Store-operated Ca2+ Influx Causes Ca2+ Release from the Intracellular Ca2+ Channels That Is Required for T Cell Activation*

The precise control of many T cell functions relies on cytosolic Ca2+ dynamics that is shaped by the Ca2+ release from the intracellular store and extracellular Ca2+ influx. The Ca2+ influx activated following T cell receptor (TCR)-mediated store depletion is considered to be a major mechanism for sustained elevation in cytosolic Ca2+ concentration ([Ca2+]i) necessary for T cell activation, whereas the role of intracellular Ca2+ release channels is believed to be minor. We found, however, that in Jurkat T cells [Ca2+]i elevation observed upon activation of the store-operated Ca2+ entry (SOCE) by passive store depletion with cyclopiazonic acid, a reversible blocker of sarco-endoplasmic reticulum Ca2+-ATPase, inversely correlated with store refilling. This indicated that intracellular Ca2+ release channels were activated in parallel with SOCE and contributed to global [Ca2+]i elevation. Pretreating cells with (-)-xestospongin C (10 μm) or ryanodine (400 μm), the antagonists of inositol 1,4,5-trisphosphate receptor (IP3R) or ryanodine receptor (RyR), respectively, facilitated store refilling and significantly reduced [Ca2+]i elevation evoked by the passive store depletion or TCR ligation. Although the Ca2+ release from the IP3R can be activated by TCR stimulation, the Ca2+ release from the RyR was not inducible via TCR engagement and was exclusively activated by the SOCE. We also established that inhibition of IP3R or RyR down-regulated T cell proliferation and T-cell growth factor interleukin 2 production. These studies revealed a new aspect of [Ca2+]i signaling in T cells, that is SOCE-dependent Ca2+ release via IP3R and/or RyR, and identified the IP3R and RyR as potential targets for manipulation of Ca2+-dependent functions of T lymphocytes.

with a foreign antigen, is required to drive the diverse transcription programs leading to T cell clonal expansion and secretion of effector cytokines necessary for coordination of the immune response (1)(2)(3). Diminished [Ca 2ϩ ] i signaling results in impaired T cell activation and, consequently, development of severe immunodeficiency (4,5).
It is established that TCR stimulation leads to production of numerous second messengers (6), including inositol 1,4,5trisphosphate (IP3), which evokes Ca 2ϩ release from the IP3R, located predominantly on the endoplasmic reticulum (ER) (7). Two IP3R isoforms (IP3R-2 and IP3R-3) are expressed in lymphocytes, thymocytes, and splenocytes, whereas Jurkat T cells express all three known IP3R isoforms (IP3R-1, IP3R-2, IP3R-3) (8,9). Binding of IP3 to the IP3R releases Ca 2ϩ from the IP3sensitive store, followed by activation of plasmalemmal storeoperated Ca 2ϩ (SOC) channels that allow Ca 2ϩ entry (SOCE) across the plasma membrane (7,10,11). Because the volume of the ER is estimated to be only ϳ1% of the volume of the T cell cytoplasm, it is believed that SOCE is the only mechanism capable of maintaining a sustained [Ca 2ϩ ] i elevation for hours and days as required for T cell activation (4,5,7,12). Therefore, the role of Ca 2ϩ release from the IP3R in T cells is believed to be limited to the generation of a signal that activates plasmalemmal SOC channels (7).
It was also shown that TCR activation stimulates production of cyclic adenosine 5Ј-diphosphate-ribose (cADPR) and/or nicotinic acid adenine dinucleotide phosphate, the endogenous agonists of RyR (13,14). Jurkat T lymphocytes express type 3 RyR (13,15), whereas the peripheral blood human T lymphocytes express type 1 and 2 ryanodine (Ry) receptors (RyR) (16). In contrast to the IP3-dependent pathway, the sequence of events leading to the RyR activation and Ca 2ϩ release from the Ry-sensitive store in T cells is less clear. Because down-regulation of IP3R is sufficient to abolish sustained [Ca 2ϩ ] i elevation and cytokine production following TCR stimulation (17), the RyR activation appears to be secondary to IP3R activation. Consistently, the formation of cADPR has been shown to be delayed compared with IP3 production following TCR engagement (13). Nevertheless, down-regulation of cADPR levels or type 3 RyR expression reduced the TCR-dependent [Ca 2ϩ ] i signaling in Jurkat T cells (13,18). In addition, RyR antagonists ruthenium red and dantrolene inhibited murine T cell proliferation and IL-2 production following TCR stimulation (19). Thus, the RyR appears to play a major role in regulating the [Ca 2ϩ ] i dynamics and, therefore, T cell functions. However, the exact mechanisms by which RyR regulates [Ca 2ϩ ] i signaling in T cells, and its functional connections with the IP3R and/or plasmalemmal SOC channels, are not established.
