Src homology region 2 domain-containing phosphatase 1 (SHP-1), a cytoplasmic protein tyrosine phosphatase, plays an important role for the regulation of signaling from various hematopoietic cell receptors. Although SHP-1 is shown to be a negative signal modulator in mast cells, its precise molecular mechanisms are not well defined. To elucidate how SHP-1 regulates mast cell signaling, we established bone marrow-derived mast cells from SHP-1-deficient motheaten and wild-type mice and analyzed downstream signals induced by cross-linking of high affinity IgE receptor, FcεRI. Upon FcεRI ligation, motheaten-derived bone marrow-derived mast cells showed enhanced tyrosine phosphorylation of Src homology region 2 domain-containing leukocyte protein of 76 kDa (SLP-76) and linker for activation of T cells, activation of mitogen-activated protein kinases and gene transcription and production of cytokine. Because the activity of Syk, responsible for the phosphorylation of SLP-76 and linker for activation of T cells, is comparable irrespective of SHP-1, both molecules might be substrates of SHP-1 in mast cells. Interestingly, the absence of SHP-1 expression disrupted the association between SLP-76 and phospholipase Cγ, which resulted in the decreased phospholipase Cγ phosphorylation, calcium mobilization, and degranulation. Collectively, these results suggest that SHP-1 regulates FcεRI-induced downstream signaling events both negatively and positively by functioning as a protein tyrosine phosphatase and as an adaptor protein contributing to the formation of signaling complex, respectively.
Mast cells play a central role in the induction of allergic and inflammatory reactions. Mast cells express the high affinity receptor for IgE (FcεRI),4 and cross-linking of FcεRI with IgE and multivalent Ag leads to mast cell activation that results in degranulation, release of granules containing preformed chemical mediators, and the synthesis and secretion of cytokines and chemokines (1, 2). FcεRI belongs to the multichain immune receptor superfamily, consisting of one IgE-binding α-chain, one β-chain, and a homodimer of disulfide-linked γ-chains (3). β- and γ-chains that contain ITAMs in their cytoplasmic regions are essential for mediating downstream signaling events (4). Upon aggregation of FcεRI, the Src-family protein tyrosine kinase (PTK), Lyn, is activated and phosphorylates tyrosine residues in the ITAMs of the β- and γ-chains. Syk PTK is then recruited to the phosphorylated ITAMs of γ-chain via its Src homology region 2 (SH2) domains, becomes tyrosine phosphorylated, and is activated by Lyn (5, 6, 7). The activated PTKs phosphorylate a number of signaling molecules, including phospholipase Cγ (PLCγ) (8, 9), Vav (10), SH2 domain-containing leukocyte protein of 76 kDa (SLP-76) (11), linker for activation of T cells (LAT) (12), and linker for activation of B cells/non-T cell activation linker (13). Phosphorylation of these proteins induces calcium (Ca2+) mobilization and activation of MAPKs and is ultimately converged into degranulation and production of cytokines. Among these cellular substrates, LAT and SLP-76 play critical roles in the integration of signaling. Following receptor ligation, tyrosine phosphorylated LAT recruits a number of SH2-containing molecules such as Grb2-Sos complex responsible for the activation of Ras-ERK pathway (14, 15), Gads-SLP-76 complex (16), and PLCγ (15, 17). SLP-76 brought to the plasma membrane by interacting with phosphorylated LAT is then tyrosine-phosphorylated by Syk, which provides docking sites for other SH2-containing molecules such as Vav (18, 19), Nck (20), and ITK (21, 22). The central proline-rich region of SLP-76 mediates a constitutive interaction with PLCγ (23). Thus, PLCγ translocated to the plasma membrane is phosphorylated by Syk/Itk PTKs and activated. Accordingly, mast cells deficient for LAT (12) or SLP-76 (24) showed impaired tyrosine phosphorylation of PLCγ and Ca2+ mobilization following FcεRI ligation.
Given that the extent of tyrosine phosphorylation of cellular proteins is strictly balanced by PTKs and protein tyrosine phosphatases (PTPs), much attention has recently been focused on a role of PTPs in immune receptor-mediated signaling (25). The cytosolic SH2-domain containing phosphatase-1 (SHP-1) has two SH2 domains in N terminus, one catalytic domain and the C-terminal regulatory domain and is highly expressed in hematopoietic cells. Mutation of the Shp-1 gene is responsible for the motheaten (me) and viable motheaten phenotypes that develop a severe autoimmune and immunodeficiency syndrome (26, 27). SHP-1 is generally considered as a negative regulator of receptor-mediated signaling in lymphocytes, because B and T cells derived from SHP-1-deficient me mice are hyperresponsive to Ag receptor stimulation (28, 29). In mast cells, following FcεRI engagement, SHP-1 is recruited to the phosphorylated tyrosine residues in ITIMs found in the cytoplasmic regions of inhibitory receptors (30, 31, 32, 33, 34). As a result, SHP-1 is activated and is predicted to dephosphorylate various signaling molecules including the β/γ subunits of FcεRI and Syk (35), mediating negative regulation of FcεRI-initiated downstream signal transduction. However, knowledge of the molecular mechanisms underlying negative regulation by SHP-1 is still limited.
