Gliomas are characterized by a high degree of vascularization. These blood vessels are necessary for tumor growth, as they are involved in providing nutrients and removing metabolic waste. Over the past few years, some studies have demonstrated that TNC is mainly found in the perivascular niche of gliomas (72, 108), particularly in hyperplastic blood vessels (113, 114). The microvessel endothelial cells (114), pericytes (115) and smooth muscle cells (116) all have the capacity to synthesize and release TN/TNC. In GBM, perivascular TNC is related to microvascular density and vascular endothelial growth factor (VEGF) expression (108). In addition, in melanoma, TNC can regulate the expression of VEGF and affect angiogenesis in tumors (117). Proteome and immunohistochemical comparisons between tissues with physiological angiogenesis and GBM angiogenesis indicated aberrant upregulation of TNC in the latter (118). Therefore, TNC may play an important role in glioma neovascularization ( Figure 3 ). Recently, Rosińska et al. (119) reported that gliomas deploy diverse neovascularization means to meet a dedicated blood supply, including co-option, angiogenesis, vasculogenesis, vascular mimicry (VM), and transdifferentiation of GSCs. First, TNC is linked to angiogenesis. Zagzag et al. (120) discovered that TNC acts as a permissive substrate that promotes microvascular cell migration in vitro by triggering focal adhesion kinase (FAK) phosphorylation in endothelial cells. In TNC-knockdown GBM-neurosphere intracranial xenografts, tumor blood vessel density was lower, while the lumen was enlarged as compared to the control (121). It is important to mention that Rupp et al. (122) demonstrated the dual angiogenic effects of TNC in GBM ( Figure 3 ). The direct contact between endothelial cells and TNC represses actin polymerization, impairs YAP signaling, and downregulates pro-angiogenic factors, consequently negatively influencing endothelial cell proliferation, survival, sprouting, and tubulogenesis. TNC also induces GBM cells to secrete pro-angiogenic factors, such as ephrinB2, a soluble molecule released by MMPs and ADAM10/17 and enhances endothelial cell tubulogenesis. These opposing effects are reminiscent of the cell-specific functions of TNC. Another study suggested that direct contact between the TNC-rich glioma matrix and endothelial cells could induce endothelial detachment, anoikis, selection of a highly proliferative phenotype, and defective tubulogenesis in vitro, whereas higher FN : TNC ratios reversed these effects (123). In addition to glioma, TNC also promotes cellular processes involved in angiogenesis in fibrovascular membranes in eyes with proliferative diabetic retinopathy (124) and affects colitis-associated cancer angiogenesis through interaction with integrin αvβ3 (125). Second, TNC can also promote vasculogenic mimicry ( Figure 3 ). Cai et al. (126) showed that TNC activates the AKT/MMP2/MMP9 axis and further promotes VM in glioma, which is similar to the findings of Kang et al. (127), who reported that TNC-knockdown suppresses this process in gastric cancer by inhibiting the ERK-mediated epithelial-to-mesenchymal transition (EMT). The third process involves the mechanism of TNC-induced transdifferentiation of GSCs ( Figure 3 ). Angel et al. (128) demonstrated the mechanism by which an autocrine TNC–ephrinB2–ephrinB4 signaling pathway supports GSC differentiation into endothelial cells. Taken together, GBM neovascularization correlates with multiple complex processes, in which TNC plays a vital role. In addition to angiogenesis, VM, and transdifferentiation of GSCs, it is necessary to explore additional functions of TNC in co-option or vasculogenesis in future.

