Heterologous desensitization of opioid-stimulated Ca2+ increase by bradykinin or ATP in NG108-15 cells

Hsp90: structure and function

Hsp90 is a highly abundant and ubiquitous molecular chaperone which plays an essential role in many cellular processes including cell cycle control, cell survival, hormone and other signalling pathways. It is important for the cell's response to stress and is a key player in maintaining cellular homeostasis. In the last ten years, it has become a major therapeutic target for cancer, and there has also been increasing interest in it as a therapeutic target in neurodegenerative disorders, and in the development of anti-virals and anti-protozoan infections. The focus of this review is the structural and mechanistic studies which have been performed in order to understand how this important chaperone acts on a wide variety of different proteins (its client proteins) and cellular processes. As with many of the other classes of molecular chaperone, Hsp90 has a critical ATPase activity, and ATP binding and hydrolysis known to modulate the conformational dynamics of the protein. It also uses a host of cochaperones which not only regulate the ATPase activity and conformational dynamics but which also mediate interactions with Hsp90 client proteins. The system is also regulated by post-translational modifications including phosphorylation and acetylation. This review discusses all these aspects of Hsp90 structure and function.

Opioid elevation of intracellular free calcium: possible mechanisms and physiological relevance

Opioid receptors are seven transmembrane domain Gi/G0 protein-coupled receptors, the activation of which stimulates a variety of intracellular signalling mechanisms including activation of inwardly rectifying potassium channels, and inhibition of both voltage-operated N-type Ca2+ channels and adenylyl cyclase activity. It is now apparent that like many other Gi/G0-coupled receptors, opioid receptor activation can significantly elevate intracellular free Ca2+ ([Ca2+]i), although the mechanism underlying this phenomenon is not well understood. In some cases opioid receptor activation alone appears to elevate [Ca2+]i, but in many cases it requires concomitant activation of Gq-coupled receptors, which themselves stimulate Ca2+ release from intracellular stores via the inositol phosphate pathway. Given the number of Ca2+-sensitive processes known to occur in cells, there are therefore a myriad of situations in which opioid receptor-mediated elevations of [Ca2+](i) may be important. Here, we review the literature documenting opioid receptor-mediated elevations of [Ca2+]i, discussing both the possible mechanisms underlying this phenomenon and its potential physiological relevance.

Rat cerebellar granule cells are protected from glutamate-induced excitotoxicity by S-nitrosoglutathione but not glutathione

In cultured rat cerebellar granule cells, glutamate or N-methyl-d-aspartate (NMDA) activation of the NMDA receptor caused a sustained increase in cytosolic Ca(2+) levels ([Ca(2+)](i)), reactive oxygen species (ROS) generation, and cell death (respective EC(50) values for glutamate were 12, 30, and 38 microM) but no increase in caspase-3 activity. Removal of extracellular Ca(2+) blocked all three glutamate-induced effects, whereas pretreatment with an ROS scavenger inhibited glutamate-induced cell death but had no effect on the [Ca(2+)](i) increase. This indicates that glutamate-induced cell death is attributable to [Ca(2+)](i) increase and ROS generation, and the [Ca(2+)](i) increase precedes ROS generation. Apoptotic cell death was not seen until 24 h after exposure of cells to glutamate. S-nitrosoglutathione abolished glutamate-induced ROS generation and cell death, and only a transient [Ca(2+)](i) increase was seen; similar results were observed with another nitric oxide (NO) donor, S-nitroso-N-acetylpenicillamine, but not with glutathione, which suggests that the effects were caused by NO. The transient [Ca(2+)](i) increase and the abolishment of ROS generation induced by glutamate and S-nitrosoglutathione were still seen in the presence of an ROS scavenger. Glial cells, which were present in the cultures used, showed no [Ca(2+)](i) increase in the presence of glutamate, and glutamate-induced granule cell death was independent of the percentage of glial cells. In conclusion, NO donors protect cultured cerebellar granule cells from glutamate-induced cell death, which is mediated by ROS generated by a sustained [Ca(2+)](i) increase, and glial cells provide negligible protection against glutamate-induced excitotoxicity.

