| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1Department of Physiology, University of Toronto, Toronto, Ontario M5S 1A8, Canada; 2Instituto Butantan, Laboratory of Pharmacology, Biochemistry and Immunochemistry, São Paulo, SP, 05503-900 Brazil; and 3Samuel Lunenfeld Research Institute, Mount Sinai Hospital, Toronto, Ontario M5G 1X5, Canada
Submitted 19 February 2003; accepted in final form 25 July 2003
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMETHODSRESULTSDISCUSSIONDISCLOSURESACKNOWLEDGMENTSREFERENCES |
|---|
-phonetoxin-IIA (-Ptx-IIA, 8360 MW), which blocks Ca2+ channels. At frog neuromuscular junctions (NMJ) bathed in normal Ca2+ (1.8 mM) saline, -Ptx IIA did not affect spontaneous transmitter release but reversibly reduced evoked transmitter release by 75 and 95% at 12 and 24 nM, respectively. In contrast, toxin effects were irreversible in low-Ca2+ (0.5 mM) saline. Ca2+ imaging in normal-Ca2+ saline showed that -Ptx-IIA partially blocked stimulus-dependent presynaptic Ca2+ signals, and the blockade was almost completely reversible. Increases in spontaneous release frequency induced by high extracellular K+ were blocked by -Ptx-IIA. Therefore -Ptx-IIA blocks N-type Ca2+ channels, which admit Ca2+ that triggers transmitter release at the frog NMJ. Additional evidence predicts that -Ptx-IIA binds to N-type Ca2+ channels at a different site from that of -Conotoxin-GVIA. -Ptx-IIA also gave a low-affinity partial blockade of transmitter release and presynaptic Ca2+ signals at crayfish NMJs where P-type channels are blocked by -agatoxin-IVA. The Ca2+-dependent reversibility and promiscuity of this toxin may make it highly useful experimentally and therapeutically. |
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONDISCLOSURESACKNOWLEDGMENTSREFERENCES |
|---|
-conotoxin-GVIA (-CgTx-GVIA), a toxin from the marine snail Conus geographus, and P-type channels are inhibited by -agatoxin-IVA (-Aga-IVA), a toxin from the funnel-web spider Agelenopsis aperta. These -toxins bind irreversibly or are slowly reversible (see review in Olivera et al. 1994
). These toxins block transmitter release at the frog neuromuscular junction (NMJ) and at the mammalian NMJ, respectively, where N- and P-type channels admit Ca2+ to trigger transmitter release (see review in Meir et al. 1999). Because they have affinity for distinct types of Ca2+ channels, each toxin alone cannot eliminate Ca2+ influx at synapses where there are multiple types of Ca2+ channels. For instance, although one of either N- or P/Q-type Ca2+ channels usually contributes at each type of nerve terminal, it is known that many nerve terminals possess more than one type of Ca2+ channel (for reviews, see Fisher and Bourque 2001; Meir et al. 1999). In the study of Ca2+-dependent processes, it would thus be advantageous to employ a nonselective, as well as reversible and impermeant, channel blocker. This study describes physiological effects on transmitter release of such a Ca2+ channel blocker obtained from the venom of the South American armed spider, Phoneutria nigriventer.
The venom of P. nigriventer contains many peptides with distinct pharmacological actions that can block Na+, K+, and Ca2+ channels (Araujo et al. 1993; Figueiredo et al. 1995; and reviewed in Gomez et al. 2002). A fraction of the venom called PF3 blocked brain dopamine release (Troncone et al. 1995) and blocked acetylcholine (ACh) release at the mouse NMJ (Souccar et al. 1995). The effect of PF3 at mouse NMJs was antagonized by increased concentration of extracellular Ca2+, and it was hypothesized that PF3 blocks presynaptic Ca2+ channels. Two other toxins, which have been isolated from the same spider venom and are likely to be the same as PF3, are known as PhTx3–4 and -phonetoxin-IIA (-Ptx-IIA). Henceforth we will refer to PF3 as -Ptx-IIA. PhTx3–4 blocks ACh release and PhTx3–4/-Ptx-IIA blocks N-type Ca2+ channels in rat dorsal root ganglion neurons and L-type channels in rat pancreatic 
cells (Cassola et al. 1998; Cordeiro et al. 1993; Gomez et al. 1995). While -Ptx-IIA blocks Ca2+ channels in neurons, the mechanism of blockade of synaptic transmission is unknown.
The present study characterizes the mechanism of -Ptx-IIA toxin action at intact synapses. We demonstrate that -Ptx-IIA partially blocks stimulus-dependent Ca2+ signals in presynaptic terminals containing either N- or P-type Ca2+ channels. At the frog NMJ, the block of presynaptic N-type channels by -Ptx-IIA was quickly reversible when the physiological saline contained 1.8 mM Ca2+, but was practically irreversible under low extracellular Ca2+ (0.5 mM) conditions. -Ptx-IIA did not block binding of -CgTx-GVIA. Finally, we show that -Ptx-IIA also blocks presynaptic P-type channels in an arthropod.
