INTRODUCTION

Top Abstract Introduction Methods Results Discussion References

INTRACELLULAR CALCIUM mobilization plays a central role in coupling the activation of muscarinic receptors with the regulation of cellular function in a variety of cells, including exocrine gland cells (2, 6, 7, 28-30). In exocrine gland cells, such as those from salivary glands, muscarinic receptor stimulation leads to a biphasic change in cytosolic Ca²⁺ concentration ([Ca²⁺]_i), with an initial rapid transient increase due to internal Ca²⁺ release and a lower more sustained increase primarily due to Ca²⁺ influx (1, 10, 24, 26, 28-32, 34, 35). The increase in [Ca²⁺]_i results in the activation of various ion channels; these include K⁺ and Cl channels, some nonselective cation channels, and ion transporters such as the Na⁺-K⁺-Cl cotransporter (7, 14, 23, 28-30). These studies have suggested that, although a transient activation of these ion channels can be achieved by internal Ca²⁺ release, their sustained activation is dependent on Ca²⁺ influx from the extracellular medium. Ca²⁺ influx in exocrine gland cells is primarily mediated via a store-operated Ca²⁺ influx pathway (24, 26, 31, 34) believed to be localized in the basolateral plasma membrane of these cells (10, 22, 25, 26). Other Ca²⁺ influx pathways might also exist such as receptor-operated pathways (20) or nonspecific cation channels (7, 28, 29). Ca²⁺-activated K⁺ channels (K_Ca) have also been proposed to be localized in the basolateral plasma membrane of exocrine gland cells, and, because of the tight regulation by [Ca²⁺]_i, the K_Ca activity has been used to monitor the changes in [Ca²⁺]_i in the subplasma membrane region (7, 10, 12, 14, 23, 28, 29).

The HSG cell line is a cloned cell line from the human submandibular gland and has been widely used as a model to study receptor-mediated signaling and salivary gland pathology (15, 20, 27, 35). A numbers of ion channels have been found in HSG cells, including a hypotonically activated Cl channel (17), an outwardly rectifying Cl channel (9), and a K_Ca (18). It was suggested in an earlier report that activation of the muscarinic receptor causes an increase in [Ca²⁺]_i that in turn activates a K_Ca (18). The channel was identified to be either of the large (BK) or intermediate (IK) conductance type on the basis of its sensitivity to charybdotoxin (ChTX) and quinine but relative insensitivity to tetraethylammonium and apamin. Furthermore, simultaneous measurements of intracellular Ca²⁺ and K⁺ current demonstrated that agonist-induced K⁺ current was very tightly correlated with changes in [Ca²⁺]_i in HSG cells (18). In general, these characteristics are largely similar to those of K⁺ channels in a variety of exocrine gland cells such as salivary and lachrymal, but not rodent pancreatic, acinar cells. In addition, HSG cells also have muscarinic receptor-stimulated Ca²⁺ signaling mechanisms similar to those seen in exocrine acinar cells. Stimulation of these cells with the muscarinic agonist carbachol (CCh) induces a biphasic increase in [Ca²⁺]_i, which is dependent on inositol 1,4,5-trisphosphate (IP₃)-induced intracellular Ca²⁺ release and Ca²⁺ influx. It was previously reported that HSG cells have two types of Ca²⁺ influx pathways: a large component that is dependent on internal Ca²⁺ store depletion, i.e., store-operated Ca²⁺ influx, and a relatively minor component that is dependent on muscarinic receptor activation, likely via a G protein (20). Our studies showed that CCh-stimulated [Ca²⁺]_i elevation in thapsigargin (TG)-treated cells, i.e., via store-independent Ca²⁺ influx pathway, did not induce further hyperpolarization of HSG cells, i.e., via activation of the K⁺ channel. On the basis of these data, we suggested that the sustained hyperpolarization in CCh-stimulated HSG cells is primarily regulated by the store-operated Ca²⁺ influx pathway.