The mRNA expression of serotonin 2C subtype receptors uncoupled with inositol hydrolysis in NG108-15 cells

Cell culture systems seem to be useful for clarifying the cellular physiological mechanisms of serotonin 2C subtype receptors (5-HT2CR) and related drug action mechanisms. However, there are still few reports about cells that contain intrinsic 5-HT2CR. This report demonstrates by using RT/PCR that 5-HT2CR mRNA exists in splicing variant forms in NGI08-15 cells. The PCR results using a pair of primers that recognized sequences near the third intracellular loop site showed two neighboring bands at about 500 bp upon electrophoresis in acrylamide gels. The sequence analysis demonstrated that one band was the rat 5-HT2CR sequence and the other one was that of the mouse. Serotonin, however, did not enhance the inositol phosphates formation in NG108-15 cells. It has been reported that post-translational modifications of RNA, splicing and editing, occur at the site of the second intracellular loop domain in 5-HT2CR mRNA. Accordingly, a pair of primers that recognized this site were designed. The molecular size of the PCR product was shorter than that expected based on the sequence of the native 5-HT2CR. The fragment lacked the 95 nucleotides of native 5-HT2CR mRNA. This seems to be the reason why serotonin did not enhance inositol phosphates formation in NG108-15 cells.

Distinct effects of different calcium-mobilizing agents on cell death in NG108-15 neuroblastoma X glioma cells

The effects of different calcium-mobilizing agents on cell death were characterized in NG108-15 neuroblastoma x glioma hybrid cells. Carbonyl cyanide p-trifluoromethoxyphenylhydrazone (FCCP) increased the cytosolic Ca(2+) concentration ([Ca(2+)](i)) and caused cell death. Thapsigargin (TG) not only increased the [Ca(2+)](i) and caused cell death but also induced neurite outgrowth via activation of phospholipase A(2) and cytochrome P450 epoxygenase. In contrast, bradykinin increased the [Ca(2+)](i), but had no effect on cell morphology or cell death. Cell death occurred by two different mechanisms, one of which was caspase-3-dependent and the other caspase-3-independent. Caspase-3 activation was Ca(2+)-dependent, whereas neurite outgrowth was Ca(2+)-independent. TG- or FCCP-induced caspase-3 activation occurred at the same time, but the cell death induced by TG was delayed. TG treatment did not enhance the generation of nitric oxide or cAMP or secretion of glial-derived neurotrophic factor or neurotrophin-3, but activated sphingosine kinase. Furthermore, inhibition of sphingosine kinase accelerated TG-induced cell death, and exogenous sphingosine 1-phosphate (S1P) protected cells from FCCP-induced cell death by about 60%. These results indicate that, in these cells, depletion of intracellular nonmitochondrial or mitochondrial Ca(2+) stores causes cell death, that TG activates phospholipase A(2) and sphingosine kinase, and that arachidonic acid induces neurite outgrowth, whereas S1P delays cell death.

Pharmacological characterisation of pyrimidinoceptor responses in NG108-15 cells

In the present study, the P2Y receptor(s) mediating the effects of the pyrimidines UTP and UDP on phospholipase C activation in the mouse neuroblastoma x rat glioma hybrid cell line NG108-15 was investigated. Reverse Transcriptase-Polymerase Chain Reaction (RT-PCR) analysis detected transcripts for the P2Y(6) and P2Y(2) receptors, but not for P2Y(1) and P2Y(4.) UTP and UDP were equipotent agonists and their effects were partially additive. Suramin, reactive blue 2 and pyridoxal phosphate-6-azophenyl-2',4'disulfonic acid (PPADS) antagonised the phospholipase C response to both UTP and UDP. High micromolar concentrations of adenosine, 2-p-(2-carboxyethyl)phenethylamino-5'-N-ethylcarboxamidoadenosine (CGS-21680), 2',3'-O-isopropylideneadenosine (iPAdo) and adenosine 3':5'-cyclic monophosphate (3',5'-cAMP) were able to antagonise the effect of UTP on phospholipase C but not that of UDP. The additivity of the UTP and UDP responses, novel P2 receptor antagonist profile and the distinguishing action of adenosine may indicate the expression of a pyrimidine selective P2Y receptor in addition to the P2Y(6) type in these cells.