|
METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONDISCLOSURESACKNOWLEDGMENTSREFERENCES |
|---|
Purification of the -Ptx-IIA toxin originally called PF3 is described elsewhere (Troncone et al. 1995). Briefly, spider venom was obtained in the Laboratory of Arthropods of the Butantan Institute (São Paulo, Brazil) by electric milking, desiccated, resuspended in 2% (vol/vol) acetic acid, centrifuged to remove solids and applied to a Sephadex G50f chromatographic column. Fractions carrying the characteristic flaccid paralysis activity were pooled and lyophilized, resuspended in water and submitted to RP-HPLC using a µC18 Boundapack column with linear gradient of trifluoroacetic acid (0.1% in water, vol/vol) and acetonitrile (90% in phosphoric acid, vol/vol). The gradient ran from 5–35% of acetonitrile in 20 min. The -Ptx-IIA/PF3 toxin obtained showed only one chromatographic peak and one amino-terminal amino acid. Further chemical identification by mass spectrometry resulted in the characteristic values of m/z of 1394, 1195, and 1046 corresponding to 6–8 protons, respectively, compatible with a molecular weight of 8,360. Cassola et al. (1998) described the full sequence of this toxin and called it -Ptx-IIA. The primary sequence of -Ptx-IIA is identical to PF3 (amino acids 15 and 19 were mistyped in Troncone et al. 1995). Furthermore, electron-spray mass spectrometry revealed the same molecular mass obtained by Cassola et al. (1998), and we now refer to PF3 as -Ptx-IIA. While -Ptx-IIA and -Aga-IIIA have 49% amino acid identity, -Ptx-IIA has 14 cysteines, probably forming seven disulfide bridges whereas -Aga-IIIA has only six disulfide bridges (Cassola et al. 1998; Venema et al. 1992). Another toxin that is probably identical to -Ptx-IIA is called Tx3–4 (Cordeiro et al. 1993; see Cassola et al. 1998). For information on availability of -Ptx-IIA please contact L.R.P. Troncone.
Electrophysiology
All frog (Rana pipiens) experiments were performed on the cutaneus pectoris nerve-muscle preparation. Animals were maintained and killed (by double pithing) in accordance with the recommendations outlined in the Canadian Council on Animal Care's Guide to the Care and Use of Experimental Animals (www.ccac.ca). Electrophysiology was conducted in normal Ringer solution (normal saline) containing (in mM) 120 NaCl, 2 KCl, 1.8 CaCl2, 1 NaHCO3, and 5 HEPES. In some experiments, a low-Ca2+ saline was used that contained 0.5 mM CaCl2 and 5 mM MgCl2. The pH was adjusted to 7.2 with NaOH. The muscle-selective Na+ channel blocker µ-conotoxin-GIIIA (10 µg/ml; RBI or Bachem) was sometimes used to avoid contractions in the normal (1.8 mM)-Ca2+ saline (Robitaille et al. 1993). Transmitter release was assayed by measuring endplate potentials (EPPs) evoked by nerve stimulation (twin pulses at 100 Hz every 5 s). Standard electrophysiological techniques using intracellular microelectrodes (3 M KCl, resistance: 10–20 M
) were employed to record electrical responses in individual muscle cells caused by transmitter release. Spontaneous transmitter release was estimated by measuring miniature EPPs (MEPPs). Responses were acquired by a Labmaster DMA interface and averaged in groups of 6 or 10 by TOMAHACQ, a PC-based data-acquisition program. MEPPs were recorded onto tape using a VR10B digitizer (Instrutech) and later analyzed using ANMEPP software. Both programs were written by T. A. Goldthorpe (Univ. of Toronto). The dose dependency of -Ptx-IIA blockade was determined by pooling together data from separate experiments; each increase in toxin concentration was applied long enough to reach equilibrium as established by on-line graphing features in the acquisition software.
The compound EPP (CEPP) from frog muscle was recorded via two platinum recording wires at the bottom of the physiological chamber. The dorsal side of the muscle was oriented directly over the recording wires. These electrodes were connected to a differential AC amplifier (Grass P15) and the CEPP signal was amplified and further processed by computer. After initially setting up the preparation in continuous saline flow (1 pass, 0.5 ml/min), flow was stopped and the bath saline was recirculated with a peristaltic pump. This assured constant mixing of solution without disturbing microelectrode recordings as the toxins were added directly to the bath.
Evoked transmitter release from crayfish (Procambarus clarkii) NMJs was measured as excitatory junction potential (EJP) amplitude (Hua et al. 1998). Experiments were conducted on the opener muscle of the first walking leg of small (5 cm) crayfish (Wojtowicz and Atwood 1984). This muscle is innervated by a single excitatory axon and a single inhibitory axon, the former of which was separated from the rest of the nerve bundle at the meropodite segment. All other nerves were cut to ensure that only the selected axon was stimulated. Crayfish saline contained (in mM) 205 NaCl, 5.4 KCl, 13.5 CaCl2, 2.7 MgCl2, 10 D-glucose, and 10 HEPES (pH adjusted to 7.4 with NaOH). In low-Ca2+ (2.0 mM) crayfish saline, equinormal NaCl was substituted for CaCl2, and when applied, the bathing saline volume was exchanged several times to ensure equilibration. In recordings of spontaneous miniature EJPs (MEJPs), 10 µM picrotoxin (Sigma) was added to block GABA receptor-mediated miniature inhibitory potentials (Takeuchi and Takeuchi 1969).
Nerve terminal Ca2+ imaging
Excised frog nerve-muscle preparations were pinned into specially adapted 25-mm petri dishes containing two wells, one 15-mm square and the other 4-mm square, both 4 mm deep. The wells were formed by two plastic blocks inserted into the uncured silicone elastomer (Sylgard; Dow Corning) poured into a petri dish. After the Sylgard cured, the plastic blocks were removed thus leaving the two wells. The muscle was pinned into the larger well containing frog Ringer solution, and the attached cut end of the nerve was submerged in the adjacent well containing a 5 mM solution of fluorescent Ca2+ indicator Ca2+ Green-1 dextran 3000 MW (Kd 260 nM; Molecular Probes, Eugene, OR). The end of the nerve was cut again while submerged in this solution. The petri dish was covered and nerve terminals were allowed to load with fluorescent indicator overnight (14 h at 15°C) by forward (anterograde) filling. Frog muscle contractions were eliminated using the acetylcholine receptor blocker 
-bungarotoxin (5 µg/ml). For crayfish preparations, Ca2+ Green-1 dextran was microinjected directly into excitatory axons and allowed to diffuse into synaptic boutons (Hua et al. 1998). Ca2+ responses were obtained over a range of stimulation frequencies (20, 40, and 60 Hz for frog NMJs and 5, 10, 20, and 30 Hz for crayfish NMJs), and within these frequency ranges the response size increased with frequency. Thus our conclusions can be made from the lower frequency data with the assurance that the indicator was not saturated.