In this study, we have examined the role of intracellular Ca²⁺ release and store-operated Ca²⁺ influx in the regulation of the K⁺ channel in HSG cells by CCh. By using the standard patch-clamp whole cell technique, we show that the channel activation is dependent on CCh-stimulated intracellular Ca²⁺ release, via IP₃-sensitive channels, and that its sustained activation is determined by Ca²⁺ influx, via the store-operated Ca²⁺ influx pathway. Importantly, we have examined K_Ca activity during the Ca²⁺ influx that occurs during reloading of internal Ca²⁺ stores, i.e., in the absence of internal Ca²⁺ release. The results show that Ca²⁺ influx alone cannot support activation of the K_Ca because of the rapid buffering of [Ca²⁺]_i in the subplasma membrane region by the activity of the intracellular Ca²⁺ pump. Thus we suggest that the Ca²⁺ influx-dependent modulation of K_Ca activity in CCh-stimulated HSG cells is not directly due to an elevation of [Ca²⁺]_i at the site of Ca²⁺ influx but rather is mediated via uptake of Ca²⁺ into the intracellular Ca²⁺ store and IP₃-dependent release.

	METHODS

Top Abstract Introduction Methods Results Discussion References

Cell culture. HSG cells were a gift from Dr. Mitsunobu Sato of the Second Department of Oral and Maxillofacial Surgery, Tokushima University, Tokushima, Japan. Cells were grown in Eagle's minimum essential medium with Earle's balanced salt solution (Biofluids, Rockville, MD) with 5% CO₂ in air at 37°C in the presence of 10% fetal calf serum, 2 mM L-glutamine, 100 U/ml penicillin, and 100 µg/ml streptomycin (all from Biofluids). Cells were fed three times a week and passaged when confluent. Cells were passaged by detaching them from the tissue culture dish with 0.25% trypsin-1.0 mM EDTA (Biofluids). A single cell suspension was reseeded on coverslips, kept in a 35-mm culture dish (Corning), and cultured for 24 h before use.

Patch-clamp experiments. The coverslips were cut to ~0.5 × 0.5 mm and placed in a perfusion chamber (Warner Instrument, Hamden, CT). The perfusion rate, ~5 ml/min, was achieved by gravity-fed plastic tubes in a bath solution that was continuously and simultaneously removed through a vacuum line. Complete solution changes were obtained within 15 s. The standard extracellular solution contained (in mM) 145 NaCl, 5 KCl, 1 MgCl₂, 1 CaCl₂, 10 glucose, 0.1 EGTA, and 5 HEPES, pH 7.4. The pipette was filled with (in mM) 150 KCl, 2 MgCl₂, 1 ATP, and 5 HEPES, pH 7.2. In some experiments, 150 mM KCl was replaced with 150 mM CsCl, and, in others, either 10 mM 1,2-bis(2-aminophenoxy)ethane-N,N,N',N'-tetraacetic acid (BAPTA) or 10-100 µM IP₃ was included in the pipette solution.

Ca²⁺ measurements. The fluorometric system used for intracellular Ca²⁺ measurement using indo 1 (Molecular Probes, Eugene, OR) has been described previously (19). Briefly, a single indo 1-loaded HSG cell was excited at 355 nm. The fluorescence emissions at 410 and 485 nm (F₄₁₀ and F₄₈₅, respectively) were measured simultaneously using two photomultipliers. The output from each photomultiplier was digitized at 2 Hz. [Ca²⁺]_i was calculated with the use of the F₄₁₀ / F₄₈₅ emission ratio by a custom-designed program using a calibration curve based on different Ca²⁺ buffer solutions. The iris diaphragm was set to a small field immediately covering a single HSG cell so that the fluorescence was recorded from one cell only.

