Abstract
Glioblastoma, the most aggressive type of primary brain tumor, portends a poor prognosis, despite current treatment modalities, due to recurrence of disease. Resistance to conventional therapies is caused by both extensive genetic abnormalities and dysregulation of the transcription landscape. A major cause of tumor recurrence, growth, and invasion is the presence of a unique population of cancer stem cells (CSCs) in the tumor and surrounding area. Consequently, CSCs have emerged as targets of interest in new treatment paradigms. The mechanistic target of rapamycin (mTOR), a serine/threonine kinase, forms two multiprotein complexes, mTORC1 and mTORC2, which regulate cell proliferation and migration. The pathogenesis of glioblastoma is largely due to the frequent loss of the tumor-suppressor gene phosphatase and tensin homolog (PTEN), leading to aberrant activation of the mTOR pathway in glioblastoma and its CSCs. Strategies to treat glioblastoma may involve inhibition of the mTOR pathway to target CSCs. Here, we explore the role of mTOR and related signaling pathways in the regulation of glioblastoma stem cells and define their roles as therapeutic targets in the treatment of glioblastoma.
Introduction
Glioblastoma is the deadliest type of brain cancer, occurring with an annual incidence of 3.23 per 100,000 population in the United States and accounting for 57.7% of gliomas and 82% of malignant gliomas (1, 2). Extensive clinical evidence indicates the standard of care for patients with glioblastoma includes maximal surgical resection followed by radio- and chemotherapy with the brain-penetrable alkylating agent temozolomide, accompanied by targeting of discrete genetic or altered pathways (3). Median survival remains 14-16 months, with a survival rate of 5% within 5 years of diagnosis (4-6). Glioblastoma displays considerable inter- and intratumoral heterogeneity due to epigenetic, genetic, protein, and microenvironmental changes that make glioblastoma in one patient different from that in another. This inter-lesion diversity provides select tumors with unique functions that allow them to grow aggressively, survive hypoxic stresses, and resist chemotherapy (7). Glioblastoma remains deadly, owing to a high recurrence rate due to its infiltrative nature, rendering gross total resection virtually unachievable. Despite therapeutic intervention, the outcomes reflect a high recurrence rate of 90% or more, leading to a consistently fatal prognosis (8).
The World Health Organization (WHO) classifies glioblastoma as a grade 4 glioma, according to histopathological features, including high vascularity, pseudopalisading necrosis, and adjacent normal brain tissue invasion. The 2021 (fifth) edition of the WHO Central Nervous System (WHO CNS5) tumor grading system incorporated genetic mutations or alterations and categorized glioblastoma based on genetic markers. These genetic alterations included epidermal growth factor receptor (EGFR) gene amplification, telomerase reverse transcriptase (TERT) promoter mutation, isocitrate dehydrogenase (IDH) mutation, or chromosome copy-number variations, in addition to its histological appearance (9, 10). Therefore, the WHO CNS5 classifications underscore the significance of genetic drivers associated with the development and progression of glioblastoma and are being increasingly utilized to stratify this disease and predict prognosis, as well as in precision medicine (11). With advancements in molecular genetics, the WHO CNS5 has classified adult-type, diffuse gliomas into three subcategories, namely, astrocytoma, IDH-mutant; oligodendroglioma, IDH-mutant and 1p/19q-codeleted; and glioblastoma, IDH-wildtype (9, 10). According to the WHO CNS5, all IDH-mutant diffuse astrocytic tumors are considered a single type (astrocytoma, IDH-mutant), followed by grading as CNS WHO grade 2, 3, or 4. In WHO grade 4 glioblastoma, IDH mutations are also frequently found in secondary disease, which accounts for 73% of clinical cases, whereas they are less frequently seen in primary glioblastoma (3.7%), with the most common mutation occurring at arginine 132 (R132) (12). In secondary glioblastoma, Parsons et al. observed an IDH1 mutation that is also found in more than 70% of WHO grade 2 and 3 astrocytomas and oligodendrogliomas (13). Moreover, tumors devoid of IDH1 mutations frequently possess alterations at analogous amino acid sites of the IDH2 gene. These mutated IDH enzymes have distinct metabolic and epigenetic characteristics and respond differently to treatments with favorable prognoses (14).