Recently, by introducing catalytically inactive form of SHP-1 (SHP-1-C/S) into rat basophilic leukemia cell line, RBL-2H3, we have shown that SHP-1 dephosphorylates Tyr-536 of SHP-1, leading to the decrease of its PTP activity (36). Because the overexpression of dominant negative mutants might have nonspecific effects and the presence of endogenous molecules might diminish the effects of mutants, it is necessary to use cells lacking the expression of specific molecules for understanding their physiological roles in cell signaling. In the present study, to elucidate the role of SHP-1 in FcεRI-signaling, we established bone marrow-derived mast cells (BMMC) from me and wild-type (WT) mice and analyzed tyrosine phosphorylation states of cellular proteins, activation of MAPKs, Ca2+ mobilization, and final outputs such as degranulation and cytokine production, all of which are induced by FcεRI cross-linking. Our results indicate that SHP-1 plays a negative role in FcεRI-induced activation of MAPKs and cytokine gene transcription and production possibly by dephosphorylating SLP-76 and LAT, whereas the phosphorylation of PLCγ, Ca2+ mobilization, and degranulation were positively regulated by SHP-1. For the positive regulation, SHP-1 appears to function as an adaptor molecule mediating the association of SLP-76 with PLCγ.
Materials and Methods
Abs and reagents
Anti-DNP IgE mAb (SPE-7), and DNP-human serum albumin (HSA) were purchased from Sigma-Aldrich. HRP-conjugated anti-phosphotyrosine (PY) mAb (PY20) and Abs against SHP-1, Lyn, Syk, ERK2, JNK2, p38, PLCγ-1, and PLCγ-2 were from Santa Cruz Biotechnology. Anti-LAT Ab and Abs against phosphotyrosine 191-human LAT (equivalent to pY195 murine LAT) and phosphotyrosine 226-human LAT (equivalent to pY235 murine LAT) were from Upstate Biotechnology. Anti-phospho-specific JNK, anti-phospho-specific p38, and anti-phosphotyrosine 171-human LAT (equivalent to pY175 murine LAT) Abs were from Cell Signaling Technology. Anti-phospho-specific ERK Ab was from Promega. Anti-SLP-76 Ab was raised in rabbits by immunizing with SH2 domain of mouse SLP-76 (aa 421–533) fused with glutathione S-transferase. All HRP-conjugated secondary Abs were described (37). ECL Western blotting detection kit and protein G Sepharose were obtained from GE Healthcare Bioscience.
Mice for the present experiments were obtained by mating C3HeBFeJ me/+ breeding pairs originally provided by The Jackson Laboratory. To detect Shp-1 gene mutation, PCR-based genotyping was performed as previously described (38). All animal experiments were conducted according to the guideline of Tokyo Metropolitan Institute for Neuroscience.
Cell culture and stimulation
Bone marrow cells from femurs of +/+ and me/me mice were cultured in IL-3-containing medium for 4–6 wk to generate WT- and me-BMMC, respectively, with >95% purity (c-Kit+, FcεRI+ as evaluated by flow cytometry). For cell stimulation, cells were incubated with 100 ng/ml SPE-7 overnight. After washing cells twice with PBS to remove excess SPE-7, the cells were precultured in the medium in the absence of IL-3 for 4 h and, then, stimulated with DNP-HSA (50 ng/ml otherwise indicated) for various periods of times.
Immunoprecipitation and Western blot analysis
After simulation, the reaction was stopped with ice-cold PBS containing 1 mM Na3VO4 and 2 mM EDTA (PBS-VE). The cells were washed twice with PBS-VE and lysed in TNE buffer (1% Nonidet P-40, 10 mM Tris-HCl (pH 7.5), 150 mM NaCl, 2 mM Na3VO4, and 2 mM EDTA). Subcelular fractionation to obtain cytoplasmic and membrane fractions was performed as described (39). The total cell lysates (TCL), cytoplasmic, or membrane fraction thus prepared were subjected to immunoprecipitation and Western blot analysis as described (37). The intensities of phosphorylated protein bands were measured with a Bio-Rad Imaging Densitometer (Bio-Rad Laboratories), and the results were standardized with the intensities of unphosphorylated protein bands and expressed as fold increase with the intensity of unstimulated cells being 1.0.
In vitro kinase assay
In vitro kinase assay for Syk was performed as described previously (37). In brief, immunoprecipitated Syk was washed three times with TNE buffer and then twice with kinase buffer (20 mM HEPES (pH 8.0), 150 mM NaCl, 10 mM MgCl2, 20 mM MnCl2). After washing, the immunoprecpitates were suspended in 10 μl of kinase buffer containing 10 μCi of [γ-32P] ATP (6000 Ci/ml) and myelin basic protein as an exogenous substrate, and incubated for 15 min at 37°C. The samples were resolved by SDS-PAGE and the resulting gels were dried and subjected to autoradiography.
