| Glycobiology | Pages |
Lex glycosphingolipids-mediated cell aggregation
Introduction
Results
Discussion
Materials and methods
Acknowledgments
Abbreviations
References
Lex glycosphingolipids-mediated cell aggregation
Glycoconjugates bearing oligosaccharide Lex, Gal[beta]1[rarr]4(Fuc[alpha]1[rarr]3)GlcNAc[beta]1[rarr]3R, are found on the surface of several cell types. Although recent studies have indicated that Lex on both glycosphingolipids (GSL) and polylactosaminoglycans can mediate under certain experimental conditions Lex-Lex interactions, cell-cell interactions based exclusively on Lex GSLs have not been demonstrated. In this study we show that preincubation of nonaggregating rat basophilic leukemia (RBL) cells with purified Lex GSLs resulted in incorporation of the GSLs into plasma membrane, as determined by immunostaining, and formation of aggregates in the presence of Ca2+; no aggregates were formed after preincubation of the cells with globoside or sphingomyelin. Lex-mediated aggregation was inhibited by removal of Ca2+ or by addition of lactofucopentaose III but not by lactose or lacto-N-fucopentaose II. In a mixture of Lex-positive and Lex-negative RBL cells most of the aggregates were composed exclusively of Lex-positive cells. The combined data suggest that interactions between Lex GSL on opposite cell surfaces are strong enough to allow formation of stable cell-cell contacts. Key words: cell aggregation/glycosphingolipids/ Lex/RBL cells
Introduction
Glycoconjugates at the cell surface appear to mediate a variety of cellular activities including cell recognition and adhesion (Fenderson et al., 1990; Shur, 1994). Although most evidence has accumulated for the role of lectin/enzyme-carbohydrate binding in cell-cell and cell-substratum interactions, recent experiments have indicated that carbohydrate-carbohydrate interactions could also play a role (for review, see Hakomori, 1992). These experiments stem from the finding that compaction of the mouse embryo correlates with the expression of stage-specific embryonic antigen 1 (SSEA-1), recognized by an antibody specific for Lex (Solter and Knowles, 1978; Gooi et al., 1981; Hakomori et al., 1981). The compaction process could be inhibited and the compacted embryos could be decompacted by Lex and more efficiently by multivalent Lex (Bird and Kimber, 1984; Fenderson et al., 1984). Several lines of evidence suggested that Lex GSLs-Lex GSLs interactions could play a role in this system (Eggens et al., 1989; Hakomori, 1992; Kojima et al., 1994). Thus, homotypic aggregation of mouse embryonal carcinoma (EC) cells F9, carrying Lex, could be inhibited by multivalent Lex. Second, liposomes containing Lex GSLs bound preferentially to Lex-positive F9 cells and to Lex GSLs coated on plastic surface. Third, liposomes or plastic microbeads containing Lex GSLs exhibited self aggregation, to an extent which was not observed with other glycolipid liposomes. Finally, comparative adhesion studies of Lex-expressing tumor cells and their Lex-nonexpressing variants showed that only Lex-expressing cells adhere to Lex GSLs-coated plates, and are involved in cell aggregation.
However, the major carrier of Lex in EC cells is not GSL but polylactosaminoglycan (embryoglycan) (Childs et al., 1983). Recent data showed autoaggregation of purified embryoglycan in the presence of Ca2+; this aggregation was inhibited by EDTA or by pretreatment of the embryoglycan by fucosidase, corroborating the previous results that Lex could play a role in cell-cell aggregation phenomena (Kojima et al., 1994). Thus, Lex-mediated interactions may reflect the presence of Lex on embryoglycan, which expresses Lex at high density, rather than the involvement of Lex GSLs in cell-cell interactions. However, it is difficult to dissect the contribution of Lex carried by GSLs, embryoglycan and/or other surface carriers in aggregation of F9 and other Lex-expressing cells. Furthermore, EC cells express on their surface the Ca2+-dependent adhesion molecule, E-cadherin, which may complicate the interpretation of aggregation experiments. It has been shown that variants of EC cells defective in the expression of Lex or embryoglycan exhibit autoaggregation comparable to that of Lex-positive parental cells, suggesting that the contribution of Lex to cell-cell interactions may be marginal in this system (Buckalew et al., 1985; Dráber and Maly, 1987). This conclusion was supported by an analysis of autoaggregation of P19 EC cells defective in the expression of E-cadherin (Sakalian and Dráber, 1991) and by an analysis of early aggregation events in embryoglycan- and/or Lex-deficient P19 EC cell mutants (Boubelík et al., 1996).
