Journal of Hepatology
Volume 50, Issue 3 , Pages 592-603, March 2009

Alteration of protein glycosylation in liver diseases☆

  • Bram Blomme

      Affiliations

    • Department of Hepatology and Gastroenterology, Ghent University Hospital, B-9000 Ghent, Belgium
    • ,
  • Christophe Van Steenkiste

      Affiliations

    • Department of Hepatology and Gastroenterology, Ghent University Hospital, B-9000 Ghent, Belgium
    • ,
  • Nico Callewaert

      Affiliations

    • Unit for Molecular Glycobiology, Department for Molecular Biomedical Research, VIB, Ghent University, Ghent, Belgium
    • Department of Biochemistry, Physiology and Microbiology, Ghent University, Ghent, Belgium
    • ,
  • Hans Van Vlierberghe

      Affiliations

    • Department of Hepatology and Gastroenterology, Ghent University Hospital, B-9000 Ghent, Belgium
    • Corresponding Author InformationCorresponding author.

published online 29 December 2008.

Associate Editor: M.P. Manns

Article Outline

Chronic liver diseases are a serious health problem worldwide. The current gold standard to assess structural liver damage is through a liver biopsy which has several disadvantages. A non-invasive, simple and non-expensive test to diagnose liver pathology would be highly desirable. Protein glycosylation has drawn the attention of many researchers in the search for an objective feature to achieve this goal. Glycosylation is a posttranslational modification of many secreted proteins and it has been known for decades that structural changes in the glycan structures of serum proteins are an indication for liver damage. The aim of this paper is to give an overview of this altered protein glycosylation in different etiologies of liver fibrosis / cirrhosis and hepatocellular carcinoma. Although individual liver diseases have their own specific markers, the same modifications seem to continuously reappear in all liver diseases: hyperfucosylation, increased branching and a bisecting N-acetylglucosamine. Analysis at mRNA and protein level of the corresponding glycosyltransferases confirm their altered status in liver pathology. The last part of this review deals with some recently developed glycomic techniques that could potentially be used in the diagnosis of liver pathology.

Abbreviations: ER, endoplasmic reticulum, Asn, asparagine, Ser, serine, Thr, threonine, GlcNAc, N-acetylglucosamine, HCC, hepatocellular carcinoma, ALD, alcoholic liver diseases, CDT, carbohydrate-deficient transferrin, ST6GalI, β-galactoside α2,6 sialyltransferase, TSA, total sialic acid, FSA, free sialic acid, Hp, haptoglobin, Con A, Concanavalin A, Apo E, Apolipoprotein E, GnT-III, N-acetylglucosaminyltransferase III, LAL, lysosomal acid lipase, HPLC, high performance liquid chromatography, MALDI-TOF, matrix assisted laser desorption/ionization time-of-flight, AGP, α1-acid glycoprotein, GnT-V, N-acetylglucosaminyltransferase V, HBx, HBV x-protein, α1-6 FT, α1-6 fucosyltransferase, LCA, lens culinaris agglutinin, E-PHA, Phaseolus vulgaris: erythroagglutinating, L-PHA, Phaseolus vulgaris: leuco-agglutinating, AAT, α-1-antitrypsin, TF, transferrin, DEN, diethylnitrosamine, DSA-FACE, DNA sequencer-assisted fluorophore-assisted carbohydrate electrophoresis, IgG, immunoglobulin G

Keywords: Glycosylation, Liver fibrosis, Hepatocellular carcinoma, Glycomics, Bio-marker

 

Back to Article Outline

1. Introduction 

Over the years it has become apparent that changes in protein glycosylation play an important role in the pathogenesis and progression of various liver diseases. In order to comprehend the relationship between glycosylation and liver diseases, some basic insight into this complex phenomenon is necessary. Therefore, a short introduction is provided, which covers basic biochemical aspects of glycosylation (see Fig. 1).

  • View full-size image.
  • Fig. 1. 

    N-Glycan with indication of the individual monosaccharides and the different binding types between the monosaccharides. a1–3: alpha 1–3 binding; a1–6: alpha 1–6 binding; a2–6: alpha 2–6 binding; b1–2: beta 1–2 binding; b1–4: beta 1–4 binding; b1–6: beta 1–6 binding.

In general, glycosylation consists of co- and posttranslational modification steps, in which individual glycans are added to proteins translated into the endoplasmic reticulum (ER), forming oligosaccharide chains. This is an enzyme-directed and a site-specific process. Two types of protein glycosylation exist: N-glycosylation to the amide nitrogen of asparagine (Asn) side chains and O-glycosylation to the hydroxyl groups of serine (Ser) and threonine (Thr) side chains. Most proteins in human serum contain one or more N-linked glycans, with the exception of albumin and C-reactive protein. O-glycans are found in mucins, which are abundantly present on mucosal surfaces and saliva. Most modifications of glycosylation in liver diseases that have been studied affect N-glycosylated proteins and these will primarily be discussed. The N-oligosaccharide chain is attached to asparagine occurring in the tripeptide sequence Asn-X-Ser, in which X could be any amino acid except proline [1], [2], [3], [4].

The biosynthesis of N- and O-glycans take place in the ER and the Golgi apparatus and it can be roughly divided in three steps [5]. The first step in N-glycosylation is carried out in the ER and consists of the formation of an oligosaccharide-lipid complex containing three glucoses, nine mannoses and two N-acetylglucosamines (GlcNAc). The lipid portion (dolichol) acts as a carrier molecule. The second step is the transfer of the oligosaccharide portion to a growing nascent polypeptide and the simultaneous removal of the three glucose residues and 1 mannose residue. The premature glycoprotein is then mediated to the Golgi apparatus where residual monosaccharides are removed until a (mannose)5(GlcNAc)2 heptasaccharide chain is formed. From here on, specific glycosyltransferases (Table 1 and [9], [10], [11]) and glycosidases will further modify the core-structure by adding or removing monosaccharides, respectively [6]. Glycosyltransferases make use of nucleotides sugars (donors) in order to incorporate monosaccharides into the N-glycan. Glycosidases on the other hand, catalyze the hydrolysis of the glycosidic linkage. In addition, glycans can have various branches (2–5) and are termed bi-, tri-, tetra- and penta-antennary, respectively. The enzymatic addition and removal of monosaccharides allow the formation of glycans with various length, composition and structure [7], [8].

Table 1. Glycosyltransferases that are important in the modification of N-glycans on serum proteins.
Glycosyltransferase familyMammalian glycosyltransferasesSubstrate specificity
α1,2-Fucosyltransferase⁎α1,2 Linkage to the terminal Gal residue in N- or O-glycans
Fucosyltransferases [9]α1,3/4-Fucosyltransferase⁎α1,3 or α1,4 Linkage to GlcNAc in GlcNAc-Gal structures
α1,6-Fucosyltransferaseα1,6 Linkage to the innermost (core) GlcNAc in N-glycans
N-Acetylglucosaminyltransferase IIIGnT-III catalyzes the addition of GlcNAc via β1-4 linkage to the β-mannose core of N-glycans
N-Acetylglucosaminyltransferases [10]N-Acetylglucosaminyltransferase IV⁎GnT-IV catalyzes the formation of GlcNac-β1-4 branches at the Man α1-3 side of the trimannosyl core of N-glycans
N-Acetylglucosaminyltransferase VGnT-V catalyzes the formation of GlcNAc-β1-6 branches at the Man α1-6 side of the trimannosyl core of N-glycans
Sialyltransferases [11]α2,6-Sialyltransferase⁎ST6GalI mediates the transfer of sialic acid residue with an α2,6-linkage to a terminal Gal residue
α2,3-Sialyltransferase⁎ST3GalI mediates the transfer of sialic acid to a Gal residue of a terminal Galβ1-3GalNAc oligosaccharide

⁎Different glycosyltransferases of this class are known.

