Pancreatic acinar cells exhibit a remarkable polarization of Ca2+ release and Ca2+ influx mechanisms. In the present brief review, we discuss the localization of channels responsible for Ca2+ release [mainly IP3 (inositol 1,4,5-trisphosphate) receptors] and proteins responsible for SOCE (store-operated Ca2+ entry). We also place these Ca2+-transporting mechanisms on the map of cellular organelles in pancreatic acinar cells, and discuss the physiological implications of the cellular geography of Ca2+ signalling. Finally, we highlight some unresolved questions stemming from recent observations of co-localization and co-immunoprecipitation of IP3 receptors with Orai channels in the apical (secretory) region of pancreatic acinar cells.
Pancreatic acinar cells specialize in the secretion of both digestive enzymes and the precursors of digestive enzymes. The cells also secrete fluid, facilitating the transport of secreted proteins along the system of pancreatic ducts. Both protein and fluid secretion is triggered by intracellular Ca2+ signals. The secretagogues ACh (acetylcholine) and CCK (cholecystokinin), utilize Ca2+ signalling cascades to trigger and regulate secretion . The apical region of an acinar cell contains a large number of secretory granules [2–4], whereas the basal region contains exceptionally well-developed rough ER (endoplasmic reticulum) [2,3,5–7]. Ca2+ signals in pancreatic acinar cells initiate in the apical region of the cell [8,9]; some of these signals remain localized to this region (local Ca2+ responses), whereas others propagate to the basal part of the cell in the form of regenerative Ca2+ waves ([3,8–11] and reviewed in ). Stimulation of pancreatic acinar cells with physiological doses of secretagogues results in the preferential formation of local Ca2+ signals, although, even at physiological concentrations, secretagogues can trigger global Ca2+ responses . The complex and polarized Ca2+ signalling present in the acinar cells is possible because of the very specific distribution of Ca2+-release and Ca2+-influx sites.
Apical Ca2+-release mechanisms and surrounding organelles
The basal region of acinar cells contains a very high density of ER, although there are also ER strands that project into the apical region. Experiments with local uncaging of caged substances (including caged Ca2+) in the ER lumen  and local reloading experiments  demonstrated continuity between the apical and basal regions of ER. IP3Rs [IP3 (inositol 1,4,5-trisphosphate) receptors] are well-defined Ca2+-release channels located in the apical region of acinar cells [13–17]. Experiments with knockout mice conducted by Futatsugi et al.  in Katsuhiko Mikoshiba's laboratory (RIKEN Brain Science Institute, Japan) convincingly demonstrated that both Ca2+ responses and the secretion of digestive enzymes in pancreatic acinar cells crucially depend on IP3R2 and IP3R3. The functional role of IP3R1 in pancreatic acinar cells is not yet clear. All three subtypes of IP3Rs (1, 2 and 3) are present in pancreatic acinar cells and are preferentially located in the apical region; more specifically, the receptors are positioned very close to tight junctions [13–17]. The organelle responsible for IP3-mediated Ca2+ release in the apical region has not yet been identified. The ER is usually considered to be the organelle responsible for this form of Ca2+ release, but it can also occur from the Golgi apparatus [13,19], although this has not been tested in acinar cells. Acidic organelles of an unspecified origin (suggested to be secretory granules) have also been shown to release Ca2+ in the apical part of the acinar cells via an IP3-mediated mechanism [20,21]. Other second messengers also contribute to the initiation of apical Ca2+ responses [22,23]. Ca2+-release sites in the acinar cells are located apically to the Golgi apparatus, and local Ca2+ responses form Ca2+ gradients across the Golgi . The Golgi apparatus is enveloped by mitochondria from the basal and lateral sides . This prominent group of mitochondria, termed the mitochondrial belt, plays an important role in shaping Ca2+ responses in the acinar cells [3,24–26], importing Ca2+ via mitochondrial uniporters and, in so doing, preventing the spread of Ca2+ into the basal part of the cell. By spatially restricting Ca2+ signals, the mitochondria play a crucial role in the formation of local apical Ca2+ responses [3,11,25,26]. This ‘firewall’ action of mitochondria can be overwhelmed by stronger stimulation (e.g. higher doses of secretagogues or higher concentrations of IP3), in this case, Ca2+ responses propagate into the basal region and form global Ca2+ signals [3,11,25,26]. Ryanodine receptors, which have been characterized in the acinar cells, play an important role in the propagation of Ca2+ signals [11,27–30], and Ca2+-induced Ca2+ release has indeed been documented in this cell type . It is important to note that the proximity of mitochondria to the apical Ca2+ signals allows very efficient co-ordination of Ca2+ responses and cellular bioenergetics. Mitochondrial Ca2+ uptake increases the rate of the tricarboxylic acid cycle, reflected by changes in NAD(P)H fluorescence  and the rate of ATP production . Ca2+ therefore both initiates energy-consuming secretory responses and simultaneously up-regulates ATP production to meet the increased energy demand . The Ca2+-dependent regulation of bioenergetics in the acinar cells is very efficient: during stimulation with Ca2+-releasing secretagogues, the rate of ATP production increases approximately 2.5 times. Remarkably, this occurs without any decrease in ATP concentration; in fact, the cytosolic ATP concentration increases slightly . Reciprocally, the bioenergetics of the cell have a powerful effect on Ca2+ signalling, with ATP depletion resulting in the inhibition of secretagogue-induced Ca2+ signalling [33,34] and Ca2+ influx . One could therefore conclude that the positioning of apical Ca2+-release sites with respect to the mitochondria is of strategic importance to both cellular bioenergetics and Ca2+ signalling.
