Characterization of Convergent Suppression by UCL-2077 (3-(Triphenylmethylaminomethyl)pyridine), Known to Inhibit Slow Afterhyperpolarization, of erg-Mediated Potassium Currents and Intermediate-Conductance Calcium-Activated Potassium Channels

Abstract

UCL-2077 (triphenylmethylaminomethyl)pyridine) was previously reported to suppress slow afterhyperpolarization in neurons. However, the information with respect to the effects of UCL-2077 on ionic currents is quite scarce. The addition of UCL-2077 decreased the amplitude of erg-mediated K⁺ current (I_K(erg)) together with an increased deactivation rate of the current in pituitary GH₃ cells. The IC₅₀ and K_D values of UCL-2077-induced inhibition of I_K(erg) were 4.7 and 5.1 μM, respectively. UCL-2077 (10 μM) distinctly shifted the midpoint in the activation curve of I_K(erg) to less hyperpolarizing potentials by 17 mV. Its presence decreased the degree of voltage hysteresis for I_K(erg) elicitation by long-lasting triangular ramp pulse. It also diminished the probability of the opening of intermediate-conductance Ca²⁺-activated K⁺ channels. In cell-attached current recordings, UCL-2077 raised the frequency of action currents. When KCNH2 mRNA was knocked down, a UCL-2077-mediated increase in AC firing was attenuated. Collectively, the actions elaborated herein conceivably contribute to the perturbating effects of this compound on electrical behaviors of excitable cells.

1. Introduction

UCL-2077 (triphenylmethylaminomethyl)pyridine), which was discovered as a small-molecule blocker, has been particularly reported to suppress the slow or late afterhyperpolarizations existing in different regions of either central neurons or brain slices [1,2,3,4,5,6,7,8]. Earlier work has also demonstrated the capability of this compound to perturb the activity of KCNQ channels [4]. However, another paper showed that it was unable to suppress the slow afterhyperpolarization recorded from thalamic midline neurons [9]. Whether this compound produces any modifications on different types of membranous ion channels thus remains unresolved.

The I_K(erg) encoded by three different subfamilies of the gene KCNH is known to give rise to the pore-forming α-subunit of erg-mediated K⁺ (i.e., K_erg or K_V11) channels [10], which is regarded to constitute the cloned counterpart of the rapidly activating delayed-rectifying K⁺ currents in heart cells. These currents inherently existing in neurons or in different types of electrically excitable cell, such as endocrine or neuroendocrine cells, can highly influence the maintenance of the resting potential as well as the increase in subthreshold excitability [11,12], thereby leading to marked changes in the discharge rate of spontaneous action potentials (APs) [13,14,15,16]. As a consequence, the resultant perturbations in the activity of such K⁺ channels have the propensity to change the stimulus–secretion coupling of these cells [13,17,18,19]. Previous work has also revealed that the activity of K_erg channels can be a regulator of tumor cell apoptosis and proliferation [20]. Whether UCL-2077 produces any modifications in this type of ion channel is largely not explored, although it has been demonstrated to suppress slow afterhyperpolarization [2,3,4,5,6,7,8].

The intermediate-conductance Ca²⁺-activated K⁺ (IK_Ca) channels (also known as K_Ca3.1, SK4, IKCa1 or KCNN4), which are encoded by the KCNN4 gene, have been increasingly studied. However, the research into these channels has been focused mostly in different non-excitable or neoplastic cells with regard to their potential roles in cell behaviors that include hormonal secretion, cell migration, cell proliferation, and regulation of Ca²⁺ influx, K⁺ efflux, or both [17,20,21,22,23]. Specifically, these channels possess single-channel conductance of 20–60 pS and their pharmacological profiles are distinct from those of large- or small-conductance Ca²⁺-activated K⁺ channels [22,24,25,26].

The modulators of functional IK_Ca channels have recently been demonstrated to perturb functional activities in different types of central neurons, such as membrane afterhyperpolarization [3,9,21,26,27,28,29]. Importantly, previous studies have reported the ability of UCL-2077 to disturb the slow afterhyperpolarization found in central neurons, suggesting a possible role in the regulation of Ca²⁺-dependent K⁺ conductance [1,5,9]. Whether the presence of UCL-2077 effects any substantial perturbation in the activity of Ca²⁺-activated K⁺ channels is therefore worthy of being thoroughly explored.

Therefore, in the current study, we wanted to assess whether the presence of UCL-2077 is capable of exerting any modifications on different types of membrane ion current. Findings from the present experimental observations highlight, for the first time, the notion that UCL-2077 can interact with the kinetic gating of K_erg channels to suppress the amplitude of I_K(erg). This compound was effective at suppressing the activity of IK_Ca channels, although minimal regulatory action on the open probability of large-conductance Ca²⁺-activated K⁺ (BK_Ca) channels was seen. It is thus important to mention that the inhibition of both I_K(erg) and IK_Ca-channel activity caused by UCL-2077 may potentially converge to act on the functional activities of neurons, and neuroendocrine or endocrine cells (e.g., afterhyperpolarization following repetitive firing) [8,9].

