Androctonus australis insect toxin (AaIT) is an insect-selective neurotoxic polypeptide from scorpion venom used to probe insect Na+ channels and to design insecticidal recombinant baculoviruses. When injected into susceptible insects (such as flies or cockroaches), nanogram doses of the toxin induce a rapid paralysis within seconds. More tolerant insects respond to microgram doses by developing either a slow progressive paralysis, as in lepidopterous larvae, or a rapid but reversible paralysis, as in Trachyderma philistina, a tenebrionid beetle. Using toxicity and binding assays, microscopy and chromatography, we show that the tolerance of insects to AaIT occurs at both the pharmacokinetic and pharmacodynamic levels. Pharmacokinetic effects occur in Trachyderma philistina in which the toxin undergoes a progressive process of degradation and elimination from the hemolymph, resulting in the loss of 95–97 % of toxin activity 6 h after injection. The pharmacodynamic aspect was demonstrated in studies of the kinetics of binding dissociation of [125I]AaIT from neuronal membranes of susceptible and tolerant insects. Stable binding is shown in susceptible insects such as cockroaches and locusts, which have a dissociation half-time of approximately 9 and 5 min, respectively. This contrasts strongly with the fast half-time of dissociation of 7 s for Spodoptera littoralis larvae and 9 s for Trachyderma philistina, which are both relatively tolerant to AaIT. These differences in binding kinetics may reflect a structural and functional diversity of Na+ channels in different insects that is responsible for their diverse susceptibility to neurotoxic polypeptides.

AaIT is a single-chain neurotoxic polypeptide (Mr 8×103) representative of the excitatory insect-selective neurotoxins derived from Buthinae scorpion venoms (Zlotkin, 1991). The toxin exclusively affects insects, modifying their neuronal Na+ conductance by binding to voltage-dependent Na+ channels (Gordon et al. 1992). Its strict selectivity for insects, demonstrated by toxicity, by electrophysiology and by binding assays (Zlotkin, 1991), has strongly encouraged its use in pest-control technology. This approach, motivated by the public health and environmental hazards that result from the massive utilization of industrial insecticides, led to the genetic combination of a baculovirus and the gene encoding the AaIT toxin. The resulting recombinant baculoviruses, when compared with the wild-type viruses, significantly enhance the lethality to insect pests, thus reducing their feeding damage (Stewart et al. 1991; McCutchen et al. 1991). Furthermore, the AaIT-expressing recombinant baculovirus has recently reached the stage of field trials, providing encouragement for the further employment of insecticidal polypeptide neurotoxins in the design and employment of selective bioinsecticides (Cory et al. 1994).

With this background, and because of its neuropharmacological and applicative significance, the basis for differences in toxicity of AaIT between species deserves clarification. This may reveal structural differences in ion channels between insect species and lead to inferred chemical modifications of the clonable neurotoxins.

The present study reveals that insect tolerance to AaIT, in addition to the common phenomena of degradation and elimination, also occurs at the pharmacodynamic level in the nature of the interaction of the toxin with its neuronal receptor.

Materials

Crude venom from the scorpion Androctonus australis was purchased from Latoxan (Rosans, France). The A. australis insect toxin (AaIT) was purified by column chromatography according to a previously described method (Zlotkin et al. 1971b).

Inulin [14C]carboxylic acid, for the determination of hemolymph volume and Na[125I] for toxin radioiodination were purchased from Amersham International plc (England).

Test animals

Tenebrionid beetles (Trachyderma philistina and Adesmia sp.), of both sexes and with a body mass of 800–1100 mg, were collected in the region of the Dead Sea and Arad southeast of Jerusalem. Larvae of the blowfly Sarcophaga falculata were bred according to the method of Zlotkin et al. (1971a).

Locusts (Locusta migratoria) and cockroaches (Periplaneta americana) were kindly donated by Professor M. Pener and Professor J. Camhi, respectively, from the Life Sciences Institute of the Hebrew University. Lepidopterous larvae (Spodoptera littoralis) were purchased from The Volcani Center for Agriculture.

Toxicity assays

Blowfly larvae were used to assay the ability of AaIT to cause immediate and fast contraction paralysis. The assay was performed as described by Zlotkin et al. (1971a) and resulted in the determination of the contraction paralysis unit (CPU), which is defined as the mass of injected material per 100 mg of body mass inducing an immediate and transient paralysis in 50 % of the test larvae. In the present study, this bioassay was employed to monitor the activity of the neurotoxin in the hemolymph of the insects. The paralytic and lethal potencies of AaIT and of the crude venom to tenebrionid beetles were determined by a subcuticular injection of various doses through an intersegmental membrane from the ventral side of the hind abdomen. To prevent excessive bleeding, a patch of melted paraffin was placed over the site of injection. Each dose was tested on 5–7 individuals. Paralysis, defined as the inability of the animal to move or turn, was determined 15 min after injection for both Trachyderma philistina and Adesmia sp. Lethality was determined 5 days after injection. The paralytic (PD50) and lethal (LD50) toxicities were determined according to the method of Reed and Muench (1938).