Measurement of cytokines
SPE-7-sensitized cells were stimulated with 50 ng/ml DNP-HSA for 0.5 h (real-time PCR analysis) or 3 h (enzyme-linked immunosorbent assay). To quantify the transcription of TNF-α and IL-6 genes, real-time PCR was performed as described (36). Levels of TNF-α and IL-6 were first normalized with GAPDH levels and determined relative to the unstimulated WT-BMMC, which was arbitrarily designated a value of 1.0. The amounts of TNF-α and IL-6 in the cell culture supernatants were measured with a Cytoscreen ELISA kit (BioSource International). The results were expressed as amounts/ml per 1 × 106 cells.
Measurement of intracellular Ca2+ concentration
BMMC were labeled with the fluorescent Ca2+ indicator, fluo-3/AM as previously described (40), after which cells were washed and resuspended in HBSS containing 1 mM CaCl2. For measuring Ca2+ store release, cells were suspended in HBSS containing 1 mM EGTA. Fluo-3 fluorescence was monitored at 5-s intervals up to 3 min with a microplate fluorometer (Fluoroskan Ascent CF; Labsystems; excitation and emission at 485 and 527 nm, respectively). The cytosolic free calcium concentration ([Ca2+]i) was calculated using the equation: [Ca2+]i = KDa [(F − Fmin)/(Fmax − F)], where KDa is the dissociation constant of the Ca2+-Fluo-3 complex (450 nM), Fmax represents the maximum fluorescence (obtained by treating cells with 5 μM A23187), Fmin represents the minimum fluorescence (obtained for A23187-treated cells in the presence of 1 mM EGTA), and F is the actual sample fluorescence.
Degranulation was determined by measuring β-hexosaminidase release. Cells were sensitized with SPE-7 overnight, washed, and precultured in Tyrode’s buffer (10 mM HEPES, pH 7.4, 130 mM NaCl, 5 mM KCl, 1.4 mM CaCl2, 1 mM MgCl2, 5.6 mM glucose and 0.1% (wt/vol) BSA) for 4 h. After stimulation of cells with 5 or 50 ng/ml DNP-HSA at 37°C for 30 min, 40 μl of supernatant or cell lysates and 100 μl of 2 mM p-nitrophenyl-N-acetyl-β-D-glucosaminide (in 0.4 M citrate and 0.2 M phosphate buffer (pH 4.5)) were added to each well of a 96-well plate, and color was developed for 30 min at 37°C. Enzyme reaction was terminated by adding 200 μl of 0.2 M glycine-NaOH (pH 10.7). The absorbance at 405 nm was measured in a microplate reader (Bio-Rad 550; Nippon Bio-Rad Laboratories). Cells were lysed with 0.1% Triton X-100, and the β-hexosaminidase activity of the extracts was measured (total). The β-hexosaminidase activity in nonstimulated cells (the spontaneous release, <5% of the total was subtracted from the enzyme activity in stimulated cells (test). The percentage of β-hexosaminidase released into the supernatant was calculated using the following formula: release (%) = (test − spontaneous)/(total − spontaneous) × 100.
Tyrosine phosphorylation of 75- and 39-kDa proteins is increased in me-derived BMMC
We have previously identified SHP-1 as a substrate of SHP-1 itself based on the findings that tyrosine phosphorylation of SHP-1 is increased in RBL-2H3 cells transfected with SHP-1-C/S and that SHP-1 dephosphorylates Tyr-536 of SHP-1 (36). To identify further substrates of SHP-1 in mast cells, we established BMMC from SHP-1-deficient me and control WT mice, yielding me- and WT-BMMC, respectively. We cultured bone marrow cells in the presence of IL-3 for 4 wk to generate highly pure BMMC as assessed by the expression of FcεRI and c-Kit (data not shown). In addition, the expression level of FcεRI and c-Kit in me-BMMC was comparable to that in WT-BMMC (data not shown), suggesting that IL-3-dependent mast cell differentiation in vitro is not impaired in the absence of SHP-1. At first, WT-BMMC thus established were incubated with DNP-specific IgE mAb, SPE-7 (100 ng/ml), overnight, precultured in the medium without IL-3 for 4 h, and stimulated with increasing amounts of DNP-HSA (up to 100 ng/ml) for 5 min to determine the optimal condition of stimulation. Cells cultured overnight in the absence of SPE-7 were used as a negative control. The TCL from these cells were subjected to SDS-PAGE and Western blotting with anti-phosphotyrosine (PY) mAb. Tyrosine phosphorylation was comparable between cells after overnight incubation with or without SPE-7 (Fig. 1,A, lanes 1 and 2) and was increased as an Ag dose-dependent manner to 50 ng/ml DNP-HSA (Fig. 1,A). Based on this result, we concluded that SPE-7 itself failed to induce considerable tyrosine phoshporylation of cellular proteins and that cells would be stimulated with 50 ng/ml DNP-HSA for additional experiments otherwise indicated. We next compared tyrosine phosphorylation of WT- and me-BMMC after stimulation for the indicated periods. As shown in Fig. 1,B, increased tyrosine phosphorylation was observed in proteins of 75 kDa (pp75), 70–72 kDa, 55 kDa, and 39 kDa (pp39) in me-BMMC before and after FcεRI cross-linking, whereas tyrosine phosphorylation of 130-kDa protein was enhanced in WT-BMMC (pp130). Almost equal protein loading was verified by anti-ERK2 immunoblotting (Fig. 1, A and B, bottom panels). Among the proteins whose tyrosine phosphorylation was increased in me-BMMC, we focused on pp75 and pp39 because, based on the m.w., they are predicted to be SLP-76 and LAT, respectively, both of which play critical roles to transduce signals from FcεRI to downstream. Although tyrosine phosphorylation of 70–72-kDa and 55-kDa proteins was strongly increased in me-BMMC before and after FcεRI ligation, the nature of these proteins are currently unidentified.
