U-shaped dose-dependent effects of BmK AS, a unique scorpion polypeptide toxin, on voltage-gated sodium channels

Abstract

Background and purpose:

Buthus martensi Karsch (BmK) AS is a scorpion polypeptide toxin, said to target the voltage-gated sodium channels (VGSCs). However, the mechanism of action of BmK AS on the VGSCs has yet to be defined.

Experimental approach:

We examined the electrophysiological effects of BmK AS in a wide dose range on the rat brain-type VGSC α-subunit, rNav1.2a, heterologously expressed in Xenopus oocytes and on the VGSCs endogenously expressed in the dorsal root ganglion neuroblastoma ND7-23 cell line.

Key results:

In the oocytes, BmK AS depolarized the voltage dependence of activation and inactivation of rNav1.2a at 0.1 and 500 nM whereas these parameters were hyperpolarized at 1 nM. In ND7-23 cells, BmK AS hyperpolarized the voltage dependence of activation and inactivation at 0.1, 1 and 100 nM but not 10 nM. BmK AS also hyperpolarized the voltage dependence of recovery from inactivation at 0.1 and 100 nM and slowed the recovery kinetics at all concentrations, but the effects of 1 and 10 nM were relatively smaller than those at 0.1 and 100 nM. Moreover, the inactivation of VGSCs was potentiated by 10 nM BmK AS in both systems, whereas it was inhibited by 0.1 or 100 nM BmK AS in the oocytes or ND7-23 cells respectively.

Conclusions and implications:

BmK AS modulated the VGSCs in a unique U-shaped dose-dependent manner, which could be due to the opposing effects of binding to two distinct receptor sites on the VGSCs.

Introduction

The venom of the Asian scorpion Buthus martensi Karsch (BmK) is a rich source of biologically active peptides, with selective actions on a variety of ion channels. Over the past decade, 51 long-chain peptides have been isolated with action on voltage-gated sodium channels (VGSC). Among these, 34 are related to the α-toxin family, 4 related to the excitatory insect toxin family, 10 related to the depressant insect toxin family, 1 β-like toxin plus 2 unclassified polypeptides, BmK AS and AS-1 (Goudet et al., 2002). BmK AS and AS-1 consist of 66 amino acids cross-linked by four disulphide bridges and their sequences exhibit 86.3% identity (Ji et al., 1999). To date, no other natural toxin has been found to share sequence homology with them, except AaH IT4, an anti-insect toxin isolated from Androctonus australis Hector (AaH), active against the VGSCs (Loret et al., 1991). An earlier study showed that BmK AS could significantly stimulate the binding of [³H]-ryanodine to partially purified ryanodine receptors from rabbit skeletal muscle with an EC₅₀ of 8 ± 1.6 µM, through an indirect mechanism (Kuniyasu et al., 1999). Then BmK AS was found to promote noradrenaline release from rat hippocampus slices (0.05–1 mM) by augmentation of sodium influx (Ji et al., 1997). Thus BmK AS seems likely to enhance cell excitability through modulation of the VGSCs. However, later experiments on rat behaviour demonstrated that BmK AS could produce an anti-nociceptive effect on the inflammation-induced spontaneous pain as well as mechanical and thermal allodynia by intrathecal injection of 0.02–1.0 µg (Chen and Ji, 2002;Chen et al., 2006; Liu et al., 2008). Most recently, it was observed that hippocampal injection of 0.5–1 µg BmK AS repressed pentylenetetrazole-induced seizures in rat epilepsy models (R. Zhao and C.C. Weng, unpublished data). Electrophysiological recordings from neuroblastoma B104 cells showed that 50–500 nM of BmK AS could suppress VGSC currents by 15–25% (Tan et al., 2003). The reduction of VGSC currents was also observed without any effects on the voltage-dependent potassium and calcium currents in the acutely dissociated small dorsal root ganglion (DRG) neurons from rats in the presence of BmK AS at concentrations greater than 100 nM, leading to the suggestion that BmK AS could suppress neuronal excitability by inhibiting the sodium currents (I_Na) (Chen et al., 2006).

Little has been done to explain the differences in reported effects of BmK AS on cell excitability. Such differences might be attributed to the differential modulation by BmK AS of the various VGSC subtypes, or to the different doses used. To resolve these differences it is necessary to elucidate the mode of action of BmK AS before analysing the effects on the amplitude of I_Na. Moreover, binding assays indicated that the equilibrium dissociation constant (K_d) for BmK AS was ∼0.46 nM on rat brain synaptosomes (Li et al., 2000), much lower than the concentrations tested in much of the earlier work.

In this study therefore, we investigated the modulating effects of BmK AS over a wide dose range from 0.05 to 500 nM on the rat brain-type VGSC α-subunit II [rNav1.2a; nomenclature follows Alexander et al. (2008)], heterologously expressed in Xenopus oocytes, as well as the VGSCs constitutively expressed in the DRG neuroblastoma ND7-23 cell line. The results showed BmK AS was distinguished from the other known polypeptide toxins by its unexpected, U-shaped dose-dependent effects on both brain-type and DRG-type VGSCs, although the detailed dose–response relationship differed in the two experimental systems.

