Abstract

Background

   Inflammatory bowel disease (IBD) is a gastrointestinal inflammatory
   disorder characterized by disturbed interactions between gut microbiota
   and host immune response. Barley leaf (BL) is a traditional Chinese
   herb recorded to have health-promoting effects. However, little is
   known about the beneficial role of BL against enteric infection-induced
   intestinal inflammation. Here, we uncover that BL protects against
   Citrobacter rodentium (C. rodentium)-induced infectious colitis by
   improving host-microbiota interactions.

Methods

   C3H/HeN mice were fed a diet with/without BL and infected with C.
   rodentium. Transcriptome sequencing, anti-CD4 antibody treatment, and
   flow cytometry were conducted to investigate the mechanisms of T cell
   immune modulation. The intervention involved administering anti-CD4
   antibody at 500 µg each time for three times before and during C.
   rodentium infection. Analysis of gut microbiota composition was
   performed by 16S rRNA gene sequencing on fecal samples. Fecal
   microbiota transplantation was conducted by administering microbiota
   from donor group to recipient group via oral gavage to investigate the
   role of intestinal microbiota in immune modulation.

Results

   BL ameliorated the severity of C. rodentium-induced colitis, and this
   effect was linked to improved gut homeostasis and enhanced mucosal
   barrier function. BL enriched the pathways of T helper 1 (Th1)/Th2 and
   Th17 cell differentiation in the colon, suggesting the involvement of
   CD4^+ T cells. Consistent with this, anti-CD4 antibody treatment
   abrogated the effect of BL and flow cytometry analysis revealed that BL
   mitigated C. rodentium-induced pro-inflammatory Th1 immune response.
   Moreover, the protective effect of BL was associated with alleviation
   of gut microbiota dysbiosis and increased abundance of Lactobacillus.
   Our in vivo studies further revealed that live Lactobacillus plantarum
   (L. plantarum) administration attenuated the pathogenic effects induced
   by C. rodentium infection, whereas heat-inactivated L. plantarum did
   not show the same results. Mechanistically, BL supplementation enriched
   L. plantarum, which subsequently released nanosized extracellular
   vesicles (EVs) that serve as a key mediator in alleviating C.
   rodentium-associated pathology and Th1 cell dysregulation.

Conclusions

   Our work thus provides evidence for utilizing BL and L.
   plantarum-derived EVs to manage enteric infection-associated IBD.

Supplementary Information

   The online version contains supplementary material available at
   10.1186/s12951-025-03504-w.

   Keywords: Barley leaf, Citrobacter rodentium, Colitis, Th1 immune
   response, Extracellular vesicles

Introduction

   Inflammatory bowel disease (IBD) is a non-specific chronic recurrent
   inflammatory disease characterized by abdominal pain and diarrhea,
   bloody stool, and weight loss [[38]1, [39]2]. Epidemiological evidence
   has shown that the prevalence and incidence of IBD are raising every
   year, especially in newly industrialized countries; and thus becomes a
   public threat and burden to human health [[40]3–[41]5]. The currently
   drug-based therapies in treating IBD showed limited effectiveness and
   potential side effects [[42]6]. Therefore, there is an urgent need to
   develop novel and effective strategies for the management of IBD.

   Though the etiology of IBD is still unclear, dysbiosis of gut
   microbiota have been implicated in the onset and progression of IBD
   [[43]7–[44]9]. The composition and function of gut microbiota could be
   shaped by diet, which is an important environmental factor associated
   with the epidemiology and pathogenesis of IBD [[45]10, [46]11]. A
   recent study has shown that diet-based therapy is effective in
   improving gut microbiota dysbiosis and alleviating clinical outcomes in
   patients with ulcerative colitis [[47]12]. Thus, dietary intervention
   may be a promising strategy that can be used in the prevention and
   treatment of IBD [[48]13].

   A wide range of pathogenic microbes are capable of entering the
   gastrointestinal tract, causing enteric infection and intestinal
   inflammation. Applying bacterial infection model is important to
   understand the fundamental role of dietary factors in host-microbe
   interactions [[49]14–[50]20]. Citrobacter rodentium (C.
   rodentium)-induced bacterial colitis has long been utilized as a robust
   model to study the role of commensal bacteria in pathogen-host
   interactions. C. rodentium infection triggers microbiota dysbiosis that
   is characterized by a reduced bacteria diversity and an over-expansion
   of C. rodentium by day 7 following infection [[51]21]. C. rodentium
   colonization can be resisted by the commensal gut microbiota and the
   colonization resistance of the gut microbiota to C. rodentium is
   reliant on nutrient availability [[52]22, [53]23]. For instance,
   deprivation of dietary fiber has been reported to promote
   susceptibility to C. rodentium via diet-driven changes in the gut
   microbiota [[54]24]. Thus, dietary nutrients may serve as crucial
   factors to protect against C. rodentium-induced enteric infection.

   Barley leaf (BL), the young leaf of Hordeum vulgare L., is a
   traditional Chinese herb with historically documented potential health
   benefits. Dietary fiber, protein, polyphenols, vitamins, and minerals
   are the major ingredients of BL [[55]25]. The antioxidative,
   antidepressant, and lipid-lowering properties have been reported
   [[56]26, [57]27]. Our previous studies have demonstrated that dietary
   BL modulates gut microbiota, and protects against chemically-induced
   colitis and colon tumorigenesis [[58]28–[59]30]. However, the
   beneficial effect of BL on intestinal microbial communities in the
   context of bacterial colitis is unclear. Whether and how gut microbiota
   modulates the potential effects of BL on host susceptibility to
   enteropathogenic infection and intestinal inflammation remains elusive.

   In this study, we applied murine pathogen C. rodentium to mimic a
   similar pathogenesis to human enterohemorrhagic and enteropathogenic E.
   coli. The beneficial effect of BL against C. rodentium infection and
   intestinal inflammation was investigated. Our results provided evidence
   to show that BL alleviated C. rodentium-induced colonic inflammation,
   barrier damage, and gut microbiota dysbiosis. Additionally, BL-induced
   protection was found to be dependent on CD4^+ T cells. We further
   revealed a crucial role of BL-induced gut microbiome in the modulation
   of protection against C. rodentium infection. Colonization of mice with
   live but not heat-inactivated Lactobacillus plantarum (L. plantarum), a
   commensal bacterium greatly enriched by BL, was sufficient to
   recapitulate the protection against C. rodentium infection.
   Mechanistically, L. plantarum-released extracellular vesicles (EVs)
   alleviated C. rodentium-induced bacterial colitis and T cell
   dysregulation in mice. This study provided novel insights into the
   potential immunoregulatory properties of BL and probiotic-derived EVs
   for developing effective therapeutic treatment in enteric infection and
   inflammation.

Results

BL attenuates intestinal pathology in C. rodentium-infected mice

   To evaluate whether BL conferred protective effects against C.
   rodentium-induced colonic inflammation, mice were fed with a standard
   chow diet or an isocaloric diet supplemented with BL for 3 weeks and
   inoculated or not with C. rodentium (Fig. [60]1A). Food consumption and
   water intake were monitored daily prior and during the infection. Our
   results showed that there was no significant difference in food intake
   or water intake between BL- and chow diet (CD)-fed mice, suggesting
   that BL supplementation had no adverse effect on the mouse eating and
   drinking habits (Figure [61]S1). Compared to control uninfected mice,
   mice infected with C. rodentium exhibited significant body weight loss
   and increased disease activity index (DAI) score (Fig. [62]1B and C).
   The colon length was shorter and the spleen weight was greater in C.
   rodentium-infected mice compared to control uninfected mice
   (Fig. [63]1D and E; Figure [64]S2). Nevertheless, the above deleterious
   effects caused by C. rodentium were attenuated in BL-fed mice
   (Fig. [65]1B-E). Histological analysis of colonic sections showed that
   C. rodentium infection led to pathologic changes, such as colonic
   thickening, epithelial injury, and inflammatory cell infiltration;
   these histologic damages were remarkably attenuated in BL-fed mice
   (Fig. [66]1F and G; Figure [67]S2). We next analyzed the effect of BL
   on colonic hyperplasia, a hallmark of C. rodentium infection [[68]15].
   The number of colonic Ki67^+ cells was dramatically reduced in C.
   rodentium-infected mice fed with BL (Fig. [69]1F and H).

Fig. 1.

