Lactobacillus reuteri maintains intestinal epithelial regeneration and repairs damaged intestinal mucosa

Abstract

Little is known about the regulatory effect of microbiota on the proliferation and regeneration of ISCs. Here, we found that L. reuteri stimulated the proliferation of intestinal epithelia by increasing the expression of R-spondins and thus activating the Wnt/β-catenin pathway. The proliferation-stimulating effect of Lactobacillus on repair is further enhanced under TNF -induced intestinal mucosal damage, and the number of Lgr5⁺ cells is maintained. Moreover, compared to the effects of C. rodentium on the induction of intestinal inflammation and crypt hyperplasia in mice, L. reuteri protected the intestinal mucosal barrier integrity by moderately modulating the Wnt/β-catenin signaling pathway to avoid overactivation. L. reuteri had the ability to maintain the number of Lgr5⁺ cells and stimulate intestinal epithelial proliferation to repair epithelial damage and reduce proinflammatory cytokine secretion in the intestine and the LPS concentration in serum. Moreover, activation of the Wnt/β-catenin pathway also induced differentiation toward Paneth cells and increased antimicrobial peptide expression to inhibit C. rodentium colonization. The protective effect of Lactobacillus against C. rodentium infection disappeared upon application of the Wnt antagonist Wnt-C59 in both mice and intestinal organoids. This study demonstrates that Lactobacillus is effective at maintaining intestinal epithelial regeneration and homeostasis as well as at repairing intestinal damage after pathological injury and is thus a promising alternative therapeutic method for intestinal inflammation.

Keywords: Lactobacillus; Wnt/β-catenin pathway; intestinal epithelia; intestinal stem cells; proliferation.

Figure 1.

L. reuteri D8 upregulated intestinal organoid proliferation under physiological conditions. (a) Coculture model of L. reuteri D8 and organoids. (b) Crypts from small intestines were seeded onto Matrigel and cultured for 3 days to obtain well-developed organoids. Scale bars, 100 μm. (c) Organoids were treated with or without L. reuteri D8 (10⁶ CFU per well) for 48 h. The surface area of organoids was calculated. Scale bars, 50 μm; n = 6. (d) RT-qPCR analysis of the fold induction of the proliferation genes c-Myc, cyclin and Ki67 in organoids treated with/without D8; n = 6. (e) Confocal images of organoid staining with Hoechst (blue) and EdU (red); scale bar, 10 μm. The mean density of EdU-positive cells in each organoid was calculated. n = 6. Data are presented as the mean ± SD. *P < .05, **P < .01. Data were combined from at least three independent experiments unless otherwise stated.

Figure 1.

L. reuteri D8 upregulated intestinal organoid proliferation under physiological conditions. (a) Coculture model of L. reuteri D8 and organoids. (b) Crypts from small intestines were seeded onto Matrigel and cultured for 3 days to obtain well-developed organoids. Scale bars, 100 μm. (c) Organoids were treated with or without L. reuteri D8 (10⁶ CFU per well) for 48 h. The surface area of organoids was calculated. Scale bars, 50 μm; n = 6. (d) RT-qPCR analysis of the fold induction of the proliferation genes c-Myc, cyclin and Ki67 in organoids treated with/without D8; n = 6. (e) Confocal images of organoid staining with Hoechst (blue) and EdU (red); scale bar, 10 μm. The mean density of EdU-positive cells in each organoid was calculated. n = 6. Data are presented as the mean ± SD. *P < .05, **P < .01. Data were combined from at least three independent experiments unless otherwise stated.

Figure 2.

