INTRODUCTION

Top Abstract Introduction Procedures Results Discussion References

CALCIUM WAVES HAVE been observed in many cell types and may play a role in the regulation of different cell functions (5). In pancreatic acinar cells, the agonist-induced increase in cytosolic free Ca²⁺ concentration ([Ca²⁺]_i) starts in the luminal cell pole and spreads to the basal cell side (11). The initiation of Ca²⁺ waves takes place in a small trigger zone within the granular area (12, 31). With the use of high spatial Ca²⁺ imaging techniques, hot spots, which act as pacemakers for oscillatory Ca²⁺ release, could be identified in this trigger zone (32). Recently, it had been suggested that the zymogen granules themselves, which are located in the luminal cell region, might release Ca²⁺ in response to inositol 1,4,5-trisphosphate [Ins(1,4,5)P₃] (8). However, in purified preparations of zymogen granules, Ins(1,4,5)P₃-induced Ca²⁺ release could not be shown and also immunoblottings with antisera specific to type I, II, and III Ins(1,4,5)P₃ receptors failed to demonstrate the presence of Ins(1,4,5)P₃ receptors in zymogen granules (37). Therefore, it is still unclear whether zymogen granules can serve as trigger pools for Ins(1,4,5)P₃-induced Ca²⁺ release.

After an initial release of Ca²⁺ from the luminal Ins(1,4,5)P₃-sensitive pool, spreading of the Ca²⁺ wave through the cytosol is carried by further release of Ca²⁺ from neighboring Ca²⁺ pools. It has been suggested that Ca²⁺-induced Ca²⁺ release (CICR), originally found in skeletal muscle, might also be the mechanism underlying Ca²⁺ wave propagation in exocrine cells (14, 18, 34). In isolated pancreatic acinar cells, it has been shown that agonist-evoked Ca²⁺ oscillations could reversibly be reduced by ryanodine (30), which is known to arrest the ryanodine receptor from skeletal muscle on a sublevel of conductance (23). Caffeine (0.5-1 mM), which sensitizes the ryanodine receptor to cytosolic Ca²⁺, could evoke Ca²⁺ spikes during intracellular subthreshold Ca²⁺ infusion through the patch pipette (30, 34). Furthermore, ryanodine (50 µM) as well as caffeine (20 mM) slowed down the speed of both acetylcholine (ACh)- and cholecystokinin-induced Ca²⁺ waves (17). These data suggest that CICR based on ryanodine receptors also exists in pancreatic acinar cells.

Genes encoding for ryanodine receptors have not only been found in muscle but also in brain and in a variety of peripheral tissues (9, 28). They could also be demonstrated in nonexcitable cells in permanent cell culture (1). However, observation of some untypical effects of ryanodine and caffeine in nonexcitable cells suggests that there might be fundamental differences in the functional properties of ryanodine receptors expressed in skeletal muscle and in nonexcitable cells (1). In mouse pancreatic acinar cells, it could be shown that, at resting Ca²⁺ concentrations, caffeine alone, which is able to release Ca²⁺ from intracellular stores in skeletal muscle (25), could not induce Ca²⁺ release (12). Furthermore, ryanodine increased the frequency of Ins(1,4,5)P₃-induced Ca²⁺ spikes and heparin not only blocked Ins(1,4,5)P₃-induced but also cyclic ADP-ribose-induced Ca²⁺ spikes (30). These data suggest that in pancreatic acinar cells there might be a cross-link between the Ins(1,4,5)P₃- and the Ca²⁺-induced Ca²⁺ release mechanism.

In the present study, we have investigated control mechanisms involved in the propagation of ACh- and bombesin-induced Ca²⁺ waves. We could show that in mouse pancreatic acinar cells the speed of agonist-induced Ca²⁺ waves depends on the applied hormone and is modulated by protein kinase C (PKC) activity. Whereas PKC inhibition increased the speed of bombesin-induced Ca²⁺ waves up to the level seen with ACh, activation of PKC decreased ACh-induced Ca²⁺ wave propagation down to the level seen with bombesin. Furthermore, spreading of ACh-induced Ca²⁺ waves could be slowed down by ruthenium red, a blocker of CICR from sarcoplasmic reticulum, and this effect was not additive to PKC activation. We therefore assume that a CICR-like mechanism that is affected by PKC activation might be involved in generation of global Ca²⁺ signals in mouse pancreatic acinar cells.

