Modulatory role for retinoid-related orphan receptor alpha in allergen-induced lung inflammation

Abstract

Rationale: Nuclear receptors play a critical role in the regulation of inflammation, thus representing attractive targets for the treatment of asthma.

Objective: In this study, we assess the potential regulatory function of retinoid-related orphan receptor alpha (RORalpha) in the adaptive immune response using ovalbumin (OVA)-induced airway inflammation as a model.

Methods: Allergen-induced inflammation was compared between wild-type (WT) and staggerer (RORalpha(sg/sg)) mice, a natural mutant strain that is deficient in RORalpha expression.

Measurements and main results: Despite robust increases in OVA-specific IgE, RORalpha(sg/sg) mice developed significantly less pulmonary inflammation, mucous cell hyperplasia, and eosinophilia compared with similarly treated WT animals. Induction of Th2 cytokines, including interleukin (IL)-4, IL-5, and IL-13, was also significantly less in RORalpha(sg/sg) mice. Microarray analysis using lung RNA showed increased expression of many genes, previously implicated in inflammation, in OVA-treated WT mice. These include mucin Muc5b, the chloride channel calcium-activated 3 (Clca3), macrophage inflammatory protein (MIP) 1alpha and 1beta, eotaxin-2, serum amyloid A3 (Saa3), and insulin-like growth factor 1 (Igf1). These genes were induced to a greater extent in OVA-treated WT mice relative to RORalpha(sg/sg) mice.

Conclusions: Our study demonstrates that mice deficient in RORalpha exhibit an attenuated allergic inflammatory response, indicating that RORalpha plays a critical role in the development of Th2-driven allergic lung inflammation in mice, and suggests that this nuclear receptor should be further evaluated as a potential asthma target.

Figure 1.

Comparison of inflammatory scores between wild-type (WT) and RORα^sg/sg mice. Hematoxylin and eosin–stained sections of lungs from WT and RORα^sg/sg mice (n = 25–28 in each group) that were challenged by either saline or ovalbumin (OVA) were scored in an unbiased manner from 0 to 4 for the extent of inflammatory cell infiltration, as described in Methods. 0–1 indicates no or little inflammation, and 4 represents severe inflammation. Perivascular/peribronchiolar infiltrates (black bars) and intraalveolar infiltrates (gray bars) were scored separately. The average histology inflammatory scores from OVA- and saline-challenged WT and RORα^sg/sg mice were calculated and plotted. Significant differences between OVA-challenged WT and RORα^sg/sg mice are indicated (*p < 0.001; ^∼p < 0.001).

Figure 2.

OVA-induced mucous cell hyperplasia in airway epithelium of WT and RORα^sg/sg mice. Sections of airways from saline- and OVA-challenged WT and RORα^sg/sg mice (n = 25–28 in each group) were stained by periodic acid Schiff (PAS) and then scored in an unbiased manner as described in Methods. The average PAS scores were plotted. The difference between the average PAS score of OVA-challenged WT and RORα^sg/sg mice was statistically significant (*p < 0.001).

Figure 3.

Decreased OVA-induced inflammation in RORα^sg/sg mice correlates with reduced accumulation of inflammatory cells in bronchoalveolar lavage fluid (BALF). Mice (n = 25–28 in each group) were sensitized with OVA and 2 wk later challenged with either OVA or saline. Twenty-four hours after the final saline or OVA challenge, mice were killed, BALF collected, and the total number of inflammatory cells determined in a Coulter counter. A significant difference in the total number of inflammatory cells was observed between OVA-challenged WT and RORα^sg/sg mice (*p < 0.0001).

Figure 4.

Reduced infiltration of eosinophils and neutrophils in BALF from OVA-challenged RORα^sg/sg mice compared with WT mice. After saline and OVA exposure, BALF (n = 25–28 in each group) was collected, and the number of macrophages, neutrophils, and eosinophils were determined as described in Methods. (A) The number of macrophages decreased in both OVA-challenged WT and RORα^sg/sg mice (^p < 0.001; ^∼p < 0.001). Significant differences (*p < 0.0001) were observed in the number of (B) neutrophils and (C) eosinophils between OVA-challenged WT and OVA-challenged RORα^sg/sg mice.

