It has been reported that galectin-3 may be activated in chronic myeloid leukemia (CML), following interaction of leukemic cells in the TME with stromal and mesenchymal stem cells (MSCs) (34–37). A study demonstrated that galectin-3 was upregulated, particularly when leukemic cells were co-cultured with bone marrow stromal cells (BMSCs) in vitro, while its expression was predominant in CML cells (35). Additionally, the expression of galectin-3 was significantly increased when CML cells were co-cultured with MSCs, and the protein expression was the highest during the chronic phase (35). In addition, overexpression of galectin-3 in CML cells promoted CML cell and BMSC proliferation, thereby accelerating the deposition of leukemia cells in the bone marrow. In acute myeloid leukemia, high galectin-3 expression was associated with poor prognosis (38). It has been suggested that galectin-3 supports leukemic cell survival in the TME via the AKT-mediated inactivation of glycogen synthase kinase (GSK)3, which is involved in the anti-apoptotic pathway, pro-cell proliferation cascade, metabolic pathway, and other processes (39–43). In addition, Krause et al demonstrated that galectin-3 was induced when t(1;19)-positive acute lymphoblastic leukemia (ALL) cells were co-cultured with glioma-derived U343 cells. Galectin-3 was considered as ligand of Mer tyrosine kinase and the feedback mechanism between those two elements may mediate the relapse of ALL in the central nervous system (CNS) (44).

The TME, a hypoxic environment, may be regulated by the primary regulator hypoxia inducible factor-1α (HIF-1α), which is known to upregulate the expression of several genes, including galectin-3, to maintain cellular homeostasis and promote cell survival in skeletal tissues (45). Galectin-3 was found to be increased in hypoxic/nutrient-deprived areas from both glioblastoma and mammary tumor tissues (26,46–48). In addition, several studies have demonstrated that the interference of nuclear factor (NF)-κB activation inhibits galectin-3 expression, resulting in cell apoptosis (49). Taken together, the interactions among galectin-3, NF-κB, HIF-1α and common stress conditions in the TME are crucial for tumor progression (Fig. 2).

Galectin-3 plays a key role in promoting tumor-driven immunosuppression. The specific effects of galectin-3 on the innate and adaptive immunity are summarized in Table I (15). A study revealed that co-culture of T cells from the peripheral blood with autologous tumor cells suppressed galectin-3 expression in tumor cells and mediated galectin-3-induced expansion of tumor-reactive T cells (6). In addition, tumor cell-secreted galectin-3 may regulate the polarization of macrophages from the M1 (antitumor) to the M2 (tumor-promoting) subtype, trigger CD8⁺ T-cell apoptosis, and downregulate the expression of TCRs (50). Emerging evidence has suggested that galectin-3 binds to lymphocyte-activation gene-3 (LAG-3) on activated terminally differentiated T cells, and functional LAG-3 is required for galectin-3-mediated T-cell suppression (51). In addition, depletion of galectin-3 was associated with increased activation of the proinflammatory signaling pathways in CD8⁺ T cells (51). Galectin-3 suppressed T-cell function via inducing T-cell anergy, and this effect was rescued by depleting surface galectin-3 (52,53). The mechanism underlying galectin-3-induced immunosuppression is mainly mediated by triggering apoptosis via its binding to antitumor T cells (54,55), and shielding the ligands on the surface of tumor cells from the activated receptors of natural killer (NK) cells (56,57). Furthermore, galectin-3 is a soluble ligand of NKp30, which is expressed on the surface of NK cells and acts as an immunomodulator to mediate immune escape of tumor cells from NK cells (58).

