Abstract Aberrant glycan epitopes are a classic hallmark of malignant transformation, yet their full clinical potential in cancer diagnostics and therapeutics is yet to be realized. This is partly because our understanding of how these epitopes are regulated remains poorly understood. In this review cancer glycan epitopes for the major glycan classes are summarized with a focus on their biosynthesis, structure and role in cancer progression. Their application as cancer biomarkers, in particular the more recent work on cancer glycoforms, and the advantages these offer over the glycan or protein alone are discussed. Finally, emerging concepts which expand on the current view of the cancer glycan epitope beyond the single structure, to patterns and the whole glycocalyx, are described. These new approaches that consider the cancer glycan epitope as a glycoform, or as a pattern of many epitope structures, are providing new targets both for cancer biomarkers and therapeutics currently in development at the bench and the clinic. biosynthesis, cancer, glycan, immunity, in silico models Introduction It is well established that altered glycosylation is a hallmark of malignant transformation. The relationship between glycan epitopes and cancer were first identified in hematological studies and some epithelial malignancies during the early and mid 1900s (Friedenrich 1930; Aird et al. 1953; Rasch 1953; Krokfors and Kinnunen 1954; McConnell et al. 1954; Moreau et al. 1957). During this time, correlative studies on blood group type and cancer risk were being investigated for some carcinomas (Krokfors and Kinnunen 1954) and the T and Tn antigens (discussed later under O-GalNAc epitopes) had been identified. The former was unmasked on erythrocytes by sialidase-producing microbes (Friedenrich 1930), and the latter in a patient with hemolytic anemia (Dausset et al. 1959). Later works in the 1970 and 1980s further cemented aberrant glycosylation with cancer progression (For an excellent overview of these earlier works up until the mid 1980s, with a thorough discussion on T and Tn glycan antigens, see Springer 1984), diagnosis (Koprowski et al. 1981; Herlyn et al. 1984), prognosis (Leathem et al. 1983; Springer 1984) and therapy (Koprowski et al. 1979). Aberrant glycan epitopes were found early on to have prognostic value (Koprowski et al. 1981; Magnani et al. 1981). In one study, altered patterns of glycosylation were observed (Leathem et al. 1983) by comparing several lectins on normal or diseased breast tissue. Here, it was noted specifically that the location of lectin binding was altered, and that the ConA lectin (which binds the nonreducing end of α-mannosyl and α-glucosyl residues) only bound to malignant tissue (Leathem et al. 1983). Importantly the lectin binding pattern seen was found to be prognostic for evaluating long-term outcome in breast cancer patients (Leathem and Brooks 1987). At the same time, targeting glycan epitopes was showing therapeutic efficacy in cancer models. For example, overexpression of the glycosphingolipid, GD3, had been identified as a melanoma associated epitope (Pukel et al. 1982). An anti-GD3 antibody, R24, was trialed, with “Major tumor regression” observed in 25% of patients (Houghton et al. 1985), long before the field of cancer immunotherapy had come of age (Mellman et al. 2011). We now have a much better idea about the nature of the glycan structures that are present, which provide more potential targets for treatment. These structures are in some cases unique to disease making them excellent targets. Additionally, most cancer biomarkers used in clinical assessment are glycoproteins overexpressed by malignant cells. Often the detection of these biomarkers is directed against either the protein backbone or the glycan moiety separately. However, the “glycoform” (defined by the combined protein and glycan structure) of these biomarkers is often altered in malignancy, which has only recently begun to be applied to improve biomarker specificity. Advances in cancer glycobiology have been somewhat slowed, because studying glycan structure in disease remains only available to the specialist. Nevertheless, there have been significant advances in our understanding of glycan structure, function, and regulation in cancer. These studies demonstrate the great potential of the cancer glycan epitope for understanding and treating malignant disease. In this review cancer glycan epitopes identified across the major glycan classes will be summarized. Where possible their biosynthesis, structure and function will be described. In the second part of the review the potential of the aberrant cancer glycoform for diagnostics will be discussed. Finally, emerging concepts of how cancer glycoform regulation, structure and function is being investigated, including in silico modeling, will be presented. Cancer glycan epitopes This section is laid out to follow the order of glycan assembly, and in general follows the model set by the “Essentials of Glycobiology 3rd edition”, and relevant chapters of this text are referenced within the background information on each glycan type. In each case the cancer epitope is discussed, and the specific changes in glycan processing that produce it, and its function in cancer progression noted, if known. First, the major core oligomer classes, O-linked (O-GalNAc), N-linked and glycosphingolipid, are discussed under core structures. These core structures can then be further modified, which is discussed under Cancer epitopes within structures common to different core glycans. Here, cancer epitopes found in (poly)N-acetyllactosamines and Lewis antigens are discussed. The terminus of these mature glycan oligomers are often capped with sialic acid residues. Sialic acid and its recognition receptors are discussed by Adams et al. within this review series, and only a brief summary is mentioned here. Under Glycan polymer epitopes the glycosaminoglycans (GAGs), a family of polymeric repeating disaccharide units, which can be O-Xyl, N-linked or free polymers are discussed. Finally in this section, under Xeno antigens we will look briefly at glycan epitopes from dietary sources that are not endogenously expressed in humans. Specific glycan epitopes mentioned are referenced in Table I where their structure is given. Figure 1 provides a simplified summary of N- and O-linked processing from gene through to function. The receptors for the cancer glycan structures highlighted are generally not discussed in detail in this review. For the Siglec receptors please see Adams et al, and Selectins see Borsig, L. both reviews are part of this series. The general process of how the major glycan classes are biosynthesized is out of the scope of this review and not discussed in detail here, only noted alterations that accompany malignant transformation that explain the aberrant glycoform produced. For background reading on glycan processing please see the text “Essentials of Glycobiology 3rd edition”. Table I. Cancer glycan epitopes mentioned in this review. With the exception of the O-GalNAc glycan’s, only the epitope structure is shown, which would be part of a larger glycan structure. *The structure shown does not define PSGL1 (PSGL1 includes the protein encoded by the SELPLG gene), but shows the P-selectin binding part of the epitope Cancer glycan epitope Class Active glycan epitope structure Identified in (cancer type) Role in cancer Tn O-glycan Pan-cancer Adhesion and invasion STn Pan-cancer Immunosuppression T-antigen Pan-cancer Metastasis MUC1-ST Pan-cancer Immunosuppression KL-6 Breast Identified on MUC1 H type 1 N-glycan Ovarian, melanoma Invasion through Gal-1 signaling Type 1 Ovarian, melanoma A1G (4)2S2-3 Melanoma Immunosuppression HMOCC-1 antigen Ovarian Unknown N-glycan branching Pan-cancer Cell planar asymmetry, invasive potential, sensitization to growth factors Bisecting GlcNAc Pan-cancer Cell signaling through galectins. Control of proliferation HCELL N-glycan and O-glycan/SLex expressing HCELLs variant shown. The precise structure of HCELLv is not known. HCELLs (N-glycan): Acute myeloid leukemia HCELLv (O-glycan): Colon, breast Metastasis through E-selectin binding GloboH GSL Pan-cancer Immunosuppression. Angiogenesis P1 Ovarian Tumor cell migration GM3 Pan-cancer Adhesion and motility GD2 Neuroblastoma, melanoma EMT, putative cancer stem cell marker GD3 Melanoma Connected to many processes including adhesion, motility, proliferation, angiogenesis RM2 Prostate Unknown CSE UNIT GAG Osteosarcoma cell line Adhesion to laminin. Tumor invasion HSNS4F5 Melanoma and ovarian cancer cell lines Adhesion Lewisa Glycolipids/O-glycan’s/N-glycan’s Pancreatic and gastrointestinal Unknown Sialyl Lewisa Colon, lung, gastric, pancreatic Ligand for selectins. Sequestering from blood. Metastasis Lewisb Colon, gastrointestinal Pathogen binding for colonization of carcinogenic pathogens Lewisx Pan-cancer Survival, proliferation, migration, adhesion Sialyl Lewisx Pan-cancer Ligand for selectins. Sequestering from blood. Metastasis LewisY Pan-cancer Signaling expression of growth factors, proteases, and adhesion molecules, which all aid tumor progression PSGL-1* O-glycan/SleX expressing Pan-cancer Metastasis though P-selectin. Platelet interaction in metastasis, and established tumor? α-Gal XENO Diet related Humans have high anti-α-Gal response Neu5Gc Diet related Humans have high anti-Neu5Gc response. Is a xenoautoantigen. Incorporates into cell glycocalyx Cancer glycan epitope Class Active glycan epitope structure Identified in (cancer type) Role in cancer Tn O-glycan Pan-cancer Adhesion and invasion STn Pan-cancer Immunosuppression T-antigen Pan-cancer Metastasis MUC1-ST Pan-cancer Immunosuppression KL-6 Breast Identified on MUC1 H type 1 N-glycan Ovarian, melanoma Invasion through Gal-1 signaling Type 1 Ovarian, melanoma A1G (4)2S2-3 Melanoma Immunosuppression HMOCC-1 antigen Ovarian Unknown N-glycan branching Pan-cancer Cell planar asymmetry, invasive potential, sensitization to growth factors Bisecting GlcNAc Pan-cancer Cell signaling through galectins. Control of proliferation HCELL N-glycan and O-glycan/SLex expressing HCELLs variant shown. The precise structure of HCELLv is not known. HCELLs (N-glycan): Acute myeloid leukemia HCELLv (O-glycan): Colon, breast Metastasis through E-selectin binding GloboH GSL Pan-cancer Immunosuppression. Angiogenesis P1 Ovarian Tumor cell migration GM3 Pan-cancer Adhesion and motility GD2 Neuroblastoma, melanoma EMT, putative cancer stem cell marker GD3 Melanoma Connected to many processes including adhesion, motility, proliferation, angiogenesis RM2 Prostate Unknown CSE UNIT GAG Osteosarcoma cell line Adhesion to laminin. Tumor invasion HSNS4F5 Melanoma and ovarian cancer cell lines Adhesion Lewisa Glycolipids/O-glycan’s/N-glycan’s Pancreatic and gastrointestinal Unknown Sialyl Lewisa Colon, lung, gastric, pancreatic Ligand for selectins. Sequestering from blood. Metastasis Lewisb Colon, gastrointestinal Pathogen binding for colonization of carcinogenic pathogens Lewisx Pan-cancer Survival, proliferation, migration, adhesion Sialyl Lewisx Pan-cancer Ligand for selectins. Sequestering from blood. Metastasis LewisY Pan-cancer Signaling expression of growth factors, proteases, and adhesion molecules, which all aid tumor progression PSGL-1* O-glycan/SleX expressing Pan-cancer Metastasis though P-selectin. Platelet interaction in metastasis, and established tumor? α-Gal XENO Diet related Humans have high anti-α-Gal response Neu5Gc Diet related Humans have high anti-Neu5Gc response. Is a xenoautoantigen. Incorporates into cell glycocalyx Table I. Cancer glycan epitopes mentioned in this review. With the exception of the O-GalNAc glycan’s, only the epitope structure is shown, which would be part of a larger glycan structure. *The structure shown does not define PSGL1 (PSGL1 includes the protein encoded by the SELPLG gene), but shows the P-selectin binding part of the epitope Cancer glycan epitope Class Active glycan epitope structure Identified in (cancer type) Role in cancer Tn O-glycan Pan-cancer Adhesion and invasion STn Pan-cancer Immunosuppression T-antigen Pan-cancer Metastasis MUC1-ST Pan-cancer Immunosuppression KL-6 Breast Identified on MUC1 H type 1 N-glycan Ovarian, melanoma Invasion through Gal-1 signaling Type 1 Ovarian, melanoma A1G (4)2S2-3 Melanoma Immunosuppression HMOCC-1 antigen Ovarian Unknown N-glycan branching Pan-cancer Cell planar asymmetry, invasive potential, sensitization to growth factors Bisecting GlcNAc Pan-cancer Cell signaling through galectins. Control of proliferation HCELL N-glycan and O-glycan/SLex expressing HCELLs variant shown. The precise structure of HCELLv is not known. HCELLs (N-glycan): Acute myeloid leukemia HCELLv (O-glycan): Colon, breast Metastasis through E-selectin binding GloboH GSL Pan-cancer Immunosuppression. Angiogenesis P1 Ovarian Tumor cell migration GM3 Pan-cancer Adhesion and motility GD2 Neuroblastoma, melanoma EMT, putative cancer stem cell marker GD3 Melanoma Connected to many processes including adhesion, motility, proliferation, angiogenesis RM2 Prostate Unknown CSE UNIT GAG Osteosarcoma cell line Adhesion to laminin. Tumor invasion HSNS4F5 Melanoma and ovarian cancer cell lines Adhesion Lewisa Glycolipids/O-glycan’s/N-glycan’s Pancreatic and gastrointestinal Unknown Sialyl Lewisa Colon, lung, gastric, pancreatic Ligand for selectins. Sequestering from blood. Metastasis Lewisb Colon, gastrointestinal Pathogen binding for colonization of carcinogenic pathogens Lewisx Pan-cancer Survival, proliferation, migration, adhesion Sialyl Lewisx Pan-cancer Ligand for selectins. Sequestering from blood. Metastasis LewisY Pan-cancer Signaling expression of growth factors, proteases, and adhesion molecules, which all aid tumor progression PSGL-1* O-glycan/SleX expressing Pan-cancer Metastasis though P-selectin. Platelet interaction in metastasis, and established tumor? α-Gal XENO Diet related Humans have high anti-α-Gal response Neu5Gc Diet related Humans have high anti-Neu5Gc response. Is a xenoautoantigen. Incorporates into cell glycocalyx Cancer glycan epitope Class Active glycan epitope structure Identified in (cancer type) Role in cancer Tn O-glycan Pan-cancer Adhesion and invasion STn Pan-cancer Immunosuppression T-antigen Pan-cancer Metastasis MUC1-ST Pan-cancer Immunosuppression KL-6 Breast Identified on MUC1 H type 1 N-glycan Ovarian, melanoma Invasion through Gal-1 signaling Type 1 Ovarian, melanoma A1G (4)2S2-3 Melanoma Immunosuppression HMOCC-1 antigen Ovarian Unknown N-glycan branching Pan-cancer Cell planar asymmetry, invasive potential, sensitization to growth factors Bisecting GlcNAc Pan-cancer Cell signaling through galectins. Control of proliferation HCELL N-glycan and O-glycan/SLex expressing HCELLs variant shown. The precise structure of HCELLv is not known. HCELLs (N-glycan): Acute myeloid leukemia HCELLv (O-glycan): Colon, breast Metastasis through E-selectin binding GloboH GSL Pan-cancer Immunosuppression. Angiogenesis P1 Ovarian Tumor cell migration GM3 Pan-cancer Adhesion and motility GD2 Neuroblastoma, melanoma EMT, putative cancer stem cell marker GD3 Melanoma Connected to many processes including adhesion, motility, proliferation, angiogenesis RM2 Prostate Unknown CSE UNIT GAG Osteosarcoma cell line Adhesion to laminin. Tumor invasion HSNS4F5 Melanoma and ovarian cancer cell lines Adhesion Lewisa Glycolipids/O-glycan’s/N-glycan’s Pancreatic and gastrointestinal Unknown Sialyl Lewisa Colon, lung, gastric, pancreatic Ligand for selectins. Sequestering from blood. Metastasis Lewisb Colon, gastrointestinal Pathogen binding for colonization of carcinogenic pathogens Lewisx Pan-cancer Survival, proliferation, migration, adhesion Sialyl Lewisx Pan-cancer Ligand for selectins. Sequestering from blood. Metastasis LewisY Pan-cancer Signaling expression of growth factors, proteases, and adhesion molecules, which all aid tumor progression PSGL-1* O-glycan/SleX expressing Pan-cancer Metastasis though P-selectin. Platelet interaction in metastasis, and established tumor? α-Gal XENO Diet related Humans have high anti-α-Gal response Neu5Gc Diet related Humans have high anti-Neu5Gc response. Is a xenoautoantigen. Incorporates into cell glycocalyx Fig. 1. View largeDownload slide Schematic of selected O-GalNAc, and N-glycan processing pathways connecting gene, structure and function together. The inner ring shows the change at gene level of glycan processing enzymes and molecules. The following ring shows primary glycan epitope biosynthesis, and the signaling molecules that act upon it, and feedback to gene expression. The next ring deals with secondary processing. Outside of the ring shows the role of the epitope in disease progression. As an example, the diagram shows for branched core N-glycans, these are overexpressed though GNT5 upregulation. This diagram also shows GNT5 can be upregulated through IL6 signaling and GNT5 is involved in Pi3K/ATK signaling. The core branched N-glycan can be further modified with N-acetyllactosamines. Functionally the N-acetyllactosamine structures are involved in galectin signaling, cell survival and drug resistance. The N-acetyllactosamines can be further modified to a Lewis antigen through the action of fucosyltransferases, and sialyltransferases. Functionally the sialyl Lewis antigens are ligands for selectins which are important receptors in metastasis (see review by Borsig, L, in this series). Fig. 1. View largeDownload slide Schematic of selected O-GalNAc, and N-glycan processing pathways connecting gene, structure and function together. The inner ring shows the change at gene level of glycan processing enzymes and molecules. The following ring shows primary glycan epitope biosynthesis, and the signaling molecules that act upon it, and feedback to gene expression. The next ring deals with secondary processing. Outside of the ring shows the role of the epitope in disease progression. As an example, the diagram shows for branched core N-glycans, these are overexpressed though GNT5 upregulation. This diagram also shows GNT5 can be upregulated through IL6 signaling and GNT5 is involved in Pi3K/ATK signaling. The core branched N-glycan can be further modified with N-acetyllactosamines. Functionally the N-acetyllactosamine structures are involved in galectin signaling, cell survival and drug resistance. The N-acetyllactosamines can be further modified to a Lewis antigen through the action of fucosyltransferases, and sialyltransferases. Functionally the sialyl Lewis antigens are ligands for selectins which are important receptors in metastasis (see review by Borsig, L, in this series). Core structures In this section O-GalNAc, N-linked and glycosphingolipid cancer epitopes are discussed. O-linked (O-GalNAc) O-GalNAc glycans are attached to proteins at serine or threonine sites (Brockhausen and Stanley 2015). Although there is no defined sequon where an O-glycan is attached, they are often found within variable number of tandem repeat (VNTR) domains, which are high in serine and threonine repeats. On a mucin, hundreds of O-glycan’s can be present within the VNTR regions, expressed in a variety of glycoforms (Brockhausen and Stanley 2015). Mucins are produced primarily by epithelial cells on the surfaces of various membranes, and secreted into the extracellular space. In healthy cells mucins are presented on the apical surface, but cells loose this polarization during malignant transformation, which supports