点击显示 收起
【关键词】 Mammalian
Among scorpion - and -toxins that modify the activation and inactivation of voltage-gated sodium channels (Navs), depressant -toxins have traditionally been classified as anti-insect selective on the basis of toxicity assays and lack of binding and effect on mammalian Navs. Here we show that the depressant -toxins LqhIT2 and Lqh-dprIT3 from Leiurus quinquestriatus hebraeus (Lqh) bind with nanomolar affinity to receptor site 4 on rat skeletal muscle Navs, but their effect on the gating properties can be viewed only after channel preconditioning, such as that rendered by a long depolarizing prepulse. This observation explains the lack of toxicity when depressant toxins are injected in mice. However, when the muscle channel rNav1.4, expressed in Xenopus laevis oocytes, was modulated by the site 3 -toxin LqhIT, LqhIT2 was capable of inducing a negative shift in the voltage-dependence of activation after a short prepulse, as was shown for other -toxins. These unprecedented results suggest that depressant toxins may have a toxic impact on mammals in the context of the complete scorpion venom. To assess whether LqhIT2 and Lqh-dprIT3 interact with the insect and rat muscle channels in a similar manner, we examined the role of Glu24, a conserved "hot spot" at the bioactive surface of -toxins. Whereas substitutions E24A/N abolished the activity of both LqhIT2 and Lqh-dprIT3 at insect Navs, they increased the affinity of the toxins for rat skeletal muscle channels. This result implies that depressant toxins interact differently with the two channel types and that substitution of Glu24 is essential for converting toxin selectivity.
Voltage-gated sodium channels (Navs) are critical in generation and propagation of action potentials in excitable cells and are targeted by a large variety of chemically distinct compounds that bind at several receptor sites on the poreforming -subunit (Gordon, 1997; Catterall, 2000). Most lipid-soluble Nav activators, including pyrethroid insecticides, toxic alkaloids (e.g., veratridine and batrachotoxin), and marine cyclic polyether toxins (e.g., brevetoxins), affect Navs of both insects and mammals. Yet, despite the general conservation of Nav structure, certain scorpion toxins show preference for Nav subtypes in mammals or insects (Gordon et al., 1998, 2007; Cestèle and Catterall, 2000; Gurevitz et al., 2007), which raised the idea of using some representatives for insect pest control (reviewed in Gurevitz et al., 2007).
Scorpion toxins that modulate Nav gating are divided between the and classes according to their mode of action and binding features to distinct receptor sites (Catterall, 2000). -Toxins prolong the action potential by inhibiting the fast inactivation of Navs upon binding to receptor site 3 (e.g., LqhIT from Leiurus quinquestriatus hebraeus; Eitan et al., 1990; Martin-Eauclaire and Couraud, 1995; Gordon et al., 1996) assigned mainly to extracellular loops in domains 1 and 4 (Catterall, 2000). -Toxins shift the voltage dependence of channel activation to more hyperpolarized membrane potentials upon binding to receptor site 4, assigned mainly to external loops in domains 2 and 3 (Marcotte et al., 1997; Cestèle et al., 1998, 2006; Shichor et al., 2002; Leipold et al., 2006). The -toxins are further classified to 1) anti-mammalian -toxins (e.g., Css2 and Css4 from Centruroides suffusus suffusus; Martin-Eauclaire and Couraud, 1995; Gordon et al., 1998; Gurevitz et al., 2007); 2) -toxins that affect both insect and mammalian Navs (e.g., Ts1 from Tityus serrulatus and Lqh1; Possani et al., 1999; Gordon et al., 2003); 3) Anti-insect selective excitatory -toxins (e.g., AahIT from Androctonus australis hector and Bj-xtrIT from Buthotus judaicus), typified by the symptoms of contraction paralysis they produce in blowfly larvae (Zlotkin et al., 1978; Froy et al., 1999), and by their unique structures (Oren et al., 1998; Li et al., 2005; Gurevitz et al., 2007); and 4) anti-insect depressant toxins, which upon injection to blowfly larvae induce flaccid paralysis as a result of sustained depolarization of the axonal membrane, leading to block in the evoked action potentials and loss of muscle tonus (Lester et al., 1982; Zlotkin et al., 1991; Ben Kalifa et al., 1997; Strugatsky et al., 2005). These 61-residue polypeptides were classified as -toxins because of their ability to modulate Nav activation and to compete with excitatory toxins on binding to receptorsite 4 in insect neuronal membranes (Gordon et al., 1984, 1992). Yet depressant toxins did not compete with anti-mammalian -toxins on binding to rat brain membranes and were harmless when injected to mice (Lester et al., 1982; Herrmann et al., 1995; Strugatsky et al., 2005).
The bioactive surfaces of the anti-insect excitatory -toxin Bj-xtrIT, the anti-mammalian -toxin Css4, and the anti-insect depressant -toxin LqhIT2 have been described previously (Cohen et al., 2004, 2005; Karbat et al., 2007). These studies highlighted a conserved "pharmacophore" composed of a key negatively charged Glu in the -helix (Fig. 1), flanked by hydrophobic residues that may isolate the point of interaction with a counterpart channel residue from the bulk solvent. An additional hydrophobic cluster of bioactive residue, at the C-tail of Bj-xtrIT and on the loop connecting the second and third -strands of Css4, was suggested to determine toxin selectivity (Cohen et al., 2004, 2005). Although the key residues involved in Css4 and Bj-xtrIT activity form two topologically distinct domains, the bioactive surface of LqhIT2 is continuous but involves the conserved pharmacophore (Glu24) and its vicinity (Karbat et al., 2007). The subset of residues common to the bioactive surfaces of Css4, Bj-xtrIT, and LqhIT2 raised the possibility that these toxins might compete in binding for receptor site 4 on different Navs.