The TME is composed of heterogeneous cell types, including endothelial, stromal, multiple immune, and tumor cells (25). As an ECM glycoprotein, TNC has been shown to exert different effects on different cells during tumor proliferation. To date, little information has been available on the TNC-mediated signaling pathways involved in cell proliferation. As mentioned above, TNC appears to play a dual role in endothelial cells in glioma (122, 123). Furthermore, the number of microglia in control and TNC-knockdown tumors was not significantly different in xenografts (121). In contrast to microglia, T-cell proliferation can be suppressed by TNC-containing exosomes produced by GSCs (129). Many studies have reported contradictory results regarding the effects of TNC on glioma cells ( Figure 4 ). Initially, we observed that TNC expression is associated with a high proliferation index (108, 109). Since then, some of the mechanisms underlying the glioma cell proliferation have begun to be revealed. Huang et al. (130) found that TNC was able to bind to FNIII13 in the heparin-binding site II and interfere with cell binding to FN by syndecan-4, thereby leading to human glioma proliferation ( Figure 4 ). Martin et al. (131) proposed that TNC promotes the growth of tumor cells by inducing the expression of 14-3-3 tau. Differential RNA expression analysis has revealed that some TNC-mediated growth-promoting signaling pathways are activated in glioma cells (132). Other in vivo and in vitro experiments have also reported the proliferative effects of TNC (128, 133, 134). Moreover, the growth of GSCs is related to TNC-activated NOTCH signaling (87). Nevertheless, the opposite phenomenon was described in another study. TNC failed to affect GBM neurosphere cell growth in vitro, whereas TNC-knockdown enhanced tumor cell proliferation in vivo (121). One possible explanation for these different functions is that different domains of TNC have different proliferative properties (27, 61). The EGF-like repeats and FBG region can bind to the EGF receptor and integrinαvβ3, respectively, contributing to growth. Certain alternatively spliced domains, such as AD1, AD2, and C, are also responsible for proliferation (27, 135). However, a fragment composed of all FNIII domains induced a reduction, whereas the integral TNC molecule led to an increase in glioma cell proliferation ( Figure 4 ) (61). Another explanation is that the absence or decrease of a specific receptor compromises the proliferative effect. The lack of PDGF-Rβ in the glioma cell lines U87, U251, and GL261 resulted in the attenuation of TNIIIA2-related proliferation (136). Recently, Fujita et al. (136–138) showed that TNIIIA2 derived from TNC plays a vital role in the malignant behavior of glioma cells. They presented a positive spiral loop in which the peptide TNIIIA2 promoted PDGF-dependent proliferation by activating integrin β1 in GBM cells expressing PDGF-receptor β. The consequent upregulation of PDGF stimulated TNC expression, which induced MMP-2-mediated TNIIIA2 liberation (136). Further research is needed to explore the relationship between other domains and cell proliferation.

A major contributor to the poor prognosis of GBM patients is the invasive nature of tumor cells, which readily invade healthy brain tissues. This invasive ability of GBM cells makes maximum safe resection nearly impossible, thereby causing therapy-resistant tumor relapse, even in distant parts of the contralateral hemisphere (139). Generally speaking, glioma cell invasion involves multiple molecular mechanisms, including ECM components, adhesion proteins, proteinases, and cytoskeletal changes (139).

A good balance between assembly and disassembly of cell-matrix adhesion sites is a key determinant of cell adhesion and migration. Appropriate adhesion is essential for migration, yet too strong an adhesion has an adverse effect on cell motility. Hence, intermediate adhesion may be most beneficial for cell migration (140). To date, TNC has been found to play complex modulatory roles in glioma cell adhesion ( Figure 4 ). Briefly, the dual effects of both pro- and anti-adhesion are dependent on the cell type, cellular context, different receptors, and the structure itself (28, 43, 44). On the one hand, TNC is not a good adhesive substrate for glioma cells, and it can impair fibronectin-mediated cell adhesion and spreading (130). However, some studies have found that TNC acts as a surface-coating molecule that supports cell adhesion (141), perhaps in an Arg-Gly-Asp (RGD)-dependent manner (142). Additionally, adhesion of glioma cells to tenascin is mediated by different coating concentrations and integrin receptors (143). Thus, TNC appears to facilitate regulation of glioma cell adhesion turnover.

Furthermore, many reports have demonstrated that TNC expression is related to the infiltrative phenotype of many tumors, including gliomas (43, 44, 89, 144, 145). Hirata et al. (146) revealed that endogenous TNC enhanced glioma invasiveness through compositional changes in the surrounding brain parenchyma. TNC not only directly supports glioma cell migration, but also augments this role, mediated by FN, through interaction with integrin α2β1 (147). In addition, several mechanisms that are involved in the regulation of TNC expression, such as IL-33/NF-κB/TNC (148) and NOTCH/RBPJκ/TNC (84), were found to increase the motility of glioma cells. Some growth factors, including lysophosphatidic acid and PDGF, strongly induce glioma cell migration via actin cytoskeleton remodeling in the TNC microenvironment ( Figure 4 ) (149). Remarkably, although TNC can activate high levels of phosphorylated FAK in endothelial cells, leading to microvascular migration (120), it stimulates low levels of FAK phosphorylation in glioma cells (120, 121). Cell migration is a complex and dynamic process, involving the establishment of polar structures, adhesion formation and disassembly, and formation of protrusions at the front, and contractile structures at the rear of the cell (150). FAK phosphorylation is associated with focal adhesion. Therefore, we speculated that the anti-attachment effect of TNC allows glioma cells to detach from ECM molecules and thereby contribute to migration.