Discrete genetic alterations have contributed to the identification of multiple molecular subtypes (15-17), including alterations of EGFR and phosphatase and tensin homolog (PTEN), which lead to activation of the mammalian target of rapamycin (mTOR; mechanistic target of rapamycin) signaling pathway, promoting cancer cell growth, proliferation, motility, and survival (15, 18, 19). mTOR is a serine/threonine kinase that functions by forming two multiprotein complexes, mTOR complex 1 (mTORC1) and mTOR complex 2 (mTORC2) (19, 20). Loss of PTEN, seen in about 36%, causes upregulation of both phosphoinositide 3-kinase (PI3K) and downstream AKT serine/threonine kinase 1 (AKT)/mTOR signaling (15, 21). In addition, extensive genomic studies showed that receptor tyrosine kinase/PI3K activation is present in approximately 86% of glioblastomas (15). Increased AKT/mTOR pathway activity accounts for key features that make glioblastoma such a formidable entity, namely its relentless growth and resistance to therapy, and this means it represents a promising target for therapy (19, 22, 23).
It has been shown that recurrent mutations in IDH1/2 are seen in about 50-80% of secondary glioblastoma, and 5-14% of primary glioblastoma (24-26). Histological features remain identical in both subtypes of glioblastoma. As mentioned earlier, primary glioblastoma is generally characterized by EGFR amplification, PTEN mutation, and p16INK4A deletion and occurs in older individuals, while secondary glioblastoma typically shows mutations in tumor protein p53 (TP53) and retinoblastoma-related gene 2 (RB2) (that regulates cell-cycle progression and also induces astrocyte differentiation in secondary glioblastoma), and occurs in younger individuals, with an average onset age of 45 years. Loss of chromosome 10q is found in both subtypes, with a varying percentage of occurrence (26).
Extensive genomic analysis of glioblastoma was examined by the Cancer Genome Atlas, which recognized the three most prevalent somatic alterations leading to their aberrant signaling pathways. These genetic changes included TP53 (78%), RB1 (87%), and RTK/RAS/PI3K signaling pathways (88%) (15, 17). In one study where glioblastoma classification was clinically relevant, validating The Cancer Genome Atlas data, glioblastoma was clustered into four distinct subtypes, namely, oligodendrocyte precursor type, differentiated oligodendrocyte type, astrocytic mesenchymal type, as well as mixed type. It is important to note that the enhanced expression of nestin, CD44, and podoplanin, with a high glial fibrillary acidic protein score, was associated with shorter survival of 12.8 months in the astrocytic mesenchymal type compared to the oligodendrocyte precursor type that displayed highly positive scores for oligodendrocyte transcription factor 2 (OLIG2), platelet-derived growth factor receptor alpha (PDGFRA), p16, P53, and synaptophysin (SYP), where survival was 19.9 months. These observations stratify glioblastoma classification by genomic-based immunohistochemical analysis (13, 15, 17, 27). Such studies and other genomic and proteomic analysis studies, therefore, support the updated guidelines for the WHO classification of CNS tumors. Mutations of IDH1, O-6-methylguanine-DNA methyltransferase (MGMT), and 1p/19q co-deletion or ATRX chromatin remodeler (ATRX) loss, which have significant diagnostic and predictive abilities, are other markers considered in defining patient prognosis (10).