	EXPERIMENTAL PROCEDURES

Top Abstract Introduction Procedures Results Discussion References

Preparation of pancreatic acinar cells. Acini and pancreatic acinar cells were prepared from adult male CD-1 mice by collagenase treatment of the excised pancreas as described previously (13). Briefly, the pancreas was removed from a mouse, which had been anesthetized with ether and then killed by cervical dislocation. One milliliter of preparation buffer [130 mM NaCl, 4.7 mM KCl, 1.3 mM CaCl₂, 1 mM MgCl₂, 1.2 mM KH₂PO₄, 10 mM N-2-hydroxyethylpiperazine-N'-2-ethanesulfonic acid (HEPES), 0.2% albumin, 0.01% trypsin inhibitor, and 10 mM glucose, pH 7.4 adjusted with NaOH] supplemented with 30 U/ml of collagenase type V was injected into the pancreas, and the tissue was subsequently incubated in 2 ml of the same buffer with collagenase for 10 min at 37°C. After enzymatic digestion, the tissue was mechanically dissociated by gentle pipetting, and the resulting cell suspension was washed three times with preparation buffer without collagenase.

Fluorescence measurements. Freshly prepared pancreatic acini and acinar cells were loaded with 3 µM fluo 3-acetoxymethyl ester (AM) for 30 min at room temperature (20). After dye loading, the cells were stored at 4°C and used for experiments within 3 h. For measurement of hormone-evoked Ca²⁺ signals, cells were placed onto polylysine-coated glass coverslips attached to the bottom of a perfusion chamber. Single cells as well as cell clusters consisting of two to a maximum of six adherent cells were analyzed. The cells were continuously superfused with a "standard NaCl buffer" (in mM: 140 NaCl, 4.7 KCl, 1.3 CaCl₂, 1 MgCl₂, 10 HEPES, and 10 glucose, pH 7.4 adjusted with NaOH). For hormonal stimulation of the cells, 500 nM ACh or 1 nM bombesin was added to the perfusate. Lag time of the bath perfusion system was determined by comparing the time between ionophoretic application of ACh from a pipette placed near the cell (<10 µm) and ACh application with the bath perfusion system. The time for hormone application was corrected for the lag time of the perfusion system. With the use of a confocal laser-scanning system (Bio-Rad, MRC 600), fluorescence images of 128 × 128 pixels with a resolution of 0.463 µm/pixel were recorded every 0.28 s. To monitor local changes in [Ca²⁺]_i, small rectangular areas were selected in the luminal and basolateral cell regions. The time point for the initial increase of the Ca²⁺ signal in the respective area was determined, and the speed (µm/s) of the hormone-induced Ca²⁺ wave was calculated from the distance of the areas divided by the time between an increase in the luminal and basolateral Ca²⁺ concentrations. In single experiments, Ca²⁺ signals of several individual cells could be analyzed at the same time. For a higher temporal resolution, necessary to trace rapid spreading of Ca²⁺ waves in the presence of thapsigargin, the line-scan mode was used. In this mode, intracellular fluorescence signals were determined along a line oriented in the luminal-to-basolateral axis of an acinar cell. The line was scanned with a frequency of 50 Hz. Mean values ± SD were calculated from Ca²⁺ signals of individual cells. The number of performed experiments and analyzed cells and the number of the cell preparations used for the experiments are given. P was calculated with the Student's t-test. All experiments were carried out at room temperature (25°C).

Determination of Ins(1,4,5)P₃. Freshly prepared acinar cells from eight mouse prancreata were sedimented by low-speed centrifugation and resuspended in 2 ml of preparation buffer without bovine serum albumin (BSA). Two hundred microliters of the cell suspension were incubated for 5 min at 30°C before incubation with agonists (500 nM ACh or 1 nM bombesin) or without agonist (control). The incubation was stopped at indicated times by addition of an equal volume of ice-cold trichloroacetic acid (1 M). The samples were then kept on ice for 15 min. After centrifugation at 7,500 g for 10 min at 4°C, the supernatants were extracted three times with water-saturated diethylether and subsequently neutralized by addition of 200 µl of a 65 mM NaHCO₃ solution. For mass determination of Ins(1,4,5)P₃, a specific receptor binding assay was used (2). Measurements of Ins(1,4,5)P₃ were performed in 500 µl assay buffer containing 60 mM tris(hydroxymethyl)aminomethane, 2.4 mM EDTA (pH 9.0), 2.4 mg/ml BSA fraction V, 100 pM [³H]Ins(1,4,5)P₃ (sp act 50 Ci/mmol), 0.2 mg binding protein prepared according to Bentz and Hildebrandt (2), and 200 µl of Ins(1,4,5)P₃ standards (0.2-50 pmol/tube) or 200 µl of the test sample, respectively. After 1 h of incubation on ice, tubes were centrifuged at 7,500 g for 10 min. The supernatants were discarded, and the pellets were resuspended in 1 ml scintillation fluid (Ultima Flo, Packard) and subsequently counted in a liquid scintillation analyzer. The Ins(1,4,5)P₃ content of the samples was determined by comparing the inhibition of [³H]Ins(1,4,5)P₃ binding with a calibration curve obtained with known amounts of Ins(1,4,5)P₃.