Galectin-3 regulates T-cell function through several mechanisms, including the negative regulation of the TCR-mediated cell response (13). Demotte et al demonstrated that treating CD8⁺ tumor-infiltrating T lymphocytes (TILs) with an anti-galectin-3 antibody could restore their ability of IFN-γ secretion (52). Furthermore, the proliferation of tumor-reactive T cells was improved following treatment with supernatants isolated from galectin-3-depleted cells, indicating an important role of the tumor-secreted galectin-3 in the suppression of T-cell activation (59). CD146/MCAM has been reported to act as a functional binding ligand of galectin-3 on the surface of endothelial cells, and is responsible for galectin-3-induced secretion of metastasis-promoting cytokines (60). Certain cytokines, such as interleukin (IL)-1, IL-6, tumor necrosis factor-α and INF-γ, are associated with metastasis and prognosis in several types of cancer (61). The interaction of galectin-3 with endothelial CD146/MCAM in the circulation resulted in increased secretion of IL-6, granulocyte colony-stimulating factor and other cytokines; therefore, they may exert an important effect on the progression and metastasis of cancer (60).

Extracellular and intracellular galectin-3 exert different effects on lymphocytes; therefore, understanding the function of galectin-3 is complicated (6). Interestingly, the cellular localization of galectin-3 determines whether it exerts apoptotic or anti-apoptotic effects on T cells. It has been reported that extracellular galectin-3 induces apoptosis, whereas intracellular galectin-3 inhibits apoptosis by promoting cell proliferation and stimulating TCR signaling (5). Therefore, extracellular galectin-3 may induce apoptosis of human thymocytes and T cells via directly binding to the glycoprotein receptors CD45 and CD71 (62). On the contrary, overexpression of galectin-3 in the intracellular compartment of Jurkat T cells was associated with the inhibition of apoptosis induced by an anti-Fas antibody and staurosporine (63). In addition, a study revealed that depletion of galectin-3 in CD4⁺ T cells upregulated TCR expression and IFN-γ secretion compared with wild-type CD4⁺ T cells (13).

An increasing number of studies have investigated the expression levels of galectin-3 in different types of cancer, and its expression was found to differ among diverse malignancies (64). Galectin-3 was shown to be upregulated in thyroid, liver, stomach and CNS cancers, and downregulated in breast, ovarian, uterine and prostate cancers (65–67). During tumor progression, galectin-3 is often localized in the cytoplasm, as has been reported for tongue and prostate cancer (64), and its expression was decreased in the nucleus during the transition of tongue tissue from normal to cancerous (64). It was, therefore, hypothesized that the nuclear translocation of galectin-3 observed during tumor progression may be a prognostic factor for patients with tongue cancer (68). A similar research, including 145 prostate cancer patients, reported that galectin-3 was usually not expressed or decreased in prostate cancer compared with normal prostate tissues (64). When galectin-3 was detected in cancer cells, it was always absent from the nucleus and was only present in the cytoplasm (69). In addition, it has been reported that the expression of galectin-3 in the cytoplasm is closely associated with vascular invasion, cell differentiation and tumor progression (70).

Galectin-3 exerts opposite biological effects, depending on its cellular localization; therefore, nuclear and cytoplasmic galectin-3 exert antitumor and tumorigenic effects, respectively (69,71,72). It has been suggested that galectin-3 is involved in several different signal transduction cascades and pro-survival processes, including the Ras, Bcl-2 and Myc signaling pathways (73–75). For example, galectin-3 regulated Bcl-2 and other members of the Bcl-2 family by directly binding to them (76). The expression of galectin-3 between the cytoplasmic and nucleal regions differs in different types of skin cancer. For example, the cytoplasmic galectin-3 expression in cutaneous squamous cell carcinoma was significantly higher compared with that in circumscribed and infiltrative basal cell carcinoma. Furthermore, the immunoreactivity of galectin-3 in the cytoplasm was increased compared with that in the nuclei of non-melanoma skin cancer cells (77). In addition, it has been hypothesized that tumor size is associated with the cytoplasmic expression of galectin-3. The nuclear and cytoplasmic expression of galectin-3 has been considered as an important factor in the malignant progression of non-melanoma skin cancer. Therefore, the expression of galectin-3 was reported to be decreased in the nucleus and increased in the cytoplasm during the transition of normal cells to tumor cells (50). Consistently, melanoma patients with low survival rate exhibited increased cytoplasmic galectin-3 expression compared with its nuclear expression (78).