Nerve terminals were examined with a water immersion objective (x40, 0.55 NA, Nikon) on a Biorad 600 confocal laser-scanning microscope configured to deliver 488 nm wavelength excitation and to detect emission >500 nm. Photomultiplier gain and black level settings were held constant between repetitions. Fluorescence intensity was averaged for the area of interest, and Ca2+ signals were expressed as a relative change in fluorescence intensity using the formula 
F/F = (Fstimulation - Frest)/Frest.
Data analysis
Data are presented as means ± SE. Normalization of data were based on baseline measurements obtained over a 5- to 10-min period prior to start of each experiment. Statistical significance was evaluated using Student's t-test (P < 0.05).
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONDISCLOSURESACKNOWLEDGMENTSREFERENCES |
|---|
-Ptx-IIA blocks neurotransmission at frog synapses
Presynaptic N-type Ca2+ channels are responsible for transmitter release at frog NMJs (Robitaille et al. 1990). To determine whether Phoneutria toxin -Ptx-IIA acts at frog NMJs, we performed electrophysiological recordings on excised cutaneus pectoris nerve-muscle preparations. -Ptx-IIA did not affect the muscle resting membrane potential, which remained between -70 and -90 mV. -Ptx-IIA blocked the EPP in a dose-dependent fashion, with an apparent EC50 of 
11 nM (Fig. 1A; n = 5–7). When -Ptx-IIA was applied at concentrations >30 nM, responses were vastly reduced within 5 min or were eliminated altogether. High-frequency nerve stimulation (100 Hz) failed to reverse the blockade (data not shown).
|
|
|
-Ptx-IIA in normal saline (containing 1.8 mM Ca2+ and µ-conotoxin-GIIIA to block muscle contractions) eliminated EPPs; on subsequent wash with fresh saline lacking µ-conotoxin-GIIIA, EPPs reappeared and when responses reached a level (>5 mV) sufficient to generate muscle contractions, recordings were lost (data not shown). We next performed similar experiments in low-Ca2+ (0.5 mM) saline, a condition in which transmitter release is reduced and in which muscle contractions do not occur. Figure 1B shows data from a typical experiment performed in low-Ca2+ saline in which EPPs were largely eliminated within a few minutes of -Ptx-IIA addition (120 nM) but the blockade did not reverse after washing with fresh low-Ca2+ saline (n = 5). In contrast, the blockade was relieved after a wash with normal saline (1.8 mM Ca2+). Similar results were obtained in three experiments and in similar experiments using extracellular recordings of the CEPP (Fig. 4A and Table 2). Thus at frog neuromuscular synapses, -Ptx-IIA is a high-affinity blocker of synaptic transmission, the actions of which can be reversed by washing with saline containing sufficient (1.8 mM) Ca2+.
-Ptx-IIA acts at the presynaptic nerve terminal
To help determine the site of -Ptx-IIA action, we studied its effect on spontaneous neurotransmitter release. When spontaneous MEPPs were recorded in normal saline for 30 min before and after addition of -Ptx-IIA (120 nM), there was no change in MEPP frequency or amplitude (Fig. 2 and data not shown). The lack of effect of -Ptx-IIA on MEPPs is similar to that observed with -CgTx-GVIA (personal observations) and suggests that -Ptx-IIA does not affect the postsynaptic muscle cells and does not by itself alter intracellular [Ca2+]. This observation contrasts with previous work on mouse NMJs by Souccar et al. (1995) showing that -Ptx-IIA reduced both amplitude and frequency of MEPPs. The difference in the effect on MEPP frequency may be due to differences in the membrane potential of the presynaptic terminals as discussed in Hua et al. (1998). To test further the hypothesis that -Ptx-IIA acts presynaptically, we examined its ability to interfere with nerve depolarization-induced increases in MEPP frequency. MEPP frequency was increased by the introduction to the saline of 8 mM KCl (with equimolar reduction in NaCl to maintain osmolarity) for 30 min and -Ptx-IIA was subsequently added to the bath for the same period. When K+ concentration was increased, muscle cells were depolarized by 30 mV and MEPP frequency increased nearly fivefold; subsequent addition of -Ptx-IIA reduced MEPP frequency to almost that observed originally in normal saline (Fig. 2; n = 3 for each comparison). The data suggest -Ptx-IIA blocks presynaptic voltage-dependent Ca2+ channels that were opened by the high extracellular K+-induced presynaptic depolarization.
