Modulation of Histamine-induced Ca2+ Release by Protein Kinase C

In HeLa cells, histamine induces production of inositol 1,4,5-trisphosphate (InsP3) and release of Ca2+ from the endoplasmic reticulum (ER). Ca2+ release is typically biphasic, with a fast and brief initial phase, followed by a much slower and prolonged one. In the presence of inhibitors of protein kinase C (PKC), including staurosporine and the specific inhibitors GF109203X and Ro-31-8220, the fast phase continued until the ER became fully empty. On the contrary, treatment with phorbol 12,13-dibutyrate inhibited Ca2+ release. Staurosporine had no effect on InsP3-induced Ca2+ release in permeabilized cells and did not modify either histamine-induced InsP3 production. These data suggest that histamine induces Ca2+ release and with a short lag activates PKC to down-regulate it. Consistently, Ca2+ oscillations induced by histamine were increased in amplitude and decreased in frequency in the presence of PKC inhibitors. We show also that mitochondrial [Ca2+] was much more sensitive to changes in ER-Ca2+ release induced by PKC modulation than cytosolic [Ca2+]. PKC inhibitors increased the histamine-induced mitochondrial [Ca2+] peak by 4-fold but increased the cytosolic [Ca2+] peak only by 20%. On the contrary, PKC activation inhibited the mitochondrial [Ca2+] peak by 90% and the cytosolic one by only 50%. Similarly, the combination of PKC inhibitors with the mitochondrial Ca2+ uniporter activator SB202190 led to dramatic increases in mitochondrial [Ca2+] peaks, with little effect on cytosolic ones. This suggests that activation of ER-Ca2+ release by PKC inhibitors could be involved in apoptosis induced by staurosporine. In addition, these mechanisms allow flexible and independent regulation of cytosolic and mitochondrial [Ca2+] during cell stimulation.

Inositol 1,4,5-trisphosphate receptors (InsP 3 R) 1 are a family of Ca 2ϩ channels of the endoplasmic reticulum (ER) that is widely distributed in different tissues. These channels become activated when agonists acting on specific plasma membrane receptors activate phospholipase C, which cleaves phosphatidylinositol 4,5-bisphosphate to generate inositol 1,4,5-trisphosphate (InsP 3 ) and diacylglicerol. InsP 3 rapidly diffuses inside the cell and triggers opening of InsP 3 R, which releases the Ca 2ϩ stored in the ER into the cytosol, generating a transient increase of the cytosolic [Ca 2ϩ ] ([Ca 2ϩ ] c ). Simultaneously, part of the released Ca 2ϩ is taken up by some mitochondria placed close to the ER, thus generating also a transient peak of mitochondrial [Ca 2ϩ ] ([Ca 2ϩ ] M ). The dynamics of both [Ca 2ϩ ] c and [Ca 2ϩ ] M peaks play an important role to determine the response of the cell to each particular stimulus. The increase in [Ca 2ϩ ] c controls the activation of many cytosolic Ca 2ϩ -dependent enzymes and plasma membrane channels and triggers phenomena such as neurosecretion or contraction (1). On the other hand, Ca 2ϩ uptake by mitochondria serves as a fast and transient Ca 2ϩ buffer able to modulate the development of [Ca 2ϩ ] c -dependent phenomena (2-4) but may also perform direct functions inside mitochondria, such as activating mitochondrial energy production (5) or opening the permeability transition pore as an initial step to apoptosis (6,7).
The [Ca 2ϩ ] c and [Ca 2ϩ ] M responses induced by an agonist depend closely on the kinetics of Ca 2ϩ release from the ER. These kinetics are conditioned first by the amount of InsP 3 produced, but many other signaling pathways contribute to modulate the response. Among them, phosphorylation of InsP 3 R by a series of kinases has been reported. Most of the available data refer to modulation by protein kinase A (PKA) (8 -16), but other protein kinases have been also involved: protein kinase C (PKC) (17)(18)(19)(20), Ca 2ϩ /calmodulin-dependent protein kinase II (18), cyclic GMP-dependent protein kinase (21), and nonreceptor protein tyrosine kinases (22).
