[Frontiers in Bioscience E3, 1443-1455, June 1, 2011]
ST6GalNAc-I controls expression of sialyl-Tn antigen in gastrointestinal tissues
Nuno T. Marcos1, Eric P. Bennett2, Joana Gomes1, Ana Magalhaes1, Catarina Gomes1, Leonor David1,3, Imran Dar2, Charlotte Jeanneau2, Shawn DeFrees4, Dorrit Krustrup5, Lotte K. Vogel2, Elin H. Kure6, Joy Burchell7, Joyce Taylor-Papadimitriou7, Henrik Clausen2, Ulla Mandel2, Celso A. Reis1,3
1
TABLE OF CONTENTS
1. ABSTRACT
Sialyl-Tn is a simple mucin-type carbohydrate antigen aberrantly expressed in gastrointestinal adenocarcinomas and in the precursor lesion intestinal metaplasia. Sialyl-Tn tumour expression is an independent indicator of poor prognosis. We have previously shown in vitro that ST6GalNAc-I and ST6GalNAc-II sialyltransferases can synthesize sialyl-Tn. The aim of the present study was to establish whether ST6GalNAc-I is the major enzyme responsible for the expression of sialyl-Tn. We used a model of CHO-ldlD cells producing only MUC1-Tn glycoform and showed that ST6GalNAc-I is the key-enzyme leading to sialyl-Tn biosynthesis. We developed novel monoclonal antibodies specific for ST6GalNAc-I and evaluated its expression in gastrointestinal tissues. ST6GalNAc-I was detected in normal colon mucosa co-localized with O-acetylated sialyl-Tn. Expression was largely unaltered in colorectal adenocarcinomas. In contrast, we found that ST6GalNAc-I is weakly expressed in normal gastric mucosa, but over-expressed in intestinal metaplasia, co-localized with sialyl-Tn. In gastric carcinomas ST6GalNAc-I was also associated with sialyl-Tn, but with heterogeneous staining and partial co-localization. Our results showed ST6GalNAc-I as the major enzyme controlling the expression of cancer-associated sialyl-Tn antigen in gastrointestinal tissues.
2. INTRODUCTION
Altered glycosylation is a common feature of cancer cells. The sialyl-Tn (STn) antigen (Neu5Acα2-6GalNAcα-O-Ser/Thr), also known as CD175s, is a simple mucin-type carbohydrate antigen which has attracted much attention because it is highly expressed in most gastric (1-3), colorectal (4), ovarian (5), breast (6, 7) and pancreatic (8) carcinomas, but has limited expression in normal tissues. The expression of sialyl-Tn is an established indicator of poor prognosis in patients with gastric (9-11), colorectal (12, 13), and ovarian cancer (14) and contributes to the aggressive phenotype of carcinoma cells by altering their behaviour (15-18). Sialyl-Tn expression is also aberrantly expressed in precursor and early lesions of carcinomas of the gastrointestinal tract, such as intestinal metaplasia (IM) of the stomach (2), adenomatous polyps (12), chronic ulcerative colitis (13) and pancreatic intraepithelial neoplasias (8).
The synthesis of the mucin-type carbohydrate antigens depends upon expression of the enzymes that initiate O-glycosylation: UDP-GalNAc:polypeptide N-acetylgalactosaminyltransferases (GalNAc-transferases) (19). GalNAc-transferases control the first step in mucin glycosylation and controls the level of complexity of mucin glycans, i.e., the sites and density of O-glycan occupancy of the mucin tandem repeats (20, 21). Processing of GalNAcα-O-Ser/Thr, also known as the Tn antigen, by different glycosyltransferases is dependent on the cell type and tissue of origin. Monosaccharides are added sequentially to the GalNAc in the Golgi apparatus, forming a variety of O-glycan stuctures that include the Core1 (Galβ1-3GalNAcα-O-Thr/Ser), also known as T antigen, and the Core2 (GlcNAcβ1-6 (Galβ1-3)GalNAcα-O-Thr/Ser) glycans. These glycoforms may be glycosylated further forming larger and more complex structures. A common feature in carcinoma cells is the activation of an alternative glycosylation pathway that caps the GalNAcα-O-Thr/Ser structures with α2,6-sialic acid. Once sialylated, the resulting sialyl-Tn glycoform, Neu5Acα2-6GalNAcα-O-Ser/Thr, cannot be glycosylated further thereby preventing normal glycan elongation (22-24). The underlying mechanisms responsible for the induction of the sialyl-Tn glycoform in cancer cells are not fully understood. Two α2,6-sialyltransferases have been shown to be capable to sialylate the GalNAc residue O-linked on proteins: ST6GalNAc-I, and ST6GalNAc-II (25-27). However, over-expression of the ST6GalNAc-I glycosyltransferase in cells can override the normal O-glycosylation pathways leading to the formation of sialyl-Tn O-glycans as the dominant mucin phenotype (28-30). We have previously shown that both ST6GalNAc-I and ST6GalNAc-II can synthesize sialyl-Tn, in vitro; however, only ST6GalNAc-I could create large amounts of sialyl-Tn when over-expressed in a cancer cell line (28). Previous studies have also correlated the amount of ST6GalNAc-I mRNA with the amount of sialyl-Tn expression in breast cancer (30) and several cell lines (31), while the level of ST6GalNAc-II mRNA did not correlate with the expression of sialyl-Tn in gastric (unpublished results) and breast cancer cell lines (26) or in breast tumours (30). However, the role of each sialyltransferase, ST6GalNAc-I and II, in the synthesis of the cancer-associated sialyl-Tn structure in cells and pathological tissues remains unclear.
A novel monoclonal antibody (MAb) directed to ST6GalNAc-I was used to evaluate its expression in normal gastric and colonic mucosa, in intestinal metaplasia of the stomach, and in gastric and colorectal adenocarcinomas. The results show that over-expression of ST6GalNAc-I in intestinal metaplasia of the stomach co-localizes with the expression of sialyl-Tn. ST6GalNAc-I was not over-expressed in colon adenocarcinomas when compared to normal mucosa, suggesting that the acetylation status of sialyl-Tn may mask this antigen in normal colonic mucosa and exposed in adenocarcinomas. In contrast, in gastric carcinomas expression of sialyl-Tn was partially, but not exclusively, associated with ST6GalNAc-I. Our results demonstrate that ST6GalNAc-I is the major regulator of expression of cancer-associated sialyl-Tn O-glycosylation.
3. MATERIAL AND METHODS
3.1.Recombinant expression of soluble ST6GalNAc-I
An expression construct of the human ST6GalNAc-I were designed to encode an enzyme lacking the cytoplasmic and the transmembrane region. This soluble construct of ST6GalNAc-I encoding amino acid residues 43-601was prepared as described previously (28). The soluble construct of ST6GalNAc-I was expressed in Sf9 cells using the Baculo-virus expression system, and purified to near homogeneity (28).