2. Results

2.1. Effect of UCL-2077 on I_K(erg) in Pituitary GH₃ Cells

The whole-cell voltage-clamp experiments were initially applied to assess the effect of UCL-2077 on ionic currents in these cells. To record I_K(erg) and avoid the contamination of Ca²⁺-activated K⁺ currents, we suspended cells in high-K⁺, Ca²⁺-free solution, and the pipette was filled with K⁺-containing solution. When the cell was maintained at −10 mV and 1-sec hyperpolarizing voltage, pulses ranging between −10 and −90 mV were applied. As cells were exposed to different concentrations of UCL-2077, the I_K(erg) amplitude was progressively decreased and the deactivation time course of the current concomitantly became accelerated (Figure 1A). For example, upon hyperpolarization to −90 mV from a holding potential of −10 mV, UCL-2077 at a concentration of 10 μM produced a substantial reduction in the peak amplitude of I_K(erg) from 276 ± 13 to 109 ± 8 pA (n = 11, p < 0.05). After the washout of this compound, the current amplitude returned to 246 ± 11 (n = 9). The addition of either E-4031 (10 μM) or azimilide (10 μM) suppressed the amplitude of I_K(erg), while chlorotoxin (1 μM) failed to cause any effect on it. The inhibitory effect of azimilide on I_K(erg) is illustrated in Figure 1B. Moreover, the azimilide- or E-4031-sensitive component of I_K(erg) during long-lasting membrane hyperpolarization was depicted in Figure 1C. However, neither the further addition of diazoxide (30 μM) nor cilostazol (10 μM) was able to reverse UCL-2077-mediated inhibition of this current. Diazoxide or cilostazol are activators of ATP-sensitive K⁺ or BK_Ca channels [30], respectively, while chlorotoxin is a blocker of Cl^- channels.

We further determined the relationship between the UCL-2077 concentration and the percentage inhibition of I_K(erg). According to Equation (1), by virtue of a nonlinear least-squares fit to the results, the IC₅₀ value required for the inhibitory effect of this compound on I_K(erg) in GH₃ cells was calculated to be 4.7 μM, and, at a concentration of 100 μM, it almost completely suppressed current amplitude (Figure 1D). These experimental results reflect that UCL-2077 exerts a depressant action on hyperpolarization-evoked I_K(erg) in GH₃ cells.

During cell exposure to UCL-2077, in addition to the depressed amplitude of I_K(erg), the deactivation time course of I_K(erg) elicited by large membrane hyperpolarization tended to be fastened. Therefore, in attempts to obtain more quantitative evidence for UCL-2077-induced blocking of I_K(erg), the time constants for I_K(erg) deactivation taken with or without addition of different concentrations of this compound were measured by appropriately fitting each deactivation trajectory of hyperpolarization-elicited current to a single exponential function. Based on the reaction scheme (Equations (2) and (3)), the relationship of 1/τ_deact versus [B] (i.e., UCL-2077 concentration) became linear (Figure 1E); hence, the blocking (k₊₁^*) and unblocking (k₋₁) rate constants were estimated to be 1.345 sec⁻¹μM⁻¹ and 6.894 sec⁻¹, respectively, which yielded a value of 5.1 μM for the dissociation constant (K_D = k₋₁/k₊₁^*). Notably, this value is nearly close to the effective IC₅₀ elaborated above.

We further examined the effect of UCL-2077 on the I-V relationship of I_K(erg) in GH₃ cells. Figure 2A illustrates original records obtained after applying 1-sec hyperpolarizing steps from a holding potential of −10 mV to membrane potentials between −100 and −20 mV in 10-mV steps taken with or without the addition of this compound. The magnitude of UCL-2077-mediated inhibitory effect on I_K(erg) measured at the beginning of hyperpolarizing pulses notably became higher than that at the end of voltage steps (Figure 2B). The presence of UCL-2077 (30 μM) led to a resultant decrease in the slope of linear fit of the peak I_K(erg) amplitudes to voltages between −100 and −60 mV from 4.88 ± 0.12 to 1.26 ± 0.06 nS (n = 8, p < 0.05). These results indicate that I_K(erg) remained functionally active in GH₃ cells [15,18], and that its blocking effect was mainly exerted on the I_K(erg), which is responsible for inward rectification.

2.2. Effect of UCL-2077 on the Gating Charge of I_K(erg) Activation and Deactivation

Based on the limiting slope method described in Equation (4), we further performed the estimation of the gating charge (z_a) involved in I_K(erg) activation. Under our experimental conditions, we applied this equation to create a lower bound to the total gating charge by approaching it in the limit of the negative potential applied. As illustrated in Figure 2C, the slope of the linear region ranging between −100 and −50 mV without or with the application of 30 μM UCL-2077 was derived and then estimated to be −0.018 or −0.016, respectively. As we applied those values to Equation (4), the z_a value taken in the control or during cell exposure to UCL-2077 was then computed to be 0.46 ± 0.02 or 0.40 ± 0.02 e (n = 8), respectively. Thus, it is clear that the z_a value used for I_K(erg) activation in the presence of UCL-2077 became significantly smaller (p < 0.05).