It is known that following store depletion the SOCE supplies Ca 2ϩ for the store replenishing from the cytosol via sarco-endoplasmic reticulum Ca 2ϩ -ATPase (SERCA) (20). Consequently, store replenishing exerts negative feedback inhibition of SOC channels and termination of SOCE (21). However, the store replenishing via SERCA can be accomplished only if the intracellular Ca 2ϩ release channels (IP3R and RyR) are inactivated. Elevation in [Ca 2ϩ ] i and/or conformational changes of plasmalemmal Ca 2ϩ channels may stimulate Ca 2ϩ release from the IP3R and/or RyR either by direct interactions or via Ca 2ϩinduced Ca 2ϩ release (CICR) mechanism (22). Although it is well established that in some cell types, such as skeletal and cardiac muscles, the Ca 2ϩ release from the intracellular store plays a crucial role in generation of [Ca 2ϩ ] i signals and is controlled by activation of plasmalemmal Ca 2ϩ channels, the effect of SOC channel activation on intracellular Ca 2ϩ release channels in T cells has not been elucidated. Recently, the activation of CICR by Ca 2ϩ influx via SOC channels was demonstrated in parotid acinar cells (23), although the specific type of intracellular channels that support the CICR and the functional significance of the CICR in non-excitable cells were not established.
In this study we explored whether Ca 2ϩ release from the IP3R and/or RyR is triggered by activation of plasmalemmal SOC channels and whether it is essential for regulation of Ca 2ϩdependent T cell functions. The study was performed on Jurkat T cells (human leukemia T cell line) that express TCR and produce IL-2 in response to stimulation with mitogenic lectins (24). We found that Ca 2ϩ release from both IP3R and RyR was activated in parallel with SOCE and significantly amplified [Ca 2ϩ ] i signaling in T lymphocytes. Accordingly, inhibition of IP3R or RyR reduced [Ca 2ϩ ] i elevation following TCR crosslinking and down-regulated proliferation and IL-2 production, the major functions of activated T cells.
Before experiments, cells were plated on poly-L-lysinecoated glass-bottom clambers and then loaded with 1 M fura-2/AM and pluronic F-127 for 5 min in modified Tyrode solution containing (in mM): 130 NaCl, 5.6 KCl, 1 MgCl 2 , 2 CaCl 2 , 10 HEPES, 10 D-glucose, pH 7.3. After washing, cells were incubated for an additional 30 min at 37°C. Some cells were incubated in modified Tyrode solution containing 10 M (-)-XeC, or 400 M Ry, or vehicle (methanol or DMSO) alone for 30 min at 37°C. In nominally Ca 2ϩ -free solution, Ca 2ϩ was omitted. In some experiments La 3ϩ was added to modified Tyrode solution and/or nominally Ca 2ϩ -free solution.
Calcium Imaging-Fluorescence images were acquired from adherent cells using a SenSys CCD camera (Roper Scientific, Tucson, AZ) and a ϫ40 oil immersion Zeiss objective on a Zeiss Axiovert 200 inverted microscope (Thornwood, NY). A Lambda DG-4 filter changer (Sutter Instrument, Novato, CA) was used for switching between 340 and 380 nm excitation wavelengths. All experiments were performed at room temperature. Solution exchange was performed via a gravity-driven perfusion system. Solution exchange in the recording chamber was completed within 5 s. The vehicle-containing solutions (methanol or DMSO) were applied in place of the drugs in all control experiments. Data acquisition was performed using MetaFluor v7.0 software (Universal Imaging, Downingtown, PA). [Ca 2ϩ ] i values were estimated from fura-2 calibration as described previously (26). K d was taken as 248 nM (27). Unless otherwise indicated, in each experiment changes in [Ca 2ϩ ] i were recorded from 15-20 cells and then averaged.