3.2. Generation of monoclonal antibodie
The recombinant ST6GalNAc-I enzyme was used as immunogen. Balb/c mice were immunized with one subcutaneous injection of 10 �g native protein in Freunds complete adjuvant, followed by two injections with Freunds incomplete adjuvant, and finally an intravenous boost without adjuvant. Eye bleeds were taken 7 days after the third immunization, and the titre and specificity of antibodies were evaluated by immunocytology with baculo-virus infected Sf9 cells expressing recombinant human ST6GalNAc-I or irrelevant enzymes. Spleen cells from one immunized mouse were fused to NS-1 myeloma cells to produce hybridomas and the cloning procedure was performed as described previously (32, 33). Hybridomas were selected by initial screening on baculo-virus infected Sf9 cells expressing either one of the following human sialyltransferases: ST6GalNAc-I, ST6GalNAc-II or ST3Gal-I. Further characterization of the hybridoma clones was done on human cell lines expressing full-length coding ST6GalNAc-I or ST6GalNAc-II enzymes as previously described (28). Antibodies were also tested by SDS-PAGE Western blot analysis.
3.3. CHO ldlD-MUC1F cells stably transfected with ST6GalNAc-I or ST6GalNAc-II
CHO ldlD cells were kindly provided by M. Krieger (MIT, USA) (34). Stable CHOldlD cell line expressing full length FLAG-epitope tagged MUC1 (MUC1F) has been previously established (CHO ldlD-MUC1F) (35). Based on existing cDNA constructs (28, 30), full length human ST6GalNAc-I and ST6GalNAc-II-Myc constructs were produced by PCR, using PfuUltra™ polymerase (Stratagene). Regions encoding the membrane anchoring (ST6GalNAc-I; aa 1-293 and ST6GalNAc-II; aa 1-80) and catalytic domains (ST6GalNAc-I; aa 294-601 and ST6GalNAc-II; aa81-374) were generated using two sets of primers. For ST6GalNAc-I membrane anchor ST6I-1 (5�-GCG GAT CCA CCA TGA GGT CCT GCC TGT GGA GAT GCA GGC-3�)/ST6I-3 (5�-GAG CGC TCT AGA CCA CAG CGA CTT GGA GGC TTT GAT CTT C-3�) and catalytic domain ST6I-2 (5�-GAG CGC TCT AGA AAA CTC TTT CTG CCC AAC CTC ACT CTC-3�)/ST6I-4 (5�-GCG GCG GCC GCT CAG TTC TTG GCT TTG GCA GTT CCG GG-3�) were used. For ST6GalNAc-II membrane anchor ST6II-1 (5�-GCG GGA TCC ACC ATG GGG CTC CCG CGC GGG TCG TTC TTC-3�)/ST6II-3 (5�-GAG CGC TCT AGA GTG GGG GTG CCG CTG AAT GGC CAG-3�) and catalytic domain ST6II-2 (5�-GAG CGC TCT AGA GGC CTG TTC AAT CTC TCC ATT CCA GTG-3�)/ST6II-4 (5�-GGC GGC CGC GCG CTG GTA CAG CTG AAG GAT GCC GGC C-3�). Single XbaI sites were introduced in the membrane anchoring domain preceding the catalytic domain leading to L293S and Q294R mutations in ST6GalNAc-I and F80S mutation in ST6GalNAc-II-Myc, sites underlined in primers shown above. Dual insert (membrane anchor and catalytic domain) were cloned directionally into the BamHI and NotI sites of the pcDNA3-zeo (Invitrogen™). For ST6GalNAc-II the C-terminal Myc-tag construct was introduced by inserting a double stranded Myc-oligo into the Not-I site of ST6GalNAc-II-pCDNA3-zeo. All constructs were sequence verified using an ABI Avant sequenator. CHO ldlD-MUC1F cells were transfected with full length human ST6GalNAc-I or ST6GalNAc-II-Myc and grown in the presence of GalNAc.
3.4. Patients and tissues
Gastric carcinomas and gastric mucosa adjacent to carcinomas were obtained from individuals undergoing surgery at Hospital S. João, Medical Faculty (Porto, Portugal). Study was performed with the approval of the local HSJ ethical committee (from 05-09-2007). Specimens were frozen at - 80oC. Serial sections were cut and used for immunofluorescence. We evaluated 22 gastric mucosas with normal morphology, 24 IM lesions adjacent to the carcinoma cases that were classified as complete IM (n =14) and incomplete IM (n =10) according to the pattern of mucin expression (described below) and 31 cases of gastric carcinoma. Gastric carcinomas were classified according to Laurén's classification (36). Colorectal adenocarcinoma cases were obtained from the KAM cohort based on the screening group of the Norwegian Colorectal Cancer Prevention study in the county of Telemark, Norway (37) and a series of colorectal cancer cases recruited to the KAM cohort from routine clinical work at Telemark Hospital, Skien and Ulleval University Hospital, Oslo. The KAM study is approved by the Regional Ethical Committee and the Norwegian Data Inspectorate. We evaluated 15 colorectal adenocarcinomas. A sample of control tissue was taken from the surgically removed tissue close to the adenocarcinoma or as far away from the tumor as possible. The histology of adjacent normal tissue was examined independently by two specialist histopathologists, and found to be normal. The histology of the adenocarcinomas was also examined independently by two specialist histopathologists in order to determine the tumor stage. All colon cancer cases were histomorphologically classified as moderately differentiated.
3.5. Immunofluorescence
Details of MAbs used are shown in Table 1. Cells were fixed in ice-cold acetone for 5 min and kept at -20�C before staining. Tissue slides were preserved at -80�C and fixed in 4% paraformaldehyde for 15 min at RT, prior to procedure.
Double staining was used for co-localization of ST6GalNAc-I (MAb 2C3) and sialyl-Tn (MAb TKH2 and HB-STn) and ST6GalNAc-I and MUC2 (MAb PMH1). MUC5AC and MUC2 staining was performed for the characterization of IM types (44). Deacetylation (saponification) of selected samples was accomplished by treating slides immediately after fixation with 0.1N NaOH for 20 min at RT prior to performing the immunofluorescence (45). Samples designed for MUC2 detection with PMH1 were pre-treated with neuraminidase as previously described (40).
Samples were washed twice in PBS and incubated for 20 min with rabbit non-immune serum (DAKO) diluted 1:5 in PBS/10%BSA. Samples were incubated overnight at 4�C with the monoclonal antibodies 2C3 or CLH2 (Table 1) diluted in PBS/5%BSA. Sections were washed in PBS and incubated with FITC-conjugated rabbit anti-mouse immunoglobulin (DAKO) diluted 1:70 in PBS/5%BSA, for 45 min. Then the samples were washed in PBS and blocked with non-immune goat serum (DAKO) diluted 1:5 in PBS/10%BSA for 20 min. Sections were incubated with the monoclonal antibody PMH1 or TKH2 (Table 1) diluted in PBS/5%BSA overnight at 4�C. Sections were washed with PBS and incubated for 30 min with Texas-Red-conjugated goat anti-mouse IgM (in the case of PMH1) or Texas-Red-conjugated goat anti-mouse IgG1-specific (in the case of TKH2) (Jackson Immunoresearch) diluted 1:70 in PBS/5%BSA and DAPI diluted 1:100, for 30 min in the dark. Samples were washed in PBS and mounted in Vectashield (Vector Laboratories, Inc).
