Abstract

   Approximately 50% of the patients with ulcerative colitis (UC) are
   primarily nonresponsive to anti‐tumor necrosis factor (TNF) therapy or
   lose their responsiveness over time. The gut microbiota plays an
   important role in the resistance of UC to anti‐TNF therapy; however,
   the underlying mechanism remains unknown. Here, it is found that the
   transplantation of gut fecal microbiota from patients with UC alters
   the diversity of the gut microbiota in dextran sulfate sodium‐induced
   colitis mice and may affect the therapeutic responsiveness of mice to
   infliximab. Furthermore, the abundances of Romboutsia and Fusobacterium
   increase in the tissues of patients with UC who do not respond to
   anti‐TNF therapy. Differentially abundant metabolites are mainly
   enriched in nicotinate and nicotinamide metabolism in NCM460 cells
   after Fusobacterium nucleatum infection. Mechanistically, F. nucleatum
   promotes the nicotinamide adenine dinucleotide (NAD^+) salvage pathway
   by upregulating NAMPT expression, which subsequently leads to the
   activation of the p38 mitogen‐activated protein kinase (MAPK) signaling
   pathway and promotes the secretion of inflammatory factors, ultimately
   inhibiting the therapeutic response to anti‐TNF drugs. These findings
   demonstrate that the gut microbiota can influence the response to
   anti‐TNF therapy in patients with UC and highlight the therapeutic
   potential of targeting F. nucleatum and its associated pathways for
   preventing and treating drug resistance in UC.

   Keywords: anti‐tumor necrosis factor therapy, fusobacterium nucleatum,
   microbiota, NAD^+ salvage pathway, ulcerative colitis
     __________________________________________________________________

   The gut microbiota affects the response to anti‐TNF therapy in UC
   patients. F. nucleatum activates the p38 MAPK signaling pathway by
   regulating the NAD^+ salvage pathway to inhibit anti‐TNF therapy
   responsiveness. Interfering with the NAD^+ salvage pathway or
   inhibiting the growth of F. nucleatum may alleviate the resistance of
   UC to anti‐TNF therapy.

   graphic file with name ADVS-12-2413128-g006.jpg

1. Introduction

   Ulcerative colitis (UC), a major form of inflammatory bowel disease
   (IBD), is a chronic inflammatory disorder of unknown etiology with a
   rapidly increasing incidence in developing countries.^[ [40]^1 ^]
   Anti‐tumor necrosis factor (TNF) therapy is the primary treatment for
   moderate‐to‐severe UC; however, approximately half of patients with UC
   are primary non‐responders to anti‐TNF therapy, and ≈40% of patients
   who initially respond to treatment may lose their responsiveness over
   time.^[ [41]^2 ^] Therefore, exploring the mechanisms of anti‐TNF
   therapy resistance in UC is essential for optimizing current
   therapeutic strategies.

   Recent studies have suggested that the gut microbiota plays an
   important role in the response to anti‐TNF therapy.^[ [42]^2 , [43]^3 ,
   [44]^4 , [45]^5 , [46]^6 , [47]^7 ^] Baseline microbial richness
   analyses indicate a preferential response to anticytokine therapy.^[
   [48]^5 ^] A high absolute abundance of Bifidobacteriales in the gut
   microbiota of pediatric patients with IBD is associated with the
   response to infliximab (IFX) treatment.^[ [49]^6 ^] Recently, a study
   using 16S rRNA sequencing of fecal samples from patients with IBD,
   including those from patients with UC and patients with Crohn's disease
   (CD), revealed that non‐responders had a lower abundance of short‐chain
   fatty acid producers, particularly those of the class Clostridia, and a
   higher abundance of pro‐inflammatory bacteria and fungi, such as those
   of the genus Candida, than responders.^[ [50]^2 ^] Another 16S rDNA
   sequencing study of CD tissues revealed that the abundance of
   Fusobacteria was higher in the feces of patients who did not respond to
   anti‐TNF therapy than in those who did respond,^[ [51]^7 ^] suggesting
   that Fusobacteria may be related to CD anti‐TNF responsiveness.
   Fusobacterium nucleatum (F. nucleatum) infection can exacerbate UC. The
   abundance of F. nucleatum in UC tissues was found to be higher than in
   normal tissues.^[ [52]^8 , [53]^9 , [54]^10 , [55]^11 ^] F. nucleatum
   promotes intestinal epithelial cell autophagy,^[ [56]^9 ^] regulates M1
   macrophage polarization,^[ [57]^11 ^] or activates IL‐17F signaling^[
   [58]^8 ^] to aggravate intestinal barrier damage and colonic
   inflammation in UC, suggesting F. nucleatum can induce inflammation and
   immune responses in UC. However, there are currently no specific
   biological markers to evaluate the responsiveness of UC to anti‐TNF
   therapy. Moreover, the relationship between the microbiome and the
   anti‐TNF response in UC and its possible molecular mechanisms have not
   yet been investigated.

   Nicotinamide adenine dinucleotide (NAD^+) is an important coenzyme in
   biological metabolism involved in oxidative phosphorylation, DNA
   repair, and inflammation.^[ [59]^12 ^] The expression of nicotinamide
   phosphoribosyl transferase (NAMPT), a rate‐limiting enzyme of the NAD^+
   salvage pathway, is elevated in the intestinal tissues of children with
   IBD and may serve as a marker of disease severity in pediatric patients
   with IBD.^[ [60]^13 ^] The NAMPT inhibitor FK866 potently blocks the
   NAD^+ salvage pathway, reduces intestinal mucosal NAD^+ levels, reduces
   the abundance and activity of NAD‐dependent enzymes, including PARP1,
   Sirt6, and CD38, reduces nuclear factor‐kappa B activation, and
   decreases cellular infiltration by inflammatory monocytes, macrophages,
   and activated T cells, thereby inhibiting the occurrence of colitis and
   colitis‐related tumors in mice.^[ [61]^14 ^] Various metabolic
   biomarkers involving lipid, bile acid, and amino acid pathways may
   contribute to the prediction of the response to anti‐TNF therapy in
   patients with IBD;^[ [62]^15 ^] however, the correlation between NAD^+
   metabolism and anti‐TNF therapy responsiveness in patients with UC
   remains unclear. One study showed that the serum levels of NAMPT are
   increased in patients with IBD who fail to respond to anti‐TNF therapy
   and decreased in patients who are responsive to these therapies.^[
   [63]^16 ^] A study on anti‐TNF therapy for UC showed that NAMPT could
   be used to predict drug non‐responsiveness.^[ [64]^17 ^] However,
   whether NAMPT affects the efficacy of UC anti‐TNF therapy by regulating
   the NAD^+ salvage pathway and the correlation between the gut
   microbiota and NAD^+ metabolism has not yet been elucidated.

   Mitogen‐activated protein kinase (MAPK) is a key regulatory pathway for
   cell proliferation, differentiation, transformation, and apoptosis. The
   MAPK signaling pathway is associated with the responsiveness to
   anti‐TNF therapy.^[ [65]^18 , [66]^19 ^] A previous study has shown
   that p38δ protein levels are significantly increased in the synovial
   tissue of rheumatoid arthritis patients who do not respond to
   infliximab; further studies on single nucleotide polymorphisms (SNPs)
   in the MAPK pathway revealed that both upstream and downstream SNPs of
   the p38MAPK cascade are associated with anti‐TNF therapy
   responsiveness.^[ [67]^18 ^] Alterations in p‐p38 and p‐JNK levels in
   the peripheral blood of IFX‐treated patients with bipolar depression
   are strongly associated with the remission of depressive symptoms.^[
   [68]^19 ^] A previous study has shown that NAMPT promotes the
   phosphorylation of p38MAPK.^[ [69]^20 ^] The exoprotein Gbp of F.
   nucleatum promotes THP‐1 cell lipid deposition by binding to CypA and
   activating the PI3K‐AKT/MAPK/NF‐kappaB pathways.^[ [70]^21 ^] However,
   whether the gut microbiota regulates anti‐TNF responsiveness via
   p38MAPK during anti‐TNF therapy in UC remains unclear.

   Here, we explored whether and how the gut microbiota affects the
   response to anti‐TNF therapy in patients with UC. We found that
   alterations in the gut microbiota were associated with the response to
   anti‐TNF therapy in patients with UC and that the abundance of F.
   nucleatum was significantly increased in patients with UC who did not
   respond to anti‐TNF therapy. We also demonstrated that F. nucleatum
   inhibited the responsiveness of UC to anti‐TNF therapy by increasing
   NAMPT expression, modulating the NAD^+ salvage pathway, and activating
   the p38 MAPK signaling pathway.

