Modulation of Pancreatic Acinar Cell to Cell Coupling during ACh-evoked Changes in Cytosolic Ca2+ *

The temporal changes in cytosolic free Ca2+ ([Ca2+] i ), Ca2+-dependent membrane currents (Im ), and gap junctional current (Ij ) elicited by acetylcholine (ACh) were measured in rat pancreatic acinar cells using digital imaging and dual perforated patch-clamp recording. ACh (50 nm-5 μm) increased [Ca2+] i and evokedIm currents without alteringIj in 19 of 37 acinar cell pairs. Although [Ca2+] i rose asynchronously in cells comprising a cluster, the delay of the [Ca2+] i responses decreased with increasing ACh concentrations. Perfusion of inositol 1,4,5-trisphosphate (IP3) into one cell of a cluster resulted in [Ca2+] i responses in neighboring cells that were not necessarily in direct contact with the stimulated one. This suggests that extensive coupling between acinar cells provides a pathway for cell-to-cell diffusion of Ca2+-releasing signals. Strikingly, maximal (1–5 μm) ACh concentrations reduced Ij by 69 ± 15% (n = 9) in 25% of the cell pairs subjected to dual patch-clamping. This decrease occurred shortly after the Im peak and was prevented by incubating acinar cells in a Ca2+-free medium, suggesting that uncoupling was subsequent to the initiation of the Ca2+-mobilizing responses. Depletion of Ca2+-sequestering stores by thapsigargin resulted in a reduction of intercellular communication similar to that observed with ACh. In addition, ACh-induced uncoupling was prevented by blocking nitric oxide production with l-nitro-arginine and restored by exposing acinar cells to dibutyryl cGMP. The results suggest that ACh-induced uncoupling and capacitative Ca2+ entry are regulated concurrently. Closure of gap junction channels may occur to functionally isolate nearby cells differing in their intrinsic sensitivity to ACh and thereby to allow for sustained activity of groups of secreting cells.

Gap junctions are intercellular channels formed by twelve subunits of membrane proteins called connexins (Cx). 1 Six subunits are contributed by each cell to form hemichannels, the docking of which provides a low resistance pathway for exchange of ions and small molecules between cells in contact. Gap junctional coupling was shown to be involved in the control of embryonic development, cell proliferation, electrical conduction, and metabolic cooperation (1)(2)(3). Because gap junction channels allow for the potential passage of molecules of a molecular mass up to 1000 Da, it is conceivable that second messengers produced in one cell can diffuse between neighboring cells to coordinate their individual response. In support of this hypothesis, the passage of Ca 2ϩ waves has been reported in epithelial cells, glial cells, and various cultured cells (4).
The exocrine pancreas represents a valuable model to search for the role of gap junctional coupling in signal transduction of nonexcitable tissues. Acinar cells are extensively electrically and chemically coupled by Cx32-and Cx26-built gap junction channels (5). A major group of secretagogues in these cells are the Ca 2ϩ -mobilizing agonists, including cholecystokinin and acetylcholine (ACh). In the highly polarized acinar cells, cytosolic Ca 2ϩ ([Ca 2ϩ ] i ) initially rises within the apical secretory region and, when stimulation is sufficient, subsequently spreads as a wave toward the basal pole of the cell (6 -9). Intercellular propagation of Ca 2ϩ oscillations and/or Ca 2ϩ waves elicited by Ca 2ϩ -mobilizing secretagogues has been reported to correlate with gap junctional activity (7, 10 -13). Increasing evidence indicates that open gap junctions coordinate the frequency of Ca 2ϩ oscillations within individual cells of a same acinus which, in turn, regulates enzyme secretion. This hypothesis, however, is in apparent contradiction with the observation that these secretagogues also evoke acinar cell uncoupling, both in vitro and in vivo, at concentrations that maximally stimulate enzyme secretion (14 -17). Thus, the role of gap junctional coupling during acute stimulation of pancreatic acinar cells remains tantalizing.
