Phylogenetic and metabolic diversity of bacteria associated with cystic fibrosis

Abstract

In patients afflicted with cystic fibrosis (CF), morbidity and mortality are primarily associated with the adverse consequences of chronic microbial bronchial infections, which are thought to be caused by a few opportunistic pathogens. However, recent evidence suggests the presence of other microorganisms, which may significantly affect the course and outcome of the infection. Using a combination of 16S rRNA gene clone libraries, bacterial culturing and pyrosequencing of barcoded 16S rRNA amplicons, the microbial communities present in CF patient sputum samples were examined. In addition to previously recognized CF pathogens such as Pseudomonas aeruginosa and Staphylococcus aureus, >60 phylogenetically diverse bacterial genera that are not typically associated with CF pathogenesis were also detected. A surprisingly large number of fermenting facultative and obligate anaerobes from multiple bacterial phyla was present in each sample. Many of the bacteria and sequences found were normal residents of the oropharyngeal microflora and with many containing opportunistic pathogens. Our data suggest that these undersampled organisms within the CF lung are part of a much more complex microbial ecosystem than is normally presumed. Characterization of these communities is the first step in elucidating potential roles of diverse bacteria in disease progression and to ultimately facilitate advances in CF therapy.

Figure 1

(a, b) Bacterial genera detected in CF sputum samples by 16S rRNA gene clone libraries, bacterial culturing and pyrosequencing. The abundance of pyrosequencing reads attributed to different genera within samples 2, 3 and 4 is shown in bar graphs (log scale). Sample 1 did not generate adequate amounts of DNA for pyrosequencing analysis. In panel a, diamond (⧫) indicates a singleton. Genus designation was determined using the RDP Naïve Bayesian Classifier tool. Traditional CF pathogens are in red and indicated by an asterisk (^*). The color reproduction of the figure is available on the html full text version of the paper.

Figure 1

(a, b) Bacterial genera detected in CF sputum samples by 16S rRNA gene clone libraries, bacterial culturing and pyrosequencing. The abundance of pyrosequencing reads attributed to different genera within samples 2, 3 and 4 is shown in bar graphs (log scale). Sample 1 did not generate adequate amounts of DNA for pyrosequencing analysis. In panel a, diamond (⧫) indicates a singleton. Genus designation was determined using the RDP Naïve Bayesian Classifier tool. Traditional CF pathogens are in red and indicated by an asterisk (^*). The color reproduction of the figure is available on the html full text version of the paper.

Figure 2

Sequence diversity and abundance detected by pyrosequencing in CF sputum samples 2, 3 and 4. (a) Rarefaction curves for total bacterial communities in the three samples at a 99% sequence similarity level. Curvature toward the horizontal indicates the increased sequencing effort required to observe novel OTUs when only rare OTUs remain to be discovered. (b) Venn diagram showing unique and shared OTUs (97%) in each sputum sample. (c) Venn diagram showing numbers of unique and shared genera detected in CF sputum samples by pyrosequencing. Genus designation was determined using the RDP Naïve Bayesian Classifier tool. Sequences that could not be classified to genera are not included in this figure. (d) Pie charts showing the taxonomic breakdown (phylum level) of pyrosequences (grey bars indicate sequences affiliated to the genus Pseudomonas).