Radioiodination

The toxin was iodinated using Iodogen (Pierce Chem. Co., Rockland, USA) according to the method described by De Lima et al. (1989), using 0.5 mCi of carrier-free [125I]Na (Nuclear Research Center, Negev, Israel) and 5 μg of toxin. The monoiodotoxin was purified as described by Gordon et al. (1992) on an Ultrapore C3 column (Beckman) using a gradient of 10 % to 80 % B, where A is 0.1 % trifluoroacetic acid (TFA) and B is isopropanol/acetonitrile 1:1 in 0.1 % TFA, at flow rate of 0.5 ml min−1. The concentration of the radiolabeled toxin was determined from the specific radioactivity of the 125I and corresponded to 2424 disints min−1 fmol−1 monoiodotoxin.

Neuronal membranes

The central nervous system (CNS) consisting of brain, subesophageal ganglion and the entire ventral nerve cord, was dissected, homogenized and then treated according to the previously published procedure for the preparation of locust synaptosomes (Breer and Jeserich, 1980; Zlotkin and Gordon, 1985). The neuronal membranes employed in this study correspond to the so called ‘P2’ preparation, which is the pellet obtained after fast centrifugation (27 000 g, 40 min, 4 °C) of the supernatant derived from the first slow centrifugation (1500 g, 10 min, 4 °C) of the crude homogenate (Zlotkin and Gordon, 1985). The medium in which the dissection and the processing of the nervous tissue were performed was composed of 0.25 mol l−1 mannitol, 10 mmol l−1 EDTA, pH 7.4, in the presence of the following proteolytic inhibitors: phenylmethylsulfonyl fluoride (50 μg ml−1), 1 μmol l−1 pepstatin A, 1 mmol l−1 iodoacetamide and 1 mmol l−1 1,10-phenanthroline.

Binding assays

Equilibrium saturation assays were performed with the tenebrionid beetle neuronal membranes using increasing concentrations of the unlabeled toxin in the presence of a constant low concentration of the labeled toxin. The specific radioactivity and the amount of bound toxin were calculated and determined for each toxin concentration. Kinetic dissociation binding assays were performed using the dilution method (Zlotkin and Gordon, 1985). Neuronal membranes of various insects (15–25 μg protein per reaction mixture) were incubated with 0.6–0.7 nmol l−1 [125I]AaIT at 18 °C for 60 min to reach equilibrium, followed by a 100-fold dilution and collection of samples at different time intervals.

Insect neuronal membranes were suspended in binding medium containing 0.13 mol l−1 choline chloride, 1 mmol l−1 EDTA, 20 mmol l−1 Hepes–Tris, pH 7.4, and 5 mg ml−1 bovine serum albumin (BSA). Following a 1 h incubation, the reaction mixture was diluted with 2 ml or 5 ml (in dissociation assays) of ice-cold wash medium (150 mmol l−1 choline chloride, 5 mmol l−1 Hepes–Tris, pH 7.4, 1 mmol l−1 EDTA and 5 mg ml−1 BSA) and filtered over GF/F filters (Whatman, UK) under vacuum followed by two more washes of the filters. Non-specific binding was determined in the presence of 1 μmol l−1 unlabeled toxin and corresponded to 20–40 % of the total binding.

Analysis of the binding assays was performed using the iterative computer program LIGAND (P. J. Munson and D. Rodbard, modified by G. A. McPherson, 1985). Kinetic experiments were analyzed according to Weiland and Molinoff (1981). Each experiment was performed three times.

Protein determination

Protein content was determined using a procedure modified from that of Lowry et al. (1951) using BSA as the standard.

Column chromatography

Gel filtration chromatography was performed with Ultrogel AcA 202 gel (IBF Reactifs, Villeneuve, France). The gel was washed and equilibrated with the running buffer, packed in a column and employed in conditions specified in the legend to Fig. 4.

Hemolymph volume determination

Hemolymph volume was determined according to the method of Wharton et al. (1965) using a radioactive tracer (inulin [14C]carboxylic acid). The hemolymph volume was determined by injecting 2×105 cts min−1 of inulin [14C]carboxylic acid in a volume of 5 μl per insect. After 40 min, 10 μl of hemolymph was collected and its radioactivity was counted in a scintillation counter. Hemolymph volume was estimated according to the method described by Herrmann et al. (1990).

Light microscope autoradiography

Incubation (Figs 1, 3A) and injection (Fig. 3B,C) assays were performed. In the incubation assays, dissected tissue was incubated with the radioiodinated toxin in the appropriate Ringer (Fig. 1) either in the absence (total binding) or the presence (non-specific binding) of a high concentration (1 μmol l−1) of unlabeled AaIT. Further details are presented in the legend to Fig. 1. In the injection assays, insects were injected with [125I]AaIT (see legend to Fig. 3). When they revealed early transient symptoms of paralysis, the insects were immediately fixed by rapid perfusion of the fixative through their body cavity. Neuronal and muscle tissues were dissected from the experimental insects and further treated for microscope autoradiography. Dissection, fixation and embedding were performed according to Fishman et al. (1991).

