Stimulatory Effects of δ-Hexachlorocyclohexane on Ca²⁺-Activated K⁺ Currents in GH₃Lactotrophs

Materials and Methods

Cell Preparation.

GH₃ (a cell line from a rat anterior pituitary adenoma) cells were obtained from the Culture Collection and Research Center (CCRC-60015, Hsinchu, Taiwan). Cells were routinely cultured in 50-ml Ham's F-12 medium (Life Technologies, Grand Island, NY) that was supplemented with 15% horse serum (v/v), 2.5% fetal calf serum (v/v), and 2 mMl-glutamate (Life Technologies) in a 5% CO₂ atmosphere. Cells were subcultured once a week, and a new stock line was generated from frozen cells (frozen in 10% glycol in medium plus serum) every 3 months. The experiments were performed after 5 or 6 days of subcultivation (60 to 80% confluence).

Stock cultures of human neuroblastoma IMR-32 cells were also obtained from the Culture Collection and Research Center (CCRC-60014). IMR-32 cells were maintained in Eagle's minimal essential medium (Life Technologies) supplemented with 2 mM l-glutamine and Earle's balanced salt solution adjusted to contain 1.5 g/liter sodium bicarbonate, 0.1 mM nonessential amino acids, 1 mM sodium pyruvate, and 10% fetal bovine serum (v/v).

Electrophysiological Measurements.

Immediately before each experiment, GH₃ or IMR-32 cells were dissociated and an aliquot of cell suspension was placed into a recording chamber affixed to the stage of an inverted phase-contrast microscope (Diaphot-200; Nikon, Tokyo, Japan). The microscope was coupled to a video camera system with a magnification of up to 1500× to continually monitor cell size during the experiments. Cells were bathed at room temperature (20–25°C) in normal Tyrode's solution containing 1.8 mM CaCl₂. Ionic currents were recorded in the whole-cell or inside-out configuration of the patch-clamp technique, using a patch-clamp amplifier (RK-400; Biologic, Claix, France) (Hamill et al., 1981; Wu et al., 1999a). Patch pipettes (3 to 5 MΩ in bathing solution) were made from borosilicated glass capillary tubes (Kimble Products, Vineland, NJ) using a two-step pipette puller (PB-7; Narishige Scientific, Tokyo, Japan), and the tips were heat-polished with a microforge (MF-83; Narishige). A programmable stimulator (SMP-311; Biologic) was used to digitally generate voltage pulses, which were rectangular- or ramp-shaped. Tested drugs were applied by perfusion or added to the bath to obtain the final concentration indicated.

The signals, consisting of voltage and current tracings, were displayed on a digital storage oscilloscope (model 1602; Gould, Valley View, OH) and on-line recorded in a digital audiotape recorder (model 1204; Biologic). After the experiments, the data were fed back and stored in a Pentium III-grade computer (Lemel, Taipei, Taiwan) at 10 kHz through an analog/digital interface (Digidata 1200; Axon Instruments, Foster City, CA) using the Clampex subroutine of the pClamp 7.0 software (Axon Instruments). Voltage-activated currents recorded during whole-cell experiments were stored without leakage correction and analyzed using the Clampfit subroutine (Axon Instruments) or pClamp module in the Origin 6.0 software package (Microcal Software, Inc., Northampton, MA) to establish a current-voltage relationship for ionic currents.

Single Channel Analysis.

Single channel currents were analyzed using Fetchan and Pstat subroutines in the pClamp software (Axon Instruments). Multi-Gaussian adjustments of the amplitude distributions between channels were used to determine unitary currents. The functional independence among channels was verified by comparing the observed stationary probability with the values calculated according to the binomial law. The number of active channels in a patch, N, was taken as the maximum number of channels simultaneously open under conditions of maximum open probability. When there was a sufficiently large number of independent observations, the opening probabilities (N·P_o) of unitary current were evaluated by an iterative process that was continued until the χ² value was no longer changed. The single channel conductance was calculated by linear regression using mean values of the current amplitudes measured at different voltages

