Mitochondrial Ca(2+) Processing by a Unit of Mitochondrial Ca(2+) Uniporter and Na(+)/Ca(2+) Exchanger Supports the Neuronal Ca(2+) Influx via Activated Glutamate Receptors.

The current study demonstrates that in hippocampal neurons mitochondrial Ca(2+) processing supports Ca(2+) influx via ionotropic glutamate (Glu) receptors. We define mitochondrial Ca(2+) processing as Ca(2+) uptake via mitochondrial Ca(2+) uniporter (MCU) combined with subsequent Ca(2+) release via mitochondrial Na(+)/Ca(2+) exchanger (NCX). Our tool is to measure the Ca(2+) influx rate in primary hippocampal co-cultures, i.e. neurons and astrocytes, by fluorescent digital microscopy, using a Fura-2-quenching method where we add small amounts of Mn(2+) in the superfusion medium. Thus, Ca(2+) influx is measured with Mn(2+) in the bath. Ru360 as inhibitor of mitochondrial Ca(2+) uptake through MCU strongly reduces the rate of Ca(2+) influx in Glu-stimulated primary hippocampal neurons. Similarly, the Ca(2+) influx rate in Glu-stimulated neurons declines after suppression of potential-dependent MCU, when we depolarize mitochondria with rotenone. With inhibition of Ca(2+) release from mitochondria via NCX using CGP-37157 the Ca(2+) influx via N-methyl-D-aspartate (NMDA)- and kainate-sensitive receptors is slowed down. Working jointly as mitochondrial Ca(2+) processing unit, MCU and NCX, apparently sustain the Ca(2+) throughput of activated Glu-sensitive receptors. Our results revise the role frequently attributed to mitochondria in neuronal Ca(2+) homeostasis, where mitochondria function mainly as Ca(2+) buffer, and prevent excessively high cytosolic Ca(2+) concentration increase during neuronal activity. The mechanism to control Ca(2+) influx in neurons, as discovered in this study, highlights mitochondrial Ca(2+) processing as a promising pharmacological target. We discuss this pathway in relation to the endoplasmic reticulum-related mechanisms of Ca(2+) processing.

Putative roles of Ca(2+) -independent phospholipase A2 in respiratory chain-associated ROS production in brain mitochondria: influence of docosahexaenoic acid and bromoenol lactone.

Ca(2+) -independent phospholipase A2 (iPLA2 ) is hypothesized to control mitochondrial reactive oxygen species (ROS) generation. Here, we modulated the influence of iPLA2 -induced liberation of non-esterified free fatty acids on ROS generation associated with the electron transport chain. We demonstrate enzymatic activity of membrane-associated iPLA2 in native, energized rat brain mitochondria (RBM). Theoretically, enhanced liberation of free fatty acids by iPLA2 modulates mitochondrial ROS generation, either attenuating the reversed electron transport (RET) or deregulating the forward electron transport of electron transport chain. For mimicking such conditions, we probed the effect of docosahexaenoic acid (DHA), a major iPLA2 product on ROS generation. We demonstrate that the adenine nucleotide translocase partly mediates DHA-induced uncoupling, and that low micromolar DHA concentrations diminish RET-dependent ROS generation. Uncoupling proteins have no effect, but the adenine nucleotide translocase inhibitor carboxyatractyloside attenuates DHA-linked uncoupling effect on RET-dependent ROS generation. Under physiological conditions of forward electron transport, low micromolar DHA stimulates ROS generation. Finally, exposure of RBM to the iPLA2 inhibitor bromoenol lactone (BEL) enhanced ROS generation. BEL diminished RBM glutathione content. BEL-treated RBM exhibits reduced Ca(2+) retention capacity and partial depolarization. Thus, we rebut the view that iPLA2 attenuates oxidative stress in brain mitochondria. However, the iPLA2 inhibitor BEL has detrimental activities on energy-dependent mitochondrial functions. The Ca(2+) -independent phospholipase A2 (iPLA2 ), a FFA (free fatty acids)-generating membrane-attached mitochondrial phospholipase, is potential to regulate ROS (reactive oxygen species) generation by mitochondria. FFA can either decrease reversed electron transport (RET)-linked or enhance forward electron transport (FET)-linked ROS generation. In the physiological mode of FET, iPLA2 activity increases ROS generation. The iPLA2 inhibitor BEL exerts detrimental effects on energy-dependent mitochondrial functions.

