Transient Tyrosine Phosphorylation of Human Ryanodine Receptor upon T Cell Stimulation*

The ryanodine receptor of Jurkat T lymphocytes was phosphorylated on tyrosine residues upon stimulation of the cells via the T cell receptor/CD3 complex. The tyrosine phosphorylation was transient, reaching a maximum at 2 min, and rapidly declined thereafter. In co-immunoprecipitates of the ryanodine receptor, the tyrosine kinases p56 lck and p59 fyn were detected. How-ever, only p59 fyn associated with the ryanodine receptor in a stimulation-dependent fashion. Both tyrosine kinases, recombinantly expressed as glutathione S -trans-ferase (GST) fusion proteins, phosphorylated the immunoprecipitated ryanodine receptor in vitro . In permeabilized Jurkat T cells, GST-p59 fyn , but not GST-p56 lck , GST-Grb2, or GST alone, significantly and con-centration-dependently enhanced Ca 2 (cid:1) release by cyclic ADP-ribose. The tyrosine kinase inhibitor PP2 specifically blocked the effect of GST-p59 fyn . This indicates that intracellular Ca 2 (cid:1) release via ryanodine receptors may be modulated by tyrosine phosphorylation during T cell activation.

Ryanodine receptors (RyR) 1 are large tetrameric Ca 2ϩ channel-forming proteins. Three different isoforms of RyR have been identified, termed types 1, 2, and 3 (also known as skeletal muscle type RyR, cardiac muscle type RyR, and brain type RyR (reviewed in Ref. 1)). There is indeed abundant expression of these particular isoforms in the respective tissues, but there is also increasing evidence for low expression of different RyR in other tissues (2).
In skeletal and cardiac muscle, RyR are the major intracellular Ca 2ϩ release channels, which control contractility by Ca 2ϩ -induced Ca 2ϩ release (reviewed in Ref. 3). In neuronal cells, Ca 2ϩ release via RyR is involved in fundamental brain function, such as long term depression (4).
Ca 2ϩ release via RyR is controlled by Ca 2ϩ itself (5). In addition, investigations in different cell systems suggest that the cyclic nucleotide cADPR may also function as an endoge-nous ligand for RyR-mediated Ca 2ϩ -release (reviewed in Refs. 6 and 7).
Activation of T lymphocytes is a fundamental part of the immune system to ensure protection against foreign antigens. One of the intracellular signaling pathways essentially necessary for T cell activation is a sustained elevation of [Ca 2ϩ ] i (reviewed in Ref. 8). The sustained elevation of [Ca 2ϩ ] i results in activation of many proteins (reviewed in Ref. 9). Of particular interest is the protein phosphatase calcineurin, which, upon activation by Ca 2ϩ /calmodulin, dephosphorylates the transcription factor nuclear factor of activated T cells (10). Only this dephosphorylated form of nuclear factor of activated T cells can enter the nucleus to activate multiple gene expression, e.g. expression of interleukin-2 (10).
Among the first biochemical events that are observed during T cell activation is phosphorylation of multiple proteins on tyrosine residues. These phosphorylation events are mediated by a concerted action of several protein-tyrosine kinases (PTK), including Src family PTKs p56 lck and p59 fyn (11)(12)(13)(14).
Recently, we have provided evidence that cADPR and RyR play an important role in the sustained phase of Ca 2ϩ signaling in T cells (15). Since TCR/CD3 complex-mediated Tyr phosphorylation of the type 1 Ins(1,4,5)P 3 receptor has been described recently (16), in this study we investigated (i) whether also RyR might be phosphorylated on tyrosine residues in a stimulationdependent fashion, and (ii) whether such tyrosine phosphorylation might have a functional effect on Ca 2ϩ release via RyR.
All other chemicals used were of the highest purity grade available. MilliQ water (Millipore Waters, Eschborn, Germany) was used for the preparation of all buffers.
