JBC

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Originally published In Press as doi:10.1074/jbc.M610462200 on March 5, 2007

J. Biol. Chem., Vol. 282, Issue 17, 12687-12697, April 27, 2007
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Supplemental Data
Right arrow All Versions of this Article:
282/17/12687    most recent
M610462200v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Smith, J. J.
Right arrow Articles by Blumenthal, K. M.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Smith, J. J.
Right arrow Articles by Blumenthal, K. M.
Social Bookmarking
Add to CiteULike   Add to Complore   Add to Connotea   Add to Del.icio.us   Add to Digg   Add to Reddit   Add to Technorati  
What's this?

Molecular Interactions of the Gating Modifier Toxin ProTx-II with Nav1.5

IMPLIED EXISTENCE OF A NOVEL TOXIN BINDING SITE COUPLED TO ACTIVATION*Formula

Jaime J. Smith{ddagger}, Theodore R. Cummins§, Sujith Alphy{ddagger}, and Kenneth M. Blumenthal{ddagger}1

From the {ddagger}Department of Biochemistry, School of Medicine and Biomedical Sciences, State University of New York, Buffalo, New York 14214 and §Department of Pharmacology and Toxicology, Indiana University School of Medicine, Indianapolis, Indiana 46202

Received for publication, November 9, 2006 , and in revised form, February 15, 2007.


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 EXPERIMENTAL PROCEDURES
 RESULTS
 DISCUSSION
 REFERENCES
 
Voltage-gated Na+ channels are critical components in the generation of action potentials in excitable cells, but despite numerous structure-function studies on these proteins, their gating mechanism remains unclear. Peptide toxins often modify channel gating, thereby providing a great deal of information about these channels. ProTx-II is a 30-amino acid peptide toxin from the venom of the tarantula, Thrixopelma pruriens, that conforms to the inhibitory cystine knot motif and which modifies activation kinetics of Nav and Cav, but not Kv, channels. ProTx-II inhibits current by shifting the voltage dependence of activation to more depolarized potentials and, therefore, differs from the classic site 4 toxins that shift voltage dependence of activation in the opposite direction. Despite this difference in functional effects, ProTx-II has been proposed to bind to neurotoxin site 4 because it modifies activation. Here, we investigate the bioactive surface of ProTx-II by alanine-scanning the toxin and analyzing the interactions of each mutant with the cardiac isoform, Nav1.5. The active face of the toxin is largely composed of hydrophobic and cationic residues, joining a growing group of predominantly Kv channel gating modifier toxins that are thought to interact with the lipid environment. In addition, we performed extensive mutagenesis of Nav1.5 to locate the receptor site with which ProTx-II interacts. Our data establish that, contrary to prior assumptions, ProTx-II does not bind to the previously characterized neurotoxin site 4, thus making it a novel probe of activation gating in Nav channels with potential to shed new light on this process.


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
EXPERIMENTAL PROCEDURES
 RESULTS
 DISCUSSION
 REFERENCES
 
Voltage-gated cation channels are integral membrane proteins that play a critical role in electrical signaling by controlling the flow of Na+, K+, and Ca2+ across the plasma membrane in response to changes in voltage. Nav channel {alpha} subunits are composed of four homologous domains, DI-DIV, each having six transmembrane segments, S1-S6. The first four segments of each domain comprise the voltage sensor of these proteins, whereas S5 and S6 form the central ion conducting pore (for review, see Ref. 1). The homologous Kv channels are tetramers of four identical subunits, each similar to a Nav channel domain.

Site-directed fluorescent labeling has shown that Nav domains I and II move with activation and are unaffected by fast inactivation gating, whereas domains III and IV exhibit kinetic components associated with deactivation and fast inactivation (2, 3). DIV-S4 has a unique role in gating in that its charges only move during inactivation (4, 5), and there is evidence for strong cooperativity among Nav channel domains throughout the gating process. In contrast, subunit coupling is only seen late in Kv channel gating transitions (6–8). It has been suggested that this major difference between Nav and Kv channel kinetics underlies the basis of fast electrical transmission, i.e. domain cooperativity is necessary for the rapid upstroke of an action potential (6). The functional differences among Nav channel domains raise the possibility that distinct domain structures might exist as well.

Given the many distinctions between Nav and Kv channels, detailed studies of Nav channel gating mechanisms are essential. Studies using neurotoxins that bind Nav channels with high affinity to alter conductance or gating properties have been extremely useful for this purpose. Polypeptide toxins derived from the venom of spiders, sea anemones, scorpions, and snails interact with voltage sensors to modify activation or inactivation and have been tremendously useful probes of gating mechanisms. These gating modifier toxins bind to sites 3 and 4, respectively. Site 3 has been localized to the extracellular S3/S4 linker of domain IV (9, 10), whereas residues in domain II S3/S4 make a major contribution to site 4 (11, 12).

Site 3 toxins, such as those from sea anemone venom, delay channel inactivation upon binding, most likely by inhibiting the normal outward movement of gating charges in DIV, resulting in the inability of the inactivation particle to mobilize (5). In contrast, site 4 toxins enhance channel activation and shift the voltage dependence of activation to more hyperpolarized potentials via a voltage sensor-trapping mechanism (11, 13). It is thought that the open probability of the channel increases because of the toxin locking the channel in its activated conformation after a depolarizing pre-pulse. ProTx-II is a 30-amino acid peptide toxin purified from the venom of the tarantula, Thrixopelma pruriens, that modifies activation of both Nav and Cav, but not Kv, channel isoforms by inhibiting peak current and shifting the voltage dependence of activation to more depolarized potentials (14, 15). ProTx-II conforms to the inhibitory cystine knot (ICK)2 motif, a common structural fold among spider toxins targeting ion channels (see The KNOTTIN database online). ICK peptides are defined by a 1-4, 2-5, 3-6 cystine connectivity and often have limited regular secondary structure. Based on its ability to modify activation, but not inactivation kinetics, it has been suggested that ProTx-II binds to site 4, but no direct evidence exists to validate this claim (14–17).

The Nav channel isoform-specific actions of other tarantula venom ICK peptide toxins have been characterized. These toxins modify activation by inhibiting sodium current and causing a depolarizing shift in gating (16). Like ProTx-II, their receptor sites have not been identified. However, detailed studies on the Kv channel ICK toxins hanatoxin and SGTx have shown that they inhibit potassium current by shifting channel opening to more depolarized potentials via an interaction with a receptor site on the C-terminal end of S3 near the extracellular surface (18, 19). The similar functional effects of ProTx-II and these Kv channel toxins on their targets raise the possibility that their receptor sites are similar as well, although Nav channel asymmetry will likely introduce an additional level of complexity into binding site identification.

Because ProTx-II modifies activation in a manner distinct from the previously characterized site 4 toxins, it is important to understand the basis for its activity. In the present study we investigate the bioactive surface of ProTx-II by alanine-scanning the peptide and analyzing the interactions of the mutants with the cardiac isoform, Nav1.5, by whole-cell voltage clamp. In addition, we carried out extensive mutagenesis of Nav1.5 to ascertain whether ProTx-II, like other toxins that modify Nav channel activation, interacts with receptor site 4. Our results indicate that the active face of ProTx-II consists of many hydrophobic as well as cationic residues that likely interact with a receptor site on Nav channels that is separate and distinct from site 4.


