DPI 201-106 (DPI), a diphenylpiperazinylindole derivative, has been recognized to be a cardio-selective modifier of voltage-gated Na+ channels (1 - 3). Based on its effect on Na+ channels, this compound has been reported to cause a positive inotropic effect, to prolong the duration of action potential (AP), and even to induce torsades de pointes (4 - 6). It has been thought that Na+ channel enhancers can prolong the open state of Na+ channels, increasing net Na+ influx and the activity of reverse mode Na+/Ca2+ exchange (2). This leads to a decrease in net Ca2+ efflux and an increase in intracellular Ca2+ ([Ca2+]i). However, a previous report also showed that DPI might exert some effects on K+ outward currents in guinea-pig ventricular myocytes (7). In skinned fibers from failing hearts, it was shown to alter the Ca2+ sensitivity (8). Whether this compound has any effects on other types of ion currents remains largely unknown.

In addition to the effects on Na+ currents in heart cells, DPI was reported to affect Na+ current in neuroblastoma cells and in rat brain-stem synaptoneurosomes (3, 9, 10). DPI-related compounds were found to affect mammalian brain Na+ channels (11, 12). It has also been shown that DPI could inhibit the release of acetylcholine from insect central nerve terminals (13). Its structurally related compounds were recently reported to block veratridine-evoked acetylcholine release in cricket synaptosomes (14). BDF-9148, a DPI congener, was found to induce a relaxation in the KCl-contracted rat aorta through a mechanism unlinked to the opening of Na+ channels (15). However, little information is available regarding the underlying mechanism of actions of DPI on ion currents or membrane potential in neuroendocrine or endocrine cells.

Therefore, the objective of this study was to investigate whether DPI, a synthetic enhancer of Na+ channels, has any effects on ionic currents and membrane potential in pituitary tumor (GH3) cells. Interestingly, we found that in these cells, DPI is capable of producing inhibitory effects on delayed-rectifier K+ current (IK(DR)) in a concentration- and state-dependent fashion. Current inactivation of IK(DR) in the presence of this compound was also quantitatively characterized in our study. The major action of this compound on IK(DR) is thought to be through an open-channel mechanism. The effects of DPI on ion currents are thus presumably not limited to its modification of Na+ channels.


MATERIALS AND METHODS
Cell preparations

GH3, a clonal cell line derived from a rat prolactin-secreting pituitary tumor, was obtained from the Bioresources Collection and Research Center (BCRC-60015; Hsinchu, Taiwan). Cells were cultured in Ham's F-12 medium (Life Technologies, Grand Island, NY) supplemented with 15% heat-inactivated horse serum (v/v), 2.5% fetal calf serum (v/v) and 2 mM L-glutamine in a humidified environment of 5% CO2/95% air (16). To promote cell differentiation, GH3 cells were transferred to a serum-free, Ca2+-free medium. Under these conditions, cells remained 80 - 90% viable for at least two weeks. Experiments were generally performed 5 or 6 days after cells had been cultured (60 - 80% confluence).

The clonal strain NG108-15 cell line, formed by Sendai virus-induced fusion of the mouse neuroblastoma clone N18TG-2 and the rat glioma clone C6 BV-1, was originally obtained from the European Collection of Cell Cultures (ECACC-88112302; Wiltshire, UK). NG108-15 cells were kept in monolayer cultures at a density of 106/ml in plastic culture disks containing Dulbecco's modified Eagle's medium (Life Technologies) supplemented with 100 µM hypoxanthine, 1 µM aminopterin, 16 µM thymidine, and 5% fetal bovine serum (v/v) as the culture medium, in a humidified incubator equilibrated with 95% air/5 % CO2 at 37 °C (17).

Electrophysiological measurements

Before electrophysiological experiments were performed, GH3 or NG108-15 cells were dissociated with 1% trypsin/EDTA solution and an aliquot of cell suspension was transferred to a recording chamber positioned on the stage of an inverted microscope (DM IL; Leica Microsystems, Wetzlar, Germany). Cells were bathed at room temperature (20 - 25 °C) in normal Tyrode's solution containing 1.8 µM CaCl2. The recording pipettes were made from Kimax-51 glass capillaries (Kimble Glass, Vineland, NJ) using a two-step microelectrode puller (PP-830; Narishige, Tokyo, Japan) and the tips were fire-polished with a microforge (MF-83; Narishige). When filled with pipette solution, their resistance ranged between 3 and 5 M. Ion currents were recorded in whole-cell configuration of the patch-clamp technique as described previously (16), with the use of an RK-400 amplifier (Bio-Logic, Claix, France) or an Axopatch-200B amplifier (Molecular Devices, Sunnyvale, CA).