Results are based on assays analyzed by two independent observers. Statistical analysis was performed using the chi-square test with Yates correction using Statview 4.01 software.
4. RESULTS
4.1. Monoclonal antibodies to ST6GalNAc-I
Two monoclonal antibodies 2C3 (IgG2a) and 1C9 (IgG1) were selected for their specific reactivity with ST6GalNAc-I. MAb 2C3 reacted with Sf9 cells expressing ST6GalNAc-I but no reactivity was observed with Sf9 cells expressing ST6GalNAc-II. MAb 2C3 was also found to react with the human cell line K562, which expresses high levels of ST6GalNAc-I transcripts and sialyl-Tn antigen (31) (Figure 1, Panel II G, J). MAb 2C3 did not show reactivity with the denatured form of ST6GalNAc-I when analyzed by SDS-PAGE Western blot (not shown). On the other hand MAb 1C9 was selected for their specific of the reactivity with ST6GalNAc-I in reduced SDS-PAGE Western blot analysis (Figure 1, Panel I). These results are in agreement with our previous findings that MAbs to glycosyltransferases tend to react either with the native conformation of the protein or with the denatured protein (32, 46). We have generally been unable to produce MAbs that react with both the native protein in unfixed cells or tissues and by SDS-PAGE Western blot analysis.
Further analysis of MAb 2C3 showed some immunoreactivity in immunocytology with the human tumour cell lines MKN45 (31) and Hela, which express low levels of endogenous ST6GalNAc-I. However, we observed strong immunoreactivity with these cells when transfected with full coding human ST6GalNAc-I (Figure 1, Panel II). The staining pattern observed in these cells was a distinct perinuclear punctuate staining typical of Golgi vesicles. MAb 2C3 furthermore produced the same staining pattern in glandular epithelial cells of salivary glands (Figure 1, Panel III), which are known to express ST6GalNAc-I mRNA.
4.2. Glycophenotype of CHO ldlD-MUC1F cells stably transfected with ST6GalNAc-I and ST6GalNAc-II
Glycosylation-defective CHO ldlD cells lack the epimerase that transforms UDP-Glc and -GlcNAc to UDP-Gal and -GalNAc, respectively (34). This defect blocks O-linked glycosylation of proteins. However, the addition of exogenous Gal or/and GalNAc to the media of ldlD cells overcame this defect and corrected the abnormal glycosylation phenotype (34). Addition of only GalNAc to the cells produced a truncated mucin limited to Tn. No sialyl-Tn was observed. This indicated that CHO ldlD cells have no competing glycosylation pathway after the Tn glycoform was produced. Interestingly, both the parental cell line CHO K1 and the CHO ldlD cell line grown in the presence of Gal and GalNAc are known to produce mono- and to a lesser extent di-sialylated T-antigen glycoforms (47), and indicates that an ST6GalNAc-transferase activity must be present. Stably transfectanted cells (CHO ldlD-MUC1F) of a full coding MUC1 have been produced and shown to express MUC1 on the cell membrane (35, 42).
CHO ldlD-MUC1F cells cultured in the presence of GalNAc expressed the Tn antigen (GalNAc-O-Ser/Thr) (Figure 2). There was no sialyl-Tn found indicating that the endogenous sialyltransferases were not capable of producing sialyl-Tn. A lack of the presence of the sialyl-T antigen was confirmed by treating the cells with neuraminidase and then staining the cells using controlled conditions with an anti-T antibody. Stable transfection of ST6GalNAc-I into these cells resulted in production of the sialyl-Tn glycoform as detected by B72.3 (Figure 2) and other sialyl-Tn Mabs (not shown). In contrast, stable transfection with ST6GalNAc-II did not result in substantial production of sialyl-Tn, although a few faint cells (<25%) could be visualized (Figure 2). Interestingly, while the cytolocalization of ST6GalNAc-I detected by MAb 2C3 was clearly supranuclear and Golgi-like, the staining of myc-tagged ST6GalNAc-II was localized throughout the cell with no apparent Golgi localization. While tagged constructs may mislocalize recombinant proteins, previously we have shown co-localisation of native and tagged constructs of several other glycosyltransferases. A 6X His tagged ST6GalNAc-II construct has also been expressed as a secreted functional enzyme in insect cells (unpublished). We are currently trying to develop Mabs to ST6GalNAc-II and characterize the cytolocalization in more detail.
4.3. Expression of ST6GalNAc-I and sialyl-Tn in normal gastric mucosa
Gastric mucosa with normal morphology showed a complete absence of expression of sialyl-Tn antigen, as expected. Immunodetection of ST6GalNAc-I was observed in all normal gastric mucosa (Table 2) and was characterised by a weak immunofluorescence signal limited to the perinuclear area of cells from the foveolar epithelium, corresponding to a Golgi-like staining (Figure 3A-G). The process of deacetylation did not alter the expression of sialyl-Tn in normal gastric mucosa, which remained negative after treatment (data not shown).
4.4. Expression of ST6GalNAc-I and sialyl-Tn in intestinal metaplasia
The 24 foci of IM, 14 of complete IM and 10 of incomplete IM, were classified according to the pattern of expression of mucins (44). Cases with metaplastic areas which co-expressed gastric mucin MUC5AC together with intestinal mucin MUC2 were classified as incomplete IM, whereas cases with MUC2 but lacking MUC5AC were classified as complete IM.
Expression of sialyl-Tn antigen was observed in the mucin vacuoles of goblet cells of all cases of complete and incomplete IM (Figure 3C and 3D, respectively). All IM cases showed a strong expression of ST6GalNAc-I (Table 2, Figure 3A-D), clearly contrasting with the faint staining of normal mucosa areas. The staining was also restricted to goblet cells and the pattern was perinuclear (Golgi-like). Therefore, both in complete and incomplete IM, ST6GalNAc-I showed increased staining and co-localized with MUC2 (Figure 3A and 3B) and sialyl-Tn (Figure 3C and 3D).
4.5. Expression of ST6GalNAc-I and sialyl-Tn in gastric carcinoma
Expression of ST6GalNAc-I was observed in 24/31 (77%) of cases of gastric carcinoma (Table 2; Figure 3 E-G).The expression levels of enzyme (intensity) could not be assessed due to the heterogenous intensity observed within each gastric carcinoma case. However, differences were observed between the pattern of expression in gastric carcinoma and normal mucosa. A punctuated Golgi-like staining restricted to foveolar epithelium was seen in normal mucosa, opposite to the widespread perinuclear and diffuse cytoplasmic staining observed in gastric carcinoma. Due to this fact and to the focal and patchy nature of sialyl-Tn and ST6GalNAc-I expression in carcinomas, a semi-quantitative scale of percentage of positive cells was used to achieve a more informative classification.
