Rapid detection of emerging pathogens and the loss of microbial diversity associated with severe lung disease in cystic fibrosis. Microbial diversity associated with severe lung disease in cystic fibrosis

Background. Chronic respiratory infection in cystic fibrosis (CF) is recognised as polymicrobial, but standard sputum microbiology does not account for the lung microbiome or detect the loss of microbial diversity associated with severe disease. To develop a clinically applicable CF microbiome surveillance scheme, the efficacy of total sputum nucleic acids isolated for accredited viral diagnosis combined with PCR-based bacterial diversity profiling (ribosomal intergenic transcribed spacer analysis; RISA), was investigated. Methods. Conventional culture and RISA were performed on 200 paired sputum samples obtained from 93 CF adults. RISA diversity profiles were examined by cluster analysis and confirmed by pyrosequencing of the 16S rRNA gene for 59 patients. Statistical analysis of the bacterial diversity its correlation to the clinical metadata was performed. Results. RISA was successful for all 200 samples with two cluster groups standing out when compared to the microbiology data. Diversity profiles associated with patients culture positive for non-Pseudomonas pathogens, Burkholderia, Achromobacter, Stenotrophomonas, and Ralstonia (designated emerging non-fermenting Gram negatives; eNFGN) were easily identified by the presence of dominant intergenic transcribed spacer (ITS) amplicons. The second, Pseudomonas group profiles were associated P. aeruginosa culture positive patients and contained a dominant 753 bp amplicon correlating to this species or diverse ITS products. Sequence analysis confirmed the presence of eNFGN genera within this group, with Burkholderia being significantly more abundant (p < 0.008); Pseudomonas was more abundant in the Pseudomonas group (p < 0.001). The presence Burkholderia was also correlated with a loss microbial diversity (p < 0.001). Low bacterial diversity was associated with severe lung disease (p < 0.001) and an absence of Streptococcus (p = 0.045) was observed in individuals with lung function in the lowest quartile. Conclusions. Nucleic acid isolated from CF sputum for viral diagnosis can serve as a template for bacterial diversity analysis. Since routine bacteriology missed the presence of dominant eNFGN pathogens or P. aeruginosa (9 out of 93 cases) that were rapidly detected by microbiota PCR-profiling, there is a strong case for linking viral sputum molecular diagnostics with molecular bacteriology to monitor the state of CF infection.