Important modifications by sugammadex, a modified γ-cyclodextrin, of ion currents in differentiated NSC-34 neuronal cells

Background Sugammadex (SGX) is a modified γ-cyclodextrin used for reversal of steroidal neuromuscular blocking agents during general anesthesia. Despite its application in clinical use, whether SGX treatment exerts any effects on membrane ion currents in neurons remains largely unclear. In this study, effects of SGX treatment on ion currents, particularly on delayed-rectifier K+ current [I K(DR)], were extensively investigated in differentiated NSC-34 neuronal cells. Results After cells were exposed to SGX (30 μM), there was a reduction in the amplitude of I K(DR) followed by an apparent slowing in current activation in response to membrane depolarization. The challenge of cells with SGX produced a depolarized shift by 15 mV in the activation curve of I K(DR) accompanied by increased gating charge of this current. However, the inactivation curve of I K(DR) remained unchanged following SGX treatment, as compared with that in untreated cells. According to a minimal reaction scheme, the lengthening of activation time constant of I K(DR) caused by cell treatment with different SGX concentrations was quantitatively estimated with a dissociation constant of 17.5 μM, a value that is clinically achievable. Accumulative slowing in I K(DR) activation elicited by repetitive stimuli was enhanced in SGX-treated cells. SGX treatment did not alter the amplitude of voltage-gated Na+ currents. In SGX-treated cells, dexamethasone (30 μM), a synthetic glucocorticoid, produced little or no effect on L-type Ca2+ currents, although it effectively suppressed the amplitude of this current in untreated cells. Conclusions The treatment of SGX may influence the amplitude and gating of I K(DR) and its actions could potentially contribute to functional activities of motor neurons if similar results were found in vivo. Electronic supplementary material The online version of this article (doi:10.1186/s12868-016-0320-5) contains supplementary material, which is available to authorized users.


Background
Sugammadex (SGX) is recognized as a modified γ-cyclodextrin with a lipophilic core and a hydrophilic periphery, and it has been used clinically for reversal of neuromuscular blockade caused by rocuronium or vecuronium during general anesthesia [1][2][3][4][5]. A previous report showed that addition of SGX could cause neuronal apoptosis in primary cultures [6]. This compound per se was also reported to be effective at reversing neurodegenerative disorder of the lower motor neurons [7,8]. However, interestingly, how the treatment with SGX can perturb ionic currents remains largely unexplored.
The K V 3 channels, a subfamily of K V channels, are distinguished from other types of K V channels by more positively shifted voltage-dependent activation and by faster activation and deactivation rates [9,10]. These differences enable K V 3 channels to be major determinants of high-frequency firing existing in several types of central neurons [11][12][13][14][15]. The activity of K V 3.1 channels has been recently described as playing a crucial role in controlling the amplitude of action potentials at synapses [16]. The de novo mutations in KCNC1, which encodes K V 3.1 channels, have been also found to expand phenotypic spectra of this channel to progressive myoclonus epilepsy [17]. Therefore, these channels clearly are important targets used for investigations on electrical behaviors of central neurons including motor neurons [10].
The NSC-34 mouse motor neuron cell line is a hybridoma cell line derived from the fusion of mouse neuroblastoma with motor neuron-enriched spinal cord cells [18,19]. These cells may create an easily accessible and clonally uniform motor neuron-like cell line that overcomes the problems associated with the culturing of primary spinal motor neurons. It has been demonstrated to be a suitable model for investigations on the mechanisms of neuronal development and differentiation in vitro and for studying electrophysiological properties of motor neurons in spite of being not considered as an adult motor neuron [18][19][20]. The biophysical properties of delayed-rectifier K + current [I K(DR) ] in NSC-34 cells were previously found to resemble the K V 3.1-encoded current because of positive mRNA detection of K V 3.1 (KCNC1) [21,22]. As NSC-34 cells were differentiated, the density of I K(DR) was significantly enhanced. Previous work from our laboratory has shown that removal of membrane cholesterol by methyl-β-cyclodextrin, a cyclic oligosaccharide, could modify activation kinetics of I K(DR) in motor neuron-like cells [23]. Additionally, although voltage-gated Na + current (I Na ) in NSC-34 neuronal cells has been previously reported [24], few studies have been concerned with the biophysical or pharmacological properties of Ca 2+ currents in these cells.
Therefore, the purpose of this work was to test whether SGX treatment could exert any perturbations on ionic currents present in NSC-34 neuronal cells differentiated with low serum and retinoic acid. The biophysical and pharmacological properties of ionic currents including delayed-rectifier K + current [I K(DR) ], voltage-gated Na + current (I Na ) and L-type Ca 2+ current (I Ca,L ) in untreated and SGX-treated cells have been characterized and compared in this study. Interestingly, the present results indicate that SGX treatment is capable of modifying the activation kinetics of I K(DR) elicited by membrane depolarization in these cells in a concentration-, time-, and state-dependent manner.