SLP-76 and LAT are hyperphosphorylated in me-BMMC
To assess whether pp75 and pp39 are SLP-76 and LAT, respectively, we performed combined immunoprecipitation and Western blot analysis. After sensitizing with SPE-7 overnight, BMMC were either left unstimulated or stimulated with 50 ng/ml DNP-HSA for 2 and 5 min, and SLP-76 and LAT were immunoprecipitated and immunoblotted with anti-PY mAb. The results shown in Fig. 2,A demonstrated that basal as well as stimulation-induced tyrosine phosphorylation of SLP-76 and LAT was higher in me-BMMC than in WT-BMMC. To evaluate which Tyr residues in LAT showed increased phosphorylation, we made use of Abs against individual phosphotyrosine residues on LAT and found that the phosphorylation of Tyr-175, -195, and -235 was increased in me-BMMC (Fig. 2,B). Although these results might suggest that both SLP-76 and LAT are substrates of SHP-1 in mast cells, it is also possible that the lack of SHP-1 expression results in increased activation of PTKs. To examine this possibility, we determined the activity of Syk, a PTK responsible for the phosphorylation of both SLP-76 (41) and LAT (42) in mast cells. Cells were either unstimulated or stimulated for 15 min, and the immunoprecipitated Syk was subjected to anti-PY immunoblotting and in vitro kinase assay using myelin basic protein as an exogenous substrate. The results clearly demonstrated that the phosphorylation state and kinase activity of Syk were almost identical between in me- and in WT-BMMC (Fig. 3). Taken together with the previous findings that SHP-1 can directly dephosphorylate SLP-76 (39, 43) and LAT (44) in other cell types, these results suggest that both proteins are physiological substrates of SHP-1 in mast cells.
FcεRI-induced MAPK activation and cytokine production are increased in me-BMMC
Given that phosphorylated Tyr residues in LAT (14, 15, 16) and SLP-76 (18, 19, 20) would provide docking sites for several molecules that mediate MAPK activation, hyperphosphorylation of both molecules found in me-BMMC is expected to lead to an enhanced activation of MAPKs. To address this possibility, we sensitized BMMC with SPE-7 and stimulated them with 50 ng/ml DNP-HSA for 5 and 15 min, and examined the activation status of MAPKs by immunoblot with phospho-specific Abs against ERK, JNK, and p38. Consistent with the phosphorylation states of LAT and SLP-76, lack of SHP-1 expression resulted in an increased phosphorylation of ERK, JNK, and p38 (Fig. 4). It should be noted that increased phosphorylation of ERK was evident even in the absence of FcεRI cross-linking. Because FcεRI-mediated MAPK activation has been demonstrated to lead to gene transcription of several cytokines (35), we next compared the transcription of mRNA and production for TNF-α and IL-6 that are the major cytokines induced by FcεRI cross-linking in mast cells (35). Real-time PCR analysis revealed that mRNA production for TNF-α and IL-6 was strongly increased in the me-BMMC that were stimulated with SPE-7 and DNP-HSA for 0.5 h (Fig. 5,A). Accordingly, the production of both cytokines was much more increased in me-BMMC, as measured by enzyme-linked immunosorbent assay (Fig. 5 B). These results indicate that SHP-1 might negatively regulate MAPK-cytokine production pathway by possibly dephosphorylating LAT and SLP-76, critical linker/adaptor proteins for MAPK activation.