To assess the concentration-dependent effect of δ-HCH on the activation of BK_Ca channels, the opening probabilities of the channel current enhanced by various concentrations of δ-HCH (3 to 300 μM) were examined. Under symmetrical K⁺ (145 mM) conditions, the inside-out configuration in which bath medium contained 0.1 μM Ca²⁺ was performed, and the holding potential was set at +60 mV. The opening probability of BK_Cachannels in the presence of 300 μM δ-HCH was considered to be 1.0 and the channel activity produced by various concentrations of δ-HCH was compared. The curve was fitted to the Hill equation by using a nonlinear regression analysis. The following form of the Hill equation was used: y = y_max ×xⁿ/(cⁿ +xⁿ), where x is the concentration of δ-HCH, y is the relativeN·P_o,y_max is the maximal relativeN·P_o, and c andn are the concentrations required for a 50% increase and Hill coefficient, respectively.

To determine the effect of δ-HCH on the activation curve of BK_Ca channels, the ramp pulses from +20 to +140 mV with a duration of 1 s were digitally applied with the aid of a programmable stimulator (SMP-311). This made the measurements of single channel conductance and channel activation more efficient (Carl and Sanders, 1990). The activation curves were calculated by averaging current responses to 20 voltage ramps and dividing each point of the averaged current by the unitary amplitude of each potential after each leakage component was corrected. The rate of change of voltage ramps was 120 mV/s, a value that was found to be not distorted by the time constants of activation or deactivation (Carl and Sanders, 1990). The number of active channels in the patch N was also counted at the end of each experiment by perfusing a high K⁺ solution with 100 μM Ca²⁺. The number was then used to normalize the opening probability at each potential. To obtain values for the slope factor of the voltage-dependent activation and half-maximal activation voltage, the activation curves obtained before and after the addition of δ-HCH were fitted with Boltzmann functions of the form: relative N·P_o =n/{1 + exp[ − K(V −V_1/2)]}, where n is the maximal relative N·P_o,K⁻¹ is the slope factor of the voltage-dependent activation [i.e., change in potential required to produce an (exponential) e-fold increase in the activation], and V_1/2 is the voltage at which there is half-maximal activation. Curve fitting to the data presented here was performed by use of Origin 6.0 software (Microcal).

Open lifetime distributions measured before and after the addition of δ-HCH were fit with logarithmically scaled bin width by using the method of McManus et al. (1987). When the square root of the number of events in a bin was plotted against the open lifetime, each component of the open lifetime distribution appeared as a clear peak and the respective time constant would fall in the vicinity of this peak.

All values are reported as means ± S.E. The paired or unpaired Student's t test and ANOVA with a least-significance difference method for multiple comparison were used for the statistical evaluation of differences among means. Differences between the values were considered statistically significant when P was <.05 or <.01.

Drugs and Solutions.

δ-HCH (1α,2α,3α,4α,5α,6β-hexachlorocyclohexane), γ-HCH (Lindane: 1α,2α,3β,4α,5α,6β-hexachlorocyclohexane), 17β-estradiol, and tetraethylammonium chloride were purchased from Sigma. Paxilline and ryanodine were obtained from Biomol (Plymouth Meeting, PA). Dantrolene, ruthenium red, inositol 1,4,5-trisphosphate hexasodium (IP₃), ionomycin, and tetrodotoxin were obtained from Research Biochemicals (Natick, MA). Clotrimazole was purchased from Calbiochem (La Jolla, CA). All other chemicals were of the highest quality commercially available. The composition of normal Tyrode's solution was as follows (in mM); NaCl 136.5, KCl 5.4, CaCl₂ 1.8, MgCl₂ 0.53, glucose 5.5, and HEPES-NaOH buffer 5 (pH 7.4). To record K⁺ currents or membrane potential, the patch pipettes were filled with solution (in mM): KCl 140, KH₂PO₄ 1, MgCl₂ 1, EGTA 0.1, Na₂ATP 3, Na₂GTP 0.1, and HEPES-KOH buffer 5 (pH 7.2). To record Ca²⁺ current, KCl inside the pipette solution was replaced with equimolar CsCl, and the pH was adjusted to 7.2 with CsOH. For the inside-out patch-clamp recording, high K⁺-bathing solution contained (mM): KCl 145, MgCl₂ 0.53, and HEPES-KOH buffer 5 (pH 7.4), and the pipette solution contained (mM); KCl 145, MgCl₂ 2, and HEPES-KOH buffer 5 (pH 7.2).