Cyclic AMP regulates the calcium transients released from IP(3)-sensitive stores by activation of rat kappa-opioid receptors expressed in CHO cells

We analyzed intracellular Ca(2+)and cAMP levels in Chinese hamster ovary cells expressing a cloned rat kappa opioid receptor (CHO-kappa cells). Although expression of kappa(kappa)-opioid receptors was confirmed with a fluorescent dynorphin analog in almost all CHO-kappa cells, the kappa-specific agonists, U50488H or U69593, induced a Ca(2+) transient only in 35% of the cells. The Ca(2+) response occurred in all-or-none fashion and the half-maximal dosage of U50488H (812.1nM) was higher than that (3.2nM) to inhibit forskolin-stimulated cAMP. The kappa-receptors coupled to G(i/o)proteins since pertussis toxin significantly reduced the U50488H actions on intracellular Ca(2+) and cAMP. The Ca(2+) transient originates from IP(3)-sensitive internal stores since the Ca(2+) response was blocked by a PLC inhibitor (U73122) or by thapsigargin depletion of internal stores while removal of extracellular Ca(2+) had no effect. Interestingly, application of dibutyryl cAMP (+ 56.2%) or 8-bromo-cAMP (+ 174.7%) significantly increased the occurrence of U50488H-induced Ca(2+) mobilization while protein kinase A (PKA) inhibitors, Rp-cAMP (-32.3%) or myr-psi PKA (-73.9%) significantly reduced the response. Therefore, it was concluded that cAMP and PKA activity can regulate the Ca(2+) mobilization. These results suggest that the kappa receptor-linked cAMP cascade regulates the occurrence of kappa-opioid-mediated Ca(2+) mobilization.

Phosphorylation promotes the desensitization of the opioid-induced Ca2+ increase in NG108-15 cells

Using the fluorescent Ca2+ indicator fura-2, we demonstrated that, in a single NG108-15 cell, acute repetitive challenge with leucine-enkephalin (EK) results in a gradual reduction of the increase of the cytosolic Ca2+ concentration ([Ca2+]i) at agonist exposure times of 90 s or less; increasing the EK exposure time of each challenge from 30 to 90 s results in greater desensitization, with complete desensitization occurring at 90 s exposure. Similar results are seen with ATP. In opioid-desensitized cells, bradykinin can still induce a marked [Ca2+]i increase, while exposure of desensitized cell to 50 mM K+ restores the response EK-induced, suggesting a role of intracellular Ca2+ stores in the desensitization process. Pretreatment of cells with certain protein kinase inhibitors, including N-(2-guanidinoethyl)-5-isoquinolinesulfonamide (HA1004) and staurosporine, prevented desensitization, while others, 1-(5-isoquinolinesulfonyl)-2-methylpiperazine (H-7) and {1-[N, O-bis-(5-isoquinolinesulfonyl)-N-methyl-l-tyrosyl]-4-phenyl-piperazine (KN-62), had no effect. In contrast, activation of protein kinase C by phorbol 12-myristate 13-acetate promoted desensitization. Thus, the desensitization is dependent on protein phosphorylation. HA1004 alone did not alter EK- or bradykinin-induced inositol 1,4, 5-trisphosphate (IP3) generation; however, the inhibitory effect of calyculin A on EK- or bradykinin-induced IP3 generation was reversed by HA1004. In addition, in the presence of HA1004, the blockade of Ca2+ influx by either verapamil or removal of extracellular Ca2+ or the depletion of Ca2+ pools by thapsigargin still led to desensitization, suggesting that phosphorylation does not alter the activity of the Ca2+ transporters involved in Ca2+ influx and Ca2+ release. Our results imply that emptying of intracellular Ca2+ stores and protein phosphorylation in the phospholipase C signaling pathway play roles in the process of desensitization.

