Fig. 7
Bacteroides fragilis-derived 3-phenylpropionic acid confers the beneficial effects on the intestinal epithelial barrier function. a Activities of serum diamine oxidase (Ctrl control; 1 to 18, the first metabolite candidate to the 18th metabolite candidate and the detailed metabolite names were shown in the Methods section). b Levels of serum endotoxin. c Levels of serum D-lactic acid. d Venn diagram analysis of metabolite candidates that decrease the serum diamine oxidase activities, endotoxin levels, and D-lactic acid levels, respectively. The numbers of corresponding metabolite candidates were shown in the Venn diagram. e Relative abundance analysis of serum 3-phenylpropionic acid in mice (Ctrl control, BF Bacteroides fragilis). f Plasma FD4 levels (3-phenylpropionic acid, 3-PPA). g Lactulose/mannitol ratio. h Representative western blotting of ZO-1, E-cadherin, Connexin 43, and β-tubulin in the jejunal epithelium of mice. i–k Quantitation of the ZO-1 (i), E-cadherin (j), and Connexin 43 (k) levels normalized to β-tubulin levels. l The OPLS-DA of fecal metabolites. m The volcano plot analysis of fecal metabolites. n The enrichment analysis of KEGG pathways by differential fecal metabolites. o Relative abundance analysis of fecal 3-phenylpropionic acid in mice. The data are presented as the mean ± SEM and evaluated by one-way ANOVA with adjustment for multiple comparisons in a–c (n = 5). The data are presented as the mean ± SEM and evaluated by Student’s t-test in f (n = 8), g (n = 8), and i–k (n = 3). In addition, the data were assessed by the VIP ≥ 1 from OPLS-DA, absolute Log2 (fold change) ≥ 1, and p-value < 0.05 in m (n = 6). **p < 0.01, *p < 0.05