The Inositol 5′-Phosphatase SHIP-2 Negatively Regulates IgE-Induced Mast Cell Degranulation and Cytokine Production

Cells, Abs, and reagents

S49.1 murine T lymphoma cells and RBL 2H3 cells were obtained from American Type Culture Collection, whereas 293FT cells were obtained from Invitrogen Life Technologies. All cell lines were maintained in DMEM with 10% heat-inactivated FBS. Rabbit polyclonal anti-SHIP-2 Abs were generated as described (16). Mouse anti-Lyn mAb, anti-SHIP-1 and rabbit polyclonal anti-ERK1/2 and anti-Akt Abs were obtained from Santa Cruz Biotechnology. Rabbit polyclonal anti-Gab2 Abs and mouse anti-phosphotyrosine (4G10) mAb were purchased from Upstate Biotechnology. Mouse anti-SHC mAb, anti-phospho-Akt, and rabbit polyclonal anti-SHC Abs were purchased from BD Biosciences. Rabbit polyclonal anti-phospholipase C(PLC)γ1 Abs, anti-phospho-PLCγ1 Abs, mouse anti-p38 mAb, anti-phospho-p38 Abs, and rabbit anti-phospho-ERK1/2 mAbs were purchased from Cell Signalling Technology. Mouse anti-FcεRI β-chain mAb was a gift from Dr. R. Siraganian (National Institute of Dental and Craniofacial Research, National Institutes of Health, Bethesda, MD).

SHIP-2 KD lentiviral vector production

The lentiviral vector, pLL3.7 was a gift from Dr. L. Van Parijs (Massachusetts Institute of Technology, Cambridge, MA). The shRNA targeting mouse SHIP-2 was a 21-bp oligonucleotide corresponding to position 1365 to 1385 of the SHIP-2 cDNA sequence (GenBank accession no. AF162781). The oligonucleotide was synthesized and cloned into the pLL 3.7 vector as described (28) with minor modifications. In brief, pLL3.7 vector was cotransfected with the lentiviral packaging mix (Invitrogen Life Technologies) into 293FT cells according to the manufacturer’s manual. The culture supernatant, which contained the lentiviral particles, was collected 72 h post-transfection. Titers were determined by transducing 293FT cells with serial dilutions of the lentiviral culture supernatant. GFP expression of the transduced cells was examined by FACS analysis 48 h after transduction. For our preparation, the titer was ∼1–2 × 10⁶ infectious units per ml. The KD efficiency of the SHIP-2 construct was determined by real-time PCR using the primers (forward 5′-GAT TAC CGT CCG ATT GCT ATG C-3′, reverse 5′-GTT GGC GAT ACC AGT CTT C-3′) probing the region 1876–1998 of SHIP-2 coding region.

Primary cell culture and lentiviral transduction

Bone marrow cells from 6- to 8-wk-old C57BL/6 mice were cultured (37°C, 5% CO₂) in DMEM containing 10% FBS, 10 ng/ml IL-3, 20 ng/ml stem cell factor, 50 U/ml penicillin, and 50 mg/ml streptomycin. Fresh medium was changed every week. Lentiviral transduction was performed at 4 wk by spinoculation (29). In brief, 1 × 10⁶ cells were resuspended in lentiviral culture supernatant (multiplicity of infection of 5) containing 8 μg/ml polybrene. Transduction was performed by centrifuging the cells at 1500 × g at 30°C for 90 min. Cells were then resuspended in fresh medium and cultured for 3 additional weeks to allow BMMCs differentiation. Positive cells were subsequently purified by GFP sorting.

FcεRI and c-kit staining

The expressions of FcεRI and c-kit on the surface of BMMCs were detected using the rat anti-mouse FcεRI α-chain mAb (eBioscience) and mouse anti-c-kit mAb (BD Biosciences). Samples were acquired on a FACSCalibur flow cytometer (BD Biosciences) and analyzed using FlowJo software (Tree Star Technologies).