Results

Stimulatory Effect of Ca²⁺-Activated K⁺Current (I_K(Ca)) by δ-HCH in GH₃Cells.

In these experiments, GH₃ cells were bathed in normal Tyrode's solution containing 1.8 mM CaCl₂. Each cell was held at the level of 0 mV to inactivate other voltage-dependent K⁺ currents (Wu et al., 1999b). When cells were depolarized from 0 mV to various potentials with a duration of 300 ms at a rate of 0.1 Hz, a family of large noisy outward currents were elicited. The direction of this membrane current was reversed at −80 mV. The current amplitudes were increased with greater depolarization, reduced by the removal of extracellular Ca²⁺, and enhanced by the presence of ionomycin (10 μM). These outward currents were thus identified as Ca²⁺-activated K⁺ currents (I_K(Ca)) (Wu et al., 1999b). When the cell was exposed to δ-HCH (30 μM), the amplitude of outward current was profoundly increased throughout the entire voltage-clamp step (Fig.1). For example, when cells were depolarized from 0 to +70 mV, δ-HCH (30 μM) significantly increased the current amplitude measured at the end of the voltage pulses from 386 ± 96 to 1221 ± 150 pA (P < .05;n = 10). However, the effect of δ-HCH was poorly reversible after 5 min of washout. The averaged current-voltage (I–V) relationships for the current amplitude in the absence and presence of δ-HCH (30 μM) are illustrated in Fig.1B.

To assess the nature of the outward current stimulated by δ-HCH, another series of experiments were conducted in bath solution containing different extracellular K⁺concentrations. The reversal potential in each cell was measured in the presence of δ-HCH. The data were then pooled and plotted as a function of extracellular K⁺ concentrations (Fig.1C). The finding showing a best-fit line through the averaged data revealed a slope of 57 mV per 10-fold increase in extracellular K⁺. These results were interpreted to indicate that the δ-HCH-stimulated outward current in these cells followed the Nernstian behavior of a K⁺-selective channel.

Lack of Effect of δ-HCH on Voltage-Dependent K⁺Outward Current (I_K(V)) in GH₃Cells.

To determine whether δ-HCH affects the amplitude of voltage-dependent I_K in these cells, the experiments were conducted in cells bathed in Ca²⁺-free Tyrode's solution containing 1 μM tetrodotoxin and 0.5 mM CdCl₂. When the cell was held at −60 mV and various potentials ranging from −50 to +70 mV were applied, the addition of δ-HCH (30 μM) did not have effect onI_K(V). However, a higher concentration of δ-HCH (100 μM) slightly inhibited the noninactivating component ofI_K(V) (Fig.2A). For example, the current amplitude measured at the end of the voltage pulses from −60 to +70 mV was decreased by the presence of δ-HCH (100 μM) to 286 ± 25 pA from a control value of 333 ± 29 pA (P < .05,n = 5). Thus, these results indicate that δ-HCH at a concentration of 100 μM or above can suppress the amplitude ofI_K(V) in GH₃ cells.

Inhibitory Effect of δ-HCH on Voltage-Dependent L-type Ca²⁺ Current (I_{Ca, L}) in GH₃ Cells.