|
Nerve terminal Ca2+ signals are attenuated by -Ptx-IIA
To seek direct evidence of blockade of presynaptic Ca2+ influx, we imaged stimulus-dependent Ca2+ signals in nerve terminals. The Ca2+-sensitive fluorescent dye Calcium Green-1 dextran was introduced directly into nerve terminals by anterograde loading through the cut end of the nerve. This method was superior to introduction of cell-permeant fluorescent dyes in that it offered clear and sensitive monitoring of Ca2+ in nerve terminals alone; the indicator does not appear in other cells such as perisynaptic glia and muscle cells which also take up the permeant indicators (Jahromi et al. 1992). Figure 3A (left) shows Calcium Green-1 fluorescence in a typical nerve terminal under resting conditions and during 60-Hz nerve stimulation; the graph shows the time course of the Ca2+ signal obtained with 20-, 40-, and 60-Hz stimulation (10 s each). After addition of -Ptx-IIA (120 nM), the stimulation-evoked responses in the same terminal were greatly reduced (Fig. 3A, right). Incomplete blockade of the Ca2+ signal is not due to insufficient -Ptx-IIA because higher concentrations also failed to eliminate the signal. The remaining signal was not further reduced by subsequent application of -CgTx-GVIA and could therefore be due to another channel type not blocked by either toxin.
|
|
-Ptx-IIA (120 nM) attenuated the Ca2+ responses, and the remaining signal was nearly eliminated on switching to zero Ca2+ saline (containing 120 nM -Ptx-IIA). In the same -Ptx-IIA-treated nerve terminals, the fluorescence signal was then restored to pretoxin control levels after washing with normal saline (Fig. 3B, right). A summary of the Ca2+ signal data appears in Table 1. We conclude from these data that -Ptx-IIA blocks neurotransmission by reversibly interfering with Ca2+ influx at presynaptic nerve terminals.
Do -Ptx-IIA and -CgTx-GVIA bind to the same site?
To ask whether -Ptx-IIA and -CgTx-GVIA occupy the same site on the Ca2+ channel, we performed a competition experiment. The experimental design relied on the differential recovery of the two toxins, namely that the blockade by -Ptx-IIA, but not that of -CgTx-GVIA, can be washed off in normal saline. If both toxins occupy the same site, then -Ptx-IIA should reduce the ability of -CgTx-GVIA to block neurotransmission irreversibly. For these experiments, we used extracellular recordings of the CEPP because this is more suitable for long experiments than intracellular recordings. The CEPP is an average response from many endplates. First, a muscle was exposed to a very high concentration of -Ptx-IIA (1.2 µM) for 30 min; this is longer than necessary to achieve a complete blockade of the CEPP. Then, an equimolar concentration of -CgTx-GVIA was added to the bath, and both toxins were allowed to compete for 2 h. Subsequently, the muscle was washed for 
2 h with a constant flow of normal saline. Nerve stimulation was maintained during the experiment at 5-s intervals. A control muscle was incubated in the same conditions but only -Ptx-IIA was added, and a second control muscle was incubated with only -CgTx-GVIA. These initial experiments were performed in normal (1.8 mM Ca2+) saline (Fig. 4A). Application of -Ptx-IIA alone produced a blockade of synaptic responses that recovered on washout. In contrast, addition of -CgTx-GVIA alone blocked responses irreversibly. When -CgTx-GVIA was applied after -Ptx-IIA, there was no additional decrease in the amplitude of the CEPP, and there was no recovery after prolonged washing (see Table 2, 2nd column, for summary). This result indicates that -CgTx-GVIA was able to bind to channels in the presence of -Ptx-IIA as if the two toxins do not share the same binding site. However, the fact that effects of -Ptx-IIA are readily reversible in normal saline (1.8 mM Ca2+) raises the possibility that -Ptx-IIA constantly binds and unbinds under these conditions. If -Ptx-IIA and -CgTx-GVIA share a binding site, then -CgTx-GVIA may have the opportunity to bind irreversibly every time -Ptx-IIA unbinds.
To examine binding sites in more detail, we performed similar experiments in low-Ca2+ saline conditions, in which -Ptx-IIA blocking effects are irreversible. In these conditions, the ability of -CgTx-GVIA to find a binding site vacated by -Ptx-IIA would be vastly decreased. In these experiments, there was no recovery of synaptic responses after sequential application of both toxins and washing with low-Ca2+ saline (Fig. 4B; also see Table 2, 3rd column, summary). However, when washed with normal saline, which displaces the -Ptx-IIA, there again was little recovery of synaptic responses. This result indicates that -CgTx-GVIA was able to bind under conditions that prohibit unbinding of -Ptx-IIA, and therefore it is likely that the two toxins bind to different sites.
Effects of -Ptx-IIA at crayfish synapses
We next tested the effects of -Ptx-IIA at crayfish NMJs where neurotransmitter release is blocked by -agatoxin-IVA (Araque et al. 1994; Wright et al. 1996) and is likely to depend on P-type Ca2+ channels. In normal crayfish saline (13.5 mM Ca2+), a 0.5 µM concentration of -Ptx-IIA reduced synaptic responses to 35% of control, but EJP amplitude was drastically reduced when the toxin was applied at 1.2 µM (Fig. 5A, n = 8). When applied at 1.2 µM, 50-Hz stimulation could not restore neurotransmission (Fig. 5A), but the effect of -Ptx-IIA was easily reversible by washing in toxin-free saline for 5 min (data not shown). Additional observations indicated that -Ptx-IIA did not affect the presynaptic action potentials or the amplitude of spontaneous miniature EJPs recorded from muscle (n = 9, 3 respectively, data not shown). These findings reveal that -Ptx-IIA works presynaptically in crayfish as it does at frog nerve terminals.
|
Although a higher concentration of toxin was required to block neurotransmission at crayfish synapses than at frog NMJs, the lower sensitivity could be due to the presence of P-type Ca2+ channels instead of N-type channels or could also be due to the higher Ca2+ concentration (13.5 mM) in the crayfish saline. To examine this further, we next constructed a dose-response curve in normal crayfish saline and in saline that contained only 2.0 mM Ca2+. This analysis showed that the EC50 was 120 nM in 2.0 mM Ca2+ saline and 360 nM in 13.5 mM Ca2+ saline (Fig. 5B). The former value is 10-fold greater than that observed at frog NMJs in similar extracellular Ca2+ conditions. Thus the lower sensitivity of crayfish synapses to -Ptx-IIA may be due to the presence of P-type Ca2+ channels instead of N-type channels at active zones.
