Loss of bacterial diversity during antibiotic treatment of intubated patients colonized with Pseudomonas aeruginosa

Abstract

Management of airway infections caused by Pseudomonas aeruginosa is a serious clinical challenge, but little is known about the microbial ecology of airway infections in intubated patients. We analyzed bacterial diversity in endotracheal aspirates obtained from intubated patients colonized by P. aeruginosa by using 16S rRNA clone libraries and microarrays (PhyloChip) to determine changes in bacterial community compositions during antibiotic treatment. Bacterial 16S rRNA genes were absent from aspirates obtained from patients briefly intubated for elective surgery but were detected by PCR in samples from all patients intubated for longer periods. Sequencing of 16S rRNA clone libraries demonstrated the presence of many orally, nasally, and gastrointestinally associated bacteria, including known pathogens, in the lungs of patients colonized with P. aeruginosa. PhyloChip analysis detected the same organisms and many additional bacterial groups present at low abundance that were not detected in clone libraries. For each patient, both culture-independent methods showed that bacterial diversity decreased following the administration of antibiotics, and communities became dominated by a pulmonary pathogen. P. aeruginosa became the dominant species in six of seven patients studied, despite treatment of five of these six with antibiotics to which it was sensitive in vitro. Our data demonstrate that the loss of bacterial diversity under antibiotic selection is highly associated with the development of pneumonia in ventilated patients colonized with P. aeruginosa. Interestingly, PhyloChip analysis demonstrated reciprocal changes in abundance between P. aeruginosa and the class Bacilli, suggesting that these groups may compete for a similar ecological niche and suggesting possible mechanisms through which the loss of microbial diversity may directly contribute to pathogen selection and persistence.

FIG. 1.

(A and B) Comparisons of 16S rRNA clone libraries of EA (white bars) and BAL (black bars) patient samples. (A) A total of 173 clones were sequenced from the EA samples, and 153 clones were sequenced from the BAL samples. (B) A total of 154 clones were sequenced from the EA samples, and 141 clones were sequenced from the BAL samples. Species abundances (as percentages) of the sequenced clones are indicated above the bars. (C) Phylogenetic tree showing all recognized bacterial phyla/divisions. Phyla detected in EAs by cloning and PhyloChip microarray analyses are shown. The five main phyla detected by clone library are indicated. Many additional phyla, undetected by cloning, were detected by array analysis.

FIG. 2.

Temporal changes in bacterial diversity. Each bar represents the color-coded relative abundance of bacteria in a single EA. Numbers above the horizontal bars represent individual patients, and the numbers of clones analyzed for each sample are indicated directly above each bar. For each sample, Shannon's diversity statistic, which reflects both species numbers and evenness of species distribution, is plotted below the histogram.

FIG. 3.

Comparison of PhyloChip and clone library monitoring of bacterial diversity (phylotype numbers) over time for four patients. Closed symbols show numbers of bacterial phylotypes detected by PhyloChip analysis; open symbols show numbers of bacterial phylotypes detected by clone library on a logarithmic scale. The PhyloChip reveals many more bacterial groups, but the data exactly mirror the loss of diversity revealed by clone library sequencing.

FIG. 4.

PhyloChip analysis of complete bacterial communities over time in EAs. (A) Bacteria are ordered alphabetically from left to right according to taxonomic affiliation. Bars above the zero line represent bacteria that increased in abundance relative to the first EA sample being compared; bars below represent those bacteria that declined in abundance. (B) Venn diagrams demonstrate the number of bacterial subfamilies detected at each time point. Note the significant overlap in composition between successive time points.