	RESULTS

Top Abstract Introduction Methods Results Discussion References

CCh stimulation of K_Ca in HSG cells. Stimulation of HSG cells with CCh induced an increase in the outward current at a holding potential of 0 mV, the Cl equilibrium potential (Fig. 1A), in 94% of the cells. Oscillatory increases were observed at lower concentrations of CCh (1-10 µM), whereas steady-state increases in the current were obtained with higher concentrations (>100 µM). Although the initial amplitude of the current in the same cell was similar at all agonist concentrations (see Fig. 1A), the mean current increased significantly with increasing CCh concentration, from 322 ± 123 pA at 1 µM and 663 ± 175 pA at 10 µM to 1,161 ± 299 pA at 100 µM CCh (P < 0.05, n = 5). A concentration of 1 µM CCh typically evoked baseline-separated oscillations with a mean frequency of 3.7 ± 1.7 per minute (n = 5), whereas 10 µM CCh induced either similar, fast baseline-separated oscillations (mean frequency of 6.6 ± 2.1 per minute, n = 6; seen in one-half of the cells tested) or slower oscillations, which were superimposed on a sustained elevation of the current (as shown in Fig. 1A). Higher concentrations of CCh (100 µM to 1.0 mM) consistently induced a fast transient increase in the outward current that was followed by lower steady-state current (Fig. 1A). Figure 1B shows outward currents in a cell that was stimulated repeatedly by 100 µM CCh; the interval between stimulations was ~3 min. The differences in the mean amplitudes of the first three stimulations are not significant (P > 0.05, n = 4). The amplitudes of second and third responses are 88.7 ± 8.1% and 71.6 ± 14.5% (n = 4), respectively, of the first response. These data demonstrate that no rapid desensitization or inactivation of the K⁺ channel response occurs with repeated short exposure to CCh.

View larger version (16K): [in this window] [in a new window]   Fig. 1.   Carbachol (CCh)-induced outward currents. A: concentration-dependent response of CCh-induced Ca2+-activated K+ channel (KCa) current in human submandibular gland (HSG) cells at a holding potential of 0 mV. CCh and all other agents were continuously applied to the bath (indicated by bars) at a rate of ~5 ml/min. This is a representative trace of results obtained with >20 cells. B: sequential stimulation of HSG cell. CCh (100 µM) was applied repeatedly for 30 s (shown by bars), with an interval of ~3 min between applications in which the cell was washed by bath solution. This is a representative trace of data from 7 different cells. C: 1,2-bis(2-aminophenoxy) ethane-N,N,N',N'-tetraacetic acid (BAPTA), directly introduced into the cell via patch pipette, and CCh (1-100 µM) were applied to the cell (indicated by bars).

View larger version (20K): [in this window] [in a new window]   Fig. 2.   Current-voltage (I-V) relationship of CCh-induced peak KCa. A: control currents were recorded at voltages of 120 to +80 mV with 20-mV steps from a holding potential of 35 mV before application of CCh. B: 10 µM CCh-induced peak current was recorded under same conditions as control recording. C: I-V relationship (in pA and mV, respectively) of peak current induced by 10 µM CCh (n = 6).

IP₃- and TG-dependent stimulation of K_Ca in HSG cells. CCh-stimulated intracellular Ca²⁺ mobilization is mediated via increases in intracellular IP₃ and IP₃-induced release of Ca²⁺ from internal Ca²⁺ stores (1, 2, 15, 32). Thus further experiments were carried out to test whether the IP₃-induced Ca²⁺ release pathway is involved in the CCh stimulation of K_Ca. IP₃, directly applied to HSG cells via the patch pipette, typically caused oscillatory increases in K_Ca at low concentrations of IP₃ (e.g., 10 µM, Fig. 3A) with a frequency of 4.8 ± 1.6 oscillations/min (n = 5). At higher IP₃ concentrations (e.g., 100 µM), a relatively sustained increase in the current was induced that appeared to be superimposed on an oscillatory current and ran down within 2-3 min (Fig. 3B). The mean current increased significantly with increasing IP₃ concentrations [from 662 ± 258 pA at 10 µM (n = 5) to 1,379 ± 424 pA at 100 µM (P < 0.01, n = 9)]. It must be noted that the responses induced by the dialysis of IP₃ were not as stable as that induced by CCh, and 2 of 18 cells tested did not respond to IP₃ stimulation. However, the pattern of currents induced by increasing concentrations of IP₃ was similar to that induced by increasing concentrations of CCh, i.e., oscillations at relatively lower concentrations and relatively sustained increases in the current at higher concentrations. Importantly, addition of CCh during the IP₃-mediated response did not alter the IP₃-induced current (data not shown) and addition of CCh after rundown of the IP₃-stimulated K_Ca oscillations induced either a very attenuated response (Fig. 3B) or no response at all.

