|Home | About | Journals | Submit | Contact Us | Français|
We developed a confocal real-time imaging approach that allows direct observation of the subcellular localization pattern of proteins involved in proximal FcRI signaling in RBL cells and primary bone marrow-derived mast cells. The adaptor protein Src homology 2 (SH2) domain-containing leukocyte phosphoprotein of 76 kDa (SLP-76) is critical for FcRI-induced calcium flux, degranulation, and cytokine secretion. In this study, we imaged SLP-76 and found it in the cytosol of unstimulated cells. Upon FcRI cross-linking, SLP-76 translocates to the cell membrane, forming clusters that colocalize with the FcRI, the tyrosine kinase Syk, the adaptor LAT, and phosphotyrosine. The disruption of the SLP-76 interaction with its constitutive binding partner, Gads, through the mutation of SLP-76 or the expression of the Gads-binding region of SLP-76, inhibits the translocation and clustering of SLP-76, suggesting that the interaction of SLP-76 with Gads is critical for appropriate subcellular localization of SLP-76. We further demonstrated that the expression of the Gads-binding region of SLP-76 in bone marrow-derived mast cells inhibits FcRI-induced calcium flux, degranulation, and cytokine secretion. These studies revealed, for the first time, that SLP-76 forms signaling clusters following FcRI stimulation and demonstrated that the Gads-binding region of SLP-76 regulates clustering of SLP-76 and FcRI-induced mast cell responses.
Mast cells are tissue resident hematopoietic cells that express the high-affinity immunoglobulin E (IgE) receptor FcRI. This receptor contains immunoreceptor tyrosine-based activation motifs (ITAMs) that become phosphorylated by Src family kinases following cross-linking by IgE and cognate antigen. Syk kinase, which binds the phosphorylated ITAMs, then becomes activated and phosphorylates a number of critical signaling proteins, including the transmembrane adaptor linker for the activation of T cells (LAT), the guanine nucleotide exchange factor Vav, and the adaptor molecule Src homology 2 (SH2) domain-containing leukocyte phosphoprotein of 76 kDa (SLP-76). These proximal events lead to the activation of phospholipase Cγ1 (PLCγ1) and phospholipase Cγ2 (PLCγ2), which cleaves the membrane phospholipid phosphatidylinositol 4,5-bisphosphate into inositol 1,4,5-trisphosphate and diacylglycerol, inducing calcium flux and protein kinase C activation, respectively. This signaling cascade regulates diverse mast cell responses, including granule release, cytokine secretion, cytoskeletal rearrangement, and gene transcription (for reviews, see references 13, 26, and 31).
SLP-76 is a hematopoietic-cell-specific adaptor protein that functions downstream of both ITAM-containing receptors and integrins (44). Loss-of-function studies have demonstrated a critical role for SLP-76 in FcRI signaling. SLP-76-deficient bone marrow-derived mast cells (BMMCs) fail to flux calcium, degranulate, or secrete interleukin-6 (IL-6) in response to FcRI stimulation. In addition, SLP-76-deficient mice are resistant to FcRI-induced passive systemic anaphylaxis (PSA) (16, 30, 43).
SLP-76 contains three N-terminal tyrosines that inducibly bind the guanine nucleotide exchange factor Vav (16, 27, 36, 42), the adaptor Nck (45), and the Tec family kinase Btk (16); a P1 region that binds the phospholipase PLCγ (46); a C-terminal SH2 domain that inducibly binds the adaptor adhesion- and degranulation-promoting adaptor protein (10, 24) and the hematopoietic progenitor kinase 1 (33); and a proline-rich region that constitutively binds the adaptor molecule Gads (1, 21). Mutational analysis has demonstrated that the N-terminal tyrosines and the proline-rich region are absolutely required, while the SH2 domain is partially required for FcRI-induced mast cell responses (16, 43).
These findings suggest an important role for a number of signaling proteins, including the cytosolic adaptor Gads, a member of the Grb2 family of proteins. Gads is expressed specifically in hematopoietic cells and contains a central SH3 domain flanked by two SH2 domains (20). Gads constitutively binds through its SH3 domain to the RxxK motif in the proline-rich region of SLP-76, and Gads inducibly associates through its SH2 domain with phosphorylated LAT (3, 21). Mice deficient in SLP-76 or LAT have profound defects in thymocyte development (8, 48), while Gads-deficient mice have a partial block in thymocyte development that phenocopies the defects seen in mice expressing a Gads-binding mutant of SLP-76 (18, 25, 47). These studies suggest that Gads mediates the formation of a signaling complex between SLP-76 and LAT that is critical for T-cell-receptor signaling and thymocyte development.
