Calcium signalling and pancreatic cell death: apoptosis or necrosis?

Abstract

Secretagogues, such as cholecystokinin and acetylcholine, utilise a variety of second messengers (inositol trisphosphate, cADPR and nicotinic acid adenine dinucleotide phosphate) to induce specific oscillatory patterns of calcium (Ca²⁺) signals in pancreatic acinar cells. These are tightly controlled in a spatiotemporal manner, and are coupled to mitochondrial metabolism necessary to fuel secretion. When Ca²⁺ homeostasis is disrupted by known precipitants of acute pancreatitis, for example, hyperstimulation or non-oxidative ethanol metabolites, Ca²⁺ stores (endoplasmic reticulum and acidic pool) become depleted and sustained cytosolic [Ca²⁺] elevations replace transient signals, leading to severe consequences. Sustained mitochondrial depolarisation, possibly via opening of the mitochondrial permeability transition pore (MPTP), elicits cellular ATP depletion that paralyses energy-dependent Ca²⁺ pumps causing cytosolic Ca²⁺ overload, while digestive enzymes are activated prematurely within the cell; Ca²⁺-dependent cellular necrosis ensues. However, when stress to the acinar cell is milder, for example, by application of the oxidant menadione, release of Ca²⁺ from stores leads to oscillatory global waves, associated with partial mitochondrial depolarisation and transient MPTP opening; apoptotic cell death is promoted via the intrinsic pathway, when associated with generation of reactive oxygen species. Apoptosis, induced by menadione or bile acids, is potentiated by inhibition of an endogenous detoxifying enzyme NAD(P)H:quinone oxidoreductase 1 (NQO1), suggesting its importance as a defence mechanism that may influence cell fate.

Main

Acute inflammation of the pancreas, triggered by such major precipitants as alcohol and gallstones, currently affects approximately 50 per 100 000 people per year, with an increasing incidence.¹ It is characterised by autodigestion of the organ involving premature, calcium (Ca²⁺)-dependent activation of digestive enzymes, vacuole formation in the apical, granular secretory pole and colocalisation of lysosomes and zymogen granules.^{2, 3} Attacks of acute pancreatitis resolve spontaneously in a majority of individuals with no further episodes; however, approximately 20–30% of patients go on to develop more severe pancreatic injury, often with extensive pancreatic necrosis and the development of a systemic inflammatory response syndrome leading to multiple organ failure, which may prove fatal.⁴

Cell death, defined as an irreversible loss of membrane integrity,⁵ has been categorised in up to 11 types ⁶ and is effected in the pancreatic acinar cell by two principal processes, apoptosis and necrosis. Apoptosis is genetically regulated and occurs via both caspase-dependent and independent pathways,⁷ while necrosis is thought to be largely non-programmed and uncontrolled, although this view has recently been questioned.⁸ Perturbation of Ca²⁺ signalling has been linked to both apoptotic and necrotic cell death;⁹ in the pancreatic acinar cell, oscillatory global rises of cytosolic Ca²⁺ may induce apoptosis,¹⁰ while sustained elevations promote necrosis.^{11, 12} These differences may relate to effects on mitochondrial function associated with the respective Ca²⁺ signals. Under physiological conditions these organelles respond to localised increases in cytosolic Ca²⁺ by stimulus-metabolism coupling that generates NAD(P)H, reductive intermediates that fuel the respiratory transport chain with consequent production of ATP required for secretion.^{13, 14, 15, 16} In contrast, global, sustained Ca²⁺ rises can drastically reduce ATP production in pancreatic acinar cells.¹²

For many years it has been accepted that the major form of pancreatic acinar cell death is necrosis¹⁷ and current evidence suggests that the balance between apoptosis and necrosis may influence the severity of acute pancreatitis.^{18, 19} For example, induction of pancreatic acinar cell apoptosis protected mice against cerulein-induced pancreatitis,²⁰ while a recent study has indicated that suppression of the apoptotic cascade in pancreatic acinar cells, via inhibition of caspases, leads to necrotising pancreatitis.²¹ This review will focus on the role of Ca²⁺ signalling in the determination of pancreatic acinar cell fate.