The functional role(s) of the N-linked carbohydrate moieties of glycoproteins is/are often not well understood. However, glycosylation is a necessity in the correct folding of certain proteins. Aberrant protein folding affects various physiochemical and functional properties of proteins: protein stability, protein solubility, protein inter-/intracellular transport and half-life in blood. The opposite can also be true. Carbohydrate moieties on glycoproteins also fulfill a role in intercellular contact and communication, which is an important aspect of host immunity as well as cancer [12], [13], [14], [15].

The liver contains various receptors on sinusoidal and hepatocyte surfaces. A lot of proteins that bind to these receptors rely on their carbohydrate moieties. Besides changes in glycosylation patterns, the changes in receptor concentration and distribution also occur in various chronic liver diseases (cirrhosis, hepatocellular carcinoma (HCC) and alcoholic liver diseases (ALD)). This leads to an accumulation of certain glycoproteins in the circulation [16], [17].

In the following sections we will discuss the role of glycosylation in liver fibrosis and its relation to various liver pathologies (ALD, hepatitis B, bile-related diseases and obesity) and its role in HCC. The last section will deal with analytical advances in glycoresearch in recent years which now allows the rapid and detailed mapping of the complex mixtures present within biofluid samples.

Back to Article Outline

2. Alteration of glycosylation in fibrosis – cirrhosis 

The gold standard to assess liver fibrosis is through a liver biopsy, which involves the removal of a small liver sample. It is well known that this procedure is accompanied by several complications. Changes in glycosylation of serum proteins have been extensively used as a non-invasive alternative and this has resulted in the development of sensitive and discriminating clinical tests for diagnostic purposes. The rationale for these tests is that the majority of glycosylated serum proteins are synthesized by the liver and in all major liver diseases, changes in this glycosylation occur.

While early glycome studies were confined to the study of sialylation patterns, glycomics has evolved ever since [18]. In the past two decades, the main way of investigating glycosylation was by using lectins [19], [20]. These lectins bind with a particular glycan structure (core-fucosylated glycans, complex glycans,…) (Table 2). Today, with the advent of high-throughput glycomic techniques, we are progressing towards a system biology approach comprising genomics and proteomics in order to draw general conclusions about a particular pathology.

Table 2. Commonly used lectins for the study of altered glycan structures in chronic liver diseases [20].
LectinCommon nameSpecificity Application
Canavalia ensiformis (Concanavalin A, Con A)Jack BeanMan/Glc (Man>Glc>GlcNAc) Con A has been extensively used in the isolation and structural studies of glycoconjugates and it has some clinical uses such as crossed affinity immunoelectrophoresis
Lens culinaris (LCA)LentilMan/Glc (Man>Glc>GlcNAc) Lens lectin is used for the isolation and analysis of glycoproteins. It is also a useful tool for the determination of the degree of fucosylation of alpha-fetoprotein and the histochemical staining of glycoconjgates
Lotus tetragonolobus (LTA)Asparagus peaFuc (α-L-Fuc) The Lotus lectins have specifically been used for the recognition of fucosylated glycans
Phaseolus vulgaris: erythroagglutinating (E-type, E-PHA)Kidney beanComplex (Galβ(1,4)GlcNAcβ(1,2)Man) Especially used for the identification of glycans with a bisecting modification
Phaseolus vulgaris: leuco-agglutinating (L-type, L-PHA)Kidney beanComplex (Galβ(1,4)GlcNacβ(1,2) [Galβ(1,4)GlcNAcβ(1,6)]Man) As L-PHA is reactive with β(1,6) branched structures of trimannosyl core asparagine-linked glycans which are highly selective markers of the metastatic potential of tumor cells, this lectin is used in cancer diagnosis

2.1. Alcoholic liver disease 

Carbohydrate-deficient transferrin (CDT) is the most used marker of chronic alcohol abuse. Human serum transferrin is a glycoprotein synthesized by the liver and involved in iron transport between sites of absorption and delivery [21]. Chronic ethanol intake alters the normal microheterogeneity pattern of transferrin as a consequence of changes in the sialic acid contents [22], [23] (See Table 3 for an overview of the assays that were used to investigate the glycan status). A decreased level of dolichol has been observed in rats fed ethanol [24]. The abnormal terminal sialylation can be explained by a decrease in β-galactoside α2,6 sialyltransferase (ST6GalI) mRNA and protein expression and/or an increase in hepatocyte membrane associated sialidase observed during chronic alcohol abuse [25], [26], [27]. Oxidation products of ethanol such as acetaldehyde interfere with the N-glycan biosynthesis and/or transfer by binding the involved enzymes. Therefore, CDT is likely the result of changes in glycosylation during biosynthesis and catabolism. Although CDT is recognized as a marker of chronic alcohol consumption, the reliability of this marker is largely dependent on the analytical fractionation method (capillary electrophoresis). Alterations in glycosylation and lack of clinical analytical standard methods might contribute to the discrepancy and sensitivity of CDT in clinical settings [28], [29].

Table 3. Overview of the main assays used to investigate the glycan status at protein or genetic level in different etiologies of chronic liver diseases.
Etiologies
Alcoholic liver diseasesFatty liver diseases – bile related diseasesViral liver diseasesHepatocellular carcinoma
HPAEC-PAD [32]MALDI-TOF MS [41]DSA-FACE [94]DSA -FACE [95]
CIAE with Con A [34] CIAE with Con A [60], [61], [68], [71], [77], LCA [52], [53], [60], [63], [64], [68]
HPLC [37]HPLC [38]HPLC [100]HPLC [60], [64], [67], [69]
Enzyme Assays for GT [24], [36], [37], MT [37], ST [37]Enzyme Assay for GnT-III [39] and α1,6FT [40]Enzyme Assay of GnT-III [42], [43], [44], GnT-IV [42], [43], GnT-V [42], [43], GT [42]Enzyme Assay of α1-6 FT [59], [62], GnT-I [73], GnT-III [62], [71], [72], [73], [79], [80], [82], GnT-IV [62], [80], [82], GnT-V [62], [77], [79], [80], [82], [83], ST [49]
Assays usedFucose-binding lectin coupled to Sepharose beads [33]Lectin Blotting with E-PHA [38], [39], L-PHA [39], AAL [40], [41], AOL [41], Con A [41], SSA [41], MAM [41]Lectin FLISA [100]Lectin Blotting with Con A [72], LCA [59], E-PHA [72], [73], [74], L-PHA [84] Allo A [66]
RT-PCR [27] RT-PCR [44], [45]RT-PCR [69], [72], [77]
Northern Blot [25]Northern Blot [38], [39]Northern Blot [42], [43], [45]Northern Blot [59], [72], [74], [79], [82], [84]
Chromatofocusing [23] Affinity Column Chromatography [54], [66], [70]
FACS analysis [42], [43]Lectin-Affinity Electrophoresis Con A [55], LCA [55], [59], E-PHA [55], Allo A [55]
In situ hybridization [48]
Immunocytochemical determination [82], [84]

Desialylation is the most important alteration observed in ALD. Besides transferrin, many other proteins are know to be desialylated in ALD including orosomucoid, α1-antitrypsin, ceruloplasmin [30]. Therefore, the overall serum desialylation pattern was studied in which serum total sialic acid (TSA) and serum free sialic acid (FSA) were studied as potential markers of alcohol abuse [31]. Both TSA en FSA were significantly increased during alcohol abuse, but as markers, they have a low sensitivity (46%) and negative predictive value (27%). Consequently, the clinical utility for screening alcoholic patients is limited.