We have studied here the modulation of histamine-induced Ca 2ϩ release from the ER in HeLa cells. We have shown before that the kinetics of Ca 2ϩ release in this case is biphasic, with a fast initial phase followed in a few seconds by a much slower one (23). We show here evidence that PKC is responsible for the shift from the fast to the slow phases of Ca 2ϩ release during histamine action. In addition, by studying in parallel the dynamics of [Ca 2ϩ ] in cytosol, mitochondria, and ER under different conditions, we show that [Ca 2ϩ ] M is much more sensitive to changes in Ca 2ϩ release than [Ca 2ϩ ] c . Together with the specific modulation of the mitochondrial Ca 2ϩ uniporter that we have reported recently (24), regulation by PKC of Ca 2ϩ release may allow cells to modulate [Ca 2ϩ ] c and [Ca 2ϩ ] M in a nearly independent way.

EXPERIMENTAL PROCEDURES
Cell Culture and Transfection-The construction strategy of the mutated mitochondrially targeted aequorin chimera has been described * This work was supported by Dirección General de Enseñ anza Superior Grant BFI2002-01397 and Junta de Castilla y León Grant VA 005/02. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18  previously (4). HeLa cells were grown in Dulbecco's modified Eagle's medium supplemented with 5% fetal calf serum. The HeLa cell clone MM5 that stably expresses mitochondrially targeted mutated aequorin has been described previously (24). Similar data were obtained using wild type cells transiently transfected with the mitmutAEQ/pcDNA3.1 plasmid. Constructs for cytosolic aequorin and ER-targeted mutated aequorin have been also described previously (24,25). They were cloned into the pcDNA3.1. plasmid and used to transfect wild type HeLa cells. Transfections were carried out using Metafectene (Biontex, Munich, Germany).
[ ] ER measurements were carried out using HeLa cells transiently transfected with the corresponding plasmid. The cells were plated onto 13-mm round coverslips. For aequorin reconstitution, HeLa cells expressing cytosolic aequorin were incubated for 1-2 h at room temperature with 1 M of wild type coelenterazine, and cells expressing mitochondrially targeted mutated aequorin were incubated for 1-2 h at room temperature with 1 M of wild type coelenterazine, in standard medium (145 mM NaCl, 5 mM KCl, 1 mM MgCl 2 , 1 mM CaCl 2 , 10 mM glucose, and 10 mM HEPES, pH 7.4). In some experiments (see Fig. 9), mitochondrially targeted mutated aequorin was reconstituted with coelenterazine n, to reduce its Ca 2ϩ affinity and allow measurement of higher [Ca 2ϩ ] M . The cells were then placed in the perfusion chamber of a purpose-built luminometer thermostatized at 37°C. In the case of aequorin targeted to the ER, Ca 2ϩ depletion of the ER was required prior to reconstitution (25). For this purpose, the cells were incubated for 10 min in medium containing 145 mM NaCl, 5 mM KCl, 1 mM MgCl 2 , 0,5 mM EGTA, 10 M 2,5-di-tert-buthyl-hydroquinone (an inhibitor of ER Ca 2ϩ -ATPase), 10 mM glucose, and 10 mM HEPES, pH 7.4. Then after washing once, the cells were placed in the same medium in the presence of 1 M coelenterazine n for 1-2 h. Before starting the experiment, the cells were placed in the perfusion chamber of the luminometer, thermostatized at 37°C, and perfused with the same medium without 2,5-di-tert-buthyl-hydroquinone for at least 5 min. Then standard medium containing 1 mM CaCl 2 was perfused to refill the ER with Ca 2ϩ , prior to the addition of the agonists. For experiments with permeabilized cells, the cells expressing ER-targeted mutated aequorin reconstituted with coelenterazine n were placed in the luminometer as above. Then standard medium containing 0.5 mM EGTA was perfused for 5 min, followed by a 1-min perfusion of intracellular medium (130 mM KCl, 10 mM NaCl, 1 mM MgCl 2 , 1 mM K 3 PO 4 , 0,5 mM EGTA, 1 mM ATP, 20 M ADP, 2 mM succinate, 20 mM HEPES, pH 7) containing 100 M digitonin. Then intracellular medium without digitonin was perfused for 2 min, followed by intracellular medium containing a fixed 100 nM [Ca 2ϩ ] prepared using an EGTA/Ca 2ϩ mixture. Then, once the [Ca 2ϩ ] ER level had reached a steady state, solutions containing different concentrations of InsP 3 prepared in 100 nM [Ca 2ϩ ] intracellular medium were perfused. Calibration of the luminescence data into [Ca 2ϩ ] was made using an algorithm as previously described (24,25). The statistical data are given as the means Ϯ S.E.