Manganese Quench of Fura-2 Fluorescence-Because Mn 2ϩ readily passes through SOC channels (28) and is unlikely to be transported from the cytosol (29,30), the rate of fura-2 quenching in the presence of Mn 2ϩ was used as a measure of divalent cation influx into the bulk cytoplasm in intact cells. The extracellular solution used in all Mn 2ϩ quench experiments contained (in mM): 130 NaCl, 5.6 KCl, 1 MgCl 2 , 0.3 CaCl 2 , 0.5 MnCl 2 , 10 HEPES, 10 D-glucose, pH 7.3. Changes in Ca 2ϩ -independent fluorescence due to Mn 2ϩ quench were determined using the method described by Zhou and Neher (31). Briefly, background-corrected fura-2 fluorescence excited at 340 nm (F 340 ) was plotted against background-corrected fura-2 fluorescence excited at 380 nm (F 380 ) during the initial 10 s following application of 30 M cyclopiazonic acid (CPA). The slope of the F 340 versus F 380 relationship equals the "isocoefficient" ␣. The fluorescence value, F i , which is unaffected by changes in [Ca 2ϩ ] i , was determined as F i ϭ F 340 ϩ ␣F 380 . The rate of F i decline is proportional to the rate of Mn 2ϩ entry into the cytosol and was used as an indicator of plasma membrane Ca 2ϩ permeability.
Proliferation Assay-Cell division track assay was conducted using the CFSE proliferation kit (Molecular Probes). Stock solution of CFSE (4 mM) was prepared in DMSO. Cells were washed, resuspended in phosphate-buffered saline (PBS) containing 4 M CFSE at density 1 ϫ 10 6 cells/ml, and incubated at 37°C for 10 min. Labeling was quenched by adding 5ϫ volume of RPMI 1640 culture medium containing 10% FBS. After washing three times with RPMI 1640 ϩ 10% FBS, cells were placed into FBS-free cell culture medium to abolish cell division and kept overnight in 5% CO 2 at 37°C. After overnight incubation, the CFSE-labeled cells were pelleted down and a fraction of them (ϳ50 ϫ 10 3 ) were fixed with 1% paraformaldehyde in PBS and analyzed by flow cytometry to establish the CFSE fluorescence profile of undivided cells. The remaining CFSE-labeled cells were resuspended in cell culture medium supplemented with 10% FBS at a density 0.05 ϫ 10 6 cells/ml and seeded into 24-well tissue culture plate (1 ml/well). Some cells were preincubated for 30 min at 37°C with (-)-XeC (10 M), or Ry (400 M), or dantrolene (30 M) prior to transfer into FBS-containing cell culture medium. Drugs were also added to the same cells on each consecutive day of culturing. The vehicle-containing solutions (methanol or DMSO) were applied in place of the drugs in all control experiments. After 72 h in culture, cells were harvested, washed with PBS, and fixed with 1% paraformaldehyde in PBS. Fluorescence intensity was measured at 488 nm using FACScan flow cytometer and CellQuest software (BD Biosciences). The number of cell divisions was calculated using FlowJo software (Tree Star Inc., Ashland, OR).
IL-2 Production Assay-Jurkat T cells were washed three times with PBS, resuspended in FBS-free cell culture medium at a density of 0.8 ϫ 10 6 cells/ml, seeded into 96-well tissue culture plates (0.4 ml/well), and then stimulated with 100 M mitogenic lectin phytohemeagglutinin P (PHA). Some cells were preincubated for 30 min at 37°C with (-)-XeC (10 M), or Ry (400 M), or dantrolene (30 M) prior to stimulation with PHA. After 24 h of incubation, cell culture supernatants were collected and contents of IL-2 were determined by enzyme-linked immunosorbent assay using Quantikine kits (R&D Systems, Minneapolis, MN) according to the manufacturer's instructions.
