ILC2-derived IL-9 inhibit s colorectal cancer progression by activating CD8+ T cells

Background Type 2 innate lymphoid cells (ILC2s), characterized by secreting type 2 cytokines, regulate multiple immune responses. ILC2s are found in different tumor tissues and ILC2-derived interleukin (IL)-4, IL-5, and IL-13 act on the cells in tumor microenvironment to participate in tumor progression. ILC2s are abundant in colorectal cancer (CRC) tissue, but the role of ILC2s in CRC remains unclear. ILC2s were sorted from the spleen using microbeads combined with flow cytometry and tumor infiltrating CD8 + T cells were isolated from tumor tissue by microbeads. Flow cytometry and immunofluorescence were used to detect the percentage of ILC2s and CD8 + T cells in the spleen and CRC tissue. Effects of IL-9 and IL-9-stimulated CD8 + T cells on CT26 cells were measured by proliferation, apoptosis, and migration assays in vitro. GEPIA was used to detect the ILC2s chemokines in CRC tissue and adjacent normal tissue. We found that In vitro experiments showed that activate CD8 + the of CT26 ILC2s were the main IL-9-secreting as shown by flow analysis. vivo experiments while tumor inhibition occurred by IL-9. 0.2 mg/ml hyaluronidase and 0.015 mg/ml DNase I USA) 37 °C for h. The samples were then filtered through a 70 µm strainer before staining with fluorescence antibodies. Mouse Percy5-CD45, FITC-lineage, PE-CD90.2, and APC-KLRG1 (Biolegend) were used to analyze the percentage of ILC2s in the mouse tumor tissue and human Percy5-CD45, FITC-lineage, PE-CRTH2, and APC-CD127 (Biolegend) were used to analyze the percentage of ILC2s in the human tumor tissue. Mouse CD8 cells were stained with Percy5-CD45, FITC-CD3, and PE-CD8 (Biolegend). For analyzing the percentage of MDSCs, FITC-CD11b and APC-Gr-1 (Biolegend) were used. All surface

The apoptosis kit FITC-Annexin V and APC-7-AAD (Multisciences, Hangzhou, China) was used following the manufacturer's protocols to analyze the apoptosis of CT26 cells.

Immunofluorescence analysis
The frozen sections were incubated at room temperature for 1 hour and treated with 4% paraformaldehyde for 20 min. Sections were then washed three times with 1 × PBS and incubated with primary antibodies anti-lineage and anti-CRTH2 at a dilution of 1:200 in blocking buffer (PBS with 5% BSA, 1% saponin and 1% Triton 100) at 4 °C overnight. The sections were then washed three times in 1 × PBS and incubated with DAPI (1:1000; eBioscience) for 30 min, followed by three more washes in 1 × PBS. Neutral resin was used to seal the sections, which were left to air dry. Sections were then visualized directly with a confocal microscope.
Enzyme-linked immunosorbent assay (ELISA) IFN-γ, IL-5, IL-9 and IL-13 in mouse serum and IFN-γ in culture supernatant were measured using an ELISA kit (MultiSciences) following the manufacturer's protocols. All samples were measured in triplicate and the mean concentration was calculated from the standard curve.
Cell proliferation assays IL-9-stimulated CT26 cells were cultured in 96-well plates (8 × 10 3 cells/well) in 5% CO 2 at 37 °C for 24 h. The proliferation ability of CT26 cells was detected using a MTT proliferation assay kit (MultiSciences) as recommended by the manufacturer.

Migration assay
For the tumor migration assay, 2 × 10 5 IL-9-stimulated CD8 + T cells were co-cultured with 1 × 10 5 CT26 cells in wells of a 24-well plate for 24 h. Adherent CT26 cells were collected and 2 × 10 4 cells in 0.2 ml of serum-free medium were added to the upper chamber of a transwell system; the lower chamber was filled with 0.6 ml serum-containing culture medium. The plate was incubated in 5% CO 2 at 37 °C for 20 h, and then the chamber was removed and cells were fixed with 4% paraformaldehyde for 15 min and stained with purple crystal for 10 min. Migration ability was quantified using direct microscopic visualization and cell counting.

Gene Expression Profiling Interactive Analysis (GEPIA)
GEPIA was used to analyze the expression of CXCL16 and CCL25 genes in the CRC tissue and adjacent normal tissue.

Statistical analysis
GraphPad Prism Version 7 software (GraphPad Software, Inc., La Jolla, CA, USA) was used to analyze all the data. The data are expressed as the mean ± SD. Comparisons between groups were assessed by unpaired Student's t-test or analysis of variance (ANOVA). A P value less than 0.05 was considered to indicate a statistically significant difference.

