Stimulatory Actions of Caffeic Acid Phenethyl Ester, a Known Inhibitor of NF- (cid:1) B Activation, on Ca 2 (cid:2) -activated K (cid:2) Current in Pituitary GH 3 Cells*

Caffeic acid phenethyl ester (CAPE), a phenolic antioxidant derived from the propolis of honeybee hives, is known to be an inhibitor of activation of nuclear transcript factor NF- (cid:1) B. Its effects on ion currents have been investigated in pituitary GH 3 cells. This compound increased Ca 2 (cid:2) -activated K (cid:2) current ( I K(Ca) ) in a concentration-dependent manner with an EC 50 value of 14 (cid:3) 2 (cid:4) M . However, the magnitude of CAPE-induced stimulation of I K(Ca) was attenuated in GH 3 cells preincubated with 2,2 (cid:1) -azo-bis-(2-amidinopro-pane) hydrochloride (100 (cid:4) M ) or t -butyl hydroperoxide (1 m M ). CAPE (50 (cid:4) M ) slightly suppressed voltage-de-pendent L-type Ca 2 (cid:2) current. In inside-out configuration, CAPE (20 (cid:4) M ) applied to the intracellular face of the detached patch enhanced the activity of large conductance Ca 2 (cid:2) -activated K (cid:2) (BK Ca ) channels with no modification in single-channel conductance. After BK Ca channel activity was increased by CAPE (20 (cid:4) M ), subsequent application of nordihydroguaiaretic acid (20 (cid:4) The between the potentials and the probability of channel and after the application of fitted a function of the where n the maximal open probability, the is is half-maximal activation, and k is the slope factor of the activation curve ( i.e. the voltage dependence of the activation process in mV per e -fold change). The averaged results are presented as the mean values (cid:7) S.E. The paired Student’s t test was used for the statistical analyses. To further clarify the statistical difference among the two or four treatment groups, analyses of variance with Duncan’s multiple range test for multiple comparisons were also performed. Differences between values were considered significant when p (cid:8) 0.05.

matory reaction in brain (6). Several lines of evidence also indicate that CAPE may modify the redox state in transformed fibroblast cells and in leukemic HL-60 cells (7)(8)(9). Furthermore, it has been reported that this compound inhibited the contractile response to phenylephrine or to high K ϩ solution in isolated rat thoracic aorta (10). However, to our knowledge, the effects of CAPE on ion currents have not been thoroughly studied.
Large conductance Ca 2ϩ -activated K ϩ (BK Ca ) channels play important roles in controlling the excitability of nerve, muscle, and other cells by stabilizing cell membrane at negative potentials (11). Their gating is known to be controlled by intracellular Ca 2ϩ and/or membrane depolarization. The challenging of cells with oxidizing agents has been found to suppress the channel activity of these channels (12). Pituitary GH 3 cells have been demonstrated to exhibit the activity of these channels (13). Riluzole and ciglitazone, both of which were reported to prevent neuronal injuries, could enhance the activity of BK Ca channels functionally expressed in these cells (14,15). Importantly, the opener of these channels has been shown to counteract the deleterious effects of excitatory neurotransmitters following neurotoxic or ischemic injuries (16). Previous studies also revealed that the BK Ca channel might be a relevant target of DNA synthesis in cultured Mü ller glial cells (17).
Therefore, the objective of this study was to (a) address the question of whether CAPE could affect Ca 2ϩ -activated K ϩ currents (I K(Ca) ) in GH 3 cells; (b) determine the effects of this compound on the activity of BK Ca channels; and (c) examine whether it can influence the membrane potential. Interestingly, the present results indicate that in GH 3 lactotrophs, CAPE does not appear to affect the activation of NF-B exclusively, despite its ability to inhibit NF-B activation in these cells (18). The CAPE-induced increase in BK Ca channel activity may account, at least in part, for its effects on cellular functions in neurons or neuroendocrine cells.

MATERIALS AND METHODS
Cell Culture-The clonal strain GH 3 cell line, originally derived from a rat anterior pituitary adenoma, was obtained from the Culture Collection and Research Center (CCRC-60015; Hsinchu, Taiwan). The detailed methods have been previously described (19). Briefly, the cells were cultured in Ham's F-12 medium (Invitrogen) supplemented with 15% heat-inactivated horse serum (v/v), 2.5% fetal calf serum (v/v), and 2 mM L-glutamine (Invitrogen) in a humidified environment of 5% CO 2 /95% air. The experiments were generally performed 5 or 6 days after cells were subcultured (60 -80% confluence).
