F. nucleatum further aggravated inflammation and gut barrier damage in DSS-induced colitis mice

To investigate whether F. nucleatum promotes inflammation and gut barrier damage in DSS-induced colitis mice, ATCC25586 strains were orally administrated daily to SPF C57BL/6J mice besides 2.5% DSS solution intake. The animal experiment procedure is shown in Fig. 1A. Body weight, survival rate, and disease activity index were monitored among groups. In HC mice, the body weight increased gradually over time, DSS and OR mice witnessed a slight increase at the first four days, and decreased from the fifth day. The body weight of OR mice decreased obviously in contrast with HC and DSS mice at the end of the experiment (p<0.01, Fig. 1B). All mice survived in the HC group, which was significantly higher than DSS and OR mice. In addition, DSS mice manifested a higher survival rate than OR mice (p<0.01, Fig. 1C). The disease activity index of OR mice was significantly higher than HC and DSS mice (p<0.05, Fig. 1D). HE staining showed the intestinal structure of HC mice was continuous and intact, with no inflammatory cell infiltration. Disrupted intestinal structure and inflammatory cell infiltration were observed in DSS mice. In the OR group, extensive mucosal damage and severe gland mucosal loss, with the confluence of inflammatory cell infiltration were observed. Histopathological scores were calculated as shown in Fig. 1E. Immunohistochemical staining revealed that Occludin protein was continuously expressed in the epithelium in HC mice, but the expression was interrupted in DSS and OR mice. In addition, the expression in DSS mice was significantly higher than in OR mice (p<0.05, Fig. 1E). We also observed increased levels of inflammatory cytokines TNF-α, IL-1β, and IL-6 in two colitis groups (Fig. 1F). Moreover, the cytokines levels were remarkably increased in OR mice in contrast with DSS mice (p<0.05, Fig. 1F).

Fig. 1

Oral-infected F. nucleatum further aggravated inflammation and gut barrier damage in DSS-induced colitis mice. A, A flowchart of animal experiments in this study. B, Variations of mice body weight among HC, DSS, and OR groups. C, Survival rate of the three groups (%). D, Disease activity index scores of the three groups. E, Representative images of histopathological by HE and immunohistochemical staining for Occludin protein. The histopathological scores and relative expression of Occludin protein were also calculated. F, Colonic levels of TNF-α, IL-1β and IL-6 by ELISA. B, Data are presented as mean ± S.D. p values were determined by analysis of variance, **p<0.01 vs. HC, ##p<0.01 vs. OR. C, Data were analyzed by Pearson’s Chi-square test. **p<0.01 vs. HC, ##p<0.01 vs. OR. D-F, Data are presented as mean ± S.D. p values were determined by analysis of variance, *p<0.05, **p<0.01

F. nucleatum promotes dysbiosis in DSS-induced colitis mice

Next, we sought to investigate the effect of F. nucleatum on disturbing the microbiota in DSS mice. We performed oral incubation of ATCC25585 strains to C57BL/6J mice. The Venn diagram analysis of OTUs revealed that 408 OTUs were identified in the three groups. HC, DSS, and OR groups contained 30, 5, and 11 unique OTUs, respectively (Fig. 2A). The species rank curve revealed a smoother and wider curve in the HC group, and the sharpest decline was found in the OR group (Fig. 2B). The rarefaction curve indicated the sequencing quality was in according with the requirement for further analysis (Fig. 2C). We further analyzed the fecal microbiota compositions at phylum and genus levels (Fig. 2D and E). The dominant bacterium in HC mice were Firmicutes, Bacteroidota, and Verrucomicrobiota. However, an increased abundance of Firmicutes and Proteobacteria was found in DSS and OR mice. The dominant bacterium in DSS and OR mice were Firmicutes, Bacteroidota, and Proteobacteria. At the same time, alpha and beta diversity were utilized to evaluate the differences in microbiota composition and diversity. The alpha diversity including the Shannon index and Chao1 index was remarkably reduced in DSS and OR mice in contrast with HC mice, and OR mice were significantly lower than DSS mice (p<0.05, Fig. 2F and G). The beta diversity including Non-Metric Multidimensional Scaling (NMDS) and Principal Coordinates Analysis (PCoA) reflected the intestinal microbiota distributions were cluster separated among groups (Fig. 2H and I). Histogram from the phylum level to the genus level revealed the significantly enriched bacterium in OR mice were mainly Dubosiella, Peptostreptococcaceae, Romboutsia, Enterococcaceae, and Escherichia coli. DSS mice enriched bacterium mainly include Lactobacillaceae, Allobaculum, and Muribaculaceae (Fig. 2J).

