Fig 2: Post-antibiotic supplementation with Akkermansia muciniphila worsened the gut barrier and aggravated colonic and systemic inflammation during tumorigenesis. (A) Colonic relative mRNA levels of Tjp1, Ocln, and Cdh1 in the four groups. (B) Colonic relative mRNA expression level of MUC2 in the four groups. (C) Serum LBP level in the four groups. (D) Colonic relative mRNA expression levels of Il1b, Il6, and Tnfa in the four groups. (E) Serum levels of IL-1ß, IL-6, and TNF-a in the four groups. Data are shown as means ± SEM. *P < 0.05.

Fig 3: Aberrant fecal metabolic patterns in children with T1D.a Heatmap of the relative abundance of 26 significantly different metabolites between the NC and T1D group. Metabolites with VIP = 1 and fold change =2 or fold change =0.5 were considered differential metabolites. b, c The concentration of total and seven fecal short-chain fatty acids, respectively (total SCFAs, p = 0.004; AA, p = 0.024; BA, p = 0.005). d–g The serum concentration of d GLP-1 (p < 0.001), e FGF19 (p = 0.002), f LBP (p = 0.003), and g IL-1ß (p = 0.042) in the NC and T1D group (NC: n = 40, T1D: n = 34). h Heatmap of the Spearman’s correlation between 28 discriminatory metabolites and 35 key bacteria species as well as clinical parameters (*FDR < 0.05). The red squares indicate positive correlations, whereas the blue squares indicate negative correlations. i Metabolomic and metagenomic markers for detecting the T1D group from the NC group were identified from Random-forest classifiers based on the combination of dual-omics markers. Markers are ranked in descending order of their importance to the accuracy of the model. j Receiver operating characteristic (ROC) curves and their corresponding area under curve (AUC), employing the combination of dual-omics markers. AA acetic acid, BA butyric acid, PA propionic acid, IBA isobutyric acid, VA valeric acid, IVA isovaleric acid, HA hexanoic acid. GLP-1 glucagon-like peptide 1, FGF19 fibroblast growth factor 19, LBP lipopolysaccharide-binding protein. For metabonomic analysis, NC: n = 77, T1D: n = 64 in the discovery set, NC: n = 29, T1D: n = 29 in the validation set. Violin plots show the median, quartiles, and min/max values. Two-sided Wilcoxon rank-sum test. *p < 0.05, **p < 0.01, ***p < 0.001. Source data are provided as a Source Data file.

Fig 4: Effects of the JZG on bacterial function. (a) PCoA plot of the bacterial functions based on Bray-Curtis distances; (b) Venn diagrams of differentially KEGG metabolic pathways; (c) heatmap of different pathways between groups; (d) H2S content in feces; (e) plasmatic concentration of LBP; (f) fecal SCFAs in cecal contents. #p < 0.05 compared with the NCD group. *p < 0.05 compared with the HFD group; (g) Spearman's correlations between gut microbial community at the family level and vital metabolic parameters linked to NAFLD; +p < 0.05 and *p < 0.01. Violin plots display medians with interquartile ranges.

Fig 5: Berberine repaired the colonic barrier damage in GVHD mice. (A-C) Expression of Occludin in each group (200×) and (D-F) expression of ZO1 in each group (200×) (N = 10/group). Protein expression of Claudin-1 (G) and Claudin-2 (H) in colon tissue (N = 3/group). (I) Serum LBP concentration was determined by ELISA (N = 5/group). Data are shown as mean ± SD and were analysed by one-way ANOVA. * p < 0.05; ** p < 0.01