Glycobiology Advance Access originally published online on September 28, 2005
Glycobiology 2006 16(2):146-154; doi:10.1093/glycob/cwj045
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Phenotypic changes induced by expression of ß-galactoside 
2,6 sialyltransferase I in the human colon cancer cell line SW948
Dipartimento di Patologia Sperimentale, Università di Bologna, Via S. Giacomo 14, 40126 Bologna, Italy
1 To whom correspondence should be addressed; e-mail: fabio.dallolio{at}unibo.it
Received on August 2, 2005; revised on September 19, 2005; accepted on September 20, 2005
 |
Abstract |
|---|
 Top Abstract IntroductionResultsDiscussionMaterials and methodsAcknowledgmentsReferences |
|---|
ß-Galactoside
2,6 sialyltransferase (ST6Gal.I), the enzyme which adds sialic acid in 2,6-linkage on lactosaminic termini of glycoproteins, is frequently overexpressed in cancer, but its relationship with malignancy remains unclear. In this study, we have investigated the phenotypic changes induced by the expression of 2,6-sialylated lactosaminic chains in the human colon cancer cell line SW948 which was originally devoid of ST6Gal.I. Clones derived from transfection with the ST6Gal.I cDNA were compared with untransfected cells and mock transfectants. The ST6Gal.I-expressing clones show (1) increased adherence to fibronectin and collagen IV but not to hyaluronic acid. Treatment with Clostridium perfrigens neuraminidase reduces the binding to fibronectin and collagen IV of ST6Gal.I-expressing cells but not that of ST6Gal.I-negative cells; (2) accumulation and more focal distribution of ß1 integrins on the cell surface; (3) different distribution of actin fibers; (4) flatter morphology and reduced tendency to multilayer growth; (5) improved ability to heal a scratch wound; (6) reduced ability to grow at the subcutaneous site of injection in nude mice. Our data suggest that the presence of 2,6-linked sialic acid on membrane glycoconjugates increases the binding to extracellular matrix components, resulting in a membrane stabilization of ß1 integrins, further strengthening the binding. This mechanism can provide a basis for the flatter morphology and the reduced tendency to multilayer growth, resulting in a more ordered tissue organization. These data indicate that in the cell line SW948, the effect of ST6Gal.I expression is consistent with the attenuation of the neoplastic phenotype. Key words: cell adhesion / collagen / fibronectin / integrins / in vivo growth
|
Introduction |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterials and methodsAcknowledgmentsReferences |
|---|
The oligosaccharide structures present at the surface of mammalian cells are involved in highly specific recognition phenomena (Varki, 1993
) and are frequently terminated by sialic acids, nine carbon sugars bearing a negative electric charge at physiological pH values (Schauer, 2000). Early indications on a role played by sialic acids in cancer biology and, in particular on the adhesive properties of cancer cells, came from experimental studies showing that the extent of cell surface sialylation of various cancer cell lines positively correlated with their invasive properties (Dennis et al., 1982; Collard et al., 1986; Morgenthaler et al., 1990), even though the conclusions reached were often conflicting.
The addition of sialic acids is mediated by sialyltransferases, enzymes which share a common donor substrate, the cytidine monophosphate-sialic acid, but are distinguished by the oligosaccharide acceptor on which they act and for the linkage type they elaborate (2,3; 2,6 or 2,8) (reviewed in Dall’Olio and Chiricolo, 2001; Harduin-Lepers et al., 2001). The addition of sialic acid in 2,6-linkage to N-acetyllactosaminic chains of glycoproteins is mediated by ß-galactoside 2,6 sialyltransferase (Weinstein et al., 1982, 1987) (ST6Gal.I according to the most widely accepted nomenclature; Tsuji et al., 1996). Among the sialyltransferases, ST6Gal.I is peculiar, because its regulation is closely associated with neoplastic transformation, in part because the enzyme is positively controlled by the ras pathway (Dalziel et al., 2004). Overexpression of ST6Gal.I has been reported to occur in several human malignancies; several clinical and experimental studies suggest a positive correlation between high ST6Gal.I and invasive behavior of cancer cells (reviewed in Dall’Olio, 2000), but other studies report an opposite conclusion (Yamamoto et al., 1997, 2001). The purpose of this work is to investigate the impact of ST6Gal.I expression on the phenotype of a human colon cancer cell line originally devoid of this enzyme activity. We found that ST6Gal.I expression induces an increased binding to fibronectin and collagen IV but not to hyaluronic acid; an accumulation and a different distribution of ß1 integrins on the cell surface through a posttranscriptional mechanism, a flatter morphology and a reduced tendency to multilayer growth; a reduced in vivo growth ability. Together, our data indicate that the expression of ST6Gal.I in this cell line attenuates the neoplastic phenotype.
|
Results |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterials and methodsAcknowledgmentsReferences |
|---|
ST6Gal.I-transfected clones express ST6Gal.I mRNA, sialyltransferase activity toward asialotransferrin and
2,6-sialylated sugar chains on the cell surfaceIn Figure 1A is shown the northern blot analysis of untransfected cells, mock-transfected and ST6Gal.I-transfected clones. For comparison is shown also the analysis of the human hepatocarcinoma cell line HepG2, which expresses ST6Gal.I at very high levels. All ST6Gal.I-transfected cells produced, albeit at a different level, a 1.2–1.3 Kb transcript which is much smaller than the 4.3–4.7 Kb transcript expressed by HepG2 cells. This difference is due to the absence in the cDNA used for transfection of the long-3'-untranslated region, which is present in the endogenous transcript (Paulson et al., 1989
; Wang et al., 1993). Neither the endogenous nor the transduced transcripts were expressed by untransfected and mock-transfected clones. The enzyme activity toward asialotransferrin (Figure 1B) was negligible in untransfected and mock-transfected clones, whereas the three ST6Gal.I-transfected clones expressed an enzyme activity ranging between 50 (for 948F4) and 20% (for 948T17) of that expressed by HepG2 cells. Fluorescence-activated cell sorter (FACS) analysis, using fluorescent lectin from Sambucus nigra (SNA) (Figure 1C) revealed a much stronger reactivity in the three ST6Gal.I-transfected clones, compared with mock-transfected and untransfected clones. Consistent with previous observations (Lee et al., 1989), this data indicates that the very low level of expression of 2,6-sialylated sugar chains in cells not expressing ST6Gal.I, undergoes a dramatic up-regulation upon ST6Gal.I expression.