One of the first questions to address is whether rises in [Ca 2ϩ ] i and changes in junctional coupling evoked by acinar cell stimulation are parallel events. The application of the patch-clamp technique to dissociated acinar cells has revealed that the kinetic of Ca 2ϩ -dependent membrane current activation reflects that of the [Ca 2ϩ ] i changes (18 -20). However, accurate monitoring of junctional conductance under dual whole-cell recording conditions has been limited because spontaneous uncoupling occurs within seconds, presumably as a result of cytoplasm dialysis (21)(22)(23). To bypass this problem, we applied here a dual perforated patch-clamp approach, which preserves the integrity of the internal milieu (24, 25), to mon-itor pairs of acinar cells stimulated with increasing ACh concentrations for both Ca 2ϩ -dependent membrane and gap junctional currents. In parallel experiments, changes in [Ca 2ϩ ] i were monitored by digital imaging of fluo-3-loaded acinar cells. Under these conditions, we observed that ACh-induced closure of gap junction channels parallels the phase plateau of the [Ca 2ϩ ] i response, but not the initial peak. We further show that acinar cell uncoupling requires the presence of extracellular Ca 2ϩ and parallels capacitative Ca 2ϩ entry.

EXPERIMENTAL PROCEDURES
Preparation of Acinar Cells-Acinar cells were prepared as described previously (26). Briefly, acini were first isolated by collagenase (CLS 3, Worthington Biochemical Corp.) digestion from the pancreas of male Wistar or Sprague-Dawley rats (about 200 g), which were killed by decapitation. Single and paired acinar cells were prepared by resuspending the intact acini in a Ca 2ϩ -and Mg 2ϩ -free Krebs Ringer-bicarbonate medium buffered to pH 7.4 with 12.5 mM Hepes-NaOH and containing 3 mM EGTA. The resulting cell suspension was repeatedly passed through an 18-gauge needle, centrifuged for 3 min at 100 ϫ g in Krebs-Ringer bicarbonate medium (KRB) supplemented with 4% bovine serum albumin. Cells were then resuspended in RPMI 1640 culture medium (Life Technologies, Inc.) supplemented with 0.1% bovine serum albumin and 0.01% trypsin inhibitor (Sigma), plated on bacterial Petri dishes (60 ϫ 15 mm), and kept at 4°C up to 6 h.
Perforated Patch-Clamp-Acinar cells were rinsed by centrifugation and allowed to attach for 15 min at room temperature onto glass coverslips, previously coated with 0.5 mg/ml poly-L-lysine (M r 150,000 -300,000, Sigma) in distilled water. Coverslips with attached acinar cells were transferred to a chamber mounted on the stage of an inverted microscope (TMD-300, Nikon AG). Throughout the experiments, cells were continuously superfused with a solution containing 136 mM NaCl, 4 mM KCl, 1 mM CaCl 2 , 1 mM MgCl 2 , and 2.5 mM glucose, which was buffered to pH 7.4 with 10 mM Hepes-NaOH (control solution). In some experiments, CaCl 2 was omitted from the control solution and 1 mM EGTA was added (Ca 2ϩ -free solution). To allow long-term recording with minimal cell damage, we chose a dual perforated configuration of the patch-clamp technique (25). To increase the success of seal formation, the tip of electrodes was first filled with a solution containing 139 mM KCl, 1 mM NaCl, 2 mM MgCl 2 , 0.5 mM EGTA, 10 mM Hepes-KOH (pH 7.2) and then back-filled with the same solution supplemented with 120 g/ml amphotericin B (24). This antibiotic concentration was prepared from a 60 mg/ml amphotericin B stock solution in dimethyl sulfoxide (Me 2 SO) and thoroughly sonicated before use. Under these conditions, the electrode resistance averaged 2.9 Ϯ 0.08 M⍀ (mean Ϯ S.E., n ϭ 56) as measured in the bathing control solution, and the reversal potential for chloride and cations was about ϩ3 mV. Once a gigaohm seal was obtained, fast capacitative current transients were compensated. Under these conditions, steady series resistances of 10.6 Ϯ 0.5 M⍀ were reached within 20 -30 min (n ϭ 37).
