Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

Smith and Hindle (1931) as well as Minton (1968) suggested that the venom of Trimeresurus wagleri (subgenus Tropidolaemus; Brattstrom, 1964), an arboreal pit viper indigenous to the islands and peninsulas of Southeast Asia, acted on the nervous system. However, Tan and Tan (1989) observed that this venom had no effect on the mechanical response of the chick biventer cervicis muscle to either nerve stimulation or bath application of acetylcholine (ACh). Thus, Tan and Tan suggested that the two lethal nonenzymatic toxins that they isolated from the crude venom were also not neurotoxins. In a further characterization of the venom of T. wagleri, Weinstein et al. (1991) purified two lethal peptides having the exact same amino acid sequence except for residue 10; this residue was histidine and tyrosine for peptide 1 and 2, respectively. These lethal peptides constituted less than 1% of the total protein in the venom of T. wagleri. Subsequently, Schmidt et al. (1992) isolated two additional lethal nonenzymatic peptides from crude venom. These peptides were equivalent to peptides 1 and 2 except for serine-leucine residues added to the amino terminus. Schmidt et al. (1992) pointed out that the four lethal peptides in T. wagleri venom were unique because of their amino acid sequence, high proline content, thermal stability (Minton, 1968; Weinstein et al., 1991), and single disulfide bond, which was essential for lethality. Furthermore, the lack of phospholipase A, proteolytic and hemolytic activity, as well as the apparent lack of neurotoxicity suggested that the toxins must also have a unique mechanism of action. Schmidt et al. (1992) named the family of peptides isolated from T. wagleri venom the "Waglerins".

In contrast to the predictions deriving from Tan and Tan's (1989) analysis of chick muscle, a study of adult rat fast twitch muscle revealed a neurotoxic action of Waglerin-1 (Aiken et al., 1992). This study suggested that Waglerin-1 was indeed functionally novel in that it appeared to act at both pre- and postsynaptic loci. Subsequent work of Tsai et al. (1995) supported this hypothesis. Because Waglerin-1 inhibited the response of motor end-plates in mature mouse muscle to iontophoretically applied ACh (Sellin et al., 1996), its lethal effect was due, in part at least, to the block of the nicotinic ACh receptor (nAChR).

Schmidt et al. (1992) reported that the i.p. LD₅₀ value of Waglerin-1 was 0.33 mg/kg for adult mice. Therefore, we routinely administered several times this dose to mice as a bioassay for peptide activity. This led to the interesting observation that neonatal mice do not respond to Waglerin-1. One explanation for this difference was that the embryonic form of the muscle nAChR (Mishina et al., 1986) does not interact with Waglerin-1. In support of this hypothesis, preliminary biochemical studies showed that Waglerin-1 protected the heterologously expressed mature but not the immature muscle type nAChR of mouse from binding -bungarotoxin (Molles et al., 1997, 1998). To further test our hypothesis, we examined the in vitro effects of Waglerin-1 on nerve-muscle preparations isolated from adult and neonatal wild-type mice as well as knockout (KO) mice lacking the gene coding for the -subunit of the nAChR. Some of the results have appeared in abstract form (McArdle et al., 1995).

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Animals and Waglerin-1. The majority of the experiments were performed on wild-type mice. Breeding pairs of Swiss-Webster mice were obtained from Taconic Farms, Inc. (Germantown, NY) and housed in our Institutional Animal Care Facility. Their offspring were examined at various times after birth. Additional experiments were performed on 30- to 35-day-old homo- and heterozygous -subunit KO mice (Witzemann et al., 1996). Experimental protocols were approved by the Institutional Animal Care and Use Committee.

