Aberrant glycosylation as biomarker for cancer: focus on CD43

Abstract

Glycosylation is a posttranslational modification of proteins playing a major role in cell signalling, immune recognition, and cell-cell interaction because of their glycan branches conferring structure variability and binding specificity to lectin ligands. Aberrant expression of glycan structures as well as occurrence of truncated structures, precursors, or novel structures of glycan may affect ligand-receptor interactions and thus interfere with regulation of cell adhesion, migration, and proliferation. Indeed, aberrant glycosylation represents a hallmark of cancer, reflecting cancer-specific changes in glycan biosynthesis pathways such as the altered expression of glycosyltransferases and glycosidases. Most studies have been carried out to identify changes in serum glycan structures. In most cancers, fucosylation and sialylation are significantly modified. Thus, aberrations in glycan structures can be used as targets to improve existing serum cancer biomarkers. The ability to distinguish differences in the glycosylation of proteins between cancer and control patients emphasizes glycobiology as a promising field for potential biomarker identification. In this review, we discuss the aberrant protein glycosylation associated with human cancer and the identification of protein glycoforms as cancer biomarkers. In particular, we will focus on the aberrant CD43 glycosylation as cancer biomarker and the potential to exploit the UN1 monoclonal antibody (UN1 mAb) to identify aberrant CD43 glycoforms.

Figure 1

Schematic representation of N-glycosylation process. N-Glycosylation is an evolutionary conserved stepwise process that can be summarized as follows. (a) The assembling of the precursor oligosaccharide occurs at the cytoplasmic side of endoplasmic reticulum (ER) and starts with dolichol-pyrophosphate (Dol-PP), which is a nucleotide linked to a monosaccharide on a lipid carrier, an isoprenoid compound (90–100 carbon atoms total). The sequential incorporation of monosaccharides N-acetyl-glucosamine (GlcNAc) and Mannose (Man) is catalyzed by various glycosyltransferases. Once the intermediate Dol-PP-GlcNA₂cMan₅ is made, it flips to the luminal side of the ER, where four further residues of mannose and three residues of glucose (Glc) are added, leading to Dol-PP-GlcNAc₂Man₉Glc₃. (b) When the proteins containing the consensus sequence for N-glycosylation (Asn-X-Ser and Asn-X-Thr) translocate to the ER, they are glycosylated by the oligosaccharyltransferase (OST) that catalyzes the transfer of the N-glycan to specific asparagine residues included within the consensus sequence (Asn-X-Ser/Thr) of the target proteins. After the oligosaccharide is transferred to the target protein, specific enzymes remove the three Glucose residues and one particular Mannose. The ER lumen contains a specific glycosyltransferase that binds to target protein, catalyzing the addition of Glucose residues only when the target protein is unfolded or misfolded. The newly glycosylated Glc₁GlcNAc₂Man₇₋₉ oligosaccharides are then bound by two specific lectins, the ER-membrane-attached Calnexin or ER-luminal Calreticulin, which will allow the glycoprotein folding. Once folding is completed, the glucose residue is removed. (c) N-Glycosylated proteins move from ER to Golgi, where specific enzymes catalyse their sequential modifications to the Man₈(GlcNAc)₂ chains. In particular, in the cis-Golgi compartment, most Mannose residues of the original backbone chain are removed, leading to a N-glycan core structure constituted by GlcNAc₂Man₃. In mammals, two main groups of oligosaccharides linked to proteins are found: complex N-glycans and high-Mannose N-glycans. These structures are generated during the passage of the proteins toward the trans-side by the subsequent addition of Mannose residues (for the high-mannose N-glycans) and by addition of three residues of N-acetyl glucosamine (GlcNAc), two residues of Gal (galactose), two residues of NeuAc (N-acetylneuraminic acid) or sialic acid, and a single fucose residue (for the complex N-glycans). (d) Graphic legend of described structures.

Figure 2

Mucin-type O-glycosylation. Different forms of O-glycosylation of proteins occur in animals. Mucin-type O-glycosylation is the most abundant form of protein O-glycosylation and consists of glycans attached via O-linked N-acetylgalactosamine (GalNAc) to serine and threonine residues. Other types of O-glycosylation include mannose-, galactose- (added to a residue of hydroxylysine), fucose-, glucose-, xylose-, and N-acetylglucosamine-O-glycosylation. GalNAc-O-glycosylation originates in the Golgi apparatus after protein folding, while other types of O-glycosylation of proteins in the secretory pathway initiate in the ER. Mucin-type O-glycosylation is initiated by a large family of homologous proteins, named N-acetylgalactosamine (GalNAc)-transferase that catalyses the transfer of GalNAc from UDP-GalNAc to Ser or Thr residues of target glycoprotein. These enzymes are sequentially and functionally conserved across species and their expression is time and tissue specific, suggesting a very complex regulation. Up to 20 different isoforms of polypeptide N-acetyl-α-D-galactosaminyltransferases are known and many are specific for the sites of attachment of the GalNAc to serine/threonine residues, influencing the density and the specific position of the O-glycosylation of target proteins. Thereafter, specific glycosyltransferases can catalyse the addition to GalNAc of specific monosaccharides generating four common subtypes (Core-1-, Core-2-, Core-3-, and Core-4-O-glycan structures) based on differential monosaccharide linkage reactions to the GalNAc (GalNAcα-Ser/Thr). Most O-glycans contain the Core-1 subtype (a), which is generated by the addition of galactose to the GalNAc through a β1–3 linkage by the Core-1-(β1–3) galactosyltransferase. This structure is usually further extended by the addition of monosaccharides such as N-acetylglucosamine, galactose, N-acetylneuraminic acid, and fucose. (b) Core-2-O-glycans are generated by the addition of GlcNAc to the GalNAc through a β1–6 linkage. In order to generate Core-2-O-glycans, Core-1 structure is required as a substrate; thus, the Core-2 structure includes the Core-1 structure. The Core-2-O-glycan can be further extended into either a mono- or biantennary form by addition of multiple galactose (Gal(β1–4)GlcNAc) units and terminal linkages of fucose and sialic acid. (c) Core-3 subtype is generated by the addition of GlcNAc in a β1–3 linkage to the GalNAc, and it can be extended by the addition of GlcNAc in a β1–6 linkage, generating the Core-4-O-glycan (d), which also can be extended by addition of monosaccharides, such as galactose, fucose, and sialic acid, which results in the synthesis of a wide spectrum of O-glycan structures. In some cases, the biosynthesis of O-glycans is stopped by the addition of sialic acid residues in early biosynthesis leading to “dead ends” of O-glycan structures that cannot be further modified ((e), (f)). Truncated O-glycan structures are frequently found in cancer cells as tumour antigens, suggesting that aberrant glycosylation may contribute to cancer progression by modifying cell signalling, adhesion, and antigenicity. (g) Graphic legend of described structures.