Both Basic and Acidic Amino Acid Residues of IpTxa Are Involved in Triggering Substate of RyR1

Imperatoxin A (IpTxa) is known to modify the gating of skeletal ryanodine receptor (RyR1). In this paper, the ability of charged aa residues of IpTxa to induce substate of native RyR1 in HSR was examined. Our results show that the basic residues (e.g., Lys19, Lys20, Lys22, Arg23, and Arg24) are important for producing substate of RyR1. In addition, other basic residues (e.g., Lys30, Arg31, and Arg33) near the C-terminus and some acidic residues (e.g., Glu29, Asp13, and Asp2) are also involved in the generation of substate. Residues such as Lys8 and Thr26 may be involved in the self-regulation of substate of RyR1, since alanine substitution of the aa residues led to a drastic conversion to the substate. The modifications of the channel gating by the wild-type and mutant toxins were similar in purified RyR1. Taken together, the specific charge distributions on the surface of IpTxa are essential for regulation of the channel gating of RyR1.

Imperatoxin A (IpTx a ) from the African scorpion Pandinus imperator is a high-affinity modulator of skeletal RyR (RyR1).It greatly increases open probability (P o ) and [ 3 H]ryanodine binding to RyR1 at nanomolar concentration [11,12].Moreover, binding of IpTx a to RyRs reconstituted in planar lipid bilayers generates marked occurrence of long-lasting openings in subconductance state (substate) [13].Confocal imaging of skeletal muscle fibers to monitor IpTx a -induced Ca 2+ sparks demonstrate that the toxin induces long-duration and low-amplitude local Ca 2+ release consistent with the observation of the prolonged substate in the presence of IpTx a [14].Another structurally related scorpion toxin, maurocalcine (MCa), and one specific small fragment of II-III loop region of skeletal DHPR (Peptide A) also bind to RyR1 and modify channel activity [15][16][17][18].MCa also strongly enhances [ 3 H]ryanodine binding to RyR1 and induces long-lasting substate with the current amplitude of 48% of the full conductance [16,18].Peptide A could bind to RyR1 and could either activate or inhibit the activity of RyR1.It could also induce the long-lasing substate [15,17,19].
A comparison of the amino acid (aa) sequences of these RyR1-modifing probes shows a common basic aa domain and the C-terminal hydroxyl-containing side chain.The aa sequences may also contribute to the essential structure for activating RyR1 [15,16].Especially, a cluster of positively charged aa residues on the surface of the peptide A is critical for activation of RyR1 [19,20].The mutations of the specific basic residues of MCa and IpTx a have failed to induce longlasting substate and to potentiate [ 3 H]ryanodine binding [20,21].To date, the ability of a single aa residue of IpTx a to control the substate of RyR1 is not fully understood.Recently we have found that several basic aa residues of IpTx a (e.g., Lys 19 , Arg 23 , and Arg 33 ) are necessary for increasing open probability and inducing substate in rabbit skeletal RyR1 [22].
In the present study, to evaluate the roles of the charged aa residues of IpTx a in modifying the RyR1 gating, synthetic wild-type and alanine-scanning mutants of IpTx a were tested on planar lipid bilayer-incorporated RyR1.The basic aa mutants (e.g., K19A, K20A, K22A, R23A, and R24A) resulted in a significant loss of production of substate in RyR1, consistent with the previous suggestion that the critical basic domain of toxin determines its binding to the channel [15,16,22].The effective domain encompassing these basic residues involved in producing substate is structurally conserved with both MCa and Peptide A [19,23].This suggests a common role of the highly clustered positive charges for their action on RyR1 channel gating.The mutations of some acidic residues (e.g., Asp 2 , Asp 13 , and Glu 29 ) and basic residues within C-terminal region of IpTx a (e.g., Lys 30 , Arg 31 , and Arg 33 ) also led to a significant inhibition on the gating.When Lys 8 and Thr 26 were replaced by alanine, the substate was predominant indicating that these two residues are essential for the functions of the toxin.In addition, the effects of the wild-type and mutant toxins on the gating behavior of RyR1 are strikingly similar when the native RyR1 in SR and the purified RyR1 are used for the incorporation into bilayers, suggesting that generation of the substate is due to a direct binding of the toxin to RyR1.