View larger version (21K): [in this window] [in a new window]   Fig. 3.   Inositol 1,4,5-trisphosphate (IP3)-induced activation of KCa in HSG cells. A: IP3 (10 µM) was applied to cell at a holding potential of 0 mV through the patch pipette. Oscillations of KCa were recorded in continuous presence of IP3 (dashed line) and were seen as soon as whole cell configuration was established. This is a representative trace of data obtained from 5 different cells. B: IP3 (100 µM) evoked sustained increases in KCa. Addition of CCh in the presence of IP3 is shown by bar. This is a representative trace of data from 11 cells.

View larger version (12K): [in this window] [in a new window]   Fig. 4.   Activation of KCa by thagsigargin (TG) in HSG cells. A: cell exposed to TG (10 µM, indicated by dashed line) after a control response to CCh (100 µM, indicated by first solid bar) was recorded. Response to CCh in the presence of TG was also recorded (indicated by second solid bar). B: cells were treated with TG (10 µM, dashed line) in a Ca2+-free medium (solid line), following which Ca2+ was reintroduced into the medium. Traces are representative of data from 10 cells.

Effect of extracellular Ca²⁺ on the regulation of the K_Ca current in CCh-stimulated HSG cells. As discussed above, activation of the muscarinic receptor in HSG cells induces a biphasic increase in [Ca²⁺]_i: an initial rapid transient increase and a subsequent lower sustained elevation (15, 20, 35). The initial elevation of Ca²⁺ is due to intracellular Ca²⁺ release from IP₃-sensitive Ca²⁺ stores, whereas the sustained elevation is dependent on Ca²⁺ influx from extracellular medium. Consistent with these previous discoveries, Fig. 5A shows the CCh-stimulated biphasic [Ca²⁺]_i increase in a single HSG cell loaded with indo 1. A concentration of 100 µM CCh induced a rapid transient increase followed by a sustained elevation of Ca²⁺. The resting and peak levels of [Ca²⁺]_i following addition of 100 µM CCh were 138 ± 8.5 nM (n = 8) and 375 ± 59 nM (n = 8). The sustained elevation of [Ca²⁺]_i was dependent on Ca²⁺ influx, since removal of extracellular Ca²⁺ reduced [Ca²⁺]_i to the resting level and reintroduction of extracellular Ca²⁺ restored sustained [Ca²⁺]_i. The pattern of CCh-induced increases in the K_Ca current was similar to that of [Ca²⁺]_i (Fig. 5B). The sustained K_Ca current was decreased to resting levels when external Ca²⁺ was removed and recovered when Ca²⁺ was reintroduced into the medium. These data, together with the effect of BAPTA (Fig. 1C), demonstrate that the K⁺ current reflects, and is dependent on, the underlying changes in [Ca²⁺]_i induced following CCh stimulation of the cells.