Single-cell [Ca 2ϩ ] c Measurements-The cells were plated onto 13-mm round coverslips. HeLa cells were loaded with fura-2 by incubation with 1 M fura-2 acetoxymethyl ester for 30 min at 30°C. Then the coverslips were mounted in a perfusion chamber placed on the stage of a Nikon Diaphot 300 inverted microscope, and the cells were superfused with standard medium (see above). The temperature was kept at 37°C. Dual wavelength measurements of fura-2 fluorescence were performed using the two-way wavelength illumination system DX-1000 (Solamere Technology Group). A 100 W Hg lamp was used as light source (Optiquip). Light was focused and collected through a Nikon Fluor 40/1.30 objective. The wavelength for dye excitation was alternated between 340 and 380 nm, and fluorescence emission at 540 nm was collected with a SensiCam digital Camera (PCO CCD imaging). A binning 4 ϫ 4 was applied to get ratio images of 320 ϫ 256 pixels (12 bits/pixel) at 0.5 Hz. The illumination system and the camera were driven by Axon Imaging Workbench 4.0 (Axon Instruments) running in a Pentium computer. The ratio images were computed off-line from the background subtracted f340 and f380 images.
Measurement of Inositol 1,4,5-Trisphosphate-HeLa cells were cultured to near confluence in a 75-ml culture flask, dissociated by treatment with a trypsin (0.05%) and EDTA (0.5 mM) solution (Invitrogen; 25300), pelleted by centrifugation (220 ϫ g for 5 min), and resuspended in medium containing 30 mM LiCl (115 mM NaCl, 30 mM LiCl, 5 mM KCl, 1 mM MgCl 2 , 1 mM CaCl 2 , 10 mM glucose, 10 mM HEPES, pH 7.4) at 1 ϫ 10 6 cells/ml. Then cells were incubated at 37°C for 15 min, with constant shaking, before the addition of histamine in the presence or in the absence of staurosporine (final volume, 250 l). The stimulation was terminated at different times by the addition of 50 l of 100% trichloroacetic acid. The samples were then centrifuged for 1 min at 13,000 ϫ g, and 200 l of the supernatants were extracted with 400 l of a mixture 3:1 of 1,1,2-trichloro-1,2,2-trifluoroethane and trioctylamine. The aqueous layer was then removed and assayed for inositol 1,4,5-trisphosphate using a commercial radioreceptor assay (kit NEK064 from PerkinElmer Life Sciences).

RESULTS
In HeLa cells, histamine triggers production of InsP 3 and release of Ca 2ϩ from the ER to the cytosol. By studying the behavior of [Ca 2ϩ ] ER during histamine stimulation, we have shown before that the release of Ca 2ϩ from the ER follows typical biphasic kinetics (23). Immediately after perfusion of the agonist, there is a drop of [Ca 2ϩ ] ER of about 100 M that occurs during the first 5-7 s (6.3 Ϯ 0.4s, n ϭ 8) of stimulation (Fig. 1A). Then this fast phase suddenly stops, and Ca 2ϩ release continues at a much slower rate for as long as histamine is present. Fig. 1A shows that stimulation or inhibition of protein kinase C completely modified this kinetics. Inhibition of protein kinase C with staurosporine transformed the biphasic kinetics into a monophasic one, where the transition between fast and slow phases disappeared, and a fast and complete emptying of the ER was observed after histamine stimulation. On the contrary, stimulation of protein kinase C with phorbol 12,13-dibutyrate produced a large inhibition of Ca 2ϩ release. These findings suggest that activation of protein kinase C by histamine stimulation is able to modulate Ca 2ϩ release through InsP 3 receptors. On the other hand, the effect of staurosporine resembles that of loading the cells with a Ca 2ϩ chelator such as BAPTA. We have shown previously that loading the cells with BAPTA dramatically increased the rate of Ca 2ϩ release induced by histamine, an effect that was attributed to the abolition of local high Ca 2ϩ microdomains responsible for autoinhibition of the InsP 3 receptor channels (23). In Fig. 1B we compare the rate of Ca 2ϩ release in control cells, in the presence of staurosporine, and in cells loaded with BAPTA either treated or not with the inhibitor. We can observe that Ca 2ϩ release in cells loaded with BAPTA was much faster than that obtained in the presence of staurosporine. In addition, staurosporine did not increase the rate of Ca 2ϩ release in BAPTA-loaded cells, although it made emptying of the ER more prolonged in the continuous presence of histamine. Therefore, the stimulating effects on Ca 2ϩ release of both BAPTA and inhibition of PKC may at least in part rely on different mechanisms.