2. Results and Discussion

2.1. The Gut Microbiota Affects the Therapeutic Response of DSS‐Induced
Colitis to IFX

   To investigate the potential relationship between alterations in the
   gut microbiota and the therapeutic effect of anti‐TNF drugs in the
   treatment of UC, we performed fecal microbiota transplantation (FMT) in
   broad‐spectrum antibiotic (ABX) model mice using feces from patients
   with UC who responded to anti‐TNF therapy (RUC) and patients with UC
   who did not respond to anti‐TNF therapy (NRUC) (Figure [71] 1A).
   Compared with mice co‐treated with dextran sodium sulfate (DSS) and
   IFX, mice transplanted with fecal microbiota from NRUC patients
   (NR+DSS+IFX) presented rapid weight loss (p < 0.01; Figure [72]1B),
   higher disease activity index (DAI) (p < 0.05; Figure [73]1C), more
   severe disease and disruption of mucosal structures (Figure [74]1D),
   shorter colon lengths (p < 0.01; Figure [75]1E) and higher histological
   scores (HS) (p < 0.01; Figure [76]1F). These data suggest that gut
   microbiota transplantation from NRUC patients may affect the response
   to IFX treatment in mice with DSS‐induced colitis. Additionally, we
   detected a more significant weight loss (p < 0.01; Figure [77]1B),
   higher DAI (p < 0.01; Figure [78]1C), more severe disease and
   disruption of mucosal structures (Figure [79]1D), shorter colon length
   (p < 0.01; Figure [80]1E), and a significant increase in HS (p < 0.01;
   Figure [81]1F) in the NR+DSS+IFX group than in mice co‐treated with DSS
   and IFX after transplantation with fecal microbiota from RUC patients
   (R+DSS+IFX). These findings indicate that the transplantation of gut
   fecal microbiota from UC patients of different origins leads to
   different responses to IFX treatment in mice with DSS‐induced colitis.

Figure 1.

   Figure 1
   [82]Open in a new tab

   Gut microbiota affects the response of DSS‐induced colitis to IFX
   therapy. A) Schematic representation of experimental setup for mice.
   B,C) Statistical analysis of body weight change B) and DAI C) of mice
   (n = 6 per group; ns = no significance, *p < 0.05, **p < 0.01, and ***p
   < 0.001; the dividing lines indicate any joins; one‐way ANOVA combined
   with Bonferroni's post hoc test. Error bars indicate SD). D)
   Representative image of colons and H&E staining of colon sections from
   mice of each group (4 × magnification and 200 × magnification). E,F)
   Statistical analysis of colon length E) and histopathological score of
   the colon F) (n = 6 per group; ns = no significance, *p < 0.05, **p
   < 0.01, and ***p < 0.001; the dividing lines indicate any joins;
   nonparametric Mann‐Whitney U test. Error bars indicate SD).

   To further investigate whether transplanting fecal microbiota from RUC
   patients could alleviate the drug resistance caused by transplanting
   fecal microbiota from NRUC patients, we simultaneously transplanted
   fecal microbiota from RUC and NRUC patients (R+NR+DSS+IFX). Compared
   with the NR+DSS+IFX group, the weights of the mice in the R+NR+DSS+IFX
   group were not significantly lower (p > 0.05; Figure [83]1B), the DAI
   was not significantly lower (p > 0.05; Figure [84]1C), the severity of
   the disease and destruction of the mucosal structure were not
   significantly different (Figure [85]1D), and the differences in colon
   length (p > 0.05; Figure [86]1E) and HS were not significant (p > 0.05;
   Figure [87]1F). These results suggest that simultaneous transplantation
   of fecal microbiota from RUC and NRUC patients did not affect the
   responsiveness of mice with DSS‐induced colitis to IFX treatment.
   Collectively, our data indicate that the fecal microbiota of patients
   with UC may affect the therapeutic responsiveness of DSS‐induced
   colitis to IFX.

2.2. FMT Alters the Diversity of the Gut Microbiota in Mice with DSS‐Induced
Colitis

   To analyze the bacterial composition in the intestinal tissues of each
   group of mice, we compared the sequencing data generated using the
   Illumina NovaSeq 6000 from DSS+IFX, NR+DSS+IFX, R+DSS+IFX, and
   R+NR+DSS+IFX mouse tissues. The alpha diversity was then analyzed for
   each group. The Chao1 index of the NR+DSS+IFX group was significantly
   higher than that of the DSS+IFX group (p < 0.01; Figure [88]S1,
   Supporting Information). Moreover, the Simpson and Shannon indices were
   significantly higher in the NR+DSS+IFX group than in the R+DSS+IFX
   group (p < 0.01; Figure [89]S1E,F, Supporting Information). However,
   there was no significant difference in alpha diversity between the
   NR+DSS+IFX and NR+R+DSS+IFX groups (p > 0.05; Figure [90]S1G–I,
   Supporting Information).

   We further evaluated the β diversity of the gut microbiota in the
   intestinal tissues of each group. Compared with those in the DSS+IFX
   group, the abundances of Firmicutes, Fusobacteriota, Deferribacterota,
   Campilobacterota and Bacteroidota in the tissues of the NR+DSS+IFX
   group were significantly higher, whereas the abundances of Myxococcota,
   Proteobacteria, Gemmatimonadota, Acidobacteriota and Actinobacteriota
   were significantly lower (p < 0.05; Figure [91] 2A). Clostridia,
   Lachnospirales, Lactobacillus, and Fusobacteriaceae were abundant in
   the NR+DSS+IFX group, whereas Proteobacteria, Ralstonia,
   Bifidobacterium, and Streptococcus were abundant in the DSS+IFX group
   (p < 0.05; Figure [92]2B). Compared with those in the NR+DSS+IFX group,
   the abundances of Firmicutes, Desulfobacterota, Bacteroidota, and
   Verrucomicrobiota in the NR+R+DSS+IFX group were significantly higher,
   whereas the abundances of Actinobacteriota, Proteobacteria,
   Acidobacteriota, and Fusobacteriota were lower (p < 0.05;
   Figure [93]2C). Hydrogenophilus, Thalassospira, Anaerovorax, and
   Rikenella were abundant in the NR+R+DSS+IFX group, whereas
   Fusobacteriota, Actinobacteriota, Fusobacterium, and Moraxellaceae were
   abundant in the NR+DSS+IFX group (p < 0.05; Figure [94]2D). Compared
   with the R+DSS+IFX group, the abundances of Acidobacteriota,
   Proteobacteria, Actinobacteriota, and Fusobacteriota were significantly
   higher, and the abundances of Firmicutes and Bacteroidota were
   significantly lower in the NR+DSS+IFX group (p < 0.05; Figure [95]2E).
   At the genus level, the abundances of Nitrosomonadaceae, Lactobacillus,
   Fusobacterium, and Bamesiella significantly increased in the NR+DSS+IFX
   group, whereas the abundances of Blautia, Oscillibacter, and
   Papillibacter significantly decreased compared with those in the
   R+DSS+IFX group (p < 0.05; Figure [96]2F–K). These data indicate that
   FMT alters the gut microbiota diversity in mice with DSS‐induced
   colitis.

Figure 2.

   Figure 2
   [97]Open in a new tab

   Changes of microbial diversity in intestinal tissues of mice with
   DSS‐induced colitis. A, C, E) Hierarchically clustered heat map
   analysis of the highly represented bacterial taxa (phylum level) in
   tissues from ID (DSS+IFX) and NID (NRUC+DSS+IFX) mice A), NID and NRID
   (NRUC+RUC+DSS+IFX) mice C), NID and RID (RUC+DSS+IFX) mice E) by 16S
   rDNA sequencing. B, D) LEfSe analysis identifies the relative taxa
   abundance from ID (red) and NID (green) mice B), NID (red), and NRID
   (green) mice D). Only taxa with values greater than the LDA threshold
   of 3.6 are shown. F) A cladogram representation of data from NID (red)
   and RID (green) mice by 16S rDNA sequencing. G–K) The abundance of
   Escherichia G), Nitrosomonadaceae H), Rhodobacteraceae I),
   Lactobacillus J), Fusobacterium K) from NID and RID tissues by 16S rDNA
   sequencing.

2.3. The Gut Microbiota in the Tissues of Patients with UC is Associated with
IFX Response

   Considering the differences in the microbiota between the patients’
   feces and tissues, as well as the abundance of bacteria colonized from
   the patients’ stool in the gut of the mice, we further analyzed the
   bacterial composition in the intestinal tissues of NRUC (n = 8) and RUC
   patients (n = 8) via 16S rDNA sequencing. There was no statistically
   significant difference in the alpha diversity of the gut microbiota in
   the intestinal tissues of these patients (p > 0.05; Figure [98] 3A–C).
   Compared with those in RUC patients, the abundances of Fusobacteriota,
   Patescibacteria, and Spirochaetes were significantly higher in the
   tissues of NRUC patients, whereas the abundances of Chloroflexi,
   Deferribacteres, and Tenericutes were significantly lower (p < 0.05;
   Figure [99]3D). Further linear discriminant analysis revealed that the
   abundances of Romboutsia and Fusobacterium in NRUC patients were
   significantly higher than in RUC patients. In contrast, the abundances
   of Akkermansia, Ralstonia, Lachnoclostridium, and Steroidobacter
   decreased significantly (p < 0.05; Figure [100]3E,F)

Figure 3.