Cell stimulation, lysis, and Western blotting

BMMCs were primed with 2.5 μg/ml anti-trinitrophenyl (TNP) IgE (BD Biosciences) overnight at 37°C. Cells were washed two times with HBSS and resuspended in Tyrode’s buffer (135 mM NaCl, 5 mM KCl, 1.8 mM CaCl₂, 1 mM MgCl₂, 5.6 mM glucose, and 0.1% BSA in 20 mM HEPES (pH 7.4)). After adapting to 37°C for 20 min, cells were stimulated by different concentrations of Ag, TNP-BSA (Biosearch Technologies) for the indicated times at 37°C. For RBL cell stimulation, 50 ng/ml anti-TNP IgE priming for 20 min at 37°C was performed instead of the overnight priming in the stimulation of BMMCs. Resting and activated cells were lysed in cell lysis buffer (50 mM Tris (pH 8.0), 10 mM EDTA, 125 mM NaCl, 10 mM Na₄P₂O₇, 10 mM NaF, 10 mM Na₃VO₄, 1% Triton X-100, and protease inhibitor mixture (Sigma-Aldrich)). Proteins were separated by SDS-PAGE, transferred to PVDF membranes, probed with the Abs of interest, and developed by ECL.

Degranulation assay

Degranulation was monitored by the release of β-hexosaminidase into the culture supernatants as reported earlier (30). BMMCs, sensitized as above, were stimulated with different concentrations (0.04 to 25 ng/ml) of TNP-BSA in 96-well plates (1 × 10⁵ cells/well, 250 μl final volumes) for 1 h at 37°C. Reactions were terminated by centrifugation at 500 × g at 4°C for 5 min, and 200 μl of supernatant was collected for β-hexosaminidase assay. Unstimulated cell lysate prepared by Tyrode’s buffer containing 1% Triton X-100 was used to obtain the total β-hexosaminidase content.

RNA extraction, single-strand cDNA synthesis and real-time PCR

Total RNA was extracted from resting and Ag-stimulated BMMCs using Trizol reagent (Invitrogen Life Technologies). Single-strand cDNA synthesis was performed using SuperScript III first-strand synthesis system (Invitrogen Life Technologies) according to the manufacturer’s instruction. A 10-fold diluted single-strand cDNA was used as template and mixed with the iQ SYBR Green supermix (Bio-Rad) using the following primers:

IL1β forward 5′-CCTGTGTTTTCCTCCTTGCCT-3′; IL-1β reverse 5′-GCCTAATGTCCCCTTGAATCAA-3′; IL-4 forward 5-ATTGTCTCTCGTCACTGACGGA-3′; IL-4 reverse 5-CGTGGATATGGCTCCTGGTACA-3′; IL-6 forward 5′-CAGAAGGAGTGGCTAAGGACCA-3′; IL-6 reverse 5′-ACGCACTAGGTTTGCCGAGTAG-3′; IL-13 forward 5′-ACAAAGCAACTGTTTCGCCAC-3′; IL-13 reverse 5′-TCTCCCCAGCAAAGTCTGATG-3′; TNF-α forward 5′-CCTTGTTGCCTCCTCTTTTGC-3′; TNF-α reverse 5′-TCAGTGATGTAGCGACAGCCTG-3′; MCP1 forward 5′-AAGGCAAAGGCAGTTCTCAACA-3′; MCP1 reverse 5′-AAGTGACCTGCTTCTCCTGCAG-3′; β-actin forward 5′-ACCAACTGGGACGATATGGAGAAGA-3′; β-actin reverse 5′-TACGACCAGAGGCA TACAGGGACAA-3′.

Real-time PCR was performed using the iCycler (Bio-Rad). The data were expressed as the C_T of cytokines normalized with the C_T of β-actin of each sample.