In terms of changes in glycosylation status, haptoglobin (Hp) has also been proposed as a candidate marker of ALD, especially alcoholic cirrhosis [32], [33]. This hemoglobin-scavenger showed two major changes in glycosylation: increased branching and increased fucosylation (hyperfucosylation i.e. the increased presence of fucose residues in the glycan structure). Hyperfucosylation is predominantly present at the α1,3 position linked to the subterminal GlcNAc instead of the core fucose position. The activity of α1,3 fucosyltransferase in blood is directly associated with elevated Hp concentrations. Changes in branching were less frequent but still significant in ALD. This was determined by an increased N-acetylglucosamine content of the Hp molecule (mol/mol Hp). However, these alterations were not specific for alcoholic cirrhosis and were also observed in cases of primary billary cirrhosis and chronic alcohol abusers, but were absent in chronic active hepatitis. Besides haptoglobin, other glycoproteins are known to exhibit an increased branching during alcoholic liver disease: α1 acid glycoprotein, α2-HS-glycoprotein and transferrin [34]. In the alcoholic groups, the proportion of Con A-unreactive subpopulations of these glycoproteins increased.

The Golgi apparatus plays an important role in the alteration of glycosylation patterns of all liver diseases. This is especially well studied in ALD [35], [36]. Characteristic is the significant accumulation of hepatic protein caused by impaired glycosylation and glycoprotein trafficking. An example of impaired glycosylation is a decrease in the activity of galactosyltransferase [36]. A proposed explanation for this reduced Golgi functioning is the deficient polymerization of microtubular protein as a downstream consequence of hepatic acetaldehyde accumulation due to ethanol oxidation [35].

A study by Ghosh et al. [37] summarized the previously mentioned alterations of N-glycosylation in an experimental rat model of ALD: decreased enzyme activities of mannosyltransferase and galactosyltransferase, lowered intracellular dolichol concentration, strong decreased synthesis and activity of ST6GalI. In addition to these findings, an increase of 30% in liver weight was observed compared to the body weight of this rat model. This is due to a significant accumulation of hepatic lipids and proteins, which leads to fatty liver and even steatosis. The novelty of this study is that they also showed an alteration in O-glycosylation. The protein apolipoprotein E was used as a model to study this. As in N-glycosylation, an impairment of mannosylation and sialylation was shown. The relative ratio of labeled sugar to leucine incorporation (glycosylation index) showed a 50% decrease for relative mannosylation of Apo E molecule at both the microsomal and Golgi level. Glycan structures on apolipoprotein E were hypothesized to play a role in the association between this apolipoprotein and high density lipoproteins (HDL) and very low density lipoproteins. The observed impairment (namely in mannosylation and sialylation of the protein molecule) might be responsible for the defective clearance of HDL and therefore lead to a defective cholesterol transport to the liver and subsequent hepatic lipid accumulation. Apolipoproteins and their attached glycans are also involved in the development of a fatty liver, which is discussed in the next section.

2.2. Fatty liver diseases 

Transgenic mice that specifically express N-acetylglucosaminyltransferase III (GnT-III) in the liver had hepatocytes with a swollen oval-like morphology, with many lipid droplets [38], [39]. GnT-III transfers a GlcNAc residue to the trimannoside core of N-glycans. The aberrant glycosylation of apolipoprotein-B (increased level of bisecting GlcNAc) disturbs the function of this protein and causes a decrease in the release of lipoproteins and an accumulation of apolipoprotein-B in the liver. These transgenic mice showed microvesicular fatty alterations with abnormal lipid accumulation in the hepatocytes. Beside apolipoprotein-B, apolipoprotein A-1 was also significantly increased in liver tissue in these mice. These studies reveal that glycosylation is a factor in the regulation of lipoprotein metabolism and that an aberrant N-glycan structure can modify certain biochemical parameters of lipid metabolism in the liver.

Not only GnT-III, but also ectopic expression of α1,6 fucosyltransferase causes steatosis in the liver and kidney [40]. Similarly, numerous small vacuoles filled with lipid droplets were histologically observed. In contrast to the studies with GnT-III, a significant increase in cholesterol esters and triglycerides was observed. Crucial in this model was the lowered activity of lysosomal acid lipase (LAL), caused by an increased fucosylation of this enzyme. The accumulation of inactive LAL might contribute to the lipid accumulation observed in the lysosomes of the liver.

2.3. Bile-related liver diseases 

Fucosylation of N-glycans on glycoproteins seems to be the most important alteration in bile-related liver diseases. An increase in fucosylation of biliary glycoproteins was reported by Nakawaga et al. [41] who analyzed the oligosaccharide structure by used 2D mapping HPLC (high performance liquid chromatography) and MALDI-TOF (matrix assisted laser desorption/ionization time-of-flight) mass spectrometry in bile and serum. The binding of biliary glycoproteins to a lectin which recognizes fucose residues, was enhanced and specific glycoproteins (α1-antitrypsin, α1-acid glycoprotein and haptoglobin) were stronger fucosylated in bile opposite serum of patients with gallbladder stones. It was suggested that fucosylation of glycoproteins could be a possible signal for secretion into bile ducts in the liver. α1-6 fucosyltransferase (Fut8)-deficient mice showed decreased levels of α1-antitrypsin and α1-acid glycoprotein (AGP) in bile and were relocated to the liver of Fut8-deficient mice [41]. To date, no other study has confirmed these results in an experimental setting.

2.4. Viral liver diseases 

Among the viral liver diseases, the hepatitis B virus (HBV) infection has been most extensively investigated in search of changes in glycosylation. Alteration in glycosylation between HBV and HCC show many similarities, because of the correlation of long-term HBV-infection and the increased risk of HCC. GnT-III is likely to play a prominent role in the alteration of glycosylation in these viral infected population.

An in vitro experiment using human hepatoblastoma cell line (Huh6) transfected with the hepatitis B viral genome (HB611) showed a specific decrease in GnT-III activity opposite the untransfected Huh6 cell line [42]. Glycosyltransferases other than GnT-III were unaltered. Subsequent Northern blot analysis showed that this difference was due to a decreased transcript of GnT-III. GnT-III and GnT-V are known to compete for the same substrate and by suppressing GnT-III, HBV might therefore push the hepatoblastoma cells towards a more malignant phenotype with branched complex-type N-linked oligosaccharides proteins. On the other hand the GnT-III gene was transfected into the HB611 cell line, which resulted in decreased secretion of HBV-related proteins and a marked suppression of HBV-related mRNAs [43]. This suggests a possible anti-viral role for this enzyme in in vivo circumstances. These results indicate that some glycoproteins whose oligosaccharide structures are changed by overexpression of GnT-III suppress HBV protein expression. This might be a unique approach to prevent HBV replication.

However, contradictory results exist concerning GnT-III expression observed in HBV-transfected hepatocyte cell lines [44], [45]. For instance, in contrast to the hepatoblastoma cell line, a HBV-transfected fetal hepatocyte cell line showed an increased GnT-III expression. In addition, HBV x-protein (HBx)-transfected cell line showed a much stronger GnT-III expression in contrast to hepatocarcinoma cells and the GnT-III-transfected cell line [46]. This result suggests strong GnT-III promoter enhancing activity for the HBx-gene. These HBx-transfected cells also showed an accumulation of aberrant glycosylated apolipoprotein B’s (increased levels of bisecting GlcNAc), triglyceride and cholesterol. These findings were supported by the previous mentioned transgenic mice in fatty liver diseases. The observed discrepancy between these studies is caused by the difference in experimental setting (used cell line, transfected genes). It is therefore unclear whether GnT-III exhibits a positive or negative influence on the HBV infection.

One of the few reports where differences in glycosylation could serve as a discriminating factor between liver diseases was performed by Anderson et al. [46]. They used AGP as the model protein to test this. Hyperfucosylation (predominantly α1-3) occurred in all liver diseases, especially in hepatitis B and C virus infected patients, but an additional N-acetylgalactosamine was detected in the majority of hepatitis C patients as determined by high performance anion exchange chromatography analysis. This is unexpected because this monosaccharide is rarely present in N-glycans.

