The pharmacological effect of Bothrops neuwiedii pauloensis ( jararaca-pintada ) snake venom on avian neuromuscular transmission

! " " " # $ % ! & $' ( ) *+ $$ ) " $ ) $ ) # ! & " & *+ # % %(,+& ! " ' . / 0& 1 2 ' $% " " *+ 3 & " " & 4" " " Correspondence L. Rodrigues-Simioni Departamento de Farmacologia FCM, UNICAMP Caixa Postal 6111 13083-970 Campinas, SP Brasil Fax: +55-19-3289-2968 E-mail: simioni@obelix.unicamp.br Research supported by FAPESP, CNPq, CAPES and FAEP-UNICAMP. Received July 3, 2002 Accepted November 5, 2002


Introduction
Snakes of the genus Bothrops are the most important cause of snakebites in Brazil.The main complications in lethal cases are acute renal failure, shock, acute respiratory failure, and sepsis (1,2).The mechanism of respiratory failure is not clearly understood since Bothrops venoms do not produce apparent signs of neurotoxicity after snake-bites.In some cases, respiratory failure was associated with pulmonary edema (1).Nevertheless, Zamuner et al. (3) reported that Bothrops neuwiedii venom caused head drop, loss of balance and respiratory failure in chicks in vivo after injection of a 0.55 mg/kg.The same report described inhibition of the twitch-tension response to B. neuwiedii venom in chick biventer cervicis preparations, but found no decrease in the responses to exogenous acetylcholine or KCl.Subsequently, Borja-Oliveira et al. (4) observed intraspecific variation in the neuromuscular activity of 16 lots of B. neuwiedii venoms in chick nerve-muscle preparations.At low concentrations (5-20 µg/ml), most of the venoms reduced the twitch-tension without completely abolishing the contracture to exogenous acetylcholine, thus suggesting a presynaptic action.The neuromuscular blockade with these venoms varied from 2 to 100% and the electrophoretic profile of the venoms which had the highest neuromuscular potency also had an additional electrophoretic band compared to the other venoms.B.n. pauloensis venom also markedly increased the frequency of miniature endplate potentials (mepps) and the incidence of giant mepps in the mouse phrenic nervediaphragm (Durigon AM, Borja-Oliveira CR, Dal Belo CA, Oshima-Franco Y, Cogo JC, Lapa AJ, Souccar C and Rodrigues-Simioni L, unpublished data).In the present study, we examined the neurotoxic action of one of the most neurotoxic samples of B.n. pauloensis venom screened in chick biventer cervicis preparations.

Isolated chick biventer cervicis nerve-muscle preparation
The biventer cervicis muscle was removed from chicks as described by Ginsborg and Warriner (5) and mounted under a tension of 0.5 g in a 5 ml organ bath containing aerated (95% O 2 , 5% CO 2 ) Krebs solution, pH 7.5, 37ºC, of the following composition: 118.6 mM NaCl, 4.69 mM KCl, 1.88 mM CaCl 2 , 1.17 mM KH 2 PO 4 , 1.17 mM MgSO 4 , 25.0 mM NaHCO 3 , and 11.65 mM glucose.Indirect (0.1 Hz, 0.2 ms, 6-7 V) and direct (0.1 Hz, 0.2 ms, 50 V) stimulation with a Grass S4 stimulator (Quincy, MA, USA) was used to stimulate the muscle and contractions and contractures were recorded via a force displacement-transducer (BG 25 GM Kulite, Leonia, NJ, USA) coupled to a Gould RS 3400 recorder (Cleveland, OH, USA).The preparations were allowed to stabilize for at least 15 min before the addition of venom (1,5,10,20,50 or 100 µg/ml).Contractures to exogenously applied submaximal concentrations of 110 µM acetylcholine (for 60 s or cumulative concentrations of 1 µM to 10 mM) and 13.4 mM KCl (for 120-160 s) were obtained in the absence of nerve stimulation prior to the addition of venom and at the end of the experiment, in order to test for the presence of neurotoxic and myotoxic activities (6).Some experiments were done at 20-24ºC or after the addition of 11.7 µM d-tubocurarine.The venom concentration of 20 µg/ml (N = 7) was not included in the

Statistical analysis
Each experiment was repeated at least three times.The results are reported as the mean ± SEM.The Student t-test was used for statistical analysis of the data, with values of P<0.05 indicating statistical significance.neuromuscular blockade in indirectly stimulated chick biventer cervicis preparations (Table 1, Figures 1 and 2).In most experiments, the venom (5-50 µg/ml) caused an initial inhibition and a secondary increase of indirectly evoked twitches followed by a progressive neuromuscular blockade (Figures 1 and 2).