Fig. 1. Sequence alignment and three-dimensional structure of -toxin representatives. A, sequences were aligned according to the conserved cysteine residues. The disulfide bonds formed between cysteine pairs in all "long-chain" scorpion toxins are designated by lines. Dashes indicate gaps. Secondary structure motifs (B, -strand; H, -helix) in Css4 follow the published structure of the -toxin Cn2 (from Centruroides noxius; Pintar et al., 1999). B, the C structure of Css4 and LqhIT2 (in gray) covered by a semitransparent molecular surface of the toxins. LqhIT2 is derived from Protein Data Bank accession 2I61. The Css4 model is from Cohen et al. (2005) and is spatially aligned with that of LqhIT2. The sulfur atoms in the disulfide bonds are highlighted in yellow, and the conserved glutamate is in red (see also A). The figure was prepared using PyMOL (http://www.pymol.org).
By analyzing the ability of depressant toxins to bind and affect various mammalian Navs, we found that despite the lack of effect on subcutaneously injected mice, depressant toxins bound to receptor site 4 on rat muscle membranes with nanomolar affinity. We show that LhqIT2 and Lqh-dprIT3 are effective on rNav1.4 when the channel is excited before toxin application by either a long depolarizing prepulse or by modulation with an -toxin. These results suggest that the reported selectivity of depressant toxins to insect Navs rests on the way they were tested and that depressant toxins may have a toxic impact on mammals in the context of the complete venom.
Toxins and Their Mutagenesis
Production of Bj-xtrIT, LqhIT2, Css4, and Lqh-dprIT3 variant c in recombinant forms, polymerase chain reaction-driven mutagenesis, expression in Escherichia coli, in vitro folding, and purification of toxin derivatives have been described in detail (Turkov et al., 1997; Froy et al., 1999; Cohen et al., 2005; Strugatsky et al., 2005)
Binding Experiments
Neuronal membranes from cockroach were prepared from whole heads of adult Periplaneta americana according to a previously described method (Froy et al., 1999). Rat skeletal muscle membranes were prepared from adult albino Wistar strain (300 g, laboratory bred) as described previously (Gordon et al., 1988). Mammalian brain synaptosomes were prepared from the same rats as described previously (Gilles et al., 2001). Membrane protein concentration was determined by a Bio-Rad Protein Assay (Bio-Rad Laboratories, Hercules, CA), using bovine serum albumin as standard. Bj-xtrIT and Css4 (with a His-tag attached; His-Css4) were radioiodinated by lactoperoxidase (Sigma, St. Louis, MO; 7 units per 60 µl of reaction mix) using 10 µg of toxin and 0.5 mCi of carrier-free Na125I (GE Healthcare, Chalfont St. Giles, Buckinghamshire, UK), and the monoiodotoxin was purified as described previously (Cohen et al., 2004, 2005). The media composition used in the binding assays and termination of the reactions were described elsewhere (Gilles et al., 2000, 2001). Nonspecific toxin binding was determined in the presence of 1 to 10 µM unlabeled toxin and consisted typically of 10 to 30% of total binding. Equilibrium competition binding assays were performed and analyzed as described previously (Cohen et al., 2005). Each experiment was performed in duplicate and repeated at least three times as indicated (n). Data are presented as mean ± S.D. of the number of independent experiments.
Expression of Sodium Channels in Oocytes and Two-Electrode Voltage Clamp Experiments
The genes encoding the Drosophila melanogaster sodium channel -subunit (DmNav1) and the auxiliary TipE subunit were kindly provided by J. Warmke (Merck, Whitehouse Station, NJ) and M. S. Williamson (IACR-Rothamsted, Harpenden, Hertfordshire, UK), respectively. The gene encoding the rat skeletal muscle sodium channel, rNav1.4, in the pAlter vector was a gift from Dr. R. G. Kallen (University of Pennsylvania, Philadelphia, PA). These genes and that for the auxiliary subunit h1 were transcribed in vitro using T7 RNA-polymerase and the mMESSAGE mMACHINE system (Ambion, Austin, TX) and were injected into Xenopus laevis oocytes as described previously (Shichor et al., 2002).
Two-Electrode Voltage-Clamp Recording
Currents were measured 1 to 2 days after injection using a two-electrode voltage clamp and a Gene Clamp 500 amplifier (Molecular Devices, Sunnyvale, CA). Data were sampled at 10 kHz and filtered at 5 kHz. Data acquisition was controlled by a Macintosh PPC 7100/80 computer (Apple Corp., Cupertino, CA), equipped with ITC-16 analog/digital converter (Instrutech Corp., Port Washington, NY), using Synapse (Synergistic Systems, Stockholm, Sweden). The bath solution contained 96 mM NaCl, 2 mM KCl, 1 mM MgCl2, 2 mM CaCl2, and 5 mM HEPES, pH 7.85. Oocytes were washed with bath solution flowing from a BPS-8 perfusion system (ALA Scientific Instruments, Westbury, NY) with a positive pressure of 4 psi. Toxins were diluted with bath solution and applied directly to the bath at their final desired concentration.
Data Analysis
Leak Subtraction. Capacitance transients and leak currents were removed by subtracting a scaled control trace using a P/6 protocol (Armstrong and Bezanilla, 1974).