The presence of UCL-2077 tends to speed up the deactivation rate of I_K(erg) in response to long-lasting step command hyperpolarization. For this reason, we also further analyzed the effect of this compound on the z_deact value from GH₃ cells. According to Equation (5), we estimated the z_deact values obtained with or without the addition of UCL-2077. As depicted in Figure 2D, we further constructed the relationships of ln(λ) versus membrane potential and the slopes of linear region between −100 and −60 mV in the control (i.e., UCL-2077 was not present) and during exposure to 30 μM UCL-2077 were thereafter noted to be −0.051 and −0.038, respectively. When we applied those values to Equation (5), the z_deact value in the control or during cell exposure to UCL-2077 was calculated to be 1.27 ± 0.03 or 0.94 ± 0.02 e (n = 9), respectively. During the exposure to UCL-2077, the z_deact value involved in I_K(erg) deactivation became significantly smaller (p < 0.05). The magnitude of decrease in z_deact (0.35 e) caused by UCL-2077 was much higher than that in z_a (0.06 e), reflecting that the action of UCL-2077 on I_K(erg) tended to be preferentially exerted on the deactivation process of K_erg channels.

2.3. Steady-State Activation of I_K(erg) Altered by the Presence of UCL-2077

To further characterize the suppressive effect of UCL-2077 on I_K(erg), we next tested the voltage dependence of the effect of UCL-2077 on I_K(erg) in GH₃ cells. Figure 3 shows the steady-state activation curve of I_K(erg) in the absence or presence of UCL-2077 (10 μM). In these experiments conducted with a two-step command voltage protocol, a 1-sec conditioning pulse to different voltages was delivered to precede the test pulse to −100 mV from a holding potential of −10 mV. We then constructed the relationship between the conditioning potentials and the normalized amplitudes of I_K(erg) and then fitted the data set with a Boltzmann function (see equation (6)). In the control, the voltage for half-maximal activation (V_1/2) or the corresponding slope factor (k) was −72.1 ± 1.9 or 8.6 ± 0.9 mV(n = 9), respectively; in the presence of UCL-2077 (10 μM), V_1/2 or k was −55.3 ± 1.8 or 8.9 ± 0.9 mV(n = 9), respectively; and in the presence of azimilide (10 μM), V_1/2 or k was −55.1 ± 1.7 or 8.7 ± 0.9 mV(n = 9), respectively. It thus became notable that the addition of UCL-2077 (10 μM) distinctly shifted the midpoint of the activation curve along the voltage axis toward the depolarizing voltage by roughly 17 mV (p < 0.05) with no substantial change in the slope factor (i.e., k value) of the current. Strikingly, under our experimental conditions, the data showed that, in addition to the decreased I_K(erg) amplitude, there was a notable voltage dependence of the steady-state activation curve of this current in its presence.

2.4. Effect of UCL-2077 on the Voltage Hysteresis Elicited in Response to Triangular Ramp Pulse

The voltage hysteresis of ionic currents has been demonstrated to have an impact on electrical behaviors of AP firing [31,32]. We thus explored whether there is possible voltage hysteresis existing in I_K(erg) measured from GH₃ cells. In this set of experiments, a long-lasting triangular ramp pulse with a duration of 2 sec (i.e., ±0.15 V/sec) was delivered as the whole-cell configuration was firmly achieved. Interestingly, as illustrated in Figure 4A, the trajectories of I_K(erg) elicited by the upsloping (i.e., depolarized from −150 to 0 mV) and downsloping (hyperpolarized from 0 to −150 mV) ramp pulse as a function of time were distinguishable between them. The current amplitude elicited by the upsloping limb of the triangular voltage ramp was less than that by the downsloping limb, strongly indicating that there was a voltage hysteresis for this current, as depicted in Figure 4B (i.e., the relationship of current amplitude versus membrane potential). As the ramp speed was reduced, the hysteresis degree for I_K(erg) became progressively increased. Notably, as the examined cell was exposed to UCL-2077 (10 μM), the I_K(erg) amplitude evoked in the upsloping limb of long-lasting triangular ramp was found to be significantly decreased to a lesser extent than that measured from the downsloping limb. For example, in the controls, the I_K(erg) amplitudes at −120 mV elicited upon the upward and downward end of triangular ramp pulse were 35.6 ± 2.4 and 97.6 ± 4.2 pA (n = 7), respectively, and the values were found to differ significantly between them (p < 0.05). However, during exposure to UCL-2077 (10 μM), the amplitudes of forward and backward I_K(erg) measured at the same level of membrane potential were substantially reduced to 19.2 ± 1.9 and 39.5 ± 2.3 pA (n = 7, p < 0.05), respectively.

We next quantified the degree of voltage hysteresis on the basis of the difference in areas under the curve in the forward (upsloping) and reverse (downsloping) direction as indicated by the arrows in Figure 4B. It was seen that for I_K(erg) in GH₃ cells, the degree of voltage hysteresis increased with slower ramp speed, and that the presence of UCL-2077 led to a conceivable reduction in the amount of such hysteresis. Figure 4C illustrates a summary of the data showing the effect of UCL at different concentrations on the areas under the curve between the forward and backward current traces. For example, the addition of UCL-2077 (10 μM) substantially decreased the area by about 50%, elicited by such a long-lasting triangular voltage ramp.