Electrophysiological Measurements-Immediately before each experiment, the cells were dissociated, and an aliquot of cell suspension was transferred to a recording chamber positioned on the stage of an inverted microscope (DM IL; Leica, Wetzlar, Germany). The cells were bathed at room temperature (20 -25°C) in normal Tyrode's solution containing 1.8 mM CaCl 2 . The recording pipettes were pulled from Kimax-51 capillaries (Kimble Glass, Vineland, NJ) using a two-stage 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 measured in the cell-attached, inside-out, and whole cell configurations of the patch-clamp technique, using an RK-400 patchclamp amplifier (Bio-Logic, Claix, France) (19).
Data Recording and Analysis-The signals were displayed on an analog/digital oscilloscope (HM 507; Hameg, East Meadow, NY) and on a liquid crystal projector (PJ550-2; ViewSonic, Walnut, CA). The data were stored in a Pentium III grade laptop computer (Slimnote VX 3 ; Lemel, Taipei, Taiwan) at 10 kHz through a Digidata 1322A interface (Axon Instruments, Union City, CA). This device was controlled by a commercially available software (pCLAMP 9.0; Axon Instruments). The currents were low pass filtered at 1 or 3 kHz. Ion currents obtained during whole cell experiments were stored without leakage correction and analyzed using the pCLAMP 9.0 software (Axon Instruments) or the Origin 6.0 software (Microcal, Northampton, MA).
To calculate the percentage of stimulation of CAPE on I K(Ca) , each cell was depolarized from 0 to ϩ50 mV, and current amplitude during cell exposure to CAPE was measured and compared. The amplitude of I K(Ca) in the presence of this compound at a concentration 200 M was taken as 100%. The concentration of CAPE required to increase 50% of current amplitude was then determined using a Hill function, y ϭ (E max ϫ is the CAPE concentration, EC 50 is the concentration required for a 50% increase; n h is the Hill coefficient, and E max is the CAPE-induced maximal increase in the amplitude of I K(Ca) .
The amplitudes of single BK Ca channel currents were determined by fitting Gaussian distributions to the amplitude histograms of the closed and the open state. The channel open probability in a patch was expressed as N⅐P o , which can be estimated using the following equation: where N is the number of active channels in the patch, A 0 is the area under the curve of an all points histogram corresponding to the closed state, and A 1 . . . . A n represent the histogram areas reflecting the levels of distinct open state for 1 to n channels in the patch.
The relationships between the membrane potentials and the probability of channel openings obtained before and after the application of CAPE (20 M) were fitted with a Boltzmann function of the form: where n P ϭ the maximal open probability, V ϭ the membrane potential in mV, V1 ⁄2 is the voltage at which there is half-maximal activation, and k is the slope factor of the activation curve (i.e. the voltage dependence of the activation process in mV per e-fold change).
The averaged results are presented as the mean values Ϯ S.E. The paired Student's t test was used for the statistical analyses. To further clarify the statistical difference among the two or four treatment groups, analyses of variance with Duncan's multiple range test for multiple comparisons were also performed. Differences between values were considered significant when p Ͻ 0.05.
The composition of normal Tyrode's solution was 136.5 mM NaCl, 5.4 mM KCl, 1.8 mM CaCl 2 , 0.53 mM MgCl 2 , 5.5 mM glucose, and 5.5 mM HEPES-NaOH buffer, pH 7.4. To record the K ϩ currents or membrane potential, the recording pipette was backfilled with a solution consisting of 140 mM KCl, 1 mM MgCl 2 , 3 mM Na 2 ATP, 0.1 mM Na 2 GTP, 0.1 mM EGTA, and 5 mM HEPES-KOH buffer, pH 7.2. The free Ca 2ϩ concentration of this solution was estimated to be 230 nM, assuming that the residual contaminating Ca 2ϩ concentration was 70 M, and the ratiometric fura-2 measurement with F-2500 fluorescence spectrophotometer (Hitachi, Tokyo, Japan) showed that this solution contained 205 Ϯ 12 nM free Ca 2ϩ for three different experiments. To measure voltagedependent Ca 2ϩ current, KCl inside the pipette solution was replaced with equimolar CsCl, and pH was adjusted to 7.2 with CsOH, whereas bathing solution contained 1 M tetrodotoxin and 10 mM tetraethylammonium chloride.
