To investigate the role of SLAP in the regulation of mast cell (MC) functions, bone marrow-derived mast cells (BMMCs) from Balb/C (WT) and SLAP knockout (KO) mice were generated. Immunoblot analysis of mature BMMCs from SLAP KO shows a complete absence of the SLAP protein compared to WT BMMCs (Figure 1A). Since the role of SLAP is well-characterized in the regulation of surface receptor expressions in other immune cell types (13, 18), we investigated the surface expression of characteristic mast cell receptors. FcεRI, cKit (CD117) and SIRPα (CD172a) showed comparable expression levels in both genotypes (n = 3) (Figure 1B). Mast cell protease-1 (MCPT-1), FcεRIα, FcεRIβ, and FcεRIγ showed similar mRNA expression levels in MC from WT and SLAP-KO, as determined by quantitative PCR (Figure 1C). This suggested that SLAP KO BMMCs do not have any maturation defect. To understand the role of SLAP in the IgE (FcεRI) receptor-mediated MC function, we measured the FcεRI-mediated degranulation response post-stimulation with the antigen, DNP-HSA. SLAP KO BMMCs showed an enhanced FcεRI mediated degranulation response in a dose dependent manner, compared to control BMMCs based on β-hexosaminidase release (Figure 1D). To confirm the enhanced degranulation of SLAP KO MCs, cell surface phosphatidylserine externalization was measured using annexin V staining on degranulated mast cells as previously described (19). SLAP KO BMMCs showed a significantly higher level of annexin V staining upon IgE cross-linking, as compared to WT cells (Figure 1E). In line with this finding, surface expression of CD107a (LAMP-1), a marker of granular exocytosis (20), was also significantly higher in the SLAP KO BMMCs post-stimulation (Figure 1F). Collectively, these results provide evidence that SLAP is a negative regulator of FcεRI-dependent MC degranulation.

In addition to a rapid degranulation response, FcεRI mediated signaling also regulates the late phase reaction of cytokine secretion via the de novo synthesis of cytokines (21). To investigate the role of SLAP in cytokine secretion, BMMCs were sensitized with anti-DNP IgE and stimulated with DNP-HSA (10 ng/ml) for 6 and 24 h, supernatants were collected, and cytokine levels were determined by ELISA. The levels of secreted IL-6, MCP-1, and TNFα were significantly higher in SLAP KO BMMCs at 6 h compared to the WT control (Figures 2A–C) indicating that SLAP negatively regulates cytokine production post-DNP-HSA stimulation.

To further characterize the role of SLAP in IgE-mediated mast cell function, we conducted a passive cutaneous-anaphylaxis assay in ear skin as previously described (22–24). WT and SLAP KO mice were sensitized with the IgE antibody for 48 h followed by the topical application of dinitrofluorobenzene (DNFB) (left ear) or vehicle control (right ear). After 24 h, swelling in DNFB treated ears was observed while no reaction occurred in the vehicle treated ear. SLAP KO mice (n = 7) had significantly increased ear skin thickness compared to WT mice (n = 6) (Figure 3A). Histological examination of ear tissue sections further confirmed that DNFB application resulted in inflammatory swelling in both genotypes, however, inflammation was more pronounced in the SLAP KO ear skin (Figure 3B). To confirm that the enhanced response of SLAP KO mice was due to mast cell activation, rather than elevated mast cell numbers, histological sections of ear skin from WT and SLAP KO mice were examined. Average mast cell numbers/field were comparable in SLAP KO and WT mice in DNFB treated ear skin (Figures 3C,D). This suggests that SLAP negatively regulates the IgE-mediated cutaneous anaphylaxis functional response but does not impact the viability, expansion or the recruitment of mast cells.