View larger version (15K): [in this window] [in a new window]   Fig. 5.   CCh-induced increases in cytosolic Ca2+ concentration ([Ca2+]i) and KCa in HSG cells. A: intracellular Ca2+ was determined by monitoring indo 1 fluorescence. CCh was added to cell (indicated by bar) after a basal level of [Ca2+]i was recorded. During the CCh-induced sustained elevation of [Ca2+]i, extracellular Ca2+ was removed (indicated by bar) and reintroduced. B: KCa was measured using a similar experimental paradigm as in Fig. 4A in a different cell from the same cell preparation. These are representative traces of data obtained with 12 cells.

View larger version (15K): [in this window] [in a new window]   Fig. 6.   Role of intracellular Ca2+ release in CCh-stimulated KCa activity in HSG cells. A: a similar protocol to that in Fig. 1B was applied except that Ca2+-free medium was perfused before and during the second or third exposure to CCh (indicated by the corresponding bars). B: experimental protocol similar to that in the Fig. 4 was employed. Control currents were recorded before extracellular Ca2+ was removed (indicated by dashed line). Note that Ca2+-free buffer plus 1 mM EGTA was perfused to the bath before CCh was removed.

Ca²⁺ influx-dependent regulation of K_Ca in HSG cells. The role of Ca²⁺ influx in CCh-stimulated oscillations of the K_Ca current was next examined. Removal of external Ca²⁺ abolished CCh-induced sustained oscillations of K_Ca (Fig. 7A), which were recovered when Ca²⁺ was reintroduced to the medium. Similarly, the sustained oscillations and steady-state increases in K_Ca induced by introducing IP₃ in the patch pipettes were also inhibited by removal of extracellular Ca²⁺ (n = 6, data not shown). To more directly demonstrate the involvement of Ca²⁺ influx, La³⁺ (1 mM), which is an effective Ca²⁺ channel antagonist and blocker of Ca²⁺ influx in a wide variety of nonexcitable cells including salivary gland cells (26, 30), was introduced into the cell medium. CCh-induced sustained oscillations in K_Ca were first decreased and then abolished in the continued presence of La³⁺. The current recovered once La³⁺ was removed from the medium (Fig. 7B). Sustained elevation of [Ca²⁺]_i in CCh-stimulated HSG cells was also blocked by addition of La³⁺ to the cell medium (data not shown). These data suggest that Ca²⁺ influx regulates the sustained activation of K_Ca in CCh-treated HSG cells. As mentioned above, internal Ca²⁺ store refill is achieved by Ca²⁺ influx via the store-dependent pathway. Thus, to further demonstrate that La³⁺ blocks K_Ca by inhibiting Ca²⁺ influx, the effect of La³⁺ on the refill of internal Ca²⁺ stores was examined (Fig. 8A). The cells were first stimulated with CCh, and then La³⁺ was added before removal of CCh. Cells were then restimulated with CCh in the continued presence of La³⁺. The amplitude of the second response to CCh was significantly reduced to 15.1 ± 5.1% of that in the control response (P < 0.01, n = 6). The inhibition was partially recovered to 37.6 ± 9.6% when La³⁺ was washed out.

View larger version (19K): [in this window] [in a new window]   Fig. 7.   Effect of Ca2+ influx on CCh-induced sustained oscillations of KCa. A: current was measured in continued presence of 10 µM CCh (dashed line). Perfusion with Ca2+-free medium and EGTA is shown by the bar. B: experimental conditions were similar to those in A. Cell was continuously perfused with medium containing CCh (10 µM), indicated by dashed line. Addition of 1 mM La3+ to the bath is also indicated (solid bar).

View larger version (15K): [in this window] [in a new window]   Fig. 8.   La3+ and Gd3+, but not Zn2+, block Ca2+ influx-dependent regulation of KCa. Same protocol as described in Fig. 1B was used. Perfusion with CCh-containing medium is shown by solid bars. Addition of the test cations (1 mM) is shown by dashed lines, that is, with La3+ (A), Gd3+ (B), and Zn2+ (C). Data are representative of results obtained with 16 cells.