FcεRI-induced Ca2+ mobilization is impaired in me-BMMC
After binding to phosphorylated LAT and SLP-76 via SH2 domain, PLCγ-1 and -2 were tyrosine phosphorylated by Syk and ITK, and activated. This leads to the generation of inositol 1,4,5-triphosphate (IP3) and the release of Ca2+ from endoplasmic reticulum (ER), followed by the influx of extracellular Ca2+ (45). We next analyzed another downstream signaling event induced by FcεRI ligation, Ca2+ mobilization. As in the case of MAPK activation, it is also likely that hyperphosphorylation of LAT and SLP-76 in me-BMMC might enhance Ca2+ signaling because increased amount of PLCγ would be recruited to strongly phosphorylated Tyr residues of LAT and SLP-76. Cells were sensitized with SPE-7 overnight, labeled with Fluo-3/AM, and stimulated with 0.5–50 ng/ml DNP-HSA. Surprisingly, the results shown in Fig. 6,A clearly demonstrated that Ca2+ mobilization was almost completely abrogated with lower than 5 ng/ml DNP-HSA in me-BMMC. When cells were stimulated with 50 ng/ml Ag, an initial Ca2+ spike was comparable between WT- and me-BMMC (Fig. 6,A), but a sustained Ca2+ plateau was decreased in me-BMMC, suggesting that extracellular Ca2+ influx induced by ER Ca2+ depletion was impaired in me-BMMC. To test this possibility, in addition to Ag, we used thapsigargin (TG), a strong and selective inhibitor of the ER Ca2+-ATPase (46). In the medium containing Ca2+, TG is capable of depleting ER Ca2+ stores, thus activating store operated Ca2+ channels in the plasma membrane. Although TG-induced Ca2+ mobilization was slightly but reproducibly reduced in me-BMMC (Fig. 6,B), the difference between me- and WT-BMMC was marginal compared with that observed when cells were stimulated with Ag (Fig. 6,A). Next, we stimulated BMMC with TG or Ag (50 ng/ml DNP-HSA) in the presence of the Ca2+ chelator EGTA. This condition allows Ca2+ release from intracellular stores but not extracellular Ca2+ entry. As shown in Fig. 6 C, me-BMMC showed similar Ca2+ responses induced by FcεRI ligation or TG to WT-BMMC. These results suggest that impaired Ca2+ response elicited with the high concentration of Ag (50 ng/ml) in me-BMMC was caused by defective extracellular Ca2+ influx, although ER Ca2+ depletion-dependent Ca2+ influx machinery per se seem to be intact in me-BMMC.
Although it is widely accepted that Ca2+ mobilization is required for the degranulation (47), additional evidence implies Ca2+-independent pathways in degranulation (48, 49, 50, 51). To elucidate the potential role of Ca2+ in degranulation, we next measured the release of β-hexosaminidase from WT- and me-BMMC following FcεRI ligation. As shown in Fig. 6,D, lack of SHP-1 resulted in a marked decrease of degranulation when cells were stimulated with 50 ng/ml DNP-HSA, consistent with the impaired Ca2+ influx in me-BMMC. The stimulation with 5 ng/ml DNP-HSA, by which me-BMMC showed marginal Ca2+ response (Fig. 6 A), still induced β-hexosaminidase release, although slightly but significantly less than WT-BMMC. In addition, the presence of EGTA almost completely abolished FcεRI-mediated degranulation even after stimulation with high dose of Ag. These results collectively suggest that SHP-1 is required both for the entry of Ca2+ from the extracellular fluid when cells were stimulated with higher concentration of Ag and for the initial Ca2+ release from ER after weak FcεRI ligation, and that Ca2+ mobilization, especially extracellular Ca2+ influx, is required for the degranulation process.
Tyrosine phosphorylation of PLCγ and the association of SLP-76 with PLCγ are impaired in me-BMMC
Finally, we investigated the molecular mechanisms underlying the positive regulation of Ca2+ signaling by SHP-1. At first, we examined tyrosine phosphorylation of PLCγ-1 and -2 following the stimulation with 5 ng/ml Ag for 2 and 5 min. Consistent with Ca2+ mobilization, robust tyrosine phosphorylation of immunoprecipitated PLCγ-1 and -2 was not detected in me-BMMC (Fig. 7, A and B). In WT-BMMC, there was no difference in the degree and the kinetics of phosphorylation between PLCγ-1 and -2. This increased phosphorylation of PLCγ-1 and -2 might, at least in part, explain the hyperphosphorylation of p130 as observed in TCLs from WT-BMMC (pp130 in Fig. 1). When cells were stimulated with increased concentrations of Ag for 5 min, phosphorylation of PLCγ-2 was detectable at 5 ng/ml in WT-BMMC whereas tyrosine phosphorylation was detectable at 50 ng/ml Ag but not at 5 ng/ml in me-BMMC (Fig. 7,B). In the course of these analyses, we found another tyrosine-phosphorylated protein of 75 kDa coimmunoprecipitated with PLCγ-2 (Fig. 7,C) and PLCγ-1 (data not shown) in WT- but not in me-BMMC. Sequential immunoprecipitation and immunoblot revealed that this protein was SLP-76 (Fig. 7,C). The association between SLP-76 and PLCγ was detectable in the absence of stimulation (Fig. 7,C), possibly through interaction between proline-rich domain of SLP-76 and SH3 domain of PLCγ as reported previously in T cells (23). To confirm that the observed molecular complex was indeed present in mast cells, we examined whether SHP-1 immunoprecipitates contain SLP-76 and PLCγ in WT-BMMC. A constitutive association between SHP-1 and SLP-76 was observed in WT-BMMC as in B cells (52) and PLCγ-1 (data not shown) and PLCγ-2 were coimmunoprecipitated with SHP-1 before and after FcεRI stimulation (Fig. 7,D). Although the association among three proteins appeared to be constitutive, SLP-76 and PLCγ were inducibly tyrosine-phosphorylated following FcεRI ligation (Fig. 7, C and D). This might be explained, at least in part, by the translocalization of SHP-1 from cytosol to membrane after FcεRI stimulation (Fig. 7 E), where trimolecular complex is likely to be localized in the vicinity to the activated PTKs. Taken together, these results might indicate that SHP-1 is required for the association of PLCγ with SLP-76, the event being essential for the phosphorylation and activation of PLCγ, subsequent Ca2+ mobilization and degranulation.
