Evidence for Dual Activation of IK(M) and IK(Ca) Caused by QO-58 (5-(2,6-Dichloro-5-fluoropyridin-3-yl)-3-phenyl-2-(trifluoromethyl)-1H-pyrazolol[1,5-a]pyrimidin-7-one)

QO-58 (5-(2,6-dichloro-5-fluoropyridin-3-yl)-3-phenyl-2-(trifluoromethyl)-1H-pyrazolol[1,5-a]pyrimidin-7-one) has been regarded to be an activator of KV7 channels with analgesic properties. However, whether and how the presence of this compound can result in any modifications of other types of membrane ion channels in native cells are not thoroughly investigated. In this study, we investigated its perturbations on M-type K+ current (IK(M)), Ca2+-activated K+ current (IK(Ca)), large-conductance Ca2+-activated K+ (BKCa) channels, and erg-mediated K+ current (IK(erg)) identified from pituitary tumor (GH3) cells. Addition of QO-58 can increase the amplitude of IK(M) and IK(Ca) in a concentration-dependent fashion, with effective EC50 of 3.1 and 4.2 μM, respectively. This compound could shift the activation curve of IK(M) toward a leftward direction with being void of changes in the gating charge. The strength in voltage-dependent hysteresis (Vhys) of IK(M) evoked by upright triangular ramp pulse (Vramp) was enhanced by adding QO-58. The probabilities of M-type K+ (KM) channels that will be open increased upon the exposure to QO-58, although no modification in single-channel conductance was seen. Furthermore, GH3-cell exposure to QO-58 effectively increased the amplitude of IK(Ca) as well as enhanced the activity of BKCa channels. Under inside-out configuration, QO-58, applied at the cytosolic leaflet of the channel, activated BKCa-channel activity, and its increase could be attenuated by further addition of verruculogen, but not by linopirdine (10 μM). The application of QO-58 could lead to a leftward shift in the activation curve of BKCa channels with neither change in the gating charge nor in single-channel conductance. Moreover, cell exposure of QO-58 (10 μM) resulted in a minor suppression of IK(erg) amplitude in response to membrane hyperpolarization. The docking results also revealed that there are possible interactions of the QO-58 molecule with the KCNQ or KCa1.1 channel. Overall, dual activation of IK(M) and IK(Ca) caused by the presence of QO-58 eventually may have high impacts on the functional activity (e.g., anti-nociceptive effect) residing in electrically excitable cells. Care must be exercised when interpreting data generated with QO-58 as it is not entirely KCNQ/KV7 selective.


Stimulatory Effect of QO-58 on M-Type K + Current (I K(M) ) Recorded from Pituitary GH 3 Cells
In an initial stage of experiments, we wanted to test if the exposure to QO-58 produced any adjustments on the magnitude of I K(M) identified in these cells. To amplify I K(M) amplitude [31,52], we used high-K + (145 mM), Ca 2+ -free solution as a bathing medium, and the recording pipette was filled up with a K + -containing solution. As illustrated in Figure 1, one minute after addition of QO-58 (1 or 3 µM), the I K(M) magnitude evoked in response to 1-s depolarizing step from −50 to −10 mV progressively increased. For example, under cell exposure to 3 µM QO-58, I K(M) amplitude was evidently increased, as demonstrated by a considerable raise in the amplitude to 201 ± 26 pA (n = 8, p < 0.05) from a control value of 118 ± 19 pA (n = 8). The QO-58-mediated increase in I K(M) observed herein was also accompanied by the fastened activation time course of the current, as demonstrated by a shortening in the value of activation time constant (τ act ) from 68 ± 8 to 22 ± 4 ms (n = 8, p < 0.05) during exposure to 3 µM QO-58. After washout of QO-58, current amplitude returned to 122 ± 21 pA (n = 8, p < 0.05). Moreover, QO-58-mediated stimulation of I K(M) was attenuated by further addition of thyrotropin releasing hormone (TRH, 1 µM) or linopirdine (10 µM), but not by iberiotoxin (200 nM). The results were demonstrated by a decrease of current amplitude during further exposure to 1 µM TRH or 10 µM linopirdine to 125 ± 21 pA (n = 8, p < 0.05) or 127 ± 21 pA (n = 8, p < 0.05), respectively. TRH or linopirdine was reported to suppress I K(M) effectively in pituitary lactotrophs [18,31,49], while iberiotoxin is known to block large-conductance Ca 2+ -activated K + (BK Ca ) channels.  The relationship between the QO-58 concentration and the percentage increase of I K(M) was ascertained and is hence illustrated in Figure 1B. To evoke I K(M) obtained in the control period (i.e., absence of QO-58) and during cell exposure to different concentrations (1-300 µM) of QO-58, each cell was depolarized from −50 to −10 mV with a duration of 1 s. Addition of QO-58 was noticed to increase the amplitude of I K(M) in a concentrationdependent fashion. As the data became least-squares fitted to a Hill function as stated in Section 4, the half-maximal concentration (i.e., EC 50 ) required for stimulatory effect of QO-58 on I K(M) was calculated to be 3.1 µM. The data from this set of experiments reflect that QO-58 has a stimulatory effect on I K(M) in GH 3 cells in a concentration-dependent manner.

