Glycan and Glycosylation as a Target for Treatment of Glioblastoma

1. Introduction

Glycosylation is a critical process to maturate the glycoproteins and glycolipids. Many factors were demonstrated to regulate this process, including 1) nucleotide sugar donors, 2) glycosyltransferase enzymes, and 3) glycosidase enzymes. The activated nucleotide sugars, synthesized through the hexosamine biosynthesis pathway, are served as sugar donors for the glycosylation process. More than 200 glycosyltransferase (GT) and glycosidase (GA) enzymes, residing in the endoplasmic reticulum (ER) or Golgi apparatus, are responsible for the addition and removal of sugar onto the glycoconjugates [1]. There are two major types of protein core-glycosylation, including N-linked and O-linked glycosylation (Figure 1). Both N-linked and O-linked glycans are generally terminal-modified with sialic acid and fucose via sialylation and fucosylation, respectively. Glycosylation is a sensitive process that could be influenced by several stimulants and cellular stresses. Many studies showed that altered glycosylation is associated with the carcinogenesis and progression of cancers [2, 3]. Defects in glycosylation are possibly caused by the alteration of nucleotide sugar synthesis or the imbalanced expression of glycosyltransferases or glycosidases [4]. Aberrant glycosylation in cancer cells causes the glycan truncation or the expression of uncommon glycans. These aberrantly expressed glycans are possibly used as a biomarker or a target for the treatment of cancers. Many tumor-associated glycans were demonstrated to play essential roles in tumor development, progression, and therapeutic resistance [2, 5].

Recent evidence suggests the alteration of glycosylation in glioblastoma (GBM) [3, 4, 6, 7]. GBM-associated glycans and glycoconjugates, such as the cluster of differentiation 44 (CD44), CD133, and ephrin-A1, were discovered to play important roles in tumor progression, leading to the poor prognosis of patients [8, 9]. Defects of glycosylation in GBM tumors were found in glycoproteins, glycolipids, glycosaminoglycans, or proteoglycans. The alteration of protein glycosylation occurred in both N-linked and O-linked glycosylation. Besides, aberrant terminal glycan modification of sialylation or fucosylation was also observed in GBM [3]. Moreover, the glycans and glycosylation also exhibited the functional significance in glioma stem-like cells (GSC) by regulating the stem cell-related phenotypes [10, 11].

Not only in cancer cells, the tumor microenvironment (TME) was also presented with aberrant glycosylation [12, 13]. Glycosylation changes in TME were found to promote tumor progression, immunosuppression, and therapeutic resistance [14]. Therefore, it is proposed that glycosylation changes of TME might be an alternative target for the treatment of GBM.

This chapter collectively summarizes the recent information on glycan and glycosylation changes and their roles in GBM progression and therapeutic resistance. The information provided here may fulfill our understanding of the roles of glycosylation and its potential to be a target for the treatment of GBM.

2. N-linked glycosylation

The N-linked glycosylation transfers the oligosaccharide chain to the target polypeptide by forming the linkage between the N-acetyl glucosamine (GlcNAc) residue and the amide side chain of the asparagine residue. The process starts in ER; an oligosaccharide is firstly synthesized on the dolichol phosphate carrier and transferred to the protein acceptor by the oligosaccharyltransferase enzyme. The premature glycan chain of N-linked glycoprotein is subsequently modified by the sequential reactions of sugar addition or removal, controlled by several GTs and GAs. The final steps, sialylation and fucosylation, are accomplished in the Golgi apparatus. Many studies demonstrated the alteration of N-linked glycosylation and its related enzymes in GBM (Table 1).

An increase of bi-, tri-, and tetra-anternary N-linked glycans was found to be associated with the progression of GBM [15, 16, 17, 26]. The α1,6-mannosylglycoprotein-β-N-acetylglucosaminyltransferase-5 (MGAT), an enzyme responsible for the synthesis of biantennary N-linked oligosaccharide, was found to promote the invasiveness of GSC [16]. N-linked glycosylation of the receptor protein tyrosine phosphatase type mu (RPTPmu) controlled by MGAT5 was demonstrated to suppress its function and consequently enhance the migration ability of GBM cells through phospholipase C (PLC)/protein kinase C (PKC) pathway [17]. In addition, the MGAT1 (a member of the N-linked associated N-acetylglucosaminyltransferase group) was highly detected in GBM and plays an essential role in promoting the proliferation and invasion of cancer cells [15, 17].