The effect of δ-HCH onI_{Ca, L} was also examined. The experiments were conducted with the Cs⁺-containing pipette solution. As shown in Fig. 2B, the cell was held at −50 mV, and the depolarizing pulses (300 ms in duration) to 0 mV were delivered at 0.1 Hz. The presence of δ-HCH suppressed the amplitude ofI_{Ca, L} in a concentration-dependent manner. When cells were depolarized from −50 to 0 mV, the amplitude ofI_{Ca, L} was significantly decreased by δ-HCH (30 μM) to 195 ± 29 pA from a control value of 368 ± 26 pA (P < .05; n = 7). However, under the same voltage protocol, the presence of δ-HCH (30 μM) produced no significant change in the kinetics of activation or inactivation of I_{Ca, L} [control: τ_act = 5 ± 3 ms, τ_inact(f) = 20 ± 7 ms, τ_inact(s) = 206 ± 11 ms; δ-HCH: τ_act = 5 ± 3 ms, τ_inact(f) = 21 ± 6 ms, τ_inact(s) = 208 ± 13 ms (n = 6)]. In addition, the inward tail current, which was evoked by the depolarizing pulses that activateI_{Ca, L}, was reduced by δ-HCH (Fig. 2B). However, there was no significant effect on theI–V relationship of I_{Ca, L} in the presence of δ-HCH (data not shown). Conversely, unlike δ-HCH, γ-HCH (30 μM) did not significantly affect the amplitude of I_{Ca, L}. These results indicate that, like nifedipine or tetrandrine (Wu et al., 1998), δ-HCH is capable of suppressing the amplitude of I_{Ca, L} in GH₃ cells.

Effect of δ-HCH on the Activity of Large Conductance Ca²⁺-Activated K⁺ (BK_Ca) Channels in GH₃ Cells.

BecauseI_K(Ca) is a large, noisy, voltage-dependent, Ca²⁺-sensitive current, and it results mainly from the opening of BK_Cachannels that have been previously studied (Wu et al., 1999b). Therefore, to determine whether the effect of δ-HCH onI_K(Ca) is related to the increased amplitude of unitary current, the enhanced opening probability, or both, the activity of BK_Ca channels present in these cell was measured and analyzed. As shown in Fig.3, under symmetrical K⁺ (145 mM) conditions, the activity of BK_Ca channels can be observed in an excised inside-out patch. When the membrane patch was exposed to δ-HCH, the activity of channel opening was profoundly increased (Fig. 3). The opening probability of the channel measured at the level of +60 mV in the control (i.e., in the absence of δ-HCH) was found to be 0.015 ± 0.007 (n = 8). The addition of δ-HCH (30 μM) to the bath medium significantly increased the channel activity to 0.274 ± 0.015 (P < .01;n = 8). However, there was no significant difference in the amplitude of the unitary outward current between the absence and presence of δ-HCH [12.6 ± 1.2 pA (n = 8) versus 12.8 ± 1.4 pA (n = 8), P> .05]. Thus, it is clear that the presence of δ-HCH can increase the opening probability of BK_Ca channels in GH₃ cells.

Concentration-Dependent Stimulation of BK_CaChannels by δ-HCH.

The relationship between the concentration of δ-HCH and the opening probability of BK_Cachannels was further examined. These experiments were conducted with symmetrical K⁺ concentration, the inside-out configuration in which bath medium contained 0.1 μM Ca²⁺ was performed, and the holding potential was set at +60 mV. As shown in Fig. 4A, δ-HCH (3–300 μM) increased the channel activity in a concentration-dependent manner. The EC₅₀ value for δ-HCH-induced channel activity was 20 μM. In addition, the Hill coefficient was found to be 2.3, suggesting that there was a positive cooperativity for the stimulation of BK_Cachannels.

Effect of δ-HCH on the Activation Curve of BK_CaChannels.