Localization of RyR by Confocal Microscopy-The cells were fixed using 2% (w/v) p-formaldehyde in buffer A containing (mM) 140 NaCl, 5 KCl, 1 MgSO 4 , 1 CaCl 2 , 1 NaH 2 PO 4 , 5.5 glucose, and 20 HEPES (pH 7.4) for 15 min at room temperature. Then, the cells were rinsed twice with buffer A. Permeabilization was carried out by incubation in methanol for 5 min at room temperature. Afterwards, the cells were again rinsed twice with buffer A. Subsequently, the cells were incubated with anti-RyR common mAb (Calbiochem) at 1 g/100 l for 60 min at room temperature. Then, excess of mAb was removed by rinsing the cells twice with buffer A. Finally, a secondary fluorescein isothiocyanate-labeled goat anti-mouse antiserum (Molecular Probes Europe, Leiden, The Netherlands) was added at 4 g/100 l for 60 min at room temperature in the dark. Then, excess of the antiserum was removed by rinsing the cells twice with buffer A. Cells stained in this way were kept on ice in the dark until use (usually within the next 90 min).
Confocal microscopy was performed using a monochromator-based imaging system (Improvision, Heidelberg, Germany) built around a Leica DM IRBE microscope at 100-fold magnification. Images were taken with a 12 bit gray-scale CCD camera (type C4742-95-12NRB; Hamamatsu, Enfield, United Kingdom) on 30 consecutive horizontal (z) planes, each 0.5 m, from the bottom to the top of the cell. The spatial resolution in the horizontal plane (xy plane) of the camera was 0.067 m/pixel. Excitation wavelength was 485 nm, and emission light was filtered at 525 nm. Raw data images were stored on hard disk. To obtain digital confocal images, mathematical deconvolution based on the pointspread algorithm was carried out using the Openlab Confocal Imaging software module (Ref. 19; Improvision, Heidelberg, Germany). Usually, deconvolution was carried out using two or three neighbors on each side of the central horizontal plane of the cell; the magnitude setting of removal of stray light was 0.2300 -0.2400 (on a scale between 0 (no removal) and 0.9999 (complete removal of stray light)).
Preparation of the Subcellular Fraction P10 from Jurkat T Lymphocytes-The cells (10 9 lymphocytes) were harvested, centrifuged twice (500 ϫ g, 5 min, room temperature), and washed twice in 5 ml of 20 mM HEPES (pH 7.5), 110 mM NaCl. Following an incubation period of 20 min at 37°C, the cells were homogenized on ice in the presence of protease inhibitors (AEBSF, 500 M; leupeptin, 20 M; pepstatin, 1 M; antipain, 10 M) and dephostatin (100 M) using a tight Potter-Elvehjem homogenizer (1500 units/min, 30 strokes). All further steps were carried out at 4°C. After removal of cell debris and unbroken cells by low speed centrifugation (500 ϫ g, 10 min), the supernatant was ultracentrifuged (20 min, 10000 ϫ g, 4°C) to obtain the membrane fraction P10. This fraction was stored in aliquots at Ϫ70°C.
Protein Assay-For protein determination the Bio-Rad protein assay (Bio-Rad, Mü nchen, Germany) was used as microassay with bovine serum albumin (fraction V; Sigma, Deisenhofen, Germany) as standard.
Immunoprecipitation of RyR-RyR from P10 membranes, known to be the major source for RyR in Jurkat T cells (15), were solubilized by addition of lysis buffer (25 mM HEPES, pH 7.2, 150 mM NaCl, 0.25% CHAPS (w/v), protease inhibitors as above). 20 g of anti-RyR common mAb (Calbiochem) or anti-RyR common polyclonal antiserum (Santa Cruz) were coupled to Protein G-Sepharose (50 l of a 50% (w/v) suspension) in 500 l of lysis buffer (1.5 h, 4°C, continuous shaking). Anti-RyR common coupled to Protein G-Sepharose (50 l) was then incubated with solubilized P10 membranes (400 g of protein) in lysis buffer for 1.5 h at 4°C under continuous shaking. Then, the Protein G-Sepharose beads were centrifuged (13,000 ϫ g, 1 min, 4°C), the supernatant was discarded, and the immunoprecipitate rinsed with lysis buffer; the washing procedure was then repeated three times (for coprecipitation experiments, two times).
Western Blotting-Immunoprecipitated RyR protein was boiled at 95°C for 5 min and subjected to SDS-PAGE in a 6% gel (3% stacking gel) under reducing conditions. To ensure migration of the large RyR subunits (ϳ550 kDa) into the gel, the gel run was extended for another 60 min after the bromphenol blue band had reached the end of the gel. In co-immunoprecipitation experiments, 7.5% polyacrylamide gels were used; also, in these experiments, gel runs were terminated as soon as the bromphenol blue band had reached the end of the gel.