Methods

cRNA preparation and expression

Plasmid pNa200 in combination with rNav1.2a cDNA was a gift from Dr Alan L Goldin (University of California, Irvine, CA, USA) and sequenced before RNA transcription. cRNAs of rNav1.2a were synthesized from Not I linearized DNA templates with T7 RNA polymerase message machine transcription kit (Ambion, Austin, TX, USA). The synthesized mRNA was analysed by agarose gel and reclaimed before storing in individual ampoules at −20°C.

Female Xenopus laevis frogs were provided by the animal centre of Shanghai Institute of Neuroscience. Oocytes were surgically removed and were incubated with collagenase (2 mg·mL⁻¹, type IA, Sigma, Saint Louis, MO, USA) in calcium-free OR₂ medium (in mM: 96 NaCl, 2 KCl, 1 MgCl₂ and 5 HEPES, pH 7.5) at 20°C for around 3 h. After washing, healthy oocytes at stage V–VI were selected for cRNA injection at a dose of 1–4 ng per oocyte and then incubated in ND96 medium (in mM: 96 NaCl, 2 KCl, 1.8 CaCl₂, 1 MgCl₂ and 5 HEPES, pH 7.5, supplemented with 5 pyruvate and 0.1 mg·mL⁻¹ gentamicin) at 20°C for 40 h (Goldin, 1991).

Two-electrode voltage clamp recordings

Whole-cell currents were recorded at room temperature with TURBO TEC-03X amplifier (NPI Electronic Instruments, Tamm, Germany) and Cellworks E 5.5 software (NPI Electronic Instruments). The electrodes were filled with 3 M KCl and balanced in bath solution (ND96) for at least 30 min before recording. Data were sampled at 10 kHz and low-pass filtered at 1.3 kHz. Data acquisition was performed by Cellworks Reader 3.6 software (NPI Electronic Instruments). Capacitance transients and linear leakages were removed by subtracting the traces before and after 500 nM tetrodotoxin addition. Oocytes were not used when the leakages were beyond 0.10 µA.

Cell culture and whole-cell patch clamp recordings

ND7-23 cells (Shanghai cell bank of Chinese Academy of Science, Shanghai, China) were obtained from the Shanghai cell bank of the Chinese Academy of Science. The cells were cultured in Dulbecco's modified Eagle medium (DMEM; Gibco, Invitrogen, Grand Island, NY, USA) supplemented with 2 mM l-glutamine, 10% heat-inactivated fetal bovine serum (FBS; Gibco, Invitrogen). Culture dishes were incubated at 37°C in a humidified atmosphere containing 5% CO₂ and subcultured approximately every 2–3 days.

Whole-cell I_Na recordings were performed as described previously (John et al., 2004), using an EPC-9 amplifier (HEKA Elektronik, Lambrecht, Germany) at room temperature. Patch pipettes were fabricated from glass capillary tubes by PP-830 Puller (Narishige, Tokyo, Japan) with the resistance of 2–3 MΩ. The internal solution contained (in mM): 120 CsF, 10 HEPES, 10 EGTA, 15 NaCl, pH 7.25. The external solution contained (in mM): 140 NaCl, 5 HEPES, 1.3 MgCl₂, 1 CaCl₂, 11 glucose, 4.7 KCl, pH 7.4 (Sigma). The internal and external solutions were adjusted to osmolarities of 285–290 and 295–300 mOsm respectively.

Data acquisition and stimulation protocols were controlled by Pulse/PusleFit 8.3 software (HEKA Elektronik). Capacitance transients and series resistance errors were compensated by 75%. Cells were discarded when the series resistance values were over 20 MΩ. Linear leakage was subtracted using P/4 protocol. Data were sampled at 20 kHz and low-pass filtered at 10 kHz. The rate of solution exchange was studied using solutions with different NaCl concentrations and found to be about 95% complete within 60 s.

Electrophysiological protocols and data analysis

In Xenopus oocytes, the holding potentials were −100 mV. The I_Na were elicited by step pulses ranging from –100 to +65 mV for 100 ms with increments of 5 mV. The amplitudes of transient sodium currents (I_NaT) before and after BmK AS application were normalized to the peak I_NaT before BmK AS application to generate the I–V curves of the I_NaT. For determining the voltage dependence of activation, the sodium conductance was calculated using the formula:

(1)

where I(V) is the I_NaT at the command voltage V, and Er_Na is the reversal potential of sodium flow estimated from the I–V curve. The conductances were normalized to the maximal value between −90 and +20 mV and fitted to a Boltzmann equation:

(2)

where V_1/2 is the voltage at which half-maximal activation occurs, and k describes the slope of the fit. The inactivation kinetics was analysed by fitting the decay course of I_NaT to a tertiary exponential function:

(3)

where t is time, t₀ is the time when the currents were just starting to exponentially decrease, A₁, A₂ and A₃ represent the amplitudes of channels inactivating with the time constants τ₁, τ₂ and τ₃, respectively, and C is the steady-state asymptote approximating to the non-inactivation persistent sodium current (I_NaP). The voltage dependence of fast and slow inactivation were analysed by two-pulse protocols, composed of a 10 or 100 ms prepulse, respectively, to potentials ranging from −100 to +30 mV with the increments of 10 mV followed by a test pulse of 0 mV for 50 ms. The amplitudes of the I_Na were normalized to their maximal value and plotted as channel availability versus prepulse potential. Data were then fitted to a Boltzmann equation:

(4)

where V_1/2 is the voltage at which 50% of the fast- or slow-inactivation component is inactivated, k is the slope factor, and C is the steady-state asymptote.

The ND7-23 cells were held at the holding potential of −120 mV. I_Na were elicited by the step pulses ranging from −90 to +70 mV for 50 ms with increments of 5 mV. The voltage dependence of activation was analysed by the same procedure as in the Xenopus oocytes. The inactivation kinetics was analysed by fitting the decay course of I_NaT to a double exponential function:

(5)

where t is time, t₀ is the time when the currents were just starting to exponentially decrease, A₁ and A₂ represent the amplitudes of the channels inactivating with the time constants τ₁ and τ₂, respectively, and C is the steady-state asymptote approximating to the leakage. Recovery kinetics were analysed by a traditional two-pulse protocol consisting of a 200 ms prepulse to 0 mV to inactivate all the VGSCs, followed by resting at −120 mV for the time in range of 1–39 ms with the increments of 2 ms, and a 10 ms test pulse to 0 mV. The current amplitude of the test pulse was divided by that of the prepulse, plotted as the recovery percentage versus the resting duration and fitted to a single exponential function:

(6)

where t is time, A denotes the maximal recovery percentage with the time constant τ. The voltage dependence of recovery and fast inactivation was analysed by a multi-pulse protocol, composed of a 200 ms prepulse to 0 mV followed by a 5 ms rest at −120 mV to recover the fast component, a 3 ms prepulse to the potentials ranging from −150 to +30 mV with increments of 10 mV, and a 10 ms test pulse to 0 mV. The amplitudes of the I_Na were normalized to their maximal value and plotted versus the second prepulse potential. Data were then fitted to a double Boltzmann equation:

(7)

where f is the fraction of the recovery component, V_{1/2_1} and V_{1/2_2} are the voltages at which 50% of the channels are recovered and inactivated, respectively, k₁ and k₂ are the slope factors, and C is the steady-state asymptote. The voltage dependence of the slow inactivation was analysed by a traditional two-pulse protocol, composed of a 30 ms prepulse to potentials ranging from −150 to +30 mV with the increments of 10 mV followed by a 5 ms rest at −120 mV to recover the fast component, and a test pulse to 0 mV for 10 ms. The amplitudes of the I_Na were normalized to their maximal value, plotted versus the prepulse potential and then fitted to the Eqn 4.

The effects of BmK AS took about 5–10 min to appear and more than 30 min to achieve a stable effect (data not shown). In this study, unless indicated, the effects of each concentration of toxin were examined after perfusion for 20 min.

Data analysis

The raw data were analysed by Origin 8.0 (OriginLab, USA). The results are shown as means ± SEM with the number of experiments shown as n in the figure legends. Differences between means were analysed by Student's test or by one-way anova, with P < 0.05 taken to show a significant difference.

Materials

The crude BmK venom was purchased from a scorpion culture farm in He-Nan Province, China. The venom was filtered with a Sephadex G-50 column and purified as described by Liu et al. (1996). The purity of the toxin was confirmed by mass spectrometry as well as peptide sequencing. Stock solutions of BmK AS (0.1 mM BmK AS plus 25 g·L⁻¹ BSA) was dissolved in the bath solution to yield the desired concentration before the electrophysiological recording.

Results

U-shaped dose-dependent shift of the voltage dependence of activation

The effects of 0.1, 1, 10, 100 and 500 nM of BmK AS were determined on the rNav1.2a subunits expressed heterologously in Xenopus oocytes. The I_Na of rNav1.2a were evoked by 100 ms step pulses ranging from −100 to +65 mV (Figure 1A–C). Before application of BmK AS, the samples were selected if they were provided with accordant I–V relationship of the I_NaT with the peak I_NaT appearing at voltages between −30 and −20 mV (Figure 1D). After perfusion with 500 nM BmK AS for 20 min, the I–V curve of the I_NaT was shifted in a depolarized direction (Figure 1E). A similar but not significant effect could be observed in the presence of 0.1 nM BmK AS, whereas 1 nM BmK AS imposed a slight opposite effect (Figure 1E). Taking into consideration the changes in the Er_Na, the voltage dependence of activation was then analysed by fitting the conductance–voltage (G–V) relationship of the I_NaT to a Boltzmann function (see Methods). As a result, only the highest concentration of BmK AS (500 nM) significantly moved the half-maximal voltage (V_1/2) of activation (Table 1) towards depolarization by 10.3 mV, compared with the control value, before BmK AS application. However, the V_1/2 values for both 0.1 and 500 nM BmK AS were significantly different from each other.

