The Impact of Spatial Network Topologies on the Adaptive Evolution of Pseudomonas aeruginosa Metapopulations

Abstract

Whether and how the spatial arrangements of populations can influence their dynamics of adaptation is not well understood. This is primarily because different theoretical frameworks comparing the dynamics of spread of a beneficial mutation in spatially structured and fully connected (well-mixed) populations often make contrasting predictions. Although classical population genetic theories find little to no impact of spatial structure on adaptation, Evolutionary Graph Theory (EGT) models that treat spatially structured populations as graphs show that specific patterns of connectivity or network topologies of a population can accelerate or decelerate adaptation compared to well-mixed populations. This thesis aims to improve our understanding about the impact of network topology of spatially structured populations on their dynamics of adaptive evolution. To do this, I first review the experimental literature till date and find that restricted migration in a spatially structured population tends to slow down the pace and extent of adaptive evolution when compared to a well-mixed population. I also find that there are no direct empirical tests that assess the impact of network topologies of populations on their adaptation. EGT predicts that certain topologies, specifically a four-patch star with bidirectional migration through a central hub to each of three peripheral populations, can accelerate the rate at which a beneficial mutation spreads through a population relative to an unstructured, well-mixed population. I directly test this prediction by tracking the dynamics of a beneficial mutation as it spreads in a four-patch metapopulation of Pseudomonas aeruginosa propagated by either star or well-mixed topology. I find that star topologies can accelerate adaptation but only when there is a high chance that the rare beneficial mutation will be lost to genetic drift. Next, I evaluate whether star topologies can still accelerate adaptation when more than one beneficial mutation is competing for fixation in a microbial population. I propagate clonal four-patch metapopulations of Pseudomonas aeruginosa by either star or well-mixed topologies under high or low beneficial mutation supply rate, as they adapt to the fluoroquinolone drug ciprofloxacin. I find that star topology can accelerate adaptation only when the beneficial mutation supply rate is kept low. By performing whole genome sequencing of the evolved metapopulations, I show that star topologies achieve this acceleration under low mutation supply because they substitute rare beneficial mutations with a higher repeatability than well-mixed topologies. In other words, evolution is accelerated because rare beneficial mutations are less likely to be lost in star topologies, not because the strength of selection itself increases. In sum, this thesis has successfully answered both the 'whether' and 'how' the network topologies of spatially structured populations can influence their dynamics of adaptive evolution. My work emphasizes the importance of including spatial structure as a key ecological complexity in evolutionary models of adaptation to more accurately predict the dynamics of adaptation in natural metapopulations, to improve the evolutionary forecasting and control of the spread of pathogenic and antibiotic resistance genes in spatially structured microbial populations such as biofilms, or through more complex transportation and hospital networks.