The formation of multimolecular signaling complexes has been proposed to regulate FcRI signaling, leading to diverse mast cell responses. Previous studies of RBL cells using immunoelectron microscopy indicated that FcRI cross-linking induces the formation of a “primary signaling domain” that includes FcRI, Syk, PLCγ2, Gab2, and the p85 regulatory subunit of phosphatidylinositol-3 kinase and a “second signaling domain” that includes LAT, PLCγ2, and p85 (39-41). These studies offered high-resolution “snapshots” of signaling complexes but did not allow direct observation of the kinetics of complex formation. While these experiments have described the spatial relationships of a number of key FcRI signaling molecules, the localization of SLP-76 was not addressed. We have therefore developed a real-time confocal imaging approach that facilitates investigation of the temporal as well as spatial regulation of proximal FcRI signaling.
The mechanisms that regulate the formation of these multimolecular signaling complexes remain poorly understood. Adaptor molecules have been proposed to play a critical role in the formation of signaling complexes in numerous cell types. We therefore investigated the subcellular localization of the adaptor protein SLP-76 in real time in the RBL cell line and in primary mast cells. Our results demonstrated that SLP-76 translocates from the cytosol to the cell membrane, where it forms signaling clusters that colocalize with the FcRI, LAT, Syk, and phosphotyrosine-containing proteins. Further, we showed that formation of the dynamic signaling clusters requires interaction between SLP-76 and Gads. Disruption of this interaction blocks SLP-76 translocation to the membrane, abrogates formation of signaling clusters, and inhibits FcRI-induced mast cell responses.
RBL cells were cultured in RPMI medium, 15% fetal calf serum, 100 U/ml penicillin, 100 μg/ml streptomycin, and 2.92 mg/ml glutamine. Bone marrow cells were cultured in RPMI medium, 20% fetal calf serum, 100 U/ml penicillin, 100 μg/ml streptomycin, 2.92 mg/ml glutamine, 25 mM HEPES, 0.1 mM nonessential amino acids, 1 mM sodium pyruvate, and 50 μg/ml gentamicin (complete RPMI medium) supplemented with 20 ng/ml of recombinant IL-3 and 20 ng/ml of recombinant stem cell factor (SCF) (Peprotech). After 4 weeks of culture, >90% of cells were mast cells, as determined by flow cytometric analysis for the expression of FcRI. Functional assays were performed on cells that had been in culture for 4 to 8 weeks.
Syk-green fluorescent protein (GFP) chimera was kindly provided by Robert Geahlen (Purdue University) and has been described previously (22). LAT-GFP chimera was kindly provided by Larry Samelson (NIH) and has been described previously (5). The monomeric red fluorescent protein (mRFP) vector was kindly provided by Roger Tsien (UCSD). It was PCR amplified from pRSET-mRFP1 (7) by using the primers 5′-CCT TAA GCC ACC ATG GCC TCC TCC GAG GAC GTC AT-3′ and 5′-CCC AAG CTT GGC GCC GGT GGA GTG GCG GC-3′. The PCR product was digested with EcoRI, blunted with Klenow fragment, and then digested with HindIII. The digested product was placed upstream of the multiple cloning site of pEGFP-C1 (Clontech) after the removal of enhanced GFP by digestion with AgeI, blunting, and digestion of HindIII. RFP-SLP-76 chimera was generated by PCR amplifying SLP-76 from GFP-SLP-76 (35) and adding EcoRI and Xma linkers with primers 5′-GGA ATT CCC ATG GCC TTG AAG AAT G-3′ and 5′-TTT CCC GGG CTA CAG ACA GCC TGC AGC-3′. SLP-76 was then digested with EcoRI and XmaI and was cloned in frame with mRFP. This construct was confirmed by sequencing. The SLP-76 G2 mutant and the GBF-DsRed2 and DsRed2 murine stem-cell virus (MSCV)-based vectors have been described previously (35). High-titer retroviral supernatants were produced via cotransfection of 293-T cells with retroviral construct and helper virus packaging construct (Imgenex).
Freshly isolated bone marrow from B6/129 mice was cultured overnight in complete RPMI medium supplemented with 10 ng/ml IL-3, 10 ng/ml IL-6 (R&D Systems), and 50 ng/ml SCF. Retroviral supernatant was then added, and cells were spin-infected at 2,500 rpm for 90 min at room temperature in the presence of 8 μg/ml of Polybrene (Sigma). Cells were again incubated overnight at 37°C with 5% CO2, and the retroviral spin infection was repeated the following day. Following overnight incubation, cells were cultured to generate mast cells. After 3 to 4 weeks, DsRed2-expressing cells were sorted using a FACSVantage SE flow cytometer (Beckton Dickinson).
Cells were stained according to standard protocols by using the following labeled antibodies: mouse anti-2,4-dinitrophenol (DNP) IgE (Sigma), anti-mouse IgE-biotin (Pharmingen), and streptavidin-APC (Pharmingen). Two-color flow cytometry was performed with a FACSCalibur (Becton Dickinson).
