In 2019, Jun Wang et al. discovered for the first time that FGL1 could specifically bind to the inhibitory receptor LAG-3, revealing its immunological function (14). LAG-3 is expressed primarily on activated CD8⁺ T cells, CD4⁺ T cells, Tregs, NKT cells, NK cells, and plasmacytoid dendritic cells (pDCs), with minimal expression in myeloid cell populations (32).

LAG-3, a type I transmembrane protein, is composed of an extracellular region, a transmembrane region, and an intracellular region. The extracellular region of LAG-3 contains four immunoglobulin (Ig)-like domains (D1 to D4), similar to CD4 (33). LAG-3 interacts with its ligands through distinct domains. Specifically, a loop structure within the D1 domain (loop 1, approximately 30 amino acids) binds to major histocompatibility complex class II (MHCII). LAG-3 selectively recognizes the conformationally stable peptide-MHCII (pMHCII) complex (34). LAG-3 forms cis-homodimers through a conserved hydrophobic structure within the D2 domain of its extracellular region, which is crucial for its suppressive activity (35). FGL1 binds to another loop structure (loop 2) within the D1 domain of the extracellular region of LAG-3 through its FD structure, promoting the formation of higher-order oligomers from LAG-3 homodimers on the cell surface. Loops 1 and 2 within the D1 domain of LAG-3 exhibit partially overlapping binding surfaces. This is consistent with preclinical evidence showing that antibodies such as relatlimab and 15011 can block LAG-3 interactions with both FGL1 and MHCII (36). Physiological concentrations of FGL1 (10 nM) can induce the formation of higher-order oligomers from LAG-3 dimers. However, high concentrations of FGL1 (100 nM) do not effectively promote the oligomerization of LAG-3 homodimers, and the underlying mechanism remains unclear; this may be related to the spatial structure of the protein (36). Beyond canonical ligands, galectin-3 and liver sinusoidal endothelial cell lectin (LSECtin) have been suggested to interact with glycans on LAG-3 (Table 1).

LAG-3 engages diverse ligands in a context-dependent manner; MHCII may predominate in lymphoid tissues and autoimmune contexts, whereas FGL1 may play more prominent roles in the liver and MHCII-deficient tumor microenvironment. Importantly, multiple ligands can coexist within the same niche, establishing redundant or synergistic inhibitory axes (Table 1). Although either FGL1 deficiency or LAG-3 deficiency exhibits comparable antitumor effects in mouse models, this does not rule out the existence of additional functional receptors for FGL1 (14). Indeed, Maruhashi et al. reported that stable pMHCII, rather than FGL1, serves as the functional ligand of LAG-3 that suppresses T cells in autoimmunity and antitumor immunity, but their study does not completely exclude the possibility that the FGL1-LAG-3 interaction is required for the stable pMHCII-mediated suppression of antitumor immunity by LAG-3 (34). Notably, the detection of robust FGL1 binding to LAG-3-deficient T cells suggests that FGL1 may associate with molecule(s) other than LAG-3 to suppress T cells. Further analyses are expected to reveal the actual role of FGL1 in regulating T cell function.

LAG-3 lacks classical inhibitory motifs, and its inhibitory function relies on three evolutionarily conserved intracellular motifs, the FSALE motif, the KIEELE motif, and the C-terminal EP-repeat motif. The functions of these domains remain to be fully defined. Upon binding to pMHCII, the FSALE and EP motifs in the intracellular region of LAG-3 transmit two distinct inhibitory signals that suppress T cell activation (37). The KIEELE motif is essential for the expansion and activation of effector T cells and for hybridoma T cells to produce IL-2 (38, 39). LAG-3 migrates to immunological synapses upon activation of CD4⁺ T or CD8⁺ T cells and is positioned adjacent to the TCR-CD3 complex. Its EP motif lowers the local pH at the synapse, causing CD4 or CD8 coreceptor molecules to dissociate from lymphocyte-specific protein tyrosine kinase, thereby inhibiting TCR signaling (40). This process can occur ligand-independently. Nevertheless, evidence shows that the critical role of ligand engagement in facilitating LAG-3-mediated inhibitory function depends on the cytoplasmic FSALE motif (41). The interaction of MHCII or membrane-bound FGL1 (not soluble FGL1) with antigen-presenting cell-mediated ligand engagement triggers rapid LAG-3 ubiquitination, releasing the LAG-3 signaling tail, particularly the FSALE motif, from its association with the cell membrane, consequently unleashing its inhibitory function (42). As a soluble ligand, FGL1 promotes the oligomerization of LAG-3 dimers, inhibiting T lymphocyte proliferation and the secretion of cytokines such as IL-2, interferon-γ (IFN-γ) and tumor necrosis factor-α (TNF-α) (14, 36). The exact mechanisms by which FGL1 affects the intracellular signaling pathways of LAG-3 are still under investigation.

