Suppression of Th17-polarized airway inflammation by rapamycin

Abstract

Because Th17-polarized airway inflammation correlates with poor control in bronchial asthma and is a feature of numerous other difficult-to-treat inflammatory lung diseases, new therapeutic approaches for this type of airway inflammation are necessary. We assessed different licensed anti-inflammatory agents with known or expected efficacy against Th17-polarization in mouse models of Th17-dependent airway inflammation. Upon intravenous transfer of in vitro derived Th17 cells and intranasal challenge with the corresponding antigen, we established acute and chronic murine models of Th17-polarised airway inflammation. Consecutively, we assessed the efficacy of methylprednisolone, roflumilast, azithromycin, AM80 and rapamycin against acute or chronic Th17-dependent airway inflammation. Quantifiers for Th17-associated inflammation comprised: bronchoalveolar lavage (BAL) differential cell counts, allergen-specific cytokine and immunoglobulin secretion, as well as flow cytometric phenotyping of pulmonary inflammatory cells. Only rapamycin proved effective against acute Th17-dependent airway inflammation, accompanied by increased plasmacytoid dendritic cells (pDCs) and reduced neutrophils as well as reduced CXCL-1 levels in BAL. Chronic Th17-dependent airway inflammation was unaltered by rapamycin treatment. None of the other agents showed efficacy in our models. Our results demonstrate that Th17-dependent airway inflammation is difficult to treat with known agents. However, we identify rapamycin as an agent with inhibitory potential against acute Th17-polarized airway inflammation.

Figure 1

Model for Th17-dependent airway inflammation and flow cytometry gating strategies. (a) Mice receive polarized Th17-cells on d0 i.v. and antigen challenges are performed with OVA and KLH on d1, 2.Within the acute model, sacrifice takes place on d4. For the chronic model further challenges with OVA or KLH are performed on d18 and d19 and sacrifice takes places on d21.(b) Transfer of Th17 cells into ROR-gamma-t reporter mice demonstrates priming and polarization of endogenous IL-17A-positive lymphocytes in lung and lymph nodes. LN: lymph nodes. (c) Example of gating strategy: After gating out debris and gating on single cells we focused on live, (DAPI⁻, lineage negative (CD3⁻CD19⁻) leukocytic populations (CD45⁺). pDCs (CD11c^lowB220⁺Ly6C/G⁺ cells), cDCs (MHCII^highCD11c^high cells) subdivided into steady-state cDCs (Ly6C/G⁻) and moDCs (Ly6C/G ⁺) and neutrophils (lin⁻MHCII⁻ CD11c⁻Ly6C/G⁺) were identified by appropriate surface markers. d: gating strategy T reg and cytokine production: After gating out debris and gating on single cells we focused on live (Pacific Orange negative) lymphocytes (FSC, SSC) T helper cells (CD3⁺CD4⁺). These were either analyzed regarding their IL-17A and IFNγ production or for the percentage of Tregs (CD25⁺Foxp3⁺).

Figure 2

Methylprednisolone treatment fails to reduce Th17-dependent acute airway inflammation. Methylprednisolone (Methpred) treatment on d1 and 2 of the acute model fails to reduce (a) BAL cell count (absolute number ± SEM), antigen-specific IL-17A secretion (pg/ml + SEM, ELISA) of lung (b) or lymph node (c) cells. Pooled data of 3 independent experiments, total number of animals/group = 7–9. One-way ANOVA with the post-hoc Bonferroni’s multiple comparison test, **p < 0.01.

Figure 3

Roflumilast, azithromycin and AM80 fail to reduce Th17-dependent acute airway inflammation. Roflumilast (ROFLU), azithromycin (AZMY) and AM80 treatment on d1 and 2 (d-5, d-3, d0 and d3 for azithromycin) of the acute model fails to reduce BAL cell count (±SEM) (a–c), Ag-specific IL-17A secretion (pg/ml + SEM, ELISA) of lung (d–f) or lymph node lymphocytes (g–i). Pooled data from 2 independent experiments with a total of 7–10 animals/group. One-way ANOVA with the post-hoc Bonferroni’s multiple comparison test, *p < 0.05, **p < 0.01, ***p < 0.001.

Figure 4

Rapamycin reduces Th17-dependent acute airway inflammation dose-dependently. Rapamycin treatment on d1 and 2 of the acute model resulted in a reduction of (a,b) BAL cell count (±SEM). (c) BAL neutrophils (±SEM) (asessessed by microscopic analyses of BAL cytospins) and (d) lung neutrophils (±SEM) (assessed by flow cytometry). Assessment of antigen specific cytokine secretion from lung (e–h) and lymph node cells revealed a dose-dependent reduction of IL-17A secretion in lung (e,f, ELISA) but not IFNγ (g,h, ELISA). Negligible amounts of IL-17A secretion were observed from lymph node cells (I,k, ELISA) with significant amounts of IFNγ secretion in this compartment (l,m, ELISA) but neither was regulated by rapamycin treatment. Pooled data of 1 (b,d,f,h,k,m) to 3 (a,c,e,g,i,l) independent experiments with a total number of 4–12 animals/group. One-way ANOVA with the post-hoc Bonferroni’s multiple comparison test, *p < 0.05, **p < 0.01, ***p < 0.001.

Figure 5

Rapamycin treatment decreases IL-17A⁺, IFNγ⁺ and regulatory T cells: Amount (% of T cells + SD) of IL-17A-producing (a,d), IFNγ-producing (b,e) and Tregs (c,f) were determined by means of flow cytometry in lung and LN from mice that underwent the acute model (RAPA, 8 mg/kg, d4). Data from one experiment with 5 animals per group. One-way ANOVA with the post-hoc Bonferroni’s multiple comparison test, *p < 0.05, **p < 0.01, ***p < 0.001.

Figure 6

Dendritic cells’ dynamic upon Rapamycin treatment: pDCs (% + SEM) return to numbers comparable to the negative control group (a) and, while the total cDC numbers remain constant (b), there is a significant decrease in the percentage of moDCs (% + SEM) (c). Data from one experiment with 4 animals/group. One-way ANOVA with the post-hoc Bonferroni’s multiple comparison test, *p < 0.05, **p < 0.01, ***p < 0.001.

Figure 7

Influence of Rapamycin on Th17/IL-17A-related cytokines and chemokines in BALF, lung and LN: Mediator concentrations (pg/ml + SD) were measured in BALF (a–g) and supernatants of OVA-specific stimulated LN (h,i) or lung (j,k) cells from mice that underwent the acute model (RAPA 8 mg/kg, d4) by means of cytometric beads array (b–k) or ELISA (a). Data from one experiment with 5 animals per group. Mann-Whitney t-test, **p < 0.01.

Figure 8

Rapamycin fails to reduce Th17-dependent chronic airway inflammation. In the chronic inflammation model rapamycin treatment did not impinge either total BAL cellularity (a–e), Ag-specific IL-17A production (pg/ml + SEM, ELISA) in the lung (f), LN (g) or Ag-specific IgG2a (U/ml) and IgG1 (ng/ml) production (h,i). Data from one representative experiment with a total of 5–6 animals/group. One-way ANOVA with the post-hoc Bonferroni’s multiple comparison test, *p < 0.05, **p < 0.01, ***p < 0.001.