The signals were displayed on an HM-507 oscilloscope (Hameg, East Meadow, NY) and on a Dell 2407WFP-HC LCD monitor (Round Rock, TX). The data were stored online in a Slimnote VX3 computer (Lemel, Taipei, Taiwan) via a universal serial bus port at 10 kHz through a Digidata-1322A interface (Molecular Devices). This device was controlled by pCLAMP 9.0 software (Molecular Devices). Cell-membrane capacitance of 15 - 32 pF (23.4 ± 5.2 pF; n = 23) was compensated. Series resistance, always in the range of 5 - 13 M, was electronically compensated. Leak and capacitative currents during the experiments were not corrected. Ion currents were low-pass filtered at 1 or 3 kHz. The signals were digitally stored and analyzed subsequently by use of pCLAMP 9.0 (Molecular Devices), Origin 7.5 software (OriginLab, Northampton, MA), or custom-made macros in Microsoft Excel (Redmont, WA). The pCLAMP-generated voltage-step profiles were used to measure the current-voltage (I-V) relationships for ion currents. Action potential duration was measured at 90% of repolarization.

Data analyses

To measure percentage inhibition of DPI on IK(DR), each cell was depolarized from -50 to +50 mV with a duration of 1 s. Current amplitudes measured at the end of depolarizing pulses in the presence of various concentrations (1 - 100 µM) of DPI were compared with the control value. The concentration-response data for DPI-induced inhibition of IK(DR) shown in Fig. 1D were fitted to the Hill equation:



where (C) represents the DPI concentration; and IC50 and n are the concentration required for a 50% inhibition and Hill coefficient, respectively; and Emax is DPI-induced maximal block of IK(DR). Microsoft Solver in Excel or Origin 7.5 software was used to fit data by a least-squares algorithm.

Values were provided as means ± SEM with sample sizes (n) indicating the number of cells from which the data were obtained. The paired or unpaired Student's t-test and one-way analysis of variance with the least-significance-difference method for multiple comparisons were used for the statistical evaluation of differences among means. Level of significance was set to P = 0.05.

Drugs and solutions

DPI 201-106 (SDZ 201-106; DPI; (±)-4-(3-(4-diphenylmethylpiperazine-1-yl)-2-hydroxypropoxy)-1H-indol-2-carbonitrile) was purchased from Biomol (Plymouth Meeting, PA). DPI 201 - 106 was prepared as 10 - 30 mM stock solutions in dimethyl sulfoxide (DMSO) and added to the bath solution at the indicated final concentration. The final concentration (0.1%) of DMSO had no effect on ion currents. Nimodipine, and tetraethyammonium chloride were obtained from Sigma Chemicals (St. Louis, MO), and fura-2 acetoxymethyl ester (fura-2-AM) and ionomycin were from Molecular Probes (Eugene, OR). Tetrodotoxin was purchased from Alomone Labs (Jerusalem, Israel), and cilostazol was from Tocris Cookson Ltd. (Bristol, UK). Tissue culture media and trypsin/EDTA were obtained from Life Technologies.

The composition of normal Tyrode's solution was 136.5 mM NaCl, 5.4 mM KCl, 1.8 mM CaCl2, 0.53 mM MgCl2, 5.5 mM glucose, and 5.5 mM HEPES-NaOH buffer, pH 7.4. To record K+ currents or membrane potential, patch pipette was filled with a solution consisting of 140 mM KCl, 1 mM MgCl2, 3 mM Na2ATP, 0.1 mM Na2GTP, 0.1 mM EGTA, and 5 mM HEPES-KOH buffer, pH 7.2. To measure Ca+2 current, K+ ions inside the pipette solution were replaced with equimolar Cs+ ions, and the pH was adjusted to 7.2 with CsOH.