Materials.
Porcine brain phosphatidylethanolamine and phosphatidylserine were purchased from Avanti Polar Lipids, Inc.All other reagents were from Sigma.

Chemical Synthesis of Wild-Type and Mutant IpTx a
Peptides.The peptide synthesis was conducted by a peptide synthesizer (Applied Biosystems model 433A).The linear precursors of wild-type and mutant IpTx a were synthesized by solid-phase Fmoc chemistry starting from Fmoc-Arg (2,2,5,7,8-pentamethylchroman-6-sulphonyl)-Alko or Fmoc-Ala-Alko resin using a variety of blocking groups for amino acid protection.After cleavage by trifluoroacetic acid, crude linear peptides were extracted with 2 M ethanoic acid, diluted to final peptide concentration of 25 μM in a solution of 1 M ammonium acetate and 2.5 mM reduced/0.25 mM oxidized glutathione adjusted to pH 7.8 with aqueous NH 4 OH, and stirred slowly at 4 • C for 2-3 days.In the redox buffer system, oxidized glutathione acts as oxidase and assists in the formation of disulfide bonds whereas reduced glutathione functions as disulfide isomerase and facilitates formation of correct disulfide bonds by promoting rapid reshuffling of incorrect disulfide parings.A 10 : 1 mixture of reduced and oxidized glutathione was suggested to be an efficient redox buffer system for producing disulfide bonds in IpTx a [21].The folding reactions were monitored by HPLC.The crude oxidized products were purified by successive chromatography with CM-cellulose CM-52 and preparative HPLC with C18 silica columns.The purity of all analogues were in the range of 60 to 95% as measured by analytical HPLC and MALDI-TOF-MS (matrix-assisted laser desorption ionization-time-of-flight MS) measurements (see Supplementary Figure 1 (in Supplementary material available online at doi: 10.1155/2011/386384)).

Preparation of Junctional SR Vesicles from Rabbit Skeletal
Muscle.A heavy fraction of fragmented SR vesicles (HSR) containing junctional SR was prepared from rabbit fasttwitch back and leg muscles as described previously [24].
2.4.Planar Lipid Bilayers.Single-channel recordings of rabbit skeletal RyR1 incorporated into planar lipid bilayers were carried out as described previously [25][26][27].Lipid bilayers, consisting of brain tissue phosphatidylethanolamine and phosphatidylserine (1 : 1) in decane (20 mg/mL) were formed across a hole of approximately 200 μm diameter.Thinning of the bilayer was monitored by bilayer capacitance.The basic composition of the cis/trans solution consisted of 300 mM cesium methanesulfonate, 10 mM Tris/Hepes (pH 7.2), 2 mM EGTA, and 1.998 mM CaCl 2 ([Ca 2+ ] free = 10 μM) [27].[Ca 2+ ] free was calculated using the "Chelator" program (Theo Schoenmaker).Cs + was selected as the charge carrier to ensure precise control of free [Ca 2+ ], to increase the channel conductance, and to avoid any contribution from potassium channels present in the SR membrane [12].Chloride channels were inhibited by using the impermeant anion methanesulfonate [12].Incorporation of ion channels was carried out as described by Miller and Racker [25] and confirmed by recording the characteristically high single-channel conductance of RyRs [27,28].The trans side was maintained at ground and the cis side was clamped at −30 mV relative to the ground.After addition of the IpTx a to the cis chamber, the single channel data were collected at −30 mV for 2-5 min.The channel activity was recorded on a DTR-1204 Digital Recorder (Biologic Science Instrument) and displayed on a Tektronix TDS 340A oscilloscope.Recordings were filtered with an 8-pole low-pass Bessel filter at 1 kHz and digitalized through a Digidata 1200 series interface (Axon Instruments).Data acquisition and analysis were done with the Axon Instruments software, pClamp v7.0.
Data were analyzed using the Hill equation described previously [12,13,15]:  as the time spent in the substate divided by a given total recording time.The durations of the substate were obtained by manual positioning of the cursors and constructing all point histograms.Mean duration of substate was measured by total substate time divided by total substate frequencies.

Statistical Analysis.
Results are given as means ± SE.Significant differences were analyzed using Student's t-test.
Differences were considered to be significant when P < 0.05.The fitting of the data to the graphs were carried out using the software, Origin v7.