A total of 5 × 105 RBL cells per well were sensitized overnight at 37°C with 1 μg/ml anti-DNP IgE in a 96-well, flat-bottom tissue culture plate. Cells were then washed twice with Tyrode's buffer (130 mM NaCl, 10 mM HEPES, 1 mM MgCl2, 5 mM KCl, 1.4 mM CaCl2, 5.6 mM glucose, 1 mg/ml bovine serum albumin, pH 7.4) and then stimulated with DNP-human serum albumin (HSA) or phorbol myristate acetate (PMA) and ionomycin for 60 min at 37°C. Total cellular granule contents were released by lysing cells with Tyrode's buffer with 1% Triton X-100. A total of 30 μl of supernatant was collected and transferred to a 96-well, flat-bottom plate. A total of 30 μl of 1 mM p-nitrophenyl-N-acetyl-β-d-glucosamide was then added to each supernatant and mixed before incubation for 1 h at 37°C. The reaction was terminated by the addition of 200 μl of 0.1 M Na2CO3-NaHCO3 buffer, and the optical density was read on a plate reader at a wavelength of 405 nm. BMMCs (1 × 106/ml) were starved of SCF overnight and then sensitized at 1 × 107/ml in complete RPMI medium without cytokines and with 1 μg/ml anti-DNP IgE (clone SPE-7; Sigma) for 4 h at 37°C with 5% CO2. Cells were then washed once in Tyrode's buffer and resuspended at 2 × 106/ml in the buffer. A total of 200 μl of cells was then stimulated with various amounts of DNP (0 to 1,000 ng/ml) for 1 h at 37°C. Cells were spun down, and 30 μl of supernatant was transferred to a 96-well, flat-bottom plate. A total of 30 μl of 1 mM p-nitrophenyl-N-acetyl-β-d-glucosamide was then added to each supernatant and mixed before incubation for 1 h at 37°C. The reaction was terminated by the addition of 200 μl of 0.1 M Na2CO3-NaHCO3 buffer, and the optical density was read on a plate reader at a wavelength of 405 nm.
BMMCs (1 × 106/ml) were starved of SCF overnight and then sensitized at 1 × 107/ml in complete RPMI medium without cytokines and with 1 μg/ml anti-DNP IgE (clone SPE-7, Sigma) for 4 h at 37°C with 5% CO2. Cells were then washed once and resuspended at 1 × 106/ml in complete RPMI medium. A total of 5 × 104 cells in complete RPMI medium was then incubated with various concentrations of DNP overnight at 37°C plus 5% CO2 in a total volume of 100 μl in a 96-well, flat-bottom plate. Each sample was assayed in triplicate. The following day, the plate was removed from the incubator and frozen at −20°C. An enzyme-linked immunosorbent assay (ELISA) was performed on thawed supernatants by using a murine IL-6 ELISA kit (Pierce/Endogen).
RBL cells and BMMCs were washed, pelleted, and lysed in ice-cold 1% NP-40 containing proteinase (50 μg/ml aprotinin, 10 μg/ml leupeptin, 50 μg/ml pepstatin A, and 1 mM Pefablock) inhibitors. Western blotting was performed using anti-DsRed2 (Clontech).
RBL cells were sensitized overnight with 1 μg/ml anti-DNP IgE. RBL cells were then incubated with Versene (sodium chloride, 8 g/liter; potassium chloride, 0.2 g/liter; dibasic sodium phosphate, 1.15 g/liter; EDTA, 0.2 g/liter; and phenol red, 0.1 g/liter; at pH 7.34) for 2 min and detached from tissue culture dishes and washed once in RBL medium. BMMCs were sensitized with 1 μg/ml anti-DNP IgE for 4 h at 37°C. Cells were then washed once in Tyrode's buffer and resuspended at 1 × 107 cells/ml in Tyrode's buffer containing 6.25 mM Probenecid (Sigma) and 2 mg/ml Indo-1 (Molecular Probes). Cells were protected from light and incubated at 37°C for 30 min. Indo-1-loaded cells were washed twice and resuspended in warm Tyrode's buffer. Data were collected using an LSR flow cytometer (Becton Dickinson). Baseline Ca2+ levels were measured for 30 s prior to the addition of DNP (100 ng/ml). The sample was collected for a total of 260 s with approximately 500 to 700 events collected per second. Ionomycin was added 35 s prior to the end of the assay. FL5 represents Indo-1 bound by Ca2+, and FL4 represents Indo-1 not bound by Ca2+. Data were analyzed using FlowJo software (TreeStar), and Ca2+ flux is defined by the mean ratio of FL5/FL4 over time.
RBL cells were detached with Versene and washed once in RBL medium. The cells were then incubated with 10 μg of plasmid DNA for 10 min and electroporated (900 μF/ 250 V) in Gene Pulser cuvettes (Bio-Rad). Cells were then rested for 10 min before plating in RBL medium. Transiently transfected cells were assayed 18 to 72 h after transfection. Stable cell lines were generated by selection in G418 (0.1 mg/ml; Mediatech) followed by fluorescence-activated cell sorting (FACS) for equal levels of GFP of DsRed2.