RESULTS
Inhibitory effect of DPI on delayed-rectifier K+ current (IK(DR)) in pituitary GH3 cells

In an initial set of experiments, the whole-cell configuration of the patch-clamp technique was used to evaluate the effect of DPI on ion currents in GH3 cells. To measure K+ outward currents, cells were bathed in Ca2+-free Tyrode's solution containing tetrodotoxin (1 µM) and CdCl2 (0.5 mM). Tetrodotoxin and CdCl2 were used to block voltage-gated Na+ and Ca2+ currents, respectively. As shown in Fig. 1, when the cell was held at -50 mV, depolarizing voltage pulses from -50 to +50 mV in 20-mV increments were applied with a duration of 1 s, a family of large K+ outward currents with little inactivation was elicited. The threshold and reversal potentials of these K+ currents were also found to be around 0 and -75 mV, respectively. These outward currents were thus identified as IK(DR) (16, 18). When cells were exposed to DPI (10 µM), the amplitude of IK(DR) measured at the end of the voltage pulses was reduced at the potentials ranging from -10 to +50 mV. For example, when the depolarizing pulses from -50 to +30 mV were applied, DPI (10 µM) significantly decreased current amplitude at the end of the voltage pulses from 408 ± 35 to 101 ± 17 pA (n = 8, P < 0.05). After washout of this compound, the amplitude of IK(DR) at the level of +30 mV was partially recovered to 373 ± 14 pA (n = 5). Averaged I-V relationships for the amplitude of initial and steady-state components of IK(DR) in the absence and presence of DPI (10 µM) are illustrated in Fig. 1B. We also found no significant difference in the percentage inhibition of DPI (10 µM) on IK(DR) across the voltages examined (Fig. 1C). These results suggest that block by DPI of IK(DR) in GH3 cells shows little or no voltage dependence, although a time-dependent block can be observed (Fig. 1A).

The relationship between the concentration of DPI and the percentage inhibition of IK(DR) was constructed. In these experiments, each cell was depolarized from -50 to +50 mV with a duration of 1 s. Current amplitudes in the different concentrations of DPI were measured at the end of depolarizing pulses. This compound was found to suppress the steady-state component of IK(DR) in a concentration-dependent manner. The half-maximal concentration (i.e., IC50) required for inhibitory effect of DPI on IK(DR) was calculated to be 9.4 µM, and at a concentration of 100 µM, it almost completely suppressed the steady-state component of IK(DR) (Fig. 1D). Therefore, it is apparent that DPI can exert a significant action on the inhibition of IK(DR) in pituitary GH3 cells.

Fig. 1. Inhibitory effect of DPI on IK(DR) in pituitary GH3 cells. Cells were bathed in Ca2+-free Tyrode's solution containing tetrodotoxin (1 µM) and CdCl2 (0.5 mM). In panel A, superimposed current traces were obtained in the absence and presence of DPI (10 µM). Current traces shown on the upper part are control, and those on the lower part were obtained 2 min after addition of DPI (10 µM). Panel B shows averaged I-V relationships for initial (circles) and steady-state (squares) components of IK(DR) in the absence (upper) and presence (lower) of 10 µM DPI. Mean ± SEM (n = 7-12). In panel C, bar graph shows a lack of voltage-dependence for DPI-induced block of IK(DR) in GH3 cells. The percentage inhibition of IK(DR) by DPI (10 µM) was shown over the voltage range of -10 to +50 mV (mean ± SEM, n = 5-10). Panel D shows concentration-response curve for DPI-induced inhibition of IK(DR). The amplitude of IK(DR) measured at the end of depolarizing pulse in the presence of various concentrations (1-100 µM) of DPI was compared with the control value, i.e., in the absence of DPI (mean ± SEM, n = 4-10). The smooth line represents the best fit to a Hill function. The values for IC50, maximally inhibited percentage of IK(DR) and the Hill coefficient were 9.4 µM, 100% and 1.1, respectively.

Kinetic studies of DPI-induced block of IK(DR)

Because IK(DR) observed in the presence of DPI tends to exhibit a pronounced peak followed by an exponential decay to a steady-state level, it is thus important to determine the kinetics of DPI-induced block of IK(DR) in GH3 cells. The concentration dependence of IK(DR) decay by DPI is illustrated in Fig. 2A and 2B. Although the initial activation phase of IK(DR) elicited by depolarizing pulse from -50 to +50 mV was unchanged during the exposure to this compound, its effects on IK(DR) were found to be concentration-dependent increase in the rate of current decay accompanied by a decrease in the residual, steady-state current. Similarly, when cells were depolarized from -50 to +30 mV, the inactivation time constants of IK(DR) obtained during exposure to 3 and 10 mM DPI were fitted by a single exponential with the values of 83 ± 7 and 51 ± 6 msec (n = 5), respectively. Therefore, increasing the concentration of DPI not only reduces the peak current, but also enhances the apparent inactivation. The results suggest that the inhibitory effect of DPI on IK(DR) in GH3 cells can be explained by state-dependent block where it binds to the open state of the channel according to the minimal kinetic scheme (19, 20):



where and ß are the voltage-dependent rate constants for the opening and closing of the delayed-rectifier K+ (KDR) channel, kOB and kBO, those for block and unblock by the DPI molecule, and (B) is the blocker (i.e., DPI) concentration. C, O and O·B shown in this scheme represent the closed, open, and open-blocked states, respectively.