L. reuteri activated the Wnt/β-catenin pathway and promoted ISC proliferation. (a) Organoids were treated with or without L. reuteri D8 (10⁶ CFU per well) for 48 h respectively. RT-qPCR analysis of the fold induction of Wnt3 and Lrp5 expression in organoids cultured with or without D8 (10⁶ CFU); n = 6. (b) Western blot analysis of active β-catenin and β-catenin expression in organoids; n = 6. (c) Organoids were co-cultured with L. reuteri D8 (10⁶ CFU) or Wnt agonist 1(10 μM) for 24 h respectively. Confocal images of organoid staining with Hoechst (blue) and active β-catenin (green). The average fluorescence intensity of active β-catenin was analyzed by Image-Pro Plus, n = 6. (d) Fold induction of active ISC markers (Lgr5, Ascl2, Olfm4) and quiescent ISC markers (Bmi1, Msi1) in organoids; n = 6. (e) Western blot results of Lgr5 protein expression in organoids; n = 6. (f) Confocal images of organoid staining with Hoechst (blue) and Lgr5 (green). The Lgr5-positive cells in each crypt were detected. N = 6. (g) The Wnt3, Lrp5 and c-Myc mRNA expression in different groups was detected by RT-qPCR, n = 6. ENR = EGF+Noggin+R-spondin, EN = EGF+Noggin. (h) r-spondin-1, r-spondin-2 and r-spondin-3 mRNA expression in organoids was detected by RT-qPCR; n = 6. (i), Organoids were stained with EdU (red). Nuclei were stained with Hoechst (blue); c. Data are presented as the mean ± SD. *P < .05, **P < .01. Data were combined from at least three independent experiments unless otherwise stated.

Figure 2.

L. reuteri activated the Wnt/β-catenin pathway and promoted ISC proliferation. (a) Organoids were treated with or without L. reuteri D8 (10⁶ CFU per well) for 48 h respectively. RT-qPCR analysis of the fold induction of Wnt3 and Lrp5 expression in organoids cultured with or without D8 (10⁶ CFU); n = 6. (b) Western blot analysis of active β-catenin and β-catenin expression in organoids; n = 6. (c) Organoids were co-cultured with L. reuteri D8 (10⁶ CFU) or Wnt agonist 1(10 μM) for 24 h respectively. Confocal images of organoid staining with Hoechst (blue) and active β-catenin (green). The average fluorescence intensity of active β-catenin was analyzed by Image-Pro Plus, n = 6. (d) Fold induction of active ISC markers (Lgr5, Ascl2, Olfm4) and quiescent ISC markers (Bmi1, Msi1) in organoids; n = 6. (e) Western blot results of Lgr5 protein expression in organoids; n = 6. (f) Confocal images of organoid staining with Hoechst (blue) and Lgr5 (green). The Lgr5-positive cells in each crypt were detected. N = 6. (g) The Wnt3, Lrp5 and c-Myc mRNA expression in different groups was detected by RT-qPCR, n = 6. ENR = EGF+Noggin+R-spondin, EN = EGF+Noggin. (h) r-spondin-1, r-spondin-2 and r-spondin-3 mRNA expression in organoids was detected by RT-qPCR; n = 6. (i), Organoids were stained with EdU (red). Nuclei were stained with Hoechst (blue); c. Data are presented as the mean ± SD. *P < .05, **P < .01. Data were combined from at least three independent experiments unless otherwise stated.

Figure 3.

The inflammatory response induced by TNF was reduced by L. reuteri. (a) Organoids were treated with TNF (60 ng/mL) for 12 h alone or cocultured with L. reuteri (10⁶ CFU) for another 36 h. Organoid morphology was assessed by light microscopy; n = 6. The number of damaged organoids per well was counted; n = 6. Scale bars, 100 μm. (b) TNF mRNA expression in organoids was detected by RT-qPCR; n = 6. (c) Annexin V-PI double staining was performed to distinguish early apoptotic cells (Annexin V⁺, PI⁻) from late apoptotic cells (Annexin V⁺, PI⁺); n = 6. (d) Confocal images of nuclear staining (blue) and EdU staining (red) in organoids. n = 6. Scale bar, 100 μm. (e) RT-qPCR analysis of the fold induction of the proliferation genes c-Myc, cyclin, Ki67 in organoids; n = 6. Data are presented as the mean ± SD. *P < .05, **P < .01. Data were combined from at least three independent experiments unless otherwise stated.

Figure 3.

The inflammatory response induced by TNF was reduced by L. reuteri. (a) Organoids were treated with TNF (60 ng/mL) for 12 h alone or cocultured with L. reuteri (10⁶ CFU) for another 36 h. Organoid morphology was assessed by light microscopy; n = 6. The number of damaged organoids per well was counted; n = 6. Scale bars, 100 μm. (b) TNF mRNA expression in organoids was detected by RT-qPCR; n = 6. (c) Annexin V-PI double staining was performed to distinguish early apoptotic cells (Annexin V⁺, PI⁻) from late apoptotic cells (Annexin V⁺, PI⁺); n = 6. (d) Confocal images of nuclear staining (blue) and EdU staining (red) in organoids. n = 6. Scale bar, 100 μm. (e) RT-qPCR analysis of the fold induction of the proliferation genes c-Myc, cyclin, Ki67 in organoids; n = 6. Data are presented as the mean ± SD. *P < .05, **P < .01. Data were combined from at least three independent experiments unless otherwise stated.