To establish the mechanism of action of -Ptx-IIA at crayfish NMJs, we performed Ca2+ imaging experiments (Hua et al. 1998). The fluorescent indicator Calcium Green-1-dextran was introduced by microinjection directly into crayfish axons and allowed to accumulate at synaptic boutons (Fig. 6A). The increase in Calcium Green-1 fluorescence was monitored during short trains of stimuli at 5, 10, 20, and 30 Hz. Simultaneous electrophysiological recordings revealed that at all frequencies tested, the EJP was completely blocked during the tetani (Fig. 6A). The graph in Fig. 6B shows that -Ptx-IIA (1.2 µM) reduced Ca2+ signals at 5 Hz but had progressively smaller effects as frequency was increased to 30 Hz (data shown represents mean of 6 experiments). The Ca2+ signals were completely blocked when preparations were bathed in crayfish saline containing zero Ca2+ and 1 mM EGTA (n = 5). Therefore while -Ptx-IIA does block presynaptic Ca2+ signals, this block is incomplete and is frequency sensitive.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONDISCLOSURESACKNOWLEDGMENTSREFERENCES |
|---|
-Ptx-IIA blocks neurotransmitter release by blocking stimulus-dependent Ca2+ entry in intact presynaptic terminals. A direct measurement of intracellular Ca2+ by fluorescent imaging showed great reduction in stimulus-dependent signals in presynaptic terminals. Furthermore, -Ptx-IIA did not block presynaptic action potentials because after application of -Ptx-IIA to block transmitter release, these were unaffected in crayfish NMJs and partial Ca2+ signals were still detected in frog terminals. Because -Ptx-IIA did not alter the amplitude of spontaneous synaptic potentials, it does not block postsynaptic ACh (frog NMJ) or glutamate (crayfish NMJ) receptors. There could be other effects of this toxin on Ca2+ signaling that could not be detected with our assays.
Previous work showed that -Ptx-IIA is effective in mammals. For instance intraperitoneal injection of 3 µg of -Ptx-IIA in 25 g mice causes flaccid paralysis and death (Rezende et al. 1991). At mouse phrenic-hemidiaphragm NMJ (Souccar et al. 1995) and rat brain striatal tissue (Troncone et al. 1995), -Ptx-IIA caused strong blockade of neurotransmitter release. Both of these effects are likely to be due to blockade of P-type Ca2+ channels that normally cause transmitter release at the mammalian NMJ (Katz et al. 1997; Meir et al. 1999; Protti et al. 1996; Sugiura et al. 1995) as well as in brain tissue (Carvalho et al. 1995; Kimura et al. 1995). In addition to N- and P-type channels, -Ptx-IIA blocks L-type Ca2+ channels expressed in cultured undifferentiated PC12 cells (Wolff et al. 1997) and in rat dorsal root ganglion cells (Cassola et al. 1998). Although -Ptx-IIA/Tx3–4 partially blocked 45Ca2+ influx in synaptosomes depolarized by Tityus scorpion toxin (Miranda et al. 2001), it is not clear that this influx is connected to Ca2+ channels that trigger exocytosis or that would be opened by action potentials.
Reversible action of -Ptx-IIA at frog neuromuscular synapses
To characterize the potency of -Ptx-IIA at frog NMJs, a dose-response relationship was constructed by adding increasing amounts of -Ptx-IIA to normal saline. An approximate EC50 of 11 nM was obtained (Fig. 1A). Similar EC50 values for -CgTx-GVIA at frog NMJ have been obtained by others (Katz et al. 1995; Sano et al. 1987; Zengel et al. 1993). Thus -Ptx-IIA is a high-affinity blocker of N-type Ca2+ channels. -Ptx-IIA did not alter spontaneous release at frog NMJs but blocked most of the fourfold increase in frequency induced by 8 mM K+. This observation suggests that -Ptx-IIA blocks Ca2+ channels opened by depolarization but is not displaced by depolarization.
The blockade of transmitter release by -Ptx-IIA was partially reversed after a few minutes of washing in normal Ca2+ saline but was irreversible in low-Ca2+ saline. Owing to the onset of contractions we could not follow recovery in normal-Ca2+ saline to quantitate the maximal extent of recovery. However, using Ca2+ imaging and -bungarotoxin to block muscle contraction, we found that there was 81–97% recovery of the Ca2+ signal when -Ptx-IIA was washed off with normal-Ca2+ saline. Therefore -Ptx-IIA has reversible action at the frog NMJ.
Toxins reveal multiple binding sites on N-type Ca2+ channels
Application of -Ptx-IIA in low-Ca2+ saline conditions, in which its effects were irreversible, did not prevent the irreversible binding of -CgTx-GVIA; reversal of blockade, as expected for -Ptx-IIA alone (Fig. 4), did not occur when -CgTx-GVIA was also applied. This indicates that binding of -Ptx-IIA cannot prevent binding of -CgTx-GVIA and that therefore the two toxins may bind to different sites on the channel. On the other hand, Dos Santos et al. (2002) showed that -Ptx-IIA can totally occlude binding of -CgTx-GVIA to recombinant Ca2+ channels expressed in baby hamster kidney cells. The comparison of binding studies on recombinant channels with our physiological studies is difficult and may be complicated by differences between channels in different animals.