   [70]Fig. 1
   [71]Open in a new tab

   Barley leaf (BL) attenuates C. rodentium (CR)-induced colitis. (A)
   Study design of in vivo mouse experiment. Mice were fed with a standard
   chow diet (CD) or an isocaloric BL-supplemented diet for 3 weeks prior
   to the infection with CR for 10 days. (B) Body weight change. (C)
   Disease activity index (DAI) scores. (D) Colon length. (E) Spleen
   weight. (F) Representative hematoxylin and eosin (H&E) staining images
   of colon tissues. Scale bar, 100 μm. Representative Ki67-stained
   immunofluorescence images of colon tissues. Scale bar, 50 μm. (G)
   Histological scores. (H) Quantification of Ki67 positive cells. (I) The
   mRNA expression of Ifnγ, Il1β, and Tnfα in the colon. (J) The levels of
   IFN-γ, IL-1β, and TNF-α in the colon. (K) The mRNA expression of CD4
   and F4/80 in the colon. In (B) - (H), (n = 8 per group). In (I) - (K),
   (n = 6 per group). Data are mean ± SEM. *P < 0.05, **P < 0.01,
   ***P < 0.001 and ****P < 0.0001. For body weight change, a repeated
   measure two-way analysis of variance (ANOVA) was performed and the rest
   of the statistics were performed with one-way ANOVA followed by Tukey’s
   multiple comparison’s test

   In addition to assessing pathological alterations in the colonic
   epithelium, the effect of BL on inflammatory factors was also examined.
   Compared to control uninfected mice, C. rodentium-infected mice showed
   significantly increased expression of Ifnγ, Il1β, and Tnfα in the colon
   at day 10 after infection; while BL downregulated these inflammatory
   cytokines but had no effect on the mRNA level of IL-4 (Fig. [72]1I).
   Correspondingly, the elevated levels of IFN-γ, IL-1β, and TNF-α caused
   by C. rodentium infection were mitigated in BL-fed mice (Fig. [73]1J).
   Moreover, BL reduced the upregulated expression of CD4^+ T cell marker
   CD4 and macrophage marker F4/80 in the colon of C. rodentium-infected
   mice, suggesting that BL suppresses C. rodentium-induced inflammatory
   responses (Fig. [74]1K). Collectively, these findings suggest that BL
   alleviates colonic injury and improves intestinal inflammation in C.
   rodentium-infected mice.

BL attenuates pathogen colonization and improves gut barrier function

   After the oral gavage, C. rodentium initially colonizes the cecum and
   then migrates to the colon [[75]17]. The severity of intestinal
   pathology during infection is primarily driven by colonization and
   virulence potential. To investigate whether BL attenuated C. rodentium
   colonization during infection, we measured the pathogen burden in C.
   rodentium-infected mice. BL supplementation significantly reduced the
   fecal C. rodentium loads on days 1, 4, 7, and 10 post-infection
   (Fig. [76]2A). Consistently, the bacterial counts in cecum content were
   significantly reduced in mice fed with BL on day 10 post-infection
   (Fig. [77]2B). Notably, the bacterial levels of C. rodentium in spleen
   and liver were also significantly lower in BL-fed mice than that in
   mice fed the control diet (Fig. [78]2C and D), suggesting that the
   systemic dissemination of enteric pathogen is mitigated by BL.

Fig. 2.

   [79]Fig. 2
   [80]Open in a new tab

   Barley leaf (BL) reduces the bacterial burden and improves the gut
   barrier function in C. rodentium (CR)-infected mice. (A) Number of CR
   in feces on days 1, 4, 7, and 10 post-infection. (B) Number of CR in
   cecum content on day 10 post-infection. (C) Number of CR in the spleen
   on day 10 post-infection. (D) Number of CR in liver on day 10
   post-infection. (E) Real-time polymerase chain reaction (PCR) assay for
   the expression of Occludin, Claudin1, Claudin2, and ZO-1 in the colon.
   (F) Representative images of alcian blue-stained colonic sections.
   Scale bar, 100 μm. (G) Quantification of mucusproducing goblet cells.
   (H) Real-time PCR assay for the expression of EspA, Map, and Tir in the
   colon. (I) Real-time PCR assay for the expression of EspA, Map, and Tir
   in the cecum. (J) Real-time PCR assay for the expression of Car4 and
   Slc26a3 in the colon. (n = 8 per group). Data are mean ± SEM.
   *P < 0.05, **P < 0.01 and ***P < 0.001. For (E), (G), and (J),
   statistical analysis was performed using one-way analysis of variance
   (ANOVA) followed by Tukey’s multiple comparison’s test, and the rest of
   the statistics were performed with Student’s t-test

   We speculated that BL might improve intestinal barrier function to
   inhibit the systemic translocation of C. rodentium. To test this
   hypothesis, we used real-time PCR to detect the expression of genes
   encoding tight junction proteins. We found that the mRNA expression of
   Claudin1, Claudin2 and ZO-1 was lower in C. rodentium-infected mice
   compared to control uninfected mice; while the expression of these
   genes was significantly elevated in mice fed with BL (Fig. [81]2E).
   Alcian blue staining revealed a reduction of mucus-producing goblet
   cells in the colon of C. rodentium-infected mice compared to that in
   control uninfected mice; whereas BL markedly ameliorated the loss of
   goblet cells caused by C. rodentium (P < 0.05) (Fig. [82]2F and G).

   The transcriptional activation of Lee pathogenicity island encoding
   T3SS is required for the colonization of C. rodentium [[83]16]. The
   LEE-encoded T3SS enables pathogens to inject their effectors into
   colonocytes [[84]16]. We further examined the impact of BL on
   Lee-encoded virulence factors. As expected, the expression of EspA,
   Map, and Tir was increased in colon and cecum of C. rodentium-infected
   mice (Fig. [85]2H and I). However, these effects were significantly
   attenuated in BL-fed mice (Fig. [86]2H and I). In addition, BL
   supplementation significantly abrogated the reduced colonic expression
   of chloride anion exchanger Sla26a3 and electrolyte absorption protein
   Car4 in C. rodentium-infected mice (Fig. [87]2J). These results
   indicate that BL improves gut barrier function to prevent C. rodentium
   colonization and attenuates infection.

BL modulates T cell immune response pathway

   To further dissect the mechanisms underlying the beneficial effect of
   BL, we performed a genome-wide transcriptome sequencing analysis on
   mouse colonic tissues. There were significant differences in mRNA gene
   expression profiles among the groups as revealed by PCA analysis and
   heat map of differentially expressed genes (DEGs) (Fig. [88]3A; Figure
   [89]S3). Venn analysis showed that there were 2063 DEGs overlapped
   between CD group vs. CD + C. rodentium group and CD + C. rodentium
   group vs. BL + C. rodentium group (Fig. [90]3B). Volcano plot revealed
   that a total of 3841 (1815 upregulated genes and 2026 downregulated
   genes) and 3001 (1647 upregulated genes and 1354 downregulated genes)
   genes differed significantly in CD versus CD + C. rodentium group and
   in CD + C. rodentium versus BL + C. rodentium group (P value < 0.05 and
   Fold change > 2), respectively (Fig. [91]3C). Notably, Th1/Th2 cell
   differentiation and Th17 cell differentiation were identified as the
   top significantly enriched KEGG pathways (Fig. [92]3D). Heat map
   analysis of DEGs on these pathways showed that C. rodentium treatment
   resulted in an increased expression of Jag2, Rela, Cd247, Cd4, Lck,
   Cd3e, Jak3, Cd3d, Lat, H2-Ea, H2-Aa, H2-Ab1, H2-Eb1, H2-DMb2, H2-DMa,
   H2-DMb1 and Ifng, and a decreased expression of Gata3, Mapk12 and
   Mapk10, and this was effectively reversed by BL (Fig. [93]3E). The
   interaction network of key genes in pathways of Th1/Th2 and Th17 cell
   differentiation was depicted by STRING analysis (Fig. [94]3F).
   Furthermore, real-time PCR was used to verify the expression of DEGs
   and the results were consistent with the transcriptome sequencing data
   (Fig. [95]3G). These data suggest that T cell immune response may
   contribute to the attenuated colitis in C. rodentium-infected mice fed
   with BL.

Fig. 3.