In Vitro Electrophysiology. Recordings (Axoclamp-2A; Axon Instruments, Foster City, CA) of synaptic potentials were made at 21-24°C with conventional sharp electrode techniques (McArdle and Albuquerque, 1973). Briefly, mice were anesthetized with diethyl ether. Either the soleus nerve-muscle preparation and/or the Triangularis sterni (TS) muscle was dissected and pinned to a Sylgard-lined Plexiglas chamber perfused with a standard physiological solution containing (mM): NaCl (135), KCl (5 or 2.5 for end-plate potential recording from crushed fiber preparations), MgCl₂ (1), NaHCO₃ (15), Na₂HPO₄ (1), CaCl₂ (2), and D-glucose (11). This solution was bubbled with 95% O₂/5% CO₂ to maintain pH at 7.3 to 7.4. Miniature end-plate potentials (mepps) and end-plate potentials (epps) were recorded in the soleus muscle. Soleus fibers were crushed (McArdle, 1975) to allow stable recording of epps. The sensitivity of the motor end-plate to iontophoretically applied ACh was assayed in the TS muscle because of its favorable anatomy; end-plates in the TS muscle were visualized at 500X with Nomarski optics (McArdle et al., 1981). Iontophoretic pipettes had a resistance of approximately 200 Mohms when filled with 3 M ACh; a constant braking current of 10 to 30 nA prevented leakage of ACh from this pipette. Iontophoretically and neurally evoked synaptic potentials were recorded before, during, and after superfusion of the end-plate region with bath solution containing varying concentrations of Waglerin-1. Thus, each muscle fiber served as its own control. The experimental solution was locally applied to the end-plate region by manually opening and closing a valve regulating flow of the Waglerin-1 containing solution (Ye and McArdle, 1997); perfusion of the recording chamber with control bathing solution was maintained at a flow rate of 5 ml/min throughout an experiment. Muscle fibers whose resting membrane potential changed by more than 5 mV during an experiment were not analyzed. All potentials were acquired and analyzed with PCLAMP software (Axon Instruments). Further off-line statistical analyses and data plots were made with commercially available software (Excel, Microsoft Corp., Redmond WA; Sigmaplot, Jandel Scientific Software, San Rafael, CA). P values were calculated with two-sided Student's t tests; p < .05 indicated a statistically significant difference.

Binding of -Bungarotoxin. TS muscles from P11 as well as adult wild-type mice were dissected and placed into two separate groups. Group 1 was exposed to the physiologic buffer described above containing Texas Red-labeled -bungarotoxin (Molecular Probes, Eugene, OR; 1:1000, 1:2000). After a 1-h incubation (22-24°C), muscles were washed 4 times with PBS and fixed for 15 min with 4% paraformaldehyde in 120 mM phosphate buffer (4% sucrose, pH 7.3), washed 4 times with PBS, and mounted on glass slides with Fluorsave (Calbiochem, San Diego, CA). Group 2 was pretreated for 20 min with physiologic solution plus 5 µM Waglerin-1 and then exposed for an additional 1 h to 5 µM Waglerin-1 plus Texas Red--bungarotoxin before fixation and mounting on slides. Muscles were viewed and photographed on a Zeiss Axiophot immunofluorescence microscope.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

As described (Schmidt et al., 1992), i.p. injection of 6 to 8 times the LD₅₀ value of Waglerin-1 caused tachypnea, tremors, loss of the righting reflex, myoclonus, and exophthalmus before death within 4 min for 100% of the adult wild-type mice examined. In contrast, P8 to P10 mice exhibited none of these symptoms when injected with the same amount of Waglerin-1. At P11 to P13, some mice transiently lost muscular coordination but retained the righting reflex; 5 of the 22 mice injected between P11 and P13 died within 1 h after Waglerin-1. The percentage of mice succumbing to Waglerin-1 increased with age so that by P20 100% died (Fig. 1). In addition, the interval between the time of injection and death decreased as a function of age.

View larger version (17K): [in this window] [in a new window]   Fig. 1.   Neonatal mice are resistant to the lethal effect of Waglerin-1. Mice were administered an i.p. injection of 50 µg of Waglerin-1. The percentage of mice dying in response to such injections is plotted as a function of the days after birth. The number of mice injected was 23 at P8-P10, 22 at P11-P13, 13 at P14-P16, 14 at P17-P18, and 14 at P20 to Adult.