Severe disturbance in the Ca2+ signaling in astrocytes from mouse models of human infantile neuroaxonal dystrophy with mutated Pla2g6.

Infantile neuroaxonal dystrophy (INAD; OMIM #no. 256600) is an inherited degenerative nervous system disorder characterized by nerve abnormalities in brain, spinal cord and peripheral nerves. About 85% of INAD patients carry mutations in the PLA2G6 gene that encodes for a Ca(2+)-independent phospholipase A(2) (VIA iPLA(2)), but how these mutations lead to disease is unknown. Besides regulating phospholipid homeostasis, VIA iPLA(2) is emerging with additional non-canonical functions, such as modulating store-regulated Ca(2+) entry into cells, and mitochondrial functions. In turn, defective Ca(2+) regulation could contribute to the development of INAD. Here, we studied possible changes in ATP-induced Ca(2+) signaling in astrocytes derived from two mutant strains of mice. The first strain carries a hypomorphic allele of the Pla2g6 that reduces transcript levels to 5-10% of that observed in wild-type mice. The second strain carries a point mutation in Pla2g6 that results in inactive VIA iPLA(2) protein with postulated gain in toxicity. Homozygous mice from both strains develop pathology analogous to that observed in INAD patients. The nucleotide ATP is the most important transmitter inducing Ca(2+) signals in astroglial networks. We demonstrate here a severe disturbance in Ca(2+) responses to ATP in astrocytes derived from both mutant mouse strains. The duration of the Ca(2+) responses in mutant astrocytes was significantly reduced when compared with values observed in control cells. We also show that the reduced Ca(2+) responses are probably due to a reduction in capacitative Ca(2+) entry (2.3-fold). Results suggest that altered Ca(2+) signaling could be a central mechanism in the development of INAD pathology.

Proinflammatory treatment of astrocytes with lipopolysaccharide results in augmented Ca2+ signaling through increased expression of via phospholipase A2 (iPLA2).

Many Ca(2+)-regulated intracellular processes are involved in the development of neuroinflammation. However, the changes of Ca(2+) signaling in the brain under inflammatory conditions were hardly studied. ATP-induced Ca(2+) signaling is a central event of signal transmission in astrocytic networks. We investigated primary astrocytes after proinflammatory stimulation with lipopolysaccharide (LPS; 100 ng/ml) for 6-24 h. We reveal that Ca(2+) responses to purinergic ATP stimulation are significantly increased in amplitude and duration after stimulation with LPS. We detected that increased amplitudes of Ca(2+) responses to ATP in LPS-treated astrocytes can be explained by substantial increase of Ca(2+) load in stores in endoplasmic reticulum. The mechanism implies enhanced Ca(2+) store refilling due to the amplification of capacitative Ca(2+) entry. The reason for the increased duration of Ca(2+) responses in LPS-treated cells is also the amplified capacitative Ca(2+) entry. Next, we established that the molecular mechanism for the LPS-induced amplification of Ca(2+) responses in astrocytes is increased expression and activity of VIA phospholipase A(2) (VIA iPLA(2)). Indeed, both gene silencing with specific small interfering RNA and pharmacological inhibition of VIA iPLA(2) with S-bromoenol lactone reduced the load of the Ca(2+) stores and caused a decrease in the amplitudes of Ca(2+) responses in LPS-treated astrocytes to values, which were comparable with those in untreated cells. Our findings highlight a novel regulatory role of VIA iPLA(2) in development of inflammation in brain. We suggest that this enzyme might be a possible target for treatment of pathologies related to brain inflammation.

Neurons and astrocytes in an infantile neuroaxonal dystrophy (INAD) mouse model show characteristic alterations in glutamate-induced Ca2+ signaling.