Effect of QO-58 on Average Current Versus Voltage (I-V) Relationship and Steady-State Activation Curve of I K(M)
We next continued to study if the presence of QO-58 can modify the I K(M) amplitude measured at different levels of membrane potentials. The average I-V relationship of I K(M) with or without the QO-58 application is illustrated in Figure 2A. The current amplitude noticeably arose as the membrane potential became depolarized to −30 mV, and the magnitude of QO-58-stimulated I K(M) at the level of −10 mV was higher than that at −20 or −30 mV. The relationship of relative I K(M) conductance versus membrane potential acquired in the control period (i.e., QO-58 was not present) and during cell exposure to 3 µM QO-58 was constructed ( Figure 2B). The continuous sigmoidal curve derived from experimental data sets was optimally fitted with a modified Boltzmann function (described in Section 4). In control, V 1/2 = −18.3 ± 0.7 mV (n = 8), q = 6.2 ± 0.2 e (n = 8), and in the presence of 3 µM QO-58, V 1/2 = −28.5 ± 0.8 mV (n = 8), q = 6.1 ± 0.2 e (n = 8). The results enabled us to reflect that in addition to increasing I K(M) conductance, the presence of QO-58 could exert a leftward shift (approximately 10 mV) along the voltage axis in the activation curve of the current, albeit with no change in the gating charge of the activation curve.

Effect of QO-58 on IK(M) Triggered by Triangular Ramp Pulse (Vramp) with Varying Durations
Earlier reports have shown the capability of IK(M) strength to modulate the patterns of bursting firing in central neurons [13,[15][16][17]20]. Therefore, we continued to evaluate how the presence of QO-58 could have any perturbations on IK(M) responding to upright triangular Vramp with varying durations. These Vramp waveforms were specifically designed and, during the measurements, thereafter, delivered to the tested cell through digital-toanalog conversion. In the current experimental scenario, we voltage-clamped the cell at −50 mV, and the upsloping (forward) limb from −50 to 0 mV followed by the downsloping  Earlier reports have shown the capability of I K(M) strength to modulate the patterns of bursting firing in central neurons [13,[15][16][17]20]. Therefore, we continued to evaluate how the presence of QO-58 could have any perturbations on I K(M) responding to upright triangular V ramp with varying durations. These V ramp waveforms were specifically designed and, during the measurements, thereafter, delivered to the tested cell through digital-to-analog conversion. In the current experimental scenario, we voltage-clamped the cell at −50 mV, and the upsloping (forward) limb from −50 to 0 mV followed by the downsloping (reverse) limb back to −50 mV with varying duration (0.4-3.2 s) was imposed over it. As demonstrated in Figure 3, the peak amplitude of I K(M) became progressively declined with increasing the V ramp 's duration (or decreasing the V ramp 's speed); however, as the V ramp 's slope decreased, the maximal strength of I K(M) triggered by the upsloping limb of triangular V ramp progressively increased. Moreover, it can be seen that under cell exposure to 3 µM QO-58, the current magnitude responding to both rising and falling V ramp was increased ( Figure 3A,B). For example, as the during of triangular V ramp was set at 3.2 s (i.e., slope = 31.25 mV/s), the application of 3 µM QO-58 increased the current amplitude measured from the upsloping or downsloping limb at the level of −10 mV from 63 ± 4 to 81 ± 5 pA (n = 8, p < 0.05) or from 91 ± 6 to 148 ± 11 pA (n = 8, p < 0.05), respectively. The experimental observations project that the strength of I K(M) in the upsloping limb was obviously increased as the duration of triangular V ramp increased, while that in the downsloping end progressively decreased, and that the presence of QO-58 contributed to an increase in I K(M) in a time-and state-dependent manner in GH 3 cells.
The voltage-dependent hysteresis (V hys ) of ionic currents have been growingly noticed to exert important impacts on electrical behaviors of action potential firing [29,31,53,54]. As illustrated in Figure 3A,B, the I K(M) amplitude triggered by the upsloping limb of upright triangular V ramp was evidently lower than that by the downsloping end, strongly reflecting that a V ramp -induced V hys behavior resides in I K(M) observed in GH 3 cells. As the duration of triangular V ramp increased from 0.4 to 3.2 s (i.e., the V ramp 's slope became decreased), the V hys degree was reduced. Of notice, by adding QO-58 (3 µM), I K(M) evoked during the upsloping limb of triangular V ramp arose to a less extent than that measured at the downsloping ramp. For example, in control period (i.e., absence of QO-58), I K(M) at the level of −15 mV during the upsloping and downsloping ends of triangular V ramp were 53 ± 6 pA (n = 8) and 91 ± 8 pA (n = 8), respectively, the values between which were noticed to differ significantly (p < 0.05). Moreover, by adding QO-58 (3 µM), the amplitudes of forward and backward I K(M) at the same level of voltage noticeably increased to 69 ± 7 pA (n = 8, p < 0.05) and 138 ± 11 pA (n = 8, p < 0.05), respectively. Therefore, the magnitude of QO-58mediated current stimulation at the upsloping (forward) and downsloping (backward) limbs of triangular V ramp differ significantly. The presence of 3 µM QO-58 increased I K(M) amplitude at −15 mV during the upsloping or downsloping limb of triangular V ramp by about 16% or 52%, respectively.
We further quantified the degree (i.e., V hys 's ∆area) of V ramp -induced V hys of I K(M) . The results demonstrated that the amount of V hys responding to 3.2 s triangular V ramp was considerably increased in the presence of QO-58. Figure 3C summarizes the data demonstrating the effects of QO-58 (1 or 3 µM) on the area encircling the forward and backward current trajectory of V ramp -evoked I K(M) . For example, in addition to its stimulation of I K(M) amplitude, cell exposure of 3 µM QO-58 resulted in an increase in the V hys strength responding to long-lasting triangular V ramp , as illustrated by a considerable increase in V hys 's ∆area arising from 508 ± 26 to 939 ± 41 mV·pA (n = 8, p < 0.05). Moreover, during the continued exposure to 3 µM QO-58, subsequent exposure to 10 µM linopirdine appreciably attenuated the ∆area to 612 ± 33 mV·pA (n = 8, p < 0.05). Linopirdine was reported to be a blocker of K M channels [18]. It is plausible to assume, therefore, that QO-58 is effective in stimulating I K(M) residing in GH 3 cells in a V hys -dependent manner.