   [96]Fig. 3
   [97]Open in a new tab

   RNA sequencing analysis of the gene expression profiles in colonic
   tissues. (A) Principle component analysis (PCA) of transcriptional
   profiling. (B) Venn analysis of the differentially expressed genes
   (DEGs). (C) Volcano plot of the DEGs. (D) Kyoto Encyclopedia of Genes
   and Genomes (KEGG) pathway enrichment analysis of DEGs. (E) Heat map of
   DEGs in Th1/Th2 cell differentiation signaling pathway. Genes with fold
   changes of > 1.5 and P adjust (Padj) of < 0.05 were considered to be
   differentially expressed. (F) Search tool for recurring instances of
   neighboring genes (STRING) network visualization of the DEGs in the
   Th1/Th2 cell differentiation signaling pathway. (G) Real-time PCR assay
   for the DEGs in Th1/Th2 cell differentiation signaling pathway (n = 6
   per group). Data are mean ± SEM. *P < 0.05 and **P < 0.01. Statistical
   analysis was performed using one-way analysis of variance (ANOVA)
   followed by Tukey’s multiple comparison’s test

BL attenuates C. rodentium infection in a CD4^+ T cell-dependent manner

   Previous studies have shown that adaptive T cell immune response is
   essential for defending against pathogen infection [[98]19, [99]20].
   However, it is unclear whether CD4^+ T cells were involved in
   protective effect of BL against C. rodentium infection. To test the
   involvement of CD4^+ T cells in beneficial effect of BL, the anti-CD4
   antibody was administered prior and during C. rodentium infection
   (Fig. [100]4A). Consistent with previous results, mice fed with BL
   protected against C. rodentium-induced body weight loss, increased DAI
   scores, colon shortening, and splenomegaly; notably, this was not seen
   in BL-fed mice treated with anti-CD4 antibody (Fig. [101]4B-E).
   Furthermore, C. rodentium-induced colonic epithelial damage and mucosal
   barrier dysfunction were alleviated in mice fed with BL; while this was
   not observed when mice were treated with anti-CD4 antibody
   (Fig. [102]4F-H). The relieving levels of IFN-γ and IL-1β in the colon
   were also not observed in BL-fed mice treated with anti-CD4 antibody
   (Fig. [103]4I). Consistently, BL failed to reduced C. rodentium loads
   in feces and liver in mice treated with anti-CD4 antibody (Fig. [104]4J
   and K). Together, these results demonstrate that CD4^+ T cells play a
   crucial role in modulating BL-mediated protection against C. rodentium
   infection.

Fig. 4.

   [105]Fig. 4
   [106]Open in a new tab

   Barley leaf (BL) protects against C. rodentium (CR)-induced colitis in
   a CD4^+ T cell-dependent manner. (A) Study design of in vivo mouse
   experiment. Mice were fed with a standard chow diet (CD) or an
   isocaloric BL-supplemented diet for 2 weeks prior to the infection with
   CR for 10 days. For clearance of CD4^+ T cells, mice were injected
   intraperitoneally with an anti-CD4 monoclonal antibody or its isotype
   control antibody once every three days at 500 µg each time for three
   times before and during CR infection. (B) Body weight change. (C)
   Disease activity index (DAI) scores. (D) Colon length. (E) Spleen
   weight. (F) Representative hematoxylin and eosin (H&E) staining images
   of colon tissues. Scale bar, 100 μm. Representative images of alcian
   blue-stained colonic sections. Scale bar, 100 μm. (G) Histological
   scores. (H) Quantification of mucusproducing goblet cells. (I) The
   levels of IFN-γ, IL-1β, and TNF-α in the colon. (J) Number of CR in
   feces on days 1, 4, and 7 post-infection. (K) Number of CR in liver on
   day 7 post-infection. n = 8 per group. Data are mean ± SEM. *P < 0.05,
   **P < 0.01 and ***P < 0.001. For body weight change, a repeated measure
   two-way analysis of variance (ANOVA) was performed and the rest of the
   statistics were performed with one-way ANOVA followed by Tukey’s
   multiple comparison’s test

BL prevents C. rodentium-induced T cell dysregulation

   To investigate the effect of BL on T cell immune response, the
   different CD4^+ T cell subsets, including Th1 cells (CD4^+IFN-γ^+), Th2
   cells (CD4^+IL-4^+), Th17 cells (CD4^+IL-17 A^+) and T regulatory cells
   (Tregs) (CD4^+CD25^+Foxp3^+), were analyzed by flow cytometry (Figure
   [107]S4 and Figure [108]S5). Compared to control uninfected mice, C.
   rodentium infection led to significant expansion of CD4^+IFN-γ^+ T
   cells in mesenteric lymph nodes (P = 0.016), which was attenuated by BL
   (P = 0.017) (Fig. [109]5A and B). Consistently, the population of
   CD4^+IFN-γ^+ T cells in the spleen was also significantly increased in
   C. rodentium-infected mice (P < 0.001), but was reduced in mice fed
   with BL (P < 0.01) (Fig. [110]5F and G). However, there was no
   significant difference in percentage of Th2, Th17, and Tregs between
   CD + C. rodentium and BL + C. rodentium groups (Fig. [111]5C-E and
   H-J). Collectively, these data demonstrate that BL attenuates Th1
   immune response, which corresponds with attenuated inflammation and
   intestinal pathology in C. rodentium-infected mice.

Fig. 5.

   [112]Fig. 5
   [113]Open in a new tab

   Flow cytometric analysis of CD4^+ T cell subsets in mouse mesenteric
   lymph node and spleen. (A) Representative flow cytometry profile of
   CD4^+IFN-γ^+ cells, CD4^+IL-4^+ cells, CD4^+IL-17 A^+ cells and
   CD4^+CD25^+Foxp3^+ cells in mouse mesenteric lymph node. (B)
   Quantification of CD4^+IFN-γ^+ cells. (C) Quantification of CD4^+IL-4^+
   cells. (D) Quantification of CD4^+IL-17 A^+ cells. (E) Quantification
   of CD4^+CD25^+Foxp3^+ cells. (F) Representative flow cytometry profile
   of CD4^+IFN-γ^+ cells, CD4^+IL-4^+ cells, CD4^+IL-17 A^+ cells and
   CD4^+CD25^+Foxp3^+ cells in mouse colon. (G) Quantification of
   CD4^+IFN-γ^+ cells. (H) Quantification of CD4^+IL-4^+ cells. (I)
   Quantification of CD4^+IL-17 A^+ cells. (J) Quantification of
   CD4^+CD25^+Foxp3^+ cells. n = 6 per group. Data are mean ± SEM.
   **P < 0.01 and ***P < 0.001. Statistical analysis was performed using
   one-way analysis of variance (ANOVA) followed by Tukey’s multiple
   comparison’s test

BL protects against gut dysbiosis in C. rodentium-infected mice

   The gut microbiota plays an essential role in resistance and clearance
   of C. rodentium [[114]22, [115]23]. To explore the modulatory effect of
   BL on the gut microbiota, fecal bacterial composition was evaluated
   utilizing Illumina sequencing of the V3-V4 region of 16 S rRNA genes.
   Alpha diversity analysis with Chao1 and Shannon index showed that C.
   rodentium treatment led to a reduction in gut microbial richness and
   diversity, which was abolished by BL supplementation (Fig. [116]6A and
   B). Partial Least Squares Discriminant Analysis (PLS-DA) of the
   microbial community showed that there were distinct group-based
   clustering patterns among the different treatment groups
   (Fig. [117]6C). The abundances of predominant phyla and genus were
   further compared. At the phylum level, the relative abundance of
   Proteobacteria was increased in C. rodentium-infected mice and was
   reduced with BL supplementation (Fig. [118]6D; Figure [119]S6).
   Further, we observed an overgrowth of Citrobacter in C.
   rodentium-infected mice, whereas the relative abundance of
   Lactobacillus was increased in C. rodentium-infected mice fed with BL
   (Fig. [120]6E; Figure [121]S6). Correspondingly, Wilcoxon rank-sum test
   confirmed that BL supplementation induced a significant increase in
   bacteria belonging to the Lactobacillus genus in both C.
   rodentium-infected (P < 0.05) or -uninfected (P < 0.05) mice
   (Fig. [122]6F).

Fig. 6.

   [123]Fig. 6
   [124]Open in a new tab

   Barley leaf (BL) prevents C. rodentium (CR)-induced gut microbiota
   dysbiosis. (A) Chao1 index. (B) Shannon index. (C) Partial least
   squares-discriminant analysis (PLS-DA) score plots of the gut
   microbiota composition. (D) Taxonomic distributions of gut bacterial
   composition at the phylum level. (E) Taxonomic distributions of gut
   bacterial composition at the genus level. (F) Comparison of gut
   bacterial composition at the genus level. (G) Spearman correlations
   analysis between the gut bacteria and colitis-related indexes. (H)
   Concentrations of short-chain fatty acids in feces. In (A) - (G),
   (n = 6 per group). In (H), (n = 8 per group). Data are mean ± SEM.
   *P < 0.05, **P < 0.01 and ***P < 0.001. For (F), statistical analysis
   was performed using a two-tailed Wilcoxon rank-sum test by R Project.
   For (G), a statistically significant correlation was performed using
   linear regression analyses. The rest of the statistics were performed
   using a one-way analysis of variance (ANOVA) followed by Tukey’s
   multiple comparison’s test

   Next, spearman correlation analysis was performed to identify the
   bacteria that was responsible for BL-mediated protection against C.
   rodentium enteric infection. We found that Lactobacillus showed strong
   negative correlations with DAI, IL-1β, TNF-α, and histological score;
   while Citrobacter displayed a strong positive association with these
   pathologic parameters (Fig. [125]6G). We next assessed whether the
   impact of BL on gut microbiota was associated with changes in the
   CD4^+IFN-γ^+ T cells during C. rodentium infection. We found that the
   population of CD4^+IFN-γ^+ T cells was positively correlated with
   Citrobacter and negatively correlated with Lactobacillus (Figure
   [126]S7).