To monitor ACh-evoked Ca 2ϩ -dependent membrane and gap junctional currents, voltage sweeps of 1 s were repeatedly applied at 3-s intervals in both cells of a pair using an EPC-9 (HEKA elektronik) and a PC-501A (Warner Instrument Corp.) patch-clamp amplifier. First, both cells (cell 1 being connected to the EPC-9, and cell 2 connected to the PC501A) were hyperpolarized for 200 ms from a holding potential of ϩ3 mV to Ϫ30 mV. This voltage step allowed for the monitoring of Ca 2ϩ -dependent chloride and cation membrane currents (I m1 and I m2 ) evoked during ACh stimulation. After returning to the holding potential for 100 msec, a negative pulse of 10 mV and 300 ms was applied to cell 2 to elicit a transjunctional potential that allowed for the measurement of junctional current (I j Ј) in cell 1. Currents were acquired at a sampling rate of 1 kHz. Series resistance (R s1 and R s2 ) was compensated in both cells. Off-line analysis of the electrophysiological data was performed using the Pulse software (HEKA elektronik). For each sweep, average values of membrane currents recorded at ϩ3 mV and Ϫ30 mV were calculated in cells 1 and 2, and amplitude of I j Ј was determined in cell 1. These values were then expressed as a function of time. Final display of the traces as well as plots were generated using the IGOR software (WaveMetrics Inc.). Thus, I m1 and I m2 were defined as the difference between the current measured at Ϫ30 mV and ϩ3 mV. True junctional current (I j ) was calculated by I j ϭ I j Ј [1ϩ(R s1 /r 1 )], where r 1 is the input resistance of cell 1 evaluated at ϩ3 mV, and I j Ј and R s1 are as described above (21).
Measurements of [Ca 2ϩ ] i -Acinar cells were incubated in the presence of 10 M fluo-3 acetoxymethylester (fluo-3/AM, Molecular Probes) and allowed to adhere for 15 min at room temperature to coverslips coated with poly-L-lysine. Coverslips were then transferred to the stage of an upright microscope fitted with differential interference contrast optics (Axioskop FS, Zeiss). A peristaltic pump was used to continuously superfuse the cells with KRB. Global stimulation of acinar cells was achieved by superfusing KRB supplemented with various concentrations of ACh (Sigma), as indicated under "Results." Stimulation of individual cells was achieved by injecting 10 -40 M inositol 1,4,5trisphosphate (IP 3 ) into one cell within a cluster using a patch electrode in the whole-cell configuration of the patch-clamp technique. In this case, the electrode-filling solution was supplemented with 10 mM Na 2phosphocreatine, 25 units/ml creatine phosphokinase, and 1 mM MgATP to minimize the rapid and spontaneous uncoupling of acinar cells (26). ACh-evoked mobilization of [Ca 2ϩ ] i was measured using a real time (30 images/s with averaging 16 frames) confocal laser scanning microscope equipped with an Ar/Kr laser (Odyssey XL with In-terVision, Ver. 1.4.1 software, Noran Instruments Inc.). Cells were viewed with a 63 ϫ 0.9 numerical aperture achroplan water immersion objective lens (Zeiss). A 100 m slit was used for [Ca 2ϩ ] i signals, giving bright images with a 3.1 m axial resolution. Fluo-3 was excited through a 488-nm band pass filter, and the emitted fluorescence was collected through a 515 nm barrier filter. To follow the time course of fluo-3 emission changes, the "bright over time" tool of the InterVision software package was applied to areas that surrounded cells on live images using an Indy R4600SC/133 MHz Silicon Graphics station. Because fluo-3 is a single-wavelength dye, its emission is a function of both intracellular Ca 2ϩ and dye concentration. [Ca 2ϩ ] i changes were therefore expressed as the F 1 /F 0 ratio where F 0 was the initial fluorescence intensity measured during the recording (27,28). Acquired data were then processed for analysis using either the Indy station or a PowerPC 8100/100 MHz (NIH Image, Ver. 1.6.0; Adobe Photoshop, Ver. 3.0.5; or Igor Pro, Ver. 2.03).