Necrosis

Necrosis is characterised by severe pathophysiological changes including mitochondrial swelling, plasmalemmal disruption and ultimately leakage of cellular contents. The latter triggers acute exudative inflammation of the surrounding tissue; subsequent activation and infiltration of neutrophils is thought to increase intracellular digestive enzyme activation and exacerbate pancreatitis.²²

Over a decade ago we proposed that abnormal, prolonged elevation of cytosolic Ca²⁺ is the crucial trigger of pancreatitis.²³ Under physiological conditions, Ca²⁺ is tightly regulated in discrete cellular microdomains, which when disrupted may result in severe pathological consequences.²⁴ Ca²⁺ signals in the pancreatic acinar cell, generated by the second messengers inositol trisphosphate (IP₃), nicotinic acid adenine dinucleotide phosphate (NAADP) and cyclic ADP-ribose (cADPR), are modulated in a precise spatiotemporal manner necessary for normal secretory function, and are generally confined to the apical pole.¹⁶ A superficial perigranular mitochondrial buffer barrier restricts the spread of Ca²⁺ signals to the basolateral area of the cell,²⁵ an action that may be shared by the closely situated Golgi apparatus.²⁶

Hyperstimulation with cholecystokinin (CCK) causes acute pancreatitis in all species so far examined,¹ and in pancreatic acinar cells obtained from mice repeatedly injected with the CCK analogue cerulein, there is a loss of apical to basolateral progression of secretagogue-induced Ca²⁺ signals, suggesting that normal homeostatic control becomes compromised.²⁷ Furthermore, supra-maximal concentrations of CCK-induced elevation of baseline Ca²⁺ in isolated pancreatic acinar cells which was strongly linked to premature intracellular activation of trypsinogen, the hallmark of acute pancreatitis,^{2, 3} an effect that may involve activation of the vacuolar ATPase.²⁸ This action to raise cytosolic Ca²⁺ is shared by other diverse pathological stimuli that cause pancreatic injury, including non-oxidative ethanol metabolites,^{11, 12} duct ligation²⁹ and bile acids.^{30, 31, 32} In addition, recent evidence has shown that the enzyme phosphatidylinositol 3-kinase (PI3K) mediates prolonged elevations of cytosolic Ca²⁺, under conditions of agonist stimulation, via an inhibitory action on the sarco-endoplasmic reticulum Ca²⁺ ATPase (SERCA) pump;³³ the PI3K system has been implicated in the development of acute pancreatitis since pharmacological inhibition or knockout of this enzyme was protective in experimental animal models.^{34, 35} Importantly, abrogation of sustained rises of Ca²⁺, using the intracellular Ca²⁺ chelator 1,2-bis(O-aminophenoxy)ethane-N,N,N′,N′-tetraacetic acid (BAPTA), prevents zymogen activation induced by CCK yperstimulation,² bile acid-induced cell damage and death³⁰ and necrosis caused by non-oxidative ethanol metabolites.¹¹

Apoptosis

Apoptosis is a genetically regulated, programmed form of cell death that can occur by two principal routes: extrinsic (receptor-mediated) and intrinsic (classical: mitochondrial) pathways.⁹ It is characterised by distinct features that involve a cascade of events, and which ultimately lead to removal of the dead cell from the tissue; unlike necrosis, apoptosis does not involve release of intracellular contents and thus does not elicit inflammation.⁶ Cell and organellar shrinkage, membrane blebbing, condensation of nuclear chromatin and DNA cleavage, and flipping of phosphatidylserine moieties from the inner to the outer side of the plasmalemmal membrane are typical events, the latter permitting detection of acinar cell apoptosis experimentally via annexin V staining using fluorescence microscopy.^{10, 36} Apoptotic pathways employ caspases, aspartate-specific cysteine proteases, as the key elements of the initiation and execution of programmed cell death, with at least 14 isoforms currently resolved.³⁷ Caspases are highly conserved through human evolution and are currently divided into ‘initiator’ (caspases 8 and 9), which induce the proteolytic cascade that results in activation of ‘executioner’ (caspases 3, 6 and 7), that cleave numerous target proteins. Cellular apoptosis may also occur independently of caspases via the action of lysosomal cathepsins, such as cathepsin D, which has been shown to translocate from lysosomes to the cytosol in response to apoptotic stimuli.³⁸ Furthermore, the scenario appears even more complex, since activation of the intrinsic route by the extrinsic pathway has been recently demonstrated in pancreatic acinar cells.²¹