GV Analysis. Mean conductance (G) was calculated from peak current/voltage relations using the equation G = I/(V – Vrev), where I is the peak current elicited upon depolarization, V is the membrane potential, and Vrev is the reversal potential. Normalized conductance voltage relationship were fit with either a one or two component Boltzmann distribution according to the equation G/Gmax = (1 – A)/(1 + exp[(V11/2 – V)/k1]) + A/(1 + exp[(V21/2 – V)/k2]), where V11/2 and V21/2 are the respective membrane potentials for two populations of channels for which the mean conductance is half-maximal; k1 and k2 are their respective slopes, and A defines the proportion of the second population (amplitude) with respect to the total. For fits in which only one population of channels was apparent, A was set to zero.
Steady-State Fast Inactivation. The voltage dependence of steady-state fast inactivation is described by a single Boltzmann distribution: I/Imax = 0 + 1/(1 + exp[(V – V1/2)/k]), where I is the peak current measured during the test depolarization step, Imax is the current obtained without a preceding conditioning step; V is the membrane potential of the conditioning step; V1/2 is the membrane potential at which half-maximal inactivation is achieved, k is the slope factor, 0 is the remaining normalized peak current at very high depolarizing conditioning potentials, and 1 is the normalized amplitude (Chen and Heinemann, 2001).
Binding of Depressant Toxins to Rat Skeletal Muscle Navs. Scorpion depressant toxins have traditionally been considered insect-selective on the basis of their exclusive toxicity to insects and high binding affinity for insect Navs (Lester et al., 1982; Zlotkin et al., 1991; Gordon et al., 1992; Strugatsky et al., 2005). Still, the elucidation of a pharmacophore common to the bioactive surface of scorpion -toxins active on insects and mammals (Cohen et al., 2005; Karbat et al., 2007) prompted us to analyze whether the toxins that show selectivity for insects would compete with the anti-mammalian -toxin Css4 on binding to receptor site 4 on rat muscle and brain Navs. Although the excitatory toxin Bj-xtrIT did not displace 125I-Css4 in concentrations up to 10 µM, the depressant toxins LqhIT2 and Lqh-dprIT3 inhibited in a dose-dependent manner the binding of 125I-Css4 to rat muscle membranes with Ki values of 45 ± 7 and 30 ± 3.2 nM, respectively (Fig. 2, Table 1). In contrast, the excitatory and the depressant toxins did not displace 125I-Css4 from rat brain synaptosomes at concentrations up to 10 µM (inset in Fig. 2). The ability of depressant toxins to bind with an apparent high affinity to rat skeletal muscle Navs raised the inevitable question of why these toxins were inactive when injected subcutaneously into mice (Lester et al., 1982; Zlotkin et al., 1991, 1993). Therefore, we analyzed the effects of LqhIT2 and Lqh-dprIT3 on the rat muscle channel rNav1.4 and compared them with those obtained on the Drosophila melanogaster channel DmNav1, expressed in X. laevis oocytes.
Fig. 2. Competition of -toxins with Css4 on binding to rat muscle membranes. Membranes were incubated 60 min at 22°C with 0.1 nM 125I-Css4 and increasing concentrations of the various -toxins. Nonspecific binding, determined in the presence of 1 µM Css4, was subtracted. The Ki values are (in nanomolar, n 3): Css4, 3.9 ± 1.17; Lqh-dprIT3, 30 ± 3.2; LqhIT2, 45 ± 7; Bj-xtrIT, >>10,000. The binding of Css4, LqhIT2, and Bj-xtrIT to rat brain synaptosomes under the same conditions is shown in the inset. The Ki values are (in nanomolar, n 3): Css4, 0.98 ± 0.1; LqhIT2 and Bj-xtrIT, >>10,000. Representative experiments are shown.
TABLE 1 Changes in apparent binding affinity of depressant -toxins and selected mutants to rat muscle and cockroach neuronal membranes
The Ki values obtained from competition binding studies using 125I-Bj-xtrIT (cockroach neuronal membranes) and 125I-Css4 (rat muscle membrane preparation). See Fig. 4 for details.
Fig. 4. Binding of LqhIT2 and its mutants to cockroach and rat muscle membrane. Competition of LqhIT2 and mutants with 125I-Bj-xtrIT on binding to cockroach neuronal membranes (A) and with 125I-Css4 to rat muscle membranes (B). Membranes were incubated 60 min at 22°C with 0.1 nM 125I-Bj-xtrIT or 125I-Css4 and increasing concentrations of the various mutants. Nonspecific binding, determined in the presence of 1 µM Bj-xtrIT or Css4, was subtracted. The Ki values in nanomolar, n 3, are given in Table 1. Representative experiments are shown.
Fig. 3. LqhIT2 and Lqh-dprIT3 effect on activation of DmNav1 and rNav1.4. A1 and A2, current-voltage (I-V) relations of DmNav1 under control conditions and in the presence of 1 µM LqhIT2 (A1) or 0.2 µM Lqh-dprIT3 (A2) with or without a 50-ms PP to +60 mV from a –100-mV holding potential. B1 and B2, current-voltage relations of rNav1.4 under control conditions and in the presence of 5 µM LqhIT2 (B1), or 5 µM Lqh-dprIT3 (B2) with a 50- or 500-ms PP to +60 mV from a –100-mV holding potential. A3 and B3, analysis of the effect of depressant toxins on DmNav1 (A3) and rNav1.4 (B3) activation after various PP durations (0–1 s) to +60 mV and measuring at –50 mV. The current at each point was normalized to the maximal effect. Representative experiments are shown.
