Fig. 5: AIPs regulate IBD mouse fecal-derived ex vitro microbial community.

a Graphical depiction of the ex vivo experimental setup. b Relative abundance of microbial genera after 48 h of treatment with vancomycin, BF_280_c, and BF_398_c at concentrations of 0.1, 1, and 10 μg/mL. DMSO was used as a blank control. Only 20 abundant species are displayed. c Bar plots show the alpha diversity of the microbial community at the species level, as represented by the Shannon diversity index. Data (n = 3) are mean ± standard deviation. Significances between treatment groups and blank group were indicated by using two-sided Welch’s t-test. The exact p values are as follows (order as 0.1, 1, 10 μg/mL): vancomycin, 0.082, 0.054, 0.85e−3; BF_280_c, 0.13, 0.067, 0.25e-3; BF_398_c, 0.010, 0.23, 0.76e-2. Three biological replicates (n = 3) were included for each group. d Principal coordinate analysis (PCoA) of microbial community based on the Bray-Curtis dissimilarity at the species level. The enclosing ellipses are estimated using the Khachiyan algorithm by R function “geom_mark_ellipse”, representing the distinct clustering of groups. e Relative abundance of four representative taxa. All box plots include center lines representing the median, box limits representing upper and lower quartiles, whiskers representing the 1.5x interquartile range, and points representing outliers. The exact p values are as follows (order as vancomycin, BF_280_c, BF_398_c): Paramuribaculum intestinale, 0.41e−3, 0.036, 0.039; Oscillospiraceae bacterium, 0.0020, 0.021, 0.013; Heminiphilus faecis, 0.35e−3, 0.0044, 0.0076; Prevotella sp MGM1, 0.55, 0.036, 0.027. MaAsLin2, a tool relying on general linear models to find multivariable association, was used to compare the species between vancomycin/AIP and the control group. The Benjamini-Hochberg (“BH”) method was adopted to adjust p values for multiple comparisons. Three biological replicates (n = 3) were included for each group. *p < 0.05; **p < 0.01; ***p < 0.001; ns, not significant.