   EXPERIMENTAL PROCEDURES
TOP
 ABSTRACT
 INTRODUCTION
 EXPERIMENTAL PROCEDURES
RESULTS
 DISCUSSION
 REFERENCES
 
Molecular Biology
ProTx-II—The ProTx-II coding sequence and upstream enterokinase site were amplified from a previously constructed expression vector in our laboratory (15) using standard PCR procedures and cut with EcoRI/HindIII using sites introduced in the primers. The cleaved product was then subcloned into an octahistidine version of the pMALc2x vector (New England Biolabs) between EcoRI and HindIII using standard molecular biology protocols. As described previously, two additional N-terminal amino acids derived from a StuI restriction site remain in the ProTx-II coding sequence (15). Site-directed mutagenesis was performed using the QuikChange method (Stratagene, La Jolla, CA) to create all recombinant toxin mutants described in this paper, and all constructs were verified by sequencing.

Molecular Biology
NaV1.5—The pBluescript plasmid containing the SCN5A gene encoding human Nav1.5 {alpha} subunit was used for all channel mutant constructs. We initially sought to swap the extracellular S3/S4 linkers in domains II and IV to assess the functional effects of toxins that are known to bind these regions. To replace DIV S3/S4 with its DII counterpart, five residues (SPTLF) were deleted in DIV S3/S4 using a loop out PCR procedure to obtain a DIV S3/S4 linker that matched the length of DII S3/S4. The remaining linker residues were then mutated to match the sequence of DII-S3/S4 using the QuikChange method. We named this construct II:II. To create the construct in which the DIV S3/S4 linker replaced the corresponding DII linker, splicing by overlap extension (20) was used to insert five residues (SPTLF) downstream of the DII S3/S4 linker to match the length of DIV S3/S4. This was followed by mutation of the eight upstream residues in DII S3/S4 to match the complete sequence to that of DIV S3/S4. We called this construct IV:IV. The swapped construct was created by digesting the IV:IV construct with AgeI/NheI to remove the newly created S3/S4 linker region in DII and subcloning that region into the II:II construct digested with AgeI/NheI. This new construct became IV:II.

Sites in the remaining extracellular linker regions were targeted by mutating residues in Nav1.5 that differ from Nav1.7, an isoform for which ProTx-II has a 100-fold higher affinity (14, 21). 1–7 amino acid mutations were made in a single primer set to account for any binding determinant in that particular linker using an inverse PCR protocol. Transmembrane segment mutations were made as single amino acid replacements.

Production of Recombinant Toxins
Wild-type and mutant toxins were expressed as fusion proteins containing maltose-binding protein upstream of the toxin in Escherichia coli BL21 (DE3) as described (15). After lysis in a French press, the supernatants obtained were purified on Ni2+-nitrilotriacetic acid resin and reduced with 10 mM dithiothreitol for 1 h at 37 °C. After diluting the proteins to 0.2 mg/ml, they were dialyzed against 2.5 mM GSH, 50 mM Tris, 100 mM NaCl, pH 8.3. After dialysis, the proteins were oxidized by dropwise addition of GSSG to a final concentration of 0.5 mM and allowed to incubate for 72 h. Fusion proteins were then dialyzed against 50 mM NH4HCO3 and cleaved overnight at room temperature with enterokinase (Novagen/EMD Biosciences). Toxins were purified via RP-HPLC as described (15). Molecular weights of purified toxins were confirmed by MALDI-TOF mass spectroscopy analysis on a Bruker Biflex IV spectrometer. Because of folding difficulties encountered with a subset of mutants, some positions were mutated to an amino acid with a larger side chain to facilitate proper packing. These include K4Q, R13Q, W24L, K26Q, and K27Q.

Folding of Synthetic Toxins
In addition to recombinant mutant toxins, some synthetic mutants were studied, including Y1A, S11A, K14A, E17A, L23A, K28A, L29A, and W30A (GenScript Corp., Piscataway, NJ). Lyophilized peptides were resuspended to a peptide concentration of 5 mg/ml in nitrogen saturated 8 M urea, 50 mM Tris, 50 mM NH4HCO3, 120 mM GSH, pH 7.8. The peptides were then diluted to a concentration of 0.5 mg/ml and a GSH concentration of 12 mM. A final dilution brought the peptide concentration to 0.125 mg/ml and urea to 2 M, whereas GSH remained at 12 mM. GSSG was then added dropwise to a final concentration of 1.2 mM and allowed to incubate at 4 °C for 48 h. Samples were taken at various stages throughout the folding reactions for RP-HPLC analysis. MALDI-TOF analysis of samples confirmed that the peptides were oxidized. To purify folded peptides, we used cation exchange chromatography (HiTrap SP FF, GE Healthcare) followed by RP-HPLC. We were unable to produce significant amounts of the G18A mutant, presumably because of its inability to fold.

Cell Culture and Electrophysiology for Wild-type and Mutant ProTx-II Studies
All cell culture reagents were purchased from Invitrogen. Standard whole-cell voltage-clamp recordings were made from all cells. To analyze the effects of wild-type and mutant ProTx-II on the human cardiac Nav channel, a stable cell line expressing Nav1.5 was constructed in HEK 293 cells as previously described (22). To verify previously reported affinity data for ProTx-II on the peripheral nerve Nav channel, a stable HEK 293 cell line expressing human Nav1.7 was utilized (23). To study the effects of ProTx-II on neuronal Nav channels, the murine neuroblastoma cell line, N1E-115 was obtained from ATCC (Manassas, VA). Because these cells predominantly express Nav1.2, we ascribe ProTx-II modification to this isoform (24). Cells were maintained in Dulbecco's modified Eagle's medium supplemented with 10% fetal bovine serum and 1% penicillin/streptomycin at 37 °C in a 5% CO2 atmosphere. 200 µg/ml G418 was used for selection of HEK 293 cells expressing Nav1.5 or Nav1.7 channels. Toxin affinities for Nav1.5 and Nav1.2 channels were measured after introduction by gravity perfusion into a 300-µl bath chamber at a flow rate of ~3 ml/min. For Nav1.7 measurements, toxin was diluted into a 250-µl recording chamber and mixed by repeatedly pipetting 25 µl over ~5 s to achieve the specified concentration. All toxin solutions contained 1 mg/ml bovine serum albumin to prevent adsorption to tubing. Single cell recordings were made at room temperature using an Axopatch 200B amplifier with a Digidata 1322A analogue to digital converter and pCLAMP software (Axon Instruments). Pipettes were pulled from borosilicate glass (World Precision Instruments) and fire-polished to a final resistance of 1–3 megaohms when filled with recording solution. Solutions used for sodium current measurements through Nav1.5 channels contained the following: bath solution, 10 mM NaCl, 130 mM CsCl, 2 mM CaCl2, 10 mM HEPES, pH 7.4 with CsOH; pipette solution, 95 mM CsF, 30 mM CsCl, 5 mM NaCl, 10 mM EGTA, 10 mM HEPES, pH 7.0 with CsOH. Solutions used for Nav1.7 channels contained the following: bath solution, 140 mM NaCl, 3 mM KCl, 1 mM MgCl2, 1 mM CaCl2, 10 mM HEPES, pH 7.3; pipette solution, 140 mM CsF, 1 mM EGTA, 1 mM MgCl2, 10 mM NaCl, 10 mM HEPES, pH 7.3. Solutions used for Nav1.2 channels contained: bath solution, 70 mM NaCl, 70 mM CsCl, 2 mM CaCl2, 10 mM HEPES, pH 7.4; pipette solution, 10 mM NaCl, 90 mM CsF, 30 mM CsCl, 10 mM EGTA, 10 mM HEPES, pH 7.0. Recordings were initiated 5–8 min after patch rupture.