Fig. 2. Evaluation of the kinetics of DPI-induced block of IK(DR) and time course of IK(DR) recovery in the presence of DPI. In panel A, typical IK(DR) was elicited by depolarizing pulses to +50 mV in the presence of DPI. The time course of current decay in the presence of 3 and 10 µM DPI was well fitted by a single exponential with a value of 79 and 47 msec, respectively. The upper part indicates the voltage protocol used. a: control; b: 3 µM DPI; c: 10 µM DPI. In panel B, the reciprocal of the time constant of the rate of block, obtained by a single-exponential fit of the decay phase of IK(DR), was plotted against the DPI concentration. Data points were fitted by a linear regression, indicating that block occurs with a molecularity of 1. Block (kOB) and unblock (kBO) rate constants, given by the slope and the y-axis intercept of the interpolated line, were 1.231 msec-1mM-1 and 0.00897 msec-1, respectively. Each point represents mean ± SEM (n = 5 - 8). In panel C, superimposed current traces were obtained by a two-pulse protocol used to examine IK(DR) recovery. Pituitary GH3 cells, bathed in Ca2+-free Tyrode's solution, were depolarized from -50 to +50 mV with a duration of 300 ms and different interpulse durations were applied. Voltage protocol is shown in the upper part of panel C. Panel D depicts the time course of recovery from inactivation of IK(DR) caused by DPI. : 10 µM; : 30 µM. The time course in the presence of 10 and 30 µM DPI was fitted to a single exponential with a time constant of 771 and 1457 msec, respectively. Each point represents mean ± SEM (n = 4 - 5). Notably, the abscissa in panels C and D was shown at logarithmic scale.

Block and unblock rate constants, kOB and kBO, were determined from the time constants of current decay evoked by the depolarizing pulses from -50 to +50 mV with a duration of 1 sec (Fig. 2A). Block and unblock rate constants could be estimated using a first-order blocking scheme:



In particular, kOB and kBO, respectively, result from the slope and from the y-axis intercept at (B) = 0 of the linear regression interpolating the reciprocal time constants (i.e., 1/b) versus different DPI concentrations. As predicted by this scheme, the relationship between 1/b and (B) was linear with a correlation coefficient of 0.95 (Fig. 2B), and the block and unblock rate constants were calculated to be 1.231 msec-1mM-1 and 0.00897 msec-1, respectively. Based on these rate constants, a value of 7.3 µM for the dissociation constant (KD = kBO/kOB) could be derived. Notably, this value was found to be in agreement with the IC50 value determined from the concentration-response curve (Fig. 1D). However, the rate constant of the inverse reaction (i.e., unblock rate constant), kBO, showed little dependence on [B]; kBO was 0.00878 ± 0.004 msec-1 (n = 5) at 3 µM and 0.00891 ± 0.005 msec-1 (n = 5) at 10 µM.

Recovery from block induced by DPI

Recovery from block was also determined with a double-pulse protocol consisting of a first (conditioning) depolarizing pulse sufficiently long to allow block to reach a steady-state. During the exposure to DPI, the membrane potential was then taken to +50 mV from -50 mV for a variable time, after which a second depolarizing pulse (test pulse) was applied at the same potential as the conditioning pulse (Fig. 2C). The ratios of the peak current amplitudes of IK(DR) evoked in response to the test and the conditioning pulse were then taken as a measure of recovery from block, and plotted versus interpulse interval. Recovery obtained in the presence of 10 µM DPI was generally complete, and its time course described by a single-exponential function with a time constant of 771 ± 57 msec (n = 5). In addition, the mean time constant for recovery from inactivation was significantly increased to 1457 ± 129 msec (n = 4, P < 0.05) when cells were exposed to 30 µM DPI (Fig. 2D). These results led us to propose that the slowing of recovery caused by DPI might be due to open channel block.