We observed a significant association between ST6GalNAc-I expression and sialyl-Tn expression, but not with the histopathologic type of gastric carcinoma (Table 2). In the majority of carcinoma cases where co-expression of sialyl-Tn and ST6GalNAc-I was observed, both sialyl-Tn and ST6GalNAc-I were largely expressed in the same cells. However, in 8 cases the score for sialyl-Tn expression exceeded the score observed for ST6GalNAc-I. In addition, there were 4 cases with expression of sialyl-Tn where we did not observe ST6GalNAc-I (Table 2).
4.6. Expression of ST6GalNAc-I and sialyl-Tn in normal colorectal mucosa
In all 15 cases, normal appearing colorectal mucosa showed a strong perinuclear, Golgi-like, staining pattern of ST6GalNAc-I throughout all compartments of the colonic crypts (Figure 4B). Before deacetylation with sodium hydroxide, sialyl-Tn was only expressed sporadically in lower compartments of the crypts in about half of the cases (Figure 4C). After deacetylation, all cases expressed sialyl-Tn intracellularly in lower compartments; however, in some cases the expression of sialyl-Tn was also observed in goblet cells throughout the crypts (Figure 4D).
4.7. Expression of ST6GalNAc-I and sialyl-Tn in colorectal adenocarcinoma
ST6GalNAc-I was observed in 15 cases of colorectal adenocarcinoma (Table 2, Figure 4E and 4F). Expression of ST6GalNAc-I was observed in 50% of cases and in >50% of the tumour cells. Generally the staining intensity was weaker in tumour cells than in the normal cells. Exceptionally strong staining of both ST6GalNAc-I and sialyl-Tn was observed in transitional tissue, a tissue which is immediately adjacent to tumour edge but with no histological features of malignancy. Transitional tissue was observed in 4 of the colon biopsies (not shown).
Sialyl-Tn was expressed in the same cases as ST6GalNAc-I and in the same areas. Deacetylation of the tissue enhanced the staining of sialyl-Tn in most of the cases, but this effect was far less pronounced than in normal colon samples. Sialyl-Tn was observed in the apical cell membranes, cytoplasm, and luminal contents of the colorectal adenocarcinomas. There was no correlation of Duke's clinical staging with the expression of ST6GalNAc-I; however, a strong association with sialyl-Tn expression was observed (Table 2). We could not detect tumour cells that expressed sialyl-Tn and does not express ST6GalNAc-I. On the other hand we could occasionally detect few tumour cells that expressed ST6GalNAc-I and lacking expression of sialyl-Tn.
5. DISCUSSION
Sialyl-Tn antigen is a simple mucin-type carbohydrate antigen whose aberrant expression is common in several human carcinomas and precursor lesions of cancer (1-8, 12, 13). Sialyl-Tn antigen is the product of an abnormal glycosylation pathway, corresponding to the early α2,6-sialylation of GalNAcα-O-Ser/Thr. The molecular mechanism leading to the activation of this pathway in carcinomas remains to be clarified.
We have previously shown in vitro that ST6GalNAc-I and ST6GalNAc-II can synthesize sialyl-Tn, but only ST6GalNAc-I can induce high amounts of sialyl-Tn when over-expressed in cancer cell lines (28). In most cells multiple glycosylation pathways compete for the same substrate and hence concordance between in vitro activity and in vivo function may not be evident. We used the CHO ldlD cell system to determine if the function of ST6GalNAc-II was inhibited by other glycosyltransferases competing for the same substrate. Indeed, ST6GalNAc-II did not produce substantial amounts of sialyl-Tn in a cell capable of only producing the enzymes substrate, the Tn glycoform as part of the abundant MUC1 mucin. In contrast, ST6GalNAc-I produced sialyl-Tn under the same conditions. Therefore, it may be concluded that the low activity of ST6GalNAc-II at producing sialyl-Tn in cells was not due to competing glycosylation. Different molecular mechanisms leading to sialyl-Tn biosynthesis have been previously described. In the LS174T colon carcinoma cell line model, Brockhausen and co-workers demonstrated that the expression of sialyl-Tn in a subset of cells is associated with a lack of core1 β3Gal-transferase activity, and not with α2,6Sialyl-transferase activity (22). Whereas in the LMCR colon carcinoma cell line, they showed that the reverse was true, i.e., sialyl-Tn expressing cells had increased α2,6Sialyl-transferase activity (23). Therefore, using two cell lines from colonic origin, it was shown two different mechanisms leading to sialyl-Tn synthesis: one due to ST6GalNAc-transferase activity overexpression, and another due to the inability of synthesizing the core1 glycoform (T-antigen), as a result of inactivation of the molecular chaperone, Cosmc (48). The Cosmc chaperone is required for expression of an active β3Gal-transferase and, subsequently, core1 synthesis and proper O-glycosylation (49). Cosmc mutations have been found in cancer cell lines and primary cervical cancers where its inactivation has been associated with sialyl-Tn expression (48, 49). Cosmc mutations in gastric carcinomas remain unidentified. In fact, the action of competing glycosyltransferases, namely core 1 β3Gal-transferase that uses the same substrate as ST6GalNAc-I (GalNAc-O-Ser/Thr), and core 3 (β3GnT6), may be a mechanism interfering with sialyl-Tn biosynthesis. Another factor may be the sub cellular localization and Golgi compartment localization of these glycosyltransferases. Previous studies have shown that sialyl-Tn synthesis in colorectal adenocarcinoma was localized throughout the Golgi apparatus, including the early compartments (50). This result was further supported by another study showing that ST6GalNAc-I was found in all Golgi sections in carcinoma cells (30). In fact, the biosynthetic pathway leading to the production of disialyl-T structures (27-29) involves the assembly of the monosialyl-T glycoform (NeuAc2-3Galβ1-3GalNAcα-O-Thr/Ser) prior to the action of ST6GalNAc-transferase. This sequence of glycan addition to the T-antigen is supported by the topological location of ST6GalNAc-I in the Golgi compartment; whereas, core 1 β3Gal-transferase and a 2,3sialyltransferases, are located in earlier cellular compartments. Studies showing that transfection of cells with ST6GalNAc-I can result in overriding of the natural O-glycosylation pathway with resulting truncated sialyl-Tn O-glycans therefore suggest that it is the overexpression and likely the resulting altered topology of the enzyme that drives the premature a 2,6sialylation of Tn before core 1 extension (30). This interpretation would be in agreement with the recent finding that ST6GalNAc-II, while capable in vitro of using Tn substrates, is incapable of inducing substantial amounts of sialyl-Tn synthesis in cells with abundant Tn substrates and without competitive alternative glycosylation pathways (28). Presumably this is due to an entirely different subcellular topology (Figure 2).