SHP-1 is generally considered as a negative signal transducer, essentially by dephosphorylating critical signaling molecules. Accordingly, our present study using me- and WT-BMMC has shown that SHP-1 functions as a negative signal modulator for MAPK-cytokine production pathway possibly by dephosphorylating LAT and SLP-76 in mast cells. After FcεRI ligation, Lyn and Syk PTKs are rapidly activated (5, 6, 7), and activated Syk phosphorylates LAT and SLP-76, which then provide docking sites for SH2-containing Grb2/Sos complex (14, 15) and Vav (18, 19)/Nck (20), respectively. We conclude that SLP-76 and LAT are physiological substrates of SHP-1 based on the following findings. First, tyrosine phosphorylation of both molecules induced by FcεRI aggregation was increased in me-BMMC. Secondly, Syk activity was comparable between WT- and me-BMMC. Direct dephosphorylation of SLP-76 (39, 43) and LAT (44) by SHP-1 demonstrated in other cell types may support our conclusion. FcεRI-induced Lyn activation leads to tyrosine phosphorylation and activation of SHP-1, which results in the dephosphorylation and inactivation of SHP-1 itself (36). Thus, it is likely that the tyrosine phosphorylation state of LAT and SLP-76 is strictly balanced by the actions of three enzymes, Lyn, Syk, and SHP-1. Regarding the localization of SHP-1 in mast cells, many reports demonstrate that, after stimulation, SHP-1 is recruited to phosphorylated tyrosine residues in ITIMs of inhibitory receptors (30, 31, 32, 33, 34). Alternatively, a previous report by Xiao et al. (53) demonstrates another possible localization of SHP-1. They showed that SHP-1 is constitutively associated with FcεRI-β-chain (53), thus exerting an inhibitory control on the FcεRI-β ITAM (54) and FcεRI-β-associated Lyn (53). For the strict and effective regulation on the function of LAT and SLP-76 by Lyn, Syk, and SHP-1, all three enzymes should be localized properly during the course of FcεRI signaling. Although the fraction of membrane-localized SHP-1 appeared to increase after stimulation (Fig. 7 E), the experiments to determine the precise subcellular localization of SHP-1 before and after FcεRI ligation are currently in progress.
In T cells, a previous study demonstrated that Grb2 associates with LAT following TCR stimulation by binding its SH2 domain to three distal phospho-Tyr residues, Tyr-171, -191, and -226 (equivalent to Tyr-175, -195, and -235 in murine LAT) and that phospho-Tyr-171 and 191 in LAT are necessary for Gads binding (55). We found that SHP-1 deficiency resulted in an increase in phosphorylation of all three Tyr residues of LAT (Fig. 2,B), which might lead to enhancement of LAT-Grb2-Sos and LAT-Gads-SLP-76 complex formation and of MAPK activation (Fig. 4). Consistent with the published evidence that MAPK activation leads to gene transcription of several cytokines in mast cells (35), mRNA transcription and production of TNF-α and IL-6 were increased in me-BMMC (Fig. 5). Based on these results, we concluded that SHP-1 regulates MAPK-cytokine pathway negatively, possibly by dephosphorylating LAT and SLP-76. However, Xie et al. (35) demonstrated that the FcεRI-induced JNK activation and mRNA transcription of TNF-α were increased in RBL-2H3 cells expressing WT-SHP-1 whereas the expression of phosphatase-deficient SHP-1 had the opposite effect, indicating a positive regulation of these signaling events by SHP-1. The possible explanations for the discrepancy in the role of SHP-1 for the regulation of MAPK-cytokine pathway might be the difference in cell types used and/or experimental system. In contrast to me-BMMC that lack SHP-1 expression, RBL-2H3 cells expressing WT- or dominant negative SHP-1 also contain endogenous SHP-1. Thus, it is possible that the introduction of PTP activity-deficient SHP-1 enhances the activity of endogeous SHP-1, leading to strong suppression of certain signaling pathway as suggested our previous report (36).
