Protective function of interleukin 27 in colitis-associated cancer via suppression of inflammatory cytokines in intestinal epithelial cells

Abstract

Numerous studies have demonstrated that inflammation contributes to a variety of cancer formation, among them, colitis-associated cancer (CAC) represents a typical inflammation-related cancer. Interleukin 27 (IL-27) has been demonstrated to play an important role in inflammation-related disease. The effect of IL-27 in intestinal inflammation is controversial and its role in CAC is not elucidated yet. In our present study, we found that IL-27 has protective function in murine model of CAC through suppression of inflammatory cytokines in intestinal epithelial cells (IECs). IL-27Rα (WSX-1) deficiency promotes the CAC development in mice, which is driven by enhanced tumor cell proliferation, more intensive myeloid-derived suppressor cells (MDSC) accumulation in colon lamina propria and higher level of inflammatory cytokines and chemokines in IECs. The levels of IL-6, TNF-α, GM-CSF and CXCL1 triggered in vitro by toll-like receptor ligands are significantly upregulated in IECs from WSX-1 KO mice. Removal of commensal microorganism through antibiotic treatment in mice to eliminate TLR ligands deprives the protective function of IL-27 on CAC tumor growth. Thus, IL-27 suppresses CAC formation through an anti-inflammation mechanism targeting IECs and in turn resists the tumorigenesis. Hence, our study explained how IL-27 exerts its anti-inflammatory function on epithelial cells to fight against chronic-inflammation-associated cancer, which might provide new insights on the potential therapeutic strategies for cancer.

Keywords: CXCL1; Colitis-associated cancer; inflammation; interleukin 27; intestinal epithelial cells.

Figure 1.

Both IL-27 and its receptors are upregulated in tissue of AOM/DSS-induced colitis-associated cancer. (A) Schematic representation of the AOM/DSS protocol. (B) (qRT-PCR) and (C) (semi-quantification) Determination of p28, EBI3 and WSX-1 mRNAs in the normal tissues, tumor adjacent tissues and CAC tumor tissues by RT-PCR (n = 3). (D) Immunofluorescence of CAC tumor tissue sections stained with anti-IL-27 and DAPI. (E) Western blot analysis of IL-27 level in the intestinal epithelial cells (IECs) and colon lamina propria (LP) in both control and CAC model mice (n> = 3). Data are mean ± SEM. *p < 0.05; **p < 0.01.

Figure 2.

IL-27/WSX-1 signaling is important for inflammatory colorectal cancer formation and development in mice. (A) Macroscopic view of the representative colons and spleens from WT or WSX-1 KO mice on day 100 of the CAC model. (B–D) Colon tumor load, tumor number and average tumor size from WT (n = 13) and WSX-1 KO (n = 13) mice on day 100 of the CAC model. (E) The size distribution of tumors in WT and WSX-1 KO mice with CAC. (F) Ki67 staining of the representative mouse colons from WT or WSX-1 KO mice on day 100 of the CAC model. (G) TUNEL staining of the representative mouse colons from WT or WSX-1 KO mice on day 100 of the CAC model. (H) mRNA quantification of IL-6 and TNF-α in the distal colon tumor from CAC models. (I) Macroscopic view of the representative colons from WT or WSX-1 KO mice on day 8 of the DSS colitis model. (J) Statistical analysis of colon length of WT and WSX-1 KO mice on day 8 of DSS colitis model. (K) Percentage of weight loss in DSS-induced colitis model. Body weight was measured every day during DSS administration. Data are mean ± SEM. *p < 0.05; **p < 0.01.

Figure 3.

WSX-1 deficiency results in increased MDSC accumulation in colon lamina propria. (A and B) Statistical results of immune cells subsets by FACS analysis in lamina propria (n > 3). (C) Representative FACS results of MDSCs in lamina propria, and statistical results of MDSCs in lamina propria (n > 3). (D) Representative FACS results of CXCR2 expression on MDSCs, and statistical results of CXCR2 expression on MDSCs (n > 3). Data are mean ± SEM. *p < 0.05; **p < 0.01.