2.5. Inhibitory Effect of UCL-2077 on Delayed-Rectifier K⁺ Current (I_K(DR))

We further examined another type of voltage-gated K⁺ current, namely I_K(DR), inherently present in GH₃ cells with or without the addition of UCL-2077. To measure I_K(DR), we bathed cells in Ca²⁺-free Tyrode’s solution containing 1 μM tetrodotoxin (TTX), and the recording pipette was filled with K⁺-containing solution. As these cells were exposed to UCL-2077 at different concentrations, the amplitude of I_K(DR) evoked by 1-sec step depolarization from −50 to +50 mV was slightly but significantly suppressed (Figure 5A). The addition of UCL-2077 at a concentration of 1 μM was not found to have any effects on I_K(DR). However, as the membrane potential was depolarized from −50 to +50 mV, this compound (30 μM) significantly decreased the peak amplitude of I_K(DR) from 2117 ± 181 to 1741 ± 119 pA (n = 8, p < 0.05). Concomitant with this finding, the inactivation time constant of I_K(DR) in response to membrane depolarization was substantially shortened to 36.3 ± 1.3 msec (n = 7, p < 0.05) from a control value of 21.7 ± 1.1 msec (n = 7), when cells were exposed to 30 μM UCL-2077. We next assessed the relationship between the UCL-2077 concentration and the percentage inhibition of I_K(DR). According to Equation (1), by virtue of a nonlinear least-squares fit to the results, the IC₅₀ value required for the inhibitory effect of this compound on I_K(erg) in GH₃ cells was calculated to be 13.4 μM (Figure 5B). By comparison to its effect on I_K(erg), the presence of UCL-2077 became less effective in suppressing I_K(DR).

2.6. Failure of UCL-2077 to Alter the Activity of BK_Ca Channels

The activity of BK_Ca channels (maxi-K⁺ channels, K_Ca1.1, KCNMA1, Slo1) functionally expressed in excitable cells has been reported to be intimately linked to the afterhyperpolarization of neurons [3,9,33]. As such, we intended to examine whether the presence of UCL-2077 perturbed the activity of BK_Ca channels in GH₃ cells. In these experiments, we conducted inside-out current recordings with symmetrical K⁺ solution (145 mM). The bath medium contained 0.1 μM Ca²⁺, and the potential was set at +60 mV. As shown in Figure 6, when we applied this compound at a concentration of 10 μM to the bath (i.e., cytoplasmic side of the excised patches), no measurable change in the probability of BK_Ca-channel openings maintained at +60 mV (0.031 ± 0.005 (in the control) versus 0.032 ± 0.005 (in the presence of UCL-2077), n = 8, p > 0.05) was found. The single-channel conductance of BK_Ca channels did not differ significantly (168 ± 8 pS [in the control] versus 168 ± 7 pS [in the presence of 10 μM UCL-2077]; n = 8, p > 0.05). However, the further application of PF1022A (10 μM), still in continued presence of UCL-2077, was effective in suppressing the activity of BK_Ca channels without any changes in single-channel amplitude, as evidenced by a significant decrease in channel opening probability to 0.011 ± 0.002 (n = 8, p < 0.05). PF1022A, a nematocidal agent, has been reported to suppress the activity of BK_Ca channels [34]. However, it needs to be noted that a different substance (i.e., cyclooctadepsipeptide emodepside) was reported to activate the Slo-1 channel [34]. The reason for this difference in stimulatory or inhibitory actions on BK_Ca channels is currently unknown; therefore, it remains to be further studied. Nevertheless, unlike effects of UCL-2077 on I_K(erg) or I_K(DR), the data showed that there was minimal effect on the probability of BK_Ca-channel openings observed in GH₃ cells.

2.7. Suppressive Effect of UCL-2077 on the Activity of IK_Ca Channels

Earlier reports have demonstrated that the magnitude of IK_Ca-channel activity could play essential roles in the occurrence of slow afterhyperpolarization appearing in endocrine cells, CA1 pyramidal neurons and enteric neurons [3,5,21,26,27]. Therefore, the possible effect of UCL-2077 on IK_Ca channels in GH₃ cells was further explored. In these experiments, we immersed cells in normal Tyrode’s solution which contained 1.8 mM CaCl₂, and we held the examined cell at the potential of 0 mV relative to the bath. In keeping with previous observations [21], the IK_Ca-channel activity, which displayed rapid open-close transitions, was robustly detected in GH₃ cells. However, the addition of UCL-2077 was able to decrease the probability of IK_Ca-channel openings with no substantial change in unitary amplitude of the channel (Figure 7A). For example, the presence of this compound (10 μM) significantly decreased the channel opening probability from 0.018 ± 0.002 to 0.005 ± 0.001 (n = 9, p < 0.05). No change in the single-channel amplitude of IK_Ca channels between the absence and presence of 10 μM UCL-2077 was demonstrated (3.87 ± 0.03 pA (control) versus 3.85 ± 0.03 pA (in the presence of UCL-2077); n = 9, p > 0.05). In the continued presence of UCL-2077 (30 μM), further addition of 9-phenanthrol (10 μM), an activator of IK_Ca channels [35], significantly reversed the UCL-2077-mediated suppression of IK_Ca channels in these cells (Figure 7B). Therefore, in addition to its perturbation on the amplitude and gating of I_K(erg), UCL-2077 was capable of diminishing the probability that IK_Ca channel would be open.

