Stimulus strength determines the BTK-dependence of the SHIP1-deficient phenotype in IgE/antigen-triggered mast cells

Abstract

Antigen (Ag)-mediated crosslinking of IgE-loaded high-affinity receptors for IgE (FcεRI) on mast cells (MCs) triggers activation of proinflammatory effector functions relevant for IgE-associated allergic disorders. The cytosolic tyrosine kinase BTK and the SH2-containing inositol-5'-phosphatase SHIP1 are central positive and negative regulators of Ag-triggered MC activation, respectively, contrarily controlling Ca²⁺ mobilisation, degranulation, and cytokine production. Using genetic and pharmacological techniques, we examined whether BTK activation in Ship1-/- MCs is mandatory for the manifestation of the well-known hyperactive phenotype of Ship1-/- MCs. We demonstrate the prominence of BTK for the Ship1-/- phenotype in a manner strictly dependent on the strength of the initial Ag stimulus; particular importance for BTK was identified in Ship1-/- bone marrow-derived MCs in response to stimulation with suboptimal Ag concentrations. With respect to MAPK activation, BTK showed particular importance at suboptimal Ag concentrations, allowing for an analogous-to-digital switch resulting in full activation of ERK1/2 already at low Ag concentrations. Our data allow for a more precise definition of the role of BTK in FcεRI-mediated signal transduction and effector function in MCs. Moreover, they suggest that reduced activation or curtate expression of SHIP1 can be compensated by pharmacological inhibition of BTK and vice versa.

Figure 1

Antigen-triggered phosphorylation of BTK and PLCγ in BMMCs is independent of PI3K. (a) WT BMMCs were pre-treated for 20 min with 100 nM Wortmannin (WM) or vehicle (DMSO) and were then left untreated (con) or stimulated with the indicated concentrations of Ag (DNP-HSA [DNP]) for 1 and 5 min. Btk−/− BMMCs were pre-treated with DMSO and stimulated accordingly. Whole-cell lysates were subjected to Western Blot analysis with antibodies against P-BTK (Tyr223), P-PKB (Ser473), and p85 (loading control). (b) WT BMMCs were pre-treated for 20 min with 100 nM WM or vehicle (DMSO) and were then left untreated (con) or stimulated with Ag (20 ng/ml DNP-HSA) for the indicated times. Btk−/− BMMCs were pre-treated with DMSO and stimulated accordingly. Whole-cell lysates were subjected to Western Blot analysis with antibodies against P-BTK (Tyr551), BTK, P-PKB (Ser473), and p85 (loading control). (c) WT and Btk−/− BMMCs were pre-treated for 20 min with 100 nM WM or vehicle (DMSO) and were then left untreated (con) or stimulated with the indicated concentrations of Ag for 1 and 5 min. Whole-cell lysates were analysed by immunoblotting against P-PLCγ1 (Tyr783), PLCγ1, P-PKB (Ser473), and p85 (loading control). Densitometry of P-PLCγ1 with reference to PLCγ1 was performed using ImageJ software (NIH, USA) and normalised to WT cells pretreated with DMSO and stimulated with 1 ng/ml Ag (1 min). Comparable results were obtained in at least three experiments with different BMMC cultures. Fine vertical lines were inserted to allow for better discrimination of cells/conditions.

Figure 2

Degranulation of Ship1−/− BMMCs in response to suboptimal Ag concentrations is strictly dependent on BTK. (a) WT (red), Btk−/− (blue), Ship1−/− (green), and DKO (yellow) BMMCs were left untreated (con) or stimulated with increasing concentrations of Ag (DNP-HSA) and analysed for the amount of degranulation. Each bar is the mean of duplicates ± SD. (b) WT and Ship1−/− BMMCs were pre-treated for 30 min with 0.3 µM of the BTK-Inhibitor Ibrutinib or vehicle (DMSO), subsequently left untreated (con) or stimulated with 2 or 20 ng/ml Ag and analysed for the amount of degranulation. Each bar is the mean of triplicates ± SD. (c) Ship1−/− BMMCs pre-treated with Ibrutinib as in (b), were left untreated (con) or stimulated with the indicated amounts of IgE (SPE-7) for the analyses of degranulation. Each bar is the mean of triplicates ± SD. (d) Ca²⁺ mobilisation was measured in WT, Btk−/−, Ship1−/−, and DKO BMMCs upon stimulation with 20 ng/ml DNP-HSA. (e) WT, Btk−/−, Ship1−/−, and DKO BMMCs were stimulated with suboptimal (1 ng/ml) and optimal (20 ng/ml) Ag concentrations for 1 and 5 min or were left untreated (con). Whole-cell lysates were subjected to immunoblot analysis with antibodies against P-PLCγ1 (Tyr783), P-PLCγ2 (Tyr759), P-LAT (Tyr191), and p85 (loading control). (f) WT and Ship1−/− BMMCs were pre-treated for 30 min with 0.3 µM Ibrutinib or vehicle (DMSO) and then stimulated with the indicated concentrations of Ag. Whole-cell lysates were analysed with antibodies against P-PLCγ1 (Tyr783), P-LAT (Tyr191), and p85 (loading control). Fine vertical lines were inserted to allow for better discrimination of cells/conditions. Because proteins of comparable size were analysed, the same lysates were separated on two gels with an anti-p85 loading control on each gel (marked with/without asterisk). Comparable results were obtained in at least three experiments with different BMMC cultures. Statistical data were analysed for (a) n = 6, (b) n = 3, and (c) n = 3 experiments with n.s. (non significant), *p < 0.05, **p < 0.005, ***p < 0.0005.