A new subclass of N-glycosylation called-Paucimannosylation, producing a truncated N-glycan (Man₃GlcNAc₂Fuc), was found to elevate in GBM compared with non-tumor tissues [27, 28]. The glycan was found to be involved in the proliferation, migration, and invasion of cancer cells [27].

Another N-link-associated enzyme, a β1,3-N-acetylglucosaminyltransferase-8 (B3GnT8), an enzyme that controls the formation of polylactosamine on β1–6 branched N-glycans, was found to regulate the proliferation and metastatic ability of cancer cells [22]. The β1,4-galactosyltransferase-5 (B4Gal-5) producing highly branched N-glycans was found to regulate the sensitivity of cancer cells to anticancer drugs-etoposide and arsenic trioxide. Suppression of B4GalT-5 could enhance the apoptosis induction effects of these drugs in cancer cells, suggesting its potential to improve the therapeutic efficiency for malignant glioma [19, 20]. Moreover, the B4GalT-5 was also found to regulate self-renewal and tumorigenicity of glioma stem-like cells [20].

Inhibition of N-glycan synthesis by the specific siRNA or inhibitors significantly suppresses tumor growth, metastasis, and radioresistance of GBM [15, 16, 17, 18, 29, 30, 31]. This information suggested the potential of N-glycosylation to be a target for the treatment of GBM.

3. O-linked glycosylation

Golgi-resident glycosyltransferases are responsible for the synthesis of O-glycans via O-linked glycosylation. A particular serine (Ser) and threonine (Thr) residues can be O-glycosidic linked with various kinds of oligosaccharides. This chapter focuses on the mucin-type O-glycosylation or O-GalNAcylation, an O-linked modification of Ser/Thr by N-acetylgalactosamine (GalNAc), followed by the formation of complex oligosaccharide structure. There are 20 isoforms of polypeptide GalNAc transferase (ppGalNAcT or GALNT) identified in humans; the enzymes catalyze the transferring of GalNAc from activated nucleotide sugar donor to initially modify the Ser or Thr residues of a specific glycoprotein [32]. Alteration of O-linked, especially O-GalNAc, glycosylation was observed in many types of cancer [1, 5, 32]. Truncated O-glycans and their associated mucin glycoproteins were applicable as a marker for diagnosis, monitoring, and prognostic prediction of cancer [1].

In GBM, the alteration of O-linked glycosylation played a significant role in the tumor progression and therapeutic resistance [23, 24]. The significance of GALNT enzymes in the progression of GBM has been revealed, suggesting their possibility of being a new target for GBM treatment (Table 1). GALNT-2 was demonstrated to promote the migration and invasion of cancer cells [23]. Expression of GALNT-12 was associated with poor prognosis of GBM patients as it promotes cancer cell proliferation, migration, and invasion via PI3K/Akt/mTOR cascade [24]. The tumor-associated truncated O-linked glycan and its receptor, macrophage galactose-type lectins, were found to modulate the function of tumor-associated macrophages and microglia in GBM [33]. Using lectin from Dolichos biflorus, the GalNAc-associated glycan was highly detected in GSC compared with its differentiated form, suggesting its potential to be a GSC marker (Table 2) [11]. This information suggested the possibility of using GALNTs as a biomarker and a therapeutic target for GBM.