For single-channel current recordings, the high K ϩ bathing solution contained 145 mM KCl, 0.53 MgCl 2 and 5 mM HEPES-KOH buffer, pH 7.4, and the pipette solution contained 145 mM KCl, 2 mM MgCl 2 , and 5 mM HEPES-KOH buffer, pH 7.2. The value of free Ca 2ϩ concentration was calculated assuming a dissociation constant for EGTA and Ca 2ϩ (at pH 7.2) of 0.1 M. To provide 0.1 M free Ca 2ϩ in bath solution, 0.5 mM CaCl 2 and 1 mM EGTA were added. 3 Cells-In the first series of experiments, the whole cell configuration of the patch-clamp technique was used to investigate the effect of CAPE on ion currents in these cells. The cells were bathed in normal Tyrode's solution containing 1.8 mM CaCl 2 , and the pipette solution contained 0.1 mM EGTA and 3 mM ATP. ATP (3 mM) included in the pipette solution was effective at suppressing ATP-sensitive K ϩ channels (20). To inactivate other types of voltage-dependent K ϩ currents, each cell was held at the level of 0 mV. As illustrated in Fig. 1, when the cell was held at 0 mV and different potentials ranging from Ϫ10 to ϩ60 mV with 10-mV increments were applied, a family of large, noisy, outward currents was elicited. These outward currents have been previously identified as I K(Ca) (19). Interestingly, within 1 min of exposing the cells to CAPE (20 M), the amplitude of outward currents was greatly increased throughout the entire range of voltage-clamp step. For example, when the cells were depolarized from 0 to ϩ50 mV, current amplitudes measured at the end of the depolarizing pulses were increased to 552 Ϯ 36 pA from a control of 228 Ϯ 25 pA (p Ͻ 0.05; n ϭ 8). The relationship between the CAPE concentration and the percentage increase of I K(Ca) has been constructed (Fig. 1C). This compound could increase the amplitude of I K(Ca) in a concentration-dependent manner with an EC 50 value of 14 Ϯ 2 M. At a concentration of 200 M, it fully increased I K(Ca) . The Hill coefficient was found to be 1.8, suggesting that there was a positive cooperation for its stimulation of I K(Ca) . These results indicate that CAPE can produce a stimulatory action on I K(Ca) in these cells.

Effect of CAPE on Ca 2ϩ -activated K ϩ Current (I K(Ca) ) in GH
Effect of CAPE on Voltage-dependent L-type Ca 2ϩ Current (I Ca,L ) in GH 3 Cells-I K(Ca) can be functionally coupled with Ca 2ϩ influx through plasmalemmal voltage-dependent Ca 2ϩ channels (21). A recent report also demonstrated that the action of CAPE on vasorelaxation in rat thoracic aorta could be due to the blockade of Ca 2ϩ movement through the cell membrane (10). For these reasons, we further investigated whether it could exert any effect on I Ca,L that was previously described in these cells (21,22). These experiments were conducted with a Cs ϩ -containing solution. The exposure to CAPE (20 M) was found to be have little or no effect on I Ca,L . However, this compound at a concentration of 50 M slightly suppressed I Ca,L , although it did not modify the I-V relationship of I Ca,L (Fig. 2). For example, CAPE (50 M) decreased the amplitude of I Ca,L to 42 Ϯ 3 pA from a control value of 51 Ϯ 6 pA (p Ͻ 0.05; n ϭ 7), when cells were depolarized from Ϫ50 to 0 mV. Therefore, this compound stimulated I K(Ca) in a manner conceivably unlikely to be linked to an increase in the amplitude of I Ca,L .