Regulation of K_Ca by Ca²⁺ influx is dependent on internal Ca²⁺ release in HSG cells. The data presented above demonstrate that the sustained K_Ca current in CCh-stimulated HSG cells is primarily regulated by Ca²⁺ influx. To examine the effect of Ca²⁺ influx in the absence of internal Ca²⁺ release, the K_Ca current was measured during refill of internal Ca²⁺ stores. Cells were first stimulated with CCh in a Ca²⁺-free medium, and atropine was then added to terminate the muscarinic receptor-mediated signaling (i.e., IP₃-dependent intracellular release was inactivated). Reintroduction of Ca²⁺ in the cell medium did not induce any change in K_Ca (Fig. 9B, also see trace in Fig. 6, A and B). However, under these conditions, Ca²⁺ influx did occur, resulting in the refill of internal Ca²⁺ stores. This is shown by the response to a subsequent addition of TG that was larger than that obtained in cells in which the internal stores were not allowed to fully refill (compare data in Fig. 9B with Fig. 9A; in A, TG was added ~1 min after perfusion with CCh-containing medium was stopped). In the absence of atropine, the IP₃-mediated Ca²⁺ release pathway remains activated, and in this case readdition of Ca²⁺ to the medium induced rapid activation of K_Ca. Subsequent removal of Ca²⁺ from the medium and addition of TG induced a small increase in [Ca²⁺]_i due to release of Ca²⁺ from partially refilled stores or from stores not mobilized by CCh. Intracellular Ca²⁺ accumulation has been suggested to strongly buffer Ca²⁺ in the subplasma membrane region in exocrine acinar cells (21, 25). Furthermore, previous studies have indicated that refill of internal Ca²⁺ stores is achieved without significant increases in [Ca²⁺]_i (24, 26, 31, 34). The data in Fig. 9 are consistent with these previous findings and indicate that the K_Ca activity monitors [Ca²⁺]_i in the region of Ca²⁺ influx. To examine the role of intracellular Ca²⁺ pump activity on the regulation of K_Ca, cells were treated with CCh, followed by atropine and then TG (i.e., Ca²⁺ influx activated but IP₃ receptor and Ca²⁺ pump inhibited). Reintroduction of Ca²⁺ into the cell medium induced rapid activation of K_Ca (Fig. 9C). Thus the activation of K_Ca by Ca²⁺ influx alone is achieved only when internal Ca²⁺ release is activated (CCh or TG treated) or internal Ca²⁺ accumulation is inhibited (TG treated, also see Fig. 10).

View larger version (17K): [in this window] [in a new window]   Fig. 9.   Ca2+ influx-dependent regulation of KCa in CCh-stimulated HSG cells is mediated via internal Ca2+ release. A: KCa regulation by Ca2+ influx in CCh-stimulated cells. Cells were stimulated with CCh in a Ca2+-free medium. Addition and removal of 1 mM Ca2+ are indicated. Status of internal Ca2+ stores was checked by addition of TG. B: effect of Ca2+ influx on KCa during refill of internal Ca2+ stores. Additions of CCh, atropine, Ca2+, and TG are indicated. C: Ca2+ influx-dependent regulation of KCa following inhibition of intracellular Ca2+ pump. Additions of CCh, TG, and Ca2+ are indicated. Data represent 3-5 experiments in each condition.

View larger version (15K): [in this window] [in a new window]   Fig. 10.   Model for regulation of KCa by Ca2+ influx in CCh-stimulated HSG cells. See DISCUSSION for details.