Back to Article Outline

3. Alteration of glycosylation in hepatocellular carcinoma 

Alterations to the normal function of the glycosylation machinery are increasingly recognized as an indication of malignant cellular transformation. In relation to HCC, glycosyltransferases are most intensively studied. Three glycosyltransferases are considered crucial: N-acetylglucosaminyltransferase V (GnT-V), N-acetylglucosaminyltransferase III (GnT-III) and α1-6 fucosyltransferase (α1-6FT). These enzymes have been suggested in the development of fibrosis, but the relation with HCC is much clearer and well documented. The activity of these three enzymes is significantly altered in hepatoma tissues or serum of HCC patients and suggest a subsequent change in glycan structures.

Some evidence can also be found that the expression of ST6GalI is up-regulated in human HCC patients and in a transgenic mouse model of HCC [47], [48], [49]. We already mentioned this enzyme in ALD where it was down-regulated. It seems that ST6GalI is differently regulated in cirrhosis and HCC.

3.1. Fucosyltransferases 

The best known marker in HCC is the elevated serum concentration of fucosylated α-fetoprotein (AFP) and this has become a standard in cancer diagnostics over the years. The presence of molecular variants of AFP with different carbohydrate chains has first been demonstrated by reactions with lectins such as concanavalin A and lentil lectin (Lens culinaris agglutinin) [50].

It is known that the serum AFP concentration levels also increase in benign liver diseases such as acute hepatitis and cirrhosis. Therefore, the serum AFP concentration alone is of little use in the differential diagnosis of HCC and benign liver diseases [51]. Consequently, it is important to develop a method that distinguishes between HCC and non-neoplastic liver diseases [52], [53]. The fucosylation index could be a possible aid to achieve this goal. Fucosylated AFPs are specifically recognized by the lentil lectin and were determined by crossed immunoaffino-electrophoresis as the lentil lectin reactive fraction in total AFP. The fucosylation index is then the percentage of the lentil lectin reactive fraction in total AFP. A highly significant increase in fucosylation index was observed in cirrhotic patients after the development of HCC. This index is useful in the early detection of HCC, especially in patients with cirrhosis which are at risk of developing HCC in the following years.

Similar results were obtained by other research groups in patients with low serological levels of AFP [54], [55]. The latter study determined AFP using different lectins: Con A, LCA-A, E-PHA and allo A. The AFP bands separated with different lectins in this study appear to have sugar-chain heterogeneity with little overlap. The presence of this tumor heterogeneity forms the basis for the combined evaluation of tumor markers. A fraction of AFP obtained with the lectin Lens culinaris agglutinin A (AFP-L3) and another fraction of AFP obtained with lectin erythroagglutinating phytohemagglutinin (AFP-P4) were simultaneous analyzed and resulted in a very high sensitivity (97%) and specificity (99,7%) in monitoring the evolution of HCC in cirrhotic patients. However, when AFP-L3 is evaluated individually, it is known to have a high specificity but not enough sensitivity to be considered a surveillance tool for HCC [56]. Because AFP-L3 has been related to tumors with rapid growth, larger size, portal vein invasion and metastasis [57], [58], its utility may be more of a prognostic marker for HCC rather than for a screening test. In addition to the increase in fucosylation patterns, mRNA of the responsible enzyme α1-6FT was shown to be enhanced in proportion to the enzymatic activity in HCC patients [59].

Besides an increment in fucosylation, formation of new antennae is also a characteristic feature of the carbohydrate chains of AFP from patients with neoplastic liver diseases [60]. Fucosylated tri-antennary, tetra-antennary and penta-antennary glycans attached to transferrin synthesized by the human hepatocarcinoma cell line Hep G2 have been reported by Campion et al. [61], Ohno et al. provided additional results which confirmed these findings in vitro [62]. AFPs were purified from 2 different hepatoma cell lines (Hep G2 and HuH-7) and a hepatoblastoma cell line (HuH-6). Simultaneously, the activities of α1-6FT, GnT-III and GnT-V were assayed in these cell lines. Fucosylated biantennary structure and α1-6FT were most frequently detected in all three cell lines. In addition, Hep G2 cells contained a high level of GnT-V, which catalyzes the formation of a triantennary structure, while GnT-III was elevated in HuH-6 cells. These data indicate that the sugar structures of AFP in these cell lines correlate well with the activities of α1-6FT, GnT-III and GnT-V.

Besides AFP, α1-antitrypsin (AAT) is known to express fucosylated biantennary glycans distinctive for HCC. Similar to AFP, the percentage of reactive species with the lectin Lens culinaris agglutinin was shown to be significantly higher in HCC than in benign liver diseases and normal controls [63]. Further structural analysis on the AAT-attached glycans was performed and it was shown that the fucosylated biantennary glycans were increased two-fold opposite non-fucosylated glycans [64]. Under normal circumstances, non-fucosylated biantennary glycans are the most abundant glycans on AAT [65]. Therefore, an increment in fucosylation on AAT is a characteristic of patients with HCC. Transferrin (TF) also shows significant changes in glycosylation patterns (fucosylation) but shows too many structural variation in contrast to AFP and AAT [66], [67], [68]. Both glycoproteins are useful alternatives in case of low AFP-levels in order to detect HCC. Although, in contrast to APF, LCA-reactive AAT and TF are not able to discriminate between HCC and liver cirrhosis.

Very recently, a study appeared that links N-glycan alterations to drug resistance in HCC. Especially α1–6 FT was overexpressed in drug-resistant cell lines and the glycomics analysis showed high core-fucosylation [69]. This further emphasizes the existence and importance of fucosylation anomalies occur in HCC.

3.2. N-acetylglucosaminyltransferases 

3.2.1. N-acetylglucosaminyltransferase III 

N-acetylglucosaminyltransferase III adds a “bisecting” GlcNAc in β1,4 linkage to the β-linked mannose Asn-linked core structure. This enzyme was determined in hepatic nodules of humans and rats and the GnT-III activity in these nodules were shown to be significantly higher in contrast to its surrounding and the control liver [70], [71], [72]. These studies suggest a possible correlation between the GnT-III activity and the precancerous stage of hepatocarcinogenesis. The activity of GnT-III in serum samples of HCC patients was determined by HPLC and it was significantly higher up-regulated in these patients in contrast to patients showing liver cirrhosis, chronic hepatitis and healthy controls. The GnT-III activity could be useful in HCC treatment follow-up because GnT-III activity is significantly reduced in HCC following percutaneous ethanol infection therapy and/or transcatheter arterial chemoembolization [71].

GnT-III exerts different (pathological) activity in an experimental mouse and rat model [70], [73]. Hepatic tumor formation in the mouse model was not accompanied by an increase in GnT-III activity nor glycoproteins expression showing a bisecting GlcNAc. They could also show that in Mgat3−/− mice (GnT-III coding gene) initiation of hepatic neoplasms was normal, but progression was severely retarded. A glycoprotein with the bisected GlcNAc synthesized outside of the liver might act as a necessary growth factor in tumor progression. Moreover, diethylnitrosamine (DEN) induced HCC in transgenic mice (Mgat3Δ/Δ, carrying a deletion in the Mgat3 gene) remained unaltered by high levels of GnT-III in the liver. Ectopic expression of the Mgat3 gene in the liver followed by DEN induced HCC and subsequent phenobarbital (PB) treatment did not result in significant changes in the number of tumors or liver weight [74]. The retarded tumor progression in the Mgat3−/− mice was showed a decrease in hepatocyte proliferation rather than an increase in hepatocyte apoptosis [75]. When the glycoprotein factor lacks the bisecting GlcNAc, it can not exert its normal function. The growth factor acting glycoprotein with the bisecting GlcNAc has still to be identified.

3.2.2. N-acetylglucosaminyltransferase V 

β1-6-GlcNAc branching providing by GnT-V is directly associated with metastasis [76] and might serve as marker for tumor invasiveness in HCC patients [77]. GnT-V is coded on the Mgat5 gene which is regulated by the Ras signaling pathway. This pathway is commonly up-regulated in various tumor cells [78]. In analogy with α1-6 fucose, β1-6 branching is also a molecular signal that exerts a particular biological function. Phosphorylation is another posttranslational modification and its importance in inter-/intracellular signal transduction is well known. It seems that glycosylation might have a similar signaling function.

