| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Department of Plant Sciences, George S. Wise Faculty of Life Sciences, Tel-Aviv University, Ramat-Aviv 69978, Tel-Aviv, Israel; and
* CEA, Départment d’Ingéniérie et d’Etudes des Protéines, C.E. Saclay, F-91191, France
2Correspondence: E-mail: mamgur@post.tau.ac.il and dgordon{at}post.tau.ac.il
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Key Words: scorpion ß-toxin • sodium channel • voltage sensor • competitive antagonist
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
. NaChs are membrane proteins composed of a pore-forming 
subunit of 
260 kDa, organized in four repeat domains (DI-DIV), each containing six trans-membrane -helical segments (S1 to S6) connected by internal and external loops. The S4 segments contain four to eight positively charged residues at a three-residue interval, and function as a voltage-sensing device. Upon membrane depolarization, these segments move outward relative to the membrane electric field, imposing a conformational change that leads to channel activation. Channel inactivation is mediated by the short intracellular loop connecting domains III and IV (2)
.
Although most molecular determinants associated with channel gating reside within the membrane and at the intracellular face of the channel, polypeptide toxins from scorpions, sea anemones, spiders, and other venomous organisms modify channel gating by binding to specific receptor sites on extracellular channel loops (2
3
4)
. Scorpion ß-toxins are single-chain polypeptides of 61–76 amino acid residues cross-linked by four disulfide bridges (5)
. Their binding to neurotoxin receptor site 4, associated with domain II of the NaCh, shifts the voltage dependence of activation in the hyperpolarizing direction (2)
. Recent work using the ß-toxin Css4 from the Mexican scorpion, Centruroides suffusus suffusus identified residues in DIIS3-S4 of the rat brain NaCh Nav1.2a required for high-affinity binding and channel modulation by the toxin (6)
. In these experiments, toxin binding was voltage independent, and a conditioning depolarizing prepulse was required before the negative shift in voltage dependence of activation could be observed. On the basis of these results, a "voltage sensor trapping" model was proposed to explain ß-toxin action. According to this model, the toxin binds to the channel at resting membrane potential. Upon strong depolarization, the DIIS4 segment moves outward, tightly associates with the prebound toxin, and is trapped in an outward, activated position. Trapping of the voltage sensor underlies the negative shift in the voltage dependence of activation, which is characteristic of scorpion ß-toxin effect. Subsequent work has shown that neutralization of the two outermost Arg residues in DIIS4 of channel Nav1.2a, by their substitution to Gln, enhanced the Css4 effect and abolished the requirement for a depolarizing prepulse (7)
. It has been proposed that neutralization of the gating charges removes kinetic barriers for Css4 binding to DIIS4 by elimination of unfavorable electrostatic interactions with positively charged toxin residues. An alternative suggestion was that putative interactions of the gating charges with acidic residues in S2 and S3 were abolished, thus facilitating S4 movement in the membrane (7)
. To distinguish between these two alternatives, toxin residues involved in the interaction with DIIS4 need to be identified.
The only scorpion ß-toxin with known structure that is currently amenable for expression and mutagenesis is the excitatory anti-insect selective toxin, Bj-xtrIT, from Buthotus judaicus (8
, 9)
. It is the longest toxin affecting NaChs that has been described in scorpion venom, and is similar to other scorpion ß-toxins in its core structure (-helix packed against a three-stranded ß-sheet stabilized by three spatially conserved disulfide bonds). However, Bj-xtrIT varies prominently in its C-tail region and the spatial arrangement of its forth disulfide bond (Fig. 1
; 8, 10). Although excitatory toxins (e.g., AahIT and Bj-xtrIT) differ from anti-mammalian ß-toxins (e.g., Cn2 and Css4) in structure and selectivity to insect and mammalian NaChs (8
, 11)
, they also shift the voltage dependence of activation in the hyperpolarizing direction and bind receptor site 4 associated with domain-2 of insect NaChs (9
, 12
13
14)
. These similarities suggest that excitatory toxins may be used as a suitable model for studying the interaction of ß-toxins with sodium channels.
