An osteoclastogenesis system, the RANKL/RANK signalling pathway, contributes to aggravated allergic inflammation

Abstract

Background and purpose: As an osteoclast differentiation factor, receptor activator of NF-κB ligand (RANKL) is produced by various immune cells and may be involved in the pathogenesis of osteoporosis and inflammation. Although RANKL is expressed in most immune cells and tissues, it is not clear how this might affect allergic inflammation.

Experimental approach: The roles of RANKL in allergic rhinitis (AR) were analysed in an ovalbumin (OVA)-induced animal model, human subjects, and a human mast cell line (HMC-1). Small interfering RNA experiments were performed in an OVA-induced AR model.

Key results: RANKL and RANKL receptor (RANK) were up-regulated in serum or nasal mucosal tissues of AR patients and AR mice. RANKL and RANK were colocalised in mast cells of nasal mucosa tissue. Depletion of RANKL by RANKL siRNA ameliorated AR symptoms and reduced AR-related biomarkers, including thymic stromal lymphopoietin (TSLP), IgE, histamine, and inflammatory cell infiltration, whereas recombinant RANKL increased AR responses and TSLP levels. In addition, functional deficiency of TSLP decreased AR responses induced by RANKL. In human mast cells, interaction of RANKL with RANK increased production of TSLP and inflammatory cytokines. Production of TSLP by RANKL stimulation was mediated through activation of the PI3K, MAPK, caspase-1, and NF-κB pathways. Furthermore, dexamethasone alleviated RANKL-induced inflammatory reactions in AR models.

Conclusion and implications: Collectively, these data suggest that RANKL may induce development of AR through up-regulation of TSLP.

Figure 1

RANKL is up‐regulated in AR patients and localized on mast cells. (a) RANKL in serum (n = 40) and (b) homogenized nasal mucosal tissues (n = 20) of AR patients was analysed by elisa. ^# P < 0.05; significantly different from normal. (c) RANKL (upper panel) and RANK (lower panel) expression in the nasal mucosa of AR patients was determined by western blot analysis. (d) Immunostaining for RANKL (green) and staining for C‐kit (red) is shown in the AR patients (magnification, ×294). Mice were sensitized on Days 1, 5, and 14 by i.p. injections of 100 μg of OVA emulsified with 20 mg of aluminium hydroxide and then challenged with 1.5 mg of OVA. (e) Immunostaining for RANKL (green) and staining for C‐kit (red) is shown in nasal mucosal tissues from normal mice and AR mice (magnification, ×294). (f) C‐kit⁺RANKL⁺ cells were counted in AR patients and animal tissues by three individuals (n = 5 per group). Data shown are the means ± SEM. ^# P < 0.05; significantly different from normal. (g) Co‐localization of C‐kit (red), RANKL (green), and RANK (purple) in the nasal mucosa of AR animal (magnification, ×100). Arrows indicate C‐kit⁺RANKL⁺RANK⁺ cells

Figure 2

Knockdown of RANKL alleviates AR symptoms in AR mice. Mice were sensitized on Days 1, 5, and 14 by i.p. injections of 100 μg of OVA emulsified in 20 mg of aluminium hydroxide, after which they received local injections of RANKL siRNA and control siRNA, or PBS alone into the nasal cavity, and were challenged with intranasal OVA for 10 days. (a) The production of RANKL in the nasal mucosal tissues of mice was determined by elisa and (b) western blot analysis. (c) The number of nose rubs that occurred 10 min after OVA intranasal provocation. (d) OVA‐specific IgE levels in serum were measured using the elisa method. (e) The levels of histamine in serum were measured by a histamine assay method. (f–h) The levels of cytokines in serum were measured using the elisa method. (i) OVA‐specific IgE levels in the nasal mucosa were measured by elisa. (j–n) The levels of cytokine in the nasal mucosa were measured by elisa. (o,p) The levels of CXCL2 and ICAM in nasal mucosa tissue were measured using elisa method. Data shown are the means ± SEM from n = 10 mice per group. ^# P < 0.05; significantly different from the OVA‐unsensitized mice. *P < 0.05; significantly different from the control siRNA OVA‐sensitized mice

Figure 3

Knockdown of RANKL reduces infiltration of eosinophils, mast cells, and Treg cells into the AR nasal mucosa tissues. Mice were sensitized on Days 1, 5, and 14 by i.p. injections of 100 μg of OVA emulsified in 20 mg of aluminium hydroxide, after which they received local injections of RANKL siRNA and control siRNA, or PBS alone into the nasal cavity, and were challenged with intranasal OVA for 10 days. (a) Nasal mucosa were stained with haematoxylin and eosin (h,e) for eosinophils, Alcian blue and Safranin O (a,s) for mast cells, and immunohistochemical diaminobenzidine stain (for Foxp3) for Treg cells. (b) Mast cells, eosinophils, and Foxp3+ Treg cells were counted by two individuals, after which five randomly selected tissue sections per mouse were counted. The absolute number of cells is shown as the mean ± SEM. ^# P < 0.05; significantly different from OVA‐unsensitized mice. *P < 0.05; significantly different from control siRNA OVA‐sensitized mice. Scale bar = 100 μm

Figure 4

Exogenous RANKL exaggerates Th2 immune responses in AR mice. We sensitized mice on Days 1, 5, and 14 by i.p. injections of 100 μg of OVA emulsified in 20 mg of aluminium hydroxide and challenged mice with 1.5 mg of OVA or recombinant mouse RANKL for 10 days. (a) The number of nose rubs that occurred 10 min after OVA or RANKL intranasal provocation. (b) IgE levels in the serum were measured by elisa. (c) The levels of histamine in serum were measured by a histamine assay method. (d–l) The levels of cytokines in the serum were measured by elisa. Data shown are the means ± SEM from n = 5 mice per group. ^# P < 0.05; significantly different from OVA‐unsensitized mice (PBS)