The effects of staurosporine were not due to a direct modulation of InsP 3 receptors by this inhibitor. Fig. 2 shows experiments performed in permeabilized cells to study Ca 2ϩ release induced directly by InsP 3 . It can be observed that Ca 2ϩ release induced by either submaximal or maximal InsP 3 concentrations was not modified in the presence of staurosporine. The effects of staurosporine cannot be attributed either to an increase of InsP 3 production induced by histamine. Fig. 3 shows that treatment with staurosporine did not modify histamineinduced InsP 3 production. Therefore, we can reasonably conclude that the effects of staurosporine are due to inhibition of a protein kinase activated by histamine that modulates Ca 2ϩ release through InsP 3 R. However, staurosporine is not an specific inhibitor of protein kinase C and may inhibit a broad spectrum of kinases, particularly protein kinases A, C, and G and myosin light chain kinase. We have then used some specific inhibitors of protein kinase C to test the involvement of this particular kinase. Fig. 4 shows that both GF109203X and Ro-31-8220, two potent and selective inhibitors of protein kinase C, produced the same effects as staurosporine. On the contrary, H89, a potent inhibitor of protein kinase A, produced little or no effect on histamine-induced Ca 2ϩ release (data not shown, but see Fig. 8).
The data shown above suggest that histamine stimulation of HeLa cells triggers first activation of phospholipase C with the subsequent production of InsP 3 and diacylglicerol. Although InsP 3 activates Ca 2ϩ release, diacylglicerol activates protein kinase C that within few seconds phosphorylates the InsP 3 receptor or a modulatory protein to down-regulate Ca 2ϩ re-lease. This mechanism could be involved in the modulation of Ca 2ϩ oscillations induced by histamine, because it has been reported that PKC activation also oscillates during HeLa cell stimulation with histamine, and the oscillations occur in phase with those of Ca 2ϩ , albeit for a delay of few seconds (26). From our results we could predict that inhibition of PKC should increase Ca 2ϩ release during each oscillation, leading to an increase in the amplitude of the oscillations. This should probably imply also a decrease in frequency, because the ER would require more time to refill and start a new oscillation. Both predictions were found to be correct. Fig. 5 shows several single-cell records of HeLa cells undergoing Ca 2ϩ oscillations in the presence of histamine. The addition of GF109203X modified the oscillation pattern as predicted, and the effect could be reversed and seen again after a second addition of the inhibitor.
We have now studied in the rest of this work how the modulation of Ca 2ϩ release by PKC affects the cytosolic [Ca 2ϩ ] peak and the transference of Ca 2ϩ from endoplasmic reticulum to mitochondria. Fig. 6   peaks to changes in agonist-induced Ca 2ϩ release to study the dose-response curve of the effect of the PKC inhibitors GF109203X and Ro-31-8220. Fig. 9 shows that GF109203X was the most effective compound, having a EC 50 for the increase in the histamine-induced [Ca 2ϩ ] M peak of ϳ100 nM. Ro-31-8220 was 5-fold less effective, with an EC 50 of ϳ0.5 M. We include also data for H89, a potent inhibitor of protein kinase A, which had little effect on the [Ca 2ϩ ] M increase even at concentrations as high as 5 M. The high potency of the PKC inhibitors, particularly GF109203X, suggests that the effect is specific for PKC.