In the experiments presented in this article, we attempted to determine whether Lex GSLs alone could contribute to an increased adhesiveness of cells. As a model we used rat basophilic leukemia (RBL) cells, which do not express Lex and do not exhibit autoaggregation, and analyzed their aggregation after incorporation of exogenous Lex GSLs into plasma membranes. We also analyzed the formation of heterotypic aggregates in mixtures of RBL cells with incorporated Lex GSLs and original Lex-negative cells.
Results
Uptake of Lex GSLs
Lex GSLs were isolated by two different methods as described in Materials and methods. In Method I, GSLs were fractionated on HPLC column of silica gel, and Lex GSLs were detected by TEC-01 monoclonal antibody (mAb) immunostaining in fractions 93-110 (Figure 1a). Lex GSLs were further purified on a prewashed silica gel column and fractions containing Lex GSLs were identified by immunostaining on aluminum sheets of silica gel (Merck) after four times repeated development; under these high resolution conditions more than 10 bands were detected by TEC-01 mAb (Figure 1b). The bands with lower mobility which were not detected in orcinol-H2SO4 developed sheets could reflect a higher activity of TEC-01 mAb with Lex GSLs having longer oligosaccharide chain with repeated Gal[beta]1[rarr]4(Fuc[alpha]1[rarr]3)GlcNAc[beta]1[rarr]3. When anti-H hapten or anti-Ley antibodies were used, no staining was observed (not shown).
Figure
High pure preparations of Lex GSLs were isolated by Method II. When separated by HPTLC and developed with orcinol-H2SO4 only two bands of lipids were observed in these preparations. Immunostaining with SH1 mAb revealed that both of these bands contained Lex GSLs (Figure 1c).
Incubation of RBL cells for 30 min at 37°C in the presence of 200 µg/ml Lex GSLs, isolated according to Method I, resulted in an increased binding of TEC-01 mAb to the cells (Figure 2). Lex GSLs were firmly incorporated into plasma membranes of the cells and were detectable for at least 2 h following incubation of the cells with Lex and extensive washing. Compared to untreated cells, cells pretreated with Lex GSLs, expressed the same amount of glycosylphosphatidylinositol-anchored glycoprotein Thy-1, detected by MRCOX7 mAb (Figure 2), but did not react with anti-H hapten mAb (not shown). Incubation of RBL cells with sphingomyelin (200 µg/ml, Figure 2) or globoside (200 µg/ml, not shown) had no effect on TEC-01 mAb binding.
Figure
Aggregation of Lex-positive RBL cells
To test for a possible adhesion function of Lex GSLs incorporated into the plasma membrane of RBL cells, a sensitive assay for cell adhesion was applied (Urushihara et al., 1979). This assay measures the ability of single cells to form aggregates in suspension as a function of time. The results shown in Figure 3 indicate that control RBL cells did not form aggregates during 60 min incubation in stirring cultures in the presence (Figure 3a) or absence (not shown) of Ca2+. Preincubation of the cells with Lex GSLs at a concentration of 200 µg/ml resulted in aggregate formation in about 20% of cells after 60 min incubation. With an increased concentration of Lex used for preincubation (400 µg/ml), still more cells formed aggregates. Highly purified Lex GSLs, isolated according to Method II, and used at a concentration of 100 µg/ml induced aggregation in more than 30% of RBL cells after 60 min (Figure 3b). The aggregation was dependent on the presence of the Ca2+; when Ca2+ was substituted by EDTA, no aggregation of Lex GSLs-preincubated cells was observed (Figure 3a). The absence of aggregation in Ca2+-free media was not attributable to the release of Lex GSLs from the surface because the cells were still able to bind TEC-01 mAb as determined by flow cytometry (not shown). No aggregation was observed in cells pretreated with sphingomyelin or globoside (Figure 3b), indicating the specificity of aggregation for Lex GSLs.