   Figure 3
   [101]Open in a new tab

   Changes of microbial diversity in intestinal tissues of UC patients.
   A–C) Alpha diversity boxplot. D) Hierarchically clustered heat map
   analysis of the highly represented bacterial taxa (phylum level) in
   tissues from RUC and NRUC patients by 16S rDNA sequencing. E) LEfSe
   analysis identifies the relative taxa abundance from RUC (green) and
   NRUC (red) patients. F) The abundance of Fusobacterium from RUC and
   NRUC tissues. G) Statistical analysis of the abundance of F. nucleatum
   in RUC (n = 10) and NRUC (n = 8) patients (*p < 0.05, and **p < 0.01,
   and ***p < 0.001; the dividing lines indicate any joins; nonparametric
   Mann‐Whitney U test. Error bars indicate SD). H) Representative images
   of FISH to assess the amount of F. nucleatum in RUC tissues (n = 40),
   NRUC tissues (n = 15), and control. EUB338 (red) is a Cy3‐conjugated
   “universal bacterial” oligonucleotide probe; FUS664 (green) is a
   FITC‐conjugated F. nucleatum oligonucleotide probe. 200 ×
   magnification. I) Stratification of F. nucleatum abundance in 89 UC
   patients and its significant association with clinicopathological
   factors.

   Given the increased abundance of Fusobacterium in the intestinal
   tissues of NR+DSS+IFX mice and NRUC patients (Figure [102]S2,
   Supporting Information) and the role of Fusobacterium, especially F.
   nucleatum, in UC progression, we further explored the abundance of F.
   nucleatum in UC tissues. Consistent with the sequencing data, RT‐PCR
   data revealed that F. nucleatum was more highly enriched in UC tissues
   than in normal tissues (n = 5; p < 0.01; Figure [103]3G). Furthermore,
   the abundance of F. nucleatum in NRUC tissues (n = 8) was higher than
   that in RUC tissues (n = 10) (p < 0.05; Figure [104]3G). We further
   detected F. nucleatum abundance in the tissues of 89 patients with UC,
   including 40 RUC patients and 15 NRUC patients, using fluorescence in
   situ hybridization (FISH) to investigate the role of F. nucleatum in
   anti‐TNF therapy in patients with UC. F. nucleatum was detected in a
   higher percentage of NRUC tissues (73.33%) than in RUC tissues (42.5%;
   p < 0.05; Figure [105]3H). Next, we evaluated the relationship between
   F. nucleatum abundance and the clinicopathological features of
   patients. The abundance of F. nucleatum was associated with the
   clinical course, IFX response, clinical activity, and hs‐CRP levels (p
   < 0.05; Figure [106]3I). These data suggest that F. nucleatum abundance
   is possibly associated with the therapeutic responsiveness of UC to
   IFX.

2.4. F. nucleatum Inhibits the Therapeutic Response of DSS‐induced Colitis
Mice to IFX

   To investigate whether F. nucleatum affects the therapeutic
   responsiveness to IFX in vivo, we administered F. nucleatum by gavage
   to ABX‐treated mice, examined the colonization of F. nucleatum
   (Figure [107]S3, Supporting Information), and treated them with DSS or
   IFX (Figure [108] 4A). We found that the DSS and IFX groups presented
   slower body weight loss (p < 0.01; Figure [109]4B), lower DAI (p
   < 0.01; Figure [110]4C), lower colon size ratio (p < 0.05;
   Figure [111]4E), and lower HS (p < 0.05; Figure [112]4F) compared with
   the DSS group. After F. nucleatum infection, weight loss, cecal edema,
   colon shortening, and colitis significantly increased in the DSS+IFX
   group (p < 0.01; Figure [113]4B–F). Additionally, western blotting
   revealed that the expression levels of ZO‐1 and Occludin in the
   intestinal tissues of the mice in the DSS+IFX group were significantly
   higher than those in the DSS group (p < 0.05; Figure [114]4G). However,
   the expression levels of ZO‐1 and Occludin in the intestinal tissues of
   the F. nucleatum+DSS+IFX group were significantly lower than those in
   the intestinal tissues of the DSS+IFX group (p < 0.01; Figure [115]4G).
   These data indicate that F. nucleatum may affect the therapeutic
   response to IFX in DSS‐induced colitis mice.

Figure 4.

   Figure 4
   [116]Open in a new tab

   F. nucleatum inhibits the therapeutic response of DSS‐induced colitis
   to IFX. A) Schematic representation of experimental setup for mice.
   B,C) Statistical analysis of body weight change B) and the disease
   activity index (DAI) C) of mice (n = 5 per group; *p < 0.05, **p
   < 0.01, and ***p < 0.001; the dividing lines indicate any joins;
   one‐way ANOVA combined with Bonferroni's post hoc test. Error bars
   indicate SD). D) Representative image of colons and H&E staining of
   colon sections from mice of each group (4 × magnification and 200 ×
   magnification). E,F) Statistical analysis of colon length E) and
   histopathological score of the colon F) (n = 5 per group; *p < 0.05,
   **p < 0.01, and ***p < 0.001; the dividing lines indicate any joins;
   nonparametric Mann‐Whitney U test. Error bars indicate SD). G) Western
   blotting was performed to detect ZO‐1 and Occludin expression in
   tissues of different groups of mice. H–M) ELISA detected the expression
   of inflammatory factors in the serum of different groups of mice (n = 5
   per group; *p < 0.05, **p < 0.01, and ***p < 0.001; the dividing lines
   indicate any joins; nonparametric Mann‐Whitney U test. Error bars
   indicate SD).

   Moreover, we detected the secretion of inflammatory cytokines in the
   serum of the mice in each group (Figure [117]4H–M). Compared with that
   in the DSS group, the secretion of TNF‐α in the serum of the DSS+IFX
   group was significantly reduced (p < 0.001; Figure [118]4M), whereas
   the secretion of IL‐10 was increased (p < 0.05; Figure [119]4K),
   indicating that IFX could inhibit the secretion of TNF‐α in DSS‐induced
   colitis. However, compared with those in the DSS+IFX group, the
   secretion of IL‐6, IL‐18, and TNF‐α in the serum of the F.
   nucleatum+DSS+IFX group was significantly higher (p < 0.001;
   Figure [120]4I,L,M), while the secretion of IL‐10 was lower (p < 0.05;
   Figure [121]4K). These results suggest that F. nucleatum increases the
   secretion of relevant inflammatory cytokines, thereby affecting the
   responsiveness of DSS‐induced colitis to IFX treatment.

2.5. F. nucleatum Promotes Resistance to Anti‐TNF Therapy in DSS‐Induced
Colitis through Activating the p38 MAPK Signaling Pathway

   Recent studies have suggested that immune cell infiltration plays a
   significant role in the development of UC.^[ [122]^22 ^] Macrophages,
   activated dendritic cells, and neutrophils play crucial roles in the
   occurrence and progression of UC.^[ [123]^23 ^] To verify the
   correlation between immune cell infiltration and the abundance of F.
   nucleatum in UC patient tissues, we used immunofluorescence and found
   that a high abundance of F. nucleatum was accompanied by a significant
   increase in CD68‐positive macrophage infiltration (p < 0.01) and a
   significant decrease in CD83‐positive dendritic cells (p < 0.05;
   Figure [124]S4A,B, Supporting Information) in UC tissues. These data
   suggest that F. nucleatum abundance is associated with macrophage
   infiltration in the tissues of patients with UC. RNA‐seq was then
   performed to analyze the gene expression profiles of THP‐1 cells
   co‐cultured with or without F. nucleatum. In total, 1939 upregulated
   genes and 1155 downregulated genes were detected (adjusted p < 0.05).
   Single‐sample gene set enrichment analysis (ssGSEA) revealed that
   differential gene enrichment was not significant in F.
   nucleatum‐infected THP‐1 cells (Figure [125]S4C, Supporting
   Information). These data indicate that F. nucleatum affects UC disease
   progression independent of immune cells.

   Furthermore, we investigated the effects of F. nucleatum infection on
   intestinal epithelial cells. Transcriptome sequencing revealed that F.
   nucleatum significantly upregulated the expression of 2596 genes and
   downregulated the expression of 1253 genes in NCM460 cells (adjusted p
   < 0.05). ssGSEA revealed that after F. nucleatum infection, genes
   associated with the MAPK, PI3K‐AKT, and TNF signaling pathways were
   significantly enriched in NCM460 cells (adjusted p < 0.01; Figure [126]
   5A). To test these results, we incubated NCM460 cells and fetal human
   colon (FHC) cells with F. nucleatum, E. coli (DH5α), or PBS (control,
   Con). Compared with E. coli and PBS treatment, F. nucleatum treatment
   substantially increased the phosphorylation level of p38 in a
   time‐dependent manner (p < 0.01), whereas there was no significant
   difference in the phosphorylation levels of JNK and ERK (p > 0.05;
   Figure [127]5B; E. coli or PBS as the control). These data suggest that
   F. nucleatum activates the MAPK signaling pathway in epithelial cells.
   Moreover, we detected the expression of phosphorylated p38 in F.
   nucleatum or IFX‐treated DSS‐induced colitis mice. Western blotting
   revealed a lower level of phosphorylated p38 in colitis tissues from
   DSS+IFX mice than in those from DSS mice. Compared with the DSS+IFX
   group, the F. nucleatum+DSS+IFX group showed significantly increased
   expression of phosphorylated p38 (p < 0.05; Figure [128]5C). These
   results suggest that F. nucleatum promotes resistance to IFX through
   activating the p38 MAPK signaling pathway.

