CB2 receptors regulate natural killer cells that limit allergic airway inflammation in a murine model of asthma

Abstract

Background: Allergic asthma is a chronic airway inflammatory disease involving the complementary actions of innate and adaptive immune responses. Endogenously generated cannabinoids acting via CB2 receptors play important roles in both homeostatic and inflammatory processes. However, the contribution of CB2-acting eicosanoids to the innate events preceding sensitization to the common house dust mite (HDM) allergen remains to be elucidated. We investigated the role of CB2 activation during allergen-induced pulmonary inflammation and natural killer (NK) cell effector function.

Methods: Lung mucosal responses in CB2-deficient (CB2^-/- ) mice were examined and compared with wild-type (WT) littermates following intranasal exposure to HDM allergen.

Results: Mice lacking CB2 receptors exhibited elevated numbers of pulmonary NK cells yet were resistant to the induction of allergic inflammation exemplified by diminished airway eosinophilia, type 2 cytokine production and mucus secretion after allergen inhalation. This phenomenon was corroborated when WT mice were treated with a CB2-specific antagonist that caused a pronounced inhibition of HDM-induced airway inflammation and goblet cell hyperplasia. Unexpectedly, the preponderance of NK cells in the lungs of CB2^-/- mice correlated with reduced numbers of group 2 innate lymphoid cells (ILC2s). Depletion of NK cells restored the allergen responsiveness in the lungs and was associated with elevated ILC2 numbers.

Conclusions: Collectively, these results reveal that CB2 activation is crucial in regulating pulmonary NK cell function, and suggest that NK cells serve to limit ILC2 activation and subsequent allergic airway inflammation. CB2 inhibition may present an important target to modulate NK cell response during pulmonary inflammation.

Keywords: CB2 receptors; animal models; asthma; innate immunity; natural killer cells.

Figure 1

CB2^−/− mice exhibit elevated numbers of pulmonary natural killer (NK) cells and a defect in HDM-induced airway inflammation. (A) Lungs from naïve CB2^−/− mice or C57BL/6 wild type (WT) animals (6 mice per group) were collected and lung mononuclear cells (LMC) isolated by collagenase dispersion of lung tissue. The number of LMC CD3⁻CD19⁻NK1.1⁺ NK cells expressing DX5 was determined using flow cytometry (by gating on CD3⁻CD19⁻ cells). Results denote mean ±SEM (n=6) and expressed as total number of lung NK cells per mouse (**p<0.01). (B) Bronchoalveolar lavage fluid (BALF) from CB2^−/− or wild type (WT) mice that were challenged by intranasal instillation of house dust mite (HDM) allergen or PBS (control) was collected and cell differential counts determined and expressed as absolute cell numbers per mouse of lymphocytes (LYM), macrophages (MAC), eosinophils (EOS), and polymorphonuclear neutrophils (PMN). Cell-associated eosinophil peroxidase (EPO) levels were assessed by colorimetric analysis. Results are mean ±SEM (n=6), **p<0.01 and ***p<0.001, and representative of four independent experiments. (C) Number of CD11b⁺Siglec-F⁺Gr1⁻ eosinophils (by gating on CD11b⁺ & F4/80⁻ cells) and (D) CD3⁺CD4⁺ T cells were determined by flow cytometry. (E) Peribronchial inflammation and mucus production were determined using hematoxylin and eosin (H&E) stain and periodic acid-Schiff (PAS) stain, respectively (20x). Data are representative of two to four independent experiments.

Figure 2

CB2 antagonist AM630 inhibits allergen-induced airway inflammation. C57BL/6 mice (6 per group) were challenged with house dust mite (HDM) allergen and either untreated (HDM group) or treated intranasally with AM630 (HDM + AM630 group). Control mice did not inhale HDM allergen but were exposed to PBS and either untreated (PBS control) or treated with AM630 (PBS + AM630). (A) Bronchoalveolar lavage fluid (BALF) was collected and cell differential counts expressed as absolute cell numbers per mouse of lymphocytes (LYM), macrophages (MAC), eosinophils (EOS), and polymorphonuclear neutrophils (PMN). Cell-associated eosinophil peroxidase (EPO) levels were assessed by colorimetric analysis. Data are mean ±SEM (n=6), ***p<0.001 and **p<0.01. (B) Number of CD11b⁺Siglec-F⁺Gr1⁻ eosinophils (by gating on CD11b⁺ & F4/80⁻ cells) or CD3⁺CD4⁺ T cells analyzed by flow cytometry and representative of three independent experiments. (C) Peribronchial inflammation and mucus production determined using hematoxylin and eosin (H&E) stain and periodic acid-Schiff (PAS) stain, respectively (20x). Results are representative of three or four independent experiments.

