Abstract
In vitro fermentation strategies with fecal inocula are considered cost-effective methods to gain mechanistic insights into fecal microbiota community dynamics. However, all in vitro approaches have their limitations due to inherent differences with respect to the in vivo situation mimicked, introducing possible biases into the results obtained. Here, we aimed to systematically optimize in vitro fermentation conditions to minimize drift from the initial inoculum, limit growth of opportunistic colonizers, and maximize the effect of added fiber products (here pectin) when compared to basal medium fermentations. We evaluated the impact of varying starting cell density and medium nutrient concentration on these three outcomes, as well as the effect of inoculation with fresh vs. stored fecal samples. By combining GC-MS metabolite profiling and 16 s rRNA gene-based amplicon sequencing, we established that starting cell densities below 10(10) cells/ml opened up growth opportunities for members the Enterobacteriaceae family. This effect was exacerbated when using fecal samples that were stored frozen at -80°C. Overgrowth of Enterobacteriaceae resulted in lowered alpha-diversity and larger community drift, possibly confounding results obtained from fermentations in such conditions. Higher medium nutrient concentrations were identified as an additional factor contributing to inoculum community preservation, although the use of a less nutrient dense medium increased the impact of fiber product addition on the obtained metabolite profiles. Overall, our microbiome observations indicated that starting cell densities of 10(10) cells/ml limited opportunities for exponential growth, suppressing in vitro community biases, whilst metabolome incubations should preferably be carried out in a diluted medium to maximize the impact of fermentable substrates.