Center for Causal Discovery
University of Pittsburgh, Carnegie Mellon University,
Pittsburgh Supercomputing Center and Yale University

Title: Applying graphical causal models to the study of Pseudomonas aeruginosa pathoadaptation during cystic fibrosis chronic rhinosinusitis

Speaker: Catherine R. Armbruster  – Postdoctoral Scholar, The Laboratory of Jennifer M. Bomberger, Department of Microbiology and Molecular Genetics, University of Pittsburgh School of Medicine, Pittsburgh, PA, USA.

Time: 10:00 AM on Thursday, July 19, 2018
Location: Room 407A/B, Baum

Abstract: Complex polymicrobial communities are present in the respiratory tract of cystic fibrosis (CF) patients, but chronic Pseudomonas aeruginosa infections account for the majority of morbidity and mortality. Chronic rhinosinusitis (CRS) is highly prevalent in CF; recent studies suggest the paranasal (PN) sinuses are a site of primary colonization and dissemination of P. aeruginosa into the lower airways. Despite a robust host immune response and aggressive antimicrobial treatment, P. aeruginosa adapts to these conditions via mutations that promote chronic respiratory infections, but little is known about specific pressures in the PN sinuses or lungs that repeatedly select for these mutations across many patients. Respiratory viral infections account for at least 40% of pulmonary exacerbations of CF adults and viral infection has been linked with the development of chronic infections that are dominated by P. aeruginosa. The Bomberger Lab recently described how respiratory virus co-infection promotes colonization of the airway epithelium by P. aeruginosa, through a dysregulation of host nutritional immunity, leading to increased iron released from airway epithelial cells. This increased iron induced P. aeruginosa to rapidly form biofilms, a form of growth in which bacteria are notoriously recalcitrant to killing by antimicrobials and immune effectors. However, the impact of viral infections on pathoadaptation of P. aeruginosa, and on the respiratory tract microbiome as a whole, is largely unknown.

Our goal was to investigate evolution of P. aeruginosa during CF CRS, and identify pathoadaptive traits associated with persistence of P. aeruginosa in the PN sinuses, sinus microbiota changes, or several measures of clinical disease severity. In a 2-year longitudinal study of CF adults, we collected sinus swabs for bacterial culture and microbiome analysis at quarterly clinic visits and during exacerbations. We genotyped P. aeruginosa isolates from patients and screened for pathoadaptive in vitro phenotypes observed in P. aeruginosa CF lung isolates, including biofilm matrix overproduction, loss of acute virulence factors, and antibiotic resistance. Whole genome sequencing of four patients’ longitudinal isolates revealed that while specific mutations varied across all patients, all lineages acquired non-synonymous mutations in iron acquisition, amino acid metabolism, interbacterial competition, antibiotic resistance, and global regulatory proteins, suggestive of mutational parallelism between patients similar to observations in the lung. Preliminary searches of a mixed dataset including patient clinical variables and bacterial in vitro phenotyping in Tetrad, a program which searches for causal explanations represented by directed graphs, suggested a causal relationship between respiratory viral infection and levels of iron and transferrin present in patient sinus samples, in agreement with our previous in vitro and mouse studies of transferrin mistrafficking during virus co-infection. Continuing efforts are aimed at using graphical causal modeling on a larger combined dataset that also includes microbiome and bacterial whole genome sequencing information, to generate hypotheses regarding drivers of CF respiratory disease progression that are testable in our in vitro models. The long-term goal of this work is to develop interventions that target initial colonization of the paranasal sinuses by P. aeruginosa, as a strategy to prevent downstream colonization of the lower respiratory tract by P. aeruginosa in CF.


If you need to join the CCD Colloquium remotely, please follow the instructions below.

To join the meeting on a computer or mobile phone:

DBMI has invited you to a video meeting.
Just want to dial in on your phone?
1) +1.408.740.7256 (United States)
+1.408.317.9253 (Alternate number)
Global Numbers:
2) Enter Meeting ID: 148 972 068
3) Press #
Want to test your video connection?