Role of Tim-3 in regulating tumorigenesis,inflammation,and antitumor immunity therapy

Immunotherapy is being increasingly considered as a vital part of cancer therapy. It has been evidenced to enhance the human immune response to cancer cells, hinders tumor growth, and has attained remarkable success in the field of cancer therapy. The treatment by targeting immune checkpoint inhibitors (ICRs), such as programmed cell death-1 (PD-1) and cytotoxic T-lymphocyte antigen-4 (CTLA-4), has been successfully accomplished in kidney cancer, melanoma, and lung cancer. Nevertheless, due to a poor response rate of certain cancer treatments, it is necessary to seek other co-inhibitory receptors or ICRs to enhance the outcome of immunotherapy ineffective or refractory to anti-CTLA-4 and anti-PD-1 therapy. Tim-3 is a member of the T-cell immunoglobulin and mucin domain-containing protein 3 (TIM) family. Tim-3 is a co-inhibitory receptor, generally expressed in CD4 + T, CD8 + T, dendritic, and macrophagic cells that secrete IFN- γ . The interaction between Tim-3 and its ligands has been proved to induce a negative regulatory pathway of the normal immune response, causing amplified autoimmunity and loss of tolerance and rendering Tim-3 a negative regulatory molecule. These immune tolerance mechanisms are part of the latest immunotherapy, which work by blocking immune checkpoints to annul the immune response of functional inhibition, activate T-cells, and eventually promote antitumor immunity. Tim-3 is one such prospective receptor for tumor immunotherapy. Blocking Tim-3 using the checkpoint inhibitors is known to augment antitumor immunity and inhibit tumor growth by several studies. Sun et al. [1] have shown that intravenous injection of oncolytic vaccinia virus (oVV) has a synergistic antitumor immunotherapy effect with both PD-1 and Tim-3 blockade in a mouse lung cancer model, rather than blocking single PD-1 or Tim-3. Studies have shown that Tim-3 combined with PD-1 blocking may help to prevent the depletion of CD8 + T cells in patients with advanced acute myeloid leukemia (AML) and other malignant hematological tumors [2]. Even though, Tim-3 has been shown to play a certain role in the occurrence and development of liver cancer [3], colon cancer [4]. and so on. In order to gain an in-depth understanding of Tim-3 as a potential immune target, the biological characteristics of Tim-3 have been reviewed initially, and its role in regulating tumorigenesis, tissue inflammation, and antitumor immunity is detailed next, followed by its usage as a target for cancer therapy.

Otherwise known as hepatitis a virus cellular receptor-2 (Havcr-2), Tim-3 is a member of the TIM family [5]. Whereas eight TIM genes (Tim-1–Tim-8) have been found in mice, only three genes encoding Tim-1 (Havcr-1), Tim-3 (Havcr-2), and Tim-4 (Timd4) have been identified in humans [6]. Tim-3’s structure has four distinct regions, namely, variable immunoglobulin domain (IgV), mucin domain, a transmembrane region, and the cytoplasmic tail [7]. The IgV domain includes two antiparallel chains, N-linked glycosylation sites, and four cysteines [8]. As against the traditional IgV domain, the Tim-3 IgV domain consists of two disulfide bonds in the four cysteines conserved by TIM protein, which repositions the CC’ loop closer to the FG loop, forming a unique cleft in the IgV domain of the family of proteins [9] (Fig. 1). Among the TIM family, Tim-3 is the molecule with the smallest mucin domain, containing O-linked glycosylation sites, especially rich in serine, proline, and threonine [9]. A tail, mainly composed of hydrophobic amino acids and embedded in a lipid bilayer, is utilized by Tim-3 proteins to pass through the cell membrane, which can penetrate into the cell with a cytoplasmic domain of ∼ 42–77 amino acids. Tim-3 also occurs in a soluble form (sTim-3), which, as compared with the complete Tim-3 structure, lacks mucin and transmembrane domain [10].