Figure
Inhibition of aggregation by LFP III
The aggregation of Lex-positive RBL cells was inhibited by inclusion of lacto-N-fucopentaose III (LFP III) into aggregation medium (Figure 4a,b). Inclusion of lactose or lacto-N-fucopentaose II (LFP II) was without any effect on Lex GSLs-mediated aggregation (Figure 4b). These data can be taken as an evidence that the Lex oligosaccharide plays a crucial role in this type of aggregation.
Figure
Aggregation in mixtures of Lex GSL-positive and Lex GSL-negative RBL cells
Lex GSLs incorporated into the cell surface could mediate cellular adhesion via an interaction between Lex-Lex (Eggens et al., 1989), Lex - another oligosaccharide (such as H hapten) (Hakomori, 1992), or Lex - lectin/lectin-like receptor. To determine which of these interactions are involved in the aggregation of Lex-positive RBL cells, we analyzed the formation of aggregates in mixtures of Lex-positive and Lex-negative cells. If Lex-Lex interactions are the cause of cell aggregation, then aggregates should be composed mostly of Lex-positive cells; Lex-negative cells should be excluded from the aggregates. However, should other structures present on the surface of RBL cells be responsible for the formation of the aggregates, then mixed aggregates would form. To distinguish between Lex-positive and Lex-negative RBL cells, one population of the cells was labeled with Hoechst 33258 dye before the aggregation assay. When Lex-positive cells labeled with Hoechst 33258 dye were mixed with unlabeled RBL cells, most of the aggregates were composed exclusively of labeled cells (Figure 5a,d). In reciprocal experiment in which RBL cells were labeled with Hoechst 33258, most of the cells in aggregates were unstained (Figure 5b,e), supporting the notion that Lex-Lex interactions underlie the cell aggregation. Staining with Hoechst dye did not change the aggregation properties of the cells as indicated by mixed aggregates formed in a control experiment in which a mixture of Hoechst 33258-labeled and -unlabeled RBL-Lex cells was analyzed (Figure 5c).
Figure
Discussion
The results presented in this articled indicate that Lex GSLs added to RBL cells can be incorporated into the plasma membrane and subsequently mediate the cell-cell interactions, as evidenced by the formation of aggregates during 60 min incubation in stirring suspensions. Several lines of evidence suggest that this aggregation is mediated via Lex GSL-Lex GSL interactions. First, homotypic aggregation was observed in cells pretreated with Lex GSLs, but not in cells pretreated with sphingomyelin or globoside. It should be pointed out that although the extent of aggregation was relatively weak, compared, for example, with the extent of aggregation mediated by E-cadherin in P19 EC cells (Boubelík et al., 1996), it was highly reproducible and was dependent on the amount of Lex GSLs used for Lex 'painting." Second, aggregation was completely inhibited by LFP III but not with lactose or LFP II (Figure 4). Third, analysis of aggregates in mixtures of RBL and RBL-Lex cells indicated that aggregates were formed almost exclusively by Lex-positive cells (Figure 5); this excluded the possibility that aggregates were formed due to an interaction of Lex GSL with some other carbohydrate, such as H-hapten, which also interacts with Lex (Hakomori, 1992), or with selectins and/or other Lex receptor-like molecules expressed on the surface of RBL cells. The possibility that the H-hapten plays a role in the analyzed aggregation was also excluded by the absence of H-hapten in preparation of GSLs used, and by the immunostaining procedure, indicating that RBL cells do not express the H-hapten.