To measure time courses of modification and dissociation, cells were held hyperpolarized at -130 mV then stepped to either -30 mV (Nav1.5 channels) or -10 mV (Nav1.2 channels) for 10 ms at a frequency of 1 Hz as either toxin-containing or toxin-free solution was introduced into the bath. To generate current-voltage (I-V) relationships, cells were stepped in 5-mV increments from -80 to +20 mV (Nav1.5 channels), from -80 to +70 mV (Nav1.7 channels), or from -50 to +50 mV (Nav1.2 channels) from a holding potential of -120 or -130 mV for 30 ms (Nav1.5 and 1.2) or 50 ms (Nav1.7), and the resulting peak currents were plotted against voltage. Current-voltage relationships were obtained just before toxin application and after steady-state inhibition was achieved. Only cells having a >4.2-mV slope were included. Currents were capacity-corrected using MatLab 6.5 (The MathWorks, Inc., Natick, MA). Only cells with a leak resistance of >750 megaohms were included in analyses. Toxin test concentrations ranged from 250 to 20 µM and were determined empirically for each toxin mutant.

Conductance-voltage (g-V) relationships quantitating the voltage dependence of activation were obtained from peak current-voltage (I-V) relationships according to g = INa/V - Vr, where INa is the peak Na+ current at test potential V, and Vr represents reversal potential. To assess the voltage dependence of inactivation, a 2-step protocol was used in which cells were stepped in 5-mV increments from -130 to -30 mV for 300 ms followed by a step to the test potential -30 mV for 20 ms to evaluate channel availability. Normalized activation and inactivation curves were fit to a Boltzmann function y = 1/[1 + e(V - V0.5)/k], where y is normalized gNa or INa, V is membrane potential, V0.5 is the midpoint of activation or inactivation, and k is the slope factor. Data are depicted as ±S.E.

The half-blocking concentration (IC50) for ProTx-II on Nav1.7 was calculated based on the single-site Langmuir inhibition isotherm using the following function: (Itoxin/I0) x [toxin]/(1 - Itoxin/I0), where I0 and Itoxin are the peak sodium currents measured with a test pulse to -30 mV before and after application of toxin, respectively, and[toxin] is the concentration of toxin.

Cell Culture and Electrophysiology for Mutant NaV1.5 Studies
Mutant channel DNA was transiently transfected into HEK 293 cells followed by whole-cell voltage clamp analysis to assess mutant channel function and interaction with wild-type ProTx-II. Cells were maintained in Dulbecco's modified Eagle's medium supplemented with 10% fetal bovine serum and 1% penicillin/streptomycin. Lipofectamine reagent in combination with PLUS reagent (Invitrogen) was used for transient transfections, and after 24–36 h of incubation cells were trypsinized and moved to coverslips for analysis. Because of lower expression levels of mutant channels, [Na+] was increased in the bath solution to either 70 mM (medium Na+) or 140 mM (high Na+), and [Cs+] was adjusted accordingly. Other components were kept constant as described above. Corresponding pipette solutions contained either 35 mM NaCl, 65 mM CsF (medium Na) or 70 mM NaCl, 30 mM CsF (high Na). The voltage dependence of activation and inactivation of linker swap mutants was analyzed and compared with wild-type Nav1.5 to ensure normal channel function using protocols described above. A wild-type toxin concentration of 1 µM was used to assess modification of mutant channels. We looked for a loss of inhibition during simple step protocols from -130 to -30 mV that would indicate the receptor site on the channel had been disrupted.

Kinetic Calculations
Peak currents were plotted versus time from the point of toxin wash-in or wash-out. The data were fit to a first order exponential decay equation using Origin 6.1 software (Microcal Software Inc., Northampton, MA). We calculated the kinetic constants for channel modification (kmod) and toxin dissociation (koff) using the inverse of {tau} of the fit for toxin wash-in and wash-out, respectively. To determine the rate of toxin association, the following equation was used: kon = kmod - koff/[toxin]. The dissociation constant was determined using koff/kon = KD. Data are reported as ±S.E.

Molecular Modeling
An energy-minimized molecular model of ProTx-II was created using the Protein Data Bank coordinates for HpTx-2, an ICK motif peptide targeting Kv4 channels (PDB code 1emx; Refs. 25 and 26). Conversion to the ProTx-II sequence was done in the Biopolymer module of InsightII, and the resulting model was then subjected to energy minimization (initially using a steepest descents protocol followed by at least 2500 cycles of conjugate gradients) in Discover to remove steric clashes. After minimization, the total energy of the model structure was ~300 kcal/mol, and it retained the backbone structure typical of the ICK motif.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 EXPERIMENTAL PROCEDURES
 RESULTS
DISCUSSION
 REFERENCES
 
Functional Characterization of Wild-type and Mutant Forms of ProTx-II—The first aim of the present study was to characterize the effects of ProTx-II mutants on Nav1.5 to isolate the pharmacophore of the toxin molecule. In addition, we hoped to identify the channel receptor site and ultimately establish a mechanism of action for this novel acting peptide toxin. We produced wild-type ProTx-II and several mutated forms either recombinantly or synthetically and purified them to homogeneity using RP-HPLC. To characterize the effects of wild-type or mutated ProTx-II on Nav1.5, we used a whole-cell voltage clamp on HEK 293 cells stably expressing this channel. We evaluated the extent of channel modification by depolarizing the cell membrane to -30 mV as toxin-free solution was replaced with toxin-containing solution. At 1 µM ProTx-II, we observed rapid and near complete inhibition of sodium current ({tau} ~ 2.5 s) (Fig. 1A). This inhibition was completely reversible upon toxin wash-out ({tau} ~ 40 s) (Fig. 1B). To verify that toxin binding is concentration-dependent, we examined channel modification over the range of 250–1000 nM ProTx-II. Modification decreased accordingly at both concentrations ({tau}250 nM = 9.3 ± 0.23 S, n = 3; {tau}500 nM = 5.48 ± 0.17 S, n = 3). In contrast, dissociation remained a zero-order reaction as expected ({tau}250 nM = 53.3 ± 3.2 S, n = 3; {tau}500 nM = 41.7 ± 2.2 S, n = 3). ProTx-II also shifts the voltage dependence of gating to more depolarized potentials, indicating that the toxin does not inhibit through a pore-blocking mechanism but, rather, interferes with the energetics of gating (15). Analysis of ProTx-II on steady-state activation and inactivation kinetics revealed that the midpoint of the activation curve shifted by 23 mV in the depolarizing direction, whereas inactivation remained unaffected by the toxin (Fig. 2, A and B). This mode of channel modification is similar to that of other ICK toxins that target Nav and Kv channels but very different from site 4 toxins targeting Nav channels. Our kinetic analysis yielded an equilibrium dissociation constant for ProTx-II of 93 nM, similar to the value obtained using natural material (14).