|
By probing the role of conserved charged residues for Bj-xtrIT function, we demonstrate that none of the basic residues is important, whereas two negatively charged residues, Glu15 and Glu30, play a mandatory role. Charge inversion of Glu15 abolishes toxicity with no effect on the binding affinity, thereby generating an efficient antagonist of the wild-type toxin. Based on the similarities in binding site and mode of action, and in concert with the model proposed by Catterall and co-workers, we propose that voltage sensor trapping by excitatory and anti-mammalian scorpion ß-toxins is mediated by direct interaction of a conserved acidic residue (Glu15 in Bj-xtrIT) with the positive gating charges of the voltage sensor as it moves outward in response to membrane depolarization.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
cells were used for plasmid constructions. E. coli BL21 cells were used for expression of recombinant toxins using the pET-11c vector as was described (9)
. Sarcophaga falculata blowfly larvae were bred in the laboratory.
Toxicity assays
Four-day-old blowfly larvae (Sarcophaga falculata; 150±20 mg body weight) were injected intersegmentally. A positive result was scored when a characteristic paralysis (immobilization and contraction) was observed up to 5 min after injection. Five concentrations of each toxin were injected to larvae (nine larvae in each group) in three independent experiments. ED50 values (effective dose 50%) were calculated according to the sampling and estimation method of Reed and Muench (15)
.
Binding studies
Insect synaptosomes were prepared from whole heads of adult cockroaches (Periplaneta americana) according to a described method (16)
. Bj-xtrIT was radioiodinated by Iodogen using 5 µg toxin and 0.5 mCi carrier free Na125I as was described. Concentration of labeled monoiodotoxin was determined according to the specific activity of 125I (2500 to 3000 dpm/fmol) (17
, 18)
. Equilibrium competition binding experiments were carried out with increasing concentrations of the unlabeled toxin in the presence of a constant low concentration of the radioactive toxin. Competition experiments were analyzed by KaleidaGraph (Synergy Software Version 3.08) and the inhibitory concentration 50% (IC50)was determined using a nonlinear fit to the Hill equation (17)
. Ki values were calculated by the equation Ki = IC50/1 + [L*/Kd], where L* is the concentration of the radioactive ligand and Kd is its dissociation constant. Each experiment was performed at least three times. Data are presented as mean ± SE of the number (n) of independent experiments.
CD spectroscopy
CD spectra were recorded at 25°C using a model 202 circular dichroism spectrometer (Aviv Instruments, Lakewood, NJ, USA). Toxins (140 µM) were dissolved in 5 mM sodium phosphate buffer, pH 7.0, and their spectrum (190–260 nm) was the average of three measurements using a quartz cell of 0.1 mm light path. Blank spectrum of the buffer was run under identical conditions and subtracted from each of the toxin spectra.
Site-directed mutagenesis and production of recombinant toxin variants
Expression of Bj-xtrIT was described (9)
. Mutations in Bj-xtrIT were generated by PCR using complementary oligonucleotide primers (purchased from Sigma, Israel) with expression vectors as DNA templates as described (9)
. Sequences were verified prior to expression, which was performed essentially as described (9)
. Toxin mutants were produced similarly to the unmodified toxin. Quantification of purified recombinant toxins was performed by amino acid analysis using an ABI system 420A/130A (Applied Biosystems Inc., Foster City, CA) after hydrolysis by 6M HCl under vacuum (18 h at 110°C).
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
). To examine whether any of them has a role in toxin function, their charge was neutralized by point mutagenesis. For each of the resulting toxin mutants, the CD spectrum was acquired and compared with that of the wild-type toxin in order to detect possible structural perturbations caused by the mutation. The CD spectra of all mutants were very similar to that of the unmodified recombinant toxin, indicating that none of the induced mutations altered significantly the toxin structure (Fig. 2
).