Cell preparation and differentiation
NSC-34 neuronal cells were originally produced by fusion of the motor neuron-enriched, embryonic mouse spinal cords with the mouse neuroblastoma [19]. These cells were kindly provided by Professor Yuh-Jyh Jong (Department of Pediatrics, Kaohsiung Medical University Hospital, Kaohsiung City, Taiwan). They were routinely grown in 1:1 mixture of Dulbecco's modified Eagle medium (DMEM) and Ham's F12 medium that was supplemented with 10% (v/v) FBS and 1% penicillin-streptomycin. Cultures were incubated at 37 °C in a humidified environment of 5% CO 2 /95% air. The medium was replenished every 2-3 days for removal of non-adhering cells. To slow cell proliferation and enhance their maturation towards a differentiated state [20], before confluence, cells were grown in 1:1 DMEM plus Ham's F12 medium supplemented with low serum (1% FBS) and 1 μM retinoic acid. The SGX-treated cells were incubated at 37 °C for 1 h in normal Tyrode's solution containing different concentrations of this compound. The reason that the duration was set at 1 h was to ensure that SGX could exert its interaction with cell membrane.

Electrophysiological measurements
Shortly before each experiment, cells were dissociated, and an aliquot of cell suspension was transferred to a homemade recording chamber positioned on the stage of a CKX-41 inverted microscope (Olympus, Tokyo, Japan). Cells were immersed at room temperature (20-25 °C) in normal Tyrode's solution containing 1.8 mM CaCl 2 . The patch electrodes used were prepared from Kimax capillary tubes (#34500; Kimble Glass, Vineland, NJ, USA) using a vertical two-step electrode puller (PP-83 or PP-830; Narishige, Tokyo, Japan), and their tips were then fire-polished with an MF-83 micro-forge (Narishige). Experiments were performed using the whole-cell configuration of standard patch-clamp technique using either an RK-400 (Bio-Logic, Claix, France) or an Axopatch 200B (Molecular Devices, Sunnyvale, CA, USA) patch-clamp amplifier [19]. Junctional potentials that developed when the composition of the pipette solution was different from that in the bath were nulled.

Data recordings
The signals consisting of voltage and current tracings were displayed and recorded online using an ASUSPRO-BU401LG computer (ASUS, Taipei City, Taiwan) equipped with a Digidata 1440A device (Molecular Devices), and the experiments were controlled by pCLAMP 10.2 software (Molecular Devices). Current signals were low-pass filtered at 3 kHz and digitized at 10 kHz. In some experiments of verifying analog-to-digital conversion, signals are digitized using a PowerLab acquisition system with Lab-Chart 7.0 programs (AD Instruments, Gerin, Tainan City, Taiwan). The resultant data achieved during this experiment were analyzed off-line by use of various analytical tools including the LabChart 7.0 program (Gerin), Origin 8.0 (OriginLab, Northampton, MA, USA) and custommade macro procedures run under Excel 2013 (Microsoft, Redmond, WA, USA). The voltage-step profiles digitally created from pCLAMP 10.2 were employed to evaluate current-voltage (I-V) relationships or steady-state inactivation of ionic currents [e.g., I K(DR) ].

