Epithelial-derived IL-33 promotes intestinal tumorigenesis in Apc Min/+ mice

Abstract

Increased expression of Interleukin (IL)-33 has been detected in intestinal samples of patients with ulcerative colitis, a condition associated with increased risk for colon cancer, but its role in the development of colorectal cancer has yet to be fully examined. Here, we investigated the role of epithelial expressed IL-33 during development of intestinal tumors. IL-33 expression was detected in epithelial cells in colorectal cancer specimens and in the Apc ^Min/+ mice. To better understand the role of epithelial-derived IL-33 in the intestinal tumorigenesis, we generated transgenic mice expressing IL-33 in intestinal epithelial cells (V33 mice). V33 Apc ^Min/+ mice, resulting from the cross of V33 with Apc ^Min/+ mice, had increased intestinal tumor burden compared with littermate Apc ^Min/+ mice. Consistently, Apc ^Min/+ mice deficient for IL-33 receptor (ST2), had reduced polyp burden. Mechanistically, overexpression of IL-33 promoted expansion of ST2⁺ regulatory T cells, increased Th2 cytokine milieu, and induced alternatively activated macrophages in the gut. IL-33 promoted marked changes in the expression of antimicrobial peptides, and antibiotic treatment of V33 Apc ^Min/+ mice abrogated the tumor promoting-effects of IL-33 in the colon. In conclusion, elevated IL-33 signaling increases tumor development in the Apc ^Min/+ mice.

Figure 1

IL-33 expression in human colorectal cancer patients and in Apc ^Min/+ mice model. (A) Paraffin sections from a tissue microarray of human colorectal cancer (CRC) patients were stained with IL-33 (red) and Pan keratin (green). Representative photographs show expression of IL-33 by intestinal epithelial cells (IECs) in poorly differentiated adenocarcinomas. Scale bars, 100 μm. (B) The percentage of IL-33 immunolabeling in the IECs from 21 well-differentiated CRC patients (W) and 68 moderate- and poor-differentiated CRC patients (M/P). *P < 0.05, Fisher exact test. (C) Relative expression levels of IL-33 mRNA were analyzed by qPCR in the gut of Apc ^Min/+ mice at various ages. Data were normalized to the expression levels of the Ubiquitin transcript. Means ± s.e.m., n = 4 per group. **P < 0.01, nonparametric Mann-Whitney test. (D) Immunofluorescence staining for IL-33 (red) and Pan keratin (green) in Apc ^Min/+ mice at different ages. Zoomed-in boxed area shows IL-33 immunolabeling in the IECs in Apc ^Min/+ mice. Scale bars, 50 μm.

Figure 2

Generation of transgenic mice expressing IL-33 in the gut epithelium. (A) Scheme for generation of V33 mice. A transgene encoding IL-33 mature form (m-IL-33) under the control of the murine villin promoter (9 kb) was used to generate V33 mice. (B) Relative expression levels of transgenic IL-33 mRNA were analyzed by qPCR in the small intestine (SI) and large intestine (LI) of wild-type (WT) and V33 mice. Data were normalized to the expression levels of the Ubiquitin transcript. Means ± s.e.m., n = 6 per group. ***P < 0.001, one-way ANOVA. (C) Enzyme linked immunosorbent assay of IL-33 in the gut explants from WT and V33 mice. Data were normalized to the weight of the intestine explant. Means ± s.e.m., n = 4 per group. **P < 0.01, ***P < 0.001, one-way ANOVA. (D) Immunofluorescence staining for IL-33 (red) in the gut of WT and V33 mice. Cell nuclei were counterstained with DAPI (blue). Notice that transgenic expression of IL-33 in the cytoplasm of intestinal epithelial cells in V33 mice. Scale bars, 50 μm.

Figure 3

IL-33 promotes intestinal polyp growth in Apc ^Min/+ mice. (A) V33 Apc ^Min/+ mice carried a significantly higher number of polyps in the SI than Apc ^Min/+ mice at d120. (B) Representative hematoxylin and eosin (H&E) staining of ileum of Apc ^Min/+ mice and V33 Apc ^Min/+ mice at d120. Individual tumors are highlighted by black lines. (C) Tumor number (left) and tumor size (right) of LI in Apc ^Min/+ mice and V33 Apc ^Min/+ mice at d120. (D) Representative gross specimens from the indicated genotype analyzed at d120. (E) Representative H&E staining of LI of Apc ^Min/+ mice and V33 Apc ^Min/+ mice at d120. Individual tumors are highlighted by black lines. Graphs show means ± s.e.m. *P < 0.05, **P < 0.01, ***P < 0.001, nonparametric Mann-Whitney test.