|
Neutralization of the basic residues Lys1, Lys2, Lys12, Lys33, Lys56, Lys66, and Lys67 (Fig. 1
) by substitution to Ala had little effect on toxicity to blowfly larvae and on the apparent binding affinity to cockroach neuronal membranes (Table 1
). Neutralization of the negatively charged amino acid residues Asp8, Glu38, Asp54, Asp55, Asp63, and Asp70 had little effect on toxin activity. Moreover, neutralization or inversion of three adjacent negative charges, Glu53, Asp54, and Asp55, which changed substantially the overall toxin charge, had practically no effect on toxin activity (Table 1)
. This suggests that the overall positive charge of the toxin is not important for activity.
|
In contrast, neutralization of Glu15 and Glu30 (mutants E15A and E30Q) had a marked effect on toxicity of Bj-xtrIT. Strikingly, the binding affinity of mutant E15A to cockroach neuronal membranes remained almost as high as that of the unmodified toxin whereas the binding affinity of mutant E30Q decreased 74.5-fold (Table 1
; Fig. 3
). To further examine this peculiarity, each residue was substituted by amino acids of different chemical properties. We first analyzed Glu30, which is conserved in the -helix of all known scorpion ß-toxins (Fig. 1
). Substitution of this acidic residue by Leu reduced the toxicity and binding affinity 27- and 158-fold, respectively, whereas a conserved substitution to Asp had a smaller effect (8.3- and 43-fold, respectively; Table 1
). This result suggested a roll for a negative charge at this position. In the case of interaction between Glu30 and a positively charged residue on the receptor, one might expect an E30R mutant to be even less potent than the E30Q mutant because it would introduce repulsive electrostatic interactions. Indeed, the toxicity and binding affinity of mutant E30R were one order of magnitude lower than those observed upon charge neutralization (E30Q and E30L; Table 1
; Fig. 3
). It is therefore conceivable that the conserved Glu30 interacts with a positively charged residue of receptor site 4.
|
Analysis of the functional roll of Glu15 using a similar mutagenic approach provided intriguing results. Whereas mutagenesis of Glu30 affected both the toxicity and binding affinity, mutations E15A and E15F decreased the toxicity 48.5 and 786-fold, respectively, but had only a minor effect on the binding affinity (Table 1
; Fig. 3
). This uncoupling of toxicity from binding affinity was most prominent when the negative charge of Glu15 was inversed. Mutant E15R was practically nontoxic to blowfly larvae in concentrations as high as 100 µg/100 mg body weight (more than four orders of magnitude decrease in toxicity). In sharp contrast, the binding affinity of E15R to cockroach neuronal membranes was only 6.8-fold lower than that of the unmodified recombinant toxin (Table 1
; Fig. 3
). To ensure that the different results in toxicity to flies vs. the binding to cockroach neuronal membranes were not the outcome of a difference in the sensitivity of blowfly and cockroach sodium channels to the Bj-xtrIT variants, the toxicity of Bj-xtrIT and the E15R mutant was examined on adult cockroaches. Wild-type Bj-xtrIT at a dose of 240 ng/100 mg adult P. americana immediately paralyzed the cockroach legs, followed by complete loss of motor coordination after 3–5 min and death after 100 min. In contrast, 1000 ng/100 mg body weight of the E15R mutant did not generate any toxic symptoms even after 24 h. Moreover, 400 ng Bj-xtrIT/100 mg cockroach was ineffective when coinjected with 10-fold higher dose of the E15R mutant (w/w). These results are in concert with the in vitro binding assays (Fig. 3
) and suggest that the blowfly and cockroach sodium channels are sensitive to Bj-xtrIT in a similar manner.
Mutant E15R is a potent competitive antagonist of the unmodified toxin
The pharmacological relevance of the high-affinity binding of mutant E15R was examined by coinjection with the unmodified toxin to blowfly larvae (in vivo competition). Dose-response curves were established using increasing doses of Bj-xtrIT with different fixed doses of E15R. The series of parallel right-shifted curves (Fig. 4
A) obtained indicates a direct competition between the two toxin derivatives. Schild regression analysis of the data yields a slope of 0.97, indicating that E15R is indeed a competitive antagonist (Fig. 4B
). The apparent IC50 value of E15R estimated from the Schild analysis is 75.72 ng/100 mg body weight of larvae, which is 5.71-fold higher than that obtained for the wild-type toxin (13.25 ng/100 mg; Fig. 4B
, inset). This ratio concurs with the 6.8 Ki ratio obtained by in vitro binding competition assays between mutant E15R and 125I-Bj-xtrIT using cockroach neuronal membranes (Table 1)
.