Intracellular calcium measurements

IgE-sensitized BMMCs were loaded with 5 μM Fura red-AM (Molecular Probes) for 30 min at 37°C. Cells were then washed 2 times with HBSS and resuspended in Tyrode’s buffer. TNP-BSA (1 ng/ml) was added for stimulation, and the resulting changes in intracellular calcium were monitored by FACS.

Microtubules (polymeric tubulin) assay

Ag-stimulated RBL cells were lysed in cell lysis buffer for 15 min on ice. Cell lysates were centrifuged at 12,000 × g for 10 min. Supernatants were discarded and the remaining pellets (Triton X-100 insoluble fraction) were resuspended in SDS lysis buffer (25 mM Tris (pH 7.4), 0.4 M NaCl, and 0.5% SDS) and boiled for 10 min. The samples were centrifuged at 12,000 × g for 10 min and the microtubule-containing supernatants were collected, and their protein concentrations were determined by BCA protein assay (Pierce). Ten micrograms of total proteins from each sample were subjected to SDS-PAGE, and the microtubules were detected by immunoblotting with anti-α-tubulin Ab.

F-actin quantification

RBL cells were seeded on 6-well plates (1 × 10⁶ cells/well) the day before measurement. To measure the total F-actin change during mast cell activation, cells were sensitized as above. The IgE-sensitized cells were washed two times with HBSS and stimulated with 10 ng/ml TNP-BSA for different time points. Stimulation was stopped by removing the TNP-BSA containing medium and cells were fixed with 3.7% paraformaldehyde in PBS for 20 min at room temperature. Cells were then permeabilized by 0.5% Triton X-100 in PBS for 5 min. After two washes with HBSS, 200 μl of 0.33 μM Alexa Fluor 488-conjugated phalloidin (Molecular Probes) was added to each well, and the plates were incubated for 30 min in dark. After two washes with HBSS, F-actin bound phalloidins were then eluted by incubating the cells with 500 μl of 100% methanol for 2 h in the dark. Supernatants were collected and the amount of eluted phalloidins was measured by fluorometry (31).

Rac-1 activation assay

Ag-stimulated RBL cells were lysed in cell lysis buffer for 15 min on ice. Cell lysates were then centrifuged at 12,000 × g for 10 min and the supernatants were collected. Total protein concentrations of the samples were determined by BCA protein assay. Equal amounts of proteins from each sample were incubated with PAK-1-PBD agarose (Cell Biolabs) at 4°C for 1 h. The samples were washed two times with cell lysis buffer, followed by boiling in 2× SDS sample buffer for 10 min. The boiled samples were subsequently centrifuged at 12,000 × g for 10 min. Active Rac-1 (GTP bound) in the supernatants was detected by Western blot using the mouse anti-Rac-1 mAb (Cell Biolabs).

GATA-1 ELISA

Ag-stimulated RBL cells were incubated for 10 min at 4°C in buffer A (10 mM HEPES (pH 7.6), 3 mM MgCl₂, 10 mM KCl, 5% glycerol, 10 mM Na₄P₂O₇, 10 mM NaF, 10 mM Na₃VO₄, protease inhibitor mixture, and 0.5% Nonidet P-40). After centrifugation, nuclear pellets were lysed by the complete lysis buffer provided by the TransAM GATA transcription factor assay kit (Active Motif). The activity of GATA-1 from the nuclear samples was determined by ELISA, according to the manufacturer’s instructions.