Effect of CAPE on I K(Ca) in Cells Preincubated with 2,2Ј-Azobis(2-aminopropane) Dihydrochloride AAPH or t-Butyl Hydroperoxide-CAPE is known to be an antioxidant flavonoid (1,23). Therefore, we next evaluated whether changes in reactive oxygen species can influence CAPE-induced stimulation of I K(Ca) in GH 3 cells. Interestingly, the results showed that CAPE-stimulated I K(Ca) was attenuated in GH 3 cells preincubated with either 100 M AAPH or 1 mM t-butyl hydroperoxide (Fig. 3). t-Butyl hydroperoxide is an oxidative agent, whereas AAPH is known to be an azo compound that can generate free radicals (24). A subsequent application of dithiothreitol (10 M) increased I K(Ca) in cells treated with AAPH or t-butyl hydroperoxide. When the AAPH-treated cells were depolarized from 0 to ϩ50 mV, CAPE (200 M) increased the density of I K(Ca) by about 15%. Conversely, in control cells, CAPE (200 M) nearly fully increased the density of these currents. These results suggest that the stimulation of I K(Ca) caused by this compound can be modified in the presence of these oxidizing agents.
Effect of CAPE on the Activity of BK Ca Channels in GH 3 Cells-The results from our whole cell experiments suggest that I K(Ca) may be K ϩ flux through the BK Ca channel (13,19), because CAPE-induced increase in I K(Ca) was suppressed by paxilline yet not by glibenclamide or apamin. To elucidate how it could act to affect I K(Ca) , the effect of this compound on BK Ca channels was further investigated. In these experiments, the single-channel recordings with inside-out configuration were performed in symmetrical K ϩ concentration (145 mM). The bath solution contained 0.1 M Ca 2ϩ , and the potential was held at ϩ60 mV. As shown in Fig. 4, the activity of BK Ca channels could be readily observed in an excised patch. An increase in channel activity could also be obtained in cellattached patches when cells were exposed to ionomycin (10 M) or squamocin (10 M). These two agents were previously reported to be Ca 2ϩ ionophores (25). When CAPE (20 M) was applied to the intracellular face of the detached patch, the channel open probability was increased (Fig. 4). The open probability obtained at the level of ϩ60 mV in the control was 0.112 Ϯ 0.005 (n ϭ 6). The application of CAPE (20 M) significantly increased channel activity to 0.289 Ϯ 0.035 (p Ͻ 0.05; n ϭ 6). When this compound was washed out, the open probability returned to the control level. However, the single-channel amplitude remained unaltered in the presence of 20 M CAPE (Fig. 4C). Moreover, curcumin (20 M) applied to the intracellular face of the excised patch was not found to have any effects on the probability of channel openings, whereas cilostazol (20 M) could increase the channel activity effectively. Similar to CAPE, curcumin has been shown to inhibit the activation of NF-B (26). Cilostazol has been recently found to stimulate I K(Ca) in human neuroblastoma SK-N-SH cells (27).
Effect of Nordihydroguaiaretic Acid on BK Ca Channels in GH 3 Cells-Nordihydroguaiaretic acid was previously reported to stimulate BK Ca channels (28). We also examined whether the stimulatory effects of CAPE and nordihydroguaiaretic acid on these channels are additive. Interestingly, as shown in  (14). Taken together, the results indicate that the stimulatory effects of CAPE and nordihydroguaiaretic acid on a single BK Ca channel are not additive in GH 3 cells.
Lack of Effect of CAPE on Single-channel Conductance of BK Ca Channels-In the next series of experiments, the effect of CAPE on BK Ca single-channel conductance was investigated. In inside-out configuration, the cells were bathed in symmetrical K ϩ concentration (145 mM), and the bath solution contained 0.1 M Ca 2ϩ . Fig. 5 (C and D) illustrates the I-V relationships of BK Ca channels obtained in the absence and presence of CAPE (20 M). The single BK Ca channel conductance calculated from a linear I-V relationship in control was 196 Ϯ 12 pS (n ϭ 11) with a reversal potential of 0 Ϯ 3 mV (n ϭ 11). Notably, the value of single-channel conductance did not differ from that (197 Ϯ 11 pS; p Ͼ 0.05, n ϭ 10) obtained in the presence of CAPE (20 M). These results indicate that CAPE causes no modification in single-channel conductance, despite its ability to increase the channel open probability.  