EVOLVIR Project

The bacterial pathogen Legionella pneumophila, the etiological agent of Legionnaires’ Disease (LD), has evolved virulence mechanisms that allow it to replicate within protozoa, its natural host. Many of these tactics also enable its replication inside human lung macrophages which causes severe and often fatal pneumonia. Among the Legionella genus, only few species cause disease in humans, such as L. pneumophila serogroup 1 (Lp1) responsible for 85% of LD worldwide. As this prevalence cannot be explained by predominance in the environment, we can therefore ask what are the evolutionary processes that led to the emergence of the pathogenicity of Lp1?
To answer this question, we propose to develop experimental evolution strategies with L. pneumophila with the objective of evolving its virulence properties. We will initiate several lineages in parallel all founded from a common ancestor derived from Lp1 which presents an attenuated virulent phenotype towards both amoebae and macrophages. The objective and the originality of our project is to re-evolve this attenuated virulent strain under five different environments: axenic liquid media, Acanthamoeba castelanii amoebae, the human macrophage-like cell line U937, an alternation of the amoebae and macrophage hosts and finally a co-culture with another Legionella strain inside amoebae in order to investigate the impact of horizontal gene transfer between both strains that may occur during evolution experiment. Bacterial lineages will be propagated for hundreds of generations by serial passages with large-population bottlenecks. These experimental conditions are known to select for the most adapted individuals in each environment resulting in the increase of bacterial fitness (ability of one genotype to give offspring). In the context of pathogenicity, increased fitness can lead to alteration in both virulence and host spectrum properties.
After several hundred generations of evolution, the evolved clones will be compared to the ancestor for various phenotypic traits. First we will measure the abilities of the evolved lineages to infect different hosts as well as in axenic medium. This will enable us to determine if evolved clones have evolved a broader or narrower host spectrum (evolution toward generalist or specialist pathogen). When significant differences will be detected, we will further investigate which steps of the infectious cycle are differentially affected. Then, the genetic modifications underlying this phenotypic diversification will be investigated. Genome of appropriately chosen evolved clones isolated from the different environments will be sequenced and compared to the ancestor as well as changes in global gene expression profiles. These approaches aim to make connections between genetic changes and phenotypic outcomes and overall to identify the target genes of the virulence/host spectrum evolution. In parallel to the experimental evolution, we propose the computational reconstruction of the evolutionary history of Legionella based on the comparative analysis of several hundred genomes of Legionella and phylogenetic methods and tools developed by partner 2. We will match the evolution of gene repertoires along the history of Legionella with their reported phenotype to thus identify candidate genes of pathogenicity. These candidates will be compared to those obtained in experimental evolution strategies to complement or restrict our final list of candidate genes for further testing.
Finally, the target genes of the evolution identified by both computational and experimental strategies will be deleted in the ancestor’s genome or replaced by the evolved allele. This will generate a set of isogenic strains that will be assessed for the various phenotypic traits. This enables to rigorously identify the molecular bases of the adaptation and overall to validate these genes or functions as markers of increased virulence.