Fig. 4B shows the activation curve of BK_Ca channels in the absence and presence of δ-HCH (30 μM). In these experiments, the activation curves of BK_Ca channels were obtained with the aid of the voltage ramp protocols. The ramp pulses were delivered from +20 to +140 mV with a duration of 1 s. The plots of opening probability of BK_Ca channels as a function of membrane potential were constructed and fit with Boltzmann function as described underMaterials and Methods. In control, n = 0.43 ± 0.03, V_1/2 = 95.3 ± 1.3 mV, and K⁻¹ = 11.9 ± 0.5 mV (n = 6), whereas in the presence of δ-HCH (30 μM),n = 1.10 ± 0.05, V_1/2= 81.2 ± 1.1 mV, and K⁻¹ = 12.0 ± 0.4 mV (n = 6). Thus, the presence of δ-HCH (30 μM) not only caused a 2.5-fold increase in the maximal opening probability of BK_Ca channels but also significantly shifted the activation curve to a less positive membrane potential by approximately 15 mV. However, there was no significant effect on the slope (i.e., K⁻¹) of the activation curve in the presence of δ-HCH. These results indicate that δ-HCH enhanced the activity of BK_Cachannels in a voltage-dependent fashion in GH₃cells.

Lack of Effect of δ-HCH on Single Channel Conductance of BK_Ca Channels.

It was examined whether δ-HCH affects the single channel conductance of BK_Ca channels. To construct the plots of current amplitude as a function of membrane potential, the voltage ramp pulses from +30 to +90 mV with a duration of 1 s were applied at a rate of 0.1 Hz. Figure 4C illustrates theI–V relationships of BK_Ca channels in the absence and presence of δ-HCH (10 and 30 μM). The single channel conductance of BK_Ca channels calculated from the linear I–V relationship in control (i.e., in the absence of δ-HCH) was 208 ± 8 pS (n = 12) with a reversal potential of 0 ± 1 mV (n = 12). The value of unitary conductance for these channels was found to be similar to that reported previously (Wu et al., 1999b) but not significantly different from that (209 ± 9 pS; P > .05,n = 10) measured in the presence of δ-HCH (30 μM). Thus, δ-HCH produced no significant change in the single channel conductance of BK_Ca channels, but enhanced the channel activity in these cells.

Effect of δ-HCH on Kinetic Behavior of BK_CaChannels.

Because it was observed that δ-HCH tended to prolong the open-time duration of BK_Ca channels, the effect of δ-HCH on the kinetic properties of BK_Ca channels was further characterized. As shown Fig. 5, in the absence of δ-HCH, the open-time histogram of BK_Ca channels at +60 mV could be fitted by a single-exponential curve with a mean open time of 1.9 ± 0.2 ms (n = 5). However, the presence of δ-HCH (10 μM) was found to increase the lifetime of the open state. A two-exponential function was thus needed to fit the open-time histogram obtained in the presence of δ-HCH (10 μM) (Fig. 5). When the membrane patches were exposed to δ-HCH (10 μM) intracellularly, the time constants for fast and slow components of open-time histogram were 1.9 ± 0.2 and 9.6 ± 0.5 ms, respectively (n = 5). Thus, δ-HCH can enhance the channel activity by increasing mean open time.

Comparison between the Effects of δ-HCH and Those of γ-HCH, 17β-Estradiol, Ryanodine, Dantrolene, IP₃, Ruthenium Red, and Paxilline.

Effects of γ-HCH, 17 β-estradiol, ryanodine, dantrolene, IP₃, ruthenium red, and paxilline on the activity of BK_Ca channels in GH₃ cells were also examined and compared. As shown in Fig.6, ryanodine (10 μM), dantrolene (10 μM), or IP₃ (10 μM) applied intracellularly had no significant effect on the channel activity. These compounds can affect Ca²⁺ release from intracellular Ca²⁺ stores. However, γ-HCH (10 μM) was found to suppress the activity of BK_Ca channels significantly. Likewise, ruthenium red (10 μM) or paxilline (1 μM) produce a profound inhibition of channel activity. Both ruthenium red and paxilline were reported to be a blocker of BK_Ca channels (Sanchez and McManus, 1996; Wu et al., 1999a). On the other hand, like δ-HCH, 17β-estradiol was also noted to enhance the activity of BK_Ca channels significantly.

Effect of δ-HCH and γ-HCH on Spontaneous Action Potentials in GH₃ Cells.