Fig. 5E shows the activation curve of BK Ca channels in the absence and presence of CAPE (20 M). The plot of open probability of BK Ca channels as a function of membrane potential was fitted with a Boltzmann function as described under "Materials and Methods." In control, n P ϭ 0.35 Ϯ 0.04, V1 ⁄2 ϭ 75.4 Ϯ 1.6 mV, and k ϭ 10.7 Ϯ 0.4 mV (n ϭ 6), whereas in the presence of CAPE (20 M), n P ϭ 0.71 Ϯ 0.07, V1 ⁄2 ϭ 61.2 Ϯ 1.9 mV, and k ϭ 10.9 Ϯ 0.6 mV (n ϭ 6). The data showed that the activation curve was shifted along the voltage axis to less positive potentials in the presence of CAPE. In contrast, no significant change in the slope (i.e. k value) of the activation curve was detected in the presence of this compound. Taken together, these results indicate that CAPE applied to the intracellular surface of the channel is capable of increasing the open probability in a voltage-dependent fashion. 3 Cells-Whether the CAPE-induced increase in the activity of these channels is associated with internal Ca 2ϩ concentration was also studied. In these experiments, when an excised membrane patch was formed, various concentrations of Ca 2ϩ in the bath before and during exposure to CAPE (20 M) were applied. As shown in Fig. 5F, the stimulatory effect of CAPE on BK Ca channel activity was exposed to CAPE (20 M), the membrane became hyperpolarized, and the repetitive firing of action potentials was gradually reduced (Fig. 7, A and B). CAPE (20 M) decreased the firing frequency from 1.05 Ϯ 0.08 to 0.36 Ϯ 0.05 Hz (p Ͻ 0.05; n ϭ 6). Paxilline (1 M), a known blocker of BK Ca channels, reversed the CAPE-induced decrease of firing frequency to 0.86 Ϯ 0.07 Hz (p Ͻ 0.05; n ϭ 5). Thus, it is clear that this compound can regulate the firing of action potentials in these cells.

Effect of Internal Ca 2ϩ Concentration on CAPE-stimulated BK Ca Channel Activity in GH
Effect of CAPE on I K(Ca) That Is Active in Normal Action Potential Waveforms-To determine whether CAPE affects I K(Ca) that is active during normal action potentials, each cell was held at Ϫ50 mV, and the ramp hyperpolarization pulses from ϩ20 to Ϫ50 mV with a duration of 100 ms at a rate of 0.05 Hz were delivered to mimic action potential-like waveforms of GH 3 cells (22). As shown in Fig. 7 (C and D), when cells were bathed in normal Tyrode's solution containing 1.8 mM CaCl 2 , the current traces representing the I-V relationships of I K(Ca) were observed in response to a voltage ramp protocol ranging from ϩ20 to Ϫ50 mV. The application of CAPE (20 M) increased peak outward currents from 203 Ϯ 15 to 506 Ϯ 34 pA (p Ͻ 0.05; n ϭ 6). A subsequent application of paxilline (1 M) could decrease the CAPE-stimulated I K(Ca) from 506 Ϯ 34 to 312 Ϯ 26 pA (p Ͻ 0.05; n ϭ 6). Thus, consistent with its inhibition of spontaneous action potentials, the results indicate that CAPE can increase I K(Ca) that is active during normal action potentials. DISCUSSION This study shows that CAPE (a) increases the amplitude of I K(Ca) in a concentration-dependent manner in pituitary GH 3 cells, (b) enhances the activity of BK Ca channel in a voltagedependent manner, and (c) reduces the repetitive firing of action potentials. This compound increased the probability of these channels in a mechanism unlikely to be linked to its inhibition of activation of NF-B. The stimulation by CAPE of I K(Ca) conceivably could be one of the mechanisms underlying CAPE-induced actions, if similar results occur in neurons or neuroendocrine cells in vivo.
The EC 50 value of CAPE required for the stimulation of I K(Ca) was 14 M in this study. This value is similar to that required for the inhibition of NF-B activation (2). Therefore, there might be a link between the actions of CAPE on neurons or neuroendocrine cells and its observed effects on ion channels, although further experiments are required to find out whether CAPE can interact with the BK Ca channel to influence I K(Ca) in other types of cells. However, a lack of effect of glibenclamide excludes the involvement of activation of ATP-sensitive K ϩ channel in CAPE-stimulated I K(Ca) in these cells.