The dynamics of signaling complexes in mast cells were monitored using an adaptation of a T-cell spreading assay (6). Delta T microscope dishes (Bioptechs) were incubated overnight with 1 ml 0.01% poly-l-lysine (Sigma) at 37°C. Poly-l-lysine was aspirated, and plates were dried at 37°C for 10 min. The microscope dishes were then incubated with 1 ml of 1 μg/ml DNP-HSA in carbonate-binding buffer (35 mM sodium bicarbonate and 75 mM sodium carbonate) for 1 h at 37°C. DNP-HSA was then aspirated, and the microscope dishes were dried for 10 min at 37°C. RBL cells were prepared by detaching them from tissue culture plates with Versene, washing them in RBL medium, and sensitizing them with 1 μg/ml anti-DNP IgE for 1 h in RBL medium at 37°C. BMMCs were prepared by sensitizing them in cytokine-free mast cell media with 1 μg/ml anti-DNP IgE for 1 h at 37°C. Cells were then washed and resuspended in RBL medium or cytokine-free mast cell medium. To initiate an assay, RBL cells or BMMCs were dropped onto a microscope dish coated with poly-l-lysine alone or poly-l-lysine and DNP-HSA. As the cells initiated contact with the antigen-coated coverslip, multiple z-axis images were collected starting from just below the plane of interaction of the cell and the coverslip. Images were collected for the indicated times. Live-cell imaging was performed on a Perkin-Elmer 5 wavelength laser UltraVIEW LCI spinning disk confocal microscope (Yokogawa) that was attached to a Nikon TE-300 inverted microscope equipped with a 60× objective, a z-axis servo controller (Physik Instrumente), a Delta T4 culture dish controller (Bioptechs), and an objective heater (Bioptechs) to maintain the sample temperature at 37°C. Samples were excited, and emission wavelengths were collected using an argon ion laser emitting 488- and 514-nm lines and an argon-krypton laser emitting 568- and 647-nm lines in conjunction with a 488/568 RGB dichroic mirror. A Hamamatsu Orca-ER charge-coupled-device camera (Hamamatsu) was used to collect images. Image analysis was performed using IP Labs version 3.9.3 r4.
For fixed images, the RBL cells, BMMCs, and delta T microscope dishes were prepared as described for the live images. Cells were added to the microscope dishes and incubated at 37°C for the indicated times. The dishes were then aspirated and washed in phosphate-buffered saline to remove medium and nonadherent cells. The cells were fixed in 4% paraformaldehyde for 20 min, quenched in 50 mM ammonium chloride for 3 min, permeabilized in 0.5% Triton X-100 for 1 min, and blocked (phosphate-buffered saline, 0.01% saponin, 0.25% gelatin, and 0.02% sodium azide) overnight. Rhodamine-conjugated PY20 antiphosphotyrosine antibody was used for immunofluorescence staining. Multiple z-axis images were acquired and then analyzed using IP Labs software to enhance contrast and to merge color channels. Representative images were chosen for the figures.
Cells were counted as containing GFP-SLP-76 clusters if distinct areas of GFP fluorescence above the background were observed specifically at the plane of interaction between the cell and the coverslip. More than 75 cells from at least 10 different high-powered fields were analyzed from each group. Cells were scored for clustering by a blinded observer. The percentages of cells with GFP-SLP-76 clustering were compared between GBF-DsRed2 and DsRed2 vector by Student's t test.
GBF-antennapedia peptide (RQIKIFQNRRMKKKNHSPLSPPHPNHEEPSRSGNNKTAKLPAPSIDRSTKPPLDRSLAPLDREPF) was synthesized by the W. M. Keck Biotechnology Resource Center (New Haven, CT). The peptide was initially dissolved in dimethyl sulfoxide and then diluted to the appropriate concentration in water.
To investigate SLP-76 subcellular localization following FcRI cross-linking, stable RBL cell lines expressing GFP alone or GFP fused to SLP-76 were generated by transfection followed by FACS for equal expression of GFP. RBL cells were preincubated with IgE specific for DNP-HSA and then were dropped onto a coverslip coated with poly-l-lysine (PLL) as a negative control or with DNP-HSA. As the cells contacted the coverslip, the antigen DNP-HSA cross-linked the FcRI receptor and initiated signaling. Live confocal visualization at the plane of interaction between the cell and the stimulatory coverslip allowed direct observation of proximal FcRI signaling events in real time. These images were collected and analyzed with high temporal and spatial resolution.
To determine the pattern of SLP-76 localization in response to FcRI activation, we compared the subcellular localization of GFP-SLP-76 on PLL-coated versus that on DNP-HSA-coated coverslips. In RBL cells interacting with the PLL surface, GFP fluorescence remained diffusely cytoplasmic (Fig. (Fig.1A)1A) (see Movie S1 in the supplemental material), whereas a dramatic clustering of GFP fluorescence was observed when the cells came in contact with DNP-HSA-coated coverslips (Fig. (Fig.1B)1B) (see Movie S2 in the supplemental material). GFP-SLP-76 formed numerous clusters, which remained for 15 to 30 min, at the cell membrane as early as 15 s after contacting the stimulatory surface (data not shown) (see Movie S2 in the supplemental material). Analysis of multiple confocal images demonstrated that these clusters were found at only the interface between the cell and the coverslip and that SLP-76 clustering was not observed in the interior of the cell or at the plasma membrane in areas not interacting with the coverslip (data not shown). In contrast, fluorescence in RBL cells expressing GFP not fused to SLP-76 remained diffusely cytoplasmic and nuclear when cells contacted PLL- or DNP-HSA-coated coverslips (data not shown). These results demonstrated that FcRI cross-linking leads to rapid and sustained clustering of SLP-76 at the antigen contact site.