Generation of SHIP-2 KD BMMCs by a lentiviral RNA interference (RNAi) system

Cross-linking of FcεRI on mast cells induces a cascade of kinase-triggered protein tyrosine phosphorylation and lipid phosphorylation that has been previously shown to be regulated by the phosphatases SHP-1 and SHIP-1 (32, 33). We have observed that the phosphoinositol phosphatase SHIP-2 becomes tyrosine phosphorylated upon FcεRI engagement with IgE and Ag (Fig. 1⇓A). Thus, we decided to investigate the role of SHIP-2 in regulating BMMCs activation. For this purpose we generated SHIP-2 KD BMMCs using a lentiviral-based RNAi technique. Six different shRNA sequences were inserted into the lentiviral vector pLL3.7 (28) and tested for KD activities by transduction into the T-lymphoma S49.1 cell line. Real-time PCR revealed that three of six of the SHIP-2 KD constructs strongly reduced the expression of SHIP-2 mRNA (60–80%) (Fig. 1⇓B). The most efficient KD construct, shRNA-4, was picked for the rest of the experiments and the lentiviral vector alone was used as control. BMMCs isolated from 6- to 8-wk-old C57BL/6 mice were then transduced with shRNA-4 (SHIP-2 KD) and lentivirus control. FACS analysis to detect GFP-positive cells showed that the transduction efficiencies were 30–50% (data not shown). GFP-positive cells were sorted and SHIP-2 protein KD efficiency was evaluated by immunoblotting. SHIP-2 KD BMMCs showed highly reduced SHIP-2 protein expression when compared with either the shRNA control BMMCs or the wild-type cells. Transduction of shRNA-4 had no effect on the expression of the homologous protein SHIP-1 (Fig. 1⇓C).

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 1.

Generation of SHIP-2 KD mast cells to assess its role in FcεRI-induced activation. A, SHIP-2 is tyrosine phosphorylated upon Ag stimulation. BMMCs from C57BL/6 mice were primed with 2.5 μg/ml anti-TNP IgE overnight at 37°C and stimulated with 1 ng/ml TNP-BSA for the indicated times. Anti-SHIP-2 immunoprecipitates were analyzed by Western blot analysis using the anti-phosphotyrosine (4G10) and reprobed with anti-SHIP-2 for normalization. Results shown are representative of two experiments. B, Lentiviral-based shRNA constructs for SHIP-2 in mouse created according to Robinson et al. (28 ). Six different shRNAs were inserted into the pLL3.7 lentiviral vector and SHIP-2 KD efficiency at the transcriptional level was examined in the murine T cell line S49.1 by real-time PCR. C, Western blot was performed to confirm the reduction of SHIP-2 protein expression in shRNA-4 transduced BMMCs. The blot was normalized by reprobing with anti-SHIP-1 Abs. D, FcεRI and c-kit receptor surface expression was tested in SHIP-2 KD and control BMMCs. Solid lines, SHIP-2 KD; dashed lines, shRNA-control transduced BMMCs; and shaded histograms are unstained controls.

To determine whether SHIP-2 KD would affect mast cell development, we analyzed the expression of surface markers FcεRI and c-kit by FACS. As shown in Fig. 1⇑D, both shRNA control and SHIP-2 KD-transduced BMMCs expressed comparable levels of both markers, indicating that the two cell types had no developmental differences. For all experiments tested, we obtained the same results using BMMCs that were transduced either before or after in vitro differentiation.

SHIP-2 negatively regulates mast cell degranulation and cytokine gene expression

It is well established that SHIP-1 acts as a negative regulator of IgE-induced mast cell degranulation by maintaining a proper PIP₃ level which in turn restricts calcium flux into the cells (8, 11, 13). To investigate whether SHIP-2 would also participate in regulating mast cell degranulation, wild-type, shRNA control and SHIP-2 KD BMMCs were primed with anti-TNP IgE overnight. Degranulation was induced by incubating with different concentrations of TNP-BSA. The activity of the released β-hexosaminidase was measured as an indicator of mast cell degranulation (Fig. 2⇓A). Both wild-type and shRNA control cells degranulated similarly for the whole range of TNP-BSA stimulation while SHIP-2 KD cells showed consistently more efficient mast cell degranulation. The level of degranulation in SHIP-2 KD cells was always two fold greater than that obtained with control cells at any concentration of Ag. Meanwhile, stimulation of all three cell groups with PMA (10 nM) and ionomycin (1 μM) gave similar degree of degranulation of almost maximal levels (Fig. 2⇓B).

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 2.

