Interleukin-33 Contributes to the Induction of Th9 Cells and Antitumor Efficacy by Dectin-1-Activated Dendritic Cells

We recently discovered that dectin-1-activated dendritic cells (DCs) drive potent T helper (Th) 9 cell responses and antitumor immunity. However, the underlying mechanisms need to be further defined. The cytokine microenvironment is critical for Th cell differentiation. Here, we show that dectin-1 activation enhances interleukin (IL)-33 expression in DCs. We found that blocking IL-33/ST2 inhibits dectin-1-activated DC-induced Th9 cell differentiation. More importantly, the addition of IL-33 further promotes Th9 cell priming and antitumor efficacy induced by dectin-1-activated DCs. Mechanistically, in addition to the promotion of Th9 and Th1 cells, dectin-1-activated DCs combined with IL-33 abolish the activity of IL-33 in the induction of regulatory T cells. Furthermore, the combined treatment of dectin-1-activated DCs and IL-33 increases the frequencies of CD4+ T cells by fostering their proliferation and inhibiting their exhaustive differentiation. Thus, our results demonstrate the important role of IL-33 in dectin-1-activated DC-induced Th9 cell differentiation and antitumor efficacy, and suggest that the combination of dectin-1-activated DCs and IL-33 may present a new effective modality of DC-based vaccines in tumor immunotherapy.

through the activation of other effectors, such as tumor-specific CD8 + CTLs and mast cells (10,11). Th9 cells may also exert a direct antitumor activity at the tumor site (12,13). These findings suggest that strategies to preferably stimulate DC induction of Th9 cells may greatly improve the antitumor efficacy of DC-based tumor therapy.
Il33, the gene encoding IL-33, is the third highest expressed gene among the 42 upregulated cytokine and costimulatory molecule genes in dectin-1-activated DCs (12). IL-33 is a member of the IL-1 superfamily cytokines and a ligand for IL-1 receptor-like 1 (IL1RL1, also known as ST2) (23,24). IL-33 has been reported to be implicated in multiple allergic disorders by stimulating Th2associated cell responses (25). In cancer immunology, IL-33 inhibits tumor growth through activation of CD8 + T and NK cells (26,27). IL-33 enhances the intratumor accumulation of type 2 innate lymphoid cells and eosinophils, which mediate potent antitumor activity (28,29). IL-33 can also promote the antitumor immunity by restoring the activation and maturation of DCs and inhibiting the differentiation and function of granulocytic myeloid-derived suppressor cells in tumors (30,31). IL-33 stimulates DC expression of OX40L (25), which may favor the induction of Th9 cells (12,15). Recently, Ramadan et al. showed that IL-33-treated IL-9-producing T (T9) cells have more potent antileukemic capacity than untreated T9 cells (32). Based on these observations, we hypothesized that IL-33 would greatly contribute to dectin-1-activated DC-induced Th9 cell differentiation and antitumor efficacy.
In this study, we showed that dectin-1 activation stimulates IL-33 expression in DCs. We found that IL-33 contributes to dectin-1-activated DC-induced Th9 cell differentiation. More importantly, the addition of IL-33 further promotes dectin-1-activated DC induction of Th9 cells and antitumor efficacy. Our results demonstrate the important role of IL-33 in dectin-1-activated DC-induced antitumor immunity and may have important clinical implications.

MaTerials anD MeThODs
Mice and cell lines C57BL/6, OT-II [C57BL/6 − Tg(TcraTcrb)425Cbn/J], and Balb/c mice were purchased from the Jackson Laboratory. Phenotype of OT-II transgenic mice was examined and confirmed (Figures S1A,B in Supplementary Material). Dectin-1 −/− mice were provided by G. Brown (University of Aberdeen, Aberdeen, Scotland). All mice were housed and bred under specific pathogenfree conditions at Animal Center of The First Hospital of Jilin University. Mice were used for experiments at age 6-8 weeks. All animal studies were conducted according to the ethical guidelines of the Animal Ethical Committee of First Hospital of Jilin University.

gene expression Profiling Data
Microarray analyses of the gene expression profiles of BMDCs and CurDCs were performed previously (12). Data are stored in the GEO repository and is accessible under the accession number GSE81111.

enzyme-linked immunosorbent assay
Concentrations of IL-33, IL-9, and IFN-γ in culture supernatants were detected by ELISAs as previously described (12). IL-33 capture/detection Abs were purchased from R&D Systems. Recombinant mouse IL-33 (aa109-266) (ELISA standard) was purchased from R&D Systems. Capture/detection Abs for IL-9 and IFN-γ were purchased from BD Biosciences. Recombinant mouse IL-9 and IFN-γ used as the standards in ELISAs were purchased from R&D Systems and BD Biosciences, respectively. Avidin-HRP was purchased from BioLegend.