Effect of QO-58 on M-Type K + Channel (K M ) Channels Measured from GH 3 Cells
The QO-58-stimulated whole-cell I K(M) detected above in these cells could arise from changes occurring in either channel open probability, single-channel amplitude, gating kinetics, or the number of K M channel. The reasons therefore enabled us to investigate the single-channel recordings of the channel with or without the presence of QO-58. In this set of cell-attached current recordings, we placed cells in high-K + , Ca 2+ -free solution, and the recording pipette was filled with low-K + (5.4 mM) solution. As demonstrated in Figure 4A, when the examined cell was maintained at +20 mV relative to the bath, the activity of single K M channel was robustly observed [31,55]. Of particular interest, as QO-58 was applied to the bath, the probabilities of K M -channel openings progressively increased. For example, the presence of 3 µM QO-58 significantly increased the channel open probability from 0.087 ± 0.021 to 0.238 ± 0.041 (n = 8, p < 0.05); however, there was devoid of changes in single-channel amplitude ( Figure 4B). Moreover, in continued presence of 3 µM QO-58, linopirdine (10 µM) resulted in an attenuation of QO-58-stimulated channel activity, as demonstrated in a significant reduction in channel activity to 0.164 ± 0.025 (n = 8, p < 0.05).

Effect of QO-58 on the Single-Channel Conductance and Activation Curve of K M Channels
We further examined if K M -channel activity measured at different levels of membrane potentials could be altered by the presence of QO-58. As demonstrated in Figure 4C, the single channel conductance of K M channels achieved with or without application of QO-58 did not significantly differ (27 ± 3 pS [in control] versus 28 ± 3 pS [in the presence of 3 µM QO-58]; n = 8, p > 0.05), despite the increased probability of K M -channel openings in its presence. A summary showing effects of QO-58 and QO-58 plus linopirdine on K M -channel activity in GH 3 cells is also presented in Figure 4D. The results led us to reflect that as GH 3 cells were continually exposed to QO-58 (3 µM), the channel open probability was significantly attenuated by subsequent addition of linopirdine (10 µM).