Figure 4.

L. reuteri enhanced ISC regeneration via moderately activating the Wnt/β-catenin pathway. Organoids were treated with TNF (60 ng/mL) for 12 h alone or cocultured with L. reuteri (10⁶ CFU) for another 36 h. (a) RT-qPCR analysis of the fold induction of Wnt3 and Lrp5 expression in organoids; n = 6. (b) Western blot analysis of β-catenin and active β-catenin expression in organoids; n = 6. (c) Fold induction of Lgr5, Olfm4, Ascl2, Bmi1 and Msi1; n = 6. (d) Western blot and densitometry analysis of Lgr5 expression; n = 6. (e) Confocal images of Lgr5 staining (green) and Hoechst staining (blue) in organoids. The number of Lgr5⁺ cells in each crypt was counted; n = 6. Scale bar, 100 μm. (f) Organoids were stained with lysozyme (red) and Hoechst (blue); n = 6; the number of lysozyme⁺ cells per crypt was calculated. (g) Western blot and densitometry analysis of lysozyme expression in organoids; n = 6. Data are presented as the mean ± SD. *P < .05, **P < .01. Data were combined from at least three independent experiments unless otherwise stated.

Figure 4.

L. reuteri enhanced ISC regeneration via moderately activating the Wnt/β-catenin pathway. Organoids were treated with TNF (60 ng/mL) for 12 h alone or cocultured with L. reuteri (10⁶ CFU) for another 36 h. (a) RT-qPCR analysis of the fold induction of Wnt3 and Lrp5 expression in organoids; n = 6. (b) Western blot analysis of β-catenin and active β-catenin expression in organoids; n = 6. (c) Fold induction of Lgr5, Olfm4, Ascl2, Bmi1 and Msi1; n = 6. (d) Western blot and densitometry analysis of Lgr5 expression; n = 6. (e) Confocal images of Lgr5 staining (green) and Hoechst staining (blue) in organoids. The number of Lgr5⁺ cells in each crypt was counted; n = 6. Scale bar, 100 μm. (f) Organoids were stained with lysozyme (red) and Hoechst (blue); n = 6; the number of lysozyme⁺ cells per crypt was calculated. (g) Western blot and densitometry analysis of lysozyme expression in organoids; n = 6. Data are presented as the mean ± SD. *P < .05, **P < .01. Data were combined from at least three independent experiments unless otherwise stated.

Figure 5.

L. reuteri reduced C. rodentium colonization and ameliorated intestinal inflammation in mice. (a) To verify colonization of L. reuteri D8 in the intestine, mice were orally administered L. reuteri D8 (10⁸ CFU) only once, and the number of L. reuteri D8 colonies in mouse feces on MRS plates containing tetracycline (500 μg/ml) was detected at the indicated time points. (b) L. reuteri D8 labeled with BacLight™ Bacterial Green Stain (10⁸ CFU) was administered to mice, intestinal tissue was collected 16 h post administration, and the distribution of L. reuteri in the intestine was detected by confocal microscopy; D8 (green), Hoechst (blue). (c) A schematic of the animal treatment strategy. Four-week-old mice were orally administered 200 μl of L. reuteri (10⁸ CFU), HK-L. reuteri (10⁸ CFU) and PBS for 28 days continuously, and the infection groups were orally administered 200 μl of C. rodentium (10⁹ CFU) on the 14th day. The mice were sacrificed after subjection to the different treatments on the 28th day. (d) The number of C. rodentium colonies in mouse feces was detected with MacConkey agar plates. (e) Mouse body weight changes during the course of the experiments. (f, g) Photomicrographs of mice and ileum pathology scores. Scale bars, 200 μm. (h) Segments of ileum were processed to measure the crypt depth. (i) The LPS concentration in serum samples was detected by ELISA; n = 6. (j) IL-β and TNF secretion in ileal tissue supernatant was detected by ELISA, n = 6. Data are presented as the mean ± SD. *P < .05, **P < .01. Data were combined from at least three independent experiments unless otherwise stated.