DPI-induced decay of tail current and crossover

Assuming that DPI interacts with the open channel, the dissociation of this compound from the blocked channel may result in a transient conducting channel that subsequently closes over time. Fig. 3 shows the superimposed currents evoked in response to the depolarizing pulse from -50 to +50 mV and the tail currents at -20 mV in the absence and presence of 3 µM DPI. In the absence of DPI, the depolarizing pulses activated IK(DR) that decayed upon repolarization to -20 mV. Cell exposure to DPI induced a decrease in current during the depolarizing pulse accompanied by peak amplitude of the tail current. In addition, it slowed the decay of the tail current. This resulted in the crossover phenomenon that was thought to be an indicator of state-dependent binding (17, 20). Under control conditions, the tail current upon repolarization to -30 mV decayed with a time constant of 52.7 ± 3.1 msec (n = 6). However, after the addition of DPI (3 µM), the decay of the tail current at -30 mV was fitted to a two-exponential with a time constant of 4.9 ± 0.8 and 91 ± 8.9 msec (n = 6). Thus, the tail current observed in the presence of DPI consists of two components. One component decays faster than the tail current in the control, while the other declines slower. This slowing in the decay of tail current and the crossover phenomenon are in agreement with the presence of transient unblocking and provide additional evidence for open channel block.

Fig. 3. Effect of DPI on the relaxation of tail current. In these experiments, the voltage pulses were depolarized from -50 to +50 mV and stepped back to -20 mV at which tail currents were measured. Notably, the tail current in control decays more rapidly and crosses over the tail current recorded in the presence of 3 µM DPI. a: control; b: 3 µM DPI.

Effect of DPI on APs in GH3 cells

In another set of experiments, we examined the effect of DPI on changes in membrane potential. In these experiments, cells were bathed in normal Tyrode's solution containing 1.8 mM CaCl2 and 1 µM tetrodotoxin. The resting membrane potential and AP duration were -71 ± 7 mV and 83 ± 19 msec, respectively (n = 12). The APs of these cells were sensitive to extracellular Ca2+ and could be abolished by nimodipine (10 µM), yet not by tetrodotoxin (1 µM) (16). The effects of DPI on APs in GH3 cells are illustrated in Fig. 4A and 4B. Within 2 min of exposing cells to DPI (10 µM), the AP duration was significantly prolonged to 142 ± 26 msec from a control value of 78 ± 18 msec (n = 9, P < 0.05). However, neither initial rise of AP (34.9 ± 3.5 vs. 35.1 ± 3.4 mV/msec; n = 9, P > 0.05) nor resting membrane potential (-72.6 ± 2.1 vs. -72.5 ± 1.9 mV; n = 9, P > 0.05) was significantly altered during the exposure to 10 µM DPI.

The effect of DPI on spontaneous firing of APs was further investigated. The typical effect of DPI (3 and 10 µM) on APs in these cells is illustrated in Fig. 4D. Notably, when cells were exposed to DPI, the duration of spike discharge was increased (Fig. 4E). For example, DPI at a concentration of 3 µM significantly increased the duration from 134 ± 31 to 238 ± 38 msec (n = 7, P < 0.05). However, no change in firing frequency could be detected in the presence of DPI (Fig. 4F). Therefore, the effect of DPI on spike broadening in GH3 cells was paralleled by a time-dependent decrease in IK(DR).

Fig. 4. Effect of DPI on APs of GH3 cells. Cells were bathed in normal Tyrode's solution containing 1.8 mM CaCl2 and tetrodotoxin (1 µM). The pipette was filled with a K+-containing solution. The experiments were conducted under current-clamp recordings. Panel A shows the potential traces obtained in control (a) and in the presence of 3 (b) and 10 (c) µM DPI. A 5-msec supra-threshold stimulus was used to induce AP. In panels B and C, bar graphs show effects of DPI (3 and 10 µM) on APD90 and initial rise of AP, respectively. Each point represents the mean ± SEM (n = 7-12). *Significantly different from control (P < 0.05). Notably, the presence of DPI produces a significant increase in AP duration with no change in initial rise. In panel D, original potential traces show the effect of DPI on the spontaneous firing of APs. a: control; b: 3 µM DPI; c: 10 µM DPI. In panels E and F, bar graphs indicate the effects of DPI (3 and 10 µM) on duration and firing frequency of APs, respectively. Each point represents the mean ± SEM (n = 5 - 7). *Significantly different from control (P < 0.05).

Effect of DPI on IK(DR) in NG108-15 neuronal cells

Finally, to verify whether DPI-induced inhibition of IK(DR) could also be observed in neuronal cells, we examined the effects of this compound on NG108-15 neuronal cells. As shown in Fig. 5, DPI (10 µM) significantly decreased the amplitude of IK(DR) measured at the end of depolarizing pulses by 59 ± 8 % (n = 5, P < 0.05). Similar to the results obtained in GH3 cell, block of IK(DR) by DPI was not instantaneous, but developed with time after the channels were opened, thus producing an apparent inactivation of the current. Consistent with the observations in GH3 cells, these data thus indicate that DPI can induce the block of IK(DR) as well as accelerate current decay in NG108-15 cells.