The role of Ca²⁺ in the induction of apoptosis has been recognised for several decades. For example, early studies have shown that the Ca²⁺ ionophore A23187³⁹ and the SERCA pump inhibitor thapsigargin,⁴⁰ which raise cytosolic Ca²⁺, both induced apoptosis, an effect prevented by Ca²⁺ chelation with BAPTA. Ca²⁺ is known to regulate the intrinsic apoptotic pathway,⁴¹ and Ca²⁺-dependent opening of the mitochondrial permeability transition pore (MPTP) is considered to be a critical feature that leads to cytochrome c release into the cytosol.⁴² Once released, cytochrome c, apoptotic peptidase activating factor-1 and pro-caspase 9 combine to form an apoptosome that activates caspase 9, with subsequent activation of caspase 3 that cleaves specific apoptotic targets causing cell death.⁴³ However, recent evidence has suggested that mitochondria may undergo MPTP formation and cytochrome c release in cells that lack isoforms of the adenine nucleotide translocator (ANT); the pore was no longer regulated by ANT ligands, while higher Ca²⁺ was required for permeability transition.⁴⁴ Importantly, hepatocytes without ANT remained responsive to various initiators of cell death, questioning the importance of the MPTP in induction of apoptosis. Furthermore, cells from cyclophilin-D-deficient mice did not undergo cyclosporin A-sensitive MPTP formation but died normally in response to various apoptotic stimuli. These animals showed resistance to necrotic cell death induced by reactive oxygen species (ROS) and Ca²⁺ overload, suggesting the importance of the MPTP in necrosis, but not apoptosis.⁴⁵

The features of the endoplasmic reticulum (ER) Ca²⁺ store in relation to secretagogue signalling have been extensively characterised in the pancreatic acinar cell.¹⁶ The ER and mitochondria possess a close spatial relationship and recent evidence has pointed to the potential importance of ER-mitochondrial Ca²⁺ transfer to cell death,⁴⁶ with the proposition that localised Ca²⁺ elevations, triggered via release from the ER, might induce or facilitate mitochondrial features of apoptosis. In particular, activation of IP₃ receptors may exert a pro-apoptotic role,⁴⁷ while complex effects of anti-apoptotic Bcl-2/Bcl-xL and pro-apoptotic Bax/Bak proteins on ER membrane Ca²⁺ permeability and mitochondrial function point to an important relationship between ER stress and cellular apoptosis.^{46, 48}

Bile acid effects on cell fate

In 1901, Opie⁴⁹ proposed the common duct theory, which suggested that gallstones that become impacted at the ampulla of Vater cause bile reflux into the pancreas, which in turn causes acute pancreatitis. The detrimental effects of bile on the pancreas have subsequently been confirmed in experimental animal models.^{50, 51} In isolated pancreatic acinar cells, bile acids induce elevations of cytosolic Ca²⁺ that are oscillatory or sustained in nature (Figure 1a)³¹ and which result in Ca²⁺-dependent cell death.³⁰ The effects of bile are associated with partial mitochondrial membrane depolarisation, when assessed using the sensitive ‘dequench’ mode of tetramethyl rhodamine methyl ester (TMRM) measurement but not by the classical mode (Figure 1a), an effect that is inhibited by chelation of intracellular Ca²⁺ with BAPTA.³²

Figure 1

(a) TLC-S (100 μM) induces (i) variable patterns of cytosolic Ca²⁺ elevations in three individual pancreatic acinar cells from the same acinar triplet³¹ and (ii) partial mitochondrial depolarisation, seen as changes of TMRM fluorescence (‘dequench’ mode); full depolarisation produced by subsequent application of the protonophore CCCP (10 μM).³² (b) (i) Schematic model of second-messenger interactions with acidic and ER Ca²⁺ stores. IP₃ activates IP₃Rs in both stores, whereas cADPR and NAADP activate RyRs in both stores, but via separate binding sites and/or activation mechanisms.⁵² (ii) Images showing the two-photon permeabilisation technique of a doublet of pancreatic acinar cells (loaded with Fluo-5N AM; white dot shows region of permeabilisation by two-photon laser beam, as described by Gerasimenko et al.⁵²). Note that only the lower cell has been permeabilised and is initially bright, due to diffusion of Texas Red dextran into the cytosol, but becomes paler on washout. (iii) TLC-S (200 μM), added in the presence of thapsigargin (10 μM), induced further additional Ca²⁺ release from an acidic store located in the secretory granule (apical) area (blue), whereas no changes were detected in the basolateral area (red).⁵⁶ (c) Effects of bile acids on cell fate. (i) Typical light-transmitted and R110-aspartic acid amide fluorescent images and (ii) mean data showing that TLC-S (300 μM) induced caspase activation in pancreatic acinar cells that was greatly potentiated when the detoxifying enzyme NQO1 was inhibited by DMN (30 μM), whereas DMN alone had no effect³⁶