an invasive phenotype (for background information and further reading on this phenomenon see elsewhere (Kufe 2009; Varki et al. 2015). For example, membrane type I matrix metalloproteinase (MT1-MMP) polarization in malignant transformation is lost on the apical surface of epithelial cells and is found to concentrate in specific membrane structures that sit close to the basement membrane, and aid invasion (Nakahara et al. 1997; Sato et al. 1997), through degradation of the basement membrane (Woskowicz et al. 2013), and activation of other MMPs which are capable of degrading the collagen rich extracellular matrix that often surrounds malignant cells (Taniwaki et al. 2007). The first step in O-glycan (O-GalNAc) synthesis is UDP-GalNAc transferred to a Ser/Thr by ppGalNAcTs, a family of enzymes consisting of ~20 members (Gerken et al. 2011). O-glycan’s are characterized across eight core structures (Brockhausen and Stanley 2015). Overexpression and/or aberrant expression of mucins by carcinomas has been known for many years (Varki et al. 2015). In general, mucins act as anti-adhesins (Komatsu et al. 2000; Kufe 2009; Varki et al. 2015), and therefore aid displacement of malignant cells during metastasis. Below the main O-GalNAc cancer epitopes are discussed, starting with the changes in glycoprocessing that produces them. Altered O-glycan processing Mucin O-glycan’s are generally truncated in malignant transformation, producing simpler and fewer types of glycan structure (Lloyd et al. 1996). Truncation of O-GalNAc glycans is controlled through mutation (Ju and Cummings 2002; Schietinger et al. 2006), gene expression (Sewell et al. 2006; Beatson et al. 2015) often during inflammation (Sproviero et al. 2012; Colomb et al. 2014), enzyme relocation (Gill et al. 2013) or intracellular environmental effects (Hassinen et al. 2011). Mutation: In malignant cells, mutation (Ju et al. 2008) or epigenetic silencing of COSMC (Mi et al. 2012; Radhakrishnan et al. 2014), is a pathway to cancer epitope production (Ju and Cummings 2002; Schietinger et al. 2006). COSMC is a chaperone molecule required for activity of a β1–3 galactosyltransferase, that in normal cells is essential for the synthesis of the T-antigen (for a review of COSMC, see Ju et al. 2011). Therefore, mutation or silencing of COSMC ablates cell access to elongated core 1 and 2 structures. COSMC mutation is associated with pancreatic (Radhakrishnan et al. 2014; Hofmann et al. 2015), and colon (Yu et al. 2015) cancers. However, COSMC silencing does not appear to be an essential process for cancer progress, since it is not a consistent feature of these cancer types, and in some cases COSMC is upregulated (Madsen et al. 2013; Huang et al. 2014). COSMC mutation or silencing leads to upregulation of Tn and STn antigens (discussed later) (Ju and Cummings 2002; Schietinger et al. 2006). Of note, whilst other malignant cell mutations often result in the expression of a single tumor antigen to which an immune response may be raised, because the COSMC mutation effects glycoforms on many proteins, it is an example of one mutation, producing many antigens, resulting in multiple immunological responses. Gene expression: Truncated O-glycans on malignant cells can also be explained through changes in expression levels of processing enzymes (Marcos et al. 2004; Sewell et al. 2006). For example, cells with active COSMC will still generate the STn epitope through ST6GalNAc-1 upregulation (Sewell et al. 2006; Beatson et al. 2015) (Figure 1). Overexpression of the 2,3-sialyltransferase ST3Gal-1 is often upregulated by malignant cell lines (Mungul et al. 2004; Picco et al. 2010). This can lead to overexpression of the ST-antigen (discussed later), even in the presence of core-2 branching enzymes C2GnT1 (Dalziel et al. 2001). The regulation of glycan processing enzymes in malignancy is not well understood. However, inflammatory signaling molecules (discussed under Lewis antigens synthesis) have been shown to upregulate some sialyltransferases, but this has so far only been conclusively shown for Lewis antigens (Padró et al. 2011). One mechanism by which O-glycan processing enzymes may be regulated at transcription has been shown to occur through intracellular mucin signaling (Solatycka et al. 2012). In this study, MUC1 inhibited expression of C2GnT1 glycan processing enzyme, that would usually lead to core-2 O-glycans, thus pushing the glycan processing machinery towards simplier and truncated structures (Solatycka et al. 2012). Enzyme relocation: Some studies have found higher concentrations of certain glycan processing enzymes alone do not necessary correlate with expression of expected glycan epitopes (Yang et al. 1994; Gill et al. 2010). For example, in one study, glycan processing enzyme levels, assayed in cancer tissues, did not correlate well with the presence of complementary epitopes (Yang et al. 1994). Relocation of glycan processing enzymes within the Golgi offers an explanation for this (Gill et al. 2010, 2013). For example, the ppGalNAcT enzyme, GalNAc-T2, has been shown to relocate from the Golgi to the endoplasmic reticulum in several cancer cell lines (Gill et al. 2013). Relocation directly caused a 7–10-fold increase in Tn antigen compared to a 2-fold increase from over expressing a GalNAc-T2 within the Golgi (Gill et al. 2013). This study demonstrates the importance of enzyme location during glycan processing, and not simply enzyme concentrations. The mechanism (s) by which this occurs is under investigation. One study found GalNAc-Ts were redistributed through the Src-kinase pathway, that was activated through the action of growth factors PDGF and EGF, in a process termed GALA (Gill et al. 2010). However, a later, recent study was unable to repeat the GALA mechanism, instead finding PDGF and EGF had no effect on enzyme location (Herbomel et al. 2017). The authors of the original manuscript have, however, replied to this work indicating they are able to reproduce their findings and make some suggestions for why the observed relocation was not seen (Bard and Chia 2017). Further studies need to be conducted now to further support the GALA mechanism, and in particular its relevance in malignant cell glycoform biosynthesis. Intracellular environmental changes: The second mechanism, outside of gene processing, that has been shown to effect O-glycan processing is changes in the pH of the Golgi apparatus (Axelsson et al. 2001; Kellokumpu et al. 2002; Hassinen et al. 2011). In one study (Axelsson et al. 2001), adding small amounts of a weak base (which showed no obvious sign of restructuring the Golgi or endoplasmic reticulum) to cell cultures, caused several O-glycan processing enzymes to relocate, which over a few days lead to a general truncation of the observed O-glycome (Axelsson et al. 2001). Supportive of this idea, under physiological conditions, it has been shown that O-glycan processing enzymes are in heteromeric complexes, which are regulated in part though Golgi acidity (Hassinen et al. 2011). Additionally, cancer cell lines have altered Golgi acidity (a higher pH than normal) (Rivinoja et al. 2006). Altered Golgi pH may therefore explain truncated O-glycosylation in some cases, and also provides a supporting mechanism for the relocation of enzymes. Malignant cells are genetically very heterogeneous, so it seems reasonable that many answers are likely correct for how the truncation of O-glycans occurs. Whichever the mechanism, the major O-glycan epitopes synthesized are Tn, STn, T and ST, which are discussed in the next section. O-glycan epitopes Tn antigen: Or simply the O-GalNAc epitope, is the simplest aberrant O-glycan and a pan-cancer epitope which may indicate aggressiveness of disease (Rambaruth et al. 2012). In addition to the four biosynthesis mechanisms described above, the presence of the Tn antigen may inhibit further elongation of neighboring glycan structures, providing a fifth mechanism by which glycan structure is truncated in disease (Brockhausen et al. 2009). It is important in cell adhesion and invasion (Freire-de-Lima et al. 2011; Gill et al. 2013; Bapu et al. 2016). The increased invasiveness seen here may result from malignant cells adopting an EMT phenotype, with which Tn expression is associated (Freire-de-Lima et al. 2011). In one study, an EMT state was induced by TGFB1, which lead to upregulation of an oncofetal form of fibronectin, that is defined by inclusion of a Tn modification (Freire-de-Lima et al. 2011). Inhibition of the Tn epitope on oncofetal fibronectin was enough to inhibit the EMT phenotype, demonstrating that a simple glycan modification can direct biological function. It is not clear which part of the EMT process the Tn epitope is involved, but the authors suggested it may be important in the mesenchymal adhesion process (Freire-de-Lima et al. 2011). In healthy tissues the Tn antigen is present within the mucinous layers of the colon, and over expressed significantly in ulcerative colitis (Fu et al. 2011; Bergstrom et al. 2016). Truncation of the glycans on the mucus barrier of the colon, allows pathogens to breach and colonize the gut (Bergstrom et al. 2016). This can lead to colitis, and eventually onset to carcinoma (Bergstrom et al. 2016). This is an interesting finding because it demonstrates the relative steric bulk of glycosylation is an important consideration, which will be discussed later in emerging concepts. Often Tn upregulation is accompanied by the STn antigen, the sialylated form of Tn antigen. STn and Tn are often found alongside each other, with the STn epitope being more frequently found (Irimura et al. 1999). STn expression is highly restricted in normal adult tissue. Like the Tn antigen, STn is associated with an increase in metastasis (Julien et al. 2006), perhaps due to a reduced interaction of malignant cells with tissue resident galectins, that are no longer able to bind because of the sialyl residue (for a review of galectin interactions, see Takenaka et al. 2002). STn may also play a role in protecting blood borne tumor cells from the host immune response (Ogata et al. 1992). Currently the Tn and STn antigens are therapeutic targets in immunotherapy. For example CAR-T cell therapies targeting MUC1-Tn (Wilkie et al. 2008; Posey et al. 2016) and the Theratope anti-STn vaccine (Miles et al. 2011). The T-antigen (Friedenrich 1930) is synthesized from the Tn antigen, through COSMC and C1GALT1 action (Brockhausen and Stanley 2015). It is commonly associated with malignant transformation in epithelial cells, upregulated by C1GALT1 enzyme (Figure 1) (Varki et al. 2015). T-antigen expression is associated with metastatic potential, which may result from disrupted galectin signaling, recently reviewed elsewhere (Sindrewicz et al. 2016). Another explanation for the association of T-antigen with metastasis, comes from observations that C1GALT upregulation (that accompanies T-antigen production) can stimulate nuclear accumulation of MUC1C (Chou et al. 2015). MUC1C can signal with β-catenin within the MUC1C/β-catenin pathway, that stimulates cell growth and invasion (Chou et al. 2015) (Figure 1). The sialylated version of the T-antigen, the sialyl-T (ST) antigen, is found in several normal adult tissues (Cao et al. 1996). ST expression is highly upregulated on MUC1 (the MUC1-ST glycoform), that is expressed by many cancer types (Irimura et al. 1999). The MUC1-ST antigen is associated with tumor progression (Mungul et al. 2004). It is important in development of the immunosuppressive tumor microenvironment (Beatson et al. 2016). In this study, MUC1-ST binding to Siglec-9 activated calcium flux and MEK/ERK signaling in macrophages (Beatson et al. 2016). The secretome of MUC1-ST activated macrophages was found to be altered to a tumor-associated macrophage, TAM, phenotype (Beatson et al. 2016). TAMs are associated with dysregulated immune cell maturation, including upregulation of several chemokine and cytokine molecules that stimulate further immune cell infiltration (and potentially feedback to alter glycan processing (Padró et al. 2011). The TAM population was identified as CD206, CD163, IDO and PD-L1 expressing. The latter PDL1 expression is suggestive of the potential of the innate immune response to further educate the infiltrating adaptive immune response. There is increasing interest in priming the innate immune landscape to better facilitate adaptive immune cell tumor killing. The “MUC1-ST Siglec-9 signaling axis” may provide a targetable route to achieve this goal (a summary is shown in Figure 2). Fig. 2. View largeDownload slide Summary of glycan epitope interactions with immune cells. Inflammatory signaling molecules within the tumor microenvironment alter glycan processing, upregulating sialylation of glycan epitopes. Some of these sialylated epitopes are ligands for innate immune cell Siglecs which limits tumor immunosurveilance and inflammation. The MUC1-ST glycoform has been shown to activate macrophages through Siglec-9 to generate a TAM phenotype which may potentially signal to infiltrating adaptive immune cells. The TAM phenotype is also associated with an altered secretome that includes chemokines/cytokines that feeds back into the altered glycoprocessing. Poly-N-acetyllactosamines (PLAs) expressed by malignant cells are a physical block against NK cell immunosurveillance. Fig. 2. View largeDownload slide Summary of glycan epitope interactions with immune cells. Inflammatory signaling molecules within the tumor microenvironment alter glycan processing, upregulating sialylation of glycan epitopes. Some of these sialylated epitopes are ligands for innate immune cell Siglecs which limits tumor immunosurveilance and inflammation. The MUC1-ST glycoform has been shown to activate macrophages through Siglec-9 to generate a TAM phenotype which may potentially signal to infiltrating adaptive immune cells. The TAM phenotype is also associated with an altered secretome that includes chemokines/cytokines that feeds back into the altered glycoprocessing. Poly-N-acetyllactosamines (PLAs) expressed by malignant cells are a physical block against NK cell immunosurveillance. In general, and regardless of the alteration in glycan processing, malignant cells appear to drive towards truncated Tn, STn, T or ST expression. Because malignant cells use a variety of mechanisms to access these epitopes, it seems very likely they provide a significant survival advantage. As described here this is likely through aiding metastasis, and inhibiting immunosurveillance. Further modifications to O-glycans, in particular the N-acetyllactosamine modification and further functionalization to terminal Lewis antigen expression, is discussed later under Cancer epitopes within structures common to different core glycans. N-linked N-glycosylation follows a strictly ordered assembly, and the site of modification is predictable to asparagine residues (N) of a peptide/protein only when an NXT/S sequon is present (where X is any residue accept proline) (Mellquist et al. 1998). There are two major changes that can occur to the core N-glycan structure, which are increased frequency of a bisecting GlcNAc, or β1,6 and β1,4 branching to the core pentasaccharide (discussed below). Other notable changes occur to the epitopes of secondary structures that are attached to the core N-glycan structure, namely the N-acetyllactosamine units and their further functionalizations (discussed under Cancer epitopes within structures common to different core glycans). Additionally, whilst O-glycan epitopes (discussed above) are usually discussed as distinct disease specific epitopes, N-glycans tend to be discussed in terms of a change to the pattern of the N-glycome. In other words, the structures identified are synthesized in normal tissues, but the pattern is altered in disease. Below the bisecting GlcNAc and branching core N-glycome patterns are discussed, followed by other changes in the pattern of the N-glycome in various cancers. Specific epitope changes to N-acetyllactosamines on N-glycans are covered later, as mentioned earlier. Bisecting GlcNAc and truncation patterns: Simple N-glycans with no branching is a common pan-cancer N-glycan epitope (Mellis and Baenziger 1983), identified on tumor-associated membrane proteins (Wang, Zhang et al. 2012). The bisecting GlcNAc is associated with GNT3 upregulation (Narasimhan 1982; Taniguchi and Kizuka 2015) and inhibits other GlcNAc transferase activity, effectively blocking N-glycan branching (Figure 1) (Brisson and Carver 1983; André et al. 2004; André et al. 2007). Regulation of GNT3 may likely result from its epigenetic activation in malignant transformation (Anugraham et al. 2014; Kohler et al. 2016). Upregulation of bisecting GlcNAc in the N-glycome alters galectin signaling, probably through inhibition of branched poly-N-acetyllactosamines (PLAs) residues (North et al. 2010) (discussed later), cell proliferation (reviewed elsewhere Miwa et al. 2012) through growth factor receptor signaling (Song et al. 2010; de-Freitas-Junior et al. 2013), inhibition of immunosurveillance (Yoshimura et al. 1996), and adhesion through integrins (Isaji et al. 2004) (reviewed elsewhere Isaji et al. 2010). For example, in ovarian cancer, the bisecting GlcNAc is associated with tumor supporting notch, WNT and TGFB pathways (Allam et al. 2015). Whilst bisecting GlcNAc can be tumor supporting, other studies show its upregulation generally results in a relative reduction in metatstatic potential (Yoshimura et al. 1995), though inhibition of N-glycan branching. Supportive of this finding, in murine models of breast cancer, loss of GNT3 (and therefore loss of bisecting GlcNAc) was associated with enhanced tumor progression, perhaps through the permissive synthesis of branched PLAs which facilitated galectin signaling (Miwa et al. 2013). Similar observations have been made in colorectal cancer cell lines (Sethi et al. 2014) and colorectal cancer tissues (Balog et al. 2012). Overall, these studies paint a paradoxical picture of bisecting GlcNAc in cancer, on one hand stimulating a protumour phenotype through cell proliferation signaling, that correlates with poor prognosis (Anugraham et al. 2014; Kohler et al. 2016), and a more aggressive phenotype (Bhaumik et al. 1998; Song et al. 2001) in some cancers. Whilst on the other hand, bisecting GlcNAc associates with reduced metastatic potential in others (Balog et al. 2012). Whether bisecting GlcNAc associates with a more, or less, aggressive phenotype, may be dependent on how mature the N-glycan processing is that’s present. For example, where bisecting GlcNAc is associated with less aggressive phenotypes, this also coincides with mostly high mannose processing of the N-glycome (Sethi et al. 2014). Whereas cancers presenting bisecting GlcNAc with complex-type glycans present (highly sialylated bisecting GlcNAc glycans) tend to associate with a more aggressive phenotype (Lee, Thaysen-Andersen et al. 2014). Therefore, cancers expressing bisecting GlcNAc with complex-type processing, may benefit from the advantages complex-type processing offers (immune evasion, metastatic potential) whereas bisecting GlcNAc glycans bearing truncated high mannose glycans do not have access to these advantages. Increased branching: Tumor cells expressing no or low bisecting GlcNAc, will likely have increased N-glycan branching (reviewed recently elsewhere Taniguchi and Kizuka 2015). This is associated with GNT5 upregulation (Miyoshi et al. 1993), which may in part result from IL6 signaling (Nakao et al. 1990), that is associated with cancer inflammation (Netea et al. 2017) (Figure 1). GNT4 also catalyzes branching through β1,4 linkage, that is associated with hepatocellular carcinoma (Fan et al. 2012). Functionally, GNT5 not only upregulates branching, it also signals through a GNT5/PTEN complex to regulate PI3K/ATK signaling, which drives cell planar asymmetry (Cheung and Dennis 2007) seen in invasive malignant cells. This provides a cell morphology explanation for why N-glycomes with increased branching are in general associated with cell lines that have higher metastatic potential (Granovsky et al. 2000; Sethi et al. 2014). More traditionally, there is significant evidence demonstrating the branched N-glycome on malignant cells enhances metastasis through galectin lattices (André et al. 2004, 2007) (for a review on the galectin lattice see elsewhere Nabi et al. 2015), and through activation of adhesion molecules such as integrins. For example, in the latter, N-glycomes on αvβ3 integrin on melanoma, were related to metastatic potential, with more sialylated and branched N-glycans on the αvβ3 integrin associated with more metastatic cells (Pocheć et al. 2015). The core N-glycome branching in itself however, is not the binding ligand for the galectin interactions that enhance metastasis. This is due to the attached PLA structures (Suzuki et al. 2005; Mendelsohn et al. 2007), discussed later under PLA modifications. Other altered N-glycome patterns: High mannose-type N-glycans are associated with immature processing within the Golgi, and are associated with several cancers including bladder (Yang et al. 2015), head and neck (Braig et al. 2017), breast (de Leoz et al. 2011), colon (Balog et al. 2012) and pancreatic (Park et al. 2015). A high mannose N-glycome can be associated with normal cells undergoing TGFB1 induced EMT (Tan et al. 2014). High mannose N-glycan’s were recently found to be involved in a resistance mechanism against cetuximab, a mAb immunotherapy that blocks EGF-EGFR cell proliferation in EGFR positive tumors. The acquired resistance occurred through a single nucleotide polymorphism, SNP, within the sequence where cetuximab binds. The resulting allele, EGFR-K521, showed altered post-translational modification in high mannose instead of complex type N-glycan’s (Braig et al. 2017) demonstrating how small changes to the protein backbone can affect the final glycoform. In this case the SNP resulted in reduced sialylation, which was found to impact structure stability and cetuximab binding, which likely accounts for the reduced cetuximab activity (Figure 1). In addition to altered N-glycome patterns discussed here so far, a distinct N-glycan glycoform has been identified on L1CAM (Hoja-Lukowicz et al. 2013), which is a membrane protein usually expressed in normal adult neurons. L1CAM is upregulated in several cancers (Doberstein, Bretz et al. 2014; Doberstein, Milde-Langosch et al. 2014; Grage-Griebenow et al. 2014; Ito et al. 2014), and is associated with metastatic potential, particularly to the brain (Kiefel et al. 2012; Valiente et al. 2014). In a recent study (Hoja-Lukowicz et al. 2013) a Galβ1,4-Galβ1 motif bearing 2,3-linked sialyl residues (the A1G (4)2S2-3 epitope) was identified, which is the first time it has been observed in cancer. So far this epitope has only been identified in melanoma, and its function and biosynthesis has not been studied. However, the di or tri 2,3-linked sialyl residues on the epitope likely plays a role in the invasive potential of the cell (Hoja-Lukowicz et al. 2013). Interestingly the human xeno antigen Neu5Gc, as discussed later in the Xeno epitopes section, was also detected in this epitope, which is a rare example of Neu5Gc being incorporated into a human N-glycan structure. Taken together, with perhaps the exception of the specific epitopes mentioned here (and later under PLAs), it seems that the type of N-glycan profile present is indicative of the differentiation stage of the malignant cells, as was concluded in an older study with neuroblastoma cell lines (Motoyoshi et al. 1993), and consistently observed more recently across several cancer types (Liu et al. 2013; Kaprio et al. 2015; Sethi et al. 2015). On the other hand, it may be possible to use N-glycomes as distinct cancer fingerprints. In a recent study N-glycomes acquired on several malignant cell types could be used to distinguish cancer origin and subtype (Hua et al. 2014). Glycosphingolipids The glycosphingolipids (GSLs) have been associated with malignant transformation for many decades, with truncation or “incompleteness of the carbohydrate chain” (Hakomori and Murakami 1968) on lipid noted as one of the molecular events accompanying malignant transformation. They are split into two main families, the glucosyl and galactosyl ceramides. GSLs in cancer, including their biosynthesis and other epitopes have recently been reviewed in detail elsewhere (Jennemann and Gröne 2013; Groux-Degroote et al. 2015; Pearce and Laubli 2016), and therefore a summary of the prominent epitopes that are in development as therapeutic targets, or newer antigens not covered in those reviews, are presented here. The Globo H (GH) epitope, remains a key target for cancer vaccine development (recently reviewed elsewhere Danishefsky et al. 2015). GH has been used in vaccination trials in ovarian, breast and prostate cancers (Tsai et al. 2013), and analogues of the molecule have been made to increase immunogenicity, breaking self-tolerance, and induce antibody class switching to IgG (Lee, Chen et al. 2014). In these cases, the desired outcome is to induce antibody dependent cell-mediated cytotoxicity (ADCC) tumor cell killing, by stimulating the correct immune response against the GH epitope. More recently epitopes of the P blood group antigens have been investigated. The P1 antigen was recently identified on the surface of ovarian cancer cells, where it may play a role in tumor cell migration (Jacob et al. 2014). In this same study, anti-P1 IgM was found in ascites fluid further suggesting that P1, whilst a naturally occurring GSL on erythrocytes, when expressed in this unusual glycoform on malignant epithelial cells became immunogenic. Like the GH studies, it is necessary to consider producing class switched antibodies in the development of vaccine strategies against the P1 structure. The gangliosides, GSLs with one or more sialic acid residue attached, have attracted interest as vaccine targets for cancer, and were some of the earlier targets for cancer immunotherapy. GD2 is associated with several cancers including neuroblastoma (Matthay et al. 2012), melanoma (Dobrenkov and Cheung 2014) and breast cancer (Battula et al. 2012), and is considered a putative cancer stem cell marker (Battula et al. 2012). Because of its restricted expression in healthy adult tissues, it is being developed as a target for cancer immunotherapy (for a review on the trials involving GD2 see elsewhere Dobrenkov and Cheung 2014). In particular, CAR T-cells against GD2 are showing significant promise for melanoma (Yvon et al. 2009; Gargett et al. 2016; Hoseini et al. 2017), sarcoma (Long et al. 2016), and neuroblastoma (Craddock et al. 2010; Sun et al. 2010; Louis et al. 2011; Singh et al. 2014; Prapa et al. 2015). The anti-GD2 system was recently further enhanced using a combination with a whole cell vaccine approach (Caruana et al. 2015), and also combining GD2 CAR-T cells against anti-PDL1 treatment, which prolongs efficacy and persistence in patients (Gargett et al. 2016; Heczey et al. 2017). GD3 is associated with melanoma, but not carcinomas of epithelial origin (Pukel et al. 1982). GD3 was one of the first immunotherapy targets (Houghton et al. 1985). Like the globo series of epitopes, the GD3 vaccine candidates have been designed to improve immunogenicity and class switching (Helling et al. 1994). GM3, the monosialyated version of GD3 and the simpliest of the ganglioside structures was first associated with cancer progression in the 1960s (Hakomori and Murakami 1968) (for a recent focused review of GM3 and cancer, see Hakomori and Handa 2015). GM3 has an inhibitory effect on several growth factor receptors, including PDGF, FGFR and EGFR, all of which are associated with cancer progression. Therefore, in some cases GM3 may inhibit tumor progression. Therefore, using GM3 and analogue structures to inhibit these signaling pathways and inhibit tumor progression have (Bremer and Hakomori 1982) and continue to be designed and tested (Fujikawa et al. 2008; Kawashima et al. 2014). For example, in recent work analogues of GM3 have been designed to aid antibody class switching against the antigen (Wang, Zhou et al. 2012) (Delgado et al. 2002; Wang et al. 2009). The xeno epitope Neu5Gc (discussed briefly later under xenoepitopes) has also been found in GM3, and antibodies can be specifically generated against the xeno version of the epitope (Krengel et al. 2004), which continue to be investigated as cancer immunotherapeutics (reviewed recently elsewhere Pearce and Laubli 2016). Additionally, fucosylated GMs have also been identified as targets. Synthesis of Fuc-GM1 has been investigated as a vaccine target (Mong et al. 2003). Processing of the GSLs has been well characterized (for recent discussion on glycoprocessing, see Jennemann and Gröne 2013 and Schnaar and Kinoshita 2015). GSL processing enzyme B3GNT5, has been identified as an essential component in the formation of the neoGSLs, associated with many malignancies. In ovarian cancer cell lines B3GNT5 expression was deleted using CRISPR/CAS9 resulting in loss of expression of neoGSL. Interestingly this deletion of B3GNT5 also alterated N-glycan sialyl modifications suggesting the presence of a shared network of processing between GSLs and N-glycan biosynthesis (Alam et al. 2017). Overall the GSLs have historically been targeted for cancer vaccines and the major challenge has been the design of immunogens that can drive antibody class switching to facilitate significant tumor killing. Cancer epitopes within structures common to different core glycans In this section N-acetyllactosamines and Lewis antigens, which can be attached to the core structures of N-, O- and GSLs are discussed. For further background reading on these structures please see elsewhere (Schnaar and Kinoshita 2015; Stanley and Cummings 2015). N-acetyllactosamines and poly-N-acetyllactosamines N-acetyllactosamine units can be added to all three core structures (Stanley and Cummings 2015). If the terminal residue is a GlcNAc, a β1-4GalT or β1-3GalT transferase may add a β4-Gal (type-2 chain, ubiquitously expressed) or β3-Gal (type-1 chain, tissue specific) moiety respectively. The resulting GlcNAc-Gal is termed a N-acetyllactosamine unit, which can be polymerized with N-acetylglucosaminyltransferases to make PLAs. For further background information, please see Stanley and Cummings (2015). PLA on O-glycans: Under the core O-glycans section, malignant transformation generally leads to truncation of O-glycan structure, which was associated with invasive and metastatic potential. However, there may be a cost to pay for these advantages. In cell lines where the Tn antigen is predominantly expressed, the truncation may leave malignant cells more sensitive to both NK cell killing and ADCC (Madsen et al. 2013). In this same study production of tumor expressed mucins were found to protect from immune cell killing resulting from O-glycan truncation. Complementary to these findings, in a separate study in the same year, in tumor cells with intact COSMC biosynthetic pathway showed better resistance to immunosurvellance, and in particular inhibition of NK cell immunity (Okamoto et al. 2013). This was found to be through production of PLA modified core-2 O-glycan’s. In this work MUC1 protein expressed on pancreatic cell lines, was decorated with extended PLAs. These bulky glycan motifs blocked NK cell interaction, through steric hinderance of the natural killer receptor (NKR) and TRAIL on NK cells, with corresponding ligands, NKR-L and DR4, respectively, that are expressed by malignant cells (Figure 2). Core-2 O-glycan synthesis, through upregulation of GCNT1, facilitates the branching of the core-1 structures, and is associated with invasive potential (Kim et al. 2012). C2GnT, which also forms branching core-2 O-glycan structures, is an important step in the elongation of PLAs on O-glycan’s (Suzuki et al. 2012). These PLAs have also been shown to physically block NK cell killing, supporting tumor progression (Suzuki et al. 2012). PLAs on N-glycans: Two specific PLA epitopes have recently been found on N-glycans from ovarian cancer cell lines (Choo et al. 2017). These are oncofetal H type 1 and type 1 LacNAc, both of which are associated with stem or stem-like cells (Choo et al. 2017). In this study these two epitopes were identified using a monoclonal antibody (mAb-A4) which was generated initially against human embryonic stem cells (Choo et al. 2008), and therefore the epitopes may be cancer stem cell markers. PLAs have also been identified on metastatic melanoma cell lines (Kinoshita et al. 2014). Functionally, PLAs on N-glycans may play a role in migration of tumor cells. For example, in mammary morphogenesis, sialylated PLAs have been reported to signal intracellularly through relocation to the nucleus by a galectin-1 interaction, which triggers an invasive phenotype (Bhat et al. 2016), in a mechanism reminiscent of extracellular galectin-1 signaling in anti-VEGF resistant tumors (Croci et al. 2014). PLAs on N-glycan’s can bind galectin-3 within the extracellular space, which facilitates metastasis (André et al. 2004; Srinivasan et al. 2009; Miwa et al. 2013). PLAs on N-glycan’s likely also block immune cell interactions, as seen with PLAs that decorate O-glycan’s. For example, in mice that lacked PLA biosynthesis (β3GnT2 KO), T-cells, B-cells and macrophages no longer expressed PLA on N-glycan’s, which associated with a more immunosensitive phenotype, further suggesting these molecules play an immunoregulatory role (Togayachi et al. 2007). Fucosylation of N-glycan PLAs (Kawasaki et al. 2009; Powlesland et al. 2009) is associated with multiple drug resistant tumor cell lines (Feng et al. 2016). This resistance is associated with FUT4 upregulation (Feng et al. 2016). Fucosylation of sialyl-N-acetyllactosamines in terminal positions on the PLA antenna, generates sialyl Lewis antigens (discussed in Lewis antigens). N-acetyllactosamine units can also be further functionalized through sulfation (Stanley and Cummings 2015). In ovarian cancer an N-acetyllactosamine unit with an altered sulfation pattern, the HMOCC-1 epitope, has been reported (Shibata et al. 2012). The HMOCC-1 epitope is expressed on a core N-glycan. It is formed through overexpression of GAL3ST3, B3GNT7 and CHST1 enzymes (Shibata et al. 2012). This epitope has so far only been identified on an ovarian carcinoma cell line. Lewis antigens Where a terminal β-Gal residue is present from a N-acetyllactosamine unit (that is found on N-glycans, O-glycans or GSLs (Hakomori and Andrews 1970; Hakomori and Jeanloz 1964; Hakomori and Strycharz 1968; Stanley and Cummings 2015), these moieties can be further functionalized with fucose units to make the Lewis blood group antigens (Stanley and Cummings 2015). There are two types which differ in the linkage of the β-Gal terminal residue, either β3 or β4 linked, types 1 and 2, respectively (see the last section on N-acetyllactosamines). Type 1 includes Lewisa (Lea, monofucosylated), and Lewisb (Leb, difucosylated). Type 2 includes Lewisx (Lex, monofucosylated), and Lewisy (Ley, difucosylated) (Stanley and Cummings 2015). Fucosylation of sialyl type 1 or 2 N-acetyllactosamines with FUT3/4 produces SLea and SLex antigens, respectively. Additionally the Lea, SLea, Lex and SLex antigens can also be further modified with sulfates (Stanley and Cummings 2015). Synthesis of the Lewis antigens has been well studied (Stanley and Cummings 2015). The fucosyltransferases are essential in generating the terminal Lewis moiety by addition of the α-linked fucose moiety to the GlcNAc (Lea and Lex) and also the Gal residue (Leb and Ley) of the terminal N-acetyllactosamine moiety. FUT3 generates the Lea antigen, FUT3 and FUT2 generate Leb. Lex and Ley antigens can be synthesized by a combination of FUT4, 5, 6 and 7 (Stanley and Cummings 2015). FUT5 strongly supports type 1 Lewis antigen biosynthesis, and is specific to the type of O-glycan core structure (Holgersson and Löfling 2006). Cytokine/chemokine stimulation is in part responsible for Lewis antigen upregulation within the tumor microenvironment (Figure 1). Recent work has demonstrated FUT enzymes are upregulated in tumor cell lines that are stimulated with various inflammatory cytokines (Bassaganas et al. 2015). In this work, IL-1β stimulated production of FUT5-7 genes to produce Lex. IL-6 and TNFα upregulated ST3GAL3-4, producing SLeX. In another study IL6 and TNFα were found to upregulate FUT1-2 and FUT6, which corresponded with biosynthesis of SLex and Ley (Padró et al. 2011). Ley may also be upregulated through the c-Jun transcription factor (Gao et al. 2014). In keeping with these findings, it has been known for many years that Lewis structures are found expressed mostly in areas where inflammation from infection is present (Mahdavi et al. 2002; Martins et al. 2006; Moran 2008). In this case, the inflammation is driven in response to malignancy. Generally the sialylated Lewis antigens are considered ligands for selectins, and therefore, their expression on malignant cells is associated with increased homing and metastasis. This is discussed in Borsig, L. in this same issue, and therefore selectins in cancer will not be reviewed here. Lea has been detected upregulated on MUC1, and also MUC2 carrier proteins on pancreatic and gastrointestinal cancer cell lines (Pour et al. 1988; Burdick et al. 1997). It also appears to be expressed in precancerous stomach tissues (Kaczmarek 2010), but the SLea epitope is dominantly expressed in malignant transformation. Its expression in malignancy is likely regulated through the FUT3 fucosyltransferase (Escrevente et al. 2006). A functional role for Lea expression in cancer (if any) has not yet been identified. Slea, or CA19-9, has been shown to be upregulated in hypoxic conditions, through the HIF family of molecules (Koike et al. 2004), through the action of FUT3 fucosyltransferase (Escrevente et al. 2006). SLea is synthesized through fucosylation of type 1 sialyl lactose. It is associated with colon (Koprowski et al. 1979, 1981), lung (Togayachi et al. 1999), gastric (Isozaki et al. 1998) and pancreatic cancer (Haglund et al. 1986; Ho et al. 1995). In the latter, SLea has been used as a tumor marker in the management of the disease (Locker et al. 2006). SLea upregulation on malignant cells enhances their adhesive properties through selectin and some extracellular matrix protein binding (Koike et al. 2004), enhancing metastatic potential, though E-selectin (Kłopocki et al. 1996; Kłopocki et al. 1998). Leb has been found expressed on colon and gastrointestinal cancers (Inagaki et al. 1990; Itzkowitz 1992; Murata et al. 1992), and may be expressed where Ley is also present (Noble et al. 2013). Its expression is linked with FUT3 overexpression in malignancy (Escrevente et al. 2006). Unlike Ley (discussed