Fig. 5. Effects of DPI on IK(DR) in NG108-15 cells. Cells were bathed in Ca2+-free Tyrode's solution. In panel A, whole-cell currents were evoked by 300-msec depolarizing pulses from -50 to +50 mV. a: control; b: 10 µM DPI; c: 30 µM DPI. In panel B, bar graph shows current amplitude at the end of depolarizing pulse in the absence and presence of 10 and 30 µM DPI. Mean ± SEM (n = 5). *Significantly different from control (P < 0.05).

DISCUSSION
Results of this study demonstrated that DPI inhibited the amplitude of IK(DR) in pituitary tumor (GH3) cells in a concentration- and time-dependent manner, although its current inhibition was voltage-independent. Unlike the effect on Ca2+-activated K+ current (see Supplementary Information), the inhibitory action on IK(DR) was correlated in time with a significant increase in the inactivation of the currents, while the activation kinetics remained unaltered. In these cells, DPI-induced block of IK(DR) is likely an important mechanism by which it prolongs the duration of APs.

Previous work has demonstrated the ability of DPI to elevate [Ca2+]i or intracellular Na+ (1, 2, 21). DPI was recently reported to increase the Ca2+ sensitivity of left ventricular skinned fibers from failing human hearts (8). However, in our study, DPI at a concentration of 10 µM was found to have no effect on basal [Ca2+]i in GH3 cells (see Supplementary Information). When the recording pipette was loaded with a high concentration of EGTA (10 mM), blockade of IK(DR) with DPI still was observed (data not shown). Taken together, it is unlikely that time-dependent block of IK(DR) caused by DPI in GH3 cells is associated with changes in [Ca2+]I. It is also tempting to attribute the DPI effect to an interference with a channel-associated binding site.

DPI inhibited IK(DR) in a concentration-dependent manner with an IC50 value of 9.4 µM (Fig. 1D). This value is similar to those required for increased prolongation of cardiac AP and enhanced cardiac contractility (4-6). In our study, the time- and state-dependent block of IK(DR) caused by DPI could explain its effect on the firing of APs in GH3 cells. In addition to voltage-gated Na+ currents (1-3), IK(DR) may be a relevant target for the action of this compound, if these types of K+ currents that are sensitive to DPI are present in heart cells. Blockade of IK(DR) combined with inhibition of Na+ channel inactivation can be responsible for its prolongation of APs and increase of stimulus-secretion coupling in neuroendocrine or endocrine cells.

The time-dependent block of IK(DR) in the presence of DPI was observed in our study. The DPI-sensitive current could be also fitted in an exponential fashion (Fig. 2). The rate of current inactivation was enhanced as the DPI concentration increased. Our experimental results thus suggest the presence of open channel block. In our study, we also found that when internally applied, DPI caused little or no effect on IK(DR) in GH3 cells Therefore, block of IK(DR) caused by DPI is likely to be associated with an interaction with the channel molecule from the exterior.

In our study, recovery from inactivation in the presence of DPI (10 µM) was slow because more than 1 sec was required for the channel to recover completely (Fig. 2). Consequently, DPI-induced block of IK(DR) will even become very significant when a train of APs occurs, because of the fact that under these conditions, the availability of KDR channels is decreased as a function of firing frequency (22, 23). However, the time scale of block onset in the presence of DPI appears to be too slow to be relevant for a single spike repolarization. The issue of whether blockade of KDR channels is responsible for the DPI-induced AP widening remains to be further investigated.

An important feature of the block of IK(DR) by DPI in GH3 or NG108-15 cells is that the initial rising phase of the current (i.e., the activation time course) remained unaffected in the presence of DPI. At the beginning of the voltage pulse, dI/dt is generally proportional to the number of channels available for activation. Our experimental results showing that dI/dt was unaltered during cell exposure to DPI, suggest that before channel activation, there should be the absence of any significant resting block of KDR channels. However, this compound tends to accelerate IK(DR) inactivation in a concentration-dependent fashion, suggesting that the DPI molecule reaches the blocking site only when the channel is in the open state. This feature can thus be explained by the minimal binding scheme, i.e., closed open open-blocked (17, 20). Inherent to this blocking scheme is that open-blocked channels are not closed unless DPI dissociates from the binding site, thus providing only one recovery path. In other words, DPI-induced block was interpreted to mean that it preferentially binds to and block an open state of the channel. DPI and its structurally related compounds might be useful pharmacological probes for gaining insights into possible mechanisms controlling the gating of KDR channels.