2.8. Effect of UCL-2077 on the Frequency of Spontaneous Action Currents (ACs)

In another set of recordings, we investigated whether UCL-2077 modulated the frequency of spontaneous ACs measured from GH₃ cells. In these cell-attached voltage-clamp recordings, we bathed cells in normal Tyrode’s solution containing 1.8 mM CaCl₂, and we maintained the potential at the resting potential of the cell. As depicted in Figure 8, the addition of UCL-2077 increased the AC frequency significantly; however, it did not alter the AC amplitude significantly. For example, as cells were exposed to UCL-2077 (10 μM), the firing frequency of ACs seen in GH₃ cells was significantly raised to 0.71 ± 0.004 Hz (n = 9, p < 0.05) from a control value of 0.16 ± 0.002 Hz (n = 9).

We then also examined whether UCL-2077 had any effects on AC firing in GH₃ cells directed against KCNH2. When we knocked this down in these cells, the frequency of ACs recorded from cell-attached current recordings was not changed in the presence of 10 μM UCL-2077 (0.71 ± 0.004 Hz (in the control) versus 0.69 ± 0.004 Hz (in the presence of 10 μM UCL-2077), n = 7, p > 0.05). In contrast, in cells transfected with a scrambled negative control of siRNA, the presence of 10 μM UCL-2077 increased AC frequency from 0.21 ± 0.002 Hz to 0.72 ± 0.004 Hz (n = 7, p < 0.05). Therefore, changes in AC frequency caused by this compound are explained by its modifications on K_erg channels.

3. Discussion

The present results showing that UCL-2077 tended to accelerate I_K(erg) deactivation suggest that the UCL-2077 molecule might reach the blocking site only when the channel remains functionally active in the open state. UCL-2077 likely has a higher affinity toward the open-deactivated K_erg channels than toward the closed (or resting) channels in GH₃ cells. Another pertinent finding in this study is that, based on the estimation of changes in the z_a and z_deact values required for I_K(erg) elicitation (Equations (4) And (5)), the addition of UCL-2077 significantly allowed those values to be smaller and the magnitude of its decrease in z_deact was greater than that in z_a. Our data suggest that, in addition to its inhibition of KCNQ channels [4], the UCL-2077 molecule may interact largely with the deactivation process to alter both the magnitude and kinetics of I_K(erg).

In our observations, the concentration-dependent inhibition of I_K(erg) with effective IC₅₀ of 4.7 μM was depicted in the presence of UCL-2077. This value is nearly close to the K_D value (5.1 μM), taken according to the first-order reaction scheme (Equations (2) and (3)). It was also noted that the values of IC₅₀ or K_D appear to be higher than used for its suppression of KCNQ1- or KCNQ2-encoded currents as reported previously [4]. However, the presence of this compound was found to distinctly shift the steady-state activation curve of I_K(erg) along the voltage axis to a less hyperpolarized potential, as well as reducing the magnitude of voltage hysteresis for I_K(erg) in response to a long-lasting triangular ramp pulse. Under cell-attached current recordings, it substantially increased the AC frequency in a concentration-dependent manner. As such, K_erg channels functionally expressed in electrically excitable cells are likely to be a viable target for its action, regardless of the detailed ionic mechanisms.

In our study, the probability of BK_Ca-channel openings in GH₃ cells was barely affected by the presence of UCL-2077. However, the magnitude of IK_Ca-channel openness was clearly subject to reduction by this compound, despite the failure to reduce the unitary amplitude. Moreover, the UCL-2077-mediated decrease in IK_Ca-channel activity was notably counteracted by the subsequent application of 9-phenanthrol, an activator of IK_Ca channels [35]. Previous observations have shown that the magnitude of IK_Ca-channel activity could have significant impacts on the slow afterhyperpolarization of neurons [21,26,27,29]. Therefore, UCL-2077-mediated actions on the electrical behaviors of neurons or endocrine cells may partly ascribe from its reduction in the probability of IK_Ca channels that would be open [3].

Voltage-hysteresis has been proposed to play a role in influencing the electrical behavior of excitable cells. In this study, consistent with previous observations in HCN channels [31,32], K_erg channels inherently existing in GH₃ cells were found to undergo either a hysteresis in their voltage-dependence, or a mode-shift, in which the voltage sensitivity of gating charge movements depends on the previous state. In the current study, we also explored the perturbating effect of UCL-2077 on such non-equilibrium property inherently in K_erg channels. Our results clearly show that the presence of this compound was able to decrease the z_a and z_deact values required for the elicitation of I_K(erg) by membrane hyperpolarization. Therefore, the detailed studies regarding whether UCL-2077 can act on a deactivated process during the elicitation of K_V11-encoded channels and modify the heart function remain to be further delineated. Although the underlying mechanism of neuronal slow afterhyperpolarization is still unclear, previous studies have demonstrated the ability of UCL-2077 to modify slow after-hyperpolarization (AHP) [9,26] could be explained by its suppression of I_K(erg) and IK_Ca-channel activity.