Cancer Stem Cells (CSCs) in Glioblastoma
Despite a growing understanding of the mechanisms driving its growth, glioblastoma remains a tumor with one of the highest resistances to current therapy. Glioblastoma recurs close to the resected site, as initially shown by computed tomography and pathologic studies of 20 patients (28). The peritumoral region characterizes the extent of invasion of glioblastoma cells, marked by alterations in the components of signaling pathways that regulate invasion, migration, and growth (29, 30). Treatment resistance has been attributed to the presence of glioblastoma stem cells (GSCs) (31). The CSC principle suggests that a small subpopulation of cells plays a major role in cancer progression and recurrence secondary to their innate properties of self-renewal and proliferation (32). As shown in Figure 1A, these CSCs, like neural stem cells (NSCs), display the ability to self-renew, differentiate, and form neurospheres in culture (33, 34). Numerous molecular markers commonly used to isolate and identify NSCs are also present in CSCs (33). The characteristics that distinguish GSCs from NSCs are that they contain many genetic mutations and chromosomal abnormalities, increasing the predisposition of GSCs to form tumors. The origin of CSCs from normal stem cells or from differentiated cells that have acquired the ability to self-renew is debatable (35). GSCs might arise from NSCs or other neural cells by acquiring genetic mutations and displaying self-renewal properties. Furthermore, GSCs isolated from human tumors and cultured in vitro display significant similarities to NSCs as they also express neural stem/progenitor markers, such as nestin, SRY-box transcription factor 2 (SOX2), and OLIG2. In addition, these GSCs are capable of differentiation through cells expressing neuronal or glial markers upon induction (36). Even following appropriate standards of care therapy, resilience and survival of CSC populations allow glioblastomas to reoccur and invade aggressively (37). Supramaximal resection of glioblastoma has shown these CSCs both within the radiographically visible tumor, as well as the peritumoral area (30). In fact, studies have demonstrated that the expression of stem cell marker nestin, together with c-Jun N-terminal kinase-mitogen-activated protein kinase (MAPK), in the peritumoral area of glioblastoma displays prognostic significance (30). Multiple investigations have established the existence of CSC populations in glioblastoma (31, 38, 39). Additionally, it has been demonstrated that CD133+, but not CD133− CSCs possess NSC properties, such as self-renewal and multipotency, and can generate tumors with identical histopathological and genetic features to the original tumor when implanted in brains of immunodeficient mice (38, 40). A study has suggested that tumor-suppressor genes PTEN and TP53 may regulate the self-renewal properties of CSCs (41). The hypothesis that only CD133+ cells have stem cell properties and are tumorigenic remains contentious. Tumor-initiating properties have also been shown to be present in cells not expressing CD133 (42). Further, certain glioblastoma cell lines possessing stem cell-like properties have been shown to be devoid of CD133 (42). In addition, both CD133+ and CD133− GSCs generated multipotent spheres, displaying self-renewal properties (43). Enhanced resistance of CD133+ GSCs was shown both in tumor xenografts as well as in neurosphere cultures, where GSCs accumulated after irradiation, implying reduced apoptosis and phosphorylation of histone H2AX (44). It has been shown that CD133+ GSCs isolated from cell lines cultured from glioblastoma tumors enhanced chemotherapeutic resistance relative to CD133− cells (45). Furthermore, CSCs exhibit enhanced DNA repair and mitochondrial reserve, leading to tumor resistance (44). Resistance to therapy is also attributed to the fact that GSCs can enter a quiescent phase that allows them to escape the effects of treatment (46). These observations affirm that GSCs can be resistant to therapy, as previously described in the model of CSC resistance in leukemia (32, 47).