INAD (infantile neuroaxonal dystrophy, OMIM#256600), an autosomal recessive inherited degenerative disease, is associated with PLA2G6 mutations. PLA2G6 encodes Ca2+-independent phospholipase A2 (VIA iPLA2). However, it is unclear how the PLA2G6-mutations lead to disease. Non-canonical functions, which were suggested for VIA iPLA2, such as regulation of cellular and mitochondrial Ca2+ are promising candidates. Therefore, we investigate glutamate (Glu)-evoked Ca2+ signals in neurons and astrocytes in co-culture obtained from three INAD mouse model strains with Pla2g6 mutations, (i) hypomorphic Pla2g6 allele with reduced transcript levels, (ii) knocked-out Pla2g6, and (iii) (G373R)-point mutation with inactive VIA iPLA2 enzyme. Homozygous offspring from these strains develop pathology similar to that observed in INAD patients. We found that in mouse neurons the Pla2g6 mutation disrupted the dependency of Glu-induced extracellular Ca2+ influx on mitochondrial Ca2+ uptake. Thus, in neurons with Pla2g6 mutation we did not detect the characteristic reduction in Glu-induced Ca2+ influx upon treatment with Ru360, a blocker of mitochondrial Ca2+ uniporter, or with rotenone. In contrast to neurons, in astrocytes, both with Pla2g6 mutation or wild-type cells, the treatment with Ru360 or rotenone reduced the rate of Glu-induced Ca2+ influx ∼2-fold. This Ca2+ influx in astrocytes represents capacitative Ca2+ entry. In astrocytes with Pla2g6 mutation, the Glu-induced Ca2+ influx was ∼2-fold lower than in wild-type controls. We suggest that this is the mechanism for strongly decreased durations of Glu-induced Ca2+ responses in astrocytes with Pla2g6 mutation. We could mimic the mutation by pharmacological inhibition of iPLA2 using S-BEL. Thus, lack of VIA iPLA2 activity caused effects in astrocytes. In summary, three INAD mouse models show comparable changes in Glu-induced Ca2+ signaling, but specific for neurons or astrocytes. This finding helps to identify pathways altered during INAD and highlights non-canonical VIA iPLA2 functions, like regulation of cellular Ca2+ fluxes by mitochondria or capacitative Ca2+-entry.

Mitochondria from a mouse model of the human infantile neuroaxonal dystrophy (INAD) with genetic defects in VIA iPLA2 have disturbed Ca(2+) regulation with reduction in Ca(2+) capacity.

Mutations in the PLA2G6 gene which encodes Ca(2+)-independent phospholipase A2 (VIA iPLA2) were detected in 85% of cases of the inherited degenerative nervous system disorder INAD (infantile neuroaxonal dystrophy, OMIM #256600). However, molecular mechanisms linking these mutations to the disease progression are unclear. VIA iPLA2 is expressed also in mitochondria. Here, we investigate Ca(2+) handling by brain mitochondria derived from mice with hypomorph Pla2g6 allele. These animals with reduced transcript levels (5% of wild type) represent a suitable model for INAD. We demonstrated significant reduction of Ca(2+) uptake rate and Ca(2+) retention capacity in brain mitochondria isolated from this mutant. This phenotype could be mimicked when in wild-type controls VIA iPLA2 was inhibited by S-BEL. Importantly, the reduction could be ameliorated partly by addition of the VIA iPLA2 product, sn-2 lysophosphatidyl-choline. Furthermore, we demonstrated in situ a reduced mitochondrial potential in neurons from mice deficient in VIA iPLA2, which could cause the reduced Ca(2+) uptake rate via the potential-dependent mitochondrial Ca(2+) uniporter. Thus, the disturbances in mitochondrial potential and the changes in Ca(2+) handling were dependent on VIA iPLA2 activity. Reduced mitochondrial Ca(2+) uptake rate and Ca(2+) retention capacity might result in increased vulnerability of mitochondria to the Ca(2+) overload and in disturbed cellular Ca(2+) signaling during INAD. For VIA iPLA2, non-canonical functions beyond sole phospholipid turnover seem to be important, such as regulation of store-operated Ca(2+) entry in cells. Thus, our findings bring new insight into molecular mechanism affected in INAD and highlight the non-canonical function of VIA iPLA2 in regulation of mitochondrial Ca(2+) handling.

Arachidonic acid in astrocytes blocks Ca(2+) oscillations by inhibiting store-operated Ca(2+) entry, and causes delayed Ca(2+) influx.