later) the function of Leb (if any) in tumor progression is not well studied. However, it may be more important in premalignant tissue. A recent study demonstrated cancer causing bacterial infections may use Leb expressed on the gastric mucosa, in attachment and colonization (Nell et al. 2014; Hage et al. 2015). Targeting Leb with monoclonal antibodies that also target Ley, may provide a therapeutic advantage through increased specificity and less off-target toxicity (Noble et al. 2013). Lex, also referred to as SSEA-1, is primarily overexpressed in malignant transformation through the upregulation of FUT4 and FUT9 fucosyltransferases (Escrevente et al. 2006), that are stimulated through cytokines TNF and IL1β within the tumor microenvironment (Kaszubska et al. 1993). Lex is a prognostic marker for several cancers including lung (Kadota et al. 1999), bladder (Konety et al. 1997), medulloblastoma (Read et al. 2009), lymphoma (Powlesland et al. 2011) and triple negative breast cancers (Koh et al. 2013). Lex expressing malignant cells are associated with increased proliferation, and a decreased tendancy to differentiate and apoptosis, and therefore may be a putative cancer stem cell marker (Read et al. 2009; Ohtsu et al. 2016). Lex is an adhesion molecule (Gooi et al. 1981), which in normal homeostasis is involved in transepithelial migration of neutrophils into tissues, during infection and inflammation (Brazil et al. 2016). Lex is expressed on several carrier proteins, including CD18, CD11b and CEA (Stocks et al. 1990). Lex motif is associated with lymphoma cell lines (Powlesland et al. 2011), where it is expressed on CD98, DEC-205 and ICAM-1. This may facilitate the interaction of lymphoma cells with DC-SIGN bearing lymphocytes and myeloid cells, which may provide an escape mechanism from immunosurveillance (van Gisbergen et al. 2005). In carcinomas, lex expression may be important in metastasis by binding endothelial scavenger receptor C-type lectin (Elola et al. 2007). As a therapeutic target, anti-Lex (anti-CD15) antibodies have been tested in leukemia (Zhong et al. 1996), and breast cancer (Vredenburgh et al. 1991). SLex: Generally on carcinomas, Slex is carried on mucin O-glycans (Burdick et al. 1997) (Hanisch et al. 1992; Hanski et al. 1995). SLeX, a ligand of E-selectin, is well known to be involved in metastasis (please refer to the review by Borsig, L. in this series). However, there are examples where Slex has other functions outside of selectin binding. For example, Slex can activate c-Met signaling (Gomes et al. 2013). c-Met signaling activates SRC and FAK, that cause the cytoskeletal changes associated with a more adhesive and invasive cell phenotype. In a separate study A SLeX antigen was found to increase cell proliferation, although the mechanism by which this signaling occurred was not elucidated (Yusa et al. 2010). Specific SLex containing glycoforms: HCELL: The ubiquitously expressed cell adhesion molecule CD44, expresses a tetra-antennary N-glycan bearing sialyllactosamines (for a review of CD44 in cancer please see elsewhere Prochazka et al. 2014). A prominent variant of the glycoform of CD44 is the hematopoietic cell E- and L-selectin ligand, HCELL glycoform, where the tetra-antennary N-glycan bearing Sialyllactosamines is fucosylated, producing terminal SLeX moieties (see Lewis antigens) (Dimitroff et al. 2000). HCELL variants are expressed on both N- and O-glycan epitopes found on CD44 (Jacobs and Sackstein 2011). In healthy individuals HCELL is restricted to hematopoietic progenitor cells (Oxley and Sackstein 1994; Sackstein and Dimitroff 2000; Dimitroff et al. 2001). However, it is expressed in several malignancies, including acute myeloid leukemia (Sackstein and Dimitroff 2000), where it is expressed on N-linked glycans as the HCELLs variant, and in breast (Zen et al. 2008) and colon (Hanley et al. 2005; Burdick et al. 2006) cancers, where it is expressed as the O-linked HCELLv variant. Both HCELL variants are strong E and L-selectin ligands (Burdick et al. 2006; Sackstein and Dimitroff 2000), through which malignant cells hijack leukocyte rolling, that aids metastasis (similar to PSGL1, which allows malignant cells to hijack leukocyte rolling though P-selectin, see SLex within the Lewis antigens). Please see the review by Borsig, L in this issue for a detailed discussion on selectins in cancer. The biosynthesis of the HCELL glycoform is generated through upregulation of fucosyltransferases VI and VII (Pachón-Peña et al. 2017), which adds a fucose unit to the lactosamine moieties present on the normal variant CD44 (for a recent review on HCELL, including its biosynthesis see elsewhere (Sackstein 2016) and in cancer, see Jacobs and Sackstein 2011). Tumor expressed HCELL may also be important in the education of tumor immune cell phenotypes in the established tumor microenvironment. For example, recent work has demonstrated transforming CD44 on mesenchymal stem cells into the HCELL glycoform enhanced macrophage homing and polarization in a fibrotic injury model (Chou et al. 2017), or through direct homing of HCELL expressed on leukocytes (Ali et al. 2017), which bind to E- and L-selectin within the tumor microenvironment. PSGL-1: Expresses SLex on core 2 O-glycans. PSGL1 is the selectin binding moiety of a P-selectin ligand, that is often overexpressed in malignant hemopoeitic cancers (Handa et al. 1995; Kappelmayer et al. 2001; Raes et al. 2007; Zheng et al. 2013; Krause et al. 2014). There are also examples of PSGL-1 expressed on solid cancers of epithelial origin, including colon (Krüger et al. 2001), pancreatic (Mathieu et al. 2004; Thomas et al. 2009), lung (Thomas et al. 2009; Gong et al. 2012) and breast (Conrad et al. 2018). PSGL-1 may play a role in tumor cell extravasation from the circulation by hijacking leukocyte rolling (Laubli and Borsig 2010) (please also see the section on the CD44 HCELL glycoform under PLAs, which is also associated with tumor metastasis and leukocyte rolling). However, nontumor expressed PSGL-1 is also important in the immune response to metastatic tumor. For example, PSGL-1 may in part regulate immune cell activation within the tumor microenvironment, though the PD1 checkpoint blockade molecule (Tinoco et al. 2016), or through recruitment of immune cells to the tumor microenvironment (Hoos et al. 2014). Further, platelet activation, in some cases, is an essential part of these interactions. For example, mucins produced by carcinomas aggregate platelet and leukocytes together, though PSGL-1, triggering microthrombi formation (Trousseau syndrome), that is associated with some malignancies (Shao et al. 2011). Platelets can also bind directly to tumor cells, stimulating an EMT phenotype (Labelle et al. 2011), and recruitment of a granulocyte subtype, that together form the early stages of the metastatic niche (Labelle et al. 2014). Further to these findings, platelet derived TGFB1 is secreted upon platelet aggregation to tumor which stimulates both primary and metastatic tumor growth (Hu et al. 2017), and survival through reduced anoikis via YAP signaling (Haemmerle et al. 2017). In the established tumor, platelets stimulate angiogenesis in the hypoxic tumor microenvironment (Haemmerle et al. 2016). Whilst PSGL-1 expression seems essential in the tumor–platelet–leukocyte interaction, it is unclear whether carcinoma expressed PSGL-1 per say is important in the mechanisms described, and HCELL expression may be more prevalent in direct tumor–platelet interactions through p-selectin (Hanley et al. 2006; Alves et al. 2008). Nevertheless, strategies that target PSGL-1 molecular interactions may have translational potential. One such approach to inhibit PSGL-1 was achieved using a small molecule synthetic nucleotide sugar analogue to downregulate cell surface expression of PSGL-1 (Kanabar et al. 2016). PSGL-1 inhibition as a therapeutic approach faces several challenges, including unwanted protumour inflammation (Li et al. 2017), and in the case of selectin mediated metastasis, tumor may bind in a PSGL-1 and/or HCELL independent manner (Goetz et al. 1996; Ma and Geng 2002). Ley: The difucosylated type 1 Ley antigen is upregulated on many cancer types, including ovarian (Yin et al. 1996), breast, lung (adenomas and squamous cell) (Miyake et al. 1992; Westwood et al. 2009), prostate (Zhang et al. 1997) and colorectal (Sakamoto et al. 1986). Ley expression in carcinomas appears to be carried mostly on mucins (including MUC1 and CA125 Yin et al. 1996). In normal tissues, the distribution of the Ley antigen appears restricted to epithelial cells, and in low levels (Zhang et al. 1997). In disease Ley signals expression of growth factors (Liu et al. 2011), including TGFB1 (Wang, Liu et al. 2012), matrix metalloproteases (Yan et al. 2010) and mucins (Hou et al. 2017). Ley expression therefore promotes growth, adhesion and invasion of malignant cells (Li et al. 2010; Yan et al. 2015). Because Ley is restricted in its expression, there has been significant interest in the development of molecules that can detect it (Kim et al. 1988; Wang et al. 2017). Therapeutically Ley is a promising target for genetically redirected T-cells (Westwood et al. 2005, 2009; Peinert et al. 2010), and mAb targeting (Herbertson et al. 2009; Noble et al. 2013; Smaletz et al. 2015; Hutchins et al. 2017). Sialic acid Malignant transformation is associated with hypersialylation of all the major glycan classes, except GAGs, discussed here. Sialylated cancer epitopes, their biosynthesis and proposed role in cancer progression have been recently reviewed elsewhere (Pearce and Laubli 2016), and therefore will not be discussed further here. Glycan polymer epitopes In this section, specific GAG cancer epitopes are summarized, and the role of hyaluronan is discussed at the end. For further information on GAGs and cancer progression, please see elsewhere (Blackhall et al. 2001; Hascall and Esko 2015; Lindahl et al. 2015). Glycosaminoglycans Heparin sulfates (HS): Are found on most proteoglycans. They form part of the basement membrane which separates the epithelium or mesothelium from the underlying connective tissue (Hynes and Naba 2012). Within the tumor microenvironment, HS bind cytokines/chemokines and growth factors, which set up gradients to attract immune cells. HS are also involved in the fine tuning of inflammatory processes, as shown recently through IFNγ signaling in macrophages (Gordts and Esko 2015). HS are ubiquitously expressed on the membrane proteoglycans of all cells. The sulfation status of HS dictates much of the molecules biological function (Lindahl et al. 2015). So far sulfated HSNS4F5 (GlcNS6S-IdoA2S)3 epitope of HS has been identified on melanoma and ovarian cancer cell lines (Smits et al. 2010; van Wijk et al. 2014). HSNS4F5 has restricted expression in adult tissues (Smits et al. 2010), but has been found on endothelial cells stimulated with inflammatory cytokines, aiding proliferation and adhesion (Smits et al. 2010). Therefore, HSNS4F5 may potentially help drive extravasation of immune cells within the tumor microenvironment. Sulfation is also associated with chondroitin sulfate (CS) GAGs (Basappa et al. 2009). Several specific cancer epitopes of CS sulfation have so far been identified. These are discussed below. The E-unit: E-units, a disulfated form of the GlcA-GalNAc repeating dissacharide promotes metastasis and invasion, through an initial improvement in adhesion (Basappa et al. 2009). Intervention with “unbound” E-units, or antibodies directed against the E-unit were able to block adhesion, and tumor invasion (Li et al. 2008). In this same study, murine lung carcinoma cell lines expressing high amounts of the E-unit associate with a highly invasive and metastatic phenotype, suggesting the potential for targeting these molecules, or using them as prognostic markers. MCSP: High molecular weight melanoma associated antigen (or melanoma associated chondroitin sulfate proteoglycan, MCSP) (Ross et al. 1983) is another example of a sulfated CS epitope. MCSP has been shown to aid attachment of melanoma cells, through integrin α4β1 (Iida et al. 1998). MCSP has been the focus of T-cell targeting responses as a melanoma specific antigen (Erfurt et al. 2007) (Erfurt et al. 2009). However, the precise structure of the glycan moieties and whether these are relevant in this context have not been investigated. The core protein peptide structure at this time is the focus of attention (Geiser et al. 1999). WF6: In ovarian cancer, the WF6 CS isotope (Pothacharoen et al. 2006) was shown to be raised in all five subtypes of ovarian cancer (Bowtell et al. 2015), however, like MCSP discussed above, the description of the epitope is limited to the core protein currently, and no detailed information on the glycoform (other than it is sulfated) has yet been reported. High sulfation on the GAG chain, nevertheless, does seem to be important in describing these GAG epitopes, which can distinguish benign from malignant cancer tissues (van der Steen et al. 2016). Hyaluronan (HA): Is an important part of the tumor microenvironment and has more recently been connected with significant developments in our understanding of malignant transformation. HA is synthesized on the inner surface of the plasma membrane, by hyaluronic acid synthase, and is not attached to either protein or lipid (Hascall and Esko 2015). In most mammals, the resulting polymer is approximately 104 disaccharides in size, roughly half the circumference of a cell (Hascall and Esko 2015). It is a major constituent of tissue matrisomes, with the average human containing about 15 g of HA, which is rapidly recycled. Whilst HA is present in large amounts in healthy tissues, it is nevertheless enriched further in tumor tissue (Hascall and Esko 2015). Classically HA has been shown to act as a barrier to block the diffusion of larger molecules, whilst allowing small molecules to diffuse freely. In normal homeostasis, HA has multiple roles, which is dependent on the size of the polymeric form, including tissue organization and development, and cell proliferation. It also acts as a backbone binding specific proteins within the tissue microenvironment (Hascall and Esko 2015). HA signals though CD44 (CD44 itself also bears altered glycoforms in cancer, the HCELL isoform discussed earlier. Please see the section on N-acetyllactosamines on N-glycans), which is an essential process in embryonic development (for a review on HA in mammalian reproduction and embryo development, see Fouladi-Nashta et al. 2017), tissue healing and regeneration (Damodarasamy et al. 2014). In the former the HA-CD44 signaling pathway effects cell proliferation and survival, but this pathway is thought to have little activity in normal adult tissues (Hascall and Esko 2015). In cancer, HA-CD44 signaling is activated providing proliferation and survival cues (Ghatak et al. 2005; Misra et al. 2008). HA may also convey resistance to some cancer therapies (Bourguignon et al. 2008). Antagonists that target HA-CD44 signaling may then sensitize cancers to chemotherapies. More recently very high molecular weight HA (HMW-HA), in the naked mole rat, was found to protect from malignant transformation (Tian et al. 2013). In this work the authors found naked mole rat fibroblasts secreted HA that was over 5-fold longer than human or mouse HA, which accumulated in abundance in naked mole rat tissues. Cells from naked mole rat were more sensitive to HA signaling than human or mouse (Tian et al. 2013). Importantly, naked mole rat fibroblasts require treatment with hyaluronidases, in addition to knockdown of oncosuppressor genes p53 and pRb, which together are usually enough to confer malignant transformation in mouse cell lines. Work continues to completely understand this observation, but this may be due to the role of HMW-HA signaling in the maintenance of stem cell niches or stabilization of cell dedifferentiation, that would also fit with the increased longevity of the species as well as its resistance to malignant transformation (Tan et al. 2017). Xeno epitopes Xeno epitopes are glycans that are not endogenously interconverted in humans, either through the carbohydrate structure, or the linkage by which it is attached. Two prominent xeno epitopes are presented here. The α-Gal epitope is an example of a glycan expressed attached through a linkage not created in humans, whereas the Neu5Gc epitope is an example of a carbohydrate with a structure not endogenously made in humans. The α-Gal epitope is well known for its association with tissue graft rejection through antibody mediated cell killing (Sandrin and McKenzie 1994). Anti-α-gal antibodies in humans are very abundant, making up approximately 1% of all immunoglobins. There has therefore been much interest in exploiting this epitope to induce tumor rejection, usually through targeting large amounts of multivalent displays of the epitope to the site of tumor. Multivalent displays of α-Gal can be acquired from rabbit red blood cell ghosts (Galili et al. 2007), engineered whole cells (Rossi et al. 2005), or chemically synthesized displays (Carlson et al. 2007). In all strategies, a localized immune reaction and tumor killing through complement dependent cytotoxicity and ADCC can be achieved in murine and in vitro models. For further reading, please see Tanemura et al. (2013). The Neu5Gc antigen has been recently reviewed elsewhere (Okerblom and Varki 2017), and therefore only a summary is included here. Whilst at first glance this molecule may appear to be similar in its biology to α-Gal they are quite different, as summarized in more detail elsewhere (Pearce et al. 2015). The “alpha linkage” in α-Gal, to which humans have circulating antibody is not retained upon metabolism of the molecule. Neu5Gc however behaves as a “trojan horse” xenoantigen, and is tolerated though cellular bioprocessing (the “Gc” moiety also ends up in other carbohydrate monomers as part of monosaccharide conversion), becoming incorporated, and displayed upon (Bergfeld, Pearce, Diaz, Lawrence et al. 2012; Bergfeld, Pearce, Diaz, Pham et al. 2012; Bergfeld et al. 2017), displayed upon the cell glycocalyx. At this point it is recognized as foreign, and an inflammatory response is generated through naturally occurring anti-Neu5Gc antibodies, which like anti-α-Gal, make up a significant part of human circulating IgG (Padler-Karavani et al. 2008). Dietary sources particularly high in Neu5Gc include red meats, cheese (in particular goats cheese) and caviar (Samraj et al. 2015). In a murine model of human Neu5Gc deficiency, a diet high in Neu5Gc increased the risk of cancer development approximately 4-fold, over mice fed a Neu5Gc null diet (Samraj et al. 2015). These finding provide a red meat specific mechanism, whereby a dietary xeno antigen explains the associated cancer risk. Glycoforms in cancer diagnostics Almost all cancer biomarkers are tumor expressed glycoproteins, or glycolipids, and therefore understanding the protein, glycosylation site, and glycan present could improve specificity and sensitivity, for diagnostics. In this section, brief examples of how specific glycan epitopes, and patterns of glycosylation, are an improvement over the traditional biomarkers are discussed. At the end, the potential advantages of glycoform detection, and the technologies to detect them are discussed. Specific glycan epitopes Whilst detection of glycoforms potentially offer high selectivity, detection