Based on the I-V relationships of IK(DR) in GH3 cells (Fig. 1), the effect of DPI on AP duration appears to be in the range of voltage (below -10 mV) where IK(DR) is not active. However, as shown in Fig. 3, the deactivating tail current observed in the presence of DPI crossed over the tail current observed in the absence of this compound when the two tracings were superimposed. This crossover suggests that DPI block can slow channel closing, and that this compound needs to be dissociated from the channel before the channel can be activated. Upon repolarization in the presence of DPI, the open state unblocks and then either closes, or binds the compound again to repeat the process; thus leading to a current increase followed by a slowing of the rate of current decay and the observed crossover. As a result, DPI-induced widening of APs can be explained by its state-dependent block of IK(DR). (17, 19, 22). Indeed, our study is consistent with previous reports showing that DPI-induced decrease in K+ conductance can partly contribute to its prolongation of cardiac action potential (7).

Previous results showing little or no effect of DPI on K+ outward current in guinea-pig ventricular myocytes (4) are somewhat different from the present study. This discrepancy remains unknown; however, it could be due to the DPI concentration. The DPI concentration used for guinea-pig heart cells is about 1 µM, while the IC50 value found in our study is 9.4 µM. It has been reported that sustained K+ currents in GH3 cells could be mediated by several different subtype Kv channels, i.e., Kv1.2, Kv1.4, Kv1.5, Kv2.1, Kv2.2, Kv3.2, Kv4.1 and Kv5.1 (24). The role of these K+ channels is to stabilize the resting membrane potential and reduce the width of APs (22, 25). It thus remains to be determined whether DPI and its structurally related compounds may affect many types of voltage-gated K+ channels in a variety of cells. It is also noteworthy that effects of DPI on ion currents in heart cells have been recognized to manifest as a prolongation of AP duration and of its electrocardiographic surrogate, the QT interval (4-6, 26, 27). It will be important to determine whether in addition to its effects on cardiac Na+ channels, DPI produces a similar action on other types of KDR channels in heart cells (28).


SUPPLEMENTARY INFORMATION
Supplementary information accompanied this paper.

Acknowledgments: This work was partly supported by grants from the National Science Council (NSC-93-2320B-006-055 and NSC-94-2320B-006-019), and the Program for Promoting Academic Excellence and Developing World Class Research Centers, Ministry of Education, Taiwan. The authors would like to thank Adonis Z. Wu for technical assistance.


