Epithelial-biofilm interaction in primary ciliary dyskinesia

Abstract

Primary ciliary dyskinesia (PCD) is an inherited disorder of motile cilia that affects around 1 in 15,000 live births and its defining feature is a lack of mucociliary clearance. This results in progressive lung disease, upper airway symptoms and, potentially, situs inversus and infertility. PCD is characterised by early colonisation of the lower airways with bacteria, resulting in progressive lung function decline. Non-typeable Haemophilus influenzae (NTHi) is the commonest of these colonising bacteria in children and is also cultured in many other lung conditions such as cystic fibrosis, chronic obstructive pulmonary disease and bronchiectasis. NTHi forms biofilms on airway epithelium; a phenotype associated with increased tolerance to antibiotic treatment and resistance to host immunity. This work hypothesised that, in addition to lack of mucociliary clearance, the PCD airway has intrinsically dysregulated responses to colonising bacteria. This may be, in part, due to extremely low airway nitric oxide (NO). Thus, exogenous NO may represent a potential therapeutic adjunct in the treatment of lower airway infection. Systematic review and metaanalysis were used to define NO levels in PCD airway and healthy/disease controls. Then an established model of a 72h NTHi biofilm in vitro and on cultured primary respiratory epithelium was employed. NO treatment of these biofilms was investigated using a targeted NO compound (PYRRO-C3D) with the NO-treated NTHi in vitro biofilms subjected to label-free proteomic analyses to identify mechanisms underlying these treatment effects. Label-free proteomic analysis was also employed to identify differences in healthy and PCD epithelial proteome following exposure to 72h NTHi biofilm, as well as identifying NTHi proteins present. Systematic review and meta-analysis showed that PCD patients had extremely low nasal NO (mean 19.4nl/min) compared to healthy (265nl/min) and CF patients (123.2nl/min). This was independent of genotype or ultrastructural defect, thus there was a common epithelial defect across almost all PCD patients. PYRRO-C3D was effective in enhancing antibiotic treatment of NTHi biofilms in vitro and on cultured respiratory epithelium (p<0.05). This effect was more pronounced on PCD epithelium (2 log fold CFU drop) than healthy (1 log fold). Inhibition of NO release and scavenging of NO showed the effect to be NO mediated. Proteomic analysis revealed NO-induced upregulation of metabolic pathways and translational machinery, as well as a D-methionine uptake lipoprotein and iron metabolism. However, methionine isomers reversed the antibiotic enhancing effect of PYRRO-C3D. Pathway analysis of proteomic data from the co-culture model demonstrated that healthy epithelium responds to NTHi colonisation by cytoskeletal remodelling, metabolic upregulation and initial hyper-proliferation as well as suppression of acute inflammation via downregulation of S100 proteins. PCD epithelium appears to show a lesser degree of hyper-proliferation and fails to downregulate pro-inflammatory S100 proteins to the same extent. NTHi proteins were also identified in the analysis, with 14 present in both the co-culture samples and in vitro work, suggesting they warrant further investigation. OMPp5 (a HSP70 protein) is of particular interest as a potential biomarker or vaccine target. In conclusion, almost all PCD patients have extremely low airway NO regardless of underlying genotype, suggesting a common epithelial dysfunction. PCD epithelium may not respond adequately to NTHi colonisation as there is a failure of both normal proliferation and inflammatory suppression through downregulation of S100 proteins. Changes in intracellular calcium flux is a potential common mechanism behind failure of ciliary beat, low nitric oxide and S100/innate immune dysfunction. NO can also be successfully used as a targeted therapeutic adjunct in NTHi biofilms and may be particularly effective in PCD due to the constitutively low NO during colonisation. NO induces metabolic and translational changes in NTHi that make it more sensitive to treatment with macrolides and, potentially, other translation-targeting antibiotics. Proteins such as OMPp5 are common to in vitro and ex vivo biofilm models, suggesting the validity of in vitro models and showing promise as potential biomarkers of significant NTHi colonisation or targets of vaccines.

More information

Identifiers

ORCID for Jane Lucas: orcid.org/0000-0001-8701-9975

Catalogue record

Export record