Effects of IpTx a on Single-Channel Gating Properties of
RyR1.To examine how IpTx a modifies RyR1 activity, RyR1 in HSR incorporated in planar lipid bilayers was tested in the presence or absence of synthetic wild-type IpTx a .
The chamber solutions for both cis (cytosolic) and trans (luminal) sides included 300 mM cesium methanesulfonate and 10 mM Tris-Hepes (pH 7.2). 10 μM free Ca 2+ was added to the cis side to activate the incorporated RyR1.After an addition of IpTx a to the cis chamber, the single channel gating properties were recorded at a holding potential of −30 mV for over 2 min at each toxin concentration.Figure 1 suggesting that increase of the full opening events of the RyR1 channel is due to interaction between the toxin and the cytosolic region of the channel.
In light of the evidence that IpTx a could induce substate both in cardiac and skeletal RyRs [13], the effects of synthetic IpTx a on the occurrence of substate were tested using native RyR1 in rabbit skeletal HSR. Figure 1(a) shows that an addition of IpTx a to the cis side of RyR1 channel could induce the substate.The probability to obtain substate (P substate ) increased, when the concentration of IpTx a increased from 6 to 100 nM in the cis chamber (Figure 1(c)).The calculated P substate, max and EC 50 for P substate were 0.92 ± 0.06 and 23.27 ± 2.37 nM, respectively.The Hill coefficient (nH) for P substate was 1.24, suggesting that IpTx a and RyR1 do not have the cooperative bindings in the concentration range.

Effects of Alanine Scanning Mutants of IpTx a on RyR1.
In the previous studies [21,22], it was proposed that specific basic amino acid residues on the surface of IpTx a are required for the electrostatic interaction of the toxin with RyR1.Particularly, Gurrola et al. [15] suggested the structural domain composed of Lys 19 -Arg 24 followed by Thr 26 is responsible for the IpTx a -RyR1 binding.A possible molecular interaction between IpTx a and RyR1 was further investigated in the present study by producing various alanine scanning mutants at charged aa.We first tested the effects of IpTx a mutants on substate of RyR1 at a holding potential of -30 mV.
The single channel traces of RyR1 activated at 30 nM wild type or mutant IpTx a are shown in Figure 2(a).Ability of the different mutant toxins to induce substate of RyR1was further tested using different toxin concentrations.Interestingly, both K8A and T26A mutants displayed a significantly increased substate lifetime, indicating a negative role of these residues in modifying RyR1 gating.P substae, max of K8A or T26A reached almost to 1, and EC 50 (nM) values were shifted from 23.27 ± 0.37 (wild type) to 5.70 ± 2.86 (K8A) or to 9.98 ± 1.92 (T26A; Figure 2(b) and Table 1).The mutations of the basic aa (Lys 19 , Lys 20 , Lys 22 , Arg 23 , Arg 24 , Lys 30 , Arg 31 , and Arg 33 ) significantly reduced the probability to obtain substate (P substate ), compared with the wild-type toxin (Figure 2(b) and Table 1).The effect of D15A mutant was similar to that of wild-type toxin, suggesting no major involvement of Asp 15 in the binding between the toxin and RyR1.On the other hand, the substitutions of some acidic aa  13 , and Glu 29 by alanine alter the probability to obtain substate significantly (Figure 2 and Table 1).These results suggest that the charged aa distributed on the surface of IpTx a contribute to the stimulatory action of the toxin and to the interaction between IpTx a and RyR1.

Mean Duration of IpTx a -Induced Substate.
To investigate the causes for the decreased P substate by the mutant toxins, the average length of substate at 30 nM wild type or mutant of IpTx a was calculated as total substate time divided by the toral frequencies of substate.The recording time in each concentration was 2 min.Despite the marked increase in P substate as IpTx a was increased from 6 nM to 100 nM, the mean duration of the IpTx a induced-substate of RyR1 appeared to be similar at different [IpTx a ] (Figure 4).The average mean duration of substate of the mutants (D2A, D13A, K19A, K22A, R23A, R24A, E29A, K30A, and R33A) was significantly less than that of wild-type IpTx a (Table 1).Some mutants (K19A, R24A, E29A, and K30A) showed more marked reduction in mean duration of substate (<2 s).The decreased mean duration of substate in the mutant toxins is due probably to the loss of their ability to induce long-lasting substate.to purified RyR1 incorporated into planar lipid bilayers (Figure 3(a)).30 nM IpTx a increased P substate both in native and purified RyR1 in a similar extent (Figures 1 and 3). Figure 4 shows that the mean durations of substate of purified RyR1 are similar to native RyR1 at the tested concentration range, suggesting that the response of RyR1 to IpTx a was not mediated by other RyR1-associated proteins.