Tumor immunotherapy experiments
BMDCs and CurDCs were pulsed with OT-II OVA peptides (5 µg/mL) for 2-4 h and then harvested for mouse immunization (n = 4-5/group). B16-OVA (2 × 10 5 cells/mouse) were injected subcutaneously into OT-II mice. On day 3 after tumor challenge, mice were randomly divided into groups and given two weekly subcutaneous immunizations with treated DCs (1 × 10 6 cells/ mouse). Mice injected with PBS served as controls. In some experimental groups, mice were given IL-33 (250 ng/mouse) every 3 days starting at 1 day after the first DC immunization. Tumor volume was calculated by the formula: 3.14 × (mean diameter) 3 /6. Mice were killed when the tumor diameter reached to the range between 1.5 and 2 cm.
We also used MPC-11 myeloma tumor model to further examine the role of dectin-1-activated DCs plus IL-33 in tumor therapy. MPC-11 tumor cells (1 × 10 6 cells/mouse) were injected subcutaneously into Balb/c mice. MPC-11 tumor cell lysates were generated by five rapid freeze-thaw cycles as described previously (33) and were used as tumor antigens. On day 3 after tumor challenge, BMDCs or CurDCs (1 × 10 6 cells/mouse) loaded with tumor cell lysates (100 μg/1 × 10 6 DCs) were used for the tumor treatment in the presence or absence of IL-33. Mice were treated twice (1 week apart). Mice injected with PBS served as controls. Tumor growth was monitored overtime.

In Vivo Functional Tests of il-33/sT2 in Dc-induced T cell Differentiation
BMDCs and CurDCs were pulsed with OT-II OVA peptides (5 µg/mL). Mice were given two weekly subcutaneous immunizations with 1 × 10 6 treated DCs. Mice injected with PBS served as controls. In some experiments, mice were given control IgG or blocking anti-ST2 mAb (αST2, 25 μg/mouse) or IL-33 (250 ng/ mouse) every 3 days starting at 1 day after the first DC immunization. On day 3 after the second DC immunization, total leukocytes from spleens and lymph nodes were restimulated with peptide-pulsed DCs for 24 h. Cells from PBS control mice were cultured without addition of DCs. Culture cells and supernatants were collected and analyzed by qPCR, ELISA, and flow cytometry.