Role of mechanistic target of rapamycin kinase (mTOR) pathway in cancer stem cell regulation in glioblastoma. (A) Glioblastoma stem cells (GSCs) may develop from neural stem cells or differentiated cells. This process involves a series of genetic changes leading to the acquisition of discrete stem cell-specific transcriptional programs attaining such conversion. Cancer stem cells within the glioblastoma or peritumoral area have the potential to rapidly repopulate a tumor mass. The failure of a cure for glioblastoma may be due to a lack of therapeutic modalities targeting the GSC population. The factors that influence the differentiation of these stem cells can also induce apoptosis or cell death. GSCs can differentiate into neuronal and glial lineages upon treatment with differentiating agents. The possibility of targeted therapies associated with specific signaling of mTOR or other pathways may lead to novel therapeutic targets and improved patient outcomes. (B) Schematic depiction of signaling pathways involving mTOR complexes mTORC1 and mTORC2. Activated phosphoinositide 3-kinase (PI3K) phosphorylates phosphatidylinositol 4,5-biphosphate (PIP2) to form PIP3. PIP3 binds to the pleckstrin homology domains of 3-phosphoinositide-dependent kinase 1/AKT serine/threonine kinase (AKT) to mediate the phosphorylation of AKTThr308. Phosphorylation of AKTSer473 is facilitated by the activation of mTORC2. Activated AKT then promotes the phosphorylation of Thr246 on proline-rich AKT substrate of 40 kDa (PRAS40). Activation of mTORC1 is achieved via AKT which inhibits the activity of the tuberous sclerosis complex 1/2 (TSC1/TSC2) complex resulting in increased level of GTP-bound RAS homolog enriched in brain (RHEB) level. Activated mTORC1 then phosphorylates multiple protein substrates, including eukaryotic translation initiation factor 4E-binding protein 1 (4E-BP1) and ribosomal protein S6 kinase beta-1 (S6K1). Phosphorylation of 4E-BP1 and S6K1 regulates numerous functions including mRNA translation, cell growth, and proliferation. DEPTOR: DEP Domain-containing mTOR-interacting protein; GRB10: growth factor receptor bound protein 10; mLST8: mTOR-associated protein; mSin1: mammalian stress-activated protein kinase-interacting protein 1; LST8 homolog; PROCTOR1: protein observed with RICTOR1; PTEN: phosphatase and tensin homolog; RAPTOR: rapamycin-sensitive adapter protein of mTOR; RICTOR: rapamycin-insensitive companion of mTOR.
The origin of GSCs is still not very clear; they may be an indication of the malignant transformation of a normal tissue stem cell, or they may also be metamorphosed from differentiated neoplastic cells, acquiring stem-like properties through specific genetic perturbations (48). It appears that GSCs display remarkable plasticity between different cellular states of the tumor. Interchange between glioblastoma CSC and non-CSC states often happens depending on factors such as nutrient deprivation, hypoxia, radiation, and others. Such interconversion may play an important role in maintenance of the proliferative or quiescent state of GSCs (48). Like glioblastoma, GSCs are genetically heterogeneous. Eradication of GSCs by traditional treatments remains challenging because of their quiescent nature, causing recurrence.
Signaling pathways involved in dictating the regulation of GSCs are generally not well defined. Multi-faceted approaches that simultaneously target various pathways and molecules may prevail in tumor resistance mechanisms. The PI3K/AKT/mTOR and the EGFR pathways that control the survival as well as the maintenance of CSCs are important potential therapeutic targets (49). The mTOR pathway regulates cell growth and migration of NSCs (50), and consequently, inhibition of the mTOR pathway in CSCs by rapamycin-family inhibitors suppresses proliferation. Moreover, the maintenance of GSCs was also shown to be regulated by the AKT/mTOR signaling pathway (51). In addition, the AKT/mTOR pathway, accompanied by TP53 signaling reactivation, can be used as an effective target for glioblastoma and its stem cells (52).
All-trans retinoic acid (ATRA), derived from retinol, is known to induce the differentiation of neuro-related progenitor cells and stem cells (50). Co-treatment using ATRA with rapamycin reduced neurosphere size and the motility of CSCs, and induced differentiation of GSCs (53). Glioblastoma reoccurs mainly due to the regeneration of tumor from remaining CSCs after initial treatment (28). Thus, targeting CSCs is an exceedingly important aspect of the clinical treatment of glioblastoma. The functional aspects of CSCs, such as cell proliferation and migration, are also important to consider since they directly correlate with the invasive nature of glioblastoma. One proposed mechanism for targeting CSCs is to induce differentiation, causing the differentiated cells to be more amenable to other therapeutic agents. Recent studies have demonstrated this approach by illustrating that mTOR inhibition alone or in combination with a differentiating agent, such as ATRA, can target CSCs (51, 54). Treatment with ATRA caused differentiation of CSCs, as demonstrated by the depletion of stem-cell marker nestin. Furthermore, treatment of glioblastoma cells with the mTORC1 inhibitor rapamycin led to nuclear localization of nestin. These observations were confirmed by western blotting, which demonstrated a time-dependent decrease in nestin expression following ATRA treatment. Proliferation of CSCs, measured by neurosphere diameter, was reduced following treatments with ATRA alone and in combination with rapamycin. Of particular importance, it was shown that the combined treatment of cells with mTOR inhibition and ATRA had a synergistic negative effect on CSC migration (51, 54). This synergism is shown to be mediated by the mitogen-activated protein kinase kinase (MEK)/extracellular-regulated kinase (ERK) pathway, given that treatment of cells with ATRA and MEK1/2 inhibitors resulted in the least cell migration, perhaps due to their influence on differentiation (55). This is of particular interest because resistance to the gold standard chemotherapeutic agent for glioblastoma, temozolomide, was also shown to be mediated via MEK–ERK-induced activation of MGMT (56). One of the mechanisms of resistance to temozolomide is the high expression of the gene encoding MGMT, which removes the methyl groups attached to the N7 and O6 sites on guanine and O3 site on adenine in genomic DNA (57). In addition, a study demonstrated that MEK inhibition reduced murine double minute 2 (MDM2) expression, which resulted in activation of TP53, leading to TP53-dependent downregulation of MGMT expression in CSCs, thereby overcoming temozolomide resistance. This further suggests that inclusion of MEK inhibitor in treatment with temozolomide would make resistant GSCs sensitive to temozolomide (56).