SHIP-2 KD increases efficiency of mast cell degranulation and cytokine production upon FcεRI engagement. A, Wild-type, shRNA control, and SHIP-2 KD BMMCs were primed with 2.5 μg/ml anti-TNP IgE overnight at 37°C and stimulated with the indicated concentrations of TNP-BSA for 1 h at 37°C. Experiments were performed in triplicate. Results shown are representative of six independent cultures. B, SHIP-2 KD, shRNA control or untransduced BMMCs were tested for degranulation by addition of 10 nM PMA and 1 μM ionomycin for 10 min at 37°C. Shadowed bars, untransduced cells; black bars, shRNAi control; white bars, SHIP-2 KD cells. C, BMMCs derived from FcγRIIB-KO mice were transduced with SHIP-2 KD shRNA or control shRNA and tested as in panel A. Degranulation was detected as the percentage of β-hexosaminidase released to the supernatant compared with total β-hexosaminidase content. Experiments were performed in triplicate. Results shown are representative of six independent cultures. D, SHIP-2 KD and control BMMCs were primed with 2.5 μg/ml anti-TNP IgE overnight at 37°C and then either left unstimulated (0 h) or stimulated with 10 ng/ml TNP-BSA for 4 h at 37°C. Black bars, shRNA control; white bars, SHIP-2 KD. Real-time PCR was performed to assay the mRNA expression of the indicated cytokine. Experiments were performed in triplicate. Results shown are representative of two experiments using independent sets of lentiviral-transduced BMMCs. ∗, p < 0.05; ∗∗, p < 0.001.

Because SHIP-2 has previously been shown to be recruited to the low affinity inhibitory FcγRIIB receptor (19, 20), and the activity of FcγRIIB can modulate FcεRI-mediated IgE reponses, we decided to test whether the mast cell degranulation increase observed in SHIP-2 KD mast cells was dependent on the presence of the inhibitory Fc receptor. We generated BMMCs derived from FcγRIIB-KO mice and transduced them with shRNA-4 and vector-control lentiviral constructs. Overall, degranulation of FcγRIIB-deficient mast cells was more efficient at all concentrations of Ag compared with wild type cells, but there was still a two fold increase in those FcγRIIB-KO cells transduced with the SHIP-2 KD lentivirus (Fig. 2⇑C). This result demonstrates that SHIP-2 regulates IgE-triggered mast cell degranulation independently of the inhibitory Fcγ receptor.

IgE-induced activation of mast cells leads to de novo synthesis and secretion of cytokines (34). To investigate whether the absence of SHIP-2 would affect cytokine production in mast cells, the mRNA levels of different cytokines in both IgE-activated SHIP-2 KD and control BMMCs were determined by real-time PCR (Fig. 2⇑D, primers shown in Material and Methods). Our results indicate that IgE + TNP-stimulated SHIP-2 KD cells only show a moderate elevation in TNF-α and IL-6 mRNA expression when compared with shRNA control cells, while there is no observable difference in IL-1β and MCP-1 mRNA levels between the two cell groups. Meanwhile, the expression levels of IL-4 and IL-13 were significantly increased (two fold or more) in activated SHIP-2 KD mast cells compared with shRNA control cells (Fig. 2⇑D).

Absence of SHIP-2 does not alter either FcεRI surface expression and internalization, or IgE-induced calcium mobilization and MAPK phosphorylation

It has previously been reported that the surface expression of FcεRI increases upon IgE priming and that this receptor is internalized after coligation with Ag during mast cell degranulation (35). Because an increase in surface expression of FcεRI would potentially affect the intensity of degranulation, we investigated whether the absence of SHIP-2 has an effect on membrane levels of FcεRI. As shown in Fig. 3⇓A, the expression pattern of FcεRI before and after receptor stimulation was the same in SHIP-2 KD and control BMMCs. This result indicates that the up-regulation of mast cell degranulation in SHIP-2 KD cells is unrelated to any change in FcεRI surface expression or in activation-mediated internalization.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 3.