of solely the glycan epitope still shows marked improvement over the protein carrier (Ju et al. 2016), both in specificity and sensitivity (Robbe-Masselot et al. 2009; Biskup et al. 2013). For example, prostate-specific antigen, used to diagnose prostate cancer, is associated with high incidence of false positives, because it is not disease specific, and its blood concentration varies under normal homeostasis (Roetzheim and Herold 1992). However, the RM2 GSL antigen provides a significantly more reliable diagnostic readout, because it is disease specific (Saito et al. 2005). The type of glycan present may also be useful in distinguishing grade and stage of disease (Chen et al. 2014; Vitiazeva et al. 2015), including early detection of cancers (Samuel et al. 1990; Remmers et al. 2013). For example, T-antigen which is expressed in adult colon, increases in expression in ulcerative colitis, prior to malignancy (Campbell et al. 1995), and the CS epitope “E-unit” (discussed under Glycosaminoglycans), is potentially useful in the subtyping of ovarian cancers (Vallen et al. 2012). Detection of functionalization on epitopes Modifications such as acetylation and sulfation are sometimes disease specific. For example, in breast cancer, a 6-sulfo modification of the T-antigen has been identified (Seko et al. 2012). Antibodies against the 6-sulfo variant are highly specific, and do not detect the normal 3-sulfo variant (seen in both normal and diseased breast tissue) (Seko et al. 2012). Patterns of glycosylation Detection of multiple glycan epitopes in parallel offers specificity advantages over one glycan moiety (Tang et al. 2015). As mentioned briefly earlier in the N-glycans section, the N-glycome can be used to fingerprint cancer cells, offering improved sensitivity and specificity (Hua et al. 2014). N-glycome profiling for diagnostics is showing promise in colon (Balog et al. 2012; Park et al. 2012), ovarian (Saldova et al. 2013) (Arnold et al. 2008), pancreatic (Zhao et al. 2017) and prostate (Jorgensen et al. 1995). In the latter, altered N-glycan glycoforms, and SLeX antigens, on the carrier protein prostate specific antigen, are much more reliable at detecting disease than the carrier protein (Jorgensen et al. 1995; Peracaula et al. 2003). Towards glycoform recognition for diagnostics Biomarkers that recognize the glycoform of the epitope, as opposed to only the core protein or glycan (Silsirivanit et al. 2013; Tanaka-Okamoto et al. 2016; Zhao et al. 2017), could improve specificity for diagnostics and prognostics. For example, in a recent study of serum proteins from pancreatic cancer patients, glycoforms on four proteins were characterized (Drabik et al. 2017). The authors characterized the N-glycan binding site (identifying some unusual N-glycan sites) and the N-glycan motif present. Combining these analyses provided an extremely high sensitivity and specify for cancer detection (Drabik et al. 2017). This work is an example of identifying unique disease-associated glycoforms, that do not appear to be expressed in normal tissues, at least not in detectable amounts. A major challenge to advancing the application of glycoforms to diagnostics, is the development of methods that allow the simultaneous identification of the core protein, glycosylation site and glycan epitope. One such approach, in situ proximity ligation assays (Gremel et al. 2013; Raykova et al. 2016), are in development for several well-known tumor expressed mucins (Pinto et al. 2012). In this approach, primary antibodies against a glycan epitope and a core protein are incubated with human tissue samples. The 2° antibodies with short DNA strands attached (the proximity ligation probes), are then added. In places where the glycan and protein epitopes are present together, the DNA strands can be amplified via rolling circle DNA synthesis, therefore dramatically amplifying the amount of DNA at these sites of close proximity. Amplified DNA are then detected with complementary fluorescently labeled oligonucleotides. This technique therefore provides a useful way to both identify specific glycoforms whilst also providing locational information for glycoform presence within the tissue (Pinto et al. 2012). This technique could be used in tandem with glycoproteomics to aid detection of useful, diagnostic glycoforms (Barallobre-Barreiro et al. 2016; Yang et al. 2017). Emerging concepts As described in this review, glycan epitopes play significant functional roles in disease progression, and therefore have high potential value for cancer therapy and diagnostics. The major attraction of targeting the glycoform is the high specificity offered. Cancer therapy and diagnostics is now starting to capitalize on the advantages that the glycoform offers. For example, a new study has demonstrated the importance of the glycosylation status of PD-L1 for contact with its receptor PD-1 (Li et al. 2018). Here, antibody targeting of the PD-L1 glycoform caused internalization of the ligand, which could be used in a antibody-drug carrier system to induce tumor cell death, specifically, and relatively safely in murine model of triple negative breast cancer (Li et al. 2018). This study demonstrates the translational potential offered by considering tumor ligands as glycoforms, which comes partly through understanding the glycan processing involved. In this last part of the review, in silico modeling of the cancer glycome as a whole, is discussed, including the functional information this can produce, and how we might use this information to understand glycoprocessing, with application in identifying targets for the next generation of therapeutics. Cancer glycoform structure, pattern and bulk glycocalyx Patterns or clusters of glycan epitopes generally convey their biological activity (Springer et al. 1983; Inoue et al. 1994). These aberrant glycan patterns are generally recognized by receptors on cells of the innate immune response, and classically are described as damage associated molecular patterns, pathogen associated molecular patterns and potentially self-associated molecular patterns (Varki 2011). To further complicate matters, the altered glycan pattern may result from changes in multiple glycan types. For example, tumor-associated mucins can have multiple glycan types attached, and as shown recently, both N- and O-linked glycan’s can be altered on the same mucin (Saeland et al. 2012). In this example, truncation of O-glycan’s, branching of N-glycan’s, and overall upregulation of terminal sialic acid on both was found (Saeland et al. 2012). An understanding of the overall pattern and the biological, chemical and physical information that this presents could provide information to better inform the development of therapeutics. An excellent example of one such approach to this used in silico models to predict biological function from physical changes, associated with the bulk glycocalyx of malignant cells (Paszek et al. 2014). In this work the bulky glycocalyx was determined from gene expression data, and a model constructed which predicted how the bulky glycocalyx would cluster receptors together, based on generation of a kinetic trap. For example, MUC1 was one such glycoprotein found to drive receptor clustering. The authors tested the bulky glycocalyx concept by intercalating glycomimetic mucin structures into nonmalignant breast epithelial cell lines, which clustered integrins supporting tumor metastasis and survival (Paszek et al. 2014). In this way genetic, chemical and physical information was used to explore biological function of the overall cell membrane glycan pattern. This example explains how the whole glycocalyx may organize itself and work together. Other models are concerned with understanding how the glycome is regulated. In this sense glycan processing is often described as being separate from template driven processes such as RNA and protein synthesis (although there is certainly a template driven part to glycan synthesis). As shown in Figure 3, a simplified overview of the processes to consider in the formation of a glycoprotein, there are three arms to the glycan processing machinery. Firstly, there is template driven production of the protein backbone, the glycan processing enzymes and scavenging receptors (the red line). Sugars are acquired from dietary sources and recycling of the glycocalyx. The acquired sugars are transported and interconverted to other monomer forms, before introduction within the glycosylation pathway (the yellow line). Finally, there are environmental considerations which act upon both these processes (the purple line). Whilst it is not shown in Figure 3 for simplicity, differences in enzyme kinetics will likely play a role (Paquet et al. 1984). However, to understand altered glycome regulation in cancer, do all of these processes need to be considered? In other words, can expression levels of glycan processing enzymes alone explain the cancer glycome, including biosynthesis of unusual cancer glycan epitopes? In this review there are examples where glycan epitopes are explained simply by upregulation of processing enzymes (Julien et al. 2006; Picco et al. 2010; Chen et al. 2014), or altered enzyme location or mutation (Yang et al. 1994; Ju and Cummings 2002; Gill et al. 2013; Hofmann et al. 2015), and finally, examples where environmental mediators effect glycan biosynthesis (Hassinen et al. 2011; Padró et al. 2011; Bassaganas et al. 2015). To help us better understand glycome regulation, computational models are in development. Generally these models are applied to controlling glycome heterogeneity in biopharmaceutical products (Hossler et al. 2007; Liu and Neelamegham 2014; St Amand et al. 2014; Krambeck et al. 2017). These in silico reconstructions model the Golgi as a set of reaction vessels, where virtual proteins spend time mixing with other virtual proteins, glycans and processing enzymes. In these examples, gene expression levels of glycan processing enzymes alone seem to fit the model very well. Some models have been applied to investigate the cancer glycome. In one study, the N-glycome in cancer was investigated using gene expression values that were integrated with glycome data (Bennun et al. 2013). As part of this study enzyme gene array data was used to make calculated mass spectral data for glycans, which correlated very well with measured observations (Bennun et al. 2013). In a separate study, enzyme catalyzed reactions for O-glycosylation were modeled as enzyme-reaction networks, using 25 simulated O-glycan processing enzymes (McDonald et al. 2016). The model was validated using experimentally determined O-glycomes from literature datasets. To validate the model, its ability to reverse glycosylate experimental O-glycomes was investigated, with an overall success rate of 87% for unique glycan structures (McDonald et al. 2016). These studies would seem to suggest that glycan processing, including production of cancer epitopes (e.g., Tn, STn) could mostly be predicted from gene data of glycan processing enzymes alone, and that altered glycosylation resulting from locational changes of enzymes, or other factors outside of expression levels, must be less frequent (but no less important). Currently in silico models of glycan processing do not include glycoproteomics, which would allow glycoform biosynthesis to be investigated. There is also no locational information, for example, in a cancer tissue it would be useful to know where particular glycoforms are expressed. Further, regulation of glycosylation is a relatively fast dynamic process (turnover of glycosylation could be achieved within minutes, to a few hours based on one in silico model Hossler et al. 2007), allowing a cell or organism to adapt quickly to a biological pressure without having to adapt its genome. For example, a mammalian pathogen will alter its coat glycoproteins (and vice versa) in response to changing immunological pressures; a host vs. pathogen “arms-race” (Coss et al. 2016). Similarly, this arms race may also manifest itself in the interaction of malignant cells with the host response. Therefore, studies which aim to understand how to disrupt or control glycan processing may accelerate discovery of targets for cancer therapeutics. One approach to study glycan processing could be integration of glycomics or ideally glycoproteomic databases against several other cellular processes within the tumor microenviroment, including transcription (inc. epigenetic alterations) (Menni et al. 2013), translation (Naba et al. 2017), metabolomics (for a review on latest technology, see Buescher et al. 2015), cellular composition (Mlecnik et al. 2016), biophysical computational models (Paszek et al. 2014) and imaging (Powers et al. 2015; Hadjialirezaei et al. 2017). The latter is important to identify both the location of enzymes and the spatial distribution of a glycan epitope within the tissue. In particular mass spectrometry imaging could be used, along with more traditional immunohistochemical analysis including lectins, to simultaneously identify the glycan, and its distribution within tissues (Powers et al. 2015). Whilst gathering all these data is daunting, all of the analytical techniques to analyze the processes or features of glycoform biosynthesis are available (Figure 3). Additionally, methods to map “glycosites” within the O-glycome (Steentoft et al. 2013; Vakhrushev et al. 2013) are in development, which provide site specific information on where glycan epitopes are attached within a protein backbone. Therefore, the challenge is to combine these data together as an integrated whole to model the process of glycoform regulation. Machine learning approaches, that use algorithms to model large or small databases to identify patterns and predictive variables, is a tool already being employed to better understand “omics” datasets across many scientific disciplines, including analysis of cancer databases such as gene expression vs. drug sensitivity profiles (Huang et al. 2017), and the diagnostic potential of circulating RNAs in cancer patients with very high selectivity (Elias et al. 2017). In our own work, we applied machine learning to investigate how higher order features, such as tissue biomechanics and architecture of the tumor microenvironment are formed (Pearce et al. 2018). This approach identified a group of 22 molecules consisting of glycoproteins, proteoglycans, and collagens that define the composition of a tumor matrisome that supports disease progression, and appears to be a conserved feature of many cancers. Machine learning could be applied to the processes involved in glycoform regulation (Figure 3). The advantage of this approach over the others described here, would be the identification of targets outside of the normal glycoprocessing pathway, a more comprehensive overview of glycome regulation, and potentially new therapeutic targets. Fig. 3. View largeDownload slide Schematic of the processes involved (white boxes) in glycoform biosynthesis. The red path indicates the synthesis of the protein backbone and glycan processing enzymes and molecules (template driven). The purple path shows the contribution of extracellular signaling which acts upon the red path. The yellow path shows the introduction of the glycan moiety into this process (not template driven). All three paths play a significant role the final glycoform that is secreted or displayed within the extracellular space. The methods to analyze these processes are shown in grey. Integration (green box) of these datasets could be used to understand how a single glycoform, pattern of glycoforms, or the glycocalyx as a whole are generated. Fig. 3. View largeDownload slide Schematic of the processes involved (white boxes) in glycoform biosynthesis. The red path indicates the synthesis of the protein backbone and glycan processing enzymes and molecules (template driven). The purple path shows the contribution of extracellular signaling which acts upon the red path. The yellow path shows the introduction of the glycan moiety into this process (not template driven). All three paths play a significant role the final glycoform that is secreted or displayed within the extracellular space. The methods to analyze these processes are shown in grey. Integration (green box) of these datasets could be used to understand how a single glycoform, pattern of glycoforms, or the glycocalyx as a whole are generated. Summary Changes in glycosylation with disease occur across all of the major glycan groups. These changes included truncation, elongation, branching, and functionalization. However, the changes that occur still obey the rules governing glycan regulation which would indicate the processing enzymes involved retain their specificity, but their location in some cases is altered. The altered glycosylation seen is also predictable in most cases, based on the expression or mutation of particular enzymes in the regulatory process. As shown in Figure 1, aberrant glycan processing can be committed down a particular biosynthetic route. For example, in N-glycan processing addition of the bisecting GlcNAc inhibits the branching pathway, and sialylation of Tn antigen pathway inhibits any further procession to core-1, or core-2 epitopes. On the other hand, some epitopes are generated in many cancers but through different glycan processing routes, such as Tn, STn, T, SLex, etc., suggesting these epitopes provide biological function that is particularly useful to survival. Whilst our overall picture of glycoform regulation still remains unclear, their biological function is easier to summarize. From what we have reviewed here cancer glycan epitope function fit into four of the hallmarks of cancer categories (Hanahan and Weinberg 2011), but mostly either “activating invasion and metastasis”, “avoiding immune disruption” and “tumor promoting inflammation” (Figure 4). In the latter the xeno antigens, which come under “tumor promoting inflammation”, could be used therapeutically in significant quantity to tip the inflammatory curve (Pearce et al. 2014) from tumor promoting to tumor inhibiting. Fig. 4. View largeDownload slide Adaption of the hallmarks of cancer figure (adapted from the original figure presented elsewhere (Hanahan and Weinberg 2011) to show where cancer glycan epitopes have currently had a strong involvement. There are other examples where glycan’s are involved in these hallmark processes, such as angiogenesis, however, this figure highlights aberrant glycan epitope involvement only. Fig. 4. View largeDownload slide Adaption of the hallmarks of cancer figure (adapted from the original figure presented elsewhere (Hanahan and Weinberg 2011) to show where cancer glycan epitopes have currently had a strong involvement. There are other examples where glycan’s are involved in these hallmark processes, such as angiogenesis, however, this figure highlights aberrant glycan epitope involvement only. Glycoproteins produced my malignant cells have a long history of use as biomarkers in the clinic, and still have significant potential to increase specificity and sensitivity for cancer prognosis and diagnosis, including early detection (early detection is discussed further elsewhere RodrÍguez et al. 2018). Currently, either the protein backbone or the glycan motif are used separately. However, recent work reviewed here promise a dramatic improvement in cancer diagnostics and prognostics, in the near future. Finally, recent advances in how we view the cancer glycan epitope, as the cancer glycome, is providing new insights into our basic understanding of cancer glycobiology, and in the long term potentially targets for new therapeutic approaches. Acknowledgements Thanks to Dr. Katie Doores (Kings College, London), Dr. Richard Beatson (Kings College, London) and Dr. Lingquan Deng (GlycoMimetics, Inc.) for critical reading of the article. Dr. Richard Beatson (Kings College, London) for design of Figure 2, and Christina Corbaci (The Scripps Research Institute, La Jolla, CA) for graphical design of Figures 1–4. Funding O. M. T. P is a recipient of an Against Breast Cancer grant (Registered Charity No. 1121258). Conflict of interest statement None declared. Abbreviations ADCC antibody dependent cell-mediated cytotoxicity CS chondroitin sulfate GAGs glycosaminoglycans GH Globo H GSLs glycosphingolipids HA Hyaluronan HMW-HA high molecular weight HA MCSP melanoma associated chondroitin sulfate proteoglycan MT1-MMP membrane type I matrix metalloproteinase NKR natural killer receptor PLAs poly-N-acetyllactosamines TAM tumor-associated macrophage VNTR variable number of tandem repeat References Aird I, Bentall HH, Roberts JA. 1953. A relationship between cancer of stomach and the ABO blood groups. Br Med J . 