The Role of the mTOR Pathway in Regulation of GSCs
mTOR serves as the protein target of rapamycin, both an immunosuppressant and an anti-fungal macrolide isolated from Streptomyces hygroscopicus (20). Structurally, mTOR kinase is comprised of multiple functional domains, including an N-terminal domain containing at least 20 repeats of Huntingtin elongation factor 3 A subunit of PP2A TOR1, serving as the site for regulatory proteins to interact, forming mTORC1 and mTORC2 (Figure 1B). Other important functional domains include transactivation/transformation-associated domain (FAT), FK506-binding protein 12 (FKBP12) rapamycin-binding domain (FRB), the C-terminal kinase domain (KD), and the C-terminal small FAT domain (FATC). The KD domain possesses conserved sequences that are homologous to the catalytic domain of the PI3K family and contains phosphorylation sites that control PI3K activity (Figure 1B) (58).
The canonical pathway of PI3K/AKT/mTOR controls protein synthesis and contributes to cellular growth in cancer however, the role of mTOR in the regulation of stem cells is becoming increasingly recognized. mTOR is a key downstream component of the pathway by which proto-oncogene Wnt-1 (WNT1) activation can lead to cell growth and tissue aging (59). Studies have demonstrated that WNT inhibits glycogen synthase kinase 3, which in turn activates mTOR via intermediates such as tuberous sclerosis complex 2 (TSC2) (60). This interaction appears to be important in regulating epidermal stem cells. Our previous finding that suppression of mTOR and MAPK altered the expression of the stem cell marker NANOG homeobox (NANOG) suggests that glioblastoma differentiation is altered by interaction of MAPK and mTOR pathways (61). NANOG is a homeodomain transcription factor that is associated with the propagation of undifferentiated human embryonic stem cells and mediates induction of pluripotency (62). While expression of NANOG is typically cytoplasmic, nuclear expression of NANOG has been demonstrated in advanced cancer types (63, 64). In contrast, the expression of NANOG is localized to both the nucleus and cytoplasm in GSCs; however, its role has yet to be elucidated (65, 66). Studies have demonstrated that suppression of MAPK plays a critical role in the maintenance of pluripotency of stem cells (67, 68). Significantly, one study demonstrated this phenomenon in an animal model of brain tumors where dedifferentiation appeared to occur (69). These observations describe the complex mechanism of stem cell regulation.