Finally, proteolytic degradation of the ECM is another important cause of glioma invasion. Sarkar et al. used a three-dimensional matrix and revealed that some proteinases are associated with TNC-mediated invasiveness, such as MMP-12 in U178 and U251 glioma cell lines (151, 152), as well as ADAM-9 in glioma patient-derived GSC lines ( Figure 4 ) (153). In addition, in vivo experiments also demonstrated its influence on invasion. In TNC-knockdown xenografts, gliomas were confined to well-defined tumor boundaries, and decreased TNC led to inhibition of glioma invasion (121). Considering the diverse actions of distinct domains of this protein, some studies have focused on alternatively spliced domains. TNIIIA2 enhances the disseminative migration of GBM (136, 138) and confers anoikis resistance (137), but these activities are both abrogated by peptide FNIII14, which inactivates β1-integrins (136–138, 154).

Thus, TNC is a potential enhancer of glioma invasiveness. It may be responsible for an intermediate cell adhesion state. Moreover, the aggressiveness of gliomas is further enhanced in the presence of TNC. Further studies involving different domains and isoforms of TNC are required.

Accumulating evidence has indicated that immune cells fail to function properly in gliomas. By employing immune escape mechanisms, including creating an immunosuppressive microenvironment, gliomas can bypass immunosurveillance and hamper the effect of immunotherapy (155). Recently, it has been proposed that serum TNC could be used as an indicator of the immunosuppressive microenvironment status of low-grade gliomas as well as to predict the efficacy of immunotherapy (156). The effect of TNC on various immune cells has been extensively reviewed elsewhere (28, 157). We have focused on the main findings. Generally, TNC activates innate immune cells, but exerts immunosuppressive effects on lymphocytes, such as T cells ( Figure 5 ). TNC is regarded as an inducer of the neuroinflammatory response in stroke, particularly SAH (56, 157). By interacting with TLR-4 of microglia or macrophages, the FBG domain of TNC leads to the production and release of pro-inflammatory cytokines, including IL-1β, IL-6, and TNF-α (56, 157). In chronic neuroinflammation in the brains of Alzheimer’s disease model mice, TNC deficiency alleviates neuroinflammation and enhances the anti-inflammatory response (69). Hence, it seems reasonable to conclude that TNC has the capacity to induce M1 proinflammatory or anti-tumorigenic phenotypes in macrophages and microglia. However, this phenomenon contrasts with the tumorigenic properties of TNC and its higher expression in high-grade gliomas. One possible explanation is that this dual effect on macrophages is likely to be dependent on the cellular source. In murine models of breast cancer, Deligne et al. (158) found that host-derived TNC enhances antitumor immunity by recruiting proinflammatory macrophages, whereas tumor-derived TNC drives macrophages to produce an immunosuppressive response. Another possibility, proposed by Yalcin et al. (28), is that TNC has different functions in distinct phases of carcinogenesis. TNC plays a pro-inflammatory and anti-tumor role in tumor initiation, and tissue remodeling and tumor-promoting roles during tumor progression. This is similar to the phenomenon in which microglia can inhibit tumor growth and exhibit a tumor-promoting state in the early and late stages of glioma progression, respectively (159). Many studies have illustrated that tumor-associated macrophages (TAMs) are re-educated by glioma cells and show remarkable heterogeneity (160–162). Thus, it is increasingly thought that the M1/M2 dichotomy is an oversimplification and is no longer applicable to gliomas. Due to the complex heterogeneity of TAMs, it is difficult to draw definitive conclusions regarding the role of TNC in TAMs. A study by Ma et al. (86) suggested an intricate relationship between TNC and CD47-mediated macrophage phagocytosis. Decreased CD47, a “don’t eat me” signal, not only recruits more M2-like TAMs, but also upregulates TNC expression, which further facilitates phagocytic ability and secretion of proinflammatory factors. Furthermore, TNC-induced cytokines, such as IL-1β and IL-6, and transcription factors, such as NF-κB and signal transducer and activator of transcription 3 (STAT3), have been implicated in cancer-related inflammation (86, 161, 163). These cytokines or transcription factors also induce TNC expression (7). This positive feedback is consistent with the fact that TNC expression increases with an increase in glioma grade. Hence, a more detailed study may be required to clarify the role of TNC in cancer-related inflammation.

