) Ps end omona s a erug ino s a cy totoxin stimulates prostacyclin production in cultured pulmonary artery endothelial cells: membrane attack and calcium influx. J. Cell. Physiol. 123. 6 4-7 2. Symington J. (1984) Electrophoretic transfer of proteins from two-dimensional gels to sheets and their detection, pp. 128-168 in (Celis J.E. and Bravo R., eds) Two-Dimensional Gel Electrophoresis of Proteins. Academic Press, New York. Szerb J.C. and 0'Regan P.A. (1985) Effect of glutamine on glutamate release from hippocampal slices induced by high K or by electrical stimulation: interaction with different Ca concentrations. J. Neurochem. 44. 1724-1731. Takahashi M . , Ohizumi Y ., and Yasumoto T. (1982) Maitotoxin, a calcium channel activator candidate. 252 J. Biol. Chem. 257 . 7 287-7 289 . Tamaki G. and Butt B.A. (1978) Impact of Perillus bioculatus on the Colorado potato beetle and plant damage. Biosis Previews Abstr. 16051067. Tauc L. (1979) Are vesicles necessary for release of acetylcholine at cholinergic synapses? Biochem. Pharmacol. 2 7 , 3493—3498. Tauc L. (1982) Nonvesicular release of neurotransmitter. Physiol. Rev. 6 2 . 857-893 . Tauc L. and Baux G. (1982) Are there intracellular acetylcholine receptors in the cholinergic synaptic nerve terminals? J. Physiol. (Paris) 7 8 . 366-372. Tauc L. and Baux G. (1985) Mechanism of acetylcholine release at neuro-neuronal synapses, pp. 253-269 in (Ro thman S.S. and Ho J.J.L., eds) Nonvesicular Transport. John Wiley and Sons, New York. Tazieff-Depierre F ., Metezeau P., and Choucavy M . (1978) Pharmacological investigations on purified toxins isolated from the sea anemone Anemonia sulcata, pp. 557-564 _in (Rosenberg P., ed) Toxins: Animal, Plant, and Microbial. Pergamon Press, New York. Thesleff S. (1984) Transmitter release in botulinum-poisoned muscles. J. Physiol. (Paris) 79. 192-195. Thesleff S. and Mo1go J . (1983) A new type of transmitter release at the neuromuscular junction. Neuroscience 9., 1-8. Thief fry M . , Morel N ., and Manaranche R. ( 1984) Binding of Glycera convoluta neurotoxin to cholinergic nerve terminals triggers a Ca-dependent acetylcholine release. J. Physiol. (Paris) 7 9 . 269-274. Tobias G.S. (1982) /6-bungarotoxin inhibits synaptosomal calcium uptake. Soc. Neurosci. Abstr. Vol. 8_ p. 794. Tomono T., Suzuki T., and Tokunaga E. (1983) Analysis of peptic fragmentation of human immunoglobulin G using high-performance liquid chromatography, pp. 9-16 in (Hearn M.T.W., Regnier F.E., and Weir T.C., eds) High Performance Liquid Chromatography of 253 Proteins and Peptides. Academic Press, New York. Toth G. and Suszkiw J.B. (1985) Effects of AH5183 on ACh release from synaptosomes. Trans. Amer. Soc. Neurochem. 16. 163. (ABSTR.). Tsien R.Y. (1980) New calcium indicators and buffers with selectivity against magnesium and protons: Design, synthesis, and properties of prototype structures. Biochemistry 19. 23 96-2404. Tu A.T. (1977) Venoms. Chemistry and Molecular Biology. John Wiley and Sons, New York. Turner T.J. and Goldin S.M. (1985) Calcium channels in rat brain synaptosomes: identification and pharmacological characterization. J. Neurosci. 5_, 841-849. Tzeng M.-C. and Siekevitz P. ( 197 8) The effect of the purified major protein factor lactrotoxin) of black widow spider venom on the release of acetylcholine and norepinephrine from mouse cerebral cortex slices. Brain Res. 13 9 . 190-196. Tz eng M.-C., Cohen R.S., and Siekevitz P. ( 1978) Release of neurotransmitters and depletion of synaptic vesicles in cerebral cortex slices by oC-latrotoxin from black widow spider venom. Proc. Natl. Acad. Sci. 7_5, 4016-4020. Van der Kloot W. and^Latta R. (1983) How elevated extracellular Ca inhibits quantal acetylcholine release at frog neuromuscular junctions in high K . Pflugers Arch. 397. 85-89. Van Eldik L.J.and Watters on D.M, (1981) Reproducible production of antiserum against vertebrate calmodulin and determination of the immunoreactive site. J. Biol. Chem. 256. 420 5-4210 . Vickroy T.W., Roeske W.R., Gehlert D.R., Wamsley J.K., and Yamamura E.I. (1985) Quantitative light macroscopic autoradiography of I H ]hemicholinium-3 binding sites in the rat central nervous system: a novel biochemical marker for mapping the distribution of cholinergic nerve terminals. Brain Res. 3 2 9 . 368-373. Vizi E.S., Torok T., Sergei A., Serfozo P., and 254 Adam-Vizi V. (1982) Na-K-activated ATPase and the release of acetylcholine and noradrenaline. J. Physiol. (Paris) 7 8 . 3 99-406 . Volkov E.M. and Frosin V.N. (1984) Nonquantal acetylcholine release from motor nerve endings and denervation changes in rat muscle fiber membranes after axonal transport blockage. Neurophysiol. (Engl, transi.) 