The effect of δ-HCH and γ-HCH on membrane potentials was also examined. Under the current-clamp conditions, GH₃ cells, bathed in normal Tyrode's solution containing 1.8 mM CaCl₂, had a resting membrane potential of −48 ± 7 mV (n = 26). The typical effects of δ-HCH and γ-HCH on spontaneous action potentials in these cells are illustrated in Fig. 7. About 70% of GH₃ cells were found to exhibit the repetitive firing of action potentials, which was Ca²⁺-sensitive and inhibited by tetrandrine, a blocker of Ca²⁺ channel blocker (Wu et al., 1999b). When cells were exposed to δ-HCH (30 μM), spontaneous spiking discharge was significantly decreased to 0.4 ± 0.1 Hz from a control value of 0.9 ± 0.2 Hz (P < .05,n = 8). Cells were also hyperpolarized to −53 ± 9 mV from a control value of −46 ± 8 mV (P < .05, n = 8). In contrast, the firing frequency of action potential was increased by the addition of γ-HCH (30 μM). The presence of γ-HCH (30 μM) increased the repetitive firing of action potentials from 0.9 ± 0.2 to 1.3 ± 0.3 Hz (P < .05, n = 7).

Stimulatory Effect of δ-HCH onI_K(Ca) in Neuroblastoma IMR-32 Cells.

Because I_K(Ca) or BK_Ca channels observed in GH₃ cells may be different from those in neurons, the effect of δ-HCH in neuroblastoma IMR-32 cells was also examined. As shown in Fig. 8, when cells were bathed in normal Tyrode's solution containing 1.8 mM CaCl₂ and the voltage pulses from 0 mV to various potentials ranging from +10 to +70 mV in 20-mV increments were applied, the addition of δ-HCH (30 μM) produced an increase in the amplitude of I_K(Ca) throughout the entire voltage-clamp step. For example, when the voltage pulses from 0 to +70 mV were evoked, 30 μM δ-HCH significantly increased the current amplitude to 432 ± 42 pA from a control value of 202 ± 35 pA (P < .05, n = 5).

To further characterize the effect of δ-HCH onI_K(Ca), the single channel experiments with an inside-out configuration were also performed in neuroblastoma IMR-32 cells. In these experiments, cells are bathed in symmetrical K⁺ concentration, the holding potential was +60 mV and bath medium contained 0.1 μM Ca²⁺. As shown in Fig. 9, the activity of intermediate-conductance K_Ca(IK_Ca) channels in these cells was observed. When clotrimazole (10 μM) was applied to the bath, the channel activity was significantly suppressed (data not shown). However, when the patch membrane was intracellularly exposed to δ-HCH (30 μM), the channel activity was greatly increased. The opening probability of IK_Ca channels measured at +60 mV in control was found to be 0.009 ± 0.001 (n = 5). After addition of 30 μM δ-HCH into the bath, the channel activity was significantly increased to 0.546 ± 0.008 (P < .01, n = 5). However, as shown in Fig. 9B, the single channel conductance of IK_Ca channels between the absence and presence of δ-HCH did not differ significantly [72 ± 2 pS (n = 5) versus 73 ± 2 pS (n = 5), P > .05]. These data indicate that δ-HCH is also capable of stimulating the activity of IK_Ca channels present in neuroblastoma IMR-32 cells.

Discussion

The results presented here show that: 1) in GH₃ lactotrophs, δ-HCH can enhance the amplitude of Ca²⁺-activated K⁺ current (I_K(Ca)); 2) δ-HCH does not affect the amplitude of voltage-dependent K⁺ current; however, it suppresses voltage-dependent L-type Ca²⁺ inward current; 3) δ-HCH stimulates the activity of BK_Ca channels in a concentration-dependent manner, but does not change single channel conductance; 4) δ-HCH shifts the activation curve of BK_Ca channels to a less positive potential; 5) the δ-HCH-mediated increase in the opening probability is mainly caused by an increase in the number of long-lived openings; and 6) δ-HCH also stimulates I_K(Ca) and enhances the activity of intermediate-conductance K_Ca(IK_Ca) channels in human neuroblastoma IMR-32 cells. This stimulatory action of BK_Ca and IK_Ca channels will cause membrane hyperpolarization, thus affecting the neuronal or neuroendocrine function, if the δ-HCH action in neurons or neuroendocrine cells in vivo is the same as those on these cells shown in this study.