The Effects of LqhIT2 and Lqh-dprIT3 on the Activation of DmNav1 and rNav1.4. The voltage-dependent activation of DmNav1 and rNav1.4 expressed in X. laevis oocytes was monitored by two-electrode voltage-clamp in the absence and presence of LqhIT2 or Lqh-dprIT3 (Fig. 3, Table 2). Because the negative shift of voltage-dependent activation induced by -toxins is better observed after a preconditioning depolarizing prepulse (PP) (Cestèle et al., 1998; Tsushima et al., 1999), we first examined the PP duration required to observe such a shift in the insect Nav in the presence of LqhIT2. In the absence of toxin, DmNav1 was not activated by a 50-ms test pulse to –50 mV independent of whether or not a depolarizing PP was provided, whereas in the presence of 1 µM LqhIT2, a hyperpolarizing shift in current-voltage relations at DmNav1 was observed only after a preconditioning depolarizing PP (Fig. 3, A1). This shift was highly dependent on the PP length (Fig. 3, A). A 100-ms PP to +60 mV provided the maximal effect measured by the developing currents elicited by a test pulse to –50 mV (Fig. 3, A3). Similar results were obtained with 200 nM Lqh-dprIT3 (Fig. 3, A2), a depressant toxin with higher toxicity to blowfly larvae (Strugatsky et al., 2005). It is noteworthy that Lqh-dprIT3 already induced a maximal effect on DmNav1 activation after 10 ms PP (Fig. 3, A3). In comparison, a very short preconditioning depolarizing PP (2 ms) was sufficient to induce a hyperpolarizing shift in activation of rNav1.2a by the anti mammalian -toxin Css4 (Cestèle et al., 1998). These results have indicated that the effects of various -toxins on the hyperpolarizing shift in channel activation clearly depend on the PP duration.
TABLE 2 Alterations of the mid-voltage of activation (V0.5) derived from conductance-voltage (G-V) curves of DmNav1 and rNav1.4 induced by depressant toxins and their mutants
The V0.5 values are from the G-V curves presented in Fig. 5. Depressant toxin effect on rNav1.4 exhibit two components: a minor negative shift in the V0.5 of the entire channel population (upper number) and a stronger shift in the V0.5 of a fraction of the toxin-modified channel population (lower number), indicated by the numbers in parentheses. The data represent the mean ± S.E.M. of at least six independent experiments.
Fig. 5. Alterations in conductance-voltage (G-V) relations induced by LqhIT2 and Lqh-dprIT3 mutants at DmNav1 and rNav1.4. The effects of LqhIT2 and its mutants K23A, E24A, and E24N on DmNav1 (A) and rNav1.4 (B). The effects of Lqh-dprIT3 and its mutant E24N on DmNav1 (C) and rNav1.4 (D). The labels in A cover B and those in C cover D. Toxin concentrations and the activation parameters (V0.5) are as described in Table 2. Conductance-voltage relations were determined as described in Fig. 3 with a 50-ms PP for DmNav1 and a 500-ms PP for rNav1.4. The data represent the mean ± S.E.M. of at least six independent experiments.
Based on the observation that LqhIT2 requires a relatively long PP to induce a hyperpolarizing shift in DmNav1 activation, and that LqhIT2 and Lqh-dprIT3 bind with relatively high affinity to the rat muscle membranes (Fig. 2), we examined whether longer PP durations would facilitate the effect of these toxins on rNav1.4 activation. We found that LqhIT2 was indeed capable of shifting the rNav1.4 activation in the hyperpolarizing direction, but it required a longer PP to +60 mV, and even after a 2-s PP, the effect was not maximal (Fig. 3B). In the absence of toxin, a 500-ms PP had no effect on the peak current, and a 1- or 2-s PP induced a 10 or 20% decrease, respectively, in the peak current elicited at –10 mV (data not shown), probably as a result of the development of slow inactivation (Featherstone et al., 1996; Mitrovic et al., 2000). Similar results were obtained with Lqh-dprIT3 (Fig. 3, B2 and B3), suggesting that depressant toxins not only bind with high affinity to rat muscle Navs but also have a clear effect under the appropriate conditions. In contrast, regardless of the PP length, the depressant toxins did not shift the activation curve of the rat brain channel rNav1.2a expressed in oocytes (data not shown), which was in concert with their inability to bind to rat brain synaptosomes (Fig. 2, inset). These data demonstrated that receptor site 4 varies in different Navs.
Effect of Substitutions at the "Hot Spot" on Selectivity of LqhIT2 and Lqh-dprIT3. We have shown that substitution of a conserved Glu residue on the bioactive surfaces of the anti-mammalian -toxin Css4 (Glu28; Fig. 1) and the anti-insect selective excitatory -toxin Bj-xtrIT (Glu30) abolished their binding and activity at the brain and insect Navs, respectively. Substitution of the adjacent Arg27 in Css4 and His25 in Bj-xtrIT also decreased the binding and activity of both toxins (Cohen et al., 2004, 2005). Therefore we analyzed whether the spatially equivalent residues in LqhIT2, Glu24 and Lys23 (Karbat et al., 2007), have a role in toxin binding and activity at rNav1.4. Substitution K23A decreased LqhIT2 affinity for both cockroach neuronal and rat muscle membranes (Fig. 4, Table 1) as well as abolished the activity at DmNav1 and rNav1.4 (Fig. 5, Table 2). Although substitutions E24A/N decreased the toxin affinity for cockroach neuronal membranes by 450-fold (Fig. 4B, Table 1) and no activity was observed at DmNav1 with up to 5 µM toxin (Fig. 5A, Table 2), LqhIT2E24A/N affinity for the rat muscle increased 5-fold (Fig. 4A, Table 1), and the effect at rNav1.4 after a 500-ms PP was higher than that of LqhIT2 (Fig. 5B, Table 2). Substitution E24N in Lqh-dprIT3 had a similar effect on toxin activity in that the effect at DmNav1 declined but increased at rNav1.4 (Fig. 5, C and D, Table 2). These results not only indicate that depressant toxins have different requirements for modifying the activation of insect versus mammalian Navs, but also that a single amino acid substitution is able to invert the preference of depressant toxins and make them selective to skeletal muscle Navs (Tables 1 and 2).