In the present study, we have produced and characterized two novel MAbs that specifically react with either the native or the denatured ST6GalNAc-I enzyme protein and have utility in immunohistology applications and SDS-PAGE Western blot. Although the homologous enzymes ST6GalNAc-I and ST6GalNAc-II share a degree of sequence similarity we did not identify clones with cross-reactivity between the enzymes. This result is similar to our past experience with the immunogenicity of homologous glycosyltransferases (32). In agreement with previous studies we have observed the absence of sialyl-Tn expression in normal gastric mucosa. Nevertheless, weak immunoreactivity for the ST6GalNAc-I enzyme was observed in the foveolar epithelium of gastric mucosa. The ST6GalNAc-I staining was restricted to the perinuclear region suggesting Golgi localization (Table 2; Figure 3). A recent study has shown that ST6GalNAc-I mRNA is weakly expressed in various normal tissues despite that these tissues generally do not express the sialyl-Tn antigen (30). In addition, we have previously shown that gastric cell lines devoid of sialyl-Tn, still express basal levels of ST6GalNAc-I transcripts (31). It is therefore clear that expression of low levels of ST6GalNAc-I mRNA as well as basal enzyme levels, as shown here, may not be sufficient per se to induce sialyl-Tn expression and further supports the importance of the levels of expression of the ST6GalNAc-I enzyme as well as the Golgi compartment localization.
The expression of sialyl-Tn in colonic mucosa appears to represent a different scenario. ST6GalNAc-I is expressed in normal colon both at the mRNA and protein levels. Nevertheless, sialyl-Tn is not normally detectable at the surface colon cells, being restricted to perinuclear staining at the base of the crypt in few cases (Figure 4). In normal colonic epithelial cells, sialic acid residues are often modified by O-acetyl groups, thereby precluding antibody recognition of the sialyl-Tn antigen. Deacetylation leads to sialyl-Tn exposure and consequent detection, and MAb staining largely increases after this procedure, as can be seen in Figure 4. Yet, this mechanism is not true for normal gastric mucosa. We observed no sialyl-Tn expression after deacetylation of normal gastric mucosa (data not shown). O-acetylation in the human stomach is very rare and limited to pathological conditions (51) so the absence of sialyl-Tn detection in normal gastric mucosa cannot be explained by such mechanism.
In the human stomach, Intestinal Metaplasia is a pre-malignant lesion characterized by a global trans-differentiation of gastric epithelium into intestinal epithelium and is associated with an increased risk for gastric carcinoma development. Intestinal metaplasia expresses several intestinal markers that are foreign to normal gastric epithelial cells, such as MUC2 and sialyl-Tn, which co-localize at the mucinous vacuoles of goblet cells (44, 52). We observed that ST6GalNAc-I was expressed in all intestinal metaplasia cases (24/24), and this expression co-localizes both with MUC2 and sialyl-Tn (Figure 3). The ST6GalNAc-I staining was in the perinuclear area of goblet cells observed in all metaplastic glands, regardless of the histological sub-type, and was noticeably more intense than in normal gastric mucosa (Figure 3). It is therefore clear that an overexpression of ST6GalNAc-I exists in intestinal metaplasia when compared to normal gastric mucosa, and supports the hypothesis that overexpression of ST6GalNAc-I is a mechanism leading to sialyl-Tn antigen expression in intestinal metaplasia.
In the present study, the abundant sialyl-Tn expression observed in intestinal metaplasia without deacetylation indicated that O-acetylation did not interfere with sialyl-Tn detection in intestinal metaplasia as it does on normal colon (51). Nevertheless, our results showed that the intestinal metaplasia model recapitulates normal colon as ST6GalNAc-I and sialyl-Tn are co-expressed in the same cells.
In gastric carcinoma, sialyl-Tn expression was detected in 87% of the cases similarly to previously published series (2), and ST6GalNAc-I expression was found on 77% of gastric carcinoma cases. ST6GalNAc-I showed a heterogeneous staining, either clustered in focal areas or widely dispersed, and showed variable intensity of staining within each case, thus justifying for a different classification criteria than in intestinal metaplasia cases, based on the percentage of positive cells in the tumour. Using this classification, ST6GalNAc-I expression was significantly associated with sialyl-Tn expression, independent of the histologic type of the tumour. Expression of ST6GalNAc-I in >50% of tumour cells was observed in 7/31 cases, all of which also expressed sialyl-Tn in >50% of cells. This equals saying that all gastric carcinoma cases that were high enzyme expressers were also so for sialyl-Tn. However, the inverse was not true, i.e. there were several cases that expressed sialyl-Tn in >50% of their tumour cells but have <50% cells with ST6GalNAc-I (8/18) or are even negative (3/18). This means that, although the majority of gastric carcinoma cases expressed sialyl-Tn and ST6GalNAc-I in the same cells, there were areas/cases where sialyl-Tn was expressed without ST6GalNAc-I detection. Therefore, although there was an association between sialyl-Tn and ST6GalNAc-I at the case level, the expression of both did not always co-localize at the cellular level. This apparent discrepancy at the cellular level may be explained by different hypothesis. One possibility is that ST6GalNAc-I may be present but technically undetectable due to low levels of expression. Supporting this hypothesis, Sewell and co-workers have found that, in breast carcinomas, ST6GalNAc-I expression correlated at the mRNA level with sialyl-Tn expression. Nevertheless, cases that were weak to moderate positives for sialyl-Tn (50% of cases) did not express any detectable ST6GalNAc-I RNA by Northern Blot analysis, but were positive by RT-PCR (30). These results suggest that enzyme detection at the protein level may only occur in cases with abundant sialyl-Tn expression. Another possibility is that ST6GalNAc-I may not be constitutively expressed in the tumour cells, resulting in a temporary presence of the enzyme and longer expression of sialyl-Tn bearing glycoprotein. Finally, we could not exclude the possibility that ST6GalNAc-II, which was demonstrated to be enzymatically capable of synthesizing sialyl-Tn in this study and in a previous one (28), may co-adjuvate ST6GalNAc-I in tumour cells.
A clear scenario was observed in colon adenocarcinomas where a major overlap of enzyme and glycan staining was observed. Colon adenocarcinomas showed a significant association between ST6GalNAc-I expression and sialyl-Tn expression that is independent of the clinical stage (Table 2). The presence of both enzyme and product was similar to normal colon mucosa with the difference that for a complete overlap, deacetylation procedures must be applied. In line with this, a previous study has shown that normal and cancerous colon tissues have similar ST6GalNAc enzymatic activity (53). Interestingly enough, transitional tissues have been shown to display increased enzymatic activity when compared to normal mucosa and cancer (54). We found identical results reflected by the exceptionally strong staining of ST6GalNAc-I in our transitional colon tissues.