Effects of IpTx a on Purified
Purity of RyR1 at the final purification step was verified by Coomassie blue staining (Supplementary Figure 2).

Effects of IpTx a Mutants on Substate of RyR1.
The highly positive charges of the basic residues of IpTx a could contribute to the formation of its functional surface area having uniquely oriented charge distribution [15,19,[21][22][23].In the present study we tested the hypothesis that electrostatic force mediates the IpTx a -RyR1 interaction by studying the effects of alanine scanning mutations of charged aa residues in IpTx a on RyR1 functions.Single point mutations of charged residues in IpTx a generally affected the probability of occurring substate (P substate ) in RyR1 (Figure 2 and Table 1).Previously, it was shown that mutations in a cluster of basic residues (Lys 19 -Arg 24 ) decreased the ability of the toxin to activate [ 3 H]ryanodine binding to RyR1 [15,21].The recombinant mutant toxins (e.g., K19A, R23A, and R24A) were less effective to increase open probability (P o ) and to induce substate of the channel [22].Our present results demonstrate that a cluster of the basic residues 19-24 is necessary for inducing substate of RyR1, confirming the functional importance of the clustered basic residues.In addition, replacing other basic residues located in C-terminal region of IpTx a with alanine (e.g., K30A, R31A, and R33A) also reduced the effects of the toxin on channel modification (Figure 2(b)).As described previously, the C-terminal basic residues (Lys 22 , Arg 23 , and Arg 24 ) are aligned in the central domain of IpTx a and possibly are responsible for activating RyR1 [21].Our present findings agree with the previous suggestion that the positively charged region within the C-terminus is involved in the interaction with RyR1.K8A and T26A, the two mutated analogs of IpTx a , showed dramatically decreased EC 50 of P substate compared to that of wild-type toxin, indicating higher binding affinity to RyR1 (Figures 2(b) and 5(b)).Although the ability of K8A to increase the substate lifetime of RyR1 was previously studied using recombinant IpTx a mutant [22], the effect of K8A on [ 3 H]ryanodine binding to RyR1 is controversial [15,21].T26A mutant was reported to reduce toxin-activated ryanodine binding to RyR1 [15].The inconsistency between occurrence of substate and ryanodine binding to RyR1 affected by K8A or T26A suggests multiple independent actions of IpTx a on different modes of channel gating.In fact, Dulhunty et al. [17] proposed an existence of multiple toxin binding sites within RyR1 including the transient activation site and substate site [17].

The Effects of IpTx a and Peptide A on RyR1 Gating.
Marked functional similarity of the three peptides, IpTx a , MCa, and Peptide A has been proposed on the basis of their primary structural homology of a specific domain consisting of basic amino acids (Lys 19 -Arg 23 of IpTx a or MCa, and Arg 681 -Lys 685 of Peptide A) [15,17,18,23,29].Stretches of these positively charged residues tend to adopt different secondary structures such as α-helical structure for Peptide A and β-sheet structure for IpTx a and MCa.However, their orientation on the surface of the peptides could be similar [19,21].Peptide A was shown to share the common binding site on RyR1 with IpTx a and MCa and mimicked the toxin effects on RyR1 gating [15,23,29].However, evidence for noncompetitive binding of MCa and peptide A to RyR1 was shown by [ 3 H]ryanodine binding and realtime Surface Plasmon resonance (SPR) studies.MCa and peptide A induced distinct modification of channel gating in an additive, but not competitive, manner.This indicates the possibility of an existence of independent binding sites for the two peptides on RyR1 [16].These different results have been further understood by the hypothesis that the toxins and peptide A binding sites within RyR have both common activation site and independent substate sites [17].Although the present results show that the mutations within the structural motif shared by Peptide A inhibited substate induction (Figure 2), it is hard to find direct evidence of competitive function in the common region of RyR1.Even though our investigations were undertaken without Peptide A, it could be suggested that the prolonged substate opening triggered by IpTx a is independent of the action of Peptide A [17].Further study will be necessary to clarify whether the structurally conserved domains of IpTx a and Peptide A compete for the induction of substate in RyR1.