   Short-chain fatty acids (SCFAs) serve as critical microbial metabolites
   of SCFAs-producing bacteria such as Lactobacillus in modulating the gut
   homeostasis [[127]31–[128]34]. We found that the fecal levels of
   acetate, propionate, butyrate, isobutyrate, valerate, and isovalerate
   were reduced in C. rodentium-infected mice compared to control
   uninfected mice. However, the levels of SCFAs, which were negatively
   correlated with colitis-related indexes, were significantly elevated in
   BL-fed mice (Fig. [129]6H; Figure [130]S8). Together, these findings
   suggest that the protective effect of BL against C. rodentium-induced
   immune dysregulation is closely linked to improvement of gut microbiota
   dysbiosis.

Gut microbiota confers protection of BL against C. rodentium-induced enteric
infection

   To further confirm the role of gut microbiota in the protective effects
   of BL against C. rodentium, fecal microbiota transplantation (FMT) was
   performed (Fig. [131]7A). In this study, antibiotic treatment was used
   to deplete the gut microbes and the antibiotic-treated recipient mice
   were then gavaged with the mixture of stool pellets from donor mice
   during the FMT (Fig. [132]7A). 16 S rRNA sequencing was used to
   determine whether the gut microbiota of donor mice was successfully
   transplanted into the recipient mice (Figure [133]S9). Our result
   revealed clear discrimination between the FMT-treated two groups
   (Figure [134]S9). Importantly, there is a high similarity in the gut
   microbiota community between the donor and the corresponding recipient
   mice (Figure [135]S9). Consistent with our previous results, mice that
   received microbiota from BL-fed mice displayed an increased abundance
   of Lactobacillus compared with control recipient mice (Fig. [136]7B).

Fig. 7.

   [137]Fig. 7
   [138]Open in a new tab

   Transplantation of feces from barley leaf (BL)-fed mice protects
   against C. rodentium (CR)-induced colitis. (A) Study design of in vivo
   mouse fecal microbiota transplantation (FMT) experiment. The recipient
   mice received a two-week treatment of combined antibiotics (neomycin,
   penicillin, metronidazole, 1000 mg/l; vancomycin and streptomycin
   500 mg/l) before FMT to remove the native microbiota. The donor mice
   were fed a standard chow diet (CD) or an isocaloric BL-supplemented
   diet for 2 weeks. Fresh feces from donor mice were collected and
   resuspended in sterile PBS at 1:10 (m/v), vortexed, and rested, then
   the supernatant was taken for transplantation for three weeks. (B) Body
   weight change. (C) Disease activity index (DAI) scores. (D) Colon
   length. (E) Spleen weight. (F) Representative hematoxylin and eosin
   (H&E) staining images of colon tissues. Scale bar, 100 μm.
   Representative images of alcian blue-stained colonic sections. Scale
   bar, 100 μm. (G) Histological scores. (H) Quantification of
   mucusproducing goblet cells. (I) The levels of IFN-γ, IL-1β, and TNF-α
   in the colon. (J) Number of CR in feces on days 1, 4, and 7
   post-infection. (K) Number of CR in the spleen on day 7 post-infection.
   (L) Number of CR in the liver on day 7 post-infection. n = 8 per group.
   Data are mean ± SEM. *P < 0.05, **P < 0.01 and ***P < 0.001. For body
   weight change, a repeated measure two-way analysis of variance (ANOVA)
   was performed and the rest of the statistics were performed with
   Student’s t-test

   We further studied the functional significance of BL-mediated gut
   microbiome in the modulation of host response to C. rodentium
   infection. We found that mice received microbiota from BL-fed mice
   showed attenuated body weight loss and reduced DAI score compared to
   mice received microbiota from control CD-fed mice (Fig. [139]7C and D).
   Moreover, a longer colon length and a lower spleen weight were detected
   in mice received microbiota from BL-fed mice compared to mice received
   microbiota from control mice (Fig. [140]7E and F). Histological
   analysis of colon tissues revealed that mice received microbiota from
   BL-fed mice exhibited attenuated tissue damage and alleviated goblet
   cell depletion compared to control recipient mice (Fig. [141]7G-I). The
   severity of inflammation was improved in recipient mice that were
   colonized by BL-induced microbiota, as evidenced by reduced levels of
   IFN-γ and IL-1β in colonic tissues (Fig. [142]7J). Correspondingly, the
   bacterial level of C. rodentium in feces, spleen, and liver was reduced
   in mice received microbiota from BL-fed mice compared to control
   recipient mice (Fig. [143]7K-M). Collectively, these results confirm
   the crucial role of gut microbiota in protective effect of BL against
   C. rodentium infection.

L. plantarum attenuates C. rodentium infection-induced colonic inflammation

   Lactobacillus was identified as the most enriched genus in BL-fed mice
   and mice received microbiota from BL-fed mice (Fig. [144]6F and
   Fig. [145]7B). By using real-time polymerase chain reaction (PCR), 9
   common Lactobacillus strains L. rhamnosus, L. brevis, L. murinus, L.
   fermentum, L. reuteri, L. delbrueckii, L. casei, L. salivarius, and L.
   plantarum in the genome of the intestinal microbiota were quantified.
   Notably, L. plantarum was the only species that significantly enriched
   in BL-fed mice compared to the control CD-fed mice (Figure [146]S10A).
   We hypothesized that BL may act as a prebiotic to enrich the commensal
   bacterium L. plantarum and assessed the direct effects of BL on L.
   plantarum growth. We found that the growth of L. plantarum was
   significantly promoted by BL treatment (Figure [147]S10B). These data
   indicate that L. plantarum may play a beneficial role in alleviating C.
   rodentium-associated pathology.

   To investigate the role of L. plantarum in C. rodentium-induced
   colitis, we administered phosphate-buffered saline (PBS), live L.
   plantarum or heat-inactivated L. plantarum to C. rodentium-infected
   mice (Fig. [148]8A). The qPCR analysis confirmed that administration of
   live L. plantarum resulted in increased abundance of L. plantarum in
   mouse feces (Fig. [149]8B). We found that mice administered with L.
   plantarum has significantly attenuated manifestations of colitis
   compared to mice gavaged with PBS, as demonstrated by body weight loss
   and DAI score (Fig. [150]8C and D). Although L. plantarum had no impact
   on the spleen weight, the colon length was significantly longer in mice
   administered with L. plantarum compared to mice gavaged with PBS
   (Fig. [151]8E and F). Histological analysis revealed that C.
   rodentium-induced pathological damage and mucosal barrier dysfunction
   were also abrogated by L. plantarum (Fig. [152]8G-I). The colonic
   expression of Sla26a3 and Car4 was increased and T-bet expression was
   reduced in mice received L. plantarum administration (Fig. [153]8J and
   K). Furthermore, C. rodentium infection led to elevated levels of C.
   rodentium in feces, liver, and cecum, which were significantly reduced
   by L. plantarum (Fig. [154]8L-N). Notably, heat-inactivated L.
   plantarum failed to confer the above beneficial effects on C. rodentium
   infection (Fig. [155]8C-N). Collectively, these results indicate that
   L. plantarum enriched with BL supplementation plays a crucial role in
   protection against C. rodentium-induced intestinal inflammation.

Fig. 8.

