One versus Many: Polymicrobial Communities and the Cystic Fibrosis Airway

Abstract

Culture-independent studies have revealed that chronic lung infections in persons with cystic fibrosis (pwCF) are rarely limited to one microbial species. Interactions among bacterial members of these polymicrobial communities in the airways of pwCF have been reported to modulate clinically relevant phenotypes. Furthermore, it is clear that a single polymicrobial community in the context of CF airway infections cannot explain the diversity of clinical outcomes. While large 16S rRNA gene-based studies have allowed us to gain insight into the microbial composition and predicted functional capacities of communities found in the CF lung, here we argue that in silico approaches can help build clinically relevant in vitro models of polymicrobial communities that can in turn be used to experimentally test and validate computationally generated hypotheses. Furthermore, we posit that combining computational and experimental approaches will enhance our understanding of mechanisms that drive microbial community function and identify new therapeutics to target polymicrobial infections.

Keywords: antibiotics; biofilms; chronic infection; cystic fibrosis; metabolic modeling; microbial communities; microbiome.

FIG 1

Characterization of the polymicrobial communities of sputum samples from cystic fibrosis inpatients and outpatients. (A) Fraction of 454 pyrosequencing reads assigned to each of the top 10 genera detected in the sample set as a whole is shown for sputum samples analyzed from inpatients (INPT; n = 13) and outpatients (OUTPT; n = 22) from a cross-sectional study. The legend on the right indicates the color assigned to each indicated genus. (B) Heat map of samples from pwCF based on Pearson hierarchical clustering of relative bacterial abundance using the data in panel A for the most prevalent four genera, which account for ∼86% of the total pyrosequencing reads. The legend at the bottom indicates the four clusters or “groups” assigned by hierarchical clustering, as reported (20). The airway infection communities from pwCF can be described by one of three profiles: (i) high Pseudomonas (group 1), (ii) high Streptococcus (group 3), and (iii) mixed communities with a relatively even distribution of taxa (groups 2 and 4). Modified from reference .

FIG 2

Predicted metabolite cross-feeding relationships between the four most abundant genera. One thousand model simulations were performed and split into 500 cases with relatively high Pseudomonas abundances and 500 cases with relatively low Pseudomonas abundances. The arrow thickness represents the magnitude of the metabolic exchange rate between the microbial species of the consortia. The color of the arrows is defined by the species producing and cross-feeding the metabolite(s) as follows: black arrow, Staphylococcus; red arrow, Streptococcus; blue arrow, Pseudomonas; green arrow, Prevotella. (A) Schematic representation of predicted metabolite exchange for Pseudomonas-dominated communities. (B) Schematic representation of predicted metabolite exchange for Streptococcus-dominated communities. Modified from reference .