Discussion
In Brazil, the snakes of the genera Crotalus and Micrurus are the only ones that produce signs of peripheral neurotoxicity after snakebites, such as palpebral ptosis and respiratory paralysis.Nevertheless, Bothrops snakebites may also produce respiratory failure associated with pulmonary edema in some cases (1).
Zamuner et al. ( 3) reported the neurotoxic action of B. neuwiedii venom in chicks in vivo and in vitro.In isolated chick nervemuscle preparations, B. neuwiedii venom (10-50 µg/ml) completely blocked neuromuscular transmission without depressing the response to acetylcholine.These results suggested a presynaptic site of action for this venom.However, not all B. neuwiedii venoms exhibit significant neurotoxicity on isolated neuromuscular preparations (4).
The present report describes the neuromuscular activity of the venom of one of the most toxic subspecies of B. neuwiedii, B.n. pauloensis, screened by Borja-Oliveira et al. (4).This venom produced concentrationdependent neuromuscular blockade in chick biventer cervicis preparations.Complete blockade at a low concentration (5 µg/ml) was not accompanied by inhibition of the responses to KCl and acetylcholine.These observations indicate that at low concentrations the venom had no inhibitory effect on postsynaptic acetylcholine receptors and its action was not dependent on myotoxicity.
At high concentrations (>10 µg/ml), the venom has a myotoxic effect, including the inhibition of KCl-induced contractures, pronounced muscle contracture and the partial inhibition of contractions in response to direct muscle stimulation in curarized preparations.Recently, Soares et al. (7)    this venom, namely BnSP-7, on chick biventer cervicis preparations.The phospholipase A 2 BnSP-7 inhibited the twitch-tension and KCl-induced contractures only at high concentrations.In addition, both BnSP-7 and the crude venom released creatine kinase from the mouse gastrocnemius muscle and induced a dose-dependent edema.
Since the blockade of the responses to indirect stimulation and to KCl as well as the incidence of contractures were temperature dependent, enzyme activity may be involved in the neuromuscular action of B.n. pauloensis venom.B. neuwiedii venom contains phospholipases (7)(8)(9)(10)(11) and the venom used in the present study had significant phospholipase A 2 activity (Borja-Oliveira CR, Kassab BH, Durigon AM, Soares AM, Toyama MH, Novello JC, Giglio JR, Marangoni S and Rodrigues-Simioni L, unpublished results).The data available do not permit us to identify the putative enzyme.
The neuromuscular action produced by B.n. pauloensis venom is consistent with the available data about phospholipase A 2 neurotoxins.Crotoxin, the main neurotoxin of Crotalus durissus terrificus venom, causes a triphasic change (depression, facilitation and final blockade) of acetylcholine release by the nerve terminals (12)(13)(14)(15), similar to that observed with other snake ß-neurotoxins, such as ß-bungarotoxin (16)(17)(18), notexin (19), taipoxin (20,21), and textilotoxin (22,23).In mouse hemi-diaphragm nerve-muscle preparations, by reducing the external Ca 2+ concentration, ß-bungarotoxin classically produces an initial transient inhibition of twitches (phase 1) followed by a prolonged facilitatory phase (phase 2) and finally a blocking phase (phase 3).The facilitatory effect of these toxins on mammalian nerve-muscle preparations is independent of phospholipase A 2 activity.Indeed, the increase in the number of twitches produced by B.n. pauloensis venom was present at 20-24ºC, although the neuromuscular blocking action (phase 3) was inhibited.The third phase (complete and irreversible inhibition of neurotransmission) produced by ß-neurotoxins depends on temperature and on the presence of Ca 2+ in the medium and is probably due to phospholipase A 2 -mediated destruction of membrane phospholipids in motor nerve terminals.The enzymatic activity of ß-neurotoxins is not significantly correlated with their toxicity (24), but is obligatory for the full expression of neurotoxic activity since its inhibition prevents lethality (25).