Recently, we described an acidic, thapsigargin-insensitive and bafilomycin A-sensitive Ca²⁺ store located in the secretory granule area of pancreatic acinar cells, which is responsive to the second messengers IP₃, cADPR and NAADP (Figure 1b).⁵² While IP₃ and cADPR cause Ca²⁺ release by activation of IP₃-dependent receptors (IP₃Rs) and ryanodine receptors (RyRs), respectively, NAADP may activate a novel Ca²⁺ channel in the acidic compartment⁵³ or open RyRs via a mechanism distinct from that of cADPR.⁵⁴ The exact location of the acidic store is still uncertain, with endosomes and lysosomes as possible candidates;⁵³ however, the most likely association has been made with zymogen granules (ZGs), which store the inactive digestive enzyme precursors.^{52, 55} Our recent results show that bile acids, in addition to inducing Ca²⁺ release from the endoplasmic reticulum (ER) store, are able to stimulate the acidic store in the apical ZG area.⁵⁶ Using a novel two-photon permeabilisation technique, we have found that the bile acid taurolithocholic acid sulphate (TLC-S) specifically activated RyRs, via an NAADP-dependent mechanism, to release stored Ca²⁺ (Figure 1b).

It is generally recognised that the crucial step in the development of pancreatitis is activation of precursor enzymes in the zymogen granules.¹ A high intra-granular concentration of Ca²⁺ is required for stability of granule contents,⁵⁷ most of which is tightly bound together with H⁺ ions within the granular matrix. IP₃ and cADPR induce Ca²⁺ release from the zymogen granules,⁵⁵ a feature common to other types of secretory granules,^{58, 59} and it is feasible that a local perigranular rise of Ca²⁺ induced by these second messengers, or by bile acids acting at RyRs,⁵⁶ would in turn activate Ca²⁺-dependent K⁺ channels present in the granular membrane, permitting the uptake of K⁺ into the granule. Since the matrix behaves as an ion exchanger,⁵⁹ Ca²⁺ and H⁺ would be replaced by K⁺, causing disaggregation of the matrix that may facilitate activation of trypsinogen to trypsin. While disaggregation is necessary for normal secretion, excessive cytosolic Ca²⁺ concentrations may induce premature disaggregation and pathological intracellular digestive enzyme activation, a hypothesis that awaits thorough evaluation in pancreatic acinar cells.

Recently, we have directly studied the effects of acute application of TLC-S on pancreatic acinar cell fate, and have shown that it causes caspase activation, consistent with induction of the apoptotic death pathway (Figure 1c).³⁶ Interestingly, this action is markedly potentiated when an endogenous detoxifying enzyme NAD(P)H:quinone oxidoreductase (NQO1; DT-diaphorase) is inhibited, and our results suggest that acute generation of ROS by bile acids may be important for the promotion of pancreatic acinar cell death. Bile acids are recognised precipitants of acute pancreatitis the detrimental actions of which may involve oxidant stress in vivo,^{51, 60} and our results are in accordance with a recent detailed study in hepatocytes, showing that TLC-S activated caspases 8, 9 and 3 via NADPH oxidase-mediated ROS production.⁶¹