Because prior work (Aiken et al., 1992; Sellin et al., 1996) suggested that the lethal action of Waglerin-1 was due, at least in part, to inhibition of the nAChR, we hypothesized that lethality depended on the presence of the mature  form of this receptor (Mishina et al., 1986; Witzemann et al., 1987; Missias et al., 1996; Villarroel and Sakmann, 1996). As an initial test of this hypothesis, we injected 30- to 35-day-old heterozygous and homozygous -subunit KO mice (Witzemann et al., 1996) with 50 µg of Waglerin-1. In accord with our hypothesis, the phenotypically wild-type heterozygous mouse died within 4 min of injection, whereas the homozygous KO mouse exhibited no response. Thus, the Waglerin-1 sensitivity of adult wild-type and heterozygous KO mice was identical but dramatically different from that of neonatal wild-type and adult homozygous KO mice. To further test our hypothesis, we compared the in vitro effect of Waglerin-1 on the response of motor end-plates in muscles from these animals to neurally and iontophoretically applied ACh. On the basis of the lethality study of Fig. 1, neonatal wild-type mice at P11 or younger were selected for these studies unless indicated otherwise.

Within 4 min of superfusing adult muscle with 1 µM Waglerin-1, mepps were lost in the baseline noise; this effect reversed slowly upon washout of Waglerin-1 (Fig. 2). In contrast, mepps were unaltered for neonatal muscles exposed to 1 µM Waglerin-1. For the neonatal end-plate presented in Fig. 2, mepp frequency was 0.2 s¹ both before and 5 min after 5 µM Waglerin-1; the corresponding mepp amplitudes were 1.5 ± 0.1 mV (n = 12) and 0.7 ± 0.1 mV (n = 10). Mepp amplitude in the neonatal muscle recovered to 1.2 ± 0.2 mV (n = 6) within 10 min of washout. Thus, a higher Waglerin-1 concentration reduced mepp amplitude without producing the complete block observed for muscles of adult mice. Muscle of adult heterozygous KO mice responded to Waglerin-1 similarly to adult wild-type muscle; 4 min of superfusion with 1 µM Waglerin-1 completely suppressed mepps, which slowly recovered with washout. On the other hand, muscle of homozygous KO mice responded to Waglerin-1 exactly as did wild-type neonatal mice because mepps remained unaltered at 4 min after 1 µM Waglerin-1.

View larger version (26K): [in this window] [in a new window]   Fig. 2.   Mepps in the soleus muscle of adult wild-type and heterozygous -subunit KO mice are more sensitive to Waglerin-1 block than are mepps of neonatal wild-type or adult homozygous KO mice. Mepps are abolished within 4 min after exposing adult wild-type muscle to 1 µM Waglerin-1. Small amplitude mepps reappeared 6 min after washout of Waglerin-1. In contrast, exposing neonatal muscle to 5 µM Waglerin-1 (WTX-1) for 5 min reduced mepp amplitude from 1.5 ± 0.1 mV (n = 12) to 0.7 ± 0.1 mV (n = 10), whereas mepp frequency remained unchanged (0.2 s1); mepp amplitude recovered to 1.2 ± 0.2 mV (n = 6) 10 min after washout. As for the adult wild-type muscle, 1 µM Waglerin-1 reversibly blocked mepps of the adult heterozygous, but not the homozygous,  subunit KO mouse. Each trace is the superposition of several 1-s episodes. The calibration at the lower right applies to all records and represents 0.4 mV and 35 ms.

The effect of Waglerin-1 on epps was qualitatively the same as for mepps (Fig. 3). That is, epps of the adult wild-type soleus muscle declined to 20% of control amplitude within 100 s of exposure to 1 µM Waglerin-1; this effect slowly reversed with washin of toxin-free solution. Waglerin-1 also decreased the epps of the heterozygous -subunit KO mouse to 20% of control. On the other hand, epps of soleus muscle from neonatal wild-type and adult homozygous KO mice remained stable at 80% of control amplitude at 100 s after 1 µM Waglerin-1. This 20% decline of epp amplitude for homozygous KO and neonatal mice was not studied further because Waglerin-1 may act presynaptically to alter quantal release of ACh (Aiken et al., 1992; Tsai et al., 1995). Thus, it was essential to explore the effect of Waglerin-1 on the response of the motor end-plate to iontophoretically applied ACh.

