Dual therapy with corticosteroid ablates the beneficial effect of DP2 antagonism in chronic experimental asthma

Abstract

Prostaglandin D2 (PGD2) signals via the DP1 and DP2 receptors. In Phase II trials, DP2 antagonism decreased airway inflammation and airway smooth muscle (ASM) area in moderate-to-severe asthma patients. However, in Phase III, DP2 antagonism failed to lower the rate of exacerbations, and DP2 as a target was shelved. Here, using a preclinical model of chronic experimental asthma, we demonstrate that rhinovirus-induced exacerbations increase PGD2 release, mucus production, transforming growth factor (TGF)-β1 and type-2 inflammation. DP2 antagonism or DP1 agonism ablates these phenotypes, increases epithelial EGF expression and decreases ASM area via increased IFN-γ. In contrast, dual DP1-DP2 antagonism or dual corticosteroid/DP2 antagonism, which attenuates endogenous PGD2, prevented ASM resolution. We demonstrate that DP2 antagonism resolves ASM remodelling via PGD2/DP1-mediated upregulation of IFN-γ expression, and that dual DP2 antagonism/corticosteroid therapy, as often occurred in the human trials, impairs the efficacy of DP2 antagonism by suppressing endogenous PGD2 and IFN-γ production.

Fig. 1. Treatment with a DP2 antagonist, but not a corticosteroid, ameliorates airway remodeling during a RV-triggered exacerbation of CEA.

A Study design. Seven days old mice were inoculated with PVM and exposed to CRE 3 days later. Mice were re-infected with PVM six weeks later and exposed to CRE weekly for four weeks. Four weeks later, mice were inoculated with RV-1b. Separate groups of mice were treated with a DP2 antagonist (OC000459), DP1 agonist (BW245c), fluticasone, soluble IL-13Rα2 or vehicle. Mice exposed to vehicle instead of virus or allergen were referred to as naïve. Mice were then inoculated with RV-1b and euthanized at 1, 3 and 7 days post infection (dpi). B Lung PGD2 levels. C Representative lung histology of α-smooth muscle actin (SMA) expression. Scale bars = 50 µm. D, E Airway smooth muscle (ASM) area. F, G Collagen deposition. H, I Mucus score. Data are presented as mean ± SEM or box-and-whisker plots showing individual data points with the boxes representing quartiles and whiskers indicating the range and are pooled data from two independent experiments (n = 4–16 mice per group). Statistical significance between different time points or different groups was determined using one-way ANOVA with Dunnett’s multiple comparison test. * denotes p < 0.05, ** denotes p < 0.01 and *** denotes p < 0.001 compared to Vehicle group. # denotes p < 0.05, ## denotes p < 0.01 and ### denotes p < 0.001 compared to RV-infected group at corresponding time point. Source data are provided as a Source Data file.

Fig. 2. DP2 antagonism or DP1 agonism decreases lung TGF-β1 expression and increases EGF expression.

A TGF-β1 immunoreactivity (red coloration). Scale bars = 50 µm. B, C Quantification of TGF-β1 immunoreactivity around the airways. D, E Lung TGF-β1 levels in homogenates. F, G Lung EGF levels in homogenates. H EGF immunoreactivity (red coloration). Scale bars = 50 µm. I, J Quantification of EGF immunoreactivity in airway epithelial cells (AECs). Data are presented as mean ± SEM or box-and-whisker plots showing individual data points with the boxes representing quartiles and whiskers indicating the range and are pooled data from two independent experiments (n = 4–9 mice per group). Statistical significance between different time points or different groups was determined using one-way ANOVA with Dunnett’s multiple comparison test. ** denotes p < 0.01 and *** denotes p < 0.001 compared to vehicle group. ## denotes p < 0.01 and ### denotes p < 0.001 compared to RV-infected group at corresponding time point. Source data are provided as a Source Data file.

Fig. 3. DP2-antagonist mediated resolution of airway remodeling is dependent on endogenous PGD2 and ablated by dual therapy with a corticosteroid.

A–E Mice with CEA were Inoculated with RV-1b and given vehicle, DP2 antagonist or DP1 antagonist or both DP2 and DP1 antagonist daily from day 99 and euthanized at 7 dpi. Quantification of (A) ASM, (B) collagen deposition, (C) Mucus score and (D) TGF-β1 expression. Lung (E) EGF and (F) PGD2 levels in homogenates. G–L Mice with chronic asthma were inoculated with RV-1b and treated with vehicle, DP2 antagonist, fluticasone, or both drugs daily from day 99 and euthanized at 7 dpi. Quantification of (G) ASM, (H) collagen deposition, (I) Mucus score and (J) TGF-β1 expression. Lung (K) EGF and (L) PGD2 levels in homogenates. Data are presented as box-and-whisker plots showing individual data points with the boxes representing quartiles and whiskers indicating the range and are representative of two independent experiments (n = 5–7 mice per group). Statistical significance between different groups was determined using one-way ANOVA with Dunnett’s multiple comparison test. * denotes p < 0.05, ** denotes p < 0.01 and *** denotes p < 0.001 compared to vehicle group. δ denotes p < 0.05 and δδ denotes p < 0.01 compared to DP2 antagonist treated group. Source data are provided as a Source Data file.

Fig. 4. DP2 antagonism or DP1 agonism attenuates RV-induced type-2 inflammation and promotes type-1 immunity.