|
The binding properties of ST6Gal.I-transfected clones
Figure 2A–C reports the dose-dependent adhesion of untransfected, mock-transfected and ST6Gal.I-transfected cells to collagen IV, fibronectin and hyaluronic acid, respectively. The adhesion to the three different substrates of untransfected (white bars) and mock-transfected cells (grey bars) was very similar, whereas ST6Gal.I-transfectants (black bars) displayed a much higher adhesion to fibronectin and to a lesser extent, to collagen IV. On the contrary, adhesion to hyaluronic acid was not dependent on the expression of 2,6-linked sialic acid on the cell surface.
|
These data are consistent with the view that adhesion to substrates recognized through ß1 integrin receptors can be modulated by sialic acid 2,6-linked to lactosamine, whereas the adhesion to substrates recognized by nonintegrin receptors (such as hyaluronic acid) is not modulated by such sugar epitope.
ST6Gal.I-expressing clones display higher amounts and a different distribution of ß1 integrins on the cell surface
Because ST6Gal.I-expressing clones showed an increased adhesion to extracellular matrix components interacting with ß1 integrin receptors, we investigated the expression of such molecules on the surface of neo- and ST6Gal.I-expressing clones. In Figure 3A is shown the FACS analysis with anti ß1 integrin antibodies of neo- and ST6Gal.I-transfected clones. In upper panels is reported the pattern of the cells analyzed just after trypsin treatment. It is evident that ß1 integrins were expressed at a slightly but consistently higher level by all ST6Gal.I-transfected clones. This unexpected observation prompted us to investigate whether such ß1 integrin overexpression was due to increased ß1 integrin gene expression in ST6Gal.I-transfected cells. This possibility was not supported by northern analysis (Figure 3B) which revealed that ß1 integrin mRNA expression, normalized for the ß-actin signal was not consistently overexpressed in ST6Gal.I-expressing clones. Thus, we hypothesized that the increased expression of ß1 integrins was due to a mechanism of posttranslational stabilization. To test the hypothesis, we left trypsin-released cell clones in suspension for 16 h before FACS-analysis. As shown in the lower panels of Figure 3A, this treatment did not affect the expression of ß1 integrins in neo-clones, whereas in ST6Gal.I-expressing clones the culture in nonadherent conditions down-regulated the expression of ß1 integrins to a level very close, if not identical, to that of neo-clones. This suggests that the adhesion to extracellular matrix components results in a stabilization of the presence of ß1 integrins in cells expressing 2,6-linked sialic acid on the cell surface but not in cells lacking this type of carbohydrate structure. To further investigate this point, the presence and distribution of ß1 integrins on the surface of control and ST6Gal.I-expressing clones was investigated by fluorescence microscopy. In the upper panels of Figure 4 is shown that the regions of cell–cell contact of the six cell lines displayed clear quantitative and qualitative differences. In fact, the reactivity of the three ST6Gal.I-expressing clones was not only higher than that of neo-clones but was also differentially distributed. Although the reactivity in neo-clones was evenly distributed on the cell membrane, in ST6Gal.I-expressing clones, the reactivity showed areas of focal distribution (arrows). This observation is consistent with the view that ST6Gal.I transfection induces a focal redistribution of ß1 integrin receptors on the cell membrane.
|
|
In focal adhesions, the actin fibers of the cytoskeleton are connected to the cytoplasmic portions of integrin receptors. To establish whether the different membrane distribution of ß1 integrins we observed in ST6Gal.I-expressing clones leads to differences in the cytoskeletal organization, we investigated the distribution of polymeric actin fibers using phalloidin as a probe (Figure 4, lower panels). The most remarkable difference between neo and ST6Gal.I clones was the more ordered association, in the last group of cell lines, of the actin fibers with the plasma membrane. In fact, in the three ST6Gal.I-expressing clones, the actin fibers showed a close and ordered association with the internal side of the plasma membrane. This organization was rarely observed in control transfectants.