QO-58-Mediated Stimulation of Ca 2+ -Activated K + Currents (I K(Ca) ) by the Presence of QO-58
Several small molecules (e.g., BMS-204352, naringenin, and QO-40) that demonstrated to stimulate K M -channel activity have been reportedly noted to regulate other types of K + currents (e.g., I K(Ca) ) [9,47,48]. For these reasons, we next explored if QO-58 is able to modify the amplitude of I K(Ca) residing in GH 3 cells. In these experiments we voltageclamped the tested cell at a holding potential of 0 mV to prevent the interference by other type of ionic currents (i.e., voltage-gated Ca 2+ currents) [56,57]. As demonstrated in Figure 5A, one minute after cell exposure to 3 µM QO-58, I K(Ca) amplitudes measured at different levels of membrane potentials increased. Average I-V relationships of I K(Ca) with or without the QO-58 (3 µM) presence are illustrated in Figure 5B. The concentrationdependent stimulation by QO-58 of macroscopic I K(Ca) amplitude was established and is hence depicted in Figure 5C. According to the Hill equation described in Section 4, the EC 50 value required for QO-58-stimulated effect on I K(Ca) was calculated to be 4.2 µM.

Stimulatory Effect of QO-58 on the Activity of Large-Conductance Ca 2+ -Activated K + (BK Ca ) Channels Identified in GH 3 Cells
The QO-58-induced raise in whole-cell I K(Ca) described above could be mediated through either adjustment in channel open probability, single-channel amplitude, gating kinetics of the BK Ca channels, or in any combinations. Therefore, these reasons urged us to assess the single-channel activities of the channels functionally active in GH 3 cells. In this set of inside-out current recordings, we bathed the tested cells in high-K + solution containing 0.1 µM Ca 2+ , and the recording pipette was filled up with K + -containing solution. As demonstrated in Figure 6A, as the excised patch was voltage-clamped at +60 mV, the activity of BK Ca channels occurring in rapid and independent open-closed transitions was robustly detected. One minute after bath addition of QO-58, the channel opening probability was conceivably increased. For example, under inside-out configuration, QO-58 at a concentration of 1 or 3 µM applied to bath medium led to a respective increase in channel opening probability to 0.118 ± 0.005 (n = 8, p < 0.05) or 0.174 ± 0.006 (n = 8, p < 0.05) from a control value of 0.073 ± 0.004 (n = 8). However, single-channel amplitude of BK Ca channel was not noticed to differ significantly between the absence and presence of 3 µM QO-58 (12.8 ± 2 pA [in control] versus 12.9 ± 2 pA [in the presence of 3 µM QO-58]; n = 8, p > 0.05). Under the exposure to 3 µM QO-58, the slow component of mean closed time of the channel became considerably shortened to 19 ± 3 ms (n = 8, p < 0.05) from a control value of 32 ± 5 ms (n = 8). Of additional notice, as the excised patch was continually exposed to 3 µM QO-58, subsequent addition of verruculogen (1 µM) considerably decreased channel open probability to 0.074 ± 0.004 (n = 8, p < 0.05), although further application of linopirdine (10 µM) produced minimal effect on it (0.174 ± 0.006 [in the presence of 3 µM QO-58 alone] versus 0.173 ± 0.006 [in presence of QO-58 plus linopirdine]; n = 8, p > 0.05) ( Figure 6B). Verruculogen is a tremorgenic mycotoxin known to effectively suppress the activity of BK Ca channels [58,59]. Therefore, the experimental results strongly indicate that QO-58-activated channel activity is mainly through its activation of BK Ca channels, rather than that of K M channels. They also enable us to project that the activation is attributed primarily to the shortening of mean closed time of the channel, despite no change in single-channel amplitude in the presence of QO-58. hence depicted in Figure 5C. According to the Hill equation described in Section 4, the EC50 value required for QO-58-stimulated effect on IK(Ca) was calculated to be 4.2 μM.