Previous reports have shown that δ-HCH can modulate ryanodine-sensitive Ca²⁺ channels and stimulate Ca²⁺ release from ryanodine-sensitive Ca²⁺ stores (Pessah et al., 1992; Rosa et al., 1997a). However, in our study performed in the inside-out configuration, δ-HCH applied intracellularly can enhance the activity of BK_Ca channels. Ryanodine, dantrolene, or IP₃ caused no significant change in the channel activity as compared with the control data. Therefore, it is unlikely that the δ-HCH-mediated increase in the activity of BK_Ca observed in GH₃ cells results from an increase in intracellular Ca²⁺that is induced by Ca²⁺ release from internal stores, Ca²⁺ entry from the cell exterior, or both. Furthermore, the present finding, demonstrating that δ-HCH suppressed the amplitude of I_{Ca, L}, excludes the possibility that the effect of δ-HCH on BK_Ca channels depends on the increased Ca²⁺ influx, which is caused by the depolarizing stimuli that lead to the activation of I_{Ca, L}.

HCH (β-isomer) was reported to exert an estrogen-like effect in human breast cancer cells (Steinmetz et al., 1996). In our study, 17 β-estradiol (30 μM) did also stimulate the activity of BK_Ca, when it was applied intracellularly. This result is consistent with a previous study indicating the direct stimulation of BK_Ca channels in endothelial cells (Rusko et al., 1995). It remains to be clarified whether δ-HCH and 17 β-estradiol might act on the same recognition site to interact with the BK_Ca channels expressed in GH₃ cells. However, the effect of δ-HCH on BK_Ca channels might be direct and independent of its binding to estrogen receptors, because the single channel experiments were performed in an excised inside-out membrane patch. In addition, the present result showing that no significant effect of IP₃ on the activity of BK_Cachannels was found suggests that the action of δ-HCH does not appear to be relevant to its structural similarity to IP₃ (Mohr et al., 1995).

The δ-HCH concentration used to produce a neurodepressant effect was found to be close to its EC₅₀ value for the stimulation of BK_Ca channels (Pomes et al., 1994;Nagata and Narahashi, 1995; Aspinwall et al., 1997; Rosa et al., 1997a;Belelli et al., 1999). δ-HCH was also effective in stimulating the activity of IK_Ca channels in neuroblastoma IMR-32 cells. On the other hand, the EC₅₀ value for the potentiation of GABA-evoked currents in oocytes expressing the human α₃β₁γ_2Lsubunit combination and the mutant ρ_11307Sreceptor was 3.4 and 38 μM, respectively (Belelli et al., 1996,1999). Thus, there might be a link between the effects of δ-HCH on neurons and its stimulating effect on BK_Cachannels. The present experiments also found that δ-HCH is independent of the presence of internal Ca²⁺(data not shown). It is likely that δ-HCH does not exert its effect via an increase in the affinity of Ca²⁺ ions for the Ca²⁺ binding site in the membrane. However, in our study, the presence of δ-HCH could produce a shift of 15 mV to a less positive potential in the activation curve of BK_Ca channels. Therefore, δ-HCH can enhance the activity of BK_Ca channels in a voltage-dependent fashion, and its interaction with these channels would be dependent on the pre-existing level of membrane potential or the concentration of δ-HCH used.

A recent study (Silvestroni et al., 1997) showed that the γ-isomer of HCH produced membrane depolarization in human sperm. The present results showing the inhibitory effect of γ-HCH on the activity of BK_Ca channels can account for this finding. However, in our study, there is no evidence showing that γ-HCH can increase the amplitude of Ca²⁺ inward current in GH₃ cells. It is thus possible that the effects of γ-HCH on the cell viability or the c-fos expression (Vendrell et al., 1992; Tusell et al., 1994; Barron et al., 1995;Silvestroni et al., 1997) are related to its indirect stimulation of Ca²⁺ channels that can be evoked by membrane depolarization.