In conclusion our results demonstrate that the novel monoclonal antibody 2C3 is highly specific for the ST6GalNAc-I enzyme and detects this protein in cells and tissues expressing it. We also demonstrate that ST6GaNAc-I expression is associated with sialyl-Tn expression in gastrointestinal tissues supporting the ST6GalNAc-I as the major regulator of expression of cancer-associated sialyl-Tn antigen on O-linked carbohydrate chains.
6. ACKNOWLEDGMENTS
We thank Paula Silva for technical assistance. This work was supported by Fundação para a Ciência e a Tecnologia - FCT (PIC/IC/82716/2007), Association for International Cancer Research (Grant 05-088), and European Union - Seventh Framework Program - Health (Grant agreement n�: 201381). NTM acknowledges GABBA and FCT (Ref. SFRH/BD/11764/2003). JG (SFRH/BD/40563/2007) and AM (SFRH/BD/36339/2007) acknowledge FCT. Work at CCG funded by the University of Copenhagen's Centre of Excellence program, the Danish Research Council, and the Carlsberg Foundation. The Norwegian Colorectal Cancer Prevention (NORCCAP) study and the Eastern Norway Regional Health Authority supported the KAM study.
7. REFERENCES
1. SE Baldus, FG Hanisch: Biochemistry and pathological importance of mucin-associated antigens in gastrointestinal neoplasia. Adv Cancer Res 79, 201-248 (2000)
doi:10.1016/S0065-230X(00)79007-5
2. L David, JM Nesland, H Clausen, F Carneiro, M Sobrinho-Simoes: Simple mucin-type carbohydrate antigens (Tn, sialosyl-Tn and T) in gastric mucosa, carcinomas and metastases. APMIS Suppl 27, 162-172 (1992)
PMid:1520525
3. M Victorzon, S Nordling, O Nilsson, PJ Roberts, C Haglund: Sialyl Tn antigen is an independent predictor of outcome in patients with gastric cancer. Int J Cancer 65, 295-300 (1996)
doi:10.1002/(SICI)1097-0215(19960126)65:3<295::AID-IJC3>3.0.CO;2-V
4. SH Itzkowitz, EJ Bloom, WA Kokal, G Modin, S Hakomori, YS Kim: Sialosyl-Tn. A novel mucin antigen associated with prognosis in colorectal cancer patients. Cancer 66, 1960-1966 (1990)
doi:10.1002/1097-0142(19901101)66:9<1960::AID-CNCR2820660919>3.0.CO;2-X
5. H Kobayashi, T Terao, Y Kawashima: Serum sialyl Tn as an independent predictor of poor prognosis in patients with epithelial ovarian cancer. J Clin Oncol 10, 95-101 (1992)
PMid:1727929
6. M Leivonen, S Nordling, J Lundin, K von Boguslawski, C Haglund: STn and prognosis in breast cancer. Oncology 61, 299-305 (2001)
doi:10.1159/000055337
PMid:11721177
7. S Yonezawa, T Tachikawa, S Shin, E Sato: Sialosyl-Tn antigen. Its distribution in normal human tissues and expression in adenocarcinomas. Am J Clin Pathol 98, 167-174 (1992)
PMid:1510031
8. GE Kim, HI Bae, HU Park, SF Kuan, SC Crawley, JJ Ho, YS Kim: Aberrant expression of MUC5AC and MUC6 gastric mucins and sialyl Tn antigen in intraepithelial neoplasms of the pancreas. Gastroenterology 123, 1052-1060 (2002)
doi:10.1053/gast.2002.36018
PMid:12360467
9. T Nakagoe, T Sawai, T Tsuji, MA Jibiki, A Nanashima, H Yamaguchi, T Yasutake, H Ayabe, K Arisawa, H Ishikawa: Predictive factors for preoperative serum levels of sialy Lewis(x), sialyl Lewis(a) and sialyl Tn antigens in gastric cancer patients. Anticancer Res 22, 451-458 (2002)
PMid:12017332
10. JL Werther, S Rivera-MacMurray, H Bruckner, M Tatematsu, SH Itzkowitz: Mucin-associated sialosyl-Tn antigen expression in gastric cancer correlates with an adverse outcome. Br J Cancer 69, 613-616 (1994)
doi:10.1038/bjc.1994.114
PMid:8123499 PMCid:1968851
11. JL Werther, M Tatematsu, R Klein, M Kurihara, K Kumagai, P Llorens, J Guidugli Neto, C Bodian, D Pertsemlidis, T Yamachika, T Kitou, S Itzkowitz: Sialosyl-Tn antigen as a marker of gastric cancer progression: an international study. Int J Cancer 69, 193-199 (1996)
doi:10.1002/(SICI)1097-0215(19960621)69:3<193::AID-IJC8>3.0.CO;2-V
12. SH Itzkowitz, EJ Bloom, TS Lau, YS Kim: Mucin associated Tn and sialosyl-Tn antigen expression in colorectal polyps. Gut 33, 518-523 (1992)
13. P Karlen, E Young, O Brostrom, R Lofberg, B Tribukait, K Ost, C Bodian, S Itzkowitz: Sialyl-Tn antigen as a marker of colon cancer risk in ulcerative colitis: relation to dysplasia and DNA aneuploidy. Gastroenterology 115, 1395-1404 (1998)
doi:10.1016/S0016-5085(98)70018-6
PMid:10871727
14. LA Holmberg, DV Oparin, T Gooley, K Lilleby, W Bensinger, MA Reddish, GD MacLean, BM Longenecker, BM Sandmaier: Clinical outcome of breast and ovarian cancer patients treated with high-dose chemotherapy, autologous stem cell rescue and THERATOPE STn-KLH cancer vaccine. Bone Marrow Transplant 25, 1233-1241 (2000)
doi:10.1038/sj.bmt.1702430
PMid:15383613
15. M Clement, J Rocher, G Loirand, J Le Pendu: Expression of sialyl-Tn epitopes on beta1 integrin alters epithelial cell phenotype, proliferation and haptotaxis. J Cell Sci 117, 5059-5069 (2004)
doi:10.1242/jcs.01350
PMid:1582597 PMCid:1374070
16. S Julien, C Lagadec, MA Krzewinski-Recchi, G Courtand, X Le Bourhis, Delannoy P: Stable expression of sialyl-Tn antigen in T47-D cells induces a decrease of cell adhesion and an increase of cell migration. Breast Cancer Res Treat 90, 77-84 (2005)
doi:10.1007/s10549-004-3137-3
PMid:15770530
17. S Julien, E Adriaenssens, K Ottenberg, A Furlan, G Courtand, AS Vercoutter-Edouart, FG Hanisch, P Delannoy, X Le Bourhis: ST6GalNAc I expression in MDA-MB-231 breast cancer cells greatly modifies their O-glycosylation pattern and enhances their tumourigenicity. Glycobiology 16, 54-64. (2006)
doi:10.1093/glycob/cwj033