Determination of diacylglycerol. One milliliter of pancreatic acinar cell suspension (1.6 mg protein) was incubated for 5 min at 30°C before stimulation with either 500 nM ACh or 1 nM bombesin. For control experiments, standard NaCl buffer was added instead of the hormone. The reaction was stopped at the indicated times by addition of 3.75 ml ice-cold chloroform-methanol (1:2, vol/vol), and the samples were kept on ice for 10 min. For lipid extraction, 1.25 ml NaCl solution (0.9%) and 1.25 ml chloroform were added (3). After centrifugation at 500 g for 5 min at 4°C, the upper aqueous phase was discarded and the lipophilic phase was evaporated under nitrogen. The amount of diacylglycerol (DAG) in the residue was determined by a method described by Preiss et al. (21). Briefly, the residue was resuspended in 20 µl of a solution containing 5 mM cardiolipin, 7.5% (wt/vol) n-octyl--glucopyranoside, and 1 mM diethylenetriamine pentaacetate and incubated for 15 min at 25°C. Then, 70 µl of a solution [50 mM imidazole-HCl, pH 6.6, 50 mM NaCl, 12.5 mM MgCl₂, 1 mM ethylene glycol-bis(-aminoethyl ether)-N,N,N',N'-tetraacetic acid (EGTA), 10 mM dithiothreitol, and 1 mg/ml diacylglycerol kinase] were added, and radioactive labeling of DAG was started by addition of 10 µl of 1 mM [-³²P]ATP (4 × 10⁶ counts/min). The enzymatic reaction was stopped after 30 min by addition of 450 µl chloroform-methanol (1:2, vol/vol). Twenty microliters of HClO₄ (1%, vol/vol) were added to hydrolyze the remaining ATP. Samples were incubated for 10 min at room temperature and then centrifuged for 1 min at 2,000 g. For phase separation, 150 µl chloroform and 150 µl HClO₄ (1%, vol/vol) were added and the samples were thoroughly vortexed. After centrifugation for 1 min at 2,000 g, the upper phase was discarded and the organic phase was washed twice with 1 ml 1% HClO₄ to remove ³²P. Next, 150 µl of the samples were placed on SI-Amprep columns and eluted with 2 ml chloroform, 2 ml ethyl acetate-hexane (1:5, vol/vol), and 2 ml chloroform-methanol-acetic acid (65:10:10, vol/vol/vol). The eluate from the last step was collected, supplemented with 10 ml scintillation fluid (Ultima Flo, Packard), and counted for 4 min in a liquid scintillation analyzer (tri-carb 2100tr, Packard).

Materials. Pervanadate solutions were prepared by addition of 2 mM H₂O₂ to a 100 mM aqueous stock solution of orthovanadate prepared according to Gordon (10). After incubation for 15 min at room temperature, remaining H₂O₂ was removed by addition of catalase.

	RESULTS

Top Abstract Introduction Procedures Results Discussion References

ACh- and bombesin-induced Ca²⁺ waves. Figure 1 shows the time course of Ca²⁺ signals in the luminal and basolateral areas of mouse pancreatic acinar cells induced by 500 nM ACh or 1 nM bombesin from two representative experiments of 396 single cell recordings (37 cell preparations) for ACh and 195 recordings (22 preparations) for bombesin. After onset of bath perfusion with 500 nM ACh, [Ca²⁺]_i rose at the luminal cell pole with a delay of 1-2 s (Fig. 1A). From the luminal trigger zone, the Ca²⁺ signal spread through the cytosol to the basolateral cell side, which was reached ~1 s later. When 1 nM bombesin was used for stimulation (Fig. 1B), the increase in the luminal Ca²⁺ occurred with larger delay (3-10 s) and also the time between the rise in the luminal and basolateral [Ca²⁺] was increased. From the distance between the two measured areas and the time between the Ca²⁺ signals, the propagation speed of the hormone-evoked Ca²⁺ wave could be calculated. The speed of ACh-induced Ca²⁺ waves determined in 37 cell preparations (396 cells) varied between 10.62 and 33.60 µm/s, whereas the speed of bombesin-induced Ca²⁺ waves was in the range of 3.90-11.79 µm/s (22 preparations, 195 cells). The frequencies of the propagation speed of ACh- and bombesin-induced Ca²⁺ waves measured in individual cells are given in Fig. 2, A and B. The mean values for both groups were significantly different (P < 0.0001) and showed a Ca²⁺ spreading of 17.3 ± 5.4 µm/s for ACh and 8.0 ± 2.2 µm/s for bombesin (Fig. 2C). Because of the variability between different preparations, further experiments, investigating modulation of Ca²⁺ signal spreading, were always related to the control experiments performed on cells from the same preparation.