Figure 5.

   Figure 5
   [129]Open in a new tab

   F.nucleatum affects the therapeutic response of DSS‐induced colitis to
   IFX through the P38 MAPK signaling pathway. A) ssGSEA was conducted to
   reveal the relationship between the abundance of F. nucleatum and the
   related pathway activity in NCM460 cells. B) Western blot analysis was
   performed to detect MAPK element expression in NCM460 and FHC cells
   co‐cultured with F. nucleatum, E. coli, or PBS (control). C) Western
   blot analysis was performed to detect p‐p38 expression in mice treated
   with F. nucleatum, DSS, or IFX. D) Schematic representation of
   experimental setup for mice. E,F) Statistical analysis of body weight
   change E) and DAI F) of mice (n = 5–6 per group; *p < 0.05, **p < 0.01,
   and ***p < 0.001; the dividing lines indicate any joins; one‐way ANOVA
   combined with Bonferroni's post hoc test. Error bars indicate SD). G)
   Representative image of colons and H&E staining of colon sections from
   mice of each group (4 × magnification and 200 × magnification). H,I)
   Statistical analysis of colon length H) and histopathological score of
   the colon I) (n = 5–6 per group; *p < 0.05, **p < 0.01, and ***p
   < 0.001; the dividing lines indicate any joins; nonparametric
   Mann‐Whitney U test. Error bars indicate SD).

   To verify whether F. nucleatum affects the therapeutic responsiveness
   to IFX via the p38 MAPK signaling pathway in vivo, ABX‐treated mice
   were initially administered F. nucleatum and then subjected to DSS,
   IFX, or the p38 signaling pathway activation‐specific inhibitor
   adezmapimod (SB 203580) (Figure [130]5D). We found that mice in the F.
   nucleatum+DSS+IFX+SB group presented slower body weight loss (p
   < 0.001; Figure [131]5E), lower DAI (p < 0.01; Figure [132]5F), and
   lower colon size ratio (p < 0.01; Figure [133]5H) than mice in the F.
   nucleatum+DSS+IFX group. Further observation of colonic
   histopathological sections and HE staining revealed that, compared with
   the F. nucleatum+DSS+IFX group, the F. nucleatum+DSS+IFX+SB group
   presented less epithelial damage (Figure [134]5G), including mucosal
   erosion, crypt loss, and lymphocyte infiltration, and lower HS (p
   < 0.01; Figure [135]5I). These results indicate that inhibiting of the
   p38 MAPK signaling pathway alleviates the resistance of DSS‐induced
   colitis mice to IFX treatment after F. nucleatum intervention.

2.6. F. nucleatum is Associated with Epithelial NAD^+ Metabolism

   Next, we performed metabolomic analysis to assess whether F. nucleatum
   infection affected the metabolites in intestinal epithelial cells. A
   total of 53 significantly upregulated metabolites and 34 downregulated
   metabolites were detected (fold change ≥ 2; Table [136]S1 and
   Figure [137]S5A, Supporting Information). Among the primary
   metabolites, the contents of most metabolites such as nucleotide and
   its metabolomics, glycerol phosphatide (GP), coenzyme and vitamins,
   glycerides (GL), and benzene and substituted derivatives were
   significantly increased. In contrast, the levels of most metabolites in
   tryptamines, cholines, pigments, amino acids, and their metabolomics,
   as well as fatty acids (FA) decreased (Figure [138] 6A). The Kyoto
   Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analysis
   revealed that compared with those in the PBS group, the differentially
   abundant metabolites in NCM460 cells after F. nucleatum infection were
   mainly enriched in the nicotinate and nicotinamide metabolism pathways
   (adjusted p < 0.05; Figure [139]6B; Figure [140]S5B, Supporting
   Information), suggesting that F. nucleatum is associated with
   nicotinate and nicotinamide metabolism in intestinal epithelial cells.
   We further analyzed the content of nicotinamide‐related metabolites.
   The results showed that nicotinamide and 6‐methylnicotinamide levels in
   NCM460 cells treated with F. nucleatum were higher than in the PBS
   group. In contrast, the level of nicotinic acid adenine dinucleotide
   was lower in the F. nucleatum‐treated group than in the PBS group
   (Figure [141]6C).

Figure 6.

   Figure 6
   [142]Open in a new tab

   F. nucleatum infection is associated with epithelial NAD^+ metabolism.
   A) Metabolomics analysis of cells co‐cultured with F. nucleatum or PBS
   (control). B) KEGG enrichment of differential metabolites in NCM460
   cells co‐cultured with F. nucleatum. C) NAD^+ related metabolites in
   NCM460 cells co‐cultured with F. nucleatum or PBS (control). D–F)
   Statistical analysis of NAD^+, NMN, NAM content in NCM460 cells
   co‐cultured with F. nucleatum or PBS (control) (*p < 0.05, **p < 0.01,
   and ***p < 0.001; unpaired Student's t‐test; Error bars indicate SD).
   G) NAD^+ related key enzymesanalysis in NCM460 cells co‐cultured with
   F. nucleatum or PBS (control) (*p < 0.05, **p < 0.01, and ***p < 0.001;
   unpaired Student's t‐test; Error bars indicate SD). H) Western blot
   analysis was performed with NCM460 cells co‐cultured with F. nucleatum,
   E. coli, or PBS (control). I) GSE analysis for NAMPT expression with
   anti‐TNF resistance in UC using the [143]GSE12251 and [144]GSE73661
   datasets. J) Statistical analysis of NAMPT mRNA expression in tissues
   from UC patients (10 RUC and 8 NRUC) and 5 healthy controls by RT‐PCR
   (*p < 0.05, **p < 0.01, and ***p < 0.001; the dividing lines indicate
   any joins; nonparametric Mann‐Whitney U test. Error bars indicate SD).
   K,L) Representative images of NAMPT expression are shown in K) (100 ×
   magnification and 200 × magnification) and are quantified in L) (*p
   < 0.05; unpaired Student's t‐test; Error bars indicate SD). M)
   Correlation analysis of NAMPT expression and the abundance of F.
   nucleatum in UC tissues (r = 0.6854, P = 0.0003; two‐tailed,
   nonparametric Spearman correlation). N,O) Representative images showing
   that the abundance of invasive F. nucleatum in UC tissues is associated
   with high expression of NAMPT N) (200 × magnification). The NAMPT
   protein levels in 22 F. nucleatum‐low and 33 F. nucleatum‐high UC
   tissues from patients were measured by immunohistochemical analysis (O;
   *p < 0.05; unpaired Student's t‐test; Error bars indicate SD).

   Nicotinamide is the main precursor of NAD^+. The content of nicotinic
   acid adenine dinucleotide, a key metabolite of the Preiss‐handler
   pathway and de novo synthesis pathway, was significantly reduced,
   suggesting that F. nucleatum may participate in the NAD^+ salvage
   pathway in epithelial cells. High‐performance liquid chromatography
   (HPLC) was subsequently conducted to assess the levels of niacinamide
   (NAM), NAD^+, and β‐nicotinamide mononucleotide (NMN) in intestinal
   epithelial cells after infection with F. nucleatum. Compared with the
   PBS group, F. nucleatum infection increased the levels of NAM, NMN, and
   NAD^+ in NCM460 cells (p < 0.01; Figure [145]6D–F). Furthermore, we
   verified the changes in the expression of key enzymes in the NAD^+
   synthesis pathway via RT‐PCR, which revealed that F. nucleatum
   infection upregulated the expression of the NAMPT, NRK1, NRK2, and NADS
   genes in NCM460 cells (p < 0.05; Figure [146]6G). Consistent with this
   finding, western blotting revealed that the F. nucleatum‐mediated
   increase in the level of NAMPT was time‐dependent (p < 0.05;
   Figure [147]6H). Taken together, these results suggest that F.
   nucleatum promotes the NAD^+ salvage pathway in intestinal epithelial
   cells.