statistical analysis
The Student's t-test (2 groups) and one-way ANOVA (≥3 groups) were used to compare various experimental groups. A P value of less than 0.05 was considered significant.
To further explore the effects of IL-33 on dectin-1-activated DC-induced Th9 cell differentiation, cytokine IL-33 was added during Th9 induction by CurDCs. The addition of IL-33 further stimulated CurDC-induced Th9 cell development (Figures 2A,B) and increased IL-9 (Figures 2A-C  OVA-peptide-pulsed BMDCs and CurDCs. CurDC-treated mice had higher percentages of ST2-expressing CD4 + (ST2 + CD4 + ) T cells in spleen cells ( Figure 3A) and higher levels of St2 mRNA expression in CD4 + T cells ( Figure 3B) than mice treated with BMDCs or PBS controls. However, there was no difference in ST2 expression by CD4 + T cells from mice immunized with BMDCs compared to PBS controls (Figures 3A,B). In addition, the immunization of CurDCs compared to BMDCs or PBS control also led to higher frequencies of ST2 + CD4 + T cells in C57BL/6 mouse model ( Figures S4A,B (12). To determine whether IL-33 contributes to the production of the Th cells in vivo, αST2 was used during CurDC immunization. A blockade of IL-33/ST2 with αST2 partially inhibited dectin-1-activated DC-induced Th9 and Th1 cell differentiation, as demonstrated by the lower expression levels of IL-9 and IFN-γ in T cell from mice immunized with CurDCs plus αST2 compared to CurDCs alone (Figures 3C,D). Together, these data demonstrated that IL-33/ST2 axis contributed to dectin-1-activated DC induction of Th9 and Th1 cells in vivo.
Lymphocyte infiltration in tumors plays a central role in mediating the antitumor effects induced by DC-based tumor therapy. We next analyzed the tumor-infiltrating lymphocytes (TILs) in B16-OVA tumor-bearing OT-II mice immunized with CurDCs plus IL-33. As shown in Figures S6A,B in Supplementary Material, mice immunized with CurDCs plus IL-33 had higher frequencies of IL-9 + CD4 + , IFN-γ + CD4 + , and GzmB + CD4 + TILs than mice immunized with CurDCs alone or BMDCs plus IL-33, whereas mice immunized with BMDCs plus IL-33 had similar frequencies of IL-9 + CD4 + , IFN-γ + CD4 + , and GzmB + CD4 + TILs as compared to those immunized by BMDCs alone (Figures S6A,B in Supplementary Material). Together, these results demonstrated   Dectin-1-activated Dcs Plus il-33 Fail to support Treg cell Differentiation Interleukin-33 was shown to promote Treg cell development (34,35). To explore the effects of IL-33 on Treg cell differentiation in DC-immunized mice, OT-II mice were immunized with OVA peptide-pulsed BMDCs or CurDCs with or without the addition of IL-33. Indeed, mice immunized with BMDCs plus IL-33 displayed significantly higher production of Foxp3 + CD4 + Treg cells ( Figure 5A) and expression levels of Foxp3 and Il10 in T cells (Figures 5B,C) than mice receiving CurDCs, CurDCs plus IL-33, BMDCs, or PBS control. However, there was no increase of Foxp3 + CD4 + Treg cells ( Figure 5A) and Foxp3 and Il10 gene expression in T cells (Figures 5B,C) from mice immunized with CurDCs plus IL-33 compared to CurDCs alone. We next examined the effects of IL-33 on DC-induced Treg cell differentiation in vitro. Similarly, the addition of IL-33 increased Foxp3 expression in BMDC-treated Treg cells (Figure 5D), but not in CurDCtreated cells ( Figure 5D). Together, these results demonstrated that the combination of dectin-1-activated DCs and IL-33 fails to support Treg cell differentiation.
il-33 Promotes the Proliferation of Dectin-1-activated Dc-Treated Th cells The expansion of antitumor effector T cells is critical for antitumor immunity in vivo. OT-II mice were used and the effects of dectin-1-activated DCs and IL-33 on the expansion of CD4 + T cells were examined. As shown in Figure 6A, mice immunized by CurDCs compared to BMDCs had comparable frequencies of CD4 + T cells in spleen cells. Interestingly, the addition of IL-33 increased the CD4 + T cell frequencies in CurDCimmunized mice ( Figure 6A) but reduced the frequencies in BMDC-immunized mice ( Figure 6A). To determine the effects of dectin-1-activated DCs and IL-33 on Th cell proliferation, T cells from spleen cells and tumor tissues were analyzed by intracellular staining of Ki67. IL-33 increased Ki67 expression in CD4 + T cells from CurDC-treated mice ( Figure 6B; Figures (Figure 6E; Figure S6F in Supplementary Material). These results indicated that IL-33 promoted an exhaustive phenotype of Th cells in mice immunized with BMDCs but not CurDCs, suggesting the potential mechanism for the high proliferation capability of Th cells in mice immunized by CurDCs plus IL-33.