Cross-linking of the FcRI is also known to induce clustering of the receptor (39). To investigate whether the SLP-76 clusters colocalized with FcRI, GFP-SLP-76 RBL cells were incubated with Alexa 568-conjugated IgE and then were dropped onto PLL- or DNP-HSA-coated coverslips. FcRI formed distinct clusters at the cell contact sites with the DNP-HSA-coated coverslip within seconds, whereas FcRI remained diffusely distributed as the cells spread upon PLL-coated coverslips. In the stimulated RBL cells, some smaller clusters of IgE could also be seen inside the cells, which likely represented endocytosis of the receptor and bound IgE (29); however, the majority of the FcRI clusters were seen at the plane of interface between the cell and the coverslip. These FcRI clusters colocalized with the SLP-76 clusters as shown by the merged fluorescence images (Fig. (Fig.2A).2A). These experiments indicated that FcRI stimulation induces SLP-76 translocation from the cytoplasm to the cell membrane, forming clusters that colocalize with the FcRI.
Following FcRI cross-linking, Syk kinase becomes activated and phosphorylates SLP-76. We therefore investigated whether Syk colocalized with SLP-76 clusters. RBL cells stably expressing mRFP fused to SLP-76 were generated and transiently transfected with Syk-GFP. Both mRFP-SLP-76 and Syk-GFP formed clusters when the anti-DNP-HSA IgE-sensitized RBL cells contacted the DNP-HSA-coated surface (Fig. (Fig.2B).2B). mRFP-SLP-76 formed clusters with kinetics and distributions similar to those of GFP-SLP-76 (data not shown). In contrast, mRFP and GFP fluorescence remained diffusely cytoplasmic when the RBL cells interacted with the PLL-coated coverslip. Merging of the mRFP and GFP images demonstrated that SLP-76 clusters and Syk clusters colocalize (Fig. (Fig.2B2B).
We next investigated whether SLP-76 clusters also colocalize with the membrane adaptor LAT. mRFP-SLP-76-expressing RBL cells were transiently transfected with LAT-GFP and were sensitized with DNP-HSA-specific IgE. When the RBL cells interacted with a DNP-HSA-coated coverslip, both SLP-76 and LAT formed distinct clusters at the cell contact sites (Fig. (Fig.2C).2C). These clusters were found specifically at the interface between the cell and the coverslip, whereas clustering was not observed in the cytoplasm or on other cell membranes, suggesting specificity for this interaction. In unstimulated cells, LAT was evenly distributed at the periphery of the cells, as expected for a transmembrane protein, while SLP-76 was diffusely distributed in the cytoplasm. Merging of the RFP and GFP images demonstrated colocalization of the SLP-76 and LAT clusters (Fig. (Fig.2C2C).
We hypothesize that SLP-76 functions by regulating the formation of multimolecular signaling complexes that contain activated signaling proteins. Since tyrosine phosphorylation of multiple signaling intermediates is required for a productive signal, we investigated whether SLP-76 clusters would colocalize with phosphotyrosine-containing proteins at the membrane. RBL GFP-SLP-76 cells were dropped onto PLL- or DNP-HSA-coated coverslips, fixed, and stained with rhodamine-conjugated antiphosphotyrosine. RBL cells adhered to and spread upon the PLL-coated coverslips but did not exhibit phosphotyrosine staining, while cells adhering and spreading on DNP-HSA-coated coverslips exhibited dramatic phosphotyrosine clusters (Fig. (Fig.2D).2D). These clusters were seen within 2 min (the earliest time point analyzed) of dropping the cells on the coverslip and remained for >30 min (data not shown). Importantly, phosphotyrosine staining was seen primarily in the plane of interaction between the cell and the stimulatory surface. Very little phosphotyrosine staining was seen in other regions of the cell. SLP-76 colocalized with phosphotyrosine-containing proteins, as seen with the merged fluorescence images (Fig. (Fig.2D2D).
While the RBL cell line is a powerful model system for investigating FcRI signaling, it is important, whenever possible, to also examine events in primary cells. Accordingly, we investigated the subcellular localization pattern of SLP-76 in BMMCs. SLP-76-deficient BMMCs expressing GFP-SLP-76 were generated by infecting SLP-76-deficient bone marrow with the MSCV-based retrovirus containing GFP fused to SLP-76, followed by growth in IL-3- and SCF-containing media for 4 weeks to allow for differentiation and proliferation. Anti-DNP IgE-sensitized BMMCs were dropped on to coverslips coated with PLL or DNP-HSA. BMMCs adhered to PLL-coated coverslips in the absence of stimulation, but DNP-HSA stimulation increased the adhesion and spreading of BMMCs on the coverslips, likely via inside-out up-regulation of integrin function. SLP-76 was diffusely cytoplasmic when BMMCs were dropped on PLL alone but formed distinct clusters upon contact with DNP-HSA-coated coverslips. In addition, these SLP-76 clusters colocalized with phosphotyrosine staining, suggesting that they represent complexes enriched for proteins important for signaling (Fig. (Fig.33).