SHIP-2 KD has no effect on FcεRI surface expression or on FcεRI-triggered calcium flux and MAPKs phosphorylation. A, Expression of FcεRI by flow cytometry in SHIP-2 KD and shRNA control BMMCs. Dotted line, unstimulated cells; dark line, primed with 2.5 μg/ml anti-TNP IgE overnight; thin line, primed and stimulated with 1 ng/ml TNP-BSA for 1 h; gray histogram, unstained cells. B, IgE-sensitized BMMCs were incubated with 5 μM Fura red-AM at 37°C for 30 min and stimulated with 1 ng/ml TNP-BSA. The intracellular calcium change was monitored with a FACSCalibur on FL-2 fluorescence. Gray line, shRNA control; black line, SHIP-2 KD. Results shown are representative of three experiments using independent BMMCs cultures. C, Cell lysates from IgE-sensitized and TNP-BSA-stimulated BMMCs were subjected to Western blot analysis using anti-phospho-PLCγ1 Abs and reprobed with anti-PLCγ1 Abs for normalization. Relative intensities were quantified by densitometry and expressed as the ratios of phospho-PLCγ1/total PLCγ1. Results shown are representative of two experiments. D, Experimental setup is the same as in C, with the use of anti-phospho-p38 and anti-phospho-ERKs Abs. The membrane was stripped and reprobed with anti-p38 and anti-ERKs Abs for normalization. Results shown are representative of two experiments.

We also observed the same calcium flux responses with SHIP-2 KD cells as with shRNA control cells (Fig. 3⇑B). For this experiment, BMMCs were loaded with Fura-red, sensitized with anti-TNP IgE, and stimulated with TNP-BSA. The kinetics of calcium flux was monitored by the reduction in red fluorescence intensity. Both SHIP-2 KD and control BMMCs were associated with the same level of intracellular calcium rise either upon IgE plus TNP-BSA stimulation or in the presence of the ionophore, ionomycin. Next we examined the tyrosine phosphorylation of PLCγ1, which is an essential event preceding intracellular calcium mobilization. As seen in Fig. 3⇑C, there was no difference in PLCγ1 phosphorylation between the control and SHIP-2 KD cells. Moreover, we didn’t observe a change in the tyrosine-phosphorylation of ERK1/2 or p38 MAPKs in the activated SHIP-2 KD cells compared with shRNA control cells (Fig. 3⇑D).

SHIP-2 KD does not alter F-actin depolymerization but enhances microtubule formation and Rac-1 activity during mast cell degranulation

Several studies have suggested that cytoskeletal rearrangements (F-actin disassembly and microtubule polymerization) are important events in FcεRI-mediated mast cell degranulation (4, 6, 36, 37). Because SHIP-2 has previously been shown to be involved in the regulation of cytoskeleton organization in various cell types (38, 39, 40), we speculated that it might also participate in cytoskeletal rearrangements in mast cells. To test this hypothesis in an immortal cell line, we generated SHIP-2 KD RBL cells using the same lentiviral constructs described earlier. As we observed in experiments using BMMCs, SHIP-2 KD-lentiviral transduction resulted in highly reduced expression of SHIP-2 protein in RBL cells and concomitant elevation in both mast cell degranulation and IL-4/IL-13 cytokine production compared with shRNA control RBL cells (data not shown). Using the RBL cell line system, we observed that aggregation of the FcεRI receptor induced microtubule polymerization at greater levels in SHIP-2 KD cells compared with controls (Fig. 4⇓A). Total α-tubulin content (both Triton X-100 soluble and insoluble fractions) was the same in both control and SHIP-2 KD samples. In the same cells, we observed that F-actin disassembly induced by FcεRI aggregation was the same independently of the SHIP-2 KD effect (Fig. 4⇓B).

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 4.