Data analyses
The relationships between the relative I K(DR) amplitude and the membrane potential obtained with or without the treatment of SGX (30 μM) were fitted with a Boltzmann function of the following form: where I max is the maximal amplitude of I K(DR) elicited by membrane depolarization, V 1/2 the voltage at which there is half-maximal activation, q the apparent gating charge, F Faraday's constant, R the universal gas constant, and T the absolute temperature. The free energy involved in the gating of I K(DR) (ΔG 0 ) was calculated assuming that there is a 2-state (i.e., closed [resting] and open) gating model inherently in the channel. ΔG 0 for I K(DR) activation at 0 mV with or without treatment of SGX would be equal to q × F × V 1/2 [25,26]. The standard errors of ΔG 0 (i.e., σ qFV 1/2 ) were calculated according to: where σ q and σ V 1/2 represent the standard errors in q and V 1/2 respectively.
The steady-state inactivation curve of I K(DR) with or without treatment of SGX was derived and plotted against the conditioning pulses and then fitted to another Boltzmann equation: where V represents the conditioning potential in mV, V 1/2 is the membrane potential for half-maximal inactivation, and k is the slope factor of inactivation curve for I K(DR) elicited by membrane depolarization.
Linear or nonlinear curve-fitting to data sets presented herein was performed using either Microsoft Solver function embedded in Excel (Microsoft) or Origin 8.0 program (OriginLab). The values are provided as the mean ± standard error of the mean (SEM) with sample sizes (n) indicating the cell number from which the results were obtained. The paired or unpaired Student's t test and a one-way analysis of variance with the leastsignificant difference method for multiple comparisons were used for statistical evaluation of differences among means. Statistical analyses were performed using the Statistical Package for Social Science 20 (SPSS; IBM Corp., Armonk, New York). Statistical significance was determined at a P value of <0.05.

Effect of SGX treatment on delayed-rectifier K + currents [I K(DR) ] in differentiated NSC-34 neuronal cells
In the first set of experiments, the whole-cell configuration of the patch-clamp technique was conducted to investigate effects of SGX on ionic currents in these cells. Cells were bathed in Ca 2+ -free Tyrode's solution that contained tetrodotoxin (1 μM) and 0.5 mM CdCl 2 , and the recording pipette was filled with K + -containing solution described in "Methods". Figure 1A depicts the I K(DR) amplitudes when cells were exposed to different concentrations (10, 30 and 100 μM) of SGX. Because the SGX concentration at 30 μM was close to the IC 50 value for inhibition of I K(DR) , most experiments shown below used such concentrations. As illustrated in Fig. 1B, C, as the cell was held at −50 mV and different voltage steps ranging from −50 to +60 mV in 10-mV increments were applied, a family of large outward currents was readily elicited. Current amplitudes were increased with greater depolarizations in an outward-rectifying manner with a reversal potential of −76 ± 2 mV (n = 12). These outward currents, which were insensitive to inhibition by either iberiotoxin (200 nM) or apamin (200 nM), have been referred to as I K(DR) , the biophysical properties of which resemble the K V 3.1-encoded K + currents [12,22]. In particular, as the cells were exposed to SGX (30 μM), the amplitude and gating of I K(DR) were altered throughout the entire range of voltage-clamp steps as compared with those taken from untreated cells. For example, when the voltage step from −50 to +50 mV was evoked, SGX (10 μM) treatment significantly decreased the amplitude of initial I K(DR) (i.e., the 60th millisecond after the beginning of voltage step) by 41 ± 3% from 1059 ± 207 to 625 ± 167 pA (n = 11, P < 0.05). However, such treatment slightly but significantly decreased I K(DR) amplitude at the end of depolarizing pulse by 22 ± 2% from 976 ± 197 to 766 ± 89 pA (n = 11, P < 0.05). No significant difference in cell capacitance between untreated (17.9 ± 1.8 pF, n = 13) and SGX-treated (18.1 ± 1.9 pF, n = 13, P > 0.05) cells could be clearly demonstrated. The averaged amplitude versus voltage relationships of I K(DR) were measured and plotted in untreated and SGX-treated cells (Fig. 1C).
The I-V relationships with or without SGX treatment were then obtained at the beginning [ Fig. 1C(a)] and end [ Fig. 1C(b)] of voltage pulses; therefore, SGX suppressed I K(DR) amplitude in a concentration-dependent manner. Additionally, neostigmine (1 μM), a typical drug used for reversing the effect of rocuronium-induced neuromuscular blockade, had no effect on the amplitude of gating of I K(DR) in these cells (Additional file 1: Fig. S1).
The activation curve of I K(DR) obtained with or without treatment of SGX Figure 2 shows the activation curve of I K(DR) in the absence and presence of SGX (30 μM) treatment. The plot of relative I K(DR) amplitude as a function of membrane potential was determined and fitted with a Boltzmann function as described under "Methods". In untreated cells, V 1/2 = 21.1 ± 2.2 mV and q = 1.93 ± 0.04 e (n = 9), while in SGX-treated cells, V 1/2 = 36.1 ± 1.9 mV and q = 2.98 ± 0.17 e (n = 9). The data showed that, as differentiated NSC-34 cells were treated with SGX (30 μM), the activation curve of this current was shifted along the voltage axis to more positive potentials by approximately 15 mV and the elementary charge for activation was elevated 1.5-fold.
According to the values of q and V 1/2 , the free energy involved in the gating of I K(DR) at 0 mV (ΔG 0 ) in the absence and presence of SGX treatment was estimated to be 3.93 ± 0.12 and 10.35 ± 0.18 kJ/mol (n = 9) respectively. The perturbation by SGX treatment of free energy (ΔΔG 0 ) involved in I K(DR) gating was calculated to be 6.42 ± 0.15 kJ/mol. It is therefore anticipated from these data that SGX treatment can increase the free energy needed for I K(DR) activation observed in differentiated NSC-34 cells.