Figure 4

Genetic ablation of ST2 reduces polyp burden in Apc ^Min/+ mice. (A) Scheme for generation of ST2 ^−/− mice by CRISPR/Cas9n technology. The two g-RNA targeting sequences are underlined in red and blue. Sequencing confirmed the deletion of exons 4 and 5. (B) Flow cytometric analysis of the expression of ST2 on ILC2 (CD45⁺ Lin⁻ IL7R⁺ KLRG1⁺) isolated from the mesenteric fat of WT and ST2 ^−/− mice. Notice the absence of ST2 expression in the ILC2 isolated from ST2 ^−/− mice. (C,D) Quantification of the number of tumors in the small (C) and large (D) intestine of Apc ^Min/+ ST2 ^+/+, Apc ^Min/+ ST2 ^+/− and Apc ^Min/+ ST2 ^−/− mice. Graphs show means ± s.e.m. *P < 0.05, **P < 0.01, N.S. means not significant, nonparametric Mann-Whitney test.

Figure 5

Epithelial-derived IL-33 signaling expands colonic ST2⁺ regulatory T (Treg) cells and induces a Th2 cytokine profile. (A) Relative and absolute number of Treg cells (CD4⁺Foxp3⁺) in the colon of Apc ^Min/+ mice and V33 Apc ^Min/+ mice at d120. Left, representative flow cytometry plots gated on CD45⁺ cells; right, statistical data show means ± s.e.m. (B) Relative and absolute number of ST2⁺ Treg cells in the colon of Apc ^Min/+ mice and V33 Apc ^Min/+ mice at d120. Left, representative flow cytometry plots gated on CD45⁺ CD4⁺ cells; right, statistical data show means ± s.e.m. (C) Relative expression levels of ST2⁺Treg and Th2 signature genes were analyzed by qPCR in sorted T cells of colonic lamina propria of WT, V33, Apc ^Min/+ mice and V33 Apc ^Min/+ mice at d120. Graphs show means ± s.e.m., n = 4–5/group. *P < 0.05, **P < 0.01, ***P < 0.001, one-way ANOVA.

Figure 6

IL-33 regulates alternative activation of colonic macrophages. (A) Relative number of macrophages (CD45⁺CD11b⁺F480⁺) in the colon of WT, V33, Apc ^Min/+ mice and V33 Apc ^Min/+ mice at d120. Graphs show means ± s.e.m., n = 3–5/group. (B) Relative expression levels of M1 and M2 macrophage signature genes were analyzed by qPCR in sorted macrophages of colonic lamina propria of WT mice and V33 mice at d120. (C) Relative expression levels of M2 macrophage signature genes were analyzed by qPCR in sorted macrophages of colonic lamina propria of WT mice, Apc ^Min/+ mice and V33 Apc ^Min/+ mice at d120. Data were normalized to fold change from age-matched WT littermate control mice. Graphs show means ± s.e.m., n = 3–5/group. *P < 0.05, **P < 0.01, ***P < 0.001, versus WT group. ^# P < 0.05, ^## P < 0.01, versus Apc ^Min/+ group.

Figure 7

Epithelial-derived IL-33 alters antimicrobial genes expression in the colon. (A) Relative expression levels of Reg3b, Reg3g, Ang4, Retnlb and Muc2 were analyzed by qPCR in the colon of size-matched tumor area (T) and adjacent non-tumor normal area (NT) of Apc ^Min/+ mice and V33 Apc ^Min/+ mice at d120. Data were normalized to the expression levels of the Ubiquitin transcript. Graphs show means ± s.e.m. n = 4–5/group. *P < 0.05, **P < 0.01, ***P < 0.001, one-way ANOVA. (B) Number of polyps in the intestine of Apc ^Min/+ mice and V33 Apc ^Min/+ mice receiving water or antibiotic continuously from weaning to 120 days post-birth. Graphs show means ± s.e.m. *P < 0.05, **P < 0.01, ***P < 0.001, nonparametric Mann-Whitney test.