Neuraminidase treatment reduces substratum adhesion only in ST6Gal.I expressors
The increased expression of ß1 integrins on the cell surface opens the question as to whether the increased adhesion of ST6Gal.I-expressing clones to extracellular matrix glycoproteins is due to a direct effect of 2,6-linked sialic acid on the strength of the binding (qualitative hypothesis) or to the increased membrane expression of ß1 integrins (quantitative hypothesis) or both. To address this point, we analyzed the adhesion of 948neo2 and 948F4 cells to fibronectin and collagen IV after neuraminidase treatment. In Figure 5A are shown the flow cytometric profiles of mock-treated (upper panels) and neuraminidase-treated (lower panels) 948neo2 and 948F4 cells, probed with SNA or anti ß1 integrin antibodies. Neuraminidase treatment caused a marked reduction of SNA reactivity of 948F4 cells but leaved unaltered the reactivity of 948neo2 cells. However, even after neuraminidase treatment, the SNA reactivity of 948F4 cells remained much stronger than that of 948neo2 cells, indicating that 2,6-linked sialic acid had been only partially cleaved by the enzymatic treatment. On the contrary, the expression of ß1 integrins was not affected by neuraminidase treatment in neither cell line. Consistent with data reported in Figure 3, the expression of ß1 integrins was higher in 948F4 cells than in 948neo2 cells. Neuraminidase treatment caused a 50% reduction of fibronectin adhesion (Figure 5B) and a >60% reduction of adhesion to collagen IV (Figure 5C), compared with mock-treated cells. On the contrary, the adhesion of 948neo2 cells to either substrate was not affected by neuraminidase treatment. These results confirm the role played by 2,6-linked sialic acid in modulating cell adhesion to extracellular matrix glycoproteins and, more importantly, support the "qualitative hypothesis". In fact, despite the identical amounts of ß1 integrins present on the surface of neuraminidase-treated and mock-treated 948F4 cells, the adhesion to fibronectin and collagen IV of the former was much lower than that of the latter.
|
Phenotypic changes induced by ST6Gal.I expression
The three ST6Gal.I-expressing clones displayed marked morphological changes with respect to neo-clones (Figure 6). Compared with ST6Gal.I-negative cell lines, the three ST6Gal.I-expressing clones showed a much flatter morphology and a strong tendency to grow as a monolayer. In a layer of ST6Gal.I-expressing cells, the cells shape was polygonal, and the cell borders were clearly defined. By contrast, in layers of neo-clones the shape of the single cells could not be easily distinguished, and many structures appearing in phase contrast as white areas were seen (Figure 6, upper panels). This appearance is indicative of the multilayer organization of the tissue.
|
In addition, the ability to repair a scratch wound was markedly different in ST6Gal.I-expressing cells (Figure 7). In fact, at day 6 the wound was still visible in untransfected and mock-transfected cells, whereas it was completely healed in ST6Gal.I expressors. This difference cannot be explained by a different proliferation rate of the cell lines, because the doubling time of the seven clones was very similar if not identical (data not shown). Rather, the better ability to repair the wound displayed by ST6Gal.I-expressing clones seems to be due to a more ordered growth and to the reduced tendency to multilayer growth.
|
ST6Gal.I expression reduces tumorigenicity of SW948 cells
To study the impact of ST6Gal.I expression on the ability to grow in vivo, an identical number of 948neo2 or 948neo4 cells was mixed with 948F4 cells and injected subcutaneously in nude mice in a "competition assay." The genomic DNA extracted from the tumors grown in the animals was polymerase chain reaction (PCR) analyzed with a primer pair complementary to the regions flanking the multiple cloning site (MCS) of the expression vector pcDNA3. The forward primer was designed upstream the MCS, whereas the reverse primer was designed downstream of MCS. In cells transfected with the empty vector (neo-clones), the PCR amplification yielded the expected 480 bp product (Figure 8, lanes 948neo2 and 948neo4), whereas in cells transfected with the ST6Gal.I expression vector, the PCR yielded a product of 
1660 bp, as can be seen in Figure 8, lane 948F4. Sequencing of the 480 and of the 1660 bp products confirmed the presence of the expected sequences: the MCS plus the flanking regions in the 480 bp product and the ST6Gal.I insert plus the flanking regions in the 1660 bp product. When the genomic DNA extracted from tumors originated by mixed cell populations was analyzed by this technique, it turned out to be comprised only by DNA of the "neo-type." Moreover, SNA-fluorescein isothiocyanate (SNA-FITC) analysis of the cell lines re-established from the tumors revealed that they were comprised almost exclusively of SNA-negative cells (data not shown), confirming the conclusion of the PCR analysis. These data indicate that the expression of ST6Gal.I reduces the tumorigenic potential of the cell line SW948.
|
|
Discussion |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterials and methodsAcknowledgmentsReferences |
|---|
In this work, we show that the expression of sialyltransferase ST6Gal.I in a colon cancer cell line originally devoid of this glycosyltransferase activity results in: (1) an increased binding to extracellular substrates recognized by integrin receptors; (2) an accumulation and a different distribution of ß1-integrins on the cell surface; (3) a different organization of the actin fibers; (4) a reduced ability to multilayer growth and a better ability to heal a wound; and (5) a reduced ability to grow in vivo.
It has long been recognized that sialic acid can somehow modulate the adhesion of cancer cells to extracellular matrix components. However, the conclusions reached by different studies are different and, in some cases, conflicting. In fact, according to some studies, increased sialylation enhances the binding to extracellular substrates (Morgenthaler et al., 1990; Lin et al., 2002; Seales et al., 2005), whereas others report an opposite effect (Dennis et al., 1982; Le Marer and Skacel, 1999) or no effect (Kijima-Suda et al., 1988).
Integrins and particularly ß1 integrin, are substrates of several glycosyltransferases, including ß1,6 N-acetylglucosaminyltransferase V (Zheng et al., 1994; Demetriou et al., 1995), sialyltransferase ST6GalNAc I (Clement et al., 2004; Julien et al., 2005), and ST6Gal.I (Pochec et al., 2003; Seales et al., 2003, 2005) (reviewed in Bellis, 2004). Phenotypic changes have been related to a direct effect of altered glycosylation of the integrin molecules, but in some cases they have been related also to a differential glycosylation of integrin-associated cell surface molecules (Sudou et al., 1995; Wang et al., 2001). The relationship between sialylation of ß1 integrins and adhesion to extracellular matrix components is controversial. In fact, the expression of the sialyl-Tn epitope on ß1 integrins has no effect on fibronectin binding (Clement et al., 2004), whereas the sialylation of ß1 integrins reduces the binding to extracellular matrix glycoproteins in melanoma (Pochec et al., 2003) and in myeloid cells (Semel et al., 2002), but it increases it in erythroleukemia K562 cells (Symington et al., 1989), mouse melanoma cells (Oz et al., 1989; Kawano et al., 1993), and colon epithelial cells (Seales et al., 2003, 2005), suggesting that the effect induced by the presence of sialic acid on integrin receptors is strongly dependent on the cell type, on the specific type of sialyl-linkage and on the level of sialylation.