Notably, several studies have shown that the behavior of tumor-stromal cells, including endothelial cells, immune cells, cancer-associated fibroblasts, myofibroblasts and MSCs, is affected by the extracellular expression of galectin-3, whereas it has been found that these cells may also secrete galectin-3 (5,79–82). Upregulation of the galectin-3 expression increases the ability of cancer cell migration and invasion in several tumors, including breast, melanoma, lung, sarcoma, gastric cancer and CML (12,35,82–85). In addition, galectin-3 interacts with ECM glycoproteins, such as fibronectin, collagen IV, elastin and laminin, which play pivotal roles in cell migration (86–89). Studies have shown that galectin-3 also interacts with epidermal growth factor receptor (EGFR) to induce its phosphorylation and re-localization from the membrane to the cytoplasm. In the case of colon cancer cell migration, extracellular galectin-3 may bind with EGFR to affect EGFR dynamics (90). Researchers demonstrated that galectin-3 exhibited an increased affinity to β-1,6-N-acetylglucosamine branched glycans. This interaction mediated the binding of galectin-3 to several types of glycoproteins and glycolipids on the cell membrane, including carcinoembryonic antigen, mucin-1 and glycosylated transmembrane tyrosine kinase receptors of EGF (91,92). The aforementioned findings suggest that galectin-3 may play multiple roles in regulating cell-matrix and cell-cell interactions in cancer.

In addition, it has been reported that tumor-secreted galectin-3 is involved in angiogenesis via binding through carbohydrate recognition, thus affecting endothelial cell behavior and regulating capillary formation during tumor progression (79). The mechanism underlying galectin-3-mediated angiogenesis has been associated with the binding of galectin-3 with αvβ3 integrins on endothelial cells to induce aggregation of integrin clusters and the activation of several signaling pathways. As a result, galectin-3 may affect the angiogenic activity of vascular endothelial growth factor and basic fibroblast growth factor, as well as the promotion of focal adhesion kinase phosphorylation (93).

Due to the immunosuppressive effects of galectin-3, its role in promoting tumor invasion, migration and angiogenesis in the TME has been attracting increasing attention. Therefore, galectin-3 is considered as a potential target for the clinical therapy of cancer. Although the pre-clinical study of a clinical grade galectin antagonist (GM-CT-01) is still at an early stage, a report demonstrated that this antagonist restored CD8⁺ T-cell function, suggesting that this compound may be an effective approach to cancer therapy (94). Furthermore, the effects of galectin-3 antagonists, combined with immune checkpoint inhibitors or T-cell agonists, were investigated to reveal their potential role on enhancing antitumor immunity and promoting regression of solid tumors (95). Preclinical studies demonstrated that treatment with a galectin-3 inhibitor, namely GR-MD-02, a carbohydrate-based drug that binds to galectin-3, promoted antigen-specific T-cell proliferation in patients with advanced cancer (95,96). In addition, GR-MD-02 combined with an irritant (anti-OX40) improved the survival rate of MCA-205 sarcoma, 4T1 breast cancer and transgenic adenocarcinoma of mouse prostate cell (TRAMP-C1) models (95,97). In addition, GR-MD-02 attenuated liver pathological changes, collagen deposition and fibrosis in mice with non-alcoholic steatohepatitis (96). The combination of GR-MD-02 with anti-OX40 treatment also reduced lung metastasis in a 4T1 breast cancer model (98). The successful application of lectin inhibitors indicates that these inhibitors may represent a potential promising approach to cancer therapy.