Dye Coupling-For dye coupling studies, intact acini attached to plastic dishes coated with poly-L-lysine were used. Acini were incubated in KRB supplemented with either 1 M ACh, 0.5 M thapsigargin, or 1 mM dibutyryl cGMP (Sigma) for up to 30 min. Specificity of these agents was tested by depleting acinar cells for their internal Ca 2ϩ . To deplete internal Ca 2ϩ stores, acini were incubated for 15 min in a Ca 2ϩ -free KRB supplemented with 1 mM EGTA and 500 nM ACh, and rinsed for an additional 15 min in the Ca 2ϩ -free solution (29). To assess coupling, one cell per acinus was impaled with a glass microelectrode (150 -200 M⍀) filled with 4% Lucifer Yellow in 150 mM LiCl and buffered to pH 7.2 with 10 mM Hepes. The microelectrode was connected to a pulse generator for passing current and recording membrane potential (30). After successful cell impalement, 0.1 nA negative square pulses of 900 ms duration and 0.5 Hz frequency were applied to the electrode for 3 min to inject the dye. At the end of the injection period, acini were photographed under fluorescence and phase-contrast illuminations. To quantitate the extent of dye coupling, color slides of the microinjected acini were projected on a graphic tablet connected to a personal computer to measure the surfaces of the whole acinus and the Lucifer Yellow-labeled cells. Cell coupling was then expressed as percentage of the acinus area. All data are expressed as mean Ϯ S.E. and compared with controls using an unpaired t test. Values were not corrected for the slight overestimation of the Lucifer Yellow stained areas which resulted from their fluorescent labeling (30).

Temporal Relationship between ACh-induced Uncoupling and [Ca 2ϩ ] i Changes-
The dual perforated patch-clamp approach was applied to pairs of acinar cells to simultaneously monitor Ca 2ϩ -dependent membrane (I m ) and gap junctional currents (I j ). All cell pairs exhibited a large and stable initial junctional conductance, which averaged 42 Ϯ 4 ns (mean Ϯ S.E., n ϭ 37) over a 30 -90 min period of recording. Nine of the 37 cell pairs studied did not respond to ACh at concentrations ranging from 100 nM to 5 M. In 19 cell pairs, ACh triggered I m currents without affecting the extensive electrical coupling (I j traces) of the cell pairs (Fig. 1A). In the remainder nine pairs, a similar paradigm was observed with ACh concentrations Յ 1 M. However, ACh induced a marked decrease in junctional coupling, averaging 69 Ϯ 15% (n ϭ 9) of the initial conductance, at concentrations of 1-5 M. This uncoupling was reversible when the secretagogue was washed out and was inducible again during a second exposure to ACh (Fig. 1B). The decrease in junctional conductance was low, if any, at the I m peaks. In contrast, uncoupling was maximal 1-2 min after the I m peaks, taking place during the sustained phase of the I m response (Figs. 1B and 3). Close inspection of the recordings showed that Ca 2ϩ -dependent currents (I m1 and I m2 ) were synchronized in both cells of a pair when junctional coupling was high. In contrast, the pattern of I m1 and I m2 currents was different when junctional coupling was low (Fig. 1B, arrows). In the example shown in Fig. 1B, while Ca 2ϩ -dependent membrane currents were still observed in one cell (I m1 trace), they were readily discontinued in the other (I m2 trace). Similar observations were made in stimulated pairs that spontaneously uncoupled during recording (n ϭ 4) or which were experimentally uncoupled by superfusing 3.5 mM heptanol (n ϭ 2).