ILC2s were significantly increased in the tumor tissue of the CRC mouse model and the CRC patient
Previous studies have shown that ILC2s can regulate the development of a variety of tumors [12].
However, the role of ILC2s in CRC has not been investigated. We first measured the percentage of ILC2s in the tumor and adjacent normal tissue of the CRC mouse model and CRC patient. We regarded Lin − CD90.2 + KLRG1 + cells as mouse ILC2s and Lin − CD127 + CRTH2 + cells as human ILC2s [15,35].
Flow cytometry results showed that the percentage of ILC2s in mouse CRC tissue was significantly increased (P < 0.001) compared with that in adjacent normal tissue (Fig. 1A). We observed the similar result in the human CRC samples (Fig. 1B). The immunofluorescence result further demonstrated that ILC2s were increased in the human CRC tissue compared with adjacent normal tissue (Fig. 1C).
Together, these results demonstrated that ILC2s were increased in the tumor tissue of the CRC mouse model and CRC patient.

Blocking ILC2s promoted tumor growth in CT26 cell-bearing mice
To investigate the role of ILC2s in the progression of CRC, we used the murine CRC cell line CT26 to generate a CRC tumor model. Five days after CT26 cells were implanted, anti-CD90.2, an ILC2 blocking antibody was intravenously injected for 5 consecutive days ( Fig. 2A). The tumor volume was measured every 5 days and the tumor growth curves were drawn. The results showed that the tumor size was larger in anti-CD90.2-treated mice than control mice (Fig. 2B). In addition, the tumor volume and tumor weight were also increased in anti-CD90.2-treated mice (Fig. 2C,D). We next measured the percentage of ILC2s in the spleen and tumor tissue from mice treated with anti-CD90.2 and control group. ILC2s were decreased both in the spleen and tumor tissue of the anti-CD90.2 group (Fig. 2E,F). We also examined the levels of ILC2-related cytokines IL-5, IL-9, and IL-13, and the antitumor cytokine IFN-γ in peripheral blood serum. ILC2-derived cytokines, IL-5, IL-9 and IFNγ were decreased in the anti-CD90.2 group compared with the control group, but no significant change in IL-13 were found (Fig. 2G). It was suggested that the reduction in ILC2s could lead to an aggravation of tumor growth.
To preliminary examine the mechanism of increased tumor growth caused by reduced ILC2s, we 3. IL-9 inhibited the growth of CT26 cells by activating CD8 + T cells Our findings indicated that ILC2s exerted anti-tumor effect in CRC. Previous studies have shown that IL-9, a cytokine secreted by ILC2s, has a powerful anti-tumor effect in solid tumors [36]. To examine the role of IL-9 in CRC, we stimulated CT26, CRC cell line, with IL-9 and then measured the effects on apoptosis and proliferation. The results showed that IL-9 at all tested concentrations had no significant effect on the apoptosis and proliferation of CT26 cells (Fig. 3A,B).
Previous studies showed that IL-9, a T-cell growth factor, regulated tumor progression by activating CD8 + T cells [36]. We sorted CD8 + T cells from tumor tissue and stimulated cells with IL-9. A greatly enhanced IFN-γ secretion ability was observed in CD8 + T cells after stimulation by the various concentrations of IL-9 (Fig. 3C). Next, we co-cultured IL-9-stimulated CD8 + T cells with CT26 cells for inhibited the migration ability of CT26 cells (Fig. 3D). The CT26 cell number was also decreased after co-culture with IL-9-stimulated CD8 + T cells compared with controls ( Fig. 3E). This result indicated that IL-9-stimulated CD8 + T cells inhibited the proliferation of CT26 cells.
Apoptosis assay found that CD8 + T cells promoted the apoptosis of CT26 cells, and IL-9-stimulated CD8 + T cells further promoted the apoptosis of CT26 cells compared with CD8 + T cells (Fig. 3F,G).
These results demonstrated that IL-9 did not directly affect the biological activities of CT26 cells, but it could inhibit the proliferation, migration and apoptosis of CT26 cells via activating CD8 + T cells in vitro. These findings indicated an anti-tumor function of IL-9 in the CRC tumor microenviroment.
4. Anti-IL-9 promoted tumor growth of CT26 cell-bearing mice but IL-9 inhibited tumor growth Our findings indicated an anti-tumor effect of IL-9 on CT26 cells in vitro. To examine the anti-tumor effect of IL-9 in vivo, we established CT26 cell-bearing mice. At 5 days after tumor implantation, mice were injected intravenously with IL-9 or anti-IL-9 for 5 consecutive days (Fig. 4A). The results showed that neutralizing IL-9 promoted tumor growth in CT26 cell-bearing mice, while the tumor size and weight in the IL-9-treated group were decreased compared with the control group (Fig. 4B,C). These results demonstrated that IL-9 also exhibits anti-tumor effects in vivo.
To determine if IL-9 affected the tumor growth of CT26 cell-bearing mice by regulating CD8 + T cells, we analyzed the percentage of CD8 + T cells in tumor tissue with flow cytometry. The results showed that the percentage of CD8 + T cells in the anti-IL-9 group was significantly decreased compared with controls, while IL-9 increased the percentage of CD8 + T cells (Fig. 4D,E). We next measured the percentage of ILC2s in tumor tissues. We found the same result as the percentage of CD8 + T cells in tumor tissues (Fig. 4F,G). Previous studies have shown that IL-9 promotes the growth and proliferation of ILC2s [37]. This finding is consistent with our results, which showed that neutralizing IL-9 reduced the number of ILC2s, while administrating IL-9 increased ILC2 numbers. Together, these results demonstrated that IL-9 regulates the number of CD8 + T cells to regulate the CRC development. 5. ILC2 is the major IL-9-producing cell subset in murine CRC To determine whether ILC2s exert anti-tumor effects on CRC through secreting IL-9, we used flow cytometry to analyze IL-9 + cells in murine CRC tissue. More than half of the IL-9 + cells were lineage − cells and almost all the lineage − cells were CD90.2 + and KLRG1 + cells (Fig. 5A). This result indicated that ILC2s were the major IL-9-producing cell-subset in murine CRC tissue (Fig. 5B). These findings also demonstrated that ILC2s could exert anti-tumor effects by secreting IL-9.
To clarify if ILC2s activate CD8 + T cells through IL-9 to inhibit CRC progression. We sorted ILC2s and co-cultured ILC2s with CD8 + T cells. ELISA result showed that ILC2s significantly increased the secretion of IFN-γ by CD8 + T cells. However, when anti-IL-9 was added, the IFN-γ-secreting ability of CD8 + T cells was decreased compared with the control group (Fig. 5C). This result further proved that ILC2s could inhibit the progression of CRC through secreting IL-9.

