GITR drives T_H9-mediated antitumor immunity

ABSTRACT

T_H9 cells have been implicated in triggering antitumor immunity. We have identified that GITR co-stimulation inhibits iT_reg cell generation but drives T_H9 cell differentiation, thereby suppressing tumor growth via enhancing the function of DCs and CTLs in vivo. Our findings provide novel mechanisms by which GITR agonists exert antitumor activity.

KEYWORDS:

• Antitumor immunity
• GITR
• iT_reg cell
• IL-9
• T_H9 cell

Increasing evidence has indicated the advantages of CD4⁺ T helper (T_H) cells in the treatment of cancer in experimental animals as well as in humans.^1,2 The IL-9-producing CD4⁺ T (T_H9) cell population is a newly identified T_H subset and has been shown to play roles in mediating parasite expulsion, allergic inflammation and antitumor immunity.³ Notably, T_H9 cells have been shown to exert the greatest antitumor activity among the T_H subsets in melanoma rejection.⁴ Hence, the induction of T_H9 cell immunity might provide an efficacious strategy for the treatment of tumors in humans.

Glucocorticoid-induced tumor necrosis factor receptor (TNFR)—related protein (GITR) is one of the molecules in the TNFR family that co-stimulates T cells. Treatment with GITR agonists showed strong antitumor effects in various tumor models.⁵ However, it has been unclear how GITR co-stimulation on T cells generates antitumor activity. In a recent study, we have found that GITR signaling profoundly enhances T_H9 cell differentiation and that IL-9 production is required for the antitumor activity mediated by GITR agonists.⁶

First, we examined whether IL-4Ra signaling plays any role in tumor rejection mediated by the anti-GITR agonistic antibody DTA-1 in a CT26 colon cancer model because DTA-1 treatment increased the expression of IL-4 and IL-13 in tumor-bearing hosts. Although T_H2 cell immunity has been suggested as pro-tumorigenic, it is also known to mediate antitumor immunity.⁷ We found that IL-4Ra signaling was essential for GITR agonist-induced tumor regression by showing that the tumor growth in IL-4Ra knockout (Il4ra^−/−) mice was not affected by DTA-1. Notably, we observed that treatment with DTA-1 substantially enhanced IL-9 expression in WT tumor-bearing hosts, while it failed to do so in Il4ra^−/− recipients. This result prompted us to analyze the effect of IL-9 on antitumor immunity induced by GITR stimulation. GITR ligation in vivo upregulated IL-9 expression in CD4⁺ T cells as early as 2 d after DTA-1 treatment. More importantly, the inhibition of tumor growth induced by DTA-1 was markedly reversed by a neutralizing antibody to IL-9. In addition, we employed a mouse model of B16 melanoma expressing ovalbumin (OVA) and compared the antitumor activity of adoptively transferred OVA-specific T_H9 cells generated with or without GITR co-stimulation. In this experiment, we observed that GITR engagement endowed donor T_H9 cells with superior antitumor efficacy in an IL-9-dependent manner. Based on this finding, we concluded that CD4⁺ T cells are the main cell type that responds to GITR co-stimulation to induce IL-9-dependent tumor regression.

Due to the lack of IL-9 receptor expression on CT26 and B16 tumor cells, we thought IL-9 acted on the intermediary cells rather than the tumors to exert antitumor activity. When we measured CD8⁺ cytotoxic T lymphocyte (CTL) responses, we found that tumor-specific cytolytic activity and cytokine and cytolytic marker expression (granzyme B, IFN-γ, TNF-α and CD107a) were all enhanced in CTLs treated with DTA-1 in an IL-9-dependent manner. Interestingly, although IL-9 expression was rapidly upregulated upon GITR ligation as early as day 2, we hardly detected tumor-specific CTL responses and the expression of related effector molecules in tumor-specific CTLs at the early time points. However, these were detected approximately a week after DTA-1 treatment in our experimental setting. In addition, our in vitro study revealed that IL-9 did not directly affect CD8⁺ T cell cytotoxicity. These observations led us to hypothesize that there must be other mediator(s) that responds to IL-9 and stimulates CTL responses in vivo. In a recent paper, Lu et al. have demonstrated that T cell-derived IL-9 chemoattracted dendritic cells (DCs) into the tumor site, thereby mediating T_H9 antitumor immunity through the activation of tumor-specific CTLs.⁸ We observed that DCs accumulated in the tumor tissue of DTA-1-treated mice, expressed higher levels of CD80, CD86 and MHC class II, and cross-presented tumor antigen (Ag) more efficiently than those of control IgG-treated mice, which were also affected by IL-9. Collectively, we propose that GITR-triggered IL-9 production from CD4⁺ T cells promotes tumor-specific CTL responses by activating tumor-infiltrating DCs in vivo, which in turn eradicates tumors (Fig. 1).

Figure 1. A schematic illustration of T_H9 cell-mediated antitumor immunity induced by GITR co-stimulation. GITR triggering inhibits iT_reg cell generation and promotes T_H9 cell differentiation. IL-9 production triggers epithelial cells to chemoattract DCs into the tumor and enhances the cross-presentation and costimulatory capacity of the tumor-infiltrating DCs. These tumor-Ag-crosspresenting DCs then potentiate tumor-specific CD8⁺ CTL responses, thereby facilitating tumor regression.

Next, by using an in vitro culture system, we demonstrated that GITR co-stimulation preferentially enhanced mouse T_H9 differentiation in a cell-intrinsic manner. Additionally, we observed an increase in human T_H9 differentiation when human GITR was triggered by a stimulatory antibody, suggesting that GITR stimulation might be capable of inducing T_H9 responses in humans. Notably, under induced T_reg (iT_reg) cell-polarizing conditions, GITR co-stimulation inhibited iT_reg cell generation and diverted the differentiation of CD4⁺ T cells toward T_H9 effector cells (Fig. 1). Hence, GITR co-stimulation might have two advantages: (i) eliminating potential immune suppressors by inhibiting iT_reg cell generation in tumor sites and (ii) promoting antitumor T_H9 effectors that repress tumor growth. A recent study by Xiao et al. supports this notion by showing that this phenomenon could take place in vivo in lymphopenic tumor-bearing hosts.⁹ Moreover, we demonstrated that reprogramming established T_reg cells into IL-9-producing cells is possible when they are simulated with a GITR agonist in the presence of IL-4.

In summary, our study demonstrated a role for GITR co-stimulation in T_H9 cell development and its functional cascade in the tumor microenvironment. However, it remains unanswered which cell type is the direct target of IL-9 to facilitate tumor-Ag cross-presentation by DCs. We speculate that cytolytic innate cells such as eosinophils,¹⁰ macrophages, NK cells and other cell types might be involved in this process. It will be important to investigate whether the antitumor activity of the GITR-agonistic antibody in human cancer patients is also IL-9-dependent. A combination of IL-9-inducing GITR agonists with immune checkpoint blockers likely further improves antitumor immunity in vivo.

Disclosure of potential conflicts of interest

No potential conflicts of interest were disclosed.

Funding

This study was supported by the National R&D Program for Cancer Control, Ministry of Health and Welfare (No. 0720500, C.-Y. Kang) and the Mid-career Researcher Program (No. 2015R1A2A1A10055844, C.-Y. Kang) of the National Research Foundation of Korea (NRF).