Stimulatory Effect of QO-58 on the Activity of Large-Conductance Ca 2+ -Activated K + (BKCa) Channels Identified in GH3 Cells
The QO-58-induced raise in whole-cell IK(Ca) described above could be mediated through either adjustment in channel open probability, single-channel amplitude, gating kinetics of the BKCa channels, or in any combinations. Therefore, these reasons urged us to assess the single-channel activities of the channels functionally active in GH3 cells. In this set of inside-out current recordings, we bathed the tested cells in high-K + solution containing 0.1 μM Ca 2+ , and the recording pipette was filled up with K + -containing solution. As demonstrated in Figure 6A, as the excised patch was voltage-clamped at +60 mV, Figure 5. Stimulatory effect of QO-58 on the amplitude of whole-cell (i.e., macroscopic) Ca 2+ -activated K + current (I K(Ca) ) measured from GH 3 cells. In this series of voltage-clamp current recordings on these cells, we used normal Tyrode's solution containing 1.8 mM CaCl 2 as a bathing medium, and the recording pipette used was backfilled with K + -enriched solution. As whole-cell configuration was established, we evoked I K(Ca) from a holding potential of 0 mV to test potentials in the range of 0 and +70 mV (10- p > 0.05) ( Figure 6B). Verruculogen is a tremorgenic mycotoxin known to effectively suppress the activity of BKCa channels [58,59]. Therefore, the experimental results strongly indicate that QO-58-activated channel activity is mainly through its activation of BKCa channels, rather than that of KM channels. They also enable us to project that the activation is attributed primarily to the shortening of mean closed time of the channel, despite no change in single-channel amplitude in the presence of QO-58. We further explored how the presence of QO-58 alters BK Ca -channel activity at different levels of membrane potentials. As demonstrated in Figure 7A, the linear relationship of single-channel amplitude versus membrane potential (i.e., single-channel conductance) was collated under inside-out configuration. As the excised patch was exposed to 3 µM QO-58, the value of single-channel conductance obtained between the absence and presence of QO-58 did not differ significantly (213 ± 8 pS [in control] versus 215 ± 9 pS [in the presence of QO-58]; n = 8, p > 0.05). Additionally, the steady-state activation curve of BK Ca channels with or with the QO-58 application is illustrated in Figure 7B. As the smooth lines drawn by fitting the data to the Boltzmann equation stated in Section 4, the results demonstrated that during the exposure to QO-58, there was a leftward shift (approximately 14 mV) along the voltage axis in the activation curve of the channel with no appreciable modifications in gating charge. For example, in control, V 1/2 = +66 ± 8 mV and q = 4.9 ± 0.2 e (n = 8), while in the presence of 3 µM QO-58, V 1/2 = +52 ± 7 mV and q = 4.8 ± 0.2 e (n = 8). These results indicated that although neither single-channel conductance nor gating charge of the channel was changed, the steady-state activation curve of BK Ca channels measured under inside-out excised patch of GH 3 cells was shifted toward less depolarized potential, as the detached patch was exposed to QO-58. ferent levels of membrane potentials. As demonstrated in Figure 7A, the linear relationship of single-channel amplitude versus membrane potential (i.e., single-channel conductance) was collated under inside-out configuration. As the excised patch was exposed to 3 μM QO-58, the value of single-channel conductance obtained between the absence and presence of QO-58 did not differ significantly (213 ± 8 pS [in control] versus 215 ± 9 pS [in the presence of QO-58]; n = 8, p > 0.05). Additionally, the steady-state activation curve of BKCa channels with or with the QO-58 application is illustrated in Figure 7B. As the smooth lines drawn by fitting the data to the Boltzmann equation stated in Section 4, the results demonstrated that during the exposure to QO-58, there was a leftward shift (approximately 14 mV) along the voltage axis in the activation curve of the channel with no appreciable modifications in gating charge. For example, in control, V1/2 = +66 ± 8 mV and q = 4.9 ± 0.2 e (n = 8), while in the presence of 3 μM QO-58, V1/2 = +52 ± 7 mV and q = 4.8 ± 0.2 e (n = 8). These results indicated that although neither single-channel conductance nor gating charge of the channel was changed, the steady-state activation curve of BKCa channels measured under inside-out excised patch of GH3 cells was shifted toward less depolarized potential, as the detached patch was exposed to QO-58.

Minor Inhibitory Effect of QO-58 on Erg-Mediated K + Current (I K(erg) ) Seen in GH 3 Cells
In another set of experiments, we attempted to explore if the presence of QO-58 could have any influence on another type of whole-cell K + current (i.e., I K(erg) ). The measurements were conducted in these cells bathed in high-K + , Ca 2+ -free solution containing 1 µM TTX and 0.5 mM CdCl 2 , and the pipette that was used was filled up with K + -enriched solution. The tested cell was voltage-clamped at −10 mV, and a series of command voltage steps ranging between −100 and 0 mV was thereafter delivered to evoke deactivating I K(erg) [12,37,52,[60][61][62][63]. As illustrated in Figure 8, under cell exposure to 10 µM QO-58, average I-V relationship of I K(erg) became lessened in the voltages ranging between −100 and −50 mV. For example, at the level of −100 mV, the exposure to 10 µM QO-58 significantly reduced the peak amplitude of hyperpolarization-activated I K(erg) by 25 ± 2% from 1055 ± 139 to 791 ± 113 pA (n = 7, p < 0.05). After the compound's washout, current amplitude returned to 1012 ± 131 pA (n = 7, p < 0.05). Therefore, it is noticeable that unlike QO-58 effect on I K(M) or I K(Ca) , I K(erg) functionally active in GH 3 cells is susceptible to minor inhibition by the QO-58 presence.
mV. For example, at the level of −100 mV, the exposure to 10 μM QO-58 significantly reduced the peak amplitude of hyperpolarization-activated IK(erg) by 25 ± 2% from 1055 ± 139 to 791 ± 113 pA (n = 7, p < 0.05). After the compound's washout, current amplitude returned to 1012 ± 131 pA (n = 7, p < 0.05). Therefore, it is noticeable that unlike QO-58 effect on IK(M) or IK(Ca), IK(erg) functionally active in GH3 cells is susceptible to minor inhibition by the QO-58 presence. Peak or sustained IK(erg) obtained with or without the QO-58 addition was measured at the start or end-point of each hyperpolarizing step with a duration of 1 s. Of note, the presence of 10 μM QO-58 slightly inhibited IK(erg) in these cells.