Proteins were subsequently transferred onto nitrocellulose sheets by tank blotting (18 h, 550 mA constant, 4°C). The nitrocellulose sheets were blocked with fatty acid-free milk powder (5% w/v in TBS buffer (1.4 M NaCl, 0.1 M Tris-HCl, pH 8.0) with 0.5% Tween 20) for 1 h at room temperature, and immunostained with the primary antibody (in TBS buffer with 0.5% Tween 20) for 18 to 24 h at 4°C. Concentrations of the primary antibodies were: (i) anti-RyR common mAb, 5 g/ml; (ii) antiphosphotyrosine mAb PY99, 1 g/ml; (iii) anti-phosphotyrosine antiserum, 2 g/ml; (iv) anti-p56 lck mAb, 1 g/ml; (v) anti-p59 fyn mAb, 1 g/ml. Then, the membrane was incubated in TBS with 0.5% Tween 20 and 2.5% (w/v) fatty acid-free milk powder three times for 10 min. Subsequently, horseradish peroxidase-labeled goat anti-mouse antiserum (Dianova, Hamburg, Germany) was added (dilution in TBS with 0.5% Tween 20 and 2.5% (w/v) fatty acid-free milk powder as recommended by the manufacturer, e.g. 1:2000 to 1:10,000) for 1 h at room temperature. Then, the following washing procedure was carried out with the membrane: two times briefly with TBS, two times for 10 min in TBS with 0.5% Tween 20, for 10 min in TBS with 3% Tween 20, two times for 15 min in TBS with 0.5% Tween 20, two times for 30 min in TBS with 0.5% Tween 20, and two times for 30 min in TBS alone. Finally, the blots were developed using the ECL kit (Amersham Pharmacia Biotech) according to the manufacturer's instructions.
In Vitro Tyrosine Phosphorylation of RyR-P10 membranes were prepared from unstimulated Jurkat T cells as described above. The membrane pellet was resuspended in kinase buffer (20 mM MOPS, pH 7.0, 5 mM MgCl 2 , 5 mM MnCl 2 , 150 mM NaCl) at 10 mg of protein/ml, and 500 l of the suspension were incubated with 1 g of recombinant GST-p56 lck or GST-p59 fyn (20) in the presence of 1 mM ATP at 37°C. To terminate the incubation, protein samples (100 l) were removed and rapidly frozen in liquid nitrogen. The samples were then solubilized (0.25% CHAPS, w/v) and subjected to immunoprecipitation using anti-RyR common mAb (Calbiochem) and SDS-PAGE in a 6% gel (3% stacking gel) under reducing conditions. Western blotting using anti-Tyr(P) mAb (PY99) was carried out as described above.
Ca 2ϩ Release Assay in Permeabilized Jurkat T Cells-Jurkat T cells were permeabilized as described (21) with the following modifications: saponin concentration (final) 40 g/ml, incubation period with saponin 17.5 or 20 min.
Measurements with 900 l of cell suspension (containing ϳ1.2 ϫ 10 7 permeabilized cells) in the presence of 1 M fura2/free acid were carried out in standard quartz fluorescence cuvettes and were started by addition of creatine kinase (20 units/ml final concentration) and creatine phosphate (20 mM final concentration). Fluorescence was recorded in a Hitachi F2000 instrument with wavelength settings of 340 and 380 nm for excitation (alternating) and 495 nm for emission.
Subsequent addition of ATP (1 mM final concentration) completed uptake of Ca 2ϩ ions into stores, and the resulting basal Ca 2ϩ concentration was usually between 100 and 500 nM, depending on the individual preparation. At the beginning of each series of experiments, the quality of the permeabilized cell preparation was checked by its responsiveness to Ins(1,4,5)P 3 (4 M final concentration) and cADPR (10 M final concentration). Chelex resin (Sigma) was added generally to reagent solutions, e.g. cADPR, Ins(1,4,5)P 3 , GST fusion proteins, to remove contaminations of Ca 2ϩ .