Lex GSLs used in this study were extracted from human tumors by two different methods. Lex GSLs isolated according to the Method I represented a mixture of TEC-01 mAb positive GSLs which exhibited heterogeneity in size (Figure 1). Lex GSLs isolated according to the Method II represented a double-band of Lex pentasaccharide, as evidenced by their reactivity with SH1 mAb (Figure 1). Importantly, incorporation of both these preparations into the plasma membrane of RBL cells resulted in their aggregation (Figure 3). It should be noted that Lex GSLs from human tumors contained ceramide having long fatty acid chain or long-chain [alpha] hydroxy fatty acid. (Yang and Hakomori, 1971; Hakomori et al., 1984). The presence of long-chain [alpha] hydroxy fatty acid has been shown to greatly enhance the interaction between GalCer and sulfatide (Stewart and Boggs, 1993). Generally, ligand binding including carbohydrate-carbohydrate interactions are more efficient when presented by long chain [alpha] hydroxy fatty acid-containing ceramide, compared to carbohydrates carried by short fatty acid chain plus eicosasphingosine. The sequence of ligand binding may be as follows: first, long-chain [alpha] hydroxy fatty acid plus phytosphingosine >> long-chain fatty acid plus phytosphingosine > long-chain fatty acid plus sphingosine >> short-chain fatty acid plus eicosasphingosine (C20 sphingosine) (S.Hakomori, personal communication).
The mechanism of Lex GSLs-mediated aggregation is unknown but it might involve Lex - Lex recognition resulting from the interaction between the Lex hydrophobic surfaces with subsequent cross-linking by Ca2+ of the two Lex molecules on opposite cells (Kojima et al., 1994). As in other systems based on carbohydrate-carbohydrate interactions which are weak in nature (Eggens et al., 1989; Misevic and Burger, 1993; Stewart and Boggs, 1993), the Lex GSLs- Lex GSLs-mediated interactions probably involve an extensive number of interactions between Lex molecules clustered at the site of cell-cell contacts. The mechanism of this clustering is not known but may involve accumulation of Lex GSLs at the cell-cell contacts promoted by thermodynamic forces acting on a topologically constrained system, similarly as was suggested for other univalent mobile ligands (Weis et al., 1982; Metzger, 1992). Furthermore, large-scale clustering of GSLs at external surface of cell membranes or liposome lipid bilayers has been detected by electron microscopy using ferritin-labeled antibodies with the freeze-fracture technique (Rock et al., 1990, 1991). This clustering could reflect structural complementarity between interacting GSLs allowing them to form large 'patches" (Kojima and Hakomori, 1989, 1991).
Although in the artificial system analyzed in this study the Lex GSLs were probably the only component involved in cell aggregation, a complex array of cell adhesion molecules probably play a role under normal conditions. It has been suggested by Hakomori's group that specific interactions between two homotypic cells could be mediated by multiple low-affinity carbohydrate-carbohydrate interactions, followed by higher-affinity interaction of pericellular adhesive proteins and their receptors (Eggens et al., 1989). Carbohydrates may also modulate adhesive properties of protein molecules. For example, we have found that the absence of Lex and embryoglycan in P19 EC cell mutants has no effect on cell aggregation events mediated mainly by E-cadherin but that it is responsible for cell sorting in heterogeneous aggregating populations (Boubelík et al., 1996). It should be also noted that at morula stage embryo the Lex could be carried by GSLs rather than embryoglycan, and therefore could be involved in Lex-based interactions described in this article. The important role of GSLs in embryogenesis is substantiated by previous data indicating that several other developmentally regulated SSEAs, including SSEA-3 and SSEA-4, are GSLs (Krupnick et al., 1994).