Docking Results on Interaction between KCa1.1 Channel and QO-58 or between KCNQ2 and QO-58
In a final set of studies, we investigated how the protein of KCa1.1 (or α-subunit of BKCa channel) or KCNQ2 could be docked with QO-58 by exploiting PyRx software. The predicted binding sites of the QO-58 molecule were demonstrated in Figure 9. Of notice, with being docked to KCa1.1 with a binding energy of −7.1 kcal/mol, QO-58 can form Peak or sustained I K(erg) obtained with or without the QO-58 addition was measured at the start or end-point of each hyperpolarizing step with a duration of 1 s. Of note, the presence of 10 µM QO-58 slightly inhibited I K(erg) in these cells.

Docking Results on Interaction between K Ca 1.1 Channel and QO-58 or between KCNQ2 and QO-58
In a final set of studies, we investigated how the protein of K Ca 1.1 (or α-subunit of BK Ca channel) or KCNQ2 could be docked with QO-58 by exploiting PyRx software. The predicted binding sites of the QO-58 molecule were demonstrated in Figure 9. Of notice, with being docked to K Ca 1.1 with a binding energy of −7.1 kcal/mol, QO-58 can form hydrogen bond with residue Thr 277, while it is able to have hydrophobic contact with residue Thr 277, Val 302, Phe 305, Ala 306, Ala 309, and Gly 310. Alternatively, as being docked to KCNQ2 with a binding energy of −6.9 kcal/mol, QO-58 can form hydrophobic contact with residue Ile 381, Ser 382, Pro 383, Asn 384, Leu 385, and Leu 387. Of notice, these results thus prompted us to reflect that KCNQ2 and BK Ca channels are likely to share unique motifs or recognition sequences with which QO-58 or other structurally similar compounds can interact, and that QO-58 can bind to the cytoplasmic residues of KCNQ2 or K Ca 1.1 channel, which are adjacent to transmembrane segment of the channel. Therefore, care needs to be mentioned in attributing the actions of QO-58 or other structurally similar compounds exclusively to the stimulation of KCNQx-(K V 7-) channel activity as reported previously [4,8,18]. unique motifs or recognition sequences with which QO-58 or other structurally similar compounds can interact, and that QO-58 can bind to the cytoplasmic residues of KCNQ2 or KCa1.1 channel, which are adjacent to transmembrane segment of the channel. Therefore, care needs to be mentioned in attributing the actions of QO-58 or other structurally similar compounds exclusively to the stimulation of KCNQx-(KV7-) channel activity as reported previously [4,8,18].

Discussion
The noticeable conclusions are drawn from this study as follows. First, the presence of QO-58 concentration-dependently increased the amplitude of IK(M) and IK(Ca) with EC50 value of 3.1 and 4.2 μM, respectively. Second, under cell exposure to QO-58, the steadystate activation curve of IK(M) was shifted along voltage axis to a hyperpolarized potential with no change in the gating charge. Third, the Vhys strength of IK(M) activated by triangular Vramp measurably increased by the QO-58 presence. Fourth, cell exposure to QO-58 enhanced the probability of BKCa-channel openings as well as shifted the activation curve of the channel at steady state toward the less depolarized potential; however, neither the