Oxidant stress: Ca²⁺, ROS and apoptosis

Oxidative stress has been implicated in the development of acute pancreatitis in diverse animal experimental models, including fatty acid infusion, ischaemia, pancreatic duct obstruction, gallstone pancreatitis and alcohol ingestion,^{60, 62, 63, 64} and measured in patients with mild and severe acute pancreatitis.⁶⁵ Previously, we have demonstrated the vital role of Ca²⁺ on MPTP opening in pancreatic acinar cell apoptosis induced by the oxidant menadione.¹⁰ Crucially, the Ca²⁺ chelator BAPTA prevented both menadione-induced repetitive cytosolic Ca²⁺ spikes and apoptosis, instigated via the intrinsic apoptotic pathway. Our study suggested that the characteristics of Ca²⁺ signals generated by menadione and physiological secretagogues might underlie differences in their effects on cell fate. For example, physiological concentrations of CCK and acetylcholine (Ach) induce oscillatory cytosolic Ca²⁺ rises, which were initiated in the apical secretory granular pole, and the spread of which to the basolateral area was substantially delayed and diminished by the mitochondrial perigranular and perinuclear buffer barriers (Figure 2a);^{25, 66} such stimulation did not cause apoptosis. In contrast, menadione, which elicited apoptosis, produced Ca²⁺ transients that were also generated in the apical pole, but which rapidly spread to the basal and nuclear regions of the acinar cell, indicating essential differences between the oxidant and the physiological secretagogues. Menadione was found to elicit partial, transient mitochondrial depolarisation that was inhibited by BAPTA, and by bongkrekic acid, an inhibitor of the MPTP,⁶⁷ suggesting that Ca²⁺-dependent MPTP opening in pancreatic acinar cells and consequent mitochondrial inhibition might explain the rapid spread of menadione-induced Ca²⁺ waves. Interestingly, when inhibition of mitochondria with antimycin A was imposed, ACh-induced Ca²⁺ reponses effectively mimicked those of menadione; however, apoptosis did not occur¹⁰ suggesting that other factors may contribute to menadione-induced pancreatic acinar cell death (Figure 2a).

Figure 2

(A) Schematic model illustrating the differences in Ca²⁺ homeostasis and induction of the PTP after stimulation with (a) menadione, (b) ACh or (c) stimulation with ACh plus antimycin A. Generation of ROS by menadione may facilitate opening of the mitochondrial PTP and spread of Ca²⁺ waves from the apical to basolateral regions of the cell (BA, bongkrekic acid; PS, phosphatidylserine; ROS, reactive oxygen species; UN, Ca²⁺ uniporter).¹⁰ (B) (i) Transmitted light and CM-H₂DCFDA fluorescence images of a doublet of acinar cells, showing ROS generation induced by menadione (30 μM) leads to (iii) apoptosis (measured with annexin V FITC) in pancreatic acinar cells. This effect of menadione on ROS and cell fate was significantly greater when NQO1 was inhibited by DMN, whereas DMN alone did not generate ROS or cause apoptosis per se³⁶

We have recently demonstrated that generation of ROS is essential for oxidant-induced apoptosis of pancreatic acinar cells.³⁶ Quinones, such as menadione, enter fast redox cycles within the cell, which consume NAD(P)H and produce ROS. Inhibition of the enzyme NQO1, to prevent menadione detoxification by two-electron reduction, potentiated both ROS generation and consequent apoptosis, while the novel NQO1 inhibitor 4-dimethoxy-2-methylnaphthalene (DMN), a menadione analogue designed not to undergo redox cycling, neither produced ROS nor affected cell fate per se (Figures 2b and 3). NQO1 is thought to be an important cellular defence mechanism to counteract electrophile and oxidant damage,⁶⁸ possibly by maintaining co-enzyme Q in a reduced, anti-oxidant state.⁶⁹ It is overexpressed in acute pancreatitis and many cancers including pancreatic adenocarcinoma,⁷⁰ and may be an important early biomarker of disease. A recent study has shown that increased expression of NQO1 reduced ROS generation induced by tert-butyl hydroperoxide and also suppressed tumour necrosis factorα- and interferonγ-induced NO production via iNOS,⁷¹ while menadione-induced toxicity was augmented in NQO1-deficient mice.⁷² Since earlier observations have indicated that transient Ca²⁺ signals alone are insufficient to induce opening of the MPTP and apoptosis per se,¹⁰ our recent data strongly suggest generation of ROS may constitute an important additional component that promotes acinar cell death. This is in accord with a previous study in hepatocytes, demonstrating MPTP opening via menadione-induced oxidative stress⁷³ and consistent with a model in which oxidation of MPTP components sensitises Ca²⁺-dependent pore opening.^{74, 75}

