Airway Administration of Flagellin Regulates the Inflammatory Response to Pseudomonas aeruginosa

Abstract

Excessive lung inflammation and airway epithelial damage are hallmarks of human inflammatory lung diseases, such as cystic fibrosis (CF). Enhancement of innate immunity provides protection against pathogens while reducing lung-damaging inflammation. However, the mechanisms underlying innate immunity-mediated protection in the lung remain mysterious, in part because of the lack of appropriate animal models for these human diseases. TLR5 (Toll-like receptor 5) stimulation by its specific ligand, the bacterial protein flagellin, has been proposed to enhance protection against several respiratory infectious diseases, although other cellular events, such as calcium signaling, may also control the intensity of the innate immune response. Here, we investigated the molecular events prompted by stimulation with flagellin and its role in regulating innate immunity in the lung of the pig, which is anatomically and genetically more similar to humans than rodent models. We found that flagellin treatment modulated NF-κB signaling and intracellular calcium homeostasis in airway epithelial cells. Flagellin pretreatment reduced the NF-κB nuclear translocation and the expression of proinflammatory cytokines to a second flagellin stimulus as well as to Pseudomonas aeruginosa infection. Moreover, in vivo administration of flagellin decreased the severity of P. aeruginosa-induced pneumonia. Then we confirmed these beneficial effects of flagellin in a pathological model of CF by using ex vivo precision-cut lung slices from a CF pigz model. These results provide evidence that flagellin treatment contributes to a better regulation of the inflammatory response in inflammatory lung diseases such as CF.

Keywords: Pseudomonas aeruginosa; TLR5; cystic fibrosis; flagellin; inflammation.

Figure 1.

Flagellin triggers a switch-like TLR5 (Toll-like receptor 5) response. Evaluation of P65 nuclear translocation in differentiated airway epithelial cells treated with 200 μl of 0 (vehicle), 0.1-, 1-, 10-, 100-, or 1,000-ng/ml apical FliC_Δ174–400 for 0, 20, 40, 60, 80, and 100 minutes was conducted. (A) Representative confocal images showing cells before (0 min; left) and after (60 min; right) stimulation with 100-ng/ml FliC_Δ174–400. Scale bar, 40 μm. (B) Time course of P65 nuclear translocation. (C) Dose responses for the expression of the CXCL2, CXCL8, CCL20, TNFAIP3, NFκBIA, and NFκBIE genes (dashed lines) and P65 nuclear translocation (solid line, half maximal effective concentration = 13.68 ng/ml). The area under the curve was extracted for the time course data and plotted over the dose response. (D) Time course of airway epithelial cells’ inflammatory response to flagellin. Quantitative PCR analysis of differentiated airway epithelial cells from wild-type pigs treated with 0- (vehicle), 0.1-, 1-, 10-, 100-, or 1,000-ng/ml apical FliC_Δ174–400 for 0, 0.5, 1, 2, 3, and 5 hours was conducted. Gene expression is shown relative to the mock group. (E) Volcano plot of RNA-sequencing data, in which the −log₁₀ of the padj is plotted against log₂ fold change. Differentiated airway epithelial cells were stimulated or not with 100-ng/ml FliC_Δ174–400 for 2 hours. Transcripts that are highlighted in red are discussed in the text. (F) Distant Regulatory Elements of Co-regulated Genes transcription factor motif analysis for the differentially expressed genes after flagellin stimulation; the importance is a product of the motif occurrence (the fraction of regulatory elements containing the motif) and weight score (motif prevalence compared with background gene list). Data are representative of three to four experiments. NFκBIA = nuclear factor-κ-B-inhibitor alpha; NFκBIE = nuclear factor-κ-B-inhibitor ε; padj = adjusted P value; TNFAIP3 = tumor necrosis factor, α-induced protein 3.

Figure 2.