Data Analysis-Image analysis was performed using MetaFluor v7.0 software (Universal Imaging). Further measurements and statistical analysis were performed using Origin 7 software (OriginLab, Northampton, MA). All data are presented as the means Ϯ S.E. of the mean; n ϭ number of experiments. Slopes were determined using a linear fit of experimental data to minimize the sum of the squared differences between the data and the fitted line. Statistical differences between means were accepted as statistically significant at p Ͻ 0.05, using Student's paired or unpaired t test or Wilcoxon nonparametric test.

Store Refilling Inversely Correlates with Levels of Global [Ca 2ϩ
] i -To explore the contribution of Ca 2ϩ release from the internal store to the global [Ca 2ϩ ] i dynamics, we employed fura-2 Ca 2ϩ indicator and CPA, a reversible SERCA blocker, that allowed us to assess the [Ca 2ϩ ] i dynamics during SOCE activation and the store content in the individual T cells in the same experiment. As expected, when CPA (30 M) was applied in Ca 2ϩ -free solution, a transient elevation in [Ca 2ϩ ] i due to Ca 2ϩ release via "leak" channels was observed (Fig. 1A). A sec- ond application of CPA following wash in Ca 2ϩ -free solution did not produce detectable [Ca 2ϩ ] i responses, indicating that the CPAsensitive store was empty. Ionomycin, a Ca 2ϩ ionophore, applied at the end of the experiment produced only a small elevation in global [Ca 2ϩ ] i , demonstrating that CPA depleted most of the intracellular Ca 2ϩ store.
Readdition of extracellular Ca 2ϩ following the first CPA application produced a global elevation in [Ca 2ϩ ] i , presumably due to Ca 2ϩ entry via SOC channels (Fig. 1B). After removing extracellular Ca 2ϩ , the second application of CPA evoked [Ca 2ϩ ] i transients, indicating that the store was allowed to be refilled during extracellular Ca 2ϩ readdition. Although on average the amplitude of the [Ca 2ϩ ] i transient produced by the second CPA application was smaller than that produced by the first CPA application, the individual cells exhibited a variety of [Ca 2ϩ ] i responses to the second application of CPA (Fig. 1, B and C). We found that cells that displayed a large elevation in [Ca 2ϩ ] i upon Ca 2ϩ readdition exhibited a small response to the second CPA application (Fig. 1C). In contrast, in cells that displayed small or no [Ca 2ϩ ] i transient in the presence of extracellular Ca 2ϩ , both CPA applications evoked similar [Ca 2ϩ ] i transients (Fig. 1C). Ionomycin, applied following the second CPA application, produced similar responses in both cases, indicating that Ca 2ϩ retention within the store may not account for the absence of Ca 2ϩ transient upon a second CPA application.
Integration of [Ca 2ϩ ] i transients evoked by SERCA blockers has been shown to provide an accurate estimate of releasable store content (32). We took the ratio of background-subtracted integrated [Ca 2ϩ ] i transients induced by the second CPA application (CPA 2 ) to the first one (CPA 1 ) (Fig. 1C) as the measure of the extent of store refilling. The [Ca 2ϩ ] i transients recorded during Ca 2ϩ readdition that were presum- ably induced by SOCE were also integrated and denoted as [Ca 2ϩ ] SOCE . When CPA 2 /CPA 1 ratios were plotted against [Ca 2ϩ ] SOCE measured in the individual cells, an inverse relationship between the extent of store refilling and global [Ca 2ϩ ] i elevation following Ca 2ϩ readdition was revealed (Fig. 1D).
The store refilling in the absence of the global [Ca 2ϩ ] i elevation was described previously and was attributed to colocalization of the plasmalemmal Ca 2ϩ channels with SERCA (33)(34)(35)(36). The absence of store refilling in cells with high levels of [Ca 2ϩ ] i implied that in these cells intracellular Ca 2ϩ release channels are active and may contribute into [Ca 2ϩ ] i elevation following Ca 2ϩ readdition.