Decreased SHIP-2 expression enhances microtubule polymerization but does not affect F-actin ring depolymerization during FcεRI-induced mast cell activation. A, SHIP-2 KD and shRNA control RBL cells were primed with 50 ng/ml anti-TNP IgE for 20 min at 37°C and stimulated with 10 ng/ml TNP-BSA for the indicated times. Stimulated cells were lysed and the Triton X-100-insoluble fractions, which contain the microtubules, were collected and subjected to the BCA protein assay. Equal amounts of proteins from each sample were analyzed by Western blotting using anti-α-tubulin Ab. The band intensities were compared with that of the 0 min of the shRNA control. Results are representative of two independent experiments. B, IgE-primed SHIP-2 KD and shRNA control RBL cells were stimulated with TNP-BSA (10 ng/ml) for the times indicated. At the end of each time point, cells were fixed and stained with Alexa Fluor 488 phalloidin for 30 min. F-actin bound phalloidins were eluted with 100% methanol and their amounts were determined by the fluorescence signal emitted. Data were expressed as percentage relative to the signal of resting cells. Experiments were performed in duplicate. Results shown are representative of two independent experiments. C, Ag-stimulated SHIP-2 KD and shRNA control RBL cells were lysed and centrifuged, with the supernatants collected and subjected to BCA protein assay. Equal amounts of proteins from each sample were incubated with PAK-1-PBD agarose at 4°C for 1 h. Samples were washed, boiled in 2× SDS sample buffer and clarified by centrifugation. Active Rac-1 (GTP bound) in the supernatant was detected by Western blotting using the anti-Rac-1 Ab. Band intensities were compared with that of the 0 min of the shRNA control. Results shown are representative of two independent experiments.

Previous studies have shown that the small GTPase Rac-1 is involved in the regulation of cytoskeletal rearrangements (41, 42). Recently, Ai et al. reported that SHIP-2 down-regulates Rac-1 activity in macrophages (22). We hypothesized that SHIP-2 could also modulate Rac-1 activity in mast cells during its regulation of microtubule formation in mast cell degranulation. To test this hypothesis, the level of active Rac-1 (the GTP bound form) present in Ag-stimulated control and SHIP-2 KD RBL cells was compared by western blot analysis. As shown in Fig. 4⇑C, Rac-1 activation was prominent at 2 min after stimulation and gradually declined at 5 min in shRNA control cells. The same activation pattern was observed in SHIP-2 KD cells, but the level was higher than in the control cells at any time point. This result supports the hypothesis that SHIP-2 negatively regulates microtubule formation by suppressing upstream Rac-1 activation during mast cell degranulation.

Enhanced GATA-1 and Akt activities in SHIP-2 KD RBL cells

The transcription factor GATA-1 has been shown to promote IL-4 and IL-13 RNA expression in mast cells (43, 44). Because our data revealed that SHIP-2 negatively regulated IL-4 and IL-13 mRNA expression, we speculated that SHIP-2 might exert its regulatory effect on the transcriptional activity of GATA-1. To test this hypothesis, we determined GATA-1 activity in Ag-stimulated SHIP-2 KD and control RBL cells by ELISA. As shown in Fig. 5⇓A, GATA-1 activity increased predominantly 15 min after stimulation and gradually declined after 30 min in both SHIP-2 KD and control RBL cells, but the extent of activation was much higher in SHIP-2 KD cells than in the control cells.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 5.

SHIP-2 KD increases the transcriptional activity of GATA-1 and enhances Akt Ser phosphorylation. A, SHIP-2 KD and shRNA control RBL cells were primed with 50 ng/ml anti-TNP IgE for 20 min and stimulated with 10 ng/ml TNP-BSA for the times indicated. Nuclear extracts were prepared and ELISA was performed to determine the transcriptional activity of GATA-1. ∗, p < 0.05; ∗∗, p < 0.0001. Experiments were performed in duplicate. Results shown are representative of three independent experiments. B, SHIP-2 KD and shRNA control RBL cells were primed and stimulated as in Fig. 4⇑. Lysates were immunoblotted with Ab against phospho-Akt and reprobed with anti-Akt Ab. Relative intensities were expressed as the ratios of phospho-Akt/total Akt. Results shown are representative of two independent experiments.