PMid:16135558
18. S Pinho, NT Marcos, B Ferreira, AS Carvalho, MJ Oliveira, F Santos-Silva, A Harduin-Lepers, CA Reis: Biological significance of cancer-associated sialyl-Tn antigen: modulation of malignant phenotype in gastric carcinoma cells. Cancer Lett 249, 157-170 (2007)
doi:10.1016/j.canlet.2006.08.010
PMid:16965854
19. H Clausen, EP Bennett: A family of UDP-GalNAc: polypeptide N-acetylgalactosaminyl-transferases control the initiation of mucin-type O-linked glycosylation. Glycobiology 6, 635-646 (1996)
doi:10.1093/glycob/6.6.635
PMid:8922959
20. H Hassan, CA Reis, EP Bennett, E Mirgorodskaya, P Roepstorff, MA Hollingsworth, J Burchell, J Taylor-Papadimitriou, H Clausen: The lectin domain of UDP-N-acetyl-D-galactosamine: polypeptide N-acetylgalactosaminyltransferase-T4 directs its glycopeptide specificities. J Biol Chem 275, 38197-38205 (2000)
doi:10.1074/jbc.M005783200
PMid:10984485
21. KG Ten Hagen, TA Fritz, LA Tabak: All in the family: the UDP-GalNAc:polypeptide N-acetylgalactosaminyltransferases. Glycobiology 13:1R-16R (2003)
doi:10.1093/glycob/cwg007
PMid:12634319
22. I Brockhausen, J Yang, N Dickinson, S Ogata, SH Itzkowitz: Enzymatic basis for sialyl-Tn expression in human colon cancer cells. Glycoconj J 15, 595-603. (1998)
doi:10.1023/A:1006967910803
PMid:9881766
23. I Brockhausen, J Yang, M Lehotay, S Ogata, S Itzkowitz: Pathways of mucin O-glycosylation in normal and malignant rat colonic epithelial cells reveal a mechanism for cancer-associated Sialyl-Tn antigen expression. Biol Chem 382, 219-232 (2001)
doi:10.1515/BC.2001.029
PMid:11308020
24. M Dalziel, C Whitehouse, I McFarlane, I Brockhausen, S Gschmeissner, T Schwientek, H Clausen, JM Burchell, J Taylor-Papadimitriou: 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 (2001)
doi:10.1074/jbc.M006523200
PMid:11118434
25. A Harduin-Lepers, R Mollicone, P Delannoy, R Oriol. The animal sialyltransferases and sialyltransferase-related genes: a phylogenetic approach. Glycobiology 15: 805-817 (2005)
doi:10.1093/glycob/cwi063
PMid:15843597
26. B Samyn-Petit, MA Krzewinski-Recchi, WF Steelant, P Delannoy, A Harduin-Lepers. Molecular cloning and functional expression of human ST6GalNAc II. Molecular expression in various human cultured cells. Biochim Biophys Acta 1474, 201-211 (2000)
PMid:10742600
27. Y Ikehara, N Kojima, N Kurosawa, T Kudo, M Kono, S Nishihara, S Issiki, K Morozumi, S Itzkowitz, T Tsuda, SI Nishimura, S Tsuji, H Narimatsu: Cloning and expression of a human gene encoding an N-acetylgalactosamine-alpha2,6-sialyltransferase (ST6GalNAc I): a candidate for synthesis of cancer-associated sialyl-Tn antigens. Glycobiology 9, 1213-1224 (1999)
doi:10.1093/glycob/9.11.1213
PMid:10536037
28. NT Marcos, S Pinho, C Grandela, A Cruz, B Samyn-Petit, A Harduin-Lepers, R Almeida, F Silva, V Morais, J Costa, J Kihlberg, H Clausen, CA Reis: Role of the human ST6GalNAc-I and ST6GalNAc-II in the synthesis of the cancer-associated sialyl-Tn antigen. Cancer Res 64, 7050-7057 (2004)
doi:10.1158/0008-5472.CAN-04-1921
PMid:15466199
29. S Julien, MA Krzewinski-Recchi, A Harduin-Lepers, V Gouyer, G Huet, X Le Bourhis, P Delannoy: Expression of sialyl-Tn antigen in breast cancer cells transfected with the human CMP-Neu5Ac: GalNAc alpha2,6-sialyltransferase (ST6GalNac I) cDNA. Glycoconj J 18, 883-893 (2001)
doi:10.1023/A:1022200525695
PMid:12820722
30. R Sewell, M Backstrom, M Dalziel, S Gschmeissner, H Karlsson, T Noll, J Gätgens, H Clausen, GC Hansson, J Burchell, J Taylor-Papadimitriou: The ST6GalNAc-I sialyltransferase localizes throughout the Golgi and is responsible for the synthesis of the tumor-associated sialyl-Tn O-glycan in human breast cancer. J Biol Chem 281, 3586-3594. (2006)
doi:10.1074/jbc.M511826200
PMid:16319059
31. NT Marcos, A Cruz, F Silva, R Almeida, L David, U Mandel, H Clausen, S Von Mensdorff-Pouilly, CA Reis: Polypeptide GalNAc-transferases, ST6GalNAc-transferase I, and ST3Gal-transferase I expression in gastric carcinoma cell lines. J Histochem Cytochem 51, 761-771 (2003)
PMid:12754287
32. U Mandel, H Hassan, MH Therkildsen, J Rygaard, MH Jakobsen, BR Juhl, E Dabelsteen, H Clausen: Expression of polypeptide GalNAc-transferases in stratified epithelia and squamous cell carcinomas: immunohistological evaluation using monoclonal antibodies to three members of the GalNAc-transferase family. Glycobiology 9, 43-52 (1999)
doi:10.1093/glycob/9.1.43
PMid:9884405
33. T White, U Mandel, TF Orntoft, E Dabelsteen, J Karkov, M Kubeja, S Hakomori, H Clausen: Murine monoclonal antibodies directed to the human histo-blood group A transferase (UDP-GalNAc:Fuc alpha 1----2Gal alpha 1----3-N-acetylgalactosaminyltransferase) and the presence therein of N-linked histo-blood group A determinant. Biochemistry 29, 2740-2747 (1990)
doi:10.1021/bi00463a017
PMid:2112026
34. DM Kingsley, KF Kozarsky, L Hobbie, M Krieger: Reversible defects in O-linked glycosylation and LDL receptor expression in a UDP-Gal/UDP-GalNAc 4-epimerase deficient mutant. Cell 44, 749-759 (1986)
PMid:18625714 PMCid:2556014
35. PK Singh, ME Behrens, JP Eggers, RL Cerny, JM Bailey, K Shanmugam, SJ Gendler, EP Bennett, MA Hollingsworth: Phosphorylation of MUC1 by Met modulates interaction with p53 and MMP1 expression. J Biol Chem 283, 26985-26995 (2008)
doi:10.1074/jbc.M805036200
36. P Laurén: The Two Histological Main Types Of Gastric Carcinoma: Diffuse Aand So-Called Iintestinal-Type Carcinoma. An Attempt At A Histo-Clinical Classification. Acta Pathol Microbiol Scand 64, 31-49 (1965)