View larger version (26K): [in this window] [in a new window]   Fig. 1.   Agonist-specific Ca2+ wave propagation in mouse pancreatic acinar cells. Cells were loaded with fluo 3 and stimulated with either 500 nM ACh (A) or 1 nM bombesin (B). Changes in the cytosolic Ca2+ concentration ([Ca2+]i) within selected areas near the luminal (lu) and basal (ba) cell membrane were monitored with a confocal laser-scanning microscope. With both hormones, an increase in [Ca2+]i could be observed, first in the luminal and with some latency in the basal cell pole. ACh-induced Ca2+ signals traversed the cell within ~1 s, whereas bombesin-induced Ca2+ waves reached the basal cell membrane ~2 s after start at the luminal cell pole. With ACh, the typical delay between hormone application and an initial increase in the luminal Ca2+ concentration was ~1 s, whereas delay under bombesin-stimulation was ~5 s.

View larger version (23K): [in this window] [in a new window]   Fig. 2.   Propagation rate of ACh- and bombesin-evoked Ca2+ waves in single pancreatic acinar cells. Cells were stimulated with either 500 nM ACh (A) or 1 nM bombesin (B), and the speed of the Ca2+ wave, spreading from the luminal to the basolateral cell pole, was determined. Histograms comprise results from measurements of 396 individual cells from 37 cell preparations for ACh and 195 cells from 22 cell preparations for bombesin. Mean propagation rate (±SD) of the agonist-evoked Ca2+ waves is given in C. ACh caused a significantly faster spreading of Ca2+ signal than bombesin (P < 0.0001).

Effect of inhibitors of Ca²⁺-ATPases in the endoplasmic reticulum. Because spreading of cytosolic Ca²⁺ signals should not only depend on Ca²⁺ release but also on Ca²⁺ reuptake from the cytosol into intracellular stores, we have tested the effect of tBHQ, an inhibitor of sarco(endo)plasmic reticulum Ca²⁺-ATPase (SERCA) (16). Because tBHQ in the micromolar range caused rapid Ca²⁺ release from intracellular stores, we have chosen a relatively low tBHQ concentration (100 nM) that did not lead to a measurable increase in [Ca²⁺]_i within 5 min (see Fig. 4A). Preincubation with 100 nM tBHQ for 5 min accelerated spreading of ACh-induced Ca²⁺ waves from 16.1 ± 5.6 µm/s (37 experiments, 52 cells) to 32.5 ± 14.1 µm/s (35 experiments, 56 cells, 7 cell preparations; P < 0.01). In contrast, the bombesin-induced Ca²⁺ wave was unchanged in the presence of tBHQ. The mean propagation speed was 7.8 ± 2.0 µm/s (20 experiments, 40 cells) in the presence and 8.5 ± 2.4 µm/s (23 experiments, 40 cells, 5 cell preparations) in the absence of tBHQ.

Hormone-induced production of Ins(1,4,5)P₃ and DAG. Stimulation of pancreatic acinar cells with ACh, as well as stimulation with bombesin, leads to G protein-dependent activation of phospholipase C (PLC), which catalyzes cleavage of phosphatidyl inositol bisphosphate (PIP₂) to Ins(1,4,5)P₃ and DAG. Whereas DAG leads to activation of PKC, Ins(1,4,5)P₃ is known to trigger Ca²⁺ release from intracellular stores (27). It is obvious that Ca²⁺ release from intracellular stores to the cytosol can influence the spreading of cytosolic Ca²⁺ signals. We, therefore, measured generation of Ins(1,4,5)P₃ in the presence of ACh or bombesin. Figure 3A shows that after application of 500 nM ACh (6 cell preparations) a maximal increase in generation of Ins(1,4,5)P₃ could already be observed within 3 s (the earliest time point that was technically possible with the experimental setup). On the other hand, when the cell suspension was stimulated with 1 nM bombesin (n = 6), the Ins(1,4,5)P₃ production measured after 3 s was significantly smaller (P < 0.005) and a maximum in the Ins(1,4,5)P₃ concentration could be observed ~10 s after hormone application. Ins(1,4,5)P₃ determinations at 30 s (P = 0.3) and 60 s (P < 0.01) showed higher Ins(1,4,5)P₃ levels in the presence of bombesin than in the presence of ACh.

View larger version (32K): [in this window] [in a new window]   Fig. 3.   Agonist-dependent production of inositol 1,4,5-trisphosphate [Ins(1,4,5)P3] and diacylglycerol (DAG) in pancreatic acinar cells. A: production of Ins(1,4,5)P3 was determined with a specific Ins(1,4,5)P3 binding assay. B: DAG content of the samples was measured by enzymatic phosphorylation of DAG with radioactively labeled ATP. Within 3 s after hormone application, stimulation of pancreatic acinar cells with 500 nM ACh led to a higher concentration of Ins(1,4,5)P3 than stimulation with 1 nM bombesin. On the other hand, in the same time interval, bombesin induced higher production of DAG than ACh. Values are means ± SD. Symbols represent individual experiments. Experiments from same cell preparation are connected with lines.