Figure 1
View larger version (12K):
[in this window]
[in a new window]

  		FIGURE 1.
Kinetics of ProTx-II modification and dissociation. HEK 293 cells stably expressing Nav1.5 were stepped to -30 mV from a holding potential of -130 mV for a duration of 10 ms at a frequency of 1Hz. A, the addition of 1 µM ProTx-II, indicated by the arrow, results in almost complete inhibition of Na+ current ({tau} ~ 2.5 s). The first and last current traces from toxin wash-in are shown in the inset. B, ProTx-II inhibition is completely reversible upon toxin wash-out ({tau} ~ 40 s); same representative cell as shown in panel A. Data are normalized to the peak current value. (n = 12 cells).

 
To identify side chains of ProTx-II that play key functional roles, we introduced point mutations at every position outside of the cystine framework. To overcome folding difficulties encountered with some mutants, it was necessary to introduce amino acids other than alanine at some positions. We consider a mutant properly folded based on several criteria. First, an HPLC retention time similar to wild type as well as a distribution as a dominant homogenous chromatographic form suggests a correct folding pattern. Second, a molecular weight identical to the calculated mass of oxidized toxin as analyzed by MALDI-TOF indicates that all disulfide bonds have formed. Finally, a circular dichroism spectrum that overlays that of wild-type toxin is characteristic of a normal fold. Fig. 3 compares the far UV CD spectra of wild-type ProTx-II to those of the homologous ICK motif-gating modifiers HpTx-2 and GsMTx-4. The CD spectra of all ProTx-II mutants we examined overlay with that obtained from wild-type toxin and are similar to those of HpTx2 and GsMTx4. G18A was the only mutant toxin that we were unable to produce in significant yield, perhaps indicating a unique requirement for flexibility at this position during folding. All other mutants met the above criteria.


Figure 2
View larger version (15K):
[in this window]
[in a new window]

  		FIGURE 2.
ProTx-II modifies the steady-state voltage dependence of activation, but not inactivation, of NaV1.5 channels. A, ProTx-II shifts the steady-state activation curve 23 mV in the depolarizing direction. Cells (n = 10) stably expressing Nav1.5 were stepped in 5-mV increments from -80 to +20 mV from a holding potential of -130 mV for 30 ms. For control cells ({circ}), the midpoint of activation Va =-42.5 ± 0.57mV, and the slope factor k = 5.91 ± 0.22 mV; for cells treated with 1 µM ProTx-II (•), midpoint of activation Va =-19.6 ± 0.5 mV, and the slope factor k = 7.48 ± 0.32 mV. B, ProTx-II has no effect on steady-state inactivation. Cells (n = 8) were stepped in 5-mV increments from -130 mV to -30 mV for 300 ms followed by a test pulse to -30 mV for 20 ms. For control cells ({circ}), the midpoint of inactivation Vh = -79.2 ± 0.7 mV, and the slope factor k = 4.26 ± 0.20 mV; for cells treated with 1 µM ProTx-II (•), the midpoint of inactivation Vh =-79.5 ± 0.7 mV, and the slope factor k = 3.91 ± 0.44 mV. Normalized conductances (activation) and currents (inactivation) were generated for both data sets, fit to a Boltzmann function, and are shown as data ± S.E.

 


Figure 3
View larger version (19K):
[in this window]
[in a new window]

  		FIGURE 3.
Circular dichroism spectra of ICK motif toxins. Spectra of ProTx-II ({triangleup}), HpTx-2 ({circ}), and GsMTx-4 ({diamond}) were recorded at peptide concentrations of 0.15 mg/ml in a Jasco-720 spectropolarimeter at a scan rate of 20 nm/min in 5 mM sodium phosphate buffer, pH 6.9. All spectra shown represent the average of four complete sweeps and were smoothed using the Jasco data analysis package. The CD spectra of all ProTx-II mutants reported in Table 1 overlay that of the wild-type toxin.

 


Figure 4
View larger version (17K):
[in this window]
[in a new window]

  		FIGURE 4.
Modification of NaV1.5 currents by ProTx-II mutants. Onset kinetics for toxins with mutations to hydrophobic or cationic residues are dramatically decreased as compared with wild-type ProTx-II. Cells stably expressing Nav1.5 were stepped to -30 mV from a holding potential of -130 mV for 10 ms every 1 s until modification was complete. Data shown are from representative cells (n ≥ 3, except for K27Q, where insufficient toxin was available), and currents are normalized to peak current values. Results shown are representative of the three general phenotypic classes observed; that is, minimal loss of affinity, 10–20-fold loss of affinity, and >20-fold loss of affinity. Toxin concentrations used were wild type, D10A, and E12A, 1 µM; W7A, W30A, and K27Q, 5 µM; M6A, 20 µM.

 
After purification of each mutant toxin, we characterized its interaction with Nav1.5 as described for wild-type ProTx-II using an empirically determined test concentration. Mutant toxins were tested at concentrations at or above their KD except in cases where insufficient availability of toxin made this prohibitive. Onset kinetics for a subset of mutants are compared with that of wild-type ProTx-II in Fig. 4. We define a residue as essential if its mutation results in a loss of affinity of at least 10-fold. As shown in Table 1, the affinities of 11 mutants do not differ substantially from that of wild-type toxin, suggesting that these residues do not play a significant role in toxin-channel interactions. Most of these sites were polar to alanine substitutions. In contrast, 10 mutant toxins show losses of affinity from 10- to 125-fold. W5A and K26Q are inactive at concentrations up to 40 µM, indicating either a severely reduced channel affinity or a folding impairment. Interestingly, most of the essential residues are hydrophobic. Analysis of the ICK toxin SGTx, which targets Kv channels, has identified a similar number of hydrophobic residues that are important for channel modification (19). It has been suggested that hydrophobic amino acids may contribute to interactions with the surrounding lipid membrane, but this has yet to be directly examined. It is possible that these residues contribute to direct protein-protein interactions within the toxin-channel complex as well.


View this table:
[in this window]
[in a new window]

  		TABLE 1
Interactions between ProTx-II mutants and Nav1.5 Kinetic analysis of 23 ProTx-II mutants compared to wild-type toxin identifies 10 residues likely to be involved in channel modification. For five of these, the loss of affinity is dominated by kon, whereas only one is dominated by koff. The remaining residues that are essential for channel modification exhibit a mixed effect on both rates. W5A and K26Q were inactive up to 40 µM. N/D, not determined.

 
In addition to hydrophobic residues, a subset of the seven cationic residues, including Lys-27, Arg-13, and Arg-22, were identified as critical for toxin activity. Neutralization of the anionic residues, Asp-10, Glu-12, and Glu-17, resulted in no loss of affinity, indicating that positively charged channel residues are most likely not involved in direct binding interactions.

Increases in KD can be dominated by either an increased rate of dissociation or a decreased rate of association. For example, our data show that the loss of affinity observed for M6A, W7A, L29A, and W30A is wholly ascribable to decreased rates of association. Taken together with our functional analysis that shows hydrophobic residues are important binding determinants, this may have implications for how the toxin accesses a potentially membrane-restricted receptor site. If these residues were involved in initial interactions with lipid membrane before the toxin molecule reaches its channel receptor site, we might expect that mutational effects would be restricted to kon.