View larger version (19K): [in this window] [in a new window]   Fig. 3.   Epps in crushed fiber soleus muscle/nerve preparations of adult wild-type and heterozygous -subunit KO mice are more sensitive to Waglerin-1 block than epps of neonatal wild-type or adult homozygous KO mice. Top, average of five epps recorded before (largest trace), during (smallest trace), and after (middle trace) exposure to 1 µM Waglerin-1 for 1.5 min. The calibration (top, middle) is: 2 mV for all recordings, 2 ms for the adult wild-type and heterozygous KO preparations, 4 ms for the adult homozygous KO, and 6 ms for the wild-type neonatal muscle. Epps were evoked at 0.5 Hz with a supramaximal pulse of 0.1 ms duration. The graphs below the row of records summarize the time course of 1 µM Waglerin-1's effect on epp amplitude: , adult wild-type (6 fibers, 2 muscles, 2 mice) or heterozygous KO (6 fibers, 1 muscle) mice; , neonatal wild-type (21 fibers, 3 muscles, 3 mice) or homozygous KO (11 fibers, 2 muscles, 1 mouse). Each data point is the mean ± S.E.M. of the epp amplitude in the presence of Waglerin-1 expressed as percentage of the control value. Asterisks indicate a significant (p < .05) difference between adult and neonatal wild-type preparations as well as between heterozygous and homozygous KO groups.

Waglerin-1 decreased the response to a constant iontophoretic pulse of ACh applied to end-plates in the TS muscle from adult wild-type mice in a concentration-dependent fashion. For example, whereas 5 nM had little effect, 0.5 µM Waglerin-1 produced almost complete inhibition. That is, the peak response to a constant pulse of ACh decreased to 5 ± 8% of control for 21 fibers in six separate TS muscles of six mice; this effect of Waglerin-1 always reversed, at least in part, within several minutes of wash with toxin-free extracellular solution. The concentration-response curve of Fig. 4 suggests that Waglerin-1 inhibited the ACh response of the TS muscle of adult wild-type mice with an apparent IC₅₀ value of 50 nM; the corresponding Hill coefficient was 1. In contrast, Waglerin-1 had much less effect on the ACh response of end-plates of neonatal muscle. For example, although 0.1 µM Waglerin-1 decreased the ACh response of adult TS muscle to 35 ± 3% (48 fibers, 12 muscles, 12 mice) of control, the response of the neonatal end-plate remained unaltered at 92 ± 3% (36 fibers, 5 muscles, 5 mice) of control. As summarized in the concentration-response curve of Fig. 4, 1 µM Waglerin-1 reduced the ACh response to 9 ± 2% (51 fibers, 16 muscles, 16 mice) and 4 ± 1% (six fibers, one muscle, one mouse) of control for the adult wild-type () and heterozygous KO () mouse. In contrast, 1 µM Waglerin-1 had no effect on the end-plate response to ACh for the homozygous KO mouse (99 ± 2% of control; , 9 fibers, 1 muscle, 1 mouse) and reduced it to 73 ± 2% of control (, 113 fibers, 16 muscles, 16 mice) for the neonatal group (P5-P11) of wild-type mice. Figure 5 illustrates the differential effect of 1 µM Waglerin-1 on the end-plate response to ACh iontophoretically applied to the TS muscle of adult homozygous and heterozygous KO mice.

View larger version (16K): [in this window] [in a new window]   Fig. 4.   The end-plate response to iontophoretically applied ACh is more sensitive to block by Waglerin-1 for the adult than for the neonatal wild-type mouse. Left, representative records of the response of end-plates in adult and neonatal TS muscles to iontophoretically applied ACh. Each trace is the average of five responses to constant pulses of ACh applied before (largest trace), 1 min during (smallest trace), and after (middle trace) the application of 5 nM, 0.1 µM, 0.5 µM, or 1.0 µM Waglerin-1 for adult muscles; no ACh responses are shown after Waglerin-1 washout for the neonate. Pulse amplitude and duration were the same for each of the preceding conditions; the range was 10 to 70 nA and 1 to 2 ms for all of the end-plates examined. The calibration is 4 mV and 20 ms for all records except the neonatal response to 5 nM and 0.1 µM Waglerin-1 where the time calibration is 40 ms. Right, concentration-response relation for the effect of Waglerin-1 on the end-plate response to ACh for the adult wild-type muscle (). The solid line is a fit of the following form of the Hill equation to the experimental points: % control response = (1/(1 + ([WTX-1]/IC50)n)) × 100, where IC50 is the concentration of Waglerin-1 ([WTX-1]) producing 50% inhibition and n is the Hill coefficient. For the adult wild-type muscle, n and IC50 had values of 1 and 50 nM. Data points represent the following numbers of muscle fibers, muscles, and mice for the concentration of Waglerin-1 following in parentheses. Adult wild-type (): 17, 5, 5 (5 nM); 46, 8, 8 (10 nM); 38, 8, 8 (50 nM); 48, 12, 12 (0.1 µM); 21, 6, 6 (0.5 µM); 51, 16, 16 (1.0 µM); 11, 4, 4 (5 µM). Neonate (P5-P11; ): 113, 16, 16 (1.0 µM). Heterozygous KO (): 6, 1, 1 (1.0 µM). Homozygous KO (): 9, 1, 1 (1.0 µM).