Fig. 1.

Autoradiography of [125I]AaIT (3 nmol l−1) in the dissected and desheathed connectives of the abdominal ventral nerve cord of Trachyderma philistina in an incubation experiment. In the experiment, the dissected tissue is incubated in the appropriate Ringer (NaCl, 200 mmol l−1; KCl 3.1 mmol l−1; CaCl2, 5.0 mmol l−1; glucose, 110 mmol l−1, pH 7.2, 550 osmol l−1) in the absence (total binding) or presence (non-specific binding) of a high concentration (1 μmol l−1) of the unlabeled toxin. Binding specificity was thus expressed as the displaceability of the labeled toxin. (A) Radioiodinated toxin only. (B) Radioiodinated toxin in the presence of 1 μmol l−1 unlabeled toxin. The difference in the density of photographic grains in A compared with B reveals the occurrence of specific binding of the toxin in the nanomolar range. a, axon; NS, neural sheath. Scale bar, 10 μm.

Fig. 1.

Autoradiography of [125I]AaIT (3 nmol l−1) in the dissected and desheathed connectives of the abdominal ventral nerve cord of Trachyderma philistina in an incubation experiment. In the experiment, the dissected tissue is incubated in the appropriate Ringer (NaCl, 200 mmol l−1; KCl 3.1 mmol l−1; CaCl2, 5.0 mmol l−1; glucose, 110 mmol l−1, pH 7.2, 550 osmol l−1) in the absence (total binding) or presence (non-specific binding) of a high concentration (1 μmol l−1) of the unlabeled toxin. Binding specificity was thus expressed as the displaceability of the labeled toxin. (A) Radioiodinated toxin only. (B) Radioiodinated toxin in the presence of 1 μmol l−1 unlabeled toxin. The difference in the density of photographic grains in A compared with B reveals the occurrence of specific binding of the toxin in the nanomolar range. a, axon; NS, neural sheath. Scale bar, 10 μm.

Fast reversible paralysis of Trachyderma philistina with high doses of AaIT

The paralytic and lethal effects of the crude venom of A. australis and its derived AaIT toxin on several insect species are presented in Table 1. The tolerance of Trachyderma philistina to the toxin was revealed by the high PD50 dose (Table 1) and the reversibility of the effect of the toxin. Reversal of the toxic effects is demonstrated in Table 2, where the response of Trachyderma philistina to the crude venom as a function of time is compared with that of Adesmia sp. In contrast to Adesmia, where the paralysis deepens with time (decreasing the PD50 values), Trachyderma philistina shows a reversal of paralysis with a time-dependent increase in the PD50 values. Doses below 10 μg of AaIT per 100 mg of body mass did not induce any form of fast or delayed paralysis (Table 1) in Trachyderma philistina.

Table 1.

The paralytica (PD50) and lethal (LD50) effects of Androctonus australis scorpion venom and AaIT, derived from the crude venom, on a variety of insect species

The paralytica (PD50) and lethal (LD50) effects of Androctonus australis scorpion venom and AaIT, derived from the crude venom, on a variety of insect species
The paralytica (PD50) and lethal (LD50) effects of Androctonus australis scorpion venom and AaIT, derived from the crude venom, on a variety of insect species
Table 2.

Time-dependence of the paralytic effect Androctonus australis scorpion venom on the tenebrionid beetles Trachyderma philistina and Adesmia sp.

Time-dependence of the paralytic effect Androctonus australis scorpion venom on the tenebrionid beetles Trachyderma philistina and Adesmia sp.
Time-dependence of the paralytic effect Androctonus australis scorpion venom on the tenebrionid beetles Trachyderma philistina and Adesmia sp.

From the available data, the various insects listed in Table 1 can be classified into three categories. The first are susceptible insects (flies, cockroaches, locusts), which respond to relatively low doses with a fast and complete paralysis. The second category, represented by the lepidopterous larvae (Herrmann et al. 1990), consists of insects resistant to toxin which show a slow and delayed paralysis in response to high doses. Trachyderma philistina, in its response to the crude venom and to AaIT, constitutes a third category. In response to very high doses, the beetles reveal the typical fast paralysis which is, however, reversible (Table 2). These aspects of dosage and reversibility are clarified in the subsequent sections of the study.

High-affinity binding of [125I]AaIT to Trachyderma philistina neuronal membranes

The occurrence of high-affinity binding of [125I]AaIT to Trachyderma philistina neuronal membranes was demonstrated by light microscope autoradiography of a mechanically exposed connectives of the insect CNS (Fig. 1) and by the equilibrium-saturation binding assays (Scatchard analysis) (Fig. 2). The binding assays showed the presence of two binding sites in three independent assays which gave mean values of KD1= 0.43±0.075 nmol l−1, Bmax1=0.08±0.35 pmol mg−1 protein and KD2=34.5±4.32 nmol l−1, Bmax2=1.44±0.35 pmol mg−1 protein (means ± S.D.). The binding constants of site 1 resemble those obtained using AaIT on membrane preparation from other susceptible species (De Lima et al. 1989).