REFERENCES

1. Doggrell S, Hoey A, Brown L. Ion channel modulators as potential positive inotropic compound for treatment of heart failure. Clin Exp Pharmacol Physiol 1994; 21: 833-843.
2. Flesch M, Erdmann E. (2001) Na+ channel activators as positive inotropic agents for the treatment of chronic heart failure. Cardiovasc Drugs Ther 2001; 15: 379-386.
3. Denac H, Mevissen M, Kühn FJ, Guionaud CT, Scholtvsik G, Greeff NG. Molecular cloning and functional characterization of a unique mammalian cardiac Nav channel isoform with low sensitivity to the synthetic inactivation inhibitor (-)-(S)-6-amino-alpha-((4-diphenylmethyl-1-piperazinyl)-methyl)-9H-purine-9-ethanol (SDZ 211-939). J Pharmacol Exp Ther 2002; 303: 89-98.
4. Yuill KH, Convery MK, Dooley PC, Doggrell SA, Hancox JC. Effects of BDF 9198 on action potentials and ionic currents from guinea-pig isolated ventricular myocytes. Br J Pharmacol 2000; 130: 1753-1766.
5. Kühlkamp V, Mewis C, Bosch R, Seipel L. Delayed sodium channel inactivation mimics long QT syndrome 3. J Cardiovasc Pharmacol 2003; 42: 113-117.
6. Hamlin RL (2007) Animal models of ventricular arrhythmias. Pharmacol Ther 2007; 113: 276-295.
7. Amos GJ, Ravens U. The inotropic agents DPI 201-106 and BDF 9148 differentially affect potassium currents of guinea-pig ventricular myocytes. Naunyn-Schmiedeberg's Arch Pharmacol 1994; 350: 426-433.
8. Okafor C, Liao P, Perreault-Micale C, et al. Mg-ATPase and Ca2+ activated myosin AtPase activity in ventricular myofibrils from non-failing and diseases human hearts-effects of calcium-sensitizing agents MCI-154, DPI 201-106, and caffeine. Mol Cell Biochem 2003; 245: 77-89.
9. Romey G, Quast U, Pauron D, Frelin C, Renaud JF, Lazdunski M. Na+ channels as sites of action of he cardioactive agent DPI 2-1-106 with agonist and antagonist enantiomers. Proc Natl Acad Sci USA 1987; 84: 896-900.
10. Cohen-Armon M, Sokolovsky M. Evidence for involvement of the voltage-dependent Na+ channel gating in depolarizing-induced activation of G-proteins. J Biol Chem 1993; 268: 9824-9838.
11. Payne GT, Deecher DC, Soderlund DM. Structure-activity relationships for the action of dihydropyrazole insecticides on mouse brain sodium channels. Pestic Biochem Phys 1998; 60: 177-185.
12. Leong D, Bloomquist JR Bempong J, et al. Insecticidal arylakylbenzhydrolpiperidines: novel inhibitors of voltage-sensitive sodium and calcium channels in mammalian brain. Pest Manag Sci 2001; 57: 889-895.
13. Nicholson RA, Kumi CO. The effects of pesticides, brevetoxin B, and the cardiotonic drug DPI 201 106 on release of acetylcholine from insect central nerve terminals. Pestic Biochem Phys 1991; 40: 86-97.
14. Bloomquist JR, Payne GT, Kinne L, Lyga J, Leong D, Nicholson RA. Toxicity and mode of action of benzhydrolpiperidines and related compounds in insects. Pestic Biochem Phys 2002; 73: 18-26.
15. Doggrell SA, Liang LC. The relaxing effect of BDF 9148 on the KCl-contracted aorta isolated from normo- and hyper-tensive rats. Naunyn-Schmied Arch Pharmacol 1998; 357: 126-132.
16. Wu SN, Li HF, Chiang HT. Stimulatory effects of d-hexachlorocyclohexane on Ca2+-activated K+ currents in GH3 lactotrophs. Mol Pharmacol 2000; 57: 865-873.
17. Wu SN, Lo YK, Chen H, Li HF, Chiang HT. Rutaecarpine-induced block of delayed rectifier K+ current in NG108-15 neuronal cells. Neuropharmacology 2001; 41: 834-843.
18. Simasko SM. Evidence for a delayed rectifier-like potassium current in the clonal rat pituitary cell line. Am J Physiol 1991; 261: E66-E75.
19. Marom S, Abbott LF. Modeling state-dependent inactivation of membrane currents. Biophys J 1994; 67: 515-520.
20. Liu CP, Chiang HT, Jan CR, Wu SN. Mechanism of carvedilol-induced block of delayed rectifier K+ current in the NG108-15 neuronal cell line. Drug Develop Res 2003; 58: 196-208.
21. Ravens U, Himmel HM. Drugs preventing Na+ and Ca2+ overload. Pharmacol Res 1999; 39: 167-174.
22. Marom S. Slow changes in the availability of voltage-gated ion channels: effects on the dynamics of excitable membranes. J Membr Biol 1998; 161: 105-113.
23. Lien CC, Jonas P. Kv3 potassium conductance is necessary and kinetically optimized for high-frequency action potential generation in hippocampal interneurons. J Neurosci 2003; 23: 2058-2068.
24. Wulfsen I, Hauber H-P, Schiemann D, Bauer CK, Schwarz JR. Expression of mRNA for voltage-dependent and inward-rectifying K+ channels in GH3/B6 cells and rat pituitary. J Neuroendocrinol 2000; 12: 263-272.
25. Lee SH, Lee EH, Ryu SY, et al. Role of K+ channels in frequency regulation of spontaneous action potentials in rat pituitary GH3 cells. Neuroendocrinology 2003; 78: 260-269.
26. Raap A, Armah B, Mest HJ, Stenzel W, Schloos J, Blechacz W. Investigations of the mechanism of the positive inotropic action of BDF 9148: comparison with DPI 201-106 and the enantiomers. J Cardiovasc Pharmacol 1997; 29: 164-173.
27. Mevissen M, Denac H, Schaad A, Portier CJ, Scholtysik G. Identification of a cardiac sodium channel insensitive to synthetic modulators. J Cardiovasc Pharmacol Ther 2001; 6: 201-212.
28. Rasmusson RL, Morales MJ, Wang S, et al. Inactivation of voltage-gated cardiac K+ channels. Circ Res 1998; 82: 739-750.