To further investigate the impact of SLAP deficiency on MC responses, F-actin polymerization, a key process in mast cell degranulation (25) was examined. Sensitized BMMCs were stimulated with DNP-HSA for the indicated time-points prior to fixation and permeabilization and F-actin content was quantified by phalloidin staining as described previously (26, 27). Flow cytometric analysis showed significantly reduced F-actin levels in SLAP KO BMMCs prior to stimulation (time 0) compared to WT BMMCs (Figure 4A). Following DNP-HSA stimulation, both WT and SLAP KO BMMCs responded similarly, although F-actin remained significantly lower at late time points (27 min). In addition to F-actin polymerization, calcium mobilization in response to FcεRI stimulation contributes to degranulation. Therefore, we loaded the BMMCs with an Indo-1AM ratiometric Ca²⁺ fluorescent indicator and stimulated cells with DNP-HSA or ionomycin in the presence or absence of EDTA. Quantification of the intracellular calcium concentration (nM) showed comparable levels between the genotypes at indicated time points (Figure 4B).

To determine the role of SLAP in pathways that regulate degranulation and cytokine responses, we studied the proximal signaling events downstream of FcεRI. IgE-sensitized BMMCs from SLAP KO or WT mice were stimulated with DNP-HSA and phosphorylation of Lyn and Syk was determined by immunoblotting with anti-phosphotyrosine and phospho-site specific antibodies. While there was no change in phosphorylation of Lyn (Figure 5A), enhanced phosphorylation of Syk activation loop tyrosines (Y525/526) was observed in SLAP KO BMMCs at 3, 9, and 27 min after DNP-HSA stimulation compared to WT (Figures 5B,C). Increased phosphorylation of Syk (tyrosine 238) in SLAP KO BMMC was also confirmed using phospho-flow cytometry (Figure 5D). Notably, SLAP KO cells showed a modest increase in the total phosphotyrosine containing proteins, compared to WT BMMCs (Figure 5A).

Activation of Syk leads to phosphorylation of the membrane adaptor protein LAT, and subsequent recruitment and activation of multiple downstream pathways including Ras-ERK-MAPK and AKT-IKK-NFκB pathways to regulate cytokine and chemokine production (28). Enhanced phosphorylation of ERK1/2 was observed in SLAP deficient BMMCs, and this was most pronounced at later time points, evidence of sustained activation of downstream signaling from SYK and LAT (Figure 6A). While there was no significant difference in p38MAPK phosphorylation (Figure 6B), sustained activation of JNK1/2 was evident in SLAP deficient MCs (Figure 6C). In keeping with the observation that SLAP KO BMMC do not show alterations in calcium mobilization, no significant change in PLCγ1 phosphorylation post-stimulation was observed in SLAP KO BMMCs (Figure 6D). SLAP KO BMMCs also had enhanced AKT phosphorylation, particularly at early time points after antigen stimulation (Figure 7A). Similarly, a modest increase in phosphorylation of IKK and the downstream transcription factor NFkB was apparent at early time points (Figures 7B,C). Collectively, these results indicate that SLAP negatively regulates antigen induced FcεRI signaling to Syk, and its downstream effectors, ERK, JNK, and AKT.