   We performed gene set enrichment analysis (GSEA) on the [148]GSE12251
   and [149]GSE73661 datasets to evaluate whether IFX treatment
   responsiveness was associated with NAMPT expression in patients with
   UC. We found that NAMPT expression in tissues from NRUC patients was
   increased compared with tissues from RUC patients (p < 0.01;
   Figure [150]6I). Next, we examined the expression of NAMPT in 5 normal
   tissues, 8 NRUC tissues, and 10 RUC tissues from the same patients
   referenced in Figure [151]3G. Real‐time PCR results revealed that NAMPT
   expression in UC tissues was increased compared with normal tissues (p
   < 0.01; Figure [152]6J) and was higher in NRUC tissues than in RUC
   tissues (p < 0.001; Figure [153]6J). Immunohistochemistry was used to
   further validate the correlation between NAMPT expression and anti‐TNF
   therapeutic responsiveness in UC tissues. Consistent with the above
   experimental results, immunohistochemistry revealed that NAMPT
   expression was significantly increased in the tissues of NRUC patients
   (p < 0.05; Figure [154]6K,L). Next, to determine whether F. nucleatum
   affected NAMPT expression, we first conducted a correlation analysis.
   We found that the abundance of F. nucleatum (Figure [155]3G) was
   positively correlated with NAMPT mRNA expression (r = 0.6854, p
   < 0.0003, Figure [156]6M). We then used FISH and IHC to validate the
   correlation between the abundance of F. nucleatum and NAMPT expression
   in UC tissues. A high abundance of F. nucleatum was often accompanied
   by high levels of NAMPT (p < 0.05; Figure [157]6N,O). These data
   suggest that F. nucleatum affects anti‐TNF‐α therapeutic responsiveness
   through NAMPT, a key enzyme of the NAD^+ salvage pathway.

2.7. F. nucleatum Activates the p38 MAPK Signaling Pathway by Regulating the
NAD^+ Salvage Pathway to Inhibit Anti‐TNF Drug Therapy Responsiveness

   To determine whether the NAD^+ salvage pathway regulates F.
   nucleatum‐mediated p38 MAPK signaling pathway activation, we performed
   NAMPT‐targeting siRNA transfection or FK866 treatment in NCM460 cells
   co‐cultured with F. nucleatum. Western blotting revealed that F.
   nucleatum induced the upregulation of NAMPT and phosphorylated p38 and
   the downregulation of ZO‐1 and Occludin, which were blocked in NCM460
   cells, upon transfection with NAMPT‐targeting siRNA or pretreatment
   with FK866 (p < 0.05; Figure [158] 7A,B). These results demonstrate
   that blocking the NAD^+ salvage pathway reduces the increase in
   phosphorylated p38 levels and the destruction of the mucosal barrier
   induced by F. nucleatum.

Figure 7.

   Figure 7
   [159]Open in a new tab

   F.nucleatum affects the therapeutic response to IFX through NAD^+
   salvage synthesis pathway. A) Western blot analysis was performed with
   NCM460 cells transfected with siNAMPT and co‐cultured with F. nucleatum
   or PBS (control). B) Western blot analysis was performed with NCM460
   cells treated with FK866 and co‐cultured with F. nucleatum or PBS
   (control). C) Schematic representation of experimental setup for mice.
   D,E) Statistical analysis of body weight change D) and DAI E) of mice
   (n = 5–6 per group; *p < 0.05, **p < 0.01, and ***p < 0.001; the
   dividing lines indicate any joins; one‐way ANOVA combined with
   Bonferroni's post hoc test. Error bars indicate SD). F,G) Statistical
   analysis of colon length F) and histopathological score of the colon G)
   (n = 5–6 per group; *p < 0.05, **p < 0.01, and ***p < 0.001; the
   dividing lines indicate any joins; nonparametric Mann‐Whitney U test.
   Error bars indicate SD). H) Representative image of colons and H&E
   staining of colon sections from mice of each group (4 × magnification
   and 200 × magnification). I,J) Statistical analysis of the content of
   NAD and NMN in intestinal epithelial cells isolated from mice (n = 5–6
   per group; *p < 0.05, **p < 0.01, and ***p < 0.001; the dividing lines
   indicate any joins; nonparametric Mann‐Whitney U test; Error bars
   indicate SD). K) Western blotting was performed to measure the
   expression of MAPK associated proteins and mucosal barrier associated
   proteins in colon tissues from mice. L) Proposed mechanism of F.
   nucleatum affecting the therapeutic response of UC to IFX.

   To investigate whether F. nucleatum affects the therapeutic
   responsiveness of colitis to IFX by regulating the NAD^+ salvage
   pathway in vivo, we established a DSS‐induced colitis mouse model
   (Figure [160]7C). Compared with those in the F. nucleatum+DSS+IFX
   group, mice in the F. nucleatum+DSS+IFX+FK866 group exhibited slower
   body weight loss (p < 0.01; Figure [161]7D), lower DAI (p < 0.01;
   Figure [162]7E), and lower colon size ratio (p < 0.01; Figure [163]7F).
   Further observation of colonic histopathological sections and HE
   staining revealed that compared with the F. nucleatum+DSS+IFX group,
   the F. nucleatum+DSS+IFX+FK866 group presented a lower HS (p < 0.01;
   Figure [164]7G) and less epithelial damage (Figure [165]7H), including
   mucosal erosion, crypt loss, and lymphocyte infiltration. Western
   blotting revealed lower levels of phosphorylated p38 and phosphorylated
   ERK and higher levels of ZO‐1 and Occludin in colitis tissues from F.
   nucleatum+DSS+IFX+FK866 mice than in those from F. nucleatum+DSS+IFX
   mice (p < 0.05; Figure [166]7K). These data indicate that inhibiting
   the NAD^+ salvage pathway in vivo could alleviate the resistance of
   mice with DSS‐induced colitis to IFX treatment after F. nucleatum
   infection.

   To test the levels of metabolites associated with the NAD^+ salvage
   pathway, in situ, mouse intestinal epithelial cells were isolated and
   analyzed using HPLC. Compared with those in the PBS group, the levels
   of NAD^+ and precursor NMN in primary intestinal epithelial cells in
   the DSS group were significantly higher (p < 0.05; Figure [167]7I,J).
   The NAD^+ content in the intestinal primary epithelial cells of mice
   with DSS‐induced colitis decreased after IFX treatment (p < 0.01;
   Figure [168]7I). Moreover, NAD^+ and precursor NMN levels in the
   epithelial cells of the F. nucleatum+DSS+IFX+FK866 group were
   significantly lower than those in the F. nucleatum+DSS+IFX group (p
   < 0.05; Figure [169]7I,J). These results demonstrate that F. nucleatum
   regulates the levels of metabolites associated with the NAD^+ salvage
   pathway in vivo.

   Furthermore, we measured the levels of inflammatory factors in the
   serum of the model mice using ELISA. Compared with those in control
   mice, several inflammatory factors in the serum of DSS‐treated mice
   were upregulated, including IL‐1β, IL‐6, IL‐8, IL‐18, and TNF‐α, or
   downregulated, including IL10 (p < 0.001; Figure [170]S6A–F, Supporting
   Information). After treatment with IFX, the secretion of IL‐1β, IL‐6,
   and TNF‐α decreased, whereas the secretion of IL‐10 increased in the
   serum of DSS‐treated mice (p < 0.001; Figure [171]S6A,B,D,F, Supporting
   Information). In addition, compared with those in the F.
   nucleatum+DSS+IFX group, the levels of IL‐1β, IL‐6, IL‐8, IL‐18, and
   TNF‐α in the serum of the F. nucleatum+DSS+IFX+FK866 group were lower
   (p < 0.01; Figure [172]S6A–F, Supporting Information). RT‐PCR was
   performed to measure the levels of inflammatory factor mRNAs in the
   intestinal tissues of the mice. The results revealed that the
   expression of IL‐1β, IL‐6, and TNF‐α in the DSS group significantly
   increased, whereas the expression of IL‐10 was decreased compared with
   that in the control group (p < 0.05; Figure [173]S6G–J, Supporting
   Information). The expression of IL‐1β and TNF‐α was reduced in the
   intestinal tissues of DSS‐treated mice after IFX treatment (p < 0.05;
   Figure [174]S6G,J, Supporting Information). Moreover, the expression of
   IL‐6 and TNF‐α was increased in the F. nucleatum+DSS+IFX group compared
   with the DSS+IFX group. In contrast, TNF‐α expression was lower in the
   F. nucleatum+DSS+IFX+FK866 group than in the F. nucleatum+DSS+IFX group
   (p < 0.05; Figure [175]S6J, Supporting Information). These data suggest
   that F. nucleatum influences the responsiveness to anti‐TNF therapy
   through secreting inflammatory factors.

   In summary, we conclude that F. nucleatum promotes the NAD^+ salvage
   pathway by upregulating NAMPT expression, which subsequently leads to
   the activation of the p38 MAPK signaling pathway and then promotes the
   secretion of inflammatory factors, ultimately inhibiting the
   therapeutic response to anti‐TNF drugs.

3. Discussion

   Anti‐TNF therapies are recommended for patients with moderate‐to‐severe
   UC who do not respond to conventional therapies or cannot receive such
   therapies because of unacceptable side effects. However, 50% of
   patients with UC do not respond to anti‐TNF therapy or lose their
   responsiveness over time.^[ [176]^24 ^] At present, there are many
   biomarkers for evaluating the response of IBD to anti‐TNF therapy;^[
   [177]^25 , [178]^26 ^] however, most of these biomarkers lack robust
   validation.^[ [179]^27 ^] Recent studies suggest that the gut
   microbiota plays an important role in the response to anti‐TNF
   therapy,^[ [180]^3 , [181]^4 ^] highlighting the potential of
   microbiome monitoring as a companion diagnostic approach.^[ [182]^28 ^]
   Therefore, exploring the mechanisms underlying microbiological
   factor‐associated resistance in IBD is particularly important for the
   development of current therapeutic strategies.