Fig. 2.

Scatchard analysis of a representative saturation curve of AaIT binding to Trachyderma philistina neuronal membranes. Neuronal membranes (20.8 μg of protein) were incubated with 0.1 nmol l−1 [125I]AaIT and increasing concentrations of unlabeled AaIT for 50 min at 18 °C in 0.4 ml of binding buffer. The amount of specifically bound [125I]AaIT was determined following rapid filtration (see Materials and methods). Non-specific binding (20–40 % of total binding) was determined in the presence of 1 μmol l−1 AaIT and was subtracted from all data points. The specific radioactivity and the amount of bound toxin were recalculated for each toxin concentration and the Scatchard plot was drawn using the computer program LIGAND. Scatchard plot analysis yielded the best fit (P<0.05) using a two-binding-site model. The values of the binding constants obtained in this experiment correspond to KD1=0.5 nmol l−1; Bmax1=0.07 pmol mg−1 protein and KD2=28.6 nmol l−1, Bmax2= 1.15 pmol mg−1 protein.

Fig. 2.

Scatchard analysis of a representative saturation curve of AaIT binding to Trachyderma philistina neuronal membranes. Neuronal membranes (20.8 μg of protein) were incubated with 0.1 nmol l−1 [125I]AaIT and increasing concentrations of unlabeled AaIT for 50 min at 18 °C in 0.4 ml of binding buffer. The amount of specifically bound [125I]AaIT was determined following rapid filtration (see Materials and methods). Non-specific binding (20–40 % of total binding) was determined in the presence of 1 μmol l−1 AaIT and was subtracted from all data points. The specific radioactivity and the amount of bound toxin were recalculated for each toxin concentration and the Scatchard plot was drawn using the computer program LIGAND. Scatchard plot analysis yielded the best fit (P<0.05) using a two-binding-site model. The values of the binding constants obtained in this experiment correspond to KD1=0.5 nmol l−1; Bmax1=0.07 pmol mg−1 protein and KD2=28.6 nmol l−1, Bmax2= 1.15 pmol mg−1 protein.

AaIT is accessible to the terminal branches of the motor nerves in Trachyderma philistina

Since the nervous system of Trachyderma philistina has been shown to possess high-affinity binding sites, we decided to examine whether the tolerance of Trachyderma philistina is a consequence of a barrier preventing access of the toxin to the motor nerves. This possibility was examined by using light microscope autoradiography in both incubation (Fig. 3A) and injection (Fig. 3B,C) assays. As previously shown with other insects (Herrmann et al. 1990; Fishman et al. 1991), the anatomically intact central nervous system (CNS) of Trachyderma philistina is impermeable to AaIT (data not shown). However, in both species of beetle, the peripheral branches of the motor nerves in close vicinity to the skeletal muscles are labeled by the toxin (Fig. 3).

Fig. 3.

The binding of AaIT to peripheral branches of motor nerves visualized by light microscope autoradiography. (A) Incubation experiment with dissected nerves of Trachyderma philistina. In this assay, the dissected tissue was incubated in 3 nmol l−1 [125I]AaIT. Note the accumulation of photographic grains on the axon (N) but not in the skeletal muscle (M) and trachea (T). (B) Injection experiment in which [125I]AaIT was injected into the body cavity of Trachyderma philistina to a final concentration of 110 nmol l−1. Beetles which showed transient disturbances in their leg movements were then selected and processed. The figure shows a peripheral branch of a motor nerve densely labeled with the radioiodinated toxin. (C) The same injection procedure as in B with an Adesmia sp. beetle, again showing dense labeling of a motor nerve. Scale bar, 10 μm.

Fig. 3.

The binding of AaIT to peripheral branches of motor nerves visualized by light microscope autoradiography. (A) Incubation experiment with dissected nerves of Trachyderma philistina. In this assay, the dissected tissue was incubated in 3 nmol l−1 [125I]AaIT. Note the accumulation of photographic grains on the axon (N) but not in the skeletal muscle (M) and trachea (T). (B) Injection experiment in which [125I]AaIT was injected into the body cavity of Trachyderma philistina to a final concentration of 110 nmol l−1. Beetles which showed transient disturbances in their leg movements were then selected and processed. The figure shows a peripheral branch of a motor nerve densely labeled with the radioiodinated toxin. (C) The same injection procedure as in B with an Adesmia sp. beetle, again showing dense labeling of a motor nerve. Scale bar, 10 μm.