Analysis of Mutant Nav1.5 Interactions with Wild-type ProTx-II—To identify the channel receptor site responsible for interacting with ProTx-II, we employed a mutagenesis strategy. Nav channel site 4 has been suggested as the receptor site for ProTx-II and similar NaV channel toxins (14–17). Because the major site 4 epitope within the domain II S3/S4 linker was first identified (11, 12), additional components have been detected in other extracellular linker regions including DIII SS2-S6 (27). However, because the II S3/S4 site makes the largest single contribution to affinity for site 4 scorpion toxins, we chose to target this region first. We created mutant Nav1.5 constructs in which the DII and DIV S3/S4 linkers were swapped. This approach yielded a IV:II mutant in which the sequence of DIV S3/S4 replaced the DII S3/S4 linker and vice versa. Using this nomenclature, the wild-type construct would be referred to as II:IV and the swapped construct as IV:II. As described under "Experimental Procedures," obligatory intermediates in this strategy were the II:II and IV:IV channels. The availability of IV:IV allowed us to directly test the interaction of ProTx-II with site 4.

Cestele et al. (11) have shown that the site 4 scorpion toxin CssIV from the venom of Centruroides suffusus suffusus shifts the voltage dependence of NaV1.2 activation in a hyperpolarizing direction, thereby enhancing activation when currents are analyzed after a prepulse to +50 mV. In contrast, NaV1.5 activation is shifted in the opposite direction and inhibited slightly (11). To ascertain whether ProTx-II also behaves as a site 4 toxin, we first examined its effects on current-voltage relationships in NaV1.2 and 1.5. As shown in Fig. 5A, application of ProTx-II (1 µM) to HEK 293 cells expressing human NaV1.5 results in a rightward shift in the I-V curve and an ~70% inhibition of maximum current. This is qualitatively similar to the effects of CssIV, but the effects of ProTx-II on both inhibition and I-V shift are much larger. We next analyzed its effects on NaV1.2 currents using the protocol described under "Experimental Procedures." This protocol does not include any prepulse step analogous to that used by Cestele et al. (11). As shown in Fig. 5B, under these conditions ProTx-II causes a depolarizing shift in the I-V relationship and inhibits current to an extent comparable with that seen with NaV1.5. These results are clearly in contrast to those observed for CssIV and are the first indication that ProTx-II is not a classical site 4 toxin.


Figure 5
View larger version (22K):
[in this window]
[in a new window]

  		FIGURE 5.
The electrophysiologic phenotype of ProTx-II is distinct from that of the site 4 toxin, CssIV. {circ}, control; •, toxin. A, the effect of 1 µM ProTx-II on Nav1.5. Cells stably expressing Nav1.5 were stepped in 5-mV increments from -80 mV to +20 mV from a holding potential of -130 mV for 30 ms. An ~70% reduction in peak current and a 20-mV depolarizing shift in the voltage dependence of activation were observed (n = 12). B, the effect of 1 µM ProTx-II on Nav1.2. N1E-115 cells expressing predominantly Nav1.2 were stepped in 5-mV increments from -50 mV to +50 mV from a holding potential of -130 mV for 30 ms. A ~80% reduction in peak current and a 5-mV depolarizing shift in the voltage dependence of activation were observed (n = 5). The experimentally determined KD for ProTx-II on Nav1.2 was ~200 nM. C, the effect of 1 µM ProTx-II on the Nav1.5 IV:IV mutant channel which lacks the DII S3/S4 linker. Cells transiently expressing the IV:IV channel were subjected to the same step protocol as described in A. A similar extent of modification to that of wild-type Nav1.5 was observed, indicating retention of the binding site (n = 7). The experimentally determined KD for ProTx-II on the IV:IV channel was ~89 nM. D, the effect of 200 nM CssIV on the Nav1.5 mutant IV:IV channel. Cells transiently expressing the IV:IV channel were subjected to the same step protocol as described in A and C. CssIV failed to modify IV:IV due to ablation of the binding site (n = 3). Data are ±S.E.

 
Although CssIV has been shown to interact with multiple extracellular sites in NaV1.2, the most dramatic effect on its affinity is seen upon mutation of G845 in the domain II S3-S4 linker. We, therefore, analyzed the ability of our NaV1.5 IV:IV channels to be modified by ProTx-II and CssIV in separate experiments. As shown in Fig. 5C, we see near inhibition of sodium current by 1 µM ProTx-II in this construct, indicating that the toxin does not bind to site 4 and likely interacts with an unknown neurotoxin receptor site. To demonstrate that we did in fact obliterate site 4, we next tested the functional effect of CssIV on IV:IV channels. At a concentration of 200 nM, CssIV had no effect on sodium current as compared with toxin-free control (Fig. 5D), although the same CssIV concentration resulted in 65% inhibition of current and a depolarizing shift in wild-type Nav1.5 (n = 2). This effect is consistent with that reported previously by Cestele et al. (11). We then performed the same experiment using ProTx-II on the II:II channel to rule out its interaction with site 3 and saw the same extent of modification as observed with wild-type Nav1.5 and IV:IV. To confirm abolition of site 3, we tested a well characterized site 3 toxin, ApB, on II:II and reproducibly observed a 90% reduction in activity. These results verify our expectation that ProTx-II also fails to bind to site 3.

Because ProTx-II does not bind to the previously characterized gating modifier neurotoxin receptor sites 3 or 4, and because all gating modifier toxins characterized to date have been shown to bind within or proximal to extracellular linkers, we next targeted the remaining extracellular linker regions. Our mutagenesis strategy exploited the fact that ProTx-II has been reported to have an ~50-fold higher affinity for the peripheral nerve sodium channel isoform, Nav1.7, over the cardiac isoform, Nav1.5 (Fig. 6; Refs. 14 and 21). Alignment of Nav1.5 and Nav1.7 S1/S2 and S3/S4 linker sequences identified 21 residues that differ between the two isoforms. Additionally, 5 residues in DIII SS2/S6 were targeted, based on their identification as determinants of site 4 toxin isoform selectivity (27). Because we had already demonstrated that DII and DIV S3/S4 linkers are not involved in ProTx-II binding, we eliminated them from this analysis and mutated the remaining residues to alanine (Table 2). All residues targeted for mutation in a single linker were altered simultaneously rather than as single point mutations to account for all possible binding contributions within a single extracellular linker. We were able to express all mutant channels after transient transfection of DNA into HEK293 cells. Upon the addition of 1 µM ProTx-II, all channel mutants tested were modified to a similar extent as observed for wild-type channels (Fig. 7). We interpret these data as evidence that ProTx-II does not make any critical binding interactions with extracellular linker regions of sodium channels.


View this table:
[in this window]
[in a new window]

  		TABLE 2
Mutational analysis of Nav1.5 To determine the ProTx-II receptor site, a total of 76 residues were mutated in transmembrane or extracellular linker regions of Nav1.5 either singly or as part of a larger substitution. The majority of linker residues were selected for mutation based upon sequence conservation between Nav1.5 and Nav1.7, and detailed analysis of these channels is depicted in Fig. 7. Acidic transmembrane residues were targeted based on (a) the importance of several cationic residues for ProTx-II affinity, (b) the importance of several hydrophobic residues for ProTx-II affinity in conjunction with the ability of ProTx-II to bind phospholipids, both, raising the possibility of a membrane-restricted binding site, and (c) the importance of transmembrane residues for binding of the ICK toxin, hanatoxin, to Kv2.1 (34). These data are depicted in supplemental Table I. The extent of modification of each mutant channel by 1 µM ProTx-II was similar to that observed for wild-type Nav1.5, i.e., 65–85% inhibition (n = 2–3 per mutant). Mutations in bold font were made as part of a large substitution in that particular linker. All others were created as single mutations.