4. Fucosylation

The terminal glycan modification by fucose, called “Fucosylation,” is controlled by the fucosyltransferase (FUT) enzymes. In human, 13 FUTs are classified according to their activities into 1) α1,2-FUTs (FUT-1 and FUT-2), 2) α1,3-FUTs (FUT-3, FUT-4, FUT-5, FUT-6, FUT-7, FUT-9, FUT-10, and FUT-11), 3) α1,4-FUTs (FUT-3 and FUT-5), 4) α1,6-FUTs (FUT-8), and 5) O-FUTs (Pofut-1 and Pofut-2) [34]. Altered expression of FUTs and the fucosylated-glycans were found to associate with tumor development and progression [5, 35]. In GBM, the aggressiveness and malignant phenotypes GBM were associated with fucosylated Lewis antigens’ expression [36]. The enzyme FUT-8, responsible for α-1,6-fucosylation of N-glycans, was discovered to promote the growth, migration, and invasion of GBM cells [25]. Inhibition of fucosylation by the inhibitor-2F-peracetyl-fucose could sensitize the effect of temozolomide (TMZ), suggesting the potential of FUT-8 to be a target for GBM treatment [25].

5. Sialylation

Sialylation is a modification of glycoproteins and glycolipids by sialic acid (Sia). There are 20 sialyltransferase enzymes (STs) responsible for three types of sialylations: 1) α2,3-sialylation, 2) α2,6-sialylation, and 3) α2,8-sialylation (Table 2) [37, 38].

Sialylation was demonstrated to involve in the stemness maintenance of GSC and tumor progression, suggesting its possibility to be a promising target for the treatment of GBM [39, 40]. An α-2,3 sialylation was found to promote the progression, while α-2,6 sialylation suppresses the GBM. Inhibition of α-2,3 or enhancement of α-2,6 sialylation significantly suppresses the metastatic ability of GBM cells [41, 42, 43]. Using lectin from Maackia amurensis, α-2,3 sialylation was found to be enhanced in GSC and play an essential role in stemness maintenance [39]. Suppression of sialylation using ST inhibitor or sialidase leads to the apoptosis of GSC [39]. The mechanism by which α-2,3 sialylation regulates stemness of GSC is probably explained by its role in the stabilization of surface CD133, an important functional GSC marker [43]. Moreover, the lectin M. amurensis lectin-II (MAL-II) could significantly induce the apoptosis of GSC, suggesting its potential for GBM treatment [39]. In addition, suppression of sialylation by a specific inhibitor was found to enhance the sensitivity of GBM cells to the general chemo-drugs—cisplatin and 5-fluorouracil [39]. This collective evidence suggested the potential of α-2,3 sialylation as a target for the treatment of GBM.

In addition, sialidases or neuraminidases (NEU), the enzymes that remove terminal Sia from the oligosaccharide chain of glycoproteins and glycolipids, were also altered in GBM. The overexpression of NEU3 significantly suppresses cancer cells’ migration and invasion ability by promoting focal adhesions through calpain-dependent proteolysis [44]. NEU4 was found to be upregulated in GSC, and suppression of NEU4 significantly reduces cell survival and stemness properties of the cells [45].

6. Gangliosides, glycosaminoglycans, and proteoglycans

Altered syntheses of gangliosides, glycosaminoglycans, and proteoglycans were observed to play significant roles in GBM [46, 47, 48, 49, 50, 51, 52]. The GD3-gangliosides, heparan sulfate (HS) glycosaminoglycans, and their responsible enzymes were found to be altered in GBM and proposed as a potential GBM marker [46, 47, 48]. Glycosaminoglycans played essential roles in the communication between GBM cells and their TME. Alteration of HS synthesis by ablation of heparanase (HPSE) results in the significant reduction of tumor cell adhesion and invasion [48]. This information implied that HS is an important factor in promoting GBM invasion; it is therefore possibly proposed as a therapeutic target for GBM.

Alteration of proteoglycan synthesis was found to associate with the development and progression of GBM [49, 50, 51, 52]. Expression of tumor-associated proteoglycans and their related enzymes were found to facilitate the tyrosine kinase signaling pathway, which benefits the progression of GBM, suggesting their potential as a promising prognostic marker and target for GBM treatment [49]. The elevation of neuro-glial proteoglycan-NG2 was associated with the invasiveness of GBM [50]. NG2 was found to control the vascular morphology and functions, suggesting its role in facilitating metastasis via tumor vascularization of GBM [51]. Targeting NG2, in combination with GD3A (a GBM-associated ganglioside), could significantly reduce the viability of GBM cells [53]. This information suggested the significance of NG2 in the progression of GBM and its possibility of being a target for treatment. Moreover, chondroitin sulfate proteoglycans (CSPGs) play important roles in organizing the tumor microenvironment to prevent tumor invasion. CSPGs were drastically decreased in a diffusely infiltrating tumor of GBM [52].