ATP-elicited oscillations of the concentration of free intracellular Ca(2+) ([Ca(2+)](i)) in rat brain astrocytes were abolished by simultaneous arachidonic acid (AA) addition, whereas the tetraenoic analogue 5,8,11,14-eicosatetraynoic acid (ETYA) was ineffective. Inhibition of oscillations is due to suppression by AA of intracellular Ca(2+) store refilling. Short-term application of AA, but not ETYA, blocked Ca(2+) influx, which was evoked by depletion of stores with cyclopiazonic acid (CPA) or thapsigargin (Tg). Addition of AA after ATP blocked ongoing [Ca(2+)](i) oscillations. Prolonged AA application without or with agonist could evoke a delayed [Ca(2+)](i) increase. This AA-induced [Ca(2+)](i) rise developed slowly, reached a plateau after 5 min, could be reversed by addition of bovine serum albumin (BSA), that scavenges AA, and was blocked by 1 microM Gd(3+), indicative for the influx of extracellular Ca(2+). Specificity for AA as active agent was demonstrated by ineffectiveness of C16:0, C18:0, C20:0, C18:2, and ETYA. Moreover, the action of AA was not affected by inhibitors of oxidative metabolism of AA (ibuprofen, MK886, SKF525A). Thus, AA exerted a dual effect on astrocytic [Ca(2+)](i), firstly, a rapid reduction of capacitative Ca(2+) entry thereby suppressing [Ca(2+)](i) oscillations, and secondly inducing a delayed activation of Ca(2+) entry, also sensitive to low Gd(3+) concentration.

Docosahexaenoic acid and arachidonic acid release in rat brain astrocytes is mediated by two separate isoforms of phospholipase A2 and is differently regulated by cyclic AMP and Ca2+.

1. Docosahexaenoic acid (DHA) and arachidonic acid (AA), polyunsaturated fatty acids (PUFAs), are important for central nervous system function during development and in various pathological states. Astrocytes are involved in the biosynthesis of PUFAs in neuronal tissue. Here, we investigated the mechanism of DHA and AA release in cultured rat brain astrocytes. 2. Primary astrocytes were cultured under standard conditions and prelabeled with [(14)C]DHA or with [(3)H]AA. Adenosine 5'-triphosphate (ATP) (20 micro M applied for 15 min), the P2Y receptor agonist, stimulates release of both DHA (289% of control) and AA (266% of control) from astrocytes. DHA release stimulated by ATP is mediated by Ca(2+)-independent phospholipase A(2) (iPLA(2)), since it is blocked by the selective iPLA(2) inhibitor 4-bromoenol lactone (BEL, 5 micro M) and is not affected either by removal of Ca(2+) from extracellular medium or by suppression of intracellular Ca(2+) release through PLC inhibitor (U73122, 5 micro M). 3. AA release, on the other hand, which is stimulated by ATP, is attributed to Ca(2+)-dependent cytosolic PLA(2) (cPLA(2)). AA release is abolished by U73122 and, by removal of extracellular Ca(2+), is insensitive to BEL and can be selectively suppressed by methyl arachidonyl fluorophosphonate (3 micro M), a general inhibitor of intracellular PLA(2) s. 4. Western blot analysis confirms the presence in rat brain astrocytes of 85 kDa cPLA(2) and 40 kDa protein reactive to iPLA(2) antibodies. 5. The influence of cAMP on regulation of PUFA release was investigated. Release of DHA is strongly amplified by the adenylyl cyclase activator forskolin (10 micro M), and by the protein kinase A (PKA) activator dibutyryl-cAMP (1 mM). In contrast, release of AA is not affected by forskolin or dibutyryl-cAMP, but is almost completely blocked by 2,3-dideoxyadenosine (20 micro M) and inhibited by 34% by H89 (10 micro M), inhibitors of adenylyl cyclase and PKA, respectively. 6. Other neuromediators, such as bradykinin, glutamate and thrombin, stimulate release of DHA and AA, which is comparable to the release stimulated by ATP. 7. Different sensitivities of iPLA(2) and cPLA(2) to Ca(2+) and cAMP reveal new pathways for the regulation of fatty acid release and reflect the significance of astrocytes in control of DHA and AA metabolism under normal and pathological conditions in brain.

Role of Ca2+-independent phospholipase A2 and n-3 polyunsaturated fatty acid docosahexaenoic acid in prostanoid production in brain: perspectives for protection in neuroinflammation.