The toxin in the body fluids of the beetle

Hemolymph volume

Assays of the recovery of radioactivity and levels of toxicity in the hemolymph demand the estimation of the insect’s hemolymph volume. Hemolymph volume was determined in three separate assays (not shown) on six beetles each, resulting in closely similar values with an overall mean of 17.5±2.9 mg 100 mg−1 body mass (v/w) (±S.D., N=18).

Radioactivity in an intact animal

The amount of radioactivity in intact Trachyderma philistina was measured immediately and 1, 6 and 24 h after injection of [125I]AaIT into three beetles for each time interval by placing the insect in a vial and introducing the vial into a gamma counter. After 24 h, 80 % of the injected radioactivity remains in the body of the beetle, in contrast to lepidopterous larvae where the majority of the radioactivity is excreted within this period (Herrmann et al. 1990).

Toxicity and radioactivity in hemolymph

In order to follow the fate of the toxin in the body fluid of the insect, we injected a mixture (12 μl per insect) of the radioiodinated (28 ng, 921×103 cts min−1) and unlabeled toxin (8 μg) and then removed measured volumes (10–20 μl per insect) of hemolymph at 0.5, 1, 6 and 24 h after injection. Four beetles were used for each time interval. Hemolymph was collected only once from each beetle. Each hemolymph sample was diluted with 0.65 % saline and assayed for radioactivity, toxicity to blowfly larvae (Table 3) and chromatographic behavior on a molecular exclusion column to estimate the degree of degradation (Table 3). The chromatographic separation of hemolymph shown in Fig. 4 reveals that the radioiodinated toxin can be separated into three radioactive fractions. The first (peak I) represents a dimeric form of the toxin, the second (peak II) corresponds to the intact and active portion of the toxin (Herrmann et al. 1990), and the third (peak III) fraction includes free iodine or low-Mr degradation products of the toxin. The toxin is degraded in the hemolymph in a time-dependent manner as shown by the progressive increase in the size of peak III. To summarize, the data presented in Fig. 4 and Table 3 reveal that the gradual progressive loss of toxicity from the hemolymph of a beetle injected with AaIT occurs concomitantly with its elimination and degradation.

Table 3.

Recoveries of toxicity, radioactivity and degree of degradation of [125I]AaIT in the hemolymph of Trachyderma philistina

Recoveries of toxicity, radioactivity and degree of degradation of [125I]AaIT in the hemolymph of Trachyderma philistina
Recoveries of toxicity, radioactivity and degree of degradation of [125I]AaIT in the hemolymph of Trachyderma philistina
Fig. 4.

Chromatographic separation of hemolymph. The figure depicts the elution pattern of radioactivity measured from fractions of hemolymph collected from beetles injected with radioiodinated (28 ng, 921×103 cts min−1) and unlabeled (8 μg) AaIT. The hemolymph was separated on a molecular exclusion Ultragel AcA 202 column (63 cm ×1.1 cm), equilibrated and eluted with 0.2 mg ml−1 bovine serum albumin and 0.02 % NaN3 at a flow rate of 3.2 ml h−1. Fractions of 0.5 ml each were collected, and radioactivity was counted in a gamma counter. Elution of a freshly radioiodinated toxin (○) was followed by elution of hemolymph samples collected 30 min (□), 6 h (▴) and 24 h (♦) after injection. I, II and III are the three peaks of radioactivity observed.

Fig. 4.

Chromatographic separation of hemolymph. The figure depicts the elution pattern of radioactivity measured from fractions of hemolymph collected from beetles injected with radioiodinated (28 ng, 921×103 cts min−1) and unlabeled (8 μg) AaIT. The hemolymph was separated on a molecular exclusion Ultragel AcA 202 column (63 cm ×1.1 cm), equilibrated and eluted with 0.2 mg ml−1 bovine serum albumin and 0.02 % NaN3 at a flow rate of 3.2 ml h−1. Fractions of 0.5 ml each were collected, and radioactivity was counted in a gamma counter. Elution of a freshly radioiodinated toxin (○) was followed by elution of hemolymph samples collected 30 min (□), 6 h (▴) and 24 h (♦) after injection. I, II and III are the three peaks of radioactivity observed.

Kinetics of dissociation

The accessibility and high affinity of the nerves of Trachyderma philistina to the AaIT toxin (Figs 2, 3) have further emphasized the problem of how toxin tolerance is achieved. Therefore, it was decided to examine the stability of the complex between the toxin and its neuronal receptor in a series of binding-dissociation assays. To measure the dissociation constant (k−1), [125I]AaIT was incubated with the neuronal preparations until steady-state conditions were reached, and this was followed by a 100-fold dilution of the toxin. The amount of specifically bound radioligand was measured at various times following the dilution. Under these conditions, by eliminating the ligand (L)–receptor (R) forward reaction (L+R⟶LR) the first-order integrated rate equation for dissociation is (Weiland and Molinoff, 1981):
formula
where Bt is the concentration of bound ligand at time t after initiation of the dissociation and B0 is the concentration of bound ligand at time zero just prior to the dilution. If the reaction is fully reversible, the slope of the first-order plot will be equal to k−1. The half-time (t1/2) of the ligand–receptor complex (LR) is calculated as ln(2/k−1) (Bennet, 1978).

