T cell immunoglobulin and ITIM domain (TIGIT), also known as WUCAM, Vstm3, or VSIG9, is a novel IC with potential to enhance clinical outcomes in patients [21, 22]. TIGIT was identified in 2009 as a member of the poliovirus receptor-related (PVR)-like protein family and has been shown to play a central role in tumor immunity and is a promising target in immunotherapy [21, 23]. Clinical trials of some mAbs that target TIGIT did not reproduce the promising preclinical data; in fact, administration of the anti-TIGIT mAbs alone often was associated with severe treatment-related adverse events (TAREs) and low objective response rate (ORR) [24, 25]. As a result, there has been a shift in focus from anti-TIGIT mAbs monotherapy to ICs co-inhibition and other modalities, which could harness the potential therapeutic value of targeting TIGIT. This review will first discuss the mechanisms underlying the inhibitory effect of TIGIT on tumor-related immunity and combination immunotherapeutic strategies, then discuss current approaches targeting TIGIT beyond mAbs monotherapy.

TIGIT is expressed by various types of T cells, including activated CD4 T helper cells, CD8 cytotoxic T cells, γδ T cells and T regulatory cells (Tregs), but not on naive T cells [22, 26, 27]. TIGIT is also not expressed on CD11c dendritic cells, CD68 macrophages, or CD20 B lymphocytes [28,29,30]. A study compared TIGIT and PD-1 expression patterns across healthy lymphoid tissues, inflamed tissues, and tumor tissues. Results revealed co-expression in over 70% of TIGIT cells with PD-1, while reciprocally, >90% of PD-1 cells co-expressed TIGIT, suggesting a potential for dual inhibitory targeting [28]. It must be noted that the same cell type may express different levels of TIGIT depending on cellular localization. Greater than 95% of CD4 T cells in lymphoid follicles expressed TIGIT relative to only <50% of CD4 T cells residing in interfollicular compartments [28]. Interestingly, TIGIT expression levels in peripheral blood and tumor tissues in patients with primary breast cancer (PBC) seem to be related to the age of the patient, but not the tumor size or lymph node metastasis [31]. Tumor-infiltrating lymphocytes (TILs) were reported to express significantly higher levels of TIGIT than peripheral T cells [32, 33]. Some investigators have suggested that TIGIT overexpression could be closely associated with T-cell exhaustion, occurring in advanced tumor stages and following cancer cell antigen exposure [34, 35]. TIGIT expression levels also correlate with poor prognosis in many tumors, suggesting an important role in tumor progression, invasiveness, and/or metastasis [36, 37]. TIGIT knockout studies have reported a reversal of T and NK cell exhaustion coupled with enhanced anti-tumor immunity [38,39,40].

Besides its ability to inhibit tumor killing by effector T and NK cells, TIGIT is involved in regulating the function of Tregs. In vitro positive stimulation of Tregs expressing TIGIT can upregulate interleukin 10 (IL-10) and fibrinogen-like protein 2 (FGL2), which are both immunosuppressive factors inhibiting T cells [41]. Previous work has also shown that TIGIT mediates inhibitory signals by suppression of IFN-γ production and downregulation T-bet, a transcription factor (TF) influencing Treg function in type 1 inflammatory responses, which is regulated by IFN-γ. TIGIT also promotes the nuclear localization of FOXO1, which is a broadly expressed TF and important for NK/CD8 T-cell activity and differentiation of Tregs [42]. This suggests that TIGIT-delivered signals promote the suppressive function of Tregs on effector T cells. Unlike PD-1/CTLA-4, inhibition of TIGIT in tumor therapy restores the immune function of T cells through multiple pathways [43].

The extracellular structural domains of TIGIT share homology with other protein receptor members of the family, such as CD155, CD96, CD226 and nectin-4 (PVRL4) [21]. The main ligands for TIGIT are CD155 (PVR), CD112 (PVRL2, nectin-2), CD113 (PVRL3, nectin-3) and nectin-4 (PVRL4) [21, 25, 43]. The interaction between TIGIT and its ligands is shown in Fig. 1. Different ligands have different affinities for TIGIT, with CD155 having the highest affinity. CD226 is another important immune-activating receptor that competes with TIGIT for the ligand CD155, but it has much lower affinity. In addition to this, other ligands of TIGIT may also be involved in interactions, forming a complex immunoregulatory network, which should be considered holistically in immunotherapy. For example, TIGIT-CD112 axis has recently been identified as an important factor that influences immunity against tumor in neuroblastoma [44], although CD112 has a higher affinity for CD112R and has been considered as another axis in immunotherapy [22, 23]. Nectin-4 is currently considered to be an exclusive ligand for TIGIT, as no other members of the PVR-like family have been found to bind to it [45]. The Fap2 protein secreted by F. nucleatum, is a special TIGIT ligand; it can directly interact with human TIGIT and cause NK cell inhibition [46].