Figure 1.

Effect of BmK AS on the I–V relationship of rNav1.2a. The sodium currents were evoked by step pulses ranging from −100 to +65 mV for 100 ms from the holding potential of −100 mV in increments of 5 mV. (A–C) Representative sodium currents before and after application of 0.1, 10 or 500 nM BmK AS, shown at intervals of 10 mV. (D–E) Statistical plots of the I–V relationship of rNav1.2a before (D) and after (E) application of 0.1, 1, 10, 100 and 500 nM BmK AS. Each oocyte was subjected to one concentration of BmK AS for 20 min. n= 12, 10, 10, 10 and 14 respectively. BmK, Buthus martensi Karsch; I_NaT, transient sodium current.

Table 1.

Parameters for voltage dependence of activation and inactivation of rNav1.2a expressed in Xenopus laevis oocytes

Concentration	CTRL	0.1 nM	1 nM	10 nM	100 nM	500 nM
n	56	12	10	10	10	14
Activation
V1/2 (mV)	−40.9 ± 0.19	−36.0 ± 0.26#	−45.4 ± 0.15	−41.2 ± 0.16	−37.1 ± 0.24#	−30.8 ± 0.22*#
k (mV)	4.2 ± 0.16	3.6 ± 0.23	3.0 ± 0.13	3.5 ± 0.14	4.4 ± 0.21	5.6 ± 0.19
Fast inactivation
V1/2 (mV)	−42.0 ± 0.87	−36.7 ± 0.64#	−46.8 ± 0.74	−40.9 ± 1.02	−41.2 ± 1.17	−33.3 ± 0.97*#
k (mV)	5.6 ± 0.65	5.9 ± 0.48	6.4 ± 0.56	6.0 ± 0.76	5.8 ± 0.91	8.4 ± 0.74*
Slow inactivation
V1/2 (mV)	−51.9 ± 0.58	−41.7 ± 0.37*#	−55.6 ± 0.46	−51.6 ± 0.94	−51.1 ± 0.54	−38.2 ± 0.66*#
k (mV)	5.4 ± 0.43	5.7 ± 0.26	5.6 ± 0.35	5.3 ± 0.52	5.4 ± 0.24	6.9 ± 0.41

^*P < 0.05; significant difference between control and Buthus martensi Karsch (BmK) AS values; one-way anova.

^#P < 0.05; significant difference between BmK AS at 1 nM and the other concentrations; one-way anova.

In the DRG neuroblastoma ND7-23 cells, the I_Na were activated by 50 ms step pulses in the range from −90 to +70 mV (Figure 2A,B), and they shared a homogenous I–V relationship and always peaked at voltages between −10 and 0 mV in the absence of BmK AS. After perfusion with 0.1, 1 or 100 nM BmK AS, the I–V curve was significantly hyperpolarized (Figure 2C) with decreased V_1/2 of activation (Table 2). However at 10 nM, BmK AS did not affect the I–V curve or the V_1/2 of activation (Figure 2D and Table 2). As the effects of 0.1 and 100 nM BmK AS were large, we measured the effects at intermediate concentrations (0.05 and 30 nM). The results (Figure S1A) showed that these concentrations exerted correspondingly intermediate effects shifting the V_1/2 of activation by −6.0 (0.05 nM) and −9.2 mV (30 nM). Thus, the voltage dependence of activation of ND7-23 VGSCs was also modulated by BmK AS in a U-shaped dose-dependent manner, as in oocytes expressing rNav1.2a.

Figure 2.

Effect of BmK AS on the I–V relationship of the VGSCs in ND7-23cells. The sodium currents were evoked by step pulses ranging from −90 to +70 mV for 50 ms from the holding potential of −120 mV in increments of 5 mV. (A–B) Representative sodium currents before and after perfusion for 20 min with 0.1 and 100 nM (A), or 10 and 100 nM (B) BmK AS in succession, shown at intervals of 10 mV. (C–D) I–V relationship of the representative sodium currents shown in (A) and (B), respectively. Note that in (D), the Er_Na decreased after application of 100 nM BmK AS and recovered after washing, indicating that the change of the Er_Na was caused by the toxin, not by membrane leakage. BmK, Buthus martensi Karsch; Er_Na, reversal potential of sodium flow; I_NaT, transient sodium current; VGSC, voltage-gated sodium channel.

Table 2.