Of interest, our data demonstrated that HCHs showed the stereospecificity in their interactions with BK_Cachannels. Unlike δ-HCH, γ-HCH was found to suppress the activity of BK_Ca channels. Indeed, a number of studies have demonstrated that γ-HCH suppressed the amplitude of GABA-induced current, whereas δ-HCH enhanced it in cortical neurons (Pomes et al., 1994), in rat dorsal root ganglion neurons (Nagata and Narahashi, 1995), and in a human embryonic kidney cell line in which GABA receptor subunits were expressed (Nagata et al., 1996). It was also reported that the cytotoxic or cardiostimulatory effects of δ-HCH and γ-HCH could result from the differential mechanisms through which these two agents act on the Ca²⁺ release from internal stores (Pessah et al., 1992; Rosa et al., 1997a,b). More importantly, in addition to its interaction with the GABA_Areceptor Cl⁻ channel complex (Bloomquist, 1992;Cristofol and Rodriguez-Farre, 1993; Pomes et al., 1994; Narahashi, 1996; Narahashi et al., 1998), γ-HCH may produce an inhibitory effect on BK_Ca channels. This effect might also contribute to its action on the reduction of noradrenaline release (Cristofol and Rodriguez-Farre, 1993, 1994), given that there would be an increase in hormonal secretion in the presence of BK_Ca channel blockers.

Previous reports have shown that δ-HCH may disorganize the lipid bilayer in erythrocytes (Verma and Singhal, 1991; Bhalla and Agrawal, 1998), in testicular plasma membrane (Srivastava et al., 1995), and in human sperm (Silvestroni et al., 1997). Similarly, in our study, δ-HCH at a concentration of 100 μM produced an initial large increase in channel activity in inside-out patches; however, a disruption of the membrane and loss of membrane patch always accompanied this. Furthermore, the stimulation ofI_K(Ca) by δ-HCH was found to be slowly developing and not easy to fully wash out. These observations could be interpreted to mean that this compound might be able to partition into the membrane to produce its actions. The lipophilic nature of δ-HCH seemed to explain the present finding that the degree of reversibility of δ-HCH was time-dependent. Furthermore, the present results demonstrated that δ-HCH applied intracellularly produced a fraction of channel openings to shift to longer-lived openings, resulting in two open kinetic states. It would be of interest to determine whether the lipophilicity of δ-HCH or its effect on membrane disorganization is related to its prolongation in open-time duration of BK_Ca channels.

In our study, a steep Hill slope of 2.3 for the δ-HCH-stimulated activity of BK_Ca channels was found. This result suggests that the binding of more than one molecule is required for its stimulatory effect on the BK_Ca channel activity. Previous reports have demonstrated that the δ-HCH is a potent positive allosteric modulator of GABA-evoked currents (Belelli et al., 1996). The Hill coefficient for the δ-HCH-mediated potentiation of GABA-evoked currents recorded from oocytes expressing the human α₃β₁γ_2Lsubunit combination or the splice variant of the Rdl subunit was about 4 (Belelli et al., 1996). However, δ-HCH was recently found to have no effect on GABA-evoked currents in oocytes expressing the wild type ρ₁ receptor (Belelli et al., 1999). In the present study, because we measured the activity of single channel current in the inside-out configuration, it is possible that δ-HCH interacts with the channel protein per se. It thus remains to be clarified whether, in addition to binding to the distinct sites on the GABA_A receptor protein, δ-HCH directly regulates the channel protein of the GABA_Areceptor Cl⁻ channel complex, although δ-HCH alone did not induce an inward current (Belelli et al., 1996).

In summary, the present study provides evidence that δ-HCH induced the change in the activity of BK_Ca channels in GH₃ cells. This finding will be of great help in the study of the underlying mechanisms through which HCHs interact with BK_Ca or IK_Ca channels expressed in neurons or neuroendocrine cells.