Differentiated or undifferentiated normal cells rely heavily on the glycolytic pathway to generate energy. Glycolysis refers to the anaerobic conversion of glucose to pyruvate through a series of intercellular enzymatic reactions to produce adenosine triphosphate (ATP), a high-energy phosphate compound (99). The Warburg effect describes a type of mitochondrial dysfunction, where cancer cells do not allow pyruvate to enter mitochondria; instead, lactate dehydrogenase (LDH) enters mitochondria to degrade pyruvate to lactic acid, which in turn enters the Cori cycle (99,100). In tumor cells, galectin-3 overexpression was associated with HIF-1α and p53 activity (21,101), and enhanced phosphoinositide 3-kinase (PI3K) signaling to promote GLUT1-mediated aerobic glycolysis in tumor cells (28,32,102). In addition, galectin-3 promoted RAS and extracellular signal-regulated kinase (ERK) 1/2 activation to induce GLUT1 expression and the activity of hexokinase, phosphofructokinase and LDHA (103–106).

It has been reported that galectin-3 is involved in maintaining mitochondrial homeostasis (107). Therefore, in ovarian cancer, cisplatin promoted the release of cytochrome c and mitochondrial reactive oxygen species in cells with galectin-3-silencing. However, the effect of cisplatin was attenuated following galectin-3 overexpression (108). Inhibition of galectin-3 expression in colorectal cancer cells reduced epirubicin-induced ATP-binding cassette transporter expression and activated the mitochondrial apoptotic pathway (33). Galectin-3 has also been suggested to be associated with pivotal regulators of mitochondrial metabolism, such as AMP-activated protein kinase and peroxisome proliferator-activated receptor, two indicators of fatty acid oxidation in mitochondria that regulate metabolic balance in tumor cells (109,110).

Galectin-3 was found to be significantly upregulated in human glioblastoma T98G cells under conditions of hypoxia and nutrient deprivation (48). Consistently, overexpression of galectin-3 enhanced T98G cell survival and adaptation viability. These findings suggested that galectin-3 mediated tumor progression via promoting angiogenesis and maintaining homeostasis in the TME (48,111). In addition, previous studies in melanoma cells have demonstrated that extracellular galectin-3 activates the p38 mitogen-activated protein kinase pathway, thereby inducing the expression of matrix metalloproteinase (MMP)9, which in turn provides nutritional support for tumor angiogenesis (32,112). Therefore, galectin-3 overexpression may be considered as an adaptive metabolic mechanism of the tumor to maintain cellular viability and homeostasis under conditions of TME stress induced by hypoxia and nutrient deficiency.

It has been reported that the AKT-mediated PI3K signaling pathway is involved in the regulation of GLUT1, glucose uptake and phosphofructokinase activity, thus affecting cell survival, cell cycle progression and therapeutic outcome (99,113–117). Furthermore, galectin-3 has a high affinity for and is cross-linked with β1,6-GlcNAc-branched N-glycans and glycoproteins to form molecular complexes on the cell surface and ECM, thus affecting the distribution of glycoproteins and cell signal transduction (66). Kariya et al demonstrated that the reactivation of PI3K mediated by β4-integrin N-glycans was inhibited following treatment with a neutralizing antibody against galectin-3 (118). Elad-Sfadia et al revealed that galectin-3 was required for the RAS-induced PI3K/AKT activation in response to growth factor stimulation (11). Additionally, galectin-3 increased β-catenin expression and accumulation in the nucleus, thereby enhancing Wnt signaling in human colon cancer cells via regulating GSK-3β phosphorylation/activity through the PI3K/AKT signaling pathway (39).