Kinetic evaluation of I K(DR) block by SGX
During cell exposure to SGX, the activation time course of I K(DR) in response to membrane depolarization tended to become slower. The activation kinetics of SGXinduced block of I K(DR) in response to membrane depolarization was further quantitatively evaluated in cells exposed to different SGX concentrations. The concentration dependence of I K(DR) inhibition by SGX treatment is illustrated in Fig. 3. The results showed that its effects on I K(DR) could exert a concentration-dependent increase in the rate of development of inhibition. Consequently, this effect on I K(DR) in differentiated NSC-34 cells can be explained by a state-dependent blocking mechanism in which this compound may preferentially bind to the closed (resting) state of the K V channels according to a minimal kinetic scheme: where α and β represent kinetic constants for the opening and closing of the K V channel respectively; k +1 and k −1 are those for blocking and unblocking by SGX treatment; and [B] is the SGX concentration. C, O and C·B indicate the closed (resting), open, and closed-blocked state respectively.
Blocking and unblocking rate constants, k +1 and k −1 , could be determined from activation time constants of SGX-sensitive I K(DR) (i.e., difference in I K(DR) taken from untreated and SGX-treated cells) obtained in different concentrations of this compound. The rate constants were then computed using the relation: where k +1 and k −1 respectively resulted from the slope and from the y-axis intercept at [B] = 0 of the linear regression interpolating the reciprocal time constants (1/τ act ) versus the SGX concentration. The relationship between 1/τ act and [B] was computed to be linear with a correlation coefficient of 0.95 (Fig. 3B). The blocking and unblocking rate constants were thereafter calculated to be 0.0012 ms −1 μM −1 and 0.021 ms −1 respectively. According to these rate constants, dividing k −1 by k +1 yielded a dissociation constant (K D ) of 17.5 μM under SGX treatment.

Inability of SGX to modify the steady-state inactivation curve of I K(DR)
In order to further characterize effect of SGX treatment on I K(DR) , we next studied the inactivation of I K(DR) in differentiated cells using a two-step voltage protocol. The quasi-inactivation curves of I K(DR) in untreated and SGX-treated cells are illustrated in Fig. 4. In this set of experiments, a 10-s conditioning pulse in different membrane potentials preceded a test potential (300 ms in duration) to +80 mV from a holding potential of −90 mV (Fig. 4a). The relationships between the conditioning pulses and the normalized amplitudes of I K(DR) in untreated and SGX-treated cells were constructed and fitted by the Boltzmann equation described in "Methods". In untreated cells, V 1/2 = 6.2 ± 0.9 mV and k = 0.42 ± 0.02 mV (n = 9), while in SGX-treated cells, V 1/2 = 6.3 ± 0.09 mV and k = 0.44 ± 0.03 mV (n = 9). Distinguishable from the results of I K(DR) activation curve obtained in untreated and SGX-treated cells, the values for both V 1/2 and k were not found to differ significantly between the two groups of cells (P > 0.05). The experimental results therefore suggest that there is little or no modification in the voltage-dependent profile of I K(DR) inactivation following SGX treatment.