1: 799– 801. Google Scholar CrossRef Search ADS PubMed Alam S, Anugraham M, Huang YL, Kohler RS, Hettich T, Winkelbach K, Grether Y, Lopez MN, Khasbiullina N, Bovin NV et al. . 2017. Altered (neo-) lacto series glycolipid biosynthesis impairs alpha2-6 sialylation on N-glycoproteins in ovarian cancer cells. Sci Rep . 7: 45367. Google Scholar CrossRef Search ADS PubMed Ali AJ, Abuelela AF, Merzaban JS. 2017. An analysis of trafficking receptors shows that CD44 and P-selectin glycoprotein ligand-1 collectively control the migration of activated human T-cells. Front Immunol . 8: 492. Google Scholar CrossRef Search ADS PubMed Allam H, Aoki K, Benigno BB, McDonald JF, Mackintosh SG, Tiemeyer M, Abbott KL. 2015. Glycomic analysis of membrane glycoproteins with bisecting glycosylation from ovarian cancer tissues reveals novel structures and functions. J Proteome Res . 14: 434– 446. Google Scholar CrossRef Search ADS PubMed Alves CS, Burdick MM, Thomas SN, Pawar P, Konstantopoulos K. 2008. The dual role of CD44 as a functional P-selectin ligand and fibrin receptor in colon carcinoma cell adhesion. Am J Physiol Cell Physiol . 294: C907– C916. Google Scholar CrossRef Search ADS PubMed St. Amand MM, Tran K, Radhakrishnan D, Robinson AS, Ogunnaike BA. 2014. Controllability analysis of protein glycosylation in CHO cells. PLoS One . 9: e87973. Google Scholar CrossRef Search ADS PubMed André S, Kozár T, Schuberth R, Unverzagt C, Kojima S, Gabius HJ. 2007. Substitutions in the N-glycan core as regulators of biorecognition: The case of core-fucose and bisecting GlcNAc moieties. Biochemistry . 46: 6984– 6995. Google Scholar CrossRef Search ADS PubMed André S, Unverzagt C, Kojima S, Frank M, Seifert J, Fink C, Kayser K, von der Lieth CW, Gabius HJ. 2004. Determination of modulation of ligand properties of synthetic complex-type biantennary N-glycans by introduction of bisecting GlcNAc in silico, in vitro and in vivo. Eur J Biochem . 271: 118– 134. Google Scholar CrossRef Search ADS PubMed Anugraham M, Jacob F, Nixdorf S, Everest-Dass AV, Heinzelmann-Schwarz V, Packer NH. 2014. Specific glycosylation of membrane proteins in epithelial ovarian cancer cell lines: Glycan structures reflect gene expression and DNA methylation status. Mol Cell Proteomics . 13: 2213– 2232. Google Scholar CrossRef Search ADS PubMed Arnold JN, Saldova R, Hamid UM, Rudd PM. 2008. Evaluation of the serum N-linked glycome for the diagnosis of cancer and chronic inflammation. Proteomics . 8: 3284– 3293. Google Scholar CrossRef Search ADS PubMed Axelsson MA, Karlsson NG, Steel DM, Ouwendijk J, Nilsson T, Hansson GC. 2001. Neutralization of pH in the Golgi apparatus causes redistribution of glycosyltransferases and changes in the O-glycosylation of mucins. Glycobiology . 11: 633– 644. Google Scholar CrossRef Search ADS PubMed Balog CI, Stavenhagen K, Fung WL, Koeleman CA, McDonnell LA, Verhoeven A, Mesker WE, Tollenaar RA, Deelder AM, Wuhrer M. 2012. N-glycosylation of colorectal cancer tissues: A liquid chromatography and mass spectrometry-based investigation. Mol Cell Proteomics . 11: 571– 585. Google Scholar CrossRef Search ADS PubMed Bapu D, Runions J, Kadhim M, Brooks SA. 2016. N-acetylgalactosamine glycans function in cancer cell adhesion to endothelial cells: A role for truncated O-glycans in metastatic mechanisms. Cancer Lett . 375: 367– 374. Google Scholar CrossRef Search ADS PubMed Barallobre-Barreiro J, Lynch M, Yin X, Mayr M. 2016. Systems biology-opportunities and challenges: The application of proteomics to study the cardiovascular extracellular matrix. Cardiovasc Res . 112: 626– 636. Google Scholar CrossRef Search ADS PubMed Bard F, Chia J. 2017. Comment on “The GalNAc-T Activation Pathway (GALA) is not a general mechanism for regulating mucin-type O-glycosylation”. PLoS One . 12: e0180005. Google Scholar CrossRef Search ADS PubMed Basappa, Murugan S, Sugahara KN, Lee CM, ten Dam GB, van Kuppevelt TH, Miyasaka M, Yamada S, Sugahara K. 2009. Involvement of chondroitin sulfate E in the liver tumor focal formation of murine osteosarcoma cells. Glycobiology . 19: 735– 742. Google Scholar CrossRef Search ADS PubMed Bassaganas S, Allende H, Cobler L, Ortiz MR, Llop E, de Bolos C, Peracaula R. 2015. Inflammatory cytokines regulate the expression of glycosyltransferases involved in the biosynthesis of tumor-associated sialylated glycans in pancreatic cancer cell lines. Cytokine . 75: 197– 206. Google Scholar CrossRef Search ADS PubMed Battula VL, Shi Y, Evans KW, Wang RY, Spaeth EL, Jacamo RO, Guerra R, Sahin AA, Marini FC, Hortobagyi G et al. . 2012. Ganglioside GD2 identifies breast cancer stem cells and promotes tumorigenesis. J Clin Invest . 122: 2066– 2078. Google Scholar CrossRef Search ADS PubMed Beatson R, Maurstad G, Picco G, Arulappu A, Coleman J, Wandell HH, Clausen H, Mandel U, Taylor-Papadimitriou J, Sletmoen M et al. . 2015. The breast cancer-associated glycoforms of MUC1, MUC1-Tn and sialyl-Tn, are expressed in COSMC wild-type cells and bind the C-type lectin MGL. PLoS One . 10: e0125994. Google Scholar CrossRef Search ADS PubMed Beatson R, Tajadura-Ortega V, Achkova D, Picco G, Tsourouktsoglou TD, Klausing S, Hillier M, Maher J, Noll T, Crocker PR et al. . 2016. The mucin MUC1 modulates the tumor immunological microenvironment through engagement of the lectin Siglec-9. Nat Immunol . 17: 1273– 1281. Google Scholar CrossRef Search ADS PubMed Bennun SV, Yarema KJ, Betenbaugh MJ, Krambeck FJ. 2013. Integration of the transcriptome and glycome for identification of glycan cell signatures. PLoS Comput Biol . 9: e1002813. Google Scholar CrossRef Search ADS PubMed Bergfeld AK, Lawrence R, Diaz SL, Pearce OMT, Ghaderi D, Gagneux P, Leakey MG, Varki A. 2017. N-glycolyl groups of nonhuman chondroitin sulfates survive in ancient fossils. Proc Natl Acad Sci USA . 114: E8155– E8164. Google Scholar CrossRef Search ADS PubMed Bergfeld AK, Pearce OM, Diaz SL, Lawrence R, Vocadlo DJ, Choudhury B, Esko JD, Varki A. 2012. Metabolism of vertebrate amino sugars with N-glycolyl groups: Incorporation of N-glycolylhexosamines into mammalian glycans by feeding N-glycolylgalactosamine. J Biol Chem . 287: 28898– 28916. Google Scholar CrossRef Search ADS PubMed Bergfeld AK, Pearce OM, Diaz SL, Pham T, Varki A. 2012. Metabolism of vertebrate amino sugars with N-glycolyl groups: Elucidating the intracellular fate of the non-human sialic acid N-glycolylneuraminic acid. J Biol Chem . 287: 28865– 28881. Google Scholar CrossRef Search ADS PubMed Bergstrom K, Liu X, Zhao Y, Gao N, Wu Q, Song K, Cui Y, Li Y, McDaniel JM, McGee S et al. . 2016. Defective intestinal mucin-type O-glycosylation causes spontaneous colitis-associated cancer in mice. Gastroenterology . 151: 152– 164 e111. Google Scholar CrossRef Search ADS PubMed Bhat R, Belardi B, Mori H, Kuo P, Tam A, Hines WC, Le QT, Bertozzi CR, Bissell MJ. 2016. Nuclear repartitioning of galectin-1 by an extracellular glycan switch regulates mammary morphogenesis. Proc Natl Acad Sci USA . 113: E4820– E4827. Google Scholar CrossRef Search ADS PubMed Bhaumik M, Harris T, Sundaram S, Johnson L, Guttenplan J, Rogler C, Stanley P. 1998. Progression of hepatic neoplasms is severely retarded in mice lacking the bisecting N-acetylglucosamine on N-glycans: Evidence for a glycoprotein factor that facilitates hepatic tumor progression. Cancer Res . 58: 2881– 2887. Google Scholar PubMed Biskup K, Braicu EI, Sehouli J, Fotopoulou C, Tauber R, Berger M, Blanchard V. 2013. Serum glycome profiling: A biomarker for diagnosis of ovarian cancer. J Proteome Res . 12: 4056– 4063. Google Scholar CrossRef Search ADS PubMed Blackhall FH, Merry CL, Davies EJ, Jayson GC. 2001. Heparan sulfate proteoglycans and cancer. Br J Cancer . 85: 1094– 1098. Google Scholar CrossRef Search ADS PubMed Bourguignon LY, Peyrollier K, Xia W, Gilad E. 2008. Hyaluronan-CD44 interaction activates stem cell marker Nanog, Stat-3-mediated MDR1 gene expression, and ankyrin-regulated multidrug efflux in breast and ovarian tumor cells. J Biol Chem . 283: 17635– 17651. Google Scholar CrossRef Search ADS PubMed Bowtell DD, Bohm S, Ahmed AA, Aspuria PJ, Bast RC Jr., Beral V, Berek JS, Birrer MJ, Blagden S, Bookman MA et al. . 2015. Rethinking ovarian cancer II: Reducing mortality from high-grade serous ovarian cancer. Nat Rev Cancer . 15: 668– 679. Google Scholar CrossRef Search ADS PubMed Braig F, Kriegs M, Voigtlaender M, Habel B, Grob T, Biskup K, Blanchard V, Sack M, Thalhammer A, Ben Batalla I et al. . 2017. Cetuximab resistance in head and neck cancer is mediated by EGFR-K521 polymorphism. Cancer Res . 77: 1188– 1199. Google Scholar CrossRef Search ADS PubMed Brazil JC, Sumagin R, Cummings RD, Louis NA, Parkos CA. 2016. Targeting of neutrophil Lewis X blocks transepithelial migration and increases phagocytosis and degranulation. Am J Pathol . 186: 297– 311. Google Scholar CrossRef Search ADS PubMed Bremer EG, Hakomori S. 1982. GM3 ganglioside induces hamster fibroblast growth inhibition in chemically-defined medium: Ganglioside may regulate growth factor receptor function. Biochem Biophys Res Commun . 106: 711– 718. Google Scholar CrossRef Search ADS PubMed Brisson JR, Carver JP. 1983. Solution conformation of asparagine-linked oligosaccharides: Alpha (1–2)-, alpha (1–3)-, beta (1–2)-, and beta (1–4)-linked units. Biochemistry . 22: 3671– 3680. Google Scholar CrossRef Search ADS PubMed Brockhausen I, Dowler T, Paulsen H. 2009. Site directed processing: Role of amino acid sequences and glycosylation of acceptor glycopeptides in the assembly of extended mucin type O-glycan core 2. Biochim Biophys Acta . 1790: 1244– 1257. Google Scholar CrossRef Search ADS PubMed Brockhausen I, Stanley P. 2015. O-GalNAc Glycans. In: Varki A, Cummings RD, Esko JD, Stanley P, Hart GW, Aebi M, Darvill AG, Kinoshita T, Packer NH et al. , editors. Essentials of Glycobiology . NY: Cold Spring Harbor. p. 113– 123. Buescher JM, Antoniewicz MR, Boros LG, Burgess SC, Brunengraber H, Clish CB, DeBerardinis RJ, Feron O, Frezza C, Ghesquiere B et al. . 2015. A roadmap for interpreting (13)C metabolite labeling patterns from cells. Curr Opin Biotechnol . 34: 189– 201. Google Scholar CrossRef Search ADS PubMed Burdick MM, Chu JT, Godar S, Sackstein R. 2006. HCELL is the major E- and L-selectin ligand expressed on LS174T colon carcinoma cells. J Biol Chem . 281: 13899– 13905. Google Scholar CrossRef Search ADS PubMed Burdick MD, Harris A, Reid CJ, Iwamura T, Hollingsworth MA. 1997. Oligosaccharides expressed on MUC1 produced by pancreatic and colon tumor cell lines. J Biol Chem . 272: 24198– 24202. Google Scholar CrossRef Search ADS PubMed Campbell BJ, Finnie IA, Hounsell EF, Rhodes JM. 1995. Direct demonstration of increased expression of Thomsen-Friedenreich (TF) antigen in colonic adenocarcinoma and ulcerative colitis mucin and its concealment in normal mucin. J Clin Invest . 95: 571– 576. Google Scholar CrossRef Search ADS PubMed Cao Y, Stosiek P, Springer GF, Karsten U. 1996. Thomsen-Friedenreich-related carbohydrate antigens in normal adult human tissues: A systematic and comparative study. Histochem Cell Biol . 106: 197– 207. Google Scholar CrossRef Search ADS PubMed Carlson CB, Mowery P, Owen RM, Dykhuizen EC, Kiessling LL. 2007. Selective tumor cell targeting using low-affinity, multivalent interactions. ACS Chem Biol . 2: 119– 127. Google Scholar CrossRef Search ADS PubMed Caruana I, Weber G, Ballard BC, Wood MS, Savoldo B, Dotti G. 2015. K562-derived whole-cell vaccine enhances antitumor responses of CAR-redirected virus-specific cytotoxic T lymphocytes in vivo. Clin Cancer Res . 21: 2952– 2962. Google Scholar CrossRef Search ADS PubMed Chen Z, Gulzar ZG, St Hill CA, Walcheck B, Brooks JD. 2014. Increased expression of GCNT1 is associated with altered O-glycosylation of PSA, PAP, and MUC1 in human prostate cancers. Prostate . 74: 1059– 1067. Google Scholar CrossRef Search ADS PubMed Cheung P, Dennis JW. 2007. Mgat5 and Pten interact to regulate cell growth and polarity. Glycobiology . 17: 767– 773. Google Scholar CrossRef Search ADS PubMed Choo AB, Tan HL, Ang SN, Fong WJ, Chin A, Lo J, Zheng L, Hentze H, Philp RJ, Oh SK et al. . 2008. Selection against undifferentiated human embryonic stem cells by a cytotoxic antibody recognizing podocalyxin-like protein-1. Stem Cells . 26: 1454– 1463. Google Scholar CrossRef Search ADS PubMed Choo M, Tan HL, Ding V, Castangia R, Belgacem O, Liau B, Hartley-Tassell L, Haslam SM, Dell A, Choo A. 2017. Characterization of H type 1 and type 1 N-acetyllactosamine glycan epitopes on ovarian cancer specifically recognized by the anti-glycan monoclonal antibody mAb-A4. J Biol Chem . 292: 6163– 6176. Google Scholar CrossRef Search ADS PubMed Chou CH, Huang MJ, Chen CH, Shyu MK, Huang J, Hung JS, Huang CS, Huang MC. 2015. Up-regulation of C1GALT1 promotes breast cancer cell growth through MUC1-C signaling pathway. Oncotarget . 6: 6123– 6135. Google Scholar PubMed Chou KJ, Lee PT, Chen CL, Hsu CY, Huang WC, Huang CW, Fang HC. 2017. CD44 fucosylation on mesenchymal stem cell enhances homing and macrophage polarization in ischemic kidney injury. Exp Cell Res . 350: 91– 102. Google Scholar CrossRef Search ADS PubMed Colomb F, Vidal O, Bobowski M, Krzewinski-Recchi MA, Harduin-Lepers A, Mensier E, Jaillard S, Lafitte JJ, Delannoy P, Groux-Degroote S. 2014. TNF induces the expression of the sialyltransferase ST3Gal IV in human bronchial mucosa via MSK1/2 protein kinases and increases FliD/sialyl-Lewis (x)-mediated adhesion of Pseudomonas aeruginosa. Biochem J . 457: 79– 87. Google Scholar CrossRef Search ADS PubMed Conrad C, Götte M, Schlomann U, Roessler M, Pagenstecher A, Anderson P, Preston J, Pruessmeyer J, Ludwig A, Li R et al. . 2018. ADAM8 expression in breast cancer derived brain metastases: Functional implications on MMP-9 expression and transendothelial migration in breast cancer cells. Int J Cancer . 142: 779– 791. Google Scholar CrossRef Search ADS PubMed Coss KP, Vasiljevic S, Pritchard LK, Krumm SA, Glaze M, Madzorera S, Moore PL, Crispin M, Doores KJ. 2016. HIV-1 glycan density drives the persistence of the mannose patch within an infected individual. J Virol . 90: 11132– 11144. Google Scholar CrossRef Search ADS PubMed Craddock JA, Lu A, Bear A, Pule M, Brenner MK, Rooney CM, Foster AE. 2010. Enhanced tumor trafficking of GD2 chimeric antigen receptor T cells by expression of the chemokine receptor CCR2b. J Immunother . 33: 780– 788. Google Scholar CrossRef Search ADS PubMed Croci DO, Cerliani JP, Dalotto-Moreno T, Mendez-Huergo SP, Mascanfroni ID, Dergan-Dylon S, Toscano MA, Caramelo JJ, Garcia-Vallejo JJ, Ouyang J et al. . 2014. Glycosylation-dependent lectin-receptor interactions preserve angiogenesis in anti-VEGF refractory tumors. Cell . 156: 744– 758. Google Scholar CrossRef Search ADS PubMed Dalziel M, Whitehouse C, McFarlane I, Brockhausen I, Gschmeissner S, Schwientek T, Clausen H, Burchell JM, Taylor-Papadimitriou J. 2001. The relative activities of the C2GnT1 and ST3Gal-I glycosyltransferases determine O-glycan structure and expression of a tumor-associated epitope on MUC1. J Biol Chem . 276: 11007– 11015. Google Scholar CrossRef Search ADS PubMed Damodarasamy M, Johnson RS, Bentov I, MacCoss MJ, Vernon RB, Reed MJ. 2014. Hyaluronan enhances wound repair and increases collagen III in aged dermal wounds. Wound Repair Regen . 22: 521– 526. Google Scholar CrossRef Search ADS PubMed Danishefsky SJ, Shue YK, Chang MN, Wong CH. 2015. Development of Globo-H cancer vaccine. Acc Chem Res . 48: 643– 652. Google Scholar CrossRef Search ADS PubMed DAUSSET J, MOULLEC J, BERNARD J. 1959. Acquired hemolytic anemia with polyagglutinability of red blood cells due to a new factor present in normal human serum (Anti-Tn). Blood . 14: 1079– 1093. Google Scholar PubMed de Leoz ML, Young LJ, An HJ, Kronewitter SR, Kim J, Miyamoto S, Borowsky AD, Chew HK, Lebrilla CB. 2011. High-mannose glycans are elevated during breast cancer progression. Mol Cell Proteomics . 10: M110.002717. Google Scholar CrossRef Search ADS PubMed de-Freitas-Junior JC, Carvalho S, Dias AM, Oliveira P, Cabral J, Seruca R, Oliveira C, Morgado-Díaz JA, Reis CA, Pinho SS. 2013. Insulin/IGF-I signaling pathways enhances tumor cell invasion through bisecting GlcNAc N-glycans modulation. An interplay with E-cadherin. PLoS One . 8: e81579. Google Scholar CrossRef Search ADS PubMed Delgado M, Lee KJ, Altobell L, Spanka C, Wentworth P, Janda KD. 2002. A parallel approach to the discovery of carrier delivery vehicles to enhance antigen immunogenicity. J Am Chem Soc . 124: 4946– 4947. Google Scholar CrossRef Search ADS PubMed Dimitroff CJ, Lee JY, Fuhlbrigge RC, Sackstein R. 2000. A distinct glycoform of CD44 is an L-selectin ligand on human hematopoietic cells. Proc Natl Acad Sci USA . 97: 13841– 13846. Google Scholar CrossRef Search ADS PubMed Dimitroff CJ, Lee JY, Rafii S, Fuhlbrigge RC, Sackstein R. 2001. CD44 is a major E-selectin ligand on human hematopoietic progenitor cells. J Cell Biol . 153: 1277– 1286. Google Scholar CrossRef Search ADS PubMed Doberstein K, Bretz NP, Schirmer U, Fiegl H, Blaheta R, Breunig C, Müller-Holzner E, Reimer D, Zeimet AG, Altevogt P. 2014. miR-21-3p is a positive regulator of L1CAM in several human carcinomas. Cancer Lett . 354: 455– 466. Google Scholar CrossRef Search ADS PubMed Doberstein K, Milde-Langosch K, Bretz NP, Schirmer U, Harari A, Witzel I, Ben-Arie A, Hubalek M, Müller-Holzner E, Reinold S et al. . 2014. L1CAM is expressed in triple-negative breast cancers and is inversely correlated with androgen receptor. BMC Cancer . 14: 958. Google Scholar CrossRef Search ADS PubMed Dobrenkov K, Cheung NK. 2014. GD2-targeted immunotherapy and radioimmunotherapy. Semin Oncol . 41: 589– 612. Google Scholar CrossRef Search ADS PubMed Drabik A, Bodzon-Kulakowska A, Suder P, Silberring J, Kulig J, Sierzega M. 2017. Glycosylation changes in serum proteins identify patients with pancreatic cancer. J Proteome Res . 16: 1436– 1444. Google Scholar CrossRef Search ADS PubMed Elias KM, Fendler W, Stawiski K, Fiascone SJ, Vitonis AF, Berkowitz RS, Frendl G, Konstantinopoulos P, Crum CP, Kedzierska M et al. . 2017. Diagnostic potential for a serum miRNA neural network for detection of ovarian cancer. eLife . 