The cytosolic adaptor protein Gads constitutively binds SLP-76, and this interaction is mediated by the SH3 region of Gads and the RxxK motif found in the proline-rich region of SLP-76. Previous studies of BMMCs by our group and others demonstrated that the SLP-76/Gads interaction is required for FcRI-induced calcium flux, degranulation, and cytokine production (16, 43). An arginine-to-alanine substitution at amino acids 237 and 240 (G2) abrogates SLP-76 binding with Gads (3). Expression of this minimal mutation of GFP-SLP-76 was unable to rescue FcRI-induced calcium flux in SLP-76-deficent BMMCs (Fig. (Fig.4A).4A). To correlate SLP-76 trafficking and competence of FcRI-induced signaling, we investigated whether the Gads-binding region of SLP-76 is required for SLP-76 to translocate and cluster at the cell membrane. We generated a stable RBL cell line expressing GFP fused to SLP-76 containing the G2 mutation. Confocal imaging of RBL GFP-SLP-76 G2 cells dropped on to DNP-HSA-coated coverslips showed that SLP-76 clustering was dramatically impaired (Fig. (Fig.4B).4B). These data suggested that SLP-76 clustering requires its binding with Gads and, by extension, that proper subcellular localization of SLP-76 is likely required for FcRI-induced mast cell responses.
To further investigate the role of the SLP-76/Gads interaction in FcRI-induced signaling and mast cell responses, we chose a complementary approach to disrupt the SLP-76/Gads interaction without altering the structure of the SLP-76 protein (Fig. (Fig.5A).5A). Previous studies of Jurkat T cells demonstrated that the expression of 50 amino acids that correspond to the Gads-binding region of SLP-76 fused to the fluorophore DsRed2 disrupted the SLP-76/Gads interaction and also inhibited T-cell-receptor-induced SLP-76 clustering, calcium flux, and CD69 upregulation (35). We hypothesized that this fragment (called the Gads-binding fragment [GBF] of SLP-76) would inhibit the dynamic clustering of SLP-76 and disrupt FcRI signaling and mast cell responses. To test whether GBF expression would block clustering of SLP-76 in mast cells, GBF-DsRed2 or DsRed2 alone as a control was transiently transfected into GFP-SLP-76-expressing RBL cells. These cells were then sorted for coexpression of both GFP and DsRed2 and then visualized using live video confocal microscopy as they interacted with antigen-coated coverslips (Fig. (Fig.5B).5B). The collected images were analyzed and visually scored by a blinded observer for clustering of GFP-SLP-76. While SLP-76 clustered in ~80% of cells expressing DsRed2 alone, SLP-76 clustering was observed in only 40% of cells expressing the GBF (Fig. (Fig.5C).5C). Two representative images of cells expressing DsRed2 or GBF-DsRed2 are shown. Further, GFP-SLP-76 clusters observed in GBF-expressing RBL cells were typically smaller than those observed in RBL cells expressing DsRed2 only (Fig. (Fig.5C).5C). These data indicated that the GBF disrupts appropriate subcellular localization of SLP-76 in mast cells.
We next asked whether inhibition of SLP-76 dynamic clustering would disrupt FcRI signaling. To address this question, RBL cell lines stably expressing DsRed2 alone (vector) or GBF fused with DsRed2 were established. The GBF-DsRed2 and vector cell lines expressed equal levels of DsRed2 fluorescence and FcRI receptor (Fig. 6A and B). GBF-DsRed2 and vector control RBL cells were assayed for FcRI-induced calcium mobilization by flow cytometry using the calcium-sensitive dye Indo-1. Calcium flux was diminished in GBF-DsRed2-expressing RBL cells compared to that for vector control RBL cells, suggesting a partial disruption in FcRI-mediated signaling (Fig. (Fig.7A).7A). Importantly, ionomycin induced similar levels of calcium flux in GBF-DsRed2-expressing cells and vector control RBL cells, indicating that cells expressing the GBF are still capable of mobilizing calcium (data not shown).
Since degranulation is a SLP-76-dependent FcRI-induced mast cell response, we investigated whether the GBF could also inhibit mast cell degranulation. GBF-DsRed2-expressing RBL cells exhibited significantly less degranulation than control DsRed2-expressing RBL cells (Fig. (Fig.7B).7B). Importantly, PMA and ionomycin induced degranulation in both GBF-DsRed2 and vector control RBL cells, indicating that the degranulation machinery is intact in both cell lines.
