Co-infection shapes pathogen fitness in a genetically diverse population of Pseudomonas viridiflava isolates from Arabidopsis

Plants are often co-infected by multiple strains of a pathogen. Interactions between strains range from competitive to synergistic and can affect pathogen fitness. However, little is known about how the costs and benefits of co-infection vary across naturally co-occurring strains, the extent to which these costs and benefits are general or strain-specific, and the identity and evolutionary history of loci shaping these interactions. Quantifying these patterns is essential for understanding the importance of co-infection in pathogen ecology and evolution. We are measuring effects of co-infection in a population of a bacterial pathogen, Pseudomonas viridiflava, isolated from Arabidopsis thaliana, using reporter- and sequence-based methods for high-throughput quantification of pathogen abundance. Across co-infections with hundreds of pairwise strain combinations in gnotobiotic plants, we find that between-strain interactions can explain as much variation in pathogen fitness as focal strain genotype alone. Competition is pervasive and predicted by growth rate in single infections. However, we find combinations of strains that deviate from expectations based on growth rate, including both inhibitory and putatively beneficial interactions. We are now using GWAS to identify loci underlying these outcomes. The prevalence of strain-dependent effects of co-infection underscores the importance of interactions between pathogen genomes as a key force in pathogen ecology and evolution