6. Elola MT, Capurro MI, Barrio MM, Coombs PJ, Taylor ME, Drickamer K, Mordoh J. 2007. Lewis x antigen mediates adhesion of human breast carcinoma cells to activated endothelium. Possible involvement of the endothelial scavenger receptor C-type lectin. Breast Cancer Res Treat . 101: 161– 174. Google Scholar CrossRef Search ADS PubMed Erfurt C, Muller E, Emmerling S, Klotz C, Hertl M, Schuler G, Schultz ES. 2009. Melanoma-associated chondroitin sulphate proteoglycan as a new target antigen for CD4+ T cells in melanoma patients. Int J Cancer . 124: 2341– 2346. Google Scholar CrossRef Search ADS PubMed Erfurt C, Sun Z, Haendle I, Schuler-Thurner B, Heirman C, Thielemans K, van der Bruggen P, Schuler G, Schultz ES. 2007. Tumor-reactive CD4+ T cell responses to the melanoma-associated chondroitin sulphate proteoglycan in melanoma patients and healthy individuals in the absence of autoimmunity. J Immunol . 178: 7703– 7709. Google Scholar CrossRef Search ADS PubMed Escrevente C, Machado E, Brito C, Reis CA, Stoeck A, Runz S, Marmé A, Altevogt P, Costa J. 2006. Different expression levels of alpha3/4 fucosyltransferases and Lewis determinants in ovarian carcinoma tissues and cell lines. Int J Oncol . 29: 557– 566. Google Scholar PubMed Fan J, Wang S, Yu S, He J, Zheng W, Zhang J. 2012. N-acetylglucosaminyltransferase IVa regulates metastatic potential of mouse hepatocarcinoma cells through glycosylation of CD147. Glycoconj J . 29: 323– 334. Google Scholar CrossRef Search ADS PubMed Feng X, Zhao L, Gao S, Song X, Dong W, Zhao Y, Zhou H, Cheng L, Miao X, Jia L. 2016. Increased fucosylation has a pivotal role in multidrug resistance of breast cancer cells through miR-224-3p targeting FUT4. Gene . 578: 232– 241. Google Scholar CrossRef Search ADS PubMed Fouladi-Nashta AA, Raheem KA, Marei WF, Ghafari F, Hartshorne GM. 2017. Regulation and roles of the hyaluronan system in mammalian reproduction. Reproduction . 153: R43– R58. Google Scholar CrossRef Search ADS PubMed Freire-de-Lima L, Gelfenbeyn K, Ding Y, Mandel U, Clausen H, Handa K, Hakomori SI. 2011. Involvement of O-glycosylation defining oncofetal fibronectin in epithelial-mesenchymal transition process. Proc Natl Acad Sci USA . 108: 17690– 17695. Google Scholar CrossRef Search ADS PubMed Friedenrich V. 1930. Production of a specific receptor quality in red cell corpuscles by bacterial activity. In: The Thomsen Haemagglutination Phenomenon . Copenhagen Denmark: Levin and Munksgaard. Fu J, Wei B, Wen T, Johansson ME, Liu X, Bradford E, Thomsson KA, McGee S, Mansour L, Tong M et al. . 2011. Loss of intestinal core 1-derived O-glycans causes spontaneous colitis in mice. J Clin Invest . 121: 1657– 1666. Google Scholar CrossRef Search ADS PubMed Fujikawa K, Imamura A, Ishida H, Kiso M. 2008. Synthesis of a GM3 ganglioside analogue carrying a phytoceramide moiety by intramolecular glycosylation as a key step. Carbohydr Res . 343: 2729– 2734. Google Scholar CrossRef Search ADS PubMed Galili U, Wigglesworth K, Abdel-Motal UM. 2007. Intratumoral injection of -gal glycolipids induces xenograft-like destruction and conversion of lesions into endogenous vaccines. J Immunol . 178: 4676– 4687. Google Scholar CrossRef Search ADS PubMed Gao N, Liu J, Liu D, Hao Y, Yan L, Ma Y, Zhuang H, Hu Z, Gao J, Yang Z et al. . 2014. c-Jun transcriptionally regulates alpha 1, 2-fucosyltransferase 1 (FUT1) in ovarian cancer. Biochimie . 107( Pt B): 286– 292. Google Scholar CrossRef Search ADS PubMed Gargett T, Yu W, Dotti G, Yvon ES, Christo SN, Hayball JD, Lewis ID, Brenner MK, Brown MP. 2016. GD2-specific CAR T cells undergo potent activation and deletion following antigen encounter but can be protected from activation-induced cell death by PD-1 blockade. Mol Ther . 24: 1135– 1149. Google Scholar CrossRef Search ADS PubMed Geiser M, Schultz D, Le Cardinal A, Voshol H, Garcia-Echeverria C. 1999. Identification of the human melanoma-associated chondroitin sulfate proteoglycan antigen epitope recognized by the antitumor monoclonal antibody 763.74 from a peptide phage library. Cancer Res . 59: 905– 910. Google Scholar PubMed Gerken TA, Jamison O, Perrine CL, Collette JC, Moinova H, Ravi L, Markowitz SD, Shen W, Patel H, Tabak LA. 2011. Emerging paradigms for the initiation of mucin-type protein O-glycosylation by the polypeptide GalNAc transferase family of glycosyltransferases. J Biol Chem . 286: 14493– 14507. Google Scholar CrossRef Search ADS PubMed Ghatak S, Misra S, Toole BP. 2005. Hyaluronan constitutively regulates ErbB2 phosphorylation and signaling complex formation in carcinoma cells. J Biol Chem . 280: 8875– 8883. Google Scholar CrossRef Search ADS PubMed Gill DJ, Chia J, Senewiratne J, Bard F. 2010. Regulation of O-glycosylation through Golgi-to-ER relocation of initiation enzymes. J Cell Biol . 189: 843– 858. Google Scholar CrossRef Search ADS PubMed Gill DJ, Tham KM, Chia J, Wang SC, Steentoft C, Clausen H, Bard-Chapeau EA, Bard FA. 2013. Initiation of GalNAc-type O-glycosylation in the endoplasmic reticulum promotes cancer cell invasiveness. Proc Natl Acad Sci USA . 110: E3152– E3161. Google Scholar CrossRef Search ADS PubMed Goetz DJ, Ding H, Atkinson WJ, Vachino G, Camphausen RT, Cumming DA, Luscinskas FW. 1996. A human colon carcinoma cell line exhibits adhesive interactions with P-selectin under fluid flow via a PSGL-1-independent mechanism. Am J Pathol . 149: 1661– 1673. Google Scholar PubMed Gomes C, Osorio H, Pinto MT, Campos D, Oliveira MJ, Reis CA. 2013. Expression of ST3GAL4 leads to SLe (x) expression and induces c-Met activation and an invasive phenotype in gastric carcinoma cells. PLoS One . 8: e66737. Google Scholar CrossRef Search ADS PubMed Gong L, Cai Y, Zhou X, Yang H. 2012. Activated platelets interact with lung cancer cells through P-selectin glycoprotein ligand-1. Pathol Oncol Res . 18: 989– 996. Google Scholar CrossRef Search ADS PubMed Gooi HC, Feizi T, Kapadia A, Knowles BB, Solter D, Evans MJ. 1981. Stage-specific embryonic antigen involves alpha 1 goes to 3 fucosylated type 2 blood group chains. Nature . 292: 156– 158. Google Scholar CrossRef Search ADS PubMed Gordts PL, Esko JD. 2015. Heparan sulfate proteoglycans fine-tune macrophage inflammation via IFN-beta. Cytokine . 72: 118– 119. Google Scholar CrossRef Search ADS PubMed Grage-Griebenow E, Jerg E, Gorys A, Wicklein D, Wesch D, Freitag-Wolf S, Goebel L, Vogel I, Becker T, Ebsen M et al. . 2014. L1CAM promotes enrichment of immunosuppressive T cells in human pancreatic cancer correlating with malignant progression. Mol Oncol . 8: 982– 997. Google Scholar CrossRef Search ADS PubMed Granovsky M, Fata J, Pawling J, Muller WJ, Khokha R, Dennis JW. 2000. Suppression of tumor growth and metastasis in Mgat5-deficient mice. Nat Med . 6: 306– 312. Google Scholar CrossRef Search ADS PubMed Gremel G, Grannas K, Sutton LA, Pontén F, Zieba A. 2013. In situ protein detection for companion diagnostics. Front Oncol . 3: 271. Google Scholar CrossRef Search ADS PubMed Groux-Degroote S, Guérardel Y, Julien S, Delannoy P. 2015. Gangliosides in breast cancer: New perspectives. Biochemistry (Mosc) . 80: 808– 819. Google Scholar CrossRef Search ADS PubMed Hadjialirezaei S, Picco G, Beatson R, Burchell J, Stokke BT, Sletmoen M. 2017. Interactions between the breast cancer-associated MUC1 mucins and C-type lectin characterized by optical tweezers. PLoS One . 12: e0175323. Google Scholar CrossRef Search ADS PubMed Haemmerle M, Bottsford-Miller J, Pradeep S, Taylor ML, Choi HJ, Hansen JM, Dalton HJ, Stone RL, Cho MS, Nick AM et al. . 2016. FAK regulates platelet extravasation and tumor growth after antiangiogenic therapy withdrawal. J Clin Invest . 126: 1885– 1896. Google Scholar CrossRef Search ADS PubMed Haemmerle M, Taylor ML, Gutschner T, Pradeep S, Cho MS, Sheng J, Lyons YM, Nagaraja AS, Dood RL, Wen Y et al. . 2017. Platelets reduce anoikis and promote metastasis by activating YAP1 signaling. Nat Commun . 8: 310. Google Scholar CrossRef Search ADS PubMed Hage N, Howard T, Phillips C, Brassington C, Overman R, Debreczeni J, Gellert P, Stolnik S, Winkler GS, Falcone FH. 2015. Structural basis of Lewis (b) antigen binding by the Helicobacter pylori adhesin BabA. Sci Adv . 1: e1500315. Google Scholar CrossRef Search ADS PubMed Haglund C, Roberts PJ, Kuusela P, Scheinin TM, Mäkelä O, Jalanko H. 1986. Evaluation of CA 19-9 as a serum tumour marker in pancreatic cancer. Br J Cancer . 53: 197– 202. Google Scholar CrossRef Search ADS PubMed Hakomori S, Andrews HD. 1970. Sphingoglycolipids with Leb activity, and the co-presence of Lea-, Leb-glycolipids in human tumor tissue. Biochim Biophys Acta . 202: 225– 228. Google Scholar CrossRef Search ADS PubMed Hakomori SI, Handa K. 2015. GM3 and cancer. Glycoconj J . 32: 1– 8. Google Scholar CrossRef Search ADS PubMed Hakomori S, Jeanloz RW. 1964. Isolation of a glycolipid containing fucose, galactose, glucose, and glucosamine from human cancerous tissue. J Biol Chem . 239: PC3606– PC3607. Google Scholar PubMed Hakomori SI, Murakami WT. 1968. Glycolipids of hamster fibroblasts and derived malignant-transformed cell lines. Proc Natl Acad Sci USA . 59: 254– 261. Google Scholar CrossRef Search ADS PubMed Hakomori S, Strycharz GD. 1968. Investigations on cellular blood-group substances. I. Isolation and chemical composition of blood-group ABH and Le-b isoantigens of sphingoglycolipid nature. Biochemistry . 7: 1279– 1286. Google Scholar CrossRef Search ADS PubMed Hanahan D, Weinberg RA. 2011. Hallmarks of cancer: The next generation. Cell . 144: 646– 674. Google Scholar CrossRef Search ADS PubMed Handa K, White T, Ito K, Fang H, Wang S, Hakomori S. 1995. P-selectin-dependent adhesion of human cancer-cells—Requirement for coexpression of a psgl-1-like core protein and the glycosylation process for sialosyl-le (x) or sialosyl-le (a). Int J Oncol . 6: 773– 781. Google Scholar PubMed Hanisch FG, Hanski C, Hasegawa A. 1992. Sialyl Lewis (x) antigen as defined by monoclonal antibody AM-3 is a marker of dysplasia in the colonic adenoma-carcinoma sequence. Cancer Res . 52: 3138– 3144. Google Scholar PubMed Hanley WD, Burdick MM, Konstantopoulos K, Sackstein R. 2005. CD44 on LS174T colon carcinoma cells possesses E-selectin ligand activity. Cancer Res . 65: 5812– 5817. Google Scholar CrossRef Search ADS PubMed Hanley WD, Napier SL, Burdick MM, Schnaar RL, Sackstein R, Konstantopoulos K. 2006. Variant isoforms of CD44 are P- and L-selectin ligands on colon carcinoma cells. FASEB J . 20: 337– 339. Google Scholar CrossRef Search ADS PubMed Hanski C, Hanski ML, Zimmer T, Ogorek D, Devine P, Riecken EO. 1995. Characterization of the major sialyl-Lex-positive mucins present in colon, colon carcinoma, and sera of patients with colorectal cancer. Cancer Res . 55: 928– 933. Google Scholar PubMed Hascall V, Esko JD. 2015. Hyaluronan. In: Varki A, Cummings RD, Esko JD, Stanley P, Hart GW, Aebi M, Darvill AG, Kinoshita T, Packer NH, Prestegard JH et al. , editors. Essentials of Glycobiology . NY: Cold Spring Harbor. Hassinen A, Pujol FM, Kokkonen N, Pieters C, Kihlström M, Korhonen K, Kellokumpu S. 2011. Functional organization of Golgi N- and O-glycosylation pathways involves pH-dependent complex formation that is impaired in cancer cells. J Biol Chem . 286: 38329– 38340. Google Scholar CrossRef Search ADS PubMed Heczey A, Louis CU, Savoldo B, Dakhova O, Durett A, Grilley B, Liu H, Wu MF, Mei Z, Gee A et al. . 2017. CAR T cells administered in combination with lymphodepletion and PD-1 inhibition to patients with neuroblastoma. Mol Ther . 25: 2214– 2224. Google Scholar CrossRef Search ADS PubMed Helling F, Shang A, Calves M, Zhang S, Ren S, Yu RK, Oettgen HF, Livingston PO. 1994. GD3 vaccines for melanoma: Superior immunogenicity of keyhole limpet hemocyanin conjugate vaccines. Cancer Res . 54: 197– 203. Google Scholar PubMed Herbertson RA, Tebbutt NC, Lee FT, MacFarlane DJ, Chappell B, Micallef N, Lee ST, Saunder T, Hopkins W, Smyth FE et al. . 2009. Phase I biodistribution and pharmacokinetic study of Lewis Y-targeting immunoconjugate CMD-193 in patients with advanced epithelial cancers. Clin Cancer Res . 15: 6709– 6715. Google Scholar CrossRef Search ADS PubMed Herbomel GG, Rojas RE, Tran DT, Ajinkya M, Beck L, Tabak LA. 2017. The GalNAc-T Activation Pathway (GALA) is not a general mechanism for regulating mucin-type O-glycosylation. PLoS One . 12: e0179241. Google Scholar CrossRef Search ADS PubMed Herlyn M, Shen JW, Sears HF, Civin CI, Verrill HL, Goldberg EM, Koprowski H. 1984. Detection of a circulating gastrointestinal cancer antigen in sera of patients with gastrointestinal malignancies by a double determinant immunoassay with monoclonal antibodies against human blood group determinants. Clin Exp Immunol . 55: 23– 35. Google Scholar PubMed Ho JJ, Siddiki B, Kim YS. 1995. Association of sialyl-Lewis (a) and sialyl-Lewis (x) with MUC-1 apomucin ina pancreatic cancer cell line. Cancer Res . 55: 3659– 3663. Google Scholar PubMed Hofmann BT, Schluter L, Lange P, Mercanoglu B, Ewald F, Folster A, Picksak AS, Harder S, El Gammal AT, Grupp K et al. . 2015. COSMC knockdown mediated aberrant O-glycosylation promotes oncogenic properties in pancreatic cancer. Mol Cancer . 14: 109. Google Scholar CrossRef Search ADS PubMed Hoja-Lukowicz D, Link-Lenczowski P, Carpentieri A, Amoresano A, Pochec E, Artemenko KA, Bergquist J, Litynska A. 2013. L1CAM from human melanoma carries a novel type of N-glycan with Galbeta1-4Galbeta1- motif. Involvement of N-linked glycans in migratory and invasive behaviour of melanoma cells. Glycoconj J . 30: 205– 225. Google Scholar CrossRef Search ADS PubMed Holgersson J, Löfling J. 2006. Glycosyltransferases involved in type 1 chain and Lewis antigen biosynthesis exhibit glycan and core chain specificity. Glycobiology . 16: 584– 593. Google Scholar CrossRef Search ADS PubMed Hoos A, Protsyuk D, Borsig L. 2014. Metastatic growth progression caused by PSGL-1-mediated recruitment of monocytes to metastatic sites. Cancer Res . 74: 695– 704. Google Scholar CrossRef Search ADS PubMed Hoseini SS, Dobrenkov K, Pankov D, Xu XL, Cheung NK. 2017. Bispecific antibody does not induce T-cell death mediated by chimeric antigen receptor against disialoganglioside GD2. Oncoimmunology . 6: e1320625. Google Scholar CrossRef Search ADS PubMed Hossler P, Mulukutla BC, Hu WS. 2007. Systems analysis of N-glycan processing in mammalian cells. PLoS One . 2: e713. Google Scholar CrossRef Search ADS PubMed Hou R, Jiang L, Liu D, Lin B, Hu Z, Gao J, Zhang D, Zhang S, Iwamori M. 2017. Lewis (y) antigen promotes the progression of epithelial ovarian cancer by stimulating MUC1 expression. Int J Mol Med . 40: 293– 302. Google Scholar CrossRef Search ADS PubMed Houghton AN, Mintzer D, Cordon-Cardo C, Welt S, Fliegel B, Vadhan S, Carswell E, Melamed MR, Oettgen HF, Old LJ. 1985. Mouse monoclonal IgG3 antibody detecting GD3 ganglioside: A phase I trial in patients with malignant melanoma. Proc Natl Acad Sci USA . 82: 1242– 1246. Google Scholar CrossRef Search ADS PubMed Hu Q, Hisamatsu T, Haemmerle M, Cho MS, Pradeep S, Rupaimoole R, Rodriguez-Aguayo C, Lopez-Berestein G, Wong STC, Sood AK et al. . 2017. Role of platelet-derived Tgfbeta1 in the progression of ovarian cancer. Clin Cancer Res . 23: 5611– 5621. Google Scholar CrossRef Search ADS PubMed Hua S, Saunders M, Dimapasoc LM, Jeong SH, Kim BJ, Kim S, So M, Lee KS, Kim JH, Lam KS et al. . 2014. Differentiation of cancer cell origin and molecular subtype by plasma membrane N-glycan profiling. J Proteome Res . 13: 961– 968. Google Scholar CrossRef Search ADS PubMed Huang J, Che MI, Lin NY, Hung JS, Huang YT, Lin WC, Huang HC, Lee PH, Liang JT, Huang MC. 2014. The molecular chaperone Cosmc enhances malignant behaviors of colon cancer cells via activation of Akt and ERK. Mol Carcinog . 53( Suppl 1): E62– E71. Google Scholar CrossRef Search ADS PubMed Huang C, Mezencev R, McDonald JF, Vannberg F. 2017. Open source machine-learning algorithms for the prediction of optimal cancer drug therapies. PLoS One . 12: e0186906. Google Scholar CrossRef Search ADS PubMed Hutchins LF, Makhoul I, Emanuel PD, Pennisi A, Siegel ER, Jousheghany F, Guo X, Pashov AD, Monzavi-Karbassi B, Kieber-Emmons T. 2017. Targeting tumor-associated carbohydrate antigens: A phase I study of a carbohydrate mimetic-peptide vaccine in stage IV breast cancer subjects. Oncotarget . 8: 99161– 99178. Google Scholar CrossRef Search ADS PubMed Hynes RO, Naba A. 2012. Overview of the matrisome—An inventory of extracellular matrix constituents and functions. Cold Spring Harb Perspect Biol . 4: a004903. Google Scholar CrossRef Search ADS PubMed Iida J, Meijne AM, Oegema TR Jr., Yednock TA, Kovach NL, Furcht LT, McCarthy JB. 1998. A role of chondroitin sulfate glycosaminoglycan binding site in alpha4beta1 integrin-mediated melanoma cell adhesion. J Biol Chem . 273: 5955– 5962. Google Scholar CrossRef Search ADS PubMed Inagaki H, Sakamoto J, Nakazato H, Bishop AE, Yura J. 1990. Expression of Lewis (a), Lewis (b), and sialated Lewis (a) antigens in early and advanced human gastric cancers. J Surg Oncol . 44: 208– 213. Google Scholar CrossRef Search ADS PubMed Inoue M, Nakada H, Tanaka N, Yamashina I. 1994. Tn antigen is expressed on leukosialin from T-lymphoid cells. Cancer Res . 54: 85– 88. Google Scholar PubMed Irimura T, Denda K, Iida S, Takeuchi H, Kato K. 1999. Diverse glycosylation of MUC1 and MUC2: Potential significance in tumor immunity. J Biochem . 126: 975– 985. Google Scholar CrossRef Search ADS PubMed Isaji T, Gu J, Nishiuchi R, Zhao Y, Takahashi M, Miyoshi E, Honke K, Sekiguchi K, Taniguchi N. 2004. Introduction of bisecting GlcNAc into integrin alpha5beta1 reduces ligand binding and down-regulates cell adhesion and cell migration. J Biol Chem . 279: 19747– 19754. Google Scholar CrossRef Search ADS PubMed Isaji T, Kariya Y, Xu Q, Fukuda T, Taniguchi N, Gu J. 2010. Functional roles of the bisecting GlcNAc in integrin-mediated cell adhesion. Methods Enzymol . 480: 445– 459. Google Scholar CrossRef Search ADS PubMed Isozaki H, Ohyama T, Mabuchi H. 1998. Expression of cell adhesion molecule CD44 and sialyl Lewis A in gastric carcinoma and colorectal carcinoma in association with hepatic metastasis. Int J Oncol . 13: 935– 942. Google Scholar PubMed Ito T, Yamada S, Tanaka C, Ito S, Murai T, Kobayashi D, Fujii T, Nakayama G, Sugimoto H, Koike M et al. . 2014. Overexpression of L1CAM is associated with tumor progression and prognosis via ERK signaling in gastric cancer. Ann Surg Oncol . 21: 560– 568. Google Scholar CrossRef Search ADS PubMed Itzkowitz SH. 1992. Blood group-related carbohydrate antigen expression in malignant and premalignant colonic neoplasms. J Cell Biochem Suppl . 16G: 97– 101. Google Scholar CrossRef Search ADS PubMed Jacob F, Anugraham M, Pochechueva T, Tse BW, Alam S, Guertler R, Bovin NV, Fedier A, Hacker NF, Huflejt ME et al. . 2014. The glycosphingolipid P (1) is an ovarian cancer-associated carbohydrate antigen involved in migration. Br J Cancer . 111: 1634– 1645. Google Scholar CrossRef Search ADS PubMed Jacobs PP, Sackstein R. 2011. CD44 and HCELL: Preventing hematogenous metastasis at step 1. FEBS Lett . 585: 3148– 3158. Google Scholar CrossRef Search ADS PubMed Jennemann R, Gröne HJ. 2013. Cell-specific in vivo functions of glycosphingolipids: Lessons from genetic deletions of enzymes involved in glycosphingolipid synthesis. Prog Lipid Res . 52: 231– 248. Google Scholar CrossRef Search ADS PubMed Jorgensen T, Berner A, Kaalhus O, Tveter KJ, Danielsen HE, Bryne M. 1995. Up-regulation of the oligosaccharide sialyl LewisX: A new prognostic parameter in metastatic prostate cancer. Cancer Res . 