16. 192-196. Volkov E.M. and Poletaev G.I. (1983) Nonquantal acetylcholine release in the frog myoneural junction after disturbance of axoplasmic transport by colchicine. Bull. Exp. Biol. Med. 94, 13 29-1331. Von Wedel R.J., Carlson S.S., and Kelly, R.B. (1981) Transfer of synaptic vesicle antigens to the presynaptic plasma membrane during exocytosis. Proc. Natl. Acad. Sci. 7 8 . 1014-1018. Wagner J.A., Carlson S.S., and Kelly R.B. (1978) Chemical and physical characterization of cholinergic synaptic vesicles. Biochemistry 17. 1199-1206. Walker J.H., Stad1er H. , and Witzemann V. (1984) Calmodulin binding proteins of the cholinergic electromotor synapse; synaptosomes, synaptic vesicles, receptor—enriched membranes, and cytoske1eton. J. Neurochem. 42. 314-320. Waxman S.G. and Ritchie J.M. (1985) Organization of ion channels in the myelinated nerve fiber. Science 228. 1502-1507. Weiland G.A. and Molinoff P.B. (1981) Quantitative analysis of drug-receptor interactions; I. Determination of kinetic and equilibrium properties. Life Sci. Ü , 313-330 . Weiner S.A. and Collier B. (1984) Uptake, metabolism, and releasabi1ity of ethyl analogues of homocholine by rat brain. J. Neurochem. 43. 1143-1151. Wheeler D.D. (1984) Kinetics of D-aspartic acid release from rat cortical synaptosomes. Neurochem. Res. 1, 1599-1614. Wheeler K.P., Barhanin J., and Lazdunski M. (1982) Specific binding of toxin II from Centruro ides suffusus suffusus to the sodium channel in 255 electroplaque membranes. Biochem. 21. 5628-5634. White T.D. (1978) Release of ATP from a synaptosomal preparation by elevated extracellular K and by veratridine. J. Neurochem. 3 0 . 329-336 . Whittemore S.R., Graber S.G., Lenox R.H., Hendley E.D. and Ehrlich Y.H. (1984) Activation of adenylyl cyclase by preincubation of rat cerebral cortical membranes under phosphorylating conditions: role of ATP, GTP, and divalent cations. J. Neurochem. 4 2 . 1685-1696 . W i eke1gr en W.O., Leonard J.P., Grimes M.J., and Clark R.D. (1985) Ü 1trastructural correlates of transmitter release in presynaptic areas of lamprey reticulospinal axons. J. Neurosci. 5_, 1188-1201 . Wilson S.P. and Kirshner N. (1983) Calcium-evoked secretion from digitonin-permeabi1ized adrenal medullary chromaffin cells. J. Biol. Chem. 2 58. 4994-5000. Wonnacott S. (1980) Inhibition by botulinum toxin of acetylcholine release from synaptosomes: latency of action and the role of gangliosides. J. Neurochem. 34. 1567-1573. Wonnacott S. and Mar chbanks R.M. ( 1 976 ) Inhibition by bjtulinum toxin of depolarization-evoked release of [ C] acetylcholine from synaptosomes in vitro. Biochem. Ji 156. 701-712. Wu P.E., Moron M . , and Barraco R. (198^) Organic calcium channel blockers enhance [ E ]pur ine release from rat brain cortical synaptosomes. Neurochem. Res. 9., 1019-1031. Yoshino J.E. (1980) Purification and Characterization of Leptinotarsin-d, a Neurotoxin Which Stimulates the Release of Neuro transmitters from Synaptosomes. PhD Thesis. University of Southern Calif., Los Angeles. 168 pp. Young G.P.E., Young J.D.-E., Deshpande A.K., Goldstein M.j^Koide S.S., and Cohn Z.A. ( 1984) A Ca -activated channel from Xenonus 1aevis oocyte membranes reconstitued into planar bilayers. Proc. Natl. Acad. Sci. 8J_, 51 55-51 59 . Zalman L.S. and Wisnieski B.J. (1984) Mechanism of 256 insertion of diphtheria toxin: Peptide entry and pore size determinations. Proc. Natl. Acad. Sci. 81, 3341-3345. Zimmerberg J. and Whitaker M. (1985) Irreversible swelling of secretory granules during exocytosis caused by calcium. Nature 315. 581—584. Zimmermann H. and Whittaker V.P. (1974) Effect of electrical stimulation on the field and composition of synaptic vesicles of the electric organ of Torpedo: a combined biochemical, electrophysiological, and morphological study. J. Neurochem. 2 2 . 435-450 . Zimmermann H. and Whittaker V.P. (1977) Morphological and biochemical heterogeneity of cholinergic synaptic vesicles. Nature 26 7 . 633-635. 257 UMI Number: DP23751 All rights reserved INFORMATION TO ALL USERS The quality of this reproduction is dependent upon the quality of the copy submitted. In the unlikely event that the author did not send a complete manuscript and there are missing pages, these will be noted. Also, if material had to be removed, a note will indicate the deletion. Dissertation RjMisMng UMI DP23751 Published by ProQuest LLC (2014). Copyright in the Dissertation held by the Author. Microform Edition © ProQuest LLC. All rights reserved. This work is protected against unauthorized copying under Title 17, United States Code ProQuest LLC. 789 East Eisenhower Parkway P.O. Box 1346 Ann Arbor, Ml 48106- 1346