Source of bombesin-induced production of DAG. Activation of PLC should lead to equimolar production of Ins(1,4,5)P₃ and DAG from PIP₂. Differences in the ACh- and bombesin-induced Ins(1,4,5)P₃ and DAG levels therefore indicate hormone-dependent differences in the metabolism of the two second messengers. DAG is not only generated by PLC activity but can also be released from phosphatidylcholine by activation of phospholipase D (PLD). This enzyme cleaves phosphatidylcholine to choline and phosphatidic acid, which in a second step is dephosphorylated by a phosphatidic acid phosphohydrolase to DAG (6). To test whether PLD-dependent generation of DAG could play a role in the slow spreading of bombesin-induced Ca²⁺ waves, we measured the propagation speed of bombesin-evoked Ca²⁺ waves in the presence of 0.3% n-butanol. Activation of PLD in the presence of n-butanol leads to production of phosphatidylbutanol instead of DAG (7) and therefore should cause a reduction in the DAG-dependent activation of PKC. Preincubation of the cells for 5 min with 0.3% n-butanol caused acceleration of the bombesin-induced Ca²⁺ wave from 6.0 ± 2.6 µm/s (31 experiments, 58 cells) under control conditions to 8.9 ± 3.3 µm/s (18 experiments, 34 cells, 3 cell preparations) in the presence of n-butanol (P < 0.001). When instead of n-butanol the ineffective secondary alcohol 2-butanol (0.3%) was used, no changes in the propagation speed occurred. Furthermore, n-butanol had no effect on ACh-induced Ca²⁺ waves. These data indicate that stimulation of pancreatic acinar cells with bombesin besides activation of PLC also involves activation of PLD.

Regulation of agonist-induced Ca²⁺ waves by PKC. Preactivation of PKC by 1 µM of the phorbol ester PMA for 5 min led to a slowing down of the ACh-induced Ca²⁺ wave from 15.8 ± 2.0 µm/s (13 experiments, 31 cells) under control conditions to 9.3 ± 0.5 µm/s (13 experiments, 24 cells, 3 cell preparations; P < 0.003) in the presence of PMA. A similar effect could be observed when PKC was stimulated with the DAG analog 1-oleoyl-2-acetyl-sn-glycerol (10 µM, preincubation for 5 min) (Fig. 4B). On the other hand, activation of PKC by PMA before stimulation with bombesin only had a clear effect on the propagation speed of Ca²⁺ waves if the Ca²⁺ spreading in the presence of bombesin was relatively fast (8.1 ± 1.6 µm/s; 7 experiments, 12 cells). Then, there was a significant reduction of the Ca²⁺ wave propagation to 4.9 ± 1.5 µm/s (8 experiments, 14 cells, 2 cell preparations; P < 0.008) in the presence of PMA. In experiments in which the Ca²⁺ spreading in the presence of bombesin was already slow (4.6 ± 1.2 µm/s; 7 experiments, 13 cells), PMA did not further significantly reduce the speed of the Ca²⁺ wave (3.8 ± 1.9 µm/s; 6 experiments, 13 cells, 1 cell preparation). When the experiments were pooled, a mean propagation speed of 6.2 ± 2.2 µm/s for the bombesin-induced Ca²⁺ wave under control and 4.1 ± 1.8 µm/s (P < 0.007) after preincubation with PMA could be calculated.

View larger version (28K): [in this window] [in a new window]   Fig. 4.   Effect of inhibitors of Ca2+-ATPases and activators and inhibitors of protein kinase C (PKC) on the propagation rate of agonist-evoked Ca2+ waves in pancreatic acinar cells. A: inhibition of endoplasmic reticulum Ca2+-ATPases with 100 nM 2,5-di-tert-butylhydroquinone (tBHQ) caused an increase in propagation rate of ACh-induced Ca2+ waves, whereas spreading of bombesin (bom)-induced Ca2+ waves was unaffected. When propagation speed of ACh-evoked Ca2+ waves was reduced by preactivation of PKC with phorbol 12-myristate 13-acetate (PMA), inhibition of Ca2+-ATPases with tBHQ failed to accelerate the Ca2+ wave. B: activation of PKC by prestimulation of the cells with 1 µM PMA or 10 µM 1-oleoyl-2-acetyl-sn-glycerol (OAG) for 5 min produced a significant decrease in propagation rate of ACh-induced Ca2+ waves. Speed of bombesin-induced Ca2+ waves was only slightly reduced. On the other hand, inhibition of PKC by preincubation with 100 µM H-7 did not influence spreading of ACh-evoked Ca2+ waves but clearly accelerated bombesin-induced Ca2+ waves. Same effect could be observed when 1 µM RO31-8220 was used for PKC inhibition.