Fig. 2

F. nucleatum further aggravates the dysbiosis in DSS-induced colitis mice. A, The Venn diagram analysis of OTUs among HC, DSS, and OR groups. B, Rank-abundance curves reflect higher species richness was found in HC mice(red curve) than in DSS and OR mice. The species richness of DSS mice was more abundant than OR mice. C, Rarefaction curves among the three groups. D, E, Relative abundance of microbiota composition on phylum and species levels. F, G, Alpha diversity including Taxa richness (Chao1 index) and species diversity (Shannon index) reflect that the microbial richness and diversity of HC mice were higher than DSS and OR mice. H, Non-metric multidimensional scaling, NMDS analysis reflects the similarity in microbial composition of samples. I, Principal coordinates analysis (PCoA) reflected the differences in sample species diversity among groups. J, Histogram represents the enriched microbiota in DSS and OR groups from phylum level to species level. F-G, Data are presented as mean ± S.D. p values were determined by analysis of variance, *p<0.05, **p<0.01

LFMT alleviates inflammation and gut barrier damage in DSS-induced colitis mice

Since we demonstrated F. nucleatum further facilitates inflammation and gut structure damage in DSS mice, we next investigated the effect of upper FMT and lower FMT on inflammation and intestinal barrier dysfunction. The body weight of UFMT and LFMT dropped not as sharply as OR mice (Fig. 3A). Survival curves declined markedly in OR and UFMT mice, and LFMT mice survival rate remained flat curves (Fig. 3B). The histopathological changes by HE staining reflected severe mucosal damage and a large number of inflammatory cells infiltrated in OR mice. However, we observed remained integrity of intestinal epithelium and scattered inflammatory cell infiltration in LFMT mice. Although ameliorative manifestations were observed in the intestine of UFMT mice, the intestinal structure was still damaged and remained few gland structures. The histopathological scores supported the results (Fig. 3C and D). We also observed increased Occludin protein expression in the LFMT mice, which was significantly different from the OR and UFMT groups (p<0.05, Fig. 3C and E). The disease activity of LFMT mice continued to increase in the first week and then decreased significantly from the first week. At the end of the experiment, the disease activity of LFMT mice was significantly different from DSS, OR, and UFMT groups (p<0.01, Fig. 3F). The disease activity index of UFMT mice remained unchanged from the tenth day. The OR mice had the highest disease activity at the end of the experiment, which was significantly different from other groups (p<0.01, Fig. 3F). We observed a remarkable reduction of TNF-α and IL-6 levels in UFMT and LFMT mice (p<0.05, Fig. 3G and I). Compared with OR and UFMT, the level of IL1-β was significantly downregulated in LFMT mice (p<0.05, Fig. 3H).

Fig. 3

Fecal microbiota transplantation via the lower gastrointestinal tract alleviates inflammation and gut barrier damage in DSS-induced colitis mice. A, Body weight changes among OR, UFMT, and LFMT groups. B, Survival rate (%). C, Representative images of histopathological by HE and immunohistochemical staining for Occludin protein. D, E, Histopathological scores and relative expression of Occludin protein. F, Variations of disease activity index among DSS, OR, UFMT, and LFMT groups. G, H, I, Colonic levels of TNF-α, IL-1β, and IL-6 by ELISA. A, Data are presented as mean ± S.D. p values were determined by analysis of variance, **p<0.01 vs. LFMT, ##p<0.01 vs. OR. B, Data are performed by Pearson’s Chi-square test. **p<0.01 vs. OR, ##p<0.01 vs. LFMT. D, F, Data are performed by Kruskal-Wallis test. **p<0.01 vs. OR, ##p<0.01 vs. LFMT. E, G-I, Data are presented as mean ± S.D. p values were determined by analysis of variance, *p<0.05, **p<0.01