The increase of cumulative inhibition of I K(DR) activation in SGX-treated cells
In another set of experiments, we sought to determine whether in SGX-treated cells, the activation time course of I K(DR) induced by repetitive depolarizing stimuli could be altered. Under control conditions (i.e., in untreated cells), a single 100-ms depolarizing step to +50 mV from a holding potential of −50 mV was applied to produce an exponential increase of I K(DR) with a time constant of 27 ± 2 ms (n = 10). However, the activation time constant for 10-ms repetitive pulses to +50 mV, each of which lasted 10 ms with 5-ms interval at −50 mV between the depolarizing pulses, was significantly reduced to 19 ± 2 ms (n = 10, P < 0.05).
The results indicate a progressive increase in the activation rate of I K(DR) in response to repetitive depolarizing stimuli ( Fig. 5 and Additional file 2: Fig. S2). However, in SGX-treated cells, the value of activation time constant obtained during this train of short repetitive pulses became raised. In the cells exposed to 30 μM SGX, the time constants were significantly increased to 47 ± 5 ms (n = 9, P < 0.05); accordingly, the results showed that there was an excessive accumulative slowing in the activation of I K(DR) as cells were exposed to SGX.

Effect of SGX treatment on voltage-gated Na + current (I Na ) in differentiated NSC-34 cells
In the next set of experiments, we investigated effect of SGX treatment on I Na in these cells. Untreated or SGXtreated cells were bathed in Ca 2+ -free Tyrode's solution containing 10 mM TEA and the recording pipette was filled with Cs + -containing solution, the composition of which is shown in Table 1. The I Na was elicited by depolarizing pulse from −80 to −10 mV with a duration of 100 ms. In our experiments, SGX treatment at 10, 30, and 100 μM did not exert any significant effect on the peak amplitude of I Na or I Ca,L in these cells. As shown in Fig. 6, there was no significant difference in the amplitude of I Na between untreated cells (1.82 ± 0.3 nA, n = 8) and cells treated with SGX (10 μM) (1.81 ± 0.2 nA, n = 8, P > 0.05). However, in SGX-treated cells, addition of ranolazine (10 μM), a blocker of I Na [27], was capable of suppressing the peak amplitude of I Na significantly from 1.82 ± 0.2 nA to 0.98 ± 0.1 nA (n = 7, P < 0.05). Therefore, SGX treatment per se did not modify the amplitude of I Na in these cells, while the presence of ranolazine remained effective at suppressing I Na in SGX-treated cells.

Comparison of the effect of dexamethasone (DEX) on L-type Ca 2+ current (I Ca,L ) in SGX-treated cells
SGX is a modified γ-cyclodextrin and able to encapsulate steroid-like compounds [1,4]. DEX, a synthetic glucocorticoid, can produce an inhibitory effect on the reversal of rocuronium-induced neuromuscular block by SGX in functionally innervated human muscle cells [28]. Therefore, in a final set of experiments, it was further examined whether the amplitude of I Ca,L recorded from SGXtreated cells could be altered by DEX. The experiments were performed as cells were bathed in normal Tyrode's solution containing 1.8 mM CaCl 2 and 1 μM tetrodotoxin, and the recording pipette was filled with Cs + -containing solution. The peak amplitude of I Ca,L in response to membrane depolarization from −50 to 0 mV was effectively suppressed by the presence of nifedipine (1 μM), a blocker of I Ca,L , but not by ω-conotoxin GVIA (1 μM), a toxin isolated from Conus geographus and known to block N-type Ca 2+ current (Additional file 3: Fig. S3). As illustrated in Fig. 7, as untreated cells were depolarized from −50 to 0 mV, DEX (30 μM) significantly suppressed the peak amplitude of I Ca,L from 51.7 ± 5.3 to 24.5 ± 2.6 pA (n = 11, P < 0.05); however, the overall I-V relationship of this current remained unchanged in the presence of DEX. The concentration of DEX (30 μM) used in this study was fundamentally based on a previous report [29]. The results are compatible with previous observations made in pituitary tumor cells [29]. In contrast, in SGX-treated cells, DEX at the same concentration had no significant effect on the amplitude of I Ca,L [52.1 ± 5.3 pA (control), n = 11 versus 51.9 ± 5.4 pA (in the presence of DEX), n = 11, P > 0.05]; therefore, findings from these results indicate that, in SGX-treated cells, DEX-mediated inhibition of I Ca,L in response to membrane depolarization was abolished.