Both GnT-III and GnT-V activity are increased in HCC, but the activity of GnT-III is much more prominent than GnT-V [79], [80], [81]. As previously mentioned, both glycosyltransferases compete for the same substrate. An increase in GnT-III activity might suppress GnT-V activity and subsequent GlcNAc β1–6 branching. In several cancers it has been shown that GnT-III counteracts tumor progression, while GnT-V promotes it [44], [76]. Moreover, it is known that GnT-III and GnT-V activity undergo opposing changes during different stages of the cell cycle in a hepatoma cell line [82]. These changes might be the result of a change in regulatory mechanisms of the cell cycle. The peak activity of GnT-III (during G0/G1) coincided with the lowest activity of GnT-V during G0/G1-stage while the opposite is seen during G2/M-stage.

A progressive increase in GnT-V was observed during HCC and was likely correlated with the TNM-classification of HCC. GnT-V activity in advanced HCC was clearly up-regulated opposite early HCC [83]. However, these findings were contradicted by the study of Ito et al. which showed a reversed expression level in early and advanced stages of HCC [84].

Back to Article Outline

4. From glycoproteomics to glycomics 

Glycosylation has evolved from simple surface glycans for the discrimination of blood types to glycans determining tumor progression. Given the importance of glycans in various biological activities (e.g. liver diseases) and the restriction of genomics and proteomics, glycomics has emerged among other – omics domains. The glycome can be defined as the complete set of glycan structures present in specific cells, tissues or organisms. The dynamic nature of these post-translational modifications and their complex regulation indicate that direct mapping of the total pathologic changes to glycans may be more informative than characterizing the glycans on individual proteins [85], [86]. Glycomic analysis also overcomes the problem of microheterogeneity of single proteins. Individual proteins mostly do not have one characteristic alteration in liver pathology (what would be favorable in clinical testing), but multiple transitions.

Many previous analyses have been performed with the technology also commonly used in proteomics. For instance, two-dimensional gel electrophoresis of serum from alcoholic patients with or without liver cirrhosis resulted in microheterogeneity among Hp, AAT and TF proteins in these patients [87]. Further analysis using lectin blotting showed common changes in glycosylation as an early indication or result of excessive alcohol consumption. A similar approach was used by Henry et al. [88] in the study of carbohydrate deficiency syndrome. In addition, mass spectrometry provides glycan structure information based on fragmentation patterns and the mass of the different sugar components.

Nowadays, N-glycans are enriched by peptide-N-glycosidase and are investigated in a protein-independent way. The N-glycan fraction can either be analyzed by mass spectrometry or by normal phase high performance liquid chromatography (NP-HPLC) using lectin-affinity column. NP-HPLC provides N-glycan profile, visualized as an electroferogram. Improvement has been made in the N-glycan profiling using DNA sequencer-assisted fluorophore-assisted carbohydrate electrophoresis (DSA-FACE) [89]. An important advantage of this technology is that the results can be combined and compared with those obtained from MALDI-TOF of the same analytes on at least the same level of sensitivity. For a clear overview of the used techniques in glycomics we refer to the following reviews [90], [91], [92], [93].

Based on recurring, unique glycan characteristics in various liver diseases, N-glycan study (N-glycome) has become an interesting research field in search of useful carbohydrate biomarkers. Using the DSA-FACE approach, Callewaert et al. were able to successfully discriminate non-cirrhotic from cirrhotic patients based on their serological N-glycome [94]. The electroferogram particularly showed an increase in agalacto sugar and decrease in tri-antennary sugar. The logarithmic ratio of the peak heights of a biantennary, core-fucosylated and bisecting GlcNAc modified sugar (increased in cirrhosis) and a triantennary sugar (decreased in cirrhosis) showed to have a good diagnostic value for the recognition of cirrhosis (AUC=0,87, specificity=100% and sensitivity=75%) and was renamed the GlycoCirrhoTest. The main disadvantage of this technique is that it can not be used as a follow-up tool, a significant difference could only be observed between F0–F3 and F4. In addition, Liu et al. were able to diagnose HCC based on the log ratio of two identified peaks following DSA-FACE on patient sera. A fucosylated tri-antennary glycan was increased in contrast to a bisecting biantennary glycan in HCC and this is consistent with previous findings on GnT-V and GnT-III activity in HCC. This test was named GlycoHCCTest in analogy with GlycoCirrhoTest and showed an accuracy similar to AFP (∼80%) [95]. However, the individual variation in the blood serum profiles was still rather high. Therefore, more extensive N-glycan profiling studies on individuals should be carried out, including longitudinal samples over long periods.

In addition to this study, N-glycome profile studies have emerged using MALDI-TOF mass spectrometry (MS). The main idea behind this approach is the reduction of analysis time and sample used for analysis [96], [97]. Although mainly used in the structural analysis of glycans, quantitative profiling is also possible [96], [97], [98], [99]. In this regard, Morelle et al. [97] studied total serum N-glycome of cirrhotic patients applying mass spectrometry for both purposes: MALDI-TOF MS for mass profiling and electrospray ionization ion-trap MS for structural characterization. However, the results of this approach were limited to 26 different identified glycans (normally over a hundred) due to isobaric glycan compounds, which cannot be distinguished. Besides similarity with the findings of Callewaert et al. (increase in bisecting GlcNAc and core fucosylation), Morelle et al. discovered an interesting group of neutral oligosaccharides. Mass spectrometry-based are, however, less suitable for diagnostics, because of the low reproducibility.

Serum is an obvious choice to study the N-glycome in association with liver disease, because a large portion of these N-glycans originate from serum proteins and are produced in the liver. However, serum also consists largely of immunoglobulins (Ig), especially IgG. IgG is produced in B-cells and are also glycosylated; therefore the whole serum N-glycome might be biased by these IgG-related glycan structures [94]. Metha et al. [100] took this to the next level and clearly showed that the major serum glycoprotein containing altered glycosylation as a function of cirrhosis is not a liver-derived protein at all. It was rather IgG (produced by B-cells) that was specifically reactive to the alpha-Gal epitope. However, in healthy subjects, this only constitutes ∼1% of total serum IgG. Moreover, this epitope is absent in humans, but is abundantly synthesized by bacteria.

Back to Article Outline

5. General conclusions 

We have clearly shown that glycosylation changes in liver pathology. Although liver diseases of different etiology have their own markers (e.g. CDT in alcoholic liver disease or AFP in HCC), the alterations of glycosylation of these proteins are quite uniform. In other words, the same type of alterations seems to continuously reappear in the different types of chronic liver diseases: increased fucosylation, increased branching and increased bisecting GlcNAc modified glycans. The activities of the glycosyltransferases that are responsible for these alterations (α1–6 FT, GnT-V and GnT-III, respectively) are extensively studied and are known to be significantly elevated in liver pathology.

Clinical tests based on these altered glycan structures have been studied intensively, but only a few have made it into clinical practice. An important reason is that although the methods used for determining the altered sugar chain moieties are reliable, they are also time consuming. With the advent of new technologies that enable simple and rapid screening methods, we are getting a step closer to finding the holy grail of hepatic clinical testing, an non-invasive test that can replace a liver biopsy. However, a combination of markers and techniques will probably be necessary to achieve this goal.

Back to Article Outline

Acknowledgement 

The authors thank Kin Jip Cheung for editing the manuscript. BB is a recipient of a scholarship GOA BOFF07/GOA/017 of the University Ghent Research Fund (BOF).