TIGIT gene is located on q13.31 of chromosome 3 in Homo sapiens. The mRNA sequence length of TIGIT transcript is 2926 nucleotides, while its coding sequence (CDS) length is 732 bp. A TIGIT molecule consists of 244 amino acids divided into three main regions or domains: the extracellular IgV region with 141 amino acids, the transmembrane domain with 23 amino acids, and the cytoplasmic tail with 80 amino acids [16, 47,48,49,50]. The cytoplasmic tail contains an immunoreceptor tail-tyrosine (ITT)-like motif and an immunoreceptor tyrosine-based inhibitory (ITIM) motif, which are conserved in humans and mice as they are critical for mediating inhibitory signaling [48, 50]. Work by Stengel et al. [47] in 2012 has helped to resolve the structure of the TIGIT/PVR complex and suggested that it is a heterotetramer consisting of CD155-TIGIT-TIGIT-CD155, where characteristic lock-and-key structures were evident. TIGIT's Y113 and CD155's F128 constitute the 'key' of the structure, inserting inside the PVR family conservative AX6G 'lock' of each other. According to this model, dimers of TIGIT and CD155 tend to be interspersed and bound; each paired with a neighboring ligand for the formation of the signaling cluster [47].

When bound to a ligand, TIGIT can directly inhibit human NK cytotoxicity in an ITIM motif-dependent manner. This was demonstrated by generating the amino acid 231-truncated TIGIT molecule [51]. It is worth noting that direct suppression of immune function by TIGIT cytoplasmic tails appears to be present only in NK cells and is not observed in T cells [50]. Most studies addressing direct intracellular signaling by TIGIT were performed using the YTS/TIGIT cell line of human NK cells [21]. Direct inhibitory signaling of TIGIT has also been observed in mice, and this signaling can be accomplished by ITIM or ITT-like motif alone [48]. The ITT-like motif is also important for inhibiting human NK cytotoxicity, as its phosphorylation causes the recruitment of growth factor receptor-binding protein 2 (Grb2), which then recruits SH2-containing inositol phosphatase-1 (SHIP-1), leading to inhibition of P13K and MAPK signaling [52]. Another study showed that ITT-like motif recruits β-arrestin 2, leading to IFN-γ inhibition [53]. TIGIT was also reported to inhibit the phosphorylation of ERK1/2 and ZAP70/Syk in response to CD155 stimulation, thereby suppressing the NK cell function [54].

CD155 can be expressed by many normal human cell types, but it is upregulated in a variety of tumors, including non-small cell lung cancer (NSCLC), melanoma, colorectal cancer and glioblastoma [55,56,57]. It contains an ITIM motif, suggesting the possibility of signaling to the cell interior. Dendritic cells (DCs), as antigen-presenting cells (APCs), express CD155 and can hence engage TIGIT on T or NK cells. DCs can therefore receive signals via their CD155 that upregulate the synthesis of immunosuppressive cytokines (IL-10) and downregulate pro-inflammatory cytokines (IL-12); this diverts DC function towards immuno-tolerance [50]. The interplay between TIGIT and CD155 on APCs can empower TIGIT to indirectly inhibit T and NK cytotoxicity by inducing the anti-inflammatory or immune-tolerance phenotype in APCs. This is further supported by the observation that TIGIT-Fc fusion protein induces CD155-expressing macrophages to upregulate the production of IL-10 and agonistic anti-TIGIT antibody induces upregulation of several TFs, chemokine receptors and Treg effector molecules, including IL-10 or Fgl2 in Tregs, re-coding them to a more inhibitory phenotype [58].

In addition to binding to CD155, TIGIT can inhibit immune responses by competing for ligands with the immune-activating receptor CD226. CD226 has been shown to play an important role in cell contact and TCR signaling [59] and to promote T cell activation [60]. TIGIT binds CD155 with higher affinity, implying that there is a competitive relationship between TIGIT and CD226 in the case of co-expression and that TIGIT preferentially exerts an immunosuppressive effect [22]. Blocking of CD226 was reported to eliminate CD4 T cell functional activation by TIGIT knockdown [61]. Besides its ability to compete with CD226 for the ligands CD155 and CD122, TIGIT was also reported to inhibit CD226 signaling. Specifically, the result of fluorescence resonance energy transfer demonstrated that TIGIT disrupts the dimerization of CD226 on T cell surface, and thereby inhibiting its ability to bind to CD155 [30].

Although TIGIT does not seem to signal to T cells through its cytoplasmic tail, it can still suppress T cell activation through an intracellular mechanism. This is based on the observation that TCR-driven activation signals can be inhibited by TIGIT-mediated signaling [61, 62]. Moreover, TIGIT was reported to downregulate the expression of several genes that are associated with TCR and T-cell activation [62]. Together, these observations suggest that TIGIT delivers inhibitory signals to T cells through specific yet-to-be-identified mechanisms that seem to operate independently of extracellular triggers or CD226 competition.

In addition to CD4, CD8, and NK cells, Tregs also express TIGIT. Demethylation of the TIGIT locus and FOXP3 binding was reported to upregulate TIGIT expression on Tregs, whereas methylation of the TIGIT gene locus can restrict FOXP3 binding. Increased expression of TIGIT on Tregs seems to enhance their immune inhibitory phenotype [21, 23]. Several gene signatures (FOXP3, Helios, neuropilin-1, CTLA-4, PD-1 and LAG-3) are also found to be upregulated in TIGIT Tregs within the tumor microenvironment (TME), turning Tregs to a more immunosuppressive phenotype [23]. The cell types inhibited by TIGIT-expressing Tregs demonstrate specificity; in particular, they inhibit only T helper 1 (Th1) and Th17 cells, but not the Th2 cells [41].