Figure 5.

Reduced accumulation of various lymphocyte subpopulations in BALF from OVA-challenged RORα^sg/sg mice compared with WT mice. BALF was collected and analyzed for the total number of lymphocytes as described in Methods. Various lymphocyte subpopulations were examined by flow cytometry using fluorescein isothiocyanate– or phycoerythrin-conjugated CD4, CD3, CD8, and B220 antibodies. (A) Comparison of total number of lymphocytes. A significant difference in the total number of lymphocytes was observed between OVA-challenged WT and RORα^sg/sg mice (*p < 0.0001; n = 25–28 in each group). (B) Comparison of different lymphocyte subpopulations (n = 9–16 in each group). Significant differences were observed in the CD3⁺ (*p < 0.0001), CD3⁺CD4⁺ (^∼p < 0.0001), CD3⁺CD8⁺ (^#p < 0.0001), and B220⁺ (^p < 0.0001) lymphocytes between OVA-challenged WT mice and OVA-challenged RORα^sg/sg mice.

Figure 6.

Reduced susceptibility of RORα^sg/sg mice to OVA-induced inflammation correlates with decreased release of several cytokines/chemokines. BALF from saline- and OVA-challenged WT and RORα^sg/sg mice (n = 25–28 in each group) was collected and the level of (A) interleukin (IL)-13, (B) eotaxin, (C) TARC, (D) IL-4, and (E) IL-5 (E) analyzed by ELISA. Significant differences (*p < 0.0001) were observed in the level of these cytokines/chemokines between OVA-challenged WT and OVA-challenged RORα^sg/sg mice.

Figure 6.

Reduced susceptibility of RORα^sg/sg mice to OVA-induced inflammation correlates with decreased release of several cytokines/chemokines. BALF from saline- and OVA-challenged WT and RORα^sg/sg mice (n = 25–28 in each group) was collected and the level of (A) interleukin (IL)-13, (B) eotaxin, (C) TARC, (D) IL-4, and (E) IL-5 (E) analyzed by ELISA. Significant differences (*p < 0.0001) were observed in the level of these cytokines/chemokines between OVA-challenged WT and OVA-challenged RORα^sg/sg mice.

Figure 7.

Comparison of the level of IgE in serum from saline- and OVA-challenged WT and RORα^sg/sg mice. Sera from saline- and OVA-challenged WT and RORα^sg/sg mice (n = 12–16 in each group) were collected and assayed for (A) total IgE and (B) OVA-specific IgE as described in Methods. A significant difference (*p < 0.0001; »p < 0.001) was observed in total serum IgE between saline-challenged RORα^sg/sg and WT mice. A significant difference (*p < 0.0001) was observed in OVA-specific IgE between OVA-challenged WT and RORα^sg/sg mice and saline-challenged mice.

Figure 8.

OVA-induced changes in resistance (Rl) and airway hyperresponsiveness in RORα^sg/sg mice. Baseline and methacholine (MCh)-induced changes in Rl were measured in saline-exposed WT (filled squares; n = 16), saline-exposed RORα^sg/sg (open squares; n = 16), OVA-exposed WT (filled circles; n = 12), and OVA-exposed RORα^sg/sg (open circles; n = 12) mice. Five measurements were taken before MCh challenge, and measurements were taken every 20 s after each MCh dose. Data represent mean Rl at each interval ± SEM. *p < 0.05 compared with RORα^sg/sg group.

Figure 9.

Differential expression of Ccl17, Ccl24, Saa3, and Igf1 mRNAs in lungs from saline- and OVA-challenged WT and RORα^sg/sg mice. Levels of RNA were determined by real-time quantitative reverse transcriptase–polymerase chain reaction as described in Methods.