   FMT overcomes anti‐PD‐1 therapy resistance in melanoma patients.^[
   [183]^29 ^] However, whether FMT can be used as an alternative
   conversion therapy for patients who are unresponsive to IFX remains
   unknown. One study has reported that immune‐mediated colitis patients
   who failed to respond to immunosuppression (infliximab or vedolizumab)
   received FMT from healthy donors as salvage therapy. Most patients
   achieved clinical remission at the end of the study period.^[ [184]^30
   ^] Moreover, UC patients who lose responsiveness to anti‐TNF therapy
   show improvements in symptoms and mucosal inflammation after FMT.^[
   [185]^31 ^] In another study of patients with prior failure of
   infliximab in CD who received FMT as a switch therapy, those who
   received FMT had significantly lower rates of clinical relapse compared
   with patients who received IFX treatment.^[ [186]^32 ^] These studies
   indicate that FMT affects the efficacy of anti‐TNF therapy in IBD
   patients. Consistent with previous studies, our study also suggests
   that the gut microbiota can affect the response to anti‐TNF therapy in
   colitis. Herein, we performed FMT in ABX model mice using feces from
   RUC patients and NRUC patients and found that, compared with the
   R+DSS+IFX group, the NR+DSS+IFX group exhibited more severe disease and
   disruption of mucosal structures. These findings indicate that
   transplantation of gut fecal microbiota from UC patients of different
   origins leads to different responses to IFX treatment in mice with
   DSS‐induced colitis. Compared with the DSS+IFX group, the NR+DSS+IFX
   group showed rapid weight loss, higher DAI, shorter colon length, and
   higher HS, suggesting that transplanting the gut microbiota of NRUC
   patients may reduce the efficacy of IFX therapy in DSS‐induced colitis
   mice. In contrast to previous studies, we used feces from UC patients
   to perform FMT in ABX model mice, which could better mimic the
   intestinal microbial environment of patients. In addition, we
   simultaneously transplanted fecal microbiota from RUC and NRUC
   patients. The results showed that fecal bacteria from RUC patients did
   not affect the responsiveness to IFX treatment in mice with DSS‐induced
   colitis transplanted with fecal bacteria from NRUC patients, possibly
   because human fecal bacteria are partially located in the gut of mice
   and because of the difference between the intestinal mucosal flora and
   fecal bacteria.

   Fecal and mucosal microbiota, as well as their endogenous metabolites,
   could provide a predictive tool for assessing the response of patients
   with IBD to various biologics.^[ [187]^28 , [188]^33 , [189]^34 ^] A
   previous study revealed no difference in the gut bacterial alpha
   diversity between the stools of CD patients who were responsive to IFX
   and those who were not responsive to IFX. The relative abundances of
   several bacterial genera, including Clostridium XI, Clostridium XVIII,
   Eggerthella, Lachnospiracea incertae sedis, Parabacteroides, and
   Peptococcus, in IFX non‐responders were significantly higher than those
   in IFX responders.^[ [190]^35 ^] Another study revealed lower dysbiosis
   indices and a higher abundance of Faecalibacterium prausnitzii in UC
   anti‐TNF therapy responders than nonresponders at baseline.^[ [191]^36
   ^] In this study, we performed 16S rDNA sequencing to analyze the
   differences in mucosal microbiota between RUC patients and NRUC
   patients and found no statistically significant difference in alpha
   diversity. In contrast to previous studies, our study revealed that the
   abundance of Romboutsia and Fusobacterium was significantly higher in
   the tissues of NRUC patients than in those of RUC patients, whereas the
   abundance of Akkermansia, Ralstonia, Lachnoclostridium, and
   Steroidobacter was significantly lower, possibly because previous
   studies analyzed CD samples or fecal samples from patients with UC and
   did not analyze the microbiota composition in UC intestinal mucosal
   tissue. In patients with CD, the abundance of Fusobacteria was higher
   in the mucosa of patients in the non‐responsive group than that in the
   responsive group,^[ [192]^7 ^] suggesting that Fusobacteria may be
   associated with anti‐TNF responsiveness in patients with CD. Studies
   have shown that F. nucleatum is highly abundant in the tissues and
   feces of patients with UC and plays an important role in the
   development of UC.^[ [193]^8 , [194]^9 , [195]^10 ^] However, the
   effect of F. nucleatum on the response to anti‐TNF treatment in UC
   patients has not been reported. In the present study, we used RT‐PCR
   and FISH to verify the abundance of F. nucleatum in UC tissues and
   reported a significant increase in the abundance of F. nucleatum in
   NRUC tissues compared with that in RUC tissues. Furthermore, we
   generated DSS‐induced colitis model mice and found that F. nucleatum
   exacerbated colitis symptoms, including weight loss, cecal edema, colon
   shortening, and colitis, in IFX‐treated mice. Additionally, the
   expression levels of ZO‐1 and Occludin in the intestinal tissue of the
   F. nucleatum+DSS+IFX group were significantly lower than those in the
   intestinal tissue of the DSS+IFX group. Therefore, our study revealed
   that the abundance of F. nucleatum in tissues might be correlated with
   the response of patients with UC to anti‐TNF therapy and that F.
   nucleatum could inhibit the responsiveness of DSS‐induced colitis mice
   to anti‐TNF therapy.

   NAD^+ is an important coenzyme in biometabolic processes and is
   synthesized via three distinct pathways: the NAD^+ salvage,
   Preiss‐Handler, and de novo pathways, with the salvage pathway being
   the predominant source of NAD^+.^[ [196]^14 ^] In salvage pathway,
   nicotinamide is converted to NMN through the enzymatic reaction of
   NAMPT, the rate‐limiting enzyme of the NAD^+ salvage pathway. In
   several disorders, including arthritis and psoriasis, the
   NAMPT‐mediated NAD^+ salvage pathway is thought to have inflammatory
   function,^[ [197]^37 , [198]^38 ^] indicating that the NAD^+ salvage
   pathway is associated with inflammation. NAMPT is also associated with
   intestinal inflammation in patients with IBD^[ [199]^14 ^] and is
   correlated with the response to anti‐TNF therapy in these patients.^[
   [200]^16 , [201]^17 ^] In this study, we verified the changes in the
   expression of key enzymes of the NAD^+ synthesis pathway via RT‐PCR,
   which revealed that F. nucleatum infection upregulated the expression
   of the NAMPT, NRK1, NRK2, and NADS genes. In contrast, the expression
   of NAPRT and QAPRT, which are key enzymes of the Preiss‐Handler and de
   novo pathways, respectively, did not increase. Moreover, we found that
   the expression of NAMPT was significantly higher in the tissues of NRUC
   patients than in those of RUC patients and that F. nucleatum abundance
   was positively associated with NAMPT expression in UC tissues. NAD^+
   regulates glycolysis, the tricarboxylic acid cycle, and oxidative
   phosphorylation‐driven energy metabolism through redox reactions.^[
   [202]^39 ^] Recent studies have shown that F. nucleatum can increase
   intracellular glycolysis in colon cancer cells to promote
   tumorigenesis.^[ [203]^40 ^] In addition, F. nucleatum is positively
   correlated with the levels of several plasma lipid metabolites^[
   [204]^41 ^] and plays a key role in lipid homeostasis in the apoptosis
   induced by NAMPT inhibitors,^[ [205]^42 ^] suggesting that F. nucleatum
   may be involved in glycolipid metabolism. In this study, we performed
   metabolomic analysis to detect the metabolites of epithelial cells
   infected with F. nucleatum. We found that the contents of most
   metabolites in the nucleotides and their metabolites, glycerol
   phosphatides, coenzymes and vitamins, glycerides, benzene, and
   substituted derivatives were significantly increased. Furthermore,
   differentially abundant metabolites were mainly enriched in nicotinate
   and nicotinamide metabolism in epithelial cells infected with F.
   nucleatum. Additionally, the levels of NAM, NMN, and NAD^+ increased,
   and the expression of the key enzyme NAMPT in the NAD^+ salvage pathway
   increased in intestinal epithelial cells after infection with F.
   nucleatum. Thus, our study suggests that F. nucleatum, which is
   associated with anti‐TNF resistance, promots the NAD^+ salvage pathway
   in intestinal epithelial cells by upregulating the key enzyme NAMPT.
   However, we did not elucidate the mechanism by which F. nucleatum
   regulates NAMPT expression. F. nucleatum inhibits butyric acid to
   promote tumor progression via the AMPK signaling pathway.^[ [206]^43 ^]
   Previous studies have demonstrated that intestinal AMPK is a possible
   upstream mediator of intestinal NAMPT‐mediated NAD^+ biosynthesis and
   may play a central role in maintaining intestinal homeostasis,
   including microbial composition.^[ [207]^44 ^] Therefore, F. nucleatum
   might affect NAMPT‐mediated NAD^+ biosynthesis by regulating AMPK.
   However, further studies are needed to precisely understand how F.
   nucleatum regulates the expression of NAMPT.

