AUTHOR=Zhang Chen-Yang , Peng Xin-Xin , Shao Hao-Qing , Li Xiao-Ya , Wu Yi , Tan Zhou-Jin TITLE=Gut Microbiota Comparison Between Intestinal Contents and Mucosa in Mice With Repeated Stress-Related Diarrhea Provides Novel Insight JOURNAL=Frontiers in Microbiology VOLUME=12 YEAR=2021 URL=https://www.frontiersin.org/journals/microbiology/articles/10.3389/fmicb.2021.626691 DOI=10.3389/fmicb.2021.626691 ISSN=1664-302X ABSTRACT=
Repeated stress-related diarrhea is a kind of functional bowel disorders (FBDs) that are mainly stemming from dysregulation of the microbiota–gut–brain axis mediated by a complex interplay of 5-hydroxytryptophan (5-HT). Intestinal content and intestinal mucosa microbiota belong to two different community systems, and the role of the two microbiota community systems in repeated stress-related diarrhea remains largely unknown. In order to ascertain the difference in composition and the potential function between intestinal content and intestinal mucosa microbiota response on repeated stress-related diarrhea, we collected intestinal contents and mucosa of mice with repeated stress-related diarrhea for 16S rRNA PacBio SMRT gene full-length sequencing, and with the digital modeling method of bacterial species abundance, the correlations among the two microbiota community systems and serum 5-HT concentration were analyzed. We found that the microbiotal composition differences both in intestinal contents and mucosa were consistent throughout all the phylogenetic ranks, with an increasing level of resolution. Compared with intestinal content microbiota, the diversity and composition of microbiota colonized in intestinal mucosa are more sensitive to repeated stress-related diarrhea. The PICRUSt2 of metagenomic function analysis found that repeated stress-related diarrhea is more likely to perturb the intestinal mucosa microbiota metagenomic functions involved in the neural response. We further found that the mucosal microbiota-based relative abundance model was more predictive on serum 5-HT concentration with the methods of machine-learning model established and multivariate dimensionality reduction (