Effect of inhibitors and activators of the CICR. To clarify the nature of the Ca²⁺ pool, which is affected by activation of PKC, we tested ruthenium red and ryanodine on the spreading of the hormone-induced Ca²⁺ signals in mouse pancreatic acinar cells. As shown in Fig. 5, ruthenium red (100 µM), which is known to inhibit CICR in skeletal muscle (26), decreased the speed of the ACh-induced Ca²⁺ wave from 16.5 ± 4.5 µm/s (23 experiments, 51 cells) to 9.0 ± 3.1 µm/s (20 experiments, 40 cells, 3 cell preparations; P < 0.0001) but had no effect on the bombesin-induced Ca²⁺ wave (9 experiments, 19 cells, 2 cell preparations). The effect of ruthenium red on the ACh-induced Ca²⁺ wave was not additive to that of PMA, indicating that the same Ca²⁺ pools are involved in modulating the propagation speed of ACh-induced Ca²⁺ waves. Ryanodine (50 µM), which arrests the CICR channel from the sarcoplasmic reticulum in a half-open state (23), had no effect on ACh- or on bombesin-evoked Ca²⁺ waves. Also, caffeine (2 mM), an activator of the sarcoplasmic reticulum Ca²⁺ channel (22), did not reduce the propagation rate of ACh-induced Ca²⁺ waves and did not have any effect on the resting [Ca²⁺]_i.

View larger version (26K): [in this window] [in a new window]   Fig. 5.   Effect of ruthenium red (RR) and PMA on propagation rate of agonist-evoked Ca2+ waves. Ruthenium red (100 µM), an inhibitor of Ca2+-induced Ca2+ release (CICR) in skeletal muscle, significantly slowed down the propagation rate of ACh-induced Ca2+ waves. Effect of ruthenium red was in the same order of magnitude as activation of PKC with PMA. When ruthenium red and PMA were applied together, no additivity could be observed. Ruthenium red had no significant effect on bombesin-evoked Ca2+ waves.

	DISCUSSION

Top Abstract Introduction Procedures Results Discussion References

Ca²⁺ waves may be an important initial step for coordinating cellular functions. In mouse pancreatic acinar cells, local Ca²⁺ oscillations in the luminal cell pole as well as oscillatory Ca²⁺ waves spreading from the luminal to the basal cell membrane have been proposed to regulate Cl secretion into the acinar lumen (11, 31). Considering the different functions the cell has to accomplish, it appears reasonable that different hormones and neurotransmitters evoke different Ca²⁺ signaling patterns. Thus activation of receptors for ACh and cholecystokinin, both linked to breakdown of PIP₂, results in different Ca²⁺ spiking patterns with longer-lasting transients for cholecystokinin compared with ACh (38). Hormone-dependent differences in the Ca²⁺ signals can also be observed when the time course for the spreading of agonist-evoked Ca²⁺ waves is investigated. Our data indicate that stimulation with ACh causes faster propagation of Ca²⁺ waves than stimulation with bombesin. This is consistent with previous data that showed that after stimulation with supramaximal concentrations of carbachol the initial Ca²⁺ signal traversed pancreatic acinar cells within 0.92 s, whereas the bombesin-induced Ca²⁺ signal took 2.3 s to spread through the cell (36). When a 10- to 15-µm distance from the luminal to the basal cell membrane is assumed, a propagation speed close to our values for the ACh- and bombesin-evoked Ca²⁺ waves can be calculated.