Inhibition of IP3R or RyR Reduces Global [Ca 2ϩ ] i Elevation during SOCE Activation and Facilitates
Store Refilling-To explore the contribution of Ca 2ϩ release from IP3R and/or RyR into the [Ca 2ϩ ] i elevation following SOCE activation in T cells, we employed (-)-XeC and Ry that specifically block IP3R and RyR, respectively (37,38). When applied extracellularly, neither (-)-XeC (10 M) nor Ry (400 M) affected basal [Ca 2ϩ ] i levels or CPA responses (Fig. 2, A and B). Furthermore, pretreatment with either (-)-XeC or Ry did not affect the rate of the storeoperated divalent cation influx assessed by the Mn 2ϩ quench of fura-2 fluorescence (Fig. 2, C and D). When applied extracellularly, Mn 2ϩ produced a slow base-line quench of fura-2 fluorescence that was significantly accelerated following CPA application due to the activation of the store-operated Mn 2ϩ influx via SOC channels. Preincubation with either (-)-XeC or Ry did not significantly affect rates of Mn 2ϩ quench in the absence or presence of CPA. These experiments indicate that neither (-)-XeC nor Ry release Ca 2ϩ from the CPA-sensitive store or directly affect SOCE.
In addition, the sequential store depletion experiments revealed that [Ca 2ϩ ] i transients evoked by the first application of CPA were similar in shape and magnitude in both untreated cells (Control) and cells preincubated with either (-)-XeC or Ry (Fig. 2, E and F). However, the [Ca 2ϩ ] i transients during Ca 2ϩ readdition were smaller in (-)-XeC-or Ry-treated cells compared with the control. Given that neither (-)-XeC nor Ry antagonize SOCE (Fig. 2, C and D), the inhibition of Ca 2ϩ release from IP3R or RyR should account for the reduction in [Ca 2ϩ ] i elevation during Ca 2ϩ readdition in cells preincubated with (-)-XeC or Ry. Consistently, CPA application following Ca 2ϩ readdition produced larger [Ca 2ϩ ] i transients and therefore increased the CPA 2 /CPA 1 ratio in (-)-XeC-or Ry-treated cells compared with untreated cells (Fig. 2, E-H). Thus, these  ] i responses that were significantly reduced in cells preincubated with (-)-XeC (Fig. 3, A and B). On average, (-)-XeC reduced integrated [Ca 2ϩ ] i responses to PHA by ϳ40% compared with those in untreated cells (Fig. 3, A and E). The same experiments revealed that [Ca 2ϩ ] i transients in response to CPA, applied following PHA application to completely deplete the store, were up-regulated by ϳ30% in (-)-XeC-treated cells (Fig. 3, A and F), which is consistent with (-)-XeC inhibition of IP3R. Application of RyR agonist caffeine (20 mM) produced low amplitude [Ca 2ϩ ] i oscillations in the absence of extracellular Ca 2ϩ , which were essentially (by 90%) inhibited in cells preincubated with 400 M Ry (n ϭ 4, not shown). Despite inhibition of [Ca 2ϩ ] i responses to caffeine, preincubation with Ry did not significantly affect the PHA-or CPAinduced [Ca 2ϩ ] i transients (Fig. 3, B, E, and F). These results indicate that TCR cross-linking with PHA readily mobilizes Ca 2ϩ from the IP3-, but not Ry-, sensitive stores.
When Ca 2ϩ was readded following TCR stimulation to allow SOCE, the amplitudes of [Ca 2ϩ ] i transients were significantly lower in both (-)-XeC-or Ry-pretreated cells than those in the untreated cells (Fig. 3, C and D). On average, preincubation with (-)-XeC or Ry reduced the maximal amplitudes of [Ca 2ϩ ] i transients during Ca 2ϩ readdition by ϳ44 and 42%, respectively (Fig. 3G), indicating that at physiological conditions Ca 2ϩ release from IP3R and/or RyR is activated in parallel with SOCE and contributes significantly to [Ca 2ϩ ] i elevation.