One possible explanation for the synchronized I m responses observed in highly coupled pairs is that cells mobilized [Ca 2ϩ ] i simultaneously. Alternatively, junctional coupling could be so high that the patch pipettes only recorded the average current of the whole cell pair. To address this question, [Ca 2ϩ ] i changes were monitored in multiple acinar cells by digital imaging. As shown in Fig. 2A, superfusion of an acinar cell pair with 100 nM ACh induced first a rise in [Ca 2ϩ ] i in one cell followed by a delayed Ca 2ϩ response in the second cell. An increase in the concentration of ACh was associated with shortening of the delay between the onset of responses, an event which was observed in all 13 cell pairs tested. At higher ACh concentrations, the Ca 2ϩ responses exhibited a typical biphasic "peakand-plateau" profile ( Fig. 2A). Two mechanisms, not mutually exclusive, could be involved in the lack of synchronization of the Ca 2ϩ response between coupled acinar cells. The long lasting intervals between individual responses at low ACh concentrations could be because of heterogeneity in ACh responsiveness (11,13,31). Furthermore, distinct IP 3 receptor sensitivity may exist in acinar cells as illustrated in Fig. 2B. Dialysis of a cell using a patch pipette containing 10 -40 M IP 3 triggered transient rises in [Ca 2ϩ ] i , which was followed by delayed Ca 2ϩ responses in either neighboring or distant cells (n ϭ 5).
Together, these data indicate that acinar cells can exhibit asynchronous rises in cytosolic Ca 2ϩ in response to Ca 2ϩ -mobilizing mediators. In dual-patched acinar cells, the asynchrony of Ca 2ϩ -dependent membrane currents was only unmasked when junctional coupling was strongly reduced (Figs. 1B and 3).
ACh-induced Uncoupling Is Dependent on Extracellular Ca 2ϩ -In acinar and other cell types, it has been shown that the rapid phase of [Ca 2ϩ ] i changes reflects Ca 2ϩ release from internal stores, whereas the later sustained phase depends on capacitative Ca 2ϩ entry into the cells (32)(33)(34). We therefore investigated whether removal of external Ca 2ϩ could affect ACh-induced changes in junctional coupling. As shown in Fig.  3, a first exposure of acinar cells to ACh induced I m currents and a delayed reduction of I j . During a second stimulation, the amplitude of the I m currents was markedly decreased and junctional coupling was not affected. Larger ACh-induced I m and I j responses were readily restored after reintroduction of Ca 2ϩ in the superfusing solution. The ACh-induced uncoupling of acinar cells observed in cells incubated in the absence of external Ca 2ϩ was typically reduced by 49 Ϯ 8% (n ϭ 4) as compared with that measured in pairs exposed to the control solution.