Unlike the innate immunity, the immunosuppressive effects of TNC on T cells have been extensively studied in prostate cancer, breast cancer, and GBM. TNC generates barriers and retains CD8 tumor-infiltrating lymphocytes inside the tumor stroma in human breast cancer (164). TNC expressed on GBM cells decreased the T cell amoeba-like shape formation and paralyzed migration, while transmigration of T cells through the monolayer and ECM of glioma cell lines lacking TNC was obviously increased (165). GSC-derived exosomes carrying TNC attenuate T cell activity by interacting with the integrin receptors α5β1 and αvβ6. More importantly, these TNC-containing exosomes pass through the blood–brain barrier and enter into the circulation, suppressing systemic immune responses in patients with GBM (129). In prostate cancer, Jachetti et al. (166). found that CSCs can migrate early to prostate-draining lymph nodes where they overexpress TNC, inhibiting T cell activation, proliferation, and cytokine production, thereby overcoming immune surveillance. In addition, the relationship between TNC and autophagy was uncovered. TNC has been implicated in the suppression of T-cell antitumor responses caused by autophagy defects. Blockade of TNC sensitizes the efficacy of immune checkpoint inhibitors in autophagy-impaired triple-negative breast cancer (167). This finding may present prospects for anti-PD1/PD-L1 treatment of gliomas.

EMT is an essential process that confers malignant properties to cancer cells (168). Initially, EMT was defined by changes in cell morphology and behavior, such as repression of the existing epithelial characteristics and gain of mesenchymal properties. In recent years, we have found that activation of EMT programs endows cancer cells with additional properties beyond enhanced motility and invasiveness, such as cancer cell stemness, local immunosuppression, increased drug resistance, changes in genomic stability, and prevention of senescence (168). Accumulating evidence has suggested that TNC can promote “EMT-like” changes in different cancers, including breast cancer (169, 170), gastric cancer (127), colorectal cancer (171), pancreatic ductal adenocarcinoma (172), and nasopharyngeal carcinoma (173). However, the EMT observed in gliomas seems to differ from the classical EMT and therefore the term “glial to mesenchymal transition” has been proposed (174). Notably, mesenchymal GBM shares certain common features with gliomas that undergo “EMT-like” changes. They both exhibit a more aggressive nature and resistance to many treatments, and have the same master regulators, such as STAT3 (175–177). Thus, TNC-induced “EMT-like” changes in gliomas require further exploration.

Based on transcriptional signatures, GBMs can be classified into three subtypes: The Cancer Genome Atlas (TCGA) proneural (TCGA-PN), classical (TCGA-CL), and mesenchymal (TCGA-MES) subtypes (178). By applying single-cell RNA-sequencing (scRNA-seq), malignant cells in GBM converge to a limited set of four cellular states: neural-progenitor-like (NPC-like), oligodendrocyte-progenitor-like (OPC-like), astrocyte-like (AC-like), and mesenchymal-like (MES-like) (179). These subtype transitions have become increasingly important, particularly the “proneural-to-mesenchymal transition” (PMT), which is analogous to EMT. This phenomenon is associated with requiring a more aggressive treatment pattern and with resistance to treatment. After reviewing the related literature, we believe that there is an underlying relationship between TNC and MES GBM. Angel et al. (128) found that TNC is overexpressed in MES GBM and that this is positively correlated with the expression of MES markers in TCGA. They further confirmed the role of TNC in regulating glioma cell plasticity in MES GBM. Moreover, Miroshnikova et al. (180) demonstrated that hypoxia-inducible factor 1α (HIF1α) directly regulates TNC expression and that TNC modifies ECM stiffness and mechano-signaling. Interestingly, another study implied that stiff tenascin-rich stroma enhances integrin mechano-signaling to induce PMT in GBMs (181). Additionally, macrophages were found to interact with receptors (OSMR or LIFR) complexed with GP130 on GBM cells via macrophage-derived oncostatin M (OSM), thereby activating STAT3 and inducing the transition of GBM cells into MES-like states (182). As mentioned above, TNC is considered an activator of STAT3 (86, 161, 163). Hence, it may be worthwhile to study the significance of TNC in mesenchymal GBM further.

The current treatment of glioma, particularly GBM, remains a challenging problem with poor prognosis. Although many therapeutic options are available, resistance to therapy frequently causes treatment failure and tumor recurrence. Several studies have shown that TNC are likely to be related to this process. Radiation therapy is known to cause tumor cell death triggered by DNA breaks. However, ionizing radiation also induces changes in the TME, leading to an increase in TNC (183). In addition, irradiation-associated inflammation and hypoxia trigger tenascin C expression via TGF-β (80) and HIF-1 signaling (180), respectively. Therefore, it is plausible that radiation therapy can kill tumor cells, while simultaneously providing favorable conditions for tumor relapse.