Back to Article Outline

References 

  1. Ploegh H, Neefjes JJ. Protein glycosylation. Curr Opin Cell Biol. 1990;2:1125–1130
  2. Hebert DN, Garman SC, Molinari M. The glycan code of the endoplasmic reticulum: asparagine-linked carbohydrate as protein maturation and quality-control tags. Trends in Cell Biol. 2005;15:364–370
  3. Kornfeld S. Diseases of abnormal protein glycosylation: an emerging area. J Clin Invest. 1998;101:1293–1295
  4. Lis H, Sharon N. Protein glycosylation: structural and functional aspects. Eur J Biochem. 1993;218:1–27
  5. Schachter H. The subcellular sites of glycosylation. Biochem Soc Symp. 1974;40:57–71
  6. Paulson JC, Colley KJ. Glycosyltransferases – structure, localization, and control of cell type-specific glycosylation. J Biol Chem. 1989;264:17615–17618
  7. Spiro RG. Protein glycosylation: nature, distribution, enzymatic formation, and disease implications of glycopeptides bonds. Glycobiology. 2002;12:43R–56R
  8. Burda P, Aebi M. The dolichol pathway of N-linked glycosylation. Biochem Biophys Acta. 1999;1426:239–257
  9. Javaud C, Dupuy F, Maftah A, Julien R, Petit JM. The fucosyltransferase gene family: an amazing summary of the underlying mechanisms of gene evolution. Genetica. 2003;118:157–170
  10. Taniguchi N, Miyoshi E, Ko JH, Ikeda Y, Ihara Y. Implication of N-acetylglucosaminyltransferases III and V in cancer: gene regulation and signaling mechanism. Biochim Biophys Acta. 1999;1455:287–300
  11. Harduin-Lepers A, Vallejo-Ruiz V, Krzewinski-Recchi MA, Samyn-Petit B, Julien S, Delannoy P. The human sialyltransferase family. Biochimie. 2001;83:727–737
  12. Zhao Y, Takahashi M, Gu J, Miyoshi E, Matsumoto A, Kitazume S, et al. Functional roles of N-glycans in cell signaling and cell adhesion in cancer. Cancer Sci. 2008;99:1304–1310
  13. Varki A. Biological roles of oligosaccharides: all of the theories are correct. Glycobiology. 1993;3:97–130
  14. Dwek RA. Glycobiology: ‘towards understanding the function of sugars’. Biochem Soc Trans. 1995;23:1–25
  15. Taniguchi N, Miyoshi E, Gu J, Honke K, Matsumoto A. Decoding sugar functions by identifying target glycoproteins. Curr Opin Struct Biol. 2006;16:561–566
  16. Sawamura T, Nakada H, Hazama H, Shiozaki Y, Sameshima Y, Tashiro Y. Hyperasialoglycoproteinemia in patients with chronic liver-diseases and or liver-cell carcinoma – asialoglycoprotein receptor in cirrhosis and liver-cell carcinoma. Gastroenterology. 1984;87:1217–1221
  17. Burgess JB, Baenziger JU, Brown WR. Abnormal surface distribution of the human asialoglycoprotein receptor in cirrhosis. Hepatology. 1992;15:702–706
  18. Martinez J, Barsigian C. Carbohydrate abnormalities of N-linked plasma glycoproteins in liver-disease. Lab Invest. 1987;57:240–257
  19. Turner GA. N-glycosylation of serum–proteins in disease and its investigation using lectins. Clin Chim Acta. 1992;208:149–171
  20. Van Damme EJM, Peumans WJ, Bardocz S, Pusztai A. Handbook of plant lectins: properties and biomedical applications. Amsterdam: John Wiley & Sons; 1998;
  21. De Jong G, Van Dijk JP, Van Eijk HG. The biology of transferrin. Clin Chim Acta. 1990;190:1–46
  22. Flahaut C, Michalski JC, Danel T, Humbert MH, Klein A. The effects of ethanol on the glycosylation of human transferrin. Glycobiology. 2003;13:191–198
  23. Storey EL, Anderson GJ, Mack U, Powell LW, Halliday JW. Desialylated transferrin as a serological marker of chronic excessive alcohol ingestion. Lancet. 1987;1:1292–1294
  24. Cottalasso D, Domenicotti C, Traverso N, Pronzato MA, Nanni G. Influence of chronic ethanol consumption on toxic effects of 1,2-dichloroethane: glycolipoprotein retention and impairment of dolichol concentration in rat liver microsomes and Golgi apparatus. Toxicology. 2002;178:229–240
  25. Rao M, Lakshman MR. Chronic ethanol consumption leads to destabilization of rat liver β-galactoside-sialyltransferase mRNA. Metabolism. 1999;48:797–803
  26. Xin Y, Lasker JM, Lieber CS. Serum carbohydrate-deficient transferrin – mechanism of increase after chronic alcohol intake. Hepatology. 1995;22:1462–1468
  27. Gong M, Garige M, Hirsch K, Lakshman MR. Liver Gal beta 1,4GlcNAc alpha 2,6-sialyltransferase is down-regulated in human alcoholics: possible cause for the appearance of asialoconjugates. Metabolism. 2007;56:1241–1247
  28. Liu YS, Xu GY, Cheng DQ, Li YM. Determination of serum carbohydrate-deficient transferrin in the diagnosis of alcoholic liver disease. Hepatobiliary Pancreat Dis Int. 2005;4:265–268
  29. Stadheim LM, O’Brien JF, Lindor KD, Gores GJ, McGill DB. Value of determining carbohydrate-deficient transferrin isoforms in the diagnosis of alcoholic liver disease. Mayo Clin Proc. 2003;78:703–707
  30. Tsutsumi M, Wang JS, Takada A. Microheterogeneity of serum glycoproteins in alcoholics – is desialo-transferrin the marker of chronic alcohol-drinking or alcoholic liver-injury. Alcohol Clin Exp Res. 1994;18:392–397
  31. Chrostek L, Cylwik B, Korcz W, Krawiec A, Koput A, Supronowicz Z, et al. Serum free sialic acid as a marker of alcohol abuse. Alcohol Clin Exp Res. 2007;31:996–1001
  32. Mann AC, Record CO, Self CH, Turner GA. Monosaccharide composition of haptoglobin in liver-diseases and alcohol-abuse – large changes in glycosylation associated with alcoholic liver-disease. Clin Chim Acta. 1994;227:69–78
  33. Chambers W, Thompson S, Skillen AW, Record CO, Turner GA. Abnormally fucosylated haptoglobin as a marker for alcoholic liver-disease but not excessive alcohol-consumption or Nonalcoholic liver-disease. Clin Chim Acta. 1993;219:177–182
  34. Jezequel M, Seta NS, Corbic MM, Feger JM, Durand GM. Modifications of concanavalin a patterns of alpha-1-acid glycoprotein and alpha-2-Hs glycoprotein in alcoholic liver-disease. Clin Chim Acta. 1988;176:49–57
  35. Matsuda Y, Takada A, Takase S, Sato H. Accumulation of glycoprotein in the Golgi-apparatus of hepatocytes in alcoholic liver injuries. Am J Gastroenterol. 1991;86:854–860
  36. Guasch R, Renaupiqueras J, Guerri C. Chronic ethanol-consumption induces accumulation of proteins in the liver Golgi-apparatus and decreases galactosyltransferase activity. Alcohol Clin Exp Res. 1992;16:942–948
  37. Ghosh P, Liu QH, Lakshman MR. Long-term ethanol exposure impairs glycosylation of both N- and O-glycosylated proteins in rat-liver. Metabolism. 1995;44:890–898
  38. Ihara Y, Yoshimura M, Miyoshi E, Nishikawa A, Sultan AS, Toyosawa S, et al. Ectopic expression of N-acetylglucosaminyltransferase III in transgenic hepatocytes disrupts apolipoprotein B secretion and induces aberrant cellular morphology with lipid storage. Proc Natl Acad Sci USA. 1998;95:2526–2530
