Cystic fibrosis (CF) is a genetic disease that results in the accumulation of dense, dehydrated mucus in the airways. Unclearable mucus contributes to the establishment of long-term microbial communities in the CF airways. Opportunistic pathogens damage the airways by inducing the host inflammatory response and generating toxic metabolites, contributing to decreased quality of life and lifespan (~40 years) for persons with CF. In order to improve treatment of lung infections, it is necessary to understand the role of microbes and microbial metabolism in triggering CF pulmonary exacerbations (CFPE), periods of worsening lung function. While antibiotics ease CFPE symptoms, infections are almost never fully eradicated. To elucidate antibiotic failure, we must first understand how relevant conditions in the airways, particularly around the time of CFPE, shape the microbiome and its antibiotic sensitivity.
CFPE may be characterized by increases in acidity (1), fermentation metabolites (2), and anaerobic bacteria (3). We characterized CF polymicrobial metabolic interactions between the fermenting microbe, Rothia muciliganosa, and the opportunistic pathogen, Pseudomonas aeruginosa (Chapter 1). Even in nutrient-rich medium designed to mimic sputum content, P. aeruginosa utilized Rothia-derived fermentation products to produce amino acids. As both host inflammation and microbial fermentation often result in drops in pH, we next determined that acidic pH is stressful for another CF opportunistic pathogen, Stenotrophomonas maltophilia, using transcriptomics and metabolomics (Chapter 2). S. maltophilia coped with low pH by expressing stress response genes and catabolizing amino acids to synthesize polyamines. Fermentation and subsequent drops in pH can be the result of hypoxia, which is a predominant condition in CF sputum (4). We used a combination of fluorescence lifetime imaging and spectral microscopy to study bacterial metabolism in oxygen gradients (Chapter 3).
To tackle the question of how antibiotics impact the microbiome, we need objective information about which antibiotics are reaching the infection-site. Given the diverse properties of antibiotics, the type and concentration of each antibiotic should be incorporated. We developed an LC-MS method to detect 18 antibiotics in 171 sputum samples and assessed the specificity of our LC-MS assay relative to subject self-reported usage (Chapter 4).