NF-κB serves an important role in inducing pro- inflammatory cytokines in several types of cancer (119,120). Emerging evidence has suggested that HIF-1α, NF-κB, caveolin-1 and TP53-inducible glycolysis and apoptosis regulator acted as inducers of cancer-stroma metabolic coupling via modulating oxidative stress and autophagy (121). Upregulation of galectin-3 in a hypoxic microenvironment relies on transcription factors such as HIF-1α and NF-κB (48,122). Nucling, an apoptosis-associated molecule, downregulated galectin-3 mRNA and protein expression via mediating the nuclear translocation of NF-κB/p65 (Fig. 3). Therefore, nucling-deficient cells were resistant to pro-apoptotic stress, whereas the expression of galectin-3 and the incidence of inflammatory injury were increased in mice lacking nucling (49). Galectin-3 also promoted IL-8 transcription and secretion via NF-κB signaling in pancreatic stellate cells; however, treatment with a NF-κB inhibitor and integrin-linked kinase (cdp33) completely inhibited the galectin-3-mediated transcriptional activities of NF-κB and IL-8 (123).

Several signaling pathways have been found to be involved in the metabolic process in tumors. Overexpression of galectin-3 in the hypoxic TME was considered to regulate tumor cell migration, invasion and adaptability (32). In addition, the Wnt/β-catenin pathway was found to be involved in the regulation of cell migration via MMPs (124). Shimura et al demonstrated that galectin-3 interacted with the β-catenin/TCF complex and was co-localized with β-catenin in the nucleus, thereby regulating the transcriptional activity of transcription factor 4 (TCF4) in breast cancer cells (125). Consistently, Song et al reported that galectin-3 mediated β-catenin expression and TCF4 activity by regulating GSK-3β phosphorylation and activation via the PI3K/AKT pathway in colon cancer cells (39). Additionally, downregulation of galectin-3 resulted in reduced phosphorylated (p)-AKT and p-GSK-3β expression, and increased GSK-3β activity, thus mediating the phosphorylation of β-catenin. This effect was considered as a critical step for the recognition of β-catenin by the F-box protein β-Trcp (126). These findings indicated that galectin-3 participates in multiple signaling pathways to regulate tumor metabolism.

Lung cancer is one of the most common types of cancer worldwide, and non-small cell lung cancer (NSCLC) accounts for ~80% of lung cancer cases (127). The expression of galectin-3 varies among different types of lung cancer. For example, in small-cell lung cancer, galectin-3 is not expressed or downregulated compared with NSCLC, in which galectin-3 is upregulated in the majority of cases (128). mRNA microarrays revealed that forkhead box D1 (FOXD1) was associated with poor prognosis and lung cancer cell proliferation (129). FOXD1 has been shown to promote lung cancer cell invasiveness via its binding with ERK1/2 and targeting galectin-3 (95). In addition, it has been demonstrated that galectin-3 regulates the expression of FOXD1 via the integrin-β1/ERK/E26 transformation specific-1 cascade, which in turn mediates the formation of a positive ring between FOXD1 and galectin-3 and promotes lung cancer invasiveness (130). A study suggested that the expression of galectin-3 may be a potential biomarker for predicting NSCLC recurrence after radical resection (131). However, the level of galectin-3 in the serum had no prognostic value in NSCLC, and no significant correlation was observed between NSCLC and galectin-3 serum levels (131). Knockdown of galectin-3 in NSCLC cell line-derived spheres decreased the expression of stemness-associated genes, suggesting that galectin-3 may play a synergistic role by interacting with β-catenin and increasing the transcriptional activity of downstream stemness-associated genes. Furthermore, cells lacking galectin-3 were less invasive, more vulnerable to chemotherapy, and inefficient in initiating tumor formation (132).

In NSCLC, galectin-3 not only mediates the malignancy of cancer cells, but also attenuates the effect of immune cells on inducing tumor cell evasion from the immune response. It has been demonstrated that the intracellular expression of galectin-3 and galectin-1 in tumor cells may block apoptosis; however, the extracellular galectins within the TME induce T-cell apoptosis via binding with CD45 and CD7 on the surface of T cells, and exacerbating the immune escape of tumor cells (133). In particular, the galectin-3 multivalent N-glycan complex impaired TCR clustering on the T-cell surface and increased the agonist threshold for TCR signaling (133). In fact, molecular interactions between T-cell surface glycans and certain galectins are functionally capable of regulating T-lymphocyte death and inflammatory responses (25,133–136). Antigen-presenting cells (APCs) and macrophages also play an important role in establishing immune cell homeostasis (137). Significant changes in the glycan chain have been identified during dendritic cell maturation in order to regulate the binding of specific galectins to mature or immature APCs (138).