|
The antagonistic effect was further demonstrated in experiments of animal recovery from paralysis imposed by the wild-type toxin. Typically, blowfly larvae injected with doses higher than two ED50 equivalents of Bj-xtrIT contract immediately and remain paralyzed for at least 30 min. Contracted larvae (1 min after injection of 2–100 ED50 equivalents of Bj-xtrIT) recovered within seconds when injected with a sevenfold higher E15R dose (w/w). No such recovery was obtained when contracted larvae were injected with 100 µg BSA. Moreover, larvae could be protected from paralysis induced by eight ED50 equivalents of Bj-xtrIT by prior injection of excess (10-fold) E15R. A complete protective effect lasted up to 20 min after E15R injection and decayed within longer durations. This has indicated that E15R remains bound to the receptor site with an apparent half-life, which fits that calculated from the dissociation rate of 125I-Bj-xtrIT in binding assays using cockroach neuronal membranes (koff=0.91±0.25x10–3 s-1, n=4; t1/2=12.7 min). We conclude that the E15R mutant acts as a potent competitive antagonist of Bj-xtrIT.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
in a somewhat simplified form can be separated into two time-resolved steps: 1) Toxin binding to receptor site 4; this step proceeds readily at resting membrane potential, and does not involve toxin interaction with the voltage sensor; 2) trapping of the voltage sensor by the toxin. After a depolarizing conditioning prepulse, the DIIS4 segment moves outward, making its outermost residues accessible for toxin binding. Interaction of the prebound toxin with these residues traps the voltage sensor in an outward, activated position (6)
. This scenario suggests that two subsets of toxin residues are involved in the interaction with the NaCh. The first subset binds the channel to form a toxin receptor complex. Upon strong depolarization, the second subset interacts with the newly exposed DIIS4 outermost residues, leading to sensor trapping.
Assuming that at least part of the toxin residues involved in sensor trapping do not make specific interactions with the channel during the binding stage, one may predict that alteration of these residues could disrupt sensor trapping without having a substantial effect on toxin binding. This prediction was met by the mutagenesis results of Glu15. The finding that substitution of the charged Glu15 with both aliphatic (Ala) and aromatic (Phe) side chains had only minor effects on the binding affinity of Bj-xtrIT to cockroach neuronal membranes (Table 1)
suggests that this residue does not interact with the receptor during the binding stage. However, the complete loss of toxicity caused by the E15R mutation suggests that Glu15 does interact with a positively charged counterpart in receptor site 4 at a later stage that is crucial for toxin action. We propose that Glu15 of Bj-xtrIT intercepts DIIS4 by making direct electrostatic interactions with the outermost arginine(s) upon depolarization. This interception prevents the inward movement of the sensor, enhancing channel activation (6
, 7)
. Our proposal is supported by the finding that chemical modification of a cysteine residue introduced in place of the outermost arginine in DIIS4 of Nav1.4 channel caused a hyperpolarizing shift in channel activation. This effect was attributed to trapping of the voltage sensor in its outward, activated position (19)
and is similar to the typical effect of ß-toxins on sodium channel activation.
Common role of toxin negative residue(s) in voltage sensor trapping
Glu15 and Glu30, which are crucial for Bj-xtrIT activity, are conserved in other excitatory and anti-mammalian ß-toxins (Fig. 1
). Glu15 in these toxins protrudes to the solvent and is 12 Å from the second functionally important, negatively charged residue (Glu30 in Bj-xtrIT, Glu28 in the anti-mammalian ß-toxins; Fig. 5
). The latter residue is conserved in the -helix of all ß-toxins and is flanked by conserved hydrophobic/nonpolar residues (Tyr26 and Val34 in Bj-xtrIT; Fig. 1
). From the similar mode of action and binding site (receptor site 4 on insect and mammalian NaChs; 12–14) of these toxins as well as the striking resemblance in the spatial disposition of the negatively charged pair (Fig. 5
), these glutamates may play a similar role in the various ß toxins. We suggest that interaction of Glu30 with a basic residue in receptor site 4 during the binding step, determines the precise position of Glu15, in turn enabling its interaction with the emerging gating charge(s) upon membrane depolarization (voltage sensor trapping).