Hepcidin is synthesized and secreted by the liver and plays a crucial role in maintaining iron metabolic balance by regulating the absorption, distribution, and utilization of iron. Recently, FGL1 was shown to be a potent suppressor of hepcidin in vitro and in vivo. FGL1 directly binds to bone morphogenetic protein 6 (BMP6), thereby inhibiting the canonical BMP-SMAD signaling cascade to repress hepcidin transcription. Moreover, the FD domain in the C-terminus of FGL1 is responsible for hepcidin suppression, whereas the N-terminal domain is inactive (29).

Activin receptor-like kinase 5 (ALK5), also known as transforming growth factor-β receptor type I (TGFβR1), is a key receptor of the TGFβ signaling pathway. FGL1 binds to ALK5 on the surface of renal tubular epithelial cells with high affinity, thereby preventing ALK5 ubiquitination and potentiating TGF-β/SMAD signaling to promote renal fibrotic progression. The FD domain (150–312 amino acid region) of FGL1 could interact with ALK5, while the 1–150 amino acid region could not (43). Besides epithelial cells and endothelial cells, ALK5 is also expressed on macrophages and T cells, playing important roles in macrophage inflammation and T cell trafficking (44, 45). It remains unknown whether FGL1 engages ALK5 on these cells to exert immunoregulatory functions.

In addition to LAG-3 and BMP6, FGL1 may have other potential binding molecule(s). Maruhashi et al. reported that the pentameric domain of cartilage oligomeric matrix protein (Comp-FGL1) strongly binds to LAG-3-deficient T cells in vitro, indicating the existence of FGL1-binding molecule(s) other than LAG-3 on activated T cells (34). High concentrations of FGL1 have no effect on the oligomerization of LAG-3 homodimers and may regulate T cells in other ways. Meng-Meng Cao et al. demonstrated that FGL1-binding molecule(s) are present on the surface of human L02 liver cells by coincubating fluorescein-labelled recombinant human FGL1 protein with human L02 liver cells. However, the molecules that interact with FGL1 have not yet been clearly identified. FGL1 interacts with binding molecule(s) in an autocrine and dose-dependent manner, promoting the proliferation of human L02 liver cells (46).

FGL1 has also been shown to induce the phosphorylation of the tyrosine protein kinase Src and then transactivate epidermal growth factor receptor (EGFR) and downstream extracellular signal-regulated kinase (ERK) 1/2 in a ligand-independent manner, promoting the proliferation of normal liver cells (47). In addition, FGL1 can promote hepatic lipid accumulation and insulin resistance (IR) via the ERK1/2 pathway (11, 13). In addition to regulating hepatocytes, liver-derived FGL1 promotes the phosphorylation of c-Jun N-terminal kinase in skeletal muscle cells through the EGFR pathway. This impairs the insulin sensitivity of skeletal muscle cells, leading to IR (5). FGL1, which is also weakly expressed in adipose tissue, can promote lipid synthesis in the 3T3-L1 mouse embryonic fibroblast (preadipocyte) cell line in a manner dependent on ERK1/2-C/EBPβ (12).

Overall, FGL1 performs immunoregulatory functions mainly via its receptor LAG-3 and may be involved in physiological and pathological processes, such as tumor immune escape and autoimmune diseases. Other receptors, such as ALK5, are also expressed in immune cells such as macrophages and T cells, but whether FGL1 regulates these cells via ALK5 remains unclear. FGL1 may act on different cell types through different receptors, and the existence of additional receptors requires further in-depth investigation (Figure 2).