Previously, SLAP has been shown to regulate the surface expression and down regulation of receptors such as TCR and BCR via recruitment of the ubiquitin ligase Cbl (13, 29, 30). Therefore, one possible mechanism to explain the enhanced signaling in SLAP deficient BMMCs is due to impaired Cbl function in the absence of SLAP. Prior studies have indicated that Cbl-b is the dominant Cbl family member expressed in BMMC where it has been demonstrated to regulate the surface expression and signaling downstream of FcεRI (10). To assess whether SLAP loss might have effects on Cbl-b function or its activity, we first tested whether SLAP interacts with Cbl-b. GST-SLAP fusion proteins were incubated with RBL-2H3 cell lysates and bound proteins analyzed by immunoblot. GST- SLAP bound to several tyrosine phosphorylated proteins including Cbl-b and FcεRIβ in both un-stimulated and DNP-HSA stimulated cells (Figure 8A). To investigate whether SLAP affects the surface expression of FcεRI upon stimulation with antigen, IgE-sensitized BMMCs were stimulated with DNP-HSA (20 ng/ml) and cell surface levels of FcεRI were determined by flow cytometry. Quantification of MFI as a measure of receptor surface expression levels revealed that SLAP KO BMMCs retained significantly higher levels of the high affinity IgE receptor compared to WT at all time-points post-stimulation (Figure 8B), suggesting impaired internalization of FcεRI. The association of SLAP and Cbl-b and altered FcεRI internalization in SLAP deficient cells could suggest that either the adaptor or ubiquitin ligase function of Cbl-b is defective in the absence of SLAP. In the RBL-2H3 cell line, both FcεRIβ and FcεRIγ are ubiquitinated following receptor aggregation and Cbl overexpression has been shown to promote ubiquitination of both subunits and enhances receptor internalization (31, 32). To test the impact of SLAP deficiency on FcεRI ubiquitination, stimulated cell lysates from wild type or SLAP KO BMMCs were incubated with ubiquitin binding domain linked agarose (agarose-TUBE) and bound proteins were detected by immunoblot. Ubiquitinated species of both FcεRIβ and FcεRIγ were observed at 3- and 9-min following stimulation. While ubiquitination of FcεRIβ appeared unaffected in the absence of SLAP (Figure 8C), recovery of ubiquitinated species of FcεRIγ was reduced in SLAP deficient compared to wild type cells (Figure 8D). Altogether these data support a model in which SLAP is important for recruitment of Cbl-b to FcεRI, ubiquitination of FcεRIγ and receptor down regulation.

A previous study in the RBL-2H3 cell line demonstrated that SLAP expression is upregulated in response to antigen stimulation (33). In agreement, the levels of SLAP protein detected by immunoblotting increased following the antigen stimulation of murine BMMCs (Figure 9A). To further assess the contribution of SLAP in IgE receptor signaling, we determined intracellular SLAP protein expression in peripheral blood derived human basophils, upon anti-FcεRI or allergen stimulation. Basophils derived from eight human donors, uniformly up-regulated SLAP expression upon FcεRI receptor cross-linking (Figures 9B,C). Furthermore, the same response was demonstrated at an allergen specific level. Basophils from food allergy patients stimulated with relevant food allergen showed a significant increase in SLAP protein levels similar to that observed upon FcεRI cross-linking, whereas stimulation with a clinically irrelevant food allergen resulted in no change in SLAP levels (Figure 9C). All experiments were controlled for the degranulation of basophils (CD63 expression) in response to FcεRI cross linking antibody and the allergen. These results indicate that FcεRI stimulation modulates SLAP expression in human basophils and suggests that SLAP has a role in modulating allergic responses.

SLAP⁻ Balb/c (SLAP KO) have been described previously (12, 14) and Balb/c (wild type:WT) mice were bred and maintained under institute guidelines at the Toronto Centre for Phenogenomics. Experiments were performed on both male and female mice at 10–12 weeks of age. All studies and experimental procedures were approved by the Toronto Centre for Phenogenomics Animal Care Committee.

Suspended BMMC cultures were derived as previously described (24, 55). Briefly, bone marrow cells from SLAP KO and WT mice were cultured for 4 weeks to generate BMMC cultures in BMMC growth medium [IMDM, 16% (v/v) Fetal calf serum (ATCC: 30-2030), 1% (v/v) antimicrobial-antimycotic solution (Invitrogen), 1 mM sodium pyruvate (Invitrogen), 1% (v/v) minimum non-essential amino acids (Invitrogen), 2% (v/v) conditioned medium from X63-IL-3 cells and 50 μM α-monothioylglycolate (Sigma-Aldrich)]. Maturation of BMMCs were determined by analyzing the expression of surface FcεRIα and Kit receptor using Life Technologies Attune® acoustic focusing cytometer. BMMC cultures expressing >90% double positive (FcεRIα⁺/Kit⁺) were considered mature and further used for experiments. The following flow antibodies were used: Biolegend anti- IL-6 (APC; clone MP5-20F3), cKIT(CD117, APCCy7; clone 2B8), IgE (FITC; clone 23G3), CD45 (PE-Cy7;clone 30-F11), SIRPα (CD172a, PE-Cy7; clone P84), Rat IgG2b κ (Isotype, APC-Cy7; clone RTK4530), Rat IgG1, κ (Isotype, APC; clone RTK2071), Rat IgG1, κ (Isotype, PE-Cy7; clone RTK2071), BD Bioscience anti- TNFα, (PE;clone MP6-XT22), or ebiosciences anti-CD16/CD32 (FcBlock, purified, clone 93).

