WNT4 promotes the symmetric fission of crypt in radiation-induced intestinal epithelial regeneration

Abstract

Background: Radiotherapy for pelvic malignant tumors inevitably causes intestinal tissue damage. The regeneration of intestinal epithelium after radiation injury relies mainly on crypt fission. However, little is known about the regulatory mechanisms of crypt fission events.

Methods: The effects of WNT4 on crypt regeneration and the symmetry of crypt fission were examined using a mouse small intestinal organoid culture model. Three-dimensional (3D) reconstructed images of organoids were applied to assess the symmetry of crypt fission and Paneth cell localization upon manipulation of WNT4 expression. The effect of WNT4 on the expression of β-catenin target genes was analyzed by real-time quantitative polymerase chain reaction (RT-qPCR). The in vivo effect of WNT4 overexpression mediated by adeno-associated virus (AAV) on symmetric fission of crypt was investigated using a radiation-injured mouse model.

Results: WNT4 has a special function of promoting symmetric fission of small intestinal crypts, although it inhibits budding, stemness, and cell proliferation on organoids. WNT4 promotes the correct localization of Paneth cells in the crypt base by regulating the expression of EphB3, thereby promoting the symmetric fission of small intestinal crypts. WNT4 negatively regulates the canonical WNT/β-catenin signaling pathway, and it promotes symmetric crypt fission in a ROR2 receptor-dependent manner. Moreover, in patients and animal models of radiation-induced intestinal injury, we found that the regenerated crypts are irregular in size and shape, Paneth cells are mislocalized, and the expression of WNT4 is decreased while EphB3 is increased. Importantly, restoration of WNT4 expression mediated by AAV effectively promotes symmetric crypt fission and thus improves the regularity of regenerating crypts in mice with radiation-induced injury.

Conclusions: Our study highlights the critical role of WNT4 in the regulation of crypt fission and provides WNT4 as a potential therapeutic target for radiation enteritis.

Keywords: Crypt fission; Epithelial regeneration; Paneth cell; WNT4.

Fig. 1

WNT4 suppresses intestinal organoid branching. A RT-qPCR analysis of the interference efficiency of Wnt4 by siRNA in mouse small intestinal organoids. B Representative images (left) and budding numbers (right) of mouse small intestinal organoids transfected with either control siRNA (si-control) or Wnt4 siRNA (si-Wnt4) after 60 h. Scale bars, 200 μm. C RT-qPCR analysis of the overexpression efficiency of Wnt4 in mouse small intestinal organoids. D Representative images (left) and budding numbers (right) of mouse small intestinal organoids transfected with either vector control (vector) or Wnt4-overexpressing construct (Wnt4-OE) after 60 h. Scale bars, 200 μm. E Representative images of IF staining for Ki67 in mouse small intestinal organoids transfected with either control siRNA or Wnt4 siRNA. Scale bars, 50 μm. The percentages of Ki67-positive cells in the bud and non-bud areas of each organoid (n = 17 for si-control and n = 21 for si-Wnt4) were quantified separately. F Representative images of IF staining for Ki67 in mouse small intestinal organoids transfected with either vector control or Wnt4-overexpressing construct. Scale bars, 50 μm. The percentages of Ki67-positive cells in the bud and non-bud areas of each organoid (n = 16 for both Vector and Wnt4-OE) were quantified separately. G RT-qPCR analysis of the mRNA levels of stem cell marker genes (Lgr5, Ascl2 and Olfm4) in mouse small intestinal organoids after knockdown or overexpression of Wnt4. H RT-qPCR analysis of the mRNA levels of cell differentiation markers (Lyz, Muc2, Chga, Fabp1 and Fabp2) in mouse small intestinal organoids after knockdown or overexpression of Wnt4. All values are means ± SD. Statistical analyses were performed by unpaired Student’s t-test (A–H). ns, not significant, *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001