Discussion
In this study, the blocking of I K(DR) by SGX treatment was noted to be not instantaneous, but to develop with time immediately after the cell became depolarized. Such treatment produces a time-dependent increase in the activation time constant of I K(DR) in response to membrane depolarization. However, the inactivation curve of I K(DR) obtained in untreated and SGX-treated cells did not differ significantly. It thus appears that, as cells are exposed to SGX, this compound has a greater affinity for the closed (resting) state in the K V channel existing on differentiated NSC-34 cells. The activated channels during SGX treatment tended to produce a lower affinity than those residing in the closed state. As a result, the transition from closed to open state became slowed during cell exposure to SGX. Based on our study, it is therefore tempting to speculate that the treatment of SGX or other structurally similar agents binds to the closed state of the channel and/or blocks a prolonged channel closing in K V 3.1 channels. Alternatively, the challenge of cells with SGX or other structurally similar agents can modulate K V 3.1 channels where the closed channel conformation represents the high-affinity binding site. It also needs to be noted that, according to minimal reaction scheme shown here, the dissociation constant for SGX treatment was calculated to be 17.5 μM, a value that can be clinically achievable [1,3,5]. The K V 3.1-encoded currents were reported to be the major molecular components of I K(DR) in these cells [20,21]. Importantly, our study demonstrated that the activation kinetics of I K(DR) (i.e., K V 3.1-encoded current) in SGX-treated cells virtually became slowed in a timeand state-dependent manner. The results were in contrast with inhibitory effects of midazolam or aconitine on I K(DR) [12,21]. These two compounds exerted inhibitory effects via a mechanism through binding to the open state of the channel followed by increased rate of I K(DR) inactivation. During repetitive stimuli, the perturbation by SGX treatment of I K(DR) activation was potentiated.
By virtue of computational analysis, previous work has shown that changes in activation kinetics of I K(DR) might lead to generation of action potentials with spikefrequency adaptation [23]. The ability of K V 3.1-encoded currents to control the waveforms of action potentials at synapses has recently been reported [16]. Indeed, different de novo mutations in KCNC1 have been reported to display a wide variety of progressive myoclonus epilepsy [17]. However, SGX treatment had little or no effect on Effect of SGX treatment on voltage-gated Na + current (I Na ) in differentiated NSC-34 cells. The untreated and SGX-treated cells were bathed in Ca 2+ -free Tyrode's solution. I Na was evoked in response to membrane depolarization from −80 to −10 mV. a Superimposed I Na tracings obtained in the cell without (a) and with (b) the exposure to SGX (30 μM). I Na trace labeled (c) was obtained after addition of ranolazine (10 μM) in SGX-treated cells. The inset indicates an expanded record from dashed box. Notably, I Na amplitude obtained from untreated and SGX-treated cells did not differ significantly; however, ranolazine could effectively suppress I Na recorded from SGX-treated cells the peak amplitude of I Na . Therefore, the present results showing any changes in the amplitude and gating by SGX treatment of I K(DR) can be of pharmacological and clinical relevance.
Following SGX treatment, I K(DR) enriched in differentiated NSC-34 cells became activated at more depolarized voltages in comparison with that from untreated cells. Moreover, the steepness of activation curve for I K(DR) became significantly greater in cells exposed to SGX, indicating that the effective number of elementary charges during channel activation in SGX-treated cells was significantly raised. These results are important because they led us to estimate that energy change (ΔG 0 SGX ) for generation of I K(DR) was a value of 10.35 kJ/mol. This value was found to be significantly greater than that ΔG 0 Ctrl (i.e., 3.93 kJ/mol) in untreated cells. SGX treatment apparently is involved in voltage-sensitive gating functions of I K(DR) , despite no clear change in inactivation curve of I K(DR) between the two groups of cells. The results lead us to propose that following SGX treatment, the energy barrier for activation of K V 3.1 channels became elevated.
In our experimental conditions, supplementation of the medium with retinoic acid resulted in changes in cell morphology and an increase in mRNA expression of the K V 3.1 subunit in differentiated NSC-34 neuronal cells [20,21]. However, the modification of I K(DR) kinetics by SGX presented here did not appear to occur by the gene regulation of these channels, because significant changes in this current in differentiated NSC-34 cells generally occurred with a short time course. Moreover, no changes in I K(DR) density after treatment with SGX were observed, suggesting that such maneuver did not alter the main parts of ion channel permeation pathway (i.e., the S5 and S6 regions). It is thus possible that SGX treatment can regulate the gating kinetics of I K(DR) with no apparent change in the number of functional channels on plasma membrane.