Parameters for voltage dependence of activation, inactivation and recovery of the ND7-23 VGSCs

Concentration	CTRL	0.1 nM	1 nM	10 nM	100 nM
n	46	12	10	12	12
Activation
V1/2 (mV)	−21.8 ± 0.23	−31.3 ± 0.46*	−30.0 ± 0.54*	−25.5 ± 0.97	−39.9 ± 0.61*
k (mV)	7.7 ± 0.28	7.0 ± 0.35	6.7 ± 0.41	6.1 ± 0.53	9.5 ± 0.56
Fast inactivation
V1/2 (mV)	−24.0 ± 0.34	−31.7 ± 0.38	−28.7 ± 0.29	−25.5 ± 0.48	−31.8 ± 1.15
k (mV)	8.2 ± 0.27	9.3 ± 0.28	9.3 ± 0.22	8.3 ± 0.38	14.0 ± 0.77*
Slow inactivation
V1/2 (mV)	−42.1 ± 0.95	−57.0 ± 0.59*	−48.6 ± 0.87	−41.0 ± 1.22	−76.9 ± 0.56*
k (mV)	10.8 ± 0.54	12.5 ± 0.38	12.3 ± 0.33	11.8 ± 0.49	13.0 ± 0.26
Recovery
V1/2 (mV)	−98.1 ± 4.04	−110.0 ± 0.89*	−101.1 ± 1.10	−103.8 ± 2.22	−109.0 ± 1.02*
k (mV)	10.8 ± 3.38	12.9 ± 0.99	10.9 ± 0.92	11.0 ± 2.06	9.7 ± 0.95

^*P < 0.05; significant difference between control and BmK AS; one-way anova.

BmK, Buthus martensi Karsch; VGSC, voltage-gated sodium channel.

U-shaped dose-dependent shift of the voltage dependence of inactivation

When expressed in Xenopus oocytes, a proportion of the fast-gating rNav1.2a channels take on the character of the slow-gating VGSCs and inactivate in a process called slow inactivation, which is mechanistically different from fast inactivation. We could therefore analyse the effects of the toxin on the voltage dependence of fast and slow inactivation, separately. As shown in Table 1, the V_1/2 of either fast or slow inactivation was shifted to less negative values (depolarization) by 0.1 and 500 nM BmK AS. No changes in the inactivation process were induced by the other concentrations of BmK AS (10 and 100 nM)

In the ND7-23 cells, the endogenous VGSCs inactivate rapidly and completely. However, they consisted of two components inactivating with different time constants (see below), and the slower inactivation was more affected by BmK AS than the faster inactivation (Figure 3). In terms of the V_1/2 of inactivation (Table 2), the fast inactivation was not changed over the concentration range used (0.1–100 nM), but the slow inactivation process was shifted to more negative values at 0.1 and 100 nM BmK AS.

Figure 3.

Effects of BmK AS on the voltage dependence of inactivation and recovery from inactivation in ND7-23 cells. Each cell was subjected to one concentration of BmK AS. n= 46, 12, 10, 12 and 12, respectively, in control and BmK AS of 0.1, 1, 10 and 100 nM. (A) Voltage dependence of the recovery and the faster inactivation, analysed by a multi-pulse protocol (top) composed of a 200 ms prepulse to 0 mV followed by 5 ms resting at −120 mV to recover the fast component, a 3 ms prepulse to the potentials ranging from −150 to +30 mV in increments of 10 mV, and a 10 ms test pulse to 0 mV. The plots of channel availability versus voltage of the second prepulse (below) fitted well with a double Boltzmann function, where the first phase represents the voltage-dependent decrease of recovery and the second phase represents the voltage-dependent increase of inactivation. The fitting parameters are indicated in Table 2. (B) Voltage dependence of the slow inactivation, analysed by a classical two-pulse protocol (top) composed of a 30 ms prepulse to potentials ranging from −150 to +30 mV in increments of 10 mV followed by 5 ms resting at −120 mV to recover the faster component, and a test pulse to 0 mV for 10 ms. The plots of channel availability versus prepulse voltage (below) fitted well to a single Boltzmann function and the fitting parameters are indicated in Table 2. BmK, Buthus martensi Karsch; I_Na, sodium current.

It is relevant here to note that the slope factor (k) of the fast inactivation curve of both rNav1.2a (Table 1) and the ND7-23 VGSCs (Table 2) was increased by the higher concentrations of BmK AS (500 or 100 nM respectively)

Bidirectional effects on the open-state inactivation kinetics

The slowly inactivating rNav1.2a subunits produced steady-state sodium currents (I_NaSS) present at the end of the 100 ms depolarizing pulses in Xenopus oocytes. BmK AS enhanced the I_NaSS at 0.1 nM (Figure 1A) but reduced it at 10 nM (Figure 1B; mean values in Figure 4A). Kinetic analysis of the inactivation showed that BmK AS did not affect the inactivation time constants (Figure 4B) but did change the ratio between the fast- and the slow-inactivation components. In the presence of 0.1 nM BmK AS, a small but significant fraction of the VGSCs were transformed from the fast-inactivation component into the slow-inactivation component, whereas in the presence of 10 nM of BmK AS, an equal fraction of VGSCs were transformed from the slow-inactivation component into the fast-inactivation component (Figure 4C).