For the dissociation studies, we chose five insect species that revealed different degrees of susceptibility or tolerance to AaIT toxin. The data are presented in Fig. 5 and Table 4, and they show an inverse relationship between the duration of the dissociation and the degree of tolerance to the toxin (see Discussion).

Table 4.

Paralytic potency and dissociation constants of AaIT in various insects and their respective neuronal membranes

Paralytic potency and dissociation constants of AaIT in various insects and their respective neuronal membranes
Paralytic potency and dissociation constants of AaIT in various insects and their respective neuronal membranes
Fig. 5.

Time course of dissociation of [125I]AaIT from five insect neuronal preparations by the dilution method (see Results). The figure shows the analysis of the dissociation curves by linearization according to equation 1 (see text as above). The values of the dissociation constants (Table 4) were determined directly from the slope and the half-time was calculated accordingly. ○ Spodoptera littoralis larvae; ▴ Adesmia sp.; ♦ Periplaneta americana; •Trachyderma philistina; ▵. Locusta migratoria.

Fig. 5.

Time course of dissociation of [125I]AaIT from five insect neuronal preparations by the dilution method (see Results). The figure shows the analysis of the dissociation curves by linearization according to equation 1 (see text as above). The values of the dissociation constants (Table 4) were determined directly from the slope and the half-time was calculated accordingly. ○ Spodoptera littoralis larvae; ▴ Adesmia sp.; ♦ Periplaneta americana; •Trachyderma philistina; ▵. Locusta migratoria.

It has been shown by Herrmann et al. (1990) that lepidopterous larvae have an unusual tolerance of AaIT and only develop a delayed and slow progressive paralysis in response to relatively high doses of toxin (PD50= 2.4 μg 100 mg−1 body mass for Spodoptera littoralis larvae within 24–48 h). Such tolerance was largely attributed to pharmacokinetic aspects such as accessibility barriers and degradation processes (Herrmann et al. 1990). The present study was initiated by the observation of a considerable tolerance to AaIT by the tenebrionid beetle Trachyderma philistina, which has previously been shown to tolerate high doses of scorpion venom (Israeli-Zindel et al. 1970). The beetle manifests a fast but reversible paralysis induced by extremely high doses of the toxin (>10 μg 100 mg−1).

The rapidity of paralysis is attributed to (a) the existence of toxin-specific high-affinity binding sites on neuronal membranes, as demonstrated by light microscope autoradiography (Fig. 1) and binding assays (Fig. 2), and (b) the obvious accessibility of the toxin to the terminal branches of the motor nerves (Fig. 3). These aspects of accessibility and high-affinity binding have been demonstrated in the toxin-susceptible Periplaneta americana (Fishman et al. 1991).

The tolerance of Trachyderma philistina, reflected by the necessity for high doses of toxin and the reversibility in its action, can be explained at the pharmacokinetic as well as the pharmacodynamic level of the intoxication process. The pharmacokinetic level is shown in the degradation and elimination of the toxin from the hemolymph. Such elimination is attributed to binding of the toxin or complexation to pharmacologically irrelevant tissues. The data presented in Table 3 and Fig. 4 reveal that there is a close agreement between the recovery of toxicity from the hemolymph (%CPU, Table 3) and the appearance of chromatographically intact toxin (fraction II, Fig. 4, Table 3), suggesting that the chromatographically intact toxin is also the pharmacologically functional one.

The phenomenon of complexation is also demonstrated by a unique expression in the present study, in nervous tissue of Trachyderma philistina, of a second class of binding site with a relatively high capacity (Bmax=1.44 pmol mg−1) and low affinity (KD=34.5 nmol l−1) which may compete with the low-capacity (Bmax=0.08 pmol mg−1) and high-affinity (KD=0.43 nmol l−1) sites for the toxin. The high-affinity, low-capacity sites are the ones that are involved in the neurotoxic effect (De Lima et al. 1989; Gordon et al. 1992).

There is a correlation between the time course of reversibility of the paralysis and the decrease in the amount of active toxin in the hemolymph. The data presented in Table 2 indicate that recovery begins to occur approximately 6 h after the injection when only approximately 4 % of the active toxin is present in the hemolymph (Table 3).

On the basis both of the present and previous data (Herrmann et al. 1990), there is no doubt that tolerance in an insect to a neurotoxic polypeptide includes factors working at the pharmacokinetic level which are similar to the detoxification and elimination processes described for insecticide resistance (Oppenoorth, 1985; Welling and Paterson, 1985). However, we also demonstrate that factors at the pharmacodynamic level are involved in tolerance to AaIT, as shown by the dissociation binding assays. The data presented in Fig. 5 and Table 4 clearly show that the tolerance is closely correlated with the stability of the toxin–receptor complex. Thus, in susceptible insects such as the cockroaches and locusts, the half-time (t1/2) of dissociation is several minutes while in tolerant insects such as lepidopterous larvae and Trachyderma philistina this is reduced to several seconds (Table 4; Fig. 5). In conditions of such fast dissociation, the persistence of the effect demands a constant high concentration of toxin at the vicinity of target sites. However, in the case of Trachyderma philistina, such high concentrations cannot be maintained because of the above-mentioned processes of complexation, degradation and elimination.