We have finally studied what would be the effect on the histamine-induced [Ca 2ϩ ] M peak of increasing Ca 2ϩ release with GF109203X and simultaneously increasing the activity of the mitochondrial Ca 2ϩ uniporter with SB202190. We have shown previously that this p38 mitogen-activated protein kinase inhibitor potently activates the mitochondrial Ca 2ϩ uniporter without modifying ER-Ca 2ϩ release (24). On the contrary, GF109203X activates ER-Ca 2ϩ release and does not modify Ca 2ϩ uptake by mitochondria in permeabilized cells (data not shown). We would therefore expect to obtain strongly additive effects of both compounds on the [Ca 2ϩ ] M peaks, as was the case. In these experiments, mitochondrially targeted mutated aequorin was reconstituted with coelenterazine n, to allow measuring higher [Ca 2ϩ ] M (23,24). GF109203X induced a prolonged [Ca 2ϩ ] M increase at a rate similar to that observed with the control. This is consistent with a prolonged fast phase of Ca 2ϩ release. Instead, SB202190 increased the rate, but not the duration, of the [Ca 2ϩ ] M increase (the peak was reached at the same time that in the control; Fig. 10, left dotted line), because its effect consists of an increase of the activity of the mitochondrial uniporter. Finally, both compounds together produced a prolonged [Ca 2ϩ ] M increase (induced by GF109203X) at a higher rate (induced by SB202190). We can see that the peak in this case was reached at the same time as when only GF109203X was present (Fig.  10, right dotted line), but the rate of increase was faster. The large increase in the histamine-induced [Ca 2ϩ ] M peak obtained in the presence of both compounds strongly contrasted with the very small changes induced by these compounds in the [Ca 2ϩ ] c peak. The inset of Fig. 10 shows that the [Ca 2ϩ ] c peak had a similar height in the presence of both SB202190 and GF109203X. In several similar experiments carried out in parallel, the [Ca 2ϩ ] c peak increased from 0.84 Ϯ 0.02 M (n ϭ 7) in the controls to 0.93 Ϯ 0.03 M (n ϭ 13) in the presence of both inhibitors. The smaller increase in the [Ca 2ϩ ] c peak induced by both compounds together compared with that induced by GF109203X alone (see above) is consistent with the increase in mitochondrial Ca 2ϩ uptake induced by SB202190. Regarding the kinetics, the [Ca 2ϩ ] c peak was wider in the presence of the inhibitors, although it returned faster to resting levels. This is also consistent with a faster and more complete Ca 2ϩ release from the ER induced by GF109203X. DISCUSSION We show in this paper two main points. The first one is the regulation of histamine-induced Ca 2ϩ release by phosphorylation mediated by PKC, a mechanism that allows fine tuning of Ca 2ϩ oscillations. The second one refers to the special relationship in terms of Ca 2ϩ signaling between ER and mitochondria, which is clearly revealed when Ca 2ϩ release is modulated by agents acting on PKC.
Regarding the first point, it has been previously described that InsP 3 receptors are regulated by phosphorylation mediated by multiple kinases, but most of the available data correspond to modulation by protein kinase A. In fact, it has been proposed that protein kinase A and two phosphatases may be components of the InsP 3 receptor macromolecular signaling complex (14). However, the physiological consequences of this modulation by PKA may depend on the InsP 3 R type or the experimental conditions. For example, PKA-mediated phosphorylation of type I InsP 3 R from cerebellum decreased (8,9) or increased (12) Ca 2ϩ release according to different authors. Phosphorylation of type III InsP 3 R in pancreatic acinar cells reduced Ca 2ϩ release (13,15), and phosphorylation of type II InsP 3 R in liver (10,11) and parotid acinar cells (16) enhanced Ca 2ϩ release. Regarding protein kinase C, it has been shown that InsP 3 R contain consensus site(s) for PKC phosphorylation (20) and serve as a substrate for this kinase in vitro (18). However, as far as we know, only two conflicting reports have been published on the effects of PKC on InsP 3 -induced Ca 2ϩ release, showing that phorbol esters reduced Ca 2ϩ release from intracellular stores in permeabilized pancreatic acinar cells (17) but enhanced Ca 2ϩ release from isolated liver nuclei (19). Therefore, little and contradictory evidence exists at present on the possible role of PKC in the modulation of InsP 3 R. We show here that specific inhibitors of PKC potently activate histamine-induced Ca 2ϩ release from the ER. Consistently, phorbol esters inhibit histamine-induced Ca 2ϩ release. In addition, the potent PKA inhibitor H89 produced little effect on Ca 2ϩ release, and we have shown previously that inhibition of other kinases such as Ca 2ϩ -calmodulin kinase II (with KN-62) or InsP 3 kinase (with wortmannin) has no effect on the histamineinduced [Ca 2ϩ ] M peak (24).