LqhIT Modulates LqhIT2 Activity at rNav1.4. We have shown previously that the -toxin LqhIT and the depressant -toxin LqhIT2 allosterically increase the binding of one another at insect Navs, which was manifested in a strong synergism in their toxicity to insects (Cohen et al., 2006). Considering that LqhIT was shown to inhibit the fast inactivation of rNav1.4 expressed in HEK cells with an EC50 of 1.2 nM (Leipold et al., 2004), we examined whether LqhIT would modulate LqhIT2 activity at rNav1.4. LqhIT in a concentration of 200 nM shifted the voltage-dependence of steady-state fast inactivation of rNav1.4 expressed in X. laevis oocytes by +15 mV (V1/2 = –48 ± 0.3 mV in the control and –33.7 ± 0.3 mV in the presence of LqhIT), thus increasing the percentage of channels available for activation at subthreshold membrane potentials (under –40 mV) with no effect on the channels conductance-voltage relations (Fig. 6). These data suggested that LqhIT might have facilitated the activity of the depressant toxin. Upon coapplication of LqhIT (200 nM) and LqhIT2 (5 µM), a 50-ms PP to +60 mV induced a hyperpolarizing shift in rNav1.4 conductance-voltage relations (Fig. 6A). Such an effect on the channel activation was comparable with that obtained by a 500-ms PP to +60 mV when LqhIT2 was applied alone (Fig. 5B, Table 2). This result demonstrates that LqhIT binding to rNav1.4 and/or its influence on fast inactivation modulates the activity of LqhIT2 on this channel as indicated by the shorter PP required to observe the effect of the depressant toxin.
Fig. 6. Effects of combined application of LqhIT2 and LqhIT on rNav1.4 activation. A, conductance-voltage relations of rNav1.4 in the absence of toxin (V0.5 = –33 ± 0.3 mV, n = 3), in the presence of 200 nM LqhIT (V0.5 = –33 ± 0.4 mV), and in the presence of 200 nM LqhIT and 5 µM LqhIT2 (V0.5 = –38 ± 0.1 for all channels, and –52 ± 1.5 mV for 8% of the channel population; see Materials and Methods and Table 2). Conductance-voltage relations were determined as described in Fig. 2 with a 50-ms PP to +60 mV. Note that 5 µM LqhIT2 had no effect on activation (see Fig. 3B1). Inset, current traces obtained in the absence of toxins and upon coapplication of 200 nM LqhIT and 5 µM LqhIT2 at a test pulse to –45 mV after a 50-ms PP to +60 mV. B, steady-state fast inactivation was determined from holding potential of –100 mV using a series of 50-ms PP from –80 to –20 mV in 5-mV increments before the test pulse of –20 mV. The steady-state inactivation of rNav1.4 fits a Boltzmann function with V0.5 = –48 ± 0.3 mV, and in the presence of 200 nM LqhIT, V0.5 = –33.7 ± 0.3 mV. Inset, effects of 200 nM LqhIT on the current fast inactivation elicited by a test pulse to –20 mV from a holding potential of –100 mV.
The experiments described in this study reveal that the allegedly "insect-selective" scorpion depressant toxins are capable of binding with high affinity and affecting mammalian skeletal muscle Navs, a fact unnoticed for almost 3 decades. The classification of depressant toxins as insect-selective relied on the lack of toxicity when injected to mice and inability to bind rat brain synaptosomes (Zlotkin et al., 1993; Gordon et al., 1998; Gurevitz et al., 2007) or affect mammalian Navs expressed in X. laevis oocytes (Gordon et al., 2003; Bosmans et al., 2005). However, a number of recent results have suggested that the issue of selectivity of depressant toxins toward insects deserves reexamination: 1) scorpion -toxins share a common pharmacophore (Cohen et al., 2005; Karbat et al., 2007), which explains their ability to compete in binding; 2) receptor site 4 on rat skeletal muscle Navs was suggested to differ from those of various mammalian neuronal and cardiac Navs (Marcotte et al., 1997; Cestèle et al., 1998; Leipold et al., 2006; Shciavon et al., 2006; Cohen et al., 2007); and 3) scorpion - and -toxins exert synergistic effects as a e result of allosteric interactions between receptor sites 3 and 4 on insect Navs (Cohen et al., 2006).
On the basis of these considerations, we reanalyzed the activity of depressant toxins on insect and mammalian Navs. Although depressant toxins were unable to bind rat brain Navs, they exerted high affinity for the rat skeletal muscle Navs (Fig. 2), which motivated us to analyze their effects on channel activation. The unexpected observation that the effect of the depressant toxin LqhIT2 on the insect Nav DmNav1 necessitated a 10-fold longer PP than that required to observe a Css4 effect on mammalian Navs suggested that LqhIT2 should be analyzed on rNav1.4 after a longer PP. The requirement for PP has been attributed to a putative energetic barrier that needs to be overcome before the prebound -toxin can trap the DII/S4 voltage sensor in its outward activated position, thus leading to enhanced channel activation upon subsequent depolarizations (Cestèle et al., 1998). Thus far, a short PP (several milliseconds) was ample for inducing a noticeable effect of most anti-mammalian -toxins on mammalian Navs (Cestèle et al., 1998, 2006; Tsushima et al., 1999; Cohen et al., 2005). Here we show that a very long PP (>500 ms) to +60 mV made rNav1.4 vulnerable to depressant toxins. It is likely that because depressant toxins did not modulate the gating properties of rNav1.2 (Bosmans et al., 2005) and rNav1.4 (Fig. 2; Gordon et al., 2003) after a depolarizing PP up to 50 ms, their activity on mammalian Navs has not been noticed thus far.