In conclusion, the data described in this article are in accord with previous results (see above), suggesting that Lex-Lex interactions may play a role in cell-cell interactions, and provide, for the first time, an evidence that interactions between Lex GSLs on opposite cell surfaces are strong enough to allow formation of stable cell-cell contacts.
Materials and methods
Isolation of Lex glycosphingolipids
Lex GSLs were isolated by two different methods. In Method I, 60 g of human hepatocellular carcinoma was homogenized in 5 volumes of methanol, followed by addition of chloroform to get the 1:2 (v/v) chloroform/methanol ratio. Second extraction was done with chloroform/methanol/water mixture at a ratio 1:2:0.15 (v/v/v). The ratio of chloroform/methanol/water in combined extracts was adjusted to 30:60:8 (v/v/v) (Ledeen et al., 1973), and ion-exchange separation of neutral and acid GSLs on the DEAE-Sephadex column was performed as described previously (Ueno et al., 1978). The neutral fraction was evaporated to dryness and GSLs with oligosaccharide chains of more than 5 sugar units were removed from chloroform to upper methanol-water phase using Suzuki partition (Suzuki, 1965); this separation removed the bulk of phospholipids and sphingomyelin. For removal of the rest of sphingomyelin and phospholipids, separation of GSLs as acetyl derivatives on a florisil column was used as described (Saito and Hakomori, 1971). The sample was deacetylated, evaporated, dissolved in a small volume of water, and salts together with other low-molecular-weight contaminants were removed by dialysis. After lyophilization the sample was dissolved in 5 ml of chloroform/methanol/water at a ratio of 56:38:10, and fractionated by high-pressure liquid chromatography (HPLC) using silica-gel column Separon SGX, 25 (I.D.) × 250 mm, particle size 7 mm (Tessek, Praha). The sample was applied using a 5 ml rheodyne loop. The following solvents were used for elution: the A solvent was chloroform and the B solvent was methanol/water mixed at a ratio of 95:5 (v/v). Separation was achieved by a combination of linear gradients of solvents according to the following program: 180 min from 17% to 70% of solvent B and 20 min from 70% to 100% of solvent B at a flow rate of 8 ml/min. Fractions, in 16 ml aliquots, were collected, and 2 ml of every third fraction was removed, evaporated, and dissolved in 50 µl of chloroform/methanol (1:2 v/v); 25 µl of each fraction was analyzed by thin layer chromatography (TLC) using TLC plastic sheets with silica gel (Polygram-Sil G Macherey-Nagel, Duren, Germany). One sheet was used for saccharide detection with orcinol detection reagent followed by detection of phospholipids (Vaskovsky and Kostetsky, 1968). The second sheet was used for immunostaining with anti-Lex mAb TEC-01 (see below). Fractions containing Lex-positive glycolipids were collected and evaporated to dryness. The final purification step using chromatography on a small silica-gel column, prewashed and run using distilled solvents, was done as described previously (Ledeen et al., 1973). Purity was checked on high performance (HPTLC) aluminum sheets of silica gel (Merck, Darmstadt) with orcinol, phosphorus, and immunodetections.
In method II, 3 liver metastases of colon carcinoma (1899 g) were homogenized and extracted with 5 volumes of isopropanol: hexane: water (55:25:20 v/v/v), followed by another extraction with 2.5 volumes of the same solvent mixture. Solvents were evaporated with rotary evaporator, the extract was dissolved in chloroform: methanol (2:1 v/v) and partitioned as described previously (Folch et al., 1971). The upper Folch fraction was evaporated and dialyzed against distilled water for 3 days. The neutral fraction was separated from the dialysate by DEAE-Sephadex A-25 chromatography in chloroform:methanol:water (30:60:8 v/v/v) and subjected to Iatrobeads HPLC with gradient elution with isopropanol/hexane/water from 55:35:10 to 55:25:20 over 400 min (one fraction per 4 min). Lex pentasaccharide glycolipids were probed by reactivity with an SH1 mAb (Singhal et al., 1987) and pooled, and HPLC chromatography was repeated under the same conditions as above. Lex pentasaccharide glycolipids were recollected and subjected to a third HPLC chromatography with a gradient of isopropanol/hexane/water from 55:45:5 to 55:35:10 to separate contaminating nLc4 (paragloboside) and Ley glycolipid. Three fractions, named c, d, and e containing, respectively, 3.9, 2.0, and 1.3 mg of GSLs were used for experiments described here.