Activated mTOR enhances cell growth and can prompt some cell types to enter the cell cycle (70). mTOR pathway activation is a result of loss-of-function mutations in tumor suppressors, such as PTEN, tuberous sclerosis 1/2 (TSC1/2), neurofibromin 1/2 (NF1/2), or oncogenic mutations in KRAS proto-oncogene, GTPase (KRAS), PIK3CA, or AKT (71). Sustained activation of mTOR, secondary to a single point mutation, is found in many cancer types, including adenocarcinoma and renal cell carcinoma (72). Furthermore, mTOR hyperactivation due to aberrant PI3K/AKT signaling contributes to both cancer pathogenesis and resistance to therapy in numerous types of cancer (73). Although evidence demonstrated the critical role of AKT signaling in gliomagenesis, AKT expression was not sufficient to induce glioma in mouse models, as it required the coactivation of KRAS signaling (74). The altered cellular metabolism in cancer cells related to sustained activation of mTOR suggests the potential for utilizing mTOR inhibitors in suppressing tumor growth (75). Suppression of the deregulated mTOR pathway exhibits therapeutic potential against tumor proliferation in many types of cancer, such as breast and renal cancer (76, 77). It has been shown that interference of this CSC pathway by rapamycin-family inhibitors significantly reduced proliferation (68). Rapamycin, a selective mTORC1 inhibitor and its analogs (rapalogs) appear to suppress mTORC1 activity and glioblastoma proliferation incompletely (78, 79). Rapalogs bind to FKBP12, inhibiting mTORC1 from executing its downstream functions (80). On the other hand, mTORC2, which regulates the cytoskeletal organization pathway involved in stem cell migration leading to aggressive invasion in glioblastoma, is unaffected by rapamycin and its rapalogs (78, 81). Suppression of proliferation and migration was shown to be achieved by combined inhibition of multiple deregulated pathways (23). Suppression of MAPK has also been shown to play an important role in the maintenance of pluripotency (68). These observations of the AKT/mTOR pathway in the maintenance of glioma CSCs support its role as a major player in the deregulated growth and invasion of GSCs (51).
Aberrant signaling of the PI3K/mTOR pathway has demonstrated clinical relevance via its activation by EGFR, a target for anti-neoplastic drugs that demonstrates abnormal signaling in 57% of glioblastoma (15, 82-84). Alternatively, the PI3K/mTOR pathway is inhibited by tumor suppressor PTEN, loss of which is noted in up to 36-60% of glioblastoma (85, 86). Loss of PTEN is an important indicator of mTOR activity in cancer, particularly in glioblastoma (87). As such, there have been numerous investigations into the use of PI3K/mTOR inhibitors in the treatment of glioblastoma. PI3K/mTOR inhibitors have been extensively studied alone and in combination with other pathway inhibitors, predominantly in recurrent glioblastoma, with only a handful of studies in newly diagnosed glioblastoma. One trial by Cloughesy et al. investigated the use of neoadjuvant oral rapamycin in 15 patients with PTEN-deficient recurrent glioblastoma. While approximately half of the patients demonstrated reduced tumor cell proliferation because of the degree of mTOR inhibition, in the other half, AKT was activated in response to rapamycin treatment, associated with increased phosphorylation of proline-rich AKT substrate of 40 kDa (PRAS40), dysregulating mTORC1 and serving as a surrogate marker for mTORC1 activity (88). The significant reduction in mTOR pathway activity by rapamycin disrupted the negative feedback loop, leading to the activation of upstream AKT pathways, precluding a statistically significant decrease in time-to-progression. While the study failed in its primary outcome, it recognized surrogate molecular markers significant to the mTOR pathway and the evaluation of rapamycin therapy (88).
As mentioned above, rapamycin and its analogs cause incomplete inhibition of mTORC1 (78). Suppression of multiple deregulated pathways offers further inhibition of proliferation and migration (23). Therefore, approaches that target both complexes, mTORC1 and mTORC2, may better inhibit stem cell survival (89). The small molecule inhibitors Torkinib (PP242) and Torin1 were shown to affect both complexes by competitively binding their ATP-binding site; however, this effect was not achieved in a clinical trial (19, 90, 91). Torin2 was developed as a modification of Torin1, with improved water solubility, bioavailability, and half-life. It was demonstrated to effectively inhibit the mTOR pathway through its influence on mTOR complexes via the ATP-binding site and preventing compensatory reactions by secondary pathways while maintaining clinical relevance (92). In a study evaluating the efficacy of various mTOR inhibitors against GSCs, Torin2 was the only mTOR inhibitor capable of repressing the self-renewal properties of GSCs (53). Torin2, therefore, may have the potential to be a powerful tumor-suppressor. XL388, another small compound that functions as an ATP-binding site inhibitor of both complexes, showed greater selectivity, oral bioavailability, and potency than Torin2 (93). A recent study demonstrated that a significant number of glioblastoma tumors expressed nestin and activated mTOR (pmTORSer2448), with most tumor cells co-expressing both markers. Furthermore, the expression of NANOG was suppressed following rapamycin treatment of glioblastoma cell lines. Neurospheres were also disrupted following rapamycin and LY294002 treatments. Treatment with ATRA combined with rapamycin or PP242 suppressed stem cell proliferation. Treatment with Torin1 and Torin2 suppressed the proliferation of glioblastoma CSCs more effectively than using XL388. Torin1 and XL388 delayed the process of self-renewal as compared to controls, whereas Torin2 halted self-renewal and eradicated tumor cells. These findings highlight that the mTOR pathway may contribute to the maintenance of quiescent CSCs, providing a basis for manipulating GSCs and underscoring the potential for use of Torin2 in the treatment of glioblastoma (61).