RESULTS
Jurkat T lymphocytes express the RyR as shown by immunoprecipitation and Western blotting (Fig. 1A), and confocal microscopy (Fig. 1B). The amount of RyR detected by an anti-RyR common mAb did not change significantly within the first minutes upon stimulation of the TCR/CD3 complex (Fig. 1A). When 10,000 ϫ g membranes were prepared in the presence of the Tyr phosphatase inhibitor dephostatin (100 M), a transient phosphorylation of Tyr residues of the RyR in response to stimulation of the TCR/CD3 complex was observed (Fig. 1A).
Using anti-RyR common mAb and a fluorescein isothiocyanatelabeled goat anti-mouse antiserum as secondary antibody in confocal microscopy, the main localization of the RyR was found to be close to the plasma membrane in permeabilized Jurkat cells (Fig. 1B). In intact cells, no staining was observed (data not shown). Similar results were obtained using 4,4difluoro-4-bora-3a,4a-diaza-s-indacene-ryanodine (Molecular Probes) for staining (data not shown). These results indicate that RyR are localized within membrane systems close to the inner surface of the plasma membrane.
Phosphorylation of Tyr residues of the RyR was transient, showing marked phosphorylation at 2 min (Fig. 1A). A more detailed analysis of the time course revealed very weak phosphorylation by 15 and 30 s, whereas substantial phosphorylation was observed at 1 and 2 min (Fig. 1C). At later time points, e.g. 5 min (Fig. 1A) or 10 or 20 min (data not shown), no phosphorylation on Tyr residues could be detected. These results were obtained with two different anti-phospho-Tyr antibodies, the mAb PY99 (Santa Cruz Biotechnology) and a polyclonal anti-phosphotyrosine antiserum (rabbit polyclonal IgG; Upstate Biotechnology).
Membrane distribution of RyR, taken together with an ex-tremely rapid kinetics of its tyrosine phosphorylation following TCR/CD3 ligation, implicated p59 fyn and p56 lck in this event. These PTKs are localized to the membrane due to their Nterminal myristoylation (22,23) and palmitoylation (24,25); are constitutively associated with the intracellular domains of several surface proteins, including TCR/CD3 (p59 fyn ) and CD4 (p56 lck ) (26,27); and respond very rapidly to TCR/CD3 stimulation (12,28). Therefore, p59 fyn and p56 lck are candidates to be involved in Tyr phosphorylation of the RyR.
To identify PTK(s) phosphorylating RyR in Jurkat T cells, co-immunoprecipitation experiments using anti-RyR antiserum were carried out (Fig. 2). In addition to tyrosine phosphorylation of the RyR (Fig. 2, left panel, 2-min time point), marked Tyr phosphorylation of a band (or a double band) near 55-60 kDa was detected in RyR immunoprecipitates (Fig. 2,  left panel). The time course of phosphorylation of this band/ double band was more rapid as compared with Tyr phosphorylation of the RyR itself (Fig. 2, left panel), thus being consistent with the idea that this band represents the PTK(s) associated with and phosphorylating RyR. Analysis of RyR immunoprecipitates for individual PTKs demonstrated that p59 fyn and p56 lck were associated with RyR in Jurkat T cells. Interestingly, p59 fyn was not found in RyR immunoprecipitate from unstimulated cells, but was co-precipitated in a time-and stimulation-dependent fashion (Fig. 2, middle panel). The same experiment carried out for p56 lck showed co-immunoprecipitation at all time points, regardless of whether the cells had been stimulated or not (Fig. 2, right panel).
The PTKs implicated in RyR tyrosine phosphorylation, p56 lck and p59 fyn , were then expressed as GST fusion proteins and tested for the ability to phosphorylate RyR in vitro. GST-p59 fyn and GST-p56 lck both stimulated marked phosphorylation on tyrosine residues of RyR (Fig. 3). In these experiments P10 membranes were incubated with the GST kinases in the presence of ATP. Interestingly, the phosphorylation was not transient, as observed in vivo (Fig. 1), but was stable for at least 20 min. This indicates loss or inhibition of phosphotyrosine-protein phosphatase(s) in the P10 preparation on the one hand (kinase assay), or loss of direct contact of PTKs and RyR during spatial redistribution in vivo.