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

The role of microbes in the formation of the tubeworm fouling community in Fish Harbor, Los Angeles Harbor, California

PDF

Studies on the cytology of bacteria and yeasts by means of ultraviolet photolysis

PDF

Studies on the bacterial oxidation of saturated fatty acids and their derivatives

PDF

Numerical analysis of ecological survey data

PDF

Studies on the aerobic oxidation of fatty acids by bacteria

PDF

Nude mouse grown human nephroblastomas

PDF

Purification and characterization of leptinotarsin-d, a neurotoxin which stimulates the release of neurotransmitters from synaptosomes

PDF

Chemical and immunological studies on a heterophile antigen extracted from bovine erythrocyte stroma

PDF

Studies of frog retinal ganglion cell responses to temporal, spatial and chromatic stimuli

PDF

Light absorption in naturally occurring marine phytoplankton communities

PDF

An analysis of traditional birth attendant (TBA) programs in selected third world countries

PDF

Morphological consequences of sympatry in two species of 'Typosyllis' (Annelida: Polychaeta: Syllidae)

PDF

Photophysiology and cellular composition of sea ice algae

PDF

The innervation of the teeth and periodontium of the rat

PDF

The interaction of learners' spatial ability aptitudes with filmic schematic operations

PDF

The 'Widmar' (1981) decision and the status of religious liberty on the public college and university campus: Antecedents, effects, and prospects

The effects of testosterone on sexually dimorphic muscles of male frogs (<italic>Xenopus laevis</italic>)

PDF

The effects of testosterone on sexually dimorphic muscles of male frogs (Xenopus laevis)

PDF

Out of many...: A social history of the homosexual rights movement as originated and continued in Los Angeles, California

PDF

Studies on cholesterol metabolism in the rat

PDF

Cation localization in growth hormone and prolactin cells of the rat pars distalis

Asset Metadata

Creator Koenig, Michael Leo (author)

Core Title Studies of two naturally occurring compounds which effect release of acetylcholine from synaptosomes

School Graduate School

Degree Doctor of Philosophy

Degree Program Biology

Degree Conferral Date 1985-10

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag biological sciences,OAI-PMH Harvest

Language English

Contributor Digitized by ProQuest (provenance)

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c30-237264

Unique identifier UC11228106

Identifier DP23751.pdf (filename),usctheses-c30-237264 (legacy record id)

Legacy Identifier DP23751.pdf

Dmrecord 237264

Document Type Dissertation

Rights Koenig, Michael Leo

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Access Conditions The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the au...

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus, Los Angeles, California 90089, USA