View larger version (11K): [in this window] [in a new window]   Fig. 5.   The end-plate response to iontophoretically applied ACh is blocked by 1 µM Waglerin-1 for the adult heterozygous but not the homozygous -subunit KO mouse. The heterozygous records represent: a, control; b, 10 s after application of Waglerin-1; c, 1 min after application of Waglerin-1; d, 4 min of washout. The records for the homozygous KO mouse were: a, control; b, 2 min of 1 µM Waglerin-1; c, 5 min of washout; 1 µM Waglerin-1 did not significantly alter the response to ACh.

Figure 6 presents representative responses to iontophoretically applied ACh recorded from the TS muscle of wild-type mice at varying times after birth. At the earliest time point examined, P5 (four fibers, one muscle, one mouse), 1 µM Waglerin-1 had no effect; the large mepps frequently observed at this stage of development were also unaltered. With age, there was a gradual increase in sensitivity to the inhibitory action of Waglerin-1. Figure 7 summarizes the time course of this sensitivity change; the squares represent the mean ± S.E.M.; the smaller circles (·) represent individual data points. The greatest increase in Walgerin-1 sensitivity occurred between P11 and P12 (p < .05); 1 µM Waglerin-1 decreased the ACh response to 77 ± 8% (27 fibers, 3 muscles, 3 mice) and 37 ± 4% (54 fibers, 7 muscles, 7 mice), respectively. However, the scatter of individual data points reveals that end-plates with lower sensitivity to the effect of Waglerin-1 persisted for several days after P12. By P20, the distribution of data points was very close to that obtained from the adult wild-type muscle. During the transition period of Waglerin-1 sensitivity, we often noted that mepps persisted even when the response to iontophoretically applied ACh was markedly reduced.

View larger version (15K): [in this window] [in a new window]   Fig. 6.   The response to iontophoretic application of ACh becomes increasingly more sensitive to block by 1 µM Waglerin-1 during development. The records at P5, P7, P9, and P11 were obtained before and 1 min after exposure to 1 µM Waglerin-1. The mepps in the P5 records indicate their resistance to Waglerin-1. The records for P13 and P14 are: a, control; b, 1 min of 1 µM Waglerin-1; c, washout. The calibration is: 2 mV for P5 and P9, 4 mV for P7 and P11, and 3 mV for P13 and P14; 80 ms for P5 and 15 ms for all other records.

View larger version (17K): [in this window] [in a new window]   Fig. 7.   Appearance of sensitivity to the blocking action of 1 µM Waglerin-1 for the response to iontophoretically applied ACh for end-plates of wild-type mice at P5 to P20. The response to iontophoretically applied ACh, as percentage of the preWaglerin-1 control, is plotted as function of the days after birth. , mean ± S.E.M., individual data points. A two-tailed Student's t test resulted in p < .05 for the difference in Waglerin-1 sensitivity at P11 and P12. In contrast, the difference between the P11 and P10 values was not significant (p > .05). Mean ± S.E.M. represent the following numbers of muscle fibers, muscles, and mice for the animal ages following in parentheses: 4, 1, 1 (P5); 6, 1, 1, (P7); 34, 4, 4 (P8); 19, 3, 3 (P9); 23, 4, 4 (P10); 27, 3, 3 (P11); 54, 7, 7 (P12); 41, 4, 4 (P13); 30, 3, 3 (P14); 28, 3, 3 (P15); 11, 1, 1 (P16); 23, 2, 2 (P18); and 51, 16, 16 (20-Adult).