GATA-1 activity has been reported to be dependent on its serine 310 phosphorylation by Akt in erythroid cells (45) and SHIP-2 can also negatively regulate Akt activity in glioblastoma cells (46). Thus, we tested whether the up-regulation of GATA-1 activity in SHIP-2 KD cells correlated with up-regulation of Akt activity. We compared the Akt phosphorylation on Ser473 in SHIP-2 KD and control RBL cells. As shown in Fig. 5⇑B, SHIP-2 KD RBL cells exhibited much higher phosphorylation of Akt than the control cells upon IgE-Ag stimulation. This result implies that SHIP-2 negatively regulates Akt activity, which may in turn control the transcriptional activity of GATA-1.

SHIP-2 does not interact with SHC, but it associates with Lyn, Gab2 and FcεRI β-chain in IgE-activated RBL cells

We next tested molecules that could physically interact with SHIP-2 during mast cell activation. We looked first for SHIP-2 association with the adaptor molecule SHC because SHC has been shown to associate with SHIP-1 upon FcR-mediated stimulation of mast cells (47). SHC-SHIP-1 interaction seems to be essential for SHIP-1 to exert its negative regulation in RBL cells (48). We analyzed anti-SHC immunoprecipitates in cell lysates from IgE-sensitized and Ag-stimulated RBL cells for the presence of SHIP-1 or SHIP-2-associated molecules. As shown in Fig. 6⇓A, SHIP-1 was constitutively associated with SHC in RBL cells, consistent with previously published observations. In contrast, no interaction between SHIP-2 and SHC was detected either before or after IgE-induced activation. It should be noted that we did observe the interaction between SHIP-2 and SHC in the anti-mouse IgM stimulated A20 B cells using the same Abs as in the experiment in RBL cells (right panel Fig. 6⇓A), indicating that the Abs we used are functional for immunoprecipitation.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 6.

SHIP-2 does not interact with SHC, but associates with Lyn, Gab2 and FcεRI β-chain in RBL cells. A, Cell lysates from IgE-sensitized and Ag-stimulated RBL cells, or from anti-IgM-activated A20 cells, were collected for immunoprecipitation using anti-SHC mAb. The membrane was probed with anti-SHIP-1, anti-SHIP-2, and rabbit anti-SHC Abs. Total cell lysates were run in parallel, acting as a positive control for Ab staining. B, SHIP-2 KD and shRNA control RBL cells were primed and stimulated as in Fig. 4⇑. Cell lysates were immunoprecipitated with anti-SHIP-2 Ab and subjected to SDS-PAGE, followed by immunoblotting. The same membrane was probed with four different Abs: anti-FcεRI β-chain, anti-Lyn, anti-Gab2, and anti-SHIP-2. Results shown are representative of three independent experiments.

To look for molecules other than SHC that could associate with SHIP-2 upon mast cell activation, we analyzed anti-SHIP-2 immunoprecipitates in RBL cells. Several bands were detected by anti-phosphotyrosine immunoblot (data not shown). Guessing by the size of these bands, we tested SHIP-2 precipitates for the presence of Gab2, Lyn, and FcεRI-β-chain (Fig. 6⇑B). We observed that SHIP-2 constitutively associates with the FcεRI β-chain. Upon Ag stimulation, SHIP-2 was shown to interact with Gab2, Lyn, and three other unknown (37, 74, and 120 kDa) tyrosine-phosphorylated proteins (Fig. 6⇑B and data not shown). In an effort to identify these unknown proteins, membranes were probed with Abs against p38, SLP76, ITK, Syk, ZAP 70, and PLCγ1 but none of these molecules were detected in anti-SHIP-2 immunoprecipitates (data not shown).