A previous study showed that activation of NF-B could be accompanied by a decrease in the current density of smooth muscle L-type Ca 2ϩ channel (29). However, the present results indicate that the CAPE-induced increase in I K(Ca) does not depend on the increased availability of intracellular Ca 2ϩ resulting from the enhanced Ca 2ϩ influx through voltage-de- The parentheses shown with each bar indicates the number of cells examined. *, significantly different from control. **, significantly different from CAPE alone group. In C, the cell was held at the level of Ϫ50 mV, and the ramp pulses from ϩ20 to Ϫ50 mV with a duration of 100 ms were applied to mimic the shape and duration of normal action potential waveforms present in GH 3  pendent Ca 2ϩ channels, because it was not found to increase the amplitude of I Ca,L . These observations are compatible with a recent report showing the inability of CAPE to alter the decrease in intracellular Ca 2ϩ induced by low K ϩ solution in cerebellar granule cells (30). Our results demonstrating that this compound at a concentration of 50 M produced a slight reduction in the amplitude of I Ca,L can also account for its ability to inhibit high K ϩ -induced vasoconstriction in isolated rat aorta (10).
It has been demonstrated that dithiothreitol could stimulate I K(Ca) in GH 3 cells (31). In AAPH-treated cells, the stimulatory effect of CAPE on I K(Ca) was attenuated, and a subsequent application of dithiothreitol effectively increased the amplitude of I K(Ca) . These results suggest that the sulfhydryl oxidizing and reducing agents can produce an effect on I K(Ca) in GH 3 cells. It will be interesting to determine to what extent the decreased production of reactive oxygen species caused by CAPE affects the stimulatory effect of CAPE on I K(Ca) , because this compound is known to be a potent flavonoid antioxidant (1,23). Moreover, it seems likely that a decrease in the production of reactive oxygen species caused by CAPE is upstream of its stimulation of I K(Ca) . Direct activation of BK Ca channels and indirect inhibition of production of reactive oxygen species may synergistically contribute to the underlying cellular mechanisms through which this compound modifies the repetitive firing of these cells. In addition, like NS004 (32), CAPE was found to increase Ca 2ϩ sensitivity of BK Ca channels observed in GH 3 cells. Its ability to increase Ca 2ϩ sensitivity of BK Ca channels suggests that the CAPE molecule may modify the cysteine residues near the carboxyl-terminal, Ca 2ϩ bowl domain of these channels (33).
Our study demonstrated that CAPE could not modify singlechannel conductance of BK Ca channels, but it did increase the channel open probability. The increase in the amplitude of I K(Ca) caused by CAPE is primarily thought to be a result of a decrease in mean closed time. It was also seen that CAPE shifted the activation curve of BK Ca channels to the left with no modification in the slope factor of this curve. This compound thus appears to produce the stimulation of BK Ca channels by a direct effect on the channel or closely associated site, although the precise mechanisms of its action remain to be further elucidated. However, our data demonstrated that CAPE applied to the intracellular face of the excised patch produced a fraction of channel closings to shift to short-lived closings, resulting in one closed kinetic state.
It is worth mentioning that unlike the molecules of NS004, NS1619, or riluzole, the CAPE molecule has the juxtaposition of two aromatic rings, the unique structure of which is similar to those of some BK Ca channel openers, such as nordihydroguaiaretic acid and resveratrol (Fig. 8) (34). The present results demonstrated that the stimulatory effect of CAPE and nordihydroguaiaretic acid on the BK Ca channel was not additive. It is thus tempting to speculate that these two compounds, which are structurally related, may interact with the same binding site in the channel.
CAPE has been recently reported to induce the release of cytochrome c from mitochondria to cytosol in C6 glioma cells (35). Cytochrome c was found to activate K ϩ channels (36). However, in inside-out configurations, we showed that CAPE applied to the intracellular face of the excised patches enhanced BK Ca channel activity. It is unlikely that the ability of CAPE to increase the amplitude of I K(Ca) is primarily due to the release of cytochrome c from mitochondria.
Curcumin, another inhibitor of NF-B activation, was not found to have effects on BK Ca channels, when it was applied to intracellular face of the excised patch. The results lead us to suggest that CAPE could induce the change in the activity of BK Ca channels in GH 3 cells in a mechanism unlikely to be linked to its inhibition of NF-B activation. However, its change in membrane potential can be explained by the stimulatory effect on these channels. Such an effect may be responsible for its actions on neurons or neuroendocrine cells in vivo, despite the ability of this compound to inhibit activation of NF-B in GH 3 cells (18). Furthermore, CAPE and other structurally related compounds seem to be intriguing pharmacological tools used to characterize the properties of the BK Ca channels. Elucidation of the structure of the binding site for CAPE or other structurally related compounds might provide a structural basis for the pharmacological modulation of BK Ca channels.