The finding that all ST6Gal.I-transfected clones express increased levels of ß1 integrins was largely unexpected. This phenomenon, which is not supported by an increased transcription of the ß1 integrin gene, is reversible when cells are not allowed to adhere for a few hours, suggesting that adhesion to a solid substrate can stabilize ß1 integrins on the cell surface through a 2,6-linked sialic acid-dependent mechanism. This view is supported by the immunofluorescence microscopy studies which reveal a focal aggregation of ß1-integrin molecules on the surface of ST6Gal.I-transfected cells but not on that of control transfectants. Consistent with a membrane accumulation of ß1-integrins on the cell surface, we observed in ST6Gal.I transfectants also a closer association of actin fibers with the cell membrane. Nevertheless, the accumulation of ß1 integrins on the cell surface does not appear to be the only mechanism responsible for the increased adhesion of ST6Gal.I-expressing cells to collagen IV and fibronectin. In fact, neuraminidase treatment leaves unaltered the level of ß1 integrin expression in 948F4 cells but reduces the adhesion to fibronectin and collagen IV by 50–60%. This indicates that the presence/absence of 2,6-linked sialic acid on the cell surface plays a direct role on cell adhesion, which is not dependent on the increased amount of ß1 intergrins. Moreover the finding that neuraminidase treatment does not alter the adhesion of 948neo2 cells indicates that sialy- linkages other than 2,6 to galactose, which are cleaved by neuraminidase treatment, play a little role in cell-substratum adhesion in this model system. On the basis of these data, we hypothesize a model in which cells overexpressing 2,6-sialylated sugar chains on their surface adhere more strongly to extracellular matrix through a direct effect of 2,6-sialylation on binding. This stronger interaction would increase the accumulation of integrin receptors in focal adhesions, resulting in an up-regulation of the receptors and a further reinforcement of the interaction. In this view, the presence of 2,6-linked sialic acid would result in a dual effect: the first would operate immediately and consists in the increase of the strength of the binding; the second would operate only after some hours and consists in an accumulation of integrin receptors in focal adhesions. Very likely, these mechanisms are at the basis of the flatter appearance of the tissues formed by ST6Gal.I-expressing 948 clones and of their better ability to heal a wound.
Binding to hyaluronic acid is mainly mediated by a nonintegrin receptor, namely CD44. Previous studies have indicated that the glycosylation state of CD44 (Lesley et al., 1995) and especially sialylation (Katoh et al., 1995) can modulate binding to hyaluronic acid. In fact, neuraminidase treatment of a purified CD44-immunoglobulin fusion protein dramatically increases its binding (Katoh et al., 1995). The fact that in our model system the dramatic differences in 2,6-sialylation fail to induce any remarkable change in hyaluronic acid binding activity can be explained by the fact that the putative regulatory activity is exerted by sialic acids linked with a bond other than ,2,6- to galactose.
Clinical (Gessner et al., 1993; Lise et al., 2000) and experimental studies (Bresalier et al., 1990; Zhu et al., 2001; Seales et al., 2005) suggest a direct relationship between ST6Gal.I overexpression and malignancy in colon cancer. However, it should be remembered that ST6Gal.I expression is under the direct control of the Ras pathway (Dalziel et al., 2004). Thus, the phenotypic changes attributed to ST6Gal.I overexpression in these studies could be due to the overexpression of Ras, more than to a direct effect of ST6Gal.I. For this reason, to elucidate the role played by 2,6-sialylation in cancer progression, it is necessary to utilize a model system in which the level of 2,6-sialylation is regulated in a Ras-independent manner (i.e., under a constitutive promoter). To our knowledge, this study is the first in which the invasive properties of human colon cancer cells expressing ST6Gal.I under a constitutive promoter are analyzed in vivo. Unexpectedly, we found that the constitutive expression of ST6Gal.I in SW948 cells attenuates some aspects of the malignant phenotype, including multistratified growth and ability to grow in vivo resembling, for some aspects, the phenotypic changes occurring upon ST6Gal.I expression in glioma cells (Yamamoto et al., 1997, 2001).
Together, current and previous results indicate that the relationship between ST6Gal.I expression and malignant phenotype can be complex and highly dependent on the cell type. In considering the overall effects of ST6Gal.I expression on tissues, it should be remembered that ST6Gal.I knockout mice undergo a normal development, showing only an altered immunological phenotype (Hennet et al., 1998). This observation is consistent with the notion that the phenotypic changes induced by ST6Gal.I can be often subtle and highly cell type specific.
|
Materials and methods |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterials and methodsAcknowledgmentsReferences |
|---|
Cell lines
The human colon carcinoma cell line SW948 (Leibovitz et al., 1976
) was routinely grown in Leibowitz’s L-15 medium supplemented with 10% fetal calf serum (FCS), penicillin (100 U/mL) and streptomycin (100 µg/mL). To minimize cell heterogeneity, before transfection experiments, the cell line was cloned by limiting dilution, and a single clone (948A1) was randomly chosen for transfection. The whole coding region of rat ST6Gal.I (Weinstein et al., 1987) was inserted into the HindIII/XhoI sites of the eukaryotic expression vector pcDNA3 (InVitrogen, Carlsbad, CA). Transfection was performed by electroporation, followed by G418 (Sigma, St. Louis, MO) selection. Individual clones resulting from transfection with the ST6Gal.I expression vector or with the empty vector (neo-clones) were isolated and characterized. Details on transfection experiments have been published previously (Dall’Olio et al., 1995, 1997, 2001). The human hepatocarcinoma cell line HepG2, a kind gift of Dr. Joseph Lau, Roswell Park Cancer Institute, Buffalo, NY, was routinely grown in Dulbecco’s modified Eagle medium (DMEM) 10% FCS and antibiotics.