Our data suggest that the modulation we find takes place during physiological stimulation of HeLa cells with histamine. This agonist induces production of InsP 3 and Ca 2ϩ release from the ER but also triggers activation of PKC (26,28,29) and phosphorylation of InsP 3 R (30). InsP 3 diffuses fast to the ER and initiates Ca 2ϩ release within milliseconds. Instead, it has been shown recently using a fluorescent reporter that phosphorylation by PKC induced by histamine in HeLa cells follows the cytosolic Ca 2ϩ transients with a lag of ϳ10 s (26). This lag closely matches the time required to terminate the fast phase of histamine-induced Ca 2ϩ release, suggesting that phosphorylation by PKC may be responsible for the shift to the slow phase of Ca 2ϩ release, thus generating the biphasic Ca 2ϩ release response induced by histamine. The effects of PKC inhibitors and BAPTA loading shown in Fig. 1B allow us to speculate on the mechanism of the modulation produced by PKC. Loading with BAPTA should avoid Ca 2ϩ -dependent inactivation of InsP 3 R, which occurs very rapidly, with time constants below 1s (31,32). Consistently, the rate of Ca 2ϩ release is much faster from the beginning in BAPTA-loaded cells. PKC inhibitors instead block an inhibition of Ca 2ϩ release that occurs about 6 s afterward. This kind of delayed inhibition resembles the Ca 2ϩindependent inactivation that has been observed in the continuous presence of InsP 3 (31,33). This idea is reinforced by the fact that staurosporine prolongs the activation of InsP 3 R and the emptying of Ca 2ϩ of the ER even in BAPTA-loaded cells (Fig. 1B). This suggests that Ca 2ϩ -independent inactivation of InsP 3 R in BAPTA-loaded cells allows a slow recovery of [Ca 2ϩ ] ER during prolonged stimulation. In the presence of both BAPTA and staurosporine, neither Ca 2ϩ -dependent nor Ca 2ϩ - independent inactivation would be operative, and the emptying is fast and persistent.
The physiological sense of the modulation of Ca 2ϩ release by PKC observed may be multiple. On the one hand, we have shown that PKC activity was able to control the kinetics of Ca 2ϩ oscillations induced by histamine, although it was not necessary to maintain oscillations. Therefore, PKC activity is not responsible for the generation of the oscillations but allows fine tuning of their amplitude and frequency. This modulation was not detected in a previous work (30), perhaps because the concentration of staurosporine used was much lower (130 nM). In our hands, concentrations of staurosporine above 400 nM were required to observe the increase in Ca 2ϩ release (data not shown). However, Zhu et al. (30) showed a complete inhibition of Ca 2ϩ oscillations in the presence of phorbol esters, which is consistent with the large inhibition of Ca 2ϩ release by phorbol esters that we observe in this paper. It is right that phorbol esters act in many cell systems at the level of the plasma membrane receptor to induce desensitization of the response (32,33), but the H 1 receptor of HeLa cells appears to be coupled to phospholipase C and continuous InsP 3 production without detectable desensitization (34), despite the activation of PKC caused by histamine action (26,28,29).
The inhibition of Ca 2ϩ release by PKC that we describe here may be also useful to avoid full ER-Ca 2ϩ emptying after agonist stimulation, which could have deleterious effects for the cell, first because of the waste of energy that would result from having a full ER-Ca 2ϩ emptying after each agonist stimulation and second because of the own effects of ER-Ca 2ϩ depletion in terms of triggering the activation of stress signaling pathways (7). In this respect, we should note that staurosporine is a known apoptotic-inducing drug in many cell lines, including HeLa cells, although the mechanism of this effect is unclear. The activation of agonist-induced Ca 2ϩ release by staurosporine that we describe in this paper occurs very fast, in seconds or minutes, and could be one of the initial steps in the pathway leading to apoptosis.