Previous studies revealed that FGL1 is frequently downregulated in HCC because of allelic loss. A reduction in or deficiency of FGL1 enhances HCC proliferation and colony formation and is significantly associated with the degree of tumor differentiation (22). Furthermore, the activation of protein kinase B (PKB/AKT) and its downstream targets of mammalian target of rapamycin (mTOR) signals increases in Fgl1-deficient mice, significantly increasing the incidence of HCC in response to treatment with the chemical carcinogens diethylnitrosamine and low-dose phenobarbital (48). FGL1 may reduce the incidence of HCC by regulating hepatocyte proliferation, survival and metabolism. Exogenous supplementation with recombinant FGL1 protein has no significant effect on the proliferation of HCC tumor cells, suggesting that the intracellular and extracellular effects of FGL1 differ (46).

Since FGL1 has been identified as a major immune inhibitory ligand of LAG-3, its role in tumors has been further investigated. Aberrant transcriptional activation and PTMs lead to elevated FGL1 expression. Studies have shown that high expression levels of FGL1 are significantly associated with poor overall survival in HCC patients (6). FGL1 in HCC tissues is relatively stable, and the FGL1-LAG-3 interaction increases, thus suppressing the antitumor immunity of CD8⁺ T cells and promoting the development of HCC (30, 31). Increased FGL1-LAG-3 interaction in CD8⁺ tissue-resident memory T cells (T_RM) in end-stage HCC may result in CD8⁺ T_RM cell exhaustion, causing tumor immune escape (6, 49).

In addition to being highly susceptible to primary tumors, the liver is also a common site of tumor metastasis, including a wide variety of malignancies, such as colorectal cancer (CRC), melanoma, lung cancer and breast cancer (50). Compared with hepatocytes, tumor cells, including Hepa1-6, MC38 and B16-F10 cells, express lower levels of Fgl1 mRNA and FGL1 protein in vitro (15). Jia-Jun Li et al. reported that FGL1 protein expression in MC38 cells was distinctly elevated in liver metastatic tumor cells compared with cecum termini, suggesting that FGL1 expression is different in vivo and in vitro (8). Although CRC tumor cells express FGL1, our previous study revealed that FGL1 deficiency in mice inhibited CRC liver metastasis and prolonged survival. FGL1 blockade promoted the antitumor activities of CD8⁺ T and NK cells, confirming that hepatocyte-derived FGL1 negatively regulates liver metastasis to a certain extent (15). On the other hand, CRC tumor cell-derived FGL1 facilitates the progression of liver metastasis. Mechanistically, tumor-associated macrophages activate the NF-ĸB pathway by secreting TNF-α and IL-1β in the liver microenvironment and transcriptionally upregulate OTUD1 expression in CRC tumor cells, which enhances FGL1 stability via deubiquitination, promoting CRC liver metastasis (8). Clinically, high plasma FGL1 levels predict poor outcomes, and reduced immune checkpoint blockade therapy benefits CRC patients with liver metastasis (8). Both hepatocyte-derived FGL1 and tumor cell-derived FGL1 promote liver metastasis, indicating that targeting FGL1 may be a new strategy for liver metastasis.

In addition to its role in liver cancer, FGL1 is involved in the development of several tumors, including lung cancer, melanoma, urothelial carcinoma (UC), stomach adenocarcinoma (STAD), GC and adrenocortical carcinoma (ACC).

The role of FGL1 in cancer is likely contextual. On the one hand, FGL1 plays vital immunoregulatory roles in the tumor environment. FGL1 promotes the secretion of IL-2 and TNF-α by T cells in vitro, simultaneously inducing their apoptosis, suggesting that FGL1 may promote the activation-induced apoptosis of T cells in LUAD (51). Moreover, FGL1 inhibits CD8⁺ T-cell infiltration into LUAD tumors and their ability to release IFN-γ, TNF-α and perforin, promoting immune escape in LUAD (16, 52). Elevated FGL1 levels are associated with worse overall survival (OS) in patients with lung cancer, metastatic melanoma, STAD and GC (14, 53, 54). Elevated LAG-3 expression, particularly in conjunction with FGL1 coexpression, is associated with reduced responsiveness to PD-(L)1 blockade therapy and unfavorable oncological outcomes in advanced urothelial carcinoma (UC) (18). FGL1 is also upregulated in microsatellite instability (MSI)-type GC cells and can bind to LAG-3 on CD8⁺ T cells. This binding inhibits the activity of CD8⁺ T cells, rendering them unable to release cytotoxic substances such as IFN-γ and TNF-α and effectively killing tumor cells (17). Targeting FGL1 facilitates anticancer immunity against MSI-high GC.