Sialyltransferase activity
The enzyme activity was measured in whole cell homogenates in the range of linearity with respect to time and enzyme concentration by incorporation of [14C] sialic acid (American Radiolabel Company, St. Louis, MO) onto asialotransferrin as previously described (Dall’Olio et al., 2001). The protein concentration of the homogenates was determined by the Lowry method (Lowry et al., 1951).
Northern blot analysis
Total RNA was prepared by the Chomczynski and Sacchi method (Chomczynski and Sacchi, 1987). Fifteen micrograms were electrophoresed on a formaldehyde agarose gel and capillary transferred to a nylon membrane (Amersham, Little Chalfont, UK) as previously described (Dall’Olio et al., 1995). As probe for rat ST6Gal.I, the whole coding sequence cDNA was used. Probe for ß1 integrin was prepared by PCR amplification of the cDNA from clone 948neo2, obtained as described (Dall’Olio et al., 2001), using primers intb1L2 (5'-GATGCCGGGTTTCACTTTGC-3') and intb1R1 (5'-TCCGTTGCTGGCTTCACAAG-3') which amplify a 1038 bp fragment, common to all ß1 integrin mRNA isoforms. Probe for ß-actin was a kind gift of Dr. Marco Trinchera, University of Insubria, Varese, Italy. cDNA probes were [32P]-labeled by oligonucleotide priming as previously described (Dall’Olio et al., 1996), using as reverse primers FD2 (5'-ATTCTCGAGTGGAGAAGGATAAGGGTG-3'), Intb1R1 and ACTR3 (5'-GCTGGAAGGTGGACAGCGA-3') for ST6Gal.I, ß1 integrin and ß actin, respectively. Blots were prehybridized for 4 h at 60°C in hybridization solution (250 mM Na2HPO4; 15% [v/v] deionized formamide; 3% [w/v] bovine serum albumin [BSA]; 7% (w/v) sodium dodecyl sulfate (SDS); 1 mM EDTA; 0.2 mL of H3PO4 [for a 100 mL final volume]) and hybridized for 16 h at 60°C in hybridization solution containing 5 x 106 c.p.m./mL [32P] radiolabeled probes. Blots were washed twice at 60°C in washing solution (150 mM Na2HPO4; 0.1% SDS; 1.2 mL of H3PO4 (for 1 L final volume)] for 20 min and exposed.
FACS analysis
Cells were released by trypsin treatment, washed twice in phosphate-buffered saline containing 2 mg/mL of BSA (PBS–BSA) and incubated with 50 µg/mL of FITC-conjugated SNA prepared as described (Dall’Olio et al., 1995) or with FITC-conjugated anti ß1 integrin mouse monoclonal antibody (clone K20) (Dako, Glostrup, Denmark) at a concentration of 0.4 µg/mL for 20 min in ice. After a wash in PBS–BSA, the cells were resuspended in PBS–BSA and analyzed with a FACScan flow cytometer (Beckton Dickinson, Mountain View, CA).
In vitro adhesion assay
Adhesion of cells to extracellular matrix components was determined essentially as described (Demetriou et al., 1995) with some modifications. Twenty-four well plates were coated with the following amounts of adhesive substrates: fibronectin from human plasma (Sigma) (3.5, 7, 14 µg/well); collagen type IV from human placenta (Sigma) 14, 21, 42 µg/well; hyaluronic acid from human umbilical cord (Sigma) 12.5, 25, 62.5 µg/well. Plates were allowed to dry at 37°C overnight. Before use, all wells were saturated with serum-free DMEM containing 2.5 mg/mL BSA for 1 h at 37°C. Cells, released by trypsin treatment, were counted and aliquots of 5 x 105 cells were seeded in each well in serum-free medium containing 2.5% BSA and allowed to adhere for 2 h. Then, the medium was removed by aspiration, the wells were washed gently three times with PBS to remove unbound cells, and 250 µL of PBS containing 1% glutaraldehyde and 0.2% methylene blue were added and left overnight to fix and stain bound cells. The staining solution was removed and the plates were washed by complete submersion in water and then turned upside down on a paper pad. Six hundred microliters of the extraction solution (methanol : acetic acid : water [4:1:4]) were added to each well and after 1 h at room temperature, the A650 was determined. Each experimental point was performed in triplicate and the background adhesion in the absence of adhesive substrates was subtracted.
Immunofluorescence microscopy
Cells, seeded on glass coverslips 48 h before staining, were fixed in 2% formaldehyde in PBS for 20 min, washed in PBS three times for 5 min. Samples for phalloidin treatment were permeabilized with 0.05% Triton X-100 + 0.05% Tween 20 in PBS for 3 min. Samples were then washed three times in PBS for 5 min, incubated in 1% BSA in PBS for 30 min to block unspecific bindings and stained either with FITC-conjugated anti ß1 integrin mouse monoclonal antibody (clone K20) (Dako) at a concentration of 0.4 µg/mL in 1% BSA in PBS for 30 min or with rhodamine-conjugated phalloidin (1 µg/mL) for 2 min. After three washes (5 min) with PBS, coverslips were mounted in Mowiol and observed with a Nikon Eclipse E600 fluorescence microscope.