Figure 3

Structural modification of the menadione molecule to prevent redox cycling and inhibit NAD(P)H:Quinone oxidoreductase (NQO1). (a) Molecular modelling of putative interactions between NQO1 enzyme and menadione (upper) and DMN (lower) suggests that menadione and DMN are flexibly docked into the active site of NQO1 (in close proximity to the bound FADH₂ – blue), utilising the N¹ and N⁵ of the FADH₂ and the N of Gly-150 for increased interaction. In the case of menadione, the O¹ and O⁴ positions of menadione are in close proximity to N¹-FADH₂ and N⁵-FADH₂ (3.6 and 3.5 Å, respectively) enabling electron transfer to occur. However, no electron transfer is feasible for DMN, although it is stabilised by FADH₂ (MeO¹-N¹FADH₂, 3.6 Å; MeO⁴-N⁵FADH₂, 3.5 Å) and Gly-150 (MeO¹-N-Gly, 3.5 Å) interactions, and consequently DMN inhibits the effects of menadione at this site.³⁶ (b) Proposed model for the mechanism of action of DMN. Metabolism of menadione by one-electron (1e/H⁺) reducing enzymes generates an unstable semiquinone radical, with further reduction to the stable hydroquinone; back oxidation generates ROS (O₂⁻) when oxygen is present. Menadione may also be metabolised by one-step, two-electron (2e/2H⁺) reduction via NQO1 directly to the hydroquinone, with no ROS production. Inhibition of NQO1 by DMN causes preferential metabolism of menadione by one-electron reductive processes leading to enhanced ROS generation³⁶

Cell fate and energetics: the importance of mitochondrial ATP production

Under physiological conditions mitochondria respond to local increases of cytosolic Ca²⁺ by generating ATP via stimulus-metabolism coupling.^{13, 14, 16} However, Ca²⁺ overload can drastically reduce ATP production and this may constitute a vital switch between apoptosis and necrosis that ultimately determines cell fate.⁷⁶ It has been suggested that promotion of necrosis through ATP depletion might in part be mediated via an inability of the apoptosome to activate the initiator caspase 9.¹⁸ Recently, an important mechanism whereby alcohol may induce Ca²⁺-dependent necrotic acinar cell death has been identified.¹² Fatty acid ethyl esters (FAEEs), non-oxidative metabolites of ethanol, are generated at higher concentrations within the pancreas than any other organ,^{77, 78, 79} and unlike ethanol per se, are able to induce experimental pancreatitis in vivo.⁸⁰ Non-oxidative ethanol metabolites induce persistent, global, cytosolic Ca²⁺ signals in a concentration-dependent manner,¹¹ initiated via IP₃ receptor-mediated Ca²⁺ release and sustained by depolarisation of mitochondria,¹² the organelle at which FAEE accumulation and hydrolysis to fatty acids (FA) is thought to occur.⁸¹ The consequent mitochondrial impairment leads to a depletion of intracellular ATP, causing run down of the SERCA and plasma membrane Ca²⁺-dependent pumps and consequent inadequate clearance of raised cytosolic Ca²⁺ (Figure 4).

Figure 4

Excess fatty acids, which are formed from the breakdown of non-oxidative ethanol metabolites (FAEEs) in mitochondria,^{12, 81} induce sustained increases of cytosolic Ca²⁺ and inhibit mitochondrial function. Palmitoleic acid (POA; 50–100 μM) (a) depolarised mitochondrial membrane potential (see light-transmitted and TMRM fluorescence images (left), and graph red trace) and elevated cytosolic [Ca²⁺] (Fluo4 fluorescence blue trace) measured simultaneously in dual-loaded pancreatic acinar cells, (b) concomitantly decreased NADH (red) and increased FAD autofluorescence (green) in the perigranular mitochondrial region, (c) depleted cellular ATP, seen as an increase in Mg Green fluorescence; subsequent addition of the protonophore CCCP, which depolarises the inner mitochondrial membrane, caused no further change. (d) Provision of supplementary ATP to the interior of the cell, via patch-pipette, prevented the rise of cytosolic Ca²⁺ induced by POA in the patched cell (blue), whereas a typical sustained Ca²⁺ response was obtained in a nearby non-patched cell (red) that did not receive ATP¹²