4. Materials and Methods

4.1. Drugs and Solutions

UCL-2077 (3-(triphenylmethylaminomethyl)pyridine; N-trityl-3-pyridinemethanamine; C₂₅H₂₂N₂; https://pubchem.ncbi.nlm.nih.gov/compound/24868317), cilostazol and diazoxide were acquired from Tocris (Union Biomed Inc., Taipei, Taiwan), and E-4031 was from Enzo (Farmingdale, NY, USA), while PF1022A was from MedChem (Princeton, NJ, USA), and 9-phenanthrol and tetrodotoxin (TTX) were from Sigma Chemical Co. (Merck Ltd., Taipei, Taiwan). Chlorotoxin was kindly acquired from Dr. Woei-Jer Chuang (Department of Biochemistry, National Cheng Kung University Medical College, Tainan, Taiwan), while azimilide was a gift from Procter and Gamble Pharmaceuticals (Cincinnati, OH, USA). Unless specified otherwise, culture media, penicillin-streptomycin and trypsin/EDTA were obtained from HyClone^TM (Thermo Fisher Scientific, Logan, UT, USA), while all other chemicals were of laboratory grade and were taken from standard sources. We used twice-distilled water that had been deionized through a Millipore-Q system (Millipore, Bedford, MA, USA) in all experiments.

The composition of the bath solution (i.e., standard HEPES-buffered normal Tyrode’s solution) in this study was 136.5 mM NaCl, 5.4 mM KCl, 5.4 mM KCl, 1.8 mM CaCl₂, 0.53 mM MgCl₂, 5.5 mM glucose, and 5.4 mM HEPES titrated to pH 7.4 with NaOH. To record I_K(DR), we backfilled the recording electrodes with a solution consisting of 130 mM K-aspartate, 20 mM KCl, 1 mM KH₂PO₄, 1 mM MgCl₂, 3 mM Na₂ATP, 0.1 mM Na₂GTP, 0.1 mM EGTA, and 5 mM HEPES titrated to pH 7.2 with KOH; and, to measure I_K(erg), we replaced the bath solution with a high-K⁺, Ca²⁺-free solution consisting of 130 mM KCl, 10 mM NaCl, 3 mM MgCl₂, 6 mM glucose, and 10 mM HEPES titrated to pH 7.4 with KOH. For the recording of BK_Ca-channel activity, we used high-K⁺ bathing solution containing 145 mM KCl, 0.53 mM MgCl₂, and 5 mM HEPES titrated to pH 7.4 with KOH, and we used the internal solution containing 145 mM KCl, 2 mM MgCl₂ and 5 mM HEPES adjusted to pH 7.2 with KOH, while the value of free Ca²⁺ concentration was calculated assuming a dissociation constant for EGTA and Ca²⁺ (at pH 7.2) of 0.1 μM. For example, to provide 0.1 μM Ca²⁺ in the bath solution, 0.5 mM CaCl₂ and 1 mM EGTA were added.

4.2. Cell Preparations

Pituitary GH₃ cells, acquired from the Biorescources Collection and Research Center (BCRC-60015; Hsinchu, Taiwan), were routinely grown in Ham’s F12 medium, which was supplemented with 15% horse serum, 2.5% fetal calf serum, and 2 mM L-glutamine in a humidified environment of 5% CO₂/95% air [18]. We commonly changed culture medium every 2–3 days for the removal of non-adhering cells, and cells would undergo passage as they reached confluence. To observe growth, we commonly used a Nikon Eclipse Ti-E inverted microscope (Li Trading Co., Taipei, Taiwan) which was equipped with a 5-megapixel cooled digital camera.

4.3. Transfection with siRNAs

Both Custom Stealth RNAi for rat KCNH2 (sense: 5′-GACCUGCUUACUGCCCUCUACUUCA-3′, antisense: 5′-UGAAGUAGAGGGCAGUAAGCCAGGUC-3′) and a scrambled negative control of siRNA, which does not interfere with any known siRNAs, were acquired from Invitrogen (Carlsbad, CA, USA). I_K(erg) is encoded by KCNH2, which constitutes the pore-forming α-subunit of K_erg (or K_V11.1) channels. In accordance with the protocol provided by the manufacturer, we transfected GH₃ cells with 25 nM of siRNAs using a transfection reagent (Lipofectamine RNAiMAX; Invitrogen). We used small concentrations of siRNAs to prevent both off-target effects and any nonspecific responses to them. Gene-specific siRNA oligomers were diluted in reduced serum medium (Opti-MEM I; Invitrogen) and were subsequently mixed with a transfection reagent. As the complexes were incubated at room temperature for 20 min, they were thereafter added to the cells for another 24–48 h. We used electrophysiological recordings to evaluate the siRNA knockdown efficiency.