   [156]Fig. 8
   [157]Open in a new tab

   Lactobacillus plantarum (L. plantarum) attenuates colonic inflammation
   in C. rodentium (CR)-infected mice. (A) Study design of in vivo mouse
   experiment. Mice were gavaged daily with phosphate-buffered saline
   (PBS), 10⁹ CFU of live L. plantarum or 10⁹ CFU of heat-inactivated L.
   plantarum (H-L. plantarum) for 2 weeks prior to the infection with CR
   for 12 days. (B) Real-time polymerase chain reaction (PCR) assay for
   the quantification of L. plantarum in mouse feces. (C) Body weight
   change. (D) Disease activity index (DAI) scores. (E) Colon length. (F)
   Spleen weight. (G) Representative hematoxylin and eosin (H&E) staining
   images of colon tissues. Scale bar, 100 μm. Representative images of
   alcian blue-stained colonic sections. Scale bar, 100 μm. (H)
   Histological scores. (I) Quantification of mucusproducing goblet cells.
   (J) Real-time PCR assay for the expression of T-bet in the colon. (K)
   Real-time PCR assay for the expression of Slc26a3 and Car4 in the
   colon. (L) Number of CR in feces on days 1 post-infection. (M) Number
   of CR in liver on day 1 post-infection. (N) Number of CR in cecum
   content on day 1 post-infection. n = 6 per group. Data are Mean ± SEM.
   *P < 0.05, **P < 0.01 and ***P < 0.001. For (B), Student’s t-test was
   performed. For (C), a repeated measure two-way analysis of variance
   (ANOVA) was performed and the rest of the statistics were performed
   with one-way ANOVA followed by Tukey’s multiple comparison’s test

L. plantarum ameliorates C. rodentium-induced colitis through its secreted
EVs

   Bacterial EVs are membrane-enclosed lipid bilayer nanoparticles that
   play essential roles in communication between gut microbiome and human
   health [[158]35]. Recently, the EVs derived from probiotics
   particularly Lactobacillus have emerged as potential mediators of host
   immune response and anti-inflammatory effect [[159]35].

   However, the role of L. plantarum-derived EVs (L-EVs) in C. rodentium
   infection and intestinal inflammation remains unclear. We thus isolated
   EVs from the supernatants of L. plantarum using an
   ultracentrifugation-based system (Fig. [160]9A). The scanning electron
   microscopy and transmission electron microscopy showed that the
   isolated L-EVs were double-layer membrane-enclosed nanoparticles with
   spherical morphology (Fig. [161]9B and C). The nanoparticle tracking
   analysis (NTA) analysis showed that the mean particle size of L-EVs
   ranged from 100 nm to 200 nm in diameter (Fig. [162]9D), which was
   consistent with the size of previous reports [[163]35]. We further
   performed proteomic analysis to characterize the protein composition of
   L-EVs. A total of 380 proteins were identified to be present in L-EVs.
   The biological functions of these proteins were analyzed by using the
   Gene Ontology (GO) database. The top ranked proteins are localized to
   the plasma membrane (26.61%), suggesting that L-EVs might originate
   directly from the budding of plasma membrane. KEGG pathway enrichment
   analysis further indicated that L-EVs-enriched proteins were primarily
   associated with amino acid metabolism, energy metabolism, antimicrobial
   and infectious disease. Specifically, most of them were found to be
   significantly involved in the modulation of signal peptide, or via
   JAK/STAT pathway, NF-κB pathway and TLR pathway [[164]36–[165]39].

Fig. 9.

   [166]Fig. 9
   [167]Open in a new tab

   Preparation and characterization of L. plantarum-derived EVs (L-EVs).
   (A) Isolation and purification procedures of L-EVs. (B) Representative
   image of the scanning electron microscope for L. plantarum showing
   membrane vesicles on the bacterial cell surface. (C) Representative
   image of the transmission electron microscopy for L-EVs. (D)
   Nanoparticle tracking analysis was performed to detect the size
   distribution of L-EVs. (E) Go secondary classification statistical
   charts of L-EVs. (F) KEGG annotated statistical charts of L-EVs

   We evaluated the biosafety of L-EVs. There is no notable difference in
   daily food intake, body weight changes, and weight of organs (colon,
   liver, and spleen) between the L-EVs group and the control group
   (Figure [168]S11A-E). Histological analysis showed that oral
   administration of L-EVs at the different dose did not cause any colon,
   liver, and spleen damage (Figure [169]S11F). Furthermore, cytokine
   profiling analysis indicated no significant difference in serum levels
   of IFN-γ and TNF-α (Figure [170]S11G and H). In addition, we
   supplemented in vivo studies with DiR-labeled L-EVs and demonstrated
   that these vesicles remain stable in circulation and effectively reach
   and persist in the colon without significant off-target distribution
   (Figure [171]S12). These results suggest that L-EVs exhibit favourable
   stability and safety profiles.

   To investigate whether the beneficial effects of L. plantarum were
   associated with their secreted EVs, L-EVs were administered to mice by
   oral gavage followed by the infection with C. rodentium
   (Fig. [172]10A). Our results showed that the dramatic loss of body
   weight and increased DAI score in C. rodentium-infected mice were
   significantly reduced by L-EVs intervention (Fig. [173]10B and C).
   Moreover, L-EVs mitigated C. rodentium-induced colon length shortening
   and increased spleen weight (Fig. [174]10D and E). Histopathological
   analysis showed that C. rodentium-induced increase in pathology score
   and mucosal barrier injury could be alleviated by L-EVs treatment
   (Fig. [175]10F-H). Similar trends were obtained with the expression of
   Ifnγ, Tnfα, Sla26a3 and Car4 in colon tissues (Fig. [176]10I and J).
   The C. rodentium loads in feces, liver, and cecum were significantly
   reduced in L-EVs-treated mice (Fig. [177]10K-M). We then further
   analyzed the population of CD4^+ T cells. The increased population of
   CD4^+IFN-γ^+ T cells in the spleen was significantly reduced by L-EVs
   intervention (Fig. [178]10N and O). Collectively, these results suggest
   that L. plantarum attenuates C. rodentium infection and intestinal
   inflammation via its secreted EVs.

Fig. 10.

   [179]Fig. 10
   [180]Open in a new tab

   L. plantarum-derived EVs (L-EVs) ameliorates C. rodentium (CR)-induced
   colitis. (A) Study design of in vivo mouse experiment. Mice were
   gavaged daily with phosphate-buffered saline (PBS) or 50 µg of L-EVs
   for 2 weeks prior to the infection with CR for 12 days. (B) Body weight
   change. (C) Disease activity index (DAI) scores. (D) Colon length. (E)
   Spleen weight. (F) Representative hematoxylin and eosin (H&E) staining
   images of colon tissues. Scale bar, 100 μm. Representative images of
   alcian blue-stained colonic sections. Scale bar, 100 μm. (G)
   Histological scores. (H) Quantification of mucusproducing goblet cells.
   (I) Real-time PCR assay for the expression of Ifnγand Tnfα in the
   colon. (J) Real-time PCR assay for the expression of Slc26a3 and Car4
   in the colon. (K) Number of CR in feces on days 1 post-infection. (L)
   Number of CR in liver on day 1 post-infection. (M) Number of CR in
   cecum content on day 1 post-infection. (N) Representative flow
   cytometry profile of CD4^+IFN-γ^+ cells. (O) Quantification of
   CD4^+IFN-γ^+ cells. n = 5–6 per group. Data are Mean ± SEM. *P < 0.05,
   **P < 0.01 and ***P < 0.001. For (B), a repeated measure two-way
   analysis of variance (ANOVA) was performed and the rest of the
   statistics were performed Student’s t-test

Discussion

   The impacts of global increase in the prevalence of IBD on human public
   health are fundamental and challenging [[181]4]. The pathogenesis of
   IBD is thought to associate with the dysregulated interplay between
   environmental changes, gut microbiota, and host immune response
   [[182]9]. Invasion by pathogenic microbes is an important source of
   factor contributing to infectious diarrhea and colitis. Although our
   previous studies have shown that BL is effective in protecting against
   chemically-induced intestinal damage [[183]28–[184]30], the potential
   benefits of BL on enteropathogenic infection and intestinal
   inflammation is unclear. In this study, the mechanism underlying the
   protective effects of BL against C. rodentium infection and colitis was
   studied.

   The BL dosage used in this study is a human equivalent dose that has
   been reported without adverse effects [[185]40]. Our results showed
   that dietary supplementation with BL alleviated C. rodentium-induced
   gut barrier impairment and pro-inflammatory Th1 immune response. The
   protective effects of BL were associated with improvement of gut
   microbiota dysbiosis and enrichment of Lactobacillus. More
   significantly, our results demonstrated for the first time that EVs
   secreted by L. plantarum act as an important mediator of the protection
   against C. rodentium-induced bacterial colitis and suppression of host
   Th1 immune response dysregulation. Therefore, these results shed light
   on the potential use of BL and probiotic-derived EVs for the clinical
   management of infection-associated intestinal inflammatory disorders in
   humans.