To summarize, as shown in our present and previous studies (Herrmann et al. 1990; Fishman et al. 1991), it appears that, in insects, sensitivity to a neurotoxic polypeptide is determined (a) by the accessibility of the toxin to the motor neuron terminals and to degradation processes and (b) by the binding affinity and binding stability of the toxin to its neuronal receptor. It has previously been established that the neuronal receptor of AaIT is the voltage-dependent insect Na+ channel (Gordon et al. 1992; Moskowitz et al. 1994).

Three major mechanisms of resistance to toxic chemicals have been identified in insects: increased detoxification, insensitive target sites (loss or absence of receptors) and decreased penetration (Mullin and Scott, 1992). The present data reveal the occurrence of an additional mechanism of resistance, namely receptor destabilization. The ability of an insect to withstand high doses of a venom neurotoxin seems to be pre-existing and preadaptive rather than a developed feature (as in the case of insecticide resistance) and, therefore, this ability should be defined as tolerance (Oppenoorth, 1985).

Against this background, we can assume that the marked differences in the toxin–receptor dissociation constants of the tolerant versus susceptible insects represent differences in the structure and function of their Na+ channels. It has been shown (Osborne and Pepper, 1992) that resistance to pyrethroids is related to a modification of the affinity of receptor sites on the voltage-dependent Na+ channels. Therefore, it appears that the insect-selective neurotoxin may be used (1) as a pharmacological tool for probing and monitoring changes and subtypes of insect Na+ channels, and (2) as a target for structural modifications in order to overcome possible mechanisms of tolerance by insect pests.

This study was supported by grants 1S-2465-94 from the United States–Israel Binational Agricultural Research and Development Fund (BARD) and 0375-172.01 from the German–Israeli Foundation for Scientific Research (GIF). The editorial assistance of Dr Michel Hamelin (Merck & Co, Inc. Research Laboratories, Rahway, NJ, USA) is greatly appreciated.