Figure 4.

Effects of BmK AS on the inactivation kinetics. (A) The proportion of steady-state rNav1.2a current at the end of depolarization for 100 ms to −20 mV, before and after application of 0.1, 1, 10, 100 or 500 nM BmK AS for 20 min. n= 12, 10, 10, 10 and 14 respectively. The I_NaSS was measured as the mean of the remainder currents during 99–100 ms. (B,C) Kinetic analysis of the inactivation of rNav1.2a at −20 mV in the absence and presence of 0.1 (n= 8) or 10 (n= 8) nM BmK AS. The decay time course of the sodium currents at −20 mV were fitted with a ternary exponential function, by which two fast-inactivation components (τ₁ and τ₂) and two slow-inactivation components (τ₃ and I_NaP) were separated according to the inactivation time constant (B). (C) The proportion of four inactivation components of rNav1.2a at −20 mV. (D) The fraction of residual sodium current after depolarization to +10 mV for 10 ms in ND7-23 cells, before and after application of 0.1, 1, 10 or 100 nM BmK AS for 20 min. n= 12, 10, 12 and 12 respectively. (E,F) Kinetic analysis of the inactivation of ND7-23 VGSCs at +10 mV in the absence and presence of 10 (n= 7) or 100 (n= 7) nM BmK AS. The decay time course of the sodium currents at +10 mV were fitted with a double exponential functions, by which a faster-(τ₁) inactivation component and a slowish-inactivation component (τ₂) were separated according to the inactivation time constant (E). (F) The proportion of two inactivation components of the ND7-23 VGSCs at +10 mV. *P < 0.05, **P < 0.01 and ***P < 0.001; significant difference between the control and BmK AS values; paired Student's t-test. BmK, Buthus martensi Karsch; I_Na, sodium current; I_NaP, persistent sodium current; I_NaSS, steady-state sodium current; VGSC, voltage-gated sodium channel.

Likewise, the inactivation kinetics of the ND7-23 VGSCs was affected by BmK AS in a bidirectional manner. In the absence of BmK AS, after depolarization to +10 mV for 10 ms, there was only about 2% of the I_Na left. This residual current was reduced by about 40% by 10 nM, whereas it was enhanced several fold by 100 nM BmK AS (Figure 4D). The acceleration of the inactivation with 10 nM BmK AS was not only due to the slight decrease of the fast-inactivation time constant (Figure 4E) but also to the increased proportion of the faster component (Figure 4F). Meanwhile, the deceleration of the inactivation with 100 nM BmK AS could be attributed partly to the increase of the slow-inactivation time constant (Figure 4E) and partly to the increased proportion of the slower component (Figure 4F). Nevertheless, 10 nM BmK AS also increased the slower inactivation time constant (Figure 4E), and the effect was not significantly different (P= 0.89, one-way anova) from that of 100 nM BmK AS.

U-shaped dose-dependent inhibition of the recovery from inactivation

For kinetic correlations between the inactivation and recovery from inactivation of VGSC, recovery kinetics were examined on the ND7-23 VGSCs. As shown in Figure 5, the time course of recovery within 39 ms was fitted well with a single exponential function before and after application of BmK AS, and the recovery time constant was increased by BmK AS from 3.22 ± 0.18 ms to 8.34 ± 0.19, 8.20 ± 0.21, 5.94 ± 0.19 and 11.9 ± 0.31 ms, at the concentration of 0.1, 1, 10 and 100 nM respectively (Figure 5). Moreover, 0.1 and 100 nM BmK AS shifted the voltage dependence of recovery significantly towards hyperpolarization (Figure 3A and Table 2), leading to the decline of the total recovery of the VGSCs at −120 mV shown in Figure 5, whereas 1 and 10 nM BmK AS did not alter total recovery. In addition, 0.05 and 30 nM BmK AS also showed intermediate effects for recovery, compared with the effects of 0.1 and 100 nM BmK AS (Figure S1B). Overall our results showed that, in ND7-23 cells, BmK AS inhibited the recovery of VGSCs from the inactivation in a U-shaped, dose-dependent manner.

Figure 5.

Effect of Buthus martensi Karsch (BmK) AS on the recovery from inactivation in ND7-23 cells. The time course of recovery at −120 mV was determined by a two-pulse protocol (inset), consisting of a 200 ms prepulse to 0 mV followed by resting at −120 mV for the time varying from 1 to 39 ms in increments of 2 ms, and a test pulse to 0 mV for 10 ms. The recovery time course was then fitted with a single exponential function to obtain the time constant of recovery. Before application of BmK AS (CTRL), the recovery time constant was 3.22 ± 0.18 ms (n= 36). After application of 0.1, 1, 10 or 100 nM BmK AS, it increased to 8.34 ± 0.19 (n= 9), 8.20 ± 0.21 (n= 9), 5.94 ± 0.19 (n= 9) or 11.9 ± 0.31 (n= 9) ms respectively. All the concentrations of BmK AS increased the recovery time constant significantly (P < 0.05). Also the effects of 0.1 and 100 nM were significantly different from the effect of 10 nM (P < 0.05); one-way anova.

