Cooperation and reversibility in microbial evolution

Quantitative experiments with microbes are providing new opportunities to test classic theories in evolutionary dynamics. In this talk I will present projects which probe two such theories, the conditions required for the evolution of cooperative behaviors and the degree to which evolution is reversible. In the first project, we studied how yeast secrete the enzyme invertase in order to cooperatively break down and consume the sugar sucrose. We found that a cheater strain (invertase knockout) is able to take advantage of and outgrow a population of cooperators, but only when present at low frequencies. We therefore observe coexistence at equilibrium between cooperator and cheater strategies in well-mixed environments. Such coexistence is the hallmark of the snowdrift game, in which players have the incentive to cheat if their opponents are cooperating [1]. In the second project, we are investigating the evolution of antibiotic resistance in bacteria in the case that there are trade-offs between resistance against two antibiotic treatments. In particular, we are probing five point mutations in the beta-lactamase gene that collectively confer resistance to the second generation antibiotic cefotaxime [2]. However, adapting to cefotaxime also increases the susceptibility to beta-lactamase inhibitors. By measuring the sensitivity of the 32 strains (containing all possible combinations of the five mutations) in both antibiotic conditions we can experimentally measure the two fitness landscapes and probe the degree to which evolution against different antibiotics is reversible.

[1] Gore et al, Nature 459 (2009)

[2] Weinreich et al, Science 312 (2006)