Article Text
PDF
Abstract
Introduction Obtaining accurate information on the composition of Cystic fibrosis (CF) lung infections is essential for the selection of appropriate therapy and maintenance of respiratory function. Whilst significant advances have been made using molecular techniques to provide such characterisation, there is a pressing need to identify the most appropriate sampling handling protocols. Here, we investigated the relationship between the time period prior to sample freezing and the bacterial community profiles.
Methods Eleven sputum samples were collected from adult CF patients experiencing acute pulmonary exacerbations. Expectorated sputum was aliquoted into 12 equal portions, with one portion being frozen immediately and others transferred to –80°C at time points over a 72 hour period. Samples were treated with propidium monoazide prior to DNA extraction to limit analysis to viable bacterial cells. The bacterial composition of samples was determined by 16S rRNA gene Terminal Restriction Fragment Length Polymorphism (T-RFLP) profiling. Changes in bacterial community composition over the time course were then analysed using ecological statistical tools.
Results A marked change in the bacterial community composition and abundance was observed within the first hour. Using Bray Curtis similarity index, a 51% similarity change in the bacterial community was observed in the first hour when compared to the sample frozen immediately. Thereafter, the bacterial community remained relatively stable for the following 72 hours (Figure 1A). Species cumulative richness showed an increase of approximately 63 new bacterial species over 72 hours, of which 25/63 (40%) emerged within the first hour. The mean sample richness also showed a significant increase in bacterial species within the first hour (Figure 1B).
(A) Change in bacterial community composition and abundance over 72 hours, the sample at each time point was compared to the sample frozen immediately. (B) Mean cumulative richness (emerging new bacterial species) and mean sample richness (total number of bacterial species in each sample). Error bars represent the standard error of the mean.
Conclusion The data presented indicates a significant change in sample bacterial composition within the first hour of sample collection. These findings have clear implications for the handling of samples for viable community profiling.