Because the amino acid sequences of -Ptx-IIA and -Aga-IIIA are similar, it may be useful to compare them in regard to binding-site specificity. In binding studies with radiolabeled toxins and rat brain membranes, Adams et al. (1993) demonstrated that -Aga-IIIA occludes the binding of -CgTx-GVIA and -conotoxin-MVIIC but not -Aga-IVA. Furthermore McDonough et al. (2002) showed that -Aga-IIIA blocked the ability of -CgTx-GVIA to block channels in rat sympathetic neurons that otherwise would have been blocked; the authors suggested that -CgTx-GVIA and -Aga-IIIA bind to a single site on N-type channels to block the channel pore (McDonough et al. 2002). Therefore our results with -Ptx-IIA and -CgTx-GVIA competing for the N-type voltage-operated calcium channel at frog NMJ suggest differences between -Ptx-IIA and -Aga-IIIA or differences between rat and frog N-type Ca2+ channels. It must also be remembered that our assay was for evoked transmitter release and not binding. Owing to nonlinearities between Ca2+ channel function and transmitter release, comparisons of these assays are difficult (see review in Meir et al. 1999). The unbinding of -Ptx-IIA from frog N-type channels was highly Ca2+ dependent. Effects of extracellular divalent ions such as Ca2+ on toxin binding and unbinding have been noted for several toxins including -CgTx-GVIA (Liang and Elmslie 2002; Wagner et al. 1988; Witcher et al. 1993).
Action at crayfish synapses
The EC50 for -Ptx-IIA at crayfish NMJ was 120 nM in 2.0 mM Ca2+ saline and 360 nM in 13.5 mM Ca2+ saline (Fig. 5B). Araque et al. (1994) found 80% blockade of transmitter release at crayfish NMJs with 300 nM -Aga-IVA while Wright et al. (1996) found 100% blockade of presynaptic Ca2+ current and transmitter release by 100 nM -Aga-IVA. It thus appears that -Ptx-IIA is a somewhat weaker blocker of cray-fish presynaptic Ca2+ channels than is -Aga-IVA. -Aga-IIIA, which is structurally similar to -Ptx-IIA, does not affect transmission at NMJs of houseflies but does block Ca2+ channels in thoracic ganglion neurons of locusts (Bindokas et al. 1991; Ertel et al. 1994; Pocock et al. 1992). Therefore both toxins can block some Ca2+ channels in arthropods.
Calcium signals and multiple channel types
The blockade of presynaptic Ca2+ channels by -Ptx-IIA was demonstrated by Ca2+ imaging. While the blockade of evoked transmitter release at the frog NMJ was complete, the blockade of stimulation-induced fluorescence induced by -Ptx-IIA was incomplete (Fig. 3). Similar results were obtained with -CgTx-GVIA; when this toxin completely blocked transmitter release the Ca2+-triggered fluorescence signal was still significant (unpublished observations). In our previous studies (Robitaille et al. 1993) using the cell-permeant indicator fluo3-AM, the Ca2+ signal was blocked when transmitter release was blocked. This difference may be due to our use of anterogradely loaded Calcium-Green-1-dextran, which has a higher affinity (260 nM) than the fluo3 (390 nM) employed in the earlier study; this would have facilitated the detection of the remaining small Ca2+ signals in the present study.
Katz et al. (1995) tested the effects of three different toxins on neurotransmission at frog NMJs. The funnel-web spider toxin FTX (a polyamine), and -CgTx-GVIA blocked evoked transmitter release but -Aga-IVA had no effect. The effectiveness of FTX to block transmitter release and to block partially perineural Ca2+ currents at the frog NMJ (Katz et al. 1995) indicates that there is a small proportion of Ca2+ channel types in addition to N channels at these synapses (see also Thaler et al. 2001). In our experiments, the persistence of Ca2+ signals when evoked transmitter release had been blocked by -Ptx-IIA or -CgTx-GVIA indicates by a different method that these presynaptic terminals may possess multiple channel types. However, it is also possible that nonlinearities in the relation between active zone Ca2+ concentration and transmitter release combined with the low affinity of the Ca2+ sensor for transmitter release and partial channel block in normal Ca2+ saline could explain these observations. Similarly, at the crayfish NMJ, we found that presynaptic Ca2+ signals persisted when transmitter release had been blocked by -Ptx IIA. Rathmayer et al. (2002) found that there are at least two types of channels sensitive to -Aga-IVA and -CgTx-GVIA that contribute to evoked transmitter release at crab NMJs. Similarly at crayfish motorneuron cell bodies, Hong and Lnenicka (1997) found that when Ba2+ replaced extracellular Ca2+, -Aga-IVA blocked only 33% of the current, but the remainder was not blocked by -CgTx-GVIA or nifedipine. However, at presynaptic terminals of crayfish neuromuscular junctions, Wright et al. (1996) found no effect of -CgTx-GVIA on presynaptic Ca2+ current or transmitter release but total blockade with 100 nM -Aga-IVA. Therefore participation of multiple channel types at crayfish NMJs remains somewhat controversial.
Summary
The work here is the first direct demonstration that -Ptx-IIA blocks presynaptic Ca2+ channels associated with triggered exocytosis. The results presented here indicate that in addition to its known effects in mammals, -Ptx-IIA is also active in lower vertebrates and arthropods. Therefore -Ptx-IIA is a broad-spectrum reversible Ca2+ channel blocker. This toxin might be useful in pain or stroke treatment where blockade of multiple channel types could be beneficial (see reviews by McIntosh and Jones 2001; Scott et al. 2002). In basic science experiments, this toxin may be useful to block reversibly transmitter release or ion currents that are evoked by action of multiple Ca2+ channel types (see review by Fisher and Bourque 2001).
|
DISCLOSURES |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONDISCLOSURESACKNOWLEDGMENTSREFERENCES |
|---|
Present address of S.-Y. Hua: Department of Biological Science, Barnard College, New York, NY, 10027.
|
ACKNOWLEDGMENTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONDISCLOSURESACKNOWLEDGMENTSREFERENCES |
|---|
|
FOOTNOTES |
|---|
Address for reprint requests and other correspondence: M. P. Charlton, Dept. Physiology, MSB 3308, University of Toronto, 1 King's College Circle, Toronto, ON M5S 1A8, Canada (E-mail: milton{at}spine.med.utoronto.ca).