Neuraminidase treatment
Cells were released by trypsin treatment, collected by centrifugation and treated with 2 U/mL of neuraminidase from C. perfrigens (Sigma) in serum-free DMEM for 1 h at 37°C. Identical aliquots of cells were mock treated in the same conditions with heat-inactivated neuraminidase. At the end of the incubation time, cells were washed by centrifugation and subjected to FACS analysis and adhesion assay as above.
Scratch wound repair assay
A lane was denuded using a plastic pipette tip on 6 days postconfluent cultures of mock- or ST6Gal.I-transfected cell lines. The repair of the wound was checked daily, and photographs were taken under an inverted phase contrast microscope just after the wound was made and 6 days later.
Determination of tumorigenicity
Cells were released by trypsin treatment and carefully counted. 3 x 106 neo-cells (948neo2 or 948neo4) were mixed with an identical number of 948F4 cells and injected subcutaneously in the right flank of nu/nu mice. After 30 days, animals were sacrificed, tumors were carefully removed and the genomic DNA was PCR analyzed. Analysis was performed on 2 µg of genomic DNA using primer pair pcDNA3L.1 (5'-CGGTTTGACTCACGGGGATTTC-3')/pcDNA3R.2 (5'-TAGGAAAGGACAGTGGGAGTGG-3'), with 35 cycles of the following program: denaturation: 94°C 1 min; annealing 63°C 1 min; extension 72°C 1 min for 35, using InViTaq DNA polymerase (Eppendorf, Milan, Italy). PCR products were analyzed on a 1.5% agarose gel stained with ethidium bromide.
|
Acknowledgments |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterials and methodsAcknowledgmentsReferences |
|---|
This work was supported by funds from PRIN (MIUR, Rome), from the University of Bologna (funds for selected research topics), and from Pallotti Legacy for Cancer Research.
|
Abbreviations |
|---|
BSA, bovine serum albumin; DMEM, Dulbecco’s modified Eagle medium; FACS, fluorescence-activated cell sorter; FCS, fetal calf serum; FITC, fluorescein isothiocyanate; MCS, multiple cloning site; PBS, phosphate-buffered saline; PCR, polymerase chain reaction; SNA, Sambucus nigra agglutinin; ST6Gal.I, ß-galactoside
2,6 sialyltransferase, E.C. 2.4.99.1 |
References |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterials and methodsAcknowledgmentsReferences |
|---|
Bellis, S.L. (2004) Variant glycosylation: an underappreciated regulatory mechanism for beta1 integrins. Biochim. Biophys. Acta, 1663, 52–60.[Medline]
Bresalier, R.S., Rockwell, R.W., Dahiya, R., Duh, Q.Y., and Kim, Y.S. (1990) Cell surface sialoprotein alterations in metastatic murine colon cancer cell lines selected in an animal model for colon cancer metastasis. Cancer Res., 50, 1299–1307.
Chomczynski, P. and Sacchi, N. (1987) Single-step method of RNA isolation by acid guanidinium thiocyanate-phenol-chloroform extraction. Anal. Biochem., 162, 156–159.[Web of Science][Medline]
Clement, M., Rocher, J., Loirand, G., and Le Pendu, J. (2004) Expression of sialyl-Tn epitopes on beta1 integrin alters epithelial cell phenotype, proliferation and haptotaxis. J. Cell Sci., 117, 5059–5069.
Collard, J.G., Schijven, J.F., Bikker, A., La Riviere, G., Bolscher, J.G., and Roos, E. (1986) Cell surface sialic acid and the invasive and metastatic potential of T- cell hybridomas. Cancer Res., 46, 3521–3527.
Dall’Olio, F. (2000) The sialyl-2,6-lactosaminyl-structure: biosynthesis and functional role. Glycoconj. J., 17, 669–676.[CrossRef][Medline]
Dall’Olio, F. and Chiricolo, M. (2001) Sialyltransferases in cancer. Glycoconj. J., 18, 841–850.[CrossRef][Medline]
Dall’Olio, F., Chiricolo, M., Lollini, P., and Lau, J.T. (1995) Human colon cancer cell lines permanently expressing 2,6-sialylated sugar chains by transfection with rat ß-galactoside 2,6 sialyltransferase cDNA. Biochem. Biophys. Res. Commun., 211, 554–561.[CrossRef][Web of Science][Medline]
Dall’Olio, F., Malagolini, N., Guerrini, S., Lau, J.T., and Serafini-Cessi, F. (1996) Differentiation-dependent expression of human ß-galactoside 2,6-sialyltransferase mRNA in colon carcinoma CaCo-2 cells. Glycoconj. J., 13, 115–121.[CrossRef][Web of Science][Medline]
Dall’Olio, F., Mariani, E., Tarozzi, A., Meneghetti, A., Chiricolo, M., Lau, J.T., and Facchini, A. (1997) Expression of ß-galactoside 2,6-sialyltransferase does not alter the susceptibility of human colon cancer cells to NK-mediated cell lysis. Glycobiology, 7, 507–513.