Ca2+ signaling induced by sphingosylphosphorylcholine and sphingosine 1-phosphate via distinct mechanisms in rat glomerular mesangial cells

BACKGROUND

To elucidate the molecular mechanism underlying sphingosine 1-phosphate (S1P) and sphingosylphosphorylcholine (SPC) mediated signaling, we compared their effects with those of adenosine triphosphate (ATP) and angiotensin II (Ang II) on the cytosolic free Ca2+ concentration ([Ca2+]i), inositol 1,4, 5-trisphosphate (IP3) generation and arachidonic acid release in rat glomerular mesangial cells.

METHODS

The fluorescent Ca2+ indicator, Fura-2, was used to measure the [Ca2+]i changes in cultured rat glomerular mesangial cells either in suspension or attached to the coverslips.

RESULTS

SPC 5 microM, S1P 5 microM, ATP 100 microM and Ang II 90 nM all induced increases in the [Ca2+]i, and the effect showed marked homologous desensitization, while heterologous desensitization was less. After the initial exposure of the cells to SPC, the increase in [Ca2+]i induced by subsequent addition of ATP or Ang II was only reduced by about 14.3% and 4.8%, respectively. After the initial exposure to S1P, a greater reduction was seen (42. 1% and 47.7%, respectively). Both arachidonic acid release and IP3 generation were activated by all four agonists with an identical rank order of effectiveness of SPC >> S1P > ATP = Ang II; both were pertussis toxin-sensitive and cholera toxin-resistant. The arachidonic acid release induced by all four agonists showed identical susceptibility to removal of extracellular Ca2+, whereas IP3 generation displayed differential extracellular Ca2+ dependence. Only SPC-induced IP3 generation was highly sensitive to extracellular Ca2+ level, and this Ca2+ dependence was abolished after pretreatment of cells with arachidonyl trifluoromethyl ketone (AACOCF3), a phospholipase A2 inhibitor. Furthermore, the Mn2+ influx was markedly greater in SPC-stimulated cells than in either control or other agonist-stimulated cells, and was decreased by prior exposure of cells to AACOCF3. After phospholipase A2 was inhibited or in the absence of extracellular Ca2+, SPC displayed identical effectiveness as S1P on desensitizing the action of ATP or Ang II on the increase in [Ca2+]i. Conclusions. Our results indicate that all four agents primarily activate phospholipase C through their receptor occupancies, but that SPC alone also induces further significant Mn2+ influx and IP3 generation attributable to its primary stimulatory effect on arachidonic acid release. Thus, the heterologous desensitization to ATP or Ang II induced by SPC was less profound than that induced by S1P, since SPC induced a Ca2+ influx.

The renin-angiotensin system in hybrid NG108-15 cells. Renin gene is from mouse neuroblastoma, angiotensinogen and angiotensin-converting enzyme genes are of rat glioma origin

Angiotensin II (Ang II) increases the level of tyrosine phosphorylation of several proteins in nondifferentiated NG108-15 cells, a hybrid derived from the fusion of mouse neuroblastoma and rat glioma cells. Conversely, incubation of NG108-15 cells with an angiotensin-converting enzyme (ACE) inhibitor decreased the basal level of tyrosine phosphorylation of proteins, suggesting that locally secreted Ang II may act as an autocrine regulator. By RT-PCR, we found that nondifferentiated NG108-15 cells contained the mRNA transcript of the rat angiotensinogen, mouse renin and rat ACE genes, thus confirming that NG108-15 cells contain all the elements of a local renin-angiotensin system.