To investigate a possible relationship between Ca 2ϩ entry and cell uncoupling, the extent of intercellular communication was studied by injecting Lucifer Yellow in isolated acini exposed to various conditions known to deplete internal Ca 2ϩ stores and/or to activate capacitative Ca 2ϩ entry. Thapsigargin is a specific inhibitor of the endoplasmic reticulum Ca 2ϩ -ATPase (35), which induces a slow depletion of Ca 2ϩ stores and, hence, capacitative entry of Ca 2ϩ . As shown in Figs. 4 and 5, a 15-20-min exposure to 0.5 M thapsigargin resulted in a marked reduction of intercellular communication. Although Lucifer yellow rapidly spread from the injected cell into all its neighbors in control acini (Fig. 4A), the diffusion of the tracer was restricted to the site of injection in the presence of thapsigargin (Fig. 4B). Quantitative analysis revealed that the surface labeled by Lucifer yellow represented 51 Ϯ 6% (n ϭ 17) of the acinus profile, a value that is markedly reduced (p Ͻ 0.001) as compared with that measured under control conditions (Fig.  5A). A similar blockade of acinar cell coupling (p Ͻ 0.001) was observed with 1 mM dibutyryl cGMP (Figs. 4C and 5A), an agent known to activate Ca 2ϩ entry (36). The effects of both agents were inhibited when internal Ca 2ϩ stores were previously depleted and acinar cells incubated in the absence of extracellular Ca 2ϩ (Fig. 5A). We therefore studied whether Ca 2ϩ -mobilizing secretagogues could cause parallel activation of capacitative Ca 2ϩ entry and cell uncoupling. ACh has been shown previously to stimulate capacitative Ca 2ϩ entry by acti- vation of nitric-oxide synthase, leading to generation of nitric oxide (NO) which, in turn, increases intracellular cGMP concentration (29,37). When cells were incubated in the presence of 1 mM L-nitro-arginine, an inhibitor of nitric-oxide synthase, the ACh-induced uncoupling was fully prevented (Fig. 5B). In contrast, application of dibutyryl cGMP to acini pre-treated with L-nitro-arginine fully reduced (p Ͻ 0.001) acinar cell coupling by an extent similar to that observed in the presence of ACh alone (Figs. 4D and 5B). These results suggest that closure of gap junctions by ACh is closely associated with the activation of capacitative Ca 2ϩ entry. DISCUSSION Our results describe the temporal relationship between changes in [Ca 2ϩ ] i and gap junctional conductance within pairs of pancreatic acinar cells exposed to ACh. Uncoupling of pancreatic acinar cells by concentrations of Ca 2ϩ -mobilizing secretagogues that maximally stimulate exocrine secretion is well documented (14 -17). However, the intracellular mechanism that mediates this effect is not known (38). Using a dual per-  forated patch-clamp approach and confocal digital imaging, we report here that ACh-induced uncoupling and [Ca 2ϩ ] i changes had distinct kinetics. Thus, decrease in junctional conductance of acinar cells consistently develops after the initial [Ca 2ϩ ] i peak and is maximal during the [Ca 2ϩ ] i plateau. Furthermore, ACh-induced uncoupling was no longer detected when reloading of internal Ca 2ϩ stores was prevented by incubation of the cells in a Ca 2ϩ -free medium. These results suggest that depletion-activated Ca 2ϩ entry was a key determinant for the AChinduced uncoupling of acinar cells.
In many nonexcitable cells, depletion of intracellular Ca 2ϩ stores by IP 3 is the primary mechanism by which cell surface receptors activate Ca 2ϩ influx. This phenomenon, which is termed capacitative Ca 2ϩ entry (33), has been involved in the control of Ca 2ϩ oscillations (32,39), secretion (40), and enzymatic regulation (41). The signal that couples store depletion to Ca 2ϩ entry has not yet been identified (33). In pancreatic acinar cells, however, there is definite evidence that NO produced by nitric-oxide synthase mediates the stimulation of cGMP formation by cholinergic agonists. cGMP, in turn, is known to modulate Ca 2ϩ entry (29,36,37,42,43). Consistent with these data, we observed that activation of capacitative Ca 2ϩ entry by either depletion of internal Ca 2ϩ stores with thapsigargin or by exposure of acinar cells to dibutyryl cGMP decreased intercellular communication to an extent similar to that observed with ACh alone. Side effects of these agents on intercellular communication appear unlikely because uncoupling was abolished in acini that were depleted for their internal Ca 2ϩ and incubated in a Ca 2ϩ -free medium. In addition, cell uncoupling evoked by ACh was prevented in the presence of a nitric-oxide synthase inhibitor, supporting the view that gap junctional conductance and capacitative Ca 2ϩ entry are concurrently regulated. These results, however, do not rule out the possibility that capacitative Ca 2ϩ entry may activate another intracellular pathway leading to modulation of junctional conductance. Support for this idea is provided by the earlier observation that okadaic acid, a phosphatase inhibitor that modulates capacitative Ca 2ϩ entry (44,45), also prevents AChinduced uncoupling (38). Although several studies have shown that gap junction channels are blocked by cGMP (25) and nitric oxide (46), our data provide the first observation that AChinduced uncoupling is linked to capacitative Ca 2ϩ entry.