Mice with CEA were inoculated with RV-1b and treated daily with a DP2 antagonist, a DP1 agonist, or vehicle starting from day 99. Mice were euthanized at 1, 3 and 7 dpi. A Number of neutrophils (SigF⁻CD11b⁺Ly6G⁺CD11c⁻) and eosinophils (SigF⁺CD11b⁺Ly6G⁻CD11c⁻) in the lungs. B Number of T_H2 cells (GATA3⁺CD4⁺T) and ILC2s (CD3ε⁻CD19⁻CD45R⁻CD11c⁻Gr‐1⁻CD11b⁻NK1.1⁻ CD90.2⁺CD200R1⁺ GATA3⁺) in the lungs. C IL-4 and IL-5 concentration in the bronchoalveolar lavage fluid (BALF). D IL-17A expression in the BALF. E IFN-γ expression in the BALF and TNF expression in the lung. A–E Data are presented as mean ± SEM and are representative of two independent experiments showing similar results (n = 4–11 mice per group). Statistical significance between different time points or different groups was determined using one-way ANOVA with Dunnett’s multiple comparison test. * denotes p < 0.05; ** denotes p < 0.01 and *** denotes p < 0.001 compared to Vehicle group. # denotes p < 0.05, ## denotes p < 0.01 and ### denotes p < 0.001 compared to RV-infected group at corresponding time point. F RV-1b infected mice were treated with vehicle or DP2 antagonist (OC000459) or DP1 antagonist (MK0524) or both DP2 and DP1 antagonist daily from day 99 and euthanized at 7 dpi. Concentrations of IFN-γ in the BALF and TNF in the lung. G RV-1b infected mice were treated with vehicle, DP2 antagonist, fluticasone, or both DP2 antagonist and fluticasone daily from day 99 and euthanized at 7 dpi. Concentrations of IFN-γ in the BALF and TNF in the lung. F, G Data are presented as box-and-whisker plots showing individual data points with the boxes representing quartiles and whiskers indicating the range and are representative of two independent experiments showing similar results (n = 5–7 mice per group). * denotes p < 0.05 and ** denotes p < 0.01 compared to vehicle group. δ denotes p < 0.05, δδ denotes p < 0.01 and δδδ denotes p < 0.001 compared to DP2 antagonist treated group. Source data are provided as a Source Data file.

Fig. 5. DP2 antagonism or DP1 agonism enhances NK cell IFN-γ production and alters the phenotype of alveolar macrophages.

Mice with CEA were inoculated with RV-1b and treated with a DP2 antagonist or a DP1 agonist daily starting from day 99. Mice were euthanized at 1, 3 and 7 dpi. A Number of IFN-γ expressing cells in the lungs at 7 dpi. B Number of IFN-γ expressing CD4⁺ T, CD8⁺ T and NK cells in the lungs at 7 dpi. C Number of TNF expressing cells in the lungs at 7 dpi. D Number of TNF expressing CD4⁺ T, alveolar macrophages (AMs) and conventional type-2 dendritic cells (cDC2s) in the lungs at 7 dpi. E Representative flow cytometry plots showing TNF expressing AMs in the lungs. F Number of total AMs in the lungs. G Representative flow cytometry plots showing CD86 and CD206 expression on AMs and the total number of CD206-expressing AMs and CD86-expressing AMs. Data are presented as mean ± SEM or box-and-whisker plots showing individual data points with the boxes representing quartiles and whiskers indicating the range and are pooled data from two independent experiments showing similar results (n = 4–8 mice per group). Statistical significance between different time points or different groups was determined using one-way ANOVA with Dunnett’s multiple comparison test. * denotes p < 0.05; ** denotes p < 0.01 and *** denotes p < 0.001 compared to vehicle group at 0 dpi. # denotes p < 0.05, ## denotes p < 0.01 and ### denotes p < 0.001 compared to RV-infected group at corresponding time point. Source data are provided as a Source Data file.

Fig. 6. DP2 antagonism mediates airway remodeling resolution via enhanced IFN-γ.

Mice with CEA were inoculated with RV-1b mice and treated with vehicle or DP2 antagonist +/− anti-IFN-γ. Mice were euthanized at 7 dpi. Quantification of (A) ASM area and (B) collagen deposition and (C) Mucus score. (D) Lung TGF-β1 levels in homogenates. E Lung EGF levels in homogenates. F, G Number of CD86-expressing AM and CD206-expressing AMs in the lung. H Lung TNF levels. Data are presented as box-and-whisker plots showing individual data points with the boxes representing quartiles and whiskers indicating the range and are representative of two independent experiments showing similar results (n = 5–10 mice per group). Statistical significance between different groups was determined using one-way ANOVA with Dunnett’s multiple comparison test. * denotes p < 0.05; ** denotes p < 0.01 and *** denotes p < 0.001 compared to vehicle group. # denotes p < 0.05, ## denotes p < 0.01 and ### denotes p < 0.001 compared to DP2 antagonist-treated group. Source data are provided as a Source Data file.

Fig. 7. Proposed mechanism by which DP2 antagonism resolves ASM mass.

Box A PGD2, EGF, and TGF-β1 are expressed at low levels in mice with stable chronic experimental asthma. Box (B) Upon a virus-associated exacerbation, PGD2 levels increase, elevating type-2 inflammation, TGF-β1 expression, and ASM remodeling. Box (C) DP2 antagonism ameliorates the aforementioned phenotypes via increased IFN-γ production by NK and CD8⁺ T cells. Box (D) By suppressing the production of endogenous PGD2, corticosteroids ablate the PGD2/DP1 receptor-induced IFN-γ expression that mediates the resolution of ASM remodeling. Created in BioRender. Howard, D. (2023) BioRender.com/c86z618.