Desensitization of delta-opioid-induced mobilization of Ca2+ stores in NG108-15 cells

Activation of delta-opioid receptors in NG108-15 cells induces the release of calcium from an inositol 1,4,5-trisphosphate- sensitive intracellular store. We used fura-2-based digital imaging to study the effects of prolonged exposure to agonist on opioid-induced increases in [Ca2+]i. Exposure to D-Ala2-E-Leu5 enkephalin (DADLE) (1 microM) for 30 min completely desensitized NG108-15 cells to a second DADLE-induced response. The cells recovered gradually over 25 min following washout of DADLE. The desensitization was not due to depletion of intracellular calcium stores and bradykinin failed to cross-desensitize the DADLE-evoked response, although both agonists mobilized the same Ca2+ store. Desensitization induced by 100 nM DADLE was overcome by a higher concentration of DADLE (100 microM). Treatment with 8-cpt-cAMP (0.1 mM) for 30 min did not influence the DADLE-induced increases in [CA2+]i. Phorbol dibutyrate (PdBu) (1 microM) blocked the response completely. Treatment with the inhibitor of cyclic nucleotide-dependent kinases H8 (1 microM) for 45 min did not prevent DADLE-induced desensitization. Treatment with the protein kinase C (PKC) inhibitors staurosporin (10 nM) and GF-109203X (200 nM) for 45 min reduced desensitization. However, down-regulation of PKC by 24 h exposure to PdBu (1 microM) failed to prevent the DADLE-induced desensitization in NG108-15 cells. Thus, we conclude that multiple pathways participated in desensitization of delta-receptor-mediated Ca2+ mobilization, one of which includes PKC.

Signaling pathways via NK1 receptors and their desensitization in an AR42J cell line

Substance P (SP) has been shown to induce phosphatidylinositol (PI) hydrolysis and Ca2+ mobilization in AR42J cells. In this study, we confirmed the expression of NK1 but not NK2 or NK3 receptors in this cell line, and further investigated signaling pathways via NK1 receptors and their desensitization. The activation of NK1 receptors by SP affected neither basal cyclic AMP level nor cyclic AMP accumulation induced by secretin and forskolin, although it stimulated PI hydrolysis. Furthermore, SP induced Ca2+ mobilization even in the absence of extracellular Ca2+, though maximal response was reduced. U73122, a phospholipase C (PLC) inhibitor, nearly abolished Ca2+ response to SP. In addition, SP-induced Ca2+ signaling and PI hydrolysis rapidly desensitized following short exposure to SP, which did not affect the Ca2+ amount in intracellular Ca2+ stores or Ca2+ responses to carbachol and gastrin releasing peptide-10. These findings suggested that NK1 receptors do not couple to adenylate cyclase, although they induce PI response, and that NK1 receptors induce both intracellular Ca2+ release and Ca2+ influx through PLC activation. Ca2+ signaling and PI hydrolysis through NK1 receptors desensitized rapidly after the stimulation, maybe dependent on the modification of NK1 receptors.

Homologous and heterologous desensitisation of somatostatin-induced increases in intracellular Ca2+ and inositol 1,4,5-trisphosphate in CHO-K1 cells expressing human recombinant somatostatin sst5 receptors

