GL14
Phenogenomic analyses: linking mycobacterial behaviours to molecular mechanisms
L Boeck(1,2)
1:Department Biomedicine, University of Basel; 2:Pulmonary Medicine, University Hospital Basel
Mycobacterium abscessus is a newly emergent multidrug-resistant pathogen causing increasing numbers of often incurable pulmonary infections, particularly in patients with Cystic Fibrosis. However, the genetic determinants underlying M. abscessus infection, virulence and how M. abscessus survives long periods of multidrug treatment are poorly understood.
To address this gap, we developed phenogenomic analyses, a multi-modular strategy aimed to capture the complexity of genome-wide associations. We phenotyped 331 isolates across 58 phenotypic dimensions including planktonic growth, antibiotic resistance, macrophage infection, in vivo infection of Drosophila melanogaster and clinical outcomes. Our analysis revealed isolate clusters with distinct phenotypic traits which are related to clinical outcomes. We combined conventional genome-wide association studies with computational modelling to define causal genetic variants and the study of variant coevolution to identify epistatic gene networks. Several revealed mechanisms and networks, including previously unknown virulence factors, were validated by CRISPR-based silencing.
Additionally, we explored antibiotic tolerance, which allows increased bacterial survival during normally lethal conditions. To assess antibiotic killing at large scale, we established Antimicrobial Single-Cell Testing (ASCT), a single-cell imaging-based technology. We found that drug tolerance is defined by the genetic background but different from drug resistance. Moreover, drug tolerance was an independent predictor of clinical outcomes, highlighting its potential use as a novel marker of antibiotic activity. Finally, by applying phenogenomic analyses to drug tolerance phenotypes in M. abscessus, we identified several molecular mechanisms that likely confer multidrug tolerance.