Bidirectional modulation of the I_Na amplitude at different concentration

Alteration of the amplitude of the I_Narepresents the complex effects of the toxin. In the presence of 0.1 nM BmK AS, the peak I_NaT of rNav1.2a (Figure 6A) and ND7-23 cells (Figure 6B) was enhanced by 14.9 ± 1.3% and 33.7 ± 4.7% respectively. At the highest concentration used in ND7-23 cells (300 nM BmK AS), around 90% of the I_Na was blocked (Figure 6B and Figure S2), but no significant effects were detected with the other concentrations of BmK AS. Thus BmK AS increased I_Na at a low concentration and reduced I_Na at high concentrations.

Figure 6.

Effect of Buthus martensi Karsch (BmK) AS on the amplitudes of sodium currents. (A) The normalized value of peak transient sodium current (I_NaT) of rNav1.2a expressed in Xenopus oocytes in the presence of 0.1, 1, 10, 100 and 500 nM BmK AS. n= 12, 10, 10, 10 and 14 respectively. (B) The normalized value of peak I_NaT in ND7-23 cells in the presence of 0.1, 1, 10, 100 and 300 nM BmK AS. n= 12, 10, 12, 12 and 7 respectively. ***P < 0.001 significant difference between the control and BmK AS; paired Student's t-test.

Discussion

From our results, we concluded that BmK AS (0.1–500 nM) modulated the activation, inactivation and recovery of the brain and DRG-type VGSCs in a U-shaped dose-dependent manner. This U-shaped dose-dependence is a unique feature of BmK AS among the polypeptide modulators of VGSC, acting extracellularly. Such U-shaped curves are more usually found with synthetic, small molecules, bearing two or more binding sites, responsible for opposite effects. In the Xenopus oocytes, BmK AS had both depolarizing and hyperpolarizing effects on the voltage dependence of activation and inactivation of rNav1.2a. BmK AS also exerts its enhancing effects at the same time as its inhibitory effects, on the open-state VGSC inactivation in both systems. Hence, the U-shaped character of the response curve for BmK AS might be explained by its occupation of two different receptor sites, one of which is of higher affinity and responsible for the inhibition of inactivation, whereas the other is of lower affinity and responsible for the potentiation of inactivation. Binding of BmK AS to two such sites would also result in the shift of the voltage in opposite directions. The relative strength of the opposing effects would depend on the toxin concentration and the VGSC subtype.

Among the scorpion polypeptide toxins targeting the VGSCs, inhibition of the inactivation and recovery is the distinctive characteristic of the α-toxins, exerted via the receptor site-3, while a negative shift of the voltage dependence of activation is characteristic for the β-toxins, via the receptor site-4. Our results indicate that BmK AS exhibits the pharmacological activities of both the α- and β-toxins. This finding is in accord with earlier results showing that BmK AS-1 and AaH IT4 were both recognized by the anti-β-toxin antibodies (Loret et al., 1991; Jia et al., 2000) and compete with both the α- and β-toxins for binding to the VGSCs (Loret et al., 1991; Jia et al., 1999). On the other hand, Css IV, a typical β-toxin from the American scorpion Centruroides suffusus suffuses (Css) bound to the rat brain VGSCs at two types of specific high-affinity sites: a variable minor type with K_d of ∼0.1 nM and a major type with a K_d of ∼5 nM (Thomsen et al., 1995). Because our results also implied two classes of binding sites with affinities around 0.1 and 10 nM BmK AS on the rat brain- and DRG-type VGSCs, the binding sites for BmK AS and Css IV might be analogous or overlapping, and the variable minor sites for Css IV might become the major sites for BmK AS to exert a α-toxin-like modulation. However, BmK AS did facilitate activation without the depolarizing pre-stimulation that is required for Css IV (Cestele et al., 1998), suggesting the binding sites for BmK AS to exert its β-toxin-like modulation must be different from that for Css IV.

The effect of the charge of the polypeptide may be important at high concentration of BmK AS. Although the higher concentrations used here were still sub-micromolar, the positive charges in BmK AS may affect the surface potential of the cell membrane near the VGSC and thus change the VGSC microenvironment. In the presence of high concentrations of BmK AS, such a charge effect may become dominant, inducing hyperpolarization of the voltage dependence of these channels, as well as the Er_Na (Figure 2D). Nevertheless, any charge effect of BmK AS could not exclude the possibility that the polypeptide AS binds to site-4, as BmK AS facilitated activation before inducing the shift of Er_Na (Figure S3).