The mechanisms responsible for somatostatin (SRIF)-induced increases in intracellular Ca2+ concentration ([Ca2+]i) and subsequent desensitisation were studied in CHO-K1 cells expressing human sst5 receptors (CHOsst5 cells). To study the nature of the desensitisation, interactions with uridine triphosphate (UTP) were examined. SRIF (pEC50 7.10) and UTP (pEC50) 5.14) caused concentration-dependent increases in [Ca2+]i but the SRIF maximum was about 40% of that to UTP. SRIF-, but not UTP-, induced increases in [Ca2+]i were transient and abolished by pertussis toxin. SRIF and UTP caused sustained increases in Ins(1,4,5)P3 but the SRIF maximum was about 30% of that to UTP. Removal of [Ca2+]e attenuated the SRIF-induced peak rise in [Ca2+]i but had no effect on the peak increases in Ins(1,4,5)P3. UTP-induced increases in [Ca2+]i and Ins(1,4,5)P3 were attenuated in the absence of [Ca2+]e. Following pre-exposure to SRIF (1 microM) or UTP (100 microM) for 5 min, subsequent SRIF responses were desensitised. Similar results were obtained in the absence of [Ca2+]e. Pre-exposure to SRIF had no effect on subsequent responses to UTP but in the absence of [Ca2+]e, responses to UTP were attenuated. The results suggest that SRIF but not UTP-induced increases in [Ca2+]i in CHOsst5 cells are mediated by pertussis toxin sensitive G proteins and are caused by an entry of extracellular Ca2+ and release from an Ins(1,4,5)P3 sensitive Ca2+ store. Homologous or heterologous desensitisation of agonist-induced increases in [Ca2+]i could be demonstrated in the presence or absence of extracellular Ca2+ respectively, and the latter appeared to involve depletion of a common intracellular Ca2+ store.

Mobilization of Ca2+ from intracellular stores in transfected neuro2a cells by activation of multiple opioid receptor subtypes

In neuronal cell lines, activation of opioid receptors has been shown to mobilize intracellular Ca2+ stores. In this report, we describe the excitatory actions of opioid agonists on murine neuroblastoma neuro2a cells stably expressing either delta, mu, or kappa opioid receptors. Fura-2-based digital imaging was used to record opioid-induced increases in intracellular Ca2+ concentration ([Ca2+]i). Repeated challenges of delta, mu, or kappa opioid receptor expressing cells with 100 nM [D-Ala2,D-Leu5]-enkephalin (DADLE), [D-Ala2,N-Me-Phe4,Gly-ol]-enkephalin (DAMGO), or trans-(+/-)-3,4-dichloro N-methyl-N-(2-[1-pyrollidinyl] cyclohexyl) benzene acetamide (U-50488H), respectively, elicited reproducible Ca2+ responses. Non-transfected neuro2a cells did not respond to opioid agonists. Removal of extracellular Ca2+ from the bath prior to and during agonist challenge did not affect significantly the agonist-evoked increase in [Ca2+]i, indicating that the response resulted from the release of Ca2+ from intracellular stores. Naloxone reversibly inhibited responses in all three cell lines, confirming that they were mediated by opioid receptors. Expression of cloned opioid receptors in neuro2a cells, coupled with digital [Ca2+]i imaging, provides a model system for the study of opioid receptors and opioid-activated signaling processes. The fact that all three receptors coupled to the same intracellular signaling mechanism suggests that the primary functional difference between opioid responses in vivo results from their selective localization.

Cholinergic modulation of the Ca2+ response to bradykinin in neuroblastoma cells

Fura 2 imaging was used to measure intracellular Ca2+ signals in N1E-115 mouse neuroblastoma cells during combined activation of bradykinin (BK) and cholinergic receptors. BK and carbachol (CCh) both activate phospholipase C (PLC) and cause Ca2+ release from inositol 1,4,5-trisphosphate (IP3)-sensitive Ca2+ stores. The Ca2+ signal in response to CCh is prolonged by the activation of Ca2+ influx, but BK does not appear to activate the influx pathway. When BK and CCh are applied together (BK+CCh), the Ca2+ response is composed of both Ca2+ release and Ca2+ influx. Ca2+ influx is also activated by BK+CCh in a subset of cells that does not respond with a intracellular Ca2+ concentration increase when CCh is presented by itself. This suggests that CCh stimulates a Ca(2+)-silent cholinergic receptor that is not coupled to Ca2+ release but acts synergistically with BK receptors to activate Ca2+ influx. Pertussis toxin reduces influx without affecting release, indicating that the G protein that modulates the influx pathway is different from the G protein responsible for activating PLC. Cholinergic stimulation also causes progressive heterologous desensitization of BK-evoked Ca2+ release. Desensitization has the unique property of continuing to develop after the cholinergic agonist is removed and the cholinergic Ca2+ response has fully recovered. Heterologous desensitization is not the result of Ca2+ store depletion or a long-lasting inhibition of PLC or IP3-dependent Ca2+ release. Instead, it appears to involve an early step in the BK-signaling cascade, possibly at the level of the B2 receptor or associated G proteins.