Earlier studies recognized Tim-3 as a specific membrane marker for terminal differentiation of helper T-cell 1 (Th1) but is not expressed on helper T-cell (Th2) [11]. Tim-3 has also been indicated by recent research to be expressed on pro-inflammatory T-cell subsets of Th17 cells, although the level of expression is lower than that of Th1 cells [12, 13]. Tim-3 was later found to be not just expressed on cells of the acquired immune response. Tim-3 has also been reported to be expressed by other cell subsets, including human monocytes [14], macrophages [15], natural killer cells [16], dendritic cells [17], mouse dendritic cells [18], mast cells [19], macrophages [20], and even in certain tumor cells. These results indicate that Tim-3 not only negatively regulates inflammatory T-cell function but also essentially participates in inhibiting autoimmune response [21], besides supporting the formation of an immunosuppressive tumor environment [21]. Depending on the pedigree transcription factor T-BET, Tim-3 is strongly expressed in differentiated Th1 cells. T-BET encourages the expression of Tim-3 by directly binding to the Tim-3 promoter [22]. Second, immunosuppressive cytokine IL-27 is a T-cell depletion regulator, which stimulates suppressor gene modules that cohere to Tim-3 and other ICRs such as TIGIT (T cell immunoglobulin and ITIM domain), Lag-3 (Lymphocyte activation gene-3), and PD-1 [23]. The transcription factor NFIL3 can also be upregulated by IL-27 to directly trigger the expression of Tim-3, whereas NFIL3 and T-BET together reshape the chromatin structure of the Tim-3 promoter for transcription [24].

To date several Tim-3 ligands have been discovered. The first identified ligand of Tim-3 is Galectin-9 (Gal-9), which belongs the highly conserved galactose lectin family [25]. It suggests the inhibitory function of Tim-3, as Gal-9 particularly recognizes Tim-3 of Th1 cells and induces apoptosis, thereby hindering tissue inflammation and autoimmune disease EAE [26]. Gal-9 has been identified to be expressed in autoimmune diseases, tumor microenvironment, chronic infection tissues, and various other cells [26]. Phosphatidylserine (PtdSer) is an essential phospholipid found in the membranes of bacteria, yeast, plants, and mammalian cells [27]. It can bind human and mouse Tim-3 in the FG-CC’ loop of the IgV domain. As there is an FG-CC’ loop in the TIM family, PtdSer can also act as a ligand for other TIM family proteins [9]. Thus, PtdSer may be the key ligand of Tim-3 on macrophages or dendritic cells and their interaction could be essentially responsible for initiating inhibitory signals without prompting phagocytosis of apoptotic cells. Another ligand of Tim-3 is CEACAM-1. It is found to be expressed on T-cells and by way of the interaction of IgV domains on two molecules, binds to Tim-3 [28]. The negative regulatory function of Tim-3 has been shown to be defective in the absence of CEACAM-1, suggesting CEACAM-1 is essential for Tim-3 to function normally, and that CEACAM-1 enhances the expression on the surface of Tim-3 cells [28]. Another Tim-3 ligand is HMGB1 that influences innate immune response [29]. HMGB1 chiefly encourages the functioning of congenital cells by cohering to advanced glycation end products receptor (RAGE) and Toll-like receptor (TLR), in turn activating congenital cells and producing pro-inflammatory factors. Nevertheless, Tim-3, together with HMGB1, interferes with the above process, thus suppressing the activation of the innate immune response.

3. Roles of TIM-3 in cancer

A pivotal emerging hallmark of cancer is evasion of immunological destruction, in particular by T cell dysfunction or T cell exhaustion. Aiming at therapeutic targeting, the breakthrough drugs have been increasingly developed including anti-ICRs mAb, CAR-T, and oncolytic virus that are in clinical trials or approved for treating certain types of human cancer.

