Abstract

In agriculture, the plants and animals that are bred come from long-term artificial selection procedures that aim at producing individuals with desirable traits. Such procedures can be applied at levels of biological organization higher than the individual such as the community level, which is of particular interest in the field of microbial community engineering. Indeed, microbial communities play key roles in agriculture but also in the environment, health and industry, and a growing body of literature shows promising results regarding the artificial selection of microbial communities for desirable functions. However, for artificial selection procedures at the community level to be efficient, there is a need for improvements of the process, which will be achieved by a better understanding of community functioning and evolution.

The aim of this work was to identify the ways of improving the artificial selection of microbial communities through a better understanding of the consequences of considering communities as selection units and to shed light on the knowledge that it could bring to our understanding of community ecology and evolution.

In a first experiment, we applied an artificial selection procedure on a soil microbial community and showed that the reproduction step, i.e. the way offspring communities are created from parental ones, can be a key determinant of the outcome of the artificial selection as it can influence community structure and diversity. Pooling several parental communities seemed to be suitable to increase the level of a function under selection. In a second experiment, we applied an artificial selection procedure on synthetic bacterial communities varying for their initial specific richness level. We showed that community diversity positively influenced the level of the function under selection and also affected community evolutionary dynamics. We also had evidence that it could affect the efficiency of the artificial selection, but the relationship was not linear, and we evidenced a trade-off between increasing community richness and preserving between-community dissimilarity in composition. This trade-off could limit the positive effect of an increase in richness on artificial selection efficiency. In a third experiment, we assessed whether the pairwise interspecific interactions evolved during the experimental evolution of bacterial communities. Our results indicated that the evolution of the interspecific interactions was widespread in community evolution as it was detected in 50% of the studied communities. Moreover, even in the communities in which the interactions did not evolve themselves, we showed that they influenced the evolutionary responses of the strains and communities which was both abiotic environment- and community-dependent.

All together these results highlighted the close links between community ecology and evolution and gave several avenues for improving artificial selection: choosing the reproduction method according to the desired direction of change in a function, promoting community diversity, promoting the evolution of the interspecific interactions. Improving our understanding of community eco-evolutionary dynamics will open the way to an efficient artificial selection of communities and its use as a tool for microbial community engineering.

Keywords: artificial selection, community evolution, diversity, interspecific interactions, microbial community

Composition du Jury :

PRIGENT-COMBARET            Claire Directrice de recherche, CNRS, Villeurbanne           Rapportrice

SANCHEZ Alvaro                  Associate professor, Yale University, West Haven              Rapporteur

BRETAGNOLLE François        Professeur, Université de Bourgogne, Dijon                      Examinateur

SICARD Delphine                 Directrice de recherche, INRAE, Montpellier                     Examinatrice

SIMONIN Marie                   Chargée de recherche, INRAE, Angers                             Examinatrice 

JEANDROZ Sylvain              Professeur, AgroSup, Dijon                                             Invité 

BLOUIN Manuel                   Professeur, AgroSup, Dijon                                            Directeur de thèse

SPOR Aym                          Chargé de recherche, INRAE, Dijon                                Codirecteur de thèse