Meta-analysis of the Microbial Diversity Cultured in Bioreactors Simulating the Gut Microbiome

Understanding the intricate ecological interactions within the gut microbiome and unravelling its impact on human health is a challenging task. Bioreactors are valuable tools that have contributed to our understanding of gut microbial ecology. However, there is a lack of studies describing and comparing the microbial diversity cultivated in these models. This knowledge is crucial for refining current models to reflect the gastrointestinal microbiome accurately. In this study, we analysed the microbial diversity of 1512 samples from 18 studies available in public repositories that employed cultures performed in batches and various bioreactor models to cultivate faecal microbiota. Community structure comparison between samples using t-distributed stochastic neighbour embedding and the Hellinger distance revealed a high variation between projects. The main driver of these differences was the inter-individual variation between the donor faecal inocula. Moreover, there was no overlap in the structure of the microbial communities between studies using the same bioreactor platform. In addition, α-diversity analysis using Hill numbers showed that highly complex bioreactors did not exhibit higher diversities than simpler designs. However, analyses of five projects in which the samples from the faecal inoculum were also provided revealed an amplicon sequence variants enrichment in bioreactors compared to the inoculum. Finally, a comparative analysis of the taxonomy of the families detected in the projects and the GMRepo database revealed bacterial families exclusively found in the bioreactor models. These findings highlight the potential of bioreactors to enrich low-abundance microorganisms from faecal samples, contributing to uncovering the gut microbial "dark matter".