Dall’Olio, F., Chiricolo, M., Mariani, E., and Facchini, A. (2001) Biosynthesis of the cancer-related sialyl-2,6-lactosaminyl epitope in colon cancer cell lines expressing ß-galactoside 2,6-sialyltransferase under a constitutive promoter. Eur. J. Biochem., 268, 5876–5884.[Web of Science][Medline]
Dalziel, M., Dall’Olio, F., Mungul, A., Piller, V., and Piller, F. (2004) Ras oncogene induces ß-galactoside 2,6-sialyltransferase (ST6Gal I) via a RalGEF-mediated signal to its housekeeping promoter. Eur. J. Biochem., 271, 3623–3634.[Web of Science][Medline]
Demetriou, M., Nabi, I.R., Coppolino, M., Dedhar, S., and Dennis, J.W. (1995) Reduced contact-inhibition and substratum adhesion in epithelial cells expressing GlcNAc-transferase V. J. Cell Biol., 130, 383–392.
Dennis, J., Waller, C., Timpl, R., and Schirrmacher, V. (1982) Surface sialic acid reduces attachment of metastatic tumour cells to collagen type IV and fibronectin. Nature, 300, 274–276.[CrossRef][Medline]
Gessner, P., Riedl, S., Quentmaier, A., and Kemmner, W. (1993) Enhanced activity of CMP-neuAc: Gal ß, 1-4GlcNAc: 2,6-sialyltransferase in metastasizing human colorectal tumor tissue and serum of tumor patients. Cancer Lett., 75, 143–149.[CrossRef][Web of Science][Medline]
Harduin-Lepers, A., Vallejo-Ruiz, V., Krzewinski-Recchi, M., Samyn-Petit, B., Julien, S., and Delannoy, P. (2001) The human sialyltransferase family. Biochimie, 83, 727–737.[Medline]
Hennet, T., Chui, D., Paulson, J.C., and Marth, J.D. (1998) Immune regulation by the ST6Gal sialyltransferase. Proc. Natl. Acad. Sci. U. S. A., 95, 4504–4509.
Julien, S., Lagadec, C., Krzewinski-Recchi, M.A., Courtand, G., Le, B.X., and Delannoy, P. (2005) 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.[CrossRef][Web of Science][Medline]
Katoh, S., Zheng, Z., Oritani, K., Shimozato, T., and Kincade, P.W. (1995) Glycosylation of CD44 negatively regulates its recognition of hyaluronan. J. Exp. Med., 182, 419–429.
Kawano, T., Takasaki, S., Tao, T.W., and Kobata, A. (1993) Altered glycosylation of beta 1 integrins associated with reduced adhesiveness to fibronectin and laminin. Int. J. Cancer, 53, 91–96.[Web of Science][Medline]
Kijima-Suda, I., Miyazawa, T., Itoh, M., Toyoshima, S., and Osawa, T. (1988) Possible mechanism of inhibition of experimental pulmonary metastasis of mouse colon adenocarcinoma 26 sublines by a sialic acid: nucleoside conjugate. Cancer Res., 48, 3728–3732.
Le Marer, N. and Skacel, P.O. (1999) Up-regulation of 2,6 sialylation during myeloid maturation: a potential role in myeloid cell release from the bone marrow. J. Cell Physiol., 179, 315–324.[CrossRef][Web of Science][Medline]
Lee, E.U., Roth, J., and Paulson, J.C. (1989) Alteration of terminal glycosylation sequences on N-linked oligosaccharides of Chinese hamster ovary cells by expression of ß-galactoside 2,6-sialyltransferase. J. Biol. Chem., 264, 13848–13855.
Leibovitz, A., Stinson, J.C., Mccombs, W.B. III, Mccoy, C.E., Mazur, K.C., and Mabry, N.D. (1976) Classification of human colorectal adenocarcinoma cell lines. Cancer Res., 36, 4562–4569.
Lesley, J., English, N., Perschl, A., Gregoroff, J., and Hyman, R. (1995) Variant cell lines selected for alterations in the function of the hyaluronan receptor CD44 show differences in glycosylation. J. Exp. Med., 182, 431–437.
Lin, S., Kemmner, W., Grigull, S., and Schlag, P.M. (2002) Cell surface alpha2,6-sialylation affects adhesion of breast carcinoma cells. Exp. Cell Res., 276, 101–110.[CrossRef][Web of Science][Medline]
Lise, M., Belluco, C., Perera, S.P., Patel, R., Thomas, P., and Ganguly, A. (2000) Clinical correlations of 2,6-sialyltransferase expression in colorectal cancer patients. Hybridoma, 19, 281–286.[CrossRef][Web of Science][Medline]
Lowry, O.H., Rosebrough, N.J., Farr, A.L., and Randall, R.J. (1951) Protein measurement with the Folin phenol reagent. J. Biol. Chem., 193, 265–275.
Morgenthaler, J., Kemmner, W., and Brossmer, R. (1990) Sialic acid dependent cell adhesion to collagen IV correlates with in vivo tumorigenicity of the human colon carcinoma sublines HCT116, HCT116a and HCT116b. Biochem. Biophys. Res. Commun., 171, 860–866.[CrossRef][Web of Science][Medline]
Oz, O.K., Campbell, A., and Tao, T.W. (1989) Reduced cell adhesion to fibronectin and laminin is associated with altered glycosylation of beta 1 integrins in a weakly metastatic glycosylation mutant. Int. J. Cancer, 44, 343–347.[Web of Science][Medline]
Paulson, J.C., Weinstein, J., and Schauer, A. (1989) Tissue-specific expression of sialyltransferases. J. Biol. Chem., 264, 10931–10934.