Consistent with previous studies [29], we clearly demonstrated that addition of DEX suppressed the peak amplitude of I Ca,L in differentiated NSC-34 neuronal cells. It is important to note, however, that the inhibition by DEX of I Ca,L did not occur in SGX-treated cells, despite the ability of MgSO 4 to suppress I Ca,L amplitude in both untreated and SGX-treated cells (data not shown). Previous observations have shown that DEX did not increase the activity of large-conductance Ca 2+ -activated K + (BK Ca ) channels in pituitary cells treated with methyl-β-cyclodextrin [30], suggesting that the binding of DEX to the protein(s) of BK Ca channels relies on membrane cholesterol.
Whether DEX produces any significant effect on I Na or I K(DR) in NSC-34 cells or primary motor neurons needs to be further investigated. Whether the presence of SGX alters the cholesterol content in surface membranes and influences the DEX effect on I Ca,L also remains to be explored. Nonetheless, our experimental results are consistent with earlier work showing that DEX is effective at exerting inhibitory effects on SGX reversal of rocuronium-induced neuromuscular block [28]. Alternatively, removal by SGX of DEX-induced block of I Ca,L could be due mostly to the possibility that, similar to a mechanism by which it can reverse muscle relaxing effects by rocuronium, the SGX molecule can effectively encapsulate the DEX molecule [1,[3][4][5].
A recent report showed that methyl-β-cyclodextrin, a cholesterol-depleting agent, could induce activation of matrix metalloproteinase-2 (MMP-2) [31]. However, the reduction by SGX treatment of I K(DR) activation rate observed in differentiated NSC-34 cells was unable to be reversed by isobavachalcone (10 μM) known to be an inhibitor of MMP-2 activity [32]. Therefore, alterations by SGX treatment of activation kinetics of I K(DR) observed in differentiated NSC-34 cells are not closely associated with a mechanism linked to MMP-2 activation.
It should be noted that the pipette solution used in this study contained 3 mM ATP, which can adequately suppress the activity of ATP-sensitive K + (K ATP ) channels. The activity of K ATP channels did not differ between the untreated and SGX-treated cells (data not shown). Changes by SGX treatment of I K(DR) amplitude and gating observed in differentiated NSC-34 cells is unlikely to arise from inhibition of K ATP channels.
The observed block of I K(DR) caused by SGX treatment actually provides an intriguing mechanism for its inhibition that relies on the closed (resting) state of the K V 3.1encoded channels. The K V 3.1-encoded currents were enriched in many central neurons including hippocampal pyramidal neurons, auditory neurons, and Purkinje cells [9,11,13,14]. The activity of these K V channels is recognized as participating in electrical behaviors of fast-spiking neurons [9,10,15,16]. Challenging cells with SGX reduced the amplitude of I K(DR) and slowed the activation time course of this current recorded from differentiated NSC-34 cells as well. The present observations would clearly initiate further studies to understand the SGX effects on electrical activity of motor neurons. Whether SGX-induced reversal of rocuronium-induced neuromuscular blockade is due partly to its blocking of I K(DR) in motor neurons in vivo remains to be further investigated. Some adverse effect such as movement of a limb or the body may be partly explained by its inhibitory effect on I K(DR) .
It is noted that neostigmine, an inhibitor of acetylcholinesterase activity, is a typical drug used in anesthesia for reversing the effect of rocuronium-induced neuromuscular blockade. In our study, neostigmine (1 μM) did not exert any effect on the amplitude and gating of I K(DR) in differentiated NSC-34 cells (Additional file 1: Fig. S1). Findings from our study might explain previous observations showing that SGX could reverse more rapidly rocuronium-induced neuromuscular blockade [33] or that the treatment with SGX was associated with less frequent dry mouth than that of neostigmine [34]. Therefore, it remains to be further delineated whether SGX might exert differential actions when it is used with patients who have been administrated with DEX or other glucocorticoids [4], if similar findings presented here occur in vivo. Nonetheless, as motor neurons are exposed to SGX, the amplitude and gating of I K(DR) could be modified and these actions might significantly contribute to functional activities of motor neurons.

Conclusion
The SGX treatment may influence the amplitude and gating of I K(DR) and its actions could contribute to functional activities of motor neurons if similar findings occurred in vivo.