Bile signalling promotes chronic respiratory infections and antibiotic tolerance

Abstract

Despite aggressive antimicrobial therapy, many respiratory pathogens persist in the lung, underpinning the chronic inflammation and eventual lung decline that are characteristic of respiratory disease. Recently, bile acid aspiration has emerged as a major comorbidity associated with a range of lung diseases, shaping the lung microbiome and promoting colonisation by Pseudomonas aeruginosa in Cystic Fibrosis (CF) patients. In order to uncover the molecular mechanism through which bile modulates the respiratory microbiome, a combination of global transcriptomic and phenotypic analyses of the P. aeruginosa response to bile was undertaken. Bile responsive pathways responsible for virulence, adaptive metabolism, and redox control were identified, with macrolide and polymyxin antibiotic tolerance increased significantly in the presence of bile. Bile acids, and chenodeoxycholic acid (CDCA) in particular, elicited chronic biofilm behaviour in P. aeruginosa, while induction of the pro-inflammatory cytokine Interleukin-6 (IL-6) in lung epithelial cells by CDCA was Farnesoid X Receptor (FXR) dependent. Microbiome analysis of paediatric CF sputum samples demonstrated increased colonisation by P. aeruginosa and other Proteobacterial pathogens in bile aspirating compared to non-aspirating patients. Together, these data suggest that bile acid signalling is a leading trigger for the development of chronic phenotypes underlying the pathophysiology of chronic respiratory disease.

Figure 1. Bile transcriptional footprint.

(a) Pie-chart profiling of the transcriptional response to bile in P. aeruginosa. (b) Singular global analysis of the bile transcriptome from the Pearson Correlation matrix using pheatmap. Correlation values are presented as per key. Following this, the clustered transcriptome profiles were interrogated for virulence and redox associated genes, representing a subset of the global profile. Red indicates downregulation of gene(s) within the designated system, while green denotes upregulation. The red/green coloured AHL box signifies las induction and rhl suppression. Grey boxes represent gene(s) or systems that are not significantly altered in the respective transcriptome datasets.

Figure 2. Bile suppresses redox and promotes antibiotic tolerance.

(a) Physiologically relevant concentrations of bile elicit a strong suppression of redox activity in P. aeruginosa. (b) Morphological changes consistent with redox flux with wrinkled colony formation on bile treated TSA plates. (c) Increased mexAB promoter activity in the presence of bile. (d) E-strip antibiotic tolerance assays reveal bile-dependent tolerance to polymyxin and macrolide antibiotics in the presence of bile. Data is the mean of at least three independent biological replicates. Statistical analysis was performed by Student’s t-test (***p ≤ 0.001).

Figure 3. Bile salts elicit chronic lifestyle in P. aeruginosa.

(a) Bile salts trigger biofilm formation in P. aeruginosa. The 0.03% solution of bile salts is equivalent to approximately 300 μM. (b) Bile salts at 50 μM also suppress the redox response. (c) CDCA at 50 μM was capable of enhancing biofilm formation in P. aeruginosa. In each panel, data is presented as OD(_{595 nm or 510 nm}) normalised to the untreated control. All experiments are the average or representative of at least three independent biological replicates. Statistical analysis was performed by Student’s t-test, one-way ANOVA or Bootstratio (*p ≤ 0.05, **p ≤ 0.005, ***p ≤ 0.001).

Figure 4. Bile salts trigger FXR-dependent pro-inflammatory cytokine production.

(a) Induction of the pro-inflammatory cytokine IL-6 by CDCA involves the FXR receptor. (b) Induction of IL-6 occurs independent of the TGR5-PKA pathway. Data presented are the average of three independent biological replicates. Statistical analysis was performed by one-way ANOVA and Bootstratio (*p ≤ 0.05, **p ≤ 0.005, ***p ≤ 0.001).

Figure 5. Bile acids in sputum correlate with Proteobacterial dominance.

(a) Increased relative abundance of Proteobacterial pathogens was observed in aspirating samples (1–4) compared to non-aspirating (5–9). Samples 1 and 2 were taken 2 months apart from the same aspirating and suggest an exacerbation with Streptococcus. Samples 5 and 6, which were taken 3 months apart from a non-aspirating patient, reveal a stable and relatively rich microbiome of a non-aspirating patient. The x-axis numbers refer to the ID Nos. in Tables S3 and S4. (b) Biofilm analysis of clinical isolates from microbiome study performed on multi-well plates and quantified by crystal violet staining. Data presented is the average of at least three independent biological replicates. Statistical analysis was performed by Bootstratio (*p ≤ 0.05).

Figure 6. Overview of the bile response underlying chronic infection and chronic inflammation in respiratory disease patients.

(1) The aspiration of bile acids into the lungs of patients with respiratory disease (2) correlates with a markedly reduced microbiome diversity in patients with CF. (3) Bile, with CDCA in particular, elicits a chronic lifestyle in several prominent respiratory pathogens. This includes a switch towards a biofilm lifestyle and suppression of virulence systems associated with the acute phase of infection. From a clinical perspective, (4) tolerance to the polymyxin and macrolide classes of antibiotics was observed in the presence of bile. In tandem with this, bile salts also (5) destabilised HIF-1 and (6) triggered FXR-dependent production of the IL-6 pro-inflammatory cytokine, potentially contributing to the chronic inflammation that characterises the pathophysiology of respiratory diseases such as CF.