DiscUssiOn
Tumor-specific Th9 cells are potent antitumor effector cells (10,11). We recently reported that dectin-1-activated DCs promote Th9 cell differentiation and trigger potent therapeutic effects against established tumors, better than regular BMDCs (12,13). However, mechanisms underlying the induction of Th9 cells and antitumor efficacy by dectin-1-activated DCs are not fully defined. IL-33 is one of the 42 upregulated genes of cytokines and costimulatory molecules in dectin-1-activated DCs (12). And we also confirmed the upregulation of IL-33 expression in dectin-1-activated DCs in this study. However, the role of IL-33 in the induction of Th9 cells and antitumor immunity by dectin-1-activated DCs still remains unknown. IL-33 is a nuclear cytokine, which is released via cell necrosis (36). Matured DCs will undergo cell death in vitro and in vivo and IL-33 can be released from these dying dectin-1-activated DCs, which may affect T cell differentiation primed by dectin-1-activated DCs. Indeed, in this study, we found that blocking IL-33/ST2 axis inhibits dectin-1-activated DC-induced Th9 differentiation in vitro and in vivo. In addition, the addition of IL-33 further promotes the development of Th9 and Th1 cells induced by dectin-1-activated DCs. More importantly, the addition of IL-33 further increases dectin-1-activated DC-induced antitumor efficacy in mouse models. Consistently, a recent report showed that IL-33-treated IL-9-producing T cells have higher antileukemic capabilities than untreated cells (32). Thus, our data demonstrate the important role of IL-33 in dectin-1-activated DC-induced Th9 cell differentiation and antitumor efficacy.
Cytokine milieu is the major determinant for Th cell differentiation (5,37). The cytokine IL-33 itself may promote the differentiation of Th1 or Th9 cells, depending on the cytokine milieu (24,32). Regular matured DCs express high levels of IL-12, a key cytokine for Th1 cell differentiation (38), which along with IL-33 directs naïve CD4 + T cells to differentiate into Th1 cells. However, dectin-1-activated DCs produce large amounts of TNF superfamily cytokines, such as TNFSF15 and OX40L (12,14,15), which along with IL-33 may further promote Th9 cell formation. Th1 cells are terminally differentiated, shorter lived effector cells (39,40), however, Th9-derived IL-9 has the potential to enhance the survival and proliferation of T cells (12). In addition, IL-33 itself may stimulate the survival and proliferation of T cells (41). Thus, these observations may account for the capability of IL-33 in promoting Th9 cell production and antitumor immunity in dectin-1-activated DC-immunized mice but not in BMDCimmunized mice.
In this study, we found that IL-33 increases the production of Foxp3 + Treg cells in mice immunized by BMDCs but not dectin-1-activated DCs. Treg cells can be divided into naturally occurring (nTreg) and inducible (iTreg) regulatory cells (42). nTreg cells are generated in the thymus and are 5-10% of the total peripheral CD4 + T cells (43). By contrast, iTregs are peripherally induced from naive CD4 + T cells in response to foreign antigens in the presence of TGF-β1 (44,45). These observations suggest that IL-33 increases the production of iTreg but not nTreg cells in BMDC-immunized mice. It remains unclear about mechanisms of the different function roles of IL-33 in Treg cell differentiation in BMDC-and dectin-1-activated DC-immunized mice. But dectin-1-activated DCs overexpress the TNF family cytokines TNF-α, TNFSF15 (TL1A), OX40L, and TNFSF8 (12,13). Among these cytokines, OX40L inhibits Treg cell differentiation and TL1A inhibits the immunosuppressive function of Treg cells (46)(47)(48)(49). Based on these observations, we speculate that the specific profile of cytokines and costimulatory molecules expressed by dectin-1-activated DCs, especially OX40L and TL1A may contribute to the inhibition of Treg cell differentiation induced by IL-33.
In conclusion, our study demonstrates that IL-33 contributes to the induction of Th9 and Th1 cells by dectin-1-activated DCs. The addition of IL-33 further promotes dectin-1-activated DC-induced antitumor efficacy. The combined use of dectin-1-activated DCs and IL-33 may present a new effective modality of DC-based vaccines in tumor immunotherapy. Our study may have important clinical implications.

eThics sTaTeMenT
This study was carried out in accordance with the recommendations of "the ethical guidelines of the Animal Ethical Committee of First Hospital of Jilin University. " The protocol was approved by the "Animal Ethical Committee of First Hospital of Jilin University. " aUThOr cOnTriBUTiOns SW initiated the study; SW designed the experiments and wrote the paper; SW, JC, and YZ performed majority of the experiments; YJ, YW, DW, and AW performed some experiments; AW read and edited the manuscript; QY, SG, HY, and RG provided critical suggestions to this study.

FUnDing
This work was supported by funds from National Natural Science Foundation of China (81372536 to SW, 81602485 to YZ and 81671592 to HY).