End-plates in the TS muscle isolated from adult wild-type mice were labeled by Texas Red--bungarotoxin (Fig. 8). These end-plates were located in a band across the muscle and showed characteristic morphology at high magnification. When TS muscles of adult mice were preincubated with 5 µM Waglerin-1 and incubated with 5 µM Waglerin-1 plus Texas Red--bungarotoxin, fluorescence of end-plates was greatly reduced. No reduction of fluorescence after incubation with 5 µM Waglerin-1 was observed in TS muscles of P11 mice (Fig. 9). Similar results were obtained for five other pairs of control and Waglerin-1-preincubated TS muscles isolated from adult and P11 mice.

View larger version (72K): [in this window] [in a new window]   Fig. 8.   Waglerin-1 (5 µM) protects end-plates in the TS muscle of the adult wild-type mouse from binding -bungarotoxin. The photomicrographs in the right column were made at a greater magnification than those to the left. The upper row control photomicrographs were obtained after incubating muscles in 1:2000 Texas Red-labeled -bungarotoxin for 1 h. The lower row of records was obtained for muscle exposed to 5 µM Waglerin-1 for 20 min before incubation with 1:2000 -bungarotoxin plus 5 µM Waglerin-1.

View larger version (63K): [in this window] [in a new window]   Fig. 9.   Waglerin-1 (5 µM) does not protect end-plates in the TS muscle of the neonatal wild-type mouse from binding -bungarotoxin. The control and experimental treatments are as described in the legend to Fig. 8.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

This work provides both direct and indirect evidence in support of our hypothesis that Waglerin-1 selectively blocks the  form of the muscle nAChR. Direct evidence is the lack of effect of Waglerin-1 on adult homozygous -subunit KO mice or nerve-muscle preparations isolated from them. In contrast, the response to Waglerin-1 of their heterozygous litter mates, as well as nerve-muscle preparations isolated from them, was equivalent to that observed for adult wild-type mice. Indirect evidence is Waglerin-1 protection of end-plates in the TS muscle of adult, but not neonatal, wild-type mice from -bungarotoxin binding. Furthermore, the development of sensitivity to both the in vivo lethal and the in vitro neuromuscular blocking effect of Waglerin-1 for neonatal wild-type mice corresponds with the appearance of the high conductance form of this receptor (Villarroel and Sakmann, 1996) as the -subunit substitutes for the -subunit during synaptogenesis (Mishina et al., 1986; Witzemann et al., 1989, 1991). Specifically, Villarroel and Sakmann (1996) reported that low conductance nAChR channels predominate at end-plates of the flexor digitorum brevis muscle of P9 rats. P9 mice are resistant to the lethal effect of Waglerin-1, and 1 µM Waglerin-1 had no effect on the end-plate response of their TS muscle to iontophoretically applied ACh. Villarroel and Sakmann found that by P14 approximately 50% of the end-plate channels activated by ACh were high in conductance; by P21, high conductance was characteristic of more than 90% of the single channel responses to ACh. Furthermore, Villarroel and Sakmann's analysis of biionic reversal potentials of ensemble currents suggested that the appearance of high end-plate conductance to Ca²⁺ was completed at P15. Thus, the switch from the  to the  form of the nAChR is largely completed between P9 and P15 for the rat flexor digitorum brevis muscle. Between P11 and P12, the response of end-plates in the mouse TS muscle to iontophoretically applied ACh exhibited a significant increase in sensitivity to 1 µM Waglerin-1. Assuming that our hypothesis is correct, the preceding finding suggests that the  form of the nAChR becomes dominant at end-plates of the mouse TS muscle between P11 and P12. However, mepps persist in the presence of 1 µM Waglerin-1 at a time during development when the response to iontophoretically applied ACh is maximally inhibited. This suggests that distinct populations of end-plate receptors mediating the response to neurally and exogenously applied ACh (Albuquerque and Gage, 1978; Argentieri et al., 1992) undergo the  to  shift at different times after birth. Furthermore, the distribution of sensitivities to Waglerin-1 block of iontophoretically evoked ACh responses also suggests that the conversion from the  to  nAChR is not uniform for all end-plates at a given postnatal time. Given the high safety factor for neuromuscular transmission, the persistence of a small population of muscle fibers with immature receptors may protect mice from the neuromuscular block underlying the lethal effect of Waglerin-1; the contribution of such immature receptors to neuromuscular transmission is negligible at P20 when all mice succumb to the lethal effect of Waglerin-1.