   We found that mice infected with C. rodentium developed severe colitis
   manifestations including weight loss, colon shortening, and spleen
   hyperplasia, which are consistent with previous studies [[186]14,
   [187]17]. The deleterious effects caused by C. rodentium infection were
   attenuated in BL-fed mice. Crypt hyperplasia, which is characterized by
   an overproliferation of undifferentiated Ki67^+ cells, is an important
   hallmark of C. rodentium-induced infection [[188]14, [189]17]. The
   induction of colonic crypt hyperplasia is a mechanism by which C.
   rodentium facilitates aerobic respiration to compete with the gut
   microbiota during infection [[190]41]. This process is regulated by the
   T3SS, which plays a central role in determining the colonization
   efficiency of C. rodentium [[191]41]. In this study, we observed that
   C. rodentium-induced increase in the number of colonic Ki67^+ cells was
   abrogated in BL-fed mice. Moreover, the bacterial loads of C. rodentium
   in feces and cecum contents were reduced in BL-fed mice compared to the
   control mice. This was accompanied by a low expression level of
   T3SS-associated virulence factors (Tir, Map, and EspA) in the colon and
   cecum of BL-fed mice. Thus, our data suggest that BL alleviates C.
   rodentium infection by limiting the bacterial colonization and
   mitigating colonic crypt hyperplasia.

   The intestinal epithelial barrier, which composes of a single layer of
   epithelial cells through the tight junction and adhesion proteins,
   serves as a defensive system involving host-microbial interaction and
   intestinal homeostasis [[192]42]. The colonic goblet cells secret MUC2
   to form a mucus layer, which is a key defense against enteric pathogens
   [[193]43]. In this study, we found that C. rodentium infection caused a
   reduction of mucus-producing goblet cells in the colon, which was
   effectively ameliorated by BL intervention. The mRNA expression of
   tight junction proteins and adhesion junctions (Occludin, Claudin1, and
   ZO-1) as well as chloride anion exchanger and water absorption proteins
   (Sla26a3 and Car4) [[194]44] was restored in C. rodentium-infected mice
   fed with BL. These results may partially explain the reduced bacterial
   loads of C. rodentium in the spleen and liver of BL-fed mice and
   suggest that BL prevents the extraintestinal transmission of C.
   rodentium by improving the gut barrier function.

   A finely tuned inflammatory response is necessary to defend against C.
   rodentium-induced infection by enhancing gut barrier and limiting
   bacterial translocation [[195]18]. However, uncontrolled or prolonged
   inflammatory response may lead to intestinal barrier impairment and
   increased risk of infection complications. Here, we found that C.
   rodentium infection induced a robust inflammatory response
   characterized by the increased production of inflammatory cytokines
   including TNF-α, IFN-γ, and IL-1β; meanwhile, the expression level of
   macrophage marker F4/80 in colonic tissue was elevated in mice infected
   with C. rodentium. BL effectively dampened the increased expression
   levels of TNF-α, IFN-γ, and IL-1β. Thus, the reduced bacterial loads
   and improved gut barrier function in BL-fed mice may be due to the
   attenuation of C. rodentium-induced inflammatory responses.

   We next used RNA sequencing to investigate the mechanisms underlying
   the anti-inflammatory effects of BL against C. rodentium infection.
   Intriguingly, functional analysis showed that the pathways of Th1/Th2
   and Th17 cell differentiation were significantly enriched in BL-fed
   mice. As the major producers of inflammatory cytokines, adaptive immune
   CD4^+ T cells are essential for defending against C. rodentium
   infection [[196]19, [197]20]. It is reported that CD4^+ T cells can be
   subdivided into Th1, Th2, Th17, and Tregs [[198]45]. C. rodentium
   infection in mice elicits a Th1/Th17 response similar to those of
   patients with pathogenic EPEC/EHEC infections [[199]46]. Notably,
   depletion of CD4^+ T cells by anti-CD4 antibody treatment effectively
   abrogated the beneficial effects of BL on C. rodentium-induced colitis,
   suggesting that CD4^+ T cells play a role in regulating the protection
   against C. rodentium infection. To further evaluate the impacts of BL
   on C. rodentium-induced T cell immune response, flow cytometry was used
   to analyze the different CD4^+ T cell subsets in the mesenteric lymph
   node and spleen. We found that C. rodentium-infected mice developed a
   considerable increase in the proportion of Th1 cells (CD4^+IFN-γ^+),
   which was attenuated in BL-fed mice. Notably, the proportions of Th2
   cells (CD4^+IL-4^+), Th17 cells (CD4^+IL-17 A^+), and Treg cells
   (CD4^+CD25^+Foxp3^+) were not affected. These results indicate that BL
   alleviates C. rodentium-induced colitis at least in part by alleviating
   CD4^+ T cell dysregulation.

   The gut microbiota has been known as a major regulator in pathogen-host
   interactions and plays a significant role in controlling the severity
   of C. rodentium-associated colon pathology [[200]16, [201]17, [202]43].
   Accumulating evidence has shown that marked variations exist in the
   composition of the gut microbiota between IBD patients and healthy
   individuals. For instance, gut microbiota dysbiosis featured by a
   reduction of microbial diversity was observed in patients with IBD
   [[203]47, [204]48]. The reduced gut bacterial diversity may be due to
   the expansion of facultative anaerobic pathogens in Proteobacteria,
   such as Salmonella, Escherichia coli, and Shigella [[205]49]. In line
   with these findings, our 16 S rRNA gene sequencing analysis showed that
   C. rodentium infection induced a decreased bacterial diversity and an
   increased abundance of Proteobacteria. The abundance of Citrobacter was
   elevated in C. rodentium-infected mice but dropped markedly in BL-fed
   mice, suggesting that C. rodentium-induced gut dysbiosis was
   effectively abrogated. Importantly, the beneficial effect of BL against
   C. rodentium-induced colitis was transferable via FMT, suggesting that
   the consumption of BL may shape a gut microenvironment that is
   sufficient to alleviate C. rodentium infection and intestinal
   inflammation. It is important to note that the 16 S rRNA sequencing
   employed in this study primarily focuses on the bacterial microbiota,
   and therefore does not account for other microbial groups such as fungi
   or viruses. Both fungi and viruses have been implicated in the
   modulation of gut health and inflammation, and their roles in colitis
   are increasingly recognized. As such, the absence of fungal and viral
   data is a limitation of our study. Future research incorporating more
   comprehensive sequencing techniques, such as metagenomics or broader
   18 S rRNA sequencing, could provide a more holistic view of the
   microbiome and its interactions in colitis.

   Further analysis revealed that BL substantially enhanced the relative
   abundance of Lactobacillus in both uninfected and infected mice. This
   result is consistent with our previous study showing that the in vitro
   anaerobic fermentation of BL with mouse feces resulted in an enrichment
   of Lactobacillus [[206]29]. We thus assessed the effects of BL on L.
   plantarum growth. Our results showed that BL had a capacity to promote
   the growth of L. plantarum. Our previous study has shown that dietary
   fibers are the primary nutritional components of BL [[207]50].
   Together, these findings suggest that BL may act as a prebiotic agent
   to promote the growth of L. plantarum. Further studies are required to
   study the molecular mechanisms underlying the effect of BL on L.
   plantarum growth.

   Lactobacillus is an important type of probiotic and is extensively used
   in dairy food. Several Lactobacillus species have been reported to be
   effective in ameliorating C. rodentium-induced bacterial colitis. For
   instance, L. reuteri could activate Wnt/β-catenin pathway to promote
   the proliferation of intestinal organoids, thereby protecting the
   intestinal mucosal barrier against pathogenic bacterial invasion
   [[208]51]. L. johnsonii could regulate the endoplasmic reticulum stress
   signalling and attenuate inflammatory responses in C.
   rodentium-infected mice [[209]52]. However, little is known about the
   health benefits of L. plantarum in the context of C. rodentium
   infection. This study, to our knowledge, is the first to characterize
   the beneficial effects of L. plantarum on C. rodentium-induced colitis,
   highlighting its potential in future clinical application for IBD
   patients.

   EVs are nano-sized, membrane-bound particles released by various
   cellular sources and play pivotal roles in various biological
   processes, such as cell-to-cell communication, tissue homeostasis and
   immune regulation [[210]53]. Recently, commensal probiotic-derived EVs
   have emerged as important mediators of host-pathogen interactions. For
   instance, EVs derived from L. johnsonii were found to alleviate
   intestinal injury in ETECK88-infected mice through enhancing M2
   macrophage polarization [[211]54]. Moreover, administration of L.
   reuteri BBC3-derived EVs has been shown to mitigate
   lipopolysaccharide-induced intestinal inflammation in broiler chickens
   [[212]55]. These findings suggest that bacterial EVs are becoming
   promising interventions or supplements for promoting intestinal
   homeostasis, due to their unique nano-scale structure and benefits on
   safety, stability, and therapeutic efficacy in modulating host health
   [[213]56]. Consistent with these previous results, the present study
   revealed a key role of EVs secreted by L. plantarum in alleviating C.
   rodentium-induced ulcerative colitis. Further studies are needed to
   identify the specific molecules that are involved in the protective
   effects of L-EVs and investigate the underlying molecular mechanisms.

