Haemophilus influenzae infection drives IL-17-mediated neutrophilic allergic airways disease

Abstract

A subset of patients with stable asthma has prominent neutrophilic and reduced eosinophilic inflammation, which is associated with attenuated airways hyper-responsiveness (AHR). Haemophilus influenzae has been isolated from the airways of neutrophilic asthmatics; however, the nature of the association between infection and the development of neutrophilic asthma is not understood. Our aim was to investigate the effects of H. influenzae respiratory infection on the development of hallmark features of asthma in a mouse model of allergic airways disease (AAD). BALB/c mice were intraperitoneally sensitized to ovalbumin (OVA) and intranasally challenged with OVA 12-15 days later to induce AAD. Mice were infected with non-typeable H. influenzae during or 10 days after sensitization, and the effects of infection on the development of key features of AAD were assessed on day 16. T-helper 17 cells were enumerated by fluorescent-activated cell sorting and depleted with anti-IL-17 neutralizing antibody. We show that infection in AAD significantly reduced eosinophilic inflammation, OVA-induced IL-5, IL-13 and IFN-γ responses and AHR; however, infection increased airway neutrophil influx in response to OVA challenge. Augmented neutrophilic inflammation correlated with increased IL-17 responses and IL-17 expressing macrophages and neutrophils (early, innate) and T lymphocytes (late, adaptive) in the lung. Significantly, depletion of IL-17 completely abrogated infection-induced neutrophilic inflammation during AAD. In conclusion, H. influenzae infection synergizes with AAD to induce Th17 immune responses that drive the development of neutrophilic and suppress eosinophilic inflammation during AAD. This results in a phenotype that is similar to neutrophilic asthma. Infection-induced neutrophilic inflammation in AAD is mediated by IL-17 responses.

Figure 1. Characterization of NTHi infection.

The profile of infection was assessed in mice that only received NTHi (i.e. not OVA), by performing a time-course of bacterial recovery from BALF and lung homogenates (A), and airway inflammation represented by BALF neutrophils, lymphocytes and eosinophils (B). IL-5, IL-13, IL-17, IL-22 and IFN-γ (C) release from MLN T cells stimulated with killed NTHi compared to unstimulated cells was also determined. Lung function in terms of AHR (dynamic compliance (D) and transpulmonary resistance (E)) in response to increasing doses of methacholine was assessed 5, 16 and 26 days after inoculation. N.B. P values for compliance and resistance were determined for the entire dose response curve, *** p<0.001, ** p<0.01, * p<0.05 compared to indicated time points, +++ p<0.001, ++ p<0.01, + p<0.05 compared to unstimulated controls.

Figure 2. NTHi infection suppressed key features of Th2-driven eosinophilic AAD.

Groups of mice were infected during (d0 NTHi+OVA) or 10 days after (d10 NTHi+OVA) OVA sensitization (A), and AAD was analyzed (on day 16). The effects of infection on IL-5, IL-13, and IFN-γ (B) release from MLN T cells stimulated with OVA, BALF total cell and eosinophil counts (C) and the percentage of blood eosinophils (D) were assessed. AHR in terms of dynamic compliance (E) and transpulmonary resistance (F) was determined (see figure 1D-E for infection only AHR results). Data for the corresponding NTHi control groups were obtained at the same time point after infection as data from the NTHi+OVA groups. N.B. P values for compliance and resistance were determined for the entire dose response curve, ### p<0.001, ## p<0.01, # p<0.05 compared to Saline controls, *** p<0.001, ** p<0.01, * p<0.05 compared to OVA controls.

Figure 3. NTHi infection suppressed systemic IL-13 and IFN-γ responses in AAD.

Groups of mice were infected during (d0 NTHi+OVA) or 10 days after (d10 NTHi+OVA) OVA sensitization, and AAD was analyzed (on day 16). The effect of infection on IL-5 (A), IL-13 (B), and IFN-γ (C) release from splenocytes stimulated with OVA was assessed. Data for the corresponding NTHi control groups were obtained at the same time point after infection as data from the NTHi+OVA groups. ### p<0.001, ## p<0.01 compared to Saline controls, *** p<0.001, * p<0.05 compared to OVA controls.