Interestingly, FAEE-induced mitochondrial impairment in pancreatic acinar cells occurs as a result of the formation of FAs from FAEE hydrolysis, since FAEE esterase inhibition prevents FAEE-induced mitochondrial impairment, allowing ATP to be generated and thus protecting the cell from cytosolic Ca²⁺ overload.¹² This mechanism explains not only how ethanol may induce severe acute pancreatitis through mitochondrial inhibition but also provides a basis for acinar cell injury in pancreatitis induced by hyperlipidaemia, a recognised risk factor for the disease. In accord, it has previously been demonstrated that infusion of oleic acid, to induce acute pancreatitis in vivo, caused dramatic decreases of intracellular ATP,⁶² a feature also common to cerulein hyperstimulation.⁸² In the cerulein model of pancreatitis, isolated mitochondria exhibit damage, including swelling and disruption of cristae.⁸³ The importance of a decline of the ATP : ADP ratio in pancreatic acinar cells has also been shown recently from experiments in which energy-dependent necrosis is promoted by endotoxin following chronic alcohol exposure in rats.⁸⁴ Furthermore, bile salts induce prolonged, global, cytosolic Ca²⁺ signals³¹ that are associated with mitochondrial depolarisation,³² although this effect on mitochondria appears less pronounced than with non-oxidative alcohol metabolites and may indicate important differences between the toxins. Such mitochondrial inhibition appears to provoke compensatory protective measures in the cell, including an upregulation of mitochondrial ATP synthase, observed after both cerulein hyperstimulation and chronic alcohol exposure.⁸⁵ The importance of ATP depletion for pancreatic acinar cell fate is further underscored by experiments in which addition of ATP to the cell interior, administered via a patch pipette, was able to reverse the detrimental Ca²⁺ signals induced by alcohol metabolites. For example, FA-induced sustained cytosolic Ca²⁺ rises, via the release from ER Ca²⁺ stores and subsequent Ca²⁺ entry, were completely abolished in cells receiving supplementary ATP, whereas control cells produced large, sustained elevations of cytosolic Ca²⁺ (Figure 4)¹² that cause cellular necrosis.¹¹

Conclusions

It is clear that Ca²⁺ signalling is tightly regulated within subcellular microdomains in the pancreatic acinar cell for normal physiological processes,²⁴ and evidence suggests that different patterns of cytosolic Ca²⁺ rises influence both apoptotic and necrotic cell death pathways. The balance between these two principal types of cell death might influence the severity of acute pancreatitis; however, whether induction of apoptosis would be beneficial in a clinical setting remains unproven. The current data in pancreatic acinar cells indicate that transient release of Ca²⁺ from the ER and acidic stores, induced by mild stimuli, such as oxidant stress, promotes apoptosis via the intrinsic pathway, when an additional factor, for example, the generation of ROS is present. This action may depend on a partial mitochondrial depolarisation and transient opening of the MPTP, which does not adversely influence ATP production.⁴¹ More severe insults, on the other hand, cause depletion of Ca²⁺ stores with the induction of sustained global Ca²⁺ elevations that inhibit mitochondrial function with a consequent drastic fall of ATP production, paralysing energy-dependent processes such as the plasmalemmal and ER Ca²⁺ pumps, and also prematurely activate digestive enzymes (Figure 5). Interventions that address either inhibition of sustained Ca²⁺ rises or protection of mitochondrial function may prove beneficial in the treatment of acute pancreatitis.

Figure 5

A simplified schematic model illustrating how cytosolic Ca²⁺ signals might influence cell fate in the pancreatic acinar cell. Oscillatory global rises of cytosolic Ca²⁺, triggered by release of Ca²⁺ from endoplasmic reticulum (ER) and acidic Ca²⁺ stores by moderate stress to the cell, for example, menadione, cause transient mitochondrial depolarisation and promote apoptosis, when additional factors such as ROS generation are present; endogenous detoxifying protective mechanisms such as NQO1 may influence the outcome. More severe insults to the cell, however, cause depletion of the ER/acidic stores, sustained pathological elevations of cytosolic [Ca²⁺] dependent on extracellular Ca²⁺ entry, which lead to irreversible inhibition of mitochondrial function, cellular ATP depletion and paralysis of energy-dependent Ca²⁺ pumps, and to premature activation of pancreatic digestive enzymes. The net effect of such changes, induced by known precipitants of acute pancreatitis, is necrotic death of the pancreatic acinar cell. It should be noted that potentially contradictory data regarding the effects of bile acids are apparent in the literature. Bile acids induce cytosolic Ca²⁺ signals that are inherently variable³² and may be associated with apoptosis (especially under conditions of NQO1 inhibition³⁶) or apparent necrosis,³⁰ differences which may relate to the type and combination of bile acids used and/or level of stimulation