   BL is a nutrient-dense plant material rich in dietary fiber,
   polyphenols, vitamins, and antioxidants that has been shown to
   favorably influence intestinal health. Its bioactive compounds
   contribute to maintaining gut homeostasis by modulating the composition
   of the gut microbiota, enhancing intestinal barrier integrity, and
   reducing inflammation. For example, the insoluble fibers in BL serve as
   substrates for beneficial bacteria, promoting the production of
   short-chain fatty acids (SCFAs) such as butyrate, which not only
   nourish colonocytes but also play a role in regulating immune responses
   and preserving epithelial integrity. In addition, the polyphenolic
   compounds act as antioxidants and anti-inflammatory agents, further
   supporting a balanced gut environment and potentially mitigating the
   risks associated with chronic inflammatory conditions.

   In parallel, L-EVs represent a cutting-edge postbiotic approach to
   modulating gut health. These nanoscale vesicles are naturally released
   by the bacteria and carry a diverse array of bioactive molecules,
   including proteins, lipids, and nucleic acids. Through their
   interaction with host cells, L-EVs could modulate signaling pathways
   that regulate immune responses and promote tissue homeostasis. Our
   results suggest that they might stimulate regulatory pathways that lead
   to the differentiation of T regulatory cells and bolster mucosal
   immunity, thereby contributing to the maintenance of an optimal
   intestinal environment.

   Together, BL and L-EVs offer a multifaceted strategy for promoting
   intestinal homeostasis. While BL acts as a dietary supplement that
   directly nourishes the gut microbiota and supports barrier function,
   L-EVs provide a targeted means of modulating host immune responses and
   reinforcing the structural and functional integrity of the gut. This
   dual intervention approach not only enhances the resilience of the
   intestinal barrier but also supports a balanced immune response,
   suggesting promising avenues for both preventive and therapeutic
   applications in gut health management. Further studies are warranted to
   fully elucidate the mechanisms underlying these effects and to
   translate these findings into clinical practice.

   In conclusion, the results from our study reveal that dietary
   supplementation with BL ameliorates the severity of C.
   rodentium-induced enteric infection and inflammation. The beneficial
   effects of BL are related to the improvement of gut barrier function
   and suppression of pro-inflammatory Th1 immune response. Moreover, BL
   reduces the C. rodentium burden and prevents C. rodentium-induced gut
   microbiota dysbiosis by promoting the abundance of Lactobacillus. L.
   plantarum and its secreted EVs are identified as an important mediator
   required for the immunomodulatory effects of BL on pathogen infection
   (Fig. [214]11). These findings support the development of BL and L-EVs
   as novel intervention approaches for infection-associated colonic
   inflammation.

Fig. 11.

   [215]Fig. 11
   [216]Open in a new tab

   Barley leaf (BL) attenuates C. rodentium infection and intestinal
   inflammation via L. plantarum-derived extracellular vesicles (L-EVs).
   The beneficial effects of BL are related to the improvement of mucosal
   barrier function and suppression of pro-inflammatory T helper 1 (Th1)
   immune response. Moreover, BL reduces the C. rodentium burden and
   alleviates C. rodentium-induced gut microbiota dysbiosis by promoting
   the growth of L. plantarum. Mechanistically, EVs released by L.
   plantarum alleviate C. rodentium-associated pathology and T cell
   dysregulation. L. plantarum and its secreted EVs are identified as
   important mediators required for the immunomodulatory effects of BL on
   pathogen infection. This study highlights the potential use of BL, L.
   plantarum and its secreted EVs for the clinical management of
   infection-associated intestinal inflammatory disorders in humans

Materials and methods

Preparation of BL powder

   The BL powder was made in accordance with our previous studies
   [[217]28–[218]30]. Briefly, fresh Hordeum vulgare L. leaves (cultivated
   in Hangzhou, China) were washed, sliced, and dried. The powdered dried
   barley leaves were then sifted through a 300-mesh sieve. Table [219]S1
   displays the nutritional composition of the barley leaves used in this
   experiment.

Animals

   Male C3H/HeN mice aged 4–6 weeks were obtained from Beijing Vital River
   Laboratory Animal Technology Co., Ltd. (Beijing, China) and then
   acclimatized under specific pathogen-free (SPF) conditions. After
   adaptation, mice were randomly assigned and fed a standard chow diet or
   an isocaloric diet, in which 2.5% BL was added. The macronutrient
   composition of the two diets was shown in Table [220]S2. Animal
   experiments were conducted following the National Institutes of Health
   guide for the care and use of Laboratory Animals (NIH Publications No.
   8023, revised 1978), and the protocols were reviewed and approved by
   the Animal Care and Ethics Committee of China Agricultural University
   (Ethics reference number: AW32602202-4-3、AW52802202-4-1).

C. rodentium infection

   To induce bacterial colitis, mice were inoculated with approximately
   1 × 10^9 colony forming units (CFU)/mouse of C. rodentium strain DBS
   100 (ATCC 51459). Briefly, a single colony growing on a MacConkey agar
   plate (Solatbio, China) was picked and inoculated in 5 ml sterile
   Luria-Bertani (LB) medium overnight, then the suspension was diluted in
   fresh LB medium at the ratio of 1:100 and incubated at 37 ℃,
   centrifuged and resuspended in PBS for later mouse infestation.

C. rodentium determination in mouse feces and tissues

   Fresh mouse stool pellets were collected, weighed, and homogenized on
   days 1, 4, 7, and 10 following infection. Following that, sample
   homogenates were serially diluted ten-fold and plated on MacConkey
   agar, where C. rodentium colonies were counted the next day. For organ
   bacterial burden analysis, cecum contents, spleens and livers were
   collected and processed as above.

Disease activity index (DAI)

   The DAI score, which takes into account weight loss, stool consistency,
   and general health status, was used to assess the severity of colitis
   as previously described [[221]17].

Histological staining

   The paraffin-embedded samples were sectioned, and slides were subjected
   to histological hematoxylin and eosin staining (H&E) before being
   observed with a microscope. The slides were evaluated as previously
   described [[222]17].

   Paraffin sections were stained with alcian blue on the manufacturer’s
   recommendations using commercial kits (Servicebio Technology CO., Ltd.,
   China) for goblet cell analysis. Image analysis was performed by
   utilizing Image J software.

Immunofluorescence staining

   After being deparaffinized and rehydrated, the paraffin-embedded
   sections underwent antigen retrieval using citric acid antigen
   retrieval buffer (pH 6.0). Slides were then 30 min later blocked in
   goat serum. The primary antibody used was rabbit antisera made against
   Ki67 (1:200; Abcam), and to specifically label the marker,
   CY3-conjugated goat anti-rabbit IgG (1:400; Jackson) was added as a
   secondary antibody. Sections were then sealed with an anti-fluorescence
   quenching agent after being stained with 4’, 6-diamidino-2-phenylindole
   (DAPI; Solarbio). Under the Nikon fluorescence microscope, the pictures
   were observed and gathered.

Inflammatory cytokine analysis

   An enzyme-linked immunosorbent assay (ELISA) kit (Nanjing Jiancheng
   Bioengineering Institute Co., Ltd, China) was used to assess cytokine
   contents in the colon according to the manufacturer’s instructions.

Quantitative real-time PCR (RT-qPCR)

   Briefly, total RNA was extracted from the colon or cecal samples stored
   at -80 °C utilizing the Trizol reagent (Invitrogen, USA) and reversed
   into cDNA according to the instructions of the FastQuant RT Kit
   (TianGen, China). RT-qPCR was performed using SYBR Real-time PCR Kit
   (Takara, Japan) by a LightCycler 480 Real-Time PCR system (Roche,
   Switzerland). Data were analyzed by the 2^-ΔΔCT method. The sequences
   of primers used in RT-qPCR are listed in Table [223]S3.

16 S rRNA gene sequencing

   Total DNA extracted from mouse feces was used as a template for PCR
   amplification of the bacterial 16S rRNA gene’s V3-V4 region. PCR
   amplification products from an ABI GeneAmp^® 9700 PCR system
   (AppliedBiosystems, Foster City, CA, USA) were sequenced on the
   Illumina Miseq platform in Majorbio Bio-Pharm Technology Co., Ltd.
   (Shanghai, China) and classified into operational taxonomic units
   (OTUs) with a difference of 0.03 (equivalent to 97% similarity). By
   visualizing datasets of microbial diversity and abundance from various
   samples, taxonomic community structure and phylogeny were assessed. The
   Majorbio Cloud Platform was used for all data analysis.

SCFAs analysis

   As previously described, the determination of SCFAs in fecal samples
   using gas chromatography was performed [[224]50].

