Final Report Summary - SOCMICROECO (Social evolution in microbial ecosystems) Bacteria typically live at high density in diverse communities in which cells strongly affect the growth and survival of their neighbours. The nature and strength of these interactions have been shown to be key to the productivity and stability of microbial communities [2,4], but little is understood about how they originate and evolve over time. It is becoming increasingly clear that understanding the community as a whole is impossible without developing a full understanding of the social interactions within these communities and their evolutionary dynamics. The aim of the project was to combine theory and experiments to create a predictive framework of evolving social interactions within microbial ecosystems. In addition to developing a theoretical model, we planned to compare the predictions of the model to a simple test case microbial ecosystem involving five species used for the treatment of industrial waste. Over the course of the project, we have advanced significantly both on the theoretical and empirical fronts. Importantly, we have published a review paper in a renowned journal [3], which has provided general theoretical predictions regarding social interactions in microbes, and has paved the way by providing a road map for future research in the field. We have then formalised the predictions of the review paper by developing a population genetic model that predicts the equilibrium frequencies of strains with different social traits. The results of this mathematical model have then been linked to a spatially-explicit computational simulation of multi-species groups of microbes. On the empirical side, we have worked toward setting up an experimental system involving five bacterial strains to digest toxic metal-working fluids (MWF) that are used in industrial applications. In the future, we will continue to work toward disentangling the interactions between these five species and understanding their evolutionary fate. In parallel, we have also developed the necessary tools to analyse experimental data from such multi-species microbial communities. While these tools were developed with collaborators to analyse cheese-associated communities [1], they are general and will similarly apply to the MWF community and other experimental systems.The review paper together with the mathematical and computational models propose a null model, which we have called the "genotypic view" that applies generally to all microbial communities: while cooperation is common between cells of the same genotype, it will only evolve between different genotypes under restrictive ecological conditions. In contrast, different genotypes will typically compete. This remains a hypothesis. However, the theoretical work conducted in this project has provided increasing support for this view, while outlining the conditions necessary for cooperation to evolve between species. Specifically, the models predict that cooperation can be favoured between two species if they do not share resources, if the cooperative traits are cheap and if mechanisms exist with which to direct benefits toward cooperative cells of the partner species (e.g. adhesion). Even if all these conditions are met, cooperation is nevertheless expected to be unstable from an evolutionary point of view.Our empirical work has made two contributions: the first has been to set up a five-species microbial consortium with which to study the evolutionary dynamics of interspecies interactions. Our preliminary experiments have revealed factors that may influence these interactions, such as differences in growth rates or propensity to form biofilms. The second contribution has been to devise a simple model based on differential equations with which to describe the simultaneous growth of multiple species in co-culture. Fitting data to this model allows us to characterise interactions within microbial systems composed of few species. Our analysis of one such microbial system sampled from cheese rind has revealed that most inter-species interactions were weak, and none of the species pairs showed mutually cooperative growth patterns. We next plan to apply these methods to the MWF consortium.Over the duration of the project, interest in understanding and analysing microbial communities has increased dramatically, both from the basic scientific perspective of understanding complex systems, and due to the range of potential practical applications. My ambition is that the work we have accomplished in this project will result in a model system for studying microbial communities, with clear experimental results together with a theoretical toolbox that will allow us to analyse and predict the behaviour of the community. This will be useful to many in the field that focus either on the theoretical or the experimental side of studying microbial communities. Apart from the scientific impact, understanding the MWF system will allow us to control and optimise current microbial systems used to digest this and other toxic waste. This has the potential to significantly reduce the pollution load compared to current rates, in a predictable and stable manner that is robust to evolutionary change and contamination. The combination of theory and experiments will allow me to make recommendations on how to achieve such improved conditions. In the near future, then, it will be possible to form industrial collaborations to scale-up the experiments toward improving waste treatment in the large industrial bioreactors.1. Button J, Dutton RJ (2012) Cheese microbes. Curr Biol 22:R587-9.2. May RM (1973) Stability and Complexity in Model Ecosystems. Princeton Univ Press, 265 pp.3. Mitri S, Foster KR (2013) The genotypic view of social interactions in microbial communities. Ann Rev Gen, 47:265-91.4. Montoya JM, Pimm SL, Solé RV (2006) Ecological networks and their fragility. Nature 442:259-64.