Figure 5.

L. reuteri reduced C. rodentium colonization and ameliorated intestinal inflammation in mice. (a) To verify colonization of L. reuteri D8 in the intestine, mice were orally administered L. reuteri D8 (10⁸ CFU) only once, and the number of L. reuteri D8 colonies in mouse feces on MRS plates containing tetracycline (500 μg/ml) was detected at the indicated time points. (b) L. reuteri D8 labeled with BacLight™ Bacterial Green Stain (10⁸ CFU) was administered to mice, intestinal tissue was collected 16 h post administration, and the distribution of L. reuteri in the intestine was detected by confocal microscopy; D8 (green), Hoechst (blue). (c) A schematic of the animal treatment strategy. Four-week-old mice were orally administered 200 μl of L. reuteri (10⁸ CFU), HK-L. reuteri (10⁸ CFU) and PBS for 28 days continuously, and the infection groups were orally administered 200 μl of C. rodentium (10⁹ CFU) on the 14th day. The mice were sacrificed after subjection to the different treatments on the 28th day. (d) The number of C. rodentium colonies in mouse feces was detected with MacConkey agar plates. (e) Mouse body weight changes during the course of the experiments. (f, g) Photomicrographs of mice and ileum pathology scores. Scale bars, 200 μm. (h) Segments of ileum were processed to measure the crypt depth. (i) The LPS concentration in serum samples was detected by ELISA; n = 6. (j) IL-β and TNF secretion in ileal tissue supernatant was detected by ELISA, n = 6. Data are presented as the mean ± SD. *P < .05, **P < .01. Data were combined from at least three independent experiments unless otherwise stated.

Figure 6.

L. reuteri maintained normal levels of lgr5⁺ ISCs and lysozyme⁺ Paneth cells to moderately stimulate epithelial proliferation after C. rodentium infection. (a) RT-qPCR analysis of the fold induction of Wnt3, Lrp5 and Lgr5 expression; n = 6. (b) c-Myc mRNA expression was detected by RT-qPCR, n = 6. (c) Confocal images of ileal tissues stained with PCNA (red) and Hoechst (blue). The mean density of PCNA⁺ cells per crypt was detected. Scale bars, 200 μm. (d) Confocal images (Lgr5 staining, green; Hoechst staining, blue) of ileum. The mean density of Lgr5-positive cells per crypt was detected. Scale bars, 200 μm. (e) Confocal images of lysozyme staining (red) and Hoechst staining (blue) in the ileum. The mean density of lysozyme+ cells per crypt was detected. (f) Fold induction of Defa1, Defa6, and lysozyme expression; n = 6. Data are presented as the mean ± SD. *P < .05, **P < .01. Data were combined from at least three independent experiments unless otherwise stated.

Figure 6.

L. reuteri maintained normal levels of lgr5⁺ ISCs and lysozyme⁺ Paneth cells to moderately stimulate epithelial proliferation after C. rodentium infection. (a) RT-qPCR analysis of the fold induction of Wnt3, Lrp5 and Lgr5 expression; n = 6. (b) c-Myc mRNA expression was detected by RT-qPCR, n = 6. (c) Confocal images of ileal tissues stained with PCNA (red) and Hoechst (blue). The mean density of PCNA⁺ cells per crypt was detected. Scale bars, 200 μm. (d) Confocal images (Lgr5 staining, green; Hoechst staining, blue) of ileum. The mean density of Lgr5-positive cells per crypt was detected. Scale bars, 200 μm. (e) Confocal images of lysozyme staining (red) and Hoechst staining (blue) in the ileum. The mean density of lysozyme+ cells per crypt was detected. (f) Fold induction of Defa1, Defa6, and lysozyme expression; n = 6. Data are presented as the mean ± SD. *P < .05, **P < .01. Data were combined from at least three independent experiments unless otherwise stated.

Figure 7.