IL-9 reversed the tumor-promoting effects of blocking ILC2s
We demonstrated that ILC2s inhibit the progression of CRC by secreting IL-9. We next evaluated whether the tumor-promoting effect that results from blocking ILC2s could be reversed by IL-9. We generated CT26 cell-bearing mice and 5 days after tumor implantation, mice were injected intravenously with anti-CD90.2, and 30 min later, each IL-9 treatment group mouse also received IL-9 for 5 consecutive days (Fig. 6A). The results showed that IL-9 significantly inhibited tumor growth after ILC2s were blocked. The tumor size and weight were significantly reduced in the anti-CD90.2 plus IL-9 treatment group compared with the anti-CD90.2 group (Fig. 6B,C). In addition, the percentages of CD8 + T cells and ILC2s in the anti-CD90.2 plus IL-9 treatment group were increased compared with the anti-CD90.2 group (Fig. 6D,E). These results further demonstrated that ILC2s regulate the progression of CRC by secreting IL-9, which could reverse the tumor-promoting effect that results from blocking ILC2s.

Peripheral ILC2s were recruited to the tumor site in CRC
Our results suggest that during the development of CRC, the body's immune resistance mechanism would result in increasing ILC2s, leading to inhibition of tumor growth by ILC2-mediated secretion of IL-9. However, we found that the percentage of ILC2s was decreased in the spleen of CT26 cell-bearing mice and CRC mice compared with control mice (Fig. 7A,B). This result was also observed in the peripheral blood of the CRC patient (Fig. 7C) We thus used GEPIA to predict the expression of CCL25, the ligand of CCR9, and CXCL16 in CRC tissue. The results revealed no significant difference in the expression of CCL25 in CRC tissue and adjacent normal tissue, while the expression of CXCL16 was significantly increased in CRC tissue compared with adjacent normal tissue (Fig. 7D). These results indicated that decreased ILC2s in the peripheral may be due to the increased CXCL16 in CRC tissue. Together, these results suggest that the peripheral ILC2s may be recruited by CXCL16 to CRC tissue to exert anti-tumor effects.