The resistance of the  form of the nAChR to Waglerin-1 (Taylor et al., 1998) can explain the lack of neurotoxicity previously suggested. Specifically, Tan and Tan (1989) observed that 5 µg/ml of crude T. wagleri venom did not alter the mechanical response of the biventer cervicis muscle isolated from 2- to 6-day-old chicks to nerve stimulation or bath application of ACh. Our present work suggests that the chick muscle of Tan and Tan's study (1989) was exposed to a concentration of Waglerins that was below the threshold concentration needed to block the immature muscle nAChR. Like the immature receptors of our study, the nAChR of chick muscle might be responsive to higher concentrations of Waglerin-1. This is likely because Waglerin-1 inhibited synaptic phenomena in muscle isolated from adult rats (Aiken et al., 1992) even though rats are resistant to the lethal and in vitro paralytic effect of the peptide (Lin et al., 1995). Comparison of the present data from adult wild-type mice with that obtained previously from rat (Aiken et al., 1992) reveals that 1 µM Waglerin-1 reduced the epp amplitude of muscle from these two species by approximately 80% and 35%, respectively. Thus, it is likely that the - subunit interface of the mature nAChR of mouse and rat muscle differs in affinity for Waglerin-1 (Molles et al., 1997). On the other hand, Tsai et al. (1995) suggested that Waglerin-1 is a potent blocker of Ca²⁺ channels controlling release of ACh from motor nerve terminals. Their suggestion was based on the observation that although 1.2 nM Waglerin-1 potentiates curare inhibition of epps, a 1000-fold higher concentration of Waglerin-1 is needed to block the contractile response of denervated mouse muscle to bath-applied ACh. Furthermore, Tsai et al. (1995) suggested that mepp frequency was reduced in the presence of Waglerin-1 concentrations having no detectable effect on mepp amplitude. Thus, Tsai et al. concluded "... that the presynaptic effect of the toxin is more potent than its postsynaptic effect". However, Tsai et al. offered no explanation as to why a presynaptic locus might contribute to the greater Waglerin-1 sensitivity of mouse relative to rat. In contrast, our present work demonstrates that it is the  form of the muscle nAChR that renders the adult mouse sensitive to the lethal and paralytic effects of Waglerin-1. Therefore, just as for chick muscle, the expression of immature -subunit nAChR (Witzemann et al., 1987) would render the denervated mouse muscle of Tsai et al.'s study relatively resistant to Waglerin-1.

Tsai et al.'s observation that 4 µM Waglerin-1 suppressed Ca²⁺ current of nerve terminals in the mouse TS muscle confirms earlier reports that µM concentrations of this peptide inhibit similar currents of nerve and myocardial muscle cells (McArdle et al., 1992; Ye and McArdle, 1993). In addition, µM Waglerin-1 interacts, in a developmentally determined fashion, with -aminobutyric acid type A receptors of hypothalamic neurons freshly isolated from mice (Ye and McArdle, 1997). Thus, actions on the central nervous system, as well as the heart and the neuromuscular junction, may contribute to the in vivo effects of Waglerin-1. However, the present study clearly demonstrates that the most sensitive in vitro locus of Waglerin-1 action is the nAChR of adult mouse muscle.

In conclusion, this study suggests that the site underlying the lethal action of Waglerin-1 is the  subunit of the mature nAChR of mouse skeletal muscle. The apparent Hill coefficient of 1 suggests a simple bimolecular interaction of Waglerin-1 with the mature nAChR of mouse skeletal muscle. Thus, the Waglerins are novel tools for exploring new drug targets on nAChRs as well as the significance of the developmentally programmed shift of receptor subunit composition.