This study also showed that SHP-1 has a positive role in the pathway leading to the activation of PLCγ and subsequent Ca2+ mobilization in mast cells. Regarding PLCγ activation, it has been demonstrated that, following stimulation, PLCγ is recruited to multiple binding sites in the cytoplasmic region of LAT, such as phoshpo-Tyr-132, -171, -191, and -226 of human LAT, where PLCγ is phosphorylated and activated (57). Given that at least Tyr-175, -195, and -235 of LAT were hyperphosphorylated in me-BMMC (Fig. 2,B), it is expected that FcεRI stimulation should induce the increased PLCγ recruitment to LAT and activation of PLCγ in me-BMMC. However, the phosphorylation of PLCγ-1 and -2 was considerably reduced in me-BMMC (Fig. 7, A–C), implying the presence of LAT-independent pathway for the phosphorylation/activation of PLCγ. Despite ∼55% sequence identity between SHP-1 and another SH2-containing phosphatase, SHP-2, it is generally believed that they have opposite roles in cell signaling. SHP-2 is considered as a positive signal transducer mainly based on the findings that cells expressing catalytically inactive SHP-2 or SHP-2-deficient cells exhibit reduced signal activation induced by growth factors and cytokine (56, 57). However, a recent study revealed that SHP-1 as well as SHP-2 plays a positive role in epidermal growth factor signaling (58). Knock down of either SHP-1 or SHP-2 causes significant reduction of ERK activation. This effect is mediated by the linker function of SHP-1. SHP-1 and SHP-2 interact with each other and SHP-2 binds directly to tyrosine-phosphorylated Gab1, thus forming trimolecular signaling complex. In our system, positive regulation of FcεRI-signaling is likely to be mediated by protein-protein interactions. In the presence of SHP-1, PLCγ-1/2 and SLP-76 can form stable complex (Fig. 7, C and D) that might be required for the tyrosine phosphorylation of PLCγ. In fact, it has been shown that SHP-1 constitutively associated with SLP-76 in B cells (52) and that PLCγ binds directly to the proline-rich region of SLP-76 via its SH3 domain (23). Then, how does this SHP-1-mediated protein-protein interactions function in the signaling pathway? One possibility is that SHP-1 determines the localization of SLP-76/PLCγ complex where PLCγ becomes readily phosphorylated upon stimulation. Again, the association of SHP-1 with FcεRIβ-chain would localize SLP-76/PLCγ complex in the vicinity of FcεRIγ-chain where Syk is recruited and activated. Indeed, although trimolecular complex was constitutively observed, SLP-76 and PLCγ were inducibly phosphorylated in WT-BMMC (Fig. 7, C and D). Furthermore, SHP-1 translocated to the membrane fraction was increased following FcεRI ligation. Alternatively, it is also possible that the formation of trimolecular complex causes the structural change of PLCγ, which would facilitate the exposure of the phosphorylation sites of PLCγ.
When BMMC are stimulated with high concentration of Ag, the initial spike is evoked but a sustained Ca2+ plateau is impaired in me-BMMC (Fig. 6,A), suggesting that there is additional defect of Ca2+ mobilization in me-BMMC, extracellular Ca2+ influx. The initial and later phases of Ca2+ responses are reflected by the rapid release of ER Ca2+ store and extracellular Ca2+ entry, respectively. It is believed that the second stage of Ca2+ response is initiated as a result of depletion of endoplasmic Ca2+ stores, which activates store-operated Ca2+ channels such as Ca2+ release-activated Ca2+ (CRAC) channels (59, 60). Although little is known about how CRAC channels are activated, recent identification of two important components, Orai1 (61) and stromal interaction molecule 1 (STIM1) (62, 63), has shed light on the molecular mechanisms of Ca2+ entry through CRAC channels. Although Orai1 is an essential component of the CRAC channel complex, STIM1 is diffusely dispersed throughout the ER membrane and possesses Ca2+-binding motif that may sense the stores. Once the ER Ca2+ stores are depleted, STIM1 becomes redistributed into discrete spots under the plasma membrane and activates CRAC channels. The importance of STIM1 in mast cell function has been demonstrated that STIM1-deficient fetal liver-derived mast cells have impaired FcεRI-induced Ca2+ influx, degranulation, and cytokine production (64). In addition, Orai1 colocalizes with STIM1 after store depletion and both proteins move in coordinated fashion to form closely apposed clusters in the ER and plasma membrane, thereby creating the elementary unit of store-operated Ca2+ entry (65). There might be at least two possible explanations as to why Ca2+ influx is impaired in me-BMMC. One possibility is that lack of SHP-1 causes dysfunction of STIM1/Orai1. In our study, almost comparable increase in intracellular Ca2+ concentration was observed in WT- and me-BMMC when cells were treated with TG that is able to activate CRAC channel by depleting ER Ca2+ stores (Fig. 6,B). This result is in strong contrast to the previous study (64) in which lack of STIM1 completely abrogates Ca2+ influx following the treatment with TG or ionomycin, implying the relatively intact function of STIM1/Orai1 in me-derived BMMC. Because TG-induced Ca2+ response is slightly decreased in me-BMMC (Fig. 6,B) and the internal Ca2+ elevation induced by the stimulation with Ag or TG is comparable between me- and WT-BMMC (Fig. 6,C), we cannot rule out the possibility that SHP-1 exerts some effects on the STIM1/Orai1-dependent activation of Ca2+ channels such as the function of STIM1, the process of STIM1 redistribution, or puncta formation of STIM1 underneath the plasma membrane. The second possibility is that lack of SHP-1 expression might lead to insufficient Ca2+ store depletion following Ag stimulation. This possibility is supported by our finding that both me- and WT-BMMC showed similar Ca2+ mobilization when cells were stimulated with TG to completely deplete Ca2+ store in the presence of extracellular Ca2+ (Fig. 6,B). In addition, TG- or Ag-induced increase in intracellular Ca2+ concentration in the absence of extracellular Ca2+ is identical between WT- and me-BMMC (Fig. 6 C). Although these results appear to indicate that there is little difference in the Ca2+ release from the ER between two groups, it should be considered that the increase in Ca2+ concentration is also mediated by the release from other intracellular Ca2+ stores such as mitochondria. Apart from the major role of mitochondria as cellular energy sources and regulator for cell death, recent studies have placed mitochondria as a Ca2+ regulator (60). To discriminate the intracellular Ca2+ store contributing to the increase of Ca2+ concentration following the stimulation with Ag or TG, it may be required to monitor the Ca2+ concentration in ER as well as in mitochondria simultaneously. Because mitochondria have been shown to reduce Ca2+-dependent inactivation of CRAC channels by quickly removing the inflowing Ca2+ (66, 67), determination of mitochondrial Ca2+ concentration might provide further evidences on the regulation Ca2+ influx thorough CRAC channel.