7. Conclusion and perspectives

Alteration of glycosylation was predominantly observed in either cancer cells or TME in GBM. Both core-glycosylation and peripheral glycan modifications were important factors in regulating the tumor development, progression, and therapeutic resistance. Several strategies have been proposed to target glycans and glycosylation for the treatment of GBM.

Suppression of glycosylation using specific interferences or inhibitors is a potential strategy to target glycosylation [54, 55]. However, there is a limitation to using the broad-spectrum glycosylation inhibitors for cancer treatment as they also affect the neighboring non-tumor cells. Targeting glycosylation of a particular glycoprotein or glycoconjugate is a possible strategy for cancer treatment. In GBM, interference of hyaluronic acid synthesis by methylumbelliferone (4-MU), an inhibitor of hyaluronic acid synthase capable of crossing the blood-brain barrier (BBB), was found to significantly inhibit the proliferation of GBM [56].

The short peptide is recently applicable for targeting or suppressing the specific glycoform of a particular glycoprotein in cancer cells. The deglycosylated form of brevican (dg-Bcan), an ECM-associated glycoprotein upregulated in GBM, was explicitly bound by a small 8-amino acid dg-Bcan-Targeting Peptide (BTP). The radiolabeled-BTP could be internalized into the cancer cell, suggesting its potential to be used as an imaging agent to detect GBM [57]. Further studies to apply this peptide for the treatment of GBM by conjugating it with chemo-drugs or other substances are noteworthy.

Based on the sugar preferential of lectins, the plant lectins were widely used to determine the expression of GBM-associated glycans as well as the functional analyses either in vitro or in vivo model (Table 3) [11, 18, 19, 21, 26, 59, 60]. Using the lectin as a therapeutic agent for GBM is another approach, either combined with other chemo-drugs or as a single agent.

The Phaseolus vulgaris erythroagglutinin (PHA-E) was used to detect the β1,4-GlcNAc-containing N-glycans. It strongly inhibits the migration ability of GBM cells, suggesting its potential to be used for the treatment of GBM [18]. In addition, PHA-E was also found to inhibit the functions of the epidermal growth factor receptor (EGF-R) and a drug efflux pump-P-glycoprotein on GBM [59, 61]. This information suggested the involvement of β1,4-GlcNAc in cancer cells’ growth and drug resistance. Moreover, the potential of PHA-E as a chemosensitizing agent for GBM was also reported [61]. MAL-II is another lectin that can suppress the stemness maintenance and induce apoptosis of GSCs, suggesting its application as a therapeutic agent for GBM [39]. Lectin from Griffonia simplicifolia I (GSL-I) was used to identify the GBM-specific cell surface glycobiomarkers compared with the low-grade glioma. The identified markers may be applicable for diagnosis and possibly used as a target for the treatment of GBM [58]. With another type of brain tumor, the lectin from Canavalia brasiliensis seeds (ConBr) was found to suppress the ERK1/2 and Akt signaling pathways, consequently inhibiting the migration ability of rat neuroblastoma cells [62]. Besides the lectins, monoclonal antibodies against the specific glycans have been established and used to detect cancer-associated glycans. The antibodies can also suppress or activate the functions of glycans in cancer cells; this information suggests the possibility of using a glycan-specific antibody to treat the GBM patients [63, 64].

In conclusion, glycans and glycosylation have been identified to play significant roles in GBM progression and therapeutic resistance. Targeting glycans and glycosylation is possibly an alternative strategy for the treatment of GBM; however, further studies to target specific glycosylation of a particular glycoconjugate are still needed. In addition, the clinical studies or trials on the potential of using glycans and glycosylation as a target for GBM treatment are still a large gap that needs to be further evaluated.

Acknowledgments

The author would like to thank the supports from the National Research Council of Thailand and Khon Kaen University, Thailand.

Conflict of interest

The author declares no conflict of interest.