Expression of the 50-amino-acid GBF fused to the 248-amino-acid DsRed2 inhibited FcRI signal transduction and mast cell responses. To investigate whether the GBF could inhibit mast cell responses using a protein transduction system instead of a genetic expression system, the GBF was fused to a 17-amino-acid sequence of the antennapedia peptide, which facilitates protein transportation into cells (11, 23) (Fig. (Fig.7C).7C). To test the effect of this peptide on FcRI signaling, RBL cells were incubated with the GBF-antennapedia peptide or vehicle alone and then assessed for FcRI-induced calcium flux. RBL cells treated with the GBF-antennapedia peptide exhibited diminished calcium flux (Fig. (Fig.7D).7D). Further, this inhibition increased with the time of incubation, with inhibition of calcium flux first observed after 30 min of exposure to GBF peptide (data not shown). Washout experiments demonstrated that the effect of GBF peptide disappears by 24 h (data not shown). Although there is a small amount of cell death associated with peptide administration (generally less than 10%), these cells respond completely normally to ionomycin treatment with a robust calcium response (Fig. (Fig.7D7D).
GBF efficiently blocks FcRI signaling and mast cell responses in the RBL cell line. To extend these findings to primary mast cells, wild-type bone marrow cells were infected with a retroviral MSCV-based vector containing either the GBF-DsRed2 or DsRed2 alone and differentiated into BMMCs in vitro. GBF-DsRed2- or DsRed2-expressing BMMCs had similar levels of DsRed2 protein and expressed equal levels of FcRI (Fig. 8A and B). Expression of the GBF inhibits FcRI-induced calcium flux in a dose-dependent fashion, while the vector alone had no effect on calcium flux (Fig. (Fig.9A).9A). Ionomycin elicited similar responses from both GBF- and vector-expressing mast cells (data not shown).
To test the ability of the GBF to block degranulation, BMMCs expressing GBF or vector alone were sorted for DsRed2 expression. FcRI-induced degranulation was assessed by hexosaminidase release. The GBF inhibited degranulation relative to the vector, but PMA-ionomycin elicited similar levels of degranulation in both the GBF and vector control cells (Fig. (Fig.9B).9B). Cytokine secretion is also a SLP-76-dependent function of mast cells. We investigated whether the GBF could inhibit FcRI-induced IL-6 production. BMMCs expressing either GBF or vector were stimulated for 24 h by cross-linking the FcRI receptor with antigen. The GBF efficiently inhibited IL-6 secretion relative to vector-expressing cells (Fig. (Fig.9C),9C), demonstrating that the GBF efficiently blocks downstream signaling events following FcRI engagement in primary mast cells. Importantly, PMA and ionomycin induced similar levels of IL-6 secretion in GBF- and vector-expressing cells (Fig. (Fig.9C9C).
We investigated the temporal and spatial regulation of FcRI signaling by characterizing the subcellular localization pattern of the receptor, SLP-76, LAT, Syk, and phosphotyrosine-containing proteins. These imaging studies demonstrated that FcRI cross-linking induces SLP-76 translocation from the cytosol to the cell membrane, where it forms clusters that colocalize with the FcRI, the tyrosine kinase Syk, the membrane adaptor LAT, and phosphotyrosine. We also extended these findings to show that SLP-76 forms signaling clusters in primary mast cells. To characterize the mechanism for SLP-76 dynamic clustering, we disrupted the SLP-76 interaction with Gads via minimal mutation of the Gads-binding region of SLP-76 or overexpression of the GBF of SLP-76, which inhibits SLP-76 translocation to the membrane and subsequent clustering. We then demonstrated that disruption of SLP-76 dynamic clustering inhibits FcRI signaling in the RBL cell line and in primary mast cells.
In addition to the studies described in this paper, clustering of signaling proteins has been observed in mast cell (17, 39-41) and T-cell lines (2, 5, 35). These findings raise a number of questions. For instance, what is the function of these “clusters” in FcRI signaling? Since the clusters appear within seconds of stimulation and are found exclusively at the sites of stimulation, they appear appropriately located both temporally and spatially to play a role in signal transduction. In addition, critical mediators of FcRI signaling, such as FcRI, Syk, LAT, and SLP-76, are found within these clusters (9, 12, 30, 32). Loss of any of these proteins severely inhibits FcRI-induced mast cell responses. Further, these clusters are enriched in phosphorylated proteins that are required for many critical interactions downstream of FcRI engagement, such as SLP-76 binding to Btk and Vav (16) and activation of PLCγ (19). Disruption of the SLP-76 interaction with Gads inhibits SLP-76 clustering, FcRI signaling, and mast cell responses. This correlation between SLP-76 clustering and mast cell responses suggests that SLP-76 clustering at the membrane may be critical for the propagation of the FcRI signal.