  39. Lee J, Song EY, Chung TW, Kang SK, Kim KS, Chung TH, et al. Hyperexpression of N-acetylglucosaminyltransferase-III in liver tissues of transgenic mice causes fatty body and obesity through severe accumulation of Apo A-I and Apo B. Arch Biochem Biophys. 2004;426:18–31
  40. Wang W, Li W, Ikeda Y, Miyagawa JI, Taniguchi M, Miyoshi E, et al. Ectopic expression of alpha1,6 fucosyltransferase in mice causes steatosis in the liver and kidney accompanied by a modification of lysosomal acid lipase. Glycobiology. 2001;11:165–174
  41. Nakagawa T, Uozumi N, Nakano M, Mizuno-Horikawa Y, Okuyama N, Taguchi T, et al. Fucosylation of N-glycans regulates the secretion of hepatic glycoproteins into bile ducts. J Biol Chem. 2006;281:29797–29806
  42. Miyoshi E, Nishikawa A, Ihara Y, Hayashi N, Fusamoto H, Kamada T, et al. Selective suppression of N-acetylglucosaminyltransferase-Iii activity in a human hepatoblastoma cell-line transfected with hepatitis-B virus. Cancer Res. 1994;54:1854–1858
  43. Miyoshi E, Ihara Y, Hayashi N, Fusamoto H, Kamada T, Taniguchi N. Transfection of N-acetylglucosaminyltransferase-Iii gene suppresses expression of hepatitis-B virus in a human hepatoma-cell line, Hb611. J Biol Chem. 1995;270:28311–28315
  44. Shim JK, Lee YC, Chung TH, Kim CH. Elevated expression of bisecting N-acetylglucosaminyltransferase-III gene in a human fetal hepatocyte cell line by hepatitis B virus. J Gastroenterol Hepatol. 2004;19:1374–1387
  45. Kang SK, Chung TW, Lee JY, Lee YC, Morton RE, Kim CH. The hepatitis B virus x protein inhibits secretion of apolipoprotein B by enhancing the expression of N-acetylglucosaminyltransferase III. J Biol Chem. 2004;279:28106–28112
  46. Anderson N, Pollacchi A, Hayes P, Therapondos G, Newsome P, Boyter A, et al. A preliminary evaluation of the differences in the glycosylation of alpha-1-acid glycoprotein between individual liver diseases. Biomed Chromatogr. 2002;16:365–372
  47. Cao Y, Merling A, Crocker PR, Keller R, Schwartz-Albiez R. Differential expression of beta-galactoside alpha 2,6 sialyltransferase and sialoglycans in normal and cirrhotic liver and hepatocellular carcinoma. Lab Invest. 2002;82:1515–1524
  48. Dall’Olio F, Chiricolo M, D’Errico A, Gruppioni E, Altimari A, Fiorentino M, et al. Expression of beta-galactoside alpha 2,6 sialyltransferase and of alpha 2,6-sialylated glycoconjugates in normal human liver, hepatocarcinoma, and cirrhosis. Glycobiology. 2004;14:39–49
  49. Pousset D, Piller V, Bureaud N, Monsigny M, Piller F. Increased alpha 2,6 sialylation of N-glycans in a transgenic mouse model of hepatocellular carcinoma. Cancer Res. 1997;57:4249–4256
  50. Bayard B, Kerckaert JP. Characterization and isolation of nine rat alpha-fetoprotein variants by gel electrophoresis and lectin affinity chromatography. Biochem Biophys Res Commun. 1977;77:489–495
  51. Karvountzis GG, Redeker AG. Relation of alpha-fetoprotein in acute hepatitis to severity and prognosis. Ann Intern Med. 1974;80:156–160
  52. Aoyagi Y, Suzuki Y, Isemura M, Nomoto M, Sekine C, Igarashi K, et al. The fucosylation index of alpha-fetoprotein and its usefulness in the early diagnosis of hepatocellular-carcinoma. Cancer. 1988;61:769–774
  53. Aoyagi Y, Saitoh A, Suzuki Y, Igarashi K, Oguro M, Yokota T, et al. Fucosylation index of alpha-fetoprotein, a possible aid in the early recognition of hepatocellular-carcinoma in patients with cirrhosis. Hepatology. 1993;17:50–52
  54. Du MQ, Hutchinson WL, Johnson PJ, Williams R. Differential alpha-fetoprotein lectin binding in hepatocellular-carcinoma – diagnostic utility at low serum levels. Cancer. 1991;67:476–480
  55. Taketa K, Sekiya C, Namiki M, Akamatsu K, Ohta Y, Endo Y, et al. Lectin-reactive profiles of alpha-fetoprotein characterizing hepatocellular-carcinoma and related conditions. Gastroenterology. 1990;99:508–518
  56. Sterling RK, Jeffers L, Gordon F, Sherman M, Venook AP, Reddy KR, et al. Clinical utility of AFP-L3% measurement in North American patients with HCV-related cirrhosis. Am J Gastroenterol. 2007;102:2196–2205
  57. Li D, Mallory T, Satomura S. AFP-L3: a new generation of tumor marker for hepatocellular carcinoma. Clin Chim Acta. 2001;313:15–19
  58. Yamashiki N, Seki T, Wakabayashi M, Nakagawa T, Imamura M, Tamai T, et al. Usefulness of lens culinaris agglutinin A-reactive fraction of alpha-fetoprotein (AFP-L3) as a marker of distant metastasis from hepatocellular carcinoma. Oncol Rep. 1999;6:1229–1232
  59. Noda K, Miyoshi E, Uozumi N, Yanagidani S, Ikeda Y, Gao CX, et al. Gene expression of alpha 1-6 fucosyltransferase in human hepatoma tissues: a possible implication for increased fucosylation of alpha-fetoprotein. Hepatology. 1998;28:944–952
  60. Aoyagi Y, Suzuki Y, Igarashi K, Saitoh A, Oguro M, Yokota T, et al. Carbohydrate structures of human alpha-fetoprotein of patients with hepatocellular carcinoma: presence of fucosylated and non-fucosylated triantennary glycans. Br J Cancer. 1993;67:486–492
  61. Campion B, Leger D, Wieruszeski JM, Montreuil J, Spik G. Presence of fucosylated triantennary, tetraantennary and pentaantennary glycans in transferrin synthesized by the human hepatocarcinoma cell line Hep G2. Eur J Biochem. 1989;184:405–413
  62. Ohno M, Nishikawa A, Koketsu M, Taga H, Endo Y, Hada T, et al. Enzymatic basis of sugar structures of alpha-fetoprotein in hepatoma and hepatoblastoma cell lines: correlation with activities of alpha 1-6 fucosyltransferase and N-acetylglucosaminyltransferases III and V. Int J Cancer. 1992;51:315–317
  63. Sekine C, Aoyagi Y, Suzuki Y, Ichida F. The reactivity of alpha-1-antitrypsin with lens-culinaris agglutinin and its usefulness in the diagnosis of neoplastic diseases of the liver. Br J Cancer. 1987;56:371–375
  64. Saitoh A, Aoyagi Y, Asakura H. Structural analysis on the sugar chains of human alpha 1-antitrypsin: presence of fucosylated biantennary glycan in hepatocellular carcinoma. Arch Biochem Biophys. 1993;303:281–287
  65. Hodges LC, Laine R, Chan SK. Structure of the oligosaccharide chains in human alpha-1-protease inhibitor. J Biol Chem. 1979;254:8208–8212
  66. Yamashita K, Koide N, Endo T, Iwaki Y, Kobata A. Altered glycosylation of serum transferrin of patients with hepatocellular-carcinoma. J Biol Chem. 1989;264:2415–2423
  67. Matsumoto K, Maeda Y, Kato S, Yuki H. Alteration of asparagine-linked glycosylation in serum transferrin of patients with hepatocellular-carcinoma. Clin Chim Acta. 1994;224:1–8
  68. Suzuki Y, Aoyagi Y, Mori S, Suda T, Naitoh A, Isokawa O, et al. Microheterogeneity of serum transferrin in the diagnosis of hepatocellular carcinoma. J Gastroenterol Hepatol. 1996;11:358–365