 
Because we have identified several hydrophobic and cationic residues in ProTx-II as essential for modification of Nav1.5, we next focused on potential binding sites within transmembrane segments. It has been suggested that gating modifier toxins might access their receptor site on their target channel after diffusing through the lipid membrane to reach a membrane-restricted site (28–30). We, therefore, created single site charge neutralization mutations of several anionic transmembrane residues in Nav1.5 and analyzed their ability to be modified by ProTx-II (Table 2 and supplemental Table 1). All mutant channels displayed near complete inhibition of current upon the addition of toxin, indicating that ProTx-II does not make critical binding interactions with any of the targeted residues in transmembrane segments.


Figure 6
View larger version (22K):
[in this window]
[in a new window]

  		FIGURE 6.
NaV1.7 channels are more sensitive to ProTx-II than NaV1.5 channels. Representative Nav1.7 currents recorded from a HEK 293 cell are shown under control conditions (A) and after application of 100 nM ProTx-II (B). The cell stably expressing Nav1.7 was stepped in 5-mV increments from -80 mV to +70 mV from a holding potential of -120 mV for 50 ms. ProTx-II inhibition was allowed to reach steady state before recording the traces shown in B. C, the current-voltage relationship from this cell is shown under control ({circ}) conditions and after exposure to toxin (•). D, the dose-response relationship for ProTx-II inhibition of human Nav1.7 (•; n = 14) and human Nav1.5 ({circ}; n = 9) channels stably expressed in HEK 293 cells. Current amplitudes were obtained before toxin application with 50-ms test pulses to -30 mV from a -120-mV holding potential. The solid curves show the single-site Langmuir inhibition isotherm fits to the data.

 

   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 EXPERIMENTAL PROCEDURES
 RESULTS
 DISCUSSION
REFERENCES
 
The molecular mechanisms fundamental to channel modification by sites 3- and 4-gating modifier toxins have been the focus of several in depth studies, and the results have shed light on the roles of individual domains in voltage sensing (5, 11, 13). However, much less is currently known about the mechanism of action of ICK peptides on Nav channels. We, therefore, investigated the interaction of the ICK peptide toxin ProTx-II and 23 mutant forms with the human cardiac Na+ channel isoform, Nav1.5. ProTx-II modifies Nav and Cav channels in a manner distinct from previously characterized Nav channel gating modifier toxins, and its channel receptor site is unknown.

Our results identify all functionally essential residues on the toxin surface, thereby establishing the bioactive face of the molecule. Of the 24 non-cysteine residues, we were able to produce and characterize 23 mutants (Table 1), all of which have CD spectra identical to that of the wild-type toxin. Interestingly, whereas these spectra closely resemble those of homologous gating modifier toxins of the ICK family, there are also clear differences among them, and all are dramatically different from a fourth homolog, SgTx-1 (19). In addition, we note that although these spectra are atypical for native proteins, all of these toxins are in fact fully active. It is, therefore, likely that these polypeptides as a group have very little regular secondary structure, perhaps contributing to their ability to target distinct channels and channel isoforms with high affinity.

Eleven ProTx-II mutants exhibit activity similar to that of wild-type toxin, and most of the residues having severely decreased channel affinities are hydrophobic. The N-terminal mutations M6A and W7A in addition to the C-terminal mutants W24L, L29A, and W30A exhibit KD values 10–76-fold higher than wild-type toxin, clearly demonstrating that these positions contribute strongly to binding affinity. Trp-5 may contribute as well since we observed no channel inhibition by W5A up to a concentration of 40 µM, and because the corresponding position has been implicated as a binding determinant in the homologous Kv channel ICK toxin, SGTx (19). As seen in Fig. 8, these residues contribute to a hydrophobic protrusion on the surface of the ProTx-II. A similar cluster of functionally important hydrophobic residues has been shown to form a protrusive patch on the active face of SGTx (19). This common feature points toward a dominant role of hydrophobic residues in ICK toxin activity irrespective of the target channel. Furthermore, our data highlight the importance of Met-19 and Val-20 for toxin activity with affinities of their alanine replacements decreased ~35 and 77-fold, respectively. The importance of these positions in channel modification is unique to ProTx-II since Ala-20 is a nonessential residue in SGTx, and Leu-19 is a folding determinant (19).


Figure 7
View larger version (28K):
[in this window]
[in a new window]

  		FIGURE 7.
ProTx-II does not bind to extracellular linkers of NaV1.5. Top, transmembrane architecture of the Nav channel {alpha} subunit is provided as a reference to channel mutations indicated in B as well as the remaining mutations listed in Table 2. Bottom, mutant Nav1.5 channels containing alanine replacements for all extracellular linker residues which differ between Nav1.5 and Nav1.7 were transiently expressed in HEK 293 cells. Data shown are scaled currents from representative cells (n ≥ 3) stepped in 5-mV increments from -80 mV to +20 mV from a holding potential of -130 mV. Inhibition by 1 µM ProTx-II (•) ranged from 65 to 85% for all mutant channels as compared with toxin-free control ({circ}). Amino acid linker sequences are shown above each trace with the mutated residues depicted in large bold font. DII S3/S4 and DIV S3/S4 linker mutants are not represented because they were analyzed separately in the linker swap experiments.

 
In addition to illustrating the importance of hydrophobic amino acids in ProTx-II, we also demonstrate that the bioactive face is amphipathic in nature. The mutations R13Q, R22A, and K27Q cause dramatic increases in KD, suggestive of a role in electrostatic interactions with the channel or phospholipid head groups. Although K26Q also lacks activity up to a 40 µM test concentration, this is more likely because of a folding aberration since its HPLC profile displays a heterogeneous population of peptide forms. In contrast, W5A behaves chromatographically as a well folded species and is similarly devoid of activity. Because mutation of the anionic residues Asp-10, Glu-12, and Glu-17 resulted in minor changes in affinity, we interpret this to mean that cationic channel residues do not directly influence toxin binding.

Wang et al. (19) also found basic residues in SGTx that are critical for inhibition of Kv2.1, whereas mutation of acidic residues in fact increased the affinity of that toxin for its target. In the SGTx solution structure, the basic residues form a ring around the hydrophobic patch, presumably creating the active face (31). Because the solution structure for ProTx-II is unknown, we chose to construct a molecular model based upon the known solution structure of a similar ICK toxin that targets Kv4 channels, HpTx-2. Our model clearly shows the existence of a hydrophobic cluster of residues, all of which we found to be essential binding determinants (Fig. 8) Residues we identified as nonessential cluster to the opposite face of the molecule. Similar molecular pharmacophores for toxins that modify either Nav or Kv channels is of particular interest because it could point to a similar mechanism of action for these peptides on their respective targets.