Inhibition of IP3R or RyR Down-regulates T Cell Proliferation and IL-2 Production-To establish the importance of Ca 2ϩ release from IP3R and/or RyR for T cell functions, we investigated the effects of IP3R or RyR inhibitors on cell proliferation and IL-2 production. We employed the CFSE fluorescent dye, which can be readily loaded into the cytoplasm and then equally distributed between cells after each cell division, resulting in a progressive decrease in CFSE fluorescence in cells that have undergone multiple rounds of divisions (39). Because Jurkat T cells proliferate instantly without stimulation, we first incubated the CFSE-labeled cells in FBS-free medium to inhibit cell divisions. The proliferation was then induced by transferring cells into complete cell culture medium. Some cells were preincubated with (-)-XeC-, or Ry, or dantrolene, a clinically relevant RyR inhibitor (40,41), prior to the induction of proliferation. The analysis of CFSE fluorescence intensity in individual cells following 72 h of incubation revealed that IP3R or RyR inhibitors significantly (p Ͻ 0.05, Wilcoxon directional test, n ϭ 5) reduced the proliferative activity of Jurkat T cells (Fig.  4A). The fractions of undivided cells constituted 55, 60, and 54% in cell cultures incubated with (-)-XeC, Ry, or dantrolene, respectively, compared with 38% in the untreated cell population. In addition, the peaks of CFSE fluorescence profiles of cells exposed to IP3R or RyR inhibitors were shifted to the right, indicating slower cell cycle progression (Fig. 4A). We also found that preincubation with (-)-XeC-, or Ry, or dantrolene significantly inhibited PHA-induced IL-2 production (Fig. 4B). These data indicate that Ca 2ϩ release from IP3R or RyR is a powerful regulator of Jurkat T cell proliferation and TCR-mediated IL-2 production.

DISCUSSION
The major finding of this study is that in T lymphocytes activation of the plasmalemmal SOC channels directly stimulates Ca 2ϩ release from both IP3R and RyR, which significantly amplifies [Ca 2ϩ ] i signaling at the expense of store refilling (Fig.  5). These findings underscore the role of intracellular Ca 2ϩ store in maintaining the sustained [Ca 2ϩ ] i elevation, which was previously solely attributed to the extracellular Ca 2ϩ influx via SOC channels (4,5,7,12).
We found that consistent with a number of previous studies performed on other cell types (33)(34)(35)(36), intracellular store in T cells may be completely refilled without global [Ca 2ϩ ] i elevation (Fig. 1C). This indicates that in T cells SOC channels may colocalize with the SERCA, which readily pumps Ca 2ϩ from the vicinity of SOC channels into the store lumen, so that Ca 2ϩ do not appear in the cytosol. The areas of close apposition of the ER and the plasma membrane were found in T cells (42)(43)(44) and may provide the structural platforms for direct Ca 2ϩ entry into the store (Fig. 5).
We have shown that (-)-XeC or Ry specifically block Ca 2ϩ release from the IP3R or RyR, demonstrating that signaling through the respective intracellular Ca 2ϩ release channel may be selectively abolished in T cells. Using these blockers we established that in the presence of the extracellular Ca 2ϩ and activated SOC channels, the Ca 2ϩ release from either IP3R or RyR may account for at least 25-45% of global [Ca 2ϩ ] i elevation. Thus, in combination, the Ca 2ϩ release from IP3R and RyR may account for Ͼ70% of the sustained [Ca 2ϩ ] i elevation following TCR stimulation. These numbers could be substantially underestimated as neither (-)-XeC nor Ry completely blocked the [Ca 2ϩ ] i responses to IP3R or RyR agonists. These data are consistent with a previous report demonstrating that in parotid acinar cells maximal Ca 2ϩ influx via SOC channels accounts for only a small fraction of the agonist-induced [Ca 2ϩ ] i elevation in the presence of the extracellular Ca 2ϩ compared with the Ca 2ϩ release from the store (23). Thus, similarly to other cell types, plasmalemmal Ca 2ϩ channels in T cells may serve for direct store refilling and for activation of intracellular Ca 2ϩ channels but by themselves contribute little to [Ca 2ϩ ] i elevation necessary for T cell activation.
Interestingly, activation of SOC channels in T cells was sufficient to trigger Ca 2ϩ release from both IP3R and RyR without stimulation of the surface receptors (Fig. 2, E-H), and, therefore, in the absence of the endogenous agonists. Moreover, in our study activation of TCR with PHA did not activate Ca 2ϩ release from the RyR (Fig. 3B), although up-regulation of the endogenous RyR agonist cADPR has been previously observed following TCR stimulation (13). This discrepancy may be due to the fact that formation of cADPR depends on the type of stimulus. Apparently, in our study PHA failed to produce robust elevation in cADPR and/or nicotinic acid adenine dinucleotide phosphate levels and consequent Ca 2ϩ release from the RyR in the absence of extracellular Ca 2ϩ influx. Nevertheless, SOC channel activation in the presence of extracellular Ca 2ϩ was sufficient to trigger the Ca 2ϩ release from the RyR, indicating that this is a primary mechanism for RyR activation in T cells. The production of RyR endogenous agonists cADPR and/or nicotinic acid adenine dinucleotide phosphate at stronger TCR stimulation may facilitate the opening of RyR to further sustain Ca 2ϩ release.