Although the importance of Ca2+ flux in mast cell degranulation has been widely accepted (47, 68), enhanced or normal degranulation was detected in Lyn-deficient mast cells despite defective Ca2+ mobilization (48, 49, 50, 51). However, the Ca2+ response in Lyn-deficient mast cells was impaired at early transient stage of the response and was almost intact at the later prolonged phase (48, 50, 51) with one exception in which Lyn deficiency caused a marked decrease in overall Ca2+ response (49). It should be noted that Stim1-deficient mast cells showed severely impaired sustained Ca2+ increase and degranulation (64). Consistent with these published evidences, me-BMMC had a decrease in later, sustained Ca2+ response and degranulation when cells were stimulated with high dose of Ag (50 ng/ml) (Fig. 6, A and D). Furthermore, treatment of the cells with EGTA almost completely abolished extracellular Ca2+ influx and degranulation in WT- and me-BMMC. All these observations suggest a prolonged Ca2+ increase is required for degranulation. However, stimulation of me-BMMC with 5 ng/ml DNP-HSA elicited almost no Ca2+ increase but still evident degranulation (Fig. 6, A and D). Taken together with exceptional Lyn-deficient mast cells that had severely reduced Ca2+ response but normal degranulation (49), it is possible that slight Ca2+ increase in specialized region, such as fusing sites between plasma membrane and granule, might be sufficient to induce degranulatioin.
In summary, to elucidate the regulatory mechanisms of SHP-1 in mast cell signaling, we identified substrates for SHP-1 by using BMMC established from me and WT mice and analyzed downstream signaling events induced by FcεRI ligation. Tyrosine phosphorylation of SLP-76 and LAT was increased in me-BMMC. Combined with the almost identical activity of Syk in both BMMC and the previous observations that SHP-1 directly dephosphorylates SLP-76 and LAT in other cell types, we concluded that SLP-76 and LAT serve as substrates for SHP-1 in mast cells. In accordance with hyperphosphorylation of both molecules, the activation of MAPKs and transcription of cytokine genes are enhanced in me-BMMC. In contrast, FcεRI-induced Ca2+ mobilization was severely impaired in me-BMMC, especially at the later plateau phase when cells were stimulated with high dose of Ag. Upon stimulation with lower concentration of Ag, Ca2+ response was abrogated in me-BMMC. In the absence of SHP-1, PLCγ failed to associate with SLP-76, which might result in the impairment of Ca2+ responses and degranulation. These results suggest a novel molecular mechanism leading to Ca2+ mobilization independent of LAT, and provide evidence that SHP-1 can positively regulate signaling events by acting as a linker/adaptor molecule in mast cells.
The authors have no financial conflict of interest.
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 U.S.C. Section 1734 solely to indicate this fact.
This work was supported in part by grants-in-aid from the Japanese Ministry of Education, Culture, Sports, Science and Technology and by a grant-in-aid High-Tech Research Center Project (2002–2006) for Private Universities matching fund subsidy from the Ministry of Education, Culture, Sports, Science, and Technology of Japan.
Abbreviations used in this paper: FcεRI, high affinity IgE receptor; PTK, protein tyrosine kinase; SH2, Src homology region 2; PLCγ, phospholipase Cγ; SLP-76, SH2 domain-containing leukocyte protein of 76 kDa; LAT, linker for activation of T cell; PTP, protein tyrosine phosphatase; SHP-1, SH2 domain-containing phosphatase 1; me, motheaten; BMMC, bone marrow-derived mast cells; WT, wild type; PY, phosphotyrosine; HSA, human serum albumin; TCL, total cell lysates; TG, thapsigargin; CRAC, Ca2+ release-activated Ca2+; STIM1, stromal interacting molecule 1; ER, endoplasmic reticulum.