Figure 1 .
Figure 1.Effect of low concentrations of Bothrops neuwiedii pauloensis venom (1, 5 and 10 µg/ml) on chick biventer cervicis nerve-muscle preparations at 37ºC.A, Concentration-time response curves for the neuromuscular blocking activity of the venom compared to Krebs solution alone (control).Each point represents the mean ± SEM of 3-6 experiments.B, Twitch-tension responses by a control (Krebs solution alone) preparation.C and D, Venomtreated preparations (5 and 10 µg/ml, respectively).B.n. pauloensis venom was added (arrow) after allowing the preparations to stabilize for 15 min.Note the contracture induced by 10 µg of venom/ml.The responses to exogenous 110 µM acetylcholine (triangles) and 13.4 mM KCl (circles) were obtained before and after the addition of venom.These results are representative of 3-6 experiments per venom concentration.*P<0.05 compared to the control preparations (Student t-test).

Figure 2 .
Figure 2. Effect of high concentrations (50-100 µg/ml) of Bothrops neuwiedii pauloensis venom on chick biventer cervicis preparations at 37ºC.A, Concentration-time response curves for the neuromuscular blocking activity of the venom compared to Krebs solution alone (control).Each point represents the mean ± SEM of 3-6 experiments.B and C, Venom-treated preparations (50 and 100 µg/ml, respectively).B.n. pauloensis venom was added (arrow) after allowing the preparations to stabilize for 15 min.Note that the contractions induced by the venom were concentration dependent.The responses to exogenous 110 µM acetylcholine (triangles) and 13.4 mM KCl (circles) were obtained before and after the addition of venom.These results are representative of 3-4 experiments per venom concentration.*P<0.05 compared to the control preparations (Student t-test).

Figure 3 .
Figure 3.Effect of Bothrops neuwiedii pauloensis venom on muscle contractions induced by 13.4 mM KCl and 110 µM acetylcholine (ACh) in the chick biventer cervicis preparations after 120-min incubation compared to the responses observed in Krebs controls (100%).Data are reported as means ± SEM for 3-6 experiments per concentration.The preparations were incubated with venom at the concentrations and temperatures indicated on the histogram.The triangles on the X-axis indicate that the venom totally blocked the response to the agonist.*P<0.05 compared to the corresponding control (Krebs solution alone) (Student t-test).

Figure 5 .
Figure 5.Effect of Bothrops neuwiedii pauloensis venom on the twitches elicited by direct muscle stimulation.A and B, Venom-treated chick biventer cervicis preparations at 20 and 50 µg/ml, respectively (arrow, time zero).Direct stimulation (D) was applied after treatment with d-tubocurarine (7.3 µM d-Tc) and after several washes (W) with Krebs solution.These results are representative of 3 and 5 experiments for A and B, respectively.

Figure 4 .
Figure 4. Effect of Bothrops neuwiedii pauloensis venom (50 µg/ml) on chick biventer cervicis preparations at 20-24ºC compared to 37ºC.A, Concentration-time response curves for the neuromuscular blocking activity of the venom compared to Krebs solution alone (control) are shown.Each point represents the mean ± SEM of 3-6 experiments.B, Venom-treated preparations (50 µg/ml) at 37ºC.C, A low incubation temperature (20-24ºC) prevented the neuromuscular action of the venom (50 µg/ml).B.n. pauloensis venom was added (arrow) after allowing the preparations to stabilize for 15 min.The responses to exogenous 110 µM acetylcholine (triangles) and 13.4 mM KCl (circles) were obtained before and after the addition of venom.These results are representative of 3-4 experiments.*P<0.05 compared to the control preparations (Student t-test).