Finally, a third-generation mTOR inhibitor, termed RapaLink-1, was developed by linking rapamycin with ATP-binding inhibitor MLN0128, which can target mTOR via both the FRB domain and the kinase domain. This compound was generated with the goal of overcoming treatment resistance to rapalogs and mTOR kinase inhibitors in cancer. RapaLink-1 is a potent mTORC1 inhibitor, with demonstrated efficacy in targeting breast cancer cells with resistance to therapy secondary to three somatic mutations within mTOR (94). RapaLink-1 can inhibit glioblastoma cell growth both in vivo and in vitro at levels comparable to those with rapamycin alone or combined with MLN0128, via its interference with the phosphorylation of eukaryotic translation initiation factor 4E-binding protein 1 (4EBP1). Importantly, cells targeted by RapaLink-1 express specific mTOR mutations in their respective xenograft models (81, 95). Although promising, further investigation is needed into the role of RapaLink-1 in the inhibition of GSCs.
Conclusion
Gliomas are incurable malignancies notable for the presence of untreatable GSCs within the tumor mass and in peritumor areas. Most importantly, glioblastoma, which tends to recur due to presence of GSC within the tumor margin and surrounding areas, appears to be regulated by the mTOR pathway. Since the discovery of mTOR nearly two decades ago, numerous aspects of pharmacological, cellular, and molecular regulation mechanisms of mTOR complexes have been described. However, the pathway’s role in the regulation of GSCs remains to be elucidated. While numerous clinical trials using mTOR inhibitors are currently underway, investigations including targeting of GSCs remain limited. Yet our current state of knowledge provides a basis for manipulating GSCs in the treatment of glioblastoma, as the mTOR pathway contributes to the maintenance of GSC quiescence. Future research should focus on further understanding of the PI3K/AKT/mTOR molecular network in the regulation of stem cell quiescence and provide the rationale for targeting the cancer-initiating cells of glioblastoma.
Footnotes
Authors’ Contributions
Conceptualization, M.J.U.; writing – original draft preparation, M.J.U., M.C.M., S.L.Z., H.T.R., E.S. and M.D.; writing – review and editing, M.J.U., M.C.M., S.L.Z., H.T.R., E.S., M.D. and C.D.G.; visualization, M.J.U. and S.L.Z.; supervision, M.J.U. and C.D.G.; project administration, M.J.U.; funding acquisition, M.J.U. All Authors read and agreed to the published version of the manuscript.
Conflicts of Interest
The Authors declare that they have no conflicts of interest related to the preparation and submission of the manuscript, nor any other relevant disclosures.
Funding
This research was funded by The Advanced Research Foundation and The Rockefeller Foundation.
Artificial Intelligence (AI) Disclosure
No artificial intelligence (AI) tools, including large language models or machine-learning software, were used in the preparation, analysis, or presentation of this manuscript.
- Received April 22, 2025.
- Revision received May 24, 2025.
- Accepted June 6, 2025.
- Copyright © 2025 The Author(s). Published by the International Institute of Anticancer Research.
This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY-NC-ND) 4.0 international license (https://creativecommons.org/licenses/by-nc-nd/4.0).
References
In this issue
Jump to section
Related Articles
Cited By...
- No citing articles found.