Immunostaining on TLC plastic sheets
For immunostaining, the TLC sheet was blocked for 1 h at 37°C in buffer A containing phosphate-buffered saline (PBS), pH 7.2, supplemented with 1% polyvinylpyrrolidone (PVP), 1% bovine serum albumin (BSA) and 0.02% NaN3. The sheet was then incubated for 1 h at 37°C with Lex-specific mAb, TEC-01 (Dráber and Pokorná, 1984) or H-hapten-specific mAb (kindly provided by Dr. Nemec); the antibodies in the form of ascites were diluted 1:200 in buffer A without PVP. After washing with PBS, the sheet was blocked once more with buffer A for 15 min at 37°C and incubated with swine anti-mouse antibody for 1 h at 37°C. The sheet was washed and blocked again and incubated with rabbit anti-swine antibody labeled with horseradish peroxidase, diluted 1:200 in buffer A. After washing, the peroxidase activity was detected using 4-chlor-1-naphtol reagent as described (Harlow and Lane, 1988).
Immunostaining on HPTLC aluminum sheets of silica gel
Silica-gel layer of HPTLC sheet was impregnated for 75 s with 0.1% solution of polyisobutylmethacrylate in cyclohexane and dried. The chromatogram was blocked in buffer B, containing 1% BSA in PBS, for 15 min at room temperature, incubated for 1 h with TEC-01 mAb, diluted 1:50 in buffer B, washed with PBS, blocked in buffer B for 5 min, and incubated for 1 h with horseradish peroxidase-conjugated swine anti-mouse antibody, diluted 1:200 in buffer B. The peroxidase activity was detected as described above.
Cell culture
RBL cells, clone 2H3 (Siraganian et al., 1982), were cultured in complete culture medium (1:1 mixture of medium RPMI 1640 and MEM) supplemented with 10% fetal calf serum (FCS) as described (Dráberová and Dráber, 1991). Semiconfluent cultures were harvested with PBS/citrate supplemented with 0.125% trypsin (Difco). Trypsin-released cells were transferred into Puck's saline supplemented with 10% (vol/vol) FCS, and centrifuged at 150 × g for 5 min. The cells were incubated for 30 min with or without various lipids (Lex GSLs, globoside or sphingomyelin) in PBS and washed twice before further analysis. For aggregation experiments, the cells were resuspended in aggregation solution (Puck's saline containing 0.8% FCS and 10 µg/ml DNase).
Flow cytometry
Untreated or lipid-treated cells were reacted for 30 min with the first layer antibody [TEC-01, anti-H hapten, MRCOX7 (Mason and Williams, 1980) or negative control mAb] followed by the fluorescein isothiocyanate-labeled second layer antibody as described previously (Dráber et al., 1988). Cells were examined by flow cytometry using a FACScan apparatus (Beckton Dickinson, USA).
Aggregation studies
Cells in aggregation solution were passed three times through a 30-gauge needle, counted and adjusted to the concentration of 2 × 106/ml; 300 µl aliquots of the cell suspension were transferred into 2 ml cryotubes (Greiner, Germany). Some samples were supplemented with 1 mM CaCl2, 1 mM EDTA, lacto-N-fucopentaose III (LFP III), and/or other reagents as indicated in Results. The cells were magnetically stirred at 100 r.p.m. for 60 min at 37°C, and homotypic adhesion was evaluated by visual measurement in a hemocytometer. The percentage of remaining single cells was expressed as a function of time. Representative experiments from three to five performed are shown.