Mature BMMCs (10⁷ cells/time point) were starved overnight in starvation medium [IMDM, 2% (v/v) Fetal calf serum (ATCC: 30-2030), 1% (v/v) antimicrobial-antimycotic solution (Invitrogen), 1 mM sodium pyruvate (Invitrogen), 1% (v/v) minimum non-essential amino acids (Invitrogen), and 50 μM α-monothioylglycolate (Sigma-Aldrich)] with 20% (v/v) anti-DNP-IgE conditioned medium (SPE7 clone; ≈1 μg/ml) for 18 h. Sensitized BMMCs were washed 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, 0.1% BSA] twice. Mast cells were stimulated with either vehicle control or 30 ng/ml dinitrophenyl-human serum albumin (DNP-HSA; Santacruz Biotechnology) in Tyrode’s buffer at 37°C for indicated timepoints and reactions were by adding ice-cold PBS. Soluble cell lysates (SCL) were prepared in Phospho-lipase C buffer [PLC buffer; 50 mM HEPES (pH 7.5), 150 mM NaCl, 105 Glycerol, 1 mM EGTA, 1% (v/v) Triton-X, 1.5 mM Mgcl₂, complete protease inhibitor cocktail (Roche), and phosphatase inhibitor (Cell Signaling Technology)] at 12,000 g for 15 min at 4°C.

Degranulation assays were performed as described previously (56). Briefly, BMMCs of SLAP KO and WT genotypes (1 × 10⁶ cells/ml) were starved and sensitized with 20% (v/v) anti-DNP-IgE conditioned medium for 18 h, and washed twice with warm Tyrode’s buffer. In 96-well plates, BMMCs (6.25 × 10⁴ cells/well) were stimulated for 1 h in Tyrode’s buffer supplemented with vehicle control, DNP-HSA in increasing concentrations at 37°C. Supernatants and cell lysates prepared in Tyrode’s buffer supplemented with 0.5% Triton X-100 were incubated with p-nitrophenyl N-acetyl-β-D-glucosamide (1.3 mg/ml) for 1 h at 37°C and reactions were stopped with the addition of 0.2 M Glycine buffer (pH 10.7). A colorimetric assay was performed to measure β-hexosaminidase activity to calculate the % release (supernatant) determined as a percentage of the total activity in the supernatant and pellet.

Samples were boiled for 10 min and resolved on 8–12% SDS-PAGE gels. The following antibodies were used: Rabbit anti-Lyn (2796), rabbit anti-phospho-Lyn 507 (2731), rabbit anti-Syk (2712), rabbit anti-phospho-Syk (2710), rabbit anti-Cbl-b (9498), rabbit anti-AKT (2731), rabbit anti-phospho-AKT (9271), rabbit anti-PLCγ1 (5690), rabbit anti-phospho-PLCγ1 (2821), rabbit anti-IKKα (2682), rabbit anti-phospho-IKKα/β (2697), rabbit anti-phospho-Erk1/2 (9101), rabbit anti-phospo-P38 MAPK (9211), rabbit anti-p38MAPK (9212), rabbit anti-phospho-JNK1/2 (9251), rabbit anti-JNK (9252) were all from Cell signaling technology. Mouse anti–FcεRIβ (H 5), mouse anti-pTyr (pY99), goat anti-SLAP (C19), and mouse anti-Cbl-b (G1) were from Santacruz Biotechnology, rabbit anti-FcεRIγ (06-727) were from Millipore. Mouse anti-tubulin (T6034, Sigma-Aldrich), and Mouse Anti-ERK2 (Clone 33/ERK2, BD transduction Lab) were also used. Secondary Abs were HRP-conjugated donkey anti-goat IgG (Jackson immunoresearch lab), HRP-conjugated goat anti-mouse IgG, HRP-conjugated goat anti-rabbit IgG (Cell signaling technology).