Fig. 2

WNT4 promotes symmetric fission of crypt in organoids. A, B Representative time-lapse images of mouse small intestinal organoids transfected with si-control, si-Wnt4, Vector or Wnt4-OE, or treated with an inhibitory Eph fragment (Eph, 10 μg/ml) at the indicated time points. Scale bars, 100 μm. Red dashed boxes indicate symmetric fissions, and yellow dashed boxes indicate asymmetric fissions. C, D Representative 3D projections of organoids stained against F-actin (Phalloidin, red) and nuclei (Hoechst, blue) after transfection with si-control, si-Wnt4, Vector or Wnt4-OE, or treated with an inhibitory Eph fragment (Eph, 10 μg/ml). Scale bars, 50 μm. E, F The symmetry ratio was calculated by dividing the length of the short daughter crypt by the length of the long daughter crypt based on the 3D projections of organoids in C and D. n = 24 for si-control, n = 29 for si-Wnt4, n = 29 for Eph, n = 27 for vector, and n = 20 for Wnt4-OE. All values are means ± SD. Statistical analyses were performed by one-way ANOVA with Tukey’s multiple comparisons test (E) or unpaired Student’s t-test (F). **p < 0.01, ****p < 0.0001

Fig. 3

WNT4 promotes the correct localization of Paneth cells in the crypt. A, B 3D projections of mouse small intestinal organoids stained to visualize Paneth cells (lysozyme, green), F-actin (Phalloidin, red), and nuclei (Hoechst, blue). Organoids were transfected with si-control or si-Wnt4, or treated with an inhibitory Eph fragment (Eph, 10 μg/ml) in A, or transfected with Vector or Wnt4-OE in B. Scale bars, 25 μm. White arrows indicate Paneth cells positioned in the crypt base and yellow arrows indicate Paneth cells positioned away from the crypt base. C, D The frequency of Paneth cells distanced from the crypt base was quantified on the basis of the 3D projections of organoids in A and B. The tip of the organoid bud was identified as the crypt base. n = 79 Paneth cells for si-control, n = 98 Paneth cells for si-Wnt4, n = 101 Paneth cells for Eph, n = 54 Paneth cells for Vector, and n = 52 Paneth cells for Wnt4-OE

Fig. 4

WNT4 negatively regulates WNT/β-catenin signaling, and relies on ROR2 receptor to promote symmetric crypt fission. A RT-qPCR analysis of the expression changes of WNT/β-catenin target genes (Axin2, Myc, Ccnd1 and Ephb3) in mouse small intestinal organoids after knockdown (left) or overexpression (right) of Wnt4. B Colocalized expression of ROR2 (green) and WNT4 (red) in both normal mouse small intestinal tissue and organoid was detected by IF staining. Boxed regions are magnified and shown in the right column. Nuclei were counterstained with DAPI (blue). White arrows indicate colocalization. Scale bars, 100 μm (columns 1–4), 50 μm (column 5). C Lysate from the whole mouse small intestine was immunoprecipitated with anti-ROR2 or anti-IgG and analyzed by western blot with anti-WNT4 and anti-ROR2 antibodies. Yellow asterisk indicates the WNT4 band. D The symmetry ratio was calculated by dividing the length of the short daughter crypt by the length of the long daughter crypt on the basis of the 3D projections of mouse small intestinal organoids with overexpression of Wnt4 combined with or without knockdown of Ror2 (n = 20 for vector, n = 22 for Wnt4-OE, n = 24 for Wnt4-OE + si-Ctr, n = 21 for Wnt4-OE + si-Ror2). E The symmetry ratio was calculated by dividing the length of the short daughter crypt by the length of the long daughter crypt on the basis of the 3D projections of mouse small intestinal organoids with either knockdown of Wnt4 or knockdown of Ror2 (n = 19 for si-control, n = 21 for si-Wnt4, n = 20 for si-Ror2). All values are means ± SD. Statistical analyses were performed by unpaired Student’s t-test between two groups (A), and one-way ANOVA with Tukey’s multiple comparisons test (D, E). ns, not significant, *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001