Bennet
,
J. P.
, Jr
(
1978
).
Methods in binding studies
. In
Neurotransmitter Receptor Binding
(ed.
H. I.
Yamamura
), pp.
57
90
.
New York
:
Raven Press
.
Breer
,
H.
and
Jeserich
,
G.
(
1980
).
A microscale flotation technique for the isolation of synaptosomes from the nervous tissue of Locusta migratoria
.
Insect Biochem
.
10
,
457
463
.
Cory
,
J. S.
,
Hirst
,
M. L.
,
Wiliams
,
T.
,
Halls
,
R. S.
,
Goulsan
,
D.
,
Green
,
B. M.
,
Carty
,
T. M.
,
Possee
,
R. D.
,
Cayley
,
P. J.
and
Bishop
,
H. L.
(
1994
).
Field trial of a genetically improved baculovirus insecticide
.
Nature
370
,
138
140
.
De Dianous
,
S.
,
Hoaran
,
F.
and
Rochat
,
H.
(
1987
).
Reexamination of the specificity of the scorpion Androctonus australis Hector insect toxin towards arthropods
.
Toxicon
25
,
411
417
.
De Lima
,
M. E.
,
Martin-Eauclaire
,
M. F.
,
Hue
,
B.
,
Loret
,
E.
,
Diniz
,
C. R.
and
Rochat
,
H.
(
1989
).
On the binding of two scorpion toxins to the central nervous system of the cockroach Periplaneta americana
.
Insect Biochem
.
19
,
413
422
.
Fishman
,
L.
,
Kagan
,
M. L.
and
Zlotkin
,
E.
(
1991
).
Accessibility of the insect nervous system to a neurotoxic polypeptide
.
J. exp. Zool
.
257
,
10
23
.
Gordon
,
D.
,
Moskowitz
,
H.
,
Eitan
,
M.
,
Warner
,
C.
,
Catterall
,
W. A.
and
Zlotkin
,
E.
(
1992
).
Localization of receptor sites for insect selective toxins on sodium channels by site directed antibodies
.
Biochemistry
31
,
7622
7628
.
Herrmann
,
R.
,
Fishman
,
L.
and
Zlotkin
,
E.
(
1990
).
The tolerance of lepidopterous larvae to an insect selective neurotoxin
.
Insect Biochem
.
20
,
625
637
.
Israeli-Zindel
,
I.
,
Zlotkin
,
E.
and
Shulov
,
A.
(
1970
).
The resistance of insects to the venom of the yellow scorpion Leiurus quinquestriatus
. In
Toxins of Animal and Plant Origin
, vol.
3 (ed.A. de Vries and E. Kochva
), pp.
933
939
. New York, London, Paris: Gordon and Breach Science Publ.
Lowry
,
O. H.
,
Rosebrough
,
N. J.
,
Farr
,
A. L.
and
Randall
,
R. H.
(
1951
).
Protein measurements with the folin reagent
.
J. biol. Chem
.
193
,
265
275
.
Mccutchen
,
B. F.
,
Choudray
,
P. V.
,
Crenshaw
,
R.
,
Maddox
,
D.
,
Kamita
,
S. G.
,
Palekar
,
N.
,
Volrath
,
S.
,
Fowler
,
E.
,
Hammock
,
B.D.
and
Maeda
,
S.
(
1991
).
Development of a recombinant baculovirus expressing an insect-selective neurotoxin: Potential for pest control
.
Bio/Technology
9
,
848
852
.
Moskowitz
,
H.
,
Herrmann
,
R.
,
Zlotkin
,
E.
and
Gordon
,
D.
(
1994
).
Variability among insect sodium channels revealed by selective neurotoxins
.
Insect Biochem. molec. Biol
.
24
,
13
19
.
Mullin
,
C. A.
and
Scott
,
J. G.
(
1992
).
Molecular Mechanisms of Insecticide Resistance
.
Washington, DC
:
American Chemical Society
.
Oppenoorth
,
F. J.
(
1985
).
Biochemistry and genetics of insecticide resistance
. In
Comprehensive Insect Physiology, Biochemistry and Pharmacology
, vol.
12
(ed.
G. A.
Kerkut
and
L. I.
Gilbert
), pp.
731
773
.
Oxford
:
Pergamon Press
.
Osborne
,
M. P.
and
Pepper
,
D. R.
(
1992
).
Mechanisms of kdr and Super kdr resistance
. In
Molecular Mechanisms of Insecticide Resistance
(ed.
C. A.
Mullin
and
J. G.
Scott
), pp.
71
89
.
Washington, DC
:
American Chemical Society
.
Reed
,
L. J.
and
Muench
,
S.
(
1938
).
A simple method of estimating fifty percent endpoint
.
Am. J. Hyg
.
27
,
493
497
.
Stewart
,
L. M. D.
,
Hirst
,
M.
,
Ferber
,
M. L.
,
Merryweather
,
A. T.
,
Cayley
,
P. J.
and
Possee
,
R. D.
(
1991
).
Construction of an improved baculovirus insecticide containing an insect-specific toxin gene
.
Nature
352
,
85
88
.
Walther
,
C.
,
Zlotkin
,
E.
and
Rathmayer
,
W.
(
1976
).
Action of different toxins from the scorpion Androctonus australis on a locust nerve–muscle preparation
.
J. Insect Physiol
.
22
,
1187
1194
.
Weiland
,
G. A.
and
Molinoff
,
P. B.
(
1981
).
Quantitative analysis of drug–receptor interactions. I. Determination of kinetic and equilibrium properties
.
Life Sci
.
29
,
313
330
.
Welling
,
W.
and
Paterson
,
G. D.
(
1985
).
Toxicodynamics of insecticides
. In
Comprehensive Insect Physiology, Biochemistry and Pharmacology
, vol.
12
(ed.
G. A.
Kerkut
and
L. I.
Gilbert
), pp.
604
645
.
Oxford
:
Pergamon Press
.
Wharton
,
D. R. A.
,
Wharton
,
M. L.
and
Lola
,
J.
(
1965
).
Blood volume and water content of the male American Cockroach Periplaneta americana – methods and influence of age and starvation
.
J. Insect Physiol
.
11
,
391
404
.
Zlotkin
,
E.
(
1991
).
Venom neurotoxins – models for selective insecticides
.
Phytoparasitica
19
,
177
182
.
Zlotkin
,
E.
,
Fishman
,
L.
and
Shapiro
,
J. P.
(
1992
).
Oral toxicity to fleshflies of a neurotoxic polypeptide
.
Archs Insect Biochem. Physiol
.
21
,
41
52
.
Zlotkin
,
E.
,
Fraenkel
,
G.
,
Miranda
,
F.
and
Lissitzky
,
S.
(
1971a
).
The effect of scorpion venom on blowfly larvae; a new method for the evaluation of scorpion venom potency
.
Toxicon
9
,
1
8
.
Zlotkin
,
E.
and
Gordon
,
D.
(
1985
).
Detection, purification and receptor binding assays of insect selective neurotoxins derived from scorpion venoms
. In
Neurochemical Techniques in Insect Research
(ed.
H.
Breer
and
T. A.
Miller
), pp.
243
295
. Heidelberg: Springer.
Zlotkin
,
E.
,
Rochat
,
H.
,
Kupeyan
,
C.
,
Miranda
,
F.
and
Lissitzky
,
S.
(
1971b
).
Purification and properties of the insect toxin from the venom of the scorpion Androctonus australis Hector
.
Biochimie, Paris
53
,
1073
1078
.