Various diseases of the central nervous system are characterized by induction of inflammatory events, which involve formation of prostaglandins. Production of prostaglandins is regulated by activity of phospholipases A(2) and cyclooxygenases. These enzymes release the prostaglandin precursor, the n-6 polyunsaturated fatty acid, arachidonic acid and oxidize it into prostaglandin H(2). Docosahexaenoic acid, which belongs to the n-3 class of polyunsaturated fatty acids, was shown to reduce production of prostaglandins after in vivo and in vitro administration. Nevertheless, the fact that in brain tissue cellular phospholipids naturally have a uniquely high content of docosahexaenoic acid was ignored so far in studies of prostaglandin formation in brain tissue. We consider the following possibilities: docosahexaenoic acid might attenuate production of prostaglandins by direct inhibition of cyclooxygenases. Such inhibition was found with the isolated enzyme. Another possibility, which has been already shown is reduction of expression of inducible cyclooxygenase-2. Additionally, we propose that docosahexaenoic acid could influence intracellular Ca(2+) signaling, which results in changes of activity of Ca(2+)-dependent phospholipase A(2), hence reducing the amount of arachidonic acid available for prostaglandin production. Astrocytes, the main type of glial cells in the brain control the release of arachidonic acid, docosahexaenoic acid and the formation of prostaglandins. Our recently obtained data revealed that the release of arachidonic and docosahexaenoic acids in astrocytes is controlled by different isoforms of phospholipase A(2), i.e. Ca(2+)-dependent phospholipase A(2) and Ca(2+)-independent phospholipase A(2), respectively. Moreover, the release of arachidonic and docosahexaenoic acids is differently regulated through Ca(2+)- and cAMP-dependent signal transduction pathways. Based on analysis of the current literature and our own data we put forward the hypothesis that Ca(2+)-independent phospholipase A(2) and docosahexaenoic acid are promising targets for treatment of inflammatory related disorders in brain. We suggest that Ca(2+)-independent phospholipase A(2) and docosahexaenoic acid might be crucially involved in brain-specific regulation of prostaglandins.

Peroxisome proliferator-activated receptor (PPAR)ß/δ, a possible nexus of PPARα- and PPARγ-dependent molecular pathways in neurodegenerative diseases: Review and novel hypotheses.

Peroxisome proliferator-activated receptors (PPARα, -ß/δ and -γ) are lipid-activated transcription factors. Synthetic PPARα and PPARγ ligands have neuroprotective properties. Recently, PPARß/δ activation emerged as the focus of a novel approach for the treatment of a wide range of neurodegenerative diseases. To fill the gap of knowledge about the role of PPARß/δ in brain, new hypotheses about PPARß/δ involvement in neuropathological processes are requested. In this paper, we describe a novel hypothesis, claiming the existence of tight interactions between the three PPAR isotypes, which we designate the "PPAR triad". We propose that PPARß/δ has a central control of the PPAR triad. The majority of studies analyze the regulation only by one of the PPAR isotypes. A few reports describe the mutual regulation of expression levels of all three PPAR isotypes by PPAR agonists. Analysis of these studies where pairwise interactions of PPARs were described allows us to support the existence of the PPAR triad with central role for PPARß/δ. In the present review, we propose the hypothesis that in a wide range of brain disorders, PPARß/δ plays a central role between PPARα and PPARγ. Finally, we prove the advantages of the PPAR triad concept by describing hypotheses of PPARß/δ involvement in the regulation of myelination, glutamate-induced neurotoxicity, and signaling pathways of reactive oxygen species/NO/Ca(2+).

Prostaglandin synthesis in rat brain astrocytes is under the control of the n-3 docosahexaenoic acid, released by group VIB calcium-independent phospholipase A2.

In the current study, we reveal that in astrocytes the VIB Ca(2+)-independent phospholipase A(2) is the enzyme responsible for the release of docosahexaenoic acid (22:6n-3). After pharmacological inhibition and siRNA silencing of VIB Ca(2+)-independent phospholipase A(2), docosahexaenoic acid release was strongly suppressed in astrocytes, which were acutely stimulated (30 min) with ATP and glutamate or after prolonged (6 h) stimulation with the endotoxin lipopolysaccharide. Docosahexaenoic acid release proceeds simultaneously with arachidonic acid (20:4n-6) release and prostaglandin liberation from astrocytes. We found that prostaglandin production is negatively controlled by endogenous docosahexaenoic acid, since pharmacological inhibition and siRNA silencing of VIB Ca(2+)-independent phospholipase A(2) significantly amplified the prostaglandin release by astrocytes stimulated with ATP, glutamate, and lipopolysaccharide. Addition of exogenous docosahexaenoic acid inhibited prostaglandin synthesis, which suggests that the negative control of prostaglandin synthesis observed here is likely due to competitive inhibition of cyclooxygenase-1/2 by free docosahexaenoic acid. Additionally, treatment of astrocytes with docosahexaenoic acid leads to the reduction in cyclooxygenase-1 expression, which also contributes to reduced prostaglandin production observed in lipopolysaccharide-stimulated cells. Thus, we identify a regulatory mechanism important for the brain, in which docosahexaenoic acid released from astrocytes by VIB Ca(2+)-independent phospholipase A(2) negatively controls prostaglandin production.