Jayaraman et al. (16) have reported that tyrosine phosphorylation of the type 1 InsP 3 receptor by p59 fyn in the presence of Ins(1,4,5)P 3 resulted in enhanced channel opening as measured by single-channel recordings in lipid planar bilayers. Using permeabilized Jurkat T cells, a well established model to analyze Ca 2ϩ release by Ins(1,4,5)P 3 and cADPR (18, 29), addition of  side). Subsequent tank blotting and immunostaining were carried out as described under "Experimental Procedures." Antibodies used for immunostaining: A, anti-RyR common mAb (Calbiochem) and anti-phosphotyrosine rabbit polyclonal antiserum (Upstate Biotechnology); C, anti-RyR common mAb (Calbiochem) and anti-phosphotyrosine PY 99 (Santa Cruz). B, confocal microscopy was carried out as described under "Experimental Procedures" using cells fixed with p-formaldehyde and permeabilized by methanol. Anti-RyR common mAb (Calbiochem) was used as primary and fluorescein isothiocyanate-labeled goat anti-mouse antiserum as secondary antibody. Raw images were obtained at 100-fold magnification, and confocal images were calculated using the point-spread algorithm. For comparison, the bright field image of the cell is shown in the upper part of B.
GST-p59 fyn , additional GST fusion proteins, or GST alone prior to addition of cADPR did not have significant effects on the amplitude of cADPR-mediated Ca 2ϩ -release (Fig. 4, left panel). In addition, the sensitivity of the permeabilized T cells to different concentrations of cADPR (0.1 to 10 M) remained essentially unchanged with prior addition of GST-p59 fyn (data not shown); in other words, there was no shift in the concentration-response curve for cADPR. Next, we added GST-p59 fyn a few minutes later than cADPR, a time point where the response to cADPR was still highly active (Fig. 4, middle panel). Under these conditions, GST-p59 fyn concentration-dependently evoked additional Ca 2ϩ release. Importantly, GST-p56 lck , GST-Grb2, or GST alone used as controls only very slightly increased cADPR-mediated Ca 2ϩ release under these conditions (Fig. 4, right panel). The effect of GST-p59 fyn was due to its catalytic activity because addition of the tyrosine kinase inhibitor PP2 1 min before addition of GST-p59 fyn concentration-dependently blocked the enhancing effect of GST-p59 fyn without disturbing the effect of cADPR itself (Table  I). Interestingly, cADPR could be substituted completely by 4-chloro-3-ethylphenol (4-CEP, Ref. 30) and partially by Ca 2ϩ ions, known agonists of RyR. Ca 2ϩ release activated by 4-CEP and Ca 2ϩ was significantly enhanced by GST-p59 fyn (Table II), indicating that the conformational change induced by RyR channel opening might be sufficient to make regulatory tyrosine residues accessible to GST-p59 fyn .

DISCUSSION
In the present study we have shown that (i) the RyR of T cells is localized in intracellular membrane systems close to the plasma membrane, (ii) the RyR was transiently phosphorylated on Tyr residues upon stimulation of the TCR/CD3 complex, (iii) p59 fyn and p56 lck were detected in co-immunoprecipitates of the RyR, (iv) recombinant GST fusion proteins of p59 fyn and p56 lck both phosphorylated immunoprecipitated RyR on tyrosine residues in vitro, and (v) GST-p59 fyn , but not GST-p56 lck , GST-Grb2, or GST alone, enhanced Ca 2ϩ release by cADPR.
Although phosphorylation on tyrosine residues has not yet been demonstrated for RyR, there are several reports on phosphorylation of serine and threonine residues. The type 1 RyR was phosphorylated in vitro by protein kinase A and Ca 2ϩ / calmodulin-dependent protein kinase, resulting in functional recovery of the channel, which was inhibited by 2.6 mM Mg 2ϩ (31). Similarly, Ser/Thr phosphorylation also enhanced the sensitivity of the type 1 RyR toward Ca 2ϩ and ATP (32). The type 2 RyR was phosphorylated by exogenously added protein kinase A, Ca 2ϩ /calmodulin-dependent protein kinase, and cyclic GMP-dependent protein kinase. However, the Ca 2ϩ /calmodulindependent protein kinase was much more effective than the other kinases; additionally, endogenous Ca 2ϩ /calmodulin-dependent protein kinase appears to associate firmly with the type 2 RyR (33). Type 2 RyR phosphorylation by protein kinase A dissociated FK506-binding protein 12.6 from the channel complex, leading to enhanced channel opening; importantly, it has been shown in failing hearts that the type 2 RyR is hyperphosphorylated by protein kinase A, leading to markedly increased channel opening (34). Pancreatic ␤-cells also express the type 2 RyR; in those cells activation of protein kinase A enhanced activation of RyR by caffeine, 4-chloro-3-ethylphenol, or 3,9-dimethylxanthine (30). Importantly, [ 3 H]ryanodine binding was also found to be stimulated by phosphorylation of the brain type (type 3) RyR (35).