Previous studies have suggested the existence of two “signaling domains” downstream of the FcRI: a primary signaling domain that includes FcRI, Syk, PLCγ2, Gab2, and the p85 regulatory subunit of PI3K and a second signaling domain that includes LAT, PLCγ1, and p85 (39-41). SLP-76 colocalizes with FcRI, Syk, and LAT, indicating that SLP-76 interacts with proteins found in both signaling domains. In addition, biochemical studies have demonstrated that SLP-76 is phosphorylated by Syk family kinases (15, 37) and associates with LAT (21), PLCγ (46), and p85 (34). These results suggest that SLP-76 may interact with members of both signaling domains, and SLP-76 may potentially regulate events mediated by both complexes. Interestingly, recent studies provide biochemical and functional evidence for two modular FcRI signaling pathways. One pathway, which includes Lyn and LAT, is required for calcium flux and degranulation, while the other pathway, which includes Fyn, Gab2, PI3K, and AKT, is required for degranulation but not for calcium flux (28). Perhaps the primary and secondary signaling domains differentially regulate the two signaling pathways. Biochemical and functional characterization of FcRI signaling in SLP-76-deficient versus Fyn-deficient BMMCs may further shed light on the role of both molecules in these two pathways.
While most SLP-76 clusters colocalize with FcRI, Syk, LAT, and phosphotyrosine, some do not. This incomplete colocalization may indicate the dynamic and heterogeneous nature of these signaling clusters. Since we are observing a dynamic process in real time, some clusters may be in the early stages of formation and may contain phosphorylated proximal signaling proteins, such as FcRI, Src kinases, Syk, and LAT, but have not yet recruited SLP-76. As these clusters mature, they may become enriched in SLP-76, which then facilitates recruitment and/or stabilization of its binding partners, PLCγ1/2, Vav, ADAP, Btk, and Nck. There may also be signaling clusters that function independently of SLP-76. It will be important to more fully characterize the proteins present in these signaling clusters, to define their dynamic interactions, and to investigate the mechanisms for signal termination.
Another question is what function do adaptor molecules, such as SLP-76, play in generating these signaling clusters? Important elements of signaling include the careful regulation of signal initiation, the ability to efficiently transduce a signal once it has been initiated, and the appropriate signal termination. Models for signal initiation include a concept of threshold for activation which is critical for avoiding inappropriate cellular responses (31). To generate a signal downstream of FcRI, perhaps SLP-76 must be phosphorylated and recruited to the membrane, must bind its partners, and must nucleate a multimolecular signaling complex. If a signal is weak, it may not reach the threshold required for SLP-76 to form functional signaling clusters, thus avoiding inappropriate responses. If a signal is robust, it is important to efficiently transduce that signal, allowing a rapid response from the mast cell. In support of this model, strong FcRI signals are required to phosphorylate LAT and to stimulate cytokine production, while weaker signals induce Gab2 phosphorylation and chemokine secretion (14). One way that SLP-76 may regulate the threshold for signaling is by approximating enzymes and substrates. For example, since Btk activates PLCγ (38), having both proteins bound to SLP-76 may greatly enhance the efficiency of PLCγ activation, thus amplifying the original signal. In addition, SLP-76 localizes to membrane subdomains (4) that may be enriched for substrates of PLCγ, further amplifying the signal.
To investigate how SLP-76 forms dynamic signaling clusters, we asked whether a SLP-76 interaction with Gads is required. Studies in T cells suggest that this interaction regulates T-cell-receptor signaling by mediating SLP-76 clustering and recruitment to lipid rafts (4, 35). Minimal mutation of the Gads-binding region of SLP-76 or overexpression of GBF, a competitive inhibitor of this interaction, demonstrates that this interaction is required for FcRI-induced SLP-76 clustering. We then investigated whether disruption of the SLP-76 interaction with Gads would inhibit FcRI-induced mast cell responses. Overexpression of GBF inhibited mast cell responses both in the RBL cell line and in primary BMMCs, suggesting that the SLP-76/Gads interaction is critical for FcRI-induced mast cell responses. These data suggest a model in which Gads recruits SLP-76 to the cell membrane where SLP-76 mediates the formation of multimolecular signaling complexes that regulate FcRI signaling.
Investigation into the mechanisms of signaling may identify attractive therapeutic targets for pathogenic conditions which are dependent upon mast cell function, such as asthma and allergy. In this study, we demonstrated that the GBF peptide inhibits FcRI-induced calcium flux in RBL cells. While these results are encouraging, the relatively high concentration of the peptide required for inhibition suggests that improvements in peptide transport into the cells, peptide stability, and/or peptide affinity will be required before attempts to utilize the GBF peptide in vivo. We are currently defining a minimal GBF sequence that will still function as a dominant negative. As imaging technologies continue to improve, real-time simultaneous imaging of multiple proteins in primary mast cells will enhance our understanding of FcRI signaling and, perhaps, will lead to the discovery of novel therapeutic targets for allergic disorders.
We thank R. Gaehlen, L. Samelson, and R. Tsien for providing reagents. Furthermore, we thank J. Rivera, F. Abtahian, M. Jordan, T. Kambayashi, J. Maltzman, and J. Stadanlick for helpful discussions and review of the manuscript.
This work was supported by the Sandler Program for Asthma Research (G.A.K.). M.A.S. is a trainee under the MSTP grant at the University of Pennsylvania.
†Supplemental material for this article may be found at http://mcb.asm.org/.