Tim-3 can act as an attractive clinical target for checkpoint-targeted cancer immunotherapy. Consistent with the key regulatory effects of innate and acquired immune responses, blocking Tim-3 plays a positive role in restoring anti-tumor immunity (Fig. 4). However, Tim-3 may play a more complex role in regulating anti-tumor response. For instance, Tim-3 blockade by anti-Tim3 antibody reduced both population of CD4 + CD25 + Foxp3 + Treg and Tim-3 + Treg cells, drived antitumor immune response and suppressed tumor growth in head and neck cancer with mice xenograft [53, 54]. Tim-3 target is an effective target for tumor immunotherapy, and its expression is related to the poor prognosis of tumor and drug resistance of PD-1. Anti-Tim-3 antibody has a synergistic effect with CTLA-4, PD-1 and PD-L1 antibodies, and can treat patients with drug resistance to PD-1 and PD-L1. However, single Tim-3 blockade by antibody does not achieve a substantial therapeutic benefit, and the principle of the limitations of single blocking Tim-3 is not known.

The increasing data showed that there exist a common interaction between Tim-3 and other ICRs to regulate T cell function, such as frequent Tim-3 overexpression on tumor-infiltrating CD8 + PD-1 + T cells in various types of cancers [55, 56, 57]. Thus, it provides the basis of combination therapy with Tim-3 blockade and PD-1 blockade. The study showed that targeting both Tim-3 and PD-1 with their antibodies reversed T cell exhaustion and exerted a synergistic effect in inhibiting tumor growth and restoring antitumor immunity [56]. In addition, the combination blockade of Tim-3 and PD-1 reversed tumor-induced T cell exhaustion, and improved cancer antigen specific CD8 + T cell responses in melanoma [57, 58]. Co-blockade of Tim-3 and PD-1 also obtained stronger antitumor immune responses in colon cancer and AML mouse models than that of PD-1 blockade alone [2, 59, 60]. Moreover, co-targeting of Tim-3 and CTLA-4 with blocking antibodies was showed to be well-tolerated and to be more effective in T cell IFN- γ -mediated antitumor immunity and inhibition of established tumor growth compared with single blockade [59].

To be noted, Tim-3 expression is limited on terminally differentiated IFN- γ -producing T cells and intratumoral Treg cells, whereas CTLA-4 and PD-1 are broadly expressed on activated-T cells, and expression of CTLA-4 exists on all Treg cells [45]. Moreover, Tim-3-deficient mice didn’t develop experimental allergic asthma [61], whereas both PD-1-deficient and CTLA-4-deficient mice develop autoimmune disease [62, 63]. Thus, Tim-3 blockade is likely to avoid risk of toxic effects that occurred in PD-1 or CTLA-4 blockade, and the advantage of co-blockade of Tim-3 and PD-1 or CTLA-4 may be safer and more effective than that of co-blockade of PD-1 and CTLA-4 in targeting intratumoral T cells.

Recently, combinational strategy of ICR blockade therapy and oncolytic virus attracted more and more attention, and achieved satisfactory results in clinical trials for several solid tumors through the hypothesis that oncolytic virus can turn cold (such as refractory PD-1 - Tim3 - ) cancers into hot (PD-1 + Tim3 + ) ones allowing ICR blockade antibody to treat better [64]. It can be verified by Sun et al. [1] that systemically administered oncolytic vaccinia virus exerts synergetic antitumor efficacy with co-blockade of PD-1 and Tim-3 in refractory lung cancer through direct cell killing effect, recruitment of TILs and enhanced expression of PD-1 and Tim-3. In addition, we combine oncolytic adenovirus and Tim-3 blocking antibody to conduct therapeutic experiments, and found this combination could efficiently suppress tumor cell growth in liver cancer (unpublished data), indicating possible prospect for other cold solid cancer therapy.