Analysis of the type 3 RyR amino acid sequence reveals that multiple tyrosine residues, such as tyrosines 810, 1022, 2289, 4045, and 4684, could well be targeted by Src family PTKs, based on the substrate specificity of these kinases (36). Among the potential substrate sites, tyrosines 4045 and 4684 are also good candidates for binding to Src homology 2 domains of Src   FIG. 2. Tyrosine phosphorylation of p56 lck and p59 fyn co-immunoprecipitated with the RyR. Jurkat T cells were stimulated as indicated by OKT3 (10 g/ml). Homogenization, preparation, and solubilization of P10 membranes and immunoprecipitation using anti-RyR common polyclonal antiserum were carried out as described in Fig. 1. Protein was separated by SDS-PAGE (7.5% separation gel, 3% stacking gel). The mass of protein in individual lanes corresponds to the mass of RyR immunoprecipitated by 1 g of anti-RyR common polyclonal antiserum. Subsequent tank blotting and immunostaining were carried out as described under "Experimental Procedures." Antibodies used for immunostaining: left panel, antiphosphotyrosine mAb PY 99, middle panel, anti-p59 fyn mAb; right panel, anti-p56 lck mAb (all from Santa Cruz). Since the signal with anti-p56 lck was more intense as compared with mAb PY 99 or anti-p59 fyn , the exposure time of the anti-p56 lck blot (right panel) was shortened. This shorter exposure time explains the lack of (artificial) "staining" of the lower edge of the blot. family PTKs, based on the specificity of phosphotyrosine sequences recognized by various Src homology 2 domains (37).
The observed p56 lck /p59 fyn -mediated tyrosine phosphorylation of RyR is consistent with the notion of the central role of these PTKs in early events of T cell signaling. p59 fyn and p56 lck are thought to trigger the signal transduction pathways initiated by TCR/CD3 ligation by phosphorylating cytoplasmic domains of the TCR/CD3 complex and activating other PTKs involved in these events (38,39). These PTKs, in turn, phosphorylate phospholipase C-␥ and linker for activation of T cells, thus triggering one pathway of Ca 2ϩ mobilization in T cells (38,39). Furthermore, it has been shown that p59 fyn and p56 lck can phosphorylate phospholipase C-␥ directly in vitro (40). Therefore, p59 fyn and p56 lck play a crucial role in Ca 2ϩ mobilization in T cells. The tyrosine phosphorylation of RyR described in this report may represent another mechanism by which p59 fyn affects Ca 2ϩ concentrations in T cells in vivo. Interestingly, tyrosine phosphorylation of RyR at the 2-min time point was in the same order of magnitude as compared with the band likely representing p59 fyn (Fig. 2, left panel). It is difficult to relate these relative staining intensities to the proportion of RyR that are actually tyrosine-phosphorylated. However, even a smaller fraction of tyrosine-phosphorylated RyR may in fact be sufficient to act as a triggering system for Ca 2ϩ -induced Ca 2ϩrelease, thereby also engaging non-phosphorylated RyR. Thus, such a model is well compatible with a functional role of a tyrosine-phosphorylated subfraction of RyR during TCR/CD3mediated Ca 2ϩ signaling.