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONDISCLOSURESACKNOWLEDGMENTSREFERENCES |
|---|
Araque A, Clarac F, and Bu' no W. P-type Ca2+ channels mediate excitatory and inhibitory synaptic transmitter release in crayfish muscle. Proc Natl Acad Sci USA 91: 4224-4228, 1994.
Araujo DA, Cordeiro MN, Diniz CR, and Beirao PS. Effects of a toxic fraction, PhTx2, from the spider Phoneutria nigriventer on the sodium current. Naunyn-Schmiedebergs Arch Pharmacol 347: 205-208, 1993.[ISI][Medline]
Bindokas VP, Venema VJ, and Adams ME. Differential antagonism of transmitter release by subtypes of omega-agatoxins. J Neurophysiol 66: 590-601, 1991.
Carvalho CM, Ferreira IL, Duarte CB, Malva JO, Tretter L, Adam-Vizi V, and Carvalho AP. Relation of [Ca2+]i to dopamine release in striatal synaptosomes: role of Ca2+ channels. Brain Res 669: 234-244, 1995.[ISI][Medline]
Cassola AC, Jaffe H, Fales HM, Castro Afeche S, Magnoli F, and Cipolla-Neto J. omega-phonetoxin-IIA: a calcium channel blocker from the spider Phoneutria nigriventer. Pfluegers 436: 545-552, 1998.
Cordeiro MdN, de Figueiredo SG, Valentim AdC, Diniz CR, von Eickstedt VR, Gilroy J, and Richardson M. Purification and amino acid sequences of six Tx3 type neurotoxins from the venom of the Brazilian "armed" spider Phoneutria nigriventer (Keys). Toxicon 31: 35-42, 1993.[Medline]
Dos Santos RG, von Renterghem C, Martin-Moutot N, Mansuelle P, Cordeiro MN, Diniz CR, Mori Y, de Lima ME, and Seagar M. Phoneutria nigriventer -phonetoxin IIA blocks the Cav2 family of calcium channels and interacts with -conotoxin-binding sites. J Biol Chem 277: 13856-13862, 2002.
Ertel EA, Warren VA, Adams ME, Griffin PR, Cohen CJ, and Smith MM. Type III -agatoxins: a family of probes for similar binding sites on L- and N-type calcium channels. Biochemistry 33: 5098-5108, 1994.[Medline]
Figueiredo SG, Garcia ME, Valentim AC, Cordeiro MN, Diniz CR, and Richardson M. Purification and amino acid sequence of the insecticidal neurotoxin Tx4(6-1) from the venom of the "armed" spider Phoneutria nigriventer (Keys). Toxicon 33: 83-93, 1995.[Medline]
Fisher TE and Bourque CW. The function of Ca2+ channel subtypes in exocytotic secretion: new perspectives from synaptic and non-synaptic release. Prog Biophys Mol Biol 77: 269-303, 2001.[ISI][Medline]
Gomez RS, Casali TA, Romano-Silva MA, Cordeiro MN, Diniz CR, Moraes-Santos T, Prado MA, and Gomez MV. The effect of PhTx3 on the release of 3H-acetylcholine induced by tityustoxin and potassium in brain cortical slices and myenteric plexus. Neurosci Lett 196: 131-133, 1995.[ISI][Medline]
Gomez MV, Kalapothakis E, Guatimosim C, and Prado MA. Phoneutria nigriventer venom: a cocktail of toxins that affect ion channels. Cell Mol Neurobiol 22: 579-588, 2002.[ISI][Medline]
Hong SJ and Lnenicka GA. Characterization of a P-type calcium current in a crayfish motoneuron and its selective modulation by impulse activity. J Neurophysiol 77: 76-85, 1997.
Hua SY, Raciborska DA, Trimble WS, and Charlton MP. Different VAMP/synaptobrevin complexes for spontaneous and evoked transmitter release at the crayfish neuromuscular junction. J Neurophysiol 80: 3233-3246, 1998.
Jahromi BS, Robitaille R, and Charlton MP. Transmitter release increases intracellular calcium in perisynaptic Schwann cells in situ. Neuron 8: 1069-1077, 1992.[ISI][Medline]
Katz E, Ferro PA, Cherksey BD, Sugimori M, Llinas R, and Uchitel OD. Effects of Ca2+ channel blockers on transmitter release and presynaptic currents at the frog neuromuscular junction. J Physiol 486: 695-706, 1995.
Katz E, Protti DA, Ferro PA, Rosato Siri MD, and Uchitel OD. Effects of Ca2+ channel blocker neurotoxins on transmitter release and presynaptic currents at the mouse neuromuscular junction. Br J Pharmacol 121: 1531-1540, 1997.[ISI][Medline]
Kimura M, Yamanishi Y, Hanada T, Kagaya T, Kuwada M, Watanabe T, Katayama K, and Nishizawa Y. Involvement of P-type calcium channels in high potassium-elicited release of neurotransmitters from rat brain slices. Neuroscience 66: 609-615, 1995.[ISI][Medline]
Liang H and Elmslie KS. Rapid and reversible block of N-type calcium channels (CaV 2.2) by omega-conotoxin GVIA in the absence of divalent cations. J Neurosci 22: 8884-8890, 2002.