55: 1817– 1819. Google Scholar PubMed Ju T, Cummings RD. 2002. A unique molecular chaperone Cosmc required for activity of the mammalian core 1 beta 3-galactosyltransferase. Proc Natl Acad Sci USA . 99: 16613– 16618. Google Scholar CrossRef Search ADS PubMed Ju T, Lanneau GS, Gautam T, Wang Y, Xia B, Stowell SR, Willard MT, Wang W, Xia JY, Zuna RE et al. . 2008. Human tumor antigens Tn and sialyl Tn arise from mutations in Cosmc. Cancer Res . 68: 1636– 1646. Google Scholar CrossRef Search ADS PubMed Ju T, Otto VI, Cummings RD. 2011. The Tn antigen-structural simplicity and biological complexity. Angew Chem Int Ed Engl . 50: 1770– 1791. Google Scholar CrossRef Search ADS PubMed Ju L, Wang Y, Xie Q, Xu X, Li Y, Chen Z, Li Y. 2016. Elevated level of serum glycoprotein bifucosylation and prognostic value in Chinese breast cancer. Glycobiology . 26: 460– 471. Google Scholar CrossRef Search ADS PubMed Julien S, Adriaenssens E, Ottenberg K, Furlan A, Courtand G, Vercoutter-Edouart AS, Hanisch FG, Delannoy P, Le Bourhis X. 2006. ST6GalNAc I expression in MDA-MB-231 breast cancer cells greatly modifies their O-glycosylation pattern and enhances their tumourigenicity. Glycobiology . 16: 54– 64. Google Scholar CrossRef Search ADS PubMed Kaczmarek R. 2010. [Alterations of Lewis histo-blood group antigen expression in cancer cells]. Postepy Hig Med Dosw (Online) . 64: 87– 99. Google Scholar PubMed Kadota A, Masutani M, Takei M, Horie T. 1999. Evaluation of expression of CD15 and sCD15 in non-small cell lung cancer. Int J Oncol . 15: 1081– 1089. Google Scholar PubMed Kanabar V, Tedaldi L, Jiang J, Nie X, Panina I, Descroix K, Man F, Pitchford SC, Page CP, Wagner GK. 2016. Base-modified UDP-sugars reduce cell surface levels of P-selectin glycoprotein 1 (PSGL-1) on IL-1beta-stimulated human monocytes. Glycobiology . 26: 1059– 1071. Google Scholar CrossRef Search ADS PubMed Kappelmayer J, Kiss A, Karászi E, Veszprémi A, Jakó J, Kiss C. 2001. Identification of P-selectin glycoprotein ligand-1 as a useful marker in acute myeloid leukaemias. Br J Haematol . 115: 903– 909. Google Scholar CrossRef Search ADS PubMed Kaprio T, Satomaa T, Heiskanen A, Hokke CH, Deelder AM, Mustonen H, Hagström J, Carpen O, Saarinen J, Haglund C. 2015. N-glycomic profiling as a tool to separate rectal adenomas from carcinomas. Mol Cell Proteomics . 14: 277– 288. Google Scholar CrossRef Search ADS PubMed Kaszubska W, Hooft van Huijsduijnen R, Ghersa P, DeRaemy-Schenk AM, Chen BP, Hai T, DeLamarter JF, Whelan J. 1993. Cyclic AMP-independent ATF family members interact with NF-kappa B and function in the activation of the E-selectin promoter in response to cytokines. Mol Cell Biol . 13: 7180– 7190. Google Scholar CrossRef Search ADS PubMed Kawasaki N, Lin CW, Inoue R, Khoo KH, Kawasaki N, Ma BY, Oka S, Ishiguro M, Sawada T, Ishida H et al. . 2009. Highly fucosylated N-glycan ligands for mannan-binding protein expressed specifically on CD26 (DPPVI) isolated from a human colorectal carcinoma cell line, SW1116. Glycobiology . 19: 437– 450. Google Scholar CrossRef Search ADS PubMed Kawashima N, Qu H, Lobaton M, Zhu Z, Sollogoub M, Cavenee WK, Handa K, Hakomori SI, Zhang Y. 2014. Efficient synthesis of chloro-derivatives of sialosyllactosylceramide, and their enhanced inhibitory effect on epidermal growth factor receptor activation. Oncol Lett . 7: 933– 940. Google Scholar CrossRef Search ADS PubMed Kellokumpu S, Sormunen R, Kellokumpu I. 2002. Abnormal glycosylation and altered Golgi structure in colorectal cancer: Dependence on intra-Golgi pH. FEBS Lett . 516: 217– 224. Google Scholar CrossRef Search ADS PubMed Kiefel H, Bondong S, Hazin J, Ridinger J, Schirmer U, Riedle S, Altevogt P. 2012. L1CAM: A major driver for tumor cell invasion and motility. Cell Adh Migr . 6: 374– 384. Google Scholar CrossRef Search ADS PubMed Kim YS, Itzkowitz SH, Yuan M, Chung Y, Satake K, Umeyama K, Hakomori S. 1988. Lex and Ley antigen expression in human pancreatic cancer. Cancer Res . 48: 475– 482. Google Scholar PubMed Kim J, Villadsen R, Sorlie T, Fogh L, Gronlund SZ, Fridriksdottir AJ, Kuhn I, Rank F, Wielenga VT, Solvang H et al. . 2012. Tumor initiating but differentiated luminal-like breast cancer cells are highly invasive in the absence of basal-like activity. Proc Natl Acad Sci USA . 109: 6124– 6129. Google Scholar CrossRef Search ADS PubMed Kinoshita M, Mitsui Y, Kakoi N, Yamada K, Hayakawa T, Kakehi K. 2014. Common glycoproteins expressing polylactosamine-type glycans on matched patient primary and metastatic melanoma cells show different glycan profiles. J Proteome Res . 13: 1021– 1033. Google Scholar CrossRef Search ADS PubMed Koh YW, Lee HJ, Ahn JH, Lee JW, Gong G. 2013. Expression of Lewis X is associated with poor prognosis in triple-negative breast cancer. Am J Clin Pathol . 139: 746– 753. Google Scholar CrossRef Search ADS PubMed Kohler RS, Anugraham M, López MN, Xiao C, Schoetzau A, Hettich T, Schlotterbeck G, Fedier A, Jacob F, Heinzelmann-Schwarz V. 2016. Epigenetic activation of MGAT3 and corresponding bisecting GlcNAc shortens the survival of cancer patients. Oncotarget . 7: 51674– 51686. Google Scholar CrossRef Search ADS PubMed Koike T, Kimura N, Miyazaki K, Yabuta T, Kumamoto K, Takenoshita S, Chen J, Kobayashi M, Hosokawa M, Taniguchi A et al. . 2004. Hypoxia induces adhesion molecules on cancer cells: A missing link between Warburg effect and induction of selectin-ligand carbohydrates. Proc Natl Acad Sci USA . 101: 8132– 8137. Google Scholar CrossRef Search ADS PubMed Komatsu M, Tatum L, Altman NH, Carothers Carraway CA, Carraway KL. 2000. Potentiation of metastasis by cell surface sialomucin complex (rat MUC4), a multifunctional anti-adhesive glycoprotein. Int J Cancer . 87: 480– 486. Google Scholar CrossRef Search ADS PubMed Konety BR, Ballou B, Jaffe R, Singh J, Reiland J, Hakala TR. 1997. Expression of SSEA-1 (Lewis (x)) on transitional cell carcinoma of the bladder. Urol Int . 58: 69– 74. Google Scholar CrossRef Search ADS PubMed Koprowski H, Herlyn M, Steplewski Z, Sears HF. 1981. Specific antigen in serum of patients with colon carcinoma. Science . 212: 53– 55. Google Scholar CrossRef Search ADS PubMed Koprowski H, Steplewski Z, Mitchell K, Herlyn M, Herlyn D, Fuhrer P. 1979. Colorectal carcinoma antigens detected by hybridoma antibodies. Somatic Cell Genet . 5: 957– 971. Google Scholar CrossRef Search ADS PubMed Krambeck FJ, Bennun SV, Andersen MR, Betenbaugh MJ. 2017. Model-based analysis of N-glycosylation in Chinese hamster ovary cells. PLoS One . 12: e0175376. Google Scholar CrossRef Search ADS PubMed Krause DS, Lazarides K, Lewis JB, von Andrian UH, Van Etten RA. 2014. Selectins and their ligands are required for homing and engraftment of BCR-ABL1+ leukemic stem cells in the bone marrow niche. Blood . 123: 1361– 1371. Google Scholar CrossRef Search ADS PubMed Krengel U, Olsson LL, Martinez C, Talavera A, Rojas G, Mier E, Angstrom J, Moreno E. 2004. Structure and molecular interactions of a unique antitumor antibody specific for N-glycolyl GM3. J Biol Chem . 279: 5597– 5603. Google Scholar CrossRef Search ADS PubMed Krokfors E, Kinnunen O. 1954. Blood groups and gynaecological cancer. Br Med J . 1: 1305– 1306. Google Scholar CrossRef Search ADS PubMed Krüger K, Büning C, Schriever F. 2001. Activated T lymphocytes bind in situ to stromal tissue of colon carcinoma but lack adhesion to tumor cells. Eur J Immunol . 31: 138– 145. Google Scholar CrossRef Search ADS PubMed Kufe DW. 2009. Mucins in cancer: Function, prognosis and therapy. Nat Rev Cancer . 9: 874– 885. Google Scholar CrossRef Search ADS PubMed Kłopocki AG, Krop-Watorek A, Duś D, Ugorski M. 1996. Adhesion of human uroepithelial cells to E-selectin: Possible involvement of sialosyl LewisA-ganglioside. Int J Cancer . 68: 239– 244. Google Scholar CrossRef Search ADS PubMed Kłopocki AG, Laskowska A, Antoniewicz-Papis J, Duk M, Lisowska E, Ugorski M. 1998. Role of sialosyl Lewis (a) in adhesion of colon cancer cells—The antisense RNA approach. Eur J Biochem . 253: 309– 318. Google Scholar CrossRef Search ADS PubMed Labelle M, Begum S, Hynes RO. 2011. Direct signaling between platelets and cancer cells induces an epithelial-mesenchymal-like transition and promotes metastasis. Cancer Cell . 20: 576– 590. Google Scholar CrossRef Search ADS PubMed Labelle M, Begum S, Hynes RO. 2014. Platelets guide the formation of early metastatic niches. Proc Natl Acad Sci USA . 111: E3053– E3061. Google Scholar CrossRef Search ADS PubMed Laubli H, Borsig L. 2010. Selectins promote tumor metastasis. Semin Cancer Biol . 20: 169– 177. Google Scholar CrossRef Search ADS PubMed Leathem AJ, Brooks SA. 1987. Predictive value of lectin binding on breast-cancer recurrence and survival. Lancet . 1: 1054– 1056. Google Scholar CrossRef Search ADS PubMed Leathem A, Dokal I, Atkins N. 1983. Lectin binding to normal and malignant breast tissue. Diagn Histopathol . 6: 171– 180. Google Scholar PubMed Lee HY, Chen CY, Tsai TI, Li ST, Lin KH, Cheng YY, Ren CT, Cheng TJ, Wu CY, Wong CH. 2014. Immunogenicity study of Globo H analogues with modification at the reducing or nonreducing end of the tumor antigen. J Am Chem Soc . 136: 16844– 16853. Google Scholar CrossRef Search ADS PubMed Lee LY, Thaysen-Andersen M, Baker MS, Packer NH, Hancock WS, Fanayan S. 2014. Comprehensive N-glycome profiling of cultured human epithelial breast cells identifies unique secretome N-glycosylation signatures enabling tumorigenic subtype classification. J Proteome Res . 13: 4783– 4795. Google Scholar CrossRef Search ADS PubMed Li CW, Lim SO, Chung EM, Kim YS, Park AH, Yao J, Cha JH, Xia W, Chan LC, Kim T et al. . 2018. Eradication of triple-negative breast cancer cells by targeting glycosylated PD-L1. Cancer Cell . 33: 187– 201.e110. Google Scholar CrossRef Search ADS PubMed Li F, Lin B, Hao Y, Li Y, Liu J, Cong J, Zhu L, Liu Q, Zhang S. 2010. Lewis Y promotes growth and adhesion of ovarian carcinoma-derived RMG-I cells by upregulating growth factors. Int J Mol Sci . 11: 3748– 3759. Google Scholar CrossRef Search ADS PubMed Li F, Ten Dam GB, Murugan S, Yamada S, Hashiguchi T, Mizumoto S, Oguri K, Okayama M, van Kuppevelt TH, Sugahara K. 2008. Involvement of highly sulfated chondroitin sulfate in the metastasis of the Lewis lung carcinoma cells. J Biol Chem . 283: 34294– 34304. Google Scholar CrossRef Search ADS PubMed Li J, Zhou Z, Zhang X, Zheng L, He D, Ye Y, Zhang QQ, Qi CL, He XD, Yu C et al. . 2017. Inflammatory molecule, PSGL-1, deficiency activates macrophages to promote colorectal cancer growth through NFκB signaling. Mol Cancer Res . 15: 467– 477. Google Scholar CrossRef Search ADS PubMed Lindahl U, Couchman J, Kimata K, Esko JD. 2015. Proteoglycans and sulfated glycosaminoglycans. In: Varki A, Cummings RD, Esko JD, Stanley P, Hart GW, Aebi M, Darvill AG, Kinoshita T, Packer NH, Prestegard JH et al. , editors. Essentials of Glycobiology . NY: Cold Spring Harbor. Liu D, Liu J, Wang C, Lin B, Liu Q, Hao Y, Zhang S, Iwamori M. 2011. The stimulation of IGF-1R expression by Lewis (y) antigen provides a powerful development mechanism of epithelial ovarian carcinoma. Int J Mol Sci . 12: 6781– 6795. Google Scholar CrossRef Search ADS PubMed Liu G, Neelamegham S. 2014. A computational framework for the automated construction of glycosylation reaction networks. PLoS One . 9: e100939. Google Scholar CrossRef Search ADS PubMed Liu X, Nie H, Zhang Y, Yao Y, Maitikabili A, Qu Y, Shi S, Chen C, Li Y. 2013. Cell surface-specific N-glycan profiling in breast cancer. PLoS One . 8: e72704. Google Scholar CrossRef Search ADS PubMed Lloyd KO, Burchell J, Kudryashov V, Yin BW, Taylor-Papadimitriou J. 1996. Comparison of O-linked carbohydrate chains in MUC-1 mucin from normal breast epithelial cell lines and breast carcinoma cell lines. Demonstration of simpler and fewer glycan chains in tumor cells. J Biol Chem . 271: 33325– 33334. Google Scholar CrossRef Search ADS PubMed Locker GY, Hamilton S, Harris J, Jessup JM, Kemeny N, Macdonald JS, Somerfield MR, Hayes DF, Bast RC, ASCO. 2006. ASCO2006 update of recommendations for the use of tumor markers in gastrointestinal cancer. J Clin Oncol . 24: 5313– 5327. Google Scholar CrossRef Search ADS PubMed Long AH, Highfill SL, Cui Y, Smith JP, Walker AJ, Ramakrishna S, El-Etriby R, Galli S, Tsokos MG, Orentas RJ et al. . 2016. Reduction of MDSCs with all-trans retinoic acid improves CAR therapy efficacy for sarcomas. Cancer Immunol Res . 4: 869– 880. Google Scholar CrossRef Search ADS PubMed Louis CU, Savoldo B, Dotti G, Pule M, Yvon E, Myers GD, Rossig C, Russell HV, Diouf O, Liu E et al. . 2011. Antitumor activity and long-term fate of chimeric antigen receptor-positive T cells in patients with neuroblastoma. Blood . 118: 6050– 6056. Google Scholar CrossRef Search ADS PubMed Ma YQ, Geng JG. 2002. Obligatory requirement of sulfation for P-selectin binding to human salivary gland carcinoma Acc-M cells and breast carcinoma ZR-75-30 cells. J Immunol . 168: 1690– 1696. Google Scholar CrossRef Search ADS PubMed Madsen CB, Lavrsen K, Steentoft C, Vester-Christensen MB, Clausen H, Wandall HH, Pedersen AE. 2013. Glycan elongation beyond the mucin associated Tn antigen protects tumor cells from immune-mediated killing. PLoS One . 8: e72413. Google Scholar CrossRef Search ADS PubMed Magnani JL, Brockhaus M, Smith DF, Ginsburg V, Blaszczyk M, Mitchell KF, Steplewski Z, Koprowski H. 1981. A monosialoganglioside is a monoclonal antibody-defined antigen of colon carcinoma. Science . 212: 55– 56. Google Scholar CrossRef Search ADS PubMed Mahdavi J, Sondén B, Hurtig M, Olfat FO, Forsberg L, Roche N, Angstrom J, Larsson T, Teneberg S, Karlsson KA et al. . 2002. Helicobacter pylori SabA adhesin in persistent infection and chronic inflammation. Science . 297: 573– 578. Google Scholar CrossRef Search ADS PubMed Marcos NT, Pinho S, Grandela C, Cruz A, Samyn-Petit B, Harduin-Lepers A, Almeida R, Silva F, Morais V, Costa J et al. . 2004. Role of the human ST6GalNAc-I and ST6GalNAc-II in the synthesis of the cancer-associated sialyl-Tn antigen. Cancer Res . 64: 7050– 7057. Google Scholar CrossRef Search ADS PubMed Martins LC, de Oliveira Corvelo TC, Oti HT, do Socorro Pompeu Loiola R, Aguiar DC, dos Santos Barile KA, do Amaral RK, Barbosa HP, Fecury AA, de Souza JT. 2006. ABH and Lewis antigen distributions in blood, saliva and gastric mucosa and H. pylori infection in gastric ulcer patients. World J Gastroenterol . 12: 1120– 1124. Google Scholar CrossRef Search ADS PubMed Mathieu S, Prorok M, Benoliel AM, Uch R, Langlet C, Bongrand P, Gerolami R, El-Battari A. 2004. Transgene expression of alpha (1,2)-fucosyltransferase-I (FUT1) in tumor cells selectively inhibits sialyl-Lewis x expression and binding to E-selectin without affecting synthesis of sialyl-Lewis a or binding to P-selectin. Am J Pathol . 164: 371– 383. Google Scholar CrossRef Search ADS PubMed Matthay KK, George RE, Yu AL. 2012. Promising therapeutic targets in neuroblastoma. Clin Cancer Res . 18: 2740– 2753. Google Scholar CrossRef Search ADS PubMed McConnell RB, Clarke CA, Downton F. 1954. Blood groups in carcinoma of the lung. Br Med J . 2: 323– 325. Google Scholar CrossRef Search ADS PubMed McDonald AG, Tipton KF, Davey GP. 2016. A knowledge-based system for display and prediction of O-glycosylation network behaviour in response to enzyme knockouts. PLoS Comput Biol . 12: e1004844. Google Scholar CrossRef Search ADS PubMed Mellis SJ, Baenziger JU. 1983. Structures of the oligosaccharides present at the three asparagine-linked glycosylation sites of human IgD. J Biol Chem . 258: 11546– 11556. Google Scholar PubMed Mellman I, Coukos G, Dranoff G. 2011. Cancer immunotherapy comes of age. Nature . 480: 480– 489. Google Scholar CrossRef Search ADS PubMed Mellquist JL, Kasturi L, Spitalnik SL, Shakin-Eshleman SH. 1998. The amino acid following an asn-X-Ser/Thr sequon is an important determinant of N-linked core glycosylation efficiency. Biochemistry . 37: 6833– 6837. Google Scholar CrossRef Search ADS PubMed Mendelsohn R, Cheung P, Berger L, Partridge E, Lau K, Datti A, Pawling J, Dennis JW. 2007. Complex N-glycan and metabolic control in tumor cells. Cancer Res . 67: 9771– 9780. Google Scholar CrossRef Search ADS PubMed Menni C, Keser T, Mangino M, Bell JT, Erte I, Akmacic I, Vuckovic F, Pucic Bakovic M, Gornik O, McCarthy MI et al. . 2013. Glycosylation of immunoglobulin g: role of genetic and epigenetic influences. PLoS One . 8: e82558. Google Scholar CrossRef Search ADS PubMed Mi R, Song L, Wang Y, Ding X, Zeng J, Lehoux S, Aryal RP, Wang J, Crew VK, van Die I et al. . 2012. Epigenetic silencing of the chaperone Cosmc in human leukocytes expressing tn antigen. J Biol Chem . 287: 41523– 41533. Google Scholar CrossRef Search ADS PubMed Miles D, Roché H, Martin M, Perren TJ, Cameron DA, Glaspy J, Dodwell D, Parker J, Mayordomo J, Tres A et al. . 2011. Phase III multicenter clinical trial of the sialyl-TN (STn)-keyhole limpet hemocyanin (KLH) vaccine for metastatic breast cancer. Oncologist . 16: 1092– 1100. Google Scholar CrossRef Search ADS PubMed Misra S, Obeid LM, Hannun YA, Minamisawa S, Berger FG, Markwald RR, Toole BP, Ghatak S. 2008. Hyaluronan constitutively regulates activation of COX-2-mediated cell survival activity in intestinal epithelial and colon carcinoma cells. J Biol Chem . 283: 14335– 14344. Google Scholar CrossRef Search ADS PubMed Miwa HE, Koba WR, Fine EJ, Giricz O, Kenny PA, Stanley P. 2013. Bisected, complex N-glycans and galectins in mouse mammary tumor progression and human breast cancer. Glycobiology . 23: 1477– 1490. Google Scholar CrossRef Search ADS PubMed Miwa HE, Song Y, Alvarez R, Cummings RD, Stanley P. 2012. The bisecting GlcNAc in cell growth control and tumor progression. Glycoconj J . 29: 609– 618. Google Scholar CrossRef Search ADS PubMed Miyake M, Taki T, Hitomi S, Hakomori S. 1992. Correlation of expression of H/Le (y)/Le (b) antigens with survival in patients with carcinoma of the lung. N Engl J Med . 327: 14– 18. Google Scholar CrossRef Search ADS PubMed Miyoshi E, Nishikawa A, Ihara Y, Gu J, Sugiyama T, Hayashi N, Fusamoto H, Kamada T, Taniguchi N. 1993. N-acetylglucosaminyltransferase III and V messenger RNA levels in LEC rats during hepatocarcinogenesis. Cancer Res . 53: 3899– 3902. Google Scholar PubMed Mlecnik B, Bindea G, Angell HK, Maby P, Angelova M, Tougeron D, Church SE, Lafontaine L, Fischer M, Fredriksen T et al. . 2016. Integrative analyses of colorectal cancer show immunoscore is a stronger predictor of patient survival than microsatellite instability. Immunity . 44: 698– 711. Google Scholar CrossRef Search ADS PubMed Mong TK, Lee HK, Duron SG, Wong CH. 2003. Reactivity-based one-pot total synthesis of fucose GM1 oligosaccharide: A sialylated antigenic epitope of small-cell lung cancer. Proc Natl Acad Sci USA . 100: 797– 802. Google Scholar CrossRef Search ADS PubMed Moran AP. 2008. Relevance of fucosylation and Lewis antigen expression in the bacterial gastroduodenal pathogen Helicobacter pylori. Carbohydr Res . 343: 1952– 1965.
Glycobiology – Oxford University Press
Published: Mar 13, 2018
It’s your single place to instantly
discover and read the research
that matters to you.
Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.
All for just $49/month
Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly
Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.
All the latest content is available, no embargo periods.
“Whoa! It’s like Spotify but for academic articles.”@Phil_Robichaud