The ability of depressant toxins to influence the activation of the mammalian Nav after a long PP was surprising and raised the question of their putative role in vivo. In light of the enhancement of LqhIT2 binding to insect Navs in the presence of a scorpion -toxin from the same venom, LqhIT (Cohen et al., 2006), and because LqhIT is highly potent on insect as well as a variety of mammalian Nav subtypes, including rNav1.4 (Eitan et al., 1990; Chen et al., 2000; Leipold et al., 2004; Gordon et al., 2007), it was rational to analyze the joint effect of both toxins on rNav1.4. Indeed, the synergism between site 3 and site 4 toxins observed at rNav1.4 may be explained by the reduction of at least 10-fold in PP duration required for induction of LqhIT2 effect in the presence of LqhIT (Fig. 6). Such a mechanism may also apply to insect Navs, where synergism between site 3 and site-4 toxins was reported (Cohen et al., 2006). LqhIT increases neuronal excitability and neuromuscular activity upon binding to receptor site 3 by induction of long plateau potentials in axons attributed to an increase in the probability of Navs to remain in open states as a result of inhibition of their fast inactivation (Eitan et al., 1990; Gilles et al., 2000; Lee et al., 2000; Benoit and Gordon, 2001). This inhibition of channel steady-state fast inactivation expands the channel population available for activation at resting membrane potential, leading to increase in the frequency of action potentials, which may act as a prepulse to facilitate -toxin activity. This, in turn, may facilitate -toxin action on the channels in their open states. Thus, the mutual enhancement in -toxin effect by -toxin interaction with receptor site 4, and vice versa, results from an indirect modification of receptor sites 3 and 4, respectively and from alteration in the voltage-dependence of channel activation (Cohen et al., 2006). We suggest, based on these considerations, that Nav1.4 in a stung mammal is preconditioned upon binding of the -toxin, thus enabling synergistic toxicity by the joint effects of - and depressant toxins.
The venom of L. quinquestriatus hebraeus contains a number of -toxins active on mice (e.g., LqhIT, Lqh2, Lqh3, Lqh4, Lqh6, and Lqh7; for references, see Gordon et al., 2007) and -toxins that exhibit preference for insects, including various depressant toxins (e.g., LqhIT2, LqhIT5, Lqh-dprIT3, Lqh1; for references, see Gurevitz et al., 2007), which together increase the impact of stinging. In their isolated form, however, the depressant toxins may be considered selective to insects, in that their affinity for insect neuronal membranes is 2 orders of magnitude higher than the affinity for rat muscle membranes (Figs. 2, 4, and 5, Table 1; Gordon et al., 1992; Strugatsky et al., 2005).
A surprising feature in the interaction of depressant toxins with the mammalian muscle Nav is the lack of function of Glu24, found to be conserved in other scorpion -toxins and considered a hot spot on the bioactive surface of these toxins toward insect and rat brain Navs (Cohen et al., 2004, 2005; Karbat et al., 2007) (Fig. 1), suggesting that receptor site 4 on the muscle channels differs from that on DmNav1. Further understanding of how scorpion -toxins interact in a preferential manner with various receptor sites on Nav subtypes seems to await elucidation of toxin-receptor interacting surfaces. At the moment, our results indicate that a single substitution at this position (LqhIT2E24A/N and Lqh-dprIT3E24N; Figs. 4 and 5, Table 1) converts these depressant toxins from "insect-selective" to "mammal-selective."
ABBREVIATIONS: Navs, voltage-gated sodium channels; Css4, Centruroides suffusus suffusus toxin 4; LqhIT2 and Lqh-dprIT3, Leiurus quinquestriatus hebraeus anti-insect depressant toxins; PP, preconditioning depolarizing prepulse.
1 Current affiliation: Neuroscience Institute, Morehouse School of Medicine, Atlanta, Georgia.
【参考文献】
Armstrong CM and Bezanilla F (1974) Charge movement associated with the opening and closing of the activation gates of Na channels. J Gen Physiol 63: 533–552.[Abstract/Free Full Text]
Ben Khalifa R, Stankiewicz M, Pelhate M, Serrano-Hernandez SE, Possani LD, Hinkel H, and Mebs D (1997) Action of babycurus-toxin 1 from the east African scorpion Babycurus centrurimorphus on the isolated cockroach giant axon. Toxicon 35: 1069–1080.
Benoit E and Gordon D (2001) The scorpion -like toxin Lqh III specifically alters sodium channel inactivation in frog myelinated axons. Neuroscience 104: 551–559.
Bosmans F, Martin-Eauclaire MF, and Tytgat J (2005) The depressant scorpion neurotoxin LqqIT2 selectively modulates the insect voltage-gated sodium channel. Toxicon 45: 501–507.
Catterall WA (2000) From ionic currents to molecular mechanisms: the structure and function of voltage-gated sodium channels. Neuron 26: 13–25.
Cestèle S and Catterall WA (2000) Molecular mechanisms of neurotoxin action on voltage-gated sodium channels. Biochimie 82: 883–892.