The spreading of cytosolic Ca²⁺ signals is a complex process and depends on the release of Ca²⁺ from intracellular stores, influx of Ca²⁺ into the cell, diffusion of Ca²⁺ in the cytosol, and binding of Ca²⁺ to mobile and immobile Ca²⁺ binding sites (cytosolic Ca²⁺ buffer). Finally, Ca²⁺ reuptake from the cytosol into the stores as well as Ca²⁺ extrusion by active transport mechanisms will also control propagation of Ca²⁺ waves. It is very unlikely that hormonal stimulation of the pancreatic acinar cell can directly produce rapid changes in the capacity of the cytosolic Ca²⁺ buffer. Furthermore, our experiments showed that agonist-dependent spreading of Ca²⁺ waves is not dependent on the presence of extracellular Ca²⁺. Therefore, changes in the Ca²⁺ buffer as well as a modulating role of Ca²⁺ influx through the plasma membrane can be excluded as reasons for agonist-dependent differences in Ca²⁺ wave propagation. In this early phase of hormone response, differences in the propagation rate must be due to a modification of the Ca²⁺ release, reuptake, and/or extrusion mechanisms. With Ca²⁺ release, a faster spreading of Ca²⁺ signals can be due to either a higher amount of Ca²⁺ initially released from the trigger pool or involvement of secondary Ca²⁺ release from Ca²⁺ stores in series to the Ins(1,4,5)P₃-dependent trigger pool. In our experiments, we used the nonratiometric dye fluo 3 to monitor changes in [Ca²⁺]_i. Absolute values for the free [Ca²⁺]_i can be calculated according to a self-ratio method (4). When a dissociation constant of 390 nM for Ca²⁺ binding of fluo 3 and a resting [Ca²⁺]_i of 90 nM are assumed, peak values for the initial increase in [Ca²⁺]_i in the luminal cell region of 490 ± 51 nM for ACh and 559 ± 60 nM for bombesin (P = 0.4) can be calculated. Because, for both agonists, the amount of initially released Ca²⁺ is in the same order of magnitude, it is most likely that agonist-dependent differences in the propagation rate of Ca²⁺ waves depend on the interaction of differently regulated Ca²⁺ stores in series to the trigger pool. In this model, increased Ca²⁺ release from secondary Ca²⁺ pools would lead to an accelerated spreading of the Ca²⁺ signal, whereas decreased secondary Ca²⁺ release would cause a slower propagation of Ca²⁺ waves.

Ca²⁺ pools in pancreatic acinar cells. In previous studies, it has been shown that, in addition to Ins(1,4,5)P₃-induced Ca²⁺ release, CICR might also play a role in the generation of Ca²⁺ signals in exocrine acinar cells (14, 18, 30, 34). Furthermore, it has been reported that the propagation rate of ACh-induced Ca²⁺ waves in intact acini is reduced by 20 mM caffeine as well as by 50 µM ryanodine, suggesting that, as in skeletal muscle, CICR is also involved in spreading of hormone-evoked Ca²⁺ signals in exocrine cells (17). In our experiments, neither 50 µM ryanodine nor 2 mM caffeine decreased the propagation speed of ACh-induced Ca²⁺ waves. Caffeine in a higher concentration (10-20 mM) could not be used, since in high concentrations caffeine inhibits development of agonist-induced Ca²⁺ signals, probably due to blocking of Ins(1,4,5)P₃ production (33). On the other hand, ruthenium red, which is known to inhibit Ca²⁺-release channels in skeletal muscle (26) as well as voltage-dependent, Ins(1,4,5)P₃-insensitive Ca²⁺ channels in the endoplasmic reticulum (24), caused a reduction of the propagation rate of ACh-induced Ca²⁺ waves. Therefore, our data indicate that in pancreatic acinar cells agonist-dependent differences in the speed of cytosolic Ca²⁺ signal propagation are due to agonist-dependent modification of Ca²⁺ release from stores in series to the Ins(1,4,5)P₃-sensitive trigger Ca²⁺ pool. The effect of ruthenium red suggests that the mechanism underlying Ca²⁺ wave propagation might involve ryanodine receptors or related proteins.

Regulation of Ca²⁺ release by PKC. Our data indicate that progression of cytosolic Ca²⁺ waves is modulated by PKC activity. The fast spreading of Ca²⁺ waves induced by ACh can be slowed down by prestimulation of PKC, whereas the slow spreading induced by bombesin can be accelerated by PKC inhibition. In parallel experiments, we could show that stimulation of pancreatic acinar cells with ACh and bombesin led to different temporal patterns in the production of Ins(1,4,5)P₃ and DAG. In the presence of ACh, Ins(1,4,5)P₃ was produced very rapidly, leading to Ca²⁺ release from luminal Ins(1,4,5)P₃-sensitive Ca²⁺ stores within 1-3 s after hormone application (see Fig. 1A). On the other hand, in the presence of bombesin, Ins(1,4,5)P₃ was produced more slowly and luminal Ca²⁺ signals were triggered ~3-10 s after hormone application (see Fig. 1B). When hormone-induced DAG production was compared 3 s after hormone application, bombesin caused a significantly higher rise in the DAG concentration than ACh. This could mean that, in the presence of bombesin, activation of PKC by DAG could occur before triggering of Ca²⁺ release from the luminal Ca²⁺ store. The observations are consistent with a model (Fig. 6) in which rapid production of DAG in the presence of bombesin causes high activity of PKC and therefore slows down the propagation rate of cytosolic Ca²⁺ waves. On the other hand, stimulation of the cell with ACh leads to formation of little DAG and therefore to low PKC activity and fast spreading of Ca²⁺ waves. Because exogenous activation and inhibition of PKC had no effect on the height of the initial Ca²⁺ signal in the luminal cell region, the effects of PKC can be best explained with the assumption of a modulating role of PKC on Ca²⁺ release from stores that discharge subsequently to the luminal trigger pool. In this model, activation of PKC has no effect on the Ins(1,4,5)P₃-sensitive primary Ca²⁺ pool in the luminal cell pole but does diminish subsequent Ca²⁺ release from the secondary stores, which are located in series between the luminal and the basal plasma membrane. Activation of PKC therefore slows down spreading of Ins(1,4,5)P₃-triggered Ca²⁺ waves, which could not be decreased to lower values but to ~8 µm/s in the presence of PMA and ruthenium red (See Figs. 4 and 5). Inhibition of PKC leads to a faster secondary Ca²⁺ release and therefore accelerates cytosolic Ca²⁺ waves. When in the presence of ACh, the propagation rate of the hormone-induced Ca²⁺ wave was reduced by preactivation of PKC with PMA; a further reduction in the spreading speed with ruthenium red could not be achieved, indicating that activation of PKC, like application of ruthenium red, inhibits a CICR mechanism (Fig. 6A).