PMid:12825872
37. G Gondal, T Grotmol, B Hofstad, M Bretthauer, TJ Eide, G Hoff: The Norwegian Colorectal Cancer Prevention (NORCCAP) screening study: baseline findings and implementations for clinical work-up in age groups 50-64 years. Scand J Gastroenterol 38, 635-642. (2003)
38. M Nuti, YA Teramoto, R Mariani-Costantini, PH Hand, D Colcher, J Schlom: A monoclonal antibody (B72.3) defines patterns of distribution of a novel tumor-associated antigen in human mammary carcinoma cell populations. Int J Cancer 29, 539-545 (1982)
doi:10.1002/ijc.2910290509
PMid:2450649
39. T Kjeldsen, H Clausen, S Hirohashi, T Ogawa, H Iijima, S Hakomori: Preparation and characterization of monoclonal antibodies directed to the tumor-associated O-linked sialosyl-2----6 alpha-N-acetylgalactosaminyl (sialosyl-Tn) epitope. Cancer Res 48, 2214-2220 (1988)
PMid:9530956
40. CA Reis, T Sorensen, U Mandel, L David, E Mirgorodskaya, P Roepstorff, J Kihlberg, JE Hansen, H Clausen . Development and characterization of an antibody directed to an alpha-N-acetyl-D-galactosamine glycosylated MUC2 peptide. Glycoconj J 15, 51-62 (1998)
doi:10.1023/A:1006939432665
41. CA Reis, L David, PA Nielsen, H Clausen, K Mirgorodskaya, P Roepstorff, M Sobrinho-Simões: Immunohistochemical study of MUC5AC expression in human gastric carcinomas using a novel monoclonal antibody. Int J Cancer 74, 112-121 (1997)
doi:10.1002/(SICI)1097-0215(19970220)74:1<112::AID-IJC19>3.0.CO;2-H
PMid:16207894
42. AL Sorensen, CA Reis, MA Tarp, U Mandel, K Ramachandran, V Sankaranarayanan, T Schwientek, R Graham, J Taylor-Papadimitriou, MA Hollingsworth, J Burchell, H Clausen: Chemoenzymatically synthesized multimeric Tn/STn MUC1 glycopeptides elicit cancer-specific anti-MUC1 antibody responses and override tolerance. Glycobiology 16, 96-107 (2006)
doi:10.1093/glycob/cwj044
PMid:17050588
43. MA Tarp, AL Sorensen, U Mandel, H Paulsen, J Burchell, J Taylor-Papadimitriou, H Clausen: Identification of a novel cancer-specific immunodominant glycopeptide epitope in the MUC1 tandem repeat. Glycobiology 17, 197-209 (2007)
doi:10.1093/glycob/cwl061
PMid:3948246
44. CA Reis, L David, P Correa, F Carneiro, C de Bolos, E Garcia, U Mandel, H Clausen, M Sobrinho-Simões: Intestinal metaplasia of human stomach displays distinct patterns of mucin (MUC1, MUC2, MUC5AC, and MUC6) expression. Cancer Res 59, 1003-1007 (1999)
PMid:10070955
45. S Ogata, I Ho, A Chen, D Dubois, J Maklansky, A Singhal, S Hakomori, SH Itzkowitz: Tumor-associated sialylated antigens are constitutively expressed in normal human colonic mucosa. Cancer Res 55, 1869-1874 (1995)
PMid:7537175
46. EP Bennett, H Hassan, U Mandel, MA Hollingsworth, N Akisawa, Y Ikematsu, G Merkx, AG van Kessel, S Olofsson, H Clausen: Cloning and characterization of a close homologue of human UDP-N-acetyl-alpha-D-galactosamine:Polypeptide N-acetylgalactosaminyltransferase-T3, designated GalNAc-T6. Evidence for genetic but not functional redundancy. J Biol Chem 274, 25362-25370 (1999)
doi:10.1074/jbc.274.36.25362
PMid:10464263
47. H Sasaki, B Bothner, A Dell, M Fukuda: Carbohydrate structure of erythropoietin expressed in Chinese hamster ovary cells by a human erythropoietin cDNA. J Biol Chem 262, 12059-12076 (1987)
PMid:3624248
48. T Ju, GS Lanneau, T Gautam, Y Wang, B Xia, SR Stowell, MT Willard, W Wang, JY Xia, RE Zuna, Z Laszik, DM Benbrook, MH Hanigan, RD Cummings: Human tumor antigens Tn and sialyl Tn arise from mutations in Cosmc. Cancer Res 68, 1636-1646 (2008)
doi:10.1158/0008-5472.CAN-07-2345
PMid:18339842
49. T Ju, RD Cummings: A unique molecular chaperone Cosmc required for activity of the mammalian core 1 beta 3-galactosyltransferase. Proc Natl Acad Sci USA 99, 16613-16618 (2002)
doi:10.1073/pnas.262438199
PMid:12464682 PMCid:139192
50. F Wang, M Goto, YS Kim, M Higashi, K Imai, E Sato, S Yonezawa: Altered GalNAc-alpha-2,6-sialylation compartments for mucin-associated sialyl-Tn antigen in colorectal adenoma and adenocarcinoma. J Histochem Cytochem 49, 1581-1592 (2001)
PMid:11724906
51. PJ Mullen, N Carr, JD Milton, JM Rhodes: Immunohistochemical detection of O-acetylated sialomucins in intestinal metaplasia and carcinoma of the stomach. Histopathology 27, 161-167 (1995)
doi:10.1111/j.1365-2559.1995.tb00024.x
PMid:8835264
52. B Ferreira, NT Marcos, L David, J Nakayama, CA Reis: Terminal alpha1,4-linked N-acetylglucosamine in Helicobacter pylori-associated intestinal metaplasia of the human stomach and gastric carcinoma cell lines. J Histochem Cytochem 54, 585-591 (2006)
doi:10.1369/jhc.5A6836.2006
PMid:16618943
53. C Vazquez-Martin, E Gil-Martin, A Fernandez-Briera: Alterations of CMP-NeuAc:asialofetuin sialyltransferase activities in human colorectal adenocarcinoma. Oncology 64, 74-82 (2003)
doi:10.1159/000066515
54. C Vazquez-Martin, E Cuevas, E Gil-Martin, A Fernandez-Briera: Correlation analysis between tumorours associated antigen sialyl-Tn expression and ST6GalNAcI activity in human colon adenocarcinoma. Oncology 67, 159-165 (2004)
doi:10.1159/000081003
Key Words: ST6GalNAc-I, Sialyl-Tn, Glycosylation, Gastric Carcinoma, Intestinal Metaplasia
Send correspondence to: Celso A. Reis, Institute of Molecular Pathology and Immunology of the University of Porto (IPATIMUP), Rua Dr. Roberto Frias, s/n; 4200-465 Porto, Portugal, Tel: 351-225570700, Fax: 351-225570799, E-mail:celso.reis@ipatimup.pt