Bacterial colonizers of Nematostella vectensis are initially selected by the host before interactions between bacteria determine further succession

Le résumé fourni par la source

Abstract The microbiota of multicellular organisms undergoes considerable changes during development but the general mechanisms that control community assembly and succession are poorly understood. Here, we use bacterial recolonization experiments in Nematostella vectensis as a model to understand general mechanisms determining bacterial establishment and succession. We compared the dynamic establishment of the microbiome on the germfree host and on inert silica. Following the dynamic reconstruction of microbial communities on both substrates, we show that the initial colonization events are strongly influenced by the host but not by the tube, while the subsequent bacteria-bacteria interactions are the main cause of bacterial succession. Interestingly, the recolonization pattern on adult hosts resembles the ontogenetic colonization succession. This process occurs independently of the bacterial composition of the inoculum and can be followed at the level of individual bacteria, suggesting that priority effects are neglectable for early colonization events in Nematostella . To identify potential metabolic traits associated with initial colonization success and potential metabolic interactions among bacteria associated with bacterial succession, we reconstructed the metabolic networks of bacterial colonizers based on their genomes. These analyses revealed that bacterial metabolic capabilities reflect the recolonization pattern, and the degradation of chitin might be a selection factor during early colonization of the animal. Concurrently, transcriptomic analyses revealed that Nematostella possesses two chitin synthase genes, one of which is upregulated during early recolonization. Our results show that early colonization events are strongly controlled by the host while subsequent colonization depends on metabolic bacteria-bacteria interactions largely independent of host development.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.