Elevated expression of TNC affects the response of gliomas to chemotherapy with temozolomide (TMZ). In an in vitro experiment, TNC-knockdown GBM neurospheres were found to be more sensitive to TMZ (121). Another study conducted in vivo and in vitro showed that peptide FNIII14, which can inhibit the effects of TNIIIA2 through inactivation of β1-integrins, increased the susceptibility of GBM cells to TMZ by suppressing O⁶-methylguanine-DNA methyltransferase (MGMT) expression (138). These findings suggest that targeting TNC may augment the anti-tumor efficacy of TMZ.

Immunotherapy is a novel therapeutic strategy that extends beyond radio- and chemotherapy. However, the immunosuppressive microenvironment limits the application of immunotherapy in gliomas. We previously reviewed the immunosuppressive effects of TNC, which further compromise the efficacy of anti-cancer immunotherapy. Recently, TTFields therapy have shown promising prospects. Nothing is currently known about TNC function in TTFields.

In the past, neurogenesis was observed to occur mainly during the developmental period. Adult neurogenesis has been widely described. Various newborn neural cells are continuously generated from NSCs located in two canonical regions of the adult CNS: the SVZ of the lateral ventricle and the dentate gyrus of the hippocampus. These stem cells reside in a specialized environment, known as a niche, which is able to maintain the basic properties of stem cells. TNC, secreted by NSCs and OPCs, is considered to be an important component of the stem cell niche and plays a vital role in NSC development (10). The effect of TNC on adult neurogenesis has also been well-characterized (184, 185). Both in vivo and in vitro assays indicate that TNC-knockout reduces OPC proliferation in the mouse CNS, which is associated with a partial loss of response to PDGF (186). In addition, the fact that TNC-null OPCs exhibit accelerated maturation rates suggests that this molecule contributes to the maintenance of the immature state (52). The underlying mechanism is gradually being elucidated (187). However, there are distinct perspectives on the role of TNC in OPC migration. Some studies have provided direct evidence for migration-inhibiting effects and mechanisms, including the prevention of WNT signaling (188) and modulation of cell–ECM interactions (189, 190). In contrast, other studies have reported that TNC may be associated with the rostral migratory stream (191), expression of proteinases (151–153) and enhancement of WNT signaling (104, 105). Thus, it is plausible to emphasize the complex migratory role of this molecule. Overall, TNC promotes NSC/OPC proliferation, inhibits differentiation, and regulates migration ( Figure 6 ).

Currently, there is a view that SVZ-derived NSCs or OPCs may have potential glioma cellular origins ( Figure 6 ) (30, 192–195). Lee et al. (29) presented direct molecular genetic confirmation from patients and mouse models, showing that NSCs harboring driver mutations in the SVZ migrate to distant brain regions and lead to GBM development via aberrant growth of the OPC lineage. Notably, as shown above, TNC is enriched in the GSC and NSC microenvironments and is responsible for tumor progression and neurogenesis, respectively. Additionally, GSCs share many features and behaviors with NSCs, including common markers (such as nestin and CD133) and self-renewal properties (196). Therefore, it is necessary to explore whether TNC is involved in the malignant transformation of NSCs or OPCs.

Previous studies have indicated that malignant gliomas can arise from neural stem/progenitor cells carrying driver mutations (such as mutations in P53, PTEN, EGFR, and NF1), whereas more differentiated cell types are less likely to undergo malignant transformation (29, 30, 195, 197). However, the causes underlying this phenomenon remain unclear. Interestingly, a relationship between TNC and gene instability has previously been reported (198). One RNA profiling experiment showed that TNC downregulates the expression of cell cycle- and DNA repair-related genes in T98G GBM cells (132). The influence of TNC on cell cycle progression of the SVZ and spinal cord neural stem/progenitor cells has been investigated (199, 200). TNC-mediated enhancement of the proliferative capacity of tumor cells (130) and neural stem/progenitor cells (200) concomitantly increases the likelihood of accumulation of mutations. Apart from the above, other important factors, such as stemness maintenance and migration of stem cells, as well as immunosuppressive effects on T cells, also provide favorable conditions for the transformation to malignant tumors. These findings suggest a possible relationship between TNC and gliomagenesis ( Figure 6 ).