Recent studies showed that high tumor-infiltrating CD39 + CD8 + T cells has reduced secretion of TNF- α and IFN- γ along with enhanced expression of Tim-3 that decides poor prognosis in renal carcinoma [65], indicating that the ectonucleotidase CD39 may be acted as a novel combinational therapy target with anti-Tim-3 antibodies. Actually, it was early reported that co-blockade of Tim-3 with a blocking antibody and CD39 by the small molecule inhibitor POM-1 significantly inhibit melanoma growth in mouse xenograft model [66]. Moreover, some studies also have tried to use shRNA/siRNA to interfere Tim-3 expression, and showed that siTim-3 could efficiently suppress tumor growth and enhanced antitumor immune effect [39, 67], suggesting an alternative therapeutic approach targeting Tim-3. Hepatocellular carcinoma (HCC) is still one of the types of tumors with the worst prognosis, as a cancer with few systematic treatment, sorafenib is approved to be used in the treatment of liver cancer but can not eradicate liver cancer. Recent studies have shown that it is safe and effective to use cabozantinib, ramucirumab or metronomic capecitabine in the treatment of liver cancer in patients with HCC who are ineffective in first-line sorafinib failure [68, 69]. In the clinical treatment of hepatocellular carcinoma, it is expected that Tim-3 combined with chemotherapeutic drugs can further improve the therapeutic efficiency of hepatocellular carcinoma.

From bench to bedside, it is indispensable to further investigate the clinical effect of anti-Tim3 blocking antibodies due to its positive role of Tim-3 blockade in cancer therapy, especially the co-blockade of Tim-3 and PD-1 with the antibodies or oncolytic virus. Up to date, there are several ongoing and planned chinical trials with anti-Tim-3 blocking antibodies alone or in combination with anti-PD-1, PD-L1 or LAG-3 antibodies et al (Table 2). However, a single anti-Tim-3 mAb agent didn’t strongly elicit activating immune responses and significant tumor regression. It might be caused by which Tim-3 is co-expressed with other ICRs such as PD-1 on CD4 + T and CD8 + T cells, and its effect of blocking Tim-3 alone or PD-1 alone is limited. In addition, as immunosuppressive cytokine, IL-27 might play pivotal roles in driving ICRs expression including Tim-3, PD-1, TIGIT, and LAG-3 on CD4 + or CD8 + T cells [23, 24]. Thus, most clinical trials were performed by co-blockade of Tim-3 and PD-1 using their blocking antibodies. TSR-022, the first reported humanized anti-Tim3 blocking antibody by Tesaro company [70], was used in a phase I trial for patients with advanced solid tumors by single agent treatment and had no dose-limiting toxicity [71]. Furthermore, TSR-022 in combination with TSR-042 (anti-PD-1) exerted increased clinical effect in patients with colorectal cancer, post-PD-1 NSCLC and melanoma without dose-escalated toxicity [72]. Another two developed Tim-3 blocking antibodies LY3321367 [73] and MGB453 [74] derived from Eli Lilly and Company and Novartis, respectively, were proved to be interim safe, efficient and well-tolerated when combined with anti-PD-1 in a phase I trial in advanced cancers. More development of anti-Tim3 blocking antibodies, such as M6903 [75], are ongoing for pre-clinical test in worldwide (Table 2). These data emphasize the advantage of anti-Tim3 antibodies on single safety, superior efficacy and favorable pharmacokinetics with co-blockade of PD-1 pathway in clinical trials.

The complex biological characteristic endows the diverse functions of Tim-3 linked to tumorigenesis, immune regulation, chronic viral infections, and inflammation. In addition to its expression on T cells, Tim-3 is expressed by dendritic cells, macrophages, mast cells, NK cells, and myeloid-derived suppressor cells.

Furthermore, the role of Tim-3 in these non-T immune cells needs to be investigated. Besides, Tim-3 has no inhibitory signaling motifs in its structure, unlike other ICRs, provoking the debate whether Tim-3 functions as an inhibitory receptor or a stimulatory receptor. Another concern is that raised plasma sTim-3 level is linked to patients with membranous nephropathy [76], GVHD, and diverse malignancies, indicating that sTim-3 may serve as a diagnostic biomarker. Furthermore, sTim-3 may also function as a molecular sink to hinder the interaction of Tim-3 and its ligands. Nevertheless, the exact function of sTim-3 is yet to be studied.

Remarkably, blockade of the Tim-3 pathway by anti-Tim3 antibodies indicated its therapeutic potential in cancer, especially noteworthy, is its co-blockade with other ICRs such as PD-1 during the clinic trials. Yet, additional efforts may be needed to further investigate the molecular mechanism of Tim-3 function in diverse contexts for successfully developing the novel agent targeting Tim-3.