Cell sorting
For quantitative analysis of the aggregation, confluent cultures were washed with complete culture medium and the cells labeled for 30 min with Hoechst 33258 dye at a concentration of 10 µg/ml in culture medium supplemented with 1% FCS. The cells were washed, trypsinized, and centrifuged for 10 min at 250 × g through FCS to remove free Hoechst 33258 dye. Hoechst 33258-labeled and -unlabeled RBL cells were incubated for 30 min with Lex GSLs as described above. Washed cells were forced three times through a 30-gauge needle to get single-cell suspension, counted and mixed at the 1:1 ratio in aggregation solution; final concentration 4 × 106/ml in a volume of 0.5 ml. The cells were allowed to aggregate on a stirring platform for 45 min at 37°C and aggregates were fixed with 2.5% glutaraldehyde in PBS. The fixed cells were applied to microscopic slides and examined by fluorescence microscopy using 461 nm filter; Hoechst 33258-labeled cells exhibited bright fluorescence whereas unlabeled cells showed only background autofluorescence. Aggregates that were composed of 5-30 cells were randomly selected, and the percentage of labeled cells in each aggregate was determined; 100 aggregates were evaluated in each sample.
Acknowledgments
We thank S. Hakomori for his help with isolation and characterization of the highly purified Lex GSLs (Method II). This work was supported in part by Grants A5052506 and A5052704 from the Grant Agency of the Academy of Sciences of the Czech Republic; Grants 302/94/1294, 310/97/0237, and 204/97/0239 from the Grant Agency of the Czech Republic; and Grant M18-3 from Ministry of Health of the Czech Republic. The research of P. Dráber was supported in part by an International Research Scholar's award from Howard Hughes Medical Institute.
Abbreviations
BSA, bovine serum albumin; EC, embryonal carcinoma; FCS, fetal calf serum; GSL, glycosphingolipid; HPLC, high-pressure liquid chromatography; HPTLC, high performance TLC; LFP II, lacto-N-fucopentaose II, LFP III, lacto-N-fucopentaose III; mAb, monoclonal antibody; PBS, phosphate buffered saline; PVP, polyvinylpyrrolidone; RBL, rat basophilic leukaemia; SSEA-1, stage-specific embryonic antigen 1; TLC, thin layer chromatography.
References
This page is run by Oxford University Press, Great Clarendon Street, Oxford OX2 6DP, as part of the OUP Journals
Comments and feedback: www-admin{at}oup.co.uk
Last modification: 27 Jan 1998
Copyright© Oxford University Press, 1998.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  D. A. Withers and S.-i. Hakomori Human alpha (1,3)-Fucosyltransferase IV (FUTIV) Gene Expression Is Regulated by Elk-1 in the U937 Cell Line J. Biol. Chem., December 15, 2000; 275(51): 40588 - 40593. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Pykari, S. Toivonen, J. Natunen, R. Niemela, H. Salminen, O. Aitio, M. Ekstrom, P. Parmanne, M. Valimaki, J. Alais, et al. The Acceptor and Site Specificity of alpha 3-Fucosyltransferase V. HIGH REACTIVITY OF THE PROXIMAL AND LOW OF THE DISTAL Galbeta 1-4GlcNAc UNIT IN i-TYPE POLYLACTOSAMINES J. Biol. Chem., December 15, 2000; 275(51): 40057 - 40063. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Z. Cao, Z. Zhao, R. Mohan, J. Alroy, P. Stanley, and N. Panjwani Role of the Lewisx Glycan Determinant in Corneal Epithelial Cell Adhesion and Differentiation J. Biol. Chem., June 8, 2001; 276(24): 21714 - 21723. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||