View larger version (32K): [in this window] [in a new window]   Fig. 6.   Model for primary Ins(1,4,5)P3-sensitive and secondary CICR Ca2+ pools involved in Ca2+ wave spreading. A: stimulation of pancreatic acinar cells with ACh leads to production of Ins(1,4,5)P3, which releases Ca2+ from the luminal trigger pool (shaded Ca2+ pool). According to the concentration gradient, Ca2+ diffuses to the basolateral membrane. In the presence of ACh, stimulation of PKC by DAG is relatively low and Ca2+-release channels of the CICR Ca2+ pools can open. CICR from the secondary stores accelerates propagation of Ca2+ signal. Artificial activation of PKC by PMA as well as direct inhibition of CICR by ruthenium red (RR) diminishes Ca2+ release from secondary stores and therefore slows down the ACh-induced Ca2+ signal. Due to the low PKC activity, inhibitors of PKC like H-7 or RO31-8220 have no effect on ACh-induced Ca2+ waves. Ca2+ release from secondary stores is partially counterbalanced by active reuptake with a tBHQ- and thapsigargin-sensitive Ca2+-ATPase. Inhibition of this Ca2+-ATPase increases the Ca2+ wave propagation rate by an increased net efflux from the secondary stores. B: when cells are stimulated with bombesin, there is a higher production of DAG compared with ACh. DAG-dependent activation of PKC inhibits CICR from secondary stores. Therefore, in the presence of bombesin, spreading of Ca2+ signal is relatively slow, since it is carried mainly by diffusion of Ca2+ released from the luminal Ins(1,4,5)P3-triggered store. Inhibition of PKC by preincubation with H-7 or RO31-8220 abolishes the inhibitory effect of PKC on CICR and therefore accelerates the bombesin-induced Ca2+ wave. Due to the high activity of PKC in the presence of bombesin, additional stimulation of PKC with PMA as well as inhibition of CICR by ruthenium red has no further effect. When CICR is inhibited by PKC activity, active Ca2+ reuptake into the filled secondary stores is small and does not contribute significantly to Ca2+ net flux from these stores. Application of tBHQ, therefore, has no significant effect on the bombesin-induced Ca2+ waves. For further explanations, see text.

Role of Ca²⁺ reuptake for Ca²⁺ wave propagation. Our experiments with the SERCA inhibitors tBHQ and thapsigargin indicate that spreading of cytosolic Ca²⁺ signals not only depends on Ca²⁺ release but also on Ca²⁺ reuptake from the cytosol into intracellular stores. The observation that ACh-induced Ca²⁺ waves were significantly accelerated by tBHQ, whereas bombesin-induced Ca²⁺ waves were unchanged, suggests that the rate of Ca²⁺ reuptake is different in the presence of ACh and bombesin. When the cells are stimulated with bombesin, Ca²⁺ release from the secondary stores is small due to high PKC activity and, therefore, the compensatory Ca²⁺ reuptake into these stores is also small (Fig. 6B). Blocking of the Ca²⁺-ATPase, therefore, has no significant effect on the Ca²⁺ wave. In contrast, in the presence of ACh, which evokes fast Ca²⁺ release from secondary Ca²⁺ pools, the counterbalancing Ca²⁺ reuptake is high. Inhibition of Ca²⁺ reuptake with tBHQ therefore significantly increases the net Ca²⁺ flux from the pool into the cytosol and leads to a faster spreading of the Ca²⁺ signal (Fig. 6A). Consistent with this model, we observed that, when ACh-induced Ca²⁺ release from the secondary stores is reduced by PKC stimulation with PMA, tBHQ fails to increase the propagation rate of the hormone-evoked Ca²⁺ wave. Because vanadate had no effect on the propagation of cytosolic Ca²⁺ signals, we can conclude that Ca²⁺ reuptake into the secondary stores is mainly dependent on a tBHQ- and thapsigargin-sensitive but vanadate-insensitive transport mechanism.