Figure 5

RANKL induces AR via increased TSLP levels. Mice were sensitized on Days 1, 5, and 14 by i.p. injections of 100 μg of OVA emulsified in 20 mg of aluminium hydroxide, after which they received local injections of TSLP siRNA and control siRNA, or PBS alone into the nasal cavity, and were challenged with intranasal RANKL for 10 days. (a) The number of the nose rubs that occurred 10 min after OVA intranasal provocation. (b) IgE levels in the serum were measured by elisa. (c) The levels of histamine in serum were measured as described. (d–h) The levels of cytokine in the serum were measured by elisa. Data shown are the means ± SEM from n = 5 mice per group. ^# P < 0.05; significantly different from OVA‐unsensitized mice. *P < 0.05; significantly different from control siRNA RANKL‐sensitized mice

Figure 6

RANKL induces production of TSLP from human mast cells. (a) RANK was expressed on the surface of human cord blood‐derived mast cells (Original magnification ×800) and HMC‐1 cells (Original magnification ×800). (b) HMC‐1 cells were stimulated with various concentrations of RANKL (1, 10, and 100 ng·ml⁻¹) for 24 hr, after which the production of TSLP in the supernatant was measured by elisa. (c) HMC‐1 cells (3 × 10⁵) were stimulated with RANKL (10 ng·ml⁻¹) for various times. The production of TSLP in the supernatant was measured by elisa. (d) HMC‐1 cells (3 × 10⁶) were treated with RANKL (10 ng·ml⁻¹) for various times, after which TSLP mRNA expression was analysed by RT‐PCR. (e) HMC‐1 cells were treated with RANKL (10 ng·ml⁻¹), RANKL neutralizing antibodies (0.5 μg·ml⁻¹), and/or IgG isotype for 24 hr, after which the production of TSLP in the supernatant was measured by elisa. (f) Human cord blood‐derived mast cells (HCBMC ) were treated with RANKL (10 ng·ml⁻¹), RANKL neutralizing antibodies (0.5 μg·ml⁻¹), and/or IgG isotype for 24 hr, after which the production of TSLP in the supernatant was measured by elisa. Data shown are the means ± SEM from n = 5 per group. ^# P < 0.05; significantly different from unstimulated cells' value. *P < 0.05; significantly different from RANKL value

Figure 7

Involvement of MAPKs, NF‐κB, PI3K, and caspase‐1 in RANKL‐induced TSLP production. (a) HMC‐1 cells (3× 10⁵) were pretreated with 0.2 μM SB203580 (SB), 10 μM SP600125 (SP), 1 μM PD98059 (PD), 10 μM PDTC, 1 μM wortmannin (WORT), or 0.05 μM caspase‐1 inhibitor (CAS‐1) for 1 hr, then incubated with 10 ng·ml⁻¹ RANKL for 24 hr. The production of TSLP was measured by elisa. Data shown are the means ± SEM from n = 5 per group. ^# P < 0.05; significantly different from unstimulated cells' value. *P < 0.05; significantly different from RANKL' value. (b–h) HMC‐1 cells (3 × 10⁶) were stimulated with 10 ng·ml⁻¹ of RANKL for the indicated times. The activation of PI3K, MAPKs, NF‐κB, and caspase‐1 were determined by Western blot analysis. (i) HMC‐1 cells were treated with RANKL (10 ng·ml⁻¹), RANKL neutralizing antibodies (0.5 μg·ml⁻¹), and/or IgG isotype for the indicated times, after which the activation of PI3K, MAPKs, NF‐κB, and caspase‐1 were determined by Western blot analysis. (j) Relative intensities of protein levels were quantified by densitometry. Data shown are the means ± SEM from n = 5 per group. ^# P < 0.05; significantly different from unstimulated cells' value. *P < 0.05; significantly different from RANKL value

Figure 8

Regulatory effect of dexamethasone in RANKL‐induced inflammatory reactions in vivo and in vitro. Mice were treated orally with dexamethasone (DEX; 5 mg·kg⁻¹) for 10 days before the intranasal RANKL challenge. (a) The number of the nasal rubs that occurred in the 10 min after the RANKL intranasal provocation. Serum was isolated from blood and then assayed for (b) IgE, (c) histamine, (d) TSLP, (e) IL‐1β, (f) IL‐4, (g) IL‐5, and (h) IL‐13. Data shown are the means ± SEM from n = 5 mice per group. (i) HMC‐1 cells (3 × 10⁵) were pretreated with dexamethasone for 1 hr, then incubated with 10 ng·ml⁻¹ of RANKL for 24 hr. The production of TSLP was measured by elisa. Data are represented as the mean ± SEM with n = 5 per group. (j) HMC‐1 cells (3 × 10⁶) were stimulated with 10 ng·ml⁻¹ of RANKL for the indicated times. The activation of PI3K, phosphorylated ERK (pERK), and caspase‐1 was determined by Western blot analysis. ^# P < 0.05; significantly different from the untreated group. *P < 0.05; significantly different from the RANKL‐treated group

Figure 9

Schematic diagram of the mechanisms involved in allergic inflammatory signalling by the RANKL/RANK system. RANKL/RANK stimulation activates the PI3K, MAPK, caspase‐1, and NF‐κB signalling pathways in mast cells. Activated caspase‐1 and NF‐κB induces the production and transcription of TSLP. Finally, RANKL/RANK‐induced TSLP increases the allergic reactions through the up‐regulation of inflammatory mediators