Figure 4. NTHi infection does not affect T regulatory cells in the suppression of AAD.

Treg numbers in the lung (A), TGF-β (B) and IL-10 (C) mRNA expression in lung tissue were assessed (on day 16). Data for the corresponding NTHi control groups were obtained at the same time point after infection as data from the NTHi+OVA groups. ## p<0.01, compared to Saline controls, ** p<0.01, * p<0.05 compared to OVA controls.

Figure 5. NTHi infection reduced markers of antigen presentation and activation in AAD.

The numbers and proportions of MHCII and CD86 expressing pDCs (A–D) and mDCs (E–H) in MLNs and lungs were determined. Data for the corresponding NTHi control groups were obtained at the same time point after infection as data from the NTHi+OVA groups. ###p<0.001, ## p<0.01, # p<0.05 compared to Saline controls, ***p<0.001, ** p<0.01, * p<0.05 compared to OVA controls.

Figure 6. NTHi infection induces neutrophilic inflammation in AAD.

The effect of NTHi infection on BALF neutrophil numbers in AAD was determined (on day 16) and compared to groups with AAD without infection or infection alone without AAD. Data for the corresponding NTHi control groups were obtained at the same time point after infection as data from the NTHi+OVA groups. ## p<0.01 compared to Saline controls, ** p<0.01 compared to OVA controls.

Figure 7. NTHi infection induces increased IL-17 responses that correlate with neutrophil influx in neutrophilic AAD.

To investigate the association between neutrophil influx and IL-17 responses, a time-course (A), of neutrophils in BALF (B), IL-17 mRNA expression in lung tissue (C), and IL-17 release from MLN T cells stimulated with killed NTHi (D), and OVA (E) was assessed. Data for the corresponding NTHi control groups were obtained at the same time point after infection as data from the NTHi+OVA groups. ### p<0.001, # p<0.05 compared to Saline controls, *** p<0.001, ** p<0.01, * p<0.05 compared to OVA controls.

Figure 8. Early neutrophil influx is associated with enhanced neutrophil chemokine expression.

To investigate the association between early neutrophil influx and neutrophil chemokine responses, KC and MIP2 mRNA expression (on day 1 and 16, A) and protein levels (day 1 only, B) in lung tissue were assessed. Data for the corresponding NTHi control groups were obtained at the same time point after infection as data from the NTHi+OVA groups. ## p<0.01, # p<0.05 compared to Saline controls, ** p<0.01, * p<0.05 compared to OVA controls.

Figure 9. NTHi infection induces Th17 cell differentiation and IL-17 production from Th17 cells in neutrophilic AAD.

Th17 cell differentiation in lung tissues was examined by determination of the expression of the transcription factor ROR-γt (A). Th17 cell numbers and proportions were quantified in lungs (B–C) and lymph nodes (D–E). Data for the corresponding NTHi control groups were obtained at the same time point after infection as data from the NTHi+OVA groups. ## p<0.01 compared to Saline controls, ** p<0.01, * p<0.05 compared to OVA controls.

Figure 10. NTHi infection induces early IL-17 production from lung macrophages and neutrophils in neutrophilic AAD.

To determine early cellular sources of IL-17, IL-17 producing macrophage (A–B) and neutrophil (C–D) numbers and proportions in lungs were assessed. IL-17 mRNA was also assessed in isolated lung macrophages (E) and neutrophils (F). Data for the corresponding NTHi control groups were obtained at the same time point after infection as data from the NTHi+OVA groups. ### p<0.001 compared to Saline controls, *** p<0.001, ** p<0.01 compared to OVA controls.

Figure 11. NTHi infection-induced IL-17 is required for the induction of neutrophilic AAD and is partially responsible for the effects of infection on T cell responses.

To confirm the role of IL-17, anti-IL-17 monoclonal antibody was administered i.p. on days 11 and 13 of infection-induced neutrophilic AAD, and AAD was analyzed (on day 16) (A). The effects on BALF neutrophil influx (B), and KC and MIP2 mRNA expression in lung tissue (C) were assessed. # p<0.05 compared to OVA controls, *** p<0.001, * p<0.05 compared to isotype controls.