|
Some scorpion ß-toxins (e.g., depressant toxins) lack an acidic residue in the position equivalent to Glu15 in excitatory and anti-mammalian ß-toxins, and therefore may trap the voltage sensor in a different way. Still, the conservation of acidic residues in other gating modifier toxins acting on voltage-gated potassium and calcium channels (Fig. 1B
) may suggest they also intercept the voltage sensor by direct electrostatic interaction. A recent report by MacKinnon and co-workers describes an antibody that stabilizes the voltage sensor paddle and mimics the depolarization-dependent effect of the gating modifier toxin VSTX1 in a voltage-dependent K+ channel (20
, 21)
. The crystal structure of the Fab fragment in complex with this channel (KvAP, PDB accession number 1ors) reveals that it interacts with the outermost Arg (chain c, Arg117) of S4 via an Asp residue (chain B, Asp101). This finding indirectly supports our suggestion that a spatially conserved negative residue in various gating modifier toxins (Glu15 in excitatory and anti-mammalian scorpion ß-toxins) is involved in voltage sensor trapping.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
Received for publication August 4, 2003. Accepted for publication December 8, 2003.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
subunit enhances voltage-sensor trapping by a ß-scorpion toxin. J. Gen. Physiol. 118,291-301-helix at the C-terminus and its implications for interaction with insect sodium channels. Structure 6,1095-1103[Medline]
-like toxin from the scorpion Leiurus quinquestriatus hebraeus. J. Mol. Biol. 285,1749-1763[CrossRef][Medline]
-like toxins with the sodium channel receptor site inferred from toxin iodination and pH-dependent binding. J. Neurochem. 75,1735-1745[CrossRef][Medline]
This article has been cited by other articles:
|
|
 |
|
  L. Cohen, N. Lipstein, I. Karbat, N. Ilan, N. Gilles, R. Kahn, D. Gordon, and M. Gurevitz Miniaturization of Scorpion {beta}-Toxins Uncovers a Putative Ancestral Surface of Interaction with Voltage-gated Sodium Channels J. Biol. Chem., May 30, 2008; 283(22): 15169 - 15176. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G. Corzo, J. K. Sabo, F. Bosmans, B. Billen, E. Villegas, J. Tytgat, and R. S. Norton Solution Structure and Alanine Scan of a Spider Toxin That Affects the Activation of Mammalian Voltage-gated Sodium Channels J. Biol. Chem., February 16, 2007; 282(7): 4643 - 4652. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. Cohen, N. Lipstein, and D. Gordon Allosteric interactions between scorpion toxin receptor sites on voltage-gated Na channels imply a novel role for weakly active components in arthropod venom FASEB J, September 1, 2006; 20(11): 1933 - 1935. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Mantegazza and S. Cestele {beta}-Scorpion toxin effects suggest electrostatic interactions in domain II of voltage-dependent sodium channels J. Physiol., October 1, 2005; 568(1): 13 - 30. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. Cohen, I. Karbat, N. Gilles, N. Ilan, M. Benveniste, D. Gordon, and M. Gurevitz Common Features in the Functional Surface of Scorpion {beta}-Toxins and Elements That Confer Specificity for Insect and Mammalian Voltage-gated Sodium Channels J. Biol. Chem., February 11, 2005; 280(6): 5045 - 5053. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
I. Karbat, F. Frolow, O. Froy, N. Gilles, L. Cohen, M. Turkov, D. Gordon, and M. Gurevitz Molecular Basis of the High Insecticidal Potency of Scorpion {alpha}-Toxins J. Biol. Chem., July 23, 2004; 279(30): 31679 - 31686. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