In a previous study we found that ProTx-II has the ability to bind to liposomes, raising the possibility that the toxin might insert into membranes as part of its mode of action (15). Together with our kinetic analysis of mutant toxins showing a critical role of hydrophobic residues in channel modification, these results could suggest a membrane-restricted channel receptor site for toxins that modify activation gating. While the functional role of lipid binding requires further study, it is clearly a property restricted to toxins affecting activation, but not inactivation, kinetics. This result was not particularly unexpected because site 3 is known to be extracellular; thus, membrane partitioning would confer no advantage to toxins affecting inactivation kinetics.

Cohen et al. (17) recently reported that site 4 is not "dipped" in the lipid bilayer and concluded that the lipid binding activity exhibited by site 4 toxins has no functional relevance. Although it is possible that this is true for the beta-toxins with which they conducted their studies, it still remains to be seen if the essential hydrophobic patch on ProTx-II is primarily important for protein-protein interactions or for facilitating entry into the bilayer.

Elucidation of the channel receptor site with which ProTx-II interacts is necessary to further characterize the molecular mechanism through which this toxin works. We previously inferred that ProTx-II was a site 4 toxin based on its ability to modify Nav channel activation and because a well characterized site 4 toxin, CssIV, also exhibits the ability to bind to liposomes. Having already demonstrated that the effects of ProTx-II and CssIV on Nav1.2 differ when analyzed using a simple step protocol (Fig. 5B and Ref. 11), we also examined its effect with a voltage-sensor trapping protocol. Intriguingly, when we use the voltage-sensor trapping protocol developed by Cestele et al. (Ref. 11 and data not shown) to evaluate whether toxin-modified channels activate at more negative potentials than is typically observed, no ProTx-II-induced current is observed. These results confirm the mechanistic differences between ProTx-II and CssIV. To locate residues necessary for ProTx-II binding, we have carried out extensive channel mutagenesis. The results of linker swap experiments established that ProTx-II does not interact with either sites 3 or 4 in the DIV and DII S3/S4 linkers, respectively. These results were of particular interest given that all Nav channel gating modifier peptide toxins characterized to date bind to these extracellular linkers (9–11). Furthermore, linker mutagenesis targeting sites at which Nav1.5 and Nav1.7 differ clearly indicate the unlikelihood of the ProTx-II receptor site existing in these regions (Fig. 7).


Figure 8
View larger version (69K):
[in this window]
[in a new window]

  		FIGURE 8.
The bioactive surface of ProTx-II. A molecular model of ProTx-II was created by in silico mutagenesis from the Protein Data Bank coordinates for HpTx-2, a homologous ICK motif peptide targeting Kv4.2 (36). The model was subjected to energy minimization to remove steric clashes. Essential residues, shown in cyan, are defined by a >10-fold loss of affinity for Nav1.5 upon mutation and are all either hydrophobic or cationic in nature. Nonessential residues are shown in magenta and segregate to the opposite faces of the molecule than essential residues.

 
Because ProTx-II displays a bioactive surface containing hydrophobic and cationic residues and can bind phospholipids, we considered that the toxin molecule might interact first with the lipid bilayer before accessing a target wholly or partially within the membrane. To evaluate potential interactions with cationic toxin residues, we created 16 individual mutations in transmembrane segments, each neutralizing an acidic residue predicted to reside in the outer leaflet of DI, DII, or DIII (Table 2 and supplemental Table 1). DIV was excluded from this study because a great deal of evidence supports its role in inactivation, rather than activation, gating (2, 5). None of the mutations we analyzed caused any reduction in channel modification by ProTx-II. We have also observed that after pretreatment of TTX-sensitive C373Y Nav1.5 channels with sub-saturating ProTx-II, the addition of 20 nM tetrodotoxin completely blocks current, strongly suggesting that ProTx-II does not bind in the pore vestibule (data not shown and Ref. 32). It remains possible that basic toxin residues interact with charged phospholipid head groups or that hydrophobic toxin residues make contact with hydrophobic channel residues, and a very recent molecular dynamics analysis suggests that the former scenario applies in the case of SGTx (33). However, a full analysis of potential interactions of hydrophobic toxin residues with hydrophobic sites in the transmembrane regions of the channel is beyond the scope of this study.

Although ProTx-II has no affinity for Kv channels, it behaves in a manner similar to Kv channel ICK toxins and also displays a similar active face. Glu-795, Leu-796, and Leu-798 in the C-terminal end of Nav1.5 DII S3 correspond to residues in the S3b region of Kv2.1 that render that channel sensitive to hanatoxin (17). Nonetheless, mutagenesis at these sites in Nav1.5 fails to disrupt channel modification, suggesting that ProTx-II does not interact with the previously identified gating modifier "hot spot" on Nav, Cav, and Kv channels (34). It has been shown kinetically that Kv channel ICK toxins have multiple receptor sites arising from their tetrameric channel structure (35). However, there is no precedent for Nav channel gating modifier toxins binding to multiple sites with high affinity, presumably because Nav channel domains do not have identical amino acid sequences. Although mutations in several extracellular linkers have been shown to decrease site 4 toxin (CssIV) binding affinity, most of the binding energy is associated with residues in DII S3/S4, a finding we were able to confirm (11). Moreover, a single mutation in DIV S3/S4 renders Nav1.5 and 1.2 almost completely insensitive to the site 3 toxins, ApB and LqqV (9, 10). These results give credence to the notion that ProTx-II interacts with only a single channel site.

In this study we have mapped the results from our toxin scan onto a structural model of ProTx-II to establish its bioactive surface, thus providing insight into its mechanism of action. In addition, we demonstrated that ProTx-II does not interact with previously characterized gating modifier peptide toxin sites. Likewise, it is highly unlikely that ProTx-II makes any direct binding interactions with other extracellular linker regions on Nav1.5. These results establish the novelty of this toxin as a gating modifier whose mechanism of inhibition differs dramatically from those of previously characterized Nav channel toxins. This uniqueness demonstrates its potential as an extremely useful probe of gating mechanisms in Nav as well as Cav channels. Given the intriguing, if elusive mechanism of action associated with this toxin, future studies aimed at identifying the channel receptor site are imperative and are currently under way.


   FOOTNOTES
 
* The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. Back

Formula The on-line version of this article (available at http://www.jbc.org) contains supplemental Table 1. Back

1 To whom all correspondence and reprint requests should be addressed: Dept. of Biochemistry, School of Medicine and Biomedical Sciences, State University of New York, 3435 Main St., Buffalo, NY 14214. Tel.: 716-829-2727; Fax: 716-829-2725; E-mail: kblumen{at}buffalo.edu.

2 The abbreviations used are: ICK, inhibitory cystine knot; HPLC, high performance liquid chromatography; RP, reverse phase; MALDI-TOF, matrix-assisted laser desorption ionization time-of-flight; HEK cells, human embryonic kidney cells; SGTx, toxin 1 from the spider Scodra griseipes. Back


   ACKNOWLEDGMENTS
 
We are very grateful to Dr. James Garrity for creating several mutant toxin contructs, Drs. Gerardo Corzo and Lourival Possani for the generous gift of CssIV, Drs. Michael Morales and Philip Gottlieb for gifts of HpTx and GsMtx4, respectively, Dr. Dorothy Hanck for cells expressing the C373Y mutant of Nav1.5, and Drs. Morales and Hanck for valuable comments on the manuscript.