Cestèle S, Qu Y, Rogers JC, Rochat H, and Catterall WA (1998) Voltage sensor-trapping: enhanced activation of sodium channels by -scorpion toxin bound to the S3–S4 loop in domain II. Neuron 21: 919–931.
Cestèle S, Yarov-Yarovoy V, Qu Y, Sampieri F, Scheuer T, and Catterall WA (2006) Structure and function of the voltage sensor of sodium channels probed by a -scorpion toxin. J Biol Chem 281: 21332–21344.[Abstract/Free Full Text]
Chen H, Gordon D, and Heinemann SH (2000) Modulation of cloned skeletal muscle sodium channels by the scorpion toxins Lqh II, Lqh III, and LqhIT. Pflugers Arch 439: 423–432.
Chen H and Heinemann SH (2001) Interaction of scorpion alpha-toxins with cardiac sodium channels: binding properties and enhancement of slow inactivation. J Gen Physiol 117: 505–518.[Abstract/Free Full Text]
Cohen L, Ilan N, Gur M, Sthmer W, Gordon D, and Gurevitz M (2007) Design of a specific activator for skeletal muscle voltage-gated sodium channels uncovers channel architecture. J Biol Chem 282: 29424–29430.[Abstract/Free Full Text]
Cohen L, Karbat I, Gilles N, Froy O, Angelovici R, Gordon D, and Gurevitz M (2004) Dissection of the functional surface of an anti-insect excitatory toxin illuminates a putative "hot spot" common to all scorpion -toxins affecting Na+ channels. J Biol Chem 279: 8206–8211.[Abstract/Free Full Text]
Cohen L, Karbat I, Gilles N, Ilan N, Gordon D, and Gurevitz M (2005) Common features in the functional surface of scorpion -toxins and elements that confer specificity for insect and mammalian voltage-gated sodium channels. J Biol Chem 280: 5045–5053.[Abstract/Free Full Text]
Cohen L, Lipstein N, and Gordon D (2006) Allosteric interaction between scorpion toxin receptor sites on voltage-gated Na channels imply a novel role for weakly active components in arthropod venom. FASEB J 20: 1933–1935.[Abstract/Free Full Text]
Eitan M, Fowler E, Herrmann R, Duval A, Pelhate M, and Zlotkin E (1990) A scorpion venom neurotoxin paralytic to insects that affects sodium current inactivation: purification, primary structure, and mode of action. Biochemistry 29: 5941–5947.
Featherstone DE, Richmond JE, and Ruben PC (1996) Interaction between fast and slow inactivation in Skm1 sodium channels. Biophys J 71: 3098–3109.
Froy O, Zilberberg N, Gordon D, Turkov D, Gilles N, Stankiewicz M, Pelhate M, Loret E, Oren D, Shaanan B, et al. (1999) The putative bioactive surface of insect-selective scorpion excitatory neurotoxins. J Biol Chem 274: 5769–5776.[Abstract/Free Full Text]
Gilles N, Chen H, Wilson H, Le Gall F, Montoya G, Molgo J, Sch?nherr R, Nicholson G, Heinemann SH, and Gordon D (2000) Scorpion - and -like toxins differentially interact with sodium channels in mammalian CNS and periphery. Eur J Neurosci 12: 2823–2832.
Gilles N, Leipold E, Chen H, Heinemann SH, and Gordon D (2001) Effect of depolarization on binding kinetics of scorpion -toxin highlights conformational changes of rat brain sodium channels. Biochemistry 40: 14576–14584.
Gordon D (1997) Sodium channels as targets for neurotoxins: mode of action and interaction of neurotoxins with receptor sites on sodium channels, in Toxins and Signal Transduction (Lazarowici P and Gutman Y eds) pp 119–149, Harwood Press, Amsterdam.
Gordon D, Ilan N, Zilberberg N, Gilles N, Urbach D, Cohen L, Karbat I, Froy O, Gaathon A, Kallen RG, et al. (2003) An `Old World' scorpion -toxin that recognizes both insect and mammalian sodium channels: a possible link towards diversification of -toxins. Eur J Biochem 270: 2663–2670.
Gordon D, Jover E, Couraud F, and Zlotkin E (1984) The binding of an insect selective neurotoxin (AaIT) from scorpion venom to locust synaptosomal membranes. Biochim Biophys Acta 778: 349–358.
Gordon D, Karbat I, Ilan N, Cohen L, Kahn R, Gilles N, Dong K, Stühmer W, Tytgat J, and Gurevitz M (2007) The differential preference of scorpion alpha-toxins for insect or mammalian sodium channels: implications for improved insect control. Toxicon 49: 452–472.
Gordon D, Martin-Eauclaire MF, Cestèle S, Kopeyan C, Carlier E, Ben Khalifa R, Pelhate M, and Rochat H (1996) Scorpion toxins affecting sodium current inactivation bind to distinct homologous receptor sites on rat brain and insect sodium channels. J Biol Chem 271: 8034–8045.[Abstract/Free Full Text]
Gordon D, Merrick D, Wallner DA, and Catterall WA (1988) Biochemical properties of sodium channels in a wide range of excitable tissues studied with site-directed antibodies. Biochemistry 27: 7032–7038.
Gordon D, Moskowitz H, Eitan M, Warner C, Catterall WA, and Zlotkin E (1992) Localization of receptor sites for insect-selective toxins on sodium channels by site-directed antibodies. Biochemistry 31: 7622–7628.
Gordon D, Savarin P, Gurevitz M, and Zinn-Justin S (1998) Functional anatomy of scorpion toxins affecting sodium channels. J Toxicol Toxin Rev 17: 131–158.