Discussion
The noticeable conclusions are drawn from this study as follows. First, the presence of QO-58 concentration-dependently increased the amplitude of I K(M) and I K(Ca) with EC 50 value of 3.1 and 4.2 µM, respectively. Second, under cell exposure to QO-58, the steady-state activation curve of I K(M) was shifted along voltage axis to a hyperpolarized potential with no change in the gating charge. Third, the V hys strength of I K(M) activated by triangular V ramp measurably increased by the QO-58 presence. Fourth, cell exposure to QO-58 enhanced the probability of BK Ca -channel openings as well as shifted the activation curve of the channel at steady state toward the less depolarized potential; however, neither the gating charge nor single-channel conductance of the channel was affected during its exposure. Fifth, the deactivating I K(erg) activated by membrane hyperpolarization was slightly suppressed by adding QO-58. Taken together, the interaction of QO-58 with K M or BK Ca channels to stimulate I K(M) or I K(Ca) in excitable cells is expected to occur in a concentrationand voltage-dependent manner, assuming that similar in vivo findings occur.
In agreement with previous studies [2], the current observations demonstrated that with optimum EC 50 of 3.1 µM, QO-58 was capable of enhancing the magnitude of I K(M) seen in GH 3 cells. Furthermore, the V hys changes have been regarded to play an essential characteristic in electrical behaviors of different excitable cells. In the current study, in accordance with earlier studies [29,31,53,54], the I K(M) intrinsically residing in GH 3 cells was robustly observed to undergo V ramp -induced V hys , suggesting that the voltage sensitivity of gating charge movements is dependent on the previous state (or conformation) of the K M channel. In other words, as the membrane potential becomes depolarized (i.e., during initiation of action potential or the upsloping limb of the triangular V ramp ), the voltage dependence of I K(M) activation would switch to less depolarized voltages with a small current magnitude, thereby have the tendency to decrease cell excitability. However, as the membrane potential becomes negative (i.e., downward ramp of the double V ramp ), the voltage dependence of K M channels may shift the mode of V hys to one which occurs at more negative potentials, thereby leading to an increase in membrane repolarization. Furthermore, upon triangular V ramp with varying durations, the QO-58 addition noticeably increased the V hys 's strength for I K(M) elicitation. Under this scenario, we extended previous results and further provided the experimental observations, strongly indicating that there would be a perturbing stimulatory effect of QO-58 on such non-equilibrium property (i.e., V hys ) in K M (or K V 7) channels in electrically excitable cells. However, how QO-58induced changes in I K(M) 's V hys are linked to the behavior of these cells occurring in vivo remains unclear. Of importance, the main point raised is that the adjustments by QO-58 of I K(M) 's V hys residing in excitable cells are anticipated to be responsible for altering the bursting pattern of action potentials in excitable cells [13][14][15][16][17]19,20,30,60].
In the present observations, effective EC 50 value needed for QO-58-stimulated I K(Ca) present in GH 3 cells was yielded to be 4.2 µM, a value that is close to that (3.1 µM) for its activation of I K(M) . Under our inside-out current recordings, the addition of QO-58 to bath medium was able to increase the probability of BK Ca -channel openings with being void of change in single-channel conductance, suggesting that QO-58 may bind to a site located on the cytoplasmic leaflet of the α-subunit of BK Ca channels. The slow component of mean closed time of the channel decreased by adding this compound. Under its exposure, the steady-state activation curve of BK Ca channels seen in GH 3 cells became overly shifted to less depolarized potential with no appreciable change in the gating charge of the curve. The QO-58-stimulated BK Ca channel activity was also effectively counteracted by subsequent addition of verruculogen (1 µM), yet not by linopirdine (10 µM). Verruculogen is known to block BK Ca channels effectively [58,59], while linopirdine can suppress K M -channel activity [18,31,33]. Under such scenario, it is plausible to notice that apart from its effects on I K(M) , QO-58-stimulated I K(Ca) arises primarily through the observed activation of BK Ca channels, although the precise or detailed ionic mechanism of QO-58 actions on the activity and gating kinetics of BK Ca channels remains to be resolved.
According to previous pharmacokinetic studies, peak plasma concentration after the oral administration with QO58-lysine with a dose of 50, 20, 12.5 mg/kg has been reported to reach around 50 µg/mL (85 µM), 20 µg/mL (34 µM), or 4 µg/mL (6.8 µM), respectively [5,7]. Moreover, the sensitivity of voltage-clamped cells (e.g., neuroendocrine or endocrine cells) to QO-58 or other structurally similar compounds (e.g.,  can be expected to depend not only on the QO-58 concentration applied, but also greatly on different confounding variables which include the pre-existing level of resting potential and varying bursting patterns of action potential firing. It is therefore conceivable to reflect that QO-58-mediated concerted stimulation of K M (KCNQx or K V 7x) and BK Ca channels seen in GH 3 cells is of pharmacological or therapeutic relevance, presuming that similar in vivo observations are found [3,8].

Cell Preparations
The GH 3 pituitary cell line, which was originally established from a pituitary tumor carried in a 7-month-old female Wistar-Furth rat, was supplied by the Bioresource Collection and Research Center (BCRC-60015; Hsinchu, Taiwan). This cell line was derived from the American Type Culture Collection (ATCC ® [CCL-82.1 TM ]; Manassas, VA, USA). We maintained GH 3 cells in Ham's F-12 medium supplemented with 2.5% heat-inactivated fetal calf serum (v/v), 15% horse serum (v/v) and 2 mM L-glutamine [62,64]. Cells were grown in monolayer culture at 37 • C in a humidified environment of CO 2 /air (1:19). For sub-culturing made by trypsinization (0.025% trypsin solution [HyClone TM ] containing 0.01 sodium N,N-diethyldithiocarbamade and EDTA), we dissociated cells and then passaged them every 2-3 days. The measurements were undertaken when cell growth underwent 60-80% confluence (usually 5-6 days).