Fig. 5

WNT4 expression is decreased in the radiation-injured human and mouse intestinal tissues. A HE staining of the small intestines of normal control mouse (Normal) and mouse with 10.5 Gy of abdominal irradiation after 4 weeks (IR). Scale bars, 100 μm. B HE staining of the radiation-injured lesion (L) and paired adjacent non-injured intestinal tissue (N) in RE patient. Scale bars, 100 μm. C IF staining was performed using anti-lysozyme (labeled Paneth cells, red) and anti-EphB3 (green) antibodies in the small intestinal tissues of normal and IR mice. Nuclei were counterstained with DAPI (blue). Scale bars, 100 μm. D ISH analysis of Wnt4 expression in the small intestinal tissues of normal and IR mice. The crypt regions (red boxes) were magnified in the middle column, and the villous regions (black boxes) were magnified in the right column. Scale bars, 200 μm (left column) and 50 μm (middle and right columns). E The intensity of Wnt4 expression in the crypts and villus mesenchyme in the normal (n = 6) and IR (n = 6) mice was quantified using ImageJ software based on the ISH analysis in D. F RT-qPCR analysis of WNT4 expression in seven clinical samples of RE lesions (L) and paired adjacent non-injured intestinal tissues (N). G Correlation analysis between the relative WNT4 mRNA level and EphB3 mRNA level in the intestinal lesions of patients with RE (n = 18). Data are means ± SD. Statistical analysis was performed by unpaired Student’s t-test (E), paired Student’s t-test (F), and Pearson correlation coefficient test (G). ns, not significant, *p < 0.05, **p < 0.01

Fig. 6

AAV-mediated WNT4 supplementation inhibits EphB3 expression in small intestinal crypts in mice with radiation injury. The mice received 10.5 Gy of abdominal irradiation or sham-irradiation as control. The next day, AAV-vector or AAV-Wnt4 was intraperitoneally injected into the irradiated mice. At 4 weeks after irradiation, the small intestinal tissues were collected and analyzed. A The overexpression of WNT4 in the small intestines of irradiated mice treated with AAV-Wnt4 was confirmed by IF staining with anti-Flag antibody (green). Nuclei were counterstained with DAPI (blue). Scale bars, 100 μm. Dashed lines indicate crypt area. a: apical; b: basal. B RT-qPCR analysis of Wnt4 expression in the entire small intestinal tissues of the irradiated mice treated with AAV-vector (n = 5) or AAV-Wnt4 (n = 5) or the sham-irradiated control mice (n = 3). C IF staining was performed using anti-lysozyme (labeled Paneth cells, red) and anti-EphB3 (green) antibodies in the small intestinal tissues of the three groups of mice. Nuclei were counterstained with DAPI (blue). Scale bars, 100 μm. Dashed lines indicate crypt area. a: apical; b: basal. D The intensity of EphB3 expression and the height of the EphB3-positive area were quantified using ImageJ on the basis of the IF staining of EphB3 in C. E The frequency of Paneth cells distanced from the crypt base was quantified using ImageJ on the basis of the IF staining of lysozyme in C. Values are means ± SD. Statistical analyses were performed by one-way ANOVA with Tukey’s multiple comparisons test (B, D). *p < 0.05, **p < 0.01, ***p < 0.001

Fig. 7

AAV-mediated WNT4 supplementation reduces asymmetric fission of small intestinal crypts in mice with radiation injury. A Representative images of HE staining (left) of the small intestinal tissues of the irradiated mice treated with AAV-vector (n = 5) or AAV-Wnt4 (n = 5) or the sham-irradiated control mice (n = 3). The proportions of symmetric and asymmetric crypt fissions in the small intestinal tissues of mice in the three groups were calculated (right). The black and red arrow heads in the sham-irradiated control group indicate crypt not undergoing fission and crypt undergoing symmetric fission, respectively, which are magnified in the next row (left). The yellow arrows in the irradiated AAV-vector group indicate crypts undergoing asymmetric fissions, which are magnified in the next row (middle). The red arrows in the irradiated AAV-Wnt4 group indicate crypts undergoing symmetric fissions, which are magnified in the next row (right). Scale bars, 50 μm (upper row) and 25 μm (lower row). B Representative images of lysozyme and Ki67 staining of the small intestinal tissues of the irradiated mice treated with AAV-vector (n = 5) or AAV-Wnt4 (n = 5) or the sham-irradiated control mice (n = 3). Scale bars, 50 μm. The numbers of mislocated Paneth cells per 20 × field and Ki67⁺ cells per crypt were quantified. All values are means ± SD. Statistical analyses were performed by two-way ANOVA (A) or one-way ANOVA (B) with Tukey’s multiple comparisons test. ns, not significant. *p < 0.05, **p < 0.01, ****p < 0.0001