Ca2+ release and Ca2+ influx in Chinese hamster ovary cells expressing the cloned mouse B2 bradykinin receptor: tyrosine kinase inhibitor-sensitive and- insensitive processes

A cDNA encoding a mouse B2 bradykinin (BK) receptor was stably transfected in Chinese hamster ovary (CHO) cells. In two resulting transformants, mouse B2 BK receptor was found to induce a twofold elevation in the inositol-1,4,5-trisphosphate level. In a pertussis toxin-insensitive manner, BK also produced a biphasic increase in the intracellular Ca2+ concentration ([Ca2+]i). The initial elevation in [Ca2+]i was abolished by thapsigargin pretreatment in Ca(2+)-free medium. The second phase was dependent on external Ca2+. The BK/inositol trisphosphate and thapsigargin-sensitive Ca2+ stores required extracellular Ca2+ for refilling. Ca2+ influx induced by BK and thapsigargin was confirmed by Mn2+ entry through Ca2+ influx pathways producing Mn2+ quenching. Genistein, a tyrosine kinase inhibitor, partially decreased the BK-induced [Ca2+]i increase during the sustained phase and the rate of Mn2+ entry. BK had essentially no effect on the intracellular cyclic AMP level. The results suggest that the mouse B2 BK receptor couples to phospholipase C in CHO cells and that its activation results in biphasic [Ca2+]i increases, by mobilization of intracellular Ca2+ and store-depletion-mediated Ca2+ influx, the latter of which is tyrosine phosphorylation dependent.

Calcium mobilization from the intracellular mitochondrial and nonmitochondrial stores of the rat cerebellum

Two major intracellular Ca2+ stores, the mitochondrial and nonmitochondrial (microsomes) fractions isolated from rat cerebellum exhibited a Ca2+ concentration and ATP-dependent Ca2+ accumulation. The maximal Ca2+ accumulation in mitochondria was higher than in microsomes, but the affinity of the mitochondria for Ca2+ was lower. In this study, Ca2+ accumulation within the mitochondria was energized by ATP hydrolysis. Thus, the protonophore, carbonyl cyanide p-trifluoromethoxyphenylhydrazone, and the F1F0 ATP synthase inhibitor, oligomycin, blocked Ca2+ accumulation and induced the discharge of the entrapped Ca2+ in the mitochondria, whereas the metabolic inhibitor, rotenone, affected neither the Ca2+ accumulation nor discharge. On the other hand, the uniporter inhibitor, ruthenium red, blocked the mitochondrial accumulation of Ca2+, but did not cause the discharge of preloaded Ca2+. In addition, arachidonic acid (AA), sphingosylphosphorylcholine (SPC) and sphingosine (SPH) elicited the dose-dependent release of Ca2+ from microsomal stores. Although the magnitudes of the Ca2+ release induced by AA, SPC or SPH were all dependent on the presence of extravesicular Ca2+ at concentrations ranging from 0.01 to 0.1 microM Ca2+, only the AA- and SPC-evoked Ca2+ releases were insensitive to temperature. The mitochondria were more sensitive than the microsomes to the AA induced release of accumulated Ca2+. Our results indicate the existence of multiple intracellular Ca2+ stores in nerve cells which can be released by various Ca2+ mediators.