   Previous studies have shown that F. nucleatum can invade cells to
   activate the p38 MAPK signaling pathway and promote the secretion of
   inflammatory factors, such as IL‐8.^[ [208]^45 , [209]^46 ^] The
   activation of the p38 MAPK pathway is associated with the efficacy of
   anti‐TNF therapy.^[ [210]^18 , [211]^19 ^] Consistent with the findings
   of previous studies, in this study, we performed transcriptome
   sequencing and found that genes that were differentially expressed in
   intestinal epithelial cells infected with F. nucleatum were
   significantly enriched in the MAPK signaling pathway. NCM460 cells and
   FHC cells were incubated with F. nucleatum, E. coli or PBS. Compared
   with E. coli and PBS treatment, the phosphorylation level of p38 was
   increased in a time‐dependent manner after F. nucleatum infection,
   suggesting that F. nucleatum can activate the p38 MAPK signaling
   pathway in vitro. Furthermore, we treated ABX mice with DSS, IFX, SB,
   and F. nucleatum. We found that compared with F.
   nucleatum+DSS+IFX‐treated mice, F. nucleatum+DSS+IFX+SB‐treated mice
   exhibited improved epithelial damage and fewer symptoms of colitis,
   indicating that inhibition of the p38 MAPK signaling pathway alleviated
   resistance to IFX treatment in F. nucleatum‐infected DSS‐induced
   colitis mice. Studies have shown that NAMPT promotes p38 MAPK
   phosphorylation in endothelial cells and that p38 MAPK activation can
   be suppressed by knocking down NAMPT in senescent cells.^[ [212]^20 ,
   [213]^47 ^] In this study, we found that the increase in p38
   phosphorylation and the disruption of the mucosal barrier proteins ZO‐1
   and Occludin induced by F. nucleatum were blocked by the knockdown or
   inhibition of NAMPT expression in intestinal epithelial cells. After
   the NAD^+ salvage pathway was blocked by FK866 in vivo, mice in the F.
   nucleatum+DSS+IFX group exhibited slower body weight loss, lower DAI,
   lower colon size ratio, and lower HS than those in the F.
   nucleatum+DSS+IFX group. Moreover, the levels of phosphorylated p38 and
   phosphorylated ERK decreased, and the expression of ZO‐1 and Occludin
   increased in colitis tissues in the F. nucleatum+DSS+IFX+FK866 group.
   These results suggest that blocking the NAD^+ salvage pathway could
   inhibit the activation of the p38 MAPK signaling pathway induced by F.
   nucleatum, thus alleviating the resistance of DSS‐induced colitis mice
   to IFX treatment. However, further studies are needed to determine
   whether NAD^+ metabolism directly regulates the p38 MAPK signaling
   pathway.

   Finally, we explored the effect of F. nucleatum on inflammatory factor
   expression and the correlation between inflammatory factor expression
   and the response to IFX treatment in UC. Some studies have shown that
   the expression of inflammatory factors such as IL‐6, IL‐10, IL‐1B,
   CXCL8, and CCL2 is correlated with the response to anti‐TNF therapy in
   IBD.^[ [214]^48 , [215]^49 , [216]^50 ^] Consistent with previous
   studies, our research revealed that F. nucleatum promoted the secretion
   of IL‐6, IL‐18, and TNF‐α and inhibited the secretion of IL‐10 in the
   serum of DSS‐induced colitis mice treated with IFX while increasing the
   expression of IL‐6 and TNF‐α in the colon tissues. Blocking the NAD^+
   salvage pathway inhibited the secretion of IL‐1β, IL‐6, IL‐8, IL‐18,
   and TNF‐α in the serum and reduced the expression of TNF‐α in the
   tissues of F. nucleatum‐infected IFX‐treated DSS‐induced colitis mice.
   Further exploration of how these inflammatory factors regulate the
   response to anti‐TNF therapy and how F. nucleatum affects the secretion
   of these inflammatory factors during drug resistance will be highly
   important.

   The association between F. nucleatum and the development of UC has been
   extensively confirmed, and the potential mechanisms involved have been
   elucidated. Targeted therapy for F. nucleatum, for example by reducing
   its growth or eliminating it from the gut microbiota without harming
   beneficial bacteria, is a promising research approach. Probiotics show
   great potential in the prevention and treatment of many diseases.^[
   [217]^51 ^] Numerous studies have demonstrated that probiotics, such as
   Bifidobacterium, Streptococcus thermophilus,^[ [218]^52 ^]
   Lactobacillus salivarius,^[ [219]^53 ^] and Lactobacillus reuter ^[
   [220]^54 ^] inhibit F. nucleatum in vitro and in vivo. Short‐chain
   fatty acid butyrate can reduce the colonization and invasion of F.
   nucleatum in colorectal cancer tissues and relieve F. nucleatum‐induced
   chemoresistance.^[ [221]^55 ^] In our study, we found that F. nucleatum
   was more abundant in NRUC tissues than in RUC tissues, and the
   abundance of F. nucleatum was associated with IFX responsiveness. After
   F. nucleatum infection, weight loss, cecal edema, colon shortening, and
   colitis significantly increased in the DSS+IFX group. However, whether
   these probiotics can alleviate the resistance of UC patients to IFX
   treatment by inhibiting the growth and colonization of F. nucleatum
   should be verified.

   In this study, we revealed that FK866 inhibited the NAD^+ salvage
   pathway and alleviated the resistance of mice with DSS‐induced colitis
   to IFX treatment after F. nucleatum infection. A previous study has
   shown that the NAMPT inhibitor FK866 reduces mucosal NAD^+ levels and
   the activation of NF‐κB, then ameliorates DSS‐induced colitis.^[
   [222]^14 ^] Moreover, clinical trials have been performed in advanced
   solid tumor malignancies to determine the toxicity profile and
   pharmacokinetics of FK866.^[ [223]^56 ^] However, it is unclear whether
   FK866 can alleviate the resistance of UC patients to IFX. Therefore,
   FK866‐mediated NAMPT blockade is a promising approach for treating
   resistance to IFX in patients with UC, further validation is needed.

   In conclusion, our findings demonstrate that changes in the gut
   microbiota are associated with response to anti‐TNF therapy in patients
   with UC. These findings provide critical insights into the molecular
   mechanisms underlying the regulation of the NAD^+ salvage pathway by F.
   nucleatum in the anti‐TNF responsiveness of UC. Interfering with the
   NAD^+ salvage pathway or inhibiting the growth of F. nucleatum could
   alleviate the resistance of UC to anti‐TNF therapy, providing a new
   direction and target for the treatment of drug‐resistant UC.

4. Experimental Section

Clinical Samples

   All the included patients had a clinically and histologically confirmed
   diagnosis of UC. Fresh UC tissue samples (10 samples from RUC and 8
   samples from NRUC) and normal tissue samples (5 cases) were collected
   from patients who underwent colonoscopy at the Gastrointestinal
   Endoscopy Center of the Second Affiliated Hospital of Chongqing Medical
   University. The fresh samples were rinsed three times with sterile
   physiological saline at 4 °C after collection and were stored at −80 °C
   for sequencing and RT‐PCR. Paraffin‐embedded UC intestinal tissues (89
   samples) were collected from the pathology department of the same
   hospital, and the clinicopathological data of the patients were
   collected. RUC patients were defined as those with a Mayo score
   reduction of ≥ 3 points after 14 weeks of initial treatment with
   anti‐TNF (infliximab). NRUC patients were defined as those with a Mayo
   score reduction of ≤3 points after 14 weeks of initial infliximab
   treatment. Patients with clinical conditions requiring emergency
   management or involving malignancy, pregnancy, infectious diarrhea, or
   primary sclerosing cholangitis were excluded. Patients with a history
   of antibiotic drug use or fecal microbiota transplantation in the
   previous three months were excluded. Patients treated with other
   IBD‐related medications and those younger than 18 years of age were
   excluded from the NRUC and RUC groups. All participants have obtained
   informed consent, and the project was approved by the Medical Ethical
   Committee of the Second Affiliated Hospital of Chongqing Medical
   University (No. 806).

Bacterial Strains, Cell Lines and Treatments

   The F. nucleatum strain used in this study was isolated as described
   previously.^[ [224]^8 ^] The strain was stored at −80 °C in fetal
   bovine serum (FBS, Gibco, USA) and glycerol (50% v/v). F. nucleatum was
   incubated in fastidious anaerobe broth under anaerobic conditions at
   37 °C in a shaker at 200–220 rpm min^−1 for 3 to 4 d. The Escherichia
   coli strain DH5α (Tiangen, China) was incubated in Luria‐Bertani medium
   in a shaker at 200–220 rpm min^−1 for 12 h at 37 °C. THP‐1 cells were
   cultured in RPMI 1640 medium supplemented with 1%
   penicillin‐streptomycin (Beyotime, China) and 10% FBS. The normal human
   epithelial cell line NCM460 and FHC cell line (ATCC, USA) were cultured
   in high‐glucose DMEM (Gibco, USA) supplemented with 10% FBS in 5% CO[2]
   at 37 °C.