McDonough SI, Boland LM, Mintz I, and Bean BP. Interaction among toxins that inhibit N-type and P-type calcium channels. J Gen Physiol 119: 313-328, 2002.
McIntosh JM and Jones RM. Cone venom–from accidental stings to deliberate injection. Toxicon 39: 1447-1451, 2001.[Medline]
Meir A, Ginsburg S, Butkevich A, Kachalsky SG, Kaiserman I, Ahdut R, Demirgoren S, and Rahamimoff R. Ion channels in presynaptic nerve terminals and control of transmitter release. Physiol Rev 79: 1019-1088, 1999.
Miranda DM, Romano-Silva MA, Kalapothakis E, Diniz CR, Cordeiro MN, Moraes-Santos T, De Marco L, Prado MA, and Gomez MV. Spider neurotoxins block the beta scorpion toxin-induced calcium uptake in rat brain cortical synaptosomes. Brain Res Bull 54: 533-536, 2001.[ISI][Medline]
Olivera BM, Miljanich GP, Ramachandran J, and Adams ME. Calcium channel diversity and neurotransmitter release: the omega-conotoxins and omega-agatoxins. Annu Rev Biochem 63: 823-867, 1994.[ISI][Medline]
Pocock JM, Venema VJ, and Adams ME. Omega-agatoxins differentially block calcium channels in locust, chick and rat synaptosomes. Neurochem Int 20: 263-270, 1992.[ISI][Medline]
Protti DA, Reisin R, Mackinley TA, and Uchitel OD. Calcium channel blockers and transmitter release at the normal human neuromuscular junction. Neurology 46: 1391-1396, 1996.
Rathmayer W, Djokaj S, Gaydukov A, and Kriesel S. The neuromuscular junctions of the slow and the fast excitatory axon in the closer of the crab Eriphia spinifrons are endowed with different Ca2+ channel types and allow neuron-specific modulation of transmitter release by two neuropeptides. J Neurosci 22: 708-717, 2002.
Rezende L Jr, Cordeiro MN, Oliveira EB, and Diniz CR. Isolation of neurotoxic peptides from the venom of the "armed" spider Phoneutria nigriventer. Toxicon 29: 1225-1233, 1991.[Medline]
Robitaille R, Adler EM, and Charlton MP. Strategic location of calcium channels at transmitter release sites of frog neuromuscular synapses. Neuron 5: 773-779, 1990.[ISI][Medline]
Robitaille R, Adler EM, and Charlton MP. Calcium channels and calciumgated potassium channels at the frog neuromuscular junction. J Physiol Paris 87: 15-24, 1993.[ISI][Medline]
Sano K, Enomoto K, and Maeno T. Effects of synthetic omega-conotoxin, a new type Ca2+ antagonist, on frog and mouse neuromuscular transmission. Eur J Pharmacol 141: 235-241, 1987.[ISI][Medline]
Scott DA, Wright CE, and Angus JA. Actions of intrathecal omega-conotoxins CVID, GVIA, MVIIA, and morphine in acute and neuropathic pain in the rat. Eur J Pharmacol 451: 279-286, 2002.[ISI][Medline]
Souccar C, Goncalo MdC, Lapa AJ, Troncone LR, Lebrun I, and Magnoli F. Blockade of acetylcholine release at the motor endplate by a polypeptide from the venom of Phoneutria nigriventer. Br J Pharmacol 116: 2817-2823, 1995.[ISI]
Sugiura Y, Woppmann A, Miljanich GP, and Ko CP. A novel omega-conopeptide for the presynaptic localization of calcium channels at the mammalian neuromuscular junction. J Neurocytol 24: 15-27, 1995.[ISI][Medline]
Takeuchi A and Takeuchi N. A study of the action of picrotoxin on the inhibitory neuromuscular junction of the crayfish. J Physiol 205: 377-391, 1969.
Thaler C, Li W, and Brehm P. Calcium channel isoforms underlying synaptic transmission at embryonic Xenopus neuromuscular junctions. J Neurosci 21: 412-422, 2001.
Troncone LR, Lebrun I, Magnoli F, and Yamane T. Biochemical and pharmacological studies on a lethal neurotoxic polypeptide from Phoneutria nigriventer spider venom. Neurochem Res 20: 879-883, 1995.[ISI][Medline]
Venema VJ, Swiderek KM, Lee TD, Hathaway GM, and Adams ME. Antagonism of synaptosomal calcium channels by subtypes of -agatoxins. J Biol Chem 267: 2610-2615, 1992.
Wagner JA, Snowman AM, Biswas A, Olivera BM, and Snyder SH. Omega-conotoxin GVIA binding to a high-affinity receptor in brain: characterization, calcium sensitivity, and solubilization. J Neurosci 8: 3354-3359, 1988.[Abstract]
Witcher DR, De Waard M, and Campbell KP. Characterization of the purified N-type Ca2+ channel and the cation sensitivity of omega-conotoxin GVIA binding. Neuropharmacology 32: 1127-1139, 1993.[ISI][Medline]
Wojtowicz JM and Atwood HL. Presynaptic membrane potential and transmitter release a
). However, in another set of experiments (
), when asynchronous release rate was first increased (by 
) and after addition of 1.2 µM 
). Top inset: a representative image of fluorescent Ca2+ indicator Calcium Green-1-dextran at crayfish presynaptic boutons loaded via direct injection into axons. Bottom inset: simultaneously obtained electrophysiological recordings from same muscle fiber before and after toxin application. Note absence of detectable EJPs in the presence of