Figure 4.

WSX-1 deficiency leads to higher level of inflammatory cytokines and chemokines in IECs. (A) FACS analysis of WSX-1 and gp130 expression gated on EpCAM⁺ intestinal epithelial cells. (B) Quantification of the mRNAs of IL-6, GM-CSF and TNF-α in the IECs of naive and CAC models (n> = 3). (C) Quantification of CXCL1 mRNA in the IECs of CAC models (n> = 3). (D) Quantification of mRNAs of iNOS, COX-2, IL-1β, CXCL2, CXCL5 in the IECs from CAC models by qRT-PCR (n> = 3). (E) Colon tumor sections from WT or WSX-1 KO CAC mice, stained with anti-CXCL1, anti-Gr-1 or DAPI. Data are mean ± SEM. *p < 0.05; **p < 0.01.

Figure 5.

IL-6, GM-CSF and CXCL1 mediated the accumulation of MDSCs in vitro. (A) Flow cytometry analysis of cells in the lower chamber of transwell migration assay, the cell number in the Gate is counted. (B) The cells attached to the bottom side of polycarbonate membrane were stained with DAPI and analyzed. (C) Flow cytometry analysis of MDSCs that cultured in 50% LPS-stimulated CT26 supernatant in the presence of anti-IL-6 or anti-GM-CSF or control IgG. Data are mean ± SEM. *p < 0.05; **p < 0.01.

Figure 6.

The alteration of inflammatory cytokines in WSX-1 KO mice with CAC is not the consequence of severer epithelium barrier disruption. (A) Quantification of the mRNAs of tight junction-related genes (occludin, claudin-1,-2,-3) in the IECs from entire colon tissues of CAC by qRT-PCR (n> = 3). (B) Sections from colon tumors and tumor adjacent tissues in WT or WSX-1 KO mice, stained with anti-ZO-1, anti-cytokeratin or DAPI. (C) Quantification of LPS in the serum from portal vein. Data are mean ± SEM. *p < 0.05; **p < 0.01.

Figure 7.

IL-27 inhibits the LPS-induced cytokine expression in IECS in vitro. (A) WT and WSX-1 KO IECs were treated with different doses of LPS for 6 h. mRNA levels of IL-6, GM-CSF and TNF-α were determined by qRT-PCR. (B) WT IECs were treated with IL-27 (50 ng/mL) and LPS (50 ng/mL) for 6 h. Quantification of the mRNA levels of CXCL1, GM-CSF and TNF-α by qRT-PCR was showed. (C) CT26 was treated by LPS (100 ng/mL) with or without IL-27 (50 ng/mL) for 6 h. Quantification of mRNA levels of IL-6, GM-CSF, TNF-α and CXCL1 was showed. (D) CT26 was stimulated with LPS (100 ng/mL) or IL-27 (50 ng/mL) for indicated time and level of total and phosphorylated p65 was determined by Western Blot. (E) CT26 was stimulated with LPS (100 ng/mL) or IL-27 (50 ng/mL) for indicated time and level of SOCS1, SHP2, total and phosphorylated STAT3 was determined by Western Blot. Data are mean ± SEM. *p < 0.05; **p < 0.01.

Figure 8.

The microbiota removal ameliorates the CAC development in WSX-1 knockout mice. (A) Determination of phosphrylation of p65 in IECs and lamina propria CD45⁺ cells of antibiotic-treated and control CAC models. (B) Colon tumor load, tumor number and average tumor size from antibiotic-treated WT (n = 6) and WSX-1 KO (n = 5) mice on day 100 of the CAC model. (C) Representative FACS results of MDSCs in lamina propria and statistical results of MDSCs in lamina propria of antibiotic-treated CAC murine model. (n> = 5). (D) Quantification of IL-6, GM-CSF, TNF-α and CXCL1 mRNAs in the IECs of antibiotic-treated CAC model mice (n> = 5). Data are mean ± SEM. *p < 0.05; **p < 0.01.