Caecal microbiota compositions from 7-day-old chicks reared in high-performance and low-performance industrial farms and systematic culturomics to select strains with anti-Campylobacter activity.

There is growing interest in exploring the chickens' intestinal microbiota and understanding its interactions with the host. The objective is to optimize this parameter in order to increase the productivity of farm animals. With the goal to isolate candidate probiotic strains, specific culturomic methods were used in our study to culture commensal bacteria from 7-days old chicks raised in two farms presenting long history of high performance. A total of 347 isolates were cultured, corresponding to at least 64 species. Among the isolates affiliated to the Firmicutes, 26 had less than 97% identity of their partial 16S sequence with that of the closest described species, while one presented less than 93% identity, thus revealing a significant potential for new species in this ecosystem. In parallel, and in order to better understand the differences between the microbiota of high-performing and low-performing animals, caecal contents of animals collected from these two farms and from a third farm with long history of low performance were collected and sequenced. This compositional analysis revealed an enrichment of Faecalibacterium-and Campylobacter-related sequences in lower-performing animals whereas there was a higher abundance of enterobacteria-related sequences in high-performing animals. We then investigated antibiosis activity against C. jejuni ATCC 700819 and C. jejuni field isolate as a first phenotypic trait to select probiotic candidates. Antibiosis was found to be limited to a few strains, including several lactic acid bacteria, a strain of Bacillus horneckiae and a strain of Escherichia coli. The antagonist activity depended on test conditions that mimicked the evolution of the intestinal environment of the chicken during its lifetime, i.e. temperature (37°C or 42°C) and oxygen levels (aerobic or anaerobic conditions). This should be taken into account according to the stage of development of the animal at which administration of the active strain is envisaged.
Competing Interests: This study received partial funding from the company Boehringer Ingelheim Animal Health, by which the co-author Christine ANDREONI was employed at time of the study. This does not alter our adherence to PLOS ONE policies on sharing data and materials.