Pulmonary receptor for advanced glycation end-products promotes asthma pathogenesis through IL-33 and accumulation of group 2 innate lymphoid cells

Abstract

Background: Single nucleotide polymorphisms in the human gene for the receptor for advanced glycation end-products (RAGE) are associated with an increased incidence of asthma. RAGE is highly expressed in the lung and has been reported to play a vital role in the pathogenesis of murine models of asthma/allergic airway inflammation (AAI) by promoting expression of the type 2 cytokines IL-5 and IL-13. IL-5 and IL-13 are prominently secreted by group 2 innate lymphoid cells (ILC2s), which are stimulated by the proallergic cytokine IL-33.

Objective: We sought to test the hypothesis that pulmonary RAGE is necessary for allergen-induced ILC2 accumulation in the lung.

Methods: AAI was induced in wild-type and RAGE knockout mice by using IL-33, house dust mite extract, or Alternaria alternata extract. RAGE's lung-specific role in type 2 responses was explored with bone marrow chimeras and induction of gastrointestinal type 2 immune responses.

Results: RAGE was found to drive AAI by promoting IL-33 expression in response to allergen and by coordinating the inflammatory response downstream of IL-33. Absence of RAGE impedes pulmonary accumulation of ILC2s in models of AAI. Bone marrow chimera studies suggest that pulmonary parenchymal, but not hematopoietic, RAGE has a central role in promoting AAI. In contrast to the lung, the absence of RAGE does not affect IL-33-induced ILC2 influx in the spleen, type 2 cytokine production in the peritoneum, or mucus hypersecretion in the gastrointestinal tract.

Conclusions: For the first time, this study demonstrates that a parenchymal factor, RAGE, mediates lung-specific accumulation of ILC2s.

Keywords: Asthma; IL-33; group 2 innate lymphoid cells; receptor for advanced glycation end-products.

Figure 1. IL-33 up-regulation in response to HDM extract is dependent on RAGE expression

(A) Immunoblot probing for IL-33 in whole lung homogenate after 7 weeks of HDM treatments and (B) summary of normalized IL-33:β-actin signal intensity ratios (saline control set arbitrarily to 1.0). (C) qRT-PCR of whole lung homogenate RNA after 7 weeks of HDM treatments probing for IL-33. IL-33 signal was normalized to the GAPDH signal and results are expressed as fold change over wild type saline values. In all cases, results are expressed as mean±S.E.M. (n=3–6 mice per strain/treatment group, *P < 0.05 versus comparison).

Figure 2. Absence of RAGE attenuates the cellular inflammatory and airway hyperreactivity responses to Alternaria alternata in an acute model of allergic airway disease

(A) Representative H&E and (B) PAS stains (200X magnification) of wild-type and RAGE KO mouse lung sections. (C) BALF total cell and eosinophil counts, normalized to 1 mL. (D) Pulmonary function test demonstrating changes airway resistance during methacholine challenge. All samples/functional analyses were collected on Day 10. Results are expressed as mean±S.E.M. (n=7–14 mice per strain/treatment group for A-C; n=3–4 mice per group for pulmonary function tests, *P < 0.05 versus comparison).

Figure 3. Absence of RAGE results in blunted IL-5, IL-13, and IL-33 cytokine responses to Alternaria alternata in an acute model of allergic airway disease

(A&B) ELISA analyses of IL-5 and IL-13 levels in the BALF at Day 10 in Alternaria model. (C) Immunoblot probing for IL-33 in whole lung homogenate after 10-day Alternaria treatment, and (right) summary of normalized IL-33:β-actin signal intensity ratios (saline controls set arbitrarily to 1.0). In all cases, results are expressed as mean±S.E.M. (n=7–14 mice per strain/treatment group, *P < 0.05 versus comparison).

Figure 4. Development of inflammatory effects in the lung downstream of exogenous IL-33 depends on RAGE

(A) Representative H&E and (B) PAS stains of lung sections (200X magnification) from wild type and RAGE KO mice following IL-33 administration. (C) BALF total cell and eosinophil counts, normalized to 1 mL. (D) ELISA analyses of IL-5 and IL-13 levels in the BALF after treatment with rIL-33. All samples were collected after four days of IL-33 treatment. In all cases, results are expressed as mean±S.E.M. (n=3–11 mice per strain/treatment group, *P < 0.05 versus comparison).

Figure 5. RAGE is required for ILC2 accumulation in the lung

(A) Representative flow plots of the live, lineage negative cell populations, gated for expression of CD90.2 and ST2. Numbers in each quadrant are percentages of parent population. ILC2s were identified as lineage negative cells that co-express CD90.2 and ST2. (B) Graphical summary of flow cytometrically-enumerated ILC2s in one lung after four days of treatment with rIL-33 or (C) ten days of treatment with Alternaria alternata. Results are expressed as mean±S.E.M. (n=3–11 mice per strain/treatment group, *P < 0.05 versus comparison).

Figure 6. Stromal, not hematopoietic, RAGE drives allergic airway inflammation

(A) BALF total cell counts and (B) eosinophil counts of bone marrow chimeric mice treated intranasally for 7 weeks with HDM extract or saline control. Treatment groups are labeled as: recipient strain (bone marrow donor strain). In all cases, results are expressed as mean±S.E.M. (n=3–7 mice per strain/treatment group, *P < 0.05 versus comparison).

Figure 7. RAGE is not required for ILC2 induction, proliferation, or secretory function in the alimentary tract after intraperitoneal rIL-33 administration

(A) Immunoblot probing for RAGE in mouse whole organ homogenates. (B) Representative PAS stains of mouse ileum sections (100X magnification). (C) IL-5 and IL-13 levels in peritoneal lavage fluid. (D) Representative lung sections stained with H&E (100X magnification). (E) ILC2 (Lineage-, CD90.2+, ST2+) cell counts in the spleen. All samples were collected after 3 days of IL-33 injections. Results are expressed as mean±S.E.M. (n=3–5 mice per strain/treatment group, *P < 0.05 versus comparison).