Figure 3

Btk deficiency impacts on Ag-induced cytokine secretion in a dose-dependent manner. (a,b) WT (red), Btk−/− (blue), Ship1−/− (green), and DKO (yellow) BMMCs or (c,d) WT and Ship1−/− BMMCs, pre-treated for 30 min with 0.3 µM Ibrutinib or vehicle (DMSO), were left untreated (con) or stimulated with the indicated concentrations of Ag (DNP-HSA). (e,f) WT and Ship1−/− BMMCs were pre-loaded with IgE overnight (for the following DNP-HSA stimulation) or left without IgE (for the following IgE stimulation). Subsequently, the cells were treated for 30 min with 0.3 µM Ibrutinib or vehicle (DMSO). Cells were then left unstimulated (con) or stimulated with Ag (DNP-HSA [DNP]; 2 ng/ml) or with 2 µg/ml IgE (SPE-7). Supernatants were subjected to IL-6 (a,c,e) and TNF-α (b,d,f) ELISAs. Each bar is the mean of triplicates ± SD; comparable results were obtained in at least three experiments with different BMMC cultures. Statistical data were analysed for (a) n = 5, (b) n = 4, (c) n = 6 or n = 4 (2000 ng/ml DNP-HSA), (d) n = 5 or n = 3 (2000 ng/ml DNP-HSA), (e) and (f) n = 3 experiments with n.s. (non significant), *p < 0.05, **p < 0.005, ***p < 0.0005.

Figure 4

SHIP1-deficiency compensates for signalling defects in BTK-deficient BMMCs. (a) WT, Btk−/−, Ship1−/−, and DKO BMMCs or (b) WT and Ship1−/− BMMCs, pre-treated with 0.3 µM Ibrutinib or vehicle (DMSO) for 30 min, were left untreated (con) or stimulated with suboptimal and optimal concentrations of Ag (DNP-HSA [DNP]) for 1 and 5 minutes. Whole-cell lysates were subjected to Western Blot analysis with the indicated antibodies against P-IKKα/β (Ser176/Ser177), P-PKB (Ser473), P-ERK1/2 (Thr202/Tyr204), P-p38 (Thr180/Tyr182), P-IκBα (Ser32), and p85 (loading control). Comparable results were obtained in at least three experiments with different BMMC cultures. Fine vertical lines were inserted to allow for better discrimination of cells/conditions. Because proteins of comparable size were analysed, the same lysates were separated on two gels with an anti-p85 loading control on each gel (a; marked with/without asterisk).

Figure 5

BTK exerts an activating and PI3K a permissive role on ERK1/2 activation in MCs stimulated by suboptimal FcεRI crosslinking. (a,b) WT (+) and Btk−/− (−) PMCs were left untreated (con) or stimulated for 1 and 5 min or (b) for 5 min with the indicated amounts of Ag (DNP-HSA [DNP]). (c) WT BMMCs were pre-treated for 30 min with 1 µM of the PLC-inhibitor U73122, its inactive compound U73343, or their vehicle (DMSO) and subsequently stimulated for 1 min with suboptimal (0.5 ng/ml) and optimal (20 ng/ml) concentrations of Ag. (d) WT and Btk−/− BMMCs were left untreated (con) or either stimulated for 1 min in parallel with 2 and 20 ng/ml Ag and increasing amounts of PMA (ng/ml) or vehicle (DMSO). Between lanes 12 and 13, no lysate was applied. (e) WT and Btk−/− BMMCs were pre-treated for 20 min with 100 nM Wortmannin (WM) before stimulation with 1 and 20 ng/ml Ag for 1 and 5 min. (f,g) WT BMMCs were pre-treated for 30 min with 0.3 µM Ibrutinib (f) or 20 min with 100 nM Wortmannin (g) before stimulation for 1 and 5 min with 2 µg/ml IgE (SPE-7). (a–g) Whole cell-lysates were subjected to immunoblot analysis with the respective antibodies against P-PKB (Ser473), P-ERK1/2 (Thr202/Tyr204), and p85 or GAPDH as loading controls. Comparable results were obtained in at least three experiments with different PMC or BMMC cultures. Fine vertical lines were inserted to allow for better discrimination of cells/conditions. (h) Model integrating the positive and negative effects of BTK and PI3K, respectively, on activation of ERK1/2. KO phenotypes and pharmacological activators/inhibitors are highlighted in fawn.

Figure 6

Effects of BTK- and SHIP1-deficiency on Ag-triggered effector functions. WT, Btk−/− (or WT cells + Ibrutinib), Ship1−/−, and DKO MCs (or Ship1−/− cells + Ibrutinib) are opposed to each other in an exemplary manner. Extent of FcεRI crosslinking is specified as suboptimal (sub) or optimal (opt). The level of induction of the analysed effector functions, degranulation and cytokine production, is represented by letter size (deg < Deg < DEG; cyto < Cyto < CYTO).