Fecal microbiota transplantation (FMT)

   FMT was performed as previously described. Briefly, the recipient mice
   received a two-week treatment of combined antibiotics (neomycin,
   penicillin, metronidazole, 1000 mg/l; vancomycin and streptomycin
   500 mg/l) before FMT to remove the native microbiota. The donor mice
   were fed CD or isocaloric BL chow, respectively. Fresh feces from donor
   mice were collected and resuspended in sterile PBS at 1:10 (m/v),
   vortexed, and rested, then the supernatant was taken for
   transplantation for three weeks. Donor and recipient mouse feces were
   collected for 16 S rRNA sequencing analysis upon completion of
   transplantation.

RNA-sequencing analysis

   Total RNA was extracted from mouse colon tissue. The quality and RNA
   integrity numbers (RINs) of the extracted RNA were determined using
   Nanodrop2000 and Agilent2100, respectively. Then RNA sequence libraries
   were constructed according to the Illumina TruseqTM RNA sample prep Kit
   protocol. Majorbio Bio-Pharm Technology Co., Ltd. (Shanghai, China)
   performed sequencing of the libraries on an Illumina HiSeq2000
   instrument and individually assessed for quality using FastQC.
   Differential expression analysis was performed by DESeq2. Genes with
   fold changes (FC) > 1.5 and P adjust (Padj) < 0.05 between groups were
   considered as differentially expressed genes (DEGs). Statistical
   significance was assessed using the negative binomial Wald test,
   followed by correction for multiple hypothesis testing using the
   Benjamini-Hochberg method. Functional enrichment clustering analysis
   was used for the Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway
   enrichment analysis.

CD4^+ T cell depletion

   For the clearance of CD4^+ T cells, mice were injected
   intraperitoneally with anti-CD4^+ monoclonal antibody or its isotype
   control antibody once every three days at 500 μg each time for three
   times before and during C. rodentium infection.

Flow cytometry analysis for T cells

   Single-cell suspensions were obtained by grinding mouse mesenteric
   lymph node and spleen tissues, incubated and stimulated with PMA,
   Ionomycin, Monensin, and RPMI 1640 medium (Solarbio, China), and
   cultured at 37 °C for 5 h under 5% CO[2] condition. Surface markers
   were stained with PerCP/Cyanine5.5 anti-mouse CD3 (Biolegend, USA),
   PE/Dazzle™ 594 anti-mouse CD4 antibody (Biolegend, USA), and APC
   anti-mouse CD25 (Biolegend, USA) for 0.5 h at 4 °C against light, then
   fixed and permeabilized, followed by PE/Cyanine7 anti-mouse IFN-γ
   (Biolegend, USA), Alexa Fluor^® 488 anti-mouse IL-4 (Biolegend, USA),
   APC/Cyanine7 anti-mouse IL-17 A (Biolegend, USA), and PE anti-mouse
   FOXP3 (Biolegend, USA) intracellular staining was performed for 0.5 h.
   Finally, cells were detected by flow cytometry.

L. plantarum cultivation and treatment

   L. plantarum was previously isolated by our laboratory and routinely
   incubated in de Man, Rogosa, and Sharpe (MRS) broth at 37 °C under
   anaerobic conditions. This strain was identified by morphological and
   physicochemical characterizations and 16 S rRNA sequence analysis. The
   growth of L. plantarum was monitored by measuring the optical density
   at 600 nm (OD[600nm]). For L. plantarum treatment, L. plantarum was
   centrifuged, enriched, and resus pended in phosphate buffer solution
   (PBS). Mice were administered with 0.2 mL L. plantarum by oral gavage
   (1 × 10^9 CFU /mL) per day for two weeks. Regarding the heat
   inactivation of L. plantarum, viability assays (plating on MRS media)
   were performed to verify that no residual live bacteria remained after
   the heat inactivation process.

Isolation of L-EVs

   The isolation of EVs from L. plantarum strain culture supernatant was
   performed by a series of centrifugation steps as described [[225]54,
   [226]55]. Briefly, after growing in MRS broth for 24 h, the bacterial
   culture solution was centrifuged at 1,000 g for 10 min. The
   bacteria-free culture supernatant was collected and centrifuged at
   12,000 g for 30 min. Following that, the supernatant was filtered via a
   0.22 μm bottle top filter (Corning, NY) to remove any cell debris from
   the medium. The L-EVs pellets were obtained by ultracentrifugation at
   12,000 g for 3 h at 4 °C. Finally, purified L-EVs were suspended in PBS
   and stored at -80 °C until further use. For L-EVs treatment, mice were
   orally administered 50 µg of L-EVs per day for two weeks.

Characterization of L-EVs

   Morphological characteristics of L. plantarum and L-EVs were detected
   by scanning electron microscopy (SEM). Samples were fixed with 2.5%
   PBS-diluted glutar aldehyde. After being rinsed three times in PBS,
   samples were post-fixed in 1% osmium tetroxide for 1 h, dehydrated in
   alcohol, and then critical-point-dried using CO[2]. The samples were
   coated with gold and observed in a Hitachi S-3400 scanning electron
   microscope (Hita chi, Japan).

   The morphology of the L-EVs was observed using transmission electron
   microscope (TEM). For sample preparation, L-EVs were fixed with 2.5%
   glutaraldehyde and were then fixed with 1% osmium tetroxide at 4 °C for
   2 h, dehydrated in acetone. Next, the samples were post-fixed in 2%
   osmium tetroxide for 1 h and 1% uranylacetate. The sections were
   post-stained with uranyl acetate and lead citrate then examined under a
   H-7650 transmission electron microscope (Hitachi, Japan).

Nanoparticle tracking analysis

   The particle size and number of L-EVs was quantified using the ZetaView
   PMX 110 particle analyzer under 405 nm excitation light. Laser light
   was used to irradiate the nanoparticle suspension, and the scattered
   light from the nanoparticles was detected. The nanoparticle
   concentration was calculated by counting the scattered particles.

Proteomic analysis

   Briefly, L-EVs were lysed on ice by sonication in buffer containing 8 M
   urea, 1% protease inhibitor cocktail and 2 mM EDTA. Following
   centrifugation at 12 000×g for 15 min at 4℃, protein concentrations of
   the supernatants were determined by BCA assay. Extracted proteins
   (100 µg per replicate) were reduced, alkylated and then digested with
   sequencing-grade trypsin (Promega, #V5111) at a 1:50 (w/w) enzyme:
   protein ratio overnight at 37℃. The resulting peptide mixtures were
   fractionated by high-pH reversed-phase HPLC and each fraction analyzed
   by UPLC-MS/MS on a Q Exactive™ Plus mass spectrometer (Thermo Fisher
   Scientific) coupled online to an Easy-nLC 1200 system. Data-independent
   acquisition was employed. Raw spectra were searched against the draft
   genome of L. plantarum in the UniProt database using MaxQuant (v1.6).
   The biological function of the identified proteins was classified by
   Gene Ontology (GO) annotation from the UniProt-GOA database. KEGG
   pathway database was used to analyze the metabolic pathways involved in
   the differentially expressed proteins.

Distribution of L-EVs in vivo

   L-EVs were incubated with the fluorescent dye DiR at 37 ℃for 1 h, after
   which DiR-labeled L-EVs were isolated by ultracentrifugation and
   resuspended in PBS. Each mouse was gavaged with DiR-labeled L-EVs at
   500 μg/200µL PBS. The distribution of L-EVs was visualized by in vivo
   imaging using an Odyssey CLx imaging system (LI-COR Biosciences,
   Lincoln, NE, USA) at 8 h after administration.

Biosafety evaluation of L-EVs in vivo

   Six-week-old C57BL/6 mice were randomly divided into four groups:
   Control, L-L-EVs (25 µg/d), M-L-EVs (50 µg/d) and H-L-EVs (100 µg/d).
   These L-EVs were administered daily for a total of 12 times. The
   changes in the body weight and food intake of the mice were recorded.
   Then, blood from the mice was collected, and the concentrations of
   TNF-α and IFN-γ in the serum were measured. The major organs (liver,
   spleen and colon) were subsequently collected for further H&E staining.

Statistics

   Statistical analysis was carried out using Prism 6.0 (GraphPad
   Software, CA). Data are presented as means ± SEM. Significant
   differences between the two groups were evaluated by a two-tailed
   unpaired Student’s t-test. Significant differences in more than two
   groups were evaluated by one-way or two-way analysis of variance
   (ANOVA) followed by Tukey’s multiple comparisons test.

Electronic supplementary material

   Below is the link to the electronic supplementary material.
   [227]Supplementary Material 1^ (3.2MB, docx)

Acknowledgements