	DISCUSSION

Top Abstract Introduction Methods Results Discussion References

The data presented describe the [Ca²⁺]_i-dependent regulation of a large-conductance K_Ca in CCh-stimulated HSG cells. The data demonstrate that there is a strong association between the increase in [Ca²⁺]_i and the increase in the K⁺ current in HSG cells stimulated with CCh. We have shown that the initial increase in K_Ca is dependent on the initial elevation of [Ca²⁺]_i, which is due to the release of Ca²⁺ from intracellular stores. This activation is mimicked by agents that induce release of Ca²⁺ from intracellular stores, such as IP₃, TG, and BHQ. Furthermore, consistent with previous [Ca²⁺]_i measurements, the initial amplitude of K_Ca is not altered by removal of extracellular Ca²⁺. However, under conditions in which the internal Ca²⁺ store is depleted, CCh activation of K_Ca is decreased, i.e., in cells treated with IP₃, TG, or BHQ or when internal Ca²⁺ store refill is prevented by removal of external Ca²⁺ or addition of La³⁺ or Gd³⁺. Importantly, our data show that Ca²⁺ influx is required for CCh-induced sustained oscillations and steady-state increases in the K⁺ current in HSG cells. Removal of extracellular Ca²⁺ reduced the sustained elevation of [Ca²⁺]_i, resulting in a corresponding decrease in K_Ca (Figs. 4, 7, and 9). These data are consistent with several reports showing that the sustained, oscillatory, or steady-state increases in [Ca²⁺]_i in a number of cells, including salivary gland cells, require Ca²⁺ influx (10, 15, 18, 20, 23, 24, 26, 32). Several different Ca²⁺ influx pathways have been proposed to be present in nonexcitable cells, including store-operated (capacitative) Ca²⁺ entry, second messenger (i.e., IP₃)-operated Ca²⁺ entry, and receptor-operated Ca²⁺ entry (1, 2, 6, 20, 31). We have previously reported the presence of store-operated Ca²⁺ entry in HSG cells. In addition, we had also reported a small Ca²⁺ entry component that appeared to be independent of the store status and was regulated by the muscarinic receptor, either directly or via a G protein (20). In the present study, CCh did not stimulate further increases in K_Ca in TG- or BHQ-stimulated cells, suggesting that only the store-operated Ca²⁺ influx pathway is primarily involved in sustaining the K⁺ current in HSG cells. This is consistent with our earlier results showing membrane potential changes in CCh-stimulated HSG cell by using membrane potential-sensitive fluorescent dyes (20).

The molecular mechanism involved in mediating Ca²⁺ influx in nonexcitable cells is not yet known. However, it has been reported recently that the store-operated Ca²⁺ entry pathway, where depletion of intracellular Ca²⁺ stores stimulates Ca²⁺ influx across the plasma membrane, is mediated via an I_CRAC channel (2, 6, 8, 16). Electrophysiological studies with mast cells and T lymphocytes have shown that I_CRAC has a very low conductance: ~1,000-fold lower than the conductance of classical voltage-sensitive Ca²⁺ channels (6, 8). I_CRAC is activated by various stimuli, such as the Ca²⁺-mobilizing agonists (e.g., CCh) or second messengers (e.g., IP₃) or the inhibitors of the Ca²⁺ pump (e.g., BHQ or TG). It is highly Ca²⁺ selective and is strongly inhibited by La³⁺ or low concentrations of Zn²⁺ and by high [Ca²⁺]_i. Although we have not shown direct measurements of the Ca²⁺ influx current in HSG cells here, we have shown that the sustained activation of K_Ca, which reflects a sustained elevation of [Ca²⁺]_i, is induced by stimulation of the cells with CCh, BHQ, or TG or by introduction of IP₃ into the cells. These results are similar to our previously reported data in which Ca²⁺ entry into fura 2-loaded HSG cells was measured. Furthermore, we have also shown here that 1) the sustained activation of K_Ca is dependent on extracellular Ca²⁺, i.e., on Ca²⁺ influx, and is blocked by La³⁺ and Gd³⁺, but not by Zn²⁺, and that 2) the inhibition of K_Ca by the divalent cations is due to the inhibition of Ca²⁺ influx. Zn²⁺ has been reported to effectively block I_CRAC in RBL mast cells. Thus the Ca²⁺ influx pathway in HSG cells does not appear to show typical characteristics of I_CRAC. On the other hand, Gd³⁺, which blocks Ca²⁺ influx into HSG cells (data not shown), has been used extensively to block stretch-activated and voltage-gated cation channels (5, 33). More recently, it has been shown to block cation influx mediated by the Trp gene product, which has been proposed as a candidate protein for the store-operated Ca²⁺ influx activity (3). However, further studies are required to fully describe the electrophysiological characteristics of the Ca²⁺ influx pathway in HSG cells.