BMMCs (WT and SLAPKO; 5 × 10⁶ cells/ml per sample) were starved and sensitized with 20% (v/v) anti-DNP-IgE conditioned medium (SPE7 clone; ≈1 μg/ml) for 18 h, and then stimulated (2.5 × 10⁶ cells/ml) with either vehicle control or 10 ng/ml DNP-HSA for 6 h at 37°C. Cells supernatants were analyzed to quantify the amount of IL-6, MCP-1, and TNFα by ELISA kits (ebiosciences) as per manufacture instructions.

IgE sensitized BMMCs (WT and SLAPKO) were stimulated with DNP-HSA (30 ng/ml) for 0, 2, 4, 6, 8, and 10 min and fixed in ice cold 2% paraformaldehyde/5 mM EGTA/5 mM EDTA and subsequently treated with permeabilization buffer (PBS/0.1% saponin/2% BSA). Polymerized actin was quantified with staining for phalloidin (10 nM; molecular probes) and analyzed by flow cytometry as previously described (56, 57).

WT (n = 6) and SLAP KO mice (n = 7) mice from two independent experiments were sensitized with 2 μg anti-dinitrophenyl-IgE (Sigma-Aldrich) by i.v. injection. After 48 h, dinitrofluorobenzene [20 μl volume of 0.2% (w/v) of DNFB (Sigma-Aldrich); in acetone-olive oil] was applied on the right ear while the left ear was treated with the vehicle alone as previously described (58). Difference in ear thickness (Δ ear thickness) between DNFB- and vehicle-treated skin was quantified using digital calipers. H&E stained images of DNFB- and vehicle-treated ear skin were acquired with an EVOS FL color imaging system (Thermo Fisher scientific) with 20 × objective (air).

Starved and sensitized BMMCs [WT and SLAP KO (20 × 10⁶ cells)] were washed twice in PBS and then re-suspended in RPMI (Phenol red free) supplemented with 10 mM HEPES 7.4, 1 mM sodium pyruvate and 1% (v/v) minimum non-essential amino acids for 4 h. Subsequently, cells were stimulated in Tyrode’s buffer with vehicle or DNP-HSA (30 ng/ml) for indicated time-points and fixed with 2% paraformaldehyde for 10 min at 37°C, followed by washing with PB (PBS supplemented with 1% bovine serum albumin) buffer twice before permeabilization with cold (−20°C) BD phosphoflow perm buffer III for 30 min. After permeabilization, cells were stained with antibody solution (1:100 dilution) for 30 min at room temperature before analysis on a cytometer. The following BD Bioscience antibodies for phosphoflow analysis were used: pSyk (pY348, PE; clone 1120-722) and pNFkB p65 (S529, PE; clone K10-895.12.5).

Starved and sensitized BMMCs (WT and SLAP KO; 60 × 10⁶) were incubated with MG132 (25 mM; Merck) and chloroquine (50 mM; Sigma-Aldrich) for 2 h before DNP–HSA stimulation. BMMCs (20 × 10⁶ cells per sample) were lysed in 500 ml ice-cold ubiquitin lysis buffer [50 mM Tris-Cl, pH 7.5, 5 mM EDTA, 10 mM Na₄P₂O₇, 100 mM NaCl, 1% Triton X-100, protease inhibitor, phosphatase inhibitors and 20 μM deubiquitlyase inhibitor (PR-619; lifesensors]. Agarose-Tandem Ubiquitin Binding Entities (TUBEs) (Lifesensors) (20 μl/sample) were incubated with cell lysates at 4°C for 2 h and were then washed three times with cold tris-buffered saline with Tween-20 (TBS-T). Bound ubiquitinated proteins were eluted and resolved with SDS–PAGE, followed by immunoblot with FcεRIβ, and FcεRIγ chain antibodies.