4.4. Electrophysiological Measurements

Shortly before the recordings were prepared, GH₃ or NSC-34 cells were harvested with 1% trypsin/EDTA solution and an aliquot of cell suspension was thereafter transferred to a home-made recording chamber which was firmly mounted on the stage of an inverted DM-IL microscope (Leica Microsystems, Wetzlar, Germany). The examined cells were suspended at room temperature (20–25 °C) in HEPES-buffered normal Tyrode’s solution, the solution of which is elaborated above. To fabricate the patch electrodes from Kimax-51 capillaries (#34500; Kimble, Vineland, NJ, USA), we used either a PP-83 vertical puller (Narishige, Tokyo, Japan) or a P-97 Flaming/Brown programmable puller (Sutter, Novato, CA, USA), and we then fire-polished the tips with an MF-83 microforge (Narishige). During the measurements, the electrodes, which bore resistances between 3 and 5 MΩ when filled with different internal solutions, were maneuvered by the use of an MX-4 manipulator (Narishige) and precisely delicately operated by an MHW-3 hydraulic micromanipulator (Narishige). We performed patch-clamp recordings in whole-cell, cell-attached, or inside-out configuration by using either an RK-400 (Bio-Logic, Claix, France) or Axopatch-200B (Molecular Devices, Sunnyvale, CA, USA) amplifier [13,18]. Liquid junction potentials, which commonly develop at electrode tip when the composition of the internal solution was distinct from that in the bath, were nulled shortly before the seal formation was made, and whole-cell data were corrected by such junction potentials. The tested compounds were applied through perfusion or were added to the bath to achieve the final concentration indicated.

Action currents (ACs), which represent action potentials (APs), were achieved in the cell-attached current recordings [15,36,37]. For AC measurements, the potential was held at the level of the resting potential (~−70 mV). AC measurements were made to enable the quantification of the underlying firing frequency with no significant changes in intracellular milieu. We also assessed the frequency and amplitude of ACs recorded from GH₃ or NSC-34 cells by using Mini Analysis Program (Synaptosoft, Leonia, NJ, USA).

4.5. Data Recordings and Analyses

The data, including voltage and current tracings, were simultaneously monitored and thereafter digitally stored in a TravelMate-6253 laptop computer (Acer, Taipei, Taiwan) at 10 kHz through a Digidata-1322A acquisition system (Molecular Devices). The device was equipped with an Adaptec Slim SCSI card (Milpitas, CA, USA) through the PCMCIA slot and, during the recordings, it was controlled by pCLAMP 9.2 (Molecular Devices). The current signals collected were low-pass filtered at 1 or 3 kHz to minimize electrical noise. Through digital-to-analog conversion, the pCLAMP-generated voltage profiles containing different types of rectangular or ramp waveform were particularly designed to assess the current-voltage (I-V) relationships, steady-state activation curve, and voltage hysteresis of ionic currents specified. If necessary for further analyses, the digitized signals were subsequently exported, either to 64-bit OriginPro 2016 (OriginLab, Northampton, MA, USA) or to the spreadsheets of Microsoft Excel (Redmond, WA, USA).

To calculate the percentage decrease in the peak amplitude of I_K(erg) obtained with or without the addition of UCL-2077, each cell was held at −10 mV and the hyperpolarizing potential to −90 mV with a duration of 1 sec was applied, while to determine the percentage decrease in I_K(DR) amplitude, each cell was held at −50 mV and the 1-sec depolarizing potential to +50 mV was applied. The peak amplitudes of hyperpolarization-induced I_K(erg) measured during cell exposure to different concentrations (0.3–100 μM) of UCL-2077 were compared with those measured after subsequent addition of E-4031 (10 μM). To evaluate the concentration-dependent relation of UCL-2077 on I_K(erg) or I_K(DR) amplitude, we fitted the data sets to the modified form of the Hill equation by using nonlinear regression analysis. That is:

$$\%\mathrm{decrease}=\left(E_{\max }\times {\left[\mathrm{UCL}\right]}^{{\mathrm{n}}_{\mathrm{H}}}\right)/\left({\left[\mathrm{UCL}\right]}^{{\mathrm{n}}_{\mathrm{H}}}+{\mathrm{IC}}_{50}^{{\mathrm{n}}_{\mathrm{H}}}\right)$$

(1)

where [UCL] is the concentration of UCL-2077 applied; IC₅₀ and n_H are the concentration needed to have a 50% inhibition and the Hill coefficient, respectively; and E_max is the UCL-2077-induced maximal reduction in the peak amplitude of I_K(erg) or I_K(DR) during hyperpolarizing or depolarizing steps.

The UCL-2077-mediated inhibition of I_K(erg) observed in GH₃ cells can be predicted by a state-dependent blocking mechanism, in which the UCL-2077 molecule preferentially binds to the opening state of the channel on the basis of minimal first-order scheme [15]:

$$\mathrm{C}\begin{array}{c}\alpha \\ \rightleftarrows \\ \beta \end{array} O \begin{array}{c}{k}_{+1}^{\ast }.\left[B\right]\\ \rightleftarrows \\ k_{-1}\end{array} O·B$$

(2)

where α or β denotes voltage-dependent rate constants for the opening or closing of K_erg channels; k₊₁^* or k₋₁ is blocking (forward) or unblocking (reverse) rate constants; and [B] is the UCL-2077 concentration. C, O or O·B is the closed, open or open-blocked states, respectively.