Calcium homeostasis modulates the TLR5 response. (A) Histamine-dependent cytosolic Ca^2 + responses in differentiated airway epithelial cells pretreated (red) or not (blue) with 100-ng/ml apical FliC_Δ174–400 for 24 hours and stimulated with histamine (100μM) for 1 minute. Differences were evaluated by using the Kolmogorov-Smirnov (K-S) test. (B) Boxplot showing the fold change increase in the histamine-induced peak Ca^2 + response in cells pretreated or not with FliC_Δ174–400 for 24 hours. “Basal” indicates the mean value before histamine stimulation. (C) Quantitative PCR analysis of differentiated airway epithelial cells from wild-type pigs pretreated or not with 1μM thapsigargin for 24 hours and later stimulated or not with flagellin (100 ng/ml) for 5 hours. “Mock” indicates PBS-treated and nonstimulated airway epithelial cells. Gene expression is shown relative to the mock group. Intergroup differences were analyzed by using the Kruskal-Wallis test followed by a pairwise Mann-Whitney test with Benjamini-Hochberg false discovery rate correction. Data are representative of three experiments. a.u.= arbitrary unit; ns = not significant.

Figure 3.

Flagellin treatment decreases the TLR5 response to a second flagellin input. (A) Quantitative PCR analysis of differentiated airway epithelial cells from wild-type pigs pretreated or not with 200 μl of 100-ng/ml apical FliC_Δ174–400 for 24 hours and later stimulated or not with a second flagellin input (100 ng/ml) for 0, 0.5, 1, 2, and 5 hours. “Mock” indicates PBS-treated and nonstimulated airway epithelial cells. Gene expression is shown relative to the mock group. Intergroup differences were analyzed by using two-way ANOVA with time and treatment as factors, which was followed by a Tukey honestly-significant-difference (HSD) post hoc test. Data are representative of three experiments. (B) Gaussian probability density functions and (C) violin plots for P65 nuclear translocation responses in differentiated airway epithelial cells pretreated or not with 100-ng/ml apical FliC_Δ174–400 for 24 hours and later stimulated with 100-ng/ml FliC_Δ174–400 for 60 minutes. Differences were evaluated by using the K-S test and the Mann-Whitney Wilcoxon test. (D) Quantitative PCR analysis of differentiated airway epithelial cells from wild-type pigs pretreated or not with 200 μl of 100-ng/ml apical FliC_Δ174–400 for 24 hours. Differences were evaluated by using the Mann-Whitney Wilcoxon test. Data are representative of four to six experiments.

Figure 4.

Flagellin treatment decreases the immune response to Pseudomonas aeruginosa strain K (PAK) in wild-type airway epithelial cells. Quantitative PCR analysis of wild-type airway epithelial cells pretreated (Flagellin–PAK) or not (PBS–PAK) with 200 μl of 100-ng/ml apical FliC_Δ174–400 or left untreated for 24 hours and later infected with the P. aeruginosa strain PAK (multiplicity of infection = 0.2) for 5 hours. Gene expression is shown relative to the mock group (treated with the vehicle). Intergroup differences were analyzed using the Kruskal-Wallis test followed by a pairwise Mann-Whitney test with Benjamini-Hochberg false discovery rate correction. Data are representative of 12 to 15 experiments.

Figure 5.

Flagellin treatment before P. aeruginosa infection dampens lung inflammation in wild-type and cystic fibrosis pigs. (A–D) Lung sections were stained 24 hours after infection with hematoxylin and eosin. Images are shown of PBS-treated and infected pig (PBS–PAK) (A and B) and flagellin-treated and infected pig (Flagellin–PAK) (C and D). (E) Wild-type pigs (n = 5) received a bronchial administration of 0.012-mg/kg FliC_Δ174–400 or PBS and were infected or not with 10 ml of a 10⁷-cfu/ml suspension of the P. aeruginosa strain PAK 24 hours later. The inflammatory response was evaluated 24 hours after infection. CCL20 and CXCL8 gene expression amounts were evaluated by using quantitative PCR analysis. Gene expression is shown relative to the PBS-treated, noninfected group. Data were analyzed by two-way ANOVA, using flagellin administration and P. aeruginosa infection as factors, which was followed by a Tukey HSD post hoc test. (F) Quantitative PCR analysis of precision-cut lung slices from newborn CFTR^+/+ and CFTR^−/− piglets treated or not with 100-ng/ml FliC_Δ174–400 for 24 hours and later infected with the P. aeruginosa strain PAK (multiplicity of infection = 0.2) for 0, 0.5, 1, 2, and 5 hours. Gene expression is shown relative to the mock group (nontreated and noninfected samples). Intergroup differences were analyzed by two-way ANOVA using time and treatment as factors, which was followed by a Tukey HSD post hoc test. Data are representative of three experiments. Scale bars: A and C, 500 μm; B and D, 50 μm.