We provide also in this paper more evidence on the existence of a privileged relationship in terms of Ca 2ϩ signaling between ER and mitochondria. The main Ca 2ϩ influx pathway of mitochondria is the Ca 2ϩ uniporter, which requires [Ca 2ϩ ] in the high micromolar range in its cytosolic side to open significantly (3,4). Thus, for many years it was thought that mitochondria could hardly take much Ca 2ϩ during InsP 3 -induced Ca 2ϩ release transients, which usually produce global [Ca 2ϩ ] c peaks not higher than 1 M. However, new techniques developed to specifically measure [Ca 2ϩ ] M (35) showed that mitochondria were actually able to take vast amounts of Ca 2ϩ during ER-Ca 2ϩ release events through both InsP 3 and ryanodine receptors (3,4,36). Evidence accumulated in the last few years suggests that the explanation of this finding relies probably in two ideas (3,4,37,38). First, the fact that opening of a particular Ca 2ϩ channel in the ER creates a local microdomain of [Ca 2ϩ ] where the concentration may be much higher than in the mean of the cytosol. Second, the possibility that ER-Ca 2ϩ release channels and mitochondrial Ca 2ϩ uniporters may be somehow colocalized. If these two conditions apply, there would be a sort of discrete communication site among ER and mitochondria that would allow a significant proportion of the Ca 2ϩ released by the ER to be taken up by mitochondria. Accordingly, Ca 2ϩ uptake by mitochondria should take place only while the [Ca 2ϩ ] microdomain exists, that is, while the ER-Ca 2ϩ channel is open. Evidence for this requirement of a sustained Ca 2ϩ release from the ER to maintain Ca 2ϩ uptake by mitochondria has been provided recently (39). We show in this paper additional evidence for this idea by showing a drastic dissociation between the cytosolic and mitochondrial [Ca 2ϩ ] responses after histamine stimulation in two conditions: one in which it produces fast but brief Ca 2ϩ release, and the other in which it produces fast and sustained Ca 2ϩ release. The cytosolic [Ca 2ϩ ] peaks differed in both conditions by only 20%, but the mitochondrial [Ca 2ϩ ] peak became 4-fold larger in the case of sustained Ca 2ϩ release. On the contrary, when Ca 2ϩ release was inhibited with phorbol ester (resulting in a sustained but much slower Ca 2ϩ release), mitochondrial Ca 2ϩ uptake was nearly abolished, but the cytosolic [Ca 2ϩ ] peak was only reduced to half of the control value. In the case of the Ca 2ϩ release induced by carbachol, the rate of ER-Ca 2ϩ release was slower than with histamine but sustained, and this resulted in a slower but sustained Ca 2ϩ uptake into mitochondria. In conclusion, the amount of Ca 2ϩ taken up by mitochondria depends on both the rate and duration of ER-Ca 2ϩ release, and mitochondria may be able to take up most of the Ca 2ϩ released by the ER only if Ca 2ϩ release is fast and sustained.
Finally, we would like to comment on the perfect additivity observed among the stimulation of histamine-induced [Ca 2ϩ ] M increase by PKC inhibitors and by the activator of the mitochondrial Ca 2ϩ uniporter SB202190. This compound, which is considered a specific inhibitor of p38 mitogen-activated protein kinase, strongly activates the mitochondrial Ca 2ϩ uniporter at low [Ca 2ϩ ] c (24). Such additivity contrasts with the scarce effect of these compounds on the cytosolic [Ca 2ϩ ] peak, suggesting that the cell has a variety of modulatory mechanisms able to produce large changes in Ca 2ϩ uptake by mitochondria with only small modifications of the [Ca 2ϩ ] c peak. These mechanisms may therefore allow performance of a nearly independent modulation of [Ca 2ϩ ] c and [Ca 2ϩ ] M by varying the activity of the uniporter or the rate and duration of Ca 2ϩ release from the ER. On the other hand, the availability of these pharmacological tools opens now the possibility to modulate in a simple way the increase in [Ca 2ϩ ] M induced by agonists, creating a good model to study the consequences of such increase for cell metabolism or initiation of apoptosis.