Thus, taken together, our data demonstrated that activation of SOC channels directly stimulates Ca 2ϩ release from both IP3R and RyR, which is a major determinant for [Ca 2ϩ ] i elevation following TCR stimulation. We further speculate that the long-lasting store depletion via IP3R and/or RyR may be necessary to prevent inactivation of plasmalemmal SOC channels (Fig. 5). Our findings are essential because they emphasize the role of intracellular Ca 2ϩ release channels in shaping [Ca 2ϩ ] i FIGURE 5. Hypothetical scheme of Ca 2؉ dynamics in T cells upon TCR activation. Engagement of TCR with the antigen (A) triggers Ca 2ϩ release from IP3R and possibly from RyR, leading to a transient or localized store depletion, which in turn activates SOC channels in the plasma membrane (PM). Activation of SOCE via SOC channels may elevate [Ca 2ϩ ] i in the vicinity of IP3R and/or RyR, which activates CICR from both IP3R and RyR, further elevating [Ca 2ϩ ] i levels. Alternatively, SOC channels may directly interact with IP3R and/or RyR to trigger Ca 2ϩ release (not shown). Because store is continuously refilled with extracellular Ca 2ϩ at the junctional sites of the ER with the plasma membrane, the Ca 2ϩ release from IP3R and/or RyR is active as long as SOC channels are activated and extracellular Ca 2ϩ is present. The Ca 2ϩ release from IP3R and/or RyR depletes the store, which keeps SOC channels open. The Ca 2ϩ release from IP3R and/or RyR and Ca 2ϩ influx via SOC channels act synergistically to sustain [Ca 2ϩ ] i elevation necessary for activation of transcription factors. The location of IP3R, RyR, SOC channel, and SERCA on the diagram is for convenience of the presentation only. signaling in T cells and identify IP3R and RyR as novel potential targets for immunosuppression. This is especially important in view of the absence of specific SOC channel blockers and availability of a number of the IP3R and RyR blockers, one of which (dantrolene) is clinically approved (40,41).
Our study demonstrated that pharmacological inhibition of IP3R or RyR exerts profound effects on major functions of human T cells such as proliferation and IL-2 production (Fig.  4). These findings are compatible with previous reports that down-regulation of IP3R expression in Jurkat T cells (17) and RyR inhibition with Ry or dantrolene in murine T cells (19) suppressed TCR-mediated proliferation and IL-2 production. Thus, pharmacological blockers of IP3R or RyR may be used as powerful tools for immunosuppression.
The important question that emerged from this study is the nature of the relationship between Ca 2ϩ release channels in the ER and plasmalemmal SOC channels. The subplasmalemmal localization of IP3R and RyR in Jurkat T cells (45) and the dependence of Ca 2ϩ release via RyR and/or IP3R on SOC channel activation imply that the IP3R and/or RyR may be coupled to the plasmalemmal SOC channels. In muscle, interaction between the RyR and plasmalemmal Ca 2ϩ channels occurs in highly specialized junctional regions where the sarcoplasmic reticulum is located in close apposition to the plasma membrane (46). In T cells, the functional SOC channels were found at the junctional sites of the ER with the plasma membrane (43,44), implying that functional interactions between the SOC channels and IP3R and/or RyR may take place. We speculate that similarly to muscle cells (47), the interaction between SOC channels and IP3R and/or RyR may be mediated either by conformational changes of SOC channel that do not require extracellular Ca 2ϩ entry or via CICR mechanism (Fig. 5). The existence of the physical coupling between SOC channels and IP3R and/or RyR in T cells, and whether SOC channel activation may trigger Ca 2ϩ release from the intracellular channels in the absence of Ca 2ϩ entry, remains to be determined.