Discussion
Current studies have shown that ILC2s are involved in the development of a variety of tumors in a tumor-specific manner [40]. However, the role of ILC2s in CRC has not been investigated. Reports have shown that the function of ILC2s in tumors depends on the cytokines that are secreted, and the roles of ILC2-derived IL-4, IL-5 and IL-13 in cancers have been previously reported [15,41,42]. The function of IL-9, a characteristic cytokine of ILC2s in tumors has not been examined [43]. Here we revealed that ILC2s are recruited to CRC tissue and secrete IL-9 to activate CD8 + T cells, leading to inhibition of CRC development.
In this study, we observed increased ILC2s in tumor tissue compared with adjacent normal tissue in CRC mice. This is consistent with most previous studies showing that ILC2s are increased in tumor sites [15]. Previous studies have shown that there may be a large number of pro-inflammatory cytokines such as IL-33 and IL-25 in the tumor microenvironment [44]. Pro-inflammatory cytokines can lead to activation and proliferation of ILC2s [45,46]. To investigate the function of the increased ILC2s in CRC, we used anti-CD90.2, an ILC2 blocking antibody, to determine the role of ILC2s in CT26 cellbearing mice. Tumor growth was significantly increased in CT26 cell-bearing mice treated with anti-CD90.2, indicating that ILC2s serve an anti-tumor role in CRC. However, anti-CD90.2 is not an ILC2-specific neutralizing antibody, this is an limitation of our research. Our subsequent experiments with specific knockout of ILC2s are required to verify our results.
After showing that blocking ILC2s exacerbated CRC development, we investigated the anti-tumor mechanism of ILC2. IL-9, a characteristic cytokine of ILC2s, plays a positive role in the antitumor immunity of solid tumors. As a lymphoid cell growth factor, IL-9 directly inhibits tumor growth by up- However, the role of ILC2-derived IL-9 has remained unclear. We found that that IL-9 had no effect on the apoptosis and proliferation of CT26 cells. However, IL-9-stimulated CD8 + T cells significantly inhibited the migration, proliferation and apoptosis of CT26 cells. These results demonstrate an antitumor role of IL-9 in CRC. To demonstrate that the anti-tumor role of ILC2s was due to the secretion of IL-9 in CRC, we investigated the IL-9-secreting cells in CRC and found that ILC2s were the main IL-9secreting cell subset. These results demonstrated that ILC2s inhibit the progression of CRC by secreting IL-9.
We next investigated the anti-tumor role of ILC2-derived IL-9 in the CT26 cell-bearing murine model in vivo. Intravenous injection ofIL-9 in mice exerted a powerful anti-tumor effect while injection of anti-IL-9 in mice promoted tumor growth. These findings demonstrate the anti-tumor role of IL-9 in CRC.
Next, we examined whether IL-9 could rescue the tumor-promoting effect caused by ILC2 blocking.
We used the CT26 cell-bearing mice model and found that IL-9 reversed the tumor-promoting role of ILC2s blocking. This result proved that ILC2s indeed exhibit a tumor suppressive function by producing IL-9 in CRC. However, these experiments have a limitation.We should use the mice CRC model to detect the role of ILC2s and IL-9 rather than just used the CT26-bearing mice model. In our next study, we plan to investigate the role of ILC2s and IL-9 in a CRC model.
In our research, we found that ILC2s was decreased in the peripheral of the CRC model. Previous studies demonstrated that ILC2s express the chemokine receptors CCR6 and CXCR6 and can be recruited to tissues that highly expressed CCL25 and CXCL16 [38]. Based on these data and the increased ILC2s in CRC tissue, we examined whether CRC tissue highly expressed chemokines CCL25 and CXCL16. GEPIA results showed that CXCL16 is highly expressed in CRC tissues compared with adjacent normal tissues. This result explains the decrease of ILC2s in peripheral and increase in tumor tissue of CRC and further suggests the possibility that ILC2s may migrate from the periphery to CRC tissue by chemotaxis induced by CXCL16.

Conclusions
In this study, we demonstrated the role of ILC2s and ILC2-derived IL-9 in CRC. We found that ILC2s were increased in CRC tissue, and it was a main IL-9-secreting cell-subset. ILC2s inhibited the growth of CRC by secreting IL-9. We also found that that the CXCL16 was increased in CRC tissue and may recruit peripheral ILC2s to CRC tissue to exert anti-CRC effect.      Anti-IL-9 promoted the tumor growth of CT26 cell-bearing mice but IL-9 inhibited tumor growth. A, At 5 days after CT26 cell implantation, mice were injected intravenously with 1 µg IL-9 or 5 µg anti-IL-9 for 5 consecutive days. B-C, Tumor volume and weight in the different treatment groups. D-E, The percentages of CD8+ T cells in the tumors of IL-9-and anti-IL-9-treated mice and control mice were detected by flow cytometry. F-G, The percentages of ILC2s in the tumors of IL-9-and anti-IL-9-treated mice and control mice were detected by flow cytometry. Data are presented as mean±SD, *P < 0.005, **P < 0.001 compared with controls.

Figure 4
Anti-IL-9 promoted the tumor growth of CT26 cell-bearing mice but IL-9 inhibited tumor growth. A, At 5 days after CT26 cell implantation, mice were injected intravenously with 1 µg IL-9 or 5 µg anti-IL-9 for 5 consecutive days. B-C, Tumor volume and weight in the different treatment groups. D-E, The percentages of CD8+ T cells in the tumors of IL-9-and anti-IL-9-treated mice and control mice were detected by flow cytometry. F-G, The percentages of ILC2s in the tumors of IL-9-and anti-IL-9-treated mice and control mice were detected by flow cytometry. Data are presented as mean±SD, *P < 0.005, **P < 0.001 compared with controls.