The University of Edinburgh and Department of Zoology, Oxford Unversity, Oxford OX1 3PS, UNITED KINGDOM.
In recent decades we have learned that cooperation is an important and pervasive feature of microbial life. This revelation raises exciting possibilities. On the one hand, we can now augment our understanding of how social phenomena evolve by using microbial model systems to test our theories. On the other hand, we can use concepts from social evolution to gain insight into the biology of the microbes we hope to control or kill.
In this thesis I explore both possibilities. First, I consider the theoretical problem of how and when microbial cooperation might be subject to frequency- and density-dependence. Formerly, vague theory and a scant, sometimes contradictory empirical literature made it unclear when such patterns could be expected. Here, I develop theory tailored to a microbial context, and in each case, I test key predictions from the theory in laboratory experiments, using as my model trait the production of siderophores by the bacterium Pseudomonas aeruginosa.
Secondly, I consider the ecological consequences of cooperator-cheat dynamics in the context of an infection. Specifically, I use experimental infections of diverse host models to investigate the role of two cooperative traits, siderophore production and quorum sensing, in the pathogenesis of P. aeruginosa. When a successful infection requires cooperation among pathogens, theory predicts that conflict among coinfecting strains can undermine cooperation and hence decrease virulence; whereas, in the absence of cooperation, conflict could lead to heightened exploitation and hence increased virulence. This exciting idea has received little empirical attention to date but here I address this using multiple pathogen strains, multiple social traits, and multiple model hosts.