Confirmation of the protective effect of L. reuteri against C. rodentium infection in mice and intestinal organoid. (a) A schematic of the animal treatment strategy. The mice were sacrificed after subjection to the different treatments on the 28th day. (b) Mouse body weight changes during the course of the experiments. (c) Photomicrographs of mice and ileum pathology scores. Scale bars, 200 μm. (d) The number of C. rodentium colonies in mouse feces was detected with MacConkey agar plates. (e) Confocal images of ileal tissues stained with PCNA (green) and Hoechst (blue). The mean density of PCNA⁺ cells per crypt was detected. Scale bars, 200 μm. (f–g) RT-qPCR analysis of the fold induction of Wnt3, Lgr5 and R-spondin1 expression; n = 6. (h) Confocal images of lysozyme staining (red) and Hoechst staining (blue) in the ileum. The mean density of lysozyme+ cells per crypt was detected. (i–j) RT-qPCR analysis of the fold induction of Lyz-1, Defa1 and Defa6 expression; n = 6. (k) IL-β and TNF secretion in ileal tissue supernatant was detected by ELISA, n = 6. (l) The LPS concentration in serum samples was detected by ELISA; n = 6. Data are presented as the mean ± SD. *P < .05, **P < .01. Data were combined from at least three independent experiments unless otherwise stated.

Figure 7.

Confirmation of the protective effect of L. reuteri against C. rodentium infection in mice and intestinal organoid. (a) A schematic of the animal treatment strategy. The mice were sacrificed after subjection to the different treatments on the 28th day. (b) Mouse body weight changes during the course of the experiments. (c) Photomicrographs of mice and ileum pathology scores. Scale bars, 200 μm. (d) The number of C. rodentium colonies in mouse feces was detected with MacConkey agar plates. (e) Confocal images of ileal tissues stained with PCNA (green) and Hoechst (blue). The mean density of PCNA⁺ cells per crypt was detected. Scale bars, 200 μm. (f–g) RT-qPCR analysis of the fold induction of Wnt3, Lgr5 and R-spondin1 expression; n = 6. (h) Confocal images of lysozyme staining (red) and Hoechst staining (blue) in the ileum. The mean density of lysozyme+ cells per crypt was detected. (i–j) RT-qPCR analysis of the fold induction of Lyz-1, Defa1 and Defa6 expression; n = 6. (k) IL-β and TNF secretion in ileal tissue supernatant was detected by ELISA, n = 6. (l) The LPS concentration in serum samples was detected by ELISA; n = 6. Data are presented as the mean ± SD. *P < .05, **P < .01. Data were combined from at least three independent experiments unless otherwise stated.

Figure 8.

Confirmation of the protective effect of L. reuteri against C. rodentium infection in mice and intestinal organoids. (a) Organoids were treated with C. rodentium (10⁶ CFU) alone or cocultured with L. reuteri (10⁶ CFU) and Wnt-C59 (100 nM). Organoid morphology was assessed by light microscopy; n = 6. Scale bars, 100 μm. (b) The number of C. rodentium colonies per well was detected with MacConkey agar plates. (c–e) RT-qPCR analysis of the fold induction of Wnt3, Lgr5, R-spondin1 and Lyz-1 expression; n = 6. (f) Organoids were stained with lysozyme (red) and Hoechst (blue); n = 6; the number of lysozyme⁺ cells per crypt was calculated. (g–h) RT-qPCR analysis of the fold induction of Defa1, Defa6, IL-1β and TNF expression; n = 6. Data are presented as the mean ± SD. *P < .05, **P < .01. Data were combined from at least three independent experiments unless otherwise stated.

Figure 8.

Confirmation of the protective effect of L. reuteri against C. rodentium infection in mice and intestinal organoids. (a) Organoids were treated with C. rodentium (10⁶ CFU) alone or cocultured with L. reuteri (10⁶ CFU) and Wnt-C59 (100 nM). Organoid morphology was assessed by light microscopy; n = 6. Scale bars, 100 μm. (b) The number of C. rodentium colonies per well was detected with MacConkey agar plates. (c–e) RT-qPCR analysis of the fold induction of Wnt3, Lgr5, R-spondin1 and Lyz-1 expression; n = 6. (f) Organoids were stained with lysozyme (red) and Hoechst (blue); n = 6; the number of lysozyme⁺ cells per crypt was calculated. (g–h) RT-qPCR analysis of the fold induction of Defa1, Defa6, IL-1β and TNF expression; n = 6. Data are presented as the mean ± SD. *P < .05, **P < .01. Data were combined from at least three independent experiments unless otherwise stated.