Comparison of the Active Sites with MCA. IpTx a and
MCa share 82% aa identity in their primary structures.In addition to the similar β-sheet structure of the common stretch of the basic residues (Lys 19 -Arg 24 ), the solution structures of IpTx a and MCa exhibit similar overall molecular folding [21,30].Because of this structural homology, these two toxins share functional similarities.Both IpTx a and MCa strongly induce SR Ca 2+ release and activate ryanodine binding to skeletal RyR.In addition, both peptides have the ability to induce reversible transition of RyR1 gating mode between substate and fast full open gating [12,13,[15][16][17].Previously, it was demonstrated that mutations of each basic residue within Lys 19 -Thr 26 and mutation of Lys 8 of MCa decreased the ability of the toxin to induce Ca 2+ release and potentiate [ 3 H]ryanodine binding in the SR [29].Moreover, the occurrence of long-lasting substate was markedly prevented by mutations of the basic residues within Lys 20 -Arg 24 of MCa [29].This inhibitory effect was reduced, if the mutation was farther from Lys 24 while alanine replacement completely inhibited the substate event of skeletal RyR [17,29].Therefore, the previous results showing the effects of MCa mutants are partially coherent with our observations of the effects of IpTx a mutants on RyR1 substate.In the present study, IpTx a mutant, R24A, was the most effective in decreasing mean duration of substate of RyR1 (Figure 3 and Table 1).R24A mutant showed comparable B max value of P substate with those of other mutants, R23A, K22A, and K20A, although the values were significantly less than that of wild-type IpTx a (Figure 3 and Table 1).This suggests that the common domain clustered by positively charged residues (Lys 20 -Arg 24 ) are responsible for the actions of two scorpion toxins to induce long-lasting substate opening of skeletal RyR1.
In spite of the high sequence identity, a significant functional difference between IpTx a and MCa has been observed.Two toxins induced different degree of substate of RyR1 at +40 mV holding potential (28% and 48% of full conductance state for IpTx a and MCa, resp.)[10,13].In addition, comparison of 3D structures of two peptides showed significantly different structural motifs near the Nterminal regions, where MCa but not IpTx a has a β-strand and makes the hydrophobic core by connecting to the side chains of four cysteine residues, Cys 10 , Cys 16 , Cys 21 , and Cys 32 [21].Thus, the difference in the functional effect between IpTx a and MCa on RyR1 gating appears to be due to the subtle change in the local charge distribution or a structural dissimilarity.Here we further suggest and it appears that both acidic (e.g., Asp 2 , Asp 13 , and Glu 29 ) and basic residues of C-terminal region of IpTx a (e.g., Lys 30 , Arg 31 , and Arg 33 ) are involved in functional interaction with RyR1 in the case of IpTx a (Figure 2(b)).To our knowledge, to date there is no report regarding to the involvement of the acidic aa residues of IpTx a are related to the occurrence of P substate of RyR1.

Conclusions
In this study, the ability of charged aa residues of IpTx a to induce substate of RyR1 was examined in detail.Both basic and acidic aa residues are involved in producing substate of RyR1 supporting the hypothesis that the structural domain constituting a local cluster of charged aa is important for modifying the mode of channel gating [15,16,21].Residues such as Lys 8 and Thr 26 of IpTX a are important in terms of their inhibitory role in producing substate of RyR1.The modified channel gating properties induced by wild-type and mutant toxins were found both in native and purified RyR1.Taken together, the specific charge distributions on the surface of IpTx a may directly regulate the gating behavior of RyR1.