Regulation of intracellular calcium levels by polyunsaturated fatty acids, arachidonic acid and docosahexaenoic acid, in astrocytes: possible involvement of phospholipase A2.

Pathological conditions in the brain, such as ischemia, trauma and seizure are accompanied by increased levels of free n-6 and n-3 polyunsaturated fatty acids (PUFA), mainly arachidonic acid (AA, 20:4n-6) and docosahexaenoic acid (DHA, 22:6n-3). A neuroprotective role has been suggested for PUFA. For investigation of the potential molecular mechanisms involved in neuroprotection by PUFA, we studied the regulation of the concentration of intracellular Ca2+ ([Ca2+]i) in rat brain astrocytes. We evaluated the presence of extracellular PUFA and the release of intracellular PUFA. Interestingly, only the constitutive brain PUFA AA and DHA, but not eicosapentaenoic acid (EPA) had prominent effects on intracellular Ca2+. AA and DHA suppressed [Ca2+]i oscillation, inhibited store-operated Ca2+ entry, and reduced the amplitudes of Ca2+ responses evoked by agonists of G protein-coupled receptors. Moreover, prolonged exposure of astrocytes to AA and DHA brought the cells to a new steady state of a moderately elevated [Ca2+]i level, where the cells became virtually insensitive to external stimuli. This new steady state can be considered as a mechanism of self-protection. It isolates disturbed parts of the brain, because AA and DHA reduce pathological overstimulation in the tissue surrounding the damaged area. In inflammation-related events, frequently AA and DHA exhibit opposite effects. However, in astrocytes AA and DHA exerted comparable effects on [Ca2+]i. Extracellularly added AA and DHA, but not EPA, were also able to induce the release of [3H]AA from prelabeled astrocytes. Therefore, we also suggest the involvement of phospholipase A2 activation and lysophospholipid generation in the regulation of intracellular Ca2+ in astrocytes.

Arachidonic acid and docosahexaenoic acid suppress thrombin-evoked Ca2+ response in rat astrocytes by endogenous arachidonic acid liberation.

Arachidonic (AA) and docosahexaenoic acid (DHA) are the major polyunsaturated fatty acids (PUFAs) in the brain. However, their influence on intracellular Ca2+ signalling is still widely unknown. In astrocytes, the amplitude of thrombin- induced Ca2+ response was time-dependently diminished by AA and DHA, or by the AA tetraynoic analogue ETYA, but not by eicosapentaenoic acid (EPA). Thrombin-elicited Ca2+ response was reduced (20-30%) by 1-min exposure to AA or DHA. Additionally, 1-min application of AA or DHA together with thrombin in Ca2+-free medium blocked Ca2+ influx, which followed after readdition of extracellular Ca2+. EPA and ETYA, however, were ineffective. Long-term treatment of astrocytes with AA and DHA, but not EPA reduced the amplitude of the thrombin-induced Ca2+ response by up to 80%. AA and DHA caused a comparable decrease in intracellular Ca2+ store content. Only DHA and AA, but not EPA or ETYA, caused liberation of endogenous AA by cytosolic phospholipase A2 (cPLA2). Therefore, we reasoned that the suppression of Ca2+ response to thrombin by AA and DHA could be due to release of endogenous AA. Possible participation of AA metabolites, however, was excluded by the finding that specific inhibitors of the different oxidative metabolic pathways of AA were not able to abrogate the inhibitory AA effect. In addition, thrombin evoked AA release via activation of cPLA2. From our data we propose a novel model of positive/negative-feed-back in which agonist-induced release of AA from membrane phospholipids promotes further AA release and then suppresses agonist-induced Ca2+ responses.