Figure 3

CB2^−/− mice exhibit elevated number of IFN-γ-producing pulmonary natural killer (NK) cells but reduced monocyte-derived dendritic cells (DC) numbers after allergen inhalation: effect of cannabinoids on IFN-γ production by NK cells. CB2^−/− or wild type (WT) mice (6 per group) were challenged by intranasal instillation of house dust mite (HDM) allergen or PBS (control). (A) Number of CD11b⁺CD11c⁺MHCII^bright DC (by gating on CD11b⁺ cells) and (B) CD3⁻CD19⁻NK1.1⁺ NK cells expressing DX5 or NKp46 (by gating on CD3⁻CD19⁻ cells) in the lung mononuclear cells (LMC) prepared by collagenase dispersion of lung tissue was determined by flow cytometry. Results denote mean ±SEM (n=6) and expressed as total number of CD11c⁺MHC-II^bright DC or CD3⁻CD19⁻NK1.1⁺ NK cells per mouse (**p<0.01). (C) IFN-γ production by LMC from WT or CB2^−/− mice stimulated in vitro with anti-NK1.1 antibody (PK136, 20 μg/ml) for 24h was determined using ELISA. (D) To examine the effect of the cannabinoid 2-AG on cytokine production by NK cells, LMC from C57BL/6 mice were stimulated in vitro with anti-NK1.1 antibody (20μg/ml) in the absence (vehicle control) or presence of 2-AG (2 μM and 10 μM) for 24h. Following culture, IFN-γ production was determined in the supernatant using ELISA. Results are mean ±SEM (n=6), *p<0.05 and **p<0.01. Data are representative of three or four independent experiments.

Figure 4

CB2^−/− mice exhibit impaired type 2 cytokine and CCL2 production after allergen inhalation. CB2^−/− or wild type (WT) mice (6 per group) were challenged with either intranasal house dust mite (HDM) allergen or PBS (control). (A) IL-4, IL-5, IL-13 and IFN-γ, and (B) CCL2 levels were measured in the bronchoalveolar lavage fluid (BALF) using ELISA or V-PLEX assay. Results are mean ±SEM (n=6), **p<0.01. Data are representative of three or four independent experiments.

Figure 5

Transfer of CB2^−/− natural killer (NK) cells suppress house dust mite (HDM) allergen-induced airway inflammation. Purified CD3⁻CD19⁻NK1.1⁺ NK cells from CB2^−/− mice were adoptively transferred into wild type (WT) hosts (6 mice per group) that were then challenged with intranasal HDM allergen (HDM + CB2^−/− NK) or PBS (PBS + CB2^−/− NK). Allergic inflammation was compared with HDM-challenged Sham group that did not receive NK cells. (A) Bronchoalveolar lavage fluid (BALF) was collected and cell differential counts were determined and expressed as absolute numbers of lymphocytes (LYM), macrophages (MAC), eosinophils (EOS) and polymorphonuclear neutrophils (PMN). Cell-associated eosinophil peroxidase (EPO) levels were assessed by colorimetric analysis. Results are mean ±SEM (n=6), **p<0.01 and *p<0.05. (B) CD3⁺CD4⁺ T cell numbers in the BALF, and (C) CD11b⁺CD11c⁺MHCII^bright dendritic cell (DC) numbers, and (D) CD3⁻CD19⁻NK1.1⁺DX5⁺ NK cell numbers in lung mononuclear cells (LMC) analyzed by flow cytometry. Data are representative of three independent experiments.

Figure 6

Depletion of natural killer (NK) cells in vivo restores allergen-induced airway inflammation in CB2^−/− mice. CB2^−/− or wild type (WT) mice (6 per group) were challenged to house dust mite (HDM) allergen, and either depleted of NK cells by treatment with anti-NK1.1 depleting antibody (PK136 clone) or treated with isotype IgG2a control (control Ig). (A) Bronchoalveolar lavage fluid (BALF) was collected and cell differential counts were expressed as absolute numbers per mouse of lymphocytes (LYM), macrophages (MAC), eosinophils (EOS) and polymorphonuclear neutrophils (PMN). Cell-associated eosinophil peroxidase (EPO) and cytokine IL-13 levels were measured in the BALF. Results are mean ±SEM (n=6), ***p<0.001 and *p<0.05, (ns = not significant). (B) Peribronchial inflammation and goblet cell hyperplasia were determined using hematoxylin and eosin (H&E) stain and periodic acid-Schiff (PAS) stain, respectively (20x). (C) Inverse correlation between lineage negative (Lin⁻) CD45⁺CD90.2⁺ ILC2 numbers expressing IL-33Rα and (D) CD3⁻CD19⁻NK1.1⁺DX5⁺ NK cell numbers in lung mononuclear cells (LMC) of naïve CB2^−/− compared to WT mice. (E) Total number of Lin⁻CD45⁺CD90.2⁺IL-33Rα⁺ ILC2s in LMC of HDM-challenged WT and CB2^−/− mice that were treated with control Ig or anti-NK1.1 antibody (expressed as cell number per mouse). Results are mean ±SEM (n=6), **p<0.01. (F) Number of Lin⁻CD45⁺CD90.2⁺ ILC2 expressing IL-13 in LMC of HDM-challenged WT or CB2^−/− mice that were treated with control Ig or anti-NK1.1 antibody was determined by intracellular staining using flow cytometry. Data are representative of three independent experiments.