A recent meta-analysis has suggested that galectin-3 may be considered as a potentially useful immune marker for distinguishing patients with papillary thyroid cancer (PTC) from those with non-PTC (139). PTC patients with positive galectin-3 expression are prone to lymph node metastasis (139). A study compared the serum galectin-3 levels between patients with thyroid cancer and healthy individuals. The results revealed that the serum galectin-3 levels in patients with thyroid cancer were significantly higher compared with those in patients with benign thyroid lesions or healthy controls (140,141). PTC and papillary thyroid micro-carcinoma (PTMC) are the most common types of thyroid malignancies (142,143). However, distinguishing PTC and PTMC from thyroid papillary hyperplasia is challenging due to tumor heterogeneity (144,145). Furthermore, the expression profiles of galectin-3, cytokeratin 19, CD56, thyroid peroxidase and BRAF mutations are commonly used for the diagnosis of PTC and PTMC (145–148).

It has been suggested that the galectin family plays an important role in Ras membrane anchoring and Ras-mediated cell transformation (11,149). Ras proteins (H-Ras, K-Ras and N-Ras) are important members of the GTPase family, which regulate cell differentiation, proliferation and cell death (150). Ras mutations are known to be involved in 25–30% of all human cancers (151,152). In addition, galectin-3 interacts with oncogenic Ras proteins, preferentially with K-Ras, to promote the activation of important signaling cascades, including serine/threonine kinase (RAF1), PI3K and Ras signaling pathways, and to regulate gene expression at the transcriptional level (11). A study demonstrated that the combination of galectin-3 inhibitor, S-trans,trans-farnesylthiosalicylic acid and modified citrus pectin was able to induce cell cycle arrest and apoptosis, thereby inhibiting the growth of anaplastic thyroid carcinoma in vitro and in vivo (153–155).

Melanoma is the most aggressive skin cancer and is considered as a highly immunogenic tumor (156). Over the past few years, studies have identified the characteristics of progressive biomarkers and their underlying mechanisms based on deep research on the invasion and chemoresistance abilities of melanoma cells, and the association between galectin-3 expression and melanoma pathogenesis (59). The results demonstrated that, compared with benign nevus, the galectin-3 expression levels were increased in thin primary melanomas. However, this expression pattern was lost during tumor progression, and galectin-3 expression was decreased in both thick and metastatic melanomas (70,157). Other studies indicated that the expression of galectin-3 in melanocytes and melanoma cells treated with a mutated BRAF inhibitor, vemurafenib, exerted a pivotal effect on reducing autophagic activity and determining cell fate (158,159).

It was previously demonstrated that the expression levels of galectin-3 and its nuclear:cytoplasmic ratio was higher in metastatic lesions compared with that in primary melanoma lesions. Additionally, an association between the nuclear expression of galectin-3 and prognosis was proposed (78). Mourad-Zeidan et al demonstrated that melanoma cells lacking galectin-3 expression exhibited reduced tumorigenic potential and decreased expression of tumor markers (160). However, other studies reported the opposite result. Therefore, a study using a xenograft melanoma model constructed with human melanoma cell lines demonstrated that the expression of galectin-3 was upregulated in thin primary melanoma lesions compared with benign pigmented skin lesions or metastases, and was negatively correlated with cell invasiveness (2). It was also suggested that, in more advanced melanoma lesions, attenuated galectin-3 expression may be associated with high risk of tumor metastasis. Related results indicate that melanoma cells may separate from the basement membrane and enter the circulation via attenuating their interaction with the ECM (2).