LFMT restores intestinal microbiota dysbiosis in DSS-induced colitis mice

To investigate the effect of FMT on intestinal microbiota dysbiosis in DSS-induced colitis mice, we performed upper and lower FMT in C57BL/6 mice and collected feces samples for 16 S-rRNA full-length sequencing. The Venn diagram of OTUs revealed 410 OTUs were identified in the OR, UFMT, and LFMT mice. Besides, OR, UFMT, and LFMT groups contain 3, 1, and 33 unique OTUs, respectively (Fig. 4A). The species rank curve revealed a smoother and wider curve was observed in the LFMT group, and the sharpest declines were found in the OR group (Fig. 4B). The rarefaction curve is shown in Fig. 4C. The species diversity revealed an increased Shannon index in the LFMT group, but no significant difference compared with OR and UFMT groups (p>0.05, Fig. 4D). Taxa richness (Chao1 index) was remarkably increased in UFMT and LFMT, and OR mice were significantly lower than DSS mice (p<0.01, Fig. 4E). We further analyzed the microbial compositions of the feces samples at phylum and genus levels. As was shown in Fig. 4F, an increased abundance of Bacteroidota and Verrucomicrobiota was found in UFMT and LFMT mice. We also observed a decreased abundance of Firmicutes and Proteobacteria in UFMT and LFMT mice. In addition, a histogram from the phylum level to the genus level revealed the significantly enriched bacterium in UFMT were Lactobacillales, Enterococcaceae, and Clostridium. LFMT mice enriched bacterium mainly include Bacteroidota, Lachnospiraceae, and Prevotellaceae (Fig. 4G). NMDS and PCoA analysis indicated the intestinal microbiota distributions of LFMT were cluster-separated from DSS and UFMT groups (Fig. 4H and I).

Fig. 4

Fecal microbiota transplantation from the lower gastrointestinal tract restores intestinal microbiota dysbiosis in DSS-induced colitis mice. A, The Venn diagram analysis of OTUs among OR, UFMT, and LFMT groups. B, Rank-abundance curves reflect higher species richness in LFMT mice (green curve) than in OR (red curve) and UFMT (blue curve) mice. C, Rarefaction curves among the three groups. D, E, Taxa richness (Chao1 index), and species diversity (Shannon index) reflect that the microbial richness and diversity of LFMT mice were higher than OR and UFMT mice. F, Relative abundance of microbiota composition on phylum and species levels. G, Histogram represents the enriched microbiota in UFMT and LFMT groups from phylum level to species level. H, NMDS analysis reflects the similarity in the microbial composition of samples. I, PCoA analysis reflected the differences in sample species diversity among groups. D-E, Data are presented as mean ± S.D. p values were determined by analysis of variance, *p<0.05, **p<0.01

F. nucleatum and virulence fadA levels are negatively correlated with LFMT therapeutic times

The correlation heat map illustrated the relationship between the inflammatory factors TNF-α, IL-1β, and IL-6 and the abundance of the top 50 genera among five groups and between UFMT and LFMT groups. Genera usually deficient probiotics in IBD such as Lactobacillus, Allobaculum, and Bacteroidales were found negative correlation with TNF-α, IL-1β, and IL-6. Genera like Romboutsia, Escherichia Shigella, Enterococcus, and Clostridium were found positively correlated with TNF-α, IL-1β, and IL-6 (Fig. 5A and B). We also analyzed nusG and fadA levels at different time points using real-time PCR. We observed that nusG and fadA were continuously increased in OR mice, which was significantly higher than UFMT and LFMT (p<0.01, Fig. 5C and D). Remarkable reduction was found in UFMT and LFMT, but the declined trend of UFMT is not as obvious as in LFMT mice (p<0.05, Fig. 5C and D). In addition, fecal nusG and fadA gene levels were negatively correlated with LFMT therapeutic times (r = 0.9531, p<0.001, Fig. 5E) (r = 0.9610, p<0.001, Fig. 5F).

Fig. 5

Lower gastrointestinal FMT can reduce the levels of F. nucleatum (shown as nusG gene) and virulence fadA, and the levels of F. nucleatum and virulence fadA are negatively correlated with FMT therapeutic times. A, The Correlation heatmap reveals the correlation between inflammatory cytokines TNF-α, IL-1β, IL-6, and the top 50 genera in abundance among the five groups. B, Correlation heatmap among OR, UFMT, and LFMT groups. C, The changes of nusG gene level with the increased times of FMT among groups. D, The changes of fadA gene level with the increased times of FMT among groups. E, Fecal nusG level was negatively correlated with FMT times (r = 0.9531, p<0.001). F, Fecal fadA level was negatively correlated with FMT times (r = 0.9610, p<0.001). C, D, Data are presented as mean ± SD, p values were determined by analysis of variance, *p<0.05, **p<0.01 vs. OR, ##p<0.01 vs. LFMT. E, F, and Data are performed by Spearman correlation