Total RNA was isolated using the RNEasy mini kit (Qiagen) and reverse transcribed with the High-Capacity cDNA Reverse Transcription kit (Applied Biosystems) per the manufacturer’s instructions. For quantitative PCR analysis, 1 μL cDNA was combined with SsoAdvanced Universal SYBR Green Supermix (BioRad) and 0.5 μM forward and reverse primers. Reactions were set up in triplicate and run on an Applied Biosystems StepOnePlus thermal cycler, using cycling conditions described by the manufacturer and an annealing temperature of 60°C. Reactions were normalized to a GAPDH expression to calculate ΔCt values. The following primers were used for the qPCR reaction: FcεRIα (For:AGAGCAAACCTGTGTACTTG, Rev:GTGGACTTTCCATTTCTTCC), FcεRIβ (For: AGGCTACCCATTCTGGGGTG, Rev: GGCTGCCTCTCACCAGATAC), FcεRIγ (For: CTCCTTTTGGTGGAAGAAGC, Rev:TGAGTCGACAGTAGAGCAGG), MCPT1 (For:TTCCCTTGCCTGGTCCCT, Rev:GTTTTCCCCCAGCCAGCT).

Starved and sensitized BMMCs (WT and SLAP KO; 2 × 10⁶/sample) loaded with Indo-1AM as described previously (59). In 96 well-black plate with clear bottom, cells were stimulated with vehicle, DNP-HSA (20 ng/ml), and ionomycin (5 μM) in presence or absence of EGTA (10 nM). Cells loaded with Indo-1AM were excited with a UV lamp tuned to 338 nm while capturing emissions at 405 and 480 nm every 20 ms at Molecular Devices SpectraMax Gemini EM. The concentration of intracellular free Ca²⁺ was determined as previously (59).

GST or GST-SLAP fusion proteins were purified as described previously (14). Fusion proteins were quantified by SDS-PAGE followed by Coomassie staining and compared with BSA standards. For GST pull down assay, 500 μg of RBL-2H3 protein lysate was incubated with GST or GST-SLAP bound beads (2 μg/sample) for 2 h at 4°C. Beads were washed three times with PLC buffer and samples were resolved on 10% SDS-PAGE gel and immunoblotted with anti- pY, anti- FcεRIβ chain (Santacruz Biotechnology), and anti-Cbl-b (Cell Signaling Technology) antibodies.

Blood samples were obtained from atopic children aged 4–6 years with informed consent and according to the protocol approved by the Research Ethics Board of the Hospital for Sick Children (REB#1000041089). After blocking with Fc block (BD biosciences 564220) cells were subjected to stimulation with stimulation buffer (Bühlmann B-CCR-STB) or anti-FcεRI mAb (Bühlmann B-CCR-STCON), or with relevant (allergic) or irrelevant (not allergic) allergen (100 ng/mL), and stained for the surface marker CCR3 (Biolegend 310708) simultaneously for 15 min at 37°C. Thereafter the reaction was stopped with fix/lysis buffer 5X (BD biosciences 558049) for 10 min at 37°C and cells were treated with permeabilization buffer III (BD biosciences 558050) overnight at −20°C. Cells were then incubated for 40 min at 4°C with mouse anti-SLAP mAb (sc-135802) and for 20 min with Cy3-conjugated goat anti-mouse secondary antibody. SLAP median fluorescence in basophils (n = 500; SSC^low/CCR3^pos) was quantified on a CytoFLEX flow cytometer (Beckman Coulter).

For analyzing the flow cytometry data, flowjo 10.0.8 was employed. BMMC phospho-flow data was further analyzed by Cytobank software to quantify the fold change in phospho-profile. A two-tailed unpaired Student’s t-test was used for all statistical analyses except the human basophil analysis where a Wilcoxon match paired t-test was applied. Graph Pad prism software was used to produce graphs and for the quantification of statistical significance (^*P < 0.05 and ^**P < 0.01).

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.