The blocking and unblocking rate constants were determined from the deactivation time constants (τ_deact) of I_K(erg) during hyperpolarized steps, as the cells were exposed to different concentrations of UCL-2077. The deactivation time constants (τ_deact) of I_K(erg) were calculated by fitting each current-trace trajectory to a single-exponential function. The apparent rate constants for binding (k₊₁^*) and unbinding (k₋₁) were achieved from the following equation:

$$\frac{1}{{\tau }_{deact}}={k_{+1}}^{\ast }\times [B]+k_{-1}$$

(3)

Specifically, the values of k₊₁^* and k₋₁ are estimated from the fit of 1/τ_deact versus different UCL-2077 concentration. Based on k₊₁^* and k₋₁, we were able to yield the value of dissociation constant (K_D = k₋₁/k₊₁^*).

To determine the number of equivalent gating charges (z_a) involved in I_K(erg) activation obtained with or without UCL-2077 addition, a less model-dependent “limiting slope” method [38,39] was used. That is,

$$z_a=\underset{P_O\to 0}{\lim }{k}_b\cdot T\cdot \frac{d\ln (P_O)}{dV}$$

(4)

where z_a is equivalent gating charge (in e) transferred during the K_erg-channel activation; dln(P_O)/dV is the slope of relationship between ln(P_O) and testing potential V; and k_b and T are the Boltzmann constant and absolute temperature in Kelvin, respectively. This equation can provide a lower bound of the total gating charge, hence approaching it in the limit of very negative potential. This method was previously described to apply for a sequential model with one open state, where the transition rates are exponentially dependent on the voltage [38,39].

A linear regression analysis of the reciprocal of τ_deact (i.e., λ) versus V_test obtained with or without the addition of UCL-2077 was approximately implemented to yield an estimate of the gating charge (z_deact) involved in the deactivation process according to the following equation:

$${\lambda }_V={\lambda }_0\cdot e^{\left[Z_i\cdot V\cdot \left(\frac{F}{RT}\right)\right]}$$

or

$$\ln \left({\lambda }_V\right)=\ln \left({\lambda }_0\right)+Z_i\cdot V\cdot \left(\frac{F}{RT}\right)$$

(5)

where λ_V and λ₀ correspond to the rate constants of deactivation (i.e., 1/τ_deact) at voltage V_test and 0 mV, respectively, and F, R and T are Faraday’s constant, the universal gas constant and the absolute temperature, respectively.

To construct the effect of UCL-2077 on the steady-state activation of I_K(erg) identified in GH₃, the relationship of the conditioning potential versus the normalized amplitude of I_K(erg) derived with or without the UCL-2077 (10 μM) addition was elaborated as a Boltzmann function, as indicated in the following form:

$$I=\frac{I_{\max }}{1+{\mathrm{e}}^{\left(\frac{-(\mathrm{V}-{\mathrm{V}}_{1/2})}{\mathrm{k}}\right)}}$$

(6)

where I_max denotes the maximal activated I_K(erg), V is the voltage in mV, V_1/2 the voltage required for a half-maximal activation, and k the slope factor appearing in the inactivation curve for I_K(erg).

4.6. Single-Channel Analyses

Unitary current events of BK_Ca or IK_Ca channels digitized were evaluated by pCLAMP software (Molecular Devices). Multi-gaussian adjustments of the amplitude distributions occurring among channel events were applied to precisely determine the opening event of single channels. The functional independence among the channels was achieved as the observed stationary probabilities were compared. We evaluated the channel opening probabilities by using an iterative process to minimize the χ² values calculated from a sufficiently large number of independent observations.

4.7. Statistical Analyses

The linear or nonlinear fitting to data sets (e.g., Hill and Boltzmann functions, and different time courses of I_K(erg) deactivation) was performed with the goodness of fit by virtue of different methods that include the Solver add-in function embedded inside Microsoft Excel 2013 (Redmond, WA, USA) and OriginPro 2016 (OriginLab). The results of the experiments are expressed as the mean ± standard error of the mean (SEM), in which sample sizes (n) are the cell number from which the data in separate set of experiments were achieved. The paired or unpaired t test followed by one-way ANOVA, together with the least-significance difference method for multiple-group comparisons among means, were implemented for statistical evaluation. However, if the Shapiro–Wilk normality test showed that the data were not normally distributed, a Kruskal–Wallis H test followed by Dunn’s multiple comparison tests would be done. The analyses were performed using SPSS version 20.0 (IBM, Armonk, NY, USA). The differences between values were considered significant when p < 0.05, unless stated otherwise.

5. Conclusions

The findings from the present study are as follows: (a) in pituitary GH₃ cells, UCL-2077 decreases I_K(erg) in a dose- and state-dependent manner; (b) UCL-2077 decrease the z_a and z_deact values required for I_K(erg) elicitation; (c) this compound attenuates the voltage hysteresis of I_K(erg) evoked in response to a long-lasting triangular ramp pulse; (d) its presence suppressed the activity of IK_Ca channels; (e), under cell-attached current recordings, it decreased AC frequency. Therefore, these results strongly indicate that the convergent suppression by UCL-2077 of K_erg and IK_Ca channels may be responsible for its effects on the functional activities in these cells.