A marked reduction of the I_Na can block the firing of action potentials, and this has been suggested to be the mechanism of BmK AS in inducing analgesic effects (Chen et al., 2006). The amplitude of I_Na is influenced by many factors including the amount of available channels, the single channel conductance, the voltage dependence, kinetics and the concentration of intracellular and extracellular sodium ions. In this study, we found that BmK AS increased the I_Na at an extremely low concentration of 0.1 nM in both systems. We expected this effect to be U-shaped dose-dependent. However, high concentrations of BmK AS reduced the I_Na in ND7-23 cells, as in the acutely dissociated DRG neurons. This may be due partly to the marked negative shift of the Er_Na reducing the driving force of sodium influx. At the same time it seems that the strong hyperpolarizing shift of the voltage dependence of inactivation (Figure 3) contributed more to the abolition of the I_Na by high dose of BmK AS. As shown in Figure S2, the I_Na was abolished by 300 nM BmK AS when the holding potential was −120 mV, but some of the I_Na recovered when the holding potential was changed to −160 mV, demonstrating that inactivation of the closed-state VGSCs is an important component of the action of BmK AS on the DRG-type VGSCs. Higher concentrations of BmK AS no longer increased the I_Na amplitude in Xenopus oocytes expressing rNav1.2a. This may be because the voltage dependence of activation of rNav1.2a was more shifted than the Er_Na in a depolarizing direction by 500 nM BmK AS, thus reducing the driving force of sodium influx.

Although high concentrations of BmK AS severely reduced the I_Na via hyperpolarizing the voltage dependence of DRG neuroblastoma VGSCs, the polypeptide may simultaneously activate I_Na near the resting potential, which might result in a raised intracellular sodium concentration ([Na⁺]_i). In the neuroblastoma B104 cells, 500 nM BmK AS elevated the [Na⁺]_i by 18% at the resting potential (Tan et al., 2004). Increased [Na⁺]_i can induce increased intracellular calcium concentration ([Ca²⁺]_i), which in turn activates downstream signalling pathways affecting ion channels and transporters. Thus, the effect of BmK AS on cell excitability are not only regulated by the concentration of BmK AS, but are also dependent upon the expression of various proteins within different cell types that functionally influence the [Na⁺]_i and [Ca²⁺]_i.

Glossary

Abbreviations:

AaH

Androctonus australis Hector

BmK

Buthus martensi Karsch

[Ca²⁺]_i

intracellular calcium concentration

Css

Centruroides suffusus suffusus

Er_Na

reversal potential of sodium flow

I_Na

sodium current

I_NaP

persistent sodium current

I_NaSS

steady-state sodium current

I_NaT

transient sodium current

K_d

equilibrium dissociation constant

[Na⁺]_i

intracellular sodium concentration

V_1/2

half-maximal voltage

VGSC

voltage-gated sodium channel

Conflict of interest

The authors state no conflict of interest.

Supporting Information

Additional Supporting Information may be found in the online version of this article:

Figure S1 Effect of 0.05 nM (n = 8) and 30 nM (n = 8) Buthus martensi Karsch (BmK) AS on the activation (A) and recovery (B) of the ND7-23 voltage-gated sodium channels (VGSCs), analysed as described in Methods. The fitting parameters (V_1/2 and k) of activation and the time constant of recovery are indicated as follows: −20.3 ± 0.37 mV, 8.0 ± 0.25 mV, 3.44 ± 0.22 ms (CTRL); −26.3 ± 0.13 mV, 7.3 ± 0.08 mV, 6.48 ± 0.22 ms (0.05 nM); −29.5 ± 0.26 mV, 8.6 ± 0.19 mV, 7.56 ± 0.12 ms (30 nM). 10 nM (n = 4, 6.08 ± 0.13 ms) and 100 nM (n = 4, 9.97 ± 0.27 ms) BmK AS were used for two positive controls to make sure the consistency of the toxin efficiency of BmK AS.

Figure S2 (A–C) Raw sodium currents before (A) and after (B and C) application of 300 nM Buthus martensi Karsch (BmK) AS for 20 min in ND7-23 cells. The currents were activated by step pulses ranging from −90 to +70 mV with the increments of 5 mV from the holding potential of −120 mV (A and B) or −160 mV (C). (D) I–V relationship of the currents in (A–C). Note the I_Na was completely abolished by 300 nM BmK AS when holding potential was −120 mV, but some of it recovered when holding potential was changed to −160 mV.

Figure S3 Representative example of the two-phase impact of 100 nM Buthus martensi Karsch (BmK) AS. At the first phase, the toxin only shifts the voltage dependence without changing the reversal potential of sodium flow (Er_Na). At the second phase however, the toxin severely shifts the Er_Na with further hyperpolarization of the voltage dependence. Note the time for BmK AS shifting the Er_Na is variable among the cells.

Please note: Wiley-Blackwell are not responsible for the content or functionality of any supporting materials supplied by the authors. Any queries (other than missing material) should be directed to the corresponding author for the article.