Figure 1 :
Figure 1: Properties of single channel gating in IpTx a -modified native RyR1.(a) Channel current traces of a single skeletal RyR1 in planar lipid bilayers activated by various concentrations of IpTx a .IpTx a at 6-100 nM (final) was added to the cis solution to activate the channel.Channel openings are shown as downward deflections.Channel activities were recorded at a holding potential of −30 mV.(b) A plot of P o versus concentration of IpTx a to activate rabbit skeletal RyR1.Data points are means ± SE of ten experiments.(c) A plot of P substate versus IpTx a concentration for rabbit skeletal RyR1.Data points are means ± SE of 9 experiments.

Figure 3 :
Figure 3: Effects of IpTx a on purified rabbit RyR1 incorporated in planar lipid bilayers.(a) Single channel activity of purified RyR1 incorporated in planar lipid bilayers at 10 μM Ca 2+ with or without IpTx a (6-100 nM) was recorded.Single channel currents are shown as downward deflections from the closed level (indicated by c).The holding potential was −30 mV.(b) The plot of P o of purified RyR1 versus various concentrations of IpTx a is shown.(c) The plot of P substate of purified RyR1 versus various concentrations of IpTx a is shown.The results are shown as the means ± SE for 7 experiments.

Figure 4 :
Figure 4: Effects of IpTx a on the mean durations of substate of native and purified RyR1.Mean durations of the substate events in native RyR1 versus purified RyR1 were determined in the presence of 6-100 nM IpTx a .Data points are means ± SE for 10 experiments.

Figure 5 :
Figure 5: Properties of single channel gating in wild-type and mutant IpTx a -modified purified RyR1.(a) Single channel currents of purified RyR1 activated in the presence of wild-type or mutant IpTx a were measured at −30 mV holding potential.The single channel opening is shown as a downward deflection.The recordings of single channel currents were measured at 30 nM wild-type or mutant IpTx a in the cis solution.(b) The plots of P substate of purified RyR1 versus wild-type or mutant IpTx a at various concentrations.
where (P o, max ) is the P o observed at saturating concentrations of IpTx a , EC 50 is the IpTx a concentration for which 50% of P o,max is obtained and nH is the Hill coefficient.
Also, P substate, max is the P substate observed at saturating concentrations of IpTx a , EC 50 is the IpTx a concentration for which 50% of P substate, max is obtained, and nH is the Hill coefficient.The probability of full open state of the channel (P o ) was defined as the ratio of the time spent in the open state to the total time exclusive of time spent in the substate.The probability to obtain substate (P substate ) was calculated time spent in the substate.The time for recorded P o was in the range of 30 s-2 min.Figure1(b) shows a plot of P o versus IpTx a concentration at 10 μM Ca 2+ .The steady-state P o was 0.18 ± 0.02 at 10 μM Ca 2+ .When the concentration of IpTx a was increased in cis chamber, P o increased markedly to 0.7, suggesting that the cytosolic IpTx a enhanced the channel activity by increasing open probability in a dose-dependent manner.Using the Hill equation (1) the parameters such as P o, max , EC 50 , and Hill coefficient (nH) were calculated.The calculated P o, max and EC 50 for P o were 0.72 ± 0.02 and 17.35 ± 4.67 nM, respectively.The Hill coefficient (nH) for P o was 1.14.Application of IpTx a to the trans (SR lumen) side of chamber did not show any effect (data not shown), (a) shows traces from continuous recordings in the absence and in the presence of 6, 15, 30, 50, and 100 nM IpTx a .When the toxin concentration increased, the occurrence of substate of RyR1 was remarkably increased.To examine the effects of IpTx a on full open state, we calculated the probability to obtain the full open state (P o ) as the time spent in the open state divided by the total time exclusive of the

Table 1 :
Effects of the mutant IpTx a on P substate of native RyR1 in SR.P substate, max and EC 50 for P substate, max were calculated using (1) as described in "Section 2." Values are means ± SE of 3-9 experiments.The average lengths of the substate events were determined at 30 nM concentration.Asterisks indicate significant differences from the wild-type toxin for each parameter (Student t-test, P < 0.05).ND: not determined.

Table 2 :
Effects of the mutant IpTx a on P substate of purified RyR1.P substate, max and EC 50 for P substate of purified RyR1 were calculated as described in the legend to Table1.Values are means ± SE of 4-10 experiments.The average lengths of the substate events were determined at 30 nM concentration.Asterisks indicate significant differences from the wild type for each parameter (Student t-test, P < 0.05).ND: not determined.