  69. Kudo T, Nakagawa H, Takahashi M, Hamaguchi J, Kamiyama N, Yokoo H, et al. N-glycan alterations are associated with drug resistance in human hepatocellular carcinoma. Mol Cancer. 2007;6:32
  70. Narasimhan S, Schachter H, Rajalakshmi S. Expression of N-acetylglucosaminyltransferase-Iii in hepatic Nodules during rat-liver carcinogenesis promoted by orotic-acid. J Biol Chem. 1988;263:1273–1281
  71. Mori S, Aoyagi Y, Yanagi M, Suzuki Y, Asakura H. Serum N-acetylglucosaminyltransferase III activities in hepatocellular carcinoma. J Gastroenterol Hepatol. 1998;13:610–619
  72. Song EY, Kang SK, Lee YC, Park YG, Chung TH, Kwon DH, et al. Expression of bisecting N-acetylglucosaminyltransferase-III in human hepatocarcinoma tissues, fetal liver tissues, and hepatoma cell lines of Hep3B and HepG2. Cancer Invest. 2001;19:799–807
  73. Bhaumik M, Harris T, Sundaram S, Johnson L, Guttenplan J, Rogler C, et al. 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. 1998;58:2881–2887
  74. Yang XP, Bhaumik M, Bhattacharyya R, Gong S, Rogler CE, Stanley P. New evidence for an extra-hepatic role of N-acetylglucosaminyltransferase III in the progression of diethylnitrosamine-induced liver tumors in mice. Cancer Res. 2000;60:3313–3319
  75. Yang XP, Tang J, Rogler CE, Stanley P. Reduced hepatocyte proliferation is the basis of retarded liver tumor progression and liver regeneration in mice lacking N-acetylglucosaminyltransferase III. Cancer Res. 2003;63:7753–7759
  76. Dennis JW, Laferte S, Waghorne C, Breitman ML, Kerbel RS. Beta-1-6 branching of Asn-linked oligosaccharides is directly associated with metastasis. Science. 1987;236582–236585
  77. Yanagi M, Aoyagi Y, Suda T, Mita Y, Asakura H. N-acetylglucosaminyltransferase V as a possible aid for the evaluation of tumor invasiveness in patients with hepatocellular carcinoma. J Gastroenterol Hepatol. 2001;16:1282–1289
  78. Dennis JW, Granovsky M, Warren CE. Glycoprotein glycosylation and cancer progression. Biochim Biophys Acta. 1999;1473:21–34
  79. Miyoshi E, Nishikawa A, Ihara Y, Gu JG, Sugiyama T, Hayashi N, et al. N-acetylglucosaminyltransferase-Iii and N-acetylglucosaminyltransferase-V messenger-RNA levels in Lec rats during hepatocarcinogenesis. Cancer Res. 1993;53:3899–3902
  80. Nishikawa A, Gu J, Fujii S, Taniguchi N. Determination of N-acetylglucosaminyltransferase-Iii, N-acetylglucosaminyltransferase-Iv and N-acetylglucosaminyltransferase-V in normal and hepatoma tissues of rats. Biochim Biophys Acta. 1990;1035:313–318
  81. Koenderman AH, Koppen PL, Koeleman CA, van den Eijnden DH. N-acetylglucosaminyltransferase III, IV and V activities in Novikoff ascites tumour cells, mouse lymphoma cells and hen oviduct. Application of a sensitive and specific assay by use of high-performance liquid chromatography. Eur J Biochem. 1989;181:651–655
  82. Guo HB, Jiang AL, Ju TZ, Chen HL. Opposing changes in N-acetylglucosaminyltransferase-V and -III during the cell cycle and all-trans retinoic acid treatment of hepatocarcinoma cell line. Biochim Biophys Acta. 2000;1495:297–307
  83. Yao M, Zhou DP, Jiang SM, Wang QH, Zhou XD, Tang ZY, et al. Elevated activity of N-acetylglucosaminyltransferase V in human hepatocellular carcinoma. J Cancer Res Clin Oncol. 1998;124:27–30
  84. Ito Y, Miyoshi E, Sakon M, Takeda T, Noda K, Tsujimoto M, et al. Elevated expression of UDP-N-acetylglucosamine: alphamannoside beta1,6 N-acetylglucosaminyltransferase is an early event in hepatocarcinogenesis. Int J Cancer. 2001;91:631–637
  85. Bosques CJ, Raguram S, Sasisekharan R. The sweet side of biomarker discovery. Nat Biotechnol. 2006;24:1100–1101
  86. Raman R, Raguram S, Venkataraman G, Paulson JC, Sasisekharan R. Glycomics: an integrated systems approach to structure-function relationships of glycans. Nat Methods. 2005;2:817–824
  87. Gravel P, Walzer C, Aubry C, Balant LP, Yersin B, Hochstrasser DF, et al. New alterations of serum glycoproteins in alcoholic and cirrhotic patients revealed by high resolution two-dimensional gel electrophoresis. Biochem Biophys Res Commun. 1996;220:78–85
  88. Henry H, Froehlich F, Perret R, Tissot JD, Eilers-Messerli B, Lavanchy D, et al. Microheterogeneity of serum glycoproteins in patients with chronic alcohol abuse compared with carbohydrate-deficient glycoprotein syndrome type I. Clin Chem. 1999;45:1408–1413
  89. Laroy W, Contreras R, Callewaert N. Glycome mapping on DNA sequencing equipment. Nat Protoc. 2006;1:397–405
  90. Turnbull JE, Field RA. Emerging glycomics technologies. Nat Chem Biol. 2007;3:74–77
  91. Miyamoto S. Clinical applications of glycomic approaches for the detection of cancer and other diseases. Curr Opin Mol Ther. 2006;8:507–513
  92. Morelle W, Michalski JC. Glycomics and mass spectrometry. Curr Pharm Des. 2005;11:2615–2645
  93. Pilobello KT, Mahal LK. Deciphering the glycocode: the complexity and analytical challenge of glycomics. Curr Opin Chem Biol. 2007;11:300–305
  94. Callewaert N, Van Vlierberghe H, Van Hecke A, Laroy W, Delanghe J, Contreras R. Noninvasive diagnosis of liver cirrhosis using DNA sequencer-based total serum protein glycomics. Nat Med. 2004;10:429–434
  95. Liu XE, Desmyter L, Gao CF, Laroy W, Dewaele S, Vanhooren V, et al. N-glycomic changes in hepatocellular carcinoma patients with liver cirrhosis induced by hepatitis B virus. Hepatology. 2007;46:1426–1435
  96. Kam RKT, Poon TCW, Chan HLY, Wong N, Hui AY, Sung JJY. High-throughput quantitative profiling of serum N-glycome by MALDI-TOF mass spectrometry and N-glycomic fingerprint of liver fibrosis. Clin Chem. 2007;53:1254–1263
  97. Morelle W, Flahaut C, Michalski JC, Louvet A, Mathurin P, Klein A. Mass spectrometric approach for screening modifications of total serum N-glycome in human diseases: application to cirrhosis. Glycobiology. 2006;16:281–293
  98. Stahl B, Thurl S, Zeng JR, Karas M, Hillenkamp F, Steup M, et al. from human milk as revealed by matrix-assisted laser desorption/ionization mass spectrometry. Anal Biochem. 1994;223:218–226
  99. Harvey DJ. Quantitative aspects of the matrix-assisted laser-desorption mass-spectrometry of complex oligosaccharides. Rapid Commun Mass Spectrom. 1993;7:614–619
  100. Mehta AS, Long RE, Comunale MA, Wang M, Rodemich L, Krakover J, et al. Increased levels of galactose-deficient anti-Gal immunoglobulin G in the sera of hepatitis C virus-infected individuals with fibrosis and cirrhosis. J Virol. 2008;82:1259–1270

☆ The authors declare that they do not have anything to disclose regarding funding from industries or conflict of interest with respect to this manuscript.

PII: S0168-8278(08)00802-7

doi:10.1016/j.jhep.2008.12.010

Journal of Hepatology
Volume 50, Issue 3 , Pages 592-603, March 2009

Article Tools

* Image Usage & Resolution