Marks and colleagues (16) have shown recently that the type 1 Ins(1,4,5)P 3 receptor is tyrosine-phosphorylated in Jurkat T cells upon stimulation of the TCR/CD3 complex. In addition, they could show that this phosphorylation increased the open probability of the isolated Ins(1,4,5)P 3 receptor as measured in lipid planar bilayer experiments (16). To investigate a possible similar influence of tyrosine phosphorylation of the RyR of T cells, we have used permeabilized Jurkat T cells (29). In such permeabilized cells, recombinant GST-p59 fyn , but not GST-p56 lck , GST-Grb2, or GST alone, concentration-dependently and specifically enhanced cADPR-mediated Ca 2ϩ release in the presence of ATP and an ATP-regenerating system. Surprisingly, this effect required initial activation of RyR by cADPR. This activation changed two parameters of the system; first, [Ca 2ϩ ] was elevated, and second, the RyR Ca 2ϩ channels switched to the open conformation. Thus, one possible explanation would be that an elevated Ca 2ϩ concentration alone is sufficient for the catalytic effect of GST-p59 fyn on RyR. However, GST-p59 fyn does not require Ca 2ϩ ions for its catalytic activity. On the other hand, it is well demonstrated that Ca 2ϩ ions act as agonists at the RyR and thereby induce a conformational change of the RyR/Ca 2ϩ channel, as visualized by cryo-electron microscopy and subsequent image processing techniques (41). This view is supported by the finding that activation of RyR by the agonists 4-CEP and Ca 2ϩ was en- GST-p59 fyn enhanced cADPR-mediated Ca 2؉ release in permeabilized Jurkat T cells. Jurkat T cells were permeabilized as detailed under "Experimental Procedures." [Ca 2ϩ ] was measured by ratiometric fluorimetry in the presence of fura2/free acid (1 M) after reloading of Ca 2ϩ stores by addition of creatine kinase, creatine phosphate, and ATP. Left panel shows representative tracings obtained with additions of GST or GST fusion proteins (2 g each) followed by cADPR (10 M); the bar chart below is the data summary as mean Ϯ S.E. (n ϭ 3). Middle panel represents the concentration-response curve for the enhancing effect of GST-p59 fyn when added after cADPR; data in the lower part are mean Ϯ S.E. (n ϭ 3-5). In the the right panel, additional GST fusion proteins or GST alone (each 2 g) were added as indicated; the bar chart represents mean values Ϯ S.E. (n ϭ 5-8).  Fig. 4 (middle panel) using 2 g of GST-p59 fyn to enhance the cADPR-mediated Ca 2ϩ release, except that the tyrosine kinase inhibitor PP2 was added 1 min after cADPR, but 1 min before GST-p59 fyn .  ] was measured in permeabilized Jurkat T cells by ratiometric fluorimetry in the presence of fura2/free acid (1 M) as described under "Experimental Procedures." Initial Ca 2ϩ release via RyR was activated by cADPR (10 M), 4-CEP (500 M), or addition of CaCl 2 (720 nM). One minute later 2 g of GST-p59 fyn was added. Data are Ca 2ϩ increases subsequent to GST-p59 fyn addition, expressed as percentage of the effect obtained with GST-p59 fyn subsequently to cADPR as initial Ca 2ϩ releasing agent. hanced by GST-p59 fyn in a similar way as for cADPR. Thus, conformational changes are very likely to result from binding of these three ligands (this statement does not necessary imply that cADPR binds directly to RyR; additional proteins may be involved here). A similar observation was made when the enhancement of Ins(1,4,5)P 3 stimulation of the type 1 Ins(1,4,5)P 3 receptor from T cells by p59 fyn was measured in bilayer experiments. In that case, the effect was observed only at high [Ca 2ϩ ] at the cytosolic side, indicating a similar Ca 2ϩ -mediated conformational change of the type 1 InsP 3 receptor (16). Taken together, these data indicate that the tyrosine residue(s) that could theoretically be phosphorylated by GST-p59 fyn are not accessible when the RyR Ca 2ϩ channel is in the closed conformation.
In conclusion, we have shown transient tyrosine phosphorylation of the RyR of T cells in response to stimulation of the TCR/CD3 complex. In addition, in permeabilized cells Ca 2ϩ release by cADPR and 4-CEP were significantly enhanced by GST-p59 fyn . Our demonstration that the effect of GST-p59 fyn was specific, concentration-dependent, and could be inhibited by the tyrosine kinase inhibitor PP2 provides strong evidence that tyrosine phosphorylation of RyR may play a crucial role in the early phase of T cell Ca 2ϩ signaling and thereby may constitute an additional element of fine tuning of T lymphocyte Ca 2ϩ signaling. In a broader context, our data provide first evidence that, in addition to serine/threonine protein kinases, PTKs are also involved in regulation of RyR.