Airway epithelial NF-κB activation promotes the ability to overcome inhalational antigen tolerance

Abstract

Background: Inhalational antigen tolerance typically protects against the development of allergic airway disease but may be overcome to induce allergic sensitization preceding the development of asthma.

Objectives: We examined in vivo whether pre-existing inhalational antigen tolerance could be overcome by activation of the transcription factor NF-κB in conducting airway epithelial cells, and used a combination of in vivo and in vitro approaches to examine the mechanisms involved.

Methods: Wild-type and transgenic mice capable of expressing constitutively active IκB kinase β (CAIKKβ) in airway epithelium were tolerized to inhaled ovalbumin. Twenty-eight days later, the transgene was transiently expressed and mice were exposed to inhaled OVA on Day 30 in an attempt to overcome inhalational tolerance.

Results: Following ovalbumin challenge on days 40-42, CAIKKβ mice in which the transgene had been activated exhibited characteristic features of allergic airway disease, including airway eosinophilia and methacholine hyper-responsiveness. Increases in the CD103(+) and CD11b(HI) lung dendritic cell populations were present in CAIKKβ mice on Day 31. Bronchoalveolar lavage from mice expressing CAIKKβ mice induced CD4(+) T cells to secrete T(H)2 and T(H)17 cytokines, an effect that required IL-4 and IL-1 signalling, respectively. CAIKKβ mice on Dox demonstrated increased numbers of innate lymphoid type 2 cells (ILC2) in the lung, which also exhibited elevated mRNA expression of the T(H)2-polarizing cytokine IL-4. Finally, airway epithelial NF-kB activation induced allergic sensitization in CAIKKβ mice on Dox that required IL-4 and IL-1 signalling in vivo.

Conclusions: Our studies demonstrate that soluble mediators generated in response to airway epithelial NF-κB activation orchestrate the breaking of inhalational tolerance and allergic antigen sensitization through the effects of soluble mediators, including IL-1 and IL-4, on pulmonary dendritic cells as well as innate lymphoid and CD4(+) T cells.

Keywords: NF-κB; T regulatory cells; allergic airway disease; dendritic cells; inhalational antigen tolerance.

Figure 1. Models for inducing and attempting to break inhalational tolerance

(a) C57BL/6 mice were exposed on days 0, 1, and 2 to either oropharyngeal aspiration (o.a.) of saline or 100 mg low-endotoxin (LE-OVA), followed by i.p. injection on Day 30 with Alum/OVA. All mice were challenged on Day 40, 41, and 42 with a 30 minute exposure of nebulized (Neb) 1% w/v OVA, and were analyzed 48 hours later on Day 44. (b) In contrast, wild type and CAIKKβ mice underwent the same tolerization and challenge regimen, but were administered Dox on days 28-30, followed by a single aerosolized exposure to 1% OVA for 30 minutes on day 30.

Figure 2. Inducible airway epithelial NF-κB activation promotes antigen-induced sensitization to allergic airway disease, regardless of tolerized status

Differential BAL cell counts (a) of wild type (WT) or transgenic CAIKKβ (TG) mice that were sensitized via 60 hours of Dox chow administration followed by aerosolized OVA, with (O/O) or without (sal/O) previous OVA tolerization (see Fig. 1b). TGF-β was measured from BAL (b) and gene expression was measured by quantitative PCR from whole lung cDNA (c-e). Splenocytes were harvested from mice 48 hours after challenge, and restimulated with OVA for 96 hours before IL-17A from cell-free supernatants was analyzed by ELISA (f). n = 6-7 mice per/group. *p=<0.05, **p=<0.01, ***p=<0.001, compared to wild type tolerized mice on Dox, WT O/O.

Figure 3. Inducible airway epithelial NF-κB activation promotes airway hyperresponsiveness to inhaled methacholine, regardless of tolerized status

Forced oscillations using the flexiVent were used to assess baseline measures of R_N, G, and H (a-c) as well as the responsiveness to inhaled methacholine (d-f). n = 3-6 mice per/group. *p=<0.05 for WT vs. TG (CAIKKβ) exposed to the same treatment regimen.

Figure 4. Airway epithelial NF-κB activation drives dendritic cell maturation and promotes an inflammatory lung DC phenotype when overcoming inhalational tolerance

Mice were analyzed on day 3 (immediately following tolerization) or day 31 (immediately following sensitization) of the inhalational tolerance protocol and lung single cell suspensions were analyzed by flow cytometry for dendritic cell surface markers (a). Cells were gated for dendritic cell phenotype (CD11c⁺, FITC autofluorescence^LO) and analyzed for co-positivity of CD103/MHC II and CD11b/MHC II positive populations. Naive dendritic cells were isolated from wild type mouse spleens and exposed to cell-free bronchoalveolar lavage fluid from wild type (WT) or CAIKKβ mice after 60 hours of Dox administration and analyzed by flow cytometry for MHC II and CD86 (b). CAIKKβ and WT littermate mice were administered Dox for 60 hours. Cell-free BAL was collected and concentrated using centrifugal filtration. Cytokines from concentrate were analyzed by Milliplex panel (c). Cytokines that were measured but undetectable include: IL-1β, MCP-1, MIP-2, TNF-α, and IL-12p70. n= 4-8 mice/group, * = p<0.05, **= p<0.01 *** = p<0.001 compared to WT.

Figure 5. Airway epithelial NF-κB activation induces release of acute inflammatory mediators into the lavageable airspace that polarize CD4⁺ T cell cytokine secretion

CD4⁺ T cells isolated from naive C57BL/6 mouse spleen and lymph nodes were plated in the presence of anti-CD3 and anti-CD28 and treated with cell-free BAL from CAIKKβ mice after 60 hours Dox administration, in the presence or absence of anti-IL-4 (10 μg/ml), or Anakinra (200 ng/ml). Cells were incubated for 72 hours and supernatants were analyzed by ELISA. n = 9 mice/group. * = p<0.05, ** = p<0.01, *** = p<0.001.

Figure 6. Airway epithelial NF-κB activation induces increased lung ILC2 populations, and ILC2 increases in IL-4 and IL-5 mRNA expression

Lung single cell suspensions from wild type and CAIKKβ mice that received 60 hours of Dox administration were analyzed for the presence of ILC2 cells (Lin-CD90.2⁺Sca-1⁺) (a, b). mRNA expression from these lung ILC2s was analyzed by RT-PCR for IL-4, IL-5, and IL-13 (c). n=3-4 mice/group. * = p<0.05, ** = p<0.01, ns = not significant, compared to WT.

Figure 7. Airway epithelial activation requires IL-4 and IL-1-signaling to drive allergic sensitization to OVA

Wild type and transgenic CAIKKβ littermates were sensitized to OVA as previously published (19), with the addition of neutralizing antibodies during the sensitization phase (a). On Day 0 and 3, transgenic mice received either 10 mg/mouse anti-rat IgG1 isotype control (Iso), 10 mg/mouse anti-IL-4 (αIL-4), or 110 μg/mouse IL-1 TRAP (TRAP). Total BAL cell counts (b) as well as differential counts from cytospins (c) were performed. Splenocytes were restimulated with OVA for 96 hours and supernatants were analyzed by ELISA (d). Quantitative-PCR from whole lung was examined for gene expression of Muc5ac and Clca1 (e). n=4-6 mice per group. *=p<0.05, **=p<0.01, ***=p<0.005, ****=p<0.001.