On the other hand, FGL1 is involved in the proliferation, migration and invasion of tumor cells. For example, FGL1 is highly expressed in the cytoplasm of LUAD cells and affects LUAD proliferation by regulating MYC target genes and myosin heavy chain 9 (MYH9) (51, 55). FGL1 promotes the proliferation, invasion, migration, and epithelial−mesenchymal transition (EMT) of GC cells (54). Elevated expression of FGL1, which physically interacts with centromere protein M, may regulate the migration and invasion of ACC cells, potentially by influencing collagen organization and cell adhesion (56).

The roles of FGL1 in the tumor environment are complex, and dual immunological and tumor biological regulation by FGL1 should be considered when FGL1 is targeted for tumor therapy.

FGL1 is upregulated in patients with SLE and liver damage (SLE-LD), RA and pSjD, playing important roles in the crosstalk between the liver and other organs and maintaining the balance of immunity and tolerance (20, 21).

The concentration of FGL1 is closely related to the serum IL-6 level in patients with autoimmune diseases. In patients with SLE-LD, IL-6 produced by hepatocytes acts on hepatocytes themselves, further promoting the production of FGL1. In the pSjD mouse model, the serum levels of FGL1 are significantly increased due to increased activation of IL-6 signaling in CD4⁺ T cells of the salivary glands and spleen. Higher expression of the FGL1 receptor LAG-3 is detected in Tregs from patients with SLE-LD. The FGL1-LAG-3 signaling axis inhibits Treg proliferation and impairs the suppressive activity of Tregs by limiting IL-10 secretion. Furthermore, FGL1-LAG-3 signaling promotes the production of pathogenic IL-17A and IL-21 by CD4⁺ T cells in patients with SLE. High levels of FGL1 aggravate disease in patients with SLE-LD (19). However, FGL1 may control the onset of autoimmune pathology in the target organ of pSjD model mice by regulating naïve/memory T cell homeostasis after T-cell activation such that Fgl1-deficient mice exhibit more severe pathological lesions (21).

In a collagen-induced arthritis (CIA) mouse model, FGL1 recombinant protein administered intraperitoneally significantly attenuated T cell-mediated autoimmune disease-related arthritis (57). Moreover, FGL1-overexpressing exosomes (FGL1-Exos) effectively suppressed inflammation scores, joint destruction, and the inflammatory response in an RA rat model (58). The increased levels of FGL1 in patients with RA may be related to liver inflammation or injury. Thus, targeting FGL1 may be a therapeutic strategy for RA, and more experimental basis is needed.

FGL1 has distinct regulatory effects on different immune cell subsets, which may be related to the expression levels of FGL1 and its receptor LAG-3. The outcome of its regulation of Tregs and effector T cell functions determines the development and progression of autoimmune diseases.

Notably, the liver is a unique immune-tolerant organ. Compared with other organ transplants, liver transplants result in minimal rejection. Moreover, graft rejection is significantly reduced if the recipient undergoes a liver transplant from the same donor prior to receiving another organ transplant. The interaction of hepatocytes and immune cells plays an important role in inducing liver transplant tolerance. FGL1 is specifically expressed by hepatocytes, and its expression level is significantly increased after partial hepatectomy (23). T cell-mediated cellular immunity plays an important role in transplant rejection, so it can be hypothesized that FGL1-mediated suppression of T cell function may play a role in liver transplantation.

In a heart allograft model, small extracellular vesicles expressing FGL1 and programmed death ligand 1 (PD-L1) could reestablish immune tolerance and significantly alleviate immune rejection (59). With the occurrence of transplantation rejection, the expression levels of LAG-3 and PD-1 increase significantly in T cells, suggesting the activation of these T cells. However, neither the PD-L1 nor FGL1, was increased in the spleens of transplanted mice; thus, enhancing these inhibitory signals may alleviate immune rejection. Targeting FGL1 may be a therapeutic strategy to reduce immune rejection in organ transplantation.