Pochec, E., Litynska, A., Amoresano, A., and Casbarra, A. (2003) Glycosylation profile of integrin alpha 3 beta 1 changes with melanoma progression. Biochim. Biophys. Acta, 1643, 113–123.[Medline]
Schauer, R. (2000) Achievements and challenges of sialic acid research. Glycoconj. J., 17, 485–499.[CrossRef][Web of Science][Medline]
Seales, E.C., Jurado, G.A., Singhal, A., and Bellis, S.L. (2003) Ras oncogene directs expression of a differentially sialylated, functionally altered beta1 integrin. Oncogene, 22, 7137–7145.[CrossRef][Web of Science][Medline]
Seales, E.C., Jurado, G.A., Brunson, B.A., Wakefield, J.K., Frost, A.R., and Bellis, S.L. (2005) Hypersialylation of ß1 integrins, observed in colon adenocarcinoma, may contribute to cancer progression by up-regulating cell motility. Cancer Res., 65, 4645–4652.
Semel, A.C., Seales, E.C., Singhal, A., Eklund, E.A., Colley, K.J., and Bellis, S.L. (2002) Hyposialylation of integrins stimulates the activity of myeloid fibronectin receptors. J. Biol. Chem., 277, 32830–32836.
Sudou, A., Ozawa, M., and Muramatsu, T. (1995) Lewis X structure increases cell substratum adhesion in L cells. J. Biochem. (Tokyo), 117, 271–275.
Symington, B.E., Symington, F.W., and Rohrschneider, L.R. (1989) Phorbol ester induces increased expression, altered glycosylation, and reduced adhesion of K562 erythroleukemia cell fibronectin receptors. J. Biol. Chem., 264, 13258–13266.
Tsuji, S., Datta, A.K., and Paulson, J.C. (1996) Systematic nomenclature for sialyltransferases. Glycobiology, 6, v–vii.
Varki, A. (1993) Biological roles of oligosaccharides: all of the theories are correct. Glycobiology, 3, 97–130.
Wang, X., Vertino, A., Eddy, R.L., Byers, M.G., Jani-Sait, S.N., Shows, T.B., and Lau, J.T. (1993) Chromosome mapping and organization of the human ß-galactoside 2,6-sialyltransferase gene. Differential and cell-type specific usage of upstream exon sequences in B-lymphoblastoid cells. J. Biol. Chem., 268, 4355–4361.
Wang, X., Sun, P., Al Qamari, A., Tai, T., Kawashima, I., and Paller, A.S. (2001) Carbohydrate-carbohydrate binding of ganglioside to integrin 5 modulates 5ß1 function. J. Biol. Chem., 276, 8436–8444.
Weinstein, J., de Souza e Silva, U., and Paulson, J.C. (1982) Purification of a Gal ß 1,4GlcNAc 2,6 sialyltransferase and a Gal ß 1,3(4) GlcNAc 2,3 sialyltransferase to homogeneity from rat liver. J. Biol. Chem., 257, 13835–13844.
Weinstein, J., Lee, E.U., Mcentee, K., Lai, P.H., and Paulson, J.C. (1987) Primary structure of ß-galactoside 2,6-sialyltransferase. Conversion of membrane-bound enzyme to soluble forms by cleavage of the NH2-terminal signal anchor. J. Biol. Chem., 262, 17735–17743.
Yamamoto, H., Kaneko, Y., Rebbaa, A., Bremer, E.G., and Moskal, J.R. (1997) 2,6-Sialyltransferase gene transfection into a human glioma cell line (U373 MG) results in decreased invasivity. J. Neurochem., 68, 2566–2576.[Web of Science][Medline]
Yamamoto, H., Oviedo, A., Sweeley, C., Saito, T., and Moskal, J.R. (2001) 2,6-Sialylation of cell-surface N-glycans inhibits glioma formation in vivo. Cancer Res., 61, 6822–6829.
Zheng, M., Fang, H., and Hakomori, S. (1994) Functional role of N-glycosylation in alpha 5 beta 1 integrin receptor. De-N-glycosylation induces dissociation or altered association of alpha 5 and beta 1 subunits and concomitant loss of fibronectin binding activity. J. Biol. Chem., 269, 12325–12331.
Zhu, Y., Srivatana, U., Ullah, A., Gagneja, H., Berenson, C.S., and Lance, P. (2001) Suppression of a sialyltransferase by antisense DNA reduces invasiveness of human colon cancer cells in vitro. Biochim. Biophys. Acta, 1536, 148–160.[Medline]
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  P. M. Drake, W. Cho, B. Li, A. Prakobphol, E. Johansen, N. L. Anderson, F. E. Regnier, B. W. Gibson, and S. J. Fisher Sweetening the Pot: Adding Glycosylation to the Biomarker Discovery Equation Clin. Chem., February 1, 2010; 56(2): 223 - 236. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. Malagolini, M. Chiricolo, M. Marini, and F. Dall'Olio Exposure of {alpha}2,6-sialylated lactosaminic chains marks apoptotic and necrotic death in different cell types Glycobiology, February 1, 2009; 19(2): 172 - 181. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Lee, H.-J. Lee, S. Bae, and Y.-S. Lee Protein Sialylation by Sialyltransferase Involves Radiation Resistance Mol. Cancer Res., August 1, 2008; 6(8): 1316 - 1325. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Hedlund, E. Ng, A. Varki, and N. M. Varki {alpha}2-6 Linked Sialic Acids on N-Glycans Modulate Carcinoma Differentiation In vivo Cancer Res., January 15, 2008; 68(2): 388 - 394. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
N. Malagolini, D. Santini, M. Chiricolo, and F. Dall'Olio Biosynthesis and expression of the Sda and sialyl Lewis x antigens in normal and cancer colon Glycobiology, July 1, 2007; 17(7): 688 - 697. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||