The involvement of store-operated Ca²⁺ influx in CCh-dependent regulation of K_Ca in HSG cells is demonstrated by the following. 1) TG and BHQ mimic CCh-induced increases in K_Ca conductance and attenuate the response induced by CCh and vice versa. 2) Inhibition of Ca²⁺ influx prevents initial activation of K_Ca by preventing refill of internal Ca²⁺ store(s). 3) Inhibition of Ca²⁺ influx prevents sustained activation of K_Ca due to loss of sustained [Ca²⁺]_i elevation. Our model for the regulation of K_Ca by Ca²⁺ influx in HSG cells is shown in Fig. 10. We have shown that Ca²⁺ influx alone, in the absence of internal Ca²⁺ release (i.e., during refill of internal Ca²⁺ stores), does not activate K_Ca (Fig. 10B, see data in Fig. 9B). When the intracellular Ca²⁺ accumulation is inhibited, K_Ca is activated by Ca²⁺ influx (Fig. 10C, see data in Fig. 9C). These data clearly indicate that the intracellular Ca²⁺ store membrane and the plasma membrane are likely to be in close proximity, consistent with previous studies (10, 25). However, presently we cannot rule out the possibility that other Ca²⁺ stores may be present that are not closely situated to the plasma membrane and thus likely not involved in the regulation of K_Ca activity.

In aggregate, the data presented above suggest that the [Ca²⁺]_i increase in the region of Ca²⁺ influx appears to be strongly buffered by intracellular Ca²⁺ accumulation and that there is minimal diffusion of Ca²⁺ from this region under conditions when Ca²⁺ influx is activated. Such buffering has been recently suggested in pancreatic cells, where it was shown that influx of Ca²⁺ induced refill of internal Ca²⁺ stores without giving rise to elevations in [Ca²⁺]_i, unless the intracellular Ca²⁺ accumulation was inhibited by TG (Ref. 25, also see footnote¹). The present studies provide further evidence for this buffering by using the K_Ca activity as a readout for subplasma membrane changes in [Ca²⁺]_i. The data (see Fig. 9A) indicate that, when the IP₃-dependent Ca²⁺ release pathway in the internal Ca²⁺ store is activated, Ca²⁺ entering the cell via the Ca²⁺ influx pathway in the plasma membrane can reach the K⁺ channel and activate it. We suggest that, in CCh-stimulated HSG cells, this is mediated by the uptake of Ca²⁺ into the internal store and release via the IP₃-sensitive channel, without significant accumulation in the store (Fig. 10A). An assumption in our model is that IP₃-dependent release of Ca²⁺ from the store does not induce a change in the Ca²⁺ pump activity (decrease) or in the diffusion (increase) of Ca²⁺ from the site of influx. An important question that arises from the above model is why the cell would expend considerable energy to pump Ca²⁺ into the store while the IP₃-sensitive release channel is activated. A possible explanation is that such a mechanism allows the cell to direct localized release of Ca²⁺ and also prevents significant increase in [Ca²⁺]_i at the site of influx. Localized sites for intracellular Ca²⁺ uptake and release have recently been proposed in salivary and pancreatic acinar cells (21, 25). Further studies will be required to determine whether the subcellular localization of the Ca²⁺ influx protein(s), the K⁺ channel, the IP₃ receptor, and the internal Ca²⁺ pump in the sub-plasma membrane region of the HSG cell determine the regulation of K_Ca by [Ca²⁺]_i.