Final Report Summary - ZEBRAPROTECT (Molecular and ecological characterisation of bacterial community-mediated protection against infection in the zebrafish) All animals harbour diverse microbial populations mostly composed of commensal, non-pathogenic species. While these animal-associated ecosystems play important roles in development and physiology of their hosts they also play a key role in the outcome and control of infectious disease. Indeed, commensal microbiota are the first point of contact with pathogenic microorganisms and prevent invading or resident pathogens to settle and proliferate. A number of hypotheses have been proposed to explain these protective effects called bacterial interference or colonisation resistance. Commensal microbiota can directly compete for nutrients and adhesion sites with pathogens (barrier effect). They can also limit pathogen growth by induction or alteration of the host immune defense (immuno-stimulation). However, to date, it is not clear whether these interactions are a concerted effort of the community, or due to the presence of certain individual species (membership effect). Hence, although it is now established that commensal microbiota provide a critical defense against infection, mechanisms underlying this community protective effect remain poorly understood.The ZEBRAPROTECT project’s general aim was to characterise the molecular and ecological mechanisms underlying protection against pathogens provided by natural host microbiota. This bacterial community-mediated protective effect was being studied in an established and biologically controlled model of gnotobiotic zebrafish larvae challenged with infection by fish pathogens. Before the start of the project, a community-protection phenotype was identified in this zebrafish model developed in the laboratory: whilst axenic zebrafish died upon exposure to the lethal fish pathogen F. columnare, conventional fish survived as well as non-infected germ-free controls. The project’s more detailed aims were to explore whether the observed community-mediated protective effects were consistent with any of the current hypotheses in the field, and ZEBRAPROTECT was designed in work packages to address these hypotheses.First, in order to confirm that the protection phenotype was indeed due to the natural larva microbiota, we reconventionalised axenic larvae with bacteria from the zebrafish facility, and exposed these larvae to F. columnare. Reconventionalisation led to full protection of the germ-free larvae, confirming the protective role of the microbiota. In order to characterize this bacterial community responsible for the protection phenotype, we analysed the natural zebrafish larva microbiota by culture and culture-independent means. Analysis of the 16S bacterial content identified a bacterial community composed of strains that were also all successfully isolated by culture. In order to discern which of these bacterial species were responsible for the protection against the pathogen, we carried out experiments where axenic larvae were pre-incubated with either individual species identified in the protective community or with different combinations of species mixes, prior to exposure to F. columnare. The combination of all identified species fully restored protection against the pathogen, also confirming that the species we identified were indeed responsible for protection. Further, we identified 2 keystone species that were capable of protecting the larvae when added individually, and a mix of 8 species that did not protect on their own but as a mix. We also collected zebrafish eggs from another 4 facilities and tested whether the natural microbiota from larvae in these facilities would protect against F. columnare. We found that the microbiota from all facilities led to full protection against infection with the pathogen, and are currently awaiting the results of the bacterial content analysis of these facilities. In order to investigate whether the abundance of a particular species was responsible for protection, we carried out analysis of community structure dynamics, and are awaiting results. The molecular characterization of the killing mechanisms of F. columnare and the protection mechanisms of the 3 different protection scenarios identified proved challenging as neither for the pathogen or the keystone protector species genetic tools are available, and both have proven highly refractory to genetic manipulation. This work is currently ongoing, as the fellow’s contract was prolonged in the host laboratory in order to complete the project.Furthermore, we were interested whether the zebrafish immune response was contributing to the protection phenotype and whether there was evidence for immuno-stimulation of the host by the bacterial community. We investigated a number of routes such as the use of zebrafish reporter lines and immune effector monitoring. We did not see evidence for a particular cell-type dependent immune response, but observed induction of certain cytokines in response to commensal microbiota compared to axenic controls, and induction of a number of cytokines in response to F.columnare alone. Overall, we successfully identified and isolated the indigenous microbiota of zebrafish larvae, and demonstrated that these isolated natural host microbiota were capable of (re)producing community-mediated protection of the larvae against F. columnare. We were able to demonstrate that protection was observed with all microbiota from all larvae tested from different facilities, and that different protection mechanisms can exist within the natural host microbiota. Once all results are obtained, we will further be able to determine whether community structure within the host microbiota is important for protection and whether a “universal protective” element is observed in the larvae microbiota from all facilities. We also characterized the host immune response to infection, and observed an induction of pro-inflammatory markers upon pathogen exposure that appeared to be modulated by the addition of indigenous microbiota. These results are not only of a fundamental interest to ecological theory, but they also represent informative data and a robust model system that can be extrapolated to a range of applications from medical to industrial and ecological sectors. In particular, information gathered in this project on the nature of commensal-host interactions will ultimately contribute to the design of more efficient antibiotic treatments or alternative prevention strategies using knowledge-based engineering of host-beneficial microbiota.