Localization of Ca2+-influx mechanisms
Secretagogue-induced Ca2+ signalling in pancreatic acinar cells is accompanied by substantial Ca2+ extrusion. Experiments utilizing the microdroplet technique resolved Ca2+ extrusion from single isolated acinar cells and revealed that extrusion can reduce total Ca2+ concentration in the cell by a few hundred micromoles per minute . This level of extrusion results in the rapid depletion of Ca2+ stores in cells placed in a Ca2+-free extracellular solution, leading to the termination of Ca2+ responses [33,35]. In an extracellular solution containing a normal physiological Ca2+ concentration, acinar cells stimulated with physiological concentrations of secretagogues display continued Ca2+ signalling for a prolonged time period, suggesting that Ca2+ stores are efficiently reloaded by Ca2+ imported from the extracellular milieu . The concept of SOCE (store-operated Ca2+ entry) (initially termed capacitative Ca2+ influx) was formulated in 1986  (for a review, see ). Later, an interesting hypothesis of ‘conformational coupling’ based on an interaction between IP3Rs and Ca2+-influx channels was put forward [38,39]. The notion of a physical interaction between Ca2+ sensors in the ER and Ca2+-influx channels leading to SOCE was later confirmed, but with STIM1 (stromal interaction molecule 1) not IP3Rs as the Ca2+ sensor, and Orai as the Ca2+-influx channel [40–43].
We have investigated the localization and translocation of STIM1 and Orai1 channels in pancreatic acinar cells. In unstimulated cells, the distribution of STIM1 was very similar to that of other markers of the ER. Upon store depletion, STIM1 translocated preferentially to the lateral and basal plasma membrane regions of the cell, forming subplasmalemmal puncta in these regions (Figure 1A). Very little STIM1 was found in the apical region of stimulated cells. One can therefore conclude that the apical and basolateral parts of the acinar cell specialize in different processes: the apical region in Ca2+ release, and the basolateral region in Ca2+ reloading. The basolateral localization of STIM1 in pancreatic acinar cells with depleted Ca2+ stores was paradoxical. The basolateral part of these cells is dominated by rough ER, which is studded with ribosomes [2,44]. A ribosome is too large to allow direct interaction between STIM1 and Orai1; the basolateral ER should therefore be unsuitable for the initiation of SOCE. Gyorgy Lur in our laboratory has resolved this paradox by observing strands of rough ER losing ribosomes and forming very tight (approximately 11–12 nm) junctions with the plasma membrane . The ribosomes were absent only from the junction, but decorated the rest of the ER strand approaching the plasma membrane. By expressing fluorescently labelled Orai1 and STIM1, we were able to demonstrate the formation of STIM1–Orai1 puncta in the basolateral region of thapsigargin-treated acinar cells, confirming the important role played by the basolateral plasma membrane region in SOCE in this cell type. Surprisingly, both exogenously expressed Orai1 and endogenous Orai1 (revealed using antibodies produced by Stefan Feske, New York University School of Medicine) were highly concentrated in the apical region of pancreatic acinar cells  (Figure 1A). This surprising apical concentration of Orai was later confirmed by Hong et al. .
Distribution of STIM1, Orai1 and IP3Rs in pancreatic acinar cells
The mystery of apical Orai
In the apical region of acinar cells, Orai1 did not co-localize with its activator STIM1, but was found to co-localize and interact (revealed by co-immunoprecipitation) with IP3Rs [14,44]. This close co-positioning of IP3Rs and Orai1 channels (Figure 1B) satisfies the structural requirements of the original conformation-coupling hypothesis. It was therefore important for us to test whether the activation of IP3Rs could influence the SOCE in this cell type. The results of our experiments have been largely negative. Caffeine, which very effectively inhibits IP3-induced Ca2+ responses in this cell type, had no effect on Ca2+ influx. ACh, a Ca2+-releasing secretagogue that stimulates IP3 production in pancreatic acinar cells, induced a small inhibition rather than potentiation of SOCE . Crucially, robust SOCE was recorded in pancreatic acinar cells from mice with double knockout of IP3R2 and IP3R3, which are the main functional IP3Rs in pancreatic acinar cells. Furthermore, in these knockout animals, the apical Orai was preserved in spite of the absence of IP3Rs. In our experiments, Orai was found not only in the apical region, but also at the lateral and basal domains of the plasma membrane (Figure 1). However, whereas lateral and basal functioning of Orai1 can be attributed to an interaction with STIM1 and formation of SOCE channels, the role of apical Orai1 remains unresolved. Perhaps the question should now be reversed and, instead of asking what IP3Rs can do for Orai1-mediated Ca2+ influx, we should instead be asking what can Orai1 do for IP3-mediated Ca2+ release?.
Signalling 2011: a Biochemical Society Centenary Celebration: A Biochemical Society Focused Meeting held at the University of Edinburgh, U.K., 8–10 June 2011. Organized and Edited by Nicholas Brindle (Leicester, U.K.), Simon Cook (The Babraham Institute, U.K.), Jeff McIlhinney (Oxford, U.K.), Simon Morley (University of Sussex, U.K.), Sandip Patel (University College London, U.K.), Susan Pyne (University of Strathclyde, U.K.), Colin Taylor (Cambridge, U.K.), Alan Wallace (AstraZeneca, U.K.) and Stephen Yarwood (Glasgow, U.K.).
Our laboratory is supported by funding from the Medical Research Council and the Wellcome Trust.