Gurevitz M, Karbat I, Cohen L, Ilan N, Kahn R, Turkov M, Stankiewicz M, Stühmer W, Dong K, and Gordon D (2007) The insecticidal potential of scorpion beta-toxins. Toxicon 49: 473–489.
Herrmann R, Moskowitz H, Zlotkin E, and Hammock BD (1995) Positive cooperativity among insecticidal scorpion neurotoxins. Toxicon 33: 1099–1102.
Karbat I, Turkov M, Cohen L, Kahn R, Gordon D, Gurevitz M, and Frolow F (2007) X-ray structure and mutagenesis of the scorpion depressant toxin LqhIT2 reveals key determinants crucial for activity and anti-insect selectivity. J Mol Biol 366: 586–601.
Lee D, Gurevitz M, and Adams ME (2000) Modification of synaptic transmission and sodium channel inactivation by the insect-selective scorpion toxin LqhIT. J Neurophysiol 83: 1181–1187.[Abstract/Free Full Text]
Leipold E, Hansel A, Borges A, and Heinemann SH (2006) Subtype specificity of scorpion toxin Tz1 interaction with voltage-gated sodium channels is determined by the pore loop of domain-3. Mol Pharmacol 70: 340–347.[Abstract/Free Full Text]
Leipold E, Lu S, Gordon D, Hansel A, and Heinemann SH (2004) Combinatorial interaction of scorpion toxins Lqh-2, Lqh-3, and LqhIT with sodium channel receptor sites-3. Mol Pharmacol 65: 685–691.[Abstract/Free Full Text]
Lester D, Lazarovici P, Pelhate M, and Zlotkin E (1982) Purification, characterization and action of two insect toxins from the venom of the scorpion Buthotus judaicus. Biochim Biophys Acta 701: 370–381.
Li C, Guan RJ, Xiang Y, Zhang Y, Wang DC (2005) Structure of an excitatory insect-specific toxin with an analgesic effect on mammals from the scorpion Buthus martensii Karsch. Acta Crystallogr D Biol Crystallogr 61: 14–21.
Marcotte P, Chen LQ, Kallen RG, and Chahnin M (1997) Effects of Tityus serrulatus scropion toxin on voltage-gated Na+ channels. Circ Res 80: 363–369.[Abstract/Free Full Text]
Martin-Eauclaire MF, and Couraud F (1995) Scorpion neurotoxins: effects and mechanisms. in Handk Neurotoxicology (Chang LW and Dyer RS eds) pp 683–716, Marcel Dekker, New York.
Mitrovic N, George AL, and Horn R (2000) Role of domain 4 in sodium channel slow inactivation. J Gen Physiol 115: 707–717.[Abstract/Free Full Text]
Oren DA, Froy O, Amit E, Kleinberger-Doron N, Gurevitz M, and Shaanan B (1998) An excitatory scorpion toxin with a distinctive feature: an additional alpha helix at the C terminus and its implications for interaction with insect sodium channels. Structure 6: 1095–1103.
Pintar A, Possani LD, and Delepierre M (1999) Solution structure of toxin 2 from Centruroides noxius Hoffmann, a -scorpion neurotoxin acting on sodium channels. J Mol Biol 287: 359–367.
Possani LD, Becerril B, Delepierre M, and Tytgat J (1999) Scorpion toxins specific for Na+-channels. Eur J Biochem 264: 287–300.
Shichor I, Zlotkin E, Ilan N, Chikashvili D, Stuhmer W, Gordon D, and Lotan I (2002) Domain 2 of Drosophila Para voltage-gated sodium channel confers insect properties to a rat brain channel. J Neurosci 22: 4364–4371.[Abstract/Free Full Text]
Strugatsky D, Zilberberg N, Stankiewicz M, Ilan N, Turkov M, Cohen L, Pelhate M, Gilles N, Gordon D, and Gurevitz M (2005) Genetic polymorphism and expression of a highly potent scorpion depressant toxin enables refinement of the effects on insect Na channels and illuminates the key role of Asn-58. Biochemistry 44: 9179–9187.
Tsushima RG, Borges A, and Backx PH (1999) Inactivated state dependence of sodium channel modulation by beta-scorpion toxin. Pflugers Arch 437: 661–668.
Turkov M, Rashi S, Noam Z, Gordon D, Ben Khalifa R, Stankiewicz M, Pelhate M, and Gurevitz M (1997) In vitro folding and functional analysis of an anti-insect selective scorpion depressant neurotoxin produced in Escherichia coli. Prot Expr Purif 9: 123–131.
Zlotkin E, Eitan M, Bindokas VP, Adams ME, Moyer M, Burkhart W, and Fowler E (1991) Functional duality and structural uniqueness of depressant insect-selective neurotoxins. Biochemistry 30: 4814–4821.
Zlotkin E, Fowler E, Gurevitz M, and Adams ME (1993) Depressant insect selective neurotoxins from scorpion venom: chemistry, action, and gene cloning. Arch Insect Biochem Physiol 22: 55–73.
Zlotkin E, Miranda F, and Rochat H (1978) Chemistry and pharmacology of Buthinae scorpion venoms, in Arthropods Venoms (Bettini S ed) pp 317–369, Springer Verlag, New York.
作者单位：Department of Plant Sciences, George S. Wise Faculty of Life Sciences (L.C., M.T., N.I., D.G., M.G.), and Department of Physiology and Pharmacology, Sackler School of Medicine (Y.T., M.B.), Tel-Aviv University, Tel-Aviv, Israel; and Commissariat à l'Energie Atomique, Department d'Ingenierie et d'Etu