Previous in vitro and in vivo studies have documented a relationship between intercellular communication and the secretory activity of pancreatic acinar cells (17,30,47). In this context, gap junctional coupling is thought to coordinate the Ca 2ϩ response of individual acinar cells within an acinus and thereby to regulate exocytosis (11,48). This idea is supported by our present finding that the large and stable junctional conductance observed in most acinar cell pairs was not altered during changes in [Ca 2ϩ ] i evoked by ACh. In agreement with a previous study using cholecystokinin (13), we observed that rises in [Ca 2ϩ ] i were asynchronous in acinar cells stimulated with low concentrations of ACh. Increasing the agonist concentration was associated with shortening of the delay between the onset of the Ca 2ϩ responses, suggesting that these cells were coupled in terms of Ca 2ϩ mobilization. Also, perfusion of IP 3 into one cell evoked a rise in [Ca 2ϩ ] i in neighboring cells even though not necessarily in those that directly contacted the stimulated one. This observation suggests that rat pancreatic acinar cells differ in their ability to mobilize Ca 2ϩ from internal stores, as indicated by previous studies reporting similar heterogeneity in [Ca 2ϩ ] i mobilization (31) and amylase secretion (48,49). This differential responsiveness may be essential to provide a properly modulable response to agonist-specific stimulation (11,50).
These results, however, are not immediately reconcilable with the observation that ACh decreases gap junctional coupling while maximally stimulating the secretory activity of acinar cells (30,47,48). The blockade of acinar cell-to-cell communication is known to enhance the basal release of amylase in vitro and in vivo (16,26,30,47). Under conditions of gap junction blockade, the potency of several agonists in stimulating exocytosis has also been found to be reduced when the effect of acinar cell uncoupling on basal secretion was taken into account (11,16,30). Therefore, uncoupling may provide acinar cells with a mechanism to sustain enzyme release during acute stimulation by increasing their rate of basal secretion. In this context, the delayed uncoupling evoked by ACh may compartmentalize cells that are highly sensitive to ACh from cells that are less sensitive to the secretagogue, therefore decreasing the effective volume of cytoplasm of interconnected cells. This regulation may be essential to ensure that the intracellular levels of critical factor(s) are maintained to allow for sustained activity of groups of actively secreting cells. Future studies should determine whether uncoupling induced by Ca 2ϩ store depletion FIG. 5. Role of capacitative Ca 2؉ influx on acinar cell uncoupling. A, the extent of dye coupling between acinar cells incubated in the presence of 0.5 M thapsigargin (Tg) or 1 mM dibutyryl cGMP (dB-cGMP) was markedly reduced (p Ͻ 0.001) as compared with that observed under control conditions (CONT or Me 2 SO (DMSO) see "Experimental Procedures"). The uncoupling effect induced by both Tg and dB-cGMP was prevented when Ca 2ϩ -sequestering stores were depleted and acini incubated in a Ca 2ϩ -free medium (Ca 2ϩ -depleted). B, the uncoupling evoked by 1 M ACh could be prevented by a 10-min preincubation of acini with 1 mM L-nitro-Arginine (L-NA). Addition of dB-cGMP to acini exposed to L-NA and ACh restored the uncoupling effect (p Ͻ 0.001). Stars indicate differences at p Ͻ 0.001 levels compared with controls. is a common mechanism to control junctional communication in other types of nonexcitable cells.