   The cell lines were infected with F. nucleatum or E. coli at a
   multiplicity of infection of 1:100. 2% DSS (MP Biomedicals, USA) or
   100 µg mL^−1 IFX was added to the culture medium, and cells were
   incubated for 24 h. The FK866 inhibitor (Sigma, USA) was dissolved in
   dimethyl sulfoxide (DMSO; Thermo Fisher Scientific, USA), and the cells
   were treated with the FK866 inhibitor at the recommended concentration
   of 100 nm for 1 h before F. nucleatum infection.

Animal Models and Experimental Design

   All animal experiments were approved by the Animal Care and Use
   Committee of the Second Affiliated Hospital of Chongqing Medical
   University (approval number: #806). Three‐ to four‐week‐old male
   C57BL/6J mice were purchased from the Vital River Laboratory Animal
   Technology Company (Beijing, China), housed in a specific pathogen‐free
   environment with a natural light‐dark cycle, and provided with
   autoclaved food and water. All mice were acclimatized for 7 days and
   then randomized into the different treatment groups.

   We performed FMT in DSS‐induced colitis mice to investigate the effect
   of the gut microbiota on the responsiveness to anti‐TNF therapy. The
   administration scheme is illustrated in Figure [225]1A. NRUC or RUC
   patients included in this study were randomly selected as fecal
   bacterial donors. Freshly collected donor fecal samples were
   immediately mixed in a biosafety cabinet under aerobic conditions,
   aliquoted into 2 mL freezing tubes, and stored at −80 °C. Freshly
   thawed human feces were resuspended in sterile phosphate‐buffered
   saline (PBS; 100 mg mL^−1), filtered through a sterile 100 µm filter,
   vigorously shaken for 5 min, and centrifuged at 500 × g for 5 min,
   after which a fecal suspension was obtained.

   The bacterial abundance was identified prior to gavage. Briefly, 200 µL
   of the fecal suspension at equal concentrations (1–3 × 10^8 cells
   mL^−1) were administered to mice via oral gavage. All mice (including
   those that did not receive FMT) were administered antibiotics,
   including 500 mg L^−1 vancomycin, 200 mg kg^−1 neomycin sulfate, 200 mg
   kg^−1 metronidazole, and 200 mg kg^−1 ampicillin, via gavage once a day
   for one week to establish an ABX mouse model. The mice were allowed to
   rest for 3 days after antibiotic treatment. FMT was performed via
   gavage with the fecal suspension (200 µL, 1–3 × 10^8 cells mL^−1) once
   a day for 2 weeks in the FMT groups. Moreover, the mice in the control
   and DSS groups were gavaged with 200 µL of a 0.9% saline solution.
   Three days after FMT completion, the mice received either normal
   drinking water or drinking water containing 3% DSS solution for 7 days.
   On day 3 after 3% DSS treatment, the mice were injected
   intraperitoneally with IFX (5 mg kg^−1 twice a week) according to the
   group assignment.

   To investigate the effect of F. nucleatum on the responsiveness to
   anti‐TNF therapy, we constructed a DSS‐induced colitis mouse model with
   F. nucleatum colonization. Briefly, on the third day after the ABX
   mouse model was established, mice were administered PBS containing F.
   nucleatum (10^9 CFU mL^−1) or PBS alone via daily gavage for 2 weeks.
   Three days after F. nucleatum infection, the mice received either
   normal drinking water or drinking water containing 3% DSS solution for
   7 days. Depending on the group, the mice were administered FK866 (10 mg
   kg^−1) intraperitoneally twice a day or the p38 MAPK‐specific inhibitor
   SB203580 (SB, MedChemExpress, USA) (5 mg kg^−1) once a day, 60 min
   after DSS treatment. On day 3 after 3% DSS treatment, the mice were
   injected intraperitoneally with IFX (5 mg kg^−1 twice a week) according
   to the group assignment.

   Body weight, fecal characteristics, rectal bleeding, fecal occult
   blood, and the DAI were recorded daily during the modeling period. On
   day 10, the mice were sacrificed after 3% DSS treatment. Serum was
   collected via centrifugation (1000 × g, 15 min, 4 °C), and the colon
   length was measured. The distal colon was collected and fixed in 4%
   paraformaldehyde for FISH and HE staining. The remaining colonic
   samples were quickly frozen in liquid nitrogen and stored at −80 °C for
   subsequent western blotting and PCR analyses.

DNA Extraction and High‐throughput Sequencing

   A TIANamp Bacteria DNA Kit (Tiangen, China) was used according to the
   manufacturer's protocol to extract genomic DNA from frozen human
   clinical samples. The extracted DNA was sent to Shanghai OE Biotech
   Co., Ltd. (Shanghai, China) for sequencing. Agarose gel electrophoresis
   and a NanoDrop 2000 spectrophotometer (Thermo Fisher Scientific, USA)
   were used to measure DNA integrity and concentration, respectively. PCR
   amplification of the V3‐V4 variable regions of the bacterial 16S rRNA
   gene was carried out in 25 µL reactions using the universal primers
   343F (5′‐TACGGRAGGCAGCAG‐3′) and 798R (5′‐AGGGTATCTAATCCT‐3′).
   Agencourt AMPure XP beads (Beckman Coulter Co., USA) were used to
   purify the PCR products, which were quantified using a Qubit dsDNA
   assay kit (Life Technologies, USA). After adjusting the PCR product
   concentration, sequencing was performed using an Illumina NovaSeq 6000
   (Illumina Inc., San Diego, CA; OE Biotech Company; Shanghai, China). To
   detect and eliminate ambiguous bases (N), paired‐end reads were
   preprocessed using Trimmomatic software. Low‐quality sequences were
   removed using a sliding window trimming approach. FLASH software was
   used to assemble the paired‐end reads according to the following
   parameters: minimum overlap of 10 bp, maximum overlap of 200 bp, and
   maximum mismatch rate of 20%. QIIME software (version 1.8.0) was used
   to retain reads with 75% bases above Q20 after eliminating unsuitable
   reads, such as sequences smaller than 200 bp and homologous or
   ambiguous sequences. Clean reads with sequence similarities ≥97% were
   clustered into operational taxonomic units (OTUs) using VSEARCH
   software. The QIIME software package was used to select representative
   sequences for each OTU and all representative sequences were annotated
   against the Silva database (Version 132).

Metabolomics Analysis

   The analysis was performed with NCM460 cells co‐cultured with F.
   nucleatum or PBS (control). Ultra‐performance liquid chromatography and
   tandem mass spectrometry were used to analyze the sample extracts.
   Using a triple quadrupole linear ion trap mass spectrometer, each ion
   pair was scanned and detected based on the optimized declustering
   potential and collision energy. Mass spectrometry data were processed
   using Analyst 1.6.3 software. The metabolites in the samples were
   qualitatively and quantitatively analyzed using mass spectrometry data
   in local metabolic databases. Based on the retention time and peak type
   of each metabolite, the chromatographic peak of each metabolite
   detected in the different samples was corrected to ensure qualitative
   and quantitative accuracy. Principal component analysis was performed
   on the samples to determine the overall metabolic differences between
   the groups and the variability between the samples within the groups.
   Metabolite content data were normalized using unit variance scaling,
   and the accumulation patterns of metabolites among different samples
   were subjected to hierarchical cluster analysis using R software.
   Differential metabolites between groups were determined by a VIP ≥1 and
   fold change ≥2 or ≤ 0.5. The KEGG compound database was used to
   annotate the differential metabolites, and the annotated metabolites
   were mapped to the KEGG pathway database.

Statistical Analyses

   The paired or unpaired Student's t‐test and the Mann–Whitney U test
   were performed to compare the quantitative data between different
   groups. The relationship between F. nucleatum abundance and mRNA
   expression was analyzed by linear regression. The association between
   patient characteristics and F. nucleatum abundance was determined using
   Pearson's chi‐square test or Fisher's exact test. The differences
   between multiple groups were evaluated using one‐way analysis of
   variance. Categorical data were expressed as percentages. All p values
   were two‐tailed, and differences with a p value of <0.05 were
   considered significant (*p < 0.05, ** p < 0.01, and *** p < 0.001).
   Data were analyzed using SPSS Statistics software (version 26.0; IBM
   Inc., Chicago, Illinois, USA) and GraphPad Prism 8 software (GraphPad
   Software, Inc., San Diego, CA, USA).

Conflict of Interest

   The authors declare no conflict of interest.

Supporting information

   Supporting Information
   [226]ADVS-12-2413128-s002.docx^ (1.5MB, docx)

   Supporting Table 1
   [227]ADVS-12-2413128-s001.xlsx^ (76.9KB, xlsx)

Acknowledgements