   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 EXPERIMENTAL PROCEDURES
 RESULTS
 DISCUSSION
 REFERENCES
 

  1. Catterall, W. A., Goldin, A. L., and Waxman, S. G. (2005) Pharmacol. Rev. 57, 397-409[Abstract/Free Full Text]
  2. Cha, A., Ruben, P. C., George, A. L., Fujimoto, E., and Bezanilla, F. (1999) Neuron 22, 73-87[CrossRef][Medline] [Order article via Infotrieve]
  3. Chanda, B., and Bezanilla, F. (2002) J. Gen. Physiol. 120, 629-645[Abstract/Free Full Text]
  4. Chahine, M., George, A. L., Zhou, M., Ji, S., Sun, W., Barchi, R. L., and Horn, R. (1994) Neuron 12, 281-294[CrossRef][Medline] [Order article via Infotrieve]
  5. Sheets, M. F., Kyle, J. W., Kallen, R. G., and Hanck, D. A. (1999) Biophys. J. 77, 747-757[Medline] [Order article via Infotrieve]
  6. Chanda, B., Asamoah, O. K., and Bezanilla, F. (2004) J. Gen. Physiol. 123, 217-230[Abstract/Free Full Text]
  7. Zagotta, W. N., Hoshi, T., and Aldrich, R. W. (1994) J. Gen. Physiol. 103, 321-362[Abstract/Free Full Text]
  8. Schoppa, N. E., and Sigworth, F. J. (1998) J. Gen. Physiol. 111, 313-342[Abstract/Free Full Text]
  9. Rogers, J. C., Qu, Y., Tanada, T. N., Scheuer, T., and Catterall, W .A. (1996) J. Biol. Chem. 271, 15950-15962[Abstract/Free Full Text]
  10. Benzinger, G. R., Kyle, J. W., Blumenthal, K. M., and Hanck, D. A. (1998) J. Biol. Chem. 273, 80-84[Abstract/Free Full Text]
  11. Cestele, S., Qu, Y., Rogers, J. C., Rochat, H., Scheuer, T., and Catterall, W. A. (1998) Neuron 21, 919-931[CrossRef][Medline] [Order article via Infotrieve]
  12. Cestele, S., Scheuer, T., Mantegazza, M., Rochat, H., and Catterall, W. A. (2001) J. Gen. Physiol. 118, 291-301[Abstract/Free Full Text]
  13. Cestele, S., Yarov-Yarovoy, V., Qu, Y., Sampieri, F., Scheuer, T., and Catterall, W. A. (2006) J. Biol. Chem. 281, 21332-21344[Abstract/Free Full Text]
  14. Middleton, R. E., Warren, V. A., Kraus, R. L., Hwang, J. C., Liu, C. J., Dai, G., Brochu, R. M., Kohler, M. G., Gao, Y. D., Garsky, V. M., Bogusky, M. J., Mehl, J. T., Cohen, C. J., and Smith, M. M. (2002) Biochemistry 41, 14734-14747[CrossRef][Medline] [Order article via Infotrieve]
  15. Smith, J. J., Alphy, S., Seibert, A. L., and Blumenthal, K. M. (2005) J. Biol. Chem. 280, 11127-11133[Abstract/Free Full Text]
  16. Bosmans, F., Rash, L., Zhu, S., Diochot, S., Lazdunski, M., Escoubas, P., and Tytgat, J. (2006) Mol. Pharmacol. 69, 419-429[Abstract/Free Full Text]
  17. Cohen, L., Gilles, N., Karbat, I., Ilan, N., Gordon, D., and Gurevitz, M. (2006) J. Biol. Chem. 281, 20673-20679[Abstract/Free Full Text]
  18. Swartz, K. J., and MacKinnon, R. (1997) Neuron 18, 675-682[CrossRef][Medline] [Order article via Infotrieve]
  19. Wang, J. M., Roh, S. H., Kim, S., Lee, C. W., Kim, J. I., and Swartz, K. J. (2004) J. Gen. Physiol. 123, 455-467[Abstract/Free Full Text]
  20. Horton, R. M., Hunt, H. D., Ho, S. N., Pullen, J. K., and Pease, L. R. (1989) Gene (Amst.) 77, 61-68[CrossRef][Medline] [Order article via Infotrieve]
  21. Kraus, R. L., Warren, V. A., Smith, M. M., Middleton, R. E., Blumenthal, K. M., and Cohen, C. J. (2002) Biophysical Society 46th Annual Meeting, February 23–27, San Francisco, Abstract 85c
  22. Seibert, A., Liu, J. R., Hanck, D. A., and Blumenthal, K. M. (2003) Biochemistry 42, 14515-14521[CrossRef][Medline] [Order article via Infotrieve]
  23. Cummins, T. R., Howe, J. R., and Waxman, S. G. (1998) J. Neurosci. 18, 9607-9619[Abstract/Free Full Text]
  24. Hirsch, J. K., and Quandt, F. N. (1996) Brain Res. 706, 343-346[CrossRef][Medline] [Order article via Infotrieve]
  25. Brahmajothi, M. V., Campbell, D. L., Rasmusson, R. L., Morales, M. J., Trimmer, J. S., Nerbonne, J. M., and Strauss, H. L. (1999) J. Gen. Physiol. 113, 581-600[Abstract/Free Full Text]
  26. Bernard, C., Legros, C., Ferrat, G., Bischoff, U., Marquardt, A., Pongs, O., Francisco, and Darbon, H. (2000) Protein Sci. 9, 2059-2067[Abstract]
  27. Leipold, E., Hansel, A., Borges, A., and Heinemann, S. H. (2006) Mol. Pharmacol. 70, 340-347[Abstract/Free Full Text]
  28. Lee, S. Y., and MacKinnon, R. (2004) Nature 430, 232-235[CrossRef][Medline] [Order article via Infotrieve]
  29. Jung, H. J., Lee, J. Y., Kim, S. H., Eu, Y., Shin, S. Y., Milescu, M., Swartz, K. J., and Kim, J. I. (2005) Biochemistry 44, 6015-6023[CrossRef][Medline] [Order article via Infotrieve]
  30. Bemporad, D., Wee, C. L., Sands, Z., Grottesi, A., and Sansom, M. S. P. (2006) Biochemistry 45, 11844-11855[CrossRef][Medline] [Order article via Infotrieve]
  31. Lee, C. W., Kim, S., Roh, S. H., Endoh, H., Kodera, Y., Maeda, T., Kohno, T., Wang, J. M., Swartz, K. J., and Kim, J. I. (2004) Biochemistry 43, 890-897[Medline] [Order article via Infotrieve]
  32. Satin, J., Kyle, J. W., Chen, M., Bell, P., Cribbs, L. L., Fozzard, H. A., and Rogart, R. B. (1992) Science 256, 1202-1205[Abstract/Free Full Text]
  33. Wee, C. L., Bemporad, D., Sands, Z., Gavaghan, D., and Sansom, M. S. P. (2006) Biophys. J. 92, 7-9[CrossRef]
  34. Winterfield, J. R., and Swartz, K. J. (2000) J. Gen. Physiol. 116, 637-644[Abstract/Free Full Text]
  35. Swartz, K. J., and MacKinnon, R. (1997) Neuron 18, 665