Optimizing ex vivo culture conditions to study human gut microbiome.

The inter-individual variations of gut microbiome contribute to the different responses toward drug therapy among populations, developing a reliable ex vivo culture method for mixed bacteria is the urgent need for predicting personal reaction to drug therapy. Unfortunately, very few attentions have been paid to the bias that could be introduced during the culture process for mixed bacteria. Here we systemically evaluated the factors that may affect the outcomes of cultured bacteria from human feces. We demonstrated that inter-individual difference of host gut microbiome was the main factor affecting the outcomes of cultured bacteria, followed by the culture medium and time point. We further optimized a new medium termed GB based on our established multi-dimensional evaluation method, which could mimic the status of in situ host gut microbiome to the highest extent. Finally, we assessed the inter-individual metabolism by host gut microbiome from 10 donors on three frequently used clinical drugs (aspirin, levodopa and doxifluridine) based on the optimized GB medium. Our results revealed obvious variation in drug metabolism by microbiome from different donors, especially levodopa and doxifluridine. This work suggested the optimized culture medium had the potential for exploring the inter-individual impacts of host gut microbiome on drug metabolism.