Electrophysiological Measurements
Shortly before the experiments, we carefully dissociated cells with a 1% trypsin/EDTA solution, and an aliquot of the suspension containing cell clumps was rapidly placed in a recording chamber adherently attached to the working stage of a DM-IL inverted microscope (Leica; Highrise Instrument, Taichung, Taiwan). The electrodes which were used to record were fabricated from Kimax-51 ® capillaries with 1.5-1.8 mm in diameter (Kimble ® 34500-99; Merck, Taipei, Taiwan) by using a PC-10 vertical puller (Narishige; Taiwan Instrument, Tainan, Taiwan), and their tips were then fire-polished with MF-83 microforge (Narishige). When the electrodes were filled up with different internal solutions described above, their resistance was measured to range between 3 and 5 MΩ, for the purpose of making good GΩ-seal formation. We performed patch clamp recordings in cell-attached, inside-out or whole-cell configuration by using either an RK-400 (Bio-Logic, Claix, France) or an Axopatch-200B amplifier (Molecular Devices; Bestgen Biotech, New Taipei City, Taiwan), as elaborated elsewhere [29,31,52,59]. Whole-cell current recordings were established by rupturing the patch of membrane isolated with GW sealing by the patch pipette, then bringing the cell interior into contact with the pipette interior.

Data Recordings and Analyses
The signals were monitored and stored online in a Sony VAIO CS series laptop computer (VGN-CS110E; Tainan, Taiwan), equipped with a low-noise 1440A digitizer (Molecular Devices). During the measurements with analog-to-digital and digital-to-analog conversion, the latter device was controlled by pCLAMP 10.6 software (Molecular Devices) run under Microsoft Windows 7 (Redmond, WA, USA).
To assess the percentage increase of QO-58 on I K(M) or I K(Ca) , we measured the amplitudes of I K(M) or I K(Ca) during cell exposure to different QO-58 concentrations (1-300 µM). The amplitude of I K(M) or I K(Ca) during cell exposure to QO-58 at a concentration of 300 µM was considered to be 100%, and the current amplitudes after application of different QO-58 concentrations were expressed relative to this value. The data sets with respect to concentration-dependent effect of QO-58 on the activation of I K(M) or I K(Ca) were satisfactorily fitted to the Hill equation with a nonlinear least-squares' algorithm. Thus: The steady-state activation curve (i.e., the relationship of the membrane potential versus the I K(M) conductance) acquired with or without exposure to QO-58 was satisfactorily approximated by a modified Boltzmann function of the following form: where G = the I K(M) conductance; G max = the maximal conductance of I K(M) ; V 1/2 = the voltage at which half-maximal activation of the current is achieved; q = the apparent gating charge; F = Faraday's constant; R = the universal gas constant; and T = the absolute temperature.
The sigmoidal relationship between the membrane potentials and relative open-state probability of BK Ca channels (i.e., the steady-state activation curve) with or without the QO-58 (3 µM) addition was collated and thereafter fitted by the Boltzmann equation using the goodness-of-fitness test; relative open probability = n {1 + exp[−(V − V 1/2 )qF/RT]} (1) where n = the maximal relative open probability; V = the membrane potential; V 1/2 = the potential for half-maximal activation; q = apparent gating charge; and F, R, and T are similarly stated above in the activation curve of I K(M) .

Curve-Fitting Approximations and Statistical Analyses
Linear (e.g., single-channel conductance) or nonlinear (e.g., Hill or Boltzmann equation and single exponential) curves fitted to different experimental data sets were made with chi-squared goodness-of-fit test using either the Solver add-in bundled with Excel ® 2021 (Microsoft, Redmond, WA, USA)) or OriginPro ® 2021 (OriginLab, Scientific Formosa, Kaohsiung, Taiwan). The values are provided as means ± error of the mean (SEM) with the sizes of experimental observations, which represent the cell number sampled. The Student's t-test (paired or unpaired) for two different group, or analyses of variance (ANOVA-1 or ANOVA-2) followed by post-hoc Fisher's least-significance difference among more than two different groups studied for multiple comparisons, were made for the statistical evaluation. Probability with p < 0.05 was considered statistically significant (as indicated with * or ** in the figures), unless noted otherwise.

Data Availability Statement:
The original data is available upon reasonable request to the corresponding author.

Acknowledgments:
The authors gratefully acknowledge to Meng-Cheng Yu for his assistance in cell preparations.

Conflicts of Interest:
The authors declare no conflict of interests that are directly relevant to the present study. The content and writing of this paper are solely the responsibility of the authors.